Vitamin D, Volume 1-2, Second Edition

List of Contributors

Numbers in parentheses indicate the page(s) on which the authors’ contributions begin.

Steven A. Abrams (811)

USDA/ARS Children’s Nutrition Research Center, Houston, Texas 77030, USA John S. Adams (341, 1379)

Division of Endocrinology, Diabetes, and Metabolism, University of California, 8700 Beverly Blvd, Los Angeles, Cedars-Sinai Medical Center, Los Angeles, California 90048, USA Judith E. Adams (967)

Clinical Radiology, Imaging Science and Biomedical Engineering, Stopford Building, The University, Manchester M13 9PT, United Kingdom Luciano Adorini (631, 1511, 1833)

BioXell SpA, 20132 Milano, Italy Paul H. Anderson (711)

Institute of Medical and Veterinary Science, Adelaide, South Australia, Australia Gerald J. Atkins (711)

Hanson Institute, Adelaide, South Australia, Australia Jane E. Aubin (649)

Department of Molecular and Medical Genetics, Faculty of Medicine, University of Toronto, Toronto, Ontario M5S 1A8, Canada Isabelle Bailleul-Forestier (599)

INSERM E110, Université Paris VII, IFR58, Cordeliers Biomedical Institute, 75270 Paris Cedex 06, France Julia Barsony (363)

Laboratory of Cellular Biochemistry and Biology, NIDDK/NIH, Bethesda, Maryland 20892-0850, USA Thomas K. Barthel (219)

Department of Biochemistry and Molecular Biophysics, University of Arizona, Tucson, Arizona 85721, USA

Norman H. Bell (789)

Department of Medicine, Medical University of South Carolina, Charleston, South Carolina 29425, USA Ariane Berdal (599)

INSERM E110, Université Paris VII, IFR58, Cordeliers Biomedical Institute, 75270 Paris Cedex 06, France Joel J. Bergh (751)

Department of Biological Sciences, University of Delaware, Newark, Delaware 19716, USA Jacqueline L. Berry (1293)

University of Manchester, Vitamin D Research Group, Department of Medicine, Manchester Royal Infirmary, Manchester M13 9WL, UK Daniel D. Bikle (609)

Endocrine Research Unit, Veterans’ Affairs Medical Center, University of California-San Francisco, San Francisco, California 94121-1598, USA John P. Bilezikian (1355)

Department of Medicine, College of Physicians and Surgeons, Columbia University, New York, New York 10032, USA Ernst Binderup (1489)

Biological Research, Leo Pharma, DK-2750 Ballerup, Denmark Lise Binderup (1489)

Biological Research, Leo Pharma, DK-2750 Ballerup, Denmark Nicholas J. Bishop (803)

Academic Department of Child Health, University of Sheffield, Sheffield Children’s Hospital, Sheffield S10 2TH, United Kingdom Ilse Bogaerts (135)

Laboratory Analytische Chemie, Van Evenstraat 4; B-3000 Leuven, Belgium

xiv Ricardo L. Boland (883)

Department de Biología, Bioquímica & Farmacía, Universidad Nacional del Sur, San Juan 670, (8000) Bahía Blanca, Argentina Adele L. Boskey (477)

Mineralized Tissues Laboratory, Hospital for Special Surgery, Affiliated with Weil College of Cornell Medical School, New York, New York 10021, USA Roger Bouillon (135, 429, 1763)

Legendo, Onderwijs en Navorsing 902, U.Z. Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium Barbara D. Boyan (575)

Department of Biomedical Engineering at Georgia Tech and Emory University, Georgia Institute of Technology, Atlanta, Georgia 30332, USA Philippe Brachet (1779)

INSERM U 643, Centre Hospitalier Universitaire, 30 bd Jean Monnet, 44093 Nantes, France Alex J. Brown (1313, 1449)

Renal Division, Washington University School of Medicine, St. Louis, Missouri 63110, USA Edward M. Brown (551)

Division of Endocrinology, Diabetes and Hypertension, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts 02115, USA Carsten Carlberg (313)

Department of Biochemistry, University of Kuopio, PFIN-70211 Kuopio, Finland Geert Carmeliet (429)

Legendo, Onderwijs en Navorsing 902, U.Z. Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium Thomas O. Carpenter (1049)

Department of Pediatrics, Yale University, School of Medicine, New Haven, Connecticut 06520-8064, USA Marie-Claire Chapuy (1085)

INSERM Unit 403, Faculty Laennec and Department of Rheumatology and Bone Disease, Edouard Herriot Hosptial, Lyon, France Fredriech K. W. Chan (1355)

Department of Medicine, Queen Elizabeth Hospital, Hong Kong Tai C. Chen (1599)

Vitamin D, Skin, and Bone Research Laboratory, Boston University Medical Center, Boston, Massachusetts 02188, USA Sylvia Christakos (721)

Department of Biochemistry and Molecular Biology, University of Medicine and Dentistry of New Jersey, New Jersey Medical School, Newark, New Jersey 07103, USA

LIST

OF

CONTRIBUTORS

Margaret Clagett-Dame (1543)

Department of Biochemistry, University of Wisconsin–Madison, Madison, Wisconsin 53706, USA Thomas L. Clemens (899)

Department of Cell Biology and Physiology, Department of Internal Medicine, University of Cincinnati School of Medicine, Cincinnati, Ohio 45276, USA Je-Yong Choi (327)

Department of Biochemistry, Kyungpook National University, Daegu, Korea Fredric L. Coe (1339)

Nephrology Section, The University of Chicago Pritzker School of Medicine, Chicago, Illinois 60637, USA Kay Colston (1663)

Department OGEM, St. George’s Hospital Medical School, London, SW17 0RE, United Kingdom Juliet E. Compston (951)

Department of Medicine, Level 5, University of Cambridge, School of Clinical Medicine, Addenbrooke’s Hospital, Cambridge, England CB2 2QQ, United Kingdom Nancy E. Cooke (117)

Department of Genetics and Medicine, University of Pennsylvania, Philadelphia, Pennsylvania 19104-6149, USA Clara Crescioli (1833)

Endocrinology Unit, Department of Clinical Physiopathology, University of Florence, Florence 50139, Italy Heide S. Cross (1709)

Department of Pathophysiology, Medical, University of Vienna, A-1090 Vienna, Währingergürtel 18-20, Austria Michael Danilenko (1635)

Department of Clinical Biochemistry, Faculty of Health Science, Ben-Gurion University of the Negev, Beer-Sheva 84105, Israel Jean-Luc Davideau (599)

INSERM E110, Université Paris VII, IFR58, Cordeliers Biomedical Institute, 75270 Paris Cedex 06, France Michael Davies (1293)

Vitamin D Research Group, University Department of Medicine, Manchester Royal Infirmary, Manchester, M13 9WL, United Kingdom Hector F. DeLuca (3, 1543)

Department of Biochemistry, University of Wisconsin–Madison, Madison, Wisconsin 53706, USA Marie B. Demay (341)

Endocrine Unit, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts 02114, USA

LIST

OF

xv

CONTRIBUTORS

Puneet Dhawan (721)

Department of Biochemistry and Molecular Biology, University of Medicine and Dentistry of New Jersey, New Jersey Medical School, Newark, New Jersey 07103, USA Carlos Encinas Dominguez (219)

Department of Biochemistry and Molecular Biophysics, University of Arizona, Tucson, Arizona 85721, USA Diane R. Dowd (291)

Department of Pharmacology–W334, Case Western Reserve University, Cleveland, Ohio 44106, USA Marc K. Drezner (1159)

Department of Medicine, Endocrinology, Diabetes, and Metabolism Section, University of Wisconsin–Madison, Madison, Wisconsin 53792, USA Thomas W. Dunlop (313)

Department of Biochemistry, University of Kuopio, PFIN-70211 Kuopio, Finland Adriana S. Dusso (1313)

Renal Division, Washington University School of Medicine, St. Louis, Missouri 63110, USA Richard Eastell (1101)

University of Sheffield Clinical Sciences Centre, Northern General Hospital, Sheffield South Yorkshire S5 7AU, United Kingdom Michael J. Econs (1189)

Indiana University School of Medicine, Department of Medicine and Medical and Molecular Genetics, Indianapolis, Indiana 46202, USA John Eisman (193)

Bone and Mineral Program, Garvan Institute of Medical Research, St. Vincent’s Hospital, Sydney, New South Wales 2010, Australia Sol Epstein (1253)

Division of Endocrinology, Diabetes, and Bone Diseases, Department of Medicine, Mount Sinai School of Medicine, New York, New York 10029, USA Erik Fink Eriksen (1805)

Osteoporosis Team of Lilly Research Laboratories, Lilly Corp Center, Indianapolis, Indiana 46285, USA Luis M. Esteban (193)

Bone and Mineral Program, Garvan Institute of Medical Research, St. Vincent’s Hospital, Sydney, New South Wales 2010, Australia Dan Faibish (477)

Mineralized Tissues Laboratory, Hospital for Special Surgery, Affiliated with Weil College of Cornell Medical School, New York, New York 10021, USA

Yue Fang (1121)

Department of Internal Medicine, Genetic Laboratory, Erasmus Medical Centre, NL-3015 GE Rotterdam, The Netherlands Mary C. Farach-Carson (751)

Department of Biological Sciences, University of Delaware, Newark, Delaware 19716, USA Murray J. Favus (1339)

Section of Endocrinology, University of Chicago, Chicago, Illinois 60637, USA David Feldman (1207, 1679)

Stanford University School of Medicine, Division of Endocrinology, Stanford, California 94305-5103, USA David Findlay (711)

Department of Orthopedic Surgery and Trauma, University of Adelaide, Adelaide 5000, South Australia, Australia Christian Frank (313)

Department of Biochemistry, University of Kuopio, PFIN-70211 Kuopio, Finland Leonard P. Freedman (263)

Department of Molecular Endocrinology and Bone Biology, Merck Research Laboratories, West Point, Pennsylvania 19486-0004, USA Ryuji Fujiki (305)

University of Tokyo, Institute of Molecular and Cellular Biosciences, 1-1-1 Yayoi-cho, Bunkyo-ku, Tokyo, 113-0032, Japan Masafumi Fukagawa (1821)

Division of Nephrology and Dialysis Center, Kobe University School of Medicine, 7-5-2 Kusunoki-cho, Chuo-ku, Kobe 650-0017, Japan Robert F. Gagel (687)

Section of Endocrinology, University of Texas M.D. Anderson Cancer Center, Houston, Texas, USA Emmanuel Garcion (1779)

INSERM U 646, 10 rue André Boquel, 49100 Angers, France Edith M. Gardiner (193)

Bone and Mineral Program, Garvan Institute of Medical Research, St. Vincent’s Hospital, Sydney, New South Wales 2010, Australia Marielle Gascon-Barré (47)

Département de Pharmacologie, Faculté de Médecine, Université de Montréal, and Centre de recherche de l’Université de Montréal, Montréal, Québec H2X 1P1, Canada Edward Giovannucci (1617)

Harvard School of Public Health, Department of Nutrition, Boston, Massachusetts 02115, USA Henning Glerup (1805)

Aarhus Kommunehospital, Dept V, Noerrebrogade 44, DK-8000 Aarhus C, Denmark

xvi Francis H. Glorieux (1197)

Genetics Unit, Shriners Hospital for Children, Departments of Surgery, Pediatrics, and Human Genetics, McGill University, Montréal, Québec H3G 1A6, Canada Wagn O. Godtfredsen (1489)

Medicinal Chemistry, Leo Pharma, DK-2750 Ballerup, Denmark David Goltzman (737)

Department of Medicine, McGill University and McGill University Health Center, Montréal, Québec H3A 1A1, Canada Soraya Gutierrez (327)

Departamento de Biología Molecular, Universidad de Concepción, Concepción, Chile Conny Gysemans (1763)

Legendo, Onderwijs en Navorsing 902, U.Z. Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium Bernard P. Halloran (823)

Division of Endocrinology, Veterans Affairs Medical Center, San Francisco, California 94121, USA Carol A. Haussler (219)

Department of Biochemistry and Molecular Biophysics, University of Arizona, Tucson, Arizona 85721, USA Mark R. Haussler (219)

Department of Biochemistry and Molecular Biophysics, University of Arizona, Tucson, Arizona 85721, USA Robert P. Heaney (773)

Creighton University, Omaha, Nebraska 68131, USA Johan Heersche (649)

Faculty of Dentistry, University of Toronto, Toronto M5S 1A8, Ontario, Canada Helen L. Henry (69)

Department of Biochemistry, University of California–Riverside, Riverside, California 92521, USA Pamela A. Hershberger (1741)

Department of Pharmacology, University of Pittsburgh, Pittsburgh, Pennsylvania 15213, USA Martin Hewison (1379)

Division of Medical Sciences, The University of Birmingham, Queen Elizabeth Medical Centre, Birmingham, B15 2TH, United Kingdom Richard A. Heyman (1557)

X-Ceptor Therapeutics, San Diego, California 92121, USA Kanji Higashio (665)

Research Center for Genomic Medicine, Saitama Medical School, Saitama, 350-1241, Japan

LIST

OF

CONTRIBUTORS

Michael F. Holick (37, 1511, 1791)

Vitamin D, Skin, and Bone Research Laboratory; Department of Medicine; Endocrinology, Nutrition and Diabetes Section; Boston Medical Center and Boston University School of Medicine, Boston, Massachusetts 02118, USA Bruce W. Hollis (931)

Departments of Pediatrics, Biochemistry, and Molecular Biology, Medical University of South Carolina, Charleston, South Carolina 29425, USA Ronald L. Horst (15)

U.S. Department of Agriculture, Agricultural Research Service, National Animal Disease Center, Ames, Iowa 50010-0070, USA Jui-Cheng Hsieh (219)

Department of Biochemistry and Molecular Biophysics, University of Arizona, Tucson, Arizona 85721, USA Karl L. Insogna (1049)

Department of Medicine, Yale University School of Medicine, New Haven, CT 06520-8064, USA Elizabeth T. Jacobs (219)

College of Medicine, University of Arizona, Tucson, Arizona 85721, USA Amjad Javed (327)

Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Glenville Jones (1423)

Department of Biochemistry, Queen’s University, Kingston, Ontario K7L 3N6, Canada Candace S. Johnson (1741)

Department of Pharmacology & Therapeutics, Roswell Park Cancer Institute, Buffalo, New York 14263, USA Peter W. Jurutka (219)

Department of Biochemistry and Molecular Biophysics, University of Arizona, Tucson, Arizona 85721, USA Mehmet Kahraman (1405)

Department of Chemistry, Johns Hopkins University, Baltimore, Maryland 21218-2685, USA S. Kaleem Zaidi (327)

Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Heidi J. Kalkwarf (839)

Division of General and Community Pediatrics, Children’s Hospital Medical Center, Cincinnati, Ohio 45229, USA

LIST

OF

xvii

CONTRIBUTORS

Shigeaki Kato (305)

University of Tokyo, Institute of Molecular and Cellular Biosciences, 1-1-1 Yayoi-cho, Bunkyo-ku, Tokyo, 113-0032, Japan Anne-Marie Kissmeyer (1489)

Biological Research, Leo Pharma, DK-2750 Ballerup, Denmark Hirochika Kitagawa (305)

University of Tokyo, Institute of Molecular and Cellular Biosciences, 1-1-1 Yayoi-cho, Bunkyo-ku, Tokyo, 113-0032, Japan Lilia M. C. Koberle (1355)

Health Sciences Department, Federal University, Sao Carlos, Brazil H. Phillip Koeffler (1727)

Hematology/Oncology Division, Cedars-Sinai Medical Center, Los Angeles, California 90048, USA Ruth Koren (761)

Felsenstein Medical Research Center, Beilinson Campus, Rabin Medical Center, Petah Tikva 49100, Israel Barbara E. Kream (703)

Department of Medicine, University of Connecticut Health Center, Farmington, Connecticut 06030-1850, USA Richard Kremer (737)

Department of Medicine, McGill University and McGill University Health Center, Montréal, Québec H3A 1A1, Canada Aruna V. Krishnan (1679)

Stanford University School of Medicine, Division of Endocrinology, Stanford, California 94305-5103, USA Noboru Kubodera (1525)

Department of Product Planning, Chugai Pharmaceutical Co. Ltd., 2-1-9 Kyobashi, Chuo-ku, Tokyo, 104-8301, Japan Rajiv Kumar (515)

Departments of Medicine, Biochemistry and Molecular Biology and Mayo Proteomics Research Center, Divisions of Nephrology, and Endocrinology, Diabetes, Metabolism, and Nutrition, Mayo Clinic and Foundation, Rochester, Minnesota 55905-0002, USA Kiyoshi Kurokawa (1821)

Research Center for Advanced Science and Technology, The University of Tokyo, Tokyo, Japan Christopher J. Laing (117)

Department of Genetics and Medicine, University of Pennsylvania, Philadelphia, Pennsylvania 19104-6144, USA Jacques Lemire (1753)

Pediatric Nephrology, University of California-San Diego, La Jolla, California 92093-0831, USA

Frédéric Lézot (599)

INSERM E110, Université Paris VII, IFR58, Cordeliers Biomedical Institute, 75270 Paris Cedex 06, France Yan Chun Li (721, 871, 1511)

Department of Medicine/GI Section, University of Chicago, Chicago, Illinois 60637, USA Jane B. Lian (327)

Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Alexander C. Lichtler (703)

Department of Genetics and Developmental Biology, The University of Connecticut Health Center, Farmington, Connecticut 06030, USA Paul Lips (1019)

Department of Endocrinology, Vrijie University Medical Center, Amsterdam, 1007 MB, The Netherlands Yan Liu (721)

Department of Biochemistry and Molecular Biology, New Jersey Medical School, Newark, New Jersey 07103, USA Paul MacDonald (291)

Department of Pharmacology–W334, Case Western Reserve University, Cleveland, Ohio 44106, USA Hubert Maehr (1511)

BioXell, Inc., Hoffmann-La Roche, Inc., Nutley, New Jersey 07110-1199, USA Mario Maggi (1833)

Andrology Unit, Department Clinical Physiopathology, University of Florence, 50139 Florence, Italy Peter J. Malloy (1207)

Stanford University School of Medicine, Division of Endocrinology, Stanford, California 94305-5103, USA David J. Mangelsdorf (863)

Department of Pharmacology, Howard Hughes Medical Institute, University of Texas Southwestern Medical Center at Dallas, Dallas, Texas 75390-9050, USA Chantal Mathieu (1763)

Legendo, Onderwijs en Navorsing 902, U.Z. Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium Brian May (85)

School of Molecular and Biomedical Science, University of Adelaide, South Australia 5005, Australia Andrew P. Mee (1293)

Vitamin D Research Group, University Department of Medicine, Manchester Royal Infirmary, Manchester, M13 9WL, United Kingdom

xviii Pierre J. Meunier (1085)

INSERM Unit 403, Faculty Laennec and Department of Rheumatology and Bone Disease, Edouard Herriot Hospital, Lyon, France Toshimi Michigami (851)

Department of Environmental Medicine, Osaka Medical Center and Institute for Maternal and Child Health, Osaka, Japan Martin Montecino (327)

Departamento de Biología Molecular, Universidad de Concepción, Concepción, Chile Dino Moras (279)

Département de Biologie et de Génomique Structurales, CNRS/INSERM/Université Louis Pasteur 1, BP 10142, 67404 Illkirch Cedex, France Roberta Morosetti (1727)

Pediatric Oncology Division, Catholic University of Rome, Rome, Italy Howard A. Morris (711)

Hanson Institute, Adelaide, South Australia, Australia Daniel L. Motola (863)

Department of Pharmacology, Howard Hughes Medical Institute, University of Texas Southwestern Medical Center at Dallas, Dallas, Texas 75390-9050, USA Josephia Muindi (1741)

Department of Medicine, Roswell Park Cancer Institute, Buffalo, New York 14263, USA Shigeo Nakajima (851)

Department of Developmental Medicine (Pediatrics), D-5, Osaka University Graduate School of Medicine, Osaka 565-0871, Japan Tally Naveh-Many (537)

Minerva Center for Calcium and Bone Metabolism, Hebrew University Hadassah Medical Center, Ein Karem, Jerusalem, 91120, Israel Philippe Naveilhan (1779)

INSERM U 643, Centre Hospitalier Universitaire, 30 bd Jean Monnet, 44093 Nantes, France Isabelle Neveu (1779)

INSERM U 643, Centre Hospitalier Universitaire, 30 bd Jean Monnet, 44093 Nantes, France Anthony W. Norman (381)

Department of Biochemistry, University of California, Riverside, Riverside, California 92521-0129, USA Anders Nykjaer (153)

Institute of Medical Biochemistry, University of Aarhus, Ole Worms Allee, DK-8000 Aarhus C, Denmark James O’Kelly (1727)

Hematology/Oncology Division, Cedars-Sinai Medical Center, Los Angeles, California 90048, USA

LIST

OF

CONTRIBUTORS

John L. Omdahl (85)

Office of Research, University of New Mexico School of Medicine, Albuquerque, New Mexico 87131-5166, USA Peter Ordentlich (1557)

X-Ceptor Therapeutics, San Diego, California 92121, USA Keiichi Ozono (851)

Department of Developmental Medicine (Pediatrics), D-5, Osaka University Graduate School of Medicine, Osaka 565-0871, Japan A. Michael Parfitt (497, 1029)

Division of Endocrinology and Center for Osteoporosis and Metabolic Bone Disease, University of Arkansas for Medical Science, Little Rock, Arkansas 72205, USA Donna M. Peehl (1679)

Stanford University School of Medicine, Division of Endocrinology, Stanford, California 94305-5103, USA Sara Peleg (1471)

Department of Endocrine Neoplasia and Hormonal Diseases, Unit 435, University of Texas, M.D. Anderson Cancer Center, Houston, Texas 77030-4009, USA Xiaorong Peng (721)

Department of Biochemistry and Molecular Biology, University of Medicine and Dentistry of New Jersey, New Jersey Medical School, Newark, New Jersey 07103, USA John M. Pettifor (1065)

Department of Pediatrics, Chris Hani Baragwanath Hospital, Mineral Metabolism Research Unit, P.O. Bertsham, Johannesburg, Gauteng 2013, South Africa J. Wesley Pike (167, 1207, 1403, 1543)

Department of Biochemistry, University of Wisconsin–Madison, Madison, Wisconsin 53706, USA Elizabeth A. Platz (1617)

Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland 21205, USA Lori A. Plum (1543)

Department of Biochemistry, University of Wisconsin–Madison, Madison, Wisconsin 53706, USA Shirwin Pockwinse (327)

Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Huibert A.P. Pols (1121, 1571)

Department of Internal Medicine, Erasmus Medical Centre, NL-3015 GD Rotterdam, The Netherlands

LIST

OF

xix

CONTRIBUTORS

Anthony A. Portale (453, 823)

Department of Pediatrics, University of California – San Francisco, San Francisco, California 94121, USA Gary H. Posner (1405)

Department of Chemistry, Johns Hopkins University, Baltimore, Maryland 21218-2685, USA Mehrdad Rahmaniyan (789)

Department of Medicine, Medical University of South Carolina, Charleston, South Carolina 29425, USA Amiram Ravid (761)

Felsenstein Medical Research Center, Beilinson Campus, Rabin Medical Center, Petah Tikva 49100, Israel G. Satyanarayana Reddy (15, 1511)

Brown University, Department of Chemistry, Providence, Rhode Island, USA Jörg Reichrath (1791)

The Saarland University Hospital, Department of Dermatology, Kirrberger Str., 66421 Homburg/Saar, Germany Timothy A. Reinhardt (15)

U.S. Department of Agriculture, Agricultural Research Service, National Animal Disease Center, Ames, Iowa 50010-0070, USA Alfred A. Reszka (263)

Department of Molecular Endocrinology and Bone Biology, Merck Research Laboratories, West Point, Pennsylvania 19486-0004, USA B. Lawrence Riggs (1101)

Division of Endocrinology, Mayo Clinic and Mayo Foundation, Rochester, Minnesota 55905, USA Natacha Rochel (279)

Département de Biologie et de Génomique Structurales, CNRS/INSERM/Université Louis Pasteur 1, BP 10142, 67404 Illkirch Cedex, France Mishaela R. Rubin (1355)

Department of Medicine, College of Physicians and Surgeons, Columbia University, New York, New York 10027, USA Andrew F. Russo (687)

Department of Physiology and Biophysics, University of Iowa, Iowa City, Iowa 52242, USA Philip Sambrook (1239)

Institute of Bone & Joint Research, University of Sydney, Royal North Shore Hospital, Sydney 2065, Australia Adina E. Schneider (1253)

Division of Endocrinology, Diabetes, and Bone Diseases, Department of Medicine, Mount Sinai School of Medicine, New York, New York 10029, USA

Gary G. Schwartz (1599)

Comprehensive Cancer Center, Wake Forest University School of Medicine, Winston-Salem, North Carolina 27157, USA Zvi Schwartz (575)

Department of Periodontics, Hebrew University Hadassah Faculty of Dental Medicine, Jerusalem, Israel Jiali Shen (327)

Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Nirupama K. Shevde (167, 1543)

Department of Biochemistry, University of Wisconsin–Madison, Madison, Wisconsin 53706, USA Rafal R. Sicinski (1543)

Department of Chemistry, University of Warsaw, ul. L. Pasteura 2, Warsaw 02-093, Poland Justin Silver (537)

Minerva Center for Calcium and Bone Metabolism, Hebrew University Hadassah Medical Center, Ein Karem, Jerusalem, 91120, Israel Eduardo A. Slatopolsky (1313, 1449)

Renal Division, Washington University School of Medicine, St. Louis, Missouri 63110, USA Bonny L. Specker (839)

Martin Program in Human Nutrition, South Dakota State University, Brookings, South Dakota 57007, USA René St-Arnaud (105, 1197)

Genetics Unit, Shriners Hospital for Children, Montréal, Quebec H3G 1A6, Canada Gary S. Stein (327)

Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Janet L. Stein (327)

Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Paula H. Stern (565)

Department of Molecular Pharmacology, Northwestern University Feinberg School of Medicine, Chicago, Illinois 60611, USA George P. Studzinski (1635)

UMD-New Jersey Medical School, Newark, New Jersey 07103, USA Tatsuo Suda (665)

Research Center for Genomic Medicine, Saitama Medical School, Hidaka-shi, Saitama 350-1241, Japan

xx

LIST

Amelia L.M. Sutton (291)

Department of Pharmacology–W334, Case Western Reserve University, Cleveland, Ohio 44106, USA Peter Tebben (515)

Departments of Medicine, Biochemistry and Molecular Biology and Mayo Proteomics Research Center, Divisions of Nephrology, and Endocrinology, Diabetes, Metabolism, and Nutrition, Mayo Clinic and Foundation, Rochester, Minnesota 55905-0002, USA Harriet S. Tenenhouse (453)

Departments of Pediatrics and Human Genetics, McGill University and Montréal Children’s Hospital Research Institute, Montréal, Québec, H3Z 2Z3, Canada Michelle L. Thatcher (219)

Department of Biochemistry and Molecular Biophysics, University of Arizona, Tucson, Arizona 85721, USA Susan Thys-Jacobs (1355)

Department of Medicine, College of Physicians and Surgeons, Columbia University, New York, New York 10027, USA Dwight A. Towler (899)

Department of Medicine, Washington University School of Medicine, St. Louis, Missouri 63110, USA Donald L. Trump (1741)

Department of Medicine, Roswell Park Cancer Institute, Buffalo, New York 14263, USA André G. Uitterlinden (1121)

Genetic Laboratory, Department of Internal Medicine, Erasmus Medical Centre, NL-3015 GD Rotterdam, The Netherlands Milan R. Uskokovi´c (1511)

BioXell, Inc., Nutley, New Jersey 07110-1199, USA Sami Väisänen (313)

Department of Biochemistry, University of Kuopio, PFIN-70211 Kuopio, Finland Hugo Van Baelen (135)

Legendo, Onderwijs en Navorsing 902, U.Z. Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium Sophie Van Cromphaut (429)

Legendo, Onderwijs en Navorsing 902, U.Z. Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium Johannes P.T.M. van Leeuwen (1571)

Department of Internal Medicine, Erasmus MC, 3000 DR Rotterdam, The Netherlands

OF

CONTRIBUTORS

Joyce B.J. van Meurs (1121)

Genetic Laboratory, Department of Internal Medicine, Erasmus Medical Centre, NL-3015 GD Rotterdam, The Netherlands Andre J. van Wijnen (327)

Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Christel Verboven (135)

Laboratory Analytische Chemie, Van Evenstraat 4; B-3000 Leuven, Belgium Reinhold Vieth (995)

Pathology and Laboratory Medicine, Mount Sinai Hospital, Toronto, Ontario, M5G 1X5, Canada Robert H. Wasserman (411)

Department of Biomedical Sciences, VRT-08-20, Cornell University, Ithaca, New York 14853, USA JoEllen Welsh (1663)

Department Biological Science, University Notre Dame, Notre Dame, Indiana 46556, USA G. Kerr Whitfield (219)

Department of Biochemistry and Molecular Biophysics, University of Arizona, Tucson, Arizona 85721, USA Michael P. Whyte (913)

Center for Metabolic Bone Disease and Molecular Research, Shriners Hospital for Children, St. Louis, Missouri 63131, USA and Division of Bone and Mineral Diseases, Washington University School of Medicine at Barnes-Jewish Hospital, St. Louis, Missouri 63110, USA. Thomas Willnow (153)

Division of Molecular Cardiovascular Research, Max-Delbrueck-Center for Molecular Medicine, Robert-Roessle-Strasse 10, D-13125 Berlin, Germany Didier Wion (1779)

INSERM U318, Centre Hospitalier Michallon, 38043 Grenoble cedex 09, France Hisataka Yasuda (665)

Center for Experimental Medicine, Institute of Medical Science, University of Tokyo, Tokyo 108-8639, Japan Daniel Young (327)

Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Chi Zhang (291)

Department of Pharmacology–W334, Case Western Reserve University, Cleveland, Ohio 44106, USA

Preface to the 2nd Edition

Those interested in the vitamin D field will not be surprised that this second edition is considerably larger than the first edition. A great deal of progress has been made since the first edition was published in 1997. However, our goal in planning this updated version remains the same. We have endeavored to provide investigators, clinicians, and students with a comprehensive, definitive, and up-to-date compendium of the diverse scientific and clinical aspects of vitamin D, each area covered by experts in the field. Our hope for the second edition is that this book will continue to serve as both a resource for current researchers, as well as a guide to assist those in related disciplines to enter the realm of vitamin D research. We hope that this book will illuminate the vitamin D field and help investigators identify areas where new research is needed as well as educate them about what is currently known. We believe that the first edition succeeded in stimulating interactions between researchers and clinicians from different disciplines and that it facilitated collaborations. As we move from basic science and physiology to the use of vitamin D and its analogs as pharmacological agents to treat various diseases, the need for cross-collaborations between researchers and clinicians from different disciplines will increase. We hope that this new volume will continue to be a valuable resource that plays a role in this advancement and stimulates and facilitates these interactions. Enormous progress in the study of vitamin D has been made in the approximately eight years since the first edition was written and we hope that this book has contributed in some way to this progress. The first edition proved to be highly valuable to its readers and the chapters have been cited frequently as authoritative reviews of the field. However, it became clear to us that the time was ripe to organize a second edition. Building on the original, we hope this second edition

will incorporate all of the progress made in the field since the first edition was published so that our objective of an up-to-date compendium containing everything you wanted to know about vitamin D will continue. The second edition is essentially a new and reinvigorated book. We have changed the symbol on the cover to reflect its updated content and the field's continued evolution into the molecular world. This new edition now includes 104 chapters. In order to cover the growth of new information on vitamin D, we have had to publish this new edition in two volumes. In addition, the book has undergone some major remodeling. There are 33 completely new chapters and 18 other chapters have had major changes in authorship and are totally rewritten. While approximately half of the chapters have some of the same authors, all have had major updates and many have new co-authors with new perspectives. We have endeavored to attract the leading investigators in each field to author the chapter covering their area. We are especially pleased with the roster of authors who have written for the second edition. They really are the leaders in their respective fields. We wish to express our thanks to Tari Paschall, Judy Meyer, and Sarah Hajduk as well as the rest of the Elsevier/Academic Press staff for their expertise and indispensable contribution to bringing this revised edition to fruition. Most of all, we thank the authors for their contributions. We hope that our readers will find this updated volume useful and informative and that it will contribute to the burgeoning growth of the vitamin D field. DAVID FELDMAN J. WESLEY PIKE FRANCIS H. GLORIEUX

Preface to the 1st Edition

Our reasons for deciding to publish an entire book devoted to vitamin D can be found in the rapid and extensive advances currently being made in this important field of research. Enormous progress in investigating many aspects of vitamin D, from basic science to clinical medicine, has been made in recent years. The ever-widening scope of vitamin D research has created new areas of inquiry so that even workers immersed in the field are not fully aware of the entire spectrum of current investigation. Our goal in planning this book was to bring the diverse scientific and clinical fields together in one definitive and up-to-date volume. It is our hope that this compendium on vitamin D will serve as both a resource for current researchers and a guide to stimulate and assist those in related disciplines to enter this field of research. In addition, we hope to provide clinicians and students with a comprehensive source of information for the varied and extensive material related to vitamin D. The explosion of information in the vitamin D sphere has led to new insights into many different areas, and in our treatment of each subject in this book we have tried to emphasize the recent advances as well as the established concepts. The classic view of vitamin D action, as a hormone limited to calcium metabolism and bone homeostasis, has undergone extensive revision and amplification in the past few years. We now know that the vitamin D receptor (VDR) is present in most tissues of the body and that vitamin D actions, in addition to the classic ones, include important effects on an extensive array of other target organs. To cover this large number of subjects, we have organized the book in the following manner: Section I, the enzymes involved in vitamin D metabolism and the activities of the various metabolites; Section II, the mechanism of action of vitamin D, including rapid, nongenomic actions and the role of the VDR in health and disease; Section III, the

effects of vitamin D and its metabolites on the various elements that constitute bone and the expanded understanding of vitamin D actions in multiple target organs, both classic and nonclassic; Section IV, the role of vitamin D in the physiology and regulation of calcium and phosphate metabolism and the multiplicity of hormonal, environmental, and other factors influencing vitamin D metabolism and action; and Sections V and VI, the role of vitamin D in the etiology and treatment of rickets, osteomalacia, and osteoporosis and the pathophysiological basis, diagnosis, and management of numerous clinical disorders involving vitamin D. The recent recognition of an expanded scope of vitamin D action and the new investigational approaches it has generated were part of the impetus for developing this volume on vitamin D. It has become clear that in addition to the classic vitamin D actions, a new spectrum of vitamin D activities that include important effects on cellular proliferation, differentiation, and the immune system has been identified. This new information has greatly expanded our understanding of the breadth of vitamin D action and has opened for investigation a large number of new avenues of research that are covered in Sections VII and VIII of this volume. Furthermore, these recently recognized nonclassic actions have led to a consideration of the potential application of vitamin D therapy to a range of diseases not previously envisioned. This therapeutic potential has spawned the search for vitamin D analogs that might have a more favorable therapeutic profile, one that is less active in causing hypercalcemia and hypercalciuria while more active in a desired application such as inducing antiproliferation, prodifferentiation, or immunosuppresion. Since 1α,25-dihydroxyvitamin D [1,25(OH)2D] and its analogs are all presumed to act via a single VDR, a few years ago most of us probably would have thought that it was impossible to achieve a

xxiv separation of these activities. Yet today, many analogs that exhibit different profiles of activity relative to 1,25(OH)2D have already been produced and extensively studied. The development of analogs with an improved therapeutic index has opened another large and complex area of vitamin D research. This work currently encompasses three domains: (1) the design and synthesis of vitamin D analogs exhibiting a separation of actions with less hypercalcemic and more antiproliferative or immunosuppressive activity, (2) the interesting biological question of the mechanism(s) by which these analogs achieve their differential activity, and (3) the investigation of the potential therapeutic applications of these analogs to treat various disease states. These new therapeutic applications, from psoriasis to cancer, from immunosuppression to neurodegenerative diseases, have drawn into the field an expanded population of scientists and physicians interested in vitamin D. Our goal in editing this book was to create a comprehensive resource on vitamin D that would be of use to a mix of researchers in different disciplines. To achieve this goal, we sought authors who had contributed greatly to their respective fields of vitamin D research. The book has a large number of chapters to accommodate

PREFACE

TO THE

1ST EDITION

many contributors and provide expertise in multiple areas. Introductory chapters in each section of the book are designed to furnish an overview of that area of vitamin D research, with other chapters devoted to a narrowly focused subject. Adjacent and closely related subjects are often covered in separate chapters by other authors. While this intensive style may occasionally create some redundancy, it has the advantage of allowing the reader to view the diverse perspectives of the different authors working in overlapping fields. In this regard, we have endeavored to provide many crossreferences to guide the reader to related information in different chapters. We express our thanks to Jasna Markovac (Editor-inChief), for encouraging us to develop this book and guiding us through the process; to Tari Paschall (Acquisitions Editor) and the Academic Press staff for their diligence, expertise, and patience in helping us complete this work. Most of all, we thank the authors for their contributions that have made this book possible. DAVID FELDMAN J. WESLEY PIKE FRANCIS H. GLORIEUX

Abbreviations

AA AC ACE ACF ACTH ADH ADHR ADP AHO AI AIDS Aj.AR ALP ANG II ANP APC APD AR ARC ATP ATRA AUC Bmax BARE bFGF BFU BGP BLM BMC BMD BMI BMP BMU bp BPH

arachadonic acid adenyl cyclase angiotensin converting enzyme activation frequency adrenocorticotropin antidiuretic hormone (vasopressin) autosomal dominant hypophosphatemic rickets adenosine diphosphate Albright’s hereditary osteodystrophy adequate intake acquired immunodeficiency syndrome adjusted apposition rate alkaline phosphatase angiotensin II atrial natriuretic peptide antigen presenting cell aminohydroxypropylidene bisphosphonate androgen receptor activator recruited cofactor adenosine triphosphate all-trans-retinoic acid area under the curve maximum number of binding sites bile acid response element basic fibroblast growth factor burst-forming unit bone Gla protein (osteocalcin) basal lateral membrane bone mineral content bone mineral density body mass index bone morphogenetic protein basic multicellular unit base pairs benign prostatic hyperplasia

BSA BUA [Ca2+]i

bovine serum albumin bone ultrasound attentuation internal calcium ion molar concentration CaBP calcium binding protein CAD coronary artery disease CaM calmodulin cAMP cyclic AMP CaSR or CaR calcium sensing receptor CAT chloramphenicol acetyltransferase CBG corticosteroid-binding globulin CBP competitive protein binding assay CC chief complaint CDCA chenodeoxycholic acid CDK or Cdk cyclin-dependent kinase cDNA complementary DNA CDP collagenase-digestible protein Cdx-2 caudal-related homeodomain protein CFU colony-forming unit cGMP cyclic GMP CGRP calcitonin gene-related peptide CHF congestive heart failure CK-II casein kinase-II CLIA competitive chemiluminescence immunoassay cM centimorgans Cm. Ln. cement line CNS central nervous system CPBA competitive protein binding assays cpm counts per minute CRE cAMP response element CREB cAMP response element binding protein CRF chronic renal failure CsA cyclosporin A CSF colony-stimulating factor

xxvi CT CTR CTX CVC CYP CYP24 DAG DBD DBP DBP DC DCA DCT DEXA or DXA 7-DHC DHEA DHT DIC DMSO DR DRIP DSP E1 E2 EAE EBT EBV EC EC50 or ED50 ECaC ECF EDTA EGF ELISA EMSA EP1 ER ERE ERK Et FACS FAD FCS FDA FFA FIT

ABBREVIATIONS

calcitonin or computerized tomography calcitonin receptor cerebrotendinous xanthomatosis calcifying vascular cells cytochrome P450 cytochrome P450, 24-hydroxylase diacylglycerol DNA binding domain diastolic blood pressure vitamin D binding protein dendritic cell deoxycholic acid distal convoluted tubule dual energy X-ray absorptiometry 7-dehydrocholesterol dehydroepiandrosterone dihydrotachysterol or dihydrotestosterone disseminated intravascular coagulation dimethyl sulfoxide direct repeat vitamin D receptor interacting protein dental sialoprotein estrone estradiol experimental autoimmune encephalitis electron-beam computed technology Epstein-Barr virus endothelial cells effective concentration (dose) to cause 50% effect epithelium calcium channel extracellular fluid ethylenediaminetetraacetic acid epidermal growth factor enzyme-linked immunosorbent assay electrophoretic mobility shift assay PG receptor-1 estrogen receptor or endoplasmic reticulum estrogen response element extracellular signal-regulated kinase endothelin fluorescence-activated cell sorting or sorter flavin adenine dinucleotide fetal calf serum U.S. Food and Drug Administration free fatty acid Fracture Intervention Trial

FMTC FP FRAP FS FSK FXR g g G0, G1, G2 GAG GC-MS G-CSF GDNF GFP GFR GH GHRH GIO GM-CSF GnRH GR GRE GRTH HAT HDAC HEK HHRH HIV HNF HPI HPLC HPV hr HRE HSA Hsp HSV HVDRR HVO IBMX IC50 ICA ICMA IDBP IDDM IDM IFN

familial medullary thyroid carcinoma formation period fluorescence recovery after photobleaching Fanconi syndrome forskolin farnesol X receptor gram acceleration due to gravity gap phases of the cell cycle glycosaminoglycan gas chromatography-mass spectrometry granulocyte colony-stimulating factor glial cell-derived neurotrophic factor green fluorescent protein glomerular filtration rate growth hormone growth hormone-releasing hormone glucocorticoid induced osteoporosis granulocyte-macrophage colonystimulating factor gonadotropin-releasing hormone glucocorticoid receptor glucocorticoid response element generalized resistance to thyroid hormone histone acetyltransferase histone deacetylase human embryonic kidney hereditary hypophosphatemic rickets with hypercalciuria human immunodeficiency virus hepatocyte nuclear factor history of present illness high-performance liquid chromatography human papilloma virus hour hormone response element human serum albumin heat-shock protein herpes simplex virus hereditary vitamin D-resistant rickets hypovitaminosis D osteopathy isobutylmethylxanthine concentration to inhibit 50% effect intestinal calcium absorption immunochemiluminometric assay intracellular vitamin D–binding protein insulin-dependent diabetes mellitus infants of diabetic mothers interferon

xxvii

ABBREVIATIONS

Ig IGFBP IGF-I, -II IGF-IR IL i.m. IMCal i.p. IP3 IRMA IU IUPAC i.v. JG JNK Kd Km kb kbp kDa KO LBD LCA LDL Li. Ce. LIF LNH LOD LPS LT LXR M M MAPK Mab MAR MAR MARRS MCR M-CSF MEN2 MGP MHC min MLR Mlt MR MRI mRNA MTC NADH

immunoglobulin IGF binding protein insulin-like growth factor type I, II IGF-I receptor interleukin (e.g., IL-1, IL-1β, etc.) intramuscular intestinal membrane calcium binding complex intraperitoneal inositol trisphosphate immunoradiometric assay international units International Union of Pure and Applied Chemists intravenous juxtaglomerular c-Jun NH2-terminal kinase dissociation constant Michaelis constant kilobases kilobase pairs kilodaltons knock out ligand binding domain lithocholic acid low-density lipoprotein lining cell leukemia inhibitory factor late neonatal hypocalcemia logarithm of the odds lipopolysaccharide leukotriene liver X receptor mitosis phase of cell cycle molar mitogen-activated protein kinase monoclonal antibody matrix attachment region mineral apposition rate membrane-associated rapid response steroid metabolic clearance rate macrophage colony-stimulating factor multiple endocrine neoplasia type 2 matrix Gla protein major histocompatibility complex minute mixed lymphocyte reaction mineralization lag time mineralcorticoid receptor magnetic resonance imaging messenger ribonucleic acid medullary thyroid carcinoma nicotinamide adenine dinucleotide

NADPH

nicotinamide adenine dinucleotide phosphate NAF nuclear accessory factor NBT nitroblue tetrazolium NcAMP nephrogenous cAMP NCP noncollagen protein NGF nerve growth factor NHANES III National Health and Nutrition Examination Survey III NHL Non-Hodgkin’s lymphoma NIDDM non-insulin-dependent diabetes mellitus NIH National Institutes of Health NK cell natural killer cell NLS nuclear localization signal NMR nuclear magnetic resonance NPT sodium/phosphate cotransporter NR nuclear receptor Ob osteoblast Oc osteocalcin or osteoclast OCIF osteoclastogenesis inhibitory factor (same as OPG) OCT 22-oxacalcitriol ODF osteoclast differentiation factor (same as RANKL) 1α-OHD3 1α-hydroxyvitamin D3 25OHD3 25-hydroxyvitamin D3 1,25(OH)2D3 1α,25-dihydroxyvitamin D3 24,25(OH)2D3 24,25-dihydroxyvitamin D3 OHO oncogenic hypophosphatemic osteomalacia Omt osteoid maturation time OPG osteoprotegerin OPN osteopontin OSM oncostatin M OVX ovariectomy Pi inorganic phosphate PA2 phospholipase A2 PAD peripheral arterial vascular disease PAM pulmonary alveolar macrophage PBL peripheral blood lymphocyte PBS phosphate-buffered saline PC phophatidyl choline PCNA proliferating cell nuclear antigen PCR polymerase chain reaction PCT proximal convoluted tubule PDDR pseudovitamin D deficiency rickets PDGF platelet-derived growth factor PEIT percutaneous ethanol injection therapy PHEX phosphate regulating gene with homologies to endopeptidases on the X chromosome PG prostaglandin

xxviii PHA PHP PIC PKA PKC PKI PLA2 PLC PMA PMCA PMH p.o. poly(A) PPAR PR PRA PRL PSA PSI PT PTH PTHrP PTX PUVA QCT QSAR 9-cis-RA RA RA RANK RANKL RAP RAR RARE RAS RBP RCI RDA RFLP RIA RID RNase ROCs ROS RPA RRA RT-PCR RXR

ABBREVIATIONS

phytohemagglutinin pseudohypoparathyroidism preinitiation complex protein kinase A protein kinase C protein kinase inhibitor phospholipase A2 phospholipase C phorbol 12-myristate 13-acetate plasma membrane calcium pump past medical history oral polyadenosine peroxisome proliferator-activated receptor progesterone receptor plasma renin activity prolactin prostate-specific antigen psoriasis severity index parathyroid parathyroid hormone parathyroid hormone-related peptide parathyroidectomy psoralen-ultraviolet A quantitative computerized tomography quantitative structure-activity relationship 9-cis-retinoic acid retinoic acid rheumatoid arthritis receptor activator NF-κB receptor activator NF-κB ligand receptor-associated protein retinoic acid receptor retinoic acid response element renin–angiotensin system retinol-binding protein relative competitive index recommended dietary allowance restriction fragment length polymorphism radioimmunoassay receptor interacting domain ribonuclease receptor operated calcium channels reactive oxygen species ribonuclease protection assay radioreceptor assay reverse transcriptase-polymerase chain reaction retinoid X receptor

RXRE SBP SD SDS SE SEM SH SHBG SLE SOS Sp1 SPF SRC-1 SSCP SV40 SXA t1/2 T3 T4 TBG TBP TC TF TFIIB TG TGF TIO TK TLR TmP or TmPi TNF TPA TPN TPTX TR TRAP TRAP TRP TRE TRE TRH Trk TSH TSS UF US USDA UTR UV VDDR-I

retinoid X receptor response element systolic blood pressure standard deviation sodium dodecyl sulfate standard error standard error of the mean social history sex hormone–binding globulin systematic lupus erythematosus speed of sound selective promoter factor 1 sun protection factor steroid receptor coactivator-1 single strand conformational polymorphism simian virus 40 single energy X-ray absorptiometry half-time triiodothyronine thyroxine thyroid-binding globulin TATA binding protein tumoral calcinosis tubular fluid general transcription factor IIB transgenic transforming growth factor tumor-induced osteomalacia thymidine kinase toll-like receptor tubular absorptive maximum for phosphorus tumor necrosis factor 12-O-tetradecanoylphorbol-13-acetate total parenteral nutrition thyroparathyroidectomized thyroid hormone receptor tartrate-resistant acid phosphatase thyroid hormone receptor associated proteins transient receptor potential thyroid hormone response element TPA response element thyrotropin-releasing hormone tyrosine kinase thyrotropin transcription start site ultrafiltrable fluid ultrasonography U.S. Department of Agriculture untranslated region ultraviolet vitamin D-dependent rickets type I (see PDDR)

xxix

ABBREVIATIONS

VDDR-II VDR VDRE VDRL VEGF VERT VICCs VSMC

vitamin D-dependent rickets type II (see HVDRR) vitamin D receptor vitamin D response element vitamin D receptor ligand vascular endothelial growth factor Vertebral Efficacy with Risedronate Therapy studies voltage-insensitive calcium channels vascular smooth muscle cell

VSSCs WHI WRE WSTF XLH XRD ZEB

voltage-senstive calcium channels Women’s Health Initiative Wilms’ tumor gene, WT1, responsive element Williams syndrome transcription factor X-linked hypophosphatemic rickets X-ray diffraction zinc finger, E box-binding transcription factor

Approximate Normal Ranges for Serum Values in Adultsa Measure

SI Units

Ionized calcium Total calcium Phosphorous, inorganic 25(OH)D 1,25(OH)2D

1.12–1.32 mmol/L 2.17–2.52 mmol/L 0.77–1.49 mol/L 24.9–169.5 nmol/L 60–108 pmol/L

Conventional Units

Conversion Factorb

4.5–5.3 mg/dL 8.7–10.1 mg/dL 2.4–4.6 mg/dL 10–68 ng/mL 25–45 pg/mL

0.2495 0.2495 0.3229 2.496 2.40

Approximate Normal Ranges for Serum Values in Childrena Measure Ionized calcium Total calcium Phosphorous, inorganic 25(OH)D 1,25(OH)2D aNormal

SI Units

Conventional Units

Conversion Factorb

1.19–1.29 mmol/L 2.25–2.63 mmol/L 1.23–1.62 mol/L 34–91 nmol/L 65–134 pmol/L

4.8–5.2 mg/dL 9.0–10.5 mg/dL 3.8–5.0 mg/dL 14–37 ng/mL 27–56 pg/mL

0.2495 0.2495 0.3229 2.496 2.40

ranges differ in various laboratories and these values are provided only as a general guide. factor X conventional units = SI units

bConversion

Useful Equivalencies of Different Units Vitamin D Calcium Phosphorus

1 µg = 40 IU 1 mmol = 40 mg 1 mmol = 30 mg

CHAPTER 1

Historical Perspective HECTOR F. DELUCA Department of Biochemistry, University of Wisconsin-Madison, Madison, Wisconsin

I. Discovery of the Vitamins II. Discovery That Vitamin D Is Not a Vitamin III. Isolation and Identification of Nutritional Forms of Vitamin D

IV. Discovery of the Physiological Functions of Vitamin D V. Discovery of the Hormonal Form of Vitamin D References

I. DISCOVERY OF THE VITAMINS

animals whereas purified materials could not [3]. Funk had found similar results for the prevention of neuritis and reasoned that there were “vital amines” present in foods from natural sources and actually provided the basis for the term “vitamins” used later to describe essential micronutrients [5].

A. Early Nutritional Views The field of nutrition was largely dominated in the nineteenth century by German chemists, led by Justus von Liebig [1]. They taught that adequacy of the diet could be described by an analysis of protein, carbohydrate, fat, and mineral. Thus, a diet containing 12% protein, 5% mineral, 10–30% fat, and the remainder as carbohydrate would be expected to support normal growth and reproduction. This view remained largely unchallenged until the very end of the 19th century and the beginning of the 20th century [2–5]. However, evidence opposing this view began to appear. One of the first was the famous study of Eijkman who studied prisoners in the Dutch East Indies maintained on a diet of polished rice [6]. A high incidence of the neurological disorder beri-beri was recorded in these inmates. Eijkman found that either feeding whole rice or returning the hulls of the polished rice could eliminate beri-beri. Eijkman reasoned that polished rice contained a toxin that was somehow neutralized by the rice hulls. Later, a colleague, Grijns [7], revisited the question and correctly demonstrated that hulls contained an important and required nutrient that prevented beri-beri. Other reports revealed that microorganic nutrients might be present. The development of scurvy in mariners was a common problem. This disease was prevented by the consumption of limes on British ships (hence, the term “Limey” to describe British sailors) and sauerkraut and fruits on other ships. This led Hoist and Frohlich to conclude that scurvy could be prevented by a nutrient present in these foods [8]. Experiments by Lunin, Magendie, Hopkins, and Funk showed that a diet of purified carbohydrate, protein, fat, and salt is unable to support growth and life of experimental animals [2–5]. This suggested that some unknown or vital factor present in natural foods was missing from the purified diets. Hopkins developed a growth test in which natural foods were found to support rapid growth of experimental VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX,

B. McCollum and Osborne and Mendel’s Discovery of Vitamin A and B Complex A key experiment demonstrating essential micronutrients was one carried out at the Wisconsin Agricultural Experiment Station, engineered by Stephen Moulton Babcock and carried out by E. B. Hart supported by McCollum and Steenbock [9]. Herds of dairy cows were maintained on a diet composed individually only of corn, oats, or wheat or were fed a mixture of all of these grains, all receiving the same amount of carbohydrate, protein, fat, and salts and all providing equal analysis according to the German chemists [1]. The animals on the corn diet did very well, produced milk in large amounts, and reproduced normally. Those on the wheat diet failed to thrive and soon were unable to reproduce or lactate. The oat group was found to be intermediate between the corn and wheat groups, and the mixture approximated the growth and reproduction found with corn. Yet all these diets had the same proximate analysis. The conclusion of the Wisconsin Experiment Station study was that there are unknown nutrients present in corn and not found in wheat that are essential for life and reproduction. This led E. B. Hart, Chairman of Biochemistry at Wisconsin, to conceive that a search for these nutrients must begin. Professor McCollum was asked to search for these nutrients using small experimental animals. McCollum and Davis demonstrated there was present in butter fat a substance that prevented xerophthalmia and was also required for growth. They termed this “a lipin-soluble growth factor” [10]. McCollum later named this factor “vitamin A” [11]. Copyright © 2005, Elsevier, Inc. All rights reserved.

4 This substance was absent from lard and other fats but was found in large amounts in cod liver oil. In constructing the diets, McCollum obtained the carbohydrates and salts from milk whey, which, unknown to him, supplied the vitamin B complex group of micronutrients that permitted him to observe a vitamin A deficiency. McCollum at Wisconsin [11] and Osborne and Mendel [12] at the Connecticut Experiment Station carried out experiments in which cod liver oil was used as a source of fat in the diet, but the minerals were supplied from pure chemicals mixed to approximate the mineral composition of milk. Starch or sugar was used as the carbohydrate. These animals developed a different group of symptoms, namely, neuritis, which could be cured by the provision of the milk components. McCollum and Osborne and Mendel correctly concluded that this activity was due to a different micronutrient called “vitamin B.” This ushered in the concept of the organic micronutrients known as vitamins.

C. History of Rickets The disease rickets was very likely known in antiquity but was described in the 15th century as revealed by later writings. Whistler first provided a clear description of rickets in which the skeleton was poorly mineralized and deformed [13]. Rickets undoubtedly in ancient times appeared only on rare occasions and hence was not considered a problem. However, at the end of the 19th century, the industrial revolution had taken place: A highly agrarian population had become urbanized, and smoke from the industrial plants polluted the atmosphere. Thus, in low-sunlight countries such as England, rickets appeared in epidemic proportions. In fact, it was known as “the English disease” [14]. Some reports of the beneficial action of cod liver oil had appeared. However, they were not given scientific credence. With the discovery of the vitamins, Sir Edward Mellanby in Great Britain began to reason that rickets might also be a disease caused by a dietary deficiency [15]. Mellanby fed dogs a diet composed primarily of oatmeal, which was the diet consumed where the incidence of rickets was the highest (i.e., Scotland). McCollum inadvertently maintained the dogs on oatmeal indoors and away from ultraviolet light. The dogs developed severe rickets. Learning from the experiments of McCollum, Mellanby provided cod liver oil to cure or prevent the disease. Mellanby could not decide whether the healing of rickets was due to vitamin A known to be present in the cod liver oil or whether it was a new and unknown substance. Therefore, the activity of healing rickets was first attributed to vitamin A.

HECTOR F. DELUCA

D. Discovery of Vitamin D McCollum, who had moved to Johns Hopkins from Wisconsin, continued his experiments on the fat-soluble materials. McCollum used aeration and heating of cod liver oil to destroy the vitamin A activity or the ability to support growth and to prevent xerophthalmia [16]. However, cod liver oil treated in this manner still retained the ability to cure rickets. McCollum correctly reasoned that the activity in healing rickets was due to a new and heretofore unknown vitamin that he termed “vitamin D.” On the basis of the experiments of McCollum and of Mellanby, vitamin D became known as an essential nutrient.

II. DISCOVERY THAT VITAMIN D IS NOT A VITAMIN At the same time that Sir Edward Mellanby was carrying out the experiments in dogs, Huldshinsky [17] and Chick et al. [18] independently found that rickets in children could be prevented or cured by exposing them to sunlight or to artificially induced ultraviolet light. Thus, the curious findings were that sunlight and ultraviolet light somehow equaled cod liver oil. These strange and divergent results required resolution. Steenbock and Hart had noted the importance of sunlight in restoring positive calcium balance in goats [19]. At Wisconsin, with McCollum carrying out experiments in small experimental animals (i.e., rats), Steenbock was required to work with larger animals. Steenbock then began to study goats because they would consume less materials and could serve as better experimental animals than cows. Steenbock began to study the calcium balance of lactating goats and found that those goats maintained outdoors in the sunlight were found to be in positive calcium balance, whereas those maintained indoors lost a great deal of their skeletal calcium to lactation [19]. Steenbock and Hart, therefore, noted the importance of sunlight on calcium balance. This work then undoubtedly led Steenbock to realize that the ultraviolet healing properties described by Huldschinsky might be related to the calcium balance experiments in goats. By irradiating the animals and diets, Steenbock and Black found that vitamin D activity could be induced and rickets could be cured [20]. A similar finding was reported soon thereafter by Hess and Weinstock [21]. Steenbock then traced this to the nonsaponifiable fraction of the lipids in foods [22]. He found that ultraviolet light activated an inactive substance to become a vitamin D–active material. Thus, ultraviolet light could be used to irradiate foods, induce vitamin D activity, and fortify foods to eliminate rickets as a major

5

CHAPTER 1 Historical Perspective

medical problem. This discovery also made available a source of vitamin D for isolation and identification.

III. ISOLATION AND IDENTIFICATION OF NUTRITIONAL FORMS OF VITAMIN D From irradiation of mixtures of plant sterols, Windaus and colleagues isolated a material that was active in healing rickets [23]. This substance was called “vitamin D1,” but its structure was not determined. Vitamin D1 proved to be an adduct of tachysterol and vitamin D2, and thus vitamin D1 was actually an error in identification. The British group led by Askew was successful in isolating and determining the structure of the first vitamin D, vitamin D2 or ergocalciferol, from irradiation of plant sterols [24]. A similar identification by the Windaus group confirmed the structure of vitamin D2 [25]. Windaus and Bock also isolated the precursor of vitamin D3 from skin, namely, 7-dehydrocholesterol [26]. Furthermore, 7-dehydrocholesterol was synthesized [27] and converted to vitamin D3 (cholecalciferol) as identified by the Windaus group [28]. Thus, the structures of nutritional forms of vitamin D became known (Fig. 1). Windaus and Bock, having isolated 7-dehydrocholesterol from skin, provided the presumptive evidence that vitamin D3 is the form of vitamin D produced in skin, a discovery that was later confirmed by the chemical identification of vitamin D3 in skin by Esvelt et al. [29] and of a previtamin D3 in skin by Holick et al. [30]. Synthetic vitamin D as produced by the irradiation process replaced the irradiation of foods as a means of fortifying foods with vitamin D and was also rapidly applied to a variety of diseases including rickets and tetany and in the provision to domestic animals such as chickens, cows, and pigs. Windaus’ group provided chemical syntheses of the vitamin D compounds, confirming their structures and

thus ending the era of the isolation and identification of nutritional forms of vitamin D and making them available for the treatment of disease. For his contributions, Windaus received the 1938 Nobel Prize in chemistry.

IV. DISCOVERY OF THE PHYSIOLOGICAL FUNCTIONS OF VITAMIN D A. Intestinal Calcium and Phosphorus Absorption Besides bone mineralization, the earliest discovered function of vitamin D is its important role in the absorption and utilization of calcium. The first report of this finding was in the early 1920s by Orr and colleagues [31]. Kletzien et al. [32] demonstrated that vitamin D plays an important role in the utilization of calcium from the diet, and a number of experiments were carried out on the utilization of calcium and phosphorus from cereal diets. Nicolaysen was responsible, however, for demonstrating unequivocally the role of vitamin D in the absorption of calcium and independently of phosphorus from the diet [33]. Nicolaysen also followed the early work of Kletzien et al. [32] in which animals adapted to a low calcium diet were better able to utilize calcium than were animals on an adequate calcium diet. This work was confirmed by Nicolaysen, who postulated the existence of an “endogenous factor” that would inform the intestine of the skeletal needs for calcium [34]. This endogenous factor later proved to be largely the active form of vitamin D, 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] [35]. Strong support for this concept was provided by the studies of Ribovich et al. [36] that showed animals maintained on a constant exogenous source of 1,25(OH)2D3 are unable to change intestinal calcium transport in response to changes in dietary calcium levels.

B. Mobilization of Calcium from Bone

FIGURE 1 Nutritional forms of vitamin D.

For many years, investigators have attempted to show that vitamin D plays a role directly on the mineralization process of the skeleton. However, early work by Rowland and Kramer [37], later work by Lamm and Neuman [38], and more recent work by Underwood and DeLuca [39] demonstrated very clearly that vitamin D does not play a significant role in the actual mineralization process of the skeleton but that the failure to mineralize the skeleton in vitamin D deficiency is due to inadequate levels of calcium and phosphorus in the plasma. Thus, the action of vitamin D in mineralizing

6 the skeleton and in preventing hypocalcemic tetany is the elevation of plasma calcium and phosphorus [40]. These discoveries laid to rest the concept of a role of vitamin D in mineralization. However, Carlsson [41] and Bauer et al. [42] were the first to realize that a major function of vitamin D is to induce the mobilization of calcium from bone when required. Thus, in animals on a low-calcium diet, the rise in serum calcium induced by vitamin D is the result of actual mobilization of calcium from bone [43]. This important function is known to be essential for the provision of calcium to meet soft-tissue needs, especially those of nerves and muscle, on a minute-to-minute basis when it is in insufficient supply from the diet. It is likely that the function of vitamin D in mobilizing calcium from bone is an osteoclastic-mediated process [44]. It is clear, however, that both vitamin D and parathyroid hormone are required for this function [45]. Furthermore, it is clear that vitamin D plays an important role in osteoclasticmediated bone resorption [46], which is certainly the first event in bone remodeling and an essential event in bone modeling [47].

C. Renal Reabsorption of Calcium and Phosphorus A final site of vitamin D action to elevate plasma calcium is in the distal renal tubule. Although experiments were suggestive of a role for vitamin D in increasing renal tubule reabsorption of calcium, a clear demonstration of this did not occur until the 1980s at the hands of Yamamoto et al. [48]. The renal tubule reabsorbs 99% of the filtered calcium even in the absence of vitamin D. However, reabsorption of the last 1% of the filtered load requires both vitamin D and parathyroid hormone. Thus, these agents work in concert in the renal reabsorption of calcium as well as in the mobilization of calcium from bone. Both agents are required to carry out this function.

D. Discovery of New Functions of Vitamin D With discovery of the receptor for the vitamin D hormone (described in Section V,G below) came the surprising result that this receptor could be found in a variety of tissues not previously appreciated as targets of vitamin D action. It localizes in the distal renal tubule cells, enterocytes of the small intestine, bone lining cells, and osteoblasts in keeping with its known role in calcium metabolism [49,50]. However, its appearance

HECTOR F. DELUCA

in tissues such as parathyroid gland, islet cells of the pancreas, cells in bone marrow (i.e., promyelocytes), lymphocytes, and certain neural cells raised the question of whether the functions of vitamin D might be broader than previously anticipated [49,50]. As a result of those findings, new functions of vitamin D have been found. For example, vitamin D plays a role in causing differentiation of promyelocytes to monocytes and the subsequent coalescing of the monocytes into multinuclear osteoclast precursors and ultimately into active osteoclasts [51,52]. Suppression of parathyroid cell growth and suppression of parathyroid hormone gene expression represent other new vitamin D actions [53,54]. In keratinocytes of skin, vitamin D appears to play a role in suppression of growth and in cellular differentiation [55]. Likely, discoveries of many new functions of 1,25(OH)2D3 will be made and are well on their way, as described in later chapters of this volume.

V. DISCOVERY OF THE HORMONAL FORM OF VITAMIN D A. Early Work of Kodicek The true pioneer of vitamin D metabolism was Egan MA Kodicek working at the Dunn Nutritional Laboratory in Cambridge U.K. Kodicek used a bioassay at first to study the fate of the vitamin D molecule and found that much vitamin D was converted to biologically inactive products [56]. Clearly, however, this approach of assaying vitamin D activity following administration of known doses of vitamin D was of limited value in determining metabolism.

B. Radiolabeled Vitamin D Experiments Professor Kodicek then began to synthesize radiolabeled vitamin D2. Unfortunately, the degree of labeling was not sufficient to permit the administration of truly physiological doses of vitamin D. Nevertheless, Professor Kodicek continued investigations into this important area. At the conclusion of 10 years of work, he concluded that vitamin D was active without metabolic modification and that the metabolites that were found were biologically inactive [57]. This conclusion was reached even as late as 1967, when it was concluded that vitamin D3 itself was the active form of vitamin D in the intestine [58]. However, chemical synthesis of vitamin D3 of high specific activity in the laboratory of the author proved to be of key importance in the demonstration of biologically active metabolites [59].

7

CHAPTER 1 Historical Perspective

By providing a truly physiological dose of vitamin D, it could be learned that the vitamin D itself disappeared and instead polar metabolites could be found in the target tissues before those tissues responded [60]. The polar metabolites proved to be more biologically active and acted more rapidly than vitamin D itself [61]. Thus, presumptive evidence of conversion of vitamin D to active forms had been obtained as early as 1967.

C. Isolation and Identification of the Active Form of Vitamin D By 1968, the first active metabolite of vitamin D was isolated in pure form and chemically identified as 25-hydroxyvitamin D3 (25OHD3) [62]. Its structure was confirmed by chemical synthesis [63] that provided it for study to the medical and scientific world. For a couple of years, 25OHD3, was visualized as the active form of vitamin D. However, when it was synthesized in radiolabeled form, it was found to be rapidly metabolized to yet more polar metabolites [64]. By this time, the Kodicek laboratory reawakened their interest in metabolism of vitamin D and began to study the metabolism of lα-tritium-labeled vitamin D [65]. Furthermore, polar metabolites of vitamin D were found by Haussler, Myrtle, and Norman [66]. The Wisconsin group labeled these metabolites as peak 5 [64], the Norman group called it peak 4B [66], and Lawson, Wilson, and Kodicek described it as peak P [65]. Kodicek et al. claimed that the metabolite of vitamin D found in intestine was deficient in tritium at the 1-position [65]. However, Myrtle et al. reported that peak 4B did not lose its tritium [67]. Thus, the suggestion of a modification at the 1-position could not be confirmed. The DeLuca group, however, isolated the active metabolite from intestines of 1600 chickens given radiolabeled vitamin D, and, by means of mass spectrometric techniques and specific chemical reactions, the structure of the active form of vitamin D in the intestine was unequivocally demonstrated as 1,25(OH)2D3 [68]. Of great importance was the finding of Fraser and Kodicek that the peak P metabolite could be produced by homogenates of chicken kidney and that anephric animals are unable to produce the peak P metabolite [69]. They correctly concluded that the site of synthesis of the active form of vitamin D is the kidney. The Wisconsin group then chemically synthesized both 1α25(OH)2D3 [70] and 1β,25(OH)2D3 [71] and provided unequivocal proof that the active form is 1α,25(OH)2D3. Furthermore, this group was able to synthesize lαOHD3, an important

analog that assumed great importance as a therapeutic agent throughout the world [72].

D. Proof That 1,25(OH)2D3 Is the Active Form of Vitamin D Proof that 1,25(OH)2D3 and not 25OHD3 is the active form was provided by experiments in which anephric animals respond to 1,25(OH)2D3 by increasing intestinal absorption of calcium and bone calcium mobilization, whereas animals receiving 25OHD3 at physiological doses did not [73–75]. Furthermore, the experiment of nature, namely, vitamin D–dependency rickets type I, an autosomal recessive disorder, provided final proof [76]. This disease could be corrected by physiological doses of synthetic 1,25(OH)2D3, whereas large amounts of vitamin D3 or 25OHD3 were needed to heal the rickets. The exact defect in this disease is now clearly known and is described elsewhere in this volume. 25OHD3 at pharmacological doses likely acts as an analog of the final vitamin D hormone, 1,25(OH)2D3 (Fig. 2).

E. Discovery of the Vitamin D Endocrine System Immediately after the identification of 1,25(OH)2D3 as the active form of vitamin D came studies in which it could be shown that animals on a low-calcium diet produce large quantities of 1,25(OH)2D3, whereas those on a high-calcium diet produce little or no 1,25(OH)2D3 [77]. A reciprocal arrangement was found for the metabolite 24R,25(OH)2D3. Thus, when calcium is needed, production of 1,25(OH)2D3 is markedly stimulated and the 24-hydroxylation degradation reaction is suppressed. When adequate calcium is present, production of 1,25(OH)2D3 is shut off and the 24-hydroxylation reaction is turned on. This discovery also satisfactorily provided evidence that 1,25(OH)2D3 is the likely endogenous factor originally described by Nicolaysen et al. [34]. The next important step was the demonstration that it is parathyroid hormone that activates 1α-hydroxylation of 25OHD3 in the kidney [78]. Thus, parathyroidectomy eliminates the hypocalcemic stimulation of 1α-hydroxylation and suppression of 24-hydroxylation, whereas administration of parathyroid hormone restores that capability. Fraser and Kodicek also provided evidence that, in intact chickens, injection of parathyroid hormone stimulated the 1α-hydroxylation reaction [79]. Thus, the basic vitamin D endocrine system was largely discovered and reported in the early 1970s, being completed by 1974.

8

HECTOR F. DELUCA

OH

Liver

Kidney

Microsomes (Mitochondria)

Mitochondria

HO

HO Vitamin D3

OH

HO 25-hydroxyvitamin D3

OH 1α,25-dihydroxyvitamin D3

FIGURE 2 Activation of the vitamin D molecule.

F. Other Metabolites of Vitamin D During the course of identification of 1,25(OH)2D3, 21,25(OH)2D3 was reported as a metabolite, as was 25,26(OH)2D3 [80,81]. However, the identification of 21,25(OH)2D3 was in error and was corrected to 24,25(OH)2D3, with the correct stereochemistry as 24R,25(OH)2D3 [82]. Over the late 1970s and early 1980s, as many as 30 metabolites of vitamin D were identified [83]. These are covered in other chapters in this volume. Of great importance was the use of the fluoro derivatives of vitamin D to illustrate that the only activation pathway of vitamin D is 25-hydroxylation followed by 1-hydroxylation [84]. Thus, 24-difluoro25OHD3 supported all known functions of vitamin D for at least two generations of animals [85]. 24-Difluoro25OHD3 cannot be 24-hydroxylated. Furthermore, other fluoro derivatives such as 26,27-hexafluoro-25OHD3 [86] and 23-difluoro-25OHD3 [87] are all fully biologically active, illustrating that 26-hydroxylation, 24-hydroxylation, and 23-hydroxylation are not essential to the function of vitamin D.

G. Discovery of the Vitamin D Receptor Zull and colleagues provided evidence that the function of vitamin D is blocked by transcription and protein inhibitors [88]. Thus, it became clear very early that a nuclear activity is required for vitamin D to carry out its functions. This work confirmed and extended the earlier work of Eisenstein and Passavoy [89]. With the discovery of the active forms of vitamin D came new attention to the idea that vitamin D may work through a nuclear mechanism. Thus, Haussler et al. reported vitamin D compounds to be associated with chromatin [66]. However, these experiments did not

exclude the possibility that the vitamin D compounds might be bound to the nuclear membrane. The first clear demonstration of the existence of a vitamin D receptor was at the hands of Brumbaugh and Haussler [90]. Furthermore, the experiments of Kream et al. [91] provided strong and unequivocal evidence of the existence of a nuclear receptor for 1,25(OH)2D3. Intense efforts toward purification of the receptor and its study appeared with the knowledge that it is a receptor protein with a molecular weight of approximately 55,000. In 1987, a partial cDNA sequence for the chicken vitamin D receptor was determined [92]. This was followed by isolation of the full coding sequence for the human [93] and rat [94,95] receptors. Cloning of the cDNAs encoding the vitamin D receptor in a human and mouse permitted the isolation of the gene encoding the vitamin D receptor [96–98]. The human gene was completely described and the mouse promoter was isolated and shown to be a TATA-less Sp1driven promoter [98]. The human gene appears to have alternate promoters [96,97]. Two groups have prepared receptor Null mutant mice, permitting extensive experiments with vitamin D receptorless animals [99,100]. From a historical point of view, one of the most important discoveries was vitamin D-dependency rickets type II [101], which is now known to be due to a defect in the receptor gene [102,103] (discussed in Chapter 11). This discovery essentially provided receptor knockout experiments in humans, allowing unequivocal proof of the essentiality of the vitamin D receptor for the functions of vitamin D. The nature of the receptor and how it functions are described in subsequent chapters along with current thinking on the molecular mechanism of action of 1,25(OH)2D3. However, the discovery of vitamin D responsive elements in the osteocalcin, osteopontin, preproparathyroid, and 24-OHase genes represent important

CHAPTER 1 Historical Perspective

historical developments [104–107]. This led to a consensus sequence and, most important, the development of the 3, 4, 5-rule of Umesono et al. [108]. It is now clear that vitamin D–responsive elements represent two imperfect repeat sequences separated by three nonspecified nucleotides. The vitamin D receptor will bind to these response elements, but it requires the presence of another nuclear factor, which proved to be the retinoid-X receptor (RXR) [109,110]. It is quite clear that the vitamin D receptor forms a heterodimer on the vitamin D responsive elements with the RXR protein on the 5′ arm of the responsive element and the vitamin D receptor on the 3′ segment [111]. The work of Rosenfeld, Glass, and colleagues [112] has demonstrated that the RXR protein when complexed with the vitamin D receptor on the responsive elements will not accept an RXR ligand, thus acting as a silent partner. Details of what is known concerning the role of the vitamin D receptor in transcription are described fully in subsequent chapters.

Acknowledgment This work was supported in part by a fund from the Wisconsin Alumni Research Foundation.

References 1. von Liebig J 1957 Animal Chemistry or Organic Chemistry in Its Application to Physiology and Pathology. In: Glass HB (ed) A History of Nutrition. The Riverside Press, Cambridge, MA. 2. Lunin N 1881 Über die Bedeutung der anorganischen Salze für die Ernährung des Tieres. Z Physiol Chem 5:31–39. 3. Hopkins G 1912 Feeding experiments illustrating the importance of accessory food factors in normal dietaries. J Physiol 44:425–460. 4. Magendie F 1816 Ann Chim Phys. 3:86–87. In: Glass HB (ed) 1957 A History of Nutrition by McCollum EV. The Riverside Press, Cambridge MA. 5. Funk C 1911 The chemical nature of the substance that cures polyneurities in birds produced by a diet of polished rice. J Physiol (London) 43:395–402. 6. Eijkman C 1897 Ein beri-beri Anliche der Hühner. Virchows Arch 148:523–527. 7. Grijns G 1957 Concerning polyneuritis geneeskundig gallinarum tijschrift voor ned indie 1901. In: McCollum EV, Glass HB (eds) A History of Nutrition. Houghton Mifflin, Boston, p. 216. 8. Hoist A. Frolich T 1907 Experimental studies relating ship-beriberi to scurvy. II. On the etiology of scurvy. J Hyg 7: 634–671. 9. Hart EB, McCollum EV, Steenbock H, Humphrey GC 1911 Physiological effect on growth and reproduction of rations balanced from restricted sources. Wis Agric Exp Sta Res Bull 17:131–205. 10. McCollum EV, Davis M 1913 The necessity of certain lipins in the diet during growth. J Biol Chem 25:167–175.

9 11. McCollum EV, Simmons N, Pitz W 1916 The relation of the unidentified dietary factors, the fat-soluble A and watersoluble B of the diet to the growth promoting properties of milk. J Biol Chem 27:33–38. 12. Osborne TB, Mendel LB 1917 The role of vitamins in the diet. J Biol Chem 31:149–163. 13. Whistler D 1645 De morbo puerli anglorum, quem patrio ideiomate indigenae vocant “the rickets” (Lugduni, Batavorum 1645). As cited by Smerdon GT,1950 Daniel Whistler and the English Disease. A translation and biographical note. J Hist Med 5:397–415. 14. Hess A 1929 The history of rickets. In: Rickets, Including Osteomalacia and Tetany. Lea & Febiger, Philadelphia, pp. 22–37. 15. Mellanby E 1919 An experimental investigation on rickets. Lancet 1:407–412. 16. McCollum EV, Simmonds N, Becker JE, Shipley PG 1922 An experimental demonstration of the existence of a vitamin which promotes calcium deposition. J Biol Chem 53:293–298. 17. Huldshinsky K 1919 Heilung von Rachitis durch Künstlich Hohen-sonne. Deut Med Wochenschr 45:712–713. 18. Chick H, Palzell EJ, Hume EM 1923 Studies of rickets in Vienna 1919–1922. Medical Research Council, Special Report No. 77. 19. Steenbock H, Hart EB 1913 The influence of function on the lime requirements of animals. J Biol Chem 14:59–73. 20. Steenbock H, Black A 1924 Fat-soluble vitamins. XVII. The induction of growth-promoting and calcifying properties in a ration by exposure to ultraviolet light. J Biol Chem 61:405–422. 21. Hess AF, Weinstock M 1924 Antirachitic properties imparted to lettuce and to growing wheat by ultraviolet irradiation. Proc Soc Exp Biol Med 22:5–6. 22. Steenbock H, Black A 1925 Fat-soluble vitamins. XXIII. The induction of growth-promoting and calcifying properties in fats and their unsaponifiable constituents by exposure to light. J Biol Chem 64:263–298. 23. Windaus A, Linsert O 1928 Vitamin D1 . Ann Chem 465:148. 24. Askew FA, Bourdillon RB, Bruce HM, Jenkins RGC, Webster TA 1931 The distillation of vitamin D. Proc R Soc 8107:76–90. 25. Windaus A, Linsert O, Luttringhaus A, Weidlich G 1932 Crystalline-vitamin D2. Ann Chem 492:226–241. 26. Windaus A, Bock F 1937 Über das Provitamin aus dem Sterin der Schweineschwarte. Z Physiol Chem 245:168–170. 27. Windaus A, Lettre H, Schenck F 1935 7-Dehydrocholesterol. Ann Chem 520:98–107. 28. Windaus A, Schenck F, von Werder F 1936 Über das antirachitisch wirksame Bestrahlungs-produkt aus 7-Dehydrocholesterin. Hoppe-Seylers Z Physiol Chem 241:100–103. 29. Esvelt RP, Schnoes HK, DeLuca HF 1978 Vitamin D3 from rat skins irradiated in vitro with ultraviolet light. Arch Biochem Biophys 188:282–286. 30. Holick MF, MacLaughlin JA, Clark MB, Holick SA, Potts JT Jr, Anderson RR, Blank IH, Parrish JA, Elias P 1980 Photosynthesis of previtamin D3 in human skin and the physiologic consequences. Science 210:203–205. 31. Orr WJ, Holt LE Jr, Wilkins L, Boone FH 1923 The calcium and phosphorus metabolism in rickets, with special reference to ultraviolet rat therapy. Am J Dis Children 26:362–372. 32. Kletzien SWF, Templin VM, Steenbock H, Thomas BH 1932 Vitamin D and the conservation of calcium in the adult. J Biol Chem 97:265–280. 33. Nicolaysen R 1937 Studies upon the mode of action of vitamin D. III. The influence of vitamin D on the absorption of calcium and phosphorus in the rat. Biochem J 31:122–129. 34. Nicolaysen R, Eeg-Larsen N, Malm OJ 1953 Physiology of calcium metabolism. Physiol Rev 33:424–444.

10 35. Boyle IT, Gray RW, Omdahl JL, DeLuca HF 1972 Calcium control of the in vivo biosynthesis of 1,25-dihydroxyvitamin D3: Nicolaysen’s endogenous factor. In: Taylor S (ed) Endocrinology 1971. Heinemann Medical Books, London, pp. 468–476. 36. Ribovich ML, DeLuca HF 1975 The influence of dietary calcium and phosphorus on intestinal calcium transport in rats given vitamin D metabolites. Arch Biochem Biophys 170:529–535. 37. Rowland J, Kramer B 1921 Calcium and phosphorus in the serum in relation to rickets. Am J Dis Children 22:105–119. 38. Lamm M, Neuman WF 1958 On the role of vitamin D in calcification. Arch Pathol 66:204–209. 39. Underwood JL, DeLuca HF 1984 Vitamin D is not directly necessary for bone growth and mineralization. Am J Physiol 246:E493–E498. 40. DeLuca HF 1967 Mechanism of action and metabolic fate of vitamin D. Vit Horm 25:315–367. 41. Carlsson A 1952 Tracer experiments on the effect of vitamin D on the skeletal metabolism of calcium and phosphorus. Acta Physiol Scand 26:212–220. 42. Bauer GCH, Carlsson A, Lindquist B 1955 Evaluation of accretion, resorption, and exchange reactions in the skeleton. Kungl Fysiograf Sallskapets I. Lund Forhandlingar 25:3–18. 43. Blunt JW, Tanaka Y, DeLuca HF 1968 The biological activity of 25-hydroxycholecalciferol, a metabolite of vitamin D3. Proc Natl Acad Sci USA 61:1503–1506. 44. Morony S, Capparelli C, Lee R, Shimamoto G, Boone T, Lacey DL, Dunstan CR 1999 A chimeric form of osteoprotegerin inhibits hypercalcemia and bone resorption induced by IL-1β, TNF-α, PTH, PTHrP, and 1,25(OH)2D3. J Bone Miner Res 14:1478–1485. 45. Garabedian M, Tanaka Y, Holick MF, DeLuca HF 1974 Response of intestinal calcium transport and bone calcium mobilization to 1,25-dihydroxyvitamin D3 in thyroparathyroidectomized rats. Endocrinology 94:1022–1027. 46. Raisz LG, Trummel CL, Holick MF, DeLuca HF 1972 1,25-Dihydroxycholecalciferol: A potent stimulator of bone resorption in tissue culture. Science 175:768–769. 47. Frost HM 1966 Bone Dynamics in Osteoporosis and Osteomalacia. Henry Ford Hospital Surgical Monograph Series. Thomas, Springfield, IL. 48. Yamamoto M, Kawanobe Y, Takahashi H, Shimazawa E, Kimura S, Ogata E 1984 Vitamin D deficiency and renal calcium transport in the rat. J Clin Invest 74:507–513. 49. Stumpf WE, Sar M, Reid FA, Tanaka Y, DeLuca HF 1979 Target cells for 1,25-dihydroxyvitamin D3 in intestinal tract, stomach, kidney, skin, pituitary and parathyroid. Science 206:1188–1190. 50. Stumpf WE, Sar M, DeLuca HF 1981 Sites of action of 1,25 (OH)2 vitamin D3 identified by thaw-mount autoradiography. In: Cohn DV, Talmage RV, Matthews JL (eds) Hormonal Control of Calcium Metabolism. Excerpta Medica, Amsterdam, pp. 222–229. 51. Suda T, Takahashi N, Martin TJ 1992 Modulation of osteoclast differentiation. Endocr Rev 13:66–80. 52. Suda T 1992 The role of 1α,25-dihydroxyvitamin D3 in the myeloid cell differentiation. Proc Soc Exp Biol Med 191:214–220. 53. Demay MB, Kiernan, MS, DeLuca HF, Kronenberg HM 1992 Sequences in the human parathyroid hormone gene that bind the 1,25-dihydroxyvitamin D3 receptor and mediate transcriptional repression in response to 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 89:8097–8101. 54. Silver J 1994 Regulation of parathyroid hormone production by 1α,25-(OH)2D3 and its analogues. Their therapeutic usefulness in secondary hyperparathyroidism. In: Vitamin D and

HECTOR F. DELUCA

55.

56.

57. 58. 59. 60. 61. 62. 63. 64. 65.

66. 67. 68.

69. 70. 71. 72. 73.

Its Analogues. The Second International Forum on Calcified Tissue and Bone Metabolism. Chugai Pharmaceutical, Tokyo, pp. 60–63. Smith EL, Walworth NC, Holick MF 1986 Effect of 1α,25-dihydroxyvitamin D3 on the morphologic and biochemical differentiation of cultured human epidermal keratinocytes grown in serum-free conditions. J Invest Dermatol 86:709–714. Kodicek E 1956 Metabolic studies on vitamin D. In: Wolstenholme GWE, O’Connor CM (eds) Ciba Foundation Symposium on Bone Structure and Metabolism. Little, Brown, and Co., Boston, pp. 161–174. Kodicek E 1960 The metabolism of vitamin D. In: Umbreit W, Molitor H (eds) Proceedings of the Fourth International Congress of Biochemistry. Pergamon, London, Vol. 11, pp. 198–208. Haussler MR, Norman AW 1967 The subcellular distribution of physiological doses of vitamin D3. Arch Biochem Biophys 118:145–153. Neville PF, DeLuca HF 1966 The synthesis of [1,2-3H]vitamin D3 and the tissue localization of a 0.25 µg (10 IU) dose per rat. Biochemistry 5:2201–2207. Lund J, DeLuca HF 1966 Biologically active metabolite of vitamin D3 from bone, liver, and blood serum. J Lipid Res 7:739–744. Morii H, Lund J, Neville PF, DeLuca HF 1967 Biological activity of a vitamin D metabolite. Arch Biochem Biophys 120:508–512. Blunt JW, DeLuca HF, Schnoes HK 1968 25-Hydroxycholecalciferol. A biologically active metabolite of vitamin D3. Biochemistry 7:3317–3322. Blunt JW, DeLuca HF 1969 The synthesis of 25-hydroxycholecalciferol. A biologically active metabolite of vitamin D3. Biochemistry 8:671–675. DeLuca HF 1970 Metabolism and function of vitamin D. In: DeLuca HF, Suttie JW (eds) The Fat-Soluble Vitamins. Univ. of Wisconsin Press, Madison, pp. 3–20. Lawson DEM, Wilson PW, Kodicek E 1969 Metabolism of vitamin D. A new cholecalciferol metabolite, involving loss of hydrogen at C-l, in chick intestinal nuclei. Biochem J 115:269–277. Haussler MR, Myrtle JF, Norman AW 1968 The association of a metabolite of vitamin D3 with intestinal mucosa chromatin in vivo. J Biol Chem 243:4055–4064. Myrtle JF, Haussler MR, Norman AW 1970 Evidence for the biologically active form of cholecalciferol in the intestine. J Biol Chem 245:1190–1196. Holick MF, Schnoes HK, DeLuca HF, Suda T, Cousins RJ 1971 Isolation and identification of 1,25–dihydroxycholecalciferol. A metabolite of vitamin D active in intestine. Biochemistry 10:2799–2804. Fraser DR, Kodicek E 1970 Unique biosynthesis by kidney of a biologically active vitamin D metabolite. Nature 228: 764–766. Semmler EJ, Holick MF, Schnoes HK, DeLuca HF 1972 The synthesis of 1α,25-dihydroxycholecalciferol—A metabolically active form of vitamin D3. Tetrahedron Lett 40:4147–4150. Paaren HE, Schnoes HK, DeLuca HF 1977 Synthesis of 1α,25-dihydroxyvitamin D3. J Chem Soc Chem Commun, 890–892. Holick MF, Semmler EJ, Schnoes HK, DeLuca HF 1973 1αHydroxy derivative of vitamin D3: A highly potent analog of 1α,25-dihydroxyvitamin D3. Science 180:190–191. Boyle IT, Miravet L, Gray RW, Holick MF, DeLuca HF 1972 The response of intestinal calcium transport to 25-hydroxy and 1,25-dihydroxyvitamin D in nephrectomized rats. Endocrinology 90:605–608.

11

CHAPTER 1 Historical Perspective

74. Holick MF, Garabedian M, DeLuca HF 1972 1,25Dihydroxycholecalciferol: Metabolite of vitamin D3 active on bone in anephric rats. Science 176:1146–1147. 75. Wong RG, Norman AW, Reddy CR, Coburn JW 1972 Biologic effects of 1,25-dihydroxycholecalciferol (a highly active vitamin D metabolite) in acutely uremic rats. J Clin Invest 51:1287–1291. 76. Fraser D, Kooh SW, Kind HP, Holick MF, Tanaka Y, DeLuca HF 1973 Pathogenesis of hereditary vitamin D dependent rickets: An inborn error of vitamin D metabolism involving defective conversion of 25-hydroxyvitamin D to 1α,25dihydroxyvitamin D. N Engl J Med 289:817–822. 77. Boyle IT, Gray RW, DeLuca HF 1971 Regulation by calcium of in vivo synthesis of 1,25-dihydroxycholecalciferol and 21,25-dihydroxycholecalciferol. Proc Natl Acad Sci USA 68:2131–2134. 78. Garabedian M, Holick MF, DeLuca HF, Boyle IT 1972 Control of 25-hydroxycholecalciferol metabolism by the parathyroid glands. Proc Natl Acad Sci USA 69:1673–1676. 79. Fraser DR, Kodicek E 1973 Regulation of 25-hydroxycholecalciferol-1-hydroxylase activity in kidney by parathyroid hormone. Nature (New Biol) 241:163–166. 80. Suda T, DeLuca HF, Schnoes HK, Ponchon G, Tanaka Y, Holick MF 1970 21,25-Dihydroxycholecalciferol. A metabolite of vitamin D3 preferentially active on bone. Biochemistry 9:2917–2922. 81. Suda T, DeLuca HF, Schnoes HK, Tanaka Y, Holick MF 1970 25,26-Dihydroxycholecalciferol, a metabolite of vitamin D3 with intestinal calcium transport activity. Biochemistry 9:4776–4780. 82. Tanaka Y, Frank H, DeLuca HF, Koizumi N, Ikekawa N 1975 Importance of the stereochemical position of the 24-hydroxyl to biological activity of 24-hydroxyvitamin D3. Biochemistry 14:3293–3296. 83. DeLuca HF, Schnoes HK 1983 Vitamin D: Recent advances. Annu Rev Biochem 52:411–439. 84. Brommage R, DeLuca HF 1985 Evidence that 1,25dihydroxyvitamin D3 is the physiologically active metabolite of vitamin D3. Endocr Rev 6:491–511. 85. Brommage R, Jarnagin K, DeLuca HF, Yamada S, Takayama H 1983 1- but not 24-hydroxylation of vitamin D is required for skeletal mineralization in rats. Am J Physiol 244: E298–E304. 86. Tanaka Y, Pahuja DN, Wichmann JK, DeLuca HF, Kobayashi Y, Taguchi T, Ikekawa N 1982 25-Hydroxy-26,26,26,27,27,27hexafluorovitamin D3: Biological activity in the rat. Arch Biochem Biophys 218:134–141. 87. Nakada M, Tanaka Y, DeLuca HF, Kobayashi Y, Ikekawa N 1985 Biological activities and binding properties of 23,23difluoro-25-hydroxyvitamin D3 and its lα-hydroxy derivative. Arch Biochem Biophys 241:173–178. 88. Zull JE, Czarnowska-Misztal E, DeLuca HF 1965 Actinomycin D inhibition of vitamin D action. Science 149:182–184. 89. Eisenstein R, Passavoy M 1964 Actinomycin D inhibits parathyroid hormone and vitamin D activity. Proc Soc Exp Biol Med 117:77–79. 90. Brumbaugh PF, Haussler MR 1973 lα,25-Dihydroxyvitamin D3 receptor: Competitive binding of vitamin D analogs. Life Sci 13:1737–1746. 91. Kream BE, Reynolds RD, Knutson JC, Eisman JA, DeLuca HF 1976 Intestinal cytosol binders of 1,25-dihydroxyvitamin D3 and 25-hydroxyvitamin D3. Arch Biochem Biophys 176: 779–787. 92. McDonnell DP, Mangelsdorf DJ, Pike JW, Haussler MR, O’Malley BW 1987 Molecular cloning of complementary

93.

94. 95. 96.

97. 98. 99.

100.

101.

102.

103.

104.

105.

106.

107.

DNA encoding the avian receptor for vitamin D. Science 235:1214–1217. Baker AR, McDonnell DP, Hughes M, Crisp TM, Mangelsdorf DJ, Haussler MR, Pike JW, Shine J, O’Malley BW 1988 Cloning and expression of full-length cDNA encoding human vitamin D receptor. Proc Natl Acad Sci USA 85: 3294–3298. Burmester JK, Maeda N, DeLuca HF 1988 Isolation and expression of rat 1,25-dihydroxyvitamin D3 receptor cDNA. Proc Natl Acad Sci USA 85:1005–1009. Burmester JK, Wiese RJ, Maeda N, DeLuca HF 1988 Structure and regulation of the rat 1,25-dihydroxyvitamin D3 receptor. Proc Natl Acad Sci USA 85:9499–9502. Miyamoto K, Kesterson RA, Yamamoto H, Nishiwaki E, Tatsumi S, Taketani Y, Morita K, Pike JW, Takeda E 1997 Structural organization of the human vitamin D receptor chromosomal gene and its promoter. Mol Endocrinol 11:1165–1179. Jehan F, DeLuca HF 1997 Cloning and characterization of the mouse vitamin D receptor promoter. Proc Natl Acad Sci USA 94:10138–10143. Jehan F, DeLuca HF 2000 The mouse vitamin D receptor is mainly expressed through an Sp1-driven promoter in vivo. Arch Biochem Biophys 377:273–283. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Genet 16:391–396. Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997 Targeted ablation of the vitamin D receptor: An animal model of vitamin D-dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94:9831–9835. Brooks MH, Bell NH, Love L, Stern PH, Orfei E, Queener SF, Hamstra AJ, DeLuca HF 1978 Vitamin D-dependent rickets Type II. Resistance of target organs to 1,25-dihydroxyvitamin D. N Engl J Med 298:996–999. Hughes MR, Malloy PJ, Kieback DG, Kesterson RA, Pike JW, Feldman D, O’Malley BW 1988 Point mutations in the human vitamin D receptor gene associated with hypocalcemic rickets. Science 242:1702–1706. Wiese RJ, Goto H, Prahl JM, Marx SJ, Thomas M, Al-Aqeel A, DeLuca HF 1993 Vitamin D–dependency rickets Type II: Truncated vitamin D receptor in three kindreds. Mol Cell Endocrinol 90:197–201. Demay MB, Gerardi JM, DeLuca HF, Kronenberg HM 1990 DNA sequences in the rat osteocalcin gene that bind the 1,25-dihydroxyvitamin D3 receptor and confer responsiveness to 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 87:369–373. Noda M, Vogel RL, Craig AM, Prahl J, DeLuca HF, Denhardt DT 1990 Identification of a DNA sequence responsible for binding of the 1,25-dihydroxyvitamin D3 receptor and 1,25-dihydroxyvitamin D3 enhancement of mouse secreted phosphoprotein 1 (Spp1, osteopontin) gene expression. Proc Natl Acad Sci USA 87:9995–9999. Demay MB, Kiernan MS, DeLuca HF, Kronenberg HM 1992 Sequences in the human parathyroid hormone gene that bind the 1,25-dihydroxyvitamin D3 receptor and mediate transcriptional repression in response to 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 89:8097–8101. Zierold C, Darwish HM, DeLuca HF 1995 Two vitamin D response elements function in the rat 1,25-dihydroxyvitamin D 24-hydroxylase promoter. J Biol Chem 270:1675–1678.

12 108. Umesono K, Murakami KK, Thompson CC, Evans RM 1991 Direct repeats as selective response elements for the thyroid hormone, retinoic acid, and vitamin D3 receptors. Cell 65:1255–1266. 109. Yu VC, Delsert C, Andersen B, Holloway JM, Devary OV, Naar AM, Kim SY, Boutin J-M, Glass CK, Rosenfeld MG 1991 RXRβ: A coregulator that enhances binding of retinoic acid, thyroid hormone, and vitamin D receptors to their cognate response elements. Cell 67:1251–1266. 110. Munder M, Herzberg IM, Zierold C, Moss VE, Hanson K, Clagett-Dame M, DeLuca HF 1995 Identification of the porcine intestinal accessory factor that enables DNA

HECTOR F. DELUCA

sequence recognition by vitamin D receptor. Proc Natl Acad Sci USA 93:2796–2799. 111. Jin CH, Pike JW 1996 Human vitamin D receptor-dependent transactivation in Saccharomyces cerevisiae requires retinoid X receptor. Mol Endocrinol 10:196–205. 112. DiRenzo J, Soderstrom M, Kurokawa R, Ogliastro M-H, Ricote M, Ingrey S, Horlein A, Rosenfeld MG, Glass CK 1997 Peroxisome proliferator-activated receptors and retinoic acid receptors differentially control the interactions of retinoid X receptor heterodimers with ligands, coactivators, and corepressors. Mol Cell Biol 17: 2166–2176.

CHAPTER 2

Vitamin D Metabolism RONALD L. HORST AND TIMOTHY A. REINHARDT U.S. Department of Agriculture, Agricultural Research Service, National Animal Disease Center, Ames, Iowa

G. SATYANARAYANA REDDY Brown University, Department of Chemistry, Providence, Rhode Island

I. Introduction II. Vitamin D Metabolism III. Vitamin D Toxicity

IV. Species Variation in Vitamin D Metabolism and Action V. Conclusions References

I. INTRODUCTION

II. VITAMIN D METABOLISM

In 1919, when the field of experimental nutrition was still in its infancy, Sir Edward Mellanby conducted a classic experiment that for the first time associated the supplementation of various growth-promoting fats with the prevention of rickets [1]. He credited the cure to the presence of a fat-soluble substance called vitamin A. McCollum et al. [2], however, later discovered that the factor responsible for healing rickets was distinct from vitamin A. McCollum named this new substance vitamin D. It was also during this period when scientists realized that there were two antirachitic factors with distinct structures. As discussed by Norman [3], the first factor to be identified was designated vitamin D2 (also known as ergocalciferol), whereas the structure of vitamin D3 (cholecalciferol) became evident some 4 to 5 years later. Vitamins D3 and D2 are used for supplementation of animal and human diets in the United States. Vitamin D3 is the form of vitamin D that is synthesized by vertebrates, whereas vitamin D2 is the major naturally occurring form of the vitamin in plants. Animals that bask in the sun such as amphibia, reptiles, and birds therefore synthesize sufficient endogenous vitamin D3 to meet their daily needs. However, herbivores may have evolved utilizing vitamin D2 as their predominant source. This chapter focuses on the general control and function of key enzymes involved in the regulation of vitamin D2 and vitamin D3 metabolism. Species differences in vitamin D metabolism, as well as vitamin D toxicity, are also discussed. The reader is also directed toward a number of additional reviews regarding vitamin D metabolism and action [3–8]. This chapter gives an overview of vitamin D metabolism; critical steps are discussed in further detail in the subsequent chapters of this section. Metabolism of vitamin D analogs is covered in Chapter 81.

A. Overview

VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX,

Vitamin D refers to a group of compounds that possess antirachitic activity. Technically vitamin D is classified as a secosteroid. Secosteroids are those in which one of the rings has been broken; in vitamin D, the 9,10 carbon–carbon bond of ring B is broken, and it is indicated by the inclusion of “9,10-seco” in the official nomenclature. There are several known nutritional forms of vitamin D, however, the best known examples are cholecalciferol (vitamin D3), which is produced in the skin, and ergocalciferol (vitamin D2), which is derived from plant tissues (Fig. 1). Therefore, when reference is made to vitamin D, the lack of a subscript usually implies either vitamin D2 or vitamin D3. The vitamin Ds are named according to the rules of the International Union of Pure and Applied Chemists (IUPAC) for steroid nomenclature [9]. Because vitamin D is derived from a steroid, the structure retains its numbering from the parent compound cholesterol. Configurations at asymmetric centers are designated by using the R and S notation applying the sequence-rule procedure [10]. Configuration of the double bonds are notated E for “entgegen” or trans, and Z for “zuzammen” or cis [11]. Thus the official name of vitamin D3, by relation to cholesterol, is 9,10-seco(5Z,7E)5,7,10(19) cholestatriene-3β-ol, and the official name of vitamin D2 is 9,10-seco(5Z,7E)-5,7,10(19), 22-ergostatetraene-3β-ol. Contemporary views categorize vitamin D3 not as a vitamin but, rather, as a prosteroid hormone. This concept is supported by the fact that in mammals vitamin D3 is derived from 7-dehydrocholesterol (the precursor of cholesterol) present in the skin. The direct action of sunlight on 7-dehydrocholesterol results in cleavage Copyright © 2005, Elsevier, Inc. All rights reserved.

16

RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY

FIGURE 1

Important nutritional forms of vitamin D.

of the B ring of the steroid structure that on thermoisomerization yields vitamin D3 (see Chapter 3). The significance of vitamin D as a prosteroid hormone became clearer in 1967 when Morii et al. [11a] isolated a new metabolite of vitamin D3 from rats that was as effective as vitamin D3 in healing rickets, raising blood calcium, and increasing intestinal calcium transport. This compound acted more rapidly than vitamin D3, requiring only 8 to 10 hr after oral administration to initiate its response. This metabolite was identified as 25-hydroxyvitamin D3 (25OHD3) [11b]. The liver was demonstrated to be important in the production of this most abundant circulating form of vitamin D3 that, under normal conditions, is present at 20 to 50 ng/ml [4]. Shortly following the discovery of 25OHD3, a number of laboratories showed that this metabolite is specifically hydroxylated at the lα-position in the kidney to yield 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] [11c–13]. The latter metabolite is now generally accepted as the hormonally active form of vitamin D3. Its importance is evidenced by the genetic disorder pseudo vitamin D–deficiency rickets (PDDR), which is caused by mutations in the 1α-hydroxylase gene. PDDR results in the inability to produce 1,25(OH)2D leading to severe rickets (see Chapters 71 and 72). In normal human plasma, 1,25(OH)2D3 circulates at approximately 1000-fold lower concentrations than 25OHD3 and is generally present at 20 to 65 pg/ml [14]. This simplistic picture outlined for vitamin D3 activation is complicated by the fact that vitamin D3 can be oxidatively metabolized to a variety of products. Most of these numerous metabolites have no identifiable

biological function, and indeed many have been isolated from animals fed abnormally high amounts of vitamin D3. Nevertheless, the evidence collected to date indicates that 25-hydroxylated vitamin D3 metabolites are preferentially metabolized at the side chain. In particular, carbon centers C-23, C-24, and C-26 are readily susceptible to further oxidation. Figure 2 illustrates products of these oxidative pathways. As indicated, these pathways are shared by both 25OHD3 and 1,25(OH)2D3, and their physiological importance is still a matter of controversy. For example, there is evidence that 24,25-dihydroxyvitamin D3 [24,25(OH)2D3] may function to stimulate bone mineralization [15,16], suppress parathyroid hormone (PTH) secretion [17], and maintain embryonic development [18]. For the most part, however, the C-24 hydroxylation and other side-chain modifications are generally considered to be catabolic in nature and play a key role in maintaining vitamin D homeostasis [19]. Although these side-chain oxidative pathways yield metabolites that are considered “nonfunctional,” the presence of these compounds in circulation could pose serious problems in the analysis for 25OHD3 and 1,25(OH)2D3 [20]. Further complicating the issue of understanding vitamin D activation, catabolism, and metabolite analysis is the presence of vitamin D2. Vitamin D2 has been shown to contribute significantly to the overall vitamin D status in humans and other mammals consuming supplemental vitamin D2 [21–23]. Vitamin D2 can also be metabolized in a similar fashion to produce several metabolites analogous to the vitamin D3 endocrine system, including the hormonally

17

CHAPTER 2 Vitamin D Metabolism

OH

−H2 O

OH (1α)

OH

(3β)HO

OH

OH(1α)

(3α)HO

1,25(OH)2D3

OH(1α)

1,25(OH)2-3-epi-D3 −H2O

OH

OH (1α)

FIGURE 2 Pathways of vitamin D3 metabolism.

active form of vitamin D2, 1,25-dihydroxyvitamin D2 [1,25(OH)2D2] [24]. Simple inspection of the side chain, however, would imply that differences between metabolism of vitamin D2 and vitamin D3 may exist. The presence of unsaturation at carbon centers C-22/C-23, along with the additional methyl group at C-24, would seem to preclude the existence of the same metabolic pathways for the two vitamins. Figure 3 outlines some of the known pathways of vitamin D2 metabolism that have been shown to date. Deviations in the vitamin D2 and vitamin D3 pathways are discussed in detail in the following sections.

B. 25-Hydroxylase The 25-hydroxylation of vitamin D is the initial step in vitamin D activation. The enzyme responsible for production of this metabolite is located in the liver (see Chapter 4). Extrahepatic sources of 25-hydroxylation have been described [25]; however, experiments with hepatectomized rats provided evidence that the liver is the major, if not the sole, physiologically relevant site of 25-hydroxylation of vitamin D [26]. Subsequent studies

also described the existence of the 25-hydroxylase in both liver mitochondria and microsomes [27–31]. In early work, the microsomal enzyme was described as an enzyme of low capacity and high affinity and, therefore, the enzyme of greatest physiological importance [31]. In contrast, the mitochondrial enzyme was described as a high-capacity, low-affinity enzyme thought to be relevant only under conditions of high vitamin D concentration such as vitamin D toxicity [32]. Early evidence that the microsomal enzyme was the physiologically relevant enzyme came from experiments that suggested this enzyme could be regulated by vitamin D status [31]. It is now clear that liver production of 25-hydroxyvitamin D (25OHD) is not significantly regulated. 25OHD production is primarily dependent on substrate concentration. An important consequence of this lack of physiological regulation of 25OHD is that measurement of blood 25OHD is an excellent measure of vitamin D nutritional status. The purification and cloning of putative liver 25-hydroxylases have been reviewed several times [33–37]. Examination of the literature shows that most of the focus is on the mitochondrial 25-hydroxylase

18

RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY

FIGURE 3 Pathways of vitamin D2 metabolism.

designated CYP27A1, which is a cytochrome P450 capable of C-26(27) hydroxylation of sterols involved in bile acid synthesis and the 25-hydroxylation of vitamin D3. The rat, rabbit, and human enzymes have been cloned [38–41]. The CYP27A1 clone has been expressed

in COS cells [39,42], and its activity has been isolated from the mitochondria of these cells. The expressed enzyme was found to 27-hydroxylate cholestanetriol and 25-hydroxylate vitamin D3. However, CYP27A1 does not 25-hydroxylate vitamin D2 [39]. Rather, CYP27A1

CHAPTER 2 Vitamin D Metabolism

was found to 24-hydroxylate and 26(27)-hydroxylate vitamin D2. These activities could explain the presence of 24-hydroxyvitamin D2 (24OHD2), 1,24-dihydroxyvitamin D2 [1,24(OH)2D2], and 24,26-dihydroxyvitamin D2 [24,26(OH)2D2] [43–45] in the plasma of rats and cows. Since rats fed vitamin D–deficient diets and supplemented with physiological amounts of vitamin D2 have 25OHD2 as their predominant monohydroxylated vitamin D2 metabolite in the plasma [44] and targeted disruption of CYP27A1 does not decrease serum 25OHD3 [46], CYP27A1 is likely not the physiologic enzyme responsible for the 25-hydroxylation of vitamin D. The rat liver microsomal 25-hydroxylase (CYP2C11) has also been studied, but it has been shown to be male-specific [47]. Data have also been presented indicating that microsomes do [48] or do not [49] possess 25-hydroxylase activity. Therefore, conclusions regarding the importance of the CYP27A1 and other ostensible microsomal 25-hydroxylases require additional research. Data obtained studying pig liver 25-hydroxylation of vitamin D3 [36,37,48,50,51] suggest that a third liver 25-hydroxylase exists that is microsomal in origin. In the pig, this enzyme, CYP2D25, is present equally in males and females and is markedly different from CYP27A1 and CYP2C11 based on a terminal amino acid sequence [51]. Most important is the finding that this pig microsomal enzyme 25-hydroxylates vitamin D2 and vitamin D3 equally. The 25-hydroxylation of vitamin D is not yet completely understood. Several enzymes may play a role in the 25-hydroxylation of vitamin D. Whether one enzyme is more physiologically relevant than others remains to be determined. Studies in primary cultures of pig hepatocytes suggest that both CYP2D25 and CYP27A1 can play a role in 25-hydroxylation of vitamin D3 [51]. Nevertheless, it is clear that mammals can use vitamin D2 as a sole source of vitamin D. Therefore, any 25-hydroxylase proclaimed as the key enzyme(s) in the 25-hydroxylation of vitamin D must be capable of 25-hydroxylating vitamin D2 as well as vitamin D3. If mammals, for example, possessed an enzyme with specificity for the vitamin D2 side chain, this enzyme may be missed due to the almost exclusive use of vitamin D3 or vitamin D3 analogs as substrates for the 25-hydroxylating reaction.

C. 1α-Hydroxylase (CYP27B1) In the late 1960s, 25OHD3 was believed to be the metabolically active form of vitamin D. However, the presence of a more polar metabolite, which accumulated

19 in the intestinal mucosa chromatin of chicks administered 3H-labeled vitamin D3, suggested a new candidate for the active form of vitamin D [52]. Subsequent work by Lawson et al. [53] showed that during the formation of this metabolite the 1α-3H was lost. This led them to suggest that the new metabolite had an oxygen function inserted at C-1 in addition to the hydroxyl group at C-25. The enhanced biological activity of this new metabolite was evident before its structure could be determined [54–56]. Fraser and Kodicek [56] demonstrated that nephrectomy abolished production of the new metabolite, and this active vitamin D compound was synthesized by kidney mitochondria. In 1971, three laboratories identified the active form of vitamin D as 1,25(OH)2D3 [11,12,57]. Subsequently, the vitamin D2 form was also isolated and identified [24]. The CYP27B1 is located in the inner mitochondrial membrane of the proximal convoluted tubule cells of the kidney [58] and is discussed in detail in Chapter 5. Extrarenal sites of 1α-hydroxylation have been reported in bone, liver, placenta, macrophages, and skin [59]. The physiological significance of these sites on systemic calcium metabolism is in doubt, as nephrectomy and/or severe renal failure results in very low to undetectable circulating 1,25(OH)2D3 levels [60]. The regulation of 1,25(OH)2D3 production is reciprocally regulated with respect to 24,25(OH)2D3 [61]. Hypocalcemia caused by calcium-deficient diets, vitamin D deficiency, or pathological factors results in increased production of 1,25(OH)2D3 [61–67]. This hypocalcemic-mediated induction of 1,25-dihydroxyvitamin D [1,25(OH)2D] production is secondary to increased PTH. Administration of PTH to animals results in increased 1,25(OH)2D3 production [64,67,68]. PTH treatment in vitro induces CYP27B1 in renal slices [63] and cultured kidney cells [67,69] and is cAMP dependent [63,67,69,70]. Thyroparathyroidectomy (TPTX) or parathyroidectomy (PTX) results in the loss of the ability to synthesize 1,25(OH)2D3. In humans, acute administration of PTH or primary hyperparathyroidism results in increased production of 1,25(OH)2D [14,71], which is evidenced by elevations in plasma 1,25(OH)2D. However, in animal studies where PTH was administered chronically to goats and calves, a transient rise in plasma 1,25(OH)2D was observed followed by a rapid decline to nearly undetectable levels [72,73]. These results could be attributed to hypercalcemic feedback on the renal CYP27B1. When plasma calcium in these animals reached 13 mg/dl, 1,25(OH)2D production appeared to cease. This same group conducted similar experiments in rats and showed that chronic PTH infusion did not result in a reduction of plasma 1,25(OH)2D3, but rather a

20

RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY

modest rise [74]. Clearly, there are varying degrees of direct calcium-mediated feedback on the renal CYP27B1. It is possible that species and age affect the set points at which plasma calcium becomes a direct negative regulator of 1,25(OH)2D3 production. In contrast to the indirect role of plasma calcium in inducing 1,25(OH)2D3 production, the role of plasma phosphate appears more direct. As plasma phosphate declines, animals shift from 24,25(OH)2D production to increased 1,25(OH)2D3 production [75,76]. Since phosphate-deficient animals are hypercalcemic, serum PTH is down and therefore cannot be providing the signal to increase 1,25(OH)2D3 production. Furthermore, TPTX phosphate-deficient animals produce 1,25(OH)2D3 similarly to intact phosphate-deficient animals [66]. Gray [77] demonstrated that hypophysectomy abolished the increase in plasma 1,25(OH)2D concentrations that normally accompanied dietary phosphate deprivation. Gray demonstrated that growth hormone or triiodothyronine replacement to hypophysectomized rats restored elevations in plasma 1,25(OH)2D associated with low dietary phosphorus, therefore suggesting a permissive role of these hormones in regulation of the renal CYP27B1 during phosphorus deficiency. A direct negative effect of 1,25(OH)2D on its own production has been reported. The inhibitory effect of 1,25(OH)2D3 on renal CYP27B1 activity occurs both in vivo and in vitro [64,78]. This repressive activity is probably indirect [79]. In vivo, this effect may be partially mediated through the ability of 1,25(OH)2D3 to inhibit PTH secretion [80]. Similarly, Beckman et al. [81] showed that vitamin D toxicity mildly inhibited renal CYP27B1 activity while low-calcium diets significantly induced the CYP27B1. They further demonstrated that administration of toxic doses of vitamin D to animals fed a calcium-deficient diet reduced CYP27B1 activity by 90%. This result occurred in spite of the fact that these animals were hypocalcemic and had serum PTH levels equal to those of control animals receiving calcium-deficient normal vitamin D diets. These data suggest that high plasma concentration of vitamin D metabolites may act directly to suppress CYP27B1 activity. There are many additional factors such as calcitonin (CT), acidosis, sex steroids, prolactin, growth hormone, glucocorticoids, thyroid hormone, and pregnancy that are potential regulators of 1,25(OH)2D production. One of the most recent and interesting is the requirement for the endocytic receptor megalin in the proximal tubular cells to allow uptake of 25OHD for 1α-hydroxylation [82]. Discussion of these is beyond the scope of this general review of vitamin D metabolites. The characteristics and regulation of the CYP27B1 are described further in several reports [37,83–85].

The cloning of the renal CYP27B1 has been achieved [86–89]. Studies with CYP27B1 knockout mice [90,91] suggest that these animals have abnormalities similar to those observed in PDDR. These knockout models will undoubtedly provide new insight into the functions of 1,25(OH)2D and are the subject of further review in Chapter 7.

D. 24-Hydroxylase (CYP24A1) The 24-hydroxylation of 25OHD3 and 1,25(OH)2D3 to form 24,25(OH)2D3 [92] and 1,24,25(OH)3D3 [93,94] is the primary mechanism and the first step in a metabolic pathway to inactivate and degrade these vitamin D metabolites. It now appears that CYP24A1 is ubiquitous and may be present in every cell and tissue that contains the vitamin D receptor (VDR). In the kidney, CYP24A1 is found on the inner mitochondrial membrane of the renal tubules [95]. The primary regulators of renal CYP24A1 activity are PTH and 1,25(OH)2D3. Normal and TPTX animals receiving injections or infusions of 1,25(OH)2D3 show marked increases in both renal CYP24A1 mRNA levels and activity [63,74,96,97]. Administration of PTH partially or completely blocks expression of CYP24A1 mRNA and activity in these animals [63,68,74,96,97]. PTH acts on the kidney via adenylate cyclase and cAMP, and it has been shown that infusions of cAMP in vivo block l,25(OH)2D3mediated inductions of the renal CYP24A1 [96,98]. Animals on calcium-deficient diets have elevated plasma 1,25(OH)2D concentrations, which are accompanied by suppressed or undetectable renal CYP24A1 activity [78,96], as well as reduced VDR concentrations [99]. The reasons for the inability of 1,25(OH)2D3 to up-regulate renal CYP24A1 during calcium deficiency are not clear. Iida et al. [100] have proposed that the down-regulation of renal VDR during calcium deficiency may be responsible for preventing the l,25(OH)2D3mediated induction of renal CYP24A1. In vivo studies by Reinhardt and Horst [74], however, suggest that under these conditions PTH is probably the more important mediator of renal CYP24A1 regulation rather than downregulation of VDR. In their experiments, Reinhardt and Horst [74] showed that 1,25(OH)2D3 treatment of animals on normal calcium diets resulted in significant up-regulation of renal CYP24A1 as well as VDR. However, when PTH was infused simultaneously with 1,25(OH)2D3, VDR up-regulation was still observed (albeit to a lesser degree), whereas CYP24A1 up-regulation was completely blocked. The importance of PTH in preventing the l,25(OH)2D3-mediated up-regulation of the renal CYP24A1 is also apparent

21

CHAPTER 2 Vitamin D Metabolism

by observation in aged rats. With advancing age, renal PTH receptors are down-regulated [101,102], while VDR remains unchanged [103]. The reduction in renal PTH receptors makes the kidney less responsive to PTH [64], which is associated with significant elevations in CYP24A1 mRNA [103,104]. These data suggest that renal responsiveness to PTH, not a decline in VDR, is the major physiological regulator of the renal CYP24A1. In the intestine, 1,25(OH)2D3 is the primary regulator of CYP24A1. In vivo administration of 1,25(OH)2D3 rapidly induces intestinal CYP24A1 activity [105]. This activity peaks by 6 hr postinjection, and rapidly declines thereafter to control levels 24 hr postinjection. Time-course experiments show that CYP24A1 mRNA peaks 4 to 6 hr post-injection and then rapidly disappears [104]. This is in contrast to the renal CYP24A1 mRNA, which peaks 12 to 24 hr post-l,25(OH)2D3 treatment and declines much more slowly. Shinki et al. [96] proposed that the intestinal CYP24A1 was 100 times more sensitive to 1,25(OH)2D3 stimuli than the renal CYP24A1. However, they examined CYP24A1 mRNA only 3 hr after a 1,25(OH)2D3 dose. Because renal CYP24A1 requires an additional 6 to 12 hr to reach peak expression, they likely underestimated the true sensitivity of the kidney to a 1,25(OH)2D3 dose. In contrast to their effect in the kidney, TPTX, PTH administration, or cAMP infusion does not affect intestinal expression of CYP24A1 induced by 1,25(OH)2D3 [96]. Animals fed low-calcium diets, with the associated secondary hyperparathyroidism and high plasma 1,25(OH)2D3 concentrations, have marked inductions of both intestinal CYP24A1 mRNA and activity [96,105]. Another contrast between intestinal and renal CYP24A1 expression is seen in the aging rat model. Intestinal CYP24A1 mRNA and activity decline or change very little in the aged animal. This contrasts to the largely increased expression of renal CYP24A1 observed in the aged animal [103]. Calcitonin has been shown to be a potent suppressor of intestinal CYP24A1 expression [106]. In these experiments, Beckman et al. [81,106] showed that vitamin D toxicity was a potent inducer of CYP24A1 mRNA and enzyme expression in both intestine and kidney. They also showed that if the hypervitaminosis D3-induced hypercalcemia was prevented by feeding low-calcium diets, the intestinal CYP24A1 expression was enhanced four-fold over hypercalcemic animals receiving the same toxic doses of vitamin D3 but consuming a normal calcium diet. These observations prompted the examination of the possibility that CT released in response to the hypercalcemia may have suppressed the induced expression of the intestinal CYP24A1. In their series of experiments, Beckman et al. [106] clearly demonstrated that CT was a potent suppressor of intestinal CYP24A1 activity. Conceivably, the CT-mediated suppression of

CYP24A1 activity could enhance l,25(OH)2D-mediated activities by prolonging its half-life. This could exacerbate conditions that manifest hypercalcemia, such as hypervitaminosis D, by preventing 24-hydroxylation and catabolism of active vitamin D metabolites. The CYP24A1 has been purified [107,108] and cloned [107,109] and the clone has been expressed [107,109]. Analysis of the amino acid sequence of the rat and human CYP24A1s showed that the sequences were 90% similar. The 21-amino acid heme binding region was found to be 100% identical [109]. Ohyama et al. [110] isolated the gene encoding the rat CYP24A1. This single-copy gene was approximately 15 kb and was composed of 12 exons. Several putative vitamin D response elements have been identified and are currently under study. Details of the purification and cloning of the CYP24A1 have been reviewed previously [35], and additional review of the molecular analysis and regulation of the CYP24A1 can be found in Chapter 6 in this book.

E. Physiological Role of CYP24A1 The major site for 24-hydroxylation appears to be the kidney. This is based on the observation that nephrectomy reduced or eliminated plasma 24,25(OH)2D3 [111]. However, nephrectomy also eliminates the production of 1,25(OH)2D3, the primary stimulator of CYP24A1. Therefore, the possibility remains that 24,25(OH)2D3 may reappear in plasma of nephrectomized subjects treated with therapeutic doses of 1,25(OH)2D3. Some studies suggest that 25OHD3 is the primary substrate for CYP24A1 [112]; however, it is now generally accepted that CYP24A1 is distributed throughout the body and that 1,25(OH)2D3, rather than 25OHD3, is the preferred substrate for CYP24A1 [96,113,114]. Napoli et al. [115] and Napoli and Horst [116] identified the formation of 24-oxo-1,25(OH)2D3 and 24-oxo-l,23,25-trihydroxyvitamin D3 [24-oxo-1,23, 25(OH)3D3] from intestinal homogenates incubated with physiological amounts of 1,25(OH)2D3. The formation of 24-oxo-1,23,25(OH)3D3 from 1,25-(OH)2D3 was enhanced by treatment of experimental animals with exogenous 1,25-(OH)2D3. They suggested that 24-hydroxylation, followed by C-23 oxidation, most likely represents a mechanism for terminating the cellular action of 1,25(OH)2D3. In a review, Haussler [117] proposed a model for the cellular action of 1,25(OH)2D3 in which he suggested that receptor-mediated, selfinduced catabolism of 1,25(OH)2D3 modulates the action of 1,25(OH)2D3. The work of Lohnes and Jones [118], Reddy and Tserng [119], and Makin et al. [120] provided further support for this proposal by showing

22

RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY

the ubiquitous presence of catabolic pathways initiated by CYP24A1 in 1,25(OH)2D3 target tissues and the complete destruction of 1,25(OH)2D3 by these pathways. In fact, CYP24A1 has been shown to do more than just initiate this catabolic cascade. Akiyoshi-Shibata et al. [121] expressed the rat CYP24A1 cDNA in Escherichia coli. They found that this enzyme not only 24-hydroxylates but catalyzes the dehydrogenation of the 24-OH group and performs 23-hydroxylation resulting in 24-oxo-l,23,25(OH)3D3 production. Only the cleavage at C-23/C-24 resulting in the 24,25,26,27tetranor-1OH,23COOHD3 was not demonstrated. However, it is now recognized from the work of several groups that CYP24A1 is capable of the complete catabolism of 25OHD3 and 1,25(OH)2D3 via a fivestep reaction process that includes 24-hydroxylation, 24-oxidation, 23-hydroxylation, side-chain cleavage, and subsequent production of the final degradative product, calcitroic acid [121–125]. Direct evidence that self-induced metabolism of 1,25(OH)2D3 suppresses the action of 1,25(OH)2D3 on target cells was reported by Pols et al. [126,127] and Reinhardt and Horst [128,129]. Both laboratories showed that ketoconazole inhibited l,25(OH)2D3induced metabolism. This inhibition resulted in increased specific accumulation of 1,25(OH)2D3 in target cells and a significant increase in the cellular half-life of l,25(OH)2D3-occupied VDR [129]. A result of blocking the self-induced metabolism of 1,25(OH)2D3 was up-regulation of the VDR. Reinhardt and Horst [128] extended these studies by demonstrating that selfinduced metabolism of 1,25(OH)2D3 in target cells limits the response of target cells to a primary 1,25(OH)2D3 stimulus by reducing occupancy of VDR by 1,25(OH)2D3 and by preventing VDR up-regulation. Additionally, their data showed that entry of 1,25(OH)2D3 into the cell is restricted due to extensive metabolism of the 1,25(OH)2D3. In whole-cell VDR assays, hormone was degraded so rapidly that VDR binding was prevented. Reinhardt et al. [130] confirmed the inhibitory effects of self-induced induction of CYP24A1 on the cellular action of 1,25(OH)2D3 in vivo by demonstrating that ketoconazole potentiates the 1,25(OH)2D3 up-regulation of VDR in rat intestine and bone. Clearly, one of the primary roles of CYP24A1 catabolic pathway is terminating the actions of 1,25(OH)2D3. Recent work using CYP24A1 knockout mice has provided additional evidence for the role of CYP24A1 in the regulation of 1,25(OH)2D activity via catabolism [19]. These mice were unable to clear 1,25(OH)2D, as was evidenced by abnormally high circulating concentrations of this metabolite. These data provided support to the contention that the primary role of CYP24A1 is

initiating the catabolic pathway and terminating the actions of 1,25(OH)2D. The role of CYP24A1 as an enzyme responsible for the production of a biologically active compound, i.e., 24,25(OH)2D3, has been controversial. The work of St-Arnaud et al. [19] has also addressed part of this issue. Deficient mineralization of intramembranous bone was found in CYP24A1-ablated mice. The genetic cross of CYP24A1-ablated mice with VDR-ablated mice rescued the defective bone phenotype, strongly suggesting that the high concentrations of 1,25(OH)2D acting through VDR and not 24,25(OH)2D was the cause of the bone defect. Constitutive expression of CYP24A1 in transgenic rats surprisingly resulted in low plasma 24,25(OH)2D and 25OHD with no effect on plasma 1,25(OH)2D [131,132]. These rats developed albumineria and hyperlipidemia and suffered from reduced bone mass. The authors demonstrated that excreted albumin appeared to compete for the binding and reabsorption of the DBP-25-OHD3 complex with megalin, resulting in a loss of 25OHD3 into the urine and subsequent reduction of plasma 24,25(OH)2D3. Supplementation of these rats with 25OHD3 prevented the bone loss without changing plasma 1,25(OH)2D.

F. Other Vitamin D3 Derivatives Functionalized at C-24 In a series of experiments conducted by Wichmann et al. [133,134], a number of 24-hydroxylated derivatives were isolated from plasma of chicks made toxic with vitamin D3. These metabolites included 24OHD3, 23,24,25-trihydroxyvitamin D3 [23,24,25(OH)3D3], and 24,25,26-trihydroxyvitamin D3 [24,25,26(OH)3D3]. These metabolites have not been described in animals receiving physiological amounts of vitamin D3, and their biological significance is unknown; however, it is likely that 23,24,25(OH)3D3 and 24,25,26(OH)3D3 are metabolites of 24,25(OH)2D3. 24,25(OH)2D3 is also the probable precursor to the formation of the side chain cleavage product, 25,26,27-tri-norvitamin D324-carboxylic [135]. This metabolite has been shown to be a product of in vitro kidney perfusion using 25OHD3 as substrate. The analogous pathway, however, could not be demonstrated using 1,25(OH)2D3 as substrate (S. Reddy, personal communication, 1996). Another metabolite isolated in the experiments of Wichmann et al. [133] was 23-dehydro-25OHD3. The immediate precursor, site(s), and biological activity of this compound are unknown. Plausible sources of the 23-dehydro compound are dehydration of 24,25(OH)2D3

CHAPTER 2 Vitamin D Metabolism

or 23,25-dihydroxyvitamin D3 [23,25(OH)2D3]. It is not certain if any of the metabolites are important under physiological conditions.

G. 23-Hydroxylation The discovery of a C-23 oxidative pathway emerged much later than the other pathways and was ushered in by the identification of 23(S),25(R)25OHD3-26,23lactone [136,137], 23(S),25(R)1,25(OH)2D3-26,23lactone [138], and their respective precursors 23(S),25(OH)2D3 and 1,23(S),25(OH)3D3 [139,140]. To date, there has been no specific 23-hydroxylase identified for the vitamin D system. Rather, like other side-chain modifications, 23-hydroxylation is likely carried out by CYP24A1 [122]. The compound, 25OHD3-26,23-lactone, can be detected in plasma from normal rats, pigs, and chicks [140,141]. However, in several species, this metabolite is not expressed unless animals are consuming excessive amounts of vitamin D3 [142]. This metabolite has unique activity in that it is three- to fivefold more competitive than 25OHD3 for binding to the plasma vitamin D-binding protein (DBP) [136]. It was, therefore, the first modification of 25OHD3 that led to enhanced binding to the plasma DBP. The metabolite, 1,25(OH)2D3-26,23-lactone, has also been demonstrated under normal conditions [143], with elevated plasma concentrations occurring during exogenous administration of pharmacological amounts of 1,25(OH)2D3 [144]. The major locus for formation of C-23 hydroxylated derivatives appears to be the kidney. Horst and Littledike [142] and Napoli et al. [140] demonstrated that nephrectomy eliminated or greatly impaired the biosynthesis of 25OHD326,23-lactone when animals were treated with excess vitamin D3 or 25OHD3. They showed that this response was due to the inability of the animals to synthesize 23(S),25(OH)2D3. However, when 23(S),25(OH)2D3 was given to nephrectomized animals, the synthesis of 25OHD3-26,23-lactone was restored. These data suggested that C-23-hydroxylation occurred predominantly, but not exclusively, in the kidney, whereas extrarenal tissues are quantitatively important in the pathway leading to 25OHD3-26,23-lactone synthesis, which includes formation of the lactone intermediates 23,25,26(OH)3D3 [145] and 25OHD3-26,23-lactol [146]. Although ambiguities remain regarding the biological effects of C-24 oxidation, 23-hydroxylation appears to clearly be a deactivation event. 23-Hydroxylation is the first side-chain modification of 25OHD3 noted to substantially reduce its affinity for the plasma DBP [147]. 23-Hydroxylation of 1,25(OH)2D3 also leads to its

23 increased plasma clearance and reduced VDR binding and biological activity [148]. The role of 23-hydroxylation as a primary oxidation event for the further metabolism of 25OHD3 and 1,25(OH)2D3 is relatively minor to its role in the further metabolism of vitamin D3 metabolites that have been previously oxidized at C-24. In other words, very little production of 23,25(OH)2D3 or 1,23,25(OH)3D3 would be expected under normal conditions. Rather, the convergences of the C-24 and C-23 oxidative pathways would lead predominantly to the formation of 24-keto-1,23,25(OH)2D3 and 24-keto-1,23,25(OH)3D3 [116], which subsequently cleave to form C-23 acids [119,149]. Therefore, as indicated in Fig. 2, the C-23 oxidative pathway can lead to two different patterns of side-chain modifications for both 25OHD3 and 1,25(OH)2D3. One pathway, which is relatively minor under physiological conditions and more predominant during hypervitaminosis D3, leads through 23-hydroxylation to formation of the lactones, whereas the other more physiologically significant pathway leads through 24-hydroxylation to 23-hydroxylation of 24-keto metabolites. Other oxidized C-23 metabolites that have been identified include 23-keto derivatives of 25OHD3 and 1,25(OH)2D3. 23-Keto-25-hydroxyvitamin D3 was synthesized in vitro from 23(S)25(OH)2D3 and 23(R),25(OH)2D3 and has unique properties in that it binds with twofold higher affinity than 25OHD3 for binding sites on the plasma DBP [150]. This affinity should be compared to that for 23(S),25(OH)2D3, which binds with 6- to 10-fold less affinity. 23-Keto25OHD3 is also about fourfold more competitive than 25OHD3 for binding to the VDR. 23-Ketonization is, therefore, the first example of a side-chain modification enhancing the affinity of 25OHD3 for the VDR. This high affinity of 23-keto-25OHD3 for VDR prompted biosynthesis of 23-keto-l,25(OH)2D3 to determine if this modification might enhance binding of 1,25(OH)2D3 to VDR. Horst et al. [151] prepared this metabolite by incubating 23-keto-25OHD3 in kidney homogenates prepared from vitamin D–deficient chicks. The major metabolite was isolated and identified as 23-keto-1,25(OH)2D3 and was shown to possess about 40% the activity of 1,25(OH)2D3 for VDR binding. 23-Ketonization of 1,25(OH)2D3, therefore, reduced the affinity of 1,25(OH)2D3 to VDR rather than increased binding. 23-Keto metabolites do not appear to be synthesized under physiologic conditions, as Napoli et al. [115] could not demonstrate the presence of these metabolites from rat intestinal homogenates incubated with 1,25(OH)2D3 or from intestinal extracts from rats dosed with 1,25(OH)2D3.

24

RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY

H. C-26 Hydroxylation 26-Hydroxylation of 25OHD3 and 1,25(OH)2D3 produces 25,26(OH)2D3 [152] and 1,25,26(OH)3D3 [153], respectively. The natural product was originally assigned the 25(R) configuration. However, Partridge et al. [154] gave the assignment as 25(S). Ikekawa et al. [155] later discovered that the naturally occurring 25,26(OH)2D3 actually existed as a mixture of 25(S) and 25(R) isomers. Although this assignment seems somewhat trivial, it was important in unraveling a controversy that existed regarding the physiological precursor to the in vivo synthesis of 25OHD3-26,23-lactone. Hollis et al. [156] demonstrated that 25,26(OH)2D3 isolated from in vivo sources could act as a precursor to the formation of the 25OHD3-26,23-lactone. Subsequent research, however, suggested that synthetic 25(S),26(OH)2D3 (which at the time was thought to be the natural configuration) did not act as precursor to the formation of 25OHD3-26,23-lactone [157], but synthetic 25(R),26(OH)2D3 could act as a precursor [140,158]. As naturally occurring 25,26(OH)2D3 is a mixture of the R and S isomers, this research validated the conclusion of Hollis et al. [156] suggesting that formation of 25OHD3-26,23-lactone could indeed proceed through 25,26(OH)2D3. This pathway has been shown to be relatively minor [159,160], with the major pathway to 25OHD3-26,23-lactone synthesis proceeding through 23(S),25(OH)2D3 [140,145,158]. The major locus for the 26-hydroxylase is unknown. Blood concentrations of 25,26(OH)2D3 are not depressed in nephrectomized humans or pigs [142,161,162]. Therefore, production of these metabolites must take place at extrarenal sources. 26-Hydroxylase activity has, however, been demonstrated in microsomes isolated from rat and pig kidneys [163]. The only extrarenal source was reported in liver mitochondria [164]. The physiological role of the C-26 oxidative pathway remains elusive. However, 25,26(OH)2D3 and 1,25,26(OH)3D3 have been shown to possess biological activity with regard to stimulating bone calcium resorption and intestinal calcium absorption, albeit to a lesser degree than either 25OHD3 or 1,25(OH)2D3 [152,165]. Therefore, it seems unlikely that 26-hydroxylation is essential for calcium uptake from the gut or release of calcium from bone.

I. C-3 Epimerization Reddy et al. [159,160] have reported the metabolism of 1,25(OH)2D3 in primary cultures of neonatal human keratinocytes and rat osteosarcoma cells into the novel A-ring modified metabolite, 1,25(OH)2-3-epi-D3.

This epimer is formed as a result of the change in the orientation of the C-3 hydroxy group from β to α. Other investigators also confirmed this finding [166,167]. Epimerization of hydroxy groups is a wellknown phenomenon in bile acid metabolism [168] and the reaction is conducted by bile acid hydroxysteroid dehydrogenase. Figure 4 outlines the proposed pathways of C-3 epimerization of 1,25(OH)2D3 as described by Reddy et al. [159]. Although there are two potential pathways for the production of 1,25(OH)2-3-epi-D3 from 1,25(OH)2D3, the most likely pathway is through keto intermediates. The C-3 epimerization has been shown to play a major role in hormone activation and inactivation in other steroid systems [169]. Indeed, 1,25(OH)2-3-epi-D3 binds to the cellular 1,25(OH)2D3 receptor (VDR) with less affinity [170,171] and has minimal activity at activating intestinal calcium transport and bone calcium resorption than 1,25(OH)2D3 [170]. On the other hand, 1,25(OH)2-epi-D3 is equipotent to 1,25(OH)2D3 at suppressing parathyroid hormone secretion in bovine parathyroid cells [171] and at inhibiting keratinocyte proliferation [170,172]. While 1,25(OH)2-3-epi-D3 does undergo side-chain metabolism [159,171], its conversion to C-23 and C-24 oxidized metabolites occurs at a slower rate than 1,25(OH)2D3 [171]; therefore, enhanced metabolic stability of the 1,25(OH)2-3-epi-D3 has been proposed as a possible explanation for the high in vitro activity in spite of its reduced binding affinity for VDR [171]. Thus, the enzyme(s) responsible for C-3 epimerization appears to play an important role not only in the regulation of intracellular concentration of 1,25(OH)2D3 but also in the formation of metabolites with a different biological activity profile in specific target tissues. These differences in intracellular concentration of 1,25(OH)2D3 and its metabolites from one tissue to another may be one possible explanation for the well-known tissuespecific actions of 1,25(OH)2D3.

J. Unique Aspects of Vitamin D2 Metabolism The 24 position of vitamin D2, in contrast to the similar position in vitamin D3, can be considered to be highly reactive. It is a tertiary carbon as well as an allylic position, and the formation of a reactive intermediate (radical, cation) at this position would be highly stabilized. The proximity of this reactive center to the 25 position would afford the possibility of C-24-hydroxylation of vitamin D2, but the presence of the C-24 methyl would preclude further oxidation to C-24-keto compounds as is known to occur in vitamin D3 metabolism. Jones et al. [43] were the first to demonstrate C-24 oxidation when they isolated 24OHD2

25

CHAPTER 2 Vitamin D Metabolism

FIGURE 4

Possible pathways of epimerization at C-3 of 1,25(OH)2D3 (adapted from [159]).

from the plasma of male rats treated with 100 IU of radiolabeled vitamin D2. Engstrom and Koszewski [173] have determined that production of 24OHD2 can occur in liver homogenates from a variety of species, and actually exceeds the formation of 25OHD2. Horst et al. [44] have shown that the concentration of 24OHD2 in plasma was about 20% that of 25OHD2 in rats receiving physiological doses of vitamin D2, and was equivalent to 25OHD2 in rats receiving pharmacological doses of vitamin D2. They also demonstrated that 1α-hydroxylation of 24OHD2 to form 1,24(OH)2D2 represented a minor but significant pathway for vitamin D2 activation. In their experiments, they determined that 1,24(OH)2D2 rivaled both 1,25(OH)2D2 and 1,25(OH)2D3 in biopotency. Both 24OHD2 and 25OHD2 and their 1α-hydroxylated metabolites can undergo subsequent hydroxylation to form 24,25-dihydroxyvitamin D2 [24,25(OH)2D2] and 1,24,25-trihydroxyvitamin D2 [1,24,25(OH)3D2]. The formation of 1,24,25(OH)3D2 represented an unequivocal deactivation of the vitamin D2 molecule [174]. Conversely, the comparable vitamin D3 metabolite, 1,24,25(OH)3D3, maintains significant biological activity and must undergo further side-chain oxidation to be rendered totally inactive [174]. Although vitamin D2 is known to undergo side-chain oxidation, only recently has evidence emerged suggesting that vitamin D2 (like vitamin D3) undergoes side-chain cleavage.

The paucity of information regarding vitamin D2 sidechain metabolism is primarily due to the lack of appropriate radiolabeled and cold substrates. Zimmerman et al. [175] have used [9,11-3H]-1,25-(OH)2D2 and cold 1,25(OH)2D2 as substrates to determine if 1,25-(OH)2D2 undergoes side-chain cleavage. Similar experiments were done using [3α-3H]-25OHD2 as substrate [176]. In these metabolism experiments it became clear that aqueous-soluble metabolites were being produced from 1,25(OH)2D2 [175] and 25OHD2 [176]. Furthermore, Zimmerman et al. [175] demonstrated that calcitroic acid was the major aqueous-soluble metabolite being produced from cell culture and kidney perfusion experiments. In a more recent study Horst et al. [177] studied the metabolism of 1,25(OH)2D3 and 1,25(OH)2D2 using a purified rat CYP24A1 system. As expected, 1,23(OH)2-24,25,26,27-tetranor D and calcitroic acid were the major lipid- and aqueous-soluble metabolites, respectively, when 1,25(OH)2D3 was used as substrate. However, when 1,25-(OH)2D2 was used as substrate, 1,24,25(OH)3D2 was the major lipid-soluble metabolite with no evidence for the production of either 1,23(OH)2-24,25,26,27-tetranor D or calcitroic acid. Apparently, the CYP24A1 was able to 24-hydroxylate 1,25(OH)2D2, but was unable to cause further changes that would result in side-chain cleavage. When the analog 1,25-(OH)2-22-ene-D3 was used as substrate, CYP24A1 was again able to effect 24-hydroxylation

26

RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY

but not side-chain cleavage (R. Horst, S. Reddy, and J. Omdahl, 2003, unpublished data). Sunita Rao et al. [178] demonstrated that 1,25(OH)2-22-ene-D3 could be metabolized to calcitroic acid by RWLue-4 cells and rat kidney. They suggested that the 1,25(OH)2-22-ene-D3 was first hydroxylated at C-24, followed by further oxidation to 1,25(OH)2-24-oxo-22-ene-D3 prior to side-chain, double-bond reduction to form 1,25(OH)224-oxo-D3. The 1,25(OH)2-24-oxo-D3 was then further metabolized to calcitroic acid, presumably by CYP24A1. The compound 1,25(OH)2D4 [a.k.a. 22,23 dihydro1,25-(OH)2D2] has also been shown to undergo sidechain oxidation similar to that of 1,25(OH)2D2 in vitro [179] and metabolized to calcitroic acid in vivo [180]. These results suggest that metabolism of 1,25(OH)2D2 to calcitroic acid clearly involves enzymes other than CYP24A1. The compounds 25,26-dihydroxyvitamin D2 [25,26(OH)2D2] and 1,25,26-trihydroxyvitamin D2 [1,25,26(OH)3D2] have been chemically synthesized [181,182]. The metabolite 25,26(OH)2D2 has also been tentatively, but not exhaustively, identified from in vivo sources [141,183], but it could not be demonstrated in kidneys perfused with 25OHD2 [184]. Similarly, 1,25,26(OH)3D2 could not be demonstrated [185]. However, it is clear that 26-hydroxylation does occur when vitamin D2 metabolites have been previously 24-hydroxylated. For example, when 24,25(OH)2D2 and 1,24,25(OH)3D2 were used as precursors, the metabolites 24,25,26(OH)3D2 and 1,24,25,26(OH)4D2, respectively, were produced in significant amounts [184,185]. Similarly, Koszewski et al. [45] and Jones et al. [186] demonstrated that C-26 hydroxylation was the major metabolic pathway for the further metabolism of 24OHD2 and 1,24(OH)2D2. Oxidation at C-24, therefore, appears to be a prerequisite for C-26 oxidation of vitamin D2 compounds. A similar situation also appears to exist for C-28 oxidation as demonstrated by Reddy and Tserng [184,185]. In their experiments, they isolated 24,25,28(OH)3D2 and 1,24,25,28(OH)4D2 from rat kidney perfusions and, through the use of various substrates, were able to show C-28 hydroxylation of vitamin D2 metabolites occurs only after C-24 hydroxylation [187].

III. VITAMIN D TOXICITY A. Overview The first documented reports of vitamin D intoxication were made in the late 1920s by Kreitmeir and Moll [188] and Putscher [189]. These cases resulted from the ingestion of large quantities of vitamin D in

the diet. Vitamin D intoxication, however, has never been reported following prolonged sunlight exposure. Holick et al. [190] suggested that nature has provided various control points that prevent the overproduction of vitamin D3 by the skin. The most important point of control is the diversion of vitamin D3 production from 7-dehydrocholesterol to non–biologically active overirradiation products such as lumisterol and tachysterol. In addition, these authors suggested that skin pigmentation and latitude were also significant determinants (albeit to a lesser degree) that limit the cutaneous production of vitamin D3. As discussed earlier, once vitamin D is in circulation, the conversion to 25OHD is relatively uncontrolled. Normally, 25OHD circulates at 30 to 50 ng/ml in most species [141]. However, when vitamin D is given in excess, plasma 25OHD can be elevated to concentrations of 1000 ng/ml or greater [191,192], while plasma 1,25(OH)2D remains at or below normal concentrations [193]. When circulating at very high concentrations, 25OHD can compete effectively with 1,25(OH)2D for binding to the VDR. Therefore, during vitamin D toxicosis, 25OHD can induce actions usually attributed to 1,25(OH)2D [193]. High circulating 25OHD can, therefore, explain how humans with low circulating concentrations of 1,25(OH)2D can show signs of vitamin D toxicity [193], and why anephric humans [who are incapable of producing 1,25(OH)2D] can become vitamin D toxic [194]. Clinical aspects of vitamin D toxicity are discussed in Chapter 78. Although it is generally accepted that 1,25(OH)2D is reduced during hypervitaminosis, a notable exception to this generalization is the ruminant [195,196]. Horst and co-workers [195,197] have shown that vitamin D3 intoxication initiated by giving 15 million IU of vitamin D3 intramuscularly (i.m.) results in significant elevations in plasma 1,25(OH)2D3. In contrast, pigs given the same i.m. dose showed a reduction in plasma 1,25(OH)2D3, as was observed in other species [197]. Therefore, elevations in plasma 1,25(OH)2D may play a significant role in the pathogenesis of vitamin D toxicity in ruminants.

B. Differences in Toxicity between Vitamins D2 and D3 Most research dealing with utilization of vitamins D2 and D3 assumes that these two forms are equally potent in most mammals. However, when large and potentially toxic doses were administered orally to rhesus monkeys [198] and horses [199], or were used in treating childhood osteodystrophy [200], vitamin D2

27

CHAPTER 2 Vitamin D Metabolism

presented fewer hypercalcemic side effects than vitamin D3. In addition, 1OHD2, which is as effective as the 1OHD3 in standard bioassays, was shown to be five- to 15-fold less toxic than 1OHD3 [201]. Studies by Horst et al. [44] provide some insight into the difference between vitamin D2 and vitamin D3 toxicity. They demonstrated that under physiological conditions the predominant monohydroxylated form of both vitamins D2 and D3 is 25OHD. In the vitamin D2–dosed rats, 24OHD2 accounted for approximately 20% of the monohydroxylated metabolites, whereas 24OHD3 could not be detected in the vitamin D3–dosed rats. When a modest superphysiological dose (800 IU/day) of vitamin D3 was given to rats, 25OHD3 remained the predominant metabolite in vitamin D3–dosed rats and was present at 26.3 ng/ml. Under these conditions, there was still no evidence for the presence of 24OHD3. However, when the same amount of vitamin D2 was given, the concentrations of 24OHD2 (14.1 ng/ml) nearly matched those of 25OHD2 (15.9 ng/ml). Interestingly, the combined concentrations of 24OHD2 and 25OHD2 in the vitamin D2–dosed animals (~30 ng/ml) was similar to the 25OHD3 concentration (26.3 ng/ml) in the vitamin D3–dosed rat. In standard assays, 25OHD2 and 25OHD3 are equipotent at displacing 3H-1,25(OH)2D3 from the calf thymus VDR. However, 24OHD2 has been shown to have at least a twofold lower affinity for binding to the calf thymus VDR (R. L. Horst, unpublished data, 1996). Therefore, the reduced toxicity of vitamin D2 is probably a result of diverting metabolism away from the production of 25OHD2 in favor of 24OHD2, which has a relatively limited affinity for VDR (a step necessary for the initiation of a biological response). Further differences between vitamins D2 and D3 were noted in their ability to up-regulate the VDR. Beckman et al. [202] found that VDR was significantly more enhanced in animals fed excess vitamin D3 relative to those animals receiving an equivalent amount of vitamin D2. Increased VDR would potentially accentuate toxic side effects by enhancing the responsiveness of intestinal tissue to the elevated 25OHD.

C. Factors Affecting Toxicity The severity of the effects and pathogenic lesions in vitamin D intoxication depend on such factors as the type of vitamin D (vitamin D2 versus vitamin D3), the dose, the functional state of the kidneys, and the composition of the diet. Vitamin D toxicity is enhanced by a rich dietary supply of calcium and phosphorus, and it is reduced when the diet is low in calcium and phosphorus [203,204]. Toxicity is also reduced when the

vitamin is accompanied by high intakes of vitamin A or by thyroxine injections [205]. The route of administration also influences toxicity. Parenteral administration of 15 million IU of vitamin D3 in a single dose caused toxicity and death in many pregnant dairy cows [195]. On the other hand, oral administration of 20 to 30 million IU of vitamin D2 daily for 7 days resulted in little or no toxicity in pregnant dairy cows [206]. Napoli et al. [207] have shown that rumen microbes are capable of metabolizing vitamin D to the inactive 10-keto-19-norvitamin D3. Parenteral administration would circumvent the deactivation of vitamin D by rumen microbes and may partially explain the difference in toxicity between oral and parenteral vitamin D. Various measures have been used in human medicine for treatment of vitamin D toxicity. These measures are mainly concerned with management of hypercalcemia. Vitamin D withdrawal is obviously indicated. It is usually not immediately successful, however, owing to the long plasma half-life of vitamin D (5 to 7 days) and 25OHD (20 to 30 days). This is in contrast to the short plasma half-life of 1OHD3 (1 to 2 days) and 1,25(OH)2D3 (4 to 8 hr). Because intestinal absorption of calcium contributes to hypercalcemia, a prompt reduction in dietary calcium is indicated. Sodium phytase, an agent that reduces intestinal calcium absorption, has also been used successfully in vitamin D toxicity management in monogastrics [208]. This treatment would be of little benefit to ruminants because of the presence of rumen microbial phytases. There have also been reports that CT [209], glucagon [210], etidronate [211], and glucocorticoid therapy [212] reduce serum calcium levels or prevent the calcinosis resulting from vitamin D intoxication (see Chapter 78).

IV. SPECIES VARIATION IN VITAMIN D METABOLISM AND ACTION Most concepts of vitamin D metabolism and function have been developed with the rat and/or chick as experimental models. Studying vitamin D metabolism is hampered by the paucity of data on the normal circulating levels of vitamin D metabolites in birds, mammals, and reptiles under normal conditions. Most recent research has focused on the analysis of 25OHD and 1,25(OH)2D as indicators of vitamin D status or aberrant physiological states. Table I summarizes the concentrations of the two metabolites that have been reported for several species by various laboratories. Close inspection of the information suggests that some mammals (mole rat, wild wood vole, horse, and wild wood mouse) and aquatic species (lamprey, carp,

28

RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY

TABLE I Plasma 25-Hydroxyvitamin D and 1,25-Dihydroxyvitamin D Concentrations in Several Species of Animals Concentration Species

25OHD (ng/ml) 1,25(OH)2D (pg/ml) Ref.

Human Rhesus monkey Rhesus monkey Marmoset Marmoset Wild woodmouse Wild bank vole Mole rat Lamprey Shark Leopard shark Horned shark Carp Bastard halibut Atlantic cod Bullfrog (mature) Soft-shelled turtle Turkey Chicken Cow Sheep Pig Horse aND,

32 188 50 90 64 <5 <5 <2 NDa ND 56 33 ND ND <2 2 16 26 27 43 27 76 6

31 207 95 400 640 <10 <5 17 274 87 3 6 174 192 59 21 12 52 21 38 36 60 19

[162] [231] [232] [232] [233] [234] [234] [213] [235] [235] [236] [236] [235] [235] [237] [235] [235] [141] [141] [141] [141] [141] [238]

Not done.

halibut, and bullfrog) appear to have very low or undetectable concentrations of 25OHD, and yet these animals appeared to be normal with no evidence of vitamin D deficiency. It is questionable whether some of these species have a requirement for vitamin D. The damara mole rat, for example, is a subterranean herbivore that in its natural habitat has no access to any obvious source of vitamin D and consumes a diet of roots and tubers [213]. These animals exhibit a high apparent calcium absorption efficiency (91%) and, like the horse and rabbit, actually use renal calcium excretion as the major regulator of calcium homeostasis [214,215]. In studies with rabbits consuming adequate amounts of calcium, it is very difficult to develop any overt or histological signs of vitamin D deficiency, and vitamin D may play a minor, if any, role in normal day-to-day functions in these animals [216].

In the wild, most animals do not have a dietary need for vitamin D, as sufficient vitamin D3 can be synthesized in the skin on irradiation by sunlight. However, indoor confinement of humans and other animals has resulted in the diet becoming the main source of vitamin D, leading to considerable research to determine the amount of dietary vitamin D required to substitute for lack of exposure to sunlight. Photochemically produced vitamin D3 enters the circulation and becomes immediately available, whereas dietary vitamin D3 may undergo modifications prior to becoming available for use by the body. One species where significant modification of vitamin D occurs before absorption is the ruminant. Within 24 hr, as much as 80% of vitamin D can undergo metabolism in vitro in rumen incubation media [217]. At least four metabolites are produced by the rumen microbes [217,218]. Two of these metabolites have been identified as the cis (5Z) and trans (5E) isomers of 10-keto-19-norvitamin D3 (Fig. 2) [207]. The trans isomer has also been identified in cow plasma (R. L. Horst, unpublished data, 1983). Neither compound has agonistic activity with regard to promoting bone calcium resorption [219] or intestinal calcium absorption [207]. Rather, this novel metabolism is likely a detoxification process, as evidenced by the ability of ruminants to tolerate large oral doses of vitamin D3 that would be toxic if given parenterally. The presence of the rumen, therefore, represents a major control point in vitamin D metabolism that may differ from monogastrics. Such a control point may have survival value, because the ruminant evolved as a grazing animal with the opportunity for long periods of sunlight exposure, as well as consumption of large quantities of irradiated plants. If left uncontrolled, such a combination might result in vitamin D toxicity. Shortly after the discovery of vitamin D, it seemed apparent that vitamins D2 and D3 had similar biological activities in most mammals and that birds and New World monkeys discriminated against vitamin D2 in favor of vitamin D3 [220,221]. More recent research, fostered by the discovery of sensitive analytical techniques and the availability of high specific activity 3H-labeled vitamin D species, indicated that differences in the metabolism of vitamins D2 and D3 in mammals are perhaps widespread. Most notable were the apparent discrimination against vitamin D2 by pigs [222], cows [218], and humans [223] and the apparent preference for vitamin D2 by rats [222,224]. Vitamin D and its metabolites are transported in the blood of vertebrates attached to a specific protein commonly known as the vitamin D binding protein or DBP [225]. Baird et al. [226] have shown that protein binding increases the solubility of steroids and that the

29

CHAPTER 2 Vitamin D Metabolism

metabolic clearance rate of steroids is in part dependent on their binding to specific plasma proteins. Affinity of metabolites to the plasma transport proteins may, therefore, provide a means for determining which species would utilize vitamin D2 poorly. For example, if the binding protein showed lower affinity toward 25OHD2 relative to 25OHD3, then one would predict that 25OHD2 would be removed from the circulation faster than 25OHD3. This is indeed the case for the chick. Hoy et al. [227] showed that chick discrimination against vitamin D2 was probably a result of enhanced clearance of the vitamin D2 metabolites 25OHD2 and 1,25(OH)2D2, and that the enhanced clearance was associated with weaker binding of these vitamin D2 metabolites (relative to the vitamin D3 forms) to DBP. In one of the most comprehensive studies reported to date, Hay and Watson [228] studied the affinities of DBP for 25OHD2 and 25OHD3 in 63 vertebrate species. They found that the DBP in fish, reptiles, and birds discriminated against 25OHD2 in favor of 25OHD3, which is consistent (at least in birds) with the discrimination against vitamin D2. One notable exception to this hypothesis, however, is the New World monkey. Hay and Watson [228] found that in New World monkeys, the plasma transport protein has equal affinity for 25OHD2 and 25OHD3, which is inconsistent with the well-documented discrimination against vitamin D2. Factors other than affinity of the binding protein for 25OHD are, therefore, important in determining how efficiently the different forms of vitamin D can be utilized by animals. Another example of species discrimination against the different vitamin D forms is in the rat. However, in this species, discrimination is against vitamin D3 in favor of vitamin D2 [222]. The rat DBP is known to have equal affinity for 25OHD2 and 25OHD3, but a lower affinity for vitamin D2 relative to vitamin D3 [229]. Reddy et al. [230] suggested that the lower affinity for vitamin D2 resulted in its enhanced availability for liver 25-hydroxylation. Hence, in the presence of DBP, more 25OHD2 was made relative to 25OHD3 when equal amounts of vitamin D2 or vitamin D3 substrate were perfused into rat livers. This observation is consistent with the higher circulating concentrations of 25OHD2 observed in acute experiments with vitamin D–deficient rats dosed with equal amounts of vitamins D2 and D3 [222]. In the experiments conducted by Reddy et al. [230], if binding protein was eliminated from the perfusion media, equal amounts of 25OHD2 and 25OHD3 were synthesized. Collectively, these data suggest that discrimination against the different forms of vitamin D could likely result from variations in the affinity of DBP for the parent compound and/or one or

more of their metabolites. Regardless of the mechanism for discrimination, it appears that these differences are present to afford animals the most efficient utilization of the most abundant antirachitic agents available in their environment.

V. CONCLUSION Vitamin D metabolism still remains an exciting area of research with much more to be learned. Critical questions remain unanswered regarding complete elucidation of the vitamin D2 metabolic pathway, and species differences between vitamins D2 and D3 metabolism are still virtually unexplored. The introduction of vitamin D analogs has also resulted in a totally different set of issues regarding their metabolism, tissue kinetics, mechanism of action, and potential therapeutic uses.

References 1. Mellanby E 1919 An experimental investigation on rickets. Lancet I 4985:407–412. 2. McCollum EV, Simmonds N, Becker JE, Shipley PG 1922 Studies on experimental rickets. XXI. An experimental demonstration of the existence of a vitamin which promotes calcium deposition. J Biol Chem 53:293–312. 3. Norman AW 1979 Vitamin D: The calcium homeostaic hormone. In: Darby WJ (ed) Nutrition: Basic and Applied Science. Academic Press, New York. 4. Napoli JL, Horst RL 1984 Vitamin D metabolism. In: Kumar R (ed) Vitamin D: Basic and Applied Aspects. Martinus Nijhoff, Boston, pp. 91–123. 5. Lawson DEM 1978 Vitamin D. Academic Press, London. 6. Kodichek E 1974 The story of vitamin D from vitamin to hormone. Lancet I 7853:325–329. 7. Haussler MR, McCain TA 1977 Basic and clinical concepts related to vitamin D metabolism and action. N Engl J Med 297:974–983, 1041–1050. 8. Deluca HF, Schnoes HK 1976 Metabolism and action of vitamin D. Annu Rev Biochem 45:631–666. 9. IUPAC 1969 IUPAC Commission on the Nomenclature of Organic Chemistry and IUPAC–IUB Commission on Biochemical Nomenclature. Revised tentative rules for nomenclature of steroids. Biochem J 113:5–28. 10. Cahn RS, Ingold C, Prelog V 1966 Specification of molecular chirality. Angew Chem, Int Ed Engl 5:385–415. 11. Blackwood JE, Gladys CL, Loening KL, Petrarca AE, Rush JE 1968 Unambiguous specification of stereoisomerism about a double bond. J Am Chem Soc 90:509–510. 11a. Morii H, Lund J, Neville PF, DeLuca HF 1967 Biological activity of a vitamin D metabolite. Arch Biochem Biophys 120:508–512. 11b. Blunt JW, DeLuca HF, Schnoes HK 1968 25-Hydroxycholecalciferol: A biologically active metabolite of vitamin D3. Biochemistry 7:3317–3322. 11c. Norman AW, Myrtle JF, Midgett RJ, Nowicki HG 1971 1,25-Dihydroxycholecalciferol: Identification of the

30

12.

13.

14.

15. 16. 17.

18. 19.

20. 21.

22. 23.

24. 25. 26. 27. 28. 29.

RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY

proposed active form of vitamin D3 in the intestine. Science 173:51–54. Lawson DEM, Fraser DR, Kodicek E, Morris HR, Williams DH 1971 Identification of 1,25-dihydroxycholecalciferol, a new kidney hormone controlling calcium metabolism. Nature 230:228–230. Holick MF, Schnoes HK, DeLuca HF, Suda T, Cousins RJ 1971 Isolation and identification of 1,25-dihydroxycholecalciferol. A metabolite of vitamin D active in intestine. Biochemistry 10:2799–2804. Broadus AE, Horst RL, Lang R, Littledike ET, Rasmussen H 1980 The importance of circulating 1,25-dihydroxyvitamin D in the pathogenesis of hypercalciuria and renal-stone formation in primary hyperparathyroidism. N Engl J Med 302:421–426. Bordier P, Rasmussen H, Marie P, Miravet L, Gueris J, Ryckwaert A 1978 Vitamin D metabolism and bone mineralization in man. J Clin Endocrinol Metab 46:284–294. Ornoy A, Goodwin D, Noff D, Edelstein S 1978 24,25Dihydroxyvitamin D is a metabolite of vitamin D essential for bone formation. Nature 276:517–519. Canterbury JM, Lerman S, Claflin AJ, Henry H, Norman A, Reiss E 1977 Inhibition of parathyroid hormone secretion by 25-hydroxycholecalciferol and 24,25-dihydroxycholecalciferol in the dog. J Clin Invest 78:1375–1383. Henry HL, Norman AW 1978 Vitamin D: Two dihydroxylated metabolites are required for normal chicken egg hatchability. Science 201:835–837. St-Arnaud R, Arabian A, Travers R, Barletta F, Raval-Pandya M, Chapin K, Depovere J, Mathieu C, Christakos S, Demay MB, Glorieux FH 2000 Deficient mineralization of intramembranous bone in vitamin D-24-hydroxylase-ablated mice is due to elevated 1,25-dihydroxyvitamin D and not to the absence of 24,25-dihydroxyvitamin D. Endocrinology 141:2658–2666. Horst RL, Littledike ET, Riley JL, Napoli JL 1981 Quantitation of vitamin D and its metabolites and their plasma concentrations in five species of animals. Anal Biochem 116:189–203. Hartwell D, Hassager C, Christiansen C 1987 Effect of vitamin D2 and vitamin D3 on the serum concentrations of 1,25(OH)2D2 and 1,25(OH)2D3 in normal subjects. Acta Endocrinol 115:378–384. Hollis BW, Pittard WB 1984 Relative concentrations of 25-hydroxyvitamin D2/D3 and 1,25-dihydroxyvitamin D2/D3 in maternal plasma at delivery. Nutr Res (N Y) 4:27–32. Reinhardt TA, Horst RL, Orf JW, Hollis BW 1984 A microassay for 1,25-dihydroxyvitamin D not requiring high performance liquid chromatography: Application to clinical studies. J Clin Endocrinol Metab 58:91–98. Jones G, Schnoes HK, DeLuca HF 1975 Isolation and identification of 1,25-dihydroxyvitamin D2. Biochemistry 14:1250–1256. Tucker G, Gaganon R, Haussler M 1973 Vitamin D3-25hydroxylase: Tissue occurrence and apparent lack of regulation. Arch Biochem Biophys 155:47–57. Ponchon G, DeLuca HF 1969 Metabolites of vitamin D3 and their biological activity. J Nutr 99:157–167. Bhattacharyya MH, DeLuca HF 1973 The regulation of rat liver calciferol-25-hydroxylase. J Biol Chem 248: 2969–2973. Bhattacharyya MH, DeLuca HF 1973 Comparative studies on the 25-hydroxylation of vitamin D3 and dihydrotachysterol. J Biol Chem 248:2974–2977. Bjorkhem I, Holmberg I 1979 On the 25-hydroxylation of vitamin D3 in vitro studied with a mass fragmentographic technique. J Biol Chem 254:9518–9524.

30. Bjorkhem I, Holmberg I 1978 Assay and properties of a mitochondrial 25-hydroxylase active on vitamin D3. J Biol Chem 253:842–849. 31. Madhok TC, DeLuca HF 1979 Characteristics of the rat liver microsomal enzyme system converting cholecalciferol into 25-hydroxycholecalciferol. Evidence for the participation of cytochrome P-450. Biochem J 184:491–499. 32. Bjorkhem I, Holmberg I, Oftebro H, Pedersen JI 1980 Properties of a reconstituted vitamin D3 25-hydroxylase from rat liver mitochondria. J Biol Chem 255:5244–5249. 33. Okuda K 1992 Vitamin D3 hydroxylase—its occurrence, biological importance and biochemistry. In: Ruckpaul K, Rein H (eds) Cytochrome P-450 Dependent Biotransformation of Endogenous Substrates: Frontiers in Biotransformation, Vol. 6. Akademic Verlag, Berlin, pp.114–147. 34. Okuda K 1994 Liver mitochodrial p450 involved in cholesterol catabolism and vitamin activation. J Lipid Res 35:361–372. 35. Okuda K-I, Usui E, Ohyama Y 1995 Recent progress in enzymology and molecular biology of enzymes involved in vitamin D metabolism. J Lipid Res 36:1641–1652. 36. Postlind H, Axen E, Bergman T, Wikvall K 1997 Cloning, structure, and expression of a cDNA encoding vitamin D3 25-hydroxylase. Biochem Biophys Res Commun 241: 491–497. 37. Wikvall K 2001 Cytochrome P450 enzymes in the bioactivation of vitamin D to its hormonal form (review). Int J Mol Med 7:201–209. 38. Andersson S, Davis DL, Dahlbach H, Jornvall H 1989 Cloning, structure, and expression of the mitochondrial cytochrome P-450 sterol 26-hydroxylase, a bile acid biosynthetic enzyme. J Biol Chem 264:8222–8229. 39. Guo Y-D, Strugnell S, Back DW, Jones G 1993 Transfected human liver cytochrome P-450 hydroxylates vitamin D analogs at different side-chain positions. Proc Natl Acad Sci USA 90:8668–8672. 40. Cali J, Russell D 1991 Characterization of human sterol 27-hydroxylase. J Biol Chem 266:7774–7778. 41. Usui E, Noshiro M, Okuda K 1990 Molecular cloning of cDNA for vitamin D3 25-hydroxylase from rat liver mitochondria. FEBS Lett 262:135–138. 42. Usui E, Noshiro M, Ohyama Y, Okuda K 1990 Unique property of liver mitochondrial P450 to catalyze the two physiologically important reactions involved in both cholesterol catabolism and vitamin D activation. FEBS Lett 274: 175–177. 43. Jones G, Schnoes HK, Levan L, DeLuca HF 1980 Isolation and identification of 24-hydroxyvitamin D2 and 24,25-dihydroxyvitamin D2. Arch Biochem Biophys 202:450–457. 44. Horst RL, Koszewski NJ, Reinhardt TA 1990 1α-Hydroxylation of 24-hydroxyvitamin D2 represents a minor physiological pathway for the activation of vitamin D2 in mammals. Biochemistry 29:578–582. 45. Koszewski NJ, Reinhardt TA, Napoli JL, Beitz DC, Horst RL 1988 24,26-Dihydroxyvitamin D2: A unique physiological metabolite of vitamin D2. Biochemistry 27:5785–5790. 46. Rosen H, Reshef A, Maeda N, Lippoldt A, Shpizen S, Triger L, Eggertsen G, Bjorkhem I, Leitersdorf E 1998 Markedly reduced bile acid synthesis but maintained levels of cholesterol and vitamin D metabolites in mice with disrupted sterol 27-hydroxylase gene. J Biol Chem 273: 14805–14812. 47. Hayashi S, Usui E, Okuda K 1988 Sex-related difference in vitamin D3 25-hydroxylase of rat liver mitochondria. J Biochem (Tokyo) 103:863–866.

CHAPTER 2 Vitamin D Metabolism

48. Hosseinpour F, Wikvall K 2000 Porcine microsomal vitamin D(3) 25-hydroxylase (CYP2D25). Catalytic properties, tissue distribution, and comparison with human CYP2D6. J Biol Chem 275:34650–34655. 49. Saarem K, Bergseth S, Oftebro H, Petersen JI 1984 Subcellular localization of vitamin D3 25-hydroxylation in human liver. J Biochem (Tokyo) 259:10936–10946. 50. Axen E, Bergman T, Wikvall K 1994 Microsomal 25-hydroxylation of vitamin D2 and vitamin D3 in pig liver. J Steroid Biochem Mol Biol 51:97–106. 51. Hosseinpour F, Ibranovic I, Tang W, Wikvall K 2003 25-Hydroxylation of vitamin D3 in primary cultures of pig hepatocytes: evidence for a role of both CYP2D25 and CYP27A1. Biochem Biophys Res Commun 303:877–883. 52. Haussler MR, Myrtle JF, Norman AW 1968 The association of a metabolite of vitamin D3 with intestinal mucosa chromatin in vivo. J Biol Chem 243:4055–4064. 53. Lawson DEM, Wilson PW, Kodicek E 1969 Metabolism of vitamin D. A new cholecalciferol metabolite, involving loss of hydrogen at C-1, in chick intestinal nuclei. Biochem J 115:269–277. 54. Kodicek E, Lawson DEM, Wilson PW 1970 Biological activity of a polar metabolite of vitamin D3. Nature 228: 763–764. 55. Haussler MR, Boyce DW, Littledike ET, Rasmussen H 1971 A rapidly acting metabolite of vitamin D3. Proc Natl Acad Sci USA 68:177–181. 56. Fraser DR, Kodicek E 1970 Unique biosynthesis by kidney of a biologically active vitamin D metabolite. Nature 228:764–766. 57. Holick MF, Schnoes HK, Deluca HF 1971 Identification of 1,25-dihydroxycholecalciferol, a form of vitamin D3 metabolically active in the intestine. Proc Natl Acad Sci USA 68:803–804. 58. Paulson SK, Deluca HF 1985 Subcellular location and properties of rat renal 25-hydroxyvitamin D3-1α-hydroxylase. J Biol Chem 260:11488–11492. 59. Armbrecht HJ, Okuda K, Wongsurawat N, Nemani RK, Chen ML, Boltz MA 1992 Characterization and regulation of the vitamin D hydroxylases. J Steroid Biochem 43: 1073–1081. 60. Gray RW, Weber HP, Dominguez JH, Lemann J, Jr. 1974 The metabolism of vitamin D3 and 25-hydroxyvitamin D3 in normal and anephric humans. J Clin Endocrinol Metab 39:1045–1056. 61. DeLuca HF 1974 Vitamin D: The vitamin and the hormone. Fed Proc 33:2211–2219. 62. Boyle I, Gray R, Deluca HF 1971 Regulation by calcium of in vivo synthesis of 1,25-dihydroxycholecalciferol and 21,25-dihydroxycholecalciferol. Proc Natl Acad Sci USA 68:2131–2137. 63. Armbrecht HJ, Wongsurawat N, Zenser TV, Davis BB 1984 Effect of PTH and 1,25-(OH)2D3 on renal 25(OH)D3 metabolism, adenylate cyclase, and protein kinase. Am J Physiol 246:E102–E107. 64. Armbrecht HJ, Wongsurawat N, Zenser TV, Davis BB 1982 Differential effects of parathyroid hormone on the renal 1,25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3 production of young and adult rats. Endocrinology 111:1339–1344. 65. Horst RL, Eisman JA, Jorgensen NA, DeLuca HF 1977 Adequate response of plasma 1,25-dihydroxyvitamin D to parturition in paretic (milk fever) dairy cows. Science 196:662–663.

31 66. Hughes MR, Brumbaugh PF, Haussler MR, Wergedal JE, Baylink DJ 1975 Regulation of serum 1α,25-dihydroxyvitamin D3 by calcium and phosphate in the rat. Science 190:578–580. 67. Henry HL 1979 Regulation of the hydroxylation of 25-hydroxyvitamin D3 in vivo and in primary cultures of chick kidney cells. J Biol Chem 254:2722–2729. 68. Omdahl JL 1978 Interaction of the parathyroid and 1,25dihydroxyvitamin D3 in the control of renal 25-hydroxyvitamin D3 metabolism. J Biol Chem 253:8474–8487. 69. Henry HL 1981 Insulin permits parathyroid hormone stimulation of 1,25-dihydroxyvitamin D3 production in cultured kidney cells. Endocrinology 108:733–735. 70. Armbrecht HJ, Forte LR, Wongsurawat N, Zenser TV, Davis BB 1984 Forskolin increases 1,25-dihydroxyvitamin D3 production by rat renal slices in vitro. Endocrinology 114:644–649. 71. Aarskog D, Aksnes L 1980 Acute response of plasma 1,25-dihydroxyvitamin D to parathyroid hormone. Lancet 1(8164):362–363. 72. Hove K, Horst RL, Littledike ET, Beitz DC 1984 Infusions of parathyroid hormone in ruminants: Hypercalcemia and reduced plasma 1,25-dihydroxyvitamin D concentrations. Endocrinology 114:897–903. 73. Hustmyer FG, Beitz DC, Goff JP, Nonnecke BJ, Horst RL, Reinhardt TA 1994 Effects of in vivo administration of 1,25-dihydroxyvitamin D3 on in vitro proliferation of bovine lymphocytes. J Dairy Sci 77:3324–3330. 74. Reinhardt TA, Horst RL 1990 Parathyroid hormone down-regulates 1,25-dihydroxyvitamin D receptors (VDR) and VDR messenger ribonucleic acid in vitro and blocks homologous up-regulation of VDR in vivo. Endocrinology 127:942–948. 75. Henry HL, Midgett RJ, Norman AW 1974 Regulation of 25-hydroxyvitamin D3-1-hydroxylase in vivo. J Biol Chem 249:7584–7592. 76. Tanaka Y, DeLuca HF 1973 The control of 25-hydroxyvitamin D metabolism by inorganic phosphorus. Arch Biochem Biophys 154:566–574. 77. Gray RW 1987 Evidence that somatomedins mediate the effect of hypophosphatemia to increase serum 1,25-dihydroxyvitamin D3 levels in rats. Endocrinology 121:504–512. 78. Armbrecht HJ, Zenser TV, Davis BB 1982 Modulation of renal production of 24,25- and 1,25-dihydroxyvitamin D3 in young and adult rats by dietary calcium, phosphorus, and 1,25-dihydroxyvitamin D3. Endocrinology 110: 1983–1988. 79. Brenza HL, DeLuca HF 2000 Regulation of 25-hydroxyvitamin D3 1α-hydroxylase gene expression by parathyroid hormone and 1,25-dihydroxyvitamin D3. Arch Biochem Biophys 381:143–152. 80. Slatopolsky E, Lopez-Hilker S, Delmez J, Dusso A, Brown A, Martin KJ 1990 The parathyroid-calcitriol axis in health and chronic renal failure. Kidney Int 38:S41–S47. 81. Beckman MJ, Johnson JA, Goff JP, Reinhardt TA, Beitz DC, Horst RL 1995 The role of dietary calcium in the physiology of vitamin D toxicity: Excess dietary vitamin D3 blunts parathyroid hormone induction of kidney 1-hydroxylase. Arch Biochem Biophys 319:535–539. 82. Leheste JR, Melsen F, Wellner M, Jansen P, Schlichting U, Renner-Muller I, Andreassen TT, Wolf E, Bachmann S, Nykjaer A, Willnow TE 2003 Hypocalcemia and osteopathy in mice with kidney-specific megalin gene defect. FASEB J 17:247–249.

32

RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY

83. Kumar R 1984 Metabolism of 1,25-dihydroxyvitamin D3. Physiol Rev 64:478–504. 84. Henry HL 1992 Vitamin D hydroxylases. J Cell Biochem 49:4–9. 85. Portale AA, Miller WL 2000 Human 25-hydroxyvitamin D-1α-hydroxylase: cloning, mutations, and gene expression. Pediatr Nephrol 14:620–625. 86. Fu GK, Portale AA, Miller WL 1997 Complete structure of the human gene for the vitamin D 1α-hydroxylase, P450c1α. DNA Cell Biol 16:1499–1507. 87. Glorieux FH, St-Arnaud R 1998 Molecular cloning of (25-OH D)-1α-hydroxylase: an approach to the understanding of vitamin D pseudo-deficiency. Recent Prog Horm Res 53:341–349; discussion 350. 88. Shinki T 1998 Cloning and expression of 25-hydroxyvitamin D3-1α-hydroxylase gene. Nippon Rinsho 56:1440–1446. 89. Yoshida T, Yoshida N, Nakamura A, Monkawa T, Hayashi M, Saruta T 1999 Cloning of porcine 25-hydroxyvitamin D3 1α-hydroxylase and its regulation by cAMP in LLC-PK1 cells. J Am Soc Nephrol 10:963–970. 90. Panda DK, Miao D, Tremblay ML, Sirois J, Farookhi R, Hendy GN, Goltzman D 2001 Targeted ablation of the 25-hydroxyvitamin D 1α-hydroxylase enzyme: evidence for skeletal, reproductive, and immune dysfunction. Proc Natl Acad Sci USA 98:7498–7503. 91. Dardenne O, Prud’homme J, Arabian A, Glorieux FH, St-Arnaud R 2001 Targeted inactivation of the 25-hydroxyvitamin D3-1α-hydroxylase gene (CYP27B1) creates an animal model of pseudovitamin D-deficiency rickets. Endocrinology 142:3135–3141. 92. Holick MF, Schnoes HK, DeLuca HF, Gray RW, Boyle IT, Suda T 1972 Isolation and identification of 24,25-dihydroxycholecalciferol, a metabolite of vitamin D3 made in the kidney. Biochemistry 11:4251–4255. 93. Kumar R, Schnoes HK, DeLuca HF 1978 Rat intestinal 25-hydroxyvitamin D3- and 1α,25-dihydroxyvitamin D324-hydroxylase. J Biol Chem 253:3804–3809. 94. Kumar R, Harnden D, DeLuca HF 1976 Metabolism of 1,25-dihydroxyvitamin D3: Evidence for side-chain oxidation. Biochemistry 15:2420–2423. 95. Iwata K, Yamamoto A, Satoh S, Ohyama Y, Tashiro Y, Setoguchi T 1995 Quantitative immunoelectron microscopic analysis of the localization and induction of 25-hydroxyvitamin D3 24-hydroxylase in rat kidney. J Histochem Cytochem 43:255–262. 96. Shinki T, Jin CH, Nishimura A, Nagai Y, Ohyama Y, Noshiro M, Okuda K, Suda T 1992 Parathyroid hormone inhibits 25-hydroxyvitamin D3-24-hydroxylase mRNA expression stimulated by 1α,25-dihydroxyvitamin D3 in rat kidney but not in intestine. J Biol Chem 267:13757–13762. 97. Nishimura A, Shinki T, Jin CH, Ohyama Y, Noshiro M, Okuda K, Suda T 1994 Regulation of messenger ribonucleic acid expression of 1α,25-dihydroxyvitamin D3-24-hydroxylase in rat osteoblasts. Endocrinology 134:1794–1799. 98. Shigematsu T, Horiuchi N, Ogura Y, Miyahara T, Suda T 1986 Human parathyroid hormone inhibits renal 24-hydroxylase activity of 25-hydroxyvitamin D3 by a mechanism involving adenosine 3′,5′-monophosphate in rats. Endocrinology 118: 1583–1589. 99. Goff JP, Reinhardt TA, Beckman MJ, Horst RL 1990 Contrasting effects of exogenous 1,25-dihydroxyvitamin D [1,25-(OH)2D] versus endogenous 1,25-(OH)2D, induced by dietary calcium restriction, on vitamin D receptors. Endocrinology 126:1031–1035.

100. Iida K, Shinki T, Yamaguchi A, DeLuca HF, Kurokawa K, Suda T 1995 A possible role of vitamin D receptors in regulating vitamin D activation in the kidney. Proc Natl Acad Sci USA 92:6112–6116. 101. Hanai H, Liang CT, Cheng L, Sacktor B 1989 Desensitization to parathyroid hormone in renal cells from aged rats is associated with alterations in G-protein activity. J Clin Invest 83:268–277. 102. Hanai H, Brennan DP, Cheng L, Goldman ME, Chorev M, Levine MA, Sacktor B, Liang CT 1990 Downregulation of parathyroid hormone receptors in renal membranes from aged rats. Am J Physiol 259:F444–F450. 103. Johnson JA, Beckman MJ, Pansini-Porta A, Christakos S, Bruns ME, Beitz DC, Horst RL, Reinhardt TA 1995 Age and gender effects on 1,25-dihydroxyvitamin D3-regulated gene expression. Exp Gerontol 30:631–643. 104. Matkovits T, Christakos S 1995 Variable in vivo regulation of rat vitamin D-dependent genes (osteopontin, Ca,Mg-adenosine triphosphatase, and 25-hydroxyvitamin D3 24-hydroxylase): Implications for differing mechanisms of regulation and involvement of multiple factors. Endocrinology 136: 3971–3982. 105. Goff JP, Reinhardt TA, Engstrom GW, Horst RL 1992 Effect of dietary calcium or phosphorus restriction and 1,25-dihydroxyvitamin D administration on rat intestinal 24-hydroxylase. Endocrinology 131:101–104. 106. Beckman MJ, Goff JP, Reinhardt TA, Beitz DC, Horst RL 1994 In vivo regulation of rat intestinal 24-hydroxylase: Potential new role of calcitonin. Endocrinology 135: 1951–1955. 107. Ohyama Y, Noshiro M, Okuda K 1991 Cloning and expression of cDNA encoding 25-hydroxyvitamin D3 24-hydroxylase. FEBS Lett 278:195–198. 108. Ohyama Y, Hayashi S, Okuda K 1989 Purification of 25-hydroxyvitamin D3 24-hydroxylase from rat kidney mitochondria. FEBS Lett 255:405–408. 109. Chen K-S, DeLuca HF 1995 Cloning of the human 1α,25dihydroxyvitamin D-3 24-hydroxylase gene promoter and identification of two vitamin D-responsive elements. Biochim Biophys Acta 1263:1–9. 110. Ohyama Y, Noshiro M, Eggertsen G, Gotoh O, Kato Y, Bjorkhem I, Okuda K 1993 Structural characterization of the gene encoding rat 25-hydroxyvitamin D3 24-hydroxylase. Biochemistry 32:76–82. 111. Horst RL, Littledike ET, Gray RW, Napoli JL 1981 Impaired 24,25-dihydroxyvitamin D production in anephric human and pig. J Clin Invest 67:274–280. 112. Taniguchi T, Eto TA, Shiotsuki H, Sueta H, Higashi S, Iwamura T, Okuda KI, Setoguchi T 2001 Newly established assay method for 25-hydroxyvitamin D3 24-hydroxylase revealed much lower Km for 25-hydroxyvitamin D3 than for 1α,25-dihydroxyvitamin D3. J Bone Miner Res 16:57–62. 113. Jones G, Tenenhouse HS 2002 1,25(OH)2D, the preferred substrate for CYP24. J Bone Miner Res 17:179–181. 114. Burgos-Trinidad M, DeLuca HF 1991 Kinetic properties of 25-hydroxyvitamin D- and 1,25-dihydroxyvitamin D-24hydroxylase from chick kidney. Biochim Biophys Acta 1078:226–230. 115. Napoli JL, Pramanik BC, Royal PM, Reinhardt TA, Horst RL 1983 Intestinal synthesis of 24-keto-1,25-dihydroxyvitamin D3. A metabolite formed in vivo with high affinity for the vitamin D cytosolic receptor. J Biol Chem 258:9100–9107. 116. Napoli JL, Horst RL 1983 C(24)- and C(23)-oxidation, converging pathways of intestinal 1,25-dihydroxyvitamin D3

33

CHAPTER 2 Vitamin D Metabolism

117. 118. 119. 120.

121.

122.

123.

124.

125. 126.

127.

128. 129.

130.

131.

132.

metabolism: Identification of 24-keto-1,23,25-trihydroxyvitamin D3. Biochemistry 22:5848–5853. Haussler MR 1986 Vitamin D receptors: Nature and function. Annu Rev Nutr 6:527–562. Lohnes D, Jones G 1987 Side chain metabolism of vitamin D3 in osteosarcoma cell line UMR-106. Characterization of products. J Biol Chem 262:14394–14401. Reddy GS, Tserng KY 1989 Calcitroic acid, end product of renal metabolism of 1,25-dihydroxyvitamin D3 through C-24 oxidation pathway. Biochemistry 28:1763–1769. Makin G, Lohnes D, Byford V, Ray R, Jones G 1989 Target cell metabolism of 1,25-dihydroxyvitamin D3 to calcitroic acid: Evidence for a pathway in kidney and bone involving 24-oxidation. Biochem J 262:173–180. Akiyoshi-Shibata M, Sakaki Y, Ohyama Y, Noshiro M, Okuda K, Yabusaki Y 1994 Further oxidation of hydroxycalcidiol by calcidiol 24-hydroxylase. Eur J Biochem 224:335–343. Sakaki T, Sawada N, Komai K, Shiozawa S, Yamada S, Yamamoto K, Ohyama Y, Inouye K 2000 Dual metabolic pathway of 25-hydroxyvitamin D3 catalyzed by human CYP24. Eur J Biochem 267:6158–6165. Sakaki T, Sawada N, Nonaka Y, Ohyama Y, Inouye K 1999 Metabolic studies using recombinant Escherichia coli cells producing rat mitochondrial CYP24. CYP24 can convert 1α,25-dihydroxyvitamin D3 to calcitroic acid. Eur J Biochem 262:43–48. Inouye K, Sakaki T 2001 Enzymatic studies on the key enzymes of vitamin D metabolism; 1α-hydroxylase (CYP27B1) and 24-hydroxylase (CYP24). Biotechnol Annu Rev 7:179–194. Beckman MJ, Tadikonda P, Werner E, Prahl J, Yamada S, DeLuca HF 1996 Human 25-hydroxyvitamin D3-24-hydroxylase, a multicatalytic enzyme. Biochemistry 35:8465–8472. Pols HAP, van Leeuwen JPTM, Schilte JP, Visser TJ, Birkenhager JC 1988 Heterologous up-regulation of the 1,25-dihydroxyvitamin D3 receptor by parathyroid hormone (PTH) and PTH-like peptide in osteoblast-like cells. Biochem Biophys Res Commun 156:588–594. Pols HAP, Birkenhager JC, Schilte JP, Visser TJ 1988 Evidence that the self-induced metabolism of 1,25-dihydroxyvitamin D3 limits the homologous up-regulation of its receptor in rat osteosarcoma cells. Biochim Biophys Acta 970:122–129. Reinhardt TA, Horst RL 1989 Self-induction of 1,25-dihydroxyvitamin D3 metabolism limits receptor occupancy and target tissue responsiveness. J Biol Chem 264:15917–15921. Reinhardt TA, Horst RL 1989 Ketoconazole inhibits selfinduced metabolism of 1,25-dihydroxyvitamin D3 and amplifies 1,25-dihydroxyvitamin D3 receptor up-regulation in rat osteosarcoma cells. Arch Biochem Biophys 272:459–465. Reinhardt TA, Horst RL, Engstrom GW, Atkins KB 1988 Ketoconazole potentiates 1,25-(OH)2D3-directed upregulation of 1,25-(OH)2D3-receptors in rat intestine and bone. In: Norman AW, Schaefer K, Grigoleit H-G, Herrath DV (eds) Vitamin D: Molecular, Cellular and Clinical Endocrinology. Walter de Gruyter, Berlin, pp. 233–234. Hosogane N, Shinki T, Kasuga H, Taketomi S, Toyama Y, Suda T 2003 Mechanisms for the reduction of 24,25-dihydroxyvitamin D3 levels and bone mass in 24-hydroxylase transgenic rats. FASEB J 17:737–739. Kasuga H, Hosogane N, Matsuoka K, Mori I, Sakura Y, Shimakawa K, Shinki T, Suda T, Taketomi S 2002 Characterization of transgenic rats constitutively expressing

133.

134.

135.

136.

137.

138.

139. 140.

141.

142. 143.

144.

145. 146.

147.

vitamin D-24-hydroxylase gene. Biochem Biophys Res Commun 297:1332–1338. Wichmann J, Schnoes HK, DeLuca HF 1981 Isolation and identification of 24(R)-hydroxyvitamin D3 from chicks given large doses of vitamin D3. Biochemistry 20: 2350–2353. Wichmann JK, Schnoes HK, DeLuca HF 1981 23,24, 25-Trihydroxyvitamin D3, 24,25,26-trihydroxyvitamin D3, 24-keto-25-hydroxyvitamin D3, and 23-dehydro-25-hydroxyvitamin D3: New in vivo metabolites of vitamin D3. Biochemistry 20:7385–7391. DeLuca HF, Schnoes HK 1979 Recent developments in the metabolism of vitamin D3. In: Norman AW (ed) Vitamin D: Basic Research and Its Clinical Application. Walter de Gruyter, Berlin, pp. 445–458. Horst RL 1979 25-OHD3-26,23-lactone: A metabolite of vitamin D3 that is 5 times more potent than 25-OHD3 in the rat plasma competitive protein binding radioassay. Biochem Biophys Res Commun 89:286–293. Wichmann JK, DeLuca HF, Schnoes HK, Horst RL, Shepard RM, Jorgensen NA 1979 25-Hydroxyvitamin D3 26,23lactone: A new in vivo metabolite of vitamin D. Biochemistry 18:4775–4780. Ohnuma N, Norman AW 1982 Production in vitro of 1α,25dihydroxyvitamin D3-26,23-lactone from 1α,25-dihydroxyvitamin D3 by rat small intestinal mucosa homogenates. Arch Biochem Biophys 213:139–147. Napoli JL, Horst RL 1983 (23S)-1,23,25-Trihydroxycholecalciferol, an intestinal metabolite of 1,25-dihydroxycholecalciferol. Biochem J 214:261–264. Napoli JL, Pramanik BC, Partridge JJ, Uskokovic MR, Horst RL 1982 23S,25-Dihydroxyvitamin D3 as a circulating metabolite of vitamin D3. Its role in 25-hydroxyvitamin D326,23-lactone biosynthesis. J Biol Chem 257:9634–9639. Horst RL, Littledike ET 1982 Comparison of plasma concentrations of vitamin D and its metabolites in young and aged domestic animals. Comp Biochem Physiol 73B: 485–489. Horst RL, Littledike ET 1980 25-OHD3-26,23 lactone: Demonstration of kidney-dependent synthesis in the pig and rat. Biochem Biophys Res Commun 93:149–154. Ishizuka S, Ohba T, Norman AW 1988 1,25-(OH)2D3-26,23lactone is a major metabolite of 1,25-(OH)2D3 under physiological conditions. In: Norman AW, Schafer K, Grigoleit H-G, Herrath DV (eds) Vitamin D: Molecular, Cellular and Clinical Endocrinology. Walter de Gruyter, Berlin, pp. 143–144. Ishizuka S, Ishimoto S, Norman AW 1984 Isolation and identification of 1α,25-dihydroxy-24-oxovitamin D3, 1α,25dihydroxyvitamin D3 26,23-lactone, and 1α,24(S),25-trihydroxyvitamin D3: In vivo metabolites of 1α,25-dihydroxyvitamin D3. Biochemistry 23:1473–1478. Napoli JL, Horst RL 1982 23,25,26-Trihydroxycholecalciferol. A 25-hydroxycholecalciferol-26,23-lactone precursor. Biochem J 206:173–176. Yamada S, Nakayama K, Takayama H, Shinki T, Takasaki Y, Suda T 1984 Isolation, identification, and metabolism of (23S,25R)-25-hydroxyvitamin D3 26,23-lactol. A biosynthetic precursor of (23S,25R)-25-hydroxyvitamin D3 26,23-lactone. J Biol Chem 259:884–889. Horst RL, Pramanik BC, Reinhardt TA, Shiuey SJ, Partridge JJ, Uskokovic MR, Napoli JL 1982 Binding properties of 23S,25dihydroxyvitamin D3: An in vivo metabolite of vitamin D3. Biochem Biophys Res Commun 106:1006–1011.

34

RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY

148. Horst RL, Wovkulich PM, Baggiolini EG, Uskokovic MR, Engstrom GW, Napoli JL 1984 (23S)-1,23,25-Trihydroxyvitamin D3: Its biologic activity and role in 1,25-dihydroxyvitamin D3 26,23-lactone biosynthesis. Biochemistry 23:3973–3979. 149. Esvelt RP, Schnoes HK, DeLuca HF 1979 Isolation and characterization of 1α-hydroxy-23-carboxytetranorvitamin D: A major metabolite of 1,25-dihydroxyvitamin D3. Biochemistry 18:3977–3983. 150. Horst RL, Reinhardt TA, Pramanik BC, Napoli JL 1983 23-Keto-25-hydroxyvitamin D3: A vitamin D3 metabolite with high affinity for the 1,25-dihydroxyvitamin D specific cytosol receptor. Biochemistry 22:245–250. 151. Horst RL, Reinhardt TA, Napoli JL 1982 23-Keto-25hydroxyvitamin D3 and 23-keto-1,25-dihydroxyvitamin D3: Two new vitamin D3 metabolites with high affinity for the 1,25-dihydroxyvitamin D3 receptor. Biochem Biophys Res Commun 107:1319–1325. 152. Suda T, DeLuca HF, Schnoes HK, Tanaka Y, Holick MF 1970 25,26-Dihydroxycholecalciferol, a metabolite of vitamin D3 with intestinal calcium transport activity. Biochemistry 9:4776–4780. 153. Reinhardt TA, Napoli JL, Pramanik B, Littledike ET, Beitz DC, Partridge JJ, Uskokovic MR, Horst RL 1981 1α,25,26-Trihydroxyvitamin D3: An in vivo and in vitro metabolite of vitamin D3. Biochemistry 20:6230–6235. 154. Partridge JJ, Shiuey SJ, Chadha HK, Baggiolini ET, Bolount JE, Uskokovic MR 1981 Synthesis and structure proof of a vitamin D3 metabolite, 25(S),26-dihydroxycholecalciferol. J Am Chem Soc 103:1253–1255. 155. Ikekawa N, Koizumi N, Ohshima E, Ishizuka S, Takeshita T, Tanaka Y, DeLuca HF 1983 Natural 25,26-dihydroxyvitamin D3 is an epimeric mixture. Proc Natl Acad Sci USA 80:5286–5288. 156. Hollis BW, Roos BA, Lambert PW 1980 25,26Dihydroxycholecalciferol: A precursor in the renal synthesis of 25-hydroxycholecalciferol-26,23-lactone. Biochem Biophys Res Commun 95:520–528. 157. Napoli JL, Horst RL 1981 25,26-Dihydroxyvitamin D3 is not a major intermediate in 25-hydroxyvitamin D3-26,23-lactone formation. Arch Biochem Biophys 212:754–758. 158. Tanaka Y, DeLuca HF, Schnoes HK, Ikekawa N, Eguchi T 1981 23,25-Dihydroxyvitamin D3: A natural precursor in the biosynthesis of 25-hydroxyvitamin D3-26,23-lactone. Proc Natl Acad Sci USA 78:4805–4808. 159. Reddy GS, Muralidharan KR, Okamura WH, Tserng KY, McLane JA 2001 Metabolism of 1α,25-dihydroxyvitamin D3 and its C-3 epimer 1α,25-dihydroxy-3-epi-vitamin D3 in neonatal human keratinocytes. Steroids 66:441–450. 160. Siu-Caldera ML, Sekimoto H, Weiskopf A, Vouros P, Muralidharan KR, Okamura WH, Bishop J, Norman AW, Uskokovic MR, Schuster I, Reddy GS 1999 Production of 1α,25-dihydroxy-3-epi-vitamin D3 in two rat osteosarcoma cell lines (UMR 106 and ROS 17/2.8): Existence of the C-3 epimerization pathway in ROS 17/2.8 cells in which the C-24 oxidation pathway is not expressed. Bone 24:457–463. 161. Horst RL, Shepard RM, Jorgensen NA, DeLuca HF 1979 The determination of 24,25-dihydroxyvitamin D and 25,26-dihydroxyvitamin D in plasma from normal and nephrectomized man. J Lab Clin Med 93:277–285. 162. Shepard RM, Horst RL, Hamstra AJ, DeLuca HF 1979 Determination of vitamin D and its metabolites in plasma from normal and anephric man. Biochem J 182: 55–69.

163. Bergman T, Postlind H 1991 Characterization of mitochondrial cytochromes P-450 from pig kidney and liver catalysing 26-hydroxylation of 25-hydroxyvitamin D3 C27 steroids. Biochem J 276:427–432. 164. Napoli JL, Okita RT, Masters BS, Horst RL 1981 Identification of 25,26-dihydroxyvitamin D3 as a rat renal 25-hydroxyvitamin D3 metabolite. Biochemistry 20:5865–5871. 165. Tanaka Y, Schnoes HK, Smith CM, DeLuca HF 1981 1,25,26-Trihydroxyvitamin D3 isolation, identification and biological activity. Arch Biochem Biophys 210: 104–109. 166. Astecker N, Reddy GS, Herzig G, Vorisek G, Schuster I 2000 1α,25-Dihydroxy-3-epi-vitamin D3 a physiological metabolite of 1α,25-dihydroxyvitamin D3: its production and metabolism in primary human keratinocytes. Mol Cell Endocrinol 170:91–101. 167. Masuda S, Kamao M, Schroeder NJ, Makin HL, Jones G, Kremer R, Rhim J, Okano T 2000 Characterization of 3-epi-1α,25-dihydroxyvitamin D3 involved in 1α,25-dihydroxyvitamin D3 metabolic pathway in cultured cell lines. Biol Pharm Bull 23:133–139. 168. Hylemon PB 1985 Metabolism of bile acids in intestinal microflora. In: Danielsson H, Sjoevall J (eds) Sterols and Bile Acids, Elsevier, pp. 331–344. 169. Penning TM, Bennett MJ, Smith-Hoog S, Schlegel BP, Jez JM, Lewis M 1997 Structure and function of 3α-hydroxysteroid dehydrogenase. Steroids 62:101–111. 170. Norman AW, Bouillon R, Farach-Carson MC, Bishop JE, Zhou LX, Nemere I, Zhao J, Muralidharan KR, Okamura WH 1993 Demonstration that 1β,25-dihydroxyvitamin D3 is an antagonist of the nongenomic but not genomic biological responses and biological profile of the three A-ring diastereomers of 1α,25-dihydroxyvitamin D3. J Biol Chem 268: 20022–20030. 171. Brown AJ, Ritter C, Slatopolsky E, Muralidharan KR, Okamura WH, Reddy GS 1999 1α,25-dihydroxy-3-epi-vitamin D3, a natural metabolite of 1α,25-dihydroxyvitamin D3, is a potent suppressor of parathyroid hormone secretion. J Cell Biochem 73:106–113. 172. Schuster I, Astecker N, HE, Hertig G, Reddy GS, Schmid J, Vorisek G 1997 Vitamin D metabolism in human keratinocytes and biological role of products. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D, Chemistry, Biology and Clinical Applications of the Steroid Hormone. Walter de Gruyter, New York, pp. 551–558. 173. Engstrom GW, Koszewski NJ 1989 Metabolism of vitamin D2 in pig liver homogenates: Evidence for a free radical reaction. Arch Biochem Biophys 270:432–440. 174. Horst RL, Reinhardt TA, Ramberg CF, Koszewski NJ, Napoli JL 1986 24-Hydroxylation of 1,25-dihydroxyergocalciferol: An unambiguous deactivation process. J Biol Chem 261:9250–9256. 175. Zimmerman DR, Reinhardt TA, Kremer R, Beitz DC, Reddy GS, Horst RL 2001 Calcitroic acid is a major catabolic metabolite in the metabolism of 1α-dihydroxyvitamin D2. Arch Biochem Biophys 392:14–22. 176. Rao LS, Ray R, Holick MF, Horst RL, Uskokovic MR, Reddy GS 2000 Metabolism of [3α-3H] 25-hydroxyvitamin D2 in kidneys isolated from normal and vitamin D2-intoxicated rats. J Nutr Sci Vitaminol (Tokyo) 46:222–229. 177. Horst RL, Omdahl JA, Reddy S 2003 Rat cytochrome P450C24 (CYP24) does not metabolize 1,25-dihydroxyvitamin D2 to calcitroic acid. J Cell Biochem 88: 282–285.

CHAPTER 2 Vitamin D Metabolism

178. Sunita Rao D, Balkundi D, Uskokovic MR, Tserng K, Clark JW, Horst RL, Satyanarayana Reddy G 2001 Double bond in the side chain of 1α,25-dihydroxy-22-ene-vitamin D3 is reduced during its metabolism: studies in chronic myeloid leukemia (RWLeu-4) cells and rat kidney. J Steroid Biochem Mol Biol 78:167–176. 179. Byford V, Strugnell S, Coldwell R, Schroeder N, Makin HL, Knutson JC, Bishop CW, Jones G 2002 Use of vitamin D4 analogs to investigate differences in hepatic and target cell metabolism of vitamins D2 and D3. Biochim Biophys Acta 1583:151–166. 180. Tachibana Y, Tsuji M 2001 Study on the metabolites of 1α,25-dihydroxyvitamin D4. Steroids 66:93–97. 181. Williams DH, Morris DS, Gilhooly MA, Norris AF 1982 Synthesis of 25(R)-hydroxycholecalciferol-26,23(S)-lactone and of 25(S),26-dihydroxyergocalciferol. In: Norman AW, Schaefer K, Herrath DV, Grigoleit H-G (eds) Vitamin D: Chemical, Biochemical and Clinical Endocrinology of Calcium Metabolism. Walter de Gruyter, Berlin, pp. 1067–1072. 182. Mazur Y, Segev D, Jones G 1982 Synthesis and determination of absolute configuration of kidney metabolites of vitamin D2. In: Norman AW, Schaefer K, Herrath DV, Grigoleit H-G (eds) Vitamin D: Chemical, Biochemical and Clinical Endocrinology of Calcium Metabolism. Walter de Gruyter, Berlin, pp. 1101–1106. 183. Hay AWM, Jones G 1979 The elution profile of vitamin D2 metabolites from Sephadex LH-20 columns. Clin Chem 25:473–475. 184. Reddy GS, Tserng KY 1990 24,25,28-Trihydroxyvitamin D2 and 24,25,26-trihydroxyvitamin D2: Novel metabolites of vitamin D2. Biochemistry 29:943–949. 185. Reddy GS, Tserng KY 1986 Isolation and identification of 1,24,25-trihydroxyvitamin D2, 1,24,25,28-tetrahydroxyvitamin D2, and 1,24,25,26-tetrahydroxyvitamin D2: New metabolites of 1,25-dihydroxyvitamin D2 produced in rat kidney. Biochemistry 25:5328–5336. 186. Jones G, Byford V, Makin H, Kremer R, Rice R, deGraffenried LA, Knutson JC, Bishop CW 1996 Antiproliferative activity and target cell catabolism of the vitamin D analog 1α24(S)-(OH)2D2 in normal and immortalized human epidermal cells. Biochem Pharmacol 52:133–140. 187. Rao DS, Siu-Caldera ML, Uskokovic MR, Horst RL, Reddy GS 1999 Physiological significance of C-28 hydroxylation in the metabolism of 1α,25-dihydroxyvitamin D2. Arch Biochem Biophys 368:319–328. 188. Kreitmeir H, Moll T 1928. Hypervitaminose durch grosse Dosen Vitamin D Munch. Med. Wochenschr 75:637–639. 189. Putscher W 1929 Über Vigantolschädigung der Niere einum Kinde. Z Kinderheilkd 48:269. 190. Holick MF, MacLaughlin JA, Doppelt SH 1981 Regulation of cutaneous previtamin D3 photosynthesis in man: Skin pigment is not an essential regulator. Science 211:590–593. 191. Clark MB, Potts JT 1977 25-Hydroxyvitamin D3 regulation. Calcif Tissue Res 22(Supplement):29–34. 192. Shepard RM, DeLuca HF 1980 Plasma concentrations of vitamin D3 and its metabolites in the rat as influenced by vitamin D3 or 25-hydroxyvitamin D3 intakes. Arch Biochem Biophys 202:43–53. 193. Hughes MR, Baylink DJ, Jones PG, Haussler MR 1976 Radioligand receptor assay for 25-hydroxyvitamin D2/D3 and 1α,25-dihydroxyvitamin D2/D3: Application to hypervitaminosis D. J Clin Invest 58:61–70. 194. Counts SJ, Baylink DJ, Shen FH, Sherrard DJ, Hickman RO 1975 Vitamin D intoxication in an anephric child. Ann Intern Med 82:196–200.

35 195. Littledike ET, Horst RL 1982 Vitamin D3 toxicity in dairy cows. J Dairy Sci 65:749–759. 196. Reinhardt TA, Conrad HR 1980 Mode of action of pharmacological doses of cholecalciferol during parturient hypocalcemia in dairy cows. J Nutr 110:1589–1594. 197. Horst RL, Reinhardt TA 1983 Vitamin D metabolism in ruminants and its relevance to the periparturient cow. J Dairy Sci 66:661–678. 198. Hunt RD, Garcia FG, Walsh RJ 1972 A comparison of the toxicity of ergocalciferol and cholecalciferol in Rhesus monkeys (Masaca mulatta). J Nutr 102:975–986. 199. Harrington DD, Page EH 1983 Acute vitamin D3 toxicosis in horses: Case reports and experimental studies of the comparative toxicity of vitamins D2 and D3. J Am Vet Med Assoc 182:1358–1369. 200. Hodson EM, Evans RA, Dunstan CR, Hills E, Wong SYP, Rosenberg AR, Roy LP 1985 Treatment of childhood renal osteodystrophy with calcitriol or ergocalciferol. Clin Nephrol 24:192–200. 201. Sjoden G, Smith C, Lingren U, DeLuca HF 1985 1α-hydroxyvitamin D2 is less toxic than 1α-hydroxyvitamin D3 in the rat. Proc Soc Exp Biol Med 178:432–436. 202. Beckman MJ, Horst RL, Reinhardt TA, Beitz DC 1990 Up-regulation of the intestinal 1,25-dihydroxyvitamin D receptor during hypervitaminosis D: A comparison between vitamin D2 and vitamin D3. Biochem Biophys Res Commun 169:910–915. 203. Mortensen JT, Brinck P, Binderup L 1993 Toxicity of vitamin D analogues in rats fed diets with standard or low calcium contents. Pharmacol Toxicol 72:124–127. 204. Hines TG, Jacobson NL, Beitz DC, Littledike ET 1985 Dietary calcium and vitamin D: Risk factors in the development of atherosclerosis in young goats. J Nutr 115: 167–178. 205. Payne JM, Manston R 1967 The safety of massive doses of vitamin D3 in the prevention of milk fever. Vet Rec 81:214–216. 206. Hibbs JW, Pounden WD 1955 Studies on milk fever in dairy cows. IV. Prevention by short-time, prepartum feeding of massive doses of vitamin D. J Dairy Sci 38:65–72. 207. Napoli JL, Sommerfeldt JL, Pramanik BC, Gardner R, Sherry AD, Partridge JJ, Uskokovic MR, Horst RL 1983 19-Nor-10-ketovitamin D derivatives: Unique metabolites of vitamin D3, vitamin D2, and 25-hydroxyvitamin D3. Biochemistry 22:3636–3640. 208. Recker RR, Schoenfeld JL, Slatopolsky R, Goldsmith R, Brickman A 1979 25-Hydroxyvitamin D (calcidiol) on renal osteodystophy: Long term results of a six center trial. In: Norman AW, Schaefer K, Herrath DV, Grigoleit HG, Coburn JW, DeLuca HF, Mawer EB, Suda T (eds) Vitamin D: Basic Research and Its Clinical Application. Walter de Gruyter, Berlin, pp. 869–875. 209. West TET, Joffe M, Sinclair L, O’Riordan JLH 1971 Treatment of hypercalcemia with calcitonin. Lancet, 675–678. 210. Ulbrych-Jablonska A 1972 The hypocalcemic effect of glucagon in cases of hypercalcemia. Helv Paediatr Acta 27:613–615. 211. Kingma JG, Roy PE 1990 Effect of ethane-I-hydroxyl, I-diphosphonate on arterial calcinosis induced by hypervitaminosis D: a morphologic investigation. J Exp Path (Oxford) 71:145–153. 212. Streck WF, Waterhouse C, Haddad JG 1979 Glucocorticoid effects in vitamin D intoxication. Arch Intern Med 139: 974–977.

36

RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY

213. Buffenstein R, Skinner DC, Yahav S, Moodley GP, Cavaleros M, Zachen D, Ross FP, Pettifor JM 1991 Effect of oral cholecalciferol supplementation at physiological and supraphysiological doses in naturally vitamin D3-deficient subterranean damara mole rats (Cryptomys damarensis). J Endocrinol 131:197–202. 214. Cheeke PR, Amberg JW 1973 Comparative calcium excretion by rats and rabbits. J Anim Sci 37:450–454. 215. Schryver HF, Hintz HF, Lowe JE 1974 Calcium and phosphorus in the nutrition of the horse. Cornell Vet 64:493–515. 216. Bourdeau JE, Schwer-Dymerski DA, Stern PH, Langman CB 1986 Calcium and phosphorus metabolism in chronically vitamin D-deficient laboratory rabbits. Miner Electrolyte Metab 12:176–185. 217. Sommerfeldt JL, Horst RL, Littledike ET, Beitz DB 1979 In vitro degradation of cholecalciferol in rumen fluid. J Dairy Sci 62(Supplement 1):192. 218. Sommerfeldt JL, Napoli JL, Littledike ET, Beitz DC, Horst RL 1983 Metabolism of orally administered [3H]ergocalciferol and [3H]cholecalciferol by dairy calves. J Nutr 113:2595–2600. 219. Stern PH, Horst RL, Gardner R, Napoli JL 1985 10-Keto or 25-hydroxy substitution confer equivalent in vitro boneresorbing activity to vitamin D3. Arch Biochem Biophys 236:555–558. 220. Hunt RD, Garcia FG, Hegsted DM, Kaplinsky N 1967 Vitamin D2 and vitamin D3 in New World primates: Influence on calcium absorptions. Science 158:943–945. 221. Steenbock H, Kletzien SWF, Haplin JG 1932 Reaction of chicken to irradiated ergosterol and irradiated yeast as contrasted with natural vitamin D of fish liver oils. J Biol Chem 97:249–264. 222. Horst RL, Napoli JL, Littledike ET 1982 Discrimination in the metabolism of orally dosed ergocalciferol and cholecalciferol by the pig, rat and chick. Biochem J 204:185–189. 223. Sebert JL, Garabedian M, deChasteigner C, Defrance D 1991 Comparative effects of equal doses of vitamin D2 and vitamin D3 for the correction of vitamin D deficiency in the elderly. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D: Gene Regulation, Structure–Function Analysis and Clinical Application. Walter de Gruyter, Berlin, pp. 765–766. 224. Reddy GS, Ray R, Holick MF 1990 Serum 1,25-dihydroxyvitamin D2 levels are two times higher than 1,25-dihydroxyvitamin D3 levels in vitamin D-deficient rats, dosed acutely with equal amounts of vitamin D2 and vitamin D3. J Bone Miner Res 5(Supplement 2):S265. 225. Haddad JG, Walgate J 1976 25-Hydroxyvitamin D transport in human plasma. J Biol Chem 251:4803–4809. 226. Baird DT, Horton R, Longcope C, Tait JF 1969 Steroid hormone dynamics under steady state conditions. Recent Prog Horm Res 25:611–664. 227. Hoy DA, Ramberg CF Jr., Horst RL 1988 Evidence that discrimination against ergocalciferol by the chick is the

228. 229. 230.

231.

232.

233.

234.

235.

236.

237.

238.

result of enhanced metabolic clearance rates for its monoand dihydroxylated metabolites. J Nutr 118:633–638. Hay AWM, Watson G 1977 Vitamin D2 in vertebrate evolution. Comp Biochem Physiol 56B:375–380. Horst RL, Reinhardt TA, Beitz DC, Littledike ET 1981 A sensitive competitive protein binding assay for vitamin D in plasma. Steroids 37:581–592. Reddy GS, Napoli JL, Hollis BW 1984 Preferential 25-hydroxylation of vitamin D2 vs vitamin D3 by rat liver is dependent upon vitamin D-binding protein. Calcif Tissue Int 36:524 (Abstract). Arnaud SB, Young DR, Cann C, Reinhardt TA, Henrickson R 1985 Is hypervitaminosis D normal in the rhesus monkey? In: Norman AW, Schaefer K, Grigoleit H-G, Herrath DV (eds) Vitamin D. Chemical, Biochemical and Clinical Update. Walter de Gruyter, Berlin, pp. 585–586. Shinki T, Shiina Y, Takahashi N, Tanioka Y, Koizumi H, Suda T 1983 Extremely high circulating levels of 1α,25dihydroxyvitamin D3 in the marmoset, a New World monkey. Biochem Biophys Res Commun 114:452–457. O’Loughlin PD, Morris HA 1991 Duodenal transport of calcium in the marmoset. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D. Gene Regulation, Structure– Function Analysis and Clinical Application. Walter de Gruyter, Berlin, pp. 345–346. Shore RF, Hayes ME, Balment RJ, Mawer EB 1988 Serum 25(OH)D3 and 1,25(OH)2D3 levels in wild and laboratorybred wood mice and bank voles. In: Norman AW, Schaefer K, Grigoleit H-G, Herrath DV (eds) Vitamin D. Molecular, Cellular and Clinical Endocrinology. Walter de Gruyter, Berlin, pp. 633–634. Kobayashi T, Takeuchi A, Okano T 1991 An evolutionary aspect in vertebrates from the viewpoint of vitamin D3 metabolism. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D. Gene Regulation, Structure–Function Analysis and Clinical Application. Walter de Gruyter, Berlin, pp. 679–680. Glowacki J, Deftos LJ, Mayer E, Norman AW, Henry H 1982 Chondrichthyes cannot resorb implanted bone and have calcium-regulating hormones. In: Norman AW, Schaefer K, Herrath DV, Grigoleit H-G (eds) Vitamin D. Chemical, Biochemical and Clinical Endocrinology of Calcium Metabolism. Walter de Gruyter, Berlin, pp. 613–615. Sundell K, Bishop JE, Björnsson BT, Horst RL, Hollis BW, Norman AW 1991 Organ distribution of a high affinity receptor for 1,25(OH)2 vitamin D3 in the Atlantic cod (Gadus morhua), a marine teleost. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D. Gene Regulation, Structure– Function Analysis and Clinical Application. Walter de Gruyter, Berlin, pp. 114–115. Maenpaa PH, Koskinen T, Koskinen E 1988 Serum profiles of vitamins A, E and D in mares and foals during different seasons. J Anim Sci 66:1418–1423.

CHAPTER 3

Photobiology of Vitamin D MICHAEL F. HOLICK

I. II. III. IV.

Vitamin D, Skin, and Bone Research Laboratory; Department of Medicine; Endocrinology, Nutrition and Diabetes Section; Boston Medical Center and Boston University School of Medicine, Boston, Massachusetts

Introduction Historical Perspective Photobiology of Vitamin D Role of Sunlight and Dietary Vitamin D in Bone Health, Overall Health, and Well-Being

I. INTRODUCTION Vitamin D is neither a vitamin nor a hormone when adequate exposure to sunlight is available to promote the synthesis of vitamin D in the skin [1]. First produced in ocean-dwelling phytoplankton and zooplankton, vitamin D has been made by life forms on earth for almost 1 billion years [2]. Although the physiological function of vitamin D in these lower life forms is uncertain, it is well recognized that the photosynthesis of vitamin D became critically important for landdwelling vertebrates that required a mechanism to increase the efficiency of absorption of dietary calcium. The major physiological function of vitamin D in vertebrates is to maintain extracellular fluid concentrations of calcium and phosphorus within a normal range. Vitamin D accomplishes this by increasing the efficiency of the small intestine to absorb dietary calcium and phosphorus, and by stimulating the mobilization of calcium and phosphorus stores from the bone [3].

II. HISTORICAL PERSPECTIVE It was the lack of appreciation of the beneficial effect of sunlight in preventing rickets that caused the widespread endemic outbreak of this bone-deforming disease in the 17th through 19th centuries. As early as 1822, Sniadecki [4] had suggested that the high incidence of rickets that occurred in Warsaw, Poland, in the early 1800s was likely caused by the lack of adequate exposure to sunlight. This was based on his observations that whereas children who lived in Warsaw had a very high incidence of rickets, children who lived in the rural areas outside of the city did not suffer the same malady. Seventy years later Palm [5] also recognized that lack of sunlight was a common denominator associated with the high incidence of rickets in children VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Sunlight, Vitamin D, and Skin Cancer VI. Conclusion Acknowledgment References

living in the inner cities when compared to children living in underdeveloped countries. Both Sniadecki [4] and Palm [5] encouraged systematic sunbathing as a means of preventing and curing rickets, but their recommendations went unheeded. Huldschinsky [6] was the first to clearly demonstrate that it was exposure of the skin to ultraviolet radiation that was responsible for the antirachitic activity of sunlight that had been postulated by Sniadecki [4] and Palm [5]. He exposed four rachitic children to radiation from a mercury arc lamp and reported dramatic reversal of rickets within 4 months. He further reasoned that the radiation responsible for producing the antirachitic factor was the same radiation that caused melanization of the skin. The intimate relationship between sunlight and its antirachitic properties in humans was finally demonstrated in 1921 when Hess and Unger [7] took rachitic children and exposed them to sunlight on the roof of a New York City hospital and demonstrated that within 4 months the rachitic lesions had healed. The concept that rickets was caused by a nutritional deficiency was supported by the observation in 1827 by Bretonneau, who treated a 15-month-old child with acute rickets with cod liver oil and noted the incredible speed at which the patient was cured [8]. His student Trousseau used liver oils from a variety of fish and aquatic mammals including herring, whales, and seals for the treatment of rickets and osteomalacia. In the early 20th century, Mellanby [9] conducted classic studies where he demonstrated that rachitic beagles that were fed cod liver oil were cured of their bone-deforming disease. Originally, it was thought that the active antirachitic factor in cod liver oil was vitamin A. However, McCollum et al. [10] heated cod liver oil to destroy the vitamin A activity and demonstrated that the preparation maintained its antirachitic activity. He, therefore, coined the term “vitamin D” for the antirachitic factor. Copyright © 2005, Elsevier, Inc. All rights reserved.

38 The appreciation that exposure to sunlight could cure rickets in children prompted Steenbock and Black [11] and Hess and Weinstock [12] in 1924 to expose a variety of foods and other substances including vegetable oils, animal feed, lettuce, wheat, and grasses to ultraviolet radiation. They found that this process imparted antirachitic activity to all of the substances. These observations led Steenbock [13] to conclude that there would be great utility in irradiation of food substances for the prevention and cure of rickets in children. This concept initially led to the addition of provitamin D (ergosterol or 7-dehydrocholesterol) to milk and its subsequent irradiation to impart antirachitic activity. Once vitamin D (D without a subscript refers to either D2 or D3) was inexpensive to produce, vitamin D was directly added to milk. This simple concept led to the eradication of rickets as a significant health problem in the United States and other countries that followed this practice.

MICHAEL F. HOLICK

SUN

SUN

SUN H HO 7-DEHYDROCHOLESTROL

SUN H

HO

LUMISTEROL3 SUN HO

PREVITAMIN3 CH3

H BLOOD VESSEL

OH TACHYSTEROL3

DBP HO SUN

VITAMIN3

SUN

SUN

III. PHOTOBIOLOGY OF VITAMIN D A. Photosynthesis of Vitamin D3 When human skin is exposed to sunlight, it is the solar ultraviolet B photons with energies between 290 and 315 nm that are responsible for causing the photolysis of 7-dehydrocholesterol (provitamin D3; the immediate precursor in the cholesterol biosynthetic pathway) to previtamin D3 [1,14] (Fig. 1). This photochemical process occurs in the plasma membrane of skin cells; as a result, the thermodynamically unstable cis,cis isomer of previtamin D3 is rapidly transformed by a rearrangement of double bonds to form vitamin D3 [15] (Fig. 2). Approximately 50% of previtamin D3 is converted to vitamin D3 within 2 hr. As vitamin D3 is formed in the membrane, its open flexible structure is likely jettisoned from the plasma membrane into the extracellular space. Once vitamin D3 enters the extracellular fluid space, it is attracted to the vitamin D binding protein (DBP) in the circulation, and thus enters the dermal capillary bed.

B. Regulation of Vitamin D Synthesis by Sunlight In the late 1960s, a theory was popularized that human skin pigmentation evolved to play a critically important role in regulating the cutaneous photosynthesis of vitamin D3. Loomis [16] speculated that melanin pigmentation in humans evolved for the purpose of preventing excessive production of vitamin D3 in the skin due to chronic excessive exposure to

OH SUPRASTEROL I

HO SUPRASTEROL II OH 5,6-TRANSVITAMIN D

FIGURE 1 Photochemical events that lead to the production and regulation of vitamin D3 in the skin. Reprinted with permission from Holick [1]. DBP is the plasma vitamin D binding protein.

sunlight especially in peoples who live near the equator. Although it is likely that increased pigmentation was important for the prevention of skin cancer, there is little evidence that skin pigmentation evolved to prevent vitamin D intoxication. The reason for this is that once previtamin D3 is formed in the skin, it either can isomerize to vitamin D3 or can absorb solar ultraviolet B radiation and isomerize into biologically inactive photoisomers lumisterol and tachysterol [17] (Fig. 1). Furthermore, when vitamin D3 is made, it can either enter into the circulation or absorb solar ultraviolet photons and isomerize to at least three photoproducts including suprasterol I, suprasterol II, and 5,6-transvitamin D3 [18] (Fig. 1). Thus, sunlight itself can regulate the total output of vitamin D3 in the skin by causing the photodegradation of previtamin D3 and vitamin D3. This is the likely explanation for why there are no reported cases of vitamin D intoxication–induced hypercalcemia from chronic excessive exposure to sunlight. Melanin, on the other hand, provides the body with an effective natural sunscreen. Thus, increased melanin pigmentation will reduce the efficiency of the sun-mediated photosynthesis of previtamin D3 [19]. This is particularly

39

CHAPTER 3 Photobiology of Vitamin D

OH

s-trans,s-cis-PRE-D3

SUN

C A

D

B

HO

HEXANE

PRO-D3

19 C D 10 9 3 5 6 7 HO

T1/2= 91 HRS

s-cis,s-cis-PRE-D3 25°C

CH3

SUN HO

VITAMIN D3 T1/2= 8 HRS

SKIN

HO PRO-D3

HO s-cis,s-cis-PRE-D3

FIGURE 2

Photolysis of provitamin D3 (pro-D3) into previtamin D3 (pre-D3) and its thermal isomerization to vitamin D3 in hexane and in lizard skin. In hexane, pro-D3 is photolyzed to s-cis,s-cis-pre-D3. Once formed, this energetically unstable conformation undergoes a conformational change to the s-trans,s-cis-pre-D3. Only the s-cis, s-cis-pre-D3 can undergo thermal isomerization to vitamin D3. The s-cis,s-cis conformer of pre-D3 is stabilized in the phospholipid bilayer by hydrophilic interactions between the 3/3-hydroxyl group and the polar head of the lipids, as well as by van der Waals interactions between the steroid ring and side-chain structure and the hydrophobic tail of the lipids. These interactions significantly decrease the conversion of the s-cis,s-cis conformer to the s-trans,s-cis conformer, thereby facilitating the thermal isomerization of s-cis,s-cis-pre-D3 to vitamin D3. Reprinted with permission from Holick et al. [15].

important in Blacks who live in northern latitudes and who ingest very little dietary vitamin D. It is the likely explanation for why Blacks have lower circulating concentrations of 25-hydroxyvitamin D3 (25OHD) and are more prone to developing vitamin D deficiency [20] (see Chapter 33).

C. Influence of Latitude, Season, and Time of Day on Vitamin D Synthesis It was recognized at the beginning of the 20th century that the incidence of rickets was much higher during the winter and early spring months and that rickets was less prevalent during the summer and autumn [21]. An evaluation of the effect of season on the cutaneous production of vitamin D3 in Boston revealed that during the summer months of June and July, 7-dehydrocholesterol was most efficiently converted to previtamin D3 [22] (Fig. 3). There was a gradual decline in the production of previtamin D3 after August, and there was essentially no previtamin D3 formed in human skin

after November (Fig. 3). Previtamin D3 photosynthesis commenced in the middle of March. To evaluate the influence of latitude on the cutaneous production of vitamin D3, similar studies were conducted in Edmonton, Canada (52°N), Los Angeles (34°N), and San Juan, Puerto Rico (18°N). In Edmonton, the photosynthesis of previtamin D3 essentially ceased by mid-October and did not resume until mid-April. However, in Los Angeles and San Juan, the production of previtamin D3 in the skin occurred throughout the year (Fig. 3). The cutaneous production of vitamin D3 was also evaluated in Boston every hour from the time the sun rose to the time the sun set in the middle of the month on cloudless days for a full year. It was found that during the summer, sunlight was capable of producing vitamin D3 in the skin from 0700 to as late as 1700 hr Eastern Standard Time (EST) (Fig. 4). However, as the zenith angle of the sun increased in the spring and autumn, previtamin D3 photosynthesis in the skin began at approximately 0900 and ceased at approximately 1600 EST [1].

40

MICHAEL F. HOLICK

D. Influence of Sunscreen Use, Melanin, Clothing, Glass, and Plastics on Vitamin D Synthesis

25

20

15

10

5

0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Month

FIGURE 3

Photosynthesis of previtamin D3 after exposure of 7-dehydrocholesterol (7-DHC) to sunlight. Measurements were as follows: in Boston (42°N) after 1 hr (
) and 3 hr () and total photo products (previtamin D3, lumisterol, and tachysterol) after 3 h in Boston (); in Edmonton, Canada (52°N), after 1 hr (); in Los Angeles (34°N) () and Puerto Rico (18°N) in January (). Reprinted with permission from Webb et al. [22].

Boston Edmonton Bergen

A

% Previtamin D3 formation

8

Any substance such as melanin, clothing, or a sunscreen that absorbs ultraviolet B radiation will reduce the cutaneous production of vitamin D3 [1,23–25]. Sunscreen use is now widely accepted as an effective method for diminishing the damaging effects due to chronic excessive exposure to sunlight. Sunscreens are of great benefit because they can prevent sunburn, skin cancer, and skin damage associated with exposure to sunlight. The solar radiation that is most responsible for causing damage to the skin is the ultraviolet B (UVB) radiation. Because the major function of sunscreens is to absorb solar UVB radiation on the surface of the skin before it penetrates into the deeper viable layers, sunscreen use diminishes the total number of UVB photons that can reach the 7-dehydrocholesterol stores in the viable epidermis to form previtamin D3. Topical application of a sunscreen will substantially diminish or completely prevent the cutaneous production of previtamin D3 [25]. The topical application of

8

C

6

6

4

4

2

2

0

0 Jan

Mar

May

Jul

Sep

8:00

Nov

Buenos Aires Johannesburg Cape Town Ushuaia

10

% Previtamin D3 formation

% Previtamin D3 formation

B

8 6 4 2

10:00 12:00 14:00 16:00 18:00

Hour of Day

Month 10

Boston Edmonton Bergen

% Previtamin D3 formation

% Photoproducts from 7-DHC

30

Buenos Aires Johannesburg Cape Town Ushuaia

D

8 6 4 2 0

0 Jan

Mar

May

Jul

Month

Sep

Nov

8:00

10:00 12:00 14:00 16:00 18:00

Hour of Day

FIGURE 4 Influence of season, time of day, and latitude on the synthesis of previtamin D3 in Northern (A and C) and southern hemispheres (B and D). The hour indicated in C and D is the end of the 1-hr exposure time. (Reproduced with permission from Chen TC 1998 The Photobiology of Vitamin D. In: Holick MF (ed) Vitamin D—Physiology, Molecular Biology and Clinical Applications. Humana Press, Totowa, NJ, pp. 17–37.)

41

CHAPTER 3 Photobiology of Vitamin D

Serum Vitamin D (nmol/L)

80

Without Sunscreen

60 40

With Sunscreen

20 0

−2

0

2

4

8 6 Days

10

12

14

16

FIGURE 5 Effect of sunscreen on vitamin D status. Circulating concentrations of vitamin D were measured in young adults after application of a cream either with a sun protection factor of 8 or without sunscreen (topical placebo cream) following a single exposure to one minimal erythemal dose of simulated sunlight. Reprinted with permission from Matsuoka et al. [23].

60

Serum 25-Hydroxyvitamin D (ng/ml)

Serum Vitamin D3 (ng/ml)

a sunscreen with a sun protection factor (SPF) of 8, followed by whole-body exposure to one minimal erythemal dose of simulated sunlight, prevented any significant increase in the blood levels of vitamin D3 in healthy young adult volunteers. On the other hand, without the sunscreen, there was a marked 10- to 20-fold increase in circulating concentrations of vitamin D3 [23] (Fig. 5). Similarly, since clothing absorbs ultraviolet B radiation, the wearing of clothing on body surfaces prevents the cutaneous production of vitamin D3 in those covered surfaces [24] (Fig. 6). Thus, in cultures that require the covering of almost the entire body with

clothing, such as Bedouins living in the Negev Desert, despite the intense sunlight environment, the Bedouin women have an increased risk of vitamin D deficiency and osteomalacia and their children are more prone to developing vitamin D–deficiency rickets [26]. Although sunscreens and clothing significantly diminish the cutaneous production of vitamin D, for children and young adults there is little concern about their developing vitamin D deficiency by practicing sun protection. The main reason is that the casual everyday limited exposure of sunlight to the face and hands is sufficient to provide the vitamin D requirement. However, elderly people, who are often very concerned about their health and appearance, will consistently apply a topical sunscreen on all sun-exposed areas and/or wear clothing over most sun-exposed areas before going outdoors. Since aging significantly decreases the capacity of the skin to produce vitamin D3 because of the marked decline in 7-dehydrocholesterol in the epidermis, chronic sunscreen use by elderly people can increase the risk of vitamin D deficiency [1,23] (Fig. 7). Because glass and Plexiglas efficiently absorb most if not all ultraviolet B photons, exposure of the skin to sunlight through glass or Plexiglas will not produce any vitamin D3 (Fig. 8).

80 60 40 20 0

None

Summer

Autumn

Summer and Sunscreen

PA IL 40

30

20

10

0 Without Sunscreen

FIGURE 6 Effect of clothing and sunscreen on vitamin D status. Circulating concentrations of vitamin D were measured in human subjects who wore either no clothing, summer-type clothing, autumn-type clothing, or summer-type clothing and sunscreen 24 hr after a whole-body exposure to one minimal erythemal dose of ultraviolet B radiation. Reprinted with permission from Matsuoka et al. [24].

50

FIGURE 7

With Sunscreen

Effect of chronic sunscreen use on vitamin D status. Circulating concentrations of 25-hydroxyvitamin D were measured in adults from Pennsylvania (PA) and Illinois (IL) who always wore a sunscreen or never wore a sunscreen. Reprinted with permission from Matsuoka et al. [25].

42

MICHAEL F. HOLICK

% Previtamin D3 Synthesis

7 6

5.98%

5 4 3 2 1 0

No Shield

ND Glass

ND Plexiglas

ND Plastic

FIGURE 8 Prevention of provitamin D3 formation by UV radiation by common light-shielding materials. The effects of glass, plastic, or Plexiglas (DuPont Chemical Company, Memphis, TN) placed between the simulated sunlight source and the provitamin D3 were measured. ND, Not detectable. Reprinted with permission from Holick [1].

IV. ROLE OF SUNLIGHT AND DIETARY VITAMIN D IN BONE HEALTH, OVERALL HEALTH, AND WELL-BEING Vitamin D deficiency is finally being recognized as an epidemic for all age groups in most industrialized countries [27–35]. Ninety to 95% of our vitamin D requirement comes from exposure to sunlight [36]. Most experts agree that in the absence of sun exposure, 1000 IU of vitamin D is required to satisfy the body’s vitamin D need [37]. If children and adults received sensible exposure to sunlight during the spring, summer and fall, they are able to store vitamin D in their body fat and call upon it during the winter when the sun is incapable of producing vitamin D in the skin. However, because of the increased concern about the risk of skin cancer from sun exposure and because both children and adults have fewer outdoor activities, there is a resurgence of vitamin D deficiency for the population at large. Nesby-O’Dell et al. [38] reported that 42% of African-American women aged 15–49 years were found to be vitamin D–deficient throughout the United States at the end of the winter. Tangpricha et al. reported that 32% of healthy young men and women aged 18–29 years were vitamin D deficient at the end of the winter in the Boston area [31]. Sullivan et al. [35] reported that 48% of young girls aged 9–11 years were vitamin D deficient at the end of the winter and 17% were vitamin D deficient at the end of the summer in Maine. Vitamin D deficiency has also become a major health concern for pregnant women and infants who receive their sole nutrition from breastfeeding [54].

Rickets, which was considered to be a disease of the 19th century, has become much more common, especially in children of color who are solely breastfed [52]. For preadolescent and adolescent children, vitamin D deficiency will prevent them from attaining their genetically programmed peak bone mass. For adults, it will precipitate and exacerbate osteoporosis, increase risk of fracture, and cause the painful bone disease osteomalacia [39–44]. It is now also recognized that insufficient exposure to sunlight and vitamin D deficiency increases risk of many common cancers, including colon, breast, prostate, ovarian, and esophageal cancers and non-Hodgkin’s and Hodgkin’s lymphoma [45–49]. In addition, it increases risk of children developing type 1 diabetes [50] and also puts adults at increased risk of developing hypertension and having heart disease [51–55].

V. SUNLIGHT, VITAMIN D, AND SKIN CANCER There is no question that chronic excessive exposure to sunlight and sun-burning experiences, especially during childhood and young adulthood, increases risk of squamous and basal cell carcinoma and wrinkling [56]. There is, however, little evidence that sensible limited exposure to sunlight to satisfy the body’s requirement for vitamin D will substantially increase risk of developing either basal or squamous cell carcinoma [36]. Furthermore, if basal or squamous cell carcinoma develops and is detected early, this type of cancer is easily curable. Melanoma, on the other hand, is often a deadly skin cancer. Although it has been suggested that melanoma is associated with sun exposure, it should be noted that most melanomas occur on non—sun-exposed areas, i.e., back of leg, abdomen, etc., and rarely occur on the face [57]. Furthermore, a recent study in Scandinavia suggested that melanoma is associated with increased number of nevi (moles), color of the hair, i.e., red and blonde, and number of sun-burning experiences [58]. They also reported that there may be an association with frequent exposure to a tanning bed. However, the data were not statistically significant and the authors concluded that there was only a suggestion of an association. To put this into perspective, it is estimated that approximately 1200 people per year will die in the United States of squamous and basal cell carcinomas that are clearly associated with excessive sun exposure. However, the lack of adequate sun exposure and/or adequate vitamin D nutrition puts at risk more than 150,000 lives for dying of many common cancers [59,60]. There has been no estimate of the toll that vitamin D

CHAPTER 3 Photobiology of Vitamin D

deficiency has in mortality rates due to heart disease and complications of type 1 diabetes, which has now been clearly linked to vitamin D deficiency.

VI. CONCLUSION The skin has a huge capacity to produce vitamin D [61]. Children and adults exposed to natural or artificial ultraviolet B radiation can satisfy their vitamin D requirement [62–66]. Exposure of a healthy adult in a bathing suit to one minimal erythemal dose (a light pinkness to the skin) of sunlight or tanning bed UV radiation is equivalent to taking between 10,000 IU and 20,000 IU of vitamin D [36,61]. Because melanin is such an effective sunscreen, African Americans require five to 50 times the exposure that a white person requires to satisfy their body’s vitamin D requirement [19]. Although aging markedly diminishes the capacity of the skin to produce vitamin D3, because its capacity is so high, elders exposed to sunlight can still raise their blood levels of 25(OH)D into a satisfactory range [67]. Thus, there needs to be a reexamination of the message that any exposure to sunlight requires some type of sunscreen or sun protection. This extreme position has pervaded the psyche of the population at large. It unfortunately has put many people at risk for vitamin D deficiency and many of the serious chronic diseases that have been associated with inadequate sun exposure and vitamin D deficiency. Sensible exposure to sunlight, typically no more than 5 to 10 minutes a day between 10 A.M. and 3 P.M. during seasons when vitamin D can be produced in the skin will satisfy most people’s vitamin D requirement [60,61]. Taking a multivitamin containing 400 IU of vitamin D will satisfy approximately 40% of the body’s requirement. Thus, additional supplementation and/or foods that contain vitamin D are required to satisfy the 1000 IU of vitamin D that most experts believe is necessary to raise the blood levels of 25(OH)D above 30 ng/ml, which is considered to be a healthy level.

Acknowledgment This work was supported in part by the following grants: General Clinical Research Center M01RR00533, R01 AR36963 and the ITA’s UV Foundation.

References 1. Holick MF 1994 McCollum Award Lecture, Vitamin D—New horizons for the 21st century. Am J Clin Nutr 60:619–630.

43 2. Holick MF 1989 Phylogenetic and evolutionary aspects of vitamin D from phytoplankton to humans. In: Pang PKT, Schreibman MP (eds) Vertebrate Endocrinology: Fundamentals and Biomedical Implications, Vol. 3. Academic Press, Orlando, FL, pp. 7–43. 3. Holick MF 1995 Vitamin D: Photobiology, metabolism, and clinical applications. In: DeGroot LJ, Besser M, Burger HG, Jameon JL, Loriaux DL, Marshall JC, Odell WD, Potts JT, Rubenstein AH (eds) Endocrinology, 3rd ed. Chapter 59. Saunders, Philadelphia, pp. 990–1013. 4. Sniadecki J 1939 cited by W Mozolowski, Jerdrzej Sniadecki (1768–1883) on the cure of rickets. Nature 143:121. 5. Palm TA 1980 The geographic distribution and etiology of rickets. Practitioner 45:270–279, 421–442. 6. Huldschinsky K 1919 Curing rickets by artificial UV-radiation (Heilung von Rachitis durch kunstliche Hohensonne). Deut Med Wochenschr 45:712–713 (in German). 7. Hess AF, Unger LF 1921 Cure of infantile rickets by sunlight. JAMA 77:39. 8. Mayer J 1957 Armand Trousseau and the arrow of time. Nutr Rev 15:321–323. 9. Mellanby T 1918 The part played by an accessory factor in the production of experimental rickets. J Physiol 52:11–14. 10. McCollum EF, Simmonds N, Becker JE, Shipley PG 1922 Studies on experimental rickets; and experimental demonstration of the existence of a vitamin which promotes calcium deposition. J Biol Chem 53:293–312. 11. Steenbock H, Black A 1924 The reduction of growth-promoting and calcifying properties in a ration by exposure to ultraviolet light. J Biol Chem 61:408–411. 12. Hess AF, Weinstock M 1924 Antirachitic properties imparted to inert fluids and green vegetables by ultraviolet irradiation. J Biol Chem 62:301–313. 13. Steenbock H 1924 The induction of growth-prompting and calcifying properties in a ration exposed to light. Science 60:224–225. 14. MacLaughlin JA, Anderson RR, Holick MF 1982 Spectral character of sunlight modulates photosynthesis of previtamin D3 and its photoisomers in human skin. Science 216:1001–1003. 15. Holick MF, Tian XQ, Allen M 1995 Evolutionary importance for the membrane enhancement of the production of vitamin D3 in the skin of poikilothermic animals. Proc Natl Acad Sci USA 92:3124–3126. 16. Loomis F 1967 Skin-pigment regulation of vitamin D biosynthesis in man. Science 157:501–506. 17. Holick MF, MacLaughlin JA, Doppelt SH 1981 Factors that influence the cutaneous photosynthesis of previtamin D3. Science 211:590–593. 18. Webb AR, deCosta BR, Holick MF 1989 Sunlight regulates the cutaneous production of vitamin D3 by causing its photodegradation. J Clin Endocrinol Metab 68:882–887. 19. Clemens TL, Adams JS, Henderson SL, Holick MF 1982 Increased skin pigment reduces the capacity of the skin to synthesize vitamin D. Lancet 1:74–76. 20. Bell NH, Greene A, Epstein S, Oexmann MJ, Shaw W, Shary J 1985 Evidence for alteration of the vitamin D endocrine system in Blacks. J Pediatr 76:470–473. 21. Kassowitz M 1987 Tetany and autointoxication in infants (Tetani and autointoxication in kindersalter). Wien Med Presse 97:139 (in Dutch). 22. Webb AR, Kline L, Holick MF 1988 Influence of season and latitude on the cutaneous synthesis of vitamin D3: Exposure to winter sunlight in Boston and Edmonton will not promote vitamin D3 synthesis in human skin. J Clin Endocrinol Metab 67:373–378.

44 23. Matsuoka LY, Ide L, Wortsman J, MacLaughlin JA, Holick MF 1987 Sunscreens suppress cutaneous vitamin D3 synthesis. J Clin Endocrinol Metab 64:1165–1168. 24. Matsuoka LY, Wortsman J, Dannenberg MJ, Hollis BW, Lu Z, Holick MF 1992 Clothing prevents ultraviolet-B radiationdependent photosynthesis of vitamin D3. J Clin Endocrinol Metab 75:1099–1103. 25. Matsuoka LY, Wortsman J, Hanifan N, Holick MF 1988 Chronic sunscreen use decreases circulating concentrations of 25-hydroxyvitamin D: A preliminary study. Arch Dermatol 124:1802–1804. 26. Sedrani SH, Al-Arabi KM, Abanny A, Elidrissy A 1990 Frequency of vitamin D deficiency rickets in Riyadh. In: Study of Vitamin D Status and Factors Leading to Its Deficiency in Saudi Arabia. King Saudi Univ. Press, Riyadh, pp. 281–285. 27. Gloth FM, Gundberg CM, Hollis BW, Haddad HG, Tobin JD 1995 Vitamin D deficiency in homebound elderly persons. JAMA 274:1683–1686. 28. Malabanan A, Veronikis IE, Holick MF 1998 Redefining vitamin D insufficiency. Lancet 351:805–806. 29. Lips P 2001 Vitamin D deficiency and secondary hyperparathyroidism in the elderly: consequences for bone loss and fractures and therapeutic implications. Endocr Rev 22:477–501. 30. Chapuy MC, Preziosi P, Maaner M, Arnaud S, Galan P, Hercberg S, et al. 1997 Prevalence of vitamin D insufficiency in an adult normal population. Osteopor Int 7:439–443. 31. Tangpricha V, Pearce EN, Chen TC, Holick MF 2002 Vitamin D insufficiency among free-living healthy young adults. Am J Med 112:659–662. 32. Kreiter SR, Schwartz RP, Kirkman HN, Charlton PA, Calikoglu AS, Davenport M 2000 Nutritional rickets in African American breast-fed infants. J Pediatr 137:2–6. 33. Welch TR 2000 Vitamin D deficient rickets: the reemergence of a once-conquered disease J Pediatr 137:143–145. 34. Thomas MK, Lloyd-Jones DM, Thadhani RI, Shaw AC, Deraska DJ, Kitch BT 1998 Hypovitaminosis D in medical inpatients. N Engl J Med 338:777–783. 35. Sullivan SS, Rosen CJ, Chen TC, Holick MF 2003 Seasonal changes in serum 25(OH)D in adolescent girls in Maine. J Bone Mine Res (Proceedings for Twenty-Fifth Annual Meeting of the American Society for Bone and Mineral Research), S407. 36. Holick MF 2003 Vitamin D: a millennium perspective. J Cell Biochem 88:296–307. 37. Heaney RP, Davies KM, Chen TC, Holick MF, Barger-Lux MJ 2003 Human serum 25-hydroxycholecalciferol response to extended oral dosing with cholecalciferol. Am J Clin Nutr 77:204–210. 38. Nesby-O’Dell S, Scanlon KS, Cogswell ME, Gillespie C, Hollis BW, Looker AC 2002 Hypovitaminosis D prevalence and determinants among African American and white women of reproductive age: third national health and nutrition examination survey, 1988–1994. Am J Clin Nutr 76:187–192. 39. Holick MF 2001 Sunlight “D”ilemma: risk of skin cancer or bone disease and muscle weakness. Lancet 357:4–6. 40. Glerup H, Mikkelsen K, Poulsen L, Hass E, Overbeck S, Andersen H et al. 2000 Hypovitaminosis D myopathy without osteomalacic bone involvement. Calcif Tissue Int 66:419–424. 41. Chapuy MC, Schott AM, Garnero P, Hans D, Delmas PD, Meunier J, et al. 1996 Healthy elderly French women living at home have secondary hyperparathyroidism and high bone turnover in winter. J Clin Endocrinol Metab 81:1129–1133. 42. Malabanan AO, Turner AK, Holick MF 1998 Severe generalized bone pain and osteoporosis in a premenopausal black female: Effect of vitamin D replacement. J Clin Densitometr 1:201–204.

MICHAEL F. HOLICK

43. Rimaniol J, Authier F, Chariot P 1994 Muscle weakness in intensive care patients: initial manifestation of vitamin D deficiency. Intensive Care Med 20:591–592. 44. Bischoff HA, Stähelin HB, Dick W, Akos R, Knecht M, Salis C, et al. 2003 Effect of vitamin D and calcium supplementation on falls: a randomized controlled study. J Bone Miner Res 18:343–351. 45. Garland CF, Garland FC, Shaw EK, Comstock GW, Helsing KJ, Gorham ED 1989 Serum 25-hydroxyvitamin D and colon cancer: Eight-year prospective study. Lancet 18:1176–1178. 46. Garland CF, Garland FC, Gorham ED, Raffa J 1992 Sunlight, Vitamin D, and Mortality from Breast and Colorectal Cancer in Italy. Biologic Effects of Light. Walter de Gruyter, New York, pp. 39–43. 47. Hanchette CL, Schwartz GG 1992 Geographic patterns of prostate cancer mortality. Cancer 70:2861–2869. 48. Ahonen MH, Tenkanen L, Teppo L, Hakama M, Tuohimaa P 2000 Prostate cancer risk and prediagnostic serum 25-hydroxyvitamin D levels (Finland). Cancer Causes Control 11:847–852. 49. Grant WB 2002 An ecologic study of dietary and solar ultraviolet-B links to breast carcinoma mortality rates. Am Cancer Soc 94:272–281. 50. Hypponen E, Laara E, Jarvelin M-R, Virtanen SM 2001 Intake of vitamin D and risk of type 1 diabetes: a birth-cohort study. Lancet 358:1500–1503. 51. Rostand SG 1979 Ultraviolet light may contribute to geographic and racial blood pressure differences. Hypertension 30:150–156. 52. Li YC, Kong J, Wei M, Chen ZF, Liu SQ, Cao LP 2002 1,25dihydroxyvitamin D3 is a negative endocrine regulator of the renin–angiotensin system. J Clin Invest 110: 229–238. 53. Scragg R, Jackson R, Holdaway IM, Lim T, Beaglehole R 1990 Myocardial infarction is inversely associated with plasma 25-hydroxyvitamin D3 levels: a community-based study. Int J Epidemiol 19:559–563. 54. Zitterman A, Schulze Schleithoff S, Tenderich C, Berthold H, Koefer R, Stehle P 2003 Low vitamin D status: a contributing factor in the pathogenesis of congestive heart failure? J Am Coll Cardiol 41:105–112. 55. Holick MF 2002 Sunlight and vitamin D, both good for cardiovascular health [editiorial] J Gen Intern Med 17: 733–735. 56. Housman TS, Feldman SR, Williford PM, Fleischer Jr AB, Goldman ND, Acostamadiedo JM 2003 Skin cancer is among the most costly of all cancers to treat for the Medicare population. J Am Acad Dermatol 48:425–429. 57. Garland CF, Garland FC, Gorham ED. Rising trends in melanoma. An hypothesis concerning sunscreen effectiveness. Ann Epidemiol 3:103–110. 58. Veierod MB, Weiderpass E, Thorn M, Hansson J, Lund E, Armstrong B, Adami H-O 2003 A prospective study of pigmentation, sun exposure, and risk of cutaneous malignant melanoma in women. J Nat Can Inst 95:1530–1538. 59. Grant WB 2002 An estimate of premature cancer mortality in the U.S. due to inadequate doses of solar ultraviolet-B radiation. Cancer 70:2861–2869. 60. Holick MF, Jenkins M 2003 The UV Advantage. iBooks, New York 61. Holick MF 2002 Vitamin D: the underappreciated D-lightful hormone that is important for skeletal and cellular health. Curr Opin Endocrinol Diabetes 9:87–98. 62. Chel VGM, Ooms ME, Popp-Snijders C, Pavel S, Schothorst AA, Meulemans CCE, et al. 1998 Ultraviolet irradiation corrects vitamin D deficiency and suppresses

CHAPTER 3 Photobiology of Vitamin D

secondary hyperparathyroidism in the elderly. J Bone Miner Res 13:1238–1242. 63. Chuck A, Todd J, Diffey B 2001 Subliminal ultraviolet-B irradiation for the prevention of vitamin D deficiency in the elderly: a feasibility study. Photochem Photoimmun Photomed 17:168–171. 64. Lund B, Sorensen OH 1979 Measurement of 25-hydroxyvitamin D in serum and its relation to sunshine, age, and vitamin D intake in the Danish population. Scand J Clin Lab Invest 39:23–30.

45 65. Adams JA, Clemens TL, Parrish JA, Holick MF 1982 Vitamin-D synthesis and metabolism after ultraviolet irradiation of normal and vitamin-D-deficient subjects. N Engl J Med 306:722–725. 66. Koutkia P, Lu Z, Chen TC, Holick MF 2001 Treatment of vitamin D deficiency due to Crohn’s disease with tanning bed ultraviolet B radiation. Gastroenterology 121:1485–1488. 67. Reid IR, Gallagher DJA, Bosworth J 1985 Prophylaxis against vitamin D deficiency in the elderly by regular sunlight exposure. Age Ageing 15:35–40.

CHAPTER 4

The Vitamin D 25-Hydroxylase MARIELLE GASCON-BARRÉ

Département de Pharmacologie, Faculté de Médecine, Université de Montréal, and Centre de recherche de l’Université de Montréal, Montréal, Québec Canada

I. Introduction II. Hepatic Uptake of Vitamin D III. The Monooxygenases Active in the Hydroxylation of Vitamin D at C-25 IV. The Microsomal Enzymes V. The Mitochondrial Enzyme

VI. Cerebrotendinous Xanthomathosis VII. Ontogeny of the Vitamin D 25-Hydroxylases VIII. Sex Differences in the Hydroxylation of Vitamin D at C-25 IX. Conclusions References

I. INTRODUCTION

mitochondrial enzymes, believed by many to be more physiologically relevant than the mitochondrial entity. In this chapter, we review the most relevant research area on the D 25-hydroxylases and address the specificity and regulation of each enzyme in the context of its dynamic functioning, including uptake, ontogeny, sex-related differences, and hepatic and intrahepatic regionalization.

25-Hydroxyvitamin D (25OHD)* is the first hydroxylated metabolite of vitamin D (D) and the immediate precursor of the fully active and hormonal form of the vitamin, 1α,25-dihydroxyvitamin D [1,25(OH)2D]. It was discovered by DeLuca and his group, who rapidly identified the liver as the first site of activation of D3 [1–3]. Over the past 35 years, the enzyme systems involved in the C-25 hydroxylation of D3, D2, and several of their analogs have been the object of intense studies by groups in North America, Europe, and Japan. The research has allowed the identification of two intrahepatic organelles, the smooth endoplasmic reticulum (microsomes) and the mitochondrium, as sites possessing fully active but distinct D 25-hydroxylases. The mitochondrial enzyme has been cloned [4–6] and its identity as a D3 25-hydroxylase established with certainty. Moreover, its presence and activity has been positively identified in all species studied including the human [7]. The microsomal enzyme received the attention of early workers in the field. It has been identified clearly in the pig, where the enzyme has been cloned and clearly shown to hydroxylate D3 and D2 at C-25 [8]. Lately, a new microsomal D 25-hydroxylase species has also been cloned and its gene transcript shown to be present in mice and humans [9]. The latter is also reported to be active in the C-25 hydroxylation of both D3 and D2. In most species, early work has shown that the microsomal D3 25-hydroxylase is an enzyme also active in the oxidation of several endogenous and exogenous substances and, based on the enzyme kinetics of the respective microsomal and *The

term vitamin D includes both vitamin D3 and vitamin D2.

VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

II. HEPATIC UPTAKE OF VITAMIN D A. The First Prerequisite Step toward Its Activation The hepatic sequestration of vitamin D is the first prerequisite step toward its activation and subsequent C-25 hydroxylation by the liver. Under normal conditions, the fractional hepatic uptake of D3 during a single pass across the rat and dog liver lies between 40 and 60% [10–12], an uptake higher than that observed for all hydroxylated metabolites of the vitamin [10,13,14]. Its hepatic clearance is estimated to be 357 ml min−1 in dogs [11]. Studies in which total uptake has been investigated for periods varying from 18 sec to 70 min have revealed that the liver does not accumulate significantly more D3 than that observed during the first pass across the organ [13,15–19] suggesting a steady state liver/ serum concentration ratio of 0.4 to 0.6 at physiological concentrations of the vitamin. In addition, data available indicate that there is no stringent regulation of uptake by the vitamin D status [13]. Both hepatocytes and nonparenchymal liver cells are able to sequester D3 and the mitochondrial D3 25-hydroxylase gene transcript has been found in sinusoidal endothelial, stellate, and Kupffer cells of the liver [20]. In a 1981 study, Dueland et al. [16], however, found that only hepatocytes were able to transform the vitamin into 25OHD3. Copyright © 2005, Elsevier, Inc. All rights reserved.

48 The uptake of D3 is not regioselective within the hepatic acinus and not perturbed by the destruction of either the periportal (proximal) or perivenous (distal or pericentral) region [21], indicating that its extraction takes place according to the concentration gradient along the acinus. This observation suggests that the intrahepatic concentration of the vitamin should be higher in periportal than in perivenous hepatocytes. This notion may be important in light of the reported intraacinar localization of the mitochondrial D3 25-hydroxylase, the kinetic parameters of the mitochondrial and microsomal enzymes, and the influence of regioselective liver diseases on the biosynthesis of 25OHD3 (see Chapter 75 for review). In addition, the hepatic extraction of D3 has been shown to be independent of its hepatic venous or arterial route of delivery [22,23], indicating that circulating vitamin D of endogenous or exogenous origin should have similar hepatic availability. In addition, manipulation of the diet in order to obtain different proportions of putative D3 carrier proteins or lipoproteins in vivo was found not to significantly affect the hepatic extraction of D3 when evaluated in isolated perfused liver preparations [22], an observation also made in vitro by Ziv et al. [19]. However, Haddad et al. [15] found that chylomicron remnants or LDL optimized the hepatic uptake of the vitamin in vitro.

III. THE MONOOXYGENASES ACTIVE IN THE HYDROXYLATION OF VITAMIN D AT C-25 Early studies rapidly established that the liver was the site of hydroxylation of D3 at C-25 [3,22,24]. The first experimental evidence obtained indicated that the D3 25-hydroxylase was localized in both the microsomal and mitochondrial fractions of liver homogenates [25]. However, it was soon reported that the enzyme was instead found almost entirely in liver microsomes and that mitochondria exhibited no or very low activity toward D3 [26]. It was also rapidly established that oxygen-18 from molecular oxygen was incorporated at C-25, strongly suggesting that the microsomal enzyme was a mixed-function oxidase [27]. Firm demonstration for the presence of a mitochondrial D3 25-hydroxylase only came several years later [7]. In 1978, studies by Suda’s group using isolated–perfused rat liver preparations demonstrated that the activation of 1αOHD3 was not regulated (Km 2.0 µM) compared to that of the natural substrate, and that two distinct Km values could be defined for the C-25 hydroxylation of D3 (5.6 nM and 1 µM), suggesting the participation of more than one D3 25-hydroxylase [28].

MARIELLE GASCON-BARRÉ

Evaluation of enzyme activity revealed that 25OHD3 production was also present in extrahepatic tissue [29,30]. At that time, support for the extrahepatic presence of a D3 25-hydroxylase came from the in vivo studies of Ponchon et al. [2] and Olson et al. [3], who reported that hepatectomy markedly reduced but did not eliminate the formation of 25OH[3H]D3 after [3H]D3 injection. They, indeed, observed that 10% of the radioactivity appearing in the plasma of intact rats was found as [3H]25OHD3 in hepatectomized animals 4 hr after injection of the parent compound. These observations have now been confirmed at the molecular level with the demonstration of the presence of the mitochondrial D3 25-hydroxylase in numerous tissues and organs (Table I). In the pig, the microsomal D3 25-hydroxylase gene transcript and its associated protein have also been shown to be present not only in the liver but also in several other organs including the kidney [31,32], whereas the gene expression of a mouse and human microsomal D 25-hydroxylase has been reported to be mainly present in the liver and testis (Table II) [9]. The contribution of the extrahepatic enzymes to the circulating concentrations of 25OHD under normal physiological conditions is not, however, presently known, but several laboratories have clearly shown that the mitochondrial and the porcine microsomal enzymes are active in extrahepatic tissues. These observations indicate that the enzymes participate in the local production of 25OHD and possibly, as well, in the production of 1α,25(OH)2D as both enzymes have been reported to also harbor 1α-hydroxylase activity as indicated in Tables III and IV.

IV. THE MICROSOMAL ENZYMES DeLuca’s group was the first to report on the subcellular localization of the D3 25-hydroxylase. Indeed, in the mid-1970s, Bhattacharyya and DeLuca [26,33,34] observed that the D3 25-hydroxylase was present in the microsomal fraction of rat liver and that it was influenced by the vitamin D status and required the presence of a cytosolic factor for optimum activity. They also noticed that the 25-hydroxylation of the synthetic compound dihydrotachysterol (DHT3) was not regulated compared to that of the natural substrate D3 [34]. In 1979, Madhok and DeLuca [35] confirmed the presence of the enzyme in rat liver microsomes and soon after, purification and reconstitution of the enzyme was achieved in Björkhem’s [36] and DeLuca’s [37] laboratories. It was shown to be a cytochrome P450 (P450, CYP) mixed-function oxidase requiring NADPH and NADPH cytochrome c/P450 reductase. Yoon and DeLuca [37] also reported that a soluble

49

CHAPTER 4 The Vitamin D 25-Hydroxylase

TABLE I Expression Sites of the Mitochondrial CYP27A1 Species

Location

Humana

Liver (adult and fetal)

Mouse

Rabbit

Rat

Reference

Cali and Russell, 1991 [6] Gascon-Barré et al. [141] Kidney (adult and fetal) Gascon-Barré et al., 2001 [141] HepG2 hepatoma cells Guo et al., 1993 [87] Fibroblastsc Cali and Russell 1991 [6] Macrophages Hansson et al., 2003 [165] Stroma cells of the ovary Jabara et al., 2003 [166] Arterial endothelial cells Javitt et al., 2002 [167] Keratinocytes Lehmann et al., [168] Parathyroid glands Schuessler et al. [169] Parathyroid tumors Correa et al., [170] Intestine (fetal jejunum Theodoropoulos et al. [120] and colon) Liver Ichikawa et al., 1995 [171] Duodenum Calvaria Lung Skin Long bone Spleen Osteoblasts (primary culture) Liver Andersson et al., 1989 [4] Duodenum Adrenal Lung Kidney Spleen Liverd Usui et al., 1990 [5] Twisk et al., 1995c [118] Liver sinusoidal Theodoropoulos et al. [20] endothelial, stellate and Kupffer cells Intestine Theodoropoulos et al. [119] Ovaries Su et al., 1990 [86] Kidney Mullick et al., 1995 [91] Axén et al., 1995 [112]

cytosolic factor was necessary for full reconstitution of enzyme activity (see Table III). In 1983, Andersson et al. [38] purified to homogeneity a rat microsomal cytochrome P450 active on the C-25 hydroxylation of the bile acid intermediates 5β-cholestane-3α,7α,12α-triol (cholestane-triol, C-triol), 5β-cholestane-3α,7α-diol (C-diol) as well as D3, 1αOHD3 but not D2. A year later, Hayashi et al. [39] purified and partially sequenced a P450 active on the 25-hydroxylation of D3 and suggested that an inhibitor was removed from microsomes during the purification steps as previously suggested by others [36,40]. Moreover, contrary to earlier observations, the reconstituted enzyme did not require the presence of a cytoplasmic factor. The partial amino acid sequence obtained indicated that the enzyme was different from P450 (PB-1) and P450 (MC-1) (CYP2B1 and 1A1, respectively, according to P450 nomenclature [41]). The same group later obtained an N-terminal amino acid sequence and reported that their purified P450 was also active on several endo- and xenobiotics as already reported by others [38,42] as well as D3, 1αOHD3 but

TABLE II Species CYP2D25 Pig

CYP2R1 Mouse

In addition to these sites, enzyme activity has been shown in a human macrophages [172] and b bovine endothelial cells [173]. c SV40 transformed; d in situ hybridization illustrates a preferential perivenous localization within the acinus [118].

Human

Expression Sites of the Microsomal CYP2D25 and CYP2R1 Location

Reference

Liver Kidney Adrenals Brain Heart Intestine Lung Muscle Spleen Thymus

Hosseinpour and Wikvall [52]

Liver Testis Kidney Brain Epididymis Lung Spleen Heart Muscle Skin Liver

Cheng et al. [9]

Cheng et al. [9]

50

MARIELLE GASCON-BARRÉ

TABLE III Species

Sex

Substrate

The Microsomal Vitamin D 25-Hydroxylases Hydroxylation Site

Characteristics

Reference

Km: 180 nMa (D-depleted) 444 nMa (D-repleted) Km: 44 nMa 360 nMa 108 to 205 pmol 30 min−1 0.75 nmol P450−1b

Delvin et al., 1978 [174]

Rat

D3

C-25

Rat

D3 DHT3 D3

C-25 C-25 C-25

D3 D3 1αOHD3 D2 Benzphetamine D3

C-25 C-25 C-25 C-25 Demethylation C-25

D3 1αOHD3 D2 C-triole C-diol f Ethylmorphine D3

C-25 C-25 C-25 C-25 C-25 Demethylation C-25

D3 1αOHD3 Testosterone

Rat

Dehydroepiandrosterone Benzphetamine D3

C-25 C-25 C-2α C-16α C-16α

7.68 pmol hr −1 mg pro−1b,c 4.3 pmol min −1 nmol P450−1b 3 pmol min −1 nmol P450−1b 2 pmol min −1 nmol P450−1b 921 pmol min −1 nmol P450−1b 3.5–4 pmol hr −1106 hepatocytesd Km: 4 µM (D-depleted) 6 µM (D-repleted) 335 pmol min−1 nmol P450−1b 1000 pmol min−1 nmol P450−1b <10 pmol min−1 nmol P450−1b 2600 pmol min−1 nmol P450−1b 360 pmol min−1 nmol P450−1b 21,000 pmol min−1 nmol P450−1b 2.3 nmol min−1 mg pro−1b (152 pmol min−1 nmol P450) 0.21 nmol min−1 nmol P450−1b 1.73 nmol min−1 nmol P450−1b 9.34 nmol min−1 nmol P450−1b 8.36 nmol min−1 nmol P450−1b 14.6 nmol min−1 nmol P450−1b

Demethylation C-25

15.6 nmol min−1 nmol P450−1b 195 pmol min−1 nmol P450−1b

Rat

C-triol Testosterone D3 C-triol* C-diol** D3

C-25 C-16α C-25 C-25 C-25 C-25

580 pmol min−1 nmol P450−1b 1283 pmol min−1 nmol P450−1b 9 pmol min−1 nmol P450−1b 90 pmol min−1 nmol P450−1b 188 pmol min−1 nmol P450−1b 335 pmol min−1 mg pro−1b (167 pmol min−1 nmol P450−1)

D3

C-25

1αOHD3 1αOHD3 D3

C-25 C-25 C-25

6 pmol min−1 mg pro−1b (4 pmol min−1 nmol P450−1) 0.152 nmol min mg pro−1a 0.053 nmol min mg pro−1a 37 pmol min 106 hepatocytesd 29 pmol min 106 hepatocytesd

D3

C-25 C-25 C-25

Rat (phenobarbital pretreated) Rat Beef

Rat

Rat

Rat Rat

Rat Rat (D depleted) (1,25(OH)2D3treated) Rat Human

,

1αOHD3

0.46 pmol hr1 mg pro−1a,c 0.40 pmol hr1 mg pro−1a 0.8–4.7 pmol min−1 mg proa

Madhok and DeLuca, 1979 [35] Björkhem et al., 1979 [36]

Yoon and DeLuca, 1979 [37] Hiwatashi et al., 1980 [42]

Dueland et al., 1981 [16]

Andersson et al., 1983 [38]

Hayashi et al., 1984 [39] Hayashi et al., 1986 [43]

Andersson and Jörnvall, 1986 [45]

Dählback and Wikvall, 1987 [46]

Saarem and Pedersen, 1987 [47] Benbrahim et al., 1988 [175]

Thierry-Palmer et al., 1995 [49]

Continued

51

CHAPTER 4 The Vitamin D 25-Hydroxylase

TABLE III Species

Sex

Pig

,

Piglet (normal) Piglet (PDDR) Piglet (normal) Piglet (PDDR) Piglet (PDDR)

,

Castrated

COS cells (pCYP2D25)

Substrate

Hydroxylation Site

D3 D2 1αOHD3 1αOHD2 25OHD3

C-25 C-25 C-25 C-25 C-1α

1αOHD3 D3

C-25

D2 1αOHD3 25OHD3 1αOHD3 25OHD3 25OHD3 D3 D2

pCYP2D25

m/hCYP2R1

The Microsomal Vitamin D 25-Hydroxylases—Cont’d

,

Characteristics 188 pmol min−1 nmol P450−1 e 98 pmol min−1 nmol P450−1e 756 pmol min−1 nmol P450−1 f 842 pmol min−1 nmol P450−1 f 6 pmol min−1 nmol P450−1e 6 pmol min−1 nmol P450−1e 3 pmol min−1 nmol P450−1 f 1 pmol min−1 nmol P450−1 f 760 pmol min−1 nmol P450−1a 200 pmol min−1 nmol P450−1a Km: 100 nM 110 pmol min−1 nmol P450−1a 475 pmol 48 h−1 mg cell pro−1g 45 pmol 48 h−1 mg cell pro−1g 57 pmol min−1 nmol CYP2D25−1 12 pmol min−1 nmol CYP2D25−1 45 pmol min−1 nmol CYP2D25−1 ND ND

C-25 C-1α C-25 C-1α C-26 C-25 C-25

Reference Axén et al. [51]

Axén et al. [53]

Hosseinpour and Wikvall [52] Hosseinpour et al. [32] Hosseinpour et al. [32] Araya et al. [57]

Cheng et al. [9]

a Liver f Partially

microsomes. b Reconstituted system. cIn the presence of a cytosolic factor. d Freshly isolated hepatocytes. e Partially purified liver microsomal P450. purified kidney microsomal P450. g Kidney pCYP2D25. *5β-cholestane-3α,7α,12α-triol; **5β-cholestane, 3α,7α,12α-diol. ND, Not determined.

TABLE IV Species

Sex

Rat (untreated)

Substrate D3 1αOHD3 DHT3 C-triol* D3 D3

The Mitochondrial Vitamin D 25-Hydroxylase Hydroxylation Site

Rat

D3 D2 D3

C-25 C-25 C-25 C-26(27) C-25 C-25 D3 C-triol* C-25 C-25 C-25 C-25 C-25

Rat

D3

C-25

(Phenobarbital treated) (Rachitagenic diet) Human Human Rat Human

and —

D3

C-25 Rabbit

D3 1αOHD3 25OHD3 C-triol*

C-25 C-25 C-1α C-27

Characteristics 0.02 nmol min−1 nmol P450−1b 0.116 nmol min−1 nmol P450−1b 0.01 nmol min−1 nmol P450−1b 8.7 nmol min−1 nmol P450−1b 0.03 nmol min−1 nmol P450-1b 0.064 nmol min−1 nmol P450−1b 0.16 nmol min−1 nmol P450−1b 40 nmol min−1 nmol P450−1b Km: 10−5 Ma Km: 2 × 10−5 Ma,c 10 pmol min−1 mg pro−1a 2 pmol min−1 mg pro−1a 0.093 nmol hr −1 mg pro−1a 0.435 nmol hr −1 mg pro−1a 550 pmol min −1 mg pro−1a (305 pmol min nmol P450−1a) 525 pmol min nmol P450−1a 350 pmol min nmol P450−1a 395 pmol min nmol P450−1b 1200 pmol min nmol P450−1b < 10 pmol min nmol P450−1b 2500 pmol min nmol P450−1b

Reference Björkhem et al., 1980 [110]

Oftebro et al., 1981 [83] Saarem et al., 1984 [84] Saarem and Pedersen, 1985 [85] Holmberg et al., 1986 [176] Saarem and Pedersen, 1987 [35] Dählback and Wikvall, 1987 [46]

Dählback and Wikvall, 1988 [111]

Continued

52

MARIELLE GASCON-BARRÉ

TABLE IV Species

Sex

Substrate

The Mitochondrial Vitamin D 25-Hydroxylase—Cont’d Hydroxylation Site

Rat

1αOHD3

C-25

Rat Rat

hCYP27A1

C-triol* D3 1αOHD3 C-triol* D2 D3 DHD3 1αOHD2 1αOHD3 1αOHDHT3 D3 1αOHD3 25OHD3 25OHD3 C-triol* 25OHD3

hCYP27A1

D3

C-27 C-25 C-25 C-27 C-25 C-25 C-25 C-25 C-25 C-25 C-25 C-25 C-1α C-27 C-27 C-1α C-27 C-25

25OHD3

C-1α

1αOHD3

C-25 (major) C-24 (minor) C-26 (minor) C-24 (major) C-26 (minor) C-24 (major) C-26 (major) C-25 (minor) C-26 (major) C-25 (minor)

Cos cellsc

E. colid

HepG2

1αOHD2 1αOHD4 24-epi1α OHD3 ∆22-1αOHD3

Characteristics

Reference

Km: 54 µMb 3.8 nmol min−1 nmol P450−1b Km: 6.3 µMb 0.36 nmol min−1 nmol P450−1b 1.4 nmol min−1 nmol P450−1b 36 nmol min−1 nmol P450−1b N/Db 342 pmol hr−1106 cellsb 344 pmol hr−1106 cellsb 16 pmol hr−1106 cellsb 1328 pmol hr−1106 cellsb 1383 pmol hr−1106 cellsb 72 pmol min−1 nmol P450−1b 189 pmol min−1 nmol P450−1b 24 pmol min−1 nmol P450−1b 19 pmol min−1 nmol P450−1b 4337 pmol min−1 nmol P450−1b 5 pmol hr nmol hr min−1 CYP27A1−1 4.5 pmol hr nmol hr min−1 CYP27A1−1 Km = 3.2 µM Vmax = 0.27 mol CYP27A1−1 min−1 Km = 3.5 µM Vmax = 0.021 mol CYP27A1−1 min−1

Masumoto et al., 1988 [80] Okuda et al., 1988 [76] Ohyama et al., 1991 [81]

Guo et al., 1993 [87]

Axén et al., 1994 [99]

Araya et al., [57] Sawada et al., [101]

Byford et al., [105]

C-24 (major) C-26 (major) C-25 (minor)

a Mitochondria; b Reconstituted system. c In the presence of a cytosolic factor. **5β-cholestane, α,7α,12α-diol. N/D, Not detectable.

not D2 [43], further demonstrating the broad spectrum of activity of the enzyme. Sex-related differences have also been observed in the microsomal C-25 hydroxylation of D3. Indeed, Hayashi et al. [43] isolated a D3 25-hydroxylase exhibiting the same N-terminal amino acid sequence as CYP2C11 [44], a P450 known to exist in male rat liver microsomes but not in those of females. This observation may explain why microsomes (or P450 purified from the microsomal fraction) obtained from female livers have consistently been shown, as indicated in Table III, to exhibit significantly less activity in the C-25 hydroxylation of D3 than preparations

d Transfected

with hCYP27. *5β-cholestane-3α,7α,12α-triol;

obtained from their male counterparts [43,45–48] except, possibly, at low substrate concentrations as reported by Thierry-Palmer et al. [49]. Whether Yoon and DeLuca [37] and Andersson et al. [38] reported on the same isoenzyme as Hayashi et al. [43] is not known, as no amino acid sequence was proposed in the former studies, but Andersson and Jörnvall [45] later proposed a partial sequence suggestive of the malespecific testosterone 16α-hydroxylase, which is now known to be a P450 compatible with the male-specific CYP2C11. Using male pig microsomes, Segura-Aguilar [50] reported that the C-25 hydroxylation of D3 is also

CHAPTER 4 The Vitamin D 25-Hydroxylase

catalyzed by a peroxidase as indicated by the use of t-butyl hydroperoxide as electron donor in the absence of NADPH and NADPH-P450 reductase. The Km of the enzyme was found to be 30 µM, a concentration much higher than the reported values in the presence of NADPH and NADPH-P450 (see Table III).

A. The Porcine CYP2D25 Purification of the porcine microsomal D3 25-hydroxylase was achieved between 1992 and 1994 by Axén et al. [31,51]. At the time, the protein was only found in liver and kidney. However, the monoclonal antibody raised against the porcine enzyme also recognized a protein of similar molecular weight in human liver but, to date, the pig homolog of the enzyme has not been cloned in any other species. The purified enzyme was shown to catalyze the C-25 hydroxylation of vitamin D3 and vitamin D2 and both compounds competitively inhibited the C-25 hydroxylation of one another. Moreover, ketoconazole, a nonselective P450 inhibitor, also inhibited the hydroxylation of D3 as well as of D2 [51]. The enzyme has been shown to have a high affinity for D compounds and converts D3 into 25OHD3 with a reported Km of 100 nM, a concentration considered to be within the physiological range. Its order of efficiency at saturating concentrations (25 µM) is reported to be 1αD3 > D3 > D2 with 750, 200 and 110 pmol/nmol P450/min respectively [52]. A year later, Axén [31] showed that the enzyme also catalyzed the hydroxylation of 25OHD3 at C-1α, indicating that the enzyme was not only a D3 25-hydroxylase but also a microsomal 1α-hydroxylase. Interestingly, the enzyme was also shown to exhibit similar 1α-hydroxylation activity in the liver of normal pigs and of pigs suffering from pseudovitamin D deficiency rickets (PDDR) (see Chapter 71), but not in the kidney where the 1α-hydroxylation was significantly decreased by the PDDR genotype (Table III) [53]. The molecular cloning of the porcine hepatic microsomal D3 25-hydroxylase was achieved by Wikvall and his group in 1997 [8], and the gene was subsequently shown to map to chromosome 5 [54]. The gene transcript codes for a protein of 500 amino acids and has a predicted molecular weight of 56,374. At that time, Northern blot analysis revealed that the gene was found in high abundance in the liver, but the analysis also showed a clearly detectable signal in the kidney. Subsequent studies using RT-PCR revealed that the transcript is expressed in several tissues including the intestine, lung, thymus, adrenals, and the brain (Table II) [52]. Transfection in simian COS cells results in the synthesis of an enzyme that is recognized by a monoclonal antibody raised against purified D 25-hydroxylase and

53 catalyzes the C-25 hydroxylation of D. The gene transcript exhibits a 70-80% identity with members of the CYP2D family of P450 and has been termed CYP2D25. CYP2D25 has unique properties among the CYP2D enzymes. Indeed, although it metabolizes the drug tolterodine, a muscarinic receptor antagonist and a substrate for the human CYP2D6, into its 5-hydroxymethylated metabolite, tolterodine also inhibits the 25-hydroxylation of D3 while quinidine, an inhibitor of CYP2D6, does not significantly inhibit the C-25 hydroxylation of D3 [52]. Moreover, it was found that the residues in substrate recognition site 3 (SRS-3) are important determinants for its function as a D 25-hydroxylase but not for its activity in the metabolism of tolterodine [55]. Surprisingly, cyclosporine A, a known inhibitor of the mitochondrial D3 25-hydroxylase, was also shown to significantly inhibit the C-25 hydroxylation activity of CYP2D25 [52]. Phorbol 12-myristate 13-acetate (PMA) down-regulates the expression of CYP2D25 and the subsequent C-25 hydroxylation of 1αOHD3, suggesting the involvement of protein kinase C in the regulation of CYP2D25 [56]. The molecular cloning of the microsomal D3 25-hydroxylase in pig kidney was achieved by Hosseinpour et al. in 2002 [32]. Genomic Southern analysis suggests the presence of a single gene for CYP2D25 in the pig and the DNA sequence and deduced peptide sequence of the renal enzyme are homologous to those of the hepatic CYP2D25. Immunocytochemistry indicates that the enzyme is expressed almost exclusively in cells of the cortical tubules as illustrated in Fig. 1. The kidney enzyme has been shown to catalyze not only the hydroxylation of D3 at C-25 but also that of 25OHD3 at C-1α [32,57]. Moreover, recombinant CYP2D25 not only hydroxylates D3 at C-25 but also converts 25OHD3 into both 1α,25(OH)2D3 and 25,26(OH)2D3 with rates of conversion of 57, 12, and 45 pmol/nmol CYP2D25/min, respectively [57]. Microsomes obtained from human livers have been shown to hydroxylate 1αOHD3 at C-25. They, however, exhibit a wide range of activities (0.7 to 4.7 pmol/ min/mg protein). Despite a high degree of sequence identity between CYP2D25 and the human CYP2D6 (77% in its primary structure), no correlation between the microsomal CYP2D6 content and the capacity to hydroxylate 1αOHD3 at C-25 could be demonstrated [52]. In addition, the same group showed that CYP2D6 overexpressed in insect microsomes did not reveal any C-25 hydroxylation activity. Collectively, these observations clearly indicate that CYP2D6 is not a D 25-hydroxylase in human and definitive identification of the microsomal D3 25-hydroxylase, in species other than the pig, will require further kinetic, immunological, and molecular characterization. However, as indicated later, the recent identification of

54

MARIELLE GASCON-BARRÉ

A

chromosome 11, and the mouse CYP2R1 is reported to be 89% identical in sequence to the human enzyme. The enzyme is sexually dimorphic, is present in highest abundance in the liver and testis (Table II), and is active on both D2 and D3 compounds including 1αOHD3. Its sequence identity with other microsomal D 25-hydroxylases (CYP2D25, CYP2C11) has been reported to vary between 33% and 39% [9]. Biochemical characterization of the enzyme has, however, not yet been published and hence its substrate specificity is not yet known.

B

V. THE MITOCHONDRIAL ENZYME A. The C27 Sterol 27-Hydroxylase

FIGURE 1 Localization of the microsomal D3 25-hydroxylase in pig kidney. Immunohistochemical localization of CYP2D25 in sections of pig kidney incubated with polyclonal antibodies raised against the hepatic CYP2D25. Immunostaining (in red) was seen in the cells of proximal tubules but not in distal tubules (A and B). No staining was detected in glomeruli (B). Control sections not treated with primary antibodies, but otherwise treated in the same way, did not show immunostaining (not shown in figure). Reproduced from Biochimica et Biophysica Acta, 2002, volume 1580, 140 [32], by copyright permission of Elsevier Science B.V.

an orphan receptor of the CYP2 family as a putative D 25-hydroxylase may soon elucidate the role of the endoplasmic reticulum in the hepatic activation of vitamin D in humans.

B. The Human and Mouse CYP2R1 Using a cDNA library from CYP27A1 null mouse liver, Cheng et al. [9] have identified an evolutionarily conserved orphan cytochrome P450 termed CYP2R1 and demonstrated that the biochemical properties of the enzyme were consistent with a D 25-hydroxylase. The CYP2R1 gene transcript has characteristic sequence features associated with cytochromes P450 of the endoplasmic reticulum. It is conserved through evolution from fish to human [58], where it is located on

Bile acid biosynthesis represents the most important pathway in the metabolism and excretion of cholesterol. Although several enzymes participate in the oxidation of the cholesterol molecule, the C27 sterol hydroxylase catalyzes the first and probably the ratelimiting step in the oxidation of the side chain in the “acidic” bile acid biosynthesis pathway [59]. In accordance with IUPA sterol nomenclature [60] and the stereochemistry of the initial reaction catalyzed by the enzyme [61–63], Cali and Russell [6] termed the enzyme sterol 27-hydroxylase (EC 1.14.13.15), although the enzyme exhibits low substrate specificity and has been referred to as the 26-hydroxylase in most early publications [64]. CYP27A1 has now been adopted as the gene symbol. After considerable debate on the subcellular localization of the 27-hydroxylase [44,65,66], the enzyme was established in 1973 to be located in the inner membrane matrix of liver mitochondria and shown to display properties similar to that of P450, catalyzing hydroxylation in the mitochondrial membrane of the adrenal glands [67,68]. These observations were later confirmed by others [69–71]. Thus, sterol 27-hydroxylase is a mixed-function oxidase requiring the two mitochondrial-specific protein cofactors, ferrodoxin (adrenodoxin), an iron–sulfur protein [72], and ferrodoxin (adrenodoxin) reductase, a FAD-containing enzyme [73]. The enzyme catalyzes the C-27 hydroxylation of 5β-cholestane-3α,7α,12β-triol [70], 5β-cholestane3α,7α-diol [65], cholesterol [74–76], and 7-ketocholesterol [77], as well as several other substrates involved in bile acid biosynthesis. Wikvall [78] purified the enzyme to apparent homogeneity from rabbit liver mitochondria. Okuda et al. [76] using female rat liver mitochondria achieved final purification of the 27-hydroxylase and noticed that the enzyme exhibited no activity toward xenobiotics using benzphetamine, 7-ethoxycoumarin, and benzo[a]pyrene as substrates.

55

CHAPTER 4 The Vitamin D 25-Hydroxylase

In 1975, Björkhem et al. [79] observed the presence of the 27-hydroxylase in human liver mitochondria, and a few years later, the same group was the first to report that rat and human mitochondria catalyzed the C-25 hydroxylation of D3 but, at that time, attributed the D3 25-hydroxylation activity to an enzyme different from the cholestane-triol 27-hydroxylase [7].

B. The Mitochondrial D3 25-Hydroxylase Is the C27 Sterol 27-Hydroxylase Masumoto et al. [80] purified to homogeneity the D3 25-hydroxylase from rat liver mitochondria and characterized it as a P450 catalyzing the C-25 hydroxylation of both D3 and 1αOHD3. These researchers proposed that a single enzyme was involved in the C-27 hydroxylation of 5β-cholestane-3α,7α,12α-triol and in the C-25 hydroxylation of D3, although the preparation exhibited much lower activity toward D3 compounds than toward the C-27 hydroxylation of bile acid intermediates as illustrated in Table IV [76]. Ohyama et al. [81] pursued the hypothesis they earlier put forward [80] and found that D3 competitively inhibited the 27-hydroxylation of C-triol while C-triol inhibited D3 25-hydroxylation. They concluded, based on the criteria proposed by Haldane and Dixon [82], that both substrates were catalyzed at a common active site of a single protein. Meanwhile Oftebro et al. [83] and Saarem et al. [84,85] confirmed the original observation of Björkhem et al. [7] that a D3 25-hydroxylase was present in human liver mitochondria and that C27 sterol 27-hydroxylation as well as D3 25-hydroxylation activities were located probably exclusively in the inner mitochondrial membrane matrix [84,85]. Cloning of the rabbit [4], rat [5,86], human [6,87], and pig [8] CYP27A1 (chromosomal location: 2q33qter [88]) was achieved between 1989 and 1997. Cloning of the rabbit 27-hydroxylase by Andersson and co-workers [4] illustrated the hydrophobic nature of the protein (with 36% of the amino acids being either aromatic or hydrophobic side chains), and the presence of a conserved cysteine residue at position 444 that is believed to be the ligand for the heme ion in P450 enzymes. Andersson’s work also confirmed earlier biochemical observations indicating extrahepatic enzyme activity with the demonstration that CYP27A1 mRNAs were present in multiple tissues as compiled in Table I. The predicted amino acid sequence of the rabbit [4], rat [5], and human [6] mitochondrial D3 25-hydroxylase contains a mitochondrial specific presequence of 32–36 amino acids on which the destination of the hemoprotein is dependent, as illustrated by Sakaki et al. [89] using a modified CYP27A1 whose

mitochondrial targeting signal had been replaced by a microsomal N-terminal targeting signal. The construct, indeed, resulted in the microsomal localization of a fully active enzyme, providing adequate electron transfer was made available to the enzyme. The mature protein contains 444 to 501 amino acids depending on the species involved and the rat enzyme molecular weight has been estimated to be 51,182 [5]. Northern analyses demonstrate two mRNA sizes of 1.9 and 2.3 kb for CYP27A1 in liver and fibroblast cells [6,86]. Characterization of the 2.3-kb mRNA indicated that the sequence was identical to the 1.9-kb mRNA in its protein-coding region, except for a 400-nucleotide extended sequence at its 5′ end [90]. The rat CYP27A1 gene contains 11 exons of 80-415 nucleotides that are separated by 10 introns of 83 bases to 10kb [91]. The gene has been shown to be regulated by several hormones as will be discussed later [90–94]. The protein sequence of the human enzyme has been reported to be 72% identical to the rat and 81% identical to the rabbit CYP27A1 [5,6]. Heterologous cell systems confirmed the requirement for adrenodoxin and adrenodoxin reductase and not NADPH P450 reductase [4] as well as the identity of sterol 27-hydroxylase and D3 25-hydroxylase. Indeed, Usui et al. [95] prepared an expression plasmid encoding the rat liver D3 25-hydroxylase and transfected it in simian COS cells. They observed both sterol C-27 and D3 C-25 hydroxylation activities in a solubilized mitochondrial fraction supplemented with adrenodoxin and NADPH-adrenodoxin reductase. Akiyoshi-Shibata et al. [96] and Sakaki et al. [89] expressed the cDNA encoding the precursor protein of the rat liver D3 25-hydroxylase in Saccharomyces cerevisiae and found that the mitochondrial fraction exhibited both D3 25-hydroxylase and C-triol 27-hydroxylase activities. These data clearly confirmed the identity of the two enzymes and ruled out the possibility that the simian cell line contained endogenous 27-hydroxylase.

C. Specificity of the CYP27A1 CYP27A1 is a high-capacity enzyme involved in the hydroxylation of bile acid intermediates, but its substrate specificity is broad and exceeds the field of bile acid biosynthesis. As discussed ealier, the purified enzyme has been shown to be active on both the C-25 and C-27 hydroxylation of cholesterol [44,64,68], and to hydroxylate 5β-cholestane-3α,7α-diol and 5β-cholestane- 3α,7α,12α-triol at C-27. The enzyme has also been shown to carry out the oxidation of the product of the reaction, 5β-cholestane-3α,7α,12α,27-tetrol, to

56 3α,7α,12α-trihydroxy-5β-cholestanoic acid [97], as well as to substitute for the classical pathway of bile acid biosynthesis involving an NAD-coupled alcohol dehydrogenase. In a manner similar to other mitochondrial P450, CYP27A1 contains a proline-rich sequence located in the amino-terminal region of the molecule that serves to reduce the tendency of the polypeptide to misfold prior to heme binding [98]. The rabbit, rat, human, pig, and bovine enzyme hydroxylates D3 compounds at C-25 as well as at many other positions on the side chain of the molecule [5,86,87,99–101]. Indeed, CYP27A1 has been shown to catalyze the C-1α, 24, 25, 26, and 27 hydroxylation of D3, D2 and related compounds [87,100–105]. In addition, the pig and human kidney CYP27A1 catalyzes the formation of a metabolite expected to be, according to its mass spectrometric properties, 4β,25(OH)2D3 [57], a metabolite already reported by others in the rat [106,107]. While D3 substrates were found to be preferably hydroxylated at C-25, several groups have observed that when the substrate exhibits the ergocalciferol side chain such as that found in D2 or 1αOHD2, a predominance of 24-hydroxy metabolites occurs with some production of 26- and 27-hydroxylated products. These metabolites have also been found in vivo in rat, cow, chicken, and human [87,102,103,108,109]. Surprisingly, D2 is not hydroxylated at position 25 while 1αOHD2 is only poorly hydroxylated as indicated in Table IV. In addition, several laboratories have now shown that the enzyme prefers 1α-hydroxylated analogs of D3 over their nonhydroxylated counterparts, and this includes the natural substrate D3, which is hydroxylated at C-25 several times less efficiently than 1αOHD3 (see Table IV) [81,87,110,111]. The observation that the 27-hydroxylase is far more efficient at hydroxylating compounds exhibiting a 1α-hydroxyl group than toward the natural substrate is in full accordance with Ohyama et al. [81], who observed that the activity of the enzyme toward bile acid intermediates was proportional to the number of hydroxyl groups on the nucleus. Björkhem [64] also raised the hypothesis that the binding of nonpolar substrates such as vitamin D3 or cholesterol to the active site of the enzyme may differ from the corresponding binding of more polar substrates such as 5β-cholestane-3α,7α,12α-triol. Jones and co-workers [104] have also shown that lengthening the side chain of the vitamin D molecule is quite well tolerated in that the enzyme is able to continue to efficiently hydroxylate D3 compounds including its analogs. These studies led to the proposal of a model of interaction between enzyme and substrate whereby the 25-hydroxylase appears to be directed to its terminal hydroxylation site by the distance from the end of the side chain. These data were also taken to

MARIELLE GASCON-BARRÉ

suggest that the substrate binding pocket of CYP27A1 can tolerate a variety of steroidal shapes either by accommodating only the terminal carbons of the side chain, or by having a specific cleft for the side chain within a broader pocket for the vitamin D/steroidal ring structure. Using several side-chain analogs of D3, the same group showed that in the CYP27A1-containing hepatoma cell HepG2, 1αOHD3 is mainly hydroxylated at C-25 with only minor C-24 and C-26 hydroxylations. 1αOHD2, however, is mainly hydroxylated at C-24 with minor hydroxylation at C-26 while both 1αOHD4 and ∆22-1αOHD3 are mainly hydroxylated at C-24 and C-26 with only minor production of their 25-hydroxylated products. These observations led the authors to conclude that the C-24 methyl group plays a crucial role in the determination of the CYP27A1directed C-24 hydroxylation. They also concluded, based on their studies with 1αOHD4 and ∆22-1αOHD3, that the C22–C23 double bond also contributes to the altered metabolism observed with D2 or D2-derived compounds [105]. Lately, Wikvall and co-workers [57,99,112] have reported that the 27-hydroxylase purified from pig and rabbit livers as well as recombinant human CYP27A1 expressed in Escherichia coli or monkey COS cells was also able to catalyze the 1α-hydroxylation of 25OHD3, albeit at a much lower rate than that observed for the conversion of D3 into 25OHD3. Interestingly, 25OHD3 1α-hydroxylase activity has been reported previously in pig [113] and fish [114] as well as in fetal and adult rat livers [115,116]. Whether or not the reported activity was due to the 27-hydroxylase is unknown as the adult rat liver 1α-hydroxylation activity was found not in the mitochondrium but in the microsomal fraction [116].

D. Regulation of CYP27A1 and Enzyme Activity CYP27A1 mRNA t1/2 has been reported to be between 13 and 24 hr in the rat liver as well as in rat hepatocytes [20,117]. In addition, many hormones, endogenous products, hepatocyte nuclear factors (HNFs), and xenobiotics have been shown to regulate the gene encoding CYP27A1 as well as its gene product. They include bile acids, growth hormone, IGF-1, glucocorticoids, insulin, 1,25(OH)2D3, thyroxine, cyclosporine A, protease inhibitors, fibrates, lipopolysaccharides (LPSs), TNFα, IL-1β, HNF1, HNF4α, and the physiological state of the animal [20,86,91,93,94,118–126]. In the rat [92,93] but not the rabbit [127], CYP27A1 is highly sensitive to the prevailing concentrations of bile acids with increases in enzyme activity, steady-state

CHAPTER 4 The Vitamin D 25-Hydroxylase

mRNA levels, and the rate of gene transcription following interruption of the enterohepatic circulation of bile acids [93,118] while increases in bile acid concentrations are known to down-regulate CYP27A1 [93,126,128,129]. 7α(OH)-4-Cholesten-3-one is an inhibitor of CYP27A1 in the pig and, in a manner similar to the microsomal CYP2D25, PMA down-regulates the expression of CYP27A1, an observation suggesting the participation of protein kinase C in the regulation of its transcription [56]. Bile acids have been reported to suppress CYP27A1 transcription by decreasing its binding to HNF-1, a transcription factor expressed in liver, intestine, pancreas, and kidney [129,130]. Similarly, HNF4α, which is a strong stimulator of CYP27A1 gene transcription, is also down-regulated by bile acids. It has, therefore, been postulated that the inhibitory effect of bile acids on CYP27A1 could be mediated through a repression of HNF4α gene transcription and protein expression rather than through a direct effect on the bile acid response element (BARE). Cholic acid has also been reported to affect CYP27A1 mRNA stability [122]. In accordance with the latter observations, Bolt et al. [131] have shown that cholestasis induced by bile-duct ligation in rats significantly inhibited the 25-hydroxylation of D3 and that addition of bile salts to liver homogenates also had an inhibitory effect on the enzyme. By contrast, biliary cirrhosis does not seem to directly affect the total hepatic output of 25OHD3 in vivo. Indeed, Plourde et al. [132] observed by directly sampling the hepatic effluent after intraportal D3 injection that the 25-hydroxylation of D3 was similar in dogs with interruption of the enterohepatic circulation by choledococystostomy anastomosis and biliary cirrhosis secondary to bile-duct ligation. These observations still leave open any definitive conclusion as to the effect of bile salts on the C-25 hydroxylation of D3 by the mitochondrial and/or microsomal enzyme(s) in vivo (see Chapter 75). The role of insulin in CYP27A1 was illustrated when bile acid production via the sterol 27-hydroxylase was shown to be inhibited up to 58% after 24 hr incubation of hepatocytes with 140 nmol/liter insulin [94]. The decrease in enzyme activity could be explained by a concomitant reduction in CYP27A1 (–62%) mRNA level as well as in transcriptional activity (–75%) [94]. Thus physiological concentrations of insulin seem to down-regulate CYP27A1 gene transcription through a direct effect of the hormone on the hepatocytes [94], although the exact locus of interaction of insulin within the gene has not been identified. Avadhani’s group first reported regulation of CYP27A1 by pituitary-regulated steroids, growth hormone, and the diurnal rhythm with a twofold increase in enzyme activity observed in the mid-dark compared

57 to the mid-light period [93]. They subsequently showed that CYP27A1 is regulated by glucocorticoids as illustrated by a sevenfold induction in mRNA levels in isolated hepatocytes exposed to dexamethasone [117]. The response to the drug is now known to be due to a dexamethasone response element located in the promoter region of the CYP27A1 gene, which has been mapped to positions between −792 and −1095 bp [122,126]. In addition to glucocorticoids up-regulating the enzyme, other drugs known to regulate CYP27A1 include cyclosporine A, protease inhibitors, and fibrates (see also Chapters 73 and 74). At the enzyme level, cyclosporine A has been shown to inhibit the C-25 hydroxylation of both D3 and cholesterol through a direct inhibition of the gene encoding the sterol 27-hydroxylase [133,134]. Indeed, a cyclosporine A response element has been identified on the CYP27A1 gene promoter at 1087–4000 bp upstream of the putative ATG start site [122]. Protease inhibitors have also been shown to decrease the C-25 hydroxylation of D3 in a human hepatocyte cell line, suggesting that patients on antiviral therapy may have an increased risk of 25OHD insufficiency [123]. The fibrate hypolipidemic drugs, on the other hand, have been reported to decrease bile acid synthesis in rat hepatocytes and to suppress CYP27A1 mRNA and enzyme activity by transcriptional and posttranscriptional mechanisms involving the peroxisome proliferator–activated receptor α (PPARα) [124]. The presence of regulatory mechanisms related to the vitamin D status as modulators of the 25-hydroxylation of the vitamin has long been a widely accepted concept [30,33]. In addition, in the early 1980s, Baran and Milne [135] and Bell et al. [136] put forward the hypothesis that 1,25(OH)2D3 might be an inhibitor of the D3 25-hydroxylase. Later studies indicated that 1,25(OH)2D3 administration influenced the in vivo handling of D3 by accelerating its metabolism as well as by increasing its metabolic and biliary clearances, but the molecular mechanisms involved in the regulation of CYP27A1 by the D status were investigated much later [12,137–139]. Axén et al. [112] first reported that CYP27A1 located in kidney and liver was affected by 1,25(OH)2D3 administration, but they showed that CYP27A1 mRNA was decreased to a greater extent in the kidney than in the liver. On the other hand, Theodoropoulos et al. [20,119] investigated the effect of D3 nutritional status as well as that of 25OHD3 and 1,25(OH)2D3 on the rat hepatic and intestinal CYP27A1 and its associated enzyme activity. The hepatic CYP27A1 gene transcript was not found to be sensitive to the D3 or 25OHD3 status but was highly sensitive to 1,25(OH)2D3 exposure with a 60% decrease in its steady-state mRNA levels

58

MARIELLE GASCON-BARRÉ

and a parallel decrease in its protein and enzyme activity (Figs. 2 and 3) [20]. The effect of 1,25(OH)2D3 was shown to be mediated by a significant decrease in the CYP27A1 transcription rate, whereas its mRNA t1/2 remained unchanged. By contrast, intestinal CYP27A1 mRNA levels were shown to be sensitive to D3 and 25OHD3 as well as 1,25(OH)2D3 exposure. 1,25(OH)2D3 administration resulted in a decrease in both intestinal CYP27A1 mRNA t1/2 and transcription rate, and CYP27A1 enzyme activity was shown to be significantly decreased after 1,25(OH)2D3 administration in vivo [119]. In addition, in the human fetal intestine kept in primary culture, 1,25(OH)2D3 was shown to decrease CYP27A1 mRNA levels by 29% after 48 hr of incubation in the presence of the hormone [120]. Collectively, these data indicate that 1,25(OH)2D3

significantly influences CYP27A1 mRNA levels and enzyme activity in the major D-metabolizing organs as well as in target organs. It also illustrates that organspecific differences are present in the regulation of the CYP27A1 gene transcript and response to D3 or its metabolites. Interestingly, Chen and Chiang [126] have suggested that the vitamin D receptor (VDR) (which has now been shown to be present in the liver [140]) could be a mediator in the inhibition of CYP27A1 by conjugated bile acids.

E. Intraacinar Localization of CYP27A1 Several studies have indicated that the CYP27A1 gene transcript is heterogeneously distributed within the

A CYP27A

18S

B CYP27A/18S (Arbitary Units)

1.0

0.8

0.6

0.4

0.2

0.0 0

1

3

5

7

Length of repletion with 1,25 (OH)2D3 (days)

FIGURE 2

Influence of 1,25(OH)2D3 administration to vitamin D–depleted rats on the hepatic D3 25-hydroxylase gene transcript rCYP27A1. Steady-state levels of hepatic rCYP27A1 mRNA levels in hypocalcemic D-depleted rats repleted with 1,25(OH)2D3 after 1 to 7 days of repletion (28 pmol/day). (A) Representative Northern blot analyses of the CYP27A1 gene transcript. (B) Densitometric analyses. Data are presented as means ± SEM. Statistically significant differences between group means were analyzed by ANOVA, with individual contrasts evaluated by the Bonferroni post hoc test. Main effect P <0.001. Statistically significant different from D-depleted rats: *P<0.0001, **P<0.0001; n = 3 animals/group. Reproduced from the American Journal of Physiology (Endocrinology and Metabolism), 2003, volume 284, p. E143 [20] by copyright permission of the American Physiological Society.

59

CHAPTER 4 The Vitamin D 25-Hydroxylase

A

D

G

B

E

H

C

F

I

FIGURE 3 Influence of 1,25(OH)2D3 administration to vitamin D depleted rats on the hepatic intra-acinar distribution of the D3 25-hydroxylase gene transcript CYP27A1. Representative photomicrograph of rat liver sections obtained after in situ hybridization with an rCYP27A antisense riboprobe. A 10-cycle RT-PCR amplification was used. Liver sections were obtained from D-depleted rats (A and B), normal rats (D and E), and 7-day 1,25(OH)2D3-repleted rats (28 pmol/day) (G and H). Negative in situ hybridization control sections with the rCYP27A sense riboprobe are presented in C, F, and I for livers obtained from D-depleted, normal, and 1,25(OH)2D3-repleted rats, respectively. Reproduced from the American Journal of Physiology (Endocrinology and Metabolism), 2003, volume 284, p. E144 [20] by copyright permission of the American Physiological Society.

hepatic acinus with higher perivenous than periportal mRNA levels in both rat and human livers [118,141]. In the human liver, in situ hybridization illustrates a clear CYP27A1 label in hepatocytes that increases in intensity in the perivenous region of the hepatic acinus [141]. In addition, regulation studies in the rat have shown increased CYP27A1 gene transcription, mRNA levels, and production of 27-hydroxycholesterol as well as a recruitment of hepatocytes expressing CYP27A1 toward the mid- to periportal region of the liver acinus when the enterohepatic circulation was interrupted by a bile salt sequestrant [93,118]. These data clearly indicate that CYP27A1 is subject to feedback regulation (in intensity and intrahepatic regionalization) at the transcription level by bile acids returning via the portal blood [93,118]. A study on the effect of the vitamin D status has, however, shown that in the vitamin D–depleted rat liver, 1,25(OH)2D3 decreased CYP27A1 in both the periportal and perivenous regions (Fig. 3) [20]. Thus the impact of 1,25(OH)2D3 on the CYP27A1

gene transcript is more widely distributed across the hepatic acinus than that of bile acids.

VI. CEREBROTENDINOUS XANTHOMATOSIS Mutations that affect either the level of expression of CYP27A1 or its primary sequence lead to cerebrotendinous xanthomatosis (CTX) [142,143]. CTX is an autosomal recessive inherited inborn error of bile acid biosynthesis leading to defects in sterol metabolism and storage characterized by xanthomatosis, atherosclerosis, progressive neurological dysfunction, and dementia. In these patients, however, both abnormal [143–146] and normal [147] vitamin D and/or calcium metabolism have been reported, which gives credence to the presence of more than one enzyme active in the 25-hydroxylation of D in humans, or to a contra-up-regulation of CYP27A1 in cases of mild mutations [148]. Indeed, some CTX

60 patients have been shown to retain residual 27-hydroxylase activity [44], while the human cell line Hep3B has been shown to efficiently activate 1αOHD3 at C-25 despite the absence of detectable CYP27A1 mRNA when evaluated by Northern analysis [149,150]. Rosen et al. [151] were the first to report on a mouse with a disrupted CYP27A1 (CYP27A1−/−). Surprisingly, despite several biochemical markers indicating disruption of bile acid and cholesterol metabolism, no CTXrelated pathological abnormalities were observed in the mice, an observation also reported by others [152]. In addition, CYP27A1−/− mice were shown to have supranormal 25OHD3 (106 compared to 43 ng/ml in CYP27A1+/+) and normal 1,25(OH)2D3 serum concentrations. The latter observations suggest that in the mouse, as in many other species, D3 25-hydroxylase(s) other than CYP27A1 are active in the endogenous synthesis of 25OHD3, an observation that has been confirmed by the identification of CYP2R1 as a D 25hydroxylase in the liver of CYP27A1−/− mice [9]. These observations also suggest that, in the CYP27A1−/− mice, an up-regulation of these enzyme(s) takes place. However, marked species differences in the mechanisms controlling the synthesis of bile acid and cholesterol metabolism have been documented [77,152–154]. Indeed, in contrast to patients suffering from CTX, CYP27A1−/− mice exhibit a dramatic increase in CYP3A activity. This leads to a marked up-regulation in the sidechain oxidation of bile acid biosynthesis, an increase not observed in CTX despite accumulation of cholestane 3α,7α,12α-triols [154]. The increase in CYP3A is secondary to the elevation of toxic bile acid intermediates in the CYP27A1−/− mouse which are potent activators of the xenobiotic/pregnane X receptor (PXR) in the mouse but only weak activators of PXR in humans. The increase in PXR acts as a feed-forward regulatory loop by which toxic bile acid intermediates activate PXR and induce their own metabolism through a PXRmediated induction of CYP3A. Because of the reported differences in bile acid and cholesterol metabolism between a mouse and human with disrupted CYP27A1, the CYP27A1−/− mouse may, therefore, not be an adequate model of the CTX-related defects in D3 anabolism observed in patients stricken with the disease. These observations also illustrate that D 25-hydroxylases other than CYP27A are most likely present in human as suggested by the putative role of the microsomal CYP2R1 in the C-25 hydroxylation of the vitamin and the capacity of CYP3A4 to act as a D 25-hydroxylase [9,58]. In additional support for this hypothesis, a study reported by Byford et al. [105] has indicated that the human CYP27A1-negative cell line Hep 3B is able to transform [3H]1αOHD4 into its 25-hydroxylated

MARIELLE GASCON-BARRÉ

metabolite at low nanomolar concentration, further suggesting the presence of a D 25-hydroxylase other than CYP27A1 in human.

VII. ONTOGENY OF THE VITAMIN D 25-HYDROXYLASES The fetal rat liver can adequately sequester D3 [155]. Moreover, when compared to values obtained in the late fetal period, D3 uptake decreased significantly during the first 2 weeks post partum and then dramatically increased (sixfold) between days 14 and 22 post partum when it reached an uptake capacity similar to that previously observed in the adult rat. The liver of 18- (3 days before birth), 19-, and 22-day-old rat fetuses hydroxylates D3 at C-25 and the value was shown to remain unchanged at day 1 after birth [156,157]. In one study [156], the microsomal 25-hydroxylase activity was shown to sharply increase from day 2 to 11 and to remain unchanged until day 60 of chronological age, whereas in another study [157], maximal enzyme activity was reached only near weaning. These authors also noticed that a cytosolic factor was necessary for optimum 25-hydroxylase activity in fetuses, neonates, and adult rats. In that context, it has been suggested that cytosol contains a cellular D binding protein [158] that could facilitate access of the substrate to the enzyme as proposed by Tsankova et al. [159]. In the pig, expression of the microsomal CYP2D25 has been shown to be similar in the newborn and adult liver, but its expression was found to be lower in the newborn than in the adult kidney [32]. The data suggest a tissue-specific developmental regulation of CYP2D25 in the pig. In contrast, the CYP27A1 gene transcript can easily be detected in the human fetal liver, kidney, jejunum, and colon. Moreover, the CYP27A1 gene transcript was shown to be regulated by 1,25(OH)2D3 in the fetal intestine [120,141]. However, in 16- to 20-week-old human fetuses, the level of the CYP27A1 gene transcript was found to be approximately 40% lower in the fetal liver than in the adult liver [141], but the level of expression of the fetal human kidney CYP27A1 gene transcript was similar to that of the adult kidney [141]. In human, 25OHD has been shown to be lower in cord than in maternal serum, although a significant relationship was observed between the two sites [160]. Premature infants have low circulating 25OHD concentrations compared to infants born at term [161,162], which rapidly increases, however, in the subsequent 4 days after birth [162], indicating fully operational D 25-hydroxylase activity in preterm infants.

61

CHAPTER 4 The Vitamin D 25-Hydroxylase

VIII. SEX DIFFERENCES IN THE HYDROXYLATION OF VITAMIN D AT C-25 Despite the observation that the circulating 25OHD concentrations are not significantly different in males and females, several studies indicate that both the microsomal and the mitochondrial enzyme activities may be gender specific and/or regulated by sex hormones. Indeed, CYP2C11, which exhibits D3 25-hydroxylase activity, is known to be male specific in the rat and rabbit but not in human [88]. Not surprisingly, the enzyme activity reported in the rat has been largely shown to be higher in males than females [45,46]. In addition, Hayashi et al. [48] concluded, based on immunoprecipitation techniques, that the D3 25-hydroxylase found in the male rat liver was a different entity from that found in the female, suggesting that a microsomal P450 distinct from CYP2C11 is also able to catalyze the 25-hydroxylation of vitamin D in rodents. By contrast, the mitochondrial enzyme CYP27A1 has been shown in the human and rat liver to be more abundant in females than in males with a 43% higher expression in female than in male rat livers, although the regulation mechanisms of the enzyme by sex hormones is not presently known [47,141,163]. In addition, Saarem and Pedersen [47] observed that both enzyme activities increased when estradiol was injected into male rats while testosterone decreased the activities of both enzymes in female rats. Addya et al. [163] also reported that the relative 25-hydroxylase levels were reduced in castrated females but increased in castrated males. Although regulatory studies have not yet been carried out on the microsomal CYP2R1, investigation of the levels of its gene transcript in male and female mouse livers indicates the absence of sexual dimorphism [9].

IX. CONCLUSIONS In liver, both microsomes and mitochondria harbor monooxygenase systems active on the 25-hydroxylation of the endogenous substrate D3, but also on several analogs of the vitamin. This allows the design of molecules with therapeutic potential that can be activated by a single hydroxylation step by the liver. The mitochondrial sterol 27-hydroxylase is a multifunctional enzyme exhibiting high capacity with a reported Km for D3 several times higher (micromolar range) than the circulating steady-state concentration of the substrate in vivo. The dynamic liver regulation of enzyme distribution associated with uptake of the substrate favoring selective acinar accumulation seems

to allow, however, the necessary conditions within the hepatocyte microenvironment for efficient transformation of D into 25OHD in vivo. In addition, the presence of CYP27A1 in multiple extrahepatic tissues and organs is suggestive of an enzyme system displaying pleiotrophic functions that are able to support the C-25 hydroxylation of the vitamin in target organs. Whether or not these sites participate in the toxicity of the vitamin, or act, in a tissue-specific manner, for the in situ delivery of 25OHD to the renal and/or the extra-renal 1α-hydroxylase as observed in vitro, remains to be investigated. The microsomal D 25-hydroxylase was originally thought to be a high-specificity low-capacity enzyme. It has now been shown that the enzymes described to date also have low substrate specificity and exhibit regulation by sex hormones and preferential intraacinar localization to the perivenous area (CYP2C11) [164]. The porcine hepatic and kidney CYP2D25 has been characterized and shown to be a high-affinity enzyme and to hydroxylate D3 and D2 compounds at both C-25 and C-1α. The human and mouse CYP2R1 has also been reported to activate D3 and D2 compounds. However, the relative importance of the microsomal and mitochondrial enzyme in the C-25 hydroxylation of vitamin D is not presently known. Indeed, Hosseinpour et al. [56] in a study in pig hepatocytes that possess both CYP2D25 (microsomes) and CYP27A1 (mitochondria) activities found that both enzymes equally participate in the 25-hydroxylation of D3, whereas Cheng et al. [9] reported that CYP27A1 was more abundantly expressed than CYP2R1 (microsomes) in the mouse liver, but the comparative contribution of each enzyme to the formation of 25OHD is not yet known. In conclusion, the liver is essential to the normal homeostasis of vitamin D. Indeed, it activates the vitamin of endogenous and exogenous origin, and efficiently exports 25OHD to the systemic circulation where its concentration represents the best marker of the vitamin D nutritional status. The presence of D microsomal and mitochondrial 25-hydroxylases (which also exhibit C-1α activity) in extrahepatic tissues also raises the hypothesis of an auto/para/intracrine system of D action in D target organs.

X. ADDENDUM While this book was in press, the biological relevance of CYP2R1 as a D 25-hydroxyalse was illustrated by the elucidation by Cheng et al. [177] of a molecular defect in the CYP2R1 gene which was found to be responsible

62 for low serum 25OHD and the classical symptoms of rickets. The affected individual had a normal CYP27A1 gene but was found to be homozygous for a transition mutation in exon 2 of the CYP2R1 gene which resulted in the elimination of D 25-hydroxylase enzyme activity. These observations establish CYP2R1 as a biologically important D 25-hydroxylase and illustrate, for the first time, that a mutation in the CYP2R1 gene which abolishes enzyme activity can lead to a rare molecular autosomal recessive disorder, a selective 25OHD deficiency.

References 1. Blunt JW, DeLuca HF, Schnoes HK 1968 25-hydroxycholecalciferol. A biologically active metabolite of vitamin D3. Biochemistry 7:3317–3322. 2. Ponchon G, Kennan AL, DeLuca HF 1969 “Activation” of vitamin D by the liver. J Clin Invest 48:2032–2037. 3. Olson EB Jr., Knutson JC, Bhattacharyya MH, DeLuca HF 1976 The effect of hepatectomy on the synthesis of 25-hydroxyvitamin D3. J Clin Invest 57:1213–1220. 4. Andersson S, Davis DL, Dahlback H, Jornvall H, Russell DW 1989 Cloning, structure, and expression of the mitochondrial cytochrome P-450 sterol 26-hydroxylase, a bile acid biosynthetic enzyme. J Biol Chem 264: 8222–8229. 5. Usui E, Noshiro M, Okuda K 1990 Molecular cloning of cDNA for vitamin D3 25-hydroxylase from rat liver mitochondria. FEBS Lett 262:135–138. 6. Cali JJ, Russell DW 1991 Characterization of human sterol 27-hydroxylase. A mitochondrial cytochrome P-450 that catalyzes multiple oxidation reaction in bile acid biosynthesis. J Biol Chem 266:7774–7778. 7. Bjorkhem I, Holmberg I 1978 Assay and properties of a mitochondrial 25-hydroxylase active on vitamin D3. J Biol Chem 253:842–849. 8. Postlind H, Axén E, Bergman T, Wikvall K 1997 Cloning, structure, and expression of a cDNA encoding vitamin D3 25-hydroxylase. Biochem Biophys Res Commun 241: 491–497. 9. Cheng JB, Motola DL, Mangelsdorf DJ, Russell DW 2003 De-orphanization of cytochrome P450 2R1. A microsomal vitamin D 25-hydroxylase. J Biol Chem 278: 38084–38093. 10. Gascon-Barré M, Huet PM 1982 Role of the liver in the homeostasis of calciferol metabolism in the dog. Endocrinology 110:563–570. 11. Gascon-Barré M, Vallières S, Huet PM 1986 Influence of phenobarbital on the hepatic handling of [3H]vitamin D3 in the dog. Am J Physiol (Gastrointest Liver Physiol) 251:G627–G635. 12. Gascon-Barré M, Gamache M 1991 Contribution of the biliary pathway to the homeostasis of vitamin D3 and of 1,25-dihydroxyvitamin D3. Endocrinology 129:2335–2344. 13. Rojanasathit S, Haddad JG 1976 Hepatic accumulation of vitamin D3 and 25-hydroxyvitamin D3. Biochim Biophys Acta 421:12–21. 14. Gascon-Barré M, Vallières S, Huet PM 1986 Uptake of the hormone 1,25-dihydroxyvitamin D3 by the dog liver. Can J Physiol Pharmacol 64:699–702.

MARIELLE GASCON-BARRÉ

15. Haddad JG, Jennings AS, Choon Aw T 1988 Vitamin D uptake and metabolism by perfused rat liver: Influences of carrier proteins. Endocrinology 123:498–504. 16. Dueland S, Holmberg I, Berg T, Pedersen JI 1981 Uptake and 25-hydroxylation of vitamin D3 by isolated rat liver cells. J Biol Chem 256:10430–10434. 17. Dueland S, Helgerud P, Pedersen JI, Berg T, Drevon CA 1983 Plasma clearance, transfer, and distribution of vitamin D3 from intestinal lymph. Am J Physiol 245:E326–E331. 18. Gascon-Barré M, Elbaz H, Therrien-Ferland D 1985 Sequestration and microsomal C-25 hydroxylation of [3H]vitamin D3. Metabolism 34:244–250. 19. Ziv E, Bar-On H, Silver J 1985 Vitamin D3 uptake by the isolated perfused rat liver from lipoprotein fractions is separate from cholesterol and triglyceride uptake. Eur J Clin Invest 15:95–99. 20. Theodoropoulos C, Demers C, Petit JL, Gascon-Barré M 2003 High sensitivity of the rat hepatic vitamin D3-25 hydroxylase CYP27A to 1,25-dihydroxyvitamin D3 administration. Am J Physiol 284:E138–E147. 21. Gascon-Barré M, Vallières S, Benbrahim N 1992 C-25 hydroxylation of vitamin D3 in the periportal and perivenous region of the hepatic acinus. Am J Physiol (Endocrinol Metab) 262:E810–E817. 22. Gascon-Barré M, Huet PM, St-Onge Brault G, Brault A, Kassissia I 1988 Liver extraction of vitamin D 3 is independent of its venous or arterial route of delivery. Studies in isolated-perfused rat liver preparations. J Pharmacol Exp Ther 245:975–981. 23. Horsting M, DeLuca HF 1969 In vitro production of 25-hydroxycholecalciferol. Biochem Biophys Res Commun 36:251–256. 24. Ponchon G, DeLuca HF 1969 The role of the liver in the metabolism of vitamin D. J Clin Invest 48:1273–1279. 25. DeLuca HF 1971 Metabolism and mechanism of action of 25-hydroxycholecalciferol. In: Nicholes G Jr., Wasserman RH (eds) Cellular Mechanisms for Calcium and Homeostasis. Academic Press, New York, pp. 426–430. 26. Bhattacharyya MH, DeLuca HF 1974 Subcellular location of rat liver calciferol-25-hydroxylase. Arch Biochem Biophys 160:58–62. 27. Madhok TC, Schnoes HK, DeLuca HF 1978 Incorporation of oxygen-18 into the 25-position of cholecalciferol by hepatic cholecalciferol 25-hydroxylase. Biochem J 175: 479–482. 28. Fukushima M, Nishii Y, Suzuki M, Suda T 1978 Comparative studies on the 25-hydroxylations of cholecalciferol and 1α-hydroxycholecalciferol in perfused rat liver. Biochem J 170:495–502. 29. Tucker GI, Gagnon RE, Haussler MR 1973 Vitamin D3-25hydroxylase: Tissue occurrence and apparent lack of regulation. Arch Biochem Biophys 155:47–57. 30. Bhattacharyya MH, DeLuca HF 1974 The regulation of calciferol-25-hydroxylase in the chick. Biochem Biophys Res Commun 59:734–741. 31. Axén E 1995 Purification from pig kidney of a microsomal cytochrome P450 catalyzing 1α-hydroxylation of 25-hydroxyvitamin D3. FEBS Lett 375:277–279. 32. Hosseinpour F, Norlin M, Wikvall K 2002 Kidney microsomal 25- and 1α-hydroxylase in vitamin D metabolism: catalytic properties, molecular cloning, cellular localization and expression during development. Biochim Biophys Acta 1580:133–144. 33. Bhattacharyya MH, DeLuca HF 1973 The regulation of rat liver calciferol-25-hydroxylase. J Biol Chem 248:2969–2973.

CHAPTER 4 The Vitamin D 25-Hydroxylase

34. Bhattacharyya MH, DeLuca HF 1973 Comparative studies on the 25-hydroxylation of vitamin D3 and dihydrotachysterol. J Biol Chem 248:2974–2977. 35. Madhok TC, DeLuca HF 1979 Characteristics of the rat liver microsomal enzyme system converting cholecalciferol into 25-hydroxycholecalciferol. Biochem J 184:491–499. 36. Björkhem I, Hansson R, Holmberg I, Wikvall K 1979 25-Hydroxylation of vitamin D3 by a reconstituted system from rat liver microsomes. Biochem Biophys Res Commun 90:615–622. 37. Yoon PS, DeLuca HF 1980 Resolution and reconstitution of soluble components of rat liver microsomal vitamin D3-25hydroxylase. Arch Biochem Biophys 203:529–541. 38. Andersson S, Holmberg I, Wikvall K 1983 25-hydroxylation of C27-steroids and vitamin D3 by a constitutive cytochrome P-450 from rat liver microsomes. J Biol Chem 258:6777–6781. 39. Hayashi S, Noshiro M, Okuda K 1984 Purification of cytochrome P-450 catalyzing 25-hydroxylation of vitamin D3 from rat liver microsomes. Biochem Biophys Res Commun 121:994–1000. 40. Holmberg I 1984 Inhibition of reconstituted vitamin D3 25hydroxylase by a protein fraction from rat liver microsomes. Biochem Biophys Res Commun 123:1209–1214. 41. Nelson DR, Kamataki T, Waxman DJ, Guengerich FP, Estabrook RW, Feyereisen R, Gonzalez FJ, Coon MJ, Gunsalus IC, Gotoh O, Okuda K, Nebert DW 1993 The P450 superfamily: Update on new sequences, gene mapping, accession numbers, early trivial names of enzymes, and nomenclature. DNA Cell Biol 12:1–51. 42. Hiwatashi A, Ichikawa Y 1980 Purification and organspecific properties of cholecalciferol 25-hydroxylase system: Cytochrome P-450D25-linked mixed function oxidase system. Biochem Biophys Commun 97:1443–1449. 43. Hayashi S, Noshiro M, Okuda K 1986 Isolation of a cytochrome P-450 that catalyzes the 25-hydroxylation of vitamin D3 from rat liver microsomes. J Biochem 99:1753–1763. 44. Okuda KI 1994 Liver mitochondrial P450 involved in cholesterol catabolism and vitamin D activation. J Lipid Res 35:361–372. 45. Andersson S, Jornvall H 1986 Sex differences in cytochrome P-450-dependent 25-hydroxylation of C27-steroids and vitamin D3 in rat liver microsomes. J Biol Chem 261: 16932–16936. 46. Dählback H, Wikvall K 1987 25-hydroxylation of vitamin D3 in rat liver: Roles of mitochondrial and microsomal cytochrome P-450. Biochem Biophys Res Commun 142: 999–1005. 47. Saarem K, Pedersen JI 1987 Sex differences in the hydroxylation of cholecalciferol and of 5β-cholestane-3α,7α,12αtriol in rat liver. Biochem J 247:73–78. 48. Hayashi SI, Usui E, Okuda K 1988 Sex-related difference in vitamin D3 25-hydroxylase of rat liver microsomes. J Biochem 103:863–866. 49. Thierry-Palmer M, Free AL, Nagappan PR, Butler AR 1995 Cholecalciferol 25-hydroxylation is similar in liver microsomes from male and female rats when cholecalciferol concentration is low. J Nutr 125:104–111. 50. Segura-Aguilar J 1996 Peroxidase activity of liver microsomal vitamin D 25-hydroxylase and cytochrome P450 1A2 catalyzes 25-hydroxylation of vitamin D3 and oxidation of dopamine to aminochrome. Biochem Mol Med 58:122–129. 51. Axén E, Bergman T, Wikvall K 1994 Microsomal 25-hydroxylation of vitamin D2 and vitamin D3 in pig liver. J Steroid Biochem Mol Biol 51:97–106.

63 52. Hosseinpour F, Wikvall K 2000 Porcine microsomal vitamin D3 25-hydroxylase (CYP2D25): Catalytic properties, tissue distribution and comparison with human CYP2D6. J Biol Chem 275:34650–34655. 53. Axén E, Harmeyer J, Wikvall K 1998 Renal and hepatic 1α-hydroxylation of 25-hydroxyvitamin D3 in piglets suffering from pseudo vitamin D-deficiency rickets, type I. Biochim Biophys Acta 1407:234–242. 54. Ciobanu DC, Gill RK, Rothschild MF, Bell NH 2000 Rapid communication: Porcine vitamin D-25-hydroxylase maps to chromosome 5′. J Anim Sci 78:3193–3194. 55. Hosseinpour F, Hidestrand M, Ingelman-Sundberg M, Wikvall K 2001 The importance of residues in substrate recognition site 3 for the catalytic function of CYP2D25 (vitamin D 25-hydroxylase). Biochem Biophys Res Commun 288:1059–1063. 56. Hosseinpour F, Ibranovic I, Tang W, Wikvall K 2003 25-Hydroxylation of vitamin D3 in primary cultures of pig hepatocytes: evidence for a role of both CYP2D25 and CYP27A1. Biochem Biophys Res Commun 303:877–883. 57. Araya Z, Hosseinpour F, Bodin K, Wikvall K 2003 Metabolism of 25-hydroxyvitamin D3 by microsomal and mitochondrial vitamin D3 25-hydroxylases (CYP2D25 and CYP27A1): a novel reaction by CYP27A1. Biochim Biophys Acta 1632:40–47. 58. Nelson DR 2003 Comparison of P450s from human and fugu: 420 million years of vertebrate P450 evolution. Arch Biochem Biophys 409:18–24. 59. Bjorkhem I 1985 Mechanism of bile acid biosynthesis in mammalian liver. In: Danielsson H, Sjovall J. (eds) Sterols and Bile Acids. Elsevier Science, New York, pp. 231–278. 60. Popjak G, Edmond J, Anet FAL, Easton NR, Jr. 1977 Carbon-13 NMR studies on cholesterol biosynthesized from [13C]mevalonates. J Am Chem Soc 99:931–935. 61. Danielsson H 1960 On the oxidation of 3α,7α,12α-trihydroxycoprostane by mouse and rat liver homogenates. Acta Chem Scand 14:348–352. 62. Shefer S, Cheng FW, Batta AK, Dayal B, Tint GS, Salen G, Mosbach EH 1978 Stereospecific side chain hydroxylations in the biosynthesis of chenodeoxycholic acid. J Biol Chem 253:6386–6392. 63. Atsuta Y, Okuda K 1981 On the stereospecificity of cholestanetriol 26-monooxygenase. J Biol Chem 256: 9144–9146. 64. Björkhem I 1992 Mechanism of degradation of the steroid side chain in the formation of bile acids. J Lipid Res 33:455–471. 65. Björkhem I, Gustafsson J 1973 Omega-hydroxylation of steroid side-chain in biosynthesis of bile acids. Eur J Biochem 36:201–212. 66. Cronholm T, Johansson G 1970 Oxidation of 5β-cholestane3α,7α,12α-triol by rat liver microsomes. Eur J Biochem 373–381. 67. Taniguchi S, Hoshita N, Okuda K 1973 Enzymatic characteristics of CO-sensitive 26-hydroxylase system for 5β-cholestane-3α,7α,12α-triol in rat-liver mitochondria and its intramitochondrial localization. Eur J Biochem 40:607–617. 68. Bjorkhem I, Gustafsson J 1974 Mitochondrial omegahydroxylation of the cholesterol side chain. J Biol Chem 249:2528–2535. 69. Okuda KI, Weber P, Ullrich V 1977 Photochemical action spectrum of the CO-inhibited 5β-cholestane-3α,7α,12α-triol 26-hydroxylase system. Biochem Biophys Res Commun 74:1071–1076.

64 70. Sato R, Atsuta Y, Imai Y, Taniguchi S, Okuda KI 1977 Hepatic mitochondrial cytochrome P-450: isolation and functional characterization. Proc Natl Acad Sci USA 74:5477–5481. 71. Pedersen JI, Oftebro H, Vanngard T 1977 Isolation from bovine liver mitochondria of a soluble ferredoxin active in a reconstituted steroid hydroxylation reaction. Biochem Biophys Res Commun 76:666–673. 72. Okamura T, John ME, Zuber MX, Simpson ER, Waterman MR 1985 Molecular cloning and amino acid sequence of the precursor form of bovine adrenodoxin: Evidence for a previously unidentified COOH-terminal peptide. Proc Natl Acad Sci USA 82:5705–5709. 73. Hanukoglu I, Gutfinger T 1989 cDNA sequence of adrenodoxin reductase. Identification of NADP-binding sites in oxidoreductases. Eur J Biochem 180:479–484. 74. Ugele B, Locher M, Burger HJ, Gebhardt R 1986 Is there a heterogeneity of liver parenchyma in taurocholate uptake? In: Paumgartner G, Stiehl A, Gerok W (eds) Bile Acids and the Liver. MTP Press Ltd., Boston, pp. 153–160. 75. Pedersen JI, Godager HK 1978 Purification of NADPHferredoxin reductase from rat liver mitochondria. Biochim Biophys Acta 525:28–36. 76. Okuda K, Masumoto O, Ohyama Y 1988 Purification and characterization of 5β-cholestane-3α,7α,12α-triol 27-hydroxylase from female rat liver mitochondria. J Biol Chem 263:18138–18142. 77. Lyons MA, Maeda N, Brown AJ 2002 Paradoxical enhancement of hepatic metabolism of 7-ketocholesterol in sterol 27-hydroxylase-deficient mice. Biochim Biophys Acta 1581:119–126. 78. Wikvall K 1984 Hydroxylations in biosynthesis of bile acids. Isolation of a cytochrome P-450 from rabbit liver mitochondria catalyzing 26-hydroxylation of C27-steroids. J Biol Chem 259:3800–3804. 79. Björkhem I, Gustafsson J, Johansson G, Persson B 1975 Biosynthesis of bile acids in man. Hydroxylation of the C27-steroid side chain. J Clin Invest 55:478–486. 80. Masumoto O, Ohyama Y, Okuda K 1988 Purification and characterization of vitamin D 25-hydroxylase from rat liver mitochondria. J Biol Chem 263:14256–14260. 81. Ohyama Y, Masumoto O, Usui E, Okuda K 1991 Multi-functional property of rat liver mitochondrial cytochrome P-450. J Biochem 109:389–393. 82. Dixon M, Webb EC 1979 The investigation of enzyme specificity. In: Dixon M, Webb EC (eds), Enzymes. Longman, London, pp. 232–234. 83. Oftebro H, Saarem K, Bjorkhem I, Pedersen JI 1981 Side chain hydroxylation of C27-steroids and vitamin D3 by a cytochrome P-450 enzyme system isolated from human liver mitochondria. J Lipid Res 22:1254–1264. 84. Saarem K, Bergseth S, Oftebro H, Pedersen JI 1984 Subcellular localization of vitamin D3 25-hydroxylase in human liver. J Biol Chem 259:10936–10960. 85. Saarem K, Pedersen JI 1985 25-hydroxylation of 1α-hydroxyvitamin D-3 in rat and human liver. Biochim Biophys Acta 840:117–126. 86. Su P, Rennert H, Shayiq RM, Yamamoto R, Zheng Y-M, Addya S, Strauss JFI, Avadhani NG 1990 A cDNA encoding a rat mitochondrial cytochrome P450 catalyzing both the 26-hydroxylation of vitamin D3: Gonadotropic regulation of the cognate mRNA in ovaries. DNA Cell Biol 9:657–665. 87. Guo YD, Strugnell S, Back DW, Jones G 1993 Transfected human liver cytochrome P-450 hydroxylates vitamin D analogs at different side-chain positions. Proc Natl Acad Sci USA 90:8668–8672.

MARIELLE GASCON-BARRÉ

88. Nebert DW, McKinnon RA 1994 Cytochrome P450: Evolution and functional diversity. In: Boyer JL, Ockner RK (eds) Progress in Liver Diseases. WB Saunders, Philadelphia, pp. 63–97. 89. Sakaki T, Akiyoshi-Shibata M, Yabusaki Y, Ohkawa H 1992 Organella-targeted expression of rat liver cytochrome P450c27 in yeast. Genetically engineered alteration of mitochondrial P450 into a microsomal form creates a novel functional electron transport chain. J Biol Chem 267: 16497–16502. 90. Shayiq RM, Avadhani NG 1992 Sequence complementarity between the 5′-terminal regions of mRNAs for rat mitochondrial cytochrome P-450c27/25 and a growth hormone– inducible serine protease inhibitor. A possible gene overlap. J Biol Chem 267:2421–2428. 91. Mullick J, Addya S, Sucharov C, Avadhani NG 1995 Localization of a transcription promoter within the second exon of the cytochrome P-450c27/25 gene for the expression of the major species of two-kilobase mRNA. Biochemistry 34:13729–13742. 92. Wiese RJ, Uhland-Smith A, Ross TK, Prahl JM, DeLuca HF 1992 Up-regulation of the vitamin D receptor in response to 1,25-dihydroxyvitamin D3 results from ligand-induced stabilization. J Biol Chem 267:20082–20086. 93. Vlahcevic ZR, Jairath SK, Heuman DM, Stravitz RT, Hylemon PB, Avadhani NG, Pandak WM 1996 Transcriptional regulation of hepatic sterol 27-hydroxylase by bile acids. Am J Physiol 270:G646–G652. 94. Twisk J, Hoekman FM, Lehmann EM, Meijer P, Mager WH, Princen HMG 1995 Insulin suppresses bile acid synthesis in cultured rat hepatocytes by down-regulation of cholesterol 7α-hydroxylase and sterol 27-hydroxylase gene transcription. Hepatology 21:501–510. 95. Usui E, Noshiro M, Ohyama Y, Okuda K 1990 Unique property of liver mitochondrial P450 to catalyze the two physiologically important reactions involved in both cholesterol catabolism and vitamin D activation. FEBS Lett 274:175–177. 96. Akiyoshi-Shibata M, Usui E, Sakaki T, Yabusaki Y, Noshiro M, Okuda K, Ohkawa H 1991 Expression of rat liver vitamin D3 25-hydroxylase cDNA in Saccharomyces cerevisiae. FEBS Lett 280:367–370. 97. Holmberg-Betsholtz I, Lund E, Bjorkhem I, Wikvall K 1993 Sterol 27-hydroxylase in bile acid biosynthesis. Mechanism of oxidation of 5β-cholestane-3α,7α,12α,27-tetrol into 3α,7α,12α-trihydroxy-5β-cholestanoic acid. J Biol Chem 268:11079–11085. 98. Kusano K, Kagawa N, Sakaguchi M, Omura T, Waterman MR 2001 Importance of a prolife-rich sequence in the amino-terminal region for correct folding of mitochondrial and soluble microbial p450s. J Biochem 129:271–277. 99. Axén E, Postlind H, Sjoberg H, Wikvall K 1994 Liver mitochondrial cytochrome P450 CYP27 and recombinant-expressed human CYP27 catalyze 1α-hydroxylation of 25-hydroxyvitamin D3. Proc Natl Acad Sci USA 91: 10014–10018. 100. Arlazoroff A, Roitberg B, Werber E, Shidlo R, Berginer VM 1991 Epileptic seizure as a presenting symptom of cerebrotendinous xanthomatosis. Epilepsia 32:657–661. 101. Sawada N, Sakaki T, Ohta M, Inouye K 2000 Metabolism of vitamin D3 by human CYP27A1. Biochem Biophys Res Commun 273:977–984. 102. Horst RL, Koszewski NJ, Reinhardt TA 1990 1α-Hydroxylation of 24-hydroxyvitamin D2 represents a minor physiological pathway for the activation of vitamin D2 in mammals. Biochemistry 29:578–582. 103. Mawer EB, Jones G, Davies M, Still PE, Byford V, Schroeder NJ, Makin HLJ, Bishop CW, Knutson JC 1998

65

CHAPTER 4 The Vitamin D 25-Hydroxylase

104.

105.

106. 107.

108. 109. 110. 111. 112.

113.

114. 115.

116. 117.

118.

119.

Unique 24-hydroxylated metabolites represent a significant pathway of metabolism of vitamin D2 in humans: 24-hydroxyvitamin D2 and 1,24-dihydroxyvitamin D2 detectable in human serum. J Clin Endocrinol Metab 83:2156–2166. Dilworth FJ, Scott I, Green A, Strugnell S, Guo YD, Roberts EA, Kremer R, Calverley MJ, Makin HLJ, Jones G 1995 Different mechanisms of hydroxylation site selection by liver and kidney cytochrome P450 species (CYP27 and CYP24 involved in vitamin D metabolism. J Biol Chem 270: 16766–16774. Byford V, Strugnell S, Coldwell R, Schroeder N, Makin HLJ, Knutson JC, Bishop CW, Jones G 2002 Use of vitamin D4 analogs to investigate differences in hepatic and target cell metabolism of vitamins D2 and D3. Biochim Biophys Acta 1583:151–166. Thierry-Palmer M, Gray TK, Napoli JL 1988 Ring hydroxylation of 25-hydroxycholecalciferol by rat renal microsomes. J Steroid Biochem 29:623–628. Rao DS, Dayal R, Siu-Caldera ML, Horst RL, Uskokovic MR, Tserng KY, Reddy GS 1999 Isolation and identification of 4,25-dihydroxyvitamin D2: A novel A-ring hydroxylated metabolite of vitamin D2. J Steroid Biochem Mol Biol 71:63–70. Jones G, Schnoes HK, Levan L, DeLuca HF 1980 Isolation and identification of 24-hydroxyvitamin D2 and 24,25-dihydroxyvitamin D2. Arch Biochem Biophys 202:450–457. Wichmann JK, Schnoes HK, DeLuca HF 1981 Isolation and identification of 24(R)-hydroxyvitamin D3 from chicks given large doses of vitamin D3. Biochemistry 20:2350–2353. Bjorkhem I, Holmberg I, Oftebro H, Pedersen JI 1980 Properties of a reconstituted vitamin D3 25-hydroxylase from rat liver mitochondria. J Biol Chem 255:5244–5249. Dählback H, Wikvall K 1988 25-Hydroxylation of vitamin D3 by a cytochrome P-450 from rabbit liver mitochondria. Biochem J 252:207–213. Axén E, Postlind H, Wikvall K 1995 Effects on CYP27 mRNA expression in rat kidney and liver by 1α,25dihydroxyvitamin D3, 1α-hydroxylase activity. Biochem Biophys Res Commun 215:136–141. Hollis BW 1990 25-hydroxyvitamin D3–1α-hydroxylase in porcine hepatic tissue: Subcellular localization to both mitochondria and microsomes. Proc Natl Acad Sci USA 87:6009–6013. Takeuchi A, Okano T, Kobayashi T 1991 The existence of 25-hydroxyvitamin D3-1α-hydroxylase in the liver of carp and bastard halibut. Life Sci 48:275–282. Takeuchi A, Okano T, Sekimoto H, Kobayashi T 1994 The enzymatic formation of 1α,25-dihydroxyvitamin D3 from 25-hydroxyvitamin D3 in the liver of fetal rats. Comp Biochem Physiol [C] 109C:1–7. Negrea LA, Slotapolsky E, Dusso AS 1995 1,25-dihydroxyvitamin D synthesis in rat liver microsomes. Horm Metabol Res 27:461–464. Stravitz RT, Vlahcevic ZR, Russell TL, Heizer ML, Avadhani NG, Hylemon PB 1996 Regulation of sterol 27-hydroxylase and an alternative pathway of bile acid biosynthesis in primary cultures of rat hepatocytes. J Steroid Biochem Mol Biol 57:337–347. Twisk J, Hoekman MFM, Mager WH, Moorman AFM, De Boer PAJ, Scheja L, Princen HMG, Gebhardt R 1995 Heterogenous expression of cholesterol 7α-hydroxylase and sterol 27-hydroxylase genes in the rat liver lobulus. J Clin Invest 95:1235–1243. Theodoropoulos C, Demers C, Mirshahi A, Gascon-Barré M 2001 1,25-dihydroxyvitamin D3 downregulates the rat

120.

121.

122.

123. 124.

125. 126. 127.

128.

129.

130. 131. 132. 133.

134.

135.

intestinal vitamin D3-25-hydroxylase CYP27A. Am J Physiol 281:E315–E325. Theodoropoulos C, Demers C, Delvin E, Ménard D, GasconBarré M 2003 Calcitriol regulates the expression of the gene encoding all three vitamin D3 hydroxylases and the drug metabolizing enzyme CYP3A4 in the human fetal intestine. Clin Endocrinol 58:489–499. Memon RA, Moser AH, Shigenaga JK, Grunfeld C, Feingold KR 2001 In vivo and in vitro regulation of sterol 27-hydroxylase in the liver during the acute phase response. Potential role of hepatocyte nuclear factor-1. J Biol Chem 276:30118–30126. Segev H, Honigman A, Rosen H, Leitersdorf E 2001 Transcriptional regulation of the human sterol 27-hydroxylase gene (CYP27) and promoter mapping. Atherosclerosis 156:339–347. Cozzolino M, Vidal M, Arcidiacono MV, Tebas P, Yarasheski KE, Dusso AS 2003 HIV-protease inhibitors impair vitamin D bioactivation to 1,25-dihydroxyvitamin D. AIDS 17:513–520. Post SM, Duez H, Gervois PP, Staels B, Kuipers F, Princen HMG 2001 Fibrates suppress bile acid synthesis via peroxisome proliferator—Activated receptor-α-mediated downregulation of cholesterol 7α-hydroxylase and sterol 27-hydroxylase expression. Arterioscler Thromb Vasc Biol 2001:1840–1845. Araya Z, Tang W, Wikvall K 2003 Hormonal regulation of the human sterol 27-hydroxylase gene CYP27A1. Biochem J 372:529–534. Chen W, Chiang JYL 2003 Regulation of human sterol 27-hydroxylase gene (CYP27A1) by bile acids and hepatocyte nuclear factor 4α (HNF4α). Gene 313:71–82. Araya Z, Sjoberg H, Wikvall K 1995 Different effects on the expression of CYP7 and CYP27 in rabbit liver by cholic acid and cholestyramine. Biochem Biophys Res Commun 216: 868–873. Nakajima S, Hsieh JC, MacDonald PN, Galligan MA, Haussler CA, Whitfield GK, Haussler MR 1994 The C-terminal region of the vitamin D receptor is essential to form a complex with a receptor auxiliary factor required for high affinity binding to the vitamin D-responsive element. Mol Endocrinol 8:159–172. Rao YP, Vlahcevic ZR, Stravitz RT, Mallonee DH, Mullick J, Avadhani NG, Hylemon PB 1999 Down-regulation of the rat hepatic sterol 27-hydroxylase gene by bile acids in transfected primary hepatocytes: possible role of hepatic nuclear factor 1α. J Steroid Biochem Mol Biol 70:1–14. Cereghini S 1996 Liver-enriched transcription factors and hepatocyte differentiation. FASEB J 10:267–282. Bolt MJG, Sitrin MD, Favus MJ, Rosenberg IH 1981 Hepatic vitamin D 25-hydroxylase: inhibition by bile duct ligation or bile salts. Hepatology 1:436–440. Plourde V, Gascon-Barré M, Willems B, Huet PM 1988 Severe cholestasis leads to vitamin D depletion without perturbing its C-25 hydroxylation in the dog. Hepatology 8:1577–1585. Princen HMG, Meijer P, Wolthers BG, Vonk RJ, Kuipers F 1991 Cyclosporine A blocks bile acid synthesis in cultured hepatocytes by specific inhibition of chenodeoxycholic acid synthesis. Biochem J 275:501–505. Dahlback-Sjoberg H, Bjorkhem I, Princen HMG 1993 Selective inhibition of mitochondrial 27-hydroxylation of bile acid intermediates and 25-hydroxylation of vitamin D3 by cyclosporin A. Biochem J 2933:203–206. Baran DT, Milne ML 1983 1,25-Dihydroxyvitamin D-induced inhibition of 3H-25 hydroxyvitamin D production by the rachitic rat liver in vitro. Calcif Tissue Int 35:461–464.

66 136. Bell NH, Shaw S, Turner RT 1984 Evidence that 1,25-dihydroxyvitamin D3 inhibits the hepatic production of 25-hydroxyvitamin D in man. J Clin Invest 74: 1540–1544. 137. Clements MR, Davies M, Hayes ME, Hickey CD, Lumb GA, Mawer EB, Adams PH 1992 The role of 1,25-dihydroxyvitamin D in the mechanism of acquired vitamin D deficiency. Clin Endocrinol 37:17–27. 138. Haddad P, Gascon-Barré M, Brault G, Plourde V 1986 Influence of calcium or 1,25-dihydroxyvitamin D3 supplementation on the hepatic microsomal and in vivo metabolism of vitamin D3 in vitamin D-depleted rats. J Clin Invest 78:1529–1537. 139. Halloran BP, Bikle DD, Levens MJ, Castro ME, Globus RK, Holton E 1986 Chronic 1,25-dihydroxyvitamin D3 administration in the rat reduces the serum concentration of 25-hydroxyvitamin D by increasing metabolic clearance rate. J Clin Invest 78:622–628. 140. Gascon-Barré M, Demers C, Mirshahi A, Néron S, Zalzal S, Nanci A 2003 The normal liver harbors the vitamin D nuclear receptor in nonparenchymal and biliary epithelial cells. Hepatology 37:1034–1042. 141. Gascon-Barré M, Demers C, Ghrab O, Theodoropoulos C, Lapointe R, Jones G, Valiquette L, Ménard D 2001 Expression of CYP27A, a gene encoding a vitamin D-25 hydroxylase in human liver and kidney. Clin Endocrinol 54:107–115. 142. Cali JJ, Hsieh CL, Francke U, Russell DW 1991 Mutations in the bile acid biosynthetic enzyme sterol 27-hydroxylase underlie cerebrotendinous xanthomatosis. J Biol Chem 266:7779–7783. 143. Leitersdorf E, Reshef A, Meiner V, Levitzki R, Schwartz SP, Dann EJ, Berkman N, Cali JJ, Klapholz L, Berginer VM 1993 Frameshift and splice-junction mutations in the sterol 27-hydroxylase gene cause cerebrotendinous xanthomatosis in Jews of Moroccan origin. J Clin Invest 91:2488–2496. 144. Berginer VM, Shany S, Alkalay D, Berginer J, Dekel S, Salen G, Tint GS, Gazit D 1993 Osteoporosis and increased bone fractures in cerebrotendinous xanthomatosis. Metab Clin Exp 42:69–74. 145. Leitersdorf E, Safadi R, Meiner V, Reshef A, Bjorkhem I, Friedlander Y, Morkos S, Berginer VM 1994 Cerebrotendinous xanthomatosis in the Israeli Druze: molecular genetics and phenotypic characteristics. Am J Hum Genet 55:907–915. 146. Berginer VM, Shany S, Alkalay D, Berginer J, Dekel S, Salen G, Tint GS, Gazit D 1993 Osteoporosis and increased bone fractures in cerebrotendinous xanthomatosis. Metabolism 42:69–74. 147. Kuriyama M, Fujiyama J, Kubota R, Nakagawa M, Osame M 1993 Osteoporosis and increased bone fractures in cerebrotendinous xanthomatosis [Letter to editor]. Metab Clin Exp 42:1497. 148. Sawada N, Sakaki T, Kitanaka S, Kato S, Inouye K 2001 Structure-function analysis of CYP27B1 and CYP27A1. Studies on mutants from patients with vitamin D–dependent rickets type I (VDDR-I) and cerebrotendinous xanthomatosis (CTX). Eur J Biochem 268:6607–6615. 149. Guo YD, Strugnell S, Jones G 1991 Identification of a human liver mitochondrial cytochrome P-450 cDNA corresponding to the vitamin D3-25-hydroxylase. J Bone Miner Res 6:S120. 150. Strugnell S, Byford V, Makin HLJ, Moriarty RM, Gilardi R, LeVan LW, Knutson JC, Bishop CW, Jones G 1995 1α,24(S)dihydroxyvitamin D2: A biologically active product of 1α-hydroxyvitamin D2 made in the human hepatoma Hep3B. Biochem J 310:233–241.

MARIELLE GASCON-BARRÉ

151. Rosen H, Reshef A, Maeda N, Lippoldt A, Shpizen S, Triger L, Eggertsen G, Björkhem I, Leitersdorf E 1998 Markedly reduced bile acid synthesis but maintained levels of cholesterol and vitamin D metabolites in mice with disrupted sterol 27-hydroxylase gene. J Biol Chem 273:14805–14812. 152. Honda A, Salen G, Matsuzaki Y, Batta AK, Xu G, Leitersdorf E, Tint GS, Erickson SK, Tanaka N, Shefer S 2001 Side chain hydroxylations in bile acid biosynthesis catalyzed by CYP3A are markedly up-regulated in CYP27−/− mice but not in cerebrotendinous xanthomatosis. J Biol Chem 276: 34579–34585. 153. Björkhem I, Araya Z, Rudling M, Angelin B, Einarsson C, Wikvall K 2002 Differences in the regulation of the classical and the alternative pathway for bile acid synthesis in human liver. No coordinate regulation of CYP7A1 and CYP27A1. J Biol Chem 277:26804–26807. 154. Goodwin B, Gauthier KC, Umetani M, Watson MA, Lochansky MI, Collins JL, Leitersdorf E, Mangelsdorf DJ, Kliewer SA, Repa JJ 2003 Identification of bile acid precursors as endogenous ligands for the nuclear xenobiotic pregnane X receptor. Proc Natl Acad Sci USA 100: 223–228. 155. Martial J, Plourde V, Gascon-Barré M 1985 Sequestration of vitamin D3 by the fetal and neonatal rat liver. Biol Neonate 48:21–28. 156. Plourde V, Haddad P, Gascon-Barré M 1985 Microsomal C-25 hydroxylation of [3H]-vitamin D3 by the fetal and neonatal rat liver. Pediatr Res 19:1206–1209. 157. Thierry-Palmer M, Cullins S, Rashada S, Gray TK, Free A 1986 Development of vitamin D3 25-hydroxylase activity in rat liver microsomes. Arch Biochem Biophys 250:120–127. 158. Miller ML, Ghazarian JG 1980 Studies on vitamin D3 metabolism. Discrete liver cytosolic binding proteins for vitamin D3 and 25-hydroxyvitamin D3. Biochem Biophys Res Commun 96:1619–1625. 159. Tsankova V, Visentin M, Cantoni L, Carelli M, Tacconi MT 1996 Peripheral benzodiazepine receptor ligands in rat liver mitochondria: Effect on 27-hydroxylation of cholesterol. Eur J Pharmacol 299:197–203. 160. Delvin EE, Glorieux FH, Salle BL, David L, Varenne JP 1982 Control of vitamin D metabolism in preterm infants: feto-maternal relationships. Arch Dis Child 57:754–757. 161. Hillman LS, Hoff N, Salmons S, Martin L, McAlister WH, Haddad J 1985 Mineral homeostasis in very premature infants: Serial evaluation of serum 25-hydroxyvitamin D, serum minerals, and bone mineralization. J Pediatr 106:970–980. 162. Salle BL, Glorieux FH, Delvin EE, David LS, Meunier G 1983 Vitamin D metabolism in preterm infants. Serial serum calcitriol values during the first four days of life. Acta Paediatr Scand 72:203–206. 163. Addya A, Zheng YM, Shayiq RM, Fan J, Avadhani NG 1991 Characterization of a female-specific hepatic mitochondrial cytochrome P-450 whose steady-state level is modulated by testosterone. Biochemistry 30:8323–8330. 164. Bühler R, Lindros KO, Nordling A, Johansson I, IngelmanSundberg M 1992 Zonation of cytochrome P450 isozyme expression and induction in rat liver. Eur J Biochem 204: 407–412. 165. Hansson M, Ellis E, Hunt MC, Schmitz G, Babiker A 2003 Marked induction of sterol 27-hydroxylase activity and mRNA levels during differentiation of human cultured monocytes into macrophages. Biochim Biophys Acta 1593:283–289. 166. Jabara S, Christenson LK, Wang CY, McAllister JM, Javitt NB, Dunaif A, Strauss JF 2003 Stromal cells of the

CHAPTER 4 The Vitamin D 25-Hydroxylase

167. 168.

169.

170.

171.

human postmenopausal ovary display a distinctive biochemical and molecular phenotype. J Clin Endocrinol Metab 88:484–492. Javitt NB 2002 25R,26-hydroxycholesterol revisited: Synthesis, metabolism, and biologic roles. J Lipid Res 43: 665–670. Lehmann B, Tiebel O, Meurer M 1999 Expression of vitamin D3 25-hydroxylase (CYP27) mRNA after induction by vitamin D3 or UVB radiation in keratinocytes of human skin equivalents. A preliminary study. Arch Dermatol Res 291:507–510. Schuessler M, Astecker N, Herzig G, Vorisek G, Schuster I 2001 Skin is an autonomous organ in synthesis, two-step activation and degradation of vitamin D3: CYP27 in epidermis completes the set of essential vitamin D3-hydroxylases. Steroids 66:399–408. Correa P, Segersten U, Hellman P, Åkerström WG 2002 Increased 25-hydroxyvitamin D3 1α-hydroxylase and reduced 25-hydroxyvitamin D3 24-hydroxylase expression in parathyroid tumors—New prospects for treatment of hyperparathyroidism with vitamin D. J Clin Endocrinol Metab 87:5826–5829. Ichikawa F, Sato K, Nanjo M, Nishii Y, Shinki T, Takahashi N, Suda T 1995 Mouse primary osteoblasts express vitamin D3 25-hydroxylase mRNA and convert 1α-hydroxyvitamin D3 into 1α,25-dihydroxyvitamin D3. Bone 16:129–135.

67 172. Bjorkhem I, Andersson O, Diczfalusy U, Sevastik B, Xiu RJ, Duan C, Lund E 1994 Atherosclerosis and sterol 27-hydroxylase: Evidence for a role of this enzyme in elimination of cholesterol from human macrophages. Proc Natl Acad Sci USA 91:8592–8596. 173. Reiss AB, Martin KO, Javitt NB, Martin DW, Grossi EA, Galloway AC 1994 Sterol 27-hydroxylase: high levels of activity in vascular endothelium. J Lipid Res 35:1026–1030. 174. Delvin EE, Arabian A, Glorieux FH 1978 Kinetics of liver microsomal cholecalciferol 25-hydroxylase in vitamin Ddepleted and -repleted rats. Biochem J 172:417–422. 175. Benbrahim N, Dubé C, Vallières S, Gascon-Barré M 1988 The calcium ionophore A23187 is a potent stimulator of the vitamin D3-25 hydroxylase in hepatocytes isolated from normocalcemic vitamin D-depleted rats. Biochem J 255: 91–97. 176. Holmberg I, Berlin T, Ewerth S, Bjorkhem I 1986 25 hydroxylase activity in subcellular fractions from human liver. Evidence for different rates of mitochondrial hydroxylation of vitamin D2 and D3. Scand J Clin Lab Invest 46: 785–790. 177. Cheng JB, Levine MA, Bell NH, Mangelsdorf DJ, Russel DW 2004 Genetic evidence that the human CYP2R1 enzyme is a key vitamin D 25-hydroxylase. Proc Natl Acad Sciences (USA) 101:771–775.

CHAPTER 5

The 25-Hydroxyvitamin D 1α-Hydroxylase HELEN L. HENRY

Department of Biochemistry, University of California at Riverside, Riverside, California

I. Occurrence and Characteristics of 25OHD3-1α-Hydroxylase II. Characteristics of the Proteins Involved in the 1α-Hydroxylation of 25OHD3 III. Cytochrome P4501α: Cloning and Gene Structure

IV. Regulation of kidney 1α-Hydroxylase Activity and Gene Expression V. Summary References

I. OCCURRENCE AND CHARACTERISTICS OF 25OHD3 1α-HYDROXYLASE

exclusively in the proximal tubules. In situ hybridization studies of cultures of embryonic mouse kidneys confirm the presence of 25OHD3 1α-hydroxylase (CYP1α) in tubular epithelium, but not collecting ducts or glomeruli [8].

A. The Kidney as the Source of Circulating 1α,25(OH)2D3 It is now well accepted that vitamin D is a precursor of the sterol hormone 1α,25-dihydroxyvitamin D3 [1α,25(OH)2D3]. The general pathway of production of 1α,25(OH)2D3 is shown in Fig. 1. It has been appreciated for some time [1,2] that the kidney is the major site of production of circulating 1α,25(OH)2D3, although as described later and discussed more thoroughly elsewhere in this volume, many other tissues and cell types have been shown to produce 1α,25(OH)2D3 from 25-hydroxyvitamin D3 (25OHD3). Within the kidney, it was established early on by microdissection studies that in the fetal rabbit [3] and in the vitamin D–deficient rat [4] and chick [5], the proximal tubules are the region of the most robust activity of the 1α-hydroxylase. With the cloning of the cDNA for the cytochrome P450 component of the 1α-hydroxylase (see Section III) has come the ability to measure its mRNA and protein levels along the nephron. Since these determinations are more sensitive than the measurement of enzyme activity, localization studies can now be carried out under conditions of vitamin D sufficiency and normal mineral status. Thus in vitamin D–sufficient mice and humans, mRNA and/or protein has been identified by in situ hybridization or immunohistochemical staining in the more distal portions of the nephron along with relatively low expression in the proximal tubules [6,7]. These observations suggest that, while the 1α-hydroxylase occurs throughout the nephron, its regulation varies such that the effects of vitamin D status and abnormal phosphorus metabolism (see Section IV) occur primarily if not VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

B. Extrarenal Production of 1α,25(OH)2D3 Along with the demonstration that the kidney contained the enzymatic capability to produce 1α,25(OH)2D3 other work was suggesting that this metabolic step was largely confined to this organ. Physiological observations such as studies in nephrectomized animals [9] and clinical experience with patients suffering from chronic renal failure [10,11] strongly supported this concept. Thus, the kidney was the focus of investigation of the 1α-hydroxylase for the decade following its identification as the site of 1,25(OH)2D3 synthesis. In terms of the regulation of serum levels of 1α,25(OH)2D3 , the kidney remains the focus of study and most of the remainder of this chapter is devoted to the renal production of 1α,25(OH)2D3. Some extrarenal sites of 1α,25(OH)2D3 will be mentioned briefly here and discussed more thoroughly in Chapter 79. It should be noted that in general the extrarenal 1α-hydroxylation of 25OHD3 does not respond to the regulatory influences involved in calcium homeostasis that control the renal enzyme activity. In the early 1980s the first reports appeared that suggested that cells outside the kidney could produce 1α,25(OH)2D3 from 25(OH)D3. These reports described studies in calvarial cells [12,13] and were followed by those of 1α-hydroxylase and 24R-hydroxylase activity in chondrocytes [14,15]. These descriptions of 1α-hydroxylase activity in bone were followed by the demonstration that keratinocytes could also produce 1α,25(OH)2D3 [16,17]. Copyright © 2005, Elsevier, Inc. All rights reserved.

70

HELEN L. HENRY

FIGURE 1 Basic metabolic pathway for the major circulating vitamin D metabolites. The two renal hydroxylases that metabolize 25OHD3 are, in general, regulated in a reciprocal fashion to one another.

During this same time period, the hypercalcemia of sarcoidosis was traced to the production of 1α,25(OH)2D3 by pulmonary macrophages [18], and there followed many reports of the production of 1α,25(OH)2D3 by cells derived from the hematopoietic system [19–24]. These sites of the autocrine and/or paracrine production and action of 1α,25(OH)2D3 are discussed in more detail in Chapter 79. Reports that placental decidual cells [25,26] also contain 1α-hydroxylase activity were complicated by the fact the production of 1,25(OH)2D3 by trophoblastic tissues can produce 1α,25(OH)2D3 by a nonenzymatic process [27]. With the availability of the nucleotide and amino acid sequences for CYP1α, the presence of both the message and the protein in placental tissue [28,29] has now been confirmed. Other tissues in which evidence for CYP1α transcription and/or expression exists include the adrenal medulla, brain, pancreas, colon [29], parathyroid glands [30], prostate cells [31], and non-small-cell lung carcinoma [32].

C. Species Distribution The 1α-hydroxylase, not surprisingly, occurs widely among vertebrate species. None of the species examined in the original species comparison [33] have since been shown to lack the renal 1α-hydroxylase.

Since there are no reports of 1α-hydroxylase activity in invertebrates it can be tentatively concluded that the ability to produce 1α,25(OH)2D3 arose early in vertebrate evolution. There has been a special interest in the occurrence of vitamin D and its metabolites in fish, due in part to the historical importance of cod liver oil as a source of vitamin D and in part to the question of how fish produce vitamin D under conditions of limited ultraviolet light. Plasma levels of vitamin D metabolites have been measured in several species of fish [34–36]. Fish liver and kidney homogenates have been reported to convert 25OHD3 to 1α,25(OH)2D3 [33,36,37]. Both phytoplankton and zooplankton were reported to have abundant quantities of provitamin D and vitamin D and were suggested to be the dietary source of vitamin D for fish [38], although these authors have concluded that vitamin D and its metabolites are unimportant to mineral metabolism in the freshwater fish tilapia [39]. Thus, although there is no question that fish contain vitamin D and its metabolites and 1α-hydroxylase activity, the physiological role of these in fish remains unclear. Another interesting comparative aspect of vitamin D metabolism is posed by the naked mole rat. This nocturnal, cave-dwelling animal with no access to ultraviolet light nevertheless has the ability to synthesize both 1α,25(OH)2D3 and receptors for the hormone [40,41]. Thus, although these animals are usually vitamin D

71

CHAPTER 5 The 25-Hydroxyvitamin D 1α-Hydroxylase

ultimately derived from NADPH, generated intramitochondrially from tricarboxylic acid cycle intermediates. In vitro assays of 1α-hydroxylase activity in isolated mitochondria have used either malate [42] or other citric acid cycle intermediates [43] as a source of reducing equivalents. The current model for mitochondrial steroid hydroxylases involves the sequential transfer of electrons from NADPH to NADPH-ferredoxin reductase, a 50-kDa protein that, in the case of adrenal and gonadal steroid hydroxylases, is loosely associated with the inner mitochondrial membrane [44]. The electrons are then passed to ferredoxin, an 11-kDa mitochondrial matrix protein that shuttles between the NADPH-ferredoxin reductase and the terminal component of the hydroxylase machinery, cytochrome P450 (so named because of its distinct spectral characteristics when carbon monoxide is bound to it). As discussed below, each of these three proteins has been the subject of study in the context of the 1α-hydroxylation of 25OHD3.

deficient, they maintain the enzymatic machinery necessary to produce the active hormonal form and the receptor necessary to respond to it.

II. CHARACTERISTICS OF THE PROTEINS INVOLVED IN THE 1α-HYDROXYLATION OF 25OHD3 A. Mixed-Function Oxidases Enzymes that hydroxylate endogenous steroids in a stereospecific manner are called mixed-function oxidases because they reduce one atom of molecular oxygen to water and one atom to the hydroxyl group to be stereospecifically incorporated into the steroid (Fig. 2). Such reactions are important in the pathways of the production of androgens, estrogens, progestins, mineralocorticoids, and glucocorticoids. Indeed, the first step in the production of these steroid hormones, the cleavage of six carbons from the side chain of cholesterol, results from two successive hydroxylations at C-20 and C-22. Some mixed-function oxidases are microsomal (e.g., the 17α-hydroxylase/C-17-C-20 lyase and the 21-hydroxylase), and some (e.g., cholesterol side-chain cleavage and the 11β-hydroxylase) are located in the inner mitochondrial membrane. The 25OHD3 1α-hydroxylase belongs to the latter group of steroid hydroxylases. For mitochondrial mixed-function oxidases, the electrons for the reduction of molecular oxygen are

1. CYTOCHROME P450

The evidence that the 1α-hydroxylase is a member of the class of cytochrome P450-dependent steroid hydroxylases was obtained in the mid-1970s and consists of the following: (1) Solubilized preparations of chick kidney mitochondria were shown to contain cytochrome P450 and to carry out 1α-hydroxylation of 25OHD3 [45]; (2) more directly, it was shown that inhibition of 1α-hydroxylase activity in isolated chick kidney mitochondria by carbon monoxide was reversed specifically by light of 450 nm [42].

R–OH + H2O NADPH

FPox

NHI–Fe

2+

P450–Fe

3+

R Reductase

NADP

FPred

Ferredoxin

NHI–Fe 3+

Cytochrome P450

P450–Fe 2+

P450–Fe 2+−O2

O2

CO P450–Fe 2+−CO

FIGURE 2 Electron transport chain for mitochondrial steroid hydroxylases. The general class of mitochondrial mixed-function oxidases, of which the 25OHD3 1α-hydroxylase is a member, consists of three components in or associated with the inner mitochondrial membrane. Cytochrome P450 reduces molecular oxygen to water and to the hydroxyl group to be incorporated into the steroid; it confers the stereospecificity of the hydroxylation reaction. In its reduced form, cytochrome P450 binds carbon monoxide and in this form absorbs light of 450 nm. The flavoprotein (FP) is NADPH-ferredoxin reductase, and the nonheme iron (NHI) protein is ferredoxin. Ox, oxidized; red, reduced.

72

HELEN L. HENRY

There have been sporadic reports of the partial purification of CYP1α. An apparently electrophoretically pure bovine kidney cytochrome P450, with a molecular weight of 57,000, retained putative 1α-hydroxylase activity [46], but there was no confirmation of the identity of the product of 25OHD3 hydroxylation. In the early 1990s, there were two reports purporting to have purified the 1α-hydroxylase cytochrome P450 from mitochondria from either vitamin D–replete or vitamin D–deficient chick kidneys [47,48]. The purified proteins had molecular weights of 59,000 and 57,000, but amino acid compositions and N-terminal sequences, when compared with the four mammalian CYP1α sequences that are now known, make it unlikely that either of these proteins was indeed the 1α-hydroxylase cytochrome P450. More recently, a 54-kDa protein with 1α-hydroxylase activity and appropriate spectral characteristics was isolated from chick kidney [49] but no sequence information about this protein was given. In another study with pig kidney mitochondria, 1α-hydroxylase activity was partially purified [50] and separated from steroid side chain–hydroxylating enzymes [51], but in neither report were any physical characteristics of the protein given. In summary, although the primary sequence of CYP1α has been deduced from cloned cDNA from several mammalian species, purified preparations of the protein itself are still not routinely obtained. The primary sequence of the cytochrome P450 that catalyzes the 1α-hydroxylation of 25OHD3, as deduced from its cDNA sequence, reveals that it is structurally related to the mitochondrial sterol side chain hydroxylases and therefore it has been given the systematic name of CYP27B1 (see [52] for a listing of human cytochrome P450s). In this chapter, the protein and its nucleic acid sequences will be referred to by its familiar and functionally more descriptive name of CYP1α. 2.

FERREDOXIN

a. Characteristics of Mitochondrial Ferredoxin Vertebrate mitochondrial ferredoxins are nonheme iron– sulfur proteins of 114–128 amino acids, which on electrophoresis generally migrate at 11–12 kDa. cDNA analysis indicates a mitochondrial leader sequence of 58–62 amino acids, depending on the species [53]. The ferredoxin involved in the mitochondrial hydroxylation of endogenous steroids is a matrix protein that shuttles between NADPH-ferredoxin reductase and cytochrome P450 to deliver, one at a time, the two electrons required for the reduction of molecular oxygen. Ferredoxins from mammalian adrenal glands (adrenodoxins) have been studied most extensively [54,55], although the protein has been isolated from several other steroidogenic tissues [56].

As is the case for ferredoxins in other steroidogenic tissues, chick kidney ferredoxin is a mitochondrial matrix protein [57]. This submitochondrial localization is consistent with its function as an electron-carrying shuttle between the NADPH-ferredoxin reductase and CYP1α. Reconstitution assays in a number of laboratories have demonstrated the absolute requirement of 1α-hydroxylase activity for ferredoxin [58–59]. Chick kidney ferredoxin has been overexpressed in Escherichia coli and shown to be approximately twice as active as bovine adrenal ferredoxin in supporting 1α-hydroxylation in a reconstituted assay system [60], further illustrating the similarities that the 1α-hydroxylase shares with other mitochondrial mixed-function oxidases that hydroxylate endogenous steroids. cDNA cloning has shown that the pig kidney and adrenal ferredoxin share identical amino acid sequences [61]; similarly, in the chick kidney and testis ferredoxins share the same primary structure [62,63]. These observations suggest that a single gene encodes ferredoxin for the various tissues in which it is expressed. Southern blot analysis of bovine [64] and chick genomic DNA [62] supports the existence of a single ferredoxin gene in these species. In the human, two ferredoxin genes have been identified, but their coding regions for the mature protein are identical [65]. Taken together the available evidence indicates that a single ferredoxin protein serves as the electron shuttle for all mitochondrial cytochrome P450s that hydroxylate endogenous steroids in the species that have been examined thus far. b. Regulation of Mitochondrial Ferredoxin Levels Since the rate of hydroxylation of a steroid by cytochrome P450 will be governed to some extent by the supply of electrons to reduce molecular oxygen, a regulatory component at this step is possible, although such a process would clearly lack some specificity in cells that can carry out more than one hydroxylation. Mammalian ferredoxins and their mRNAs have been shown to be developmentally and hormonally regulated [66–69]. For example, hypophysectomy decreases and replacement treatment with adrenocorticotropin (ACTH) increases adrenal ferredoxin mRNA levels in the rat [67,70]. In a study of several cell types, ferredoxin mRNA was stimulated by appropriate pituitary tropic hormones, possibly through a cAMP-dependent mechanism, and ferredoxin mRNA levels decreased in the testis and adrenal during fetal development [66]. Chick kidney mitochondrial ferredoxin mRNA is up-regulated approximately 40% in the vitamin D– deficient state [62]. Although kidney mitochondrial ferredoxin is also presumably required for the 24Rhydroxylation of vitamin D, which is, in general, regulated reciprocally to the 1α-hydroxylase, 1α-hydroxylase activity in vitamin D deficiency is much higher than is

73

CHAPTER 5 The 25-Hydroxyvitamin D 1α-Hydroxylase

24R-hydroxylase activity in the vitamin D–replete chick. Therefore, increased mRNA levels of ferredoxin mRNA in vitamin D deficiency are not likely to be involved in the reciprocal regulation of these two enzymes, but rather are related to the increased demand of maximal 1α-hydroxylase activity brought about by vitamin D deficiency. Ferredoxin phosphorylation has been suggested as a posttranslational mechanism for the regulation of its steroid hydroxylation activity in the adrenal gland and kidney. Although adrenal ferredoxin can be phosphorylated in vivo by cAMP-dependent protein kinase, no convincing evidence of phosphorylated adrenal ferredoxin in vivo has appeared, nor has a role been established for such a phosphoferredoxin in the physiological regulation of adrenal or gonadal steroidogenesis. On the other hand, ferredoxin from primary cultures of chick kidney cells has been shown to exist as a phosphoprotein and to be rapidly dephosphorylated by treatment of the cells with the phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA) [57]. This dephosphorylation is accompanied by decreased 1α-hydroxylase activity, supporting for a role of ferredoxin phosphorylation and dephosphorylation in the regulation of 1-hydroxylation of 25OHD3. However, the molecular mechanisms involved in such a process remain to be determined. 3. NADPH-FERREDOXIN REDUCTASE

The flavoprotein NADPH-ferredoxin reductase, which accepts electrons from intramitochondrially generated NADPH, has been purified from pig kidney mitochondria [71] and does not appear to differ significantly in size (~52 kDa), enzymatic characteristics, or immunologically from the reductases from other mammalian tissues that carry out hydroxylations in the production of endogenous steroids such as the corpus luteum and the adrenal gland [72,73]. cDNAs for bovine and human adrenal NADPH-ferredoxin reductase have been cloned and shown to code for a mitochondrial targeting sequence in addition to the mature protein [74].

B. Accessory Proteins In addition to the three components of the enzymatic machinery that hydroxylates 25OHD3 at the 1α position, certain other cellular proteins have been implicated in the process of moving the substrate from the circulation to the site of metabolism in the inner mitochondrial membrane. Although the steroid hormones and their precursors are lipophilic and might be thought to simply diffuse through the cellular membranes

to the site of their metabolism, specific proteins to direct their intracellular movements are quite likely. One group of candidates for this type of protein, currently termed the intracellular vitamin D binding proteins, or IDBPs, was originally found in vitamin D(and other steroid hormone)-resistant New World primates [75]. The constitutive expression of binding proteins for 25OHD3 in these animals was initially thought to be the causative factor in their vitamin D resistance and/or the response to it, leading to high circulating levels of the hormone. More recently, these proteins have been cloned and sequenced and found to be homologous to human heat shock proteins of the hsp-70 family [76,77], and they are now thought to play an active role in the delivery of the substrate 25OHD3 to the 1α-hydroxylase, rather than acting to sequester the substrate as originally thought. In addition, these proteins have been shown to bind DNA and may play a function in the regulation of 24R-hydroxylase gene expression (see Chapter 21). A somewhat more definitive role in the uptake of 25OHD3 from the circulation and delivery to the 1α-hydroxylase has been proposed for the endocytotic receptor megalin, a member of the low-density lipoprotein receptor family [78]. These authors initially observed abnormal excretion of 25OHD3 in megalin knockout mice and showed that complexes of the vitamin D binding protein, DBP, and 25OHD3 are filtered by the glomerulus and reabsorbed by megalin in the proximal tubule of the kidney. In megalin knockout mice, the mRNA levels for both the 1α-hydroxylase and the 24R-hydroxylase were altered in a way consistent with severe vitamin D deficiency (that is, elevated 1α-hydroxylase and decreased 24R-hydroxylase activity; see Section IV), confirming the inability of the kidney to produce 1α,25(OH)2D3 to regulate these two enzymes [79,80]. See Chapter 10 for a more thorough discussion of the role of megalin in vitamin D metabolism.

III. CYTOCHROME P4501α: CLONING AND GENE STRUCTURE Reports of the cloning of the cDNA encoding the cytochrome P450 catalyzing the 1α-hydroxylation of 25(OH)D3 appeared from a number of laboratories in 1997. The mouse cDNA was cloned from kidneys of mice lacking the VDR, which resulted in very high expression levels of 1α-hydroxylase mRNA [81]. At approximately the same time, the rat kidney [82,83], human keratinocyte [84], and human kidney [85] sequences for the 1α-hydroxylase cytochrome P450 were published. These were followed by the porcine kidney

74 coding sequence [86]. For the mouse, rat, and human sequences, the identification of the cloned cDNA as that encoding the 1α-hydroxylase cytochrome P450 was confirmed by transfection into cells capable of supporting steroid hydroxylation (cell lines derived from kidney or testis having endogenously expressed ferredoxin and ferredoxin reductase) and demonstration that the product formed from added 25(OH)D3 was 1α,25(OH)2D3. The porcine sequence was identified on the basis of its 88% homology with the other three mammalian sequences. The sequences of cDNAs encoding the mammalian CYP1αs predict proteins with 501 to 508 amino acids. The structures of the human [85,87,88] and mouse [89] CYP27B1 genes have been determined. The mammalian CYP27B1 gene is divided into nine exons, seven

HELEN L. HENRY

of roughly equal size (170–200 bp encoding 57–67 amino acids each) and two smaller ones (26 and 37 amino acids). The putative sterol binding region (Fig. 3) is encoded by exon VI and the heme-binding region, centering around the cysteine residue that is invariant in all mitochondrial P450s that hydroxylate endogenous steroids, is encoded by exon VIII. The intron size ranges from approximately 80 to 660 bp, so altogether, the CYP27B1 gene is small in size relative to some other members of the mitochondrial cytochrome P450s, although the overall intron/exon organization is highly conserved among the genes coding for these proteins. The gene for CYP1α has been mapped to human chromosome 12 [83,87] and linkage analysis shows that it lies

FIGURE 3 Deduced amino acid sequences of CYP1α from several mammalian species. The human sequence is given in its entirety and in the other sequences, identical amino acids are indicated by dashes. Box A is the putative sterol binding site and Box B is the heme-binding region, containing the cysteine that is invariant in all cytochrome P450s.

75

CHAPTER 5 The 25-Hydroxyvitamin D 1α-Hydroxylase

very close to the defect for human vitamin D–dependent rickets (VDDR type I) (see Chapter 71). In the mouse, the CYP1α gene is on chromosome 10, the region equivalent to the localization of the human gene [89].

IV. REGULATION OF KIDNEY 1α-HYDROXYLASE ACTIVITY AND GENE EXPRESSION In numerous whole-animal and cell-culture model systems it is now well established that the two most important regulators of the activity of the 1α-hydroxylase are 1α,25(OH)2D3 itself and parathyroid hormone. In addition, there has been evidence for the involvement of calcitonin, dietary mineral levels, and hormones of other endocrine systems in the regulation of renal 25OHD3 metabolism.

A. 1α,25(OH)2D3 It was recognized at the time of the localization of the production of 1,25(OH)2D3 in the kidney that, in both avian and mammalian species, vitamin D–deficient animals have higher 1α-hydroxylase activity than do vitamin D–replete animals [1,90–92]. The ability of 1,25(OH)2D3 to replicate this effect on 1α-hydroxylase has been demonstrated in chick [93,94] and mammalian [95] cells in vitro. Both in vivo and in cell culture, the activity of the 24-hydroxylase is regulated by 1α,25(OH)2D3 in a reciprocal fashion to the 1α-hydroxylase. When 1,25(OH)2D3 is removed from the cell culture medium [93] or when its administration in vivo is terminated [90], 1α-hydroxylase activity returns to the high levels characteristic of the vitamin D–deficient state, indicating the ready reversibility of the inhibitory effect of 1,25(OH)2D3 on kidney 1α-hydroxylase activity. In a wide variety of

experimental conditions the effect of 1α,25(OH)2D3 on 1α-hydroxylase activity observed in intact animals in vivo can be duplicated by experiments in primary cultures of kidney cells. From these types of experiments, however, the mechanism of the down-regulation of 1α-hydroxylase activity by 1α,25(OH)2D3 has not been elucidated. The availability of the cDNA and genomic sequences for CYP1α has opened up new avenues of exploration of this question. There is no question that steadystate CYP1α mRNA levels are down-regulated by 1α,25(OH)2D3 in the kidneys of intact animals [96,97] and that this down-regulation requires the presence of the VDR [81,96]. When steady-state CYP1α mRNA levels are measured in cultured cells, however, the results are mixed and may depend on the cell type used, as shown in Table I. In fact, of the cell types tested, only in the MCT cells did 1α,25(OH)2D3 downregulate the endogenous expression of the mRNA for the 1α-hydroxylase cytochrome P450. Of the nonresponsive cells, both the LLC-PK1 and the HKC-8 cells have been reported to express 1α-hydroxylase activity [98,99]. However, in none of the studies in Table I were data regarding the effect of 1α,25(OH)2D3 on 1α-hydroxylase enzymatic activity presented under the same conditions in which the cells were apparently unresponsive. Thus, it is not known whether the failure to see down-regulation of CYP1α mRNA by 1α,25(OH)2D3 was due to dissociation of its effects on the transcription and activity of the cytochrome P450 or whether these cells, for unknown reasons, failed to respond to 1α,25(OH)2D3 at both the RNA and protein levels. In addition to studies of the effects of 1α,25-dihydroxyvitamin D3 on the steady-state levels of CYP1α mRNA, several laboratories have prepared and studied reporter constructs using the 5′-flanking region of the gene. In one report such a construct was active in cells derived from proximal renal tubules but not in other

TABLE I Summary of Effects of 1α,25(OH)2 D3 on Steady-State CYP1α mRNA levels Whole Animals Species Rat Mouse Mouse-VDR-KO

Effect of 1α,25(OH)2D3 ↓ ↓ →

Cell Culture Reference

Cell Line

Effect of 1α,25(OH)2D3 treatment

Reference

[96,97] [96] [81,96]

MCT cells LLC-PK1 AOK-B50 HKC-8

↓ → → →

[96,151] [86] [152] [152]

Cell types referred to are MCT:SV-40-transformed mouse proximal tubule cells [151]; LLC-PK1; AOK-B50, a derivative of LLC-PK1 stably transfected with opossum PTH/PTHrP receptor [153]; and HKC-8, SV-40-transformed human kidney cell line [154].

76

HELEN L. HENRY

TABLE II

Summary of Effects of 1α,25(OH)2D3 on CYP1α Promoter Activity

Promoter (species-kb)

Cell type

Effect

Reference

m-1.7 m-1.6 h-1.1 h-0.5

AOK-B50 OK AOK-B50 MCT

→ ↓ → ↓

[155] [156] [88] [151]

Cell types referred to are the same as in Table 1, with the addition of the opossum kidney cell line, OK.

cell types [100]. A summary of the results of this type of study is shown in Table II. No definitive vitamin D regulatory element has been identified from sequence analysis of the promoter region of mammalian CYP1α genes, although functional analysis suggests that such a regulatory site may lie within 0.5 kb of the transcriptional start site [96]. As is the case with studies of the regulation by 1α,25(OH)2D3 of endogenous CYP1α mRNA levels, it is clear from Table II that studies of the regulation of promoter activity in various cell types have yielded mixed results and have not as yet revealed whether 1α,25(OH)2D3 down-regulates 1α-hydroxylase activity through a decrease in transcription of its mRNA, through an indirect transcriptional mechanism, or through posttranscriptional and/or posttranslational processes. It is possible that more than one mechanism is involved and that the model system employed will play a role in which is revealed.

B. Parathyroid Hormone In whole animal studies, parathyroidectomy diminishes and administration of PTH increases 1α-hydroxylase activity in kidney tissue, in both avian and mammalian species [101–105]. PTH has also been shown to stimulate 1α-hydroxylase activity in cultured avian [106–108] and mammalian [109,110] kidney cells. The stimulatory effect of PTH on 1α-hydroxylase activity is mediated at least in part by the cAMP signaling pathway, as its effects can be mimicked by forskolin [109,111]. Although specific proteins with altered phosphorylation states in response to PTH have been identified [112], it has not yet been determined whether they play a physiological role in the regulation of the 1α-hydroxylation of 25OHD3. In addition to the cAMP signaling pathway, protein kinase C is also involved in the regulation of the renal

metabolism of 25OHD3 by both the 1α-hydroxylase and the 24-hydroxylase. In chick kidney cells, the phorbol ester protein kinase C (PKC) activator TPA decreases 1α-hydroxylase activity transiently [57] and over a 4-hr incubation period [113]. The latter decrease in 1α-hydroxylase activity is associated with increased 24-hydroxylase activity. The abilities of TPA and 1,25(OH)2D3 to decrease 1α-hydroxylase activity in cultured chick kidney cells are additive, indicating that they operate through distinct mechanisms. In perfused rat kidney proximal tubules, on the other hand, PKC activation by TPA resulted in a transient increase in 1α-hydroxylase activity and a sustained increase when the calcium ionophore A23187 was added with TPA [114]. PTH has been reported to stimulate the PKC pathway in rat renal cells [115]. The relative physiological importance of the cAMP and PKC signaling pathways in the control of 1α-hydroxylase activity has not yet been fully resolved. Furthermore, the fact that TPA activation of PKC decreases 1α-hydroxylase activity in chick kidney cells, both in the short term and over longer times, while increasing its activity in rat preparations, suggests that the involvement of PKC in the regulation of 1α-hydroxylase may be more complex than initially appreciated. The availability of reagents with which to study the regulation of steady-state CYP1α mRNA levels and promoter activity has led to the demonstration that PTH exerts at least part of its effects on 1α-hydroxylase activity at the transcriptional level. In addition, the role of cyclic AMP in this process has been confirmed in these studies, which are summarized in Tables III and IV. In every cell type studied, PTH stimulated the endogenous levels of CYP1α mRNA and, when tested, this effect was enhanced by cyclic AMP. Only in the MCT cells, however, did 1α,25(OH)2D3 attenuate the effect of PTH, similar to the results seen when the ability of 1α,25(OH)2D3 alone to down-regulate CYP1α mRNA was tested (Table I). PTH also stimulated the TABLE III Summary of the Regulation of Steady-State CYP1α mRNA Levels by PTH, cAMP, and Forskolin Effect of Cell line LLC-PK1 AOK-B50 HKC-8 MCT

PTH

cAMP/FSK

1α,25(OH)2D3a

Reference

↑ ↑ ↑ ↑

↑ ↑ nd nd

→ → → ↓

[86] [152] [152] [96]

a The effect of 1α,25(OH) D refers to whether the steroid attenuates 2 3 (↓) or has no effect on (→) stimulation of CYP1α mRNA levels by PTH or cAMP. nd, Not determined.

77

CHAPTER 5 The 25-Hydroxyvitamin D 1α-Hydroxylase

activity of a variety of CYP1α promoter constructs and this effect was attenuated, although only modestly in two cases, by 1α,25(OH)2D3. The observation that the effect of 1α,25(OH)2D3 is not particularly robust may account for the lack of effect on endogenous levels of 1α,25(OH)2D3 mRNA. Taken together, the results very strongly indicate that PTH stimulation of 1α-hydroxylase activity in renal cells is mediated, at least in part, by a transcriptional effect on CYP1α gene expression. Furthermore, they support the concept that a balance between PTH and 1α,25(OH)2D3 determines this level of expression. That this transcription effect may not tell the whole story, however, is suggested by a report that in adult rats, PTH administration increases CYP1α mRNA levels in the kidney severalfold, but has no effect on 1α-hydroxylase activity. Clearly the regulation of the renal production of 1α,25(OH)2D3 is complex, occurs at more than one step, and is an area that deserves more detailed investigation.

C. Other Regulators of Renal 1α,25(OH)2D3 Synthesis

levels of this message and directly influences the transcriptional activity of the CYP1α gene promoter. Shinki et al. [97] showed increased steady-state CYP1α mRNA levels when normocalcemic rats were injected with calcitonin. Interestingly, this increase in CYP1α mRNA was preceded by a marked decrease in VDR mRNA, leaving open the possibility that part of the effect of calcitonin is to render the kidney 1α-hydroxylase less sensitive to feedback inhibition by 1α,25(OH)2D3. In a second report of the stimulation of steady-state levels of CYP1α mRNA by calcitonin in rat kidney [96] it was also observed that simultaneous administration of 1α,25(OH)2D3 abrogated this effect in intact rats but not in the VDR-KO mouse model that was also used. A direct stimulatory effect of calcitonin on CYP1α mRNA levels in the MCT cells, a cell line derived from mouse proximal tubules [96], and in LLC-PK1 cells [122] has been reported. Finally, the human CYP1α promoter fused to CAT reporter is up-regulated by calcitonin in MCT cells, further underscoring a role for this calcium-regulating hormone in the renal metabolism of 25OHD3. 2. DIETARY PHOSPHORUS

1. CALCITONIN

The role of calcitonin in the regulation of renal vitamin D metabolism has a long and somewhat controversial history. Although there is a substantial number of reports suggesting that calcitonin plays a role in the regulation of vitamin D metabolism of the intact animal [97,116–118], evidence that these effects are exerted directly on the renal cell that carries out the 1α-hydroxylation of 25OHD3 has been not been consistently obtained. Thus, when added to isolated renal cell preparations in vitro calcitonin has been reported to stimulate [119], inhibit [120], or have no effect [121] on 1α-hydroxylase activity. More recent studies measuring CYP1α mRNA levels, however, strongly suggest that calcitonin does stimulate the steady-state

In mammalian animal models, low dietary phosphorus leads to increased serum concentrations of 1α,25(OH)2D3 and increased 1α-hydroxylase activity measured in renal preparations [123–126]. In humans as well, dietary phosphorus is inversely correlated with serum levels of 1α,25(OH)2D3 [127–130]. It has been demonstrated that dietary phosphorus restriction in mice leads to an increase of renal proximal tubule production of 1α,25(OH)2D3 as well as increases in both the mRNA and protein levels of CYP1α [7], providing strong evidence that the effect of phosphorus is exerted, at least in part, on the level of transcription of CYP1α. In addition, these authors also provided evidence that reduced catabolism (due to reduced 24Rhydroxylase activity) of 1α,25(OH)2D3 may also play a role in the increased serum levels of 1α,25(OH)2D3 seen when dietary phosphorus is low. 3. OTHER STEROID HORMONES

TABLE IV Promoter (species-kb) m-1.7 m-1.6 h-1.1 h-0.3 h-0.5

Summary of Effects of PTH on CYP1α Promoter Activity Cell type

AOK-B50 OK AOK-B50 AOK-B50 MCT

PTH/FSK/ cAMP 1α,25(OH)2D3 Reference ↑ ↑ ↑ ↑ ↑

↓ (marginal) ↓ (modest) ↓ nd nd

[155] [156] [88] [157] [151]

Steroid hormones of other endocrine systems have been tested for their effects, or lack thereof, on 1,25(OH)2D3 production in vivo or in cell culture. These include estrogen and the synthetic glucocorticoid dexamethasone. The stimulatory effects of estrogen on 1α-hydroxylase activity measured in extracts of kidney tissue from chicks and Japanese quail [131–133] could not be confirmed as direct ones in cell culture [134]. Similarly, although the negative influence of glucocorticoids on bone health and the vitamin D endocrine system is well documented in a number

78

HELEN L. HENRY

of animal models and clinical studies (see Chapter 72), it is apparent that these effects cannot be attributed in a substantial way to direct alteration of vitamin D metabolism. The effects of glucocorticoids on plasma levels of 1α,25(OH)2D3 are variable [135–137], and dexamethasone exhibited a slight inhibitory effect on 1α-hydroxylase activity in cultured chick kidney cells [138]. A more recent study in mice showed that under conditions of low dietary calcium only, treatment of animals slightly reduced 1α-hydroxylase mRNA and activity in the kidney [139]. Thus, although interactions between other endocrine systems and the regulation of mineral metabolism by 1α,25(OH)2D3 undoubtedly play a role in calcium homeostasis and bone mineral metabolism, they do not apparently take place in a significant way at the level of 1α,25(OH)2D3 production in the kidney. 4. ANTIMYCOTICS

The antimycotics ketoconazole and miconazole inhibit 1α-hydroxylase in cell culture [140] and were subsequently shown to substantially decrease serum levels of 1,25(OH)2D3 and active calcium transport in rats [141]. Clinical studies have suggested that there are lowered serum 1,25(OH)2D3 levels in individuals being treated with these antifungal agents [142]. In one report, ketoconazole was used to ameliorate hypercalcemia resulting from acute active tuberculosis [143]. These effects of the antimycotics are similar to those known for other steroid hormones since they occur at the level of steroid hydroxylation [144,145]. Currently the primary interest in these agents in the vitamin D field is focused on their inhibition of the 24R-hydroxylase and the catabolic pathway for 1α,25(OH)2D3 that it initiates [146–148]. 5. THE KLOTHO GENE PRODUCT

A mouse homozygous mutation in a single gene, named klotho, displays multiple disorders that are similar to those seen in human aging [149]. In addition, these mice and a similar strain that is null for the klotho gene [150] have greatly elevated plasma 1α,25(OH)2D3. The levels of mRNA encoding CYP1α are also elevated in the kidneys of these mice. The regulatory responses to 1α,25(OH)2D3, PTH, and calcitonin appear to be intact in these mice, suggesting that there may be an alternate regulatory circuit involved in governing 1α,25(OH)2D3 hydroxylase activity.

V. SUMMARY The 25(OH)D3-1α-hydroxylase is a member of the family of cytochrome P450–dependent enzymes,

specifically those mitochondrial enzymes that metabolize endogenous steroids such as cholesterol side-chain cleavage systems in the adrenal cortex and gonads and the adrenal 11β-hydroxylase. The CYP1α activity is widely distributed in vertebrates. Although the kidney is still considered to be the primary site of circulating levels of 1α,25(OH)2D3, it is clearly produced in many other cells as well, mostly as part of paracrine/ autocrine systems. The past several years has seen considerable advances in the field of vitamin D metabolism with the cloning of CYP1α so that the tools are now available to study its regulation at the molecular level. The elucidation of these mechanisms and how they differ in the kidney and extrarenal sites of 1α,25(OH)2D3 production is just one of the exciting areas in which we can expect new ideas and information in the future.

References 1. Fraser DR, Kodicek E 1970 Unique biosynthesis by kidney of a biologically active vitamin D metabolite. Nature 288:764–766. 2. Norman AW 1971 Evidence for a new kidney produced hormone, 1,25-dihydroxycholecalciferol, the proposed biologically active form of vitamin D. Am J Clin Nutr 24:1346–1351. 3. Akiba T, Endou H, Koseki C, Sakai F 1980 Localization of 25-hydroxyvitamin D3-1α-hydroxylase activity in the mammalian kidney. Biochem Biophys Res Commun 94:313–318. 4. Kawashima H, Torikai S, Kurokawa K 1981 Localization of 25-hydroxyvitamin D3 1α-hydroxylase and 24-hydroxylase along the rat nephron. Proc Natl Acad Sci USA 78:1199–1203. 5. Brunette MG, Chan M, Ferriere C, Roberts KD 1978 Site of 1,25(OH)2 vitamin D3 synthesis in the kidney. Nature 276:287–289. 6. Zehnder D, Bland R, Walker EA, Bradwell AR, Howie AJ, Hewison M, Stewart PM 1999 Expression of 25-hydroxyvitamin D3-1α-hydroxylase in the human kidney. J Am Soc Nephrol 10:2465–2473. 7. Zhang MY, Wang X, Wang JT, Compagnone NA, Mellon SH, Olson JL, Tenenhouse HS, Miller WL, Portale AA 2002 Dietary phosphorus transcriptionally regulates 25-hydroxyvitamin D-1alpha-hydroxylase gene expression in the proximal renal tubule. Endocrinology 143:587–595. 8. Yamagata M, Kimoto A, Michigami T, Nakayama M, Ozono K 2001 Hydroxylases involved in vitamin D metabolism are differentially expressed in murine embryonic kidney: application of whole mount in situ hybridization. Endocrinology 142:3223–3230. 9. Hartenbower DL, Stella FJ, Norman AW, Friedler RM, Coburn JW 1977 Impaired vitamin D metabolism in acute uremia. J Lab Clin Med 90:760–766. 10. Mawer EB, Taylor CM, Backhouse J, Lumb GA, Stanbury SW 1973 Failure of formation of 1,25-dihydroxycholecalciferol in chronic renal insufficiency. Lancet 1:626–628. 11. Eisman JA, Hamstra AJ, Kream BE, DeLuca HF 1976 1,25Dihydroxyvitamin D in biological fluids: A simplified and sensitive assay. Science 193:1021–1023. 12. Turner RT, Puzas JE, Forte MD, Lester GE, Gray TK, Howard GA, Baylink, DJ 1980 In vitro synthesis of 1α,25dihydroxycholecalciferol and 24,25-dihydroxycholecalciferol

CHAPTER 5 The 25-Hydroxyvitamin D 1α-Hydroxylase

13.

14.

15.

16. 17. 18. 19.

20.

21.

22. 23.

24.

25. 26.

27.

28.

by isolated calvarial cells. Proc Natl Acad Sci USA 77: 5720–5724. Puzas JE, Turner RT, Howard GA, Brand JS, Baylink DJ 1987 Synthesis of 1,25-dihydroxycholecalciferol and 24,25-dihydroxycholecalciferol by calvarial cells: Characterization of the enzyme systems. Biochem J 245:333–338. Pedrozo HA, Boyan BD, Mazock J, Dean DD, Gomez R, Schwartz Z 1999 TGFβ1 regulates 25-hydroxyvitamin D3, 1α- and 24-hydroxylase activity in cultured growth plate chondrocytes in a maturation-dependent manner. Calcif Tissue Int 64:50–56. Schwartz Z, Brooks B, Swain L, Del Toro F, Norman AW, Boyan BD 1992 Production of 1,25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3 by growth zone and resting zone chondrocytes is dependent on cell maturation and is regulated by hormones and growth factors. Endocrinology 130:2495–2504. Bikle DD, Nemanic MK, Gee E, Elias P 1986 1,25Dihydroxyvitamin D3 production by human keratinocytes. J Clin Invest 78:557–566. Bikle DD, Nemanic MK, Whitney JO, Elias PW 1986 Neonatal human foreskin keratinocytes produce 1,25dihydroxyvitamin D3. Biochem 25:1545–1548. Adams JS, Sharma OP, Gacad MA, Singer FR 1983 Metabolism of 25-hydroxyvitamin D3 by cultured pulmonary alveolar macrophages in sarcoidosis. J Clin Invest 72:1856–1860. Reichel H, Koeffler HP, Tobler A, Norman AW 1987 1α,25Dihydroxyvitamin D3 inhibits γ-interferon synthesis by normal human peripheral blood lymphocytes. Proc Natl Acad Sci USA 84:3385–3389. Reichel H, Bishop JE, Koeffler HP, Norman AW 1987 25-Dihydroxyvitamin D3 metabolism by lipopolysaccharidestimulated normal human macrophages. J Clin Endocr Metab 64:1–9. Adams JS, Singer FR, Gacad MA, Sharma OP, Hayes MJ, Vouros P, Holick MF 1985 Isolation and structural identification of 1,25-dihydroxyvitamin D3 produced by cultured alveolar macrophages in sarcoidosis. J Clin Endocr Metab 60:960–966. Mawer EB, Hayes ME 1992 Renal and extra-renal synthesis of 1,25-dihydroxy vitamin D. Prog Endocrinol 87:382–386. Cadranel J, Garabedian M, Milleron B, Guillozo H, Akoun G, Hance AJ 1990 1,25(OH)2D3 production by T lymphocytes and alveolar macrophages recovered by lavage from normocalcemic patients with tuberculosis. J Clin Invest 85: 1588–1593. Pryke AM, Duggan C, White CP, Posen S, Mason RS 1990 Tumor necrosis factor-α induces vitamin D-1-hydroxylase activity in normal human alveolar macrophages. J Cell Physiol 142:652–656. Delvin EE, Arabian A 1987 Kinetics and regulation of 25-hydroxycholecalciferol 1-α−hydroxylase from cells isolated from human term decidua. Eur J Biochem 164:659–662. Zerwekh JE, Breslau NA 1986 Human placental production of 1,25-dihydroxyvitamin D3: Biochemical characterization and production in normal subjects and patients with pseudohyperparathyroidism. J Clin Endocr Metab 62:192. Hollis BW, Iskersky VN, Chang MK 1989 In vitro metabolism of 25-hydroxyvitamin D3 by human trophoblastic homogenates, mitochondria, and microsomes: Lack of evidence for the presence of 25-hydroxyvitamin D3-1α-hydroxylase and 24R-hydroxylase. Endocrinology 125:1224–1230. Diaz L, Sanchez I, Avila E, Halhali A, Vilchis F, Larrea F 2000 Identification of a 25-hydroxyvitamin D3 1α-hydroxylase gene transcription product in cultures of human syncytiotrophoblast cells [In Process Citation]. J Clin Endocrinol Metab 85: 2543–2549.

79 29. Zehnder D, Bland R, Williams MC, McNinch RW, Howie AJ, Stewart PM, Hewison M 2001 Extrarenal expression of 25-hydroxyvitamin D3-1α−hydroxylase. J Clin End Metab 86: 888–894. 30. Segersten U, Correa P, Hewison M, Hellman P, Dralle H, Carling T, Akerstrom G, Westin G 2002 25-hydroxyvitamin D(3)-1alpha-hydroxylase expression in normal and pathological parathyroid glands. J Clin Endocrinol Metab 87:2967–2972. 31. Schwartz GG, Whitlatch LW, Chen TC, Lokeshwar BL, Holick MF 1998 Human prostate cells synthesize 1,25-dihydroxyvitamin D3 from 25-hydroxyvitamin D3. Cancer Epidemiol Biomarkers Prev 7:391–395. 32. Jones G, Ramshaw H, Zhang A, Cook R, White J, Petkovich M 1999 Expression and activity of vitamin D-metabolizing cytochrome P450s (CYP1α and CYP24) in human nonsmall cell lung carcinomas. Endocrinology 140:3303–3310. 33. Henry HL, Norman AW 1975 Presence of renal 25-hydroxyvitamin-D-1-hydroxylase in species of all vertebrate classes. Comp Biochem Physiol 50B:441–434. 34. Nahm TH, Lee SW, Fausto A, Sonn Y, Avioli LV 1979 25-(OH)D, a circulating vitamin D metabolite in fish. Biochem Biophys Res Commun 89:396–402. 35. Egaas E, Lambertsen G 1979 Naturally occurring vitamin D3 in fish products analyzed by high performance liquid chromatography using vitamin D2 as an internal standard. J Am Oil Chem Soc 56:188. 36. Takeuchi A, Okano T, Kobayashi T 1991 The existence of 25-hydroxyvitamin D3-1α-hydroxylase in the liver of carp and bastard halibut. Life Sci 48:275–282. 37. Brown PB, Robinson EH 1992 Vitamin D studies with channel catfish (Ictalurus punctatus) reared in calcium-free water. Comp Biochem Physiol 103A:213–219. 38. Rao DS, Raghuramulu N 1996 Food chain as origin of vitamin D in fish. Comp Biochem Physiol 114A:15–19. 39. Rao DS, Raghuramulu N 1999 Vitamin D3 and its metabolites have no role in calcium and phosphorus metabolism in Tilapia mossambica. J Nutr Sci Vitaminol (Tokyo) 45:9–19. 40. Buffenstein R, Sergeev IN, Pettifor JM 1993 Vitamin D hydroxylases and their regulation in a naturally vitamin D– deficient subterranean mammal, the naked mole rat (Heterocephalus glaber). J Endcrinol 138:59–64. 41. Sergeev IN, Buffenstein R, Pettifor JM 1993 Vitamin D receptors in a naturally vitamin D-deficient subterranean mammal, the naked mole rat (Heterocephalus glaber): Biochemical characterization. Gen Comp Endocrinol 90:338–345. 42. Henry HL, Norman AW 1974 Studies on calciferol metabolism. IX. Renal 25-hydroxyvitamin D3-1-hydroxylase. Involvement of cytochrome P-450 and other properties. J Biol Chem 249: 7529–7535. 43. Hagenfeldt Y, Pedersen JI, Bjorkhem I 1988 Properties of guinea-pig kidney 25-hydroxyvitamin D3-1α-hydroxylase assayed by isotope dilution-mass spectrometry. Biochem J 250:521–526. 44. Churchill PF, deAlvare LR, Kimura T 1978 Topological studies of the steroid hydroxylase complexes in bovine adrenocortical mitochondria. J Biol Chem 253:4924–4929. 45. Ghazarian JG, Jefcoate CR, Knutson JC, Orme-Johnson W, DeLuca HF 1974 Mitochondrial cytochrome P450. A component of chick kidney 25-hydroxycholecalciferol1α−hydroxylase. J Biol Chem 249:3026–3033. 46. Hiwatashi A, Nishii Y, Ichikawa Y 1982 Purification of cytochrome P-450-1α (25-hydroxyvitamin D3-1α hydroxylase) of bovine kidney mitochondria. Biochem Biophys Res Commun 105:320–327.

80 47. Burgos-Trinidad M, Ismail R, Ettinger RA, Prahl JM, DeLuca HF 1992 Immunopurified 25-hydroxyvitamin D 1α-hydroxylase and 1,25-dihydroxyvitamin D 24-hydroxylase are closely related but distinct enzymes. J Biol Chem 267:3498–3505. 48. Moorthy B, Mandel ML, Ghazarian JG 1991 Amino-terminal sequence homology of two chick kidney mitochondrial proteins immunoisolated with monoclonal antibodies to the cytochrome P450 of the 25-hydroxyvitamin D3-1α hydroxylase. J Bone Miner Res 6:199–204. 49. Wakino S, Meguro M, Suzuki H, Saruta T, Ogishima T, Ishimura Y, Shinki T, Suda T 1996 Evidence for 54-kD protein in chicken kidney as a cytochrome P450 with a high molecular activity of 25-hydroxyvitamin D3 1α-hydroxylase. Gerentology 42:67–77. 50. Gray RW, Ghazarian JG 1989 Solubilization and reconstitution of kidney 25-hydroxyvitamin D3 1α- and 24-hydroxylases from vitamin D-replete pigs. Biochem J 259:561–568. 51. Postlind H 1990 Separation of the cytochromes P-450 in pig kidney mitochondria catalyzing 1α-, 24- and 26-hydroxylations of 25-hydroxyvitamin D3. Biochem Biophys Res Commun 168:261–266. 52. Nelson DR 1999 Cytochrome P450 and the individuality of species. Arch Biochem Biophys 369:1–10. 53. Okamura T, John ME, Zuber MX, Simpson ER, Waterman MR 1985 Molecular cloning and amino acid sequence of the precursor form of bovine adrenodoxin: evidence for a previously unidentified COOH-terminal peptide. Proc Natl Acad Sci USA 82:5705–5709. 54. Sakihama N, Hiwatashi A, Miyatake A, Shin M, Ichikawa Y 1988 Isolation and purification of mature bovine adrenocortical ferredoxin with an elongated carboxyl end. Arch Biochem Biophys 264:23–29. 55. Cupp JR Vickery LE 1989 Adrenodoxin with a COOHterminal deletion (des 116-128) exhibits enhanced activity. J Biol Chem 264:1602–1607. 56. Coughlan VM, Cupp JR, Vickery LE 1988 Purification and characterization of human placental ferredoxin. Arch Biochem Biophys 264:376–382. 57. Tang C, Kain SR, Henry HL 1993 The phorbol ester 12-O-tetradecanoyl-phorbol-13-acetate stimulates the dephosphorylation of mitochondrial ferredoxin in cultured chick kidney cells. Endocrinology 133:1823–1829. 58. Yoon PS, DeLuca HF 1980 Purification and properties of chick renal mitochondrial ferredoxin. Biochemistry 19:2165–2171. 59. Driscoll WJ, Omdahl JL 1989. Characterization and N-terminal amino acid sequence of multiple ferredoxins in kidney and adrenal mitochondria. Eur J Biochem 185:181–187. 60. Tang C, Henry HL 1993 Overexpression in Escherichia coli and affinity purification of chick kidney ferredoxin [published erratum appears in J Biol Chem 1993 268(15):11474]. J Biol Chem 268:5069–5076. 61. Omdahl JL, Wilson K, Swerdlow H, Driscoll WJ 1992 Molecular cloning and immunological characterization of porcine kidney ferredoxin. Arch Biochem Biophys 293:213–218. 62. Blanchard RK, Henry HL 1996 Chick kidney ferredoxin: Complementary DNA cloning and vitamin D effects on mRNA levels. Comp Biochem Physiol [B] 114B:337–344. 63. Kagimoto K, McCarthy JL, Waterman MR, Kagimoto M 1988 Deduced amino acid sequence ot mature chick testis ferredoxin. Biochem Biophys Res Commun 155:379–383. 64. Kagimoto M, Kagimoto K, Simpson ER, Waterman MR 1988 Transcription of the bovine adrenodoxin gene produces two species of mRNA of which only one is translated into adrenodoxin. J Biol Chem 263:8925–8928.

HELEN L. HENRY

65. Chang C-Y, Wu D-A, Mohandas TK, Chung B-C 1990 Structure, sequence, chromosomal location and evolution of the human ferredoxin gene. DNA Cell Biol 9:205–212. 66. Voutilainen R, Picado-Leonard J, DiBlasio AM, Miller WL 1988 Hormonal and developmental regulation of adrenodoxin messenger ribonucleic acid in steroidogenic tissues. J Clin Endocrinol Metab 66:383–388. 67. John ME, John TS, Boggaram V, Simpson ER, Waterman MR 1986 Transcriptional regulation of steroid hydroxylase genes by corticotropin. Proc Natl Acad Sci USA 83:4715–4719. 68. Kramer RE, Anderson CM, Peterson JA, Simpson ER, Waterman MR 1982 Adrenodoxin biosynthesis by bovine adrenal cells in monolayer culture. J Biol Chem 257: 14921–14925. 69. Miller WL 1988 Molecular biology of steroid hormone synthesis. Endocr Rev 9:295–318. 70. Simpson ER, Waterman MR 1983 Regulation by ACTH of steroid hormone biosynthesis in the adrenal cortex. Can J Biochem Cell Biol 61:692–707. 71. Gnanaiah W, Omdahl JL 1986 Isolation and characterization of pig kidney mitochondrial ferredoxin: NADP+ oxidoreductase. J Biol Chem 261:12649–12654. 72. Chu JW, Kimura T 1973 Studies on the adrenal steroid hydroxylases. Molecular and catalytic properties of adrenodoxin reductase (a flavoprotein). J Biol Chem 248:2089–2094. 73. Brandt ME, Vickery LE 1992 Expression and characterization of human mitochondrial ferredoxin reductase in Escherichia coli. Arch Biochem Biophys 294:735–740. 74. Solish SB, Picado-Leonard J, Morel Y, Kuhn RW, Mohandas TK, Hanukoglu I, Miller WL 1988 Human adrenodoxin reductase: Two mRNAs encoded by a single gene on chromosome 17cen q25 are expressed in steroidogenic tissues. Proc Natl Acad Sci USA 85:7104–7108. 75. Gacad MA, Adams JS 1993 Identification of a competitive binding component in vitamin D-resistant New World primate cells with a low affinity but high capacity for 1,25-dihydroxyvitamin D3. J Bone Miner Res 8:27–35. 76. Wu S, Ren S., Chen H, Chun RF, Gacad MA, Adams JS 2000 Intracellular vitamin D binding proteins: Novel facilitators of vitamin D-directed transactivation. Mol Endocrinol 14: 1387–1397. 77. Wu S, Chun R, Gacad MA, Ren S, Chen H, Adams JS 2002 Regulation of 1,25-dihydroxyvitamin D synthesis by intracellular vitamin D binding protein-1. Endocrinology 143:4135–4138. 78. Nykjaer A, Dragun D, Walther D, Vorum H, Jacobsen C, Herz J, Melsen F, Christensen EI, Willnow, TE 1999 An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) vitamin D3. Cell 96:507–515. 79. Hilpert J, Wogensen L, Thykjaer T, Wellner M, Schlichting U, Orntoft TF, Bachmann S, Nykjaer A, Willnow TE 2002 Expression profiling confirms the role of endocytic receptor megalin in renal vitamin D3 metabolism. Kidney Int 62:1672–1681. 80. Leheste JR, Melsen F, Wellner M, Jansen P, Schlichting U, Renner-Muller I, Andreassen TT, Wolf E, Bachmann S, Nykjaer A, Willnow TE 2003 Hypocalcemia and osteopathy in mice with kidney-specific megalin gene defect. FASEB J 17:247–249. 81. Takeyama K, Kitanaka S, Sato T, Kobori M, Yanagisawa J, Kato S 1997 25-Hydroxyvitamin D3 1α-hydroxylase and vitamin D synthesis. Science 277:1827–1830. 82. Shinki T, Shimada H, Wakino S, Anazawa H, Hayashi M, Saruta T, DeLuca HF, Suda T 1997 Cloning and expression of rat 25-hydroxyvitamin D3-1α-hydroxylase cDNA. Proc Natl Acad Sci USA 94:12920–12925.

CHAPTER 5 The 25-Hydroxyvitamin D 1α-Hydroxylase

83. St-Arnaud R, Messerlian S, Moir JM, Omdahl JL, Glorieux FH 1997 The 25-hydroxyvitamin D 1α-hydroxylase gene maps to the pseudovitamin D deficiency rickets (PDDR) disease locus. J Bone Miner Res 12:1552–1559. 84. Fu GK, Lin D, Zhang MY, Bikle DD, Shackleton CH, Miller WL, Portale AA 1997 Cloning of human 25-hydroxyvitamin D1α-hydroxylase and mutations causing vitamin D-dependent rickets type 1. Mol Endocrinol 11:1961–1970. 85. Monkawa T, Yoshida T, Wakino S, Shinki T, Anazawa H, DeLuca HF, Suda T, Hayashi M, Saruta T 1997 Molecular cloning of cDNA and genomic DNA for human 25-hydroxyvitamin D3 1α-hydroxylase. Biochem Biophys Res Commun 239:527–533. 86. Yoshida T, Yoshida N, Nakamura A, Monkawa T, Hayashi M, Saruta T 1999 Cloning of porcine 25-hydroxyvitamin D3 1α-hydroxylase and its regulation by cAMP in LLC-PK1 cells. J Am Soc Nephrol 10:963–970. 87. Fu GK, Portale A, Miller WL 1997 Complete structure of the human gene for the vitamin D 1α−hydroxylase, P450c1α. DNA Cell Biol 16:1961–1970. 88. Kong XF, Zhu XH, Pei YL, Jackson DM, Holick MF 1999 Molecular cloning, characterization, and promoter analysis of the human 25-hydroxyvitamin D3-1α-hydroxylase gene. Proc Natl Acad Sci USA 96:6988–6993. 89. Panda DK, Kawas SA, Seldin MF, Hendy GN, Goltzman D 2001 25-Hydroxyvitamin D 1α-hydroxylase: Structure of the mouse gene, chromosomal assignment, and developmental expression. J Bone Miner Res 16:46–56. 90. Henry HL, Midgett RJ, Norman AW 1974 Studies on calciferol metabolism X: Regulation of 25-hydroxyvitamin D3-1-hydroxylase, in vivo. J Biol Chem 249:7584–7592. 91. Colston KW, Evans IM, Spelsberg TC, Macintyre I 1977 Feedback regulation of vitamin D metabolism by 1,25-dihydroxycholecalciferol. Biochem J 164:83–89. 92. Takahashi N, Miyaura C, Horiuchi N, Suda T 1981 Regulation and its refractoriness of 25-hydroxyvitamin D3 metabolism in vitamin D deficiency. J Nutr Sci Vitaminol 27:429–437. 93. Henry HL 1979 Regulation of the hydroxylation of 25hydroxyvitamin D3 in vivo and in primary cultures of chick kidney cells. J Biol Chem 254:2722–2729. 94. Spanos E 1983 Regulation of the renal 25OHD3 hydroxylases by vitamin-D3 metabolites and analogs in primary chick kidney cell cultures. Acta Endocrinol 104:25–26. 95. Turner RT, Bottemiller BL, Howard GA, Baylink DJ 1980 In vitro metabolism of 25-hydroxyvitamin D3 by isolated rat kidney cells. Proc Natl Acad Sci USA 77:1537–1540. 96. Murayama A, Takeyama K, Kitanaka S, Kodera Y, Kawaguchi Y, Hosoya T, Kato S 1999 Positive and negative regulations of the renal 25-hydroxyvitamin D3 1α-hydroxylase gene by parathyroid hormone, calcitonin, and 1α,25(OH)2D3 in intact animals. Endocrinology 140:2224–2231. 97. Shinki T, Ueno Y, DeLuca HF, Suda T 1999 Calcitonin is a major regulator for the expression of renal 25-hydroxyvitamin D3-1alpha-hydroxylase gene in normocalcemic rats. Proc Natl Acad Sci USA 96:8253–8258. 98. Condamine L, Vztovsnik F, Friedlander G, Menaa C, Garabédian M 1994 Local action of phosphate depletion and insulin-like growth factor 1 on in vitro production of 1,25dihydroxyvitamin D3 by cultured mammalian kidney cells. J Clin Invest 94:1673–1679. 99. Bland R, Walker EA, Hughes SV, Stewart PM, Hewison M 1999 Constitutive expression of 25-hydroxyvitamin D3-1αhydroxylase in a transformed human proximal tubule cell line: evidence for direct regulation of vitamin D metabolism by calcium. Endocrinology 140:2027–2034.

81 100. Yoshida T, Yoshino J, Hayashi M, Saruta T 2002 Identification of a renal proximal tubular cell-specific enhancer in the mouse 25-hydroxyvitamin D 1alpha-hydroxylase gene. J Am Soc Nephrol 13:1455–1463. 101. Fraser DR, Kodicek E 1973 Regulation of 25-hydroxycholecalciferol-1-hydroxylase activity in kidney by parathyroid hormone. Nature New Biol 241:163–166. 102. Baksi SN, Kenny AD 1979 Acute effects of parathyroid extract on renal vitamin D hydroxylases in Japanese quail. Pharmacology 18:169–174. 103. Booth BE, Tsai HC, Morris RC 1985 Vitamin-D status regulates 25-hydroxyvitamin-D3-1-α-hydroxylase and its responsiveness to parathyroid hormone in the chick. J Clin Invest 75:155–161. 104. Walker AT, Stewart AF, Korn EA, Shiratori T, Mitnick MA, Carpenter TO 1990 Effect of parathyroid hormone-like peptides on 25-hydroxyvitamin D-1α-hydroxylase activity in rodents. Am J Physiol 258:E297–E303. 105. Armbrecht HJ, Boltz MA, Hodam TL 2003 PTH increases renal 25(OH)D3-1alpha-hydroxylase (CYP1alpha) mRNA but not renal 1,25(OH)2D3 production in adult rats. Am J Physiol Renal Physiol 284:F1032–F1036. 106. Spanos E, Brown DJ, Macintyre I 1979 Regulation of 25(OH)-D3 metabolism by parathyroid hormone in primary chick kidney cell cultures. FEBS Lett 105:31–34. 107. Henry HL 1981 Insulin permits parathyroid hormone stimulation of 1,25-dihydroxyvitamin D3 production in cultured kidney cells. Endocrinology 108:733–735. 108. Rost CR, Bikle DD, Kaplan RA 1981 In vitro stimulation of 25-hydroxycholecalciferol 1α-hydroxylation by parathyroid hormone in chick kidney slices: Evidence for a role for adenosine 3′5′-monophosphate. Endocrinology 108:1002–1006. 109. Fukase M, Birge SJ, Rifas L, Avioli LV, Chase LR 1982 Regulation of 25 hydroxyvitamin D3-1-hydroxylase in serumfree monolayer culture of mouse kidney. Endocrinology 110: 1073–1075. 110. Kramer R, Goltzman D 1982 Parathyroid hormone stimulates mammalian 25-hydroxyvitamin D3-1α-hydroxylase in vitro. Endocrinology 110:294–296. 111. Henry HL 1985 Parathyroid hormone modulation of 25-hydroxyvitamin D3 metabolism by cultured chick kidney cells is mimicked and enhanced by forskolin. Endocrinology 116:503–507. 112. Noland TA, Henry HL 1983 Protein phosphorylation in chick kidney: Response to parathyroid hormone, cyclic AMP, calcium and phosphatidyl serine. J Biol Chem 258: 538–546. 113. Henry HL, Luntao EM 1989 Interactions between intracellular signals involved in the regulation of 25-hydroxyvitamin D3 metabolism. Endocrinology 124:2228–2234. 114. Ro HK, Tembe V, Favus MJ 1992 Evidence that activation of protein kinase C can stimulate 1,25-dihydroxyvitamin D3 secretion by rat proximal tubules. Endocrinology 131: 1424–1428. 115. Dunlay R, Hruska KA 1990 PTH receptor coupling to phospholipase C is an alternative pathway of signal transduction in bone and kidney. Am J Physiol 258:F223–F231. 116. Galante L, Colston KW, MacAuley SJ, Macintyre I 1972 Effect of calcitonin on vitamin D metabolism. Nature 238: 271–273. 117. Nesbitt T, Lobaugh B, Drezner MK 1987 Calcitonin stimulation of renal 25-hydroxyvitamin-D-1α-hydroxylase activity in hypophosphatemic mice—evidence that the regulation of calcitriol production is not universally abnormal in x-linked hypophosphatemia. J Clin Invest 79:15–19.

82 118. Horiuchi N, Takashita H, Matsumoto T, Takahashi N, Shimazawa E, Suda T, Ogata E 1979 Salmon calcitonininduced stimulation of 1α,25-dihydroxycholecalciferol synthesis in rats involving a mechanism independent of adenosine 3′:5′:-cyclic monophosphate. Biochem J 184:269–275. 119. Larkins RG, MacAuley SJ, Rapoport A, Martin TJ, Tullock BR, Macintyre I 1974 Effects of nucleotides,hormones, ions, and 1,25-dihydroxycholecalciferol on 1,25-dihydroxycholecalciferol production in isolated chick renal tubules. Clin Sci Mol Med 46:569–582. 120. Rasmussen H, Wong M, Bikle DD, Goodman DBP 1972 Hormonal control of the renal conversion of 25-hydroxycholecalciferol to 1,25-dihydroxycholecalciferol. J Clin Invest 51:2502–2504. 121. Henry HL, Noland TA, Al-Abdaly FA, Cunningham NS, Luntao EM, Amdahl LD 1982 Responses of cultured chick kidney cells to parathyroid hormone and calcitonin. In: Norman AW, Schaefer K, Herrath DV, Grigoleit H-G (eds.) Vitamin D: Chemical, Biochemical, and Clinical Endocrinology of Calcium Metabolism. Walter de Gruyter, Berlin. 122. Yoshida N, Yoshida T, Nakamura A, Monkawa T, Hayashi M, Saruta T 1999 Calcitonin induces 25-hydroxyvitamin D3 1αhydroxylase mRNA expression via protein kinase C pathway in LLC-PK1 cells. J Am Soc Nephrol 10:2474–2479. 123. Hughes MR, Brumbaugh PF, Hussler MR, Wergedal JE, Baylink DJ 1975 Regulation of serum 1alpha,25-dihydroxyvitamin D3 by calcium and phosphate in the rat. Science 190:578–580. 124. Baxter LA, DeLuca HF 1976 Stimulation of 25-hydroxyvitamin D3-1α-hydroxylase by phosphate depletion. J Biol Chem 251:3158–3161. 125. Gray RW, Napoli JL 1983 Dietary phosphate deprivation increase 1,25-dihydroxyvitamin D3 synthesis in rat kidney in vitro. J Biol Chem 258:1152–1155. 126. Rader JI, Baylink DJ, Hughes MR, Safilian EF, Haussler MR 1979 Calcium and phosphorus deficiency in rats: Effect on PTH and 1,25-dihydroxyvitamin D3. Am J Physiol 236: E118–E122. 127. Gray RW, Wilz DR, Caldas AE, Lemann J 1977 The importance of phosphate in regulating plasma 1,25-(OH)2-vitamin D levels in humans: Studies in healthy subjects, in calcium-stone formers and in patients with primary hyperparathyroidism. J Clin Endocrinol Metab 45:299–306. 128. Maierhofer WJ, Gray RW, Lemann J 1984 Phosphate deprivation increases serum 1,25-(OH)2-vitamin-D concentrations in healthy men. Kidney Int 25:571–575. 129. Portale AA, Halloran BP, Murphy MM, Morris FC 1986 Oral intake of phosphorus can determine the serum concentration of 1,25-dihydroxyvitamin D by determining its production rate in humans. J Clin Invest 77:7–12. 130. Portale AA, Halloran BP, Murphy MM, Morris RC, Jr. 1986 Oral intake of phosphorus can determine the serum concentration of 1,25-dihydroxyvitamin D by determining its production rate in humans. J Clin Invest 77:7–12. 131. Castillo L, Tanaka Y, DeLuca HF, Sunde ML 1977 The stimulation of 25-hydroxyvitamin D3-1α-hydroxylase by estrogen. Arch Biochem Biophys 179:211–217. 132. Baksi SN, Kenny AD 1980 Estradiol-induced stimulation of 25-hydroxyvitamin D3-1-hydroxylase in vitamin D-deficient Japanese quail. Pharmacology 20:298–303. 133. Pike JW, Spanos E, Coloston KW, Macintyre I, Haussler MR 1978 Influence of estrogen on renal vitamin D hydroxylases and serum 1α,25-(OH)2D3 in chicks. Am J Physiol 235: E338–E343.

HELEN L. HENRY

134. Henry HL 1981 25-OH-D3 metabolism in kidney cell cultures: Lack of a direct effect of estradiol. Am J Physiol 240:E119–E124. 135. Findling JW, Adams ND, Lemann J Jr., Gray RW, Thomas CJ, Tyrrell JB 1982 Vitamin D metabolites and parathyroid hormone in Cushing’s syndrome: Relationship to calcium and phosphorus homeostasis. J Clin Endocrinol Metab 54: 1039–1044. 136. Chesney RW, Hamstra AJ, Mazess RB, DeLuca HF, O’Reagan S 1978 Reduction of serum 1,25-dihydroxyvitamin D3 in children receiving glucocorticoids. Lancet 2:1123–1125. 137. Seeman E, Kumar R, Hunder GG, Scott M, Heath H, Riggs BL 1980 Production, degradation, and circulating levels of 1,25Dihydroxyvitamin D in health and in chronic glucocorticoid excess. J Clin Invest 66:664–669. 138. Henry HL 1986 Effect of dexamethasone on 25-hydroxyvitamin D3 metabolism in chick kidney cell cultures. Endocrinology 118:940–944. 139. Akeno N, Matsunuma A, Maeda T, Kawane T, Horiuchi N 2000 Regulation of vitamin D-1α-hydroxylase and 24-hydroxylase expression by dexamethasone in mouse kidney. J Endocrinol 164:339–348. 140. Henry HL 1985 Effect of ketoconazole and miconazole on 25-hydroxyvitamin D3 metabolism by cultured chick kidney cells. J Steroid Biochem Mol Biol 23:991–994. 141. Boass A, Toverud SU 1995 Suppression of circulating calcitriol and duodenal active Ca transport by ketoconazole in pregnant rats. Am J Physiol Endocrinol Metab 269: E934–E939. 142. Glass AR, Eil C 1986 Ketoconazole-induced reduction in serum 1,25-dihydroxyvitamin D. J Clin Endocr Metab 63: 766–769. 143. Saggese G, Bertelloni S, Baroncelli GI, Di Nero G 1993 Ketoconazole decreases the serum ionized calcium and 1,25-dihydroxyvitamin D levels in tuberculosis-associated hypercalcemia. Am J Dis Child 147:270–273. 144. Kowal J 1983 The effect of ketoconazole on steroidogenesis in cultured mouse adrenal cortex tumor cells. Endocrinology 112:1541–1543. 145. Loose DS, Kan PB, Hirst MA, Marcas RA, Feldman D 1983 Ketoconazole blocks adrenal steroidogenesis by inhibiting cytochrome P450-dependent enzymes. J Clin Invest 71: 1495–1499. 146. Zhao J, Tan BK, Marcelis S, Verstuyf A, Bouillon R 1996 Enhancement of antiproliferative activity of 1α,25-dihydroxyvitamin D3 (analogs) by cytochrome P450 enzyme inhibitors is compound- and cell-type specific. J Steroid Biochem Mol Biol 57:197–202. 147. Peehl DM, Skowronski RJ, Leung GK, Wong ST, Stamey TA, Feldman D 1994 Antiproliferative effects of 1,25-dihydroxyvitamin D3 on primary cultures of human prostatic cells. Cancer Res 54:805–810. 148. Peehl DM, Seto E, Feldman D 2001 Rationale for combination ketoconazole/vitamin D treatment of prostate cancer. Urology 58:123–126. 149. Yoshida T, Fujimori T, Nabeshima Y 2002 Mediation of unusually high concentrations of 1,25-dihydroxyvitamin D in homozygous klotho mutant mice by increased expression of renal 1α-hydroxylase gene. Endocrinology 143:683–689. 150. Tsujikawa H, Kurotaki Y, Fujimori T, Fukuda K, Nabeshima YI 2003 Klotho, a gene related to a syndrome resembling human premature aging, functions in a negative regulatory circuit of vitamin D endocrine system. Mol Endocrinol.

CHAPTER 5 The 25-Hydroxyvitamin D 1α-Hydroxylase

151. Murayama A, Takeyama K, Kitanaka S, Kodera Y, Hosoya T, Kato S 1998 The promoter of the human 25-hydroxyvitamin D3 1α−hydroxylase gene confers positive and negative responsiveness to PTH, calcitonin, and 1α,25(OH)2D3. Biochem Biophys Res Commun 249:11–16. 152. Brenza HL, DeLuca HF 2000 Regulation of 25-hydroxyvitamin D3 1α-hydroxylase gene expression by parathyroid hormone and 1,25-dihydroxyvitamin D3 [In Process Citation]. Arch Biochem Biophys 381:143–152. 153. Bringhurst FR, Juppner H, Guo J, Urena P, Potts JT Jr Kronenberg HM, Abou-Samra AB, Segre GV 1993 Cloned, stably expressed parathyroid hormone (PTH)/PTH-related peptide receptors activate multiple messenger signals and biological responses in LLC-PK1 kidney cells. Endocrinology 132:2090-2098. 154. Racusen LC, Monteil C, Sgrignoli A, Lucskay M, Marouillat S, Rhim JG, Morin JP 1997 Cell lines with extended in vitro

83 growth potential from human renal proximal tubule: characterization, response to inducers, and comparison with established cell lines. J Lab Clin Med 129:318–329. 155. Brenza HL, Kimmel-Jehan C, Jehan F, Shinki T, Wakino S, Anazawa H, Suda T, DeLuca HF 1998 Parathyroid hormone activation of the 25-hydroxyvitamin D3-1α-hydroxylase gene promoter. Proc Natl Acad Sci USA 95:1387–1391. 156. Armbrecht HJ, Hodam TL, Boltz MA 2003 Hormonal regulation of 25-hydroxyvitamin D3-1alpha-hydroxylase and 24-hydroxylase gene transcription in opossum kidney cells. Arch Biochem Biophys 409:298–304. 157. Gao XH, Dwivedi PP, Choe S, Alba F, Morris HA, Omdahl JL, May BK 2002 Basal and parathyroid hormone induced expression of the human 25-hydroxyvitamin D 1alpha-hydroxylase gene promoter in kidney AOK-B50 cells: role of Sp1, Ets and CCAAT box protein binding sites. Int. J Biochem Cell Biol 34:921–930.

CHAPTER 6

The 25-Hydroxyvitamin D 24-Hydroxylase JOHN OMDAHL BRIAN MAY

Office of Research, University of New Mexico School of Medicine, Albuquerque, NM 87131-5166 School of Molecular and Biomedical Science, University of Adelaide, South Australia 5005, Australia

I. Background II. Enzyme Structure and Function III. Cellular Expression and Regulation

I. BACKGROUND A. Enzyme Function and Regulated Expression Vitamin D is a secosteroid whose biological function is dependent upon its metabolic activation and turnover. These metabolite pathways contain specific hydroxylase enzymes that are members of the cytochrome P450 superfamily of mixed-function monooxygenases. Bioactivation of vitamin D involves the sequential actions of 25-hydroxylase and 1-hydroxylase enzymes leading to the synthesis of the hormonally active secosteroid 1,25-dihydroxyvitamin D [1,25(OH)2D]* (Fig. 1). These two enzymes are discussed in Chapter 4 (vitamin D 25-hydroxylase) and Chapter 5 (25-hydroxyvitamin D 1α-hydroxylase) and will be mentioned in this chapter only on a comparative basis to 25-hydroxyvitamin D24(R)-hydroxylase cytochrome P450c24 (CYP24), the enzyme that directs the side-chain metabolism of 25-hydroxylated vitamin D metabolites, which leads to their terminal physiological inactivation and turnover. Most cellular actions of vitamin D are mediated through the secosteroid hormone 1,25(OH)2D and involve the transcription of vitamin D–dependent genes. These regulatory processes involve the coordinated modulation and coupling of rapid signal-transduction pathways with slower acting ligand-dependent transcription factors [1]. In both cases, the secosteroid ligand binds to a ligand-specific receptor. The rapid response receptor is located in the cellular membrane of target tissues and initiates rapid signaling responses through a receptor that has been referred to as the membrane-associated receptor [e.g., mVDR or

*The term vitamin D will be used to denote vitamin D3 unless stated otherwise. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

IV. Molecular Aspects References

membrane-associated rapid response steroid (MSSRS) receptor complex] [2,3]. The hormone receptor for the transcription process involves the nuclear vitamin D receptor (nVDR), which is a ligand-dependent transcription factor that functions as a VDR:1,25(OH)2D heterodimeric complex with the cis-retinoic acid: RXR complex (i.e., VDR-RXR) to regulate vitamin D–dependent genes associated with development and homeostasis (see Chapters 13–17 for details). Cellular and ambient levels 1,25(OH)2D are regulated through the hormone’s synthesis and degradation. The regulated 1α-hydroxylase (i.e., P4540c1 or CYP27B1) directs the hormone’s biosynthesis. In a complementary manner, the regulatory 24(R)-hydroxylase enzyme (i.e., P450c24 or CYP24A1) mediates degradation of 1,25(OH)2D through side-chain oxidation reactions. Expression of the two enzymes is most commonly regulated in a diametric manner that functions to maintain the hormone’s level at a proper set point. For example, low-calcium states are associated with an elevation in parathyroid hormone (PTH) that functions to up-regulate P450c1 (i.e., CYP27B1) while down-regulating P450c24 (i.e., CYP24A1) expression. The associated increase in 1,25(OH)2D functions to stimulate calcium absorption and correct the low-calcium state. Control of 1,25(OH)2D overproduction and the attendant development of a hypercalcemic state is autoregulated through a negative feedback system whereby 1,25(OH)2D down-regulates its oversynthesis (i.e., suppresses CYP27B1 expression) while up-regulating expression of CYP24A1. Of particular importance is the action of CYP24A1 to catabolically inactivate 1,25(OH)2D and thereby function to counterregulate the hormone’s cellular and ambient concentration. The regulatory and expression linkages between CYP27B1 and CYP24A1 are fundamental to understanding the vitamin D regulatory pathway and will serve as a point of reference in this chapter as well as other Copyright © 2005, Elsevier, Inc. All rights reserved.

86

JOHN OMDAHL AND BRIAN MAY

Liver Liver UV light 7-dehydro cholesterol

Skin Skin

P450c25

Pre-Vitamin D

Diet

25(OH)D3 Kidney Kidney

Target Target Tissues Tissues P450c24

P450c1 Calcitroic acid

1,25(OH)2D3 24,25(OH)22D33

1,25(OH)22D33 RXR VDR

( )

P450c24

P450c1

1,24,25(OH)33D33 VDRE

Genes

P450c24 Calcitroic acid

Excretion

FIGURE 1

Activation and molecular pathways for vitamin D. Integrative schematic of vitamin D synthesis, metabolism and molecular action.

sections in this book in which various aspects of vitamin D metabolism are discussed extensively.

B. Enzyme Cellular Expression The 24(R)-hydroxylase enzyme displays a broad tissue distribution wherein it is expressed in nearly all cells. A number of examples are discussed in the regulatory section in the second half of this chapter. The major route of enzyme induction involves the combined action of 1,25(OH)2D and VDR to increase transcription of the CYP24A1 gene as discussed in the latter two sections of this chapter. Recognizing the importance of 1,25(OH)2D in the induction of CYP24A1, the enzyme is not expressed in vitamin D– deficient animals. Consequently, the cellular expression of CYP24A1 is linked tightly to the vitamin D status and coexpression of VDR. Because of the ubiquitous distribution of VDR [4,5], most cells express a basal level of CYP24A1 or respond to increased 1,25(OH)2D levels by inducing biosynthesis of the enzyme, particularly in kidney and small intestine [6–8]. Consequently, the cellular level of CYP24A1 is particularly influenced by the concentration of VDR and the level of 1,25(OH)2D, the latter reflecting both locally produced hormone [9] and hormone acquired from the circulation. Hence the broad cellular induction of CYP24A1 by 1,25(OH)2D can be envisioned as an effector arm of the negative feedback system that functions to maintain

proper hormone level and avoids the buildup of a potentially toxic secosteroid. Kidney is the major endocrine organ for 1,25(OH)2D and therefore there has been considerable interest in the regulation of renal CYP24A1 expression. Using the rat developmental model, Matkovits and Christakos [6] reported low levels of renal mRNA at birth that remained unchanged for several weeks and then increased substantially over the next 20 months, thereby establishing kidney as the major site of basal enzyme expression. In recent studies, quantitation of rat kidney mRNA levels using a real time RT-PCR protocol [10] confirmed this developmental profile for CYP24A1 mRNA expression (PH Anderson, S Iida, BK May, HA Morris, submitted). A small fall in renal CYP24A1 mRNA level was seen between 3 and 9 weeks of age followed by a continuous increase until 24 months. In a comparative measurement, CYP27B1 (i.e., 1αhydroxylase) mRNA was highest at 3 weeks of age and then decreased particularly between 3 and 15 weeks, followed by a further slight decrease over 24 months. The high production of renal 1,25(OH)2D during earlier development would ensure adequate circulating 1,25(OH)2D and hence calcium for bone growth. Serum levels of 1,25(OH)2D are determined by the expression levels of both CYP27B1 and CYP24A1 in the kidney [10]. CYP24A1 is expressed in both proximal and distal tubules [11–13], although induced enzyme expression occurs to a higher level in the proximal tubule [13].

87

CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase

Mechanistic and regulatory studies of 1,25(OH)2D3 action and CYP24A1 expression have relied on a host of cell-culture models. Investigations with cultured cells have revealed induction of the CYP24A1 gene in many different cell types. Enzyme activity and/or CYP24A1 mRNA levels have been detected in numerous primary and transformed cultures that include intestinal, osteoblast, monocyte/macrophage, keratinocyte, and cancer cells [14–25]. In most of these studies, cells incubated in a serum-free medium expressed little if any 24(R)-hydroxylase activity or CYP24A1 mRNA, but enzyme and message levels were highly induced by 1,25(OH)2D treatment. In addition to 1,25(OH)2D, other more cell-type-selective regulators of CYP24A1 gene expression have been studied, and include PTH, calcitonin, phorbol ester, dexamethasone, and gammainterferon (see Section III of this chapter). The induction of CYP24A1 mRNA in primary cultures of rat renal tubular cells was prevented by actinomycin D, verifying that the increase in CYP24A1 mRNA level was due to gene transcription [15]. Cycloheximide also inhibited the induction of CYP24A1 mRNA in these cells, demonstrating a continued need for protein synthesis [15,19], which could involve both signal transduction and transcription regulatory proteins.

C. Enzyme Properties P450c24 acquired its name 25-hydroxyvitamin D 24(R)-hydroxylase based upon its action to 24(R)hydroxylate, the prehormone 25-hydroxyvitamin D3 [25OHD] to 24(R),25-dihydroxyvitamin D [24,25 (OH)2D] [26]. The P450c24 enzyme was subsequently shown to 24(R)-hydroxylate 1,25(OH)2D3 to 1,24,25 (OH)3D3 [27]. See Figure 2. The CYP24A1 in some species expresses predominantly 23(S)-hydroxylase activity [11,28]; therefore, CYP24A1 isoenzymes direct both C23 and C24 oxidative pathways. In both instances, the dominant C23 or C24 hydroxylation activity directs the initial step associated with the enzyme’s side-chain oxidative reactions associated with the inactivation of 25-hydroxylated vitamin D metabolites as discussed in the next section on enzyme pathways. The CYP24A1 enzyme is a protoporphyrin IX hemoprotein (~52 kDa) that binds oxygen and directs the regiospecific and stereospecific hydroxylation and oxidation of secosteroid substrate molecules using electrons derived from NADPH [29] (Fig. 2). The vitamin D enzyme is distinct from the 24(S)-hydroxylase enzyme involved with cholesterol metabolism [30] (i.e., CYP2R1). CYP24A1 is an inner mitochondrial membrane enzyme that is part of a mini–electron

Cyp24A1red

Cyp24A1ox

25 (OH)D3 + O2 [1,25 (OH)2D3]

24,25 (OH)2D3 + H2O [1,24,25 (OH)3D3]

FIGURE 2

Terminal reaction of CYP24A1. The use of molecular oxygen by reduced CYP24A1 to 24(R)-hydroxylate 25-hydroxyvitamin D substrates. red, reduced; ox, oxidized.

transport chain consisting of NADPH-ferredoxin reductase, ferredoxin (an iron–sulfur protein), and the terminal cytochrome CYP24A1 [31] (Fig. 2). Electrons are delivered through a specific protein:protein interaction between the enzyme and the electron donor ferredoxin [32]. Since the vitamin D 24(R)-hydroxylase enzyme is less than 40% similar to other P450s, it was assigned to a new P450 family and given the cytochrome P450 abbreviation CYP24A1. Consistent with its unique protein sequence, CYP24A1 does not react immunologically with several liver P450 enzymes or mitochondrial P450c27 (CYP27A1) that expresses 25-hydroxylase activity and is a family member with the 25-hydroxyvitamin D 1α-hydroxylase enzyme (i.e., CYP27).

D. Enzyme Pathways CYP24A1 directs the biosynthesis of 24,25(OH)2D and 1,24,25(OH)3D. These two vitamin D metabolites express less (i.e., 24,25(OH)2D) or approximately equivalent (i.e., 1,24,25(OH)3D) biological activity compared to 1,25(OH)2D. A biological role for 24,25 (OH)2D has been studied most extensively [2,3] because of its higher cellular and ambient concentration. Both metabolites are initial reactants in the C24-oxidation pathway that leads to the side-chain oxidation of 25-hydroxy metabolites to terminal C23-carboxylic acid end products [33–35] (Fig. 3). An investigation with partially purified 24(R)-hydroxylase enzyme from pig kidney documented copurification of the 23- and 24-hydroxylase activities [31], which was consistent with one enzyme expressing both activities. All catalytic activities in the C24-oxidation pathway were subsequently shown to be expressed by recombinant rat CYP24A1 [28,36]. The multifunctional enzyme initially oxidizes the 24-hydroxyl group to a 24-keto (oxo) functionality. Subsequent steps involve the 23(S)-hydroxylation of the 24-oxo metabolites and

88

JOHN OMDAHL AND BRIAN MAY

to favor an initial C23(S)- or C24(R)-hydroxylation. It is evident, therefore, that CYP24A1 is a multicatalytic enzyme [28,36–38] that is capable of directing all reactions in the side-chain oxidation of 25-hydroxylated vitamin D metabolites to cyclic lactone (C23 pathway) [38] or C23 carboxylic-acid (C24 pathway) [37], or both [28,38] end products (Fig. 3). In this regard, CYP24A1 is the central enzyme involved with determining the biological half-life of 1,25(OH)2D and its therapeutic analogs.

II. ENZYME STRUCTURE AND FUNCTION A. Cytochrome P450 Motif and Function

FIGURE 3

Multicatalytic action of CYP24A1. Reactions of the C23 and C24 oxidative pathways catalyzed by CYP24A1 with lactol and carboxylic acid end products.

oxidative cleavage between C23 and C24 followed by oxidation of the C23 hydroxyl, via a C23 aldehyde intermediate, to 24,25,26,27-tetranor-23-COOH-vitamin D or calcitroic acid from 25(OH)2D and 1,25(OH)2D substrates, respectively (Fig. 3). However, the initial hydroxylation occurs at C23 in several CYP24A1 isoforms from different species, which result in an alternate pathway that leads to a side-chain cyclic lactone as final product (Fig. 3). For example, rat CYP24A1 expresses predominantly the C24-oxidation pathway, whereas human [28] and guinea pig [11] enzymes express significant levels of C23-hydroxylase activity that results in the synthesis of both lactone and carboxylicacid end products. Expression of predominantly 23(S)or 24(R)-hydroxylase activity in different isoforms would appear to be determined by point-residue differences in the substrate binding and catalytic center that direct substrate alignment in the heme active site so as

Cytochromes P450 constitute a superfamily of hemecontaining enzymes that direct the monooxygenation of hydrophobic substrates. More than 1000 genes have been identified and divided into 70 families [39] that metabolize preferred substrates ranging from polycyclic aromatic hydrocarbons to fatty acids and steroids. Although all P450s have a similar structural core, they have unique amino acid sequences that determine the enzyme’s substrate recognition and mediate regiospecific and stereospecific monooxygenation reactions within the active site. Molecular oxygen used in the monooxygenase reactions is activated through its interaction with the iron atom in the active site hemeprotoporphyrin IX molecule. In most instances, the oxygen molecule is activated to an oxyferryl functionality that stereospecifically directs the oxygenation/ hydroxylation of substrates that enter through a surface oriented substrate-access channel and subsequently align with the heme molecule in the enzyme’s central active site.

B. Molecular Modeling Only a limited number of P450 enzymes have been crystallized and characterized by X-ray diffraction analysis [40–44]. However, because of the molecules’ conserved structural attributes, it is possible to use molecular-modeling software (e.g., Insight-II and Sybyl) to develop enzyme similarity models that closely resemble the enzyme’s native conformation. Using the Insight-II Modeler program (Accelrys Inc.) and crystallographic coordinate values for P450-BM3 and P450-2C5, it has been possible to derive a similarity model for rat P450c24 (CYP24A1) (A Annalara, S Graham, JL Omdahl, unpublished data). Most of the substrate-interactive residues in CYP24A1 are located predominantly in the B′, F, G, and I helices, and the F-G loop.

89

CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase

FIGURE 4

Model of rat CYP24A1 active site. A similarity model for CYP24A1 active site that shows alignment of the side-chain C24 with the heme (red) and alignment of the A-ring (adjacent to the 1,25(OH)2D notation) with F-249 in the F-helix.

Substrate docking of 1,25(OH)2D in the CYP24A1 model placed the C24 position in close proximity (2 Å) to the heme oxyferryl group with the A-ring of the molecule oriented toward the F-helix (Fig. 4). The A-ring displayed close interactions (3 Å) with met-246 and phe-249 of the F-helix. A prominent role for phe-249 in binding 1,25(OH)2D was determined though site-directed mutagenesis as discussed in the following section. The CYP24A1 similarity model is being used in conjunction with site-directed mutagenesis to map the full spectrum of substrate-binding sites within the substrate-access channel and active site.

C. Metabolic, Kinetic, and Spectroscopic Analysis 1. METABOLIC ANALYSIS

Consistent with the P450 nature of CYP24A1, the enzyme’s activity is impeded by P450 azole inhibitors [45–47]. The CYP24A1 purified from rat kidney [29] was shown to metabolize 25(OH)D to 24,25(OH)2D and 1,25(OH)2D to 1,24,25(OH)3D. Subsequent studies using bacterial-expressed recombinant enzyme demonstrated the action of CYP24A1 to side-chain metabolize 1,25(OH)2D [37]. The role of CYP24A1 to

direct side-chain oxidation reactions beyond the initial 24(R)-hydroxylation was established using recombinant enzyme preparations from bacteria and baculovirus [28,36–38]. Most interestingly, the rat enzyme was demonstrated to metabolize all steps in the C24-oxidation pathway leading to C23-carboxylic acid end products [36,37]. In contrast to the rat enzyme, human CYP24A1 [28,38] was found to catalyze both the C24 and C23 oxidation pathways that resulted in the tetranor C23-carboxylic acid and the 26,23-lactone end products, respectively (Fig. 3). Evident from these studies was the multicatalytic activity of CYP24A1 isozymes to direct all reactions in the C24 and C23 oxidative pathways. In contrast to previous recombinant CYP24A1 studies that used enzyme membrane preparations or detergent extracts (i.e., impure enzyme), a more recent investigation succeeded in purifying bacterial recombinant rat CYP24A1 (A Annalora, JL Omdahl, unpublished data). The purified enzyme expressed all steps in the C24 oxidation pathways, which demonstrated clearly the multicatalytic activity of CYP24A1. 2. KINETIC ANALYSIS

The rate of substrate metabolism by CYP24A1 has been analyzed using cellular, detergent extracted, and

90

JOHN OMDAHL AND BRIAN MAY

purified enzyme preparations. Km values from cellular and detergent extract preparations [19,48–51] span a large range (0.01–1.5 µM) in which Km values for 1,25(OH)2D are 1/4 to 1/30 that of 25(OH)D, which supported the concept that CYP24A1 has a higher preference for 1,25(OH)2D hormone. However, a more recent study [52] with solubilized enzyme determined the Km for 25(OH)D to be 0.6 µM and that of 1,25(OH)2D was 35-fold higher (20.9 µM), which indicated an enzyme preference for 25(OH)D instead of 1,25(OH)2D. The diversity of Km values is due in part to the sequential–multicatalytic properties of CYP24A1 and the combined effect of electron transfer from the co-redox ferredoxin partner and dissociation of intermediate reactants from the enzyme. Using purified rat CYP24A1, it has been possible to focus on the first step of the multicatalytic reaction path and to regulate the rate of electron transfer to the enzyme (A Pastuezyn, JL Omdahl, unpublished data). Using these conditions and those that favor the 24(R)-hydroxylation step, we have determined Km values for 25(OH)D and 1,25(OH)2D that are less than 1 µM in which 25(OH)D had a slightly lower Km and greater turnover number (A Pastuezyn, JL Omdahl, unpublished data). Also evident from the kinetic analysis with pure enzyme was the slow reaction kinetics for CYP24A1 [i.e., turnover number (TN) of 2–20 min−1] [29], compared to the higher rate observed for another mitochondrial P450 from the bile-acid pathway (i.e., CYP27A1, TN = 40–50 min−1) [53,54]. 3. SPECTRAL ANALYSIS

CYP24A1 displays a typical type-I mitochondrial P450 CO-reduced difference spectrum with a 450-nm Soret band that shifts to 420 nm upon enzyme inactivation. Because of the presence of a heme molecule in the enzyme’s active site, it is possible to follow the absolute spectral shift from 420 to 390 nm that occurs upon addition of substrate (i.e., substrate-induced spectral shift) and determine the dissociation constant (Kd) for substrate with enzyme [55]. Such binding techniques revealed a high specificity of the enzyme for vitamin D secosteroids containing a 1α- and/or 25-hydroxyl group (A Annalora, JL Omdahl et al., submitted). Both 1α-hydroxyvitamin D and 1,25(OH)2D displayed the lowest Kd (highest binding affinity) for CYP24A1. However, only 1,25(OH)2D was capable of being 24(R)-hydroxylated [29], which demonstrated the necessity of a C25-hydroxyl group for the physiological side-chain 24(R)-hydroxylation and subsequent oxidation reactions catalyzed by CYP24A1. In a related manner, the side-chain metabolism of 1,25(OH)2D2 to calcitroic acid is impeded by the vitamin D2 side-chain

structure modifications (i.e., double bond at C22 and methyl group at C24) [35,56]. The Kd for 25(OH)D is sixfold higher than observed for 1,25(OH)2D. But because of its lower Km, higher turnover number, and 1000-fold higher ambient level, the conversion of 25(OH)D to 24,25(OH)2D is a favored reaction. In addition, the 24(R)-hydroxylated metabolite has a decreased affinity for CYP24A1 (A Annalora, JL Omdahl, submitted), which favors its dissociation from enzyme, especially in cells where electron transfer is limiting. The combined effects result in a significant serum level of 24,25(OH)2D (2–5 ng/ml) that is 1/5 to 1/10 the concentration of the parent prehormone 25(OH)D. In contrast, the ambient level of 1,24,25(OH)3D is insignificant because of the metabolite’s low cellular concentration and greater affinity to remain bound to CYP24A1 (i.e., lower Kd) for further oxidative metabolism. Introduction of a single mutation in the F-helix of CYP24A1 has a profound impact on substrate binding and perturbation of the heme spin-state in the active site [57] (A Annalora, JL Omdahl, unpublished data). Mutation of phenylalanine 249 to threonine (F249T) caused a three- to sixfold increase in the Kd for 25(OH)D and 1,25(OH)2D, respectively. Based upon the CYP24A1 similarity model, phenylalanine 249 is projected to interact with the A-ring of 1,25(OH)2D when bound in the active site (Fig. 4). Loss of this hydrophobic interaction through the F249T mutation prevents the hormone from binding properly in the substrate pocket. As a consequence, the last oxidative sequence in the C24 pathway (i.e., 24,25,26,27-tetranor-1,23(OH)2D to calcitroic acid) is suppressed because of the misalignment of substrate side-chain in the active site. Mutagenesis of phenylalanine 249 to alanine (F249A) or tyrosine (F249Y) also confounded the substrate binding. However, in this instance the mutation affected an early step in the C24 pathway involving conversion of 1,24,25(OH)3D to 24-oxo-1,25(OH)2D. It is evident, therefore, that the combination of molecular modeling and site-directed mutagenesis can be used to identify amino acid residues that are involved with the multiple catalytic activities associated with CYP24A1.

III. CELLULAR EXPRESSION AND REGULATION A. CYP24A1 Knockout and Transgenic Animals CYP24A1 cDNA clones have been isolated for rat [58], human [59], chicken [60], mouse [61], and pig [62].

91

CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase

The human CYP24A1 gene is located on chromosome 20 [63], but there are no known diseases that can be attributed to this gene, although aberrant overexpression is potentially oncogenic as discussed in a later section. The exon–intron structures have been determined for the rat [64] and mouse [61] CYP24A1 genes, which are “typical genes” of length about 15 kb. The biological functions of CYP24A1 gene expression in vivo have been pursued in conventional CYP24A1 knockout mice and also in transgenic rats constitutively expressing CYP24A1. Knockout mice were generated using embryonic stem cell technology so that the heme-binding domain of the mouse CYP24A1 gene was deleted ensuring that no functional CYP24A1 protein was produced [65,66]. Homozygous null mice exhibited a high circulating level of 1,25(OH)2D in keeping with the lack of catabolism through the C23/24 oxidation pathway. There was abnormal bone histology characterized by excessive undermineralized bone matrix and this raised the possibility that a deficiency of the CYP24A1 oxidation product 24,25(OH)2D may be responsible. While 24,25(OH)2D has been considered an inactive metabolite, there is evidence for a biological function of 24,25(OH)2D in bone physiology, particularly in the development of growth plate chondrocytes [67,68]. However, the mouse phenotype was rescued when the CYP24A1 null mice were mated with vitamin D receptor (VDR) knockout mice [69], demonstrating that the bone abnormality was not due to a deficiency of 24,25(OH)2D but most likely to elevated 1,25(OH)2D, which requires VDR for its biological actions [70,71]. In a different approach, constitutive expression of CYP24A1 has been achieved in the tissues of transgenic rats with the introduction of a rat CYP24A1 cDNA sequence under the control of the viral MLV-LTR promoter [72,73]. It was anticipated that characteristics of this transgenic rat would resemble those observed in CYP27B1 knockout mice where the lack of production of 1,25(OH)2D results in hypocalcemia, hyperparathyroidism, growth retardation, and rickets [74,75]. Unexpectedly, after 8 weeks the transgenic rats developed albuminuria and hyperlipidemia with a significant reduction in circulating 25(OH)D and 24,25(OH)2D levels accompanied by a lowering of bone mineral density [72,73]. On the other hand, the plasma concentration of 1,25(OH)2D in the transgenic rats was the same as in wild type, most likely because of up-regulation of CYP27B1. Excess renal excretion of albumin is probably responsible for the reduction in plasma 25(OH)D due to its prevention of 25(OH)D reuptake by renal tubular cells [73]. Infusion of transgenic rats with 25(OH)D prevented the mineralization defect by making the circulating 1,25(OH)2D

more available [73]. It is not clear how constitutive expression of CYP24A1 alters renal function, resulting in albuminuria and marked reduction of plasma 25(OH)D. Overall, these studies are significant and suggest that 1,25(OH)2D is involved in functions other than regulation of vitamin D metabolism. Thus far, there have been no reports on the ablation or overexpression of CYP24A1 in selected tissues, which could shed light on tissue specific roles for CYP24A1.

B. Regulation of Enzyme Expression 1. KIDNEY

The most important physiological regulators of renal CYP24A1 expression in the adult are 1,25(OH)2D and PTH [76–78]. These hormones will therefore influence the production of renal 1,25(OH)2D and hence the level of circulating hormone. Renal CYP24A1 gene transcription is induced by 1,25(OH)2D in a VDR-dependent fashion. This action of 1,25(OH)2D occurs chiefly in the proximal convoluted tubules [79] and will assist in the maintenance of 1,25(OH)2D levels, in particular in the prevention of hypercalcemia. In marked contrast, the peptide hormone PTH functions through its cellular receptor to suppress renal CYP24A1 expression [50] when circulating levels of calcium are low [76–78]. This inhibition occurs through alteration of CYP24A1 mRNA stability and alterations in CYP24A1 gene transcription [62,80,81] (R Serda and JL Omdahl, unpublished). However, the PTH-mediated inhibition of CYP24A1 expression is not observed in the intestine that lacks PTH receptor [50]. PTH-differentially regulates CYP24A1 in proximal and distal convoluted tubular cells with PTH inhibiting 1,25(OH)2D-stimulated expression of CYP24A1 in proximal convoluted tubules, but potentiating the initial induction of CYP24A1 by 1,25(OH)2D in distal convoluted tubular cells [79]. It has been proposed that in distal tubules a coordinated action of PTH and 1,25(OH)2D stimulates calcium reabsorption, but when levels of 1,25(OH)2D are high, PTH stimulates the induction of CYP24A1 by 1,25(OH)2D, thereby lowering cellular 1,25(OH)2D and avoiding excessive calcium uptake by these tubules [79]. In addition to 1,25(OH)2D and PTH, the peptide hormone calcitonin [82–84] is observed to also contribute to regulation of renal CYP24A1 expression. Calcitonin is secreted by the thyroid parafollicular C cells in response to elevated serum calcium levels and is well known to have hypocalcemic actions that result from an inhibition of osteoclastic bone resorption and stimulation of renal calcium excretion [83,85]. Whether these actions contribute significantly to the

92

JOHN OMDAHL AND BRIAN MAY

maintenance of ambient calcium levels in human adults is uncertain [83]. Alternatively, calcitonin could contribute to calcium homeostasis through regulating 1,25(OH)2D levels. Renal CYP27B1 expression in response to calcitonin has been studied. For example, calcitonin induces CYP27B1 expression in kidney LLC-PK1 and MCT proximal tubule cells [86,87], but there is no effect on CYP27B1 levels in the proximal convoluted HKC-8 cells [88]. In distal parts of the nephron, calcitonin stimulates CYP27B1 under normocalcemic conditions [89]. However, the physiological rationale for any induction of CYP27B1 by calcitonin is unclear, since calcitonin is produced in response to high serum calcium. Recent studies have established for the first time that expression of the rat CYP24A1 promoter can be substantially induced by calcitonin in both transiently and stably transfected kidney HEK293 cells (X-P Gao, P Dwivedi, JL Omdahl, H Morris, and B May, submitted). These findings raise the possibility that such a stimulatory action by calcitonin could be important in vivo for modulating renal 1,25(OH)2D production. 2. INTESTINE

Intestinal cells do not express PTH or calcitonin receptors. Circulating 1,25(OH)2D stimulates intestinal calcium import, and the induction of intestinal CYP24A1 by the hormone is likely to be an important factor in determining the level of circulating secosteroid [76]. In this regard, induction of CYP24A1 in the intestine is more responsive to acute changes in 1,25(OH)2D compared with the kidney [90]. 3. BONE

An important role for CYP24A1 in bone development was suggested from the phenotype of CYP24A1 null mice that were found to exhibit abnormal intramembranous bone formation as discussed earlier. This phenotype may result from a local toxic level of 1,25(OH)2D due to the loss of CYP24A1. In earlier studies, Nishimura et al. [91] showed that 1,25(OH)2D induces expression of CYP24A1 in osteoblast cell lines and observed that there is a greater induction by 1,25(OH)2D in the less mature osteoblastic cell lines. This data has been confirmed by the quantification of CYP24A1 mRNA [10] in both rat and human osteoblast-like cell lines (P Anderson, H Morris, and B May, unpublished data). Nonstimulated levels of CYP24A1 mRNA are very low but can be increased dramatically up to 400,000-fold by 1,25(OH)2D. This 1,25(OH2)D–dependent stimulation of CYP24A1 expression appears blunted in fully differentiated, mature cells. Locally produced 1,25(OH)2D, acting in an autocrine/paracrine fashion, is now considered to

have an antiproliferative and prodifferentiation effect on many cells types, both normal and cancerous [92]. It is therefore an attractive hypothesis that 1,25(OH)2D may modulate CYP24A1 expression during osteoblast development so that low levels of 1,25(OH)2D are maintained during the proliferative phase but are increased during the mineralization phase. In support of this concept, the addition of 1,25(OH)2D to proliferative osteoblasts in culture inhibits proliferation but addition to more mature osteoblasts promotes expression of genes involved in differentiation and mineralization [93]. Osteoblasts also express PTH receptor, but there is no evidence that PTH lowers expression of 1,25(OH)2D-induced CYP24A1 as seen in the kidney [94]. In fact, in rat osteoblastic UMR-106 cells, PTH synergises with 1,25(OH)2D to initially enhance CYP24A1 mRNA and protein levels by a mechanism that involves PKA activity [94,95], although mRNA expression is lowered at later time intervals (R Serda and JL Omdahl, unpublished data). It is possible that the early synergistic increase in CYP24A1 in response to PTH and 1,25(OH)2D serves to ensure lowered cellular levels of 1,25(OH)2D commensurate with early osteoblast development. In a manner similar to PTH, the potent glucocorticoid dexamethasone markedly stimulates CYP24A1 mRNA expression and CYP24A1 activity in 1,25(OH)2Dtreated UMR-106 cells [96]. In vivo, glucorticoids markedly affect bone metabolism, albeit in a complex fashion, and an enhanced catabolism of 1,25(OH)2D in the presence of glucorticoid in osteoblasts could influence net bone resorption by counteracting the anabolic effect of 1,25(OH)2D in stimulating mineralization in these cells [96]. 4. MONOCYTES/MACROPHAGES

Immune cell types such as monocytes and macrophages express both CYP27B1 and CYP24A1 with the locally synthesized 1,25(OH)2D exerting its effects on the immune system through both autocrine and paracrine mechanisms [97]. The production of 1,25(OH)2D expression by monocytes or macrophages is markedly enhanced by exposure of cells to the cytokine INF-gamma [98]. This response mimics the autoimmune disease sarcoidosis where there is an abnormally high production of 1,25(OH)2D by diseaseactivated macrophages with associated hypercalcemia. INF-gamma-stimulated production of 1,25(OH)2D results both from up-regulation of CYP27B1 expression [99,100] and from a novel repression of CYP24A1 induction by 1,25(OH)2D that suppresses the ability of 1,25(OH)2D to control its own synthesis. The molecular aspects of this repression have been defined [101] and are discussed later.

93

CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase

5. KERATINOCYTE

As mentioned, 1,25(OH)2D can influence the differentiation of many cells in culture [92] and keratinocytes are a well studied system in which locally synthesized 1,25(OH)2D acts in an autocrine/paracrine fashion to inhibit cell proliferation and to stimulate terminal keratinocyte differentiation [102,103]. As well as expressing CYP27B1 for local 1,25(OH)2D production, keratinocytes have the capacity to metabolise 1,25(OH)2D through induction of CYP24A1 [45]. At pharmacological levels, 1,25(OH)2D and its analogs can be used in the treatment of hyperproliferative diseases such as psoriasis, but their effectiveness is restricted by expression of CYP24A1, and this has led to a search for selective inhibitors of CYP24A1 [45]. It appears that a constitutive expression of CYP27B1 enzyme in keratinocytes synthesizes local 1,25(OH)2D in which this level of 1,25(OH)2D functions to finely tune CYP24A1. Interestingly, the level of CYP27B1 enzyme activity in keratinocytes markedly declines when CYP24A1 activity is induced by 1,25(OH)2D [45,104]. This inhibition of CYP27B1 enzyme activity is not due to a decrease in CYP27B1 mRNA but rather reflects an increased CYP24A1 directed catabolism of both substrate 25(OH)D and the product 1,25(OH)2D of CYP27B1 [45,104]. This finding that CYP27B1 activity is not directly regulated by 1,25(OH)2D differs from the situation in kidney, where excess 1,25(OH)2D appears to inhibit transcription of the CYP27B1 gene [105,106]. 6. CANCER CELLS

CYP24A1 may be an important player in contributing to the dysregulation of cell growth through lowering the local cellular 1,25(OH)2D concentration. In this regard, CYP24A1 has been suggested as a potential oncogene, since the gene was found coamplified with a known oncogene in human breast cancer tissue [107]. Elevated expression of CYP24A1 mRNA has been noted in highly malignant human colon carcinomas and may be due to amplification of the CYP24A1 gene [108]. A study of different prostate cancer cell lines showed that the growth inhibitory properties of 1,25(OH)2D correlated inversely with the level of CYP24A1 [109]. For example, the human prostate cancer DU-145 cell line expresses high levels of CYP24A1 in response to 1,25(OH)2D, and there is only minimal inhibition of cell proliferation. However, the antiproliferative activity of 1,25(OH)2D on these cells can be significantly restored through inhibition of CYP24A1 activity [110]. The high levels of induction of CYP24A1 observed in the prostate cancer cell lines [109] do not correlate with VDR amounts and the mechanism is not known. A significantly lower prevalence of prostate, mammary, and

colon cancers in Asia compared with Western industrialized countries has been linked to typical Asian diets with high amounts of soy products rich in the phytoestrogens, particularly isoflavonoids, suggesting that these compounds may prevent cancer [111]. Genestein, a major isoflavonoid, has been investigated recently. Administration of genestein or soy protein decreased expression of CYP24A1 mRNA in mouse colon [111] while genestein was also found to be a potent inhibitor of CYP24A1 mRNA expression in DU-145 cells [112].

IV. MOLECULAR ASPECTS In keeping with the widespread tissue distribution of CYP24A1, expression of the gene can be influenced by a broad range of regulatory agents. Molecular studies have focused on the control of CYP24A1 expression by 1,25(OH)2D, PTH, calcitonin, phorbol ester, dexamethasone, and gamma interferon, and their actions are summarized in Fig. 5. Control of CYP24A1 expression by these regulators appears to occur predominantly through modulation of gene transcription, although PTH is a notable exception in which it can alter CYP24A1 mRNA stability.

A. 1,25(OH)2D and the Rat CYP24A1 Promoter 1,25(OH)2D is the most potent inducer of CYP24A1 expression in kidney, osteoblast and other cell types. Investigations into the architecture of the rat CYP24A1 proximal promoter region have permitted a detailed understanding of the 1,25(OH)2D inductive process. Functional transcription factor binding sites have been identified that underlie basal and induced expression (see Figs. 6 and 7). Of central importance are two vitamin D response elements (VDREs) present on the non-coding strand and about 100 bp apart [113–118]. The proximal VDRE-1 at −136/−150 (5′-AGGTGAgtgAGGGCG-3′) and the distal VDRE-2 at −244/−258 (5′-GGTTCAgcgGGTGCG-3′) are DR3-type elements comprising two hexameric half-site direct repeats with a spacing of three nucleotides. Such a tandem arrangement of two VDREs remains unique to the CYP24A1 promoter and underlies transcriptional synergy between the VDREs and the high level of 1,25(OH)2D induction [117]. Electrophoretic mobility shift assays employing specific antibodies have shown that each VDRE binds VDR together with its partner RXR. Earlier experiments revealed than an oligonucleotide encompassing VDRE-1 bound the VDR/RXR heterodimer

94

JOHN OMDAHL AND BRIAN MAY

TPA (Dex) 1,25D

mVDR

calcitonin

CTR

KIDNEY

PKC

mVDR ERKs

PKA/PKC

P P AP1-like, Ets-1

P

RXR, Ets-1 other factors

PKA

Sp1, NF-Y

endonuclease

CYP24 mRNA RXR VDR

RXR VDR

VDRE-2

VDRE-1

PTHR

PTH

Induced CYP24 Gene

CYP24 protein

P mitochondria

Stat-1 VDR

NUCLEUS P Stat-1

IFN-gamma receptor

MONOCYTES/ MACROPHAGES

INF-gamma

FIGURE 5 Transcription regulation model in kidney and monocyte/macrophage cells. The transcription factors on the CYP24A1 promoter are shown for the regulatory agents phorbol ester (TPA), dexamethasone (dex), 1,25(OH)2D acting on either the nuclear VDR (nVDR) at the membrane or a novel membrane-inserted VDR (mVDRm), calcitonin acting on the calcitonin receptor (CTR), PTH acting on the PTH receptor (PTHR), and interferon gamma acting on the interferon gamma receptor. It is proposed that PTH action results in degradation of the CYP24A1 mRNA through the action of an endonuclease.

to a much lesser extent than an oligonucleotide encompassing VDRE-2 [117]. However, mutagenesis of each site within a –298 bp CYP24A1 promoter construct demonstrated that VDRE-1 contributes far more strongly to the 1,25(OH)2D inductive response in all cell types examined including COS-1, kidney HEK-293, UMR106, and ROS17/2.8 cell lines. This suggested that transcription factors bound nearby to VDRE-1 could enhance its activity.

Subsequently, a functional binding site for the ubiquitous transcription factor Ets-1 (Ets binding site, EBS; see Fig. 6) was identified downstream of VDRE-1 at −119/−128 on the noncoding strand and was required for maximal 1,25(OH)2D induction in COS-1 cells [1,119]. The EBS cooperates with VDRE-1 but not VDRE-2, presumably reflecting the close proximity of the VDRE-1 [1]. Direct protein–protein interaction between Ets-1 and VDR has been demonstrated by the

RNAP II machinery −298

RXR VDR

VDRE-2

RXR VDR

AP-1 like VDRE-1 EBS

GC

CCAAT

FIGURE 6 Cis-binding elements in CYP24A1 promoter. Functional binding sites for transcription factors within the first –298 bp of the rat CYP24A1 promoter. The arrows indicate that the sequences for VDRE-1, VDRE-2, and EBS are on the noncoding strand.

95

CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase

two-hybrid assay [1]. As shown in Fig. 7B, Ets-1 bound at the EBS probably interacts with liganded VDR/RXR on VDRE-1 [1] and also with coactivator complexes [78] bound to liganded VDR/RXR. The EBS does not contribute to the low basal CYP24A1 promoter expression observed in transfected cells in the absence of added 1,25(OH)2D [119]. A repressor complex bound to unliganded VDR/RXR on the CYP24A1 promoter maintains this low expression in the absence of secosteroid [120]. The presence of this repressor complex may sterically prevent the binding of Ets-1 (Fig. 7A). When 1,25(OH)2D binds to VDR, it is proposed that the repressor complex dissociates and is replaced first by a coactivator complex with histone acetyltransferase activity that remodels

A

Basal

AP-1

Ets-1

X

X

chromatin [78,121] and then by a mediator complex, as described later. Since the EBS does not contribute to CYP24A1 promoter-directed basal expression detected in cells, it was clear that other transcription factor binding sites must be present. A GC box at −101/−114 (5′-CCCGCCCC-3′) immediately downstream of EBS and an inverted CCAAT box further downstream at –51/–62 (5′CATTGGC-3′) are functionally active (see Fig. 5; X-H Gao, P Dwivedi, and B May, unpublished data). Supershift analysis employing isolated nuclear extracts suggest that the ubiquitous factors Sp1 and NF-Y [122] bind to the GC and CCAAT box, respectively, with no binding detected of C/EBP alpha or beta to the CCAAT motif. Mutagenesis experiments established

HDAC? Repressor complex RXR VDR

VDRE-2

B

RNAP II machinery

Repressor complex RXR VDR

AP-1 like VDRE-1 EBS

weak GC

CCAAT

1,25D-induction

HAT

HAT CBP/p300

CBP/p300

p160

p160

RXR VDR

RXR VDR

VDRE-2

AP-1 like VDRE-1 EBS

RNAP II machinery

GC

CCAAT

C

RNAP II machinery

mediator complex RXR VDR

VDRE-2

RXR VDR

AP-1 like VDRE-1 EBS

Strong GC

CCAAT

FIGURE 7 Proposed binding of regulatory trans-proteins. (A) In the absence of 1,25(OH)2D, it is proposed that a repressor complex binds a VDR/RXR heterodimer on each VDRE. This complex may repress through modifying chromatin via histone deacetylase activity (HDAC) and/or preventing access to the AP-1 like binding protein and Ets-1. Low basal expression is driven by Sp1 and NF-Y bound at the GC and CCAAT box sites, respectively, through interaction with the RNA polymerase II machinery. (B) Following ligand binding to VDR, it is proposed that coactivators p160 and CBP/p300 are recruited and through histone acetyl transferase activity (HAT) modify chromatin and enhance binding of AP-1 like protein and Ets-1. Synergy may result from an interaction of all transcription factors with CBP except NF-Y and from interaction of the two coactivator complexes. (C) The two coactivator complexes are proposed to be replaced with a coactivator mediator complex that interacts with and activates RNA polymerase II and hence stimulates CYP24A1 gene transcription.

96

JOHN OMDAHL AND BRIAN MAY

(−) 1,25 (OH)2D + 1,25 (OH)2D UMR-106 cells 60x 10.00

36x

5.00

33x

38x 24x

6x

1.00 0.50

Mutated AP-1 like

Mutated EBS

Mutated GC + CCAAT

Mutated CCAAT

Mutated GC

WTpCYP24-298-luc

Mutated AP-1 like

Mutated EBS

Mutated GC + CCAAT

Mutated CCAAT

0

Mutated GC

0.25

WT pCYP24-298-luc

Relative Luciferase Activity

15.00

FIGURE 8

Mutational analysis of CYP24A1 promoter during induction of 1,25(OH)2D. Expression of transiently transfected wild type and mutant –298 bp CYP24A1 promoter constructs in UMR-106 cells. Cells were treated with 1,25(OH)2D at 10−7M. The relative luciferase activity represents the mean ±SD firefly luciferase/Renilla luciferase ratios of triplicate samples.

that the GC and CCAAT boxes predominantly are responsible for basal expression of the CYP24A1 promoter in the HEK-293 and UMR-106 cell lines examined. In UMR-106 cells, basal expression is substantially lowered when the CCAAT and GC boxes are mutated either individually or together (Fig. 8). A similar contribution from each site to basal expression is observed in stably transfected HEK-293T cells (Fig. 9). Sp1 and NF-Y could drive basal expression through interactions with the general transcription machinery (see Fig. 7A), notably with the general transcription factor TFIID bound at the TATA box of the CYP24A1 promoter. The GC and CCAAT box elements are important for induction of the CYP24A1 promoter by 1,25(OH)2D. In UMR-106 cells, when the GC and CCAAT box sites are both inactivated, 1,25(OH)2D-induced CYP24A1 promoter activity is reduced from about 60-fold to 24-fold (Fig. 9). The GC and CCAAT box sites are not responsive per se to 1,25(OH)2D, since mutagenesis of both VDREs in the CYP24A1 promoter abolishes induction by 1,25(OH)2D. In UMR-106 cells, mutagenesis of EBS has no effect on basal expression but lowers 1,25(OH)2D induction from 60-fold to 36-fold (Fig. 8). Another sequence further upstream of VDRE-1 at –163/−171, 5′-TGTCGGTCA-3′, has recently been characterized through site-directed mutagenesis of the

CYP24A1 promoter and shown to markedly stimulate 1,25(OH)2D induction, particularly in osteoblastic cell lines (P Dwivedi, J Kaplan, JL Omdahl, H Morris, and B May, in preparation). This site resembles an AP-1 site (consensus 5′-TGAG/CTCA-3′) and is referred to here as AP-1 like (as shown in Figs. 6 and 7). In UMR106 cells, inactivation of this sequence markedly lowers 1,25(OH)2D induction of the CYP24A1 promoter from 60-fold to six-fold (Fig. 8), and further analysis has established that the site cooperates with VDRE-1 and not VDRE-2. The AP-1 like sequence does not contribute to basal CYP24A1 promoter expression, perhaps because of the presence of the repressor complex bound on VDRE-1 (Figs. 8 and 7A). Within the AP-1 like motif, the sequence 5′-CGGTCA-3′ resembles a VDRE halfsite and has attracted attention previously when it was shown that mutagenesis of this sequence lowered 1,25(OH)2D induction [114,118]. Overall, the rat CYP24A1 proximal promoter (see Fig. 6) comprises a weak VDRE-1 encircled by compensatory transcription factors and a strong upstream VDRE-2 apparently devoid of nearby transcription factors. A model for the nuclear genomic action of 1,25(OH)2D that leads to stimulation of transcription of the CYP24A1 promoter can be proposed (Fig. 7). When 1,25(OH)2D binds to VDR located at each VDRE, the

97

CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase

−calcitonin

50.00

+calcitonin

Luciferase Activity / ug Protein

14x

45.00

2.8x

5.00

1.2x

Mutated GC

Mutated CCAAT

WTpCYP24-298-luc

Mutated GC

Mutated CCAAT

WT pCYP24-298-luc

0

FIGURE 9 Mutational analysis of CYP24A1 promoter during induction by calcitonin. Expression of wild-type and mutant –298 bp CYP24A1 promoter constructs in stably transfected HEK-293T cells expressing the calcitonin receptor. Cells were treated overnight with calcitonin (10 nM). The fire fly luciferase activity was normalised to the amount of cellular protein and the values represent the mean ± SD of triplicate samples.

repressor complex is dissociated and a p160 coactivator is recruited on each VDRE, together with other coactivators such as CBP/p300 and SKIP/NCoA-62 [78,121], so that cooperative interactions between the coactivator complexes result in transcriptional synergy. At VDRE-1, it is envisaged that recruitment of a coactivator complex by liganded VDR/RXR is enhanced by the AP-1 like protein, by Ets-1, and also by Sp1. The coactivator complex through its intrinsic histone acetyl transferase activity could modify chromatin permitting access to factors and facilitating assembly of the transcription machinery. Finally, in a proposed two-step mechanism, the coactivator complex would be replaced by the multiprotein mediator complex DRIP (VDR interacting proteins) that interacts with and activates RNAPII [78]. There is evidence that the induction of CYP24A1 mRNA by 1,25(OH)2D can be blocked by the protein synthesis inhibitor cycloheximide, implicating the obligatory production of a regulatory protein [81,123]. This protein is apparently not VDR or RXR, suggesting that for induction to proceed optimally, synthesis of a transcription factor or a coactivator component or a protein that stabilizes CYP24A1 mRNA is required [81].

B. 1,25(OH)2D and Mitogen-Activated Protein Kinase Activities In addition to its classical genomic action in the nucleus, it is now clear that 1,25(OH)2D also functions at the plasma membrane to rapidly activate downstream signaling pathways critical for transcription of the CYP24A1 promoter [1,119]. Such a nongenomic action is also a feature of steroid hormones, which is currently receiving considerable attention [124]. The importance of phosphorylation events in the induction of the CYP24A1 promoter by 1,25(OH)2D was first suggested by studies in COS-1 cells cotransfected with VDR, where overexpression of H-Ras increased the inductive response and also potentiated the effect of exogenously expressed wild-type Ets-1 acting through the EBS [119]. This response agreed with the proposal that the functionality of Ets-1 was dependent on phosphorylation of threonine residue 38. A mutant form of Ets-1 where threonine residue 38 was mutated to alanine lowered the level of induction of the CYP24A1 promoter to that observed with mutated EBS [1]. Also, a dominant negative Ras mutant inhibited 1,25(OH)2D induction to a level greater than that observed following

98 EBS inactivation. This demonstrated that the action of Ras on the CYP24A1 promoter was likely to be channeled not only through Ets-1 but also through other components on the promoter. Since MAP kinase activities are downstream effectors of Ras, a detailed investigation was undertaken into the possible role of these kinases in the activation of the rat CYP24A1 promoter by 1,25(OH)2D in VDRtransfected COS-1 cells [1]. Rapid stimulation of ERK1/2 and ERK5 activities by 1,25(OH)2D occurred in these cells with maximal activities peaking at 5–10 min and inhibition of activity directed by a Ras dominant negative clone. These data raised the possibility that ERK1/2 and ERK5 could be involved in the regulation of the CYP24A1 promoter. Studies with specific pharmacological inhibitors and dominant negative mutants established that in 1,25(OH)2D treated COS-1 cells, inhibition of the ERK1/2 module markedly lowered induction of the CYP24A1 promoter while the p38 and JNK modules were not required [1]. Stimulation of ERK5 activity by 1,25(OH)2D was required for maximal induction in which the secosteroid’s action was effected through phosphorylation of Ets-1. Activated ERK1 phosphorylated RXRalpha at serine 260, and when COS-1 cells were transfected with the mutant RXRalphaS260A clone (in which serine was replaced by alanine) induction of the CYP24A1 promoter was inhibited. VDR was not phosphorylated by either kinase activity. These studies established a critical role in these cells for the Ras-ERK1/2-RXRalpha signaling pathway for induction of the CYP24A1 promoter with an enhancing contribution from the Ras-ERK5 pathway through phosphorylation of Ets-1. An analysis of the signaling pathways that contribute to 1,25(OH)2D induction of the CYP24A1 promoter using specific pharmacological inhibitors and dominant negative clones has revealed that the contribution of ERK modules and other signaling molecules is dependent on the cell type. For example, in HEK-293 cells 1,25(OH)2D induction is dependent on JNK activity and there is little activation of ERK1/2. The contribution to promoter induction of Src and PI3Kinase activities also is cell type dependent. PKC (see Section III,B,4) is important for 1,25(OH)2D induction in most cell types examined. Overall these studies show that the nongenomic signaling pathways activated by 1,25(OH)2D are vital for CYP24A1 promoter activity with PKC/Ras/ ERK modules of particular importance. The precise role of the signaling pathways and the downstream targets on the CYP24A1 promoter (i.e. AP-1-like sequence, VDR/RXR, Ets-1, Sp1 and CCAAT-box binding protein) in the different cell types will require an in-depth analysis. As discussed previously [1], it is interesting that whereas overexpressed Ras facilitates

JOHN OMDAHL AND BRIAN MAY

1,25(OH)2D induction of the CYP24A1 promoter activity in COS-1 cells through phosphorylation of serine 160 in RXR, in Ras-transformed keratinocytes 1,25(OH)2D-dependent phosphorylation of serine 160 in RXR attenuates 1,25(OH)2D transactivation [125]. The molecular basis for this difference in the action of phosphorylated RXR is not known. As discussed earlier, the inhibitory action of genestein on CYP24A1 mRNA expression supports the proposal that soy consumption may protect against cancer through elevating local 1,25(OH)2D production. Genestein is a well-known tyrosine kinase inhibitor and the observed inhibitory effect on CYP24A1 mRNA expression is likely to be at the nongenomic level.

C. 1,25(OH)2D and the Nongenomic Response The rapid activations of ERK1/2 and ERK5 and other signaling molecules by 1,25(OH)2D, as a prelude to CYP24A1 promoter stimulation, are consistent with the nongenomic actions reported for steroid hormones [124]. Studies of the estrogen receptor (ER) provide a model in which ER is recruited to the plasma membrane by interaction with caveolin proteins that serve as scaffolds in caveolae, discrete domains of the plasma membrane [126,127]. Signaling molecules including various G proteins, Src, PI3Kinase, and Raf are also recruited to these scaffolds so that the ER in response to ligand can rapidly modulate a variety of signaling pathways within a confined space. There is evidence for the ER interacting directly with Src, and PI3K and stimulating PKC and PKA activities through various G proteins. Studies on the mechanism underlying the nongenomic action of 1,25(OH)2D are less extensive than those of estrogen/ER pathway. There is evidence that nongenomic actions of 1,25(OH)2D include G-protein stimulation of adenylate cyclase, phospholipase C beta [77] and phospholipase C gamma, the last through the tyrosine kinases Src and PI3K [128]. By analogy with ER, it can be envisaged that the classical VDR is recruited to the plasma membrane where it activates signaling cascades in response to 1,25(OH)2D. Indeed, nongenomic activities are lost from osteoblast cultures derived from homozygous mice expressing a mutant VDR lacking the first zinc finger necessary for DNA binding [129]. Presumably conformational changes in the mutant VDR interfere with membrane localization or binding to the downstream signaling molecules (i.e., function as dominant-negative VDR). On the other hand, it is possible that a unique VDR may participate in nongenomic signaling. A novel progesterone receptor has been identified with seven putative transmembrane domains, thus resembling the classic G protein coupled

99

CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase

receptors [130]. This raises the possibility that a novel membrane form of VDR, perhaps together with the classical VDR acting at the plasma membrane, contribute to nongenomic biological responses. In that regard, it is interesting that PKC activation by 1,25(OH)2D is still observed in osteoblast cultures isolated from VDR null mice where the second zinc finger of the DNA binding domain has been deleted [131,132]. Although these authors suggest that the nongenomic response is independent of the nuclear VDR activity, it is still possible that the absence of the second zinc finger alters downstream interactions necessary for PKC activation. Other evidence has shown that the ligand specificity for nongenomic activities in osteoblasts is distinct from that for nuclear gene activation [133].

lowered synergy from 7.1-fold to 3.3-fold and 1.1-fold, respectively. Hence both sites play a role, but the AP-1 like sequence plays a predominant role in TPA sensitivity. It can be proposed (see Fig. 5) that activated PKC in response to TPA further stimulates the activity of MAP kinase signaling pathways above that achieved by 1,25(OH)2D alone, leading to the phosphorylation and activation of AP-1 like protein and Ets-1 and overall enhancement of 1,25(OH)2D induction. Whether the nonresponsiveness of the CYP24A1 promoter to phorbol ester action in osteoblasts mirrors the lack of activation of MAP kinase pathways by TPA remains to be determined.

E. Glucocorticoid and Protein Kinase C D. Phorbol Ester and Protein Kinase C The stimulation of CYP24A1 expression by phorbol ester treatment and the involvement of PKC activity has been of interest for many years [76]. The synergistic action of phorbol ester and 1,25(OH)2D appears to be cell type specific. In mouse renal tubules, the phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA), also referred to as PMA, increases CYP24A1 activity in a PKC-dependent fashion [134]. TPA also markedly increased the level of CYP24A1 mRNA in 1,25(OH)2Dtreated rat renal cells [15,135] and in rat intestinal epithelial cells [136,137]. However, synergy is not observed in osteoblast cells. TPA does not affect either CYP24A1 mRNA levels or CYP24A1 promoter activity in the presence or absence of 1,25(OH)2D [94,137,138] in rat osteoblastic UMR-106 cells, although some TPA stimulation of CYP24A1 mRNA in these cells in the presence of 1,25(OH)2D has been reported at earlier times after treatment [96]. Synergy also is not observed in rat osteoblastic ROS17/2.8 cells transfected with rat –298 bp CYP24A1 promoter constructs (P Dwivedi and B May, unpublished). The molecular mechanism by which TPA stimulates 1,25(OH)2D-induced CYP24A1 expression has attracted interest but has remained elusive [76]. Since it is well known that TPA can stimulate the activity of the transcription factor AP-1 and more recently Ets proteins [139], studies have been undertaken to determine the possible roles of the EBS and AP-1 like sequences in the CYP24A1 promoter in kidney 293T cells (P Dwivedi, B Nutschey, and B May, submitted). TPA alone gives minimal induction of the CYP24A1 promoter, but in the presence of 1,25(OH)2D there is marked synergy (about 7.1-fold) and this was localized to the first –186 bp of promoter sequence. Within this region, mutagenesis individually of the AP-1 like sequence and EBS

Dexamethasone markedly enhances 1,25(OH)2D induction of CYP24A1 mRNA in UMR-106 cells and to a lesser extent in renal LLC-PK1 cells with the effect of dexamethasone being blocked by a PKC inhibitor [96]. In this study, dexamethasone markedly increased c-fos mRNA and protein but only slightly induced the abundance of c-jun mRNA in UMR-106 cells. It was proposed that dexamethasone enhances CYP24A1 mRNA expression in osteoblast cells via activation of AP-1 in a process that requires PKC activity. It would be relevant to determine if dexamethasone acts through the AP-1 like sequence identified in the CYP24A1 promoter.

F. Parathyroid Hormone In recent studies employing opossum kidney (OK) cells that express both endogenous VDR and PTH receptors, Armbrecht et al. [140] showed that PTH (acting through cAMP) moderately increases expression of a reporter construct containing −580 bp of rat CYP24A1 promoter. The mechanism by which this cAMP-mediated increase in transcription occurs is not known but may involve increased VDR [138]. As expected, 1,25(OH)2D also stimulated promoter expression. The apparent paradox that PTH stimulates renal CYP24A1 promoter expression, yet substantially lowers CYP24A1 activity in vivo, has been resolved [62,80,81]. It has been demonstrated that PTH markedly destabilizes the CYP24A1 mRNA, reducing the half-life by fourfold as determined in AOK-B50 cells [81]. The precise mechanism underlying this cAMP-dependent degradation is not known but could involve PKA-directed phosphorylation of a CYP24A1 mRNA binding protein whose loss permits degradation by a specific endonuclease (Fig. 5). Interestingly, while the 3′ untranslated region is a common target for regulating mRNA stability, the studies

100 of Zierold et al. [80] showed that neither the 3′ nor the 5′ untranslated regions of the CYP24A1 mRNA contain the instability element, suggesting that such a site is located in the coding region. Alteration of mRNA stability is considered a rapid response mechanism for modulating gene expression. Hence in the hypocalcemic state, PTH production by the parathyroid gland will result in destabilization of renal CYP24A1 mRNA with a rapid increase in the production of 1,25(OH)2D by the kidney through decreased removal of 25(OH)D and 1,25(OH)2D. In contrast to this, it is accepted that the major effect of PTH on renal CYP27B1 expression is through increased promoter activity [140,141].

G. Calcitonin As discussed earlier, recent data have shown that calcitonin substantially increases expression of a rat CYP24A1 –298 bp promoter construct in both transiently and stably transfected kidney HEK-293 cells expressing the calcitonin receptor. This induction is strongly dependent on the GC and CCAAT box sites as demonstrated in stably transfected HEK-293 cells (Fig. 9) in which calcitonin induces the transfected promoter 14-fold. However, this induction is reduced to 2.8-fold and 1.2-fold when the CCAAT box and GC boxes are mutated, respectively. Pharmacological inhibitor studies indicated the involvement of both PKA and PKC activities in calcitonin stimulation but not basal expression. Although calcitonin is known to activate ERK1 and ERK2 in these cells [142], these ERK activities do not participate in calcitonin induction, since the ERK inhibitor PD98059 and the dominant negative clone ERK1K71R did not affect calcitonin induction. The precise roles of PKA and PKC remain to be determined, but it seems likely that they act through Sp1 and NF-Y (Fig. 5). On the basis of this new data, it is tempting to suggest that calcitonin in vivo contributes to the control of circulating calcium levels through regulating 1,25(OH)2D. Such an action of calcitonin could be important in the hypercalcemic state when a transient increase in serum calcium triggers the production of calcitonin. In this regard, rats consuming a high-calcium diet markedly increase renal CYP24A1 and VDR mRNA levels, although the hormone regulation is unclear [10].

H. Interferon (INF)-Gamma Increased production of 1,25(OH)2D by INF-gamma activated monocytes/macrophages is associated with a marked inhibition of CYP24A1 up-regulation

JOHN OMDAHL AND BRIAN MAY

by 1,25(OH)2D. A novel mechanism for this antagonistic action by INF-gamma has been elucidated [101]. Basal expression of a human 24-hydroxylase promoter construct encompassing the two VDREs was not affected by INF-gamma, but 1,25(OH)2D-stimulated expression was inhibited. These workers established a sequence of events where, following the binding of INF-gamma to its receptor on the cell surface, the transcription factor Stat1 is activated by phosphorylation and in its homodimeric phosphorylated form enters the nucleus and binds liganded VDR (Fig. 5). This interaction abrogates the binding of liganded VDR/RXR to the VDREs in the CYP24A1 promoter, thereby suppressing 1,25(OH)2D activation. It is interesting that INFgamma stimulated CYP27B1 gene expression is not inhibited by high levels of 1,25(OH)2D in these cells [99], although studies suggesting a negative regulation of CYP27B1 gene expression by 1,25(OH)2D have been reported in the kidney [143]. Perhaps a genomic-acting inhibitory mechanism exists only in kidney cells.

Acknowledgments The authors recognize the expert assistance of Stephanie Finlayson, Andrezej Pastuszyn, Rita Serda, and Andrew Annolara, who greatly facilitated the preparation and composition of this chapter. Also available were discussions with Howard Morris, Prem Dwivedi, and Satya Reddy on subject matter contained in this chapter.

References 1. Dwivedi PP, Hii CS, Ferrante A, et al. 2002 Role of MAP kinases in the 1,25-dihydroxyvitamin D3-induced transactivation of the rat cytochrome P450C24 (CYP24) promoter. Specific function for Erk1/Erk2 and Erk5. J Biol Chem 277:29643–29653. 2. Nemere I, Pietras RJ, Blackmore PF 2003 Membrane receptors for steroid hormones: Signal transduction and physiological significance. J Cell Biochem 88:438–445. 3. Farach-Carson MC, Nemere I 2003 Membrane receptors for vitamin D steroid hormones: potential new drug targets. Curr Drug Targets 4:67–76. 4. Pike JW 1991 Vitamin D3 receptors: structure and function in transcription. Annu Rev Nutr 11:189–216. 5. Hannah SS, Norman AW 1994 1 Alpha,25(OH)2 vitamin D3regulated expression of the eukaryotic genome. Nutr. Rev 52:376–382. 6. Matkovits T, Christakos S 1995 Variable in vivo regulation of rat vitamin D-dependent genes (osteopontin, Ca,Mg-adenosine triphosphatase, and 25-hydroxyvitamin D3 24-hydroxylase): implications for differing mechanisms of regulation and involvement of multiple factors. Endocrinology 136: 3971–3982.

101

CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase

7. Armbrecht HJ, Boltz MA 1991 Expression of 25-hydroxyvitamin D 24-hydroxylase cytochrome P450 in kidney and intestine. Effect of 1,25-dihydroxyvitamin D and age. FEBS Lett 292:17–20. 8. Shinki T, Jin CH, Nishimura A, et al. 1992 Parathyroid hormone inhibits 25-hydroxyvitamin D3-24-hydroxylase mRNA expression stimulated by 1 alpha,25-dihydroxyvitamin D3 in rat kidney but not in intestine. J Biol Chem 267:13757–13762. 9. Zehnder D, Bland R, Williams MC, et al. 2001 Extrarenal expression of 25-hydroxyvitamin D(3)-1 alpha-hydroxylase. J Clin Endocrinol Metab 86:888–894. 10. Anderson PH, O’Loughlin PD, May BK, Morris HA 2003 Quantification of mRNA for the vitamin D metabolising enzymes CYP27B1 and CYP24 and VDR in kidney using real-time reverse transcriptase-polymerase chain reaction. J Mol Endocrinol (in press). 11. Pedersen JI, Hagenfeldt Y, Bjorkhem I 1998 Assay and properties of 25-hydroxyvitamin D3 23-hydroxylase. Evidence that 23,25-dihydroxyvitamin D3 is a major metabolite in 1,25-dihydroxyvitamin D3-treated or fasted guinea pigs. Biochem J 250:527–532. 12. Kumar R, Schaefer J, Grande JP, Roche PC 1994 Immunolocalization of calcitriol receptor, 24-hydroxylase cytochrome P-450, and calbindin D28k in human kidney. Am J Physiol 266:F477–F485. 13. Iwata K, Yamamoto A, Satoh S, et al. 1995 Quantitative immunoelectron microscopic analysis of the localization and induction of 25-hydroxyvitamin D3 24-hydroxylase in rat kidney. J Histochem Cytochem 43:255–262. 14. Henry HL, Dutta C, Cunningham N, et al. 1992 The cellular and molecular regulation of 1,25(OH)2D3 production. J Steroid Biochem Mol Biol 41:401–407. 15. Chen ML, Boltz MA, Armbrecht HJ 1993 Effects of 1,25dihydroxyvitamin D3 and phorbol ester on 25-hydroxyvitamin D3 24-hydroxylase cytochrome P450 messenger ribonucleic acid levels in primary cultures of rat renal cells. Endocrinology 132:1782–1788. 16. Matsumoto T, Kawanobe Y, Ogata E 1985 Regulation of 24,25-dihydroxyvitamin D-3 production by 1,25-dihydroxyvitamin D-3 and synthetic human parathyroid hormone fragment 1-34 in a cloned monkey kidney cell line (JTC-12). Biochim Biophys Acta 845:358–365. 17. Tomon M, Tenenhouse HS, Jones G 1990 Expression of 25-hydroxyvitamin D3-24-hydroxylase activity in Caco-2 cells. An in vitro model of intestinal vitamin D catabolism. Endocrinology 126:2868–2875. 18. Kane KF, Langman MJ, Williams GR 1996 Antiproliferative responses to two human colon cancer cell lines to vitamin D3 are differently modified by 9-cis-retinoic acid. Cancer Res 56:623–632. 19. Inaba M, Burgos-Trinidad M, DeLuca HF 1991 Characteristics of the 25-hydroxyvitamin D3- and 1,25-dihydroxyvitamin D3-24-hydroxylase(s) from HL-60 cells. Arch Biochem Biophys 284:257–263. 20. Skowronski RJ, Peehl DM, Feldman D 1995 Actions of vitamin D3, analogs on human prostate cancer cell lines: comparison with 1,25-dihydroxyvitamin D3. Endocrinology 136:20–26. 21. Makin G, Lohnes D, Byford V, Ray R, Jones G 1989 Target cell metabolism of 1,25-dihydroxyvitamin D3 to calcitroic acid. Evidence for a pathway in kidney and bone involving 24-oxidation. Biochem J 262:173–180. 22. Eil C, Liberman UA, Marx SJ 1986 The molecular basis for resistance to 1,25-dihydroxyvitamin D: studies in cells

23.

24. 25. 26.

27. 28. 29.

30.

31.

32.

33. 34.

35.

36.

37.

38. 39.

cultured from patients with hereditary hypocalcemic 1,25(OH)2D3-resistant rickets. Adv Exp Med Biol 196: 407–422. Morgan JW, Reddy GS, Uskokovic MR, et al. 1994 Functional block for 1 alpha,25-dihydroxyvitamin D3-mediated gene regulation in human B lymphocytes. J Biol Chem 269:13437–13443. Naveilhan P, Neveu I, Baudet C, et al. 1993 Expression of 25(OH) vitamin D3 24-hydroxylase gene in glial cells. NeuroReport 5:255–257. Chen ML, Heinrich G, Ohyama YI, et al. 1994 Expression of 25-hydroxyvitamin D3-24-hydroxylase mRNA in cultured human keratinocytes. Proc Soc Exp Biol Med 207:57–61. Holick MF, Schnoes HK, DeLuca HF, et al. 1972 Isolation and identification of 24,25-dihydroxycholecalciferol, a metabolite of vitamin D made in the kidney. Biochemistry 11:4251–4255. Holick MF, Kleiner-Bossaller A, Schnoes HK, et al. 1973 1,24,25-Trihydroxyvitamin D3. A metabolite of vitamin D3 effective on intestine. J Biol Chem 248:6691–6696. Beckman MJ, Tadikonda P, Werner E, et al. 1996 Human 25-hydroxyvitamin D3-24-hydroxylase, a multicatalytic enzyme. Biochemistry 35:8465–8472. Ohyama Y, Okuda K 1991 Isolation and characterization of a cytochrome P-450 from rat kidney mitochondria that catalyzes the 24-hydroxylation of 25-hydroxyvitamin D3. J Biol Chem 266:8690–8695. Lund EG, Guileyardo JM, Russell DW 1999 cDNA cloning of cholesterol 24-hydroxylase, a mediator of cholesterol homeostasis in the brain. Proc Natl Acad Sci USA 96: 7238–7243. Gray RW, Omdahl JL, Ghazarian JG, Horst RL 1990 Induction of 25-OH-vitamin D3 24- and 23-hydroxylase activities in partially purified renal extracts from pigs given exogenous 1,25-(OH)2D3. Steroids 55:395–398. Pikuleva IA, Cao C, Waterman MR 1999 An additional electrostatic interaction between adrenodoxin and P450c27 (CYP27A1) results in tighter binding than between adrenodoxin and p450scc (CYP11A1). J Biol Chem 274:2045–2052. Reddy GS, Tserng KY 1989 Calcitroic acid, end product of renal metabolism of 1,25-dihydroxyvitamin D3 through C-24 oxidation pathway. Biochemistry 28:1763–1769. Esvelt RP, Schnoes HK, DeLuca HF 1979 Isolation and characterization of 1 alpha-hydroxy-23-carboxytetranorvitamin D: a major metabolite of 1,25-dihydroxyvitamin D3. Biochemistry 18:3977–3983. Horst RL, Reinhardt TA, Ramberg CF, Koszewski NJ, Napoli JL 1986 24-Hydroxylation of 1,25-dihydroxyergocalciferol. An unambiguous deactivation process. J Biol Chem 261:9250–9256. Akiyoshi-Shibata M, Sakaki T, Ohyama Y, et al. 1994 Further oxidation of hydroxycalcidiol by calcidiol 24-hydroxylase. A study with the mature enzyme expressed in Escherichia coli. Eur J Biochem 224:335–343. Sakaki T, Sawada N, Nonaka Y, Ohyama Y, Inouye K 1999 Metabolic studies using recombinant Escherichia coli cells producing rat mitochondrial CYP24. CYP24 can convert 1α,25-dihydroxyvitamin D3 to calcitroic acid. Eur J Biochem 262:43–48. Sakaki T, Sawada N, Komai K, et al. 2000 Dual metabolic pathway of 25-hydroxyvitamin D3 catalyzed by human CYP24. Eur J Biochem 267:6158–6165. Peterson JA, Graham SE 1998 A close family resemblance: The importance of structure in understanding cytochromes P450. Structure 6:1079–1085.

102 40. Poulos TL, Finzel BC, Howard AJ 1987 High-resolution crystal structure of cytochrome P450cam. J Mol Biol 195:687–700. 41. Ravichandran KG, Boddupalli SS, Hasermann CA, Peterson JA, Deisenhofer J 1993 Crystal structure of hemoprotein domain of P450BM-3, a prototype for microsomal P450’s. Science 261:731–736. 42. Hasemann CA, Ravichandran KG, Peterson JA, Deisenhofer J 1994 Crystal structure and refinement of cytochrome P450terp at 2.3 A resolution. J Mol Biol 236:1169–1185. 43. Brown AJ, Finch JL, Lopez-Hilker S, et al. 1990 New active analogues of vitamin D with low calcemic activity. Kidney Int Suppl 29:S22–S27. 44. Wester MR, Johnson EF, Marques-Soares C, et al. 2003 Structure of a substrate complex of mammalian cytochrome P450 2C5 at 2.3 Å resolution: Evidence for multiple substrate binding modes. Biochemistry 42:6370–6379. 45. Schuster I, Egger H, Astecker N, et al. 2001 Selective inhibitors of CYP24: Mechanistic tools to explore vitamin D metabolism in human keratinocytes. Steroids 66:451–462. 46. Schuster I, Egger H, Bikle D, et al. 2001 Selective inhibition of vitamin D hydroxylases in human keratinocytes. Steroids 66:409–422. 47. Peehl DM, Seto E, Feldman D 2001 Rationale for combination ketoconazole/vitamin D treatment of prostate cancer. Urology 58:123–126. 48. Bayanov AA, Brunt AR 1999 Role of hypoxia and constitutionally different resistance to hypoxia/stress as the determiners of individual profile of cytochrome P450 isozyme activity. Gen Pharmacol 33:355–361. 49. Tenenhouse HS, Yip A, Jones G 1988 Increased renal catabolism of 1,25-dihydroxyvitamin D3 in murine X-linked hypophosphatemic rickets. J Clin Invest 81:461–465. 50. Shinki T, Jin CH, Nishimura A, et al. 1992 Parathyroid hormone inhibits 25-hydroxyvitamin D3-24-hydroxylase mRNA expression stimulated by 1 alpha,25-dihydroxyvitamin D3 in rat kidney but not in intestine. J Biol Chem 267:13757–13762. 51. Burgos-Trinidad M, DeLuca HF 1991 Kinetic properties of 25-hydroxyvitamin D- and 1,25-dihydroxyvitamin D24-hydroxylase from chick kidney. Biochim Biophys Acta 1078:226–230. 52. Taniguchi T, Eto TA, Shiotsuki H, et al. 2001 Newly established assay method for 25-hydroxyvitamin D3 24hydroxylase revealed much lower Km for 25-hydroxyvitamin D3 than for 1α,25-dihydroxyvitamin D3. J Bone Miner Res 16:57–62. 53. Pikuleva IA, Bjorkhem I, Waterman MR 1997 Expression, purification, and enzymatic properties of recombinant human cytochrome P450c27 (CYP27). Arch Biochem Biophys 343:123–130. 54. Pikuleva IA, Puchkaev A, Bjorkhem I 2001 Putative helix F contributes to regioselectivity of hydroxylation in mitochondrial cytochrome P450 27A1. Biochemistry 40: 7621–7629. 55. Schenkman JB, Remmer H, Estabrook RW 1967 Spectral studies of drug interaction with hepatic microsomal cytochrome. Mol Pharmacol 3:113–123. 56. Horst RL, Omdahl JA, Reddy S 2003 Rat cytochrome P450C24 (CYP24) does not metabolize 1,25-dihydroxyvitamin D2 to calcitroic acid. J Cell Biochem 88:282–285. 57. Omdahl JL, Bobrovnikova EV, Annalora A, Chen P, Serda R 2003 Expression, structure–function, and molecular modeling of vitamin D P450s. J Cell Biochem 88:356–362. 58. Ohyama Y, Noshiro M, Okuda K 1991 Cloning and expression of cDNA encoding 25-hydroxyvitamin D3 24-hydroxylase. FEBS Lett 278:195–198.

JOHN OMDAHL AND BRIAN MAY

59. Chen KS, Prahl JM, DeLuca HF 1993 Isolation and expression of human 1,25-dihydroxyvitamin D3 24-hydroxylase cDNA. Proc Natl Acad Sci USA 90:4543–4547. 60. Jehan F, Ismail R, Hanson K, DeLuca HF 1998 Cloning and expression of the chicken 25-hydroxyvitamin D3 24-hydroxylase cDNA. Biochim Biophys Acta 1395:259–265. 61. Akeno N, Saikatsu S, Kawane T, Horiuchi N 1997 Mouse vitamin D-24-hydroxylase: molecular cloning, tissue distribution, and transcriptional regulation by 1α,25-dihydroxyvitamin D3. Endocrinology 138:2233–2240. 62. Zierold C, Reinholz GG, Mings JA, Prahl JM, DeLuca HF 2001 Regulation of the porcine 1,25-dihydroxyvitamin D3-24-hydroxylase (CYP24) by 1,25-dihydroxyvitamin D3 and parathyroid hormone in AOK-B50 cells. Arch Biochem Biophys 381:323–327. 63. Hahn CN, Baker E, Laslo P, et al. 1993 Localization of the human vitamin D 24-hydroxylase gene (CYP24) to chromosome 20q13.2—>q13.3. Cytogenet Cell Genet 62: 192–193. 64. Ohyama Y, Noshiro M, Eggertsen G, et al. 1993 Structural characterization of the gene encoding rat 25-hydroxyvitamin D3 24-hydroxylase. Biochemistry 32:76–82. 65. St Arnaud R, Arabian A, Glorieux FH 1996 Abnormal bone development in mice deficient for the vitamin D 24-hydroxylase gene. J Bone Miner Res 11:S126. 66. St Arnaud R 1999 Targeted inactivation of vitamin D hydroxylases in mice. Bone 25:127–129. 67. Boyan BD, Sylvia VL, Dean DD, Schwartz Z 2001 24, 25-(OH)2D3 regulates cartilage and bone via autocrine and endocrine mechanisms. Steroids 66:363–374. 68. Schwartz Z, Ehland H, Sylvia VL, et al. 2002 1α,25-dihydroxyvitamin D3 and 24R,25-dihydroxyvitamin D3 modulate growth plate chondrocyte physiology via protein kinase Cdependent phosphorylation of extracellular signal-regulated kinase 1/2 mitogen-activated protein kinase. Endocrinology 143:2775–2786. 69. St Arnaud R, Arabian A, Travers R, et al. 2000 Deficient mineralization of intramembranous bone in vitamin D-24hydroxylase-ablated mice is due to elevated 1,25-dihydroxyvitamin D and not to the absence of 24,25-dihydroxyvitamin D. Endocrinology 141:2658–2666. 70. Haussler MR, Whitfield GK, Haussler CA, et al. 1998 The nuclear vitamin D receptor: biological and molecular regulatory properties revealed. J Bone Miner Res 13:325–349. 71. Issa LL, Leong GM, Eisman JA 1998 Molecular mechanism of vitamin D receptor action. Inflamm Res 47:451–475. 72. Kasuga H, Hosogane N, Matsuoka K, et al. 2002 Characterization of transgenic rats constitutively expressing vitamin D-24-hydroxylase gene. Biochem Biophys Res Commun 297:1332–1338. 73. Hosogane N, Shinki T, Kasuga H, et al. 2003 Mechanisms for the reduction of 24,25-dihydroxyvitamin D3 levels and bone mass in 24-hydroxylase transgenic rats. FASEB J 17:737–739. 74. Panda DK, Miao D, Tremblay ML, et al. 2001 Targeted ablation of the 25-hydroxyvitamin D 1α-hydroxylase enzyme: Evidence for skeletal, reproductive, and immune dysfunction. Proc Natl Acad Sci USA 98:7498–7503. 75. Dardenne O, Prud’homme J, Arabian A, Glorieux FH, St Arnaud R 2001 Targeted inactivation of the 25-hydroxyvitamin D(3)-1(α)-hydroxylase gene (CYP27B1) creates an animal model of pseudovitamin D-deficiency rickets. Endocrinology 142:3135–3141. 76. Omdahl JL, May BK 1997 The 25-hydroxyvitamin D 24-hydroxylase. In: Feldman D, Glorieux F, Pike J (eds) Vitamin D. Academic Press, San Diego pp. 69–85.

103

CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase

77. Omdahl JL, Morris HA, May BK 2002 Hydroxylase enzymes of the vitamin D pathway: Expression, function, and regulation. Annu Rev Nutr 22:139–166. 78. Sutton AL, MacDonald PN 2003 Vitamin D: More than a “bone-a-fide” hormone. Mol Endocrinol 17:777–791. 79. Yang W, Friedman PA, Kumar R, et al. 1999 Expression of 25(OH)D3 24-hydroxylase in distal nephron: coordinate regulation by 1,25(OH)2D3 and cAMP or PTH. Am J Physiol 276:E793–E805. 80. Zierold C, Mings JA, DeLuca HF 2001 Parathyroid hormone regulates 25-hydroxyvitamin D3-24-hydroxylase mRNA by altering its stability. Proc Natl Acad Sci USA 98: 13572–13576. 81. Zierold C, Mings JA, DeLuca HF 2003 Regulation of 25hydroxyvitamin D3-24-hydroxylase mRNA by 1,25-dihydroxyvitamin D3 and parathyroid hormone. J Cell Biochem 88:234–237. 82. Pondel M 2000 Calcitonin and calcitonin receptors: bone and beyond. Int J Exp Pathol 81:405–422. 83. Brown EM, MacLeod RJ 2001 Extracellular calcium sensing and extracellular calcium signaling. Physiol Rev 81:239–297. 84. Inzerillo AM, Zaidi M, Huang CL 2002 Calcitonin: The other thyroid hormone. Thyroid 12:791–798. 85. Jones G, Strugnell SA, DeLuca HF 1998 Current understanding of the molecular actions of vitamin D. Physiol Rev 78:1193–1231. 86. Yoshida N, Yoshida T, Nakamura A, et al. 1999 Calcitonin induces 25-hydroxyvitamin D3 1α-hydroxylase mRNA expression via protein kinase C pathway in LLC-PK1 cells. J Am Soc Nephrol 10:2474–2479. 87. Murayama A, Takeyama K, Kitanaka S, et al. 1998 The promoter of the human 25-hydroxyvitamin D3 1α-hydroxylase gene confers positive and negative responsiveness to PTH, calcitonin, and 1α,25(OH)2D3. Biochem Biophys Res Commun 249:11–16. 88. Bland R, Zehnder D, Hughes SV, et al. 2001 Regulation of vitamin D-1α-hydroxylase in a human cortical collecting duct cell line. Kidney Int 60:1277–1286. 89. Shinki T, Ueno Y, DeLuca HF, Suda T 1999 Calcitonin is a major regulator for the expression of renal 25-hydroxyvitamin D3-1α-hydroxylase gene in normocalcemic rats. Proc Natl Acad Sci USA 96:8253–8258. 90. Henry HL 2001 The 25(OH)D3/1α,25(OH)2D3-24(R)-hydroxylase: A catabolic or biosynthetic enzyme? Steroids 66:391–398. 91. Nishimura A, Shinki T, Jin CH, et al. 1994 Regulation of messenger ribonucleic acid expression of 1α,25-dihydroxyvitamin D3-24-hydroxylase in rat osteoblasts. Endocrinology 134:1794–1799. 92. Ylikomi T, Laaksi I, Lou YR, et al. 2002 Antiproliferative action of vitamin D. Vitam Horm 64:357–406. 93. Gurlek A, Kumar R 2001 Regulation of osteoblast growth by interactions between transforming growth factor-beta and 1α,25-dihydroxyvitamin D3. Crit Rev Eukaryot Gene Expr 11:299–317. 94. Armbrecht HJ, Hodam TL 1994 Parathyroid hormone and 1,25-dihydroxyvitamin D synergistically induce the 1,25dihydroxyvitamin D-24-hydroxylase in rat UMR-106 osteoblast-like cells. Biochem Biophys Res Commun 205:674–679. 95. Armbrecht HJ, Hodam TL, Boltz MA, et al. 1998 Induction of the vitamin D 24-hydroxylase (CYP24) by 1,25-dihydroxyvitamin D3 is regulated by parathyroid hormone in UMR106 osteoblastic cells [see comments]. Endocrinology 139: 3375–3381. 96. Kurahashi I, Matsunuma A, Kawane T, Abe M, Horiuchi N 2002 Dexamethasone enhances vitamin D-24-hydroxylase

97. 98.

99. 100.

101.

102. 103. 104. 105.

106.

107. 108.

109.

110.

111.

112.

expression in osteoblastic (UMR-106) and renal (LLC-PK1) cells treated with 1α,25-dihydroxyvitamin D3. Endocrine 17: 109–118. Griffin MD, Xing N, Kumar R 2003 Vitamin D and its analogs as regulators of immune activation and antigen presentation. Annu Rev Nutr 23:117–145. Dusso AS, Kamimura S, Gallieni M, et al. 1997 γ-Interferoninduced resistance to 1,25-(OH)2 D3 in human monocytes and macrophages: A mechanism for the hypercalcemia of various granulomatoses. J Clin Endocrinol Metab 82:2222–2232. Monkawa T, Yoshida T, Hayashi M, Saruta T 2000 Identification of 25-hydroxyvitamin D3 1α-hydroxylase gene expression in macrophages. Kidney Int 58:559–568. Overbergh L, Decallonne B, Valckx D, et al. 2000 Identification and immune regulation of 25-hydroxyvitamin D-1-α-hydroxylase in murine macrophages. Clin Exp Immunol 120:139–146. Vidal M, Ramana CV, Dusso AS 2002 Stat1-vitamin D receptor interactions antagonize 1,25-dihydroxyvitamin D transcriptional activity and enhance stat1-mediated transcription. Mol Cell Biol 22:2777–2787. Bikle DD, Ng D, Tu C, Oda Y, Xie Z 2001 Calcium- and vitamin D-regulated keratinocyte differentiation. Mol Cell Endocrinol 177:161–171. Bikle DD, Tu CL, Xie Z, Oda Y 2003 Vitamin D regulated keratinocyte differentiation: Role of coactivators. J Cell Biochem 88:290–295. Xie Z, Munson SJ, Huang N, et al. 2002 The mechanism of 1,25-dihydroxyvitamin D3 autoregulation in keratinocytes. J Biol Chem 277:36987–36990. Murayama A, Takeyama K, Kitanaka S, et al. 1999 Positive and negative regulations of the renal 25-hydroxyvitamin D3 1α-hydroxylase gene by parathyroid hormone, calcitonin, and 1α,25(OH)2D3 in intact animals. Endocrinology 140:2224–2231. Murayama A, Kim M, Unno K, Takeyama K, Kato S 2001 A novel mechanism of the vitamin D dependent transcriptional repression through the human 25-hydroxyvitamin D3 1α-hydroxylase in nVDRE. J Bone Miner Res 16:S229. Albertson DG, Ylstra B, Segraves R, et al. 2000 Quantitative mapping of amplicon structure by array CGH identifies CYP24 as a candidate oncogene. Nat Genet 25:144–146. Bareis P, Bises G, Bischof MG, Cross HS, Peterlik M 2001 25-hydroxy-vitamin D metabolism in human colon cancer cells during tumor progression. Biochem Biophys Res Commun 285:1012–1017. Miller GJ, Stapleton GE, Hedlund TE, Moffat KA 1995 Vitamin D receptor expression, 24-hydroxylase activity, and inhibition of growth by 1α,25-dihydroxyvitamin D3 in seven human prostatic carcinoma cell lines. Clin Cancer Res 1:997–1003. Ly LH, Zhao XY, Holloway L, Feldman D 1999 Liarozole acts synergistically with 1α,25-dihydroxyvitamin D3 to inhibit growth of DU 145 human prostate cancer cells by blocking 24-hydroxylase activity. Endocrinology 140: 2071–2076. Kallay E, Adlercreutz H, Farhan H, et al. 2002 Phytoestrogens regulate vitamin D metabolism in the mouse colon: Relevance for colon tumor prevention and therapy. J Nutr 132: 3490S–3493S. Farhan H, Wahala K, Adlercreutz H, Cross HS 2002 Isoflavonoids inhibit catabolism of vitamin D in prostate cancer cells. J Chromatogr B Anal Technol Biomed Life Sci 777:261–268.

104 113. Ohyama Y, Ozono K, Uchida M, et al. 1994 Identification of a vitamin D-responsive element in the 5′-flanking region of the rat 25-hydroxyvitamin D3 24-hydroxylase gene. J Biol Chem 269:10545–10550. 114. Hahn CN, Kerry DM, Omdahl JL, May BK 1994 Identification of a vitamin D responsive element in the promoter of the rat cytochrome P450(24) gene. Nucleic Acids Res 22:2410–2416. 115. Zierold C, Darwish HM, DeLuca HF 1994 Identification of a vitamin D-response element in the rat calcidiol (25-hydroxyvitamin D3) 24-hydroxylase gene. Proc Natl Acad Sci USA 91:900–902. 116. Zierold C, Darwish HM, DeLuca HF 1995 Two vitamin D response elements function in the rat 1,25-dihydroxyvitamin D 24-hydroxylase promoter. J Biol Chem 270:1675–1678. 117. Kerry DM, Dwivedi PP, Hahn CN, et al. 1996 Transcriptional synergism between vitamin D-responsive elements in the rat 25-hydroxyvitamin D3 24-hydroxylase (CYP24) promoter. J Biol Chem 271:29715–29721. 118. Ohyama Y, Ozono K, Uchida M, et al. 1996 Functional assessment of two vitamin D-responsive elements in the rat 25-hydroxyvitamin D3 24-hydroxylase gene. J Biol Chem 271:30381–30385. 119. Dwivedi PP, Omdahl JL, Kola I, Hume DA, May BK 2000 Regulation of rat cytochrome P450C24 (CYP24) gene expression. Evidence for functional cooperation of rasactivated Ets transcription factors with the vitamin D receptor in 1,25-dihydroxyvitamin D3-mediated induction. J Biol Chem 275:47–55. 120. Dwivedi PP, Muscat GE, Bailey PJ, Omdahl JL, May BK 1998 Repression of basal transcription by vitamin D receptor: evidence for interaction of unliganded vitamin D receptor with two receptor interaction domains in RIP13∆1. J Mol Endocrinol. 20:327–335. 121. Barry JB, Leong GM, Church WB, et al. 2003 Interactions of SKIP/NCoA-62, TFIIB, and retinoid X receptor with vitamin D receptor helix H10 residues. J Biol Chem 278:8224–8228. 122. Mantovani R 1999 The molecular biology of the CCAATbinding factor NF-Y. Gene 239:15–27. 123. Zierold C, Mings JA, Prahl JM, Reinholz GG, DeLuca HF 2002 Protein synthesis is required for optimal induction of 25-hydroxyvitamin D3-24-hydroxylase, osteocalcin, and osteopontin mRNA by 1,25-dihydroxyvitamin D3. Arch Biochem Biophys 404:18–24. 124. Falkenstein E, Tillmann HC, Christ M, Feuring M, Wehling M 2000 Multiple actions of steroid hormones—a focus on rapid, nongenomic effects. Pharmacol Rev 52:513–556. 125. Solomon C, White JH, Kremer R 1999 Mitogen-activated protein kinase inhibits 1,25-dihydroxyvitamin D3-dependent signal transduction by phosphorylating human retinoid X receptor α. J Clin Invest 103:1729–1735. 126. Razandi M, Alton G, Pedram A, et al. 2003 Identification of a structural determinant necessary for the localization and function of estrogen receptor α at the plasma membrane. Mol Cell Biol 23:1633–1646. 127. Razandi M, Oh P, Pedram A, Schnitzer J, Levin ER 2002 ERs associate with and regulate the production of caveolin: implications for signaling and cellular actions. Mol Endocrinol 16:100–115. 128. Buitrago C, Gonzalez PV, De Boland AR 2002 Nongenomic action of 1α,25(OH)2-vitamin D3. Activation of muscle cell PLC gamma through the tyrosine kinase c-Src and PtdIns 3-kinase. Eur J BioChem 269:2506–2515.

JOHN OMDAHL AND BRIAN MAY

129. Erben RG, Soegiarto DW, Weber K, et al. 2002 Deletion of deoxyribonucleic acid binding domain of the vitamin D receptor abrogates genomic and nongenomic functions of vitamin D. Mol Endocrinol. 16:1524–1537. 130. Hammes SR 2003 The further redefining of steroid-mediated signaling. Proc Natl Acad Sci USA 100:2168–2170. 131. Wali RK, Kong J, Demay MB, et al. 2001 The Vitamin D receptor is not required for the rapid activation of PKC and rise in intracellular calcium induced by 1,25-dihydroxyvitamin D3 in mouse osteoblasts. J Bone Miner Res 16:S229. 132. Wali RK, Kong J, Sitrin MD, Bissonnette M, Li YC 2003 Vitamin D receptor is not required for the rapid actions of 1,25-dihydroxyvitamin D3 to increase intracellular calcium and activate protein kinase C in mouse osteoblasts. J Cell BioChem 88:794–801. 133. Norman AW, Henry HL, Bishop JE, et al. 2001 Different shapes of the steroid hormone 1α,25(OH)2-vitamin D3 act as agonists for two different receptors in the vitamin D endocrine system to mediate genomic and rapid responses. Steroids 66:147–158. 134. Mandla S, Boneh A, Tenenhouse HS 1990 Evidence for protein kinase C involvement in the regulation of renal 25-hydroxyvitamin D3–24-hydroxylase. Endocrinology 127: 2639–2647. 135. Armbrecht HJ, Chen ML, Hodam TL, Boltz MA 1997 Induction of 24-hydroxylase cytochrome P450 mRNA by 1,25-dihydroxyvitamin D and phorbol esters in normal rat kidney (NRK-52E) cells. J Endocrinol 153:199–205. 136. Koyama H, Inaba M, Nishizawa Y, Ohno S, Morii H 1994 Protein kinase C is involved in 24-hydroxylase gene expression induced by 1,25(OH)2D3 in rat intestinal epithelial cells. J Cell BioChem 55:230–240. 137. Armbrecht HJ, Boltz MA, Hodam TL, Kumar VB 2001 Differential responsiveness of intestinal epithelial cells to 1,25-dihydroxyvitamin D3—role of protein kinase C. J Endocrinol 169:145–151. 138. Yang W, Hyllner SJ, Christakos S 2001 Interrelationship between signal transduction pathways and 1,25(OH)2D3 in UMR106 osteoblastic cells. Am J Physiol Endocrinol Metab 281:E162-E170. 139. Reddy SP, Vuong H, Adiseshaiah P 2003 Interplay between proximal and distal promoter elements is required for squamous differentiation marker induction in the bronchial epithelium: role for ESE-1, Sp1, and AP-1 proteins. J Biol Chem 278:21378–21387. 140. Armbrecht HJ, Hodam TL, Boltz MA 2003 Hormonal regulation of 25-hydroxyvitamin D3-1α-hydroxylase and 24-hydroxylase gene transcription in opossum kidney cells. Arch Biochem Biophys 409:298–304. 141. Gao XH, Dwivedi PP, Choe S, et al. 2002 Basal and parathyroid hormone induced expression of the human 25-hydroxyvitamin D 1α-hydroxylase gene promoter in kidney AOK-B50 cells: role of Sp1, Ets and CCAAT box protein binding sites. Int J Biochem Cell Biol 34: 921–930. 142. Raggatt LJ, Evdokiou A, Findlay DM 2000 Sustained activation of Erk1/2 MAPK and cell growth suppression by the insert-negative, but not the insert-positive isoform of the human calcitonin receptor. J Endocrinol 167:93–105. 143. Takeyama K, Kitanaka S, Sato T, et al. 1997 25-Hydroxyvitamin D3 1α-hydroxylase and vitamin D synthesis. Science 277:1827–1830.

CHAPTER 7

Mutant Mouse Models of Vitamin D Metabolic Enzymes RENÉ ST-ARNAUD

Genetics Unit, Shriners Hospital for Children, and McGill University, Montreal (Quebec), Canada

I. Introduction II. Hepatic 25-Hydroxylation and cyp27A1 Knockout III. 24-Hydroxylation and cyp24A1 Knockout

I. INTRODUCTION A. Note on Cytochrome P450 Nomenclature Since 1987, efforts have been made to encourage scientists worldwide to avoid “home-made” designations for naming cytochrome P450s in order to avoid confusing the nomenclature system and the scientific literature. This chapter will follow the guidelines provided in the relevant publications [1,2] and the recently updated names listed on the Cytochrome P450 Homepage Web site (http://drnelson.utmem.edu/mouse.master.table.html). Human genes will be italicized in capital letters, while mouse genes will be in lower case italics. Proteins are referred to in capital, nonitalicized characters (Table I).

B. Vitamin D Metabolism and Mutant Mouse Models of Vitamin D Metabolic Enzymes Vitamin D, produced endogenously in the skin upon exposure to ultraviolet light (sunlight) or ingested in the diet must be metabolized twice to be activated and function as a key regulator of mineral ion homeostasis. Vitamin D, bound to the vitamin D binding protein, DBP, is transported to the liver where vitamin D 25-hydroxylases (CYP27A1 and/or CYP2R1) add a hydroxyl group on carbon 25 to produce 25-hydroxyvitamin D [25(OH)D] (see Chapter 4). The 25(OH)D metabolite also circulates in the bloodstream bound to DBP (see Chapter 8). It must be further hydroxylated in the kidney to gain hormonal bioactivity. Hydroxylation at position 1α by the enzyme 25-hydroxyvitamin D1α-hydroxylase (CYP27B1) converts 25(OH)D to 1α,25-dihydroxyvitamin D [1,25(OH)2D], the active, hormonal form of vitamin D that regulates mineral homeostasis, skeletal homeostasis, and cellular differentiation (see Chapter 5). Among several target genes, the 1,25(OH)2D hormone induces in target cells the VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

IV. 1α-Hydroxylation and cyp27B1 Knockout V. Summary and Perspectives References

expression of the gene encoding the key effector of its catabolic breakdown: 25-hydroxyvitamin D-24-hydroxylase (CYP24A1). This ensures attenuation of the 1,25(OH)2D biological signal inside target cells and helps regulate vitamin D homeostasis. All vitamin D hydroxylases characterized to date belong to the superfamily of cytochrome P450 enzymes. They are heme-containing, mixed-function oxidases that use molecular oxygen as a terminal electron acceptor. They require the accessory electron transfer proteins, ferredoxin and ferredoxin reductase, to accept reducing equivalents from nicotinamide adenine dinucleotide phosphate (NADPH) and stereospecifically hydroxylate vitamin D metabolites. Molecular genetics–based research has allowed dramatic progress in our understanding of vitamin D metabolism and function. Several mice strains, deficient for the vitamin D metabolic enzymes CYP27A1, CYP27B1, and CYP24A1, have been engineered. This chapter will review the phenotype of the mutant strains, highlight the key findings derived from the analysis of the strains, and discuss the future avenues for research.

II. HEPATIC 25-HYDROXYLATION AND cyp27A1 KNOCKOUT Until recently, only one cytochrome P450 molecule had been cloned and shown to hydroxylate vitamin D at position 25. This was the bifunctional CYP27A1 sterol 27-hydroxylase that derives its name for its ability to both 27-hydroxylate the side chains of cholesterolderived intermediates involved in bile acid biosynthesis and 25-hydroxylate vitamin D [3]. However, several pieces of experimental evidence argued for the existence of a second, physiologically relevant vitamin D 25-hydroxylase enzyme (reviewed in [4]). First, mice with a disrupted cyp27A1 gene have reduced bile acid Copyright © 2005, Elsevier, Inc. All rights reserved.

106

RENÉ ST-ARNAUD

TABLE I Cytochrome P450 Nomenclature Used Common name

Human gene symbol

Mouse gene symbol

Protein

CYP27A1 CYP2R1 CYP27B1 CYP24A1

cyp27A1 cyp2R1 cyp27B1 cyp24A1

CYP27A1 CYP2R1 CYP27B1 CYP24A1

Vitamin D-25-hydroxylase (mitochondrial) Vitamin D-25-hydroxylase (microsomal) 25-Hydroxyvitamin D-1α-hydroxylase 25-Hydroxyvitamin D-24-hydroxylase

synthesis, but normal vitamin D metabolite levels [5]. Second, patients with the inherited disease cerebrotendinous xanthomatosis, caused by mutations in CYP27A1 [6,7], exhibit normal vitamin D metabolism and no obvious 25(OH)D or 1,25(OH)2D deficiency [8]. Third, early studies using perfused rat liver revealed kinetics of vitamin D metabolism that supported two 25-hydroxylase activities: a high-affinity, low-capacity enzyme and a low-affinity, high-capacity form (now presumed to be CYP27A1) [9]. The subcellular localizations of these two enzymatic activities are thought to differ, CYP27A1 being a mitochondrial enzyme [10,11], while the high-affinity, low-capacity enzyme is associated with the endoplasmic reticulum (microsomes) [12]. Next, the promoter of CYP27A1 is regulated by bile acids, but not by vitamin D metabolites [13], a finding that is not consistent with the weak but demonstrated regulation of the hepatic 25-hydroxylase observed following vitamin D intake in previously vitamin D–depleted animals [14]. Last, cells transfected with an expression vector for CYP27A1 are not able to 25hydroxylate vitamin D2 [15]. The 25-hydroxy-D2 metabolite is synthesized in vivo, however, suggesting that an enzyme distinct from CYP27A1 is involved in its synthesis.

A. Phenotype of cyp27A1-Deficient Mice The CYP27A1 enzyme is important for bile acid biosynthesis and regulation of cholesterol homeostasis [10,11]. Mutations in the human gene cause cerebrotendinous xanthomatosis (CTX), a lipid storage disorder leading to premature atherosclerosis and progressive neurological dysfunction [6,7]. Mice deficient for cyp27A1 have been engineered [5]. Homozygous cyp27A1−/− animals have larger livers and larger adrenals [16], decreased synthesis and excretion of bile acids [5], and hypertriglyceridemia [16]. The increased formation of 25-hydroxylated bile alcohols and cholestanol observed in CTX patients was not observed in cyp27A1−/− mice, and no CTX-related pathological abnormalities were evident [5]. At present, it remains

unclear whether the cyp27A1-deficient mouse is a valid animal model for CTX. Interestingly, cyp27A1-null mice had normal serum concentrations of 1,25(OH)2D and slightly elevated levels of 25(OH)D in their blood [5]. These results strengthen the hypothesis that a distinct vitamin D 25-hydroxylase enzyme plays a physiological role in vitamin D metabolism.

B. Identification of Novel Vitamin D-25-Hydroxylases An expression-based screening strategy has been used to demonstrate that CYP2R1 has vitamin D 25-hydroxylase activity. The experimental strategy used a cDNA library from cyp27A1-null mice to avoid isolating clones encoding the sterol 27-hydroxylase (CYP27A1). Expression of CYP2R1 in transfected cells led to transcriptional activation through the vitamin D receptor when vitamin D2, vitamin D3, or 1α(OH)D3 was added to cells [17]. The results obtained strongly suggest that CYP2R1 is the microsomal vitamin D 25-hydroxylase. Mice with mutations in the cyp2R1 gene are being engineered. It will prove interesting to analyze vitamin D metabolism and circulating levels of 25(OH)D and 1,25(OH)2D in cyp2R1-null mice. Based on specific activity determination, it was estimated that the microsomal enzyme is responsible for about 30% of the total 25-hydroxylation activity, while the mitochondrial enzyme is responsible for the remaining 70% [18]. This suggests a redundancy of function that could lead to no detectable vitamin D–related phenotype in cyp2R1-deficient mice. Altering the vitamin D 25-hydroxylase activity in mice in vivo would then require crossing the cyp27A1-null and cyp2R1-deficient strains and measuring circulating vitamin D metabolites levels as a function of cyp27A1 and cyp2R1 gene dosage. Purified recombinant CYP3A4 was also recently shown to 25-hydroxylate the 1α(OH)D2 substrate in vitro, with lower enzymatic activity toward 1α(OH)D3 or vitamin D2 [19]. Interestingly, CYP3A4 expression

107

CHAPTER 7 Mutant Mouse Models of Vitamin D Metabolic Enzymes

vectors were devoid of 25-hydroxylase activity in the transfection-based strategy that led to the identification of CYP2R1 as a putative microsomal vitamin D 25-hydroxylase [17]. CYP3A4 regulates the metabolism of more than half of all commonly used drugs. It remains to be determined whether CYP3A4 plays any physiological role in vitamin D metabolism or whether its activity is limited to the conversion of the prodrug 1α(OH)D2.

III. 24-HYDROXYLATION AND cyp24A1 KNOCKOUT The CYP24A1 enzyme catalyzes the addition of an hydroxyl group on carbon 24 of the vitamin D secosteroid backbone. It can utilize several substrates: 25(OH)D, to produce 24,25-dihydroxyvitamin D [24,25(OH)2D]; 1,25(OH)2D, to produce 1,24,25trihydroxyvitamin D [1,24,25(OH)3D]; and even 1α-hydroxyvitamin D [1α(OH)D], to generate 1α,24dihydroxyvitamin D [1,24(OH)2D]. The 24,25(OH)2D metabolite is the most abundant dihydroxylated metabolite in the circulation. The C-24 oxidation pathway catalyzed by the CYP24A1 molecule comprises five enzymatic steps that involve successive hydroxylation/oxidation reactions at carbons 24 and 23, followed by cleavage of the secosteroid at the C-23/C-24 bond and subsequent oxidation of the cleaved product to calcitroic acid [20,21]. The hypothesis that the main role of the C-24 oxidation pathway is attenuation of the 1,25(OH)2D biological signal inside target cells was tested in vitro using cytochrome P450 inhibitors. Blocking P450 activity by treatment of cells with ketoconazole inhibits catabolism and results in 1,25(OH)2D accumulation and extended hormone action [22]. This hypothesis was also tested and confirmed in vivo by engineering cyp24A1-deficient mice.

A. Phenotype of cyp24A1-Deficient Mice Exons 9 and 10 of the cyp24A1 gene, encoding the heme-binding domain [23], were deleted using homologous recombination in embryonic stem (ES) cells. This strategy ensured that a true null allele was created. Heterozygous animals were phenotypically normal and fertile, and the engineered mutation was transmitted to the progeny with the expected Mendelian ratio [24]. Fifty percent of cyp24A1−/− mice die before 3 weeks of age (Table II) [24,25]. Analysis of macrophage function ruled out impaired responses to infection as the cause for post-natal death. The perinatal lethality is

TABLE II

Incomplete Penetrance of the cyp24A1−/− Perinatal Lethality Phenotype

Number of cyp24A1−/− embryos 578

Alive at weaning

Dead at weaning

274 (47%)

304 (53%)

most likely a consequence of hypercalcemia secondary to hypervitaminosis D, since the inactivation of the cyp24A1 gene in mice impaired the ability of the animals to clear 1,25(OH)2D. Bolus and chronic 1,25 (OH)2D administration resulted in a marked elevation in serum 1,25(OH)2D levels in the mutant animals [24]. Chronic 1,25(OH)2D administration in cyp24A1−/− mutants resulted in histological changes consistent with hypervitaminosis D in the kidney: cortical tubular dilation, necrotic debris, and calcification (nephrocalcinosis). The inability to regulate 1,25(OH)2D and calcium homeostasis presumably leads to fatal hypercalcemia. Indeed, extremely high levels of circulating 1,25(OH)2D and calcium were measured in runted animals that died before weaning [24]. Another condition where vitamin D homeostasis is disregulated in cyp24A1-null mice is pregnancy. Circulating 1,25(OH)2D levels rise in the late stages of gestation in rodents [26]. Since gestating cyp24A1−/− females cannot regulate this increase through CYP24A1-mediated catabolism, serum 1,25(OH)2D concentrations rise to abnormally high levels in pregnant cyp24A1−/− dams [24]. This is accompanied by concomitant elevation of the circulating calcium concentration [24] that sometimes reaches fatal levels. Since half of the mutant progeny appear unaffected by the cyp24A1 deficiency, these animals most likely use alternate means of regulating vitamin D homeostasis. We have measured clearance and metabolism of labeled 1,25(OH)2D in cyp24A1−/− survivors and heterozygote controls. These experiments have confirmed that cyp24A1-null mice have impaired clearance of 1,25(OH)2D. Surprisingly, cyp24A1−/− mice appear to lack not only 24-hydroxylated metabolites but also 1,25(OH)2D-26,23-lactone [27], supporting the view that the CYP24A1 enzyme is also responsible for hydroxylating 1,25(OH)2D at C-23 and C-26 [28,29]. These surprising findings suggest that the cyp24A1null survivors adapt to the impaired vitamin D catabolism not by using an alternative catabolic route, but by limiting the synthesis of the active compound. The survival of some cyp24A1−/− mutant animals to adulthood has also allowed experiments designed to address the effect of perturbing vitamin D metabolism during development. Bone development is

108 abnormal in homozygous mutants born of homozygous females [24,30]. Bone forms via two mechanisms during embryogenesis: intramembranous bone formation and endochondral bone formation [31,32]. The intramembranous formation allows the growth of the surface of flat bones as well as the thickening of cortical bone, while endochondral formation provides for the lengthening of long bones. The major difference between the two is the presence or absence of an intermediate cartilaginous phase. Intramembranous bone formation occurs when mesenchymal precursor cells proliferate and differentiate directly into osteoblasts that produce a mucoprotein matrix, called osteoid, in which collagen fibrils are enmeshed. The osteoblasts then begin to mineralize the osteoid by depositing inorganic crystals of calcium phosphate on, between, and within the collagen fibers, forming a primary immature bone tissue called woven bone. Woven bone is progressively replaced by mature, lamellar bone. In lamellar bone, the matrix is compact and the collagen fibers are long and highly ordered. Endochondral bone formation entails the conversion of a cartilaginous template, the “anlagen,” into bone. Mesenchymal cells condense and differentiate into chondroblasts that secrete the cartilaginous matrix. The chondroblasts embed themselves within the matrix they produce and become chondrocytes. Because of the malleable nature of the cartilaginous matrix, the chondrocytes can continue to proliferate in their lacunae. The embryonic anlagen, a template of the future bone piece, is avascular. During its early development, a ring of woven bone (the bony collar) is formed by intramembranous ossification in the future midshaft area. This calcified woven bone is then invaded by vascular tissue, and this process of angiogenesis brings to the bone rudiment the precursors of the osteoclastic lineage. The osteoclasts excavate the hematopoietic bone marrow cavity while new osteoblasts are recruited to replace the cartilage scaffold with bone matrix. At the extremities of long bones (epiphyses), longitudinal growth occurs by a similar process of endochondral bone formation at the growth plates [32]. Growth-plate chondrocytes undergo several steps of maturation from proliferating chondrocytes, forming characteristic orderly columns, to nonproliferating, hypertrophic cells. The cartilaginous matrix becomes mineralized just below the hypertrophic zone of the growth plate and the chondrocytes then undergo apoptosis. Once calcified, the cartilage matrix is partially resorbed by osteoclasts. Following resorption, osteoblasts differentiate to form a layer of woven bone on top of the mineralized cartilage remnants of the longitudinal septa. This is the first activation–resorption– formation sequence of remodeling the cartilage into

RENÉ ST-ARNAUD

woven bone. Further down in the growth plate, this immature bone will undergo subsequent remodeling to replace the woven bone and calcified cartilage remnants with lamellar bone. An extensive literature demonstrates that cells from the growth plate respond to 24,25(OH)2D in vitro in a cell maturation–dependent manner (reviewed in [33]). The less differentiated cells of the resting zone respond to 24,25(OH)2D. The more mature cells of the prehypertrophic and hypertrophic compartments respond primarily to 1,25(OH)2D. Interestingly, treatment of resting zone chondrocytes with 24,25(OH)2D induces a change in maturation state [34], supporting the hypothesis that 24,25(OH)2D plays a role in cartilage development. It is interesting to note that no significant disruption of growth plate organization was noted in cyp24A1-null mice and that all cell types could readily be identified (Fig. 1), with the expression of differentiation markers such as type II collagen and type X collagen confirmed (St-Arnaud and Beaulieu, unpublished results). Our data suggest that the absence of CYP24A1 activity does not affect growth plate development and

FIGURE 1 Normal growth plate architecture in cyp24A1-null mice. Sagittal sections through the vertebrae of cyp24A1+/− (upper panel) and cyp24A1−/− (lower panel) littermates. Sections were stained by the Goldner method. See color plate.

CHAPTER 7 Mutant Mouse Models of Vitamin D Metabolic Enzymes

that 24,25(OH)2D is not required for chondrocyte maturation in vivo. It remains formally possible, however, that a redundant endocrine system may be able to compensate for the function of the 24,25(OH)2D metabolite in living animals. Histological examination of the bones from homozygous cyp24A1−/− pups born from homozygous cyp24A1-null dams revealed an accumulation of osteoid at sites of intramembranous ossification, however [24,30]. Control heterozygote littermates showed normal bone structure. Two major hypotheses can be formulated to account for the phenotype of the cyp24A1-deficient embryos from cyp24A1 mutant mice: (1) perturbation of 1,25(OH)2D catabolism through the inactivation of the C-24 oxidation pathway affects bone development; (2) metabolites of vitamin D hydroxylated at position 24 are essential for bone development. To differentiate between these possibilities, the cyp24A1-deficient animals were bred to mice carrying an inactivating mutation of the vitamin D receptor (VDR) gene [35]. If elevated 1,25(OH)2D levels, acting through the vitamin D receptor, were responsible for the observed phenotype, then mice lacking the receptor and the cyp24A1 gene should not show the aberrant intramembranous bone development. Using this elegant genetic strategy, incontrovertible evidence was obtained that expression of the VDR is required for the impaired mineralization phenotype of the cyp24A1-deficient animals: double mutant homozygotes (cyp24A1−/− and VDR−/−) showed normal intramembranous bone formation at all sites examined [24]. This demonstrates that elevated 1,25(OH)2D levels during gestation affect mineralization and suggest that impaired vitamin D metabolism during development perturbs bone formation.

B. Future Studies The cyp24A1-deficient strain represents a useful tool to address the still controversial question of the putative biological activity of vitamin D metabolites hydroxylated at the 24 position. Does it remain possible that 24,25(OH)2D exerts biological activity in specific tissues and that genetic redundancy prevented the identification of this activity in cyp24A1-null mice? Using the cyp24A1-ablated mice as a source of costochondral chondrocytes would confirm or refute the importance of local production of vitamin D metabolites in the maturation of chondrocytic cells in this culture system [36]. Published results also suggest that 24,25(OH)2D may be important during fracture repair. The circulating levels of 24,25(OH)2D increase during fracture repair

109

in chicks because of an increase in renal CYP24A1 activity [37]. When the effect of various vitamin D metabolites on the mechanical properties of healed bones was tested, the strength of healed bones was maximized in animals fed 24,25(OH)2D3 [38]. These results support a role for 24,25(OH)2D as an essential vitamin D metabolite for fracture repair. We have started to use the murine model of distraction osteogenesis [39] to compare fracture repair between wild-type mice and cyp24A1−/− mutant mice. If a defect in fracture repair is observed in cyp24A1-null mice, we will attempt rescue using 24,25(OH)2D, 1,24,25(OH)3D, 1,24(OH)2D, or other vitamin D analogs.

IV. 1α-HYDROXYLATION AND cyp27B1 KNOCKOUT The 25-hydroxyvitamin D-1α-hydroxylase enzyme (CYP27B1) converts 25(OH)D to 1,25(OH)2D, the biologically active metabolite of vitamin D. Mutations in the CYP27B1 gene cause pseudo–vitamin D deficiency Rickets (PDDR), a rare autosomal disease characterized by growth retardation, failure to thrive, rickets, and osteomalacia [40,41] (see Chapter 71). Biochemical analysis of serum from affected patients reveals hypocalcemia, secondary hyperparathyroidism, and undetectable levels of 1,25(OH)2D3.

A. Phenotype of cyp27B1-Deficient Mice An animal model of PDDR was engineered independently by two laboratories using targeted inactivation of the cyp27B1 gene in mice [42,43]. The engineered mutation was transmitted to the progeny with the expected Mendelian ratio. Heterozygous animals had no discernable phenotype and were fertile. Homozygous mutant animals were phenotypically normal at birth but exhibited retarded growth as measured by weight gain [43] and femur length at 8 weeks [44]. Serum analysis of homozygous mutant animals confirmed that they were hypocalcemic, hypophosphatemic, and had undetectable circulating levels of 1,25(OH)2D3 [42,43]. Serum parathyroid hormone levels were highly increased as early as weaning [42] and remained elevated thereafter [42,43]. The parathyroid glands also increased in size in young adult cyp27B1 homozygous mutant mice [42,43]. From a biochemical standpoint, the only difference between the cyp27B1−/− mice and patients with PDDR is that patients with PDDR have normal serum levels of 25(OH)D [45–47] and 24,25(OH)2D [48], whereas

110 elevated levels of 25(OH)D [42,43] and very low levels of 24,25(OH)2D [42] are seen in cyp27B1 mutant mice. The 1,25(OH)2D hormone is the main in vivo regulator of the expression of the CYP24A1 enzyme that catalyzes the synthesis of 24,25(OH)2D [49]. Therefore, low circulating 24,25(OH)2D levels and undetectable cyp24A1 expression in cyp27B1−/− mice would be anticipated. The discrepancy with the human disease remains to be explained but could result from species differences. This observed inhibition of cyp24A1 expression in mice, combined with the targeted ablation of the CYP27B1 enzyme, leads to a metabolic block in mutant animals and an accumulation of the unprocessed 25(OH)D substrate [42]. The expression of vitamin D–dependent genes was also analyzed in cyp27B1-null mice. Expression of the calbindin-D9k gene was completely inhibited in the intestine of mutant mice [42,43] and severely downregulated in the kidney [43]. A reduction in the levels of mRNA for calbindin-D28k and the vitamin D receptor in kidney tissue was observed, whereas renal expression of cyp24A1 was almost completely abolished [42,43]. Interestingly, expression levels of the osteocalcin and osteopontin genes in tibia were unaffected in mutant animals [42]. The osteocalcin gene promoter contains several positive and negative regulatory elements [50]. Similarly, osteopontin gene expression is regulated by many agents acting via diverse signaling pathways in specific cell types [51]. Although 1,25(OH)2D3 has been shown to regulate the expression of these genes both in vitro and in vivo [50,51], the unaltered mRNA levels for both genes measured in cyp27B1-null animals suggest that the vitamin D hormone is not the main modulator of the steady-state expression of these genes in murine bone tissue. Histological analysis of the bones from 3-week-old mutant animals confirmed the evidence of rickets [42]. At the age of 7 to 8 weeks, femurs from cyp27B1-ablated mice present a severe disorganization in the architecture of the growth plate and marked osteomalacia [42,43]. The X-ray features of the bones from cyp27B1−/− animals also matched the clinical manifestations of PDDR. Contact radiography of femurs from mutant animals revealed diffuse osteopenia (hypomineralization) and rachitic metaphyseal changes [52]. Inactivation of cyp27B1 therefore produced a bone phenotype that is identical to the human genetic disease PDDR, confirming that mutations in CYP27B1 underlie this disorder. The animal model was further used to compared efficacy of clinical rescue regimens (see later discussion). The vitamin D hormone has been shown to have immunomodulatory effects, and local synthesis of 1,25(OH)2D could play an important autocrine or paracrine role in the differentiation or function of immune cells [53–55]. Interestingly, cyp27B1-deficient

RENÉ ST-ARNAUD

mice exhibited a reduction in CD4- and CD8-positive peripheral T lymphocyte populations [43]. Large CD4positive ectopic lymph nodes were detected in the neck of the cyp27B1-null and cyp27B1+/− heterozygous animals [43]. These findings support an important role for CYP27B1 in the regulation of immune function. This phenotype has never been reported in vitamin D receptor (VDR) knockout strains [35,56,57], which raises the hypothesis that 1,25(OH)2D may not exert its immunomodulatory effects via the VDR or that CYP27B1 influences immune function via metabolites other than 1,25(OH)2D [43]. The fertility and reproductive capacity of vitamin D– deficient rats is impaired, and normal reproductive effectiveness can be rescued by normalizing calcemia in males but not females, where 1,25(OH)2D is required for reproductive function [58,59]. cyp27B1-deficient females showed uterine hypoplasia, decreased ovarian size, compromised folliculogenesis, and infertility [43]. A similar phenotype was observed in one strain of VDR-deficient mice [57]. These results establish a critical role for cyp27B1 function in female reproduction.

B. Rescue of the Phenotype 1. 1,25(OH)2D REPLACEMENT THERAPY

The treatment of choice for PDDR patients is replacement therapy with 1,25(OH)2D3 [45]. It results in rapid and complete correction of the abnormal phenotype, restoring normocalcemia, eliminating secondary hyperparathyroidism and features of rickets. The restoration of bone mineral content is equally rapid and histological evidence of healing has been documented [45]. This therapeutic regimen was used to rescue the PDDR phenotype of cyp27B1−/− mice [60]. Animals received daily subcutaneous injections of 1,25(OH)2D3. A “rescue” concentration of 500 pg per g body weight (BW) was administered from 21 days of age until 35 days of age, followed by a “maintenance” dose of 100 pg per g BW (s.c.) from 35 days of age until they reached the age of 60 days. This treatment normalized biochemical parameters, restored the biomechanical properties of bones, and cured rickets and osteomalacia, as assessed by histomorphometric analyses [60]. PDDR patients receiving 1,25(OH)2D3-replacement therapy have normal serum concentrations of the metabolite [45], whereas chronic treatment with 1,25(OH)2D3 did not normalize circulating levels of the metabolite in cyp27B1−/− mice [60]. This difference is most likely due to the induction of the CYP24A1 enzyme observed in 1,25(OH)2D3-treated cyp27B1null mutants [60], and further highlights the species differences in vitamin D metabolic pathways that was noted in the first description of the animal model for

111

CHAPTER 7 Mutant Mouse Models of Vitamin D Metabolic Enzymes

PDDR [42,43], with regard to the regulation of cyp24A1 gene expression and circulating 25(OH)D3 or 24,25(OH)2D3 metabolite levels. 2. 1α(OH)D REPLACEMENT THERAPY

Before 1,25(OH)2D3 became available as a drug, clinicians used the monohydroxylated analog 1α(OH)D3 for treating PDDR patients [61]. This compound, which requires only liver hydroxylation at the 25 position to fully mimic 1,25(OH)2D3, is still used in Japan and some European countries. We have used a rescue regimen similar to the treatment described earlier to treat cyp27B1-deficient mice with 1α(OH)D3. The treatment was as effective as the injection protocol with 1,25(OH)2D3 to normalize calcemia, phosphatemia, and reverse the secondary hyperparathyroidism in cyp27B1−/− mice, without deleterious effects in heterozygous controls. The replacement therapy with 1α(OH)D3 also normalized growth and corrected the bone phenotype of the mutant mice, effectively curing rickets and osteomalacia and normalizing histomorphometric indices. Experiments are in progress to compare the efficacy of other vitamin D analogs, such as 1α(OH)D2 or 1,24(S)(OH)2D2 [62], in rescuing the PDDR phenotype of cyp27B1 knockout animals. 3. HIGH-CALCIUM, HIGH-PHOSPHORUS, HIGH-LACTOSE DIET

Mutation in the vitamin D receptor (VDR) gene results in a second form of vitamin D–related rickets called hereditary vitamin D–resistant rickets (HVDRR),

characterized by hypocalcemia, secondary hyperparathyroidism, rickets, osteomalacia, and alopecia [63] (see Chapter 72). Several laboratories have developed a valid mouse model for this type of hereditary rickets [35,56,57]. Rescue of rickets has been successfully performed using a 2% calcium diet, 1.25% phosphorus and 20% lactose, previously reported to prevent hypocalcemia in vitamin D–deficient rats [64]. This diet completely normalized all biochemical and histomorphometric parameters in VDR-ablated animals [65,66]. We rescued the PDDR phenotype of mice deficient for the cyp27B1 gene by feeding them with the highcalcium diet [67]. Homozygous mutant and heterozygous control animals were fed the rescue diet (2% calcium, 1.25% phosphorus, and 20% lactose) from 3 weeks of age until sacrifice at 8.5 weeks of age. Blood biochemistry analysis revealed that the rescue diet corrected the hypocalcemia and secondary hyperparathyroidism. Despite the restoration of normocalcemia, cyp27B1−/− (and heterozygous control) animals fed the rescue diet initially gained weight less rapidly than control mice fed normal mouse chow. Although cyp27B1−/− mice fed the rescue diet eventually reached the same weight as control animals, the treatment did not entirely correct bone growth as femur size remained significantly smaller than control (Fig. 2). Bone histology and histomorphometry confirmed that the rickets and osteomalacia were cured. The rescue diet also restored the biomechanical properties of the bone tissue within normal parameters. These results demonstrate that

Growth rescue in cyp27B1-deficient mice *

20

**

control

* Femur length (mm)

1,25(OH)2D3 rescue High Ca2+ diet rescue 10

0 +/−

−/−

The high-calcium rescue diet only partially restores bone growth of cyp27B1−/− mice. Femurs were measured at sacrifice (60 days old). cyp27B1−/− animals had significantly reduced femur lengths, which was only partially corrected by the high-calcium diet (High Ca2+ diet rescue). Replacement therapy with 1,25(OH)2D3 (1,25(OH)2D3 rescue) normalized femur growth. *, p < 0.05; **, p < 0.01.

FIGURE 2

112 correction of the abnormal mineral ion homeostasis by feeding with a high-calcium rescue diet is effective to rescue the PDDR phenotype of cyp27B1 mutant mice. This treatment, however, does not appear as effective as 1,25(OH)2D3 replacement therapy since bone growth remained impaired. In addition to the partial rescue of femur growth, the high-calcium rescue diet increased growth plate thickness in cyp27B1−/− and cyp27B1+/− animals [67]. This phenomenon was not observed in VDR-ablated mice receiving similar treatment [65,66]. A major difference between the two strains of animals is the ability to synthesize 1,25(OH)2D3, both systemically and locally. It has been hypothesized that local production of 1,25(OH)2D3 is important for the proliferation and maturation of chondrocytes in the growth plate, and that some of these effects involve nongenomic mechanisms of action for 1,25(OH)2D3 [33]. Autocrine synthesis and nongenomic actions of 1,25(OH)2D3 would remain active in VDR-ablated animals, whereas they are affected in cyp27B1-deficient mice. This could explain the differences in the response of the two strains to the dietary manipulation. The targeted cyp27B1 mouse model that we have engineered [42] represents an invaluable tool to characterize the putative autocrine/ paracrine roles of 1,25(OH)2D3 in the growth plate under conditions of normal mineral ion homeostasis (see later discussion).

C. Future Studies Local production of 1,25(OH)2D could play an important autocrine or paracrine role in the differentiation or function of osteoblasts, chondrocytes, keratinocytes, and macrophages [33,68,69]. The targeting strategy used to engineer the cyp27B1-deficient strain, based on the Cre/lox methodology, allows tissuespecific inactivation of the targeted gene. The technology utilizes the Cre recombinase from the P1 bacteriophage. This enzyme catalyzes recombination between two 34-base-pair recognition elements, called loxP sites, causing excision of the intervening sequences. Transgenic mice lines expressing Cre under the control of tissue-specific promoters are crossed with mice harboring loxP sites to achieve excision at will (Fig. 3) (reviewed in [70]). The cyp27B1 gene was cloned from a 129 SV mouse genomic library to construct a targeting vector in which exon 8, encoding the heme binding domain [71], was flanked by a 5′-loxP recognition site and by a 3′-loxP-neo-loxP selection cassette. Following homologous recombination at the cyp27B1 locus in ES cells and injection of targeted ES cells in host blastocysts, a mouse strain in which the cyp27B1

RENÉ ST-ARNAUD

locus is “floxed” (flanked by loxP) was established. The conditional cyp27B1 allele provides an invaluable genetic tool to analyze the putative autocrine/paracrine roles that have been hypothesized for 1,25(OH)2D in various cell types. It may be argued that a significant portion of the active 1,25(OH)2D metabolite found in target tissues is derived systemically. Against the background of circulating levels of the hormone, any local drop may be too subtle an effect to be easily detectable. However, a recent finding in the field of sex steroids is that a large proportion of androgens and estrogens are synthesized locally in peripheral target tissues such as skin or adipose tissue [72]. Despite “central” steroidogenesis in the gonads, peripheral target tissues utilize “on the spot” intracellular hormone synthesis/action to regulate their function [73]. Moreover, chondrocytes are compartmentalized away from the circulation, and local synthesis of vitamin D metabolites may contribute significantly to the regulation of the proliferation, differentiation, and function of growth plate cells. The 1,25(OH)2D hormone has been shown to stimulate macrophage function. Treatment of macrophages with 1,25(OH)2D enhances macrophage chemotaxis [74] and increases phagocytic killing of bacteria [75]. Crossing the floxed cyp27B1 mice with mice expressing Cre in macrophages under the control of the macrophage scavenger receptor A gene promoter [76] would inhibit autocrine synthesis of 1,25(OH)2D by macrophages. One possible phenotype that could be predicted following this genetic manipulation would be increased susceptibility to microbial infection. The 1,25(OH)2D hormone is likely to be an autocrine or paracrine factor for epidermal differentiation [77] since keratinocytes make 1,25(OH)2D [78], express the VDR [79], and respond to 1,25(OH)2D with changes in proliferation and differentiation [79,80]. Indeed, our results show that the expression of keratinocyte differentiation markers is reduced in the skin of cyp27B1 knockout animals and that recovery of the epidermal barrier (assessed by the transepidermal water loss assay) [81] is significantly delayed in CYP27B1-deficient skin [82]. Tissue-specific inactivation of cyp27B1 in floxed mice crossed to keratin 5-Cre [83] or keratin 14-Cre transgenicmice [84] would allow examination of the contribution of systemic versus local production of 1,25(OH)2D on skin differentiation. Chondrocytes have been shown to express cyp27B1 both in vivo [71] and in vitro [36,85]. This has led investigators to hypothesize that local production of 1,25(OH)2D could play an important autocrine or paracrine role in the differentiation or function of these cells [33]. To test this putative role of the vitamin D

CHAPTER 7 Mutant Mouse Models of Vitamin D Metabolic Enzymes

113

FIGURE 3 Tissue-specific gene inactivation using the Cre/lox strategy. Floxed (flanked by loxP) mice are crossed with transgenic mice in which the Cre recombinase is under the control of a tissue-specific promoter (TSP). In this example, loxP sites (red triangles) flank exon 8 of the target gene and the tissue-specific promoter is expressed in bone tissue. Expression of the Cre transgene in bone cells will cause excision of exon 8 between the two loxP sites (bottom left), while the target gene will remain intact in other tissues due to lack of Cre expression (bottom right panel).

hormone in growth plate cells, we have initiated cyp27B1 gene ablation in chondrocytes [86] by crossing the floxed cyp27B1 strain with transgenic mice expressing the Cre recombinase under the control of the type II collagen promoter/enhancer (Col2-Cre) [87]. Because 1,25(OH)2D regulates the growth and terminal differentiation of hypertrophic chondrocytes in vitro [33,88,89], we hypothesize that these parameters will be perturbed in vivo following the genetic manipulation that we will engineer. This should be manifested as increased chondrocyte proliferation and an abnormal architecture of the growth plate, which could even affect the primary spongiosa. Neovascularization and VEGF expression could be reduced [90]. The expression of differentiation markers could be perturbed in Col2-Cre; 1α−/fl mice.

V. SUMMARY AND PERSPECTIVES Mutant mouse models have allowed significant progress in our understanding of vitamin D metabolism and function. The observation that cyp27A1−/− (mitochondrial 25-OHase) mice have normal vitamin D homeostasis suggested that a distinct cytochrome P450 enzyme could be involved in the 25-hydroxylation of vitamin D under physiological conditions. An expressionbased cloning strategy using a cDNA library prepared from the liver of cyp27A1−/− animals was then used to identify the microsomal 25-hydroxylating enzyme, CYP2R1. It will prove interesting to analyze vitamin D metabolism and circulating levels of 25(OH)D and 1,25(OH)2D in cyp2R1-null mice and cyp27A1-cyp2R1 compound mutants to determine the relative contributions

114 of the two enzymes in vivo. The cyp24A1-deficient strain will be used to test the putative role of vitamin D metabolites hydroxylated at position 24 in mammalian fracture repair. The confirmation of such a role for 24,25(OH)2D would open up interesting avenues for clinical research using vitamin D analogs. Finally, the cyp27B1-deficient mice represent a valid animal model for pseudo–vitamin D deficiency rickets. The conditional ablation strategy used to engineer these animals will help elucidate the hypothesized autocrine/ paracrine roles of 1,25(OH)2D in target tissues such as cartilage, skin, and macrophages.

References 1. McAlpine PJ, Van Cong N, Boucheix C, Pakstis AJ, Doute RC, Shows TB 1987 The 1987 Catalog of mapped genes and report of the nomenclature committee. Cytogenet Cell Genet 46:29–101. 2. Shows TB, et al., McAlpine PJ, Boucheix C, Collins FS, Conneally PM, Frezal J, Gershowitz H, Goodfellow PN, Hall JG 1987 Guidelines for human gene nomenclature. An international system for human gene nomenclature (ISGN, 1987). Cytogenet Cell Genet 46:11–28. 3. Okuda KI, Usui E, Ohyama Y 1995 Recent progress in enzymology and molecular biology of enzymes involved in vitamin D metabolism. J Lipid Res 36:1641–1652. 4. Jones G, Strugnell SA, DeLuca HF 1998 Current understanding of the molecular actions of vitamin D. Physiol Rev 78:1193–1231. 5. Rosen H, Reshef A, Maeda N, Lippoldt A, Shpizen S, Triger L, Eggertsen G, Bjorkhem I, Leitersdorf E 1998 Markedly reduced bile acid synthesis but maintained levels of cholesterol and vitamin D metabolites in mice with disrupted sterol 27-hydroxylase gene. J Biol Chem 273:14805–14812. 6. Salen G, Shefer S, Cheng FW, Dayal B, Batta AK, Tint GS 1979 Cholic acid biosynthesis: The enzymatic defect in cerebrotendinous xanthomatosis. J Clin Invest 63:38–44. 7. Cali JJ, Hsieh CL, Francke U, Russell DW 1991 Mutations in the bile acid biosynthetic enzyme sterol 27-hydroxylase underlie cerebrotendinous xanthomatosis. J Biol Chem 266: 7779–7783. 8. Kuriyama M, Fujiyama J, Kubota R, Nakagawa M, Osame M 1993 Osteoporosis and increased bone fractures in cerebrotendinous xanthomatosis. Metabolism 42:1497–1498. 9. Fukushima M, Suzuki Y, Tohira Y, Nishii Y, Suzuki M 1976 25-Hydroxylation of 1α-hydroxyvitamin D3 in vivo and in perfused rat liver. FEBS Lett 65:211–214. 10. Cali JJ, Russell DW 1991 Characterization of human sterol 27-hydroxylase. A mitochondrial cytochrome P-450 that catalyzes multiple oxidation reactions in bile acid biosynthesis. J Biol Chem 266:7774–7778. 11. Andersson D, Davis DL, Dahlback H, Jornvall H 1989 Cloning, structure and expression of the mitochondrial cytochrome P-450 sterol 26-hydroxylase, a bile acid biosynthetic enzyme. J Biol Chem 264:8222–8229. 12. Madhok TC, DeLuca HF 1979 Characteristics of the rat liver microsomal enzyme system converting cholecalciferol into 25-hydroxycholecalciferol. Evidence for the participation of cytochrome p-450. Biochem J 184:491–499.

RENÉ ST-ARNAUD

13. Vlahcevic ZR, Jairath SK, Heuman DM, Stravitz RT, Hylemon PB, Avadhani NG, Pandak WM 1996 Transcriptional regulation of hepatic sterol 27-hydroxylase by bile acids. Am J Physiol 270:G646–652. 14. Bhattacharyya MH, DeLuca HF 1973 The regulation of rat liver calciferol-25-hydroxylase. J Biol Chem 248:2969–2973. 15. Guo YD, Strugnell S, Back DW, Jones G 1993 Transfected human liver cytochrome P-450 hydroxylates vitamin D analogs at different side-chain positions. Proc Natl Acad Sci USA 90:8668–8672. 16. Repa JJ, Lund EG, Horton JD, Leitersdorf E, Russell DW, Dietschy JM, Turley SD 2000 Disruption of the sterol 27-hydroxylase gene in mice results in hepatomegaly and hypertriglyceridemia. Reversal by cholic acid feeding. J Biol Chem 275:39685–39692. 17. Cheng JB, Motola DL, Mangelsdorf DJ, Russell DW 2003 De-orphanization of cytochrome P450 2R1: A microsomal vitamin d 25-hydroxylase. J Biol Chem 278:38084–38093. 18. Bjorkhem I, Holmberg I 1978 Assay and properties of a mitochondrial 25-hydroxylase active on vitamin D3. J Biol Chem 253:842–849. 19. Gupta RP, Hollis BW, Patel SB, Bell NH 2003 CYP3A4 is a human microsomal vitamin D-25-hydroxylase. J Bone Miner Res (in press). 20. Makin G, Lohnes D, Byford V, Ray R, Jones G 1989 Target cell metabolism of 1,25-dihydroxyvitamin D3 to calcitroic acid. Evidence for a pathway in kidney and bone involving 24-oxidation. Biochem J 262:173–180. 21. Reddy GS, Tserng KY 1989 Calcitroic acid, end product of renal metabolism of 1,25-dihydroxyvitamin D3 through C-24 oxidation pathway. Biochemistry 28:1763–1769. 22. Reinhardt TA, Horst RL 1989 Ketoconazole inhibits selfinduced metabolism of 1,25-dihydroxyvitamin D3 and amplifies 1,25-dihydroxyvitamin D3 receptor up-regulation in rat osteosarcoma cells. Arch Biochem Biophys 272:459–465. 23. Ohyama Y, Noshiro M, Eggertsen G, Gotoh O, Kato Y, Bjorkhem I, Okuda K 1993 Structural characterization of the gene encoding rat 25-hydroxyvitamin D3 24-hydroxylase. Biochemistry 32:76–82. 24. St-Arnaud R, Arabian A, Travers R, Barletta F, Raval-Pandya M, Chapin K, Depovere J, Mathieu C, Christakos S, Demay MB, Glorieux FH 2000 Deficient mineralization of intramembranous bone in vitamin D-24-hydroxylase-ablated mice is due to elevated 1,25-dihydroxyvitamin D and not to the absence of 24,25-dihydroxyvitamin D. Endocrinology 141:2658–2666. 25. St-Arnaud R 1999 Targeted inactivation of vitamin D hydroxylases in mice. Bone 25:127–129. 26. Kovacs CS, Kronenberg HM 1997 Maternal-fetal calcium and bone metabolism during pregnancy, puerperium, and lactation. Endocr Rev 18:832–872. 27. Masuda S, Byford V, Arabian A, Sakai Y, Demay MB, St-Arnaud R, Jones G 2004 Altered pharmacokinetics of 1α,25-dihydroxyvitamin D3 and 25-hydroxyvitamin D3 in blood and tissues of the vitamin D-24-hydroxylase null mouse. Submitted. 28. Miyamoto Y, Shinki T, Yamamoto K, Ohyama Y, Iwasaki H, Hosotani R, Kasama T, Takayama H, Yamada S, Suda T 1997 1α,25-dihydroxyvitamin D3-24-hydroxylase (CYP24) hydroxylates the carbon at the end of the side chain (C-26) of the C-24-fluorinated analog of 1α,25-dihydroxyvitamin D3. J Biol Chem 272:14115–14119. 29. Beckman MJ, Tadikonda P, Werner E, Prahl J, Yamada S, DeLuca HF 1996 Human 25-hydroxyvitamin D3-24-hydroxylase, a multicatalytic enzyme. Biochemistry 35:8465–8472.

CHAPTER 7 Mutant Mouse Models of Vitamin D Metabolic Enzymes

30. St-Arnaud R, Glorieux FH 1997 Vitamin D and bone development. In: Feldman D, Glorieux FH, Pike JW (eds) Vitamin D. Academic Press, San Diego, pp. 293–303. 31. Zelzer E, Olsen BR 2003 The genetic basis for skeletal diseases. Nature 423:343–348. 32. Kronenberg HM 2003 Developmental regulation of the growth plate. Nature 423:332–336. 33. Boyan BD, Dean DD, Sylvia VL, Schwartz Z 1997 Cartilage and vitamin D: genomic and nongenomic regulation by 1,25(OH)2D3 and 24,25(OH)2D3. In: Feldman D, Glorieux FH, Pike JW (eds) Vitamin D. Academic Press, San Diego, pp. 395–422. 34. Schwartz Z, Dean D, Walton J, Brooks B, Boyan B 1995 Treatment of resting zone chondrocytes with 24,25-dihydroxyvitamin D3 [24,25-(OH)2D3] induces differentiation into a 1,25-(OH)2D3-responsive phenotype characteristic of growth zone chondrocytes. Endocrinology 136:402–411. 35. Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997 Targeted ablation of the vitamin D receptor: an animal model of vitamin D-dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94:9831–9835. 36. Schwartz Z, Brooks B, Swain L, Del Toro F, Norman A, Boyan B 1992 Production of 1,25-dihydroxyvitamin D3 and 24, 25-dihydroxyvitamin D3 by growth zone and resting zone chondrocytes is dependent on cell maturation and is regulated by hormones and growth factors. Endocrinology 130:2495–2504. 37. Seo EG, Norman AW 1997 Three-fold induction of renal 25-hydroxyvitamin D3-24-hydroxylase activity and increased serum 24,25-dihydroxyvitamin D3 levels are correlated with the healing process after chick tibial fracture. J Bone Miner Res 12:598–606. 38. Seo EG, Einhorn TA, Norman AW 1997 24R,25-dihydroxyvitamin D3: An essential vitamin D3 metabolite for both normal bone integrity and healing of tibial fracture in chicks. Endocrinology 138:3864–3872. 39. Tay BK, Le AX, Gould SE, Helms JA 1998 Histochemical and molecular analyses of distraction osteogenesis in a mouse model. J Orthop Res 16:636–642. 40. Miller WL, Portale AA 1999 Genetic causes of rickets. Curr Opin Pediatr 11:333–339. 41. St-Arnaud R, Glorieux FH 2000 Hereditary defects in vitamin D metabolism and action. In: DeGroot LJ, Jameson JL (eds) Endocrinology, 4th ed., Vol. 2. W.B. Saunders, Philadelphia; pp. 1154–1168. 42. Dardenne O, Prud’homme J, Arabian A, Glorieux FH, St-Arnaud R 2001 Targeted inactivation of the 25-hydroxyvitamin D3-1α-hydroxylase gene (CYP27B1) creates an animal model of pseudovitamin D-deficiency rickets. Endocrinology 142:3135–3141. 43. Panda DK, Miao D, Tremblay ML, Sirois J, Farookhi R, Hendy GN, Goltzman D 2001 Targeted ablation of the 25-hydroxyvitamin D 1α-hydroxylase enzyme: Evidence for skeletal, reproductive, and immune dysfunction. Proc Natl Acad Sci USA 98:7498–7503. 44. St-Arnaud R, Demay MB 2003 Vitamin D biology. In: Glorieux FH, Juppner H, Pettifor JM (eds) Pediatric Bone: Biology & Diseases. Elsevier/Academic Press, San Diego, pp. 193–216. 45. Delvin EE, Glorieux FH, Marie PJ, Pettifor JM 1981 Vitamin D dependency: replacement therapy with calcitriol. J Pediatr 99: 26–34. 46. Rosen JF, Finberg L 1972 Vitamin D-dependent rickets: actions of parathyroid hormone and 25-hydroxycholecalciferol. Pediatr Res 6:552–562.

115

47. Scriver CR, Reade TM, DeLuca HF, Hamstra AJ 1978 Serum 1,25-dihydroxyvitamin D levels in normal subjects and in patients with hereditary rickets or bone disease. N Engl J Med 299:976–979. 48. Mandla S, Jones G, Tenenhouse HS 1992 Normal 24-hydroxylation of vitamin D metabolites in patients with vitamin Ddependency rickets type I. Structural implications for the vitamin D hydroxylases. J Clin Endocrinol Metab 74:814–820. 49. Shinki T, Jin CH, Nishimura A, Nagai Y, Ohyama Y, Noshiro M, Okuda K, Suda T 1992 Parathyroid hormone inhibits 25-hydroxyvitamin D3-24-hydroxylase mRNA expression stimulated by 1α,25-dihydroxyvitamin D3 in rat kidney but not in intestine. J Biol Chem 267:13757–13762. 50. Lian JB, Stein GS, Stein JL, van Wijnen AJ 1998 Osteocalcin gene promoter: Unlocking the secrets for regulation of osteoblast growth and differentiation. J Cell Biochem Suppl 31:62–72. 51. Denhardt DT, Noda M 1998 Osteopontin expression and function: role in bone remodeling. J Cell Biochem Suppl 31:92–102. 52. St-Arnaud R, Dardenne O, Glorieux FH 2001 Etiologie moléculaire des rachitismes vitamino-dépendents héréditaires. Médecine/Sciences 17:1289–1296. 53. Abe E, Miyaura C, Sakagami H, Takeda M, Konno K, Yamazaki T, Yoshiki S, Suda T 1981 Differentiation of mouse myeloid leukemia cells induced by 1α,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 78:4990–4994. 54. Nunn JD, Katz DR, Barker S, Fraher LJ, Hewison M, Hendy GN, O’Riordan JL 1986 Regulation of human tonsillar T-cell proliferation by the active metabolite of vitamin D3. Immunology 59:479–484. 55. Provvedini DM, Rulot CM, Sobol RE, Tsoukas CD, Manolagas SC 1987 1α,25-Dihydroxyvitamin D3 receptors in human thymic and tonsillar lymphocytes. J Bone Miner Res 2:239–247. 56. Van Cromphaut SJ, Dewerchin M, Hoenderop JG, Stockmans I, Van Herck E, Kato S, Bindels RJ, Collen D, Carmeliet P, Bouillon R, Carmeliet G 2001 Duodenal calcium absorption in vitamin D receptor-knockout mice: functional and molecular aspects. Proc Natl Acad Sci USA 98:13324–13329. 57. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Genet 16:391–396. 58. Halloran BP, DeLuca HF 1980 Effect of vitamin D deficiency on fertility and reproductive capacity in the female rat. J Nutr 110:1573–1580. 59. Kwiecinski GG, Petrie GI, DeLuca HF 1989 Vitamin D is necessary for reproductive functions of the male rat. J Nutr 119:741–744. 60. Dardenne O, Prudhomme J, Hacking SA, Glorieux FH, St-Arnaud R 2003 Rescue of the pseudo-vitamin D deficiency rickets phenotype of CYP27B1-deficient mice by treatment with 1,25-dihydroxyvitamin D3: biochemical, histomorphometric, and biomechanical analyses. J Bone Miner Res 18: 637–643. 61. Reade TM, Scriver CR, Glorieux FH, Nogrady B, Delvin E, Poirier R, Holick F, DeLuca HF 1975 Response to crystalline 1α-hydroxyvitamin D3 in vitamin D dependency. Pediatr Res 9:593–599. 62. Horst R, Prapong S, Reinhardt T, Koszewski N, Knutson J, Bishop C 2000 Comparison of the relative effects of 1,24dihydroxyvitamin D(2) [1,24-(OH)(2)D(2)], 1,24-dihydroxyvitamin D(3) [1,24-(OH)(2)D(3)], and 1,25-dihydroxyvitamin D(3)

116

63. 64. 65.

66.

67.

68.

69. 70. 71.

72. 73.

74.

75.

76.

77.

[1,25-(OH)(2)D(3)] on selected vitamin D-regulated events in the rat. Biochem Pharmacol 60:701–708. Malloy PJ, Pike JW, Feldman D 1999 The vitamin D receptor and the syndrome of hereditary 1,25-dihydroxyvitamin Dresistant rickets. Endocr Rev 20:156–188. Kollenkirchen U, Fox J, Walters MR 1991 Normocalcemia without hyperparathyroidism in vitamin D-deficient rats. J Bone Miner Res 6:273–278. Li YC, Amling M, Pirro AE, Priemel M, Meuse J, Baron R, Delling G, Demay MB 1998 Normalization of mineral ion homeostasis by dietary means prevents hyperparathyroidism, rickets, and osteomalacia, but not alopecia in vitamin D receptor-ablated mice. Endocrinology 139:4391–4396. Amling M, Priemel M, Holzmann T, Chapin K, Rueger JM, Baron R, Demay MB 1999 Rescue of the skeletal phenotype of vitamin D receptor-ablated mice in the setting of normal mineral ion homeostasis: formal histomorphometric and biomechanical analyses. Endocrinology 140:4982–4987. Dardenne O, Prud’homme J, Hacking SA, Glorieux FH, St-Arnaud R 2003 Correction of the abnormal mineral ion homeostasis with a high-calcium, high-phosphorus, highlactose diet rescues the PDDR phenotype of mice deficient for the 25-hydroxyvitamin D-1α-hydroxylase (CYP27B1). Bone 32:332–340. Hewison M, O’Riordan JLH 1997 Immunomodulatory and cell differentiation effects of vitamin D. In: Feldman L, Glorieux FH, Pike JW (eds) Vitamin D. Academic Press, San Diego, pp. 447–462. Bikle D 1997 Vitamin D and skin. In: Feldman D, Glorieux FH, Pike JW (eds) Vitamin D. Academic Press, San Diego, pp. 379–394. Jiang R, Gridley T 1997 Gene targeting: Things go better with Cre. Curr Biol 7:R321–323. Panda DK, Al Kawas S, Seldin MF, Hendy GN, Goltzman D 2001 25-hydroxyvitamin D 1α-hydroxylase: structure of the mouse gene, chromosomal assignment, and developmental expression. J Bone Miner Res 16:46–56. Chen W, Thiboutot D, Zouboulis CC 2002 Cutaneous androgen metabolism: basic research and clinical perspective. J Invest Dermatol 119:992–1007. Labrie F, Luu-The V, Lin SX, Simard J, Labrie C, El-Alfy M, Pelletier G, Bélanger A 2000 Intracrinology: role of the family of 17β−hydroxysteroid dehydrogenases in human physiology and disease. J Mol Endocrinol 25:1–16. Girasole G, Wang JM, Pedrazzoni M, Pioli G, Balotta C, Passeri M, Lazzarin A, Ridolfo A, Mantovani A 1990 Augmentation of monocyte chemotaxis by 1α,25-dihydroxyvitamin D3. Stimulation of defective migration of AIDS patients. J Immunol 145:2459–2464. Rook GA, Taverne J, Leveton C, Steele J 1987 The role of gamma-interferon, vitamin D3 metabolites and tumour necrosis factor in the pathogenesis of tuberculosis. Immunology 62:229–234. Horvai A, Palinski W, Wu H, Moulton KS, Kalla K, Glass CK 1995 Scavenger receptor A gene regulatory elements target gene expression to macrophages and to foam cells of atherosclerotic lesions. Proc Natl Acad Sci USA 92:5391–5395. Pillai S, Bikle DD, Elias PM 1988 Vitamin D and epidermal differentiation: Evidence for a role of endogenously produced

RENÉ ST-ARNAUD

78. 79. 80.

81. 82.

83.

84.

85.

86.

87.

88.

89.

90.

vitamin D metabolites in keratinocyte differentiation. Skin Pharmacol 1:149–160. Bikle DD, Nemanic MK, Whitney JO, Elias PW 1986 Neonatal human foreskin keratinocytes produce 1,25-dihydroxyvitamin D3. Biochemistry 25:1545–1548. Pillai S, Bikle DD, Elias PM 1988 1,25-Dihydroxyvitamin D production and receptor binding in human keratinocytes varies with differentiation. J Biol Chem 263:5390–5395. Smith EL, Walworth NC, Holick MF 1986 Effect of 1α,25dihydroxyvitamin D3 on the morphologic and biochemical differentiation of cultured human epidermal keratinocytes grown in serum-free conditions. J Invest Dermatol 86: 709–714. Vinson LJ, Singer EJ, Kowbler WR, Lehman MD, Masurat T 1965 The nature of the epidermal barrier and some factors influencing skin permeability. Tox Appl Pharm 7:7–19. Bikle DD, Chang S, Crumrine D, Elalieh H, Man M-Q, Choi EH, Dardenne O, Xie Z, St-Arnaud R, Feingold K, Elias PM 2004 25-hydroxyvitamin D 1 alpha hydroxylase is required for optimal epidermal differentiation and permeability barrier homeostasis. J Invest Dermatol 122:984–992. Tarutani M, Itami S, Okabe M, Ikawa M, Tezuka T, Yoshikawa K, Kinoshita T, Takeda J 1997 Tissue-specific knockout of the mouse Pig-a gene reveals important roles for GPI-anchored proteins in skin development. Proc Natl Acad Sci USA 94:7400–7405. Vasioukhin V, Degenstein L, Wise B, Fuchs E 1999 The magical touch: genome targeting in epidermal stem cells induced by tamoxifen application to mouse skin. Proc Natl Acad Sci USA 96:8551–8556. Pedrozo HA, Boyan BD, Mazock J, Dean DD, Gomez R, Schwartz Z 1999 TGFβ1 regulates 25-hydroxyvitamin D3 1αand 24-hydroxylase activity in cultured growth plate chondrocytes in a maturation-dependent manner. Calcif Tissue Int 64:50–56. St-Arnaud R, Dardenne O, Prud’homme J, Hacking SA, Glorieux FH 2003 Conventional and tissue-specific inactivation of the 25-hydroxyvitamin D-1α-hydroxylase (CYP27B1). J Cell Biochem 88:245–251. Terpstra L, Prud’homme J, Arabian A, Takeda S, Karsenty G, Dedhar S, St-Arnaud R 2003 Reduced chondrocyte proliferation and chondrodysplasia in mice lacking the integrin-linked kinase in chondrocytes. J Cell Biol 162:139–148. Gerstenfeld LC, Kelly CM, Von Deck M, Lian JB 1990 Comparative morphological and biochemical analysis of hypertrophic, non-hypertrophic and 1,25(OH)2D3 treated non-hypertrophic chondrocytes. Connect Tissue Res 24: 29–39. Kato Y, Shimazu A, Iwamoto M, Nakashima K, Koike T, Suzuki F, Nishii Y, Sato K 1990 Role of 1,25-dihydroxycholecalciferol in growth-plate cartilage: inhibition of terminal differentiation of chondrocytes in vitro and in vivo. Proc Natl Acad Sci USA 87:6522–6526. Lin R, Amizuka N, Sasaki T, Aarts MM, Ozawa H, Goltzman D, Henderson JE, White JH 2002 1α,25-dihydroxyvitamin D3 promotes vascularization of the chondro-osseous junction by stimulating expression of vascular endothelial growth factor and matrix metalloproteinase 9. J Bone Miner Res 17:1604–1612.

CHAPTER 8

Vitamin D–Binding Protein CHRISTOPHER J. LAING AND NANCY E. COOKE Departments of Genetics and Medicine, University of Pennsylvania, Philadelphia, Pennsylvania

I. Introduction II. Vitamin D Binding Protein and Its Gene III. Functional Features of Vitamin D Binding Protein

IV. Concluding Remarks References

I. INTRODUCTION

been cloned and sequenced. The amino acid structure of DBP was first predicted from these cDNA sequences. Human DBP mRNA contains 1690 nucleotides and encodes a 458-amino-acid secreted protein. There is a 16-amino-acid signal peptide, based upon alignment of the sequenced amino terminus of the mature protein [15,16] with the cDNA-predicted sequence of the primary translation product. The predicted mature DBP proteins in rat, mouse, and rabbit are each 460 amino acids long, two amino-acids longer than the human sequence, whereas that of T. scripta is 466 amino acids in length (see Section II,B,3). There is an N-linked glycosylation consensus sequence in human, rat, and mouse but not rabbit or T. scripta DBP. Periodic positioning of 28 cysteine residues predicts a characteristic secondary structure demarcated by internal disulfide bonds and defines the presence of three internally homologous domains [15] (Fig. 1). The crystal structure of DBP has now been described by several groups [17–21]. The three domains of DBP consist entirely of α-helices, 10 in domain 1, nine in domain 2, and four in domain 3. For a more detailed description of the tertiary structure of the DBP molecule, refer to Chapter 9.

Originally identified in 1959 by serum electrophoresis as a polymorphic protein, vitamin D binding protein (DBP) was referred to as the group-specific component of serum (Gc globulin) [1]. At that time its function was not known, although it became useful in population genetics [2] and forensic medicine [3]. DBP is structurally related to albumin and α-fetoprotein, and the DBP gene is a member of the linked albumin and α-fetoprotein gene family. Binding, solubilization, and serum transport of the vitamin D sterols have been considered the predominant functions of DBP, although there is now evidence that DBP may have a spectrum of biological activities, including roles in inflammation and the immune system. The following sections review the general structural and functional features of DBP. A more detailed review of how some of these structural features may explain DBP’s functional activities can be found in Chapter 9. Chapter 10 reviews the relationships among DBP, megalin, and cubilin in the regulation of a key step in vitamin D metabolism.

II. VITAMIN D BINDING PROTEIN AND ITS GENE A. Structural Features Serum DBP is a polymorphic, monomeric, serum α-globulin of approximately 58 kDa, its size being dependent upon glycosylation state [4]. The DNA sequences have been determined for the human [5,6] and rat [7] DBP structural genes. Rat and human DBP genes span 35 and 42 kb, respectively, and both contain 13 exons. The signal sequence of the protein is encoded on the first exon, whereas the last exon contains the entire noncoding 3′-untranslated region. Since the initial reports of human DBP cDNA structure [8,9], DBP cDNAs from the rat [10], mouse [11], rabbit [12], chicken [13], and a turtle, Trachemys scripta [14], have VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND, GLORIEUX

B. Evolutionary Relationships 1. CHROMOSOMAL LOCALIZATION AND LINKAGES

The human DBP gene was assigned to chromosome 4 by analysis of somatic cell hybrids and sublocalized to 4q11-q13 by in situ hybridization to metaphase chromosome spreads [22]. This sublocalization overlaps with the known positions of albumin and α-fetoprotein [23]. These three genes have been assigned to chromosome 13 in the rat [24] and to chromosome 5 in the mouse [11]. The rodent chromosomes encoding DBP are syntenic with human chromosome 4, thus demonstrating conservation of the DBP–α-fetoprotein–albumin linkage in three species. A linkage between DBP and albumin Copyright © 2005, Elsevier, Inc. All rights reserved.

118

CHRISTOPHER J. LAING AND NANCY E. COOKE

L

NH2

T

S

L

V

L

Q S

A

K

S C E

E

K

T

K N K

E

Y D R

G R E

C K

E

F

S H

L

G K

E D

F

A C C A

E

T

L

S

V

V

E

K

D P D C Y D

T

R

T

S

A

L

S

V S

L

V

L

Y

S R

K

V Q E

F

T G S

P

F G A

E

S N S

P

F

P V

G

DOMAIN I

L

E I

A S

V

Y

E R K

C C E L

C M

D N

T

P

E

T

P

F

E Q P Q H K

L

C E

A

F

R K D P

K

E

Y

W

S

T

C C S G V M S

L

Y

S

K

P

T

V C

F

A N Q F M

T G P H

A A E

Y

S

T

N

Y G

Y

S

V

L

L

S

L

P

A Q

S R

L

S N

L

I

K

L

A Q K

A

L

P

L

V D E

L

D A

T

V

K

L

C D N

L

S

T

K

T

K

E Q C C D E

F

K

S

L

K

E R

L

Q

L

K

V R N S

L

T

T

L

L

S

L

C S Q Y

A

A

Y G E

K

K

D E

T

A

H

V S

E C C K

S

L

I

N

T

I

S

E D C M A

K

E

L

P

E H

T

P

A N

DOMAIN II

A M D V K N

T

P

L

F

V C

T

Y

F

P

Q A

A

E

V D P

L

E

L

T

K

V M D K

Y

T

F

E G

L

S

K

L

T

P

E

L V

K

S

N A

K G P

L

L

K

E Q G K D

I

F

S

S

L

E

C A D Y

S

E N

T

F

T

E

Y

K

K

K

I

S C C N S

A

F

D S R K N V

L

K

A

P

L

E

I

D A

I

L

COOH

M P

D V C D P G N E D

V D C C S

T

T

C

F

E

L

S R R

L

F

V

E

P

L

L

A

E R

L

K

A

L

E

K

T

A D P

T H

K L K L

DOMAIN III N S

P

Y C D S

E

L

K N

P

FIGURE 1 Human DBP, primary structure. The amino acid sequence is indicated using single-letter amino acid abbreviations. Fewer than 25% of the amino acids are different in human, mouse, and rat DBP (shaded) [8–11]. The smaller circles mark the positions of amino acid residues present in albumin and α-fetoprotein, but absent in DBP [8]. The three internal domains defined by amino acid sequence similarities are indicated at left. Disulfide bond positions are marked, and the position of a disulfide bond present in domain 1 of DBP, but absent in both albumin and α-fetoprotein, is indicated by a distinctive line between residues 13 and 59. The positions of interruptions to the coding sequence by introns are indicated by arrowheads.

119

CHAPTER 8 Vitamin D–Binding Protein

20 kb upstream from the 5′ end of the α-fetoprotein gene [31,32]. The rat α-albumin gene is located 10 kb downstream from the α-fetoprotein locus, so that the order of the four genes from 5′ to 3′ with respect to the direction of albumin transcription is the same in the rat as in humans. Similar spatial relationships between the genes have also been reported in the mouse [33]. Thus, DBP is a member of a linked multigene family encoding at least four serum proteins, each of which is predominantly expressed in the fetal and/or adult liver (see Section III,A).

has also been noted in horses and in chickens [25,26]. A fourth gene in the DBP/albumin/α-fetoprotein family has been described by two groups and named afamin by one group [27] and α-albumin by the other [28]. α-Albumin is selectively expressed in the liver during the late stages of development, suggesting that it may be a phylogenetic intermediate between α-fetoprotein and albumin and may share some of the functions of these proteins [28,29]. Physical and meiotic mapping has determined that the organization of the gene family in humans is centromere–DBP–albumin–α-fetoprotein–α-albumin– telomere, the transcriptional orientation of DBP being opposite to the other members of the gene family (Fig. 2). The separation between DBP and the remaining members of the gene family is at least 1.5 Mb [30]. In contrast to the substantial distance between DBP and albumin, the albumin and α-fetoprotein genes are tightly linked in the mammalian species studied so far. In the rat and human, albumin is located only 14 to

2. EVOLUTION WITHIN THE GENE FAMILY

There is a striking pattern of structural similarity among the proteins and genes of the DBP, albumin, α-fetoprotein, and α-albumin family [6–12,15,16,20, 21,28]. The four mRNAs and encoded proteins have sequence similarities of approximately 40% at the nucleotide level and 25% at the amino acid level. The positioning of disulfide bond–forming cysteine residues

IN SITU LOCALIZATION DBP ALB/AFP/αALB

GENETIC LINKAGE (cM) IL8

1.9

ALB

1.0

DBP

>1.5 Mb

630 kb αALB AFP ALB

PHYSICAL MAP

(100 kb)

280 MY 300 MY HUMAN CHROMOSOME 4 560-600 MY

TRIPLE DOMAIN

SINGLE DOMAIN

FIGURE 2 The linked DBP multigene family. The chromosomal locations of DBP, albumin (ALB), α-fetoprotein (AFP), and α-albumin (αALB) as determined by in situ localization and recombination analyses are indicated [22,157]. Physical mapping has indicated the arrangement and transcriptional orientations of the genes to be as shown [30]. The evolutionary relationships among the members of the DBP/ALB/AFP-αALB multigene family, as predicted from amino acid and nucleic acid similarities, are indicated in the cladistic diagram [34–37]. Times of divergence are indicated in millions of years (MY). The physical maps and cladogram are not drawn to scale.

120 is nearly identical in all four proteins, predicting a shared three-domain secondary structure. Despite the amino acid sequence similarities between DBP and albumin, and the similarity in individual domain topologies, the orientation of the three domains is very different in the two proteins. This possibly contributes to the unique binding properties of DBP [20,21]. The carboxy-terminal domain of DBP is 124 amino acids shorter that those of albumin and α-fetoprotein. This is due to the loss of the exons homologous to 12 and 13 in the 15-exon albumin and α-fetoprotein genes, resulting in the 13-exon DBP gene [6,7]. With the exception of these two missing exons, the positions at which the introns interrupt the coding regions of the three genes are highly similar (Fig. 1). These striking structural parallels firmly establish the shared ancestry of the DBP, albumin, and α-fetoprotein genes. A number of models have been proposed to account for the evolution of the DBP multigene family and for the shared, triplicated internal domain structure. In one model, the ancestral internal domain is encoded by four exons that subsequently triplicated, creating the present gene encoding a three-domain protein. This domain triplication likely predated vertebrate evolution [34,35]. The separation of a unique DBP gene from this precursor is estimated to have occurred 560–600 million years ago [36,37] (Fig. 2). It has been postulated that albumin duplicated from the ancestral gene and began to diverge about 280 million years ago, just after the time of the amphibian/reptile divergence about 350 million years ago. The absence of a larval α-fetoprotein gene in amphibians and the existence of α-fetoprotein in chickens is consistent with this conclusion [36]. Duplication and subsequent evolution of the α-albumin and α-fetoprotein genes followed this [37]. Evidence of this divergence may be observed in a duplicated albumin gene in the adult amphibian and some reptiles [36,38], and the presence of α-fetoprotein in reptiles. It has been estimated that DBP has evolved at the slowest rate, and that α-fetoprotein and α-albumin have evolved at the fastest rates of the four proteins, although all have evolved at a more rapid rate than proteins of the globin family [36,37]. There has been significant invasion of noncoding sequences by repetitive DNA elements. Intron expansion in the DBP gene has been significantly greater compared with the other family members, providing further evidence of its longer period of separation from the ancestral gene [39]. Regardless of the precise details of this series of evolutionary events, DBP appears to be the oldest member of this ancient and well-conserved gene family. 3. DBP IN VERTEBRATE EVOLUTION

A specific protein carrier for vitamin D compounds has been identified in examples of all major vertebrate

CHRISTOPHER J. LAING AND NANCY E. COOKE

phyla [40–44]. The presence of a serum DBP in some species of cartilaginous fish (Cyprinus carpio), but not in others (Tilapia nilotica), suggests that the emergence of a discrete DBP may have occurred during the evolution of this phylum [41]. Overall the structures of the DBP gene, mRNA, and protein have been well conserved. There is over 70% amino acid homology among human, mouse, rat, rabbit, and T. scripta DBP. Furthermore, the affinity of reptilian, avian, and mammalian DBPs for the major vitamin D metabolite, 25-hydroxyvitamin D3, are all in the nanomolar range. However, despite this high degree of both structural and functional conservation, significant interspecies variations in DBP concentration and in DBP–ligand affinity exist. These variations appear to be more dependent on ecological variables such as geographic distribution and diet, than on phylogenetic distance [43,44]. Only mammalian DBP demonstrates equal affinity for metabolites of vitamin D2 and vitamin D3, suggesting that the ability to efficiently handle vitamin D2 metabolites may have conferred some advantage on emerging mammalian vertebrates [44]. An unusual circumstance has been observed in the emydid turtle T. scripta, where DBP is also able to transport thyroxine on a unique binding site. The DBP of this species constitutes the major carrier of vitamin D sterols, but its serum concentration is regulated by thyroxine. Comparative studies suggest that this dual binding is unique to members of Emydidae [45,46]. The case of emydid turtles notwithstanding, DBP is structurally and functionally highly conserved among vertebrates, and this, along with the apparent evolutionary responsiveness of its functional features to ecological factors, suggests an important physiological role for this protein.

C. Polymorphisms of DBP/Gc-Globulin Protein and Gene DBP (or group-specific component, Gc) was originally characterized in humans by serum electrophoresis as the product of two autosomal, codominant alleles (Gc1 and Gc2). Isoelectric focusing allowed the further characterization of slow (Gc1S) and fast (Gc1F) subtypes of Gc1 [47,48], resulting in six common phenotypes. The protein isoforms produced by the Gc1F and Gc1S alleles represent a mixed population containing or lacking a single N-acetylneuraminic acid on a threonine residue at position 420 [49]. In contrast, the Gc2 allele produces a single protein that contains a nonglycosylated lysine at position 420. The DBP gene contains an Alu middle repetitive element located at the end of DBP intron 8 resulting in four gene polymorphisms created by differences in the size of the Alu poly(A) tract. Each of these polymorphisms is equally distributed among

121

CHAPTER 8 Vitamin D–Binding Protein

the three common alleles [50]. Although the major alleles can be distinguished at the DNA level by restriction digestions or PCR-single-strand conformation polymorphism analysis [51], protein electrophoretic techniques remain the most usual initial approach to assigning protein isoform phenotypes. Although commonly used for forensic purposes in the past [3,52], DBP phenotypying is now more frequently used to map population dynamics and to assess phenotypic susceptibility to disease (see later discussion). In addition to the three common alleles, there are also more than 124 rare variant alleles described worldwide. The geographic occurrences of these variants often correspond to patterns of human population migrations and thus are of anthropological interest [2,53–56]. The molecular bases for some of these rare variants have been determined by sequencing of exons amplified by the polymerase chain reaction [57–59]. A number of genetic studies have suggested the presence of a DBP-null allele [60,61], but these have not been clearly distinguished from low-expressing “pseudo-silent” alleles [62]. The lack of a naturally occurring DBP-null homozygote in the human population suggested that such a genotype would be embryonically lethal. A homozygous, DBP-deficient mouse line has, however, been generated that is viable and fertile [63]. DBP polymorphisms have also been identified in nonhuman primates [64,65], rodents [66], a variety of ungulates [25,66–70], domestic cats [71], marsupials [72], and birds [26,73,74]. The high degree of polymorphic variation in the human population and its widespread occurrence among vertebrates make the DBP locus among the most polymorphic known. There is some suggestion that differences in serum concentration of DBP may be attributable to phenotypic isoform [75], and that different isoforms of DBP may exhibit affinity differences for vitamin D metabolites [76,77]. However, the biological significance, if any, of these differences remains to be established. DBP isoforms have also been reported to be associated with differences in susceptibility to a wide variety of diseases and syndromes, including multiple sclerosis [78,79], autoimmune thyroid disease [80], chronic obstructive pulmonary disease [81,82], low bone mineral density in women [83], and male osteoporosis [84]. A protein that is structurally related to DBP, and thus a possible variant of DBP, has been identified as a potential serum marker for prostate cancer [85]. The nature of this protein and its relationship to DBP await its isolation and further characterization. The associations between DBP isoform and disease are, for the most part, limited to specific geographic areas or ethnic groups. It is not clear whether they apply to the human population in general, or what, if any, the biological bases and implications of these associations may be.

III. FUNCTIONAL FEATURES OF VITAMIN D BINDING PROTEIN A. Synthesis and Turnover of DBP The circulating DBP concentration in adult mammals and birds is in the micromolar range [43,44]. DBP concentration increases in the postnatal period and can be influenced by a number of factors and processes. Most studies have failed to link vitamin D status or disturbances of mineral homeostasis with alterations to DBP concentration [86–89], although there is some evidence that there may be a correlation between ligand concentration and DBP concentration in the elderly [90]. DBP concentration increases during pregnancy [91,92], and under the influence of female sex hormones in humans and birds [93–95]. In rodents, DBP concentration increases in response to testosterone and is reported to be higher in males than in females [96]. Malnutrition results in a decrease in circulating DBP concentration [97], as does liver failure [91,98], pronounced proteinuria [99,100], and increased actin–DBP complexing resulting from tissue necrosis or damage (see Section III,C,2). The specific mechanisms by which these factors and processes regulate DBP concentration are not well understood. Like albumin and other serum proteins, DBP is produced primarily in the liver, and the increase in postnatal DBP concentration to adult levels is thought to reflect maturation of the liver protein synthesis processes [86,101]. It has been estimated that 10 mg/kg/day of DBP is produced in humans [102]. The relatively low concentration of hepatic DBP mRNA suggests that the DBP transcript, like that of albumin, may be very stable and exhibit a long cytosolic half-life [10]. Spliced DBP transcripts have been identified in a range of extrahepatic sites, although at concentrations approximately 100- to 1000-fold lower than those found in the liver [101]. The role of these extrahepatic transcripts is not known, as they are not thought to be in enough abundance to significantly influence the circulating concentration of DBP, and indeed it has not yet been determined whether they are translated. DBP production is up-regulated following its increased clearance in response to trauma or hepatic necrosis [103–105] (see Section III,C,2), raising the possibility that its synthesis is regulated by a positive feedback mechanism. A range of growth factors and steroids have been demonstrated to influence DBP production by cells in culture [106,107], although the mechanisms by which these factors regulate DBP synthesis are largely unknown. It is thought that the postnatal regulation of DBP production is largely at the level of gene transcription, as is the case for α-fetoprotein and albumin [108]. Recent work has identified a locus-control

122 region for the human DBP gene, located within its large first intron. The expression of a full-length human DBP transgene in mice is abolished by deletion of this region [109, and Hiroki et al., submitted, 2004]. Expression of the DBP gene is partially under the control of HNF-1α and HNF-1β, which bind to three cis elements near its proximal promoter. HNF-1α has a major stimulatory effect whereas HNF-1β is a transdominant inhibitor of HNF-1α stimulation [110]. Both are members of the POU-homeodomain family of transcription regulators [111]. Protein kinase C has been implicated in the regulation of α-fetoprotein and albumin gene expression in the liver; however, it has not been demonstrated to regulate DBP expression [112]. The regulatory mechanisms for DBP clearance from the circulation are not known. Clearance of DBP from the plasma in the rabbit, like that of albumin, is described by a multiexponential curve with an estimated t1/2 of 1.7 days. This is less than half that of albumin with a t1/2 of 5 days, but over five times faster than that of its major ligand, 25-hydroxyvitamin D3 (25(OH)D3), suggesting that ligand recycling is likely [113]. DBP is primarily cleared by the kidneys as a result of its uptake by megalin in the proximal tubule epithelium [102,114–116] (see Section III,B,3 of this chapter and Chapter 10 of this volume). DBP–actin complexes are cleared up to three times more rapidly than unliganded DBP, primarily by hepatic filtration [117,118] (see Section III,C,2). Its association with 25(OH)D3, on the other hand, has not been reported to affect the clearance rate of DBP [113]. It has been suggested that glycosylation state may be responsible for altering clearance mechanisms of other plasma proteins, including sex hormone–binding globulin (SHBG) [119]; however, differences in glycosylation of DBP (see Section II,C) have not been reported to influence its clearance from the circulation [113]. The structural similarities between, and shared ancestry of albumin, α-fetoprotein, and DBP originally suggested overlapping regulatory mechanisms of production and turnover. It is now clear, however, that regulation of DBP synthesis and catabolism may be quite independent, and these processes await further characterization.

B. Molecular Interactions DBP associates with a number of biological molecules, including vitamin D sterols, actin, fatty acids, and various cell membrane components. These interactions will be discussed in this section, and their proposed physiological significance considered in Section C. Some of these interactions are summarized in Table I.

CHRISTOPHER J. LAING AND NANCY E. COOKE

TABLE I Features and Binding Characteristics of DBP Features Isoelectric point Electrophoretic migration Size Plasma concentration Plasma half-life Daily production rate Altered plasma levels Increased Decreased Vitamin D sterol binding Plasma capacity Normal sterol occupancy Affinity (KD) ~nanomolar ~micromolar Clearance by Actin binding Plasma capacity Normal actin occupancy Affinity (KD) DBP-actin complexes Clearance by Cell associations Renal epithelium Cell “receptor” Internalization Physiology Lymphocytes Cell “receptor” Internalization Physiology Monocytes/macrophages Cell “receptor” Internalization Physiology Neutrophils Cell “receptor” Internalization Physiology

4.5–4.8 Post-albumin, inter-α-globulin 58 kDa, single-chain glycoprotein 4–8 µM (232–464 mg/liter) 2.5–3.0 days ~10 mg/kg Estrogen, pregnancy Nephrotic syndrome, liver disease, malnutrition mol/mol (2.4 mg D sterol/liter) <5% 25OHD, 24,25(OH)2D 1,25(OH)2D, vitamin D Renal proximal tubule epithelial cells mol/mol (270 mg/liter) Trace—? mg/liter Nanomolar Seen with tissue injuries, inflammation Hepatic phagocytic cells, sinusoidal endothelium

Megalin, cubilin Yes Protein clearance, delivery of ligand ?, IgG–Fc receptor complex ? ?, conversion to DBP-maf ? ? ?, activation of phagocytic function Chondroitin sulfate proteoglycan ?, no ?, enhancement of C5a chemotaxis

1. VITAMIN D LIGANDS

As well as having weak, nonspecific associations with lipoproteins and albumin, vitamin D sterols in the extracellular fluid form a specific, high-capacity, high-affinity association with DBP [120–122]. All naturally occurring and synthetic vitamin D sterols are bound by a single binding site, but with variable affinity [reviewed in 123–126]. Among the naturally occurring vitamin D metabolites, DBP shows the greatest affinity for 25(OH)D3 and 24,25-dihydroxyvitamin D3 (24,25(OH)2D3), followed by 1,25dihydroxyvitamin D3 (1,25(OH)2D3) and parent

123

CHAPTER 8 Vitamin D–Binding Protein

vitamin D3 [4,113,42,127,128] (Table I). The vitamin D–binding site of DBP was first localized to the amino terminus of the protein using biochemical techniques such as proteolysis [129,130] (Fig. 1). Molecular modeling and labeling studies have identified important contact points and molecular orientations of a variety of vitamin D sterols within the vitamin D–binding pocket [124]. The vitamin D–binding site of DBP imposes rigid steric restrictions on its ligands. The binding pocket of the nuclear vitamin D receptor shares this feature, which contrasts with many other ligand–protein interactions such as the estrogen receptor-binding site, within which ligands are relatively free to rotate [125]. However, the actual orientations of vitamin D ligands within the DBP and vitamin D receptor-binding pockets are very different. In the latter case, the sterol is completely surrounded by the receptor protein, whereas there is indication that the DBP binding site remains “open” and that part of the sterol molecule protrudes outside of the pocket [124–126,131] (see Chapter 15). Elucidation of the crystal structure of 25(OH)D3liganded DBP has provided a more detailed understanding of the molecular interaction between these two molecules [19,20]. These structural features will pave the way to understanding how different ligands are oriented within the vitamin D–binding space and are discussed in further detail in Chapter 9 of this volume. SHBG is known to bind estradiol and C19-androgen metabolites in different orientations resulting in distinct conformational changes in the binding protein, and possibly providing a structural explanation for differences in the biological activities of these ligands [132,133]. Whether similar binding differences are observed among the various vitamin D ligands of DBP remains to be fully determined. Vitamin D binding increases the chemical [134] and thermal [122] stability of DBP, diminishes its isoelectric point [134,135], and is now known to result in rotation of DBP domain I structures [20]. The association of vitamin D ligands with DBP is thought to facilitate their dispersal throughout the body, and their delivery to sites of metabolism and action (see Section III,C,1). 2. ACTIN

The ability of DBP to form a complex with actin was first observed during cytosolic contamination of preparations of DBP-containing extracellular fluid [136,137]. DBP binds globular, but not filamentous actin in a specific, saturable, high-affinity manner (Table I). It possesses a single binding site for actin that was originally localized to the C-terminal end of the molecule [130,138] (Fig.1), but which more recently, through elucidation of the crystal structure of DBP, has been identified as involving components

of all three DBP domains [17–19,21] (for more detail see Chapter 9). Binding of actin involves rotations of the domains and flexion of some of the intradomain extended structures within the DBP molecule [17]. The extent, if any, to which the binding of actin influences the affinity of DBP for vitamin D, or vice versa, remains unclear, although evidence thus far suggests that the binding sites are structurally and functionally independent [17,19,21,139]. DBP–actin complexes have been identified in normal human plasma and have been found to increase in concentration during pregnancy and following incidents of tissue damage and necrosis. This has prompted the proposal that DBP acts in concert with other actin-binding proteins, particularly gelsolin, as a component of an extracellular actin scavenging system [140–143] (see Section III,C,2). 3. CELL ASSOCIATIONS

DBP can associate with the plasmalemma of a wide variety of cells, including renal proximal tubule epithelium [114], lymphocytes [144,145], monocytes [144–147], neutrophils [148], yolk-sac endodermal cells [149], trophoblastic cells [150,151], and spermatocytes [152]. Despite the demonstration of DBP transcripts in a number of tissues other than liver (see Section III,A), it is generally thought that this cellsurface DBP is more likely to be acquired, at least partially, from the extracellular fluid, rather than completely produced de novo [147]. A search for a specific cellsurface DBP receptor has been thus far unsuccessful, although various nonspecific cell-surface structures have been found to interact with DBP (Table I). In the renal proximal tubule epithelium, megalin, a member of the low-density lipoprotein receptor gene family, and its co-receptor cubilin are responsible for binding a wide variety of molecules, including DBP [114,115,153]. DBP is thought to bind to the IgG-Fc receptor complex of lymphocytes [154], and an elastase-sensitive chondroitin sulfate proteoglycan has been implicated as a nonsaturable DBP-binding structure on neutrophil surfaces. This chondroitin sulfate, although not specifically identified, is thought to bind multiple DBP molecules [148,155]. Although a surface receptor on monocytes has yet to be identified, the existence of a monoclonal antibody for DBP that does not recognize monocyte-associated protein [146,156], suggests a discrete structural interaction of DBP with the cell surface. The fate and function of cell-associated DBP are not fully understood in most instances, although megalin-complexed DBP is thought to deliver 25-hydroxyvitamin D3 to intracellular metabolic enzymes (see Section III,C,1,b), and neutrophilassociated DBP is thought to be involved in regulation of chemotaxis (see Section III,C,2).

124 C. Physiological Roles of DBP DBP was originally thought to be an acute phase reactant until it was found to bind vitamin D sterols and the identity of Gc-globulin and DBP were confirmed [122,157,158]. However, the subsequent discovery of its range of seemingly unrelated ligand and cell associations has prompted the suggestion that DBP is a multifunctional protein [159]. Although actions of DBP in addition to the physiological economy of vitamin D have been well documented (see Section III,B), many of the biological roles suggested for DBP are speculative, or supported by circumstantial evidence. The reported functions of DBP fall into two main categories: those involving the physiological economy and function of vitamin D and calcium homeostasis, and those involved in inflammatory processes and the immune response. It is still not clear whether these functions of DBP are independent or are physiologically linked. 1. DBP AND THE PHYSIOLOGICAL ECONOMY FUNCTION OF VITAMIN D

AND

a. The Role of Plasma Proteins in Sterol Transport and Function The sterols, and their biologically functional subset, the steroid hormones, are a group of biological molecules that are structurally related to cholesterol. They include the sex hormones, corticosteroids, thyroid hormones, retinoids, and vitamin D molecules. These molecules function in a diverse range of tissues, which necessitates distribution throughout the body. Sterols are highly lipophilic, and as such disperse poorly in the aqueous environment of the extracellular space. They do, however, readily and rapidly diffuse across cell membranes. These shared limitations necessitate transport mechanisms that allow the dispersal of sterols through the circulation, and their controlled access to cells. These requirements are met partially by nonspecific carriers such as albumin, but more importantly by a heterogeneous group of ligandspecific protein carriers. In addition to DBP, this group of proteins includes ex-hormone binding globulin (SHBG), corticosteroid-binding globulin (CBG), thyroidbinding globulin (TBG), and retinol-binding protein (RBP). These carrier proteins, while structurally unrelated, share many common functional features, reflective of the shared biochemical and physiological features of their sterol ligands. This section summarizes these shared features, along with the changing views of the functional roles of carrier proteins in steroid physiology. The specific case of DBP and vitamin D transport is considered in Section III,C,1,b. The vast majority of circulating steroids (over 90%, and in many cases up to 99%) are found in association

CHRISTOPHER J. LAING AND NANCY E. COOKE

with their specific carrier proteins. The free hormone theory was developed to describe the manner in which these proteins disperse their ligands throughout the extracellular space and make them available to target cells [160,161]. According to this theory, the ease with which the steroids can diffuse across cell membranes means that it is the small, free (or unbound) fraction of steroid in the extracellular fluid that is biologically available to cells. As such, sterol carrier proteins are thought to function by sequestering their ligands and making them unavailable to cells (and hence ensuring their continued transit in the circulation). The association between a particular protein carrier and its ligand obeys the law of mass action, where bound and “free” species exist in the following equilibrium: Ligand + BP ↔ Ligand−BP The factors that determine the proportion of free (or bioavailable) sterol are the concentration of each of the equilibrium species and the affinity of the ligand for its carrier (as well as those of other ligands competing for the same site). The metabolic clearance rates of the sex steroids and the corticosteroids are inversely proportional to their affinities for their respective specific plasma binding proteins [162,163], providing experimental support for the free hormone theory. However, the free hormone hypothesis has been challenged by a number of observations that suggest it may not fully explain the relationship between sterols, their binding proteins, and target cells. One of the most important is the direct, tissue-specific association of binding proteins with cell membranes. A specific cell membrane receptor for SHBG has been well described in a number of cell types. The binding of specific ligands to the SHBG–receptor complex initiates second-messenger cascades resulting in alterations to cell growth and differentiation, and secretion of specific proteins [reviewed in 164–166]. Thus SHBG not only acts to deliver its ligands to target cells, but directly participates in a biologically functional process that is either independent of or that augments the classic genomic actions of steroid hormones [reviewed in 167]. It has been recognized that SHBG expression occurs in a range of extrahepatic sites, many of which are sex-steroid target cells and coexpress the SHBG receptor. This suggests that SHBG may act as an autocrine or paracrine factor [168,169]. Although less well described, and more hotly contested, evidence has been presented which suggests that CBG and RBP may also bind to cell membrane receptors in a tissue-specific fashion. The physiological consequences of such associations are still largely speculative [170–174].

CHAPTER 8 Vitamin D–Binding Protein

Our original views of sterol binding proteins ascribed to them a primarily passive role in the retention of their ligands in the circulation, thus creating a pool from which controlled cellular uptake could occur. They were thought to maintain adequate stores of their ligands in the face of intermittent supply, and to buffer against potentially toxic overload. This role for the binding proteins is certainly physiologically relevant. However, more recent descriptions of receptors and cell interactions for these proteins suggest a more diverse and active function in the physiological trafficking of sterols. The functional similarities within this structurally diverse group of binding proteins suggest that some of these themes may also be shared by DBP. b. The Case of Vitamin D–Binding Protein The free hormone theory has been suggested to describe the DBP–vitamin D–cell interaction [91,160,175], and a number of lines of experimental evidence support this idea. The metabolic clearance rates of vitamin D metabolites and analogs are inversely proportional to their affinity for DBP [123,176], and DBP has been shown to limit the uptake and action of 1,25(OH)2D3 in cells grown in vitro [177,178]. Analogs of 1,25(OH)2D3 with a low affinity for DBP stimulate vitamin D– responsive gene expression more rapidly than those with higher affinity [179], and mice not expressing DBP demonstrate a more rapid genomic response to 1,25(OH)2D3 than their wild-type littermates [63]. It has been proposed that the pathophysiology of vitamin D toxicity may involve the dislodging from DBP and resultant increased bioavailability of 1,25(OH)2D3 by the more abundant and avidly bound 25(OH)D3 [63,180]. Several experimental observations, however, suggest that the free hormone hypothesis may not fully explain the role of DBP in vitamin D transport and function. Its high plasma concentration and its high ligand affinity (Table I) ensure that DBP binds between 85% and 98% of each circulating vitamin D sterol, depending upon the metabolite, while the remainder is considered “free” or in association with low-affinity carriers [91,127,181]. On the other hand, only 5% of DBP molecules are occupied by a vitamin D sterol ligand [130]. Using a range of estimates of “free” 1,25(OH)2D3 concentration [91,127,176,181,182], the concentration of bioavailable 1,25(OH)2D3 in the extracellular fluid is still approximately an order of magnitude less than the dissociation constant of the intracellular vitamin D receptor. Thus if only the free fraction of hormone were available for cellular uptake, there would be very little likelihood of high receptor occupancy. This suggests that proteinmediated ligand uptake by cells may play an important role in the physiology of vitamin D metabolism and function.

125 Megalin–DBP complexes are internalized by the renal proximal tubular cells, and DBP is thought to undergo proteolysis after receptor and ligand scavenging. This results in the delivery of vitamin D sterols to intracellular binding sites such as metabolic enzymes and the vitamin D receptor, via specific cytosolic chaperones, members of the heat-shock protein-70 family [183]. This process constitutes an important step in the conversion of 25(OH)D3 to the biologically active 1,25(OH)2D3 [114,116] (see Chapter 10). The association of DBP with other cell types may involve further specific or nonspecific receptors (see Section III,C,2). These receptors may also be involved in ligand internalization, or may regulate cell signaling as has been described for cell surface receptors for SHBG (see Section III,C,1,a). 1,25(OH)2D3 and 24,25(OH)2D3 can exert effects in a variety of cells via rapid, nongenomic pathways thought to involve novel cell membrane receptors [184,185] (see Chapter 23). This is a theme that has also been described for sex hormones, which can stimulate cell-signaling cascades directly via sex-steroid membrane receptors, or via SHBG receptors [167]. It is tantalizing to speculate that some of the non-genomic actions of vitamin D metabolites may be mediated via interactions of DBP with cell surface receptors. The demonstration of DBP transcripts in a range of extrahepatic tissues (see Section III,A) suggests that DBP produced at these sites may be involved in functions not directly related to the dispersal of vitamin D ligands in the circulation and may act as an autocrine or paracrine factor as has been demonstrated for SHBG (see Section III,C,1,a). Further work is required to elucidate the mechanisms by which DBP regulates the cellular uptake, metabolism, and biological activities of its ligands. However, it is clear that whether the bioavailability of vitamin D is dependent upon the “free hormone” model, or protein-mediated pathways, or a combination of the two, DBP plays an important role in regulating the physiological economy of its sterol ligands. When fed a normal diet, mice in which the DBP locus has been inactivated do not demonstrate any specific phenotype that distinguishes them from wild-type littermates [63]. This “mild” phenotype is in contrast to mouse models in which the genes for the vitamin D receptor [186] or enzymes of metabolism of vitamin D [187,188] have been inactivated. Rather mild phenotypic differences have also been observed in mouse models and natural human mutations resulting in no or low expression of CBG [189,190], TBG [191,192], and RBP [193], suggesting that the mechanisms by which sterols arrive at and enter cells are likely to be tissue specific and to possess a certain element of redundancy. It has been suggested that the nonspecific carrier

126 proteins, such as albumin, may play a more than trivial role in the transport and processing of sterols [194]. However, in the case of both DBP–null [63] and RBPnull [193] mice, the absence of specific binding protein renders the animal more susceptible to derangements associated with abnormal intake or production of the respective ligand. The intermittent nature of vitamin D production or intake suggests that an important evolutionary advantage may be conferred upon organisms possessing an efficient plasma binding protein for the vitamin D sterols. That interspecies variations in DBP concentration and 25OHD3 affinity correlate more closely with ecological factors than with phylogenetic distance (see Section II,B,3) suggests that the properties of this protein have been important determinants of the conservation or activity of its vitamin D ligands during vertebrate evolution. Furthermore, there is some evidence to suggest that the concentration of DBP may be altered during times of increased calcium demand, such as at the initiation of lactation [44]. Possible links among malnutrition, DBP concentration, and the development of rachitic syndromes [97,195,196] also support the important role played by DBP in the regulation of the physiological economy of its vitamin D ligands. 2. THE ROLE OF DBP IN INFLAMMATORY PROCESSES SYSTEM

AND IN THE IMMUNE

Inflammation is a complex tissue reaction to physical injury, to infection, or to a local immune response and is characterized by changes to the local vasculature and the accumulation of fluid, biologically active proteins, and white blood cells. The pathogenesis and development of both local and systemic inflammation is intricately linked to these vascular and cellular responses. DBP has been associated with a number of steps within the inflammatory process, including the prevention of thromboembolic events in the microvasculature by actin, the stimulation of chemotaxis by phagocytic neutrophils, and the activation and stimulation of phagocytic function by macrophages. In this section, the evidence for a physiological role for DBP in the inflammatory process and in immune function is discussed. Tissue damage or necrosis, either due to or as a sequel to a primary insult, is an early component of the general inflammatory process and results in the release of intracellular proteins such as filamentous actin. Free actin in the circulation results in the development of thromboemboli in the microvasculature and the disruption of normal coagulation events [197]. This leads to tissue ischemia and multiple organ dysfunction syndrome. Actin may also be directly toxic to pulmonary

CHRISTOPHER J. LAING AND NANCY E. COOKE

endothelial cells [198]. DBP and another plasma protein, gelsolin, have been hypothesized to act in a coordinated fashion to dismantle actin filaments, sequester monomeric actin, and facilitate its clearance from the circulation [18,21,117,118,197]. In support of this hypothesis, circulating DBP concentrations are decreased and DBP–actin complexes increased in a variety of circumstances associated with tissue necrosis or cell disruption, such as fulminant hepatic necrosis [104,105,199], septic shock [200], and multiple trauma [103,201,202]. This reduced concentration of DBP is thought to be the result of the rapid removal of DBP–actin complexes in the liver [104,203] (see Section III,A). During recovery from liver disease or tissue trauma, plasma DBP concentration rises above the normal range, perhaps indicating a compensatory stimulation of DBP synthesis, which has been suggested to result from cytokine activity [102–105]. The priority for maintenance of plasma DBP concentration by the liver is similar to that of prothrombin index, and higher than that for plasma albumin concentration [98]. Nevertheless, the actin scavenging system appears to be saturable [141] and an association between low plasma DBP concentration and the development of multiple organ dysfunction has been documented in patients with fulminant hepatic necrosis and following severe injury [199,202]. An early response to tissue damage is the recruitment and activation of phagocytic cells. In the immune system, chemotaxis involves the directional migration of cells in response to an extracellular cue, usually in the form of a chemical gradient. A wide variety of chemoattractants have been identified, with varying degrees of specificity. A serum enhancer of C5a-stimulated chemotaxis of leukocytes was subsequently identified to be DBP. The mechanism(s) by which DBP enhances neutrophil response to C5a are unknown, but several studies document that such an effect exists [155,204–207]. Preincubation of DBP with neutrophils is sufficient for enhanced C5a-stimulated chemotaxis, which is apparently mediated via DBP–cell surface proteoglycan interaction [148,205,207] (see Section III,B,3). Although cell membrane–associated proteoglycans have been suggested to facilitate the uptake of their ligands [208], DBP bound to the surface of neutrophils does not appear to be internalized [155]. Some cell-surface proteoglycans are known to possess cytoplasmic domains that interact with components of the cytoskeleton [208], raising the possibility that the DBP–chondroitin sulfate complex is involved in cell signaling. It is now known that a neutrophil-associated serine protease, elastase, establishes the steady-state binding of DBP to neutrophil surfaces by cleaving and thus regulating the number of available chondroitin-sulfate binding sites. The establishment of steady-state binding is essential

127

CHAPTER 8 Vitamin D–Binding Protein

for cochemotactic activity, suggesting that the mechanism by which DBP acts as a cochemotaxin may involve dynamic proteoglycan signaling in the cytosol, or that localization of the DBP–receptor complex to active areas of the cell membrane is required [155]. Antigenic stimulants cause the activation of phagocytic function in leukocytes and other cells via several pathways. In one such pathway, cell death and inflammation increase the activity of membrane phospholipases, resulting in the accumulation of lysophospholipids and other phospholipid degradation products, which are potent stimulators of macrophage activation [209,210]. It has been demonstrated ex vivo and in vitro that this activation may occur through an intermediate macrophage-activating factor, the partially deglycosylated DBP (DBP-maf). DBP is converted to DBP-maf via the inducible, stepwise actions of B-lymphocytic β-galactosidase and T-lymphocytic sialidase [211–213]. DBP-maf activity is specific to Fcγ-mediated phagocytosis and does not appear to be involved in complementprimed phagocytic processes that can be stimulated by such factors as lipopolysaccharide [210]. DBP-maf activated macrophages demonstrate significant tumoricidal activity [214–216]. The specific stimulants for DBP-maf generation and the particular phagocytic pathways that it activates suggest that the principal role of DBP-maf may be in the immune response to neoplasia. The exact mechanism by which DBP-maf acts is unknown; however, the complete deglycosylation of DBP-maf by the removal of the terminal alphaN-acetylgalactosamine (GalNAc) moiety results in diminished macrophage-activating activity. Furthermore, the accumulation of tumor-derived alpha-N-acetylgalactomanidase in advanced cancer patients has been linked to immunosuppression. This suggests that DBPmaf may stimulate Fcγ translocation to the cell surface via GalNAc-specific lectin on the macrophage surface [217–219]. DBP maf has also been reported to stimulate bone-resorptive activities of osteoclasts in vitro [220]. It has been proposed that a defect in lymphocytic β-galactosidase and/or some other deficiency in DBPmaf generation may be involved in the pathogenesis of some osteopetrosis syndromes [221–223]. Most recently, it has been suggested that DBP–maf may have direct antiangiogenic effects on endothelial cells and be able to inhibit the angiogenic actions of a number of tumor-derived growth factors. This activity is proposed to operate via the CD36 receptor, expressed on both macrophages and endothelial cells [224,225], although the nature of this proposed activity awaits more thorough characterization. Given the seemingly diverse and integral roles played by DBP in inflammatory and immune processes, it was surprising that mice in which DBP expression had

been disrupted were viable [63]. Indeed, DBP-null mice show no obvious phenotype that can be related to deficient immune function. No differences in phagocytic cell recruitment in response to intraperitoneal treatment with thioglycollate medium [226] or to infection with Listeria monocytogenes or Toxoplasma gondii [White et al., submitted, 2004.] could be attributed to the inactivated DBP allele. DBP-null mice do not demonstrate osteopetrotic lesions [63], causing some uncertainty about the obligatory nature of DBP-maf in osteoclastic function. The reasons for these apparent discrepancies between in vitro or ex vivo actions of DBP and observations in the DBP-null mouse are as yet unclear. The processes of immunity and inflammation reflect complex genetic determinants. Strain differences in peritoneal recruitment of leukocytes have been identified in mice [226], and species differences are likely to be more profound still. Furthermore, phagocyte site specificity, as well as the nature of the target particles, may differentially determine the nature and regulation of phagocytic processes [227]. Therefore, the lack of observed phenotypic difference between DBP-null and wild-type mice in relation to inflammation and immunity may reflect species-, strain-, tissue-, and/or pathologyspecific factors. Immune defects related to the inactivated DBP allele, if any, may not become apparent without producing the appropriate specific challenges to the immune system. Alternatively, it is possible that the “non–vitamin D binding” actions of DBP may not represent an obligate role for DBP in inflammation and immunity, or may be biologically redundant. A third possibility is that they may be involved in an as-yetunidentified integrated physiological role for DBP. Further experimentation is required to resolve these points.

IV. CONCLUDING REMARKS Several important technical steps have advanced our understanding of the structural and functional features of DBP. The elucidation of the sequences of both the full-length DBP gene and its messenger RNA in a variety of vertebrates has revealed the relationship between this protein and the albumin/α-fetoprotein multigene family, and their place in vertebrate evolution. Comparisons of the primary structures of these proteins have identified differences in DBP that pointed toward a structural explanation for its unique functional activities. Descriptions of the crystal structure of DBP in complex with two of its ligands, 25(OH)D3 and actin, have extended this explanation to the tertiary structural level. The generation of a mouse line with an inactivated DBP locus has provided an important tool for the study

128 of the integrated physiology of DBP. It has enabled the demonstration of the important role played by DBP in the turnover and actions of its vitamin D ligands. Further investigation of non–vitamin D functions of DBP using this mouse model are underway. While our understanding of its actions and their molecular bases is increasing, further work is required to clarify the physiological role(s) of DBP in the whole organism. Functional similarities reported among the other sterol binding proteins may hold important clues for future investigations of the physiology of DBP.

References 1. Hirschfeld J, Jonsson B, Rasmusson M 1959 Immune electrophoretic demonstration of qualitative differences in human sera and their relation to the haptoglobins. Acta Pathol Microbiol 47:160–168. 2. Cleve H, Constans J 1988 The mutants of the vitamin D–binding protein: more than 120 variants of the Gc/DBP system. Vox Sang 54:215–225. 3. Westwood WA, Werret DJ 1986 Group specific component: a review of the isoelectric focusing methods and auxiliary methods available for the separation of its phenotypes. Forensic Sci Int 32:135–150. 4. Haddad JG, Walgate J 1976 25-Hydroxyvitamin D transport in human plasma. Isolation and partial characterisation of calcifidiol-binding protein. J Biol Chem 251:4803–4809. 5. Braun A, Kofler A, Morawietz S, Cleve H 1993 Sequence and organization of the human vitamin D–binding protein gene. Biochim Biophys Acta 1216:385–394. 6. Witke FW, Gibbs PEM, Zielinski R, Yang F, Bowman BH, Dugaiczyk A 1993 Complete structure of the human Gc gene: differences and similarities between members of the albumin gene family. Genomics 16:751–754. 7. Ray K, Wang XK, Zhao M, Cooke NE 1991 The rat vitamin D binding protein (Gc globulin) gene: structural analysis, functional and evolutionary correlations. J Biol Chem 266: 6221–6229. 8. Cooke NE, David EV 1985 Serum vitamin D binding protein is a third member of the albumin and α-fetoprotein gene family. J Clin Invest 76:2420–2424. 9. Yang F, Brune JL, Naylor SL, Cupples RL, Naberhaus KH, Bowman BH 1985 Human group-specific component (Gc) is a member of the albumin family. Proc Natl Acad Sci USA 82:7994–7998. 10. Cooke NE 1986 Rat vitamin D binding protein: determination of the full length primary structure from cloned cDNA. J Biol Chem 261:3441–3450. 11. Yang F, Bergeron JM, Linehan LA, Lalley PA, Sakaguchi AY, Bowman BH 1990 Mapping and conservation of the group-specific component gene in mouse. Genomics 7: 509–516. 12. Osawa M, Tsuji T, Yukawa N, Saito T, Takeichi S 1994 Cloning and sequence analysis of cDNA encoding rabbit vitamin D binding protein (Gc globulin). Biochem Mol Biol Int 34:1003–1009. 13. Burnside J, Sofer L 1999 Gallus gallus vitamin D binding protein cDNA. GenBank direct submission; accession number AF121350.

CHRISTOPHER J. LAING AND NANCY E. COOKE

14. Hunt JL, Licht P 1998 Identification and structural characterization of a novel member of the vitamin D binding protein family. Comp Biochem Physiol 121B:397–406. 15. Schoentgen F, Metz-Boutigue MH, Jolles J, Constans J, Jolles P 1986 Complete amino acid sequence of human vitamin D binding protein (group-specific component): evidence of a three fold internal homology as in serum albumin and alphafetoprotein. Biochim Biophys Acta 871:189–198. 16. Litwiller R, Fass R, Kumar R 1987 The amino acid sequence of the NH2-terminal portion of the rat and human vitamin D binding protein: evidence for a high degree of homology between rat and human vitamin D binding protein. Life Sci 38:2179–2184. 17. Head JF, Swamy N, Ray R 2002 Crystal structure of the complex between actin and human vitamin D binding protein at 2.5Å resolution. Biochemistry 41:9015–9020. 18. Otterbein LR, Cosio C, Graceffa P, Dominguez R 2002 Crystal structures of the vitamin D–binding protein and its complex with actin: structural basis of the actin-scavenger system. Proc Natl Acad Sci USA 99:8003–8008. 19. Swamy N, Head JF, Weitz D, Ray R 2002 Biochemical and preliminary crystallographic characterization of the vitamin D sterol- and actin-binding by human vitamin D–binding protein. Arch Biochem Biophys 402:14–23. 20. Verboven C, Rabijns A, De Maeyer M, Van Baelen H, Bouillon R, De Ranter C 2002 A structural basis for the unique binding features of the human vitamin D–binding protein. Nature Struct Biol 9:131–136. 21. Verboven C, Bogaerts I, Waelkens E, Rabijns A, Van Baelen H, Bouillon R, De Ranter C 2003 Actin–DBP: the perfect structural fit? Acta Crystallogr D Biol Crystallogr 59:263–273. 22. Cooke NE, Williard HF, David EV, George DL 1986 Direct regional assignment of the gene for vitamin D binding protein (Gc-globulin) to human chromosome 4q11–q13 and identification of an associated DNA polymorphism. Hum Genet 73:225–229. 23. Harper ME, Dugaiczyk A 1983 Linkage of the evolutionarilyrelated serum albumin and α-fetoprotein genes with q11–q22 of human chromosome 4. Am J Hum Genet 35:565–572. 24. Cooke NE, Levan G, Szpirer J 1987 The rat vitamin D binding protein (Gc-globulin) gene is syntenic with the rat albumin and α-fetoprotein genes on chromosome 14. Cytogenet Cell Genet 44:98–100. 25. Weitkamp LR, Allen PZ 1979 Evolutionary conservation of equine Gc alleles and of mammalian Gc/albumin linkage. Genetics 92:1347–2272. 26. Juneja RK, Sandberg K, Andersson L, Gahne B 1982 Close physical linkage between albumin and vitamin D binding protein (Gc) loci in chicken: a 300 million year old linkage group. Genet Res Cambridge 40:95–98. 27. Lichenstein HS, Lyons DE, Wurfel MM, Johnson DA, McGinley MD, Leidli JC, Trollinger DB, Mayer JP, Wright SD, Zukowski MM 1994 Afamin is a new member of the albumin, α-fetoprotein, and vitamin D–binding protein gene family. J Biol Chem 269:18149–18154. 28. Belanger L, Roy S, Allard D 1994 New albumin gene 3′ adjacent to the α1-fetoprotein locus. J Biol Chem 269:5481–5484. 29. Gabant P, Forrester L, Nichols J, Van Reeth T, De Mees C, Pajack B, Watt A, Smitz J, Alexandre H, Szpirer C, Szpirer J 2002 Alpha-fetoprotein, the major fetal serum protein, is not essential for embryonic development but is required for female fertility. Proc Natl Acad Sci USA 99:12865–12870. 30. Song Y-H, Naumova AK, Liebhaber SA, Cooke NE 1999 Physical and meiotic mapping of the region of human chromosome 4q11–q13 encompassing the vitamin D binding

CHAPTER 8 Vitamin D–Binding Protein

31. 32.

33. 34. 35. 36.

37. 38. 39. 40. 41. 42. 43.

44. 45.

46.

47. 48.

49.

protein DBP/Gc-globulin and albumin multigene cluster. Genome Res 9:581–587. Urano Y, Sakai M, Watanabe K, Tamaoki T 1984 Tandem arrangement of the albumin and alpha-fetoprotein genes in the human genome. Gene 32:255–261. Chevrette M, Guertin M, Turcotte B, Belanger L 1987 The rat α1-fetoprotein gene: characterization of the 5′-flanking region and tandem organization with the albumin gene. Nucleic Acids Res 15:1338–1339. The Jackson Laboratory, Bar Harbor, Maine 2003 Mouse Genome Informatics Web Site. URL: http://www.informatics.jax. org. August. Eiferman FA, Young PR, Scott RW, Tilghman SM 1981 Intragenic amplification and divergence in the mouse α-fetoprotein gene. Nature 294:713–718. Gibbs EM, Dugaiczyk A 1987 Origin of structural domains of the serum albumin gene family and a predicted structure of the gene for vitamin D–binding protein. Mol Biol Evol 4:364–379. Haefliger DN, Moskaitis JE, Schoenberg DR, Wahli W 1989 Amphibian albumins as members of the albumin, alpha-fetoprotein, vitamin D–binding protein multigene family. J Mol Evol 29:344–354. Gibbs PEM, Witke WF, Dugaiczyk A 1998 The molecular clock runs at different rates among closely related members of a gene family. J Mol Evol 46:552–561. Brown MA, Chambers GK, Licht P 1997 Purification and partial amino acid sequences of two distinct albumins from turtle plasma. Comp Biochem Physiol 118B:367–374. Nishio H, Hamdi HK, Dugaiczyk A 1999 Genomic expansion across the albumin gene family on human chromosome 4q is directional. Biol Chem 380:1431–1434. Hay AWM, Watson G 1976 The plasma transport of 25-hydroxycholecalciferol in fish, amphibians, reptiles and birds. Comp Biochem Physiol 53:167–172. Van Baelen H, Allewaert K, Bouillon R 1988 New aspects of the plasma carrier for 25-hydroxycholecalciferol in vertebrates. Ann NY Acad Sci 538:60–68. Laing CJ, Fraser DR 1999 The vitamin D system in iguanian lizards. Comp Biochem Physiol 123B:373–379. Laing CJ, Shea GM, Fraser DR 1999 Variation in the concentration and binding affinities of the plasma vitamin D–binding protein. In: Danks J, Dacke C, Flik G, Gay C (eds) Calcium Metabolism: Comparative Endocrinology. BioScientifica, Bristol, UK, pp. 93–98. Laing CJ 2000 Comparative Biology of Plasma Vitamin D Binding Protein. Ph.D. thesis. Department of Animal Science, University of Sydney, Sydney, Australia (http://adt.caul.edu.au). Licht P 1994 The relation of the dual thyroxine/vitamin D– binding protein (TBP/DBP) of emydid turtles to the vitamin D– binding proteins of other vertebrates. Gen Comp Endocrinol 94:215–224. Whitworth DJ, Hunt L, Licht P 2000 Widespread expression of the mRNA encoding a novel vitamin D/thyroxine dual binding protein in the turtle Trachemys scripta. Gen Comp Endocrinol 118:354–358. Constans J, Viau M 1977 Group-specific component: evidence for two subtypes of the Gc1 gene. Science 198:1070–1071. Hoste B 1979 Group specific component (Gc) and transferrin (Tf) subtypes ascertained by isoelectric focusing. A simple non-immunological staining procedure for Gc. Hum Genet 50:75–79. Cleve H, Patutschnick W 1979 Neuraminidase treatment reveals sialic acid differences in certain genetic variants of the Gc system (vitamin D binding protein). Hum Genet 47:193–198.

129 50. Braun A, Bichlmaier R, Muller B, Cleve H 1993 Molecular evaluation of an Alu repeat including a polymorphic variable poly(dA) (AluVpA) in the vitamin D binding protein (DBP) gene. Hum Genet 90:526–532. 51. Dissing J, Thymann M, Hopkinson D 2003 Simultaneous detection of ACP1 and GC genotypes using PCR/SSCP. Ann Hum Genet 67:81–85. 52. Thymann M 1981 Gc subtypes determined by agarose isoelectric focusing. Distribution in Denmark, and application to paternity cases. Hum Hered 31:214–221. 53. Lai LYC, Constans J, Archer GT 1990 A new cathodal Gc variant in Australia. Ann Hum Biol 17:526–532. 54. Chen LZ, Easteal S, Board PG, Summers KM, Bhatia KK, Kirk RL 1990 Albumin–vitamin D–binding protein haplotypes in Asian-Pacific populations. Hum Genet 85:89–97. 55. Kravchuk OI, Balanovskii OP, Nurbaev SD, Makarov SV, Spitsyn VA, Ginter EK 1998 Genogenography of the aboriginal population of Marii El (from data on immunobiochemical polymorphism). Genetika 34:1542–1554. 56. Yin J, Zhang G, Liu J 2000 Genetic polymorphism of groupspecific component/vitamin-D–binding protein subtypes in six populations from north-eastern China. Anthropol Anzeiger 58:193–198. 57. Kofler A, Braun A, Jenkins T, Serjeanston SW, Cleve H 1995 Characterization of mutants of the vitamin D–binding protein/ group specific component: GC aborigine (1A1) from Australian aborigines and South African blacks, and 2A9 from south Germany. Vox Sang 68:50–54. 58. Yuasa I, Kofler A, Braun A, Bichlmaier R, Kammerer S, Cleve H 1995 A novel sequence polymorphism in exon 8 of the human vitamin D–binding protein (GC) gene in an African population. Jap J Hum Genet 40:149–151. 59. Yuasa I, Kofler A, Braun A, Umetsu K, Bichlmaier R, Kammerer S, Cleve H 1995 Characterization of mutants of the vitamin D–binding protein/group-specific component: molecular evolution of GC*1A2 and GC*1A3, common in some Asian populations. Hum Genet 95:507–512. 60. Dykes D, Polesky H, Cox E 1981 Isoelectric focusing of Gc (vitamin D binding globulin) in parentage testing. Hum Genet 58:174–175. 61. Yoshifumi Y, Nishimoto H, and Ikemoto S 1989 Interstitial deletion of the proximal long arm of chromosome 4 associated with father-child incompatibility within the Gc-system: probable reduced gene dosage effect and partial piebald trait. Am J Med Genet 32:520–523. 62. Vavrusa B, Cleve H, Constans J 1983 A deficiency mutant of the Gc system. Hum Genet 65:102–107. 63. Safadi FF, Thornton P, Magiera H, Hollis BW, Gentile M, Haddad JG, Liebhaber SA, Cooke NE 1999 Osteopathy and resistance to vitamin D toxicity in mice null for vitamin D binding protein. J Clin Invest 103:239–251. 64. Cleve H, Patutschnick W 1979 Different phenotypes of the group-specific component (Gc) in chimpanzees. Hum Genet 50:217–220. 65. Tsuchida S, Nagai A, Ikemoto S 1990 Polymorphisms of the group-specific component (Gc) in three species of Tamarins (Sanguinus oedipus, S. labiatus, S. mystax). Jpn J Vet Sci 52:649–651. 66. Imawari M, Akanuma Y, Muto Y, Itakura H, Kosaka K 1980 Isolation and partial characterization of two immunologically similar vitamin D–binding proteins in rat serum. J Biochem 88:349–360. 67. Gahne B, Juneja RK 1978 Polymorphic post-albumin of cattle and horse plasma identified as vitamin D binding

130

68. 69. 70. 71. 72. 73. 74. 75. 76.

77. 78.

79. 80.

81.

82.

83.

84.

protein (Gc protein). Anim Blood Groups Biochem Genet 9:37–40. Tate ML, Dodds KG, Thomas KJ, McEwan KM 1992 Genetic polymorphism of plasminogen and vitamin D binding protein in red deer (Cervus elephus L). Anim Genet 23:303–313. Vankan DM, Bell K 1992 Genetic polymorphism of plasma vitamin D–binding protein (Gc) in Australian goats. Anim Genet 23:457–462. Shin JA, Yang YH, Kim HS, Yun YM, Lee KK 2002 Genetic polymorphism of the serum proteins of horses in Jeju. J Vet Sci 3:255–263. Juneja RK, Niini T, Larsson HEB, Sandberg K 1991 Genetic polymorphism of six plasma proteins in the domestic cat. J Hered 82:178–181. Arthur H, Bell K, Cooper DW 1998 Plasma protein polymorphisms in the tammar wallaby (Macropus eugenii). Aust J Zool 46:193–199. Juneja RK, Gahne B, Kuryl J, Gasparka J 1982 Genetic polymorphism of the vitamin D–binding protein and a pre-transferrin in chicken plasma. Hereditas 98:89–96. Shibata T, Abe T 1996 Linkage between the loci for serum albumin and vitamin D binding protein (Gc) in the Japanese quail. Anim Genet 27:195–197. Lauridsen AL, Vestergaard P, Nexo E 2001 Mean serum concentration of vitamin D-binding protein (Gc globulin) is related to the Gc phenotype in women. Clin Chem 47:753–756. Braun A, Brandhofer A, Cleve H 1990 Interaction of the vitamin D–binding protein (group-specific component) and its ligand 25-hydroxy-vitamin D3: binding differences of the various genetic types disclosed by isoelectric focusing. Electrophoresis 11:478–483. Arnaud J, Constans J 1993 Affinity differences for vitamin D metabolites associated with the genetic isoforms of the human serum carrier protein (DBP). Hum Genet 92:183–188. Arnason A, Jensson O, Skaftadottir I, Birgisdottir B, Gudmundsson G, Johannesson G 1980 HLA types, Gc protein and other genetic markers in multiple sclerosis and two other neurological diseases in Iceland. Acta Neurol Scand 62(Suppl 78):39–40. Niino M, Kikuchi S, Fukazawa T, Yabe I, Tashiro K 2002 No association of vitamin D–binding protein gene polymorphisms in Japanese patients with MS. J Neuroimmunol 127:177–179. Pani MA, Regulla K, Segni M, Hoffmann S, Hufner M, Pasquino AM, Usadel KH, Badenhoop K 2002 A polymorphism within the vitamin D–binding protein gene is associated with Graves’ disease but not with Hashimoto’s thyroiditis. J Clin Endocrinol Metab 87:2564–2567. Ishii T, Keicho N, Teramoto S, Azuma A, Kudoh S, Fukuchi Y, Ouchi Y, Matsuse T 2001 Association of Gc-globulin variation with susceptibility to COPD and diffuse panbronchitis. Eur Respir 18:753–757. Sandford AJ, Chagani T, Weir TD, Connett JE, Anthonisen NR, Pare PD 2001 Susceptibility genes for rapid decline of lung function in the lung health study. Am J Resp Crit Care Med 163:469–473. Ezura Y, Nakajima T, Kajita M, Ishida R, Inoue S, Yoshida H, Suzuki T, Shiraki M, Hosoi T, Orimo H, Emi M 2003 Association of molecular variants, haplotypes, and linkage disequilibrium within the human vitamin D–binding protein (DBP) gene with postmenopausal bone mineral density. J Bone Miner Res 18:1642–1649. Papiha SS, Allcroft LC, Kanan RM, Francis RM, Datta HK 1999 Vitamin D binding protein gene in male osteoporosis: association of plasma DBP and bone mineral density with

CHRISTOPHER J. LAING AND NANCY E. COOKE

85.

86.

87. 88. 89. 90.

91.

92.

93. 94.

95. 96. 97. 98.

99.

100. 101.

(TAAA)(n)-Alu polymorphism in DBP. Calcif Tissue Int 65:262–266. Hlavaty JJ, Partin AW, Shue MJ, Mangold LA, Derby J, Javier T, Kelley S, Stieg A, Briggman JV, Hass GM, Wu Y-J 2003 Identification and preliminary clinical evaluation of a 50.8-kDa serum marker for prostate cancer. Urology 61:1261–1265. Haddad JG, Walgate J 1976 Radioimmunoassay of the binding protein for vitamin D and its metabolites in human serum. Concentrations in normal subjects and patients with disorders of mineral homeostasis. J Clin Invest 67:1550–1560. Bouillon R, Van Baelen H, DeMoor P 1977 The measurement of vitamin D–binding protein in human serum. J Clin Endocrinol Metab 45:225–231. Imawari M, Goodman DS 1977 Immunological and immunoassay studies of the binding protein for vitamin D and its metabolites in human serum. J Clin Invest 59:432–442. Parviainen M, Harmoinen A, Visakorpi J 1982 Immunonephelometric quantification of serum vitamin D–binding globulin. Scand J Clin Lab Invest 42:571–575. Rejnmark L, Lauridsen AL, Vestergaard P, Heickendorff L, Andreasen F, Mosekilde L 2002 Diurnal rhythm of plasma 1,25-dihydroxyvitamin D and vitamin D–binding protein in postmenopausal women: relationship to plasma parathyroid hormone and calcium and phosphate metabolism. Eur J Endocrinol 146:635–642. Bikle DD, Gee E, Halloran B, Haddad JG 1984 Free 1,25dihydroxyvitamin D levels in serum from normal subjects, pregnant subjects, and subjects with liver disease. J Clin Invest 74:1966–1971. Abbas SK, Care AD, Van Baelen H, Bouillon R 1987 Plasma vitamin D–binding protein and free 1,25-dihydroxyvitamin D3 index in pregnant ewes and their fetuses in the last month of gestation. J Endocrinol 115:7–12. Fraser DR, Emtage JS 1976 Vitamin D in the avian egg. Its molecular identity and mechanism of incorporation into yolk. Biochem J 160:671–682. Cheema C, Grant BF, Marcus R 1989 Effects of estrogen on circulating “free” and total 1,25-dihydroxyvitamin D and on parathyroid–vitamin D axis in postmenopausal women. J Clin Invest 83:537–542. Harris SS, Dawson-Hughes B 1998 The association of oral contraceptive use with plasma 25-hydroxyvitamin D levels. J Am Coll Nutr 17:282–284. Bouillon R, Vandoren G, Van Baelen H, DeMoor P 1978 Immunochemical measurement of the vitamin D–binding protein in rat serum. Endocrinology 102:1710–1715. Laing CJ, Fraser DR 2002 Changes with malnutrition in the concentration of plasma vitamin D–binding protein (DBP) in growing rats. Br J Nutr 88:133–139. Schiodt FV, Bondesen S, Muller K, Rasmussen A, Hjortrup A, Kirkegaard P, Hansen BA, Tygstrup N, Ott P 1999 Reconstitution of the actin-scavenger system after orthotopic liver transplantation for end-stage liver disease: a prospective and longitudinal study. Liver Transplant Surg 5:310–317. Nyomba BL, Bouillon R, Lissens W, Van Baelen H, De Moor P 1985 1,25-Dihydroxyvitamin D and vitamin D–binding protein are both decreased in streptozotocin-diabetic rats. Endocrinology. 116:2483–2488. Auwerx J, De Keyser L, Bouillon R, De Moor P 1986 Decreased free 1,25-dihydroxycholecalciferol index in patients with the nephrotic syndrome. Nephron 42:231–235. McLeod JF, Cooke NE 1989 The vitamin D–binding protein, α-fetoprotein, albumin multigene family: detection of transcripts in multiple tissues. J Biol Chem 264:21760–21769.

CHAPTER 8 Vitamin D–Binding Protein

102. Kawakami M, Blum CB, Ramakrishan R, Dell RB, Goodman DS 1981 Turnover of the plasma binding protein for vitamin D and its metabolites in normal human subjects. J Clin Endocrinol Metab 53:1110–1116. 103. Dahl B, Schiodt FV, Rudolph S, Ott P, Kiaer T, Heslet L 2001 Trauma stimulates the synthesis of Gc-globulin. Inten Care Med 27:394–399. 104. Schiodt FV, Ott P, Tygstrup N, Dahl B, Bondesen S 2001 Temporal profile of total, bound, and free Gc-globulin after acetaminophen overdose. Liver Transplant 7:732–738. 105. Schiodt FV, Clemmesen JO, Bondesen S, Dahl B, Ott P 2001 Increased turnover of Gc-globulin in patients with hepatic encephalopathy. Scand J Gastroenterol 36:998–1003. 106. Egawa T, Ito H, Nakamura H, Yamamoto H, Kishimoto S 1992 Hormonal regulation of vitamin D–binding protein production by a human hepatoma cell line. Biochem Int 28:551–557. 107. Guha C, Osawa M, Werner PA, Galbraith RM, Paddock GV 1995 Regulation of human Gc (vitamin D–binding) protein levels: hormonal and cytokine control of gene expression in vitro. Hepatology 21:1675–1681. 108. Danan JL, Ribiero V, Lechner MC, Pompom D 1991 Expression of the rat vitamin D–binding protein gene. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D. Gene Regulation, Structure-Function Analysis and Clinical Application. Proceedings of the Eighth Workshop on Vitamin D. Walter de Gruyter, New York, pp. 697–698. 109. Hiroki T, Song Y-H, Naumova AK, Liebhaber SA, Cooke NE 2000 A human vitamin D binding protein gene with limited flanking sequences is sufficient for liver-specific and copy number-dependent expression in transgenic mice. The Endocrine Society, 82nd Annual Meeting, Toronto, June 21–25, (poster). 110. Song Y-H, Ray K, Liebhaber SA, Cooke NE 1998 Vitamin D– binding protein gene transcription is regulated by the relative abundance of hepatocyte nuclear factors 1α and 1β. J Biol Chem 273:28408–28418. 111. Chi YI, Frantz JD, Oh BC, Hansen L, Dhe-Paganon S, Shoelson SE 2002 Diabetes mutations delineate an atypical POU domain in HNF-1alpha. Mol Cell 10: 1129–1135. 112. Roncero C, Ventura J-J, Sanchez A, Bois-Joyeux B, Mesa ML, Thomassin H, Danan J-L, Benito M, Fabregat I 1998 Phorbol esters down-regulate α-fetoprotein gene expression without affecting growth in fetal hepatocytes in primary culture. Biochim Biophys Acta 1402:151–164. 113. Haddad JG, Fraser DR, Lawson DEM 1981 Vitamin D binding protein. Turnover and fate in the rabbit. J Clin Invest 67:1550–1560. 114. Nykjaer A, Dragun D, Walther D, Vorum H, Jacobsen C, Herz J, Melsen F, Christensen EI 1999 An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) vitamin D3. Cell 96:507–515. 115. Nykjaer A, Fyfe JC, Kozyraki R, Leheste JR, Jacobsen C, Nielsen MS, Verroust PJ, Aminoff M, De la Chapelle A, Moestrup SK, Ray R, Gliemann J, Willnow TE, Christensen EI 2001 Cubilin dysfunction causes abnormal metabolism of the steroid hormone 25(OH) vitamin D3. Proc Natl Acad Sci USA 98:13895–13900. 116. Leheste JR, Melsen F, Wellner M, Jansen P, Schlichting U, Renner-Muller I, Andreassen TT, Wolf E, Bachmann S, Nykjaer A, Willnow T 2003 Hypocalcemia and osteopathy in mice with kidney-specific megalin gene defect. FASEB J 17:247–249.

131 117. Goldschmidt-Clermont P, Van Baelen H, Bouillon R, Shook TE, Williams MH, Nel AE, Galbraith RM 1988 Role of groupspecific component (vitamin D binding protein) in clearance of actin from the circulation in the rabbit. J Clin Invest 81:1519–1527. 118. Dueland S, Nenseter MS, Drevon CA 1991 Uptake and degradation of filamentous actin and vitamin D–binding protein in the rat. Biochem J 274:237–241. 119. Cousin P, Dechaud H, Grenot C, Lejeune H, Hammond GL, Pugeat M 1999 Influence of glycosylation on the clearance of recombinant human sex hormone-binding globulin from rabbit blood. J Steroid Biochem Mol Biol 70:115–121. 120. Innes Chalk KJ, Kodicek E 1961 The association of 14Clabelled vitamin D2 with rat serum proteins. Biochem J 79:1–7. 121. Rikkers H, DeLuca HF 1967 An in vivo study of the carrier proteins of 3H-vitamins D3 and D4 in rat serum. Am J Physiol 213:380–386. 122. Bouillon R, Van Baelen H, Rombauts W, DeMoor P 1976 The purification and characterization of the human serum binding protein for the 25-hydroxycholecalciferol (transcalciferin). Identity with group specific component. Eur J Biochem 66:285–291. 123. Bouillon R, Verstuyf A, Zhao J, Tan B, Van Baelen H 1996 Nonhypercalcemic vitamin D analogs: interactions with the vitamin D–binding protein. Horm Res 45:117–121. 124. Swamy N, Dutta A, Ray R 1997 Roles of the structure and orientation of ligands and ligand mimics inside the ligandbinding pocket of the vitamin D–binding protein. Biochemistry 36:7432–7436. 125. Norman AW, Ishizuka S, Okamura WH 2001 Ligands for the vitamin D endocrine system: different shapes function as agonists and antagonists for genomic and rapid response receptors or as a ligand for the plasma vitamin D binding protein. J Steroid Biochem Mol Biol 76:49–59. 126. Carlberg C 2003 Molecular basis of the selective activity of vitamin D analogues. J Cell Biochem 88:274–281. 127. Vieth R 1994 Simple method for determining specific binding capacity of vitamin D–binding protein and its use to calculate the concentration of “free” 1,25-dihydroxyvitamin D. Clin Chem 40:435–441. 128. Teegarden D, Meredith SC, Sitrin MD 1991 Determination of the affinity of vitamin D metabolites to serum vitamin D binding protein using an assay employing lipid-coated polystyrene beads. Anal Biochem 199:293–299. 129. Ray R, Bouillon R, Van Baelen H, Holick MF 1991 Photoaffinity labelling of human serum vitamin D binding protein and chemical cleavages of the labelled protein: identification of an 11.5 kDa peptide containing the putative 25-hydroxyvitamin D3 binding site. Biochemistry 30:6638–7642. 130. Haddad JG, Hu YZ, Kowalski MA, Laramore C, Ray K, Robzyk P, Cooke NE 1992 Identification of the sterol- and actin binding domains of plasma vitamin D binding protein (Gc globulin). Biochemistry 31:7174–7181. 131. Addo JK, Swamy N, Ray R 2002 The C19 position of 25-hydroxyvitamin D3 faces outward in the vitamin D sterolbinding pocket of vitamin D–binding protein. Bioorg Med Chem Lett 12:279–281. 132. Avvakumov GV, Grishkovskaya I, Muller YA, Hammond GL 2002 Crystal structure of human sex hormone-binding globulin in complex with 2-methoxyestradiol reveals the molecular basis for high affinity interactions with C-2 derivatives of estradiol. J Biol Chem 277:45219–45225.

132 133. Grishkovskaya I, Avvakumov GV, Hammond GL, Catalano MG, Muller YA 2002 Steroid ligands bind human sex hormone-binding globulin in specific orientations and produce distinct changes in protein conformation. J Biol Chem 277:32086–32093. 134. Surarit R, Svasti J 1980 Effect of ligand binding on the conformation of human plasma vitamin D–binding protein (group specific component). Biochem J 191:401–410. 135. Koszelak S, McPherson A, Bouillon R, Van Baelen H 1985 Crystallisation and preliminary X-ray analysis of the vitamin Dbinding protein from human serum. J Steroid Biochem 23:1077–1078. 136. Cooke N, Walgate J, Haddad J 1979 Human serum binding protein for vitamin D and its metabolites. II. Specific, high affinity association with a protein in nucleated tissue. J Biol Chem 254:5965–5971. 137. Van Baelen H, Bouillon R, De Moor P 1980 Vitamin D–binding protein (Gc-globulin) binds actin. J Biol Chem 255:2270–2272. 138. Buch S, Gremm D, Wegner A, Mannherz HG 2002 Binding of a C-terminal fragment (residues 369 to 435) of vitamin D– binding protein to actin. Biol Chem 383:1621–1631. 139. McLeod JF, Kowalski MA, Haddad JG 1989 Interactions among serum vitamin D binding protein, monomeric actin, profilin and profilactin. J Biol Chem 264:1260–1267. 140. Emerson DL, Arnaud P, Galbraith RM 1983 Evidence of increased Gc:actin complexes in pregnant serum: a possible result of trophoblast embolism. Am J Reprod Immunol 4:185–189. 141. Goldschmidt-Clermont PJ, Galbraith RM, Emerson DL, Werner PAM, Nel AE, Lee WM 1985 Accurate quantitation of native Gc in serum and estimation of endogenous GcG-actin complexes by rocket immunoelectrophoresis. Clin Chim Acta 148:173–183. 142. Haddad JG, Harper KD, Guoth M, Pietra GG, Sanger JW 1990 Angiopathic consequences of saturating the plasma scavenger system for actin. Proc Natl Acad Sci USA 87:1381–1385. 143. Lee WM, Galbraith RM 1992 The extracellular actinscavenger system and actin toxicity. New Engl J Med 326: 1335–1341. 144. Petrini M, Galbraith RM, Werner, PAM, Emerson DL, Arnaud P 1984 Gc (vitamin D binding protein) binds to cytoplasm of all human lymphocytes and is expressed on B-cell membranes. Clin Immunol Immunopathol 31:282–295. 145. Machii T, Kimura H, Ueda E, Chujo T, Morikta T, Katagiri S, Tagawa S, Kitani T 1986 Distribution of Gc protein (vitamin D binding protein) on the surfaces of normal human lymphocytes and leukemic lymphocytes. Acta Haematol 75:26–29. 146. McLeod JF, Kowalski MA, Haddad JG 1986 Characterization of a monoclonal antibody to human serum vitamin D binding protein (Gc globulin): recognition of an epitope hidden in membranes of circulating monocytes. Endocrinology 119:77–83. 147. Guoth M, Murgia A, Smith RM, Prystowsky MB, Cooke NE, Haddad JG 1990 Cell surface vitamin D binding protein (Gc globulin) is acquired from plasma. Endocrinology 127:2313–2321. 148. DiMartino SJ, Kew RR 1999 Initial characterisation of the vitamin D binding protein (Gc-globulin) binding site on the neutrophil plasma membrane: evidence for a chondroitin sulfate proteoglycan. J Immunol 163:2135–2142. 149. Danan JL, Delorme AC, Ripoche MA, Bouillon R, Van Baelen H, Mathieu H 1985 Presence of immunoreactive vitamin D–binding protein in rat yolk sac endodermal cells. Endocrinology 117:243–247.

CHRISTOPHER J. LAING AND NANCY E. COOKE

150. Emerson DL, Werner PA, Cheng MH, Galbraith RM 1985 Presence of Gc (vitamin D–binding protein) and interactions with actin in human placental tissue. Am J Reprod Immunol Microbiol 7:15–21. 151. Nestler JE, McLeod JF, Kowalski MA, Strauss JF, Haddad JG 1987 Detection of vitamin D binding protein on the surface of cytotrophoblasts isolated from human placentae. Endocrinology 120:1996–2002. 152. Yu HM, Li XJ, Kadam AL, Cheng CY, Koide SS 1994 Human testis vitamin D binding protein involved in infertility. Arch Androl 33:119–128. 153. Moestrup SK, Verroust PJ (2001) Megalin- and cubilinmediated endocytosis of protein-bound vitamins, lipids, and hormones in polarized epithelia. Ann Rev Nutr 21:407–428. 154. Petrini M, Galbraith RM, Emerson DL, Nel AE, Arnaud P 1985 Structural studies of T lymphocyte Fc receptors. Association of Gc protein with IgG binding to Fc gamma. J Biol Chem 260:1804–1810. 155. DiMartino SJ, Shah AB, Trujillo G, Kew RR 2001 Elastase controls the binding of the vitamin D–binding protein (Gc-globulin) to neutrophils: a potential role in the regulation of C5a co-chemotactic activity. J Immunol 166:2688–2694. 156. Osawa M, Kimura A, Yukawa N, Seto Y, Saito T, Tsuji T, Takeichi S 1998 Epitope analysis of monoclonal antibodies to human Gc globulin (vitamin D–binding protein). Biochem Mol Biol Int 44:293–303. 157. Weitkamp LR, Rucknagel DL, Gershowitz H 1966 Genetic linkage between structural loci for albumin and group specific component (Gc). Am J Hum Genet 18:559–571. 158. Daiger SP, Schanfield MS, Cavalli-Sforza LL 1975 Human group-specific component (Gc) proteins bind vitamin D and 25-hydroxyvitamin D. Proc Natl Acad Sci USA 72:2076–2080. 159. White P, Cooke N 2000 The multifunctional properties and characteristics of vitamin D binding protein. Trends Endocrinol Metab 11:320–321. 160. Mendel CM 1989 The free hormone hypothesis: a physiologically based mathematical model. Endocr Rev 10:232–274. 161. Hammond GL 2002 Access of reproductive steroids to target tissues. Obstet Gynecol Clin N Am 29:411–423. 162. Siiteri PK, Murai JT, Hammond GL, Nisker JA, Raymoure WJ, Kuhn RW 1982 The serum transport of steroid hormones. Rec Prog Horm Res 38:457–503. 163. Vermeullen A 1988 Physiology of the testosterone binding globulin in man. Ann NY Acad Sci 538:103–111. 164. Nakhla AM, Leonard J, Hryb DJ, Rosner W 1999 Sex hormone–binding globulin receptor signal transduction proceeds via a G protein. Steroids 64:213–216. 165. Rosner W, Hryb DJ, Khan MS, Nakhla AM, Romas NA 1999 Androgen and estrogen signaling at the cell membrane via G-proteins and cyclic adenosine monophosphate. Steroids 64:100–106. 166. Rosner W, Hryb DJ, Khan MS, Nakhla AM, Romas NA 1999 Sex hormone–binding globulin mediates steroid hormone signal transduction at the plasma membrane. J Steroid Biochem Mol Biol 69:481–485. 167. Heinlein CA, Chang C 2002 The roles of androgen receptors and androgen-binding proteins in nongenomic androgen actions. Mol Endocrinol 16:2181–2187. 168. Hryb DJ, Nakhla AM, Kahn SM, St George J, Levy NC, Romas NA, Rosner W 2002 Sex hormone-binding globulin in the human prostate is locally synthesized and may act as an autocrine/paracrine effector. J Biol Chem 277: 26618–26622.

CHAPTER 8 Vitamin D–Binding Protein

169. Kahn SM, Hryb DJ, Nakhla AM, Romas NA, Rosner W 2002 Sex hormone-binding globulin is synthesized in target cells. J Endocrinol 175:113–120. 170. Orchinik M, Hastings N, Witt D, McEwen BS 1997 Highaffinity binding of corticosterone to mammalian neuronal membranes: possible role of corticosteroid binding globulin. J Steroid Biochem Mol Biol 60:229–236. 171. Hodam JR, Creek KE 1998 Comparison of the metabolism of retinol delivered to human keratinocytes either bound to serum retinol-binding protein or added directly to the culture medium. Exp Cell Res 238:257–264. 172. Sundaram M, Sivaprasadarao A, DeSousa MM, Findlay JBC 1998 The transfer of retinol from serum retinol-binding protein to cellular retinol-binding protein is mediated by a membrane receptor. J Biol Chem 273:336–3342. 173. Noy N Retinoid-binding proteins: mediators of retinoid action. Biochem J 348:481–495. 174. Breuner CW, Orchinik M 2002 Plasma binding proteins as mediators of corticosteroid action in vertebrates. J Endocrinol 175:99–112. 175. Bouillon R, Van Assche F, Van Baelen H, Heyns W, DeMoor P 1981 Influence of the vitamin D–binding protein on the serum concentration of 1,25-dihydroxyvitamin D—significance of the free 1,25-dihydroxyvitamin D concentration. J Clin Invest 67:589–596. 176. Dilworth FJ, Calverly MJ, Makin HLJ, Jones G 1994 Increased biological activity of 20-epi-1,25-dihydroxyvitamin D3 is due to reduced catabolism and altered protein binding. Biochem Pharmacol 47:987–993. 177. Vanham G, Van Baelen H, Tan BK, Bouillon R 1988 The effect of vitamin D analogs and of vitamin D-binding protein on lymphocyte proliferation. J Steroid Biochem 29:381–386. 178. Vargas S, Bouillon R, Van Baelen H, Raisz LG 1990 Effects of vitamin D-binding protein on bone resorption stimulated by 1,25-dihydroxyvitamin D3. Calcif Tissue Int 47:164–168. 179. Furuichi T, Kawata S, Asoh Y, Kumaki K, Ohyama Y 1998 Differential time course of induction of 1α,25-dihydroxyvitamin D3-24-hydroxylase mRNA expression in rats by 1α, 25dihydroxyvitamin D3 and its analogs. Life Sci 62:453–459. 180. Pettifor JM, Bikle DD, Cavaleros M, Zachen D, Kamdar MC, Ross FP 1995 Serum levels of free 1,25-dihydroxyvitamin D in vitamin D toxicity. Ann Intern Med 122:511–513. 181. Van Hoof HJ, Swinkels LM, Ross HA, Sweep CG, Benraad TJ 1998 Determination of non-protein-bound plasma 1,25dihydroxyvitamin D by symmetric (rate) dialysis. Anal Biochem 258:176–183. 182. Silver J, Fainaru M 1979 Transport of vitamin D sterols in human plasma: effect of excess vitamin D, 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D. Eur J Clin Invest 9:433–438. 183. Adams JS, Chen H, Chun RF, Nguyen L, Wu S, Ren S-Y, Barsony J, Gacad MA 2003 Novel regulators of vitamin D action and metabolism: lessons learned at the Los Angeles Zoo. J Cell Biochem 88:308–314. 184. Larsson D, Nemere I, Sundell K 2001 Putative basal lateral membrane receptors for 24,25-dihydroxyvitamin D3 in carp and Atlantic cod enterocytes: characterization of binding and effects on intracellular calcium regulation. J Cell Biochem 83:171–186. 185. Norman AW, Olivera CJ, Silva FRMB, Bishop JE 2002 A specific binding protein/receptor for 1α,25-dihydroxyvitamin D3 is present in an intestinal caveolae membrane fraction. Biochem Biophys Res Commun 298:414–419.

133 186. Donohue MM, Demay MB 2002 Rickets in VDR null mice is secondary to decreased apoptosis of hypertrophic chondrocytes. Endocrinology 143:3691–3694. 187. St-Arnaud R, Arabian A, Glorieux FH 1996 Abnormal bone development in mice deficient for the vitamin D 24-hydroxylase gene. J Bone Miner Res 11:S126. 188. St-Arnaud R, Dardenne O, Prud’homme J, Hacking SA, Glorieux FH 2003 Conventional and tissue-specific inactivation of the 25-hydroxyvitamin D-1-hydroxylase (CYP27B1). J Cell Biochem 88:245–251. 189. Torpy DJ, Bachmann AW, Grice JE, Fitzgerald SP, Phillips PJ, Whitworth JA, Jackson RV 2001 Familial corticosteroidbinding globulin deficiency due to a novel null mutation: association with fatigue and relative hypotension. J Clin Endocrinol Metab 86:3692–3700. 190. Joyner JM, Hutley LJ, Bachmann AW, Torpy DJ, Prins JB 2003 Greater replication and differentiation of preadipocytes in inherited corticosteroid-binding globulin deficiency. Am J Physiol Endocrinol Metab 284:E1049–E1054. 191. Domingues R, Bugalho MJ, Garrao A, Boavida JM, Sobrinho L 2002 Two novel variants in the thyroxine-binding globulin (TBG) gene behind the diagnosis of TBG deficiency. Eur J Endocrinol 146:485–490. 192. Su C-C, Wu Y-C, Chiu C-Y, Won JG-S, Jap T-S 2003 Two novel mutations in the gene encoding thyroxine-binding globulin (TBG) as a cause of complete TBG deficiency in Taiwan. Clin Endocrinol 58:409–414. 193. Vogel S, Piantedosi R, O’Byrne SM, Kako Y, Quadro L, Gottesman ME, Goldberg IJ, Blaner WS 2002 Retinol-binding protein-deficient mice: biochemical basis for impaired vision. Biochemistry 41:15360–15368. 194. Baker ME 2002 Albumin, steroid hormones and the origin of vertebrates. J Endocrinol 175:121–127. 195. Soliman AT, Madina EH, Morsi MR 1996 Radiological, biochemical, and hormonal changes in malnourished children with rachitic manifestations. J Trop Pediatr 42:34–37. 196. Lulseged S, Fitwi G 1999 Vitamin D deficiency rickets: sociodemographic and clinical risk factors in children seen at a referral hospital in Addis Ababa. East Afr Med J 76: 457–461. 197. Vasconcellos CA, Lind SE 1993 Coordinated inhibition of actin-induced platelet aggregation by plasma gelsolin and vitamin D–binding protein. Blood 82:3648–3657. 198. Erukhimov JA, Tang Z-L, Johnson BA, Donahoe MP, Razzack JA, Gibson KF, Lee WM, Wasserloos KJ, Watkins SA, Pitt BR 2000 Actin-containing sera from patients with adult respiratory distress syndrome are toxic to sheep pulmonary epithelial cells. Am J Respir Crit Care Med 162: 288–294. 199. Schiodt FV, Ott P, Bondesen S, Tygstrup N 1997 Reduced serum Gc-globulin concentrations in patients with fulminant hepatic failure: association with multiple organ failure. Crit Care Med 25:1366–1370. 200. Lee WM, Reines D, Watt GH, Cook JA, Wise WC, Halushka PV, Galbraith RM 1989 Alterations in Gc levels and complexing in septic shock. Circ Shock 28:249–255. 201. Osawa M, Yukawa N, Saito T, Huang XL, Kusakabe T, Takeichi S 1998 Increased complex formation of Gc globulin with actin in plasma from human cadavers. Forensic Sci Int 96:39–45. 202. Dahl B, Schiodt FV, Ott P, Wians F, Lee WM, Balko J, O’Keefe GE 2003 Plasma concentration of Gc-globulin is associated with organ dysfunction and sepsis after injury. Crit Care Med 31:152–156.

134 203. Dahl B, Schiodt FV, Kiaer T, Ott P, Bondesen S, Tygstrup N 1998 Serum Gc-globulin in the early course of multiple trauma. Crit Care Med 26:285–289. 204. Metcalf JP, Thompson AB, Gossman GL, Nelson KJ, Koyama S, Rennard SI, Robbins RA 1991 Gc globulin functions as a co-chemotaxin in the lower respiratory tract. Am Rev Respir Dis 143:844–849. 205. Kew RR, Mollison KW, Webster RO 1992 Binding of Gc globulin (DBP) to C5a or C5a des Arg is not necessary for co-chemotactic activity. J Leukocyte Biol 58:55–58. 206. Kew RR, Fisher JA, Webster RO 1995 Co-chemotactic effect of Gc-globulin (DBP) for C5a: transient conversion into an active co-chemotaxin by neutrophils. J Immunol 155: 5369–5374. 207. Binder R, Kress A, Kan G, Herrmann K, Kirschfink M 1999 Neutrophil priming by cytokines and vitamin D binding protein (Gc-globulin): impact on C5a-mediated chemotaxis, degranulation and respiratory burst. Mol Immunol 36:885–892. 208. Belting M 2003 Heparan sulfate proteoglycan as a plasma membrane carrier. Trends Biochem Sci 28:145–151. 209. Yamamoto N, Ngwenya BZ 1987 Activation of mouse peritoneal macrophages by lysophospholipids and ether derivatives of neutral lipids and phospholipids. Cancer Res 47:2008–2013. 210. Yamamoto N, St Claire DA, Homma S, Ngwenya BZ 1988 Activation of mouse macrophages by alkylglycerols, inflammation products of cancerous tissues. Cancer Res 48:6044–6049. 211. Naraparaju VR, Yamamoto N 1994 Roles of β-galactosidase of B lymphocytes and sialidase of T lymphocytes in inflammation-primed activation of macrophages. Immunol Lett 43:143–148. 212. Yamamoto N 1996 Structural definition of a potent macrophage activating factor derived from vitamin D3-binding protein with adjuvant activity for antibody production. Mol Immunol 33:1157–1164. 213. Mohamad SB, Nagasawa H, Uto Y, Hori H 2002 Preparation of Gc protein-derived macrophage activating factor (GcMAF) and its structural characterization and biological activities. Anticancer Res 22:4297–4300. 214. Yamamoto N, Naraparaju VR, Neil RN, Suyama H, Nakazato H 1997 Therapeutic efficiency of vitamin D3-binding protein-derived macrophage activating factor for prostate, breast, and colon cancers. Proc Am Assoc Cancer Res 38:31. 215. Koga Y, Naraparaju VR, Yamamoto N 1999 Antitumor effect of vitamin D–binding protein-derived macrophage activating factor on Ehrlich ascites tumor-bearing mice. Proc Soc Exp Biol Med 220:20–26.

CHRISTOPHER J. LAING AND NANCY E. COOKE

216. Yamamoto N, Urade M 2002 Therapeutic efficacy of Gc protein derived macrophage activating factor for adenocarcinoma. Proc Am Assoc Cancer Res 43:914. 217. Ono T, Ichikawa H, Asami R, Yamamoto N 1995 Enzymatically modified Gc globulin induces the translocation of FcγRI and FcγRII from intracellular storage compartments to the cell surface in murine peritoneal macrophage. Jpn J Inflamm 15:293–299. 218. Yamamoto N, Naraparaju VR, Asbell SO 1996 Deglycosylation of serum vitamin D3–binding protein leads to immunosuppression in cancer patients. Cancer Res 56:2827–2831. 219. Mohamad SB, Nagasawa H, Uto Y, Hori H 2002 Tumor cell alpha-N-acetylgalactosaminidase activity and its involvement in GcMAF-related macrophage activation. Comp Biochem Biophys 132A:1–8. 220. Swamy N, Ghosh S, Schneider GB, Ray R 2001 Baculovirusexpressed vitamin D–binding protein-macrophage activating factor (DBP-maf) activates osteoclasts and binding of 25-hydroxyvitamin D3 does not influence this activity. J Cell Biochem 81:535–546. 221. Schneider GB, Benis KA, Flay NW, Ireland RA, Popoff SN 1995 Effects of vitamin D–binding protein-macrophage activating factor (DBP-MAF) infusion on bone resorption in two osteopetrotic mutations. Bone 16:657–662. 222. Yamamoto N, Naraparaju VR 1996 A defect in β-galactosidase of B-lymphocytes in the osteopetrotic (op/op) mouse. Immunol Lett 50:35–40. 223. Datta HK, Cook DB, Kanan RM 1999 Evidence for continuous basal generation of Gc-MAF: absence in infantile osteopetrosis and restoration after bone marrow transplant. Blood 93:4026–4027. 224. Kanda S, Mochizuki Y, Miyata Y, Kanetake H, Yamamoto N 2002 Effects of vitamin D3-binding protein–derived macrophage activating factor (GcMAF) on angiogenesis. J Nat Cancer Inst 94:1311–1319. 225. Kisker O, Onizuka S, Becker CM, Fannon M, Flynn E, D’Amato R, Zetter B, Folkman J, Ray R, Swamy N, PineShepherd S 2003 Vitamin D binding protein-macrophage activating factor (DBP-maf) inhibits angiogenesis and tumor growth in mice. Neoplasia 5:32–40. 226. White P, Liebhaber SA, Cooke NE 2002 129X1/SvJ mouse strain has a novel defect in inflammatory cell recruitment. J Immunol 168:869–874. 227. Reichner JS, Fitzpatrick PA, Wakshull E, Albina JE 2001 Receptor-mediated phagocytosis of rat macrophages is regulated differentially for opsonized particles and nonopsonized particles containing β-glucan. Immunology 104:198–206.

CHAPTER 9

New Aspects of DBP ILSE BOGAERTS AND CHRISTEL VERBOVEN Lab. Analytische Chemie, Van Evenstraat 4; B-3000 Leuven, Belgium

HUGO VAN BAELEN AND ROGER BOUILLON Legendo, Onderwijs en Navorsing 902, Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium.

I. Introduction II. Three-Dimensional Structure of Vitamin D–Binding Protein III. Three-Dimensional Structures of Vitamin D–Binding Protein in Complex with Ligands

IV. A Structural Explanation for the Unique Functions of Vitamin D–Binding Protein within Its Gene Family References

SUMMARY

N-terminal part of domain I. The crystal structure of the DBP–actin complex was determined at 2.4 Å resolution. This structure reveals that the fast-growing side of actin monomers is blocked completely through a perfect structural fit with DBP, demonstrating how DBP effectively interferes with actin-filament formation. Comparisons of the DBP structures with the structure of serum albumin, another family member, reveal a similar topology but also significant differences in overall as well as local folding. These structural differences explain the unique vitamin D3–binding and actin-binding properties of DBP.

The crystal structure of human serum vitamin D– binding protein (DBP) and the complex of hDBP and actin were solved. Despite the common genetic origin and similarities in overall gene/protein structure, the 3D structure of DBP differs substantially from that of human serum albumin. The crystal structure of DBP allowed the positioning of the surface grove in domain I responsible for binding of all vitamin D metabolites. This allows us to explain or predict the relative affinity of vitamin D metabolites and analogs for DBP. This is of particular importance since DBP affinity explains, at least in part, differences in pharmacokinetics and selective profiles of vitamin D analogs. The resolution of the DBP surface interacting with G actin explains the tight binding between these two proteins and also allows us to explain the molecular basis of the DBP actin scavenging system, which is an important protection against tissue damage–induced multiple organ failure. Finally, the binding site of DBP has been identified for FFA, but other putative interactions of DBP with macrophages (DBP-MAF) and complement need further work.

I. INTRODUCTION The human serum vitamin D–binding protein (DBP) has many physiologically important functions, ranging from transporting vitamin D3 metabolites, binding and sequestering globular actin and binding fatty acids to functioning in the immune system. The crystal structure of DBP in complex with the vitamin D3 metabolite 25-hydroxyvitamin D3 was solved, as well as the structure of DBP in complex with a vitamin D3 analog. Both structures reveal the vitamin D–binding site in the VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

II. THREE-DIMENSIONAL STRUCTURE OF VITAMIN D–BINDING PROTEIN A. Overall Features of the Vitamin D–Binding Protein Structure The crystal structure of human DBP was solved to 2.3-Å resolution [1]. The two DBP molecules present in the asymmetric unit (further referred to as DBP-A and DBP-B) are not identical. One of the major differences between them is that the DBP-B molecules are in complex with 25OHD3,whereas the DBP-A molecules are not. As in human serum albumin (HSA) [2,3], DBP has an all α-helical structure and contains three structurally similar domains (Fig. 1). DBP and the other members of its protein family (human serum albumin, afamin, and α-fetoprotein) are postulated to have evolved from a progenitor that arose from the triple repeat of a 192-amino acid sequence [4]. This threedomain structure has been preserved in DBP; however, the third repeat is largely truncated at the C terminus. Copyright © 2005, Elsevier, Inc. All rights reserved.

136

ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON

DOMAIN I DOMAIN I

DOMAIN II

DOMAIN II

FIGURE 1

DOMAIN III

DOMAIN II

Stereo view of the overall fold of the DBP structure.

The HSA structure [2,3] indicates that each domain consists of 10 α-helices. In the DBP structure, the first domain (residues 1–191) has this α-helical arrangement. However, the second domain (residues 192–378) has a similar topology but helix 7 is replaced by a coil folding (Figs. 2, 3), and the third domain (residues 379–458) contains only helices 1–4. The three domains of DBP do not pack in a spherical manner but adopt a rather peculiar shape with two large grooves A more detailed comparison of both DBP-A and DBP-B structures reveals differences in local and overall folding. All pairwise Cα atoms in DBP-A and DBP-B superimpose with a root mean square (rms) deviation of 1.63 Å. Superimposing all common Cα atoms of both domains I, II, and III gives, however, mean rms deviations of 0.646 Å, 0.981 Å, and 0.416 Å, respectively. The higher value for the superposition of the overall DBP-A and DBP-B structures is caused by slightly different orientations of the respective domains in both structures. This difference of DBP-B may be

caused by the presence of 25OHD3, because a similar phenomenon is observed with the HSA structure. All three HSA domains assume a different orientation upon binding of free fatty acids [5]. Differences between DBP-A and DBP-B are also observed at the temperature factor level. The main difference is a much higher B-factor of the N terminus of domain I in the DBP-A molecule (91.2 compared to 76.8 for DBP-B). 25OHD3 is bound to helices 1–6 of domain I (see later discussion); its presence explains the higher stability in this region of DBP-B compared to the same region in DBP-A. In addition to the electron density for the 25OHD3 molecule, elongated residual density is observed that is able to accommodate oleic acid, a free fatty acid (FFA). DBP has been reported to bind unsaturated FFAs such as oleic acid [6–8]. Because the DBP25OHD3 crystals were obtained from a pooled human serum sample, the FFAs were expected to cocrystallize with DBP. Although DBP can be postulated to have several different FFA binding sites analogous with HSA [5,9], only one site was detected in DBP-A, and two fatty acids were seen in DBP-B. Because the FFAs present in this crystal are acquired from plasma rather than deliberately added to the protein sample, all possible FFA binding pockets may not be occupied.

B. Comparison with the Structure of Serum Albumin, a Member of Its Gene Family

FIGURE 2

The structure of domain II illustrating its typical topology. The numbering of the helices is kept the same as the similar HSA structure, although the small helix 7 is not present in this second domain of DBP.

Based on the presence of sequence homology and nearly identical disulfide bridge pattern in DBP, human serum albumin (HSA), α-fetoprotein, and afamin, the overall folds of these proteins are believed to be homologous [11]. Despite the sequence similarity, the only function DBP shares with the other family members is its fatty acid binding ability [6,7].

137

CHAPTER 9 New Aspects of DBP

FIGURE 3 Sequence alignment of DBP sequences of various species (DBP-H: human; DBP-RB: rabbit; DBP-R: rat; DBP-M: mouse) and the human sequences belonging to the same protein family (albu-H: human serum albumin; AFAM-H: human afamin, FETA-H: human α-fetoprotein). DBP has 23% sequence identity with serum albumin, 20% with α-fetoprotein, and 21% with afamin. Conserved residues are marked in blue; identical residues, green; and homologous residues, purple. Triangles indicate the Cys residues responsible for the typical disulfide bridge pattern observed in this protein family. For ALBU-H, AFAM-H, and FETO-H, only the sequence that aligns with the DBP sequence is shown. The secondary structure elements of DBP are also shown. The α-helices, shown at the top, have the same numbering as the similar HSA structure, although α-helix 7 of domain II has a coil folding in DBP. This figure was made with ALSCRIPT [10]. See color plate.

Superposition of the respective domains of the DBP and HSA structures shows similar topologies (Fig. 4a). Although the folding within each corresponding domain shows some parallels, the global orientation of the three domains in both molecules is strikingly different (Fig. 4b), resulting in two totally different structures. Superposition of all the aligned Cα atoms of DBP and HSA [3] gives an rms deviation as high as 10.1 Å (438 atoms superimposed). A comparison between the separate domains also indicates remarkable differences. The rms deviation between the aligned Cα atoms of domain I is 4.0 Å (175 atoms superimposed), 4.2 Å for domain II (185 atoms superimposed), and 3.9 Å for domain III

(78 atoms superimposed). These rms deviations are even higher when HSA is bound to fatty acids.

III. THREE-DIMENSIONAL STRUCTURES OF VITAMIN D–BINDING PROTEIN IN COMPLEX WITH LIGANDS A. The DBP–Vitamin D Complex 1. THE VITAMIN D BINDING SITE

Residual electron density that can accommodate 25OHD3 is observed in DBP-B close to the biochemically

138

ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON

A

B

DOM III HSA

DOM III HSA DOM I HSA

DOM I HSA

DOM I DBP

DOM I DBP

NN

N N

DOM III DBP DOM III DBP

DOM II DBP DOM II DBP

DOM II HSA

DOM II HSA

FIGURE 4 Structural differences between DBP and serum albumin. (A) Stereo representation of the superimposed domains II of DBP (light gray) and HSA (dark grey). The helices of HSA are labeled. The small helix 7 is not present in this particular DBP domain. (B) Stereo view of the overall folds of the DBP and HSA (medium gray) molecules based on the superposition of the domains II. DBP domain I is shown in dark grey, DBP domains II and III in light grey.

identified vitamin D–binding residues (35–49) [12] (Fig. 5b). Residues belonging to helices 1–6 of domain I form the complete vitamin D–binding site (Fig. 5a,b). This binding site designation was further confirmed by the elucidation of the structure of DBP in complex with the JY analog, 22-(m-hydroxyphenyl)-23,24,25,26,27pentanor vitamin D3 (Table I). The electron density map of the DBP-JY structure clearly illustrates the presence of residual density in the N-terminal part of domain I for both the DBP-A and DBP-B molecules (Fig. 5c). This density can accommodate the JY analog. In contrast with the DBP-25(OH)D3 structure (Fig. 5b), the shape of the electron density in the sidechain region of the vitamin D3 molecule illustrates the presence of an aromatic feature (Fig. 5c). In the DBP-JY structure, the vitamin D–binding regions of both DBP-A

and DBP-B have similar average B-values, which is in contrast to the DBP-25(OH)D3 structure. The vitamin D–binding site is lined predominantly by hydrophobic residues (Fig. 5b), allowing favorable interactions with the hydrophobic vitamin D3 ligand. Binding of 25(OH)D3 is further improved by hydrogen bond formation of the 25-hydroxyl with Tyr 32 (O–H…O distance 2.85 Å), and the 3-hydroxyl with Ser 76 (O…H–O distance 2.81 Å) and Met 107 (O–H…S distance 3.01 Å). For DBP-JY, the same hydrogen bonds can be formed for 3-hydroxyl, but the hydroxyl of the phenolic substituent is hydrogen bonded to Glu 15. This hydroxyl of JY is at the same position as the oxygen of a water molecule in the DBP-25(OH)D3 structure (Fig. 5b,c). However, the detailed structural features of 25(OH)D3 or JY cannot be seen in the

139

CHAPTER 9 New Aspects of DBP

density of the structures, probably because not all of the sites present in the DBP-25(OH)D3 and DBP-JY crystals are fully occupied, but it may also be ascribed to mobility of the bound ligand. The vitamin D–binding site of DBP-B in the DBP-25(OH)D3 structure is

estimated to be 55% occupied by 25(OH)D3. For the DBP-JY structure, the estimated occupancy of JY in DBP-A is 0.66, whereas that for DBP-B is 0.72. Despite the low detail in the observed electron density, the proposed orientation of the molecule in the

A

B

F24

V12 WAT WAT

V51 E8

L31

Y32

T72

L47 435

576

F36 579

M107

FIGURE 5 The conformation of vitamin D metabolites and analogs in the vitamin D–binding site of DBP. (A) Stereo view of the location of the vitamin D–binding site in the DBP molecule. Helices 1–6 of domain I of DBP-A (light blue) and DBP-B (pink) from the DBP-25OHD3 structure are superimposed based on the common Cα atoms in helix 2. The DBP-B molecule is the one in which the loop at the top and helix 1(H1) are complete. The 25OHD3 is shown in a ball-and-stick representation. This superposition illustrates that the disordered residues of the DBP-A molecule belong to the region that forms the vitamin D–binding site. Two water molecules are present in the vitamin D–binding site (red balls). (B) View of the simulated annealing Fo−Fc omit map of DBP in the DBP-25OHD3 structure showing the electron density for the omitted 25OHD3 molecule. The map is contoured at 3.0 σ. Residues closer than 4.2 Å to the ligand are shown. Two water molecules (WAT) are present in the vitamin D–binding site (red balls). (Continued)

140

ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON

C

E15

F24 L27

V12 E8

JY T72

WAT

V51 L47

S28 L31 Y32 K35 F36

L75 S76 S79 M107 D

FIGURE 5

Cont’d (C) Same as (B), but showing the electron density for the omitted JY molecule in the DBP-JY structure. The one remaining water molecule in the vitamin D–binding site is a red ball. (D) Stereo view of the ball-and-stick representation of the modeled vitamin D3 analogs in the vitamin D–binding site. 25OHD3 is shown in gray; 1,25(OH)2D3, dark blue; JX, pink; JY, purple; ZAF, yellow; ZAG, light blue; ZAQ, red; and ZAU, green. For ZAF, ZAG, and ZAQ, the presence of the substituent allows for extra hydrophobic interactions with DBP residues, probably resulting in the somewhat higher affinity compared to 1,25(OH)2D3. See color plate.

binding site is consistent with extensive biochemical and modeling data and, therefore, is assumed to be correct. The increased binding of the JY analog to DBP compared to the binding of 25(OH)D3 could result from favorable stacking of the JY aromatic side chain with Phe 24 (Fig. 5c). The aromatic residues Tyr 32, Phe 36, and Tyr 68 further stabilize the JY aromatic side chain. The JX analog, 22-(P-hydroxyphenyl)23,24,25,26,27-pentanor vitamin D3 (Table I), has an approximately threefold higher affinity for DBP than 25(OH)D3. Modeling of JX shows that the p-hydroxyl

allows for extra stacking with Phe 24 (Fig. 5d). This hydroxyl also makes a tight hydrogen bond with Ser 28. Although the modeling of 1,25(OH)2D3 illustrates that this molecule can make the same hydrogen bonds with DBP as 25(OH)D3 (Fig. 5d), it also shows that the axial 1-hydroxyl causes steric hindrance with Met 107. This may explain the lower DBP binding affinity observed for 1,25(OH)2D3. The vitamin D–binding site is a cleft; consequently, certain parts of the vitamin D3 molecule have no interaction with DBP—for example, the β-side of the C- and D-rings of

141

CHAPTER 9 New Aspects of DBP

TABLE I Structural Formulas of Vitamin D3 Metabolites and Analogs and Their Affinity for DBP 18

R1

12

R3

17 13

11

16

D

C 14

9

15

8

7

H

6 19

5 10

4 A

1

3 HO Vitamin D analog

R2

2

R2

R1 21

25OHD

24

22 20

23

25 27

R3

H

Affinity for DBP

H

667

H

1

H

H

2111

H

H

1314

26 OH 21 1,25(OH)2D3

22 20

24 23

25 27

OH

26 OH

JX OH OH JY

21 ZAF

ZAG

21

24

22 20

23

22 20

25 27

OH

2

OH

1.5

26 OH 24

23

25 27 26 OH

21 ZAQ 21 ZAU

22 20

24 23

22 20

OH

CH2=CH

OH

H

26 OH 24

23

25 27

25 27

3.8 1

To illustrate the relative affinity of the ligands for the vitamin D–binding protein, the reported Relative Competitive Index values [13] are shown. All reported values are relative to the result for 1,25(OH)D3, which is normalized to 1. The ZAU analog has the same structural formula as 1,25(OH)D3 except for the methyl group on C-13, which is replaced by a vinyl group in the β-position.

142

ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON

the vitamin D3 molecule (Table I, Fig. 5a–c). Analogs substituted in these parts of the molecule display affinities similar to the affinity between 1,25(OH)2D3 and DBP (Table I). Furthermore, modeling of the analogs ZAG (11β-phenyl-1α,25-dihydroxyvitamin D3), ZAF (11α-phenyl-1α,25-dihydroxyvitamin D3), ZAQ (11αvinyl-1α,25-dihydroxyvitamin D3), and ZAU (13β-vinyl18-nor-1α,25-dihydroxyvitamin D3) illustrates that large substituents, such as a vinyl group or a phenyl ring, can easily fit in the binding site (Fig. 5d). These observations confirm the proposed orientation of 25OHD3 in the vitamin D–binding site. 2. DIFFERENCES BETWEEN THE VITAMIN D–BINDING SITE AND THE VITAMIN D POCKET OF THE VITAMIN D RECEPTOR

The effects of the vitamin D3 hormone and its analogs are mediated through binding to an intracellular vitamin D receptor (VDR) [13]. The comparison of the DBP vitamin D–binding site with that of the VDR structure [14] clearly illustrates that they are dissimilar. In VDR, the vitamin D–binding site is a closed pocket formed in the inner structure of the receptor, whereas in DBP it is a cleft located at the surface of the molecule and partly in contact with the surrounding solvent. Moreover, no parallel for both of these physiologically important binding sites is found in ligand conformation, hydrogen bonding, or hydrophobic interactions. In the VDR pocket, the A-ring of the vitamin D3 ligand has a B-chair conformation (3-hydroxyl axial), whereas in the DBP cleft the A-ring has an A-chair conformation (3-hydroxyl equatorial). In both binding sites, the C5–C6–C7–C8 torsion angle of the ligand is nonplanar. The angle is 149° in DBP and −149° in VDR. The torsion angles responsible for the side-chain orientation of the vitamin D3 molecule are different as well (C13–C17–C20–C22 = −77° in DBP and 89° in VDR; C17–C20–C22–C23 = −70° in DBP and −156° in VDR). 3. BIOLOGICAL IMPLICATIONS

The active vitamin D3 hormone, 1,25(OH)2D3, is used in the treatment of renal osteodystrophy, hypoparathyroidism, and osteoporosis. Its analogs are used successfully in the treatment of psoriasis [15] and are currently being explored for the treatment of hyperproliferative disorders [16] and the prevention of autoimmune disorders [17,18]. However, treatment with 1,25(OH)2D3 is limited because of severe side effects (such as hypercalciurea, hypercalcemia, and increased bone resorption) caused by the required pharmacological doses [13]. In order to separate the antiproliferative and prodifferentiating effects from the calcemic and skeletal effects,

a whole generation of chemically modified 1,25(OH)2D3 molecules (analogs) has been synthesized, but with poor success rates. Many of the analogs, having a more selective activity profile, display a decreased binding affinity for DBP [19–21]. Recent progress in certain aspects of the functional mechanism of 1,25(OH)2D3 provides new perspectives for the rational design of vitamin D3 analogs with a more selective profile. For example, structural analysis of the ligand binding domain of VDR [14,22] revealed the three-dimensional arrangement of its vitamin D–binding pocket. The elucidation of the three dimensional architecture of DBP’s vitamin D–binding site provides a solid basis for detailed structural studies on DBP–vitamin D analog complexes that will lead to the better understanding of the selective action mechanisms of 1,25(OH)2D3 and its analogs. The information contained in the three-dimensional architectures of both the DBP and VDR vitamin D–binding sites may lead to the development of compounds with a more original chemical structure and a more tissue- or gene-specific biological action.

B. The DBP–Actin Complex 1. ROLE OF DBP IN THE ACTIN-SCAVENGER SYSTEM

Actin is a highly abundant intracellular protein present in all eukaryotic cells and has a pivotal role in muscle contraction as well as in cell movements. Actin also has an essential function in maintaining and controlling cell shape and architecture: It is the essential building block of the microfilament system, a cytoskeletal structure that complements two other cytoskeletal structures (the microtubules and the intermediate filaments). In low-salt buffers actin exists as a monomeric protein (globular actin, G-actin), but it polymerizes under physiological salt conditions into a double helical 10-nm-thick filament structure (filamentous actin, F-actin). In vivo, the equilibrium between G-actin and F-actin is controlled mainly by the action of several actin-binding proteins with distinct activities (e.g., gelsolin or profilin). In conditions involving severe cell injury, such as trauma, shock, sepsis, and fulminant hepatic necrosis, large quantities of actin are released in the systemic circulation. The presence of actin filaments in blood, leading to an increase in blood viscosity, is dangerous and may be fatal [23]. In addition, actin can promote clot formation by its ability to aggregate platelets [24]. Therefore, the presence of a protective system, preventing actin polymerization and ensuring actin’s fast elimination, is essential.

143

CHAPTER 9 New Aspects of DBP

The vitamin D–binding protein is an actin-binding protein [11,25] and acts as an actin-sequestering agent in extracellular space. DBP and gelsolin—the only other plasma protein that binds actin avidly—play a crucial role in the clearance of actin filaments from the circulation, a process known as the “actin-scavenger system” [23]. Gelsolin has three actin-binding properties: (i) it severs F-actin; (ii) it caps the end of F-actin; and (iii) it nucleates actin filament assemblies. Gelsolin will shift the equilibrium between actin assembly–disassembly toward actin depolymerization, but will basically not prevent actin reassembly. DBP, by forming a complex with G-actin, prevents the (re-)formation of actin filaments. Therefore DBP and gelsolin have complementary functions: Upon severing the actin filaments by gelsolin, the generated globular actin is locked in its monomeric state by DBP. The clearance of this DBP–actin complex is even substantially faster compared to the clearance of free DBP [26–29]. The complementary action of both proteins is further illustrated by the previous observation that both DBP and gelsolin are required to inhibit actinstimulated platelet aggregation [24]. 2. GENERAL FEATURES OF THE DBP–ACTIN STRUCTURE

The DBP–actin structure described here was solved by X-ray diffraction by Verboven et al. [30]. The final model of the DBP–actin structure converged to an R factor of 19.90% and an Rfree value of 25.35%. It contains an actin, DBP, and ATP molecule, one calcium ion, and 290 water molecules. The structure of the DBP–actin complex reveals that G-actin binds in one of the DBP grooves, mainly formed by helix 10 of domain I, helix 6 of domain II, and helix 3 of domain III, allowing DBP and actin to fit as two pieces of a jigsaw puzzle (Fig. 6). The surface area buried at the interface of this complex is as large as 3600 Å2. Numerous intermolecular hydrogen bonds, hydrophobic contacts, and electrostatic interactions (Table II) stabilize the complex. The actin monomer consists of two domains with each domain further subdivided in two subdomains (Fig. 6). Subdomain 1 comprises residues 1–32, 70–137, and 338–375; subdomain 2, residues 33–69; subdomain 3, residues 138–180 and 270–337; and subdomain 4, residues 181–269. Biochemical experiments delimited the DBP interface by the residues 360–372 of the actin subdomain 1 [31]. The DBP–actin structure demonstrates that actin residues of subdomains 1 and 3 constitute the DBP binding interface (Fig. 6, Table II). As in actin–gelsolin segment 1 (actin-GS1) [32] and in actin–gelsolin segment 4 to 6 (actin-GS4-6) [33], the binding of DBP occurs at the actin cleft formed at the

SUBDOMAIN2

SUBDOMAIN4

SUBDOMAIN1 SUBDOMAIN3

FIGURE 6 Overall architecture of the DBP–actin structure. Actin binds in one of the large grooves present in the DBP structure. The three domains of DBP and the four subdomains in actin are shown in different colors. See color plate.

interface of subdomains 1 and 3. The exposed hydrophobic residues on helix 341–349 of actin subdomain 1 are thereby occluded from solvent. Even more striking is that in all these complexes the apolar patch (actin residues 341–349) is masked by a solventexposed hydrophobic face of a helix present in gelsolin segment 1, in gelsolin segment 4, and in DBP. Upon complexation with DBP, the changes in the actin structure are restricted to some small regions rather than affecting the general fold. This is reflected in the small rms differences obtained from the superposition of the DBP–actin structure with other actin complexes: actin-GS1 [32], actin-DNase I [34], and uncomplexed actin [35]. Similar to other actin complex structures [32,33], the structural differences are mainly caused by folding changes in subdomains 2 and 4, whereas, despite their interaction with DBP, the backbone fold of subdomains 1 and 3 is hardly altered. Only the side chains of the residues at the interface adopt different orientations to ensure the favorable contacts with the neighboring DBP residues. The ATP molecule, bound in the cleft between the subdomains 2 and 4, and nearly all of the ATP surrounding residues in DBP–actin have the same orientation and

144

ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON

TABLE II

Intermolecular Contacts between Actin and DBP Residues

α-Actin

DBP

Residue

Atom

Residue

Atom

Distance (Å)

LYS 113 LYS 113 TYR 143 TYR 143 ALA 144 SER 145 SER 145 GLY 146 GLY 146 GLY 146 GLY 146 ARG 147 ARG 147 ARG 147 ARG 147 ARG 147 ARG 147 THR 148 THR 148 TYR 166 TYR 166 TYR 166 TYR 166 GLU 167 GLU 167 GLU 167 TYR 169 TYR 169 LEU 171 HIS 173 HIS 173 MET 176 GLU 276 TYR 279 TYR 279 ASN 280 ASN 280 ASN 280 MET 283 MET 283 MET 283 LYS 284

NZ NZ CG OH O O C O CA O O NH2 NH1 NH1 NH1 NH1 NH1 CG2 CG2 CE2 OH CD2 CD2 O OE1 CG CE1 OH CD1 NE2 CD2 SD CG OH CD2 OD1 ND2 OD1 O CB CE NZ

SER 278 ASP 282 LEU 184 LEU 188 THR 180 ASN 125 THR 180 ASN 125 THR 180 PHE 183 ARG 187 GLU 122 PRO 123 THR 124 ASN 125 ILE 128 ARG 187 ARG 187 LEU 188 LEU 195 THR 198 VAL 294 PHE 298 LYS 191 SER 194 LEU 195 GLU 286 CYS 295 PHE 298 PHE 298 PRO 300 THR 398 THR 400 PHE 399 THR 400 THR 398 PHE 399 THR 400 TYR 394 PHE 399 SER 434 SER 395

OG OD2 CD1 CD2 CG2 OD1 CG2 OD1 O CB NE O O C ND2 CD1 NH2 CB CD2 CD1 CG2 CG2 CE2 NZ OG CD1 OE1 SG CD2 O CG CG2 CG2 CE2 CG2 CB N OG1 OH CB CB O

3.51 3.49 3.64 3.87 3.21 3.45 3.87 3.14 3.65 3.57 3.25 3.59 2.92 3.64 2.84 3.80 3.23 3.93 3.85 4.07 3.25 4.07 3.72 2.45 2.63 3.34 3.85 3.28 3.81 4.02 3.32 4.05a 4.20 3.67b 4.01a 3.19 3.51 3.00 3.61 3.89 3.45 3.76 (Continued)

145

CHAPTER 9 New Aspects of DBP

TABLE II Intermolecular Contacts between Actin and DBP Residues—Cont’d α-Actin

DBP

Residue

Atom

Residue

LYS 284 ASP 286 ASP 286 ASP 286 ILE 287 ILE 287 ILE 287 ILE 287 ASP 288 ILE 289 ARG 290 ARG 290 LYS 291 LYS 291 LYS 291 ALA 295 LYS 328 LYS 328 ILE 345 LEU 346 LEU 349 LEU 349 LEU 349 THR 351 THR 351 MET 355 HIS 371

NZ OD1 OD1 OD1 CD1 CG2 CG2 CD1 OD2 CD1 NH2 NH2 CD NZ NZ CB NZ NZ CG2 CD1 CD1 CD1 CD1 OG1 CG2 CE CB

ASN 397 TYR 297 ASN 440 ASN 440 PHE 117 SER 434 ASN 435 PRO 442 ARG 202 PHE 298 TYR 394 SER 434 PRO 118 TYR 120 TYR 151 VAL 121 PRO 123 GLU 127 LEU 184 LEU 188 LEU 184 LYS 185 LEU 188 LYS 185 GLN 189 LYS 287 GLN 285

β- or γ-Actin

Atom O CE2 CG ND2 CE1 O OD1 CD NE CZ OH O O O OH CG1 CA OE1 CD2 CD1 CG CA CD1 CG CD CE OE1

Distance (Å) 3.30 3.74 3.60 2.97 3.19a 3.48b 3.79b 3.26a 3.26 4.03 4.07 2.87 3.36 2.79 2.83 3.58 3.94 3.19 3.70 4.06 4.06 4.13 3.99 4.19 3.74 4.20 2.89

DBP

Residue

Atom

Residue

Atom

Distance (Å)

CYS 272 PHE 279 VAL 287 VAL 287

SG CE2 CG1 CG1

THR 400 PHE 399 SER 434 ASN 435

OG1 CD2 O OD1

3.36c 3.55 3.40 3.66

a

These interactions are absent in DBP–β-actin or DBP–γ-actin complexes. residues are mutated in β- or γ-actin and have different interactions with neighboring DBP residues. Their interactions are shown in the second part of the table. cThis interaction is not present in DBP–α-actin. bThese

146

ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON

conformation as found for all other known ATP–actin structures. Although DBP interacts with the C-terminal part of actin, located in subdomain 1, the C-terminal actin residues 373 to 375 are disordered in DBP–actin and show no electron density. Similar to actin-GS1 [32] and actin-GS4–6 [33], residues 41 to 50 (i.e., the loop that forms a β-sheet with DNase I in the actinDNase I structure [34] are also disordered in our structure. According to limited proteolysis experiments performed with DBP [12] the actin binding site is located in its C-terminal part, and more specifically residues 350 to 403 (of domains II and III) are involved in actin binding. However, our DBP–actin structure shows that all three DBP domains interact with actin (Fig. 6, Table II). Besides residues 394–395 and 397–400, both part of the biochemically identified actin-binding region (residues 350–403), an additional 35 residues from other DBP regions also ascertain the interaction. Only minor folding differences are observed between the DBP conformations in the presence or absence of actin. The rms difference of 1.86 Å, obtained from superimposing all common α-carbons of the DBP–actin and of the DBP (PDB code 1J78) [1] structures, can mainly be ascribed to a different orientation of domain I and to a different folding in some loop regions. The domain by domain superposition gives lower rms difference values: 1.40 Å, 1.27 Å and 0.46 Å for domains I, II, and III, respectively. The larger rms difference value for domain I is caused by the interaction with actin and by crystal packing differences (Fig. 7). In order to allow the interaction between solvent-exposed hydrophobic DBP residues on the C-terminal helix of domain I (amino acid region 180–188) and the hydrophobic cleft formed by actin subdomains 1 and 3, the C-terminal part of DBP domain I has to readjust its orientation slightly (Fig. 7). In agreement with previous biochemical observations [25,36], the DBP–actin structure proves that the bound actin and the observed folding differences in domain I do not hinder the binding of vitamin D3 compounds to the vitamin D–binding site of DBP (Fig. 7). All presently missing residues in the DBP structure (98–104; 318–322; 361–364), belonging to regions that are poorly defined in the electron density map, are located opposite of the actin binding site. The DBP residues involved in actin binding are relatively well conserved between different species in all presently known DBP sequences. Together with the highly conserved amino acid sequence of actin throughout evolution (e.g., human and rabbit skeletal muscle actin are 100% identical) it explains why the

Neighboring DBP

Actin

25OHD3

Domain I DBP-25OHD3 Domain I DBP-actin

FIGURE 7

The superposition of the domains I of DBP–actin and of the DBP–25-hydroxyvitamin D3 (DBP–25OHD3) complex [1]. Domain I of DBP–actin is shown in dark grey, actin in light grey, domain I of DBP-25OHD3 in medium grey, and a neighboring DBP (at the top) present in the DBP–25OHD3 crystal is also in dark gray. Some of the interacting residues are shown in ball-and-stick representation as well as the 25-hydroxyvitamin D3. The C-terminal part of domain I (helices 7–10) of DBP–actin is shifted toward the actin to allow favorable contacts between DBP helix 10 and the cleft between actin subdomains 1 and 3. Owing to the presence of a neighboring molecule in the DBP-25OHD3 crystal, the N-terminal part of domain I (helices 1–6; i.e., the vitamin D binding site) is shifted toward this neighbor to make intermolecular hydrogen bonds and hydrophobic contacts. This superposition also illustrates that the binding of a vitamin D3 compound to DBP in the presence of actin is possible.

actin-binding property of DBP is universally observed in vertebrates. In comparison with α-actin, a twofold decrease of DBP binding affinity for the nonmuscle actin isoforms, β-actin and γ-actin, was observed [36]. The actin in this DBP–actin structure originates from rabbit muscle, i.e., α-actin. Among the α/β or α/γ sequence differences, the residues Met/Leu 176, Ala/Cys 272, Tyr/Phe 279, and Ile/Val 287 are located in the DBP binding site and the residues Asn/Thr 297, Met/Leu 299, and Thr/Ser 358 are situated in its surroundings. However, in silico replacement of these residues in the DBP–actin structure does not result in any unfavorable contacts (Table II), and there is no indication that the binding interface should fold differently to accommodate the replaced amino acids. Even, the interaction surface between β- or γ-actin and DBP, using the modeled DBP–actin β- or γ-complex for the calculations, remains similar (3609 Å2 compared to 3607 Å2 for DBP–α-actin). The lower affinity of DBP for both β- and γ-actin can only be ascribed to the loss of hydrophobic interactions between DBP residues 117 and 442 and the β or γ-actin residue 287, between DBP residue 398

147

CHAPTER 9 New Aspects of DBP

and the β- or γ-actin residue 176, and between DBP residue 400 and the β- or γ-actin residue 279 (Table II). 3. THE DBP–ACTIN BINDING INTERFACE COMPARED OTHER ACTIN COMPLEXES

TO

Several structures of G-actin in complex with an actin-binding protein have already been elucidated: actin–DNase I [34], actin–profilin [37], actin–GS1 [32], and actin–GS4-6 [33]. The actin surface buried by the complex formation with DBP is 3600 Å2, whereas the buried surfaces for the actin–DNase I, actin–GS1, actin–GS4-6, and actin–profilin complexes are 1800 Å2, 2100 Å2, 2100 Å2, and 2000 Å2, respectively. From all these complexes, DBP has the largest binding interface, indicating that the DBP–actin interaction is not merely fortuitous. The DNase I binding site of actin, consisting of residues from subdomains 2 and 4, is located opposite to the DBP binding site (subdomains 1 and 3). This corroborates the observation that the ternary complex of DBP, DNase I, and actin can be formed in vitro [25,38]. The interaction surfaces of profilin, gelsolin segment 1, and gelsolin segments 4–6 are all located at the same side of actin as the DBP binding site and are all partially overlapping. This structural comparison also demonstrates how DBP bound to actin obstructs its further interaction with profilin, as previously observed [39]. Gelsolin, consisting of six segments of 120 to 130 amino acids each (S1 to S6), has been reported to nucleate growth through the binding of two actin monomers in an appropriate orientation. DBP can prevent re-formation of actin filaments by reducing the effective G-actin concentration. The observation that DBP is capable of displacing one actin from the ternary gelsolin:actin (1:2) complex may also play a role in inhibiting the nucleation [40]. However, the current understanding of the structural organization of the gelsolin:actin (1:2) complex does not permit us to explain how DBP can remove one actin from it. For, in the proposed model of this ternary complex [33], the DBP binding interface is blocked in both actins either by segment 1 or by segment 4 of gelsolin. However, another study, involving cross-linking experiments with different gelsolin constructs, indicated that a ternary complex with the two actins in appropriate orientation for filament growth [41] is formed only with gelsolin segments 2–6, whereas with the whole gelsolin the actin monomers are in an antiparallel orientation as in the lower actin dimer. It is noteworthy that DBP, gelsolin segment 1, and gelsolin segment 4–6 have a hydrophobic helix that interacts with the hydrophobic patch in the cleft between subdomains 1 and 3 of actin. It can be postulated that other proteins binding actin at

the barbed end side and having an appropriate helix with solvent accessible hydrophobic residues, will bind to actin in such an orientation that this helix is also located in the cleft between subdomains 1 and 3. This may be the case for the N-terminal helix of the actin sequestering β-thymosins. Such an orientation of this helix would explain the cross-links between thymosin β 4 and actin: Lys 3–Glu 167 and Lys 18–Asp 1 [42]. Moreover, it was previously suggested that thymosin β 4 interacts with actin through a patch of hydrophobic residues located on the N-terminal helix [43]. This hypothesis also supports the previously suggested idea that the α3-helix of cofilin [44], an actin-severing protein, could interact with actin in a similar manner to its corresponding α-helix in gelsolin. 4. THE EXCELLENT ACTIN-SEQUESTERING PROPERTIES OF DBP

Actin monomers polymerize into double helical filaments (Fig. 8a) twisting around each other [45]. In contrast to other types of polymerization, the actin polymerization is not linear but involves a nucleation step, with a nucleus consisting of three subunits. The resulting microfilaments are polarized: the affinity for addition of new actin monomers is different (approximately 10-fold) at both ends. This can result, at intermediate actin monomer concentrations, in a polymerization of the filament at one end (the “fast growing” or “barbed” end) and a simultaneous depolymerization at the other end (the “slow growing” or “pointed” end) [46]. DBP binds G-actin at the side corresponding to the barbed or fast-growing end of actin in the F-actin filament. It thereby occludes the actin residues 166 to 169 and 286 to 289, all responsible for the longitudinal interactions between actin subunits within one F-actin strand [45]. Moreover, both regions are part of a hydrophobic pocket, which is postulated to be essential for the insertion of a hydrophobic plug from an actin subunit of the opposite actin strand. Although DBP does not make an immediate contact with actin residues 110 to 112, which are involved in interactions with an actin subunit of the opposite F-actin strand, its large volume sterically hinders the approach of the “opposite strand” subunit. The presence of DBP on G-actin therefore prevents the binding of two actin subunits: one that belongs in F-actin to the same strand, the other to the opposite strand (Fig. 8a). Hence, DBP not only blocks elongation of actin at this side, but also prohibits actin nucleation. The presence of DBP on actin seems not to hinder the interaction with other actin molecules at the pointed end side (Fig. 8a). Filament growth is, however, impossible. In conditions where the molar G-actin concentration is less than or equal to that of DBP, all G-actin subunits will be

148

ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON

A

(-)

B

(+) FIGURE 8 The binding of DBP to actin prevents actin-filament formation. (A) Superposition of the actin from the DBP–actin complex with the central actin subunit of the F-actin model [45] (present in PDB file 1alm) [58]. The five actin subunits have different grey scales. The presence of DBP (dark grey) in the DBP–actin complex prevents the binding of two subunits, which would belong in an actin filament to opposite strands: the two bottom actin subunits shown in light grey. For clarity reasons, the actin from the DBP–actin is not shown. The + and − signs indicate the fast-growing (barbed) and slow-growing (pointed) ends of F-actin, respectively. (B) Different DBP–actin complexes with the DBP always shown in dark grey and the actin in the same grey scales as the central actin subunit and the pointed-end actin subunits of (A). The presence of the bulky DBP on all actin subunits prohibits the approach of these actin subunits to another actin subunit.

captured and blocked by DBP (Fig. 8b). In addition, the presence of DBP prevents the formation of the so-called lower dimer. In this actin dimer model, the two actin molecules are in antiparallel orientation, a prerequisite for enabling actin polymerization [47]. 5. IMPLICATIONS FOR THE ACTIN-SCAVENGER SYSTEM

The observed perfect fit between actin and DBP, as illustrated here, provides the structural basis for the important role of DBP in the extracellular actin-scavenger system. Furthermore, although DBP

and gelsolin are both present in large concentrations (µM) in blood, their capacity to scavenge actin may still be overwhelmed during massive cell injury. Indeed, the saturation of the actin-scavenger system leads to thrombi formation and microangiopathy [48], and excessive amounts of actin in the circulation may lead to a condition resembling “multiple organ dysfunction syndrome.” This syndrome was found to be associated with reduced serum DBP levels [49,50]. Moreover, DBP serum concentrations are reduced in some patients with hepatic failure [49,51]. The DBP concentrations not only have some value in predicting

149

CHAPTER 9 New Aspects of DBP

survival from hepatic failure [52,53] but also may identify high-risk patients after multiple trauma [54]. Consequently, the DBP–actin structure, revealing the DBP residues essential for the interaction with actin and for the actin sequestering, is of major importance for the development of a drug that could be applied in the pathological conditions just mentioned. Despite intensive analysis no DBP-deficient individual has been identified, leading to the suggestion that certain functions of DBP might be essential for survival. However, viable DBP knockout mice have been generated [55]. These mice were only used for analysis of DBP’s role in vitamin D metabolism and action. Although DBP knockout mice are viable in normal laboratory conditions, it might well be that the role of DBP in the actin-scavenger system is confined to cases of severe tissue damage in which DBP is essential for survival. This would be in line with our striking structural observations and the perfect conservation of DBP–actin interaction during the evolution of vertebrates. In the actin–GS1 [32], actin–GS4-6 [33], actin– profilin [37], and DBP–actin complexes and in actin covalently linked to tetramethylrhodamine-5-maleimide [56] (actin-TMR), actin filament formation is blocked at the barbed-end side. The structures of all these complexes illustrate that all these actin ligands block the cleft between subdomains 1 and 3. Even a small molecule such as TMR is able to prevent filament formation. Intracellular sequestering proteins such as thymosin β4 (5 kDa) and profilin (15 kDa) are rather small compared to their extracellular counterpart, DBP (51 kDa). These remarkable differences in size of the actin interaction surfaces between the extra- and intracellular sequestering proteins might be related to the fact that in the intracellular compartment an equilibrium between G-actin and F-actin has to be maintained. In extracellular space F-actin formation has to be prevented at any cost. Therefore, DBP, a better actin-sequestering protein, is present in high concentrations in the extracellular compartment. 6. COMPARISON OF THE THREE REPORTED DBP–ACTIN STRUCTURES

Although our DBP–actin structure was obtained from a crystal grown in a different space group (P21) to that of two other very recently reported DBP–actin structures (space group P212121; PDB codes 1kxp [35] and 1lot [57]), the structures are very similar. The Cα atoms of our DBP–actin structure and the 1lot structure superimpose with an rms difference of 0.791 Å (781 atoms superimposed), while the Cα atoms of our DBP–actin structure and the 1kxp structure superimpose

with an rms difference of 0.709 Å (777 atoms superimposed). The differences between these three structures mainly reside in the loop regions and the N and C termini and are mainly the result of different lattice contacts. In the paper of Otterbein and co-workers, the possibility of a conformational change within the C-terminal actin residues 365–375 is proposed, since in their structure (1kxp) this region is poorly defined in the electron-density map. However, the seven extra residues 365–371 in our structure assume a similar conformation as in the other reported actin structures. The 1kxp structure and our structure have Mg2+-ATP at their nucleotide-binding site, while the 1lot structure has Ca2+-ATP. The presence of a different ion does not alter the conformation of ATP or the conformations of the surrounding amino acids. Only the water coordination of the ions is different: Ca2+ has a pentagonal bipyramidal coordination with five water molecules and two O atoms from the ATP β- and γ-phosphates, compared with the nearly octahedral coordination of Mg2+, which has one less water molecule. In our DBP–actin structure and the 1lot structure, His 73 of the actin molecule is methylated, while in the 1kxp structure this is not the case. Although the peptide bond between Glu 72 and His 73 of actin has the same orientation in 1lot as in our structure, there is no hydrogen bond in 1lot between the main-chain carbonyl group of Glu 72 and the side chain of Arg 183. In 1kxp, the orientation of this peptide bond is different, making such a hydrogen bond impossible.

IV. A STRUCTURAL EXPLANATION FOR THE UNIQUE FUNCTIONS OF VITAMIN D–BINDING PROTEIN WITHIN ITS GENE FAMILY The unique functions of DBP within this protein family must arise from its distinctive fold. The different helical arrangement observed in domain I of DBP allows for the binding of vitamin D3 ligands. Superposition of helix 2 of domain I of DBP with that of HSA illustrates that helix 4 of HSA makes a closer contact with helices 2 and 3, leaving no room for a vitamin D3 molecule. Compared to HSA, α-fetoprotein, and afamin, the third domain of DBP is largely truncated at the C terminus. Moreover, the comparison of the DBP and HSA structures revealed remarkably differing orientations of the three domains in both structures [1]. The shorter domain III of DBP together with its distinctive fold allows DBP to bind actin in contrast to the other family members (Fig. 9).

150

ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON

A

DBP domain I Actin HSA domain III

HSA domain I

DBP domain III DBP domain II HSA domain II

B

FIGURE 9

Superposition of the DBP from the DBP-actin structure with the HSA structure [2], illustrating that HSA cannot bind actin. (A) Stereoview of the overall structure of HSA (in gray) positioned on the DBP–actin structure based on the superposition of domains II of DBP and HSA. Only the actin subdomains 1 and 3 (light grey) are shown. Domain I of DBP is shown in dark grey, domain II, in grey and domain III in medium grey. The orientation of domains I–III in DBP and HSA is completely different. Consequently, the residues forming the actin-binding interface in DBP have a totally different arrangement in HSA. (B) Close-up view of (A) illustrating that owing to their different orientation, HSA helices 9 and 10 of domain I and helices 1–3 of domain III collide with parts of the actin structure.

References 1. Verboven C, Ragijns A, De Maeyer M, Van Baelen H, Bouillon R, De Ranter C 2002 A structural basis for the unique binding features of the human vitamin D-binding protein. Nat Struct Biol 9:131–136. 2. He XM, Carter DC 1992 Atomic-structure and chemistry of human serum-albumin. Nature 358:209–215. 3. Sugio S, Kashima A, Mochizuki S, Noda M, Koboyashi K 1999 Crystal structure of human serum albumin at 2.5 angstrom resolution. Protein Eng 12:439–446.

4. Cooke NE, Haddad JG 2003 In: “Vitamin D” Feldman D, Glorieux FH, Pike JW (eds) Vitamin D; 1st ed. Academic Press, Orlando, FL, pp. 87–101. 5. Curry S, Mandelkow H, Brick P, Franks N 1998 Crystal structure of human serum albumin complexed with fatty acid reveals an asymmetric distribution of binding sites. Nat Struct Biol 5:827–835. 6. Williams MH, Vanalstyne EL, Galbraith RM 1988 Evidence of a novel association of unsaturated fatty-acids with Gc (vitamin-D-binding protein). Biochem Biophys Res Commun 153:1019–1024.

CHAPTER 9 New Aspects of DBP

7. Ena JM, Esteban C, Perez MD, Uriel J, Calvo M 1989 Fattyacids bound to vitamin-D-binding protein (Dbp) from human and bovine sera. Biochem Int 19:1–7. 8. Bouillon R, Xiang DZ, Convents R, Vanbaelen H 1992 Polyunsaturated fatty-acids decrease the apparent affinity of vitamin-D metabolites for human vitamin D–binding protein. J Steroid Biochem Mol Biol 42:855–861. 9. Bhattacharya AA, Grune T, Curry S 2000 Crystallographic analysis reveals common modes of binding of medium and long-chain fatty acids to human serum albumin. J Mol Biol 303:721–732. 10. Barton GJ 1993 Alscript—a tool to format multiple sequence alignments. Protein Eng 6:37–40. 11. White P, Cooke N 2000 The multifunctional properties and characteristics of vitamin D-binding protein. Trends Endocrinol Metab 11:320–327. 12. Haddad JG, Hu YZ, Kowalski MA, Laramore C, Ray K, Robzyk P, Cooke NE 1992 Identification of the sterol-binding and actin-binding domains of plasma vitamin-D bindingprotein (Gc-globulin). Biochemistry 31:7174–7181. 13. Bouillon R, Okamura WH, Norman AW 1995 Structure– function relationships in the vitamin-D endocrine system. Endocr Rev 16:200–257. 14. Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179. 15. Ashcroft DM, Po ALW, Williams HC, Griffiths CEM 2000 Systematic review of comparative efficacy and tolerability of calcipotriol in treating chronic plaque psoriasis. Br Med J 320:963–967. 16. van den Bemd GJCM, Pols HAP, van Leeuwen JPTM 2000 Anti-tumor effects of 1,25-dihydroxyvitamin D-3 and vitamin D analogs. Curr Pharm Design 6:717–732. 17. Casteels K, Waer M, Laureys J, Valckx D, Depovere J, Bouillon R, Mathieu C 1998 Prevention of autoimmune destruction of syngeneic islet grafts in spontaneously diabetic nonobese diabetic mice by a combination of a vitamin D-3 analog and cyclosporine. Transplantation 65:1225–1232. 18. Cantorna MT, Hayes CE, DeLuca HF 1996 1,25-Dihydroxyvitamin D-3 reversibly blocks the progression of relapsing encephalomyelitis, a model of multiple sclerosis. Proc Natl Acad Sci USA 93:7861–7864. 19. Dusso AS, Negrea L, Gunawardhana S, Lopezhilker S, Finch J, Mori T, Nishii Y, Slatopolsky E, Brown AJ 1991 On the mechanisms for the selective action of vitamin-D analogs. Endocrinology 128:1687–1692. 20. Bouillon R, Allewaert K, Xiang DZ, Tan BK, Vanbaelen H 1991 Vitamin-D analogs with low affinity for the vitamin-D binding-protein—Enhanced in vitro and decreased in vivo activity. J Bone Miner Res 6:1051–1057. 21. Brown AJ 2000 Mechanisms for the selective actions of vitamin D analogues. Curr Pharm Design 6:701–716. 22. Tocchini-Valentini G, Rochel N, Wurtz JM, Mitschler A, Moras D 2001 Crystal structures of the vitamin D receptor complexed to superagonist 20-epi ligands. Proc Natl Acad Sci USA 98:5491–5496. 23. Lee WM, Galbraith RM 1992 Mechanisms of disease—the extracellular actin-scavenger system and actin toxicity. N Engl J Med 326:1335–1341. 24. Vasconcellos CA, Lind SE 1993 Coordinated inhibition of actin-induced platelet-aggregation by plasma gelsolin and vitamin-D-binding protein. Blood 82:3648–3657. 25. Van Baelen H, Bouillon R, De Moor P 1980 Vitamin-Dbinding protein (Gc-globulin) binds actin. J Biol Chem 255: 2270–2272.

151 26. Lind SE, Smith DB, Janmey PA, Stossel TP 1986 Role of plasma gelsolin and the vitamin-D-binding protein in clearing actin from the circulation. J Clin Invest 78:736–742. 27. Goldschmidt-Clermont PJ, Van Baelen H, Bouillon R, Shook TE, Williams MH, Nel AE, Galbraith RM 1988 Role of group-specific component (vitamin-D binding-protein) in clearance of actin from the circulation in the rabbit. J Clin Invest 81:1519–1527. 28. Dueland S, Nenseter MS, Drevon CA 1991 Uptake and degradation of filamentous actin and vitamin-D-binding protein in the rat. Biochem J 274:237–241. 29. Herrmanssdoerfer AJ, Heeb GT, Feustel PJ, Estes JE, Keenan CJ, Minnear FL, Selden L, Giunta C, Flor JR, Blumenstock FA 1993 Vascular clearance and organ uptake of G- and F-actin in the rat. Am J Physiol 265:G1071–G1081. 30. Verboven C, Bogaerts I, Waelkens E, Rabijns A, Van Baelen H, Bouillon R, De Ranter C 2003 Actin-DBP: the perfect structural fit? Acta Crystallogr D Biol Crystallogr 59: 263–273. 31. Houmeida A, Hanin V, Constans J, Benyamin Y, Roustan C 1992 Localization of a vitamin-D-binding protein-interaction site in the COOH-terminal sequence of actin. Eur J Biochem 203:499-503. 32. McLaughlin PJ, Gooch JT, Mannherz HG, Weeds AG 1993 Structure of gelsolin segment-1-actin complex and the mechanism of filament severing. Nature 364:685–692. 33. Robinson RC, Mejillano M, Le VP, Burtnick LD, Yin HL, Choe S 1999 Domain movement in gelsolin: a calciumactivated switch. Science 286:1939–1942. 34. Kabsch W, Mannherz HG, Suck D, Pai EF, Holmes KC 1990 Atomic structure of the actin–DNase-I complex. Nature 347:37–44. 35. Otterbein LR, Cosio C, Graceffa P, Dominguez R 2002 Crystal structures of the vitamin D-binding protein and its complex with actin: structural basis of the actin-scavenger system. Proc Nat Acad Sci USA 99:8003–8008 36. Mcleod JF, Kowalski MA, Haddad JG 1989 Interactions among serum vitamin-D binding-protein, monomeric actin, profilin, and profilactin. J Biol Chem 264:1260–1267. 37. Schutt CE, Myslik JC, Rozycki MD, Goonesekere NCW, Lindberg U 1993 The structure of crystalline profilin betaactin. Nature 365:810–816. 38. Goldschmidt-Clermont PJ, Galbraith RM, Emerson DL, Marsot F, Nel AE, Arnaud P 1985 Distinct sites on the the G-actin molecule bind group-specific component and deoxy-ribonuclease-I. Biochem J 228:471–477. 39. Goldschmidt-Clermont PJ, Vanalstyne EL, Day JR, Emerson DL, Nel AE, Lazarchick J, Galbraith RM 1986 Groupspecific component (vitamin-D binding-protein) prevents the interaction between G-actin and profilin. Biochemistry 25: 6467–6472. 40. Janmey PA, Stossel TP, Lind SE 1986 Sequential binding of actin monomers to plasma gelsolin and its inhibition by vitamin-D-binding protein. Biochem Biophy Res Commun 136:72–79. 41. Hesterkamp T, Weeds AG, Mannherz HG 1993 The actin monomers in the ternary gelsolin–2-actin complex are in an antiparallel orientation. Eur J Biochem 218:507–513. 42. Safer D, Sosnick TR, Elzinga M 1997 Thymosin β(4) binds actin in an extended conformation and contacts both the barbed and pointed ends. Biochemistry 36:5806–5816. 43. Van Troys M, Dewitte D, Goethals M, Carlier MF, Vandekerckhove J, Ampe C 1996 The actin binding site of thymosin β 4 mapped by mutational analysis. EMBO J 15:201–210.

152

ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON

44. Fedorov AA, Lappalainen P, Fedorov EV, Drubin DG, Almo SC 1997 Structure determination of yeast cofilin. Nat Struct Biol 4:366–369. 45. Holmes KC, Popp D, Gebhard W, Kabsch W 1990 Atomic model of the actin filament. Nature 347:44–49. 46. Wegner A 1976 Head to tail polymerization of actin. J Mol Biol 108:139–150. 47. Steinmetz MO, Stoffler D, Hoenger A, Bremer A, Aebi U 1997 Actin: from cell biology to atomic detail. J Struct Biol 119:295–320. 48. Vercaeren I, Winderickx J, Devos A, Peeters B, Heyns W 1993 An effect of androgens on the length of the poly(A)-tail and alternative splicing cause size heterogeneity of the messenger ribonucleic acids encoding cystatin-related protein [corrected and republished with original paging, article originally printed in 1992 Endocrinology 131(6):2496–2502]. Endocrinology 132:2496–2502. 49. Schiodt FV, Ott P, Bondesen S, Tygstrup N 1997 Reduced serum Gc-globulin concentrations in patients with fulminant hepatic failure: association with multiple organ failure. Crit Care Med 25:1366–1370. 50. Dahl P, Schiodt FV, Klaer T, Ott P, Bondesen S, Tygstrup N 1998 Serum Gc-globulin in the early course of multiple trauma. Crit Care Med 26:285–289. 51. Schiodt FV, Bondesen S, Tygstrup N 1995 Serial measurements of serum Gc-globulin in acetaminophen intoxication. Eur J Gastroenterol Hepatol 7:635–640.

52. Lee WM, Galbraith RM, Watt GH, Hughes RD, McIntire DD, Hoffman BJ, Williams R 1995 Predicting survival in fulminant hepatic-failure using serum Gc protein concentrations. Hepatology 21:101–105. 53. Schiodt FV, Bondesen S, Petersen I, Dalhoff K, Ott P, Tygstrup N 1996 Admission levels of serum Gc-globulin: predictive value in fulminant hepatic failure. Hepatology 23:713–718. 54. Dahl B, Schiodt FV, Nielsen M, Kiaer T, Williams JG, Ott P 1999 Admission level of Gc-globulin predicts outcome after multiple trauma. Injury—Int J Care Injured 30: 275–281. 55. Safadi FF, Thornton P, Magiera H, Hollis BW, Gentile M, Haddad JG, Liebhaber SA, Cooke NE 1999 Osteopathy and resistance to vitamin D toxicity in mice null for vitamin D binding protein. J Clin Invest 103:239–251. 56. Otterbein LR, Graceffa P, Dominguez R 2001 The crystal structure of uncomplexed actin in the ADP state. Science 293:708–711. 57. Head JF, Swamy N, Ray R 2002 Crystal structure of the complex between actin and human vitamin D–binding protein at 2.5 angstrom resolution. Biochemistry 41: 9015–9020. 58. Mendelson R, Morris EP 1997 The structure of the actomyosin subfragment 1 complex: results of searches using data from electron microscopy and x-ray crystallography. Proc Natl Acad Sci USA 94:8533–8538.

CHAPTER 10

Endocytic Pathways for 25-(OH) Vitamin D3 THOMAS E. WILLNOW ANDERS NYKJAER

Division of Molecular Cardiovascular Research Max-Delbrueck-Center for Molecular Medicine D-13125 Berlin, Germany Institute of Medical Biochemistry, University of Aarhus, DK-8000 Aarhus C, Denmark

I. Introduction II. Renal Endocytosis of 25-(OH) Vitamin D3

III. Conclusions References

I. INTRODUCTION

II. RENAL ENDOCYTOSIS OF 25-(OH) VITAMIN D3

In the 1970s, Fraser and Kodicek [1] and DeLuca and co-workers [2] were the first to demonstrate that the kidney is responsible for the conversion of 25-(OH) vitamin D3 to a more polar metabolite that acts as a potent regulator of calcium transport in the intestinal mucosa. Shortly thereafter, this metabolite was identified as 1,25-(OH)2 vitamin D3 [3,4]. Conversion was shown to take place in the epithelial cells of the proximal convoluted tubules (PCT) [5,6] by the action of 25-hydroxyvitamin D3-1α-hydroxylase, a cytochrome P450 enzyme located in the mitochondria [7]. A large body of work has firmly established the pivotal role of the PCT in the conversion of 25-(OH) vitamin D3 to 1,25-(OH)2 vitamin D3, and the regulatory steps involved in this process [reviewed in 8,9]. The route, however, that delivers the precursor to PCT cells for metabolism has been less well characterized. In the circulation, 25-(OH) vitamin D3 is transported bound to the carrier vitamin D–binding protein (DBP), also known as group-specific component of serum (Gc-globulin) [10]. Similar to other steroids, 25-(OH) vitamin D3 can dissociate from its carrier and passively diffuse through the plasma membrane of target cells. In PCT, the uptake of the free vitamin is believed to proceed from the circulation via the basolateral site of the epithelium [11,12]. Recent studies have identified an alternative route for delivery of 25-(OH) vitamin D3 to the PCT. It involves endocytosis of 25-(OH) vitamin D3/DBP complexes via endocytic receptors expressed on the luminal surface of the epithelium. This chapter describes the role of endocytic pathways for 25-(OH) vitamin D3/DBP complexes in vitamin D metabolism, and the molecular components involved in this process. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

A. Physiology 1. 25-(OH) VITAMIN D3/DBP COMPLEXES ARE SUBJECT TO GLOMERULAR FILTRATION

Patients with nephrotic syndrome suffer from hypocalcemia due to low serum levels of 25-(OH) vitamin D3 and 1,25-(OH)2 vitamin D3. Hypovitaminosis D is caused by an increased urinary loss of 25-(OH) vitamin D3 bound to DBP, depleting the plasma vitamin D pool [13–15]. The same phenotype can be observed in nephrotic rats [16]. Because no 25-(OH) vitamin D3 or DBP is present in the urine of control subjects, increased glomerular permeability in nephrotic syndrome was held responsible for an abnormal filtration of 25-(OH) vitamin D3/DBP complexes. In the healthy kidney, 25-(OH) vitamin D3/DBP complexes were believed to be excluded from glomerular filtration. This assumption has to be revised because a number of studies now demonstrate that circulating 25-(OH) vitamin D3/DBP complexes are normally filtered through the glomerulus and retrieved by PCT cells from the primary urine. Foremost, this fact has been demonstrated in patients suffering from proximal tubular resorption deficiencies (i.e, Fanconi syndrome) [17] such as in Itai–Itai disease (cadmium-induced tubulopathy) [18], Imerslund–Grasbeck syndrome (IGS) (see Section II,C,1,b) [19] or multiple myeloma [20]. In these individuals, the glomerular filtration barrier is intact and only low-molecular-weight metabolites are lost into the urine because of the inability of the PCT to retrieve physiological ligands from the tubule lumen. Patients with tubular resorption deficiencies of various etiologies excrete DBP (Fig. 1A) and 25-(OH) Copyright © 2005, Elsevier, Inc. All rights reserved.

154

THOMAS E. WILLNOW AND ANDERS NYKJAER

A

1

B

2

3

4

5

6

7

8

9

FIGURE 1 Excretion of DBP in patients and in mice with tubular resorption deficiencies. (A) Western blot analysis demonstrates the presence of DBP in urine samples from patients with multiple myeloma (light chain disease) (lane 1), cystinosis (lane 3), Fanconi–Bickel syndrome (lane 4), and Imerslund–Grasbeck syndrome (lane 5), but not from a healthy control subject (lane 2). (Samples provided by D. Müller, Humboldt University Berlin.) (B) DBP excretion is also seen in urine samples from megalindeficient (lanes 6 and 8) but not from wild-type mice (lanes 7 and 9).

vitamin D3 [20], indicating that the vitamin–carrier complex is subject to normal filtration. Thus, the urinary loss of 25-(OH) vitamin D3/DBP in nephrotic syndrome may be explained in part by the excessive amounts of filtered high-molecular-weight proteins that saturate tubular retrieval pathways. Further evidence for a physiological role of glomerular filtration in renal vitamin D metabolism stems from the fact that glomerular filtration rates directly correlate with plasma 1,25(OH)2 vitamin D3 levels [21]. 2. MEGALIN, AN ENDOCYTIC RECEPTOR FOR 25-(OH) VITAMIN D3/DBP

The identity of the tubular retrieval pathway for 25-(OH) vitamin D3/DBP and its significance for vitamin D metabolism was uncovered through studies on megalin, an endocytic receptor abundantly expressed on the brush border (luminal) surface of the PCT cells (Fig. 2) [22,23]. Megalin is a scavenging receptor with broad ligand specificity that constitutes the main pathway for tubular reabsorption of filtered plasma proteins [24]. Mice genetically deficient for this receptor suffer from low-molecular-weight proteinuria and excrete the receptor ligands [20,25–27]. Among other proteins, megalin knockout mice lose DBP into the urine, which is normally retrieved by this receptor (Fig. 1B). No uptake of DBP is seen in megalin-deficient kidneys whereas endocytosis of DBP can easily be detected in renal tissue from control animals (Fig. 2). Both mice with an obligate or with a conditional kidney-specific megalin gene defect were used to evaluate the pathophysiological consequences of urinary

loss of DBP. These studies have highlighted the importance of tubular retrieval pathways for renal and systemic vitamin D metabolism [20,25,28,29]. Megalin-deficient mice excrete approximately 0.3 mg of DBP per day [25]. Because the same amount of protein is retrieved from the glomerular filtrate in a normal kidney in 24 hr, glomerular filtration likely accounts for most of the plasma turnover of the carrier protein [30]. Massive urinary loss of DBP in megalin knockout mice results in a concomitant loss of 25-(OH) vitamin D3 bound to the carrier protein (Table I). The total 25-(OH) vitamin D3 loss amounts to 0.3 to 0.5 ng in 24 hr [25]. On a vitamin D3–enriched diet, the plasma levels of 25-(OH) vitamin D3 and 1,25-(OH)2 vitamin D3 are reduced by more than 70% in the obligate and by 50% in the conditional knockout line (Table I). The less severe phenotype in the latter mouse model is due to the residual megalin activity that is retained in animals with kidney-specific megalin gene deletion (10% of normal levels) [29]. Owing to the dietary supplementation with vitamin D3, megalin-deficient mice can balance the constant urinary loss of 25-(OH) vitamin D3 and exhibit normal plasma calcium, phosphorus, and PTH levels. However, when placed on a vitamin D–depleted chow, the animals suffer from hypocalcemia and hyperparathyroidism (Table I) [29]. Their bones are characterized by a massive increase in osteoid surfaces and by a dramatic reduction in inactive and mineralizing surfaces and in total bone mineral content (Table II). Such features of osteomalacia as a consequence of hypovitaminosis D are also observed in mice with inactivation of the DBP gene [31], highlighting

155

Anti-DBP

Anti-megalin

CHAPTER 10 Endocytic Pathways for 25-(OH) Vitamin D3

FIGURE 2 Immunodetection of megalin and DBP in murine PCT. In wild-type (+/+) PCT cells, megalin (arrowheads) is present on the luminal cell surface where it internalizes filtered DBP, detectable in intracellular vesicles (arrows). No uptake of the carrier is seen in PCT of megalin knockout mice (−/−).

TABLE I Endocrine Parameters in Wild-Type Mice (Control) and in Animals with Obligate (Megalin−/−) or Kidney-Specific Megalin Gene Defect (Conditional) Genotypes Parameters Vitamin D–enriched diet Urine 25-OH D3 (nM/mM creatinine) Plasma 25-OH D3 (nM) 1,25-(OH)2 D3 (pM) PTH (pM) Calcium (mM) Phosphorous (mM) Vitamin D–depleted diet Plasma PTH (pM) Calcium (mM) Phosphorous (mM)

Control

n

Conditional

n

Megalin−/−

n

n.d.

8

10.9 ± 1.9

13

10.4 ± 4.1

6

—

92.8 ± 8.3 120.1 ± 7.0 52.1 ± 4.0 2.3 ± 0.04 3.2 ± 0.21

9 5 5 5 5

48.1 ± 6.4 66.4 ± 14.6 67.0 ± 6.3 2.2 ± 0.04 3.0 ± 0.08

8 4 5 5 5

27.2 ± 8.3 43.2 ± 17.5 68.5 ± 1.8 — —

4 4 3

0.0006 0.009 0.08 0.6 0.56

72.6 ± 4.0 2.2 ± 0.03 3.0 ± 0.08

10 10 10

88.3 ± 1.8 1.8 ± 0.1 3.1 ± 0.1

5 9 9

— — —

P

0.04 0.03 0.3

Data are the mean ± SEM. Statistical significance of differences between values in controls and conditional knockout mice are indicated (P). n, number of animals; n.d., not detectable.

156

THOMAS E. WILLNOW AND ANDERS NYKJAER

TABLE II Bone Morphometric Analysis of Wild Type Mice (Control) or Animals with Kidney-Specific Megalin Gene Defect (Conditional) on a Vitamin D–Depleted Diet (Mean ± SEM; n, number of animals) Parameters Bone mineral content in femur and tibia (mg) Osteoid surface (% of total surface) Inactive surfaces (% of total surface) Mineralizing surface (% of total surface)

Control (n = 5)

Conditional (n = 4)

P

50.4 ± 3.3 3.0 ± 1.7 92.4 ± 1.7 31.8 ± 6.0

29.5 ± 5.0 67.2 ± 17.9 18.7 ± 16.1 2.2 ± 0.9

< 0.01 0.02 < 0.01 < 0.01

the close functional relationship of the endocytic receptor megalin and the carrier DBP in vitamin D metabolism. Further support for a crucial role of the megalin retrieval pathway in uptake of 25-(OH) vitamin D3 comes from studies in rats treated with megalin antagonists. When receptor antagonists are infused into PCT of the rat, no conversion of 25-(OH) vitamin D3 to 1,25-(OH)2 vitamin D3 is seen, indicating that the megalin receptor pathway provides PCT cells with the precursor to produce the active hormone 1,25-(OH)2 vitamin D3 [25].

B. Cell Biology According to current hypotheses, free steroid hormones enter cells by passive diffusion through the

BN16 cells

35 Celluar uptake (% of added tracer)

plasma membrane. In contrast, steroids bound to transport proteins are considered biologically inactive and blocked from cell entry by their carriers [32]. This concept does not apply to the secosteroid 25-(OH) vitamin D3 complexed with DBP as shown in rat and mouse models [25, 29] and in cultured cells [19]. When Brown Norway rat choriocarcinoma (BN16) cells are incubated with 25-(OH) vitamin D3/DBP complexes more than 30% of the complexes are taken up within 4 hr (Fig. 3). The cellular uptake of the steroid is dependent on megalin, which is abundantly expressed in these cells, and can be blocked by a megalin antagonist, the receptor-associated protein (RAP) [33], or by anti-megalin antibodies. Very little uptake of complexed 3H-25-(OH) vitamin D3 is seen in keratinocytes that lack megalin expression (Fig. 3). Similar results have been obtained in other cell lines not expressing the megalin, indicating

30 25-OH-D3 /125I-DBP 25

3H-25-OH -D/DBP 3

20 15 10

Keratinocytes

5 0

FIGURE 3

buffer

α-megalin lgG

sheep lgG

RAP

buffer

Cellular uptake of 25-(OH) vitamin D3/DBP complexes. Brown Norway rat choriocarcinoma (BN16) cells and keratinocytes were incubated for 4 hr with 25-(OH) vitamin D3/DBP complexes labeled on either the steroid (3H-25-(OH) vitamin D3/DBP, open bars) or the protein moiety (25-(OH) vitamin D3/125I-DBP, closed bars). Cellular uptake of the carrier and the vitamin can be detected in megalin-expressing BN16 cells but not in keratinocytes that lack the receptor. Megalin-mediated uptake of complexed 25-(OH) vitamin D3 is blocked by the megalin antagonist receptor-associated protein (RAP) and by sheep anti-megalin antibodies, but not by control sheep IgG.

157

CHAPTER 10 Endocytic Pathways for 25-(OH) Vitamin D3

FIGURE 4

This immunoelectron micrograph depicts internalized DBP (arrows) in a lysosome of a murine PCT cell. (Provided by E. I. Christensen, University of Aarhus.)

that the efficient uptake of metabolites complexed with DBP requires the presence of megalin [34–37]. Like most internalized proteins, DBP is transported to lysosomes for degradation (Fig. 4) [25,38]. The intracellular route taken by 25-(OH) vitamin D3 through the endocytic compartments to the mitochondria is unknown and probably involves intracellular 25-(OH) vitamin D binding proteins (IDBP) (see Chapter 21) [39].

autoantigen in an induced glomerular nephritis model in rats, the passive Heymann nephritis [40]. In this animal model, autoantibodies directed against megalin form immunodeposits in the glomerular wall, causing severe tissue damage and renal failure [40, 41]. Megalin is a giant 600-kDa cell surface protein. Its gene covers 235,000 base pairs on the human chromosome 2q24-q31 [42] and consists of 79 exons [43]. The protein is highly conserved among species and can even be found in primitive organisms such as nematodes [44]. Sequencing of the rat and human megalin cDNA revealed a close structural similarity of this receptor with members of the LDL receptor gene family, a group of multifunctional endocytic receptors [43,45,46]. Similar to other family members, megalin is characterized by a number of structural domains required to perform endocytosis (Fig. 5). Its extracellular domain consists of 36 cysteine-rich complement-type repeats,

Complement-type repeat EGF YWTD

NH2

EGF precursor homology domain

CUB domain EGF-type repeat Amphipathic hellx

Membrane anchor NPXY motif

COOH

C. Molecular Biology Investigations in rats treated with megalin antagonists [25] or in mice with induced megalin gene defect [25,29] have established a central role of this receptor in renal uptake of 25-(OH) vitamin D3 metabolites in vivo. Recent studies have uncovered the molecular details of this pathway. Besides the receptor megalin itself, it involves co-receptors (cubilin), cellular adaptors (disabled-2), and components of the endocytic machinery (voltage-gated chloride channel-5), as well as extracellular (DBP) and intracellular 25-(OH) vitamin D3-binding proteins (IDBP). DBP and IDBP will only be mentioned briefly in this chapter because they are covered elsewhere in this textbook (Chapters 8 and 21, respectively). Other components of the pathway will be described in the following sections. 1. RECEPTORS AND CO-RECEPTORS

a. Megalin Megalin, also known as glycoprotein (gp) 330 or low-density lipoprotein (LDL) receptorrelated protein-2, was initially identified as the

NH2

Membrane

Megalin

COOH

Membrane

Cubilin

FIGURE 5 (Left panel) Structure of megalin. The extracellular domain of the receptor consists of clusters of complement-type repeats and epidermal growth factor (EGF) presursor homology domains. The clusters of complement-type repeats are the site of ligand binding. YWTD repeats in the EGF precursor homology domains are responsible for pH-dependent release of ligands. A single transmembrane domain mediates membrane attachment. NPXY motifs in the cytoplasmic tail are involved in internalization via coated pits and in cytosolic adaptor binding. (Right panel) Structure of cubilin. The cubilin polypeptide consists of 27 CUB domains, a cluster of 8 EGF repeats, and an amphipathic helix. CUB domains are involved in ligand binding. The amino-terminal amphipathic helix mediates peripheral membrane attachment.

158

bil in

ali n

Cu

α-

αM eg

organized in four clusters of seven to 11 repeats. The second cluster represents the main site of ligand binding [47]. All clusters of complement-type repeats are separated by epidermal growth factor (EGF) precursor homology domains composed of EGF-type repeats and tyrosine-tryptophan-threonine-aspartate (YWTD)containing spacer regions. The later sites are involved in pH-dependent release of ligands in endosomes [48]. A 22-amino-acid transmembrane domain mediates membrane attachment of the receptor. The cytoplasmic tail contains three asparagine-proline-X-tyrosine (NPXY) elements that represent signals for coated-pit internalization [49] and binding sites for cytosolic adapter proteins such as Disabled (Dab)-2 [50]. Megalin is expressed in many absorptive epithelia of embryonic and adult tissues, the predominant sites of expression being the yolk sac and the neuroepithelium of early embryos and the PCT and the small intestine in adults [24]. In these tissues, megalin expression is restricted to the apical cell surface and to endosomal compartments in line with a role of this receptor in apical clearance pathways [23]. Many ligands have been identified that bind to megalin in vitro and that are taken up by the receptor into cultured cells or in vivo. By and large, these ligands represent three major classes of molecules: (1) proteases and protease/inhibitor complexes, (2) lipoproteins, and (3) vitamins or hormones bound to carrier proteins [reviewed in 24,51]. Besides 25-(OH) vitamin D3/DBP, the receptor reabsorbs complexes of retinol (vitamin A) with the retinol-binding protein [26] and vitamin B12 with the carrier transcobalamin [27] from the primary urine into PCT, indicating a role as a general retrieval receptor for filtered metabolites. The function of the receptor in other epithelia is less well characterized. Forebrain defects in mice with obligate megalin gene deletion argue for an important role of the protein in brain development [52]. Megalin binds most ligands with moderate affinities (Kd 100–500 nM). Ligand binding is dependent on the presence of calcium that is required to stabilize the three-dimensional structure of the complement-type repeats [53]. DBP binds to megalin with a Kd of 108 nM regardless of whether 25-(OH) vitamin D3 is bound to the carrier or not [25]. b. Cubilin When Nykjaer et al. [19] used DBPaffinity chromatography to purify membrane proteins from renal tissues that bind this carrier, two DBPinteracting proteins were identified (Fig. 6). One was megalin, the other cubilin, a peripheral membrane protein expressed in the PCT and other epithelia [54]. Cubilin was initially identified as the receptor for uptake of intrinsic factor (IF)/vitamin B12 complexes in the

THOMAS E. WILLNOW AND ANDERS NYKJAER

DBP-Column

250

96 68 45 30 1

2

3

4

5

6

7

FIGURE 6 The kidney expresses two DBP receptors. Two proteins can be recovered by DBP-affinity chromatography from renal membrane extracts. Lanes 1–5 depict elution fractions from a DBPcolumn separated by SDS–polyacrylamide gel electrophoresis and stained with silver nitrate. Two high-molecular-weight DBP-binding proteins elute from the column, megalin and cubilin. Lanes 6 and 7 are immunoblots of the fraction in lane 4 incubated with antimegalin (lane 6) or anti-cubilin antisera (lane 7).

intestine and is therefore also known as the intrinsic factor/cobalamin receptor [55]. The human cubilin gene is located on chromosome 10p12.1 [56]. The sequence gives rise to a 460-kDa protein. So far, cubilin has been cloned from humans, rats, and dogs [56–58]. It received its name from a cluster of so called CUB domains that make up most of the receptor polypeptide (Fig. 5). CUB domains are 120 amino acids modules found in components of the complement system, such as C1r/C1s, in the EGF related sea urchin protein UEGF and in the bone morphogenic protein, BMP-1 (hence the name CUB). In addition to 27 CUB domains, cubilin possesses a short amino terminal amphipathic helix and 8 EGF-like repeats [57]. In contrast to prototype endocytic receptors such as megalin, cubilin lacks a membrane anchor and a cytosolic tail and is believed to associate with the plasma membrane via its hydrophobic amino terminal helix [59]. The protein directly binds to megalin and recycles through the endocytic compartments as a co-receptor complex (see Section II,C,1,c). In mice genetically deficient for megalin, cubilin biosynthesis and intracellular transport are significantly impaired [60]. Therefore, PCTs lacking megalin are also functionally deficient for cubilin [60, 61]. Besides in the PCT, cubilin is expressed in several absorptive epithelia including the yolk sac and the ileum.

159

CHAPTER 10 Endocytic Pathways for 25-(OH) Vitamin D3

At the ultrastructural level, cubilin with megalin colocalizes in the apical part of epithelial cells, in particular in clathrin-coated pits, in dense apical tubules, and in endosomes, in agreement with the requirement of the receptor to associate with megalin [62]. Investigations into the physiological role of cubilin have been facilitated by the identification of natural mutations in the corresponding gene in patients [63]. Defects in the cubilin gene were found in individuals suffering from megaloblastic anemia 1 or ImerslundGrasbeck syndrome (IGS), a rare recessive disorder characterized by a reduced number of red blood cells displaying an increased cell volume [63,64]. Consistent with a crucial role of endocytic receptors in tubular retrieval of filtered metabolites in the kidney, several plasma proteins have been identified that are cleared from the primary urine via cubilin. These ligands include DBP [19], transferrin [65], and albumin [61], as well as Clara cell secretory protein (CCSP), a transporter for steroid hormones [60]. As a consequence of the cubilin gene defect, patients with IGS exhibit lowmolecular-weight proteinuria and excrete ligand/carrier complexes such as 25-(OH) vitamin D3/DBP [19]. c. The Two-Receptor Model Based on studies in a number of experimental systems, Nykjaer and Willnow proposed a two-receptor model for the binding of 25-(OH) vitamin D3/DBP to the PCT cell surface and for the subsequent cellular uptake of the vitamin/carrier complexes (Fig. 7) [19,66]. According to this model, complexes of DBP and 25-(OH) vitamin D3 are reabsorbed from the glomerular filtrate by association of DBP with megalin, followed by endocytic uptake. Alternatively, the complexes are bound by cubilin first and internalized via interaction of this receptor

FIGURE 7 Two-receptor model for renal uptake of 25-(OH) vitamin D3/DBP. Details of this model are given in the text.

with megalin. Because cubilin acts as a surrogate binding site for DBP on the PCT surface, cubilin deficiency partially affects renal reabsorption of 25-(OH) vitamin D3, whereas DBP or megalin gene defects completely eliminate the uptake pathway. This two-receptor model is confirmed by the moderate vitamin D deficiency observed in cubilin-deficient patients and dogs as compared to DBP or megalin knockout mice [19,25,29,31]. 2. DISABLED-2

To recycle through the endocytic compartments, receptors require the assistance of scaffold or adaptor proteins, cytoplasmic factors that bind to the tail of the receptors and regulate their trafficking. Some adaptor proteins are common to most endocytic receptors such as AP-2 that localize receptors to the clathrin coat of endocytic vesicles [67]. Other adaptors are specific for certain receptors or even cell types. Disabled (Dab)-2 is an adaptor protein required for proper function of megalin in PCT. It harbors a phosphotyrosine-binding (PTB) domain with which it attaches to an NPXY motif in the megalin tail sequences [50]. Lack of Dab-2 expression in knockout mice results in impaired tubular endocytosis and in excretion of DBP [68]. The exact mode of action of Dab-2 still remains to be elucidated. However, findings obtained in Dab-2-deficient mice suggest a direct role of the adaptor in megalin trafficking and in endocytosis of 25-(OH) vitamin D3/DBP complexes. 3. VOLTAGE-GATED CHLORIDE CHANNEL-5

Additional support for a role of endocytic pathways in the renal metabolism of 25-(OH) vitamin D3 stems from studies in patients with Dent’s disease, a rare X-linked hereditary disorder characterized by disturbances in calcium homeostasis and bone metabolism [69]. This disease is caused by mutations in the gene encoding voltage-gated chloride channel-5 (ClC-5), a Cl− transporter expressed in endosomes of PCTs [70–72]. ClC-5 is believed to be responsible for sustaining chloride conductance required for import of H+ and acidification of the endosomes. Thus, ClC-5 defects in patients result in tubular endocytic dysfunction and in the inability to reabsorb filtered plasma proteins from the lumen of the PCT [73]. One explanation for defects in calcium homeostasis in ClC-5-deficient individuals comes from observations in mice lacking this ion channel. In ClC-5 knockout animals, endocytic malfunction results in disruption of megalin-mediated uptake of 25-(OH) vitamin D3/DBP complexes, in urinary loss of these metabolites, and in a three-fold decrease in plasma levels of 25-(OH) vitamin D3 and 1,25-(OH)2 vitamin D3 [74].

160

THOMAS E. WILLNOW AND ANDERS NYKJAER

III. CONCLUSION In recent years, a number of genetic models affecting renal vitamin D metabolism in patients and laboratory animals have been elucidated. Taken together, studies in these models have identified the existence of an endocytic pathway for 25-(OH) vitamin D3 metabolites in the PCT (Fig. 8). In this pathway, circulating 25-(OH) vitamin D3 metabolites are directed to the kidney through association with DBP, internalized from the glomerular filtrate by megalin and cubilin, two DBPbinding proteins that form a two-receptor complex, and delivered to lysosomal compartments. Recycling of the receptors megalin and cubilin requires Dab-2, a cytoplasmic adaptor for megalin. Inside the cells, DBP is degraded while 25-(OH) vitamin D3 is transferred into the cytoplasm for further activation and resecretion. The intracellular mechanisms that control the release of the vitamin D metabolite from the endocytic compartments and its transport to mitochondria, the site of hydroxylation, are unknown. However, the low solubility of 25-(OH) vitamin D3 strongly argues for the presence of intracellular transport proteins that facilitate the targeting of the steroid to the correct organelle. Such intracellular transport proteins have been identified for lipophilic compounds including vitamin A (cellular retinol–binding proteins) [75] and vitamin E (α-tocopherol transfer protein) [76]. Studies by Adams and co-workers indicate that two members of the constitutively expressed heat shock protein-70 family, designated intracellular vitamin D-binding protein-1 and -2 (IDBP-1, -2) serve similar functions for vitamin D [39,77].

FIGURE 8

The existence of an endocytic pathway for 25-(OH) vitamin D3 is intriguing as it is clearly distinct from the free diffusion of the metabolite into PCT. Uptake proceeds from the luminal rather than the basolateral side of the epithelium. Cell entry occurs in the DBP-bound, but not in the free form of the hormone. There is no doubt that this endocytic pathway plays a crucial role in the retrieval of filtered vitamin D molecules from the glomerular filtrate in order to prevent uncontrolled urinary loss of essential metabolites. The contribution of this pathway to production of 1,25-(OH)2 vitamin D3 in the kidney still remains to be clarified. However, the small fraction of free 25-(OH) vitamin D3 available for passive diffusion (0.003% of total) [78] and the large amount of precursor converted by the kidney every day (0.3 to 0.5 ng) [79] strongly argue for the existence of uptake pathways whereby the kidney gains access to the pool of bound 25-(OH) vitamin D3. The expression of megalin in other epithelia such as the small intestine, breast, and prostate that also exhibit 25-hydroxyvitamin D3-1α-hydroxylase activities may even indicate a role of this endocytic pathway in extrarenal conversion of 25-(OH) vitamin D3. In conclusion, the findings discussed in this chapter do not contradict other routes (such as free diffusion) for uptake of 25-(OH) vitamin D3 into the kidney (or other tissues). In fact, residual levels of 1,25-(OH)2 vitamin D3 are found in megalin- and DBP-deficient mice [29,31]. Whether they result from renal or extrarenal conversion of the precursor is unclear at present. Whatever the source of 1,25-(OH)2 vitamin D3 may be, this pool of the hormone is not able to sustain normal calcium and bone metabolism under dietary vitamin D restriction.

Endocytic pathway for renal uptake and conversion of 25-(OH) vitamin D3/DBP complexes. Details of this pathway are discussed in the text.

161

CHAPTER 10 Endocytic Pathways for 25-(OH) Vitamin D3

References 1. Fraser DR, Kodicek E 1970 Unique biosynthesis by kidney of a biological active vitamin D metabolite. Nature 228:764–766. 2. Gray R, Boyle I, DeLuca HF 1971 Vitamin D metabolism: the role of kidney tissue. Science 172:1232–1234. 3. Lawson DE, Fraser DR, Kodicek E, Morris HR, Williams DH 1971 Identification of 1,25-dihydroxycholecalciferol, a new kidney hormone controlling calcium metabolism. Nature 230:228–230. 4. Holick MF, Schnoes HK, DeLuca HF, Suda T, Cousins RJ 1971 Isolation and identification of 1,25-dihydroxycholecalciferol. A metabolite of vitamin D active in intestine. Biochemistry 10:2799–2804. 5. Brunette MG, Chan M, Ferriere C, Roberts KD 1978 Site of 1,25(OH)2 vitamin D3 synthesis in the kidney. Nature 276:287–289. 6. Kawashima H, Torikai S, Kurokawa K 1981 Localization of 25-hydroxyvitamin D3 1α-hydroxylase and 24-hydroxylase along the rat nephron. Proc Natl Acad Sci USA 78:1199–1203. 7. Paulson SK, DeLuca HF 1985 Subcellular location and properties of rat renal 25-hydroxyvitamin D3-1α-hydroxylase. J Biol Chem 260:11488–11492. 8. Bland R, Zehnder D, Hewison M 2000 Expression of 25-hydroxyvitamin D3-1α-hydroxylase along the nephron: new insights into renal vitamin D metabolism. Curr Opin Nephrol Hypertens 9:17–22. 9. Brown AJ, Dusso A, Slatopolsky E 1999 Vitamin D. Am J Physiol 277:F157–175. 10. Daiger SP, Schanfield MS, Cavalli-Sforza LL 1975 Groupspecific component (Gc) proteins bind vitamin D and 25-hydroxyvitamin D. Proc Natl Acad Sci USA 72:2076–2080. 11. Haddad JG 1995 Plasma vitamin D–binding protein (Gc-globulin): multiple tasks. J Steroid Biochem Mol Biol 53:579–582. 12. White P, Cooke N 2000 The multifunctional properties and characteristics of vitamin D–binding protein. Trends Endocrinol Metab 11:320–327. 13. Goldstein DA, Oda Y, Kurokawa K, Massry SG 1977 Blood levels of 25-hydroxyvitamin D in nephrotic syndrome. Studies in 26 patients. Ann Intern Med 87:664–667. 14. Barragry JM, France MW, Carter ND, Auton JA, Beer M Boucher BJ, Cohen RD 1977 Vitamin-D metabolism in nephrotic syndrome. Lancet 24:629–632. 15. Colston K, Williams NJ, Cleeve HJ 1985 Studies on vitamin D binding protein in the nephrotic syndrome. Clin Chem 31:718–721. 16. Mizokuchi M, Kubota M, Tomino Y, Koide H 1992 Possible mechanism of impaired calcium and vitamin D metabolism in nephrotic rats. Kidney Int 42:335–340. 17. Harrison HE 1953 The Fanconi syndrome. J Chron Dis 7: 346–355. 18. Teranishi H, Kasuya M, Aoshima K, Kato T, Migita S 1983 Demonstration of vitamin D–binding protein (Gc-globulin) in the urine of Itai-itai disease patients. Toxicol Lett 15:7-12. 19. Nykjaer A, Fyfe JC, Kozyraki R, Leheste JR, Jacobsen C, Nielsen MS, Verroust PJ, Aminoff M, de la Chapelle A, Moestrup SK, Ray R, Gliemann J, Willnow TE, Christensen EI 2001 Cubilin dysfunction causes abnormal metabolism of the steroid hormone 25(OH) vitamin D3. Proc Natl Acad Sci USA 98:13895–13900. 20. Leheste JR, Rolinski B, Vorum H, Hilpert J, Nykjaer A, Jacobsen C, Aucouturier P, Moskaug JO, Otto A, Christensen EI,

21.

22.

23.

24. 25.

26.

27.

28.

29.

30. 31.

32. 33. 34. 35. 36.

Willnow TE 1999 Megalin knockout mice as an animal model of low molecular weight proteinuria. Am J Pathol 155: 1361–1370. St John A, Thomas MB, Davies CP, Mullan B, Dick I, Hutchison B, van der Schaff A, Prince RL 1992 Determinants of intact parathyroid hormone and free 1,25-dihydroxyvitamin D levels in mild and moderate renal failure. Nephron 61: 422–427. Orlando RA, Farquhar MG 1993 Identification of a cell line that expresses a cell surface and a soluble form of the gp330/receptor-associated protein (RAP) Heymann nephritis antigenic complex. Proc Natl Acad Sci USA 90:4082–4086. Christensen EI, Nielsen S, Moestrup SK, Borre C, Maunsbach AB, de Heer E, Ronco P, Hammond TG, Verroust P 1995 Segmental distribution of the endocytosis receptor gp330 in renal proximal tubules. EJCB 66:349–364. Christensen EI, Willnow TE 1999 Essential role of megalin in renal proximal tubule for vitamin homeostasis. J Am Soc Nephrol 10:2224–2236. Nykjaer A, Dragun D, Walther D, Vorum H, Jacobsen C, Herz J, Melsen F, Christensen EI, Willnow TE 1999 An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) vitamin D3. Cell 96:507–515. Christensen EI, Moskaug JO, Vorum H, Jacobsen C, Gundersen TE, Nykjaer A, Blomhoff R, Willnow TE, Moestrup SK 1999 Evidence for an essential role of megalin in transepithelial transport of retinol. J Am Soc Nephrol 10:685–695. Birn H, Willnow TE, Nielsen R, Norden AG, Bonsch C, Moestrup SK, Nexo E, Christensen EI 2002 Megalin is essential for renal proximal tubule reabsorption and accumulation of transcobalamin-B12. Am J Physiol Renal Physiol 282: F408–F416. Hilpert J, Wogensen L, Thykjaer T, Wellner M, Schlichting U, Orntoft TF, Bachmann S, Nykjaer A, Willnow TE 2002 Expression profiling confirms the role of endocytic receptor megalin in renal vitamin D3 metabolism. Kidney Int 62: 1672–1681. Leheste JR, Melsen F, Wellner M, Jansen P, Schlichting U, Renner-Muller I, Andreassen TT, Wolf E, Bachmann S Nykjaer A, Willnow TE 2003 Hypocalcemia and osteopathy in mice with kidney-specific megalin gene defect. FASEB J 17: 247–249. Haddad JG, Fraser DR, Lawson DE 1981 Vitamin D plasma binding protein. Turnover and fate in the rabbit. J Clin Invest 67:1550–1560. Safadi FF, Thornton P, Magiera H, Hollis BW, Gentile M, Haddad JG, Liebhaber SA, Cooke NE 1999 Osteopathy and resistance to vitamin D toxicity in mice null for vitamin D binding protein. J Clin Invest 103:239–251. Mendel CM 1989 The free hormone hypothesis: a physiologically based mathematical model. Endocr Rev 10:232–274. Christensen EI, Gliemann J, Moestrup SK 1992 Renal tubule gp330 is a calcium binding receptor for endocytic uptake of protein. J Histochem Cytochem 40:1481–1490. Bikle DD, Gee E 1989 Free, and not total, 1,25-dihydroxyvitamin D regulates 25-hydroxyvitamin D metabolism by keratinocytes. Endocrinology 124:649–654. Vanham G, Van Baelen H, Tan BK, Bouillon R 1988 The effect of vitamin D analogs and of vitamin D–binding protein on lymphocyte proliferation. J Steroid Biochem 29:381–386. Adams JS 1984 Specific internalization of 1,25-dihydroxyvitamin D3 by cultured intestinal epithelial cells. J Steroid Biochem 20:857–862.

162 37. Manolagas SC, Deftos LJ 1980 Studies of the internalization of vitamin D3 metabolites by cultured osteogenic sarcoma cells and their application to a non-chromatographic cytoreceptor assay for 1,25-dihydroxyvitamin D3. Biochem Biophys Res Commun 95:596–602. 38. Keenan MJ, Holmes RP 1991 The uptake and metabolism of 25-hydroxyvitamin D3 and vitamin D binding protein by cultured porcine kidney cells (LLC-PK1). Int J Biochem 23:1225–1230. 39. Wu S, Ren S, Chen H, Chun RF, Gacad MA, Adams JS 2000 Intracellular vitamin D binding proteins: novel facilitators of vitamin D–directed transactivation. Mol Endocrinol 14: 1387–1397. 40. Kerjaschki D, Miettinen A, Farquhar MG 1987 Initial events in the formation of immune deposits in passive Heymann nephritis. gp330-anti-gp330 immune complexes form in epithelial coated pits and rapidly become attached to the glomerular basement membrane. J Exp Med 166:109–128. 41. Yamazaki H, Ullrich R, Exner M, Saito A, Orlando RA, Kerjaschki D, Farquhar MG 1998 All four putative ligandbinding domains in megalin contain pathogenic epitopes capable of inducing passive Heymann nephritis. J Am Soc Nephrol 9:1638–1644. 42. www.ncbi.nlm.nih.gov/cgi-bin/Entrez 43. Saito A, Pietromonaco S, Loo AK, Farquhar MG 1994 Complete cloning and sequencing of rat gp330/“megalin,” a distinctive member of the low density lipoprotein receptor gene family. Proc Natl Acad Sci USA 11 9725–9729. 44. Yochem J, Greenwald I 1993 A gene for a low density lipoprotein receptor-related protein in the nematode Caenorhabditis elegans. Proc Natl Acad Sci USA 90: 4572–4576. 45. Hjalm G, Murray E, Crumley G, Harazim W, Lundgren S, Onyango I, Ek B, Larsson M, Juhlin C, Hellman P, Davis H, Akerstrom G, Rask L, Morse B 1996 Cloning and sequencing of human gp330, a Ca2+-binding receptor with potential intracellular signaling properties. Eur J Biochem 239:132–137. 46. Willnow TE, Nykjaer A, Herz J 1999 Lipoprotein receptors: new roles for ancient proteins. Nat Cell Biol 1:E157–E162. 47. Orlando RA, Exner M, Czekay R-P, Yamazaki H, Saito A, Ullrich R, Kerjaschki D, Farquhar MG 1997 Identification of the second cluster of ligand-binding repeats in megalin as a site for receptor–ligand interactions. Proc Natl Acad Sci USA 94:2368–2373. 48. Rudenko G, Henry L, Henderson K, Ichtchenko K, Brown MS, Goldstein JL, Deisenhofer J 2002 Structure of the LDL receptor extracellular domain at endosomal pH. Science 298: 2353–2358. 49. Kibbey RG, Rizo J, Gierasch LM, Anderson RG 1998 The LDL receptor clustering motif interacts with the clathrin terminal domain in a reverse turn conformation. J Cell Biol 142:59–67. 50. Oleinikov AV, Zhao J, Makker SP 2001 Cytosolic adaptor protein Dab2 is an intracellular ligand of endocytic receptor gp600/ megalin. Biochem J 347:613–621. 51. Nykjaer A, Willnow TE 2002 The low-density lipoprotein receptor gene family: a cellular Swiss army knife? Trends Cell Biol 12:273–280. 52. Willnow TE, Hilpert J, Armstrong SA, Rohlmann A, Hammer RE, Burns DK, Herz J 1996 Defective forebrain development in mice lacking gp330/megalin. Proc Natl Acad Sci USA 93:8460–8464. 53. Fass D, Blacklow S, Kim PS, Berger JM 1997 Molecular basis of familial hypercholesterolaemia from structure of LDL receptor module. Nature 388:691–693.

THOMAS E. WILLNOW AND ANDERS NYKJAER

54. Christensen EI, Birn H, Verroust P, Moestrup SK 1998 Membrane receptors for endocytosis in the renal proximal tubules. Int Rev Cytol 180:237–284. 55. Seetharam B, Levine JS, Ramasamy M, Alpers DH 1988 Purification, properties, and immunochemical localisation of a receptor for intrinsic factor–cobalamin receptor in the rat kidney. J Biol Chem 263:4443–4449. 56. Kozyraki R, Kristiansen M, Silahtaroglu A, Hansen C, Jacobsen C, Tommerup N, Verroust PJ, Moestrup SK 1998 The human intrinsic factor vitamin B12 receptor, cubilin: molecular characterization and chromosomal mapping of the gene to 10p within the autosomal recessive megaloblastic anemia (MGA1) region. Blood 91:3593–3600. 57. Moestrup SK, Kozyraki R, Kristiansen M, Kaysen JH, Rasmussen HH, Brault D, Pontillon F, Goda FO, Christensen EI, Hammond TG, Verroust PJ 1998 The intrinsic factor– vitamin B12 receptor and target of teratogenic antibodies is a megalin-binding peripheral membrane protein with homology to developmental proteins. J Biol Chem 273:5235–5242. 58. Xu D, Kozyraki R, Newman TC, Fyfe JC 1999 Genetic evidence of an accessory activity required specifically for cubilin brush-border expression and intrinsic factor–cobalamin absorption. Blood 94:3604–3606. 59. Kristiansen M, Kozyraki R, Jacobsen C, Nexo E, Verroust PJ, Moestrup SK 1999 Molecular dissection of the intrinsic factorvitamin B12 receptor, cubilin, discloses regions important for membrane association and ligand binding. J Biol Chem 274:20540–20544. 60. Burmeister R, Boe IM, Nykjaer A, Jacobsen C, Moestrup SK, Verroust P, Christensen EI, Lund J, Willnow TE 2001 A tworeceptor pathway for catabolism of Clara cell secretory protein in the kidney. J Biol Chem 276:13295–13301. 61. Birn H, Fyfe JC, Jacobsen C, Mounier F, Verroust PJ, Orskov H, Willnow TE, Moestrup SK, Christensen EI 2000 Cubilin is an albumin binding protein important for renal tubular albumin reabsorption. J Clin Invest 105:1353–1361. 62. Sahali D, Mulliez N, Chatelet F, Dupuis R, Ronco P, Verroust P 1988 Characterization of a 280-kD protein restricted to the coated pits of the renal brush border and the epithelial cells of the yolk sac. Teratogenic effect of specific monoclonal antibodies. J Exp Med 167:213–218. 63. Aminoff M, Carter JE, Chadwick RB, Johnson C, Grasbeck R, Abdelaal MA, Broch H, Jenner LB, Verroust PJ, Moestrup SK, de la Chapelle A, Krahe R 1999 Mutations in CUBN, encoding the intrinsic factor–vitamin B12 receptor, cubilin, cause hereditary megaloblastic anaemia 1. Nat Genet 21: 309–313. 64. Imerslund O 1960 Idiopathic chronic megaloblastemic anemia in children. Acta Paediatr 49:208–209. 65. Kozyraki R, Fyfe J, Verroust PJ, Jacobsen C, Dautry-Varsat A, Gburek J, Willnow TE, Christensen EI, Moestrup SK 2001 Megalin-dependent cubilin-mediated endocytosis is a major pathway for the apical uptake of transferrin in polarized epithelia. Proc Natl Acad Sci USA 98:12491–12496. 66. Willnow TE, Nykjaer A 2002 Pathways for kidney-specific uptake of the steroid hormone 25-hydroxyvitamin D3. Curr Opin Lipidol 13:255–260. 67. Bansal A, Gierasch LM 1991 The NPXY internalization signal of the LDL receptor adopts a reverse-turn conformation. Cell 67:1195–1201. 68. Morris SM, Tallquist MD, Rock CO, Cooper JA 2002 Dual roles for the Dab2 adaptor protein in embryonic development and kidney transport. EMBO J 21:1555–1564. 69. Pook MA, Wrong O, Wooding C, Norden AG, Feest TG, Thakker RV 1993 Dent’s disease, a renal Fanconi syndrome

CHAPTER 10 Endocytic Pathways for 25-(OH) Vitamin D3

70.

71.

72.

73.

with nephrocalcinosis and kidney stones, is associated with a microdeletion involving DXS255 and maps to Xp11.22. Hum Mol Genet 2:2129–2134. Lloyd SE, Pearce SH, Fisher SE, Steinmeyer K, Schwappach B, Scheinman SJ, Harding B, Bolino A, Devoto M, Goodyer P, Rigden SP, Wrong O, Jentsch TJ, Craig IW, Thakker RV 1996 A common molecular basis for three inherited kidney stone diseases. Nature 379:445–449. Igarashi T, Gunther W, Sekine T, Inatomi J, Shiraga H, Takahashi S, Suzuki J, Tsuru N, Yanagihara T, Shimazu M, Jentsch TJ, Thakker RV 1998 Functional characterization of renal chloride channel, CLCN5, mutations associated with Dent’s Japan disease. Kidney Int 54:1850–1860. Yamamoto K, Cox JP, Friedrich T, Christie PT, Bald M, Houtman PN, Lapsley MJ, Patzer L, Tsimaratos M, van’t Hoff WG, Yamaoka K, Jentsch TJ, Thakker RV 2000 Characterization of renal chloride channel (CLCN5) mutations in Dent’s disease. J Am Soc Nephrol 11:1460–1468. Gunther W, Luchow A, Cluzeaud F, Vandewalle A, Jentsch TJ 1998 ClC-5, the chloride channel mutated in Dent’s disease, colocalizes with the proton pump in endocytotically active kidney cells. Proc Natl Acad Sci USA 95:8075–8080.

163 74. Piwon N, Gunther W, Schwake M, Bosl MR, Jentsch TJ 2000 ClC-5 Cl−-channel disruption impairs endocytosis in a mouse model for Dent’s disease. Nature 408:369–373. 75. Napoli JL 1996 Biochemical pathways of retinoid transport, metabolism, signal transduction. Clin Immunol Immunopathol 80:S52–62. 76. Ouahchi K, Arita M, Kayden H, Hentati F, Ben Hamida M, Sokol R, Arai H, Inoue K, Mandel JL, Koenig M 1995 Ataxia with isolated vitamin E deficiency is caused by mutations in the alpha-tocopherol transfer protein. Nat Genet 9:141–145. 77. Adams JS, Chen H, Chun RF, Nguyen L, Wu S, Ren SY, Barsony J, Gacad MA 2003 Novel regulators of vitamin D action and metabolism: lessons learned at the Los Angeles zoo. J Cell Biochem 88:308–314. 78. Bikle DD, Gee E, Halloran B, Kowalski MA, Ryzen E, Haddad JG 1986 Assessment of the free fraction of 25-hydroxyvitamin D in serum and its regulation by albumin and the vitamin D–binding protein. J Clin Endocrinol Metab 63:954–959. 79. Bikle DD, Siiteri PK, Ryzen E, Haddad JG 1985 Serum protein binding of 1,25-dihydroxyvitamin D: a reevaluation by direct measurement of free metabolite levels. J Clin Endocrinol Metab 61:969–975.

CHAPTER 11

The Vitamin D Receptor J. WESLEY PIKE AND NIRUPAMA K. SHEVDE Department of Biochemistry, University of Wisconsin–Madison, Madison, Wisconsin

I. II. III. IV.

Introduction Discovery of the Vitamin D Receptor Characterization of the Vitamin D Receptor Structural Gene for the Vitamin D Receptor

I. INTRODUCTION Many significant advances have been made during the past several decades in our understanding of the mechanisms through which steroid hormones function within target cells [1–6]. It is now well known that both the sex and adrenal steroids, thyroid hormone, retinoic acid (RA), and 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) all bind to specific intracellular receptors. These proteins in turn transduce the environmental information carried by the small hormonal signals to the cell nucleus resulting in modifications in gene expression. Virtually all of the genes that encode nuclear receptors have now been cloned, revealing them to be part of a large gene family of proteins with highly related structural similarity. Interestingly, members of this family of nuclear receptor genes also transduce vertebrate signals generated by other forms of retinoic acid such as 9-cis-RA, fatty acids, nutritional metabolites, and invertebrate signaling molecules such as the insect molting hormone ecdysone [6] as well as the classical hormones. Perhaps most importantly, the cloning of the nuclear receptor family of genes has made available critical reagents that have led to important new insights into the structures of the proteins, an understanding of how hormonal signals activate these factors, delineation of the mechanisms by which they interact with the regulatory regions of target genes, and the discovery that numerous additional transcription factor complexes are essential for downstream gene modulation. The idea that vitamin D might function as a steroidlike hormone emerged in 1968 [7] and actually predated the documented discovery of 1,25(OH)2D3, the active hormonal form of vitamin D3 [8–10]. The lipophilic nature of the small vitamin D molecule and its capacity to localize in target tissues, as well as the fact that vitamin D responses were sensitive to transcriptional inhibitors, provided an initial basis for this hypothesis. Additional studies also defined the exquisite and highly complex nature of the regulation VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Functional Analysis of the Vitamin D Receptor VI. The Human Vitamin D–Receptor Chromosomal Gene VII. Concluding Comments References

of 1,25(OH)2D3 synthesis and production by the renal enzyme 25-hydroxyvitamin D3 1α-hydroxylase (see Chapter 5) [11]. Nevertheless, the discovery of a “binding protein” in specific target tissues that appeared to mediate nuclear localization of what eventually was determined to be the active vitamin D hormone, 1,25(OH)2D3 was perhaps the most significant [12]. Indeed, further characterization of this macromolecule over the years, and its eventual cloning in 1987 [13], confirmed the original hypothesis that 1,25(OH)2D3 was a steroid-like hormone. Its proposed mechanism in general terms is simply depicted in Fig. 1. These studies also showed that the actions of 1,25(OH)2D3 were directly linked to the gene regulatory properties of this “binding protein” or vitamin D receptor (VDR) [13–15], as alluded to in the figure. The structure and function of the VDR confirmed 1,25(OH)2D3 as an authentic steroid hormone and ushered in a new era of research on the molecular mechanisms underlying its regulation of gene expression. In this chapter, we provide an historical overview as well as contemporary summary of the VDR’s central role in mediating the actions of the vitamin D hormone. We describe the discovery of the receptor, a summary of its tissue and cellular distribution, and its biochemical characteristics. We also provide an overview of the VDR’s structural organization as well as how it functions to regulate gene expression. Finally, we describe the structural organization for the human VDR chromosomal gene. The reader is referred to the many additional chapters in the Mechanism of Action section of this book for more depth into each of the many aspects related to VDR structure, biochemistry, and biological function.

II. DISCOVERY OF THE VITAMIN D RECEPTOR Evidence for the existence of a binding protein or receptor (i.e., the VDR) was first provided by Haussler and Norman in 1969 [12]. These studies demonstrated Copyright © 2005, Elsevier, Inc. All rights reserved.

168

J. WESLEY PIKE AND NIRUPAMA K. SHEVDE

DB P-1 ,25 (O H)

2D 3

1,

25

(O H D3 )2

Target cell Nucleus R

R

mRNA

Tissue Specific Cellular Response

Mineral Homeostasis Bone Metabolism Vitamin D3 Catabolism Immunomodulation Differentiation Receptor Autoregulation

FIGURE 1

Model for the molecular mechanism of action of the vitamin D hormone. 1,25(OH)2D3 dissociates from serum vitamin D binding protein (DBP) or other carrier proteins, enters the cell, and interacts with the vitamin D receptor (VDR). Activation by the ligand leads to binding of the VDR to promoters within responsive genes and subsequent modulation of gene expression. Biological responses are indicated.

a preferential uptake of active vitamin into the intestinal chromatin fraction of chickens. They also provided further evidence that the process was saturable at low concentrations of administered active vitamin and was likely mediated by a specific protein component. Following these pioneering studies, more definitive evidence for the VDR began to emerge from the studies of Brumbaugh and Haussler [16] and Lawson and Wilson [17], who established the protein nature of the receptor via proteolytic digestion studies and equilibrium sedimentation analysis. Studies by Brumbaugh and Haussler [18] also established that the protein bound 1,25(OH)2D3 with high nanomolar affinity, an important prerequisite for mediating vitamin D action. Nevertheless, it was experiments carried out in vitro in 1975 that provided the most compelling evidence that

the protein might function as a nuclear receptor by demonstrating that cytosol-derived VDR could bind to chromatin fractions in a hormone-sensitive fashion [18]. Collectively, these findings provided the impetus for further exploration into the properties, structure, and function of the VDR.

III. CHARACTERIZATION OF THE VITAMIN D RECEPTOR A. Tissue Distribution of the VDR The VDR was first discovered in extracts prepared from chicken intestine, perhaps as a result of the considerable role the chicken as a model system had played in delineating the function of vitamin D in

169

CHAPTER 11 The Vitamin D Receptor

calcium homeostasis. Despite this, the known actions of vitamin D extended well beyond those in the intestine and were certainly not limited to the avian species. Target tissues of vitamin D included the kidney, bone, and the parathyroid glands and accordingly, the VDR was eventually identified in those tissues in avian as well as mammalian species [19–23]. The discovery of the VDR in mammalian intestine and in most of the listed tissues, nevertheless, required efforts over several years. Kream et al. [24] provided evidence for the existence of a VDR in rat intestinal tissue. Its properties were similar to those of the chicken receptor, although equilibrium sedimentation experiments suggested that rat receptor might exhibit a smaller molecular weight. Importantly, the discovery of receptors in these as well as other tissues was facilitated by tritiumlabeled 1,25(OH)2D3, whose availability with increasing radioisotopic specific activity provided the enhanced sensitivity necessary to detect VDR levels in tissues other than the intestine where it was most abundant. Thus, high specific radioactivity 1,25(OH)2D3 was a major determinant in the discovery and subsequent characterization of the VDR. 1,25(OH)2D3 with very high specific radioactivity also enabled detection of VDR in many unexpected target tissues as well as in cultured cell lines. The search for receptors in these tissue and cell sources was prompted by emerging evidence that the biological effects of 1,25(OH)2D3 extended beyond that of calcium and phosphorus homeostasis to include a role in cellular proliferation and differentiation [25,26]. Accordingly, VDRs were discovered in tissues such as pancreas [20,27], placenta [20], pituitary [28], ovary [29], testis [30], mammary gland [31], and heart [32]. Cultured cell lines that expressed the VDR were similarly extensive and now include cells of fibroblastic, chondrocytic, osteoblastic, myoblastic, hematopoietic, and lymphopoietic origin as well as cells derived from normal kidney, intestine, and skin. Of more fundamental nature, they also include cells of both tumorigenic and nontumorigenic origin [33–36]. A partial listing of vitamin D target tissues and cells is documented in Table I. The biological role of 1,25(OH)2D3 in many of these tissues and cells is considered more extensively in many additional chapters in this book. Perhaps the most interesting biological effect of vitamin D metabolites, however, is their ability to regulate proliferation and differentiation [37]. These effects of 1,25(OH)2D3 are not unlike those of most steroid hormones. Most importantly, however, the actions highlight a potential therapeutic role for 1,25(OH)2D3 and its analogs in cancer treatment, an area considered in more depth in Chapters 84–88 and 93–97, and in the regulation of the immune system [38,39], considered in Chapters 36, 98, and 99.

TABLE I Cellular and Tissue Distribution of the VDR System Gastrointestinal Hepatic Renal Cardiovascular Endocrine Exocrine Reproductive

Immune Respiratory Musculoskeletal

Epidermis/appendage Central nervous system Connective tissue

Tissue Esophagus, stomach, small intestine, large intestine, colon Liver parenchyma cells Kidney, urethra Cardiac muscle Parathyroid gland, thyroid, adrenal, pituitary Parotid gland, sebaceous gland Testis, ovary, placenta, uterus, endometrium, yolk sac, avian chorioallantoic membrane, avian shell gland Thymus, bone marrow, B cells, T cells Lung alveolar cells Bone osteoblasts and osteocytes, cartilage chondrocytes, striated muscle Skin, breast, hair follicles Brain neurons Fibroblasts, stroma

Antiproliferative effects of vitamin D are currently being exploited in the treatment of human psoriasis, a hyperproliferative disorder of the skin [40] that is another important target of vitamin D action [37,41] outlined in more detail in Chapter 101. Indeed, the clear presence of the VDR in most tissues in avian and mammalian species indicates that the biological effects of vitamin D are likely to be very widespread. While the analysis of tissue extracts led to the discovery of many VDR-containing sites, it was the use of scintillant-enhanced autoradiography employed early on by Zile et al. [42] and Jones and Haussler [43] and later by Stumpf et al. [44,45] that led to a thorough appreciation of the wide cellular distribution of vitamin D target tissues. These studies revealed such targets as the duodenum, ileum, and jejunum of the rat intestine, stomach, kidney, skin, and pituitary as well as several neurons in the brain. While the presence of 1,25(OH)2D3 (or a metabolite of the administered compound) in these tissues through autoradiographic means does not provide direct evidence of the presence of VDR, the bulk of these targets were confirmed when an antibody to the VDR was developed some years later enabling the use of immunocytochemistry [46–49]. Most recently, studies using Northern blot [13], in situ hybridization [50], or RT-PCR [51] analyses have been employed to detect the VDR mRNA in cells and tissues.

170 The availability of probes that enabled these analyses closely followed the cloning of the VDR in 1987 and 1988 (to be discussed in Section IV). Although this RNA detection approach has been useful, it often ignores the important underlying assumption that VDR mRNA transcripts correlate directly with VDR protein. The use of these nucleic techniques has been helpful, however, in the study of the transcriptional regulation of VDR production.

B. Subcellular Distribution of the VDR Although the site of action of ligand-bound VDR in the nucleus is clear, the subcellular location of unliganded VDR has remained controversial over several decades. Early studies suggested that the unliganded VDR was cytosolic (see Fig. l and Brumbaugh and Haussler [18]). Studies in the early 1980s, however, suggested that the VDR might be loosely associated with the nuclear fraction [52,53]. Indeed, the application of immunocytochemistry supports this largely nuclear localization of the apo form of the VDR, an intracellular residence site consistent with that of most other nuclear receptors. It is important to note, however, that the presence of the unliganded VDR within the nucleus does not imply that the VDR must be bound to regulatory sites of action on 1,25(OH)2D3 responsive genes. To the contrary, recent chromatin immunoprecipitation assays to be discussed later suggest that the VDR is not bound to these regulatory sites until activated by 1,25(OH)2D3 [54,55], supporting the need to identify these alternative localization sites for VDR in the nucleus in the absence of hormone. Perhaps not surprisingly, the controversy regarding the intracellular site of residency of unliganded VDR has reemerged recently. In studies using fluorescent probes and discussed in greater detail in Chapter 22, Barsony and colleagues [57,58] provide evidence that recombinantly derived VDR distributes into both the cytoplasmic and nuclear compartments in the absence of ligand and is imported into the nucleus via several essential components, including specific nuclear import signals within the VDR molecule, interaction with RXR, and interaction with nuclear import proteins lining the nuclear pores. Moreover, the VDR actively cycles between the two compartments. With the qualification that the VDR is expressed in relatively high concentrations in the cell as a chimeric protein fused to green fluorescent protein (GFP), the data seem to support strongly active receptor movement between both cellular compartments. Given these caveats, however, it is likely that the controversy will continue.

J. WESLEY PIKE AND NIRUPAMA K. SHEVDE

C. Biochemical Properties and Organization of the VDR The VDR is expressed in low concentrations in target tissues and cultured cells, a property consistent with its role as a potent transcription regulatory molecule [58,59]. Estimates of receptor abundance range from under 200 to over 25,000 copies of VDR/cell (10 to 100 fmol/mg protein) depending on the cell type or cell line examined, and up to 1 pmol/mg protein in certain tissue extracts. Differences in VDR abundance suggest that cells with higher VDR content may be more highly responsive to 1,25(OH)2D3 than those with lower levels. Little evidence supports this hypothesis, however, and in fact recent data suggest that the expression levels of many of the comodulatory complexes discussed in Chapters 14, 16, and 17 may be more significant determinants of cellular response than the receptor concentration itself [60]. Other factors may also contribute to the level of response to 1,25(OH)2D3, including cellular capacity to internalize and metabolize 1,25(OH)2D3, differential activation events that involve the VDR directly (phosphorylation or other posttranslational modifications), and finally the accessibility and individual activation properties of specific gene targets. These, as well as additional events, are likely to contribute significantly to the biological responsiveness and cellular sensitivity to 1,25(OH)2D3. Numerous physical and functional properties of the VDR were defined following the protein’s discovery in 1973, several of which are documented in Table II. Sedimentation analysis revealed a protein of 3 to 3.7 S that exhibited an elongated shape. Gel filtration estimates of protein size ranged from 50,000 to 70,000 Da

TABLE II

Biochemical Properties of the VDR and Its RNA Transcript

VDR protein Molecular weight: 50 kDa (human) to 60 kDa (chicken) Sedimentation coefficient (S): 3.2 (human, rat) to 3.7 (chicken) Equilibrium dissociation constant: 1–2 × 10–10 M Isoelectric point: 6.2 Phosphoprotein: Serine phosphorylated at residue 208 (human) in response to 1,25(OH)2D3 Lability: Proteolytically sensitive Tissue abundance: 10 to 1000 fmol of VDR/mg protein hVDR mRNA 4.8 kb (human), 3.0 (chicken)

171

CHAPTER 11 The Vitamin D Receptor

depending on species. The most important characteristic of the VDR, however, was its capacity to bind 1,25(OH)2D3 with both high affinity and selectivity [61–65]. Accordingly, the VDR displayed an equilibrium dissociation constant (KD) of approximately 10−10 M for the natural ligand 1,25(OH)2D3 and bound this ligand’s precursors as well as less active metabolites with substantially lower affinities [63,64]. The contribution of both the 25-hydroxyl and the 1α-hydroxyl groups on the 1,25(OH)2D3 molecule to specific, high-affinity binding to VDR has been studied extensively [64] and will be discussed later. An important discovery in the late 1970s was the observation that the VDR could bind to DNA, an anticipated property of a protein that activates transcription [58]. While a more thorough understanding of VDR DNA binding emerged following the identification of specific DNA-binding sites (vitamin D response elements or VDREs) within gene promoters, as discussed later in this chapter and other chapters, the finding that the VDR bound to nonspecific DNA set the stage for the ensuing studies aimed at understanding the structural organization and function of the VDR. Perhaps the most important observation made of VDR DNA binding properties was the finding that while VDR exhibited an obvious “affinity” for DNA in the absence of 1,25(OH)2D3, a feature unlike that of most steroid receptors, its “affinity” for DNA quantitatively increased following complex formation with 1,25(OH)2D3 [65,66]. The latter property implied that the structure of the VDR or perhaps the composition of the active receptor was transformed in the presence of the hormonal ligand. Recent X-ray diffraction studies to be described later in this chapter as well as in Chapter 15 define the nature of the effects of ligand on the VDR and its DNA binding capabilities at the atomic level [67–69]. Purification of the VDR initially from chicken intestine [14,58] and later from porcine intestine [15] led to the development of monoclonal antibodies to the VDR. These reagents proved essential in further immunological characterization of the VDR, provided alternative receptor isolation methods, and ultimately offered the means whereby the VDR gene was cloned. Western blotting analysis revealed for the first time that the precise molecular mass of VDR varied from species to species, ranging from 60,000 Da in the chicken [70] to approximately 50,000 Da in humans [71]. These relative sizes were confirmed through molecular cloning. Immunological analyses of this type also revealed that the VDR comprised A and B isoforms in the chicken but a single form in mammalian-derived tissues [70]. The two forms of the chicken VDR are

now known to arise from the existence of alternative start sites in the mRNA, although the functional consequence of two such proteins with differing aminoterminal extensions remains unknown [72,73]. Finally, immunoprecipitation of the VDR from in vitro translated tissue mRNA revealed that whereas the chicken receptor protein was larger than that from mammalian cells, the mRNA transcripts were considerably smaller [71]. The differential size of these transcripts has been confirmed through the use of hybridization techniques incorporating the cDNA for the VDR as a probe. Initial insights into the structural organization of the VDR emerged using techniques such as limited proteolytic digestion together with radiolabeled 1,25(OH)2D3 and immunologic probes. Initial studies by Pike [74] suggested that the VDR epitope for a specific monoclonal antibody to the VDR lay immediately adjacent to the domain necessary for DNA binding. Additional studies by Allegretto and co-workers [75,76] employing limited digestion with trypsin led to the realization that to the l,25(OH)2D3-bound VDR could be cleaved into two domains, one a large fragment of 30 to 40 kDa that retained prebound 1,25(OH)2D3 and a second of 16 to 20 kDa that could bind DNA independently. Most importantly, the findings suggested that the two functional domains (ligand-binding and DNA-binding domains) were separated by a proteolytically sensitive “hinge.” The inability following trypsin digestion to identify the domain responsible for ligand binding of hormone-free receptor argued strongly for a ligandinduced change in receptor conformation. This finding substantiated earlier observations of increased VDR stability in tissue extracts following complex formation with ligand. Additional preliminary evidence indicated that the DNA binding domain was located amino terminal to the hormone binding domain [77]. As described in the next section, these rudimentary insights into the overall structural organization of the VDR were proven correct via the subsequent cloning of the receptor’s structural gene.

IV. STRUCTURAL GENE FOR THE VITAMIN D RECEPTOR A. Cloning of the VDR The biochemical purification of the steroid receptors in the early 1980s led to the development of immunological probes that facilitated not only further characterization of the proteins, but the cloning of their structural genes as well. As a result, the receptors for virtually all the known steroid, thyroid, and

172 vitamin hormones were cloned during the latter half of the 1980s [2,3]. Interestingly, sequence analysis of the first cloned receptor cDNAs revealed regions that exhibited significant sequence similarity with the DNA binding domain of an unrelated transcription factor TFIIIA. In addition, sequence similarities within this specific region among the several nuclear receptors suggested the possibility that related transcripts might be recovered through low-stringency hybridization screening [3]. Accordingly, this led to the recovery of numerous additional cDNAs for unknown nuclear receptors or orphan receptors. For many of these, cognate ligands, if they exist, remain currently unidentified. However, investigation of several of these unknown receptors led to the eventual discovery of novel ligands and the identification of new hormonal systems. Examples include the discovery of retinoid X receptor (RXR) and its ligand 9-cis RA [78], peroxisome proliferator-activated receptor gamma (PPARγ) and one of its ligands prostaglandin J2 [79], farnesol–X receptor (FXR) and its ligand farnesol [80], the liver X receptor (LXR) and its oxysterol ligands [81,82], and several others. Interestingly, the sequence of the DNA binding domain of the glucocorticoid receptor (GR) was first noted to exhibit a high degree of similarity with the viral oncogene v-erbA [83]. The cloning of c-erbA led to the discovery that this gene encoded the first of two thyroid hormone receptor (TR) genes [84]. The development of the anti-VDR monoclonal antibody 9A7 enabled McDonnell et al. [13] to screen randomly primed chicken intestinal cDNA expression libraries and eventually to recover a single cDNA clone that encoded a zinc finger containing DNA binding domain. This domain exhibited high homology to the DNA binding domain of GR. The presence of this domain and its reactivity to the 9A7 antibody (known to interact in or near the DNA binding domain of the VDR [65]) led to initial confidence that this and several other isolated cDNAs represented a portion of the transcript encoding the chicken VDR. Subsequent hybridization-selected in vitro translation techniques using these clones substantiated the authenticity of the cDNA clones [13]. The recovery of this cDNA allowed the subsequent rapid recovery of full-length VDR cDNA transcripts from human [85] tissue sources. The rat intestinal VDR was cloned shortly thereafter by Burmester et al. [86]. Recovery of a cDNA transcript from the human leukemia HL-60 cell line [87] revealed virtual identity to that of the original human VDR cDNA cloned by Baker et al. [85]. Sequences of the VDR have been reported from mouse [88], Japanese quail [89], and Xenopus [90] and have appeared in both the human and mouse genomic data bases [91,92]. The cloning of the VDRs from various

J. WESLEY PIKE AND NIRUPAMA K. SHEVDE

sources and species was a milestone in the vitamin D field. Perhaps as important, it provided the first direct evidence of a structural relationship between the VDR and other bona fide members of the steroid receptor family of genes [93]. The VDR is highly conserved across tissue sources and species. The overall sequence similarity of rat, mouse, and avian receptors to that of human VDR is 79, 86, and 66%, respectively. This homology rises to above 95% within specific domains such as the DNA binding domain (see Chapter 13). There is, nevertheless, considerable variability in the number of residues amino terminal to the DNA binding domain. This region varies from 21 amino acids in the human VDR (the smallest of the VDRs) to approximately 57 amino acids in the chicken protein [85,89]. In addition, the chicken gene exhibits one alternative start site (the first at nucleotide 47 and the second at nucleotide 67, which corresponds to the most 5′ ATG in the human VDR transcript identified by Baker et al. [85]) that accounts for the two forms of the VDR found in the avian species [72,73]. The human VDR cDNA sequence also presents as a transcript with two potential start sites, one lying only three codons downstream of the first [85]. Whether both initiation codons are used in this transcript is unclear. However, an important polymorphism in the VDR gene has been demonstrated in the human population [94,95]. This polymorphism is a single-base-pair transition that occurs in the first initiation site reported by Baker et al. [85] (ATG to ACG). The result of this polymorphism in humans is the production of a smaller VDR of 424 amino acids that occurs in homozygous form in approximately 37% of the human population and in heterozygous form (one allele containing both initiation sites and the second containing only the downstream site) in approximately 48% of the population [96] (Chapter 72). Whether the two proteins exhibit different biochemical properties and/or different functions remains an open question, although it does appear that the smaller form may be more active [96–98]. This has been suggested to occur via a tighter contact with the transcription factor TFIIB [96]. Association studies indicate that this polymorphism may be linked to bone mineral density [97], although these data are not particularly robust [99–101] (Chapter 68).

B. VDR Is a Member of the Intracellular Receptor Gene Family The cloning of glucocorticoid [83] and estrogen [102] receptors in 1985 and 1986 were the first of a series of successful efforts to clone the intracellular receptor genes. More than 150 cloned members of this

173

CHAPTER 11 The Vitamin D Receptor

intracellular receptor gene family [5] have been identified, making it the largest transcription factor family currently known. It is in fact noteworthy that the cloning of this family of genes was so successful that publication of the sequence of both the mouse and human genomes in 2002 did not reveal a single uncloned and unidentified functional nuclear receptor gene [91,92]. The evolutionary success of this family of transcription factors is highlighted both by its abundance and by its involvement in an incredibly wide range of biological processes that include growth, development, differentiation, and adult homeostasis. The reader is referred to other reviews on the steroid receptor family for details [1–6] as well as summaries in Chapters 13–16.

A

C. Structural Domains of the VDR 1. OVERVIEW

The cloning of the estrogen receptor (ER) in 1986 and the GR shortly thereafter revealed that these proteins were made up of distinct regions or domains, leading to their designation as domains A, B, C, D, E, and F [102]. As illustrated in Fig. 2A, the highly conserved DNA binding domain is designated the C domain and provides a central focus for the remainder of the molecules. Accordingly, the nonconserved A/B domain comprises all residues amino terminal to the DNA binding domain and the D domain or hinge region comprises residues lying downstream of C. The carboxyterminal region that contains the ligand binding

NR A/B

B

C 1

24

D 89

E/F 232

427

41 6 42 3

1

9

34

31

27

23

2

C

2

hVDR

427

44 8 45 5

7

5

37

35

0 31

27

0

E/F Domain VDR

456

40 8 41 5

8

6

33

31

1 27

23

4

TRβ

462

90 7 91 4

0

8

83

80

1 76

72

1

RARα

934

ER

53 8 54 5

45 7

43 5

38 9

34 9

PR

595

FIGURE 2 Functional domain structure of the nuclear receptor superfamily. (A) The nuclear receptors (NR) are separated into four regions designated A/B, C, D, and E/F. (B) Residue boundaries of corresponding regions within the human VDR. (C) Three regions of sequence similarity and residue boundaries within the E/F domain of several nuclear receptors. Shown in addition to VDR are thyroid receptor β, (TRβ), retinoid receptor α (RARα), progesterone receptor (PR), and estrogen receptor (ER). Functions associated with the domains include transactivation (A/B), DNA binding (C), flexible hinge (D), and dimerization, ligand binding, transactivation, and repression (E/F) (see text for details).

174 domain is termed the E or E/F domain. Three stretches of sequence similarity exist in the E domain for virtually all of the nuclear receptor gene family. The F domain exhibits extensive variability and is not conserved, much like that of the A/B domain. The domain structure of mammalian VDRs is depicted in Fig. 2B. As can be seen, the A/B domain is very short relative to other members of the nuclear receptor family, representing one of the most abbreviated of the entire gene family. The C region comprises the DNA binding domain of the VDR [103]. The D domain is a highly flexible region within the VDR and links both its DNA-binding and its hormone-binding components [103]. It is not conserved among VDRs from other species or among other members of the steroid receptor family [85,86]. This region is, however, essential for intact hormone binding, a feature unlike that of most other steroid receptors. The hinge region may be more complex than that found in other receptors in that it contains an additional segment that corresponds to a specific exon in the chromosomal gene [104]. Interestingly, however, this region can be removed without compromising 1,25(OH)2D3 binding or transcriptional function [67,105]. Finally, the E/F domain contains the 1,25(OH)2D3 binding region of the VDR [103] and, equally important, serves as a highly complex protein–protein interface for a series of additional proteins that are integral to receptor activity [106–108]. This region has been termed activation function 2 or AF-2. The ability of the VDR to bind both DNA and additional transcription factors highlights its central role in nucleating the formation of transcriptionally active complexes at specific sites of gene regulation. Since the general role of all members of the receptor gene family is similar, it is not surprising that the E region is moderately conserved across the entire gene family and contains several subdomains that are even more highly conserved, as depicted in Fig. 2C. The F domain is absent in the VDR. 2. DNA BINDING DOMAIN

The DNA binding domain of the VDR is encoded in domain C. This domain consists of two similar modules, each consisting of a zinc-coordinated finger structure. The zinc atoms are individually coordinated in a tetrahedral fashion through four highly conserved cysteine residues that serve to stabilize the finger structure itself. Importantly, the finger modules are structurally unrelated to the zinc fingers found in TFIIIA wherein the zinc atom is coordinated through two cysteines and two histidines [109,110]. Surprisingly, although the two zinc modules of the VDR appear highly related to each other structurally, they are not equivalent topologically because the chirality of the residues that

J. WESLEY PIKE AND NIRUPAMA K. SHEVDE

coordinate the zinc atoms in each module is different [110]. The function of each of these modules in DNA binding is also different. Whereas the amino-terminal module directs specific DNA binding in the major grove of the DNA binding site, the carboxy-terminal module serves as a dimerization interface for interaction with partner proteins [111,112]. It seems likely that the two exons that encode these modules (see Section VI) evolved separately, although it is possible that each derived from a common ancestral gene through duplication and then diverged later. The threedimensional structure of the DNA-binding domain of several of the receptors has been determined through both nuclear magnetic resonance (NMR) spectroscopy and X-ray crystallography. The salient features of the finger structures are the presence of carboxy-terminal α-helices essential to interaction with DNA. Thus, our understanding of the structural organization of these modules as well as the mechanisms by which they function to interact with DNA and other proteins is well advanced [113–118]. 3. LIGAND BINDING DOMAIN

The E region of the VDR represents a multifunctional domain that exerts absolute regulatory control over the DNA-binding as well as transcription-modifying properties of the VDR. Historically, this region and comparable regions found in other receptors were explored for common structural features as well as for individual functional properties. As indicated in the overview in this section, several regions of homology were identified. Since cognate ligands differ substantially, it appeared likely that these regions of similarity represented motifs integral to the receptors’ role in transactivation. Closer scrutiny revealed the presence of multiple heptad repeats that were located in the ligand binding domain (LBD) of most members of the gene family including that of the VDR [119]. Despite this and many other important biochemical and molecular biologic discoveries, it is clear that the greatest insight into LBD structure emerged as a result of X-ray crystallographic studies initiated by Moras and colleagues with the retinoid receptors RARγ and RXRα [120,121]. As of this date, the three-dimensional structures of the LBDs of most of the known ligand-binding members of the nuclear receptor family including VDR have been determined [67–69,122–126]. With the exception of RXRα, the structures of the LBDs from most of the receptors have been solved in the presence of their cognate ligands. Thus, the bulk of our information of LBD structure in the absence of hormone derives from the apo structure of RXRα. RXRs play a unique role as unliganded heterodimer partners for many members of the nuclear receptor family [5,6].

175

CHAPTER 11 The Vitamin D Receptor

three-dimensional structures of numerous members of the steroid receptor superfamily [67–69,121,122–126,128]. As with many of the steroid receptors, the threedimensional structure of the LBD of the VDR has also been determined [67–69]. As this is the subject of Chapter 15, only a brief summary of the structure of the VDR will be described here. Rochel et al. [67] first crystallized the human VDR LBD in complex with 1,25(OH)2D3 as well as other ligands [68]. A recent report by Vanhooke and colleagues has confirmed and extended this important work using the rat VDR [69]. Essential to the successful crystallography effort of both groups was the removal of a portion of the D region that prevented crystal formation. The three-dimensional structure of the LBD of the rat VDR is depicted in Fig. 3. The human and rat VDR structures revealed the presence of thirteen α-helices and three β-sheets that together form at its core a highly hydrophobic ligand binding pocket. Surprisingly, the overall volume of this pocket is much greater than that seen for many of the other steroid receptors. 1,25(OH)2D3 binds in this pocket in an extended configuration with the A ring in the β chair conformation and the 1α-OH group equatorial. The ligand is anchored by six hydrogen-bonding interactions, the 1α-OH group with Ser233, Arg270, the 3β-OH group with Tyr143 and Ser274, and the 25-OH group with His301 and His393 (using the rat VDR

Accordingly, it seems unlikely that the LBD of this receptor will faithfully mimic the other apo-receptor forms. Nevertheless, the structure has revealed the presence of 12 α-helices (H1–H12) arranged in an anti-parallel α-helical sandwich. Perhaps most interesting is the location of H12, which projects away from the body of the RXRα LBD [120]. As will be discussed hereafter and in other chapters in more detail, this projection is in contrast to the location of H12 in virtually all the liganded or holo-receptor forms wherein H12 is packed tightly against the body of the LBD [67–69,121,122–126]. This presumptive change in nuclear receptor structure in the AF-2 region in response to hormone, together with many other changes evident in the overall conformation of the receptor, appear integral to its ability to attract DNA binding partners and to recruit protein complexes essential for transcriptional activation. It is important to note here, however, that the idea that ligand induced significant changes in receptor structure emerged many years earlier from both receptor stability and proteolytic digestion studies [75,76, and references therein]. The latter analyses have been refined through the use of metabolically labeled nuclear receptors derived from in vitro translated receptor cDNAs [127] and are described in detail in Chapters 18 and 83. The reader is referred for details to the many primary publications that report the

A

DNA Binding

B

AF2

N

C 1

21

88

240

Hinge Region

423 Ligand Binding AF2

Deletion

N

C 116

165 211 240

Hinge Region

DRIP205

423 Ligand Binding

6H

N-KNHPMLMNLL KDH-C 625

637

NR Box II

FIGURE 3 Structural organization of the rat VDR in complex with l,25(OH)2D3. (A) Schematic organization of the specific functional domains of the wild-type rat VDR (upper panel) and a ligand binding domain/6x-histidine fusion protein (middle panel) used to prepare rat VDR LBD crystals as described by Vanhooke et al. [69]. Shown in the lower panel is the DRIP205 NR box II peptide that was cocrystallized with the rat VDR LBD. (B) Ribbon representation of the rat VDR LBD in complex with l,25(OH)2D3 and the DRIP205 coactivator peptide.

176 sequence numbering). Several of the LBD helices, including H9 and H10, provide a binding surface for the VDR’s DNA binding partner RXR (see later discussion). Importantly, H12 is packed against the body of the LBD such that it seals the ligand binding site and along with H3 and H4 creates the AF-2 hydrophobic cleft that serves as a docking site for transcriptional comodulators of VDR action (to be discussed later as well as in detail in Chapters 13–18). Several of the residues that form H12 make direct contact with the ligand. Importantly, the recent elucidation of the wildtype VDR from the zebrafish has confirmed that the removal of the “insertion region” located in the hinge region is without consequence to the overall structure of the receptor [129]. 4. ACTIVATION FUNCTIONS

As indicated earlier, hormone binding leads to the creation of a functional AF-2. The core of this functional domain appears to be residues associated with H12 such as human residues 416 to 423 [130,131], although early mutagenesis studies indicated more extensive involvement of components located upstream in the E/F region [7]. Indeed, H3 and H4 are also involved. The hydrophobic cleft formed by these three helices provides an essential binding site for single proteins that mediate linkage to more complex sets of factors that include the p160 histone acetyltransferase (HAT) coactivator complexes [132] and the D-receptor interacting protein (DRIP) complex (Mediator-D complex) [133], both of which are discussed hereafter and in detail in Chapters 14 and 16. Proteins such as the p160 coactivators SRC-1, SRC-2, and SRC-3, and DRIP205 interact directly within the ligand-induced H12 cleft via a short helical LxxLL motif that is common to these comodulators [134]. Indeed, studies have confirmed that LxxLL-containing peptides that bind to either the VDR or RXR function as dominant negative regulators of VDR action [135]. Vanhooke et al. [69] has provided three-dimensional structural insight into the interaction between the rat VDR and an LxxLL motif through cocrystallization of the rat VDR LBD with an LxxLL peptide derived from the second coactivator motif of DRIP205. This group observed that the LxxLL peptide folds into a short α-helix that binds to the surface groove formed by H3, H4, H5, and H12. The LxxLL interaction buries some 500 square angstroms of VDR surface area and involves direct contact with residues of VDR H3, H4, and H12 including Pro412, Leu413, and Glu416 of H12. These interactions form a “charge clamp” that has been described previously for both PPARγ and its coactivator SRC-1 [125]. The side chains of many of the LxxLL motif are buried within the pocket and surrounded by hydrophobic residues.

J. WESLEY PIKE AND NIRUPAMA K. SHEVDE

Nonspecific amino acids project directly into the solvent and away from the binding cavity. These findings support a strong structural basis for the interaction of the VDR with at least one coactivator and may lead to new insights with regard to VDR–coactivator interaction and stability. 5. LIGAND BINDING AND DIFFERENTIAL ACTIVATION

Since 1,25(OH)2D3 acts as a small molecular switch to activate the VDR, it should not be surprising that the binding of ligand would induce sweeping conformational changes in the VDR LBD. Changes in conformation are also likely to be sensitive to the overall structure of the ligand, supporting the hypothesis that synthetic analogs of natural hormonal ligands may be capable of inducing unique changes in receptor, thereby altering their biological function. Indeed, synthetic ligands of the ER as well as other nuclear receptors have been identified that produce both unique receptor conformations as identified by their crystal structures [123,136,137] and that manifest surprising tissue selectivity [138,139]. Numerous vitamin D analogs have also been created. Interestingly, although these compounds manifest unusual actions both in vivo and in vitro (see Chapters 82–88), none have demonstrated their ability to induce a conformation within the VDR that is different from that induced by 1,25(OH)2D3 via crystallography [68,69]. Thus, the molecular basis for their potentially useful therapeutic actions remains unclear.

V. FUNCTIONAL ANALYSIS OF THE VITAMIN D RECEPTOR A. Osteocalcin Regulation as a Model for VDR-Regulated Transcription Osteocalcin (OC) is a relatively abundant noncollagenous protein found in bone. Genetic ablation of the OC gene in mice suggests that this osteoblast-specific protein may contribute to the density and structural integrity of bone [140]. Despite the uncertainty surrounding the function of OC, its expression at the transcriptional level is regulated by a number of cytokines, growth factors, and systemic hormones, one of which is 1,25(OH)2D3 [141,142]. This fact together with the cloning of the human OC gene and its promoter in 1986 [143] provided the raw materials for investigation of 1,25(OH)2D3 action on gene expression at the molecular level. Initial efforts focused on determining if 1,25(OH)2D3 could stimulate the activity of the human OC gene promoter linked to a reporter gene when introduced

177

CHAPTER 11 The Vitamin D Receptor

into bone cells via transfection [103,144,145]. The results suggested that 1,25(OH)2D3 strongly induced this gene promoter, indicating that the effects were direct [146]. Kerner et al. [144] localized the cis-acting element within the promoter subsequently to a region approximately 500 bp upstream of the transcriptional start site. This study and an additional one by Ozono et al. [147] precisely mapped the first vitamin D response element (VDRE) to a directly repeated hexanucleotide sequence separated by three base pairs. Subsequent studies by others using the rat OC gene promoter resulted in a similar identification [148–150]. These studies provided the first insights into DNA sequences within genes that mediated vitamin D-inducible activity. Many genes have been explored since that time for their sensitivity to 1,25(OH)2D3, including genes for mouse osteopontin (OPN) [151], mouse calbindinD28K [152], rat calbindin D9K [153], rat [154,155] and human [156] 25-hydroxyvitamin D3 24-hydroxylase (two VDREs), human p21 [157], and a host of others. A partial list of the sequences that have been shown to mediate vitamin D action together with their locations within the promoters is documented in Table III and elsewhere in this book. These sequences reveal that a “typical” VDRE comprises two hexanucleotide repeats separated by a 3-bp spacer. Although the sequence of the “spacer” is not generally conserved, recent studies have suggested that these base pairs may influence VDR binding [158]. Nevertheless, the general half-site consensus sequence is AGGTCA or GGTTCA, although considerable variability is apparent. Chapter 18 describes many of these VDREs in more detail. It also identifies additional putative VDREs in gene promoters that are an “atypical” configuration; however, the validity of these elements in the context of their natural promoters will require additional investigation.

TABLE III Gene Rat osteocalcin Human osteocalcin Mouse osteopontin Rat calbindin-D9K Mouse calbindin-D28K Rat 24 hydroxylase Human 24 hydroxylase Rat pit 1 Human p21

The efforts to define VDREs were part of a larger effort by numerous investigators to learn about steroid receptor response elements in general, with regard to specificity as well as to receptor selectivity [159]. These overall efforts led to our current understanding that nuclear receptor binding sites can be categorized into three groups: palindromic half-sites that interact with receptors for the sex steroids; directly repeated half-sites with variable spacing that interact with receptors for 1,25(OH)2D3, retinoic acid, thyroid hormone, and a number of other ligands; and single half-sites that mediate the actions of receptors such as NGF-IB that act as monomers [6]. The reader is referred elsewhere [1–6] for a complete review of the nature of these DNA binding sites. The binding of the VDR to endogenous gene promoters in living cells has been examined by several investigators including Sierra et al. [160], Zhang et al. [161]; and Pike and colleagues [54,55] using chromatin immunoprecipitation techniques. In these assays, cultured cells are treated with 1,25(OH)2D3 for specific periods of time, and the cells are then harvested and treated briefly with formaldehyde to fix the proteins that are bound to DNA. Cellular chromatin is then isolated, sonicated to produce small random fragments of 400–500 bp, and precipitated with antibodies to the protein of interest. The coprecipitated DNA is then isolated and its abundance evaluated by PCR using primers designed to amplify specific regions of gene promoters that are of interest. The concentration of promoter DNA correlates directly with immunoprecipitated protein. Importantly, 1,25(OH)2D3 stimulated the appearance of the VDR on several genes promoters including those for OC [160] as well as Cyp24 and OPN [55]. The results of a similar experiment for Cyp24 and OPN promoters are seen in Fig. 4. Although this

Location and Sequence of Positive Natural Vitamin D Response Elements Location −460/−446 −499/−485 −757/−743 −489/−475 −198/−183 −150/−136 (Proximal) −258/−244 (Distal) −169−/155 (Proximal) −291/−277 (Distal) −67/−52 −779/−765

Nucleotide sequence GGGTGA atg AGGACA GGGTGA acg GGGGCA GGTTCA cga GGTTCA GGGTGT cgg AAGCCC GGGGGA tgtg AGGAGA AGGTGA gtg AGGGCG GGTTCA gcg GGTGCG AGGTGA gcg AGGGCG AGTTCA ccg GGTGTG AGTTCA gcga AGTTCA AGGGAG att GGTTCA

178

J. WESLEY PIKE AND NIRUPAMA K. SHEVDE

A

Mouse CYP24 promoter

VDRE2

VDRE1

−236

ORF control

−51

+1

+11055

+11252

Mouse osteopontin promoter

VDRE

−854

ORF control +1

−658

B

+5181

CYP24

1,25 (OH)2D3

−

+

+5380

Osteopontin −

+

−

+

−

+

αVDR αRXR -Ab INPUT promoter

ORF

promoter

ORF

FIGURE 4 1,25(OH)2D3-induced association of the VDR and RXR to the promoters for Cyp24 and OPN. (A) Schematic diagram of the murine Cyp24 and OPN promoters. The locations of the VDREs within the two promoters are indicated with +1 representing the start site of transcription. Primer sites utilized to amplify either the VDRE in the promoter region or the open reading frame (ORF) are indicated. (B) VDR and RXR binding to the OPN or Cyp24 promoters as assessed by chromatin immunoprecipitation. For details, see [54].

approach does not permit the exact identification of the binding site for the VDR on a target gene, the results support localization of the VDR to a region within the promoter that contains a previously identified VDRE. This type of study is consistent with virtually all previous in vitro work that has led to the identification of VDREs in the OC, Cyp24, and OPN gene promoters.

B. VDR Binding to Specific DNA Involves RXR High-affinity binding of the VDR to DNA in vitro requires two regions, the DNA binding domain and the LBD. This conclusion is drawn from numerous mutagenesis experiments that revealed that amino acid alterations either in the zinc finger modules [104,162–168] or in the VDR LBD [103,107,130,169] were capable of blocking DNA binding. Interestingly, the molecular basis for altered DNA binding in the two regions is different. In the first case, changes in the VDR DNA binding domain prevent interaction directly with DNA

(see Chapter 13 for details of specific DNA binding). In the second, however, changes in LBD residues block the ability of the receptor to form the protein dimers that are essential for the high-affinity VDRE binding [107,169]. The discovery that the VDR must form a dimer to bind DNA was suggested by the structural nature of VDREs, which were composed of two repeated halfsites. Surprisingly, however, it was found that the VDR bound to DNA not as a homodimer but rather as a heterodimer as illustrated in the cartoon in Fig. 5. Liao et al. [170] and Sone et al. [162,171,172] both observed that while VDR derived from mammalian cell extracts was fully capable of binding to DNA in vitro, receptor produced either by in vitro transcription/translation or through recombinant means in nonmammalian cells failed to produce DNA binding-competent VDR. The inactivity of the VDR could be rescued, however, through the addition of general mammalian cell extract, suggesting the presence of a VDR DNA-binding facilitator. This factor(s) was discovered in a variety of tissue

179

CHAPTER 11 The Vitamin D Receptor

CBP/p300 PCAF

DRIP

SRC1

HAT 1,25(OH)2D3

RXR

Chromatin

Basal machinery

VDR

VDRE DR3

FIGURE 5

Contemporary model for VDR- and RXR-mediated gene regulation. The VDR and RXR bind to the promoter of a target gene as a heterodimer in response to 1,25(OH)2D3. Transcriptionally active complexes that include the SRC/CBP/p300 HAT coactivator complex or the DRIP205/Mediator complex or perhaps others are subsequently recruited to facilitate changes in chromatin architecture and to enhance RNA polymerase II entry. Both may be needed to modulate gene expression.

and cell sources [172] and termed nuclear accessory factor (NAF); its presence was subsequently confirmed by others [173,174]. NAFs were also discovered for other nuclear receptors including those for thyroid hormone and retinoic acid. In 1991 and 1992, Yu et al. [175], Leid et al. [176], Zhang et al. [177], and Kliewer et al. [178] all discovered that RXR, a previously cloned member of the nuclear receptor family, was the likely NAF protein partner. Three isotypes of RXR (α, β, and γ) are known and may be present in a given cell type. Considerable effort has gone into defining the nature and function of RXR-mediated DNA binding, and the reader is referred to the many reviews that deal with this complex issue [2,5]. Nevertheless, it is now clear that both VDR and RXR subunits are necessary not only to enable high-affinity DNA binding, but also to contribute to the activation process (see Chapter 13). Jin et al. [107,169] utilized an extensive series of internal deletions of the VDR to define two regions essential for interaction with RXR. These regions coincide with the two moderately conserved subregions of homology located in the E domain of the VDR and are associated directly with H9 and H10 of the crystal structure of the RXR LBD [120]. Although there has been some disagreement with regard to domain requirements for dimerization of other receptors [179,180], few have been cocrystallized together with

RXR, which would allow a more precise understanding of dimerization at the atomic level. The interaction surfaces appear to be large, however, and are widely believed to involve H9 and H10 of each protein. Unfortunately, the deletion analysis described by Jin et al. [107] was unable to distinguish between structural abnormalities induced by the deletion and the actual dimerization domain itself. Thus, elucidation of the three-dimensional structure of VDR-RXR LBD heterodimer will be required to define precisely the interaction between VDR and RXR. Despite these in vitro studies, simple experiments designed to unequivocally demonstrate an absolute requirement for RXR in VDR function in intact cells have not been carried out, primarily because of the ubiquitous expression of endogenous RXR in mammalian cell lines. This remains important since it has been proposed that the activation of the osteopontin gene by 1,25(OH)2D3 involves only VDR homodimers [181]. It is also important because the ability of 1,25(OH)2D3 to suppress several growth factor/cytokines genes such as those for IL-2 [182], GM-CSF [183], and interferon γ [184] as well as human PTH [185] may not require RXR. This suggests that the need for RXR in vitamin D gene regulation may be gene promoter–specific. Jin and Pike [106] utilized yeast, which do not express VDR and RXR′, to recreate the VDR transcriptional

180 response unit and to test for the requirement of RXRs on a typical VDRE. While VDR exhibited little capacity to activate a chimeric gene promoter containing a VDRE derived from the OC genes in the absence of RXR, the addition of RXR dramatically increased VDR-mediated activation. Despite this study, perhaps the most definitive assessment of RXR participation has come from studies by Barsony and colleagues [186] using immunofluorescence and Pike and colleagues [54,55] using the chromatin immunoprecipitation techniques. Prufer et al. [186] discovered that both VDR and RXR colocalize in response to 1,25(OH)2D3 within the nucleus in a VDR DNA-binding domain-dependent manner as discrete complexes that are believed to represent sites of transcription. As depicted in Fig. 4, the studies by Kim et al. [55] showed unequivocally that treatment of cells with 1,25(OH)2D3 leads to the rapid colocalization of both VDR and RXR on a similar fragment of the Cyp24 gene promoter, a gene that contains typical VDREs and is believed to require a VDR/RXR heterodimer for activation. Although the resolution of binding in these intact cells does not permit direct determination of heterodimer formation, this conclusion is a reasonable one given the earlier in vitro biochemical interaction studies. Interestingly, 1,25(OH)2D3 also promotes the colocalization of both VDR and RXR to the osteopontin promoter as seen in Fig. 4 [55]. These data indicate that VDR homodimer-mediated activation of the osteopontin gene is unlikely.

C. Polarity of DNA Binding The asymmetric nature of natural VDREs (direct repeats) coupled with the heteromeric nature of the receptor activation unit (VDR-RXR) indicates that the two receptor subunits must bind to the VDRE with a defined polarity. Studies by Jin and Pike [106] and Freedman and co-workers [164] addressed this question in detail. Through the use of chimeric receptors and chimeric response elements, it is now clear that RXR binds to the upstream 5′ half-element and the VDR binds to the downstream 3′ half-element of VDREs oriented on the DNA sense strand as illustrated in Fig. 5. This organization is consistent with the relative polarity noted for both RXR-TR [180] and RXR-RAR heterodimers [187,188] bound to their respective response elements.

D. Transactivation by the VDR and RXR Involves Coregulators Studies by McDonnell et al. [103] first demonstrated that the responsiveness of a transfected human

J. WESLEY PIKE AND NIRUPAMA K. SHEVDE

OC promoter to 1,25(OH)2D3 in VDR-negative cell lines such as CV-1 or COS7 required co-introduction of VDR. This biological assay enabled an extensive examination of the domains of the VDR that are required for transactivation and revealed the not surprising crucial nature of both the DNA binding and heterodimerization functions in transactivation. They also revealed the critical role of an intact AF-2 function in the context of normal DNA binding and heterodimer function for subsequent transactivation. Interestingly, this assay also established the inactive nature of many mutant VDRs that were identified in the human syndrome of hereditary l,25(OH)2D3-resistant rickets (HVDRR) considered in more detail in Chapter 72 [104,167,168]. Activation of the VDR involves a l,25(OH)2D3dependent conformational change within the LBD, which as indicated earlier creates a functional AF-2 domain. The protein components that interact within the AF-2 region of the nuclear receptors in general include the SWI/SNF complex that functions to remodel chromatin in an ATP-dependent manner [189,190], the p300/CBP coactivator complex [191,192] that contains nuclear receptor interacting proteins SRC1, SRC2, and/or SRC3 [132] and function to acetylate histones [193, 194], mediator complex that contains the nuclear receptor interacting protein DRIP205 and functions to facilitate entry of RNA polymerase II [133,195,196], and likely other [197] complexes. The potential interactions of the SRC-1/CBP and the DRIP complexes with the VDR/RXR heterodimer are illustrated in Fig. 5. The exact role of each of these enzyme-containing complexes during transcription is unclear as is the temporal order in which they are recruited to receptor-bound promoters during activation. The VDR has been shown to interact directly with specific components of each of these complexes in vitro [197–200], and the addition of many of these factors to cells via transfection assays strongly enhances l,25(OH)2D3-induced transcription. Direct evidence that the VDR recruits many of these factors to target promoters in living cells has only recently emerged, however, through the use of chromatin immunoprecipitation assays. Zhang et al. [161] and Kim et al. [55], both have demonstrated that l,25(OH)2D3 induces the recruitment of SRC-1 to the Cyp24 promoter. Kim et al. [55] also showed that SRC-2 and SRC-3 were similarly recruited, although the temporal pattern of recruitment differed among these coactivators. This group also observed that the HAT containing integrators p300 and CBP as well as mediator component DRIP205 were also recruited to the Cyp24 promoter in response to l,25(OH)2D3 [55]. Importantly, the consequence of this recruitment in the case of DRIP205 appears to be association of RNA polymerase II [55], whereas in the case of the p160 complex they result in the acetylation of histone 4 (Kim and Pike, unpublished).

181

CHAPTER 11 The Vitamin D Receptor

of the RXR H12 region by Thompson et al. [209]. The concept that RXR is not a “silent partner” but rather a transcriptionally active partner of the VDR has been reinforced also through studies by Pathrose et al. [135], who used RXR-selective LxxLL peptides to demonstrate that blockade of RXR recruitment of coactivators also leads to inhibition of transcription induced by l,25(OH)2D3. Finally, recent studies by Bettoun et al. [210] also support the contribution of RXR to VDR-induced transcription. Thus, it seems likely that both subunits of the VDR heterodimer complex are essential to the regulation of gene expression by l,25(OH)2D3.

These studies provide clear evidence not only that the VDR interacts with these coregulators on specific genes in living cells but that the recruitment of these complexes leads to functional consequences with regard to gene promoter status and transcriptional activity. A number of proteins also interact with the VDR in a ligand-independent manner in regions outside that of AF-2. These include the protein NCoA62, a coactivator that may play a role in mRNA transcript splicing [201]; WINAC [202]; Alien [203]; hairless [204]; SUG1, a 19S proteasomal component that may play a role in VDR degradation [205]; E6-AP, an E3 ligase that may play a role in VDR processing [Pike et al., unpublished]; and corepressors such as NCoR [206] and SMRT [207]. Little is known of how these proteins interact with the VDR or of the domain(s) within the VDR that mediates the interactions. The VDR also interacts with the core promoter transcription factor TFIIB [108,208], suggesting that the biochemical mechanisms by which the VDR contributes regulatory inputs into the basal transcriptional apparatus will be multifaceted and complex. The reader is referred to Chapters 13–17 for additional details regarding the interaction of the VDR with potential comodulators and the core transcription factor TFIIB.

F. VDR Dynamics Apart from the involvement of VDR and RXR as well as numerous coregulators in transcription, recent studies have suggested that the process is highly dynamic. Indeed, general studies in cultured cells using chromatin immunoprecipitation and other assays have indicated that the binding of nuclear receptors to target gene promoters is cyclic in nature [211–213]. Accordingly, hormone-induced binding of the receptor to gene promoters exhibited cycles with a periodicity ranging from 30 to 40 min. The recruitment of coregulator complexes is also cyclic, suggesting the possibility of a spatial and temporal order to the transcriptional process that is necessary for gene regulation [214]. Whether this cycling process is fundamental to the mechanism whereby genes are activated and transcription is maintained by transcription factors is currently unclear. The binding of the VDR and RXR to the Cyp24 and OPN promoters in response to l,25(OH)2D3 is also cyclic, as seen in Fig. 6 [55]. This binding is accompanied in parallel fashion by coactivators such as the p160 family of genes, CBP, and p300 as well as

E. Direct Involvement of RXR in Transactivation One question that has emerged from many of the functional studies of VDR is whether RXR contributes directly to the recruitment of the coactivators described earlier. This is particularly relevant since RXR also contains a functional AF-2 which interacts in vitro with most of the coactivator complexes described earlier. That RXR contributes directly to VDR-mediated transactivation has been demonstrated through mutagenesis

B

CYP24

Relative intensity, ×10−3

VDR

120

RXR pol II

80 40 0

OPN 120

VDR RXR pol II

80

40

Time (min)

165

150

135

120

105

90

75

60

45

30

15

0

165

150

135

120

105

90

75

60

45

30

15

0 0

Relative intensity, ×10−3

A

Time (min)

FIGURE 6 Localization of VDR, RXR, and RNA pol II to the (A) Cyp24 and (B) OPN promoters in response to 1,25(OH)2D3 as a function of time. The results are a densitometric depiction of the PCR products detected following amplification of either Cyp24 or OPN gene promoter DNA. For details, see Kim et al. [55].

182

J. WESLEY PIKE AND NIRUPAMA K. SHEVDE

Coactivator complexes

VDRE

TATA

Ligand-driven complex assembly

1,25(OH)2D3

RXR

VDR PCAF

HAT complex

RXR VDR

CBP/p300 p160s RXR VDR

Proteasome-driven disassembly

HISTONE

VDRE

HISTONE

Coactivator complexes

DRIP complex

DRIP 205 RXR VDR VDRE

TATA

FIGURE 7

Model for the 1,25(OH)2D3-driven assembly of transcriptionally active regulatory complexes on target genes and proteasome-directed disassembly of these complexes during signal processing. The order of recruitment and the influence of the gene itself on this process are currently unknown.

by the receptor interacting component of the DRIP complex, DRIP205 [55]. While current efforts are in place to determine the fundamental role of cycling in nuclear receptor activation, it is clear that this process may well enable precise real-time cellular monitoring of ambient hormone concentrations. A model for l,25(OH)2D3-driven assembly of VDR/RXR cofactor complexes on DNA and their proteasome mediated disassembly drawn from Kim et al. [55] is seen in Fig. 7.

G. General Conclusions An understanding of the functional activity of the VDR has emerged as a direct result of biochemical, molecular biological, and structural investigation. While insights gained from the three-dimensional structure of the VDR and other nuclear receptors have been exceptionally revealing, they have benefited enormously from numerous earlier biochemical and molecular biologic studies that have provided a wealth of information regarding the general domain structure and function of the nuclear receptor family. It is, in fact, a testimony to the effectiveness of these latter approaches that such great insights have emerged and that the crystallography has in many cases been largely confirmatory. The three approaches have, however, been highly complementary and have greatly extended our understanding of nuclear receptor structure and function.

VI. THE HUMAN VITAMIN D–RECEPTOR CHROMOSOMAL GENE A. Organization of the Gene The cloning of the VDR structural gene in 1987 initiated over a decade of highly productive research on the mechanism of action of vitamin D. Interestingly, unlike many of the other members of the nuclear receptor family, which exist as products of multiple genes, the VDR itself remains the apparent product of a single gene. Indeed, the recently published sequence analysis of both the mouse and human genomes appears to confirm the existence of only one VDR [91,92]. The human VDR gene is reported to lie on chromosome 12 [215]. The initial organization of the intron–exon structure of human VDR chromosomal gene corresponding to the sequence of the VDR reported by Baker et al. [85] was determined in 1988 [104]. Additional efforts have defined the complete structure of the gene [216]. Since the VDR gene is considered in some depth in Chapter 12, it will only be described briefly in this section. Restriction mapping of several lambda clones and a series of four recovered human cosmid clones coupled to nucleotide sequence analysis of relevant portions of these clones revealed a gene spanning over 75 kb of DNA. Eight exons comprise the coding sequence of the VDR protein. The first of these is exon 2, which contains the most proximal 3 bp of the 5′ noncoding

183

CHAPTER 11 The Vitamin D Receptor

sequence, the translation start site, and nucleotide sequence that encodes the first zinc finger module. Exon 3, which lies approximately 15 kb downstream, encodes the second zinc finger module. Exons 4, 5, and 6 encode the D region or hinge. Exons 6, 7, 8, and 9 encode a portion of the hinge and the carboxy-terminal E/F region as well as approximately 3200 nucleotides of 3′ noncoding sequence. The relationship between these exons, the full-length mRNA, and the VDR protein itself is illustrated in Fig. 8. It is clear that the human chromosomal gene for the VDR is not unlike other steroid receptor genes in size and exon organization. Insight into the 5′ end of this large gene and its promoter has also emerged over the past decade, revealing the promoter and regulatory regions of the human VDR to be highly complex (Fig. 8). This seems to be untrue of the rodent VDR genes where a single promoter is likely responsible for VDR expression [217]. Two short exons lie upstream of exon 2 and account for the known 5′ noncoding sequence reported by Baker et al. [85]. These exons, termed 1a and 1c, are 77 and 81 bp in length. A promoter (P1) lies immediately upstream of exon 1a and is clearly characterized by its GC-rich nature and the absence of a TATA box. Interestingly, an exon of 121 bp not found in the originally reported sequence of Baker et al. [85] and termed exon 1b is located 4.5 kb downstream of exon 1a. Variable use of exons 1b and 1c leads to the production

−281 −204

of alternatively spliced mRNAs whose nature and function remain unknown. More recently, additional upstream exons and their associated promoters have been ascribed to the human VDR gene [218]. These additions, as well as the transcripts and protein products that they produce [219], are described more completely in Chapter 12. Since the relative abundance of the bulk of these transcripts is exceedingly low, their contribution to overall VDR expression and to the various potential forms of the VDR proteins that are produced remains controversial. Clearly, additional work will be required to establish the bona fides of these novel forms of the VDR. Nevertheless, the presence of these transcripts suggests unique and possibly tissue-selective regulation of VDR expression. The identity of regulatory control elements in the human gene as well as those in the mouse gene is only now beginning to emerge.

B. Polymorphisms within the VDR Gene Genetic polymorphisms have been identified within the human VDR gene. The first represents a C to T transition in the translation initiation site located in exon 2 [94,95]. The presence of a C in this position results in the initiation of translation at a site three codons downstream and in the production of a gene

−83 −2

FIGURE 8 Structural organization of the human chromosomal VDR gene. The human VDR gene locus (DNA) is composed of as many as 14 exons (1a–1f, 2 through 9) spanning over 75 kb of DNA. A 10-kb scale is indicated at the right. The location of exons relative to the mRNA transcript of ~4800 nucleotides (mRNA) and the encoded VDR protein of 427 amino acids (hVDR) is illustrated. With regard to the hVDR mRNA, negative numbers indicate 5′ noncoding nucleotides and positive numbers indicate protein encoding nucleotides beginning with +1 indicated by Baker et al. [85]. Numbers below the hVDR protein indicate the amino acid residue boundaries of shaded homology domains. Regions of functionality are designated A/B, C, D, and E/F as in Figs. 2 or 3.

184 product three amino acids shorter. The distribution of this polymorphism in the human population and a potential relationship between the frequency of this polymorphism and bone mineral density were discussed in an earlier section of this chapter, and additional studies have been carried out which explore this issue in greater detail [97–101]. One finding has been the observation that the smaller form of the VDR appears to be more active transcriptionally [96–98]. Additional polymorphisms, although they are not located in regions of the gene that might directly affect the structure of the protein, have also been defined within the introns located between exons 7 and 8 as well as within the 3′ noncoding region of exon 9 [220, see Chapter 68 and references therein]. These polymorphisms appear also to correlate with bone mineral density in several human populations and are hypothesized to be predictive for osteoporosis. At presence, these conclusions are highly controversial, however, and are not widely supported by numerous recent studies that have been carried out over the past several years. This topic and associated references are considered in depth in Chapter 68.

VII. CONCLUDING COMMENTS The basic elements of the mechanism of action of vitamin D have been defined. The pace of exploration into the actions of vitamin D has accelerated enormously since the mid-1980s, largely as a result of the molecular cloning of the VDR in 1987 but also as a result of the availability of cloned vitamin D target genes. As described in this chapter, we have gained considerable insight into the structure of the VDR and its compartmentalization into definable functional domains. The now-available crystal structure of the VDR LBD has placed much of this biochemical and molecular biologic work into a clearer perspective. The availability of recombinant clones has allowed investigation of the interaction of the VDR with vitamin D–inducible gene promoters and definition of VDREs. Further studies have revealed that the VDR requires a protein partner for DNA binding in the form of RXR, a central regulator of several nuclear receptors. Although additional research will be necessary, a considerable understanding of the role of receptor and its ligand in the regulation of transcription has begun to emerge. These insights are currently being utilized to gain an understanding of the tissue-selective mechanisms of action of a new generation of vitamin D analogs under consideration as therapeutic agents for a broad range of indications that include skin diseases, immunologic disorders, and cancer. The cloning of the

J. WESLEY PIKE AND NIRUPAMA K. SHEVDE

VDR also enabled the recovery of its chromosomal gene. Subsequent investigation into the nature of the human syndrome of hereditary l,25(OH)2D3-resistant rickets (HVDRR) was made possible through characterization of the VDR gene itself [104,216] and has revealed the underlying cause to be mutations in the gene that produce dysfunctional receptor protein (described in Chapter 72). This discovery, together with the genetic ablation of the VDR gene in mice (see Chapter 20 for details) that clearly mimics physiologically the syndrome of HVDRR, confirm the central role of the VDR in the regulation of mineral metabolism. In view of our current understanding of VDR action on target gene promoters, future studies are likely to focus on the downstream roles of co-regulator complexes in mediating VDR action at the individual gene promoter level. Questions that need to be answered include those that address the individual roles of coregulator complexes in transcription, the order in which they are recruited to individual promoters, and whether features of the promoter itself play a deterministic role in the recruitment process and cellular response. Elucidation of these processes will be essential to the eventual understanding of the unique actions of vitamin D analogs that are potentially useful therapeutically. Also likely to emerge from these studies will be a better understanding of the regulation of the VDR at both the transcriptional and posttranslational levels and the contribution of this regulation to the mechanism of vitamin D action. Perhaps more important than these molecular details is the likelihood that we will achieve a better understanding of how vitamin D controls directly as well as indirectly the expression of broad networks of genes that are responsible in turn for cellular and tissue activities such as those involving proliferation and differentiation. Thus, although significant progress has been made in the past few years there is still much work that needs to be done.

References 1. Beato M 1989 Gene regulation by steroid hormones. Cell 56:335–344. 2. Evans RM 1988 The steroid and thyroid hormone receptor super-family. Science 240:889–895. 3. O’Malley BW 1990 The steroid receptor superfamily: More excitement predicted for the future. Mol Endocrinol 4:363–369. 4. Beato M, Herrliche P, Schutz G 1995 Steroid hormone receptors: Many actors in search of a plot. Cell 83:851–857. 5. Mangelsdorf DJ, Evans RM 1995 The RXR heterodimer and orphan receptors. Cell 83:841–850. 6. Mangelsdorf DJ, Thummel C, Beato M, Herrliche P, Schultz G, Umesono K, Blumberg B, Kastner P, Mark M, Chambon P,

185

CHAPTER 11 The Vitamin D Receptor

7. 8.

9.

10.

11. 12. 13.

14.

15.

16.

17. 18. 19. 20.

21. 22. 23. 24. 25.

Evans RM 1995 The nuclear receptor superfamily: The second decade. Cell 83:835–839. Norman AW 1968 The mode of action of vitamin D. Biol Rev 243:4055–4064. Lawson DEM, Wilson PW, Kodicek E, Morrison HR, Williams DH 1971 Identification of 1,25-dihydroxycholecalciferol, a new kidney hormone controlling calcium metabolism. Nature 230:228–230. Norman AW, Myrtle JF, Midgett RJ, Nowicki HG, Williams V, Popjak G 1971 1,25-Dihydroxycholecalciferol: Identification of the proposed active form of vitamin D in the intestine. Science 173:51–54. Holick MF, Schnoes HK, DeLuca HF, Suda T, Cousins RJ 1971 Isolation and identification of 1,25-dihydroxycholecalciferol. A metabolite of vitamin D active in intestine. Biochemistry 10:2799–2804. Fraser DR, Kodicek E 1970 Unique biosynthesis by kidney of a biologically active vitamin D metabolite. Nature 228:764–766. Haussler MR, Norman AW 1969 Chromosomal receptor for a vitamin D metabolite. Proc Natl Acad Sci USA 62:155–162. McDonnell DP, Mangelsdorf DJ, Pike JW, Haussler MR, O’Malley BW 1987 Molecular cloning of complementary DNA encoding the avian receptor for vitamin D. Science 235:1214–1217. Pike JW, Marion SL, Donaldson CA, Haussler MR 1983 Serum and monoclonal antibodies against the chick intestinal receptor for 1,25-dihydroxyvitamin D3: Generation by a preparation enriched in a 64,000 dalton protein. J Biol Chem 258:1289–1296. Dame MC, Pierce EA, Prahl JM, Hayes CE, DeLuca HF 1986 Monoclonal antibodies to the porcine intestinal receptor for 1,25-dihydroxyvitamin D3: Interaction with distinct receptor domains. Biochemistry 25:4523–4534. Brumbaugh PF, Haussler MR 1974 1,25-Dihydroxycholecalciferol receptors in intestine. Temperature-dependent transfer of the hormone to chromatin via a specific receptor. J Biol Chem 249:1258–1262. Lawson DEM, Wilson PW 1974 Intranuclear localization and receptor proteins for 1,25-dihydroxycholecalciferol in chick intestine. Biochem J 144:573–583. Brumbaugh PF, Haussler MR 1975 Specific binding of 1,25dihydroxycholecalciferol to nuclear components of chick intestine. J Biol Chem 250:1588–1594. Brumbaugh PF, Hughes MR, Haussler MR 1975 Cytoplasmic and nuclear components for 1,25-dihydroxyvitamin D3 in chick parathyroid glands. Proc Natl Acad Sci USA 72:4871–4875. Pike JW, Gooze LL, Haussler MR 1980 Biochemical evidence for 1,25-dihydroxyvitamin D receptor macromolecules in parathyroid, pancreas, pituitary, and placental tissues. Life Sci 26:407–414. Kream BE, Jose M, Yamada S, DeLuca HF 1977 A specific high affinity binding macromolecule for 1,25-dihydroxyvitamin D3 in fetal bone. Science 197:1086–1088. Chandler JS, Pike JW, Haussler MR 1979 1,25-Dihydroxyvitamin D3 receptors in rat kidney cytosol. Biochem Biophys Res Commun 90:1057–1063. Colston K, Feldman D 1980 Nuclear translocation of the 1,25-dihydroxycholecalciferol receptor in mouse kidney. J Biol Chem 255:7510–7513. Kream BE, Yamada Y, Schnoes HK, DeLuca HF 1977 Specific cytosol-binding protein for 1,25-dihydroxyvitamin D3 in rat intestine. J Biol Chem 254:9488–9491. Abe E, Miyaura C, Sakagami H, Takeda M, Konno K, Yamazaki T, Yashiki S, Suda T 1981 Differentiation of mouse

26.

27. 28. 29. 30. 31. 32. 33. 34.

35.

36. 37.

38. 39.

40. 41. 42. 43. 44.

myeloid leukemia cells induced by 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 78:4990–4994. Shavit ZB, Teitlebaum SL, Reitsma P, Hall A, Pegg LE, Trial J, Kahn AJ 1983 Induction of monocytic differentiation and bone resorption by 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 80:5907–5911. Christakos S, Norman AW 1981 Biochemical characterization of 1,25-dihydroxyvitamin D3 receptors in chick pancreas and kidney cytosols. Endocrinology 108:140–149. Haussler MR, Manolagas SC, Deftos L 1980 Evidence for a 1,25-dihydroxyvitamin D3 receptor-like macromolecule in rat pituitary. J Biol Chem 255:5007–5010. Dokoh S, Donaldson CA, Marion SL, Pike JW, Haussler MR 1983 The ovary: A target organ for 1,25-dihydroxyvitamin D3. Endocrinology 112:200–206. Merke J, Kreusser W, Bier B, Ritz E 1983 Demonstration and characterization of a testicular receptor for 1,25-dihydroxyvitamin D3 in the rat. Eur J Biochem 130:303–308. Colston K, Hirst M, Feldman D 1980 Organ distribution of the cytoplasmic 1,25-dihydroxycholecalciferol receptor in various mouse tissues. Endocrinology 107:1916–1922. Walters MR, Wicker DC, Riggle PC 1986 1,25-Dihydroxyvitamin D3 receptor identified in the rat heart. J Mol Cell Cardiol 18:67–72. Colston K, Colston MJ, Fieldsteel AH, Feldman D 1982 1,25-Dihydroxyvitamin D receptors in human epithelial cancer cell lines. Cancer Res 42:856–859. Colston K, Colston MJ, Feldman D 1981 1,25-Dihydroxyvitamin D and malignant melanoma: The presence of receptors and inhibition of cell growth in culture. Endocrinology 108:1083–1086. Eisman JA, Martin TJ, MacIntyre I, Frampton RJ, Moseley JM, Whitehead R 1980 1,25-Dihydroxyvitamin D3 receptors in a cultured human breast cancer cell line (MCF-7). Biochem Biophys Res Commun 93:9–15. Haussler MR 1986 Vitamin D receptor: Nature and function. Annu Rev Nutr 6:527–562. Feldman D, Chen T, Hirst M, Colston K, Karasek M, Cone C 1980 Demonstration of 1,25-dihydroxyvitamin D receptors in human skin biopsies. J Clin Endocrinol Metab 51: 1463–1465. Prowedini DM, Tsoukas CD, Deftos LJ, Manolagas SC 1983 1,25-Dihydroxyvitamin D3 receptors in human leucocytes. Science 221:1181–1183. Bhalla AK, Amento EP, Clemens TL, Holick MF, Krane SM 1983 Specific high affinity receptors for 1,25-dihydroxyvitamin D3 in human peripheral blood mononuclear cells: Presence in monocytes and induction in T lymphocytes following activation. J Clin Endocrinol Metab 57:1308–1310. Krageballe K 1992 Treatment of psoriasis with calcipotriol and other vitamin D analogues. J Am Acad Dermatol 27: 1001–1008. Simpson RU, DeLuca HF 1980 Characterization of a receptorlike protein for 1,25-dihydroxyvitamin D3 in rat skin. Proc Natl Acad Sci USA 77:5822–5827. Zile M, Barsness EC, Yamada S, Schnoes HK, DeLuca HF 1976 Localization of 1,25-dihydroxyvitamin D3 in intestinal nuclei in vivo. Arch Biochem Biophys 186:15–24. Jones PG, Haussler MR 1979 Scintillation autoradiography localization of 1,25-dihydroxyvitamin D3 in chick intestine. Endocrinology 104:313–321. Stumpf WE, Sar M, Reid FA, Tanaka Y, DeLuca HF 1979 Target cells for 1,25-dihydroxyvitamin D3 in intestinal tract, stomach, kidney, skin, pituitary, and parathyroids. Science 206:1188–1190.

186 45. Stumpf WE, Sar M, Clark SA, DeLuca HF 1982 Brain target sites for 1,25-dihydroxyvitamin D3. Science 215:1403–1405. 46. Clemens TL, Garrett KP, Zhou XY, Pike JW, Haussler MR, Dempster DW 1988 Immunocytochemical localization of the 1,25-dihydroxyvitamin D3 receptor in target cells. Endocrinology 122:1224–1230. 47. Berger U, Wilson P, McClelland RA, Colston K, Haussler MR, Pike JW, Coombes RC 1988 Immunocytochemical detection of 1,25-dihydroxyvitamin D3 receptors in normal human tissues. J Clin Endocrinol Metab 67:607–613. 48. Milde P, Merke J, Ritz E, Haussler MR, Rauterberg EW 1989 Immunocytochemical detection of 1,25-dihydroxyvitamin D3 receptors by monoclonal antibodies: A comparison of four immunoperoxidase methods. J Histochem Cytochem 37:1609–1617. 49. Berger U, Wilson P, McClelland RA, Colston K, Haussler MR, Pike JW, Coombes RC 1987 Immunocytochemical detection of 1,25-dihydroxyvitamin D3 receptors in breast cancer. Cancer Res 47:6793–6799. 50. Nevah-Many T, Marx R, Keshet E, Pike JW, Silver J 1990 Regulation of 1,25-dihydroxyvitamin D3 receptor gene expression by 1,25-dihydroxyvitamin D3 in the parathyroid in vivo. J Clin Invest 86:1968–1975. 51. Mocharla H, DeTogni P, Yu XP, White P, Jilka RJ, Manolagas SC 1994 In search for an association between vitamin D receptor polymorphism and VDR mRNA expression in peripheral blood mononuclear cells of normal volunteers. J Bone Miner Res 9(Suppl. 1):S416. 52. Walters MR, Hunziker W, Norman AW 1980 Unoccupied 1,25-dihydroxyvitamin D3 receptors. Nuclear/cytosol ratios depend on ionic strength. J Biol Chem 255:6799–6805. 53. Walters MR, Hunziker W, Norman AW 1981 1,25-Dihydroxyvitamin D3 receptors: Intermediates between triiodothyronine and steroid hormone receptors. Trends Biochem Sci 6:268–271. 54. Yamamoto H, Shevde NK, Warrier A, Plum LA, DeLuca HF 2003 2-Methylene-19-nor-(20S)-1,25-dihydroxyvitamin D3 potently stimulates gene-specific DNA binding of the vitamin D receptor in osteoblasts. J Biol Chem 278: 31756–31765. 55. Kim S, Yamamoto H, Shevde NK, Pike JW 2004 1,25Dihydroxyvitamin D3 stimulates dynamic vitamin D receptor/ retinoid x receptor DNA-binding and coactivator recruitment in intact osteoblasts. J Bone Min Res 18(Suppl 2): 530 (Abst). 56. Racz A, Barsony J 1999 Hormone dependent translocation of vitamin D receptors is linked to transactivation. J Biol Chem 274:19352–19360. 57. Prufer K, Barsony J 2002 Retinoid X receptor dominates the nuclear import and export of the unliganded vitamin D receptor. Mol Endocrinol 16:1738–1751. 58. Pike JW, Haussler MR 1979 Purification of chicken intestinal receptor for 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 76:5488–5494. 59. Haussler MR, Pike JW, Chandler JS, Manolagas SC, Deftos LJ 1981 Molecular actions of 1,25-dihydroxyvitamin D3: New cultured cell models. Ann NY Acad Sci 372:502–517. 60. Katzenellenbogen JA, O’Malley BW, Katzenellenbogen BS 1996 Tripartite steroid hormone receptor pharmacology: interaction with multiple effector sites as a basis for the celland promoter-specific action of these hormones. Mol Endocrinol 6:119–131. 61. Mellon WS, DeLuca HF 1979 An equilibrium and kinetic study of 1,25-dihydroxyvitamin D3 binding to chicken intestinal cytosol employing high specific activity l,25-dihydroxy[3H-26,27] vitamin D3. Arch Biochem Biophys 197:90–95.

J. WESLEY PIKE AND NIRUPAMA K. SHEVDE

62. Wecksler WR, Norman AW 1980 A kinetic and equilibrium binding study of 1,25-dihydroxyvitamin D3 with its cytosol receptor from chick intestinal mucosa. J Biol Chem 255:3571–3574. 63. Kream BE, Jose JL, DeLuca HF 1977 The chick intestinal cytosol binding protein for 1,25-dihydroxyvitamin D3: A study of analog binding. Arch Biochem Biophys 179:462–468. 64. Wecksler WR, Okamura WH, Norman AW 1978 Quantitative assessment of the structural requirements for the interaction of 1,25-dihydroxyvitamin D3 with its chick intestinal mucosa receptor system. J Steroid Biochem 9:929–937. 65. Pike JW, Haussler MR 1983 Association of 1,25-dihydroxyvitamin D3 with cultured 3T6 mouse fibroblast. J Biol Chem 258:8554–8560. 66. Hunziker W, Walters MR, Bishop JE, Norman AW 1983 Unoccupied and in vitro and in vivo occupied 1,25-dihydroxyvitamin D3 intestinal receptors. Multiple biochemical forms and evidence for transformations. J Biol Chem 258:8642–8648. 67. Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179. 68. Tocchini-Valentini G, Roche N, Wurtz JM, Mitschler A, Moras D 2001 Crystal structures of the vitamin D receptor complexed to superagonist 20-epi ligands. Proc Natl Acad Sci USA 98:5491–5496. 69. Vanhooke JL, Benning MM, Bauer CB, Pike JW, DeLuca HF 2004 Molecular structure of the rat vitamin D receptor ligand binding domain complexed with 2-carbon-substituted vitamin D3 hormone analogs and an LxxLL-containing coactivator peptide. Biochemistry, 43:4101–4110. 70. Pike JW, Sleator N, Haussler MR 1987 Chicken intestinal receptor for 1,25-dihydroxyvitamin D3. Immunologic characterization and homogeneous isolation of a 60,000-dalton protein. J Biol Chem 262:1305–1311. 71. Mangelsdorf DJ, Pike JW, Haussler MR 1987 Avian and mammalian receptors for 1,25-dihydroxyvitamin D3: In vitro translation to characterize size and hormone-dependent regulation. Proc Natl Acad Sci USA 84:354–358. 72. Pike JW, Kesterson RA, Scott RA, Kerner SA, McDonnell DP, O’Malley BW 1988 Vitamin D receptors: Molecular structure of the protein and its chromosomal gene. In: Normal AW, Schaefer K, Grigoleit H-G, Herrath DV (eds) Vitamin D: Molecular, Cellular and Clinical Endocrinology, de Gruyter, Berlin, pp. 215–224. 73. Elaroussi MA, Prahl JM, DeLuca HF 1992 The avian vitamin D receptors: Primary structures and their origins. Proc Natl Acad Sci USA 91:11596–11600. 74. Pike JW, 1984 Monoclonal antibodies to chick intestinal receptors for 1,25-dihydroxyvitamin D3: Interaction and effects of binding on receptor function. J Biol Chem 259:1167–1173. 75. Allegretto EA, Pike JW 1985 Trypsin cleavage of chick 1,25dihydroxyvitamin D3 receptors. Generation of discrete polypeptides which retain hormone but are unreactive to DNA and monoclonal antibody. J Biol Chem 260: 10139–10145. 76. Allegretto EA, Pike JW, Haussler MR 1987 Immunological detection of unique proteolytic fragments of the chick 1,25-dihydroxyvitamin D3 receptor. Distinct 20-kDa DNA binding and 45-kDa hormone-binding species. J Biol Chem 262:1312–1319. 77. Allegretto EA, Pike JW, Haussler MR 1987 C-Terminal proteo-lysis of the avian 1,25-dihydroxyvitamin D3 receptor. Biochem Biophys Res Commun 147:479–485.

187

CHAPTER 11 The Vitamin D Receptor

78. Heyman RA, Mangelsdorf DJ, Dyck JA, Stein RB, Eichele G, Evans RM, Thaller C 1992 9-cis-Retinoic acid is a high affinity ligand for the retinoid X receptor. Cell 68:397–406. 79. Forman BM, Tontonoz P, Chen C, Brun RP, Speigelman BM, Evans RM 1995 15-Deoxy-∆12,14-prostaglandin J2 is a ligand for the adipocyte determination factor PPARy. Cell 83:803–812. 80. Forman BM, Goode E, Chen J, Oro AE, Bradley DJ, Perlman T, Noonan DJ, Burka LT, McMorris T, Lamph WW, Evans RM, Weinberger C 1995 Identification of a nuclear receptor that is activated by farnesol metabolites. Cell 81:687–693. 81. Janowski BA, Willy PJ, Devi TR, Falck JR, Mangelsdorf DJ 1996 An oxysterol signalling pathway mediated by the nuclear receptor LXR alpha. Nature 383:728–731. 82. Chawla A, Repa JJ, Evans RM, Mangelsdorf DJ 2001 Nuclear receptors and lipid physiology: opening the X-files. Science 294:1866–1870. 83. Hollenberg SM, Weinberger C, Ong ES, Cerelli G, Ono AE, Lebo R, Thompson EB, Rosenfeld MG, Evans RM 1985 Primary structure and expression of a functional human glucocorticoid receptor cDNA. Nature 318:635–641. 84. Weinberger C, Thompson CC, Ong ES, Lebo R, Gruol DJ, Evans RM 1986 The c-erbA gene encodes a thyroid hormone receptor. Nature 324:641–646. 85. Baker AR, McDonnell DP, Hughes MR, Crisp TM, Mangelsdorf DJ, Haussler MR, Shine J, Pike JW, O’Malley BW 1988 Molecular cloning and expression of human vitamin D receptor complementary DNA: Structural homology with thyroid hormone receptor. Proc Natl Acad Sci USA 85:3294–3298. 86. Burmester JK, Maeda N, DeLuca HF 1988 Isolation and expression of rat 1,25-dihydroxyvitamin D receptor cDNA. Proc Natl Acad Sci USA 85:1005–1009. 87. Goto H, Chen KS, Prahl JM, DeLuca HF 1992 A single receptor identical to that for intestinal T47D cells mediates the action of 1,25-dihydroxyvitamin D3 in HL-60 cells. Biochim Biophys Acta 1132:103–108. 88. Kamei Y, Kawada T, Fukuwatari T, Ono T, Kato S, Sugimoto E 1995 Cloning and sequence of the gene encoding the mouse vitamin D receptor. Gene 152:281–282. 89. Elaroussi MA, Prahl JM, DeLuca HF 1992 The avian vitamin D receptors: Primary structures and their origins. Proc Natl Acad Sci USA 91:11596–11600. 90. Li YC, Bergwitz C, Juppner H, Demay MB 1996 Cloning and characterization of the vitamin D receptor from Xenopus laevis. J Bone Miner Res 11(Suppl. 1):S161. 91. Ciana P, Vegeto E, Beato M, Chambon P, Gustafsson JA, Parker M, Wahli W, Maggi A 2002 Looking at nuclear receptors from the heights of Erice. Workshop on nuclear receptor structure and function. EMBO Rep 3:125–129. 92. Enmark E, Gustafsson JA 2001 Comparing nuclear receptors in worms, flies and humans. Trends Pharmacol Sci 22: 611–615. 93. McDonnell DP, Pike JW, O’Malley BW 1988 Vitamin D receptor: A primitive steroid receptor related to the thyroid hormone receptor. J Steroid Biochem 30:41–47. 94. Kesterson RA, Pike JW 1991 Unpublished. 95. Saijo T, Ito M, Takeda E, Huq AHM, Naito E, Yokoto I, Sone T, Pike JW, Kuroda Y 1991 A unique mutation in vitamin D receptor gene in three Japanese patients with vitamin D– dependent rickets type II: Utility of single-strand conformation polymorphism analysis for heterozygous carrier detection. Am J Hum Genet 49:668–673. 96. Jurutka PW, Remus LS, Whitfield GK, Thompson PD, Hsieh JC, Zitzer H, Tavakkoli P, Galligan MA, Dang HT, Haussler CA,

97.

98.

99.

100.

101.

102. 103.

104.

105. 106. 107. 108. 109. 110. 111.

112. 113.

Haussler MR 2000 The polymorphic N terminus in human vitamin D receptor isoforms influences transcriptional activity by modulating interaction with transcription factor IIB. Mol Endocrinol 14:401–420. Arai H, Miyamoto K, Taketani Y, Yamamoto H, Iemori Y, Morita K, Tonai T, Nishisho T, Mori S, Takeda E 1997 A vitamin D receptor gene polymorphism in the translation initiation codon: effect on protein activity and relation to bone mineral density in Japanese women. J Bone Miner Res 12:915–921. Gross C, Krishnan AV, Malloy PJ, Eccleshall TR, Zhao XY, Feldman D 1998 The vitamin D receptor gene start codon polymorphism: a functional analysis of FokI variants. J Bone Miner Res 13:1691–1699. Chiu KC, Chuang LM, Yoon C 2001 The vitamin D receptor polymorphism in the translation initiation codon is a risk factor for insulin resistance in glucose tolerant Caucasians. BMC Med Genet 2:2. Kubota M, Yoshida S, Ikeda M, Okada Y, Arai H, Miyamoto K, Takeda E 2001 Association between two types of vitamin D receptor gene polymorphism and bone status in premenopausal Japanese women. Calcif Tissue Int 68:16–22. Whitfield GK, Remus LS, Jurutka PW, Zitzer H, Oza AK, Dang HT, Haussler CA, Galligan MA, Thatcher ML, Encinas Dominguez C, Haussler MR 2001 Functionally relevant polymorphisms in the human nuclear vitamin D receptor gene. Mol Cell Endocrinol 177:145–159. Green S, Walter P, Kumar V, Krust A, Bornert JM, Argos P, Chambon P 1986 Human oestrogen receptor cDNA: Sequence expression and homology to v-erbA. Nature 320:134–139. McDonnell DP, Scott RA, Kerner RA, O’Malley BW, Pike JW 1989 Functional domains of the human vitamin D receptor regulate osteocalcin gene expression. Mol Endocrinol 3:635–644. Hughes MR, Malloy PJ, Kieback DG, Kesterson RA, Pike JW, Feldman D, O’Malley BW 1988 Point mutations in the human vitamin D receptor gene associated with hypocalcemic rickets. Science 242:1702–1705. Rochel N, Tocchini-Valentini G, Egea PF, Juntunen K, Garnier JM, Vihko P, Moras D 2001 Eur J Biochem 268:971–979. Jin CH, Pike JW 1996 Human vitamin D receptor dependent transactivation in Saccharomyces cerevisiae requires retinoid X receptor. Mol Endocrinol 10:196–205. Jin CH, Kerner SA, Hong MH, Pike JW 1996 Transcriptional activation and dimerization functions of the human vitamin D receptor. Mol Endocrinol 10:945–957. MacDonald PN, Sherman DR, Dowd DR, Jefcoat SC Jr, DeLisle RK 1995 The vitamin D receptor interacts with general transcription factor IIB. J Biol Chem 270:4748–4752. Berg JM 1989 DNA binding specificity and steroid receptors. Cell 57:1065–1068. Berg JM 1988 Proposed structure for the zinc-binding domains from transcription factor IIIA and related proteins. Proc Natl Acad Sci USA 85:99–102. Mader S, Kumar V, deVereneuil H, Chambon P 1989 Three amino acids of the oestrogen receptor are essential to its ability to distinguish an oestrogen from a glucocorticoid responsive receptor. Nature 338:271–274. Umesono K, Evans RM 1989 Determinants of target gene specificity for steroid/thyroid hormone receptors. Cell 57:1139–1146. Schwabe JWR, Neuhaus D, Rhodes D 1990 Solution structure of the DNA binding domain of the oestrogen receptor. Nature 348:458–461.

188 114. Hard T, Kellenbach E, Boelens R, Maler BA, Dahlman K, Freedman LP, Carlstedt-Duke J, Yamamoto KR, Gustafsson JA, Kaptein R 1990 Solution structure of the glucocorticoid receptor DNA binding domain. Science 249:157–160. 115. Luisi BF, Xu W, Otwinowski Z, Freedman LP, Yamamoto KR, Sigler PB 1991 Crystallographic analysis of the interaction of the glucocorticoid receptor with DNA. Nature 352: 497–505. 116. Schwabe JWR, Chapman L, Finch JT, Rhodes D 1993 The crystal structure of the oestrogen receptor DNA binding domain bound to DNA: How receptors discriminate between their response elements. Cell 75:567–578. 117. Lee MS, Kliewer SA, Provencal J, Wright PE, Evans RM 1993 Structure of the retinoid X receptor α DNA binding domain: A helix required for homodimeric DNA binding. Science 260:1117–1121. 118. Rastinejad F, Perlmann T, Evans RM, Sigler PB 1995 Structural determinants of nuclear receptor assembly on DNA direct repeats. Nature 375:203–211. 119. Forman BM, Yang CR, Au M, Casanova J, Ghysdael J, Samuels HH 1989 A domain containing leucine-zipper-like motifs mediate novel in vivo interactions between the thyroid hormone and retinoic acid receptors. Mol Endocrinol 3:1610–1626. 120. Bourguet W, Ruff D, Chambon P, Gronemeyer H, Moras D 1995 Crystal structure of the ligand binding domain of the human nuclear receptor RXRa. Nature 375:377–382. 121. Renaud JP, Natacha R, Ruff M, Vivat V, Chambon P, Gronemyer H, Moras D 1995 Crystal structure of the RARy ligand binding domain bound to all-trans-retinoic acid. Nature 378:681–689. 122. Wagner RL, Apriletti JW, McGrath ME, West BL, Baxter JD, Fletterick RJ 1995 A structural role for hormone in the thyroid hormone receptor. Nature 378:690–697. 123. Shiau AK, Barstad D, Loria PM, Cheng L, Kushner PJ, Agard DA, Greene GL 1998 The structural basis of estrogen receptor/coactivator recognition and the antagonism of this interaction by tamoxifen. Cell 95:927–937. 124. Williams S, Sigler P 1998 Atomic structure of progesterone complexed with its receptor. Nature 393:392–396. 125. Nolte RT, Wisely GB, Westin S, Cobb JE, Lambert MH, Kurokawa R, Rosenfeld MG, Willson TM, Glass CK Milburn MV 1998 Ligand binding and co-activator assembly of the peroxisome proliferator-activated receptor-gamma. Nature 395:137–143. 126. Downes M, Verdecia MA, Roecker AJ, Hughes R, Hogenesch JB, Kast-Woelbern HR, Bowman ME, Ferrer JL, Anisfeld AM, Edwards PA, Rosenfeld JM, Alvarez JG, Noel JP, Nicolaou KC, Evans RM 2003 A chemical, genetic, and structural analysis of the nuclear bile acid receptor FXR. Mol Cell 11:1079–1092. 127. Peleg S, Sastry M, Collins ED, Bishop JE, Norman AW 1995 Distinct conformational changes induced by 20-epi analogues of 1,25-dihydroxyvitamin D3 are associated with enhanced activation of the vitamin D receptor. J Biol Chem 270: 10551–10558. 128. Quack M, Carlberg C 1999 Selective recognition of vitamin D receptor conformations mediates promoter selectivity of vitamin D analogs. Mol Pharmacol 55:1077–1087. 129. Moras D 2003 Twelfth Workshop on Vitamin D Maastricht, The Netherlands, June 2003, p. 3 (Abstract). 130. Whitfield GK, Hsieh JC, Nakajima S, MacDonald PN, Thompson PD, Jurutka PW, Haussler CA, Haussler MR 1995 A highly conserved region in the hormone-binding domain

J. WESLEY PIKE AND NIRUPAMA K. SHEVDE

131.

132.

133. 134.

135.

136.

137.

138.

139.

140.

141. 142. 143. 144.

145.

of the human vitamin D receptor contains residues vital for heterodimerization with retinoid X receptor and for transcriptional activation. Mol Endocrinol 9:1166–1179. Danielian PS, White R, Lees JA, Parker MG 1992 Identification of a conserved region required for hormone dependent transcriptional activation by steroid hormone receptors. EMBO J 11:1025–1033. Xu J, O’Malley BW 2002 Molecular mechanisms and cellular biology of the steroid receptor coactivator (SRC) family in steroid receptor function. Rev Endocr Metab Disord 3:185–192. Rachez C, Freedman LP 2001 Mediator complexes and transcription. Curr Opin Cell Biol 13:274–280. McInerney EM, Rose DW, Flynn SE, Westin S, Mullen TM, Krones A, Inostroza J, Torchia J, Nolte RT, Assa-Munt N, Milburn MV, Glass CK, Rosenfeld MG 1998 Determinants of coactivator LXXLL motif specificity in nuclear receptor transcriptional activation. Genes Dev 12:3357–3368. Pathrose P, Barmina O, Chang CY, McDonnell DP, Shevde NK, Pike JW 2002 Inhibition of 1,25-dihydroxyvitamin D3dependent transcription by synthetic LXXLL peptide antagonists that target the activation domains of the vitamin D and retinoid X receptors. J Bone Miner Res 17:2196–2205. [Erratum in 2003 J Bone Miner Res 18:390.] Brzozowski AM, Pike AC, Dauter Z, Hubbard RE, Bonn T, Engstrom O, Ohman L, Greene GL, Gustafsson JA, Carlquist M 1997 Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 389:753–758. Shiau AK, Barstad D, Radek JT, Meyers MJ, Nettles KW, Katzenellenbogen BS, Katzenellenbogen JA, Agard DA, Greene G 2002 Structural characterization of a subtypeselective ligand reveals a novel mode of estrogen receptor antagonism. Nature Struc Biol 9:359–364. Tzukerman MT, Esty A, Santiso-Mere D, Danielian P, Parker MG, Stein RB, Pike JW, McDonnell DP 1994 Human estrogen receptor transcriptional capacity is determined by both cellular and promoter context and mediated by two functionally distinct intramolecular regions. Mol Endocrinol 8:21–30. McDonnell DP, Clemm DL, Hermann T, Goldman ME, Pike JW 1995 Analysis of estrogen receptor function in vitro reveals three distinct classes of anti-estrogens. Mol Endocrinol 9:659–669. Ducy P, Desbois C, Boyce B, Pinero G, Story B, Dunstan C, Smith E, Bonadio J, Goldstein S, Gundberg C, Bradley A, Karsenty G 1996 Increased bone formation in osteocalcindeficient mice. Nature 382:448–452. Price PA, Baukol SA 1980 1,25-Dihydroxyvitamin D3 increases synthesis of the vitamin D-dependent bone proteins by osteosarcoma cells. J Biol Chem 225:11660–11663. Pan LC, Price PA 1986 1,25-Dihydroxyvitamin D3 stimulates transcription of bone gla protein. J Bone Miner Res 1(Suppl. 1):A20. Celeste AJ, Rosen V, Buecker JL, Kriz R, Wang EA, Wozney JM 1986 Isolation of the human gene for bone gla protein utilizing mouse and rat cDNA clones. EMBO J 5:1885–1890. Kerner SA, Scott RA, Pike JW 1989 Sequence elements in the human osteocalcin gene confer basal activation and inducible response to hormonal vitamin D3. Proc Natl Acad Sci USA 86:4455–4459. Morrison NA, Shine J, Fragonas J-C, Verkest V, McMenemy ML, Eisman JA 1989 1,25-Dihydroxyvitamin D3 responsive element and glucocorticoid repression in the osteocalcin gene. Science 246:1158–1161.

CHAPTER 11 The Vitamin D Receptor

146. Yoon K, Rutledge SJC, Buenaga RF, Rodan GA 1988 Characterization of the rat osteocalcin gene: Stimulation of promoter activity by 1,25-dihydroxyvitamin D3. Biochemistry 27:8521–8526. 147. Ozono K, Liao J, Scott RA, Kerner SA, Pike JW 1990 The vitamin D responsive element in the human osteocalcin gene: Association with a nuclear proto-oncogene enhancer. J Biol Chem 265:21881–21888. 148. Demay MB, Gerardi JM, DeLuca HF, Kronenberg HM 1990 DNA sequences in the rat osteocalcin gene that bind the 1,25-dihydroxyvitamin D receptor and confer responsiveness to 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 87:369–373. 149. Lian JB, Stewart C, Puchacz E, Mackowiak S, Shalhoub V, Colart D, Sambetti G, Stein G 1989 Structure of the rat osteocalcin gene and regulation of vitamin D–dependent expression. Proc Natl Acad Sci USA 86:1143–1147. 150. Terpening CM, Haussler MR, Jurutka PW, Galligan MA, Komm BS, Haussler MR 1991 The vitamin D responsive element in the rat bone gla protein gene is an imperfect direct repeat that cooperates with other cis elements in 1,25dihydroxyvitamin D3 mediated transcriptional activation. Mol Endocrinol 5:373–385. 151. Noda M, Vogel RL, Craig AM, Prahl J, DeLuca HF, Denhardt D 1990 Identification of a DNA sequence responsible for binding of the 1,25-dihydroxyvitamin D3 receptor and 1,25-dihydroxyvitamin D3 enhancement of mouse secreted phosphoprotein 1 (Supp-1 or osteopontin) gene expression. Proc Natl Acad Sci USA 87:9995–9999. 152. Gill RK, Christakos S 1993 Identification of sequence elements in the mouse calbindin-D28K gene that confer 1,25-dihydroxyvitamin D- and butyrate inducible responses. Proc Natl Acad Sci USA 90:2984–2988. 153. Darwish HM, DeLuca HF 1992 Identification of a 1,25-dihydroxyvitamin D3 response element in the 5’ flanking region of the rat calbindin D-9K gene. Proc Natl Acad Sci USA 89:603–607. 154. Ohyama Y, Ozono K, Uchida M, Shinki T, Kato S, Suda T, Yamamoto O, Noshiro M, Kato Y 1994 Identification of a vitamin D responsive element in the 5′-flanking region of the rat 25-hydroxyvitamin D3 24-hydroxylase gene. J Biol Chem 269:10545–10550. 155. Zierold C, Darwish HM, DeLuca HF 1995 Two vitamin D response elements function in the rat 1,25-dihydroxyvitamin D3–24-hydroxylase promoter. J Biol Chem 269: 1675–1678. 156. Chen KS, DeLuca HF 1995 Cloning of the human 1,25dihydroxyvitamin D3–24-hydroxylase gene promote and identification of two vitamin D responsive elements. Biochim Biophys Acta 1263:1–9. 157. Liu M, Lee M-H, Cohen M, Freedman LP 1996 Transcriptional activation of the p21 gene by vitamin D3 leads to the differentiation of the myelomonocytic cell line U937. Genes Dev 10:142–153. 158. van den Bemd GJ, Jhamai M, Staal A, van Wijnen AJ, Lian JB, Stein GS, Pols HA, van Leeuwen JP 2002 A central dinucleotide within vitamin D response elements modulates DNA binding and transactivation by the vitamin D receptor in cellular response to natural and synthetic ligands. J Biol Chem 277:14539–14546. 159. Umesono K, Murikami KK, Thompson CC, Evans RM 1991 Direct repeats as selective response elements for the thyroid hormone, retinoic acid, and vitamin D3 receptors. Cell 65:1225–1266.

189 160. Sierra J, Villagra A, Paredes R, Cruzat F, Gutierrez S, Javed A, Arriagada G, Olate J, Imschenetzky M, Van Wijnen AJ, Lian JB, Stein GS, Stein JL, Montecino M 2003 Regulation of the bone-specific osteocalcin gene by p300 requires Runx2/Cbfa1 and the vitamin D3 receptor but not p300 intrinsic histone acetyltransferase activity. Mol Cell Biol 23:3339–3351. 161. Zhang C, Dowd DR, Staal A, Gu C, Lian JB, van Wijnen AJ, Stein GS, MacDonald PN 2003 Nuclear coactivator-62 kDa/ Ski-interacting protein is a nuclear matrix-associated coactivator that may couple vitamin D receptor-mediated transcription and RNA splicing. J Biol Chem 278:35325–35336. 162. Sone T, Kerner SA, Pike JW 1991 Vitamin D receptor interaction with specific DNA. Association as a 1,25-dihydroxyvitamin D3-modulated heterodimer. J Biol Chem 266: 23296–23305. 163. Freedman LP, Towers T 1991 DNA binding properties of the vitamin D receptor zinc fingers region. Mol Endocrinol 5:1815–1826. 164. Towers T, Luisi BL, Asianov A, Freedman LP 1993 DNA target selectivity by the vitamin D receptor: Mechanism for dimer binding to an asymmetric repeat element. Proc Natl Acad Sci USA 90:6310–6314. 165. Lemon B, Freedman LP 1996 Selective effects of ligands on vitamin D3 receptor and retinoid X receptor mediated gene activation in vivo. Mol Cell Biol 16:1006–1016. 166. Nishikawa J, Kitaura M, Matsumoto M, Imagawa M, Nishihara T 1994 Difference and similarity of DNA sequence recognized by VDR homodimer and VDR/RXR heterodimer. Nucleic Acids Res 22:2902–2907. 167. Sone T, Scott R, Hughes M, Malloy P, Feldman D, O’Malley BW, Pike JW 1989 Mutant vitamin D receptors which confer hereditary resistance to 1,25-dihydroxyvitamin D3 in humans are transcriptionally inactive in vitro. J Biol Chem 264: 20230–20234. 168. Hughes MR, Malloy PJ, O’Malley BW, Pike JW, Feldman D 1991 Genetic defects of the 1,25-dihydroxyvitamin D receptor. J Receptor Res 11:699–716. 169. Nakajima S, Hsieh JC, MacDonald PN, Galligan MA, Haussler CA, Whitfield GK, Haussler MR 1994 The C-terminal region of the vitamin D receptor is essential to form a complex with a receptor auxiliary factor required for high affinity binding to the vitamin D responsive element. Mol Endocrinol 8:159–172. 170. Liao J, Ozono K, Sone T, McDonnell DP, Pike JW 1990 Vitamin D receptor interaction with specific DNA requires a nuclear protein and 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 87:9751–9755. 171. Sone T, McDonnell DP, O’Malley BW, Pike JW 1990 Expression of the human vitamin D receptor in Saccharomyces cerevisiae: Purification properties and generation of polyclonal antibodies. J Biol Chem 265:21997–22003. 172. Sone T, Ozono K, Pike JW 1991 A 55-kilodalton accessory factor facilitates vitamin D receptor DNA binding. Mol Endocrinol 5:1578–1586. 173. MacDonald PN, Haussler CA, Terpening CM, Galligan MA, Reeder MC, Whitfield GK, Haussler MR 1991 Baculovirusmediated expression of the human vitamin D receptor: Functional characterization, vitamin D response element interactions and evidence for a receptor auxiliary factor. J Biol Chem 266:18808–18813. 174. Ross TK, Moss VE, Prahl JM, DeLuca HF 1992 A nuclear protein essential for binding of rat 1,25-dihydroxyvitamin D3 receptor to its response elements. Proc Natl Acad Sci USA 89:256–260.

190 175. Yu V, Delsert C, Andersen B, Holloway JM, Devary OV, Naar AM, Kirn SY, Boutin JM, Glass CK, Rosenfeld MG 1991 RXR/3: A coregulator that enhances binding of retinoic acid, thyroid hormone, and vitamin D receptors to their cognate response elements. Cell 67:1251–1266. 176. Leid M, Kastner P, Lyons R, Nakshatro H, Saunders M, Zacharewski T, Chen J-Y, Staub A, Gamier J-M, Mader S, Chambon P 1992 Purification, cloning and RXR identity of the HeLa cell factor with which RAR or TR heterodimerizes to bind target sequences efficiently. Cell 68:377–395. 177. Zhang XK, Hoffman B, Tran PB, Graupner G, Pfahl M 1992 Retinoid X receptor is an auxiliary protein for thyroid hormone and retinoic acid receptors. Nature 355:441–446. 178. Kliewer SA, Umesono K, Mangelsdorf DJ, Evans RM 1992 Retinoid X receptor interacts with nuclear receptors in retinoic acid, thyroid, and vitamin D signaling. Nature 355:446–449. 179. Perlmann T, Umesono K, Rangarajan PN, Forman BM, Evans RM 1996 Two distinct dimerization interfaces differentially modulate target gene specificity of nuclear hormone receptors. Mol Endocrinol 10:958–966. 180. Perlmann T, Rangarajan PN, Umesono K, Evans RM 1993 Determinants for selective RAR and TR recognition of direct repeat HREs. Genes Dev 7:1411–1422. 181. Freedman LP, Arce V, Perez Fernandez R 1994 DNA sequences that act as high affinity targets for the vitamin D3 receptor in the absence of the retinoid X receptor. Mol Endocrinol 8:265–273. 182. Alroy I, Towers TL, Freedman LP 1995 Transcriptional repression of the interleukin-2 gene by vitamin D3: direct inhibition of NFATp/AP-1 complex formation by a nuclear hormone receptor. Mol Cell Biol 15:5789–5799. 183. Towers TL, Staeva TP, Freedman LP 1999 A two-hit mechanism for vitamin D3-mediated transcriptional repression of the granulocyte-macrophage colony-stimulating factor gene: vitamin D receptor competes for DNA binding with NFAT1 and stabilizes c-Jun. Mol Cell Biol 19:4191–4109. 184. Staeva-Vieira TP, Freedman LP 2002 1,25-Dihydroxyvitamin D3 inhibits IFN-gamma and IL-4 levels during in vitro polarization of primary murine CD4+ T cells. J Immunol 168: 1181–1189. 185. Mackey SL, Heymont JL, Kronenberg HM, Demay MB 1996 Vitamin D receptor binding to the negative human parathyroid hormone vitamin D response element does not require the retinoid x receptor. Mol Endocrinol 10:298–305. 186. Prufer K, Racz A, Lin GC, Barsony J 2000 Dimerization with retinoid X receptors promotes nuclear localization and subnuclear targeting of vitamin D receptors. J Biol Chem 275:41114–41123. 187. Kurokawa B, Yu VC, Naar A, Kyakumoto S, Han Z, Silverman S, Rosenfeld MG, Glass CK 1993 Differential orientation of the DNA-binding domain and carboxy terminal dimerization interface regulate binding site selection by nuclear receptor heterodimers. Genes Dev 7:1423–1435. 188. Sucov HM, Murakami KK, Evans RM 1990 Characterization of an autoregulated response element in the mouse retinoic acid receptor type beta gene. Proc Natl Acad Sci USA 87: 5392–5396. 189. Fyodorov DV, Kadonaga JT 2001 The many faces of chromatin remodeling: SWItching beyond transcription. Cell 106:523–525. 190. Narlikar GJ, Fan HY, Kingston RE 2002 Cooperation between complexes that regulate chromatin structure and transcription. Cell 108:475–487. 191. Chakravarti D, LaMorte VJ, Nelson MC, Nakajima T, Schulman IG, Juguilon H, Montminy M, Evans RM 1996

J. WESLEY PIKE AND NIRUPAMA K. SHEVDE

192. 193. 194.

195.

196.

197.

198.

199. 200.

201.

202.

203.

204.

205.

206.

Role of CBP/P300 in nuclear receptor signalling. Nature 383:99–103. Ogryzko VV, Schiltz RL, Russanova V, Howard BH, Nakatani Y 1996 The transcriptional coactivators p300 and CBP are histone acetyltransferases. Cell 87:953–959. Imhof A, Yang XJ, Ogryzko VV, Nakatani Y, Wolffe AP, Ge H 1997 Acetylation of general transcription factors by histone acetyltransferases. Curr Biol 7:689–692. Spencer TE, Jenster G, Burcin MM, Allis CD, Zhou J, Mizzen CA, McKenna NJ, Onate SA, Tsai SY, Tsai MJ, O’Malley BW 1997 Steroid receptor coactivator-1 is a histone acetyltransferase. Nature 389:194–198. Jiang YW, Veschambre P, Erdjument-Bromage H, Tempst P, Conaway JW, Conaway RC, Kornberg RD 1998 Mammalian mediator of transcriptional regulation and its possible role as an end-point of signal transduction pathways. Proc Natl Acad Sci USA 95:8538–8543. Rachez C, Lemon BD, Suldan Z, Bromleigh V, Gamble M, Naar AM, Erdjument-Bromage H, Tempst P, Freedman LP 1999 Ligand-dependent transcription activation by nuclear receptors requires the DRIP complex. Nature 398:824–828. Masuyama H, Brownfield CM, St-Arnaud R, MacDonald PN 1997 Evidence for ligand-dependent intramolecular folding of the AF-2 domain in vitamin D receptor-activated transcription and coactivator interaction. Mol Endocrinol 11:1507–1517. Kraichely DM, Collins JJ 3rd, DeLisle RK, MacDonald PN 1999 The autonomous transactivation domain in helix H3 of the vitamin D receptor is required for transactivation and coactivator interaction. J Biol Chem 274:14352–14358. Leo C, Li H, Chen JD 2000 Differential mechanisms of nuclear receptor regulation by receptor-associated coactivator 3. J Biol Chem 275:5976–5982. Issa LL, Leong GM, Barry JB, Sutherland RL, Eisman JA 2001 Glucocorticoid receptor-interacting protein-1 and receptor-associated coactivator-3 differentially interact with the vitamin D receptor (VDR) and regulate VDR-retinoid X receptor transcriptional cross-talk. Endocrinology 142:1606–1615. Baudino TA, Kraichely DM, Jefcoat SC Jr, Winchester SK, Partridge NC, MacDonald PN 1998 Isolation and characterization of a novel coactivator protein, NCoA-62, involved in vitamin D–mediated transcription. J Biol Chem 273: 16434–16441. Kitagawa H, Fujiki R, Yoshimura K, Mezaki Y, Uematsu Y, Matsui D, Ogawa S, Unno K, Okubo M, Tokita A, Nakagawa T, Ito T, Ishimi Y, Nagasawa H, Matsumoto T, Yanagisawa J, Kato S, 2003 The chromatin-remodeling complex WINAC targets a nuclear receptor to promoters and is impaired in Williams syndrome. Cell 113:905–917. Polly P, Herdick M, Moehren U, Baniahmad A, Heinzel T, Carlberg C 2000 VDR-Alien: a novel, DNA-selective vitamin D3 receptor-corepressor partnership. FASEB J 14:1455–1463. Hsieh JC, Sisk JM, Jurutka PW, Haussler CA, Slater SA, Haussler MR, Thompson CC 2003 Physical and functional interaction between the vitamin D receptor and hairless corepressor, two proteins required for hair cycling. J Biol Chem 278:38665–38674. Masuyama H, MacDonald PN 1998 Proteasome-mediated degradation of the vitamin D receptor (VDR) and a putative role for SUG1 interaction with the AF-2 domain of VDR. J Cell Biochem 71:429–440. Perissi V, Staszewski LM, McInerney EM, Kurokawa R, Krones A, Rose DW, Lambert MH, Milburn MV, Glass CK,

CHAPTER 11 The Vitamin D Receptor

207.

208.

209.

210.

211. 212.

Rosenfeld MG 1999 Molecular determinants of nuclear receptor-corepressor interaction. Genes Dev 13:3198–3208. Tagami T, Lutz WH, Kumar R, Jameson JL 1998 The interaction of the vitamin D receptor with nuclear receptor corepressors and coactivators. Biochem Biophys Res Commun 253:358–363. Blanco JCG, Wang I-M, Tsai SY, Tsai M-J, O’Malley BW, Jurutka PW, Haussler MR, Ozato K 1995 Transcription factor TFIIB and the vitamin D receptor cooperatively active ligand-dependent transcription. Proc Natl Acad Sci USA 92:1535–1539. Thompson PD, Remus LS, Hsieh JC, Jurutka PW, Whitfield GK, Galligan MA, Encinas Dominguez C, Haussler CA, Haussler MR 2001 Distinct retinoid X receptor activation function-2 residues mediate transactivation in homodimeric and vitamin D receptor heterodimeric contexts. J Mol Endocrinol 27:211–227. Bettoun DJ, Burris TP, Houck KA, Buck DW 2nd, Stayrook KR, Khalifa B, Lu J, Chin WW, Nagpal S 2003 Retinoid X receptor is a nonsilent major contributor to vitamin D receptor-mediated transcriptional activation. Mol Endocrinol 17:2320–2328. Shang Y, Hu X, DiRenzo J, Lazar MA, Brown M 2000 Cofactor dynamics and sufficiency in estrogen receptorregulated transcription. Cell 103:843–852. Reid G, Hubner MR, Metivier R, Brand H, Denger S, Manu D, Beaudouin J, Ellenberg J, Gannon F 2003 Cyclic, proteasomemediated turnover of unliganded and liganded ERalpha on responsive promoters is an integral feature of estrogen signaling. Mol Cell 11:695–707.

191 213. Shang Y, Myers M, Brown M 2003 Formation of the androgen receptor transcription complex. Mol Cell 9:601–610. 214. Burakov D, Crofts LA, Chang CP, Freedman LP 2002 Reciprocal recruitment of DRIP/mediator and p160 coactivator complexes in vivo by estrogen receptor. J Biol Chem 277:14359–14362. 215. Szpirer S, Szpirer C, Riviere M, Levan G, Maryen P, Cassiman JJ, Wiese R, DeLuca HF 1991 The SP1 transcription factor gene (SP1) and the 1,25-dihydroxyvitamin D receptor gene (VDR) are colocalized on human chromosomal arm 12q and rat chromosome 7. Genomics 11: 168–173. 216. Miyamoto K, Kesterson RA, Yamamoto H, Taketani Y, Nishiwaki E, Tatsumi S, Inoue Y, Morita K, Takeda E, Pike JW 1997 Structural organization of the human vitamin D receptor chromosomal gene and its promoter. Mol Endocrinol 11:1165–1179. 217. Jehan F, DeLuca HF 2000 The mouse vitamin D receptor is mainly expressed through an Sp1-driven promoter in vivo. Arch Biochem Biophys 377:273–283. 218. Crofts LA, Hancock MS, Morrison NA, Eisman JA 1998 Multiple promoters direct the tissue-specific expression of novel N-terminal variant human vitamin D receptor gene transcripts. Proc Natl Acad Sci USA 95:10529–10534. 219. Sunn KL, Cock TA, Crofts LA, Eisman JA, Gardiner EM 2001 Novel N-terminal variant of human VDR. Mol Endocrinol 15:1599–1609. 220. Morrison NA, Qi JC, Tokita A, Kelly PJ, Crofts L, Nguyen TV, Sambrook PN, Eisman JA 1994 Prediction of bone density from vitamin D receptor alleles. Nature 367:284–287.

CHAPTER 12

Vitamin D Receptor Promoter and Regulation of Receptor Expression LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER Bone and Mineral Research Program, Garvan Institute of Medical Research, Sydney, Australia

I. Introduction II. Tissue Distribution III. The Vitamin D Receptor Gene Locus

IV. Regulation of Vitamin D Receptor Expression and Abundance V. Concluding Remarks References

I. INTRODUCTION

Current understanding of this regulation is often confounded by inconsistencies between findings in different experimental systems, which may relate to the particular species, tissue, or cell type represented by the model under study. This chapter therefore describes current understanding of the complexities of VDR gene regulation, with the goal of providing a foundation for interpretation of recent findings and a basis for further exploration.

The cellular and physiological actions of 1,25dihydroxyvitamin D [1,25(OH)2D3] are mediated primarily through the nuclear vitamin D receptor (VDR) as described in Chapter 11, although there is in vitro evidence for rapid actions of the hormone that appear to be mediated by a distinct nonnuclear receptor, as described in Chapter 23. Alteration in circulating levels of 1,25(OH)2D3 is a primary mechanism by which receptorexpressing cells are subjected to physiological hormone responses, as described in Chapters 46, 49, 65, and 66. Another important regulatory mechanism, however, is variation in cellular content of functional receptor, by alteration in VDR abundance or by post-translational modification. Such regulation allows multiple target cells to have graded responses to 1,25(OH)2D3 in a given hormonal environment, depending on the VDR levels. This chapter addresses the transcriptional mechanisms by which VDR abundance is regulated. Regulation of receptor function by post-translational mechanisms is discussed in Chapter 13. The nuclear VDR gene structure is complex, with several functioning promoters and alternatively spliced 5′ coding and noncoding exons in some species. There are consensus transcription factor binding elements and response motifs in the promoter regions. In some cases, there is experimental evidence that these motifs are functional and respond to known regulators of VDR transcription. However, there are also consensus element matches that suggest previously unsuspected regulatory inputs and await experimental investigation. Moreover, in some cases the promoter regions lack consensus response elements relevant to signal transduction pathways that are known to regulate VDR expression, suggesting indirect mechanisms of regulation. Study of the transcriptional control of the VDR gene therefore provides both expected and unexpected insights into the physiological regulation and function of this receptor. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

II. TISSUE DISTRIBUTION Specific receptors for 1,25(OH)2D3 were initially identified in the intestine of the chick by sucrose density gradient separation and assays of binding to [3H]-1,25(OH)2D3 [1,2]. Such receptors were subsequently identified in rat and human intestine, rat and chick fetal calvaria, chick kidney and pancreas, and rat skin [3–7], with sedimentation coefficients of 3.2–3.7 S. This experimental approach failed to detect receptors in brain, lung, thymus, adipose tissue, skeletal muscle, uterus, and myocardium, leading to the initial impression that there was a selective organ distribution. In addition to these studies, which detected VDR primarily in calciotropic tissues, early reports also indicated the presence of a receptor with physicochemical properties similar to those of chicken intestine receptor in the breast cancer cell line MCF-7 [8] and in a number (23 of 33) of cultured human cancer cell lines [9], including seven of 10 malignant melanoma lines and eight of nine colonic carcinoma lines. In the 1980s monoclonal antibodies generated against VDR protein isolated from chicken and pig intestine, or against synthetic peptides from the human VDR [10–12], were used to detect VDR in protein extracts from a variety of tissues and cell lines, and to clone the chicken VDR cDNA [13,14]. Copyright © 2005, Elsevier, Inc. All rights reserved.

194 A monoclonal antibody to the chick intestinal receptor detected nuclear VDR in histological sections of normal human epithelial tissues, including liver, kidney, thyroid, adrenal, gastrointestinal tract, breast, and skin [15]. A quantitative immunoradiometric assay (IRMA) indicated the presence of VDR protein in rat tissues [16], with the highest level in proximal small intestine and colon, moderate levels in ileum and kidney, and lower amounts in tissues of the vitamin D endocrine system, e.g., bone, thyroid/parathyroid, and skin. The IRMA proved more sensitive than the earlier sedimentation and radioligand approach mentioned above, revealing even lower amounts of VDR protein in the nonclassical vitamin D target tissues lung, heart, stomach, spleen, and liver, but not in cerebrum, cerebellum, or skeletal muscle [16]. More recent studies have detected VDR in pituitary, activated B and T lymphocytes, ovary, and testes, as well as in brain [17]. The availability of even more sensitive techniques such as reverse-transcription polymerase chain reaction (RT-PCR) has made it clear that VDR is expressed even in liver, which previously was considered to have very low or even no VDR [18]. VDR mRNA and protein were very low in hepatocytes, but Kupffer, stellate, and endothelial cells expressed functional VDR that mediated a positive CYP24 promoter response to 1,25(OH)2D3, suggesting that selective hepatic cell populations are also targets for the vitamin D endocrine system [18]. Based on these findings, it appears that there are more than 50 vitamin D target tissues, with the actions of 1,25(OH)2D3 extending far beyond its role in calcium homeostasis. The levels of VDR vary greatly between tissues and even between different cell types within a given tissue. Regulation of VDR expression therefore plays a fundamental role in determining physiological vitamin D responsiveness.

III. THE VITAMIN D RECEPTOR GENE LOCUS The many biological actions of vitamin D and its derivatives and the varied distribution of its receptor indicate complex regulation, which includes precise control of VDR and its abundance in a tissue- and species-specific manner. Part one of this section is focussed on the structure of the 5′ end of the VDR gene in chicken and mammals, including human, mouse, and rat, with detailed descriptions of VDR N-terminal variants that may provide functional diversity to the vitamin D endocrine system. The second part of this section presents evidence for the complex regulation of the VDR through its promoter structure and the presence of numerous transcription factor response elements.

LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER

The third part briefly reviews evidence related to the effects of common DNA sequence polymorphisms on VDR expression.

A. VDR Gene Structure 1. STRUCTURE OF THE VDR LOCUS

VDR is widely but not ubiquitously distributed. Its transcription has been analyzed primarily in mammalian and avian species, although it is present in all vertebrates and to date cDNAs have been isolated from more than 18 species including fish, amphibians, and reptiles. The human and rat mRNAs are 4.4 kb [19,20], in contrast with the approximately 2.7 kb avian [21,22] and the two Xenopus laevis (2.2 and 1.8 kb) transcripts [23]. However the sizes of the encoded receptor proteins are somewhat comparable across species, at 422 amino acid residues in Xenopus, 424 residues in rat, 427 and 477 residues in human, and 437 and 451 residues in chicken. There are eight coding exons (exons 2 to 9 according to human VDR notation [24] and a variable number of untranslated 5′ exons. The functional start codon of the mouse locus is in exon 3 [25], which is equivalent to human and rat exon 2 and chicken exon 4 (Fig. 1). Notably, in the chicken locus, translation is initiated not in exon 4, but instead in exon 3 at one of two initiation codons, giving rise to two isoforms that differ by 14 amino acids [22]. There are two known untranslated exons upstream of the first coding exons in chicken, mouse, and rat, which are highly conserved at the nucleotide level among mammalian but not chicken loci. The human VDR locus is more complex, with six transcribed exons upstream of the initiation codon in exon 2. Alternative promoter usage and alternative splicing of exons 1A, 1B, and 1C [24] and 1D, 1E, and 1F [27] give rise to 14 different human VDR transcripts [27]. Exons 1A and 1C are equivalent to mouse exons 1 and 2 [25]. Exon 1D is of particular interest, as it contains a translation initiation codon that is in frame with the conventional human VDR open reading frame [27]. An alternatively spliced transcript has been detected that includes exons 1D and 1C and encodes an N-terminally extended receptor isoform, VDRB1, which is present in human tissues and cell lines [28] (discussed further hereafter). A second exon 1Dcontaining transcript lacking exon 1C and encoding a shorter isoform VDRB2 has been detected, but to date the existence of this putative receptor has not been confirmed. Exon 1D–containing transcripts have not been reported in other species, although the DNA sequence is evolutionarily conserved in mouse and rat [29], with

195

CHAPTER 12 VDR Promoter and Abundance

ATG

ATG hVDRB1

1A 1F

1E

1D

1B

1C

2

Human hVDR

ATG 1A

1D-like

1C

2

Rat

ATG 1

1D-like

2

3

Mouse

ATG 1

2

3

4

Chicken

FIGURE 1

Structure of the 5′ flanking region of mammalian and chicken VDR gene loci. Arrows indicate exons with the initiation codon; two start codons are present in chicken exon 3. Alternative splicing produces a number of transcripts in the human VDR; two of them, which give rise to the original VDR and the VDRB1 isoform, are shown. Dotted line on the chicken locus indicates sequence not yet determined.

an overall homology of 75% with the human sequence. Chicken exon 2 has low homology with human exon 1D (50%) and its position relative to exon 1 is not conserved compared to the relative positions of human exons 1A and 1D (Fig. 1). Primer extension analysis of mouse VDR transcripts did give rise to products with a different sequence 5′ to exon 2, perhaps originating from an upstream promoter or a differentially spliced transcript [30]. To date, there is no evidence from homology searches of the mouse and rat genome databases that the remaining human exons (1B, 1E, and 1F) are present in rodents. However, they do appear to be present in other primates, as genomic sequence from chimpanzee (Pan troglodytes) includes all the exons with a 98–100% identity to human sequences [29]. The evolutionary conservation of these exons and surrounding sequences suggests an important biological role for the N-terminal extension (whether for modulation of VDR expression or function). Future studies on other mammals may reveal whether these isoforms are a general phenomenon or limited to species evolutionarily closer to humans. 2. N-TERMINAL NUCLEAR RECEPTOR VARIANTS

The VDR is a member of the nuclear receptor (NR) superfamily of ligand-inducible transcription factors. It has a modular structure typical of the NRs, including a DNA binding domain (DBD) and a C-terminal ligand binding domain (LBD), linked by a flexible hinge

(for details see Chapter 11). There is a less wellconserved A/B domain at the N terminus, varying in length from 23 amino acids in the VDR to more than 600 amino acids in the mineralocorticoid receptor [31]. There is little or no A/B region homology among the NRs, in part due to the existence of N-terminal variants or some receptors. These isoforms can arise from separate genes or from a single gene by alternative splicing, usage of different promoters, and/or distinct translational start sites [32–38]. Such variants contribute to functional diversity. As an example, the two human progesterone receptor isoforms differ in promoter specificity [39] and ligand responses [40]. There are also functional differences, as the progesterone A receptors can dominantly inhibit the B receptors [41,42] and other steroid receptor superfamily members [43]. Such distinct functional characteristics of N-terminal variant nuclear receptors may contribute to modulation of different physiological responses. 3. VITAMIN D RECEPTOR VARIANTS

The broad range of physiological vitamin D responses indicates functional diversity, and although there is only one VDR locus in each species [on human chromosome (Chr) 12, mouse Chr 15, and rat Chr 7], there is some evidence for VDR isoforms. In flounder (Paralichthys olivaceus), two VDR subtypes originate from different transcripts, with nucleotide and predicted amino acid sequence variations distributed

196

LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER

sometimes termed VDRA [28,29]. These two human VDR isoforms have been detected together at similar levels in human kidney tissue and in kidney, intestine, and bone cell lines [28], with the identity of the larger protein confirmed using a VDRB1-specific antiserum. In addition, the VDRB1-encoding transcript was also detected in parathyroid adenoma tissue and in skin and mammary gland cell lines [27]. VDRB1 exhibits differential transactivation activity [29] that appears to be due, at least in part, to a strong ligand-independent transcriptional activation domain in the A/B region [50]. The differential accumulation of VDRB1 into nuclear speckles may also contribute to transactivation differences [28] although a mechanism for such an effect has not been established. The distinct transactivation properties of VDRB1 would be consistent with a role in diversification of the vitamin D physiological response. The relative abundance of VDR and VDRB1 in each tissue or cell type may therefore be regulated. The complexity of the 5′ flanking region of the human VDR locus suggests mechanisms that may provide such regulation (discussed in Section III,A,1 and later).

throughout the receptor [44]. Flounder VDRa shows a more limited tissue distribution than VDRb and, to date, no functional differences between the two have been reported. VDRs from chicken and quail exhibit two isoforms (A and B) that differ by 14 amino acids and are translated from a single mRNA species. The chicken VDR gene contains 3 in-frame ATG sites for translation initiation, with the third in a position in exon 4 equivalent to the mammalian start codon site (Fig. 1). The first ATG is in a suboptimal context (TCCATGT), whereas the second is a better fit for the optimal consensus sequence (AGCATGG) [22,45]. Mutational analysis indicates that the two translated avian VDR isoforms arise from the first and second ATG sites by a “leaky scanning” mechanism [22], but no differential functions have been identified for these two chicken receptors. Similarly, the physiological significance of a dominant-negative C-terminally truncated rat VDR resulting from aberrant splicing leading to retention of intron 8 is not known [46]. Variants of human VDR have been identified. A common start codon polymorphism that produces a VDR with a three-amino-acid N-terminal truncation [47] has been associated with elevated transactivation activity [48,49]. A human VDR isoform with a longer N-terminal extension is generated from an exon 1D-containing transcript, as described earlier (See Section III,A,1) [27,28]. Other than its 50-amino-acid N-terminal extension, the VDRB1 protein is identical to the originally described VDR [19] (Fig. 2), which is

A /B

DBD

B. VDR Promoter To analyze the regulation of the VDR gene, its 5′ flanking region has been cloned and its promoter activity characterized in mouse [25,30], human [24,27,51], and chicken [52]. In these species the flanking sequences

Hinge

LBD

AF-2

427 aa

50

hVDR

427 aa

ATG

hVDR B1

ATG

TGA

Fok I 1A 1D

1B

1C

2

3

4 5 6

7 8

9

FIGURE 2 Structure of the human VDR isoforms. Receptor protein domains are shown on top, exon structure and alternative 5′ splicing patterns below. Arrows indicate the start codons for VDR (exon 2) and VDRB1 isoform (exon 1D). VDRB1 differs from VDR only in its N-terminal extension of 50 amino acids. *FokI polymorphism gives rise to a protein of 424 amino acids. White boxes indicate the exons that encode the N-terminal extension of VDRB1 isoform. Striped boxes represent untranslated exons.

197

CHAPTER 12 VDR Promoter and Abundance

5′ to exon 1 (mouse and chicken) or exon 1A (humans) (see Fig. 1) are highly conserved [52]. Their properties are the focus of the following sections. 1. NONHUMAN VDR PROMOTERS

The sequence of the mouse VDR 5′ flanking DNA contains characteristics of a GC-rich promoter including four tandem Sp1 sites and two upstream CCAAT boxes in opposite orientation (Fig. 3). Such TATA-less promoters are not uncommon among the steroid/ thyroid hormone receptor superfamily [53]. A 1.5-kb segment of the mouse 5′-flanking region activated transcription of a reporter gene in rat osteosarcoma ROS 17/2.8 and mouse fibroblastic NIH3T3 cell lines [25]. Deletion analysis indicated that the sequence from 54 to 177 bp upstream of the transcription start site provides most of the promoter activity and a region 177 to 481 bp upstream appears to have a repression function [30]. The first, second, and fourth Sp1 sites between 56 and 112 bp upstream of the transcription start site are necessary for mVDR promoter activity, whereas the Sp1–3 site appears to be inhibitory [30]. The Sp1 site proximal to exon 1 (Sp1-1) (position 1435, Fig. 3) overlaps a perfect consensus site for two closely related zinc finger transcription factors, Krox-20 and Krox-24 [54]. Both factors were originally noted for their importance in the nervous system [55,56]; however, Krox20−/− mice exhibit abnormal bone formation with a defect in endochondral ossification [57]. This phenotype is consistent with Krox20 expression in a subpopulation of growth plate hypertrophic chondrocytes and in differentiating osteoblasts, two cell types that are known to be responsive to vitamin D and to express the VDR (see Chapters 32 and 33). Thus, the effect of the Krox20 deletion on bone formation may relate to a defect in VDR expression in cells normally responsive to 1,25(OH)2D3. To date, however, VDR levels in these mice and the function of the Krox binding site in the mouse VDR promoter have not been tested. The mouse VDR promoter contains an AP-1 site (position 1123, Fig. 3) that differs by one nucleotide from one that overlaps the VDRE in the human osteocalcin promoter [58]. AP-1 is a target for the stressactivated protein kinases p38 and JNK, two major MAPK (mitogen-activated protein kinase) pathways. The p38 and JNK pathways and the c-Jun component of AP-1 were required for VDR stimulation by stress activation, which further sensitized human breast cancer MCF-7 cells to vitamin D3-induced growth inhibition [59]. There is also an imperfect AP-1 site in the mouse VDR promoter (position 1281, Fig. 3), similar to a binding site for the osteoblast-specific transcription factor NMP-2 (nuclear matrix protein 2) in the rat osteocalcin promoter [60]. Although these similarities

are suggestive, its role in regulating mouse VDR transcription is not known. No consensus vitamin D-, cAMP-, glucocorticoid-, or estrogen-responsive element was identified in the mouse VDR promoter 1.5-kb region by computer analysis, although there are a putative cyclic AMP-response element binding protein/Ets transcription factor (CREB/ETF) binding site at position 944 and an AP-2 site at position 1043 (Fig. 3) [25]. Treatment with forskolin, a cyclic AMP agonist known to increase endogenous VDR protein and mRNA content (discussed further in Section IV,B,4,1) [61], resulted in a three to fivefold increase in promoter-reporter activity, although the CREB/ETF site itself was not tested for function [25]. Further analysis of the mouse promoter demonstrated the presence of two E-boxes (at positions 1034 and 1129, Fig. 3), both with the consensus sequence ‘CACCTG’ target for the ZEB transcription factor, and both mediating up-regulated VDR expression in vitro [62]. In this region there are also five potential binding sites for the transcription factor c-MYB that appear to function additively with ZEB [62]. In this context it is notable that both ZEB and VDR knockout mice exhibit growth retardation and skeletal abnormalities [63,64]. A WT1, Wilms’ tumor gene, responsive element (WRE) overlapping the Sp1–4 site (position 1387, Fig. 3) is functional and able to mediate WT1-enhanced VDR promoter activity in transient transactivation assays, consistent with elevation of endogenous VDR upon overexpression of WT1 in colon carcinoma and kidney cell lines [65,66]. The rat VDR promoter has not yet been experimentally examined, but high DNA sequence homology (over 85%) with the mouse promoter suggests it is likely to function similarly (Fig. 3). Three of the functional Sp1 sites of the mouse promoter are perfectly conserved in the rat 5′ region whereas the fourth, Sp1–2, is divergent at two nucleotides. The rat Krox20/24 site differs from the mouse sequence by one nucleotide. The chicken VDR promoter is also structurally similar to mouse, lacking a TATA box and possessing several GC-rich Sp1 sites. Deletion studies indicated that the proximal Sp1 sites are strong elements to drive gene expression, with upstream promoter sequences having a negative effect on gene expression [52]. This promoter also has an AP2 site, and a Krox20/24 site outside the proximal promoter region [52]. Thus far, the regulatory contributions of these sites to chicken VDR expression have not been elucidated. 2. HUMAN VDR PROMOTERS

Regulation of the expression of human VDR appears to be more complex than the rodent and avian VDR with six untranslated exons 5′ to the exon that

198

LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER

FIGURE 3 Alignment of the promoter regions of mouse, rat, and human VDR. The first exon is underlined (exon 1 of mouse and exon 1A of human and rat). Bent arrows indicate major transcriptional start sites. Superscript stars on shaded nucleotides indicate the beginning of the published cDNAs [19,20,26]. Additional sequence in the first exon was determined: for mVDR (mouse) by primer extension [30] and 5′ RACE (Esteban et al., unpublished); for rVDR (rat) by 5′ RACE (Esteban et al., unpublished); for hVDR (human) by 5′ RACE [24]. In the 600 bp upstream of the first exon there is 88% identity between mouse and rat and about 63% between rodents and human sequence. Alignment with chicken VDR promoter is not included because its identity with mammals is only about 40%, although the GC-rich region containing the Sp1 sites is highly conserved [30]. Specific transcription factor responsive elements are indicated only for mouse VDR which has been more extensively studied. Other sites that are also present in the human and chicken sequence [24,52] and in the promoter of human exon 1C [51] are discussed in the text. Alignment is based on mouse VDR promoter coordinates (gb#AF017779, 1574 bp) and the homologous region of rVDR promoter from gb#AC119476 and hVDR promoter from gb#AY342401 (nt 235-2019). Asterisks indicate sequence identity in all three promoters.

199

CHAPTER 12 VDR Promoter and Abundance

E-Box c-MYB

Sp1 1A

WRE

Forskolin, ATRA, 17β-E2, Phytoestrogens, Dex ATG

ATG hVDR

1F

1E 1A1D

2

1B 1C

Cdx-2

ATRA

FIGURE 4 Regulatory regions of human VDR locus. White circles indicate regions with functional promoter activity, bent arrows on exons 1A, 1C, 1D, and 1F are transcriptional start sites. Two major initiation codons on exons 1D (for VDRB1) and exon 2 (for VDR) are indicated by ATG above downward arrows. Experimentally confirmed transcription factor response elements are: WRE (WT1 responsive element), Sp1 transcription factor, Cdx-2 (caudal-related homeodomain). Also two E-boxes (ZEB responsive element) and several c-MYB responsive elements present in the promoter region 5′ to exon 1A are functional in mouse [62], but have not been tested in the human sequence. Other untested consensus response elements for other transcription factors [24] are not shown. Forskolin, ATRA (all-trans-retinoic acid), 17β-E2 (17β-estradiol), phytoestrogens, and Dex (dexamethasone) induce the promoter 5′ to exon 1C, but response elements have not been defined.

encodes the originally described initiation codon, exon 2 (Fig. 4) [24,27]. Up to 14 variant transcripts assembled from multiple combinations of the six upstream exons were initially identified in human kidney [27]. Their expression is regulated by up to three distinct promoters: a promoter upstream of exon 1A and equivalent to those described for rodents and chicken [24,27], a second promoter upstream of exon 1F [27], and a third promoter upstream of exon 1C [51]. The proximal 5′-flanking region upstream of exon 1A of the human VDR gene is highly conserved with chicken and mouse loci, both in its lack of a TATA box initiator and the presence of a GC-rich region with putative Sp1 and transcription factor binding sites (Fig. 4). As in mouse, proximal elements from 103 to 464 bp upstream of the transcription start site displayed high promoter activity, whereas elements further upstream repressed transcription [24]. Also comparable to the mouse, deletion of the Sp1 sites proximal to the start of the mRNA resulted in a 10-fold drop in promoter activity, indicating an important role in promoter function. This same promoter may also drive the expression of exon 1D transcripts, as available evidence indicates that upstream sequences within 300 bp of exon 1D lack promoter activity and may, in fact, contain a suppressor element [27]. Further studies are required to clarify whether the exon 1D promoter does overlap the 1A promoter and/or transcribed exon sequence. A few transcription factor binding sites have been identified in the human 1A promoter region. At position –3731 to –3720 bp relative to the exon 1A transcription start site, there is a polymorphic binding site for

the caudal-related homeodomain protein Cdx-2 (see Section III,C), which activates VDR gene transcription by binding to this element [67,68]. As it is an intestinespecific transcription factor, Cdx-2 may contribute to tissue-specific regulation of the VDR. As in the mouse VDR promoter, there are two E-box targets for the ZEB transcription factor and potential binding sites for the transcription factor c-MYB [62]. There are also several WREs: one overlapping the most distal Sp1 site [66] and three more putative distal sites that are able to bind WT1 protein [69]. The only functional WRE of the latter three, 308 bp upstream of the exon 1A transcription start site (Figs. 3 and 4), appears to be distinct from the WT1-response element defined for the mouse promoter. A distal promoter of the human VDR gene, more than 9 kb upstream of exons 1A and 1D, drives the expression of exon 1F–containing transcripts. Whereas exon 1A and 1D transcripts appear to be ubiquitous, the distribution of exon 1F–containing transcripts is more restricted, detected to date only in human kidney, parathyroid tissue, and the intestinal carcinoma cell line LIM 1863 [27]. As these sources represent major calciotropic target tissues, the distal promoter may require tissue- or cell-specific regulatory factors to mediate the restricted gene expression pattern. Interestingly, although not translated, exon 1F–containing transcripts are underexpressed in parathyroid adenomas and in hyperplastic glands of secondary hyperparathyroidism in comparison with normal parathyroid glands [70]. In addition to those initiating at exons 1F, 1A, and 1D, human VDR transcripts that initiate at exon 1C

200

LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER

have been detected in MCF-7 breast cancer cells [51]. These transcripts appear to arise from a third human VDR promoter upstream of exon 1C, as evidenced by the promoter activity of a 1.3 kb 5′ flanking segment in reporter gene assays [51]. Characteristic promoter features in this segment include GC-rich regions and a TATA box at –29 bp, as well as putative AP-2 and Sp1 transcription factor binding sites and several consensus glucocorticoid response elements (GREs). The GREs appear to be regulated according to cell type, as dexamethasone induced promoter activity in the SUM159PT but not in the MCF-7 breast cancer cell line [51]; however, specific functional GRE(s) in this region have not been defined. There are no consensus vitamin D (VDRE), retinoic acid (RARE), or estrogen (ERE) responsive elements in the functional 1.3-kb exon 1C promoter fragment, nor is there a clear forskolin responsive element [51]. This is somewhat surprising, as transcriptional regulation by this region is enhanced in the estrogen receptor (ER)-positive MCF-7 cell line by treatment with forskolin, retinoic acid, and 17β-estradiol. The latter response appears to be mediated by the estrogen receptor, as the response to estradiol is absent in the ER-negative SUM159PT cell line [51]. This promoter is also responsive to up-regulation by phytoestrogens in the ER positive human breast cancer cell lines T47D and MCF-7 [71]. A retinoic acid responsive region downstream of exon 1C may be relevant for regulation of endogenous VDR expression [24], and the effect of forskolin may be mediated through the AP-2 sites upstream of the human exon 1C [51], as some studies have implicated AP-2 elements in mediating cAMP responsiveness of other gene promoters [72,73]. Further studies are required to clarify the mechanisms of these 1C promoter hormone responses, which are of particular interest as there is currently no evidence to suggest that either the exon 1F or the exon 1A promoter is hormonally regulated [24,25,27,30].

1F

ATG 1A 1E 1D

Cdx-2

1B

ATG 2 1C

Fok I

C. VDR Polymorphisms Polymorphisms in the VDR locus have been studied in relation to skeletal biology, renal metabolic and immune functions, and cancer risk and responsiveness in a variety of populations worldwide. The most commonly studied polymorphisms and a more recently described polymorphism in a transcription factor binding site are briefly discussed here. A polymorphism has been described in the Cdx-2 binding site of the human promoter (see Section III,B,2), which resulted in lower promoter activity (70% of control transcription level) and impaired binding to Cdx-2 protein (Fig. 5). This single nucleotide polymorphism may have significant impact on target-cell responsiveness to 1,25(OH)2D3 by affecting VDR abundance. It has been associated with lower bone mineral density (BMD) at the lumbar spine in postmenopausal Japanese women [68]. Presence of a polymorphic FokI restriction site in exon 2 (the ‘f’ allele) actually affects receptor structure, allowing production of a normal-length human VDR (427 aa), whereas absence of the site (the ‘F’ allele) produces an N-terminally shortened 424 aa protein [47]. The ‘F’ allele has been associated with elevated transactivation activity rather than differences in VDR expression per se [48,49,74]. Specific FokI alleles have been correlated with differences in bone density, fracture incidence, nutritional rickets, risk of osteosarcoma, auto-immune hepatitis, urolithiasis, insulin sensitivity and drug clearance rates [75–81]. Restriction site polymorphisms for the enzymes BsmI and ApaI have been described in the last intron of the human VDR gene and a silent TaqI site in exon 9 [82,83] as well as a polyadenosine (A) microsatellite within the 3′ untranslated region (3′ UTR) [84]. These polymorphisms have been associated with differences in BMD in a number of genetic and population based studies [85–90], but not all have observed such associations [91,92]. Although unlikely to lead to changes in

TGA 3

45

6

78

Bsm I Apa I

9

Poly(A) Taq I microsatellite

FIGURE 5 Polymorphisms in human VDR gene. Single-nucleotide polymorphism has been identified in the Cdx-2 site, restriction site polymorphisms in exon 2, intron 8, and exon 9, and poly(A) microsatellite in 3′ UTR. Only the FokI polymorphism affects protein structure. The BsmI, ApaI, TaqI, and poly(A) polymorphisms are in linkage disequilibrium in some populations [74,84].

201

CHAPTER 12 VDR Promoter and Abundance

the receptor structure, these polymorphisms could affect mRNA accumulation and hence VDR protein level and function. A minigene representing a haplotype lacking the BsmI and ApaI sites but including a functional TaqI site (‘BBAAtt’) produced higher reporter gene activity than the ‘bbaaTT’ haplotype in transient transfection assays, suggesting that the ‘BBAAtt’ haplotype accumulated more VDR mRNA and protein [85]. Consistent with this finding, alleles ‘b’, ‘a’, and ‘T’ were associated with significantly lower VDR and higher PTH mRNA levels in patients with primary hyperparathyroidism [93]. Similarly, the poly(A) microsatellite ‘L’ (long) allele in the 3′ UTR was associated with higher VDR activity than the ‘S’ (short) allele in human fibroblast cell lines [74]. Allelespecific variation in mRNA stability or transcriptional regulation may be tissue or cell type specific, however, as the opposite pattern was observed in myeloid leukemia and prostate cancer cell lines, with a 30% reduction in RT-PCR product derived from the small ‘t’ allele [94]. Tissue specific allele effects may also explain the lack of relationship between VDR alleles and receptor density in duodenal mucosa of healthy premenopausal women, despite an association with bone mineral content (BMC) [95], and also the lack of association between VDR genotype and protein concentration in duodenum [96], peripheral blood mononuclear cells [97], and cultured skin fibroblasts [98]. Thus, although there appears to be no association between the BsmI, ApaI, and TaqI polymorphisms and VDR levels in many tissues and cell types studied, genotype may affect receptor levels in some tissues or under certain pathological conditions. Recent studies have indicated genetic interactions between various VDR alleles, with combined haplotypes sometimes regulating polymorphism effects that were not detected in single polymorphism analysis [74].

IV. REGULATION OF VITAMIN D RECEPTOR EXPRESSION AND ABUNDANCE An important mechanism for the modulation of cellular responsiveness to 1,25(OH)2D3 is mediated by the regulation of receptor abundance. Developmental regulation occurs, with receptors for 1,25(OH)2D3 appearing in the neonatal rat intestine shortly before weaning, coincident with a switch to vitamin D–dependent calcium absorption [99]. Target cell responsiveness is also regulated, with the ability of 1,25(OH)2D3 to stimulate 24-hydroxylase and inhibit collagen synthesis in mouse osteoblastic cells proportional to its receptor

content [100,101]. Receptor abundance also varies with cellular differentiation, exemplified in a series of clonal rat osteosarcoma cell lines of increasing osteoblasticlike character (ROS 24/1, ROS 2/3, and ROS 17/2.8), in which receptor levels correlated with the magnitude of 1,25(OH)2D3-associated inhibition of growth, transition from cuboid to spindle-like morphology, and suppression of colony formation in soft agar [102]. The following section focuses on a variety of studies which have shown that complex physiological signals involved in the regulation of VDR abundance vary among species and between the various 1,25(OH)2D3 target organs. To facilitate comprehension of this complex story, main findings are summarized in Table I.

A. Homologous Regulation VDR expression is affected by the presence of its own ligand, 1,25(OH)2D3, with ligand-mediated autoregulation of VDR documented in several experimental systems. Using a chicken cDNA probe, a receptor mRNA species in mouse 3T6 fibroblasts was substantially increased by prior exposure to 1,25(OH)2D3 [14]. In the pig kidney cell line LLC-PK1 there was homologous up-regulation by 1,25(OH)2D3 and other vitamin D analogs with a substantial increase in the number of 1,25(OH)2D3 receptors without altering the affinity of receptor for the hormone. Treatment of these cells with the RNA synthesis inhibitor actinomycin D showed that the increase in receptors was in part dependent on RNA synthesis, and in part on receptor occupancy [103]. Similar results were observed in rat osteosarcoma ROS 17/2 cells, in which there was a receptor increase upon treatment with 1,25(OH)2D3, with roughly 50–60% of the increase prevented by inhibition of RNA polymerase II activity with α-amanitin [104]. Similarly, human breast cancer cells increased specific binding of [3H]1,25(OH)2D3 upon hormone treatment, which was prevented by actinomycin D treatment [105]. Thus, although pharmacological inhibitors of transcription partially inhibit VDR induction, other 1,25(OH)2D3 responsive mechanisms also appeared to be involved in homologous regulation of receptor abundance. The typical response to 1,25(OH)2D3 treatment is characterized by an ascending phase, which represents hormone association with receptor and reaches a maximum at 90–120 min, followed by a rapid descending phase that is closely associated with a decrease of medium 1,25(OH)2D3 due to the metabolism of the hormone [106]. Treatment of the osteosarcoma cell line UMR-106 with ketoconazole (an inhibitor of 1,25(OH)2D3 metabolism) resulted, after a lag-time of about

202

LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER

TABLE I Brief Summary of Regulators of VDR Abundance

1,25(OH)2D3

Calcium Phosphorus

Fatty acids Glucocorticoids

Estrogens

Retinoic acid

Growth factors Cell cycle/ differentiation Growth hormone (GH) PTH

PKA activators

PKC activators

aD 3

Effect on VDR levels

Comment

Up-regulation in calcemic tissues (kidney, intestine, parathyroid gland), and no change in other tissues Up-regulation in cell lines

In normocalcemic animals

[113,114,116–120]

Mainly due to stabilization of the VDR by D3a Hypocalcemia prevents upregulation by D3 In low-phosphate diet

[124,125]

In hypocalcemia, down-regulation in kidney and parathyroid gland, and no effect in intestine In rat and chick increase of intestinal VDR. In kidney no change Parathyroid gland had no change Modulation of differentiation and increase of VDR by butyrate Species-dependent. Increased VDR in rat and decreased in mouse Intestine up-regulation Kidney down-regulation Testis, heart, lung, no change Increase of VDR in intestine and enhancement of Ca2+ absorption No change in kidney VDR increase in osteoblastic cells Up-regulation of hVDR exon 1C prom Up-regulation of VDR is tissue- and species-dependent Up-regulation of hVDR exon 1C prom Up-regulation of VDR and stimulation of proliferation. In vivo, tissue-specific Proliferating cells, VDR high Quiescent cells, VDR low Increased Ca2+ absorption and intestinal CaBP-9K, but no change in VDR Reduction in kidney VDR increase in osteoblastic cells VDR down-regulated in ROS17/2.8 and decreased up-regulatory effect of D3 VDR up-regulated by forskolin and other PKA activators Blocking Ca2+ channels reduces cAMP induction Phorbol esters down-regulate VDR in osteoblastic cells In rat kidney and epithelial cells potentiation of D3 action VDR phosphorylation

is 1,25(OH)2D3.

References

[119,127,129,130] [112,132]

With high-phosphate diet Synergistic effect of D3 and butyrate Effect probably not VDR specific, but on general development Effects are tissue-dependent and cell type-dependent

[133] [135,136,139]

Some effects of 17β-estradiol on Ca2+ are independent of D3

[155–158,162]

ER-dependent Synergistic effect of ATRA and D3

[51] [126,152,171–174]

[141,142,144] [109,143,150–152]

By EGF, insulin, IGF-I

[24,51] [179–182]

No up-regulation by D3 in differentiated cells May be IGF-I mediated effect

[100,101,185, 187] [190–192,194]

cAMP mediated By PTH

[196,198,199] [200,201]

Via cAMP elevation

[61,182,198,203]

When activated by PTH Via PKC activation

[204] [183,205]

Via PKC activation VDR-PKC cross-regulation

[207–211,214]

CHAPTER 12 VDR Promoter and Abundance

90 min, in a sharp increase of receptor accumulation, which was blocked by coincubation with cycloheximide and actinomycin D [107]. Most of these findings were based on receptor quantification by high specific activity tritiated ligand and, therefore, non-hormone-binding or unoccupied forms of the receptor may not have been characterized. Western blots using the monoclonal antibody IVG8C11, which has high cross-reactivity with porcine and other mammalian VDR molecules [11,108], avoided this potential problem and detected up-regulation of 1,25(OH)2D3 receptors in human HL-60 promyelocytic leukemic cells line at 12 hr and down-regulation at 48 and 72 hr in the continuous presence of hormone [13]. Furthermore, 1,25(OH)2D3 stabilized VDR mRNA levels in human MG-63 osteosarcoma cells, increasing its apparent half-life by 11 hr and increasing its VDR protein content [109]. However, the effect of 1,25(OH)2D3 on VDR levels appears to be cell specific, as treatment with the hormone down-regulated VDR mRNA level in the human megakaryoblastic leukemia cell line HIMeg, but caused no change in the human osteosarcoma cell line HOS-8603 [110]. Similarly, treatment with 1,25(OH)2D3 did not alter VDR protein expression in three phenotypically distinct osteosarcoma cell lines [111]. In general, these in vitro studies indicate that in many but not all cell line models, receptor abundance is increased by transcriptional and translational mechanisms. Specificity of VDR up-regulation by 1,25(OH)2D3 has also been observed in numerous in vivo studies, with outcomes highly dependent upon the tissue examined, the method by which 1,25(OH)2D3 levels were manipulated, and perhaps also upon species. Basal expression of chicken intestinal VDR appears not to be dependent upon 1,25(OH)2D3, as VDR mRNA concentrations were similar in vitamin D–deficient and normal chick duodena [112]. In vitamin D–deficient rats injected with 1,25(OH)2D3, there was significant up-regulation of renal VDR after 5 days of treatment, again with minimal effect on duodenal VDR [113]. Duration of treatment influenced results, with shorter treatment period more likely to yield an increase in VDR mRNA levels. In rat parathyroid gland there was a maximum increase of VDR mRNA at 6 hr and only a marginal increase at 24 hr [114], whereas daily dosing for 5 days did not alter the level of VDR mRNA in either the intestine or kidney of vitamin D–deficient rats [115]. Furthermore, mRNA levels of the receptor in the intestine of vitamin D-deficient rats increased 10-fold above deficiency levels at 6 and 12 hr after an intravenous dose of 1,25(OH)2D3, returning to predosing levels at 24 hr [116]. The protein response in this study was less marked, as there was only a twofold increase of the VDR at 12 hr, as detected by IRMA, with

203 no change detected by ligand-binding assay [116]. In normocalcemic rats, the administration of vitamin D resulted in a 5-fold increase in kidney receptor concentration, whereas in hypocalcemic animals, vitamin D treatment did not change receptor levels [117,118], perhaps reflecting a homeostatic increase in plasma 1,25(OH)2D3 levels in the latter group. A consistent set of observations was reported with exogenous administration of 1,25(OH)2D3 in rats fed a normal diet, which resulted in a 1.5-fold increase in duodenal and a 3-fold increase in renal VDR content [119]. By contrast, rats fed a low-calcium diet in the same study exhibited plasma 1,25(OH)2D3 concentrations similar to those treated with exogenous 1,25(OH)2D3, but duodenal VDR content was not up-regulated and renal VDR was 20–38% lower than in rats with normal calcium intake. Thus, VDR levels in kidney and intestine may be homeostatically regulated by dietary calcium, but not through the elevation of plasma 1,25(OH)2D3 levels [119] (see Section IV,B,1,a). Yet homologous VDR up-regulation was not a universal property of 1,25(OH)2D3 target tissues, as homologous up-regulation of VDR occurred under a wide variety of experimental conditions in rat kidney but not in non-calcium homeostatic VDRcontaining tissues such as lung, heart, or testis [120]. Clearly, the mechanisms of homologous up-regulation of VDR must be complex. Many studies addressing mechanism have focused on VDR transcript levels in cell line models after 1,25(OH)2D3 treatment. In addition to those studies mentioned above, increased VDR mRNA was also observed after treatment in mouse fibroblast 3T6 cells [121], the human osteosarcoma cell line MG-63 [109], and HL-60 cells [122]. Accumulation of VDR transcripts appears to occur in part through increased transcription, but also to a substantial extent through mRNA stabilization, as evidenced by studies using RNA polymerase II inhibitors [103–105]. Studies suggestive of stabilization of the VDR protein by its ligand include radio-ligand binding studies in human breast cancer and pig LLC-PK1 cell lines [105,123], but not in the rat osteosarcoma cell line ROS 17/2 [104]. More recent studies using IRMA indicated that VDR protein levels were elevated 3-fold after 8 hr of treatment with no change in mRNA levels in 3T6 mouse fibroblasts and IEC-6 rat intestinal epithelial cells, as well as in rat intestine and kidney [124]. Notably, with inhibition of protein synthesis using cycloheximide, 1,25(OH)2D3 markedly reduced degradation of previously formed receptor. Consistent findings were reported in the rat osteosarcoma ROS 17/2.8, and in the human breast cancer T47-D cell lines, in which administration of 1,25(OH)2D3 did not change receptor mRNA level, but protein did accumulate as a result of increased receptor half-life [125,126]. Thus, in addition

204 to increasing transcription and mRNA stabilization, it appears that 1,25(OH)2D3 can increase VDR content in vivo or in cell culture by stabilizing the receptor. This effect on protein stability may be due to hormoneinduced conformational changes that directly affect receptor degradation or that enhance its functional interactions with RXR, DNA, or transcriptional co-regulators and thereby prolong receptor half-life. As ligand effects on nuclear receptor complex formation, stabilization and turnover are a very active area of investigation (described in more detail in Chapters 13 to 16), it is likely that this aspect of VDR regulation could be significantly clarified in the near future.

B. Heterologous Regulation VDR expression can be modulated by numerous physiological stimuli such as dietary composition, steroid hormones and retinoids, growth factors, peptide hormones, and second messenger activators. This section will consider the effects of these agents in regulating VDR levels to clarify the interrelationships between their activities in calcium homeostasis and cell proliferation and differentiation. 1. DIETARY COMPOSITION

A primary function of the vitamin D hormone system is control of calcium and phosphorus homeostasis, in part through 1,25(OH)2D3 regulation of its own receptor, as described in Section IV,A. As the levels of 1,25(OH)2D3 in vivo are tightly linked with the levels of dietary calcium and phosphorus (discussed in detail in Chapters 24 to 26), this section will briefly discuss mechanistic aspects of the effects of dietary calcium and phosphorus on VDR expression. Fatty acids also play a role of in the regulation of VDR levels, in part through interactions between the VDR and other nuclear receptors and transcriptional cofactors, as described in Chapter 16. Recent studies that reveal a role for dietary lipids in the regulation of VDR abundance are also briefly discussed here. a. Calcium A number of studies have shown that calcium levels have no impact on the VDR levels in the intestine, but do regulate its abundance in kidney and parathyroid gland, as under conditions of severe hypocalcemia and resulting hyperparathyroidism, little or no VDR is present in these tissues. For example, in rat kidney, in the absence of vitamin D, total VDR protein was increased 2-fold by an increase in serum calcium [117], indicative of a vitamin D–independent regulatory pathway. Also, although vitamin D increased renal VDR levels 5-fold at normal serum calcium levels, this was not the case in hypocalcemic animals [117].

LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER

Complementing these VDR protein studies, dietary calcium restriction decreased renal VDR mRNA levels in mice, even after treatment with 1,25(OH)2D3, whereas with dietary calcium supplementation, treatment with 1,25(OH)2D3 caused VDR mRNA to increase 8- to 10-fold in the vitamin D-deficient group [127]. These findings are also consistent with reduced renal VDR protein content after 1,25(OH)2D3 treatment of rats with low dietary calcium intake compared to normocalcemic animals [118,119,128,129]. Thus, hypocalcemia markedly impairs homologous up-regulation of renal VDR by 1,25(OH)2D3, suggesting interaction between distinct 1,25(OH)2D3- and calcium-responsive mechanisms in this tissue. By contrast to these renal changes, duodenal VDR was not up-regulated after 1,25(OH)2D3 treatment in rats fed a low-calcium diet [118,119,129]. These findings suggest that in rodent models, down-regulation of VDR under hypocalcemic conditions is tissue-specific: intestinal VDR is present at both low and high serum calcium levels, whereas kidney VDR is absent or markedly reduced under hypocalcemic conditions, resulting in abolition of the 1,25(OH)2D3 response. A high-calcium diet prevented a fall in VDR mRNA levels in the parathyroid gland of rats with hyperparathyroidism, as did 1,25(OH)2D3 treatment [130]; this coordinate regulation of VDR expression by calcium and 1,25(OH)2D3 in kidney and parathyroid gland was not associated with changes in intestinal VDR mRNA. There are species differences, however, as there was a significant decrease in VDR mRNA in the duodenum of chicks fed a low-calcium diet in the presence of increased serum 1,25(OH)2D3 [112]. Thus, complex regulation of VDR levels under dietary deficiencies is likely to involve other factors or hormones such as parathyroid hormone for modulation of the 1,25(OH)2D3 response. Other physiological changes affecting calcium homeostasis are pregnancy and lactation, which both up-regulate intestinal calcium absorption. However, rat intestinal VDR mRNA did not change during gestation and lactation, although intestinal VDR content increased 2-fold in late pregnancy and lactation, suggesting a posttranscriptional mechanism of VDR up-regulation [131]. In general, these studies indicate that both 1,25(OH)2D3 and calcium are required for VDR production in the kidney above the basal level, whereas calcium has only a minor effect on intestinal VDR level. Downregulation of renal VDR in response to hypocalcemia may be a protective measure to block 1,25(OH)2D3mediated suppression of the 1α-hydroxylase and induction of the 24-hydroxylase, which would result in a net increase in serum 1,25(OH)2D3 levels [129].

205

CHAPTER 12 VDR Promoter and Abundance

The mechanisms by which 1,25(OH)2D3 and calcium regulate renal VDR expression levels may be indirect or posttranscriptional, as no vitamin D response elements have been detected in either mouse or human VDR promoter, and no response to 1,25(OH)2D3 has been observed in reporter-gene analyses, suggesting that 1,25(OH)2D3 does not directly activate expression of its own receptor [24,25,27,51]. b. Phosphorus Dietary phosphorus restriction upregulates VDR in a tissue-specific and time-dependent manner [132]. A low-phosphorus diet resulted in rapid increase of serum 1,25(OH)2D3 and a 3.5-fold increase in rat intestinal VDR by day 3, as quantitated by ligand binding assay, followed by a decrease to a plateau of 2.5-fold of control levels by day 5 to 10. This finding is consistent with an increase in VDR mRNA in duodenum of chicks fed a low-phosphorus diet [112]. By contrast, a low-phosphorus diet did not up-regulate kidney or splenic monocyte/macrophage VDR in rat [132]. In another study, rats on a high-phosphorus diet did not exhibit altered levels of VDR in the parathyroid gland, although PTH mRNA levels were elevated and there was prevention of both hypocalcemia and low calcitriol levels [133]. The mechanisms by which phosphorus may regulate VDR mRNA levels and functional receptor abundance have not yet been elucidated. c. Fatty Acids Understanding of the roles of dietary fatty acids in regulation of gene expression is increasing [134] and evidence that lipids regulate VDR levels is accumulating. In rats, an ovariectomy-associated increase in intestinal VDR protein declined after longterm dietary supplementation with essential fatty acids [135]. As lipids activate transcription by the nuclear receptor peroxisome proliferator-activated receptor (PPAR), their effects on VDR levels may be indirect, via increased competition by PPAR for the VDR heterodimerization partner RXR [134]. Fatty acid regulation of protein kinases, particularly PKC, may also modulate VDR action and expression levels (see Section IV,B,4). Butyrate, a short-chain fatty acid and normal constituent of colon content, enhanced nuclear VDR protein content in Caco-2 cells in a time- and dosedependent manner; its prodrug, tributyrin, significantly increased VDR-mRNA level (2.5-fold) and VDR binding activity [136]. This response may mediate the synergistic effect of butyrate compounds with 1,25(OH)2D3 in induction of differentiation in the Caco-2 cell line [137]. Similar butyrate-induced increases have been observed in some [138,139], but not all [139] cell culture models, perhaps reflecting tissue and species specificity. To date, however, there is limited evidence for butyrate regulation of VDR abundance or 1,25(OH)2D3 function in vivo.

2. STEROIDS AND RETINOIDS

In addition to 1,25(OH)2D3, other hormones that regulate VDR abundance are steroids and retinoids. Like calcitriol, these classes of hormones also act through nuclear receptors to regulate transcription of target genes that include the VDR gene. This section will focus on glucocorticoid, estrogen, and retinoid effects on VDR abundance, as each is clinically important, particularly in skeletal health. a. Glucocorticoids In prolonged exposure to elevated glucocorticoid levels such as in Cushing’s syndrome or in glucocorticoid therapy, glucocorticoid actions appear to oppose vitamin D effects (see Chapter 73). The result is often osteopenia, with major calcium homeostatic changes in intestine and bone contributing to accelerated bone loss [140]. The present section will focus on the role of glucocorticoids in the regulation of VDR abundance as a possible basis for this apparent antagonism between glucocorticoid and vitamin D actions. In vivo treatment with the glucocorticoid hydrocortisone has been reported to increase intestinal VDR mRNA in rat if administered between 15 and 17 days postpartum, although administration before day 14 or on days 19–21 postpartum did not affect VDR expression [141,142]. The sensitive period between days 15 and 17 postpartum coincides with a glucocorticoid-sensitive period of rat intestinal development, however, and may reflect a general glucocorticoid effect on intestinal maturation rather than a specific effect on VDR mRNA, as it is accompanied by an increase in actin mRNA [142]. The effect of glucocorticoids does appear to be tissue specific, as dexamethasone injection caused VDR up-regulation in rat intestine and down-regulation in kidney, but had no effect on receptor level in testis, heart, and lung [143]. Species differences may also exist, as intestinal VDR was decreased by hydrocortisone administration in mouse [144], whereas changes in VDR were not observed in chicken intestine after treatment with the glucocorticoid compound triamcinolone [145]. There is also species variation in bone cell response. In short-term rat calvarial organ cultures glucocorticoids prevented a fall in VDR levels [146] and in rat osteosarcoma cell cultures glucocorticoids stimulated 1,25(OH)2D3 receptor levels [147,148]. By contrast, in mouse osteoblastic cells glucocorticoids caused a decrease in VDR abundance [149]. Given these interspecies differences, extrapolation from animal models to humans about glucocorticoid effects on VDR is problematic. For mechanistic studies, human cells and tissue models are particularly important, even though outcomes can vary in a single cell line. In one study using MG-63 osteosarcoma cells, dexamethasone produced a time- and dose-dependent

206 decrease in VDR mRNA [150], whereas in another study dexamethasone increased the amount of VDR mRNA, although VDR protein levels were not affected [109]. Similarly, VDR protein levels in cultured human keratinocytes and fibroblasts were not altered after treatment with corticosteroids [151]. Evidence for functional interaction comes from MCF-7 breast cancer and Caov-4 ovarian cancer cell lines, in which dexamethasone augmented calcitriol effects, although VDR levels were not assessed in this study [152]. Thus, it seems that the effects of glucocorticoids on the 1,25(OH)2D3 response in human cells are tissue-specific and may not relate to VDR levels per se in all cases. Although the effects of glucocorticoids on VDR might be mediated by direct action of the glucocorticoid receptor on the promoter of the VDR gene, no glucocorticoid responsive elements (GREs) have been found in the exon 1 promoter region (1.5 kb) of the mouse or exon 1A upstream region of the human VDR locus (discussed in Section III,B) [24,25,27]. However, in the 5′ flanking region of human exon 1C there are putative GREs [51]. Dexamethasone did not induce activity of this promoter in the estrogen-dependent MCF-7 breast cancer cell line, although there was upregulation in the estrogen-independent SUM159PT breast cancer cell line [51], consistent with cell type specificity of glucocorticoid effects on VDR abundance. The presence of multiple promoters, as in the human VDR locus, or indirect regulation via other mechanisms (e.g., as in estrogen-dependent versus estrogenindependent breast cancer cell lines, or perhaps by competition for transcription cofactors; see Chapter 16) may account for the apparent differences in the effects of glucocorticoids on VDR regulation. Furthermore, glucocorticoids may interact with other hormones that regulate VDR abundance, such as PTH (see Section IV,B,3,c). b. Estrogens Actions of the steroid hormone estrogen are mediated by the estrogen receptor (ER), another member of the nuclear receptor superfamily. Reductions in estrogen levels at the menopause are associated with decreased bone mineral density, which is thought to occur as a result of calcium homeostatic changes. One such change is a reduction in intestinal calcium absorption [153] which is improved by estrogen replacement [154]. Although 17β-estradiol (E2) may positively affect intestinal calcium absorption independent of vitamin D [155], the actions attributed to estrogens might also be mediated at least in part by regulation of 1,25(OH)2D3 or VDR levels, as 1,25(OH)2D3 is the principal hormonal regulator of intestinal calcium absorption. Animal models support this concept, as estrogenstimulated increases in VDR expression and response were observed in rat duodenum [156,157]. Other studies

LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER

have found that estrogen and phytoestrogens such as genistein increase VDR transcript level, protein expression, and endogenous 1,25(OH)2D3 bioactivity in rat colonic mucosa and colon carcinoma cell lines [158,159]. Thus, some estrogen activities in the colonic mucosa such as growth suppression could be mediated, at least in part, by increased responsiveness to endogenous 1,25(OH)2D3. However, in another study intestinal VDR levels in immature rat were unaffected by estrogen, although there were increases in VDR abundance in uterus and liver but a decrease in kidney [160,161]. Thus, conclusions about the regulation of the VDR in one tissue cannot be extrapolated to other tissues. There have been a number of studies of estrogen effects on VDR abundance in bone-derived cells because of the relevance to postmenopausal osteoporosis. VDR expression in mouse bone decreased with aging and estrogen deprivation but recovered after E2 supplementation [162]. In vitro findings are consistent, as E2 treatment doubled total cellular VDR-binding capacity in the ROS 17/2.8 rat osteosarcoma cell line [163], increased VDR protein in human osteosarcoma cell lines [164], and increased VDR mRNA levels in the human MG-63 and N-976 osteoblastic cell lines [109,162]. Correlation between VDR mRNA and protein changes was not entirely consistent [109], although E2 treatment did potentiate the antiapoptotic effect of 1,25(OH)2D3 in the latter study [162]. Interaction between vitamin D and estrogen regulation has been observed in VDR knockout mice, in which VDR appears essential for the regulation of CYP19 (aromatase cytochrome P450, a key enzyme in estrogen biosynthesis) of both female and male gonads [165]. To date, estrogen responsive elements (EREs) have not been detected in the promoter region of human, mouse, or chicken VDR genes. However, the human exon 1C promoter region (see Section III,B,2) was up-regulated by 17β-estradiol and the phytoestrogens resveratrol and genistein, despite the apparent lack of an ERE in the promoter region [51,71]. As for glucocorticoids, estradiol effects on VDR abundance vary according to species and tissue. Evidence indicates that there are significant interactions between estradiol and vitamin D responses in bone and intestine, with some renal involvement as well. It is not clear whether these findings reflect a particular experimental focus on calcium homeostatic tissues, or a special role for an estrogen–vitamin D axis in calcium regulation. Studies of this interaction in other non-calcium-homeostatic tissues should resolve this question. c. Retinoic Acid Vitamin A and its active metabolites retinol and retinoic acid, known for their involvement in vision, reproduction, embryonic development, and epithelial growth and differentiation, are also

207

CHAPTER 12 VDR Promoter and Abundance

necessary for normal bone metabolism. The actions of the retinoic acids are mediated by nuclear receptors closely related to the VDR. There are two families of nuclear retinoid receptors, the retinoic acid receptor RAR, activated by all-trans-retinoic acid (ATRA) and/or 9-cis-retinoic acid (9-cis RA), and the retinoid X receptor RXR, activated specifically by 9-cis RA [166]. In addition, RXR can act as DNA binding partner for a number of other nuclear receptors, including those for the thyroid hormone (TR), as well as the VDR and the RAR [167,168]. Vitamin A and 1,25(OH)2D3 share a number of physiological effects, including stimulation of bone resorption and effects on bone cell differentiation [169] and on keratinocyte proliferation and differentiation [170]; thus, vitamin A effects on bone and other tissues may be mediated through VDR regulation. All-trans-retinoic acid produced a dose-dependent increase in 1,25(OH)2D3 binding in rat osteosarcoma cells after 24 hr that was inhibited by actinomycin D [171]. This increase in binding was due to increased VDR protein, rather than modification of receptor affinity for the hormone. Induction of VDR by retinoic acid was greater in proliferating than nonproliferating tumorigenic cells (ROS 17/2A and UMR-106M), but not in the nontumorigenic cell line RCJ 1.20 [172]. ATRA caused a reduction in VDR abundance in rat osteoblastic cells, but an increase in receptor level in mouse osteoblasts that required protein and RNA synthesis. These opposite effects of ATRA on mouse and rat cells suggest that species-dependent factors modulate the action of retinoids in mammalian cells [173]. Findings in human osteosarcoma (MG-63) and breast cancer (T47-D) cell lines were similar to those in mouse osteoblasts with an ATRA-induced increase in VDR mRNA and its half-life although, at least in the MG-63 cells, there was no change in VDR protein levels [109,126]. Other studies have focused on the synergistic effects of retinoic acid and 1,25(OH)2D3 as an indication of their interactions and cross-regulation of their activities. In breast and ovarian cancer cells (MCF-7 and NIH:OVCAR3), and in human promyeloid leukemia cells (HL-60), ATRA and vitamin D3 analogs have additive or synergistic effects with respect to inhibition of cell growth, DNA synthesis, and the induction of differentiation [152,174]. In addition, ATRA cooperates with 1,25(OH)2D3 to promote differentiation of HL60 cells to monocytes [175]. By contrast, although ATRA and 1,25(OH)2D3 each inhibited proliferation of normal human keratinocytes, in combination they had antagonistic rather than synergistic effect [176]. The action of retinoic acid may be through direct action of RAR on the VDR gene promoter, as the ability

of retinoic acid to up-regulate the VDR is mediated, at least in part, at the transcriptional level. There are sites that may confer retinoic acid responsiveness upstream and downstream of the human VDR exon 1C promoter [24,51] (see Section III,B,2), but retinoic acid response regions have not been identified in the rodent VDR loci. It appears, therefore, that the retinoic acid regulation of VDR may be tissue and species specific. It is possible that, like glucocorticoid and estrogen effects, retinoids may affect VDR abundance by indirect mechanisms such as changing levels or availability of RXR or of shared transcriptional co-regulators. If so, there may be specificity in the mechanism, as another hormone with a receptor that heterodimerizes with RXR, triiodothyronine, increased VDR content in MG63 osteosarcoma cells [177], and the progesterone analog R5020 increased VDR in the T47-D breast cancer cell line [126]. 3. PEPTIDE HORMONES

In addition to the steroid and retinoid hormones, peptide hormones also play important roles in determining VDR level. They include the polypeptide growth factors, growth hormone, and the calciotropic hormones PTH and PTHrP. This section focuses on the effects of these compounds, which act via membrane receptors. a. Growth Factors In general, 1,25(OH)2D3 inhibits cell proliferation and induces differentiation (Chapter 92), yet VDR levels are often correlated with cellular proliferation in vivo and in vitro. The level of VDR in primary mouse osteoblastic cell cultures appeared to be regulated in accordance with the rate of cell division [178], suggesting that vitamin D regulation may be modulated by cell cycle–associated changes in VDR content. Mitogenic growth factors such as epidermal growth factor (EGF), insulin, and insulin-like growth factor I (IGF-I), which stimulated the proliferation of NIH-3T3 mouse fibroblastic and MCF-7 human breast cancer cell lines and rat growth plate chondrocytes, also caused significant increases in VDR protein and mRNA levels [179,180]. Furthermore, subcutaneous administration of EGF increased intestinal VDR in the proximal segments of the small intestine but not in kidney of suckling rats [181]. By contrast, in UMR-106 osteoblastic cells, EGF reduced the number of 1,25(OH)2D3 binding sites without altering binding affinity, and attenuated PTH-induced VDR up-regulation [182]. Similar down-regulation of VDR expression by basic fibroblast growth factor (bFGF), EGF, and Ha-ras transformation was observed in mouse NIH3T3 fibroblasts and HC-11 mammary epithelial cells [183,184]. Changes in signaling by mitogenic growth factors may also be the cause of decreases in VDR in confluent cultures of the mouse MMB-1 osteoblastic cell line and

208 in mouse primary osteoblastic cell culture [100,101], although this is not a universal finding [101]. Involvement of growth factors in decreased VDR abundance during cellular differentiation is also likely [185,186], as growth factor deprivation can force cellular quiescence, with restoration of VDR expression upon cell cycle reentry [187]. In this context, although the mechanism is not known, it is interesting to note that apigenin, a flavonoid that induces cellular growth arrest with concomitant inhibition of intracellular signalling and decreased protooncogene expression, potently inhibited VDR mRNA and protein expression in human keratinocytes [188]. Effects of growth factors on VDR abundance therefore appeared to be determined by cell type or tissue origin, by the state of confluence, and by the intracellular signaling response. The differences in the effects of growth factors on VDR levels in proliferation and postconfluent cultures is consistent with a role for 1,25(OH)2D3 in control of cell cycle and progression of cells down a differentiation pathway. However, the mechanisms by which VDR abundance is controlled during cell cycle and the function of 1,25(OH)2D3 in the regulation of cellular differentiation remain as yet unanswered questions. b. Growth Hormone Postmenopausal osteoporosis is an age-related disease that is often associated with intestinal calcium malabsorption [189], and 1,25(OH)2D3 is its main hormonal regulator. Calcium malabsorption can be corrected by estrogen therapy [153] and intestinal absorption of calcium is also increased by growth hormone (GH). GH increased intestinal absorption of calcium in pigs and similarly increased intestinal calcium absorption and expression of the calcium binding protein CaBP 9K, but not plasma concentration of 1,25(OH)2D3, in aged female rats [190,191]. The stimulation of calcium absorption by GH may result, in part, from up-regulation of intestinal VDR [192] and possibly ER [193], which up-regulates intestinal VDR, as noted in Section IV,B,2,b. Although GH also increased intestinal calcium absorption and serum 1,25(OH)2D3 in pig, GH treatment did not change intestinal VDR levels though it did reduce renal VDR in this species [194]. It is still unclear whether the GH effect on VDR occurs at the level of transcription, translation, or message stability, and whether GH effects on VDR are mediated indirectly through IGF-I [191], ER [193], or changes in serum calcium [190]. c. Parathyroid Hormone Parathyroid hormone (PTH) plays a central role in the regulation of calcium homeostasis, in part by regulating the synthesis of the other major calciotropic hormone, 1,25(OH)2D3. PTH is a primary inducer of the 1α-hydroxylation of 25(OH)D3 in kidney to produce 1,25(OH)2D3 [195],

LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER

which, in turn, feeds back to control PTH production and regulate VDR in the parathyroid gland (see Chapters 5 and 30). The PTH/PTHrP receptor and VDR are both present in kidney and osteoblasts. The mechanism of PTH action is mediated through activation of the PKA and PKC signal transduction systems [196–198], both of which regulate VDR levels in vitamin D target cells (discussed in Section IV,B,4). Furthermore, PTH-induced increases in intracellular calcium may also regulate VDR levels. This diversity of PTH second messenger responses leads to differential effects on VDR levels and interactions with the vitamin D response pathway that appear to vary according to the test system. VDR content in the preosteoblastic UMR-106 rat osteogenic sarcoma cell line and the MC3T3-E1 mouse preosteoblastic line was increased by PTH in a time- and dose-dependent manner via cAMP activation [196,198,199]. By contrast, PTH or PTHrP treatment of the more mature rat osteoblastic osteosarcoma cell line ROS17/2.8 led to decreased VDR mRNA levels and 1,25(OH)2D3 binding [200,201]. The differences between these osteoblastic models may relate to the different stages of differentiation, perhaps reflecting cell type–specific variation in factors that regulate VDR levels. In concept, this would be consistent with the observations that a PTH-dependent decrease in VDR in ROS 17/2.8 was mediated by the PKA response [201], whereas a PTH-dependent increase in VDR in rat chondrocytic cultures was mediated by the PKC response pathway [202] (discussed further hereafter). Interactions in vivo are also complex, as PTH inhibited an increase in intestinal VDR levels after 5-day coinfusion of rats with 1,25(OH)2D3 [200]. There was also a partial inhibition of renal VDR levels in this study, accompanied by abolition of CYP24 response to the 1,25(OH)2D3 treatment. Tissue-specific variation in PTH effects on VDR levels presumably relate to the activation of various signaling pathways, depending on the specific target cell type and its state of differentiation (discussed further in the next section). 4. ACTIVATORS OF SECOND MESSENGERS

The various physiological inputs that affect VDR levels involve complex interactions among the intracellular signal transduction pathways. This section briefly addresses the regulation of VDR by the second messengers protein kinase A (PKA) and protein kinase C (PKC). PKA activators include forskolin or cAMP analogs, whereas PKC activators include the phorbol esters, such as phorbol myristate acetate (PMA) and phorbol-12,13-dibutyrate. Each can affect VDR levels in isolation, and the evidence suggests that the VDRmodulating activities of the PKA and PKC pathways

CHAPTER 12 VDR Promoter and Abundance

can be mutually antagonistic. Given their central role in the integration of cellular responses to extracellular molecules, these pathways are key in the mechanisms that regulate VDR levels in target cells. a. Protein Kinase A Early studies showed that cellular response to PTH was via activation of the cAMP-dependent protein kinase A (PKA) pathway, which preceded PTH-induced increases in VDR levels in UMR-106 cells [196]. Subsequent mechanistic studies in NIH3T3 cells confirmed that the adenylate cyclase activator forskolin led to a substantial time-dependent increase in VDR abundance [61]. Similar increases by forskolin have been observed in the preosteoblastic UMR-106 and MC3T3-E1 cell lines [198,199], but not in more mature osteoblastic ROS17/2.8 cells, in which forskolin caused a decrease in VDR level and suppressed homologous up-regulation of VDR [201]. PKA activation by forskolin also increased VDR mRNA in the human HOS-8603 osteosarcoma cell line and a megakaryoblastic leukemia line [110], and a prostaglandin E2-induced increase in VDR in HL-60 human promyelocytic cell line also appeared to be via cAMP [203]. The cAMP activation effect on 1,25(OH)2D3 binding in UMR-106 cells was not affected by the increase of intracellular calcium using a calcium agonist or a calcium ionophore; however, cAMP-mediated activation was reduced by use of calcium channel blockers to lower intracellular calcium [204]. The increase in VDR levels by PKA activation thus appears to involve elevation of intracellular calcium. This is particularly interesting, as intracellular calcium is increased in some cell lines by 1,25(OH)2D3, suggesting a mechanism for synergy between the 1,25(OH)2D3 and PKA responses. b. Protein Kinase C Inhibition of VDR levels in mouse NIH3T3 fibroblastic cells by basic FGF (see Section IV,B,3,a) activated protein kinase C (PKC) [183], consistent with the finding that activation of PKC by phorbol esters down-regulated VDR mRNA and protein levels in a time- and dose-dependent manner. Staurosporine, an inhibitor of PKC, blocked the effects of phorbol esters on VDR levels. Activation of PKC also overcame the stimulation of VDR expression by mitogenic serum factors [183]. Kinetics of the VDR response to PKC activation by phorbol esters was time dependent and complex, however, with a decrease and subsequent increase in VDR mRNA levels in the preosteoblastic UMR-106 rat osteosarcoma cell line [205], but a consistent time- and dose-dependent increase in the ROS17/2.8 line, representing a more mature osteoblastic stage [206]. Further, pretreatment with phorbol esters prior to 1,25(OH)2D3 treatment resulted in a synergistic 10-fold induction of VDR mRNA in the latter cell line [206]. Such synergy in regulation of VDR may explain the functional potentiation

209 of 1,25(OH)2D3 responses by PKC activation in primary rat kidney cell [207] and intestinal epithelial cell cultures [208]. In the latter, induction of the vitamin D target gene CYP24 by pharmacological doses of 1,25(OH)2D3 was potentiated by phorbol ester treatment. Similarly, 1,25(OH)2D3 induced phosphorylation of osteopontin and induced tumorigenic transformation of JB6C141.5a epidermal cells only in the presence of phorbol esters [209]. There appears to be a complex relationship between PKC and VDR actions involving cross-regulation of both molecules [210,211]. Direct modulation of the phosphorylation state of the VDR by PKC has been suggested [212,213], with phosphorylation at serine-51 by PKC altering VDR regulation of transcriptional transactivation [214]. Taken together, these observations suggest that VDR function is regulated by functional interactions between PKC and 1,25(OH)2D3. Intracellular calcium levels may play a mechanistic role in this interaction, as both PKC activation and 1,25(OH)2D3 can increase intracellular calcium. There is evidence to support this concept, as elevation in intracellular calcium by a calcium ionophore decreased VDR levels in NIH3T3 cells in a time-dependent manner, which was further decreased after further PKC activation [183]. c. PKA and PKC Interactions Simultaneous activation of PKC and PKA responses in NIH3T3 cells abolished the PKA-mediated increase in VDR protein and mRNA [61], indicating mutual antagonism between these intracellular signal transduction pathways in the regulation of VDR levels. This concept is supported by subsequent observations that PKC activation by phorbol esters in UMR-106 and ROS 17/2.8 cells blocked a PTH- and forskolin-induced up-regulation of VDR [205] and similar findings in MC3T3-E1 and UMR-106–01 osteosarcoma cells [198]. Thus, the evidence suggests cell type-specific and timedependent effects of PKC and PKA pathway activation on VDR levels, as well as interaction between these pathways. The relative strengths of the PKA and PKC signal transduction responses will, of course, vary as relevant extracellular stimuli change. The net effects of the multiple signaling inputs on VDR abundance will depend on downstream changes in VDR transcription and on VDR posttranslational modification. Changes in VDR transcription may result from activation of transcription cofactors that act on the VDR promoter, such as CREB activation by PKA after PTH stimulation. Posttranslational effects that might affect VDR function could include PKC phosphorylation of VDR, which would alter VDR transactivation activity and could have secondary effects on VDR production by homologous or heterologous pathways. The availability

210 of VDR promoter sequences and cloned promoter regions from several species will facilitate dissection of the mechanisms responsible for second-messenger effects on VDR abundance.

V. CONCLUDING REMARKS A wide range of tissues respond to 1,25(OH)2D3 with a variety of actions including calcium homeostatic changes, regulation of cellular proliferation and differentiation, and regulation of vitamin D metabolism. Abundance of the nuclear VDR is a major determinant of the strength of these 1,25(OH)2D3 responses. Factors that influence VDR content include the hormone 1,25(OH)2D3 itself, dietary calcium, phosphorus and fatty acid content, steroid and retinoid hormones, peptide hormones including growth factors and the major calciotropic hormone PTH, and other activators of secondmessenger systems. The existence of multiple promoters in the human locus and of variant VDR isoforms in human and nonhuman species suggests other mechanisms by which nuclear VDR responses to these regulators may vary. The responses to the range of regulators vary in a species-specific manner and exhibit tissue and cell-type specificity. The mechanisms that provide specificity to VDR regulation are not fully understood but may relate to the state of target cell proliferation or differentiation, cell type–specific differences in intracellular second-messenger responses to a variety of activators, and variation in the complement of downstream nuclear proteins available for the regulation of gene transcription in various target cells. There is still much work needed to elucidate the details of these mechanisms and their importance in the regulation of cell proliferation and differentiation and in the physiological vitamin D response. The complex regulation of VDR expression by this variety of effectors in fulfilling the multiple biological roles of 1,25(OH)2D3, coupled with the increasing understanding of the complex promoter structure and regulation of the VDR locus, make this gene an ideal model for further mechanistic investigation. Importantly, understanding of the regulation of VDR abundance may also have clinical utility, as target cell responses to 1,25(OH)2D3 and vitamin D analog treatment may be limited by levels of functional VDR and thus subject to enhancement by therapeutic increases in VDR availability.

References 1. Brumbaugh PF, Haussler MR 1975 Specific binding of 1α,25-dihydroxycholecalciferol to nuclear components of chick intestine. J Biol Chem 250:1588–1594.

LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER

2. Kream BE, Reynolds RD, Knutson JC, Eisman JA, DeLuca HF 1976 Intestinal cytosol binders of 1,25-dihydroxyvitamin D3 and 25-hydroxyvitamin D3. Arch Biochem Biophys 176:779–787. 3. Kream BE, Yamada S, Schnoes HK, DeLuca HF 1977 Specific cytosol-binding protein for 1,25-dihydroxyvitamin D3 in rat intestine. J Biol Chem 252:4501–4505. 4. Kream BE, Jose M, Yamada S, DeLuca HF 1977 A specific high-affinity binding macromolecule for 1,25-dihydroxyvitamin D3 in fetal bone. Science 197:1086–1088. 5. Christakos S, Norman AW 1979 Studies on the mode of action of calciferol. XVIII. Evidence for a specific high affinity binding protein for 1,25 dihydroxyvitamin D3 in chick kidney and pancreas. Biochem Biophys Res Commun 89:56–63. 6. Wecksler WR, Ross FP, Mason RS, Norman AW 1980 Biochemical properties of the 1α, 25-dihydroxyvitamin D3 cytosol receptors from human and chicken intestinal mucosa. J Clin Endocrinol Metab 50:152–157. 7. Simpson RU, DeLuca HF 1980 Characterization of a receptorlike protein for 1,25-dihydroxyvitamin D3 in rat skin. Proc Natl Acad Sci USA 77:5822–5826. 8. Eisman JA, Martin TJ, MacIntyre I, Moseley JM 1979 1,25-dihydroxyvitamin-D-receptor in breast cancer cells. Lancet 2:1335–1336. 9. Frampton RJ, Suva LJ, Eisman JA, Findlay DM, Moore GE, Moseley JM, Martin TJ 1982 Presence of 1,25-dihydroxyvitamin D3 receptors in established human cancer cell lines in culture. Cancer Res 42:1116–1119. 10. Pike JW, Marion SL, Donaldson CA, Haussler MR 1983 Serum and monoclonal antibodies against the chick intestinal receptor for 1,25-dihydroxyvitamin D3. Generation by a preparation enriched in a 64,000-dalton protein. J Biol Chem 258:1289–1296. 11. Dame MC, Pierce EA, Prahl JM, Hayes CE, DeLuca HF 1986 Monoclonal antibodies to the porcine intestinal receptor for 1,25-dihydroxyvitamin D3: interaction with distinct receptor domains. Biochemistry 25:4523–4534. 12. Tuohimaa P, Blauer M, Jaaskelainen T, Itkonen A, Lindfors M, Mahonen A, Palvimo J, Vilja P, Maenpaa PH 1992 Characterization of human 1,25-dihydroxyvitamin D3 receptor anti-peptide antibodies. J Steroid Biochem Mol Biol 43: 649–657. 13. Lee Y, Inaba M, DeLuca HF, Mellon WS 1989 Immunological identification of 1,25 dihydroxyvitamin D3 receptors in human promyelocytic leukemic cells (HL-60) during homologous regulation. J Biol Chem 264: 13701–13705. 14. McDonnell DP, Mangelsdorf DJ, Pike JW, Haussler MR, O’Malley BW 1987 Molecular cloning of complemetary DNA encoding the avian receptor for vitamin D. Science 235:1214–1217. 15. Berger U, Wilson P, McClelland RA, Colston K, Haussler MR, Pike JW, Coombes RC 1988 Immunocytochemical detection of 1,25-dihydroxyvitamin D receptors in normal human tissues. J Clin Endocrinol Metab 67:607–613. 16. Sandgren ME, Bronnegard M, DeLuca HF 1991 Tissue distribution of the 1,25-dihydroxyvitamin D3 receptor in the male rat. Biochem Biophys Res Commun 181:611–616. 17. Walters MR 1992 Newly identified actions of the vitamin D endocrine system. Endocr Rev 13:719–764. 18. Gascon-Barre M, Demers C, Mirshahi A, Neron S, Zalzal S, Nanci A 2003 The normal liver harbors the vitamin D nuclear receptor in nonparenchymal and biliary epithelial cells. Hepatology 37:1034–1042.

211

CHAPTER 12 VDR Promoter and Abundance

19. Baker AR, McDonnell DP, Hughes M, Crisp TM, Mangelsdorf DJ, Haussler MR, Pike JW, Shine J, O’Malley BW 1988 Cloning and expression of full-length cDNA encoding human vitamin D receptor. Proc Natl Acad Sci USA 85:3294–3298. 20. Burmester JK, Wiese RJ, Maeda N, DeLuca HF 1988 Structure and regulation of the rat 1,25-dihydroxyvitamin D3 receptor. Proc Natl Acad Sci USA 85:9499–9502. 21. Elaroussi MA, Prahl JM, DeLuca HF 1994 The avian vitamin D receptors: primary structures and their origins. Proc Natl Acad Sci USA 91:11596–11600. 22. Lu Z, Hanson K, DeLuca HF 1997 Cloning and origin of the two forms of chicken vitamin D receptor. Arch Biochem Biophys 339:99–106. 23. Li YC, Bergwitz C, Juppner H, Demay MB 1997 Cloning and characterization of the vitamin D receptor from Xenopus laevis. Endocrinology 138:2347–2353. 24. Miyamoto K, Kesterson RA, Yamamoto H, Taketani Y, Nishiwaki E, Tatsumi S, Inoue Y, Morita K, Takeda E, Pike JW 1997 Structural organization of the human vitamin D receptor chromosonal gene and its promoter. Mol Endocrinol 11:1165–1179. 25. Jehan F, DeLuca HF 1997 Cloning and characterization of the mouse vitamin D receptor promoter. Proc Natl Acad Sci USA 94:10138–10143. 26. Kamei Y, Kawada T, Fukuwatari T, Ono T, Kato S, Sugimoto E 1995 Cloning and sequencing of the gene encoding the mouse vitamin D receptor. Gene 152:281–282. 27. Crofts LA, Hancock MS, Morrison NA, Eisman JA 1998 Multiple promoters direct the tissue-specific expression of novel N-terminal variant human vitamin D receptor gene transcripts. Proc Natl Acad Sci USA 95:10529–10534. 28. Sunn KL, Cock TA, Crofts LA, Eisman JA, Gardiner EM 2001 Novel N-terminal variant of human VDR. Mol Endocrinol 15:1599–1609. 29. Gardiner EM, Esteban LM, Fong C, Allison SJ, Flanagan JL, Kouzmenko AP, Eisman JA 2004 Vitamin D receptor B1 and exon 1D: functional and evolutionary analysis. J Steroid Biochem Mol Biol 89–90:233–238. 30. Jehan F, DeLuca HF 2000 The mouse vitamin D receptor is mainly expressed through an Sp1-driven promoter in vivo. Arch Biochem Biophys 377:273–283. 31. Arriza JL, Weinberger C, Cerelli G, Glaser TM, Handelin BL, Housman DE, Evans RM 1987 Cloning of human mineralocorticoid receptor complementary DNA: structural and functional kinship with the glucocorticoid receptor. Science 237:268–275. 32. Kuiper GG, Gustafsson JA 1997 The novel estrogen receptor-β subtype: potential role in the cell- and promoter-specific actions of estrogens and anti-estrogens. FEBS Lett 410:87–90. 33. Tremblay GB, Tremblay A, Copeland NG, Gilbert DJ, Jenkins NA, Labrie F, Giguere V 1997 Cloning, chromosomal localization, and functional analysis of the murine estrogen receptor β. Mol Endocrinol 11:353–365. 34. Lazar MA 1993 Thyroid hormone receptors: multiple forms, multiple possibilities. Endocr Rev 14:184–193. 35. Hoopes CW, Taketo M, Ozato K, Liu Q, Howard TA, Linney E, Seldin MF 1992 Mapping of the mouse Rxr loci encoding nuclear retinoid X receptors RXRα, RXRβ, and RXRγ. Genomics 14:611–617. 36. Giguere V, Lyn S, Yip P, Siu CH, Amin S 1990 Molecular cloning of cDNA encoding a second cellular retinoic acidbinding protein. Proc Natl Acad Sci USA 87:6233–6237. 37. Dreyer C, Krey G, Keller H, Gival F, Helftenbein G, Wahli W 1992 Control of the peroxisomal β-oxidation pathway by

38.

39.

40.

41.

42.

43.

44.

45. 46.

47.

48.

49.

50.

a novel family of nuclear hormone receptors. Cell 68: 879–887. Conneely OM, Kettelberger DM, Tsai MJ, Schrader WT, O’Malley BW 1989 The chicken progesterone receptor A and B isoforms are products of an alternate translation initiation event. J Biol Chem 264:14062–14064. Kastner P, Krust A, Turcotte B, Stropp U, Tora L, Gronemeyer H, Chambon P 1990 Two distinct estrogenregulated promoters generate transcripts encoding the two functionally different human progesterone receptor forms A and B. EMBO J 9:1603–1614. Giangrande PH, McDonnell DP 1999 The A and B isoforms of the human progesterone receptor: two functionally different transcription factors encoded by a single gene. Recent Prog Horm Res 54:291–313. Vegeto E, Shahbaz MM, Wen DX, Goldman ME, O’Malley BW, McDonnell DP 1993 Human progesterone receptor A form is a cell- and promoter-specific repressor of human progesterone receptor B function. Mol Endocrinol 7:1244–1255. Tung L, Mohamed MK, Hoeffler JP, Takimoto GS, Horwitz KB 1993 Antagonist-occupied human progesterone B-receptors activate transcription without binding to progesterone response elements and are dominantly inhibited by A-receptors. Mol Endocrinol 7:1256–1265. McDonnell DP, Shahbaz MM, Vegeto E, Goldman ME 1994 The human progesterone receptor A-form functions as a transcriptional modulator of mineralocorticoid receptor transcriptional activity. J Steroid Biochem Mol Biol 48: 425–432. Suzuki T, Suzuki N, Srivastava AS, Kurokawa T 2000 Identification of cDNAs encoding two subtypes of vitamin D receptor in flounder, Paralichthys olivaceus. Biochem Biophys Res Commun 270:40–45. Kozak M 1996 Interpreting cDNA sequences: some insights from studies on translation. Mamm Genome 7:563–574. Ebihara K, Masuhiro Y, Kitamoto T, Suzawa M, Uematsu Y, Yoshizawa T, Ono T, Harada H, Matsuda K, Hasegawa T, Masushige S, Kato S 1996 Intron retention generates a novel isoform of the murine vitamin D receptor that acts in a dominant negative way on the vitamin D signaling pathway. Mol Cell Biol 16:3393–3400. Saijo T, Ito M, Takeda E, Huq AH, Naito E, Yokota I, Sone T, Pike JW, Kuroda Y 1991 A unique mutation in the vitamin D receptor gene in three Japanese patients with vitamin Ddependent rickets type II: utility of single-strand conformation polymorphism analysis for heterozygous carrier detection. Am J Hum Genet 49:668–673. Arai H, Miyamoto K, Taketani Y, Yamamoto H, Iemori Y, Morita K, Tonai T, Nishisho T, Mori S, Takeda E 1997 A vitamin D receptor gene polymorphism in the translation initiation codon: effect on protein activity and relation to bone mineral density in Japanese women. J Bone Miner Res 12:915–921. Jurutka PW, Remus LS, Whitfield GK, Thompson PD, Hsieh JC, Zitzer H, Tavakkoli P, Galligan MA, Dang HT, Haussler CA, Haussler MR 2000 The polymorphic N terminus in human vitamin D receptor isoforms influences transcriptional activity by modulating interaction with transcription factor IIB. Mol Endocrinol 14:401–420. Flanagan JL, Sunn KL, Leong GM, Fong C, Kouzmenko AP, Eisman JA, Gardiner EM 2002 Enhanced transactivation activity of vitamin D receptor B1 associated with focal nuclear accumulation and cofactor binding. J Bone Miner Res suppl 1 17:S216.

212 51. Byrne IM, Flanagan L, Tenniswood MP, Welsh J 2000 Identification of hormone-responsive promoter immediately upstream of exon 1c in the human vitamin D receptor gene. Endocrinology 141:2829–2836. 52. Lu Z, Jehan F, Zierold C, DeLuca HF 2000 Isolation and characterization of the chicken vitamin D receptor gene and its promoter. J Cell Biochem 77:92–102. 53. Gronemeyer H, Laudet V 1995 Trancription factors 3: nuclear receptors. Protein Profile 2:1173–1308. 54. Lemaire P, Vesque C, Schmitt J, Stunnenberg H, Frank R, Charnay P 1990 The serum-inducible mouse gene Krox-24 encodes a sequence-specific transcriptional activator. Mol Cell Biol 10:3456–3467. 55. Schneider-Maurouny S, Topilko P, Seitandou T, Levi G, Cohen-Tannoudji M, Pournin S, Babinet C, Charnay P 1993 Disruption of Krox-20 results in alteration of rhombomeres 3 and 5 in the developing hindbrain. Cell 75:1199–1214. 56. Topilko P, Schneider-Maunoury S, Levi G, Baron-Van Evercooren A, Chennoufi AB, Seitanidou T, Babinet C, Charnay P 1994 Knox-20 controls myelination in the peripheral nervous system. Nature 371:796–799. 57. Levi G, Topilko P, Schneider-Maunoury S, Lasagna M, Mantero S, Cancedda R, Charnay P 1996 Defective bone formationin Krox-20 mutant mice. Development 122:113–120. 58. Schule R, Umesono K, Mangelsdorf DJ, Bolado J, Pike JW, Evans RM 1990 Jun-Fos and receptors for vitamins A and D recognize a common response element in the human osteocalcin gene. Cell 61:497–504. 59. Qi X, Pramanik R, Wang J, Schultz RM, Maitra RK, Han J, DeLuca HF, Chen G 2002 The p38 and JNK pathways cooperate to trans-activate vitamin D receptor via c-Jun/AP-1 and sensitize human breast cancer cells to vitamin D3-induced growth inhibition. J Biol Chem 277:25884–25892. 60. Merriman HL, van Wijnen AJ, Hiebert S, Bidwell JP, Fey E, Lian J, Stein J, Stein GS 1995 The tissue-specific nuclear matrix protein, NMP-2, is a member of the AML/CBF/ PEBP2/runt domain transcription factor family: interactions with the osteocalcin gene promoter. Biochemistry 34: 13125–13132. 61. Krishnan AV, Feldman D 1992 Cyclic adenosine 3′,5′monophosphate up-regulates 1,25-dihydroxyvitamin D3 receptor gene expression and enhances hormone action. Mol Endocrinol 6:198–206. 62. Lazarova DL, Bordonaro M, Sartorelli AC 2001 Transcriptional regulation of the vitamin D3 receptor gene by ZEB. Cell Growth Differ 12:319–326. 63. Takagi TMH, Kondoh H, Higashi Y 1998 δEF1, a zinc finger and homeodomain transcription factor, is required for skeleton patterning in multiple lineages. Development 125:21–31. 64. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Gene 16:391–396. 65. Wagner KD, Wagner N, Sukhatme VP, Scholz H 2001 Activation of vitamin D receptor by the Wilms’ tumor gene product mediates apoptosis of renal cells. J Am Soc Nephrol 12:1188–1196. 66. Maurer U, Jehan F, Englert C, Hubinger G, Weidmann E, DeLuca HF, Bergmann L 2001 The Wilms’ tumor gene product (WT1) modulates the response to 1,25-dihydroxyvitamin D3 by induction of the vitamin D receptor. J Biol Chem 276: 3727–3732.

LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER

67. Yamamoto H, Miyamoto K, Li B, Taketani Y, Kitano M, Inoue Y, Morita K, Pike JW, Takeda E 1999 The caudalrelated homeodomain protein Cdx-2 regulates vitamin D receptor gene expression in the small intestine. J Bone Miner Res 14:240–247. 68. Arai H, Miyamoto KI, Yoshida M, Yamamoto H, Taketani Y, Morita K, Kubota M, Yoshida S, Ikeda M, Watabe F, Kanemasa Y, Takeda E 2001 The polymorphism in the caudalrelated homeodomain protein Cdx-2 binding element in the human vitamin D receptor gene. J Bone Miner Res 16: 1256–1264. 69. Lee TH, Pelletier J 2001 Functional characterization of WT1 binding sites within the human vitamin D receptor gene promoter. Physiol Genomics 7:187–200. 70. Correa P, Akerstrom G, Westin G 2002 Exclusive underexpression of vitamin D receptor exon 1f transcripts in tumors of primary hyperparathyroidism. Eur J Endocrinol 147:671–675. 71. Wietzke JA, Welsh J 2003 Phytoestrogen regulation of a Vitamin D3 receptor promoter and 1,25-dihydroxyvitamin D3 actions in human breast cancer cells. J Steroid Biochem Mol Biol 84:149–157. 72. Duan C, Clemmons DR 1995 Transcription factor AP-2 regulates human insulin-like growth factor binding protein-5 gene expression. J Biol Chem 270:24844–24851. 73. Johnson W, Jameson JL 1999 AP-2 (activating protein 2) and Sp1 (selective promoter factor 1) regulatory elements play distinct roles in the control of basal activity and cyclic adenosine 3′,5′-monophosphate responsiveness of the human chorionic gonadotropin-β promoter. Mol Endocrinol 13:1963–1975. 74. Whitfield GK, Remus LS, Jurutka PW, Zitzer H, Oza AK, Dang HT, Haussler CA, Galligan MA, Thatcher ML, Encinas-Dominguez C, Haussler MR 2001 Functionally relevant polymorphisms in the human nuclear vitamin D receptor gene. Mol Cell Endocrinol 177:145–159. 75. Masi L, Cimaz R, Simonini G, Bindi G, Stagi S, Gozzini A, Malentacchi C, Brandi ML, Falcini F 2002 Association of low bone mass with vitamin d receptor gene and calcitonin receptor gene polymorphisms in juvenile idiopathic arthritis. J Rheumatol 29:2225–2231. 76. Fischer PR, Thacher TD, Pettifor JM, Jorde LB, Eccleshall TR, Feldman D 2000 Vitamin D receptor polymorphisms and nutritional rickets in Nigerian children. J Bone Miner Res 15:2206–2210. 77. Ruza E, Sotillo E, Sierrasesumaga L, Azcona C, PatinoGarcia A 2003 Analysis of polymorphisms of the vitamin D receptor, estrogen receptor, and collagen Iα1 genes and their relationship with height in children with bone cancer. J Pediatr Hematol Oncol 25:780–786. 78. Vogel A, Strassburg CP, Manns MP 2002 Genetic association of vitamin D receptor polymorphisms with primary biliary cirrhosis and autoimmune hepatitis. Hepatology 35:126–131. 79. Chen WC, Chen HY, Lu HF, Hsu CD, Tsai FJ 2001 Association of the vitamin D receptor gene start codon Fok I polymorphism with calcium oxalate stone disease. BJU Int 87:168–171. 80. Chiu KC, Chuang LM, Yoon C 2001 The vitamin D receptor polymorphism in the translation initiation codon is a risk factor for insulin resistance in glucose tolerant Caucasians. BMC Med Genet 2:2. 81. Kishi S, Yang W, Boureau B, Morand S, Das S, Chen P, Cook EH, Rosner GL, Schuetz EG, Pui CH, Relling MV 2004 Effects of prednisone and genetic polymorphisms on etoposide disposition in children with acute lymphoblastic leukemia. Blood 103:67–72.

CHAPTER 12 VDR Promoter and Abundance

82. Morrison NA, Yeoman R, Kelly PJ, Eisman JA 1992 Contribution of trans-acting factor alleles to normal physiological variability: Vitamin D receptor gene polymorphisms and circulating osteocalcin. Proc Natl Acad Sci USA 89:6665–6669. 83. Faraco JH, Morrison NA, Baker A, Shine J, Frossard PM 1989 Apa I dimorphism at the human vitamin D receptor gene locus. Nucleic Acids Res 17:2150. 84. Ingles SA, Haile RW, Henderson BE, Kolonel LN, Nakaichi G, Shi CY, Yu MC, Ross RK, Coetzee GA 1997 Strength of linkage disequilibrium between two vitamin D receptor markers in five ethnic groups: implications for association studies. Cancer Epidemiol Biomarkers Prev 6:93–98. 85. Morrison NA, Qi JC, Tokita A, Kelly PJ, Crofts L, Nguyen TV, Sambrook PN, Eisman JA 1994 Prediction of bone density from vitamin D receptor alleles. Nature 367: 284–287. 86. Yamagata Z, Miyamura T, Iijima S, Asaka A, Sasaki M, Kato J, Koizumi K 1994 Vitamin D receptor gene polymorphism and bone mineral density in healthy Japanese women. Lancet 344:1027. 87. Ferrari S, Rizzoli R, Chevalley T, Slosman D, Eisman JA, Bonjour JP 1995 Vitamin-D-receptor-gene polymorphisms and change in lumbar-spine bone mineral density. Lancet 345:423–424. 88. Spector TD, Keen RW, Arden NK, Morrison NA, Major PJ, Nguyen TV, Kelly PJ, Baker JR, Sambrook PN, Lanchbury JS, Eisman JA 1995 Influence of vitamin D receptor genotype on bone mineral density in postmenopausal women: a twin study in Britain. Bri Med J 310:1357–1360. 89. Krall EA, Parry P, Lichter JB, Dawson-Hughes B 1995 Vitamin D receptor alleles and rates of bone loss: influences of years since menopause and calcium intake. J Bone Miner Res 10:978–984. 90. Grundberg E, Brandstrom H, Ribom EL, Ljunggren O, Kindmark A, Mallmin H 2003 A poly adenosine repeat in the human vitamin D receptor gene is associated with bone mineral density in young Swedish women. Calcif Tissue Int (in press). 91. Hustmyer FG, Peacock M, Hui S, Johnston CC, Christian J 1994 Bone mineral density in relation to polymorphism at the vitamin D receptor gene locus. J Clin Invest 94: 2130–2134. 92. Melhus H, Kindmark A, Amer S, Wilen B, Lindh E, Ljunghall S 1994 Vitamin D receptor genotypes in osteoporosis. Lancet 344:949. 93. Carling T, Rastad J, Akerstrom G, Westin G 1998 Vitamin D receptor (VDR) and parathyroid hormone messenger ribonucleic acid levels correspond to polymorphic VDR alleles in human parathyroid tumors. J Clin Endocrinol Metab 83:2255–2259. 94. Verbeek W, Gombart AF, Shiohara M, Campbell M, Koeffler HP 1997 Vitamin D receptor: no evidence for allelespecific mRNA stability in cells which are heterozygous for the Taq I restriction enzyme polymorphism. Biochem Biophys Res Commun 238:77–80. 95. Barger-Lux MJ, Heaney RP, Hayes J, DeLuca HF, Johnson ML, Gong G 1995 Vitamin D receptor gene polymorphism, bone mass, body size, and vitamin D receptor density. Calcif Tissue Int 57:161–162. 96. Kinyamu HK, Gallagher JC, Knezetic JA, DeLuca HF, Prahl JM, Lanspa SJ 1997 Effect of vitamin D receptor genotypes on calcium absorption, duodenal vitamin D receptor concentration, and serum 1,25(OH)2D3 dihydroxyvitamin D levels in normal women. Calcif Tissue Int 60:491–495.

213 97. Mocharla H, Butch AW, Pappas AA, Flick JT, Weinstein RS, De Togni P, Jilka RL, Roberson PK, Parfitt AM, SC M 1997 Quantification of vitamin D receptor mRNA by competitive polymerase chain reaction in PBMC: lack of correspondence with common allelic variants. J Bone Miner Res 12:726–733. 98. Gross C, Musiol IM, Eccleshall TR, Malloy PJ, Feldman D 1998 Vitamin D receptor gene polymorphisms: analysis of ligand binding and hormone responsiveness in cultured skin fibroblasts. Biochem Biophys Res Commun 242:467–473. 99. Halloran BP, DeLuca HF 1981 Appearance of the intestinal cytosolic receptor for 1,25-dihydroxyvitamin D3 during neonatal development in the rat. J Biol Chem 256:7338–7342. 100. Walters MR, Rosen DM, Norman AW, Luben RA 1982 1,25-Dihydroxyvitamin D receptors in an established bone cell line. Correlation with biochemical responses. J Biol Chem 257:7481–7484. 101. Chen TL, Li JM, Ye TV, Cone CM, Feldman D 1986 Hormonal responses to 1,25-dihydroxyvitamin D3 in cultured mouse osteoblast-like cells: modulation by changes in receptor level. J Cell Physiol 126:21–28. 102. Dokoh S, Donaldson CA, Haussler MR 1984 Influence of 1,25-dihydroxyvitamin D3 on cultured osteogenic sarcoma cells: correlation with the 1,25-dihydroxyvitamin D3 receptor. Cancer Res 44:2103–2109. 103. Costa EM, Hirst MA, Feldman D 1985 Regulation of 1,25-dihydroxyvitamin D3 receptors by vitamin D analogs in cultured mammalian cells. Endocrinology 117:2203–2210. 104. Pan LC, Price PA 1987 Ligand-dependent regulation of the 1,25-dihydroxyvitamin D3 receptor in rat osteosarcoma cells. J Biol Chem 262:4670–4675. 105. Sher E, Frampton RJ, Eisman JA 1985 Regulation of the 1,25-dihydroxyvitamin D3 receptor by 1,25-dihydroxyvitamin D3 in intact human cancer cells. Endocrinology 116:971–977. 106. Pols HAP, Schilte HP, Visser TJ, Birkenhager JC 1987 Effect of ketoconazole on metabolism and binding of 1,25dihydroxyvitamin D3 by intact rat osteogenic sarcoma cells. Biochim Biophys Acta 931:115–119. 107. Pols HAP, Birkenhager JC, Schilte JP, Visser TJ 1988 Evidence that the self-induced metabolism of 1,25-dihydroxyvitamin D3 limits the homologous up-regulation of its receptor in rat osteosarcoma cells. Biochim Biophys Acta 970:122–129. 108. Dame MC, Pierce EA, DeLuca HF 1985 Identification of the porcine intestinal 1,25-dihydroxyvitamin D3 receptor on sodium dodecyl sulfate/polyacrylamide gels by renaturation and immunoblotting. Proc Natl Acad Sci USA 82: 7825–7829. 109. Mahonen A, Maenpaa PH 1994 Steroid hormone modulation of vitamin D receptor levels in human MG-63 osteosarcoma cells. Biochem Biophys Res Commun 205:1179–1186. 110. Song LN 1996 Demonstration of vitamin D receptor expression in a human megakaryoblastic leukemia cell line: regulation of vitamin D receptor mRNA expression and responsiveness by forskolin. J Steroid Biochem Mol Biol 57:265–274. 111. Bland R, Sammons RL, Sheppard MC, Williams GR 1997 Thyroid hormone, vitamin D and retinoid receptor expression and signalling in primary cultures of rat osteoblastic and immortalised osteosarcoma cells. J Endocrinol 154:63–74. 112. Meyer J, Fullmer CS, Wasserman RH, Komm BS, Haussler MR 1992 Dietary restriction of calcium, phosphorus, and vitamin D elicits differential regulation of the mRNAs for avian intestinal calbindin-D28k and the 1,25-dihydroxyvitamin D3 receptor. J Bone Miner Res 7:441–448.

214 113. Costa EM, Feldman D 1986 Homologous up-regulation of the 1,25(OH)2 vitamin D3 receptor in rats. Biochem Biophys Res Commun 137:742–747. 114. Naveh-Many T, Marx R, Keshet E, Pike JW, Silver J 1990 Regulation of 1,25-dihydroxyvitamin D3 receptor gene expression by 1,25-dihydroxyvitamin D3 in the parathyroid in vivo. J Clin Invest 86:1968–1975. 115. Huang YC, Lee S, Stolz R, Gabrielides C, Pansini-Porta A, Bruns ME, Bruns DE, Miffin TE, Pike JW, Christakos S 1989 Effect of hormones and development on the expression of the rat 1,25-dihydroxyvitamin D3 receptor gene. Comparison with calbindin gene expression. J Biol Chem 264:17454–17461. 116. Strom M, Sandgren ME, Brown TA, DeLuca HF 1989 1,25-dihydroxyvitamin D3 up-regulates the 1,25-dihydroxyvitamin D3 receptor in vivo. Proc Natl Acad Sci USA 86:9770–9773. 117. Sandgren ME, DeLuca HF 1990 Serum calcium and vitamin D regulate 1,25-dihydroxyvitamin D receptor concentration in rat kidney in vivo. Proc Natl Acad Sci USA 87:4312–4314. 118. Zineb R, Zhor B, Odile W, Marthe RR 1998 Distinct, tissuespecific regulation of vitamin D receptor in the intestine, kidney, and skin by dietary calcium and vitamin D. Endocrinology 139:1844–1852. 119. Goff JP, Reinhardt TA, Beckman MJ, Horst RL 1990 Contrasting effects of exogenous 1,25-dihydroxyvitamin D [1,25-(OH)2D] versus endogenous 1,25-(OH)2D, induced by dietary calcium restriction, on vitamin D receptors. Endocrinology 126:1031–1035. 120. Gensure RC, Antrobus SD, Fox J, Okwueze M, Talton SY, Walters MR 1998 Homologous up-regulation of vitamin D receptors is tissue specific in the rat. J Bone Miner Res 13:454–463. 121. Mangelsdorf DJ, Pike JW, Haussler MR 1987 Avian and mammalian receptors for 1,25-dihydroxyvitamin D3: in vitro translation to characterize size and hormone-dependent regulation. Proc Natl Acad Sci USA 84:354–358. 122. Pan P, Reddy K, Lee S, Studzinski GP 1991 Differentiationrelated regulation of 1,25-dihydroxyvitamin D3 receptor mRNA in human leukaemia cells HL-60. Cell Prolif 24:159–170. 123. Costa EM, Feldman D 1987 Measurement of 1,25-dihydroxyvitamin D3 receptor turnover by dense amino acid labeling: changes during receptor up-regulation by vitamin D metabolites. Endocrinology 120:1173–1178. 124. Wiese RJ, Uhland-Smith A, Ross TK, Prahl JM, DeLuca HF 1992 Up-regulation of the vitamin D receptor in response to 1,25-dihydroxyvitamin D3 results from ligand-induced stabilization. J Biol Chem 267:20082–20086. 125. Arbour NC, Prahl JM, DeLuca HF 1993 Stabilization of the vitamin D receptor in rat osteosarcoma cells through the action of 1,25-dihydroxyvitamin D3. Mol Endocrinol 7:1307–1312. 126. Davoodi F, Brenner RV, Evans SR, Schumaker LM, Shabahang M, Nauta RJ, Buras RR 1995 Modulation of vitamin D receptor and estrogen receptor by 1,25(OH)2vitamin D3 in T-47D human breast cancer cells. J Steroid Biochem Mol Biol 54:147–153. 127. Healy KD, Zella JB, Prahl JM, DeLuca HF 2003 Regulation of the murine renal vitamin D receptor by 1,25-dihydroxyvitamin D3 and calcium. Proc Natl Acad Sci USA 100: 9733–9737. 128. Uhland-Smith A, DeLuca HF 1993 The necessity for calcium for increased renal vitamin D receptor in response to 1,25dihydroxyvitamin D. Biochim Biophys Acta 1176:321–326.

LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER

129. Beckman MJ, DeLuca HF 2002 Regulation of renal vitamin D receptor is an important determinant of 1α,25-dihydroxyvitamin D3 levels in vivo. Arch Biochem Biophys 401: 44–52. 130. Brown AJ, Zhong M, Finch J, Ritter C, Slatopolsky E 1995 The roles of calcium and 1,25-dihydroxyvitamin D3 in the regulation of vitamin D receptor expression by rat parathyroid glands. Endocrinology 136:1419–1424. 131. Zhu Y, Goff JP, Reinhardt TA, Horst RL 1997 Pregnancy and lactation increase vitamin D–dependent intestinal membrane calcium adenosine triphosphatase and calcium binding protein messenger ribonucleic acid expression. Endocrinology 139:3520–3524. 132. Sriussadaporn S, Wong MS, Pike JW, Favus MJ 1995 Tissue specificity and mechanism of vitamin D receptor up-regulation during dietary phosphorus restriction in the rat. J Bone Miner Res 10:271–280. 133. Hernandez A, Concepcion MT, Rodriguez M, Salido E, Torres A 1996 High phosphorus diet increases preproPTH mRNA independent of calcium and calcitriol in normal rats. Kidney Int 50:1872–1878. 134. Jump DB, Clarke SD, Thelen A, Liimatta M, Ren B, Badin MV 1997 Dietary fat, genes, and human health. Adv Exp Med Biol 422:167–176. 135. Leonard F, Haag M, Kruger MC 2001 Modulation of intestinal vitamin D receptor availability and calcium ATPase activity by essential fatty acids. Prostaglandins Leukot Essent Fatty Acids 64:147–150. 136. Gaschott T, Stein J 2003 Short-chain fatty acids and colon cancer cells: the vitamin D receptor—butyrate connection. Recent Results Cancer Res 164:247–257. 137. Gaschott T, Wachtershauser A, Steinhilber D, Stein J 2001 1,25-Dihydroxycholecalciferol enhances butyrate-induced p21(Waf1/Cip1) expression. Biochem Biophys Res Commun 283:80–85. 138. Costa EM, Feldman D 1987 Modulation of 1,25-dihydroxyvitamin D3 receptor binding and action by sodium butyrate in cultured pig kidney cells (LLC-PK1). J Bone Miner Res 2:151–159. 139. Maiyar AC, Norman AW 1992 Effects of sodium butyrate on 1,25-dihydroxyvitamin D3 receptor activity in primary chick kidney cells. Mol Cell Endocrinol 84:99–107. 140. Kim CH, Cheng SL, Kim GS 1999 Effects of dexamethasone on proliferation, activity, and cytokine secretion of normal human bone marrow stromal cells: possible mechanisms of glucocorticoid-induced bone loss. J Endocrinol 162: 371–379. 141. Massaro ER, Simpson RU, DeLuca HF 1983 Glucocorticoids and appearance of 1,25-dihydroxyvitamin D3 receptor in rat intestine. Am J Physiol 244:E230–E235. 142. Lee S, Szlachetka M, Christakos S 1991 Effect of glucocorticoids and 1,25-dihydroxyvitamin D3 on the developmental expression of the rat intestinal vitamin D receptor gene. Endocrinology 129:396–401. 143. Gensure RC, Fish JM, Walters MR 1993 Dexamethasone downregulates vitamin D receptors in rat kidney, unmasking a high affinity binding site. Biochem Biophys Res Commun 195:1139–1144. 144. Hirst M, Feldman D 1982 Glucocorticoids down-regulate the number of 1,25-dihydroxyvitamin D3 receptors in mouse intestine. Biochem Biophys Res Commun 105:1590–1596. 145. Wilhelm F, Norman AW 1985 Influence of triamcinolone, estradiol-17β and testosterone on 1,25-dihydroxyvitamin D3 binding performances to its chick intestinal receptor. J Steroid Biochem 23:913–918.

CHAPTER 12 VDR Promoter and Abundance

146. Manolagas SC, Anderson DC, Lumb GA 1979 Glucocorticoids regulate the concentration of 1,25-dihydroxycholecalciferol receptors in bone. Nature 277:314–315. 147. Manolagas SC, Abare J, Deftos LJ 1984 Glucocorticoids increase the 1,25(OH)2D3 receptor concentration in rat osteogenic sarcoma cells. Calcif Tissue Int 36:153–157. 148. Chen TL, Cone CM, Morey-Hilton E, Feldman D 1983 1α,25-dihydroxyvitamin D3 receptors in cultured rat osteoblast-like cells. Glucocorticoid treatment increases receptor content. J Biol Chem 258:4350–4355. 149. Chen TL, Cone CM, Morey-Holton E, Feldman D 1982 Glucocorticoid regulation of 1,25(OH)2-vitamin D3 receptors in cultured mouse bone cells. J Biol Chem 257:13564–13569. 150. Godschalk M, Levy JR, Downs RW 1992 Glucocorticoids decrease vitamin D receptor number and gene expression in human osteosarcoma cells. J Bone Miner Res 7:21–27. 151. Midorikawa K, Sayama K, Shirakata Y, Hanakawa Y, Sun L, Hashimoto K 1999 Expression of vitamin D receptor in cultured human keratinocytes and fibroblasts is not altered by corticosteroids. J Dermatol Sci 21:8–12. 152. Saunders DE, Christensen C, Williams JR, Wappler NL, Lawrence WD, Malone JM, Malviya VK, Deppe G 1995 Inhibition of breast and ovarian carcinoma cell growth by 1,25-dihydroxyvitamin D3 combined with retinoic acid or dexamethasone. Anticancer Drugs 6:562–569. 153. Gallagher JC, Riggs BL, Eisman JA, Hamstra A, Arnaud SB, DeLuca HF 1979 Intestinal calcium absorption and serum vitamin D metabolites in normal subjects and osteoporotic patients. J Clin Invest 64:729–736. 154. Gallagher JC, Riggs BL, DeLuca HF 1980 Effect of estrogen on calcium absorption and serum vitamin D metabolites in postmenopausal osteoporosis. J Clin Endocrinol Metab 51:1359–1364. 155. Ten Bolscher M, Netelenbos JC, Barto R, Van Buuren LM, Van der vijgh WJ 1999 Estrogen regulation of intestinal calcium absorption in the intact and ovariectomized adult rat. J Bone Miner Res 14:1197–1202. 156. Liel Y, Shany S, Smirnoff P, Schwartz B 1999 Estrogen increases 1,25-dihydroxyvitamin D receptors expression and bioresponse in the rat duodenal mucosa. Endocrinology 140:280–285. 157. Colin EM, Van Den Bemd GJ, Van Aken M, Christakos S, De Jonge HR, DeLuca HF, Prahl JM, Birkenhager JC, Buurman CJ, Pols HA, Van Leeuwen JP 1999 Evidence for involvement of 17β-estradiol in intestinal calcium absorption independent of 1,25-dihydroxyvitamin D3 level in the Rat. J Bone Miner Res 14:57–64. 158. Schwartz B, Smirnoff P, Shany S, Y L 2000 Estrogen controls expression and bioresponse of 1,25-dihydroxyvitamin D receptors in the rat colon. Mol Cell Biochem 203:87–93. 159. Lechner D, Cross HS 2003 Phytoestrogens and 17β-estradiol influence vitamin D metabolism and receptor expression— relevance for colon cancer prevention. Recent Results Cancer Res 164:379–391. 160. Levy J, Zuili I, Yankowitz N, Shany S 1984 Induction of cytosolic receptors for 1α,25-dihydroxyvitamin D3 in the immature rat uterus by oestradiol. J Endocrinol 100:265–269. 161. Duncan WE, Glass AR, Wray HL 1991 Estrogen regulation of the nuclear 1,25-dihydroxyvitamin D3 receptor in rat liver and kidney. Endocrinology 129:2318–2324. 162. Duque G, Abdaimi KE, Macoritto M, Miller MM, Kremer R 2002 Estrogens (E2) regulate expression and response of 1,25-dihydroxyvitamin D3 receptors in bone cells: changes with aging and hormone deprivation. Biochem Biophys Res Commun 299:446–454.

215 163. Liel Y, Kraus S, Levy J, Shany S 1992 Evidence that estrogens modulate activity and increase the number of 1,25-dihydroxyvitamin D receptors in osteoblast-like cells (ROS 17/2.8). Endocrinology 130:2597–2601. 164. Ishibe M, Nojima T, Ishibashi T, Koda T, Kaneda K, Rosier RN, Puzas JE 1995 17β-estradiol increases the receptor number and modulates the action of 1,25-dihydroxyvitamin D3 in human osteosarcoma-derived osteoblast-like cells. Calcif Tissue Int 57:430–435. 165. Kinuta K, Tanaka H, Moriwake T, Aya K, Kato S, Seino Y 2000 Vitamin D is an important factor in estrogen biosynthesis of both female and male gonads. Endocrinology 141:1317–1324. 166. Allenby G, Bocquel MT, Saunders M, Kazmer S, Speck J, Rosenberger M, Lovey A, Kastner P, Grippo JF, Chambon P, Levin A 1993 Retinoic acid receptors and retinoid X receptors: interactions with endogenous retinoic acids. Proc Natl Acad Sci USA 90:30–34. 167. Yu VC, Delsert C, Andersen B, Holloway JM, Devary OV, Naar AM, Kim SY, Boutin JM, Glass CK, Rosenfeld MG 1991 RXRβ: a coregulator that enhances binding of retinoic acid, thyroid hormone, and vitamin D receptors to their cognate response elements. Cell 67:1251–1266. 168. Kliewer SA, Umesono K, Manglesdorf DJ, Evans RM 1992 Retinoid X receptor interacts with nuclear receptors in retinoic acid, thyroid hormone and vitamin D3 signalling. Nature 355:446–449. 169. Navia JM, Harris SS 1980 Vitamin A influence on calcium metabolism and calcification. Ann N Y Acad Sci 355:45–57. 170. Bikle DD, Pillai S 1993 Vitamin D, calcium, and epidermal differentiation. Endocr Rev 14:3–19. 171. Petkovich PM, Heersche JNM, Tinker DO, Jones G 1984 Retinoic acid stimulates 1,25-dihydroxyvitamin D3 binding in rat osteosarcoma cells. J Biol Chem 259:8274–8280. 172. Lee KL, Petkovich PM, Heersche JN 1988 The effects of sodium butyrate on the retinoic acid-induced changes in 1,25-dihydroxyvitamin D3 receptors in tumorigenic and nontumorigenic bone derived cell lines. Endocrinology 122:2399–2406. 173. Chen TL, Feldman D 1985 Retinoic acid modullation of 1,25(OH)2 vitamin D3 receptors and bioresponse in bone cells: species differences between rat and mouse. Biochem Biophys Res Commun 132:74–80. 174. Dore BT, Uskokovic MR, Momparler RL 1993 Interaction of retinoic acid and vitamin D3 analogs on HL-60 myeloid leukemic cells. Leuk Res 17:749–757. 175. Brown G, Bunce CM, Rowlands DC, Williams GR 1994 Alltrans retinoic acid and 1α,25-dihydroxyvitamin D3 co-operate to promote differentiation of the human promyeloid leukemia cell line HL60 to monocytes. Leukemia 8:806–815. 176. Gibson DF, Bikle DD, Harris J 1998 All-trans retinoic acid blocks the antiproliferative prodifferentiating actions of 1,25dihydroxyvitamin D3 in normal human keratinocytes. J Cell Physiol 174:1–8. 177. Mahonen A, Pirskanen A, Maenpaa PH 1991 Homologous and heterologous regulation of 1,25-dihydroxyvitamin D-3 receptor mRNA levels in human osteosarcoma cells. Biochim Biophys Acta 1088:111–118. 178. Chen TL, Feldman D 1981 Regulation of 1,25-dihydroxyvitamin D3 receptors in cultured mouse bone cells. Correlation of receptor concentration with the rate of cell division. J Biol Chem 256:5561–5566. 179. Krishnan AV, Feldman D 1991 Stimulation of 1,25-dihydroxyvitamin D3 receptor gene expression in cultured cells by serum and growth factors. J Bone Miner Res 6:1099–1107.

216 180. Klaus G, Weber L, Rodriguez J, Fernandez P, Klein T, Grulich-Henn J, Hugel U, Ritz E, Mehls P 1998 Interaction of IGF-I and 1α,25(OH)2D3 on receptor expression and growth stimulation in rat growth plate chondrocytes. Kidney Int 53:1152–1161. 181. Bruns DE, Krishnan AV, Feldman D, Gray RW, Christakos S, Hirsch GN, Bruns ME 1989 Epidermal growth factor increases intestinal calbindin-D9k and 1,25-dihydroxyvitamin D receptors in neonatal rats. Endocrinology 125:478–485. 182. van Leeuwen JPTM, Pols HAP, Schilte JP, Visser TJ, Birkenhager JC 1991 Modulation by epidermal growth factor of the basal 1,25(OH)2D3 receptor level and the heterologous up-regulation of the 1,25(OH)2D3 receptor in clonal osteoblast-like cells. Calcif Tissue Int 49:35–42. 183. Krishnan AV, Feldman D 1991 Activation of protein kinase-C inhibits vitamin D receptor gene expression. Mol Endocrinol 5:605–612. 184. Escaleira MT, Brentani MM 1999 Vitamin D3 receptor (VDR) expression in HC-11 mammary cells: regulation by growthmodulatory agents, differentiation, and Ha-ras transformation. Breast Cancer Res Treat 54:123–133. 185. Zhao X, Feldman D 1993 Regulation of vitamin D receptor abundance and responsiveness during differentiation of HT-29 human colon cancer cells. Endocrinology 132:1808–1814. 186. Hewison M, Freeman L, Hughes SV, Evans KN, Bland R, Eliopoulos AG, Kilby MD, Moss PAH, Chakraverty R 2003 Differential regulation of vitamin D receptor and its ligand in human monocyte-derived dendritic cells. J Immunol 170: 5382–5390. 187. Segaert S, Degreef H, Bouillon R 2000 Vitamin D receptor expression is linked to cell cycle control in normal human keratinocytes. Biochem Biophys Res Commun 279:89–94. 188. Segaert S, Courtois S, Garmyn M, Degreef H, Bouillon R 2000 The flavonoid apigenin suppresses vitamin D receptor expression and vitamin D responsiveness in normal human keratinocytes. Biochem Biophys Res Commun 268:237–241. 189. Gallagher JC 1990 The pathogenesis of osteoporosis. Bone Miner 9:215–227. 190. Denis I, Thomasset M, Pointillart A 1994 Influence of exogenous porcine growth hormone on vitamin D metabolism and calcium and phosphorus absorption in intact pigs. Calcif Tissue Int 54:489–492. 191. Fleet JC, Bruns ME, Hock JM, Wood RJ 1994 Growth hormone and parathyroid hormone stimulate intestinal calcium absorption in aged female rats. Endocrinology 134: 1755–1760. 192. Chen C, Noland KA, Kalu DN 1997 Modulation of intestinal vitamin D receptor by ovariectomy, estrogen and growth hormone. Mech Ageing Dev 99:109–122. 193. Chen C, Kalu DN 1998 Modulation of intestinal estrogen receptor by ovariectomy, estrogen and growth hormone. J Pharmacol Exp Ther 286:328–333. 194. Goff JP, Caperna TJ, Steele NC 1990 Effects of growth hormone administration on vitamin D metabolism and vitamin D receptors in the pig. Domest Anim Endocrinol 7:425–433. 195. Garabedian M, Holick MF, DeLuca HF, Boyle IT 1972 Control of 25-hydroxycholecalciferol metabolism by parathyroid glands. Proc Natl Acad Sci USA 69:1673–1676. 196. Pols HAP, van Leeuwen JP, Schilte JP, Visser TJ, Birkenhager JC 1988 Heterologous up-regulation of the 1,25-dihydroxyvitamin D3 receptor by parathyroid hormone (PTH) and PTH-like peptide in osteoblast-like cells. Biochem Biophys Res Commun 156:588–594. 197. Chakravarthy BR, Durkin JP, Rixon RH, Whitfield JF 1990 Parathyroid hormone fragment [3-34] stimulates protein kinase C (PKC) activity in rat osteosarcoma and murine

LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER

198.

199.

200.

201.

202.

203.

204.

205.

206.

207.

208.

209.

210. 211.

T-lymphoma cells. Biochem Biophys Res Commun 171: 1105–1110. Krishnan AV, Cramer SD, Bringhurst FR, Feldman D 1995 Regulation of 1,25-dihydroxyvitamin D3 receptors by parathyroid hormone in osteoblastic cells: role of second messenger pathways. Endocrinology 136:705–712. van Leeuwen JPTM, Birkenhager JC, Vink-van Wijngaarden T, van den Bemd GJCM, Pols HAP 1992 Regulation of 1,25dihydroxyvitamin D3 receptor gene expression by parathyroid hormone and cAMP-agonists. Biochem Biophys Res Commun 185:881–886. Reinhardt TA, Horst RL 1990 Parathyroid hormone downregulates 1,25-dihydroxyvitamin D receptors (VDR) and VDR messenger ribonucleic acid in vitro and blocks homologous up-regulation of VDR in vivo. Endocrinology 127:942–948. Sriussadaporn S, Wong MS, Whitfield JF, Tembe V, Favus MJ 1995 Structure–function relationship of human parathyroid hormone in the regulation of vitamin D receptor expression in osteoblast-like cells (ROS 17/2.8). Endocrinology 136: 3735–3742. Klaus G, von Eichel B, May T, Hugel U, Mayer H, Ritz E, Mehls O 1994 Synergistic effects of parathyroid hormone and 1,25-dihydroxyvitamin D3 on proliferation and vitamin D receptor expression of rat growth cartilage cells. Endocrinology 135:1307–1315. Smith SJ, Green LM, Hayes ME, Mawer EB 1999 Prostaglandin E2 regulates vitamin D receptor expression, vitamin D-24-hydroxylase activity and cell proliferation in an adherent human myeloid leukemia cell line (Ad-HL60). Prostaglandins Other Lipid Mediat 57:73–85. van Leeuwen JPTM, Birkenhager JC, Schilte JP, Buurman CJ, Pols HAP 1990 Role of calcium and cAMP in heterologous up-regulation of the 1,25-dihydroxyvitamin D3 receptor in an osteoblast cell line. Cell Calcium 11:281–289. van Leeuwen JPTM, Birkenhager JC, Buurman CJ, van den Bemd GJCM, Bos MP, Pols HAP 1992 Bidirectional regulation of the 1,25-dihydroxyvitamin D3 receptor by phorbol esteractivated protein kinase-C in osteoblast-like cells: interaction with adenosine 3′,5′-monophosphate-induced up-regulation of the 1,25-dihydroxyvitamin D3 receptor. Endocrinology 130:2259–2266. Reinhardt TA, Horst RL 1994 Phorbol 12-myristate 13-acetate and 1,25-dihydroxyvitamin D3 regulate 1,25-dihydroxyvitamin D3 receptors synergistically in rat osteosarcoma cells. Mol Cell Endocrinol 101:159–165. Chen ML, Boltz MA, Armbrecht HJ 1993 Effects of 1,25dihydroxyvitamin D3 and phorbol ester on 25-hydroxyvitamin D3 24-hydroxylase cytochrome P450 messenger ribonucleic acid levels in primary cultures of rat renal cells. Endocrinology 132:1782–1788. Armbrecht HJ, Hodam TL, Boltz MA, Chen ML 1993 Phorbol ester markedly increases the sensitivity of intestinal epithelial cells to 1,25-dihydroxyvitamin D3. FEBS Lett 327:13–16. Chang PL, Prince CW 1993 1α,25-Dihydroxyvitamin D3 enhances 12-O-tetradecanoylphorbol-13-acetate-induced tumorigenic transformation and osteopontin expression in mouse JB6 epidermal cells. Cancer Res 53:2217–2220. Martell RE, Simpson RU, Taylor JM 1987 1,25-Dihydroxyvitamin D3 regulation of phorbol ester receptors in HL-60 leukemia cells. J Biol Chem 262:5570–5575. van Leeuwen JPTM, Birkenhager JC, M vdBGJC, Buurman CJ, Staal A, Bos M, Pols HAP 1992 Evidence for the functional involvement of protein kinase C in the action of 1,25-dihydroxyvitamin D3 in bone. J Biol Chem 267:12562–12569.

CHAPTER 12 VDR Promoter and Abundance

212. Desai RK, van Wijnen AJ, Stein JL, Stein GS, Lian JB 1995 Control of 1,25-dihydroxyvitamin D3 receptor–mediated enhancement of osteocalcin gene transcription: effects of perturbing phosphorylation pathways by okadaic acid and staurosporine. Endocrinology 136:5685–5693. 213. Hara H, Yasunami Y, Adachi T 2002 Alteration of cellular phosphorylation state affects vitamin D receptor–mediated

217 CYP3A4 mRNA induction in Caco-2 cells. Biochem Biophys Res Commun 296:182–188. 214. Hsieh JC, Jurutka PW, Nakajima S, Galligan MA, Haussler CA, Shimizu Y, Shimizu N, Whitfield GK, Haussler MR 1993 Phosphorylation of the human vitamin D receptor by protein kinase C. Biochemical and functional evaluation of the serine 51 recognition site. J Biol Chem 268:15118–15126.

CHAPTER 13

Nuclear Vitamin D Receptor: Structure-Function, Molecular Control of Gene Transcription, and Novel Bioactions G. KERR WHITFIELD, PETER W. JURUTKA, CAROL A. HAUSSLER, JUI-CHENG HSIEH, THOMAS K. BARTHEL, ELIZABETH T. JACOBS,* CARLOS ENCINAS DOMINGUEZ, MICHELLE L. THATCHER, AND MARK R. HAUSSLER Department of Biochemistry and Molecular Biophysics, and *Arizona Cancer Center; College of Medicine, University of Arizona, Tucson, Arizona I. Introduction II. Gene Targets and Biological Actions of the Vitamin D Receptor III. The Vitamin D Receptor as a Member of the Nuclear Receptor Superfamily IV. Structure-Function of the Vitamin D Receptor

V. Mechanisms of Vitamin D Receptor-Mediated Control of Gene Expression VI. Implications of Vitamin D Receptor-Mediated Signaling for Human Health and Disease VII. Summary and Perspectives References

I. INTRODUCTION

found to be a 3.3–3.7S protein (depending on species) that bound 1,25(OH)2D3 with high affinity (Kd = 0.1–1 nM) [8,9] and displayed a pharmacologic profile for binding various D metabolites and analogs that is consistent with their relative biological activities [10,11]. Avian intestinal VDR was shown to be a DNA-binding protein, and this property was exploited in the first purification of VDR [12]. An enriched avian VDR preparation, containing one major and several minor proteins of molecular mass 50–70 kDa [12,13], was then utilized to generate monoclonal antibodies to the VDR that cross-reacted with all species of vitamin D receptor tested [14]. Finally, via monoclonal antibody-based screening of a λGT-11 expression vector library, a partial cDNA clone of avian VDR was obtained [15], revealing the presence of the classical (Cys2–Cys2)2 zinc finger motif of the chromatinbound steroid hormone receptors [16]. Independent biochemical evidence demonstrated the importance of reduced cysteine thiol groups for DNA association [17] and implicated zinc in maintaining the DNA binding integrity of VDR [18]. Subsequently, fulllength human [19], rat [20], and avian [21] VDRs were cloned and their sequences deduced, allowing for overexpression and functional studies of the type discussed in this chapter and elsewhere in this volume. It is notable that the cloning and classification of VDR

A. Historical Aspects and Overview of the Vitamin D Receptor The existence of the vitamin D receptor (VDR) was originally appreciated when it was observed that administration of radioactive vitamin D3 at physiologic doses to rachitic chickens elicited the selective nuclear localization of a bioactive metabolite of the parent vitamin in the target intestine [1]. The binding of this D-metabolite in the nucleus was specific, saturable, and confined to the deoxynucleoprotein chromatin subfraction [1]. Subsequent salt extraction experiments resulted in the discovery of the chromosomal protein that comprises the vitamin D receptor [2]. The vitamin D metabolite associated with VDR was shown to be more polar than 25-hydroxyvitamin D3 (25(OH)D3) [3,4] and at least five times as potent and more rapidly acting than either vitamin D3 or 25(OH)D3 [4]. These results set the stage for the chemical identification [5,6] of the VDR ligand as 1α,25-dihydroxyvitamin D3 (1,25(OH)2D3) (Fig. 1), a metabolite produced mainly in the kidney [7]. When isotopically labeled 1,25(OH)2D3 could be produced in quantity, biochemical and pharmacological characterization of VDR was accomplished. VDR was VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

220

G. KERR WHITFIELD

VDR

ET AL .

RXR OH

HO

OH

1α,25-Dihydroxyvitamin D3

COOH

9-cis-Retinoic acid

FIGURE 1 Chemical structures of the classic natural ligands for the protein subunits of the VDR–RXR heterodimer. 1α,25(OH)2D3 is a high-affinity ligand for VDR (Kd < 1 nM) [8,9]. The vitamin A metabolite 9-cis-retinoic acid is a natural ligand for RXR with a Kd of approximately 10 nM [26].

as a DNA-binding transcription factor activated by ligand not only validated the discovery of VDR as a chromosomal protein [2], but also verified the initial report of its uncharacterized hormonal ligand localized in nuclear chromatin [1].

a heterodimer of the 1,25(OH)2D3-liganded VDR and unoccupied RXR.

B. The Vitamin D Receptor Signals as a Binary Complex with Retinoid X Receptor

A. Vitamin D Metabolism and Actions in Target Tissues

More recent research on VDR included a key demonstration [22–24] that high-affinity and specific DNA binding by VDR requires the presence of a co-receptor, now known to be any of the three retinoid X receptor (RXR) isoforms found in human tissues [25]. Furthermore, it was demonstrated that VDR-RXR heterodimer activation of transcription occurs on a DNA responsive element in the natural context of the rat osteocalcin promoter in intact bone cells, and that 9-cis-retinoic acid (9-cis RA) (Fig. 1), one of the natural ligands of the RXR partner [26], operates, at least in the context of this promoter, to suppress VDR-mediated stimulation of transcription by 1,25(OH)2D3 [27]. Therefore, RXR is the functional partner of VDR in a binary heterocomplex that is switched positively by the vitamin D hormone and negatively by 9-cis RA or a synthetic RXR-specific ligand [27–32]. Apparently, 9-cis RA attenuates 1,25(OH)2D3 responsiveness by diverting RXRs away from VDR-mediated transcription toward RXR homodimers and other RXR-dependent transcriptional pathways [33,34] (see Section V,A). From these and other studies with different vitamin D responsive genes (reviewed by Haussler et al. [35]), it appears that the active signal transduction complex that mediates the biological effects of vitamin D is

II. GENE TARGETS AND BIOLOGICAL ACTIONS OF THE VITAMIN D RECEPTOR

An overview of the formation of 1,25(OH)2D3 and its actions in vertebrates with a mineralized skeleton is presented in Fig. 2. The hormonal precursor and parent compound, vitamin D3, can either be obtained in the diet or formed from 7-dehydrocholesterol in skin via a nonenzymatic, UV light-dependent reaction. Vitamin D3 is then transported to the liver, where it is hydroxylated at the 25 position of the side chain to produce 25(OH)D3, the major circulating form of vitamin D3. The final step in the production of the hormonal form occurs mainly in the kidney, via a tightly regulated 1α-hydroxylation reaction (Fig. 2). The cytochrome P450-containing (CYP) enzymes that catalyze 25- and 1α-hydroxylations are mitochondrial CYP27A1 and CYP27B1, respectively, as well as a newly characterized microsomal 25-hydroxylase, CYP2R1 [36]. These reactions are discussed in more detail in Chapters 4–5 of this volume. As depicted in Fig. 2, 1,25(OH)2D3 can circulate to various target tissues to exert its actions which, in the conception of the authors of this chapter, are largely mediated by the nuclear VDR and its RXR heterodimeric partner. Many of the long-recognized functions of 1,25(OH)2D3 involve the regulation of calcium and phosphate metabolism, raising the blood levels of these ions to facilitate bone mineralization. Thus, in the small

221

CHAPTER 13 Nuclear Vitamin D Receptor

1,25 VDR Skin

Macrophage

Involucrin

Liver CYP27A1/2R1

Immune system

24-hydroxylated D Metabolites

Circulating 25(OH)D3

VDR Circulating 1α,25(OH)2D3

25(OH)D3 1α,25(OH)2D3 CYP27B1 −

PTH

Diet

1,25 Carbonic VDR anhydrase

1,25

CYP24

Differentiation

Vitamin D3

Kidney

Vitamin D catabolism

Keratinocyte

Ca2+

1,25

To nuclear vitamin D receptor

PO3− 4

VDR

1,25 +

VDR

Blood Ca•PO4

Parathyroid

Absorption Other 1,25(OH)2D3VDR target systems: • Central nervous • Endocrine (thyroid, pancreas, etc.) • Reproductive tract • Muscle • Some tumors

Osteoclast precursor

Reabsorption

Ca2+

1,25 VDR

Mineralization of collagen

PO3− 4 Small intestine

Bone Osteoblast

1,25 VDR

Resorption OPG decoy − 1,25 receptor RANKL VDR

+ IL-6

Osteoblast

+

RANK RANKL

Low Ca2+

PTH Synthesis − 1,25 VDR

c-fms

M-C

SF

Osteocalcin Osteopontin + Cell differentiation and fusion

PO3− 4 Ca2+ Osteoclast

FIGURE 2 Vitamin D metabolism and its biologic actions via the nuclear VDR. Depicted at the upper left are the sources of vitamin D3, which are either dietary or via a UV light-mediated photolysis of 7-dehydrocholesterol in the epidermis (see Chapter 3). The central bar (shaded) depicts the enzymatic conversion of vitamin D3 to the 25-hydroxy and then to the 1α,25-dihydroxy derivative (1α,25(OH)2D3), which is the active, hormonal form. These and subsequent hydroxylation reactions are catalyzed by CYPs (see text). The 1,25(OH)2D3 hormonal form (shown as 1,25 inside a white semicircle) then circulates to the various target tissues (selected target tissues identified in italics), where it binds to the nuclear VDR (depicted in white letters inside a black semicircle). Examples of target genes that are either transcriptionally activated (no symbol shown) or repressed (– symbol in a circle) are listed inside boxes.

intestine, 1,25(OH)2D3 binds to VDR, eliciting regulation of genes involved in calcium and phosphate absorption. In the kidney, 1,25(OH)2D3 conserves blood calcium and phosphate by promoting their reabsorption from the glomerular filtrate. The parathyroid gland also expresses VDR, and when the receptor is liganded with 1,25(OH)2D3, parathyroid hormone (PTH) synthesis is suppressed by a direct action on gene transcription [37]. This negative feedback loop, which limits the stimulation of CYP27B1 by PTH under low calcium conditions (Fig. 2), serves to curb the bone-resorbing effects of PTH in anticipation of 1,25(OH)2D3-mediated increases in both intestinal calcium absorption and bone resorption, thus preventing hypercalcemia. The direct boneresorbing effects of 1,25(OH)2D3 are mediated first in osteoblasts, where the liganded VDR promotes transcription of the gene encoding receptor activator of NF-κB ligand (RANKL) [38]. RANKL, displayed on the surface of osteoblasts, is detected by a cognate receptor

(RANK) on osteoclast precursor cells to promote their differentiation into bone-resorbing osteoclasts. This process also requires the liganding of c-fms, an osteoclast precursor surface receptor for macrophage colony stimulating factor (M-CSF). As illustrated in Fig. 2, 1,25(OH)2D3-triggered osteoclastogenesis can be inhibited by osteoprotegerin (OPG) in its capacity as a decoy receptor for RANKL [39]. Other aspects of bone biology regulated by 1,25(OH)2D3 shown in Fig. 2 include the transcriptional activation in osteoblasts of the genes encoding osteocalcin [40] and osteopontin [41], two proteins involved in bone remodeling, as well as the paracrine activation of existing osteoclasts via induction of osteoblast-derived cytokines such as IL-6 [42]. In addition to bone mineral homeostasis, 1,25(OH)2D3 also has effects on differentiation of certain cell types in skin [43] and in the immune system (Fig. 2) [44]. Examples of genes regulated in these tissues are involucrin in keratinocytes [45] and carbonic anhydrase

222 in macrophages [46]. Interestingly, the skin and the immune system are now recognized as extrarenal sites of CYP27B1 action to produce 1,25(OH)2D3 locally for autocrine and paracrine effects [47,48] (not shown in Fig. 2; see also Chapter 5). As depicted in Fig. 2 using the kidney as an example, an important mechanism by which the 1,25(OH)2D3VDR-mediated signal is terminated in all target tissues is the catalytic action of CYP24, an enzyme that initiates the process of 1,25(OH)2D3 catabolism (see Chapters 6 and 7). The CYP24 gene is transcriptionally activated by 1,25(OH)2D3 [49,50]. In addition, the 1α-hydroxylase (1α-OHase) CYP27B1 gene appears to be repressed by 1,25(OH)2D3 in a short negative feedback loop to limit the production of 1,25(OH)2D3 [51]. Sections III and IV of this volume contain a more extensive discussion of 1,25(OH)2D3 target tissues, which, as indicated in the box at the lower left of Fig. 2, include numerous cell types in addition to small intestine, bone, kidney, parathyroid, skin, and the immune system. Finally, insights into novel roles for the VDRRXR heterodimer in the colon and the hair follicle are discussed in Section VI of this chapter.

B. VDR-Regulated Genes and Vitamin D-Responsive Elements The vitamin D target cells/tissues discussed above all express significant levels of VDR and one or more of its RXR heteropartners. After liganding with 1,25(OH)2D3, the next step in the bioactivity of VDRRXR is the recognition of DNA sequences in the genes that are transcriptionally regulated. Table I displays the sequences of vitamin D-responsive elements (VDREs), which are binding sites for the VDR-RXR heterodimer in the 5′ upstream regions of a variety of vitamin D target genes. Many of the VDREs listed in Table I consist of a direct hexanucleotide repeat with a spacer of three nucleotides (DR3); such VDREs can mediate either positive (upper panel) or negative (lower panel) transcriptional regulation of target genes. In positive DR3 VDREs, VDR has been shown to occupy the 3′ halfelement, with RXR residing on the 5′ half-site [28]. Below the positive DR3 VDRE sequences in Table I is shown an “optimal” VDRE that was experimentally determined via binding of randomized oligonucleotides to a VDR-RXR heterodimer [52,53]. The conclusion from random selection is in general agreement with the repertoire of natural VDREs and defines the optimal VDRE as a direct repeat of two six-base half-elements that resemble estrogen responsive element (ERE) halfsites, i.e., AGGTCA, separated by a spacer of three nucleotides, denoted as a DR3. The highest affinity

G. KERR WHITFIELD

ET AL .

3′ (VDR) half-site is PGTTCA, where P is a purine base, and the highest affinity 5′ (RXR) half-site is PGGTCA. In addition, the random selection results suggest that the guanine at position 3 of the spacer (shown in lowercase in Table I) is important for VDR binding, an observation that is consistent with the finding [23,54] that this base is partially protected by RXR-VDR in methylation interference assays. In addition, a positive VDRE from the proximal region of the human CYP3A4 gene (central panel, Table I) represents an unusual configuration in that the two hexanucleotide half-element repeats are everted relative to one another, with a spacer of six base pairs (ER6) [55,56]. However, an examination of the 5′ halfsite complement, namely gAGTTCA, reveals that it is a perfect match to the 3′ VDR binding half-site in the optimal DR3 VDRE, including the first base pair of the spacer. Likewise, the 3′ half-site of the ER6 VDRE represents a perfect match to the 5′ RXR binding site in the optimal DR3 VDRE. Whether the ER6-bound VDR and RXR can interact with each other in the same fashion as on a DR3 element, and, indeed, whether VDR and RXR occupy the expected positions in the ER6, are questions for future research. Nevertheless, it should be emphasized that the use of an ER6 element by VDR has precedent with other receptors. In fact, this same element in the CYP3A4 gene has been shown to function as a responsive element for the pregnane X receptor (PXR), which also binds DR3 elements [57–59]. ER6-type elements also bind thyroid hormone receptor (TR)-RXR heterodimers [60,61], for which the normal responsive elements in other genes are direct repeats with a spacer of four nucleotides (DR4s) [60]. In contrast to the unique ER6 element, in which both half-elements coincide with the randomly selected sequences, the half-sites that exist in natural DR3 elements usually contain one to three bases that do not match the optimal VDRE. These variant bases occur with approximately equal distribution between the two half-sites. The variant bases, shown with single underline in Table I and tabulated below the random selection DR3, are usually purines. One notable exception is a thymidine in the third position of the 5′ (RXR) half-site, actually causing this site to match the optimal 3′ (VDR) half-site. The multiple sequence variations in natural VDREs may provide a spectrum of affinities for the VDR-RXR heterodimer, thus enabling these elements to respond to differing concentrations of the receptors (or their ligands) [62]. Another possibility, for which increasing evidence is accumulating, is that variant VDRE sequences induce unique conformations in the VDR-RXR complex, thereby promoting association of the heterodimer with distinct subsets of coactivators [63],

223

CHAPTER 13 Nuclear Vitamin D Receptor

TABLE I DNA Sequences of Selected Natural VDREs from Vitamin D-Regulated Genes Tissue/cell

Type of VDRE

5′-Half

Spacer

3′-Half

Reference

Intestine Intestine Keratinocyte Kidney Many targets Many targets Myelomonocytic line U937 Osteoblast Osteoblast Osteoblast Osteoclast precursor, macrophage Osteoclast precursor, macrophage

Positive DR3 (gene) human CYP3A4 distal rat CYP3A23 human involucrin human NPT2 rat CYP24 proximal rat CYP24 distal human p21 human osteocalcin rat osteocalcin mouse osteopontin chicken carbonic anhydrase chicken integrin β3

GGGTCA AGTTCA GGCAGA GGGGCA AGGTGA GGTTCA AGGGAG GGGTGA GGGTGA GGTTCA AGGGCA GAGGCA

gca tga tct gca gtg gcg att acg atg cga tgg gaa

AGTTCA AGTTCA GGCAGA AGGGCA AGGGCG GGTGCG GGTTCA GGGGCA AGGACA GGTTCA AGTTCG GGGAGA

[56,98,108] [56,316] [45] [317] [49] [50] [88] [318] [40,319] [41] [46] [320]

Random selection Common deviations Uncommon deviations

Optimal DR3 in positive DR3s in positive DR3s

PGGTCA –– TGG– –ACAAG

nng nna nnt

PGTTCA ––GG–G ––CAG–

[52,53]

Intestine

Positive ER6 (gene) human CYP3A4 proximal

TGAACT caaagg AGGTCA

[55,56]

Parathyroid Parathyroid Osteoblast Osteoblast Kidney Tumors, other targets Tumors, other targets (Pre-)osteoblast

Negative DR3 (gene) human PTH chicken PTH mouse osteocalcin rat bone sialoprotein PKA inhibitor rat PTHrP proximal rat PTHrP distal rat Cbfa1

GGTTCA GGGTCA GGGCA A AGGGTT ATGTTG AGGTTA GGTGGA AGTACT

[37] [321] [322] [323] [324] [301] [302] [325]

aag gga atg tat ctg ctc gag gtg

CAGACA GGGTGT AGGACA AGGTCA AGGTCA AGTGAA GGGTGA AGGTCA

Each VDRE consists of 5′ and 3′ half-elements (column 3) that bind RXR-VDR and mediate transcriptional regulation of genes (column 2) in target cells (column 1). Bases in positive VDREs that differ from the optimal sequence obtained by random selection experiments (center row) are single underlined, unless they are unique to negative VDREs, in which case they are double underlined. P denotes a purine nucleotide, i.e., A or G; n is any base.

or permitting differential actions in the context of diverse tissues [62]. When the DR3 sequences that mediate negative transcriptional regulation are examined, it is apparent that these, too, display a considerable degree of deviation from the optimal VDRE sequence. Not surprisingly, the base differences occurring in negative VDREs do not correspond to those found in positive VDREs. Those variant bases that exist only in negative VDREs are double underlined in Table I and often include pyrimidines, in addition to the predominant purine substitutions that occur in both positive and negative

VDREs (single underlined). Exactly how the uniquely negative acting bases influence the VDR-RXR heterodimer to mediate transcriptional repression is not known. However, Koszewski and colleagues [64] were able to demonstrate that the negative VDRE in the chicken PTH gene can be converted into a positive element simply by changing the most 3′ two bases from GT to CA. The success of this experiment implies that the VDRE sequence alone can determine whether the bound VDR-RXR heterodimer will activate or repress transcription. These results can also be interpreted as further evidence that the VDRE sequence can induce

224 significant conformational changes in the VDR-RXR heterodimer that are sufficient to promote binding to corepressors, rather than coactivators. Another novel aspect of negative VDREs is that the variant bases are not equally distributed between halfsites (Table I). Thus, PTH genes have unique negative variations exclusively in the 3′ element, while other negatively regulated genes have variations primarily in the 5′ half-element. The significance of this asymmetry for negative regulation, or even possibly the polarity of VDR and RXR binding to the two half-sites, is not currently known. Clues may be found by examining the all-trans-retinoic acid receptor (RAR)-RXR heterocomplex, which acquires ligand-dependent repressor activity when associated in reverse polarity (RXR on the 3′ half-site) with a direct repeat with a spacer of one nucleotide (DR1) retinoic acid responsive element (RARE) [65]. It should be noted that one negative VDRE in Table I, namely the distal VDRE in the rat parathyroid hormonerelated peptide (PTHrP) gene, has no base difference that would identify it as a negative element, i.e., all variations observed in this VDRE are also evident in positive VDREs. However, this same gene possesses a proximal VDRE that contains three base variants occurring only in negative VDREs (Table I). This observation highlights another area that is not well understood with regard to VDREs, namely the means by which multiple VDREs in the same gene function together to effect transcriptional regulation. The fact that three 1,25(OH)2D3-regulated genes (CYP24, CYP3A4, and PTHrP) with multiple VDREs have been characterized to date suggests that probing the functional interaction of two or more VDREs in the same gene is an important area for further study. The conclusion, from analyzing a number of VDREs, is that the docking sequence for VDR-RXR on DNA consists of clearly defined DR3 and ER6 motifs in the promoter region of vitamin D controlled genes. Regulation of the expression of many of these genes by 1,25(OH)2D3 in traditional vitamin D target tissues such as kidney, bone, parathyroid, skin and the hematopoietic system has been well established [66], and the combination of Fig. 2 and Table I paints a picture of the biological breadth of VDR-mediated gene control. Obviously, the 1,25(OH)2D3-VDR/RXR complex modulates the expression of a multitude of genes, and recent microarray studies [67–71] suggest that the number of vitamin D-regulated genes is much greater than the well-characterized bioresponses discussed herein. As described next, the use of VDR null animals has both validated the obligatory role of VDR in 1,25(OH)2D3 signaling, and revealed which VDR responses are the most pathophysiologically relevant.

G. KERR WHITFIELD

ET AL .

C. Consequences of VDR Ablation in Rodents and Humans Important insight into the biological impact of VDR to mediate the myriad of gene regulatory effects of 1,25(OH)2D3 discussed earlier can be gained by examining the phenotype of mice in which the VDR has been ablated (VDR knockouts), as well as the phenotype of human patients with hereditary hypocalcemic vitamin D-resistant rickets (HVDRR). HVDRR is a consequence of defects in the VDR gene on human chromosome 12 [72–75]. Clinically significant HVDRR is an autosomal recessive disorder resulting in severe bowing of the lower extremities, short stature, and often alopecia [74]. The serum chemistry in HVDRR reveals hypocalcemia, secondary hyperparathyroidism, elevated alkaline phosphatase, variable hypophosphatemia, and markedly increased 1,25(OH)2D3. Many of these sequelae are the expected manifestations of the loss of 1,25(OH)2D3 regulation in the target genes outlined above, such as PTH, RANKL, and CYP24. Most aspects of HVDRR, with the exception of alopecia, mimic those of classic vitamin D-deficiency rickets, suggesting that VDR not only mediates all of the bone mineral homeostatic actions of vitamin D, but may also participate in hair follicle function in a manner independent of vitamin D status. As detailed later in this chapter, individual natural VDR mutations observed in HVDRR have been very instructive in the elucidation of structure/ function relationships within the receptor, especially with regard to DNA, ligand, and coactivator binding. Several strains of homozygous VDR knockout mice have been created [76–78], and their phenotype has been characterized under various conditions. Somewhat surprisingly, VDR−/− mice, except for hair loss, appeared phenotypically normal until weaning, after which they began to display the suite of symptoms characteristic of HVDRR, namely hypocalcemia, hyperparathyroidism, and elevated 1,25(OH)2D3 levels [76,77]. One early observation with VDR knockout animals was that most aspects of the VDR−/− phenotype, except for alopecia, could be reversed if blood calcium levels were normalized by dietary manipulations [79]. The so-called “rescue diet” contained calcium, phosphorus, and lactose supplements to restore crucial bone mineral ions not properly assimilated after weaning [80]. The successful rescue, in relation to bone mineral homeostasis, of VDR-null mice suggests that the dominant biological action of 1,25(OH)2D3-VDR is to promote intestinal calcium and phosphate absorption, especially when there is physiologic stress from the requirement for bone mineralization under conditions of low calcium availability. With respect to alopecia, it was observed that lack of VDR prevented hair follicles from entering the anagen phase of the hair cycle [81], but transgenic

CHAPTER 13 Nuclear Vitamin D Receptor

VDR expression targeted to keratinocytes in the background of a VDR-null animal was successful in restoring hair cycling [82]. These results not only confirmed the essential role of VDR in maintaining hair (and skin) health, but also pinpointed keratinocytes as an important cellular target for VDR action in skin. More recent studies using VDR−/− animals fed the rescue diet to normalize calcium have revealed additional roles of VDR that are not simply secondary effects of maintaining serum calcium and phosphate levels. Separate studies have found that VDR−/− mice, when compared to normal mice: (a) exhibit uterine hypoplasia [77], probably linked to reduced expression of aromatase (CYP19) [83]; (b) display hyperplasia of the mammary glands [84]; (c) have blunted insulin synthesis in pancreatic beta cells [85]; (d) have reduced synthesis of macrophage IL-18 and STAT4 in T helper subset type 1 (Th1) lymphocytes, leading to an impaired Th1 response [86]; and (e) are markedly more sensitive to the skin carcinogen 7,12-dimethylbenzanthracene [87]. These exciting recent results indicate that beyond bone mineral integrity and hair cycling, VDR signaling significantly affects development in certain tissues as well as the function of the endocrine pancreas. VDR action is clearly immunomodulatory, and VDR likely mediates the reported anticancer effects of vitamin D. In the context of chemoprevention, 1,25(OH)2D3 is known to possess potent antiproliferative, prodifferentiation, proapoptotic, and/or cell cycle arrest activities in epithelial cells [56]. The molecular mechanisms underlying these effects may be related to the ability of liganded VDR to arrest cells at the G1 stage via induction of cell cycle regulatory proteins such as p21 [88] (Table I) and p27 [89], to control cell growth transcription factors such as c-myc [90] and c-fos [91], or to elicit apoptosis by repression of the Bcl-2 antiapoptotic factor [92]. However, an additional intriguing possibility is that VDR is chemopreventative by inducing detoxification enzymes, e.g., CYP3A, since several of the genes encoding these CYPs in mammals possess VDREs (Table I). Thus, as discussed in depth later in this chapter, a newly recognized action of liganded VDR is that of inducing CYPs for xenobiotic detoxification.

III. THE VITAMIN D RECEPTOR AS A MEMBER OF THE NUCLEAR RECEPTOR SUPERFAMILY A. VDR in the Context of the 48 Human Nuclear Receptors A proper understanding of the molecular actions of VDR requires an appreciation of its genetic placement within the nuclear receptor superfamily. The human

225 genome sequence [93] encodes 48 nuclear receptors [94], and interrelationships between these receptors are illustrated in Fig. 3. All human nuclear receptors, along with those from other species, have been placed into six groups according to a unified system of nomenclature [95]. As with previous analyses using a subset of human receptors [95], VDR groups with the TRs, RARs, the peroxisome proliferator-activated receptors (PPARs), the reverb orphan receptors, and the retinoic acid receptor-related orphan receptors (RORs) in a single clade that has been named group 1 [95]. As illustrated on the right side of Fig. 3, a significant feature shared by group 1 receptors is that many of them form active heterodimeric complexes with the RXRs from group 2 [96]. The closest relatives of VDR within group 1 are PXR and the constitutive androstane receptor (CAR), which, with VDR, form subgroup 1I (Fig. 3) (full designations are NR1I1 for VDR, NR1I2 for PXR, and NR1I3 for CAR) [95]. The next closest group of receptors to VDR is subgroup 1H, including the liver X receptors (LXRs) and the farnesoid X receptor (FXR). One notable feature of the 1H and 1I receptors is the ability of these receptors to recognize multiple ligands [97]. The diversity of ligands for PXR is especially broad and includes not only endogenous steroids, but also an array of other lipophilic compounds such as the secondary bile acid lithocholic acid (LCA) and the antibiotic rifampicin, as well as xenobiotics such as hyperforin, the active ingredient of St. John’s wort [97]. Until recently, VDR was thought to possess a single ligand, the 1,25(OH)2D3 hormone. However, it is now known that mammalian VDR has at least two additional ligands (Fig. 4), namely LCA and its derivative, 3-ketolithocholic acid (3-ketoLCA) [98]. This finding is of considerable interest to human medicine, since it is known that LCA is a secondary bile acid with significant carcinogenic potential [98–100], and its binding to VDR could trigger the detoxification of this ligand via the induction of CYP3A4 mediated by the VDREs in this gene (Table I). The structures of both LCA and 3-ketoLCA are shown in Fig. 4, along with the structures for prototypical PXR, FXR, and LXR ligands. It is evident when comparing these compounds that the ligand binding profile of VDR, expanded to include LCA and its 3-keto derivative, now resembles the ligand profile of the closest VDR relatives in the nuclear receptor superfamily, especially when it is noted that both PXR [101,102] and FXR [98,103] are also activated by LCA to some extent. It is now appreciated that most group 1I and 1H members, with the exception of CAR and the two LXRs, can recognize bile acid ligands. Therefore, based upon an analysis of its evolutionary position among the nuclear receptors and recent data

226

G. KERR WHITFIELD

TRα 1A TRβ RARα RARβ 1B RARγ PPARα PPARβ 1C PPARγ reverbα 1D reverbβ RORα RORβ 1F RORγ VDR PXR 1I CAR LXRβ LXRα 1H FXR HNF4 2A HNF4γ RXRα 2B RXRβ RXRγ TR2 2C TR4 TLX 2E PNR COUP COUP2 2F EAR2 DAX 0B SHP GCNF1 6A ERα 3A ERβ ERRα ERRβ 3B ERRγ GR MR 3C PR AR FTZ1 5A LRH1 NURR1 4A NOR1 NGFIB Outgroup (Cnidarian)

RXR

RXR

RXR

Group 1

RXR

RXR

RXR RXR

Group 2

nb

Group 0

or

Group 6

Group 3

Group 5 Group 4

AmNR2

FIGURE 3 Human nuclear receptors categorized into subgroups (0B-6A) according to the unified nomenclature scheme [95] with subgroup classifications shown to the right of the receptor abbreviations. This cladogram was created in Clustal W [326] using protein sequences obtained from GenBank. Abbreviations for Group 1 receptors and for RXR are defined in the text, and accession numbers are taken from Whitfield et al. [104]. Other abbreviations are HNF4, hepatocyte nuclear factor-4; TR2/4, testis-specific receptors 2/4; TLX, human homolog of Drosophila tailless; PNR, photoreceptor-specific nuclear receptor; COUP, chicken ovalbumin upstream promoter transcription factor; EAR2, erbA-related receptor-2; DAX, DSS-AHC region on human X chromosome; SHP, small heterodimer partner; GCNF1, germ cell nuclear factor-1; ERα/β, estrogen receptors α/β; ERRα/β/γ, estrogen-receptor related receptors α/β/γ; GR, glucocorticoid receptor; MR, mineralocorticoid receptor; PR, progesterone receptor; AR, androgen receptor; (Continued)

ET AL .

227

CHAPTER 13 Nuclear Vitamin D Receptor

VDR COOH

COOH

O

HO Lithocholic acid

PXR

3-Ketolithocholic acid

CH3

FXR

C=O

COOH

OH

HO O 5β-Pregnane-3,20-dione

LXR

Chenodeoxycholic acid

OH O

HO HO 22(R)-Hydroxycholesterol

24,25-Epoxycholesterol

FIGURE 4 Novel VDR ligands compared with ligands for the related receptors PXR, FXR, and LXR. The top panel shows the structures of LCA and its derivative, 3-ketoLCA, which are newly recognized ligands for human VDR [98]. The central two panels display the structures of 5β-pregnane-3,20-dione and chenodeoxycholic acid (CDCA), which have been reported as natural ligands for PXR [109] and human FXR [290,291], respectively. The nuclear receptors LXRα and LXRβ bind to oxysterols such as the two shown in the lower panel [327].

with bile acids, it is apparent that VDR binds ligands beyond the vitamin D congeners that regulate bone mineral homeostasis, and it is conceivable that VDR associates with novel lipophilic ligands to effect other extraosseous bioresponses (see Section VI,B).

FIGURE 3 Continued

B. New Insights from the Evolution of VDR Recently we cloned a high-affinity receptor for 1,25(OH)2D3 from the sea lamprey [104], which, along with other lampreys, hagfishes, and jawless fishes,

FTZ1, fushi tarazu 1 factor homolog; LRH1, liver receptor homolog-1; NURR1, nur related protein-1; NOR1, neural orphan receptor-1; NGF1B, nerve growth factor-induced protein-1B. AmNR2 is Acropora millepora nuclear receptor-2, a protein from coral used as an outgroup for constructing the cladogram. To the right of the subgroup classification is a schematic depiction of the typical configuration of the corresponding receptor (monomer or dimer) that binds to its cognate DNA responsive element. Many group 1 receptors, including VDR, bind as heterodimers with RXR (primary receptor shown as solid oval, with heterodimeric partner indicated as box containing the letters RXR). In addition, RXR forms homodimers on its responsive element. Many other receptors form homodimers on their respective responsive elements, illustrated as paired solid ovals. Receptors that bind DNA as monomers are depicted with single solid ovals. The group 0B receptors do not have a DBD and, hence, do not bind directly to DNA (indicated by the letters “nb”).

228 constitutes the most ancient lineage among extant vertebrates. Because lampreys lack both a calcified skeleton/teeth and hair, this was an unexpected finding given the dominant role of VDR in bone mineralization and hair cycling in mammals. Lamprey VDR, which displays approximately 60% amino acid sequence identity with other known vertebrate VDRs, is expressed in juvenile skin, mouth, and larval protospleen, but is not well expressed in juvenile intestine [104]. The low expression in intestine, combined with the fact that lampreys lack a calcified skeleton and teeth, strongly suggests that VDR is not a major factor in intestinal calcium absorption in this species. However, the presence of lamprey VDR in skin implies that VDR has some role in this tissue, analogous to the action of mammalian VDR in keratinocytes as discussed earlier. Another clue to the function of VDR in lampreys might be the observation that, among several VDREs tested, only the VDRE from the human CYP3A4 gene was able to confer significant 1,25(OH)2D3-dependent activation of a reporter gene [104]. These findings may offer insight into the function of VDR in the lamprey, indicating this evolutionarily ancient receptor could perform the role of inducing CYPs for xenobiotic detoxification, and likely provide a link to the present and a novel action of VDR akin to that of its close evolutionary cousins. Indeed, the common thread connecting the 1I receptors, VDR, PXR and CAR, is the induction of CYP enzymes that participate in xenobiotic detoxification, as summarized in Fig. 5A. A major target for VDR and PXR in humans is CYP3A4 [55,56,98], for which the detoxification substrates include LCA [105]. Initial studies focused on VDR liganded to 1,25(OH)2D3 as a regulator of CYP3A4 [55,56], but more recent experiments have revealed that LCA is also capable of binding VDR to up-regulate expression of human CYP3A4, or its equivalent in rats (CYP3A23) or mice (CYP3A11) [98]. There is also evidence that other CYP enzymes may be VDR targets [106]. In fact, many of the traditional targets of VDR-mediated regulation of vitamin D metabolism, namely 1α- and 24-hydroxylases, are CYPs (CYP27B1 and CYP24, respectively, according to standard CYP nomenclature) [107]. These studies therefore demonstrate that, in addition to its well-known role in calcium homeostasis, VDR also participates in the regulation of intestinal CYP genes in a manner similar to its closest relative, PXR [55,56]. As summarized in Fig. 5A, VDR and PXR may regulate some of the same genes, since PXR also utilizes DR3 and ER6 responsive elements in DNA for transcriptional activation [108]. The differential transcriptional effects of VDR and PXR may reside not in the genes that are regulated, but rather in the overlapping,

G. KERR WHITFIELD

ET AL .

yet distinct, ligand profile (Fig. 5A, second column); for example, PXR does not respond to 1,25(OH)2D3 [109]. It should be noted that CAR, the third human receptor in subgroup 1I, is also implicated in CYP regulation [110,111]. Whereas CAR normally binds DR4 elements (Fig. 5A), this relatively uncharacterized receptor may exhibit some cross over binding to DR3 elements; conversely, PXR may cross over to DR4 elements [112]. There is some preliminary evidence that VDR may also cross over to DR4 elements [106], as it was reported that VDR can activate transcription from at least two elements that strongly resemble degenerate DR4s [38,113]. Another aspect to consider in the differential activities of VDR, PXR, and CAR is the tissue distributions of these receptors. In humans, VDR is expressed in a large number of tissues (reviewed in Haussler et al. [35]; see also Fig. 2), including small intestine, colon, kidney, skin, certain immune cells (e.g., macrophages and Th1 lymphocytes), prostate, pancreas, ovary, and heart, but is present only in very low levels in mammalian liver [114]. The expression patterns of PXR and CAR are more limited; PXR is found predominantly in liver, small intestine, and colon [59], and CAR is restricted mainly to liver [115]. Thus, it would appear that VDR likely plays a significant role in CYP induction in intestine, colon, kidney, skin, and possibly also prostate and ovary, but defers to PXR and CAR in liver, a major site of xenobiotic detoxification and excretion. VDR shares fewer common characteristics with the more distantly related subgroup 1H receptors. As with PXR and CAR, liver is a predominant site of action for FXR and LXRα, along with other tissues such as adipose [116,117]. However, in contrast to the VDR/PXR/CAR grouping, xenobiotic detoxification does not appear to be a primary function of these receptors. Rather, as shown in Fig. 5A, FXR and the LXRs appear to be important regulators of cholesterol metabolism to bile acids and transport of cholesterol metabolites [118,119]. The general conclusion based upon examination of the evolutionary position of VDR among the 48 nuclear receptors is that it is extremely closely related to PXR, both structurally and functionally. In one respect, VDR appears to have evolved as a “specialty” regulator of intestinal calcium absorption and hair growth in terrestrial animals, providing both a mineralized skeleton for locomotion in a calciumscarce environment, and physical protection against the harmful UV radiation of the sun. Yet VDR also has retained its PXR-like ability to effect xenobiotic detoxification via CYP induction. VDR could complement PXR by serving as a guardian of epithelial cell integrity, especially at environmentally or xenobiotically exposed sites such as skin, intestine, and kidney.

229

CHAPTER 13 Nuclear Vitamin D Receptor

IV. STRUCTURE-FUNCTION OF THE VITAMIN D RECEPTOR VDR possesses the classical (Cys2–Cys2)2 zinc finger motif for DNA binding, and a multifunctional C-terminal domain for ligand binding, RXR heterodimerization, and transcriptional coactivator binding, but lacks the long N-terminal extension typical of the traditional steroid hormone receptors [120]. Figure 5B displays in schematic form the structural similarities between the group 1I and 1H receptors, highlighting the major functional domains for DNA-, ligand-, and heterodimer-binding. The percent identities within these functional domains mirror the evolutionary classification of these receptors, with PXR showing the highest identities to VDR in all three functional regions. The group 1H receptors (LXR and FXR) exhibit identities that are somewhat lower in the DNA binding and heterodimerization domains when compared to VDR, but display relatively strong identities in the ligand binding domain (LBD) (at 27–34%, approximating that of PXR and CAR), perhaps reflecting the fact that all 1I and 1H receptors recognize cholesterol derivatives with intact (or nearly intact) side chains (see Fig. 4). The human thyroid hormone receptor α (TRα) and human RXRα are included in Fig. 5B for comparison. Note that amino acid identities remain relatively high in the DNA binding and heterodimerization domains, even when VDR is compared to RXR, a group 2 receptor. However, identities in the LBD decline precipitously when TRα and particularly RXRα, which displays only 5% identity with VDR in the LBD, are compared to VDR. These low identities are likely a consequence of the fact that TR and RXR both bind nonsterol ligands. A unique feature of VDR and PXR that is probably related to their ligand binding profile is the fact that both of these receptors have especially lengthy amino acid segments between the DNA binding domain (DBD) and the highly conserved regions of the LBD (residues “missing” in the other receptors in this area relative to VDR are indicated with numbers preceded by a minus sign in Fig. 5B). The presence of additional residues following the DBD in VDR and PXR has been proposed to confer upon these two receptors more spacious binding pockets for accommodating a range of ligands, such as xenobiotics [121,122].

A. The Zinc Finger DNA Binding Domain of hVDR Throughout the nuclear receptor superfamily, the DBD is the region of highest sequence similarity, with

Ligands

A VDR 1I

PXR CAR

oxysterols

DR4

oxysterols

DR4

1° bile acids 1 (e.g., CDCA)

IR1,ER8

LXRα FXR

Genes

xenobiotic xenobiotics DR3,ER6 LCA, 3-ketoLCA (DR4, ER8) transport, CYPs xenobiotic DR4 xenobiotics transport, CYPs (DR3)

LXRβ 1H

Elements

1,25(OH)2D3 Ca++ homeostasis, LCA, 3-ketoLCA DR3,ER6 CYPs, hair cycle

cholesterol/lipid homeostasis cholesterol/lipid homeostasis bile acid homeostasis

Ligand binding Heterodimerization

B

VDR

DBD −5

+17

PXR

1I

−13

2B

RXRα

FIGURE 5

=34%

=41%

=30%

=41%

−46

=27%

=37%

−43

=16%

=35%

=5%

=31%

−47

−24

+86 51% +126

TRα

=49% −4

53%

FXR

1A

=27%

−56

+97

LXRα LXRβ

=50%

64%

CAR

1H

=35%

67%

56% +15 49% −73

+134 47%

Comparative features of VDR and closely related nuclear receptors from subgroups 1I and 1H. Abbreviations for each receptor are defined in the text. (A) Comparison of cognate ligands, DNA-responsive elements, and target gene function. Ligand abbreviations are defined in the text and the legend to Fig. 4. DNA responsive elements are classified as direct repeats with spacers of three or four nucleotides (DR3, DR4), elements in which the two half-site repeats are everted with a spacers of 6 or 8 nucleotides (ER6, ER8), or inverted repeats with a spacer of one nucleotide (IR1). Under target genes, CYP refers to cytochrome P450containing enzymes that are involved in detoxification of xenobiotics, or in the case of the CYP24 gene, degradation of 1,25(OH)2D3 and 25(OH)D3. (B) Schematic representations of each receptor from panel A along with the human TRα and RXRα. The subgroup designation of each receptor is listed at the left (see Fig. 3). The zinc finger DBD is indicated by a solid black box with white lettering that includes the percent amino acid identity in the DBD between the indicated receptor and human VDR. Other domains for which percent identities were computed are those that mediate ligand binding (three subdomains, lightly shaded) and heterodimerization (two subdomains indicated by hatched boxes). The remaining regions of the receptors are depicted as white boxes with numbers indicating a greater (+) or lesser (−) quantity of amino acids in those segments compared with human VDR.

230 identities ranging from 67% between VDR and PXR, to 47% between VDR and RXRα (Fig. 5Β). This high degree of similarity is a consequence of the fact that all nuclear receptors, with the exception of those that do not directly bind DNA, possess a DBD composed of two zinc fingers of the Cys2–Cys2 type. As illustrated schematically in Fig. 6A, the two zinc finger modules in hVDR are followed in the primary sequence by a C-terminal extension (CTE), which contains a long α-helix. These subregions are color coded in Fig. 6A for recognition in Fig. 6B–D (see color plate insert). The following subdomains are highlighted: a portion of the N-terminal domain (only two residues, 22 and 23, shown in dark blue), two zinc fingers (colored green and light blue, respectively) separated by an intervening region (brown), and a CTE shown in yellow. Nuclear localization signals are depicted in purple and naturally occurring inactivating mutations that alter amino acids in this region are denoted in red. Coordinated zinc atoms are shown as black circles with white lettering. The entire DBD, extending from positions 22 to 110 in the human VDR amino acid sequence, has been crystallized [123], and representative views of this region, constructed in Protein Explorer [124] using the published coordinates, are shown in Fig. 6B–D. It is important to consider that the hVDR DBD fragment was crystallized by Shaffer and Gewirth as a homodimer on three different DR3 VDREs [123]; the views shown here (Figs. 6B–D) are from VDR DBD fragment homodimers bound to the mouse osteopontin VDRE (Table I) [41]. While the native VDR functions not as a homodimer, but as an RXR heterodimer as described below and detailed elsewhere in this chapter, it is reasonable to assume that the DNA contacts established by homodimeric VDR DBD fragments approximate those of at least the core zinc finger region of the VDR partner when it binds to a VDRE as part of the full-length VDR-RXR heterodimer. Accordingly, as depicted in Fig. 6B, which illustrates only the 3′ half-element of the VDRE with the docked hVDR DBD fragment, the first zinc finger (green in color plate insert) of hVDR possesses an α-helix on the C-terminal side containing amino acid residues (specifically E42, K45, R49, and R50) that contact VDRE 3′ half-element bases in the major groove of DNA. Recent data [123,125] indicate that hVDR residues E42 and R49 make contact with the central two base pairs of the canonical AGTTCA VDRE 3′ half-element, whereas K45 and R50 contact the second and fifth G-C base pairs, respectively, in this half-element. Accordingly, the α-helix that contains these four hVDR base-pair contact residues (E42, K45, R49 and R50) is referred to as the specific DNA

G. KERR WHITFIELD

ET AL .

recognition α-helix and extends from E42 to K53. As depicted schematically in Fig. 6A, E42 forms part of a previously reported proximal box (P-box), which consists of three amino acids that distinguish the nuclear receptors which bind to AGGTCA ERE-type half-elements (EGG P-box) from those that bind to AGAACA GREtype half-elements (GSV P-box) [125]. Thus, exchange of only these three amino acids can switch ER homodimers to recognize GRE inverted repeats and, conversely, GR homodimers to recognize ERE inverted repeats [126]. However, for RXR heterodimerizing receptors such as VDR, an additional level of DNAbinding specificity is manifest, with an arginine at position 49 (Fig. 6A) dictating heterodimeric binding to direct repeat hormone responsive elements (homodimerizing nuclear receptors possess a lysine in the position corresponding to R49 in hVDR) [125]. Finally, in hVDR, the terminal residue of the DNA recognition α-helix, K53 (Fig. 6A), is also required for VDRE binding, and along with the CTE, confers VDR, as well as its closely related receptor PXR, with selective DR3 docking capacity [125]. Because heterodimerizing nuclear receptors such as VDR employ residues beyond the P-box to generate selective responsive element motif binding, we have grouped the highly conserved R49 along with the K53 residue in hVDR (or their positional equivalents in other receptors) into a new “specificity” box (S-box) [125]. The S-box therefore complements and extends the traditional P-box, and together they endow RXR heterodimerizing nuclear receptors with the capacity to specifically recognize their cognate direct repeat DNA responsive elements. An additional α-helix situated on the C-terminal side of the second zinc finger (blue in color plate insert) is positioned to interact with the DNA phosphate backbone (Fig. 6B), and there is extensive contact between residues at one end of this α-helix (specifically R73, R74, and R80) and DNA phosphates. The combined energy provided by the DNA recognition and phosphate backbone binding α-helices is apparently what is required for specific VDRE binding by VDR, and this conclusion is supported by the location of VDR point mutations found in patients with HVDRR [74] as illustrated in red in Fig. 6B (see color plate insert). All of these VDR-inactivating point mutations occur in the two zinc fingers, and many are localized to DNA base- or phosphate backbone-contact residues (e.g., K45, R50, R73, and R80). The consequences of the other HVDRR mutations in the DBD are less obvious, but still instructive. The planar side chain of phenylalanine 47 interacts extensively with the aliphatic portion of arginine 80 on one side and serine 51 on the other. Replacement of this residue with a branched chain amino acid (F47I) may

231

CHAPTER 13 Nuclear Vitamin D Receptor

A

DBD

CTE

D

Ligand binding/Heterodimerization

46 47

35

33

50

73

Ser51

Second Zn Finger

First Zn Finger

P-box E42 G43 G46 Zn 22

51 49-50 P 53-55

24 45 46 47

DNA Binding Mutations

Zn

82-83 102-104

60 80

50 R49 K53 S-box

89

C

Positively charged residues that mediate nuclear localization of hVDR

Natural mutations in hVDR that inhibit DNA binding and cause vitamin D resistance

B

110

CTE

Nuclear Localization

22

22 73

Zn

Zn Zn

82

33 45 35

46

Zn

83

N37

80 47

RXR

DNA 50

54

110 Heterodimeric orientation of CTE

110

DNA

K91 50 49 E92

Homodimeric orientation of CTE

110

103 102

55

53 104 103 102 104

Homodimeric orientation of CTE

110

Heterodimeric orientation of CTE

FIGURE 6 Structure–function relationships in the human VDR DBD (see color plate insert and text for details).

produce steric hindrances in this critical region. Glycines 33 and 46 are oriented toward the phosphate backbone; their mutation to negatively charged (and bulkier) aspartates (G33D and G46D) would likely lead to repulsion and/or steric hindrance, preventing proper DNA contact. Finally, histidine 35 forms a watermediated hydrogen bond complex with both a DNA phosphate and with lysine 45; its mutation to glutamine (H35Q) presumably disrupts these interactions. Therefore, with respect to the core zinc finger DBD of hVDR and its contact with the VDRE, the interpretation of the X-ray structure of the VDR-DNA cocrystal is consistent with the observed vitamin D-resistant rachitic and alopecic phenotype of the patients with DBD-inactivating VDR mutations. This is perhaps the most compelling evidence not only for the obligatory

role for VDR in vitamin D action, but also for the absolute requirement for DNA binding as a mechanism for VDR signaling. The depiction of the hVDR DBD fragment co-crystal with a VDRE in Fig. 6B includes two alternative representations for the positioning of the CTE. The CTE location in the VDR homodimer-VDRE structure is shown at the lower left [123]. An alternative positioning of the CTE, which is taken from a model of a VDRDBD/RXR-DBD heterodimer on a DR3 VDRE that was constructed by Rastinejad and colleagues [127] based on X-ray crystallographic coordinates of a TRDBD/RXR-DBD bound to a DR4 TRE, is illustrated at the lower right. The heterodimeric positioning of the CTE suggested by Rastinejad et al. [127] would be more consistent with a VDR-RXR heterodimer rather

232 than a VDR-VDR homodimer as the physiologic mediator of 1,25(OH)2D3 action [27,32,54,128–131]. The heterodimeric orientation of the VDR CTE is also more reconcilable with mutagenesis results [132] for the hVDR CTE, which indicate a crucial role for CTE residues in VDRE binding, including clusters of basic residues at amino acids 102–104 and 109–111. Although the X-ray crystallographic coordinates of the VDR/RXR heterodimer docked to a DR3 VDRE are not available, the model as presented by Rastinejad et al. [127] places the VDR CTE in a position nearly parallel to the DNA phosphate backbone, extending to the right in Fig. 6B in a 3′ direction relative to the VDRE sense strand. This heterodimeric positioning of the CTE would allow VDR basic residues in the 109–111 cluster to contact the DNA phosphate groups, unlike in the VDR homodimer cocrystal of Shaffer and Gewirth [123], in which the VDR CTE crosses the DNA phosphate backbone at nearly a right angle, and VDR residues 109–111 are placed too far from the phosphate backbone for any contact to occur. Therefore, pending the cocrystallization of a VDR-RXR heterodimer on a VDRE, we conclude [125] that, whereas the hVDR DBD fragment homodimer crystal [123] provides valid results for the DNA base and phosphate backbone contacts of the core zinc finger residues, it distorts the positioning of the CTE, which likely conforms instead to the heterodimeric model of RXR-VDR [127]. In order to bind to DNA and regulate gene expression, transcription factors require nuclear translocation after biosynthesis on the polyribosomes, and nuclear localization is dependent upon various arrangements of basic amino acids in short stretches within the primary amino acid sequence [133]. Four such segments of positively charged amino acids that mediate hVDR nuclear translocation [134,135] exist in the DBD as designated in purple in Fig. 6A (see color plate insert). Figure 6C contains the same view of the VDR DBD as in Figure 6B, but again highlights (in purple) the hVDR basic residues that mediate nuclear localization [134,135]. These residues align vertically on one surface of hVDR when the CTE is in the heterodimeric orientation, perhaps creating an interaction surface for nuclear import factors. Some of the same hVDR amino acids that contact DNA also participate in nuclear localization, namely the basic residue clusters at positions 49–50 [134] and 102–104 [135]. The coidentity of DNA binding and nuclear localization residues is not a general feature of transcriptionally active proteins, but has been reported previously for a variety of DNA- and RNA-binding proteins [136]. Two additional basic clusters are also involved in hVDR nuclear translocation, located at residues 53–55 [134] and 82–83 [135]. Exactly how these positively charged residues mediate

G. KERR WHITFIELD

ET AL .

nuclear localization is not known, although the fundamental nuclear membrane machinery, including importin, has been characterized for this general process [137]. Interestingly, Barsony and colleagues [138] have confirmed the essential role of hVDR residues 49–50 and 53–55 [134], and demonstrated that positionally equivalent basic residues in RXR are also important in nuclear localization, especially of the unliganded VDR-RXR heterodimer [138] (see also Chapter 22). Thus, it appears that a quasi-stable, hormone-unoccupied VDR-RXR heterodimer is the species of receptor that localizes in the nucleus of vitamin D target cells, probably bound nonspecifically to DNA with relatively low affinity. The weak association between unliganded VDR and its RXR partner may be facilitated by hydrogen bonding of N37 in hVDR to residues in the second zinc finger in RXR, as well as by salt bridges between K91 and E92 in hVDR to oppositely charged amino acids in the second zinc finger of RXR (N37, K91, and E92 are shown in orange in Fig. 6C (see color plate insert)). These three residues located on the 5′ side of the tertiary structure of the VDR DBD on the VDRE constitute a proposed [127] RXR interface (as suggested by the orange arrow, Fig. 6C) that is likely buttressed when the VDR-RXR heterodimer is bound to DNA. However, as discussed in the next section, a much stronger dimerization interface between VDR and RXR, that is 1,25(OH)2D3dependent, exists in the LBD of VDR. A final structure-function aspect of the VDR DBD relates to regulation of nuclear localization and DNA binding by phosphorylation. Serine 51, located near the C terminus of the first zinc finger and within the base recognition α-helix, is a site for phosphorylation by protein kinase C (PKC) [139]. Whereas some investigators have observed a potentiation of VDR activity after PKC activation in intact cells [140], probably as a consequence of phosphorylation of a VDR coactivator, results from our laboratory indicate that direct phosphorylation of human VDR by PKC at serine 51 has negative effects on both nuclear localization and DNA binding [139]. Figure 6D (inset at upper right) presents a crystallographic view of a portion of the first zinc finger of hVDR bound to a DR3 VDRE [123]. It is evident that the serine 51 side chain faces toward a DNA phosphate group (in color plate, phosphorus atom is shaded orange; oxygens are magenta). We hypothesize that phosphorylation of the serine 51 side chain which is located only 4.7 nm from one of the phosphate oxygens, not only introduces a steric hindrance to this positioning of the VDR DNA binding helix, but also creates a charge repulsion between the two negative phosphate groups that would prevent the DNA-binding α-helix of VDR from contacting the VDRE. Presumably, introduction

CHAPTER 13 Nuclear Vitamin D Receptor

of a negative charge at position 51 would also neutralize the nearby basic nuclear translocation residues. Supporting this conclusion is our observation [139] that replacing serine 51 with a constitutive negatively charged amino acid, namely aspartate, precludes both nuclear localization and VDRE binding by hVDR. Therefore, although the biological significance of PKCcatalyzed phosphorylation of hVDR is unknown, this modification clearly modulates the activity of the receptor, perhaps attenuating it under certain conditions of cell growth and differentiation.

B. Ligand Binding/Heterodimerization/ Transactivation Domains of VDR and Comparison with PXR To date, an X-ray crystallographic structure of the DBD and the LBD together in the same receptor protein has not been reported, and therefore it is not possible to visualize how the DBD and the ligand binding/ heterodimerization domains are arranged relative to one another. However, an X-ray crystallographic structure of the VDR ligand binding/heterodimerization domain bound to 1,25(OH)2D3 has been obtained by Rochel et al. [121]. As shown schematically in Fig. 7A, and in a view created in Protein Explorer [124] in Fig. 7B, the crystallized VDR structure contains two discontinuous segments, from residues 118 to 164 and from 216 to 425, spliced together to form a peptide that could be purified and crystallized [121] (the position of the deletion is indicated by a white box with crossed lines in Fig. 7A and by red highlighting of residues 164 and 216 on either side of the deletion in the Fig. 7B color plate insert). Despite the discontinuity, the structure of the VDR ligand binding/heterodimerization domain displayed in Fig. 7B bears a strong resemblance to the X-ray crystal structures of the corresponding regions of other nuclear receptors [141], consisting of what has been described as a “sandwich” of 12–13 α-helices with a centrally located hydrophobic ligand binding pocket (see also Chapter 15). Within the general structure of the α-helical sandwich, the ninth and 10th helices (depicted in green in Fig. 7, see color plate insert) have been shown in several nuclear receptors to contain heterodimer contacts [142]. The surface formed by these helices in VDR appears similar to that observed in crystals of the LBD of RARα, RXRα, PPARγ, and ER [143–146]. Moreover, there is a human HVDRR patient harboring a VDR in which arginine 391 (helix 10) is mutated to cysteine (colored blue in Fig. 7B, see color plate insert), and in vitro experiments have confirmed that

233 RXR heterodimerization by the R391C mutant hVDR is severely impaired [147]. The R391C HVDRR patient, like all those with DNA-binding mutations in hVDR (see Fig. 6AB), displayed an alopecic phenotype, indicating that RXR heterodimerization is functionally linked to VDRE binding, and that both of these actions of VDR are required for hair cycling. Independently supporting an obligatory role for the RXRα heteropartner in hair cycling is the observation that mice with an RXRα conditional knockout in skin exhibit alopecia indistinguishable from that in VDRnull animals [128]. Crystallographic studies with a number of nuclear receptors have revealed that helices 3, 5, and 12 combine to form a platform for transcriptional coactivator binding [148–152], illustrated for hVDR in yellow in Fig. 7B (see color plate insert). The involvement of helix 12 in VDR action is also supported by the identification of an HVDRR patient, in whom glutamate 420 is altered to lysine [153]. Further testing has demonstrated that natural mutant E420K hVDR [153], as well as its synthetic counterpart, E420A [154], have abrogated transactivation capacity caused by their inability to interact with the comodulators steroid receptor coactivator-1 (SRC-1) and vitamin D receptor interacting protein 205 (DRIP205). The third natural hVDR mutant highlighted in Fig. 7B is an arginine 274 to leucine alteration that leads to a loss in binding of the 1,25(OH)2D3 ligand [147]. Arginine 274 is located in helix 5, and directly contacts the 1,25(OH)2D3 ligand via hydrogen bonding of the 1α-hydroxyl moiety [121]. Thus, the three natural point mutations in the hVDR LBD highlighted in Fig. 7B represent loss-of-function alterations for each of the three molecular actions of this domain, namely ligand binding, heterodimerization, and transactivation, with the respective disruption of contact by hVDR with 1,25(OH)2D3, RXR, and coactivators. In summary, just as clinically significant inactivating point mutations in the hVDR DBD prove the obligatory role of DNA binding in VDR action (Fig. 6B), the three HVDRR associated point mutations in the C-terminal domain of hVDR highlighted in Fig. 7B illuminate the necessity for ligand binding, RXR heterodimerization, and coactivator contact in order for VDR to transduce the signal for the prevention of rickets. Although the crystal structure of the unoccupied LBD of hVDR has not been solved, it is generally accepted [141] that unliganded nuclear receptors are configured such that the helix 12/activation function 2 (AF2) is in the “open” orientation, directed away from the α-helical sandwich and allowing for entry of the ligand into a hydrophobic pocket (Fig. 8). Upon ligand binding with 1,25(OH)2D3, helix 12 of VDR, which contains two ligand

234

G. KERR WHITFIELD

A

DBD

ET AL .

Ligand binding/Heterodimerization

CTE

Transactivation 165

215

118

425 H1

deleted

H3 H5 βstrands

H9

H10 H12

hVDR

hPXR/SXR

B

C Helix 1 Helix 9

Helix 5

C 274 420 Hel

Helix 3

391

Helix 10

RXR

r

ato tiv

c oa

ix 1

2

SR12813

1,25(OH)2D3

∆165-215

Proposed strand region deleted from hVDR

177-198 (unstructured)

FIGURE 7 Structure-function of the human VDR ligand binding/heterodimerization/transactivation domain compared to that of hPXR/SXR (see color plate insert and text for details).

contact residues (valine 418 and phenylalanine 422), is repositioned to the agonist bound “closed” orientation and interacts with helices 3, 5, and 11 [121] to create a coactivator platform (Fig. 7B). 1,25(OH)2D3 ligand binding also alters the conformation of the LBD such that the surface presented by helices 9 and 10 is more favorable for association with RXR (Fig. 7B). Facilitation by 1,25(OH)2D3 of strong heterodimerization of VDR with RXR has been demonstrated [27,32,147], and the resulting VDR-RXR heterocomplex is capable of high-affinity interaction with VDREs. Therefore VDR liganding with the 1,25(OH)2D3 hormone generates both RXR heterodimerization for specific DNA binding and coactivator docking for transcriptional activation. One fascinating aspect of the action of the VDR LBD is the following observation. Point mutations in hVDR that abrogate DNA binding and RXR heterodimerization lead to both rickets and alopecia; however,

neither 1,25(OH)2D3 ligand binding (e.g., R274L) nor coactivator interaction (e.g., E420K) mutations elicit alopecia in conjunction with the vitamin D-resistant rachitic phenotype. Since vitamin D deficiency or ablation of the 1α-hydroxylase enzyme that generates the 1,25(OH)2D3 hormone [155,156] also causes rickets without alopecia, it is clear that the 1,25(OH)2D3 ligand binding function of VDR can be dissociated from the action of the receptor to regulate the hair cycle. Similarly, transcriptional activation by VDR is not required for hair growth. It therefore appears that either the unliganded VDR-RXR heterodimer bound to DNA is sufficient for control of the hair cycle, or a novel (non-vitamin D) lipophilic ligand operates in skin to activate VDR. As discussed later in this chapter, because VDR mediation of the hair cycle does not require coactivators, it is possible that this VDR function is executed by gene repression, and a nuclear receptor corepressor such as the hairless protein (Hr) [157]

235

CHAPTER 13 Nuclear Vitamin D Receptor

may participate in VDR signal transduction in the hair cycle. The availability of a PXR structure [122] permits interesting comparisons between the ligand binding/ heterodimerization domain of VDR (complexed with 1,25(OH)2D3) and that of PXR (bound to synthetic ligand SR12813), its closest relative in the nuclear receptor superfamily. The overall structures of the domains of these two receptors are nearly superimposable [121,122] (compare Figs. 7B and 7C with matching views created in Protein Explorer [124]). The PXR LBD has a particularly large ligand binding pocket (1150 Å3) [158]; the ligand binding pocket of VDR, even with the deletion of residues 165–215, is also large (approx. 700 Å3) compared with other nuclear receptors that have been crystallized (approx. 400 Å3 [121] but ranging as high as 800 Å3 for LXRβ) [159]. This suggests that like PXR and LXR, VDR may be able to accommodate a variety of lipophilic ligands beyond 1,25(OH)2D3, including LCA/3-ketoLCA. Of additional interest is the fact that the PXR crystal, which was created from a continuous stretch of sequence from positions 142–431 of overexpressed human PXR [122], contains residues that correspond to most of the 165–215 region that is absent in the VDR LBD crystal. Thus the PXR X-ray crystallographic solution can serve as a model for the VDR structure in this region. Of particular note are the additional two β-strands in PXR (illustrated in red in color plate insert), for which the corresponding hypothetical positions in VDR are suggested by the dotted outline in Fig. 7B. Indeed, inclusion of these two β-strands, plus associated turns, in this speculative hVDR LBD structure could allow for the further expansion of the ligand binding pocket. The assertion has been made that this region (amino acids 165–215 in hVDR) may not be necessary for the transactivation function of VDR [160]. However, a mutation was reported in an HVDRR patient in which alteration of cysteine 190 to tryptophan results in loss of hormone binding by hVDR [161]. Although it is possible that the substitution of cysteine 190 by the more nonpolar and much bulkier tryptophan side chain somehow disrupts the general tertiary structure of the VDR LBD, this finding could be interpreted as an indication that the deleted region (165–215) does in fact serve some important function that is not detected by limited in vitro testing of the mutated receptor performed with only the 1,25(OH)2D3 ligand at high (10− 7 M) concentrations [160]. Therefore, as suggested by the close evolutionary relationship between VDR and PXR (Figs. 3 and 5B), as well as the ability of both receptors to bind bile acids (Fig. 5A) and functionally overlap to induce CYP

detoxification enzymes (Fig. 5A), X-ray crystallographic structures of the VDR and PXR ligand binding domains (Fig. 7B,C) strongly support the concept that VDR potentially possesses an as yet uncharacterized array of novel ligands. The repertoire of new VDR ligands could approach that of PXR if one considers the relatively large size of the hydrophobic pocket in the hypothetical expanded VDR LBD architecture pictured in Fig. 7B.

V. MECHANISMS OF VITAMIN D RECEPTOR-MEDIATED CONTROL OF GENE EXPRESSION A. Initial Biochemical Steps in the Activation of VDR Signaling As described earlier in Figs. 1–7 and Table I, several lipophilic ligands and target gene DNA sequences have been defined for VDR, and the structures of the 1,25(OH)2D3 hormone-occupied LBD as well as the VDRE-docked DBD have been elucidated. However, the precise molecular mechanism whereby VDR ligands signal the control of gene transcription has not been fully characterized. This process is best understood for VDR mediation of 1,25(OH)2D3-stimulated transcription, where RXR heterodimerization constitutes an obligatory initial step in the VDR activation pathway. At physiologic receptor concentrations and ionic strength, VDR will bind to VDREs only in the presence of both 1,25(OH)2D3 and RXR [32], and in an in vitro transcription system containing native chromatin, 1,25(OH)2D3 will transactivate via VDR only when RXR is included [129]. Figure 8 illustrates in hypothetical schematic fashion how the hormonal ligand could be influencing VDR to interact more efficaciously with its heterodimeric partner, with a VDRE, and with coactivators. Based upon experiments with VDR [66,130,162], as well as insight from the mode of action of other nuclear receptors [163–165], the key event in the allosteric model presented in Fig. 8 is the binding of a ligand, either the 1,25(OH)2D3 ligand for VDR or 9-cis RA for RXR. According to this model, the ligand that arrives initially sets the tone for the dimerization and transcriptional events that follow. If 1,25(OH)2D3 binds first, the liganded VDR is proposed to follow the pathway to VDR-RXR heterodimerization, indicated by bold arrows in Fig. 8. There are several steps that apparently are set in motion by the ligand binding event. These include high-affinity heterodimeric interaction of VDR with RXR and the recognition by the VDR-RXR heterodimer of specific

236

G. KERR WHITFIELD

ET AL .

Basal state: 9-cis RA

RXR

Platform for coactivators

1,25(OH)2D3 ligand

VDR

1,25(OH)2D3

Protein kinase

P

VDR

RXR

Non-specific DNA ‘‘sliding’’ 5'

AGGTCA

9-cis Retinoic acid ligand

Homodimeric RXR diversion

3nt AGTTCA

DR3

RXR

3'

VDRE

Coactivator recruitment

VDR Coactivator(s)

Platform for coactivators 9-cis RA

RXR

LXXLL

9-cis RA

LXXLL

RXR

1,25(OH)2D3

RXR

5'

AGGTCA 1nt AGGTCA DR1 RXRE

P

VDR

3' 5'

Coactivator recruitment

9-cis RA-stimulated transactivation via RXR-RXR

AGGTCA

3nt AGTTCA

DR3

VDRE

3'

1,25(OH)2D3-stimulated transactivation via RXR-VDR

FIGURE 8

Allosteric model of VDR-RXR activation after binding 1,25(OH)2D3, or diversion of the obligatory RXR partner by its 9-cis-retinoic acid ligand (see text for details).

VDREs upstream of 1,25(OH)2D3 target genes [27,32]. The presence of the 1,25(OH)2D3 ligand in the binding pocket also results in a dramatic conformational change in the position of helix 12 at the C terminus of VDR, bringing it to the “closed” position to serve in its AF2 role as part of a platform for coactivator binding [154,166,167]. Further, as depicted in Fig. 8 and supported experimentally [130,131,162], the liganding of VDR conformationally influences its RXR heteropartner to cause the AF2 region of RXR to pivot into the “closed” or active position. The VDR-RXR heterodimer is now poised to bind coactivators, and the allosteric repositioning of the RXR AF2 also greatly reduces the affinity of RXR for its 9-cis RA ligand. With RXR serving as a subordinate partner, VDR is referred to as a nonpermissive primary receptor within the heterodimer,

because RXR cannot bind 9-cis RA when heterodimerized to liganded VDR [130,131,162]. The conformational changes elicited in VDR by the above described interactions with ligand, RXR and/or DNA have the added effect of converting VDR into a more efficient substrate for one or more serine protein kinases [54]. The most relevant phosphorylation appears to be catalyzed by casein kinase II (CK2) on hVDR serine 208 [168,169], an event that has been shown to potentiate the transcriptional activity of the VDR-RXR heterodimer [170], likely by enhancing interactions with coactivators [171]. As discussed earlier (Fig. 7), the coactivator platform in other nuclear receptors has been shown in crystallographic studies to consist of helix 12 (the AF2) in concert with helices 3 and 5 [148–152], and it has

CHAPTER 13 Nuclear Vitamin D Receptor

been suggested that most nuclear receptors have similar interaction domains [141,142], with the possible exception of the progesterone receptor [146,172]. These crystallographic studies indicate that coactivators of the p160 class, such as SRC-1, utilize one or more LXXLL motifs for binding to nuclear receptors (where L = leucine and X = any residue). There is evidence from site-directed mutagenesis experiments that helix 12 and helix 3 of VDR are indeed involved in coactivator contact [154,173]; also, the above-mentioned natural mutation in VDR in which glutamate 420 is altered to lysine has been found not only to confer the HVDRR phenotype in a human patient, but also to abolish binding of SRC-1 to VDR in experiments with transfected cells [153]. As the “subordinate” partner receptor bound to the 5′ half-site of the VDRE, RXR nonetheless has been proven to be essential for transactivation by the 1,25(OH)2D3-liganded VDR-RXR heterodimer [130,131,162]. Accordingly, the depiction in Fig. 8 illustrates the AF2s of VDR and RXR simultaneously contacting the same coactivator. This is a plausible scenario, since SRC-1 and other p160 coactivators possess three distinct LXXLL domains that are known to interact with nuclear receptors [174,175]. However, also pictured in Fig. 8 (by the dashed vertical line dividing the coactivator) is the possibility that VDR and RXR contact separate coactivators at this stage [130,131,176], the combination of which is required for the full transcriptional response. In summary, the initial events in positive signaling by VDR are triggered by 1,25(OH)2D3 liganding, which generates the VDR-RXR heterodimer that both binds to the VDRE and recruits bridging coactivators that attract and stimulate the transcription machinery. The events described in the preceding paragraphs are all predicated on the binding of 1,25(OH)2D3, and not 9-cis RA, to the unliganded VDR-RXR heterodimer. Should 9-cis RA bind first, the scenario likely becomes quite different. In this case, as suggested in Fig. 8 (lower left), the VDR-RXR heterodimer would dissociate, leaving RXR to form homodimers that could then seek out retinoid X receptor responsive elements (RXREs) in DNA. The most common reported motif for an RXRE is a DR1 [25]. In this pathway (narrow arrows on the left side of Fig. 8), the binding of the RXR ligand would preempt action by 1,25(OH)2D3 in favor of retinoid signaling, and this model is consistent with observations in many systems using 1,25(OH)2D3- and retinoid-dependent reporter genes [27,30,32]. However, there are contrasting reports that, in certain regulatory or cellular contexts [177,178], or with different VDREs [179], 1,25(OH)2D3 and retinoid ligands can act synergistically to effect transcriptional activation. As will be

237 discussed later in relation to Fig. 10, synergistic activation of gene expression by 1,25(OH)2D3 and retinoids could be the result of distinct VDRE sequences allosterically imparting conformational variations on the VDRRXR heterodimer and/or the presence of cell-specific transcriptional comodulators.

B. A Sequential, Cyclic Working Model for Transactivation by 1,25(OH)2D3-Liganded VDR-RXR In the process of dissecting the complete mechanism whereby 1,25(OH)2D3-liganded VDR effects transcriptional activation, numerous VDR-interacting proteins (VIPs) have been identified [35,66,180]. VDR possesses interaction surfaces in the DBD (Fig. 6C) and LBD (Fig. 7B) for RXR, the most fundamental VIP. The 1,25(OH)2D3-liganding of VDR conformationally influences both subunits in the VDR-RXR heterodimer to recognize VDREs and become a binary nucleus for the attraction of coactivators to vitamin D regulated promoters (Fig. 8). The balance of the transactivation process is incompletely understood, but appears to involve numerous other VIPs that remodel chromatin, participate in transcriptional initiation, and modify VDR. Some VIPs directly contact VDR, and others act as part of supercomplexes tethered to VDR. By analogy to the molecular actions of the TR [181] and ER [182], and the observations that the regions of VDR required for interactions with p160 coactivators like SRC-1 [166], vitamin D receptor interacting proteins such as DRIP205 [167], and proteasomelinked TRIP1/SUG1 [183] physically overlap and include the AF2/helix 12 domain, we propose that these VIPs function via a sequential or combinatorial [184,185] rather than simultaneous mechanism. Figure 9 represents a current conception of the 1,25(OH)2D3-liganded VDR-RXR transcriptional activation cycle, focusing on sequential complexes formed on the VDR-RXR heterodimer [186]. A starting point for the cyclic model illustrated in Fig. 9 is that of the unliganded and loosely associated VDR-RXR heterodimer nonspecifically bound to DNA, with transcription idling in the basal state (Fig. 9, top). This depiction is similar to the basal state of the VDR-RXR heterodimer in Fig. 8, but also includes a corepressor bound to the unliganded VDR-RXR. Corepressors have a higher affinity for unliganded VDR [187], in which the AF2 domains of VDR and RXR are presumably in the open, inactive position [166]. Some corepressors, such as silencing mediator for retinoic acid and thyroid hormone receptors (SMRT)

238 [188], and the hairless gene product, Hr [157,189], associate with proteins possessing histone deacetylase (HDAC) activity that represses chromatin [188]. A final member of the apo-VDR complex is basal transcription factor IIB (TFIIB), which interacts with unliganded VDR at two sites [190–193] (indicated as I and II in Fig. 9). Thus, analogous to the findings of Li et al. [194], who showed that TR-RXR exists in a complex with both a corepressor (NCoR) and coactivator (SRC3/ACTR), we postulate that inactive VDR-RXR is associated with both a corepressor and an activator (TFIIB) of transcription. As detailed in Fig. 8, a number of transformations occur upon switching VDR to the active state by binding of the 1,25(OH)2D3 ligand, including higher affinity heterodimeric association between VDR and RXR, specific recognition of VDREs by the liganded heterodimer, repositioning of the AF2 domains in both VDR and RXR, and serine phosphorylation of VDR. These changes convert VDR and RXR into platforms for coactivator binding and, concomitant with 1,25(OH)2D3 liganding, the corepressor/HDAC complex dissociates in order to permit the docking of a coactivator(s). The coactivator shown in Fig. 9 (right) is SRC-1, which belongs to the p160 class of coregulators that interact via LXXLL domains with a hydrophobic crevice created by contact between the AF2/helix 12 and helices 3 and 5 in both VDR and RXR (as discussed above and pictured for VDR in Fig. 7B). The coactivator complex, illustrated on the right in Fig. 9, apparently contains at least three additional proteins. NCoA-62 [193,195], a nuclear receptor coactivator also known as Ski-interacting protein, or SKIP, is believed to interact directly with VDR, but at a site distinct from the AF2 domain, and forms a ternary complex with VDR and SRC-1 [193,196]. CREB binding protein (CBP) or its relative, p300, is a cointegrator that is next incorporated into the activation complex by interaction with SRC-1 [197]. CBP/p300 does not bind directly to VDR and only associates with the receptor when the SRC-1 bridge is present [186]. CBP/p300, in turn, attracts the multisubunit Brahma/ SWI2-related gene 1 associated factor (BAF), or the closely related polybromo- and BAF-containing complex (PBAF) [198], both of which can serve as the mammalian counterpart [199] to the yeast SWI/SNF complex [185]. The net result of ligand-driven assembly of the VDR-RXR/SRC-1/NCoA-62/CBP-p300/PBAF supercomplex (depicted on the right in Fig. 9) is that chromatin remodeling factors have been recruited to the promoter of a 1,25(OH)2D3-regulated gene. There is an alternative view of the interaction of PBAF with nuclear receptors such as RAR [200] and

G. KERR WHITFIELD

ET AL .

VDR [201] (Fig. 9, top), in which the ATP-driven chromatin remodeling activity of PBAF is required prior to RXR-heterodimeric binding to the RARE and VDRE, respectively, and that the remodeling SWI/SNF-like complex actually attracts the unliganded nuclear receptor heterodimer to the promoter site instead of vice versa (Fig. 9, right). Indeed, the chromatin remodeling complex WINAC that is defective in Williams syndrome [201], an inherited hypercalcemic disorder caused by a deletion in chromosome 7, has been reported [201] to target unliganded VDR to its promoter sites. WINAC [201], PBAF [198], and possibly other complexes that utilize ATP hydrolysis to rearrange nucleosomal arrays to render DNA accessible to VDR-RXR are clearly required for 1,25(OH)2D3 stimulation of transcription in the context of native chromatin [198]. However, in vivo footprinting of the rat osteocalcin gene promoter reveals that only in the presence of the 1,25(OH)2D3 ligand is the VDRE occupied by VDR [202]. Therefore, whether PBAF-like complexes facilitate VDRE recognition by VDR prior to liganding will require further proof, although roles for SWI/SNF at multiple steps in transactivation have been reported [203], including recruiting histone acetyl transferase (HAT) complexes. SWI/SNF has even been proposed to execute disassembly of nuclear receptor-coregulator complexes, a role also postulated for chaperone proteins such as p23 (not illustrated in Fig. 9) [204]. Disassembly of transcription factor complexes could facilitate “turning of the cycle” in Fig. 9 and generate rapid, dynamic interactions between VDR and its array of co-regulators. In fact, recent studies of the RNA polymerases suggest that rather than consisting of stable holocomplexes, transcription (or transcriptional control) units are assembled in a stochastic fashion from freely diffusible subunits that only transiently associate [205]. Employing fluorescence recovery after photobleaching, it was observed that the liganded glucocorticoid receptor [206] and its interacting protein (GRIP1) coactivator [207] quickly exchange between a DNA promoter and the nucleoplasm. Thus, the transcriptional activation complexes pictured in Fig. 9 may be dynamically assembled and reassembled by a “hit and run” mechanism instead of remaining as holo-complexes throughout the turning of the cycle. The task of PBAF, along with the SRC-1/CBP or SRC-1/p300 complex, is to derepress the chromatin in the vicinity of the VDRE (Fig. 9). This is accomplished by the combined action of BAF/PBAF to remodel the chromatin [198] by sliding nucleosomes along DNA, and of the SRC-1 and CBP/p300 complex to reacetylate nucleosomal histones via HAT activities [208] intrinsic to the coactivators. Histone acetylation and

239

CHAPTER 13 Nuclear Vitamin D Receptor

Basal state

Corepressor AF2

AF2

I

VDR

RXR

Hormone catabolism

HDAC

TFIIB Corepressor/HDAC

PBAF SWI/SNF

TFIIB

II Nonspecific DNA

Repressed chromatin

Corepressor/HDAC

Protein phosphatase

CYP24

SRC-1,CBP NCoA-62

1,25(OH)2D3

Chromatin remodeling

TRIP1/SUG1 SRC-1,CBP NCoA-62, TRIP1/SUG1 TFIIB,PBAF

VDR degradation in 26S proteasome

1,25(OH)2D3Ligand Protein kinase

CBP/P300

RXR

ubiquitination

AF2

N SRC-1 Co LXXLL LXXLL A62 AF2 AF2

P

1,25

VDR

U

Transcriptional recycling

VDR

3nt

TFIIB

3' 5'

Repressed chromatin

VDR target gene

monoubiquitination

SRC-1,CBP PBAF(SWI/SNF)

Mediator complex D

NCoA-62 DRIP205/Mediator RNA pol II

R

I P

PIC s

DRIP205/Mediator RNA pol II

DRIP205 NR1

NR2

AF2

AF2

P

RNA Pol II TFIIB

1,25

RXR VDR 5'

3'

Autoacetylation

Histone deacetylation

N Co A62

3nt

VDRE Specific DNA sequence

TRIP1/SUG1

Spliceosomemediated hnRNA processing

P

1,25

RXR VDR 5'

Chromatin remodeling

Histone acetylation

TRIP1/SUG 1 AF2

VDR poly-

VDRE Specific DNA sequence

FIGURE 9

PBAF SWI/SNF

3nt

3'

VDRE Specific DNA sequence

De-repressed chromatin

Transcriptional initiation

VDR target gene

The transcriptional life cycle of VDR, depicting a model for the stages in transactivation of an idealized 1,25(OH)2D3regulated gene (clockwise from top of figure). At twelve o’clock is shown the unliganded VDR-RXR heterodimer including its proposed interactions with TFIIB and with a corepressor. The corepressor, which binds to VDR when its AF2 is in the “inactive” configuration, acts by attracting HDACs that repress chromatin. Also shown is PBAF, which according to one view [201] may remodel chromatin to attract the VDR-RXR heterodimer to the promoter. Illustrated at three o’clock are the postulated initial alterations that transpire after binding of 1,25(OH)2D3 to VDR. These changes include one or more hormone-dependent phosphorylations of VDR (indicated by a circled “P” against a dark background), along with reconfiguration of the AF2 domains in VDR and RXR. The VDR-RXR heterodimer is now bound to a VDRE adjacent to a target gene. Changes in the conformation and phosphorylation status of the VDR-RXR heterodimer serve to dissociate the corepressor/HDAC complex and promote the binding of a coactivator complex that includes not only SRC-1, but also CBP/p300, nuclear receptor coactivator-62 (NCoA-62), and a SWI/SNF chromatin remodeling complex anchored by PBAF. This coactivator complex derepresses chromatin in the vicinity of the VDR target gene by catalyzing histone acetylation and chromatin remodeling. At six o’clock is illustrated the transcriptional initiation complex. Steps that are postulated to occur between three and six o’clock are (i) autoacetylation of SRC-1 and CBP/p300, leading to their dissociation from the VDR-RXR heterodimer, taking PBAF with them; (ii) binding of DRIP205 to the AF2 of VDR (and also RXR), an event that attracts other DRIPs in the Mediator complex; and (iii) delivery of TFIIB to the RNA Pol II transcription machine (indicated by a rightward arrow). The complex pictured at nine o’clock is postulated to form after the transcription machine has moved away from the promoter, with DRIP205/Mediator dissociating from the VDR-RXR heterodimer to be replaced by TRIP1, a mammalian ortholog of the yeast SUG1 protein. NCoA-62 also dissociates, and may be involved in processing of the primary heterogeneous nuclear RNA (hnRNA) transcript by the spliceosome [210]. VDR is then subject to ubiquitination by TRIP1 and its associated ubiquitin ligation complex (an attached ubiquitin is indicated by a circled “U”). The structure illustrated at nine o’clock now has three alternative fates. The “transcriptional recycling” fate encompasses dissociation of TRIP1, either with or without monoubiquitination of VDR, and reassociation of SRC-1, CBP/p300, NCoA-62, TFIIB and PBAF to regenerate the complex at three o’clock for a rapid reinitiation of transcription. A second fate would be dissociation and metabolic elimination of the 1,25(OH)2D3 ligand via catabolism by CYP24, leading to recovery of the unliganded complex at twelve o’clock after dissociation of TRIP1 and reassociation of TFIIB and corepressor. A third fate (receptor degradation), which may occur after several rounds of VDR ubiquitination and transcriptional initiation, would be transfer of the poly-ubiquitinated VDR to the 26S proteasome for degradation, effectively terminating the VDR transcriptional life cycle.

240 methylation [184] appear to reduce internucleosomal interaction in chromatin, converting it to transcriptionally active euchromatin. Acetylation by HATs is proposed to continue until coactivators undergo selfacetylation [182] and dissociate from VDR along with BAF/PBAF to free sites on the receptor for recruiting DRIPs and the RNA polymerase II preinitiation complex. The next complex to form with the VDR-RXR heterodimer (shown at the bottom of Fig. 9) is initiated by the attraction of DRIP205 to the AF2s of both VDR and RXR [167]. This occurs on a similar or overlapping docking site to that vacated by the p160-class coactivator (SRC-1) in the previous complex. The physical impossibility of both SRC-1 and DRIP205 interacting simultaneously with the AF2 platforms of VDR and RXR is, in fact, one of the chief arguments for the sequential model presented herein (see also Rachez et al. [167]). DRIP205 then attracts a Mediator complex that is still incompletely characterized, but which contains numerous other DRIPs that bridge the VDR/RXR/NCoA-62/DRIP205 complex to the RNA polymerase II (RNA Pol II) transcription machine [209]. When DRIP205 associates with VDR, we hypothesize that it displaces TFIIB and delivers it to RNA Pol II [191,192], facilitating the formation of the transcriptional preinitiation complex (PIC). After transcription begins and the initiation complex dissociates, DRIP205/Mediator is likely released from VDR, as is NCoA-62, which remains in the vicinity of the induced gene and appears to participate in spliceosomemediated heterogeneous nuclear RNA processing [210]. Once transcriptional initiation has occurred, and DRIP205/Mediator has departed the VDR-RXR platform, the absence of coactivators with either HAT or chromatin remodeling activity would presumably expose the promoter DNA and its nucleosomes in the region of the VDRE to the repressive action of histone deacetylases. Several possibilities now exist for the fate of the liganded VDR-RXR AF2 platform. Evidence from chromatin immunoprecipitation assays [211] indicates rapid oscillatory binding of p160 and DRIP coactivators to VDR, suggesting that the next event in the VDR transcriptional cycle is the rebinding of the SRC-1, NCoA-62, CBP/p300/PBAF supercomplex, basically “short-cutting,” or recycling, back to the chromatin remodeling stage shown at the right of Fig. 9. Because 1,25(OH)2D3-liganding of VDR has been reported to protect the receptor from proteasome-mediated degradation [212], sustained transcriptional activation by VDR-RXR via alternating recruitment of p160 and DRIP205 coactivators is a conceivable scenario. On the other hand, in other eukaryotic transcription systems, for example the srb10 kinase Mediator subunit

G. KERR WHITFIELD

ET AL .

which signals ubiquitination of the transcriptional activator Gcn4, and the MED8 subunit that assembles into Elongin BC-based E3 ubiquitin ligase [213], Mediator attraction to the PIC is coupled to the subsequent recruitment of ubiquitination machinery. Thus, as hypothesized on the left side of Fig. 9, in the case of VDR action, the vacant VDR-RXR AF2 platform may become occupied by TRIP1 (the mammalian ortholog of the yeast SUG1 protein [183]) as a consequence of DRIP205/Mediatorcoupled attraction of the ubiquitination machinery. TRIP1/SUG1 is a documented VDR-interacting protein [183,186] that stimulates VDR proteolysis in a process blocked by the 26S proteasome inhibitor, MG-132. If VDR follows the GR pattern of successive mono-, di-, tri-, and tetra-ubiquitination, in which the linkage of fewer than four ubiquitin moieties precludes association of the proteasome degradation machinery [214], then like GR, VDR will be capable of reengaging HAT/DRIP/PIC complexes and completing additional rounds of transcription [215]. In this model, activated VDR-RXR would traverse the transcription cycle pictured in Fig. 9 clockwise until the first ubiquitination event, and then “short circuit” the cycle by making four rounds of the lower half of the cycle prior to tetra-ubiquitination and subsequent degradation in the 26S proteasome. A final alternative fate of the TRIP1-VDR-RXR complex would be triggered by the dissociation of both TRIP1 and the 1,25(OH)2D3 ligand. The loss of ligand leads to a conformational change in VDR and RXR, in which the AF2 domains pivot to the open inactive position, leading to reassociation of corepressor with the unliganded VDR-RXR heterodimer. Unliganded VDR would now be available as a target for a protein phosphatase to remove the phosphate(s) on VDR that was covalently attached by the hormone-dependent protein kinase(s). All of these changes would return the VDR-RXR heterodimer to the state appearing at the top of Fig. 9, completing a full turn of the proposed transcription activation cycle. Termination of the VDR transcriptional cycle is postulated to be accompanied by catabolism of the 1,25(OH)2D3 ligand via CYP24 (24-hydroxylase) initiation of its conversion to calcitroic acid [47]. This enzyme is, in fact, encoded by one of the genes that is transcriptionally activated by the liganded VDR-RXR heterodimer to promote destruction of the ligand as a mechanism for self-limiting the transcriptional activation signal [216] (see also Chapter 6). The transcriptional activation cycle model for VDR illustrated in Fig. 9 is presented as a working hypothesis because the order in which each co-regulator participates has not been determined experimentally, although firm evidence exists that each VDR-interacting player binds to the receptor, at least under in vitro conditions.

CHAPTER 13 Nuclear Vitamin D Receptor

Thus, the order of events depicted in Fig. 9 will not be absolutely defined until real-time chromatin immunoprecipitation assays of the type carried out by Pike and colleagues [211] are completed for VDR on several of its endogenous target gene promoters. The cycle shown in Fig. 9 is a basic pattern subject to many variations depending upon the specific target cell and gene regulated by VDR. A schematic conception of tissue- and promoter-specific influences on the transcriptional activation cycle is shown in Fig. 10. As discussed earlier (Table I), there are many variations in the sequences of natural VDREs, and there are indications that these sequence differences may allosterically confer the VDRE-bound heterodimer with distinct conformations that attract different sets of coactivators [62,63]. The VDRE may also be located adjacent to binding sites for other transcription factors (termed cis-modulators in Fig. 10), and there are reported cases of interactions between the bound VDR-RXR heterodimer and nearby DNA-bound transmodulator proteins. Prominent examples are the human osteocalcin and involucrin promoters, in which the VDRE is located adjacent to an AP-1 binding site for the jun/fos heterodimer [45,217]. Given the obvious complexity of many other natural VDRE-containing promoters, there are likely additional examples that remain to be characterized, for instance in the promoters of the rat osteocalcin [218] and mammalian CYP24 genes [219]. As outlined earlier, phosphorylation of VDR by various serine kinases can be either stimulatory when catalyzed by CK2 at serine 208 [170], or inhibitory when catalyzed by PKC at serine 51 [139]. Moreover, we have recently provided evidence [54] that phosphorylation of hVDR by an unidentified kinase may be required for high-affinity heterodimeric binding of the receptor to a VDRE. There are other kinases that phosphorylate VDR, such as PKA [220], and because protein kinase expression is likely cell specific, the resulting phosphorylation pattern of VDR may vary from one target cell to another. Phosphorylation of RXR has been studied less extensively, but there are indications that RXR is also phosphorylated [221–223]. This posttranslational modification of RXR presumably could influence its transcriptional activity [219,221–223], but whether phosphorylation of RXR plays a role in signaling by VDR-RXR, as suggested by Fig. 10, remains to be clarified. Specific studies of RXR as a heterodimeric partner are complicated by the fact that there are three distinct RXR genes in mammals [224], with each gene producing multiple isoforms [225]. On a superficial level, all of these RXR isoforms seem interchangeable as heterodimeric partners for VDR. However, results from several laboratories [28,177,226] strongly suggest that RXR isoforms may exhibit differential abilities to

241 participate in transactivation in heterodimeric context with VDR. The VDR-mediated transactivation models presented in this chapter (Figs. 8 and 9) consider the 1,25(OH)2D3occupied VDR-RXR heterodimer to be refractory to stimulation by 9-cis RA, a natural RXR ligand. However, there have been reports of synergism between 1,25(OH)2D3 and 9-cis RA in the activation of certain genes [227]. The above-mentioned possibility of alternative RXR conformations when the VDR-RXR heterodimer is bound to different VDREs raises the possibility that some of these conformations might allow for 9-cis RA to bind to RXR and to participate in reconfiguring the RXR AF2 for coactivator binding. For many years 1,25(OH)2D3 was considered to be the sole physiologic ligand for VDR. However, with the recognition of at least two new natural bile acid ligands for VDR (Fig. 4), plus the availability of numerous chemical analogs of 1,25(OH)2D3, it is possible to screen ligands that selectively activate the VDR-RXR heterodimer [228–230]. The ability of some analogs to mimic certain transcriptional actions of 1,25(OH)2D3 (e.g., in skin differentiation) while avoiding the hypercalcemic effects has unveiled exciting possibilities for the use of these compounds in therapeutic interventions for hyperproliferative skin disorders [231], for autoimmune disorders [232], or as potential antitumor agents [233]. The basis for selective actions of VDR ligands is presumably that they, like the VDREs (which can also be considered as ligands for VDR-RXR), stabilize distinct conformations of the heterodimer that can undergo some, but not all, of the full spectrum of coactivator/corepressor interactions. This strategy would be analogous to the use of selective estrogen receptor modulators (SERMs) in targeting estrogen receptor antagonist and agonist actions to specific estrogen target tissues [234]. As has been emphasized throughout this chapter (especially Figs. 8 and 9), in order for VDR to control the expression of genes, it must interact with coregulator proteins that possess a variety of activities that transduce the signal to RNA polymerase II. Many of these proteins associate with the AF2/helix 12 of VDR, or at least influence the helix 3/5/12 platform. Clearly, the allosteric regulators of VDR just outlined impinge on the ability of the receptor to attract the sequence of comodulators illustrated from left to right across the top of Fig. 10. The several categories of comodulators that associate with VDR, classified according to their actions, are corepressors (Hr/HDACs), SWI/SNF chromatin remodelers (PBAF/WINAC), HAT coactivators/ cointegrators (SRC-1/CBP/p300), DRIP Mediators and basal transcription factors that assist in the assembly of the PIC containing RNA polymerase II

242

G. KERR WHITFIELD

ET AL .

Cell specific comodulator expression Chromatin repression: Hr/HDACs

CoChromatin remodeling: integrators: CBP/ PBAF/WINAC p300 (SWI/SNF)

Coactivators: SRC-1 GRIP1/TIF2 ACTR

Mediator/Basal factors: DRIP 205 TAFs TFIIB DRIPs

HATs

Receptor degradation: TRIP-ubiquitination 26S proteasome

NCoA-62 RXR ligands Rexinoids

Cell specific P phosphorylation

AF-2

AF-2

RXR isoforms

VDR polymorphisms

Transmodulator

VDR ligands Drug analogs

P

Cell specific phosphorylation

Transmodulator DNA

Cismodulator

VDRE sequences

Cismodulator

Promoter specific

FIGURE 10 Model showing potential allosteric regulatory inputs (large single- and double-headed, open arrows) to VDR-RXR signaling as mediated by a liganded heterodimeric receptor–DNA complex. The VDR-RXR heterodimer bound to a VDRE is considered to be the central entity into which various allosteric inputs converge. The basic program of transcriptional activation by liganded VDR presented in Figs. 8 and 9 can be modulated by variations in VDRE sequences in different target promoters (see Table I), or by adjoining DNA sequences to which may be bound other (possibly cellspecific) proteins that serve as “transmodulators” of VDR action (see text for examples). VDR activity may be intentionally modified by drug analogs, for which selective actions are believed to be the result of their ability to induce conformations of VDR that differ from the 1,25(OH)2D3-dependent conformation, and may promote VDR interactions with distinct sets of coactivators. The same principle may apply to alternative natural VDR ligands, such as LCA. Another possible molecular scenario for activation of the VDR-RXR heterodimer may be the ability of certain VDREs or ligands to permit the binding of rexinoid ligands to RXR. This possibility, plus the existence of three distinct RXR genes, each producing multiple isoforms, some cell-specific, further amplifies the variations in signaling by the VDR-RXR heterodimer. All of these allosteric inputs to VDR and its RXR coreceptor may influence subsequent steps in transcriptional regulation, including the binding of various coactivators or corepressor complexes (shown at the top), some of which may be cell-specific.

(DRIP205/TAFs/TFIIB), coactivators that participate in downstream hnRNA splicing (NCoA-62), and receptor proteolysis cofactors (TRIP1-SUG1/26S proteasome). Perusing the list of comodulators at the top of Fig. 10, one is impressed with the sheer number of protein–protein interactions that occur in target cells in order for the liganded VDR-RXR-VDRE supercomplex to repress (upper left) or stimulate (top center) DNA transcription. This complex VDR signaling network would be analogous to the estrogen receptor signaling network [235] that involves a multitude of estrogen receptor interacting proteins.

The exact set of comodulators recruited by the activated VDR-RXR heterodimer may constitute another tissue-specific permutation influencing the nature of transcriptional activation. Many coactivators, including SRC-1, GRIP1 (SRC-2), and ACTR (SRC-3), have distinct tissue distributions. Studies with knockout mice suggest that there are coactivator redundancies in many target tissues [236]. Nevertheless, the presence of varying relative and/or limiting concentrations of certain coactivators may have significant, yet incompletely characterized, effects in a given tissue [237]. The final variable illustrated in Fig. 10 that influences

CHAPTER 13 Nuclear Vitamin D Receptor

VDR-RXR signaling is common polymorphisms in the hVDR gene [238]. As discussed in Section VI,A, both the quality and quantity of VDR-mediated transactivation can be affected by polymorphic variants in the hVDR gene. Therefore, as pictured in integrated fashion in Fig. 10, the VDR-RXR “nucleus” that recruits comodulators for signal transduction is a multifaceted entity that is differentially influenced in an allosteric manner (open arrows in Fig. 10) by hVDR gene polymorphisms, variable VDR lipophilic ligands, cell-specific VDR and RXR phosphorylations perhaps resulting from cross-talk actions of cell surface-acting hormones, unique transmodulators associated with cis elements adjacent to VDREs, and gene promoter-selective VDRE DNA sequence platforms.

VI. IMPLICATIONS OF VITAMIN D RECEPTOR-MEDIATED SIGNALING FOR HUMAN HEALTH AND DISEASE A. VDR Polymorphisms: Association with Disease Risk and Functional Consequences A potentially significant factor in signaling by the VDR-RXR heterodimer is the existence of common polymorphic variants in the hVDR gene, a summary of which is presented in Fig. 11. There are essentially three major locations of polymorphic variation in hVDR: (i) a dimorphic site toward the 5′ end of the gene between exons 1D and 1G within a docking sequence for Cdx-2, a caudal-related homeobox protein; (ii) a dimorphic sequence at the translation start site in exon II; and (iii) a cluster of linked sites in or near exon IX. The characteristics, frequencies, and functional consequences of these three variable loci are discussed next. Cdx-2 plays an important role in intestinal development and differentiation [239], and has been reported to participate in intestine-specific expression of VDR [240] as well as the sucrase-isomaltase complex (SI) [241]. A binding site for this factor, namely the sequence TTTAY (where Y is a pyrimidine [242]) has been identified near the predominant hVDR gene promoter located downstream of exon IA (the Cdx-2 binding site is designated by a black dot in the top register of Fig. 11 between exons ID and IG). Within this binding site is a dimorphic (A/G) variation that results in a change in the Cdx-2 binding site on the complementary strand from TTTAT to TTTAC (see inset, lower left of Fig. 11). The binding site with the G variant was shown by gel mobility shift experiments to have a slightly lower affinity for the Cdx-2 factor [243], implying that

243 intestinal hVDR expression could be suboptimal with the G allele of the receptor gene. Consistent with this in vitro result and conclusion, bone mineral density (BMD) in the lumbar spine was observed to be 12% lower in Japanese women with the GG genotype compared with AA homozygotes [243]. The distribution of the A versus G version of the VDR gene was found to be approximately 45% versus 55% in a Japanese population [243], indicating that the apparently less active G allele predominates in this ethnic group. The translational start site polymorphism in exon II of the hVDR gene produces two alternative forms of the hVDR protein, with one isoform three amino acids longer at the N terminus than the other. The three additional residues in the longer f form are Met-Glu-Ala (Fig. 11, center inset). The approximate worldwide gene frequencies of the f and F alleles are 35% and 65%, respectively [244–247]. Interestingly, the F hVDR allele is associated with enhanced BMD in three different populations of women [244,245,248]. In concert with these clinical observations, the shorter F species of hVDR is also more transcriptionally active [192,244]. Studies in our laboratory revealed that the shorter F hVDR interacts more avidly with TFIIB, and that mutation of the glutamate at position 2 of the longer f isoform to an uncharged residue (alanine) restores the transactivation ability of the longer species to that of the shorter one [192]. Since the TFIIB interaction at this site (site II in the depiction of TFIIB at the top of Fig. 9) appears to be dependent on the presence of two basic residues (Arg-18 and Arg-22) in the hVDR N-terminal domain, we have speculated that the molecular mechanism whereby the negatively charged Glu-2 residue attenuates TFIIB binding could involve either an intermolecular repulsion between f hVDR and presumed negative residues in TFIIB that bridge to hVDR Arg-18/22, or a nonproductive intramolecular interaction of Glu-2 in f hVDR with the Arg-18/22 TFIIB site II, thus precluding TFIIB contact [192]. The common 3′ polymorphisms in the hVDR gene are more complex and difficult to interpret with respect to their functional impact. There are almost two dozen reported hVDR gene polymorphisms in or near exon IX [249] that are in linkage disequilibrium, none of which result in a change in the encoded protein (Fig. 11). Also, unlike the Cdx-2 variations, none of these common 3′ polymorphisms in the hVDR gene are known to occur in the binding site for a factor that could affect receptor transcription. However, one of the areas of variation, a singlet(A) repeat microsatellite approximately 1 kb upstream of the poly(A) tail in the 3′ UTR [250] (Fig. 11, right inset), could conceivably affect the stability or translatability of the VDR mRNA. We have postulated [238] that the human

244

G. KERR WHITFIELD

ET AL .

Alternative Splicing/promoter usage (Tissue specificity?)

IF

IE IAID IG IB

IC II

DNA-bonding zinc fingers

Untranslated regions

Ligand-binding domain

III

IV V VI

VII VIII

IX

q13-14 BsmI (intron)

5′ UTR Chr 12

Alternative translation start

Linkage: ♦ BAtS or ♦ baTL

Cdx-2 binding site ♦ Cdx-2 tissue-specific enhancer for intestine ♦ binding site is polymorphic:

Clinical association: ♦ GG homozygote lower BMD

ApaI TaqI (A)S or (A)L (intron) (silent)

Clinical association: ♦ ff homozygote lower BMD Functional observation: ♦ f isoform less active in vitro

Functional observation: ♦ G allele less active in vitro

3′ UTR

Length difference in a singlet(A) repeat: S: Short (A) repeat = (A)10-15 L: Long (A) repeat = (A)17-24 Clinical associations: ♦ lower Ca absorption, BMD (BAt) ♦ higher rate of bone loss (B) ♦ risk factor for prostate (TL), breast cancers (b) ♦ risk factor for psoriasis (A)

FIGURE 11

Summary of common polymorphic variations in the human VDR gene that may influence receptor activity. At the top is the arrangement of VDR exons [277]. Exons shown as black rectangles contain coding sequences; hatched rectangles represent the 5′ and 3′ untranslated regions (UTRs). Exon IX contains the terminal portion of the coding region along with the entire 3′ UTR [277]. A black dot indicates the position of a binding site for the intestinal Cdx-2 transcription factor [241] that has been reported to affect VDR expression in the intestine [240]. The composition of the 5′ UTR is complex [328], but a predominant form of VDR mRNA appears to be produced from the promoter just upstream of exon IA and contains exons IA and IC spliced to exons II–IX (this promoter and two alternative promoters are indicated by arrows above exons IF, IA, and ID) [277]. There are two variants at the Cdx-2 binding site, two variants in exon II, and numerous polymorphic forms in or near exon IX. The two Cdx-2 binding site variants are proposed to have different affinities for the Cdx-2 intestine-specific transcription factor [243]. SI is the abbreviation for sucraseisomaltase, another gene with a binding site for Cdx-2. The two exon II sequence variants dictate alternative translational start sites in the VDR mRNA, beginning at methionine-1 (M1) or methionine-4 (M4); these alternative forms are detected by the presence ( f) or absence (F) of a FokI restriction site in the genomic DNA (or in cDNA made from the hVDR mRNA). The multiple polymorphic forms at the 3′ end of the gene form a linkage group, with the two most common haplotypes (BAtS and baTL) shown in the black inset at right. None of these 3′ gene variations, including the two TaqI forms, result in any change in the VDR protein sequence. However, we have postulated that the L/S polymorphic forms, which affect the length of a singlet(A) repeat in the 3′ UTR, may influence translatability of the hVDR mRNA (see text). Text boxes at the bottom indicate clinical and functional correlations that have been observed for the indicated hVDR polymorphic variants.

poly(A) binding protein (PABP) binds to this singlet(A) repeat, especially in its longer versions. It has been reported that mammalian and yeast PABPs can enhance translatability of mRNAs via an interaction with other proteins that bind at the 5′ end of the message [251,252]. Usually, PABPs bind to the poly(A) tail of mRNAs, with each monomer occupying approximately 27 adenylate residues [253]. Intriguingly, further studies with human PABP suggest that as few as 11 consecutive As can bind to PABP, with 25 As giving maximum

affinity [254]. Thus, we have speculated that binding of PABP may be enhanced in long (L) alleles (17–24 As) versus short (S) alleles (10–15 As) of the hVDR gene. Greater binding of PABP to L alleles would then lead to more efficacious hVDR mRNA translation via a superior interaction with factors such as EF-4B [251]. This proposal is consistent with our preliminary observation that hVDR L alleles generate a higher concentration of transcriptionally active receptor than do S alleles [238]. The gene frequencies of hVDR L alleles

CHAPTER 13 Nuclear Vitamin D Receptor

(with the remainder being S) vary in human populations from a high of 91% in Asians to a low of 59% in Caucasians [255], indicating that the apparently more active L allele and its linked 3′ polymorphisms predominate worldwide. Given our determination that the L alleles, like the F alleles, produce VDR with a higher activity than the respective alternative alleles S and f, we have devised a system for classifying VDR genotypes at both loci using a single variable [238]. Accordingly, the genotype FFLL is assigned an allele score of 4, with the opposite ffSS genotype receiving a score of 0. Intermediate genotypes receive a score that equals the sum of the more active alleles, L and F; for example, a genotype of FfLS would receive a score of 2. Using allele scores, the average VDR activity in a group of patients can be estimated, with clinical implications as discussed next. The ability of allele score to predict the VDR activity in an individual patient is limited by the fact that there appears to be a more active and a less active subgroup within each allele score grouping [238]. This observation strongly suggests that at least one more functionally significant VDR gene polymorphism remains to be discovered. There is, in fact, a third functionally significant polymorphism within the Cdx-2 binding site between exons 1D and 1G; however, this polymorphism would be predicted to affect VDR activity only in intestine, not in skin fibroblasts that were the subject of our investigation [238]. A significant amount of literature is devoted to identifying correlations between hVDR gene polymorphisms and various clinical parameters or metabolic disorders, for instance BMD or prostate cancer risk [249]. Without reviewing this entire body of work (the reader is referred to Chapter 66 and to other recent reviews [249,256,257]), we offer several generalizations as follows. A broad summary of the accumulating data implicates the less transcriptionally active ff and BB/tt (BAtS haplotype) homozygous alleles as risk factors for low BMD [248], although these are not universal findings and the field of hVDR gene polymorphisms is somewhat controversial [258,259]. Nevertheless, an intuitive interpretation that less active hVDR alleles constitute a risk factor for skeletal hypomineralization is consistent with the physiologic role of VDR in calcium metabolism and bone biology (Fig. 2). The studies probing cancer prevalence versus hVDR gene polymorphisms are more difficult to reconcile and do not offer the physiological correlations supplied by bone mineral homeostasis. In the case of cancer incidence, it appears that the more active hVDR alleles are actually risk factors, especially for prostate and breast cancer [250,260–263]. This is counterintuitive considering the reported antiproliferative and

245 prodifferentiation roles of VDR in prostate [264,265], skin [43], immune cells [88,266,267], and other tissues [268–270]. There are exceptions to the more transcriptionally active baTL hVDR haplotype (see Fig. 11, inset at right center for linkage of alleles in this 3′ region of the gene) having an association with higher cancer incidence; for example the baTL genotype correlates with a lower susceptibility to the initiation and progression of malignant melanoma [271], as well as with a lower breast cancer risk of Latinas living in the United States [272]. Finally, the baTL haplotype associates with a lower incidence of the hyperproliferative skin disease psoriasis [273]. Nevertheless, the majority of data in the case of both prostate [250,260,262] and breast [263] cancer indicate that the more active F/baTL genotypes are procarcinogenic. An explanation of these findings is still elusive, but one mechanism to account for this paradox could be the known immunomodulatory influence of 1,25(OH)2D3-liganded VDR, which consists (in part) of a modification of T-cell function through suppression of the Th1 response [274]. This action of VDR is beneficial in combating graft rejection and autoimmune diseases, but it could conceivably have a negative impact on immune surveillance against malignant cells and tumors [275,276]. Another speculative explanation for the association of more active hVDR polymorphisms with prostate and breast cancer is that VDR, by its action to induce CYP enzymes which can catalyze the formation as well as degradation of carcinogenic metabolites, may be facilitating the synthesis of androgenic and estrogenic compounds that promote cell proliferation in the corresponding endocrine cancer, but as yet there is no evidence of this occurrence. An omission in the literature on hVDR polymorphisms and their pathophysiologic impact is that many investigators limit their analyses to only the F/f or the 3′ UTR polymorphisms in isolation, instead of determining the complete genotype of the patient cohort. Examining the genotype at only one site cannot accurately predict the bioactivity of the hVDR produced from that allele, and will thus give an incomplete picture of the true impact of these polymorphisms for human disease. Published results in which only one site is analyzed (or multiple 3′ UTR sites without either the FokI or the Cdx-2 sites) tend to be inconsistent and even contradictory. This, it could be argued, may be a contributing factor to the current state of controversy in the field. Recently, monitoring the transcriptional potency of endogenous hVDR in cultured human skin fibroblasts, we determined that a combination of FFLL homozygous alleles produced hVDR with the highest transcriptional activity (up to 30-fold more potent than ffSS), and genetic mixtures containing

246 the less active hVDR alleles ( f and S) correspondingly compromised the transcriptional potency of hVDR in heterozygotes such as FfLS [238]. Thus, because 5′ and 3′ hVDR polymorphisms can counteract one another in terms of functional activity and clinical impact, the influence of combined polymorphisms should be examined in the future, perhaps utilizing the allele score concept [238] to account for genetic variation at all relevant sites. The ultimate genotype analysis would be to determine complete hVDR gene haplotypes [249]. This analysis presents considerable technical difficulties; indeed, it may be quite impractical to include the FokI restriction endonuclease site determining f versus F and the 3′ cluster of variants, given the large distance between them in the genome (the FokI site in exon II is >30 kb from the BsmI and ApaI sites between exons VIII and IX and nearly 40 kb from the singlet(A) repeat in exon IX) [277]. Nevertheless, results from such studies, even if limited in scope, could provide valuable insights. Several groups have sought to identify polymorphic variants in VDR target genes that are likely associated with human disease, including osteocalcin [278], CYP19 [279], the cyclin-dependent kinase inhibitors p21 and p27 [280], and CYP3A4 [281]. This approach is promising, especially when the target gene is actually involved in a physiologic process related to the disease in question, for example CYP19 (aromatase) in the etiology of osteoporosis in postmenopausal women [279]. In fact, a complete assessment of the genetic component for disease risk would ideally account for variations not only in VDR, but also in RXR, VDR coactivators, and relevant target genes. Finally, disorders with newly reported relationships to VDR polymorphisms include: type I diabetes mellitus (higher risk with genotypes At and Bt) [282], susceptibility to the hemorrhagic form of dengue fever (with the t genotype displaying greater resistance) [283], anemia in hemodialysis patients (the BB genotype has lower hemoglobin levels) [284], and various dental indices (AA patients had the highest rates of alveolar bone loss) [285]. Many of these disorders have an immune component, and renewed interest in the immune effects of the VDR-RXR signaling system has received impetus from the recent observation that VDR knockout mice, contrary to early reports of a normal immune system, display an abnormal T-cell response [86].

B. Emerging Clinically Relevant Actions of VDR in the Colon and Hair Follicle In addition to the noncalcemic functions of VDR that have been discussed earlier in this chapter, such as

G. KERR WHITFIELD

ET AL .

modulation of the immune system and prevention of epithelial cell cancers, two striking new extraosseous actions of VDR have come into focus recently, as illustrated in Fig. 12. One of these is to drive the hair cycle in skin, which is unique in that, as explained later, it requires neither a vitamin D-derived ligand [155] nor the transactivation function of VDR [153] (Fig. 12B). Therefore, VDR impinges on hair cycle signal transduction in keratinocytes either in an unliganded form or perhaps occupied by an undiscovered lipophilic ligand present in skin, and does so apparently by repressing a VDR target gene(s) [157]. This gene repression in the hair follicle may be mediated by the recently demonstrated physical and functional interaction between VDR-RXR and the Hr nuclear receptor corepressor product of the hairless gene [157]. The second novel noncalcemic action of VDR, the detoxification of carcinogenic LCA in the colon (Fig. 12A), entails the binding of the newly discovered LCA ligand to VDR to induce CYPs [98]. Thus VDR has emerged as a secondary bile acid sensor as well as calcemic endocrine nuclear receptor [98]. VDR and vitamin D have been implicated in the prevention of colon cancer through epidemiologic observations that regions of the world with the lowest sunlight exposure coincide with high incidences of both rickets and colon cancer [286]. Furthermore, vitamin D supplementation reduces dietary fat-promoted colon carcinogenesis in rats [287], and VDR-null mice exhibit precocious hyperproliferation in the colon descendens where VDR is normally expressed in levels that rival those in skin and small intestine [288]. The paradigm whereby VDR/vitamin D suppresses colon cancer has not been fully characterized, and could involve the ability of VDR to arrest cells at the G1 stage of the cell cycle via induction of p21 [88] and p27 [289], to repress cell growth transcription factors such as c-myc [90] and c-fos [91], or to elicit apoptosis by diminishing the Bcl-2 anti-apoptotic factor [92]. A newly discovered mechanism through which VDR could prevent colon cancer, particularly that induced by high-fat Western diets, is that of chemoprevention by inducing CYP3A4 to detoxify LCA in the colon cells [56,98]. Figure 12A illustrates the role for VDR-RXR in the regulation of intestinal CYP enzymes, in particular CYP3A4. This concept is based upon the recent appreciation that the secondary bile acid, LCA, and its 3-keto metabolite, are novel VDR ligands [98]. Figure 12A also depicts the hepatic synthesis of primary bile acids, focusing on chenodeoxycholic acid, the precursor of LCA. The rate limiting enzyme in this biosynthetic process, CYP7A1, is under positive as well as negative control by VDR-related receptors LXR and FXR, respectively [290]. High liver cholesterol leads

247

CHAPTER 13 Nuclear Vitamin D Receptor

A

Oxysterols

+

LIVER

LXR CYP 7A1 Cholesterol FXR − Chenodeoxycholic acid

Gut bacteria 7-dehydroxylation COO−

LCA HO

H VDR

RXR

LCA

+

Colonocyte

CYP3A4 LCA 6αOH LCA

COLON Colon xenobiotic detoxification

B

ABC transporter 6αOH LCA in lumen

Noggin

Wnt ligand Fz Dsh

BMP4

PTHrP

BMPR

PTHR

Keratinocyte ? β-catenin

SMADs

?

1 BULB

HAIR FOLLICLE

Dermal papilla

2

RXRα

Shh, Hoxc13, msx-1, msx-2

Telogen

FIGURE 12

Hr

Lef1/TCF

novel ligand VDR ?

Anagen

Catagen

Two novel extraosseous actions of the VDR-RXR heterodimer: xenobiotic detoxification in the colon and control of hair cycling. (A) Role of VDR-RXR in detoxification of the secondary bile acid, LCA. The rate-limiting step in the formation of bile acids from cholesterol in liver is catalyzed by CYP7A1 [116]. One of the major primary bile acids, chenodeoxycholic acid, can be converted by gut bacteria to LCA, which may accumulate in the colon since it is not effectively reabsorbed. In the colonocytes, LCA-bound VDR (or 1,25(OH)2D3liganded VDR, not shown), heterodimerized with RXR, activates transcription of the gene encoding CYP3A4. The CYP3A4 enzyme then converts LCA to its 6α-hydroxy form, which is a substrate for the ABC efflux transporter. (B) Proposed action of the VDR-RXR heterodimer in the hair follicle. Keratinocytes or their precursors in the bulge of the follicle are thought to receive signals that stimulate the follicle to exit a resting phase (telogen) and enter a phase of active growth (anagen). Recent research [295,299,300] has identified a number of factors that influence this transition, and a simplified diagram is presented here. Abbreviations not defined in the text are Wnt, ortholog of Drosophila wingless and mouse int-1; Lef1, lymphoid enhancer factor-1; TCF, T cell-specific factor; and msx-1 and msx-2, orthologs of Drosophila muscle-specific homeobox protein. Factors that are membrane receptors are boxed. Solid arrows indicate activation and dotted lines ending in a solid perpendicular line denote inhibition. The black circled 1 and 2 in the BMP/SMAD pathway represent sequential negative and positive signaling as described in the text. The precise point(s) at which VDR-RXR or the RXR-VDR-Hr complex acts has not been characterized, and it is not known whether VDR functions unliganded or occupied by a novel, non-vitamin D ligand.

248 to the production of oxysterols (Fig. 4), which are known to bind to LXR and induce the CYP7A1 gene that catalyzes the first step in bile acid synthesis via 7α-hydroxylation [116]. An opposing regulation of CYP7A1 occurs as a result of high levels of primary bile acids in liver (e.g., chenodeoxycholic acid), which bind to FXR (Fig. 4) and, via an indirect mechanism, inhibit the transcription of CYP7A1 [291]. Formation of the secondary bile acid, LCA, by intestinal bacteria is effected by 7-dehydroxylation of chenodeoxycholic acid [292,293]. LCA then travels to the colon, where its potentially toxic or carcinogenic effects [99,100] are proposed to be ameliorated by the catabolic action of CYP3A4 in the colonocyte (Fig. 12A) [105]. The CYP3A4 gene is transactivated by LCA-bound [98] (or 1,25(OH)2D3bound [55,56]) VDR-RXR, leading to 6α-hydroxylation of LCA [105] and its secretion via an ATP binding cassette (ABC) transporter into the colonic lumen for excretion [290]. Regulation of CYP3A4 was previously known to occur via PXR [59], but experiments with knockout mice have clearly demonstrated that another receptor, presumably VDR [98], is capable of mediating this regulation in PXR−/− mice [55,98]. Therefore, we propose that, in addition to its antiproliferation/ prodifferentiation and proapoptotic actions, VDR specifically operates to prevent high dietary fat-induced colon cancer by detoxifying LCA. This chemoprotective action of VDR via CYP3A4 induction occurs when the receptor is occupied by its alternative ligands, LCA or 1,25(OH)2D3, explaining both how LCA is detoxified, and why this process is amplified in the presence of supplemental vitamin D. A second noncalcemic site of VDR action, in this case clinically relevant to the prevention of hair loss, is the hair follicle (Fig. 12B). As described earlier, alopecia in VDR knockout mice cannot be corrected by excess dietary calcium [79]. Also, vitamin D- or 1,25(OH)2D3deficiency, the latter caused by ablation of the CYP27B1 1α-OHase enyzme [155,156], does not confer alopecia. Thus, the action of VDR to trigger normal hair cycling is independent of vitamin D status, indicating that VDR functions in this capacity either in the unliganded conformation, or occupied by a nonvitamin D-derived lipid constitutively present in skin. Interestingly, conditional knockout of RXRα in skin [128] elicits alopecia indistinguishable from that occurring in VDR-null mice, implying that the functional VDR unit, as with 1,25(OH)2D3-dependent actions, is the VDR-RXR binary complex. Further experiments utilizing VDR knockout animals [294], and with VDR−/− mice in which VDR expression is transgenically targeted to keratinocytes [82], have demonstrated that the absence of VDR function renders hair follicles unable to

G. KERR WHITFIELD

ET AL .

enter the growth (anagen) phase. Therefore, as depicted in Fig. 12B, VDR appears to function in keratinocytes to trigger the telogen (resting phase) to anagen (growth phase) transition in the hair cycle. How does VDR impinge upon hair cycle signaling? According to the hypothetical model presented in Fig. 12B, VDR-RXR could intervene at any of a number of steps in the complex process of anagen initiation. Findings mainly from the laboratory of Fuchs et al. [295] have revealed that the Lef/TCF family of transcription factors, complexed with β-catenin, play a crucial role in the transmodulation of several genes which encode proteins that lead to the initiation of anagen. The list of genes directly or indirectly activated by the Lef/TCF/βcatenin complex includes sonic hedgehog (Shh), Hoxc13, msx-1, and msx-2 [295]. This complex also activates hair specific genes such as keratin 14, and represses E-cadherin, the latter effect diminishing adherens junctions to facilitate hair follicle formation [296]. As illustrated in Fig. 12B, control of Lef1/TCF transcription factor includes at least two upstream pathways, one originating with the binding of an extracellular Wnt ligand to a cell surface receptor of the Frizzled (Fz) family. This signal, acting through the Disheveled factor (Dsh), leads to a stabilization of β-catenin and formation of an active complex with Lef1/TCF. The second pathway involves an inhibitory factor, bone morphogenetic protein-4 (BMP4), that functions in a negative manner to block synthesis of Lef1/TCF, and this inhibition can be relieved by association with noggin, an extracellular ligand secreted by the dermal papilla that acts as an antagonist at the BMP receptor (BMPR) [297,298]. Based upon recent experiments [298] in which the BMPR was knocked out, a second function of BMP/SMAD signaling has been revealed subsequent to its antagonism by noggin to permit Lef1 accumulation in hair follicle matrix cells. Once Lef1 has accumulated in response to the noggin antagonist, this is followed by a positive effect of BMP/SMAD to somehow facilitate the cooperative action of β-catenin and Lef1 to modulate transcription in the cortex cells that generate the hair shaft. Therefore, these pathways all converge in the genesis of the transcriptionally active Lef1/TCF/ β-catenin complex (Fig. 12B). To date, VDR has been studied only in relation to β-catenin in colon cancer cells, and paradoxically, the receptor attenuates β-catenin signaling by competing with Lef1/TCF for β-catenin binding [270]. However, the potential actions of VDRRXR to influence either the Wnt or BMP/noggin pathways in keratinocytes have not as yet been probed, and the receptor heterodimer could affect any number of steps in hair-cycle signal transduction as postulated in Fig. 12B.

249

CHAPTER 13 Nuclear Vitamin D Receptor

A third pathway initiated by yet another extracellular ligand, PTHrP, may be involved in triggering anagen, although recent reports indicate that PTHrP accelerates entry into catagen (apoptosis of the lower follicle) rather that preventing entry into anagen [299,300]. Regardless of which hypothesis for PTHrP action is correct, effective hair growth during anagen would require either inhibition of PTHrP synthesis and secretion, or repression of its receptor (PTHR), in target cells, and as postulated in Fig. 12B, VDR could act as a repressor at either of these steps. Interestingly, transcription of PTHrP is directly inhibited by the 1,25(OH)2D3-occupied VDR-RXR heterodimer through negative VDREs [301,302] (Table I), rendering PTHrP as one of the most attractive candidate targets for repression by VDR in hair follicle signaling. The mechanism whereby VDR-RXR triggers the hair cycle remains unknown. However, naturally occurring point mutations in hVDR that confer alopecia as well as vitamin D-resistant rickets exist in the hVDR DNA binding [74,303,304] and RXR heterodimerization [147] domains, yet mutations that specifically compromise either 1,25(OH)2D3 ligand [147,305] or coactivator [153] contacts yield rickets without alopecia. Thus, neither 1,25(OH)2D3 ligand contacts nor receptor-mediated transactivation is required for VDR to drive the hair cycle; only the RXR heterodimerization and DNA binding functions of VDR appear to be necessary for this action. A tonic inhibitor of hair growth has been hypothesized to exist in telogen skin [306], and a reasonable molecular scenario is that VDR-RXR functions to repress this inhibitor, which may or may not be BMP4 or PTHrP (Fig. 12B). Mechanisms of gene repression by VDR are not well understood (see Section II), and VDR does not associate strongly with traditional nuclear receptor corepressors such as N-CoR and SMRT [307–309]. Recently, VDR has been observed to interact physically and functionally with a novel corepressor, the Hr product of the hairless (hr) gene [157]. Hr recruits HDACs to repress the actions of both TR [310] and ROR [311], and it dramatically inhibits both basal and ligand-stimulated VDRRXR-mediated transcription [157]. Intriguingly, similar to knockout of the genes encoding either VDR or RXRα, inactivating mutations in the mammalian hr gene result in congenital hair loss in both mice [312] and humans [313,314]. Remarkably, the hair loss phenotype elicited by specific mutations in hVDR resembles the atrichia resulting from mutations in the hr gene [315], suggesting that the VDR, RXRα, and Hr proteins all impact a common signaling pathway, perhaps as a trimeric complex (Fig. 12B). Coexpression of hVDR and Hr mRNAs in the matrix and outer root sheath cells of the

mouse skin hair follicle [157] supports this hypothesis. What remains to be determined is where the putative RXRα-VDR-Hr complex impinges on hair cycle signal transduction to trigger the telogen-to-anagen transition, and several possibilities are illustrated in Fig. 12B. Elucidating the molecular role of VDR in controlling the hair cycle clearly represents an important challenge, one that may lead us to fundamental new insight into novel gene repressive mechanisms of VDR action. Finally, understanding this mechanism may facilitate the development of VDR mimetics to prevent or treat hair loss.

VII. SUMMARY AND PERSPECTIVES The vitamin D receptor is a unique member of the nuclear receptor superfamily in the realm of ligand binding, mediating both the endocrine control of bone mineral metabolism and the detoxification of endobiotics such as LCA. VDR functions as a heterodimer with unliganded RXR, with the binary protein complex required for recognition and high-affinity association with VDREs in the promoters of regulated genes, as well as for the recruitment of transcriptional comodulator proteins that govern the activity of RNA polymerase II. As a nuclear receptor that heterodimerizes with RXR, VDR therefore possesses the molecular character both of its evolutionarily closest endocrine relative, the thyroid hormone receptor, and of its closest xenobiotic detoxifying receptor, PXR. The biology of VDR signaling primarily involves (i) stimulation of intestinal calcium and phosphate absorption to prevent rickets/ osteomalacia, (ii) enhancement of bone remodeling via osteoblast-induced osteoclast maturation, (iii) differentiation of skin cells and maintenance of the hair cycle, (iv) repression/induction of CYP enzymes for 1,25(OH)2D3 hormone synthesis/degradation as well as for the promotion of secondary bile acid detoxification, (v) modulation of the immune system, and (vi) potential anticancer actions via the control of epithelial cell growth, differentiation, and apoptosis. Based upon biochemical mutagenesis experiments, the X-ray crystallographic characterization of the DNAand ligand-binding domains, and naturally occurring mutations that confer on patients the vitamin D-resistant phenotype of rickets and/or alopecia, the structurefunction of VDR is reasonably well understood. Amino acid residues in an α-helical region of the zinc finger domain of VDR that contact specific DNA base pairs in VDRE docking sites have been identified, as have nuclear localization signals in this domain. The more structurally complex LBD of VDR consists of a sandwich

250 of α-helices surrounding a hydrophobic pocket that accommodates lipophilic ligands, as well as surface facets for the attraction of numerous interacting proteins including the RXR dimerization partner. The VDR-RXR heterodimer, which is allosterically influenced by 1,25(OH)2D3 and other VDR ligands, associates with (i) ATP-dependent SWI/SNF chromatin remodeling complexes (e.g., PBAF and WINAC) that apparently slide nucleosomes along DNA and are required for both induction and repression; (ii) p160 platform coactivators (e.g., SRC-1) that recruit CBP/p300 to effect HAT modification of chromatin that disrupts internucleosomal interactions to convert heterochromatin to transcriptionally active euchromatin; and (iii) DRIP205/Mediator complexes that bridge to the C-terminal domain of RNA polymerase II plus TFIIB, which links VDR-RXR to transcriptional initiation. Alternatively, VDR-RXR may associate with corepressors such as Hr that attract HDACs to return chromatin structure to the transcriptionally inactive heterochromatin state in the vicinity of gene promoters negatively regulated by VDR. The functional activity of hVDR appears to be influenced by common gene polymorphisms, with certain 5′ promoter (Cdx-2), translation start site (F/f ), and 3′-UTR (L/S) alleles yielding higher expression of (Cdx-2 A-allele and 3′-UTR L-allele), or intrinsically more transcriptionally active (translation start site F-allele), hVDR. Such polymorphic hVDR variants affect bone mineral density and may relate to the epidemiology of osteoporosis, as well as significantly alter the risk of certain epithelial cancers. In conclusion, in the time elapsed since the last edition of this volume, we have witnessed an explosion of new information enhancing our understanding of VDR, including crystallographic views of both the DBD and ligand binding/heterodimerization domains, availability of the complete human genome sequence, and thorough characterizations of VDR and RXR knockout mice. However, many questions still remain unanswered. A major structure-function breakthrough would be the X-ray crystallographic solution of fulllength VDR, or at least of a closely related nuclear receptor, so that we can begin to understand the interactions between the DBD and the ligand binding/ heterodimerization domains. A second major realm of future investigation is the identification of additional VDR-regulated genes, particularly those supporting intestinal calcium and phosphate transport, as the molecular mechanism whereby VDR exerts this primary action is currently uncharacterized. An equally exciting endeavor will be the discovery of additional natural VDR or RXR ligands, especially in the more recently identified and studied VDR target tissues such as the hair follicle/skin and immune systems. The availability

G. KERR WHITFIELD

ET AL .

of this information would allow for molecular investigations of the VDR transcriptional control cycle (Fig. 10) in the presence of novel ligands and/or in the context of other promoters in addition to those of the rat osteocalcin or mouse osteopontin genes that have heretofore been the focus of VDR transactivation studies. Such experiments will likely confirm and extend our understanding of the variety of conformations and coregulator associations that the VDR-RXR heterodimer is capable of achieving. Moreover, insights from the approaches just outlined may then enable us to comprehend the interactions between two or more VDREs in the same vitamin D regulated promoter, or between the VDR-RXR heterodimer and other transactivators that are bound to the same promoter. Finally, all of these concepts need to be applied to the poorly understood topic of negative regulation by the VDRRXR heterodimer. Transrepression by VDR is likely complex, and no doubt involves a variety of mechanisms, but its characterization will increase our knowledge of such fundamental VDR tenets as suppression of PTH and the 1α-OHase enzyme in relation to calcium/ phosphate homeostasis, control of PTHrP and other hair cycle-related genes with respect to preventing alopecia, and regulation of Bcl-2 and other genes controlling cell growth and apoptosis that may play a role in the action of VDR as a nuclear receptor sentinel in preventing epithelial cell cancers. Clearly, there are many challenges ahead in our molecular probing of VDR and, undoubtedly, a few more surprises as well.

Acknowledgments Supported by NIH grants to M.R.H. The authors thank other members of our laboratory, Hope T. L. Dang, Jamie Dawson, Neal Hall, Magdalena J. Kaczmarska, and Stephanie A. Slater, for their contributions.

References 1. Haussler MR, Myrtle JF, Norman AW 1968 The association of a metabolite of vitamin D3 with intestinal mucosa chromatin in vivo. J Biol Chem 243:4055–4064. 2. Haussler MR, Norman AW 1969 Chromosomal receptor for a vitamin D metabolite. Proc Natl Acad Sci USA 62:155–162. 3. Myrtle JF, Haussler MR, Norman AW 1970 Evidence for the biologically active form of cholecalciferol in the intestine. J Biol Chem 245:1190–1196. 4. Haussler MR, Boyce DW, Littledike ET, Rasmussen H 1971 A rapid acting metabolite of vitamin D3. Proc Natl Acad Sci USA 68:177–181. 5. Holick MF, Schnoes HK, DeLuca HF, Suda T, Cousins RJ 1971 Isolation and identification of 1,25-dihydroxycholecalciferol, a metabolite of vitamin D active in the intestine. Biochemistry 10:2799–2804.

251

CHAPTER 13 Nuclear Vitamin D Receptor

6. Lawson DEM, Fraser DR, Kodicek E, Morris HR, Williams DH 1971 Identification of 1,25-dihydroxycholecalciferol, a new kidney hormone controlling calcium metabolism. Nature 230:228–230. 7. Fraser DR, Kodicek E 1970 Unique biosynthesis by kidney of a biologically active vitamin D metabolite. Nature 228: 764–766. 8. Brumbaugh PF, Haussler MR 1974 1α,25-Dihydroxycholecalciferol receptors in intestine. I. Association of 1α,25dihydroxycholecalciferol with intestinal mucosa chromatin. J Biol Chem 249:1251–1257. 9. Brumbaugh PF, Haussler MR 1974 1α,25-Dihydroxycholecalciferol receptors in intestine. II. Temperature-dependent transfer of the hormone to chromatin via a specific cytosol receptor. J Biol Chem 249:1258–1262. 10. Brumbaugh PF, Haussler MR 1973 1α,25-Dihydroxyvitamin D3 receptor: Competitive binding of vitamin D analogs. Life Sci 13:1737–1746. 11. Norman AW, Procsal DA, Okamura WH, Wing RM 1975 Structure-function studies of the interaction of the hormonally active form of vitamin D3, 1α,25-dihydroxy-vitamin D3, with the intestine. J Steroid Biochem 6:461–467. 12. Pike JW, Haussler MR 1979 Purification of chicken intestinal receptor for 1,25-dihydroxyvitamin D. Proc Natl Acad Sci USA 76:5485–5489. 13. Pike JW, Donaldson CA, Marion SL, Haussler MR 1982 Development of hybridomas secreting monoclonal antibodies to the chicken intestinal 1α,25-dihydroxyvitamin D3 receptor. Proc Natl Acad Sci USA 79:7719–7723. 14. Pike JW, Marion SL, Donaldson CA, Haussler MR 1983 Serum and monoclonal antibodies against the chick intestinal receptor for 1,25-dihydroxyvitamin D3. J Biol Chem 258:1289–1296. 15. McDonnell DP, Mangelsdorf DJ, Pike JW, Haussler MR, O’Malley BW 1987 Molecular cloning of complementary DNA encoding the avian receptor for vitamin D. Science 235:1214–1217. 16. Evans RM 1988 The steroid and thyroid hormone receptor superfamily. Science 240:889–895. 17. Pike JW 1981 Evidence for a reactive sulfhydryl in the DNA binding domain of the 1,25-dihydroxyvitamin D3 receptor. Biochem Biophys Res Commun 100:1713–1719. 18. Haussler MR, Mangelsdorf DJ, Yamaoka K, Allegretto EA, Komm BS, Terpening CM, McDonnell DP, Pike JW, O’Malley BW 1988 Molecular characterization and actions of the vitamin D hormone receptor. In: Ringold G, ed. Steroid Hormone Action, Vol. 75. Alan R. Liss, New York, pp. 247–262. 19. Baker AR, McDonnell DP, Hughes MR, Crisp TM, Mangelsdorf DJ, Haussler MR, Pike JW, Shine J, O’Malley BW 1988 Cloning and expression of full-length cDNA encoding human vitamin D receptor. Proc Natl Acad Sci USA 85:3294–3298. 20. Burmester JK, Wiese RJ, Maeda N, DeLuca HF 1988 Structure and regulation of the rat 1,25-dihydroxyvitamin D3 receptor. Proc Natl Acad Sci USA 85:9499–9502. 21. Lu Z, Hanson K, DeLuca HF 1997 Cloning and origin of the two forms of chicken vitamin D receptor. Arch Biochem Biophys 339:99–106. 22. Sone T, Ozono K, Pike JW 1991 A 55-kilodalton accessory factor facilitates vitamin D receptor DNA binding. Mol Endocrinol 5:1578–1586. 23. MacDonald PN, Haussler CA, Terpening CM, Galligan MA, Reeder MC, Whitfield GK, Haussler MR 1991 Baculovirusmediated expression of the human vitamin D receptor: functional characterization, vitamin D response element

24.

25. 26. 27.

28. 29.

30. 31. 32.

33. 34.

35.

36. 37.

38.

39.

interactions, and evidence for a receptor auxiliary factor. J Biol Chem 266:18808–18813. Ross TK, Moss VE, Prahl JM, DeLuca HF 1992 A nuclear protein essential for binding of rat 1,25-dihydroxyvitamin D3 receptor to its response elements. Proc Natl Acad Sci USA 89:256–260. Mangelsdorf DJ, Evans RM 1995 The RXR heterodimers and orphan receptors. Cell 83:841–850. Heyman RA, Mangelsdorf DJ, Dyck JA, Stein RB, Eichele G, Evans RM, Thaller C 1992 9-Cis retinoic acid is a high affinity ligand for the retinoid X receptor. Cell 68:397–406. MacDonald PN, Dowd DR, Nakajima S, Galligan MA, Reeder MC, Haussler CA, Ozato K, Haussler MR 1993 Retinoid X receptors stimulate and 9-cis retinoic acid inhibits 1,25-dihydroxyvitamin D3-activated expression of the rat osteocalcin gene. Mol Cell Biol 13:5907–5917. Jin CH, Kerner SA, Hong MH, Pike JW 1996 Transcriptional activation and dimerization functions in the human vitamin D receptor. Mol Endocrinol 10:945–957. Cheskis B, Freedman LP 1994 Ligand modulates the conversion of DNA-bound vitamin D3 receptor (VDR) homodimers into VDR-retinoid X receptor heterodimers. Mol Cell Biol 14:3329–3338. Lemon BD, Freedman LP 1996 Selective effects of ligands on vitamin D3 receptor- and retinoid X receptor-mediated gene activation in vivo. Mol Cell Biol 16:1006–1016. Cheskis B, Freedman LP 1996 Modulation of nuclear receptor interactions by ligands: Kinetic analysis using surface plasmon resonance. Biochemistry 35:3309–3318. Thompson PD, Jurutka PW, Haussler CA, Whitfield GK, Haussler MR 1998 Heterodimeric DNA binding by the vitamin D receptor and retinoid X receptors is enhanced by 1,25-dihydroxyvitamin D3 and inhibited by 9-cis retinoic acid: evidence for allosteric receptor interactions. J Biol Chem 273:8483–8491. MacDonald PN, Dowd DR, Haussler MR 1994 New insight into the structure and functions of the vitamin D receptor. Sem Nephrol 14:101–118. Haussler MR, Jurutka PW, Hsieh J-C, Thompson PD, Selznick SH, Haussler CA, Whitfield GK 1995 New understanding of the molecular mechanism of receptor-mediated genomic actions of the vitamin D hormone. Bone 17(Suppl 2): 33S–38S. Haussler MR, Whitfield GK, Haussler CA, Hsieh J-C, Thompson PD, Selznick SH, Encinas Dominguez C, Jurutka PW 1998 The nuclear vitamin D receptor: biological and molecular regulatory properties revealed. J Bone Miner Res 13:325–349. Cheng JB, Motola DL, Mangelsdorf DJ, Russell DW 2003 De-orphanization of cytochrome P450 2R1: A microsomal vitamin D 25-hydroxylase. J Biol Chem 278:38084–38093. DeMay MB, Kiernan MS, DeLuca HF, Kronenberg HM 1992 Sequences in the human parathyroid hormone gene that bind the 1,25-dihydroxyvitamin D3 receptor and mediate transcriptional repression in response to 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 89:8097–8101. Kitazawa R, Kitazawa S 2002 Vitamin D3 augments osteoclastogenesis via vitamin D-responsive element of mouse RANKL gene promoter. Biochem Biophys Res Commun 290:650–655. Yasuda H, Shima N, Nakagawa N, Yamaguchi K, Kinosaki M, Mochizuki S, Tomoyasu A, Yano K, Goto M, Murakami A, Tsuda E, Morinaga T, Higashio K, Udagawa N, Takahashi N, Suda T 1998 Osteoclast differentiation factor is a ligand for

252

40.

41.

42. 43. 44. 45.

46. 47. 48.

49.

50.

51.

52.

53.

54.

G. KERR WHITFIELD

osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci USA 95:3597–3602. Terpening CM, Haussler CA, Jurutka PW, Galligan MA, Komm BS, Haussler MR 1991 The vitamin D-responsive element in the rat bone gla protein is an imperfect direct repeat that cooperates with other cis-elements in 1,25dihydroxyvitamin D3-mediated transcriptional activation. Mol Endocrinol 5:373–385. Noda M, Vogel RL, Craig AM, Prahl J, DeLuca HF, Denhardt DT 1990 Identification of a DNA sequence responsible for binding of the 1,25-dihydroxyvitamin D3 receptor and 1,25-dihydroxyvitamin D3 enhancement of mouse secreted phosphoprotein 1 (Spp-1 or osteopontin) gene expression. Proc Natl Acad Sci USA 87:9995–9999. Horowitz MC 1993 Cytokines and estrogen in bone: anti-osteoporotic effects. Science 260:626–627. Bikle DD, Pillai S 1993 Vitamin D, calcium and epidermal differentiation. Endocr Rev 14:3–19. Manolagas SC, Yu X-P, Girasole G, Bellido T 1994 Vitamin D and the hematolymphopoietic tissue: A 1994 update. Sem Nephrol 14:129–143. Bikle DD, Ng D, Oda Y, Hanley K, Feingold K, Xie Z 2002 The vitamin D response element of the involucrin gene mediates its regulation by 1,25-dihydroxyvitamin D3. J Invest Dermatol 119:1109–1113. Quelo I, Machuca I, Jurdic P 1998 Identification of a vitamin D response element in the proximal promoter of the chicken carbonic anhydrase II gene. J Biol Chem 273:10638–10646. Omdahl JL, Morris HA, May BK 2002 Hydroxylase enzymes of the vitamin D pathway: expression, function, and regulation. Ann Rev Nutr 22:139–166. Adams JS, Singer FR, Gacad MA, Sharma OP, Hayes MJ, Vouros P, Holick MF 1985 Isolation and structural identification of 1,25-dihydroxyvitamin D3 produced by cultured alveolar macrophages in sarcoidosis. J Clin Endocrinol Metab 60:960–966. Ohyama Y, Ozono K, Uchida M, Shinki T, Kato S, Suda T, Yamamoto O, Noshiro M, Kato Y 1994 Identification of a vitamin D-responsive element in the 5′ flanking region of the rat 25-hydroxyvitamin D3 24-hydroxylase gene. J Biol Chem 269:10545–10550. Zierold C, Darwish HM, DeLuca HF 1994 Identification of a vitamin D-responsive element in the rat calcidiol (25-hydroxyvitamin D3) 24-hydroxylase gene. Proc Natl Acad Sci USA 91:900–902. Murayama A, Takeyama K, Kitanaka S, Kodera Y, Kawaguchi Y, Hosoya T, Kato S 1999 Positive and negative regulations of the renal 25-hydroxyvitamin D3 1α-hydroxylase gene by parathyroid hormone, calcitonin, and 1α,25(OH)2D3 in intact animals. Endocrinology 140:2224–2231. Colnot S, Lambert M, Blin C, Thomasset M, Perret C 1995 Identification of DNA sequences that bind retinoid X receptor1,25(OH)2D3-receptor heterodimers with high affinity. Mol Cell Endocrinol 113:89–98. Nishikawa J, Kitaura M, Matsumoto M, Imagawa M, Nishihara T 1994 Difference and similarity of DNA sequence recognized by VDR homodimer and VDR/RXR heterodimer. Nucleic Acids Res 22:2902–2907. Jurutka PW, MacDonald PN, Nakajima S, Hsieh J-C, Thompson PD, Whitfield GK, Galligan MA, Haussler CA, Haussler MR 2002 Isolation of baculovirus-expressed human vitamin D receptor: DNA responsive element interactions and phosphorylation of the purified receptor. J Cell Biochem 85:435–457.

ET AL .

55. Thummel KE, Brimer C, Yasuda K, Thottassery J, Senn T, Lin Y, Ishizuka H, Kharasch E, Schuetz J, Schuetz E 2001 Transcriptional control of intestinal cytochrome P-4503A by 1alpha,25-dihydroxy vitamin D3. Mol Pharmacol 60: 1399–1406. 56. Thompson PD, Jurutka PW, Whitfield GK, Myskowski SM, Eichhorst KR, Encinas Dominguez C, Haussler CA, Haussler MR 2002 Liganded VDR induces CYP3A4 in small intestinal and colon cancer cells via DR3 and ER6 vitamin D responsive elements. Biochem Biophys Res Commun 299:730–738. 57. Bertilsson G, Heidrich J, Svensson K, Asman M, Jendeberg L, Sydow-Backman M, Ohlsson R, Postlind H, Blomquist P, Berkenstam A 1998 Identification of a human nuclear receptor defines a new signaling pathway for CYP3A induction. Proc Natl Acad Sci USA 95:12208–12213. 58. Blumberg B, Sabbagh W, Jr., Juguilon H, Bolado J, Jr., van Meter CM, Ong ES, Evans RM 1998 SXR, a novel steroid and xenobiotic-sensing nuclear receptor. Genes Dev 12:3195–3205. 59. Lehmann JM, McKee DD, Watson MA, Willson TM, Moore JT, Kliewer SA 1998 The human orphan nuclear receptor PXR is activated by compounds that regulate CYP3A4 gene expression and cause drug interactions. J Clin Invest 102:1016–1023. 60. Hartong R, Wang N, Kurokawa R, Lazar MA, Glass CK, Apriletti JW, Dillmann WH 1994 Delineation of three different thyroid hormone-response elements in promoter of rat sarcoplasmic reticulum Ca2+ ATPase gene. J Biol Chem 269: 13021–13029. 61. Baniahmad A, Steiner C, Kohne AC, Renkawitz R 1990 Modular structure of a chicken lysozyme silencer: involvement of an unusual thyroid hormone receptor binding site. Cell 61:505–514. 62. van Den Bemd GJ, Jhamai M, Staal A, van Wijnen AJ, Lian JB, Stein GS, Pols HA, van Leeuwen JP 2002 A central dinucleotide within vitamin D response elements modulates DNA binding and transactivation by the vitamin D receptor in cellular response to natural and synthetic ligands. J Biol Chem 277:14539–14546. 63. Staal A, van Wijnen AJ, Birkenhäger JC, Pols HAP, Prahl J, DeLuca H, Gaub M-P, Lian JB, Stein GS, van Leeuwen JPTM, Stein JL 1996 Distinct conformations of vitamin D receptor/ retinoid X receptor-α heterodimers are specified by dinucleotide differences in the vitamin D-responsive elements for the osteocalcin and osteopontin genes. Mol Endocrinol 10:1444–1456. 64. Koszewski NJ, Ashok S, Russell J 1999 Turning a negative into a positive: vitamin D receptor interactions with the avian parathyroid hormone response element. Mol Endocrinol 13:455–465. 65. Kurokawa R, Söderström M, Hörlein A, Halachmi S, Brown M, Rosenfeld MG, Glass CK 1995 Polarity-specific activities of retinoic acid receptors determined by a co-repressor. Nature 377:451–454. 66. Jurutka PW, Whitfield GK, Hsieh J-C, Thompson PD, Haussler CA, Haussler MR 2001 Molecular nature of the vitamin D receptor and its role in regulation of gene expression. Rev Endocr Metab Disord 2:203–216. 67. Lyakhovich A, Aksenov N, Pennanen P, Miettinen S, Ahonen MH, Syvala H, Ylikomi T, Tuohimaa P 2000 Vitamin D induced up-regulation of keratinocyte growth factor (FGF7/KGF) in MCF-7 human breast cancer cells. Biochem Biophys Res Commun 273:675–680. 68. Farach-Carson MC, Xu Y 2002 Microarray detection of gene expression changes induced by 1,25(OH)2D3 and a Ca2+

CHAPTER 13 Nuclear Vitamin D Receptor

69. 70. 71.

72.

73. 74. 75.

76.

77.

78.

79.

80.

81. 82.

83.

influx-activating analog in osteoblastic ROS 17/2.8 cells. Steroids 67:467–470. Krishnan AV, Peehl DM, Feldman D 2003 Inhibition of prostate cancer growth by vitamin D: Regulation of target gene expression. J Cell Biochem 88:363–371. Li X, Zheng W, Li YC 2003 Altered gene expression profile in the kidney of vitamin D receptor knockout mice. J Cell Biochem 89:709–719. Swami S, Raghavachari N, Muller UR, Bao YP, Feldman D 2003 Vitamin D growth inhibition of breast cancer cells: gene expression patterns assessed by cDNA microarray. Breast Cancer Res Treat 80:49–62. Feldman D, Malloy PJ 1990 Hereditary 1,25-dihydroxyvitamin D resistant rickets: molecular basis and implications for the role of 1,25(OH)2D3 in normal physiology. Mol Cell Endocrinol 72:C57–62. Faraco JH, Morrison NA, Baker A, Shine J, Frossard PM 1989 ApaI dimorphism at the human vitamin D receptor gene locus. Nucleic Acids Res 17:2150. Malloy PJ, Pike JW, Feldman D 1999 The vitamin D receptor and the syndrome of hereditary 1,25-dihydroxyvitamin Dresistant rickets. Endocr Rev 20:156–188. Rut AR, Hewison M, Kristjansson K, Luisi B, Hughes MR, O’Riordan JLH 1994 Two mutations causing vitamin D resistant rickets: modelling on the basis of steroid hormone receptor DNA-binding domain crystal structures. Clin Endocrinol (Oxf) 41:581–590. Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997 Targeted ablation of the vitamin D receptor: an animal model of vitamin D-dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94: 9831–9835. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Genet 16:391–396. Van Cromphaut SJ, Dewerchin M, Hoenderop JG, Stockmans I, Van Herck E, Kato S, Bindels RJ, Collen D, Carmeliet P, Bouillon R, Carmeliet G 2001 Duodenal calcium absorption in vitamin D receptor-knockout mice: functional and molecular aspects. Proc Natl Acad Sci USA 98:13324–13329. Li YC, Amling M, Pirro AE, Priemel M, Meuse J, Baron R, Delling G, Demay MB 1998 Normalization of mineral ion homeostasis by dietary means prevents hyperparathyroidism, rickets, and osteomalacia, but not alopecia in vitamin D receptor-ablated mice. Endocrinology 139:4391–4396. Amling M, Priemel M, Holzmann T, Chapin K, Rueger JM, Baron R, Demay MB 1999 Rescue of the skeletal phenotype of vitamin D receptor-ablated mice in the setting of normal mineral ion homeostasis: formal histomorphometric and biomechanical analyses. Endocrinology 140:4982–4987. Sakai Y, Demay MB 2000 Evaluation of keratinocyte proliferation and differentiation in vitamin D receptor knockout mice. Endocrinology 141:2043–2049. Chen CH, Sakai Y, Demay MB 2001 Targeting expression of the human vitamin D receptor to the keratinocytes of vitamin D receptor null mice prevents alopecia. Endocrinology 142: 5386–5389. Kinuta K, Tanaka H, Moriwake T, Aya K, Kato S, Seino Y 2000 Vitamin D is an important factor in estrogen biosynthesis of both female and male gonads. Endocrinology 141:1317–1324.

253 84. Zinser G, Packman K, Welsh J 2002 Vitamin D3 receptor ablation alters mammary gland morphogenesis. Development 129:3067–3076. 85. Zeitz U, Weber K, Soegiarto DW, Wolf E, Balling R, Erben RG 2003 Impaired insulin secretory capacity in mice lacking a functional vitamin D receptor. FASEB J 17:509–511. 86. O’Kelly J, Hisatake J, Hisatake Y, Bishop J, Norman A, Koeffler HP 2002 Normal myelopoiesis but abnormal T lymphocyte responses in vitamin D receptor knockout mice. J Clin Invest 109:1091–1099. 87. Zinser GM, Sundberg JP, Welsh J 2002 Vitamin D3 receptor ablation sensitizes skin to chemically induced tumorigenesis. Carcinogenesis 23:2103–2109. 88. Liu M, Lee MH, Cohen M, Bommakanti M, Freedman LP 1996 Transcriptional activation of the Cdk inhibitor p21 by vitamin D3 leads to the induced differentiation of the myelomonocytic cell line U937. Genes Dev 10:142–153. 89. Yang ES, Burnstein KL 2003 Vitamin D inhibits G1 to S progression in LNCaP prostate cancer cells through p27Kip1 stabilization and Cdk2 mislocalization to the cytoplasm. J Biol Chem 278:46862–46868. 90. Saunders DE, Christensen C, Wappler NL, Schultz JF, Lawrence WD, Malviya VK, Malone JM, Deppe G 1993 Inhibition of c-myc in breast and ovarian carcinoma cells by 1,25-dihydroxyvitamin D3, retinoic acid and dexamethasone. Anticancer Drugs 4:201–208. 91. Candeliere GA, Jurutka PW, Haussler MR, St-Arnaud R 1996 A composite element binding the vitamin D receptor, retinoid X receptor α, and a member of the CTF/NF-1 family of transcription factors mediates the vitamin D responsiveness of the c-fos promoter. Mol Cell Biol 16:584–592. 92. Elstner E, Linker-Israeli M, Umiel T, Le J, Grillier I, Said J, Shintaku IP, Krajewski S, Reed JC, Binderup L, Koeffler HP 1996 Combination of a potent 20-epi-vitamin D3 analogue (KH 1060) with 9-cis-retinoic acid irreversibly inhibits clonal growth, decreases bcl-2 expression, and induces apoptosis in HL-60 leukemic cells. Cancer Res 56:3570–3576. 93. Lander ES, Linton LM, Birren B, Nusbaum C, Zody MC, Baldwin J, Devon K, Dewar K, Doyle M, FitzHugh W, Funke R, Gage D, Harris K, Heaford A, Howland J, Kann L, Lehoczky J, LeVine R, McEwan P, McKernan K, Meldrim J, Mesirov JP, Miranda C, Morris W, Naylor J, Raymond C, Rosetti M, Santos R, Sheridan A, Sougnez C, Stange-Thomann N, Stojanovic N, Subramanian A, Wyman D, Rogers J, Sulston J, Ainscough R, Beck S, Bentley D, Burton J, Clee C, Carter N, Coulson A, Deadman R, Deloukas P, Dunham A, Dunham I, Durbin R, French L, Grafham D, Gregory S, Hubbard T, Humphray S, Hunt A, Jones M, Lloyd C, McMurray A, Matthews L, Mercer S, Milne S, Mullikin JC, Mungall A, Plumb R, Ross M, Shownkeen R, Sims S, Waterston RH, Wilson RK, Hillier LW, McPherson JD, Marra MA, Mardis ER, Fulton LA, Chinwalla AT, Pepin KH, Gish WR, Chissoe SL, Wendl MC, Delehaunty KD, Miner TL, Delehaunty A, Kramer JB, Cook LL, Fulton RS, Johnson DL, Minx PJ, Clifton SW, Hawkins T, Branscomb E, Predki P, Richardson P, Wenning S, Slezak T, Doggett N, Cheng JF, Olsen A, Lucas S, Elkin C, Uberbacher E, Frazier M, Gibbs RA, Muzny DM, Scherer SE, Bouck JB, Sodergren EJ, Worley KC, Rives CM, Gorrell JH, Metzker ML, Naylor SL, Kucherlapati RS, Nelson DL, Weinstock GM, Sakaki Y, Fujiyama A, Hattori M, Yada T, Toyoda A, Itoh T, Kawagoe C, Watanabe H, Totoki Y, Taylor T, Weissenbach J, Heilig R, Saurin W, Artiguenave F, Brottier P, Bruls T, Pelletier E, Robert C, Wincker P, Smith DR, Doucette-Stamm L, Rubenfield M, Weinstock K, Lee HM, Dubois J, Rosenthal A,

254

94. 95. 96. 97.

98.

99. 100. 101.

102.

103.

104.

G. KERR WHITFIELD

Platzer M, Nyakatura G, Taudien S, Rump A, Yang H, Yu J, Wang J, Huang G, Gu J, Hood L, Rowen L, Madan A, Qin S, Davis RW, Federspiel NA, Abola AP, Proctor MJ, Myers RM, Schmutz J, Dickson M, Grimwood J, Cox DR, Olson MV, Kaul R, Shimizu N, Kawasaki K, Minoshima S, Evans GA, Athanasiou M, Schultz R, Roe BA, Chen F, Pan H, Ramser J, Lehrach H, Reinhardt R, McCombie WR, de la Bastide M, Dedhia N, Blocker H, Hornischer K, Nordsiek G, Agarwala R, Aravind L, Bailey JA, Bateman A, Batzoglou S, Birney E, Bork P, Brown DG, Burge CB, Cerutti L, Chen HC, Church D, Clamp M, Copley RR, Doerks T, Eddy SR, Eichler EE, Furey TS, Galagan J, Gilbert JG, Harmon C, Hayashizaki Y, Haussler D, Hermjakob H, Hokamp K, Jang W, Johnson LS, Jones TA, Kasif S, Kaspryzk A, Kennedy S, Kent WJ, Kitts P, Koonin EV, Korf I, Kulp D, Lancet D, Lowe TM, McLysaght A, Mikkelsen T, Moran JV, Mulder N, Pollara VJ, Ponting CP, Schuler G, Schultz J, Slater G, Smit AF, Stupka E, Szustakowski J, Thierry-Mieg D, Thierry-Mieg J, Wagner L, Wallis J, Wheeler R, Williams A, Wolf YI, Wolfe KH, Yang SP, Yeh RF, Collins F, Guyer MS, Peterson J, Felsenfeld A, Wetterstrand KA, Patrinos A, Morgan MJ, Szustakowki J, de Jong P, Catanese JJ, Osoegawa K, Shizuya H, Choi S, Chen YJ 2001 Initial sequencing and analysis of the human genome. Nature 409:860–921. Robinson-Rechavi M, Carpentier AS, Duffraisse M, Laudet V 2001 How many nuclear hormone receptors are there in the human genome? Trends Genet 17:554–556. Laudet V, Auwerx J, Gustafsson J-A, Wahli W 1999 A unified nomenclature system for the nuclear receptor superfamily. Cell 97:161–163. Escriva H, Delaunay F, Laudet V 2000 Ligand binding and nuclear receptor evolution. Bioessays 22:717–727. Moore LB, Maglich JM, McKee DD, Wisely B, Willson TM, Kliewer SA, Lambert MH, Moore JT 2002 Pregnane X receptor (PXR), constitutive androstane receptor (CAR), and benzoate X receptor (BXR) define three pharmacologically distinct classes of nuclear receptors. Mol Endocrinol 16:977–986. Makishima M, Lu TT, Xie W, Whitfield GK, Domoto H, Evans RM, Haussler MR, Mangelsdorf DJ 2002 Vitamin D receptor as an intestinal bile acid sensor. Science 296: 1313–1316. Nagengast FM, Grubben MJ, van Munster IP 1995 Role of bile acids in colorectal carcinogenesis. Eur J Cancer 31A:1067–1070. Hamada K, Umemoto A, Kajikawa A, Seraj MJ, Monden Y 1994 In vitro formation of DNA adducts with bile acids. Carcinogenesis 15:1911–1915. Staudinger JL, Goodwin B, Jones SA, Hawkins-Brown D, MacKenzie KI, LaTour A, Liu Y, Klaassen CD, Brown KK, Reinhard J, Willson TM, Koller BH, Kliewer SA 2001 The nuclear receptor PXR is a lithocholic acid sensor that protects against liver toxicity. Proc Natl Acad Sci USA 98:3369–3374. Xie W, Radominska-Pandya A, Shi Y, Simon CM, Nelson MC, Ong ES, Waxman DJ, Evans RM 2001 An essential role for nuclear receptors SXR/PXR in detoxification of cholestatic bile acids. Proc Natl Acad Sci USA 98:3375–3380. Parks DJ, Blanchard SG, Bledsoe RK, Chandra G, Consler TG, Kliewer SA, Stimmel JB, Willson TM, Zavacki AM, Moore DD, Lehmann JM 1999 Bile acids: natural ligands for an orphan nuclear receptor. Science 284:1365–1368. Whitfield GK, Dang HTL, Schluter SF, Bernstein RM, Bunag T, Manzon LA, Hsieh G, Dominguez CE, Youson JH, Haussler MR, Marchalonis JJ 2003 Cloning of a functional

105. 106.

107.

108. 109.

110. 111.

112.

113.

114.

115.

116. 117.

118. 119. 120.

ET AL .

vitamin D receptor from the lamprey (Petromyzon marinus), an ancient vertebrate lacking a calcified skeleton and teeth. Endocrinology 144:2704–2716. Araya Z, Wikvall K 1999 6α-Hydroxylation of taurochenodeoxycholic acid and lithocholic acid by CYP3A4 in human liver microsomes. Biochim Biophys Acta 1438:47–54. Drocourt L, Ourlin JC, Pascussi JM, Maurel P, Vilarem MJ 2002 Expression of CYP3A4, CYP2B6, and CYP2C9 is regulated by the vitamin D receptor pathway in primary human hepatocytes. J Biol Chem 277:25125–25132. Nelson DR, Koymans L, Kamataki T, Stegeman JJ, Feyereisen R, Waxman DJ, Waterman MR, Gotoh O, Coon MJ, Estabrook RW, Gunsalus IC, Nebert DW 1996 P450 superfamily: update on new sequences, gene mapping, accession numbers and nomenclature. Pharmacogenetics 6:1–42. Goodwin B, Redinbo MR, Kliewer SA 2002 Regulation of CYP3A gene transcription by the pregnane X receptor. Annu Rev Pharmacol Toxicol 42:1–23. Kliewer SA, Moore JT, Wade L, Staudinger JL, Watson MA, Jones SA, McKee DD, Oliver BB, Willson TM, Zetterstrom RH, Perlmann T, Lehmann JM 1998 An orphan nuclear receptor activated by pregnanes defines a novel steroid signaling pathway. Cell 92:73–82. Honkakoski P, Sueyoshi T, Negishi M 2003 Drug-activated nuclear receptors CAR and PXR. Ann Med 35:172–182. Guo GL, Lambert G, Negishi M, Ward JM, Brewer HB, Jr., Kliewer SA, Gonzalez FJ, Sinal CJ 2003 Complementary roles of farnesoid X receptor, pregnane X receptor, and constitutive androstane receptor in protection against bile acid toxicity. J Biol Chem 278:45062–45071. Maglich JM, Stoltz CM, Goodwin B, Hawkins-Brown D, Moore JT, Kliewer SA 2002 Nuclear pregnane X receptor and constitutive androstane receptor regulate overlapping but distinct sets of genes involved in xenobiotic detoxification. Mol Pharmacol 62:638–646. Gill RK, Christakos S 1993 Identification of sequence elements in mouse calbindin-D28K gene that confer 1,25dihydroxyvitamin D3- and butyrate-inducible responses. Proc Natl Acad Sci USA 90:2984–2988. Clemens TL, Garrett KP, Zhou X-Y, Pike JW, Haussler MR, Dempster DW 1988 Immunocytochemical localization of the 1,25-dihydroxyvitamin D3 receptor in target cells. Endocrinology 122:1224–1230. Baes M, Gulick T, Choi HS, Martinoli MG, Simha D, Moore DD 1994 A new orphan member of the nuclear hormone receptor superfamily that interacts with a subset of retinoic acid response elements. Mol Cell Biol 14:1544–1551. Repa JJ, Mangelsdorf DJ 1999 Nuclear receptor regulation of cholesterol and bile acid metabolism. Curr Opin Biotechnol 10:557–563. Makishima M, Okamoto AY, Repa JJ, Tu H, Learned RM, Luk A, Hull MV, Lustig KD, Mangelsdorf DJ, Shan B 1999 Identification of a nuclear receptor for bile acids. Science 284:1362–1365. Edwards PA, Kast HR, Anisfeld AM 2002 BAREing it all: the adoption of LXR and FXR and their roles in lipid homeostasis. J Lipid Res 43:2–12. Lu TT, Repa JJ, Mangelsdorf DJ 2001 Orphan nuclear receptors as eLiXiRs and FiXeRs of sterol metabolism. J Biol Chem 276:37735–37738. Whitfield GK, Jurutka PW, Haussler CA, Haussler MR 1999 Steroid hormone receptors: evolution, ligands and molecular basis of biologic function. J Cell Biochem Suppls 32/33: 110–122.

CHAPTER 13 Nuclear Vitamin D Receptor

121. Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179. 122. Watkins RE, Wisely GB, Moore LB, Collins JL, Lambert MH, Williams SP, Willson TM, Kliewer SA, Redinbo MR 2001 The human nuclear xenobiotic receptor PXR: structural determinants of directed promiscuity. Science 292:2329–2333. 123. Shaffer PL, Gewirth DT 2002 Structural basis of VDR-DNA interactions on direct repeat response elements. EMBO J 21:2242–2252. 124. Martz E 2002 Protein explorer: easy yet powerful macromolecular visualization. Trends Biochem Sci 27:107–109. 125. Hsieh J-C, Whitfield GK, Jurutka PW, Haussler CA, Thatcher ML, Thompson PD, Dang HTL, Galligan MA, Oza AK, Haussler MR 2003 Two basic amino acids C-terminal of the P-box specify functional binding of the vitamin D receptor to its rat osteocalcin DNA responsive element. Endocrinology 144:5065–5080. 126. Mader S, Kumar V, de Verneuil H, Chambon P 1989 Three amino acids of the oestrogen receptor are essential to its ability to distinguish an oestrogen from a glucocorticoidresponsive element. Nature 338:271–274. 127. Rastinejad F, Perlmann T, Evans RM, Sigler PB 1995 Structural determinants of nuclear receptor assembly on DNA direct repeats. Nature 375:203–211. 128. Li M, Indra AK, Warot X, Brocard J, Messaddeq N, Kato S, Metzger D, Chambon P 2000 Skin abnormalities generated by temporally controlled RXRalpha mutations in mouse epidermis. Nature 407:633–636. 129. Lemon BD, Fondell JD, Freedman LP 1997 Retinoid X receptor:vitamin D3 receptor heterodimers promote stable preinitiation complex formation and direct 1,25-dihydroxyvitamin D3-dependent cell-free transcription. Mol Cell Biol 17:1923–1937. 130. Thompson PD, Remus LS, Hsieh J-C, Jurutka PW, Whitfield GK, Galligan MA, Encinas Dominguez C, Haussler CA, Haussler MR 2001 Distinct retinoid X receptor activation function-2 residues mediate transactivation in homodimeric and vitamin D receptor heterodimeric contexts. J Mol Endocrinol 27:211–227. 131. Pathrose P, Barmina O, Chang CY, McDonnell DP, Shevde NK, Pike JW 2002 Inhibition of 1,25-dihydroxyvitamin D3-dependent transcription by synthetic LXXLL peptide antagonists that target the activation domains of the vitamin D and retinoid X receptors. J Bone Miner Res 17:2196–2205. 132. Hsieh J-C, Whitfield GK, Oza AK, Dang HTL, Price JN, Galligan MA, Jurutka PW, Thompson PD, Haussler CA, Haussler MR 1999 Characterization of unique DNA binding and transcriptional activation functions in the carboxylterminal extension of the zinc finger region in the human vitamin D receptor. Biochemistry 38:16347–16358. 133. Cyert MS 2001 Regulation of nuclear localization during signaling. J Biol Chem 276:20805–20808. 134. Hsieh J-C, Shimizu Y, Minoshima S, Shimizu N, Haussler CA, Jurutka PW, Haussler MR 1998 Novel nuclear localization signal between the two DNA-binding zinc fingers in the human vitamin D receptor. J Cell Biochem 70:94–109. 135. Luo Z, Rouvinen J, Mäenpää PH 1994 A peptide C-terminal to the second Zn finger of human vitamin D receptor is able to specify nuclear localization. Eur J Biochem 223: 381–387. 136. LaCasse EC, Lefebvre YA 1995 Nuclear localization signals overlap DNA- or RNA-binding domains in nucleic acidbinding proteins. Nucleic Acids Res 23:1647–1656.

255 137. Stewart M, Baker RP, Bayliss R, Clayton L, Grant RP, Littlewood T, Matsuura Y 2001 Molecular mechanism of translocation through nuclear pore complexes during nuclear protein import. FEBS Lett 498:145–149. 138. Prufer K, Racz A, Lin GC, Barsony J 2000 Dimerization with retinoid X receptors promotes nuclear localization and subnuclear targeting of vitamin D receptors. J Biol Chem 275:41114–41123. 139. Hsieh J-C, Jurutka PW, Nakajima S, Galligan MA, Haussler CA, Shimizu Y, Shimizu N, Whitfield GK, Haussler MR 1993 Phosphorylation of the human vitamin D receptor by protein kinase C: Biochemical and functional evaluation of the serine 51 recognition site. J Biol Chem 268:15118–15126. 140. Yang W, Hyllner SJ, Christakos S 2001 Interrelationship between signal transduction pathways and 1,25(OH)2D3 in UMR106 osteoblastic cells. Am J Physiol Endocrinol Metab 281:E162–170. 141. Moras D, Gronemeyer H 1998 The nuclear receptor ligandbinding domain: structure and function. Curr Opin Cell Biol 10:384–391. 142. Weatherman RV, Fletterick RJ, Scanlan TS 1999 Nuclearreceptor ligands and ligand-binding domains. Ann Rev Biochem 68:559–581. 143. Bourguet W, Vivat V, Wurtz JM, Chambon P, Gronemeyer H, Moras D 2000 Crystal structure of a heterodimeric complex of RAR and RXR ligand-binding domains. Mol Cell 5:289–298. 144. Bourguet W, Ruff M, Chambon P, Gronemeyer H, Moras D 1995 Crystal structure of the ligand-binding domain of the human nuclear receptor RXR-α. Nature 375:377–382. 145. Gampe RT, Montana VG, Lambert MH, Miller AB, Bledsoe RK, Milburn MV, Kliewer SA, Willson TM, Xu HE 2000 Asymmetry in the PPARgamma/RXRalpha crystal structure reveals the molecular basis of heterodimerization among nuclear receptors. Mol Cell 5:545–555. 146. Tanenbaum DM, Wang Y, Williams SP, Sigler PB 1998 Crystallographic comparison of the estrogen and progesterone receptor’s ligand binding domains. Proc Natl Acad Sci USA 95:5998–6003. 147. Whitfield GK, Selznick SH, Haussler CA, Hsieh J-C, Galligan MA, Jurutka PW, Thompson PD, Lee SM, Zerwekh JE, Haussler MR 1996 Vitamin D receptors from patients with resistance to 1,25-dihydroxyvitamin D3: point mutations confer reduced transactivation in response to ligand and impaired interaction with the retinoid X receptor heterodimeric partner. Mol Endocrinol 10:1617–1631. 148. Feng W, Ribeiro RC, Wagner RL, Nguyen H, Apriletti JW, Fletterick RJ, Baxter JD, Kushner PJ, West BL 1998 Hormonedependent coactivator binding to a hydrophobic cleft on nuclear receptors. Science 280:1747–1749. 149. Darimont BD, Wagner RL, Apriletti JW, Stallcup MR, Kushner PJ, Baxter JD, Fletterick RJ, Yamamoto KR 1998 Structure and specificity of nuclear receptor-coactivator interactions. Genes Dev 12:3343–3356. 150. Nolte RT, Wisely GB, Westin S, Cobb JE, Lambert MH, Kurokawa R, Rosenfeld MG, Willson TM, Glass CK, Milburn MV 1998 Ligand binding and co-activator assembly of the peroxisome proliferator-activated receptor-γ. Nature 395:137–143. 151. Stehlin C, Wurtz JM, Steinmetz A, Greiner E, Schule R, Moras D, Renaud JP 2001 X-ray structure of the orphan nuclear receptor RORbeta ligand-binding domain in the active conformation. EMBO J 20:5822–5831. 152. Mi LZ, Devarakonda S, Harp JM, Han Q, Pellicciari R, Willson TM, Khorasanizadeh S, Rastinejad F 2003 Structural

256

153.

154.

155.

156.

157.

158.

159.

160.

161.

162.

163.

164. 165.

G. KERR WHITFIELD

basis for bile acid binding and activation of the nuclear receptor FXR. Mol Cell 11:1093–1100. Malloy PJ, Xu R, Peng L, Clark PA, Feldman D 2002 A novel mutation in helix 12 of the vitamin D receptor impairs coactivator interaction and causes hereditary 1,25-dihydroxyvitamin D-resistant rickets without alopecia. Mol Endocrinol 16:2538–2546. Jurutka PW, Hsieh J-C, Remus LS, Whitfield GK, Thompson PD, Haussler CA, Blanco JCG, Ozato K, Haussler MR 1997 Mutations in the 1,25-dihydroxyvitamin D3 receptor identifying C-terminal amino acids required for transcriptional activation that are functionally dissociated from hormone binding, heterodimeric DNA binding and interaction with basal transcription factor IIB, in vitro. J Biol Chem 272: 14592–14599. Panda DK, Miao D, Tremblay ML, Sirois J, Farookhi R, Hendy GN, Goltzman D 2001 Targeted ablation of the 25-hydroxyvitamin D 1alpha-hydroxylase enzyme: evidence for skeletal, reproductive, and immune dysfunction. Proc Natl Acad Sci USA 98:7498–7503. Dardenne O, Prud’homme J, Arabian A, Glorieux FH, St-Arnaud R 2001 Targeted inactivation of the 25-hydroxyvitamin D3-1α-hydroxylase gene (CYP27B1) creates an animal model of pseudovitamin D-deficiency rickets. Endocrinology 142:3135–3141. Hsieh JC, Sisk JM, Jurutka PW, Haussler CA, Slater SA, Haussler MR, Thompson CC 2003 Physical and functional interaction between the vitamin D receptor and Hairless corepressor, two proteins required for hair cycling. J Biol Chem 278:38665–38674. Watkins RE, Maglich JM, Moore LB, Wisely GB, Noble SM, Davis-Searles PR, Lambert MH, Kliewer SA, Redinbo MR 2003 2.1 A crystal structure of human PXR in complex with the St. John’s wort compound hyperforin. Biochemistry 42:1430–1438. Hoerer S, Schmid A, Heckel A, Budzinski RM, Nar H 2003 Crystal structure of the human liver X receptor beta ligandbinding domain in complex with a synthetic agonist. J Mol Biol 334:853–861. Rochel N, Tocchini-Valentini G, Egea PF, Juntunen K, Garnier JM, Vihko P, Moras D 2001 Functional and structural characterization of the insertion region in the ligand binding domain of the vitamin D nuclear receptor. Eur J Biochem 268:971–979. Malloy PJ, Hughes MR, Pike JW, Feldman D 1991 Vitamin D receptor mutants and hereditary 1,25-dihydroxyvitamin D resistant rickets. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D: Gene Regulation, Structure-Function Analysis and Clinical Application. Walter de Gruyter, New York, pp. 116–124. Bettoun DJ, Burris TP, Houck KA, Buck DW, 2nd, Stayrook KR, Khalifa B, Lu J, Chin WW, Nagpal S 2003 Retinoid X receptor is a non-silent major contributor to vitamin D receptor-mediated transcriptional activation. Mol Endocrinol 17:2320–2328. Warnmark A, Treuter E, Wright A, Gustafsson JA 2003 Activation functions 1 and 2 of nuclear receptors: molecular strategies for transcriptional activation. Mol Endocrinol 17:1901–1909. Thompson EB, Kumar R 2003 DNA binding of nuclear hormone receptors influences their structure and function. Biochem Biophys Res Commun 306:1–4. McKenna NJ, O’Malley BW 2002 Combinatorial control of gene expression by nuclear receptors and coregulators. Cell 108:465–474.

ET AL .

166. Masuyama H, Brownfield CM, St-Arnaud R, MacDonald PN 1997 Evidence for ligand-dependent intramolecular folding of the AF-2 domain in vitamin D receptor-activated transcription and coactivator interaction. Mol Endocrinol 11: 1507–1517. 167. Rachez C, Gamble M, Chang CP, Atkins GB, Lazar MA, Freedman LP 2000 The DRIP complex and SRC-1/p160 coactivators share similar nuclear receptor binding determinants but constitute functionally distinct complexes. Mol Cell Biol 20:2718–2726. 168. Jurutka PW, Hsieh J-C, MacDonald PN, Terpening CM, Haussler CA, Haussler MR, Whitfield GK 1993 Phosphorylation of serine 208 in the human vitamin D receptor: the predominant amino acid phosphorylated by casein kinase II, in vitro, and identification as a significant phosphorylation site in intact cells. J Biol Chem 268:6791–6799. 169. Hilliard GM, Cook RG, Weigel NL, Pike JW 1994 1,25Dihydroxyvitamin D3 modulates phosphorylation of serine 205 in the human vitamin D receptor: Site-directed mutagenesis of this residue promotes alternative phosphorylation. Biochemistry 33:4300–4311. 170. Jurutka PW, Hsieh J-C, Nakajima S, Haussler CA, Whitfield GK, Haussler MR 1996 Human vitamin D receptor phosphorylation by casein kinase II at ser-208 potentiates transcriptional activation. Proc Natl Acad Sci USA 93: 3519–3524. 171. Barletta F, Freedman LP, Christakos S 2002 Enhancement of VDR-mediated transcription by phosphorylation: correlation with increased interaction between the VDR and DRIP205, a subunit of the VDR-interacting protein coactivator complex. Mol Endocrinol 16:301–314. 172. Williams SP, Sigler PB 1998 Atomic structure of progesterone complexed with its receptor. Nature 393:392–396. 173. Kraichely DM, Collins JJ, 3rd, DeLisle RK, MacDonald PN 1999 The autonomous transactivation domain in helix H3 of the vitamin D receptor is required for transactivation and coactivator interaction. J Biol Chem 274:14352–14358. 174. Heery DM, Kalkhoven E, Hoare S, Parker MG 1997 A signature motif in transcriptional co-activators mediates binding to nuclear receptors. Nature 387:733–736. 175. Ding XF, Anderson CM, Ma H, Hong H, Uht RM, Kushner PJ, Stallcup MR 1998 Nuclear receptor-binding sites of coactivators glucocorticoid receptor interacting protein 1 (GRIP1) and steroid receptor coactivator 1 (SRC-1): multiple motifs with different binding specificities. Mol Endocrinol 12:302–313. 176. Jurutka PW, Remus LS, Whitfield GK, Galligan MA, Haussler CA, Haussler MR 2000 Biochemical evidence for a 170-kilodalton, AF-2-dependent vitamin D receptor/retinoid X receptor coactivator that is highly expressed in osteoblasts. Biochem Biophys Res Commun 267:813–819. 177. Kephart DD, Walfish PG, DeLuca H, Butt TR 1996 Retinoid X receptor isotype identity directs human vitamin D receptor heterodimer transactivation from the 24-hydroxylase vitamin D response elements in yeast. Mol Endocrinol 10:408–419. 178. Li XY, Xiao JH, Feng X, Qin L, Voorhees JJ 1997 Retinoid X receptor-specific ligands synergistically upregulate 1,25dihydroxyvitamin D3-dependent transcription in epidermal keratinocytes in vitro and in vivo. J Invest Dermatol 108: 506–512. 179. Kato S, Sasaki H, Suzawa M, Masushige S, Tora L, Chambon P, Gronemeyer H 1995 Widely spaced, directly repeated PuGGTCA elements act as promiscuous enhancers for different classes of nuclear receptors. Mol Cell Biol 15:5858–5867.

257

CHAPTER 13 Nuclear Vitamin D Receptor

180. Sutton AL, MacDonald PN 2003 Vitamin D: More than a “bone-a-fide” hormone. Mol Endocrinol 17:777–791. 181. Sharma D, Fondell JD 2000 Temporal formation of distinct thyroid hormone receptor coactivator complexes in HeLa cells. Mol Endocrinol 14:2001–2009. 182. Shang Y, Hu X, DiRenzo J, Lazar MA, Brown M 2000 Cofactor dynamics and sufficiency in estrogen receptorregulated transcription. Cell 103:843–852. 183. Masuyama H, MacDonald PN 1998 Proteasome-mediated degradation of the vitamin D receptor (VDR) and a putative role for SUG1 interaction with the AF-2 domain of VDR. J Cell Biochem 71:429–440. 184. Rosenfeld MG, Glass CK 2001 Coregulator codes of transcriptional regulation by nuclear receptors. J Biol Chem 276:36865–36868. 185. Huang ZQ, Li J, Sachs LM, Cole PA, Wong J 2003 A role for cofactor-cofactor and cofactor-histone interactions in targeting p300, SWI/SNF and mediator for transcription. EMBO J 22:2146–2155. 186. Encinas Dominguez C 2002 Insights into the biochemical life cycle of the vitamin D receptor: protein and DNA interactions that transduce the signal for gene expression. Biochemistry and Molecular Biophysics. University of Arizona, Tucson, pp. 175–176 [Ph.D. thesis]. 187. Dwivedi PP, Muscat GE, Bailey PJ, Omdahl JL, May BK 1998 Repression of basal transcription by vitamin D receptor: evidence for interaction of unliganded vitamin D receptor with two receptor interaction domains in RIP13delta1. J Mol Endocrinol 20:327–335. 188. Lusser A, Kadonaga JT 2003 Chromatin remodeling by ATPdependent molecular machines. Bioessays 25:1192–1200. 189. Potter GB, Zarach JM, Sisk JM, Thompson CC 2002 The thyroid hormone-regulated corepressor hairless associates with histone deacetylases in neonatal rat brain. Mol Endocrinol 16:2547–2560. 190. Blanco JCG, Wang I-M, Tsai SY, Tsai M-J, O’Malley BW, Jurutka PW, Haussler MR, Ozato K 1995 Transcription factor TFIIB and the vitamin D receptor cooperatively activate ligand-dependent transcription. Proc Natl Acad Sci USA 92:1535–1539. 191. Masuyama H, Jefcoat SC, Jr., MacDonald PN 1997 The N-terminal domain of transcription factor IIB is required for direct interaction with the vitamin D receptor and participates in vitamin D-mediated transcription. Mol Endocrinol 11:218–228. 192. Jurutka PW, Remus LS, Whitfield GK, Thompson PD, Hsieh J-C, Zitzer H, Tavakkoli P, Galligan MA, Dang HT, Haussler CA, Haussler MR 2000 The polymorphic N terminus in human vitamin D receptor isoforms influences transcriptional activity by modulating interaction with transcription factor IIB. Mol Endocrinol 14:401–420. 193. Barry JB, Leong GM, Church WB, Issa LL, Eisman JA, Gardiner EM 2003 Interactions of SKIP/NCoA-62, TFIIB, and retinoid X receptor with vitamin D receptor helix H10 residues. J Biol Chem 278:8224–8228. 194. Li X, Kimbrel EA, Kenan DJ, McDonnell DP 2002 Direct interactions between corepressors and coactivators permit the integration of nuclear receptor-mediated repression and activation. Mol Endocrinol 16:1482–1491. 195. Baudino TA, Kraichely DM, Jefcoat SC, Jr., Winchester SK, Partridge NC, MacDonald PN 1998 Isolation and characterization of a novel coactivator protein, NCoA-62, involved in vitamin D-mediated transcription. J Biol Chem 273:16434–16441. 196. Zhang C, Baudino TA, Dowd DR, Tokumaru H, Wang W, MacDonald PN 2001 Ternary complexes and cooperative

197.

198. 199. 200.

201.

202.

203. 204. 205. 206. 207.

208.

209. 210.

211.

212.

213.

interplay between NCoA-62/Ski-interacting protein and steroid receptor coactivators in vitamin D receptor-mediated transcription. J Biol Chem 276:40614–40620. Lee YH, Koh SS, Zhang X, Cheng X, Stallcup MR 2002 Synergy among nuclear receptor coactivators: selective requirement for protein methyltransferase and acetyltransferase activities. Mol Cell Biol 22:3621–3632. Lemon B, Inouye C, King DS, Tjian R 2001 Selectivity of chromatin-remodelling cofactors for ligand-activated transcription. Nature 414:924–928. Becker PB, Horz W 2002 ATP-dependent nucleosome remodeling. Ann Rev Biochem 71:247–273. Dilworth FJ, Fromental-Ramain C, Yamamoto K, Chambon P 2000 ATP-driven chromatin remodeling activity and histone acetyltransferases act sequentially during transactivation by RAR/RXR in vitro. Mol Cell 6:1049–1058. Kitagawa H, Fujiki R, Yoshimura K, Mezaki Y, Uematsu Y, Matsui D, Ogawa S, Unno K, Okubo M, Tokita A, Nakagawa T, Ito T, Ishimi Y, Nagasawa H, Matsumoto T, Yanagisawa J, Kato S 2003 The chromatin-remodeling complex WINAC targets a nuclear receptor to promoters and is impaired in Williams syndrome. Cell 113:905–917. Breen EC, van Wijnen AJ, Lian JB, Stein GS, Stein JL 1994 In vivo occupancy of the vitamin D responsive element in the osteocalcin gene supports vitamin D-dependent transcriptional upregulation in intact cells. Proc Natl Acad Sci USA 91:12902–12906. Fry CJ, Peterson CL 2002 Transcription. Unlocking the gates to gene expression. Science 295:1847–1848. Freeman BC, Yamamoto KR 2002 Disassembly of transcriptional regulatory complexes by molecular chaperones. Science 296:2232–2235. Vermeulen W, Houtsmuller AB 2002 The transcription cycle in vivo. A blind watchmaker at work. Mol Cell 10:1264–1266. McNally JG, Muller WG, Walker D, Wolford R, Hager GL 2000 The glucocorticoid receptor: rapid exchange with regulatory sites in living cells. Science 287:1262–1265. Becker M, Baumann C, John S, Walker DA, Vigneron M, McNally JG, Hager GL 2002 Dynamic behavior of transcription factors on a natural promoter in living cells. EMBO Rep 3:1188–1194. Demarest SJ, Martinez-Yamout M, Chung J, Chen H, Xu W, Dyson HJ, Evans RM, Wright PE 2002 Mutual synergistic folding in recruitment of CBP/p300 by p160 nuclear receptor coactivators. Nature 415:549–553. Rachez C, Freedman LP 2001 Mediator complexes and transcription. Curr Opin Cell Biol 13:274–280. Zhang C, Dowd DR, Staal A, Gu C, Lian JB, van Wijnen AJ, Stein GS, MacDonald PN 2003 Nuclear coactivator-62 kDa/Ski-interacting protein is a nuclear matrix-associated coactivator that may couple vitamin D receptor-mediated transcription and RNA splicing. J Biol Chem 278:35325–35336. Kim S, Yamamoto H, Shevde NK, Pike JW 2003 1,25Dihydroxyvitamin D3 stimulates dynamic vitamin D3 receptor/ retinoid receptor DNA-binding and coactivator recruitment in intact osteoblasts. J Bone Miner Res 18(Suppl 2):S30 (Abstract 1110). Li XY, Boudjelal M, Xiao JH, Peng ZH, Asuru A, Kang S, Fisher GJ, Voorhees JJ 1999 1,25-Dihydroxyvitamin D3 increases nuclear vitamin D3 receptors by blocking ubiquitin/ proteasome-mediated degradation in human skin. Mol Endocrinol 13:1686–1694. Conaway RC, Brower CS, Conaway JW 2002 Emerging roles of ubiquitin in transcription regulation. Science 296: 1254–1258.

258 214. Wang X, Pongrac JL, DeFranco DB 2002 Glucocorticoid receptors in hippocampal neurons that do not engage proteasomes escape from hormone-dependent down-regulation but maintain transactivation activity. Mol Endocrinol 16: 1987–1998. 215. Gonzalez F, Delahodde A, Kodadek T, Johnston SA 2002 Recruitment of a 19S proteasome subcomplex to an activated promoter. Science 296:548–550. 216. Ohyama Y, Ozono K, Uchida M, Yoshimura M, Shinki T, Suda T, Yamamoto O 1996 Functional assessment of two vitamin D-responsive elements in the rat 25-hydroxyvitamin D3 24-hydroxylase gene. J Biol Chem 271:30381–30385. 217. Aslam F, McCabe L, Frenkel B, van Wijnen AJ, Stein GS, Lian JB, Stein JL 1999 AP-1 and vitamin D receptor (VDR) signaling pathways converge at the rat osteocalcin VDR element: requirement for the internal activating protein-1 site for vitamin D-mediated trans-activation. Endocrinology 140:63–70. 218. Sierra J, Villagra A, Paredes R, Cruzat F, Gutierrez S, Javed A, Arriagada G, Olate J, Imschenetzky M, Van Wijnen AJ, Lian JB, Stein GS, Stein JL, Montecino M 2003 Regulation of the bone-specific osteocalcin gene by p300 requires Runx2/ Cbfa1 and the vitamin D3 receptor but not p300 intrinsic histone acetyltransferase activity. Mol Cell Biol 23:3339–3351. 219. Dwivedi PP, Hii CS, Ferrante A, Tan J, Der CJ, Omdahl JL, Morris HA, May BK 2002 Role of MAP kinases in the 1,25dihydroxyvitamin D3-induced transactivation of the rat cytochrome P450C24 (CYP24) promoter. Specific functions for ERK1/ERK2 and ERK5. J Biol Chem 277:29643–29653. 220. Hsieh J-C, Dang HTL, Galligan MA, Whitfield GK, Jurutka PW, Thompson PD, Haussler CA, Haussler MR 2001 Phosphorylation of the human vitamin D receptor by protein kinase A downregulates 1,25(OH)2D3-dependent transactivation by reducing retinoid X receptor β heterodimerization. J Bone Miner Res 16(Suppl 1):S231 (Abstract F517). 221. Bastien J, Adam-Stitah S, Plassat JL, Chambon P, RochetteEgly C 2002 The phosphorylation site located in the A region of retinoic X receptor alpha is required for the antiproliferative effect of retinoic acid (RA) and the activation of RA target genes in F9 cells. J Biol Chem 277:28683–28689. 222. Adachi S, Okuno M, Matsushima-Nishiwaki R, Takano Y, Kojima S, Friedman SL, Moriwaki H, Okano Y 2002 Phosphorylation of retinoid X receptor suppresses its ubiquitination in human hepatocellular carcinoma. Hepatology 35:332–340. 223. Lee HY, Suh YA, Robinson MJ, Clifford JL, Hong WK, Woodgett JR, Cobb MH, Mangelsdorf DJ, Kurie JM 2000 Stress pathway activation induces phosphorylation of retinoid X receptor. J Biol Chem 275:32193–32199. 224. Mangelsdorf DJ, Borgmeyer U, Heyman RA, Zhou JY, Ong ES, Oro AE, Kakizuka A, Evans RM 1992 Characterization of three RXR genes that mediate the action of 9-cis retinoic acid. Genes Dev 6:329–344. 225. Ruau D, Duarte J, Ourjdal T, Perriere G, Laudet V, RobinsonRechavi M. 2004 Update of NUREBASE: nuclear hormone receptor functional genomics. Nucleic Acids Res. 32 Database issue:D165–167. 226. Thompson PD, Hsieh J-C, Whitfield GK, Haussler CA, Jurutka PW, Galligan MA, Tillman JB, Spindler SR, Haussler MR 1999 The vitamin D receptor displays DNA binding and transactivation as a heterodimer with the retinoid X receptor, but not with the thyroid hormone receptor. J Cell Biochem 75:462–480. 227. Reinhardt TA, Koszewski NJ, Omdahl J, Horst RL 1999 1,25-Dihydroxyvitamin D3 and 9-cis-retinoids are synergistic

G. KERR WHITFIELD

228. 229. 230.

231. 232. 233.

234. 235. 236.

237. 238.

239. 240.

241. 242.

243.

244.

ET AL .

regulators of 24-hydroxylase activity in the rat and 1, 25dihydroxyvitamin D3 alters retinoic acid metabolism in vivo. Arch Biochem Biophys 368:244–248. Masuda S, Jones G 2003 Vitamin D analogs-drug design based on proteins involved in vitamin D signal transduction. Curr Drug Targets Immune Endocr Metabol Disord 3:43–66. Issa LL, Leong GM, Sutherland RL, Eisman JA 2002 Vitamin D analogue-specific recruitment of vitamin D receptor coactivators. J Bone Miner Res 17:879–890. Liu YY, Nguyen C, Peleg S 2000 Regulation of ligandinduced heterodimerization and coactivator interaction by the activation function-2 domain of the vitamin D receptor. Mol Endocrinol 14:1776–1787. Scott LJ, Dunn CJ, Goa KL 2001 Calcipotriol ointment. A review of its use in the management of psoriasis. Am J Clin Dermatol 2:95–120. Adorini L 2002 Immunomodulatory effects of vitamin D receptor ligands in autoimmune diseases. Int Immunopharmacol 2:1017–1028. Vegesna V, O’Kelly J, Said J, Uskokovic M, Binderup L, Koeffler HP 2003 Ability of potent vitamin D3 analogs to inhibit growth of prostate cancer cells in vivo. Anticancer Res 23:283–289. Riggs BL, Hartmann LC 2003 Selective estrogen-receptor modulators—mechanisms of action and application to clinical practice. N Engl J Med 348:618–629. McDonnell DP, Norris JD 2002 Connections and regulation of the human estrogen receptor. Science 296:1642–1644. Gehin M, Mark M, Dennefeld C, Dierich A, Gronemeyer H, Chambon P 2002 The function of TIF2/GRIP1 in mouse reproduction is distinct from those of SRC-1 and p/CIP. Mol Cell Biol 22:5923–5937. Xu J, Li Q 2003 Review of the in vivo functions of the p160 steroid receptor coactivator family. Mol Endocrinol 17:1681–1692. Whitfield GK, Remus LS, Jurutka PW, Zitzer H, Oza AK, Dang HTL, Haussler CA, Galligan MA, Thatcher ML, Encinas Dominguez C, Haussler MR 2001 Functionally relevant polymorphisms in the human nuclear vitamin D receptor gene. Mol Cell Endocrinol 177:145–159. Freund JN, Domon-Dell C, Kedinger M, Duluc I 1998 The Cdx-1 and Cdx-2 homeobox genes in the intestine. Biochem Cell Biol 76:957–969. Yamamoto H, Miyamoto K, Li B, Taketani Y, Kitano M, Inoue Y, Morita K, Pike JW, Takeda E 1999 The caudal-related homeodomain protein Cdx-2 regulates vitamin D receptor gene expression in the small intestine. J Bone Miner Res 14:240–247. Suh E, Chen L, Taylor J, Traber PG 1994 A homeodomain protein related to caudal regulates intestine-specific gene transcription. Mol Cell Biol 14:7340–7351. van Wering HM, Moyer L, Grand RJ, Krasinski SD 2002 Novel interaction at the Cdx-2 binding sites of the lactasephlorizin hydrolase promoter. Biochem Biophys Res Commun 299:587–593. Arai H, Miyamoto KI, Yoshida M, Yamamoto H, Taketani Y, Morita K, Kubota M, Yoshida S, Ikeda M, Watabe F, Kanemasa Y, Takeda E 2001 The polymorphism in the caudalrelated homeodomain protein Cdx-2 binding element in the human vitamin D receptor gene. J Bone Miner Res 16: 1256–1264. Arai H, Miyamoto K-I, Taketani Y, Yamamoto H, Iemori Y, Morita K, Tonai T, Nishisho T, Mori S, Takeda E 1997 A vitamin D receptor gene polymorphism in the translation initiation codon: effect on protein activity and relation to

CHAPTER 13 Nuclear Vitamin D Receptor

245.

246.

247.

248.

249.

250.

251. 252. 253. 254. 255.

256. 257.

258.

259.

260.

bone mineral density in Japanese women. J Bone Miner Res 12:915–921. Ferrari S, Rizzoli R, Manen D, Slosman D, Bonjour JP 1998 Vitamin D receptor gene start codon polymorphisms (FokI) and bone mineral density: interaction with age, dietary calcium, and 3′-end region polymorphisms. J Bone Miner Res 13:925–930. Ames SK, Ellis KJ, Gunn SK, Copeland KC, Abrams SA 1999 Vitamin D receptor gene Fok1 polymorphism predicts calcium absorption and bone mineral density in children. J Bone Miner Res 14:740–746. Gennari L, Becherini L, Mansani R, Masi L, Falchetti A, Morelli A, Colli E, Gonnelli S, Cepollaro C, Brandi ML 1999 FokI polymorphism at translation initiation site of the vitamin D receptor gene predicts bone mineral density and vertebral fractures in postmenopausal Italian women. J Bone Miner Res 14:1379–1386. Gross C, Eccleshall TR, Malloy PJ, Villa ML, Marcus R, Feldman D 1996 The presence of a polymorphism at the translation initiation site of the vitamin D receptor gene is associated with low bone mineral density in postmenopausal Mexican-American women. J Bone Miner Res 11: 1850–1855. Uitterlinden AG, Fang Y, Bergink AP, van Meurs JB, van Leeuwen HP, Pols HA 2002 The role of vitamin D receptor gene polymorphisms in bone biology. Mol Cell Endocrinol 197:15–21. Ingles SA, Ross RK, Yu MC, Irvine RA, La Pera G, Haile RW, Coetzee GA 1997 Association of prostate cancer risk with genetic polymorphisms in vitamin D receptor and androgen receptor. J Natl Cancer Inst 89:166–170. Gallie DR 1998 A tale of two termini: a functional interaction between the termini of an mRNA is a prerequisite for efficient translation initiation. Gene 216:1–11. Mangus DA, Evans MC, Jacobson A 2003 Poly(A)-binding proteins: multifunctional scaffolds for the post-transcriptional control of gene expression. Genome Biol 4:223. Baer BW, Kornberg RD 1980 Repeating structure of cytoplasmic poly(A)-ribonucleoprotein. Proc Natl Acad Sci USA 77:1890–1892. Deo RC, Bonanno JB, Sonenberg N, Burley SK 1999 Recognition of polyadenylate RNA by the poly(A)-binding protein. Cell 98:835–845. Ingles SA, Haile RW, Henderson BE, Kolonel LN, Nakaichi G, Shi CY, Yu MC, Ross RK, Coetzee GA 1997 Strength of linkage disequilibrium between two vitamin D receptor markers in five ethnic groups: implications for association studies. Cancer Epidemiol Biomarkers Prev 6:93–98. Ralston SH 2002 Genetic control of susceptibility to osteoporosis. J Clin Endocrinol Metab 87:2460–2466. Gennari L, Becherini L, Falchetti A, Masi L, Massart F, Brandi ML 2002 Genetics of osteoporosis: role of steroid hormone receptor gene polymorphisms. J Steroid Biochem Mol Biol 81:1–24. Hustmyer FG, Peacock M, Hui S, Johnston CC, Christian J 1994 Bone mineral density in relation to polymorphism at the vitamin D receptor gene locus. J Clin Invest 94: 2130–2134. Houston LA, Grant SFA, Reid DM, Ralston SH 1996 Vitamin D receptor polymorphism, bone mineral density, and osteoporotic vertebral fracture: Studies in a UK population. Bone 18:249–252. Taylor JA, Hirvonen A, Watson M, Pittman G, Mohler JL, Bell DA 1996 Association of prostate cancer with vitamin D receptor gene polymorphism. Cancer Res 56:4108–4110.

259 261. Feldman D 1997 Androgen and vitamin D receptor gene polymorphisms: the long and short of prostate cancer risk. J Natl Cancer Inst 89:109–111. 262. Xu Y, Shibata A, McNeal JE, Stamey TA, Feldman D, Peehl DM 2003 Vitamin D receptor start codon polymorphism (FokI) and prostate cancer progression. Cancer Epidemiol Biomarkers Prev 12:23–27. 263. Ruggiero M, Pacini S, Aterini S, Fallai C, Ruggiero C, Pacini P 1998 Vitamin D receptor gene polymorphism is associated with metastatic breast cancer. Oncol Res 10:43–46. 264. Peehl DM, Skowronski RJ, Leung GK, Wong ST, Stamey TA, Feldman D 1994 Antiproliferative effects of 1,25-dihydroxyvitamin D3 on primary cultures of human prostatic cells. Cancer Res 54:805–810. 265. Blutt SE, Allegretto EA, Pike JW, Weigel NL 1997 1,25dihydroxyvitamin D3 and 9-cis-retinoic acid act synergistically to inhibit the growth of LNCaP prostate cells and cause accumulation of cells in G1. Endocrinology 138: 1491–1497. 266. Mangelsdorf DJ, Koeffler HP, Donaldson CA, Pike JW, Haussler MR 1984 1,25-dihydroxyvitamin D3-induced differentiation in a human promyelocytic leukemia cell line (HL-60): Receptor-mediated maturation to macrophage-like cells. J Cell Biol 98:391–398. 267. Freedman LP 1999 Transcriptional targets of the vitamin D3 receptor-mediating cell cycle arrest and differentiation. J Nutr 129(2S Suppl):581S-586S. 268. Yabushita H, Hirata M, Noguchi M, Nakanishi M 1996 Vitamin D receptor in endometrial carcinoma and the differentiation-inducing effect of 1,25-dihydroxyvitamin D3 on endometrial carcinoma cell lines. J Obstet Gynaecol Res 22:529–539. 269. Giuliano AR, Franceschi RT, Wood RJ 1991 Characterization of the vitamin D receptor from the Caco-2 human colon carcinoma cell line: Effect of cellular differentiation. Arch Biochem Biophys 285:261–269. 270. Palmer HG, Gonzalez-Sancho JM, Espada J, Berciano MT, Puig I, Baulida J, Quintanilla M, Cano A, de Herreros AG, Lafarga M, Munoz A 2001 Vitamin D3 promotes the differentiation of colon carcinoma cells by the induction of E-cadherin and the inhibition of beta-catenin signaling. J Cell Biol 154:369–387. 271. Hutchinson PE, Osborne JE, Lear JT, Smith AG, Bowers PW, Morris PN, Jones PW, York C, Strange RC, Fryer AA 2000 Vitamin D receptor polymorphisms are associated with altered prognosis in patients with malignant melanoma. Clin Cancer Res 6:498–504. 272. Ingles SA, Garcia DG, Wang W, Nieters A, Henderson BE, Kolonel LN, Haile RW, Coetzee GA 2000 Vitamin D receptor genotype and breast cancer in Latinas (United States). Cancer Causes Control 11:25–30. 273. Park BS, Park JS, Lee DY, Youn JI, Kim IG 1999 Vitamin D receptor polymorphism is associated with psoriasis. J Invest Dermatol 112:113–116. 274. Lemire JM, Archer DC, Beck L, Spiegelberg HL 1995 Immunosuppressive actions of 1,25-dihydroxyvitamin D3: preferential inhibition of Th1 functions. J Nutr 125(6 Suppl): 1704S-1708S. 275. Vogel G 2002 Asthma research. Missing gene takes mice’s breath away. Science 295:253. 276. Pages F, Berger A, Henglein B, Piqueras B, Danel C, Zinzindohoue F, Thiounn N, Cugnenc PH, Fridman WH 1999 Modulation of interleukin-18 expression in human colon carcinoma: consequences for tumor immune surveillance. Int J Cancer 84:326–330.

260 277. Miyamoto K-i, Kesterson RA, Yamamoto H, Taketani Y, Nishiwaki E, Tatsumi S, Inoue Y, Morita K, Takeda E, Pike JW 1997 Structural organization of the human vitamin D receptor chromosomal gene and its promoter. Mol Endocrinol 11:1165–1179. 278. Yamada Y, Ando F, Niino N, Shimokata H 2003 Association of polymorphisms of interleukin-6, osteocalcin, and vitamin D receptor genes, alone or in combination, with bone mineral density in community-dwelling Japanese women and men. J Clin Endocrinol Metab 88:3372–3378. 279. Zmuda JM, Cauley JA, Danielson ME, Ferrell RE 2003 Vitamin D receptor and aromatase gene interaction and bone mass in older African-American women. Metabolism 52:521–523. 280. Kibel AS, Suarez BK, Belani J, Oh J, Webster R, BrophyEbbers M, Guo C, Catalona WJ, Picus J, Goodfellow PJ 2003 CDKN1A and CDKN1B polymorphisms and risk of advanced prostate carcinoma. Cancer Res 63:2033–2036. 281. Tayeb MT, Clark C, Haites NE, Sharp L, Murray GI, McLeod HL 2003 CYP3A4 and VDR gene polymorphisms and the risk of prostate cancer in men with benign prostate hyperplasia. Br J Cancer 88:928–932. 282. Pani MA, Knapp M, Donner H, Braun J, Baur MP, Usadel KH, Badenhoop K 2000 Vitamin D receptor allele combinations influence genetic susceptibility to type 1 diabetes in Germans. Diabetes 49:504–507. 283. Loke H, Bethell D, Phuong CX, Day N, White N, Farrar J, Hill A 2002 Susceptibility to dengue hemorrhagic fever in Vietnam: evidence of an association with variation in the vitamin D receptor and Fc gamma receptor IIa genes. Am J Trop Med Hyg 67:102–106. 284. Erturk S, Kutlay S, Karabulut HG, Keven K, Nergizoglu G, Ates K, Bokesoy I, Duman N 2002 The impact of vitamin D receptor genotype on the management of anemia in hemodialysis patients. Am J Kidney Dis 40:816–823. 285. Inagaki K, Krall EA, Fleet JC, Garcia RI 2003 Vitamin D receptor alleles, periodontal disease progression, and tooth loss in the VA dental longitudinal study. J Periodontol 74:161–167. 286. Garland CF, Garland FC, Gorham ED 1999 Calcium and vitamin D. Their potential roles in colon and breast cancer prevention. Ann N Y Acad Sci 889:107–119. 287. Pence BC, Buddingh F 1988 Inhibition of dietary fat-promoted colon carcinogenesis in rats by supplemental calcium or vitamin D3. Carcinogenesis 9:187–190. 288. Kallay E, Pietschmann P, Toyokuni S, Bajna E, Hahn P, Mazzucco K, Bieglmayer C, Kato S, Cross HS 2001 Characterization of a vitamin D receptor knockout mouse as a model of colorectal hyperproliferation and DNA damage. Carcinogenesis 22:1429–1435. 289. Kawa S, Nikaido T, Aoki Y, Zhai Y, Kumagai T, Furihata K, Fujii S, Kiyosawa K 1997 Vitamin D analogues up-regulate p21 and p27 during growth inhibition of pancreatic cancer cell lines. Br J Cancer 76:884–889. 290. Chawla A, Repa JJ, Evans RM, Mangelsdorf DJ 2001 Nuclear receptors and lipid physiology: opening the X-files. Science 294:1866–1870. 291. Lu TT, Makishima M, Repa JJ, Schoonjans K, Kerr TA, Auwerx J, Mangelsdorf DJ 2000 Molecular basis for feedback regulation of bile acid synthesis by nuclear receptors. Mol Cell 6:507–515. 292. Hirano S, Nakama R, Tamaki M, Masuda N, Oda H 1981 Isolation and characterization of thirteen intestinal microorganisms capable of 7 alpha-dehydroxylating bile acids. Appl Environ Microbiol 41:737–745.

G. KERR WHITFIELD

ET AL .

293. Takamine F, Imamura T 1995 Isolation and characterization of bile acid 7-dehydroxylating bacteria from human feces. Microbiol Immunol 39:11–18. 294. Sakai Y, Kishimoto J, Demay MB 2001 Metabolic and cellular analysis of alopecia in vitamin D receptor knockout mice. J Clin Invest 107:961–966. 295. Fuchs E, Merrill BJ, Jamora C, DasGupta R 2001 At the roots of a never-ending cycle. Dev Cell 1:13–25. 296. Barrandon Y 2003 Developmental biology: A hairy situation. Nature 422:272–273. 297. Jamora C, DasGupta R, Kocieniewski P, Fuchs E 2003 Links between signal transduction, transcription and adhesion in epithelial bud development. Nature 422:317–322. 298. Kobielak K, Pasolli HA, Alonso L, Polak L, Fuchs E 2003 Defining BMP functions in the hair follicle by conditional ablation of BMP receptor IA. J Cell Biol 163:609–623. 299. Peters EM, Foitzik K, Paus R, Ray S, Holick MF 2001 A new strategy for modulating chemotherapy-induced alopecia, using PTH/PTHrP receptor agonist and antagonist. J Invest Dermatol 117:173–178. 300. Cho YM, Woodard GL, Dunbar M, Gocken T, Jimenez JA, Foley J 2003 Hair-cycle-dependent expression of parathyroid hormone-related protein and its type I receptor: evidence for regulation at the anagen to catagen transition. J Invest Dermatol 120:715–727. 301. Falzon M 1996 DNA sequences in the rat parathyroid hormone-related peptide gene responsible for 1,25-dihydroxyvitamin D3-mediated transcriptional repression. Mol Endocrinol 10:672–681. 302. Kremer R, Sebag M, Champigny C, Meerovitch K, Hendy GN, White J, Goltzman D 1996 Identification and characterization of 1,25-dihydroxyvitamin D3-responsive repressor sequences in the rat parathyroid hormone-related peptide gene. J Biol Chem 271:16310–16316. 303. Hirst MA, Hochman HI, Feldman D 1985 Vitamin D resistance and alopecia: a kindred with normal 1,25-dihydroxyvitamin D binding, but decreased receptor affinity for deoxyribonucleic acid. J Clin Endocrinol Metab 60:490–495. 304. Hughes MR, Malloy PJ, Kieback DG, Kesterson RA, Pike JW, Feldman D, O’Malley BW 1988 Point mutations in the human vitamin D receptor gene associated with hypocalcemic rickets. Science 242:1702–1705. 305. Kristjansson K, Rut AR, Hewison M, O’Riordan JLH, Hughes MR 1993 Two mutations in the hormone binding domain of the vitamin D receptor cause tissue resistance to 1,25-dihydroxyvitamin D3. J Clin Invest 92:12–16. 306. Botchkarev VA, Botchkareva NV, Nakamura M, Huber O, Funa K, Lauster R, Paus R, Gilchrest BA 2001 Noggin is required for induction of the hair follicle growth phase in postnatal skin. FASEB J 15:2205–2214. 307. Hörlein AJ, Näär AM, Heinzel T, Torchia J, Gloss B, Kurokawa R, Ryan A, Kamei Y, Soderstrom M, Glass CK, Rosenfeld MG 1995 Ligand-independent repression by the thyroid hormone receptor mediated by a nuclear receptor co-repressor. Nature 377:397–404. 308. Tagami T, Lutz WH, Kumar R, Jameson JL 1998 The interaction of the vitamin D receptor with nuclear receptor corepressors and coactivators. Biochem Biophys Res Commun 253:358–363. 309. Wong CW, Privalsky ML 1998 Transcriptional silencing is defined by isoform- and heterodimer-specific interactions between nuclear hormone receptors and corepressors. Mol Cell Biol 18:5724–5733. 310. Potter GB, Beaudoin GM, 3rd, DeRenzo CL, Zarach JM, Chen SH, Thompson CC 2001 The hairless gene mutated in

261

CHAPTER 13 Nuclear Vitamin D Receptor

311.

312. 313.

314.

315.

316.

317.

318.

319.

congenital hair loss disorders encodes a novel nuclear receptor corepressor. Genes Dev 15:2687–2701. Moraitis AN, Giguere V, Thompson CC 2002 Novel mechanism of nuclear receptor corepressor interaction dictated by activation function 2 helix determinants. Mol Cell Biol 22:6831–6841. Stoye JP, Fenner S, Greenoak GE, Moran C, Coffin JM 1988 Role of endogenous retroviruses as mutagens: the hairless mutation of mice. Cell 54:383–391. Ahmad W, Faiyaz ul Haque M, Brancolini V, Tsou HC, ul Haque S, Lam H, Aita VM, Owen J, deBlaquiere M, Frank J, Cserhalmi-Friedman PB, Leask A, McGrath JA, Peacocke M, Ahmad M, Ott J, Christiano AM 1998 Alopecia universalis associated with a mutation in the human hairless gene. Science 279:720–724. Cichon S, Anker M, Vogt IR, Rohleder H, Putzstuck M, Hillmer A, Farooq SA, Al-Dhafri KS, Ahmad M, Haque S, Rietschel M, Propping P, Kruse R, Nothen MM 1998 Cloning, genomic organization, alternative transcripts and mutational analysis of the gene responsible for autosomal recessive universal congenital alopecia. Hum Mol Genet 7:1671–1679. Miller J, Djabali K, Chen T, Liu Y, Ioffreda M, Lyle S, Christiano AM, Holick M, Cotsarelis G 2001 Atrichia caused by mutations in the vitamin D receptor gene is a phenocopy of generalized atrichia caused by mutations in the hairless gene. J Invest Dermatol 117:612–617. Barwick JL, Quattrochi LC, Mills AS, Potenza C, Tukey RH, Guzelian PS 1996 Trans-species gene transfer for analysis of glucocorticoid-inducible transcriptional activation of transiently expressed human CYP3A4 and rabbit CYP3A6 in primary cultures of adult rat and rabbit hepatocytes. Mol Pharmacol 50:10–16. Taketani Y, Segawa H, Chikamori M, Morita K, Tanaka K, Kido S, Yamamoto H, Iemori Y, Tatsumi S, Tsugawa N, Okano T, Kobayashi T, Miyamoto K, Takeda E 1998 Regulation of type II renal Na+-dependent inorganic phosphate transporters by 1,25-dihydroxyvitamin D3. Identification of a vitamin D-responsive element in the human NAPi-3 gene. J Biol Chem 273:14575–14581. Kerner SA, Scott RA, Pike JW 1989 Sequence elements in the human osteocalcin gene confer basal activation and inducible response to hormonal vitamin D3. Proc Natl Acad Sci USA 86:4455–4459. DeMay MB, Gerardi JM, DeLuca HF, Kronenberg HM 1990 DNA sequences in the rat osteocalcin gene that bind the 1,25dihydroxyvitamin D3 receptor and confer responsiveness to

320.

321.

322.

323.

324.

325.

326.

327.

328.

1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 87: 369–373. Cao X, Ross FP, Zhang L, MacDonald PN, Chappel J, Teitelbaum SL 1993 Cloning of the promoter for the avian integrin β3 subunit gene and its regulation by 1,25-dihydroxyvitamin D3. J Biol Chem 268:27371–27380. Liu SM, Koszewski N, Lupez M, Malluche HH, Olivera A, Russell J 1996 Characterization of a response element in the 5′-flanking region of the avian (chicken) PTH gene that mediates negative regulation of gene transcription by 1,25dihydroxyvitamin D3 and binds the vitamin D3 receptor. Mol Endocrinol 10:206–215. Lian JB, Shalhoub V, Aslam F, Frenkel B, Green J, Hamrah M, Stein GS, Stein JL 1997 Species-specific glucocorticoid and 1,25-dihydroxyvitamin D responsiveness in mouse MC3T3-E1 osteoblasts: Dexamethasone inhibits osteoblast differentiation and vitamin D down-regulates osteocalcin gene expression. Endocrinology 138:2117–2127. Kim RH, Li JJ, Ogata Y, Yamauchi M, Freedman LP, Sodek J 1996 Identification of a vitamin D3-response element that overlaps a unique inverted TATA box in the rat bone sialoprotein gene. Biochem J 318:219–226. Rowland-Goldsmith MA, Henry HL 1997 Structure and regulatory elements of the vitamin D down-regulated gene for cAMP dependent protein kinase inhibitor (PKI). In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D: Chemistry, Biology and Clinical Applications of the Steroid Hormone. Vol. 1. Riverside: University of California, Printing and Reprographics, Riverside, CA, pp. 300–301. Drissi H, Pouliot A, Koolloos C, Stein JL, Lian JB, Stein GS, van Wijnen AJ 2002 1,25-(OH)2-Vitamin D3 suppresses the bone-related Runx2/Cbfa1 gene promoter. Exp Cell Res 274:323–333. Thompson JD, Higgins DG, Gibson TJ 1994 CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22:4673–4680. Janowski BA, Grogan MJ, Jones SA, Wisely GB, Kliewer SA, Corey EJ, Mangelsdorf DJ 1999 Structural requirements of ligands for the oxysterol liver X receptors LXRalpha and LXRbeta. Proc Natl Acad Sci USA 96:266–271. Sunn KL, Cock TA, Crofts LA, Eisman JA, Gardiner EM 2001 Novel N-terminal variant of human VDR. Mol Endocrinol 15:1599–1609.

CHAPTER 14

Vitamin D Receptor Cofactors: Function, Regulation, and Selectivity LEONARD P. FREEDMAN AND ALFRED A. RESZKA Department of Molecular Endocrinology and Bone Biology, Merck Research Laboratories, West Point, Pennsylvania

I. Introduction II. The Vitamin D Receptor and the Basal Machinery of Transcription III. Coactivators IV. Integration of Signaling Pathways

V. Molecular Basis for Tissue-Selective Vitamin D Receptor Ligands VI. Conclusions References

I. INTRODUCTION

multistep process initiated at the promoter region of expressed genes. It is catalyzed by RNA polymerase II (RNA Pol II) and requires the assembly of general transcription factors (GTFs) including TFIIA, -B, -D, -E, -F, -H, at the promoter (illustrated in Fig. 1; for review see [1–4]). This process is regulated by a combination of transcription factors recruited to a given promoter by direct DNA binding to specific response elements. DNA-bound factors mediate protein–protein interactions with components of the transcription machinery, ultimately targeting the recruitment and/or control of RNA Pol II. Several direct contacts have been identified between VDR or other nuclear receptors and the basal transcription apparatus. TFIIB was reported to interact directly with VDR-LBD. Interestingly, this interaction does not include the transactivation motif (AF-2) of VDR [5,6]. Moreover, the ligand may have a distinct effect on this interaction, depending on the cell type [5,7,8]. Taken together these data may suggest the requirement of additional targets for VDR transcription activity. For example, another basal factor, TFIIA, can also directly bind to VDR. This effect is strongly stimulated by ligand and occurs in the context of VDR bound to a promoter DNA template [9]. VDR also binds several TBP-associated factors TAFs (TATA box-binding protein [TBP] associated factors) that comprise the basal factor TFIID. TAFII135 and TAFII55 bind to VDR, RAR, and TR and enhance their activity [10,11]. TAFII28 has a potentiating or repressive effect on VDR and ER activity, depending on the cell type (COS versus HeLa cells, respectively) [12]. Finally, VDR binds to a number of newly discovered factors that appear to bridge or recruit other activities important for the activation process, as described at length later.

Vitamin D3 receptor (VDR) regulates transcription in direct response to its hormonal ligand, 1α,25(OH)2D3. Ligand binding leads to the recruitment of coactivators, defined as proteins that potentiate the activity of specific transcription factors. Many of these cofactors, as part of large complexes, act to remodel chromatin through ATP-dependent subunits or intrinsic histone-modifying activities. In addition, other ligand-recruited complexes appear to act more directly on the transcriptional apparatus. This suggests that transcriptional regulation by VDR and other nuclear receptors may involve a process of both chromatin alteration and direct recruitment of key initiation components at regulated promoters. This chapter will review the major cofactor complexes found to have significant effects on VDR function. The vast number of these coactivators and coactivator complexes provide potential targets for the relatively specific effects that have been achieved with some clinically important ligand analogs of nuclear receptors (for example, raloxifene for the estrogen receptor). This has spurred a tremendous amount of interest and effort in the quest to generate highly selective nuclear receptor modulators that would confer as drugs desirable effects on target tissues in the absence of potentially deleterious side effects. We will review efforts to achieve this kind of profile with vitamin D analogs, with an emphasis on how such selective ligands might function at the level of cofactor regulation.

II. THE VITAMIN D RECEPTOR AND THE BASAL MACHINERY OF TRANSCRIPTION Transcriptional activation of genes regulated by nuclear receptors and other transcription factors is a VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

264 III. COACTIVATORS A. The SRC/p160 Family of Coactivators Over the past several years, a growing number of proteins have been identified as coactivators for various nuclear receptors (for review see [3,13,14]). Many of these putative coactivators have been cloned by yeast two-hybrid or GST pull-down assays by virtue of their interaction with members of the nuclear receptor family, and their ability to potentiate transcriptional activity. Among the many nuclear receptor coactivators characterized thus far, a homologous family of proteins has emerged. It has been alternatively named SRC, NCoA, or more generically, p160, based on one of its first identified members, the 160-kDa protein steroid receptor coactivator-1 (SRC-1) [15]. The SRC/p160 family comprises three types of factors, on the basis of their homologies, including SRC-1/NCoA-1, GRIP1/ TIF2/NCoA-2, and pCIP/RAC3/ACTR/AIB1/TRAM-1 (for review, see [16,17]). The original enzymatic activity found to be common to these factors is histone acetyltransferase (HAT) activity, which catalyzes the acetylation of lysine residues at the N-terminal tails of histones [18,19]. Acetylation is thought to destabilize the interactions between DNA and the histone cores that form its nucleosomal structure in the nucleus. The SRC/p160 family members would thereby act as coactivators by loosening the repressive effect of chromatin on gene expression. In the original publications describing these proteins, ligand-dependent binding to VDR was only documented with the coactivator ACTR, but subsequent analyses have shown that VDR is also the target of GRIP1/TIF-2 and SRC-1 [20,21] (also our observations). Specific inactivation of SRC-1 by gene targeting reveals that there may be at least partial functional redundancy between the different SRC/p160 family members, since only a partial resistance to hormonal response is observed in SRC-1(−/−) mice. This phenomenon is concomitant with increased mRNA levels of other coactivators such as TIF2, perhaps compensating for the loss of SRC-1 [22]. The coactivator effects of SRC-1 on nuclear receptors have also been demonstrated in vitro in the presence of chromatin assembled templates [23]. Under these conditions, SRC-1 strongly potentiates the ligand-dependent activity of the progesterone receptor (PR-B). Interestingly, this potentiation also occurs to a certain extent on naked DNA, in the absence of chromatin. These results may suggest a dual effect of SRC-1, both on chromatin remodeling and on other activities or interactions yet to be identified. Beyond this homologous family, several proteins that have been found to act as co-regulators of VDR

LEONARD P. FREEDMAN AND ALFRED A. RESZKA

cannot be classified in any of the previous categories. NCoA-62 /SKIP, identified by yeast two-hybrid, coactivates VDR in a ligand-dependent fashion [24]. ChIP analyses reveal that NCoA-62 associates with liganded VDR after the binding of SRC-1 [25]. Interestingly, NCoA-62 was also shown to interact with components of the spliceosome, and a dominant negative NCoA-62 inhibits splicing of 1α,25(OH)2VD3 -induced transcripts, suggesting that it somehow couples VDR-mediated transcription to the process of RNA splicing. TIF1 was found to interact with VDR and other nuclear receptors [26]. TIF1 also interacts with heterochromatin-associated proteins and has a kinase activity that can phosphorylate several general transcription factors (TFIIEa, TAFII28, and TAFII55), suggesting a novel mechanism of transcription regulation for nuclear receptors [27].

B. Large Coactivator Assemblies with Multifunctional HAT Activity The SRC/p160 family of coactivators not only binds to nuclear receptors, such as VDR, but also recruits and forms complexes with CBP/p300 [19]. The cointegrators CBP and p300 also bind to a large panel of transcription factors [28], and although they do bind to nuclear receptors, this interaction is much weaker than is seen with SRC/p160. Indeed, they appear to interact with nuclear receptors cooperatively along with SRC/ p160 and other components such as p/CIP and PCAF, together forming a larger coactivator complex [13,29]. Alternatively, CBP and SRC-1 may be stabilized by a specific RNA coactivator, SRA (steroid receptor RNA activator). Interestingly, SRA was found to be part of a 600–700 kDa ribonucleoprotein structure that includes SRC-1 [30]. CBP/p300 display HAT activity [31,32], which plays an integral role in chromatin remodeling. The fact that different components of the complex possess HAT activity suggests some sort of a cooperative effect or/and an increased array of specificities. Recent reports suggest additional functions for CBP’s HAT activity besides chromatin remodeling. CBP/p300 can acetylate nonhistone proteins, such as the transcription factor p53, thus enhancing its DNA binding activity [33]. Components of the basal machinery (TFIIEa, TFIIF) are also acetylated by p300, PCAF, and TAFII250, but the effects of this modification have not yet been elucidated [34]. Intriguingly, CBP/p300 can regulate the association between ACTR and the estrogen receptor by directly acetylating ACTR at two lysines near to one of its NR boxes, thereby disrupting its association with ER [35]. Currently, our view of how SRC/p160 functions is primarily to act as means of recruiting

265

CHAPTER 14 Vitamin D Receptor Cofactors: Function, Regulation, and Selectivity

CBP/p300 to a nuclear receptor. It is CBP/p300 rather than SRC/p160 that appears to be the primary source of HAT activity.

C. The Human Mediator Complex: DRIP 1. IDENTIFICATION OF THE COMPLEX

Besides the SRC/p160 coactivators, several laboratories, including our own, identified a novel type of nuclear receptor coactivator complex, alternatively called DRIP, TRAP, ARC, NAT, or mammalian Mediator (reviewed

in [36]) (Table I), with each name depending on the purification process and the activators tested as targets. These complexes are required for activation of transcription in vitro, in different transcription assays involving purified components of the transcription machinery. At the time of their respective discoveries, each of these complexes was thought to be specific for a distinct transcription factor. DRIP (vitamin D receptor interacting proteins) [37,38] and ARC (activator recruited cofactor) [39,40] complexes were originally purified from nuclear extracts by in vitro pull-down assays using GST fusions with VDR-LBD or using the activation motifs of several

TABLE I Subunit Composition of General Coactivator Complexes. DRIP

DRIP250 DRIP240 DRIP205 DRIP150 DRIP130 DRIP100 DRIP97 DRIP92 DRIP77 DRIP70–2

ARC CBP/p300 ARC250 ARC240 ARC205 ARC150 ARC130 ARC105/TIG-1 ARC100 ARC92 ARC77 ARC70 ARC42

SMCC/TRAP

TRAP240 TRAP230 TRAP220 TRAP170 TRAP150 TRAP100 TRAP97 TRAP95 TRAP93 TRAP80

CRSP

p230 CRSP200 CRSP150 CRSP130

ARC36 ARC34

DRIP33

ARC33 ARC32 p28

p150 p140/hSur2

p95 p90 CRSP77 CRSP70

Mediator (mouse)

Mediator (Yeast S.c.)

p160a p160b Rgr1/p110

Nut1 Gal11 Rgr1

Ring3/p96a p96b

Sin4 Srb4 Med1

p78 p70 p55

hSrb10 DRIP36 DRIP34

NAT

hMed7 hMed6 hTRF hSrb11 hSoh1

hSrb7 hNut2

CRSP34 CRSP33

p56/Cdk8 p45 p37 p36 p33 p31/Cycl.C p30 p23 p22 p21 p17 p14

p34 Med7/p36 Med6/p32 TRF/p28a p28b

Srb7/p21

Subunits with highlight are equivalent in the different complexes, or homologs between mammalian and yeast complexes. Subunits in the same lanes have similar molecular weights regardless of any homology.

Med2 Pgd1/Hrs1 Srb10 Med4 Med7 Srb5 Med6 Med8 Srb11 Rox3/Ssn7 Srb2 Med9/Cse2 Srb7 Med10/Nut2 Med11 Srb6

266 transcription factors (SREBP-1a, NF-kB, and VP16), respectively. The TRAP complex (TR associated proteins [41,42]) was isolated by coimmunoprecipitation of epitope-tagged TR stably expressed in HeLa cells. TRAP was subsequently found to be identical to SMCC (Srb/Med-containing cofactor complex [43], a complex purified by coimmunoprecipitation with antibodies directed against epitope-tagged Srb10. The cloning of these component subunits by independent groups revealed the near-identity of their sequences, suggesting that these different complexes might actually constitute a single, universal one. Importantly, some subunits do differ from complex to complex. For example, TRAP150 has no homology with any of its candidate counterparts. In addition, DRIP/ARC/CRSP130, also identified as hSur2 [44], has not been identified in the TRAP complex. The various subunit compositions are summarized in Table I. Other complexes exist that have close identities with the DRIP, ARC, and TRAP complexes. CRSP (cofactor required for Sp1 activation) [45], purified by multiple chromatographic steps, appears to be a subset of nine subunits of the DRIP/ARC complex and might represent a stable core of subunits or a conserved subcomplex among various functionally related complexes. The CRSP complex differs, however, by its two unrelated 34- and 70-kDa subunits (Table I). The homology of CRSP70’s N terminus with the elongation factor TFIIS is a unique feature among all the complexes described so far. Despite its apparently limited number of subunits, CRSP potentiates the activity of the transcription factor Sp1 in vitro. The NAT complex (negative regulator of activated transcription) was identified by coimmunoprecipitation out of HeLa nuclear extracts with an antibody against hSrb10/CDK8 [46]. NAT shares many common subunits with DRIP, ARC, and TRAP. However, when the NAT complex was tested in vitro in a purified transcription assay in the presence of RNA Pol II, general transcription factors (TFIIA, to -H), and a cofactor activity PC4 (see later discussion), it exhibited a repressing effect on transcription driven by various activators, without any influence on basal transcription. This unexpected result suggests that these complexes may not only have an activation potential, but also repressive activities on transcription in vitro, as discussed later. The mouse Mediator complex was identified [47] through biochemical purification out of nuclear extracts of murine cells. Its name reflects its homologies with a yeast Mediator counterpart (Table I; see later discussion). 2. FUNCTIONALITY OF THE DRIP COMPLEX

The functional role of DRIP and related complexes in human cells can be postulated on the basis of a series

LEONARD P. FREEDMAN AND ALFRED A. RESZKA

of classic studies in yeast (for review, see [48,49]). Genetic and biochemical analyses in yeast revealed the importance of several types of factors for transcription of target genes by RNA Pol II in response to activators [49]. They include the products of SRB genes, identified in a genetic screen as suppressors of the effect of truncations within the CTD of RNA Pol II [50]. Srb proteins point to the importance of CTD phosphorylation in the transition between transcription initiation at the promoter and elongation of the RNA [51] (transcript, and references therein). Biochemical fractionation resulted in the isolation of Mediator as a multisubunit complex [52] that contained Srb proteins, and another set of components named Med proteins [53,54]. Detailed functional studies in yeast revealed the importance of both Srb and Med components of the Mediator complex in either general transcription (Srb4 [55]), or activation of transcription by specific factors both in vivo and in vitro, such as Gcn4, Gal4, or VP16 (observed for Med2, Med6, Pgd1/Hrs1, and Sin4 [53,54]). This raised the definition of the yeast Mediator as a “global transcription coactivator” [56]. The existence of human homologs of a number of yeast Mediator proteins suggested that a corresponding complex would exist in higher organisms. Mammalian Mediator was first identified in mice by Kornberg’s group [47]. Its homology to the yeast Mediator is also suggested by some of its functional characteristics: the complex can bind the CTD of RNA Pol II, and it stimulates in vitro CTD phosphorylation, catalyzed by TFIIH. Some of its subunits have yeast counterparts (Rgr1, Med6, Med7, Srb7), but several other subunits lack any yeast homologs. Interestingly some, but not all, of the mouse Mediator subunits are also present within DRIP, ARC, TRAP/SMCC, and NAT (see Table I). That DRIP, ARC, and TRAP/SMCC complexes all contain Med components [40,43,57] may suggest that these complexes are likely to be human homologs of the Mammalian Mediator. However, a side-by-side comparison of the respective subunits within human and mouse complexes favors a scenario where “mammalian Mediator” encompasses subcomplexes purified as DRIP, ARC, and TRAP/SMCC (Table I). A larger diversity of complexes in higher eukaryotes relative to yeast was previously suggested for the RNA Pol II holoenzyme on the basis of its different subunit compositions in mammalian preparations (i.e., variations in a subset of components) [48]. The potential differences in composition of the DRIP, ARC, TRAP/SMCC, NAT, and mammalian Mediator complexes may reflect this diversity. Whatever the case, the presence of Srb/Med subunits in DRIP, ARC, and TRAP/SMCC strongly suggests that at least part of how this complex functions is through recruitment of RNA Pol II. Indeed, experiments demonstrate

CHAPTER 14 Vitamin D Receptor Cofactors: Function, Regulation, and Selectivity

that DRIP bound to liganded VDR indeed can recruit RNA Pol II and associated subunits [58]. As mentioned previously, activities of all these complexes were tested in highly purified transcription assays in vitro, in response to specific activators. The DRIP complex strongly potentiated ligand-dependent VDR-RXR transcription on DNA templates assembled into chromatin, but interestingly it had little or no effect on the same transcription in the absence of chromatin (naked templates). None of the previously identified SRC/p160 coactivators have been found to be part of the purified DRIP complex [38], and the absence of any HAT activity in the DRIP complex suggests that it may contain distinct chromatin remodeling activities, or may recruit them. The ARC complex tested on chromatin-assembled templates exhibits cooperativity between different activators, such as Sp1 and SREBP-1a, on the same template promoter, in conditions where ARC has no effect on the same activators tested individually [40]. This is an interesting demonstration of the ability of these complexes to integrate multiple transcription pathways into a synergistic effect on gene activation. TRAP/SMCC also enhances activatordependent transcription in a purified in vitro system with RNA Pol II, GTFs, and PC4, but only in the absence of or with limiting concentrations of TFIIH [43]. In a transcription system including both PC4 and TFIIH, the SMCC complex can repress transcription. This effect was specific for the presence of activators such as Gal4-AH, since no strong effect was observed on basal transcription. By analogy with the effect of Mediator in yeast on phosphorylation of RNA Pol II CTD, the kinase activity of SMCC was tested in vitro. Although the CTD was phosphorylated, the major substrate was the cofactor PC4, a single-stranded DNA binding protein that is required for activated transcription [59,60]. PC4 can interact in vitro with TFIIA and VP16 [60]. More importantly, PC4 binding and transcription activities are lost upon phosphorylation [59,60]. The conservation of PC4 in yeast (as Tsp1) highlights its general requirement among eukaryotic systems [61]. Interestingly, the repression effect by SMCC was also observed when a CTD-less form of RNA Pol II was used. Based on this, the authors suggested that regulation of transcription by SMCC could be mediated in concert with PC4, but independently of modifications of the RNA Pol II CTD [43].

D. ATP-Dependent Remodeling Cofactors In addition to HAT and HDAC-containing activities, complexes found biochemically to contain ATPdependent chromatin remodeling activities, which

267

could be expected to produce more significant chromatin rearrangements, have also been linked to steroid and nuclear receptors. Yeast genetics originally identified a group of gene products, SNF/Swi, which reduced expression of the SUC2 and HO genes involved in sucrose fermentation and mating-type switching, while among suppressors of these genes, SIN or switchindependent mutations, were identified to be histones and other chromatin-associated proteins. SNF2/Swi2 homologs were subsequently found in Drosophila (Brahma) and mammalian cells (hbrm and BRG-1) [62]. Purified multisubunit SNF/Swi complexes were later found to contain DNA-dependent ATPase activity intrinsic to SNF2/Swi2, which simultaneously disrupted nucleosomes and enhanced transcription factor binding [63,64]. Purification of hbrm and BRG-1 revealed they are present in distinct mammalian complexes associated with selective BRG-1 associated factors (BAFs) [65]. ATP-dependent remodeling by these complexes has been shown to be independent of histone tails and is both persistent upon removal of ATP and reversible upon readdition of SNF/Swi [66,67]. An early connection of SNF/Swi to nuclear receptormediated transcription was revealed by the loss of steroid-induced reporter gene activation by GR and ER in Swi-1, -2, and -3-mutant yeast strains [68]. Additionally, when overexpressed in cells, hbrm was found to cooperate with transfected GR in mammalian cell lines devoid of endogenous hbrm [69] and ER was observed to interact with both hbrm and BRG-1 in yeast in twohybrid experiments [70]. Furthermore, conditional BRG-1 knockout F9 cell lines were found to be inviable, while heterozygote lines could not proliferate yet remained sensitive to RA-induced differentiation [71]. One extensively examined model for ligand-dependent perturbations of chromatin structure in preinitiation complex (PIC) formation, as well as antagonism between the progesterone and glucocorticoid receptors (PR and GR), has been the integrated composite MMTV promoter/enhancer (reviewed in [72]). Transiently transfected MMTV DNA is activated by PR ligands; however, random integration of this same DNA into the mammalian genome with concomitant packaging into a phased array of chromatin eliminated PR-dependent gene activation by preclusion of binding site accessibility, but also induced progestin-specific repression of glucocorticoid-mediated transcription from MMTV, suggesting competition for common molecular targets. One report has described a model for this competition mediated through BRG-1 [73]. Interestingly, both progestins and anti-progestins inhibited GR stimulation of integrated MMTV DNA and decreased its association with limiting amounts of BRG-1, as assessed by coimmunoprecipitation. These same compounds were unable

268 to reduce transactivation of transiently transfected promoters, nor were they able to destabilize GR bound to the p160 coactivators NCoA-1/SRC-1 and NCoA2/GRIP-1, as well as with p300. Thus the observed squelching was a function of recruitment of remodeling complex(es) and not through association of other coactivator complexes, and may reflect an initial binding threshold for target site accessibility. In addition to SNF/Swi, other ATP-dependent remodeling complexes have been isolated, including the distinct yeast RSC complex, and a related family of complexes which share homologs of the SNF2/Swi2related ISWI ATPase, namely the Drosophila complexes NURF, ACF, and CHRAC and the human RSF complex (recently reviewed in [74]). The ISWI-containing complexes were initially purified biochemically based on their abilities to catalyze the assembly of nucleosomal DNA (ACF [75]), physiologically space salt or polyglutamate-deposited nucleosomes (ACF [75], CHRAC [76], RSF [77]), enhance transcription factor binding (NURF [78]), and increase transcription from nucleosomal templates in vitro (NURF [79], RSF [77]). Recently, another distinct ATP-dependent remodeling activity was found associated with an HDAC-containing complex, NURD, which also possesses a subunit with limited homology to N-CoR, MTA-1 [80,81]. Intriguingly, antibodies directed against the CHD4/Mi-2 ATPase subunit of NURD partially relieve ligand-independent TR/RXR repression of the TRβA promoter in Xenopus oocytes by microinjection experiments [80]. Recently, two new chromatin remodeling complexes have been characterized through profound functional effects on VDR. VDR has been found to associate with a novel complex called WINAC, which shares subunits of the SWI/SNF and ISWI complexes [82]. WINAC interacts directly with VDR, albeit in a ligandindependent manner, through one constituent subunit, the Williams syndrome transcription factor (WSTF). Interestingly, the WSTF gene is deleted in patients with Williams syndrome, a neurodevelopmental disorder that, in addition to causing cognitive disorders, also results in congenital heart disease. Kitagawa and coworkers found, using chromatin immunoprecipitation (ChIP) assays, that WINAC is recruited to both negative and positive VDREs by VDR, and that WSTF overexpression potentiated VDR-mediated transactivation or transrepression. Presumably, this occurs through the ability of the WINAC complex to enhance both the assembly and disassembly of nucleosomal arrays. As it contains BRG1, WINAC functions in an ATP-dependent fashion. A second chromatin-remodeling complex recently described also appears to be important for VDR transcriptional function, at least in a purified, reconstituted

LEONARD P. FREEDMAN AND ALFRED A. RESZKA

transcription system. Using conventional biochemical purification, Lemon and co-workers [83] purified an activity that is required for ligand-dependent transactivation by VDR using a chromatin-dependent reconstituted in vitro transcription assay. This activity was identified as a multisubunit complex called PBAF, which shares many, but not all, of the same subunits as SWI/SNF-A. Using antibody depletion, it was found that only PBAF, but not SWI/SNF, supported VDRdependent transactivation, suggesting that at least in this experimental system, which is dependent on the presence of the DRIP complex, these two chromatin remodeling complexes are not functionally equivalent, nor does one complement the activity of the other. The ever-expanding number of chromatin-remodeling, histone-modifying, and Pol II–recruiting complexes that interact with and are utilized by nuclear receptors in general and VDR in particular makes this a particularly challenging scientific question to study. It also suggests that this wide repertoire could be the basis for tissue selectivity by VDR analogs and other nuclear receptor modulators (see next section). However, it also suggests that the process of transcriptional regulation involves multiple steps mediated by multiple complexes conferring distinct functions, whereby VDR and other nuclear receptors serve as ligand-regulated platforms for these complexes to the DNA in the proximity of regulated gene promoters. Some chromatin immunoprecipitation experiments are consistent with this so-called ordered-recruitment model, but it may also turn out that specific promoters (or more accurately, specific VDREs) impose the recruitment of a selective repertoire of cofactor complexes through allosteric effects perpetuated through DNA-bound VDR. The delineation of tissue selectivity by VDR ligands, then, will be an exceedingly complicated paradigm to define, but it may be that this complexity is actually what nature had in mind to achieve maximal regulatory flexibility.

IV. INTEGRATION OF SIGNALING PATHWAYS The WINAC, SRC-1/p160, and DRIP complexes represent three unrelated protein complexes, together carrying several distinct activities that have been shown to potentiate transcription activity of VDR (and other nuclear receptors). These activities may function together to provide a synergistic effect of 1,25(OH)2D3mediated activation, or to provide specificity of targeting to its regulation. VDR transcription regulation may require a combination of chromatin remodeling activities, as well as efficient recruitment of the RNA Pol II

269

CHAPTER 14 Vitamin D Receptor Cofactors: Function, Regulation, and Selectivity

motif in the DRIP complex as a combination of two subunits, and might explain the presence of two unrelated activation functions in the same receptor. In the TRAP/SMCC complex, both p53 and VP16 interact via TRAP80, but binding of VP16 and TR (which binds TRAP220) to the TRAP complex is not mutually exclusive, suggesting that this complex may support activation by a combination of distinct transcription factors. The transactivation domain (conserved region 3, CR3) of E1A interacts directly with hSur-2, which corresponds to DRIP130 [44]. This defines yet another novel and distinct target motif for an activator within the complexes. Several other classes of transcription factors have been found to functionally interact with DRIP subunits, and presumably the entire complex. One noteworthy example is interferon-activated ISGF3 transcription factor. ISGF3 is a heterotrimeric complex comprising STAT1, STAT2, and a DNA-binding subunit, IRF9. Upon induction by interferons, resulting in tyrosine phosphorylation of STATs, the complex enters the nucleus, binds to interferon–stimulated response elements, and activates transcription of regulated promoters. Lau and co-workers [88] have reported that ISGF3stimulated transcription is dependent on direct STAT2 interactions with DRIP150. This interaction leads to the recruitment of other DRIP subunits as well as RNA Pol II to ISGF3 target promoters, as demonstrated by chromatin immunoprecipitation. These results

machinery, via several of its basal factors. We have suggested a stepwise model that combines these distinct activities (Fig. 1), where WINAC and/or CBP/p160 coactivator complexes might be required for chromatin remodeling, followed by the direct recruitment of the transcription machinery by the DRIP complex. Definitive experimental analysis has yet to confirm this view [14,84]. In vitro, however, the SRC/p160 and DRIP coactivators appear to have no real intrinsic difference in their ability to interact with nuclear receptors. They both utilize the AF-2 of nuclear receptors and bind with similar affinities (TRAP220 versus TIF2 for TR [85]; DRIP205 and GRIP1 for VDR [86]. Thus we could envision a cooperative model where both CBP/p160 and DRIP complexes simultaneously occupy a promoter, and through their combined actions facilitate activation of transcription (Fig. 1). The observation that CBP can acetylate ACTR, leading to the latter’s dissociation from liganded ER [35], suggests a mechanism for the sequential model, whereby the first complex functions to acetylate histones and disrupt chromatin structure, whereupon it itself dissociates from the receptor, allowing the DRIP complex to bind and act at the level of direct recruitment. Recent studies have revealed multiple binding motifs for coactivators within the DRIP and TRAP complexes. For example, GR interacts with the DRIP complex via both AF-1 and AF-2 motifs with DRIP150 and DRIP205, respectively [87]. These results define the GR binding

DRIP/mediator (direct recruitment) p53

HAT complexes (nuclesome modification) SWI/SNF complexes (chromatin remodeling) CBP P300

BAF60 BAF155

STAT GR AR

80 150

SRC-1 p160

E1A 130

230

Med6 205

100

NFκB SREBP Sp1

Med7

PCAF

BAF53 BAF180

VP16

BRG/hBRM

HAT

ATP

RXR-VDR CTD TFIIH TFIIE TFIIA

Nuclesomes

FIGURE 1

DR3

RNA pol II

TFIID TFIIB

VDR associations with coactivator complexes

TFIIF

270 implicate an entirely distinct signaling system as functioning through the DRIP complex, and suggest that diverse pathways, such as the vitamin D and interferon signaling systems, may intersect or compete for limiting cofactors (such as the DRIP complex) by sharing subunits required for their optimal transactivation function.

V. MOLECULAR BASIS FOR TISSUE-SELECTIVE VITAMIN D RECEPTOR LIGANDS A. Established Paradigms The molecular basis for tissue selectivity of various VDR ligands is an area of considerable clinical interest. Most intriguing is our expanding understanding that mechanisms conferring tissue selectivity for one type of nuclear receptor may not fully apply to another. The concept of a tissue-selective ligand for a nuclear receptor was first used to describe ligands for the estrogen receptor (ER) that maintained certain estrogenic effects in the absence of others. To a great extent, such ligands, SERMs such as tamoxifen and raloxifene, act as antagonists in certain tissues and cell types (e.g., breast) while conferring some degree of agonism in others (e.g., bone). For the most part, receptor agonism by these ligands is less than that seen with the natural ligand, estradiol. Ligands for the VDR differ in the respect that superagonists have been described in addition to antagonists. For the superagonists, no equivalent ER or PR paradigm exists to help explain the genesis of such activity. For the antagonists, we find that, as with SERMs, some residual agonism is left intact, suggesting VDR modulator-like activity. Interestingly, the molecular basis for partial VDR agonism seems to differ from that described for ERα. Earlier work with ERα sought to explain tissue selectivity through X-ray crystallography analysis. Crystallography comparing the structure of ERα when liganded with estradiol or raloxifene revealed that the position of helix 12 differed, depending on which ligand (agonist or antagonist) was bound into the LBD [89]. Similar differences are predicted to exist for the progesterone receptor when bound to progesterone and the selective progesterone receptor modulator (SPRM), RU486 [90]. The shifting of H12 after modulator (SERM, SPRM, etc.) binding changes the topology of the AF-2 domain and alters the ability of the receptor to recruit coactivators. It also enhances the ability of the receptor to recruit co-repressors, and thus suppresses gene transcription. Although crystallographic analyses of the VDR have been reported, none have yet examined the

LEONARD P. FREEDMAN AND ALFRED A. RESZKA

conformation of the receptor in the presence of an antagonist. The X-ray structure in the presence of the natural ligand, 1α,25(OH)2D3, shows that VDR has a somewhat larger ligand-binding pocket (697 Å3) than that for ERα (369 Å3) or PR (422 Å3) [91]. This means that 1α,25(OH)2D3 occupies a smaller proportion of the pocket (56%) than with either ERα (63%) or PR (67%). This may suggest that a number of different ligands could bind to VDR and allow the induction of several conformations upon ligand binding. Although we do not have information regarding the structure of VDR in the presence of an antagonist, data do exist for the binding of superagonists [92]. Interestingly, regardless of whether it is bound by 1α,25(OH)2D3 or by either of two superagonists (MC1288 or KH1060), VDR seems to respond in a similar fashion. The overall protein structure is identical and the A- to D-ring moieties of each ligand form identical contacts, bound in identical orientations, with key residues within the VDR LBD. Conclusions based on this work include that superagonist-bound VDR might be more stable and exhibit a longer half-life than seen with the natural ligand. Other work suggests that superagonists can selectively enhance coactivator recruitment. This does resemble the selective effects of ERα and PR ligands on coactivator and corepressor recruitment, although several fundamental differences separate VDR from these nuclear receptors, as discussed below. Selective coactivator recruitment in and of itself is only part of how SERMs and SPRMs elicit tissuespecific responses. To a certain extent, a ligand can elicit only one conformation of the receptor, and this confers a specific affinity of the receptor for each individual coactivator and co-repressor. After this, the altered responses of nuclear receptors to these partial agonists vary cell-type to cell-type, based on the relative levels of expression of each of the coregulators. Thus in a cell with high levels of the right coactivator, e.g.,SRC-1, an agonistic response might be seen, whereas in other cell types, reduced expression of the required coactivator can result in better co-repressor recruitment and resulting transcriptional repression. This can also be achieved through higher levels of co-repressor expression within a given cell type. As noted above for ERα, the altered positioning of H12 upon binding to raloxifene changes the relative affinity of the receptor for the various co-regulators. The clearest evidence for ligand-induced and tissue-selective recruitment of coactivators was generated using chromatin immunoprecipitation (ChIP) analyses [93]. Using MCF-7 and Ishikawa cells as models for breast and uterine stimulation, estradiol was compared to both tamoxifen and raloxifene. Whereas both SERMs display similar antagonism of breast cancer growth, tamoxifen is

CHAPTER 14 Vitamin D Receptor Cofactors: Function, Regulation, and Selectivity

considered to be quite uterotropic. Not surprisingly, both breast cancer antagonists displayed a similar lack of ability to recruit coactivators, such as SRC-1, GRIP1, and AIB1, to estrogen-responsive promoters in MCF-7 breast cancer cells. In addition, both effectively recruited the SMRT and NCoR corepressors, along with two histone deacetylases, all of which act to suppress gene transcription. Interestingly, and consistent with their divergent effects on the uterus, raloxifene maintained this general profile in Ishikawa cells, whereas the response to tamoxifen in these cells resembled that of the natural agonist, estradiol. However, of the two promoters tested (c-Myc and cathepsin D) only the proliferative c-Myc promoter showed the agonist recruitment profile. How then is selectivity generated? In the case of MCF-7 versus Ishikawa cells, the selectivity can be traced back to the relative expression of the coactivator, SRC-1, whose expression is diminished in MCF-7 cells (versus that seen in Ishikawa) [93]. Increased expression (via transfection) of SRC-1, but not GRIP1 or AIB1, enhanced the ability of tamoxifen, but not raloxifene, to induce gene expression off of the c-Myc (and IGF-I) promoter. This type of selective transcriptional response, as influenced by altered ratios of coactivators and co-repressors, was also reported for PR in its response to RU486 [94]. Following the observation that RU486 acts as an agonist in T47D (breast cancer) cells and an antagonist in HeLa (cervical carcinoma) cells, it was observed that the T47D cell line has diminished expression of both NCoR and SMRT corepressors, although expression levels of SRC-1 appeared to be normal. Consistent with its ability to convert tamoxifen into an agonist in MCF-7 cells, overexpression of SRC-1 in HeLa cells can convert RU486 into a partial agonist. RU486 can also be converted into an antagonist in T47D cells, where it otherwise displays partial agonism, by the overexpression of SMRT. This results in the suppression of RU486, but not progesterone, recruitment of SRC-1, to PR and, consequently, suppressed ligand-induced transcription. Distinguished from the SPRM-like activities of RU486, the PR pure antagonist, ZK98, is refractive to SRC-1 overexpression and maintains its pure antagonist profile. Antagonism for this ligand does not require overexpression of any co-repressor in these cells.

B. The VDR Paradigm(s) 1. ANTAGONISTS/PARTIAL AGONISTS

With a paradigm in place for two well-characterized nuclear receptors, it seems logical that VDR should act in an identical fashion in response to partial agonists.

271

Agonists and superagonists should favor binding of coactivators over co-repressors, while partial antagonists should favor recruitment of co-repressors. To some extent this paradigm holds true, except where co-repressors are concerned. The binding of VDR to co-repressors in the unliganded state is relatively weak, and the physiological role for this interaction, if any, remains unknown. Indeed, mammalian two-hybrid analyses directly comparing VDR to thyroid hormone receptor (TR) showed that TR recruitment of SMRT is up to 26-fold stronger than that for VDR, while the recruitment of NCoR was an incredible 335-fold stronger [95]. One might expect that the binding of an antagonist ligand to VDR could enhance recruitment of either of these co-repressors, but this has not been seen. Indeed, the antagonists, ZK159222 and TEI-9647, actually displace a weakly bound NCoR in GST pulldown analyses. A similar response has also been reported for the Gemini ligand, which displaces NCoR binding in both GST pull-down and mammalian two-hybrid analyses (Gonzales et al., in preparation, 2003). Interestingly, however, overexpression of NCoR along with VDR and RXR has been found to partially suppress Gemini and EB1436-mediated transcription to a greater extent than that seen with the natural ligand, 1,25(OH)2VD3 [96]. This provides some evidence for ligand-specific effects of a co-repressor on VDRmediated gene transcription. Unresolved is the means by which this transcriptional repression is mediated, given the displacement of the co-repressor in the presence of the Gemini ligand. Meanwhile other recent studies have shown that the hairless co-repressor, which suppresses VDR-mediated transcription in the presence of 1,25(OH)2VD3, also binds to the VDR [97]. Consistent with its ability to suppress agonist-induced transcription, the hairless co-repressor binds VDR with equal affinity in the presence or absence of the natural ligand. Whether or not the hairless co-repressor could play a role in partial antagonism by vitamin D analogs remains to be established. 2. SUPERAGONISTS AND SELECTIVE COACTIVATOR RECRUITMENT

In considering the actions of VDR superagonists, several mechanisms seem to play a role, each changing, depending on the ligand tested (Fig. 2). For the superagonist ligand, numerous effects have been described relating to enhanced VDR-RXR heterodimer formation and enhanced recruitment of several of the coactivators. The formation of a VDR-RXR heterodimer in and of itself is an interesting aspect of VDR biology [98]. For many years, RXR was considered a silent partner for VDR, as transcription off of the heterodimer was dependent on VDR ligands alone. Recently, however,

272

LEONARD P. FREEDMAN AND ALFRED A. RESZKA

CBP P300 CBP P300

GRIP-1

80

PCAF 150

PCAF CBP P300

AIB-1

230

130 Med6

SRC-1 205

100

PCAF

Med7

CTD

DR3

Selective and partial agonists Natural ligand and selective/super agonists Super agonists RXR

FIGURE 2

VDR

O Ligand

Ligand-dependent coactivator complex recruitment to VDR

several lines of evidence have suggested that RXR plays a prominent role in VDR biology in both determining subcellular localization and, indeed, participating in the control of gene transcription. For instance, the dimerization domain of RXR encodes a nuclear localization sequence that is required for VDR translocation to the nucleus [99]. Mutation of this sequence traps VDR in the cytoplasm when in the unliganded state. Interestingly, however, 1,25(OH)2VD3 triggers the localization of NLS-deficient RXR-VDR into the nucleus. Paradoxically, it also enhances the export rate of the normal heterodimer from the nucleus. Within the nucleus, RXR also contributes to 1,25(OH)2VD3-induced transcription. This has been demonstrated by the ability of RXR-specific LXXLL motif peptides to suppress 1,25(OH)2VD3-induced transcription [100]. Very recently, VDR ligands have been shown to elicit a “phantom ligand effect” on RXR and thus to enable the recruitment of p160 family coactivators to RXR itself [101]. This could play a role in superagonist activity by certain VDR ligands, as suggested by the ability of certain of these ligands to induce heterodimer formation at concentrations well below that required by the natural ligand. This has

been demonstrated for the 20-epi ligands, MC1627, MC1288, and 2-methylene-19-nor-(20S)-1,25-dihydroxyvitamin D3 (2-MD), which bind VDR with equal affinity to 1,25(OH)2VD3, yet induce heterodimer formation at concentrations 10 to 100-fold lower [102–105]. It remains unknown whether or not the RXR moiety of the VDR heterodimer recruits coactivators under these circumstances. Enhanced heterodimer formation may contribute to the enhanced transcriptional activity of MC1288, but this is likely only part of the story (Fig. 2). Perhaps more important is the ability of this ligand to enhance recruitment of the DRIP complex to VDR [102]. Interestingly, while MC1288 induces enhanced DRIP205 recruitment, recruitment of SRC-1 and GRIP-1 coactivators is no greater than that seen with the natural ligand [102,103]. Meanwhile, 2-MD shows enhanced recruitment of DRIP205, SRC-1, and GRIP-1 [105], whereas MC1627 shows enhanced DRIP205 and GRIP-1 recruitment. In the context of myeloid cell differentiation, which is potently induced by both MC1288 and MC1627, it is the recruitment of DRIP205 that best correlates with the phenomenon. With regard to 2-MD, it remains unknown whether or not any specific

CHAPTER 14 Vitamin D Receptor Cofactors: Function, Regulation, and Selectivity

coactivator recruitment is responsible for enhanced bone anabolic activity. Selective recruitment of DRIP205 and p160 family members can also be regulated as part of the normal differentiation process, as demonstrated in the keratinocyte [106]. In this study, Vitamin D3 analogs were not examined, but by inference, analogs displaying selective coactivator recruitment are likely to show enhanced tissue selectivity. Using GST-VDR as bait, coactivator complexes were purified from both proliferating and differentiated keratinocytes in a 1α,25(OH)2VD3 dependent fashion. Interestingly, whereas DRIP205 and associated proteins predominated in the proliferating cells, AIB-1 (SRC-3) and GRIP-1 predominated after differentiation. This was associated with reciprocal changes in coactivator expression, DRIP205 declining with differentiation and AIB-1 increasing. Although overexpression of either cofactor could enhance transcription in the proliferation phase, only AIB-1 was effective postdifferentiation. We see this as intriguing, since ligands such as MC1288 can enhance recruitment of DRIP-205 during their induction of differentiation, as described earlier. Putatively, such selective coactivator recruitment, coupled with selective expression in the keratinocyte, could enhance the differentiation response in this system. Indeed, MC1288 suppresses proliferation and stimulates keratinocyte differentiation at concentrations 600- and 700-fold lower than seen with the natural ligand, respectively [107]. The enhanced potency of MC1288 in the keratinocyte versus myeloid cells, whereby MC1288 is approximately 100-fold more potent than 1α,25(OH)2VD3, may suggest a somewhat synergistic effect of the ligand specificity with the heightened and selective expression of DRIP-205 in the proliferating keratinocyte. That VDR should show different capacities to recruit one coactivator versus another in a ligand-dependent state is consistent with somewhat different means of recruitment for each coactivator. Indeed, a direct comparison of GRIP-1 and AIB-1 showed that different sequences within VDR are required to recruit each coactivator [108]. Furthermore, RXR seems to play a more active role in recruitment of AIB-1 than GRIP-1 in VDR-RXR-mediated transcription. Nonetheless, there is greater evidence for enhanced GRIP-1 recruitment by VDR ligands than for any other coactivator. Indeed, whereas several ligands can induce an interaction between VDR and SRC-1, GRIP-1, and AIB-1, OCT (22-oxa 1α,25(OH)2VD3) only poorly recruits either SRC-1 or AIB-1, while its recruitment of GRIP-1 is comparable to that of the natural ligand [109]. Consistent with this, GRIP-1, but not SRC-1, overexpression

273

substantially enhances OCT-induced transcription in a fashion reminiscent of SERM effects on ERα, as described earlier. Analyses of a broad panel of vitamin D analogs showed that GRIP-1 was more consistently recruited at levels closer to that of the natural ligand than was AIB-1 [104]. For these ligands, transcriptional activity was enhanced by overexpression of GRIP-1, whereas, AIB-1 overexpression actually suppressed transcription. To date, we know of only one example of cell type–dependent coactivator recruitment by VDR modulators. This type of activity was demonstrated for the noncalcemic analog Ro-26–9228, which recruits GRIP-1 in human osteoblasts, but not in CaCo-2 cells [110]. In this study, the coactivator was tested in vitro as a GST-fusion protein; therefore the differential recruitment of GRIP-1 to VDR in these two different cell types could not be attributed to altered expression of the coactivators. Since similar selective recruitment of DRIP-205 and SRC-1 was also reported, it seems that VDR is somehow modified by its expression in these different cell types. This also distinguishes this analog from numerous others, which show more or less selective recruitment of GRIP-1 over other tested coactivators. In summary, the VDR-RXR heterodimer seems to have its own mechanisms for the genesis of VDRMlike activity. As with ERα and PR, VDR can display somewhat selective coactivator recruitment in the presence of a broad range of tissue-selective ligands. Although the data are limited, the preponderance of evidence suggests that co-repressor recruitment is not at all involved in the genesis of VDRM activity. Binding of NCoR and SMRT is weak in the unliganded state, and this is further reduced by the binding of agonist and antagonist ligands. Coactivator-dependent effects on superagonism are also seen with certain ligands. These ligands display both enhanced VDRRXR heterodimer formation and enhanced recruitment of DRIP205 and, perhaps, GRIP-1 in a ligand-specific manner (Fig. 2). Overall, this suggests that the modulation of VDR transcriptional activity is coactivatorspecific. This sets a new paradigm for VDR that is distinct from that of ERα and PR, which respond to coactivator and co-repressors, depending on relative expression levels.

VI. CONCLUSIONS Transcriptional regulation by 1,25(OH)2D3 can be dissected into several functional activities that are mediated by VDR. Generally, transactivation requires that a repressed state of chromatin has to be disrupted

274 in a given target gene’s regulatory regions, together with the ability to promote productive elongation of RNA products by the RNA Pol II machinery at the site of transcription initiation. Several candidates for such activities that are recruited by nuclear receptors in direct response to ligand binding have been identified, as has been described in this review. The HAT activitycontaining coactivators (SRC/p160 family, CBP/p300, PCAF, etc.) appear at face value to act primarily in the disruption of chromatin through histone modifications, although a series of additional, provocative targets corresponding to ATP-dependent chromatin remodelers have more recently been identified. The regulation of RNA Pol II and its ability to respond to activators appears to be mediated by a distinct type of cofactor complex (DRIP/ARC/TRAP/SMCC/Mediator) whose functions are not yet completely elucidated. Given the fact that many classes of activators interact with DRIP beyond VDR and nuclear receptors, this complex must be considered as a regulatory panel for RNA Pol II rather than an exclusive target for nuclear receptors. DRIP may be therefore viewed as a downstream target of multiple transcription activators, perhaps conferring on RNA Pol II the ability to simultaneously integrate multiple signaling pathways onto a single promoter in vivo. This model may provide a key for an interpretation of some unresolved mechanisms involving the simultaneous cross-talk of several signaling systems.

References 1. Zawel L, Reinberg D 1995 Common themes in assembly and function of eukaryotic transcription complexes. Annu Rev Biochem 64:533–561. 2. Roeder RG 1996 The role of general initiation factors in transcription by RNA polymerase II. Trends Biochem Sci 21:327–335. 3. Lemon BD, Freedman LP 1999 Nuclear receptor cofactors as chromatin remodelers. Curr Opin Genet Dev 9:499–504. 4. Lee TI, Young RA 2000 Transcription of eukaryotic proteincoding genes. Ann Rev Genet 34:77–137. 5. Blanco JC, Wang IM, Tsai SY, Tsai MJ, O’Malley BW, Jurutka PW, Haussler MR, Ozato K 1995 Transcription factor TFIIB and the vitamin D receptor cooperatively activate ligand-dependent transcription. Proc Natl Acad Sci USA 92:1535–1539. 6. Jurutka P, Hsieh J, Remus L, Whitfield G, Thompson P, Haussler C, Blanco J, Ozato K, Haussler M 1997 Mutations in the 1,25-dihydroxyvitamin D3 receptor identifying C-terminal amino acids required for transcriptional activation that are functionally dissociated from hormone binding, heterodimeric DNA binding, and interaction with basal transcription factor IIB, in vitro. J Biol Chem 272:14592–14599. 7. Masuyama H, Jefcoat SC, Jr., MacDonald PN 1997 The N-terminal domain of transcription factor IIB is required for direct interaction with the vitamin D receptor and participates in vitamin D-mediated transcription. Mol Endocrinol 11:218–228.

LEONARD P. FREEDMAN AND ALFRED A. RESZKA

8. MacDonald P, Sherman D, Dowd D, Jefcoat SJ, DeLisle R 1995 The vitamin D receptor interacts with general transcription factor IIB. J Biol Chem 270:4748–4752. 9. Lemon BD, Fondell JD, Freedman LP 1997 Retinoid X receptor:vitamin D3 receptor heterodimers promote stable preinitiation complex formation and direct 1,25-dihydroxyvitamin D3-dependent cell-free transcription. Mol Cell Biol 17:1923–1937. 10. Mengus G, May M, Carre L, Chambon P, Davidson I 1997 Human TAF(II)135 potentiates transcriptional activation by the AF-2s of the retinoic acid, vitamin D3, and thyroid hormone receptors in mammalian cells. Genes Dev 11:1381–1395. 11. Lavigne AC, Mengus G, Gangloff YG, Wurtz JM, Davidson I 1999 Human TAF(II)55 interacts with the vitamin D3 and thyroid hormone receptors and with derivatives of the retinoid X receptor that have altered transactivation properties. Mol Cell Biol 19:5486–5494. 12. May M, Mengus G, Lavigne AC, Chambon P, Davidson I 1996 Human TAF(II28) promotes transcriptional stimulation by activation function 2 of the retinoid X receptors. EMBO J 15:3093–3104. 13. Torchia J, Glass C, Rosenfeld MG 1998 Co-activators and co-repressors in the integration of transcriptional responses. Curr Opin Cell Biol 10:373–383. 14. Freedman LP 1999 Increasing the complexity of coactivation in nuclear receptor signaling. Cell 97:5–8. 15. Onate SA, Tsai SY, Tsai MJ, O’Malley BW 1995 Sequence and characterization of a co-activator for the steroid hormone receptor superfamily. Science 270:1354–1357. 16. McKenna NJ, Xu J, Nawaz Z, Tsai SY, Tsai MJ, O’Malley BW 1999 Nuclear receptor coactivators: multiple enzymes, multiple complexes, multiple functions. J Steroid Biochem Mol Biol 69:3–12. 17. Xu L, Glass CK, Rosenfeld MG 1999 Coactivator and corepressor complexes in nuclear receptor function. Curr Opin Genet Dev 9:140–147. 18. Spencer TE, Jenster G, Burcin MM, Allis CD, Zhou JX, Mizzen CA, McKenna NJ, Onate SA, Tsai SY, Tsai MJ, Omalley BW 1997 Steroid receptor coactivator-1 is a histone acetyltransferase. Nature 389:194–198. 19. Chen HW, Lin RJ, Schiltz RL, Chakravarti D, Nash A, Nagy L, Privalsky ML, Nakatani Y, Evans RM 1997 Nuclear receptor coactivator ACTR is a novel histone acetyltransferase and forms a multimeric activation complex with P/CAF and CBP/p300. Cell 90:569–580. 20. Hong H, Kohli K, Garabedian MJ, Stallcup MR 1997 GRIP1, a transcriptional coactivator for the AF-2 transactivation domain of steroid, thyroid, retinoid, and vitamin D receptors. Mol Cell Biol 17:2735–2744. 21. Masuyama H, Brownfield CM, StArnaud R, MacDonald PN 1997 Evidence for ligand-dependent intramolecular folding of the AF-2 domain in vitamin D receptor-activated transcription and coactivator interaction. Mol Endocrinol 11:1507–1517. 22. Xu J, Qiu Y, DeMayo FJ, Tsai SY, Tsai MJ, O’Malley BW 1998 Partial hormone resistance in mice with disruption of the steroid receptor coactivator-1 (SRC-1) gene. Science 279:1922–1925. 23. Liu Z, Wong J, Tsai SY, Tsai MJ, O’Malley BW 1999 Steroid receptor coactivator-1 (SRC-1) enhances ligand-dependent and receptor-dependent cell-free transcription of chromatin. Proc Natl Acad Sci USA 96:9485–9490. 24. Baudino TA, Kraichely DM, Jefcoat SC, Jr., Winchester SK, Partridge NC, MacDonald PN 1998 Isolation and characterization of a novel coactivator protein, NCoA-62, involved

CHAPTER 14 Vitamin D Receptor Cofactors: Function, Regulation, and Selectivity

25.

26.

27.

28. 29. 30.

31. 32. 33. 34. 35. 36. 37.

38.

39.

40. 41. 42.

in vitamin D–mediated transcription. J Biol Chem 273: 16434–16441. Zhang C, Dowd DR, Staal A, Gu C, Lian JB, van Wijnen AJ, Stein GS, MacDonald PN 2003 Nuclear coactivator-62 kDa/Ski-interacting protein is a nuclear matrix-associated coactivator that may couple vitamin D receptor-mediated transcription and RNA splicing. J Biol Chem 278: 35325–35336. Le Douarin B, vom Baur E, Zechel C, Heery D, Heine M, Vivat V, Gronemeyer H, Losson R, Chambon P 1996 Liganddependent interaction of nuclear receptors with potential transcriptional intermediary factors (mediators). Phil Trans R Soc Lond B Biol Sci 351:569–578. Fraser RA, Heard DJ, Adam S, Lavigne AC, Le Douarin B, Tora L, Losson R, Rochette-Egly C, Chambon P 1998 The putative cofactor TIF1alpha is a protein kinase that is hyperphosphorylated upon interaction with liganded nuclear receptors. J Biol Chem 273:16199–16204. Goldman PS, Tran VK, Goodman RH 1997 The multifunctional role of the co-activator CBP in transcriptional regulation. Recent Prog Horm Res 52:103–119. McKenna NJ, Nawaz Z, Tsai SY, Tsai M-J, O’Malley BW 1998 Distinct steady-state nuclear receptor coregulator complexes exist in vivo. Proc Natl Acad Sci USA 95:11697–11702. Lanz RB, McKenna NJ, Onate SA, Albrecht U, Wong J, Tsai SY, Tsai MJ, O’Malley BW 1999 A steroid receptor coactivator, SRA, functions as an RNA and is present in an SRC-1 complex. Cell 97:17–27. Bannister AJ, Kouzarides T 1996 The CBP co-activator is a histone acetyltransferase. Nature 384:641–643. Ogryzko VV, Schiltz RL, Russanova V, Howard BH, Nakatani Y 1996 The transcriptional coactivators p300 and CBP are histone acetyltransferases. Cell 87:953–959. Gu W, Roeder RG 1997 Activation of p53 sequence-specific DNA binding by acetylation of the p53 C-terminal domain. Cell 90:595–606. Imhof A, Yang XJ, Ogryzko VV, Nakatani Y, Wolffe AP, Ge H 1997 Acetylation of general transcription factors by histone acetyltransferases. Curr Biol 7:689–692. Chen H, Lin RJ, Xie W, Wilpitz D, Evans RM 1999 Regulation of hormone-induced histone hyperacetylation and gene activation via acetylation of an acetylase. Cell 98:675–686. Rachez C, Freedman LP 2000 Mediator complexes and transcription. Curr Opin Cell Biol 13:274–280. Rachez C, Suldan Z, Ward J, Chang CP, Burakov D, Erdjument-Bromage H, Tempst P, Freedman LP 1998 A novel protein complex that interacts with the vitamin D3 receptor in a ligand-dependent manner and enhances VDR transactivation in a cell-free system. Genes Dev 12:1787–1800. Rachez C, Lemon BD, Suldan Z, Bromleigh V, Gamble M, Naar AM, Erdjument-Bromage H, Tempst P, Freedman LP 1999 Ligand-dependent transcription activation by nuclear receptors requires the DRIP complex. Nature 398:824–828. Naar AM, Beaurang PA, Robinson KM, Oliner JD, Avizonis D, Scheek S, Zwicker J, Kadonaga JT, Tjian R 1998 Chromatin, TAFs, and a novel multiprotein coactivator are required for synergistic activation by Sp1 and SREBP-1a in vitro. Genes Dev 12:3020–3031. Naar AM, Beaurang PA, Zhou S, Abraham S, Solomon W, Tjian R 1999 Composite co-activator ARC mediates chromatindirected transcriptional activation. Nature 398:828–832. Fondell JD, Ge H, Roeder RG 1996 Ligand induction of a transcriptionally active thyroid hormone receptor coactivator complex. Proc Natl Acad Sci USA 93:8329–8333. Ito M, Yuan CX, Malik S, Gu W, Fondell JD, Yamamura S, Fu ZY, Zhang X, Qin J, Roeder RG 1999 Identity between

43.

44. 45. 46.

47.

48. 49. 50. 51.

52.

53.

54.

55. 56.

57.

58.

59.

275

TRAP and SMCC complexes indicates novel pathways for the function of nuclear receptors and diverse mammalian activators. Mol Cell 3:361–370. Gu W, Malik S, Ito M, Yuan CX, Fondell JD, Zhang X, Martinez E, Qin J, Roeder RG 1999 A novel human SRB/MEDcontaining cofactor complex, SMCC, involved in transcription regulation [published erratum appears in Mol Cell 1999 Apr;3(4):following 541]. Mol Cell 3:97–108. Boyer TG, Martin ME, Lees E, Ricciardi RP, Berk AJ 1999 Mammalian Srb/Mediator complex is targeted by adenovirus E1A protein [see comments]. Nature 399:276–279. Ryu S, Zhou S, Ladurner AG, Tjian R 1999 The transcriptional cofactor complex CRSP is required for activity of the enhancer-binding protein Sp1. Nature 397:446–450. Sun X, Zhang Y, Cho H, Rickert P, Lees E, Lane W, Reinberg D 1998 NAT, a human complex containing Srb polypeptides that functions as a negative regulator of activated transcription. Mol Cell 2:213–222. Jiang YW, Veschambre P, Erdjument-Bromage H, Tempst P, Conaway JW, Conaway RC, Kornberg RD 1998 Mammalian mediator of transcriptional regulation and its possible role as an end-point of signal transduction pathways. Proc Natl Acad Sci USA 95:8538–8543. Parvin JD, Young RA 1998 Regulatory targets in the RNA polymerase II holoenzyme. Curr Opin Genet Dev 8:565–570. Myer VE, Young RA 1998 RNA polymerase II holoenzymes and subcomplexes. J Biol Chem 273:27757–27760. Koleske AJ, Young RA 1994 An RNA polymerase II holoenzyme responsive to activators [see comments]. Nature 368:466–469. Hengartner CJ, Myer VE, Liao SM, Wilson CJ, Koh SS, Young RA 1998 Temporal regulation of RNA polymerase II by Srb10 and Kin28 cyclin-dependent kinases. Mol Cell 2:43–53. Kim YJ, Bjorklund S, Li Y, Sayre MH, Kornberg RD 1994 A multiprotein mediator of transcriptional activation and its interaction with the C-terminal repeat domain of RNA polymerase II. Cell 77:599–608. Lee YC, Min S, Gim BS, Kim YJ 1997 A transcriptional mediator protein that is required for activation of many RNA polymerase II promoters and is conserved from yeast to humans. Mol Cell Biol 17:4622–4632. Myers LC, Gustafsson CM, Bushnell DA, Lui M, ErdjumentBromage H, Tempst P, Kornberg RD 1998 The Med proteins of yeast and their function through the RNA polymerase II carboxy-terminal domain. Genes Dev 12:45–54. Thompson CM, Young RA 1995 General requirement for RNA polymerase II holoenzymes in vivo. Proc Natl Acad Sci USA 92:4587–4590. Myers LC, Gustafsson CM, Hayashibara KC, Brown PO, Kornberg RD 1999 Mediator protein mutations that selectively abolish activated transcription [see comments]. Proc Natl Acad Sci USA 96:67–72. Rachez C, Lemon BD, Suldan Z, Bromleigh V, Gamble M, Naar AM, Erdjument Bromage H, Tempst P, Freedman LP 1999 Ligand-dependent transcription activation by nuclear receptors requires the DRIP complex. Nature 398:824–828. Chiba N, Suldan Z, Freedman LP, Parvin JD 2000 Binding of liganded vitamin D receptor to the vitamin D receptor interacting protein coactivator complex induces interaction with RNA polymerase II holoenzyme. J Biol Chem 275:10719–10722. Kretzschmar M, Kaiser K, Lottspeich F, Meisterernst M 1994 A novel mediator of class II gene transcription with homology to viral immediate-early transcriptional regulators. Cell 78:525–534.

276 60. Ge H, Zhao Y, Chait BT, Roeder RG 1994 Phosphorylation negatively regulates the function of coactivator PC4. Proc Natl Acad Sci USA 91:12691–12695. 61. Henry NL, Bushnell DA, Kornberg RD 1996 A yeast transcriptional stimulatory protein similar to human PC4. J Biol Chem 271:21842–21847. 62. Carlson M, Laurent BC 1994 The SNF/SWI family of global transcriptional activators. Curr Opin Cell Biol 6:396–402. 63. Cote J, Quinn J, Workman JL, Peterson CL 1994 Stimulation of GAL4 derivative binding to nucleosomal DNA by the yeast SWI/SNF complex. Science 265:53–60. 64. Imbalzano AN, Kwon H, Green MR, Kingston RE 1994 Facilitated binding of TATA-binding protein to nucleosomal DNA. Nature 370:481–485. 65. Wang W, Xue Y, Zhou S, Kuo A, Cairns BR, Crabtree GR 1996 Diversity and specialization of mammalian SWI/SNF complexes. Genes Dev 10:2117–2130. 66. Schnitzler G, Sif S, Kingston RE 1998 Human SWI/SNF interconverts a nucleosome between its base state and a stable remodeled state. Cell 94:17–27. 67. Guyon JR, Narlikar GJ, Sif S, Kingston RE 1999 Stable remodeling of tailless nucleosomes by the human SWI-SNF complex. Mol Cell Biol 19:2088–2097. 68. Yoshinaga SK, Peterson CL, Herskowitz I, Yamamoto KR 1992 Roles of SWI1, SWI2, and SWI3 proteins for transcriptional enhancement by steroid receptors. Science 258: 1598–1604. 69. Muchardt C, Yaniv M 1993 A human homologue of Saccharomyces cerevisiae SNF2/SWI2 and Drosophila brm genes potentiates transcriptional activation by the glucocorticoid receptor. EMBO J 12:4279–4290. 70. Ichinose H, Garnier JM, Chambon P, Losson R 1997 Liganddependent interaction between the estrogen receptor and the human homologues of SWI2/SNF2. Gene 188:95–100. 71. Sumi-Ichinose C, Ichinose H, Metzger D, Chambon P 1997 SNF2beta-BRG1 is essential for the viability of F9 murine embryonal carcinoma cells. Mol Cell Biol 17:5976–5986. 72. Beato M, Sanchez Pacheco A 1996 Interaction of steroid hormone receptors with the transcription initiation complex. Endocr Rev 17:587–609. 73. Fryer CJ, Archer TK 1998 Chromatin remodelling by the glucocorticoid receptor requires the BRG1 complex. Nature 393:88–91. 74. Kornberg RD, Lorch Y 1999 Chromatin-modifying and -remodeling complexes. Curr Opin Genet Dev 9:148–151. 75. Ito T, Bulger M, Pazin MJ, Kobayashi R, Kadonaga JT 1997 ACF, an ISWI-containing and ATP-utilizing chromatin assembly and remodeling factor. Cell 90:145–155. 76. Varga-Weisz PD, Wilm M, Bonte E, Dumas K, Mann M, Becker PB 1997 Chromatin-remodelling factor CHRAC contains the ATPases ISWI and topoisomerase II. Nature 388: 598–602. 77. LeRoy G, Orphanides G, Lane WS, Reinberg D 1998 Requirement of RSF and FACT for transcription of chromatin templates in vitro. Science 282:1900–1904. 78. Tsukiyama T, Wu C 1995 Purification and properties of an ATPdependent nucleosome remodeling factor. Cell 83:1011–1020. 79. Mizuguchi G, Tsukiyama T, Wisniewski J, Wu C 1997 Role of nucleosome remodeling factor NURF in transcriptional activation of chromatin. Mol Cell 1:141–150. 80. Xue Y, Wong J, Moreno GT, Young MK, Cote J, Wang W 1998 NURD, a novel complex with both ATP-dependent chromatin-remodeling and histone deacetylase activities. Mol Cell 2:851–861.

LEONARD P. FREEDMAN AND ALFRED A. RESZKA

81. Zhang Y, LeRoy G, Seelig HP, Lane WS, Reinberg D 1998 The dermatomyositis-specific autoantigen Mi2 is a component of a complex containing histone deacetylase and nucleosome remodeling activities. Cell 95:279–289. 82. Kitagawa H, Fujiki R, Yoshimura K, Mezaki Y, Uematsu Y, Matsui D, Ogawa S, Unno K, Okubo M, Tokita A, Nakagawa T, Ito T, Ishimi Y, Nagasawa H, Matsumoto T, Yanagisawa J, Kato S 2003 The chromatin-remodeling complex WINAC targets a nuclear receptor to promoters and is impaired in Williams syndrome. Cell 113:905–917. 83. Lemon B, Inouye C, King DS, Tjian R 2001 Selectivity of chromatin-remodelling cofactors for ligand-activated transcription. Nature 414:924–928. 84. Fondell JD, Guermah M, Malik S, Roeder RG 1999 Thyroid hormone receptor-associated proteins and general positive cofactors mediate thyroid hormone receptor function in the absence of the TATA box-binding protein-associated factors of TFIID. Proc Natl Acad Sci USA 96:1959–1964. 85. Treuter E, Johansson L, Thomsen JS, A Wr, Leers J, Pelto-Huikko M, Sjberg M, Wright AP, Spyrou G, Gustafsson J 1999 Competition between thyroid hormone receptor-associated protein (TRAP) 220 and transcriptional intermediary factor (TIF) 2 for binding to nuclear receptors. Implications for the recruitment of trap and p160 coactivator complexes. J Biol Chem 274:6667–6677. 86. Rachez C, Gamble M, Chang C-PB, Atkins GB, Lazar MA, Freedman LP 2000 The DRIP complex and SRC-1/p160 coactivators share similar nuclear receptor binding determinants but constitute functionally distinct complexes. Mol Cell Biol 20:2718–2726. 87. Hittelman AB, Burakov D, JA Ii-L, Freedman LP, Garabedian MJ 1999 Differential regulation of glucocorticoid receptor transcriptional activation via AF-1-associated proteins. EMBO J 18:5380–5388. 88. Lau JF, Nusinzon I, Burakov D, Freedman LP, Horvath CM 2003 Role of metazoan mediator proteins in interferonresponsive transcription. Mol Cell Biol 23:620–628. 89. Brzozowski AM, Pike AC, Dauter Z, Hubbard RE, Bonn T, Engstrom O, Ohman L, Greene GL, Gustafsson JA, Carlquist M 1997 Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 389:753–758. 90. Williams SP, Sigler PB 1998 Atomic structure of progesterone complexed with its receptor. Nature 393:392–396. 91. Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179. 92. Tocchini Valentini G, Rochel N, Wurtz JM, Mitschler A, Moras D 2001 Crystal structures of the vitamin D receptor complexed to superagonist 20-epi ligands. Proc Natl Acad Sci USA 98:5491–5496. 93. Shang Y, Brown M 2002 Molecular determinants for the tissue specificity of SERMs. Science 295:2465–2468. 94. Liu Z, Auboeuf D, Wong J, Chen JD, Tsai SY, Tsai MJ, O’Malley BW 2002 Coactivator/corepressor ratios modulate PR-mediated transcription by the selective receptor modulator RU486.Proc Natl Acad Sci USA 99:7940–7944. 95. Tagami T, Lutz WH, Kumar R, Jameson JL 1998 The interaction of the vitamin D receptor with nuclear receptor corepressors and coactivators. Biochem Biophys Res Commun 253:358–363. 96. Herdick M, Carlberg C 2000 Agonist-triggered modulation of the activated and silent state of the vitamin D(3) receptor by interaction with co-repressors and co-activators. J Mol Biol 304:793–801.

CHAPTER 14 Vitamin D Receptor Cofactors: Function, Regulation, and Selectivity

97. Hsieh J-C, Sisk JM, Jurutka PW, Haussler CA, Slater SA, Haussler MR, Thompson CC 2003 Physical and functional interaction between the vitamin D receptor and hairless corepressor, two proteins required for hair cycling. J Biol Chem M304886200. 98. Cheskis B, Freedman LP 1994 Ligand modulates the conversion of DNA-bound vitamin D3 receptor (VDR) homodimers into VDR-retinoid X receptor heterodimers. Mol Cell Biol 14:3329–3338. 99. Prufer K, Barsony J 2002 Retinoid X receptor dominates the nuclear import and export of the unliganded vitamin D receptor. Mol Endocrinol 16:1738–1751. 100. Pike JW, Pathrose P, Barmina O, Chang CY, McDonnell DP, Yamamoto H, Shevde NK 2003 Synthetic LXXLL peptide antagonize 1,25-dihydroxyvitamin D3-dependent transcription. J Cell Biochem 88:252–258. 101. Bettoun DJ, Burris TP, Houck KA, Buck DW, II, Stayrook KR, Khalifa B, Lu J, Chin WW, Nagpal S 2003 Retinoid X receptor is a non-silent major contributor to vitamin D receptor-mediated transcriptional activation. Mol Endocrinol me.2003-0148. 102. Yang W, Freedman LP 1999 20-Epi analogues of 1,25-dihydroxyvitamin D3 are highly potent inducers of DRIP coactivator complex binding to the vitamin D3 receptor. J Biol Chem 274:16838–16845. 103. Liu YY, Nguyen C, Peleg S 2000 Regulation of ligandinduced heterodimerization and coactivator interaction by the activation function-2 domain of the vitamin D receptor. Mol Endocrinol 14:1776–1787. 104. Issa LL, Leong GM, Sutherland RL, Eisman JA 2002 Vitamin D analogue-specific recruitment of vitamin D receptor coactivators. J Bone Miner Res 17:879–890.

277

105. Yamamoto H, Shevde NK, Warrier A, Plum LA, DeLuca HF, Pike JW 2003 2-Methylene-19-nor-(20S)-1,25-dihydroxyvitamin D3 potently stimulates gene-specific DNA binding of the vitamin D receptor in osteoblasts. J Biol Chem 278:31756–31765. 106. Oda Y, Sihlbom C, Chalkley RJ, Huang L, Rachez C, Chang C-PB, Burlingame AL, Freedman LP, Bikle DD 2003 Two distinct coactivators, DRIP/Mediator and SRC/p160, are differentially involved in VDR transactivation during keratinocyte differentiation. Mol Endocrinol me.2003-0063. 107. Gniadecki R 1997 Effects of 1,25-dihydroxyvitamin D3 and its 20-epi analogues (MC 1288, MC 1301, KH 1060), on clonal keratinocyte growth: evidence for differentiation of keratinocyte stem cells and analysis of the modulatory effects of cytokines. Br J Pharmacol 120:1119–1127. 108. Issa LL, Leong GM, Barry JB, Sutherland RL, Eisman JA 2001 Glucocorticoid receptor-interacting protein-1 and receptor-associated coactivator-3 differentially interact with the vitamin D receptor (VDR) and regulate VDR-retinoid X receptor transcriptional cross-talk. Endocrinology 142: 1606–1615. 109. Takeyama K, Masuhiro Y, Fuse H, Endoh H, Murayama A, Kitanaka S, Suzawa M, Yanagisawa J, Kato S 1999 Selective interaction of vitamin D receptor with transcriptional coactivators by a vitamin D analog. Mol Cell Biol 19: 1049–1055. 110. Peleg S, Ismail A, Uskokovic MR, Avnur Z 2003 Evidence for tissue- and cell-type selective activation of the vitamin D receptor by Ro-26–9228, a noncalcemic analog of vitamin D3. J Cell Biochem 88:267–273.

CHAPTER 15

Vitamin D Nuclear Receptor Ligand-Binding Domain Crystal Structures NATACHA ROCHEL AND DINO MORAS Département de Biologie et de Génomique Structurales, IGBMC, CNRS/INSERM/Université Louis Pasteur, Illkirch, France

I. II. III. IV. V.

Introduction Biological Properties of hVDR∆ Solution Studies Crystal Structure of hVDR∆ Bound to 1α,25(OH)2D3 Mutant Analysis

I. INTRODUCTION All nuclear receptors (NRs) are modular proteins that harbor one DNA-binding domain and one ligandbinding domain (LBD) (Fig. 1) [1,2]. NRs act as agonistinduced factors that enhance or suppress transcription of their target genes. Certain receptors can act as silencers of transcription in the absence of ligands or the presence of certain antagonists. Agonists induce a change in the structure of the NR that allows interaction with coactivators that can acetylate histones, which prepare target gene promoters through decondensation of the corresponding chromatin [3]. Following decondensation, a second complex, called TRAP (thyroid receptor–associated proteins) or DRIP (Vitamin D receptor–interacting proteins) appears to establish linkage to the basal transcriptional machinery [4,5]. The ligand-binding domain of nuclear receptors harbors ligand-dependent activation function or AF-2, a major interface for dimerization with RXR and interface for coactivators as well as co-repressors. This domain is highly structured and encodes several functions in a ligand-dependent manner. Detailed molecular insights into the structure–function of nuclear receptors have been gained by the elucidation of the crystal structures of the LBD alone or in complexes with agonists, antagonists, and co-regulator peptides (Fig. 2). The first 3D structures reported for NR LBDs were those of the unliganded RXRα [6], the all-trans-retinoic acid–bound RARγ [7], and the agonist-bound thyroid receptor TRβ [8]. Unliganded receptors are referred to as apo receptor forms while liganded referred to as holo forms. To date, the crystal structures of 23 distinct NR LBDs VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

VI. VII. VIII. IX.

Crystal Structure of zVDR Bound to 1α,25(OH)2D3 Structure of hVDR Complexed to Superagonist Ligands Structure of zVDR in Complex with Gemini Conclusion References

and four LBD heterodimers have been reported. The later concern the EcR LBD bound to an antagonist and RXRα LBD bound to phospholipid ligand [9] and LXβ and RXRβ bound to agonists ligands [10]. The general fold of nuclear receptors consists of a threelayered α-helical sandwich. The helices have been designated H1 to H12, according to the first crystal structures of RXR and RAR [11]. The different crystal structures of apo and holo forms of RXR LBDs as well as extensive NR mutagenesis has demonstrated that the LBD undergoes a major conformational change upon ligand binding (Fig. 2). Ligand-induced conformational changes for NRs have been likened to the “mousetrap mechanism” described for RAR [7]. Upon ligand binding, helix H11 is repositioned in the continuity of helix H10, and helix H12 swings to seal the binding cavity while the ω-loop flips over underneath helix H6 carrying along the N-terminal part of helix H3. Helix 12, also referred as the activating domain AD of the AF-2 function, stabilizes ligand binding by contributing to the hydrophobic environment, in some cases making additional contacts with the ligand. The structural data reveal that H12, when folded back onto the core of the LBD, forms a hydrophobic cleft together with other surface-exposed residues that accommodates the “NR box” of coactivators such as members of the SRC-1/TIF2 family [12–14]. VDRs have been characterized from mammals [15–17], birds [18], Xenopus laevis [19], Paralichtus olivaceus [20], zebrafish (GenBank accession number AAF21427), and recently from lampreys [21]. Sequence analysis of the VDR subfamily members reveals that VDR presents a large insertion domain at the N-terminal Copyright © 2005, Elsevier, Inc. All rights reserved.

280

NATACHA ROCHEL AND DINO MORAS

A/B

C

D

E

N

F AD

C

DNA binding dimerization

DIMERIZATION NLS LIGAND BINDING

AF-1 Activation function ligand independent

AF-2 Activation function ligand-dependent interaction with cofactor

FIGURE 1 Structural and functional organization of nuclear receptors. NRs consist of six domains (A–F) based on regions of conserved sequence and function.

N

H9

H8

H1

H4 H10

H5

H3 H3 H7

H2

H12 9-cisRA

H11 H11

H6

part of the LBD in the peptide connecting helices H1 to H3. The length of this connecting region varies between 72 and 81 residues compared to 15–25 residues for the other nuclear receptors. This insertion region, for which no biological function has been found, is poorly conserved between the VDR species with only 9% identity. One mutation associated with the genetic disease vitamin D–resistant rickets type II has been found in this region (Cys190Trp of hVDR) with no effect on ligand binding [22]. A phosphorylation site also has been identified at Ser208 [23]. Secondary structure prediction reveals that this region is not structured. Among the NR LBD crystal structures solved, it has been reported frequently that the region connecting helices H1 to H3 is poorly ordered and shows little if any secondary structure. In hVDR, this region is sensitive to proteolysis with a trypsin cleavage site at the C terminus of Arg174 [24]. In order to stabilize the overall structure of the hVDR LBD, a VDR LBD mutant has been engineered by removing 50 residues in the region connecting helices H1 to H3 [25,26]. This VDR mutant, hVDR∆ (118-427∆166-216), stabilized the protein by lowering the number of conformations adopted by the insertion region. This hVDR∆ has been fully characterized in solution and compared to the wild-type hVDR.

H12

II. BIOLOGICAL PROPERTIES OF hVDR∆

FIGURE 2 Superimposition of unliganded (green and yellow) and liganded (blue and red) hRXRα LBD monomers. The main conformational differences affect helices H3, H11 and H12. The arrows show the main structural changes upon ligand binding. Adapted from Fig. 4A of Egea PF, Mitschler A, Rochel N, Ruff M, Chambon P, Moras D 2000 Crystal structure of the human RXRalpha ligand-binding domain bound to its natural ligand: 9-cis retinoic acid. EMBO J 19:2592–2601 [48].

The ability of hVDR∆ to bind ligands was determined by Scatchard analysis. This VDR mutant exhibits the same binding affinity for 1α,25(OH)2D3 and 1α,25(OH)2D3 analogs as the hVDR wild type [25,26]. Functional studies of the hVDR∆ and hVDR wild type were performed in order to compare their transactivation properties. In a first system, the hVDR LBD mutant and wild type were fused to the DNA binding domain of the yeast activator Gal4. The chimeric proteins were

CHAPTER 15 Vitamin D Nuclear Receptor Ligand-Binding Domain Crystal Structures

expressed by transient transfection in COS cells and transactivation was measured with a Gal4-responsive reporter. Both proteins exhibit comparable transactivation [25]. In a second system, the full-length hVDR mutant or wild type was cotransfected with a luciferase reporter plasmid containing the osteopontin gene VDRE under the control of the thymidine kinase promoter [26]. Similar results with regard to activation were obtained for the two VDR constructs, confirming the functional integrity of the hVDR∆ protein.

III. SOLUTION STUDIES Like the wild-type receptor, the hVDR∆ behaves as a monomeric species and is able to heterodimerize with RXR. The hVDR∆/RXRs LBDs complexed to 1α,25(OH)2D3 and 9-cis-retinoic acid form a stable species and the two partners are able to bind their ligands as seen by native electrospray ionization mass spectrometry and analytical ultracentrifugation [26]. Low-resolution structural information on the hVDR structure has been obtained by a small-angle X-ray scattering study [26,27]. The scattering function is time and space averaged. The radii of gyration (Fig. 3) of hVDR and hVDR∆ complexed to 1α,25(OH)2D3 were measured and are 26.2 Å and 23.4 Å, respectively. The difference of 2.8 Å, which would correspond to an increase in volume of 35% for a spherical object, is large and would suggest an appended insertion domain weakly connected to the main core LBD. The longest distance in the protein was also estimated in this study. A difference of 15 Å between the Dmax of hVDR and hVDR∆ suggests again that this insertion region may be positioned along the axis collinear to a N- to C-terminal axis including helix H12. This solution study confirms that the insertion domain is mobile and weakly connected to the rest of the LBD, which is itself structured similarly to that of the hVDR wild type.

IV. CRYSTAL STRUCTURE OF hVDR∆ BOUND TO 1α,25(OH)2D3 Deletion of the hinge region insertion in hVDR∆ stabilizes the protein and allowed the crystallization of the hVDR–1α,25(OH)2D3 complex [25]. The crystal structure was solved at 1.8 Å by a combination of molecular replacement using a homology model based on the retinoic acid receptor RARγ [7] and isomorphous replacement with a mercurial derivative. A higher resolution (1.5 Å) data set was collected later [28]. The overall topology of the hVDR∆ LBD (Fig. 4) is that of the canonical LBD with 13 α helices sandwiched in

281

three layers and a three-stranded β sheet. Helices H1 and H3 are connected by two small helices H2 and H3n, H3n replacing the ω loop of the RARγ structure [7]. The truncation of the hVDR∆ construct is positioned just before H3n as shown in Fig. 4. The VDR LBD structure is closely related to that of the holo hRARγ LBD structure [7] with a root mean square deviation (rmsd) of the superimposed structure of 1.2 Å over 179 residues. The connection between helices H1 and H3 follows a path between H3 and the tip of the β-sheet similar to that of the ERα structure [29]. The tip of the β sheet is consequently shifted outward and enlarges the ligand-binding cavity. The β sheet tip is stabilized by hydrogen bonds with residues of the H2-H3n loop. Helix 12 is in the agonist position and stabilized by two interactions with the ligand. Helix 12 is also stabilized by several hydrophobic contacts with residues of H3, H5, and H11, and two polar interactions with residues of H3 and H4 (Fig. 5). Some of these residues contact the ligand, thus indicating an additional indirect ligand control of the position of helix H12. In the hVDR∆ structure, a strong crystalline contact is observed between helix H3n and helices H3, H4, and H12 of a symmetrically related molecule, with H3n mimicking the coactivator SRC-1 peptide contacts [12–14]. Active vitamin D resides in a chair B conformation with the 19-methylene “up” and the 1α-OH and 3β-OH groups in an equatorial and axial orientation, respectively. A chair A conformation of the A-ring would disrupt the hydrogen bonds formed by the hydroxyl and the protein. The conjugated triene system connecting A-ring to C- and D-rings accommodates an almost trans conformation with the C6–C7 bond exhibiting a torsion angle of –149° that deviates significantly from the planar geometry, which results in the curved shape of the ligand bound to the receptor. The deviation explains the lack of biological activity of the analogs with a trans or a cis conformation of the C6–C7 bond [30]. The α face of the C-ring is lined by Trp286, whereas the methyl C18 on the β face points toward Val234 (H3). The ligand-binding pocket is lined by hydrophobic residues (Fig. 6). The elongated ligand embraces helix H3 with its A-ring oriented toward the C termini of helix H5 and the 25-OH close to helices H7 and H11. The 1-OH group forms two hydrogen bonds with Ser237 (H3) and Arg274 (H5), while the 3-OH group is hydrogen-bonded to Ser278 (H5) and Tyr143 (loop H1-H2). The 25-OH moiety forms two hydrogen bonds with His305 (loop H6-H7) and His397 (H11). The triene is tightly fitted in a hydrophobic channel sandwiched between Ser275 (loop H5-β) and Trp286 (β1) on one side and Leu233 (H3) on the other side. The aliphatic chain at position 17 of the D-ring adopts

282

NATACHA ROCHEL AND DINO MORAS

8.5

In I(Q)

A

7.5 6.5 5.5 0

0.002

0.004

Q2 (Å 8

B

9

C

7

8

In I(Q)

In I(Q)

6 5 4 3 2

0.006

−2)

7 6 5 4 3

1 0

0.05

0.1

Q (Å

D

0.15

0.2

0.05

−1)

0.1

Q (Å

E

1 0.9

0.9

0.8

0.8

0.7

0.7

0.6

0.6

0.5

0.2

0.25

0.5

0.4

0.4

0.3

0.3

0.2

0.2

0.1

0.1

0

0.15

−1)

1

P(r)

P(r)

0

0.25

0 0

50

r (Å)

0

50

r (Å)

FIGURE 3 Solution scattering analysis of hVDR∆ (▲) and hVDR wild type (◆) bound to 1α,25(OH)2D3. (A) Guinier analysis. (B) and (C) X-ray scattering curves of hVDR∆ and wild type, respectively. (D) and (E) Distance distribution function P(r). Adapted from Fig. 6 of Rochel N, Tocchini-Valentini G, Egea PF, Juntunen K, Garnier JM, Vihko P, Moras D 2001 Functional and structural characterization of the insertion region in the ligand-binding domain of the vitamin D nuclear receptor. Eur J Biochem 268:971–979 [26].

an extended conformation parallel to the C13–C18 bond with the C13–C17–C20–C22 torsion angle close to 90°; it is surrounded by hydrophobic residues. The ligand binding cavity of hVDR∆ is large (697 Å3) with the ligand occupying only 56% of this volume. A channel of water molecules near position 2 of the A-ring makes an additional space that can accommodate ligands with a methyl group at position 2. The fourfold increase in binding affinity of the 2α−methyl analog is in agreement with this observation [31]. Additional space around the aliphatic chain is also observed.

V. MUTANT ANALYSIS In order to investigate structurally and functionally important amino acid interactions within the ligandbinding pocket of the wild-type hVDR in the presence of several synthetic vitamin D analogs, Vaisanen et al. [32] have combined side-directed alanine mutagenesis with limited proteolytic digestion, electrophoretic mobility shift assay, and reporter gene assay. They have shown that structurally different agonists have distinct ligand–receptor interactions and that residues

283

CHAPTER 15 Vitamin D Nuclear Receptor Ligand-Binding Domain Crystal Structures

H4 H1 H9

H8

K264

H4 H10 H3

H5 E420 H3

H12

H11

S235

H12

H2 H7

D232 H6

T415

L414

Loop6–7

H2

H3n M412

R154

Insertion Domain 220ELTVS216................164GGGND160

FIGURE 4 Overall fold of hVDR∆ ligand-binding domain. The helices are represented as cylinders and β sheets as arrows. The whole structure is colored in gray except the helix H12 in purple. The ligand is depicted in yellow. The insertion region location is shown in green. The reconnected residues are indicated, together with their sequence numbering. Adapted from Fig. 1 of Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179 [25]. (See color plate).

FIGURE 5 Intramolecular interactions of helix H12 in VDR. The backbone of the protein is colored in gray except for helix H12 in purple. The side chains residues involved in polar stabilization of H12 are shown together with the hydrogen bonds (green dots). The ligand is depicted in yellow and red for the oxygen atoms. Adapted from Fig. 3 of Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179 [25]. (See color plate).

F150

F150

S278

O3

R274

Y147

Y143

Y147

Y143

S278

O3

R274

Y293

Y293 S237

S237 O1 L233

S275

I271

W286

I271 M272

V234

V234

V418

S275

O1 L233

W286

I268

V418

L313

M272 L230

I268

V300

L227

L313 V300

L227 F422

F422 H397 O25

H397 O25 L309

L309 Y401

H305

Y401

H305

FIGURE 6 Stereo view of 1α,25(OH)2D3 in the hVDR binding pocket. The ligand molecule is shown in the experimental electron density omit map contoured at 1.0 standard deviation. The hydroxyl groups are depicted as red spheres. Water molecules are shown as purple spheres. Hydrogen bonds are shown as green dotted lines. Carbon atoms are colored in gray, and oxygen and nitrogen atoms are colored in red and blue, respectively. Adapted from Fig. 3 of Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179 [25]. (See color plate).

284 H229, D232, E269, F279, and Y295 are critical for the agonist conformation [32]. Using a two-dimensional alanine scanning mutational analysis, Choi et al. [33] have studied the interactions between VDR and various vitamin D ligands. Eighteen alanine mutants of residues forming the ligand-binding pocket were studied. These residues form either hydrogen bonds with the ligands (Y143, S237, R274, S278, H305, H397) or hydrophobic interactions (D144, L233, V234, I238, I268, I271, S275, C288, W286, V300, Q400, Y401). The transactivation potencies of these mutants with the natural hormone and 11 other ligands were evaluated through functional studies with luciferase reporter assay and a mouse osteopontin response element. The comparison of six mutants of hVDR∆ and wild-type hVDR has shown that both LBPs are similar, thus providing additional evidence to support identical structures of VDR mutant and wild type. The importance of these residues in the interactions with specific ligands was evaluated in this study. Eight residues (Y143, D144, L233, I271, R274, W286, H397, Y401) have been shown to be essential for transactivation by vitamin D ligands, and mutations to less bulky hydrophobic residues increase the potency of 20-epi ligands [33].

VI. CRYSTAL STRUCTURE OF zVDR BOUND TO 1α,25(OH)2D3 In order to validate the conclusions made on the hVDR∆ structure and to find another crystal packing, the VDR from zebrafish (Danio rerio) was also studied. The zVDR exhibits 100% identity in the LBP lining residues and its transactivation potency is 50% of that of hVDR [34]. This activity is consistent with that of Xenopus laevis VDR or lamprey VDR, which show 50% [19] and 25% [21] of the activity of the hVDR, respectively. The zVDR LBD exhibits 69% identity and 79% similarity in its sequence with that of the hVDR LBD, while the insertion region (191–252 of zVDR) exhibits only 34% identity and 47% similarity. The zVDR LBD in complex with 1α,25(OH)2D3 was crystallized in presence of a coactivator peptide that contains the second NR box of SRC-1 [34]. The zVDR LBD (Fig. 7) adopts the canonical active conformation observed for agonist-bound receptors. The SRC-1 peptide (687-696 HKILHRLLQE) forms an amphipatic α-helix interacting with a hydrophobic cleft on the LBD surface. These interactions are similar to those described for other NRs [12–14]. In particular, Glu446 from H12 [Glu420] forms hydrogen bonds with the backbone amide nitrogen of Leu690 and Leu691.

NATACHA ROCHEL AND DINO MORAS

H1

H12

FIGURE 7 Overall fold of zVDR ligand-binding domain. The helices are represented as cylinders and β sheets as arrows. The whole structure is colored in gray and the SRC-1 peptide in dark gray. The ligand 1α,25(OH)2D3 is depicted in yellow.

At the other end, Lys274 from H3 [Lys246] forms a hydrogen bond to the main chain oxygen of Leu694. The structures of zVDR LBD and hVDR∆ LBD complexes with 1α,25(OH)2D3 are similar with a root mean square deviation of 0.72 Å over 236 main-chain atoms. The binding pocket is identical and the natural ligand adopts the same conformation and forms the same interactions with the protein in the two structures (Fig. 8).

H11

FIGURE 8 Superimposition of the LBP of hVDR∆-1α,25(OH)2D3

in green and zVDR-1α,25(OH)2D3 in yellow. The side-chain residues forming the binding pocket are shown together with the hydrogen bonds (dots).

CHAPTER 15 Vitamin D Nuclear Receptor Ligand-Binding Domain Crystal Structures

Notably the channel of water molecules at position 2 of the ligand, which was observed in the hVDR structure, is also observed in the zVDR. The differences observed when superimposing the two structures are small and primarily involve the loops. The ligandbinding pocket is conserved in both sequence and structure. The VDR-specific insertion region between helices H2 and H3 present in the zVDR construct and deleted in the hVDR mutant construct is not visible in the electron density map. This probably reflects the mobility of this region resulting in a static disorder. In the crystal structure of a VDR subfamily member, PXR [35,36], which also contains an insertion region between helices H1 and H3, the loop is also partially unresolved. The structural similarities between zVDR wild-type and hVDR∆ LBDs bound to 1α,25(OH)2D3 validate the overall biological relevance of the mutant hVDR structure.

VII. STRUCTURE OF hVDR COMPLEXED TO SUPERAGONIST LIGANDS Among the several synthetic analogs of vitamin D, the 20-epi compounds [37], which exhibit an inverted stereochemistry at position 20 in the flexible aliphatic chain, have attracted much attention. They are potent growth inhibitors and inducers of cell differentiation, while showing an affinity similar to that of 1α,25(OH)2D3 for VDR [37,38]. KH1060 (1α,25-dihydroxy20epi-22oxa-24,26,27-trihomovitamin D3 [37]), a member of this 20-epi family, exhibits similar properties with decreased calcemic side effects. The 20-epi compounds induce VDR-dependent transcription at concentrations at least 100-fold lower than the natural ligand and provoke antiproliferative activity several orders of magnitude higher than the natural ligand [37–40]. The differences in biological activity of 1α,25(OH)2D3 and the 20-epi molecules in general, and KH1060 in particular, are known to be VDR-LBD dependent. However, they are not yet understood. The ability of 20-epi analogs complexed to VDR to induce transcription appears to correlate with the ability of these compounds to promote coactivator interaction [40]. Differing proteolytic digestion patterns of VDR in the presence of the natural ligand and the 20-epi compounds have been interpreted to reflect large conformational changes in the receptor upon binding of the later class of molecules. In order to investigate the binding mode of the 20-epi analogs to the VDR-LBD, we have determined the high-resolution crystal structures of the hVDR∆ in complex with MC1288 (20-epi-1α,25(OH)2D3 [37]) and KH1060 and compared these structures to that obtained with the natural ligand [28] (Fig. 9).

285

When compared to hVDR∆-1α,25(OH)2D3 complex, the atomic models show a rms deviation on Cα atoms of 0.08 Å and 0.14 Å for hVDR∆-MC1288 complex and hVDR∆−KH1060 complex, respectively. Variations involve only some side chains located at the surface of the protein. In opposition to the view that the 20-epi analogs induce a different agonist conformation, the overall conformation and especially the position of helix H12 are strictly maintained in all three complexes. Furthermore, the ligand-binding cavity is unique and conserved for all three complexes. The rms deviations of all atoms forming the ligand pocket are 0.09 Å for hVDR∆-MC1288 complex and 0.12 Å for the hVDR∆-KH1060 complex. The sizes of the three ligands are 381 Å3, 375 Å3, and 392 Å3 for 1α,25(OH)2D3, MC1288, and KH1060, whereas the volume of the ligand pocket remains unaltered in the three complexes (660 Å3). The ligands occupy only 57% of the volume of the pocket for the VDR1α,25(OH)2D3 and VDR-MC1288 complexes and 59% for the VDR-KH1060 complex. The A, seco-B, and C/D rings form identical contacts as previously described for the 1α,25(OH)2D3-VDR complex. The hydroxyl groups make the same hydrogen bonds: 1-OH with Ser237 and Arg274, 3-OH with Tyr143 and Ser278, and 25-OH with His305 and His397. The deletion of 1-OH or 25-OH leads to the largest changes with a significant decrease in binding (1/1000), while that of the 3-OH has a smaller effect (1/20) [41]. The specific interactions observed in the three ligand–protein complexes involve the hydrophobic contacts of the 17β-aliphatic chains (Fig. 9B). When comparing the natural ligand and MC1288, the main difference is the positioning of the methyl group C21 which results in different contacts with Val300, Leu309, and His397. In MC1288, the C21 moiety is closer to His397. Other protein–ligand contacts differ, but to a lesser extent they involve the methyl groups at positions 23 (His305), 24 (His397), and 27 (His397 and Val418). In these two complexes, the carbon C22 makes no contact at a distance closer than 4.2 Å. In the case of KH1060, the methyl group C21 is quite close to that of the corresponding atom in 1α,25(OH)2D3. Note that the oxygen atom at position 22 of KH1060 forms a van der Waals contact with Val300. The major differences observed between KH1060 and the two other ligands are the tighter and more numerous ligand–protein contacts. The methyl groups C26α and C27α, specific to KH1060, form additional contacts with H3, loop 6–7, H11, and H12. A weak density is observed for the C26α methyl group, suggesting a structural disorder, whereas the C27α methyl group is clearly defined. In the three complexes, the ligands adopt an elongated conformation (Fig. 9C) similar to that described

286

NATACHA ROCHEL AND DINO MORAS

A

A

3-OH 25-OH

B

1-OH

FIGURE 9 Crystal structures of the VDR LBD complexed to 1α,25(OH)2D3, MC1288, and KH1060. (A) Experimental electron density omit map contoured at 2.0 standard deviation of (A) 1α,25(OH)2D3, (B) MC1288, and (C) KH1060. (B) Closeup view of KH1060 in the ligand-binding pocket. Secondary structure features are represented in blue (α-helices) and green (β-strands). The ligand is colored in yellow with the oxygen atoms in red. The volume of the cavity is represented in gray. (C) Superposition of 1α,25(OH)2D3 (yellow), MC1288 (green), and KH1060 (blue) ligands after superimposed VDR complexes. Oxygen atoms are colored in red. Adapted from Fig. 2 and 3 of Tocchini-Valentini G, Rochel N, Wurtz JM, Mitschler A, Moras D 2001 Crystal structures of the vitamin D receptor complexed to superagonist 20-epi ligands. Proc Natl Acad Sci USA 98:5491–5496 [28]. (See color plate).

C

B

C

1-OH

3-OH A D 25-OH C

CHAPTER 15 Vitamin D Nuclear Receptor Ligand-Binding Domain Crystal Structures

in the hVDR∆-1α,25(OH)2D3 structure. In the 20-epi complexes, the aliphatic chain is less constrained, thus allowing alternative conformations of the 17β-chain. The distances between the 25-OH group and the 1-OH and 3-OH are similar for the three complexes. These represent the anchoring points that must be maintained in order to obtain an active conformation. The aliphatic side chain of 1α,25(OH)2D3 adopts an extended conformation parallel to the C13–C18 bond with the C13–C17–C20–C22 torsion angle close to 90°. The two 20-epi analogs use different strategies to fit into the cavity. KH1060 accommodates its longer chain by adopting an eclipsed conformation around the O22–C23 bond, whereas MC1288 adopts a gauche conformation and has to compensate for a chain that is too short to maintain the 25-OH interactions because of the C20 inverted stereochemistry. Our preliminary docking experiments of the 20-epi analogs correctly positioned the methyl C21 in the same cavity as 1α,25(OH)2D3, but failed to give the correct geometry of the aliphatic chains. The crystallographic structures provide the exact information regarding these changes and the details of the interactions with the protein. Compared to the natural ligand, the aliphatic chains of the 20-epi analogs are lining the opposite side of the pocket. The different orientation of the chains is reflected by the different values of the torsion angles C16–C17–C20–C22 and C17–C20–C22(O22 for KH1060)–C23. Farther down the chain, while 1α,25(OH)2D3 adopts an extended conformation, the MC1288 and KH1060 exhibit gauche and eclipsed conformations, respectively, to properly orient the 25 hydroxyl group. The energetically unfavorable eclipsed conformation observed in KH1060 (C20–O22–C23–C24 of KH1060 equal to 16°) is made possible by the strong interactions formed by O22 with C12 (2.9 Å) and C18 (3.24 Å). A methylene group at this position as in MC1301 would be shifted away from the D-ring. A crude modeling analysis suggests that the methylene group would move away from C12, adopting a gauche (+) conformation instead of the eclipsed one observed for KH1060. As a result, additional contacts within the protein pockets are formed (Val300) that may contribute to the higher affinity of MC1301; other contributions such as solvation cannot be excluded. Similarly, the naturally occurring hydroxylation of C24α in KH1060 and C23 in 1α,25(OH)2D3 would form steric clashes with His305, affecting the hydrogen bond network with 25-OH. Based on the crystal structures this activity would depend on the capacity of the pocket to accommodate these additional groups. Both conformations and additional interactions of the 20-epi analogs are predicted to afford higher stability and longer half-life of the active complexes. Indeed,

287

within 3 hr, 60% of 1α,25(OH)2D3 dissociates from the VDR complex, whereas only 5–20% of MC1288 is dissociated [38]. Furthermore, receptors lacking the C-terminal helix 12 exhibit different dissociation behavior for the 20-epi analogs that are still capable of interacting with a mutated receptor [38,39]. Limited proteolytic digestion studies show that the 20-epi analog–receptor complexes are more resistant to digestion, suggesting that they are more stable. Additionally, an important role might be played by the rate of assimilation, i.e., how synthetic agonists are metabolized as compared to the natural ligand in different cells types [42,43].

VIII. STRUCTURE OF zVDR IN COMPLEX WITH GEMINI Among more than 3000 synthetic analogs of 1α,25(OH)2D3 synthesized to increase the potency and specificity of the physiological effects of vitamin D, Gemini (1α,25-dihydroxy-21-(3-hydroxy-3-methylbutyl)vitamin D3) is an interesting molecule with two identical side chains branching at carbon 20. Although Gemini binds less efficiently to VDR (38% compared to 1α,25(OH)2D3 [44]), its transactivation potency is similar to that of 1α,25(OH)2D3 in ROS cells [44] and 10-fold higher in HeLa and COS-7 cells [45]. It has been shown that in presence of an excess of co-repressor, the VDR–Gemini complex shifts from an agonist to an inverse agonist conformation through the recruitment of N-Cor and mediates repression [45,46]. Its stereochemistry and its 25% increase in volume when compared to the natural ligand create a major docking problem using the original LBP as a template. The packing constraints of the crystal form obtained for the hVDR∆ discriminate complexes with even small conformational changes near the ligand-binding pocket. In order to overcome this problem and obtain crystals of complexes that could not be crystallized with hVDR∆, we used the zVDR LBD construct. The zVDR in complex with Gemini and a coactivator peptide was crystallized and the structure solved at resolution 2.6 Å [34]. At the backbone level, the overall structure of zVDR-Gemini is almost identical to that of zVDR1α,25(OH)2D3 with a rms deviation of 0.37 Å over 249 main-chain atoms. However, the binding of the ligand’s second side chain results in a significant adaptation of the binding pocket that would have been difficult to predict [47]. A new pocket is created by the combined effect of a backbone shift and a side-chain conformational reorientation. The positions of helices H11 and H12 are unchanged and the peptide coactivator makes the same interactions with the protein, underlying the similar agonist character of both structures.

288 The most striking effect that emerges from this study is the “formation” of a new channel that extends the original pocket. This ligand-dependent pocket emphasizes the adaptability of the LBD and confirms the induced-fit mechanism of ligand binding. Despite large discrepancies in the shape and size of the ligand pocket, the overall structure of VDR and more importantly the position of H12 and of the coactivator peptide remain unaltered, consistent with the agonist character of Gemini. Thus the molecular mechanism of the agonism of Gemini is probably identical to that of 1α,25(OH)2D3.

IX. CONCLUSION The crystal structures of VDR LBD explain most features of ligand binding. These structures show the adaptability of the ligands and of the pocket in some specific cases. In all experimentally determined structures, the ligands were agonists and the protein conformation, notably H12, is conserved. The available structural information does not permit an understanding of the functional and structural role of the insertion domain. The loop 2-3, truncated in the hVDR and only partially visible in the zVDR structures, apparently does not influence the function of this nuclear receptor. Since this unique feature of VDR has not been lost during evolution, the question of its function remains an interesting one. However, the next step of structural studies should address the more general problem of the full-length receptor structure in different functional states.

Acknowledgment This work was supported by grants from CNRS, INSERM, Ministére de la Recherche et de la Technologie and by the Genopole and SPINE programs.

References 1. Moras D, Gronemeyer H 1998 The nuclear receptor ligandbinding domain: structure and function. Curr Opin Cell Biol 10:384–391. 2. Chambon P 1996 A decade of molecular biology of retinoic acid receptors. FASEB J 10:940–954. 3. McKenna NJ, Lanz RB, O’Malley BW 1999 Nuclear receptor coregulators: cellular and molecular biology. Endocr Rev 20:321–344. 4. Glass CK, Rosenfeld MG 2000 The coregulator exchange in transcriptional functions of nuclear receptors. Genes Dev 14:121–141.

NATACHA ROCHEL AND DINO MORAS

5. Freedman LP 1999 Increasing the complexity of coactivation in nuclear receptor signaling. Cell 97:5–8. 6. Bourguet W, Ruff M, Chambon P, Gronemeyer H, Moras D 1995 Crystal structure of the ligand-binding domain of the human nuclear receptor RXR-alpha. Nature 375:377–382. 7. Renaud JP, Rochel N, Ruff M, Vivat V, Chambon P, Gronemeyer H, Moras D 1995 Crystal structure of the RARgamma ligand-binding domain bound to all-trans retinoic acid. Nature 378:681–689. 8. Wagner RL, Apriletti JW, McGrath ME, West BL, Baxter JD, Fletterick RJ 1995 A structural role for hormone in the thyroid hormone receptor. Nature 378:690–697. 9. Billas IM, Iwema T, Garnier JM, Mitschler A, Rochel N, Moras D 2003 Nature 426:91–96. 10. Svensson S, Ostberg T, Jacobsson M, Norstrom C, Stefansson K, Hallen D, Johansson IC, Zachrisson K, Ogg D, Jendeberg L 2003 EMBO J. 22:4625–4633. 11. Wurtz JM, Bourguet W, Renaud JP, Vivat V, Chambon P, Moras D, Gronemeyer H 1996 A canonical structure for the ligand-binding domain of nuclear receptors. Nat Struc Biol 3:87–94. 12. Darimont BD, Wagner RL, Apriletti JW, Stallcup MR, Kushner PJ, Baxter JD, Fletterick RJ, Yamamoto KR 1998 Structure and specificity of nuclear receptor-coactivator interactions. Genes Dev 12:3343–3356. 13. Nolte RT, Wisely GB, Westin S, Cobb JE, Lambert MH, Kurokawa R, Rosenfeld MG, Willson TM, Glass CK, Milburn MV 1998 Ligand binding and co-activator assembly of the peroxisome proliferator-activated receptor-gamma. Nature 395:137–143. 14. Shiau AK, Barstad D, Loria PM, Cheng L, Kushner PJ, Agard DA, Greene GL 1998 The structural basis of estrogen receptor/coactivator recognition and the antagonism of this interaction by tamoxifen. Cell 95:927–937. 15. Baker AR, McDonnell DP, Hughes M, Crisp TM, Mangelsdorf DJ, Haussler MR, Pike JW, Shine J, O’Malley BW 1988 Cloning and expression of full-length cDNA encoding human vitamin D receptor. Proc Natl Acad Sci USA 85: 3294–3298. 16. Burmester JK, Maeda N, DeLuca HF 1988 Isolation and expression of rat 1,25-dihydroxyvitamin D3 receptor cDNA. Proc Natl Acad Sci USA 85:9499–9502. 17. Kamei Y, Kawada T, Fukuwatari T, Ono T, Kato S, Sugimoto E 1995 Cloning and sequencing of the gene encoding the mouse vitamin D receptor. Gene 152:281–282. 18. Lu Z, Hanson K, DeLuca HF 1997 Cloning and origin of the two forms of chicken vitamin D receptor. Arch Biochem Biophys 339:99–106. 19. Li YC, Bergwitz C, Juppner H, Demay MB 1997. Cloning and characterization of the vitamin D receptor from Xenopus laevis. Endocrinology 138:2347–2353. 20. Suzuki T, Suzuki N, Srivastava AS, Kurokawa T 2000 Identification of cDNAs encoding two subtypes of vitamin D receptor in flounder, Paralichthys olivaceus. Biochem Biophys Res Commun 270:40–45. 21. Whitfield GK, Dang HT, Schluter SF, Bernstein RM, Bunag T, Manzon LA, Hsieh G, Dominguez CE, Youson JH, Haussler MR, Marchalonis JJ 2003 Cloning of a functional vitamin D receptor from the lamprey (Petromyzon marinus), an ancient vertebrate lacking a calcified skeleton and teeth. Endocrinology 144:2704–2716. 22. Malloy PJ, Pike JW, Feldman D 1999 The vitamin D receptor and the syndrome of hereditary 1,25-dihydroxyvitamin D-resistant rickets. Endocrinol Rev 20:156–188.

CHAPTER 15 Vitamin D Nuclear Receptor Ligand-Binding Domain Crystal Structures

23. Jurutka PW, Hsieh JC, Nakajima S, Haussler CA, Whitfield GK, Haussler MR 1996 Human vitamin D receptor phosphorylation by casein kinase II at Ser-208 potentiates transcriptional activation. Proc Natl Acad Sci USA 93:3519–24. 24. Vaisanen S, Juntunen K, Itkonen A, Vihko P, Maenpaa PH 1997 Conformational studies of human vitamin-D receptor by antipeptide antibodies, partial proteolytic digestion and ligand binding. Eur J Biochem 248:156–162. 25. Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179. 26. Rochel N, Tocchini-Valentini G, Egea PF, Juntunen K, Garnier JM, Vihko P, Moras D 2001 Functional and structural characterization of the insertion region in the ligand-binding domain of the vitamin D nuclear receptor. Eur J Biochem 268:971–979. 27. Juntunen K, Rochel N, Moras D, Vihko P 1999 Large-scale expression and purification of the human vitamin D receptor and its ligand-binding domain for structural studies. Biochem J 344:297–303. 28. Tocchini-Valentini G, Rochel N, Wurtz JM, Mitschler A, Moras D 2001 Crystal structures of the vitamin D receptor complexed to superagonist 20-epi ligands. Proc Natl Acad Sci USA 98:5491–5496. 29. Brzozowski AM, Pike AC, Dauter Z, Hubbard RE, Bonn T, Engstrom O, Ohman L, Greene GL, Gustafsson JA, Carlquist M 1997 Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 389:753–758. 30. Norman AW, Okamura WH, Hammond MW, Bishop JE, Dormanen MC, Bouillon R, van Baelen H, Ridall AL, Daane E, Khoury R, Farach-Carson MC 1997 Comparison of 6-s-cisand 6-s-trans-locked analogs of 1α,25-dihydroxyvitamin D3 indicates that the 6-s-cis conformation is preferred for rapid nongenomic biological responses and that neither 6-s-cis- nor 6-s-trans-locked analogs are preferred for genomic biological responses. Mol Endocrinol 11:1518–1531. 31. Fujishima T, Liu Z, Miura D, Chokki M, Ishizuka S, Konno K, Takayama H 1998 Synthesis and biological activity of 2-methyl-20-epi analogues of 1α,25-dihydroxyvitamin D3. Bioorg Med Chem Lett 8:2145–2148. 32. Vaisanen S, Ryhanen S, Saarela JT, Perakyla M, Andersin T, Maenpaa PH 2002 Structurally and functionally important amino acids of the agonistic conformation of the human vitamin D receptor. Mol Pharmacol 62:788–794. 33. Choi M, Yamamoto K, Itoh T, Makishima M, Mangelsdorf DJ, Moras D, DeLuca HF, Yamada S 2003 Interaction between vitamin D receptor and vitamin D ligands. Two-dimensional alanine scanning mutational analysis. Chem Biol 10:261–270. 34. Ciesielski F, Rochel N, Mitschler A, Kouzmenko A, Moras D 2004 J Steroid Biochem Mol Biol. 89–90:55–59. 35. Watkins RE, Wisely GB, Moore LB, Collins JL, Lambert MH, Williams SP, Willson TM, Kliewer SA, Redinbo MR 2001 The human nuclear xenobiotic receptor PXR: structural determinants of directed promiscuity. Science 292:2329–2333.

289

36. Watkins RE, Maglich JM, Moore LB, Wisely GB, Noble SM, Davis-Searles PR, Lambert MH, Kliewer SA, Redinbo MR 2003 2.1 A crystal structure of human PXR in complex with the St. John’s wort compound hyperforin. Biochemistry 42: 1430–1438. 37. Binderup L, Latini S, Binderup E, Bretting C, Calverley M, Hansen K 1991 20-Epi-vitamin D3 analogues: A novel class of potent regulators of cell growth and immune responses. Biochem Pharmacol 42:1569–1575. 38. Peleg S, Sastry M, Collins ED, Bishop JE, Norman AW 1995 Distinct conformational changes induced by 20-epi analogues of 1α,25-dihydroxyvitamin D3 are associated with enhanced activation of the vitamin D receptor. J Biol Chem 270: 10551–10558. 39. Liu YY, Collins ED, Norman AW, Peleg S 1997 Differential interaction of 1α,25-dihydroxyvitamin D3 analogues and their 20-epi homologues with the vitamin D receptor. J Biol Chem 272:3336–3345. 40. Yang W, Freedman LP 1999 20-Epi analogues of 1,25dihydroxyvitamin D3 are highly potent inducers of DRIP coactivator complex binding to the vitamin D3 receptor. J Biol Chem 274:16838–16845. 41. Bouillon R, Okamura WH, Norman AW 1995 Structurefunction relationships in the vitamin D endocrine system. Endocr Rev 16:200–257. 42. Dilworth FJ, Williams GR, Kissmeyer AM, Nielsen JL, Binderup E, Calverley MJ, Makin HL, Jones G 1997 The vitamin D analog, KH1060, is rapidly degraded both in vivo and in vitro via several pathways: principal metabolites generated retain significant biological activity. Endocrinology 138: 5485–5496. 43. Shankar VN, Dilworth FJ, Makin HL, Schroeder NJ, Trafford DJ, Kissmeyer AM, Calverley MJ, Binderup E, Jones G 1997 Metabolism of the vitamin D analog EB1089 by cultured human cells: redirection of hydroxylation site to distal carbons of the side chain. Biochem Pharmacol 53:783–793. 44. Norman AW, Manchand PS, Uskokovic MR, Okamura WH, Takeuchi JA, Bishop JE, Hisatake JI, Koeffler HP, Peleg S 2000 Characterization of a novel analogue of 1α,25(OH)2vitamin D3 with two side chains: interaction with its nuclear receptor and cellular actions. J Med Chem 43:2719–2730. 45. Bury Y, Herdick M, Uskokovic MR, Carlberg C 2001 Gene regulatory potential of 1α,25-dihydroxyvitamin D3 analogues with two side chains. J Cell Biochem 81:179–190. 46. Carlberg C 2003 Molecular basis of the selective activity of vitamin D analogues. J Cell Biochem 88:274–281. 47. Väisänen S, Peräkylä M, Kärkkäinen JI, Uskokovic R, Carlberg C 2003 Structural evaluation of the agonistic action of a vitamin D analog with two side chains binding to the nuclear vitamin D receptor. Mol Pharmacol 63:1230–1237. 48. Egea PF, Mitschler A, Rochel N, Ruff M, Chambon P, Moras D 2000 Crystal structure of the human RXRα ligand-binding domain bound to its natural ligand: 9-cis retinoic acid. EMBO J 19:2592–2601.

CHAPTER 16

Comodulators of Vitamin D Receptor–Mediated Gene Expression DIANE R. DOWD, AMELIA L. M. SUTTON, CHI ZHANG, AND PAUL N. MACDONALD Department of Pharmacology, Case Western Reserve University, Cleveland, Ohio

I. Introduction II. Coactivators III. Co-repressors

IV. Conclusion—Integrated Model of Comodulator Activity References

I. INTRODUCTION

with vitamin D response elements (VDREs) in the promoters of target genes. VDREs are characterized by two direct hexameric repeats with a spacer of three nucleotides (DR-3 elements) (Chapter 14). Although purified recombinant VDR interacts weakly with DR-3 elements, nuclear extracts containing one or more accessory factors were shown to dramatically enhance the binding affinity of the VDR to these VDREs [8,9]. Subsequent studies revealed that the nuclear accessory factor was the retinoid X receptor, or RXR (Fig. 1) [10,11]. RXR is currently well accepted as the biologically relevant heterodimeric partner for the VDR and other class II nuclear receptors (NRs) (reviewed in [12]). The association of 1,25(OH)2D3 with VDR promotes both the heterodimerization with RXR and high-affinity binding to DR-3 VDREs, with the VDR occupying the 3′ half-site and the RXR occupying the 5′ halfsite [13,14]. In this manner, the interaction of the liganded VDR/RXR with a VDRE confers target gene selectivity and ultimately influences the rate of RNA polymerase II (RNA Pol II)–directed transcription (Chapter 14). About 15 years ago, a phenomenon termed “squelching” was observed in which the ligand binding domain (LBD) of one NR interfered with the transcriptional activation mediated by a second NR [15,16]. The theory was that there are limiting quantities of accessory factors or adapter proteins that interact with the receptor’s LBD and are necessary for NR-mediated transcription. Thus, the LBD of one NR competes with the other intact liganded receptor for binding to these proteins. Indeed, it is becoming increasingly clear that additional protein–protein interactions are required to regulate 1,25(OH)2D3-dependent, VDR/RXR-mediated transcription. The regulation of VDR-mediated transcription likely involves a complex series of macromolecular interactions occurring in a temporally coordinated fashion. Association between the liganded

Nearly a century has passed since the discovery of fat-soluble vitamin D as a micronutrient that is essential to maintain appropriate bone mineralization. In 1970, the bioactive form of vitamin D was isolated and identified as 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) [1–3]. Its main physiological role is to promote the intestinal absorption of dietary calcium and phosphate. Thus, 1,25(OH)2D3 ensures that the serum concentration of these ions is optimal for normal mineralization of the collagen matrix. Moreover, 1,25(OH)2D3 may directly affect bone remodeling by causing osteoblasts to terminally differentiate into osteocytes and deposit calcified matrix [4]. 1,25(OH)2D3 also promotes the differentiation of precursor cells into mature osteoclasts, which function to resorb bone and maintain appropriate bone remodeling [5]. In addition to its traditional role in maintaining calcium and phosphate homeostasis, the vitamin D endocrine system is also involved in a number of other physiological processes including blood pressure regulation, immune function, mammary gland development, and hair follicle cycling (reviewed in [6]). 1,25(OH)2D3 is generated by two sequential hydroxylations of vitamin D3 (cholecalciferol), a secosteroid precursor that is obtained in the diet or produced in the skin upon exposure to UV light (Chapter 2). 1,25(OH)2D3 is transported in the serum bound to the serum vitamin D binding protein (Chapters 8 and 9). 1,25(OH)2D3 dissociates from this transport protein and, because of its lipophilic nature, is thought to enter the cell by passive diffusion. Once inside the cell, 1,25(OH)2D3 is bound selectively by the vitamin D receptor or VDR (Chapter 11). VDR was first cloned in 1987 and was found to be a member of the superfamily of nuclear receptors (NRs) that regulates gene expression in a 1,25(OH)2D3-dependent manner [7]. The binding of 1,25(OH)2D3 to VDR enhances the association of VDR VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

292

DIANE R. DOWD, AMELIA L. M. SUTTON, CHI ZHANG, AND PAUL N. MACDONALD

RXR

VDR

D

D RXR VDR

VDRE

D RXR VDR

VDRE

FIGURE 1 1,25(OH)2D3-induced heterodimerization between VDR and RXR. VDR binds to its ligand, 1,25(OH)2D3, heterodimerizes with RXR, and binds to a DR-3 type VDRE in the promoter of 1,25(OH)2D3-responsive genes. VDRs occupy the proximal or 3′ half-site while RXR occupies the distal or 5′ half-site.

VDR/RXR heterodimer and other transcriptional components may be classified into two general categories: general transcription factors and the comodulatory proteins. The interaction of VDR with the first group results in direct contacts with the preinitiation complex (PIC), which may facilitate assembly or recruitment of the PIC and thereby stimulate transcription by RNA Pol II. TFIIB and TATA binding-protein-associated factors (TAFs) are examples of this class of transcription factor, which modulates the activity of liganded VDR. In addition to direct contacts with the general transcription machinery, the liganded VDR is also linked to the transcriptional PIC by the NR comodulatory factors. NR comodulators are proteins that interact directly with NRs and modulate, either positively or negatively, their abilities to regulate their transcriptional activity. NR comodulators are classified either as coactivators or co-repressors, and they aid in the induction or repression, respectively, of ligand/receptormediated transcription. Coactivator proteins function to augment transcription by one of three proposed mechanisms. First, coactivators may function as macromolecular bridges between the liganded receptor and the general transcriptional machinery. In this capacity, they may recruit components of the PIC, aid in the

assembly of the PIC, or promote the stability of the complex. Second, some coactivators possess histone acetyltransferase (HAT) activity or recruit HAT activity to the promoter, which may loosen chromatin packaging, thereby making it more accessible to the basal transcriptional machinery. A third possible mechanism of coactivator function is to increase the rate of coupling between RNA Pol II–directed transcription and more distal events such as transcription elongation and RNA processing. For example, several coactivator proteins express domains that are consistent with RNA processing proteins, thus pointing to a role in both activated transcription and RNA splicing mechanisms (Section II, E). Thus, these multifunctional proteins may be able to interact with the nuclear receptor, general transcription factors, chromatin remodeling activities, and the splicing machinery to couple transcription to RNA processing and dramatically influence the rate at which a gene product is expressed. NR co-repressors are generally defined as proteins that interact with unliganded receptors to repress basal expression of hormone-responsive genes. They are distinguished from other general transcriptional repressors that interfere with NRs through distinct mechanisms such as binding to the receptor or to DNA to disrupt NR response-element binding. In this chapter, NR co-repressors are discussed as nuclear factors that interact with the unliganded NRs and directly modify histones or recruit modifying enzymes, such as histone deacetylases (HDACs), to maintain chromatin in a tightly packaged state and silence transcription from the promoter in an NR-dependent manner. Research spanning the past two decades reveals that 1,25(OH)2D3-mediated transcription is more complex than the simple binding of the receptor to DNA and the recruitment of RNA Pol II to initiate transcription. VDR/RXR-activated transcription involves complex interactions that may occur in a spatially distinct and temporally coordinated fashion to increase the rate at which 1,25(OH)2D3-responsive genes are transcribed and at which the resulting RNA transcript is processed. An integrated model of coactivator/co-repressor function (Fig. 2) proposes that type II NRs maintain a transcriptionally inactive state at a promoter by recruiting co-repressors and their associated HDAC activity. Upon binding ligand, the co-repressors either dissociate from the receptor or are displaced from the complex by the entrance of coactivator molecules. These coactivators then acetylate core histones while recruiting other coactivators and transcription factors, thereby creating a transcriptionally permissive environment at the promoter. At later stages, they may also promote the recruitment or stability of the RNA splicing machinery to enhance the rate at which mature RNAs are made and subsequently translated. The focus of this chapter

CHAPTER 16 Comodulators of Vitamin D Receptor–Mediated Gene Expression

cor

epr

ess

RXR VDR

ors

Ac

VDRE TATA

D

D RXR

coa

ctiv ato

rs

VDR RNA Pol II

VDRE

TBP TATA Ac Ac Ac Ac

FIGURE 2

Model of comodulator activity on VDR-mediated gene expression. The VDR/RXR heterodimer is loosely bound to a VDRE in the absence of ligand. In this state, it may interact with co-repressors, which function to keep gene transcription repressed, in part, by keeping histone proteins deacetylated. Upon binding ligand, the co-repressor is released and replaced by coactivators. The coactivators then function to remodel the chromatin and aid in the recruitment of RNA Pol II and other key components of the preinitiation complex.

is on the roles of the comodulator proteins of vitamin Ddependent transcription in the induction and repression of 1,25(OH)2D3-regulated gene expression.

II. COACTIVATORS The binding of 1,25(OH)2D3 to the VDR initiates an orchestrated cascade of protein assembly ultimately leading to transcriptional activation of select target genes. Ligand-induced coactivator recruitment to the VDR/RXR heterodimer is one of the early events in this transactivation process. These initial interactions between VDR and coactivators are the seed for the assembly of intricate multiprotein complexes that remodel the chromatin structure, recruit the core transcriptional machinery, and induce expression of 1,25(OH)2D3-regulated genes. Thus, understanding the molecular details of 1,25(OH)2D3-induced assembly of VDR and various coactivator complexes is central to the process of 1,25(OH)2D3/VDR-activated transcription.

A. Mechanism of Interaction VDR and other NRs exhibit a modular structure with three principle domains [17]. The amino terminus

293

of most NRs contains a transactivation function, although for VDR, this region is small and its function is poorly understood. Adjacent to this domain is the DNA-binding domain, which contains nine strictly conserved cysteine residues forming two zinc-binding modules that mediate high-affinity and highly selective binding to DNA. The large carboxyl-terminal ligandbinding domain (LBD) of VDR is organized into 12 alpha helices [18]. The LBD mediates association of VDR not only with 1,25(OH)2D3 [17], but also with its heterodimeric partner RXR [10,11,19] and comodulatory proteins [20]. Helix 12 of the VDR LBD contains the ligand-dependent activation function-2 (AF-2), which is essential for transactivation mediated by the VDR [19–21]. The molecular mechanisms involved in AF-2-dependent transactivation mediated by VDR are becoming increasingly clear based on structural analysis of VDR and other NRs. As 1,25(OH)2D3 nestles into the ligand-binding cavity of VDR, the AF-2 helix undergoes a subtle yet important conformational change. Helix 12 folds over the LBD [22] and, together with helices 3, 4, and 5, creates a hydrophobic crevice that selectively interacts with a complementary leucine-rich hydrophobic domain on many nuclear receptor coactivators (Fig. 3; [23]). This domain in the coactivators is termed the NR box and is composed of the consensus core sequence LXXLL [24]. The NR box forms an amphipathic α-helix whose orientation in the hydrophobic crevice of VDR is probably stabilized by a “charge clamp” composed of two conserved charged residues, one in the AF-2 helix 12 (E420) and the other in helix 3 of the VDR (K246) [25]. Besides the three highly conserved leucines, other residues in the NR box are also important for high-affinity binding and selectivity of coactivators for individual nuclear receptors [26–28]. While many coactivators interact with NRs via LXXLL motifs, others do not. Those coactivators that lack LXXLL motifs appear to interact with VDR and other nuclear receptors in an AF-2-independent manner (discussed in Section II,D). Thus, it is apparent that NRs can bind multiple coactivators using distinct motifs.

B. SRC Family of Coactivators The first nuclear receptor coactivator identified was steroid receptor coactivator-1 (SRC-1) [29]. SRC-1 was initially identified as a progesterone receptor–interacting protein that selectively enhanced hormone-dependent transcription. Subsequent studies have shown that SRC-1 is important for the transactivation of nearly all nuclear receptors including VDR [20,30,31]. SRC-1 constitutes the founding member of the p160 or SRC family, which also includes GRIP-1/TIF2/SRC-2 [32,33] and p/CIP/RAC3/ACTR/AIB-1/TRAM-1/SRC-3 [34–38].

294

DIANE R. DOWD, AMELIA L. M. SUTTON, CHI ZHANG, AND PAUL N. MACDONALD

FIGURE 3 The vitamin D receptor undergoes a conformational change upon binding hormone. The VDR binds ligand, and this induces a conformational change in helix 12 of the AF-2 domain. Helix 12 folds to trap the ligand in the binding pocket. This change also creates a hydrophobic cleft or surface on VDR composed of helices H3, H4, H5, and H12 that LXXLL motifs in coactivator proteins use for docking to the VDR.

SRCs are composed of several domains that are highly conserved among the coactivator family (Fig. 4). A basic helix–loop–helix domain and a PAS domain exists in the amino terminus of SRCs. The PAS domain is characteristic of the Per/Arnt/Sim family of transcription

factors. This amino-terminal region exhibits intrinsic transcriptional activity when tethered to a heterologous DNA-binding domain and is thought to mediate protein–protein interactions with other transcription factors, perhaps including other PAS proteins [39]. Three LXXLL-containing NR boxes are located in the mid-region of the SRCs and these mediate the liganddependent interaction with NRs [24]. Although these three NR boxes were first presumed to be functionally redundant, subsequent mutational studies showed that the individual LXXLL motifs and surrounding sequences confer some degree of receptor selectivity [27,40,41]. In the case of SRC-3, NR box III appears to be most important for interaction with VDR [42]. The carboxyl termini of SRCs contain a second autonomous transactivation domain [39]. This domain is characterized by a glutamine-rich region common to many transcriptional activators. The large carboxylterminal region of SRC-1 also mediates interactions with the transcription factors CREB-binding protein (CBP)/p300 [27,43], p300/CBP-associated factor (p/CAF) [44], and coactivator-associated arginine methyltransferase 1 (CARM1) [45]. These interactions involve additional LXXLL motifs distinct from the NR boxes described above. The precise mechanisms of transcriptional activation by SRCs are not entirely clear, but most studies point to the central role of SRCs in modifying the histone acetylation state and chromatin structure of hormoneresponsive promoters [46–48]. Once targeted to the appropriate promoter through interactions with NRs, SRCs are thought to remodel the chromatin, creating a template that is more accessible to the transcriptional machinery. This remodeling occurs through the covalent addition of acetyl groups onto the carboxyl-terminal lysine residues of histones. This acetyl modification weakens the electrostatic interaction between the positively charged histone tails and the negatively charged phosphate backbone of the DNA, thus inducing decondensation of the chromatin (reviewed in [49]). Such HAT activity is contained both in the SRCs themselves [36,47] and in several of the other transcriptional proteins that SRCs recruit, such as p300/CBP [50,51] and p/CAF [52]. In addition, p300/CBP binds directly to nuclear receptors and, together with SRC coactivators, they synergistically stimulate transcription [53,54]. Furthermore, SRCs also promote histone methylation, and presumably loosening of the chromatin structure, through their interactions with CARM1 [45]. Murine knockout models of each of the SRCs have been generated [55–57]. These models show a degree of functional redundancy in the SRC family, especially in the development and maintenance of the female reproductive system. However, the individual SRCs

CHAPTER 16 Comodulators of Vitamin D Receptor–Mediated Gene Expression

295

FIGURE 4

Schematic of the conserved SRC coactivator domains. bHLH, basic helix–loop–helix domain; PAS, region characteristic of the Per/Arnt/Sim family of transcription factors; NR boxes, nuclear receptor interaction boxes containing the LXXLL motifs; CBP interaction, region that interacts with CREB-binding protein; Q-rich, region rich in glutamine residues.

may also have distinct physiological functions perhaps related to nuclear hormone receptor selectivity. For example, deletion of SRC-2 results in aberrant spermatogenesis and testicular defects, whereas deletion of either SRC-1 or SRC-3 has no effect on the male reproductive tract [56]. Mutation of SRC-3 causes decreased estrogen-mediated protection against vascular damage [57]. Since the impact of these genetic deletions on VDR/ 1,25(OH)2D3-targeted systems has not been investigated, it is unknown which SRCs are important in vivo in VDR-mediated transcription. However, in vitro evidence suggests that SRC-2 may be a favored coactivator for VDR since SRC-2 is expressed in 1,25(OH)2D3 target cells such as osteoblasts and preferentially augments 1,25(OH)2D3-induced expression of the osteocalcin gene [58].

C. Mediator-D Complex In addition to SRCs, a large multiprotein complex called DRIP (vitamin D receptor interacting proteins) is a coactivator for VDR and other nuclear receptors [59,60]. DRIP shares many components with the transcriptional coactivator complexes thyroid hormone receptor (TR) activating proteins, activator recruited cofactor, and the mammalian Mediator complex [61–63]. In fact, these four complexes are most probably one in the same. These complexes interact with a variety of mammalian transcriptional activators, most notably NRs, SREBP-1a, and E1A. Thus, they likely have fundamental roles in activator-induced transcriptional processes well beyond the VDR and other NRs. Because of the considerable similarity between proteins comprising these various complexes, a unified nomenclature was proposed that

uses the Mediator complex as the basis for naming the multiple complexes and subunits [64]. Using this nomenclature, DRIP is referred to as Mediator-D. Mediator-D is proposed to interact with the liganded VDR bound to its enhancer element and then to recruit RNA Pol II, thereby acting as a bridge between VDR and the PIC [64,65]. Although mediator complexes may recruit RNA Pol II to the promoter, the polymerase is not tightly bound, thereby allowing for its release and the efficient initiation of transcription. Moreover, yeast mediator does not interact with the hyperphosphorylated form of RNA Pol II involved in transcriptional elongation [66]. These data suggest that mediator complexes are associated with RNA Pol II only during preinitiation and the transition to elongation. Moreover, in vitro transcription studies indicate that following the release of RNA Pol II, mediator remains bound to the promoter to function in the reinitiation of a second round of transcription [67]. Chromatin immunoprecipitation studies on hormone-responsive promoters allude to an even more dynamic cycling process of unknown function [68]. While the Mediator-D complex efficiently stimulates VDR-mediated transcription in vitro on chromatinized templates, Mediator-D and other mammalian mediator complexes do not contain detectable HAT activity [60]. However, recent evidence from yeast mediator complexes suggests they may assist in maintaining chromatin in a hyperacetylated, open conformation [69]. This raises the question of when and how the recruitment of chromatin remodeling complexes and mediator complexes takes place. A model that addresses this question proposes the sequential recruitment of coactivators and is described in more detail later (Section IV).

296

DIANE R. DOWD, AMELIA L. M. SUTTON, CHI ZHANG, AND PAUL N. MACDONALD

10 36 240 17 cdk8 95 130 230 33 97 150 220 78 34 70 100 D VDR

FIGURE 5 Model of the Mediator-D complex interacting with the liganded VDR. Med-220 uses an LXXLL NR box motif for docking to the VDR.

The Mediator-D complex is composed of at least 10 different proteins anchored by Med220, which interacts directly with ligand-activated VDR/RXR heterodimers through the second of two LXXLL motifs (Fig. 5, [70]). While Med220 promotes interaction with the VDR, another component of Mediator-D, Med130, may stimulate assembly of the PIC on the promoter [71]. The interaction between VDR and Med220 is enhanced by phosphorylation of VDR, and this correlates with an increase in 1,25(OH)2D3-mediated transcription [72]. Further support for the importance of this interaction comes from studies from Yang and Freedman [73] using 20-epi analogs of 1,25(OH)2D3. They showed a link between the increased transcriptional activity of these higher potency 20-epi compounds and increased binding of VDR to Med220 when compared to the effects of the 1,25(OH)2D3 ligand. This enhanced binding and transcriptional activity was also reflected in an increase in cellular differentiation and a decrease in proliferation. No differences were observed between 1,25(OH)2D3 and the 20-epi analogs when comparing the ability of these ligands to promote binding of VDR to SRC-2, thus highlighting the importance of mediator complexes in agonist-induced VDR activity. Genetic ablation of Med220 in mice causes embryonic lethality [74]. Moreover, fibroblasts derived from these animals exhibit attenuated thyroid hormone– stimulated transcription, but these cells retain normal retinoic acid responses, suggesting both receptor selectivity and a functional redundancy of NR coactivators in vivo [74]. Heterozygous mutants display growth retardation, impaired transcription and hypothyroidism, underscoring the primary importance of this mediator complex in TR-regulated processes and possibly other NR-mediated transcriptional events. Since VDRmediated transcriptional processes have not been examined in this model, it is unclear whether Med220 is required for 1,25(OH)2D3 action in vivo.

D. NCoA62/SKIP NCoA62 is a VDR and NR coactivator that was isolated as a VDR-interacting protein by a yeast two-hybrid screen [75]. Expression of NCoA62 in mammalian cells enhances vitamin D-, retinoic acid-, estrogen-, and glucocorticoid-activated transcription. NCoA62 is structurally and mechanistically distinct from SRCs and Mediator-D. However, it is highly related to Bx42, a Drosophila melanogaster nuclear protein putatively involved in ecdysone-stimulated transcription [76]. NCoA62 was independently identified as a factor that interacts with the Ski oncoprotein and was termed Ski-interacting protein, or SKIP [77]. Thus, it is referred to as NCoA62/SKIP throughout this chapter. NCoA62/SKIP has subsequently been shown to interact with a diverse array of transcription regulatory factors including CBF1 and silencing mediator for retinoic acid and thyroid hormone receptors (SMRT) [78], Smad2 and Smad3 [79], poly(A)-binding protein 2 [80], human papillomavirus (HPV-16) E7 oncoprotein [81], MIBP1, a member of the major histocompatibility complex binding protein family [82], and the retinoblastoma tumor suppressor protein [83]. These studies highlight a potential common role for NCoA62/SKIP in a more general aspect of target gene regulation by different classes of transcription factors. Importantly, the interaction between NCoA62/SKIP and VDR is independent of the AF-2 domain [75]. Moreover, ligand is not required for the interaction in vitro, although ligand does enhance the interaction. Unlike the SRC family of coactivators, NCoA62/SKIP does not contain obvious LXXLL motifs. NCoA62 also exhibits a marked preference for binding to the VDR-RXR heterodimer relative to VDR alone [84]. The region of NCoA62/SKIP between residues 274 and 342 mediates the interaction with the VDR-RXR heterodimer, and this region is referred to as the receptor interacting domain or RID (Fig. 6). The highly charged C-terminal domain is responsible for NCoA62/SKIP coactivator function in NR-mediated transcription. This region expresses an autonomous transactivation domain that is referred to as TAD-1 (reviewed in [85]). The nuclear localization sequence (NLS) of NCoA62/SKIP is contained in the C-terminal residues 531–536 and is necessary and sufficient for nuclear targeting of NCoA62/SKIP [86]. A strictly conserved LPXP motif is present in all reported NCoA62/SKIP orthologs from a variety of organisms [87], but the function of this domain or region is unknown at present. The mechanisms through which NCoA62/SKIP may function to augment VDR-mediated transcription are becoming more apparent. For example, selective,

CHAPTER 16 Comodulators of Vitamin D Receptor–Mediated Gene Expression

297

FIGURE 6 Schematic of NCoA62/SKIP. LPXP, a strictly conserved motif of unknown function; SNW, region containing a central conserved serine, asparagine, and tryptophan, as well as the receptor interaction domain (RID); TAD-1, transcription activation domain; NLS, nuclear localization sequence.

multiprotein complexes form between VDR, RXR, NCoA62/SKIP, and SRC coactivators [84]. NCoA62 and SRCs contact different domains within the VDRRXR heterodimer. Indeed, both NCoA62/SKIP, and SRC simultaneously interact with liganded VDR to form a ternary complex in vitro, which suggests that liganded VDR can act as a bridge to recruit both SRCs and NCoA62/SKIP to the promoter complex (Fig. 7) [84]. Coexpression of NCoA62/SKIP and SRCs leads to a synergistic induction of VDR-mediated transcription, and protein interference studies indicate that both coactivators are required for optimal VDR-mediated transcription [84]. Perhaps the most promising clues to the significance and mechanism of NCoA62/SKIP in VDR-activated transcription are found in recent chromatin immunoprecipitation (ChIP) approaches [86]. These studies show that NCoA62/SKIP is physically recruited in a 1,25(OH)2D3-dependent manner to the 1,25(OH)2D3responsive regions of native VDR target genes in osteoblasts including the osteocalcin and 24-hydroxylase promoters. Interestingly, NCoA62/SKIP is associated with the promoter region after the entry of both VDR and SRC, suggesting it may function at more distal steps of the transactivation process compared to the SRCs. As mentioned previously, the SRCs exhibit histone acetyltransferase (HAT) activity and recruit other proteins such as CBP/P300 that possess HAT

NCoA62/ D SRCs RXR VDR

VDRE

FIGURE 7

CBP/ p300 PIC

Ac

Ac Ac Ac Ac

Hypothetical model for VDR-SRC-NCoA62/SKIP complexes in VDR-activated transcription.

activity [47,88]. No significant HAT activity has been detected for NCoA62/SKIP (Zhang and MacDonald, unpublished observations). Thus, one possible interpretation is that NCoA62/SKIP enters the promoter region after the chromatin remodeling step. Additionally, NCoA62/SKIP is also known to interact with basal transcription factors such as TFIIB [89], suggesting that NCoA62/SKIP may also be involved in connecting VDR to the general transcription machinery. Regardless, NCoA62/SKIP and SRCs apparently function through different mechanisms to enhance VDR-mediated transcription, and the function of both distinct classes of coactivators is needed for appropriate vitamin D– activated target gene expression.

E. Nuclear Receptor Coactivators: Potential Links to RNA Processing In addition to chromatin remodeling and recruiting the transcription machinery, NR coactivators have been implicated in more distal steps of gene expression including transcription elongation and RNA processing. In this regard, NR coactivators have been proposed to be important coupling factors linking transcription and RNA processing [90]. The coupling concept is that RNA processing events (e.g., RNA capping, polyadenylation, and splicing) are physically connected to RNA Pol II and that RNA processing begins on the nascent RNA emerging from the polymerase complex (reviewed in [91–94]). Putative candidates that may aid in this coupling process are predicted to associate with and alter the functional properties of proteins involved in both transcription and RNA processing. In fact, many NR coactivator proteins contain domains or regions that are associated with RNA processing proteins, making some NR coactivators strong candidates as transcription-splicing couplers [90]. Indeed, Auboeuf et al. have demonstrated that progesterone receptor (PR)-activated transcription influences splicing decisions of alternatively spliced transcripts in a PR- and

298

DIANE R. DOWD, AMELIA L. M. SUTTON, CHI ZHANG, AND PAUL N. MACDONALD

progesterone response element–dependent manner [90]. NR coactivators such as CoAA and PGC-1 have been shown to alter RNA processing events in NRmediated transcription [90,95]. Thus, the idea put forth in these papers is that activated steroid receptors bind to target gene promoters and recruit factors that are involved in both transcription regulation and splicing regulation. This is thought to ensure efficient coordination between RNA synthesis and the final mature mRNA. NCoA62/SKIP may also play a central role in coupling transcription to mRNA slicing. Several observations support this hypothesis. First, endogenous NCoA62/SKIP is recruited to native VDR-responsive promoters in a 1,25-(OH)2D3-dependent manner in osteoblast-like target cells proving its physical association with the transcriptional regulatory machinery at the responsive promoter regions [86]. Second, studies characterizing protein components of the spliceosome and interchromatin granules, where many splicing factors are present, identified NCoA62/SKIP as a component that is associated with the splicing machinery in general and at discrete steps in the splicing mechanism [86,96–98]. Finally, a dominant-negative NCoA62/SKIP mutant effectively interferes with appropriate splicing of transcripts derived from a VDR-activated mini-gene cassette [86]. These studies indicate that disrupting the function of endogenous NCoA62/SKIP inhibits 1,25(OH)2D3-activated transcription, in part, by interfering with appropriate splicing of the mRNA transcripts. This supports the hypothesis that NCoA62/SKIP may couple VDR-mediated transcription to RNA splicing and ensure appropriate, efficient processing of vitamin D– regulated mRNA transcripts.

F. Other Coactivators In addition to these three classes of coactivators, several other proteins interact with VDR in a liganddependent manner and stimulate VDR-mediated transactivation. As mentioned earlier, VDR directly associates with several components of the transcriptional machinery. For example, VDR selectively binds to TFIIB [99,100], a component of the basal transcriptional complex whose entry is rate-limiting in the preinitiation complex formation. The LBD and amino-terminal regions of VDR are involved in contacting an aminoterminal region of TFIIB [99–101], and these two proteins function cooperatively to enhance ligand-dependent transcription [99]. VDR also interacts with TFIID and TFIIA, two essential components of the basal transcriptional machinery [102,103]. TFIID consists of the TATA-box-binding protein and associated factors

or TAFs. VDR associates with the TFIID subunit hTAF(II)135, and TAF(II)135 selectively potentiates VDR-mediated transcription. Direct interaction of VDR with these general factors further strengthens the link between VDR and the core transcriptional machinery and thus represents a putative mechanism in VDR-activated transcription. VDR also contacts Smad3, a transcription factor activated by TGF-β signaling [104]. Smad3 interacts with VDR in a ligand-dependent manner and stimulates VDR-mediated transcription. Additionally, SRC-1 enhances the binding of Smad3 to VDR, suggesting cooperation between these two coactivators. Given its function in two diverse signaling networks, Smad3 may serve as a bridge between 1,25(OH)2D3 and TGF-β signal transduction pathways. Interestingly, NCo62/SKIP serves as a coactivator for Smad-dependent transcription, lending additional support for the concept of crosstalk between these two pathways [79]. An additional binding partner for VDR is the helix–loop–helix transcription factor Ets-1 [105]. This interaction occurs between the DNA-binding domains of both proteins. Unlike other VDR coactivators, Ets-1 can stimulate VDR-mediated transcription in the absence of ligand on select promoters, such as the prolactin promoter.

III. CO-REPRESSORS The role of the co-repressor in regulating gene expression is generally the converse of that of the coactivator (reviewed in [106]). Co-repressors interact with DNA-bound transcription factors and play essential roles in silencing or repressing transcription. In the case of NR-regulated gene expression, co-repressors generally function to lower basal promoter activity in the presence of unliganded receptor; however, in some cases, repression can occur in the presence of NRs bound to antagonists.

A. SMRT and NCoR Perhaps the best characterized NR co-repressors are the ubiquitously expressed proteins SMRT [107] and nuclear receptor co-repressor (NCoR) [108]. SMRT and NCoR co-repressors strongly repress basal promoters in systems regulated by TR and the retinoic acid receptor (RAR). While the SRC family of coactivators interact with NRs via NR boxes (described earlier), SMRT and NCoR co-repressor interaction is mediated through co-repressor NR (CoRNR) boxes [109]. CoRNR boxes are generally composed of the

CHAPTER 16 Comodulators of Vitamin D Receptor–Mediated Gene Expression

sequence Φ-X-X-Φ-Φ, where Φ is a hydrophobic residue and X is any amino acid. Mutation of these sequences abolishes interaction of the co-repressors with unliganded NR [109]. In contrast to the SRC family of coactivators, SMRT and NCoR interaction with VDR is localized to the ligand binding domain but is independent of the AF-2 domain [110,111]. In fact, deletion of the AF-2 helix enhances co-repressor binding [107]. Although VDR is in the same NR family as TR and RAR, the silencing effects of SMRT and NCoR on VDR-regulated templates are weaker than in the other systems [112]. This is most probably due to a reduced binding of SMRT and NCoR to VDR relative to that of TR and RAR. It is possible that there may be antagonists for the VDR that could enhance the binding of the receptor to the co-repressors and thus increase transcriptional silencing. This means of interaction has been documented for co-repressor recruitment to the ER [113,114] and the PR [115]. The mechanism of repression of the SMRT/NCoR class involves interaction with Sin3, a protein shown to be part of a complex that includes HDAC1 and 2 [116–118]. Thus, the Sin3 proteins Sin3A and Sin3B recruit HDAC activity to target gene promoters and their subsequent activities result in compact or tightly wound nucleosomal packaging that ultimately represses basal expression of the promoter.

B. Hairless As mentioned earlier, the VDR plays a role in mammalian hair follicle cycling and disruption of the VDR gene, in some cases, leads to total alopecia (for review see [119]). Inactivation of the hairless gene (hr) also results in a similar phenotype [120]. Hr can repress transcription mediated by VDR, TR and RARrelated orphan receptor, and this discovery provides a molecular basis for its role in the maintenance of hair growth [121–123]. Although Hr lacks homology with SMRT and NCoR, it functions in a similar manner. Like these other co-repressors, Hr binds to VDR and the other NRs in a region localized to the central portion of the ligand binding domain and independent of AF-2 [123], the docking site for some coactivators. This interaction requires two Φ-X-X-Φ-Φ hydrophobic motifs in Hr [123]. Although Hr does not exhibit histone deacetylase activity, it is present in complexes with multiple HDACs. Hr also is found in matrixassociated deacetylase bodies together with SMRT and multiple HDACs [121]. Whereas SMRT and NCoR are ubiquitously expressed, the expression of Hr is restricted primarily to brain and skin, and thus may serve a more specialized role (reviewed in [124]).

299

C. Alien In 1999, Dressel et al. described a novel co-repressor, Alien, which is unrelated to the SMRT/NCoR family of co-repressors [125]. In these early studies, Alien interacted with TR in the absence of ligand and repressed TR-mediated basal transcription. Alien does not interact with the glucocorticoid receptor, RXR, or RAR. In this system, the mechanism of Alien-mediated silencing appears to be through the recruitment of Sin3A and its associated histone deacetylase activity. Alien did not interact with SMRT or NCoR and thus represents an independent pathway to Sin3 recruitment. Polly et al. [126] extended these findings to demonstrate that Alien also interacts with VDR and represses basal transcription from a DR-3 VDRE, but not an atypical IP9-type VDRE, suggesting a level of specificity for repression. The interaction of Alien with unliganded VDR is independent of the AF-2 domain of VDR and appears to use a different interaction surface than does NCoR. In contrast to the studies by Dressel, a deacetylase inhibitor had little effect on the ability of Alien to repress transcription. Thus, in the VDR system, Alien appears to be using alternate molecular pathways to exert its corepressor effect.

IV. CONCLUSION—INTEGRATED MODEL OF COMODULATOR ACTIVITY Gene expression is a complex process involving the coordinated repression and activation of transcription. According to the present model of nuclear hormone signaling, unliganded VDR, like other NRs, are DNAbound and complexed with co-repressor molecules that keep chromatin condensed and the promoter inaccessible to the transcription machinery. Upon binding 1,25(OH)2D3, the co-repressors are displaced by coactivators to begin the transcriptional process. To date, more than 30 nuclear receptor coactivators have been identified and recent effort focuses on distinguishing the redundant and unique functions of these proteins. Immunodepletion and dominant negative approaches indicate that SRCs, Mediator-D, and NCoA62/SKIP have different functions and all are required for robust VDR-mediated transcription. However, the temporal and spatial details of how these diverse coactivators and other accessory transcription factors are assembled onto a 1,25(OH)2D3-responsive promoter remain to be determined. Chromatin immunoprecipitation (ChIP) assays have proven instrumental in providing a limited picture of how endogenous VDR and coactivators convene on a native promoter template by capturing transcriptional complexes assembled on the promoter

300

DIANE R. DOWD, AMELIA L. M. SUTTON, CHI ZHANG, AND PAUL N. MACDONALD

at various times. ChIP assays demonstrate that both nuclear receptors and their coactivators cycle on and off the promoter [68,86,127]. Upon ligand addition, VDR, or other nuclear receptors, enters the transcriptional complex first, followed by SRCs (Fig. 8; [68,86]). SRCs likely loosen the chromatin structure by both bringing intrinsic HAT activity and recruiting extrinsic HAT activity to the complex, through their interaction with p300/CBP. SRCs then dissociate, allowing for binding

LX

SRCs

XL

D RXR VDR

L

p300 VDRE

Ac

p300 SRCs Med 220

L

XL

LX

D RXR VDR

RNA Pol II TBP

VDRE NCoA62/ SKIP

Ac Ac Ac Ac

spliceosome NCoA62/ SKIP

References

Med 220

L

VDRE

XL LX

D RXR VDR

Ac Ac Ac Ac

TBP

RNA Pol II Ac Ac Ac Ac

of the Mediator-D multimeric complex [68,128]. Mediator-D is thought to recruit the PIC and RNA Pol II holoenzyme to initiate transcription of target genes [65]. NCoA62/SKIP also enters the VDR complex following SRC-1 [86], but the relative entrance of NCoA62/SKIP and Mediator-D to the transcriptional complex is currently unknown. It is possible that SRCs help to recruit NCoA62/SKIP to the complex as suggested by the data that these two proteins can form a stable ternary complex with VDR [84]. Once NCoA62/SKIP is bound, it may target the spliceosome complex to the actively transcribed gene [86], allowing for efficient splicing of the nascent transcript. Taken together, these ChIP studies suggest a temporal model of coactivator action in which all three major classes of coactivators enter the complex at distinct times and provide three different functions: chromatin remodeling, recruitment of the core transcriptional machinery, and coupled transcriptional activation and RNA splicing. Importantly, one must be cautious not to oversimplify what is obviously a hugely complex process that requires a variety of macromolecular machines and coordination of many complex pathways. The model presented in Fig. 8 is meant to represent one global means to integrate the known actions of comodulator proteins, but the reality will likely turn out to be a much more complicated scenario.

Ac Ac Ac Ac

FIGURE 8 Model of the sequential occupation of the promoter site with coactivators, the PIC and the spliceosome. First, ligandactivated VDR/RXR binds to VDREs in target genes and recruits coactivators such as SRCs and p300/CBP. The HAT activity of these coactivators loosens the chromatin structure, allowing for a more transcriptionally permissive environment. Second, SRCs and p300/CBP dissociate, allowing for binding of the Mediator-D multimeric complex. This complex is thought to recruit RNA Pol II and the core transcriptional machinery to initiate active transcription of the target gene. Finally, NCoA62/SKIP binds to the VDR/RXR complex and tethers the splicing machinery to the activated promoter, allowing for immediate splicing of the nascent transcript.

1. Fraser DR, Kodicek E 1970 Unique biosynthesis by kidney of a biologically active vitamin D metabolite. Nature 228: 764–766. 2. Kodicek E, Lawson DEM, Wilson PW 1970 Biological activity of a polar metabolite of vitamin D3. Nature 228: 763–765. 3. Holick MF, Schnoes HK, DeLuca HF, Suda T, Cousins RJ 1971 Isolation and identification of 1,25-dihydroxycholecalciferol: a metabolite of vitamin D active in the intestine. Biochemistry 10:2799–2804. 4. Owen TA, Aronow MS, Barone LM, Bettencourt B, Stein GS, Lian JB 1991 Pleiotropic effects of vitamin D on osteoblast gene expression are related to the proliferative and differentiated state of the bone cell phenotype: dependency upon basal levels of gene expression, duration of exposure, and bone matrix competency in normal rat osteoblast cultures. Endocrinology 128:1496–504. 5. Bar-Shavit Z, Teitelbaum SL, Reitsma P, Hall A, Pegg LE, Trial S, Kahn AJ 1983 Induction of moncytic differentiation and bone resorption by 1,25(OH)2D3. Proc Natl Acad Sci USA 80:5908–5911. 6. Sutton AL, MacDonald PN 2003 Vitamin D: more than a “bone-a-fide” hormone. Mol Endocrinol 17:777–791. 7. McDonnell DP, Mangelsdorf DJ, Pike JW, Haussler MR, O’Malley BW 1987 Molecular cloning of complementary DNA encoding the avian receptor for vitamin D. Science 235:1214–1217. 8. Liao J, Ozono K, Sone T, McDonnell DP, Pike JW 1990 Vitamin D receptor interaction with specific DNA requires

CHAPTER 16 Comodulators of Vitamin D Receptor–Mediated Gene Expression

9.

10.

11.

12. 13.

14. 15. 16.

17. 18. 19.

20.

21.

22.

23.

24.

a nuclear protein and 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 87:9751–9755. MacDonald PN, Haussler CA, Terpening CM, Galligan MA, Reeder MC, Whitfield GK, Haussler MR 1991 Baculovirusmediated expression of the human vitamin D receptor. Functional characterization, vitamin D response element interactions and evidence for a receptor auxiliary factor. J Biol Chem 266:18808–18813. MacDonald PN, Dowd DR, Nakajima S, Galligan MA, Reeder MC, Haussler CA, Ozato K, Haussler MR 1993 Retinoid X receptors stimulate and 9-cis retinoic acid inhibits 1,25-dihydroxyvitamin D3-activated expression of the rat osteocalcin gene. Mol Cell Biol 13:5907–5917. Kliewer SA, Umesono K, Mangelsdorf DJ, Evans RM 1992 Retinoid X receptor interacts with nuclear receptors in retinoic acid, thyroid hormone, and vitamin D3 signalling. Nature 355:446–449. Mangelsdorf DJ, Evans RM 1995 The RXR heterodimers and orphan receptors. Cell 83:841–850. Kurokawa R, Yu VC, Naar A, Kyakumoto S, Han Z, Silverman S, Rosenfeld MG, Glass CK 1993 Differential orientations of the DNA-binding domain and carboxy-terminal dimerization interface regulate binding site selection by nuclear receptor heterodimers. Genes Dev 7:1423–1435. Lemon BD, Freedman LP 1996 Selective effects of ligands on vitamin D3 receptor- and retinoid X receptor-mediated gene activation in vivo. Mol Cell Biol 16:1006–1016. Tasset D, Tora L, Fromental C, Scheer E, Chambon P 1990 Distinct classes of transcriptional activating domains function by different mechanisms. Cell 62:1177–1187. Meyer M-E, Gronemeyer H, Turcotte B, Bocquel M-T, Tasset D, Chambon P 1989 Steroid hormone receptors compete for factors that mediate their enhancer function. Cell 57:433–442. McDonnell DP, Scott RA, Kerner SA, O’Malley BW, Pike JW 1989 Functional domains of the human vitamin D3 receptor regulate osteocalcin gene expression. Mol Endocrinol 3:635–644. Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179. Nakajima S, Hsieh J-C, MacDonald PN, Galligan MA, Haussler CA, Whitfield GK, Haussler MR 1994 The C-terminal region of the vitamin D receptor is essential to form a complex with a receptor auxiliary factor required for high affinity binding to the vitamin D–responsive element. Mol Endocrinol 8:159–172. Masuyama H, Brownfield CM, St.-Arnaud R, MacDonald PN 1997 Evidence for ligand-dependent intramolecular folding of the AF-2 domain in vitamin D receptor-activated transcription and coactivator interaction. Mol Endocrinol 11:1507–1517. Danielian PS, White R, Lees JA, Parker MG 1992 Identification of a conserved region required for hormone dependent transcriptional activation by steroid hormone receptors. EMBO J 11:1025–1033. Renaud J-P, Rochel N, Ruff M, Vivat V, Chambon P, Gronemeyer H, Moras D 1995 Crystal structure of the RAR-γ ligand-binding domain bound to all-trans retinoic acid. Nature 378:681–689. Feng W, Ribeiro RC, Wagner RL, Nguyen H, Apriletti JW, Fletterick RJ, Baxter JD, Kushner PJ, West BL 1998 Hormone-dependent coactivator binding to a hydrophobic cleft on nuclear receptors. Science 280:1747–1749. Heery DM, Kalkhoven E, Hoare S, Parker MG 1997 A signature motif in transcriptional co-activators mediates binding to nuclear receptors. Nature 387:733–736.

301

25. Nolte, RT, Wisely GB, Westin S, Cobb JE, Lambert MH, Kurokawa R, Rosenfeld MG, Willson TM, Glass CK, Milburn MV 1998 Ligand binding and co-activator assembly of the peroxisome proliferator–activated receptor-gamma. Nature 395:137–143. 26. Darimont BD, Wagner RL, Apriletti JW, Stallcup MR, Kushner PJ, Baxter JD, Fletterick RJ, and Yamamoto KR 1998 Structure and specificity of nuclear receptor–coactivator interactions. Genes Dev 12:3343–3356. 27. McInerney EM, Rose DW, Flynn SE, Westin S, Mullen TM, Krones A, Inostroza J, Torchia J, Nolte RT, Assa-Munt N, Milburn MV, Glass CK, Rosenfeld MG 1998 Determinants of coactivator LXXLL motif specificity in nuclear receptor transcriptional activation. Genes Dev 12:3357–3368. 28. Pathrose P, Barmina O, Chang CY, McDonnell DP, Shevde NK, Pike JW 2002 Inhibition of 1,25-dihydroxyvitamin D3dependent transcription by synthetic LXXLL peptide antagonists that target the activation domains of the vitamin D and retinoid X receptors. J Bone Miner Res 17:2196–2205. 29. Onate SA, Tsai SY, Tsai M-J, O’Malley BW 1995 Sequence and characterization of a coactivator for the steroid hormone receptor superfamily. Science 270:1354–1357. 30. Kraichely DM, Collins JJ III, DeLisle RK, MacDonald PN 1999 The autonomous transactivation domain in helix H3 of the vitamin D receptor is required for transactivation and coactivator interaction. J Biol Chem 274:14352–14358. 31. Gill RK, Bell NH 2000 Steroid receptor co-activator-1 mediates 1,25-dihydroxyvitamin D3–stimulated alkaline phosphatase in human osteosarcoma cells. Calcif Tissue Int 66:370–374. 32. Hong H, Kohli K, Trivedi A, Johnson DL, Stallcup MR 1996 GRIP1, a novel mouse protein that serves as a transcriptional coactivator in yeast for the hormone binding domains of steroid receptors. Proc Natl Acad Sci USA 93:4948–4952. 33. Voegel JJ, Heine MJ, Zechel C, Chambon P, Gronemeyer H 1996 TIF2, a 160 kDa transcriptional mediator for the liganddependent activation function AF-2 of nuclear receptors. EMBO J 15:3667–3675. 34. Li H, Gomes PJ, Chen JD 1997 RAC3, a steroid/nuclear receptor–associated coactivator that is related to SRC-1 and TIF2. Proc Natl Acad Sci USA 94:8479–8484. 35. Anzick SL, Kononen J, Walker RL, Azorsa DO, Tanner MM, Guan XYGS, Kallioniemi OP, Trent JM, Meltzer PS 1997 AIB1, a steroid receptor coactivator amplified in breast and ovarian cancer. Science 277:965–968. 36. Chen H, Lin RJ, Schiltz RL, Chakravarti D, Nash A, Nagy L, Privalsky ML, Nakatani Y, Evans RM 1997 Nuclear receptor coactivator ACTR is a novel histone acetyltransferase and forms a multimeric activation complex with P/CAF and CBP/p300. Cell 90:569–580. 37. Takeshita A, Cardona GR, Koibuchi N, Suen CS, Chin WW 1997 TRAM-1, a novel 160-kDa thyroid hormone receptor activator molecule, exhibits distinct properties from steroid receptor coactivator-1. J Biol Chem 272:27629–27634. 38. Torchia J, Rose DW, Inostroza J, Kamei Y, Westin S, Glass CK, Rosenfeld MG 1997 The transcriptional co-activator p/CIP binds CBP and mediates nuclear-receptor function. Nature 387:677–684. 39. Onate SA, Boonyaratanakornkit V, Spencer TE, Tsai SY, Tsai MJ, Edwards DP, O’Malley BW 1998 The steroid receptor coactivator-1 contains multiple receptor interacting and activation domains that cooperatively enhance the activation function 1 (AF1) and AF2 domains of steroid receptors. J Biol Chem 273:12101–12108. 40. Ding XF, Anderson CM, Ma H, Hong H, Uht RM, Kushner PJ, Stallcup MR 1998 Nuclear receptor-binding sites of coactivators

302

41.

42. 43.

44.

45. 46.

47.

48. 49. 50. 51. 52. 53.

54.

55.

56.

57.

DIANE R. DOWD, AMELIA L. M. SUTTON, CHI ZHANG, AND PAUL N. MACDONALD

glucocorticoid receptor interacting protein 1 (GRIP1) and steroid receptor coactivator 1 (SRC-1): multiple motifs with different binding specificities. Mol Endocrinol 12:302–313. Leers J, Treuter E, Gustafsson JA 1998 Mechanistic principles in NR box–dependent interaction between nuclear hormone receptors and the coactivator TIF2. Mol Cell Biol 18: 6001–6013. Leo C, Li H, Chen JD 2000 Differential mechanisms of nuclear receptor regulation by receptor-associated coactivator 3. J Biol Chem 275:5976–5982. Voegel JJ, Heine MJ, Tini M, Vivat V, Chambon P, Gronemeyer H 1998 The coactivator TIF2 contains three nuclear receptor-binding motifs and mediates transactivation through CBP binding-dependent and -independent pathways. EMBO J 17:507–519. Korzus E, Torchia J, Rose DW, Xu L, Kurokawa R, McInerney EM, Mullen TM, Glass CK, Rosenfeld MG 1998 Transcription factor–specific requirements for coactivators and their acetyltransferase functions. Science 279:703–707. Chen D, Ma H, Hong H, Koh SS, Huang SM, Schurter BT, Aswad DW, Stallcup MR 1999 Regulation of transcription by a protein methyltransferase. Science 284:2174–2177. Utley RT, Ikeda K, Grant PA, Cote J, Steger DJ, Eberharter A, John S, Workman JL 1998 Transcriptional activators direct histone acetyltransferase complexes to nucleosomes. Nature 394:498–502. Spencer TE, Jenster G, Burcin MM, Allis CD, Zhou J, Mizzen CA, McKenna NJ, Onate SA, Tsai SY, Tsai MJ, O’Malley BW 1997 Steroid receptor coactivator-1 is a histone acetyltransferase. Nature 389:194–198. Chen H, Lin RJ, Xie W, Wilpitz D, Evans RM 1999 Regulation of hormone-induced histone hyperacetylation and gene activation via acetylation of an acetylase. Cell 98:675–686. Workman JL, Kingston RE 1998 Alteration of nucleosome structure as a mechanism of transcriptional regulation. Annu Rev Biochem 67:545–579. Ogryzko VV, Schiltz RL, Russanova V, Howard BH, Nakatani Y 1996 The transcriptional coactivators p300 and CBP are histone acetyltransferases. Cell 87:953–959. Bannister AJ, Kouzarides T 1996 The CBP co-activator is a histone acetyltransferase. Nature 384:641–643. Yang XJ, Ogryzko VV, Nishikawa J, Howard BH, Nakatani Y 1996 A p300/CBP-associated factor that competes with the adenoviral oncoprotein E1A. Nature 382:319–324. Kamei Y, Xu L, Heinzel T, Torchia J, Kurokawa R, Gloss B, Lin S-C, Heyman RA, Rose DW, Glass CK, Rosenfeld MG 1996 A CBP integrator complex mediates transcriptional activation and AP-1 inhibition by nuclear receptors. Cell 85: 403–414. Chakravarti D, LaMorte VJ, Nelson MC, Nakajima T, Schulman IG, Juguilon H, Montminy M, Evans RM 1996 Role of CBP/P300 in nuclear receptor signalling. Nature 383:99–103. Xu J, Qiu Y, DeMayo FJ, Tsai SY, Tsai MJ, O’Malley BW 1998 Partial hormone resistance in mice with disruption of the steroid receptor coactivator-1 (SRC-1) gene. Science 279:1922–1925. Gehin M, Mark M, Dennefeld C, Dierich A, Gronemeyer H, Chambon P 2002 The function of TIF2/GRIP1 in mouse reproduction is distinct from those of SRC-1 and p/CIP. Mol Cell Biol 22:5923–5937. Yuan Y, Liao L, Tulis DA, Xu J 2002 Steroid receptor coactivator-3 is required for inhibition of neointima formation by estrogen. Circulation 105:2653–2659.

58. Issa LL, Leong GM, Barry JB, Sutherland RL, Eisman JA 2001 Glucocorticoid receptor-interacting protein-1 and receptorassociated coactivator-3 differentially interact with the vitamin D receptor (VDR) and regulate VDR-retinoid X receptor transcriptional cross-talk. Endocrinology 142:1606–1615. 59. Rachez C, Suldan Z, Ward J, Chang CP, Burakov D, Erdjument-Bromage H, Tempst P, Freedman LP 1998 A novel protein complex that interacts with the vitamin D3 receptor in a ligand-dependent manner and enhances VDR transactivation in a cell-free system. Genes Dev 12: 1787–1800. 60. Rachez C, Lemon BD, Suldan Z, Bromleigh V, Gamble M, Naar AM, Erdjument-Bromage H, Tempst P, Freedman LP 1999 Ligand-dependent transcription activation by nuclear receptors requires the DRIP complex. Nature 398:824–828. 61. Fondell JD, Ge H, Roeder RG 1996 Ligand induction of a transcriptionally active thyroid hormone receptor coactivator complex. Proc Natl Acad Sci USA 93:8329–8333. 62. Jiang YW, Veschambre P, Erdjument-Bromage H, Tempst P, Conaway JW, Conaway RC, Kornberg RD 1998 Mammalian mediator of transcriptional regulation and its possible role as an end-point of signal transduction pathways. Proc Natl Acad Sci USA 95:8538–8543. 63. Naar AM, Beaurang PA, Zhou S, Abraham S, Solomon W, Tjian R 1999 Composite co-activator ARC mediates chromatindirected transcriptional activation. Nature 398:828–832. 64. Rachez C, Freedman LP 2001 Mediator complexes and transcription. Curr Opin Cell Biol 13:274–280. 65. Chiba N, Suldan Z, Freedman LP, Parvin JD 2000 Binding of liganded vitamin D receptor to the vitamin D receptor interacting protein coactivator complex induces interaction with RNA polymerase II holoenzyme. J Biol Chem 275: 10719–10722. 66. Sun X, Zhang Y, Cho H, Rickert P, Lees E, Lane W, Reinberg D 1998 NAT, a human complex containing Srb polypeptides that functions as a negative regulator of activated transcription. Mol Cell 2:213–222. 67. Yudkovsky N, Ranish JA, Hahn S 2000 A transcription reinitiation intermediate that is stabilized by activator. Nature 408:225–229. 68. Burakov D, Crofts LA, Chang CP, Freedman LP 2002 Reciprocal recruitment of DRIP/mediator and p160 coactivator complexes in vivo by estrogen receptor. J Biol Chem 277: 14359–14362. 69. Lorch Y, Beve J, Gustafsson CM, Myers LC, Kornberg RD 2000 Mediator-nucleosome interaction. Mol Cell 6:197–201. 70. Rachez C, Gamble M, Chang CP, Atkins GB, Lazar MA, Freedman LP 2000 The DRIP complex and SRC-1/p160 coactivators share similar nuclear receptor binding determinants but constitute functionally distinct complexes. Mol Cell Biol 20:2718–2726. 71. Cantin GT, Stevens JL, Berk AJ 2003 Activation domainmediator interactions promote transcription preinitiation complex assembly on promoter DNA. Proc Natl Acad Sci USA 100:12003–12008. 72. Barletta F, Freedman LP, Christakos S 2002 Enhancement of VDR-mediated transcription by phosphorylation: correlation with increased interaction between the VDR and DRIP205, a subunit of the VDR-interacting protein coactivator complex. Mol Endocrinol 16:301–314. 73. Yang W, Freedman LP 1999 20-Epi analogues of 1,25-dihydroxyvitamin D3 are highly potent inducers of DRIP coactivator complex binding to the vitamin D3 receptor. J Biol Chem 274:16838–16845.

CHAPTER 16 Comodulators of Vitamin D Receptor–Mediated Gene Expression

74. Ito M, Yuan CX, Okano HJ, Darnell RB, Roeder RG 2000 Involvement of the TRAP220 component of the TRAP/SMCC coactivator complex in embryonic development and thyroid hormone action. Mol Cell 5:683–693. 75. Baudino TA, Kraichely DM, Jefcoat SC Jr, Winchester SK, Partridge NC, MacDonald PN 1998 Isolation and characterization of a novel coactivator protein, NCoA-62, involved in vitamin D-mediated transcription. J Biol Chem 273: 16434–16441. 76. Wieland C, Mann S, Besser H, Saumweber H 1992 The Drosophila nuclear protein Bx42, which is found in many puffs on polytene chromosomes, is highly charged. Chromosoma 101:517–525. 77. Dahl R, Wani B, Hayman MJ 1998 The Ski oncoprotein interacts with Skip, the human homolog of Drosophila Bx42. Oncogene 16:1579–1586. 78. Zhou S, Fujimuro M, Hsieh JJ, Chen L, Miyamoto A, Weinmaster G, Hayward SD 2000 SKIP, a CBF1-associated protein, interacts with the ankyrin repeat domain of NotchIC to facilitate NotchIC function. Mol Cell Biol 20:2400–2410. 79. Leong GM, Subramaniam N, Figueroa J, Flanagan JL, Hayman MJ, Eisman JA, Kouzmenko AP 2001 Ski-interacting protein interacts with Smad proteins to augment transforming growth factor-beta-dependent transcription. J Biol Chem 276:18243–18248. 80. Kim YJ, Noguchi S, Hayashi YK, Tsukahara T, Shimizu T, Arahata K 2001 The product of an oculopharyngeal muscular dystrophy gene, poly(A)-binding protein 2, interacts with SKIP and stimulates muscle-specific gene expression. Hum Mol Genet 10:1129–1139. 81. Prathapam T, Kuhne C, Banks L 2001 The HPV-16 E7 oncoprotein binds Skip and suppresses its transcriptional activity. Oncogene 20:7677–7685. 82. Fukuda S, Yamasaki Y, Iwaki T, Kawasaki H, Akieda S, Fukuchi N, Tahira T, Hayashi K 2002 Characterization of the biological functions of a transcription factor, c-myc intron binding protein 1 (MIBP1). J Biochem (Tokyo) 131:349–357. 83. Prathapam T, Kuhne C, Banks L 2002 Skip interacts with the retinoblastoma tumor suppressor and inhibits its transcriptional repression activity. Nucleic Acids Res 30:5261–5268. 84. Zhang C, Baudino TA, Dowd DR, Tokumaru H, Wang W, MacDonald PN 2001 Ternary complexes and cooperative interplay between NCoA-62/Ski-interacting protein and steroid receptor coactivators in vitamin D receptor-mediated transcription. J Biol Chem 276:40614–40620. 85. MacDonald PN, Baudino TA, Tokumaru H, Dowd DR, Zhang C 2001 Vitamin D receptor and nuclear receptor coactivators: crucial interactions in vitamin D–mediated transcription. Steroids 66:171–176. 86. Zhang C, Dowd DR, Staal A, Gu C, Lian JB, Van Wijnen AJ, Stein GS, MacDonald PN 2003 NCoA62/SKIP is a nuclear matrix-associated coactivator that may couple vitamin D receptor-mediated transcription and RNA splicing. J Biol Chem 278:35325–35336. 87. Kostrouchova M, Housa D, Kostrouch Z, Saudek V, Rall JE 2002 SKIP is an indispensable factor for Caenorhabditis elegans development. Proc Natl Acad Sci USA 99:9254–9259. 88. Orgryzko VV, Schiltz RL, Russanova V, Howard BH, Nakatani Y 1996 The transcriptional coactivators p300 and CBP are histone acetyltransferases. Cell 87:953–959. 89. Barry JB, Leong GM, Church WB, Issa LL, Eisman JA, Gardiner EM 2003 Interactions of SKIP/NCoA-62, TFIIB, and retinoid X receptor with vitamin D receptor helix H10 residues. J Biol Chem 278:8224–8228.

303

90. Auboeuf D, Honig A, Berget SM, O’Malley BW 2002 Coordinate regulation of transcription and splicing by steroid receptor coregulators. Science 298:416–419. 91. Manley JL 2002 Nuclear coupling: RNA processing reaches back to transcription. Nat Struct Biol 9:790–791. 92. Orphanides G, Reinberg D 2002 A unified theory of gene expression. Cell 108:439–451. 93. Proudfoot NJ, Furger A, Dye MJ 2002 Integrating mRNA processing with transcription. Cell 108:501–512. 94. Maniatis T, Reed R 2002 An extensive network of coupling among gene expression machines. Nature 416:499–506. 95. Monsalve M, Wu Z, Adelmant G, Puigserver P, Fan M, Spiegelman BM 2000 Direct coupling of transcription and mRNA processing through the thermogenic coactivator PGC-1. Mol Cell 6:307–316. 96. Makarov EM, Makarova OV, Urlaub H, Gentzel M, Will CL, Wilm M, Luhrmann R 2002 Small nuclear ribonucleoprotein remodeling during catalytic activation of the spliceosome. Science 298:2205–2208. 97. Mintz PJ, Patterson SD, Neuwald AF, Spahr CS, Spector DL 1999 Purification and biochemical characterization of interchromatin granule clusters. EMBO J 18:4308–4320. 98. Neubauer G, King A, Rappsilber J, Calvio C, Watson M, Ajuh P, Sleeman J, Lamond A, Mann M 1998 Mass spectrometry and EST-database searching allows characterization of the multi-protein spliceosome complex. Nat Genet 20:46–50. 99. Blanco JCG, Wang I-M, Tsai SY, Tsai M-J, O’Malley BW, Jurutka PW, Haussler MR, Ozato K 1995 Transcription factor TFIIB and the vitamin D receptor cooperatively activate ligand-dependent transcription. Proc Natl Acad Sci USA 92:1535–1539. 100. MacDonald PN, Sherman DR, Dowd DR, Jefcoat SC, DeLisle RK 1995 The vitamin D receptor interacts with general transcription factor IIB. J Biol Chem 270:4748–4752. 101. Jurutka PW, Remus LS, Whitfield GK, Thompson PD, Hsieh JC, Zitzer H, Tavakkoli P, Galligan MA, Dang HT, Haussler CA, Haussler MR 2000 The polymorphic N terminus in human vitamin D receptor isoforms influences transcriptional activity by modulating interaction with transcription factor IIB. Mol Endocrinol 14:401–420. 102. Lemon BD, Fondell JD, Freedman LP 1997 Retinoid X receptor:vitamin D3 receptor heterodimers promote stable preinitiation complex formation and direct 1,25-dihydroxyvitamin D3-dependent cell-free transcription. Mol Cell Biol 17:1923–1937. 103. Mengus G, May M, Carre L, Chambon P, Davidson I 1997 Human TAF(II)135 potentiates transcriptional activation by the AF-2s of the retinoic acid, vitamin D3, and thyroid hormone receptors in mammalian cells. Genes Dev 11: 1381–1395. 104. Yanagisawa J, Yanagi Y, Masuhiro Y, Suzawa M, Watanabe M, Kashiwagi K, Toriyabe T, Kawabata M, Miyazono K, Kato S 1999 Convergence of transforming growth factor-beta and vitamin D signaling pathways on SMAD transcriptional coactivators. Science 283:1317–1321. 105. Tolon RM, Castillo AI, Jimenez-Lara AM, Aranda A 2000 Association with Ets-1 causes ligand- and AF2-independent activation of nuclear receptors. Mol Cell Biol 20:8793–8802. 106. Rosenfeld MG, Glass CK 2001 Coregulator codes of transcriptional regulation by nuclear receptors. J Biol Chem 276:36865–36868. 107. Chen JD, Evans RM 1995 A transcriptional co-repressor that interacts with nuclear hormone receptors. Nature 377: 454–457.

304

DIANE R. DOWD, AMELIA L. M. SUTTON, CHI ZHANG, AND PAUL N. MACDONALD

108. Horlein AJ, Naar AM, Heinzel T, Torchia J, Gloss B, Kurokawa R, Ryan A, Kamei Y, Soderstrom M, Glass CK, Rosenfeld MG 1995 Ligand-independent repression by the thyroid hormone receptor mediated by a nuclear receptor co-repressor. Nature 377:397–404. 109. Hu X, Lazar MA 1999 The CoRNR motif controls the recruitment of corepressors by nuclear hormone receptors. Nature 402:93–96. 110. Nagy L, Kao HY, Love JD, Li C, Banayo E, Gooch JT, Krishna V, Chatterjee K, Evans RM, Schwabe JW 1999 Mechanism of corepressor binding and release from nuclear hormone receptors. Genes Dev 13:3209–3216. 111. Perissi V, Staszewski LM, McInerney EM, Kurokawa R, Krones A, Rose DW, Lambert MH, Milburn MV, Glass CK, Rosenfeld MG 1999 Molecular determinants of nuclear receptor-corepressor interaction. Genes Dev 13:3198–3208. 112. Tagami T, Lutz WH, Kumar R, Jameson JL 1998 The interaction of the vitamin D receptor with nuclear receptor corepressors and coactivators. Biochem Biophys Res Commun 253:358–363. 113. Smith CL, Nawaz Z, O’Malley BW 1997 Coactivator and corepressor regulation of the agonist/antagonist activity of the mixed antiestrogen, 4-hydroxytamoxifen. Mol Endocrinol 11:657–666. 114. Jackson TA, Richer JK, Bain DL, Takimoto GS, Tung L, Horwitz KB 1997 The partial agonist activity of antagonistoccupied steroid receptors is controlled by a novel hinge domain-binding coactivator L7/SPA and the corepressors N-CoR or SMRT. Mol Endocrinol 11:693–705. 115. Zhang X, Jeyakumar M, Petukhov S, Bagchi MK 1998 A nuclear receptor corepressor modulates transcriptional activity of antagonist-occupied steroid hormone receptor. Mol Endocrinol 12:513–524. 116. Nagy L, Kao HY, Chakravarti D, Lin RJ, Hassig CA, Ayer DE, Schreiber SL, Evans RM 1997 Nuclear receptor repression mediated by a complex containing SMRT, mSin3A, and histone deacetylase. Cell 89:373–380. 117. Heinzel T, Lavinsky RM, Mullen T-M, Soderstrom M, Laherty CD, Torchia J, Yang WM, Brard G, Ngo SD, Davie JR, Seto E, Eisenman RN, Rose DW, Glass CK, Rosenfeld MG 1997 A complex containing N-CoR, mSin3 and histone deacetylase mediates transcriptional repression. Nature 387:43–48.

118. Alland L, Muhle R, Hou H, Potes J, Chin L, Schreiber-Agus N, Depinho RA 1997 Role for N-CoR and histone deacetylase in Sin3-mediated transcriptional repression. Nature 387: 49–55. 119. Malloy PJ, Pike JW, Feldman D 1999 The vitamin D receptor and the syndrome of hereditary 1,25-dihydroxyvitamin D– resistant rickets. Endocr Rev 20:156–188. 120. Miller J, Djabali K, Chen T, Liu Y, Ioffreda M, Lyle S, Christiano AM, Holick M, Cotsarelis G 2001 Atrichia caused by mutations in the vitamin D receptor gene is a phenocopy of generalized atrichia caused by mutations in the hairless gene. J Invest Dermatol 117:612–617. 121. Potter GB, Beaudoin GM 3rd, DeRenzo CL, Zarach JM, Chen SH, Thompson CC 2001 The hairless gene mutated in congenital hair loss disorders encodes a novel nuclear receptor corepressor. Genes Dev 15:2687–2701. 122. Moraitis AN, Giguere V, Thompson CC 2002 Novel mechanism of nuclear receptor corepressor interaction dictated by activation function 2 helix determinants. Mol Cell Biol 22:6831–6841. 123. Hsieh JC, Sisk JM, Jurutka PW, Haussler CA, Slater SA, Haussler MR, Thompson CC 2003 Physical and functional interaction between the vitamin D receptor and hairless corepressor, two proteins required for hair cycling. J Biol Chem 278:38665–38674. 124. Panteleyev AA, Paus R, Ahmad W, Sundberg JP, Christiano AM 1998 Molecular and functional aspects of the hairless (hr) gene in laboratory rodents and humans. Exp Dermatol 7:249–267. 125. Dressel U, Thormeyer D, Altincicek B, Paululat A, Eggert M, Schneider S, Tenbaum SP, Renkawitz R, Baniahmad A 1999 Alien, a highly conserved protein with characteristics of a corepressor for members of the nuclear hormone receptor superfamily. Mol Cell Biol 19:3383–3394. 126. Polly P, Herdick M, Moehren U, Baniahmad A, Heinzel T, Carlberg C 2000 VDR-Alien: a novel, DNA-selective vitamin D3 receptor-corepressor partnership. FASEB J 14:1455–1463. 127. Shang Y, Hu X, DiRenzo J, Lazar MA, Brown M 2000 Cofactor dynamics and sufficiency in estrogen receptorregulated transcription. Cell 103:843–852. 128. Sharma D, Fondell JD 2002 Ordered recruitment of histone acetyltransferases and the TRAP/Mediator complex to thyroid hormone-responsive promoters in vivo. Proc Natl Acad Sci USA 99:7934–7939.

CHAPTER 17

Promoter Targeting of Vitamin D Receptor through a Chromatin Remodeling Complex SHIGEAKI KATO,* RYOJI FUJIKI, AND HIROCHIKA KITAGAWA* Institute of Molecular and Cellular Biosciences, University of Tokyo, Japan, and *SOREST, Japan Science and Technology, Saitama, Japan

I. Introduction II. Chromatin Remodeling Is a Prerequisite for Transcriptional Controls by the Vitamin D Receptor III. Purification and Identification of Williams Syndrome Transcription Factor as a Vitamin D Receptor Interactant IV. Purification and Identification of a Novel Williams Syndrome Transcription Factor Complex Associating with Vitamin D Receptor V. WINAC Is a Novel Multifunctional ATP-Dependent Chromatin Remodeling Complex That Rearranges

a Nucleosome Array around a Vitamin D Responsive Element in Vitro VI. Williams Syndrome Transcription Factor Coactivated the Ligand-Induced Transactivation Function of Vitamin D Receptor VII. Molecular Mechanism of VDR Promoter Targeting of Vitamin D Receptor by WINAC and Cooperative WINAC Function with the Co-regulator Complexes References

I. INTRODUCTION

These co-regulatory complexes contain enzymes that modify histones through acetylation, which results in rearrangement of nucleosome arrays [9]. However, prior to such necleosome array reorganization by co-regulator complexes, chromatin remodeling is believed to be coupled with the promoter targeting of VDR, like all of the DNA binding transcription factors (Fig. 1).

The calciotropic hormone 1,25(OH)2D3, the active form of vitamin D3, regulates calcium homeostasis as well as cellular proliferation and differentiation [1]. Most biological actions of 1,25(OH)2D3 are believed to be mediated through transcriptional controls of a particular set of target genes by the vitamin D receptor (VDR) [2,3]. VDR is a member of the nuclear receptor (NR) gene superfamily, which acts as a ligand-inducible transcriptional factor [4]. Like other members of the nuclear receptor superfamily, VDR structure is divided into several functional domains, with the most highly conserved DNA-binding domain (C) located centrally and the less highly conserved ligand-binding domain (E) located at the C-terminal end [5]. Most nuclear receptors harbor both an N-terminal activation function 1 (AF-1) and a C-terminal AF-2 domain [6]. The VDR, however, appears to lack the significant N-terminal AF-1 function because of its relatively short A/B domain. In the promoters of target genes that are controlled by liganded VDR, VDR/RXR heterodimer recognizes and directly binds to cognate vitamin D responsive elements (VDREs) [6]. Liganded VDR bound upon the VDREs induces the recruitment of a number of histone acetyltransferase (HAT) and non-HAT coactivators and coactivator complexes to activate transcription [7,8]. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

II. CHROMATIN REMODELING IS A PREREQUISITE FOR TRANSCRIPTIONAL CONTROLS BY THE VITAMIN D RECEPTOR At transcriptional initiation sites in promoters, rearrangement of the nucleosome array is thought to be indispensable for controlling transcription of sequencespecific regulators such as VDR through chromatin remodeling and histone modifications [9,10]. These complexes (Fig. 2) modify the chromatin configuration in a highly regulated manner. One class contains several discrete subfamilies of transcription coregulatory complexes that harbor either HAT or histone deacetylase (HDAC) activities to covalently modify histones through acetylation [11]. In NR ligand-induced transactivation processes, the complexes containing HDAC first act to corepress transactivation of unliganded NRs. Upon ligand binding, two HAT complexes, p160/CBP and TRRAP/GCN5, are Copyright © 2005, Elsevier, Inc. All rights reserved.

306

SHIGEAKI KATO, RYOJI FUJIKI, AND HIROCHIKA KITAGAWA

Co-activator complexes

Co-repressor complexes

HAT complexes

HDAC complexes HDAC1/2

PCAF complex

mSin3

p300

NCoR/ SMRT

non-HAT co-activators

p160 family protein

DRIP/TRAP complexes Chromatin remodeling

Chromatin remodeling

VDR RXR

VDRE

FIGURE 1 VDRs coordinate the activities of chromatin remodeling complexes to facilitate. Because the promoter is assembled into a nucleosomal structure that is transcriptonally inactive the chromatin environment limits the accessibility of DNA-binding transcription factor and their co-regulators to promoter. To conquer this difficulty, an interaction of VDR with unknown chromatin remodeling complexes reorganizes nucleosomal arrays at VDR targeting promoters, which results in stabilized binding of transcription co-regulator.

A SWI/SNF family

ISWI family

ATPase

Bromo

hSWI/SNF complex Brg1/ BAF250 hBrm BAF60a/b/c BAF155 BAF170 BAF47 Ini1 BAF50

B

ATPase

SANT

hSpt3 hPCAF TAF30 hAda3 PAF65β

chromo

ATPase

hNuRD complex

hAcf1

MTA2 HDAC1 MBD3

hISWI

RbA MTA1 p46 Mi-2 RbA HDAC2 p48

HDAC complexes CBP/p160 complex

TAF15 PAF65α hAda2 TRRAP

PHD

hACF complex

HAT complexes hPCAF complex

Mi-2 family

CBP/ p300 p160 family protein

mSin3 complex HDAC1SAP18 RbA p46 mSin3 SAP30 HDAC2 RbA p48

TAF31

FIGURE 2 Chromatin remodeling complexes. (A) ATP-dependent chromatin remodeling complexes. These types of complexes can be divided into three major classes based on catalytic ATPase subunit: Brg1/hBrm, hISWI, and Mi-2. Shown here are some representative members of the SWI/SNF, ISWI, and Mi-2 families of human ATP-dependent chromatin remodeling complexes. (B) Some representative members of the histone modifying complexes. Purple, ATPase; red, histone acetyltransferase; green, histone deacetylase. (See color plate).

CHAPTER 17 Promoter Targeting of Vitamin D Receptor through a Chromatin Remodeling Complex

recruited with dissociation of the HDAC complexes, and coactivate the NR function [12–14]. Another class of complexes contains chromatin remodeling complexes that use ATP hydrolysis to rearrange nucleosomal arrays in a noncovalent manner, thereby making chromosomal DNA accessible for sequence-specific regulators such as the VDR [11]. Besides of transcriptional controls, these ATP-dependent chromatin remodeling complexes are supposed to act on DNA repair and DNA replication. These complexes are further classified into subfamilies based on the major catalytic components, the ATPases (Brg1/hBrm, hISWI, and Mi-2) [15,16]. Recently, the ligand-induced transactivation function of VDR in vitro has been shown to require a SWI2/SNF2-type chromatin remodeling complex containing pBAF180 [17].

III. PURIFICATION AND IDENTIFICATION OF WILLIAMS SYNDROME TRANSCRIPTION FACTOR AS A VITAMIN D RECEPTOR INTERACTANT To search for novel co-regulatory complexes that interact with VDR, HeLa cell nuclear extracts were incubated with a chimeric VDR-hinge-ligand binding domain region (VDR-DEF) fused to glutathione-Stransferase (GST) in the presence or absence of 1,25(OH)2D3 (Fig. 3A). Proteins that interacted with VDR-DEF were separated by SDS-PAGE and silver stained (Fig. 3B). Mass spectrometry coupled with the apparent molecular weights of the different proteins

GST

P11

kDa

HNE

HeLa nuclear extract (HNE)

VDR(D3+)

B

VDR(D3−)

A

TRRAP/PAF400 P11 Phosphocellulose column DRIP240/TRAP230 200 GST

DRIP205/TRAP220 WSTF TRAP130

116 GST–VDR(D3−)

GST–VDR(D3+)

WINAC

WINAC DRIP/TRAPs

307

DRIP100/TRAP100 97 DRIP70 66

WSTF TRAP220 TIF2 Brg1 hSNF2h

FIGURE 3 Purification and identification of human proteins interacting with 1,25(OH)2D3–unbound and bound VDR. (A) Purification scheme for VDR interacting proteins. The eluted fraction from a P11 phosphocellulose column was incubated with immobilized GST-VDR(DEF) in the absence or presence of 1,25(OH)2D3 (D3, 10−6 M). (B) Identification of ligand-independent and -dependent VDR interacting proteins. In the upper panel, fractions were subjected to SDS-PAGE, followed by silver staining. Total HeLa S3 nuclear extract [HNE] (lane 1), a fraction eluted from the P11 column [P11] (lane 2), fractions from GST [GST] (lane 3), and unliganded and liganded GST-VDR(DEF) columns [VDR(−);VDR(+)] (lanes 4 and 5) were examined by mass spectrometry. Identified proteins are indicated at the right side of the panel. The lower panel shows Western blot analysis using specific antibodies shown in the panel.

308

SHIGEAKI KATO, RYOJI FUJIKI, AND HIROCHIKA KITAGAWA

associated with the ligand activated VDR-DEF led to the identification of several known components of the DRIP/TRAP/SMCC complex (Fig. 3B) that were in agreement with previous observations [12,14]. One ligand-independent VDR-specific interactant turned out as the Williams syndrome transcription factor (WSTF)/ WBSCR9/BAZ1B [18,19]. WSTF is a candidate gene whose product is potentially responsible for diverse WS disorders [18,20]. WSTF is highly homologous to hACF1 as one of the Bromodomain-adjacent-to-Zn finger (BAZ) family proteins [19], and hACF1 is a partner of hSNF2h (a Drosophila ISWI homolog), which forms well-characterized ISWI-based chromatin remodeling complexes (see Fig. 2) [21]. These facts raise the possibility that a WSTF-containing complex is an ATPdependent chromatin complex.

than 670 kDa contained WSTF, indicating that WSTF forms a stable nuclear complex. With mass fingerprinting, we identified all the components of the purified complex containing WSTF (Fig. 4C) and designated this complex as WINAC (WSTF including nucleosome assembly complex) [22]. WINAC consists of at least 13 components, but unexpectedly, does not contain hSNF2h as an ATPase subunit (Fig. 4C). Rather, the SWI/SNF type ATPases (Brg1 and hBrm) and several BAF components are shared with the SWI2/SNF2-based complexes [9]. Interestingly, WINAC appears to harbor three additional components (TopoIIb, FACTp140, and CAF-1p150) [23–25], which have not been observed in any of the known ATP-dependent chromatin remodeling complexes, but are supposed to be involved in other nuclear processes such as DNA repair and replication.

IV. PURIFICATION AND IDENTIFICATION OF A NOVEL WILLIAMS SYNDROME TRANSCRIPTION FACTOR COMPLEX ASSOCIATING WITH VITAMIN D RECEPTOR

V. WINAC IS A NOVEL MULTIFUNCTIONAL ATP-DEPENDENT CHROMATIN REMODELING COMPLEX THAT REARRANGES A NUCLEOSOME ARRAY AROUND A VITAMIN D RESPONSIVE ELEMENT IN VITRO

WSTF-containing complexes were purified by sequential columns (Fig. 4A) and fractionated on glycerol density gradients (Fig. 4B, upper panel). The fractionated complex with a molecular weight of greater

A

B Nuclear extract from f–WSTF cells GST

An ATP-dependent chromatin assembly reaction is clearly induced by WINAC as assessed by a standard micrococcal nuclease assay (Fig. 5A), indicating that

C kDa 200 116

BAF250 Brg1/hBrm TopoIIb WSTF

97 GST–VDR(D3−)

Glycerol density gradient

66

BAF170 BAF155 FACT p140 CAF–1 p150

WSTF Brg1

aFLAG M2 resin

WINAC

VDR GST–VDR BAF60a BAF57 BAF53 Ini1 WINAC

FIGURE 4 Purification and identification of a human WSTF-containing multiprotein complex “WINAC”. (A) Purification scheme of WINAC from MCF7 stable transformants. (B) Fractionation of purified complexes on glycerol density gradient. In the lower panel, Western blot analysis of each fraction using specific antibodies is shown. (C) The purified complex was subjected to SDS-PAGE, followed by silver staining and identified by mass spectrometry (indicated in the left of the panel).

309

CHAPTER 17 Promoter Targeting of Vitamin D Receptor through a Chromatin Remodeling Complex

A

ATP WINAC

B

+ − + + − +

Proximal probe

Distal probe

C

Proximal Probe

GAL–VDR −

+

−

+

−

+

−

+

GAL–ERα

+

+

−

−

−

−

+

+

−

−

+

+

GAL–PPAR −

−

+

+

−

+

−

+

WINAC

WINAC

FIGURE 5 WINAC as an ATP-dependent chromatin remodeling complex. (A) Chromatin reconstitution activity of WINAC. The reacted samples were subjected to partial micrococcal nuclease digestion. The molecular mass marker is the 200-bp ladder. (B) Chromatin disruption by WINAC is specifically VDR dependent. Oligonucleotide probe corresponds to either a sequence between the GAL4 sites and the RNA start site (Proximal Probe) or 900 bp upstream of the start site (Distal Probe). (C) WINAC is unable to affect the chromatin architecture through other nuclear receptors (ER and PPAR).

Brg1/hBrm in WINAC serves as an ATPase for this ATP-dependent chromatin remodeling process. WINAC appeared to have a chromatin assembly activity (data not shown) similar to RSF [22,26]. We then examined the ability of WINAC to disrupt nucleosome arrays through VDR bound VDRE since the known ATP-dependent chromatin remodeling complexes are able to recognize a nucleosomal array around binding sites of a sequence specific regulator [27]. Disruption of the nucleosome arrays in the vicinity of the GAL4 binding site was induced only when both VDR and WINAC were present (Fig. 5B), while other regions appeared unaffected in the nucleosome array. Reflecting VDR-specific nucleosome disruption by WINAC (among several tested receptors) (Fig. 5C), WINAC potentiated ligand-induced transactivation in vitro only for VDR, but not for either ERα or PPARγ (data not shown)[22].

VI. WILLIAMS SYNDROME TRANSCRIPTION FACTOR COACTIVATED THE LIGAND-INDUCED TRANSACTIVATION FUNCTION OF VITAMIN D RECEPTOR In a reporter assay, 1,25(OH)2D3 (10−9 M) was effective in inducing VDR AF-2 transactivation function, and WSTF coactivated this ligand-induced AF-2

function of VDR, but not ERα (Fig. 6A). Coactivation by either Brg1 or hBrm was expectedly detected in VDR and ERα (Fig. 6A) [28,29]; however, such coactivator-like activity of WSTF was found only for VDR, but not for ERα, even in the presence of Brg1/hBrm (Fig. 6A). ChIP analysis of the positive VDRE in the human 25-hydroxyvitamin D3 24-hydroxylase gene promoter revealed that associations of VDR and the WINAC components with the promoter did not require ligand binding. In contrast, ligand binding to VDR appeared to be a prerequiste for occupancy of known coactivators TRAP220 and TIF2 on the promoter (Fig. 6B), together with ligand-induced histone H4 acetylation (data not shown) [22] though ligand-induced TRAP220 and TIF2 occupancy was cyclic (data not shown), as already described [28]. Such ligand-dependent and -independent recruitment of factors to the promoter and histone modification were robustly attenuated by WSTF-RNAi expression (Fig. 6B). As the VDR/RXR heterodimer also represses transcription in a ligand-dependent manner through a negative VDRE (nVDRE), the action of WSTF in ligandinduced transrepression was examined in a naturally occurring nVDRE in the human 25-hydroxyvitamin D3 1α-hydroxylase [1 (OH)ase gene promoter [30]. ChIP analysis revealed that VDR and WINAC appear to bind to the nVDRE in a ligand-independent manner, while cyclic ligand-induced recruitments of N-CoR and HDAC2 were

310

SHIGEAKI KATO, RYOJI FUJIKI, AND HIROCHIKA KITAGAWA

B Transcription activity (fold)

A GAL4-VDR-DEF

WSTF

− − − − − +

− − − − +

− − − − +

− − − − − − + − − − − − − + + +

+ − + − − +

− + + − − +

− − − + − +

− + − + − +

− − − − + +

− + − − + +

− − + − − + − − − − −

WSTF-RNAi − − − + − − + − + − + Brg1

− − − − + + + − − − −

TRAP220

− − − − − − − + + − −

TIF2 E2

− − − − − − − − − + + − + + + + + + + + + +

Input

Input

VDR

VDR

fWSTF

fWSTF

Brg1

Brg1

TRAP220

NCoR

TIF2

HDAC2

hSNF2H

hSNF2H

D3

−

+

−

+

+

D3

−

+

−

+

+

WSTF-RNAi

−

−

−

−

+

WSTF-RNAi

−

−

−

−

+

D GAL4-ERα-DEF

1α(OH)ase gene promoter negative distal VDRE

Transcription activity (fold)

− − − − − −

Transcription activity (fold)

WSTF WSTF-RNAi Brg1 TRAP220 TIF2 D3

C 24(OH)ase gene promoter positive distal VDRE

WSTF WSTF-RNAi Brg1 NCoR HDAC2 D3

− − − − − −

− − − − − +

− − − − +

− − − − +

− − − − − − + − − − − − − + + +

+ − + − − +

− + + − − +

− − − + − +

− + − + − +

− − − − + +

− + − − + +

FIGURE 6

Ligand-dependent promoter targeting of co-regulators through WINAC-VDR association. (A) VDR-specific facilitation of coactivator accessibility by WINAC. MCF7 cells were transfected with the expression vectors of a luciferase reporter plasmid containing the GAL4 upstream activation sequence (UAS) [17mer(×2)] driven by the β-globin promoter (0.5 µg). PML-CMV (2 ng), GAL4-DBD-VDR-DEF (0.2 µg), GAL4-DBD-ERa-DEF (0.2 µg), pDNA3-FLAG-WSTF (+; 0.1 µg: ++; 0.3 µg), pSV-Brg1 (0.2 µg), pSV-hBrm (0.2 µg), pcDNA3-TRAP220 (0.3 µg), pcDNA3-TIF2 (0.3 µg), siRNA (+; 0.1 µg: ++; 0.2 µg) of WSTF-RNAi, or control RNAi or their combinations were transfected as indicated in the panels in the absence or presence of ligand (10−9 M). Bars in each graph show the fold change in luciferase activity relative to the activity of the receptor transactivation in the presence of ligand. (B, C) ChIP analysis on the 24(OH)ase promoter and 1α(OH)ase promoter of WSTF stable transformants. Soluble chromatin was prepared from WSTF stable transformants treated with 1,25(OH)2D3 (10−9 M) for 45 min and immunoprecipitated with indicated antibodies. (D) The coregulator-like actions of WSTF on the naturally occurring negative vitamin D response elements. MCF7 cells were transfected with the expression vectors of either the luciferase reporter plasmid containing a human 1α(OH)ase promoter containing a negative VDRE and the factors shown in (A) or together with pcDNA3-N-CoR (0.3 µg), pcDNA3-HDAC2 (0.3 µg).

observed (Fig. 6C). Ligand-dependent repression was dependent on the expression levels of WSTF (Fig. 6D). Thus, it is likely that WINAC association with VDR facilitates targeting of a putative co-repressor complex to the nVDRE. Thus, these findings indicate that WINAC rearranges the nucleosome array around the positive and negative VDREs, thereby facilitating accessibility of the coregulatory complexes accessible to VDR for further transcription controls (Fig. 7).

VII. MOLECULAR MECHANISM OF VDR PROMOTER TARGETING OF VITAMIN D RECEPTOR BY WINAC, AND COOPERATIVE WINAC FUNCTION WITH THE CO-REGULATOR COMPLEXES Both WSTF and the ATPase subunits coactivated the ligand-induced VDR transactivation, a finding similar

311

CHAPTER 17 Promoter Targeting of Vitamin D Receptor through a Chromatin Remodeling Complex

WINAC WSTF

VDR RXR

VDR RXR

HDAC complexes HAT complexes

HDAC1/2

WINAC

mSin3

PCAF complex

NCoR/ WSTF SMRT VDR RXR

Gene activation

WSTF

VDR RXR

Target gene

gene represseion

WINAC

p300

p160 family protein

1α,25(OH)2D3

Target gene

FIGURE 7

Model of WINAC action in chromatin reorganization. WINAC mediates the recruitment of unliganded VDR to VDR target sites in promoters, whereas subsequent binding of co-regulators requires ligand binding. This recruitment order exemplifies that an interaction of a sequence-specific regulator with a chromatin remodeling complex can organize nucleosomal arrays at specific local sites in order to make promoters accessible for co-regulators.

to the reported coactivator-like actions of the SWI2/SNF2-type complex components, ATPases and BAF57, for the ligand-induced ERα transactivation [28,29]. Notably, VDR coactivation by the liganddependent NR coactivators (TIF2 and TRAP220) was abrogated by WSTF-RNAi expression. However, such WSTF coactivator-like actions were not observed for ERα and the other receptors tested (data not shown), nor were they reduced by WSTF-RNAi expression, supporting the observed direct and selective interaction of WSTF with VDR, though it remains unclear whether promoter targeting of VDR requires only WINAC or the other chromatin remodeling complexes. More remarkably, we found that WSTF could potentiate the ligand-induced transrepression of VDR on the 1α(OH)ase negative VDRE (Fig. 6D), where ablation of endogenous WSTF by RNAi expression led to a significant reduction in ligand-induced co-repressor recruitment (Fig. 6C). Thus, ligand-independent association of WINAC and VDR in VDR target promoters appears to facilitate local nucleosomal array accessibility for ligand-dependent co-regulators, following histone tail modifications by the recruited co-regulator complexes [11] (Fig. 7). From our ChIP analysis, VDR appears to be selectively targeted through WINAC to the promoters in a ligand-independent fashion or following recruitment of co-regulator complexes. This ligand-independent association of VDR with the target promoter is surprisingly distinct from the ligand-induced promoter targeting of many other steroid receptors [28,29]. It is possible that other nonsteroidal receptors such as

RAR, RXR, and TR may associate with their target promoters in the absence of hormone, and as-yetunidentified chromatin remodeling complexes may assist the promoter targeting. As HAT and HDAC complexes appear not to associate with unliganded VDR bound upon the tested promoters, WINAC targeting to the VDR target promoters does not appear to require specific histone tail modifications by other co-regulators. Thus, it is possible that WINAC binding to promoters facilitates VDR recognition and specific binding to VDREs, through nucleosomal mobilization by WINAC, presumably cooperating with the other chromatin complexes [17]. Alternatively, once VDR binds to VDREs during nonspecific chromatin remodeling, WINAC might be recruited to VDR on specific promoters to engage in local nucleosome reorganization. The latter possibility coincides well with a recent report regarding a sequence-specific regulator, SATB1 [16]. As a result of WINAC recruitment, the local chromatin structure near VDREs may transit into an active chromosomal state that appears competent for receipt of both coactivator and co-repressor complexes dependent on the VDRE sequences and the tertiary positions of DNA-bound VDR. This scheme may not be considered, however, with VDREs composed of unrelated DNA sequences with the canonical VDREs. Regardless, our hypothesis is not consistent with current understanding that the chromatin remodeling complexes are recruited only after acetylation/ deacetylation of histone tails by the coregulatory complexes [11]. However, the orders of the complex targetings are supposed to be dependent on regulator type

312

SHIGEAKI KATO, RYOJI FUJIKI, AND HIROCHIKA KITAGAWA

and promoter context [31,32]. This point should be addressed with VDR together with WINAC.

activator interactions with the ATM-related Tra1 subunit. Science 292:2333–2337. Fyodorov DV, Kadonaga JT 2001 The many faces of chromatin remodeling: SWItching beyond transcription. Cell 106:523–525. Yasui D, Miyano M, Cai S, Varga-Weisz P, Kohwi-Shigematsu T 2002 SATB1 targets chromatin remodeling to regulate genes over long distances. Nature 419:641–645. Lemon B, Inouye C, King DS, Tjian R 2001 Selectivity of chromatin-remodeling cofactors for ligand-activated transcription. Nature 414:924–928. Lu X, Meng X, Morris CA, Keating MT 1998 A novel human gene, WSTF, is deleted in Williams syndrome. Genomics 54:241–249. Jones MH, Hamana N, Nezu J, Shimane M 2000 A novel family of bromodomain genes. Genomics 63:40–45. Peoples RJ, Cisco MJ, Kaplan P, Francke U 1998 Identification of the WBSCR9 gene, encoding a novel transcriptional regulator, in the Williams–Beuren syndrome deletion at 7q11.23. Cytogenet Cell Genet 82:238–246. Poot RA, Dellaire G, Hulsmann BB, Grimaldi MA, Corona DF, Becker PB, Bickmore WA, Varga-Weisz PD 2000 HuCHRAC, a human ISWI chromatin remodelling complex contains hACF1 and two novel histone-fold proteins. EMBO J 19: 3377–3387. Kitagawa H, Fujiki R, Yoshimura K, Mezaki Y, Uematus Y, Matsui D, Ogawa S, Unno K, Okubo M, Tokita A, Nakagawa T, Ito T, Ishimi Y, Nagasawa H, Matsumoto T, Yanagisawa J, Kato S 2003 The chromatin remodeling complex WINAC targets a nuclear receptor to promoters and is impaired in Williams Syndrome. Cell 113:1–13. Smith S, Stillman B 1989 Purification and characterization of CAF-I, a human cell factor required for chromatin assembly during DNA replication in vitro. Cell 58:15–25. Varga-Weisz PD, Wilm M, Bonte E, Dumas K, Mann M, Becker PB 1997 Chromatin-remodelling factor CHRAC contains the ATPases ISWI and topoisomerase II. Nature 388:598–602. LeRoy G, Orphanides G, Lane WS, Reinberg D 1998 Requirement of RSF and FACT for transcription of chromatin templates in vitro. Science 282:1900–1904. Loyola A, LeRoy G, Wang YH, Reinberg D 2001 Reconstitution of recombinant chromatin establishes a requirement for histonetail modifications during chromatin assembly and transcription. Genes Dev 15:2837–2851. Ito T, Bulger M, Pazin MJ, Kobayashi R, Kadonaga JT 1997 ACF, an ISWI-containing and ATP-utilizing chromatin assembly and remodeling factor. Cell 90:145–155. Shang Y, Hu X, DiRenzo J, Lazar MA, Brown M 2000 Cofactor dynamics and sufficiency in estrogen receptor-regulated transcription. Cell 103:843–852. Belandia B, Orford RL, Hurst HC, Parker MG 2002 Targeting of SWI/SNF chromatin remodelling complexes to estrogenresponsive genes. EMBO J 21:4094–4103. Murayama A, Takeyama K, Kitanaka S, Kodera Y, Hosoya T, Kato S 1998 The promoter of the human 25-hydroxyvitamin D3 1α-hydroxylase gene confers positive and negative responsiveness to PTH, calcitonin, and 1α,25(OH)2D3. Biochem Biophys Res Commun 249:11–16. Lomvardas S, Thanos D 2002 Modifying gene expression programs by altering core promoter chromatin architecture. Cell 110:261–271. Soutoglou E, Talianidis I 2002 Coordination of PIC assembly and chromatin remodeling during differentiation-induced gene activation. Science 295:1901–1904.

Acknowledgment We sincerely thank all collaborators for the WINAC projects, and the laboratory members for helpful discussions and technical support. We are grateful to Miss Y. Nagasawa for preparation of the manuscript.

References

15. 16. 17. 18. 19. 20.

1. Deluca HF 1986 The metabolism and functions of vitamin D. Adv Exp Med Biol 196:361–375. 2. Bouillon R, Okamura WH, Norman AW 1995 Structurefunction relationships in the vitamin D endocrine system. Endocr Rev 16:200–257. 3. Walters MR 1992 Newly identified actions of the vitamin D endocrine system. Endocr Rev 13:719–764. 4. Mangelsdorf DJ, Thummel C, Beato M, Herrlich P, Schutz G, Umesono K, Blumberg B, Kastner P, Mark M, Chambon P, et al. 1995 The nuclear receptor superfamily: the second decade. Cell 83:835–839. 5. Huges MR, Malloy PJ, Kieback DG, Kesterson RA, Pike JW, Feldman D, O’Malley BW 1988 Point mutations in the human vitamin D receptor associated with hypocalcemic rickets. Science 242:1702–1705. 6. Haussler MR, Whitfield GK, Haussler CA, Hsieh JC, Thompson PD, Selznick SH, Dominguez CE, Jurutka PW 1988 The nuclear vitamin D receptor: biological and molecular regulatory properties revealed. J Bone Miner Res 13: 325–349. 7. Rachez C, Lemon D, Suldan Z, Bromleigh V, Gamble M, Naar M, Erdjument H, Tempst P, Freedman P 1999 Ligand-dependent transcription activation by nuclear receptors requires the DRIP complex. Nature 398:824–828. 8. Glass CK, Rosenfeld MG 2000 The co-regulator exchange in transcriptional functions of nuclear receptors. Genes Dev 14:121–141. 9. Narlikar GJ, Fan HY, Kingston RE 2002 Cooperation between complexes that regulate chromatin structure and transcription. Cell 108:475–487. 10. Emerson BM 2002 Specificity of gene regulation. Cell 109:267–270. 11. Hassan AH, Prochasson P, Neely KE, Galasinski SC, Chandy M, Carrozza MJ, Workman, JL 2002 Function and selectivity of bromodomains in anchoring chromatinmodifying complexes to promoter nucleosomes. Cell 111: 369–379. 12. Yanagisawa J, Kitagawa H, Yanagida M, Wada O, Ogawa S, Nakagomi M, Oishi H, Yamamoto Y, Nagasawa H, McMahon SB, Cole MD, Tola L, Takahashi N, Kato S 2002 Nuclear receptor function requires a TFTC-type histone acetyltransferase complex. Mol Cell 9:553–562. 13. Lanz RB, McKenna NJ, Onate SA, Albrecht U, Wong J, Tsai SY, Tsai MJ, O’Malley BW 1999 A steroid receptor coactivator, SRA, functions as an RNA and is present in an SRC-1 complex. Cell 97:17–27. 14. Brown CE, Howe L, Sousa K, Alley SC, Carrozza MJ, Tan S, Workman JL 2001 Recruitment of HAT complexes by direct

21.

22.

23. 24.

25. 26.

27. 28. 29. 30.

31. 32.

CHAPTER 18

Molecular Basis of the Diversity of Vitamin D Target Genes CARSTEN CARLBERG, THOMAS W. DUNLOP, CHRISTIAN FRANK, AND SAMI VÄISÄNEN Department of Biochemistry, University of Kuopio, Kuopio, Finland

I. Molecular Basis of the Genomic Actions of 1,25(OH)2D3 II. Classical Vitamin D–Receptor Binding Sites

III. Complex Vitamin D–Receptor Binding Sites IV. Conclusion References

I. MOLECULAR BASIS OF THE GENOMIC ACTIONS OF 1,25(OH)2D3

the p160-family, such as SRC-1, TIF2, and RAC3 [8]. These CoAs link the ligand-activated VDR to enzymes displaying histone acetyltransferase (HAT) activity that cause chromatin opening. Subsequently, ligand-activated VDR changes rapidly from the CoAs of the p160-family and those of the DRIP/TRAP family. The latter are part of a mediator complex of approximately 15 proteins that build a bridge to the basal transcription machinery [9]. In this way ligand-activated VDR fulfills two tasks, opening chromatin and activating transcription. The LBD of the VDR can be stabilized by 1,25(OH)2D3 or its analogs in an agonistic, antagonistic, or inverse agonistic conformation [10]. The position of helix 12 is the most critically important feature of these conformations, because it determines the distance that separates the charge clamp amino acids K246 in helix 3 and E420 in helix 12 that are both essential for VDR– CoA interaction (see Chapters 13–15). Most VDR ligands have been identified as agonists and only a few as pure or partial antagonists [11]. Two side-chain analogs, such as Gemini and its derivatives, have conditional inverse agonistic properties [12]. Antagonists induce CoR dissociation from the VDR but completely or partially prevent CoA interaction and thus transactivation [13]. At supramolar CoR concentrations inverse agonists actively recruit CoRs to the VDR and thus mediate repression of 1,25(OH)2D3 target genes. These ligand-triggered protein–protein interactions are the central molecular events of nuclear 1,25(OH)2D3 signaling.

A. Ligand-Triggered Protein–Protein Interactions of the VDR The vitamin D3 receptor (VDR) is the only nuclear protein that binds with high affinity to the biologically most active vitamin D metabolite, 1α,25-dihydroxyvitamin D3 (1,25(OH)2D3) [1]. The VDR is one of 11 members of the nuclear receptor (NR) superfamily that function as classical endocrine receptors. These include the receptors for the nuclear hormones retinoic acid, thyroid hormone, estradiol, progesterone, testosterone, cortisol, and aldosterol, which bind their specific ligand with a Kd of 1 nM or less [2]. Like most members of the NR superfamily, the VDR contains two zinc finger structures forming a characteristic DNA-binding domain (DBD) of 66 amino acids [3] and a carboxyterminal ligand-binding domain (LBD) of approximately 300 amino acids formed by 12 α-helices [4] (see Chapters 11–15). Ligand binding causes a conformational change within the LBD, whereby helix 12, the most carboxy-terminal α-helix, closes the ligand-binding pocket via a “mousetrap-like” intramolecular folding [5]. The LBD is also involved in a variety of interactions with nuclear proteins, such as other members of the NR superfamily, coactivator (CoA) and co-repressor (CoR) proteins [6]. CoR proteins, such as NCoR, SMRT, and Alien, link nonliganded, DNA-bound VDR to enzymes with histone deacetylase activity that cause chromatin condensation [7]. This provides VDR with intrinsic repressive properties comparable to those of retinoic acid and thyroid hormone receptors (RARs and T3Rs). The conformational change within VDR’s LBD after binding of 1,25(OH)2D3 or one of its agonistic analogs results in replacement of CoR by a CoA protein of VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

B. DNA Binding of the VDR An essential prerequisite for a direct modulation of transcription via 1,25(OH)2D3-triggered protein– protein interactions is the location of activated VDR close to the basal transcriptional machinery. This is achieved Copyright © 2005, Elsevier, Inc. All rights reserved.

314

CARSTEN CARLBERG, THOMAS W. DUNLOP, CHRISTIAN FRANK, AND SAMI VÄISÄNEN

through the specific binding of the VDR to a 1,25(OH)2D3 response element (VDRE) in the regulatory region of a primary 1,25(OH)2D3 responding gene [14]. The DBD of the VDR contacts the major grove of a hexameric sequence, referred to as core binding motif, with the consensus sequence RGKTSA (R = A or G, K = G or T, S = C or G). The affinity of monomeric VDR to a single core binding motif is not sufficient for the formation of a stable protein–DNA complex and thus VDR requires formation of homo- and/or heterodimeric complexes with a partner NR in order to allow efficient DNA binding [15]. In most cases the heterodimeric partner of VDR is the retinoid X receptor (RXR), another NR family member. The protein–DNA complex of a VDR-RXR heterodimer sitting on a VDRE therefore can be considered as a molecular switch for primary 1,25(OH)2D3 responding genes. Most NRs are able to dimerize in solution via their LBDs, but the DBDs dimerize only in the presence of DNA [16]. The DBD and the LBD of all NRs are linked by a hinge region of 35 to 50 amino acid residues that form a long α-helical structure according to the crystal structure of the DNA-bound T3R-RXR heterodimer [17]. The loop between this α-helix and the second zinc finger of the DBD contains a short six-amino-acid residue region, referred to as a T-box, which has been suggested to provide a dimerization interface for the interaction with the DBD of RXR [18]. Steric constrains allow dimerization of DBDs only on response elements (REs) with properly spaced core binding motifs. Modeling of the DBDs of VDR and RXR on DNA [17] suggested that an asymmetric arrangement, i.e., head-to-tail, as a direct repeat with three intervening nucleotides (DR3) provides the most efficient interface of the core DBDs. This fits with the 3-4-5 rule of Umesono et al. [19], in which VDR-RXR heterodimers should show optimal binding to DR3-type REs, whereas other NRs prefer DR4-type REs [e.g., T3R, constitutive androstane receptor (CAR) and pregnane X receptor (PXR)] and DR5-type REs (e.g., RARs). On DR3-, DR4-, and DR5-type REs, the different heterodimers bind with the same polarity, in which RXR always binds to the upstream hexamer and the partner receptor, e.g., VDR, to the downstream hexamer [20,21]. This specific and directed dimerization of the DBDs appears to be the major discriminative parameter for selective RE recognition.

II. CLASSICAL VITAMIN D–RECEPTOR BINDING SITES A. Classification of DR3-type REs Numerous studies (e.g., [15,22]) have confirmed Umesono’s suggestion [19] that VDR binds well to

DR3-type REs. Most of these studies also demonstrated that VDR preferentially or even exclusively forms heterodimers with RXR on these REs. DR3-type REs are therefore widely accepted as the classical structure of a VDRE. Every transcriptionally responsive primary 1,25(OH)2D3 target gene has to contain at least one VDRE in its promoter region and these VDREs are generally located relatively close to the transcription start site (TSS) of these genes. It is assumed that matrix attachment regions (MARs) subdivide genomic DNA into units of an average length of 100 kB containing the coding region of at least one gene [23]. DNA looping should be able to bring any DNA site within the same chromatin unit close to the basal transcriptional machinery that is assembled on the TSS (Fig. 1). This model suggests that also very distant sequences can serve as VDREs and that even sequences downstream of the TSS could serve as functional VDR binding sites. The list of the presently known natural VDREs (Table I) indicates that most of them have a DR3-type structure and are located within the first 1000 bp of promoter sequence upstream of the TSS. In addition, this VDRE list suggests that the consensus VDR core binding motif is RGKTSA, although some natural hexameric sequences show a significant deviation from this consensus sequence (Table I). However, one has to take into account that all these VDREs had been identified before the genomes of Homo sapiens and other species had been sequenced, and only limited promoter sequences were available. Moreover, only a very few of these VDREs, such as that of the rat osteocalcin gene, are understood in their promoter context, i.e., in the context of chromatin organization and flanking binding sites for other transcription factors (TFs). Therefore, the present VDRE list (Table I) has to be taken with some precaution. Results from in silico promoter screening and a more detailed understanding of chromatin organization of primary 1,25(OH)2D3 target genes will revise this list soon. On the basis of presently published data, the strongest VDR-RXR heterodimer binding DR3-type VDREs has been identified within the rat atrial natriuretic factor (ANF) promoter [24], the mouse and pig osteopontin promoter [25,26], and the chicken carbonic anhydrase II promoter [27]. These four elements were categorized into class I (see Table I). The DR3-type VDREs of the human and rat 24-hydroxylase (CYP24) promoter [28,29], the human Na+-dependent inorganic phosphate transporter type II promoter [30], the rat osteocalcin promoter [31], the human parathyroid hormone (PTH) promoter [32], and the first VDRE of the rat PTH related peptide (PTHrP) promoter [33] demonstrated only 6–30% of the binding strength of the rat ANF VDRE and were grouped into class II. The binding of VDR-RXR heterodimers to the ten

315

CHAPTER 18 Diversity of Vitamin D Target Genes

VDRE R X R

MAR

V D R

TSS Basal transcriptional machinery

Coding region

? V D R MAR

R X R

VDRE

FIGURE 1

Schematic structure of a chromatin unit. A chromatin unit is a region between two MARs and often contains only one gene. DNA looping should permit that every DNA sequence within the same chromatin unit to be located near the basal transcriptional machinery. The occurrence of VDREs downstream of a coding region has not yet been proven experimentally.

DR3-type VDREs in classes I and II was found to be enhanced by 1,25(OH)2D3 by a factor of 2 to 5 [34]. In contrast, the second VDRE of the rat PTHrP promoter [35], the rat calbindin D9k promoter [36] and the quail slow myosin heavy chain promoter [37], the human growth hormone promoter [38], the chicken integrin β3 promoter [39], the chicken PTH promoter [40], and the human p21WAF1/CIP1 promoter [41] were grouped together into class III, as they showed less than 5% of the affinity for VDR-RXR heterodimers compared to the rat ANF VDRE and no significant ligand inducibility. According to these stringent in vitro criteria, the core sequences of the class III members cannot be considered as functional VDREs. The in vitro binding affinity of VDR-RXR heterodimers to VDREs was shown to be proportional to their in vivo functionality in transiently transfected cells, i.e., to mediate in a heterologous promoter context induction of reporter gene activity [21,42]. In this assay system class III VDREs did not show any in vivo functionality. The hexameric sequences of the class III VDREs show a significant deviation from the RGKTSA consensus and explains their relatively low ability of complex formation with VDR-RXR heterodimers. However, it is possible that class III VDREs may gain responsiveness to 1,25(OH)2D3 in their natural promoter context through the help of flanking partner proteins. Moreover, the functionality of a 1,25(OH)2D3 responding gene will also

depend on a potential cooperative action of two or more VDREs, such as in the case of the CYP24 gene [43].

B. Do DR3-Type REs Differ in Their Function? Since the pleiotropic physiological effects of 1,25(OH)2D3 are based in the final analysis on transcriptional regulation of primary 1,25(OH)2D3 responding genes, the genes should be explained through activation of VDR-RXR heterodimers bound to DR3-type VDREs. Therefore, there have been attempts to explain the various effects of 1,25(OH)2D3 by a multiplicity of 1,25(OH)2D3 signaling pathways that are based on DR3-type VDREs [44]. This leads to the question of whether each DR3-type VDRE has an individual functionality that may explain the specific effects of 1,25(OH)2D3 or whether all DR3-type VDREs function in the same way. The affinity of the known natural DR3-type VDREs for VDR-RXR heterodimers in vitro (at a fixed protein– DNA ratio) seems to be their major discriminating parameter, which allowed the grouping of the VDREs into classes as discussed earlier (Table I). VDR-RXR heterodimers appear to form identical complexes on the 10 VDREs of classes I and II, since indistinguishable VDR-RXR heterodimer conformations were observed on these VDREs [34]. This is in contrast to a report that VDR-RXR heterodimers take different conformations

316

CARSTEN CARLBERG, THOMAS W. DUNLOP, CHRISTIAN FRANK, AND SAMI VÄISÄNEN

TABLE I Natural VDRCS Gene

Species

Class I DR3-type VDREs ANF Osteopontin Osteopontin Carbonic anhydrase II

Rat Mouse Pig Chicken

AGAGGTCATGAAGGACA AAGGTTCACGAGGTTCA ATGGGTCATATGGTTCA GAAGGGCATGGAGTTCG

−907 −759 −2261 −62

[24] [26] [25] [27]

Human Human Human

CGAGGTCAGCGAGGGCG GGAGTTCACCGGGTGTG CAGGGGCAGCAAGGGCA

−171 −291 −1977

[28] [28] [30]

Human Human Rat Rat Rat Rat

ATGGTTCAAAGCAGACA CCGGGTGAACGGGGGCA CTGGGTGAATGAGGACA CGAGGTGAGTGAGGGCG AGGGTTCAGCGGGTGCG TAAGGTTACTCAGTGAA

−122 −500 −457 −152 −259 −805

[32] [52] [31] [29] [29] [35]

Class III DR3-type VDREs p21 Growth hormone Bone sialo protein PTH related peptide Calbindin D9k Integrin β3 PTH Slow myosin heavy chain

Human Human Rat Rat Rat Chicken Chicken Quail

GTAGGGAGATTGGTTCA TGGGGTCAACAGTGGGA GAAGGGTTTATAGGTCA AGGGTGGAGAGGGGTGA GAGGGTGTCGGAAGCCC GCGAGGCAGAAGGGAGA GAGGGTCAGGAGGGTGT GAAGGACAAAGAGGGGA

−779 −59 −30 −1107 −490 −772 −76 −801

[41] [38] [78] [33] [36] [39] [40] [37]

DR4-type VDREs Pit-1 Calbindin D28k

Rat Mouse

GAAGTTCAGCGAAGTTCA CTGGGGGATGTGAGGAGA

−683 −200

[48] [95]

DR6-type VDREs Osteocalcin Phospholipase C-γ1 CYP24 Fibronectin

Human Human Rat Mouse

TTTGGTGACTCACCGGGTGA GCAGGTCAGACCACTGGACA CGGGTCGAGCCCAGGGTTCA CCGGGTGACGTCACGGGGTA

−514 −805 −231 −152

[51] [56] [55] [54]

ER9-type VDREs Calbindin D9k Osteocalcin c-fos p21

Human Rat Mouse Mouse

TGCCCTTCCTTATGGGGTTCA TGCACTGGGTGAATGAGGACA TGACCCTGGGAACCGGGTCCA TGACCTGAAAGTGGAAGGTGA

−147 −461 −482 −3811

[21] [21] [57] [96]

Complex VDREs Osteocalcin Osteocalcin c-fos

Human Rat Mouse

TTTGGTGACTCACCGGGTGAACGGGGGCA TGCACTGGGTGAATGAGGACA AGGTGAAAGATGTATGCCA AGACGGGGGTTGAAAG

−514 −461 −178

[42] [21] [77]

Class II DR3-type VDREs CYP241 CYP242 Na+-dependent inorganic Phosphate transporter type II PTH Osteocalcin Osteocalcin CYP241 CYP242 PTH related peptide

Sequence

Position

Reference

The core sequence of the different types of VDREs and their position in relation to the TSS are indicated. Hexameric core binding motifs are in bold and deviations from the consensus sequence RGKTSA are underlined.

317

CHAPTER 18 Diversity of Vitamin D Target Genes

on the mouse osteopontin and the rat osteocalcin DR3type VDRE [44]. In that study, nuclear extracts from bone cells provided protein complexes on the two VDREs that were differentially recognized by antibodies. One could assume that the VDR-RXR heterodimers in these complexes interact with different types of cofactors as has been described for estrogen REs [45]. However, VDRRXR heterodimers formed on the different DR3-type VDREs display no significant differences in their interaction with a given CoA or CoR protein. The ligand sensitivity of VDR-RXR heterodimers bound to the different DR3-type VDREs showed no significant deviation from the average value of 0.1 nM. This suggests that the ligand-triggered VDR-RXRVDRE complex formation does not depend on ligand concentration or VDR conformation [34]. In addition, VDR-RXR heterodimers show no significant differences in their 1,25(OH)2D3-triggered interaction with CoA and CoR proteins. These findings suggest that heterodimers are able to differentiate between individual DR3-type VDREs on the basis of the different protein-DNA affinities. However, when the complexes are formed, they appear to function in exactly the same fashion. This means that at least comparative in vitro data cannot support any indications of multiple DR3type VDRE-mediated 1,25(OH)2D3 signaling pathways. Investigations of VDR-RXR heterodimers and their conformations have provided multiple new insights on the functionality of these molecular switches that may well explain their function in living cells [7,46]. This makes it likely that the lack of variation of the in vitro functionality of DR3-type VDREs (from positively as well as negatively regulated genes) can be transferred to the in vivo situation. This facilitates further investigation of the principles of DR3-type VDRE-mediated gene regulation, since an observation that has been made with one specific DR3-type VDRE can be generalized for the whole family of DR3-type VDREs. However, an explanation for the multiplicity of 1,25(OH)2D3 signaling has to be found at a different level.

C. Other DR-Type VDREs Since a variety of DR3-type VDREs seems unable to explain the pleiotropic physiological action of 1,25(OH)2D3 and the dissociated profile that some synthetic 1,25(OH)2D3 analogs display, other VDRE structures such as direct repeats with four and six intervening nucleotides (DR4 and DR6) or everted repeats with nine spacing nucleotides (ER9) may offer an alternative explanation. A comparison of the individual VDRE core sequences (Table I) with their classification according to the affinity for VDR-RXR heterodimers

suggests that the degree of deviation from the core binding motif consensus sequence RGKTSA [47] is proportional to the loss of in vitro functionality [34]. Interestingly, the DR4-type RE of the rat pit-1 gene [48], which contains two perfect core binding motifs, was found to be stronger than any known natural DR3-type VDRE [34]. Does this suggest that DR4-type REs are even better VDREs than DR3-type structures? One has to consider that a DR4-type REs is also recognized by the heterodimeric complexes of T3R, CAR, PXR, and other orphan NRs with RXR [49,50], whereas the same complexes bind to DR3-type REs less tightly than VDR-RXR heterodimers. The competitive situation on DR4-type REs may therefore be the reason why in vivo VDR-RXR heterodimers still prefer DR3-type REs. Moreover, VDR-RXR heterodimers bind to DR4-type REs in the same conformation as to DR3-type REs [50], i.e., there seems to be no differential action of VDR on these elements due to a differential complex formation with RXR. The VDRE of the human osteocalcin promoter was the first identified natural binding site for the VDR [51,52]. It was initially described as a DR6-type structure, but later a third cryptic hexamer was identified at a distance of three nucleotides, so that the whole VDRE is more likely a complex DR6/DR3-type RE (Table I). The DR6 part of the VDRE has been shown to bind VDR homodimers [15] and VDR-RAR heterodimers [53], whereas the DR3 part weakly binds VDR-RXR heterodimers. Other examples of a DR6-type VDREs that bind VDR homodimers and VDR-RAR heterodimers have been identified in the promoters of mouse fibronectin [54], rat CYP24 [55], and human phospholipase C [56]. Their functionality remains to be determined.

D. Everted Repeats Dimerization facilitates cooperative, high-affinity interaction of NRs, such as VDR and RXR, with specific hexameric core binding motifs. As schematically depicted in Fig. 2, VDR-RXR heterodimers bind to DR-type VDREs in a nonsymmetrical head-to-tail tandem arrangement. In contrast, like all palindromic sequences, ER-type VDREs are per se symmetric, since the heterodimeric partner receptors bind in a tail-to-tail arrangement. However, natural ER9-type REs were found to be sufficiently asymmetric in their core binding sequences, in order to allow a polarity-determined binding of heterodimers [21]. So far, ER9-type VDREs have been identified within the promoter regions of the human calbindin D9k gene [21], the mouse c-fos gene [57], and the rat osteocalcin gene [21]. Interestingly, the last VDRE shares one core binding motif with the

318

CARSTEN CARLBERG, THOMAS W. DUNLOP, CHRISTIAN FRANK, AND SAMI VÄISÄNEN

Direct repeats

Everted repeat 9

3 or 4

DBD

Hinge

LBD Helix 12

Helix 12

RXR

VDR

RXR

VDR

FIGURE 2 DNA complex formation of VDR-RXR heterodimers. The DNA binding of VDR-

RXR heterodimers to DR- or ER-type VDREs is schematically indicated. Some critical α-helices within the LBD of VDR and the DBDs of VDR and RXR are indicated by cylinders.

DR3-type VDRE identified by Demay et al. [31], and there appears to be a complex ER9/DR3-type RE. On DR3-type VDREs, both receptor DBDs are located at roughly the same side of the DNA (tilted by 51.4°), whereas on ER9-type VDREs the DBDs are at nearly opposite sites of the DNA (tilted by 154.3°) [18]. Moreover, the distance between the DBDs along the axis of the DNA is threefold greater on ER9-type VDREs than on DR3-type VDREs (Fig. 2). Computer modeling and experimental studies have both shown that the T-box of VDR contributes to the dimerization interface of the extended VDR DBD with the RXR DBD, i.e., on a DR3-type VDRE the VDR DBD directly contacts the RXR DBD, whereas a direct contact of both DBDs is not possible on an ER9-type VDRE. The lack of DBD dimerization on an ER9-type RE suggests that DNA-driven cooperativity between the partner DBDs is unlikely. In fact, the spacing of the two core binding motifs is less stringent than on DRtype REs, as ER7-, ER8-, and ER10-type structures are also able to bind dimeric VDR complexes [58,59]. However, the Kd value for the binding of VDR-RXR heterodimers to DR3- and ER9-type REs has been determined to be in a similar range of 0.5 to 1 nM [21]. According to our model of multiple 1,25(OH)2D3 signaling pathways [60], the pleiotropic function of 1,25(OH)2D3 is based on a variety of dimeric VDR complexes bound to different types of VDREs. The model assumes that each of these VDR-VDRE

complexes may be representative for a group of primary 1,25(OH)2D3 target genes that are involved in the regulation of a distinct proportion of the pleiotropic actions of this nuclear hormone. In support of the model, some 1,25(OH)2D3 analogs have shown the tendency to preferentially activate VDR-RXR heterodimers that are bound to ER9-type VDREs [61], whereas other analogs prefer DR3-type VDRE-bound VDR complexes [62]. The selective biological profile of the analog EB1089, i.e., having both potent antiproliferative potential and a reduced calcemic effect [63], was associated with a higher selectivity (approximately 15 times) to activate ER9-type VDREs compared to DR3-type VDREs [61]. Because of the relatively low number of known primary 1,25(OH)2D3 target genes with characterized VDREs, this idea has to be proven. However, the genes of mouse c-fos [57] and human and mouse p21WAF1/CIP1 [14] each contain an ER9-type VDRE in their regulatory regions. A VDRE-selective in vitro stabilization of VDR-RXR heterodimers was demonstrated by ligand-dependent gel shift assays and showed that EB1089 mediated the stabilization of VDR-RXR heterodimers on an ER9-type VDREs at approximately eightfold lower concentrations than on a DR3-type VDRE [64]. In contrast, the natural hormone 1,25(OH)2D3 showed no significant selectivity. Taken together, the selective activation of ER9-type VDREs by EB1089, i.e., the observation of promoter selectivity of a 1,25(OH)2D3 analog, seems to be based on the enhanced DNA binding affinity of

319

CHAPTER 18 Diversity of Vitamin D Target Genes

a subset of all VDR-RXR heterodimers to this type of VDRE.

E. Do Hexameric Core Binding Motifs Provide Sufficient Specificity for DNA Binding of VDR-RXR Heterodimers? It has been known for some time that some NRs, such as T3R, which can also bind as monomers to DNA, have a clear preference for certain 5′-flanking sequences to the hexameric core binding motif [65,66]. In addition, it had been demonstrated that in particular the dinucleotide sequence 5′-flanking of the downstream hexamer of a RE have a significant effect on the complex formation of NR heterodimers, such as RARRXR, T3R-RXR, and VDR-RXR [66–68]. This provides the sequence of the spacer between the hexameric motifs of natural REs with a critical role for determining the specific recognition of the RE and the regulation of the respective gene. The dinucleotide AG, as found, e.g., in front of the downstream motif of the DR4-type RE of the rat pit-1 gene, seems to be optimal for VDRRXR heterodimers [68]. This preference should be independent of the type of RE and apply to all VDRE types. The crystal structure of DNA-bound T3R-RXR heterodimers [17] demonstrated a contact of amino acids of the carboxy-terminal extension of the T3R-DBD with the two 5′-flanking nucleotides of the downstream core binding motif. Similar assumptions could be made for the VDR. Interestingly, variations of the dinucleotide sequence 5′-flanking to the upstream core binding motif also have a lower but still significant influence on complex formation and functional activity of VDR-RXR heterodimers [68]. This suggests that RXR also specifically contacts 5′-flanking nucleotides to its binding motif. In summary, 5′-flanking sequences should be considered as an integral part of a RE, so that more likely octameric motifs instead of hexameric sequences are specific VDR binding sites. A reanalysis of previously characterized VDREs (Table I) for their 5′-flanking sequences will provide a better understanding of their relative strength. Even more important is the possibility of a more accurate prediction of putative VDREs from large amounts of sequence data now available from the human genome.

F. VDR Homodimers and RXR-Independent 1,25(OH)2D3 Signaling VDR homodimers were found to bind and initiate ligand-driven transactivation from DR3- and DR6-type VDREs [15,69]. Moreover, homodimers of VDR DBDs

have even been crystallized on DR3-type REs [3] confirming the view that they may have some physiological relevance. The in vitro binding of VDR homodimers on idealized DR3- and DR6-type REs is not as high as that of VDR-RXR heterodimers on DR3-type REs [42,70]. However, in vitro binding affinity of VDR-RXR heterodimers to natural DR3-type VDRE of class III is also weak [34] and VDR homodimers may be specifically stabilized by CoA proteins [71]. In vitro complex formation experiments as well as yeast or mammalian two-hybrid experiments leave no doubts that within the NR superfamily, RXR is the most efficient protein– protein interaction partner of the VDR. This interaction preference is evolutionarily conserved and also applies to the 15 closest members of VDR within its superfamily [72]. Moreover, at least one of the three RXR subtypes seems to be expressed in every mammalian cell, which makes investigations of putative RXRindependent 1,25(OH)2D3 signaling pathways difficult. In vitro as well as in a Drosophila cell system the formation of VDR homodimers has been shown [15]. Evidence for VDR-RAR and VDR-T3R heterodimers has also been reported [42,53,59]. The concentration of protein was very high, however. VDR-RAR and VDR-T3R heterodimers showed preference for DR5and DR6-type REs [42] and have also been observed on ER7- to ER10-type REs [58,59]. Taken together, on most VDREs and in the vast majority of physiological situations, VDR-RXR heterodimers seem to mediate 1,25(OH)2D3 signaling. There may be exceptions, however. In these cases, their in vitro relevance needs to be studied in the context of the natural promoter in a chromatin environment.

III. COMPLEX VITAMIN D–RECEPTOR BINDING SITES A. Simple versus Complex VDREs With the exception of the VDREs in the human and rat osteocalcin gene, all natural VDREs are formed by only two hexameric core binding motifs in a DR3, DR4, DR6, and ER9 arrangement (Table I). These latter targets may be therefore considered as “simple” VDREs. The question is, whether the occurrence of one simple VDRE within a promoter is sufficient for gene responsiveness to 1,25(OH)2D3. Because of its optimized 5′-flanking dinucleotide and core binding motif sequences the DR4-type RE of the rat pit-1 gene is the most efficient known VDRE in vitro [34,50]. However, the chromatin in the region of the pit-1 gene promoter containing this RE seems to be closed in the adult rat, so that the responsiveness of the gene to 1,25(OH)2D3

320

CARSTEN CARLBERG, THOMAS W. DUNLOP, CHRISTIAN FRANK, AND SAMI VÄISÄNEN

is lower than expected [73]. This indicates that a high in vitro binding affinity of VDR-RXR heterodimers for a VDRE is not sufficient for responsiveness to 1,25(OH)2D3. When the promoter region that contains the VDRE is covered by condensed chromatin, VDRRXR heterodimers are unable to bind there. This makes sufficiently decondensed chromatin an essential prerequisite for a functional VDRE. Chromatin decondensation is achieved by the activity of HATs, which are recruited to their local chromatin target by CoA proteins. In turn, these CoAs are transiently attracted to a promoter region by ligand-activated NRs and other active TFs. Therefore, the more TF binding sites a given promoter region contains and the more of these TFs are expressed in the respective cell, the higher is the chance that this area of the promoter gets locally decondensed. One example is the VDRE of the rat osteocalcin gene, which is flanked on both sides with a binding site for the TF Runx2/Cbfa1 [74]. By contacting CoA proteins and HATs Runx2/Cbfa1 seems to mediate the opening of chromatin locally, which allows efficient binding of VDR-RXR heterodimers to this decondensed region to occur. This mechanism suggests that VDREs are better targets for VDR-RXR heterodimers, if other TFs are bound to the same chromatin region. In this respect, promoter context and cell-specific expression of other TFs may be of greater importance to VDRE functionality and specificity than its in vitro binding profile. If this idea is true, it will apply to other members of the NR superfamily and will question the validity of isolated simple VDREs. Therefore, in the future VDREs will have to be understood as complex structures with multiple TF binding sites. Some of these TF binding sites will be other NR core binding motifs. The DR6/DR3 and ER9/DR3 structures of the VDREs of the human and rat osteocalcin genes, respectively, are the first examples of complex VDREs. These two complex VDREs show only limited homology to each other, although they are derived from orthologous genes. This suggests that for an important primary 1,25(OH)2D3 responding gene such as osteocalcin, there may be limited evolutionary pressure for a specific VDRE structure. It seems to be more important to guarantee an efficient binding of the VDR to a promoter in competition with the tight packaging of nucleosomes. It is interesting to note that the complex VDRE of the human osteocalcin gene is overlaid by a binding site of the TF AP-1 [75], which provides the RE with an increased activity. These types of REs are also observed for other NRs and often referred to as composite REs [76]. Another interesting example of a complex/composite VDRE has been reported in the mouse c-fos promoter [77]. Within this VDRE three hexameric core binding motifs are forming a DR7/DR7 structure, which contains an internal

binding site for the TF NF-1. Additional examples are the VDRE of the human and mouse fibronectin gene, which contains an internal binding site for the TF CREB [54], or the VDRE of the rat bone sialo protein, which also seems to bind the general TF TBP [78].

B. RE Clusters In mammals, the most responsive known primary 1,25(OH)2D3 target gene is CYP24, which gene product is the key enzyme involved in the catabolism of 1,25(OH)2D3. Activation of this gene provides a negative feedback loop mechanism controlling the level of the hormonal ligand. The reason for the strong responsiveness of the CYP24 gene appears to be that it contains two DR3-type VDREs separated by a distance of less than 100 bp in close proximity to the TSS. These DR3-type VDRE clusters are evolutionarily conserved between humans and rodents [43,79]. The sequences of both VDREs are not optimal and are categorized into class II (Table I). However, these less optimal structures are more than compensated by the fact that they are located in close vicinity to each other and to the TSS of the gene. In addition, binding sites for the TF Ets-1 have been characterized within the CYP24 promoter and seem to interfere synergistically with the two VDREs, i.e., the cluster of VDR and Ets binding sites form another type of complex VDRE. Such types of RE clusters are also known for other genes that are primary targets of additional members of the NR superfamily, such as VDR’s close relatives CAR and PXR. Investigation of natural CAR and PXR target gene, such as CYP2B6 or CYP3A4, respectively [80,81], indicate that a single RE is insufficient for mediating the regulatory role of the receptors and that more likely at least two CAR or PXR REs in close proximity to each other are required. In the case of CAR, these multiple RE clusters are called phenobarbital response enhancer modules (PBREMs), because they mediate the responsiveness to the CAR activator phenobarbital. The CYP2B6 gene contains two DR4-type REs with an additional binding site for the TF NF-1 [80], whereas the PBREM of the UDPglucuronosyltransferase 1A1 gene is formed by three CAR REs [82]. In contrast, the CYP3A4 gene contains an ER6-type RE proximal to the TSS and a more distal DR3/ER6 cluster [81]. Taken together, these assemblies represent two or more simple NR REs together with binding sites for other types of TFs. Their overall function seems to follow the same rules, i.e., the greater number of NRs and other TFs bind to such promoter regions, the greater is the chance that they induce histone acetylation and chromatin decondensation.

CHAPTER 18 Diversity of Vitamin D Target Genes

Another interesting observation in relation to the RE clusters is that they are rather promiscuously bound by related members of the NR superfamily, such as VDR, CAR, and PXR. VDR was shown to activate the CYP2B6, CYP3A4, and CYP2C9 genes by replacing CAR and PXR within the respective RE clusters [83]. In particular, the CYP3A4 gene, which contains a DR3-type RE within its RE cluster, was shown to be stimulated effectively by 1,25(OH)2D3 and can be considered as a primary 1,25(OH)2D3 responding gene. This suggests that 1,25(OH)2D3 and the VDR may have an impact on the metabolism of prescription drugs, of which 60% are metabolized by the CYP3A4 enzyme [84]. In this context, it is interesting to note that the VDR seems to be able to act as an intestinal bile acid sensor, because certain bile acids have been identified as low-affinity VDR ligands [85]. These actions are also mediated by CYP3A4. It is very likely that future investigations will reveal more of these 1,25(OH)2D3 target genes, thus enlarging the physiological importance of this hormone to an even greater extent.

C. Negative VDREs A relatively undercharacterized aspect of 1,25(OH)2D3 signaling is the mechanism of down-regulation of 1,25(OH)2D3 responsive genes. Microarray experiments have indicated that nearly half of all primary 1,25(OH)2D3 responding genes are down-regulated by the hormone, but few of them have been studied in more detail. It is obvious that only genes which show basal activity can be down-regulated, i.e., these genes exhibit basal activity due to other TFs binding to their promoter. There are several different models that attempt to explain how 1,25(OH)2D3 and the VDR can mediate down-regulation of genes, but the common theme is that VDR counteracts the activity of specific TFs. In the situation where these activating TFs are other NRs or TFs that bind to composite NR REs, VDR could simply compete for DNA binding sites [86,87]. In a similar way, VDR could also compete for binding to partner proteins, such as RXR, or for common CoAs, such as SRC-1 or p300 [88]. In all these situations the down-regulating effects of the VDR should be of general impact, i.e., the mechanism could apply to other genes in the same way. So far, however, no general down-regulating effects of 1,25(OH)2D3 have been reported. The binding of 1,25(OH)2D3 to the VDR results in a ligand-dependent conformational change, which leads to an exchange of proteins that bind to the LBD of the VDR, i.e., altering protein-protein interaction profiles but no changes in VDR-DNA interaction properties.

321 In most cases, however, 1,25(OH)2D3-dependent down-regulation of a gene involves specific DNA binding sites on the specific promoter, which are referred to as negative VDREs. Experiments suggest that DNA binding of the VDR to down-regulated genes such as the PTH gene, does not involve RXR [89,90], whereas other studies, such as on the ANF gene [91], come to opposite conclusions. As discussed above, there are also alternative partner proteins for the VDR, such as RAR or VDR itself, and one might speculate that cell- and gene-specific TFs could help the VDR in binding efficiently to DNA. Investigation of the 1,25(OH)2D3-mediated repression of the PTHrP gene indicates that protein kinases may be involved in the down-regulation of this gene [92]. Some of the natural VDREs in Table I were identified within genes that are down-regulated by 1,25(OH)2D3 and are therefore more likely to be negative VDREs. These sequences are not revealing, however, because in the absence of a natural promoter context and normal chromatin organization, these VDREs behave like positive VDREs. This again highlights the fact that promoter context, the chromatin status, and the cluster of neighboring TFs are important for the function of a simple VDRE. Thus simple VDREs cannot act as negative VDREs. In the case of the calcitonin gene, two separate promoter regions are suggested to be responsible for the down-regulation of gene transcription by 1,25(OH)2D3 [93] (see Chapter 39). Together, this information indicates that mechanisms governing 1,25(OH)2D3-mediated down-regulation is complex. The concept that VDRE clusters together with and other TF binding sites regulate primary 1,25(OH)2D3 responding genes suggests that a promoter may contain both negative and positive VDREs. The activities of the different VDREs are determined by the promoter context and may not be simultaneously active. One might imagine that prior to stimulation with 1,25(OH)2D3 only the negative VDREs bind the VDR and recruit CoRs. This would actively condense the chromatin on a particular promoter region. The addition of ligand induces the release of CoR proteins and reduces chromatin density. The VDR may then be transiently released from the negative VDRE and bind to a positive VDRE, which may be uncovered through 1,25(OH)2D3-dependent local nucleosome acetylation. The VDR then interacts with CoAs of the DRIP/TRAP mediator protein complex on this positive VDRE leading to transient transcriptional activation. After a certain period of time, newly synthesized, unliganded VDR again binds to the negative VDRE, which initiates chromatin closing and inactivation of the positive VDRE. In this or even more complex scenarios, the balance between negative and positive

322

CARSTEN CARLBERG, THOMAS W. DUNLOP, CHRISTIAN FRANK, AND SAMI VÄISÄNEN

VDREs could explain the time course of the activation of primary 1,25(OH)2D3 responding genes.

have to be included in more effective versions of in silico screening software.

D. Whole-Genome Screening for Putative VDREs

IV. CONCLUSION

Considering both orientations of DNA, the NR core binding motif RGKTSA should be found every 256 bp of genomic DNA. Dimeric assemblies of such hexamers should show up as direct repeats every 65,536 bp of promoter sequence and as everted repeats every 32,768 bp. Since VDR-RXR heterodimers bind comparably well to DR3-, DR4-, and ER9-type REs, an in silico screening is expected to identify a putative VDR binding site every 16,384 bp. This would predict nearly 200,000 putative VDREs within the human genome, and as a consequence on average every gene would contain several VDREs in its promoter and should be responsive to 1,25(OH)2D3. Similar calculations apply to other members of the NR superfamily, and for TFs with a shorter specific binding site even higher numbers would be predicted. A realistic number of 1,25(OH)2D3 responding genes is far less than this, perhaps a few hundred. The number of VDR molecules varies from hundreds to several thousand molecules. These calculations make it obvious that not every putative VDR binding site is used in nature in any cell at any given time. The most obvious reason is that most of these sequences are effectively covered by nucleosomes, so that they are not accessible to the VDR. This applies in particular to those sequences that are isolated from other NR or TF binding sites or lie distant from the promoter. This perspective strongly discourages the idea that isolated, simple VDREs may be functional in vivo. Therefore, the presently identified simple VDREs (Table I) may be parts of more complex VDREs as already demonstrated for the CYP24 and osteocalcin gene. An effective in silico prediction of novel VDREs has to focus on the identification of complex VDREs. Unfortunately, presently available in silico screening software, such as NUBISCAN [94], is unable to identify complex VDREs. Other available programs such as TRANSFAC have the capability to identify larger numbers of REs for different TFs, making the identification of potential complex REs possible. However, even these programs are unable to predict chromatin condensation states and nucleosome positioning, which is essential information in determining the likelihood that a given RE lies in an accessible region of a promoter. General parameters, such as nucleosome positioning and interspecies homology screening of regulatory sequences as well as the binding sites of all other TFs and their cell-specific expression patterns,

The sequencing of the complete human genome and the genomes of other species, i.e., the availability of all regulatory sequences, enables a more mature understanding of the diversity of 1,25(OH)2D3 target genes. Perhaps the idea of simple isolated VDREs, such as those listed in Table I, should shift to the concept of complex VDREs, of which the simple VDRE represents the core. Depending on the temporal presence of cellspecific TFs, these complex REs may act positively or negatively in respect to 1,25(OH)2D3. The coordinated action of these different types of VDREs could explain the individual response of target genes to 1,25(OH)2D3.

Acknowledgment The Academy of Finland (grant 50319) supported the work.

References 1. Sutton AL, MacDonald PN 2003 Vitamin D: more than a “bone-a-fide” hormone. Mol Endocrinol 17:777–791. 2. Chawla A, Repa JJ, Evans RM, Mangelsdorf DJ 2001 Nuclear receptors and lipid physiology: opening the X-files. Science 294:1866–1870. 3. Shaffer PL, Gewirth DT 2002 Structural basis of VDR–DNA interactions on direct repeat response elements. EMBO J 21:2242–2252. 4. Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 Crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179. 5. Moras D, Gronemeyer H 1998 The nuclear receptor ligandbinding domain: structure and function. Curr Opin Cell Biol 10:384–391. 6. Glass CK, Rosenfeld MG 2000 The coregulator exchange in transcriptional functions of nuclear receptors. Genes Dev 14:121–141. 7. Polly P, Herdick M, Moehren U, Baniahmad A, Heinzel T, Carlberg C 2000 VDR-Alien: a novel, DNA-selective vitamin D3 receptor-corepressor partnership. FASEB J 14:1455–1463. 8. Leo C, Chen JD 2000 The SRC family of nuclear receptor coactivators. Gene 245:1–11. 9. Rachez C, Lemon BD, Suldan Z, Bromleigh V, Gamble M, Näär AM, Erdjument-Bromage H, Tempst P, Freedman LP 1999 Ligand-dependent transcription activation by nuclear receptors requires the DRIP complex. Nature 398:824–828. 10. Carlberg C 2003 Molecular basis of the selective activity of vitamin D analogues. J Cell Biochem 88:274–281. 11. Toell A, Gonzalez MM, Ruf D, Steinmeyer A, Ishizuka S, Carlberg C 2001 Different molecular mechanisms of vitamin D3 receptor antagonists. Mol Pharmacol 59:1478–1485.

CHAPTER 18 Diversity of Vitamin D Target Genes

12. Macias Gonzalez M, Samenfeld P, Peräkylä M, Carlberg C 2003 Corepressor excess shifts the two side chain vitamin D analogue Gemini from an agonist to an inverse agonist of the vitamin D receptor. Mol Endocrinol 17:2028–2038. 13. Väisänen S, Ryhänen S, Saarela JT, Peräkylä M, Andersin T, Maenpää PH 2002 Structurally and functionally important amino acids of the agonistic conformation of the human vitamin D receptor. Mol Pharmacol 62:788–794. 14. Carlberg C, Polly P 1998 Gene regulation by vitamin D3. Crit Rev Eukaryot Gene Expr 8:19–42. 15. Carlberg C, Bendik I, Wyss A, Meier E, Sturzenbecker LJ, Grippo JF, Hunziker W 1993 Two nuclear signalling pathways for vitamin D. Nature 361:657–660. 16. Glass CK 1994 Differential recognition of target genes by nuclear receptor monomers, dimers, and heterodimers. Endocr Rev 15:391–407. 17. Rastinejad F, Perlmann T, Evans RM, Sigler PB 1995 Structural determinants of nuclear receptor assembly on DNA direct repeats. Nature 375:203–211. 18. Quack M, Szafranski K, Rouvinen J, Carlberg C 1998 The role of the T-box for the function of the vitamin D receptor on different types of response elements. Nucleic Acids Res 26: 5372–5378. 19. Umesono K, Murakami KK, Thompson CC, Evans RM 1991 Direct repeats as selective response elements for the thyroid hormone, retinoic acid, and vitamin D3 receptors. Cell 65: 1255–1266. 20. Quelo I, Kahlen JP, Rascle A, Jurdic P, Carlberg C 1994 Identification and characterization of a vitamin D3 response element of chicken carbonic anhydrase-II. DNA Cell Biol 13:1181–1187. 21. Schräder M, Nayeri S, Kahlen JP, Müller KM, Carlberg C 1995 Natural vitamin D3 response elements formed by inverted palindromes: polarity-directed ligand sensitivity of vitamin D3 receptor–retinoid X receptor heterodimer-mediated transactivation. Mol Cell Biol 15:1154–1161. 22. MacDonald PN, Dowd DR, Nakajima S, Galligan MA, Reeder MC, Haussler CA, Ozato K, Haussler MR 1993 Retinoid X receptors stimulate and 9-cis retinoic acid inhibits 1,25-dihydroxyvitamin D3-activated expression of the rat osteocalcin gene. Mol Cell Biol 13:5907–5917. 23. Hancock R 2000 A new look at the nuclear matrix. Chromosoma 109:219–225. 24. Kahlen JP, Carlberg C 1996 Functional characterization of a 1,25-dihydroxyvitamin D3 receptor binding site found in the rat atrial natriuretic factor promoter. Biochem Biophys Res Commun 218:882–886. 25. Zhang Q, Wrana JL, Sodek J 1992 Characterization of the promoter region of the porcine opn (osteopontin, secreted phosphoprotein 1) gene. Eur J Biochem. 207:649–659. 26. Noda M, Vogel RL, Craig AM, Prahl J, DeLuca HF, Denhardt DT 1990 Identification of a DNA sequence responsible for binding of the 1,25-dihydroxyvitamin D3 receptor and 1,25-dihydroxyvitamin D3 enhancement of mouse secreted phosphoprotein 1 (Spp-1 or osteopontin) gene expression. Proc Natl Acad Sci USA 87:9995–9999. 27. Quelo I, Machuca I, Jurdic P 1998 Identification of a vitamin D response element in the proximal promoter of the chicken carbonic anhydrase II gene. J Biol Chem 273: 10638–10646. 28. Chen KS, DeLuca HF 1995 Cloning of the human 1α,25dihydroxyvitamin D3 24-hydroxylase gene promoter and identification of two vitamin D-responsive elements. Biochem Biophys Acta 1263:1–9.

323 29. Zierold C, Darwish HM, DeLuca HF 1994 Identification of a vitamin D-response element in the rat calcidiol (25-hydroxyvitamin D3) 24-hydroxylase gene. Proc Natl Acad Sci USA 91:900–902. 30. Taketani Y, Segawa H, Chikamori M, Morita K, Tanaka K, Kido S, Yamamoto H, Iemori Y, Tatsumi S, Tsugawa N, Okano T, Kobayashi T, Miyamoto K, Takeda E 1998 Regulation of type II renal Na+-dependent inorganic phosphate transporters by 1,25-dihydroxyvitamin D3: identification of a vitamin Dresponsive element in the human NaPi-3 gene. J Biol Chem 273:14575–14581. 31. Demay MB, Gerardi JM, DeLuca HF, Kronenberg, HM 1990 DNA sequences in the rat osteocalcin gene that bind the 1,25-dihydroxyvitamin D3 receptor and confer responsiveness to 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 87:369–373. 32. Demay MB, Kieran MS, DeLuca HF, Kronenberg HM 1992 Sequences in the human parathyroid hormone gene that bind the 1,25-dihydroxyvitamin D3 receptor and mediate transcriptional repression in response to 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 89:8097–8101. 33. Kremer R, Sebag M, Champigny C, Meerovitch K, Hendy GN, White J, Goltzman D 1996 Identification and characterization of 1,25-dihydroxyvitamin D3-responsive repressor sequences in the rat parathyroid hormone-related peptide gene. J Biol Chem 271:16310–16316. 34. Toell A, Polly P, Carlberg C 2000 All natural DR3-type vitamin D response elements show a similar functionality in vitro. Biochem J 352:301–309. 35. Falzon M 1996 DNA sequences in the rat parathyroid hormone-related peptide gene responsible for 1,25-dihydroxyvitamin D3-mediated transcriptional repression. Mol Endocrinol 10:672–681. 36. Darwish HM, DeLuca HF 1992 Identification of a 1,25dihydroxyvitamin D3-response element in the 5′-flanking region of the rat calbindin D9k gene. Proc Natl Acad Sci USA 89:603–607. 37. Wang GF, Nikovits W, Schleinitz M, Stockdale FE 1998 A positive GATA element and a negative vitamin D receptorlike element control atrial chamber-specific expression of a slow myosin heavy-chain gene during cardiac morphogenesis. Mol Cell Biol 18:6023–6034. 38. Alonso M, Segura C, Dieguez C, Perez-Fernandez R 1998 High-affinity binding site to the vitamin D receptor DNA binding domain in the human growth hormone promoter. Biochem Biophys Res Commun 247:882–887. 39. Cao X, Ross FP, Zhang L, MacDonald PN, Chappel J, Teitelbaum SL 1993 Cloning of the promoter for the avian integrin β3 subunit gene and its regulation by 1,25-dihydroxyvitamin D3. J Biol Chem 268:27371–27380. 40. Liu SM, Koszewski N, Lupez M, Malluche HH, Olivera A, Russell J 1996 Characterization of a response element in the 5′-flanking region of the avian (chicken) PTH gene that mediates negative regulation of gene transcription by 1,25dihydroxyvitamin D3 and binds the vitamin D3 receptor. Mol Endocrinol 10:206–215. 41. Liu M, Lee MH, Cohen M, Bommakanti M, Freedman LP 1996 Transcriptional activation of the Cdk inhibitor p21 by vitamin D3 leads to the induced differentiation of the myelomonocytic cell line U937. Genes Dev 10:142–153. 42. Schräder M, Müller KM, Carlberg C 1994 Specificity and flexibility of vitamin D signaling. Modulation of the activation of natural vitamin D response elements by thyroid hormone. J Biol Chem 269:5501–5504.

324

CARSTEN CARLBERG, THOMAS W. DUNLOP, CHRISTIAN FRANK, AND SAMI VÄISÄNEN

43. Kerry DM, Dwivedi PP, Hahn CN, Morris HA, Omdahl JL, May BK 1996 Transcriptional synergism between vitamin Dresponsive elements in the rat 25-hydroxyvitamin D3 24-hydroxylase (CYP) promoter. J Biol Chem 271:29715–29721. 44. Staal A, van Wijnen AJ, Birkenhäger JC, Pols HAP, Prahl J, DeLuca HF, Gaub MP, Lian JB, Stein GS, van Leeuwen JPTM, Stein JL 1996 Distinct conformations of vitamin D receptor/ retinoid X receptor-a heterodimers are specified by dinucleotide differences in the vitamin D-responsive elements of the osteocalcin and osteopontin genes. Mol Endocrinol 10:1444–1456. 45. Wood JR, Greene GL, Nardulli AM 1998 Estrogen response elements function as allosteric modulators of estrogen receptor conformations. Mol Cell Biol 18:1927–1934. 46. Herdick M, Bury Y, Quack M, Uskokovic M, Polly P, Carlberg C 2000 Response element- and coactivator-mediated conformational change of the vitamin D3 receptor permits sensitive interaction with agonists. Mol Pharmacol 57:1206–1217. 47. Carlberg C 1995 Mechanisms of nuclear signalling by vitamin D3. Interplay with retinoid and thyroid hormone signalling. Eur J Biochem 231:517–527. 48. Rhodes SJ, Chen R, DiMattia GE, Scully KM, Kalla KA, Lin SC, Yu VC, Rosenfeld MG 1993 A tissue-specific enhancer confers Pit-1-dependent morphogen inducibility and autoregulation on the pit-1 gene. Genes Dev 7:913–932. 49. Mangelsdorf DJ, Evans RM 1995 The RXR heterodimers and orphan receptors. Cell 83:841–850. 50. Quack M, Carlberg C 2000 Ligand-triggered stabilization of vitamin D receptor/retinoid X receptor heterodimer conformations on DR4-type response elements. J Mol Biol 296: 743–756. 51. Morrison NA, Shine J, Fragonas JC, Verkest V, McMenemey ML, Eisman JA 1989 1,25-Dihydroxyvitamin D-responsive element and glucocorticoid repression in the osteocalcin gene. Science 246:1158–1161. 52. Ozono K, Liao J, Kerner SA, Scott RA, Pike JW 1990 The vitamin D–responsive element in the human osteocalcin gene. Association with a nuclear proto-oncogene enhancer. J Biol Chem 265:21881–21888. 53. Schräder M, Bendik I, Becker-Andre M, Carlberg C 1993 Interaction between retinoic acid and vitamin D signaling pathways. J Biol Chem 268:17830–17836. 54. Polly P, Carlberg C, Eisman JA, Morrison NA 1996 Identification of a vitamin D3 response element in the fibronectin gene that is bound by a vitamin D3 receptor homodimer. J Cell Biochem 60:322–333. 55. Kahlen JP, Carlberg C 1994 Identification of a vitamin D receptor homodimer-type response element in the rat calcitriol 24-hydroxylase gene promoter. Biochem Biophys Res Commun 202:1366–1372. 56. Xie Z, Bikle DD 1997 Cloning of the human phospholipase C-γ1 promoter and identification of a DR6-type vitamin Dresponsive element. J Biol Chem 272:6573–6577. 57. Schräder M, Kahlen JP, Carlberg C 1997 Functional characterization of a novel type of 1α,25-dihydroxyvitamin D3 response element identified in the mouse c-fos promoter. Biochem Biophys Res Commun 230:646–651. 58. Schräder M, Müller KM, Becker-André M, Carlberg C 1994 Response element selectivity for heterodimerization of vitamin D receptors with retinoic acid and retinoid X receptors. J Mol Endocrinol 12:327–339. 59. Schräder M, Müller KM, Nayeri S, Kahlen JP, Carlberg C 1994 VDR-T3R receptor heterodimer polarity directs ligand sensitivity of transactivation. Nature 370:382–386. 60. Carlberg C 1996 The concept of multiple vitamin D signaling pathways. J Investig Dermatol Symp Proc 1:10–14.

61. Nayeri S, Danielsson C, Kahlen JP, Schräder M, Mathiasen IS, Binderup L, Carlberg C 1995 The anti-proliferative effect of vitamin D3 analogues is not mediated by inhibition of the AP-1 pathway, but may be related to promoter selectivity. Oncogene 11:1853–1858. 62. Danielsson C, Mathiasen IS, James SY, Nayeri S, Bretting C, Hansen CM, Colston KW, Carlberg C 1997 Sensitive induction of apoptosis in breast cancer cells by a novel 1,25-dihydroxyvitamin D3 analogue shows relation to promoter selectivity. J Cell Biochem 66:552–562. 63. Mørk Hansen C, Mäenpää PH 1997 EB1089—a novel vitamin D analog with strong antiproliferative and differentiation inducing effects on target cells. Biochem Pharmacol. 54:1173–1179. 64. Quack M, Carlberg C 1999 Selective recognition of vitamin D receptor conformations mediates promoter selectivity of vitamin D analogs. Mol Pharmacol 55:1077–1087. 65. Carlberg C, Hooft van Huijsduijnen R, Staple JK, DeLamarter JF, Becker-Andre M 1994 RZRs, a new family of retinoid-related orphan receptors that function as both monomers and homodimers. Mol Endocrinol 8:757–770. 66. Schräder M, Becker-Andre M, Carlberg C 1994 Thyroid hormone receptor functions as monomeric ligand-induced transcription factor on octameric half-sites. Consequences also for dimerization. J Biol Chem 269:6444–6449. 67. Mader S, Leroy P, Chen JY, Chambon P 1993 Multiple parameters control the selectivity of nuclear receptors for their response elements. J Biol Chem 268:591–600. 68. Quack M, Frank C, Carlberg C 2002 Differential nuclear receptor signalling from DR4-type response elements. J Cell Biochem 86:601–612. 69. Cheskis B, Freedman LP 1994 Ligand modulates the conversion of DNA-bound vitamin D3 receptor (VDR) homodimers into VDR-retinoid X receptor heterodimers. Mol Cell Biol 14:3329–3338. 70. Schräder M, Wyss A, Sturzenbecker LJ, Grippo JF, LeMotte P, Carlberg C 1993 RXR-dependent and RXR-independent transactivation by retinoic acid receptors. Nucleic Acids Res 21:1231–1237. 71. Takeshita A, Ozawa Y, Chin WW 2000 Nuclear receptor coactivators facilitate vitamin D receptor homodimer action on direct repeat hormone response elements. Endocrinology 141:1281–1284. 72. Robinson-Rechavi M, Garcia HE, Laudet V 2003 The nuclear receptor superfamily. J Cell Sci 116:585–586. 73. Castillo AI, Jimenez-Lara AM, Tolon RM, Aranda A 1999 Synergistic activation of the prolactin promoter by vitamin D receptor and GHF-1: role of coactivators, CREB-binding protein and steroid hormone receptor coactivator-1 (SRC-1). Mol Endocrinol 13:1141–1154. 74. Sierra J, Villagra A, Paredes R, Cruzat F, Gutierrez S, Javed A, Arriagada G, Olate J, Imschenetzky M, Van Wijnen AJ, Lian JB, Stein GS, Stein JL, Montecino M 2003 Regulation of the bonespecific osteocalcin gene by p300 requires Runx2/Cbfa1 and the vitamin D3 receptor but not p300 intrinsic histone acetyltransferase activity. Mol Cell Biol 23:339–3351. 75. Schüle R, Umesono K, Mangelsdorf DJ, Bolado J, Pike JW, Evans RM 1990 Jun-Fos and receptors for vitamins A and D recognize a common response element in the human osteocalcin gene. Cell 61:497–504. 76. Miner JN, Yamamoto KR 1992 The basic region of AP-1 specifies glucocorticoid receptor activity at a composite response element. Genes Dev 6:491–2501. 77. Candeliere GA, Jurutka PW, Haussler MR, St-Arnaud R 1996 A composite element binding the vitamin D receptor, retinoid X receptor α, and a member of the CTF/NF-1 family of

325

CHAPTER 18 Diversity of Vitamin D Target Genes

78.

79. 80.

81.

82.

83.

84.

85. 86.

87.

transcription factors mediates the vitamin responsiveness of the c-fos promoter. Mol Cell Biol 16:584–592. Kim RH, Li JJ, Ogata Y, Yamauchi M, Freedman LP, Sodek J 1996 Identification of a vitamin D3-response element that overlaps a unique inverted TATA box in the rat bone sialoprotein gene. Biochem J 318:219–226. Zierold C, Darwish HM, DeLuca HF 1995 Two vitamin D response elements function in the rat 1,25-dihydroxyvitamin D 24-hydroxylase promoter. J Biol Chem 270:1675–1678. Honkakoski P, Zelko I, Sueyoshi T, Negishi M 1998 The nuclear orphan receptor CAR-retinoid X receptor heterodimer activates the phenobarbital-responsive enhancer module of the CYP2B gene. Mol Cell Biol 18:5652–5658. Lehmann JM, McKee DD, Watson MA, Willson TM, Moore JT, Kliewer SA 1998 The human orphan nuclear receptor PXR is activated by compounds that regulate CYP3A4 gene expression and cause drug interactions. J Clin Invest 102:1016–1023. Sugatani J, Kojima H, Ueda A, Kakizaki S, Yoshinari K, Gong QH, Owens IS, Negishi M, Sueyoshi T 2001 The phenobarbital response enhancer module in the human bilirubin UDP-glucuronosyltransferase UGT1A1 gene and regulation by the nuclear receptor CAR. Hepatology 33:1232–1238. Drocourt L, Ourlin JC, Pascussi JM, Maurel P, Vilarem MJ 2002 Expression of CYP3A4, CYP2B6, and CYP2C9 is regulated by the vitamin D receptor pathway in primary human hepatocytes. J Biol Chem 277:25125–25132. Thummel KE, Brimer C, Yasuda K, Thottassery J, Senn T, Lin Y, Ishizuka H, Kharasch E, Schuetz J, Schuetz E 2001 Transcriptional control of intestinal cytochrome P-4503A by 1α,25-dihydroxy vitamin D3. Mol Pharmacol 60:1399–1406. Makishima M, Lu TT, Xie W, Whitfield GK, Domoto H, Evans RM, Haussler MR, Mangelsdorf DJ 2002 Vitamin D receptor as an intestinal bile acid sensor. Science 296:1313–1316. Towers TL, Freedman LP 1998 Granulocyte-macrophage colonystimulating factor gene transcription is directly repressed by the vitamin D3 receptor: implications for allosteric influences on nuclear receptor structure and function by a DNA element. J Biol Chem 273:10338–10348. Towers TL, Staeva TP, Freedman LP 1999 A two-hit mechanism for vitamin D3-dediates transcriptional repression of the

88. 89.

90.

91.

92.

93.

94.

95.

96.

granulocyte-macrophage colony-stimulating factor gene: vitamin D receptor completes for DNA binding with NFAT1 and stabilizes c-Jun. Mol Cell Biol 19:4191–4199. Polly P, Carlberg C, Eisman JA, Morrison NA 1997 1α,25Dihydroxyvitamin D3 receptor as a mediator of transrepression of retinoid signaling. J Cell Biochem 67:287–296. Mackey SL, Heymont JL, Kronenberg HM, Demay MB 1996 Vitamin D receptor binding to the negative human parathyroid hormone response element does not require the retinoid X receptor. Mol Endocrinol 10:298–305. Nishishita T, Okazaki T, Ishikawa T, Igarashi T, Hata K, Ogata E, Fujita T 1998 A negative vitamin D response DNA element in the human parathyroid hormone-related peptide gene binds to vitamin D receptor along with Ku antigen to mediate negative gene regulation by vitamin D. J Biol Chem 273: 10901–10907. Chen S, Costa CHRM, Nakamura K, Ribero RCJ, Gardner DG 1999 Vitamin D–dependent suppression of human atrial natriuretic peptic gene promoter activity requires heterodimer assembly. J Biol Chem 274:11260–11266 Okazaki T, Nishimori S, Ogata E, Fujita T 2003 Vitamin D– dependent recruitment of DNA-PK to the chromatinized negative vitamin D response element in the PTHrP gene is required for gene repression by vitamin D. Biochem Biophys Res Commun 304:632–637. Peleg S, Abruzzese RV, Cooper CW, Gagel RF 1993 Downregulation of calcitonin gene transcription by vitamin D requires two widely separated enhancer sequences. Mol Endocrinol 7:1999–1008. Podvinec M, Kaufmann MR, Handschin C, Meyer UA 2002 NUBIScan, an in silico approach for prediction of nuclear receptor response elements. Mol Endocrinol 16: 1269–1279. Gill RK, Christakos S 1993 Identification of sequence elements in mouse calbindin-D28k gene that confer 1,25-dihydroxyvitamin D3- and butyrate-inducible responses. Proc Natl Acad Sci USA 90:2984–2988. Carlberg C 1997 Critical analysis of 1α,25-dihydroxyvitamin D3 response elements. In: Norman AW, Bouillon B, Thomasset M (eds) Proc 10th Workshop on Vitamin D, pp. 268–275.

CHAPTER 19

Intranuclear Organization of the Regulatory Machinery for Vitamin D–Mediated Control of Skeletal Gene Expression GARY S. STEIN, JANE B. LIAN, MARTIN MONTECINO,* JANET L. STEIN, ANDRE J. VAN WIJNEN, AMJAD JAVED, JE-YONG CHOI,** S. KALEEM ZAIDI, SORAYA GUTIERREZ,* JIALI SHEN, SHIRWIN POCKWINSE, AND DANIEL YOUNG Department of Cell Biology and Cancer Center, University of Massachusetts Medical School, Worcester, Massachusetts; ∗Departamento de Biologia Molecular, Universidad de Concepcion, Concepcion, Chile; ∗∗Department of Biochemistry, Kyungpook National University, Daegu, Korea I. Introduction II. Requirements for Physiologically Responsive Control of Skeletal Gene Expression In Vivo III. Gene Expression within the Three-Dimensional Context of Nuclear Architecture IV. Chromatin Remodeling Facilitates Vitamin D–Mediated Promoter Accessibility and Integration of Regulatory Activities V. Nuclear Microenvironments: Accommodating The Rules That Govern In Vivo Transcriptional Control

VI. Scaffolding of Regulatory Elements for Combinatorial Control of Gene Expression VII. Intranuclear Trafficking of Skeletal Regulatory Factors to Subnuclear Sites That Support Transcription: “To Be in the Right Place at the Right Time” VIII. The Regulated and Regulatory Parameters of Subnuclear Organization References

I. INTRODUCTION

factors abrogate competency for vitamin D control of skeletal gene expression during development and fidelity of gene expression in tumor cells.

The architecturally associated subnuclear organization of nucleic acids and cognate regulatory factors illustrates functional interrelationships between nuclear structure and gene expression. Mechanisms that contribute to the spatial distribution of transcription factors within the dynamic three-dimensional context of nuclear architecture control the sorting of regulatory information and the transcriptionally competent or repressed chromatin configuration of gene promoters as well as the assembly and activities of sites within the nucleus that support gene expression. Vitamin D control of gene expression serves as a paradigm for experimentally addressing mechanisms that govern the intranuclear targeting of regulatory factors to nuclear domains where chromatin remodeling and transcription of developmental as well as tissue-specific genes occur. We will provide an overview of molecular, cellular, biochemical, and in vivo genetic approaches that provide insight into the trafficking of regulatory factors that mediate vitamin D–controlled gene expression to transcriptionally active subnuclear sites. Examples will be presented that suggest modifications in the intranuclear targeting of transcription VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

II. REQUIREMENTS FOR PHYSIOLOGICALLY RESPONSIVE CONTROL OF SKELETAL GENE EXPRESSION IN VIVO Bone formation during development and skeletal remodeling throughout life requires the complex and interdependent expression of cell growth and phenotypic genes (reviewed in [1,2]). There is a requirement for responsiveness to a broad spectrum of regulatory cues that transduce physiological signals from the extracellular matrix to sites within the nucleus where genes that mediate skeletogenesis reside [3–6]. As our understanding of gene regulatory mechanisms expands, it becomes increasingly evident that there are unique parameters of transcriptional control that support the transient activation and suppression of genes for skeletal development and bone homeostasis. Other mechanisms Copyright © 2005, Elsevier, Inc. All rights reserved.

328 are invoked for long-term obligations to gene expression that sustain the specialized properties of the bone cells. Vitamin D serves as a principal modulator of skeletal gene transcription necessitating an understanding of interfaces between activity of this steroid hormone with regulatory cascades that are functionally linked to the regulation of skeletal genes [7]. There is growing appreciation for the repertoire of factors that influence gene expression for commitment to the osteoblast lineage. It is well documented that sequentially expressed genes support progression of osteoblast differentiation through developmental transition points where responsiveness to phosphorylationmediated regulatory cascades determine competency for establishing and maintaining the structural and functional properties of bone cells [2,4,8,9]. The catalog of promoter elements and cognate regulatory proteins that govern skeletal gene expression offers essential but insufficient insight into mechanisms that are operative in intact cells. Gene promoters serve as regulatory infrastructure by functioning as blueprints for responsiveness to the flow of cellular regulatory signals. But to access the specific genetic information necessitates understanding transcriptional control of skeletal genes within the context of the subnuclear organization of nucleic acids and regulatory proteins. Explanations are required for (1) convergence of multiple regulatory signals at promoter sequences; (2) the integration of regulatory information at independent promoter domains; (3) selective utilization of redundant regulatory pathways; (4) thresholds for initiation or down-regulation of transcription with limited intranuclear representation of promoter elements and regulatory factors; (5) mechanisms that render the promoters of cell growth and phenotypic genes competent for protein–DNA and protein–protein interactions in a physiologically responsive manner; (6) the composition, organization, and assembly of sites within the nucleus that support transcription; and (7) the intranuclear trafficking of regulatory proteins to transcriptionally active foci.

III. GENE EXPRESSION WITHIN THE THREE-DIMENSIONAL CONTEXT OF NUCLEAR ARCHITECTURE Evidence is accumulating that the architectural organization of nucleic acids and regulatory proteins within the nucleus support functional interrelationships between nuclear structure and gene expression (Fig. 1 and 2). There is increasing acceptance that components of nuclear architecture are functionally linked to the organization and sorting of regulatory information in a manner that permits selective utilization [10–20]. The primary

GARY S. STEIN,

ET AL .

level of nuclear organization, the representation and ordering of genes and promoter elements, provides alternatives for physiological control. The molecular organization of regulatory elements, the overlap of regulatory sequences within promoter domains, and the multipartite composition of regulatory complexes increase options for responsiveness. Chromatin structure and nucleosome organization reduce distances between regulatory sequences, facilitate crosstalk between promoter elements, and render elements competent for interactions with positive and negative regulatory factors [21]. The components of higher order nuclear architecture that includes nuclear pores [22,23], the nuclear matrix, and subnuclear domains contribute to the subnuclear distribution and activities of genes and regulatory factors (reviewed in [24–27]). Compartmentalization of regulatory complexes is illustrated by focal organization of PML bodies [28–30], Runx bodies [11,12,16,31,32], and the nucleolus, chromosomes [33], as well as by the punctate intranuclear distribution of sites for replication [34–36], DNA repair, transcription [37–42], and the processing of gene transcripts [26,43,44]. There is emerging recognition that nuclear structure and function are causally interrelated. With mounting evidence for organization of nucleic acids and regulatory proteins into subnuclear domains that are associated with components of nuclear architecture, the perception of a dichotomy between nuclear architecture and control of gene expression is difficult to justify. Rather, it is necessary to design experiments to define mechanisms that direct genes and regulatory factors to sites within the nucleus where localization integrates regulatory parameters of gene expression and establishes microenvironments with boundaries between regulatory complexes that are required for fidelity of activity. The bone-specific osteocalcin gene and skeletalrestricted Runx2 (AML3/Cbfa1/PEBP24) transcription factor serve as paradigms for obligatory relationships between nuclear structure with physiological control of skeletal gene expression [32,45–48]. The modularly organized promoter of the bone-specific osteocalcin gene contains proximal and distal regulatory elements that support basal, and tissue-specific as well as growth factor, homeodomain, signaling protein, and steroid hormone responsive transcriptional control (reviewed in [45,46,49–55]) (see Fig. 3). Modulation of osteocalcin gene expression during bone formation and remodeling requires physiologically responsive accessibility of these proximal and upstream promoter sequences to regulatory and coregulatory proteins as well as protein–protein interactions that integrate independent promoter domains. The nuclear matrixassociated Runx transcription factors contribute to the control of skeletal gene expression by sequence-specific

CHAPTER 19 Intranuclear Organization of the Regulatory Machinery for Vitamin D

329

FIGURE 1 Multiple levels of chromatin architecture within the nucleus. Higher order chromatin organization in the nucleus results from sequential packaging of DNA from a linear double helix (upper panel). The sequence of the Runx binding element is shown. A loop domain emanating from the highly compact chromatin is schematically illustrated (lower panel). Elements at the base of the loop structure designated MARs [matrix attachment regions, or alternatively locus control regions (LCR) or scaffold attachment regions (SCA)] mediate association of these genomic domains with the nuclear scaffold. Genes within the loop domain undergo local chromatin remodeling to support transcriptional activation or suppression. The nuclear matrix provides anchorage for both nucleic acids and regulatory as well as co-regulatory factors that control transcription.

binding to promoter elements of target genes and serving as scaffolds for the assembly and organization of co-regulatory proteins that mediate biochemical and architectural control of promoter activity.

IV. CHROMATIN REMODELING FACILITATES VITAMIN D–MEDIATED PROMOTER ACCESSIBILITY AND INTEGRATION OF REGULATORY ACTIVITIES It is well recognized that genomic DNA is packaged as chromatin. These “bead on a string” structures designated nucleosomes are structurally remodeled to accommodate requirements for transcription, emphasizing the extent to which architectural organization of genes is causally related to functional activity. The

identification and characterization of proteins that catalyze histone acetylation, deacetylation, methylation, and phosphorylation [56–63], as well as the SWI/SNFrelated proteins [56,64–68] that facilitate chromatin remodeling and potentially the accessibility of promoter sequences to regulatory and coregulatory factors, represent an important dimension in control of the structural and functional activities of genes and promoter regulatory elements [64,69–75]. Relationships of regulatory signaling pathways to enhance activities that modulate gene, chromatin, and chromosome organization can now be directly investigated. Additional levels of specificity are provided by structural modifications of gene promoters that influence competency for factor interactions. Simply stated, changes in the architectural properties of promoter elements determine effectiveness of gene regulatory sequences as substrates for interactions with regulatory factors. The regulatory and regulated parameters of chromatin

330

GARY S. STEIN,

ET AL .

Structural components

Apoptosis

Survivin

Replication sites

RPA

Chromosomal territories

Chromosomes

Nuclear envelope

PML bodies

Nucleoli

Transcription SWI/SNF complex sites

BRCA1

CAF-1

Replication and repair

RUNX domains

VDR SC 35 Domains

Coiled bodies

TLE

Transcription

Splicing FIGURE 2 Components of nuclear architecture are functionally linked to the organization and sorting of regulatory information. Nuclear functions are organized into distinct, nonoverlapping subnuclear domains. Nuclear matrix (shown in the center as revealed by electron microscopy), the underlying network of anastomizing network of filaments and fibers provides structural basis for the functional compartmentalization of nuclear functions. Immunofluorescence microscopy of the nucleus has revealed the distinct subnuclear distribution of vital nuclear processes, including (but not limited to) DNA replication sites and proteins involved in replication such as CAF-1 and RPA; DNA damage/repair as shown by BRCA1; chromatin remodeling, e.g., mediated by the SWI/SNF complex; structural parameters of the nucleus, such as the nuclear envelope, chromosomes, and chromosomal territories; chromatin organization and tissue specific transcriptional control, for example Runx, TLE, and VDR domains; and RNA synthesis and processing involving, for example, transcription sites; SC35 domains, coiled bodies, and nucleoli as well as proteins involved in cell survival e.g., survivin. Subnuclear PML bodies of unknown function have been examined in numerous cell types. All these domains are associated with the nuclear matrix. (See color plate).

remodeling and the rate limiting steps in the relevant signaling cascades are being actively pursued and will unquestionably provide insight into skeletal gene regulatory mechanisms from structural and functional perspectives. The chromatin organization of the osteocalcin gene illustrates dynamic remodeling of a promoter to accommodate requirements for phenotype-related developmental and steroid hormone responsive activity.

Nuclease digestion and ligation-mediated PCR analysis as well as in vitro nucleosome reconstitution studies establish the placement of nucleosomes in the proximal basal/tissue specific domain and at the upstream vitamin D–responsive element, blocking accessibility of these promoter sequences to regulatory proteins in immature bone cells when this skeletal-restricted gene is suppressed [60,76–80]. In response to developmental and skeletal regulatory signals the striking

CHAPTER 19 Intranuclear Organization of the Regulatory Machinery for Vitamin D

A

Inactive Osteocalcin Gene

B

Basal Transcription OC Gene Distal DHS

C

331

Proximal DHS

Vitamin D Induced Transcription of OC Gene Distal DHS

D

Proximal DHS

Proximal DNasel HS

Nuclear Matrix Distal DNasel HS

FIGURE 3 Remodeling of the Osteocalcin gene promoter during developmental progression of the osteoblast phenotype. The transcriptionally silent rat Osteocalcin gene is schematically illustrated with nucleosomes placed in the proximal tissue-specific and distal enhancer region of the promoter (A). Factors that support basal tissue-specific transcription are recruited to the OC gene promoter and are organized in proximal and distal promoter domains. Modifications in chromatin structure that mediate assembly of the regulatory machinery for the nuclease hypersensitive sites reflect OC gene transcription. A positioned nucleosome resides between the proximal basal and distal enhancer regions of the promoter (B). In response to vitamin D, chromatin remodeling renders the upstream VDRE competent for binding the VDR/RXR heterodimer with its cognate element (C). Higher order chromatin organization permits cross talk between basal transcription machinery and the vitamin D receptor complex that involves direct interactions of the vitamin D receptor, Runx2, and TFIIB (D).

removal of a nucleosome and modifications in chromatin structure render the proximal promoter of the OC gene accessible to regulatory and co-regulatory proteins that support basal level activity [48,48,60,78]. Vitamin D enhancement of osteocalcin gene transcription is associated with removal of the nucleosome at the upstream vitamin D–responsive element that permits binding of the vitamin D receptor–RXR heterodimer [48,60,78,81]. The retention of a nucleosome

between the proximal and upstream enhancer domain reduces distance between the basal and vitamin D–responsive element and supports a promoter configuration that is conducive to protein–protein interactions between the vitamin D receptor and the basal TFIIB transcription factor [82–84]. Interaction of the vitamin D receptor at the distal promoter region of the bone-specific osteocalcin gene requires nucleosomal remodeling [85].

332 Thus, insight into control of skeletal gene expression can be obtained from an understanding of mechanisms that alter osteocalcin gene chromatin organization under biological conditions. Site-directed mutagenesis of osteocalcin genes that are genetically integrated in stable cell lines has established that RUNX elements flanking the proximal and upstream promoter sequences are responsible for developmental and vitamin D– induced chromatin remodeling [48]. The recent demonstration of functional interactions between the VDR and flanking Runx proteins is consistent with linkage of VDRE organization and Runx regulatory elements [86]. Reduced CpG methylation is associated with transcriptional activation of the bone-specific osteocalcin gene in osteoblasts [87]. In vitro and in vivo genetic approaches have demonstrated that RUNX2 controls developmental and steroid hormone–responsive chromatin reconfiguration of the osteocalcin gene promoter [48,80]. Chromatin immunoprecipitation analyses have shown that developmental and vitamin D–linked remodeling of osteocalcin gene promoter organization is accompanied by acetylation of histones in the proximal basal and upstream vitamin D responsive element domains [88,89]. This posttranslational modification of histone proteins reduces the tenacity of histone DNA interactions in a manner that is conducive to an open chromatin organization with increased access to regulatory factors. The most compelling evidence for a functional involvement of chromatin organization in skeletal gene expression is the obligatory relationship of dynamic changes in the biochemical and structural properties of osteocalcin gene promoter organization with competency for bone tissue-restricted and enhanced transcription in response to vitamin D [48]. Yet, despite the cogent support for a central role of chromatin remodeling in transcriptional control of the osteocalcin gene, there are open-ended questions. It is not justifiable to extrapolate from these findings to conclude that all genes that are activated and suppressed during skeletogenesis employ identical mechanisms. From a broader biological perspective there are multiple levels of control that must be mechanistically characterized to explain physiologically responsive regulation of chromatin structure within restricted and global genomic contexts.

V. NUCLEAR MICROENVIRONMENTS: ACCOMMODATING THE RULES THAT GOVERN IN VIVO TRANSCRIPTIONAL CONTROL Key components of the basal transcription machinery and several tissue-specific transcription factor complexes

GARY S. STEIN,

ET AL .

are functionally compartmentalized as specialized subnuclear domains [11–13,16,27,39,40,90–102]. Such compartmentalization may, at least in part, accommodate biological constraints on the control of transcription in nuclei of intact bone cells. The low representation of promoter regulatory elements and cognate transcription factors necessitates a subnuclear organization of nucleic acids and regulatory proteins that supports threshold concentrations for the activation and repression of gene expression. From an historical perspective, compartmentalization of the regulatory machinery for ribosomal genes in nucleoli and the organization of chromosomes during mitosis provide paradigms for intranuclear localization of genes and regulatory complexes. During the past several years there has been growing recognition that the organization of nucleic acids and regulatory proteins is functionally linked to the assembly, organization, and activity of gene regulatory machinery. Cellular, molecular, biochemical, and genetic evidence indicates an obligatory relationship between sites within the nucleus where regulatory complexes reside and fidelity of transcriptional control. The biological relevance for the intranuclear distribution of regulatory complexes is directly reflected by aberrant nuclear structure–gene expression interrelationships that are associated with perturbations in skeletal development [18] and leukemia [16].

VI. SCAFFOLDING OF REGULATORY COMPONENTS FOR COMBINATORIAL CONTROL OF GENE EXPRESSION Functional interrelationships between nuclear structure and gene expression are strikingly reflected by dual recognition of regulatory proteins, such as RUNX transcription factors, for interactions with both promoter elements and co-regulatory proteins; such interactions modulate the structural and functional properties of targeted genes at microenvironments within the nucleus. Sequence-specific interactions with promoter elements result in placement of RUNX proteins at strategic sites where they provide scaffolds for protein–protein interactions that mediate the organization of machinery for a broad spectrum of regulatory requirements. These include histone modifications and chromatin remodeling that establish competency for transcription factor binding and genomic conformations that interface activities at proximal and upstream promoter domains, as well as the integration of regulatory cues from signaling pathways that activate or suppress gene expression in a physiologically responsive manner. As a consequence, the RUNX proteins are posttranslationally modified

CHAPTER 19 Intranuclear Organization of the Regulatory Machinery for Vitamin D

(e.g., phosphorylated) to further influence the extent to which they engage in regulatory activity. The complexity of RUNX regulatory proteins that assemble as supercomplexes of transcriptional regulatory factors illustrates the potential impact on skeletalrelated gene expression. Recent documentation that RUNX proteins are components of a stable complex that includes basal transcription factors, chromatin remodeling factors, and histone-modifying factors indicates the scope of RUNX-mediated combinatorial control. A key component of the RUNX complex is the p300/CBP coactivator that functions as a transcriptional adaptor. Interactions with several transcription factors results in the formation of multimolecular complexes that regulate expression of a broad spectrum of genes [103]. p300 contains a domain with intrinsic histone acetyltransferase (HAT) activity [104,105], which has been implicated in chromatin structure alterations associated with modulation of gene expression [106]. p300 interacts with additional proteins containing HAT activity that include P/CAF, SRC-1, and ACTR. A basis is thereby provided for formation of large multiprotein complexes that contribute multiple HAT activities with options for specificity [107–111]. It has been established that RUNX2 and p300 are components of the same nuclear complexes in osteoblastic cells [112]. Furthermore, when recruited to the osteocalcin gene promoter by RUNX2, p300 stimulates both basal and vitamin D-enhanced osteocalcin promoter activity. Thus interactions of RUNX2 with p300 supports assembly of multisubunit complexes with several HAT-containing proteins at a series of regulatory regions of the bone-specific osteocalcin gene promoter. In a parallel manner, Kitabayashi et al. [113] have shown that in myeloid cells RUNX1, a homolog of the bonespecific RUNX2, interacts with p300 and together they up-regulate myeloid-specific genes. It was also determined that a C-terminal region of the Runt domain in both RUNX1 and RUNX2, is critical for their interactions with p300 [112,113]. Considering the high degree of homology between these two members of the RUNX transcription factor family it is likely that the structural determinants for RUNX interactions with p300 are conserved. In addition to functioning as transcriptional activators, RUNX proteins suppress gene expression (transcription). Repression requires the recruitment of transcriptional repressors and co-repressors with histone deacetylase activity (HDACs) to promoter regulatory elements of genes that are down-regulated. Combinatorial control that dampens transcription is illustrated by interaction of RUNX2 with the transcriptional co-repressors TLE/ Groucho through a conserved VWRPY domain located at the C terminus of the protein, which represses the

333

expression of the bone-sialo protein (BSP) gene in osteoblastic cells [32]. Another example of combinatorial control that results in transcriptional suppression by RUNX2 is down-regulation of the p21CIP/WAF promoter in fibroblastic and osteoblastic cells. Here HDAC6 interacts with a second repression domain that also resides in the C-terminal region of RUNX2 and is recruited to chromatin by RUNX2 [114]. Taken together, these results are consistent with combinatorial control that is mediated by RUNX-dependent recruitment of coactivator and co-repressors proteins that are associated with and organized as multiprotein complexes to activate or repress target genes in a physiologically responsive manner. p300 can also be recruited to gene promoters by the transcription factor C/EBP [115,116]. Interestingly, a C/EBP-responsive regulatory element has been identified in the proximal promoter region of the rat OC gene adjacent to the RUNX2 site C [117]. C/EBPβ physically interacts with RUNX2 and synergistically activates the osteocalcin promoter [117], suggesting that both proteins form a complex with p300 and together up-regulate basal tissue-specific transcription. C/EBPβ has additionally been shown to interact with ATPdependent chromatin remodeling complexes of the SWI/SNF family [118], recruiting these complexes to promoter sequences and activating cell-specific expression. Taken together, these findings indicate that RUNX factors engage in protein–DNA and protein–protein interactions that collectively determine the composition and organization of promoter regulatory complexes. The inclusion of chromatin remodeling activity in these multisubunit complexes provides a biochemical basis for conformational modifications of promoter elements as well as combinatorial specificity for transcription. Transcription factors that function as scaffolds for interaction with co-regulatory proteins provide an architectural basis for accommodating the combinatorial requirements of biological control. Combinatorial control supports replication, transcription, and repair by two mechanisms. Context-dependent combinations and permutations of regulatory proteins are assembled into multipartite complexes that increase specificity. Scaffold-associated protein–DNA and protein–protein interactions permit integration of regulatory activities. Nuclear microenvironments are thereby organized, with gene promoters as focal points, where threshold concentrations of regulatory macromolecules are attained. The complexity that is achieved by these architecturally organized oligomeric factors can maximize options for responsiveness to diverse regulatory requirements for transient and long term biological control.

334 VII. INTRANUCLEAR TRAFFICKING OF SKELETAL REGULATORY FACTORS TO SUBNUCLEAR SITES THAT SUPPORT TRANSCRIPTION: ‘‘TO BE IN THE RIGHT PLACE AT THE RIGHT TIME’’ There is a need to gain insight into mechanisms that direct skeletal factors to subnuclear sites where regulatory events occur. Association of osteoblast, myeloid, and lymphoid RUNX transcription factors that mediate tissue-specific transcription with the nuclear matrix has permitted direct examination of mechanisms for targeting regulatory proteins to transcriptionally active subnuclear domains [12,45–47,95,119–126]. Both biochemical and immunofluorescence analyses have shown that RUNX transcription factors exhibit a punctate nuclear distribution that is associated with the nuclear matrix in situ [11,12,127,128]. Taken together, these observations are consistent with the concept that the nuclear matrix is functionally involved in gene localization and in the concentration and subnuclear localization of regulatory factors [12,93,94,129–132]. The initial indication that nuclear matrix association of RUNX factors is required for maximal activity was provided by the observation that transcriptionally active RUNX proteins associate with the nuclear matrix but inactive C-terminally truncated RUNX proteins do not [6,12,32,128,133] (Fig. 3). This localization of RUNX was established by biochemical fractionation and in situ immunofluorescence as well as by green fluorescent protein tagged RUNX proteins [31] in living cells. Colocalization of RUNX1, 2, and 3 at nuclear matrix–associated sites indicates a common intranuclear targeting mechanism may be operative for the family of RUNX transcription factors [31,32,127]. Variations in the partitioning of transcriptionally active and inactive RUNX between subnuclear fractions permitted development of a strategy to identify a region of the RUNX transcription factors that directs the regulatory proteins to nuclear matrix–associated foci. A series of deletions and internal mutations was constructed and assayed for competency to associate with the nuclear matrix by Western blot analysis of biochemically prepared nuclear fractions and by in situ immunostaining following transfection into intact cells. Association of osteogenic and hematopoietic RUNX proteins with the nuclear matrix is independent of DNA binding and requires a nuclear matrix targeting signal, a 31-amino-acid segment near the C terminus that is distinct from nuclear localization signals [12]. The nuclear matrix targeting signal functions autonomously and is necessary as well as sufficient to direct the transcriptionally

GARY S. STEIN,

ET AL .

active RUNX transcription factors to nuclear matrix– associated sites where gene expression occurs [12]. These findings indicate mechanisms involved in the selective trafficking of proteins to specialized domains within the nucleus where they become components of functional regulatory complexes. At least two trafficking signals appear to be required for subnuclear targeting of RUNX transcription factors; the first supports nuclear import (nuclear localization signal) and a second mediates association with the nuclear matrix (nuclear matrix targeting signal). The multiplicity of determinants for nuclear localization and alternative splicing of RUNX messenger RNA may provide the requisite complexity to support targeting to specific sites within the nucleus in response to diverse biological conditions. Furthermore, because gene expression by RUNX involves contributions by factors and co-regulatory proteins that include CBFβ [46,134–139] and C/EBP [117,140], Groucho/TLE [32,141,142], HES, and SMAD [6,143], RUNX may facilitate recruitment of these factors to the nuclear matrix.

A. Properties of Transcriptionally Active Subnuclear Compartments Association of genes and cognate factors with the nuclear matrix may support the formation and/or activities of nuclear domains that facilitate transcriptional control [10,45,55,132,144–151]. Results from our laboratory indicate that the association of RUNX transcription factors with the nuclear matrix is obligatory for activity [11,18]. The promoter recognition function of the runt homology domain of RUNX, and thus the consequential interactions with RUNX-responsive genes, is essential for formation of transcriptionally active foci containing RUNX and RNA polymerase II that are nuclear matrix associated [11]. Additionally, the nuclear matrix targeting signal supports transactivation when associated with an appropriate promoter, and transcriptional activity of the nuclear matrix targeting signal depends on association with the nuclear matrix [11]. Taken together, targeting of RUNX transcription factors to the nuclear matrix is important for their function and transcription. However, components of the nuclear matrix that function as acceptor sites remain to be established. Characterization of such nuclear matrix components will provide an additional dimension to characterizing molecular mechanisms associated with gene expression—the targeting of regulatory proteins to specific spatial domains within the nucleus. An initial indication of transcription factor interactions with the nuclear matrix is provided by crystal structure of the

CHAPTER 19 Intranuclear Organization of the Regulatory Machinery for Vitamin D

RUNX nuclear matrix targeting signal that was determined by X-ray diffraction analysis at 2.7 Å [127,152].

B. Subnuclear Targeting and Integration of Signaling Pathways Gene expression during skeletal development and bone remodeling is controlled by a broad spectrum of regulatory signals that converge at promoter elements to activate or repress transcription in a physiologically responsive manner. The subnuclear compartmentalization of transcription machinery necessitates a mechanistic explanation for directing signaling factor to sites within the nucleus where gene expression occurs under conditions that support integration of regulatory cues. The interactions of YAP and SMAD co-regulatory proteins with C-terminal segments of the RUNX2 transcription factor permit assessment of requirements for recruitment of cSRC and BMP/TGFb-mediated signals to skeletal target genes. Our findings indicate that nuclear import of YAP and SMAD co-regulatory factors is agonist dependent. However, there is a stringent requirement for fidelity of RUNX subnuclear targeting for recruitment of these signaling proteins to transcriptionally active subnuclear foci. Our results demonstrate that the interactions and spatial–temporal organization of RUNX and SMAD as well as YAP co-regulatory proteins are essential for assembly of transcription machinery that supports expression of skeletal genes [6,128]. Competency for intranuclear trafficking of RUNX proteins has similarly been functionally linked with the subnuclear localization and activity of TLE/Groucho co-regulatory proteins [32]. These findings are consistent with proteins serving as a scaffold for interactions with co-regulatory proteins that contribute to biological control.

C. In Vivo Consequences of Aberrant Intranuclear Trafficking of RUNX Transcription Factors Using RUNX2 and its essential role in osteogenesis as a model, we investigated the fundamental importance of fidelity of subnuclear localization for tissue differentiating activity by deleting the intranuclear targeting signal via homologous recombination. Mice homozygous for the deletion (RUNX2∆C) do not form bone because of perturbed maturation or arrest of osteoblasts. Heterozygotes do not develop clavicles, but are otherwise normal. These phenotypes are

335

indistinguishable from those of the RUNX2 homozygous and heterozygous null mutants, indicating that the intranuclear targeting signal is a critical determinant for function. The expressed truncated RUNX2∆C protein enters the nucleus and retains normal DNA-binding activity, but shows complete loss of intranuclear targeting. These results establish that the multifunctional N-terminal region of the RUNX2 protein is not sufficient for biological activity. Our results demonstrate that subnuclear localization of RUNX factors in specific foci together with associated regulatory functions is essential for control of RUNXdependent genes involved in tissue differentiation during embryonic development [18]. The importance of subnuclear localization of RUNX transcription factors for biological control is further indicated by compromised subnuclear organization and activity of RUNX1 hematopoietic regulatory proteins in acute myelogenous leukemia [16].

VIII. THE REGULATED AND REGULATORY PARAMETERS OF SUBNUCLEAR ORGANIZATION Multiple lines of evidence suggest that components of nuclear architecture contribute both structurally and enzymatically to control gene expression during osteoblast differentiation. Sequences have been identified that direct RUNX transcription factors to nuclear matrix–associated sites that support transcription in a cell cycle–dependent manner [153]. Insight is thereby provided into mechanisms linked to the assembly and activities of subnuclear domains where transcription occurs. In a restricted sense, the foundation has been provided for experimentally addressing intranuclear trafficking of gene regulatory factors and control of association with the nuclear matrix to establish and sustain domains that are competent for transcription. The unique sequences [11,12] and crystal structure for the 31-amino-acid nuclear matrix targeting signal of RUNX transcription factors [127,152] support specificity for localization at intranuclear sites where the regulatory machinery for gene expression is assembled, rendered operative, and/or suppressed. In a broader context, there is a growing appreciation for involvement of nuclear architecture in a dynamic and bidirectional exchange of gene transcripts and regulatory factors between the nucleus and cytoplasm, as well as between regions and structures within the nucleus [13,27,35,154–156]. It would be presumptuous to propose a single model to account for the specific pathways that direct transcription

336 factors to sites within the nucleus that support transcription. However, findings suggest that parameters of nuclear architecture functionally interface with components of transcriptional control. The involvement of nuclear matrix–associated transcription factors with recruitment of regulatory components to modulate transcription remains to be defined. Working models that serve as frameworks for experimentally addressing components of transcriptional control within the context of nuclear architecture can be compatible with mechanisms that involve architecturally or activitydriven assembly of transcriptionally active intranuclear foci. The diversity of targeting signals must be established to evaluate the extent to which regulatory discrimination is mediated by encoded intranuclear trafficking signals. It will additionally be important to biochemically and mechanistically define the checkpoints, which are operative during subnuclear distribution of regulatory factors, and the editing steps, which are invoked to ensure that structural and functional fidelity of nuclear domains, where replication and expression of genes occur. There is emerging recognition that placement of regulatory components of gene expression must be temporally and spatially coordinated to optimally mediate biological control. It is realistic to anticipate that further understanding of mechanisms that position genes and regulatory factors for establishment and maintenance of the bone cell phenotype will clarify nuclear structure–function interrelationships that are operative during osteoblast differentiation and vitamin D modulation of regulatory activity.

GARY S. STEIN,

4. 5.

6. 7.

8. 9.

10. 11.

12.

13. 14.

Acknowledgments 15.

Results presented in this chapter were in part supported by grants for the National Institutes of Health (AR45688, PO1CA82834, DE12528, AR39588, AR45689, PO1AR48818, FIRCA R03 TW00990, FONDECYT 1030479/DK32520). The authors appreciate the editorial assistance of Elizabeth Bronstein and Karen Concaugh with the preparation of this manuscript.

16.

17.

18.

References 1. Bilezikian JP, Raisz LG, Rodan GA 2002 Principles of Bone Biology. Academic Press, San Diego. 2. Stein GS, Lian JB, Montecino M et al. 2002 Involvement of nuclear architecture in regulating gene expression in bone cells. In: Bilezikian JP, Raisz LG, Rodan GA (eds) Principles of Bone Biology. 1. Academic Press, San Diego, pp. 169–188. 3. Stein GS, van Wijnen AJ, Stein JL, Lian JB, Montecino M 1997 Nuclear architecture in developmental transcriptional

19. 20. 21.

ET AL .

control of cell growth and tissue-specific genes. In: Bird RC, Stein GS, Lian JB, Stein JL (eds) Nuclear Structure and Gene Expression. Academic Press, New York, pp. 178–214. Lian JB, Stein GS 1999 The cells of bone. In: Seibel MJ, Robins SP, Bilezikian JP (eds) Dynamics of Bone and Cartilage Metabolism. Academic Press, San Diego, pp. 165–185. Xiao G, Gopalkrishnan RV, Jiang D et al. 2002 Bone morphogenetic proteins, extracellular matrix, and mitogenactivated protein kinase signaling pathways are required for osteoblast-specific gene expression and differentiation in MC3T3-E1 cells. J Bone Miner Res 17:101–110. Zaidi SK, Sullivan AJ, van Wijnen AJ et al. 2002 Integration of Runx and Smad regulatory signals at transcriptionally active subnuclear sites. Proc Natl Acad Sci USA 99:8048–8053. Lian JB, Staal A, van Wijnen AJ, Stein JL, Stein GS 1999 Biologic and molecular effects of vitamin D on bone. In: Holick MF (ed) Vitamin D: Molecular Biology, Physiology, and Clinical Applications. Humana Press, Totowa, NJ, pp. 175–193. Aubin JE, Liu F 1996 The osteoblast lineage. In: Bilezikian JP, Raisz LG, Rodan GA (eds) Principles of Bone Biology. Academic Press, San Diego, pp. 51–68. Schinke T, Karsenty G 2002 Transcriptional control of osteoblast differentiation and function. In: Bilezikian JP, Raisz LG, Rodan GA (eds) Principles of Bone Biology. 1. Academic Press, San Diego, pp. 83–91. Berezney R, Mortillaro M, Ma H et al. 1996 Connecting nuclear architecture and genomic function. J Cell Biochem 62:223–226. Zeng C, McNeil S, Pockwinse S et al. 1998 Intranuclear targeting of AML/CBFα regulatory factors to nuclear matrixassociated transcriptional domains. Proc Natl Acad Sci USA 95:1585–1589. Zeng C, van Wijnen AJ, Stein JL et al. 1997 Identification of a nuclear matrix targeting signal in the leukemia and bonerelated AML/CBFα transcription factors. Proc Natl Acad Sci USA 94:6746–6751. Lamond AI, Earnshaw WC 1998 Structure and function in the nucleus. Science 280:547–553. Ma H, Siegel AJ, Berezney R 1999 Association of chromosome territories with the nuclear matrix. Disruption of human chromosome territories correlates with the release of a subset of nuclear matrix proteins. J Cell Biol 146:531–542. Ma H, Samarabandu J, Devdhar RS et al. 1998 Spatial and temporal dynamics of DNA replication sites in mammalian cells. J Cell Biol 143:1415–1425. McNeil S, Zeng C, Harrington KS et al. 1999 The t(8;21) chromosomal translocation in acute myelogenous leukemia modifies intranuclear targeting of the AML1/CBFalpha2 transcription factor. Proc Natl Acad Sci USA 96:14882–14887. McNeil S, Guo B, Stein JL et al. 1998 Targeting of the YY1 transcription factor to the nucleolus and the nuclear matrix in situ: the C-terminus is a principal determinant for nuclear trafficking. J Cell Biochem 68:500–510. Choi J-Y, Pratap J, Javed A et al. 2001 Subnuclear targeting of Runx/Cbfa/AML factors is essential for tissue-specific differentiation during embryonic development. Proc Natl Acad Sci USA 98:8650–8655. DeFranco DB 2002 Navigating steroid hormone receptors through the nuclear compartment. Mol Endocrinol 16: 1449–1455. Gasser SM 2002 Visualizing chromatin dynamics in interphase nuclei. Science 296:1412–1416. Strahl BD, Grant PA, Briggs SD et al. 2002 Set2 is a nucleosomal histone H3-selective methyltransferase that mediates transcriptional repression. Mol Cell Biol 22:1298–1306.

CHAPTER 19 Intranuclear Organization of the Regulatory Machinery for Vitamin D

22. Iborra FJ, Jackson DA, Cook PR 2000 The path of RNA through nuclear pores: apparent entry from the sides into specialized pores [In Process Citation]. J Cell Sci 113 (Pt 2): 291–302. 23. Mattaj IW, Englmeier L 1998 Nucleocytoplasmic transport: the soluble phase. Annu Rev Biochem 67:265–306. 24. Berezney R, Jeon KW 1995 Structural and Functional Organization of the Nuclear Matrix. Academic Press, New York. 25. Penman S 1995 Rethinking cell structure. Proc Natl Acad Sci USA 92:5251–5257. 26. Misteli T 2000 Cell biology of transcription and pre-mRNA splicing: nuclear architecture meets nuclear function. J Cell Sci 113:1841–1849. 27. Stein GS, van Wijnen AJ, Stein JL et al. 2000 Intranuclear trafficking of transcription factors: implications for biological control. J Cell Sci 113:2527–2533. 28. Dyck JA, Maul GG, Miller WH et al. 1994 A novel macromolecular structure is a target of the promyelocyte-retinoic acid receptor oncoprotein. Cell 76:333–343. 29. Grande MA, van der Kraan I, van Steensel B et al. 1996 PML-containing nuclear bodies: their spatial distribution in relation to other nuclear components. J Cell Biochem 63:280–291. 30. Melnick A, Licht JD 1999 Deconstructing a disease: RARalpha, its fusion partners, and their roles in the pathogenesis of acute promyelocytic leukemia. Blood 93:3167–3215. 31. Harrington KS, Javed A, Drissi H et al. 2002 Transcription factors RUNX1/AML1 and RUNX2/Cbfa1 dynamically associate with stationary subnuclear domains. J Cell Sci 115:4167–4176. 32. Javed A, Guo B, Hiebert S et al. 2000 Groucho/TLE/R-Esp proteins associate with the nuclear matrix and repress RUNX (CBFα/AML/PEBP2α) dependent activation of tissuespecific gene transcription. J Cell Sci 113:2221–2231. 33. Ma H, Siegel AJ, Berezney R 1999 Association of chromosome territories with the nuclear matrix. Disruption of human chromosome territories correlates with the release of a subset of nuclear matrix proteins. J Cell Biol 146:531–542. 34. Leonhardt H, Rahn HP, Cardoso MC 1998 Intranuclear targeting of DNA replication factors. J Cell Biochem Suppl 30–31:243–249. 35. Wei X, Samarabandu J, Devdhar RS et al. 1998 Segregation of transcription and replication sites into higher order domains. Science 281:1502–1505. 36. Cook PR 1999 The organization of replication and transcription. Science 284:1790–1795. 37. Ciejek EM, Tsai MJ, O’Malley BW 1983 Actively transcribed genes are associated with the nuclear matrix. Nature 306: 607–609. 38. van Steensel B, Jenster G, Damm K, Brinkmann AO, van Driel R 1995 Domains of the human androgen receptor and glucocorticoid receptor involved in binding to the nuclear matrix. J Cell Biochem 57:465–478. 39. Htun H, Barsony J, Renyi I, Gould DL, Hager GL 1996 Visualization of glucocorticoid receptor translocation and intranuclear organization in living cells with a green fluorescent protein chimera. Proc Natl Acad Sci USA 93:4845–4850. 40. Stenoien D, Sharp ZD, Smith CL, Mancini MA 1998 Functional subnuclear partitioning of transcription factors. J Cell Biochem 70:213–221. 41. Kimura H, Tao Y, Roeder RG, Cook PR 1999 Quantitation of RNA polymerase II and its transcription factors in an HeLa cell: little soluble holoenzyme but significant amounts of polymerases attached to the nuclear substructure. Mol Cell Biol 19:5383–5392.

337

42. Verschure PJ, van Der Kraan I, Manders EM, van Driel R 1999 Spatial relationship between transcription sites and chromosome territories. J Cell Biol 147:13–24. 43. Misteli T, Spector DL 1999 RNA polymerase II targets pre-mRNA splicing factors to transcription sites in vivo. Mol Cell 3:697–705. 44. Smith KP, Moen PT, Wydner KL, Coleman JR, Lawrence JB 1999 Processing of endogenous pre-mRNAs in association with SC-35 domains is gene specific. J Cell Biol 144: 617–629. 45. Merriman HL, van Wijnen AJ, Hiebert S et al. 1995 The tissue-specific nuclear matrix protein, NMP-2, is a member of the AML/CBF/PEBP2/runt domain transcription factor family: interactions with the osteocalcin gene promoter. Biochemistry 34:13125–13132. 46. Banerjee C, Hiebert SW, Stein JL, Lian JB, Stein GS 1996 An AML-1 consensus sequence binds an osteoblast-specific complex and transcriptionally activates the osteocalcin gene. Proc Natl Acad Sci USA 93:4968–4973. 47. Ducy P, Zhang R, Geoffroy V, Ridall AL, Karsenty G 1997 Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. Cell 89:747–754. 48. Javed A, Gutierrez S, Montecino M et al. 1999 Multiple Cbfa/AML sites in the rat osteocalcin promoter are required for basal and vitamin D responsive transcription and contribute to chromatin organization. Mol Cell Biol 19:7491–7500. 49. Demay MB, Gerardi JM, DeLuca HF, Kronenberg HM 1990 DNA sequences in the rat osteocalcin gene that bind the 1,25-dihydroxyvitamin D3 receptor and confer responsive to 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 87:369–373. 50. Markose ER, Stein JL, Stein GS, Lian JB 1990 Vitamin Dmediated modifications in protein-DNA interactions at two promoter elements of the osteocalcin gene. Proc Natl Acad Sci USA 87:1701–1705. 51. Bortell R, Owen TA, Bidwell JP et al. 1992 Vitamin Dresponsive protein-DNA interactions at multiple promoter regulatory elements that contribute to the level of rat osteocalcin gene expression. Proc Natl Acad Sci USA 89:6119–6123. 52. Hoffmann HM, Catron KM, van Wijnen AJ et al. 1994 Transcriptional control of the tissue-specific, developmentally regulated osteocalcin gene requires a binding motif for the Msx family of homeodomain proteins. Proc Natl Acad Sci USA 91:12887–12891. 53. Tamura M, Noda M 1994 Identification of a DNA sequence involved in osteoblast-specific gene expression via interaction with helix–loop–helix (HLH)-type transcription factors. J Cell Biol 126:773–782. 54. Ducy P, Karsenty G 1995 Two distinct osteoblast-specific cis-acting elements control expression of a mouse osteocalcin gene. Mol Cell Biol 15:1858–1869. 55. Guo B, Odgren PR, van Wijnen AJ et al. 1995 The nuclear matrix protein NMP-1 is the transcription factor YY1. Proc Natl Acad Sci USA 92:10526–10530. 56. Workman JL, Kingston RE 1998 Alteration of nucleosome structure as a mechanism of transcriptional regulation. Annu Rev Biochem 67:545–579. 57. Zhang Y, Sun ZW, Iratni R et al. 1998 SAP30, a novel protein conserved between human and yeast, is a component of a histone deacetylase complex. Mol Cell 1:1021–1031. 58. Ayer DE 1999 Histone deacetylases: transcriptional repression with SINers and NuRDs. Trends Cell Biol 9:193–198. 59. Kornberg RD, Lorch Y 1999 Chromatin-modifying and -remodeling complexes. Curr Opin Genet Dev 9:148–151.

338 60. Montecino M, Frenkel B, van Wijnen AJ et al. 1999 Chromatin hyperacetylation abrogates vitamin D-mediated transcriptional upregulation of the tissue-specific osteocalcin gene in vivo. Biochemistry 38:1338–1345. 61. Safadi FF, Rane SG, Gentile MA et al. 1999 Cancellous osteopenia in mice lacking cyclin dependent kinase-4 (CDK-4). J Bone Miner Res 14:S164. 62. Agalioti T, Lomvardas S, Parekh B et al. 2000 Ordered recruitment of chromatin modifying and general transcription factors to the IFN-beta promoter. Cell 103:667–678. 63. Hassan AH, Neely KE, Workman JL 2001 Histone acetyltransferase complexes stabilize swi/snf binding to promoter nucleosomes. Cell 104:817–827. 64. Imbalzano AN 1998 Energy-dependent chromatin remodelers: complex complexes and their components. Crit Rev Eukaryot Gene Expr 8:225–255. 65. Lorch Y, Cairns BR, Zhang M, Kornberg RD 1998 Activated RSC-nucleosome complex and persistently altered form of the nucleosome. Cell 94:29–34. 66. Peterson CL, Zhao Y, Chait BT 1998 Subunits of the yeast SWI/SNF complex are members of the actin-related protein (ARP) family. J Biol Chem 273:23641–23644. 67. Schnitzler G, Sif S, Kingston RE 1998 Human SWI/SNF interconverts a nucleosome between its base state and a stable remodeled state. Cell 94:17–27. 68. de la Serna I, Carlson KA, Imbalzano AN 2001 Mammalian SWI/SNF complexes promote MyoD-mediated muscle differentiation. Nat Genet 27:187–190. 69. de la Serna I, Imbalzano AN 2002 Unfolding heterochromatin for replication. Nat Genet 32:560–562. 70. Carrozza MJ, Utley RT, Workman JL, Cote J 2003 The diverse functions of histone acetyltransferase complexes. Trends Genet 19:321–329. 71. Neely KE, Workman JL 2002 Histone acetylation and chromatin remodeling: which comes first? Mol Genet Metab 76:1–5. 72. Fan HY, He X, Kingston RE, Narlikar GJ 2003 Distinct strategies to make nucleosomal DNA accessible. Mol Cell 11:1311–1322. 73. Narlikar GJ, Fan HY, Kingston RE 2002 Cooperation between complexes that regulate chromatin structure and transcription. Cell 108:475–487. 74. Jaskelioff M, Peterson CL 2003 Chromatin and transcription: histones continue to make their marks. Nat Cell Biol 5:395–399. 75. Peterson CL 2003 Transcriptional activation: getting a grip on condensed chromatin. Curr Biol 13:R195–R197. 76. Breen EC, van Wijnen AJ, Lian JB, Stein GS, Stein JL 1994 In vivo occupancy of the vitamin D responsive element in the osteocalcin gene supports vitamin D–dependent transcriptional upregulation in intact cells. Proc Natl Acad Sci USA 91:12902–12906. 77. Montecino M, Lian J, Stein G, Stein J 1994 Specific nucleosomal organization supports developmentally regulated expression of the osteocalcin gene [abstract]. J Bone Miner Res 9:S352. 78. Montecino M, Lian J, Stein G, Stein J 1996 Changes in chromatin structure support constitutive and developmentally regulated transcription of the bone-specific osteocalcin gene in osteoblastic cells. Biochemistry 35:5093–5102. 79. Paredes R, Gutierrez J, Gutierrez S et al. 2002 Interaction of the 1alpha,25-dihydroxyvitamin D3 receptor at the distal promoter region of the bone-specific osteocalcin gene requires nucleosomal remodelling. Biochem J 363:667–676. 80. Gutierrez J, Sierra J, Medina R et al. 2000 Interaction of CBFα/AML/PEBP2α transcription factors with nucleosomal

GARY S. STEIN,

81.

82.

83. 84.

85.

86.

87.

88.

89.

90. 91. 92.

93.

94.

95.

96.

ET AL .

sequences requires flexibility in the translational positioning of the histone octamer and exposure of the Cbfα site. Biochemistry 39:13565–13574. Paredes R, Gutierrez J, Gutierrez S et al. 2002 Interaction of the 1alpha,25-dihydroxyvitamin D3 receptor at the distal promoter region of the bone-specific osteocalcin gene requires nucleosomal remodelling. Biochem J 363:667–676. Blanco JCG, Wang I-M, Tsai SY et al. 1995 Transcription factor TFIIB and the vitamin D receptor cooperatively activate ligand-dependent transcription. Proc Natl Acad Sci USA 92:1535–1539. MacDonald PN, Sherman DR, Dowd DR, Jefcoat SC, Jr., DeLisle RK 1995 The vitamin D receptor interacts with general transcription factor IIB. J Biol Chem 270:4748–4752. Guo B, Aslam F, van Wijnen AJ et al. 1997 YY1 regulates VDR/RXR mediated transactivation of the vitamin D responsive osteocalcin gene. Proc Natl Acad Sci USA 94: 121–126. Paredes R, Gutierrez J, Gutierrez S et al. 2002 Interaction of the 1alpha,25-dihydroxyvitamin D3 receptor at the distal promoter region of the bone-specific osteocalcin gene requires nucleosomal remodelling. Biochem J 363:667–676. Paredes R, Arriagada G, Cruzat F, Villagra A, Olate J, Zaidi SK et al. 2003 The Runx2 transcription factor interacts with the 1α,25-dihydroxy vitamin D3 receptor and facilitates 1α,25-dihydroxy vitamin D3–dependent upregulation of the rat OC gene expression in osteoblastic cells. Unpublished. Villagra A, Gutierrez J, Paredes R et al. 2002 Reduced CpG methylation is associated with transcriptional activation of the bone-specific rat osteocalcin gene in osteoblasts. J Cell Biochem 85:112–122. Shen J, Montecino MA, Lian JB et al. 2002 Histone acetylation in vivo at the osteocalcin locus is functionally linked to vitamin D dependent, bone tissue-specific transcription. J Biol Chem 277:20284–20292. Shen J, Hovhannisyan H, Lian JB et al. 2003 Transcriptional induction of the osteocalcin gene during osteoblast differentiation involves acetylation of histones h3 and h4. Mol Endocrinol 17:743–756. Robinson SI, Nelkin BD, Vogelstein B 1982 The ovalbumin gene is associated with the nuclear matrix of chicken oviduct cells. Cell 28:99–106. Stief A, Winter DM, Stratling WH, Sippel AE 1989 A nuclear DNA attachment element mediates elevated and positionindependent gene activity. Nature 341:343–345. Schaack J, Ho WY, Freimuth P, Shenk T 1990 Adenovirus terminal protein mediates both nuclear matrix association and efficient transcription of adenovirus DNA. Genes Dev 4:1197–1208. Dworetzky SI, Wright KL, Fey EG et al. 1992 Sequencespecific DNA-binding proteins are components of a nuclear matrix-attachment site. Proc Natl Acad Sci USA 89: 4178–4182. van Wijnen AJ, Bidwell JP, Fey EG et al. 1993 Nuclear matrix association of multiple sequence-specific DNA binding activities related to SP-1, ATF, CCAAT, C/EBP, OCT-1, and AP-1. Biochemistry 32:8397–8402. Banerjee C, McCabe LR, Choi J-Y et al. 1997 Runt homology domain proteins in osteoblast differentiation: AML-3/CBFA1 is a major component of a bone specific complex. J Cell Biochem 66:1–8. Bangs P, Burke B, Powers C et al. 1998 Functional analysis of Tpr: identification of nuclear pore complex association and nuclear localization domains and a role in mRNA export. J Cell Biol 143:1801–1812.

CHAPTER 19 Intranuclear Organization of the Regulatory Machinery for Vitamin D

97. Ma H, Samarabandu J, Devdhar RS et al. 1998 Spatial and temporal dynamics of DNA replication sites in mammalian cells. J Cell Biol 143:1415–1425. 98. Tang Y, Getzenberg RH, Vietmeier BN et al. 1998 The DNAbinding and τ2 transactivation domains of the rat glucocorticoid receptor constitute a nuclear matrix targeting signal. Mol Endocrinol 12:1420–1431. 99. Misteli T, Spector DL 1999 RNA polymerase II targets pre-mRNA splicing factors to transcription sites in vivo. Mol Cell 3:697–705. 100. Wei X, Somanathan S, Samarabandu J, Berezney R 1999 Three-dimensional visualization of transcription sites and their association with splicing factor-rich nuclear speckles. J Cell Biol 146:543–558. 101. Kimura H, Tao Y, Roeder RG, Cook PR 1999 Quantitation of RNA polymerase II and its transcription factors in an HeLa cell: little soluble holoenzyme but significant amounts of polymerases attached to the nuclear substructure. Mol Cell Biol 19:5383–5392. 102. DeFranco DB 2002 Navigating steroid hormone receptors through the nuclear compartment. Mol Endocrinol 16: 1449–1455. 103. Goodman RH, Smolik S 2000 CBP/p300 in cell growth, transformation, and development. Genes Dev 14:1553–1577. 104. Bannister AJ, Oehler T, Wilhelm D, Angel P, Kouzarides T 1995 Stimulation of c-Jun activity by CBP: c-Jun residues Ser63/73 are required for CBP induced stimulation in vivo and CBP binding in vitro. Oncogene 11:2509–2514. 105. Ogryzko VV, Schiltz RL, Russanova V, Howard BH, Nakatani Y 1996 The transcriptional coactivators p300 and CBP are histone acetyltransferases. Cell 87:953–959. 106. Spencer VA, Davie JR 1999 Role of covalent modifications of histones in regulating gene expression. Gene 240:1–12. 107. Chakravarti D, LaMorte VJ, Nelson MC et al. 1996 Role of CBP/P300 in nuclear receptor signalling. Nature 383:99–103. 108. Yang XJ, Ogryzko VV, Nishikawa J, Howard BH, Nakatani Y 1996 A p300/CBP-associated factor that competes with the adenoviral oncoprotein E1A. Nature 382:319–324. 109. Chen H, Lin RJ, Schiltz RL et al. 1997 Nuclear receptor coactivator ACTR is a novel histone acetyltransferase and forms a multimeric activation complex with P/CAF and CBP/p300. Cell 90:569–580. 110. Spencer TE, Jenster G, Burcin MM et al. 1997 Steroid receptor coactivator-1 is a histone acetyltransferase. Nature 389: 194–198. 111. Torchia J, Rose DW, Inostroza J et al. 1997 The transcriptional co-activator p/CIP binds CBP and mediates nuclearreceptor function. Nature 387:677–684. 112. Sierra J, Villagra A, Paredes R et al. 2003 Regulation of the bone-specific osteocalcin gene by p300 requires Runx2/Cbfa1 and the vitamin D3 receptor but not p300 intrinsic histone acetyl transferase activity. Mol Cell Biol 23:3339–3351. 113. Kitabayashi I, Yokoyama A, Shimizu K, Ohki M 1998 Interaction and functional cooperation of the leukemiaassociated factors AML1 and p300 in myeloid cell differentiation. EMBO J 17:2994–3004. 114. Westendorf JJ, Zaidi SK, Cascino JE et al. 2002 Runx2 (Cbfa1, AML-3) interacts with histone deacetylase 6 and represses the p21(CIP1/WAF1) promoter. Mol Cell Biol 22:7982–7992. 115. Oelgeschlager M, Janknecht R, Krieg J, Schreek S, Luscher B 1996 Interaction of the co-activator CBP with Myb proteins: effects on Myb-specific transactivation and on the cooperativity with NF-. EMBO J 15:2771–2780.

339

116. Mink S, Haenig B, Klempnauer KH 1997 Interaction and functional collaboration of p300 and C/EBPbeta. Mol Cell Biol 17:6609–6617. 117. Gutierrez S, Javed A, Tennant D et al. 2002 CCAAT/ enhancer-binding proteins (C/EBP) β and δ Activate osteocalcin gene transcription and synergize with Runx2 at the C/EBP element to regulate bone-specific expression. J Biol Chem 277:1316–1323. 118. Kowenz-Leutz E, Leutz A 1999 A C/EBP beta isoform recruits the SWI/SNF complex to activate myeloid genes. Mol Cell 4:735–743. 119. Bae SC, Yamaguchi-Iwai Y, Ogawa E et al. 1993 Isolation of PEBP2αB cDNA representing the mouse homolog of human acute myeloid leukemia gene, AML1. Oncogene 8:809–814. 120. Meyers S, Downing JR, Hiebert SW 1993 Identification of AML-1 and the t(8;21) translocation protein (AML-1/ETO) as sequence-specific DNA-binding proteins; the runt homology domain is required for DNA binding and protein–protein interactions. Mol Cell Biol 13:6336–6345. 121. Meyers S, Lenny N, Sun W-H, Hiebert SW 1996 AML-2 is a potential target for transcriptional regulation by the t(8;21) and t(12;21) fusion proteins in acute leukemia. Oncogene 13:303–312. 122. Meyers S, Lenny N, Hiebert SW 1995 The t(8;21) fusion protein interferes with AML-1B-dependent transcriptional activation. Mol Cell Biol 15:1974–1982. 123. Wang S, Wang Q, Crute BE et al. 1993 Cloning and characterization of subunits of the T-cell receptor and murine leukemia virus enhancer core-binding factor. Mol Cell Biol 13:3324–3339. 124. Nuchprayoon I, Meyers S, Scott LM et al. 1994 PEBP2/CBF, the murine homolog of the human myeloid AML1 and PEBP2β/CBFβ proto-oncoproteins, regulates the murine myeloperoxidase and neutrophil elastase genes in immature myeloid cells. Mol Cell Biol 14:5558-5568. 125. Frank R, Zhang J, Uchida H et al. 1995 The AML1/ETO fusion protein blocks transactivation of the GM-CSF promoter by AML1B. Oncogene 11:2667–2674. 126. Satake M, Nomura S, Yamaguchi-Iwai Y et al. 1995 Expression of the Runt domain-encoding PEBP2 alpha genes in T cells during thymic development. Mol Cell Biol 15:1662–1670. 127. Tang L, Guo B, van Wijnen AJ et al. 1998 Preliminary crystallographic study of the glutathione S-transferase fused with the nuclear matrix targeting signal of the transcription factor AML-1/CBFα2. J Struct Biol 123:83–85. 128. Zaidi SK, Sullivan AJ, van Wijnen AJ et al. 2002 Functional integration of RUNX2 and BMP2 responsive regulatory signals at transcriptionally active subnuclear sites in osseous and non-osseous cells [abstract]. J Bone Miner Res 17:5439. 129. Bidwell JP, van Wijnen AJ, Fey EG et al. 1993 Osteocalcin gene promoter-binding factors are tissue-specific nuclear matrix components. Proc Natl Acad Sci USA 90:3162–3166. 130. Blencowe BJ, Nickerson JA, Issner R, Penman S, Sharp PA 1994 Association of nuclear matrix antigens with exoncontaining splicing complexes. J Cell Biol 127:593–607. 131. Mancini MA, Shan B, Nickerson JA, Penman S, Lee WH 1994 The retinoblastoma gene product is a cell cycle-dependent, nuclear matrix-associated protein. Proc Natl Acad Sci USA 91:418–422. 132. Nickerson JA, Blencowe BJ, Penman S 1995 The architectural organization of nuclear metabolism. In: Berezney R, Jeon KW (eds) Structural and Functional Organization of the Nuclear Matrix.162A. Academic Press, New York, pp. 67–123.

340 133. Choi J-Y, Javed A, Zaidi SK et al. 1999 Structure-function determinants for subnuclear targeting and transactivation in the Cbfa1/AML3 transcription factor [abstract]. J Bone Miner Res 14:S469. 134. Ogawa E, Inuzuka M, Maruyama M et al. 1993 Molecular cloning and characterization of PEBP2β, the heterodimeric partner of a novel Drosophila runt-related DNA binding protein PEBP2α. Virology 194:314–331. 135. Giese K, Kingsley C, Kirshner JR, Grosschedl R 1995 Assembly and function of a TCR alpha enhancer complex is dependent on LEF-1-induced DNA bending and multiple protein-protein interactions. Genes Dev 9:995–1008. 136. Mao S, Frank RC, Zhang J, Miyazaki Y, Nimer SD 1999 Functional and physical interactions between AML1 proteins and an ETS protein, MEF: implications for the pathogenesis of t(8;21)-positive leukemias. Mol Cell Biol 19:3635–3644. 137. Xie XQ, Pardali E, Holm M, Sideras P, Grundstrom T 1999 AML and Ets proteins regulate the I alpha1 germ-line promoter. Eur J Immunol 29:488–498. 138. Miller J, Horner A, Stacy T et al. 2002 The core-binding factor beta subunit is required for bone formation and hematopoietic maturation. Nat Genet 32:645–649. 139. Kundu M, Javed A, Jeon JP et al. 2002 Cbfbeta interacts with Runx2 and has a critical role in bone development. Nat Genet 32:639–644. 140. Zhang DE, Hetherington CJ, Meyers S et al. 1996 CCAAT enhancer-binding protein (C/EBP) and AML1 (CBFα2) synergistically activate the macrophage colony-stimulating factor receptor promoter. Mol Cell Biol 16:1231–1240. 141. Levanon D, Goldstein RE, Bernstein Y et al. 1998 Transcriptional repression by AML1 and LEF-1 is mediated by the TLE/Groucho corepressors. Proc Natl Acad Sci USA 95:11590–11595. 142. Javed A, Barnes GL, Jassanya BO et al. 2001 runt homology domain transcription factors (Runx, Cbfa, and AML) mediate repression of the bone sialoprotein promoter: evidence for promoter context-dependent activity of Cbfa proteins. Mol Cell Biol 21:2891–2905. 143. Zhang YW, Yasui N, Ito K et al. 2000 A RUNX2/PEBP2αA/ CBFA1 mutation displaying impaired transactivation and Smad interaction in cleidocranial dysplasia. Proc Natl Acad Sci USA 97:10549–10554. 144. Chen HY, Sun JM, Hendzel MJ, Rattner JB, Davie JR 1996 Changes in the nuclear matrix of chicken erythrocytes that accompany maturation. Biochem J 320:257–265.

GARY S. STEIN,

ET AL .

145. Nardozza TA, Quigley MM, Getzenberg RH 1996 Association of transcription factors with the nuclear matrix. J Cell Biochem 61:467–477. 146. Stein GS, Stein JL, Lian JB, van Wijnen AJ, Montecino M 1996 Functional interrelationships between nuclear structure and transcriptional control: contributions to regulation of cell cycle- and tissue-specific gene expression. J Cell Biochem 62:198–209. 147. Alvarez M, Long H, Onyia J et al. 1997 Rat osteoblast and osteosarcoma nuclear matrix proteins bind with sequence specificity to the rat type I collagen promoter. Endocrinology 138:482–489. 148. Davie JR 1997 Nuclear matrix, dynamic histone acetylation and transcriptionally active chromatin. Mol Biol Rep 24:197–207. 149. Grande MA, van der Kraan I, de Jong L, van Driel R 1997 Nuclear distribution of transcription factors in relation to sites of transcription and RNA polymerase II. J Cell Sci 110:1781–1791. 150. Jackson DA 1997 Chromatin domains and nuclear compartments: establishing sites of gene expression in eukaryotic nuclei. Mol Biol Rep. 24:209–220. 151. Lindenmuth DM, van Wijnen AJ, Hiebert S et al. 1997 Subcellular partitioning of transcription factors during osteoblast differentiation: developmental association of the AML/CBFα/PEBP2α-related transcription factorNMP-2 with the nuclear matrix. J Cell Biochem 66: 123–132. 152. Tang L, Guo B, Javed A et al. 1999 Crystal structure of the nuclear matrix targeting signal of the transcription factor AML-1/PEBP2αB/CBFα2. J Biol Chem 274: 33580–33586. 153. Young DW, Zaidi SK, Stein JL, Lian JB, van Wijnen AJ, Stein GS 2002 The dynamic spatial distribution of a tissuespecific transcription factor during mitosis. Unpublished. 154. Wei X, Somanathan S, Samarabandu J, Berezney R 1999 Three-dimensional visualization of transcription sites and their association with splicing factor-rich nuclear speckles. J Cell Biol 146:543–558. 155. Stein GS, Montecino M, van Wijnen AJ, Stein JL, Lian JB 2000 Nuclear structure–gene expression interrelationships: implications for aberrant gene expression in cancer. Cancer Res 60:2067–2076. 156. Gasser SM 2002 Visualizing chromatin dynamics in interphase nuclei. Science 296:1412–1416.

CHAPTER 20

Mouse Models of Vitamin D Receptor Ablation MARIE B. DEMAY

I. II. III. IV.

Endocrine Unit, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts

Introduction Effect on Growth and Mineral Ion Homeostasis Effect on Vascular System Effect on Reproduction

I. INTRODUCTION The absence of functional vitamin D receptors (VDRs) has been shown to be the molecular basis for the human disease HVDRR (see Chapter 72). Data from affected patients and their families have provided significant insight into the role of the VDR in vivo. However, the limitations of human studies have precluded detailed investigations of the abnormalities in mineral ion homeostasis and determinants of alopecia in these kindreds. Four laboratories have independently generated mice with targeted ablation of the VDR [1–4]. The resultant phenotype mirrors that seen in humans affected with HVDRR. The generation of these mice lacking functional VDRs has permitted detailed investigations into abnormalities in calcium absorption, bone formation, and parathyroid function that cannot be performed in humans. Furthermore, they have permitted novel investigations into the role of the VDR in nontraditional target organs such as the skin and vascular system.

II. EFFECT ON GROWTH AND MINERAL ION HOMEOSTASIS Mice homozygous for targeted ablation of the VDR are born with the expected Mendelian frequency, demonstrating that the VDR is not essential for embryonic development. The pups are phenotypically normal and indistinguishable from their wild-type and heterozygous littermates until the third week of life [1–4]. From 21 days of age, however, the knockout mice maintain a body weight approximately 10–15% lower than that of their sex-matched control littermates until approximately 6 to 8 months of age. From that point on, they VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Effect on the Immune System VI. Integument VII. Conclusions References

gradually begin to lose weight, and by 1 year of age are approximately half the size of their sex-matched littermates (Fig. 1). Correlating with normal immunoreactive PTH (iPTH) levels, the knockout mice are normocalcemic the first 2 weeks of life, although, by 21 days, secondary hyperparathyroidism develops, presumably as a consequence of impaired intestinal calcium absorption [1]. This time of onset of impaired intestinal calcium absorption correlates well with data obtained from studies in vitamin D–deficient rats [5]. These investigations demonstrated that intestinal calcium absorption was vitamin D–independent the first 2 weeks of life, then was gradually replaced by a vitamin D–dependent transport system, which predominated by 35 days. Studies in 10-week-old VDR-null mice confirmed a marked impairment in duodenal calcium absorption [3]. This was accompanied by a significant reduction in mRNAs encoding the calcium channels that are thought to regulate calcium entry into the enterocyte, CaT1 and ECaC (see Chapters 24 and 25). There was also a dramatic decrease in the expression of calbindin 9K, a 1,25-dihydroxyvitamin D– inducible gene thought to play a role in intracellular calcium transfer [2–4, 6]. This suggests that the 1,25-dihydroxyvitamin D–induced expression of these calcium channels and of calbindin 9K is a key determinant of intestinal calcium absorption and that this hormonal regulation requires the presence of the nuclear VDR. Interestingly, although ECaC and CaT1 mRNA levels in the kidney are unaffected by VDR status [3], renal calbindin 9K levels are dramatically decreased in the VDR-null mice [2–4, 6], correlating with an increase in renal calcium clearance. These data suggest that calbindin 9K deficiency likely plays an important role in the pathophysiology of renal calcium loss in the VDR-null mice. Copyright © 2005, Elsevier, Inc. All rights reserved.

342

MARIE B. DEMAY

FIGURE 1 Phenotype of VDR-null mice. At 1 year of age, the VDR-null mice (foreground) are significantly smaller than their wild-type littermates and demonstrate severe alopecia.

As a result of the PTH–mediated increase in urinary phosphorus clearance, the VDR knockout mice develop hypophosphatemia and by 21 days of age have serum phosphorus levels in the range of 6.5 mg/dl versus 10.5 mg/dl in control littermates. Initially, the secondary hyperparathyroidism is able to compensate for the decrease in intestinal calcium absorption and the knockout mice maintain normal serum ionized calcium levels until day 28. However, from that point onward, ionized calcium levels are decreased by 30%: those of the VDR knockout mice stabilizing at approximately 1.0 mmol/liter versus 1.3 mmol/liter in control littermates [1–4]. At 15 days of age, there is no detectable abnormality in parathyroid function in the VDR knockout mice. They have normal serum iPTH levels and the size of their parathyroid glands is not increased [1], suggesting that the transcription-repressing effects of 1,25-dihydroxyvitamin D on the PTH gene and its antiproliferative effects on parathyroid cells observed in vitro (see Chapter 30) are not required for normal parathyroid development or do not require the actions of the nuclear VDR. However, in association with the development of impaired intestinal calcium absorption, secondary hyperparathyroidism is observed in the VDR-null mice. There is a gradual onset of parathyroid hyperplasia due to increased parathyroid cellular proliferation, evidenced by a twofold increase in the

number of parathyroid cells expressing proliferating cell nuclear antigen at 35 days of age [7]. By 70 days of age, when the iPTH levels are approximately 16-fold elevated, the VDR knockout mice have a 10-fold increase in parathyroid glandular volume, accompanied by an increase in PTH mRNA levels, as assessed by in situ hybridization analyses [1]. When the VDR-null mice are treated with a special diet that enables maintenance of normal mineral ion homeostasis, hyperparathyroidism is not observed, nor is parathyroid glandular hyperplasia [7] (Fig. 2). This suggests that the nuclear VDR is not required for normal parathyroid function when serum calcium and phosphorus levels are normal. Perhaps the role of the VDR in the parathyroid cell is important as a response to hypercalcemic or hypocalcemic stress, or alternatively its role may be redundant. The lack of parathyroid abnormalities in normocalcemic VDR-null mice does, however, suggest that hypocalcemia is the major stimulus to PTH gene transcription and parathyroid glandular proliferation when the actions of 1,25-dihydroxyvitamin D are impaired. Analogous to findings in humans with VDR mutations and in states of vitamin D deficiency during growth (see Chapter 63), VDR-null mice develop rickets and osteomalacia [1–4]. The skeletons of 2-week-old VDR knockout mice are indistinguishable from those of their control littermates, both radiologically and histologically, demonstrating that the VDR is not required

343

CHAPTER 20 Mouse Models of Vitamin D Receptor Ablation

A B C FIGURE 2 Parathyroid hyperplasia in the VDR-null mice is prevented by normalizing mineral ions. (A) Parathyroid gland from a 70-day-old wild-type mouse; (B) from a 70-day-old VDR-null littermate fed regular chow; (C) from a 70-day-old VDR-null mouse with normal mineral ion homeostasis. The asterisk marks the parathyroid gland. (See color plate).

for skeletal development. However, expansion of the hypertrophic chondrocyte layer of the growth plate is evident histologically at 21 days [8], and by 35 days of age, there is a dramatic expansion in this layer with disruption of the columnar alignment of the hypertrophic chondrocytes (Fig. 3). Histological analyses at this time also reveal osteomalacia, characterized by a decrease in bone mineralization and an increase in osteoid volume. Histomorphometric analyses of the skeletons of 70-day-old mice reveal a 30-fold increase in osteoid volume in the VDR knockout mice, compared to that observed in their wild-type and heterozygous littermates. This is associated with an increase

A

B

in bone volume/tissue volume, a consequence of an increase in both trabecular thickness and trabecular number (Table I). As a result of the dramatic impairment in bone mineralization, the bones of the VDR knockout mice demonstrate reduced stiffness and decreased strength during biomechanical testing [9]. Analyses of cellular content reveal an increase in osteoblast number, presumably due to increased bone turnover associated with secondary hyperparathyroidism. Despite the marked secondary hyperparathyroidism, osteoclast number is not significantly increased in the VDR-null mice [9]. Since osteoclast differentiation in response to PTH has been shown to be normal in vitro

C

FIGURE 3 Prevention of rickets by normalizing mineral ion homeostasis. (A) Tibial growth plate from a 35-day-old wild-type mouse; (B) from a 35-day-old VDR-null littermate fed regular chow, demonstrating marked expansion of the hypertrophic chondrocyte layer with distortion of the normal columnar structure; (C) from a 35-day-old VDR-null mouse with normal mineral ion homeostasis. (See color plate).

344

MARIE B. DEMAY

TABLE I Histomorphometric Parameters in VDR Knockout Mice On regular chow Wild-type OS/BS OV/BV BV/TV

20.9 + 2.9% 1.7 + 0.16% 10.3 + 2.4%

Knockout 86.4 + 2.8% 51.3 + 3.4% 54.5 + 2.9%

On diet that normalizes mineral ions Wild-type

Knockout

22.2 + 3.6% 1.4 + 0.3% 13.1 + 1.4%

23.1 + 1.7% 1.2 + 0.2% 14.3 + 1.2%

OS, osteoid surface; BS, bone surface; OV, osteoid volume; BV, bone volume; TV, tissue volume. Adapted from [9].

in the VDR-null mice [10], an increase in osteoclast number would have been anticipated. The inappropriately normal osteoclast number may be due to an impairment in osteoclast attachment to bone in the VDR-null mice, since the osteoclast αvβ3 integrin is highly induced by 1,25-dihydroxyvitamin D [11,12]. Alternatively, since osteoclasts are unable to resorb osteoid, their attachment to this unmineralized matrix may be impaired. As a consequence of the dramatic decrease in bone mineralization in the VDR knockout mice, their bones demonstrate reduced stiffness and strength on biomechanical testing compared to those of control littermates [9]. When normal mineral ion homeostasis is maintained in the VDR-null mice, all of these skeletal abnormalities are prevented [9]. Notably, there is normalization of the mineral apposition rate and osteomalacia is not observed. There is normalization of osteoid volume and of bone volume/tissue volume (Table I), suggesting that this abnormality in the hypocalcemic VDR-null mice is a consequence of an increase in matrix synthesis due to the anabolic effects of PTH, coupled with impaired bone resorption, since osteoclasts cannot resorb unmineralized matrix. It is notable that in the setting of normal serum calcium and phosphorus levels, histomorphometric analyses and biomechanical testing demonstrate that there is no detectable consequence of VDR ablation. These studies suggest that the VDR is not required for skeletal homeostasis. However, they do not exclude the possibility that the role of the VDR in the skeleton is redundant and that, in its absence, a second receptor or alternative homeostatic pathway performs its normal functions. Notable in this respect is the effect of 1,25-dihydroxyvitamin D on the regulation of RANK ligand production by the osteoblast. Although 1,25-dihydroxyvitamin D has been shown to play an important role in inducing the synthesis of this key regulator of osteoclast differentiation [13], histomorphometric analyses in the normocalcemic VDR-null mice reveal normal osteoclast numbers and

resorption surfaces. However, in vitro studies demonstrate that osteoblasts from VDR-null mice cannot support osteoclastogenesis when cocultured with normal spleen cells and 1,25-dihydroxyvitamin D, although when these cocultures are performed with PTH and interleukin 1α, osteoclasts with resorbing activity were formed [10]. These data clearly demonstrate that the VDR plays a key role in the skeleton; however, in its absence, other regulatory molecules, including cytokines, hormones, or alternative receptors, may be called upon to maintain skeletal homeostasis. Notable in this respect is the notion that the rapid actions of 1,25-dihydroxyvitamin D, such as the rapid increase in intracellular calcium and activation of second messengers, are mediated by an alternative receptor (see Chapters 23 and 33). However, studies in osteoblasts isolated from mice lacking the DNA binding domain of the VDR demonstrate that the rapid increase in intracellular calcium in response to hormone is abolished [4]. These data suggest that at least some of the rapid actions of this steroid hormone are dependent upon the presence of a functional nuclear receptor and challenge the hypothesis that the actions of a membrane receptor are responsible for the maintenance of skeletal homeostasis in the normocalcemic mice lacking functional nuclear receptors. The presence of a normal growth plate prior to the development of impaired mineral ion homeostasis suggests that the expansion of the late hypertrophic chondrocyte layer, characteristic of rickets, is a consequence of impaired intestinal calcium absorption rather than the absence of a functional receptor in the chondrocytes of the VDR-null mice. Furthermore, maintenance of normal mineral ion homeostasis prevents the development of rachitic changes [7] (Fig. 3), demonstrating that, in an ideal metabolic environment, the VDR is not essential for the development or maintenance of a normal growth plate. Since extracellular calcium promotes the expression of markers of terminal chondrocyte differentiation and increases the production of mineralized matrix in chondrocytic cell lines [14], studies

345

CHAPTER 20 Mouse Models of Vitamin D Receptor Ablation

were performed to address whether impaired mineral ion homeostasis led to abnormalities in chondrocyte differentiation. These studies demonstrated that the growth plate changes were not associated with a disruption in the acquisition of markers of chondrocyte differentiation, as assessed by in situ hybridization of long bones [8]. Interestingly, expansion of the late hypertrophic chondrocyte layer was observed at a time when the VDR null mice were normocalcemic, but had hypophosphatemia due to secondary hyperparathyroidism. Although no detectable change in chondrocyte proliferation was observed, there was marked impairment in apoptosis of the late hypertrophic chondrocytes in the hypophosphatemic VDR null mice [8]. In vitro studies have demonstrated that inorganic phosphate induces chondrocyte apoptosis in a time and dosedependent manner [15–17], suggesting that the development of hypophosphatemia is a critical event in the etiology of rickets in the VDR-null mice.

III. EFFECT ON VASCULAR SYSTEM Based on epidemiological studies demonstrating an inverse relationship between plasma 25-hydroxyvitamin D levels and blood pressure, investigations were performed in the VDR knockout mice to determine whether, in fact, these mice were predisposed to developing hypertension. The mean blood pressure in the VDR knockout mice was found to be 20 mmHg higher than that of wild-type mice [18]. This was associated with an increase in renin mRNA levels and in plasma angiotensin II levels. Despite an increase in these parameters under basal conditions, salt loading, a suppressor of renin expression, was capable of decreasing renin mRNA levels in the VDR knockout mice. However, the VDR knockout mice still maintained higher renin mRNA levels than controls treated in the same fashion. These data suggest that the regulatory mechanisms required for normal salt and water balance are operative in the VDR knockout mice, but that VDR ablation leads to an increase in basal renin activity. To determine whether this increase in renin was due to hormone deficiency, wild-type mice were rendered vitamin D deficient. Interestingly, a 50% elevation in renin mRNA levels was observed in the vitamin D–deficient mice with functional VDRs, demonstrating that this was a ligand-dependent effect [18]. Studies in cell models demonstrated that overexpression of the VDR in renal cells resulted in a decrease in renin mRNA levels and that regulatory regions in the renin gene could mediate transcriptional repression in response to 1,25-dihydroxyvitamin D (see Chapter 54).

IV. EFFECT ON REPRODUCTION Initial evaluation of the reproductive potential of VDR-null mice revealed variability among the knockout lines generated. Female mice with deletion of the second zinc finger were reported to be fertile [1], whereas those with deletion of the first zinc finger had infertility associated with uterine hypoplasia and impaired folliculogenesis [2]. Subsequent studies revealed that male VDR-null mice from this same line also had reproductive dysfunction, characterized by a decreased sperm count and a reduction in sperm motility. Aromatase activity in the ovary and testis were found to be reduced and evaluation of gonadotropins revealed hypergonadotropic hypogonadism, pointing to a defect at the level of the gonad, rather than reproductive dysfunction secondary to decreased body mass or intercurrent illness [19]. However, studies in mice lacking the DBD of the VDR showed normal uterine, testicular, and seminal vesicle weight as well as normal circulating levels of testosterone or estradiol [4]. Folliculogenesis and spermatogenesis were not impaired in this line of mice, nor was fertility. Interestingly, subsequent analyses revealed that calcium supplementation partially reversed the decrease in aromatase activity and normalized estrogen levels in the mice with ablation of the first zinc finger [19]. When VDR-null females from this same line of mice were fed a high-calcium diet, they had normal-sized litters and normal pup survival [20]; their fertility was shown to be directly related to the calcium content of the diet [21]. These studies once again point to the important contribution of abnormal mineral ion homeostasis to the phenotype of the VDR-null mice and raise the question as to what specific abnormality is responsible for the development of hypergonadotropic hypogonadism in the hypocalcemic mice with targeted ablation of the first zinc finger. Studies of mammary gland development have demonstrated an enhancement of growth and budding in the absence of a functional VDR. The glands of the VDR-null females were shown to have an increase in the number of secondary branch points and as well as in the number of terminal end buds [22]. The lack of VDR also enhanced the proliferative response to estrogen and progesterone. The effect of 1,25-dihydroxyvitamin D on growth of mammary glands in vitro demonstrated a receptor-dependent suppression of the sex steroid– induced branching enhancement in the wild-type mice, but not in those lacking a functional VDR [22]. These observations suggest that hormone-dependent effects of the VDR serve to attenuate mammary development and raise the question of the long-term consequences

346

MARIE B. DEMAY

of VDR ablation or vitamin D deficiency on mammary gland development and involution (see Chapter 93).

V. EFFECT ON THE IMMUNE SYSTEM A number of investigations have demonstrated that 1,25-dihydroxyvitamin D induces the expression of several genes with immunomodulatory properties in vitro as well as having immunomodulatory effects in vivo (see Chapters 98 and 99). These studies have demonstrated that 1,25-dihydroxyvitamin D and its analogs are able to prevent the development of autoimmune disorders and synergize with agents that prevent graft rejection in experimental models. Vitamin D–deficient animals have impaired chemotaxis and cell-mediated immunity, but the postulated immunomodulatory properties of the ligand-bound serum vitamin D binding protein raise the question as to what role the nuclear VDR plays in the immune dysfunction observed in vitamin D deficiency (see Chapters 8 and 9). Studies in the VDR-null mice have shown modest consequences of VDR ablation, most of which seem to be reversed by normalization of mineral ion homeostasis [23]. Splenocyte proliferation in response to antiCD3 stimulation is mildly impaired, as is macrophage chemotaxis [23]; however, macrophage phagocytosis and killing are not affected. Leucocyte and lymphocyte subset composition is also normal. In vivo rejection of transplanted islet cells was comparable in the knockout and wild-type mice, demonstrating that absence of the VDR did not enhance graft rejection [23]. Since vitamin D analogs have been shown to delay the onset of diabetes in NOD mice, the effect of VDR status on susceptibility to diabetes was examined. Interestingly, the

A

B

VDR-null mice were protected from low-dose streptozotocin-induced diabetes. All these immune defects, as well as protection from streoptozotocininduced diabetes, were reversed in the setting of normal mineral ion homeostasis [23]. VDR-null mice were shown to have an increase in mature splenic dendritic cells, consistent with the known effects of 1,25-dihydroxyvitamin D on inhibiting dendritic cell maturation [24]. Although studies in cells from the VDR-null mice have demonstrated that the VDR is required for the effects of 1,25-dihydroxyvitamin D on the differentiation of bone marrow progenitors into macrophages and monocytes [25], the lack of significant in vivo consequences of VDR ablation suggest that the role of the nuclear VDR in the immune system is redundant.

VI. INTEGUMENT Analogous to what is seen in numerous HVDRR kindreds, the VDR knockout mice develop progressive alopecia. Clinical evidence of alopecia is seen as early as the 4th week of life, beginning in the periorbital region and progressing dorsally, eventually resulting in total loss of fur by 3 to 4 months of age [1–4]. Histological examination of the skin reveals dilatation of hair follicles with formation of dermal cysts (Fig. 4). This histological phenotype is analogous to that seen in mice with mutations in the hairless gene, in which progressive loss of fur is seen from 2 weeks of age [26]. However, expression of hairless mRNA in the skin and keratinocytes of neonatal VDR null mice is not decreased [27], suggesting that, if these two genes are in a common pathway, the VDR is likely to be downstream of hairless.

C

FIGURE 4 Normalization of mineral ion homeostasis does not prevent skin changes in the VDR-null mice. (A) Skin section from a 70-day-old wild-type mouse; (B) from a 70-day-old VDR-null littermate fed regular chow, demonstrating dermal cysts and dilatation of hair follicles; (C) from a 70-day-old VDR-null mouse with normal mineral ion homeostasis. (See color plate).

CHAPTER 20 Mouse Models of Vitamin D Receptor Ablation

Because normalization of mineral ion homeostasis failed to prevent the development of the skin phenotype in the VDR-null mice [7] (Fig. 4), studies were undertaken to address whether the loss of fur was a consequence of an abnormality in keratinocytes, which give rise to the epidermal component of the hair follicle. Numerous studies have demonstrated that 1,25-dihydroxyvitamin D plays a significant role in decreasing keratinocyte proliferation and promoting keratinocyte differentiation (see Chapter 35). Studies in primary keratinocytes isolated from neonatal VDR knockout mice failed to demonstrate abnormal keratinocyte proliferation or impaired differentiation [27]. However, investigations in growing mice demonstrated a decrease in expression levels of epidermal differentiation markers, including involucrin, profilaggrin, and loricrin, from birth to 3 weeks of age [28]. Taken together, these data suggest that the effects of the VDR on keratinocytes are not essential during development, but that they are required for keratinocyte differentiation and skin homeostasis postnatally. Hair follicle development begins at embryonic day 14.5 and is dependent on reciprocal interactions between the epidermal and mesodermal components of the hair follicle. This morphogenic period lasts until the 3rd week of life, which marks the end of the first hair cycle. After this period, the hair follicle continues to cycle, characterized by a phase of rapid growth (anagen), followed by a regression phase (catagen), a quiescent phase (telogen), and reentry into anagen to generate a new hair shaft. Regulation of this postmorphogenic hair cycling is also thought to require continued epidermal–mesodermal communication. Abnormalities in the dermal papilla, the mesodermal component of the hair follicle, or in epidermal–mesodermal communication, required for the maintenance of the normal hair cycle, could also lead to the development of alopecia. Depilation of fur in control mice results in initiation of a new hair cycle with regrowth of fur. Such studies, performed at 18 days of age, a time when there is no detectable histological abnormality in the skin of the VDR-null mice, demonstrated that the VDR-null mice had a defect in anagen initiation [27]. The question remained as to whether this was due to a cellular or a metabolic defect. Because 1,25-dihydroxyvitamin D down-regulates its own biosynthesis by repressing the 25-hydroxyvitamin D-1α-hydroxylase gene [29] and increases its metabolism by up-regulating the 24-hydroxylase gene through VDR-dependent actions [30, 31], VDR-ablated mice have very high levels of 1,25-dihydroxyvitamin D even when mineral ion levels are normalized. Since 1,25-dihydroxyvitamin D has been shown to inhibit keratinocyte proliferation and to promote keratinocyte

347 differentiation [32], very high levels of this hormone or its metabolites may lead to alopecia by toxic interactions with an alternative receptor. However, VDR null mice with undetectable circulating levels of both 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D still developed alopecia, demonstrating that this phenotype was not a consequence of toxic levels of circulating hormone [33]. Interestingly, wild-type littermates with undetectable circulating levels of 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D did not develop alopecia [33]. This observation suggested that absence of receptor per se rather than absence of ligand-dependent receptor effects was responsible for the alopecia in the VDR-null mice and that different skin phenotypes are seen in absence of ligand versus absence of receptor. Since the VDR is widely expressed under normal circumstances and is missing from all cells of the VDR-null mice, it was not evident whether lack of VDR expression in keratinocytes, dermal papilla cells, or other organs of the VDR null-mice was responsible for the abnormality in the hair cycle. Hair reconstitution assays, in which implantation of a mixture of keratinocytes and activated dermal papilla cells into a nude mouse host recapitulates the process of hair follicle morphogenesis [34] leading to a functional hair follicle, provided a method to address whether the VDR was required in either the epidermal or mesodermal component of the hair follicle, or both. By implanting wild-type or VDR-null dermal papilla cells with wildtype or VDR-null keratinocytes, both isolated from neonatal mice in which the hair cycle is still in the morphogenic period, hair follicles were generated, with epidermal and mesodermal components differing in VDR status (wild-type versus null) [33]. These studies confirmed that the VDR was not required for hair follicle morphogenesis. However, hair follicles reconstituted with VDR-null keratinocytes were unable to sustain postmorphogenic hair cycles. In contrast, when dermal papilla cells of either genotype were mixed with wildtype keratinocytes, a normal response to anagen induction, including generation of new hair shafts, was observed. These studies demonstrated that expression of the VDR in the keratinocytes was essential for normal postmorphogenic hair cycling. To address whether keratinocyte-specific expression of the VDR was sufficient to maintain hair follicle homeostasis, transgenic mice expressing the VDR under the regulation of a keratinocyte-specific promoter were generated [35, 36]. These mice were crossbred to the VDR-null mice to generate mice that expressed the VDR only in keratinocytes. These mice did not develop alopecia and had a normal response to anagen-initiating stimuli, providing definitive proof that expression of

348 the VDR in keratinocytes was both necessary and sufficient for the maintenance of postmorphogenic hair cycling. Future investigations will be required to determine which regions of the VDR are required for skin homeostasis, whether the actions are dependent on an unique subset of nuclear receptor coactivators or an intact ligand binding domain, and, ultimately, what molecular pathway is disrupted by VDR ablation.

VII. CONCLUSIONS Studies in mice lacking functional VDRs have clearly demonstrated that the major roles of this nuclear receptor in vivo are the promotion of intestinal calcium absorption and the maintenance of skin homeostasis. Interestingly, effects of the VDR on two traditional target organs, the parathyroid and the skeleton, seem to be redundant, in that no abnormalities are observed in these organs in VDR-null mice with normal mineral ions. The effects of the VDR on intestinal calcium absorption seem to be classical ligand-dependent effects, and significant progress has been made in the identification of genes involved in VDR-mediated intestinal calcium absorption. In contrast, data point to potentially hormone-independent effects of the VDR on the hair cycle, and the molecular pathways that mediate the actions of this nuclear receptor in the skin are yet to be identified.

References 1. Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997 Targeted ablation of the vitamin D receptor: an animal model of vitamin D-dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94:9831–9835. 2. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Alioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Genet 16:391–396. 3. VanCromphaut S, Dewerchin M, Hoenderop J, Stockmans I, VanHerck E, Kato S, Bindels R, Collen D, Carmeliet P, Bouillon R, Carmeliet G 2001 Duodenal calcium absorption in vitamin D receptor-knockout mice: Functional and molecular aspects. Proc Natl Acad Sci USA 98:13324–13329. 4. Erben R, Soegiarto D, Weber K, Zeitz U, Lieberherr M, Gniadecki R, Moller G, Adamski J, Balling R 2002 Deletion of deoxyribonucleic acid binding domain of the vitamin D receptor abrogates genomic and nongenomic functions of vitamin D. Mol Endocrinol 16:1524–1537. 5. Dostal LA, Toverud SU 1984 Effect of vitamin D3 on duodenal calcium absorption in vivo during early development. Am J Physiol 246:G528–G534.

MARIE B. DEMAY

6. Li YC, Pirro AE, Demay M 1998 Analysis of vitamin Ddependent calcium-binding protein messenger ribonucleic acid expression in mice lacking the vitamin D receptor. Endocrinology 139:847–851. 7. Li YC, Amling M, Pirro AE, Priemel M, Meuse J, Baron R, Delling G, Demay MB 1998 Normalization of mineral ion homeostasis by dietary means prevents hyperparathyroidism, rickets, and osteomalacia, but not alopecia in vitamin D receptor–ablated mice. Endocrinology 139:4391–4396. 8. Donohue MM, Demay MB 2002 Rickets in VDR null mice is secondary to decreased apoptosis of hypertrophic chondrocytes. Endocrinology 143:3691–3694. 9. Amling M, Priemel M HT, Chapin K, Rueger JM, Baron R, Demay MB 1999 Rescue of the skeletal phenotype of vitamin D receptor ablated mice in the setting of normal mineral ion homeostasis: Formal histomorphometric and biomechanical analyses. Endocrinology 140:4982–4987. 10. Takeda S, Yoshizawa T, Nagai Y, Yamato H, Fukumoto S, Sekine K, Kato S, Matsumoto T, Fujita T 1999 Stimulation of osteoclast formation by 1,25-dihydroxyvitamin D requires its binding to vitamin D receptor (VDR) in osteoblastic cells: Studies using VDR knockout mice. Endocrinology 140:1005–1008. 11. Cao X, Ross FP, Zhang L, MacDonald PN, Chappel J, Teitelbaum SL 1993 Cloning of the promoter for the avian integrin β3 subunit gene and its regulation by 1,25-dihydroxyvitamin D3. J Biol Chem 268:27371–27380. 12. Medhora MM, Teitelbaum S, Chappel J, Alvarez J, Mimura H, Ross FP, Hruska K 1993 1α,25-Dihydroxyvitamin D3 up-regulates expression of the osteoclast integrin αvβ3. J Biol Chem 268:1456–1461. 13. Yasuda H, Shima N, Nakagawa N, Yamaguchi K, Kinosaki M, Mochizuki S, Tomoyasu A, Yano K, Goto M, Murakami A, Tsuda E, Morinaga T, Higashio K, Udagawa N, Takahashi N, Suda T 1998 Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci USA 95:3597–3602. 14. Chang W, Tu C, Pratt S, Chen T, Shoback D 2002 Extracellular Ca2+-Sensing receptors modulate matrix production and mineralization in chondrogenic RCJ3.15.18 cells. Endocrinology 143:1467–1474. 15. Adams C, Mansfield K, Perlot R, Shapiro I 2001 Matrix regulation of skeletal cell apoptosis: Role of calcium and phosphate ions. J Biol Chem 276:20316–20322. 16. Mansfield K, Rajpurohit R, Shapiro I 1999 Extracellular phosphate ions cause apoptosis of terminally differentiated epiphyseal chondrocytes. J Cell Physiol 179:276–286. 17. Mansfield K, Teixeira C, Adams C, Shapiro I 2001 Phosphate ions mediate chondrocyte apoptosis through a plasma membrane transporter mechanism. Bone 28:1–8. 18. Li YC, Kong J, Wei M, Chen ZF, Liu SQ, Cao LP 2002 1,25Dihydroxyvitamin D3 is a negative endocrine regulator of the renin-angiotensin system. J Clin Invest 110:229–238. 19. Kinuta K, Tanaka H, Moriwake T, Aya K, Kato S, Seino Y 2000 Vitamin D is an important factor in estrogen biosynthesis of both female and male gonads. Endocrinology 141:1317–1324. 20. Johnson LE, DeLuca HF 2001 Vitamin D receptor null mutant mice fed high levels of calcium are fertile. J Nutr 131: 1787–1791. 21. Johnson LE, DeLuca HF 2002 Reproductive defects are corrected in vitamin D-deficient female rats fed a high calcium, phosphorus and lactose diet. J Nutr 132:2270–2273.

CHAPTER 20 Mouse Models of Vitamin D Receptor Ablation

22. Zinser G, Packman K, Welsh J 2002 Vitamin D3 receptor ablation alters mammary gland morphogenesis. Development 129:3067–3076. 23. Mathieu C, VanEtten E, Gysemans C, Decallonne G, Kato S, Laureys J, Depovere J, Valckx D, Verstuyf A, Bouillon R 2001 In vitro and in vivo analysis of the immune system of vitamin D receptor knockout mice. J Bone Miner Res 16:2057–2065. 24. Xing N, L-Maldonado ML, Bachman LA, McKean DJ, Kumar R, Griffin MD 2002 Distinctive dendritic cell modulation by vitamin D3 and glucocorticoid pathways. Biochem Biophys Res Commun 297:645–652. 25. O’Kelly J, Hisatake J, Hisatake Y, Bishop J, Norman A, Koeffler HP 2002 Normal myelopoiesis but abnormal T lymphocyte responses in vitamin D receptor knockout mice. J Clin Invest 109:1091–1099. 26. Mann SJ 1971 Hair loss and cyst formation in hairless and rhino mutant mice. Anat Rec 170:485–499. 27. Sakai Y, Demay MB 2000 Evaluation of keratinocyte proliferation and differentiation in vitamin D receptor knockout mice. Endocrinology 141:2043–2049. 28. Xie Z, Komuves L, Yu QC, Elalieh H, Ng DC, Leary C, Chang S, Crumrine D, Yoshizawa T, Kato S, Bikle DD 2002 Lack of the vitamin D receptor is associated with reduced epidermal differentiation and hair follicle growth. J Invest Dermatol 118:11–16. 29. Takeyama K, Kitanaka S, Sato T, Kobori M, Yanagisawa J, Kato S 1997 25-Hydroxyvitamin D3 1α-hydroxylase and vitamin D synthesis. Science 277:1827–1830.

349 30. Kerry DM DP, Hahn CN, Morris HA, Omdahl JL, May BK 1996 Transcriptional synergism between vitamin D-responsive elements in the rat 25-hydroxyvitamin D 24-hydroxylase (CYP24) promoter. J Biol Chem 22:29715–29721. 31. Ohyama Y, Ozono K, Uchida M, Shinki T, Kato S, Suda T, Yamamoto O, Noshiro M, Kato Y 1994 Identification of a vitamin D–responsive element in the 5′-flanking region of the rat 25-hydroxyvitamin D3 24-hydroxylase gene. J Biol Chem 269:10545–10550. 32. Bikle DD, Pillai S 1993 Vitamin D, calcium, and epidermal differentiation. Endocr Rev 14:3–19. 33. Sakai Y, Kishimoto J, Demay MB 2001 Metabolic and cellular analysis of alopecia in vitamin D receptor knockout mice. J Clin Invest 107:961–966. 34. Kishimoto J, Ehama R, Wu L, Jiang S, Jiang N, Burgeson RE 1999 Selective activation of the versican promoter by epithelial-mesenchymal interactions during hair follicle development. Proc Natl Acad Sci USA 96:7336–7341. 35. Chen C, Sakai Y, Demay MB 2001 Targeting expression of the human vitamin D receptor to the keratinocytes of vitamin D receptor null mice prevents alopecia. Endocrinology 142: 5386–5389. 36. Kong J, Li XJ, Gavin D, Jiang Y, Li YC 2002 Targeted expression of human vitamin D receptor in the skin promotes the initiation of the postnatal hair follicle cycle and rescues the alopecia in vitamin D receptor null mice. J Invest Dermatol 118:631–638.

CHAPTER 21

Intracellular Vitamin D Response Element Binding Proteins JOHN S. ADAMS

Burns and Allen Research Institute and the Division of Endocrinology, Diabetes and Metabolism, Cedars-Sinai Medical Center, The David Geffen School of Medicine at UCLA, Los Angeles, California

I. Introduction II. New World Primates III. The Biochemical Nature of Vitamin D Resistance in New World Primates IV. New World Primate–like Vitamin D Resistance in Man V. Characterization of the Human Response Element Binding Protein (REBiP)

I. INTRODUCTION Since the publication of the first edition of Vitamin D, much has been learned regarding the cellular machinery for intracellular vitamin D trafficking (see Chapters 10 and 22), genomic action (see Chapters 11–20), and metabolism (see Chapters 4–7). This chapter will chronicle the discovery of a family of heterogeneous nuclear ribonucleoproteins (hnRNPs) that constitute one such previously unrecognized mode of control over the transactivation of vitamin D–regulated genes. As so often happens in science and medicine, this discovery evolved from the molecular analysis of a successful “experiment of nature.” The story begins with this author’s investigation of a man-made disturbance in vitamin D homeostasis in several species of New World primates resident at the Los Angeles Zoo in the mid1980s and concludes with the discovery of a previously unrecognized form of human vitamin D–resistant rickets.

II. NEW WORLD PRIMATES A. Early Primate Evolution In the Eocene period the great southern hemispheric landmass, Pangea, ruptured. This tectonic event resulted in the American landmass and Madagascar moving away from Africa. This continental separation occurred early in the process of primate evolution, trapping primordial primates in South America, Africa, and Madagascar, respectively. As a consequence, the three major primate infraorders, platyrrhines or New World primates, catarrhines or Old World primates, and lemurs, evolved independently of one another [1] (Fig. 1). VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

VI. Heterogeneous Nuclear Ribonucleoproteins (hnRNPs) VII. Compensation for the Dominant-Negative Acting, Response Element Binding Proteins VIII. Conclusion Reference

Unlike Old World primates, including our own species, which have populated virtually every landmass on our planet, New World primates have remained confined to Central and South America for the past 50 million years. Compared to Old World primates, especially some of the terrestrial species such as gorilla, New World primates are smaller in stature, a trait well-suited to their lifestyle as plant-eating arboreal sunbathers residing in the canopy of the periequatorial rain forests of the Americas.

B. Simian Bone Disease The appearance of generalized metabolic bone disease in captive primates has been recognized for the past 150 years [2]. The disease, which has not been well studied from a histopathological standpoint, carries the clinical and radiological stigmata of rickets and osteomalacia [3]. Compared to Old World primates reared in captivity, New World primates or platyrrhines are particularly susceptible to the disease. The disorder affects primarily young, growing animals and results in muscle weakness, skeletal fragility, and in many instances death of the affected individual. Rachitic bone disease of this sort has long presented a problem to veterinarians caring for captive platyrrhines, particularly in North American and European zoos [4], because death of preadolescent and adolescent primates prior to sexual maturity severely limits on-site breeding programs. Because the disease was reported to be ameliorated either by the oral administration of vitamin D3 in large doses or by ultraviolet B irradiation of affected Copyright © 2005, Elsevier, Inc. All rights reserved.

352

JOHN S. ADAMS

Old World

New World

Madagascar

Hominoidea Platyrrhini

Superorder

Catarrhini

ANTHRIPOIDEA PRIMATES

Lemuridae

Infraorder

PROSIMII

Suborder Order

FIGURE 1 New World primate evolution. Shown in geographic terms is the independent evolution of the three primate infraorders, Platyrrhini, Catarrhini, and Lemuridae, in South America (the New World), Africa (the Old World), and Madagascar, respectively. The superorder Hominoidea represents the human precursors.

primates, it was presumed to be caused by vitamin D deficiency [4]. The frequent occurrence of rickets and osteomalacia in New World primates was also ascribed to the relative inability of platyrrhines to effectively employ vitamin D2 in their diet [5]; a similar observation had been made for chickens [6]. Using assay technology that does not discriminate between 25-hydroxylated vitamin D2 and vitamin D3 metabolites, investigators [7] determined that 25-hydroxyvitamin D (25-OHD) levels were two- to threefold higher when platyrrhines were dosed with supplemental vitamin D3 than with vitamin D2. These data suggested that 25-hydroxylation of vitamin D substrate in New World primates was much more effective when vitamin D3 was employed as substrate. However, in the same study two species of Old World primates demonstrated similar discrimination against vitamin D2, in favor of vitamin D3. In summary, these results seemed to indicate that all subhuman primates, whether Old or New World, were relatively resistant to vitamin D2 in terms of its ability to engender an increase in serum levels of 25-OHD. Finally, Hay [8] suggested that New World primates may transport 25-OHD in the serum by means and via proteins that are dissimilar from those encountered in Old World primate species. This hypothesis was disproven by Bouillon et al. [9], who showed that the

vitamin D binding protein was the major carrier of 25-OHD in their serum of both New and Old World primates. The question of why platyrrhines were more susceptible to vitamin D deficiency than were catarrhines began to be answered with the detection of extraordinarily high circulating levels of the active vitamin D metabolite, 1,25-dihydroxyvitamin D, in New World primates [10–12]. These data confirmed that New World primates were resistant to the vitamin D hormone.

C. Outbreak of Rickets in the New World Primate Colonies of the Los Angeles Zoo The index case in the original studies was a preadolescent New World primate of the Emperor tamarin species (Fig. 2A). When investigated radiographically (Fig. 2B), this tamarin and those like him displayed classical rickets complete with growth retardation and metaphyseal cupping and fraying characteristic of rickets. In order to investigate this rachitic syndrome, blood and urine was collected from involved monkeys as well as from control, nonrachitic New and Old World primates. As shown in Fig. 3, that comparison yielded a biochemical phenotype that was most remarkable for

353

CHAPTER 21 Response Element Binding Proteins

A

B

FIGURE 2 A rachitic New World primate resident of the Los Angeles Zoo. (A) A preadolescent emperor tamarin with rickets. (B) The characteristic rachitic “cupping” and “fraying” of the tibial metaphysis (arrows) of this monkey.

an elevated serum 1,25-(OH)2D in rachitic New World primates [11]. In fact, with the exception of nocturnal primates in the genus Aotus, New World primates in all genera had vitamin D hormone levels ranging to 100fold higher than that observed in Old World primates including man [13–16]. In the initial analysis New World primates affected with rickets were those with the lowest 1,25-(OH)2D levels, while their healthy counterparts were those with the highest serum 1,25-(OH)2D. These data were interpreted to mean that most New World primate genera were naturally resistant to the vitamin D hormone, and that the resistant state could be compensated by maintenance of high 1,25-(OH)2D levels. If this was true, then an increase in the serum 1,25-(OH)2D concentration in affected primates should result in biochemical

compensation for the resistant state and resolution of their rachitic bone disease. When rachitic New World primates were exposed to 6 months of artificial sunlight, both serum 25-hydroxyvitamin D (25-OHD) and product 1,25-(OH)2D levels rose significantly (P ≤ 0.02), resulting in cure of rickets [15]. In summary, New World primates are periequitorial sunbathers for a reason. As depicted by the oversized arrows in a simplified scheme of vitamin D synthesis and metabolism in Fig. 4, New World primates require much cutaneous vitamin D synthesis in order to push their 25-OHD and 1,25-(OH)2D levels high enough to effectively interact with the VDR. The question remained: why were these primates resistant to all but the highest levels of the vitamin D hormone?

7-DHC

Rachitic Phenotype

Cholesterol

UVB Blood calcium.............

Slightly decreased

Urine calcium..............

Slightly decreased

Blood phosphate................

Normal

Urine phosphate.................

Normal

Serum creatinine................

Normal

Liver function......................

Normal

25-OHD..............................

Normal

1,25-(OH)2D.......................

Very high

FIGURE 3 Biochemical phenotype of rachitic New World primates. Demonstrated are the biochemical indices of bone health in New World primates suffering from rickets compared to developmental age- and sex-matched, nonrachitic Old World primates. The outstanding characteristic is a 1,25-dihydroxyvitamin D (1,25(OH)2D) level two to three orders of magnitude greater than that observed in Old World primates, including man.

Vitamin D3

25-OHase 25-OHD3

24,25-(OH)2D3

1-OHase 1,25-(OH)2D3

1,24,25-(OH)3D3

Calcitroic acid

VDR

FIGURE 4 Simplified scheme of vitamin D synthesis and metabolism in New World primates. The bold arrows describe the means by which high 1,25-dihydroxyvitamin D3 (1,25-(OH)2D3) levels are achieved and maintained in New World primates compared to their Old World and human primate counterparts.

354

JOHN S. ADAMS

III. THE BIOCHEMICAL NATURE OF VITAMIN D RESISTANCE IN NEW WORLD PRIMATES In order to answer the question of resistance in New World primates, cultured fibroblasts and immortalized cell lines from both resistant and hormone-responsive New and Old World primates were used to track, step by step, the path taken by the vitamin D hormone from the serum vitamin D binding protein (DBP) in the blood in route to the nucleus and transactivation of hormone-responsive genes [11,13–22]. It was shown that the movement of hormone from DBP, across the cell membrane and through the cytoplasm and nuclear membrane was indistinguishable from that observed in Old World primate cells. It was also determined that the ability of the New World primate VDR to bind 1,25-(OH)2D3 or 1,25-(OH)2D2 and induce receptor dimerization with the retinoid X receptor (RXR) was normal. In fact, when removed from the intranuclear environment and in distinction to previous reports [12], the VDR in New World primates was similar to the Old World primate VDR in all biochemical and functional respects [22]. That which was not the same was the reduced ability of VDR-RXR complex to bind to its cognate cis element and transactivate the expression of genes. In order to elucidate nuclear receptor events in New World primate cells, the nuclei of New World primate cells were isolated and extracted. In addition to the VDR-RXR, it was determined that these extracts contained a second protein bound to VDREs. This protein

A

was coined the vitamin D response element binding protein or VDRE-BP [21]. In electrophoretic mobility shift assay (EMSA) using the VDRE as probe, Old World primate cell extract contained only the VDR-RXR bound to the VDRE probe, while the New World primate extract contained two probe-reactive bands, one compatible with the VDR-RXR and a second, more pronounced VDRE-BP-VDRE band. This VDRE-BP-VDRE binding reaction was specific, as the VDRE-BP was competed from VDRE probe by the addition of excess unlabeled VDRE. These data suggested that VDRE-BP might function as a dominant-negative inhibitor of receptorresponse element binding by competing in trans with receptor, “knocking it off” the VDRE (Fig. 5), thus preventing VDR-RXR binding to the VDREs. When recombinant human VDR and RXR were permitted to interact in EMSA with increasing amounts of nuclear extract from vitamin D-resistant cells containing a VDRE-BP or from normal vitamin D–responsive cells, the addition of more control extract only amplified the VDR-RXR-retarded probe on the gel. By contrast, increasing amounts of the hormone-resistant extract competed away receptor–probe binding in favor VDRE-BP–probe binding. To date, two distinct VDRE-BPs have been identified, purified, cloned, and characterized in New World primates [21]. Both are members of the heterogeneous nuclear ribonucleoprotein A (hnRNPA) family of single strand mRNA binding proteins [23]. However, as just pointed out, VDRE-BPs can also bind specifically to double-strand DNA. In fact, it is by virtue of their ability to bind DNA that they can be distinguished from

B 1,25-D

RXR

VDR

VDRE

Wildtype

VDRE-BP

VDRE

Dominant-negative

FIGURE 5 The dominant-negative action of the New World primate vitamin D response element binding protein (VDRE-BP). (A) The “wild-type” events by which the retinoid X receptor (RXR) and hormone-liganded vitamin D receptor (VDR) dimer pair interact in trans with the vitamin D response element (VDRE) to regulate transcription of 1,25-dihydroxyvitamin D (1,25-D)-responsive genes. (B) The proposed “dominant-negative” event leading to competition for binding to the VDRE between the VDRE-BP and RXR-VDR; the net result is a blockade of transcriptional regulation.

355

CHAPTER 21 Response Element Binding Proteins

A. Index Case

Luciferase activity (arbitrary units)

500

*p < 0.001 250

0 OWP

+VDRE-BP

NWP

FIGURE 6 Squelching of RXR-VDR-directed, VDRE-reporterdriven transactivation. Shown is the significant decrease in VDREdirected luciferase reporter activity, assayed in relpicate (n = 4), in a subclone of Old World primate (OWP) cells after stable transfection with the New World primate VDRE-BP-2 (solid bar) compared to reporter activity in vector-alone transfected OWP cells (open bar) and New World primate (NWP) cells (shaded bar). Data are shown as luciferase units relative to an internal β-galactosidase standard.

traditional co-repressor proteins [24]. When overexpressed, they can effectively squelch VDR-directed transactivation. Depicted in Fig. 6 is VDRE-directed reporter activity in a subclone of wild-type Old World primate cells stably overexpressing the New World primate VDRE-BP-2 as well as in naturally hormone-resistant New World primate cells. Stable overexpression of VDRE-BP-2 squelched luciferase activity substantially compared to untransfected, wild-type cells to levels observed in hormone-resistant New World primate cells that naturally overexpress the protein; VDRE-BP-1 overexpression reduced, but not significantly, transactivation. This is strong confirmatory evidence that when overexpressed in vivo, VDRE-BP-2 is the cause of vitamin D resistance in these monkeys.

IV. NEW WORLD PRIMATE–LIKE VITAMIN D RESISTANCE IN MAN The Adams laboratory has been studying a vitamin D– resistant state in New World primates for more than 15 years with the expectation that it would provide valuable insight into how steroid hormones regulate gene expression in human primates. Prior to this time it was unknown whether there existed a human homolog to the vitamin D–resistant state in New World primates.

In 1993 a patient with the classical signs of type II hereditary vitamin D–resistant rickets (HVDRRII), including alopecia, was reported [25]. The biochemical phenotype of this patient included hypocalcemia (2.03 mmol/L corrected for albumin [normal range, 2.25–2.55 mmol/L]); raised serum alkaline phosphatase (1101 U/L [normal range, <300 U/L]); and raised circulating levels of 1,25-(OH)2D [466–650 pmol/L (normal range, 48–156 pmol/L]). Despite this, the patient had normal VDR expression; sequence analyses indicated that the coding regions, as well as the 5′ and 3′ untranslated regions of the VDR gene, were normal. Furthermore, when extracted from cells, VDR from the patient displayed normal binding capacity and affinity for the active form of vitamin D, 1,25-(OH)2D3. Transfection of VDR cDNA from the patient into receptor-deficient (human) CV-1 cells resulted in normal transactivation in response to 1,25-(OH)2D3 [25]. However, in the subject’s own cells the VDR was incapable of stable nuclear localization and transactivation after exposure to hormone 1,25-(OH)2D3. These data suggested that an extra-VDR factor was interfering with the nuclear translocation and transactivation of the VDR in cells from this patient. The patient was without evidence of any other form of steroid–thyroid–retinoid hormone resistance, and correction of rachitic bone disease was accomplished with high-dose 1,25-(OH)2D3 treatment (12 µg/day) and calcium supplements (1–3 g/day), although the alopecia persisted. It was recently discovered [26] that the underlying cause of insensitivity to 1,25-(OH)2D in this patient resulted from over-expression of a vitamin D response element binding protein (REBiP) with similarity to hnRNPA proteins that cause receptornormal hormone resistance in New World primates.

B. Suppression of RXR-VDR-Mediated Transactivation and VDRE Binding in HVDRR Cell Lines To investigate the presence of such an extra-VDR factor in cells from this HVDRR patient, promoterreporter assays were carried out using normal and HVDRR fibroblasts as transfection recipients [26]. Data confirmed the suppression of VDRE-mediated transactivation in HVDRR cells in the presence or absence of wild-type VDR. Furthermore, in contrast to control cells, HVDRR fibroblasts showed no induction of luciferase activity following treatment with 10 nM 1,25-(OH)2D3. Based on these data shown, as was the case in hormoneresistant New World primate cells, we hypothesized that cells from the HVDRR patient constitutively overexpress

356 a nuclear protein that competes with the VDR-RXR for binding to heterodimer cis recognition sequences (Fig. 5). To test this hypothesis subsequent assays were carried out using EBV-transformed HVDRR and control B-lymphocytes, which could be cultured in large volumes to facilitate maximal protein recovery from nuclear extracts. In contrast to control lymphocytes, cells from the HVDRR patient did not demonstrate any significant antiproliferative response to increasing doses of 1,25-(OH)2D3 following mitogen (PMA) activation. To investigate this further, EMSAs were carried out with recombinant human VDR and RXR and the 3-nucleotide-spaced direct repeat VDRE (VDRE-DR3; cis sequence AGGTCAcagAGGTCA) probe in the presence or absence of increasing amounts of nuclear extract from control or HVDRR cells. Data showed that addition of the HVDRR nuclear extract competitively displaced RXR-VDR binding to the VDRE in a dosedependent fashion. In contrast, nuclear extracts from control cells were without effect. These results were confirmed by densitometric analysis of EMSA band density performed in triplicate; significant (p < 0.05), stepwise diminishment of probe-VDR-RXR complex formation with increasing doses of HVDRR NE (4, 8, and 12 µg) was achieved compared to equivalent control extracts.

JOHN S. ADAMS

Nuclear extracts from HVDRR and control cells were subjected to SDS-PAGE and the separated proteins analyzed for their ability to bind radiolabeled VDRE probe and RXRE probe in Southwestern blots. A VDRE-DR3 and RXRE-reactive protein was observed only in the nuclear extracts from the HVDRR cells. In view of the RNA binding capacity of hnRNPs, further studies were carried out to assess REBiP binding to single-stranded nucleotide sequences [26]. When HVDRR nuclear extracts were incubated with a single-strand radiolabeled DNA probe consisting of the upper strand of the RXRE, the REBiP complex could only be competed away using an excess of unlabeled single- or doublestranded RXRE. Results presented above suggested that resistance to 1,25-(OH)2D3 in the HVDRR patient correlated with overexpression of an REBiP(s). To confirm a functional link between the cis binding of REBiP and VDR-mediated transactivation, further reporter studies were carried out using the VDR-positive HKC-8 human renal cell line cotransfected with expression constructs for human hnRNPA1 or hnRNPA2 and with the NWP dominant-negative-acting hnRNP-related VDRE-BP2, which show strong homology to one another [21]. Both of these REBiPs suppressed VDREdirected transcription to a similar degree (Fig. 7).

V. CHARACTERIZATION OF THE HUMAN RESPONSE ELEMENT BINDING PROTEIN (REBiP) A. REBiP Binding in Cis The cis-element specificity of the HVDRR cell extract-VDRE EMSA complex was assessed by competition analyses. Data confirmed that RXRE and VDRE competed away the hormone response element–binding complex; however, consensus sequences for other types of cis elements such as the ERE showed no displacement of the response element binding complex [26]. The peptide nature of the complex component(s) was confirmed using HVDRR nuclear extracts that had been digested by trypsin; unlike untreated HVDRR extracts, these preparations were unable to compete away the EMSA complex formed by recombinant VDR-RXR [26]. Furthermore, VDRE probe binding by HVDRR nuclear extracts was competed away by the addition of antihnRNPA1 and -A2 antibody, suggesting that, in a fashion similar to that in New World primate cells (Fig. 5), HVDRR cells expressed an hnRNP-related REBiP capable of interacting with the VDRE. Characterization of the REBiP by Western blot analysis showed increased expression of hnRNPA1 and -A2 immunoreactive proteins in HVDRR cells.

** p < 0.001 Relative VDRE-luciferase activity

**

Vector

Human hnRNPA1

**

Primate VDRE-BP2

FIGURE 7 Equivalent squelching of VDRE-directed luciferase activity by human and New World primate hnRNP-related response element binding proteins. Shown are vitamin D receptor (VDR)-positive human kidney cells (HKC-8) transfected with a VDRE-driven reporter plasmid alone (solid bar) or cotransfected with expression constructs for human hnRNPA1 (dark-shaded bar) or New World primate VDRE-BP2 (light-shaded bar). Cells were precultured in the presence of 10 nM 1,25-(OH)2D3 for 24 hr. Data are shown as luciferase units relative to an internal β-galactosidase standard. Values are the mean ± standard deviation of three separate assays.

357

CHAPTER 21 Response Element Binding Proteins

B. Lessons Learned from Overexpression of REBiP In summary, we have described a new form of human sterol/steroid hormone insensitivity that results from overexpression of a dominant-negative-acting response element binding protein or REBiP. In this case the phenotype of vitamin D resistance could not be attributed to mutant forms or abnormal amounts of either the VDR or its principal binding partners, RXRα and RXRβ, and there was no clinical evidence for malfunction of other steroid hormone receptors. The fact that REBiP is both a single- and double-strand nucleic acid binding protein and not a receptor binding protein indicates that vitamin D resistance in this patient was not due to abnormal expression of a VDR accessory protein. As with New World primates, the underlying cause of insensitivity to 1,25-(OH)2D3 appears to be overexpression of an hnRNPA, with both the human hnRNPA1 and the New World primate VDRE-BP2 equivalent able to squelch VDRE-mediated transactivation. The fact that the hypocalcemia and rachitic bone disease in this HVDRR patient were responsive to high-dose 1,25-(OH)2D3 treatment [25] indicates that vitamin D resistance is not absolute, but instead is likely to be determined by the relative abundance of the hnRNPA-related REBiP and competent (i.e., 1,25(OH)2D3-liganded) RXR-VDR heterodimer present in the target cell. For example, if the balance in cis element binding favored REBiP, because of either a relative abundance of this protein and/or a relative lack of the competitive RXR-VDR, then one would predict a 1,25-(OH)2D3-reversible vitamin D-resistant phenotype in vivo. In fact, this prediction was compatible with the patient’s response to treatment with high-dose 1,25-(OH)2D3; hypocalcemia and rachitic bone disease were corrected as long as the serum 1,25-(OH)2D3 level remained high [25]. These data also suggest that it is the VDRE-DR3 cis element that legislates the antirachitic action of the hormone. In contrast, failure of the patient’s alopecia to be improved with an increase in serum 1,25-(OH)2D3 levels suggests that (1) the VDRE-DR3 does not participate in control of genes rendering the hairless phenotype; (2) maternal levels of 1,25-(OH)2D3 are not high enough for fetal hair development; and/or (3) as suggested by previous data from HVDRR patients and VDR-ablated animals, the role of 1,25-(OH)2D3-VDR in hair follicle development involves only a prenatal mechanism. That the REBiP complex could not be competed out with an ERE or YY1-RE suggests that the hnRNPA family members have greater specificity for direct repeat half-sites. This is further supported by the elucidation of a double-strand DNA target sequence for hnRNPA1 that includes two cryptic direct repeat half-sites

(ggctggtcttgaactcctgA/GCTCAAA/GGTGAtcctcc; [27]) and no overlap with established RNA target sequences for hnRNPA1 [23]. Furthermore, studies from Chen et al. have shown that the estrogen resistance-causing estrogen response element binding protein in New World primates belongs to the distinct hnRNPC family (see Section VII) and possesses an entirely different target sequence on double-strand DNA. This provides an explanation for the highly specific phenotype observed with the HVDRR patient.

VI. HETEROGENEOUS NUCLEAR RIBONUCLEOPROTEINS (hnRNPs) A. hnRNPs: The Classical View The hnRNPs were initially recognized as proteins that interacted in trans with specific cis elements in the 3′ end of pre-mRNAs [28]. This interaction of the hnRNPs with the pre-mRNA begins cotranscriptionally and may persist as an mRNA–protein complex (mRNP) through the process of splicing, posttranscriptional modification, and nuclear export to the ribosome. The classic view [23] held that these proteins were critical in the “stabilization” of those mRNAs that were destined for translation; on the other hand, the absence of these stabilizing factors routed mRNAs to destruction in advance of translation. However, the data presented here for both the New World primate VDRE-BPs and the human REBiPs indicate that some members of the hnRNP family of proteins, at least when overexpressed, can also alter the process of transcription through a direct protein–DNA interaction. This kind of information and that from a variety of other sources [28] now clearly demonstrates that the hnRNPs are more than just single strand RNA binding proteins that bind to 3′ end of pre-mRNAs, “stabilizing” those transcripts for eventual translation.

B. hnRNPs as Multifunctional Proteins Whereas work presented so far points to a role of hnRNPs in transcription, the work of others is beginning to show that these proteins subserve a number of functions both inside and outside of the eukaryotic cell nucleus. There are at least four protein machines in the nucleus and another in the cytoplasm of the human cell dedicated to the process of making transcripts of genes and converting those transcripts to proteins: the nucleosome, transcription apparatus, spliceosome, and exosome in the nucleus and the ribosome outside of the nuclear compartment. There is now growing evidence that the hnRNPs are involved in the operation of each of these [29].

358

JOHN S. ADAMS

The nucleosome is the basic structural element in chromatin [30]. It consists of 146 bp of DNA wound tightly around a core of histones. The nucleosomes and their chromatin are differentially compacted into highly condensed heterochromatin and euchromatin, the latter of which is less densely packed and representative of the sections of the genome that are accessible to transcription activators. It is the insertion of single-strand and double-strand DNA binding proteins, including members of the hnRNP family [31,32], that initiates decompaction and provides accessibility to the regulatory elements of genes destined for transcriptional modulation. The transcription apparatus is classically viewed as the collection of proteins associated with RNA polymerase II (RNAPII). Included among these proteins are a number of transcription factors that interact with specific cis elements in the promoter regions of genes targeted for transcription. These transcription factors include the sterol/sterol receptors as well as members of the hnRNP family of single/double-strand DNA proteins [20,21,33] which can compete with the receptor proteins for the same cis element. DNA strand separation is a crucial element of the transcription process. Strand separation is initiated by RNAPII as it is tethered to the promoter by a bridge of co-regulator proteins [34]. Once initiation is fixed, the process of “elongation” ensues and the co-regulator tether is broken or dislodged from its DNA anchor site in the promoter. The RNAPII is then free to make its way down the coding strand of the DNA template (Fig. 8). This chain of events makes it possible for the human hnRNP-related REBiP which can occupy the VDRE in the promoter (at left in Fig. 8) and squelch hormonedirected transactivation [26], to also interact with the C terminal domain (CTD) (middle of Fig. 8) of RNAPII,

1,25-D

RNAP

Pre-mRNA

CTD

Co-activator RXR

TATA

attracting the RNAPII to the transcription start site. By successfully competing in cis for the VDRE, the liganded VDR-RXR can dislodge the REBiP, permitting the coactivator tether to assemble and interact with RNAPII. Once initiation of transcription is fixed, it is possible for the REBiP to reoccupy the VDRE releasing the VDR-RXR-anchored tether and permitting elongation (of right in Fig. 8). On the other hand, if the REBiP remains bound to the CTD of RNAPII, then by virtue of its ability to bind single-strand RNA it has the potential to influence the processing, export, and translation of the transcript by binding to its 3′ UTR (see later discussion). The spliceosome is one of the largest of the nuclear machines. It functions to recognize the intron–exon junctions in eukaryotic genes, remove the introns and then variably re-“splice” the transcribed exons in a cellspecific manner. The spliceosome comprises 145 distinct proteins. At least 30 of these proteins, including 12 in the hnRNP family, have recognized functions outside of the realm of splicing [35]. The function of these many spliceosome-associated hnRNPs is unknown. It is almost certain that their ability to recognize both RNA and other proteins is key in demarcating the scission and resplice junctions in pre-mRNA as part of the so-called exon–exon junction complex or EJC. There is now convincing evidence that the processes of transcription, splicing, 3′ polyadenylation and 5′ capping are functionally coupled and temporally coincident [36] with the CTD of RNAPII proposed to serve as a platform for the ordered assembly of the different families of pre-mRNA processing machines ([29]; see Fig. 8). An especially relevant example is the recent study of Aboeuf and co-workers [37]. They showed that the liganded glucocorticoid receptor (GR) exerted coincident control over RNAPII and alternative splicing of

REBiP

VDR

RNAP CTD

REBiP 5′

REBiP RNAP 3′

CTD REBiP

FIGURE 8 Simplified schematic of the potential for serial interaction of the vitamin D-resistance-causing, human vitamin D–resistant rickets (HVDRR) response element binding protein (REBiP) with the VDRE in the promoter of a 1,25(OH)2D-VDR-RXR-regulated gene. Shown is the transition of REBiP interaction with the VDRE in the basal state (at left), the C-terminal domain (CTD) of the RNA polymerase II (RNAP; middle), and 3′ untranslated end of the pre-mRNA transcript of that gene (at right). Arrows indicate the direction of transcription.

CHAPTER 21 Response Element Binding Proteins

the transcript initiated under the influence of the ligand. It is possible that VDRE-reactive REBiP and VDR-RXR coordinately alter transcription and splicing in a similar fashion. The exosome constitutes the protein machinery responsible for the export of mRNA from the nucleus [38]. These events do not occur in isolation. Rather they are functionally linked to upstream events in RNA splicing and processing and downstream processes such as nonsense-mediated messenger decay [39]. The hnRNPs are proposed to be principal players in the exosome, acting as a connector between the 3′ end of the mRNA and constituent proteins of the nuclear pore [40]. Hence, it is possible, as described in Fig. 8, that the hnRNPA-related VDRE-BPs and REBiP, by virtue of their protein- and nucleic acid-binding capacity, could be passed from promoter to RNAPII- to 3′ end of mRNA to exosome and contribute to the delivery of that same gene’s mRNA to the cytoplasm. These hypotheses are yet to be confirmed but are under active investigation. The ribosome and the process of translation is the final stop for an mRNA. hnRNPs serve as ribosome recognition proteins [41]. They participate in parking transcripts with tRNAs anchored in the 30s subunit of the ribosome. Therefore, it is possible that the influence of the human REBiPs and the nonhuman primate homolog VDRE-BPs can extend all the way from the promoter of a gene to the translation of that gene’s transcript into a functional protein; as previously noted it is now clear that the process of making and translating RNA is not the result of a series of independent processes, but the outcome of functionally-linked simultaneous events. It is also clear, as demonstrated by the hnRNPs, that proteins thought once to be involved with specific single steps in the making, processing and translating RNA are not limited to single steps in that chain of events. Finally, as exemplified by the 154 proteins making up the spliceosome machinery of which only half have a clearly defined splicing function [35], it is also clear that there remains much to be learned about how these various complex machines described above function in concert with one another.

VII. COMPENSATION FOR THE DOMINANT-NEGATIVE ACTING, RESPONSE ELEMENT BINDING PROTEINS A. Intracellular Vitamin D Binding Proteins During the process of discovery of the VDRE-BPs in New World primate cells, it was also observed that

359 these cells were extraordinarily efficient at accumulating 25-hydroxylated vitamin D metabolites in the cytoplasmic space. Accumulation here was the result of expression of a second set of resistance-associated proteins. These intracellular vitamin D bindings proteins [42,43], or IDBPs as they have come to be called, exhibit both high capacity and high affinity for 25-hydroxylated vitamin D metabolites. In fact, among all of the vitamin D metabolites that have been tested, the IDBP(s) purified from vitamin D–resistant New World primate cells bind 25-OHD3 and 25-OHD2 best [14,43]; in a competitive displacement assay using radioinert 25-OHD3 as competitor and [3H]25-OHD3 as labeled ligand, the concentration of metabolite required to achieve half-maximal displacement of labeled hormone (EC50) was <1 nM. Although normally present in Old World primate (including human) cells, these proteins can be overexpressed some 50-fold in New World primate cells. These IDBPs are highly homologous to proteins in the heat shock protein-70 family [44]. The first four members of this family, IDBP-1, -2, -3 and -4, have been cloned and characterized [21,45]. They bear a high degree of sequence identity with constitutively expressed human heat shock protein-70, endoplasmic reticulum–targeted glucose regulated protein-78 (grp78), mitochondrial-targeted grp 75, and heat-shock inducible heat shock protein-70, respectively. The general domain structure of the IDBPs [44] is shown in Fig. 9A. They all contain an ATP-binding-ATPase domain ahead of a protein-protein interaction domain. Some like IDBP-2 and IDBP-3 also harbor an N-terminal organelle-targeting domain. Preliminary studies indicate that the vitamin D ligand binding domain resides in the middle of the molecule near the C-terminal aspect of the ATPase domain [46]. What are these IDBPs doing inside the hormoneresistant New World primate cell? Two countervailing hypotheses were considered to explain the function of these proteins (Fig. 9B). One hypothesis held that these IDBPs were “sink” molecules that worked in cooperation with the VDRE-BP in the nucleus to exert vitamin D resistance by disallowing access of the hormone to the VDR and the nucleus of the cell. The opposing hypothesis held that these were “swim” molecules that actually promoted the delivery of ligand to the vitamin D receptor, improving the ability of the VDR to dimerize and bind to DNA, antagonizing the actions of the VDRE-BP that was overexpressed in New World primate cells. In order to determine which of these hypotheses was correct, the most abundant of these IDBPs, IDBP-1, was stably overexpressed in vitamin Dresponsive (i.e., wild-type) Old World primate cells and shown to be pro-transactivating [46]; the endogenous

360

JOHN S. ADAMS

A

B

ATPase

Hormone

‘‘swim’’

Vitamin D binding Organelle targeting

Protein:protein interaction

“sink’’ RXR Target cell

VDR

VDRE

FIGURE 9 General domain structure of the hsp70-related intracellular vitamin D binding proteins (IDBPs) and their proposed function in vitamin D-resistant New World primate cells. (A) Domain structure of hsp70-like proteins. All contain an ATP-binding-ATPase domain ahead of two protein–protein interaction domains. Some also harbor an N-terminal organelle-targeting domain. Molecular mapping studies place the vitamin D sterol binding domain in the ATPase domain of the protein. (B) The countervailing “sink” and “swim” functional hypotheses for these IDBPs.

transcriptional activity of three different 1,25-(OH)2Dresponsive genes, the vitamin D-24-hydroxylase, osteopontin, and osteocalcin genes, in Old World primate wild-type cells was significantly enhanced. It was concluded from these studies, at least for the function of transactivation, that IDBP-1 was a “swim” molecule for the vitamin D hormone, promoting delivery of ligand to the VDR; a clear role for IDBP-2 and IDBP-3 in this regard has not yet been determined. Considering the facts that New World primates are required to maintain very high serum levels of 1,25(OH)2D in order to avert rickets (Fig. 2), it was hypothesized that the IDBPs, which are known to bind 25-OHD even better than 1,25-(OH)2D, will also promote the synthesis of the active vitamin D metabolite via promotion of the 25-OHD-1-hydroxylase. When human kidney cells expressing the 25-OHD-1-hydroxylase gene were stably transfected with IDBP-1 and incubated with substrate 25-OHD3, 1,25-(OH)2D3 production went up four- to eightfold compared to untransfected, wildtype cells [47]. This increase in specific 25-OHD1-hydroxylase activity occurred independent of a change in expression of the 25-OHD-1-hydroxylase gene [46]. In fact, current data (see next section) now strongly indicate that this increase in hormone production is the result of the ability of IDBPs to promote the delivery of substrate 25-OHD to the inner mitochondrial membrane and the 25-OHD-1-hydroxylase stabled there.

B. A New Model for Intracellular Vitamin D Trafficking Dogma has held that sterol/steroid hormones such as vitamin D, by nature of their lipid solubility, diffuse through the plasma membrane of the target cell and

“Ping-Pong” around the cell interior until they encounter another specific binding protein such as 25-OHD-1-hydroxylase or the VDR with which to bind. Recent results, developed from a compendium of confocal imaging studies with fluorescently labeled IDBPs and vitamin D metabolites as well as with protein–protein interaction experiments [45], indicate that the hormone does not haphazardly “Ping-Pong” around the cell interior. Rather, the hormone enters the cell and is distributed to specific intracellular destinations by a series of protein–protein interactions which involve the hsp70 family of IDBPs. For example, we now know from the work of Willnow et al. (see Chapter 10; [48,49]) that vitamin Ds can enter target cells via internalized vesicles. The vitamin D stays bound to DBP, which is in turn bound by megalin and cubulin, members of the LDL superfamily of proteins. Once inside the cell there is interaction between the C-terminal domain of megalin, which protrudes into the cytoplasm, and the N-terminal domain of IDBP-1 [45]. In fact, if one then incubates IDBP-1 overexpressing cells with a fluorescently labeled 25-hydroxylated vitamin D metabolite, one will observe a significant increase in the uptake of the labeled hormone [45].

VIII. CONCLUSION As our knowledge base increases, the mechanisms underlying the control of vitamin D action and metabolism within target cells continue to expand, sometimes in unexpected directions. The discovery of the hnRNP-related vitamin D response element binding proteins in subhuman (VDRE-BPs) in human (REBiPs) primate is one such unexpected direction. The potential that these multifunctional nucleic acid binding proteins can alter hormone-directed transcription of genes in

361

CHAPTER 21 Response Element Binding Proteins

addition to controlling the posttranscriptional fate of mRNAs transcribed from those genes indicates that the traditional views of the various nuclear machines (i.e., transcriptosome, sliceosome, etc.) working independently of one another is no longer tenable. Adding to the untenable nature of these traditional views is the ability of these multifunctional hnRNPs to recruit the expression of other classes of proteins, such as the hsp70 family of intracellular vitamin D binding proteins, not previously known to be important in the chaperoning of small molecules, to further modify to the cellular responsiveness to vitamin Ds. Although originally discovered as elements in the cellular vitamin D regulatory system, the hnRNP and hsp family of proteins are also active in the control of estrogen action. Chen and colleagues [20,33] have discovered proteins in the hnRNP C family that act in a dominant-negative mode to squelch estrogen–estrogen response element (ERE)-directed transactivation. In a manner similar to that observed for the VDRE-BPs and REBiPs with the VDRE, these estrogen response element binding proteins (ERE-BPs) act in trans to prevent estrogen receptor (ER) dimer binding to the ERE. Most recently, Chen et al. [26] have also discovered an intracellular estrogen binding protein (IEBP). This protein is in the hsp27 family. Although the co-overexpression of IEBP is observed in the presence of ERE-BP overexpression, the role the IEBP plays, if any, in ERERE-directed transactivation is still under analysis.

11.

12. 13.

14. 15. 16. 17.

18.

19.

20.

References 1. Pilbeam D 1984 The descent of hominoids and hominids. Sci Am 250:84–96. 2. Bland Sutton JB 1884 Observation on rickets etc. in wild animals. J Anat 18:363–397. 3. Krook L, Barrett RB 1962 Simian bone disease—A secondary hyperparathyroidism. Cornell Vet 52:459–492. 4. Hershkovitz P 1977 Living New World Monkeys (Platyrrhini): With an Introduction to Primates. University of Chicago Press, Chicago. 5. Hunt RD, Garcia FG, Hegsted DM 1967 A comparison of vitamin D2 and D3 in New World primates. I. Production and regression of osteodystrophia fibrosa. Lab Anim Care 17:222–234. 6. Steenbock H, Kletzein SWF, Halpin JG 1932 The reaction of the chicken to irradiated ergosterol and irradiated yeast as contrasted with the natural vitamin D of fish liver oils. J Biol Chem 97:249–266. 7. Marx SJ, Jones G, Weinstein RS, Chrousos GP, Renquist DM 1989 Differences in mineral metabolism among nonhuman primates receiving diets with only vitamin D3 or only vitamin D2. J Clin Endocrinol Metab 69:1282–1290. 8. Hay AW 1975 The transport of 25-hydroxycholecalciferol in a New World monkey. Biochem J 151:193–196. 9. Bouillon R, van Baelen H, de Moor P 1976. The transport of vitamin D in the serum of primates. Biochem J 159:463–466. 10. Shinki T, Shiina Y, Takahashi N, Tanioka Y, Koizumi H, Suda T 1983 Extremely high circulating levels of 1 alpha,

21.

22.

23. 24. 25.

26.

27.

25-dihydroxyvitamin D3 in the marmoset, a New World monkey. Biochem Biophys Res Commun 114:452–457. Adams JS, Gacad MA, Baker AJ, Gonzalez B, Rude RK 1985 Serum concentrations of 1,25-dihydroxyvitamin D3 in platyrrhini and catarrhini: A phylogenetic appraisal. Am J Primatol 19: 219–224. Liberman UA, de Grange D, Marx SJ 1985 Low affinity of the receptor for 1-alpha,25-dihydroxyvitamin D3 in the marmoset, a New World monkey. FEBS Lett 182:385–388. Adams JS, Gacad MA, Baker AJ, Kheun G, Rude RK 1985 Diminished internalization and action of 1,25-dihydroxyvitamin D3 in dermal fibroblasts cultured from New World primates. Endocrinology 116:2523–2527. Gacad MA, Adams JS 1991 Endogenous blockade of 1,25dihydroxyvitamin D–receptor binding in New World primate cells. J Clin Invest 87:996–1001. Gacad MA, Deseran MW, Adams JS 1992 Influence of ultraviolet B radiation on vitamin D3 metabolism in New World primates. Am J Primatol 28:263–270. Gacad MA, Adams JS 1992 Specificity of steroid binding in New World primate B95–8 cells with a vitamin D-resistant phenotype. Endocrinology 131:2581–2587. Adams JS, Gacad MA 1988 Phenotypic diversity of the cellular 1,25-dihydroxyvitamin D3–receptor interaction among different genera of New World primates. J Clin Endocrinol Metab 66:224–229. Gacad MA, Adams JS 1993 Identification of a competitive binding component in vitamin D-resistant New World primate cells with a low affinity but high capacity for 1,25-dihydroxyvitamin D3. J Bone Miner Res 8:27–35. Arbelle JE, Chen H, Gacad MA, Allegretto EA, Pike JW, Adams JS 1996 Inhibition of vitamin D receptor–retinoid X receptor-vitamin D response element complex formation by nuclear extracts of vitamin D–resistant New World primate cells. Endocrinology 137:786–789. Chen H, Arbelle JE, Gacad MA, Allegretto EA, Adams JS 1997 Vitamin D and gonadal steroid-resistant New World primate cells express an intracellular protein which competes with the estrogen receptor for binding to the estrogen response element. J Clin Invest 99:669–675. Chen H, Hu B, Allegretto EA, Adams JS 2000 The vitamin D response element–binding protein. A novel dominant-negative regulator of vitamin D–directed transactivation. J Biol Chem 275:35557–35564. Chun RF, Chen H, Boldrick L, Sweet C, Adams JS 2001 Cloning, sequencing, and functional characterization of the vitamin D receptor in vitamin D–resistant New World primates. Am J Primatol 54:107–118. Dreyfuss G, Matunis MJ, Pinol-Roma S, Burd CG 1993 hnRNP proteins and the biogenesis of mRNA. Annu Rev Biochem 62:289–321. Horwitz KB, Jackson TA, Bain DL, Richer JK, Takimoto GS, Tung L 1996 Nuclear receptor coactivators and corepressors. Mol Endocrinol 10:1167–1177. Hewison M, Rut AR, Kristjansson K, Walker RE, Dillon MJ, Hughes MR, O’Riordan JL 1993 Tissue resistance to 1,25-dihydroxyvitamin D without a mutation of the vitamin D receptor gene. Clin Endocrinol (Oxf) 39: 663–670. Chen H, Hewison M, Hu B, Adams JS 2003 Heterogeneous nuclear ribonucleoprotein (hnRNP)-binding to hormone response elements: a novel cause of vitamin D resistance. Proc Natl Acad Sci USA 100:6109–6114. Donev RM, Doneva TA, Bowen WR, Sheer D 2002 HnRNPA1 binds directly to double-stranded DNA in vitro within a 36 bp sequence. Mol Cell Biochem 233:181–185.

362 28. Dreyfuss G, Kim VN, Kataoka N 2002 Messenger-RNAbinding proteins and the messages they carry. Nat Rev Mol Cell Biol 3:195–205. 29. Orphanides G, Reinberg D 2002 A unified theory of gene expression. Cell 108:439–451. 30. Horn PJ, Peterson CL 2002 Molecular biology. Chromatin higher order folding—wrapping up transcription. Science 297: 1824–1827. 31. Michelotti GA, Michelotti EF, Pullner A, Duncan RC, Eick D, Levens D 1996 Multiple single-stranded cis elements are associated with activated chromatin of the human c-myc gene in vivo. Mol Cell Biol 16:2656–2669. 32. Martens JH, Verlaan M, Kalkhoven E, Dorsman JC, Zantema A 2002 Scaffold/matrix attachment region elements interact with a p300-scaffold attachment factor A complex and are bound by acetylated nucleosomes. Mol Cell Biol 22: 2598–2606. 33. Chen H, Hu B, Gacad MA, Adams JS 1998 Cloning and expression of a novel dominant-negative-acting estrogen response element-binding protein in the heterogeneous nuclear ribonucleoprotein family. J Biol Chem 273:31352–31357. 34. Yin YW, Steitz TA 2002 Structural basis for the transition from initiation to elongation transcription in T7 RNA polymerase. Science 298:1387–1395. 35. Zhou Z, Licklider LJ, Gygi SP, Reed R 2002 Comprehensive proteomic analysis of the human spliceosome. Nature 419:182–185. 36. Proudfoot NJ, Furger A, Dye MJ 2002 Integrating mRNA processing with transcription. Cell 108:501–512. 37. Auboeuf D, Honig A, Berget SM, O’Malley BW 2002 Coordinate regulation of transcription and splicing by steroid receptor coregulators. Science 298:416–419. 38. Hilleren P, McCarthy T, Rosbash M, Parker R, Jensen TH 2001 Quality control of mRNA 3′-end processing is linked to the nuclear exosome. Nature 413:538–542.

JOHN S. ADAMS

39. Reed R, Hurt E 2002 A conserved mRNA export machinery coupled to pre-mRNA splicing. Cell 108:523–531. 40. Nakielny S, Dreyfuss G 1999 Transport of proteins and RNAs in and out of the nucleus. Cell 99:677–690. 41. Zhu J, Mayeda A, Krainer AR 2001 Exon identity established through differential antagonism between exonic splicing silencer-bound hnRNP A1 and enhancer-bound SR proteins. Mol Cell 8:1351–1361. 42. Gacad MA, Chen H, Arbelle JE, LeBon T, Adams JS 1997 Functional characterization and purification of an intracellular vitamin D–binding protein in vitamin D–resistant new world primate cells. Amino acid sequence homology with proteins in the hsp-70 family. J Biol Chem 272:8433–8440. 43. Gacad MA, Adams JS 1998 Proteins in the heat shock-70 family specifically bind 25-hydroxyvitamin D3 and 17betaestradiol. J Clin Endocrinol Metab 83:1264–1267. 44. Hartl FU 1996 Molecular chaperones in cellular protein folding. Nature 381:571–579. 45. Adams JS, Chen H, Chun RF, Nguyen L, Wu S, Ren SY, Barsony J, Gacad MA 2003 Novel regulators of vitamin D action and metabolism: Lessons learned at the Los Angeles Zoo. J Cell Biochem 88:308–314. 46. Wu S, Ren S, Chen H, Chun RF, Gacad MA, Adams JS 2000 Intracellular vitamin D binding proteins: novel facilitators of vitamin D-directed transactivation. Mol Endocrinol 14:1387–1397. 47. Wu S, Chun R, Gacad MA, Ren S, Chen H, Adams JS 2002 Regulation of 1,25-dihydroxyvitamin D synthesis by intracellular vitamin D binding protein-1. Endocrinology 143:4135. 48. Nykjaer A, Dragun D, Walther D, Vorum H, Jacobsen C, Herz J, Melsen F, Christensen EI, Willnow TE 1999 An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) vitamin D3. Cell 96:507–515. 49. Christensen EI, Willnow TE 1999 Essential role of megalin in renal proximal tubule for vitamin homeostasis. J Am Soc Nephrol 10:2224–2236.

CHAPTER 22

Vitamin D Receptor and Retinoid X Receptor Subcellular Trafficking JULIA BARSONY

Laboratory of Cell Biochemistry and Biology, National Institutes of Diabetes, Digestive and Kidney Diseases, NIH, Bethesda, Maryland

I. Evolving Concepts of Receptor Localization II. Nucleocytoplasmic Trafficking of Vitamin D Receptor and Retinoid X Receptor

III. Intranuclear Trafficking of Vitmain D Receptor and Retinoid X Receptor IV. Summary and Future Directions References

I. EVOLVING CONCEPTS OF RECEPTOR LOCALIZATION

proposed by E. V. Jensen (Fig. 1A) for the mechanism of steroid receptor actions in the 1960s [11]. Later, antibodies were raised against steroid receptors and were used for immunocytochemical detection. Results of these studies indicated that even the unliganded nuclear receptors reside exclusively in the nucleus. This initiated a dramatic shift in the paradigm of nuclear receptor localization. The evidence for receptor translocation was widely criticized as artifactual, and the preferred model advocated that both unliganded and liganded nuclear receptors are anchored to target gene sequences [12] (Fig. 1B). Opponents of this static model demonstrated that detergent treatment of cell membranes, which was necessary for getting the antibodies inside the cell, washed out many of the cytoplasmic receptors [13], and they considered the exclusively nuclear receptor localization an artifact. Soon after the cloning of VDR, antibodies were generated and used for the detection of VDR immunoreactivity in fixed cells [14]. Not too surprisingly, VDR immunoreactivity was initially detected in the nucleus both before and after calcitriol treatment [14], which was consistent with the model depicted in Fig. 1B. Subsequently, we developed a more sensitive immunocytology technique by replacing chemical fixation with microwave fixation. This fixation allowed antibody access to receptors without detergent use. Our studies with this method showed unliganded VDR in the cytoplasm and an apparent translocation of cytoplasmic VDR into the nucleus after calcitriol treatment [15], consistent with the model depicted in Fig. 1A. This concept was further supported by our studies with a biologically active fluorescent labeled calcitriol in living cells. Results of these experiments also indicated that high affinity binding sites reside in the cytoplasm and accumulate in the nucleus over time [16], which is consistent with the model depicted in Fig. 1A. Debate about VDR localization

It is well established that vitamin D receptor (VDR) mediates most, if not all, actions of calcitriol in target cells. Vitamin D receptors (VDRs) belong to the type II nuclear receptor family of proteins, which all act as heterodimers with retinoid X receptors (RXRs) to regulate transcription of their target genes. VDR undergoes a rapid activation process upon binding of calcitriol. This activation process involves multiple protein–protein interactions, conformational change, phosphorylation, and interaction with specific target gene sequences. An important aspect of VDR functioning is the temporal and spatial aspect of activation events. These events include protein–protein interactions: dimerization with RXR, interactions with coactivators [1], co-repressors [2,3], chaperones [4], kinases [5,6], calreticulin [7], and with nuclear import and export receptors [8]. Several of these protein interactions influence the binding of VDR and RXR to target genes and the interaction with the transcription machinery. Other protein interactions influence VDR localization and stability. Our understanding of the mechanisms of VDR functions has changed considerably over the past few years, in part due to new insights into VDR and RXR subcellular localization and trafficking. Even before the identification of VDR in the early 1980s, there was great interest in locating the receptor for calcitriol as a way to gain insight into the mechanism of calcitriol actions. Early studies with biochemical methods, such as radioligand and cell fractionation, detected the unliganded receptor for calcitriol mostly in the cytosolic fractions of target cells, whereas the liganded VDR was recovered mostly in nuclear fractions [9,10]. These findings supported the model of VDR translocation, which was similar to the model

364

JULIA BARSONY

RXR/VDR

VDR

A

Translocation

RXR/VDR

B

Anchored

RXR/VDR

RXR/VDR

C

Shuttling

FIGURE 1 Models of VDR/RXR Localization. (A) Translocation model. Model depicts the RXR in the nucleus and the unliganded VDR in both the cytoplasm and the nucleus. It stipulates that binding of calcitriol (red diamond) initiates VDR translocation from the cytoplasm into the nucleus, where dimerization of VDR with RXR would take place. (B) Anchored model. Model depicts both the unliganded and the liganded VDR and the RXR in the nucleus, attached to specific DNA recognition sequences. It assumes that ligand binding does not change receptor localization but initiates VDR recruitment of RXR and coactivators and an interaction with the transcription machinery. (C) Shuttling model. Model depicts both unliganded and liganded VDR forming dimers with RXR in the cytoplasm and in the nucleus. It assumes that change in the shuttling speed of VDR after hormone binding reflects the formation of a distinct macromolecular complex. See CD-ROM for color.

persisted until recently, and was only resolved when the use of green fluorescent protein (GFP) chimeras and modern imaging methods revealed the highly dynamic nature of VDR, RXR, and most other nuclear receptors. Transcriptionally active fluorescent protein chimeras of VDR were first generated in our laboratory [17] and later in other laboratories [18,19]. We detected the unliganded GFP-VDR both in the cytoplasm and the nucleus, and showed that after calcitriol addition GFPVDR accumulates in the nucleus [17]. A subsequent study found more unliganded GFP-VDR in the nucleus after fixation than we did in living cells [18]. Nonetheless, this study [18] and another study by Sunn et al. [19] both detected hormone-induced GFP-VDR translocation into the nucleus. These initial experiments with GFP-VDR, like previous studies using immunocytology, observed the steady-state distribution of receptors and appeared to favor the translocation model (Fig. 1A). More recently, we conducted kinetic analysis of VDR trafficking using photobleaching and in vitro transport assays. Results of these experiments revealed that unliganded VDR and liganded VDR both shuttle constantly between the cytoplasm and the nucleus [20] and suggested a novel dynamic model for VDR and RXR localization (Fig. 1C). This kinetic analysis revealed that the nucleocytoplasmic exchange rate is faster for the liganded GFP-VDR than for the unliganded GFPVDR [21], indicating that a faster nuclear import of the liganded VDR could give the appearance of hormoneinduced VDR translocation. Almost certainly, different protein–protein interactions explain the difference between the transport characteristics of the unliganded and the liganded VDR [22]. The most likely interacting partner of VDR that could influence hormonedependent change in VDR distribution is RXR. Dimerization with RXR is essential for both constitutive and calcitriol-dependent functions of VDR. Recently this has been elegantly demonstrated by knockout mouse studies [23–26]. Therefore, the subcellular localization of RXR and the location of dimerization are important aspects of calcitriol actions. Historically, immunocytology experiments led to the general consensus favoring exclusively nuclear localization of RXR. Also in living cells, microscopy showed fluorescent protein chimeras of RXR predominantly in the nucleus [27,28]. The translocation model for VDR actions was consistent with these findings (Fig. 1A), and assumed that VDR dimerizes with RXR in the nucleus after hormone binding. However, three lines of evidence contradicted this model. First, our fluorescence resonance energy transfer (FRET) experiments demonstrated the presence of VDR/RXR heterodimer not only in the nucleus, but also in the cytoplasm [27] (Fig. 2). Second, coexpression experiments showed that the wild-type RXR could bring

365

CHAPTER 22 VDR and RXR Subcellular Trafficking

A

B

FIGURE 2

Fluorescence Energy Transfer (FRET) Visualizes Interactions of Unliganded GFP-VDR and RXR-BFP. FRET signal is generated when the green VDR and blue RXR are closer than 50 Å. High FRET intensities are labeled with green pseudocolor and low background intensities with blue pseudocolor. Cells expressed fluorescent protein chimeras of wild-type VDR and either the wild-type RXR-BFP (A) or the heterodimerization incompetent mutant of RXR-BFP (B). Cells were cultured without serum or calcitriol before the experiment. FRET is most abundant in the nuclei, but also visible in the cytoplasm. Images were taken from living cells by confocal microscope (Zeiss LSM410). Bars = 10 µm. See CD-ROM for color.

into the nucleus a nuclear import mutant of VDR [20] (see detail in Section II). Third, photobleaching and in vitro transport assays demonstrated that unliganded and liganded RXR rapidly shuttle between the cytoplasm and the nucleus [21] (Fig. 3). Taken together these findings supported the novel, dynamic model for VDR/RXR localization (Fig. 1C). According to this model, both VDR and RXR constantly interact with large protein complexes to facilitate ongoing nuclear import and export and intranuclear trafficking.

II. NUCLEOCYTOPLASMIC TRAFFICKING OF VITAMIN D RECEPTOR AND RETINOID X RECEPTOR A. Nuclear Import and Export Mechanisms Nucleocytoplasmic transport of macromolecules through the nuclear pore complex (NPC) is a fundamental cellular process requiring energy. Whereas the NPC permits unrestrained passage of small molecules (up to 9 nm in diameter) through a central aqueous channel, transport of most proteins (over 20–40 kDa) and RNA can only pass through the NPC in both directions by receptor-mediated and Ran-GTP gradient

dependent active transport [29–32]. Rapid progress in cell biology has revealed the major components of this transport machinery (reviewed in [33,34]). The soluble transport receptors specifically recognize their cargoes. In addition, they often act as chaperone proteins to prevent aggregation of basic proteins [35]. Multiple classes of nuclear transport receptors exist, including the family of importin-β-like proteins (importins/exportins, also known as karyopherins) [36]. There are more than 20 members of this family and they mediate both nuclear import and export of proteins and RNA. The small nuclear transport factor 2 (NTF2/p10) imports the small GTPase Ran into the nucleus [37]. Proteins of the third transport receptor class, a heterodimer of TAP/NXF and p15/NXT, mediate mRNA export from the nucleus. All transport receptors interact with components of the NPC and contain a phenylalanine/glycine-rich motif. How do these transport receptors work? Transport receptors specifically bind their cargo in the originating compartment and then release them in the destination compartment. The regulation of this process by RanGTPase is best understood for transport mediated by importins and exportins (Fig. 4). Both importins and exportins bind to Ran-GTP through their conserved amino-terminal domains. The concentration of RanGTP is high in the nucleus, whereas the concentration of Ran-GDP is high in the cytoplasm. This gradient is maintained by Ran-GTPase activating protein and

366

JULIA BARSONY

Pre-bleach

Post-bleach

Recovery

Time (min)

GFP-VDR Vehicle

15–30

GFP-VDR Calcitriol

5–15

YFP-RXR Vehicle

20–30

FIGURE 3 Fluorescence recovery after photobleaching demonstrates that VDR and RXR shuttle between the cytoplasm and nucleus. COS-7 cells expressing GFP-VDR (upper and middle panels) were treated with vehicle (0.1% ethanol) or calcitriol (5 nM) for 1 hr before microscopy. COS7 cells expressing YFP-RXR (lower panels) were treated with vehicle. Cells with two nuclei were selected (left) and a 2 µm2 spot (4×4 pixel) within one of the nuclei (arrow) was exposed to a focused laser beam at full power for 1 sec/pixel. This intense illumination caused irreversible photobleaching (darkening) of all the receptors within the exposed nucleus, whereas the receptors in the other nucleus remained bright. After photobleaching, serial images were taken for measurements of fluorescence recovery. The images taken immediately after photobleaching are shown in the center panels and images taken at complete recovery are shown in the right panels. Range of recovery time (right) was calculated from 10 independent experiments. This is the time for reaching equilibrium of brightness between the two nuclei of the cell, which occurs due to the passage of bright and dark receptors through the cytoplasm. Bars, 10 µm. (See color plate).

Ran guanine nucleotide exchange factor. The gradient of Ran-GTP and its role in mediating cargo release from import/export receptors have been confirmed in living cells using FRET [38]. Nuclear import is mediated by import receptors, such as importin β (Fig. 4). Importins bind either directly or indirectly through adapter proteins to their cargo proteins in the cytoplasm without binding to Ran. Importin α is the most important adapter protein, which binds to basic amino acid–rich sequences on their cargo proteins, so-called nuclear localization sequences (NLSs). Importin α then binds to proteins of the importin β family, and the import receptor complex interacts with proteins of the nuclear pore complex in a Ran-dependent manner. It was reported that large transport cargoes might need more than one receptor

molecule for efficient translocation, because otherwise partitioning of the transport complex into the hydrophobic phase of the NPC may be hindered [39]. After crossing the nuclear pore complex, cargoes are released in the nucleus upon Ran-GTP binding. The importin/Ran-GTP complex is then recycled back to the cytoplasm. Export from the nucleus is mediated by export receptors (Fig. 4). Ran-GTP induces cargo binding to export receptors exclusively in the nucleus. After export to the cytoplasm, GTP hydrolysis by Ran-GAP and RanBP1/2 allows separation of the exportin/cargo/RanGTP complex. Ran is then recycled back to the nucleus by the importer NTF2. Nuclear export of proteins often involves multiple mechanisms; many of these mechanisms are poorly understood. The best-characterized

367

CHAPTER 22 VDR and RXR Subcellular Trafficking

GTP hydrolysis GDP

Cytoplasm

Importin

Exportin

NPC Nucleus Exportin

Importin GTP

GTP

Ran-GTP

GDP

Ran-GDP

GTP

Ran-GTP binding

Cargo

FIGURE 4 Schematic representation of nuclear import and export mechanisms. Ran-GTP gradient confers directionality of nuclear import and export receptor movement across the nuclear pore complex (NPC). Adapter proteins, such as importin α, recognize nuclear localization signals (NLS) on their cargo in the cytoplasm. This complex is stabilized by the binding of the import receptors importin β, which interacts with the importin β-binding (IBB) domain of the cargoloaded importin α. This triple import complex then docks to the cytoplasmic side of the NPC via importin β and translocates to the nuclear side of the NPC. There binding of Ran-GTP terminates translocation, leading to the displacement of importin α–cargo protein from the import complex and the formation of an importin β–Ran-GTP complex. The C-terminal-IBB-domain of the cargofree importin α binds to the export receptor CAS cooperatively with Ran-GTP, and is recycled back to the cytoplasm. The Ran-GTP bound importin β translocates back to the cytoplasm either alone or with cargoes that can directly bind to importin β. In the cytoplasm, the activities of Ran-binding protein 1 (RanBP1) and the Ran-specific GTPase-activating protein 1 (Ran-GAP) result in the hydrolysis of GTP on Ran, which leads to the disassembly of export complexes. CAS recycles back to the nucleus by passing through the NPC. Another import receptor, NTF2, together with its dimerization partner takes Ran into the nucleus. Export receptors bind with their cargo and with Ran-GTP in the nucleus. After this complex crosses the NPC, hydrolysis of GTP leads to the disassembly of the export complex, and exportin recycles back to the nucleus. See CD-ROM for color.

export pathway uses the export receptor called chromosomal region maintenance 1 protein (Crm-1) or exportin 1. Cargoes for Crm-1 are proteins that carry a leucinerich nuclear export signal (NES) [40] and RNA [41]. Leptomycin B (LMB) is a specific inhibitor of Crm-1 mediated export; it prevents binding of cargo proteins to Crm-1 [42]. Even without an NES, proteins can be exported from the nucleus by Crm-1 using adaptor proteins [43]. In addition, other Ran-independent export mechanisms carry mRNA and ribosomal subunits

out of the nucleus [44]. Newly identified exportins are CAS, an importin family member that binds to the large NES in the importin family of proteins, and exportin-t, which is responsible for tRNA export [45]. Calreticulin could also be an export receptor. It binds to specific sequences in the DNA-binding domains of nuclear receptors, and this binding is essential for the efficient export of these receptors [8,46]. However, the direct interaction of calreticulin with the nuclear pore complex has not been demonstrated, leaving open the possibility

368

JULIA BARSONY

that it is not an export receptor but an adaptor protein. Adding to the complexity of nuclear import and export mechanisms is indirect transport, utilizing a “piggyback” mechanism through cargo association with other exported proteins or RNA [47]. The recent rapid progress in the identification of import and export mechanisms for nuclear receptors is fueled by the understanding that receptor interactions with the import and export machinery represent a critical mechanism that regulates their transcriptional activities. During the past 2 years, several kinetic studies with fluorescent chimeras of nuclear receptors demonstrated that most of them shuttle constantly between cytoplasm and nucleus [48]. Their import is mediated through binding to importins. One or two bipartite NLSs have been found in the DNA-binding domain (DBD) of most nuclear receptors. In addition to the NLSs in the DBD, constitutive NLS was identified in the hinge region of the androgen receptor (AR) [49]. Ligandregulated NLS was found in the ligand-binding domain (LBD) of the glucocorticoid receptor (GR) [50] and the AR [51]. Several nuclear proteins utilize other import pathways; their arginine-rich NLS binds directly to importin β [52]. Finally, indirect nuclear import through a piggyback mechanism is also possible. Export of nuclear receptors is still not well defined. Early studies indicated that the export of progesterone receptors and estrogen receptors are mediated by passive diffusion [53]. More recent studies in living cells with fluorescent protein chimeras revealed that the export of nuclear receptors is also mediated by active transport. First, we demonstrated that Crm-1 mediates export of VDR. This prompted studies that revealed the Crm-1-mediated export of estrogen receptor (ER), TR, and retinoic acid receptor (RAR) [54]. GR, AR, and progesterone receptor (PR) did not appear to be DBD 24

exported by Crm-1 [46,55,56]. however, sequences important for the seemingly NLS-dependent export of GR, AR, and PR have been identified in the LBD [56]. Other reports indicated that calreticulin regulate export by interacting with conserved amino acids in the DBD of nuclear receptors. In addition to direct export by export receptors, nuclear receptors may be exported by a piggyback mechanism through association with other shuttling proteins or RNA [47]. Such indirect export has been reported for the RXR, via the phosphorylationdependent export of the orphan nuclear receptor NGFI-B that contains leucine-rich NESs [57]. Another example is the 14-3-3 protein that interacts with RIP140 co-repressor and exports it from the nucleus, thus enhancing transcriptional activity of GR [58]. Because RIP140 binds to both VDR [59] and RXR [60], this mechanism may be important for the regulation of calcitriol action. Current research is likely to elucidate novel export pathways of nuclear receptors and to resolve existing controversies. The complexities of nuclear import and export mechanisms for nuclear receptors provide a versatile regulatory pathway of hormone actions. Ongoing research will elucidate the spatial–temporal relationship between receptors and co-regulators [61,62] and the regulation and the functional significance of receptor export [63,64].

B. Interactions of VDR and RXR with Nuclear Import and Export Receptors Using wheat germ agglutinin in permeabilized cells, we were the first to demonstrate that VDR interacts with the NPC [15]. Subsequently, putative nuclear localization sequences were identified within the VDR (Fig. 5).

Hinge region 89

LBD 190

N

AF-2 427 C

NLS1

NLS2

NLS3

NLS4 154RPPVRVNDGGGSHPSRPNSR

69TKDNRRH 102RKREMILKRK

45KGFFRRSMKRKA

FIGURE 5 Schematic representation of hVDR functional domains and the putative nuclear localization sequences (NLSs). Amino acids representing the NLSs are shown in bold letters. Mutations of NLS2, NLS3, and NLS4 did not cause cytoplasmic retention of the receptor in living cells; their location in the VDR is marked with pink. See CD-ROM for color.

CHAPTER 22 VDR and RXR Subcellular Trafficking

The bipartite NLS between the two zinc-fingers of the DBD is homologous to the NLS found in steroid receptors (NLS1). Mutations of this region resulted in cytoplasmic retention of both unliganded and liganded VDR [65]. We later confirmed that this is a functional NLS in living cells by generating mutations in NLS1 of the GFP-VDR (K53Q, R54G, K55E) [66]. Another cluster of basic amino acids (NLS2) in the second zinc finger is homologous to the NLS found in the orphan nuclear receptor TR2 [67]. Mutations of these basic amino acids to lysine did not cause cytoplasmic retention of GFP-VDR (Barsony J. and Prufer K., unpublished); thus this segment is unlikely to function as NLS. Two more clusters of basic amino acids within the VDR appeared initially to function as NLSs. Michigami and colleagues showed that a short segment of 20 amino acids in the hinge-region of VDR (NLS4; 154–173) enabled cytoplasmic GFP-tagged alkaline phosphatase to translocate to nuclei [18]. Surprisingly, in the context of the intact receptor we were unable to confirm that this segment functions as NLS using mutational analysis. Mutations at amino acids 154–158 (R154G, P155A, P156A, R158G) and at amino acids 169–173 (R169G, P170A, R173G) did not prevent nuclear import of GFPVDR (Barsony J. and Prufer K., unpublished). Another peptide representing amino acids 102–111 of VDR has also been shown to enable nuclear accumulation of the fluorescein-labeled IgG [68]. We deleted these amino acids 102–110 (NLS3) of the GFP-VDR without significant impairment of the nuclear import of the unliganded VDR (Fig. 6). In addition to constitutive NLSs, VDR segments that are important for the calcitriol-induced redistribution have been defined (Fig. 6). We have shown that liganddependent acceleration of GFP-VDR import is disturbed by deletion or mutations in the activation-function 2 (AF2) domain [17,21]. This finding indicated that coactivators could play a role in the regulation of VDR import. Deletion of NLS3 also abolished hormonedependent nuclear accumulation of GFP-VDR (Fig. 6). Coincidentally, the residues 105–109 within this putative NLS of VDR binds to the hsp70 [69]. The loss of hsp70 binding to VDR has no effect on hormone binding and coactivator interactions [69]. Therefore, this finding could indicate a role for hsp70 in the regulation of nuclear import of the liganded VDR. RXR also interacts with import receptors. Figure 7 shows a putative NLS in the first zinc finger of RXR, which is a homolog of the NLS1 of VDR. To test the functional significance of this NLS, we introduced point mutations at amino acids 160, 161, 164, and 165 (K160Q, R161G, R164G, and K165Q) [20]. Microscopy showed that the mutant nls1YFP-RXR is retained in the cytoplasm (Fig. 7). An additional NLS is located between amino acids 181 and 186 within the second

369 zinc finger of RXR (NLS2). Mutations in this segment (R182E, R184E) caused cytoplasmic retention of the nls2YFP-RXR (Fig. 7). Taken together, these findings show that both VDR and RXR interact with nuclear import receptors via multiple binding sites. We used coexpression experiments with wild-type and NLS mutants of VDR and RXR to study the impact of dimerization on receptor nuclear import (Fig. 8). Coexpression of wild-type RXR-BFP restored nuclear localization of nls1GFP-VDR, whereas coexpression of nls1RXR-BFP retained the wild-type GFP-VDR in the cytoplasm. Moreover, calcitriol treatment induced nuclear accumulation nls1RXR-BFP when it was coexpressed with GFP-VDR. These data indicate that RXR dominates the nuclear import of unliganded VDR, whereas liganded VDR dominate the nuclear import of RXR. In addition, these data strongly support the notion that VDR and RXR dimerize in the cytoplasm. Similar to other nuclear receptors, VDR and RXR also exit the nucleus through the NPC [21]. We studied this export by conventional permeabilization experiments and by monitoring fluorescence recovery after photobleaching (FRAP) (Fig. 3). This later technique allowed us to characterize the nucleocytoplasmic shuttling speed; the recovery half-times were 15–30 min for the unliganded VDR and 5–15 min for liganded VDR, and 20–30 min for unliganded RXR [20]. Results indicated that both VDR and RXR have to interact frequently with nuclear import and export receptors. Nuclear export of unliganded GFP-VDR was inhibited by LMB treatment, resulting in an exclusively nuclear receptor accumulation (Fig. 9). This finding indicates that the unliganded VDR is exported by Crm-1. Calcitriol treatment prevented LMB sensitivity of VDR, suggesting that this faster VDR export is mediated by a Crm-1 independent mechanism. We searched for the NES of VDR by mutating the leucine-rich sequences. Mutations of the leucines 320-325 to alanines in the LBD of VDR abrogated LMB sensitivity [21]. In addition, nuclear export of GFP-VDR was inhibited by mutations at the conserved calreticulin-binding site in the DBD (F47A, F48A). Nuclear export of RXR was also prevented by mutations in the calreticulin-binding site (F158A, F159A), but the RXR export was not inhibited by treatment with LMB [21]. Interestingly, RXR can also be exported by an indirect mechanism. Katagiri and colleagues showed that treatment with NGF induced phosphorylation of orphan nuclear receptor NGFI-B (also called nur77) at amino acid 105, and this phosphorylation induced NGFI-B export together with the export of the dimerization partner, RXR [57]. It is possible that RXR export by other dimerization partners could influence VDR nucleocytoplasmic trafficking. The research of nuclear export pathways is rapidly progressing and is likely to yield a better insight into

370

JULIA BARSONY

Vehicle

Calcitriol

GFP-VDR

FIGURE 6

Intact binding sites for coactivators and heat shock protein 70 are required for calcitriol-dependent nuclear accumulation of VDR. Leucines 417 and 420 were changed to alanines to create cofGFP-VDR, and amino acids 102-110 were deleted to create ∆hsp70GFP-VDR (also referred to as NLS3). COS-7 cells were transfected with expression vectors 24 hr before microscopy and were treated with 10 nM calcitriol for the last 1 hr in serum- and phenol red-free medium. Calcitriol induced nuclear accumulation of GFP-VDR, but failed to change the nucleocytoplasmic distribution of cofGFPVDR and ∆hsp70GFP-VDR. Images were taken from living cells by confocal microscope (Zeiss LSM410). Bars, 10 µm. See CD-ROM for color.

cof GFP-VDR

∆hsp 70GFP-VDR

DBD 140

Hinge 206

LBD

AF-2

265

467

N

C

FIGURE 7 Schematic represen-

NLS1 NLS2

tation of RXR functional domains and the putative nuclear localization sequences (NLSs). Images are shown from HEC293 cells stably expressing wild-type YFP-RXR or its mutants as indicated. Mutation in NLS1 and NLS2 both cause cytoplasmic localization of YFPRXR. Images were taken from living cells by confocal microscope (Zeiss LSM410). Bars, 10 µm. (See color plate).

YFP-RXR

NLS1YFP-RXR

NLS2YFP-RXR

371

CHAPTER 22 VDR and RXR Subcellular Trafficking

GFP-VDR

GFP-VDR+ nlsRXR-BFP

the mechanisms of VDR and RXR export in the near future.

C. Importance of VDR/RXR Shuttling for Transcriptional Activities Vehicle

Calcitriol

nlsGFP-VDR

nlsGFP-VDR +RXR-BFP

Vehicle

Calcitriol

In vivo, the importance of nuclear import and export for calcitriol actions has been demonstrated by experiments on fibroblasts from patients with hereditary resistance to calcitriol [15]. As expected, mutations in the NLSs of the VDR and RXR inhibited calcitriol-induced transcriptional activities. Luciferase reporter assays (Fig. 10) showed that coexpression of nls1GFP-VDR with nls1RXR-BFP completely blocked transcriptional activity, and even the nls1GFP-VDR expressed alone had minimal transcriptional activity. The activity of nls1GFP-VDR was substantially increased by coexpression of wild-type RXR. Coexpression of nls1RXR-BFP with wild-type GFP-VDR reduced calcitriol-induced transcription of 24Ohase/Luc. These results showed a close correlation between transcriptional activities and nuclear localization, indicating the physiological relevance of receptor compartmentalization [20]. Unexpectedly, inhibition of nuclear export also blocked transcriptional activity (Fig. 9C). Treatment of cells expressing GFP-VDR with LMB caused nuclear retention of receptors and prevented the effect of calcitriol on reporter gene expression. LMB was ineffective to change receptor export and transcriptional activity when it was added after calcitriol [20]. Based on these results we suggested for the first time that receptor export contributes to transcriptional activity of VDR. Subsequently, the importance of nuclear export for signaling was demonstrated for other proteins, such as Smad 1[70].

D. Regulation of VDR and RXR Nucleocytoplasmic Trafficking FIGURE 8

VDR and RXR nuclear import by “piggyback” mechanism. HEK293 cells expressed wild-type GFP-VDR alone or coexpressed GFP-VDR and mutant nlsRXR-BFP (upper panels). HEK293 cells expressed mutant nlsGFP-VDR alone or coexpressed nlsGFP-VDR and wild-type RXR-BFP (lower panels). Images show the green signal (GFP), whereas the blue (BFP) signal is not shown. Alone, both nlsGFP-VDR and nlsRXR-BFP sequestered in the cytoplasm. Coexpression of nlsRXR-BFP caused cytoplasmic retention of GFP-VDR. Calcitriol addition induced nuclear accumulation of GFP-VDR despite the presence of nlsRXR-BFP. Calcitriol failed to induce nuclear accumulation of nlsGFP-VDR. Coexpression of RXR-BFP allowed nuclear accumulation of nlsGFP-VDR. These findings suggest that RXR and VDR form dimers in the cytoplasm. Furthermore, these data indicate that RXR dominates the import of the unliganded VDR and liganded VDR dominates the import of RXR. Bars, 10 µm. See CD-ROM for color.

Nucleocytoplasmic partitioning is a regulatory tool for signaling utilized in a wide variety of species, including yeast, plants, and vertebrates. Several regulatory mechanisms have been described: oligomerization; multiple NLSs and NESs acting in concert; importin or exportin overexpression; increase or decrease in the cargo affinity to importins or exportins; cargo sequestration by binding to cytoplasmic or nuclear docking proteins; and cargo NLS or NES masking by conformational change, phosphorylation, acetylation, or protein binding. Various signaling pathways make use of this regulatory potential. Well-characterized examples

372

JULIA BARSONY

A

DBD

Hinge region

24

89

LBD 190

AF-2 427

N

C 47-48 Calreticulin

320-325 Crm-1 Vehicle

Leptomycin B

B

C

16

24-Ohase/Lucactivities Fold Induction

14 12 10 8 6 4 2 0 Vehicle

LeptomycinB

FIGURE 9 Nuclear export of VDR. (A) Schematic representation of VDR functional domains indicating the sequences important for export by calreticulin and by Crm-1 receptors. (B) Treatment with leptomycin B caused nuclear accumulation of GFP-VDR by preventing its Crm-1 mediated export. GL48 cells (a HEK293 derived line stably expressing GFP-VDR) were treated with vehicle (DMSO) or leptomycin B (2 nM for 3 hr) before microscopy. Images of living cells were taken by confocal microscope (Zeiss LSM410). Bar, 10 µm. (C) Inhibition of VDR export decreased calcitriol-induced transcriptional activity. ROSA I cells, which stably express a 24-hydroxylase-luciferase reporter, were first treated with either vehicle (DMSO) or with leptomycin B (2 nM for 3 hr and then treated with 10 nM calcitriol for 24 hr. Luminescence values are expressed as mean ± 1 S.D. See CD-ROM for color.

are NF-κB [71], MAP kinases [72], STAT [73], Smad1 [70], and p53 [74,75]. To date, very little is known about the regulation of VDR and RXR nucleocytoplasmic shuttling, and data regarding the regulation of other nuclear receptors are just starting to emerge. The first insight in the regulation of VDR nucleocytoplasmic trafficking came from coexpression experiments. We found that dimerization with RXR promotes nuclear import and nuclear retention of

VDR [76]. Similarly, dimerization with RXR has also been shown to increase nuclear retention of TR [28]. Calcium/calmodulin is one of the important regulators of nuclear import. This mechanism has been shown to play a critical role for NF-κB/Rel signaling. In most cells, binding of inhibitory I-kB protein (IκB) to NF-κB/Rel masks their NLS and thus NF-κB/Rel are sequestered in the cytoplasm. About 150 different signals can stimulate NF-κB/Rel signaling, including UV radiation,

373

CHAPTER 22 VDR and RXR Subcellular Trafficking

450000

400000

Normalized luciferase activity

350000

300000 250000

200000 150000

100000 50000

0 GFP-VDR

nls GFP-VDR

nls GFP-VDR + RXR-BFP

GFP-VDR + nls RXR-BFP

FIGURE 10

Nuclear import of VDR and RXR correlates with calcitriol-induced transcriptional activities. The mutant nlsGFP-VDR lost almost all transcriptional activities. Coexpression of RXR-BFP partially restored transcriptional activities of nlsGFP-VDR. Coexpression of nlsRXR-BFP with GFP-VDR decreased transcriptional activity of GFPVDR. COS-7 cells were transiently transfected with the receptor expression vectors as indicated, together with 24-hydroxylase-luciferase reporter and ß-galactosidase expression vectors, using Lipofectamine 2000 reagents as described [21]. Normalized luciferase activities from vehicle-treated cells are shown as open bars, and from calcitriol-treated (10 nM for 24 hr) cells are shown as shaded bars. Luminescence data were normalized with β-galactosidase activities, and data are expressed as mean ±1 S.D.

oxidative stress, inflammation, viruses, and bacterial lipopolysaccharides. Cellular stimulation results in phosphorylation, ubiquitination, and degradation of IkB, allowing the translocation of NF-κΒ/Rel to the nucleus and the action on target genes. Calmodulin (CaM) is a highly conserved, ubiquitously expressed Ca2+ binding protein that serves as a key mediator of intracellular Ca2+ signals. After signal-dependent release of Rel-c from the docking, CaM binds Rel-c adjacent to its NLS in a calcium-dependent manner and blocks its nuclear translocation [77]. This mechanism plays an important role in calcium-mediated inhibition of inflammatory response. Another example is the regulation of nuclear factor of activated T cell (NFAT) activity by calcium [78]. NFAT proteins are expressed in most immune system cells and regulate the transcription of cytokine genes critical for the immune response. The activity of NFAT proteins is tightly regulated by the

calcium/calmodulin-dependent protein phosphatase 2B and calcineurin (CaN). Dephosphorylation of NFAT by CaN is required for NFAT nuclear localization. Current immunosuppressive drugs such as cyclosporine A and FK506 block CaN activity, thus inhibiting nuclear translocation of NFAT and consequent cytokine gene transcription. We found that the ligand-dependent nuclear import of GR was not affected by depletion of intracellular calcium stores [79]. Nuclear import of VDR was also unperturbed by calcium depletion (Barsony J., unpublished). In contrast, the export of nuclear receptors is regulated by calcium-dependent binding of calreticulin to their DBD. Depletion of calcium caused nuclear retention of GR, whereas overexpression of calreticulin or cell fusion–induced release of calreticulin from the endoplasmic reticulum enhanced GR export [46]. Similar calcium-dependent regulation of VDR export is expected, since VDR

374 export is also mediated by a calreticulin binding sensitive pathway [20]. Docking in the cytoplasm can retard import and docking in the nucleus can retard export of proteins. Docking of unliganded steroid receptors in the cytoplasm is caused by their binding to heat shock protein 90 (hsp90), which masks their NLS [80]. Whereas hsp90 binding does not retain unliganded VDR and RXR in the cytoplasm, binding to microtubules could serve as potential docking mechanism for VDR in the cytoplasm [81]. There are indications for nuclear docking of steroid receptors. Tethering to the nuclear matrix by the small RING finger protein SNURF has been shown to retard the export of AR after hormone withdrawal [51]. A similar mechanism could be involved in the regulation of VDR export, as an association of VDR with the nuclear matrix has been described [82,83]. Our Triton X-100 permeabilization experiments indicated that about 2% of VDR is docked in the nucleus [21]. Protein interactions that facilitate export of nuclear receptors are emerging. Binding to 14-3-3 protein has been shown to facilitate nuclear export of GR, AR, ER, and TR, but not PPAR and RXR [58,84]. Increased expression of export receptors (calreticulin, Crm-1) has been shown to facilitate export of nuclear receptors [85]. Interaction with cell-cycle regulators could also be important. We observed cell-cycle regulated changes in VDR compartmentalization in HEK293 cells stably expressing GFP-VDR [86]. Nuclear import and export mechanisms that transport coactivators and corepressors in and out of the nucleus adds to the complexity of the regulation of VDR trafficking. For example, SRC-1 has been shown to shuttle between the cytoplasm and the nucleus, exported from the nucleus by Crm-1 [87]. SMRT phosphorylation and nuclear export is associated with inhibition of corepressor function [88]. These and other protein interactions are likely to be important for the control of VDR and RXR trafficking. Receptor modifications, such as acetylation and phosphorylation, represent additional potential regulatory mechanisms of VDR and RXR trafficking. Receptor acetylation at lysine is one of the mechanisms that control transcriptional activities. Important studies have demonstrated that a conserved lysine-rich motif within the hinge region/LBD of the ER and AR is directly acetylated [89,90]. Ligand-induced acetylation promoted coactivator binding and reduced co-repressor binding. Coincidently, the same sequence functions as hormone-dependent NLS of the AR. A potential acetylation signal sequence is also present in the VDR [89]. Therefore, this region could be a regulatory site for nuclear import as well. Phosphorylation is another potential regulatory mechanism of subcellular trafficking. For example, the nuclear retention of thyroid hormone

JULIA BARSONY

receptor alpha1 (TRα1) is mediated by receptor phosphorylation [91]. Exploration of these and other regulatory mechanisms for VDR and RXR trafficking is expected to reveal important and often cell type–specific pathways that modulate hormone actions.

III. INTRANUCLEAR TRAFFICKING OF VITAMIN D RECEPTOR AND RETINOID X RECEPTOR In our original studies using photobleaching in the nuclei of GFP-GR expressing cells, we found that the receptor is in rapid circulation within the nucleus (Fig. 3) [79,92]. Since then, several studies described intranuclear motion of various steroid receptors based on photobleaching experiments [61,62,93–95]. These studies characterized the speed of receptor motion by measuring the half-maximal time for fluorescence recovery, after photobleaching an area within the nucleus. Estrogen binding decreased the mobility of GFP-ER; the t1/2 was 1.6 sec for unliganded, 5.8 sec for estrogen bound [96]. In another study by Stenoien et al., the t1/2 for unliganded ER was less than 1 sec, for estrogen-bound ER 5–6 sec. A similar decrease in the intranuclear mobility of GFP-RAR has been observed after retinoic acid treatment, but not in the mobility of GFP-TR [96]. We found that calcitriol binding also decreased the intranuclear mobility of GFP-VDR; the t1/2 for the unliganded receptor was about 2 sec and for the liganded receptor was 5–7 sec (Barsony et al, unpublished). Hormone treatment induces focal accumulation of most nuclear receptors, including the VDR (Fig. 11) [27]. Mutational analysis indicated that the focal accumulation of GFP-VDR signifies binding to regulatory DNA sequences [76]. A mutation that prevented DNA binding of VDR prevented the formation of discrete bright foci in the nuclei of GFP-VDR–expressing cells (Fig. 11). In addition, microscopy showed that GFPVDR and RXR-BFP bind to the same DNA-binding sites (foci). In cells that coexpressed GFP-VDR and RXRBFP, calcitriol induced accumulation of both receptors in the same foci [97]. Moreover, FRET experiments confirmed that GFP-VDR and RXR-BFP bind to foci as heterodimers [97] (Fig. 11). We used two types of photobleaching techniques to study the dynamics of GFPVDR exchange between foci and areas of diffuse signal. For FRAP, a 0.2 µm2 area (less than 5% of total nuclear area) corresponding to a discrete bright spot within the nucleus was subjected to intense laser illumination for 0.5 sec, which caused the receptors in the illuminated area to completely lose fluorescence. Following this

375

CHAPTER 22 VDR and RXR Subcellular Trafficking

GFP-VDR

dna GFP-VDR

A

GFP-VDR + RXR-BFP

GFP-VDR + hd RXR-BFP

B

FIGURE 11 Calcitriol induces focal accumulation of GFP-VDR and RXR-BFP in the nucleus. (A) GFP-VDR and dnaGFP-VDR were stably expressed in COS-7 cells. Without calcitriol, GFP-VDR is diffusely distributed in the nucleus (not shown). Treatment with calcitriol (10 nM calcitriol for 1 hr) caused formation of bright foci (left). A mutation (R77Q) was introduced in the DNA-binding region of GFP-VDR to create dnaGFPVDR [76]. Calcitriol did not cause focal accumulation of dnaGFP-VDR (right), indicating that intranuclear foci signify DNA binding. (B) FRET signals from the nuclei of COS7 cells stably expressing either GFP-VDR and RXR-BFP (left) or GFP-VDR and a heterodimerization incompetent mutant hdRXR-BFP (right). Cells were treated with calcitriol (10 nM for 1 hr) before microscopy. The high-intensity FRET signal shows that GFP-VDR and RXR-BFP accumulate in the same foci (left). Low, background intensity signal was generated in nuclei of cells expressing GFP-VDR and hdRXR-BFP, indicating lack of dimerization. Images were taken from living cells by confocal microscope (Zeiss LSM410). Bars, 10 µm. (See color plate).

photobleaching, serial images were taken and fluorescence recovery was calculated in the area that was bleached. Data were obtained from at least 20 nuclei of COS-7 cells. For fluorescence loss in photobleaching (FLIP) measurements, a 0.4-µm2 area corresponding to diffuse nuclear signal distant from the selected bright spot was subjected to intense laser illumination for increasing times (0.5 sec to 1 min) in 1-sec increments. Loss of fluorescence was measured over the selected bright spot to detect the time-dependent decrease in

fluorescence intensities, as more and more receptors turned black in the nucleus because of photobleaching. As expected, shorter photoexposure caused loss of fluorescence outside the foci (6.5 sec), and longer photoexposure caused complete loss of fluorescence even in the foci (42 sec). From this difference, the exchange rate between receptor and DNA-binding sites can be calculated. Similar FLIP experiments were performed in COS-7 cells that express YFP-RXR (Barsony et al., unpublished).

376 Results of our photobleaching experiments are in good agreement with results of FRAP and FLIP experiments using GFP-GR. Two groups used fluorescent chimeras of steroid receptors in cells that contain enlarged well-defined DNA-binding sites for these receptors. Interaction of GFP-GR with a large tandem array of MMTV-LTR-ras-BPV have been studied by Hager and colleagues [62,95]. Measuring fluorescence loss in photobleaching (FLIP) and fluorescence recovery after photobleaching (FRAP) they found that the exchange rate between the array and receptors GFP-GR outside the array was about 5 sec [62]. In contrast, measuring the interaction of CFP-LacER with an array of lac operator elements Stenoien and colleagues [93] found that even 20 min of bleaching failed to result in loss of fluorescence over the array after estradiol treatment, suggesting that this receptor was immobilized on the array. Nonetheless, the same group found that coactivator interaction with this array was short; SRC recovery was observed within 20–30 sec, and CBP recovery within 20 sec. These results argued against the belief that steroid receptor complexes are statically bound to regulatory sites in hormone-responsive promoters, and suggested that receptors and their cofactors must interact with the response element repeatedly during transcriptional activation. The short duration of receptor–DNA interaction has been further supported by chromatin immunoprecipitation studies [98].

IV. SUMMARY AND FUTURE DIRECTIONS A significant advance during the past 3 years is the understanding that VDR, RXR, and most other nuclear receptors and co-regulators constantly shuttle in and out of the nucleus and remain in rapid motion within the nucleus. During this movement, VDR forms complexes with multiple protein machines in a sequential manner. Earlier, we learned which proteins can bind VDR, and now is the time to learn where and when these protein interactions take place. The ability to monitor receptor movement in living cells by fluorescent techniques in real time gives a new insight into the “rules of the game,” the hormonal effects. With multispectral imaging, several “players” can be monitored simultaneously, and their interaction can be detected using FRET. Expression of fluorescent protein chimeras of VDR and RXR is already permitting mechanistic understanding of nuclear import and export. Multiple nuclear localization signals have been identified in both receptors and in their coactivators and co-repressors. A better understanding of the export pathways is expected shortly.

JULIA BARSONY

Available data strongly suggest that transcriptional regulation is a highly dynamic, sequential process. The technologies that we developed have already been applied to studies on the trafficking of steroid receptors and other RXR partners, such as TR. Ongoing research is likely to elucidate the regulatory mechanisms that influence VDR and RXR trafficking in the near future. These mechanisms will generate a broad interaction “field” with multiple signaling processes converging to modulate calcitriol actions. In addition, our ability to monitor receptor–DNA interactions at a time resolution of seconds provides us with an opportunity to explore the dynamic assembly of transcription complexes at the regulatory sites.

References 1. MacDonald PN, Baudino TA, Tokumaru H, Dowd DR, Zhang C 2001 Vitamin D receptor and nuclear receptor coactivators: crucial interactions in vitamin D–mediated transcription. Steroids 66:171–176. 2. Polly P, Herdick M, Moehren U et al. 2000 VDR-Alien: A novel, DNA-selective vitamin D3 receptor–corepressor partnership. FASEB J 14:1455–1463. 3. Zhang C, Baudino TA, Dowd DR et al. 2001 Ternary complexes and cooperative interplay between NCoA-62/Ski-interacting protein and steroid receptor coactivators in vitamin D receptormediated transcription. J Biol Chem 276:40614–40620. 4. Lutz W, Kohno K, Kumar R 2001 The role of heat shock protein 70 in vitamin D receptor function. Biochem Biophys Res Commun 282:1211–1219. 5. Barletta F, Freedman LP, Christakos S 2002 Enhancement of VDR-mediated transcription by phosphorylation: Correlation with increased interaction between the VDR and DRIP205, a subunit of the VDR-interacting protein coactivator complex. Mol Endocrinol 16:301–314. 6. Jurutka PW, MacDonald PN, Nakajima S et al. 2002 Isolation of baculovirus-expressed human vitamin D receptor: DNA responsive element interactions and phosphorylation of the purified receptor. J Cell Biochem 85:435–457. 7. Wheeler DG, Horsford J, Michalak M, White JH, Hendy GN 1995 Calreticulin inhibits vitamin D3 signal transduction. Nucleic Acids Res 23:3268–3274. 8. Holaska JM, Black BE, Love DC et al. 2001 Calreticulin is a receptor for nuclear export. J Cell Biol 152:127–140. 9. Putkey JA, Wecksler WR, Norman AW 1978 The interaction of 1,25-dihydroxyvitamin D3 with its intestinal mucosa receptor: kinetic parameters and structural requirements. Lipids 13: 723–729. 10. Bloor JH, Dasmahapatra A, Weiser MM, Klohs WD 1982 Kinetics of subcellular distribution in rat intestine of 1,25dihydroxycholecalciferol administered in vivo. Evidence for concentration within 5 min into purified nuclei. Biochem J 208:567–575. 11. Jensen EV, Suzuki T, Kawashima T et al. 1968 A two-step mechanism for the interaction of estradiol with rat uterus. Proc Natl Acad Sci USA 59:632–638. 12. Gorski J, Welshons WV, Sakai D et al. 1986 Evolution of a model of estrogen action. Recent Prog Horm Res 42:297–329.

CHAPTER 22 VDR and RXR Subcellular Trafficking

13. Wikstrom AC, Bakke O, Okret S, Bronnegard M, Gustafsson JA 1987 Intracellular localization of the glucocorticoid receptor: Evidence for cytoplasmic and nuclear localization. Endocrinology 120:1232–1242. 14. Clemens TL, Garrett KP, Zhou XY et al. 1988 Immunocytochemical localization of the 1,25-dihydroxy vitamin D3 receptor in target cells. Endocrinology 122: 1224–1230. 15. Barsony J, Pike JW, DeLuca HF, Marx SJ 1990 Immunocytology with microwave-fixed fibroblasts shows 1α,25-dihydroxyvitamin D3–dependent rapid and estrogen-dependent slow reorganization of vitamin D receptors. J Cell Biol 111:2385–2395. 16. Barsony J, Renyi I, McKoy W 1997 Subcellular distribution of normal and mutant vitamin D receptors in living cells. Studies with a novel fluorescent ligand. J Biol Chem 272:5774–5782. 17. Racz A, Barsony J 1999 Hormone-dependent translocation of vitamin D receptors is linked to transactivation. J Biol Chem 274:19352–19360. 18. Michigami T, Suga A, Yamazaki M et al. 1999 Identification of amino acid sequence in the hinge region of human vitamin D receptor that transfers a cytosolic protein to the nucleus. J Biol Chem 274:33531–33538. 19. Sunn KL, Cock TA, Crofts LA, Eisman JA, Gardiner EM 2001 Novel N-terminal variant of human VDR. Mol Endocrinol 15:1599–1609. 20. Prufer K, Barsony J 2002 Retinoid X receptor dominates the nuclear import and export of the unliganded vitamin D receptor. Mol Endocrinol 16:1738–1751. 21. Prufer K, Barsony J 2002 Retinoid x receptor dominates the nuclear import and export of the unliganded vitamin d receptor. Mol Endocrinol 16:1738–1751. 22. Barsony J, Prufer K 2002 Vitamin D receptor and retinoid X receptor interactions in motion. Vitam Horm 65:345–376. 23. Kato S, Takeyama K, Kitanaka S et al. 1999 In vivo function of VDR in gene expression-VDR knock-out mice. J Steroid Biochem Mol Biol 69:247–251. 24. Mathieu C, Van Etten E, Gysemans C et al. 2001 In vitro and in vivo analysis of the immune system of vitamin D receptor knockout mice. J Bone Miner Res 16:2057–2065. 25. Chen CH, Sakai Y, Demay MB 2001 Targeting expression of the human vitamin D receptor to the keratinocytes of vitamin D receptor null mice prevents alopecia. Endocrinology 142: 5386–5389. 26. Li M, Indra AK, Warot X et al. 2000 Skin abnormalities generated by temporally controlled RXRalpha mutations in mouse epidermis. Nature 407:633–636. 27. Prufer K, Racz A, Lin GC, Barsony J 2000 Dimerization with retinoid X receptors promotes nuclear localization and subnuclear targeting of vitamin D receptors. J Biol Chem 275: 41114–41123. 28. Baumann CT, Maruvada P, Hager GL, Yen PM 2001 Nuclear cytoplasmic shuttling by thyroid hormone receptors. Multiple protein interactions are required for nuclear retention. J Biol Chem 276:11237–11245. 29. Mattaj IW, Englmeier L 1998 Nucleocytoplasmic transport: the soluble phase. Annu Rev Biochem 67:265–306. 30. Gorlich D, Kutay U 1999 Transport between the cell nucleus and the cytoplasm. Annu Rev Cell Dev Biol 15:607–660. 31. Weis K 2003 Regulating access to the genome: Nucleocytoplasmic transport throughout the cell cycle. Cell 112:441–451. 32. Gorlich D, Seewald MJ, Ribbeck K 2003 Characterization of Ran-driven cargo transport and the RanGTPase system by kinetic measurements and computer simulation. EMBO J 22:1088–1100.

377 33. Chook YM, Blobel G 2001 Karyopherins and nuclear import. Curr Opin Struct Biol 11:703–715. 34. Kuersten S, Ohno M, Mattaj IW 2001 Nucleocytoplasmic transport: Ran, beta and beyond. Trends Cell Biol 11:497–503. 35. Jakel S, Mingot JM, Schwarzmaier P, Hartmann E, Gorlich D 2002 Importins fulfil a dual function as nuclear import receptors and cytoplasmic chaperones for exposed basic domains. EMBO J 21:377–386. 36. Macara IG 2001 Transport into and out of the nucleus. Microbiol Mol Biol Rev 65:570–94, table. 37. Smith A, Brownawell A, Macara IG 1998 Nuclear import of Ran is mediated by the transport factor NTF2. Curr Biol 8:1403–1406. 38. Kalab P, Weis K, Heald R 2002 Visualization of a Ran-GTP gradient in interphase and mitotic Xenopus egg extracts. Science 295:2452–2456. 39. Ribbeck K, Gorlich D 2002 The permeability barrier of nuclear pore complexes appears to operate via hydrophobic exclusion. EMBO J 21:2664–2671. 40. Stade K, Ford CS, Guthrie C, Weis K 1997 Exportin 1 (Crm1p) is an essential nuclear export factor. Cell 90:1041–1050. 41. Fornerod M, Ohno M. 2002 Exportin-mediated nuclear export of proteins and ribonucleoproteins. Results Probl Cell Differ 35:67–91. 42. Ossareh-Nazari B, Bachelerie F, Dargemont C 1997 Evidence for a role of CRM1 in signal-mediated nuclear protein export. Science 278:141–144. 43. Trotta CR, Lund E, Kahan L, Johnson AW, Dahlberg JE 2003 Coordinated nuclear export of 60S ribosomal subunits and NMD3 in vertebrates. EMBO J 22:2841–2851. 44. Clouse KN, Luo MJ, Zhou Z, Reed R 2001 A Ran-independent pathway for export of spliced mRNA. Nat Cell Biol 3:97–99. 45. Yi R, Bogerd HP, Wiegand HL, Cullen BR 2002 Both ran and importins have the ability to function as nuclear mRNA export factors. RNA 8:180–187. 46. Holaska JM, Black BE, Rastinejad F, Paschal BM 2002 Ca2+-dependent nuclear export mediated by calreticulin. Mol Cell Biol 22:6286–6297. 47. Segal SP, Graves LE, Verheyden J, Goodwin EB 2001 RNARegulated TRA-1 nuclear export controls sexual fate. Dev Cell 1:539–551. 48. Bunn CF, Neidig JA, Freidinger KE et al. 2001 Nucleocytoplasmic Shuttling of the Thyroid Hormone Receptor alpha. Mol Endocrinol 15:512–533. 49. Zhou ZX, Sar M, Simental JA, Lane MV, Wilson EM 1994 A ligand-dependent bipartite nuclear targeting signal in the human androgen receptor. Requirement for the DNA-binding domain and modulation by NH2-terminal and carboxyl-terminal sequences. J Biol Chem 269:13115–13123. 50. Picard D, Yamamoto KR 1987 Two signals mediate hormonedependent nuclear localization of the glucocorticoid receptor. EMBO J 6:3333–3340. 51. Poukka H, Karvonen U, Yoshikawa N et al. 2003 The RING finger protein SNURF modulates nuclear trafficking of the androgen receptor. J Cell Sci 113(Pt 17):2991–3001. 52. Palmeri D, Malim MH 1999 Importin beta can mediate the nuclear import of an arginine-rich nuclear localization signal in the absence of importin alpha. Mol Cell Biol 19:1218–1225. 53. Guiochon-Mantel A, Lescop P, Christin-Maitre S et al. 1991 Nucleocytoplasmic shuttling of the progesterone receptor. EMBO J 10:3851–3859. 54. Maruvada P, Baumann CT, Hager GL, Yen PM 2003 Dynamic shuttling and intranuclear mobility of nuclear hormone receptors. J Biol Chem 278:12425–12432.

378 55. Liu J, DeFranco DB 2000 Protracted nuclear export of glucocorticoid receptor limits its turnover and does not require the exportin 1/CRM1-directed nuclear export pathway. Mol Endocrinol 14:40–51. 56. Saporita AJ, Zhang Q, Navai N et al. 2003 Identification and characterization of a ligand-regulated nuclear export signal in androgen receptor. J Biol Chem. 57. Katagiri Y, Takeda K, Yu ZX et al. 2001 Modulation of retinoid signalling through NGF-induced nuclear export of NGFI-B. Nat Cell Biol 2:435–440. 58. Zilliacus J, Holter E, Wakui H et al. 2001 Regulation of glucocorticoid receptor activity by 14-3-3-dependent intracellular relocalization of the corepressor rip140. Mol Endocrinol 15:501–511. 59. Masuyama H, Jefcoat SCJ, MacDonald PN 1997 The N-terminal domain of transcription factor IIB is required for direct interaction with the vitamin D receptor and participates in vitamin D–mediated transcription. Mol Endocrinol 11:218–228. 60. Wiebel FF, Steffensen KR, Treuter E, Feltkamp D, Gustafsson JA 1999 Ligand-independent coregulator recruitment by the triply activatable OR1/retinoid X receptor–alpha nuclear receptor heterodimer [In Process Citation]. Mol Endocrinol 13:1105–1118. 61. Stenoien DL, Mancini MG, Patel K et al. 2000 Subnuclear trafficking of estrogen receptor-alpha and steroid receptor coactivator-1. Mol Endocrinol 14:518–534. 62. Becker M, Baumann C, John S et al. 2002 Dynamic behavior of transcription factors on a natural promoter in living cells. EMBO Rep 3:1188–1194. 63. Xiao Z, Brownawell AM, Macara IG, Lodish HF 2003 A novel nuclear export signal in Smad1 is essential for its signaling activity. J Biol Chem 278:34245–34252. 64. Black BE, Holaska JM, Rastinejad F, Paschal BM 2001 DNA binding domains in diverse nuclear receptors function as nuclear export signals. Curr Biol 11:1749–1758. 65. Hsieh JC, Shimizu Y, Minoshima S et al. 1998 Novel nuclear localization signal between the two DNA-binding zinc fingers in the human vitamin D receptor. J Cell Biochem 70: 94–109. 66. Prufer K, Racz A, Lin GC, Barsony J 2000 Dimerization with retinoid X receptors promotes nuclear localization and subnuclear targeting of vitamin D receptors. J Biol Chem 275:41114–41123. 67. Yu Z, Lee CH, Chinpaisal C, Wei LN 1998 A constitutive nuclear localization signal from the second zinc-finger of orphan nuclear receptor TR2. J Endocrinol 159:53–60. 68. Luo Z, Rouvinen J, Maenpaa PH 1994 A peptide C-terminal to the second Zn finger of human vitamin D receptor is able to specify nuclear localization. Eur J Biochem 223:381–387. 69. Swamy N, Mohr SC, Xu W, Ray R 1999 Vitamin D receptor interacts with DnaK/heat shock protein 70: Identification of DnaK interaction site on vitamin D receptor. Arch Biochem Biophys 363:219–226. 70. Xiao Z, Brownawell AM, Macara IG, Lodish HF 2003 A novel nuclear export signal in Smad1 is essential for its signaling activity. J Biol Chem 278:34245–34252. 71. Birbach A, Gold P, Binder BR et al. 2002 Signaling molecules of the NF-kappa B pathway shuttle constitutively between cytoplasm and nucleus. J Biol Chem 277:10842–10851. 72. Aplin AE, Hogan BP, Tomeu J, Juliano RL 2002 Cell adhesion differentially regulates the nucleocytoplasmic distribution of active MAP kinases. J Cell Sci 115:2781–2790.

JULIA BARSONY

73. Meyer T, Marg A, Lemke P, Wiesner B, Vinkemeier U 2003 DNA binding controls inactivation and nuclear accumulation of the transcription factor Stat1. Genes Dev 17:1992–2005. 74 O’Keefe K, Li H, Zhang Y 2003 Nucleocytoplasmic shuttling of p53 is essential for MDM2-mediated cytoplasmic degradation but not ubiquitination. Mol Cell Biol 23:6396–6405. 75. Wadhwa R, Yaguchi T, Hasan MK et al. 2002 Hsp70 family member, mot-2/mthsp70/GRP75, binds to the cytoplasmic sequestration domain of the p53 protein. Exp Cell Res 274:246–253. 76. Prufer K, Racz A, Lin GC, Barsony J 2000 Dimerization with retinoid X receptors promotes nuclear localization and subnuclear targeting of vitamin D receptors. J Biol Chem 275:41114–41123. 77. Antonsson A, Hughes K, Edin S, Grundstrom T 2003 Regulation of c-Rel nuclear localization by binding of Ca2+/calmodulin. Mol Cell Biol 23:1418–1427. 78. Trevillyan JM, Chiou XG, Chen YW et al. 2001 Potent inhibition of NFAT activation and T cell cytokine production by novel low molecular weight pyrazole compounds. J Biol Chem 276:48118–48126. 79. Barsony J, Carroll J, McKoy W, Renyi I, Gould DL, Htun H et al. 1997 Intracellular Traffic of Glucocorticoid Receptors: Studies with Green Fluorescent Protein Chimeras in Living Cells. Springer-Verlag, New York. 80. Pratt WB, Dalman FC, Meshinchi S, Scherrer LC 1990 The relationship between glucocorticoid receptor binding to Hsp90 and receptor function. Nippon Naibunpi Gakkai Zasshi 66: 1185–1197. 81. Barsony J, McKoy W 1992 Molybdate increases intracellular 3′,5′-guanosine cyclic monophosphate and stabilizes vitamin D receptor association with tubulin-containing filaments. J Biol Chem 267:24457–24465. 82. Lian JB, Stein JL, Stein GS et al. 2001 Contributions of nuclear architecture and chromatin to vitamin D–dependent transcriptional control of the rat osteocalcin gene. Steroids 66:159–170. 83. Zhang C, Dowd DR, Staal A et al. 2003 Nuclear coactivator-62 kDa/Ski-interacting protein is a nuclear matrix– associated coactivator that may couple vitamin D receptor–mediated transcription and RNA splicing. J Biol Chem 278:35325–35336. 84. Kino T, Souvatzoglou E, De Martino MU et al. 2003 Protein 14-3-3sigma interacts with and favors cytoplasmic subcellular localization of the glucocorticoid receptor, acting as a negative regulator of the glucocorticoid signaling pathway. J Biol Chem 278:25651–25656. 85. Walther RF, Lamprecht C, Ridsdale A et al. 2003 Nuclear export of the glucocorticoid receptor is accelerated by cell fusion-dependent release of calreticulin. J Biol Chem 278: 37858–37864. 86. Prufer K, Schroder C, Racz A, Barsony J 2000 Cell Cycle Dependence of Vitamin D Receptor Expression. Vitamin D Workshop, Riverside, CA. 87. Amazit L, Alj Y, Tyagi RK et al. 2003 Subcellular localization and mechanisms of nucleocytoplasmic trafficking of steroid receptor coactivator-1. J Biol Chem 278: 32195–32203. 88. Hong SH, Privalsky ML 2000 The SMRT corepressor is regulated by a MEK-1 kinase pathway: inhibition of corepressor function is associated with SMRT phosphorylation and nuclear export. Mol Cell Biol 20:6612–6625.

CHAPTER 22 VDR and RXR Subcellular Trafficking

89. Wang C, Fu M, Angeletti RH et al. 2001 Direct acetylation of the estrogen receptor alpha hinge region by p300 regulates transactivation and hormone sensitivity. J Biol Chem 276:18375–18383. 90. Fu M, Wang C, Zhang X, Pestell R 2003 Nuclear receptor modifications and endocrine cell proliferation. J Steroid Biochem Mol Biol 85:133–138. 91. Nicoll JB, Gwinn BL, Iwig JS et al. 2003 Compartmentspecific phosphorylation of rat thyroid hormone receptor alpha1 regulates nuclear localization and retention. Mol Cell Endocrinol 205:65–77. 92. Hager GL, Smith CL, Fragoso G et al. 1998 Intranuclear trafficking and gene targeting by members of the steroid/nuclear receptor superfamily. J Steroid Biochem Mol Biol 65:125–132. 93. Stenoien DL, Patel K, Mancini MG et al. 2001 FRAP reveals that mobility of oestrogen receptor-alpha is ligand- and proteasome-dependent. Nat Cell Biol 3:15–23.

379 94. Akiyama TE, Baumann CT, Sakai S, Hager GL, Gonzalez FJ 2002 Selective intranuclear redistribution of PPAR isoforms by RXR alpha. Mol Endocrinol 16:707–721. 95. McNally JG, Muller WG, Walker D, Wolford R, Hager GL 2000 The glucocorticoid receptor: rapid exchange with regulatory sites in living cells. Science 287:1262–1265. 96. Maruvada P, Baumann CT, Hager GL, Yen PM 2003 Dynamic shuttling and intranuclear mobility of nuclear hormone receptors. J Biol Chem 278:12425–12432. 97. Barsony J, Prufer K 2002 Vitamin D receptor and retinoid X receptor interactions in motion. Vitam Horm Adv Res Appl 65:345–376. 98. Yamamoto H, Shevde NK, Warrier A et al. 2003 2-Methylene19-nor-(20S)-1,25-dihydroxyvitamin D3 potently stimulates gene-specific DNA binding of the vitamin D receptor in osteoblasts. J Biol Chem 278:31756–31765.

CHAPTER 23

1α,25(OH)2-Vitamin D3–Mediated Rapid and Genomic Responses Are Dependent upon Critical Structure–Function Relationships for Both the Ligand and Receptor(s) ANTHONY W. NORMAN

Department of Biochemistry and Division of Biomedical Sciences, University of California, Riverside, California

I. Introduction to Vitamin D3 and 1α,25(OH)2D3 II. Vitamin D Endocrine System III. Proteins with Ligand Binding Domains for 1α,25(OH)2D3

IV. Structure–Function Evaluation of Selected Rapid Responses Mediated by 1α,25(OH)2D3 V. Summary VI. Addendum References

I. INTRODUCTION TO VITAMIN D3 AND 1α,25(OH)2D3

certainty whether any or all of them occur via signal transduction pathways that do not involve immediate interaction of 1α,25(OH)2D3 with its nuclear receptor.

The secosteroid 1α,25(OH)2-vitamin D3 [1α,25(OH)2D3] initiates biological responses via both regulation of gene transcription as well as via rapid response pathways some of which include opening of voltage-gated Ca2+ and Cl− channels and involvement with protein kinase C (PKC) [1]. The general objective of this chapter is to provide an introduction to rapid or nongenomic responses mediated by 1α,25(OH)2D3. The principal focus will be to compare structure– function relationships for ligands for rapid responses, genomic responses mediated by the nuclear vitamin D receptor (VDR) with the plasma vitamin D binding protein (DBP). There is a problem of definition: what is implied by the term “rapid responses”? For the purposes of this chapter, “rapid responses” is used to describe biological responses to 1α,25(OH)2D3 that appear to occur too rapidly to be simply explained via interaction of the ligand with the vitamin D receptor (VDR) present in the cell nucleus. Frequently these responses are generated within 1–2 min to 15–45 min to, in some cases, 1–4 hr. However, since there is not yet available a detailed molecular mechanism(s) to describe all the observed rapid responses to 1α,25(OH)2D3, it is not known with VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

A. Structure of 1α,25(OH)2D3 1α,25(OH)2D3 is one of 37 known metabolites of vitamin D3 [2,3]. The structure of 1α,25(OH)2D3 is presented in Fig. 1A. The structure and chemistry of vitamin D compounds are related to those of the classical steroid hormones, i.e., the glucocorticoids, estrogens, androgens, progestins, and mineralocorticoids. However, there are three structural aspects of the vitamin D compounds that make them distinct from classic steroids: (a) The vitamin D3 compounds have a side chain comprising eight carbons, whereas the classical steroids have no side chain or only a two-carbon side chain; (b) the 9,10 carbon–carbon bond of the β-ring is broken in this family of related compounds and, accordingly, they are designated as being secosteroids; the term “seco” designates that one of the rings of the cyclopentanoperhydrophenanthrene ring structure of classical steroids has been broken; and (c) the vitamin D3 family of molecules is unusually conformationally flexible in contradistinction to the classic steroids (see later discussion). Copyright © 2005, Elsevier, Inc. All rights reserved.

382

ANTHONY W. NORMAN

A

22

21

24

18 17 12 11 13 14 9

5

4

16 15

8H 7

6

Side chain

25 OH

Rotation, around B-ring 6,7 bond

19 10

3 HO

B

A-ring chair–chair interconversion

OH

2

TOP VIEW (as shown in structure)

C

IN-PLANE VIEW

Side-chain Side-chain H H

HO

HO

H

H H

HO α-chair

D

21 18 12 11 13 14 9

6 4 3

510

8H 7

H

A-ring conformations (chair–chair inversion)

OH β-chair

22

24 26 20 23 25 OH 17 27 16

OH HO

15

19

H

A

19 HO

D

C 9

Fast

7 6

HO

OH 2 6-s-trans confirmation

6-s-cis confirmation

FIGURE 1 Structural aspects of 1α,25(OH)2D3 which contribute to its conformational flexibility. (A) Structure of 1α,25(OH)2D3 and an indication of the location of the three structural aspects of vitamin D secosteroids that contribute to the conformational flexibility of these molecules. (B) The dynamic single bond rotation of the cholesterol-like side chain of the hormone shown on the center panel (i.e., rotations about the six single bonds are indicated by the curved arrows). Energy minimization calculations to define the most stable orientation of the side chain indicate that there are 394 conformers within 4 kcal/mol of the global minimum (lowest energy state). The dots indicate the position in threedimensional space of the 25-hydroxyl group for the 394 minima; collectively these dots define the volume in space that the side chain occupies. The left panel illustrates a “top view” looking down on the C/D rings while the right panel presents an edge-on view of the C/D rings. A discussion of the consequences of the side-chain conformational mobility has been previously presented [4,5,116]. (C) The cyclohexane-like A-ring is free to interchange rapidly between a pair of chair-chair conformers effectively equilibrating the key 1α- and 3β-hydroxyls between equatorial and axial orientations [6,7]. Thus when the 1α-hydroxyl is axial, the 3β-hydroxyl is equatorial, and vice versa. (D) Rotational freedom about the 6–7 carbon–carbon single bond of the seco B-ring allows conformations ranging from the more steroid-like 6-s-cis conformation to the open and extended 6-s-trans form of the hormone [8].

CHAPTER 23 Function Relationships for Both the Ligand and Receptor(s)

B. Conformational Flexibility

II. VITAMIN D ENDOCRINE SYSTEM

The intact eight-carbon side-chain of vitamin D3 and related secosteroids can easily assume numerous shapes and positions by virtue of rotation about its many carbon–carbon single bonds; see Fig. 1B. The dots indicate the position in three-dimensional space of the 25-hydroxyl group for 394 readily identifiable side-chain conformations. A discussion of the consequences of the side-chain conformational mobility has been previously presented [4,5]. Breakage of the 9,10 carbon–carbon bond releases the A-ring from being held in place to the B-ring and, accordingly, the cyclohexane-like A-ring is free to rapidly interchange (many thousands of times per second) between a pair of chair–chair conformers (Fig. 1C). This has the consequence of changing the orientation of the key 1α- and 3β-hydroxyls between either the equatorial or axial orientation [6,7]. Rotational freedom about the 6–7 carbon–carbon bond of the seco B-ring allows a continuum of conformations ranging from the more steroid-like 6-s-cis conformation to the open and extended 6-s-trans form of the hormone [8]; these conformations are illustrated in Fig. 1D and also Fig. 11. It has been generally assumed for receptor–ligand interactions that the ligand is frozen in a single conformation dictated by both the structural constraints of the ligand and the three-dimensional architecture of the peptide chains that create the ligand binding domain (LBD) of the receptor(s). The ligands for the RAR and TR, as for the VDR, are all conformationally flexible, and the X-ray crystallographic structure for each receptor indicated that only one definitive conformer was present in their ligand binding domain [9,10]. This clearly demonstrates that the capability of steroid receptors to capture one ligand conformation from a large population of available flexible conformers is a general feature of several proteins. Certainly the most important polar functional group of 1α,25(OH)2D3 is the 1α-hydroxyl; it is this functional group that makes the molecule distinct from 25(OH)D3, which is the major circulating form of vitamin D3 present in the blood. 1α,25(OH)2D3 binds some 666-fold more tightly to the nuclear receptor than 25(OH)D3 [1]. Also, the orientation of the hydroxyl on carbon-1 is important to the nuclear receptor; thus the nuclear receptor binds 1α,25(OH)2D3 3000-fold better than 1β,25(OH)2D3 [1]. Thus there is no question that receptors for 1α,25(OH)2D3 must be able to distinguish both the presence or absence of this hydroxyl and its orientation (1α versus 1β) as well as to capture it to form a stable receptor–ligand complex.

A. General Description

383

The concept of the existence of the vitamin D endocrine system is now firmly established (see Fig. 2) [1,12,13]. This endocrine system includes four key components. First, the kidney functions as an endocrine gland to produce in a physiologically regulated manner the two secosteroid hormones 1α,25(OH)2D3 and 24R,25-dihydroxyvitamin D3 [24R,25(OH)2D3]. Second, a feedback loop regulates the secretion of parathyroid hormone (PTH) in response to the serum ionized Ca2+ level. A prime function of PTH is to interact in the proximal tubule of the kidney to increase or (in the face of falling PTH levels) decrease the activity of the 25(OH)D3-1α-hydroxylase, which, in turn, determines the output of 1α,25(OH)2D3 available to generate biological responses. A third key component in the operation of the vitamin D endocrine system is the plasma vitamin D binding protein (DBP), which carries vitamin D3 and all its metabolites to their various target organs. The DBP is known to have a specific ligand binding domain for vitamin D-related ligands that is different in specificity from the ligand binding domain of the nuclear vitamin D receptor [11]. Fourth and most importantly, target organs for 1α,25(OH)2D3 possess receptors for the hormones that, when occupied with ligand, initiate the signal transduction pathways leading to the appearance of biological responses. Many of these topics are reviewed in detail in other chapters of this book. Understanding how 1α,25(OH)2D3 can produce biological responses via regulation of gene transcription [14–16] as well as via rapid membrane-initiated pathways [17–19], some of which involve opening of voltage-gated Ca2+ channels [20] or protein kinase C (PKC) [21] (see Fig. 3), is one of the areas of current intensive research investigation. The regulation of gene transcription by 1α,25(OH)2D3 is known to be mediated by interaction of this ligand with a nuclear receptor protein, termed the VDRnuc, while the membrane-initiated responses are postulated to be mediated through interaction of the 1α,25(OH)2D3 with a protein receptor located on the external surface of the cell; this receptor is referred to as the VDRmem. There is also emerging evidence that the plasma membrane binding protein [VDRmem] may be associated with caveolae. Caveolae are flask-shaped invaginations on the plasma membrane that serve as platforms for accumulation of signal transduction proteins; the following citations will provide an entrée into this large literature [22–24].

384

ANTHONY W. NORMAN

VITAMIN D3

LIVER

Kidney as an endocrine gland 25(OH)2D3

25(OH) D3

24-HYDROXYLASE (+) 24R,25(OH)2D3 PLACENTAL PRODUCTION 1α,25(OH)2D3

(+)

(−) 1-HYDROXYLASE (+) 1α,25(OH)2D3 H

PARACRINE PRODUCTION OF 1α,25(OH)2D3

H

HO

OH H

OH

OH

H

OH

H

24R,25(OH)2D3

PARATHYROID HORMONE

(−) (−) SHORT LONG FEEDBACK FEEDBACK LOOP LOOP

HO

1α,25(OH)2D3

24R,25(OH)2D3

BLOOD BLOOD 1α,25(OH)2D3

FETUS Development

CELLULAR GROWTH & DEFFERENTION

RAPID ACTIONS

1α,25(OH)2D3 NUCLEAR RECEPTORS

BLOOD 24R,25(OH)2D3

CLASSIC TARGET ORGANS 1α,25(OH)2D3 or ANALOGS

Pi Ca2+

24R,25(OH)2D3 RECEPTORS

BONE INTESTINE KIDNEY

REABSORPTION OF Ca2+ & Pi ABSORPTION OF Ca2+ SELECTED BIOLOGICAL RESPONSES

MOBILIZATION / ACCRETION OF Ca2+ & Pi

FIGURE 2 Summary of the vitamin D endocrine system. The parent vitamin D3 is first converted in the liver to 25(OH)D3 and then in the kidney to one of two hormones, 1α,25(OH)2D3 or 24R,25(OH)2D3. 1α,25(OH)2D3 produces biological responses in its target organs via either interacting with a nuclear receptor (VDRnuc) or with a system which generates rapid responses and is believed to involve a membrane receptor (VDRmem) [1]. The precise biological role(s) of 24R,25(OH)2D3 are not yet defined; however, several studies suggest the presence of receptors in chondrocytes [59,117,118] and bone [61,119].

B. Tissues with the Nuclear VDR Table I summarizes the target organs which are known to possess the VDRnuc. The nuclear responses to 1α,25(OH)2D3 are generated in a manner homologous to that of classical steroid hormones, i.e., cortisol, progesterone, estradiol, testosterone, and aldosterone. In the general model, the hormone is produced in an endocrine gland in response to a physiological stimulus, then circulates in the blood bound to a protein carrier (the vitamin D binding protein or DBP), which delivers it to target tissues where the hormone enters the cell and interacts with a specific, high-affinity intracellular receptor(s). The receptor–hormone complex then localizes in the nucleus, undergoes some type of activation perhaps involving phosphorylation [25–29], and binds to a hormone response element (HRE) on the DNA to modulate the expression of hormone-sensitive genes. The modulation of gene transcription results in either the induction or the repression of specific mRNAs, ultimately resulting in changes in protein

expression needed to produce the required biological response [15,30,31]. High-affinity receptors for 1α,25(OH)2D3 have been identified in at least 35 target tissues [14,32,33], and more than 100 genes are known to be regulated by 1α,25(OH)2D3 [15]. The mechanism of the regulation of these genes may be direct, as described earlier, through HREs, or indirect. See the following reviews for further information [16,34,35]. Since the discovery of 1α,25(OH)2D3 in 1968 [36] and its VDRnuc in 1969 [37], research has focused on defining how these agents collaborate to generate biological responses; of course the principal focus was on understanding regulation of gene transcription. Table II summarizes the new biological actions of 1α,25(OH)2D3 that are known to be mediated by the VDRnuc. In many instances, the first hint of the biological process involved was derived from discovery and then biochemical description and characterization of the VDR in a new location. For example (see Table II), the observation that the VDR was present in the pancreas [38,39] and more particularly the β cells [40]

385

CHAPTER 23 Function Relationships for Both the Ligand and Receptor(s)

Natural steroid hormone

Conformationally flexible analogs

Conformationally restricted analogs

TABLE I Tissue Distribution of the Nuclear Receptor (VDR) for 1α,25(OH)2D3 Target cell

Nucleus

Receptors

Plasma membrane

Ligand-receptor complex

MAP kinase activation

DNA

cAMP PKC Ca2+ channel CI−channel (open)

“Cross talk”

Gene expression Up regulation

mRNA

Down regulation

mRNA

(minutes-hours-days)

(seconds-minutes)

BIOLOGICAL RESPONSES

FIGURE 3 Pathways for generation of biological responses by 1α,25(OH)2D3. In the genomic pathway, occupancy of the nuclear receptor for 1α,25(OH)2D3 (VDRnuc) by a ligand leads to an up- or down-regulation of genes subject to hormone regulation. In the membrane-initiated pathway, occupancy of a membrane receptor for 1α,25(OH)2D3 by a ligand can lead to activation of protein kinase C (PKC), adenylate cyclase, phospholipase C, and/or opening of voltage-gated Ca2+ or chloride (Cl−) channel openings, which are coupled to generation of the end biological response(s). Also, this figure emphasizes that rapid responses can engage in cross-talk to modulate genomic responses.

suggested that it was responsible for the production of calbindin. Another discovery was by Tsoukas et al. [41], who reported that B and T lymphocytes, when activated, both expressed a VDRnuc. This observation is the foundation of a major new field of research concerning the role of vitamin D on the immune system [42,43].

C. Rapid Responses Starting in approximately 1983, a series of observations were reported which suggested that some of the biological responses generated by 1α,25(OH)2D3 occurred too rapidly to be simply explained via genomic events.

Adipose Bone Brain Cancer cells (many) Colon Ganglion Hair follicle Kidney Lung Muscle (smooth) Osteoclast Pancreas β cell Parotid Placenta Skin Testis Thyroid Yolk sac

Reference a

Target cell

Reference a

[123] [125,126] [128–130] [131,133–138]

Adrenal Bone marrow Breast Cartilage

[124] [127] [131,132] [139–141]

[133,142] [145] [147] [38] [153,154] [108,157,158]

Eggshell gland Epididymis Intestine Liver Muscle (cardiac) Osteoblast

[143,144] [146] [148,149] [150–152] [155,156] [159,160]

[161,162] [38,39]

Ovary Parathyroid

[163,164] [165–167]

[168] [169] [175–179] [180,181] [184] [185]

Pituitary Prostate Stomach Thymus Uterus

[169–171] [172–174] [170] [182,183] [134,146]

aIn the interest of economy, only selected reference citations are included. bAn alternative approach to the biochemical characterization of receptors in a proposed target organ for 1α,25(OH)2D3 is the technique of autoradiography using high specific activity [3H]1,25(OH)2D3; W. D. Stumpf has published many papers proposing the presence of specific receptors for 1α,25(OH)2D3 based on his autoradiographic studies [170,186].

Accordingly, there is now active research on how 1α,25(OH)2D3 generates biological responses via both genomic and rapid pathways; a model is presented in Fig. 3. Table III presents a tabulation of the rapid responses mediated by 1α,25(OH)2D3. In comparison to our understanding of the interaction of 1α,25(OH)2D3 with its nuclear VDR and the plethora of details concerning regulation of gene transcription, it is clear at the time of preparation of this chapter that the field of nongenomic responses is still in its descriptive phase. As a consequence of many of these studies, it was originally proposed, based on this laboratory’s studies on transcaltachia, that some of the actions of 1α,25(OH)2D3 may be mediated at the cell membrane or by extranuclear subcellular components

386 TABLE II

ANTHONY W. NORMAN

New Biological Actions of 1α,25(OH)2D3

Target cells or organ actions related to VDRnuc Adipose Bone marrow Cancer cells

Cells with a VDRmem Immune system Intestine Leukemia NB4 cells

Osteoblasts Pancreas Pancreas B cells Vascular smooth muscle

Biological process Adipocyte metabolism Monocyte differentiation to macrophages Antiproliferation, apoptosis, cell differentiation Rapid responses Immunosuppression Transcaltachia Activation of MAP-kinase; cell differentiation Opening of Ca2+ and Cl− channels Insulin secretion Insulin secretion Cell migration; activation of phosphatidyl inositol-3′-kinase

TABLE III Organ/cell/system Cartilage cells

Colon Cross-talk between VDR and phosphatase Endothelial cells: skeletal muscle cells, leukemia cells Fibroblasts Intestine

Keratinocytes Kidney Lipid bilayer Liver Muscle Osteoblast

Parathyroid cells Promyelocytic leukemic cells

References [187] [188,189] [190–193]

See Table III [41,194] [44,88] [94,195]

[20,96] [40,196] [120,197] [121]

near the plasma membrane [44–46]. A candidate membrane receptor (VDRmem) for 1α,25(OH)2D3 has been identified [47]; see later discussion. An alternative possibility is that the nuclear vitamin D receptor (VDRnuc) may, under some circumstances, be associated with plasma membrane acceptor sites [48] or with caveolae in the plasma membrane [49]. As a point of comparison between 1α,25(OH)2D3 and the other steroid hormones, it is well known that they, too (estradiol, progesterone, testosterone, aldosterone, and cortisol), can all generate a wide array of rapid responses; the following references provide an entrée into this large literature [50–52]. As with 1α,25(OH)2D3, it is not clear for the other steroid hormones whether they all utilize novel membrane receptors or nuclear receptors associated with plasma membrane caveolae that are linked to the generation of rapid responses (or both). There is strong experimental evidence for both a progesterone G–protein coupled receptor [53,54] and a nuclear estrogen receptor associated with caveolae in endothelial cells that is linked to rapid responses mediated by nitric oxide (NO) [55–58]. Both of these physiological settings could be a paradigm for 1α,25(OH)2D3-mediated rapid responses.

Distribution of Rapid Responses to 1α,25(OH)2D3 Response studied

Reference

Activation of PLCβ Regulation of matrix vesicle PKC Arachidonic acid is an autocoid mediator of growth plate cells PKC effects Activation of Ser/Thr phosphatase P79 S6 kinase via interaction with catalytic subunit Activation of PI3-kinase

[198] [199–201] [202] [203] [204]

Accumulation of cGMP near VDRnuc Rapid transport of intestinal Ca2+ (transcaltachia); CaCo-2 cells, PKC, G proteins Activation of PKC Alter PKC subcellular distribution PKC effects Activation of highly purified PKC Increased intracellular Ca2+ and activation of PKC and MAP kinase PKC and Ca2+ effects ROS 17/2.8 cells, Ca2+ channel opening Cl− channel opening, UMR-106 cells Ca2+ channel opening by 24R,25(OH)2D3 Phospholipid metabolism Cytosolic Ca2+ Aspects of cell differentiation PKC effects

[207] [44,208,209] [210] [211] [212] [213] [214] [215,216] [217] [20] [218] [203] [219] [220] [221,222] [21,223]

[121,205,206]

387

CHAPTER 23 Function Relationships for Both the Ligand and Receptor(s)

Thus, each possesses a unique ligand binding domain which specifically recognizes secosteroids. More specifically, the ligand with highest affinity to DBP is 25(OH)D3. The physiological responsibility of DBP is to transport vitamin D and all its daughter metabolites through the blood compartment delivering appropriate metabolites to appropriate target organs. In contrast, the VDR has a more specific ligand binding domain in that it exclusively focuses on binding 1α,25(OH)2D3 (666-fold more tightly than 25(OH)D3).

III. PROTEINS WITH LIGAND BINDING DOMAINS FOR 1α,25(OH)2D3 A. Introduction There are only a limited number of proteins that have ligand binding domains for vitamin D3 secosteroids; these include the plasma transport protein or vitamin D– binding protein (DBP), the various cytochrome P450 hydroxylases that mediate the myriad of metabolic transformations of vitamin D3 and its metabolites, and receptors for 1α,25(OH)2D3, including the classic nuclear receptor (VDRnuc) and membrane receptors (VDRmem). There is also a report from Seo et al. [59] describing a putative membrane receptor for 24R,25(OH)2D3 [60,61]. An ongoing project in the laboratory of the author is to determine the shape of the vitamin D3 metabolite ligand(s) for all these classes of proteins. Since these proteins are not believed to be structurally related, it is not anticipated that there necessarily will be structural homology of the various ligand binding domains. In this chapter we will first detail structural comparison of the VDRnuc with the DBP. Figure 4 presents a comparison of the domains of the VDRnuc with the domains of DBP. The details are provided in the figure legend. It is to be emphasized that there is not any amino acid sequence homology between these two proteins.

B. Vitamin D Receptor (VDRnuc) Based on evaluation of the human genome database, it is known that there are a total of 48 members of the nuclear receptor superfamily [62]. In this superfamily, 28 receptors have an LBD which binds a small ligand molecule; of these 28 receptors, 8 are classical hormone receptors and 20 are orphan receptors [62]. The VDR belongs to the subfamily of nuclear receptors with an LBD which binds a classical hormone; this includes as ligands the glucocorticoids (GR), progestins (PR), estrogens (ER), aldosterone (MR for mineral corticoid), androgens (AR), thyroid hormone (TR), the hormonal forms of vitamin A (RAR), and the VDR [63,64].

A VDR

HINGE

DNA BINDING

H2N

1

24

A/B

89 118 C

“LOOP”

165

1α,25(OH)2D3 BINDING

215

COOH

404 427

D

E

F

B DBP

25(OH)D3 BINDING

H2N 1

ACTIN BINDING DOMAIN

110 I

192

COOH

368 II

458 III

FIGURE 4 Schematic models of the VDR and DBP. (A) The VDR comprises 427 amino acid residues that are divided into six domains (A–F). The numbers below the VDR indicate the amino acid residue boundaries for the various domains. The VDR belongs to a superfamily of nuclear receptors that have the same general A–F domain organization. The C domain, the most highly conserved, which contains the DNA-binding domain defines the superfamily; it contains two zinc finger motifs. The E domain or ligand binding domain (LBD) is less conserved and is responsible for binding 1α,25(OH)2D3 or its analogs and transcriptional activation. The A/B domain of the VDR is much smaller than other members of the superfamily. The portion of the intact VDR that was crystallized and subjected to X-ray crystallographic analysis included residues 118–427, but with deletion of the loop region of the hinge domain D, specifically residues 165–215 (see text). (B) DBP consists of 458 amino acid residues and is divided into three domains (I, II, and III). The numbers below the DBP indicate the amino acid residue boundaries for the various domains. Domains I, II, and III have been postulated to have evolved from a progenitor that arose from the triple repeat of a 192 amino acid sequence [78]. However, domain III is significantly truncated at the C terminus. The 25(OH)D3 binding cleft is associated with the first six α-helices or residues 1–110 of domain I. The actin binding property of DBP is associated with a portion of domains I and III, which clamp the actin while it rests on domain II.

388 A dramatic advance in understanding of the threedimensional structure of the LBD of steroid receptors has occurred over the past 7 years with the X-ray crystallographic structure determination of the LBD of five steroid hormone–related receptors. These include the LBDs of the TR, RAR, ER, PR, GR, MR, PPARγ (peroxisome proliferator activated receptor), and the closely related TR and RAR; this work has been reviewed by Weatherman et al. [65]. They all have a common structural organization consisting of six domains [65,66] with significant amino acid sequence homologies across all the domains (see Fig. 4A). The different domains act as distinct functional modules that can function independently of each other [67]; see Fig. 3A. Also, an X-ray structure is available for the LBD of the unoccupied 9-cis-retinoic acid receptor RXR [68]. Further, the ER LBD X-ray structures are known for a ligand (raloxifene) that can act as an antagonist of the transcriptional activation function of the nuclear receptor. In 2000, the crystal structure of a modified version of the ligand binding domain of the nuclear receptor for vitamin D, bound to 1α,25(OH)2D3, was determined at a 1.8-Å resolution [69]. A follow-up X-ray crystallographic report compared the VDR LBD and bound ligand for 1α,25(OH)2D3 with that of four superagonist analogs of 1α,25(OH)2D3 [70]. The structure of the LBD of the human VDRnuc spans amino acid residues 143–427, [COOH terminus] but without residues 165–215, which were in an undefined loop in the hinge region of domain D (see Fig. 4A). In comparison to other nuclear receptors, the length of the VDR loop region is significantly longer. The removal of the flexible insertion domain in the VDR LBD produced a more soluble protein that was more amenable to crystallization. The VDR LBD protein structure is very similar to the LBD of the five other X-ray crystallographic determined nuclear hormone receptor structures [65]. The VDR LBD structure consists of a three-stranded β-sheet and 12 α-helices that are arranged to create a three-layer sandwich that completely encompasses the ligand 1α,25(OH)2D3 in a hydrophobic core (see Fig. 6A,C). Comparison of panels A, B, and C of Fig. 5 strikingly illustrates the organizational structure of the VDR ligand binding domain. Panel A shows the conventional ribbon structure; VDR comprises 12 helical ribbons that collectively create a three-layered sandwich structure. However, when one converts the ribbon structure to a Corey–Pauling representation of all of the atoms associated with the VDR (Fig. 5B), it can be seen that no portion of the ligand [hormone 1α,25(OH)2D3] is exposed to the surface environment. Said differently, the 1α,25(OH)2D3 is buried in the interior of the ligand binding domain, as illustrated

ANTHONY W. NORMAN

in Fig.5C. The dominant structural features of the ligand molecule, the 1α-, 3β-, and 25-hydroxyl groups are all interacting in a steroid-specific manner with various hydrogen bonding partners provided by the R-groups of the appropriate amino acids of the primary amino acid sequence of the VDR. The functional groups of the amino acids comprising the interior surface of the VDR ligand cavity interact very selectively with the ligand 1α,25(OH)2D3 to form a stable ligand–LBD complex (Fig. 5C). 1α,25(OH)2D3 is positioned in the LBD so that the A-ring enters the binding cavity first and then moves to the deep interior of the receptor so that the trailing side chain and the 25-hydroxyl group are just inside the LBD, which is closed as a consequence of helix 12 moving from its open position (characteristic of the unoccupied VDR LBD). The 1α-hydroxyl participates in two hydrogen bonds with Ser237 (helix-3) and Arg274 (helix-5). The 3β-hydroxyl group participates in two hydrogen bonds with Ser278 (helix-5) and Tyr143 (helix-2). The 25-hydroxyl group is hydrogen bonded to both His305 (loop helices 6/7) and His397 (helix-11). The C6-C7 single bond exhibits a trans conformation that deviates by 30° from the planar extended form of the ligand (see Fig. 12). With respect to conformational flexibility of the vitamin D secosteroids, it is generally accepted that the molecular mobility is displayed in organic solvents as well as in an aqueous environment similar to that encountered in biological systems. It is generally assumed for receptor–ligand interactions that the ligand is frozen out in a single shape or conformation that is dictated both by the structural constraints of the ligand and by the threedimensional architecture of the peptide chains that create the ligand binding domain of the receptor.

C. Vitamin D Binding Protein (DBP) The vitamin D–binding protein (DBP), also known as group-specific component (GC-globulin) [71,72], is the serum protein that serves as the transporter and reservoir for the principal vitamin D metabolites throughout the vitamin D endocrine system [73,74]. These include 25(OH)D3, the major circulating metabolite [KD ~ 6 × 10−9 M] [75], and the steroid hormone 1α,25(OH)2D3 [KD ~ 6 × 10−8 M]. DBP can be up to 5% glycosylated and is known to be one of the most polymorphic proteins, with three common allelic variants and more than 124 rare variants known [76]. DBP’s plasma concentration (4–8 µM) is approximately 20-fold higher than that of the total circulating vitamin D metabolites (~10−7 M). DBP binds 88% of the total serum 25(OH)D3 and 85% of serum

389

CHAPTER 23 Function Relationships for Both the Ligand and Receptor(s)

A

B

C

H1

H8 H10/11 H5

H3

1,25D H7 1,25D H12 H12

FIGURE 5 Three-dimensional structure of the VDR ligand binding domain for 118–425 (∆165–215 is the loop deletion) with its bound ligand 1α,25(OH)2D3 as determined by X-ray crystallography. (A) Ribbon structure of the VDR LBD bound to 1α,25(OH)2D3 as its ligand. Each of the twelve helices [e.g., helix 1 (H1)] can be identified by individual colors (as well as the labels on panel 3A). The colors of the individual helices are as follows: H1, green; H2, gray; H3, light blue; H4, purple; H5, yellow; H6/α-sheet, pink; H7, red; H8, aqua; H9, bronze; H10/H11, gold; H12, black; ∆165–215 (loop deletion), rust. The white regions represent loops and other flexible regions of the receptor molecule. The ligand, 1α,25(OH)2D3 has its atoms colored so that oxygen is red and carbon blue. (B) The VDR LBD is shown as a Corey–Pauling–Koltun (CPK) space-filling model. The position of helix-12 (black) in the closed position [69] effectively sequesters the ligand from the external environment of the protein, indicated by the absence of visible carbon and oxygen atoms from 1α,25(OH)2D3 in this view. (C) A view of the interior of the ligand binding domain of the VDR. This panel represents a “slice” across the middle plain of the view of panel (B). Clearly visible in the interior of the VDR LBD is the agonist ligand 1α,25(OH)2D3. (See color plate.)

1,25(OH)2D3, yet only 5% of the total circulating DBP actually carries vitamin D metabolites [77]. The three-domain structure of DBP is shown in Fig. 6A. Domains I, II, and III have been postulated to have evolved from a progenitor that arose from the triple repeat of a 192 amino acid sequence [78]; however, domain III is significantly truncated at the C-terminus. The position of the vitamin D metabolite and actin binding domains are specified in domains I and portions of domains I, II, and III, respectively. The X-ray crystallographic structures of the human DBP with a bound ligand of either 25(OH)D3 or 22-(m-hydroxyphenyl)-23,24,25,26,27-pentanor vitamin D3 has been determined [79] (see Fig. 6). Both crystals contained two nonidentical DBP molecules present in the asymmetric unit [79]. 25(OH)D3 was absent in the two recent X-ray crystal structures of DBP complexed with an actin monomer [80,81]. A second biological function of DBP is to serve as a scavenger in the circulating system for actin [80–82]. In the DBP: actin complex, the actin is bound in a groove formed

by the three domains, where domain I and III flank the actin monomer as it sits on top of domain II. DBP differs structurally from other serum transport proteins such as SHBG [83], retinol binding protein (RBP) [84], and thyroid hormone binding globules (TBG) [85], which all are β-barrel proteins. Uteroglobulin (UG), a progesterone binding protein, is small (17-kDa, α-helical) and lacks any homology to DBP. The most striking difference between DBP and the proteins in its family as well as other serum binding proteins is the topology of the ligand binding domain. The N-terminal region of DBP, helix 1–helix 6 of domain I, forms the ligand binding domain (LBD) where 25(OH)D3 and other vitamin D metabolites bind. When bound to DBP, vitamin D sterols including 1α,25(OH)2D3 remain highly exposed to the external environment, which is not the case for internally sequestered ligands bound to SHBG, RBP, progesterone binding globins UG, and TBG. Although virtually one whole surface of the 25(OH)D3 molecule (Fig. 6B) is exposed to the

390

ANTHONY W. NORMAN

I

A H2N–

Vitamin D Binding Domain

II

III

–COOH

FIGURE 6 Three-dimensional structure of the vitamin D binding protein (DBP). (A) Amino acid sequence of the 476 amino acids that form the primary amino acid sequence of DBP. This protein consists of three domains that have high sequence homology and are believed to have occurred evolutionarily as a consequence of gene duplication. The bold black bars indicate the presence of 14 disulfide linkages. See CD-ROM for 6A in color. (Continued)

CHAPTER 23 Function Relationships for Both the Ligand and Receptor(s)

B

391

C

FIGURE 6 Cont’d (B) The X-ray structure of DBP. This figure illustrates the three domains (I, yellow; II, light blue; III, green) of the DBP in a ribbon structure representation. The atoms of the ligand, 25(OH)D3, are colored so that oxygen is red and carbon blue. The ligand binding domain of the DBP is a crevice located on the surface of domain I. (C) This illustrates the CPK space-filling structure of DBP, with white regions indicating flexible regions of the molecule. Virtually the entire top surface (β-face) of the 25(OH)D3 (blue and red) is exposed to the external environment. (See color plate for parts B and C).

external environment when bound to the DBP, the protein’s affinity for 25(OH)D3 still remains high (~6 × 10−9 M), presumably because of the relative strength of the protein–ligand interactions on the α-face of 25(OH)D3. When 25(OH)D3 is bound to DBP, the region below the plane of the A, C, and D rings (Fig. 1), with its LBD, 25(OH)D3, in the α-chair (3β-OH equatorial) conformation, lies in the cleft with the α-face and the entire side chain making favorable hydrophobic contacts with the LBD. The DBP/25(OH)D3 complex is further stabilized by hydrogen bonds made by the two hydroxyls of the ligand. The conformation of 25(OH)D3 when bound to the DBP cleft has an overall L-type shape (hook-like) with the exocyclic alkene (C19) 31° below the CD-ring plane (see Fig. 11) and the side chain oriented almost perpendicular to and below the trans-hydrindane (CD-ring) (see Fig. 12) [79]. The distance between C17-O25 in the DBP/25-OH complex is 3.8 Å; thus the 25-hydroxyl group of the ligand is not extended away from the side chain, but rather curled back toward Tyr32 (Fig. 12). Table IV presents the important similarities and differences in the protein structures of the VDR and DBP as well as the shapes of their ligands. It is clear that each binding protein (DBP and VDR) has its own individual specification for generating a stable protein– ligand complex that is appropriate for its role in the biological system.

D. Membrane Receptor for 1α,25(OH)2D3 [VDRmem] The concept of the existence of a membrane receptor for 1α,25(OH)2D3 has its origins in the study of the

process of 1α,25(OH)2D3-mediated response of transcaltachia, or the rapid hormonal stimulation of intestinal Ca2+ absorption in the perfused chick intestine [86]. The most potent agonist is the natural hormone 1α,25(OH)2D3, which can stimulate the transport of calcium across the intestine within 1–2 min [44,87]. The current view of this laboratory and others is that transcaltachia is initiated by 1α,25(OH)2D3 via a membrane-localized receptor, the VDRmem. One hallmark of rapid responses that will be described in detail in a later portion of this chapter is that a different shape of 1α,25(OH)2D3 is optimal for genomic responses (bowl shaped 6-s-trans 1α,25(OH)2D3) as contrasted with the ligand shape for rapid responses with a 6-s-cis-shaped 1α,25(OH)2D3 (see Fig. 12). Thus, 1α,25(OH)2D3, which is conformationally flexible, is the known optimal agonist for both genomic responses and rapid responses. However, a major breakthrough occurred with the chemical synthesis of 1α,25(OH)2D3-lumisterol D3 (JN). This compound is permanently locked in the 6-s-cis position and while it is a full agonist for rapid responses binds only weakly to the VDRnuc [88]. An important verification of the concept of the existence of the proposed VDRmem is the isolation and purification of such a protein. Our laboratory described in 1994 a protein present in the basal lateral membrane of chick intestinal mucosa cells, which displays a 4400-fold enrichment in binding activity for 1α,25(OH)2D3 [47]. We believed at that time that the purified protein represented the VDRmem. Since that time, there have been two additional efforts to further purify and characterize this protein. One involves the work of Farach-Carson and Nemere, who have reported the existence of a membrane associated response system (MARRS) that is present in chick intestinal

392

ANTHONY W. NORMAN

TABLE IV

Comparison of VDR and DBP Crystal Structures

Receptor parameter Mol. wt. (kDa) Number of amino acid residues of intact protein Number of residues in X-ray structure Location of LBD on the protein Ligand–protein contacts Ligand A ring conformation 3β-OH C5–C6–C7–C8 torsion angle Ligand general shape A ring position C17–C20–C22–C23 torsion angle Side-chain orientation Distance from C-17 to oxygen on C-25 Overall ligand shape

VDRnuc 51 427 158 Interior pocket Uniquea β-Chair Axial +211° Bowl-shaped 6-s-trans 30° above C/D ring −156° Extended 6.9 Å Bowl

DBP 58 458 458 Surface cleft Uniquea α-Chair Equatorial +149° Twisted 6-s-trans 31° below C/D ring −70° Curled down 3.8 Å Hook

a The descriptor unique is used to indicate that the amino acid residues of the protein involved with the stabilizing hydrogen bond contact points with the respective ligands, 1α,25(OH)2D3 for VDR and 25(OH)D3 for DBP, are totally different.

cells to be a candidate as a ligand binding protein capable of initiating rapid responses to 1α,25(OH)2D3 and phosphate absorption [19]. Recent work in our laboratory has focused on isolation of the intestinal, kidney, or lung membrane fractions that contain caveolae. Caveolae are flask-shaped membrane invaginations that are enriched in sphingolipids and cholesterol which are commonly found in both caveolae and/or lipid rafts [22–24]. The caveolae enriched membrane fraction (CMF) is isolated from chick or rat intestine, kidney, or lung tissue. We find that under in vitro conditions there is a saturable binding of high-specific-activity tritiated 1α,25(OH)2D3 in the CMF. This binding activity is steroid specific for 1α-hydroxylated analogs and for 1α,25(OH)2D3 has a KD equal to 1.4 nM [49]. Current studies are focusing on the physiological relationship of the CMF 1α,25(OH)2D3 and the generation of physiological responses.

IV. STRUCTURE–FUNCTION EVALUATION OF SELECTED RAPID RESPONSES MEDIATED BY 1α,25(OH)2D3 A. Molecular Tools for Study of 1α,25(OH)2D3 Mediated Rapid Responses Figure 7 presents structures of two families of 1α,25(OH)2D3 analogs that have been found to be useful to study both rapid responses and genomic responses

mediated by both 1α,25(OH)2D3. The natural hormone, 1α,25(OH)2D3, is an agonist of both rapid responses (VDRmem) as well as genomic responses (VDRnuc). In the upper row, the 6-s-cis lumisterol analogs (JN and MZ) are full agonists for rapid responses mediated by 1α,25(OH)2D3. In contrast, the 1β,25(OH)2D3 (HL) is an antagonist of rapid responses mediated by 1α,25(OH)2D3 [89]. In the lower row, analogs MK and ML have been found to be antagonists of the nuclear receptor for 1α,25(OH)2D3 (VDRnuc) blocking gene activation [90,91]. These analogs are structurally related to the vitamin D metabolite 1α,25(OH)2D3-2623-lactone D3 [92,93]. The analogs MK and ML have no agonist actions on rapid responses [94] . Thus, this family of molecular tools has facilitated study of both rapid responses and genomic response systems to identify crucial structure–function relationships. In the following paragraphs, three case studies will be presented that rely heavily on the use of these molecular tools.

B. Rapid Response Case Studies 1. CASE STUDY 1: PATCH CLAMP ANALYSIS OF ROS 17/2.8 CELLS

It has been demonstrated in osteoblasts that the steroid hormone 1α,25(OH)2D3 facilitates the opening of L-type Ca2+ channels as well as prolongation of the open state [20,95]; these responses occur in seconds after the addition of 1α,25(OH)2D3 and represent

393

CHAPTER 23 Function Relationships for Both the Ligand and Receptor(s)

RAPID RESPONSES VDRmem ANTAGONIST HO H

OH

H

OH

OH

H

H

1α,25(OH)2D3

H

OH

OH

H

H

HO

HO

H

OH OH

H

H

HO

1α,25(OH)2-Lumisterol3 JN

HO 1α,24(OH)2-Lumisterol3 MZ

1β,25(OH)2D3 HL

GENOMIC RESPONSES VDRNuc

H

H OH H

H

HO 1α,25(OH)2D3

OH

HO

H O

O O

ANTAGONISTS

O

OH

HO

OH

TEI-9647 MK

H

TEI-9648 ML

FIGURE 7 Molecular tools for study of rapid responses and genomic responses mediated by 1α,25(OH)2D3 and related analogs. The structures shown for 1α,25(OH)2D3 indicate (top) the 6-cis conformational shape and (bottom) the 6-trans conformations of 1α,25(OH)2D3. The natural hormone is a full agonist for all known rapid responses and genomic responses. In the remainder of the figure, the upper panel presents three structures of analogs in the 6-s-cis shape of 1α,25(OH)2D3 that have proven very useful for study of rapid responses. The analogs JN and MZ are agonists for mediation of rapid responses (see text in later figures). Analog HL or 1β,25(OH)2D3 is an antagonist only of rapid responses [88,89]. The structures shown on the bottom row (MK and ML) are analogs of a naturally occurring metabolite of 1α,25(OH)2D3, namely the 1α,25(OH)2-26,23-lactone-D3. In analogs MK the 25-hydroxyl group has been dehydrated to give a 26–27 double bond. These analogs have been shown to be antagonists of the nuclear VDR and are without effect on rapid responses [90,94].

breakthrough studies. In follow-up studies, Zanello and Norman [96] carried out a structure–function analysis in ROS 17/2.8 cells to identify the structural components of 1α,25(OH)2D3 that were crucial for opening the L-gated Ca2+ channels. In these studies we also learned that 1α,25(OH)2D3 was also able to open chloride channels ROS 17/2.8 cells. Figure 8A summarizes a structure–function study evaluating the ability of 1α,25(OH)2D3 and related analogs to open chloride channels. The results emphasize the structural specificity inherent in the receptor that binds these analogs to initiate rapid responses. Thus, the 6-s-cis-locked 1α,25(OH)2D3-lumisterol is a full agonist in comparison to 1α,25(OH)2D3 for opening of chloride channels. However, the companion 6-s-trans–locked analog 1α,25(OH)2-tachysterol3 is without agonist activity. These results emphasize that rapid responses are initiated only by vitamin D secosteroids that have a shape

similar to the 6-s-cis. Also, it is clear that other nonsecosteroids such as cholesterol or 17β estradiol are impotent in this system. Use of the analog HL [1β,25(OH)2D3] indicates that the membrane receptor has no agonistic response to this ligand. However, as shown by the combined use of 1β,25(OH)2D3 with 1α,25(OH)2D3, it was found that the beta isomer functions as an antagonist. This observation is studied in further detail and the results shown in Fig. 8B. Here a dose-response experiment employing 1α,25(OH)2D3 as an agonist was performed; as is typical of most rapid responses to this hormone, the dose-response curve is biphasic with maximal activity at about 5–20 nM with a fall-off in response at concentrations > 50 nM. However, in the combined presence of 1β,25(OH)2D3 and 1α,25(OH)2D3, it is clear that 1β,25(OH)2D3 is able to block all dose levels of the natural hormone. These results were in agreement with earlier studies

394 A

ANTHONY W. NORMAN

5

OH

B

H OH

4 3

Fold increase of the 2 currents

H

5

HO 1α,25(OH)2-Lumisterol3 JN

H OH H

1α,25(OH)2-Tachysterol3 JB

HO

OH

1 0 1α,25(OH)2D3 HL 1α,25D+HL JN 1β,25D

1α,25-D3 alone 1α,25-D3 +1β,25-D3

6

H

ANTAGONIST

Fold 4 increase 3 of chloride currents 2

HO

0 Cholesterol 17βEstradiol

OH H 1α,25(OH)2D3 HL

1

JB

H

OH

0.05

0.5 5 1α,25(OH)2D3 (nM )

50

FIGURE 8 Structure–function evaluation of ligand opening of chloride currents in ROS 17/2.8 cells. (A) Comparative affects of analogs of 1α,25(OH)2D3 and other steroids on chloride currents and ROS 17/2.8 cells. For further details see (AWN 660). (B) Fold increase of chloride currents mediated by 1α,25(OH)2D3 in the absence and presence of the antagonist 1 nM 1β,25(OH)2D3. This data resulted from patch clamping of single isolated cells. 1α,25(OH)2D3 or analogs HL [1β,25(OH)2D3] were added at least 3 min prior to the electrical measurements. For further details see [96].

where analog HL was also found to be capable of inhibiting 1α,25(OH)2D3 mediated transcaltachia [89]. In data not presented, the ability of 1β,25(OH)2D3 to bind to the VDRnuc was evaluated in comparison to 1α,25(OH)2D3; it can only bind approximately 0.5% as well as the natural hormone. From these studies, we conclude that the agonist effect of 1α,25(OH)2D3 in osteoblasts at the site of the membrane level seems to be determined by some structural features of the hormone molecule that may be crucial for its interaction with the putative membrane receptor in the cell surface. 2. CASE STUDY 2: 1α,25(OH)2D3 STIMULATION OF INSULIN SECRETION

Pancreatic β-cells have been found to have a receptor for vitamin D and to be a target of 1α,25(OH)2D3 [39,97]. Although there are many studies on the effect of 1α,25(OH)2D3 on insulin secretion [98–102], the precise role of vitamin D in β-cell function is not yet clear. Most investigation has been restricted to the chronic effects of 1α,25(OH)2D3 and its analogs on insulin secretion. There have been only a small number of investigations concerning the rapid nongenomic effect of 1α,25(OH)2D3 on pancreatic β-cell function. Among these studies, Sergeev and Rhoten [103] have reported that the administration of 1α,25(OH)2D3 evoked oscillations of intracellular Ca2+ in a pancreatic β-cell line within a few minutes. However, the detailed mechanism of the nongenomic effect of 1α,25(OH)2D3 through signal transduction via VDRmem in pancreatic β-cells remains unclear. In this study, we have compared the effect of two 1α,25(OH)2D3 analogs, 1α,25(OH)2lumisterol and 1β,25(OH)2D3, in the rapid action of 1α,25(OH)2D3 on insulin secretory mechanisms, and we find evidence of nongenomic stimulation

by vitamin D on insulin secretion in normal rat pancreatic β-cells. As shown in Fig. 9, 1α,25(OH)2lumisterol3 (JN) augmented the 16.7 nM glucose–induced insulin release from rat islets in a dose-dependent manner. This fact was primarily apparent in the second phase of insulin release induced by 16.7 nM glucose, which occurs after 10 min. Also in data not shown, the stimulation was reversible within 15 min after withdrawal of analog JN from the medium. Strikingly, the stimulatory effect of insulin release was completely abolished by addition of 1β,25(OH)2D3 (HL), whereas analog HL alone did not affect glucose-induced insulin release. This result strongly suggests that 1α,25(OH)2D3 and its analog JN function in a nongenomic fashion in the pancreas β-cell. In pancreatic β-cells, signals generated through intracellular glucose metabolism are believed to close the KATP channel on the plasma membrane. The resulting decrease in K+ conductance leads to depolarization of the membrane, with subsequent opening of a voltagedependent Ca2+ channel. Ca2+ influx through these channels then increases, leading to a rise in [Ca2+]i, which eventually triggers exocytosis of insulin granules [103,104]. The secretory enhancement by 1α,25(OH)2− lumisterol3 was observed also in the presence of diazoxide (an opener of KATP channels), a depolarizing concentration of K+, and 16.7 mM glucose (which is known to produce so-called KATP channel-independent insulin release through metabolic signaling derived from glucose) [105,106]. On the other hand, both insulin release and [Ca2+]i dynamics were not affected by 1α,25(OH)2lumisterol3 in the presence of a basal concentration (3.3 mM) of glucose, indicating that 1α,25(OH)2lumisterol3 itself is not a depolarizing agent,

CHAPTER 23 Function Relationships for Both the Ligand and Receptor(s)

Insulin release (ng/islet/ 30 min)

6

4

2

0 JN (M) 0 HL (M)

10−10

10−9

10−8

10−7

10−8 10−8

10−8

FIGURE 9 Glucose-induced insulin secretion from rat pancreatic islets stimulated by 6-s-cis 1α,25(OH)2D3 2-lumisterol (analog JN). In this study [120] islets of Langerhans were isolated from rats by collagenase digestion. Insulin release from intact islets was monitored using either static incubation or diffusion system. The results shown in this panel are obtained from a study of the concentration-dependent effect of 1α,25(OH)2-lumisterol (JN) on 16.7 nM glucose-induced insulin secretion from rat pancreatic islets. In some incubations, the analog of 1β,25(OH)2D3 (HL) was added simultaneously with the analog JN. Values are expressed at the means ± SE of five determinations in the same experiment. *P < 0.05 vs control (16.7 nM glucose alone).

like sulfonylurea, which induces KATP channel closure to produce membrane depolarization. 1α,25(OH)2lumisterol3, therefore, acts distal to membrane depolarization in glucose-induced insulin secretion, and its enhancement of insulin secretion could result from rising [Ca2+]i or augmentation of Ca2+ efficacy in the exocytotic system. However, in the presence of depolarizing concentration of K+ with diazoxide, 1α,25(OH)2lumisterol3 did not enhance insulin release without a high concentration of glucose, and even with a high concentration of glucose, 1α,25(OH)2lumisterol3 failed to enhance insulin release from diazoxidehyperpolarized islets. 1α,25(OH)2lumisterol3-enhanced insulin release, therefore, requires both activated glucose metabolism and plasma membrane depolarization, as suggested also by the observation that mannoheptulose prevents 1α,25(OH)2lumisterol3 from enhancing insulin secretion, even in the depolarizing condition. 3. CASE STUDY 3: ACTIVATION OF PHOSPHATIDYL INOSITOL 3-KINASE BY 1α,25(OH)2D3

Recent studies suggest that 1α,25(OH)2D3 plays an important role in the cardiovascular system through its receptors in the heart and in vascular smooth muscle cells (VSMC) [107,108]. In particular, 1α,25(OH)2D3

395

has been shown to regulate calcium homeostasis, modulate growth, and increase calcification in smooth muscle cells [109–111]. Indeed, the mitogenic role of 1α,25(OH)2D3 in VSMC has been previously reported [110–112], although its effect on the migration has not been investigated. Therefore, the purpose of this study was to determine whether 1α,25(OH)2D3 can promote VSMC migration, and if so, to determine whether the mechanism is mediated by the genomic or nongenomic effects of VDR. As shown in Fig. 10A, the effect of 1α,25(OH)2D3 on vascular smooth muscle cell (VSMC) migration was assessed using a modified Boyden transwell cell culture chamber. The addition of 1α,25(OH)2D3 to the lower compartment of the Boyden chamber induced a concentration-dependent increase of VSMC migration in the upper chamber over the time frame of 4 hr; in this assay there is actual physical movement of the VSMC cell, i.e., a chemotaxis effect happens quite rapidly. Indeed, the rapid response occurred over physiological concentration range of 0.1–100 nM 1α,25(OH)2D3. Overall, the migration rate increased as a consequence of the dose response studies from 19% to 42% over control levels. A role for PI3-kinase in VSMC migration has been previously reported [113,114]. As shown in Figure 10B, treatment of VSMC with 1α,25(OH)2D3 activated PI3kinase in a time course–dependent and dose-dependent fashion (upper two rows of Fig. 10). The PI3-kinase was dramatically activated over a concentration range of 0.12–1 nM 1α,25(OH)2D3. Shown in the bottom row of Fig. 10B, the two inhibitors of PI3-kinase, namely wortmannin and LY294002, both blocked activation by 1α,25(OH)2D3. In contrast, the RNA polymerase inhibitor, 5,6-dichloro-1-β-D-ribofuranosylben3imidazole (DRB), was without effect on 1α,25(OH)2D3 activation of PI3-kinase. In data not shown, DRB, which is an inhibitor of genomic responses, was shown to block gene transcription induced by TMF-α. In additional data not presented, using 1α,25(OH)2lumisterol (JN) duplicated all of the results shown in Fig. 10B. In the results shown in Fig.10C, a comparison of the effect of the two specific molecular tool antagonists, namely analog HL, rapid responses, and analog MK, genomic responses, on 1α,25(OH)2D3 activation of PI3-kinase. The rapid response measured as a consequence of 1α,25(OH)2D3 induced cell migration as well as the activation of PI3-kinase are shown; both rapid responses were blocked by 1β,25(OH)2D3 (HL), but were not affected by the genomic VDR receptor antagonist, analog MK. Collectively, these findings suggest there exists a rapid response membrane receptor that mediates the nongenomic actions of 1α,25(OH)2D3 VSMC and

396

1,25D3-induced cell migration (increased in %)

A

ANTHONY W. NORMAN

B

50

*

40

*

Time course PIP3

* C

30

10′

30′

60′

1

10

1,25D3

*

20

5′

Dose response PIP3

10

0

C

0

0.01

0.1 1 1,25D3 (nmol/L)

10

0.01

100

0.1

1,25D3 (nmol/L) Inhibitors PIP3

C

−

Wort

LY

−

DRB

1,25D3

C 1,25D3-induced cell migration (increase in %)

40

30

* *

20

10

0

−

Control

HL

MK

1,25D3 PIP3 C

−

HL

MK

1,25D3

FIGURE 10 Evaluation of 1α,25(OH)2D3 mediated rapid responses in vascular endothelial cells with concordant activation of phosphatidyl-inositol-3-kinase activation. (A) 1α,25(OH)2D3 induces vascular smooth muscle cell (VSMC) migration. The effect of increasing concentrations of 1α,25(OH)2D3 on VSMC migration is shown. The experiments were performed three times in triplicates. *P < 0.05 compared with unstimulated cells. (B) 1α,25(OH)2D3 induces PI3-kinase activity in VSMCs. Time course: Time-dependent 5–60 min increase in PI3-kinase activity induced by 1α,25(OH)2D3 (1 nmol/L) compared with unstimulated or control cells. Dose–response: Effect of increasing concentrations of 1α,25(OH)2D3 of PI3-kinase activity. Cells were stimulated for 10 min with 1α,25(OH)2D3 (0.01 to 10 nmol/liter) were left untreated. Inhibitors: Effect of wortmanin (WORT, 30 nmol/liter), LY294002 (L, 10 µmol/liter), or DRB (50 µmol/liter) on PI3-kinase activity induced by 1α,25(OH)2D3 (10 nmol/liter). All experiments in this panel were performed three times. Results taken from [120]. (C) Effect of 1β,25(OH)2D3 (HL) and (23S)-25-dehydro-1α-OH-D3-26,23-lactone (MK on 1α,25(OH)2D3induced VSMC migration and PI3-kinase activity. (Top) VSMC migration was induced by 1 nmol/liter of 1α,25(OH)2D3 in the presence or in the absence of 50 nmol/liter of HL or MK. Experiments were performed three times in triplicates. *P < 0.05 compared with unstimulated cells (control). (Bottom) Effect of HL or MK (50 nmol/liter) on PI3-kinase activity induced by 1α,25(OH)2D3 (1 nmol). Experiments were performed three times with comparable results (from the following study [121]).

397

CHAPTER 23 Function Relationships for Both the Ligand and Receptor(s)

PI3-kinase. The mechanism underlying this nongenomic response is not completely understood. The PI3kinase assays suggest that a signaling complex containing both the VDR and PI3-kinase is formed after the addition of 1α,25(OH)2D3. It is possible that this activation process could occur in the plasma membrane or caveolaeenriched membrane fraction; alternatively, it is possible that the activation process of PI3-kinase occurred in the cell cytosol. A similar observation to that discussed earlier for 1α,25(OH)2D3 was reported concerning the nongenomic/rapid response effects of estradiol and endothelial cells [115]. The estrogen receptor (ER) was also shown to activate PI3-kinase. However, additional evidence suggested that the ER formed a heterodimer with the p85α regulating subunit of PI3-kinase.

C. Summary of Structure–Function Studies Table V represents a structure–function summary analysis of the process of transcaltachia (the rapid hormonal stimulation of calcium absorption) as well as the three case studies (insulin secretion, smooth muscle cell migration, and opening of chloride channels). In each of these systems, the natural hormone 1α,25(OH)2D3 was a full agonist. Further, the 6-s-cis locked 1α,25(OH)2lumisterol3 was also a full agonist. In the two systems that were evaluated, transcaltachia [88] and chloride channel opening [96], the 6-s-trans restricted analog 1α,25(OH)2tachysterol3 was without any agonist effect for these rapid responses. Finally, 1β,25(OH)2D3 (analog HL) was a potent antagonist of rapid responses mediated by the natural hormone. Collectively, the structure–function studies carried out in these four systems with respect to rapid responses

clearly indicate that there is a membrane-associated receptor that can distinguish structural differences in potential ligand agonists (e.g., 6-s-cis vs 6-s-trans).

V. SUMMARY This chapter has focused on key structure–function relationships that bear on differences between the classical genomic responses mediated by the VDR in contrast to the putative VDR membrane receptor which is responsible for rapid responses. A third important player in the biological systems where both genomic and rapid responses occur is the plasma transport protein, the vitamin D binding protein (DBP). All of the experimental results summarized here emphasize that genomic responses and rapid responses are mediated by different shapes of the agonist hormone 1α,25(OH)2D3. Although the agonist 1α,25(OH)2D3 is conformationally flexible because of three facets of its structure (see Fig. 1), the dominant structural aspect relevant to the present discussion is the fact that it is a secosteroid. More specifically, the 9–10 carbon bond is broken in vitamin D–related compounds, thus imparting complete 360° capability to rotate around the 6–7 single carbon bond (see Fig. 11). This figure emphasizes by 90° increments the change in shapes moving from the planar-6-s-cis shape at 0° through 90° (α triene) to 180° (planar 6-s-trans shape), and to 270° (β-triene) and then finally 90° back to the starting point of 0°. In solutions (including the cell cytosol) the free 1α,25(OH)2D3 ligand would present literally millions of shapes to potential receptors. However the nature of receptor– ligand interaction is such that only one preferred shape will be selected to form a receptor–ligand complex.

TABLE V Summary of Case Studies: 1α,25(OH)2D3 Structure–Function Analysis of Rapid Responses Case studies Response → Agonist ↓ 1α,25(OH)2D3 conformationally flexible (JN) 6-s-cis locked (JB) 6-s-trans locked (HL) 1β,25(OH)2D3 Reference

Chick intestine: Transcaltachia (rapid Ca2+ absorption)

Osteoblasts: Opening Cl− channels & secretion

Pancreatic β-cells: Insulin secretion

Endothelial cells: Smooth muscle cell migration

Yes

Yes

Yes

Yes

Yes No Antagonism Norman, Bouillon, Farach-Carson [88]

Yes No Antagonism Zanello, Norman [96]

Yes — Antagonism Ishida, Norman [120]

Yes — Antagonism Liao, Norman [121]

398

ANTHONY W. NORMAN

360° Rotation around 6,7-single bond of 1α,25(OH)2D3 Planar 6-s-cis SC D H C H

VDRmem ligand agonist

A OH OH 0° 315°

HO β-Triene

SC D C

A H

OH 270°

45°

Rotation above CDplane

SC 90°

Free 1,25-D3

211°

180°

H H

α-Triene

A

Rotation below CDplane 135°

225°

D C

OH

HO

149°

SC D C

Bound 1,25-D3 VDR-nuc

OH

H H

Bound 25-D3 DBP-LBD

A

Planar 6-s-trans

OH

FIGURE 11 Biologically relevant ligand shapes of 1α,25(OH)2D3 for rapid responses, genomic responses, and binding to vitamin D binding protein (DBP). This illustrates the consequences of 360° rotation about the 6,7 carbon bond of 1α,25(OH)2D3. Four steps of successive 90° rotations are illustrated. Each intermediate ligand structure is a different shape, particularly with respect to the position of the critical 1α-hydroxyl and the orientation of the A-ring in relation to the plane of the C/D rings. The three arrows indicate the degrees of rotation about the 6,7 carbon bond of the ligand for the VDRmem (0°), the ligand for DBP (+149°), and the ligand for VDRnuc (+211°).

A

B

1α,25(OH)2D3 VDRnuc

C

1α,25(OH)2-lumisterol VDRmem

25(OH)D3 DBP

FIGURE 12 Space-filling and stick representations illustrating the clear differences in shapes for the optimal VDRmem ligand, 1α,25(OH)2-lumisterol (planar molecule; analog JN in Fig. 11), the optimal DBP 25(OH)D3 or 1α,25(OH)2D3 ligand (twisted 6-s-trans with the A-ring C10-C19 exocyclic alkene 30° below the plane of the C/D ring), and the optimal VDRnuc ligand 1α,25(OH)2D3 (bowl shaped twisted 6-s-trans with the A-ring 30° above the plane of the C/D ring). Each of the illustrated molecules has its 25-OH pointed to the right and the A-ring to the left. Also, there are significant differences in the orientation of the side chain for the VDRnuc (north) and DBP (south); the optimal side-chain orientation of the VDRmem ligand has not yet been precisely defined.

399

CHAPTER 23 Function Relationships for Both the Ligand and Receptor(s)

Careful inspection of Fig. 11 and 12 will identify the three optimal shapes of the conformationally flexible natural hormone 1α,25(OH)2D3. The optimal ligand shape for the VDRnuc is the bowl shaped 6-s-trans where the A ring is rotated 211° from that of the planar 6-s-trans 1α,25(OH)2D3. In contrast, the optimal ligand for the DBP is a structure which has been rotated 149° from that of the planar 6-s-cis ligand at 0°. Finally, the optimal ligand for the VDRmem is the planar 6-s-cis shape of 1α,25(OH)2D3. The contrast in shapes between the optimal ligand for VDRnuc, VDRmem, and DBP is summarized in Fig. 12. The difference in shapes is dramatic and emphasizes an opportunity as well as a challenge for shape-selective drug development. It is possible to carry out chemical syntheses of conformationally restricted analogs such that a product putative drug could be developed that only interacts with VDRnuc without significant actions

on VDRmem and vice versa. An overriding drug development challenge relates to whether any drug that might be optimal for VDRmem would also bind significantly to DBP, which might be a necessity for transport of the drug to its cellular sites of action. Figure 13 is a summary figure that attempts to integrate the signal transduction pathways which are used by the nuclear receptor (VDRnuc) and the membrane receptor (VDRmem) of 1α,25(OH)2D3. Figure 13 emphasizes the complexity of overlapping and interconnecting signal transduction pathways. Whereas there is a solid and secure foundation describing 1α,25(OH)2D3mediated genomic actions, 1α,25(OH)2D3-mediated rapid responses are still in their developmental phase. One extraordinary challenge will be to identify and fully biochemically characterize the VDRmem. An equally daunting task will be to define which signal transduction pathways mediate rapid responses in the

1α,25(OH)2D3 = RAPID RESPONSES Ligand shape matters

Affects target cell biology

1α,25(OH)2D3

PLASMA MEMBRANE Ca 2+

Cell nucleus

Ca 2+

Caveolae Signal transduction PI3K

Tissues with rapid responses Pancreas B cell Adipocytes Vascular smooth muscle Intestine Monocytes Osteoblasts

G PROTEIN

GENE EXPRESSION

Altered genomic responses Osteocalcin promoter 24-OHase promoter Alkaline phosphatase NB4 cell differentiation Micro array analysis

PKC

Phospholipase C

Systems PKC RAS/MAP Kinase

Second messengers Phosphoproteins “Cross-talk” RAF/MAP Kinase

PI3 ′Kinase

PtdIns-3,4,5-P3

FIGURE 13 Schematic model describing how the conformationally flexible 1α,25(OH)2D3 can interact with a nuclear receptor to generate genomic responses or a plasma membrane receptor to generate rapid responses. Binding of 1α,25(OH)2D3 to the membrane surface receptor may result in the activation of one or more second messenger systems, including phospholipase C, protein kinase C, G protein coupled receptors, or phosphatidyl-inositol-3-kinase (PI3). There are a number of possible outcomes including opening of the voltage-gated calcium channel or generation of the indicated second messengers. Some of these second messengers, particularly RAF/MAP kinase, may engage in cross-talk with a nucleus to modulate gene expression. Evidence has been presented that rapid responses can modulate the list of specific genes tabulated in the figure [122].

400

ANTHONY W. NORMAN

various cell systems. A further important question is to answer whether the VDRmem can communicate with the nucleus of the cell to modulate gene transcription. In data not shown but summarized in Fig. 13, there are at least five different systems where molecular biological evidence has been obtained for the process of crosstalk. In these studies, 6-s-cis locked analogs have been shown to modulate genomic responses at relatively early time intervals (2–4 hr). Without a doubt, it will be intriguing to monitor developments in this field in the future.

to the flickering gate properties of certain membrane channels that open and close rapidly. In contrast, over a long time-frame (steady state equilibrium), 1α,25(OH)2D3 occupancy of the genomic pocket is significantly favored over the alternative pocket because the increase in hydrophobic stability prolongs the half-life of the ligand in the genomic pocket. A related model to describe ligand/receptor induced dissociation of rapid from genomic responses was included in the comprehensive analysis of nongenotropic, sex-nonspecific signaling by the ER and AR [228].

VI. ADDENDUM

References

Recent evidence suggests that both the VDR and the estrogen receptor (ER) are localized to plasma membrane caveolae and they are also required for initiation of non-genomic rapid responses [224,225]. Using the atomic coordinates of the X-ray structure of the VDR and computer modeling, we have been able to identify the presence of a putative alternative LBD in the VDR that can accommodate via computer docking either the conformationally flexible 1,25(OH)2D3 or the 6-s-cis locked 1α,25(OH)2lumisterol; both are known to be full agonists for rapid responses [226]. As a consequence we have proposed that a conformational ensemble receptor model can give insight into the complex ligand-receptor relationships for the VDR that determine whether the genomic (G) or alternative (A) pocket becomes occupied by a ligand leading to either a genomic or rapid response [227]. In this model the unbound receptor macromolecules exist in the cytoplasm as multiple, equilibrating VDR conformations that follow the laws associated with standard statistical distributions. Thus, 1,25(OH)2D3 or a drug ligand would sample the existing population of VDR conformations available thus generating a new statistical distribution of the ligand-receptor conformers. The population of each VDR-ligand conformer is dependent on the ligand’s structure and chemistry and the subsequent receptor A- versus G-pocket stabilities. The genomic pocket portal is believed to be controlled by the position of helix-12 and a finite time interval will be required for the helix to move to an open position; thus, entrance to the genomic pocket is gated. While details of access to the alternative pocket are not yet known, it is proposed that given the increased loop character of the amino acid residues forming the lid of the alternative pocket, the energy barrier between the opened and closed states is far less than that required for helix-12 repositioning. This effectively makes the alternative pocket more accessible to the ligand than the genomic pocket. This mechanism may be analogous

1. Bouillon R, Okamura WH, Norman AW 1995 Structure– function relationships in the vitamin D endocrine system. Endocr Rev 16:200–257. 2. Henry HL, Norman AW 1991 Metabolism of vitamin D. In: Coe FL, Favus MJ (eds) Disorders of Bone and Mineral Metabolism. Raven Press, New York. 3. Henry HL, Norman AW 1984 Vitamin D: Metabolism and biological action. Ann Rev Nutr 4:493–520. 4. Midland MM, Plumet J, Okamura WH 1993 Effect of C20 stereochemistry on the conformational profile of the side chains of vitamin D analogs. Bioorg Med Chem Lett 3: 1799–1804. 5. Okamura WH, Palenzuela JA, Plumet J, Midland MM 1992 Vitamin D: Structure–function analyses and the design of analogs. J Cell Biochem 49:10–18. 6. Wing RM, Okamura WH, Pirio MR, Sine SM, Norman AW 1974 Vitamin D3: Conformations of vitamin D3, 1α,25dihydroxyvitamin D3, and dihydrotachysterol3. Science 186: 939–941. 7. Okamura WH, Norman AW, Wing RM 1974 Vitamin D: Concerning the relationship between molecular topology and biological function. Proc Natl Acad Sci USA 71:4194–4197. 8. Norman AW, Okamura WH, Farach-Carson MC, Allewaert K, Branisteanu D, Nemere I, Muralidharan KR, Bouillon R 1993 Structure–function studies of 1,25-dihydroxyvitamin D3 and the vitamin D endocrine system. 1,25-Dihydroxypentadeuterio-previtamin D3 (as a 6-s-cis analog) stimulates nongenomic but not genomic biological responses. J Biol Chem 268:13811–13819. 9. Renaud JP, Rochel N, Ruff M, Vivat V, Chambon P, Gronemeyer H, Moras D 1995 Crystal structure of the RARgamma ligand-binding domain bound to all-trans retinoic acid. Nature 378:681–689. 10. Wagner RL, Apriletti JW, McGrath ME, West BL, Baxter JD, Fletterick RJ 1995 A structural role for hormone in the thyroid hormone receptor. Nature 378:690–697. 11. Bishop JE, Collins ED, Okamura WH, Norman AW 1994 Profile of ligand specificity of the vitamin D binding protein for 1α,25(OH)2-vitamin D3 and its analogs. J Bone Miner Res 9:1277–1288. 12. Norman AW, Hurwitz S 1993 The role of the vitamin D endocrine system in avian bone biology. J Nutr 123:310–316. 13. Norman AW 1998 Receptors for 1α,25(OH)2D3: Past, present, and future. J Bone Miner Res 13:1360–1369. 14. Lowe KE, Maiyar AC, Norman AW 1992 Vitamin D-mediated gene expression. Crit Rev Eukar Gene Exp 2:65–109.

CHAPTER 23 Function Relationships for Both the Ligand and Receptor(s)

15. Hannah SS, Norman AW 1994 1α,25(OH)2-vitamin D3regulated expression of the eukaryotic genome. Nutr Rev 52:376–382. 16. Norman AW, Collins ED 2001 Vitamin D and gene expression. In: Moussa NN, Berdanier C (eds) Nutrient–Gene Interactions in Health and Disease. CRC Press, Orlando, FL. 17. Norman AW, Bishop JE, Bula CM, Olivera CJ, Mizwicki MT, Zanello LP, Ishida H, Okamura WH 2002 Molecular tools for study of genomic and rapid signal transduction responses initiated by 1α,25(OH)2-vitamin D3. Steroids 67:457–466. 18. Norman AW 2002 Nuclear receptor structure (VDR) and ligand specificities for genomic and rapid biological responses. In: Bilezikian JP, Raisz LG, Rodan GA (eds) Principles of Bone Biology. Academic Press, San Diego. 19. Nemere I, Farach-Carson MC 1999 Membrane receptors for steroid hormones: A case for specific cell surface binding sites for vitamin D metabolites and estrogens. Biochem Biophys Res Commun 248:443–449. 20. Caffrey JM, Farach-Carson MC 1989 Vitamin D3 metabolites modulate dihydropyridine-sensitive calcium currents in clonal rat osteosarcoma cells. J Biol Chem 264:20265–20274. 21. Berry DM, Antochi R, Bhatia M, Meckling-Gill KA 1996 1,25-Dihydroxyvitamin D3 stimulates expression and translocation of protein kinase Cα and Cδ via a nongenomic mechanism and rapidly induces phosphorylation of a 33-kDa protein in acute promyelocytic NB4 cells. J Biol Chem 271:16090–16096. 22. Anderson RG, Jacobson K 2002 A role for lipid shells in targeting proteins to caveolae, rafts, and other lipid domains. Science 296:1821–1825. 23. Isshiki M, Ying YS, Fujita T, Anderson RG 2002 A molecular sensor detects signal transduction from caveolae in living cells. J Biol Chem 277:43389–43398. 24. Razani B, Woodman SE, Lisanti MP 2002 Caveolae: from cell biology to animal physiology. Pharmacol Rev 54:431–467. 25. Ortí E, Bodwell JE, Munck A 1992 Phosphorylation of steroid hormone receptors. Endocr Rev 13:105–128. 26. Darwish HM, Burmester JK, Moss VE, DeLuca HF 1993 Phosphorylation is involved in transcriptional activation by the 1,25-dihydroxyvitamin D3 receptor. Biochim Biophys Acta Lipids Lipid Metab 1167:29–36. 27. Jurutka PW, Hsieh J-C, Haussler MR 1993 Phosphorylation of the human 1,25-dihydroxyvitamin D3 receptor by cAMP-dependent protein kinase, in vitro, and in transfected COS-7 cells. Biochem Biophys Res Commun 191: 1089–1096. 28. Hsieh JC, Jurutka PW, Nakajima S, Galligan MA, Haussler CA, Shimizu Y, Shimizu N, Whitfield GK, Haussler MR 1993 Phosphorylation of the human vitamin D receptor by protein kinase C. Biochemical and functional evaluation of the serine 51 recognition site. J Biol Chem 268:15118–15126. 29. Jurutka PW, Hsieh J-C, MacDonald PN, Terpening CM, Haussler CA, Haussler MR, Whitfield GK 1993 Phosphorylation of serine 208 in the human vitamin D receptor. J Biol Chem 268:6791–6799. 30. Freedman LP, Arce V, Perez Fernandez R 1994 DNA sequences that act as high affinity targets for the vitamin D3 receptor in the absence of the retinoid X receptor. Mol Endocrinol 8:265–273. 31. Liu M, Freedman LP 1994 Transcriptional synergism between the vitamin D3 receptor and other nonreceptor transcription factors. Mol Endocrinol 8:1593–1604. 32. Walters MR 1992 Newly identified actions of the vitamin D endocrine system. Endocr Rev 13:719–764.

401

33. Norman AW, Roth J, Orci L 1982 The vitamin D endocrine system: Steroid metabolism, hormone receptors and biological response (calcium binding proteins). Endocr Rev 3:331–366. 34. Carlberg C, Quack M, Herdick M, Bury Y, Polly P, Toell A 2001 Central role of VDR conformations for understanding selective actions of vitamin D3 analogues. Steroids 66:213–221. 35. Haussler MR, Whitfield GK, Haussler CA, Hsieh JC, Thompson PD, Selznick SH, Dominguez CE, Jurutka PW 1998 The nuclear vitamin D receptor: Biological and molecular regulatory properties revealed. J Bone Miner Res 13:325–349. 36. Haussler MR, Myrtle JF, Norman AW 1968 The association of a metabolite of vitamin D3 with intestinal mucosa chromatin, in vivo. J Biol Chem 243:4055–4064. 37. Haussler MR Norman AW 1969 Chromosomal receptor for a vitamin D metabolite. Proc Natl Acad Sci USA 62:155–162. 38. Christakos S, Norman AW 1981 Studies on the mode of action of calciferol XXIX—Biochemical characterization of 1,25-dihydroxyvitamin D3 receptors in chick pancreas and kidney cytosol. Endocrinology 108:140–149. 39. Ishida H, Norman AW 1988 Demonstration of a high affinity receptor for 1,25-dihydroxyvitamin D3 in rat pancreas. Mol Cell Endocrinol 60:109–117. 40. Roth J, Bonner-Weir S, Norman AW, Orci L 1982 Immunocytochemistry of vitamin D-dependent calcium binding protein in chick pancreas: Exclusive localization in B-cells. Endocrinology 111:2216–2218. 41. Tsoukas CD, Provvedini DM, Manolagas SC 1984 1,25Dihydroxyvitamin D3: A novel immunoregulatory hormone. Science 224:1438–1440. 42. Adorini L 2002 Immunomodulatory effects of vitamin D receptor ligands in autoimmune diseases. Int Immunopharmacol 2:1017–1028. 43. Mathieu C, Adorini L 2002 The coming of age of 1,25-dihydroxyvitamin D3 analogs as immunodulatory agents. Trends Mol Med 8:174–179. 44. Nemere I, Yoshimoto Y, Norman AW 1984 Studies on the mode of action of calciferol. LIV. Calcium transport in perfused duodena from normal chicks: enhancement with 14 minutes of exposure to 1α,25-dihydroxyvitamin D3. Endocrinology 115: 1476–1483. 45. Norman AW, Nemere I, Williams G, King M 1984 1α,25Dihydroxyvitamin D3 mediates biological responses both as a steroid hormone and as a membrane-active agent. In: Cohn DV, Potts JT Jr, Fujita T (eds) Endocrine Control of Bone and Calcium Metabolism. Elsevier Science, Amsterdam. 46. Nemere I, Zhou L-X, Norman AW 1993 Nontranscriptional effects of steroid hormones. Receptor 3:277–291. 47. Nemere I, Dormanen MC, Hammond MW, Okamura WH, Norman AW 1994 Identification of a specific binding protein for 1α,25-dihydroxyvitamin D3 in basal-lateral membranes of chick intestinal epithelium and relationship to transcaltachia. J Biol Chem 269:23750–23756. 48. Kim YS, MacDonald PN, Dedhar S, Hruska KA 1996 Association of 1α,25-dihydroxyvitamin D3-occupied vitamin D receptors with cellular membrane acceptance sites. Endocrinology 137:3649–3658. 49. Norman AW, Olivera CJ, Barreto Silva FR, Bishop JE 2002 A specific binding protein/receptor for 1α,25-dihydroxy D3 is present in an intestinal caveolae membrane fraction. Biochem Biophys Res Commun 298:414–419. 50. Losel RM, Falkenstein E, Feuring M, Schultz A, Tillmann HC, Rossol-Haseroth K, Wehling M 2003 Nongenomic steroid

402

51. 52.

53.

54.

55.

56. 57. 58.

59.

60.

61.

62. 63. 64.

65. 66.

67. 68. 69.

ANTHONY W. NORMAN

action: controversies, questions, and answers. Physiol Rev 83:965–1016. 2002 Proceedings of the Second International Meeting on Rapid Responses to Steroid Hormones. Steroid 67:413–548. Cato AC, Nestl A, Mink S 2002 Rapid actions of steroid receptors in cellular signaling pathways. Science STKE 2002 http://stke.sciencemag.org/cgi/content/full/sigtrans; 2002/138/re9:RE9. Zhu Y, Rice CD, Pang Y, Pace M, Thomas P 2003 Cloning, expression, and characterization of a membrane progestin receptor and evidence it is an intermediary in meiotic maturation of fish oocytes. Proc Natl Acad Sci USA 100:2231–2236. Zhu Y, Bond J, Thomas P 2003 Identification, classification, and partical characterization of genes in humans and other vertebrates homologous to a fish membrane progestin receptor. Proc Natl Acad Sci USA 100:2237–2242. Chambliss KL, Yuhanna IS, Mineo C, Liu P, German Z, Sherman TS, Mendelsohn ME, Anderson RG, Shaul PW 2000 Estrogen receptor alpha and endothelial nitric oxide synthase are organized into a functional signaling module in caveolae. Circ Res 87:E44–E52. Mendelsohn ME 2000 Nongenomic, ER-mediated activation of endothelial nitric oxide synthase: How does it work? What does it mean? Circ Res 87:956–960. Shaul PW 2000 Novel role of estrogen receptors in vascular endothelium. Semin Perinatol 24:70–74. Wyckoff MH, Chambliss KL, Mineo C, Yuhanna IS, Mendelsohn ME, Mumby SM, Shaul PW 2001 Plasma membrane estrogen receptors are coupled to endothelial nitricoxide synthase through Gai. J Biol Chem 276:27071–27076. Seo E-G, Kato A, Norman AW 1996 Evidence for a 24R,25(OH)2-vitamin D3 receptor/binding protein in a membrane fraction isolated from a chick tibial fracture-healing callus. Biochem Biophys Res Commun 225:203–208. Nemere I, Yazzie-Atkinson D, Johns DO, Larsson D 2002 Biochemical characterization and purification of a binding protein for 24,25-dihydroxyvitamin D3 from chick intestine. J Endocrinol 172:211–219. Boyan BD, Bonewald LF, Sylvia VL, Nemere I, Larsson D, Norman AW, Rosser J, Dean DD, Schwartz Z 2002 Evidence for distinct membrane receptors for 1α,25-(OH)2D3 and 24R,25-(OH)2D3 in osteoblasts. Steroids 67:235–246. Willson TM, Moore JT 2002 Genomics verus orphan nuclear receptors: A half-time report. Mol Endocrinol 16:1135–1144. Evans RM 1988 The steroid and thyroid hormone receptor superfamily. Science 240:889–895. Mangelsdorf DJ, Thummel C, Beato M, Herrlich P, Schütz G, Umesono K, Blumberg B, Kastner P, Mark M, Chambon P, Evans RM 1995 The nuclear receptor superfamily: The second decade. Cell 83:835–839. Weatherman RV, Fletterick RJ, Scanlon TS 1999 Nuclear receptor ligands and ligand-binding domains. Annu Rev Biochem 68:559–582. Krust A, Green S, Argos P, Kumar V, Walter P, Bornert JM, Chambon P 1986 The chicken oestrogen receptor sequence: Homology with v-erbA and the human oestrogen and glucocorticoid receptors. EMBO J 5:891–897. Beato M 1989 Gene regulation by steroid hormones. Cell 56:335–344. Bourguet W, Ruff M, Chambon P, Gronemeyer H, Moras D 1995 Crystal structure of the ligand-binding domain of the human nuclear receptor RXRα. Nature 375:377–382. Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179.

70. Tocchini-Valentini G, Rochel N, Wurtz JM, Mitschler A, Moras D 2001 Crystal structures of the vitamin D receptor complexed to superagonist 20-epi ligands. Proc Natl Acad Sci USA 98:5491–5496. 71. Daiger SP, Schanfield MS, Cavalli-Sforza LL 1975 Group-specific component (Gc) proteins bind vitamin D and 25-hydroxyvitamin D. Proc Natl Acad Sci USA 72: 2076–2080. 72. Delmez JA, Tindira C, Grooms P, Dusso A, Windus DW, Slatopolsky E 1989 Parathyroid hormone suppression by intravenous 1,25-dihydroxyvitamin D. A role for increased sensitivity to calcium. J Clin Invest 83:1349–1355. 73. Bouillon R, Van Baelen H 1981 Transport of vitamin D: Significance of free and total concentrations of the vitamin D metabolites. Calcif Tissue Int 33:451–453. 74. Haddad JG 1995 Plasma vitamin D–binding protein (Gc-globulin): Multiple tasks. J Steroid Biochem Mol Biol 53:579–582. 75. Haddad JG Jr, Walgate J 1976 25-Hydroxyvitamin D transport in human plasma. Isolation and partial characterization of calcifidiol-binding protein. J Biol Chem 251:4803–4809. 76. Song YH, Naumova AK, Liebhaber SA, and Cooke NE 1999 Physical and meiotic mapping of the region of human chromosome 4q11-q13 encompassing the vitamin D binding protein DBP/Gc-globulin and albumin multigene cluster. Genome Res 9:581–587. 77. Bouillon R, Van Baelen H, De Moor P 1977 The measurement of the vitamin D-binding protein in human serum. J Clin Endocr Metab 45:225–231. 78. Gibbs PEM, Dugaiczyk A 1987 Origin of structural domains of the serum albumin gene family and a predicted structure of the gene for vitamin D binding protein. Mol Biol Evolu 4:364–379. 79. Verboven C, Rabijns A, De Maeyer M, Van Baelen H, Bouillon R, De Ranter C 2002 A structural basis for the unique binding features of the human vitamin D-binding protein. Nat Struct Biol 9:131–136. 80. Head JF, Swamy N, Ray R 2002 Crystal structure of the complex between actin and human vitamin D–binding protein at 2.5 Å resolution. Biochemistry 41:9015–9020. 81. Otterbein LR, Cosio C, Graceffa P, Dominguez R 2002 Crystal structures of the vitamin D-binding protein and its complex with actin: Structural basis of the actin-scavenger system. Proc Natl Acad Sci USA 99:8003–8008. 82. Haddad JG, Hu YZ, Kowalski MA, Laramore C, Ray K, Robzyk P, Cooke NE 1992 Identification of the sterol- and actin-binding domains of plasma vitamin D binding protein (Gc-globulin). Biochem 31:7174–7181. 83. Grishkovskaya I, Avvakumov GV, Sklenar G, Dales D, Hammond GL, Muller YA 2000 Crystal structure of human sex hormone-binding globulin: steroid transport by a laminin G-like domain. EMBO J 19:504–512. 84. Zanotti G, Calderone V, Beda M, Malpeli G, Folli C, Berni R 2001 Structure of chicken plasma retinol-binding protein. Biochim Biophys Acta 1550:64–69. 85. Janssen OE, Lahner H, Grasberger H, Spring SA, Saller B, Mann K, Refetoff S, Einspanier R 2002 Characterization and primary structures of bovine and porcine thyroxine-binding globulin. Mol Cell Endocrinol 186:27–35. 86. Norman AW 1995 Transcaltachia (the rapid hormonal stimulation of intestinal calcium transport): A component of adaptation to calcium needs and calcium availability. Am Zool 35:483–489. 87. De Boland AR, Nemere I, Norman AW 1990 Ca2+-channel agonist Bay K8644 mimics 1,25(OH)2-vitamin D3 rapid

CHAPTER 23 Function Relationships for Both the Ligand and Receptor(s)

88.

89.

90.

91.

92.

93.

94.

95. 96.

97. 98.

99.

100.

enhancement of Ca2+ transport in chick perfused duodenum. Biochem Biophys Res Commun 166:217–222. Norman AW, Okamura WH, Hammond MW, Bishop JE, Dormanen MC, Bouillon R, Van Baelen H, Ridall AL Daane E, Khoury R, Farach-Carson MC 1997 Comparison of 6-s-cis and 6-s-trans locked analogs of 1α,25(OH)2-vitamin D3 indicates that the 6-s-cis conformation is preferred for rapid nongenomic biological responses and that neither 6-s-cis nor 6-s-trans locked analogs are preferred for genomic biological responses. Mol Endocrinol 11: 1518–1531. Norman AW, Bouillon R, Farach-Carson MC, Bishop JE, Zhou L-X, Nemere I, Zhao J, Muralidharan KR, Okamura WH 1993 Demonstration that 1β,25-dihydroxyvitamin D3 is an antagonist of the nongenomic but not genomic biological responses and biological profile of the three A-ring diastereomers of 1α,25-dihydroxyvitamin D3. J Biol Chem 268: 20022–20030. Bula CM, Bishop JE, Ishizuka S, Norman AW 2000 25-Dehydro-1α-hydroxyvitamin D3–26,23S-lactone antagonizes the nuclear vitamin D receptor by mediating a unique noncovalent conformational change. Mol Endocrinol 14:1788–1796. Ishizuka S, Miura D, Eguchi H, Ozono K, Chokki M, Kamimura T, Norman AW 2000 Antagonistic action of novel 1α,25-dihydroxyvitamin D3-26,23-lactone analogs on 25-hydroxyvitamin-D3-24-hydroxylase gene expression induced by 1α,25-dihydroxy-vitamin D3 in human promyelocytic leukemia (HL-60) cells. Arch Biochem Biophys 380:92–102. Ishizuka S, Ishimoto S, Norman AW 1984 Isolation and identification of 1α,25-dihydroxy-24-oxo-vitamin D3, 1α,25-dihydroxyvitamin D3-26,23-lactone and 1α,24(S),25trihydroxyvitamin D3: in vivo metabolites of 1α,25-dihydroxyvitamin D3. Biochemistry 23:1473–1478. Ishizuka S, Norman AW 1987 Metabolic pathways of 1α,25dihydroxyvitamin D3 to 1α,25-dihydroxyvitamin D3-26,23lactone: Stereo-retained and stereo-selective lactonization. J Biol Chem 262:7165–7170. Miura D, Manabe K, Gao Q, Norman AW, Ishizuka S 1999 1α,25-Dihydroxyvitamin D3–26,23-lactone analogs antagonize differentiation of human leukemia cells (HL-60 cells) but not of human acute promyelocytic leukemia cells (NB4 cells). FEBS Lett 460:297–302. Yukihiro S, Posner GH, Guggino SE 1994 Vitamin D3 analogs stimulate calcium currents in rat osteosarcoma cells. J Biol Chem 269:23889–23893. Zanello LP, Norman AW 1997 Stimulation by 1α,25(OH)2vitamin D3 of whole cell chloride currents in osteoblastic ROS 17/2.8 cells: A structure–function study. J Biol Chem 272:22617–22622. Clark SA, Stumpf WE, Sar M, DeLuca HF, Tanaka Y 1980 Target cells for 1,25-dihydroxyvitamin D3 in the pancreas. Cell Tissue Res 209:515–520. Kadowaki S, Norman AW 1984 Studies on the mode of action of calciferol (XLIX). Dietary vitamin D is essential for normal insulin secretion from the perfused rat pancreas. J Clin Invest 73:759–766. Labriji-Mestaghanmi H, Billaudel B, Garnier PE, Malaisse WJ, Sutter BC 1988 Vitamin D and pancreatic islet function. I. Time course for changes in insulin secretion and content during vitamin D deprivation and repletion. J Endocrinol Invest 11:577–584. Tanaka Y, Seino Y, Ishida M, Yamaoka K, Satomura K, Yabuuchi H, Imura H 1986 Effect of 1,25-dihydroxyvitamin D3

101. 102.

103. 104. 105.

106.

107. 108. 109.

110. 111.

112.

113.

114. 115.

116.

117.

403

on insulin secretion: Direct or mediated. Endocrinology 118(5): 1971–1976. Beaulieu C, Kestekian R, Havrankova J, Gascon-Barré M 1993 Calcium is essential in normalizing intolerance to glucose that accompanies vitamin D depletion in vivo. Diabetes 42:35–43. Lee S, Clark SA, Gill RK, Christakos S 1994 1,25Dihydroxyvitamin D3 and pancreatic beta-cell function: Vitamin D receptors, gene expression, and insulin secretion. Endocrinology 134:1602–1610. Sergeev IN, Rhoten WB 1995 1,25-Dihydroxyvitamin D3 evokes oscillations of intracellular calcium in a pancreatic β-cell line. Endocrinology 136:2852–2861. Henquin JC 1994 Cell biology of insulin secretion. In: Kahn CR, Weir GC (eds) Joslin’s Diabetes Mellitus. Lea & Febiger, Philadelphia. Gembal M, Gilon P, Henquin JC 1992 Evidence that glucose can control insulin release independently from its action on ATP-sensitive K+ channels in mouse B cells. J Clin Invest 89:1288–1295. Gembal M, Detimary P, Gilon P, Gao ZY, Henquin JC 1993 Mechanisms by which glucose can control insulin release independently from its action on adenosine triphosphate-sensitive K+ channels in mouse B cells. J Clin Invest 91:871–880. Walters MR, Wicker DC, Riggle PC 1986 1,25Dihydroxyvitamin D3 receptors identified in the rat heart. J Mol Cell Cardiol 18:67–72. Merke J, Hofmann W, Goldschm D, Ritz E 1987 Demonstration of 1,25(OH)2 vitamin-D3 receptors and actions in vascular smooth-muscle cells in vitro. Calcif Tissue Int 41:112–114. Inoue T, Kawashima H 1988 1,25-Dihydroxyvitamin D3 stimulates 45Ca2+-uptake by cultured vascular smooth muscle cells derived from rat aorta. Biochem Biophys Res Commun 152:1388–1394. Mitsuhashi T, Morris RC Jr, Ives HE 1991 1,25Dihydroxyvitamin D3 modulates growth of vascular smooth muscle cells. J Clin Invest 87:1889–1895. Jono S, Nishizawa Y, Shioi A, Morii H 1998 1,25Dihydroxyvitamin D3 increase in vitro vascular calcification by modulating secretion of endogenous parathyroid hormone-related peptide. Circulation 98:1302–1306. Koh E, Morimoto S, Nabata T, Takamoto S, Kitano S, Ogihara T 1990 Effects of 1,25-dihydroxyvitamin D3 on the syntheses of DNA and glycosaminoglycans by rat aortic smooth muscle cells in vitro. Life Sci 46:1545–1551. Duan C, Bauchat JR, Hsieh T 2000 Phosphatidylinositol 3-kinase is required for insulin-like growth factor-I-induced vascular smooth muscle cell proliferation and migration. Circ Res 86:15–23. Yamboliev IA, Chen J, Gerthoffer WT 2001 PI 3-kinases and Src kinases regulate spreading and migration of cultured VSMCs. Am J Physiol Cell Physiol 281:C709–C718. Simoncini T, Hafezi-Moghadam A, Brazil DP, Ley K, Chin WW, Liao JK 2000 Interaction of oestrogen receptor with the regulatory subunit of phosphatidylinositol-3-OH kinase. Nature 407:538–541. Okamura WH, Midland MM, Hammond MW, Rahman NA, Dormanen MC, Nemere I, Norman AW 1994 Conformation and related topological features of vitamin D: Structure– function relationships. In: Vitamin D, A pluripotent steroid hormone: structural studies, molecular endocrinology and clinical applications. Norman AW, Bouillon R, Thomasset M (eds) Walter de Gruyter, Berlin. Corvol M, Ulmann A, and Garabedian M 1980 Specific nuclear uptake of 24,25-dihydroxycholecalciferol, a vitamin D3 metabolite biologically active in cartilage. FEBS Lett 116:273–276.

404 118. Pedrozo HA, Schwartz Z, Rimes S, Sylvia VL, Nemere I, Posner GH, Dean DD, Boyan BD 1999 Physiological importance of the 1,25(OH)2D3 membrane receptor and evidence for a membrane receptor specific for 24,25(OH)2D3. J Bone Miner Res 14:856–867. 119. Kato A, Seo EG, Einhorn TA, Bishop JE, Norman AW 1998 Studies on 24R,25-dihydroxyvitamin D3: Evidence for a non-nuclear membrane receptor in the chick tibial fracturehealing callus. Bone 23:141–146. 120. Kajikawa M, Ishida H, Fujimoto S, Mukai E, Nishimura M, Fujita J, Tsuura Y, Okamoto Y, Norman AW, Seino Y 1999 An insulinotropic effect of vitamin D analog with increasing intracellular Ca2+ concentration in pancreatic β-cells through nongenomic signal transduction. Endocrinology 140: 4706–4712. 121. Rebsamen MC, Sun J, Norman AW, Liao JK 2002 1α,25dihydroxyvitamin D3 induces vascular smooth muscle cell migration via activation of phosphatidylinositol 3-kinase. Circ Res 91:17–24. 122. Norman AW, Song XD, Bishop JE, Okamura WH, Ishizuka S 2000 1α,25(OH)2-vitamin D3 mediated rapid and genomic responses in NB4 cells: Evidence for cross-talk from rapid responses to genomic effects. In: Vitamin D Endocrine System. Structural, Biological, Genetic and Clinical Aspects. University of California, Riverside, Riverside, CA. 123. Shi H, Norman AW, Okamura WH, Sen A, Zemel MB 2002 1α,25-Dihydroxyvitamin D3 inhibits uncoupling protein 2 expression in human adipocytes. FASEB J 16:1808–1810. 124. Clark SA, Stumpf WE, Bishop CW, DeLuca HF, Park DH, Joh TH 1986 The adrenal—a new target organ of the calciotropic hormone 1,25-dihydroxyvitamin-D3. Cell Tissue Res 243:299–302. 125. Kream BE, Jose M, Yamada S, DeLuca HF 1977 A specific high-affinity binding macromolecule for 1,25-dihydroxycholecalciferol in fetal bone. Science 197:1086–1088. 126. Chen TL, Hirst MA, Feldman D 1979 A receptorlike binding macromolecule for 1α,25-dihydroxycholecalciferol in cultured mouse bone cells. J Biol Chem 254: 7491–7494. 127. Bellido T, Girasole G, Passeri G, Yu X-P, Mocharla H, Jilka RL, Notides A, Manolagas SC 1993 Demonstration of estrogen and vitamin D receptors in bone marrow-derived stromal cells: Up-regulation of the estrogen receptor by 1,25-dihydroxyvitamin-D3. Endocrinology 133:553–562. 128. Walters MR, Fischette CT, Fetzer C, May B, Riggle PC, Tibaldo-Bongiorno M, Christakos S 1992 Specific 1,25dihydroxyvitamin D3 binding sites in choroid plexus. Eur J Pharmacol 13:309–311. 129. Stumpf WE, O’Brien LP 1987 1,25(OH)2 vitamin D3 sites of action in the brain. Histochemistry 87:393–406. 130. Bidmon H-J, Stumpf WE 1994 Distribution of target cells for 1,25-dihydroxyvitamin D3 in the brain of the yellow bellied turtle Trachemys scripta. Brain Res 640:277–285. 131. Colston KW, Berger U, Wilson P, Hadcocks L, Naeem I, Earl HM, Coombes RC 1988 Mammary gland 1,25-dihydroxyvitamin D3 receptor content during pregnancy and lactation. Mol Cell Endocrinol 60:15–22. 132. Berger U, McClelland RA, Wilson P, Greene GL, Haussler MR, Pike JW, Colston K, Easton D, Coombes RC 1991 Immunocytochemical determination of estrogen receptor, progesterone receptor, and 1,25-dihydroxyvitamin D3 receptor in breast cancer and relationship to prognosis. Cancer Res 51:239–244. 133. Harper KD, Iozzo RV, Haddad JG 1989 Receptors for and bioresponses to 1,25-dihydroxyvitamin D in a human

ANTHONY W. NORMAN

134.

135. 136.

137.

138. 139.

140.

141.

142. 143. 144.

145. 146. 147.

148. 149.

150. 151.

colon carcinoma cell line (HT-29). Metabolism 38(11): 1062–1069. Reichrath J, Rafi L, Müller SM, Mink D, Reitnauer K, Tilgen W, Schmidt W, Friedrich M 1998 Immunohistochemical analysis of 1,25-dihydroxyvitamin D3 receptor in cervical carcinoma. Histochem J 30:561–567. Miyaguchi S, Watanabe T 2001 The role of vitamin D3 receptor mRNA in the proliferation of hepatocellular carcinoma. Hepato-Gastroenterology 47:468–472. Kawa S, Yoshizawa K, Tokoo M, Imai H, Oguchi H, Kiyosawa K, Homma T, Nikaido T, Furihata K 1996 Inhibitory effect of 22-oxa-1,25-dihydroxyvitamin D3 on the proliferation of pancreatic cancer cell lines. Gastroenterology 110: 1605–1613. Folgueira MA, Federico MH, Roela RA, Maistro S, Katayama ML, Brentani MM 2000 Differential regulation of vitamin D receptor expression in distinct leukemic cell lines upon phorbol ester-induced growth arrest. Braz J Med Biol Res 33:559–568. Masood R, Nagpal S, Zheng T, Cai J, Tulpule A, Smith DL, Gill PS 2001 Kaposi sarcoma is a therapeutic target for vitamin D3 receptor agonist. Blood 96:3188–3194. Klaus G, Merke J, Eing H, Hügel U, Milde P, Reichel H, Ritz E, Mehls O 1991 1,25(OH)2D3 receptor regulation and 1,25(OH)2D3 effects in primary cultures of growth cartilage cells of the rat. Calcif Tissue Int 49:340–348. Balmain N, Hauchecorne M, Pike JW, Cuisinier-Gleizes P, Mathieu H 1993 Distribution and subcellular immunolocalization of 1,25-dihydroxyvitamin D3 receptors in rat epiphyseal cartilage. Cell Mol B 39:339–350. Berry JL, Farquharson C, Whitehead CC, Mawer EB 1996 Growth plate chondrocyte vitamin D receptor number and affinity are reduced in avian tibial dyschondroplastic lesions. Bone 19:197–203. Vandewalle B, Adenis A, Hornez L, Revillion F, Lefebvre J 1994 1,25-Dihydroxyvitamin D3 receptors in normal and malignant human colorectal tissues. Cancer Lett 86:67–73. Biroc SL 1969 A possible receptor site for a metabolite of vitamin D3 in the supernatant fraction of hen egg shell gland. [Journal unknown.] Striem S, Bar A 1991 Modulation of quail intestinal and egg shell gland calbindin (Mr 28,000) gene expression by vitamin D3, 1,25-dihydroxyvitamin D3 and egg laying. Mol Cell Endocrinol 75:169–177. Stumpf WE, Clark SA, O’Brien LP, Reid FA 1988 1,25(OH)2 vitamin D3 sites of action in spinal cord and sensory ganglion. Anat Embryo 177:307–310. Walters MR 1982 1,25-Dihydroxyvitamin D receptors in rat testes, epididymis, and uterus. Fed Proc 41(4):1165. Reichrath J, Schilli M, Kerber A, Bahmer FA, Czarnetzki BM, Paus R 1994 Hair follicle expression of 1,25-dihydroxyvitamin D3 receptors during the murine hair cycle. Brit J Dermatol 131:477–482. Haussler MR, Norman AW 1969 Chromosomal receptor for a vitamin D metabolite: A nutrition classic. Nutr Rev 43:181–183. Tsai HC, Norman AW 1972 Studies on calciferol metabolism. VIII. Evidence for a cytoplasmic receptor for 1,25-dihydroxyvitamin D3 in the intestinal mucosa. J Biol Chem 248:5967–5975. Duncan WE, Glass AR, Wray HL 1991 Estrogen regulation of the nuclear 1,25-dihydroxyvitamin D3 receptor in rat liver and kidney. Endocrinology 129:2318–2324. Segura C, Alonso M, Fraga C, Caballero T, guez C, Andrez F 1999 Vitamin D receptor ontogenesis in rat liver. Histochem Cell Biol 112:163–167.

CHAPTER 23 Function Relationships for Both the Ligand and Receptor(s)

152. Drocourt L, Ourlin JC, Pascussi JM, Maurel P, Vilarem MJ 2002 Expression of CYP3A4, CYP2B6, and CYP2C9 is regulated by the vitamin D receptor pathway in primary human hepatocytes. J Biol Chem. 277:25125–25132. 153. Nguyen TM, Guillozo H, Marin L, Dufour ME, Tordet C, Pike JW, Garabedian M 1990 1,25-Dihydroxyvitamin D3 receptors in rat lung during the perinatal period: Regulation and immunohistochemical localization. Endocrinology 127:1755–1762. 154. Miller CW, Morosetti R, Campbell MJ, Mendoza S, Koeffler HP 1997 Integrity of the 1,25-dihydroxyvitamin D3 receptor in bone, lung, and other cancers. Mol Carcinog 19: 254–257. 155. O’Connell TD, Simpson RU 1996 Immunochemical identification of the 1,25-dihydroxyvitamin D3 receptor protein in human heart. Cell Biol Int 20:621–624. 156. Weishaar RE, Simpson RU 1987 Vitamin D3 and cardiovascular function in rats. J Clin Invest 79:1–7. 157. Simpson RU, Thomas GA, Arnold AJ 1985 Identification of 1,25 dihydroxyvitamin D3 receptors and activities in muscle. J Biol Chem 260:8882–8891. 158. Boland R, Norman AW, Ritz E, Hasselbach W 1992 Presence of a 1,25-dihydroxyvitamin D3 receptor in chick skeletal muscle myoblasts. Biochem Biophys Res Commun 128: 305–311. 159. Chen TL, Feldman D 1981 Regulation of 1,25-dihydroxyvitamin D3 receptors in cultured mouse bone cells. J Biol Chem 256:5561–5566. 160. Walters MR, Rosen DM, Norman AW, Luben RA 1982 1,25Dihydroxyvitamin D receptors in an established bone cell line: Correlation with biochemical responses. J Biol Chem 257:7481–7484. 161. Reddy SV, Kurihara N, Menaa C, Landucci G, Forthal D, Koop BA, Windle JJ, Roodman GD 2001 Osteoclasts formed by measles virus-infected osteoclast precursors from hCD46 transgenic mice express characteristics of pagetic osteoclasts. Endocrinology 142:2898–2905. 162. Menaa C, Barsony J, Reddy SV, Cornish J, Cundy T, Roodman GD 2002 1,25-Dihydroxyvitamin D3 hypersensitivity of osteoclast precursors from patients with Paget’s disease. J Bone Miner Res 15:228–235. 163. Dokoh S, Donaldson CA, Marion SL, Pike JW, and Haussler MR 1983 The ovary: A target organ for 1α,25dihydroxyvitamin D3. Endocrinology 112(1):200–206. 164. Ahonen MH, Zhuang YH, Aine R, Ylikomi T, Tuohimaa P 2000 Androgen receptor and vitamin D receptor in human ovarian cancer: Growth stimulation and inhibition by ligands. Int J Cancer 86:40–46. 165. Henry HL, Norman AW 1975 Localization of 1,25dihydroxy-vitamin D3 in chick parathyroid glands. Biochem Biophys Res Commun 62:781–788. 166. Wecksler WR, Henry HL, Norman AW 1977 Studies on the mode of action of calciferol XIII: Subcellular localization of 1,25-dihydroxy vitamin D3 in chicken parathyroid glands. Arch Biochem Biophys 183:168–175. 167. Wecksler WR, Ross FP, Mason RS, Posen S, Norman AW 1980 Studies on the mode of action of calciferol. XXIV. Biochemical properties of the 1α,25-dihydroxyvitamin D3 cytoplasmic receptors from human and chick parathyroid glands. Arch Biochem Biophys 201:95–103. 168. Peterfy C, Tenenhouse A 1982 Vitamin D receptors in isolated rat parotid and acinar cells. Biochim Biophys Acta 721:158–163. 169. Pike JW, Gooze LL, Haussler MR 1980 Biochemical evidence for 1,25-dihydroxyvitamin D3 receptor macromolecules

170.

171. 172.

173.

174.

175. 176.

177. 178. 179.

180. 181.

182.

183. 184. 185.

186.

405

in parathyroid, pancreatic, pituitary, and placental tissues. Life Sci 26:407–414. Stumpf WE, Sar M, Reid FA, Tanaka Y, DeLuca HF 1979 Target cells for 1,25-dihydroxyvitamin D3 in intestinal tract, stomach, kidney, skin, pituitary, and parathyroid. Science 206:1188–1190. Haussler MR, Manolagas SC, Deftos LJ 1982 Receptor for 1,25-dihydroxyvitamin D3 in GH3 pituitary cells. J Steroid Biochem Mol Biol 16:15–19. Skowronski RJ, Peehl DM, Feldman D 1993 Vitamin D and prostate cancer: 1,25 Dihydroxyvitamin D3 receptors and actions in human prostate cancer cell lines. Endocrinology 132:1952–1960. Kivineva M, Bläuer M, Syvälä H, Tammela T, Tuohimaa P 1998 Localization of 1,25-dihydroxyvitamin D3 receptor (VDR) expression in human prostate. J Steroid Biochem Mol Biol 66:121–127. Miller GJ, Stapleton GE, Ferrara JA, Lucia MS, Pfister S, Hedlund TE, Upadhya P 1992 The human prostatic carcinoma cell line LNCaP expresses biologically active, specific receptors for 1α,25-dihydroxyvitamin D3. Cancer Res 52:515–520. Simpson RU, DeLuca HF 1980 Chacterization of a receptorlike protein for 1,25-dihydroxyvitamin D3 in rat skin. Proc Natl Acad Sci USA 77(10):5822–5826. Feldman D, Chen T, Cone C, Hirst M, Shani S, Benderli A, Hochberg Z 1982 Vitamin D resistant rickets with alopecia: Cultured skin fibroblasts exhibit defective cytoplasmic receptors and unresponsiveness to 1,25(OH)2D3. J Clin Endo 55(5):1020–1022. Pillai S, Bikle DD, Elias PM 1988 1,25-Dihydroxyvitamin D production and receptor binding in human keratinocytes varies with differentiation. J Biol Chem 263:5390–5395. Segaert S, Degreef H, Bouillon R 2000 Vitamin D receptor expression is linked to cell cycle control in normal human keratinocytes. Biochem Biophys Res Commun 279:89–94. Chen CH, Sakai Y, Demay MB 2001 Targeting expression of the human vitamin D receptor to the keratinocytes of vitamin D receptor null mice prevents alopecia. Endocrinology 142: 5386–5389. Habib FK, Maddy SQ, Gelly KJ 1990 Characterisation of receptors for 1,25-dihydroxyvitamin D3 in the human testis. J Steroid Biochem Mol Biol 35(2):195–199. Osmundsen BC, Huang HFS, Anderson MB, Christakos S, Walters MR 1989 Multiple sites of action of the vitamin D endocrine system: FSH stimulation of testis 1,25-dihydroxyvitamin D3 receptors. J Steroid Biochem Mol Biol 34(1–6): 339–343. Provvedini DM, Sakagami Y, Manolagas SC 1989 Distinct target cells and effects of 1α,25-dihydroxyvitamin D3 and glucocorticoids in the rat thymus gland. Endocrinology 124:1532–1538. Stumpf WE, Downs TW 1987 Nuclear receptors for 1,25(OH)2 vitamin D3 in thymus reticular cells studied by autoradiography. Histochemistry 87:367–369. Berg JP, Torjesen PA, Haug E 1991 Calcitriol receptors in rat thyroid follicular cells (FRTL-5). Acta Endocr (Copenh) 125:574–580. Danan J, Delorme A, Cuisinier-Gleizes P 1981 Biochemical evidence for a cytoplasmic 1α,25-dihydroxyvitamin D3 receptor-like protein in rat yolk sac. J Biol Chem 256: 4847–4850. Stumpf WE, Sar M 1975 Autoradiograph techniques for localizing steroid hormones. Methods Enzymol 36: 135–140.

406 187. Shi H, Norman AW, Okamura WH, Sen A, Zemel MB 2001 1α,25-Dihydroxyvitamin D3 modulates human adipocyte metabolism via nongenomic action. FASEB J 15: 2751–2753. 188. Abe E, Miyaura C, Sakagami H, Takeda M, Konno K, Yamazaki T, Yoshiki S, Suda T 1981 Differentiation of mouse myeloid leukemia cells induced by 1α,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 78:4990–4994. 189. Mangelsdorf DJ, Koeffler HP, Donaldson CA, Pike JW, Haussler MR 1984 1,25-Dihydroxyvitamin D3-induced differentiation in a human promyelocytic leukemia cell line (HL-60): Receptor-mediated maturation to macrophage-like cells. J Cell Biol 98:391–398. 190. Park WH, Seol JG, Kim ES, Hyun JM, Jung CW, Lee CC, Binderup L, Koeffler HP, Kim BK, Lee YY 2000 Induction of apoptosis by vitamin D3 analogue EB1089 in NCI-H929 myeloma cells via activation of caspase 3 and p38 MAP kinase. Br J Haematol 109:576–583. 191. Manolagas SC, Provvedi DM, Murray EJ, Tsoukas CD, Deftos LJ 1986 The antiproliferative effect of calcitriol on human peripheral-blood mononuclear-cells. J Clin Endocr Metab 63:394–400. 192. Elstner E, Lee YY, Hashiya M, Pakkala S, Binderup L, Norman AW, Okamura WH, Koeffler HP 1994 1α,25Dihydroxy-20-epi-vitamin D3: An extraordinarily potent inhibitor of leukemic cell growth in vitro. Blood 84:1960–1967. 193. Zhou JY, Norman AW, Akashi M, Chen D-L, Uskokovic MR, Aurrecoechea JM, Dauben WG, Okamura WH, and Koeffler HP 1991 Development of a novel 1,25(OH)2vitamin D3 analog with potent ability to induce HL-60 cell differentiation without modulating calcium metabolism. Blood 78:75–82. 194. van Etten E, Branisteanu DD, Verstuyf A, Waer M, Bouillon R, Mathieu C 2000 Analogs of 1,25-dihydroxyvitamin D3 as dose-reducing agents for classical immunosuppressants. Transplantation 69:1932–1942. 195. Song X, Bishop JE, Okamura WH, Norman AW 1998 Stimulation of phosphorylation of mitogen-activated protein kinase by 1α,25-dihydroxyvitamin D3 in promyelocytic NB4 leukemia cells: A structure–function study. Endocrinology 139:457–465. 196. Norman AW, Frankel BJ, Heldt AM, Grodsky GM 1980 Vitamin D deficiency inhibits pancreatic secretion of insulin. Science 209:823–825. 197. Cade C, Norman AW 1987 Rapid normalization/stimulation by 1,25(OH)2-vitamin D3 of insulin secretion and glucose tolerance in the vitamin D–deficient rat. Endocrinology 120:1490–1497. 198. Schwartz Z, Shaked D, Hardin RR, Gruwell S, Dean DD, Sylvia VL, Boyan BD 2003 1α,25(OH)2D3 causes a rapid increase in phosphatidylinositol-specific PLC-beta activity via phospholipase A2 dependent production of lysophospholipid. Unpublished. 199. Schwartz Z, Sylvia VL, Larsson D, Nemere I, Casasola D, Dean DD, Boyan BD 2002 1α,25(OH)2D3 regulates chondrocyte matrix vesicle protein kinase C (PKC) directly via G-protein-dependent mechanisms and indirectly via incorporation of PKC during matrix vesicle biogenesis. J Biol Chem 277:11828–11837. 200. Schwartz Z, Ehland H, Sylvia VL, Larsson D, Hardin RR, Bingham V, Lopez D, Dean DD, Boyan BD 2002 1α,25Dihydroxyvitamin D3 and 24R,25-dihydroxyvitamin D3 modulate growth plate chondrocyte physiology via protein

ANTHONY W. NORMAN

201.

202.

203.

204.

205. 206.

207. 208.

209.

210.

211.

212.

213. 214.

kinase C–dependent phosphorylation of extracellular signal– regulated kinase 1/2 mitogen-activated protein kinase. Endocrinology 143:2775–2786. Schwartz Z, Sylvia VL, Del Toro F, Hardin RR, Dean DD, Boyan BD 2000 24R,25-(OH)2D3 mediates its membrane receptor-dependent effects on protein kinase C and alkaline phosphatase via phospholipase A(2) and cyclooxygenase-1 but not cyclooxygenase-2 in growth plate chondrocytes. J Cell Physiol 182:390–401. Boyan BD, Sylvia VL, Curry D, Chang Z, Dean DD, Schwartz Z 1998 Arachidonic acid is an autocoid mediator of the differential action of 1,25-(OH)2D3 and 24,25-(OH)2D3 on growth plate chondrocytes. J Cell Physiol 176:516–524. Bissonnette M, Wali RK, Hartmann SC, Niedziela SM, Roy HK, Tien X-Y, Sitrin MD, Brasitus TA 1995 1,25Dihydroxyvitamin D3 and 12-O-tetradecanoyl phorbol 13-acetate cause differential activation of Ca2+-dependent and Ca2+-independent isoforms of protein kinase C in rat colonocytes. J Clin Invest 95:2215–2221. Bettoun DJ, Buck DW, Lu J, Khalifa B, Chin WW, Nagpal S 2002 A vitamin D receptor-Ser/Thr phosphatase-p70 S6 kinase complex and modulation of its enzymatic activities by the ligand. J Biol Chem 277:24847–24850. Buitrago C, Gonzalez PV, De Boland AR 2002 Nongenomic action of 1α,25(OH)2-vitamin D3. Eur J Biochem 269: 2506–2515. Hmama Z, Nandan D, Sly L, Knutson KL, Herrera-Velit P, Reiner NE 1999 1α,25-Dihydroxyvitamin D3–induced myeloid cell differentiation is regulated by a vitamin D receptor– phosphatidylinositol 3-kinase signaling complex. J Exp Med 190:1583–1594. Barsony J, Marx SJ 1991 Rapid accumulation of cyclic GMP near activated vitamin D receptors. Proc Natl Acad Sci USA 88:1436–1440. De Boland AR, Norman AW 1990 Evidence for involvement of protein kinase C and cyclic adenosine 3′,5′-monophosphatedependent protein kinase in the 1,25-dihydroxyvitamin D3– mediated rapid stimulation of intestinal calcium transport (transcaltachia). Endocrinology 127:39–45. De Boland AR, Norman AW 1990 Influx of extracellular calcium mediates 1,25-dihydroxyvitamin D3–dependent transcaltachia (the rapid stimulation of duodenal Ca2+ transport). Endocrinology 127:2475–2480. Khare S, Tien X-Y, Wilson D, Wali RK, Bissonnette BM, Scaglione-Sewell B, Sitrin MD, Brasitus TA 1994 The role of protein kinase-Cα in the activation of particulate guanylate cyclase by 1α,25-dihydroxyvitamin D3 in CaCo-2 cells. Endocrinology 135:277–283. Bissonnette M, Tien X-Y, Niedziela SM, Hartmann SC, Frawley BP Jr, Roy HK, Sitrin MD, Perlman RL, Brasitus TA 1994 1,25(OH)2 vitamin D3 activates PKC-α in Caco-2 cells: A mechanism to limit secosteroid-induced rise in [Ca2+]i. Am J Physiol Gastrointest Liver Physiol 267: G465-G475. Yada Y, Ozeki T, Meguro S, Mori S, Nozawa Y 1989 Signal transduction in the onset of terminal keratinocyte differentiation induced by 1α,25-dihydroxyvitamin D3: Role of protein kinase C translocation. Biochem Biophys Res Commun 163:1517–1522. Simboli-Campbell M, Franks DJ, Welsh JE 1992 1,25(OH)2D3 increases membrane associated protein kinase C in MDBK cells. Cell Signal 4:99–109. Slater SJ, Kelly MB, Taddeo FJ, Larkin JD, Yeager MD, McLane JA, Ho C, Stubbs CD 1995 Direct activation of

CHAPTER 23 Function Relationships for Both the Ligand and Receptor(s)

215.

216.

217.

218.

219.

220. 221.

protein kinase C by 1α,25-dihydroxyvitamin D3. J Biol Chem 270:6639–6643. Baran DT, Sorensen AM, Honeyman RW, Ray R, Holick MF 1990 1α,25-Dihydroxyvitamin D3–induced increments in hepatocyte cytosolic calcium and lysophosphatidylinositol: Inhibition by pertussis toxin and 1β,25-dihydroxyvitamin D3. J Bone Miner Res 5:517–524. Beno DWA, Brady LM, Bissonnette M, Davis BH 1995 Protein kinase C and mitogen-activated protein kinase are required for 1,25-dihydroxyvitamin D3–stimulated Egr induction. J Biol Chem 270:3642–3647. Sellés J, Boland R 1991 Evidence on the participation of the 3′,5′-cyclic AMP pathway in the non-genomic action of 1,25-dihydroxy-vitamin D3 in cardiac muscle. Mol Cell Endocrinol 82:229–235. Zanello LP, Norman AW 1996 1α,25(OH)2 vitamin D3– mediated stimulation of outward anionic currents in osteoblast-like ROS 17/2.8 cells. Biochem Biophys Res Commun 225:551–556. Bourdeau A, Atmani F, Grosse B, Lieberherr M 1990 Rapid effects of 1,25-dihydroxyvitamin D3 and extracellular Ca2+ on phospholipid metabolism in dispersed porcine parathyroid cells. Endocrinology 127:2738–2743. Sugimoto T, Ritter C, Ried I, Morrissey J, Slatopolsky E 1988 Effect of 1,25-dihydroxyvitamin D3 on cytosolic calcium in dispersed parathyroid cells. Kidney Int 33:850–854. Bhatia M, Kirkland JB, Meckling-Gill KA 1995 Monocytic differentiation of acute promyelocytic leukemia cells in response to 1,25-dihydroxyvitamin D3 is independent of nuclear receptor binding. J Biol Chem 270:15962–15965.

407

222. Bhatia M, Kirkland JB, Meckling-Gill KA 1996 1,25Dihydroxyvitamin D3 primes acute promyelocytic cells for TPA-induced monocytic differentiation through both PKC and tyrosine phosphorylation cascades. Exp Cell Res 222:61–69. 223. Biskobing DM, Rubin J 1993 1,25-Dihydroxyvitamin D3 and phorbol myristate acetate produce divergent phenotypes in a monomyelocytic cell line. Endocrinology 132:862–866. 224. Chambliss KL, Yuhanna IS, Anderson RG, Mendelsohn ME, Shaul PW 2002 ERβ has nongenomic action in caveolae. Mol Endocrinol 16:938–946. 225. Huhtakangas JA, Olivera CJ, Bishop JE, Zanello LP, Norman AW. The vitamin D receptor is present in caveolae-enriched plasma membranes and binds 1,25(OH)2-vitamin D3 in vivo and in vitro. Mol Endocrinol. In Press. 2004. 226. Mizwicki MT, Bula CM, Keidel D, Bishop JE, Zanello LP, Wurtz JM, Moras D, Norman AW 2004 Identification of a novel alternative ligand binding pocket in the nuclear vitamin D receptor and its functional importance in 1,25(OH)2vitamin D3 cellular signaling. Proc Nat Acad Sci USA. In Press. 227. Norman AW, Mizwicki MT, Norman DPG 2004 Steroid hormone rapid actions, membrane receptors and a conformational ensemble model. Nature Reviews Drug Discovery 3:27–41. 228. Kousteni S, Bellido T, Plotkin LI, O'Brien CA, Bodenner DL, Han L, Han K, DiGregorio GB, Katzenellenbogen JA, Katzenellenbogen BS, Roberson PK, Weinstein RS, Jilka RL, Manolagas SC 2001 Nongenotropic, sex-nonspecific signaling through the estrogen or androgen receptors: dissociation from transcriptional activity. Cell 104:719–730.

CHAPTER 24

Vitamin D and the Intestinal Absorption of Calcium: A View and Overview ROBERT H. WASSERMAN I. II. III. IV. V.

Department of Biomedical Sciences, Cornell University, Ithaca, New York

Introduction The Vitamin D Hormone Overview Transcellular Calcium Absorption Vitamin D and the Paracellular Path

I. INTRODUCTION Calcium is the fifth most prominent element in the human body, constituting 0.3% of the total number of atoms therein and about 2% by weight, with 99% of this calcium present in the skeleton in the form of highly insoluble calcium phosphate crystals [1]. The calcium mineral deposited into the organic matrix of bone contributes to the rigidity and strength of bone and is a storehouse of calcium that can be removed to maintain the circulating levels of calcium within the normal range of about 2.5 mM (10 mg per dl). Besides bone formation, calcium is essential for the functioning of a multitude of physiological and biochemical processes. As shown by Sydney Ringer in the latter part of the 19th century, the calcium ion is essential for heart contractility, for the development of fertilized eggs, and for cell adhesion. In addition, other functions requiring calcium include blood coagulation, neuromuscular excitability, nerve impulse transmission, the maintenance of the permeability of cell and cellular membranes, enzyme activation, and the secretion and functioning of hormones [2]. The only way that calcium ions gain access to the body to serve these functions is by the ingestion and absorption of dietary calcium. The secosteroid vitamin D is required in most vertebrate species for the optimization and regulation of the calcium absorption process. The importance of vitamin D in human and animal biology is emphasized by the consequences of its lack. A deficiency of vitamin D predisposes the young growing individual to the disease of rickets and, in the adult, to the disease of osteomalacia. Rickets is characterized VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

VI. Vesicular Transport of Calcium VII. Commentary on Segment-Specific Intestinal Absorption of Calcium VIII. Summary References

by hypocalcemia and/or hypophosphatemia, bone deformities of the long bones, scoliosis of the spine, and enlargement of the costrochondral junctions (“rachitic rosary”) [3]. In the adult, vitamin D deficiency leads to the disease of osteomalacia, which is a failure of mineralization of osteoid produced during bone remodeling. The earlier studies by investigators such as E. Mellanby, R. Nicolaysen, E. Kodicek, A. Carlsson, and B. Lindquist established that an important site of vitamin D action was the intestinal tract and the proper absorption of calcium (see Chapter 1 for historical notes). A number of uncertainties, as one might expect, surrounded these pioneer studies. Hypotheses were offered that vitamin D directly affected calcium absorption and secondarily phosphate absorption, or that the direct effect was on phosphate absorption, with calcium secondarily affected. No special physiological and biochemical mechanisms to explain how vitamin D effects the absorption of calcium were evident at those times. In recent years, a considerable amount of research has been accomplished that has advanced our understanding of these transport processes, but a full description of the absorption processes is still lacking. This chapter describes some of the earlier as well as more current information on calcium absorption, emphasizing the role of vitamin D. The amount of material to be included in this chapter was necessarily limited, and important contributions unfortunately had to be omitted. However, the reader is referred to various reviews [4–14] for more detailed discussions and contrasting views on the role of vitamin D in specific epithelial transport processes. Copyright © 2005, Elsevier, Inc. All rights reserved.

412

ROBERT H. WASSERMAN

II. THE VITAMIN D HORMONE

16

As discussed in detail in this volume, vitamin D, either from dietary sources or by endogenous synthesis following ultraviolet irradiation of the skin, is transformed into its most active hormonal form, 1,25-dihydroxyvitamin D [1,25(OH)2D], by two sequential hydroxylation reactions [14]. The first, 25-hydroxylation, takes place in the liver, and the second, lα-hydroxylation, occurs in the kidney. The rate of renal synthesis of 1,25(OH)2D3 is feedback regulated. The metabolite, 1,25(OH)2D3, often referred to as the vitamin D hormone, with PTH and calcitonin constitute the traditional calciotropic hormones that maintain serum calcium within narrow limits through effects, direct or indirect, on the intestine, kidney, and skeleton. In addition to its homeostatic function, the vitamin D hormone also influences a number of other biological systems, as discussed elsewhere in this volume. In intestine, kidney, and other calcium-transporting organs, 1,25(OH)2D3 elicits genomic effects, inducing the synthesis of specific proteins following its interaction with the nuclear vitamin D receptor (VDRnuclear) (refer to Chapters 11–14 for details). Nongenomic effects of 1,25(OH)2D3 have also been described which are proposed to operate through activation of a membraneassociated vitamin D receptor (VDRmembrane) (refer to Chapter 23 by A.W. Norman for details).

14

III. OVERVIEW A. Relation of Calcium Intake to Calcium Absorption The duality of the mechanism of calcium absorption was clearly demonstrated in earlier studies on the relationship between dietary calcium intake (or intraluminal calcium concentrations) and the amount or rate of calcium absorption. The form of this relationship is shown in Fig. 1 [15], and a theoretical description of this relationship is depicted in Fig. 2 [16]. The curve can be resolved into two components. One component of calcium absorption is saturable and follows Michaelis– Menten-type kinetics. The other mode, diffusional absorption, is nonsaturable, with the amount of absorption a direct function of the luminal concentration of calcium. The equation describing this phenomenon, in Michaelis–Menten terms, is as follows: Jms = {(Jmax)[Ca]}/{[Ca] + Kt} + D[Ca]

EFFLUX

12

Vitamin D3 10

Calcium flux (µ moles / cm. hour)

8

6 4

Rachitic

2

0

Rachitic 2

4 INFLUX

Vitamin D3

6 0

20

40 60 80 100 Intraluminal Ca conc. (mM)

120

140

FIGURE 1

Effect of vitamin D3 and intraluminal calcium concentration on the unidirectional fluxes of calcium across chick duodenum in vivo. Efflux represents transfer of calcium from lumen to blood and influx from blood to lumen. Note that the vitamin D increased both calcium outflux and influx and increased both the saturable and nonsaturable mode of absorption (see Fig. 2). Each point represents the mean of experiments in five to six chicks. (From Wasserman and Kallfelz [15]).

calcium absorbed by a saturable process; [Ca] = calcium concentration in lumen (or the dietary calcium intake); Kt = binding constant of calcium to the ratelimiting component of the saturable process; and D = proportionality factor that relates calcium absorption to calcium intake or luminal calcium concentrations. As examples, reference is made to two reports relating calcium absorption to calcium intake in the human. Wilkinson [17] collected data from the literature on calcium balance studies of healthy individuals. Their ages varied from 19 to 83 years. The equation relating calcium intake to net calcium absorbed in the Wilkinson study, as given by Nordin and Marshall [18], is:

(1)

y = [(491x)/(287 + x)] + 0.06x − 206 (mg/day) (2)

where Jms = total calcium flux across the intestinal membranes (mucosal to serosal); Jmax = maximum flux of

where x is dietary calcium and y is net calcium absorbed. The units are mg/day. The value of 206 mg/day

CHAPTER 24 Vitamin D and the Intestinal Absorption of Calcium: A View and Overview

5 I+II

Jms =

Ca Jms (Arbitrary units)

4

(Jmax )[Ca] + D[Ca] [Ca] + Kt I+II

3

II

Jms = D[Ca] II

2

I

Jms =

1

0

1

(Jmax )[Ca] [Ca] + Kt

2 3 4 Intraluminal Ca concentration (Arbitrary units)

5

I

6

FIGURE 2 Theoretical curves for calcium flux across intestinal membranes (mucosal to serosal). Curve I (JI) represents the active component of the process and shows typical saturation kinetics. The second component, curve II (JII), represents simple diffusion and is a function of the diffusion coefficient of the system (D) and the calcium concentration in the precursor compartment (Ca). The dashed line (I + II) represents the sum of the two separate processes and is not unlike experimentally observed results (see Fig. 1). (From Wasserman and Taylor [16].)

represents the fecal excretion of calcium when calcium intake is zero. Heaney et al. [19] examined the relationship between calcium intake and calcium absorption in 180 healthy individuals. Included in the study were nuns of average age of about 41 years and a group of men and women of about 30–52 years of age. The calcium absorption (Ca Abs) equation that gave the best fit included an exponential term for the saturable component, as shown below: Ca Abs = 0.3127[exp(−1.0539CaD)]CaD + 0.1541CaD (g/day)

(3)

where CaD = grams of calcium ingested per day (g/day). The diffusion, nonsaturable factor in the Heaney study differs from that of the Wilkinson study, which might be due to differences in experimental technique and the different makeup of the study groups.

B. Relative Absorption of Calcium by Segments of the Small Intestine The degree of calcium absorption in any particular segment of the gastrointestinal tract is determined by the rate of absorption in that segment and the transit

413

time spent in that segment. The latter is related to segment length. Using the rat as experimental animal, Marcus and Lengemann [20,20a] approached this question by first determining transit rate along the small intestine with the radioactive nonabsorbable indicator yttrium-91. The transit half-time in the duodenum was 6 min for either a liquid or a solid meal. The transit half-time increased with progression aborally, maximizing in the distal small intestine at 102 min for the solid meal and 120 min for the liquid dose. With this information and data obtained after dosing rats with 45Ca-labeled liquid or solid meals, the degree of absorption in the various segments of the small intestine was estimated. With the labeled liquid dose, the duodenum, jejunum, and ileum absorbed 15%, 23%, and 62% of the ingested calcium, respectively. With labeled solid dose, the duodenum, jejunum, and ileum absorbed 8%, 4%, and 88%, respectively. In both cases, the stomach absorbed 0%. The relative sites of absorption of radiostrontium in the rat, as determined by Cramer and Copp [21], yielded similar results, with 7% of the effective absorption occurring in the duodenum, 17% in the jejunum, 65% in the ileum, and 2% in the stomach. In the dog, the relative effect sites of calcium absorption in the small intestine were 4% in duodenum, 16% in the jejunum, and 80% in the ileum [22]. More recently, Duflos et al. [23] essentially confirmed the earlier work, finding the sojourn time in the ileum was considerably longer than in the duodenum by a factor of about 55, and that most of the calcium that was absorbed was absorbed in the ileum. Thus, in these mammals, the sites where most of the ingested calcium was absorbed were the lower segments of the small intestine, and only about 10% of calcium absorption takes place in the duodenum despite its capacity to absorb calcium at a more rapid rate than any other segment. Undoubtedly a similar situation pertains to the human. This emphasizes the need to understand more thoroughly the mechanisms of calcium absorption that occurs in the mammalian ileum where most of the dietary calcium is absorbed. This is discussed more fully in Section VII, A. In avian species, the intestinal regions where most of the calcium is absorbed are the jejunum and duodenum with the lowest relative absorption taking place in the lower regions of the small intestine [24,25].

C. Relevant Structural Features of the Enterocyte The absorption of nutrients, including calcium, occurs by their movement across the epithelium cell

414 layer of the intestinal tract, which is composed of absorbing enterocytes, goblet cells, and other cell types. The surface of the absorbing enterocyte that faces the intestinal lumen is the brush border membrane, the microvillus. Attached to it exteriorly is the glycocalyx, comprised of anionic glycoproteins and mucopolysaccharides. Within the core of the microvillus are microtubules that contribute to the maintenance of its structure and integrity. These microtubules extend into the dense terminal web region located subjacent to the microvillus. Beneath the terminal web and within the cytosolic compartment are the usual intracellular organelles: mitochondria, endoplasmic reticulum, and nucleus. The basolateral membrane defines the exterior basal and lateral cell boundaries. The lamina propria is the structure beneath the enterocyte which contains blood capillaries and the lymphatic vessels that transfer absorbed nutrients parenterally. With this brief description of the intestinal epithelium, it is apparent that the transcellular absorption of calcium is a complex process, requiring the movement of calcium from the intestinal lumen through the glycocalyx region, across the brush border membrane, into and through the cell interior (with the possibility of transient sequestration by intracellular organelles), followed by the mediated extrusion from the cell across the basolateral membrane into the extracellular fluid of the lamina propria. The capillaries resident in the lamina propria transfer the newly absorbed ion into the main circulatory system for distribution of calcium throughout the body proper. The nonsaturable, diffusional absorption of calcium moves through the paracellular path that exists between adjacent cells of the epithelial cellular membrane.

IV. TRANSCELLULAR CALCIUM ABSORPTION A. Thermodynamic Parameters The physicochemical parameters that pertain to transcellular calcium transport are the intracellular ion calcium concentration of the order of 10−7 M, and an intracellular electropotential of about −50 mV with respect to the intestinal lumen. Thus, the calcium ion, usually in the millimolar range within the intestinal lumen, can enter the enterocyte by diffusion down a steep electrochemical gradient. The blood side of the epithelial cellular membrane has a potential of about +6 mV with respect to the intestinal lumen, and therefore the electropotential gradient from cell interior to lamina propria is about +56 mV. There is also a considerable calcium concentration

ROBERT H. WASSERMAN

gradient with the intracellular Ca2+ at 10−7 M and extracellular unbound Ca2+ at 1.25 × 10−3 M Ca2+. Thus, an input of energy is required to transport intracellular Ca2+ uphill against a considerable electrochemical gradient into the extracellular fluid within the lamina propria.

B. Evidence for the Active Intestinal Transport of Calcium The active intestinal calcium transport across the intestine was first shown by Schachter [26], using the in vitro everted gut sac technique. Calcium in the mucosal (lumen) side was transported into the serosal compartment uphill against a concentration gradient. The active transport process was highly dependent on the vitamin D status of the donor animal. The presence of an active Ca2+ transport system was also demonstrated in the intact animal [16]. The unidirectional Ca2+ fluxes across the intestine of the rat were measured, and knowing the electrochemical potential difference across the duodenal segment, fluxes could be subjected to analysis by the method of Ussing (see Wasserman and Taylor [16] for a more complete discussion and references). The analyses showed that calcium was absorbed by an active, energy-requiring mechanism in both the duodenum and the lower segments of the rat small intestine. This was also demonstrated by others [27,28]. A similar approach was employed with chick intestine in situ and the presence of active calcium transport by chick duodenum was demonstrated [15]. In addition, it was shown that the uphill transport of Ca2+ was enhanced by vitamin D repletion of vitamin D–deficient chicks, verifying the earlier findings of Schachter [26].

C. Vitamin D and Calcium Entry, the First Step in Transcellular Calcium Absorption 1. VITAMIN D AND BRUSH BORDER PERMEABILITY

An increase in the permeability of the microvillar (brush border) membrane of the intestine by vitamin D was demonstrated more than 40 years ago with the in vitro everted gut sac procedure [26] and with the ligated loop procedure in the intact animal [15]. An enhanced calcium uptake by isolated brush border membrane vesicles from vitamin D–adequate animals as compared to vesicles from vitamin D–deficient animals was shown by several groups; among those were Bikle et al. [29] and Fontaine et al. [30].

CHAPTER 24 Vitamin D and the Intestinal Absorption of Calcium: A View and Overview

2. MOLECULAR BASES OF CALCIUM ENTRY

a. Calcium Entry Channels1 Hoenderop and colleagues [31], using expression cloning technology, identified a protein from rabbit kidney associated with maximal Ca2+ uptake by a Xenopus oocyte test system. The protein was appropriately named epithelial calcium channel (ECaC). Sequencing of the ECaC cDNA revealed a protein comprising 730 amino acids with a molecular weight of about 83 kDa. Topolographic analysis predicted the presence of three structural domains: a hydrophobic domain composed of six transmembrane segments, and two hydrophilic domains. The C-terminal hydrophilic domain contained two potential phosphorylation sites. The N-terminal domain had ankyrin binding repeats and several phosphorylation sites. Besides the kidney, ECaC mRNA was highly expressed in rabbit duodenum, less in jejunum, but not in the ileum. The apparent affinity constant for Ca2+ uptake by ECaC is about 0.2 mM. Peng et al. [32], using a similar Xenopus oocyte expression system, screened a rat duodenal library and identified a cDNA encoding another epithelial calcium transport protein (CaT1) that was closely related to ECaC. The CaT1 cDNA encodes a protein of 727 amino acids with a predicted relative mass of 83.2 kDa, similar to the Hoenderop ECaC protein. The topology of CaT1 was also similar. CaT1 mRNA, like ECaC mRNA, was expressed in duodenum, proximal jejunum but not in ileum or distal jejunum. CaT1 mRNA, unlike ECaC mRNA, was not expressed in kidney. The apparent affinity constant for Ca2+ uptake by CaT1 is about 0.4 mM. Screening a rat kidney cortex library with a CaT1 probe, Peng et al. (33) isolated a cDNA encoding a protein with 84.2% homology with rabbit kidney ECaC. This protein, CaT2, localized specifically in the kidney and was not expressed in intestine or any other tissue examined. The Michaelis constant (Km) was 0.66 mM when expressed in Xenopus oocytes. By in situ hybridization, CaT2 (also known as ECaC) was reported to colocalize with calbindin-D28K in distal renal tubules. Similar epithelial calcium channels were identified in human intestine, with CaT1 transcripts highly expressed in duodenum and less so in stomach and jejunum [34]. No CaT1 mRNA was expressed in the 1The epithelial calcium channel identified by Hoenderop et al. [31]

is known as ECaC (original designation), CaT2, and TRPV5. The channel identified by Peng et al. [32] is known as CaT1 (original designation), ECaC2, and TRPV6. The TRP designations indicate a close relationship to the transient receptor potential (TRP) channel proteins.

415

ileum, but very low levels were seen in ascending but not descending colon. Calcium transport studies done on oocytes injected with human CaT1 cDNA showed Ca2+ uptake that was concentration dependent, saturable, and with an apparent Michaelis constant (Km) of 0.25 mM. The apparent Michaelis constant of the rat CaT1 was higher, at 0.44 mM [32]. Muller et al. [35] also obtained the full-length cDNA of human ECaC (hECaC) with properties very similar to those of rabbit ECaC. hECaC was expressed in duodenum, jejunum, colon, and rectum, but again not in ileum. Barley et al. [36] identified a cDNA from human duodenum giving a predicted amino acid sequence with a 97% homology with rat CaT1. This protein, named ECaC2 (CaT1), was present in duodenum but not ileum, colon, or kidney. ECaC2 was not correlated with 1,25(OH)2D3 levels, suggesting that the formation of ECaC2 (CaT1) was not vitamin D–dependent in the human. This appears to conflict with the observations of others who have demonstrated a vitamin D dependency of channel synthesis. Caco-2 cells are derived from a colon adenocarcinoma cell line, which differentiates in culture and at confluence assumes the characteristics of a welldifferentiated small intestine–like absorbing cellular membrane [37]. These cells have a vitamin D receptor (VDR) and, in response to 1,25(OH)2D3, there is an increase in transepithelial calcium transport and an increase in calbindin-D9K expression [38,39]. Of particular interest here is the recent study on the expression of the epithelial calcium channel, CaT1, in response to 1,25(OH)2D3 in this cell line [38]. CaT1 mRNA was present at low levels in the absence of 1,25(OH)2D3. After the addition of the hormone to the culture system, an increase in CaT1 mRNA occurred at 2 hr and further elevated with time. The rapid increase in the 1,25(OH)2D3-mediated expression of CaT1 mRNA preceded that of calbindin-D9K mRNA. b. Other Factors Possibly Involved in Calcium Entry i. Brush border membrane lipids. Rasmussen and colleagues [40] observed that 1,25(OH)2D3 altered the composition of brush border phospholipids. Using labeled precursors, the vitamin D hormone stimulated the synthesis of phosphatidylcholine (PC) with an accompanying decrease in the synthesis of phosphatidylethanolamine (PE), a response to 1,25(OH)2D3 that was independently verified [41]. The 1,25(OH)2D3 mediated lipid modification closely corresponded with an increase of Ca2+ uptake by isolated brush border membrane vesicles. There was also an increase in the number of polyunsaturated fatty acids groups in the PC fractions, which could result in an increase membrane fluidity, but neither Putkey et al. [42] using electron

416 spin resonance nor Bikle et al. [43] using fluorescence polarization were unable to observe an effect of vitamin D on the fluidity state of the brush border membranes. On the other hand, Brasitus et al. [44], using specific probes, did find an early effect of 1,25(OH)2D3 on brush border properties, but an increase in overall Ca2+ absorption was not discernible in the Brasitus study until much later, at 5 hr after 1,25(OH)2D3. These findings are consistent with other observations that an early modification of brush border membrane properties might increase Ca2+ entry, but a less rapid vitamin D–dependent reaction is required before an increase in overall calcium absorption can take place (cf. IV,C,3). ii. Alkaline phosphatase. This microvillar enzyme hydrolyzes a number of organic ortho- and pyrophosphates, and vitamin D treatment of deficient animals increases the activity of this enzyme [45–49]. Relating vitamin D–enhanced calcium transport with increases in alkaline phosphatase has not yielded a consistent and meaningful correspondence. For these reasons, the role of this brush border enzyme in vitamin D– mediated calcium absorption is yet to be clearly defined. iii. Intestinal membrane calcium-binding protein (IMCal). A particulate calcium-binding complex was identified in brush border membranes isolated from rat intestinal mucosa by Kowarski and Schachter [50,51]. The calcium-binding component of the complex, termed IMCal, was resolvable from an enzymatically active component and shown to have a molecular weight of about 200,000 by gel filtration. It was composed of subunits with a molecular weight of about 20,500, similar in size to a vitamin D–dependent protein described by Miller et al. [52]. The concentration of IMCal, as measured by its calcium binding activity, varied with factors that influence the degree of calcium absorption, including dietary calcium levels, age, intestinal distribution, and vitamin D status [53]. The proposal was put forth that IMCal functioned by mediating the transfer of calcium from intestinal lumen across the brush border membrane into the cytosolic compartment. The molecular weight data clearly differentiates IMCal from the epithelial calcium channels, CaT1 and ECaC. Unfortunately, little information on IMCal has been forthcoming since the appearance of the early reports. iv. Calmodulin. This calcium-binding protein, identified by W. Y. Cheung in 1970, activates many enzyme systems. This protein is present in relatively high concentration in the microvillar region of the small intestine [54–56]. Although there is no evidence that 1,25(OH)2D3 affects the synthesis of calmodulin, there are reports that the vitamin D hormone does

ROBERT H. WASSERMAN

cause a redistribution of calmodulin in the intestinal cell, increasing its association with the brush border complex [57,58] and specifically with a 105-kDa protein [58]. The 105-kDa protein of Bikle is most likely the 110-kDa protein-calmodulin complex studied by Mooseker and Coleman [59]. The 110-kDa protein is myosin I, an ATPase mechanoenzyme that tethers the F-actin filaments to the microvillar membrane, and it was speculated that the actin–myosin I–calmodulin complex via a Ca2+-dependent reaction might influence microvillar membrane permeability [60]. v. Protein sulfhydryl (-SH) groups. Selenite binds tightly to sulfhydryl groups of proteins, and it was shown that selenite uptake by brush border vesicles was increased by 1,25(OH)2D3 [61]. This was a rapid effect; a significant increase in selenite uptake was seen at 10 min after intravenous administration of 1,25(OH)2D3 to rachitic chicks [62]. Using a fluorescent probe, it was shown that feeding diets deficient in calcium or phosphorus to vitamin D–adequate chicks resulted in about a 2.5-fold increase in the mucosal concentration of calbindin-D28K and about a 3.1-fold increase in reactive sulfhydryl groups of isolated brush border membranes [63]. In the presence of sodium dodecyl sulfate (SDS), the number of reactive groups was increased by a factor of about 4.7, indicating that most of the available sulfhydryl groups were inaccessible to reagent and most likely buried within the lipid phase of the brush border membrane. Proteins prominently labeled by the fluorescent sulfhydryl probe had molecule masses of about 43,000 and 110,000. The sulfhydryl content of the 110,000-Da protein was positively affected by 1,25(OH)2D3. 1,25(OH)2D3-mediated increases in sulfhydryl groups occurred in intestinal and kidney brush border membranes, but not mitochondria, intestinal basolateral membranes, erythrocyte membranes, or cardiac sarcolemma, indicating some degree of specificity [64]. Other hormones also increase the number of available number of sulfhydryl groups [64], and these effects are yet to be explained. 3. CALCIUM ENTRY IS SOMETIMES LIMITING, SOMETIMES NOT

The calcium entry step can, in some situations, be the limiting step in overall calcium absorption and not in others. An example of the latter, shown in Fig. 3, comes from the study of the time course of events that take place when 1,25(OH)2D3 is administered to a vitamin D–deficient animal as reported by Fullmer [65]. Using the in situ ligated duodenal loop procedure, 47Ca placed in the gut lumen was shown to accumulate in the mucosal tissue rather rapidly, within 30 min, after the 1,25(OH)2D3 dose (Fig. 3). Only later, at 3–4 hr after hormone, was

417

CHAPTER 24 Vitamin D and the Intestinal Absorption of Calcium: A View and Overview

80

a

40 b 30 20

b

10 0

0 1

2 4 Time after 1,25(OH)2D3 (hr)

2.4

A

70

8

FIGURE 3

Intestinal absorption and tissue retention of luminally administered 47Ca2+ in vitamin D–deficient chicks at varying times following a single dose of 1,25(OH)2D3. Data are means ± SEM for six chicks. Lowercase letters represent statistically significant results: a denotes significant difference from the zero time control (p < 0.05) in 47Ca2+ tissue retention, and b denotes significant difference ( p < 0.05) in 47Ca2+ absorption above the control value. (From Fullmer et al. [82]; reprinted with permission.)

Duodenal 47Ca Absorption (% Dose)

47Calcium

(% dose)

50

Absorbed Retained

2.0

Absorption

60

1.6

50

1.2

40

0.8

CaBP

30

0.4 Trace

20

B

50

0

47Ca Absorption

2.0

40

1.6

30

1.2

20

0.8 CaBP

10

0.4

Trace

0

0 0

there an increase in overall 47Ca absorption. The 3 to 4-hr lag period coincided with the period required for the appearance of calbindin-D28K, the vitamin D–induced high-affinity calcium-binding protein, as shown in Fig. 4. (The identification, properties, and possible functions of the calbindins are more fully described in Section IV,D,1 and by Christakos, et al. in Chapter 42.) To illustrate a circumstance in which calcium entry was the limiting step, vitamin D–deficient chicks were given per os a dose of vitamin D one-third of their requirement [41]. A small amount of calbindin was synthesized, considerably less than is seen under more normal conditions and less than needed for the maximal absorption of calcium. 1,25(OH)2D3 given to these animals resulted in an increase in calcium absorption at 2 hrs whereas 3 to 4 hr is required for the synthesis of calbindin-D28K to become evident (Fig. 4), indicating in this case that the limiting step is calcium entry. The studies by the Belgian group [66] on the behavior of VDR knockout (VDR-KO) mice bears on this question, as summarized by Bouillon et al. [11]. The VDR knockout mice have characteristics expected of vitamin D–resistant individuals. It was reported that the calbindin-D9K in the duodenum of the Leuven strain of VDR-KO mice on a normal calcium diet was about the same as in the wild-type mice, but the expression of duodenal ECaC2 (CaT1) was considerably downregulated, as was the absorption of calcium. In this

47Ca

CaBP (µg/mg protein)

a

1

2 4 Time (Hrs.) after 1,25(OH)2D3

8

FIGURE 4 Effect of 1,25(OH)2D3 on duodenal absorption of 47Ca and on CaBP synthesis in rachitic chicks. (A) Absorption period, 30 min; 1,25(OH)2D3, 1 µg per chick. (B) Absorption period, 15 min; 1,25(OH)2D3, 0.3 µg per chick. The data represent the mean ± SEM for 47Ca absorption for five or six chicks. 1,25(OH)2D3 significantly increased 47Ca absorption at 4 and 8 hr (above the time 0 control; P < 0.01). (From Wasserman et al. [41].)

situation, it was proposed that calcium entry via the calcium channel(s) was the rate-limiting step in transcellular calcium transport [11].

D. Cytosolic Transfer of Calcium, the Second Step in Transcellular Calcium Absorption Two current models have been proposed to explain the vitamin D–enhanced transcytosolic movement of calcium. One is based on the proposal that the vitamin D– induced, high-affinity, calcium-binding proteins, the calbindins, serve as intracellular facilitators of calcium diffusion. The other proposes that intracellular calcium is transported in membrane-bound vesicles, the vesicular transport model. Each model could contribute to the protection of the cell against Ca2+-mediated cytotoxicity by sequestration of in-transit cytosolic Ca2+.

418 1. THE CALBINDINS AND THE CALBINDIN-BASED MODEL OF INTESTINAL CALCIUM TRANSPORT

a. The Calbindins The calbindins were first identified in chick intestine in 1966 [67] and then in rat intestine in 1967 [68]. The calbindins have been shown to be widely distributed in many species and present in many different tissues, including kidney, pancreas, and brain [8,69]. Many properties and characteristics of the calbindins have been defined, and the induction of their synthesis by the vitamin D hormone was established [10,69,70,71]. The avian form, calbindin-D28K, has a molecular weight of about 30,000 and binds four Ca2+ ions per molecule with high affinity. The mammalian form, calbindin-D9K, has a molecular weight of about 10,000 and binds two Ca2+ ions with high affinity. The high-affinity binding sites of both proteins have an average association constant (Ka) of 2–4 × 106 M−1 and therefore a dissociation constant (Kd) of 2.5–5 × 10−7 M. The binding affinities of the proteins are in a range that are appropriate for these proteins to serve as effective intracellular Ca2+ buffers. The high affinity calcium binding sites of the calbindins are in the helix–loop–helix EF-hand configuration of Kretsinger and Nockolds [72]. At this site, each calcium atom is bound to five hydrophobic amino acid residues and one water molecule. Although the larger calbindin, calbindin D28K, has six EF hand motifs, only four participate in high-affinity calcium binding. b. Calbindin Synthesis The primary stimulus for the syntheses of calbindin-D9K and calbindin-D28K is interactions of 1,25(OH)2D3 with the nuclear vitamin D receptor (VDR). However, other factors appear to be able to participate in the synthetic process. For example, in vitamin D receptor (VDR)-ablated mice, intestinal calbindin-D28K mRNA was substantially reduced but was normalized by a high local concentration of calcium, phosphorus, and/or lactose (the so-called “rescue diet”) in the intestinal lumen [73]. As another example, in the shell gland of the laying hen, there is a high Ca2+ flux across the shell gland membrane of the laying hen in order to provide the large amount of calcium required for the formation of the egg shell. Bar et al. [74] obtained results consistent with a dependency of calbindin mRNA synthesis on the rate at which calcium is transported across the egg shell gland. 1,25(OH)2D3 in the egg shell gland system is apparently required for the translation of calbindin-D28K mRNA into the protein [75]. c. The Relation of Calbindin to Calcium Absorption Direct correlations between the mucosal concentration of calbindin and the efficiency of calcium absorption under a wide variety of experimental conditions have been demonstrated, giving support to an

ROBERT H. WASSERMAN

important role of the calbindins in vitamin D–mediated calcium absorption. In the avian species, early work on the calcium-binding protein in the chick intestine revealed its vitamin D dependency and the presence of the protein in all parts of the small intestine [76]. The concentration of calbindin-D28K followed the following sequence: duodenum > jejunum > ileum. The vitamin D– dependent increase in calcium absorption followed the same sequence, i.e., duodenum > jejunum > ileum [77]. This provided in this species a reasonable site-to-site correlation between the calcium-binding protein concentrations and the magnitude of the effect of vitamin D on the calcium absorption process. Another correlation is the temporal correspondence between a 1,25(OH)2D3-mediated increase in calcium absorption and appearance of calbindin-D28K in chick duodenum of vitamin-D-deficient chicks (Fig. 4). As reviewed elsewhere [69,78], there are corresponding increases in intestinal calbindin and calcium absorption during adaptation to low-calcium or low-phosphorus diets, during growth, and in the transition from nonlaying to egg-laying in hens. Both of these entities decrease during aging [79,80]. Cortisol treatment, which inhibits calcium absorption, appears to do so by depressing calbindin synthesis [49]. There are situations, however, where noncorrelations exist. One example is the persistence of calbindin in the cytosol after vitamin D–enhanced calcium absorption has descended toward the pre–vitamin D baseline [69]. Noncorrelations of this sort can be explained in terms of the multifunctional role of 1,25(OH)2D3 on calcium absorption, indicating that under certain circumstances, vitamin D–dependent reactions or factors other than the calbindins become limiting, such as calcium entry as discussed in Section IV,C,3. 2. FUNCTIONS OF THE CALBINDINS

There are at least three, perhaps four, functions proposed to be served by the calbindins. One is participation in the control of the opened–closed status of the microvillar calcium channel. Another is the acceleration of the diffusion of calcium through the cytosol, and the third, an increase in the availability of calcium for extrusion by calcium pumps at the basolateral membrane. The fourth is a direct stimulatory effect on calcium pump activity. a. Calbindin and Activation of the Calcium Entry Process A clue that the transfer of calcium across the microvillus membrane is controlled by calcium itself comes from the report of Sampson et al. [81]. Using electron microscopic and Ca autoradiographic techniques, Sampson and colleagues showed that in vitamin D– deficient rat, luminal Ca2+ can cross the microvillar membrane, binding predominately to the inner aspect

CHAPTER 24 Vitamin D and the Intestinal Absorption of Calcium: A View and Overview

of the microvillus. After vitamin D repletion, Ca2+ moves away from this region, some sequestered by mitochondria. Confirmation and additional detail came 20 years later when the movement of calcium during absorption was visualized by the use of a stable calcium isotope (44Ca2+) and ion microscopy [82]. Like the Sampson observations, it was shown in the vitamin D– deficient chick intestine that luminal Ca can enter the intestinal cell but remains bound to elements within the microvillar region. After vitamin D–repletion, absorbed Ca2+ also appeared to interact transiently with structures in this region before moving throughout the cell as a calbindin complex. The calcium ions initially associated with the intracellular elements of the microvillar membrane might have a role in the control of calcium entry. This view is consistent with the observations of Yue et al. [83], who showed that the transport function of the epithelial calcium channel, CaT1, was inactivated by high Ca2+ concentrations and activated by a decrease in intracellular Ca2+ concentrations. Feedback control of calcium channel activity by intracellular calcium was also put forth by Hoenderop et al. [12] and Peng et al. [13]. Kinase activation by cyclic nucleotides could also control channel activity by phosphorylation of sites present in the epithelial calcium channels. Possibly related to this scheme is the control of a CaT-like (CaT-L) calcium channel of selected exocrine tissues closely related to CaT1 and ECaC as reported by Niemeyer et al. [84]. This channel has a calmodulin binding site that is also a site for protein kinase C phosphorylation. It was shown that the calcium-dependent calmodulin binding to CaT-L facilitates channel inactivation, and protein kinase C–mediated phosphorylation at the calmodulin site restores channel activity. b. Calbindins and Cytosolic Diffusion of Calcium Ninety percent of the calbindin within the enterocyte is readily released from intestinal tissue upon homogenization and therefore assumed to be primarily in a soluble state. Because of its soluble nature and its Ca2+-binding characteristics, the idea that the calbindins could serve as facilitators of Ca2+ flux through the cytosol was put forth [85]. Levine and Williams [86] and Levine et al [87] suggested that calbindin acts as an “aqueous phase ionophore,” i.e., a diffusional facilitator, increasing the flux of Ca2+ from apical to the basal region of the cell by a factor of 102. Kretsinger et al. [88] provided a theoretical basis for calbindin to act as a cytosolic diffusional facilitator. With this mechanism of calbindin action, the total amount of Ca2+ moving from the apical to this basal region would be the sum of the concentrations of free Ca2+ and calbindinbound Ca2+.

419

Feher [89] and Feher et al. [90] tested the diffusional facilitator hypothesis by examining the rate of diffusion of Ca2+ in a three-compartment system, from a precursor (lumen) compartment through the center “cytosol” compartment and then into the product (“extracellular”) compartment. The compartments were separated by semipermeable membranes. By placing calbindin in the “cytosol” middle compartment, there was a significant increase in the rate of diffusion of Ca2+ as compared to albumin, giving experimental support to the diffusion facilitator hypothesis. Other observations provide additional support for this concept of calbindin function. Pansu et al. [91] had reported that theophylline inhibits Ca2+ binding to calbindin and depresses Ca2+ absorption with no observable effect on Ca2+ uptake by isolated brush border membranes or basolateral membrane vesicles. More recently, Koster et al. [92] incorporated a soluble Ca2+ ligand, BAPTA, into isolated rabbit cortical collecting tubules and showed that the intracellular BAPTA “fully mimicked” the stimulatory effect of CaBP on transcellular Ca2+ transport. Detailed analysis of the intestinal absorption system by Bronner et al. [7] and computer modelling by Feher et al. [93] provide additional support for the functioning of calbindins as diffusional facilitators.

E. Calcium Extrusion Across the Basolateral Membrane, the Third Step in Transcellular Calcium Absorption The transfer of Ca2+ from the cell interior to the extracellular space requires energy input in order to overcome the considerable electrochemical potential difference. As depicted in Fig. 5, two systems associated with the basolateral membrane are available to extrude Ca2+ against this considerable gradient, the ATP-dependent Ca2+ pump (Ca2+-ATPase) and a Na+/Ca2+ exchanger. Also depicted is the process of exocytosis, the terminal event of the vesicular transport model. 1. THE PLASMA MEMBRANE CALCIUM PUMP

An ATP-dependent plasma membrane Ca2+ pump (PMCA) was first identified in erythrocyte membranes, and the properties of PMCA have been summarized by Carafoli [94] and Penniston and Enyedi [95]. PMCA has a molecular weight of 130–140 kDa, transports one Ca2+ per ATP, and has a Km of approximately 0.2 µM in the presence of calmodulin. Isolated basolateral vesicle membranes were initially used to demonstrate the presence of an ATP-dependent uphill transporter in intestinal membranes [96–98].

420

ROBERT H. WASSERMAN

Ca-related events the basolateral membrane Ca

Ca Channel Ca Pump

Ca

Ca Ca-Calbindin

Ca

Ca

Diffusion Ca

Ca

Free Ca

Na/Ca Exchange

Na

Ca

Diffusion

Na

Na/K Pump Vesicular

K

Ca

Transport Ca

Exocytosis

FIGURE 5 Calcium-related events at the basolateral membrane of the enterocyte. These include the uphill transport by the plasma membrane calcium pump and the Na+/K+ exchange pump. The Na+/K+ pump maintains a transmembrane gradient of Na+ for the operation of the Na+/Ca2+ exchanger. Also shown is the vesicular mode of transport. From [10].

The transport capacity of the ATP-dependent Ca2+ uptake by rat basolateral vesicles varies with maturation and aging [80,99], intestinal segment [100], and crypt-to-villus axis [101]. The stimulatory effect of vitamin D on the ATP-dependent uptake of Ca2+ by isolated basolateral membranes was shown [96,98], and kinetic analysis indicated that 1,25(OH)2D increases the Vmax of the transport process by a factor of about 3, whereas the Km is unaffected by treatment [10]. Zelinski et al. [102] were the first to show that 1,25(OH)2D3 stimulated the synthesis of Ca2+ pump mRNA. This was followed by our studies done with an antibody produced against the erythrocyte plasma membrane Ca2+ pump that cross-reacts with a Ca2+ epitope in rat and avian intestine [103,104]. Immunohistochemical analysis demonstrated clearly that the Ca2+ pump was associated with the basolateral membrane [105]. Vitamin D stimulated the synthesis of the avian Ca2+ pump [105] and, as shown by Armbrecht et al. [106], also of the calcium pump in rat intestine. Adaptation of

vitamin D–replete chicks to either a low-calcium diet or a low-phosphorus diet also increased the amount of the Ca2+ pump units associated with the basolateral membranes [107]. The increase in PMCA gene expression by 1,25(OH)2D3 is due, at least in part, to a transcriptional event [108]. 2. MODULATORS OF CALCIUM PUMP ACTIVITY

Calmodulin, the ubiquitous high-affinity calcium binding protein, is known to stimulate the activity of the PMCA [94]. Several studies have shown that calbindin-D9K and calbindin-D28K, under certain conditions, can also stimulate the activity of the basolateral Ca2+ pump (refer to Wasserman et al. [104] for references and discussion). Timmermans et al. [109] verified that calbindin-D28K can stimulate PMCA activity of rat duodenal basolateral membrane vesicles, an effect shared by another high-affinity calcium-binding protein, parvalbumin.

CHAPTER 24 Vitamin D and the Intestinal Absorption of Calcium: A View and Overview

The exact mechanism by which these high-affinity calcium binding proteins other than calmodulin affect PMCA activity is not known. One possibility is a direct interaction with the PMCA molecule, in a calmodulinlike response. Another is a means of efficiently transferring Ca2+ to the calcium binding site of PMCA, similar to the stimulatory effect of the calcium chelator, EGTA, on PMCA activity (refer to Carafoli [94] for a discussion of the EGTA effect). Like the microvillar membrane, the lipid composition of the basolateral membranes is affected by vitamin D status [110]. There is a relative increase in cholesterol content of the basolateral membranes and an increase in the arachidonic acid content of membrane phospholipids. Calcium and phosphorus deficiencies also altered lipid composition and fluidity [111]. Such lipid changes could affect the operation of enzymes, including and the plasma membrane calcium pump in particular. 3. PLASMA MEMBRANE SODIUM–CALCIUM EXCHANGER

The extrusion of Ca2+ by the Na+/Ca2+ exchanger is linked to the downhill movement of extracellular Na+ into the cell. The exchanger is rheogenic, transferring three Na+ per one Ca2+, and therefore is responsive to the cytosolic negative electropotential with respect to the external fluid phase. The Ca2+ binding affinity of the Na+/Ca2+ exchanger was reported to vary from 0.1 to 8 µM in different systems [10], in a range appropriate for the extrusion of intracellular Ca2+ across the basolateral membrane. Ghijsen et al. [112] estimated that the Na+/Ca2+ exchanger in rat intestine might account for 20% of the Ca2+ extrusion capacity of the basolateral membrane. They also reported that vitamin D did not affect exchanger activity.

V. VITAMIN D AND THE PARACELLULAR PATH Calcium is also absorbed by the nonsaturable diffusional mode by way of the paracellular path that exists between adjacent cells of the intestinal cellular membrane. The permeability of the pathway is regulated by the proteins that make up the tight junctional complex at the apical region of the adjacent cells, and also by structures present within the intercellular space [113,114]. The proteins that form these structures are synthesized by the adjacent intestinal cells and cojoin within the intercellular space to serve as part of the “glue” holding the cells together. Proteins involved in the function of the tight junction complex have been well described, and these include primarily occludin and a protein that is a member of the claudin family [115].

421

Claudin has been specifically implicated in creating charge-selective channels across the tight junctional complex [116]. Another protein, cadherin, is associated with the adherens junctions located in the intercellular space below the apical tight junctional complex. These and other paracellular-associated proteins are discussed in detail in reviews by Anderson and Van Itallie [117] and Balda and Matter [118].

A. Thermodynamic Parameters The thermodynamic parameters that pertain to the calcium flow through the paracellular path include the electropotential difference across the intestinal epithelium of about +6 mV, and the calcium concentration in the product compartment, the lamina propria, which is assumed to be about 1.25 mM, is the same as in blood plasma. Using these values, the concentration of calcium in the intestinal lumen required for there to be net diffusion of calcium via the paracellular path has been estimated to be in the range of 2–6 mM Ca2+ [5,16].

B. Vitamin D Dependency of the Diffusional Process The diffusional, nonsaturated mode of calcium absorption is considered by some to be independent of vitamin D action [119–121]. There are, however, observations that are consistent with an enhancement of paracellular calcium absorption by vitamin D. The earlier studies in which unidirectional fluxes of calcium across the avian duodenum were measured showed that vitamin D, in addition to enhancing active Ca2+ transport, also increases the nonsaturable phase of calcium absorption (Fig. 1) [15]. Further, an increase in the diffusional flow of Ca2+ from plasma to intestinal lumen by vitamin D was also observed in situ in chick ileum and rat duodenum [122], supporting the view that vitamin D has a stimulatory effect on the diffusional process. Subsequently, a number of reports appeared that provided further support of the ability of vitamin D to enhance the paracellular absorption of calcium. For instance, Hurwitz and Bar [25] reported that vitamin D increased both outflux (lumen to plasma) and influx (plasma to lumen) in the intact chick “which provides some evidence that vitamin D acts to increase the permeability of the intestine for calcium.” Karbach [123] measured the fluxes of calcium across different segments of rat intestine, using the Ussing chamber technique. Both fluxes (mucosa to

422 serosa and serosa to mucosa) were increased by 1,25(OH)2D3 and, in Karbach’s words, there was a “stimulation of paracellular calcium flux in both directions in all three intestinal segments.” Studies on rat intestine by Jungbluth and Binswanger [124] yielded a similar view that, in the duodenum and jejunum, the “overall effect of 1,25(OH)2D3 on intestinal Ca and Pi transport is to increase both cell-mediated active mucosal-to-serosal transport and paracellular diffusional serosal-to-mucosa ion movement.” The work of Chirayath et al. [125], with the Caco-2 cell line, demonstrated a stimulation of diffusional transport of Ca2+ through the paracellular pathway by 1,25(OH)2D3. The transfer of Ca2+ across the confluent cell layer was inversely related to transepithelial resistance, i.e., directly related to conductance. The bidirectional fluxes of calcium were also increased. These observations and others reported by Chirayath et al. [125] appear to establish a vitamin D–dependent paracellular route in the confluent Caco-2 cells. Consistent with this are Caco-2 cell line experiments of Fleet et al. [126,127] in which 1,25(OH)2D3 increased the bidirectional calcium fluxes. Li et al. [128], in examining events occurring in vitamin D receptor–null mice, noted a 50% reduction in intestinal calbindin-D9K without a significant change in calcium absorption. These authors surmised that the passive paracellular pathway compensated for the “reduction in the calbindin-mediated transport” and put forth the view that the paracellular calcium transport is increased by 1,25(OH)2D3 in these mice. The mechanism by which 1,25(OH)2D3 alters the paracellular permeability of intestinal epithelium is not precisely known. Stenson et al. [129] have reported that activation of protein kinase C (PKC) increases paracellular permeability, and 1,25(OH)2D3 activation of PKC has been shown in various systems [130–132]. Perez et al. [133] suggest that epithelial permeability is controllable by intracellular mediators, such as Ca2+, cAMP, and protein kinase C, probably by affecting cytoskeleton activity.

VI. VESICULAR TRANSPORT OF CALCIUM The movement of luminally derived Ca2+ through the cytosolic compartment within membrane-bound structures was proposed by Jande and Brewer in 1974 [134] and by Warner and Coleman in 1975 [135] on the basis of histological and electron probe analysis, respectively. More recently, Nemere and Norman [136] also implicated vesicular transport as a means by which Ca2+ moves through the cell, to be released from the enterocyte by exocytosis.

ROBERT H. WASSERMAN

VII. COMMENTARY ON SEGMENT-SPECIFIC INTESTINAL ABSORPTION OF CALCIUM A. Duodenum and Ileum In attempts to understand the mechanism by which vitamin D increases calcium absorption, the duodenum of various species has attracted most research attention because of its highly active transport system. From this effort, the major vitamin D–dependent components of the active process have been identified. These include the epithelial calcium channels, the calbindins, and the plasma membrane calcium pump. But it is the ileum rather than the duodenum where perhaps 70–80% of the ingested calcium is absorbed (cf. III,B). Because of this, it is important to attempt to understand the processes by which this distal segment transports calcium, and different views have been expressed as to the mechanisms of ileal calcium absorption. On one hand, Pansu et al. [120] could not detect a saturable, active component of calcium absorption in the rat ileum, or a response of the segment to 1,25(OH)2D3, or the presence in this segment of CaBP, i.e., calbindinD9K. In a recent report, Bronner et al. [137] stated that “in the ileum … all calcium is absorbed by the passive route,” a route of absorption which these investigators consider to be “vitamin D-independent.” On the other hand, in situ intestinal perfusion studies in the rat disclosed evidence of the ability of all segments of the small intestine, the duodenum, jejunum, and ileum, to actively transport calcium, using unidirectional flux data and Ussing criteria [16]. Krawitt and Schedl [27], also using an in situ perfusion system, observed net movement of calcium from lumen to plasma in the proximal, middle, and distal segments of the small intestine, indicative of active calcium transport in each of these segments. In a study of patients with chronic renal disease, Vergne-Marini et al. [138] showed that 1,25(OH)2D3 treatment increased the unidirectional calcium flux ratio in the ileum from 1.3 to 2.3, the latter exceeding the theoretical Ussing flux ratio of 1.2, thereby providing evidence of an active transport system in the human ileum. Armbrecht et al. [80] verified the presence of calbindin-D9K in rat ileum and the dependency of its synthesis on 1,25(OH)2D3. Armbrecht et al. [106] further demonstrated that the concentration of the plasma membrane calcium pump in rat ileum is elevated by 1,25(OH)2D3, the response varying with age. Thus, although these are present in lesser quantities than in the duodenum, the ileum does contain two essential components of the vitamin D–dependent transcellular active transport system, calbindin-D9K and

Transcellular

Paracellular Vitamin D deficient

A

D

Ca

Ca Ca Ca Ca Ca Ca Ca Ca Ca Ca Ca Ca

Ca Ca Ca Ca

−D

Ca

Ca

Ca Ca Ca Ca Ca

Ca pump

Vitamin D adequate—Duodenum

B

E Ca pump

Ca Ca

CaT1 Ca

Ca

Ca

+D

Ca

Ca Ca

Ca

CaBP Ca

Ca

CaBP Ca

Ca

Ca

Vitamin D adequate—Ileum

C

F Ca pump

Ca Ca Ca

Ca

CaBP +D

Ca

Ca

Ca

Ca Ca

CaBP

Ca

FIGURE 6 Models of vitamin D–mediated processes of intestinal absorption of calcium. Transcellular: (A) In the absence of vitamin D, calcium entry is by simple diffusion across the microvillar membrane (in duodenum, a basal level of an epithelial calcium channel might be present). Entered calcium might transiently associate with the inner aspect of the microvillar complex before diffusing to the basolateral membrane, followed by extrusion into the lamina propria by the plasma membrane calcium pump, the latter also at basal levels. (B) In the duodenum after vitamin D repletion, the rate of calcium entry is accelerated because of the synthesis of epithelial calcium channels, mediated by 1,25(OH)2D3 action. Calcium entry is controlled by the calcium concentration in the microvillar region with high [Ca] inhibiting calcium entry and low [Ca] increasing channel permeability. A means of lowering [Ca] is by complexation to 1,25(OH)2D3induced calbindins, the high-affinity calcium-binding proteins. Calbindin (mammal, calbindin-D9K; avian, calbindin-D28K) also functions as a transcytosolic diffusion facilitators, increasing the rate of transfer of bound calcium to the basolateral membrane. Calcium, dissociated from calbindin in the microvicinity of the calcium pump, is transported uphill into the extracellular fluid and at an accelerated rate due to increased synthesis of pump units mediated by 1,25(OH)2D3. (C) An epithelial calcium channel has not been identified in the ileum, and therefore calcium entry into the enterocyte is by simple diffusion. After vitamin D, the ileal enterocyte contains the 1,25(OH)2D3-induced calbindin and PMCA, already present, is augmented by 1,25(OH)2D3. As designated by the lighter shading, calbindin and PMCA are at lower concentrations compared to the duodenum, accounting for the slower rate of active transport in the ileum. The relatively long ileal transit time largely compensates for the relatively slow rate of ileal calcium absorption. Paracellular: (D) In vitamin D deficiency, calcium is absorbable by the nonsaturable diffusional process via the paracellular path if the concentration of calcium in the intestinal lumen is sufficiently high to overcome the thermodynamic barrier. (E,F) Though controversial, the evidence summarized in the text supports a 1,25(OH)2D3-mediated increase in the calcium permeability of the paracellular pathway. Modified from [10].

424 the plasma membrane calcium pump. Not identified in the ileum, however, are the recently discovered epithelial calcium channels (Section IV,C,2,a). Perhaps there is a calcium channel in the rat ileum different from CaT1 or ECaC but yet to be identified. In any event, luminal calcium can diffuse into the enterocyte by simple diffusion across the microvillar membrane as shown in the ion microscopy studies of Fullmer et al. [82]. After entry, calcium can now be transferred through and out of the cell in the same manner as in the duodenum but at a slower rate. Despite this slower rate, ileal calcium absorption by the active process could make a significant impact on overall calcium absorption because of the relatively long transit time of calcium in the ileal segment.

B. Colon The absorption of dietary calcium in the human colon is a small percentage of totally absorbed calcium, perhaps 3–6%, depending on calcium intake [139]. In the rat, the colonic contribution to total absorption was about 8% [21]. The colon has the capability of actively transporting calcium by a vitamin D–mediated mechanism [140–143].

VIII. SUMMARY Vitamin D is required for the optimal absorption of calcium in order. Physiologically, the understanding of vitamin D action began with the demonstration circa 1960 that vitamin D stimulated the saturable, energydependent absorption of calcium. Biochemically, among the first components of the calcium transport system to be identified were vitamin D–induced high-affinity Ca binding proteins, the calbindins: the avian type, calbindinD28K, and the mammalian type, calbindin-D9K. One of the proposed function of the calbindins is to facilitate the transcytosolic movement of calcium. Other components of trans-cellular calcium absorption are the basolateral plasma membrane calcium pump, the number of pump units enhanced by vitamin D, and the recently identified epithelial calcium channels that facilitate calcium entry across the apical microvillar membrane. These channels also have a vitamin D-dependency and their permeability and their activity is feedback regulated, directly or indirectly, by intracellular calcium concentrations. With these three components, a model of the operation of active transcellular calcium absorption system is presented in Fig. 6. In brief, calcium within the intestinal lumen enters the enterocyte by simple diffusion across the lipid layer of the microvillar membrane or through the epithelial calcium channels. The calcium, perhaps transiently interacting with components of the

ROBERT H. WASSERMAN

microvillus region, binds to calbindin and moves through the cytosol in this bound form, or as the free ion. At the basolateral membrane, the calcium released from calbindin is transported actively into the extracellular space by the plasma membrane calcium pump. Calcium absorbed by a nonsaturable diffusion process, moves paracellularly across the intestinal cellular layer. Although this is not universally accepted, there is considerable evidence that vitamin D does increase this diffusional mode of calcium absorption. Thus, it appears that vitamin D, through its hormonal form, affects calcium absorption both by the active transport system and by the diffusional system. In addition, calcium is absorbed in the absence of vitamin D, although at a much slower rate and to a lesser extent than in the presence of vitamin D, and most often in insufficient amounts to prevent bone disease. Which mode of absorption dominates depends on a number of different variables. These variables include vitamin D status, dietary and luminal calcium concentrations, the vigor of the active transport system in any given segment, and the transit time in that segment. By whichever mechanism, there is experimental support for the view that, in normal, vitamin D–adequate individuals, most (or all) of the ingested calcium is absorbed by vitamin D– mediated processes which include both transcellular active transport and transparacellular diffusion.

Acknowledgments The author gives special thanks to Norma Jayne for her capable secretarial assistance and to the National Institutes of Health (DK-04652) for many years of research support.

References 1. Robertson WG 1988 Chemistry and biochemistry of calcium. In: Nordin BEC (ed) Calcium in Human Biology. SpringerVerlag, London. 2. Avioli LV, Birge SJ 1978 Mechanisms of calcium absorption: A reappraisal. In: Andreoli TE, Hoffman JF, Fanestil DD (eds) Physiology of Membrane Disorders. Plenum, New York, pp. 919–940. 3. Harrison HE 1975 Vitamin D malnutrition. Prog Food Nutr Sci 1:349–357. 4. Bronner F, Pansu D 1999 Nutritional aspects of calcium absorption. J Nutr 129(1):9–12. 5. Walling MW 1982 Regulation of intestinal calcium and inorganic phosphate absorption. In: Parsons JA (ed) Endocrinology of Calcium Metabolism. Raven Press, New York, pp. 87–101. 6. Favus MJ 1985 Factors that influence absorption and secretion of calcium in the small intestine and colon. Am J Physiol 248(1):G147–G157.

CHAPTER 24 Vitamin D and the Intestinal Absorption of Calcium: A View and Overview

7. Bronner F, Pansu D, Stein WD1986 An analysis of intestinal calcium transport across the rat intestine. Am J Physiol 250(1):G561–G569. 8. Christakos S, Gabrielides C, Rhoten WB 1989 Vitamin D– dependent calcium binding proteins: chemistry, distribution, functional considerations, and molecular biology. Endocr Rev 10(1):3–26. 9. Gross M, Kumar R 1990 Physiology and biochemistry of vitamin D-dependent calcium binding proteins. Am J Physiol 259:F195–F209. 10. Wasserman RH, Fullmer CS 1995 Vitamin D and intestinal calcium transport: facts, speculations and hypotheses. J Nutr 125(7 Suppl):1971S–1979S. 11. Bouillon R, Van Cromphaut S, Carmeliet G 2003 Intestinal calcium absorption: Molecular vitamin D mediated mechanisms. J Cell Biochem 88(2):332–339. 12. Hoenderop JG, Nilius B, Bindels RJ 2002 ECaC: The gatekeeper of transepithelial Ca2+ transport. Biochim Biophys Acta 1600(1–2):6–11. 13. Peng JB, Brown EM, Hediger MA 2003 Apical entry channels in calcium-transporting epithelia. News Physiol Sci 18(4):158–163. 14. Holick MF 2003 Vitamin D: A millenium perspective. J Cell Biochem 88:296–307. 15. Wasserman RH, Kallfelz FA 1962 Vitamin D3 and the unidirectional calcium fluxes across the rachitic chick duodenum. Am J Physiol 203:221–224. 16. Wasserman RH, Taylor AN 1969 Some aspects of the intestinal absorption of calcium, with special reference to vitamin D. In: Comar CL, Bronner F (eds) Mineral Metabolism, An Advanced Treatise, Vol. III. Academic Press, New York, pp. 321–403. 17. Wilkinson R 1976 Absorption of calcium, phosphorus, and magnesium. In: Nordin BEC (ed) Calcium Phosphate and Magnesium Metabolism. Churchill Livingstone, Edinburgh. 18. Nordin BEC, Marshall DH 1988 Dietary requirements for calcium. In: Nodin BEC (ed) Calcium in Human Biology Springer-Verlag, London. 19. Heaney RP, Saville PD, Recker RR 1975 Calcium absorption as a function of calcium intake. J Lab Clin Med 85(6):881–890. 19a. Marcus, CS, & Lengemann, FW 1962 Use of radioyttrium to study food movement in the small intestine of the rat. J Nutr 76:179–182. 20. Marcus CS, Lengemann FW 1962 Absorption of Ca45 and Sr85 from solid and liquid food at various levels of the alimentary tract of the rat. J Nutr 77:155–160. 21. Cramer CF, Copp DH 1959 Progress and rate of absorption of radiostrontium through intestinal tracts of rats. Proc Soc Exp Biol Med 102:514–517. 22. Cramer CF 1965 Sites of calcium absorption and the calcium concentration of gut contents in the dog. Can J Physiol and Pharmacol 43:75–78. 23. Duflos C, Bellaton C, Pansu D, Bronner F 1995 Calcium solubility, intestinal sojourn time and paracellular permeability codetermine passive calcium absorption in rats. J Nutr 125(9):2348–2355. 24. Hurwitz S, Bar A 1966 Rate of passage of calcium-45 and yttrium-91 along the intestine, and calcium absorption in the laying fowl. J Nutr 89:311–316. 25. Hurwitz S, Bar A 1972 Site of vitamin D action in chick intestine. Am J Physiol 222(3):761–767. 26. Schachter D 1962 Vitamin D and the active transport of calcium by the small intestine. In: Wasserman RH (ed) The Transfer of Calcium and Strontium across Biological Membranes. Academic Press, New York, pp. 197–210.

425

27. Krawitt EL, Schedl HP 1968 In vivo calcium transport by rat small intestine. Am J Physiol 214(2):232–236. 28. Younoszai MK, Urban E, Schedl HP 1973 Vitamin D and intestinal calcium fluxes in vivo in the rat. Am J Physiol 225(2):287–292. 29. Bikle DD, Morrissey RL, Zolock DT 1979 The mechanism of action of vitamin D in the intestine. Am J Clin Nutr 32(11):2322–2328. 30. Fontaine O, Matsumoto T, Goodman DB, Rasmussen H 1981 Liponomic control of Ca2+ transport: relationship to mechanism of action of 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 78(3):1751–1754. 31. Hoenderop JG, van der Kemp AW, Hartog A, van de Graaf SF, van Os CH, Willems PT, Bindels RJ 1999 Molecular identification of the apical Ca2+ channel in 1,25-dihydroxyvitamin D3-responsive epithelia. J Biol Chem 274(13): 8375–8378. 32. Peng JB, Chen XZ, Berger UV, Vassilev PM, Tsukaguchi H, Brown EM, Hediger MA 1999 Molecular cloning and characterization of a channel-like transporter mediating intestinal calcium absorption. J Biol Chem 274(32): 22739–22746. 33. Peng JB, Chen XZ, Berger UV, Vassilev PM, Brown EM Hediger MA 2000 A rat kidney-specific calcium transporter in the distal nephron. J Biol Chem 275(36):28186–28194. 34. Peng JB, Chen XZ, Berger UV, Weremowicz S, Morton CC, Vassilev PM, Brown EM, Hediger MA 2000 Human calcium transport protein CaT1. Biochem Biophys Res Commun 278(2):326–332. 35. Muller D, Hoenderop JG, Meij IC, van den Heuvel LP, Knoers NV, den Hollander AI, Eggert P, Garcia-Nieto V, Claverie-Martin F, Bindels RJ 2000 Molecular cloning, tissue distribution, and chromosomal mapping of the human epithelial Ca2+ channel (ECaC1). Genomics 67(1):48–53. 36. Barley NF, Howard A, O’Callaghan D, Legon S, Walters JR 2001 Epithelial calcium transporter expression in human duodenum. Am J Physiol 280(2):G285-G290. 37. Wood RJ, Tchack L, Taparia S 2001 1,25-Dihydroxyvitamin D3 increases the expression of the CaT1 epithelial calcium channel in the Caco-2 human intestinal cell line. BMC Physiol. 1(1):11. 38. Giuliano AR, Wood RJ 1991 Vitamin D–regulated calcium transport in Caco-2 cells: unique in vitro model. Am J Physiol 260(2 Pt 1):G207–G212. 39. Fleet JC, Wood RJ 1994 Identification of calbindin D-9k mRNA and its regulation by 1,25-dihydroxyvitamin D3 in Caco-2 cells. Arch Biochem Biophys 308(1):171–174. 40. Rasmussen H, Matsumoto T, Fontaine O, Goodman DB 1982 Role of changes in membrane lipid structure in the action of 1,25-dihydroxyvitamin D3. Fed Proc 41(1):72–77. 41. Wasserman RH, Brindak ME, Meyer SA, Fullmer CS 1982 Evidence for multiple effects of vitamin D3 on calcium absorption: Response of rachitic chicks, with or without partial vitamin D3 repletion, to 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 79(24):7939–7943. 42. Putkey JA, Spielvogel AM, Sauerheber RD, Dunlap CS, Norman AW 1982 Vitamin D–mediated intestinal calcium transport. Effects of essential fatty acid deficiency and spin label studies of enterocyte membrane lipid fluidity. Biochim Biophys Acta 688(1):177–190. 43. Bikle DD, Whitney J, Munson S 1984 The relationship of membrane fluidity to calcium flux in chick intestinal brush border membranes. Endocrinology 114(1):260–267. 44. Brasitus TA, Dudeja PK, Eby B, Lau K 1986 Correction by 1,25-dihydroxycholecalciferol of the abnormal fluidity

426

45. 46. 47.

48. 49.

50. 51.

52. 53. 54. 55.

56.

57. 58.

59. 60.

61. 62.

ROBERT H. WASSERMAN

and lipid composition of enterocyte brush border membranes in vitamin D-deprived rats. J Biol Chem 261(35): 16404–16409. Pileggi VJ, DeLuca HF, Steenbock H 1955 The role of vitamin D and intestinal phytase in the prevention of rickets in rats on cereal diets. Arch Biochem Biophys 58:194–204. Holdsworth ES 1970 The effects of vitamin D on the enzyme activities in the mucosal cells of the chick small intestine. J Membrane Biol 3:43–53. Norman AW, Mircheff AK, Adams TH, Spielvogel A 1970 Studies on the mechanism of action of calciferol. 3. Vitamin D– mediated increase of intestinal brush border alkaline phosphatase activity. Biochim Biophys Acta 215(2):348–359. Haussler M, Nagode LA, Rasmussen H 1970 Induction of intestinal brush border alkaline phosphatase by vitamin D: Identity with Ca-ATPase. Nature 228(277),1199–1201. Feher JJ, Wasserman RH 1979 Intestinal calcium-binding protein and calcium absorption in cortisol-treated chicks: effects of vitamin D3 and 1,25-dihydroxyvitamin D3. Endocrinology 104(2):547–551. Kowarski S, Schachter D 1975 Vitamin D–dependent, particulate calcium-binding activity and intestinal calcium transport. Am J Physiol 229:1198–1204. Kowarski S, Schachter D 1980 Intestinal membrane calciumbinding protein. Vitamin D–dependent membrane component of the intestinal calcium transport mechanism. J Biol Chem 255(22):10834–10840. Miller A, Ueng TH, Bronner F 1979 Isolation of a vitamin D– dependent, calcium-binding protein from brush border of rat duodenal mucosa. FEBS Lett 103:319–322. Schachter D, Kowarski S 1982 Isolation of the protein IMCal, a vitamin D–dependent membrane component of the intestinal transport mechanism for calcium. Fed Proc 41:84–87. Glenney JR Jr, Glenney P 1985 Comparison of Ca++regulated events in the intestinal brush border. J Cell Biol 100(3):754–763. Howe CL, Keller TCS, III, Mooseker MS, Wasserman RH 1982 Analysis of cytoskeletal proteins and Ca2+-dependent regulation of structure in intestinal brush borders from rachitic chicks. Proc Natl Acad Sci USA 79:1134–1138. Kaune R, Munson S, Bikle DD 1994 Regulation of calmodulin binding to the ATP extractable 110 kDa protein (myosin I) from chicken duodenal brush border by 1,25-(OH)2D3. Biochim Biophys Acta 1190(2):329–336. Bikle DD, Munson S, Chafouleas J 1984 Calmodulin may mediate 1,25-dihydroxyvitamin D–stimulated intestinal calcium transport. FEBS Lett 174(1):30–33. Bikle D, Munson S, Christakos S, Kumar R, Buckendahl P 1989 Calmodulin binding to the intestinal brush-border membrane: comparison with other calcium-binding proteins. Biochim Biophys Acta 1010(1):122–127. Mooseker MS, Coleman TR 1989 The 110-kDa proteincalmodulin complex of the intestinal microvillus (brush border myosin I) is a mechanoenzyme. J Cell Biol 108:2395–2400. Mooseker MS, Wolenski JS, Coleman TR, Hayden SM, Cheney RD, Espreafico E, Heintzelman MB, Peterson MD 1991 Structural and functional dissection of a membranebound mechanoenzyme brush border myosin I. Current Top Membr 33:31–55. Mykkanen HM, Wasserman RH 1989 Uptake of 75Se-selenite by brush border membrane vesicles from chick duodenum stimulated by vitamin D. J Nutr 119(2):242–247. Mykkanen HM, Wasserman RH 1990 Relationship of membrane-bound sulfhydryl groups to vitamin D–stimulated

63.

64.

65. 66.

67. 68. 69.

70. 71. 72. 73.

74.

75.

76.

77. 78. 79.

80.

uptake of [75Se] selenite by the brush border membrane vesicles from chick duodenum. J Nutr 120(8):882–888. de Talamoni NT, Mykkanen H, Wasserman RH 1990 Enhancement of sulfhydryl group availability in the intestinal brush border membrane by deficiencies of dietary calcium and phosphorus in chicks. J Nutr 120(10):1198–1204. Tolosa de Talamoni N, Mykkanen H, Cai Q, Wasserman RH 1991 Hormonal effects on the sulfhydryl groups associated with intestinal brush border membrane proteins. Biochim Biophys Acta 1094(2):224–230. Fullmer CS 1992 Intestinal calcium absorption: calcium entry. J Nutr 122(3 Suppl):644–650. Van Cromphaut SJ, Dewerchin M, Hoenderop JG, Stockmans I, Van Herck E, Kato S, Bindels RJ, Collen D, Carmeliet P, Bouillon R, Carmeliet G 2001 Duodenal calcium absorption in vitamin D receptor-knockout mice: Functional and molecular aspects. Proc Natl Acad Sci USA 98(23):13324–13329. Wasserman RH, Taylor AN 1966 Vitamin D3–induced calcium-binding protein in chick intestinal mucosa. Science 152:791–793. Kallfelz FA, Taylor AN, Wasserman RH 1967 Vitamin D– induced calcium-binding factor in rat intestinal mucosa. Proc Soc Exptl Biol Med 125:54–58. Wasserman RH, Fullmer CS 1982 Vitamin D–induced calcium-binding protein (CaBP). In: Cheung WY (ed) Calcium and Cell Function, Vol. 2. Academic Press, New York, pp. 175–216, Wasserman RH, Fullmer CS 1983 Calcium transport proteins, calcium absorption, and vitamin D. Annu Rev Physiol 45:375–390. Feher JJ, Wasserman RH 1979 Calcium absorption and calcium-binding protein: Quantitative relationship. Am J Physiol 236(5):E556-E561. Kretsinger RH, Nockolds CE 1973 Carp muscle calciumbinding protein. II. Structure determination and general description. J Biol Chem 248(9):3313–3326. Li YC, Pirro AE, Demay MB 1998 Analysis of vitamin D– dependent calcium-binding protein messenger ribonucleic acid expression in mice lacking the vitamin D receptor. Endocrinology 139(3):847–851. Bar A, Striem S, Vax E, Talpaz H, Hurwitz S 1992 Regulation of calbindin mRNA and calbindin turnover in intestine and shell gland of the chicken. Am J Physiol 262 (5 Pt 2):R800–R805. Striem S, Bar A 1991 Modulation of quail intestinal and egg shell gland calbindin (Mr 28,000) expression by vitamin D3, 1,25-dihydroxyvitamin D3 and egg laying. Mol Cell Endocrinol 75(2):169–177. Taylor AN, Wasserman RH 1967 Vitamin D3–induced calcium-binding protein: Partial purification, electrophoretic visualization, and tissue distribution. Arch Biochem Biophys 119(1):536–540. Wasserman RH 1962 Studies on vitamin D3 and the intestinal absorption of calcium and other ions in the rachitic chick. J Nutr 77:69–80. Wasserman RH, Fullmer CS 1989 On the molecular mechanism of intestinal calcium transport. Adv Exp Med Biol 249:45–65. Armbrecht HJ, Boltz MA, Christakos S, Bruns ME 1998 Capacity of 1,25-dihydroxyvitamin D to stimulate expression of calbindin changes with age in the rat. Arch Biochem Biophys 352(2):159–164. Armbrecht HJ, Boltz M, Strong R, Richardson A, Bruns ME, Christakos S 1989 Expression of calbindin-D decreases with

CHAPTER 24 Vitamin D and the Intestinal Absorption of Calcium: A View and Overview

81.

82.

83. 84.

85. 86.

87.

88.

89. 90.

91. 92.

93. 94. 95. 96.

97.

age in intestine and kidney. Endocrinology 125(6): 2950–2956. Sampson HW, Matthews JL, Martin JH, Kunin AS 1970 An electron microscopic localization of calcium in the small intestine of normal, rachitic, and vitamin-D-treated rats. Calcif Tissue Res 5(4):305–316. Fullmer CS, Chandra S, Smith CA, Morrison GH, Wasserman RH 1996 Ion microscopic imaging of calcium during 1,25-dihydroxyvitamin D–mediated intestinal absorption. Histochem Cell Biol 106(2):215–222. Yue L, Peng JB, Hediger MA, Clapham DE 2001 CaT1 manifests the pore properties of the calcium-release-activated calcium channel. Nature 410(6829):705–709. Niemeyer BA, Bergs C, Wissenbach U, Flockerzi V, Trost C 2001 Competitive regulation of CaT-like-mediated Ca2+ entry by protein kinase C and calmodulin. Proc Natl Acad Sci USA 98(6):3600–3605. Schachter D 1969 Toward a molecular description of active transport. In: Dowben RM (ed) Biological Membranes. Little, Brown, Boston, pp. 157–176. Levine BA, Williams RJP 1982 Calcium binding to proteins and other large biological anion centers. In: Cheung WY (ed) Calcium and Cell Function, Vol. II. Academic Press, New York, pp. 2–38. Levine BA, Williams RJP, Fullmer CS, Wasserman RH 1977 NMR studies of various calcium-binding proteins. In: Wasserman RH, Corradino RA, Carafoli E, Kretsinger RH, MacLennan DH, Siegel FL (eds) Calcium Binding Proteins and Calcium Function Elsevier/North-Holland, Amsterdam, pp. 29–37. Kretsinger RH, Mann JE, Simmonds JG 1982 Model of facilitated diffusion of calcium by the intestinal calcium binding protein. In: Norman AW, Schaefer K, von Herrath D, Grigoleit H-G (eds) Vitamin D: Chemical, Biochemical and Clinical Endocrinology of Calcium Metabolism. de Gruyter, Berlin, pp. 233–248. Feher JJ 1983 Facilitated calcium diffusion by intestinal calcium-binding protein. Am J Physiol 244:303–307. Feher JJ, Fullmer CS, Fritzsch GK 1989 Comparison of the enhanced steady-state diffusion of calcium by calbindin-D9K and calmodulin: Possible importance in intestinal calcium absorption. Cell Calcium 10:189–203. Pansu D, Bellaton C, Roche C, Bronner F 1989 Theophylline inhibits transcellular Ca transport in intestine and Ca binding by CaBP. Am J Physiol 257(6 Pt 1):G935-G943. Koster HPG, Hartog A, Van Os CH, Bindels RJM 1995 Calbindin-D28K facilitates cytosolic calcium diffusion without interfering with calcium signaling. Cell Calcium 18:187–196. Feher JJ, Fullmer CS, Wasserman RH 1992 Role of facilitated diffusion of calcium by calbindin in intestinal calcium absorption. Am J Physiol 262(2 Pt.1):C517–C526. Carafoli E 1991 Calcium pump of the plasma membrane. Physiol Rev 71(1):129–153. Penniston JT, Enyedi A 1998 Modulation of the plasma membrane Ca2+ pump. J Membrane Biol 165(2):101–109. Freedman RA, Weiser MM, Isselbacher KJ 1977 Calcium translocation by Golgi and lateral-basal membrane vesicles from rat intestine: Decrease in vitamin D–deficient rats. Proc Natl Acad Sci USA 74:3612–3616. Ghijsen WEJM, De Jong MD, Van Os CH 1982 ATP-dependent calcium transport and its correlation with Ca2+-ATPase activity in basolateral plasma membranes of rat duodenum. Biochim Biophys Acta 689:327–336.

427

98. Hildmann B, Schmidt A, Murer H 1982 Ca++-transport across basal-lateral plasma membranes from rat small intestinal epithelial cells. J Membrane Biol 65:55–62. 99. Ghisan FK, Leonard D, Pietsch J 1988 Calcium transport by plasma membranes of enterocytes during development: Role of 1,25(OH)2 vitamin D3. Pediatr Res 24:338–341. 100. Van Corven EJJM, de Jong MD, van Os CH 1986 Enterocyte isolation procedure specificially effects ATP-dependent Ca2+transport in small intestinal plasma membranes. Cell Calcium 7:89–99. 101. Walters JR, Weiser MM 1987 Calcium transport by rat duodenal villus and crypt basolateral membranes. Am J Physiol 252:G170-G177. 102. Zelinski JM, Sykes DE, Weiser MM 1991 The effect of vitamin D on rat intestinal plasma membrane Ca-pump mRNA. Biochem Biophys Res Commun 179:749–755. 103. Borke JL, Caride A, Verma AK, Penniston JT, Kumar R 1990 Cellular and segmental distribution of Ca2+-pump epitopes in rat intestine. Pflügers Arch 417(1):120–122. 104. Wasserman RH, Chandler JS, Meyer SA, Smith CA, Brindak ME, Fullmer CS, Penniston JT, Kumar R 1992 Intestinal calcium transport and calcium extrusion processes at the basolateral membrane. J Nutr 122:662–671. 105. Wasserman RH, Smith CA, Brindak ME, de Talamoni N, Fullmer CS, Penniston JT, Kumar R 1991 Vitamin D and mineral deficiencies increase the plasma membrane calcium pump of chicken intestine. Gastroenterology 102:886–894. 106. Armbrecht HJ, Boltz MA, Kumar VB 1999 Intestinal plasma membrane calcium pump protein and its induction by 1,25(OH)2D3 decrease with age. Am J Physiol 277: G41–G47. 107. Cai Q, Chandler JS, Wasserman RH, Kumar R, Penniston JT 1993 Vitamin D and adaptation to dietary calcium and phosphate deficiencies increase intestinal plasma membrane calcium pump gene expression. Proc Natl Acad Sci USA 90:1345–1349. 108. Pannabecker TL, Chandler JS, Wasserman RH 1995 Vitamin D– dependent transcriptional regulation of the intestinal plasma membrane calcium pump. Biochem Biophys Res Commun 213:499–505. 109. Timmermans JA, Bindels RJ, Van Os CM 1995 Stimulation of plasma membrane Ca2+ pump by calbindin-D28k and calmodulin is additive in EGTA-free solutions. J Nutr 125 (7 Suppl):1981S–1986S. 110. Alisio A, Canas F, de Bronia DH, Pereira R, Tolosa de Talamoni N 1997 Effect of vitamin D deficiency on lipid composition and calcium transport by basolateral membrane vesicles from chick intestine. Biochem Mol Biol Int 42(2):339–347. 111. Tolosa de Talamoni NG 1996 Calcium and phosphorous deficiencies alter the lipid composition and fluidity of intestinal basolateral membranes. Comp Biochem Physiol A Physiol 115(4):309–315. 112. Ghijsen WEJM, De Jong MD, Van Os CH 1983 Kinetic properties of Na+/Ca2+ exchange in basolateral plasma membranes of rat small intestine. Biochem Biophys Acta 730:85–94. 113. Wong V, Goodenough DA 1999 Paracellular channels! Science 285:62. 114. Tang VW, Goodenough DA 2003 Paracellular ion channel at the tight junction. Biophys J 84(3):1660–1673. 115. Itoh M, Furuse M, Morita K, Kubota K, Saitou M, Tsukita S 1999 Direct binding of three tight junction-associated MAGUKs, ZO-1, ZO-2, and ZO-3, with the COOH termini of claudins. J Cell Biol 147(6):1351–1363.

428 116. Colegio OR, Van Itallie CM, McCrea HJ, Rahner C, Anderson JM 2002 Claudins create charge-selective channels in the paracellular pathway between epithelial cells. Am J Physiol 283(1):C142–C147. 117. Anderson JM, Van Itallie CM 1995 Tight junctions and the molecular basis for regulation of paracellular permeability. Am J Physiol 269(4 Pg 1):G467–G475. 118. Balda MS, Matter K 2000 Transmembrane proteins of tight junctions. Semin Cell Dev Biol 11(4):281–289. 119. Pansu D, Bellaton C, Bronner F 1981 Effect of Ca intake on saturable and nonsaturable components of duodenal Ca transport. Am J Physiol 240(1):G32–G37. 120. Pansu D, Bellaton C, Roche C, Bronner F 1983 Duodenal and ileal calcium absorption in the rat and effects of vitamin D. Am J Physiol 244(6):G695–G700. 121. Nellans HN 1990 Intestinal calcium absorption. Interplay of paracellular and cellular pathways. Miner Electrolyte Metab 16(2–3):101–108. 122. Wasserman RH, Taylor AN, Kallfelz FA 1966 Vitamin D and transfer of plasma calcium to intestinal lumen in chicks and rats. Am J Physiol 211:419–423. 123. Karbach U 1992 Paracellular calcium transport across the small intestine. J Nutr 122:672–677. 124. Jungbluth H, Binswanger U 1989 Unidirectional duodenal and jejunal calcium and phosphorus transport in the rat: effects of dietary phosphorus depletion, ethane-1-hydroxy1,1-diphosphonate and 1,25-dihydroxycholecalciferol. Res Exp Med (Berl) 189(6):439–449. 125. Chirayath MV, Gajdzik L, Hulla W, Graf J, Cross HS, Peterlik M 1998 Vitamin D increases tight-junction conductance and paracellular Ca2+ transport in Caco-2 cell cultures. Am J Physiol 274(2 Pt 1):G389–G396. 126. Fleet JC, Wood RJ 1999 Specific 1,25(OH)2D3-mediated regulation of transcellular calcium transport in Caco-2 cells. Am J Physiol 276(4 Pt 1):G958–G964. 127. Fleet JC, Eksir F, Hance KW, Wood RJ 2002 Vitamin D– inducible calcium transport and gene expression in three Caco-2 cell lines. Am J Physiol 283(3):G618–G625. 128. Li YC, Bolt MJ, Cao LP, Sitrin MD 2001 Effects of vitamin D receptor inactivation on the expression of calbindins and calcium metabolism. Am J Physiol 281(3):E558–E564. 129. Stenson WF, Easom RA, Riehl TE, Turk J 1993 Regulation of paracellular permeability in Caco-2 cell monolayers by protein kinase C. Am J Physiol 265:G955–G962. 130. Corradino RA 1974 Embryonic chick intestine in organ culture: interaction of adenylate cyclase system and

ROBERT H. WASSERMAN

131.

132.

133. 134. 135. 136.

137. 138.

139. 140. 141. 142. 143.

vitamin D3–mediated calcium absorptive mechanism. Endocrinology 94(6):1607–1614. Wali RK, Baum CL, Sitrin MD, Brasitus TA 1990 1,25(OH)2 vitamin D3 stimulates membrane phosphoinositide turnover, activates protein kinase C, and increases cytosolic calcium in rat colonic epithelium. J Clin Invest 85:1296–1303. Obeid LM, Okazaki T, Karolak LA, Hannun YA 1990 Transcriptional regulation of protein kinase C by 1,25dihydroxyvitamin D3 in HL-60 cells. J Biol Chem 265, 2370–2374. Perez M, Barber A, Ponz F 1997 Modulation of intestinal paracellular permeability by intracellular mediators and cytoskeleton. Can J Physiol Pharmacol 75(4):287–292. Jande SS, Brewer LM 1974 Effects of vitamin D3 on duodenal absorptive cells of chicks. An electron microscopic study. Z Anat Entwickl-Gesch 144, 249–265. Warner RR, Coleman JR 1975 Electron probe analysis of calcium transport by small intestine. J Cell Biol 64: 54–74. Nemere I, Norman AW 1990 Transcaltachia, vesicular calcium transport, and microtubule-associated calbindin-D28K: Emerging views of 1,25-dihiydroxyvitamin D3–mediated intestinal calcium absorption. Miner Electrol Metab 16:109–114. Bronner F, Slepchenko B, Wood RJ, Pansu D 2003 Role of passive transport in calcium absorption. J Nutr 133:1426. Vergne-Marini P, Parker TF, Pak CY, Hull AR, DeLuca HF, Fordtran JS 1976 Jejunal and ileal absorption in patients with chronic renal disease. Effect of 1-alpha-hydroxycholecalciferol. J Clin Invest 57(4):861–866. Barger-Lux MJ, Heaney RP, Recker RR 1989 Time course of calcium absorption in humans: evidence for a colonic component. Calcif Tissue Int 44(5):308–311. Favus MJ, Kathpalia SC, Coe FL, Mond AE 1980 Effects of diet calcium and 1,25-dihydroxyvitamin D3 on colon calcium active transport. Am J Physiol 238(2):G75–G78. Favus MJ, Langman CB 1984 Effects of 1,25-dihydroxyvitamin D3 on colonic calcium transport in vitamin D-deficient and normal rats. Am J Physiol 246(3 Pt 1):G268–G273. Grimstead WC, Pak CY, Krejs GJ 1984 Effect of 1,25-dihydroxyvitamin D3 on calcium absorption in the colon of healthy humans. Am J Physiol 247(2 Pt 1):G189–G192. Petith MM, Wilson HD, Schedl HP 1979 Vitamin D dependence of in vivo calcium transport and mucosal calcium binding protein in rat large intestine. Gastroenterology 76(1):99–104.

CHAPTER 25

Intestinal Calcium Absorption: Lessons from Knockout Mice and Men ROGER BOUILLON, GEERT CARMELIET, AND SOPHIE VAN CROMPHAUT Laboratory for Experimental Medicine and Endocrinology, Katholieke Universiteit Leuven, Leuven, Belgium

I. Introduction II. The Epithelial Calcium Channels: Gatekeepers for Calcium Influx III. Active Intestinal Calcium Absorption: Vitamin D–Dependent Mechanisms

IV. Active Calcium Absorption During Reproduction V. Active Calcium Absorption and Corticosteroids VI. Summary References

I. INTRODUCTION

(KO) [12,13] and 1α(OH)ase KO mice [14,15] and normalized bone structure and strength. Based on these data, it is obvious that vitamin D action on bone is redundant, whereas the small intestine plays an indispensable role in 1,25(OH)2D action on calcium homeostasis and bone mineralization. Calcium is absorbed via two different pathways in the small intestine (Fig. 1). The passive, paracellular pathway through tight junctions is more important in distal parts of the small intestine [16]. The active form of intestinal calcium absorption is energy dependent and is most developed in duodenum. 1,25(OH)2D, through its genomic actions, is regarded as one of the major stimulators of this process, which requires the combined action of at least three groups of calcium transport proteins: (1) calcium influx, which was elusive until the epithelial calcium channels ECaC1 [17–22] and ECaC2 [23–25] were described as possible gatekeepers for calcium entering the cell; (2) intracellular calcium transfer by calbindin-D9K [26–28]; and (3) calcium extrusion by the plasma membrane calcium ATPase (PMCA1b) [29–31]. Furthermore, female mice of one VDR KO strain, namely the Leuven VDR KO, proved to have an estrus cycle and to be fertile. This offered the opportunity to investigate whether estrogens or hormonal changes during the female reproductive cycle influence gene expression regarding duodenal calcium absorption, and if so, whether these effects are vitamin D–VDR dependent or independent and have functional implications. This chapter describes new insights on calcium absorption, which were obtained from the evaluation of VDR [7,9,32–36], 1α(OH)ase [15], estrogen receptor alpha (ERKOα) and beta (ERKOβ) [32], and Na/Pi cotransporter (Npt2) [37] knockout mice.

Human diseases can often be mimicked by mouse genetic studies, which have become powerful tools in unraveling physiological processes at the molecular level. Inevitably, this approach has also been applied in mineral homeostasis research. 1,25(OH)2 vitamin D [1,25(OH)2D] is considered as one of the main hormones for calcium homeostasis and bone mineralization [1]. The genomic effects underlying part of these processes are mediated by the interaction of 1,25(OH)2D with the nuclear vitamin D receptor (VDR) in a ligand-dependent manner [2]. A specific hereditary defect in the gene coding for the 25-hydroxyvitamin D-1α hydroxylase [1α(OH)ase] or CYP27B1 impairs the final, critical step in the biosynthesis of 1,25(OH)2D resulting in vitamin D–dependent rickets type I (VDDR-I), also called pseudo–vitamin D deficiency (PDDR) [3] (see Chaper 71). Mutations in the VDR are responsible for vitamin D–dependent rickets type II (VDDR-II), also called hereditary hypocalcemic vitamin D–resistant rickets (HVDRR) [4] (see Chapter 72). Both of these diseases are transmitted as autosomal recessive traits. Clinical features and response to administrated 1,25(OH)2D are distinct. The phenotypes of mice with targeted ablation of 1α(OH)ase [5,6] or VDR [7–9] exhibit the clinical abnormalities observed in the VDDR-I and -II patients, respectively, and offer the opportunity to investigate the in vivo functions of 1,25(OH)2D and the molecular basis of its actions. The bone pathology in HVDRR can be cured by frequent intravenous calcium infusions [10] and prevented by high oral doses of calcium [11]. Consistent with this standard therapy, a high-calcium diet prevented rickets and hyperparathyroidism in VDR knockout VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

430

ROGER BOUILLON,

3Na1+

ET AL .

NCX1

Ca2+ 1,25(OH)2D

VDR

ECaC2 ATP CaBP9K ADP ECaC1

+

−

PMCA1b

−

+

FIGURE 1 Active intestinal calcium absorption. Mechanisms and 1,25(OH)2D-VDR dependent effects.

II. THE EPITHELIAL CALCIUM CHANNELS: GATEKEEPERS FOR CALCIUM INFLUX A. Introduction The first step in transcellular calcium (re)absorption was not clearly defined until 1999. Moreover, this step was not regarded primordial because ion microscopy images revealed that luminal calcium entered the enterocyte of vitamin D–deficient chickens [38,39]. It became generally accepted that repletion with 1,25(OH)2D induces the synthesis of calbindin D, which subsequently facilitates the diffusion of calcium in a bound form from the brush border region through the cytosol to the vicinity of the PMCA1b [39]. However, in 1999, the epithelial calcium channels ECaC1 and ECaC2 (or calcium transport protein type 1, CaT1)—in the unified nomenclature renamed transient receptor potential vanilloid 5 (TRPV5) and TRPV6 [40]—were cloned from kidney [17] and small intestine [23], respectively. They are two new members of the Transient Receptor Potential Vanilloid (TRPV) family, and they are the most calcium selective of the TRP superfamily.

B. The Transient Receptor Potential (TRP) Channel Superfamily The TRP superfamily consists of a variety of calcium permeable cation channels, which may be involved in

calcium influx. Calcium influx across the plasma membrane is important for several physiologic processes, such as apoptosis, cell proliferation, and modulation of cell cycle, but also for ion homeostasis. These non-voltage-gated cation channels vary significantly in their selectivity of activation, but share structural features (amino-terminal ankyrin repeats, six transmembrane domains, a short hydrophobic stretch between S5 and S6 representing the pore region; Fig. 2) and the permeability for cations and calcium. The TRP superfamily can be divided into six subfamilies [24, 41, 42]. The first is composed of the classical TRPs: TRPC (C for canonical). The archetypical TRP, Drosophila TRP, belongs to the TRPC family and is required for phototransduction. Some TRPCs may be store-operated channels (SOCs), which are activated by release of calcium from internal stores. The mammalian TRPC proteins also are expressed in the central nervous system or are implicated in the pheromone response. Most members of the second subfamily TRPV (V for vanilloid) function in sensory physiology. These include vanilloid receptor VR1, human vanilloid receptor-like protein VRL-1, and OSM-9, which respond to heat, osmolarity, odorants, and mechanical stimuli. The third subfamily TRPN includes proteins with many ankyrin repeats, one of which, NOMPC, participates in mechanotransduction. The fourth subfamily, TRPM, is named after a putative tumor suppressor: melastatin. PKD2 and mucolipidin are the founding members of the TRPP and TRPML. Mutations in PKD2 and mucolipidin result in polycystic kidney disease and a severe neurodegenerative disorder, respectively.

431

CHAPTER 25 Calcium Absorption and KO Models

A Outside

Inside

A A A

COOH

NH2

B

Topside

FIGURE 2 ECaC architecture. The predicted topology of the ECaC1/2 monomer (A) and architecture of the ECaC1/2 (hetero)tetrameric channel complex (B). (A) Transmembrane segments are numbered and ankyrin repeats in the N-terminal tail are indicated with A. (B) Four monomeric ECaC1/2 subunits form a tetrameric complex; as a consequence, four aspartate residues (D542) form the calcium-binding site within the pore of the channel. (With permission, Cell Calcium, 2003, 33:504 [52].)

account for the basal calcium uptake by these cells. Noteworthy is the fact that ECaC2 mRNA is much more abundant in rodent placenta than ECaC1 mRNA, as assessed by quantitative RT-PCR (unpublished personal data) and its expression is not influenced by the genotype of the mother. Data on ECaC2 expression in placenta of VDR KO pups of VDR KO mothers are presently lacking. In mice, ECaC2 is expressed more ubiquitously than in humans, whereas ECaC1 expression is restricted to the kidney. The relative mRNA levels of ECaC1 and ECaC2 in tissues coexpressing these channels are different. In duodenum and other parts of the small intestine, ECaC2 mRNA level is much higher than that of ECaC1 mRNA [9] and ECaC1 seems to be the major isoform in kidney, although both isoforms are coexpressed in these tissues. Detailed investigations of protein levels and activity measurements will provide more clarity on this issue. In the mouse and human gastrointestinal tract, ECaC2 expression is restricted essentially to the epithelial cells [45]. In the duodenum, ECaC2 and ECaC1 proteins are present in a thin layer along the apical membrane of the villus tips [18,45]. As this is the major site for calcium absorption, ECaC1/2 expression patterns are in line with the role of these proteins in dietary calcium uptake. In the kidney, ECaC1 is localized in the apical domain of the epithelial cells lining the late part of the distal convoluted tubule and the connecting tubule [18,43]. In these nephron segments ECaC1 colocalizes with other proteins involved in calcium transport, such as calbindin-D28K, sodium-calcium exchanger (NCX), and PMCA1b [18].

C. Tissue Distribution of ECaC1 and ECaC2

D. Functional Properties of ECaC1 and ECaC2

The tissue distribution of ECaC1 and ECaC2 has been studied extensively by Northern blot, RT-PCR analysis, and immunohistochemistry [19–21,23,36,43,44]. In humans, both channels are coexpressed in the organs that mediate transcellular calcium transport, such as duodenum, jejunum, colon, and kidney, but also in exocrine tissues such as pancreas, prostate, mammary gland, sweat gland, and salivary gland. Epithelial calcium channels present in the placenta might be involved in placental transfer of maternal calcium. The placental syncytiotrophoblast layer carries out this crucial process in fetal development. The fact that mRNA of ECaC1 and ECaC2 is expressed in this layer may indicate that these calcium channels

When expressed in Xenopus laevis oocytes or human embryonic kidney (HEK) 293 cells, ECaC1 and ECaC2 mediate a saturable calcium uptake and manifest distinct electrophysiological features, including hyperpolarization-dependent calcium entry and calcium-dependent inactivation [17,46–49]. ECaC1 and ECaC2-expressing HEK293 cells display large inward currents that are strongly dependent on extracellular calcium and reverse at high positive membrane potentials [49]. The current/voltage relationship shows prominent inward rectification. Thus, under unstimulated physiological conditions, when the membrane potential is typically around –70 mV in the distal part of the nephron, ECaC1 constitutes a substantial

432 calcium conductance permitting basal calcium influx. Studies have shown that ECaC1 and ECaC2 exhibit a similar ion permeation sequence for divalent cations (Ca2+ > Sr2+ ~ Ba2+ > Mn2+) [49]. The characteristic pore region of ECaC1 and ECaC2 is unique for its high calcium selectivity. A single aspartatic residue in the pore region at position 542 (D542) is crucial for calcium permeation. Mutation of D542 to alanine abolishes calcium permeation, but does not affect the permeation of monovalent cations [48]. This amino acid is completely conserved during evolution in ECaC2 [21]. From an electrophysiological point of view ECaC1 and ECaC2 are highly similar, in line with the 75% homology at the amino acid level between these channels [20]. However, three distinctive differences have been found. Calcium-dependent inactivation of ECaC2 consists of a fast initial phase followed by a slower phase of inactivation, whereas ECaC1 displays only this slow inactivation behavior [50]. A second difference is the barium selectivity that results in a significantly higher Ba2+/Ca2+ current ratio for ECaC1 than for ECaC2. Finally, a pharmacological distinction can be made between ECaC1 and ECaC2 with ruthenium red, for which ECaC2 has a 100-fold lower affinity than ECaC1.

E. Molecular Architecture of the Epithelial Calcium Channels The oligomerization of ECaC1 and ECaC2 has only recently been unraveled [51]. Cross-linking studies, co-immunoprecipitation, and molecular mass determination of ECaC1/2 using sucrose gradient sedimentation have shown that ECaC1 and ECaC2 form homo- and heterotetrameric channel complexes (Fig. 2). This tetrameric architecture of ECaC1/2 implies that four of the aspartatic residues (D452) form a negatively charged ring that functions as a selectivity filter for calcium in analogy with voltage-gated calcium channels [52]. ECaC1 and ECaC2 are co-expressed in several tissues, which allows oligomerization of these channels in vivo, which subsequently may influence the channels properties. ECaC1 and ECaC2 exhibit different channel kinetics with respect to calcium-dependent inactivation, barium selectivity, and sensitivity for inhibition by ruthenium red. Concatemers were constructed consisting of four ECaC1 and ECaC2 subunits in a head-to-tail fashion. An increased number of ECaC1 subunits in such a concatemer displayed more ECaC1-like properties, indicating that the stoichiometry of ECaC1/2 heterotetramers influences the channel

ROGER BOUILLON,

ET AL .

properties [51]. Consequently, regulation of the relative expression levels of ECaC1 and ECaC2 may be a mechanism for fine-tuning calcium transport kinetics in ECaC1/2 expressing tissues.

F. Association of the S100A10–Annexin 2 Complex Is Required for Functional Expression Using yeast two-hybrid and GST pull-down assays, the S100A10 protein was identified as the first auxiliary protein of the epithelial calcium channels. S100A10 and annexin 2 were consistently detected in ECaC1 expressing distal convoluted and connecting tubules in the kidney [53]. Furthermore, the S100A10– annexin 2 complex was present along the brush border membrane of duodenum, which is in agreement with the ECaC2 localization [53]. Actually, S100A10 forms a heterotetrameric complex with annexin 2 and associates with the conserved sequence VATTV located in the C-terminal tail of ECaC1 and ECaC2. Of these five amino acids, the first threonine plays a crucial role. The activity of ECaC1 and ECaC2 was abolished when this particular threonine was mutated, demonstrating that this motif is essential for channel function. Malfunctioning of these mutant channels was accompanied by redistribution to a sub–plasma membrane area [53]. These findings suggest a role for the S100A10-annexin 2 heterotetramer in the trafficking process of ECaC1 and ECaC2 to the plasma membrane. Whether this involves facilitated translocation of ECaC1 and ECaC2 channels to the plasma membrane or enhancement of channel retention remains to be elucidated. Moreover, down-regulating annexin 2 expression using specific small interfering RNA inhibited ECaC1 and ECaC2 mediated currents in transfected HEK293 cells. Finally, quantitative PCR analysis of kidney samples from 1α(OH)ase KO mice showed that treatment with 1,25(OH)2D significantly up-regulated the mRNA abundance of S100A10, analogous to ECaC1 expression [53].

G. Summary In conclusion, the epithelial calcium channels are primarily expressed in calcium-transporting epithelia. Structural analysis and electrophysiological studies suggest that they play an essential role in calcium entry as a part of the calcium (re)absorption, thus contributing to calcium homeostasis.

433

CHAPTER 25 Calcium Absorption and KO Models

III. ACTIVE INTESTINAL CALCIUM ABSORPTION: VITAMIN D–DEPENDENT MECHANISMS A. Vitamin D Receptor Knockout Mice 1. GENERATION OF VDR-NULL MICE

Four different groups (Tokyo, Boston, Leuven, and Munchen, respectively [7–9,36]) created VDR null mice by ablation of either exon II/first zinc finger [7,9,36] or exon III/second zinc finger [8] in the DNA binding site of this nuclear receptor. The Leuven VDR-null mice were generated using the Cre-lox system. VDR null mice of the Boston, Leuven, and Munchen strain exhibit nearly normal survival rates for up to 6 months, whereas affected homozygotes of the Tokyo strain die within 15 weeks [7]. The differing survival curves are not completely related to diet or environmental variations, but may partly be related to strain differences: the Tokyo VDR-null mice housed in the facilities of the Leuven VDR-null mice still suffer from a more important failure to thrive compared with the Leuven strain [9]. However, all VDR-null mouse strains show growth retardation starting after weaning. 2. MINERAL HOMEOSTASIS AND RICKETS OF VDR-NULL MICE

VDR-null mice are born phenotypically normal, despite expression of VDR in fetal life from embryonic day 10.5 onward in several tissues of WT mice. After weaning, VDR-null mice develop hypocalcemia and hyperparathyroidism, as evidenced by elevated parathyroid hormone (PTH) levels in serum, an increase in the size of the parathyroid glands as well as an increase in PTH mRNA levels [7–9]. Concomitantly, hypophosphatemia develops and 1,25(OH)2D serum levels become 10- to 100-fold elevated, coincident with extremely low 24,25(OH)2D serum levels. These latter data are consistent with earlier findings that the activities of 1α(OH)ase and 24-hydroxylase are regulated negatively and positively, respectively, by 1,25(OH)2D [54,55]. Long bones showed a reduced length and the typical features of rickets, including widening of epiphyseal growth plates, thinning of the cortex, and cupping and widening of the metaphysis. Amling and co-workers [13] studied histomorphometric and biomechanical parameters in Boston VDR-null mice fed regular chow. The following observations were made in 70-day-old animals: (a) 85% of bone surface covered with osteoid; (b) a marked increase in bone volume, due to increased osteoid; (c) an increase in osteoblast number and increased surface of bone covered by osteoblasts; (d) normal osteoclast number and surface

of bone covered by osteoclasts; (e) impaired mineral apposition rate; (f) reduced stiffness and increased ultimate deformation; (g) expansion of length of growth plate (+48%) with disorganization of the chondrocyte columns and increased matrix volume. The time of onset of symptoms at 3 weeks of age fits with the observation that in neonatal rats intestinal calcium absorption occurs by a 1,25(OH)2D-independent mechanism and is gradually replaced by a 1,25(OH)2D-dependent mechanism [56]. Normalization of mineral ion homeostasis (calcemia and phosphatemia) by feeding these mice a diet enriched with 2% calcium, 20% lactose, and 1.25% phosphorus (the so-called “rescue diet”) prevented hyperparathyroidism and rickets [9,12,13,33]. Normal morphology in the growth cartilage and adjacent metaphysis and normal biomechanical competence of cortical bone were demonstrated. This is consistent with the prevention and treatment of human VDRR type II with high doses of calcium [10,11] (see Chapter 72). These data suggest that, although 1,25(OH)2D has been shown to have significant effects on the expression of genes by osteoblasts and osteoclasts, the principal action of VDR in skeletal growth, maturation, and remodeling is its role in intestinal calcium absorption (Fig. 1) and/or renal calcium reabsorption. 3. CALCIUM ABSORPTION IN VDR-NULL MICE: FUNCTIONAL ASPECTS

The beneficial effect of the rescue diet aroused the question whether intestinal calcium absorption was indeed impaired in VDR-null mice. Boston VDR-null exhibit a similar calcium disappearance from the whole intestinal tract, measured 1.5, 3, and 6 hr after administration, compared with WT littermates [34]. The amount of 45Ca in the sera and femurs at these time points is three to four times lower in these VDR-null mice, consistent with the hypocalcemia. Boston VDR-null mice display a normal urinary calcium excretion, which is to be considered inappropriately high taking the hypocalcemia into account. Urinary calcium excretion of the VDR-null mice becomes even twice as high as WT levels after 1 week of a rescue diet that normalizes serum calcium. This suggests that VDR inactivation has little effect on the total intestinal calcium absorption under these experimental conditions and that the VDR-null phenotype is due to a defect in renal calcium conservation. A similar or even higher urinary calcium/creatinine ratio is also noted in Leuven [32] and Tokyo VDR-null [33] mice on a normal diet, compared to the respective WT littermates. By contrast, this ratio is reduced to one fifth in Tokyo VDR-null mice on a rescue diet since weaning,

434

ROGER BOUILLON,

compared to Tokyo WT mice on the same high-calcium diet [33]. Moreover, the setup of the study in the Boston VDR-null mice allows only a crude estimation of calcium absorption, and cannot exclude a reduction in active duodenal calcium absorption. Marcus and co-workers determined that following administration of a single oral dose of tracer calcium, 62% was absorbed in the ileum, 23% in jejunum, and only 15% in duodenum. The amount absorbed depended on the residence time in the particular segment and the rate of absorption. The latter is highest in duodenum, followed by jejunum, and is lowest in ileum [57]. Furthermore, it is generally accepted that active duodenal calcium absorption rather intervenes in maintaining mineral homeostasis when the organism is deprived of a consistent and plentiful supply of calcium. This occurs in mammals mainly after weaning. As already previously mentioned, the VDR KO phenotype only becomes apparent after weaning [7–9] on a normal diet and is rescued by administration of the high-calcium diet. Hence, defects in active calcium

A

ECaC2

absorption could definitely have important implications for mineral homeostasis. Active calcium absorption was assessed in the Leuven and Tokyo VDR-null strain by measuring appearance of 45Ca in serum at several time points between 2 and 10 min after oral gavage [9] (Fig. 3B, right panel). The obtained area under the curve (AUCpo0–10 min) is used to quantify the absorption. This AUCpo0–10 min is, compared to the respective WT mice, reduced to one-third of normal, whether measured in the Leuven VDR-null mice (on a low-calcium diet for 1 week (AUCpo0–10 min WT: 13.3 ± 0.9 ∆µmol min vs. KO: 4.4 ± 0.4 ∆µmol min)) or in Tokyo VDR-null mice [kept on a normal calcium diet (AUCpo0–10 min WT: 7.0 ± 0.4 ∆µmol min vs. KO: 2.3 ± 0.3 ∆µmol min)]. On the other hand, calcium kinetics during the first 10 min after intravenous bolus in the Leuven VDR-null animals is indistinguishable from kinetics in WT mice (AUCiv1–10 min WT: 10.4 ± 0.6 ∆µmol min vs. KO: 9.2 ± 0.4 ∆µmol min) (Fig. 3B). Hence, low serum 45Ca concentrations in the VDR KO mice are not due to an enhanced clearance from the blood compartment.

Calbindin-D9K

*

ET AL .

PMCA1b

50

*

*§

* *§

0

2

100 * 50

1

*§

*§ * * *§

0

0

WT KO

WT KO

WT KO

WT KO

WT KO

WT KO

Low Ca

Normal

Rescue

Low Ca

Normal

Rescue

∆µMol serum calcium

B

WT Low Ca diet WT Normal diet WT Rescue diet

2.5

*

2.0

*

*

* *

1.5 1.0 0.5 0.0 0

* 2

*

*

*

*

4

6

8

10

Time (minutes)

∆µMol serum calcium

Diet:

3

150

mRNA (% WT Nl diet)

100

4

*

200

Calb/Tot prot (%)

650

mRNA (% WT Nl diet)

mRNA (% WT Nl diet)

700

100 *

50

*

*§

0 WT KO

WT KO

WT KO

Low Ca

Normal

Rescue

VDR WT per os VDR KO per os VDR WT i.v. VDR KO i.v.

2.5 2.0 1.5 1.0 0.5

x

x

x x

x

8

10

0.0 0

2

4

6

Time (minutes)

FIGURE 3 Effect of VDR inactivation and dietary intervention on active intestinal calcium absorption. (A) Gene expression patterns. (B) Functional aspects: in vivo calcium absorption assay. Calbindin-D9K expression: mRNA levels: rising right, solid bars, and rising left pattern; protein levels: diagonal cross-hatch, horizontal lines, and horizontal cross-hatch pattern. *P < 0.05 vs. WT normal diet, §P < 0.05 vs. KO normal diet, xP < 0.05 vs. WT low-calcium diet.

435

CHAPTER 25 Calcium Absorption and KO Models

Interestingly, the recorded absolute values in calcium appearance and area under the curve in the Tokyo strain are 50% lower than in the Leuven strain. The older age of these animals, the normal calcium diet, and strain differences are possible explanations for these findings [9]. Song et al. assessed active duodenal calcium absorption in Tokyo VDR-null mice by means of the in situ ligated loop technique [33]. In this study, active duodenal calcium absorption is strongly age dependent and falls to a third when comparing 2- and 3-month-old WT mice. Compared to WT mice, calcium absorption in VDR KO mice fed a normal calcium diet is reduced by 71% in 2-month-old mice (WT: 55.4 ± 3.8% vs KO: 16.3 ± 1.6%) and by 87% in 3-month-old mice (WT: 18.0 ± 0.9% vs KO: 2.3 ± 1.1%) [33]. Moreover, in the 3-month-old VDR KO mice calcium absorption is equal to the calculated rate of paracellular diffusion (13.8%/hr [58]). Feeding the high-calcium diet to the WT group caused a 65% reduction in the efficiency of intestinal calcium absorption in 2- and 3-month-old WT mice (WT 2 month: 20 ± 1.7%, WT 3 month: 6.3 ± 0.5%). This rescue diet induced relatively higher levels of calcium absorption in VDR KO mice, compared to KO mice on normal diet (KO 2 month: 18.4 ± 1.5%, KO 3 month: 10.0 ± 1.4%), which is more or less comparable to levels in VDR WT animals on the high-calcium diet [33]. The authors suggest that a constraint imposed upon transcellular intestinal calcium absorption in VDR KO mice fed the normal calcium diet is removed by the rescue diet. However, they also mention that between 2 and 3 months of age VDR KO mice do not gain any body weight when they are raised on a normal calcium diet; hence it is not excluded that the extremely low rate of calcium absorption seen at 3 months is caused by a general failure to thrive. These two different studies thus indicate that active duodenal calcium absorption is clearly reduced in Leuven [9] and Tokyo VDR KO [9,33] mice to about one-third of WT levels. Administration of a rescue diet also leads to a 60% decrease in active duodenal calcium absorption in WT mice [9,33]. Tokyo VDR WT and KO mice on this diet absorb comparable amounts of calcium. One may postulate that the values recorded on the high calcium diet are indicative of a basal level of calcium absorption, such as occurs when the calcium uptake and conservation mechanisms are shut down (in the case of vitamin D endocrine system) because the organism is supplied with plenty of calcium. Song et al. [33] calculated that this level of calcium absorption is too important to be exclusively passive. Further investigation will be necessary to determine whether, even when the drive for active calcium absorption is low, part of the calcium is taken up via a transcellular

pathway, which would be constitutively present and independent from vitamin D–VDR action. 4. CALCIUM ABSORPTION IN VDR-NULL MICE ON NORMAL CALCIUM DIET: MOLECULAR ASPECTS

Subsequently, attempts were made to unravel the molecular basis of this impaired active duodenal calcium absorption. Obviously, the three first targets to investigate were (a) ECaC2 and ECaC1; (b) calbindin-D9K; and (c) PMCA1b (Fig. 1). The expression of the epithelial calcium channels [9,33,36] and PMCA1b [9] was only quantified at RNA level. Calbindin-D9K expression was assessed at the RNA and protein levels (by Western blotting and RIA) [7,9,33–35]. Previous studies showed that 1,25(OH)2D repletion of vitamin–deficient chickens or rats led to an increase in calbindin-D9K [26–28] and PMCA1b expression [16,29–31]. Intriguingly, PMCA1b expression in Leuven and Tokyo VDR-null mice is comparable to WT littermates (Fig. 3A, Table I) [9]. The defect in calbindin-D9K expression is not consistent. Different groups established decreased duodenal calbindin-D9K RNA and protein levels in Tokyo VDR-null [7,9,33] (between 50 and 30% of WT levels; Table I). Initially, the same observation was made in Boston VDR-null mice [35] with a decrease of calbindin-D9K RNA expression from 50 to 15% between 2 and 11 weeks of age (Table I). A later study did not confirm these data: expression is reduced only in young animals, but by the age of 6 or 9 weeks the expression levels are comparable to those of WT littermates (Table I) [34]. Previously, it was already shown in Leuven VDR-null mice that the normal calbindin-D9K protein [with a slight decrease in calbindin-D9K RNA level (62%)] content could not account for the HVDRR phenotype of these mice (Fig. 3A, Table I) [9]. In contrast, duodenal ECaC2, and to a lesser degree ECaC1, mRNA levels are markedly reduced in the Leuven and the Tokyo VDR-null mice to less than 10% [9] and 5% [9,33] of their WT littermates, respectively (Fig. 3A, Table I). In duodenum of Munchen VDR-ablated mice, ECaC2 is also significantly decreased, but only by about threefold compared with WT mice (Table I) [36]. Nevertheless, the latter investigators used a different method for quantification of duodenal RNA samples. Data on ECaC2 expression in the Boston VDR-null mice are lacking at present. Hence, unexpectedly, among the calcium absorption mechanisms, only the expression of the epithelial calcium channels is dramatically and consistently downregulated in three VDR KO strains. At first glance, this seems contradictory to former functional and molecular data suggesting that regulation of calbindin-D9K

TABLE I Differences in Gene Expression Pattern of the Three Calcium Transporter Genes in the 4 Different VDR KO Strains on Varying Calcium Diets: An Overview of the Literature Mouse

Leuven VDR KO [9] (10 w)

Tokyo VDR KO [7,9,33]

Boston VDR KO [34,35]

Diet

Low Ca

Normal

Rescue

Normal

Rescue

ECaC2 mRNA

↓1/500

↓1/10

↓1/3

10 w: ↓1/30 [9] 2 m: ↓1/20 [33]

2 m: ↓1/26 [33]

Calb-D9K mRNA

↓1/20

↓62%

=

3 and 7w: ↓ 1/3 [7] 10 w: ↓1/3 [9] 2 m: ↓1/2 [33]

Calb-D9K protein

PMCA1b mRNA

↓1/5

↓1/2

=

=

=

↓1/2

Normal

10 w: ↓1/3 [9] 2 m: ↓1/2 [33]

2 m: =

[33]

2 m: ↓69% [33]

2→11w: [35] ↓1/2→1/6 3 w: 1/5 [34] 6–18 w: = [34] 3–9 w: 1/2 [34] 13–18 w: = [34]

10 w: = [9]

Expression levels are compared to levels of WT littermates under the same dietary conditions. Age: weeks (w) and months (m).

Rescue

5→11w: = [35]

Munchen VDR KO [36] (Age?) Normal

Rescue

↓1/3

=

437

CHAPTER 25 Calcium Absorption and KO Models

expression is rate-limiting for vitamin D–dependent active calcium absorption. Nonetheless, revision of the literature reveals some flaws in this theory, which could be completed with the following new notions. For instance, after 1,25(OH)2D injection, a lack of correlation was found between the early rise in calcium transport and the later increase in calbindin-D9K content [26,59]. Likewise, the first event uncovered on ion microscopic images following 1,25(OH)2D-repletion of vitamin D–deficient animals is an intensification in calcium accumulation at the brush border region [38]. There is a significant lag between the time of hormone administration and redistribution of calcium away from the brush border, which coincides with the induction of calbindin-D9K synthesis [38]. In normal, vitamin D replete mice, a single pharmacological injection of 1,25(OH)2D (2 µg/kg body weight) induced the expression of all three calcium transporters in duodenum (Table II), which correlated with a doubling of the functional calcium absorption [9]. Interestingly, when evaluating the duodenal gene expression pattern over time, ECaC2 expression was already maximal at 6 hr (sixfold increase), whereas calbindin-D9K protein levels only peaked at 24 hr (threefold induction) PMCA1b expression was raised, as well, but to a lesser extent ([9], Table II, and unpublished personal data). In agreement with this data, Wood and co-workers showed that treatment of the human colon adenocarcinoma cell line Caco-2 with 1,25(OH)2D rapidly up-regulated ECaC2 expression (4 hr: fourfold; 34 hr: 10-fold, without any detection of ECaC1), which preceded by several hours the doubling of calbindinD9K [60]. Finally, calcium absorption in vivo and calbindin-D9K protein levels correlated on the one hand, fairly well in several experimental conditions (Fig. 3) [9,33], on the other hand, this correlation could not be established in Tokyo VDR KO mice on a 0.5% calcium diet [33].

TABLE II 1,25(OH)2D Injection and Duodenal Gene Expression Patterns in WT Mice

ECaC2 mRNA Calbindin-D9K mRNA Calbindin-D9K RIA PMCA1b mRNA

Vehicle

1,25(OH)2D3 6 hr

1,25(OH)2D3 24 hr

1 ± 0.2 1 ± 0.2 0.9 ± 0.2 1 ± 0.1

6.5 ± 0.8* 2.2 ± 0.3* 0.8 ± 0.18 1.8 ± 0.2*

6.2 ± 2.2* 42.5 ± 6.7* 3.1 ± 0.1* 1.4 ± 0.2

Time-response differences measured in WT mice (n = 6/group). mRNA levels are expressed relative to levels in vehicle-treated mice. CalbindinD9K protein levels are expressed as % of total duodenal protein levels. *P < 0.05 vs. WT vehicle.

In summary, the study of gene expression patterns in duodena of VDR-null mice implies a crucial role for the epithelial calcium channels in active calcium absorption, probably combined with a more complex regulation driven by the interaction between calcium influx and intracellular calcium transfer (Fig. 1). In this way the calcium buffering capacity of the cell keeps step with the increase in apical calcium entry and serves to maintain a low calcium concentration in close proximity of the influx pathway. Indeed, increased intracellular free calcium concentration would lead to negative feedback regulation of ECaC2and ECaC1-mediated channel activity [49,61,62]. Hence, in the absence of sufficient calcium buffering capacity, ECaC2 and ECaC1 functional activity would decrease and nullify the increased expression of ECaC channels at the apical surface. This is a prerequisite for all cells involved in transcellular calcium transport since the increase in intracellular calcium concentration associated with enhanced apical calcium uptake would otherwise be cytotoxic. Further research, including the characterization of the phenotype of ECaC1, ECaC2, or calbindin-D9K null mice or calcium channel rescue of VDR KO mice, will be necessary to support this concept. 5. DIETARY INTERVENTIONS AND ADAPTATION CALCIUM ABSORPTION IN VDR WT AND KO MICE

OF INTESTINAL

It is generally accepted that the primary homeostatic mechanism for stimulating active calcium absorption in response to low dietary calcium intake is an increase in 1,25(OH)2D. Conversely, when calcium is administered abundantly, no drive for active calcium transport would be present. This pattern is indeed observed when in vivo calcium absorption is measured in WT Swiss mice. One week of 0.02% calcium diet doubles, min whereas 1 week of rescue diet halves, the AUC0–10 po when compared with values of Swiss mice on a normal calcium diet (Fig. 3B, left panel) [9]. The in situ ligated loop procedure in 3-month-old Tokyo WT mice demonstrates analogous effects: feeding a 2% calcium diet for 1 week suppresses calcium absorption by 75%, while feeding a 0.02% calcium diet for 1 week increases calcium absorption 2.3-fold (which was associated with concordant changes in plasma 1,25(OH)2D [33]. The influence of dietary calcium on duodenal expression levels of the calcium transporter genes was extensively studied in the Leuven strain (Fig. 3A, Table I) [9]. First, WT animals present with a significant 2.5-fold increase of calbindin-D9K protein content on the low-calcium diet and a more than 70% decrease in calbindin-D9K RNA and protein level on the rescue diet. In the KO animals consuming either high- or

438 low-calcium diet, calbindin-D9K RNA level and protein levels decrease to less than 40% of KO levels on a normal diet. Second, PMCA1b gene expression is significantly decreased (by two to three times) in WT mice on the low-calcium diet and in KO mice on both low- and high-calcium diet. Third, ECaC2 expression varies considerably in WT mice: a sixfold increase with calcium restriction and a 90% reduction with calcium abundance, compared to levels of WT mice on normal diet. Although ECaC2 expression is already severely impaired in VDR KO animals, it has still a tendency to further decline on both low- and highcalcium diet (Fig. 3A, Table I). Despite less pronounced differences and lower detection levels, a similar pattern is found for ECaC1 RNA levels in the different conditions [9]. Administration of the rescue diet results in normalization of ECaC2 RNA expression in the Munchen VDR KO [36], and of calbindin-D9K RNA levels in the Boston KO (Table I) [35]. Though calbindin-D9K levels are almost indistinguishable from WT levels, ECaC2 expression remains markedly lower in rescued Tokyo VDR KO animals, compared to WT littermates on the same diet [33] (Table I). On the whole, duodenal expression patterns in the different VDR KO models tend to resemble the WT profile on the rescue diet (except for ECaC2 expression in the Tokyo VDR KO). By contrast, lack of adaptation to the low-calcium diet accentuates the difference in duodenal gene expression pattern between Leuven VDR WT and KO animals, and especially, the discrepancy in ECaC2 expression (Table I). Hence, ECaC2 (and ECaC1) expression is not only strongly VDR-dependent and consistently downregulated in VDR-KO strains, but it fluctuates also markedly in WT animals according to dietary calcium-driven variations in 1,25(OH)2D [9,33]. This provides additional evidence that calcium influx, interacting with intracellular calcium transfer mechanisms, should be considered as the rate-limiting step in the process of vitamin D–dependent duodenal calcium absorption.

B. Deficiency in 1,25(OH)2D Production (vitamin D–Dependent Rickets Type I) 1. GENERATION OF 1α(OH)ASE-NULL MICE

The engineering of 1α(OH)ase-null mice was reported by two different groups. The 1α(OH)ase gene was ablated by targeting part of exon 7 and 8 [5], or exons 6 to 9 [6]. In both cases the heme-binding domain and part of or the complete hormone-binding

ROGER BOUILLON,

ET AL .

domain were knocked out. No exact data are available on life expectancy in these mice, but both strains demonstrate failure to thrive (see Chapter 7). 2. MINERAL HOMEOSTASIS IN 1α(OH)ASE-NULL MICE

Regarding calcium homeostasis both 1α(OH)ase null models are identical. They demonstrate hypocalcemia, severe secondary hyperparathyroidism, hypophosphatemia, elevated serum alkaline phosphatase and increased phosphaturia, analogous to the phenotype of the VDR-null mice [5,6]. However, 1α(OH)ase-null mice genuinely lack the capacity to synthesize 1,25(OH)2D: circulating levels of 1,25(OH)2D were undetectable, whereas circulating levels of 25(OH)D doubled [5,6]. Typical features of advanced rickets are observed histologically in bone of 1α(OH)ase-null mice: disorganization and widening of the columnar alignment of hypertrophic chondrocytes, resulting in increased width of the growth plate, impaired calcification of the hypertrophic cartilage, and accumulation of osteoid in trabecular and cortical bone [5,6]. This emphasizes the nonredundant role played by the renal 1α(OH)ase in producing the hormonally active metabolite of vitamin and shows that the pathology of the 1α(OH)ase-null mice is the equivalent of the human VDDR-I disease. The treatment of choice for VDRR-I is long-term replacement therapy with 1,25(OH)2D. This strategy was also applied on the 1α(OH)ase-null mice [15,63]; the effect of the rescue diet was also evaluated [14,15,64]. Either treatment with subcutaneous 1,25(OH)2D, or feeding the high-calcium and lactose diet effectively normalizes biochemical parameters (calcemia, hyperparathyroidism), restores the biomechanical properties of bones, corrects growth plate architecture, and cures rickets and osteomalacia in the 1α(OH)ase-null mice generated by the group of R. St-Arnaud [14,63]. Intriguingly, under the rescue diet, the overall size of the femur is the only parameter that is not normalized in this 1α(OH)ase KO strain [14]. Presumably, 1,25(OH)2D treatment stimulates active calcium absorption in these mice, whereas on the rescue diet calcium is delivered via the passive pathway. Passive calcium uptake may not be able to adequately meet the demands for calcium during rapid growth spurts in this animal model. However, the latter phenomenon is only observed in the 1α(OH)ase-null mice and not in the VDR KO mice. The team of Goltzman report also that despite normalization of calcemia and bone mineral content with the rescue diet, PTH levels remain elevated and the growth plate remains larger and more distorted in their specific 1α(OH)ase-null strain [64].

439

CHAPTER 25 Calcium Absorption and KO Models

It is not excluded that strain-specific effects intervene or that the background of the KO mice is different. As already mentioned, a basal level of active calcium transport may be present (as the three steps of transcellular calcium transport are still expressed in the respective KO mice, though at lower levels), which may be strain-related. When comparing VDR KO, 1α(OH)ase KO and compound VDR-1α(OH)ase KO mice, Goltzman and colleagues found that hypocalcemia correlates with the absence of VDR and bone mineralization correlates with ambient calcium levels, but that the increase in trabecular volume and osteoblast number preferentially correlates with the absence of 1α(OH)ase activity [64]. Moreover, the compound VDR–1α(OH)ase KO is definitely more affected; suffering from more extreme hypocalcemia and failure to thrive, compared with the single KO models [64]. Unfortunately, data regarding the duodenal phenotype of these compound VDR-1α(OH)ase KO mice are lacking at present. Subsequently, St-Arnaud et al. engineered a mouse with selective inactivation of the 1α(OH)ase enzyme in growth plate chondrocytes in order to investigate the putative autocrine/paracrine roles of 1,25(OH)2D in the growth plate under conditions of normal mineral ion homeostasis. First examination of the bones from these mutant animals does not reveal any gross abnormalities [65]. 3. CALCIUM ABSORPTION IN 1α(OH)ASE-NULL MICE

Precise histomorphometric analysis is required to assess the impact of chondrocyte-specific inactivation on 1α(OH)ase on the maturation and function of growth plate cells. However, the preliminary results, which could not distinguish a phenotype in these mice, and the therapeutic effect of the rescue diet in 1α(OH)ase KO mice point once again toward the intestine as major site of action of 1,25(OH)2D. Moreover, one can postulate that the defects in the duodenal expression pattern of 1α(OH)ase KO mice should be analogous to the findings in VDR KO mice. Unfortunately, comparative data concerning mRNA expression levels of the calcium transporter genes in 1α(OH)ase WT and KO mice are currently limited to assessment of calbindin-D9K: similar to the Northern analysis of duodenum extracts from the Tokyo and Boston VDR-null mice, RNA levels are clearly downregulated in both models of 1α(OH)ase-null mice [5,6]. Strangely enough, high dietary calcium intake significantly increases the expression of ECaC2 (10-fold) and calbindin-D9K (25-fold) [15]. The mechanism that underlies this vitamin D–independent calcium-regulated pathway is still unknown. Calcium response elements

and/or transcription factors could be involved in the calcium-mediated regulation of gene expression found in that study [15]. As already mentioned, in Leuven and Tokyo VDR WT and KO mice, the high-calcium diet reduces the expression levels of the calcium transporter genes [9,33], which is in line with the hypothesis that active calcium absorption is shut down under these conditions. The data regarding the Boston [35] and Munchen [36] VDR KO are equivocal: the respective authors simply stipulate that calbindin-D9K and ECaC2 levels, respectively, “normalize” compared to WT levels, but it is not clear if WT levels per sé are altered on the rescue diet. Supplementation of 1α-OHase KO mice with 1,25(OH)2D results in a tremendous increase (more pronounced than on the rescue diet) in duodenal mRNA levels of ECaC2 (50-fold), calbindin-D9K (165-fold), and PMCA1b (3.4-fold). ECaC1 mRNA only reaches detectable levels after 1,25(OH)2D administration [15]. These findings support the idea that 1,25(OH)2D is a significant regulator of both epithelial calcium channels in duodenum and that the 1,25(OH)2D–VDR axis stimulates active intestinal calcium absorption by increasing the rate of calcium influx across the intestinal brush border membrane (Fig. 1). At present, comparative data on duodenal expression patterns between WT, 1α(OH)ase KO, VDR KO, and compound VDR1α(OH)ase KO mice, as well as functional analysis of calcium absorption in these mice, are eagerly anticipated.

C. Disturbed Phosphate Homeostasis 1. GENERATION AND CHARACTERISTICS OF NPT2-NULL MICE AND HYP MICE

(See Chapter 26.) Disruption of the major renal sodium-phosphate (Na/Pi) cotransporter (Npt2) gene by targeted ablation of exons 6 to 10 results in increased urinary phosphate excretion, hypophosphatemia, and a unique age-related skeletal phenotype [66]. In addition, Npt2-null mice exhibit increased renal 1α(OH)ase activity, a threefold increase in serum concentrations of 1,25(OH)2D (consistent with the hypothesis that hypophosphatemia is an important signal for renal production of 1,25(OH)2D), hypercalcemia (WT mice 2.27 ± 0.07 mM, vs. null mice 2.99 ± 0.11 mM), and calciuria. Fasting normalizes serum and urinary calcium [37]. Mice with X-linked hypophosphatemia (Hyp) harbor a large deletion in the PHEX “Phosphate regulating gene with homology to Endopeptidases on the X chromosome” gene. They suffer from renal phosphate wasting (due to a 50% reduction in renal Npt2 expression)

440 and hypophosphatemia. However, intestinal calcium absorption in Hyp mice is impaired secondary to inappropriate renal production of 1,25(OH)2D [67], resulting in mild hypocalcemia (normal mice 2.35 ± 0.1 mM, vs Hyp mice 2.03 ± 0.01 mM) [37]. Whereas primary defects of the 1,25(OH)2D–VDR axis intervene in VDR and 1α(OH)ase KO mice, a perturbed phosphate homeostasis triggers the modifications in calcium homeostasis in Npt2-null and Hyp mice. Nonetheless, functional and molecular analysis of active calcium absorption in the latter mouse models are interesting to corroborate the hypothesis based on the VDR and 1α(OH)ase KO mice. 2. CALCIUM ABSORPTION IN NPT2-NULL AND HYP MICE

The absorption of 45Ca by isolated duodenal loops and the appearance of 45Ca in plasma 2 and 4 min after its administration are significantly (about 1.5-fold of WT levels) higher in Npt2-null mice and significantly (less than 35% of levels in normal mice) lower in Hyp mice. This is accompanied by a doubling of ECaC2, ECaC1, and calbindin-D9K mRNA abundance in Npt2-null mice, as assessed by ribonuclease protection assay [37]. On the other hand, the relative abundance of both epithelial calcium channel transcripts is significantly lower in Hyp mice (less than 55%) than in normal littermates [37]. Another study described already a reduction in immunoreactive calbindin-D9K abundance in these mice [68]. These findings clearly indicate that altered intestinal calcium absorption is responsible, at least in part, for the variations in calcium homeostasis in the Npt2-null and Hyp mice. In addition, they provide further support for the hypothesis that ECaC2, in combination with calbindin-D9K, mediates alterations in active intestinal calcium absorption (Fig. 1).

IV. ACTIVE CALCIUM ABSORPTION DURING REPRODUCTION A. Introduction Though 1,25(OH)2D is regarded as the major regulator of active duodenal calcium absorption, other steroid hormones, such as estrogens, or other reproductionrelated hormones may influence the process also. The mineralization of the fetal skeleton and rapid skeletal growth of the newborn require hormone-mediated adjustments in maternal calcium metabolism. This maternal adaptation, during pregnancy [69–72] as well as during lactation [73,74], presumably includes a variable combination of renal calcium conservation, increased calcium resorption from the skeleton, and enhanced

ROGER BOUILLON,

ET AL .

intestinal calcium absorption. Moreover, reduced intestinal calcium absorption is considered to contribute to the pathogenesis of postmenopausal osteoporosis [75–79]. Calcium absorption may be restored or stimulated during estrogen therapy. Most animal and human studies have been hampered by the fact that physiological or therapeutic changes in estrogen status may be accompanied by changes in total and/or free 1,25(OH)2D levels [69,79–82]. Hypothetically, alterations in intestinal calcium absorption could be related to variations in 1,25(OH)2D [70,76,83], to changes in intestinal responsiveness to 1,25(OH)2D [77,84], or to direct action of other hormones on the intestine [73,78,79]. Both estrogen receptor (ER) alpha and beta are present in the intestine [85], as well as the prolactin [86] and the PTH/PTH-related protein (PTH/PTHrP) receptor [87]. The Leuven VDR KO mouse [9] is a unique animal model because it combines a relatively normal fertility with the absence of 1,25(OH)2D genomic actions. Therefore, this animal model can contribute important new data to our understanding of the physiological role of estrogens in the regulation of intestinal calcium transport and by extension the molecular mechanisms underlying regulation of calcium transport in reproduction [88].

B. Gestation and Intestinal Calcium Absorption in VDR WT and KO Mice By the end of gestation (day 18.5), VDR WT and KO mice receiving a normal calcium diet display a slight but significant increase in serum calcium concentration, compared to the appropriate nonpregnant controls (Table III) [32,89]. The concept that cancellous bone is accumulated during gestation as a store for calcium release during subsequent lactation has been recognized for many years [69]. Acquisition of trabecular bone mass was substantiated not only in VDR WT mice, but also in the KO animals (Table III) [32,89]. In duodenum, a significant two-fold stimulation of PMCA1b, calbindin-D9K mRNA, and calbindin-D9K protein expression is seen only in pregnant WT mice (Fig. 4A). This can be assigned to a 2.9-fold increase in circulating 1,25(OH)2D levels [32], an effect that has also been described in rats [90]. In contrast, an impressive (12-fold) and consistent up-regulation of duodenal ECaC2 expression is noted in both pregnant VDR WT and KO animals [32] (Fig. 4A). Consequently ECaC2 expression in pregnant VDR KO mice reaches the level of WT control mice. Additionally, in WT mice, no significant influence of pregnancy on VDR mRNA levels is uncovered (Fig. 4A) [32]. Pregnancy in rats is associated with a doubling of intestinal absorption

441

CHAPTER 25 Calcium Absorption and KO Models

TABLE III

Reproduction and Calcium Homeostasis in VDR WT and KO Mice VDR WT mice

Control Pregnant Control Lactating Sham Ovx Estrogen

VDR KO mice

Serum Ca2+ (mg/dl)

Femur Ca2+ (mg/g dw)

Trabecular density (mg/cm3)

8.9 ± 0.1 10.0 ± 0.3a 9.9 ± 0.3 10.5 ± 0.2 8.7 ± 0.1 8.6 ± 0.1 8.9 ± 0.1

nd nd 20.7 ± 0.5 19.6 ± 0.2 20.3 ± 0.3 20.7 ± 0.3 22.3 ± 0.7c

133 ± 19 204 ± 15a 122 ± 18 86 ± 10 92 ± 11 52 ± 7c 792 ± 29c

Serum Ca2+ (mg/dl)

Femur Ca2+ (mg/g dw)

Trabecular density (mg/cm3)

8.1 ± 0.3a 9.1 ± 0.2b 9.3 ± 0.2 5.3 ± 0.4a,b 7.3 ± 0.4c 7.2 ± 0.3c 7.3 ± 0.4c

nd nd 19.4 ± 0.6 16.7 ± 0.7a,b 17.7 ± 0.6c 18.1 ± 0.6c 21.3 ± 0.6d

138 ± 16 188 ± 25a 130 ± 19 63 ± 13a,b 91 ± 18 79 ± 15 579 ± 61c,d

Dry weight (dw). aP < 0.05 vs WT control. bP < 0.05 vs. KO control. cP < 0.05 vs. WT sham. dP < 0.05 vs. KO sham.

500

600

P

C

P

*

5

*

3

300

§

150 0

C

L

300 200 100

* *

*§

C

L

0

100

* 50

1

*

*

0

0

S O E S O E ECaC2

ERKOα

C P * 0

50 0

0.4 0.2

0

0.0 WT KO

WT KO

50

100

C

P

150

300

100

200

50

100

0

C L

C L

*

*

0

0

C

L

100 75 50 25 0

S O E

50

* 100

25 50 0

0

S O E

Vitamin D status

* 40

200

0

C P

150 100

100

mRNA (% WT control)

0

S O E

0.6

80

1,25(OH)2D3(pg/ml)

mRNA (% WT control)

50

Calb-D9K PMCA1b

120

WT KO

100

300

150

2

150

S O E

mRNA (% WT)

2 1

*

200

400

0

4

450

C L

**

C

150

*

mRNA (% WT control)

*

0

0

*

mRNA (% WT sham)

§

150

C L

1

VDR WT 150

1,25(OH)2D3(pg/ml)

300

2

VDR KO

1,25(OH)2D3(pg/ml)

*

3

100

C P

1350

0

200

Vitamin D status

200

50 1,25(OH)2D3(pg/ml)

1500

C P

4

mRNA (% WT control)

0

*

5

300

mRNA (% WT sham)

§

150

* *

VDR WT Calb/Tot prot (%)

300

VDR KO

Calb/Tot prot (%)

mRNA (% WT control)

1200

mRNA (% WT control)

mRNA (% WT control) mRNA (% WT control)

1350

S O E

B

VDR WT

mRNA (% WT sham)

Ovx ± E2

mRNA (% WT sham)

Lactation

VDR KO

*

Calb/Tot prot (%)

VDR WT Gestation

PMCA1b

Calbindin-D9K

120 80

25 40

mRNA (% WT)

ECaC2

Calb/Tot prot (%)

A

0

0 WT

KO

FIGURE 4 Effect of reproduction on active intestinal calcium absorption: gene expression patterns. (A) Gestation, lactation, ovariectomy, and estrogen therapy in VDR WT and KO mice: duodenal gene expression pattern and 1,25(OH)2D-VDR status. (B) ERα WT and ERKOα mice: duodenal gene expression pattern and 1,25(OH)2D-VDR status. Calbindin-D9K expression: mRNA levels: solid bars; protein levels: horizontal lines. 1,25(OH)2D levels: rising right pattern. VDR mRNA levels: diagonal cross-hatch pattern. Control (C); Pregnant (P); Lactating (L); Sham (S); Ovariectomy (Ovx / O); Estrogen therapy (E). *P<0.05 vs. WT control or vs WT sham, § P < 0.05 vs. KO control or vs. KO sham.

442

ROGER BOUILLON,

+

−

−

ET AL .

+ 1,25(OH)2D

Ca2+

VDR ?

E2 Prolactin PTH-rP/PTH ATP

Placental lactogen

?

CaBP9K ECaC2

ADP

PMCA1b

FIGURE 5 Active intestinal calcium absorption and gestation. Gene expression pattern.

of calcium, even in the absence of vitamin D [91,92]. Accordingly, an increased ECaC2 expression in duodenum of VDR KO mothers may be considered part of the mechanisms to accommodate the calcium demands of late gestation, in view of the fact that calcemia and trabecular bone density are increased to the same extent as in pregnant WT littermates (Fig. 5). This model implies that hormonal adjustments of pregnancy may have a physiological role in promoting calcium absorption by direct action on the intestine, besides an indirect 1,25(OH)2D-dependent effect (Fig. 5) [69,70]. Possible candidate hormones involved in pregnancy-related adaptation of ECaC2 mRNA expression are, for instance, high circulating estrogens, PTHrP [69], and placental lactogen [93,94], as well as prolactin [69,95,96] (Fig. 5). On the other hand, prolactin injections do not increase calcium absorption in late pregnancy in rats, nor does suppression of prolactin secretion by bromocriptine reduce the high calcium transport ratios immediately before parturition [94]. Further studies are needed to evaluate the relative contribution of these pregnancy hormones and the functional repercussions of this changed gene expression on intestinal calcium absorption [88].

C. Lactation and Intestinal Calcium Absorption in VDR WT and KO Mice In order to minimize the stress of gestation-related calcium demand for VDR KO mothers, all mice were studied consuming a high-calcium diet during gestation.

Once the pups were delivered, the diet was switched to a normal calcium diet till analysis at day 19 of lactation. Pups of VDR KO mothers exhibit normal survival and growth during 19 days of lactation. During lactation maternal calcium requirements are high in the setting of a fall in estrogen levels. A PTHrP mediated temporary demineralization of the skeleton is postulated as the main compensatory mechanism to fulfill these needs [69]. In this mouse model bone loss is undoubtedly observed in lactating VDR WT mice, but especially in VDR KO mice (Table III). Moreover, calcemia and calcium content in femur are not affected in lactating WT mice, whereas lactating VDR KO mice suffer from severe hypocalcemia and low calcium content in femur, compared to their control counterparts (Table III). Nevertheless, molecular and functional analysis indicates that active intestinal calcium absorption is additionally enhanced in a VDR-independent manner. Duodenal ECaC2 expression is manifestly raised (13-fold) in lactating WT and KO mice compared to nonlactating littermates (Fig. 4A). Hence, as in pregnant VDR KO animals, ECaC2 expression levels of lactating VDR KO mice become similar to those of control WT mice. Calbindin-D9K RNA and protein levels are increased five and two times, respectively, in lactating WT animals (Fig. 4A). In lactating VDR KO mice, calbindin-D9K protein levels double (Fig. 4A). PMCA1b mRNA levels increase about 1.6-fold in lactating WT and KO mice [32]. This VDR-independent effect was supported by unaltered 1,25(OH)2D-levels in VDR WT animals (Fig. 4.A) [32], in contrast to

443

CHAPTER 25 Calcium Absorption and KO Models

+

−

−

+ 1,25(OH)2D

Ca2+

VDR

?

Prolactin PTH-rP/PTH Placental lactogen

? ATP CaBP9K ADP

ECaC2

PMCA1b

FIGURE 6 Active intestinal calcium absorption and lactation. Gene expression pattern.

∆µMol serum calcium

previous reports of elevated 1,25(OH)2D-levels in lactating rodents [73,97]. Likewise, a study in lactating rats reported that vitamin D deprivation does not appreciably affect enhanced calcium transport in duodenal sacs, but that vitamin D was required to prevent detrimental hypocalcemia [73]. See Fig. 6. Accordingly, the in vivo calcium absorption technique [9] validated the assumption that these changes in gene expression patterns reflect enhanced calcium absorption in this stage of lactation (unpublished personal data). Figure 7 illustrates the increase of calcium appearance in serum of lactating mothers, resulting in a 3.7-fold increase of area under the curve in lactating WT mice (AUCpo0–10 min lactating WT: 18.4 ± 1.7 ∆µmol min vs control WT: 4.9 ± 0.9 ∆µmol min) and a 2.2-fold increase in lactating KO mice (AUCpo0–10min lactating KO: 7.7 ± 1.4 ∆µmol min vs. control KO: 3.6 ± 0.3 ∆µmol min). 2.5

*

*

*

*

*

2.0 1.5

§

§

4 6 8 Time (minutes)

10

§

1.0

§

§

0.5

WT lactating KO lactating WT control KO control

0.0 0

2

FIGURE 7 Active intestinal calcium absorption and lactation. Functional aspects: in vivo calcium absorption assay. *P < 0.05 vs. WT control, §P < 0.05 vs. KO control.

Since estrogens are low during lactation [70], further studies are needed to unravel the observed vitamin D– independent stimulatory mechanism on calcium absorption (Fig. 6). Parathyroidectomy does not influence active calcium absorption during lactation [73], and in spite of elevated PTH levels, active duodenal calcium absorption is impaired in VDR KO mice [9]. Apart from prolactin [69,96], PTHrP levels remain high during lactation, and both are therefore candidate hormones [69,98]. When PTHrP is selectively knocked out from the lactating mammary gland, circulating PTHrP levels are decreased, but not absent, and bone mass is preserved [99]. Remarkably, plasma calcium levels are stable in these dams despite normal milk calcium concentrations, normal PTH levels, reduced 1,25(OH)2D and PTHrP levels. Apparently, dietary calcium serves as a more important source of the total calcium used for milk production than does skeletal calcium in mice. However, the duodenal phenotype of these lactating mice has not been described yet, and the exact stimulatory mechanism on calcium absorption during lactation still remains elusive [99]. Finally, ileal active calcium absorption in late lactation has previously been described in rats [94]; however, the presence of such an additional calcium absorption mechanism has not yet been verified in lactating VDR KO mice. In contrast to gestation, lactation is a setting in which not only the mRNA abundance of ECaC2 but also the expression of the two other calcium transporter proteins is positively influenced in VDR KO mice. In view of the prolonged high calcium needs during lactation, it is not totally unexpected that evolutionary

444

ROGER BOUILLON,

mechanisms have created more than one mechanism to stimulate calcium absorption (Fig. 5).

D. Direct Genomic Effects of Estrogens on Intestinal Calcium Absorption 1. ESTROGEN TREATMENT AND INTESTINAL CALCIUM ABSORPTION

In order to expand the notion of VDR-independent, reproduction-related stimulation of calcium absorption it is important to elaborate on estrogen effects on the intestine. Implantation of 17β-estradiol pellets during four weeks (following ovariectomy) results in a doubling of the uterine wet weight in VDR WT and KO animals, indicating that the administered estrogen dose was supraphysiological and that both genotypes have the same responsiveness to estrogens [32]. In duodenum a significant increase in ECaC2 mRNA expression is found, not only in VDR WT females (fourfold), but also in VDR KO females (eightfold). In contrast, neither expression of calbindin-D9K mRNA and protein, nor of PMCA1b mRNA varies significantly in estrogen-treated animals (Figs. 4A and 8). Last, duodenal VDR mRNA expression is not modified by estrogen status in VDR WT animals (Fig. 4A). Although the suggested increased dietary calcium absorption has not yet been confirmed by functional data, the manifest increase in bone mass (six- to eightfold increase in trabecular bone density), which is paralleled by markedly enhanced calcium content in femur

+

(10% more Ca in WT mice, and up to 20% in KO mice) provides indirect evidence for it (Table III). Although estrogen administration is known to affect the skeleton directly [100], the source for this increase in bone calcium is most likely dietary. Reduced renal calcium loss is less likely as the urinary calcium/creatinine ratio is not changed in VDR WT mice [32] or is even increased in VDR KO mice [32,34]. Furthermore, this VDR-independent induction of ECaC2 mRNA expression agrees with the observation that a pharmacological dose of estradiol-benzoate significantly increased duodenal calcium absorption during in vivo duodenal single pass perfusion in rats, an effect that was blocked by the ER antagonist ICI 182780 [101]. In addition, rat duodenal enterocytes in vitro respond directly to 17β-estradiol with enhanced calcium uptake that is suppressed by transcription and protein synthesis inhibitors [102]. Moreover, Weber et al. described an estrogen response element in the promoter sequence of the mouse ECaC2 gene [36]. Additionally, estradiolmediated induction of intestinal ECaC2 mRNA expression has also been reported in ovariectomized rats (fourfold) and in 1α(OH)ase KO mice (ninefold) [15]. Consequently, the physiologically high concentrations of estrogens during pregnancy may definitely play a role in VDR independent induction of ECaC2 expression. Finally, estrogen replacement therapy has a long history of use in the therapy of postmenopausal bone loss. Thus revisiting the physiology of estrogen action on calcium transport may trigger the development of new therapies targeted not only at the bone, but also at the intestine (Fig. 8) [88].

−

−

+ 1,25(OH)2D

Ca2+ ERβ

VDR

ERα

E2 ATP

CaBP9K ECaC2

ET AL .

ADP

PMCA1b

FIGURE 8 Active intestinal calcium absorption and estrogen treatment. Gene expression pattern.

445

CHAPTER 25 Calcium Absorption and KO Models

2. ESTROGEN DEFICIENCY IN VDR WT AND KO MICE

As pharmacological doses of estrogen interfere with intestinal calcium absorption, the effect of estrogen deficiency could not be overlooked either. Several human studies report a decreased efficiency of global calcium absorption in normal postmenopausal women [75,103], which is more pronounced in postmenopausal osteoporosis [76,104]. Data on active calcium absorption in humans are lacking so far. In VDR WT mice, ovariectomy causes bone loss as evidenced by the 43% reduction in trabecular bone density, compared to sham-operated littermates (Table III), while loss of ovarian function does not significantly alter this parameter in VDR KO mice. Total and free 1,25(OH)2D levels double in ovariectomized VDR WT mice, which may represent a compensatory mechanism for the lack of endogenous estrogens (Table III). Duodenal expression pattern of the three known calcium transporter genes is not altered by ovariectomy in VDR WT females, or in VDR KO females, though (Fig. 4A). These findings agree with a normal active calcium absorption in ovariectomized rats, as revealed previously by in vivo single-pass perfusion [101]. However, the observations in these rodent studies should be interpreted with caution in light of the limitations inherent to the design of the study: the validity of extrapolation toward older animals and human postmenopausal osteoporosis has yet to be proven. 3. ESTROGEN RECEPTOR ALPHA (ERKOα) AND BETA (ERKOβ) HULL MICE

Following ovariectomy, endogenous estrogen production, although significantly reduced, is not absent, as adrenal androgens can be aromatized to estrogens. By contrast, in ERKOα and ERKOβ mice, genomic actions of estrogens mediated by their respective ERα and ERβ, are completely abolished. ERKOα mice were obtained by targeted ablation of exon 2 [105] and ERKOβ mice via homologous recombination replacing exon 2 and part of exon 3 (encoding the first Zn finger) by the neo selection gene [32]. ERα mRNA expression is much more abundant in murine duodenum than ERβ (unpublished personal data), a pattern opposite to human and rat intestine [106]. This finding probably explains the unchanged duodenal gene expression pattern in ERKOβ mice [32]. In contrast, ERKOα mice displayed a significant reduction of duodenal ECaC2 mRNA expression to 45% of their WT littermates, while calbindin-D9K and PMCA1b levels were unaffected (Fig. 4B). Furthermore, serum calcium and 1,25(OH)2D3 levels, as well as duodenal VDR expression, were not different from those in WT littermates (Fig. 4B), pleading for direct genomic actions of estrogens on mouse duodenal ECaC2 mRNA expression, mediated by the ERα (Fig. 8).

V. ACTIVE CALCIUM ABSORPTION AND CORTICOSTEROIDS A. Corticosteroids and Intestinal Calcium Absorption Since besides 1,25(OH)2D, estrogens also have distinct VDR-independent effects on intestinal calcium absorption, it is appealing to explore the effects of a third group of steroid hormones on intestinal calcium absorption: corticosteroids. “Malabsorption of calcium” is considered to be a fairly consistent finding in steroid-treated patients [107–109] and animals [109–113]. It is demonstrable within the first 2 weeks of steroid treatment, when concentrations of vitamin D metabolites are either normal or increased. It is generally believed that this calcium malabsorption is due to reduced active calcium absorption. However, a molecular substrate for this malabsorption has never been irrefutably identified. Moreover, it is still unclear whether corticosteroids act through the 1,25(OH)2DVDR system, or whether they exert direct effects on the intestine via their own nuclear receptor.

B. Dexamethasone and Intestinal Calcium Absorption in WT Mice One previous study in vitamin D–deficient Swiss mice showed an important reduction in calbindin-D9K mRNA expression (−86%, compared with vehicletreated mice) after treatment with 2 mg/kg dexamethasone for 5 days [114]. This test was repeated with vitamin D–replete WT mice: no difference could be detected between dexamethasone or vehicle-treated mice at the molecular or at the functional level, not even when the mice were fed a 0.02% calcium diet for 1 week prior to sacrifice (Fig. 9). Because of this lack of an evident effect in WT animals using available techniques, evaluation in VDR KO mice was not further explored (unpublished personal data).

VI. SUMMARY Duodenal gene expression of the epithelial calcium channels clearly react more dynamically to changes in calcium intake [9], vitamin D status [9,33,37], estrogen administration, gestation, or lactation [32] than any of the other calcium transporter genes (Table IV). This is paralleled by similar functional changes in active intestinal calcium absorption, when assessed in VDR WT and KO animals, under varying dietary calcium intake [9,33] and at the end of lactation. Hence it seems that these intestinal calcium influx mechanisms

Lessons from Intervention Studies in VDR WT and KO Mice

VDR KO

6 hr

24 hr

Low calcium diet

↑↑ECaC2 ↑Calbindin-D9K ↑PMCA1b

↑↑ ECaC2 ↑↑ Calbindin-D9K ↑ PMCA1b

↑↑ ECaC2 ↑ Calbindin-D9K

no effect

no effect

↑ ↑ ↑ ↑

VDR WT

Dietary intervention Rescue diet ECaC2

Calbindin-D9K

(PMCA1b ECaC2 Calbindin-D9K PMCA1b

↑ ↑ ↑↑

1,25(OH)2D3

↑ ↑↑

TABLE IV

ECaC2

Reproduction Gestation

Lactation

E2 treatment

↑↑↑ ECaC2 ↑ Calbindin-D9K ↑ PMCA1b ↑↑↑ ECaC2

↑↑↑ ECaC2 ↑ Calbindin-D9K ↑ PMCA1b ↑↑↑ ECaC2 ↑ Calbindin-D9K ↑ PMCA1b

↑ ECaC2

Calbindin-D9K

PMCA1b

↑ ECaC2

VDR dependent

VDR independent and dependent

VDR in- and dependent

VDR independent

VDR independent

1,25(OH)2D3

1,25(OH)2D3 + ?

Hormone(s) ?

Hormone(s) ?

E2-ERα dependent

Importance of regulation is related to font size. VDR-dependent effects are underlined.

Factor(s) ?

447

CHAPTER 25 Calcium Absorption and KO Models

Calb-D9K PMCA1b

80

1.0

40

0.5

0

0.0 V D

V

D

B

1.5

∆µMol serum calcium

mRNA (% Vehicle)

ECaC2 120

Calb/Tot prot (%)

A

2.5 2.0 1.5 1.0

WT Vehicle WT Dexa

0.5 0.0 0

V D

2

4

6

8

10

Time (minutes)

FIGURE 9 Active intestinal calcium absorption and dexamethasone treatment. (A) Molecular aspects. (B) Functional aspects: in vivo calcium absorption assay. Vehicle (V); Dexamethasone (D).

are the feedback-controlled part of the actually known steps in the transcellular calcium transport system. At first sight, calcium entry was unlikely to be the critical or rate-limiting factor of active intestinal calcium absorption, because of the favorable electrochemical and concentration gradient between gastrointestinal lumen and intracellular milieu. Since the epithelial calcium channels are so tightly regulated by vitamin D, reproduction and estrogens, calcium entry appears to be the rate-limiting step. Such a natural selection mechanism is in fact obvious as it avoids the massive entry of calcium when subsequent intracellular calcium buffering and shuttling mechanisms would fail. In such circumstances high intracellular calcium concentrations would soon become lethal for the enterocyte. On the other hand, actual uptake of calcium from the gut into the body definitely requires intracellular calcium buffering and shuttling toward the basolateral membrane with subsequent extrusion from the enterocyte (Fig. 1, Table IV). Thus, it is not surprising and even crucial that calbindin-D9K and PMCA1b expression are also induced when active calcium absorption is activated. This, however, was observed to be at a lesser extent than ECaC2 expression. In addition, analysis of female VDR KO mice has provided irrefutable evidence for VDR-independent stimulation of the expression of calcium transporter genes by estrogens and lactation [32], whereas during gestation a combination of VDR-dependent and -independent mechanisms intervene [32] (Table IV). Presumably, active intestinal calcium absorption is a more complex process than the actually known three major steps. Apart from the S100A10-annexin 2 complex, regulatory elements/genes are presently enigmatic. Further unraveling of the calcium absorption process promises to be a fascinating challenge for further research. Moreover, future studies using ECaC1, ECaC2, and calbindin-D9K KO mice to examine intestinal calcium absorption will be indispensable to

provide in vivo evidence of the contribution of each of these proteins to the process of intestinal calcium transport [88].

References 1. Reichel H, Koeffler HP, Norman AW 1989 The role of the vitamin D endocrine system in health and disease. N Engl J Med 320:980–991. 2. Haussler MR, Whitfield GK, Haussler CA, Hsieh JC, Thompson PD, Selznick SH, Dominguez CE, Jurutka PW 1998 The nuclear vitamin D receptor: biological and molecular regulatory properties revealed. J Bone Miner Res 13:325–349. 3. Kato S, Yanagisawa J, Murayama A, Kitanaka S, Takeyama K 1998 The importance of 25-hydroxyvitamin D3 1α-hydroxylase gene in vitamin D–dependent rickets. Curr Opin Nephrol Hypertens 7:377–383. 4. Malloy PJ, Pike JW, Feldman D 1999 The vitamin D receptor and the syndrome of hereditary 1,25-dihydroxyvitamin D–resistant rickets. Endocr Rev 20:156–188. 5. Dardenne O, Prud’homme J, Arabian A, Glorieux FH, St Arnaud R 2001 Targeted inactivation of the 25-hydroxyvitamin D3-1α-hydroxylase gene (CYP27B1) creates an animal model of pseudovitamin D–deficiency rickets. Endocrinology 142:3135–3141. 6. Panda DK, Miao D, Tremblay ML, Sirois J, Farookhi R, Hendy GN, Goltzman D 2001 Targeted ablation of the 25-hydroxyvitamin D 1α-hydroxylase enzyme: evidence for skeletal, reproductive, and immune dysfunction. Proc Natl Acad Sci USA 98:7498–7503. 7. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Genet 16:391–396. 8. Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay, MB 1997 Targeted ablation of the vitamin D receptor: an animal model of vitamin D–dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94:9831–9835. 9. Van Cromphaut SJ, Dewerchin M, Hoenderop JG, Stockmans I, Van Herck E, Kato S, Bindels RJ, Collen D, Carmeliet P, Bouillon R, Carmeliet G 2001 Duodenal

448

10.

11. 12.

13.

14.

15.

16.

17.

18. 19.

20.

21. 22.

23.

24.

ROGER BOUILLON,

calcium absorption in vitamin D receptor-knockout mice: functional and molecular aspects. Proc Natl Acad Sci USA 98:13324–13329. Balsan S, Garabedian M, Larchet M, Gorski AM, Cournot G, Tau C, Bourdeau A, Silve C, Ricour C 1986 Long-term nocturnal calcium infusions can cure rickets and promote normal mineralization in hereditary resistance to 1,25-dihydroxyvitamin D. J Clin Invest 77:1661–1667. Hochberg Z, Tiosano D, Even L 1992 Calcium therapy for calcitriol-resistant rickets. J Pediatr 121:803–808. Li YC, Amling M, Pirro AE, Priemel M, Meuse J, Baron R, Delling G, Demay MB 1998 Normalization of mineral ion homeostasis by dietary means prevents hyperparathyroidism, rickets, and osteomalacia, but not alopecia in vitamin D receptor-ablated mice. Endocrinology 139:4391–4396. Amling M, Priemel M, Holzmann T, Chapin K, Rueger JM, Baron R, Demay MB 1999. Rescue of the skeletal phenotype of vitamin D receptor-ablated mice in the setting of normal mineral ion homeostasis: formal histomorphometric and biomechanical analyses. Endocrinology 140:4982–4987. Dardenne O, Prud’homme J, Hacking SA, Glorieux FH, St Arnaud R 2003 Correction of the abnormal mineral ion homeostasis with a high-calcium, high-phosphorus, highlactose diet rescues the PDDR phenotype of mice deficient for the 25-hydroxyvitamin D-1α-hydroxylase (CYP27B1). Bone 32:332–340. Van Abel M, Hoenderop JG, van der Kemp AW, van Leeuwen JP, Bindels RJ 2003 Regulation of the epithelial Ca2+ channels in small intestine as studied by quantitative mRNA detection. Am J Physiol Gastrointest Liver Physiol 285:G78–G85. Wasserman RH 1997 Vitamin D and the intestinal absorption of calcium and phosphorus. In: Feldman D, Glorieux FH, and Pike JW (eds) Vitamin D. Academic Press, San Diego, pp. 259–273. Hoenderop JG, van der Kemp AW, Hartog A, van de Graaf SF, van Os CH, Willems PH, Bindels RJ 1999 Molecular identification of the apical Ca2+ channel in 1:25-dihydroxyvitamin D3responsive epithelia. J Biol Chem 274:8375–8378. Hoenderop JG, Hartog A, Stuiver M, Doucet A, Willems PH, Bindels RJ 2000 Localization of the epithelial Ca2+ channel in rabbit kidney and intestine. J Am Soc Nephrol 11:1171–1178. Muller D, Hoenderop JG, Meij IC, van den Heuvel LP, Knoers NV, den Hollander AI, Eggert P, Garcia-Nieto V, Claverie-Martin F, Bindels RJ 2000 Molecular cloning, tissue distribution, and chromosomal mapping of the human epithelial Ca2+ channel (ECAC1). Genomics 67:48–53. Hoenderop JG, Muller D, Kemp AW, Hartog A, Suzuki M, Ishibashi K, Imai M, Sweep F, Willems PH, Os CH, Bindels RJ 2001 Calcitriol controls the epithelial calcium channel in kidney. J Am Soc Nephrol 12:1342–1349. Hoenderop JG, Nilius B, Bindels RJ 2003 Epithelial calcium channels: from identification to function and regulation. Pflügers Arch 446:304–308. Muller D, Hoenderop JG, Merkx GF, van Os CH, Bindels RJ 2000 Gene structure and chromosomal mapping of human epithelial calcium channel. Biochem Biophys Res Commun 275:47–52. Peng JB, Chen XZ, Berger UV, Vassilev PM, Tsukaguchi H, Brown EM, Hediger MA 1999 Molecular cloning and characterization of a channel-like transporter mediating intestinal calcium absorption. J Biol Chem 274:22739–22746. Peng JB, Brown EM, Hediger MA 2001 Structural conservation of the genes encoding CaT1:CaT2:and related cation channels. Genomics 76:99–109.

ET AL .

25. Peng JB, Hediger MA 2002 A family of calcium-permeable channels in the kidney: distinct roles in renal calcium handling. Curr Opin Nephrol Hypertens 11:555–561. 26. Thomasset M 1997 Calbindin-D9K. In: Feldman D, Glorieux FH, and Pike JW (ed) Vitamin D. Academic Press, San Diego, pp. 223–232. 27. Christakos S, Gabrielides C, Rhoten WB 1989 Vitamin D– dependent calcium binding proteins: chemistry, distribution, functional considerations, and molecular biology. Endocr Rev 10:3–26. 28. Dupret JM, Brun P, Perret C, Lomri N, Thomasset M, Cuisinier-Gleizes P 1987 Transcriptional and posttranscriptional regulation of vitamin D–dependent calciumbinding protein gene expression in the rat duodenum by 1,25-dihydroxycholecalciferol. J Biol Chem 262:16553–16557. 29. Cai Q, Chandler JS, Wasserman RH, Kumar R, Penniston JT 1993 Vitamin D and adaptation to dietary calcium and phosphate deficiencies increase intestinal plasma membrane calcium pump gene expression. Proc Natl Acad Sci USA 90:1345–1349. 30. Pannabecker TL, Chandler JS, and Wasserman RH 1995 Vitamin-D-dependent transcriptional regulation of the intestinal plasma membrane calcium pump. Biochem Biophys Res Commun 213:499–505. 31. Wasserman RH, Smith CA, Brindak ME, De Talamoni N, Fullmer CS, Penniston JT, Kumar R 1992 Vitamin D and mineral deficiencies increase the plasma membrane calcium pump of chicken intestine. Gastroenterology 102:886–894. 32. Van Cromphaut SJ, Rummens K, Stockmans I, Van Herck E, Dijcks FA, Ederveen AG, Carmeliet P, Verhaeghe J, Bouillon R, Carmeliet G 2003 Intestinal calcium transporter genes are upregulated by estrogens and the reproductive cycle through vitamin D receptor-independent mechanisms. J Bone Miner Res 18:1725–1736. 33. Song Y, Kato S, Fleet JC 2003 Vitamin D receptor (VDR) knockout mice reveal VDR-independent regulation of intestinal calcium absorption and ECaC2 and calbindin D9k mRNA. J Nutr 133:374–380. 34. Li YC, Bolt MJ, Cao LP, Sitrin MD 2001 Effects of vitamin D receptor inactivation on the expression of calbindins and calcium metabolism. Am J Physiol Endocrinol Metab 281:E558–E564. 35. Li YC, Pirro AE, Demay MB 1998 Analysis of vitamin D–dependent calcium-binding protein messenger ribonucleic acid expression in mice lacking the vitamin D receptor. Endocrinology 139:847–851. 36. Weber K, Erben RG, Rump A, Adamski J 2001 Gene structure and regulation of the murine epithelial calcium channels ECaC1 and 2. Biochem Biophys Res Commun 289:1287–1294. 37. Tenenhouse HS, Gauthier C, Martel J, Hoenderop JG, Hartog A, Meyer MH, Meyer RA Jr, Bindels RJ 2002 Na/Pi cotransporter (Npt2) gene disruption increases duodenal calcium absorption and expression of epithelial calcium channels 1 and 2. Pflügers Arch 444:670–676. 38. Fullmer CS, Chandra S, Smith CA, Morrison GH, Wasserman RH 1996 Ion microscopic imaging of calcium during 1,25-dihydroxyvitamin D–mediated intestinal absorption. Histochem Cell Biol 106:215–222. 39. Chandra S, Fullmer CS, Smith CA, Wasserman RH, Morrison GH 1990 Ion microscopic imaging of calcium transport in the intestinal tissue of vitamin D-deficient and vitamin D-replete chickens: a 44Ca stable isotope study. Proc Natl Acad Sci USA 87:5715–5719. 40. Montell C, Birnbaumer L, Flockerzi V, Bindels RJ, Bruford EA, Caterina MJ, Clapham DE, Harteneck C, Heller S, Julius D,

CHAPTER 25 Calcium Absorption and KO Models

41. 42. 43.

44.

45. 46. 47.

48.

49.

50.

51.

52. 53.

54.

55. 56. 57.

Kojima I, Mori Y, Penner R, Prawitt D, Scharenberg AM, Schultz G, Shimizu N, Zhu MX 2002 A unified nomenclature for the superfamily of TRP cation channels. Mol Cell 9:229–231. Montell C 2001 Physiology, phylogeny, and functions of the TRP superfamily of cation channels. Sci STKE 90:RE1. Hoenderop JG, Nilius B, Bindels RJ 2002 ECaC: the gatekeeper of transepithelial Ca2+ transport. Biochim Biophys Acta 1600:6–11. Loffing J, Loffing-Cueni D, Valderrabano V, Klausli L, Hebert SC, Rossier BC, Hoenderop JG, Bindels RJ, Kaissling B 2001 Distribution of transcellular calcium and sodium transport pathways along mouse distal nephron. Am J Physiol Renal Physiol 281:F1021–F1027. Peng JB, Chen XZ, Berger UV, Weremowicz S, Morton CC, Vassilev PM, Brown EM, Hediger MA 2000 Human calcium transport protein CaT1. Biochem Biophys Res Commun 278:326–332. Voets T, Nilius B 2003 The pore of TRP channels: trivial or neglected? Cell Calcium 33:299–302. Voets T, Janssens A, Prenen J, Droogmans G, Nilius B 2003 Mg2+-dependent gating and strong inward rectification of the cation channel TRPV6. J Gen Physiol 121:245–260. Watanabe H, Vriens J, Janssens A, Wondergem R, Droogmans G, Nilius B 2003 Modulation of TRPV4 gating by intra- and extracellular Ca2+. Cell Calcium 33: 489–495. Nilius B, Vennekens R, Prenen J, Hoenderop JG, Droogmans G, Bindels RJ 2001 The single pore residue Asp542 determines Ca2+ permeation and Mg2+ block of the epithelial Ca2+ channel. J Biol Chem 276:1020–1025. Vennekens R, Hoenderop JG, Prenen J, Stuiver M, Willems PH, Droogmans G, Nilius B, Bindels RJ 2000 Permeation and gating properties of the novel epithelial Ca2+ channel. J Biol Chem 275:3963–3969. Hoenderop JG, Vennekens R, Muller D, Prenen J, Droogmans G, Bindels RJ, Nilius B 2001 Function and expression of the epithelial Ca2+ channel family: comparison of mammalian ECaC1 and 2. J Physiol 537:747–761. Hoenderop JG, Voets T, Hoefs S, Weidema F, Prenen J, Nilius B, Bindels RJ 2003 Homo- and heterotetrameric architecture of the epithelial Ca2+ channels TRPV5 and TRPV6. EMBO J 22:776–785. den Dekker E, Hoenderop JG, Nilius B, Bindels RJ 2003 The epithelial calcium channels, TRPV5 & TRPV6: from identification towards regulation. Cell Calcium 33:497–507. van de Graaf SF, Hoenderop JG, Gkika D, Lamers D, Prenen J, Rescher U, Gerke V, Staub O, Nilius B, Bindels RJ 2003 Functional expression of the epithelial Ca2+ channels (TRPV5 and TRPV6) requires association of the S100A10annexin 2 complex. EMBO J 22:1478–1487. Murayama A, Takeyama K, Kitanaka S, Kodera Y, Hosoya T, Kato S 1998 The promoter of the human 25-hydroxyvitamin D3 1α-hydroxylase gene confers positive and negative responsiveness to PTH, calcitonin, and 1 alpha,25(OH)2D3. Biochem Biophys Res Commun 249:11–16. Omdahl J, May B 1997 The 25-hydroxyvitamin D 24-hydroxylase. In: Vitamin D, 1st Edition Feldman D, Glorieux FH, and Pike JW (eds) Academic Press, San Diego, pp. 69–85. Dostal LA, Toverud SU 1984 Effect of vitamin D3 on duodenal calcium absorption in vivo during early development. Am J Physiol 246:G528–G534. Marcus CS, Lengemann FW 1962 Absorption of Ca45 and Sr85 from solid and liquid food at various levels of the alimentary tract of the rat. J Nutr 77:155–160.

449 58. Bronner F, Yost JH 1985 Saturable and nonsaturable copper and calcium transport in mouse duodenum. Am J Physiol 249:G108–G112. 59. Bishop CW, Kendrick NC, Deluca HF 1983. Induction of calcium-binding protein before 1,25-dihydroxyvitamin D3 stimulation of duodenal calcium uptake. J Biol Chem 258:1305–1310. 60. Wood RJ, Tchack L, Taparia S 2001 1,25-Dihydroxyvitamin D3 increases the expression of the CaT1 epithelial calcium channel in the Caco-2 human intestinal cell line. BMC Physiol 1:11–18. 61. Yue L, Peng JB, Hediger MA, Clapham DE 2001 CaT1 manifests the pore properties of the calcium-release-activated calcium channel. Nature 410:705–709. 62. Nilius B, Prenen J, Vennekens R, Hoenderop JG, Bindels RJ, Droogmans G 2001 Modulation of the epithelial calcium channel, ECaC, by intracellular Ca2+. Cell Calcium 29:417–428. 63. Dardenne O, Prudhomme J, Hacking SA, Glorieux FH, St Arnaud R 2003 Rescue of the pseudo-vitamin D deficiency rickets phenotype of CYP27B1-deficient mice by treatment with 1,25-dihydroxyvitamin D3: biochemical, histomorphometric, and biomechanical analyses. J Bone Miner Res 18:637–643. 64. Panda DK, Miao D, Huo R, Hendy GN, Goltzman D 2003 Comparison of mice with deletion of the 25OH Vitamin D3 1αHydroxylase, of the vitamin D receptor and of both, suggest an in vivo role for VDR-dependent and independent regulation of calcium and skeletal homeostasis. J Bone Miner Res 17 (Suppl 1):1029. 65. St Arnaud R, Dardenne O, Prud’homme J, Hacking SA, Glorieux FH 2003 Conventional and tissue-specific inactivation of the 25-hydroxyvitamin D-1α-hydroxylase (CYP27B1). J Cell Biochem 88:245–251. 66. Beck L, Karaplis AC, Amizuka N, Hewson AS, Ozawa H, Tenenhouse HS 1998. Targeted inactivation of Npt2 in mice leads to severe renal phosphate wasting, hypercalciuria, and skeletal abnormalities. Proc Natl Acad Sci USA 95: 5372–5377. 67. Meyer RA Jr, Meyer MH, Gray RW, Bruns ME 1987 Evidence that low plasma 1,25-dihydroxyvitamin D causes intestinal malabsorption of calcium and phosphate in juvenile X-linked hypophosphatemic mice. J Bone Miner Res 2:67–82. 68. Bruns ME, Meyer RA Jr, Meyer MH 1984 Low levels of intestinal vitamin D-dependent calcium-binding protein in juvenile X-linked hypophosphatemic mice. Endocrinology 115:1459–1463. 69. Kovacs CS, Kronenberg HM 1997 Maternal–fetal calcium and bone metabolism during pregnancy, puerperium, and lactation. Endocr Rev 18:832–872. 70. Ritchie LD, Fung EB, Halloran BP, Turnlund JR, Van Loan MD, Cann CE, King JC 1998 A longitudinal study of calcium homeostasis during human pregnancy and lactation and after resumption of menses. Am J Clin Nutr 67:693–701. 71. Kent GN, Price RI, Gutteridge DH, Rosman KJ, Smith M, Allen JR, Hickling CJ, Blakeman SL 1991 The efficiency of intestinal calcium absorption is increased in late pregnancy but not in established lactation. Calcif Tissue Int 48:293–295. 72. Kent GN, Price RI, Gutteridge DH, Allen JR, Blakeman SL, Bhagat CI, St John A, Barnes MP, Smith M, Evans DV 1991 Acute effects of an oral calcium load in pregnancy and lactation: findings on renal calcium conservation and biochemical indices of bone turnover. Miner Electrolyte Metab 17:1–7.

450 73. Boass A, Toverud SU, Pike JW, Haussler MR 1981 Calcium metabolism during lactation: enhanced intestinal calcium absorption in vitamin D–deprived, hypocalcemic rats. Endocrinology 109:900–907. 74. Boass A, Toverud SU 1997 Enhanced nonsaturable calcium transport in the jejunum of rats during lactation, but not during pregnancy. J Bone Miner Res 12:1577–1583. 75. Heaney RP, Recker RR, Stegman MR, Moy AJ 1989 Calcium absorption in women: relationships to calcium intake, estrogen status, and age. J Bone Miner Res 4:469–475. 76. Gallagher JC, Riggs BL, Eisman J, Hamstra A, Arnaud SB, Deluca HF 1979 Intestinal calcium absorption and serum vitamin D metabolites in normal subjects and osteoporotic patients: effect of age and dietary calcium. J Clin Invest 64:729–736. 77. Gennari C, Agnusdei D, Nardi P, Civitelli R 1990 Estrogen preserves a normal intestinal responsiveness to 1,25-dihydroxyvitamin D3 in oophorectomized women. J Clin Endocrinol Metab 71:1288–1293. 78. O’Loughlin PD, Morris HA 1998 Oestrogen deficiency impairs intestinal calcium absorption in the rat. J Physiol 511:313–322. 79. Colin EM, Van Den Bemd GJ, Van Aken M, Christakos S, De Jonge HR, Deluca HF, Prahl JM, Birkenhager JC, Buurman CJ, Pols HA, van Leeuwen JP 1999 Evidence for involvement of 17β-estradiol in intestinal calcium absorption independent of 1,25-dihydroxyvitamin D3 level in the rat. J Bone Miner Res 14:57–64. 80. Kalu DN, Chen C 1999 Ovariectomized murine model of postmenopausal calcium malabsorption. J Bone Miner Res 14:593–601. 81. Chen C, Noland KA, Kalu DN 1997 Modulation of intestinal vitamin D receptor by ovariectomy, estrogen and growth hormone. Mech Ageing Dev 99:109–122. 82. Bouillon R, Van Assche FA, Van Baelen H, Heyns W, De Moor P 1981 Influence of the vitamin D-binding protein on the serum concentration of 1,25-dihydroxyvitamin D3. Significance of the free 1,25-dihydroxyvitamin D3 concentration. J Clin Invest 67:589–596. 83. Prince RL, Smith M, Dick IM, Price RI, Webb PG, Henderson NK, Harris MM 1991 Prevention of postmenopausal osteoporosis. A comparative study of exercise, calcium supplementation, and hormone-replacement therapy. N Engl J Med 325:1189–1195. 84. Liel Y, Shany S, Smirnoff P, Schwartz B 1999 Estrogen increases 1,25-dihydroxyvitamin D receptors expression and bioresponse in the rat duodenal mucosa. Endocrinology 140:280–285. 85. Campbell-Thompson M 1997 Estrogen Receptor alpha and beta expression in upper gastrointestinal tract with regulation of Trefoil Factor Family 2 mRNA levels in ovariectomized rats. Biochem Biophys Res Commun 240:478–483. 86. Nagano M, Chastre E, Choquet A, Bara J, Gespach C, Kelly PA 1995 Expression of prolactin and growth hormone receptor genes and their isoforms in the gastrointestinal tract. Am J Physiol 268:G431–G442. 87. Li H, Seitz PK, Thomas ML, Selvanayagam P, Rajaraman S, Cooper CW 1995 Widespread expression of the parathyroid hormone-related peptide and PTH/PTHrP receptor genes in intestinal epithelial cells. Lab Invest 73:864–870. 88. Christakos S, Prince R 2003 Estrogen, vitamin D, and calcium transport. J Bone Miner Res 18:1737–1739. 89. Rummens K, Van Cromphaut SJ, Carmeliet G, Van Herck E, van Bree R, Stockmans I, Bouillon R, Verhaeghe J 2003

ROGER BOUILLON,

90.

91. 92. 93. 94. 95.

96. 97. 98.

99.

100.

101.

102. 103. 104. 105.

106.

ET AL .

Pregnancy in mice lacking the vitamin D receptor: normal maternal skeletal response, but fetal hypomineralization rescued by maternal calcium supplementation. Pediatr Res 54:466–473. Zhu Y, Goff JP, Reinhardt TA, Horst RL 1998 Pregnancy and lactation increase vitamin D–dependent intestinal membrane calcium adenosine triphosphatase and calcium binding protein messenger ribonucleic acid expression. Endocrinology 139:3520–3524. Halloran BP, Deluca HF 1980 Calcium transport in small intestine during pregnancy and lactation. Am. J Physiol 239:E64–E68. Brommage R, Baxter DC, Gierke LW 1990 Vitamin D– independent intestinal calcium and phosphorus absorption during reproduction. Am J Physiol 259:G631–G638. Mainoya JR 1975 Effects of bovine growth hormone, human placental lactogen and ovine prolactin on intestinal fluid and ion transport in the rat. Endocrinology 96:1165–1170. Boass A, Lovdal JA, Toverud SU 1992 Pregnancy- and lactation-induced changes in active intestinal calcium transport in rats. Am J Physiol 263:G127–G134. Lotinun S, Limlomwongse L, Krishnamra N 1998 The study of a physiological significance of prolactin in the regulation of calcium metabolism during pregnancy and lactation in rats. Can J Physiol Pharmacol 76:218–228. Pahuja DN, Deluca HF 1981 Stimulation of intestinal calcium transport and bone calcium mobilization by prolactin in vitamin D-deficient rats. Science 214:1038–1039. Pike JW, Parker JB, Haussler MR, Boass A, Toverud SV 1979 Dynamic changes in circulating 1,25-dihydroxyvitamin D during reproduction in rats. Science 204:1427–1429. Lippuner K, Zehnder HJ, Casez JP, Takkinen R, Jaeger P 1996 PTH-related protein is released into the mother’s bloodstream during location: evidence for beneficial effects on maternal calcium-phosphate metabolism. J Bone Miner Res 11:1394–1399. VanHouten JN, Dann P, Stewart AF, Watson CJ, Pollak M, Karaplis AC, Wysolmerski JJ 2003 Mammary-specific deletion of parathyroid hormone-related protein preserves bone mass during lactation. J Clin Invest 112:1429–1436. Edwards MW, Bain SD, Bailey MC, Lantry MM, Howard GA 1992 17β Estradiol stimulation of endosteal bone formation in the ovariectomized mouse: an animal model for the evaluation of bone-targeted estrogens. Bone 13:29–34. Ten Bolscher M, Netelenbos JC, Barto R, Van Buuren LM, Van der vijgh WJ 1999 Estrogen regulation of intestinal calcium absorption in the intact and ovariectomized adult rat. J Bone Miner Res 14:1197–1202. Arjmandi BH, Salih MA, Herbert DC, Sims SH, Kalu DN 1993 Evidence for estrogen receptor–linked calcium transport in the intestine. Bone Miner 21:63–74. Heaney RP, Recker RR, Saville PD 1978 Menopausal changes in calcium balance performance. J Lab Clin Med 92:953–963. Gallagher JC, Aaron J, Horsman A, Marshall DH, Wilkinson R, Nordin BE 1973 The crush fracture syndrome in postmenopausal women. Clin Endocrinol Metab 2:293–315. Korach KS, Couse JF, Curtis SW, Washburn TF, Lindzey J, Kimbro KS, Eddy EM, Migliaccio S, Snedeker SM, Lubahn DB, Schomberg DW, Smith EP 1996 Estrogen receptor gene disruption: Molecular characterization and experimental and clinical phenotypes. Recent Prog Horm Res 51:159–186. Campbell-Thompson M, Lynch IJ, Bhardwaj B 2001. Expression of estrogen receptor (ER) subtypes and ERβ isoforms in colon cancer. Cancer Res 61:632–640.

CHAPTER 25 Calcium Absorption and KO Models

107. Reid IR 1997 Glucocorticoid osteoporosis—mechanisms and management. Eur J Endocrinol 137:209–217. 108. Shrivastava A, Lyon A, Mcintosh N 2000 The effect of dexamethasone on growth, mineral balance and bone mineralisation in preterm infants with chronic lung disease. Eur J Pediatr 159:380–384. 109. Smith BJ, Phillips PJ, Pannall PR, Cain HJ, Leckie WJ 1993 Effect of orally administered beclomethasone dipropionate on calcium absorption from the gut in normal subjects. Thorax 48:890–893. 110. Aloia JF, Semla HM, Yeh JK 1984 Discordant effects of glucocorticoids on active and passive transport of calcium in the rat duodenum. Calcif Tissue Int 36:327–331.

451 111. Glade MJ, Krook L, Schryver HF, Hintz HF 1982 Calcium metabolism in glucocorticoid-treated pony foals. J Nutr 112:77–86. 112. Weiler HA, Wang Z, Atkinson SA 1995 Dexamethasone treatment impairs calcium regulation and reduces bone mineralization in infant pigs. Am J Clin Nutr 61:805–811. 113. Yeh JK, Aloia JF 1986 Influence of glucocorticoids on calcium absorption in different segments of the rat intestine. Calcif Tissue Int 38:282–288. 114. Li H, Christakos S 1991 Differential regulation by 1,25-dihydroxyvitamin D3 of calbindin-D9k and calbindin-D28k gene expression in mouse kidney. Endocrinology 128: 2844–2852.

CHAPTER 26

Phosphate Homeostasis HARRIET S. TENENHOUSE ANTHONY A. PORTALE

I. II. III. IV. V.

Departments of Pediatrics and Human Genetics, McGill University and The Montreal Children’s Hospital Research Institute, Montreal, Quebec Department of Pediatrics, University of California San Francisco, San Francisco, California

Introduction Phosphate Homeostasis Intestinal Phosphate Absorption Renal Phosphate Transport Role of Phosphate in Regulation of Renal Vitamin D Metabolism

I. INTRODUCTION Inorganic phosphate (Pi) is fundamental to cellular metabolism and, in vertebrates, to skeletal mineralization. To accomplish these functions, transport systems have evolved to permit the efficient transfer of negatively charged Pi ions across hydrophobic membrane barriers. Ingested Pi is absorbed by the small intestine, deposited in bone, and filtered by the kidney where it is reabsorbed and excreted in amounts that are determined by the specific requirements of the organism. The kidney is a major determinant of Pi homeostasis by virtue of its ability to increase or decrease its Pi reabsorptive capacity to accommodate Pi need. Accordingly, significant advances have been made in our understanding of the molecular mechanisms involved in renal tubular Pi reabsorption and its hormonal regulation and modulation by dietary Pi intake. This chapter focuses on Pi chemistry and homeostasis, the cellular and molecular aspects of intestinal and renal Pi transport and their regulation, and Pi transport in bone. In addition, the role of novel Pi regulating genes, PHEX and FGF23, in renal Pi handling and vitamin D metabolism will be described. For a more detailed discussion of renal Pi wasting disorders in humans, the reader is referred to Chapters 69 and 70 in this volume.

II. PHOSPHATE HOMEOSTASIS

VI. Phosphate Transport in Bone VII. Disorders Causing Phosphate Deficiency VIII. Summary and Conclusions References

Pi is an important constituent of bone mineral, and thus in growing individuals, the balance of Pi must be positive to meet the needs of skeletal growth and consolidation; in the adult, Pi balance is zero. Pi deficiency results in osteomalacia in both children and adults.

B. Phosphate Chemistry Phosphate exists in the plasma in two forms, an organic form, consisting principally of phospholipids and phosphate esters, and an inorganic form [2]. Of the total plasma Pi concentration of approximately 14 mg/dl (4.52 mM), about 4 mg/dl (1.29 mM) is in the inorganic form. Of this, about 10% to 15% is protein bound and the remainder, which is freely filtered by the renal glomerulus, exists principally either as “free” Pi ions or as Pi complexed with sodium, calcium, or magnesium. In principle, four forms of inorganic Pi are present in biological solutions: H3PO4, H2PO4−, HPO42−, and PO43−, which are in a pH-dependent equilibrium. However, at physiological pH, only HPO42− and H2PO4− are present at significant concentrations. The ratio of the divalent to monovalent forms can be determined by the HendersonHasselbalch equation, pH = pKa + log (HPO42−/H2PO4−). The dissociation constant, pKa, for Pi is 6.8. Thus, at a pH of 7.4, the ratio of divalent to monovalent Pi anions is essentially 4:1. In clinical settings, only the inorganic orthophosphate form of Pi is routinely measured.

A. Phosphate Distribution in the Body Pi accounts for approximately 0.6% of body weight at birth and about 1% of body weight, or 600 to 700 g, in the adult [1]. Approximately 85% of body Pi is in the skeleton and teeth, approximately 15% is in soft tissue, and the remainder (∼0.3%) is in extracellular fluid. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

C. Extracellular Phosphate Homeostasis In the adult in zero Pi balance, net intestinal absorption of Pi (dietary Pi minus fecal Pi) is approximately 60% to 65% of dietary intake. To satisfy the demands Copyright © 2005, Elsevier, Inc. All rights reserved.

454

HARRIET S. TENENHOUSE AND ANTHONY A. PORTALE

of rapid growth of bone and soft tissue, net intestinal absorption of Pi in infants is higher than in the adult and can exceed 90% of dietary intake [3,4]. Metabolic balance studies in normal adult humans reveal that over the customary range of dietary Pi, net absorption is a linear function of intake [5], with no indication of saturation. Thus, inadequate net Pi absorption results primarily from decreased Pi availability rather than from changes in the intrinsic capacity of intestinal Pi transport. A small amount of Pi is secreted into the intestinal lumen in digestive fluids. Absorbed Pi enters the extracellular Pi pool, which is in equilibrium with the bone and soft-tissue Pi pools. Pi is filtered at the glomerulus and is reabsorbed to a large extent by the renal tubule. In subjects in zero Pi balance, the amount of Pi excreted by the kidney is equal to the net amount absorbed by the intestine and in growing children is less than the net amount absorbed due to deposition of Pi in bone. An overall schema of Pi homeostasis is depicted in Fig. 1. Renal tubular reabsorption of Pi plays a central role in the regulation of plasma Pi concentration and Pi homeostasis. In response to a decrease in the extracellular

Pi concentration, urine excretion of Pi decreases promptly. This acute response reflects both a decrease in the filtered Pi load and an intrinsic adaptive increase in proximal tubule Pi reabsorption induced by hypophosphatemia or decreased dietary Pi intake (see Section IV,E,3). Hypophosphatemia also is a potent stimulus for the renal synthesis of 1,25-dihydroxyvitamin D3 (1,25(OH)2D) [6–8] (see Section V), and the resulting increase in serum 1,25(OH)2D concentration acts to stimulate intestinal absorption of Pi and calcium absorption and their mobilization from bone. Hypophosphatemia also can directly promote mobilization of Pi and calcium from bone. With an increase in plasma calcium concentration, PTH release is suppressed, which leads to a further decrease in renal Pi excretion but an increase in calcium excretion. These homeostatic adjustments result in an increase in extracellular Pi concentration toward normal values, with little change in serum calcium concentration. Conversely, in response to an increase in plasma Pi concentration, production of 1,25(OH)2D is decreased and release of PTH is increased. The effects of hyperphosphatemia on bone, kidney, and intestine are opposite to those

DIET 1400 mg

ECF Pi Absorption

Formation

1100 mg GUT

350 mg 550 mg

BONE

Secretion

Resorption

200 mg

350 mg

7000 mg

6100 mg

FECES 500 mg

KIDNEY

URINE 900 mg

FIGURE 1 Phosphate fluxes between body pools in the normal human adult in zero phosphate balance. Reprinted with permission from [189].

455

CHAPTER 26 Phosphate Homeostasis

occurring with hypophosphatemia, the net result being a decrease in plasma Pi concentration to normal values. In healthy subjects ingesting typical diets, the serum Pi concentration exhibits a circadian rhythm, characterized by a rapid decrease in early morning to a nadir shortly before noon, a subsequent increase to a plateau in late afternoon, and a small further increase to a peak shortly after midnight [9,10]. The amplitude of the rhythm (nadir to peak) is approximately 1.2 mg/dl (0.39 mM), or 30% of the 24-hr mean level. Restriction or supplementation of dietary Pi induces a substantial decrease or increase, respectively, in serum Pi concentrations during the late morning, afternoon, and evening, but induces less or no change in the morning fasting Pi concentration [10]. To minimize the impact of changes in dietary Pi on the serum Pi concentration, specimens for analysis should be obtained in the morning fasting state. Specimens obtained in the afternoon are more likely to be affected by diet and thus may be more useful to monitor the effect of manipulation of dietary Pi on serum Pi concentrations, as in patients with renal insufficiency receiving phosphorus-binding agents to treat hyperphosphatemia. Other factors can also affect the serum Pi concentration. Presumably because of movement of Pi into cells, the serum Pi concentration can be decreased acutely by intravenous infusion of glucose or insulin, ingestion of carbohydrate rich meals, acute respiratory alkalosis, or infusion or endogenous release of epinephrine. Serum Pi concentration can be increased acutely by metabolic acidosis and by intravenous infusion of calcium [11]. There are substantial effects of age on the fasting serum Pi concentration. In infants in the first 3 months of life, Pi levels are highest (4.8 to 7.4 mg/dl, mean 6.2 mg/dl [2 mM]) and decrease at age 1–2 years to 4.5 to 5.8 mg/dl (mean 5.0 mg/dl [1.6 mM]) [12]. In midchildhood, values range from 3.5 to 5.5 mg/dl (mean 4.4 mg/dl [1.42 mM]) and decrease to adult values by late adolescence [13,14]. In adult males, serum Pi is ∼3.5 mg/dl (1.13 mM) at age 20 years and decreases to ∼3.0 mg/dl (0.97 mM) at age 70 [14,15]. In women, the values are similar to those of men until after the menopause, when they increase slightly from ∼3.4 mg/dl (1.1 mM) at age 50 years to 3.7 mg/dl (1.2 mM) at age 70.

III. INTESTINAL PHOSPHATE ABSORPTION A. Cellular Aspects Dietary Pi is absorbed in the small intestine, with most of the absorption occurring in the duodenum and jejunum and minimal absorption in the ileum.

Pi absorption occurs by a nonsaturable, paracellular pathway as well as by a saturable, energy-requiring Na+-dependent process that has been localized to the mucosal surface. The movement of Pi from the small intestinal lumen into the blood involves transport across the mucosal brush border membrane against an electrochemical gradient, transport through the cytoplasm, and efflux across the serosal membrane [16]. A variety of approaches have been used to investigate Pi absorption by the small intestine. These include in vivo methods where the appearance of labeled Pi in the blood is monitored as a function of time after its oral administration, and in vitro studies where the Pi fluxes across purified mucosal or serosal membrane vesicle preparations are determined. Mucosal brush border membrane Pi transport is saturable, Na+-dependent, and driven by a Na+-gradient (outside > inside) that is maintained by the basolateral membrane–associated Na+,K+-ATPase. The exit of Pi at the serosal (basolateral) surface occurs down an electrochemical gradient and has not been well characterized. However, evidence suggests that the process is carrier-mediated, Na+-independent, and electrogenic. Mucosal Na/Pi cotransport is pH-dependent, with transport activity higher at pH 6 than at pH 7.4.

B. Regulation Pi transport across the mucosal membrane is regulated by 1,25(OH)2D and dietary intake of Pi [16]. Administration of the vitamin D hormone to either vitamin D–deficient or -replete animals elicits a significant increase in net Pi absorption that is associated with a corresponding increase in Na/Pi cotransport Vmax across the mucosal brush border membrane. These findings, combined with the demonstration that Na+-dependent Pi transport across the mucosal brush border membrane correlates with intestinal transepithelial Pi transport, uphold the notion that Pi transport across the mucosal membrane is the rate-limiting step in intestinal Pi absorption. The response of intestinal Pi transport to 1,25(OH)2D is dependent on protein synthesis, occurs several hours after its administration, and apparently occurs after weaning. In a pig model in which mutant animals exhibit defective renal synthesis of 1,25(OH)2D, intestinal brush border membrane Na/Pi cotransport Vmax is similar in newborn mutant and wild-type animals but is significantly reduced in weanling mutants when compared to age-matched wild-type animals [17]. Moreover, while 1,25(OH)2D had no effect on mucosal Pi transport in newborn mutants, administration of 1,25(OH)2D corrected the Pi transport defect in weanling mutants [17].

456 Intestinal Na/Pi cotransport is highly age-dependent and decreases significantly in the first 7 days post partum in both wild-type and mutant piglets [17]. In addition to the more frequently documented genomic mechanisms for 1,25(OH)2D-stimulated intestinal Pi absorption, there is increasing evidence for nonnuclear actions of the vitamin D hormone on duodenal Pi transport. Nemere demonstrated that low levels of 1,25(OH)2D elicit a rapid increase in Pi transport in the perfused duodenal loop of normal chicks [18]. The acute nature of the stimulation, which is apparent in 4–8 min after 1,25(OH)2D administration, suggests a signal transduction pathway independent of genomic regulation. Low-Pi diet also elicits an increase in mucosal Na/Pi cotransport. However, the response to Pi restriction is likely mediated by 1,25(OH)2D. It is well known that the renal synthesis of the vitamin D hormone is stimulated by hypophosphatemic states [8] and that the adaptive intestinal response to Pi restriction is blunted in vitamin D–deficient animals [19].

C. Molecular Mechanisms In an effort to identify intestinal Na/Pi cotransporters by expression cloning, mRNA was isolated from intestine of 1,25(OH)2D-treated rabbits and shown to stimulate Na/Pi cotransport when expressed in Xenopus oocytes [20]. In subsequent studies, a Pi uptake stimulator (PiUS) that increases Na/Pi cotransport in Xenopus oocytes was cloned from rabbit intestine [21]. However, since hydropathy analysis of the 425-amino-acid protein suggested that PiUS lacks transmembrane segments, it was not considered to be a Pi transporter per se. Recently, a full-length cDNA was generated from an EST clone that exhibited sequence homology with the most abundant, renal type II Na/Pi cotransporter [22] (see Section IV,C). When expressed in Xenopus oocytes, the cDNA induced Na/Pi cotransport that was electrogenic, with a pH dependence that resembled that of intestinal brush border membrane Na/Pi cotransport, i.e., higher at pH 6 that at pH 7.4 [22]. Based on its high homology to the renal type II transporter, it was designated type IIb (solute carrier series SLC34A2) and the renal isoform was renamed type IIa. These transporters will be referred to as NPT2a and NPT2b throughout the present chapter. The NPT2b/Npt2b genes map to human and mouse chromosome regions 4p15.2 and 5C1, respectively (Table I). NPT2b mRNA is expressed in a variety of tissues including small intestine, but not in kidney [22]. The demonstration that the NPT2b protein is localized to the apical

HARRIET S. TENENHOUSE AND ANTHONY A. PORTALE

TABLE I Nomenclature and Chromosomal Localization of Human and Mouse Na/Phosphate Cotransporter Genes Common name

Gene namea

NPT1d Npt1d NPT2a Npt2a NPT2b Npt2b NPT2c Npt2c GLVR-1, PIT-1 Glvr-1, Pit-1 RAM-1, PIT-2 Ram-1, Pit-2

SLC17A1 Slc17a1 SLC34A1 Slc34a1 SLC34A2 Slc34a2 NA NA SLC20A1 Slc20a1 SLC20A2 Slc20a2

Human locusb

Mouse locusc

6p22 13A3.2 5q35.2 13B2 4p15.2 5C1 NA 2A3 2q14.1 2F3 8p11.21 8A3

aGene

nomenclature is available at www.gene.ucl.ac.uk/nomenclature. gene loci are available at www.ensembl.org/homo. cMouse gene loci are available at www.ensembl.org/mus. dUpper and lower case refer to human and mouse genes, respectively. NA indicates that neither the gene name nor the gene locus is assigned. bHuman

membrane of enterocytes suggested that it plays a role in the absorption of Pi by the small intestine [22].

D. Molecular Mechanisms Governing NPT2b Regulation Western blotting and immunohistochemical analysis revealed that NPT2b protein expression in the small intestine is regulated by dietary Pi intake [23] (Table II). Chronic Pi deprivation elicits an increase in Na/Pi cotransport in small intestinal brush border membrane vesicles and a corresponding increase in NPT2b protein abundance in the intestinal apical membrane. In contrast, both parameters were decreased by chronically feeding a high-Pi diet and, under these conditions, NPT2b protein was no longer detectable in the intestinal apical membrane [23]. The correlation between apical Na/Pi cotransport and NPT2b protein expression in these studies suggested that NPT2b plays an important role in intestinal Pi absorption and its regulation by dietary Pi intake. Injection of cholecalciferol elicits comparable increases in small intestinal brush border membrane Na/Pi cotransport and apical abundance of NPT2b protein [23] (Table II). However, corresponding increases in intestinal NPT2b mRNA were not evident [23], suggesting that the effect of vitamin D cannot be

457

CHAPTER 26 Phosphate Homeostasis

TABLE II Factor Pi restriction High Pi intake 1,25(OH)2D EGF Glucocorticoids Aging

Factors Involved in the Regulation of Intestinal Pi Absorption and NPT2b Gene Expression Mucosal Na/Pi cotransport

NPT2b protein abundance

NPT2b mRNA abundance

Ref.



  



ND  ND

NC NC NC,
  

[23] [23] [23,24] [25] [26] [24,26]

NC refers to no change in values. ND indicates that the measurement was not done.

explained by increased NPT2b gene transcription. In another study, however, the stimulation in Na/Pi cotransport elicited by 1,25(OH)2D was associated with comparable increases in both NPT2b mRNA and protein in 14-day-old rats [24]. In addition, in cultured intestinal epithelial cells, 1,25(OH)2D elicited an increase in NPT2b mRNA abundance (Table II) as well as NPT2b promoter activity [24]. Intestinal Na/Pi cotransport and apical NPT2b expression decreases with age in rats, either in the absence or presence of vitamin D [24] (Table II). Small intestinal NPT2b expression is also subject to regulation by epidermal growth factor (EGF) and glucocorticoids (Table II), hormones that are known to influence Pi homeostasis. EGF decreases NPT2b mRNA abundance in rat intestine and human intestinal cell (Caco-2) cultures [25]. EGF also inhibits NPT2b promoter activity in Caco-2 cells, and the latter response is abolished by actinomycin D, confirming transcriptional regulation [25]. The EGF response element in the NPT2b promoter was identified and evidence obtained for its interaction with the c-myb transcription factor [25]. Methylprednisolone, a glucocorticoid analog, also elicits a significant decrease in small intestinal brush border membrane Na/Pi cotransport and a concomitant decrease in NPT2b mRNA and protein expression [26] (Table II). The inhibition by methylprednisolone was evident in mice ranging from 14 days to 9 months of age, although apical transport activity and intestinal NPT2b expression were remarkably higher in the suckling mice than in adult animals [26] (Table II).

IV. RENAL PHOSPHATE TRANSPORT A. Physiology and Tubular Localization Considerable effort has been devoted to the study of Pi transport in the kidney, and much of the material

discussed here is covered in greater detail in review articles (see [27–34]). The proximal tubule is the major site of Pi reabsorption, with approximately 60% of the filtered load reclaimed in the proximal convoluted tubule and approximately 15% in the proximal straight tubule. In addition, a small but variable portion (<10%) of filtered Pi is reabsorbed in more distal segments of the nephron. Clearance studies in humans and experimental animals show that when the filtered load of Pi is progressively increased, Pi reabsorption increases until a maximum tubular reabsorptive rate for Pi, or TmP, is reached, after which Pi excretion increases in proportion to its filtered load. The measurement of TmP varies among individuals and within the same individual, due in part to variation in GFR. Thus the ratio, TmP/GFR, or the maximum tubular reabsorption of Pi per unit volume of GFR, is the most reliable quantitative estimate of the overall tubular Pi reabsorptive capacity [35] and can be considered to reflect the quantity of Na/Pi cotransporters available per unit of kidney mass. The serum Pi concentration at which Pi reabsorption is maximal is called the “theoretical renal Pi threshold”; this value is equal to the ratio, TmP/GFR, and closely approximates the normal fasting serum Pi concentration. Thus, the renal reabsorptive capacity for Pi is the principal determinant of the serum Pi concentration.

B. Cellular Aspects Transepithelial Pi transport is essentially unidirectional and involves uptake across the brush border membrane, translocation across the cell, and efflux at the basolateral membrane (Fig. 2). Pi uptake at the apical cell surface is the rate-limiting step in the overall Pi reabsorptive process and the major site of its regulation. As in the intestine, it is mediated by Na+-dependent

458

HARRIET S. TENENHOUSE AND ANTHONY A. PORTALE

Luminal

3Na+

Basolateral

type IIa

H2PO42−

Na+ type III Pi

xNa+ H2PO42−

type I

Pi A

Cl− / org. anions

−

Pi ?

FIGURE 2

Location of identified and postulated Na+-dependent and Na+-independent Pi transporters in the proximal tubule cell. Available data indicate that most of proximal Pi reabsorption occurs via the type IIa cotransporter (NPT2a), which is localized in the brush border membrane and is the major target for physiologic regulation of renal Pi reabsorption. Org. anions, organic anions; ?, as yet unidentified luminal Pi transporters; A− to the anion that is exchanged with Pi at the basolateral membrane. Reprinted with permission from [190].

Pi transporters that depend on the basolateral membrane-associated Na+,K+-ATPase. Na/Pi cotransport is electrogenic and sensitive to changes in pH, with 10- to 20-fold increases documented when the pH is raised from 6 to 8.5. Little is known about the translocation of Pi across the cell except that Pi anions rapidly equilibrate with intracellular inorganic and organic Pi pools. There are few data regarding the mechanisms involved in the efflux of Pi at the basolateral cell surface. The latter appears to be a passive process that is driven by the electrical gradient existing across the membrane and occurs via an anion exchange mechanism (Fig. 2).

C. Molecular Aspects cDNAs encoding Na/Pi cotransporters, designated type I (NPT1,SLC17A1) and type IIa (NPT2a, SLC34A1; formerly type II, NPT2), have been identified in mammalian kidney and share only 20% identity [36,37]. The NPT1 transporters are approximately 465 amino acids in length with seven to nine membranespanning segments, whereas the NPT2a transporters

comprise approximately 635 amino acids, with eight predicted membrane-spanning segments. The NPT1 and NPT2a transporters account for ∼85 and ∼13%, respectively, of renal Na/Pi cotransporter mRNAs [38]. The chromosomal localization of the NPT1 and NPT2a genes in human and mouse has been determined (Table I) and the human [39] and murine [40] genes cloned and characterized. NPT1 and NPT2a transcripts have been localized to the proximal tubule and NPT1 and NPT2a immunoreactive proteins to the brush border membrane of proximal tubular cells (Fig. 2). Whereas NPT1 is uniformly expressed in all segments of the proximal nephron, NPT2a expression is highest in the S1 segments of the proximal tubule (see [30]). Functional studies of NPT1 and NPT2a have been conducted in cRNA-injected oocytes and in cDNAtransfected renal cell lines (see [30]). Both NPT1 and NPT2a mediate high-affinity Na/Pi cotransport. However, their pH profiles differ significantly, that of NPT2a, and not NPT1, bearing close resemblance to the pH dependence of Na/Pi cotransport in isolated renal brush border membrane vesicles. Moreover, NPT1 exhibits broader substrate specificity than NPT2a and can mediate the transport of Cl− and organic anions. On the basis of these findings it was suggested that NPT1 not only mediates brush border membrane Na+-dependent Pi transport but also serves as an apical channel for Cl− transport and the excretion of anionic xenobiotics. The precise physiological role of NPT1 will thus require further study. Most recently, a Na/Pi cotransporter with homology to NPT2a and NPT2b was identified in rat, human [41], and mouse [42] kidney and designated NPT2c. Although the human and mouse NPT2c/Npt2c genes have been mapped, gene names have not yet been assigned (Table I). NPT2c is approximately one order of magnitude less abundant than NPT2a at the mRNA level, is expressed exclusively in the brush border membrane of proximal tubular cells, and is regulated by dietary Pi [41,42]. The relative abundance of Npt2c protein is significantly higher in kidneys of 22-day-old rats than in those of 60-day-old rats, suggesting that Npt2c is a growth-related renal Na/Pi cotransporter [41]. In addition, hybrid depletion studies suggested that Npt2c accounts for approximately 30% of Na/Pi cotransport in kidneys of Pi-deprived adult mice [42]. Further studies are necessary to determine the precise contribution of renal NPT2c to overall renal Pi reabsorption. The restricted tissue-specific expression of NPT1, NPT2a, and NPT2c is of interest. NPT1 is expressed primarily in kidney but has also been detected in liver and brain. NPT2a mRNA and protein have only been

459

CHAPTER 26 Phosphate Homeostasis

found in kidney and the multinucleated bone-resorbing cell, the osteoclast [43] (see Section VI). To date NPT2c has only been detected in kidney. Two additional Na/Pi cotransporters, designated type III, have been detected in mammalian kidney and account for <1% of renal Na/Pi cotransporter mRNAs. Both are cell surface viral receptors [gibbon ape leukemia virus (Glvr-1, Pit-1, SLC20A1) and murine amphotropic virus (Ram-1, Pit-2, SLC20A2)] that mediate high-affinity, electrogenic Na+-dependent Pi transport when expressed in oocytes and in mammalian cells [33,44]. Although Glvr-1 and Ram-1 show no sequence similarity to NPT1 or NPT2a, they share 60% sequence identity with a putative Pi permease of Neurospora crassa. Chromosomal localization of the two transporters in human and mouse is shown in Table I. Glvr-1 and Ram-1 are widely expressed in mammalian tissues (see Section VI) and may serve as “housekeeping” Na/Pi cotransporters. Both Glvr-1 and Ram-1 proteins are likely expressed in the basolateral membrane of polarized epithelial cells (Fig. 2) where they function to maintain cellular Pi homeostasis.

D. Structure–Function Studies of NPT2a NPT2a-mediated Na/Pi cotransport is electrogenic and involves the inward flux of three Na+ ions and one Pi anion (preferentially divalent) [45]. Binding of the first Na+ ion to the negatively charged carrier is followed by the interaction of two additional Na+ ions and Pi with the carrier [45,46]. The transfer of the fully loaded carrier is electroneutral. Electrogenicity is achieved by the reorientation of the empty carrier after the discharge of the Na+ and Pi ions to the cell interior [46]. Similar to other Na/solute cotransporters, there is a significant Na+ leak after interaction with the first Na+ ion [45]. However, the latter is of minimal physiological significance since the transporter is preferentially in its fully loaded transport cycle in the presence of Pi [45]. There is considerable information on the structures of the NPT2a and NPT2b transporters, and a schematic representation of NPT2a is shown in Fig. 3. Several different analytical approaches were used to arrive at this structure, and these are summarized as follows: (i) hydrophobicity predictions [37]; (ii) antibody accessibility combined with epitope insertion [47]; (iii) cysteine insertion and accessibility of permeant and impermeant sulfhydryl reagents [48,49]; and (iv) glycosylation studies [50]. These studies demonstrated several important features of NPT2a (Fig. 3): The transporter has a large extracellular loop that separates it into two domains. There is intramolecular homology

within the ICL1 and ECL3 domains. Both the NH2 and COOH termini are oriented intracellularly. The studies just described also addressed the functional significance of specific domains of NPT2a (Fig. 3). Cysteine-insertion studies suggested that ICL1 and ECL3 form an important part of a “permeation pore” that participates in both “cotransport” and “Na+ leak” function [48,49]. Chimera construction—based on different transport properties of NPT2a and NPT2b [22,37]— suggested the involvement of three amino acid residues in determining the pH dependence of NPT2a, namely increased transport at higher pH [51,52]. The chimera approach also suggested the importance of two basic amino acid residues in ICL3 for PTH-dependent internalization [53] (see Section IV,E,1). Deletion studies documented that the COOH terminus contains information for brush border membrane expression, i.e., a terminal PDZ-binding motif and a membrane internalization signal [54]. Both the NH2 and COOH termini portions of NPT2a are required for transport activity. However, functional studies revealed that cleavage of the NPT2a protein backbone, between the two glycosylation sites in the large extracellular loop, does not interfere with transport function [55,56]. It is assumed that under this condition a disulfide bridge within this large extracellular loop stabilizes the transporter (Fig. 3). Although NPT2a might be part of a multimeric complex (see later discussion), one transporter unit is sufficient to mediate Na/Pi cotransport [57].

E. Regulation of NPT2a Regulation of renal Pi reabsorption has been the subject of intense investigation (see Table III and [58–78]). The regulation is achieved primarily by alterations in the abundance of NPT2a protein in the brush border membrane of proximal tubular cells and occurs largely in the absence of changes in NPT2a mRNA (for review, see [30]). Changes in the amount of apical NPT2a protein are accomplished by either membrane insertion or retrieval/lysosomal degradation of the transporter. Although membrane trafficking of NPT2a is a key element in the regulation of renal Pi handling, little is known about the molecular signals that confer specificity to this process, i.e., why other transporters are not inserted into or retrieved from the membrane along with NPT2a. Several proteins, such as PDZK1 (NaPi-Cap1, diphor-1), NaPi-Cap2, and NHERF-1 [79], which interact with NPT2a, have been identified and likely play a key role in NPT2a trafficking in response to factors that stimulate or inhibit renal Pi transport [80].

460

HARRIET S. TENENHOUSE AND ANTHONY A. PORTALE

s s

ECL-3

ICL-1

ICL-3

NH2

TRL-COOH Glycosylation sites Na/Pi cotransport/Na+ slippage

PDZ interactions

α-helical pH-dependence PTH regulation/internalization Scaffolding/positioning

FIGURE 3 Structure–function relationship of the renal NPT2a Na/Pi cotransporter. See text for details. Reprinted with permission from [32].

1. PARATHYROID HORMONE

PTH is the major hormonal regulator of renal Pi reabsorption. PTH acts directly on proximal tubular cells and inhibits both apical and basolateral Na/Pi cotransport (Table III) by mechanisms that involve the internalization of cell surface NPT2a protein [60] and its subsequent lysosomal degradation [81]. Considerable progress has been made in the identification of the signaling pathways that mediate the PTH response and the proteins involved in the retrieval of NPT2a from the plasma membrane.

PTH binding to receptors on the basolateral membrane activates protein kinase A (PKA) and/or protein kinase C (PKC) signaling pathways, whereas its binding to apical receptors activates PKC [82]. The extracellular signal-regulated kinase/mitogen-activated protein kinase (ERK/MAPK) pathway also participates in PTH-induced signaling [83], and recent studies show that the PKA and PKC signaling pathways converge on the ERK/ MAPK pathway to internalize NPT2a protein [84]. Although the downstream targets for ERK/MAPKmediated phosphorylation remain unknown, changes

461

CHAPTER 26 Phosphate Homeostasis

in the phosphorylation state of NPT2a are not associated with its PTH-induced internalization [85]. Rather, it has been postulated that the phosphorylation of proteins that associate with NPT2a may determine its regulation. AKAP79, an A kinase anchoring protein [86], and RAP, a receptor-associated protein [87], have been shown to participate in the PTH-mediated retrieval of NPT2a from the plasma membrane of proximal tubular cells. In opossum kidney (OK) cells, AKAP79 associates with NPT2a and the regulatory and catalytic subunits of PKA, and this process is necessary for PKA-dependent inhibition of Na/Pi cotransport [86]. In RAP-deficient mice, PTH-induced internalization of NPT2a is significantly delayed, whereas regulation by dietary Pi (see Section IV,E,3) is not affected [87]. In contrast, NHERF-1, which associates with NPT2a and MAP-17 in mouse kidney [88], does not appear to play a role in PTH-induced internalization of NPT2a [89]. 2. OTHER HORMONAL REGULATORS

Other hormones also contribute to the regulation of proximal tubular Pi transport (Table III). Growth hormone, insulin-like growth factor-I, insulin, 1,25(OH)2D, and thyroid hormone all stimulate Pi reabsorption,

TABLE III Factor Pi restriction High Pi intake PTH IGF-1 Insulin 1,25(OH)2D T3 PTHrP ANF EGF TGF-α and -β Glucocorticoids Stanniocalcin FGF-23 sFRP-4 Aging

whereas PTH-related peptide, calcitonin, atrial natriuretic factor, epidermal growth factor, transforming growth factor-α, and glucocorticoids inhibit Pi reclamation (for review, see [30]). The increase in brush border membrane Na/Pi cotransport by thyroid hormone is associated with an increase in NPT2a mRNA [66], whereas both dexamethasone [73] and epidermal growth factor [69] decrease NPT2a mRNA abundance. Because thyroid hormone and glucocorticoids exert their effects on gene transcription via ligandactivated receptors that bind to response elements in the promoter region of target genes, it was suggested that these hormones exert their effects by modulating NPT2a gene transcription. However, neither thyroid hormone nor dexamethasone has an effect on NPT2a promoter–reporter gene expression (see [90]), suggesting that transcriptional mechanisms are not involved. Moreover, recent studies have been unable to detect a thyroid hormone–mediated increase in NPT2a mRNA (see [90]). Administration of 1,25(OH)2D to vitamin D–deficient rats elicits an increase in brush border membrane Na/Pi cotransport that is accompanied by an increase in renal NPT2a mRNA and protein abundance (Table III) [64]. Although these results are consistent with direct

Factors Involved in the Regulation of Renal Na/Pi Cotransport and NPT2a Gene Expression Na/Pi cotransporta

NPT2a gene expressionb

Ref.


 



        


  ND ND

ND ND  ND  ND  ND 

[58,59] [58,59] [60,61] [62] [63] [64] [65,66] [67] [68] [69,70] [71,72] [73] [74] [75,77] [76] [66,78]

aNa/P cotransport studies were performed in cultured proximal tubular cells or in renal brush border membrane vesicles i prepared from appropriate animal models. bRefers to changes in renal Npt2a mRNA or protein abundance. ND indicates that the measurement was not done. For other abbreviations, see text.

462 effects of 1,25(OH)2D on NPT2a-mediated renal Na/Pi cotransport, the effects may result from a 1,25(OH)2Ddependent decrease in PTH levels. However, the finding that 1,25(OH)2D increased the activity of a NPT2a promoter–luciferase reporter gene construct suggests a direct effect of the vitamin D hormone on NPT2a gene transcription [64]. Other hormones that regulate renal Pi handling include stanniocalcin, 5-hydroxytryptamine (5-HT), fibroblast growth factor (FGF)-23, and secreted frizzledrelated protein (sFRP)-4 (Table III). Stanniocalcin is a peptide hormone that counteracts hypercalcemia and stimulates Pi reabsorption in bony fish, and is also produced by humans. Infusion of stanniocalcin in rats stimulates renal Pi reabsorption and brush border membrane Na/Pi cotransport [74], suggesting a role for stanniocalcin in the maintenance of Pi homeostasis in mammals as well as fish. 5-HT is synthesized in the kidney and locally generated 5-HT was shown to interfere with PTH-mediated inhibition of renal Na/Pi cotransport [91]. These findings suggested that 5-HT is a paracrine modulator of renal Pi transport. FGF-23 is a secreted peptide contributing to renal Pi wasting in autosomal dominant hypophosphatemic rickets and oncogenic hypophosphatemic osteomalacia (see Sections VII,B and C). Administration of FGF-23 elicits a decrease in serum Pi in mice that is attributed to renal Pi wasting [92], a reduction in brush border membrane Na/Pi cotransport [75], and decreased renal Npt2a expression (Table III) [77]. In addition, rats receiving intrahepatic injection of FGF-23 cDNA develop hypophosphatemia and a significant decrease in both brush border membrane Na/Pi cotransport and NPT2c protein abundance [93]. FGF-23-dependent inhibition of NPT2a-mediated Na/Pi cotransport in OK cells occurs by activation of MAP kinase [94]. sFRP-4, which, like FGF-23, is highly expressed in tumors from patients with oncogenic hypophosphatemic osteomalacia [76] (see Section VII,C), has also been tested for its phosphaturic action. sFRP-4 induced a specific increase in the renal fractional excretion of Pi and hypophosphatemia when infused in rats and inhibited Na/Pi cotransport in vitro when added to OK cells (Table III) [76]. 3. REGULATION BY DIETARY PHOSPHATE

Dietary Pi intake is a key determinant of renal Pi handling. Pi deprivation elicits an increase in Pi reabsorption and brush border membrane Na/Pi cotransport. Both acute and chronic exposure to low dietary Pi induce an increase in transport Vmax as well as in NPT2a protein but not NPT2a mRNA [59] (Table III). These findings are consistent with recent data demonstrating

HARRIET S. TENENHOUSE AND ANTHONY A. PORTALE

that low dietary Pi has no effect on NPT2a promoter activity [90]. The acute response to Pi deprivation is mediated by microtubule-dependent recruitment of existing NPT2a protein to the apical membrane [95]. In contrast, exposure to high dietary Pi leads to the internalization of cell surface NPT2a protein into the endosomal compartment by a microtubule-independent mechanism [95]. Internalized NPT2a protein is then delivered to the lysosome, by a microtubuledependent process, for degradation [59]. A Pi response element (PRE) was identified in the mouse NPT2a promoter by DNA footprint analysis [96]. The PRE was shown to bind a mouse transcription factor, TFE3, and the expression of TFE3 is increased in the kidney in response to Pi deprivation [96]. On the basis of these results, it was suggested that TFE3 participates in the transcriptional regulation of the NPT2a gene by dietary Pi. Further work is necessary to assess the physiological significance of these findings.

F. Lessons from Mouse Models with Renal Phosphate Transport Defects 1. THE NPT2a KNOCKOUT MOUSE

The critical role of NPT2a in the maintenance of Pi homeostasis was clearly demonstrated in mice in which the Npt2a gene (lowercase refers to the mouse gene) was knocked out by targeted mutagenesis [97]. Mice homozygous for the disrupted gene (Npt2−/−) exhibit decreased renal Pi reabsorption, an ∼80% loss of brush border membrane Na/Pi cotransport, hypophosphatemia, an appropriate adaptive increase in the renal synthesis [98] and serum concentration of 1,25(OH)2D [97,98] and associated hypercalcemia, hypercalciuria, and hypoparathyroidism, and an agedependent skeletal phenotype [97,99]. Of interest was the demonstration that renal brush border membrane Na/Pi cotransport is not increased in Npt2−/− mice maintained on a low-Pi diet [100] and is not decreased in Npt2−/− mice following PTH administration [101]. These findings demonstrate unequivocally that NPT2a is the target for regulation of renal Pi handling by PTH and dietary Pi. Recent studies show that Npt2c, but not Npt1, is upregulated in Npt2−/− mice and that the 2.8-fold increase in renal brush border membrane Npt2c protein in the mutants is not accompanied by an increase in Npt2c mRNA [102]. The failure of Npt2−/− mice to respond to Pi restriction with a further increase in Npt2c protein suggests that Npt2c protein is maximally up-regulated and likely accounts for residual brush border membrane Na/Pi cotransport in Npt2−/− mice [102].

463

CHAPTER 26 Phosphate Homeostasis

2. THE X-LINKED HYP AND GY MOUSE MODELS

Hyp and Gy are spontaneous and radiation-induced mutations, respectively, on the mouse X chromosome, and mice harboring these mutations have served as valuable models to elucidate the pathophysiology of the homologous human disorder, X-linked hypophosphatemia (XLH) (Table IV) [103]. Mutations in Phex/ PHEX, a Pi regulating gene with homologies to endopeptidases on the X chromosome, have been identified in Hyp [104] and Gy [105] mice and XLH patients [106]. Like patients with XLH, Hyp and Gy mice are characterized by growth retardation, rickets and osteomalacia, hypophosphatemia, and renal defects in Pi transport and the regulation of vitamin D metabolism [103] (see Section V,C). In Hyp and Gy mice, the increased urinary Pi excretion can be attributed to a 50% decrease in renal brush border membrane Na/Pi cotransport Vmax [107,108] and a comparable reduction in the renal expression of Npt2a mRNA and immunoreactive protein [109–111]. In addition, renal expression of both Npt1 and Npt2c mRNA and protein is significantly decreased in Hyp mice when compared to that in normal littermates [102]. These data suggest that loss of Phex function is responsible for the down-regulation of all three renal Na/Pi cotransporters in the mutant strain. The Pi transport defect in Hyp mice is not intrinsic to the kidney but rather is dependent on a circulating factor, and this conclusion is based on several findings. Parabiosis of normal mice to Hyp mice leads to the development of hypophosphatemia, reduced renal Pi reabsorption, and a PTH-independent decrease in renal brush-border membrane Na/Pi cotransport in the normal mouse of the normal–Hyp pair [112,113]. Furthermore, renal cross-transplantation revealed that hypophosphatemia and renal Pi wasting persist in Hyp mice that

TABLE IV Disorder XLH Hyp Gy Ska1 ADHR OHO HHRH Npt2−/−

Serum Pi        

have received a normal kidney, whereas normal serum Pi and normal renal Pi handling persist in normal mice that have received a Hyp kidney [114]. In further support for this hypothesis, serum from Hyp mice but not normal mice inhibited Na/Pi cotransport in primary mouse renal cell cultures [115]. Inhibition of Na/Pi cotransport in target renal cells was also elicited with conditioned medium derived from cultured Hyp osteoblasts, whereas no inhibition was apparent with conditioned medium from cultured normal osteoblasts [115]. The latter data suggest that the Hyp osteoblast is responsible for the release and/or modification of a humoral factor that accumulates in the absence of Phex function and inhibits Na/Pi cotransport in renal epithelial cells. A likely candidate for this factor is FGF-23 (see Section VII). Of interest is the recent report of a point mutation in the mouse Phex gene that was introduced by random mutagenesis with ethyl-nitrosourea [116]. The mutation, designated Ska1, causes skipping of exon 8 and results in clinical and biochemical phenotypes that are similar to those in Hyp mice and XLH patients, including hypophosphatemia, rickets, and osteomalacia (Table IV) [116].

V. ROLE OF PHOSPHATE IN THE REGULATION OF RENAL VITAMIN D METABOLISM A. Lessons from Mouse Models Dietary Pi intake and serum Pi concentration are critically important determinants of the renal metabolism of 1,25(OH)2D. Tanaka and DeLuca first demonstrated that rats maintained on a low-Pi diet

Features of Renal Phosphate Wasting Disorders in Humans and Mice Serum 1,25(OH)2D NC NC NC ND NC NC



Urine Ca/creatinine

Primary defect

Ref.

NC NC NC ND NC NC



PHEX mutation 3′Phex deletion 5′Phex deletion Phex mutation FGF23 mutation Tumor-mediated Unknown Npt2 deletion

[106] [104] [105] [116] [168] [188] [160] [97]


, an increase in values; , a decrease in values; NC, no change in values; ND, measurement not done.

464

HARRIET S. TENENHOUSE AND ANTHONY A. PORTALE

Studies in intact mice reveal that restriction of dietary Pi induces a rapid, sustained, approximately six- to eightfold increase in renal mitochondrial 1α-hydroxylase activity (Fig. 4A, C), which can be attributed to an approximately sevenfold increase in Vmax of the enzyme, with no measurable change in its Km [130]. The increase in 1α-hydroxylase activity can be attributed to an increase in the renal abundance of 1α-hydroxylase (P450c1α, CYP27B1) mRNA (Fig. 4B, D) and protein [130,131]. Immunohistochemical analysis demonstrates that the increase in P450c1α gene expression occurs exclusively in the proximal renal tubule and is due, at least in part, to increased transcription of the P450c1α gene [130]. Degradation of 1,25(OH)2D to calcitroic acid and cholacalcioic acid via the 24-oxidation pathway is initiated by the enzyme 25-hydroxyvitamin D-24-hydroxylase (24-hydroxylase, P450c24, CYP24). Pi restriction induces a significant decrease in renal 24-hydroxylase activity [132] and mRNA abundance [130,133].

synthesized predominately 1,25(OH)2D, and that this effect was not diminished by parathyroidectomy [6]. They further observed an inverse relationship between the synthesis of 1,25(OH)2D and the serum Pi concentration. Numerous studies in animals demonstrate that hypophosphatemia induced by dietary Pi restriction induces an increase in the production rate of 1,25(OH)2D measured in vivo [117,118] and in 1α-hydroxylase activity measured in vitro [7,8,119]. Studies in healthy human subjects are consistent with those in experimental animals: Restriction of dietary Pi induces an increase [10,120–123] and supplementation of dietary Pi [10,123,124] a decrease in the serum concentration of 1,25(OH)2D and in its in vivo production rate [123]. It is of interest that the stimulation of 1,25(OH)2D production by Pi depletion is abolished by hypophysectomy and is restored by administration of growth hormone or insulin-like growth factor I (IGF-1) [125–129], indicating that an intact growth hormone/IGF-1 axis is required for the effect; however, the molecular mechanisms are unknown.

40 # 20

0.02%

0.6%

1α-Hydroxylase activity (pg/mg/15 min)

125 100 75 50

#

25 # 0

FIGURE 4

WT Npt2−/−

800 600

#

400 200 0

1.6%

#

WT Hyp

1000 P450c1α mRNA abundance (% control)

1α-Hydroxylase activity (pg/mg/15 min)

WT Npt2−/−

0

C

B

60

0.02%

0.6%

1.6%

D P450c1α mRNA abundance (% control)

A

#

WT Hyp

600

400 #

200 # 0

0.02%

1% Dietary phosphorus

1.6%

0.02%

1% Dietary phosphorus

1.6%

Effects of dietary Pi on renal mitochondrial 1α-hydroxylase activity (A, C) and renal P450c1α mRNA abundance (B, D) in wildtype mice (white bars) and either Npt2−/− (A, B) or Hyp (C, D) mice (hatched bars). Mice were fed the low (0.02%), control (0.6–1%), or high (1.6%) Pi diet for 4-5 days. Renal mitochondria were prepared and incubated with 25-hydroxyvitamin D3 to determine 1α-hydroxylase activity, and renal total RNA was prepared to estimate the abundance of P450c1α mRNA, relative to β-actin mRNA, by ribonuclease protection assay. Bars depict mean ± SEM. *, Compared with the 0.6% or 1% Pi diet, within each species, P < 0.05. #, Compared with wild-type animals, within each diet group, P < 0.05.

CHAPTER 26 Phosphate Homeostasis

B. The Npt2a Knockout Mouse Mice homozygous for the disrupted Na/Pi cotransporter gene (Npt2−/−) have been used to address the question of the relationship between renal reabsorption of Pi, per se, and regulation of P450c1α gene expression. It has been postulated that normal transepithelial transport of Pi in the proximal renal tubule is essential for the regulation of renal 1,25(OH)2D synthesis by Pi. Beck et al. reported that renal Pi wasting and hypophosphatemia in Npt2−/− mice is associated with an increase in serum 1,25(OH)2D concentration [97]. The increase in serum 1,25(OH)2D in Npt2−/− mice can be attributed to an increase in renal 1α-hydroxylase activity and P450c1α mRNA abundance (Fig. 4A, B) [98]. Thus, renal 1α-hydroxylase activity and mRNA expression in Npt2−/− mice increase appropriately in response to the hypophosphatemia that results from renal Pi wasting. Moreover, restriction or supplementation of dietary Pi in Npt2−/− mice elicits a further increase or a decrease, respectively, in serum 1,25(OH)2D concentrations, in renal mitochondrial 1α-hydroxylase activity and in P450c1α mRNA abundance (Fig. 4A, B) [98]. In both wild-type and Npt2−/− mice, values of P450c1α mRNA abundance vary inversely and significantly with serum Pi concentrations. These studies provide unequivocal evidence that normal renal Na/Pi cotransport is not necessary for the regulation of renal 1,25(OH)2D production by Pi, and that renal Pi wasting in itself does not contribute to dysregulated 1,25(OH)2D production, thereby dispelling a previously held notion. These findings suggest that changes in serum Pi concentration per se are sufficient to initiate the signaling pathways involved in the up-regulation and down-regulation of the P4501α gene by manipulation of dietary Pi.

C. The X-Linked Hyp Mouse In contrast to the response to Pi depletion in wildtype and Npt2−/− mice, the hypophosphatemia that results from renal Pi wasting in XLH is attended by abnormal regulation of vitamin D metabolism. In patients with XLH, serum concentrations of 1,25(OH)2D are nominally within the normal range [134,135]; however, such values are inappropriately low given the attendant hypophosphatemia, and serum 1,25(OH)2D concentrations fail to increase with dietary Pi restriction [121]. Regulation of vitamin D metabolism is similarly disordered in the Hyp mouse model of XLH. In Hyp mice, the serum concentration of 1,25(OH)2D is not different from that in normal littermates and, therefore, is inappropriately low for the degree of hypophosphatemia [136,137].

465 Renal 1α-hydroxylase activity also is inappropriately low for the degree of hypophosphatemia [119], and Pi restriction induces a paradoxical decrease in both serum concentration of 1,25(OH)2D [136,137] and renal 1α-hydroxylase activity [138]. The molecular basis for the disordered regulation of vitamin D metabolism in Hyp mice was explored in wild-type and Hyp mice subjected to manipulation of dietary Pi. In wild-type mice a low-Pi diet elicits sixfold and threefold increases in renal 1α-hydroxylase activity and P450c1α mRNA expression, respectively, whereas in the Hyp strain the diet-induced changes were of similar magnitude but opposite in direction (Fig. 4C, D) [139]. A high-Pi diet was without effect in wild-type mice, whereas in Hyp mice the same diet induced threefold and twofold increases, respectively, in enzyme activity and P450c1α mRNA abundance (Fig. 4C, D) [139]. In wild-type mice, both renal 1α-hydroxylase activity and P450c1α mRNA abundance varied inversely and significantly with serum Pi concentrations, whereas in Hyp mice the relationship between both renal parameters and serum Pi concentration was direct [139]. The regulation of 24-hydroxylase also is abnormal in Hyp mice, with renal enzyme activity being twofold higher in Hyp than in wild-type littermates and paradoxically increasing further with Pi restriction [137]. In Hyp mice, Pi restriction induced a significant increase in renal P450c24 mRNA abundance [139–141], in contrast to the decrease observed in wild-type mice [139]. These findings suggest that the decrease in the serum concentration of 1,25(OH)2D in phosphorus-restricted Hyp mice can be attributed to an increase in its renal catabolism as well as to a reduction in its renal synthesis [139]. Nevertheless, current data demonstrate that regulation of both the P450c1α and P45024 genes by Pi is disordered in Hyp mice at the level of renal 1α-hydroxylase activity and renal P450c1α and P450c24 mRNA expression and suggest that loss of Phex function in Hyp mice gives rise to disordered transcriptional regulation of both the P450c1α and P450c24 genes. The extent to which posttranscriptional or posttranslational mechanisms contribute to regulation of expression of these genes remains to be determined.

VI. PHOSPHATE TRANSPORT IN BONE Pi is essential for the mineralization of extracellular matrix by osteogenic cells and is a vital element for osteogenic cell function. Thus, a considerable effort has been devoted to the characterization of Pi transport in osteoblasts and chondrocytes, the cells that produce

466 and secrete the extracellular matrix. Pi is also crucial for energy metabolism in osteoclasts, multinucleated cells involved in bone resorption. Recent progress on Pi transport in osteoblasts and osteoclasts is briefly described below (for review, see [142]).

A. Osteoblasts and Matrix Vesicles Studies of Pi transport in bone have primarily employed osteoblast cell lines, such as human osteosarcoma–derived SaOS-2 cells, rat osteosarcoma– derived UMR-106 and ROS 17/2.8 cells, and murine and rat calvarial–derived MC3T3-E1 and PyMS cells. Since these cell lines may differ, for example in their state of differentiation, the extent to which data generated from such in vitro studies reflects the in vivo situation remains to be determined. Na/Pi cotransport in cultured osteoblasts differs from that in renal proximal tubular cells in several respects [143]. Km values for Pi are higher in osteoblasts (300–500 µM vs 50–100 µM) and transport in osteoblasts is greater at acidic pH [143]. Moreover, several factors that inhibit Na/Pi cotransport in renal proximal tubular cells (Table III) stimulate transport in osteoblasts (Table V). In both cell types, Na/Pi cotransport is regulated by the extracellular Pi concentration, with transport increasing significantly as the Pi concentration in the culture medium is reduced [144] (Tables III and V). Na/Pi cotransport in osteoblasts is also stimulated by a variety of hormones and agents [145–154] (see Table V), which are of interest because of their potential therapeutic applications. The mechanisms involved in the stimulation of Na/Pi cotransport by many of these agents have been investigated. The actions of PTH and PTHrP are dependent on the cAMP-PKA pathway, whereas the stimulation by IGF-1, PDGF, bFGF, and fluoride is dependent on tyrosine phosphorylation, and the action of epinephrine, endothelin-1, and prostaglandin E2 is associated with the activation of protein kinase C– mediated phosphorylation. GLVR-1(PIT-1), the ubiquitous high-affinity type III Na/Pi cotransporter (see Section IV,C), is expressed in SaOS-2 cells [147]. The increase in Na/Pi cotransport elicited by IGF-1 in SaOS-2 cells is associated with a comparable increase in the relative abundance of GLVR-1 mRNA [147]. Furthermore, in cultured MC3T3-E1 cells, Glvr-1/Pit-1 mRNA abundance increased during cell differentiation and correlated with increases in osteocalcin mRNA and mineralization [155]. Although these findings suggest that Glvr1/Pit-1 mRNA may be used as a marker for osteoblast differentiation, the relative contribution of Glvr-1/Pit-1 to Pi transport in osteoblasts and the role of as yet

HARRIET S. TENENHOUSE AND ANTHONY A. PORTALE

TABLE V Factors Stimulating Na/Pi Cotransport in Cultured Osteoblasts Factor Pi restriction PTH IGF-1 1,25 (OH)2D PTHrPa TGF-αa PDGFa bFGFa Epinephrine Endothelin-1 Prostaglandin E2 Fluoride

Cell line

Ref.

MC3T3-E1 UMR-106 SaOS-2 PyMS UMR-106 UMR-106 MC3T3-E1 MC3T3-E1 MC3T3-E1 MC3T3-E1 PyMS UMR-106

[142,144] [61,145] [147] [148] [146] [146] [149] [150] [151] [152] [153] [154]

aPTHrP, parathyroid hormone related peptide; TGF-α, transforming growth factor-α; PDGF, platelet-derived growth factor; bFGF, basic fibroblast growth factor.

unidentified Na/Pi cotransporters in the osteoblast remain to be determined. Pi transport has also been examined in matrix vesicles, which are structures that derive from osteoblasts by budding from elongated tubular extensions that project from the plasma membrane of these cells [142]. Pi transport in matrix vesicles is Na+ dependent and reflects the transport activity of the cells from which they originated. It has been proposed that Pi accumulation inside the matrix vesicles is involved in the formation of nucleation sites that are necessary for initiation of the calcification process [142].

B. Osteoclasts Osteoclasts exhibit Na/Pi cotransport activity that is stimulated by the addition of bone particles to the culture medium and is inhibited by peptides with the RGDS (arg-gly-asp-ser) motif, suggesting that integrins and cell–matrix interactions are involved in the regulation of the transport process [156]. There is evidence that an NPT2a-like Na/Pi cotransporter that is present in osteoclasts is localized to discrete vesicles in unpolarized osteoclasts and to the basolateral membrane in actively resorbing cells [157]. Furthermore, the addition of phosphonoformic acid to osteoclast cultures led to inhibition of Na/Pi cotransport, diminished ATP production, and decreased bone resorption [157]. Taken together, these findings suggest that Na/Pi cotransport in osteoclasts is essential for energy metabolism and osteoclast-mediated bone resorption.

CHAPTER 26 Phosphate Homeostasis

Moreover, the data suggest that some of the Pi released from bone during the resorptive process might be utilized by the osteoclast to maintain cellular ATP content during the cylical processes of migration, attachment, and resorption. In light of the finding that NPT2a is expressed in the osteoclast, it was of interest to characterize the skeletal phenotype in mice homozygous for the disrupted Npt2a gene (see Section IV,F,1). Indeed, histomorphometric analysis revealed that Npt2−/− mice exhibit an age-dependent skeletal phenotype that is attributable not only to hypophosphatemia, secondary to the renal defect in brush border membrane Na/Pi cotransport, but also to an intrinsic osteoclast defect [99]. In addition to Npt2a, Pit-1 is expressed in the mouse osteoclast [43]. Furthermore, Npt2a and Pit-1 appear to associate with NHERF-1, ezrin, and actin, suggesting that such interactions play an important role in membrane sorting and regulation of Npt2a- and Pit-1mediated Na/Pi cotransport in the osteoclast [43].

VII. DISORDERS CAUSING PHOSPHATE DEFICIENCY Since Pi is sufficiently abundant in the diet, Pi deficiency is unlikely to develop except under unusual circumstances (for reviews, see [158,159]). The following review will be limited to well-characterized hypophosphatemic disorders associated with renal Pi wasting in humans. Three of the disorders, X-linked hypophosphatemia (XLH), autosomal dominant hypophosphatemic rickets (ADHR), and hereditary hypophosphatemic rickets with hypercalciuria (HHRH), are inherited, and the fourth, oncogenic hypophosphatemic osteomalacia (OHO), is acquired. Each is characterized by rickets and osteomalacia and, with the exception of HHRH, by abnormal regulation of renal vitamin D metabolism (Table IV). The discussion will focus primarily on the renal Pi transport phenotype and the underlying pathophysiology. For a more detailed review of these disorders, the reader is referred to Chapters 69 and 70 in this volume.

A. X-Linked Hypophosphatemia (XLH) The features that distinguish XLH from the other renal Pi wasting disorders are its X-linked dominant mode of inheritance, its higher prevalence (1 in 20,000), and the availability of two murine homologs (see Section IV,F,2), which serve as valuable models to study the pathophysiology of XLH [103,160]. The gene responsible for XLH [106] was identified and designated PHEX (formerly PEX) to depict a phosphate-regulating gene

467 with homology to endopeptidases on the X chromosome (Table IV). However, the mechanism whereby loss of PHEX function leads to renal Pi wasting and hypophosphatemia is not understood. The PHEX gene encodes a 749-amino-acid protein [106] that exhibits significant homology to the M13 family of zinc metallopeptidases, which are type II membrane glycoproteins, with a large extracellular domain containing 10 highly conserved cysteine residues and a zinc-binding motif, essential for conformational integrity and catalytic activity, respectively [161]. The 171 mutations in the PHEX gene identified in XLH patients, which are cataloged in a locus-specific database [162] available online (www.phexdb.mcgill.ca), are predicted to result in loss of function of the PHEX protein. Some disease-causing missense mutations in the PHEX gene give rise to proteins that remain trapped in the endoplasmic reticulum, where they are degraded [163,164]. In some cases, however, the trapped PHEX protein can be rescued from the endoplasmic reticulum [163] and exhibit full catalytic activity [164]. These findings provide a mechanism for loss of PHEX function and a basis for the development of novel therapeutic approaches for patients carrying such trafficking mutations. PHEX is expressed predominantly in osteoblasts, osteocytes, and odontoblasts, but not in kidney (see [103]). Although it is not immediately apparent how loss of PHEX function results in a decrease in renal Pi reabsorption, it was suggested that PHEX is involved in the inactivation of phosphaturic factors or the activation of a Pi-conserving hormone [104,106,160,165]. In either case, renal Na/Pi cotransporters would be downregulated and Pi wasting would ensue. Although recombinant PHEX can cleave PTHrP107-139 [166] and peptides derived from MEPE (matrix extracellular phosphoglycoprotein) and FGF-23 [167], the precise function of these PHEX substrates in vivo and their role in the regulation of brush border membrane Na/Pi cotransporters and renal Pi handling remain to be established. Patients with XLH are treated with oral Pi supplements and 1,25(OH)2D [160]. Since this form of therapy is often associated with undesirable side effects, the aim of future studies is to develop novel therapeutic approaches based on knowledge of the mechanism of action of the PHEX gene product.

B. Autosomal Dominant Hypophosphatemic Rickets (ADHR) ADHR is far less common than XLH, it exhibits male-to-male transmission consistent with autosomal dominant inheritance, and it is characterized by incomplete penetrance and variable age of onset [160].

468 The gene responsible for ADHR encodes a new member of the fibroblast growth factor (FGF) family, FGF23, a secreted peptide that is processed to amino- and carboxy-terminal peptides at a consensus proprotein convertase (furin) site, RHTR (ArgHisThrArg) [168] (Table IV). Missense mutations involving the two R residues in this proteolytic cleavage site have been identified in patients with ADHR [168] and were shown to abrogate FGF-23 processing [169–171]. FGF-23 induces hypophosphatemia and renal Pi wasting in vivo [77,92]. Moreover, the increase in renal Pi excretion elicited by the hydrolysis-resistant mutant FGF-23 is associated with decreased brush border membrane Na/Pi cotransport and decreased brush border membrane abundance of Npt2a [77] and Npt2c [93] proteins (see Section IV,E,2). Furthermore, the phosphaturic action of hydrolysis-resistant mutant FGF-23 is significantly greater than that of wild-type FGF-23 [77], and the amino- and carboxy-terminal processed FGF-23 fragments do not induce hypophosphatemia in vivo [171]. Although the latter studies attribute the phosphaturic action of FGF-23 only to the intact protein, differences in the pharmacokinetics of the intact and processed FGF-23 peptides may have contributed to these findings [171]. Regardless, the findings to date provide a mechanism for hypophosphatemia and renal Pi wasting in ADHR. The mechanisms through which FGF-23 affects Pi homeostasis remain unclear, since conflicting data are reported regarding its action on Na/Pi cotransport in OK cells [92,160,169]. In addition, the site of FGF-23 production is unknown, although it is abundantly expressed in tumors removed from patients with OHO, an acquired renal Pi wasting disorder with features of XLH and ADHR [92,169,172] (see Section VII,C). Therapy for ADHR is similar to that of XLH (see Section VII,A). However, once the mechanisms of FGF-23 processing, action, and degradation are well understood, novel therapeutic approaches will likely be developed.

C. Oncogenic Hypophosphatemic Osteomalacia (OHO) Also known as tumor-induced osteomalacia, OHO is an acquired and rare form of renal Pi wasting, with clinical and biochemical features of XLH and ADHR (Table IV). Patients with OHO also exhibit muscle weakness, fatigue, and fractures. Although several types of tumors are associated with this syndrome, the majority appear to arise from mesenchymal elements [173]. Evidence suggests that the phenotypic features of OHO result from a humoral phosphaturic factor, designated

HARRIET S. TENENHOUSE AND ANTHONY A. PORTALE

phosphatonin [174], that is secreted by the tumor, since removal of the tumor leads to correction of the clinical and biochemical phenotype. Consistent with this hypothesis is the demonstration that extracts from OHO tumors inhibit Na/phosphate cotransport in OK cells [175]. The analysis of abundantly and differentially expressed genes in OHO tumors led to the identification of phosphatonin candidates, including FGF-23 [92,169,172,176–178], sFRP-4 [76,178] and MEPE [176,179]. As discussed earlier (Section VII,B), patients with ADHR harbor mutations in FGF-23 that prevent its processing and degradation [169–171], and both wildtype and mutant forms of FGF-23 induce hypophosphatemia and renal Pi wasting in vivo [77,92,171]. Moreover, sFRP-4 decreases serum Pi levels and inhibits renal Pi reabsorption in rats and mice [76]. Thus, FGF-23 and sFRP-4 likely contribute to the pathogenesis of OHO. However, the role of MEPE in this disorder remains to be determined. Further work is necessary to establish whether the OHO tumor factor is related to the circulating phosphaturic factor responsible for the renal Pi wasting in the Hyp mouse model of XLH (see Section IV,F,2). Although the obvious therapy for patients with OHO is surgical removal of the tumor, the tumors are small, present in obscure areas, difficult to locate without sophisticated imaging procedures, and often inaccessible. Treatment with Pi, in combination with 1,25(OH)2D, can be initiated and continued until the tumor is removed [160].

D. Does Renal Phosphate Wasting in XLH, ADHR and OHO Involve a Common Metabolic Pathway? Several findings have led to the intriguing hypothesis that defects in a common metabolic pathway can account for impaired renal Pi reabsorption and hypophosphatemia in XLH, ADHR, and OHO. First, FGF-23 mutations that prevent its processing to amino- and carboxy-terminal fragments, and presumably result in the accumulation of FGF-23, cause ADHR [168]. Second, wild-type and mutant FGF-23 induce renal Pi wasting and hypophosphatemia in vivo [77,92,171]. Third, serum FGF-23 levels are greatly increased in OHO patients and are normalized after tumor excision, a procedure that corrects the clinical and biochemical abnormalities of the disorder [180,181]. Fourth, serum FGF-23 levels are increased in some XLH patients [180,181] and are increased >10-fold in Hyp mice when compared to levels in wild-type littermates [182]. Fifth, a peptide derived from FGF-23 is a PHEX

469

CHAPTER 26 Phosphate Homeostasis

substrate [167]. Although the evidence that FGF-23 is a contributing factor for impaired renal Pi reabsorption and hypophosphatemia in ADHR and OHO is strong, it is more difficult to explain the mechanism for increased serum FGF-23 levels in XLH. It is possible that loss of PHEX function leads to the accumulation of an FGF-23 peptide, confirmed to be a PHEX substrate [167], which in turn acts as a feedback inhibitor of FGF-23 processing by proprotein convertase. Further work is necessary to establish the validity of these hypotheses, as well as to determine the contribution of other phosphaturic factors such as sFRP-4 to the underlying pathophysiology and the mechanisms involved in the down-regulation of renal brush border membrane Na/Pi cotransporters.

E. Hereditary Hypophosphatemic Rickets with Hypercalciuria (HHRH) HHRH can be distinguished from XLH, ADHR, and OHO by the appropriately increased serum concentration of 1,25(OH)2D, in response to hypophosphatemia, and the associated hypercalciuria [160,183,184] (Table IV). Since Pi supplementation alone corrects all the clinical and biochemical abnormalities in HHRH, with the exception of the renal Pi leak, it was suggested that HHRH is a primary renal Pi wasting disorder [184]. Given that the biochemical features of HHRH closely resemble those of mice homozygous for the disrupted Npt2a gene (see Section IV,F,1) and that Npt2a is an important determinant of Pi homeostasis and a target for its regulation [30], we hypothesized that mutations in the human ortholog, NPT2a, may be responsible for HHRH. To test this hypothesis, the NPT2a coding region and a fragment of the NPT2a promoter were screened for mutations [185]. No putative diseasecausing mutations in the NPT2a gene were found [185]. Two single nucleotide polymorphisms (SNPs), a silent substitution in exon 7 and a nucleotide substitution in intron 4, were identified and neither segregated with HHRH in a large Bedouin kindred [185]. Furthermore, linkage analysis demonstrated that the two SNPs, as well as five microsatellite markers flanking NPT2a in the chromosome 5q35 region, were not linked to HHRH in the Bedouin kindred [185]. These data exclude NPT2a as a candidate gene for HHRH and suggest that mutations in another renal Na/Pi cotransporter, or a regulator thereof, is responsible for the disorder. It is of interest that heterozygous mutations in the NPT2a gene have been identified in two individuals with renal Pi wasting and hypophosphatemia associated with either urolithiasis or bone demineralization [186]. Although the authors suggest that one copy of

the mutant NPT2a gene is sufficient for the expression of the respective phenotypic abnormalities, more recent functional studies of the mutant NPT2a cDNAs in Xenopus oocytes failed to provide evidence for altered kinetics or dominant negative effects [187]. Thus, the heterozygous mutations in NPT2a likely cannot explain the renal Pi wasting and hypophosphatemia in the patients studied [186]. In support of this hypothesis is the demonstration that mice heterozygous for the disrupted Npt2a gene fail to exhibit a defect in renal brush border membrane Na/Pi cotransport [97].

VIII. SUMMARY AND CONCLUSIONS Several Na/Pi cotransporters have been identified in intestinal, renal, and bone cells. The type IIa Na/Pi cotransporter (Npt2a) is expressed in the brush border membrane of proximal tubular cells, where the bulk of filtered Pi is reabsorbed, and plays a crucial role in the maintenance of Pi homeostasis. Npt2a is a target for regulation by PTH and dietary Pi, both of which induce changes in Npt2a protein abundance in the brush border membrane. Several proteins essential to membrane insertion, retrieval, and degradation of Npt2a have recently been described. In addition, two novel Pi-regulating genes, PHEX and FGF23, which are mutated in patients with Mendelian Pi wasting disorders, have been identified. Future studies are necessary to uncover additional Na/Pi cotransporters in the intestine, kidney, and bone that play an important role in determining Pi economy and to define the precise mechanism whereby PHEX and FGF-23 regulate Pi homeostasis.

References 1. Nordin BEC 1976 Nutritional considerations. In: Nordin BEC (ed) Calcium, Phosphate and Magnesium Metabolism. Churchill Livingstone, New York, pp. 1–35. 2. Marshall RW 1976 Plasma fractions. In: Nordin BEC (ed) Calcium, Phosphate and Magnesium Metabolism. Churchill Livingstone, New York, pp. 162–185. 3. Rowe J, Rowe D, Horak E, Spackman T, Saltzman R, Robinson S, Philipps A, Raye J 1984 Hypophosphatemia and hypercalciuria in small premature infants fed human milk: evidence for inadequate dietary phosphorus. J Pediatr 104:112–117. 4. Giles MM, Fenton MH, Shaw B, Elton RA, Clarke M, Lang M, Hume R 1987 Sequential calcium and phosphorus balance studies in preterm infants. J Pediatr 110:591–598. 5. Wilkinson R 1976 Absorption of calcium, phosphorus and magnesium. In: Nordin BEC (ed) Calcium, Phosphate and Magnesium Metabolism. Churchill Livingstone, New York, pp. 36–112.

470 6. Tanaka Y, DeLuca HF 1973 The control of 25-hydroxyvitamin D metabolism by inorganic phosphorus. Arch Biochem Biophys 154:566–574. 7. Baxter LA, DeLuca HF 1976 Stimulation of 25-hydroxyvitamin D3-1α-hydroxylase by phosphate depletion. J Biol Chem 251:3158–3161. 8. Gray RW, Napoli JL 1983 Dietary phosphate deprivation increases 1,25-dihydroxyvitamin D3 synthesis in rat kidney in vitro. J Biol Chem 258:1152–1155. 9. Markowitz M, Rotkin L, Rosen JF 1981 Circadian rhythms of blood minerals in humans. Science 213:672–674. 10. Portale AA, Halloran BP, Morris RC Jr 1987 Dietary intake of phosphorus modulates the circadian rhythm in serum concentration of phosphorus. Implications for the renal production of 1,25-dihydroxyvitamin D. J Clin Invest 80:1147–1154. 11. Suki WN, Rouse D 1996 Renal transport of calcium, magnesium, and phosphate. In: Suki WN, Rouse D (eds) The Kidney. W.B. Saunders, Philadelphia, pp. 472–509. 12. Brodehl J, Gellissen K, Weber HP 1982 Postnatal development of tubular phosphate reabsorption. Clin Nephrol 17:163–171. 13. Arnaud SB, Goldsmith RS, Stickler GB, McCall JT, Arnaud CD 1973 Serum parathyroid hormone and blood minerals: interrelationships in normal children. Pediatr Res 7:485–493. 14. Greenberg BR, Winters RW, Graham JB 1960 The normal ranges of serum inorganic phosphorus and its utility as a discriminant in the diagnosis of congenital hypophosphatemia. J Clin Endocrinol Metab 20:364. 15. Keating FR Jr, Jones JD, Elveback LR, Randall RV 1969 The relation of age and sex to distribution of values in healthy adults of serum calcium, inorganic phosphorus, magnesium, alkaline phosphatase, total proteins, albumin, and blood urea. J Lab Clin Med 73:825–834. 16. Cross HS, Debiec H, Peterlik M 1990 Mechanism and regulation of intestinal phosphate absorption. Miner Electrolyte Metab 16:115–124. 17. Schroder B, Hattenhauer O, Breves G 1998 Phosphate transport in pig proximal small intestines during postnatal development: Lack of modulation by calcitriol. Endocrinology 139:1500–1507. 18. Nemere I 1996 Apparent nonnuclear regulation of intestinal phosphate transport: Effects of 1,25-dihydroxyvitamin D3, 24,25-dihydroxyvitamin D3, and 25-hydroxyvitamin D3. Endocrinology 137:2254–2261. 19. Danisi G, Murer H 1991 Inorganic phosphate absorption in small intestine. In: Field M, Frizzel RA (eds) Handbook of Physiology—The Gastrointestinal System IV. Oxford University Press, New York, pp. 323–336. 20. Yagci A, Werner A, Murer H, Biber J 1992 Effect of rabbit duodenal mRNA on phosphate transport in Xenopus laevis oocytes: dependence on 1,25-dihydroxy-vitamin-D3. Pflügers Arch 422:211–216. 21. Norbis F, Boll M, Stange G, Markovich D, Verrey F, Biber J, Murer H 1997 Identification of a cDNA/protein leading to an increased Pi-uptake in Xenopus laevis oocytes. J Membrane Biol 156:19–24. 22. Hilfiker H, Hattenhauer O, Traebert M, Forster I, Murer H, Biber J 1998 Characterization of a murine type II sodiumphosphate cotransporter expressed in mammalian small intestine. Proc Natl Acad Sci USA 95:14564–14569. 23. Hattenhauer O, Traebert M, Murer H, Biber J 1999 Regulation of small intestinal Na–Pi type IIb cotransporter by dietary phosphate intake. Am J Physiol 277:G756–G762. 24. Xu H, Bai L, Collins JF, Ghishan FK 2002 Age-dependent regulation of rat intestinal type IIb sodium-phosphate

HARRIET S. TENENHOUSE AND ANTHONY A. PORTALE

25.

26.

27.

28. 29. 30. 31. 32. 33. 34. 35.

36.

37.

38.

39.

40.

41. 42.

cotransporter by 1,25-(OH)2 vitamin D3. Am J Physiol Cell Physiol 282:C487–C493. Xu H, Inouye M, Hines ER, Collins JF, Ghishan FK 2003 Transcriptional regulation of the human NaPi-IIb cotransporter by EGF in Caco-2 cells involves c-myb. Am J Physiol Cell Physiol 284:C1262–C1271. Arima K, Hines ER, Kiela PR, Drees JB, Collins JF, Ghishan FK 2002 Glucocorticoid regulation and glycosylation of mouse intestinal type IIb Na-Pi cotransporter during ontogeny. Am J Physiol Gastrointest. Liver Physiol 283: G426–G434. Berndt TJ, Knox FG 1992 Renal regulation of phosphate excretion. In: Seldin DW, Giebisch G (eds) The Kidney, Physiology and Pathophysiology. Raven Press, New York, pp. 2511–2532. Tenenhouse HS 1997 Cellular and molecular mechanisms of renal phosphate transport. J Bone Mineral Res 12: 159–164. Tenenhouse HS 1999 Recent advances in epithelial sodiumcoupled phosphate transport. Curr Opin Nephrol Hypertens 8:407–414. Murer H, Hernando N, Forster I, Biber J 2000 Proximal tubular phosphate reabsorption: molecular mechanisms. Physiol Rev 80:1373–1409. Murer H, Hernando N, Forster I, Biber J 2003 Regulation of Na/Pi transporter in the proximal tubule. Annu Rev Physiol 65:531–542. Tenenhouse HS, Murer H 2003 Disorders of renal tubular phosphate transport. J Am Soc Nephrol 14:240–247. Collins JF, Bai L, Ghishan FK 2004 The SLC20 family of proteins: dual functions as sodium-phosphate cotransporters and viral receptors. Pflügers Arch, 447:647–652. Murer H, Forster I, Biber J 2004 The sodium phosphate cotransporter family SLC34. Pflügers Arch, 763–767. Bijvoet O 1976 The importance of the kidney in phosphate homeostasis. In: Avioli L, Bordier P, Fleisch H, Mary S, Slatopolsky E (eds) Phosphate Metabolism, Kidney and Bone. Armour-Montagu, Paris, p. 421. Werner A, Moore ML, Mantei N, Biber J, Semenza G, Murer H 1991 Cloning and expression of cDNA for a Na/Pi cotransport system of kidney cortex. Proc Natl Acad Sci USA 88: 9608–9612. Magagnin S, Werner A, Markovich D, Sorribas V, Stange G, Biber J, Murer H 1993 Expression cloning of human and rat renal cortex Na/Pi cotransport. Proc Natl Acad Sci USA 90: 5979–5983. Tenenhouse HS, Roy S, Martel J, Gauthier C 1998 Differential expression, abundance and regulation of Na+phosphate cotransporter genes in murine kidney. Am J Physiol 44:F527–F534. Hartmann CM, Hewson AS, Kos CH, Hilfiker H, Soumoumou Y, Murer H, Tenenhouse HS 1996 Structure of murine and human renal type II Na+-phosphate cotransporter genes (Npt2 and NPT2). Proc Natl Acad Sci USA 93: 7409–7414. Soumounou Y, Gauthier C, Tenenhouse HS 2001 Murine and human type I Na-phosphate cotransporter genes: structure and promoter activity. Am J Physiol Renal Physiol 281: F1082–F1091. Segawa H, Kaneko I, Takahashi A, Kuwahata M, Ito M, Ohkido I, Tatsumi S, Miyamoto K 2002 Growth-related renal type II Na/Pi cotransporter. J Biol Chem 277:19665–19672. Ohkido I, Segawa H, Yanagida R, Nakamura M, Miyamoto K 2003 Cloning, gene structure and dietary regulation of the

471

CHAPTER 26 Phosphate Homeostasis

43.

44.

45. 46. 47. 48.

49.

50. 51.

52.

53.

54. 55. 56.

57. 58.

59.

type-IIc Na/Pi cotransporter in the mouse kidney. Pflügers Arch 446:106–115. Khadeer MA, Tang Z, Tenenhouse HS, Eiden MV, Murer H, Hernando N, Weinman EJ, Chellaiah MA, Gupta A 2003 Na+-dependent phosphate transporters in the murine osteoclast: Cellular distribution and protein interactions. Am J Physiol Cell Physiol 284:1633–1644. Kavanaugh MP, Miller DG, Zhang W, Law W, Kozak SL, Kabat D, Miller AD 1994 Cell-surface receptors for gibbon ape leukemia virus and amphotropic murine retrovirus are inducible sodium-phosphate symporters. Proc Natl Acad Sci USA 91:7071–7075. Forster IC, Loo DD, Eskandari S 1999 Stoichiometry and Na+ binding cooperativity of rat and flounder renal type II Na+–Pi cotransporters. Am J Physiol 276:F644–F649. Forster I, Hernando N, Biber J, Murer H 1998 The voltage dependence of a cloned mammalian renal type II Na+/Pi cotransporter (NaPi-2). J Gen Physiol 112:1–18. Lambert G, Traebert M, Hernando N, Biber J, Murer H 1999 Studies on the topology of the renal type II NaPi-cotransporter. Pflügers Arch Eur J Physiol 437:972–978. Lambert G, Forster IC, Stange G, Kohler K, Biber J, Murer H 2001 Cysteine mutagenesis reveals novel structure–function features within the predicted third extracellular loop of the type IIa Na+/Pi cotransporter. J Gen Physiol 117:533–546. Kohler K, Forster IC, Stange G, Biber J, Murer H 2002 Identification of functionally important sites in the first intracellular loop of the NaPi-IIa cotransporter. Am J Physiol 282:F687–F696. Hayes G, Busch A, Lotscher M, Waldegger S, Lang F, Verrey F, Biber J, Murer H 1994 Role of N-linked glycosylation in rat renal Na/Pi-cotransport. J Biol Chem 269:24143–24149. de la Horra C, Hernando N, Lambert G, Forster I, Biber J, Murer H 2000 Molecular determinants of pH sensitivity of the type IIa Na/Pi cotransporter. J Biol Chem 275: 6284–6287. de la Horra C, Hernando N, Forster I, Biber J, Murer H 2001 Amino acids involved in sodium interaction of murine type II Na+–Pi cotransporters expressed in Xenopus oocytes. J Physiol 531:383–391. Karim-Jimenez Z, Hernando N, Biber J, Murer H 2000 A dibasic motif involved in parathyroid hormone-induced down-regulation of the type IIa NaPi cotransporter. Proc Natl Acad Sci USA 97:12896–12901. Karim-Jimenez Z, Hernando N, Biber J, Murer H 2001 Molecular determinants for apical expression of the renal type IIa Na+/Pi-cotransporter. Pflügers Arch 442:782–790. Ehnes C, Forster IC, Kohler K, Biber J, Murer H 2002 Functional studies on a split type II Na/Pi-cotransporter. J Membrane Biol 188:227–236. Kohl B, Wagner CA, Huelseweh B, Busch AE, Werner A 1998 The Na+-phosphate cotransport system (NaPi-II) with a cleaved protein backbone: implications on function and membrane insertion. J Physiol 508:341–350. Kohler K, Forster IC, Lambert G, Biber J, Murer H 2000 The functional unit of the renal type IIa Na+/Pi cotransporter is a monomer. J Biol Chem 275:26113–26120. Levi M, Lotscher M, Sorribas V, Custer M, Arar M, Kaissling B, Murer H, Biber J 1994 Cellular mechanisms of acute and chronic adaptation of rat renal phosphate transporter to alterations in dietary phosphate. Am J Physiol 267: F900–F908. Pfister MF, Hilfiker H, Forgo J, Lederer E, Biber J, Murer H 1998 Cellular mechanisms involved in the acute adaptation

60.

61.

62. 63. 64.

65. 66.

67.

68.

69. 70. 71.

72. 73.

74.

75.

of OK cell Na/Pi-cotransport to high- or low-Pi medium. Pflügers Arch 435:713–719. Kempson SA, Lotscher M, Kaissling B, Biber J, Murer H, Levi M 1995 Parathyroid hormone action on phosphate transporter mRNA and protein in rat renal proximal tubules. Am J Physiol 268:F784–F791. Lotscher M, Scarpetta Y, Levi M, Halaihel N, Wang H, Zajicek HK, Biber J, Murer H, Kaissling B 1999 Rapid downregulation of rat renal Na/Pi cotransporter in response to parathyroid hormone involves microtubule rearrangement. J Clin Invest 104:483–494. Caverzasio J, Bonjour JP 1993 Growth factors and renal regulation of phosphate transport. Pediatr Nephrol 7: 802–806. Abraham MI, McAteer JA, Kempson SA 1990 Insulin stimulates phosphate transport in opossum kidney epithelial cells. Am J Physiol 258:F1592–F1598. Taketani Y, Segawa H, Chikamori M, Morita K, Tanaka K, Kido S, Yamamoto H, Iemori Y, Tatsumi S, Tsugawa N, Okano T, Kobayashi T, Miyamoto K, Takeda E 1998 Regulation of type II renal Na+-dependent inorganic phosphate transporters by 1,25-dihydroxyvitamin D3. Identification of a vitamin D–responsive element in the human NaPi-3 gene. J Biol Chem 273:14575–14581. Sorribas V, Markovich D, Verri T, Biber J, Murer H 1995 Thyroid hormone stimulation of Na/Pi cotransport in opossum kidney cells. Pflügers Arch 431:266–271. Alcalde AI, Sarasa M, Raldua D, Aramayona J, Morales R, Biber J, Murer H, Levi M, Sorribas V 1999 Role of thyroid hormone in regulation of renal phosphate transport in young and aged rats. Endocrinology 140:1544–1551. Pizurki L, Rizzoli R, Moseley J, Martin TJ, Caverzasio J, Bonjour JP 1988 Effect of synthetic tumoral PTH-related peptide on cAMP production and Na-dependent Pi transport. Am J Physiol 255:F957–F961. Nakai M, Fukase M, Kinoshita Y, Fujita T 1988 Atrial natriuretic factor inhibits phosphate uptake in opossum kidney cells: as a model of renal proximal tubules. Biochem Biophys Res Commun 152:1416–1420. Arar M, Baum M, Biber J, Murer H, Levi M 1995 Epidermal growth factor inhibits Na-Pi cotransport and mRNA in OK cells. Am J Physiol 268:F309–F314. Arar M, Zajicek HK, Elshihabi I, Levi M 1999 Epidermal growth factor inhibits Na–Pi cotransport in weaned and suckling rats. Am J Physiol 276:F72–F78. Pizurki L, Rizzoli R, Caverzasio J, Bonjour JP 1990 Effect of transforming growth factor-α and parathyroid hormonerelated protein on phosphate transport in renal cells. Am J Physiol 259:F929–F935. Law F, Rizzoli R, Bonjour JP 1993 Transforming growth factor-β inhibits phosphate transport in renal epithelial cells. Am J Physiol 264:F623–F628. Levi M, Shayman JA, Abe A, Gross SK, McCluer RH, Biber J, Murer H, Lotscher M, Cronin RE 1995 Dexamethasone modulates rat renal brush border membrane phosphate transporter mRNA and protein abundance and glycosphingolipid composition. J Clin Invest 96:207–216. Wagner GF, Vozzolo BL, Jaworski E, Haddad M, Kline RL, Olsen HS, Rosen CA, Davison MB, Renfro JL 1997 Human stanniocalcin inhibits renal phosphate excretion in the rat. J Bone Miner Res 12:165–171. Saito H, Kusano K, Kinosaki M, Ito H, Hirata M, Segawa H, Miyamoto K, Fukushima N 2003 Human fibroblast growth factor-23 mutants suppress Na+-dependent phosphate

472

76.

77.

78.

79. 80. 81.

82.

83.

84.

85. 86.

87.

88.

89.

90. 91.

HARRIET S. TENENHOUSE AND ANTHONY A. PORTALE

co-transport activity and 1α,25-dihydroxyvitamin D3 production. J Biol Chem 278:2206–2211. Berndt T, Craig TA, Bowe AE, Vassiliadis J, Reczek D, Finnegan R, Jan De Beur S, Schiavi SC, Kumar R 2003 Secreted frizzled-related protein 4 is a potent tumor-derived phosphaturic agent. J Clin Invest, in press. Bai XY, Miao D, Goltzman D, Karaplis AC 2003 The autosomal dominant hypophosphatemic rickets R176Q mutation in FGF23 resists proteolytic cleavage and enhances in vivo biological potency. J Biol Chem 278:9843–9849. Traebert M, Lotscher M, Aschwanden R, Ritthaler T, Biber J, Murer H, Kaissling B 1999 Distribution of the sodium/phosphate transporter during postnatal ontogeny of the rat kidney. J Am Soc Nephrol 10:1407–1415. Gisler SM, Stagljar I, Traebert M, Bacic D, Biber J, Murer H 2001 Interaction of the type IIa Na/Pi cotransporter with PDZ proteins. J Biol Chem 276:9206–9213. Biber J 2001 Emerging roles of transporter–PDZ complexes in renal proximal tubular reabsorption. Pflügers Arch 443:3–5. Pfister MK, Ruf I, Stange G, Ziegler U, Lederer E, Biber J 1998 Parathyroid hormone leads to the lysosomal degradation of the renal type II Na/Pi cotransporter. Proc Natl Acad Sci USA 95:1909–1914. Traebert M, Volkl H, Biber J, Murer H, Kaissling B 2000 Luminal and contraluminal action of 1-34 and 3-34 PTH peptides on renal type IIa Na–Pi cotransporter. Am J Physiol 278:F792–F798. Lederer ED, Sohi SS, McLeish KR 2000 Parathyroid hormone stimulates extracellular signal-regulated kinase (ERK) activity through two independent signal transduction pathways: Role of ERK in sodium-phosphate cotransport. J Am Soc Nephrol 11:222–231. Bacic D, Schulz N, Biber J, Kaissling B, Murer H, Wagner CA 2003 Involvement of the MAPK-kinase pathway in the PTH-mediated regulation of the proximal tubule type IIa Na+/Pi cotransporter in mouse kidney. Pflügers Arch 446:52–60. Jankowski M, Hilfiker H, Biber J, Murer H 2001 The opossum kidney cell type IIa Na/Pi cotransporter is a phosphoprotein. Kidney Blood Press Res 24:1–4. Khundmiri SJ, Rane MJ, Lederer ED 2003 Parathyroid hormone regulation of type II sodium-phosphate cotransporters is dependent on an A kinase anchoring protein. J Biol Chem 278:10134–10141. Bacic D, Capuano P, Gisler SM, Pribanic S, Christensen EI, Biber J, Loffing J, Kaissling B, Wagner CA, Murer H 2003 Impaired PTH-induced endocytotic down-regulation of the renal type IIa Na+/Pi-cotransporter in RAP-deficient mice with reduced megalin expression. Pflügers Arch 446: 475–484. Pribanic S, Gisler SM, Bacic D, Madjdpour C, Hernando N, Sorribas V, Gantenbein A, Biber J, Murer H 2003 Interactions of MAP17 with the NaPi-IIa/PDZK1 protein complex in renal proximal tubular cells. Am J Physiol Renal Physiol. Lederer ED, Khundmiri SJ, Weinman EJ 2003 Role of NHERF-1 in regulation of the activity of Na-K ATPase and sodium-phosphate co-transport in epithelial cells. J Am Soc Nephrol 14:1711–1719. Hilfiker H, Hartmann CM, Stange G, Murer H 1998 Characterization of the 5′-flanking region of OK cell type II Na–Pi cotransporter gene. Am J Physiol 274:F197–F204. Hafdi Z, Couette S, Comoy E, Prie D, Amiel C, Friedlander G 1996 Locally formed 5-hydroxytryptamine stimulates

92.

93.

94.

95.

96.

97.

98.

99.

100.

101. 102.

103. 104.

105.

106.

phosphate transport in cultured opossum kidney cells and in rat kidney. Biochem J 320:615–621. Shimada T, Mizutani S, Muto T, Yoneya T, Hino R, Takeda S, Takeuchi Y, Fujita T, Fukumoto S, Yamashita T 2001 Cloning and characterization of FGF23 as a causative factor of tumor-induced osteomalacia. Proc Natl Acad Sci USA 98: 6500–6505. Segawa H, Kawakami E, Kaneko I, Kuwahata M, Ito M, Kusano K, Saito H, Fukushima N, Miyamoto KI 2003 Effect of hydrolysis-resistant FGF23-R179Q on dietary phosphate regulation of the renal type-II Na/Pi transporter. Pflügers Arch 446:585–592. Yamashita T, Konishi M, Miyake A, Inui K, Itoh N 2002 Fibroblast growth factor (FGF)-23 Inhibits renal phosphate reabsorption by activation of the mitogen-activated protein kinase pathway. J Biol Chem 277:28265–28270. Lotscher M, Kaissling B, Biber J, Murer H, Levi M 1997 Role of microtubules in the rapid regulation of renal phosphate transport in response to acute alterations in dietary phosphate content. J Clin Invest 99:1302–1312. Kido S, Miyamoto K, Mizobuchi H, Taketani Y, Ohkido I, Ogawa N, Kaneko, Harashima S, Takeda E 1999 Identification of regulatory sequences and binding proteins in the type II sodium/phosphate cotransporter NPT2 gene responsive to dietary phosphate. J Biol Chem 274:28256–28263. Beck L, Karaplis AC, Amizuka N, Hewson AS, Ozawa H, Tenenhouse HS 1998 Targeted inactivation of Npt2 in mice leads to severe renal phosphate wasting, hypercalciuria and skeletal abnormalities. Proc Natl Acad Sci USA 95: 5372–5377. Tenenhouse HS, Martel J, Gauthier C, Zhang MYH, Portale AA 2001 Renal expression of the sodium/phosphate cotransporter gene, Npt2; is not required for regulation of renal 1α-hydroxylase by phosphate. Endocrinology 142: 1124–1129. Gupta A, Tenenhouse HS, Hoag HM, Wang D, Khadeer MA, Namba N, Feng X, Hruska KA 2001 Identification of the type II Na+–Pi cotransporter (Npt2) in the osteoclast and the skeletal phenotype of Npt2−/− mice. Bone 29:467–476. Hoag HM, Martel J, Gauthier C, Tenenhouse HS 1999 Effects of Npt2 gene ablation and low-phosphate diet on renal Na/ phosphate cotransport and cotransporter gene expression. J Clin Invest 104:679–686. Zhao N, Tenenhouse HS 2000 Npt2 gene disruption confers resistance to the inhibitory action of PTH on renal Na-phosphate cotransport. Endocrinology 141:2159–2165. Tenenhouse HS, Martel J, Gauthier C, Segawa H, Miyamoto K 2003 Differential effects of Npt2a gene ablation and the X-linked Hyp mutation on renal expression of type IIc Na/Pi cotransporter. Am J Physiol, 285:F1277–F1278. Tenenhouse HS 1999 X-linked hypophosphatemia: A homologous disorder in humans and mice. Nephrol Dial Transplant 14:333–341. Beck L, Soumounou Y, Martel J, Krishnamurthy G, Gauthier C, Goodyer C, Tenenhouse HS 1997 Pex/PEX tissue distribution and evidence for a deletion in the 3′ region of the Pex gene in X-linked hypophosphatemic mice. J Clin Invest 99: 1200–1209. Strom TM, Francis F, Lorenz B, Boeddrich A, Econs M, Lehrach H, Meitinger T 1997 Pex gene deletions in Gy and Hyp mice provide mouse models for X-linked hypophosphatemia. Hum Mol Genet 6:165–171. HYP Consortium 1995 A gene (PEX) with homologies to endopeptidases is mutated in patients with X-linked hypophosphatemic rickets. Nat Genet 11:130–136.

CHAPTER 26 Phosphate Homeostasis

107. Tenenhouse HS, Klugerman AH, Neal JL 1989 Effect of phosphonoformic acid, dietary phosphate and the Hyp mutation on kinetically distinct phosphate transport processes in mouse kidney. Biochim Biophys Acta 984:207–213. 108. Tenenhouse HS, Meyer RA Jr, Mandla S, Meyer MH, Gray RW 1992 Renal phosphate transport and vitamin D metabolism in X-linked hypophosphatemic Gy mice: Responses to phosphate deprivation. Endocrinology 131:51–56. 109. Tenenhouse HS, Werner A, Biber J, Ma S, Martel J, Roy S, Murer H 1994 Renal Na+-phosphate cotransport in murine X-linked hypophosphatemic rickets: Molecular characterization. J Clin Invest 93:671–676. 110. Beck L, Meyer RA Jr, Meyer MH, Biber J, Murer H, Tenenhouse HS 1996 Renal expression of Na+-phosphate cotransporter mRNA and protein: Effect of the Gy mutation and low phosphate diet. Pflügers Arch 431:936–941. 111. Tenenhouse HS, Beck L 1996 Renal Na+-phosphate cotransporter gene expression in X-linked Hyp and Gy mice. Kidney Int 49:1027–1032. 112. Meyer RA Jr, Meyer MH, Gray RW 1989 Parabiosis suggests a humoral factor is involved in X-linked hypophosphatemia in mice. J Bone Miner Res 4:493–500. 113. Meyer RA Jr, Tenenhouse HS, Meyer MH, Klugerman AH 1989 The renal phosphate transport defect in normal mice parabiosed to X-linked hypophosphatemic mice persists after parathyroidectomy. J Bone Miner Res 4:523–532. 114. Nesbitt T, Coffman TM, Griffiths R, Drezner MK 1992 Cross-transplantation of kidneys in normal and Hyp mice: Evidence that the Hyp phenotype is unrelated to an intrinsic renal defect. J Clin Invest 89:1453–1459. 115. Lajeunesse D, Meyer RA Jr, Hamal L 1996 Direct evidence of a humorally-mediated inhibition of renal phosphate transport in the Hyp mouse: involvement of an osteoblast-derived factor. Kidney Int 50:1531–1538. 116. Carpinelli MR, Wicks IP, Sims NA, O’Donnell K, Hanzinikolas K, Burt R, Foote SJ, Bahlo M, Alexander WS, Hilton DJ 2002 An ethyl-nitrosourea-induced point mutation in phex causes exon skipping, x-linked hypophosphatemia, and rickets. Am J Pathol 161:1925–1933. 117. Hughes MR, Brumbaugh PF, Hussler MR, Wergedal JE, Baylink DJ 1975 Regulation of serum 1α,25-dihydroxyvitamin D3 by calcium and phosphate in the rat. Science 190:578–580. 118. Gray RW 1981 Control of plasma 1,25-(OH)2-vitamin D concentrations by calcium and phosphorus in the rat: effects of hypophysectomy. Calcif Tissue Int 33:485–488. 119. Lobaugh B, Drezner MK 1983 Abnormal regulation of renal 25-hydroxyvitamin D-1α-hydroxylase activity in the X-linked hypophosphatemic mouse. J Clin Invest 71:400–403. 120. Gray RW, Wilz DR, Caldas AE, Lemann J Jr 1977 The importance of phosphate in regulating plasma 1,25-(OH)2vitamin D levels in humans: studies in healthy subjects in calcium-stone formers and in patients with primary hyperparathyroidism. J Clin Endocrinol Metab 45:299–306. 121. Insogna KL, Broadus AL, Gertner JM 1983 Impaired phosphorus conservation and 1,25-dihydroxyvitamin D generation during phosphorus deprivation in familial hypophosphatemic rickets. J Clin Invest 71:1562–1569. 122. Maierhofer WJ, Gray RW, Lemann J Jr 1984 Phosphate deprivation increases serum 1,25-(OH)2-vitamin D concentrations in healthy men. Kidney Int. 25:571–575. 123. Portale AA, Halloran BP, Murphy MM, Morris RC Jr 1986 Oral intake of phosphorus can determine the serum concentration of 1,25-dihydroxyvitamin D by determining its production rate in humans. J Clin Invest 77:7–12.

473 124. Portale AA, Halloran BP, Morris RC Jr 1989 Physiologic regulation of the serum concentration of 1,25-dihydroxyvitamin D by phosphorus in normal men. J Clin Invest 83:1494–1499. 125. Gray RW 1981 Control of plasma 1,25-(OH)2-vitamin D concentrations by calcium and phosphorus in the rat: Effects of hypophysectomy. Calcif Tissue Int 33:485–488. 126. Gray RW, Garthwaite TL, Phillips LS 1983 Growth hormone and triiodothyronine permit an increase in plasma 1,25(OH)2D concentrations in response to dietary phosphate deprivation in hypophysectomized rats. Calcif Tissue Int 35: 100–106. 127. Gray RW, Garthwaite TL 1985 Activation of renal 1,25-dihydroxyvitamin D3 synthesis by phosphate deprivation: Evidence for a role for growth hormone. Endocrinology 116:189–193. 128. Halloran BP, Spencer EM 1988 Dietary phosphorus and 1,25-dihydroxyvitamin D metabolism: Influence of insulinlike growth factor I. Endocrinology 123:1225–1229. 129. Grieff M, Zhong M, Finch J, Ritter CS, Slatopolsky E, Brown AJ 1996 Renal calcitriol synthesis and serum phosphorus in response to dietary phosphorus restriction and anabolic agents. Am J Kidney Dis 28:589–595. 130. Zhang MY, Wang X, Wang JT, Compagnone NA, Mellon SH, Olson JL, Tenenhouse HS, Miller WL, Portale AA 2002 Dietary phosphorus transcriptionally regulates 25-hydroxyvitamin D-1α-hydroxylase gene expression in the proximal renal tubule. Endocrinology 143:587–595. 131. Yoshida T, Yoshida N, Monkawa T, Hayashi M, Saruta T 2001 Dietary phosphorus deprivation induces 25-hydroxyvitamin D3 1α-hydroxylase gene expression. Endocrinology 142:1720–1726. 132. Wu S, Grieff M, Brown AJ 1997 Regulation of renal vitamin D24-hydroxylase by phosphate: effects of hypophysectomy, growth hormone and insulin-like growth factor I. Biochem Biophys Res Commun 233:813–817. 133. Wu S, Finch J, Zhong M, Slatopolsky E, Grieff M, Brown AJ 1996 Expression of the renal 25-hydroxyvitamin D-24hydroxylase gene: regulation by dietary phosphate. Am J Physiol 271:F203–F208. 134. Scriver CR, Reade TM, DeLuca HF, Hamstra AJ 1978 Serum 1,25-dihydroxyvitamin D levels in normal subjects and in patients with hereditary rickets or bone disease. New Engl J Med 299:976–979. 135. Delvin EE, Glorieux FH 1981 Serum 1,25-dihydroxyvitamin D concentration in hypophosphatemic vitamin D–resistant rickets. Calcif Tissue Int 33:173–175. 136. Meyer RA Jr, Gray RW, Meyer MH 1980 Abnormal vitamin D metabolism in the X-linked hypophosphatemic mouse. Endocrinology 107:1577–1581. 137. Tenenhouse HS, Jones G 1990 Abnormal regulation of renal vitamin D catabolism by dietary phosphate in murine X-linked hypophosphatemic rickets. J Clin Invest 85:1450–1455. 138. Yamaoka K, Seino Y, Satomura K, Tanaka Y, Yabuuchi H, Haussler MR 1986 Abnormal relationship between serum phosphate concentration and renal 25-hydroxycholecalciferol-1α-hydroxylase activity in X-linked hypophosphatemic mice. Miner Electrol Metab 12:194–198. 139. Azam N, Zhang MY, Wang X, Tenenhouse HS, Portale AA 2003 Disordered regulation of renal 25-hydroxyvitamin D-1α-hydroxylase gene expression by phosphorus in X-linked hypophosphatemic (hyp) mice. Endocrinology 144: 3463–3468. 140. Roy S, Martel J, Ma S, Tenenhouse HS 1994 Increased renal 25-hydroxyvitamin D3-24-hydroxylase messenger ribonucleic

474

141.

142. 143. 144.

145. 146.

147.

148. 149.

150.

151.

152.

153.

154.

155.

156.

HARRIET S. TENENHOUSE AND ANTHONY A. PORTALE

acid and immunoreactive protein in phosphate-deprived Hyp mice: A mechanism for accelerated 1,25-dihydroxyvitamin D3 catabolism in X-linked hypophosphatemic rickets. Endocrinology 134:1761–1767. Roy S, Tenenhouse HS 1996 Transcriptional regulation and renal localization of 1,25-dihydroxyvitamin D3-24-hydroxylase gene expression: Effects of the Hyp mutation and 1,25dihydroxyvitamin D3. Endocrinology 137:2938–2946. Caverzasio J, Bonjour JP 1996 Characteristics and regulation of Pi transport in osteogenic cells for bone metabolism. Kidney Int 49:975–980. Caverzasio J, Selz T, Bonjour JP 1988 Characteristics of phosphate transport in osteoblastlike cells. Calcif Tissue Int 43:83–87. Montessuit C, Bonjour JP, Caverzasio J 1995 Expression and regulation of Na-dependent Pi transport in matrix vesicles produced by osteoblast-like cells. J Bone Miner Res 10: 625–631. Selz T, Caverzasio J, Bonjour JP 1989 Regulation of Nadependent Pi transport by parathyroid hormone in osteoblastlike cells. Am J Physiol 256:E93–100. Pizurki L, Rizzoli R, Caverzasio J, Bonjour JP 1991 Stimulation by parathyroid hormone-related protein and transforming growth factor-alpha of phosphate transport in osteoblast-like cells. J Bone Miner Res 6:1235–1241. Palmer G, Bonjour JP, Caverzasio J 1997 Expression of a newly identified phosphate transporter/retrovirus receptor in human SaOS-2 osteoblast-like cells and its regulation by insulin-like growth factor I. Endocrinology 138: 5202–5209. Veldman CM, Schlapfer I, Schmid C 1997 1α,25-dihydroxyvitamin D3 stimulates sodium-dependent phosphate transport in osteoblast-like cells. Bone 21:41–47. Zhen X, Bonjour JP, Caverzasio J 1997 Platelet-derived growth factor stimulates sodium-dependent Pi transport in osteoblastic cells via phospholipase Cgamma and phosphatidylinositol 3′ kinase. J Bone Miner Res 12:36–44. Suzuki A, Palmer G, Bonjour JP, Caverzasio J 2000 Stimulation of sodium-dependent phosphate transport and signaling mechanisms induced by basic fibroblast growth factor in MC3T3-E1 osteoblast-like cells. J Bone Miner Res 15:95–102. Suzuki A, Palmer G, Bonjour JP, Caverzasio J 2001 Stimulation of sodium-dependent inorganic phosphate transport by activation of Gi/o-protein-coupled receptors by epinephrine in MC3T3-E1 osteoblast-like cells. Bone 28:589–594. Masukawa H, Miura Y, Sato I, Oiso Y, Suzuki A 2001 Stimulatory effect of endothelin-1 on Na-dependent phosphate transport and its signaling mechanism in osteoblast-like cells. J Cell Biochem 83:47–55. Veldman CM, Schlapfer I, Schmid C 1998 Prostaglandin E2 stimulates sodium-dependent phosphate transport in osteoblastic cells via a protein kinase C–mediated pathway. Endocrinology 139:89–94. Burgener D, Bonjour JP, Caverzasio J 1995 Fluoride increases tyrosine kinase activity in osteoblast-like cells: regulatory role for the stimulation of cell proliferation and Pi transport across the plasma membrane. J Bone Miner Res 10:164–171. Nielsen LB, Pedersen FS, Pedersen L 2001 Expression of type III sodium-dependent phosphate transporters/retroviral receptors mRNAs during osteoblast differentiation. Bone 28:160–166. Gupta A, Miyauchi A, Fujimori A, Hruska KA 1996 Phosphate transport in osteoclasts: A functional and immunochemical characterization. Kidney Int 49:968–974.

157. Gupta A, Guo X, Alvarez UM, Hruska KA 1997 Regulation of sodium-dependent phosphate transport in osteoclasts. J Clin Invest 100:538–549. 158. Hodgson SF, Hurley DL 1993 Acquired hypophosphatemia. Endocrinol Metab Clin North Am 22:397–409. 159. Hruska KA, Lederer ED 1999 Hyperphosphatemia and hypophosphatemia. In: Favus MJ (ed) Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism. Lippincott Williams & Wilkins, Philadelphia, pp. 245–253. 160. Tenenhouse HS, Econs MJ 2001 Mendelian hypophosphatemias. In: Scriver CR, Beaudet AL, Sly WS, Valle D (eds) The Metabolic and Molecular Bases of Inherited Disease. McGraw-Hill, New York, pp. 5039–5067. 161. Turner AJ, Tanzawa K 1997 Mammalian membrane metallopeptidases: NEP, ECE, KELL and PEX. FASEB J 11:355–364. 162. Sabbagh Y, Jones AO, Tenenhouse HS 2000 PHEXdb, a locus-specific database for mutations causing X-linked hypophosphatemia. Hum Mut 16:1–6. 163. Sabbagh Y, Boileau G, DesGroseillers L, Tenenhouse HS 2001 Disease-causing missense mutation in the PHEX gene interfere with membrane targeting of the recombinant protein. Hum Mol Genet 10:1539–1546. 164. Sabbagh Y, Boileau G, Campos M, Carmona AK, Tenenhouse HS 2003 Structure and function of diseasecausing missense mutations in the PHEX gene. J Clin Endocrinol Metab 88:2213–2222. 165. Econs MJ, Francis F 1997 Positional cloning of the PEX gene: new insights into the pathophysiology of X-linked hypophosphatemic rickets. Am J Physiol 273:F489–F498. 166. Boileau G, Tenenhouse HS, DesGroseillers L, Crine P 2001 Characterization of PHEX endopeptidase catalytic activity: identification of parathyroid-hormone-related peptide 107-139 as a substrate and osteocalcin, PPi and phosphate as inhibitors. Biochem J 355:707–713. 167. Campos M, Couture C, Hirata IY, Juliano MA, Loisel TP, Crine P, Juliano L, Boileau G, Carmona AK 2003 Human recombinant PHEX has a strict S1′ specificity for acidic residues and cleaves peptides derived from FGF-23 and MEPE. Biochem J 373:271–279. 168. ADHR Consortium 2000 Autosomal dominant hypophosphataemic rickets is associated with mutations in FGF23. Nat Genet 26:345–348. 169. Bowe AE, Finnegan R, Jan de Beur SM, Cho J, Levine MA, Kumar R, Schiavi SC 2001 FGF-23 inhibits renal tubular phosphate transport and is a phex substrate. Biochem Biophys Res Commun 284:977–981. 170. White KE, Carn G, Lorenz-Depiereux B, Benet-Pages A, Strom TM, Econs MJ 2001 Autosomal-dominant hypophosphatemic rickets (ADHR) mutations stabilize FGF-23. Kidney Int. 60:2079–2086. 171. Shimada T, Muto T, Urakawa I, Yoneya T, Yamazaki Y, Okawa K, Takeuchi Y, Fujita T, Fukumoto S, Yamashita T 2002 Mutant FGF-23 responsible for autosomal dominant hypophosphatemic rickets is resistant to proteolytic cleavage and causes hypophosphatemia in vivo. Endocrinology 143:3179–3182. 172. White KE, Jonsson KB, Carn G, Hampson G, Spector TD, Mannstadt M, Lorenz-Depiereux B, Miyauchi A, Yang IM, Ljunggren O, Meitinger T, Strom TM, Juppner H, Econs MJ 2001 The autosomal dominant hypophosphatemic rickets (ADHR) gene is a secreted polypeptide overexpressed by tumors that cause phosphate wasting. J Clin Endocrinol Metab 86:497–500. 173. Kumar R 2000 Tumor-induced osteomalacia and the regulation of phosphate homeostasis. Bone 27:333–338.

475

CHAPTER 26 Phosphate Homeostasis

174. Econs MJ, Drezner MK 1994 Tumor-induced osteomalacia— unveiling a new hormone. N Engl J Med 330:1679–1681. 175. Jonsson KB, Mannstadt M, Miyauchi A, Yang IM, Stein G, Ljunggren O, Juppner H 2001 Extracts from tumors causing oncogenic osteomalacia inhibit phosphate uptake in opossum kidney cells. J Endocrinol 169:613–620. 176. Jan de Beur SM, Finnegan RB, Vassiliadis J, Cook B, Barberio D, Estes S, Manavalan P, Petroziello J, Madden SL, Cho JY, Kumar R, Levine MA, Schiavi SC 2002 Tumors associated with oncogenic osteomalacia express genes important in bone and mineral metabolism. J Bone Miner Res 17:1102–1110. 177. Fukumoto S, Yamashita T 2002 Fibroblast growth factor-23 is the phosphaturic factor intumor-induced osteomalacia and may be phosphatonin. Curr Opin Nephrol Hypertens 11:385–389. 178. Schiavi SC, Moe OW 2002 Phosphatnins: a new class of phosphate-regulating proteins. Curr Opin Nephrol Hypertens 11:423–430. 179. Rowe PS, de Zoysa PA, Dong R, Wang HR, White KE, Econs MJ, Oudet CL 2000 MEPE, a new gene expressed in bone marrow and tumors causing osteomalacia. Genomics 67:54–68. 180. Yamazaki Y, Okazaki R, Shibata M, Hasegawa Y, Satoh K, Tajima T, Takeuchi Y, Fujita T, Nakahara K, Yamashita T, Fukumoto S 2002 Increased circulatory level of biologically active full-length FGF-23 in patients with hypophosphatemic rickets/osteomalacia. J Clin Endocrinol Metab 87:4957–4960. 181. Jonsson KB, Zahradnik R, Larsson T, White KE, Sugimoto T, Imanishi Y, Yamamoto T, Hampson G, Koshiyama H, Ljunggren O, Oba K, Yang M, Miyauchi A, Econs M, Lavigne J, Jueppner H 2002 Fibroblast growth factor 23 in oncogenic osteomalacia and X-linked hypophosphatemia. N Engl J Med 348:1656–1663. 182. Yamazaki Y, Shimada T, Imai R, Hino R, Aono Y, Murakami J, Fukumoto S, Yamashita T 2003 Elevated circulatory and

183. 184.

185.

186.

187.

188.

189. 190.

expression level of fibroblast growth factor (FGF)-23 in hypophosphatemic mice. Bone 32, S88. Tieder M, Modai D, Samuel R, Arie R, Halabe A, Bab I, Gabizon D, Lieberman UA 1985 Hereditary hypophosphatemic rickets with hypercalciuria. N Engl J Med 312:611–617. Tieder M, Modai D, Shaked U, Samuel R, Arie R, Halabe A, Maor J, Weissgarten J, Averbukh Z, Cohen N, Edelstein S, Lieberman UA 1987 “Idiopathic” hypercalciuria and hereditary hypophosphatemic rickets. N Engl J Med 316:125–129. Jones AO, Tzenova J, Frappier D, Crumley M, Roslin NM, Kos CH, Tieder M, Langman CB, Proesmans W, Carpenter TO, Rice A, Anderson D, Morgan K, Fujiwara TM, Tenenhouse HS 2001 Hereditary hypophosphatemic rickets with hypercalciuria is not caused by mutations in the Na/Pi cotransporter NPT2 gene. J Am Soc Nephrol 12:507–514. Prie D, Huart V, Bakouh N, Planelles G, Dellis O, Gerard B, Hulin P, Benque-Blanchet F, Silve C, Grandchamp B, Friedlander G 2002 Nephrolithiasis and osteoporosis associated with hypophosphatemia caused by mutations in the type IIa sodium-phosphate cotransporter. N Engl J Med 347:983–991. Virkki LV, Forster IC, Hernando N, Biber J, Murer H 2003 Functional characterization of two naturally occurring mutations in the human sodium-phosphate cotransporter type IIa. J Bone Miner Res, 18:2135–2141. Kumar R 2002 New insights into phosphate homeostasis: fibroblast growth factor 23 and frizzled-related protein-4 are phosphaturic factors derived from tumors associated with osteomalacia. Curr Opin Nephrol Hypertens 11:547–553. Portale AA 1999 Calcium and phosphorus. In: Barrat TM, Avner ED, Harmon WE (eds) Pediatric Nephrology. Lippincott Williams and Wilkins, Baltimore, pp. 191–213. Murer H, Forster I, Hernando N, Lambert G, Traebert M, Biber J 1999 Posttranscriptional regulation of the proximal tubule NaPi-II transporter in response to PTH and dietary [Pi]. Am J Physiol 277:F676–F684.

CHAPTER 27

Mineralization DAN FAIBISH AND ADELE L. BOSKEY Hospital for Special Surgery, New York, New York, 10021; Affiliated with Weil College of Cornell Medical School, New York, New York

I. Introduction II. Mineralization and Mineral Properties in Model Systems with Vitamin D Alterations

III. Conclusions References

I. INTRODUCTION

Most biomineralization is epitaxial or heterogeneous— occurring on the surfaces of preexisting crystals or on protein and lipid templates that resemble the surface of the crystal [6]. These processes use much less energy than de novo mineralization, and require lower supersaturation. Crystal growth occurs as ions and ion clusters add on to the surface of the initial nuclei or other pre-existing crystals. Crystal growth requires less energy than nucleation and is limited in the case of biologic mineralization by the template upon which the crystals are deposited. Crystals can grow in all dimensions by the addition of ions [6]; by agglomeration [7], in which crystals accumulate, not always in an oriented fashion; or by secondary nucleation. Secondary nucleation, as is seen with hydroxyapatite crystals maturing in solutions in the presence of the dentin protein phosphophoryn [8], is a branching process in which new nuclei form on the surfaces of existing crystals, thus resulting in a new population of immature crystals. Among the vertebrate tissues, enamel contains the largest apatite crystals, but these crystals do contain carbonate [9] and other environmental contaminants (e.g., strontium, fluoride, lead [10]). Bone and tendon– bone insertions contain the smallest apatite mineral crystals [11], 6–10% carbonate [12], as well as adsorbed and incorporated citrate, fluoride, and other trace impurities. Bone apatite crystals are also hydroxide deficient [13], but are not totally devoid of hydroxyl groups [14]. Cementum and dentin mineral crystals have intermediate sizes and intermediate accumulation of foreign ions. In each of these tissues the crystals that are formed initially are smaller than those in the mature tissues because of both the growth of existing crystals and remodeling by osteoclasts [5]. In humans, mineral also deposits at abnormal (unexpected/dystrophic) locations [15]. This dystrophic/ pathologic mineral is frequently apatitic, but calcite (calcium carbonate) is found in pancreatic stones [16]; monosodium urate and sodium pyrophosphate, as well as apatite [15,17], can be deposited in cartilage; and

A. Definitions “Biologic mineralization” is the physicochemical process leading to deposition of inorganic crystals (minerals) on organic matrix within the cell or outside it. This term is more specific than “mineralization,” as biologic mineralization implies a relation between the organic matrix and the mineral. The cell-mediated biomineralization process describes events through which minerals are deposited within cells in an oriented fashion (for example, iron oxides and sulfides in magnetotactic bacteria [1]; silicates in diatoms [2]) or upon an extracellular matrix (for example, calcium carbonates in shells [3] and exoskeletons [4]) or calcium phosphates in bones and teeth [5]. The mineral in physiologically calcified vertebrate tissues is an analog of the geologic mineral, hydroxyapatite (Fig. 1). The physiologic hydroxyapatite crystals are much smaller than those found in geologic deposits and have stoichiometries different from the predicted 10Ca:6PO4 :2OH of the geologic mineral. For that reason biologic vertebrate mineral is often referred to as “apatite” or “apatitic,” meaning “like hydroxyapatite.” This chapter will focus on physiologic and dystrophic apatite formation in situ and in culture and how vitamin D affects the formation of this mineral. Crystalline deposits may form by several different mechanisms. De novo crystal deposition occurs when the solution supersaturation exceeds the solubility of the precipitating phase. Supersaturation refers to the ratio of the solution ion product to the solubility product of the phase in question. The process starts with “nucleation,” in which several ions or ion clusters come together in solution with the orientation they will have in the final crystal. This first step requires a great deal of energy and is facilitated by increasing the ion product, lowering the diffusion of ions in solution, etc. [5]. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

478

DAN FAIBISH AND ADELE L. BOSKEY

Ca10(PO4)6(OH)2

Na+ Mg++

CO3− −

F−

Sr++ K++ vacancies

FIGURE 1 The composition of the hydroxyapatite unit cell showing the major substituents occurring in biologic apatites and the ions for which they substitute.

brushite, oxalates, and uric acid occur in kidney and salivary stones (Table I). The dystrophic calcifications seen in arteries induced by vitamin D toxicity [18,19], those seen in the presence of hyperphosphatemia [20,21], and those seen in muscle calcifications [22], among others, are all apatitic. In some cases the tissue becomes bone-like and genes associated with osteogenesis are activated [23], as in vascular calcification.

B. Direct and Indirect Effects of Vitamin D and Vitamin D Metabolites on Mineralization 1. PHYSICAL CHEMISTRY OF MINERALIZATION

De novo apatite formation requires Ca2+, PO43−, and OH− ions to come together in the correct orientation with sufficient energy and in sufficient numbers to

TABLE I Dystrophic Mineral Deposits Mineral phase Apatite

Calcium carbonate (aragonite) Oxalates Pyrophosphates Urates

Found in

Affected by vitamin D

Blood vessels Kidney and bladder stones Salivary stones Muscle Skin Hyaline and articular cartilage Pancreatic stones

+ + + + + + 0

Kidney and salivary stones Soft tissues Hyaline and articular cartilage Hyaline and articular cartilage Kidney stones

0 0 0 0

+, Deposition in these tissues is accelerated in hypervitaminosis-D, 0, no effect.

form the first stable apatite crystal (nucleus). After this nucleus is formed, additional ions can add on to these small crystals (nuclei) causing crystal growth. As crystals become larger, new nuclei can branch off the surface (secondary nucleation), in a fashion analogous to glycogen formation. These new nuclei grow and form additional secondary nuclei in an exponential fashion. During biologic apatite formation, matrix molecules provide sites for accumulation of ions and templates to orient mineral growth. Some molecules may act as heterogeneous nucleators, facilitating the deposition of mineral. Others may bind to the crystals and regulate their shape and size [25]. Nucleation occurs at multiple sites along these templates, and crystal habit (shape) and size is regulated by the template and other matrix proteins that bind to the surface of the apatite crystals. In cartilage, bone, dentin, tendon, and ligaments, mineral crystals form on a collagenous matrix. This collagen is the “template” for initial mineral formation and crystal growth [24–26] and mineralization of the collagen matrix gives it increased flexibility and strength [27,28]. Differences in the distribution and properties of other extracellular matrix molecules associated with the fibrillar collagen influence the crystal size and the site of initial crystal deposition. In enamel, which does not have a collagen component, the enamelins are believed to act as nucleators [29], while amelogenin is thought to regulate crystal size and shape [30]. Tables II and III list matrix components from solution, cell culture, or animal models reviewed elsewhere [5] that have been shown to act as apatite nucleators and regulators of apatite crystal growth. Those proteins whose expression is known to be regulated by 1,25(OH)2 vitamin D3 are also indicated. It should be noted that the effects of vitamin D in regulating the expression of matrix molecules important for calcification are indirect effects. Cells control the mineralization process, both by regulation of calcium and phosphate efflux [31,32] and by production of collagen and noncollagenous proteins which guide and direct mineral deposition. There is also a physicochemical process by which supersaturated solutions lead sometimes to mineral deposition in unwanted sites. One example of this physicochemical effect is the hypervitaminosis D syndrome [18,19], where injections of vitamin D into animals cause elevations of circulating calcium and result in arterial and kidney calcification. 2. THE NATURE OF VERTEBRATE MINERAL

The mineral that forms in physiologically mineralized tissues—calcified cartilage, bone, dentin, and enamel, is an analogue of the geologic mineral, hydroxyapatite. The chemical formula for this mineral is Ca10(PO4)6(OH)2,

479

CHAPTER 27 Mineralization

TABLE II

Extracellular Matrix Proteins Associated with Mineralization and Regulated by Vitamin D: Solution Data Effect on mineralization

Ref.

Regulated by vitamin Da

Ref.

Template Inhibitor Inhibitor Inhibitor/nucleator Nucleator Weak nucleator/inhibitor Nucleator Nucleator Regulator of shape Inhibitor Inhibitor Nucleator Nucleator Inhibitor Inhibitor Hydrolyzes phosphate esters/phosphotransferase Hydrolyzes ATP, facilitates cellular Ca transport Degrade matrix molecules that inhibit mineralization

[24] [104] [91,110] [24] [97] [24] [8] [29] [30] [24] [142] [143] [144] [144] [24] [24] [145] [24]

+ + + ? + + + + + + + + ? ? + + + +

[92–95] [102,103] [110]

Protein Type I collagen Osteocalcin Osteopontin Osteonectin Bone sialoprotein (BSP) Dentin sialoprotein Dentin phosphophoryn Amelogenin Enamelin Matrix gla protein (MGP) Fetuin Proteolipid Biglycan Decorin Aggrecan Alkaline phosphatase ATPase Matrix metalloproteinases

[98–101] [150] [150] [150] [150] [151] [152] [117–119]

[153] [154] [145] [146–148]

a Regulation by vitamin D depends on concentration, cell type, and cell maturity; hence + indicates that there is an effect. Readers are referred to other chapters to see precise effects. ?, Unknown.

as there are two structural units in the smallest repeating component of the crystal structure (the unit cell). With the exception of enamel, the other physiologic mineral is deposited in an oriented fashion on a collagen template. Unlike geologic hydroxyapatite crystals

TABLE III Protein Type I collagen Osteocalcin Osteopontin Osteonectin Amelogenin Biglycan Alkaline phosphatase (bone specific) MMPs

which are quite large and visible to the naked eye, the physiologic mineral crystals are microscopic in size being <200Å in their longest dimension in bone, dentin, and related tissues, and ~400Å in the longest dimension in enamel. Because of the small crystal size (some

Extracellular Matrix Proteins Associated with Mineralization and Regulated by Vitamin D: Mineral Properties in Knockout, Transgenic, and Mutant Animals Model(s)

Phenotype

Osteogenesis imperfecta (mutant, knockout, knockin) Knockout Knockout Knockout Transgenic knockout Knockout Knockout

Brittle bones; smaller crystals Increased bone mineral content; larger crystals Increased bone mineral content, larger crystals Increased collagen maturity, larger crystals Defective enamel mineralization Decreased bone mineral content, larger crystals Poorly mineralized bone

Knockouts

Varies depending on enzyme and site

Ref. [154] [107] [111] [155] [156,157] [158] [159] [160]

480 crystals contain only a few unit cells), the physiologic mineral crystals contain a large number of surface impurities and vacancies. Bone apatite is rich in carbonate, which tends to substitute for phosphate in the lattice [33], and is deficient in hydroxide [13,14]; hence it is often referred to as “apatite.” Depending on diet and environment the bone apatite may contain magnesium, sodium, fluoride, strontium, potassium, acid phosphate, and citrate substituents (Fig. 1). The presence of such imperfections in bone, cementum, calcified cartilage, and similarly in the slightly larger dentin crystals in general reduces the crystallinity (crystal size and perfection) of the physiologic mineral deposits and tends to make the crystals more soluble than geologic hydroxyapatite [34,35]. This solubility is essential both to enable remodeling and to facilitate mineral homeostasis. The mechanisms of this homeostasis are highly dependent on vitamin D metabolites as reviewed in Section III of this book. Enamel apatite crystals are larger than bone and dentin mineral and tend to include different impurities. Enamel carbonate content is lower and carbonate is substituted mainly for hydroxide ions [33]. Fluoride is a frequent substituent, as it is included in dentifrices as a caries prophylactic, both to inhibit bacterial carbohydrate metabolism and to decrease enamel mineral solubility. This decrease in solubility is related to the hydrogen bond–induced stabilization of the apatite crystal structure [36]. 3. MINERALIZATION MECHANISMS IN BONE CARTILAGE

AND

In general, body fluids are supersaturated with respect to hydroxyapatite. In other words, the ion product [Ca]10 × [PO4]6 × [OH]2 exceeds the hydroxyapatite solubility product of 10−58 [37]. Yet biologic apatite only deposits physiologically in the so-called mineralized tissues. This is because the body fluids and tissues contain numerous mineralization inhibitors. These inhibitors prevent de novo precipitation, protect the cells from becoming engulfed in mineral, and also regulate the size and shape of the crystals that do form. Mineralization inhibitors include anionic molecules that can chelate calcium, e.g., citrate, ATP, pyrophosphate, glycosaminoglycans, and proteins that bind specifically to apatite crystals, e.g., osteonectin, fetuin, and albumin [24]. The deposition of mineral in bones and teeth occurs at specific sites where the barriers to crystal deposition are diminished either by elevating Ca × PO4 concentrations or by exposing matrix molecules or structures that facilitate mineral deposition. Mineral deposition can be facilitated when mineral crystals are already

DAN FAIBISH AND ADELE L. BOSKEY

present, in which case there is growth and proliferation of the initial crystals, when the surface structure of the matrix matches that of the precipitating phase, or when inhibitors of mineralization are removed [24]. Extracellular matrix vesicles [38], the site of initial mineral deposition in calcifying cartilage and mantle dentin (the first site of dentin mineral deposition), provide protected sites for the accumulation of calcium and phosphate ions [24,39]. Their membranes are also rich in enzymes such as ATPase, pyrophosphatase, alkaline phosphatase, and matrix metalloproteinases [38], enzymes that are required to disrupt mineralization inhibitors in the extracellular matrix [39]. However, the majority of physiologic mineral is believed not to be formed by a vesicle-mediated process. Most mineral forms at multiple discrete locations on type I collagen fibrils [40]. Collagen itself does not support de novo apatite formation; instead, specific anionic noncollagenous proteins can bind to and stabilize the initially formed crystals and regulate the mineralization process [24]. These noncollagenous proteins associate with the collagen fibrils, facilitating initiation of mineralization on the collagen template. In solutions and in cell culture, several of these proteins have been proven to be nucleators of apatite (Table II). These in vitro observations have been validated in animals in which these proteins are ablated (knocked out) or overexpressed (Table III). One of the challenges associated with deciphering the mechanism of biologic calcification is determining which proteins are absolutely essential for initiation of mineralization, and then finding the order in which these proteins are expressed.

C. Methods for Quantifying Tissue Mineralization Several important questions must be addressed when examining a mineralized/mineralizing tissue: Is the tissue mineralized? How much mineral is there? Is the mineral characteristic of physiologic mineral? (Is it a poorly crystalline apatite? Is it properly aligned with the collagen matrix? Is the composition of the mineral different from that in an age- and background-matched control tissue? What size are the crystals?) There are multiple techniques that can be used to address these questions. These techniques are illustrated by examples from vitamin D–related studies. 1. IS THE TISSUE MINERALIZED?

Radiographic methods [plain films, fine-focus radiographs, peripheral quantitative computerized tomography (pQCT), or even magnetic resonance

481

CHAPTER 27 Mineralization

imaging (MRI)] all show changes in opacity when mineral is present. For example, plain films of rachitic (vitamin D–deficient) growing animals show enlarged epiphyses and absence of mineral. Plain X-ray films are routinely used to reveal the presence of increased osteoid and bowing, characteristic of vitamin D–deficient osteomalacia. Vitamin D toxicity is similarly recognized radiographically as increased density in animals and man. More quantitative methods such as fine-focus radiographs, microcomputerized tomography (µCT), pQCT, and dual photon absorptiometry (DEXA) can provide more detail on bone mineral content and bone density. Analysis by µCT provides bone structural data including relative bone density, morphometric parameters such as trabecular number, trabecular thickness, bone volume fraction, and connectivity in three dimensions, in good but not perfect agreement with parameters measured by histomorphometric means [41]. Although there are not yet reports of µCT used to study vitamin D effects, increased availability of µCT should increase its application to metabolic bone diseases in animal models. Peripheral quantitative computerized tomography (pQCT) is related to µCT, enabling measurement of both cortical and trabecular density in peripheral sites. However, where µCT is limited to relatively small samples, whole bodies can be imaged by pQCT, although morphometric parameters cannot be determined. This technique has been applied to demonstrate bone loss in models of vitamin D–stimulated remodeling [42]. Vitamin D and calcium–deficient ovariectomized sheep characterized by quantitative computerized tomography (QCT) and mechanical testing showed decreased density and reduced mechanical strength when contrasted with ovariectomized sheep given a diet rich in calcium and vitamin D [43]. It is important to point out that in a large clinical study of elderly women, pQCT failed to find 1,25(OH)2 vitamin D3 or 25-OH vitamin D3 serum levels as predictors of bone mineral density [44]. 2. HOW MUCH MINERAL IS THERE?

The standard method for quantifying the amount of mineral in a tissue is gravimetric measurement of the weight of the residue left after the tissue is dried (110°C) and the organic components removed by ashing at 600°C. The ash weight, or ash density, if expressed per volume, reveals the total amount of mineral in the tissue. Chemical analysis of Ca and PO4 content has also been used to calculate mineral content, but because of the high organic phosphate content of these tissues these numbers may be inaccurate. Recently, infrared and Raman spectroscopic analysis of mineral to matrix

ratio has been shown to correlate linearly with ash weight of synthetic mixtures [45,46]. There are differences in mineral content of a given bone based on quantitative back scattered election imaging [47] and quantitative density fractionation of ground bone [48]. Nevertheless, quantitative and spatial information was first provided by the spectroscopic imaging techniques described hereafter. 3. IS THE MINERAL CHARACTERISTIC PHYSIOLOGIC MINERAL (IS IT A POORLY CRYSTALLINE APATITE)?

OF

The gold standard for determining the presence of bone mineral (or any mineral phase) is X-ray diffraction (XRD). The X-rays impingent on a powdered sample are reflected at specific angles characteristic of the spacing between lattice planes, and hence characteristic of the mineral phase in question (Fig. 2A). While the small crystal size and numerous imperfections in the physiologic apatite crystals make their diffraction patterns quite broad, they are still easily recognizable as apatite. Additionally, the broadening of individual peaks due to the small size and imperfections in a particular dimension can be used to calculate an average crystallite size and perfection. Electron microscopy (EM) shows the size and shape of the mineral crystals and selected area electron diffraction, analogous to X-ray diffraction, giving a characteristic pattern. The advantage of EM is that physiologic mineral crystals can be observed oriented parallel to the collagen fibrils, whereas dystrophic mineral may not always have that appearance [49]. Both XRD and electron diffraction have proven useful for characterizing mineral associated with matrices and matrix vesicles developed in culture [50]. The amount of mineral present can be quantified by neither X-ray diffraction nor electron diffraction. This can be achieved by quantitative analysis [chemical determination by microprobe or other chemical analysis] or by ashing the specimen (burning off the organic phase) to measure the gravimetric yield of mineral. Energy dispersive X-ray analysis (EDAX) can provide information on the chemical composition of the surface. Coupled to scanning electron microscopes, EDAX enables observation of the morphology of the mineral deposit and simultaneous determination of Ca:PO4 ratios of the mineral [51]. However, the precise determination of this ratio requires comparison with standards, a procedure not always followed. Information on mineral particle thickness and alignment can be obtained by scanning small-angle X-ray scattering (scanning-SAXS). The data obtained agree with the size predictions from infrared techniques [52]. As will be detailed later, scanning SAXS has been

482

DAN FAIBISH AND ADELE L. BOSKEY

A

211

300 112

002

202 210

102

0 20

22

24

26

28

30

32

34

36

Degrees 2θ

B

Phosphate

Amide I Amide II Carbonate

1800

1600

1400 Wave numbers

1200

1000

800

(cm−1)

FIGURE 2 (A) X-ray diffraction pattern of synthetic hydroxyapatite showing the major peaks used for mineral analysis. The broadening of the 002 (c-axis) peak (arrow) is routinely used to estimate crystallite size and perfection. (B) Infrared spectrum of bone mineral showing the bands characteristic of apatite and protein.

used to characterize the bone mineral in the transgenic mouse that overexpresses the vitamin D receptor [53]. Atomic force microscopy (AFM) [54,55] enables examination of the surface structure of a variety of materials and can be used to image individual bone crystals. This high-resolution technique depends on the use of a probe to visualize individual crystals. However, the organic matrix must be removed from the tissue for these measurements, and this is a difficult challenge that might cause breakdown of some crystals. To date there have been no AFM analyses of any of the bones or soft tissues of animals with vitamin D abnormalities.

The vibrations of atoms in molecules are determined both by the bonds in which they are located and by the environment surrounding these moieties. Vibrations that affect the dipole moment and those that are symmetrical, detected by infrared and Raman techniques, respectively, can be used both to identify molecular species and for quantitative analysis of mineral composition and content. As in X-ray and electron diffraction where the poorly mineralized biologic apatite has broad diffraction peaks, the IR and Raman spectra characteristic of physiologic apatite are also broadened (Fig. 2B). Analysis of the relative phosphate to protein peak area ratios (v1/amide I in

483

CHAPTER 27 Mineralization

Raman; v3/amide I in IR [46]) is used to calculate mineral content. Curve fitting or chemometric analysis of specific subbands of the compound spectral features can be used to reveal mineral crystal size and perfection (crystallinity [52,56–58]), acid phosphate content [56,57], and carbonate substitution [59,12]. Magnetic resonance imaging (MRI) and nuclear magnetic resonance (NMR) methods can also reveal the distribution of acid phosphate and features of the mineral crystal structure. Solid-state NMR spectroscopy has long been used for 1H, 13C, and 31P studies of bone mineral and calcium phosphates [60]. Magic angle spinning (MAS) and MAS coupled with cross polarization (CP-MAS) have been used to characterize initial mineral and matrix composition in developing bones [61–63]. Although there are no published studies of mineral characteristics observed by these techniques in the presence and absence of vitamin D, the high correlation between these techniques, X-ray diffraction, and spectroscopic and chemical analyses [64] indicates that NMR data would support the findings from XRD and IR described later. Quantitative backscattered electron imaging (qBEI) can be used to measure the amount of mineral in a localized region of biopsied bone. Gray-scale images are produced, and each pixel in the image is compared to osteoid and hydroxyapatite standards, from which calcium weight percent values are calculated. Data is compared based on pixel distributions of calcium peak weight percent (CaPeak). This technique has been used to describe the mineral distribution in the vitamin D receptor (VDR) knockout mice discussed later [65]. 4. METHODS FOR STUDYING IN VITRO MINERALIZATION

Studies in culture (in the presence of tissues or cells) have been extensively used to assess the function of vitamin D metabolites and the vitamin D receptor in the mineralization process. Although it is popular to use 10 mM β-glycerophosphate as a substrate [e.g., 66–68], as will be discussed later, 5–10 mM organic phosphate can cause ectopic mineral deposition, even in the absence of cells or matrix, as long as alkaline phosphatase activity is present [49,69]. Mineral can be characterized in the culture system by measuring changes in solution calcium or phosphate concentrations (often by the use of radiolabeled ions) and the amount of mineral in the matrix (chemical analysis, diffraction, or spectroscopic methods). More recently it has become very popular to quantify calcium and phosphate accumulation in culture by the use of histochemical stains. Sometimes these values are quantified by extracting the stained culture. One of the problems with this histochemical approach is that these stains

are often not specific for mineral. For example, von Kossa staining for phosphate [51,66] can equally label phosphate-containing matrix molecules, especially cell membranes [70]. Alizarin red, used to stain for calcium [51,66], can show the presence of this ion in calcium phosphates, but also calcium bound to proteoglycans. Thus conclusive proof of the presence of mineral in a culture experiment comes from diffraction methods, electron microscopy, or spectroscopic techniques as discussed earlier.

D. Direct and Indirect Effects of Vitamin D on Mineralization 1. PHYSICOCHEMICAL EFFECTS

Vitamin D’s “direct” effect is to increase serum Ca2+ levels by stimulating osteoclastic resorption, kidney Ca retention, and intestinal Ca absorption (see Section III of this book). This in turn leads to an elevation in local calcium concentration, and often an elevation in phosphate. The increased circulating levels of calcium and phosphate, though not causing marked hypercalcemia or hyperphosphatemia (because of the counter effects of parathyroid hormone), are thought to be sufficient to raise the serum levels enough to promote dystrophic calcification, especially in blood vessels [19,71] and kidneys [72]. Injection of vitamin D3 or 1,25-(OH)2 vitamin D3 into animals can cause formation of apatite deposits in articular and growth cartilage as well as bone [73]. There is also a case report showing that milk accidentally fortified with excessive vitamin D caused dental pulp calcification in a child [74], showing that mineralizing tissues other than bone or cartilage can be affected. The vitamin D–receptor knockout mouse has an impaired bone formation phenotype (see later discussion) that can be rescued by calcium treatment [75]. Similarly, vitamin D–deficient rickets in animals can be cured by increasing serum calcium (by lactose and calcium infusion) without addition of vitamin D [76,77]. This suggests that the increased mineralization associated with vitamin D may simply reflect increased calcium or calcium and phosphate concentrations. However, the mineral quality has not been examined in any of these calcium-treated models. From the perspective of physical chemistry, to precipitate a basic calcium phosphate such as bone apatite, the solution needs to have sufficient pH, and sufficient calcium and phosphate concentrations. In solution, at pH7.4, when the Ca × PO4 mM2 product exceeds 5.5 mM2, precipitation will occur even in the absence of a nucleator or a template for mineral deposition [78]. Thus, raising the local calcium or local phosphate concentration, even slightly, can cause

484 precipitation, all other factors being equivalent. This is extremely important when considering cell culture studies where ion concentrations are elevated beyond physiologic levels. Physiologic phosphate concentrations are 0.9–1.5 mM in most species [79]. β-Glycerophosphate, often used in mineralizing cultures and said to enhance osteoblastic differentiation, is a substrate for alkaline phosphatase. When 10 mM β-glycerophosphate is used in culture, the phosphate concentrations of the medium can be as high as 10 mM. With physiological calcium concentrations of 1.5-2.5mM, the Ca × PO4 concentration of the solution exceeds 5.5 mM2 and hence mineral will deposit in the absence of a matrix. As shown by Khouja et al. [69], the occurrence of this dystrophic mineralization simply demonstrates the presence of alkaline phosphatase activity, not of physiologic mineralization. Physiologic calcium concentrations are ~2.1–2.5 mM. Raising the concentration to 2.9 or 4.2 mM [67] with phosphate at even 2 mM will give a Ca × PO4 product of 5.8 mM2 or 8.4 mM2, greater than the value needed for de novo precipitation. Although there is altered gene expression caused by the presence of these elevated calcium concentrations [67], it has not yet been demonstrated that this altered gene expression is due to the presence of the calcium, rather than to the presence of the dystrophic mineral that surrounds the cells. We know that gene expression is altered when mineral surrounds the osteocyte [80], and in this case the mineral surrounding the osteocyte is physiologic mineral. Thus caution must be exercised when looking at gene expression in such cases. In situ there are more controls on calcium concentrations; thus the elevations in calcium in tissues of animals treated with vitamin D may not directly cause precipitation, but may increase the ion product sufficiently that initial mineralization is facilitated, as is crystal growth. In vitamin D–deficient animals (with rickets and/or osteomalacia) the mineral content of the bulk tissue, reflected by histochemical stains or ash weights, is decreased relative to controls (Fig. 3). The crystals present in the epiphysis and metaphysis, i.e., newly formed mineral crystals, tend to be larger than those in age- and gender-matched controls. It is not certain whether this is because osteoclast activity is increased or because in the absence of an appropriate matrix, existing crystals grow at the expense of new crystals being formed. Increased osteoclast activity has been noted in some cases of human hypophosphatemic oncogenic osteomalacia [81,82] and in vitamin D–deficient rats [83], hens [84], and pigs [85]. However, the effects of vitamin D metabolites on extracellular matrix formation and composition suggest that impaired formation may be a contributing factor.

DAN FAIBISH AND ADELE L. BOSKEY

2. EFFECTS OF VITAMIN D ON CELLS AND MATRIX MOLECULES

a. Introduction As reviewed elsewhere in this volume, vitamin D is a differentiation factor, facilitating the expression of mature chondrocyte and mature osteoblast activity, as well as osteoclastogenesis. Vitamin D alters the gene expression and protein synthesis of numerous matrix proteins which are regulators of biomineralization (Table II). This regulation occurs through both genomic and nongenomic pathways and is modulated by calcium influx. The vitamin D–regulated proteins include, but are not limited to, calcium-binding proteins and several enamel (e.g., enamelins, amelogenins), bone, and dentin proteins [e.g., dentin sialoglycoprotein (DSP) and dentin phosphoprotein (DPP)]. Vitamin D also increases expression of membrane-bound enzymes such as alkaline phosphatase and the kinases and phosphatases [86] that regulate the phosphorylation and dephosphorylation of intracellular and extracellular matrix proteins. The effects of vitamin D metabolites on cell differentiation are reviewed elsewhere in Section I of this book; however, it is important to note that in vitro 24,25-(OH)2 vitamin D stimulates chick chondrocyte [87] and rabbit osteoblast maturation [88], while 1,25(OH)2 vitamin D stimulates rat osteoblast differentiation in bone marrow stromal cells [89]. Since it is the mature chondrocyte and osteoblast that synthesize the mineralizable matrix and/or provide the enzymes necessary to modulate that matrix to facilitate mineralization, one indirect effect of vitamin D might be on the formation of the proper cells to allow mineralization. In cultured chick hypertrophic chondrocytes, for example, 1,25-(OH)2 vitamin D3 causes a dose-dependent decrease in activity of alkaline phosphatase, while 24,25(OH)2 vitamin D3 stimulates this activity [90]. The effect of 1,25(OH)2 vitamin D3 in stimulating kinase and phosphatase activity, and thereby regulating the extent of phosphorylation of osteopontin, is extremely interesting in light of data showing that the moderately phosphorylated bone osteopontin is an effective mineralization inhibitor [91]. In contrast, the highly phosphorylated milk osteopontin can induce hydroxyapatite formation in solution, and stabilize precursors of hydroxyapatite [Boskey, unpublished]. The activities of several of the other matrix proteins depend on their state of phosphorylation, stressing the importance of these enzymes. b. The Influence of Vitamin D on Bone Matrix Proteins Vitamin D modulates the production of several molecules (Table II) that are crucial for the synthesis and function of bone tissue, although there is no clear evidence that direct effects of vitamin D are required

485

CHAPTER 27 Mineralization

A

D-deficient

D-deficient

Control

Control

100

1.00

12

80

0.80

60

0.60

6 3 Ca

B

20

0.20

0

PO4

Control

0.40

Hyp

0.80 β002

80 Ash Weight

1.00

0.40

20

0.20 epiphysis metaphysis diaphysis

Control

D-deficient

80

0.80

60

0.60

40

0.40

20

0.20 metaphysis diaphysis

calvaria

Hyp

1.60 1.20 0.80 0.40 epiphysis metaphysis diaphysis

D-deficient

Control

D-deficient

2.00 1.60 1.20 0.80 0.40 0.00

0.00

0

diaphysis

0.00

Control 1.00

β002

Ash Weight

Control

epiphysis metaphysis diaphysis

100

epiphysis

D-deficient

2.00

0.00

0

C

Hyp

0.60

40

0.00

diaphysis

Control

100

60

epiphysis

Ca/PO4 (molar ratio)

0

40

Ca/PO4 (molar ratio)

9

β002

15 Ash Weight

Serum Chemistry

Control

metaphysis

diaphysis

metaphysis diaphysis calvaria

FIGURE 3 Mineral analyses in models of vitamin D deficiency. (A) Vitamin D–deficient rats. Male Sprague-Dawley rats (21 days old) were maintained in the dark on a vitamin-D and phosphate-deficient diet for 3 weeks. The first panel shows their serum Ca and PO4 levels. 1,25-(OH)2vitamin D levels were not detectable in the D-deficient animals. Tibias removed at sacrifice were used for analysis of ash weight and β002 (1/crystallinity). Reproduced with permission from Springer-Verlag [119]. (B) Female mice with hypo-phosphatemic rickets (Hyp) and their age- and sex-matched controls were 35 days old at sacrifice. Femora were used for analysis of ash weight, β002, and Ca/PO4 ratio of the ash. Reprinted with permission from Elsevier [117]. (C) Second-generation vitamin D–deficient rats and their age-matched controls were 7 weeks old at sacrifice. Femora were used for analysis of ash weight and β002 and Ca/PO4 ratio of the ash. Reprinted with permission from Elsevier [116].

for normal bone mineralization. The actions of these proteins have been described in detail elsewhere [24]; hence only a few representative proteins are described in this section. Readers are referred to references in Tables II and III for more detail. i. Collagen The scaffold of bone matrix is made predominantly of type I collagen molecules. These molecules are produced early in bone tissue formation, and their special morphology supports the mineral deposition in later stages. In addition, the collagen

molecules provide the elasticity characteristic of bone tissue, which is essential for its normal function in response to mechanical forces. The synthesis of collagen is a complex process, and it is regulated by several factors, one of them being vitamin D. The active metabolite, 1,25(OH)2 vitamin D3 downregulates the transcription of α1(I) collagen by osteoblasts in a variety of cells [92–95]. This decrease in type I collagen expression appears to be cell stage dependent, occurring in early bone nodules formed in

486

DAN FAIBISH AND ADELE L. BOSKEY

culture, but not in intermediate and mature ones [94]. Adding 1,25(OH)2 vitamin D3 after the completion of extracellular matrix formation has been reported to inhibit further mineralization in chick osteoblast cultures [95], suggesting that 1,25(OH)2 vitamin D3 affects the metabolic processes by which osteoblasts control mineralization, independent of its effects on the formation of the collagenous matrix. However, Matsumoto showed that 1,25(OH)2 vitamin D3 stimulated alkaline phosphatase activity in both early and late osteoblast cultures and reported enhanced mineral deposition based on EM and 45Ca uptake when 10−10 M 1,25(OH)2D3 was added after the start of mineralization [96]. Similarly (Fig. 4), in differentiating chick limb-bud mesenchymal cell micro-mass cultures, 1,25(OH)2D3 added continuously to cultures after the matrix had been formed and mineralization was started (day 9) increased the rate of mineral accumulation compared to mineralizing cultures that did not receive exogenous vitamin D. This increase was in contrast to cultures to which vitamin D was added continuously while cells were differentiating (day 5) or when the cells were first beginning to deposit a cartilage matrix (day 7).

Vitamin D also regulates the expression of the noncollagenous proteins in mineralized tissues (Table II). Many of these proteins are believed to play direct roles in the regulation of biomineralization [5,24], and thus they account in part for the indirect effect of vitamin D on mineralization. ii. Noncollagenous proteins The expression of bone sialoprotein (BSP), an in vitro apatite nucleator [97] and mineralization regulator [24], is suppressed by addition of 1,25(OH)2D3 to osteoblast cultures [68,95,98–101]. In osteoblast cultures without exogenous vitamin D supplementation, BSP is expressed at its highest levels immediately prior to onset of extracellular matrix mineralization [101]. Thus one wonders whether its suppression in the presence of 10−8 M 1,25(OH)2D3 may reflect a compensatory mechanism, preventing excessive initial calcification. Osteocalcin (OC) was one of the first matrix proteins whose expression was shown to be up-regulated by vitamin D [102]. OC is a member of a large family of hepatic and skeletal vitamin K–dependent proteins that undergo posttranslational modification and γ-carboxylation at key glutamic acid residues (Gla) and have mineral binding capacities. The mature 5.7-kDa

0.50 C 0.40

Ca : DNA

4P 0.30

4P +D day5

0.20

4P +D day7 4P +D day9

0.10

0.00 12

14

16 Days in Culture

19

21

FIGURE 4 Effect of 10−8 M 1,25(OH)2D3 on mineralization in differentiating chick limb bud mesenchymal cell micromass cultures. Data is expressed as µg Ca/µg DNA. Continuous vitamin D treatment was started at day 5 (when cartilage nodules formed), at day 7 (prior to the start of visible chondrocyte hypertrophy), or at day 9 (2 days before start of mineralization). Control cultures received a comparable volume (20 µl) of ethanol, the carrier for 1,25(OH)2D3. On days 12–21, Ca and DNA contents were determined in control and mineralizing cultures. The DNA contents of mineralizing and nonmineralizing cultures were not significantly different (not shown), and 1,25(OH)2 vitamin D3 treatment did not significantly accelerate proliferation. There was a slight increase in DNA content of cultures treated on day 5, but neither this increase nor the gradual increase in DNA content with time was significant, indicating that the cultures had already achieved their plateau phase of proliferation. In contrast, cultures treated on day 9 showed an increase in Ca content relative to the other cultures.

487

CHAPTER 27 Mineralization

protein contains three Gla residues and accumulates in bone as a result of its high affinity for hydroxyapatite [24]. Initial osteoblastic expression of OC occurs after the onset of extracellular matrix mineralization and increases with progressive mineralization and maturation of the osteoblast to a terminally differentiated state [68]. 1,25(OH)2D3 is a major stimulus for the transcription of human and rat OC, but it inhibits the transcription of the mouse OC gene [103], implying either a species difference or, more likely, differences in conditions under which expression was studied. OC can inhibit the formation of hydroxyapatite in vitro [104], a function that requires the presence of the Gla residues [105]. OC also is important for osteoclast recruitment [24]. OC-deficient mice have increased rates of bone formation despite a lack of detectable abnormalities in osteoclast numbers [106]. These mice also have increased mineral content and increased mineral crystallinity [107], supporting the role of OC in regulating mineral turnover. Thus OC has a role in the recruitment of osteoclasts to the surface of mineralized bone, contributing in this way to the regulation of both bone formation and resorption [101]. OC biosynthesis is suppressed throughout the life span of the culture, as a result of an arrested stage of cell differentiation when 1,25(OH)2 vitamin D3 is added to rat osteoblast cultures before the onset of mineralization. Once the cultures begin to mineralize and the cells are terminally differentiated, media OC concentrations are increased by continuous or acute application of the vitamin [108]. Osteopontin (OPN), a glycoprotein produced in a number of tissues [109], is consistently up-regulated by 1,25(OH)2 vitamin D3 in proliferating and differentiated mouse and rat osteoblasts [110]. In bone, OPN mediates autocrine–paracrine functions in the regulation of tissue formation. OPN is important for recruiting osteoclasts for bone remodeling [110] and it acts as a signaling protein in many tissues. OPN is an in vitro inhibitor of mineralization in muscle, cartilage, and bone [24,110]. In that light it is of interest to note that increasing levels of inorganic phosphate activate OPN expression [109].

II. MINERALIZATION AND MINERAL PROPERTIES IN MODEL SYSTEMS WITH VITAMIN D ALTERATIONS The effects on mineralization of vitamin D, 1,25(OH)2 vitamin D3 and other vitamin D metabolites varies with dose and type, as well as with the system being studied. Even in similar species effects depend on animal age, gender, route of delivery, and method of analysis.

In cell culture, results are also dependent on the maturity of cells or tissues, concentrations of metabolite, and whether the metabolite is given acutely or continuously. It is well known that vitamin D deficiency or even vitamin D insufficiency causes rickets (failure of growth plate cartilage to become mineralized) and osteomalacia (failure of osteoid to become mineralized) [112–114], and that vitamin D toxicity (hypervitaminosis D) causes dystrophic calcification [19,115]. But even in rodents the effects of vitamin D deficiency (or D insufficiency) and hypervitaminosis D can be variable. In cell and organ culture the same vitamin D metabolites have been reported both to prevent mineralization and to enhance it. Understanding the reasons for this diversity can provide a good deal of insight into mineralization mechanisms.

A. Animal Models and Human Diseases with Mineral Alterations 1. VITAMIN D DEFICIENCY

The mineral content (ash weight), crystallinity parameter (β002, which is inversely related to crystallite size and perfection), and Ca/P ratio of the mineral in the epiphyseal growth plates, metaphysis, diaphysis, and calvaria of several models of vitamin-D deficient rickets are compared in Fig. 3. In each model, the mineral content is decreased relative to age-matched controls, and the crystals tend to be larger, but Ca/P ratios are quite variable. The vitamin D–deficient rat (Fig. 3A [116]), the hypophosphatemic mouse (Fig. 3B [117]), and the second-generation vitamin D-deficient rat (Fig. 3C [118]) show distinct differences, although the D-deficient animals have negligible 1,25(OH)2vitamin D3 and reduced 25 (OH) vitamin D levels. In the secondgeneration vitamin D-deficient rats, born to vitamin D– deficient mothers [116], there are only small differences in bone ash weight, crystallinity, and Ca/P ratios when compared to age-matched controls. These differences are greater in comparably aged, comparable sex and background, first-generation vitamin D–deficient rats and chickens [118,119]. The initial electron microscopic studies of vitamin D– deficient bones [120] pointed out that the total number of extracellular matrix vesicles was not different from that in vitamin D–sufficient rats, but the vesicles contained less mineral and an extracelllular matrix devoid of mineral. Healing of the rickets was then associated with both matrix vesicle and extracellular matrix mineralization. In another classic study, EDAX was used to monitor changes in cellular and extracellular matrix ion concentrations in all zones of normal and rachitic growth plates [121]. Potassium levels were invariant,

488 and calcium was found throughout the extracellular matrix in both normal and rachitic growth plates. The phosphate levels in the extracellular matrix did not increase until mineralization commenced, and hence were lower in the rachitic matrix, leading the authors to point out the importance of phosphate sufficiency for mineralization. Since vitamin D is now known to regulate both calcium and phosphate levels ([122] and Section III of this book), these observations are no longer surprising, but are important in terms of the mineralization mechanisms discussed previously. In the hypophosphatemic mouse [117], in which failure to mineralize the growth plate and osteoid is due to a genetic abnormality in the endopeptidase known as PHEX (see Chapter 70), the Ca/P ratios are elevated, probably because of abnormalities in renal phosphate transport. In addition, there is less bone mineral and the mineral crystals that are present tend to be larger/more perfect. A mechanism can be postulated to explain these mineralization abnormalities in the bones of hypophosphatemic animals. It is suggested that vitamin D regulates mineralization by controlling degradation of matrix proteins based on the knowledge that a potential substrate for PHEX is matrix extracellular phosphoglycoprotein (MEPE) [123,124] and, by analogy, the structurally related phosphorylated matrix glycoproteins, OPN, BSP, and DMP-1, and that vitamin D causes down-regulation of PHEX expression [125]. In other words, vitamin D suppresses the breakdown of these PHEX substrates, enabling them to act as mineralization inhibitors. PHEX may also be important in producing nucleators by the degradation of these phosphoproteins. Although we know that many of these phosphoproteins can act both as nucleators and inhibitors [24], this mechanism is hypothetical and remains to be validated. The mineral properties of vitamin D insufficiency in human children have not been characterized. However, mineral in adults with osteomalacia has been subject to more detailed analysis (see later discussion). 2. VDR ALTERATIONS

The nuclear vitamin D receptor (VDR) binds 1,25(OH)2 vitamin D with high affinity and selectivity, as reviewed in Section II. The histologic appearance of the bone tissue of vitamin D receptor knockout (VDRKO) animals resembles that of an animal with vitamin D deficiency [75,126,127]. The mice have a normal skeleton at birth but develop hypocalcemia and hyperparathyroidism shortly after weaning, comparable to human vitamin D–resistant rickets type II [128,129] and Chapters 70 and 72). While bone volume was not decreased in these mice, the amount of unmineralized

DAN FAIBISH AND ADELE L. BOSKEY

bone (osteoid) was increased 15-fold compared to the controls [128]. However, the rachitic bone malformation and growth retardation in the VDR-KO can be rescued by dietary calcium supplementation or phosphorus restriction [129]. These animals have normal circulating levels of 25-OHvitamin D and elevated levels of 1,25(OH)2vitamin D. They have slight hypocalcemia and elevated PTH levels due to the inability of the vitamin D target organs to regulate calcium influx. There have not yet been characterizations of the mineral in these animals; however, it is anticipated that they will resemble the parameters in vitamin D–deficient bones. The VDR knockout animals have extended growth plates and an increased number of hypertrophic chondrocytes [128]. The hypertrophic chondrocytes express the correct relative amounts of phenotypic markers of chondrocyte maturity, type X collagen, vascular endothelial growth factor (VEGF), and osteopontin, but show decreased staining for annexin V–phosphatidylserine, an early marker of apoptosis [130]. This suggests that impaired mineralization might be due to the failure of chondrocytes to undergo apoptosis, but another possibility is that the calcium and phosphate levels are not sufficient to mineralize the growth plates, with apoptosis following rather than preceding mineralization. Additionally, both annexin and phosphatidylserine are membrane components of extracellular matrix vesicles; thus decreased staining for these components might suggest that chondrocytes are not producing vesicles, or that vesicles lacking these components are not functional. Osteoclast numbers are normal in the VDR knockout [75], arguing against excessive remodeling as the origin of the mineralization defect. More is known about the mineral characteristics in the vitamin D receptor transgenics. Gardiner et al. [53] overexpressed the vitamin D receptor under control of the osteocalcin promoter. The long bones in these VDR transgenic mice had increased cross-sectional area and increased strength with two- and threefold elevations of osteoblast VDR levels [53]. Based on qBEI analysis [65], cortical bone peak mineral content (CaPeak) was enhanced in a dose dependent manner with the mice with 3 × overexpression having values higher than controls (p < 0.02), and the transgenic mice with 2 × overexpression having intermediate values. In trabecular bone, the increase in CaPeak was significant for both twofold and threefold overexpression. Calcified cartilage and primary spongiosa showed parallel trends. There was also increased homogeneity of mineralization in the transgenic mice. However, FTIR microspectroscopy, XRD, and scanning-SAXS failed to show any significant alterations in the mineral crystal properties of the transgenic bones, although there was a trend for the

489

CHAPTER 27 Mineralization

Control Transgenic

Mineral : Matrix Ratio

12 9 6 3 0

CP

CC Cortical bone

CE

T Trabecular bone

FIGURE 5 Infrared microspectroscopy of the effect of three-fold VDR overexpression on the tibia of 4-month-old mice shows slight but not significant increase in mineral content (mineral:matrix ratio). Data were collected from cortical bone on the periosteal side (CP), in the center (CC), and on the endosteal side (CE), as well as within individual trabeculae (T).

4. THE 1-HYDROXYLASE KNOCKOUT

Mice that lack the 1-α-hydroxylase also present with hypocalcemia, rickets, and osteomalacia [134] and mimic human pseudovitamin D–deficiency rickets (PDDR) (see Chapter 71). Patients that lack this hydroxylase have secondary hyperparathyroidism, growth retardation, rickets, and osteomalacia [135]. The mineral in the bones of the PDDR mice has not been characterized, but their mechanical properties are typical of rachitic animals with significantly reduced stiffness and reduced ultimate load [136] (see Chapter 7). Treatment of these animals with high calcium, phosphate, and lactose diets rescues the phenotype, but the bones do not grow as well as the VDR knockout mice given the same diet [136], implying that the 1,25(OH)2 vitamin D may have properties beyond regulating serum calcium and phosphate levels. It is important to note that there were no differences in osteoclast number in the control and knockout mice given the normal or the enriched diet. 5. OSTEOPOROSIS AND OSTEOMALACIA IN HUMANS

mineral content (mineral:matrix ratio) to be increased in the transgenic mice (Fig. 5) while crystallinity was identical in both groups [65]. Thus VDR overexpression did not markedly change the mineral properties and did not mimic effects of vitamin D toxicity. In humans, hereditary 1,25(OH)2 vitamin D–resistant rickets (HVDRR), also known as vitamin D–dependent rickets type II, is a rare autosomal recessive disease that arises as a result of mutations in the gene encoding the VDR [131,132] (see Chapter 72). There have been no detailed mineral analyses in these patients. 3. VITAMIN D CARRIER PROTEIN KNOCKOUT

Vitamin D is carried to its receptor (VDR) in the target tissues by the vitamin D binding protein (DBP). The mineral properties in DBP knockout mice have not been reported, but they have an interesting phenotype [133]. DBP knockout mice have low levels of serum vitamin D metabolites but otherwise appear normal and show none of the bone abnormalities seen in vitamin D–deficiency rickets or osteomalacia. When stressed with excess exogenous vitamin D (1000 U/g body weight) to induce vitamin D toxicity, they did have elevated serum calcium levels, but did not show the toxic effects found in wild-type animals. Kidney mineral deposits were found in the wild-type but not in the knockout mice. Additionally, when stressed with a diet low in vitamin D, histomorphometry showed increased thickening of the osteoid seams in the DBP knockout mice, consistent with impaired mineralization and “hypovitaminosis D osteopathy.”

Many individuals with osteoporosis are Ca deficient, demonstrate vitamin D insufficiency, and show histochemical evidence of osteomalacia [137,138]. Osteoporotic bones without evidence of osteomalacia have decreased mineral content and increased mineral crystallinity based on FTIR microspectroscopic analysis [139,140]. Since it is likely that some of these osteoporotic patients could have osteomalacia, we tested the hypothesis that the mineral in the osteomalacic bone was different from normal. Two groups of methacrylateembedded iliac crest biopsies were examined by FTIRI. In this preliminary study, controls were seven female subjects, aged 36–57, without apparent bone disease and four female subjects with morphologically defined osteomalacia. In tissue sections from patients with osteomalacia, mineralization rates determined by histomorphometry were decreased, while osteoid parameters were increased, revealing defective primary mineralization. In those sections, the degree of mineralization measured by qBEI in old bone (calcified tissue) was slightly higher than in four of the controls, reflecting the normal evolution of the secondary mineralization of bone tissue. The FTIRI for the biopsies showed that mineral content is decreased, not significantly, in both cortical and trabecular bone of the osteomalacic patients, with slight but not significant increases in crystallinity (Fig. 6). No differences in mineral content, crystallinity, or collagen maturity parameters were found when 400 µm × 400 µm areas of mineralized regions in osteomalacic and control specimens were compared. These findings support the hypothesis that the

490

DAN FAIBISH AND ADELE L. BOSKEY

Control

Osteomalacia

5

4

3

2

1

0 Cortical Bone

Trabecular Bone

Cortical Bone

Mineral : Matrix

Trabecular Bone

Crystallinity

FIGURE 6 Infrared imaging comparing the mean mineral content (mineral: matrix ratio) and crystallinity in normal female patients to those in female patients with osteomalacia. Values are mean ± SD (n = 7 and 4, respectively) as averaged from three to five fields from each biopsy.

inferior mechanical properties of osteomalacic bone originate in the delayed pattern of primary mineralization and the resulting smaller amount of mineralized tissue.

III. CONCLUSIONS Vitamin D affects mineralization predominantly by regulating local calcium and phosphate levels. Additionally, because vitamin D has genomic and nongenomic effects on the mineralizing tissues’ cells, vitamin D’s indirect actions on membrane properties, enzyme activity, and matrix protein expression and phosphorylation can affect the mineral that is formed in its presence or absence. A recent report [141] indicates that an analog of 1,25-(OH)2 vitamin D3 stimulates osteoblast activity in vitro and in ovariectomized rats, arguing for a direct effect of vitamin D on bone cells beyond the increase in calcium and phosphate levels discussed in this chapter. The properties of the matrix formed with this new analog appear to go beyond the stimulation of matrix protein synthesis, but the properties of the mineral in animals treated with this drug are not known. Unfortunately there is a dearth of information on the mineral properties in most of the new animal models in which mineralization is altered. It is hoped that this chapter will encourage analyses of those tissues.

References 1. Glasauer S, Langley S, Beveridge TJ 2002 Intracellular iron minerals in a dissimilatory iron-reducing bacterium. Science 295:117–119. 2. Wetherbee R 2002 Biomineralization. The diatom glasshouse. Science 298:547. 3. Wilt FH 2002 Biomineralization of the spicules of sea urchin embryos. Zoolog Sci 19:253-61. 4. Pierce DC, Butler KD, Roer RD 2001 Effects of exogenous N-acetylhexosaminidase on the structure and mineralization of the post-ecdysial exoskeleton of the blue crab, Callinectes sapidus. Comp Biochem Physiol B Biochem Mol Biol 128:691–700. 5. Boskey AL 2001 Bone mineralization. In: Cowin SC (ed) Bone Biomechanics, 3rd ed. CRC Press, Boca Raton, FL, pp. 5.1–5.34. 6. Plate U, Arnold S, Stratmann U, Wiesmann HP, Hohling HJ. 1998 General principle of ordered apatitic crystal formation in enamel and collagen rich hard tissues. Connect Tissue Res 38:149–157. 7. Beshensky AM, Wesson JA, Worcester EM, Sorokina EJ, Snyder CJ, Kleinman JG 2001 Effects of urinary macromolecules on hydroxyapatite crystal formation. J Am Soc Nephrol 12:2108–16. 8. Boskey AL, Maresca M, Doty S, Sabsay B, Veis A 1990 Concentration-dependent effects of dentin phosphophoryn in the regulation of in vitro hydroxyapatite formation and growth Bone Miner 11:55–65. 9. Smith CE 1998 Cellular and chemical events during enamel maturation. Crit Rev Oral Biol Med 9:128–61. 10. Vernois V, Deschamps N, Revel G 1989 Characterization of human dental enamel by epithermal neutron activation analysis. J Trace Elem Electrolytes Health Dis 3:67–70.

CHAPTER 27 Mineralization

11. Handschin RG, Stern WB 1994 Crystallographic and chemical analysis of human bone apatite (Crista Iliaca) Clin Rheumatol 13(Suppl 1):75–90. 12. Ou-Yang H, Paschalis EP, Mayo WE, Boskey AL, Mendelsohn R 2001 Infrared microscopic imaging of bone: spatial distribution of CO32−. J Bone Miner Res 16: 893–900. 13. Loong CK, Rey C, Kuhn LT, Combes C, Wu Y, Chen S, Glimcher MJ 2000 Evidence of hydroxyl-ion deficiency in bone apatites: an inelastic neutron-scattering study. Bone 26:599–602. 14. Cho G, Wu Y, Ackerman JL 2003 Detection of hydroxyl ions in bone mineral by solid-state NMR spectroscopy. Science 300:1123–1127. 15. Boskey AL 2002 Pathogenesis of cartilage calcification: mechanisms of crystal deposition in cartilage. Curr Rheumatol Rep 4:245–251. 16. Gerbaud V, Pignol D, Loret E, Bertrand JA, Berland Y, Fontecilla-Camps JC, Canselier JP, Gabas N, Verdier JM 2000 Mechanism of calcite crystal growth inhibition by the N-terminal undecapeptide of lithostathine. J Biol Chem 275:1057–1064. 17. Fam AG 2000 What is new about crystals other than monosodium urate? Curr Opin Rheumatol 12:228–234. 18. Rajasree S, Umashankar PR, Lal AV, Sarma PS, Kartha CC 2002 1,25-Dihydroxyvitamin D3 receptor is upregulated in aortic smooth muscle cells during hypervitaminosis D. Life Sci 70:1777–1788. 19. Price PA, Buckley JR, Williamson MK 2001 The amino bisphosphonate ibandronate prevents vitamin D toxicity and inhibits vitamin D–induced calcification of arteries, cartilage, lungs and kidneys in rats. J Nutr 131:2910–2915. 20. Rodriguez-Cano L, Garcia-Patos V, Creus M, Bastida P, Ortega JJ, Castells A 1996 Childhood calcinosis cutis. Pediatr Dermatol 13:114–117. 21. Blay P, Fernandez-Martinez JM, Diaz-Lopez B 2001 Vertebral involvement in hyperphosphatemic tumoral calcinosis. Bone 28:316–318. 22. Claudepierre P, Rahmouni A, Bergamasco P, Montazel JL, Larget-Piet B, Chevalier X 1997 Misleading clinical aspects of hydroxyapatite deposits: a series of 15 cases. J Rheumatol 24:531–535. 23. Engelse MA, Neele JM, Bronckers AL, Pannekoek H, de Vries CJ 2001 Vascular calcification: expression patterns of the osteoblast-specific gene core binding factor alpha-1 and the protective factor matrix gla protein in human atherogenesis. Cardiovasc Res 52:281–289. 24. Gokhale J, Robey PG, Boskey AL 2001 The biochemistry of bone. In: Marcus R, Feldman D, Kelsey JC (eds) Osteoporosis, 2nd ed. Vol. 1. Academic Press, San Diego, pp. 107–189. 25. Boskey AL 1992 Mineral–matrix interactions in bone and cartilage. Clin Orthop 281:244–274. 26. Potter K, Leapman RD, Basser PJ, Landis WJ 2002 Cartilage calcification studied by proton nuclear magnetic resonance microscopy. J Bone Miner Res 17:652–660. 27. Burr DB 2002 The contribution of the organic matrix to bone’s material properties. Bone 31:8–11. 28. Landis WJ 1999 An overview of vertebrate mineralization with emphasis on collagen-mineral interaction. Gravit Space Biol Bull 12:15–26. 29. Hu CC, Hart TC, Dupont BR, Chen JJ, Sun X, Qian Q, Zhang CH, Jiang H, Mattern VL, Wright JT, Simmer JP 2000. Cloning human enamelin cDNA, chromosomal localization, and analysis of expression during tooth development. J Dent Res 79:912–919.

491 30. Moradian-Oldak J 2001 Amelogenins: assembly, processing and control of crystal morphology. Matrix Biol 20:293–305. 31. Stains JP, Weber JA, Gay CV 2002 Expression of Na+/Ca2+ exchanger isoforms (NCX1 and NCX3) and plasma membrane Ca2+ ATPase during osteoblast differentiation. J Cell Biochem 84:625–635. 32. Lundquist P, Ritchie HH, Moore K, Lundgren T, Linde A Phosphate and calcium uptake by rat odontoblast-like MRPC-1 cells concomitant with mineralization. J Bone Miner Res 17:1801–1813. 33. Penel G, Leroy G, Rey C, Bres E 1998 MicroRaman spectral study of the PO4 and CO3 vibrational modes in synthetic and biological apatites. Calcif Tissue Int 63:475–481. 34. Driessens FC 1988 Physiology of hard tissues in comparison with the solubility of synthetic calcium phosphates. Ann NY Acad Sci 523:131–136. 35. Barry AB, Baig AA, Miller SC, Higuchi WI 2002 Effect of age on rat bone solubility and crystallinity. Calcif Tissue Int 71:167–671. 36. Robinson C, Shore RC, Brookes SJ, Strafford S, Wood SR, Kirkham J 2000 The chemistry of enamel caries. Crit Rev Oral Biol Med 11:481–495. 37. Patel PR, and Brown WE 1975 Thermodynamic solubility product of human tooth enamel: powdered sample. J Dent Res 54:728–736. 38. Anderson HC 1995 Molecular biology of matrix vesicles. Clin Orthop 314:266–280. 39. Boskey AL, Boyan BD, Doty SB, Feliciano A, Greer K, Weiland D, Swain LD, Schwartz Z 1992 Studies of matrix vesicle-induced mineralization in a gelatin gel. Bone Miner 17:257–262. 40. Glimcher MJ 1987 The nature of the mineral component of bone and the mechanism of calcification. Instr Course Lect 36:49–69. 41. Uchiyama T, Tanizawa T, Muramatsu H, Endo N, Takahashi HE, Hara T 1997 A morphometric comparison of trabecular structure of human ilium between microcomputed tomography and conventional histomorphometry. Calcif Tissue Int 61:493–498. 42. Daci E, Verstuyf A, Moermans K, Bouillon R, Carmeliet G 2000 Bone resorption induced by 1α,25 dihydroxyvitamin D3 in vivo is not altered by inactivation of the plasminogen activator inhibitor 1. Bone 27:97–102. 43. Lill CA, Fluegel AK, Schneider E 2000 Sheep model for fracture treatment in osteoporotic bone: a pilot study about different induction regimens. J Orthop Trauma 14:559–565. 44. Boonen S, Cheng XG, Nijs J, Nicholson PH, Verbeke G, Lesaffre E, Aerssens J, Dequeker J 1997 Factors associated with cortical and trabecular bone loss as quantified by peripheral computed tomography (pQCT) at the ultradistal radius in aging women. Calcif Tissue Int 60:164–170. 45. Pienkowski D, Doers TM, Monier-Faugere MC, Geng Z, Camacho NP, Boskey AL, Malluche HH 1997 Calcitonin alters bone quality in beagle dogs. J Bone Miner Res 12: 1936–1943. 46. Carden A, Morris MD 2000 Application of vibrational spectroscopy to the study of mineralized tissues. J Biomed Opt 5:259–268. 47. Baker D, Kupke KG, Ingram P, Roggli VL, Shelburne JD 1985 Microprobe analysis in human pathology. Scan Electron Microsc 659–680. 48. Lundon K, Grynpas M 1993 The long-term effect of ovariectomy on the quality and quantity of cortical bone in the young cynomolgus monkey: a comparison of density fractionation and histomorphometric techniques. Bone 14:389–395.

492 49. Gronowicz G, Woodiel FN, McCarthy MB, Raisz LG 1989 In vitro mineralization of fetal rat parietal bones in defined serum-free medium: effect of beta-glycerol phosphate. J Bone Miner Res 4:313–324. 50. Tsukamoto Y, Fukutani S, Shin-Ike T, Kubota T, Sato S, Suzuki Y, Mori M 1992 Mineralized nodule formation by cultures of human dental pulp–derived fibroblasts. Arch Oral Biol 37:1045–1055. 51. Costa AM, Fernandes HM 2000 Long-term effects of parathyroid hormone, 1,25-dihydroxyvitamin D3, and dexamethasone on the cell growth and functional activity of human osteogenic alveolar bone cell cultures. Pharmacol Res 42:345–353. 52. Camacho NP, Rinnerthaler S, Paschalis EP, Mendelsohn R, Boskey AL, Fratzl 1999 Complementary information on bone ultrastructure from scanning small angle X-ray scattering and Fourier-transform infrared microspectroscopy. Bone 25:287–293. 53. Gardiner EM, Baldock PA, Thomas GP, Sims NA, Henderson KN, Hollis B, White CP, Sunn KL, Morrison NA, Walsh WR, Eisman JA 2000 Increased formation and decreased resorption of bone in mice with elevated vitamin D receptor in mature cells of the osteoblastic lineage. FASEB J 14:1908–1916. 54. Eppell SJ, Tong W, Katz JL, Kuhn L, Glimcher MJ 2001 Shape and size of isolated bone mineralites measured using atomic force microscopy. J Orthop Res 19:1027–1034. 55. Tong W, Glimcher MJ, Katz JL, Kuhn L, Eppell SJ 2003 Size and shape of mineralites in young bovine bone measured by atomic force microscopy. Calcif Tissue Int 72:592–598. 56. Miller LM, Vairavamurthy V, Chance MR, Mendelsohn R, Paschalis EP, Betts F, Boskey AL 2001 In situ analysis of mineral content and crystallinity in bone using infrared micro-spectroscopy of the nu(4) PO43− vibration. Biochim Biophys Acta 1527:11–19. 57. Gadaleta SJ, Paschalis EP, Betts F, Mendelsohn R, Boskey AL 1996 Fourier transform infrared spectroscopy of the solution-mediated conversion of amorphous calcium phosphate to hydroxyapatite: new correlations between X-ray diffraction and infrared data. Calcif Tissue Int 58:9–16. 58. Grynpas MD, Rey C 1992 The effect of fluoride treatment on bone mineral crystals in the rat. Bone 13:423–429. 59. Rey C, Renugopalakrishnan V, Collins B, Glimcher MJ 1991 Fourier transform infrared spectroscopic study of the carbonate ions in bone mineral during aging. Calcif Tissue Int 49:251–258. 60. Rothwell WP, Waugh JS, Yesinowski JP 1980 High-resolution variable-temperature P-31 NMR of solid calcium phosphates. J Am Chem Soc 102:2637–2643. 61. Wu Y, Ackerman JL, Kim HM, Rey C, Barroug A, Glimcher MJ 2002 Nuclear magnetic resonance spin-spin relaxation of the crystals of bone, dental enamel, and synthetic hydroxyapatites. J Bone Miner Res 17:472–480. 62. Wu Y, Ackerman JL, Strawich ES, Rey C, Kim HM, Glimcher MJ 2003 Phosphate ions in bone: identification of a calcium-organic phosphate complex by 31P solid-state NMR spectroscopy at early stages of mineralization. Calcif Tissue Int 72:610–626. 63. Wu Y, Glimcher MJ, Rey C, Ackerman JL 1994 A unique protonated phosphate group in bone mineral not present in synthetic calcium phosphates. Identification by phosphorus31 solid state NMR spectroscopy. J Mol Biol 244:423–435. 64. Bohic S, Rey C, Legrand A, Sfihi H, Rohanizadeh R, Martel C, Barbier A, Daculsi G. 2000 Characterization of the trabecular

DAN FAIBISH AND ADELE L. BOSKEY

65.

66.

67.

68.

69.

70.

71. 72. 73.

74. 75.

76.

77.

78. 79. 80.

rat bone mineral: Effect of ovariectomy and bisphosphonate treatment. Bone 26:341–348. Misof BM, Roschger P, Tesch W, Baldock PA, Valenta A, Meßmer P, Eisman JA, Boskey AL, Gardiner EM, Fratzl P, Klaushofer K 2003 Targeted overexpression of vitamin D receptor in osteoblasts increases calcium concentration without affecting structural properties of bone mineral crystals. Calcif Tissue Int June 6, e-publication. Balmain N, von Eichel B, Toury R, Belquasmi F, Hauchecorne M, Klaus G, Mehls O, Ritz E 1995 CalbindinD28K and -D9K and 1,25(OH)2 vitamin D3 receptor immunolocalization and mineralization induction in long-term primary cultures of rat epiphysial chondrocytes. Bone 17:37–45. Chang W, Tu C, Pratt S, Chen TH, Shoback D 2002 Extracellular Ca2+-sensing receptors modulate matrix production and mineralization in chondrogenic RCJ3.1C5.18 cells. Endocrinology 143:1467–1474. Lian JB, McKee MD, Todd AM, Gerstenfeld LC 1993 Induction of bone-related proteins, osteocalcin and osteopontin, and their matrix ultrastructural localization with development of chondrocyte hypertrophy in vitro. J Cell Biochem 52:206–219. Khouja HI, Bevington A, Kemp GJ, Russell RG 1990 Calcium and orthophosphate deposits in vitro do not imply osteoblastmediated mineralization: mineralization by betaglycerophosphate in the absence of osteoblasts. Bone 11:385–391. Bonewald LF, Harris SE, Rosser J, Dallas MR, Dallas SL, Camacho NP, Boyan B, Boskey A 2003 Von Kossa staining alone is not sufficient to confirm that mineralization in vitro represents bone formation. Calcif Tissue Int 72:537–547. Kingma JG Jr, Roy PE 1988 Ultrastructural study of hypervitaminosis D induced arterial calcification in Wistar rats. Artery 16:51–61. Rambeck WA, Strohle HG, Wetzel A, Zucker H 1981 Calcinogenic activity of vitamin D3 and vitamin D3 palmitate in rat and rabbit. Int J Vitam Nutr Res 51:359–364. Jiang YB, Wang YZ, Zhao J, Marchal G, Wang YX, Shen YL, Xing SZ, Zhang XY, Baert AL 1990 Metastatic calcification within bone. The main cause of osteosclerosis in hypervitaminosis D3. Radiologic–pathologic correlation. Invest Radiol 25:1188–1196. Giunta JL 1998 Dental changes in hypervitaminosis D. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 85:410–413. Amling M, Priemel M, Holzmann T, Chapin K. Rueger JM, Baron R, Demay MB 1999 Rescue of the skeletal phenotype of vitamin D receptor-ablated mice in the setting of normal mineral ion homeostasis: formal histomorphometric and biomechanical analyses. Endocrinology 140:4982–4987. Weinstein RS, Underwood JL, Hutson MS, DeLuca HF 1984 Bone histomorphometry in vitamin D–deficient rats infused with calcium and phosphorus. Am J Physiol 246: E499–E505. Holtrop ME, Cox KA, Carnes DL, Holick MF 1986 Effects of serum calcium and phosphorus on skeletal mineralization in vitamin D–deficient rats. Am J Physiol 251: E234–E240. Boskey AL 1989 Hydroxyapatite formation in a dynamic gel system: Effects of type I collagen, lipids, and proteoglycans. J Phys Chem 93:1628–1633. Giachelli CM, Jono S, Shioi A, Nishizawa Y, Mori K, Morii H 2001 Vascular calcification and inorganic phosphate. Am J Kidney Dis 38:S34–S377. Kato Y, Boskey A, Spevak L, Dallas M, Hori M, Bonewald LF 2001 Establishment of an osteoid preosteocyte-like cell

CHAPTER 27 Mineralization

81.

82.

83.

84. 85.

86.

87. 88.

89.

90.

91.

92.

93.

94.

95.

MLO-A5 that spontaneously mineralizes in culture. J Bone Miner Res 16:1622–1633. Shane E, Parisien M, Henderson JE, Dempster DW, Feldman F, Hardy MA, Tohme JF, Karaplis AC, Clemens TL 1997 Tumor-induced osteomalacia: clinical and basic studies. J Bone Miner Res 12:1502–1511. Siris ES, Clemens TL, Dempster DW, Shane E, Segre GV, Lindsay R, Bilezikian JP 1987 Tumor-induced osteomalacia. Kinetics of calcium, phosphorus, and vitamin D metabolism and characteristics of bone histomorphometry. Am J Med 82: 307–312. Holtrop ME, Cox KA, Clark MB, Holick MF, Anast CS 1981 1,25-Dihydroxycholecalciferol stimulates osteoclasts in rat bones in the absence of parathyroid hormone. Endocrinology 108:2293–2301. Newbrey JW, Truitt ST, Roland DA, Frost TJ, Untawale GG 1992 Bone histomorphometry in 1,25(OH)2D3- and vitamin D3– treated aged laying hens. Avian Dis 36:700–706. Denis I, Cournot G, Lacroix H, Colin C, Zerath E, Pointillart A 2000 In vivo bone metabolism and ex vivo bone marrow osteoprogenitors in vitamin D–deprived pigs. Bone 26: 491–498. Farach-Carson MC, Xu Y 2002 Microarray detection of gene expression changes induced by 1,25(OH)2D3 and a Ca2+ influxactivating analog in osteoblastic ROS 17/2.8 cells. Steroids 67:467–470. Binderman I, Somjen D 1984 24,25-Dihydroxycholecalciferol induces the growth of chick cartilage in vitro. Endocrinology 115:430–432. Nakamura T, Suzuki K, Hirai T, Kurokawa T, Orimo H 1992 Increased bone volume and reduced bone turnover in vitamin D–replete rabbits by the administration of 24R,25-dihydroxyvitamin D3. Bone 13:229–236. Erben RG, Scutt AM, Miao D, Kollenkirchen U, Haberey M 1997 Short-term treatment of rats with high dose 1,25-dihydroxyvitamin D3 stimulates bone formation and increases the number of osteoblast precursor cells in bone marrow. Endocrinology 138:4629–4635. Hale LV, Kemick ML, Wuthier RE 1986 Effect of vitamin D metabolites on the expression of alkaline phosphatase activity by epiphyseal hypertrophic chondrocytes in primary cell culture. J Bone Miner Res 1:489–495. Boskey AL, Maresca M, Ullrich W, Doty SB, Butler WT, Prince CW 1993 Osteopontin-hydroxyapatite interactions in vitro: inhibition of hydroxyapatite formation and growth in a gelatin-gel. Bone Miner 22:147–159. Mahonen A, Jukkola A, Risteli L, Risteli J, Maenpaa PH 1998 Type I procollagen synthesis is regulated by steroids and related hormones in human osteosarcoma cells. J Cell Biochem 68:151–163. Pavlin D, Bedalov A, Kronenberg MS, Kream BE, Rowe DW, Smith CL, Pike JW, Lichtler AC 1994 Analysis of regulatory regions in the COL1A1 gene responsible for 1,25-dihydroxyvitamin D3–mediated transcriptional repression in osteoblastic cells. J Cell Biochem 56:490–501. Bellows CG, Reimers SM, Heersche JN 1999 Expression of mRNAs for type-I collagen, bone sialoprotein, osteocalcin, and osteopontin at different stages of osteoblastic differentiation and their regulation by 1,25 dihydroxyvitamin D3. Cell Tissue Res 297:249–259. Broess M, Riva A, Gerstenfeld LC 1995 Inhibitory effects of 1,25(OH)2 vitamin D3 on collagen type I, osteopontin, and osteocalcin gene expression in chicken osteoblasts. J Cell Biochem 57:440–451.

493 96. Matsumoto T, Igarashi C, Takeuchi Y, Harada S, Kikuchi T, Yamato H, Ogata E 1991 Stimulation by 1,25-dihydroxyvitamin D3 of in vitro mineralization induced by osteoblast-like MC3T3-E1 cells. Bone 12:27–32. 97. Tye CE, Rattray KR, Warner KJ, Gordon JA, Sodek J, Hunter GK, Goldberg HA 2003 Delineation of the hydroxyapatitenucleating domains of bone sialoprotein. J Biol Chem 278:7949–7955. 98. Chen JJ, Jin H, Ranly DM, Sodek J, Boyan BD 1999 Altered expression of bone sialoproteins in vitamin D–deficient rBSP2.7Luc transgenic mice. J Bone Miner Res 14: 22122–22129. 99. Yang R, Gerstenfeld LC 1997 Structural analysis and characterization of tissue and hormonal responsive expression of the avian bone sialoprotein (BSP) gene. J Cell Biochem 64:77–93. 100. Sodek J, Kim RH, Ogata Y, Li J, Yamauchi M, Zhang Q, Freedman LP 1996 Regulation of bone sialoprotein gene transcription by steroid hormones. Connect Tissue Res 32:209–217. 101. White C, Gardiner E, Eisman J 1999 Tissue specific and vitamin D responsive gene expression in bone. Mol Biol Rep 25:45–61. 102. Price PA, Baukol SA 1981 1,25-Dihydroxyvitamin D3 increases serum levels of the vitamin K–dependent bone protein. Biochem Biophys Res Commun 99:928–935. 103. Chen TL, Fry D 1999 Hormonal regulation of the osteoblastic phenotype expression in neonatal murine calvarial cells. Calcif Tissue Int 64:304–309. 104. Boskey AL, Wians FH Jr, Hauschka PV 1985 The effect of osteocalcin on in vitro lipid-induced hydroxyapatite formation and seeded hydroxyapatite growth. Calcif Tissue Int 37:57–62. 105. Poser JW, Price PA 1979 A method for decarboxylation of gamma-carboxyglutamic acid in proteins. Properties of the decarboxylated gamma-carboxyglutamic acid protein from calf bone. J Biol Chem 254:431–436. 106. Ducy P, Desbois C, Boyce B, Pinero G, Story B, Dunstan C, Smith E, Bonadio J, Goldstein S, Gundberg C, Bradley A, Karsenty G 1996 Increased bone formation in osteocalcindeficient mice. Nature 382:448–452. 107. Boskey AL, Gadaleta S, Gundberg C, Doty SB, Ducy P, Karsenty G 1998 Fourier transform infrared microspectroscopic analysis of bones of osteocalcin-deficient mice provides insight into the function of osteocalcin. Bone 23:187–196. 108. Carpenter TO, Moltz KC, Ellis B, Andreoli M, McCarthy TL, Centrella M, Bryan D, Gundberg CM 1998 Osteocalcin production in primary osteoblast cultures derived from normal and Hyp mice. Endocrinology 139:35–43. 109. Giachelli CM and Steitz S 2000 Osteopontin: a versatile regulator of inflammation and biomineralization. Matrix Biol 19:615–622. 110. Lian JB, Shalhoub V, Aslam F, Frenkel B, Green J, Hamrah M, Stein GS, Stein JL 1997 Species-specific glucocorticoid and 1,25-dihydroxyvitamin D responsiveness in mouse MC3T3-E1 osteoblasts: Dexamethasone inhibits osteoblast differentiation and vitamin D down-regulates osteocalcin gene expression. Endocrinology 138:2117–2127. 111. Boskey AL, Spevak L, Paschalis E, Doty SB, McKee MD 2002 Osteopontin deficiency increases mineral content and mineral crystallinity in mouse bone. Calcif Tissue Int 71:145–154. 112. Berry JL, Davies M, Mee AP 2002 Vitamin D metabolism, rickets, and osteomalacia. Semin Musculoskelet Radiol 6:173–182.

494 113. Mawer EB, Davies M 2001 Vitamin D nutrition and bone disease in adults. Rev Endocr Metab Disord 2:153–164. 114. Pettifor JM 2002 Rickets. Calcif Tissue Int 70:398–399. 115. Evliyaoglu O, Berberoglu M, Ocal G, Adiyaman P, Aycan Z 2001 Severe hypercalcemia of an infant due to vitamin D toxicity associated with hypercholesterolemia. J Pediatr Endocrinol Metab 14:915–919. 116. Boskey AL, Di Carlo EF, Gilder H, Donnelly R, Weintroub S 1988 The effect of short-term treatment with vitamin D metabolites on bone lipid and mineral composition in healing vitamin D-deficient rats. Bone 9:309–318. 117. Boskey AL, Gilder H, Neufeld E, Ecarot B, Glorieux FH 1991 Phospholipid changes in the bones of the hypophosphatemic mouse. Bone 12:345–351. 118. Boskey AL, Dickson IR 1988 Influence of vitamin D status on the content of complexed acidic phospholipids in chick diaphyseal bone. Bone Miner 4:365–371. 119. Donnelly R, Bockman R, DiCarlo E, Betts F, Boskey A 1993 The effect of gallium nitrate on healing of vitamin D- and phosphate-deficient rickets in the immature rat. Calcif Tissue Int 53:400–410. 120. Anderson H.C, Cecil R, Sajdera SW 1975 Calcification of rachitic rat cartilage in vitro by extracellular matrix vesicles. Am J Pathol 79:237–254. 121. Shapiro IM, Boyde A 1984 Microdissection—elemental analysis of the mineralizing growth cartilage of the normal and rachitic chick. Metab Bone Dis Relat Res 5:317–326. 122. Xu H, Bai L, Collins JF, Ghishan FK 2002 Age-dependent regulation of rat intestinal type IIb sodium-phosphate cotransporter by 1,25-(OH)2 vitamin D3. Am J Physiol Cell Physiol 282:C487–C493. 123. Argiro L, Desbarats M, Glorieux FH, Ecarot B 2001 Mepe, the gene encoding a tumor-secreted protein in oncogenic hypophosphatemic osteomalacia, is expressed in bone. Genomics 74:342–351. 124. Campos M, Couture C, Hirata IY, Juliano MA, Loisel TP, Crine P, Juliano L, Boileau G, Carmona AK 2003 Human recombinant PHEX has a strict S1′ specificity for acidic residues and cleaves peptides derived from FGF-23 and MEPE. Biochem J [epub ahead of print] 125. Ecarot B, Desbarats M 1999 1,25-(OH)2D3 down-regulates expression of Phex, a marker of the mature osteoblast. Endocrinology 140:1192–1199. 126. Kato S, Takeyama K, Kitanaka S, Murayama A, Sekine K, Yoshizawa T 1999 In vivo function of VDR in gene expressionVDR knock-out mice. J Steroid Biochem Mol Biol 69: 247–251. 127. Masuyama R, Nakaya Y, Tanaka S, Tsurukami H, Nakamura T, Watanabe S, Yoshizawa T, Kato S, Suzuki K 2001 Dietary phosphorus restriction reverses the impaired bone mineralization in vitamin D receptor knockout mice. Endocrinology 142:494–497. 128. Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997 Targeted ablation of the vitamin D receptor: An animal model of vitamin D–dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94:9831–9835. 129. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Gene 16:391–396. 130. Donohue MM, Demay MB 2002 Rickets in VDR null mice is secondary to decreased apoptosis of hypertrophic chondrocytes. Endocrinology 143:3691–3694.

DAN FAIBISH AND ADELE L. BOSKEY

131. Mechica JB, Leite MO, Mendonca BB, Frazzatto ES, Borelli A, Latronico AC 1997 A novel nonsense mutation in the first zinc finger of the vitamin D receptor causing hereditary 1,25dihydroxyvitamin D3–resistant rickets. J Clin Endocrinol Metab 82:3892–3894. 132. Cockerill FJ, Hawa NS, Yousaf N, Hewison M, O’Riordan JL, Farrow SM 1997 Mutations in the vitamin D receptor gene in three kindreds associated with hereditary vitamin D resistant rickets. J Clin Endocrinol Metab 82:3156–3160. 133. Safadi FF, Thornton P, Magiera H, Hollis BW, Gentile M, Haddad JG, Liebhaber SA, Cooke NE 1999 Osteopathy and resistance to vitamin D toxicity in mice null for vitamin D binding protein. J Clin Invest 103:239–251. 134. Panda D, Miao ML, Tremblay J, Sirois R, Farookhi GN, Hendy D, Goltzman D 2001 Targeted ablation of the 25-hydroxyvitamin D 1α-hydroxylase enzyme: Evidence for skeletal, reproductive, and immune dysfunction. Proc Natl Acad Sci USA 98:7498–7503. 135. Fraser D, Kooh SW, Kind HP, Holick MF, Tanaka Y, DeLuca HF 1973 Pathogenesis of hereditary vitamin-D-dependent rickets. An inborn error of vitamin D metabolism involving defective conversion of 25-hydroxyvitamin D to 1α,25-dihydroxyvitamin D. N Engl J Med 289:817–822. 136. Dardenne O, Prud’homme J, Hacking SA, Glorieux FH, St-Arnaud R 2003 Correction of the abnormal mineral ion homeostasis with a high-calcium, high-phosphorus, highlactose diet rescues the PDDR phenotype of mice deficient for the 25-hydroxyvitamin D-1α-hydroxylase (CYP27B1). Bone 32:332–340. 137. Nordin BE, Morris HA 1989 The calcium deficiency model for osteoporosis. Nutr Rev 47:65–72. 138. Gennari C 2001 Calcium and vitamin D nutrition and bone disease of the elderly. Public Health Nutr 4:547–559. 139. Paschalis EP, Betts F, DiCarlo E, Mendelsohn R, Boskey AL 1997 FTIR microspectroscopic analysis of human iliac crest biopsies from untreated osteoporotic bone. Calcif Tissue Int 61:487–492. 140. Gadeleta SJ, Boskey AL, Paschalis E, Carlson C, Menschik F, Baldini T, Peterson M, Rimnac CM 2000 A physical, chemical, and mechanical study of lumbar vertebrae from normal, ovariectomized, and nandrolone decanoate-treated cynomolgus monkeys (Macaca fascicularis). Bone 27:541–550. 141. Shevde NK, Plum LA, Clagett-Dame M, Yamamoto H, Pike JW, DeLuca HF 2002 A potent analog of 1α,25-dihydroxyvitamin D3 selectively induces bone formation. Proc Natl Acad Sci USA 99:13487–13491. 142. Schinke T, Amendt C, Trindl A, Poschke O, Muller-Esterl W, Jahnen-Dechent W 1996 The serum protein α2-HS glycoprotein/fetuin inhibits apatite formation in vitro and in mineralizing calvaria cells. A possible role in mineralization and calcium homeostasis. J Biol Chem 271:20789–20796. 143. Wu LN, Genge BR, Dunkelberger DG, LeGeros RZ, Concannon B, Wuthier RE 1997 Physicochemical characterization of the nucleational core of matrix vesicles. J Biol Chem 272:4404–4411. 144. Boskey AL, Spevak L, Doty SB, Rosenberg L 1997 Effects of bone CS-proteoglycans DS-decorin, and DS-biglycan on hydroxyapatite formation in a gelatin gel. Calcif Tissue Int 61:298–305. 145. Lidor C, Edelstein S 1987 Calcitriol increases Ca2+-ATPase activity. Biochem Biophys Res Commun 144:713–717. 146. Dean DD, Schwartz Z, Schmitz J, Muniz OE, Lu Y, Calderon F, Howell DS, Boyan BD 1996 Vitamin D regulation of metalloproteinase activity in matrix vesicles. Connect Tissue Res 35:331–336.

CHAPTER 27 Mineralization

147. Uchida M, Shima M, Chikazu D, Fujieda A, Obara K, Suzuki H, Nagai Y, Yamato H, Kawaguchi H 2001 Transcriptional induction of matrix metalloproteinase-13 (collagenase-3) by 1α,25-dihydroxyvitamin D3 in mouse osteoblastic MC3T3-E1 cells. J Bone Miner Res 16:221–230. 148. Lin R, Amizuka N, Sasaki T, Aarts MM, Ozawa H, Goltzman D, Henderson JE, White JH 2002 1α,25-Dihydroxyvitamin D3 promotes vascularization of the chondro-osseous junction by stimulating expression of vascular endothelial growth factor and matrix metalloproteinase 9. J Bone Miner Res 17:1604–1612. 149. Papagerakis P, MacDougall M, Berdal A 2002 Differential epithelial and mesenchymal regulation of tooth-specific matrix proteins expression by 1,25-dihydroxyvitamin D3 in vivo. Connect Tissue Res 43:372–375. 150. Farzaneh-Far A, Weissberg PL, Proudfoot D, Shanahan CM 2001 Transcriptional regulation of matrix gla protein. Z Kardiol 90:s38–s42. 151. Elfahime E, Felix JM, Koch B 1996 Regulation of corticosteroid-binding globulin synthesis by 1α,25-dihyroxyvitamin D3 (calcitriol), 9-cis-retinoic acid and triiodothyronine in cultured rat fetal hepatocytes. J Steroid Biochem Mol Biol 57:109–115. 152. Horton WE Jr, Balakir R, Precht P, Liang CT 1991 1,25Dihydroxyvitamin D3 down-regulates aggrecan proteoglycan expression in immortalized rat chondrocytes through a posttranscriptional mechanism. J Biol Chem 266:24804–24808. 153. Rickard DJ, Kazhdan I, Leboy PS 1995 Importance of 1,25dihydroxyvitamin D3 and the nonadherent cells of marrow for osteoblast differentiation from rat marrow stromal cells. Bone 16:671–678.

495 154. Camacho NP, Carroll P, Raggio CL 2003 Fourier transform infrared imaging spectroscopy (FT-IRIS) of mineralization in bisphosphonate-treated oim/oim mice. Calcif Tissue Int 72:604–609. 155. Boskey AL, Moore DJ, Amling M, Canalis E, Delany A 2003 Infrared analysis of the mineral and matrix in bones of osteonectin-null mice and their wildtype controls. J Bone Miner Res 18:1005–1011. 156. Paine ML, Zhu DH, Luo W, Bringas P Jr, Goldberg M, White SN, Lei YP, Sarikaya M, Fong HK, Snead ML 2000 Enamel biomineralization defects result from alterations to amelogenin self-assembly. J Struct Biol 132:191–200. 157. Gibson CW, Yuan ZA, Hall B, Longenecker G, Chen E, Thyagarajan T, Sreenath T, Wright JT, Decker S, Piddington R, Harrison G, Kulkarni AB 2001 Amelogenin-deficient mice display an amelogenesis imperfecta phenotype. J Biol Chem 276:31871–31875. 158. Xu T, Bianco P, Fisher LW, Longenecker G, Smith E, Goldstein S, Bonadio J, Boskey A, Heegaard AM, Sommer B, Satomura K, Dominguez P, Zhao C, Kulkarni AB, Robey PG, Young MF 1998 Targeted disruption of the biglycan gene leads to an osteoporosis-like phenotype in mice. Nat Genet 20:78–82. 159. Narisawa S, Frohlander N, Millan JL 1997 Inactivation of two mouse alkaline phosphatase genes and establishment of a model of infantile hypophosphatasia. Dev Dyn 208: 432–446. 160. Somerville RPT, Oblander SA, Apte SS. 2003 Matrix metalloproteinases: old dogs with new tricks. Genome Biology 4:216–238.

CHAPTER 28

Modeling and Remodeling: How Bone Cells Work Together A. M. PARFITT

I. II. III. IV.

Division of Endocrinology and Center for Osteoporosis and Metabolic Bone Disease, University of Arkansas for Medical Sciences

Introduction The Structural and Cellular Basis of Bone Growth The Purposes of Bone Remodeling The Basic Multicellular Unit as the Instrument of Bone Remodeling

I. INTRODUCTION For many years bone physiology has been thought of in terms of two opposing but otherwise unrelated kinds of cell, often depicted as sitting on either side of a seesaw. The assumption that osteoclasts and osteoblasts are independent still dominates the field, although it was demolished more than 30 years ago by Harold Frost [1], who recognized that the production and activity of these cells were coordinated in time and space in different modes for different biological purposes, the major modes being redistribution, repair, and replacement (Table I). During growth, bone is formed in one location and after a few weeks or months resorbed in a location that is different relative to the increased size and altered shape of the bone, not because there is anything wrong with it, but because it is no longer needed at that location. As bones grow in length, bone formed at the junction between the growth plate and the metaphysis is resorbed at the junction between the metaphysis and the diaphysis. As bones grow in width, bone formed beneath the periosteum is resorbed at the endosteum. At these various relative locations, resorption and formation continue with only brief interruptions for extended periods; the cells are operating in the modeling mode [2]. In fracture

Parts of this chapter have been previously published in Parfitt AM 1997 Genetic effects on bone mass and turnover-relevance to black/white differences. J Am Coll Nutr 16:325–333, and Parfitt AM 2004 New concepts of bone remodeling: A unified spatial and temporal model with physiological and pathophysiologic implications. In: Agarwal S, Stout S (eds) Kluwer Academic/Plenum New York (in press). Reproduced with permission of the publishers. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Disordered Remodeling and Age-Related Bone Loss VI. Possible Targets for Vitamin D Action VII. Conclusions References

healing, damaged bone is removed and callus in the form of woven bone is laid down; the prime consideration is speed of production and the quality of the bone is of lesser importance. Dead bone can survive for thousands of years, but living bone gradually loses mechanical competence with increasing age and must be periodically replaced by new bone. The replacement mechanism in the adult skeleton, in which osteoclasts and osteoblasts are cooperating for a common purpose, is referred to as remodeling. The salient features of modeling and remodeling are compared in Table II. In modeling, bone is first formed and then resorbed soon after, but in remodeling, bone is first resorbed and then replaced at the same location. Growth is uninterrupted but remodeling occurs infrequently; episodes lasting a few months are usually separated by several years of quiescence. In both modeling and remodeling the rates of resorption and formation are both much higher than the differences between them, but in modeling the processes occur on different surfaces and any matching is genetic or systemic, whereas in remodeling the processes occur on the same surface, and there is local coupling. Modeling leads to net gain in bone mass, but remodeling usually is associated with net loss. Modeling and remodeling are different cellular mechanisms of bone turnover, which is the volume replacement of bone, usually expressed as a fractional rate constant (e.g., %/y), regardless of cellular mechanism. Modeling is the principal mode of bone cell coordination in the growing skeleton and remodeling is the principal mode in the mature skeleton, but this distinction is not absolute. The formation of secondary osteons in the long bone cortices by remodeling [1] and remodeling of cancellous bone in the central skeleton [3] both Copyright © 2005, Elsevier, Inc. All rights reserved.

498

A. M. PARFITT

TABLE I Supracellular Organization of Bone Cells Mode

Need

Redistribution (modeling) Repair (fracture callus) Replacement (remodeling)

Equal quality bone In a different place Any bone In the right place Better bone In the same place

begin during childhood and continue throughout life. After completion of longitudinal growth many bones continue to enlarge very slowly, either by maintenance at a slow rate of periosteal apposition in the modeling mode, or by very slow remodeling with prolonged interruptions [4]. However, remodeling is primarily a phenomenon of the endosteal envelope of the mature skeleton. In both modeling and remodeling osteoclastic bone resorption involves dissolution of the mineral followed soon after by depolymerization of glycosaminoglycans and digestion of collagen and other matrix proteins; at the light microscopic level, mineral and matrix appear to be removed simultaneously [1]. By contrast, osteoblastic bone formation occurs in two distinct stages separated in time and space. Bone matrix is first synthesized and organized, and exists as unmineralized osteoid for a variable time before the onset of mineralization, as explained in more detail in Chapter 63. Bone formed at a location where there is no existing bone is woven, with irregular distribution of collagen fibers, but bone laid down in apposition to an existing bone surface is lamellar and with regular orientation of collagen fibers [1].

TABLE II Comparison of Two Major Modes of Supracellular Organization of Bone Cells Feature Mode Purpose Context Sequence Timing Location Coupling Balance

Modeling

Remodeling

Redistribution Growth Youth F→R Continuous Different Systemic Net gain

Replacement Maintenance Maturity A → R →F Cyclical Same Local Net loss

F, Formation; R, resorption; A, activation. (For explanation of different definitions of this term, see text.)

II. THE STRUCTURAL AND CELLULAR BASIS OF BONE GROWTH During embryonic development the bones of the postcranial skeleton begin as cartilage models of particular size, shape, and location, based on genetic instructions acting via diffusible morphogens [5]. These are gradually replaced by bone as they enlarge, in accordance with the changing mechanical requirements that result from growth in length and increasing muscle strength [5]. The manner of replacement differs in detail between bones, but in a typical future long bone, islands of spiculated bone appear within the ends to form the future epiphyses and an elongated collar of more closely packed bone appears beneath the perichondrium to form the future diaphysis [6]. Cortical bone is mainly formed beneath the periosteum and is mainly resorbed at what has become the endosteum, so that diaphysial cortical bone represents a by-product of growth in width [7,8]. However, to maintain the flared waist-like shape of the metaphysis, there must also be localized endosteal apposition and periosteal resorption [8]. When growth stops, the most recently formed cortical bone becomes the permanent diaphysis. Endochondral ossification is described in Chapter 33, but some aspects especially pertinent to the understanding of Vitamin D effects are summarized here. Between the epiphysis and metaphysis of a growing long bone lies the cartilaginous growth plate (Fig. 1), which is organized vertically as transverse zones and horizontally as vertical columns of chondrocytes, embedded in a honeycomb of matrix [9–11]; in the proximal tibia of a young rat the plate is about 600 µm thick. The sequence of layers known as resting, proliferative, and hypertrophic zones, and the zones of provisional calcification and vascular invasion, represent a sequence of events that occur in the same absolute location, while the epiphysis retreats and the metaphysis approaches. At the upper (or distal) end of the growth plate new cartilage is produced and at the lower (or proximal) end old cartilage is replaced by new bone. For most of the growth period these processes occur at the same rate so that the width of the plate does not change. A typical chondrocyte arises from division of a stem cell in the resting zone, divides once or twice more [12], and increases in volume about 10-fold, all the while synthesizing new matrix [10]. After about 40 hr, corresponding to a growth rate of about 350 µm/day, the longitudinal cartilaginous septa become mineralized. Contrary to previous belief, this requires that the chondrocytes remain viable until mineralization in their vicinity is complete [10,13], when they undergo apoptosis. Ingrowth of capillaries brings macrophages that resorb the transverse unmineralized septa, and chondroclasts

499

CHAPTER 28 Modeling and Remodeling: How Bone Cells Work Together

Secondary ossification centre Primary spongiosum and resorption front

Articular cartilage

Epiphyseal growth plate

Metaphysis

Reserve zone (Germinal layer)

Proliferative zone

Prehypertrophic zone

Hypertrophic zone a) Maturation Diaphysis b) Degeneration (Apotosis)

c) Provisional calcification

FIGURE 1 The epiphyseal growth plate. The right-hand panel shows a detailed representation of chondrocytes in the individual zones of the growth plate. From Stevens and Williams [11], reprinted with permission.

that resorb about two-thirds of the mineralized septa. Osteoblasts migrate from the adjacent bone marrow and deposit loosely textured woven bone on the surface of the remaining one-third of the mineralized cartilage septa, to form the primary spongiosa [9–11]. Continued minimodeling [14] replaces the cartilage cores with bone to form the secondary spongiosa, and the trabeculae are progressively resorbed so that after a few weeks the metaphysis has been replaced by diaphyseal marrow [15]. As growth slows, the life span of the secondary spongiosa continues to lengthen, and when longitudinal growth ceases, the most recently formed spongiosa becomes the permanent metaphysis. Metaphyseal cancellous bone thus represents a by-product of growth in length. Bones of the central skeleton grow in a somewhat different manner. In the vertebral bodies cancellous bone formed beneath the two growth plates remains in continuity at the centrum, the adult bone resembling the fusion of two epiphyses and two metaphyses with no intervening diaphysis [16]. Unlike the long-bone metaphyses, newly made cancellous bone is not resorbed but persists into adult life and cancellous tissue fills the whole interior of the body. As in the long bones, the cortices grow outward by circumferential periosteal apposition, but unlike the long bones the inner region of the cortex is not completely resorbed, but is converted into cancellous bone that remains in

continuity with the cancellous bone produced at the growth plate [3]. The process of cancellization of the inner cortex, which is an important component of agerelated bone loss, begins during growth in the central skeleton. In the ilium (Fig. 2) the outer cortex behaves as in a vertebral body, but at the inner cortex the periosteal surface is resorbed and the endocortical surface undergoes formation, incorporating some existing trabeculae by compaction [3]. Growth of the pelvis entails movement of the entire ilium away from the midline. The final length of a bone will be influenced by the rate of growth and by the time during which growth continues prior to epiphyseal fusion. The rate of individual bone elongation is governed by the rate of cartilage cell proliferation in the resting zones, which is the primary target of growth hormone and its local mediator insulinlike growth factor-I (IGFI) [17], and growth continues until cartilage cell proliferation ceases, following which the epiphyses fuse [18]. To maintain the ratio of width to length within its narrow range, the rate and duration of periosteal apposition must be linked in some way to the rate and duration of cartilage cell proliferation. This is the main determinant of the final amount of bone in the bone, but several other mechanisms contribute to individual variation. The number of trabeculae formed will depend on the number of capillaries that invade

500

A. M. PARFITT

OUTER

INNER Ps

2y

Ec

20y Ps

8

6

Ec

Ec

4

Ps

Ec

Ps

2

0

mm

FIGURE 2 Model of iliac growth. Diagrammatic representation of iliac growth showing inferred changes in location of periosteal (Ps) and endocortical (Ec) surfaces of outer and inner cortices from age 2 to age 20. Left vertical dashed line indicates initial location of outer Ec surface and extent of expansion in cancellous width. All cancellous bone to the left of this line represents unresorbed cortical bone. Right vertical dashed line indicates final location of inner Ec surface. All initial cancellous bone to the right of this line was incorporated into the cortex by compaction. Reprinted from Parfitt et al. [3] with permission.

the zone of provisional calcification, and the thickness of trabeculae will reflect the balance between apoptotic death and survival of cartilage cells in the hypertrophic zone [19]. During transformation of the primary to the secondary spongiosa, many trabeculae are removed. The extent and timing of this process are genetically determined [20] and influenced by mechanical loading [21]. At the completion of growth, the distance from the articular cartilage to the metaphyseal–diaphyseal junction, which influences the total amount of metaphyseal cancellous bone, is broadly proportional to the length of the bone, but is also subject to individual variation. An important but inadequately studied question is the extent to which individual trabeculae become thicker during skeletal growth. In the peripheral skeleton, the life span of trabeculae is too short for their thickening to contribute to peak bone mass until a few years before growth ceases, but the situation is different in the central skeleton. In iliac bone biopsies in healthy subjects aged 2 to 17 years, there was a progressive increase in trabecular thickness of about 60 µm, the result of rapid remodeling with positive balance [3]. Interpretation of these data is complicated by the changing location of the biopsy site in relation to the acetabulum, but there is no substantial gradient in trabecular thickness in the adult ilium. Consequently, formation of initially thicker trabeculae with increasing age would not account for the data. Although this has not been demonstrated prospectively, it is probable that trabeculae do increase in thickness during growth. This inference is supported

by the modest prepubertal increase in 3D vertebral cancellous bone density determined by QCT, which accelerates during the pubertal growth spurt [22]. Since the region of interest is far from the growth plates, new trabeculae could not be formed and the increase in apparent density could only result from increased thickness of existing trabeculae. The final thickness of diaphyseal cortical bone is the difference between the thickness of bone formed at the periosteum and the thickness of bone resorbed at the endocortical surface. Periosteal apposition depends on osteoblast precursor proliferation, and as previously mentioned is tied to growth plate cartilage cell proliferation, most likely because the same genes control proliferation in response to hormonal stimulation. During the pubertal growth spurt, as well as peak height velocity, there is peak bone width velocity [23], driven by the surges in gonadal hormonal secretion [7] and potentiated by physical activity [5,7]. Bone is also added to the endocortical surface with narrowing of the marrow cavity, the only time in life when this occurs. But gain in bone width and cortical thickness can be partly disassociated from growth in length by other factors. For example, if net absorption of calcium is insufficient to meet the demands of bone growth, net endocortical resorption will increase and cortical width fail to increase or even decrease. But above a certain threshold, further increases in net absorption are not accompanied by increased calcium retention [24], indicating that endocortical resorption cannot be reduced below the level dictated by the growth timetable. No such limitation applies to physical activity, which amplifies all the effects of pubertal growth with increased endocortical as well as periosteal gain in the extremities [7], and increased trabecular thickness in the spine [25].

III. THE PURPOSES OF BONE REMODELING The reasons for bone replacement depend on the function of the bone that is replaced [26]. The primary function of bone is to resist mechanical loads; such structural bone includes all cortical bone, and cancellous bone in the long-bone metaphyses which transmits the loads on the synovial joints to the thick cortical bone above or below. All such bone is adjacent to yellow fatty marrow, in contrast to the red hematopoietic marrow adjacent to central cancellous bone [26]. The two kinds of marrow correspond to two kinds of cancellous bone, referred to as structural and metabolic to indicate their principal but not their only function (Table III); vertebral cancellous bone is load bearing, but contributes less to compressive strength than the cortical bone

CHAPTER 28 Modeling and Remodeling: How Bone Cells Work Together

TABLE III

Significance of Marrow Composition for Cancellous Bone

Feature

Yellow Marrow

Red Marrow

Bone type Location Principal Functions Cellularity Blood flow Turnover

Structural Peripheral Transmit loads Absorb energy Low Low Low

Metabolic Central Calcium homeostasis Support hematopoiesis High High High

surrounding it [27]. Note especially the difference in turnover—cortical bone in the ilium has higher turnover than cancellous bone in the extremities, so that biology trumps geometry [28]! Because structural and metabolic bones differ in their function, they differ in their reasons for replacement (Table IV). Structural bone, both cortical and cancellous, like all load-bearing materials, is subject to fatigue damage after a certain number of load-bearing cycles [29], but unlike manmade structures, it has its own mechanism of self-repair. There may also be nontargeted remodeling to maintain bone age below some upper limit, referred to here as spare, in the sense that its abrogation will not have immediate harmful effects, whereas the abrogation of targeted remodeling may allow fatigue damage to spread and accumulate into an overt fracture [30]. The reasons for the much higher turnover of metabolic cancellous bone are less well understood. The exchange of calcium ions at the bone surface, which is an essential element of plasma calcium homeostasis [31], is impaired by the gradual enlargement of the crystals at the expense of water, and it is likely that highly mineralized bone would need to be replaced. In the ilium, osteocyte viability declines TABLE IV

Reasons for Bone Replacement by Remodeling

Structural bone (cortical, fatty cancellous) Replace fatigue damaged bone — Targeted Prevent bone senescence — “Spare”a Metabolic bone (hematopoietic cancellous) Replace over-mineralized bone — Targeted Preserve osteocyte viability — ? Targeted ? Release of growth factors — “Spare” aSpare remodeling has also been referred to as redundant, surplus, stochastic, or nontargeted [30].

501

with age in interstitial bone but not in surface bone, most likely because its aging is prevented by remodeling [32]. Finally, the controlled release of growth factors stored in the bone may be necessary for optimum hematopoiesis. There are several other functions of remodeling. Bone resorption increases temporarily at night to compensate for the lack of intestinal calcium absorption [33], most likely because existing osteoclasts work a bit harder [31], and there must be some minimum number of osteoclasts for this mechanism to work. At times of temporary increase in calcium demand during growth, pregnancy, and lactation, there is a temporary increase in cortical porosity due to increased remodeling [34]. As mentioned earlier, during the adolescent growth spurt, remodeling accomplishes an increase in the thickness of trabeculae [3], an exception to the general rule that remodeling leads only to conservation or loss of bone. Remodeling discards bone made redundant by disuse [35], and malfunction of this mechanism may be one way in which estrogen deficiency has harmful effects [36]. Finally, remodeling participates in primary fracture healing in cortical bone, creating dowel pins between the two fragments [37].

IV. THE BASIC MULTICELLULAR UNIT AS THE INSTRUMENT OF BONE REMODELING All remodeling is carried out by temporary anatomic structures, first identified by Frost [38] and named by him basic multicellular units (or BMUs), to which, in the adult noninjured skeleton, all osteoclasts and osteoblasts belong. The BMU excavates and refills a tunnel through the bone [39], leaving a new osteon in its wake (Fig. 3). At the front is a team of multinucleated osteoclasts forming the cutting cone; behind them is a capillary that brings circulating monocytes to become preosteoclasts [40], surrounded by loose connective tissue containing nerve fibers and behind that successive teams of osteoblasts forming the closing cone. During its lifespan of many months, the BMU while traveling through the bone maintains the same spatial and temporal relationships between its constituent cells—osteoclasts, vascular endothelial cells, stromal connective tissue cells, and osteoblasts. Trying to understand bone remodeling while ignoring the structure of the BMU is like trying to understand endochondral ossification while ignoring the spatial and temporal relationships between the cells in the growth plate. Histologic sections most often intercept the BMU not longitudinally, as in Fig. 3, but transversely (Fig. 4). Each osteoblast team is assembled on the cement surface

502

A. M. PARFITT

FIGURE 3 Cortical BMU in longitudinal section. A typical BMU in human iliac cortical bone is shown (left panel) with a team of osteoclasts at the bottom, osteoid seams (stained purple) lined with osteoblasts above, and a capillary in between. The entire structure, which is approximately 1 to 2 mm long and 0.2 to 0.4 mm wide, is traveling downward at a rate that can be measured by the longitudinal distance between the tetracycline labels (middle panel) and which in humans is about 25 µm per day [39]. Schematic of coordinated movement in right panel. Middle and right panels reproduced by courtesy of Dr. Robert Schenk. (See color plate).

within a narrow window of time and in a narrow zone between the cutting and closing cones. Each team is like a necklace one cell deep, and about two to three new teams are formed each day. Each team refills the tunnel at a single cross-sectional location as completely as possible. New osteoblasts can join the team only while it is being assembled, not after it has begun to deposit new bone matrix [41]. As one goes further back the teams were born earlier and so are closer to completing their task (Fig. 3, right panel). As it moves through tissue space the BMU creates and leaves behind successive cross-sectional cycles of remodeling, each one slightly out of step with the one before, and at each cross-sectional location resorption moves centrifugally until the cavity is the right size, and formation moves centripetally until refilling of the cavity has been completed at that location. Until recently it was widely believed that the threedimensional organization of the BMU just described applied only to cortical bone, but it is now clear that a

similar organization applies to cancellous bone [39]. A cancellous BMU shows the same relationship between the osteoclastic cutting hemicone in front and the osteoblastic closing hemicone following behind (Fig. 5). Such structures are rarely observed since this requires precise orientation of the randomly generated section in three orthogonal planes, but considerable indirect evidence, reviewed in detail elsewhere [39], indicates that this is the normal, characteristic, manner of cancellous bone remodeling. The process is referred to as hemiosteonal because the BMU excavates a trench across the surface of the bone, rather than a tunnel through the bone, leaving in its wake what may be termed, with some geometrical latitude, a hemiosteon rather than an osteon. Osteonal and hemiosteonal remodeling both display the same relationships between cells that are maintained during BMU progression, the same localization of osteoblast assembly to the narrow zone between the cutting and closing cones or hemicones, and the same dependence on the circulation [42,43],

CHAPTER 28 Modeling and Remodeling: How Bone Cells Work Together

503

FIGURE 4

Cortical BMU in cross section. The bright field on the left shows a large cavity lined by purple staining osteoid within largely unstained mineralized bone. The fluorescence on the right shows that bone formation began eccentrically, as there are two tetracycline labels in the top left corner and only one around the rest of the circumference. The osteoblasts are plump and the osteoid seam (stained purple) is wide. In the lower left is another BMU in which bone formation at this location is almost completed. The cavity is much smaller, the osteoid seam much thinner, the cells much flatter, and the labels much closer together. The cutting cone of the BMU passed through the plane of section about 10 days ago in the right upper location and about 2–3 months ago in the left lower location. From the distribution of partly stained bone of low density that has been recently formed, the newer BMU has arisen adjacent to a not much older BMU in which secondary mineralization has not yet been completed, indicating a state of high bone turnover. Copyright A.M. Parfitt 2001: used with permission. (See color plate).

providing a source for osteoclast precursors and a conduit for ion exchange and intercellular signaling. As the BMU exists and moves in three-dimensional space, it has a beginning (called origination), a middle (called progression), and an end (called termination). Each BMU, representing a new remodeling project, begins on a quiescent surface in response to a target, which is a region of bone in need of replacement [26,30]. The site of origination must be reasonably close to a blood vessel, and in most cases it will be some distance from the target, to which the BMU must progress by excavating a trench in cancellous or a tunnel in cortical bone. Once the target has been reached, it will take some time for the machinery that has been cranked up to be cranked down again, so that the BMU will often progress for some distance beyond the target (Fig. 6). Such posttarget progression provides a convenient basis for spare remodeling, and the distance

traveled will depend on the availability of circulating osteoclast precursors, which are produced in the bone marrow under hormonal control, forming a pool whose size depends on the concentration in the blood [40]. During origination only osteoclasts are needed and during termination only osteoblasts are needed, but during progression, which is much the longest phase, both osteoclasts and osteoblasts are needed at the same time, but at different locations. How a new BMU gets started is still only dimly understood. The first step must be to recognize that some bone needs replacement. In cortical bone, most, perhaps all, remodeling is initiated by fatigue microdamage [44]. The necessary targeting is probably mediated by local death of osteocytes by apoptosis [45,46], although the signals involved are unknown. How, or even whether, targeting is achieved in metabolic bone is unknown [30]. Frost originally

504

A. M. PARFITT

FIGURE 5

Osteonal and hemiosteonal remodeling. (Upper panel) A representative cortical BMU in the dog, with the upper half shaded. (Lower panel) A cancellous BMU in a patient with secondary hyperparathyroidism, in which the spatial organization resembles the bottom half of the upper panel. Although it is possible that the appearance is specific to the particular disease state in which it was observed, reasons are given in [39] and in the text for believing that the structure is characteristic of cancellous bone remodeling in general, but is more likely to be observed when bone turnover is very high because the necessary circumstances will otherwise rarely occur. Called to author’s attention by Dr. David Baylink. Original provided by Dr. Robert Schenk and reproduced with permission. (See color plate).

defined activation as a stimulus to precursor cells to proliferate [1]. It is now often stated that the first step in remodeling is the activation of osteoclasts, but in the normal adult skeleton there are no osteoclasts waiting to be activated. Bone lining cells must be involved in site selection and it is these cells that are activated, to remove the thin endosteal membrane that normally separates them from the mineralized bone, and to retract, providing a foothold for the future BMU [39]. A new vessel then grows toward the site to allow access to circulating mononuclear precursors [40], which are presumably attracted by chemotaxis and fuse to form the first new osteoclasts [47], which begin resorbing in the right direction toward the target. In order for a BMU to progress through the bone or across the surface of the bone several conditions must be satisfied. First, access for circulating preosteoclasts must be maintained. In cortical bone this is achieved by neoangiogenesis, the central capillary keeping up with the cutting cone [42], but in cancellous bone it is

more likely achieved by vasculogenic mimicry, by lining cells that persist as a canopy over the remodeling site, forming a compartment that is in communication with a marrow sinusoid [43]. Second, signals for the arrival of new preosteoclasts must persist, analogous to the continued recruitment of leukocytes to sites of inflammation [46]. Third, the precursor cells that leave the circulation to form new osteoclasts must be replaced from the bone marrow. If all of these conditions are met, new cycles of remodeling, oriented transversely to the direction of BMU advance, will continue to be generated. Failure of one or more of these mechanisms presumably leads to BMU termination. The number of such cycles determines the total distance traveled by the BMU both toward and beyond its target [26,47], and the product of origination frequency and the average number of cycles per BMU corresponds to the histomorphometric index called activation frequency. This is the reciprocal of the average time between the onset of successive cycles of remodeling at the same location,

505

CHAPTER 28 Modeling and Remodeling: How Bone Cells Work Together

WEEKS QUIESCENCE

0

PRE-Oc 1

ORIGINATION

Ob

Oc 16

PROGRESSION

Ob

Oc

TERMINATION

32

QUIESCENCE

48

very old bone

old bone

new bone

FIGURE 6

Cancellous bone remodeling. Evolution of a single BMU through successive stages. BMU is shown as traveling from left to right, excavating a trench across the surface. It originates where a blood vessel arborizes over the bone surface (location denoted by vertical red line) and is directed to remove a region of bone that is too old to carry out its function, whether this function is primarily mechanical, as in the metaphyses of the long bones, or primarily metabolic, as in the axial skeleton [26]. The BMU progresses beyond this target as part of the stochastic or spare component of remodeling. Progression is sustained by the continued recruitment of new preosteoclasts (PRE-Oc), which need to be precisely targeted to the apex of the cutting hemicone, adjacent to the lining cells that cover the bone that is about to be resorbed. Termination occurs when the supply of PRE-Oc is turned off, but osteoblast recruitment continues until the trench is refilled. For further details see text. Copyright A. M. Parfitt (1995); used with permission. See color plate.

which is about 3 years in central cancellous bone and about 10 to 20 years at sites where bone turnover is lower. Activation frequency, which is a two-dimensional concept, is the best histologic index of the overall intensity of bone remodeling, but it is not a measure of the birth rate of new BMU, which is a three-dimensional concept. For example, one BMU that progresses for nine units of distance will have exactly the same effect on all histologic, radiokinetic, and biochemical indices of bone turnover as three BMUs that each progress for only three units of distance [26,47], but the biological significance will be quite different. The former represents a low demand for remodeling with a large surplus component, the latter a high demand for remodeling that is efficiently satisfied. Endocrine abnormalities such as estrogen deficiency, or parathyroid or thyroid hormone

excess, that increase the availability of osteoclasts can increase the extent of post target progression without increasing BMU origination, if the demand for remodeling has not changed [48]. Hormones are indifferent to bone structure and have no means of specifying where a new BMU will be formed. Bone remodeling resembles hematopoiesis in several respects. Both are examples of bone marrow cell renewal, in which changes in the number of functioning cells are more important than changes in individual cell activity. In both, the executive cells have a short life span and need to be continually replaced. In both, a basal rate of cell production can be increased on demand; in both, cell number depends not only on cell production but on the timing of cell death, which determines cell life span; and in both, dysregulation of cell number will lead to disease. The life span of the BMU

506 comprising separate stages of origination, progression, and termination, is measured in months, but the life span of osteoblasts while they are making bone is measured in weeks, and the life span of osteoclast nuclei is measured in days. All cells that originate in the marrow die after a rather short time, and all osteoclasts die by apoptosis [47], but osteoblasts, derived from local stromal cells, have a more complex fate [41]. About two-thirds die by apoptosis, so that the density of cells on the surface progressively declines, a process augmented by the incorporation of about 30% of the initial number of osteoblasts into bone as osteocytes. Because each osteoblast has to cover a progressively larger area, they become flatter and thinner and at the termination of bone formation at a particular location about 5% of the original number persist as the lining cells that cover all quiescent bone surfaces. Osteocytes may survive for decades, but eventually will either be removed as the bone is remodeled or themselves die by apoptosis [32].

V. DISORDERED REMODELING AND AGE-RELATED BONE LOSS Age-related bone loss is a near-universal characteristic of the human species, from which no group is known to be exempt. But whatever the proximate causes such as physical inactivity, altered nutrition, or sex hormone deficiency, and whatever the genetic susceptibilities or molecular mechanisms, all bone loss must be accountable for in terms of disordered remodeling. But first the distinction between its reversible and irreversible components must be clarified. Reversible bone loss is an inevitable consequence of an increase in bone remodeling, most clearly exemplified by the changes in rib cortical porosity during the antler growth cycle in deer, referred to as cyclic physiologic osteoporosis [49,50]. Some of the calcium needed for the antlers is borrowed from the skeleton by starting up a large number of new BMUs, and when they have run their course the debt is fully repaid. The aggregate volume of all the new holes is the remodeling space, but the reversible deficit also includes unmineralized bone matrix or osteoid, which has a short life span, and incompletely mineralized matrix in young recently formed bone, which has a long life span [34]. Because completion of mineralization takes a long time, mineral density increases steadily with the age of the bone. Increased remodeling reduces mean bone age and mean mineral density (Fig. 4) with converse changes when normal remodeling is restored. All three components of the reversible deficit contribute to temporary calcium needs during growth, pregnancy, and lactation [50].

A. M. PARFITT

It is useful to consider the mechanisms of irreversible bone loss in terms of the sequential changes at a single cross-sectional location that are initiated by a BMU as it travels perpendicular to the plane of section, even though this is an inaccurate description of the bone remodeling process in three dimensions (Fig. 7). The cycle begins with a quiescent surface covered by flat lining cells, continues with activation of the lining cells and recruitment of preosteoclasts, resorption to a certain depth, the reversal phase after the termination of resorption, osteoblast recruitment and bone formation, and in the ideal situation in a young healthy adult, complete refilling of the cavity and restoration of the bone surface to its previous location [51]. Because of the perpendicular movement of the bone surface, I call this the down and up model. In terms of this model, all irreversible bone loss is the result of focal imbalance between the depth of a resorption cavity and the depth of the new bone deposited within the cavity, which corresponds to the histologic measurement of wall thickness (Fig. 8). This focal imbalance can result either because the cavity is too deep or because the new bone is too shallow. The bone loss is irreversible because each cycle of remodeling, representing events within a two-dimensional slice of bone generated by the movement of the BMU in three dimensions, constitutes a transaction that once completed cannot be revoked, even though future transactions may be positive in response to treatment. Because of the three-dimensional organization of the BMU, bone resorption can be partitioned into horizontal and vertical components (Fig. 9). The horizontal component is the distance traveled by the BMU across the surface, which depends on the supply of osteoclast precursor cells, both production in the marrow and delivery by the circulation. The vertical component is the depth of the cavity, of which the main determinant is the timing of osteoclast apoptosis. Earlier apoptosis means a shallower cavity, later apoptosis a deeper cavity, with consequent perforation of trabecular plates and cancellization of the inner third of the cortex [52]. The timing of apoptosis is independent of birth rate, since delayed apoptosis can occur in the early stages of estrogen deficiency, in which osteoclast production is increased [53], or in glucocorticoid excess, in which it is decreased [54]. Unlike bone resorption, bone formation has only a vertical component, but like bone resorption, it is influenced by cell death as well as by cell birth. At each cross-sectional location, new osteoblasts assemble on the cement surface at the floor of the resorption cavity. They are columnar in shape and closely packed, and become progressively fewer, flatter, and less active until those remaining become lining cells (Fig. 4). In the elderly,

507

CHAPTER 28 Modeling and Remodeling: How Bone Cells Work Together

2. Activation Pre Oc.

3. Resorption 1. Quiescence

Oc.

Li.Ce. E.Lc. Old bone New bone 4. Reversal

Li.Ce. B.St.U.

Osteoid

?

Cm.Ln. Cm.Ln. 6. Quiescence

5. Formation Ob.

Cm.Ln.

FIGURE 7 Diagrammatic representation of the remodeling cycle in cancellous bone. Successive stages of quiescence, activation, resorption, reversal, formation, and back to quiescence at a single cross-sectional location generated by the movement of the BMU perpendicular to the plane of section are depicted, the so called down and up model. Li.Ce., Lining cell; Pre.Oc., preosteoclast; Oc., osteoclast; E.Lc., eroded lacuna; Cm.Ln., cement line; Ob., osteoblast; B.St.U., bone structural unit (hemiosteon). Refilling is assumed to be complete, and bone marrow lying above the lining cells is omitted for clarity. Reproduced from Parfitt [51] with permission of the publisher.

each team of osteoblasts on the cancellous surface makes less bone, partly because there are fewer initial members of the team and partly because some die sooner, which leads to trabecular and cortical thinning [52]. This can be the result of physical inactivity, estrogen deficiency, glucocorticoid excess, or aging per se. All bone loss occurs from an internal surface, one of the three subdivisions of the endosteal envelope, in contrast to the periosteal envelope, which may slowly gain bone, so that the decline in total bone mass is always smaller than the amount lost [55]. The absolute rate of bone loss from a surface, expressed as volume of bone per unit surface per unit time, depends on the average amount lost in each cycle of remodeling and how often the cycles occur (Fig. 10), which is why there is a general relationship between the rate of bone loss and the rate of bone turnover [56]. The rate of loss expressed as a percentage of the initial amount also depends on the surface-to-volume ratio, which is why the relative rate is slower in cortical than in cancellous bone, even though the absolute amount lost is

greater [57]. The absolute amount lost from a bone depends also on the absolute extent of surface, which declines with age in cancellous bone because whole structural elements are removed, and increases with age in cortical bone because subendocortical cavities become larger [52,58], which is why the fractional rate of loss tends to fall with age in cancellous bone but may increase with age in cortical bone [59]. A remarkable feature of age-related bone loss is its universality, affecting not only almost every person but almost every bone. Although different skeletal sites may lose bone at different rates in the short term, the wider the age range over which the data are collected, the more similar the rates become. For both central and peripheral sites, comprising various proportions of cortical and cancellous bone, the long term rates of bone loss measured cross sectionally are in the range of 1–1.5%/y [26]. Most bone lost with age is cortical and cortical, thinning is mainly the result of increased resorption depth [60,61], which is the two-dimensional reflection of deeper penetration by endocortical BMUs.

508

A. M. PARFITT

Normal – Focal Bone balance

Osteoclast mediated Bone loss

Osteoblast mediated Bone loss

Old bone

New bone

FIGURE 8 Possible mechanisms of focal remodeling imbalance. The upper panel shows a normal-depth resorption cavity on the left, completely refilled with new bone on the right. The middle panel shows a resorption cavity of excessive depth that is incompletely refilled by a normal amount of new bone. The bottom panel shows a resorption cavity of normal depth that is incompletely refilled by a subnormal amount of new bone. Note that the amount of bone lost as a result of the cycle, indicated by the clear area between the original surface location denoted by the interrupted line, and the new surface location, can be the same even though the cellular mechanism is quite different. Modified from Parfitt [52] with permission of the publisher.

Because the rates of fractional loss are so similar, the increase in resorption depth at different sites must be inversely related to the customary rate of turnover and positively related to the usual thickness of cortical bone at each site [26]. When bone loss is both generalized and sustained, as in normal aging, it appears that resorption depth at different sites increases to the extent necessary to bring about much the same rates of fractional bone loss, and to adjust for differences in bone turnover contingent on differences in marrow composition and for differences in local bone structure and geometry [26]. The only conceivable kind of explanation for such a phenomenon is biomechanical. All mechanical influences on bone remodeling are mediated by strain, the technical term for relative deformation of a structural material as the result of load bearing. Similar fractional rates of bone loss throughout the skeleton will produce similar proportional changes in the strains that occur in different bones as the result of the same pattern and intensity of

physical activity. The recruitment and activity of osteoclasts and osteoblasts are orchestrated by the mechanostat in such a way that strain is maintained within an acceptable range [36]. As a result of the sedentary lifestyle made possible by economic development, aging is in most persons accompanied by a progressive reduction in physical activity and muscle strength, of earlier onset and greater severity than is biologically mandated [62]. The risk of fracture should not be increased, as the reduced bone mass would remain appropriate to the reduced level of activity, but this does not take into account the age-related increase in the liability to falls, to which the mechanostat is blind. If this were the sole explanation for age-related bone loss, its magnitude should have been less in more physically active populations, but perhaps the mechanostat is reset, so that the bones respond not to actual but to erroneously perceived disuse [36], either because of estrogen deficiency, or as a consequence of the aging process itself [26].

509

CHAPTER 28 Modeling and Remodeling: How Bone Cells Work Together

(Direction of BMU progression)

Precursor cell supply (production and delivery)

Extent

Oc

Ob

Timing of apoptosis

Depth

FIGURE 9

Two aspects of bone resorption. A. BMU is traveling across the cancellous surface from left to right. At the apex the youngest osteoclast abuts the lining cell covering the surface that is about to be resorbed. The extent of resorption in the direction of progression depends on maintaining the supply of mononuclear osteoclast precursors that are produced in the bone marrow and delivered to the resorption site by the circulation. The depth of resorption, which determines the magnitude of the task set for the osteoblasts, depends on the timing of apoptosis. Copyright A. M. Parfitt, 2001, used with permission. See CD-ROM for color.

VI. POSSIBLE TARGETS FOR VITAMIN D ACTION

Acf.N

Acf X2

Acf X0.5

0

3 Time (years)

6

FIGURE 10 Remodeling determinants of bone loss. In normal adult human cancellous bone one cycle occurs about once every 3 years (upper panel). If the activation frequency is doubled, there will be four cycles in 6 years instead of two (middle panel), and if it is halved, there will be only one cycle (lower panel), with corresponding differences in the cumulative bone loss, indicated by the distances between the interrupted and solid lines after 6 years, even if the amount lost in each cycle remains constant. Differences in the magnitude of focal imbalance will lead to proportional differences in the rate of bone loss if there is no change in activation frequency. Copyright A. M. Parfitt, 2001, used with permission.

Vitamin D is essential for the mineralization of growth plate cartilage and osteoid as described in detail in Chapter 63. Apart from defective mineralization, the most consistent effect of vitamin D deficiency on bone remodeling is increased turnover due to secondary hyperparathyroidism, of which the pathogenesis is discussed in Chapter 30. Parathyroid hormone (PTH) increases osteoclast formation and activation frequency by mechanisms that are beyond the scope of this chapter [53], and this occurs despite the important role of calcitriol in osteoclastogenesis (Chapter 38). Serum calcitriol levels are normal or even increased in most patients with vitamin D deficiency [63] because the stimulatory effects of PTH, hypocalcemia, and hypophosphatemia on the renal 1α-hydroxylase more than offset the effect of substrate deficiency. Although bone formation rate is increased on all surfaces, the most clinically important effect is accelerated loss of cortical bone in both the peripheral and central skeleton, as a result of increased depth of resorption cavities on the

510 endocortical surface [64]. Further loss can be prevented by returning PTH secretion to normal by an appropriate combination of vitamin D and calcium. This contributes to the reduction in hip fracture risk by vitamin D [65], although improved muscle strength may be even more important (Chapter 102) [66]. Agents that increase osteoclast production in vitro, such as PTH, generally increase activation frequency and bone turnover in vivo, but such an effect has not been demonstrated for calcitriol. This could be because PTH secretion is inhibited (both directly and indirectly by an increase in plasma calcium), but calcitriol did not increase biochemical indices of bone turnover when given to patients with hypoparathyroidism [67]. PTH had its expected effects on bone turnover, and both treatments were effective in restoring normocalcemia, in part by increasing the level at which blood and bone equilibrate at quiescent bone surfaces [31]. A similar dissociation between bone turnover and calcemia was found with vitamin D treatment of hypoparathyroidism [68]. Why calcitriol and PTH, which have similar effects on osteoclast production in vitro, should have such dissimilar effects on bone turnover in vivo is an important topic for future research. Its importance resides not so much in the specific issue, but in the more general question of the relevance of in vitro studies to whole organism pathophysiology [69]. When vitamin D deficiency and secondary hyperparathyroidism occur in the context of intestinal malabsorption there is cancellous as well as cortical osteopenia [70]. This could be due to the added effect of calcium deficiency, although there may also be deficiency of some other nutrient not yet recognized as essential for bone health. A third possibility is that vitamin D deficiency might contribute more directly to the pathogenesis of osteoporosis. Since circulating calcidiol is a substrate for the extra renal synthesis of calcitriol in bone [71] and calcitriol has effects on osteoblasts such as stimulating the release of osteocalcin [72,73], it is conceivable that vitamin D deficiency could compromise osteoblast function even in the absence of secondary hyperparathyroidism, leading to incomplete refilling of resorption cavities and thinning of trabeculae to a greater extent than expected from aging alone. Although such a mechanism is speculative, it is in agreement with evidence that the level of vitamin D nutrition that is optimal for bone health is higher than is currently believed (Chapters 61,62) [74]. The hypercalcemia of vitamin D intoxication is often ascribed to increased intestinal calcium absorption together with reduced glomerular filtration rate, but effects of vitamin D on bone are also implicated. As mentioned earlier, calcitriol potentiates the effects

A. M. PARFITT

of PTH on setting the level of blood–bone equilibration [31], but whether this contributes to hypercalcemia is unclear. Increased bone resorption has been inferred in patients with vitamin D intoxication from several lines of evidence, including increased urinary hydroxyproline excretion [75], increased fasting urinary calcium excretion [76,77], negative calcium balance [78,79], and the response to bisphosphonate administration [76,77,80], but with minimal histologic verification. The single bone biopsy showed a sevenfold increase in osteoclast number but no increase in turnover; rather, there was defective mineralization and osteoid accumulation [75]. Serum levels of 25OHD are as high as 20 times normal, and levels of 1,25(OH)2D are about two to three times the normal mean [75,77]. In one case there was a close relationship between changes in plasma calcium and 1,25(OH)2D levels in response to glucocorticoid administration [75], but increased bone resorption has also occurred in an anephric patient [81]. 25-Hydroxy D levels may be high enough to activate calcitriol receptors in bone or to serve as substrate for the synthesis of calcitriol in bone; the metabolite responsible may differ between patients. Whether the increase in osteoclast number in a single case was due to increased recruitment or increased life span is unclear. The absence of increased turnover in this and other cases is more in keeping with increased activity of existing osteoclasts, which is a plausible in vivo expression of the in vitro effects of calcitriol on bone resorption (Chapter 32).

VII. CONCLUSIONS The remodeling system is highly complex [1,53] but its essential elements are few in number. First, for the primary purpose of damage repair, new BMUs must be initiated promptly and must travel rapidly in the right direction. Second, remodeling influences bone mass by the interplay of activation frequency, resorption depth, and wall thickness in different regions of the skeleton. Each of these elements depends on the recruitment, life span, and activity of teams of osteoclasts or osteoblasts, coordinated in part by signals from lining cells and osteocytes. It is regrettable that 80 years after the discovery of vitamin D and 30 years after its metabolism was elucidated, nothing definite is known about the direct effects of vitamin D or of any of its metabolites on any of the elements just enumerated—the necessary experiments have just not been conducted. A similar dearth of knowledge applies to most pharmaceutical agents [82]. This defect will not be remedied until there is more general recognition

CHAPTER 28 Modeling and Remodeling: How Bone Cells Work Together

that the role of disordered remodeling in the pathophysiology of metabolic bone disease depends on how the system actually works, and cannot be captured by the mantra-like repetition of “bone resorption” and “bone formation.”

20. 21.

References 1. Frost HM 1973 Bone remodeling and its relationship to metabolic bone diseases. Charles C. Thomas, Springfield, IL. 2. Parfitt AM 1997 Genetic effects on bone mass and turnoverrelevance to black/white differences. J Am Coll Nutr 16:325–333. 3. Parfitt AM, Travers R, Rauch F, Glorieux FH 1992 Structural and cellular changes during bone growth in healthy children. Bone 27:487–494. 4. Balena R, Shih M-S, Parfitt AM 1992 Bone resorption and formation on the periosteal envelope of the ilium: A histomorphometric study in healthy women. J Bone Miner Res 7:1475–1482. 5. Rauch F, Schoenau E. The developing bone: Slave or master of its cells and molecules? Pediatr Res 50:309–314. 6. Gardner E 1971 Osteogenesis in the human embryo and fetus. In: Bourne H (ed) The Biochemistry and Physiology of Bone, Vol. III, Development and Growth. Academic Press, New York, pp. 77–118. 7. Parfitt AM 1994 The two faces of growth: benefits and risks to bone integrity. Osteopor Int 4:382–398. 8. Sissons HA 1971 The growth of bone. In: Bourne GH (ed) The Biochemistry and Physiology of Bone, Vol. III. Development and Growth. Academic Press, New York, pp. 145–180. 9. Brighton CT 1978 Structure and function of the growth plate. Clin Orthop Rel Res 136:22–32. 10. Schenk RK, Hunziker EB 1991 Growth plate: Histophysiology, cell and matrix turnover. In: Glorieux FH (ed) Rickets, Nestle Nutrition Workshop Series Vol. 21. Raven Press, New York, pp. 63–78. 11. Stevens DA, Williams GR 1999 Hormone regulation of chondrocyte differentiation and endochondral bone formation. Mol Cell Endocrinol 151:195–204. 12. Kember NF, Kirkwood JR 1991 Cell kinetics and the study of longitudinal bone growth: a perspective. In: Dixon AD, Sarnat BG, Hoyte DA (eds) Fundamentals of Bone Growth: Methodology and Applications. CRC Press, Boca Raton, FL, pp. 153–162. 13. Boskey AL, Doty SB, Stiner D, Binderman I 1996 Viable cells are a requirement for in vitro cartilage calcification. Calcif Tissue Int 58:177–185. 14. Frost HM 1990 Structural adaptations to mechanical usage. 1. Redefining Wolff’s law: The bone remodeling problem. Anat Rec 226: 403–413. 15. Turner RT 1994 Cancellous bone turnover in growing rats: Time-dependent changes in association between calcein label and osteoblasts. J Bone Miner Res 9:1419–1424. 16. Doskocil EN, Valouch P, Pazderka V 1993 On vertebral body growth. Funct Dev. Morphol 3:149–155. 17. Daughaday WH, Rotwein P 1989 Insulin-like growth factors I and II: peptide, messenger ribonucleic acid and gene structures, serum and tissue concentrations. Endo Rev 10:68–91. 18. Parfitt AM 2002 Misconceptions (1): Epiphyseal fusion causes cessation of growth. Bone 30:337–339. 19. Amling M, Neff L, Tanaka S, Inoue D, Kuida K, Weir E, Philbrick WM, Broadus AE, Baron R 1997 Bcl-2 lies downstream

22. 23. 24. 25.

26.

27. 28. 29.

30. 31.

32. 33.

34. 35.

36. 37. 38. 39. 40.

511

to parathyroid hormone–related peptide in a signaling pathway that regulates chrondrocyte maturation during skeletal development. J Cell Biol 136:205–213. Schaffler MB, Wenzel TE, Fazzalari NL, Fyhrie DP 1993 Quantitative characteristics of cancellous bone morphogenesis. Trans Orthop Res Soc 18:150. Frost HM, Jee WSS 1994 Perspectives: A vital biomechanical model of the endochondral ossification mechanism. Anat Rec 240:435–446. Gilsanz V, Gibbens DT, Roe TF, et al. Vertebral bone density in children: Effect of puberty. Radiology 166:847–850. Garn SM 1970 The earlier gain and the later loss of cortical bone. Charles C. Thomas, Springfield, IL. Matkovic V, Heaney RP 1991 Calcium balance during human growth: Evidence for threshold behavior. Am J Clin Nutr 55:992–996. Block JE, Friedlander AL, Brooks GA, Steiger P, Stubbs HA, Genant HK 1989 Determinants of bone density among athletes engaged in weight-bearing and non-weight-bearing activity. J Appl Physiol 67:1100–1105. Parfitt AM. Skeletal heterogeneity and the purposes of bone remodeling: implications for the understanding of osteoporosis. In: Marcus R, Feldman D, Kelsey J (eds) Osteoporosis, 2nd ed. Academic Press, San Diego, pp. 433–447. Mazess RB 1990 Fracture risk: a role for compact bone. Calc Tiss Int 47:191–193. Parfitt AM 2002 Misconceptions (2): Turnover is always higher in cancellous than in cortical bone. 30:807–809. Burr DB, Forwood MR, Fyhrie DP, Martin RB, Schaffler MB, Turner CH 1997 Bone microdamage and skeletal fragility in osteoporotic and stress fractures. J Bone Miner Res 12: 6–15. Parfitt AM 2002 Targeted and non-targeted bone remodeling: Relationship to BMU origination and progression. Bone 30:5–7. Parfitt AM 1993 Calcium homeostasis. In: Mundy GR, Martin TJ (eds) Handbook of Experimental Pharmacology, Vol. 107. Physiology and Pharmacology of Bone. Springer-Verlag, Heidelberg, pp. 1–65. Qiu S, Palnitkar S, Rao DS, Parfitt AM 2002 Age and distance from the surface but not menopause reduce osteocyte variability in human cancellous bone. Bone 31:313–318. Lakatos P, Blumsohn A, Eastell R, Tarjan G, Shinoda H, Stern PH 1995 Circadian rhythm of in vitro bone-resorbing activity in human serum. J Clin Endocrinol Metab 80: 3185–3190. Parfitt AM 1980 Morphologic basis of bone mineral measurements. Transient and steady state effects of treatment in osteoporosis [Editorial]. Miner Electrolyte Metab 4:273–287. Parfitt AM 1990 Pharmacologic manipulation of bone remodelling and calcium homeostasis. In: Kanis JA (ed) Progress in Basic and Clinical Pharmacology, Vol. 4, Calcium Metabolism. S. Karger, Basel, pp. 1–27. Frost HM 1996 Perspectives: A proposed general model of the “mechanostat.” Anat Rec 244:139–147. Schenk R 1987 Cytodynamics and histodynamics of primary bone repair. In: Lane JM (ed) Fracture Healing. Churchill Livingstone, Edinburgh, pp. 23–32. Frost HM 1969 Tetracycline-based histological analysis of bone remodeling. Calcif Tissue Res 3:211–237. Parfitt AM 1994 Osteonal and hemi-osteonal remodeling: The spatial and temporal framework for signal traffic in adult human bone. J Cell Biochem 55:273–286. Parfitt AM 1998 Mini-Review. Osteoclast precursors as leucocytes: Importance of the area code. Bone 23:491–494.

512 41. Parfitt AM 1990 Bone-forming cells in clinical conditions. In: Hall BK (ed) Bone: A Treatise, Vol. 1, The Osteoblast and Osteocyte. Telford Press, Caldwell, NJ, pp. 351–429. 42. Parfitt AM 2000 Mini-Review. The mechanism of coupling— A role for the vasculature. Bone 26:319–323. 43. Parfitt AM 2001 The bone remodeling compartment: A circulatory function for bone lining cells. J Bone Miner Res 16:1583–1585. 44. Martin RB 2002 Is all cortical bone remodeling initiated by microdamage? Bone 30:8–13. 45. Verborgt O, Gibson GJ, Schaffler MB 2000 Loss of osteocyte integrity in association with microdamage and bone remodeling after fatigue in vivo. J Bone Miner Res 15:60–67. 46. Noble BS, Peet N, Stevens HY, Brabbs A, Mosley JR, Reilly GC, Reeve J, Skerry TM, Lanyon LE 2003 Mechanical loading: Biphasic osteocyte survival and targeting of osteoclasts for bone destruction in rat cortical bone. Am J Physiol Cell Physiol 284:C934–C943. 47. Parfitt AM, Mundy GR, Roodman GD, Hughes DE, Boyce B 1996 A new model for the regulation of bone resorption, with particular reference to the effects of bisphosphonates. J Bone Miner Res 11:150–159. 48. Parfitt AM 1996 Hormonal influences on bone remodeling and bone loss—application to the management of primary hyperparathyroidism. Ann Intern Med 125:413–415. 49. Banks WJ, Epling GP, Kainer RA, Davis RW 1968 Antler growth and osteoporosis. I. Morphological and morphometric changes in the costal compacta during the antler growth cycle. Anat Rec 162:387–398. 50. Parfitt AM 1981 Integration of skeletal and mineral homeostasis. In: DeLuca HF, Frost H, Jee W, Johnston C, Parfitt AM (eds) Osteoporosis: Recent Advances in Pathogenesis and Treatment. University Park Press, Baltimore, pp. 115–126. 51. Parfitt AM 1993 Morphometry of bone resorption: Introduction and overview. Bone 14:435–441. 52. Parfitt AM 1988 Bone remodeling: Relationship to the amount and structure of bone and the pathogenesis and prevention of fractures. In: Riggs BL, Melton LJ (eds) Osteoporosis— Etiology, Diagnosis and Management. Raven Press, New York, pp. 45–94. 53. Manolagas SC 2000 Birth and death of bone cells: basic regulatory mechanisms and implications for the pathogenesis and treatment of osteoporosis. Endocr Rev 21:115–137. 54. Weinstein RS, Jilka RJ, Parfitt AM, Manolagas SC 1998 Inhibition of osteoblastogenesis and promotion of apoptosis of osteoblasts and osteocytes by glucocorticoids. Potential mechanisms of their deleterious effects on bone. J Clin Invest 102:274–282. 55. Duan Y, Turner CH, Kim BT, Seeman E 2001 Sexual dimorphism in vertebral fragility is more the result of gender differences in age-related bone gain than bone loss. J Bone Miner Res 16:2267–2275. 56. Garnero P, Delmas PD 2001 Biochemical markers of bone turnover in osteoporosis. In: Marcus R, Feldman D, Kelsey J (eds) Osteoporosis, 2nd ed. Academic Press, New York, pp. 459–477. 57. Han Z-H, Palnitkar S, Rao DS, Nelson D, Parfitt AM 1996 Effect of ethnicity and age or menopause on the structure and geometry of iliac bone. J Bone Miner Res 11:1967–1975. 58. Martin RB 1972 The effects of geometric feedback in the development of osteoporosis. J Biomechanics 5:47–455. 59. Ensrud KE, Palermo L, Black DM, Cauley J, Jergas M, Orwoll ES, Nevin MC, Fox KM, Cummings SR 1995 Hip and calcaneal bone loss increase with advancing age: longitudinal

A. M. PARFITT

60.

61.

62. 63.

64. 65.

66. 67.

68. 69.

70.

71.

72.

73.

74.

results from the study of osteoporotic fractures. J Bone Miner Res 10:1778–1787. Han Z-H, Palnitkar S, Rao D Sudhaker, Nelson D, Parfitt AM 1997 Effects of ethnicity and age or menopause on the remodeling and turnover of iliac bone: Implications for mechanisms of bone loss. J Bone Miner Res 12:498–508. Parfitt AM 1990 The three organizational levels of bone remodeling: implications for the interpretation of biochemical markers and the mechanisms of bone loss. In: Christiansen C, Overgaard K (eds) Osteoporosis. Proceedings of Third International Symposium. Osteopress ApS, Denmark, pp. 429–434. Frost HM 1997 On our age-related bone loss: Insights from a new paradigm. J Bone Miner Res 12:1539–1546. Parfitt AM 1998 Osteomalacia and related disorders. In: Avioli LV, Krane SM (eds) Metabolic Bone Disease and Clinically Related Disorders, 3rd ed. Academic Press, San Diego, pp. 327–386. Parfitt AM 2003 Renal bone disease—A new conceptual framework for the interpretation of bone histomorphometry. Curr Opin Nephrol Hyperten, 12:387–408. Chapuy MC, Pamphile R, Paris E, Kempf C, Schlichting M, Arnaud S, Garnero P, Meunier PJ 2002 Combined calcium and vitamin D3 supplementation in elderly women: confirmation of reversal of secondary hyperparathyroidism and hip fracture risk: the Decalyos II study. Osteoporosis Int 13:257–264. Janssen HC, Samson MM, Verhaar HJ 2002 Vitamin D deficiency, muscle function and falls in elderly people. Am J Clin Nutr 75:611–615. Winer KK, Yanovski JA, Cutler GB 1996 Synthetic human parathyroid hormone 1-34 vs calcitriol and calcium in the treatment of hypoparathyroidism: Results of a short-term randomized crossover trial. JAMA 276:631–636. Langdahl BL, Mortensen L, Vesterby A, Eriksen EF, Charles P 1996 Bone histomorphometry in hypoparathyroid patients treated with vitamin D. Bone 18:103–108. Parfitt AM 1995 Problems in the application of in vitro systems to the study of human bone remodeling. Proceedings of the Workshop on Human Models of Skeletal Aging, NIH, February 1994. Calcif Tissue Int 56(Suppl 1):S5–S7. Parfitt AM 2001 The evolution of vitamin D related bone disease: The importance of an early stage of increased bone turnover without impaired mineralization. In: Burkhardt P, Heaney R, Dawson-Hughes B (eds) Nutritional Aspects of Osteoporosis. Academic Press, San Diego, pp. 197–208. Howard GA, Turner RT, Sherrard DJ, Baylink DJ 1981 Human bone cells in culture metabolize 25-hydroxy vitamin D, to 1,25-dihydroxyvitamin D, and 24,25-dihydroxyvitamin D. J Biol Chem 256:7738–7740. Duda RJ Jr, Kumar R, Nelson KI, Zinsmeister AR, Mann KG, Riggs BL 1987 1,25-Dihydroxyvitamin D stimulation test for osteoblast function in normal and osteoporotic postmenopausal women. J Clin Invest 79:1249–1253. Inaba M, Terada M, Koyama H, Yoshida O, Ishimura E, Kawagishi T, Okuno Y, Nishizawa Y, Otani H, Morii H 1995 Influence of high glucose on 1,25-dihydroxyvitamin D3–induced effect on human osteoblast-like MG-63 Cells. J Bone Miner Res 10:1050–1056. Vieth R 2001 Would prehistoric human 25-hydroxyvitamin D concentrations be beneficial, and how much vitamin D do we need to ensure desirable nutritional targets? In: Burckhardt P, Dawson-Hughes B, Heaney RP (eds) Nutritional Aspects of Osteoporosis. Academic Press, San Diego, pp. 173–295.

CHAPTER 28 Modeling and Remodeling: How Bone Cells Work Together

75. Davies M, Mawer EB, Freemont AJ 1986 The osteodystrophy of hypervitaminosis D: a metabolic study. Quart J Med 61:911–919. 76. Rizzoli R, Stoermann C, Ammann P, Bonjour JP 1994 Hypercalcemia and hyperosteolysis in vitamin D intoxication: Effects of clodronate therapy. Bone 15:193–198. 77. Selby PL, Davies M, Marks JS, Mawer EB 1995 Vitamin D intoxication causes hypercalcaemia by increased bone resorption which responds to pamidronate. Clin Endocrinol 43:531–536. 78. Streck WF, Waterhouse C, Haddad JG 1979 Glucocorticoid effects in vitamin D intoxication. Arch Intern Med 139: 974–977.

513

79. Down PF, Polak A, Regan RJ 1979 A family with massive acute vitamin D intoxication. Postgrad Med J 55:897–902. 80. Lee DC, Lee GY 1998 The use of pamidronate for hypercalcemia secondary to acute vitamin D intoxication. J Toxicol Clin Toxicol 36:719–721. 81. Counts SJ, Baylink DJ, Shen FH, Sherrard DJ, Hickman RO 1975 Vitamin D intoxication in an anephric child. Ann Int Med 82:196–200. 82. Parfitt AM. Idiosyncratic comments on the state of knowledge in osteoporosis, with particular emphasis on its limitations. In: Christiansen C, Overgaard K (eds) Osteoporosis. 1990 Third International Symposium. Osteopress ApS, Denmark, pp. 1845–1851.

CHAPTER 29

Vitamin D and the Kidney PETER TEBBEN AND RAJIV KUMAR

Departments of Medicine, Biochemistry and Molecular Biology and Mayo Proteomics Research Center, Divisions of Nephrology, and Endocrinology, Diabetes, Metabolism, and Nutrition, Mayo Clinic and Foundation, Rochester, Minnesota

I. Introduction II. Role of the Kidney in the Metabolism of 25OH D III. Effects of Vitamin D, 25(OH)D3, and 1,25(OH)2D3 on the Renal Handling of Calcium and Phosphorus

IV. Distribution and Regulation of Vitamin D–Dependent Proteins in the Kidney V. Conclusion References

I. INTRODUCTION

7. Finally, the kidney has an equally important role in the control of plasma phosphate and the filtration and reabsorption of phosphate. As in the case of calcium, the reabsorption and secretion of phosphate are under hormonal control and many of the same hormones and factors involved in calcium regulation play a significant role in the regulation of phosphate reabsorption [8–13].

The kidney has a unique function in mineral homeostasis and plays a vital role in the control of plasma calcium and phosphorus. While examining how the kidney controls calcium and phosphate homeostasis, it is worthwhile to keep the following facts in mind: 1. In humans, in a 24-hr period, about 8 g of calcium is filtered at the glomerulus, and about 7.8 g is reabsorbed in the proximal and distal tubules and the loop of Henle [1–5]. This is carried out in a manner such that, under normal circumstances, i.e., in states of neutral calcium balance, the amount of calcium in the urine closely approximates that absorbed in the intestine. The mechanisms by which calcium is reabsorbed in the kidney are complex and yield several insights into the cellular regulation of calcium transport. 2. The reabsorption of calcium in the kidney is controlled by several factors such as the filtered load of sodium, urine flow, and the activity of several hormones, most notably parathyroid hormone (PTH), 1,25-dihydroxyvitamin D (1,25(OH)2D3), and calcitonin, in addition to others [1–5]. 3. The kidney is the major site of synthesis of 1,25(OH)2D3, the active and hormonal form of vitamin D [6,7]. 4. The kidney expresses several vitamin D–dependent proteins such as the plasma membrane calcium pump (PMCa), the epithelial calcium channel (ECaC), the sodium calcium exchanger, and the calbindins, some of which play a vital role in calcium transport. 5. The kidney expresses the vitamin D receptor (VDR). 6. The kidney expresses 25-hydroxyvitamin D3 1αhydroxylase (1α-hydroxylase) and 1,25-dihydroxyvitamin D3-24-hydroxylase (24-hydroxylase), and other 1,25-dihydroxyvitamin D3 and vitamin D analog metabolizing enzymes [6,7]. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Given the complex role of the kidney in calcium and phosphate homeostasis, a brief overview of calcium and phosphate reabsorption in the kidney is in order.

A. Calcium Handling by the Kidney About 55% of total plasma calcium is ultrafilterable [14]. The ultrafilterable calcium concentration is about 1.35 mM (5.4 mg/dl) and closely approximates the concentrations of calcium present in glomerular fluid [15–17]. The total amount of calcium filtered at the glomerulus in a 24-hr period is about 8000 mg. Approximately 98% of the filtered load of calcium is reabsorbed in the tubules. Thus, the amount of calcium excreted in the urine in a 24-hr period is about 150–200 mg [1,14]. In the proximal tubule, about 50–60% of the filtered load of calcium is reabsorbed [2,3,18]. The reabsorption of calcium is thought to occur as result of solvent drag by a paracellular route and is sodium dependent. Volume expansion and a reduction of tubular sodium reabsorption inhibits calcium reabsorption, while volume contraction and an increase in sodium reabsorption enhance calcium reabsorption [2,19]. Inhibition of sodium–potassium ATPase activity and sodium reabsorption by ouabain reduce the amount of calcium reabsorbed as does the substitution of sodium with lithium [19]. The concentration of calcium at the end of Copyright © 2005, Elsevier, Inc. All rights reserved.

516 the proximal tubule is similar to that in the glomerular fluid. Importantly, calcium reabsorption in the proximal tubule is not influenced by thiazide diuretics, by hormones such as PTH or 1,25(OH)2D3, or by hydrogen ions [2,3,18,19]. As we shall discuss later, some vitamin D–dependent proteins such as the calbindins, ECaC, and PMCa pump either are not expressed in the proximal tubule or are expressed in low amounts when compared with the amounts expressed in the distal tubule. Calcium reabsorption in the descending loop and the thin ascending limb of the loop of Henle is minimal. In the thick ascending limb of the loop of Henle about 20% of the filtered load of calcium is reabsorbed while another 10–15% is reabsorbed in the distal tubule, with the remaining 5% being reabsorbed in the collecting ducts [2,3,18,19]. There are important distinctions between the factors influencing calcium reabsorption in the proximal tubule and the mechanisms of calcium reabsorption in the distal segments of the nephron. First, the movement of calcium in the distal nephron occurs against a concentration gradient (lumen relative to extracellular fluid). Second, the lumen of the tubule is electronegative and becomes progressively more so toward the end of the distal tubule. Third, calcium reabsorption can be dissociated from sodium reabsorption by thiazide diuretics, which inhibit sodium reabsorption but enhance calcium reabsorption. Fourth, hydrogen ions inhibit calcium reabsorption in the distal tubule, whereas they have no effect on calcium reabsorption in the proximal tubule.

B. Phosphate Handling by the Kidney Virtually all inorganic phosphate in the serum is filtered by the glomerulus [8–10]. About 80% of filtered phosphorus is reabsorbed in the kidney, mostly in the proximal tubule. The amount of phosphorus reabsorbed in the proximal tubule is greatest in the first half of the proximal tubule and exceeds that of sodium. There is evidence for further phosphorus reabsorption in the pars recta. Little or no phosphorus reabsorption occurs in the loop of Henle or the distal tubule, although there is some debate about whether there is phosphorus reabsorption in the distal tubule. The reabsorption of phosphate is sodium dependent and is mediated by a sodium–phosphate cotransporter (Na-Pi II) [20]. Na-Pi II activity is increased by a low-phosphate diet and decreased by PTH [21–24]. The recently described phosphatonins fibroblast growth factor-23 (FGF-23) and secreted frizzled related protein-4 (sFRP-4) are also able to inhibit sodium-dependent phosphate transport [13,25]. In opossum kidney (OK) cells, NaPi II is

PETER TEBBEN AND RAJIV KUMAR

TABLE I Factors that Alter Renal Phosphate Excretion Increase

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

High phosphate diet Parathyroid hormone Calcitonin Chronic vitamin D Glucagon Glucocorticoids Volume expansion Increased pCO2 Chronic Acidosis Starvation Diuretics “Phosphatonin” FGF-23 sFRP-4

Decrease

1. 2. 3. 4. 5. 6. 7. 8.

Low-phosphate diet Thyro-parathyroidectomy Thyroxine Acute vitamin D Insulin Growth hormone Volume contraction Decreased pCO2

internalized from the cell membrane in response to FGF-23 and sFRP-4, similar to the effects of PTH [26,27]. Additional factors involved in phosphorus reabsorption are noted in Table I.

II. ROLE OF THE KIDNEY IN THE METABOLISM OF 25OH D A. Formation of 1,25(OH)2D3 25-Hydroxyvitamin D3-1α-hydroxylase is a multicomponent, cytochrome P450–containing enzyme in the mitochondria of renal proximal tubular cells [28–34] (see Chapter 5). The central role of the kidney in the formation of 1,25(OH)2D3 was first noted by Fraser and Kodicek, who demonstrated that nephrectomy abolished the formation of 1,25(OH)2D3 [35,36]. This was subsequently confirmed by others [37,38]. Nephrectomy greatly decreases circulating 1,25(OH)2D3 concentration in vivo except during pregnancy, granulomatous diseases, and lymphomas associated with the ectopic production of 1,25(OH)2D3 [39–42]. While the kidney is the major site of 1,25(OH)2D3 production, 25-hydroxyvitamin D3-1α-hydroxylase activity has been found in several other cell types throughout the body [43–48]. In vitro, chick renal epithelial cells in culture, mammalian nephron segments, and homogenates derived from avian and mammalian (mostly rodent) renal cells all appear to metabolize 25(OH)D3 to 1,25(OH)2D3 [49–53]. For some time the proximal tubule was felt to be the only site of 1,25(OH)2D3 synthesis in the kidney.

517

CHAPTER 29 Vitamin D and the Kidney

TABLE II

Effect of Increased Level or Activity of Various Factors on 1,25(OH)2D3 Concentration or 1α-Hydroxylase Activity

Factor

Parathyroid hormone Serum inorganic phosphorus 1,25(OH)2D3 Calcium (direct) Calcitonin Hydrogen ion Sex steroids Prolactin Growth hormone and insulin-like growth factor-1 Glucocorticoids Thyroid hormone Fibroblast growth factor 23 Frizzled related protein 4 Pregnancy

Animals

Humans

↑ ↓ ↓ ? ↑,↓,0 ↓ ↑ ↑ ↑ ↓,0 ? ↓ ↓ ↑

↑ ↓ ↓ ↓ ↑ 0 ↑ 0 ↑,↓,0 ↑,↓,0 ↓a ? ? ↑a

Reference

[17,51,250–258] [248,259–261] [250,262] [263,264] [14,51,250,251,265,266] [252,267,268] [247,269] [270–272] [273–278] [133,279–282] [283–285] [286,287] [13] [288,289]

↑, Stimulation or increase; ↓, suppression or decrease; 0, no effect; ?, effect not known. a Effects may be secondary to changes in calcium, phosphorus or parathyroid hormone. (With permission, modified from Kumar R [6].)

However, there is evidence that other tubular segments synthesize 1,25(OH)2D3 [54,55]. Using immunohistochemistry and in situ hybridization techniques, Zehnder et al. have demonstrated 1α-hydroxylase mRNA and protein in the distal convoluted tubule, cortical collecting duct, thick ascending limb of the loop of Henle, and Bowman’s capsule [55]. Recent experiments in which the 25-hydroxyvitamin D 1α-hydroxylase cytochrome P450 gene was deleted in mice point to the central role of this enzyme in vitamin D metabolism [56]. Table II summarizes some of the key factors known to regulate the activity of this enzyme in vivo and in vitro. The major regulators appear to be PTH, inorganic phosphorus, and 1,25(OH)2D3 itself.

B. 24-Hydroxylase Activity in the Kidney The 24-hydroxylase enzyme is a multicomponent cytochrome P450 enzyme expressed in the kidney as well as many other tissues [57–75] (see Chapter 6). In the kidney, 24-hydroxylase is expressed primarily in the proximal tubule but is also present in more distal segments [76,77]. It is responsible for the conversion of 25-hydroxyvitamin D3 and 1,25-dihydroxyvitamin D3 to 24,25-dihydroxyvitamin D3 and 1,24,25-trihydroxyvitamin D3, respectively. There is conflicting evidence whether 25(OH)D3 or 1,25(OH)2D3 is the preferred substrate for 24-hydroxylase [78]. Some have

reported 1,25(OH)2D3 is the preferred substrate with a Km approximately 10-fold lower than that for 25(OH)D3 [79,80], while others have found Km values substantially lower for 25(OH)D3 [81]. It has been suggested that these metabolites have certain unique properties and actions [82–85]; however, others have not confirmed these observations [86–92]. The 24-hydroxylase enzyme activity and mRNA expression in the kidney is up-regulated by 1,25(OH)2D3 [93–99]. The effect of 1,25(OH)2D3 is blunted in vitro and in vivo by parathyroid hormone [93–96]. The kidney is also capable of transforming 25(OH)D3 to several other compounds listed in Table III [100–120]. The specific physiological roles of these various metabolites are not known with certainty. Several polar metabolites of 1,25(OH)2D3 are formed in the liver, including calcitroic acid and glucuronide and sulfate conjugates of the hormone; these and small amounts of unchanged dihydroxylated and trihydroxylated metabolites of vitamin D are excreted in the urine [121–135]. Many of the tranformations

TABLE III 25OHD3

The Metabolism of 25OHD3 by the Kidney 24R,25(OH)2D3 25S,26(OH)2D3 23S,25(OH)2D3

24-Keto-25(OH)D3 25(OH)D3-lactone 23-Keto-25(OH)D3

518

PETER TEBBEN AND RAJIV KUMAR

that occur with 25(OH)D3 also occur in the case of 1,25(OH)2D3.

III. EFFECTS OF VITAMIN D, 25(OH)D3 AND 1,25(OH)2D3 ON THE RENAL HANDLING OF CALCIUM AND PHOSPHORUS Urine Ca (µmol/100 ml GF)

Clinical studies have shown that vitamin D deficiency is associated with low urine calcium concentration, whereas vitamin D excess or intoxication is associated with hypercalciuria [1]. The concentrations of calcium and phosphorus in the urine in these in vivo situations, however, reflect the decreased or increased calcium absorption in the intestine, the presence of hypo- or hypercalcemia, and the presence of diminished or elevated concentrations of circulating PTH. Puschett et al. examined the effect of vitamin D3 and 25(OH)D3 on the renal transport of phosphate, sodium, and calcium in parathyroidectomized dogs [136–138]. They showed that short-term infusions of vitamin D3 and 25(OH)D3 were associated with decreases in the clearances of phosphate, calcium, and sodium relative to the clearance of inulin. These studies were interpreted as showing that vitamin D3 and 25(OH)D3 enhance phosphate, calcium, and sodium reabsorption. A followup study by Puschett et al. showed that 1,25(OH)2D3 had similar effects on the excretion of phosphate, calcium, and sodium in thyroparathyroidectomized (TPTX) dogs [139]. We performed similar studies examining the effects of 25(OH)D3 on renal bicarbonate and phosphate reabsorption [140]. Unlike the studies of Puschett et al., we observed that phosphate and bicarbonate reabsorption increased only in intact animals and not in parathyroidectomized animals, suggesting the need for PTH. Our observations are similar to those of Popovtzer et al. [141]. Yamamoto et al. carried out perhaps the most comprehensive examination of the effects of 1,25(OH)2D3 on the reabsorption of calcium [142]. Vitamin D–deficient rats, vitamin D–deficient rats supplemented with dietary calcium to normalize plasma calcium and PTH levels, and vitamin D–replete rats were examined following TPTX and the infusion of graded amounts of calcium. Figure 1 shows the relationship between the amount of calcium excreted in the urine and serum calcium concentrations in the three groups of animals. Urinary calcium excretion was lower in vitamin D–replete rats than in vitamin D– deficient rats, suggesting that vitamin D increased the efficiency of calcium absorption in the absence of PTH. In a second group of experiments, rats treated in the manner noted earlier were TPTX and infused with PTH. The results of this experiment show that a lower dose of PTH is needed to exhibit a comparable effect

30

20

10

0

1

1.5

2

2.5

Serum Ca (mM)

FIGURE 1 Relationships between urinary calcium excretion and serum calcium concentration in three groups of thyroparathyroidectomized (TPTX) rats. Serum concentration and urinary excretion of calcium were determined 16–19 hr after continuous infusion of an electrolyte solution containing 0–30 mM CaCl2. Each point represents the data pooled according to a continuous series of 0.25-mM changes in serum calcium concentration. Horizontal bars indicate standard error of mean serum calcium concentration, and vertical bars indicate standard error of mean urinary calcium excretion. The lines were derived from the regression analysis of the linear portion of data. (
) Group A rats fed vitamin D–deficient standard diet. ( ) Group B rats fed vitamin D–deficient diet containing high calcium and lactose. () Group C rats fed vitamin D–replete standard diet. For any given serum calcium level, the urinary calcium excretion was significantly lower in vitamin D–replete rats (group C) than in vitamin D–deficient rats (groups A and B). Thus, the apparent serum calcium threshold determined as an intercept of the regression line on the serum calcium axis was higher in vitamin D-replete rats (~1.5 mM) than in vitamin D–deficient rats (~1.0 mM). There was no significant difference in the calcium threshold between group A and group B. (With permission from Yamamoto et al. [142].

on renal calcium reabsorption in vitamin D–replete rats when compared to vitamin D-deficient rats (Fig. 2). This finding may be explained by in vitro studies of distal convoluted tubule cells in which 1,25(OH)2D3 increased PTH/PTHrP receptor mRNA levels [143].

519

CHAPTER 29 Vitamin D and the Kidney

Urine Ca (µmol/100 ml GF)

A

B

C

30 TPTX

TPTX

TPTX 20

10 PTH 2.5 U/h 0

2 2.5 1 1.5 Serum Ca (mM)

PTH 2.5 U/h 0

1

1.5 2 2.5 Serum Ca (mM)

PTH 0.75 U/h 0

1 1.5 2 2.5 Serum Ca (mM)

FIGURE 2

The effects of PTH infusion on the renal handling of calcium among three groups of TPTX rats. The data are presented as described in the legend to Fig. 1. PTH was delivered at 2.5 U/hr (
) in group A and B rats, and 0.75 U/hr ( ) in group C rats. A, B, and C illustrate the results in groups A, B, and C, respectively. The enhancement of calcium reabsorption by PTH is shown as the shifts of the lines to the right in each group. The striking difference exists between vitamin D–deficient (groups A and B) and vitamin D–replete (group C) rats in the doses of PTH required to induce a comparable shift in the calcium threshold. ×, data of TPTX rats in each group (from Fig. 1). (With permission from Yamamoto et al. [142].)

Winaver et al. used micropuncture to examine the sites along the nephron at which 25(OH)D3 exerted its antiphosphaturic and hypocalciuric effects [144]. They observed that the effects of 25(OH)D3 were not mediated at the level of the superficial proximal tubule but were likely to have an effect on other segments of the nephron, most likely the distal tubule. These studies are similar to those of Sutton et al. and others who observed distal tubular effects of 25(OH)D3 in the dog [145]. Harris and Seely examined the effects of 1,25(OH)2D3 on tubular calcium handling and concluded that the enhanced reabsorption of calcium occurred in distal tubular segments of the nephron [146]. All of these studies do not exclude an effect of the vitamin D analogs on a segment of the proximal tubule that is not accessible to micropuncture but are consistent with the localization of many vitamin D–responsive proteins exclusively in the distal nephron. In vitro studies have shown that vitamin D deficiency is associated with decreased calcium uptake in membranes derived form the distal segments of the nephron [147]. This occurs in membranes of both luminal and basolateral origin. In addition, Bindels et al. have demonstrated that cultured rabbit connecting tubule cells show an increase in the transport of calcium when exposed to 1,25(OH)2D3 [148]. Protein and mRNA expression of the recently

described epithelial calcium channel (ECaC) are diminished in rats fed a diet deficient in vitamin D [149]. EcaC protein is present in the apical membrane of the distal convoluted tubule and is responsible for uptake of calcium from the tubular fluid into the cell. Calcitriol increases mRNA and protein expression of the PMCa pump, which is involved with active transport of calcium through the basolateral membrane of distal tubule cells [150]. A synthesis of the experimental results suggests that vitamin D3 metabolites have effects on the distal tubular reabsorption of calcium through several mechanisms. Expression of proteins responsible for distal tubule fluid uptake, intracellular trafficking, and basolateral transport of calcium are responsive to vitamin D3. Figure 3 depicts the transcellular transport of calcium through a distal tubule cell.

IV. DISTRIBUTION AND REGULATION OF VITAMIN D–DEPENDENT PROTEINS IN THE KIDNEY Several vitamin D–dependent proteins are expressed in the kidney, and many of these play a role in calcium and phosphate transport. The VDR, calbindins, PMCa

520

PETER TEBBEN AND RAJIV KUMAR

1,25(OH)2D3

Shuttling

ECaC Translation

Na+

Calbindin Ca2+ VDR

Activation

Transcription P P ATP Accessory proteins

Apical

Ca2+ ADP

Basolateral

FIGURE 3

Integrated model of active Ca2+ reabsorption in the distal part of the nephron. Apical entry of Ca2+ is facilitated by ECaC; Ca2+ then binds to calbindin-D28K, and this complex diffuses through the cytosol to the basolateral membrane, where Ca2+ is extruded by a Na+/Ca2+ exchanger and a plasma membrane Ca2+-ATPase. The individually controlled steps in the activation process of the rate-limiting Ca2+ entry channel include 1,25(OH)2D3-mediated transcriptional and translational activation, shuttling to the apical membrane, and subsequent activation of apically located channels by ambient Ca2+ concentration, direct phosphorylation and/or accessory proteins. (With permission from Hoenderop et al. [241].)

pump, and the ECaC are all found in renal tubule cells and act coordinately in the regulation of calcium transport in the nephron [5,18,143,149,151–157]. We will discuss pertinent information regarding these vitamin D–dependent proteins. Additional vitamin D– responsive proteins involved in renal calcium and phosphate transport are listed in Table IV.

TABLE IV Vitamin D–Responsive Proteins in the Kidney Vitamin D receptor 24-Hydroxylase Plasma membrane calcium pump Epithelial calcium channel Calbindin D28K Calbindin D9K Calcium sensing receptor PTH/PTHrP receptor Sodium-phosphate cotransporter type II

A. 1,25-Dihydroxyvitamin D3 Receptor (VDR) in the Kidney The VDR mediates many, if not all, of the effects of 1,25(OH)2D3 in diverse organs [158–166]. The distribution of the VDR in the kidney has been assessed using a variety of techniques including ligand binding assays with protein obtained from specific microdissected nephron segments, the localization of radiolabeled 1,25(OH)2D3 by autoradiography, and the use of various antibody techniques [76,167–171]. Using protein from microdissected nephron segments, the VDR was found in proximal and distal tubules [76,169]. With autoradiographic methods, following the administration of labeled ligand in vivo, silver grains were localized over distal tubule segments [76,168]. We have used sensitive polyclonal antibodies to localize the receptor in human and rat kidneys [155]. These antibodies were raised against highly purified antigen that was expressed in bacteria. The specificity of these antibodies was determined by absorption with purified antigen and by Western analysis, which showed that

521

CHAPTER 29 Vitamin D and the Kidney

they detected a protein band with a molecular mass of approximately 50,000 [155,172]. With these antibodies, we found that the VDR was present abundantly in the distal tubule and to a lesser extent in the proximal tubule (Fig. 4). Cells expressing calbindin-D28K also express the calcium pump and the epithelial calcium channel [149,173]. Interestingly, not all cells in the distal tubule expressed the receptor. Acid-secreting cells do not express the VDR in significant amounts. Taken together, the results are consistent with the notion that the VDR is present in significant amounts in the distal tubule where it regulates the amount and the activity of several vitamin D–dependent proteins such as the PMCa pump, ECaC, and calbindin D28K. Although in lesser amounts, the proximal tubule also expresses the VDR where it regulates the activity of 1α-hydroxylase and 24-hydroxylase. We have shown that the VDR is present in cells of the developing rodent kidney [174] (Figs. 5 and 6) and in the cultured metanephros (Fig. 7). The VDR was

detected as early as day 15 post coitum (p.c.) in the developing rat kidney in vivo. Significant amounts of the receptor were found in the metanephric mesenchyme as well as in the ureteric bud. As the kidney matured, the receptor was observed in the S-shaped and comma-shaped bodies and in the developing glomerulus, specifically in the parietal and visceral epithelial cells. The VDR staining in the latter cells persists in the adult kidney as well. Despite slightly different gestational periods, similar patterns of VDR were found in the developing mouse kidney in vivo. Of great interest is the observation that calbindin D28K appears in the distal tubule only around day 18 p.c. This is when urine flow begins in the kidney. The VDR is also present in mouse metanephric cultures in the same distribution pattern as is found in vivo. By day 3 of culture, the pattern of expression of the VDR was similar to that seen in the mouse kidney in vivo at day 15. Calbindin D28K does not appear in the cultured metanephros. These results showing that the VDR

A

D

G

B

E

H

C

F

I

FIGURE 4

A–C Immunohistochemical detection of VDR in normal human kidney tissue with polyclonal anti-hVDR antibody 2-152 (A, ×200; B, ×400; C, ×400). D–F Immunohistochemical detection of 25(OH)D3 24-hydroxylase cytochrome P-450 in human kidney (D, ×200; E, ×400; F, ×400). G Immunohistochemical detection of calbindin D28k in human kidney (G, ×400). H and I Control panels showing kidney tissue stained with preimmune serum (H, ×200; I, ×400). (With permission from Kumar R et al. [155].)

522

PETER TEBBEN AND RAJIV KUMAR

A

B

appears well before the appearance of calbindin D28K suggest that it may play a role in fetal renal development. Additional evidence for the regulation of calbindin D28K and calbindin D9K expression in the kidney can be found in VDR knockout mice, where calbindin expression is reduced [175]. The VDR is regulated by several factors in diverse tissues [176]. Concentrations of the VDR in the kidney and parathyroid glands are mainly regulated by 1,25(OH)2D3, PTH, and dietary calcium. Studies have shown that somewhat different results are obtained in vivo with respect to VDR abundance when 1,25(OH)2D3 concentrations are altered by dietary manipulations when compared to the effects of the intravenous administration of the hormone [177,178]. Exogenous administration of 1,25(OH)2D3 in rats

FIGURE 5 Vitamin D receptor (VDR) (A) immunostaining in metanephros of rat fetus on gestational day 15. (B) Higher power view of boxed area in A. Branching ureteric buds (arrows) and mesenchyme (M) are indicated. Bars, 2.1 µm. (With permission from Johnson et al. [174].) (See color plate.)

increases VDR levels in duodenal and renal tissues [177,178]. When endogenous 1,25(OH)2D3 concentrations are increased by adapting an animal to a low calcium diet (Table V), VDR concentrations in the duodenum and kidney do not increase. The difference appears to be due to increases in the levels of PTH elicited by the low-calcium diet and decreased PTH following 1,25(OH)2D3 administration. PTH has been shown to down-regulate VDR in osteosarcoma cells as well as block up-regulation of VDR in rats infused with both PTH and 1,25(OH)2D3 [179]. These results demonstrate the opposing effects of PTH and 1,25(OH)2D3 on VDR expression in the kidney. The results obtained from in vivo studies are consistent with those obtained following the administration of 1,25(OH)2D3 to yeast cells transfected with a VDR

523

CHAPTER 29 Vitamin D and the Kidney

A

FIGURE 6

VDR (A) and calbindin D28k (B) immunostaining in metanephros of rat fetus on gestational day 17. Parietal epithelial cells (small arrows), visceral epithelial cells (open arrows), and tubule portion of developing comma-shaped body (T) are indicated. Bars, 2.1 µm. (With permission from Johnson et al. [174].) (See color plate.)

B

construct [180]. There is evidence that in certain cells such as fibroblasts, 1,25(OH)2D3 may have its predominant effect to increase VDR abundance by ligandinduced stabilization of the protein [181] Wiese et al. showed that the addition of 1,25(OH)2D3 to fibroblasts did not result in a significant increase in VDR mRNA concentrations but did result in increases in the amount of VDR protein, suggesting that the predominant effect of 1,25(OH)2D3 in fibroblasts is the stabilization of pre-existing VDR protein or the stabilization of preexisting vitamin D receptor mRNA. In addition to the VDR, several vitamin D– dependent proteins are expressed in the kidney. The pattern of regulation of these proteins is of great interest, inasmuch as it casts light on the different mechanisms by which calcium is transported in the kidney.

Those proteins involved in calcium transport include calbindin D28K and calbindin D9K, the PMCa pump, EC2C, and the calcium sensing receptor. Additionally, although not involved in the transport of calcium, the 24-hydroxylase and the sodium-phosphate type II cotransporter (NaPi II) are also expressed in the kidney. All of these proteins appear to be regulated by 1,25(OH)2D3.

B. Calbindins-D The calbindins-D are widely distributed in many tissues of the body and in a variety of species. Except in the brain, their synthesis is dependent on vitamin D but not calcium or phosphorus. There are two forms of

524

PETER TEBBEN AND RAJIV KUMAR

P

D

P

FIGURE 7 VDR immunostaining in mouse metanephric organ culture explant following 120 hr of incubation. Parietal epithelial cells (small arrow), visceral epithelial cells (large arrow), proximal tubule (P), and distal tubule (D) are indicated. Bars, 2.1 µm. (With permission from Johnson et al. [174]). (See color plate.)

D

calbindin-D, namely, calbindin-D28K and calbindinD9K, that are variably distributed in different tissues of the body [182–199]. The apparent molecular weight of the larger protein is about 30,000 Da, whereas that of the smaller protein is 9,000 Da [182]. The proteins are classical E-F hand proteins, the calbindin D28K having six EF hand structures and the calbindin D9K having two such motifs [200–219]. The proteins bind calcium with high affinity and in different molar amounts. Calbindin D28K binds 3–4 mol calcium per mol protein and calbindin D9K 2 mol calcium per mol protein [182]. In the mouse kidney, both forms are present and

are regulated by vitamin D. In other species, only calbindin D28K is expressed in the kidney. Both proteins undergo conformational change upon binding to calcium. This phenomenon has been studied extensively by us in the case of calbindin D28K. We have shown that the protein undergoes a two-step change in conformation that is associated with binding to a high-affinity site in EF-hand 1 and a further change upon calcium binding to sites in EF hands 4 and 5 [209,211]. The conformation change is probably what allows the protein to act as a modulator of the activity of the PMCa pump (see later discussion).

525

CHAPTER 29 Vitamin D and the Kidney

TABLE V Effect of Dietary Calcium on Unoccupied VDR Content in Rat Duodenum and Kidney Day Diet

VDRa Duodenum Kidney Calciuma (mg/dl) Phosphorusa (mg/dl) 1,25-(OH)2D3a (pg/ml)

1% calcium 0.02% calcium 1% calcium 0.02% calcium 1% calcium 0.02% calcium 1% calcium 0.02% calcium 1% calcium 0.02% calcium

2

341 ± 26 365 ± 27 ND ND 9.98 ± 0.18 9.45 ± 0.12 8.40 ± 0.17 8.15 ± 0.22 153 ± 11 180 ± 20

7

197 ± 17 226 ± 28 163 ± 11 120 ± 4* 9.82 ± 0.08 9.15 ± 0.17* 8.88 ± 0.22 8.65 ± 0.13 113 ± 13 392 ± 45**

14

202 ± 17 221 ± 16 165 ± 9 131 ± 10* 9.16 ± 0.16 9.09 ± 0.14 8.73 ± 0.17 8.37 ± 0.28 139 ± 9 682 ± 44**

21

259 ± 26 267 ± 28 124 ± 8 77 ± 3* 9.52 ± 0.41 8.70 ± 0.29* 7.92 ± 0.29 8.41 ± 0.23 160 ± 32 829 ± 59**

aValues are mean ± SEM. Unoccupied vitamin D receptor content expressed as fmol [3H]1,25(OH) D bound per mg cytosol protein. *P < 0.05; 2 3 **P < 0.001. ND, Not done. Modified from Goff et al. [177].

Calbindin D28K and calbindin D9K are regulated by 1,25(OH)2D3 in the kidney [213–220]. Both calbindins have lower expression in 1α-hydroxylase knockout mice [221]. Calbindin D28K and calbindin D9K expression was normalized by treatment with 1,25(OH)2D3; however, only calbindin D28K expression was increased when 1α-hydroxylase knockout mice were fed a highcalcium diet. In a mouse VDR-KO model, renal calbindin D9K expression was nearly abolished, whereas calbindin D28K expression returned to control values as the animals age [222]. Cao et al. demonstrated that calbindin-D9K induction by 1,25(OH)2D3 in vitro is absent in VDR-null cells [223]. Furthermore, calbindin D9K regulation by 1,25(OH)2D3 was restored after transfection of the VDR null cells with human VDR, clearly showing that the effect of 1,25(OH)2D3 is mediated through the VDR. In vitro, PTH has a synergistic effect with 1,25(OH)2D3 on calbindin D9K expression but has no effect alone [223]. In contrast, calbindin D28K expression is enhanced by PTH infusion without elevations in 1,25(OH)2D3 concentrations [224]. Using distal tubule membranes, preincubation with calbindin D28K increased calcium uptake in luminal membranes while calbindin D9K preincubation increased calcium uptake in basolateral membranes [225,226]. Calbindin D28K knockout mice fed a high-calcium (1%) diet have an elevated urinary calcium/creatinine ratio despite no difference in serum calcium or PTH compared to wild-type littermates [227,228]. However, the elevated urinary calcium/creatinine ratio in calbindin D28K knockout mice was not apparent after fasting [227].

This is consistent with our finding that fractional excretion of calcium is not altered in calbindin D28K knockout mice fed a normal calcium diet [229]. The effects of calbindin D28K on renal calcium conservation in mice are modest, perhaps because of compensatory increases in calbindin D9K. The interactions between vitamin D, PTH, calcium, and the calbindins appears to be quite complex. The exact mechanisms and their interactions with other calcium transport proteins is not yet completely understood.

C. The Plasma Membrane Calcium Pump We raised monoclonal and polyclonal antibodies directed against the PMCa pump and used them to examine the distribution of these proteins in the adult human kidney [18,151–154]. We found that epitopes for the calcium–magnesium ATPase (calcium pump) were expressed in the basolateral membrane of the distal tubular cells (Fig. 8). Similar patterns of expression were apparent in the rat [152] and rabbit [230] kidney. Interestingly, not all cells of the distal tubule stained positively for the calcium pump. Further investigation showed that cells expressing carbonic anhydrase and presumably involved in acid secretion did not express the PMCa pump, whereas the other cells of the distal tubule did so. We found that the calbindin D28K was present in the same cells of the distal tubule as the PMCa pump. Our studies on the distribution of the PMCa pump in the kidney and its localization

526

PETER TEBBEN AND RAJIV KUMAR

A

B

C

D

E

F

G

H

I

J

FIGURE 8

Immunoperoxidase localization of Ca2+-Mg2+ ATPase within human kidney distal tubules and human spleen erythrocytes. Kidney: (A and B) Monoclonal antibody JA3. (C and D) Monoclonal antibody JA8. (E and F) Negative control. (G and H) Double stain, PAS, and JA3; (arrows) PAS-positive proximal tubule brush border. Spleen: (I) Monoclonal antibody JA3. (J) Negative control. (A, C, E, and G) ×200. (B, D, F, H, I, and J) ×640. (With permission from Borke et al. [151].) (See color plate.)

527

CHAPTER 29 Vitamin D and the Kidney

predominantly in the distal tubule of the kidney are supported by others who have shown that calcium ATPases are present mostly in the distal tubule [230]. In addition, it appears that the PMCa pump is widely distributed in a large number of other calciumtransporting tissues, many of which display vitamin D– dependent calcium transport [231–237]. Table VI shows the distribution of the PMCa pump in different tissues. In MDBK (bovine distal tubule) cells, 1,25(OH)2D3 increases PMCa pump mRNA and protein [150] Bouhtiauy et al. have examined the effects of vitamin D deficiency on the activity of the PMCa pump [225,226]. They observed that vitamin D deficiency was associated with a decrease in PMCa pump activity in the distal tubule of the kidney. These same authors also observed that calbindin D9K increased the activity of the renal basolateral membrane calcium pump. In collaboration with Wasserman’s group at Cornell University, we have shown that the PMCa pump is regulated in the intestine by vitamin D [232,238,239]. We showed by Western analysis using monoclonal antibodies directed against the pump that shortly after the administration of 1,25(OH)2D3 to vitamin D– deficient chicks, there is an increase in the amount of immunoreactive PMCa pump in the cells of the duodenum, jejunum, and ileum [232]. This is associated with an increase in the amount of mRNA for the pump in the same segments of the intestine [238]. The increase occurs within 3–6 hr following the administration of 1,25(OH)2D3 and the effect is dosedependent. Furthermore, dietary calcium and phosphorus depletion are also associated with an increase in the amount of the pump expressed in the intestine. Thus, in the intestine, 1,25(OH)2D3 increases the synthesis of the PMCa pump. In addition, 1,25(OH)2D3 increases the activity of the PMCa pump in intestinal cell basolateral membranes.

TABLE VI Tissue

Kidney Intestine Trophoblast Choroid plexus Shell gland Bone Bone

The mechanism by which up-regulation of the calcium pump occurs in the kidney and intestine is uncertain, although several possibilities arise. It could be a direct effect of either 1,25(OH)2D3 or via stimulation of calbindin D9k or calbindin D28k [225,226]. There is also evidence that stimulation of the calcium sensing receptor (CaSR) decreases calcium absorption by inhibiting PMCa pump activity [240].

D. The Epithelial Calcium Channel Hoenderop et al. recently described the epithelial calcium channel (ECaC), which is expressed in the apical membrane of the distal tubule and is distinct from previously described calcium channels [173]. There are at least two members in this new family of calcium channels, ECaC1 and ECaC2 [241]. ECaC1 expression is limited to the kidney while ECaC2 is expressed in several other tissues [241–244]. The EcaCs have six putative transmembrane spanning domains including a pore-forming hydrophobic region between transmembrane domains 5 and 6 [245]. Several putative vitamin D response elements (VDREs) have been identified within the promoter region of human ECaC. Hoenderop et al. also demonstrated that ECaC1 mRNA and protein levels are increased to near control levels after vitamin D rescue in rats fed a vitamin D– deficient diet [149]. In a study using 1α-hydroxylase knockout mice, a greater than 50% reduction in ECaC1 expression was found compared to control mice [221]. Additionally, renal ECaC1 expression in 1α-hydroxylase KO mice is normalized after treatment with 1,25(OH)2D3 [221]. Similar findings were seen when examining calbindin D28K expression which co-localizes to the same distal tubule cells as ECaC1 [173,221]. However, ECaC1 is not regulated through vitamin D effects on calbindin D28K. Calbindin D28K

Distribution of Plasma Membrane Calcium Pump in Transporting Epithelia as Assessed by Immunohistochemistry

Source

Rat, human Rat, chick Rat, human Cat Human Chick Human Chick

Cell type

Distal convoluted tubule, principal cell Absorptive cell Syncytiotrophoblast Choroid plexus Secretory cell Principal cell Osteoblast Osteoclast

Location in cell

Reference

Basolateral Basolateral Basal Apical

[18,151–154] [231,232] [233] [234]

Apical Not vectorially oriented Not vectorially oriented

[235] [236] [237]

528 KO mice and cyclosporine A induced down-regulation of calbindin D28K has no effect on ECaC1 expression [227]. Others have suggested that calcium also regulates ECaC1 expression. Quantitative PRC techniques showed reduced expression of ECaC1 in VDR KO mice compared to control mice. When fed highcalcium diets, VDR KO mice had normalization of ECaC1 concentrations [244].

E. The 25-Hydroxyvitamin D3- and 1,25-Dihydroxyvitamin D3-24-Hydroxylase The 24-hydroxylase enzyme is widely distributed in a number of renal and nonrenal tissues [67–75,84,85, 246]. Pioneering work by DeLuca’s laboratory showed that the renal 24-hydroxylase was regulated by calcium and phosphorus such that elevated or normal calcium levels induced the 24-hydroxylase, whereas low calcium levels inhibited it [69–71,247]. Similarly, elevated serum phosphorus concentrations increased the synthesis of the 24-hydroxylase enzyme, whereas lowphosphorus diets decreased the activity of the enzyme [248]. We have used antibodies against the 24-hydroxylase cytochrome P450 to examine the distribution of the enzyme in the human kidney, and found exceptionally high concentrations of the cytochrome P450 in distal tubular cells [155]. Lower amounts were found in the proximal tubule. Using enzymatic methods, several investigators have found 24-hydroxylase activity in kidney cells of the proximal tubule of the rat nephron. Iida et al. were unable to find 24-hydroxylase activity in microdissected distal tubule segment [249]. The reason for this apparent discrepancy between human and rat tissues is uncertain. Certainly, it would make biologic sense for the 24-hydroxylase to be present in the distal tubule where other elements of the vitamin D responsive system are present. Other enzymes responsible for the transformation of 25-hydroxyvitamin D are also present in the kidney; they mediate the reactions shown in Table III.

V. CONCLUSION The kidney plays a vital role in the conservation of calcium and phosphorus. Besides being the site of synthesis of 1,25(OH)2D3, the kidney responds to the hormone by increasing the efficiency of calcium and phosphorus reabsorption. Elements of the calcium transport systems including calbindin D28K, calbindin D9K, the epithelial calcium channel, and the plasma membrane calcium pump all localize to the distal portion of the nephron and are regulated directly or indirectly by 1,25(OH)2D3.

PETER TEBBEN AND RAJIV KUMAR

References 1. 1993 Metabolic bone and stone disease. In: Nordin BEC, Need AG, Morris HA (eds) Churchill Livingstone, Edinburgh, p. 481. 2. Friedman PA, Gesek FA 1995 Cellular calcium transport in renal epithelia: measurement, mechanisms, and regulation. Physiol Rev 75:429–471. 3. Borke JL, Penniston JT, Kumar R 1990 Recent advances in calcium transport by the kidney. Sem Nephrol 10:15–23. 4. Johnson JA, Kumar R 1994 Vitamin D and renal calcium transport. Curr Opin Nephrol Hypertens 3:424–429. 5. Kumar R 1995 Calcium transport in epithelial cells of the intestine and kidney. J Cell Biochem 57:392–398. 6. Kumar R 1984 Metabolism of 1,25-dihydroxyvitamin D3. Physiol Rev 64:478–504. 7. Kumar R 1991 Vitamin D and calcium transport. Kidney Int 40:1177–1189. 8. Berndt TJ, Knox FG 1992 Renal regulation of phosphate excretion. In: Seldin DW, Giebisch GH (eds) The Kidney: Physiology and Pathophysiology. Raven Press, New York, pp. 2511–2532. 9. Knox FG, Haramati A 1985 Renal regulation of phosphate excretion. In: Seldin DW, Giebisch GH (eds) The Kidney: Physiology and Pathophysiology. Raven Press, New York, pp. 1351–1396. 10. Knox FG, Haas JA, Berndt T, Marchand GR, Youngberg SP 1997 Phosphate transport in superficial and deep nephrons in phosphate-loaded rats. Am J Physiol 233:F150–F153. 11. Schiavi SC, Moe OW 2002 Phosphatonins: a new class of phosphate-regulating proteins. Curr Opin Nephrol Hypertens 11:423–430. 12. Kumar R 2002 New insights into phosphate homeostasis: fibroblast growth factor 23 and frizzled-related protein-4 are phosphaturic factors derived from tumors associated with osteomalacia. Curr Opin Nephrol Hypertens 11:547–553. 13. Berndt T, Craig TA, Bowe AE, et al. 2003 Secreted frizzledrelated protein 4 is a potent tumor-derived phosphaturic agent. J Clin Invest, in press. 14. Kumar R. 1996 Calcium disorders. In: Kokko JP, Tannen RL (eds) Fluids and Electrolytes. Saunders, Philadelphia, pp. 391–419. 15. Neuman WF, Neuman MW. The chemical dynamics of bone mineral. University of Chicago Press, Chicago, p. 209. 16. Moore EW 1970 Ionized calcium in normal serum, ultrafiltrates, and whole blood determined by ion-exchange electrodes. J Clin Invest 49:318–334. 17. Raman A 1971 The calcium fractions of normal serum. Clin Biochem 4:141–146. 18. Kumar R, Penniston JT, Borke JL 1988 Ca2+-Mg2+-ATPase calcium pumps in the kidney. New Physiol Sci 3:219–222. 19. Friedman PA 1988 Renal calcium transport: sites and insights. New Physiol Sci 3:17–21. 20. Murer H, Hernando N, Forster I, Biber J 2001 Molecular aspects in the regulation of renal inorganic phosphate reabsorption: The type IIa sodium/inorganic phosphate co-transporter as the key player. Curr Opin Nephrol Hypertens 10:555–561. 21. Traebert M, Roth J, Biber J, Murer H, Kaissling B 2000 Internalization of proximal tubular type II Na-Pi cotransporter by PTH: immunogold electron microscopy. Am J Physiol Renal Fluid Electrolyte Physiol 278:F148–F154. 22. Murer H, Lotscher M, Kaissling B, Levi M, Kempson SA, Biber J 1996 Renal brush border membrane Na/Pi-cotransport:

CHAPTER 29 Vitamin D and the Kidney

23.

24.

25.

26.

27.

28.

29.

30.

31. 32.

33. 34.

35. 36.

37. 38. 39. 40.

molecular aspects in PTH-dependent and dietary regulation. Kidney Int 49:1769–1773. Keusch I, Traebert M, Lotscher M, Kaissling B, Murer H, Biber J 1998 Parathyroid hormone and dietary phosphate provoke a lysosomal routing of the proximal tubular Na/Picotransporter type II. Kidney Int 54:1224–1232. Lotscher M, Scarpetta Y, Levi M, et al. 1999 Rapid downregulation of rat renal Na/Pi cotransporter in response to parathyroid hormone involves microtubule rearrangement. J Clin Invest 104:483–494. Yamashita T, Konishi M, Miyake A, Inui K, Itoh N 2002 Fibroblast growth factor (FGF)-23 inhibits renal phosphate reabsorption by activation of the mitogen-activated protein kinase pathway. J Biol Chem 277:28265–28270. O’Brien S, Bowe AE, Hernando N, et al. 2003 Role of FGF23 in surface retrieval of sodium-phosphate co-transporter (NaPiIIa) in opossum kidney proximal tubule cells. J Am Soc Nephrol: ASN abstract 555303. O’Brien S, Bowe AE, Pragnell M, Kumar R, Schiavi SC 2003 FGF23 and FRP4 internalize sodium-phosphate co-transporter (NaPi2a) in opossum kidney proximal tubule cells. J Bone Miner Res: ASBMR Abstract. Ghazarian JG, DeLuca HF 1974 25-Hydroxycholecalciferol-1hydroxylase: a specific requirement for NADPH and a hemoprotein component in chick kidney mitochondria. Arch Biochem Biophys 160:63–72. Ghazarian JG, Jefcoate CR, Knutson JC, Orme-Johnson WH, DeLuca HF 1974 Mitochondrial cytochrome P450. A component of chick kidney 25-hydrocholecalciferol-1α-hydroxylase. J Biol Chem 249:3026–3033. Pedersen JI, Ghazarian JG, Orme-Johnson NR, DeLuca HF 1976 Isolation of chick renal mitochondrial ferredoxin active in the 25-hydroxyvitamin D3-1α-hydroxylase system. J Biol Chem 251:3933–3941. Yoon PS, DeLuca HF 1980 Purification and properties of chick renal mitochondrial ferredoxin. Biochemistry 19:2165–2171. Yoon PS, Rawlings J, Orme-Johnson WH, DeLuca HF 1980 Renal mitochondrial ferredoxin active in 25-hydroxyvitamin D3 1α-hydroxylase. Characterization of the iron-sulfur cluster using interprotein cluster transfer and electron paramagnetic resonance spectroscopy. Biochemistry 19:2172–2176. Okamoto Y, DeLuca HF 1994 Separation of two forms of chick 1,25-dihydroxyvitamin D3 and 25-hydroxyvitamin D3 24-hydroxylase. Proc Soc Exp Biol Med 205:52–55. Burgos-Trinidad M, Ismail R, Ettinger RA, Prahl JM, DeLuca HF 1992 Immunopurified 25-hydroxyvitamin D 1α-hydroxylase and 1,25-dihydroxyvitamin D 24-hydroxylase are closely related but distinct enzymes. J Biol Chem 267:3498–3505. Fraser DR, Kodicek E 1970 Unique biosynthesis by kidney of a biological active vitamin D metabolite. Nature 228:764–766. Lawson DE, Fraser DR, Kodicek E, Morris HR, Williams DH 1971 Identification of 1,25-dihydroxycholecalciferol, a new kidney hormone controlling calcium metabolism. Nature 230:228–230. Gray R, Boyle I, DeLuca HF 1971 Vitamin D metabolism: The role of kidney tissue. Science 172:1232–1234. Shultz TD, Fox J, Heath H, 3rd, Kumar R 1983 Do tissues other than the kidney produce 1,25-dihydroxyvitamin D3 in vivo? A reexamination. Pro Natl Acad Sci USA 80: 1746–1750. Weisman Y, Vargas A, Duckett G, Reiter E, Root AW 1978 Synthesis of 1,25-dihydroxyvitamin D in the nephrectomized pregnant rat. Endocrinology 103:1992–1996. Tanaka Y, Halloran B, Schnoes HK, DeLuca HF 1979 In vitro production of 1,25-dihydroxyvitamin D3 by rat placental tissue. Proc Natl Acad Sci USA 76:5033–5035.

529 41. Bell NH 1984 Sarcoidosis. In: Kumar R (ed) Vitamin D: Basic and Clinical Aspects. Nijhoff, Boston, pp. 581–589. 42. Golconda MS, Larson TS, Kolb LG, Kumar R 1996 1,25Dihydroxyvitamin D-mediated hypercalcemia in a renal transplant recipient. Mayo Clinic Proceedings 71:32–36. 43. Schuster I, Egger H, Bikle D, et al. 2001 Selective inhibition of vitamin D hydroxylases in human keratinocytes. Steroids 66:409–422. 44. Diaz L, Arranz C, Avila E, Halhali A, Vilchis F, Larrea F 2002 Expression and activity of 25-hydroxyvitamin D-1α-hydroxylase are restricted in cultures of human syncytiotrophoblast cells from preeclamptic pregnancies. J Clin Endocrinol Metab 87:3876–3882. 45. Zehnder D, Evans KN, Kilby MD, et al. 2002 The ontogeny of 25-hydroxyvitamin D3 1α-hydroxylase expression in human placenta and decidua. Am J Pathol 161:105–114. 46. Howard GA, Turner RT, Sherrard DJ, Baylink DJ 1981 Human bone cells in culture metabolize 25-hydroxyvitamin D3 to 1,25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3. J Biol Chem 256:7738–7740. 47. Dusso AS, Finch J, Brown A, et al. 1991 Extrarenal production of calcitriol in normal and uremic humans. J Clin Endocrinol Metab 72:157–164. 48. Fu GK, Lin D, Zhang MY, et al. 1997 Cloning of human 25-hydroxyvitamin D-1α-hydroxylase and mutations causing vitamin D–dependent rickets type 1. Mol Endocrinol 11:1961–1970. 49. Henry HL 1984 Regulation of the synthesis of 1,25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3 in the kidney cell culture. In: Kumar R (ed) Vitamin D: Basic and Clinical Aspects. Nijhoff, Boston, pp. 152–174. 50. Turner RL 1984 Mammalian 25-hydroxyvitamin D-1αhydroxylase: Measurement and regulation. In: Kumar R (ed) Vitamin D: Basic and Clinical Aspects. Nijhoff, Boston, pp. 175–196. 51. Rasmussen H, Wong M, Bikle D, Goodman DB 1972 Hormonal control of the renal conversion of 25-hydroxycholecalciferol to 1,25-dihydroxycholecalciferol. J Clin Invest 51:2502–2504. 52. Trechsel U, Bonjour JP, Fleisch H 1979 Regulation of the metabolism of 25-hydroxyvitamin D3 in primary cultures of chick kidney cells. J Clin Invest 64:206–217. 53. Henry HL 1979 Regulation of the hydroxylation of 25-hydroxyvitamin D3 in vivo and in primary cultures of chick kidney cells. J Biol Chemi 254:2722–2729. 54. Bland R, Zehnder D, Hughes SV, Ronco PM, Stewart PM, Hewison M 2001 Regulation of vitamin D-1α-hydroxylase in a human cortical collecting duct cell line. Kidney Int 60:1277–1286. 55. Zehnder D, Bland R, Walker EA, et al. 1999 Expression of 25-hydroxyvitamin D3-1α-hydroxylase in the human kidney. J Am Soc Nephrol 10:2465–2473. 56. Panda DK, Miao D, Tremblay ML, et al. 2001 Targeted ablation of the 25-hydroxyvitamin D 1α-hydroxylase enzyme: evidence for skeletal, reproductive, and immune dysfunction. Proc Natl Acad Sci USA 98:7498–7503. 57. Rubin LP, Yeung B, Vouros P, Vilner LM, Reddy GS 1993 Evidence for human placental synthesis of 24,25-dihydroxyvitamin D3 and 23,25-dihydroxyvitamin D3. Pediatr Res 34:98–104. 58. Staal A, van den Bemd GJ, Birkenhager JC, Pols HA, van Leeuwen JP 1997 Consequences of vitamin D receptor regulation for the 1,25-dihydroxyvitamin D3–induced 24-hydroxylase activity in osteoblast-like cells: initiation of the C24-oxidation pathway. Bone 20:237–243.

530 59. Armbrecht HJ, Hodam TL, Boltz MA, Partridge NC, Brown AJ, Kumar VB 1998 Induction of the vitamin D 24-hydroxylase (CYP24) by 1,25-dihydroxyvitamin D3 is regulated by parathyroid hormone in UMR106 osteoblastic cells [comment]. Endocrinology 139:3375–3381. 60. Kurahashi I, Matsunuma A, Kawane T, Abe M, Horiuchi N 2002 Dexamethasone enhances vitamin D-24-hydroxylase expression in osteoblastic (UMR-106) and renal (LLC-PK1) cells treated with 1α,25-dihydroxyvitamin D3. Endocr J 17:109–18. 61. Nishimura A, Shinki T, Jin CH, et al. 1994 Regulation of messenger ribonucleic acid expression of 1α,25-dihydroxyvitamin D3-24-hydroxylase in rat osteoblasts. Endocrinology 134: 1794–1799. 62. Kreutz M, Andreesen R, Krause SW, Szabo A, Ritz E, Reichel H 1993 1,25-Dihydroxyvitamin D3 production and vitamin D3 receptor expression are developmentally regulated during differentiation of human monocytes into macrophages. Blood 82:1300–1307. 63. Demers C, Lemay J, Hendy GN, Gascon-Barre M 1997 Comparative in vivo expression of the calcitriol-24-hydroxylase gene in kidney and intestine. J Mol Endocrinol 18:37–48. 64. Theodoropoulos C, Demers C, Delvin E, Menard D, GasconBarre M 2003 Calcitriol regulates the expression of the genes encoding the three key vitamin D3 hydroxylases and the drugmetabolizing enzyme CYP3A4 in the human fetal intestine. Clin Endocrinol 58:489–499. 65. Furuichi T, Kawata S, Asoh Y, Kumaki K, Ohyama Y 1998 Differential time course of induction of 1α,25-dihydroxyvitamin D3-24-hydroxylase mRNA expression in rats by 1α,25-dihydroxyvitamin D3 and its analogs. Life Sci 62:453–459. 66. Akeno N, Saikatsu S, Kawane T, Horiuchi N 1997 Mouse vitamin D-24-hydroxylase: molecular cloning, tissue distribution, and transcriptional regulation by 1α,25-dihydroxyvitamin D3. Endocrinology 138:2233–2240. 67. Tanaka Y, Lorenc RS, DeLuca HF The role of 1,25-dihydroxyvitamin D3 and parathyroid hormone in the regulation of chick renal 25-hydroxyvitamin D3-24-hydroxylase. Arch Biochem Biophys 171:521–526. 68. Tanaka Y, DeLuca HF, Ikekawa N, Morisaki M, Koizumi N 1975 Determination of stereochemical configuration of the 24-hydroxyl group of 24,25-dihydroxyvitamin D3 and its biological importance. Arch Biochem Biophys 170:620–626. 69. Holick MF, Baxter LA, Schraufrogel PK, Tavela TE, DeLuca HF 1976 Metabolism and biological activity of 24,25-dihydroxyvitamin D3 in the chick. J Biol Chem 251:397–402. 70. Holick MF, Schnoes HK, DeLuca HF, Gray RW, Boyle IT, Suda T 1972 Isolation and identification of 24,25-dihydroxycholecalciferol, a metabolite of vitamin D made in the kidney. Biochemistry 11:4251–4255. 71. Knutson JC, DeLuca HF 1974 25-Hydroxyvitamin D3-24hydroxylase. Subcellular location and properties. Biochemistry 13:1543–1548. 72. Kumar R, Schnoes HK, DeLuca HF 1978 Rat intestinal 25-hydroxyvitamin D3- and 1α,25-dihydroxyvitamin D324-hydroxylase. J Biol Chem 253:3804–3809. 73. Garabedian M, Du Bois MB, Corvol MT, Pezant E, Balsan S 1978 Vitamin D and cartilage. I. In vitro metabolism of 25-hydroxycholecalciferol by cartilage. Endocrinology 102: 1262–1268. 74. Kulkowski JA, Chan T, Martinez J, Ghazarian JG 1979 Modulation of 25-hydroxyvitamin D3-24-hydroxylase by aminophylline: A cytochrome P-450 monooxygenase system. Biochem Biophys Res Commun 90:50–57.

PETER TEBBEN AND RAJIV KUMAR

75. DeLuca H 1979 The vitamin D system in the regulation of calcium and phosphorus metabolism. Nutr Rev 37:161–193. 76. Kawashima H, Torikai S, Kurokawa K 1981 Localization of 25-hydroxyvitamin D3 1α-hydroxylase and 24hydroxylase along the rat nephron. Proc Natl Acad Sci USA 78:1199–1203. 77. Yang W, Friedman PA, Kumar R, et al. 1999 Expression of 25(OH)D3 24-hydroxylase in distal nephron: coordinate regulation by 1,25(OH)2D3 and cAMP or PTH. Am J Physiol 276:E793–E805. 78. Omdahl JL, Morris HA, May BK 2002 Hydroxylase enzymes of the vitamin D pathway: expression, function, and regulation. Ann Rev Nutr 22:139–166. 79. Akiyoshi-Shibata M, Sakaki T, Ohyama Y, Noshiro M, Okuda K, Yabusaki Y 1994 Further oxidation of hydroxycalcidiol by calcidiol 24-hydroxylase. A study with the mature enzyme expressed in Escherichia coli. Eur J Biochem 224: 335–343. 80. Burgos-Trinidad M, DeLuca HF 1991 Kinetic properties of 25-hydroxyvitamin D- and 1,25-dihydroxyvitamin D-24hydroxylase from chick kidney. Biochim Biophys Acta 1078:226–230. 81. Taniguchi T, Eto TA, Shiotsuki H, et al. 2001 Newly established assay method for 25-hydroxyvitamin D3 24-hydroxylase revealed much lower Km for 25-hydroxyvitamin D3 than for 1α,25-dihydroxyvitamin D3 [comment]. J Bone Miner Res 16:57–62. 82. Seo EG, Norman AW 1997 Three-fold induction of renal 25-hydroxyvitamin D3-24-hydroxylase activity and increased serum 24,25-dihydroxyvitamin D3 levels are correlated with the healing process after chick tibial fracture. J Bone Miner Res 12:598–606. 83. Henry HL, Norman AW 1978 Vitamin D: Two dihydroxylated metabolites are required for normal chicken egg hatchability. Science 201:835–837. 84. Henry HL, Taylor AN, Norman AW 1977 Response of chick parathyroid glands to the vitamin D metabolites, 1,25dihydroxycholecalciferol and 24,25-dihydroxycholecalciferol. J Nutr 107:1918–1926. 85. Ornoy A, Goodwin D, Noff D, Edelstein S 1978 24,25Dihydroxyvitamin D is a metabolite of vitamin D essential for bone formation. Nature 276:517–519. 86. St-Arnaud R, Arabian A, Travers R, et al. 2000 Deficient mineralization of intramembranous bone in vitamin D-24hydroxylase-ablated mice is due to elevated 1,25-dihydroxyvitamin D and not to the absence of 24,25-dihydroxyvitamin D. Endocrinology 141:2658–2666. 87. Kobayashi Y, Taguchi T, Terada T, Oshida J, Morisaki M, Ikekawa N 1979 Synthesis of 24,25-difluoro- and 24e-fluoro25-hydroxyvitamin D3. Tetrahedron 22:2023–2026. 88. Yamada S, Ohmori M, Takayama H 1979 Synthesis of 24,24difluoro-25-hydroxyvitamin D3. Tetrahedron 21:1859–1862. 89. Halloran BP, DeLuca HF, Yamada S, Takayama H, Jee WS 1981 An examination of the importance of 24-hydroxylation to the function of vitamin D during early development. Calcif Tissue Int 33:489–497. 90. Tanaka Y, DeLuca HF, Kobayashi Y, Taguchi T, Ikekawa N, Morisaki M 1979 Biological activity of 24,24-difluoro25-hydroxyvitamin D3. Effect of blocking of 24-hydroxylation on the functions of vitamin D. J Biol Chem 254: 7163–7167. 91. Mathews CHE, Parfitt AM, Brommage R, De Luca HF 1983 Only 1,25-dihydroxyvitamin D3 is needed for normal bone growth and mineralization in the rat. Am Soc Bone Miner Res (abstract):S66.

CHAPTER 29 Vitamin D and the Kidney

92. Tanaka Y, Pahuja DN, Wichmann JK, et al. 1982 25-Hydroxy26,26,26,27,27,27-hexafluorovitamin D3: Biological activity in the rat. Arch Biochem Biophys 218:134–141. 93. Shinki T, Jin CH, Nishimura A, et al. 1992 Parathyroid hormone inhibits 25-hydroxyvitamin D3-24-hydroxylase mRNA expression stimulated by 1α,25-dihydroxyvitamin D3 in rat kidney but not in intestine. J Biol Chem 267:13757–13762. 94. Zierold C, Mings JA, DeLuca HF 2001 Parathyroid hormone regulates 25-hydroxyvitamin D3-24-hydroxylase mRNA by altering its stability. Proc Natl Acad Sci USA 98:13572–13576. 95. Zierold C, Reinholz GG, Mings JA, Prahl JM, DeLuca HF 2000 Regulation of the procine 1,25-dihydroxyvitamin D324-hydroxylase (CYP24) by 1,25-dihydroxyvitamin D3 and parathyroid hormone in AOK-B50 cells. Arch Biochem Biophys 81:323–327. 96. Matsumoto T, Kawanobe Y, Ogata E 1985 Regulation of 24,25-dihydroxyvitamin D3 production by 1,25-dihydroxyvitamin D3 and synthetic human parathyroid hormone fragment 1-34 in a cloned monkey kidney cell line (JTC-12). Biochim Biophys Acta 845:358–365. 97. Armbrecht HJ, Boltz MA 1991 Expression of 25-hydroxyvitamin D 24-hydroxylase cytochrome P450 in kidney and intestine. Effect of 1,25-dihydroxyvitamin D and age. FEBS Lett 292:17–20. 98. Gray RW, Omdahl JL, Ghazarian JG, Horst RL 1990 Induction of 25-OH-vitamin D3 24- and 23-hydroxylase activities in partially purified renal extracts from pigs given exogenous 1,25-(OH)2D3. Steroids 55:395–398. 99. Chen ML, Boltz MA, Armbrecht HJ. Effects of 1,25-dihydroxyvitamin D3 and phorbol ester on 25-hydroxyvitamin D3 24-hydroxylase cytochrome P450 messenger ribonucleic acid levels in primary cultures of rat renal cells. Endocrinology 132:1782–1788. 100. Takasaki Y, Horiuchi N, Takahashi N, et al. 1980 Isolation and identificaton of 25-hydroxy-24-oxocholecalciferol: a metabolite of 25-hydroxycholecalciferol. Biochem Biophys Res Commun 95:177–181. 101. Takasaki Y, Suda T, Yamada S, Takayama H, Nishii Y 1981 Isolation, identification, and biological activity of 25hydroxy-24-oxovitamin D3: a new metabolite of vitamin D3 generated by in vitro incubations with kidney homogenates. Biochemistry 20:1681–1686. 102. Wichmann JK, Schnoes HK, DeLuca HF 1981 23,24,25Trihydroxyvitamin D3, 24,25,26-trihydroxyvitamin D3, 24-keto-25-hydroxyvitamin D3, and 23-dehydro-25hydroxyvitamin D3: new in vivo metabolites of vitamin D3. Biochemistry 20:7385–7391. 103. Redel J, Bazely N, Tanaka Y, DeLuca HF 1978 The absolute configuration of the natural 25,26-dihydroxycholecalciferol. FEBS Lett 94:228–230. 104. Partridge JJ, Shiuey SJ, Chadha NK, Baggiolini EG, Blount JF 1981 Synthesis and structure proof of a vitamin D3 metabolite, 25(S),26-dihydroxycholecalciferol. J Am Chem Soc 103:1253–1255. 105. Cesario M, Guilhem J, Pascard C, Redel J 1978 The absolute configuration of C-25 epimers of 25,26-dihydroxycholecalciferol by x-ray diffraction analysis. Tetrahedron 12: 1097–1098. 106. Cesario M, Guilhem J, Pascard C, Redel J 1980 The absolute configuration of C-25 epimers of 25,26-dihydroxycholecalciferol by x-ray diffraction analysis (Erratum). Tetrahedron 21:1588. 107. Horst RL, Littledike ET 1980 25-OHD3-26,23 lactone: demonstration of kidney-dependent synthesis in the pig and rat. Biochem Biophys Res Commun 93:149–154.

531 108. Gray RW, Caldas AE, Weber JL, Ghazarian JG 1978 Biotransformations of 25-hydroxyvitamin D3 by kidney microsomes. Biochem Biophys Res Commun 82:212–218. 109. Tanaka Y, Shepard RA, DeLuca HF, Schnoes HK 1978 The 26-hydroxylation of 25-hydroxyvitamin D3 in vitro by chick renal homogenates. Biochem Biophys Res Commun 83:7–13. 110. Horst RL 1979 25-OH D3-26,23-lactone: a metabolite of vitamin D3 that is 5 times more potent than 25-OH D3 in the rat plasma competitive protein binding radioassay. Biochem Biophy Res Commun 89:286–293. 111. Wichmann JK, DeLuca HF, Schnoes HK, Horst RL, Shepard RM, Jorgensen NA 1979 25-Hydroxyvitamin D3 26,23lactone: a new in vivo metabolite of vitamin D. Biochemistry 18:4775–4780. 112. Ishizuka S, Ishimoto S, Norman AW 1982 Metabolic pathway to 25-hydroxyvitamin D3-26,23-lactone from 25-hydroxyvitamin D3. FEBS Lett 138:83–87. 113. Tanaka Y, Wichmann JK, Schnoes HK, DeLuca HF 1981 Isolation and identification of 23,25-dihydroxyvitamin D3, an in vivo metabolite of vitamin D3. Biochemistry 20:3875–3879. 114. Horst RL, Pramanik BC, Reinhardt TA, et al. 1982 Binding properties of 23S,25-dihydroxyvitamin D3: An in vivo metabolite of vitamin D3. Biochem Biophys Res Commun 106:1006–1011. 115. Partridge JJ, Chadha NK, Shiuey SJ, et al. 1982 Synthesis and structure proof of 23S,25-dihydroxycholecalciferol, a new in vivo vitamin D3 metabolite. In: Norman AW, ed. Vitamin D: Endocrinological Aspects and Their Clinical Applications. Walter de Gruyter, New York, pp. 1073–1078. 116. Morris DS, Williams DH, Norris AF 1981 Structure and synthesis of 25-hydroxycholecalciferol-26,23-lactone, a metabolite of vitamin D3. J Org Chem 46:3422–3428. 117. Morris DS, Williams DH, Norris AF 1981 Structure and synthesis of 25-hydroxycholecalciferol-26,23-lactone, a metabolite of vitamin D3. J Chem Soc Chem Commun 9:424–425. 118. Tanaka Y, Wichmann JK, Paaren HE, Schnoes HK, DeLuca HF 1980 Role of kidney tissue in the production of 25-hydroxyvitamin D3-26,23-lactone and 1α,25-dihydroxyvitamin D326,23-lactone. Proc Natl Acad Sci USA 77:6411–6414. 119. Horst RL, Reinhardt TA, Napoli JL 1982 23-keto-25hydroxyvitamin D3 and 23-keto-1,25-dihydroxyvitamin D3: two new vitamin D3 metabolites with high affinity for the 1,25-dihydroxyvitamin D3 receptor. Biochem Biophys Res Commun 107:1319–1325. 120. Horst RL, Reinhardt TA, Pramanik BC, Napoli JL 1983 23-Keto-25-hydroxyvitamin D3: a vitamin D3 metabolite with high affinity for the 1,25-dihydroxyvitamin D specific cytosol receptor. Biochemistry 22:245–250. 121. Kumar R 1984 The metabolism of dihydroxylated vitamin D metabolites. In: Kumar R (ed) Vitamin D: Basic and Clinical Aspects. Martinus Nirjhoff, Boston, pp. 69–90. 122. Frolik CA, DeLuca HF 1972 Metabolism of 1,25-dihydroxycholecalciferol in the rat. J Clin Invest 51:2900-2906. 123. Frolik CA, DeLuca HF 1973 The stimulation of 1,25dihydroxycholecalciferol treatment. J Clin Invest 52:543–548. 124. Kumar R, Harnden D, DeLuca HF 1976 Metabolism of 1,25dihydroxyvitamin D3: evidence for side-chain oxidation. Biochemistry 15:2420–2423. 125. Esvelt RP, Schnoes HK, DeLuca HF 1979 Isolation and characterization of 1α-hydroxy-23-carboxytetranorvitamin D: a major metabolite of 1,25-dihydroxyvitamin D3. Biochemistry 18:3977–3983. 126. Ohnuma N, Kruse JR, Popjak G, Norman AW 1982 Isolation and chemical characterization of two new vitamin D

532

127.

128.

129. 130. 131. 132.

133.

134. 135.

136. 137.

138.

139.

140.

141.

142. 143.

PETER TEBBEN AND RAJIV KUMAR

metabolites produced by the intestine. 1,25-Dihydroxy-23oxo-vitamin D3 and 1,25,26-trihydroxy-23-oxo-vitamin D3. J Biol Chem 257:5097–5102. Ohnuma N, Norman AW 1982 Identification of a new C-23 oxidation pathway of metabolism for 1,25-dihydroxyvitamin D3 present in intestine and kidney. J Biol Chem 257: 8261-8271. Reinhardt TA, Napoli JL, Beitz DC, Littledike ET, Horst RL 1981 A new in vivo metabolite of vitamin D3: 1,25,26trihydroxyvitamin D3. Biochem Biophys Res Commun 99: 302–307. Kumar R, Nagubandi S, Mattox VR, Londowski JM 1980 Enterohepatic physiology of 1,25-dihydroxyvitamin D3. J Clin Invest 65:277–284. Kumar R, Nagubandi S, Londowski JM 1981 Production of a polar metabolite of 1,25-dihydroxyvitamin D3 in a rat liver perfusion system. Digest Dis Sci 26:242–246. Litwiller RD, Mattox VR, Jardine I, Kumar R 1982 Evidence for a monoglucuronide of 1,25-dihydroxyvitamin D3 in rat bile. J Biol Chem 257:7491–7494. Gray RW, Caldas AE, Wilz DR, Lemann J Jr., Smith GA, DeLuca HF 1978 Metabolism and excretion of 3H-1,25(OH)2-vitamin D3 in healthy adults. J Clin Endocrinol Metab 46:756–765. Seeman E, Kumar R, Hunder GG, Scott M, Heath H, 3rd, Riggs BL 1980 Production, degradation, and circulating levels of 1,25-dihydroxyvitamin D in health and in chronic glucocorticoid excess. J Clin Invest 66:664–669. Wiesner RH, Kumar R, Seeman E, Go VL 1980 Enterohepatic physiology of 1,25-dihydroxyvitamin D3 metabolites in normal man. J Lab Clin Med 96:1094–1100. Harnden D, Kumar R, Holick MF, Deluca HF 1976 Side chain metabolism of 25-hydroxy-[26,27-14C]vitamin D3 and 1,25-dihydroxy-[26,27-14C]vitamin D3 in vivo. Science 193:493–494. Puschett JB, Beck WS Jr. 1975 Parathyroid hormone and 25hydroxy vitamin D3: synergistic and antagonistic effects on renal phosphate transport. Science 190:473–475. Puschett JB, Beck WS Jr., Jelonek A, Fernandez PC 1974 Study of the renal tubular interactions of thyrocalcitonin, cyclic adenosine 3′,5′-monophosphate, 25-hydroxycholecalciferol, and calcium ion. J Clin Invest 53:756–767. Puschett JB, Moranz J, Kurnick WS 1972 Evidence for a direct action of cholecalciferol and 25-hydroxycholecalciferol on the renal transport of phosphate, sodium, and calcium. J Clin Invest 51:373–385. Puschett JB, Fernandez PC, Boyle IT, Gray RW, Omdahl JL, DeLuca HF 1972 The acute renal tubular effects of 1,25-dihydroxycholecalciferol. Proc Soc Exp Biol Med 141:379–384. Siegfried D, Kumar R, Arrudag, Kurtzman N 1976 Influence of vitamin D on bicarbonate reabsorption, In: Proceedings of the 2nd International Congress on Phosphate, Heidelberg. Plenum, New York. Popovtzer MM, Robinette JB, DeLuca HF, Holick MF. The acute effect of 25-hydroxycholecalciferol on renal handling of phosphorus. Evidence for a parathyroid hormone-dependent mechanism. J Clin Invest 53:913–921. Yamamoto M, Kawanobe Y, Takahashi H, Shimazawa E, Kimura S, Ogata E 1984 Vitamin D deficiency and renal calcium transport in the rat. J Clin Invest 74:507–513. Sneddon WB, Barry EL, Coutermarsh BA, Gesek FA, Liu F, Friedman PA 1998 Regulation of renal parathyroid hormone receptor expression by 1,25-dihydroxyvitamin D3 and retinoic acid. Cell Physiol Biochem 8:261–277.

144. Winaver J, Sylk DB, Robertson JS, Chen TC, Puschett JB 1980 Micropuncture study of the acute renal tubular effects of 25-hydroxyvitamin D3 in the dog. Mineral Electrolyte Metab 4:178–188. 145. Sutton RAL, N.L.M. W, Dirks JS 1975 25-Hydroxyvitamin D3 (25(OH)D3): Enhancement of distal tubular calcium reabsorption in the dog. 8th Annual Meeting of the American Society of Nephrology. 146. Harris CA, Seely JF 1979 Effects of 1,25(OH)2D3 on renal handling of electrolytes in the thyroparathyroidectomized (TPTX) D-deficient rat. Clin Res 27:417A. 147. Bouhtiauy I, Lajeunesse D, Brunette MG 1993 Effect of vitamin D depletion on calcium transport by the luminal and basolateral membranes of the proximal and distal nephrons. Endocrinology 132:115–120. 148. Bindels RJ, Hartog A, Timmermans J, Van Os CH 1991 Active Ca2+ transport in primary cultures of rabbit kidney CCD: stimulation by 1,25-dihydroxyvitamin D3 and PTH. Am J Physiol 261:F799–F807. 149. Hoenderop JG, Muller D, Van Der Kemp AW, et al. 2001 Calcitriol controls the epithelial calcium channel in kidney. J Am Soc Nephrol 12:1342–1349. 150. Glendenning P, Ratajczak T, Dick IM, Prince RL 2000 Calcitriol upregulates expression and activity of the 1β isoform of the plasma membrane calcium pump in immortalized distal kidney tubular cells. Arch Biochem Biophys 380: 126–132. 151. Borke JL, Minami J, Verma A, Penniston JT, Kumar R 1987 Monoclonal antibodies to human erythrocyte membrane Ca++-Mg++ adenosine triphosphatase pump recognize an epitope in the basolateral membrane of human kidney distal tubule cells. J Clin Invest 80:1225–1231. 152. Borke JL, Caride A, Verma AK, Penniston JT, Kumar R 1989 Plasma membrane calcium pump and 28-kDa calcium binding protein in cells of rat kidney distal tubules. Am J Physiol 257:F842–F849. 153. Borke JL, Minami J, Verma AK, Penniston JT, Kumar R 1988 Co-localization of erythrocyte Ca++-Mg++ ATPase and vitamin D-dependent 28-kDa-calcium binding protein. Kidney Int 34:262–267. 154. Johnson JA, Kumar R 1994 Renal and intestinal calcium transport: roles of vitamin D and vitamin D-dependent calcium binding proteins. Semin Nephrol 14:119–128. 155. Kumar R, Schaefer J, Grande JP, Roche PC 1994 Immunolocalization of calcitriol receptor, 24-hydroxylase cytochrome P450, and calbindin D28k in human kidney. Am J Physiol 266:F477–F485. 156. Canaff L, Hendy GN 2002 Human calcium-sensing receptor gene. Vitamin D response elements in promoters P1 and P2 confer transcriptional responsiveness to 1,25-dihydroxyvitamin D. J Biol Chem 277:30337–30350. 157. Wald H, Dranitzki-Elhalel M, Backenroth R, Popovtzer MM 1998 Evidence for interference of vitamin D with PTH/PTHrP receptor expression in opossum kidney cells. Pflüg Arch Eur J Physiol 436:289–294. 158. Ozono K, Sone T, Pike JW 1991 The genomic mechanism of action of 1,25-dihydroxyvitamin D3. J Bone Miner Res 6:1021–1027. 159. Pike JW 1991 Vitamin D3 receptors: Structure and function in transcription. Ann Rev Nutr 11:189–216. 160. Hughes MR, Malloy PJ, O’Malley BW, Pike JW, Feldman D 1991 Genetic defects of the 1,25-dihydroxyvitamin D3 receptor. J Receptor Res 11:699–716. 161. Feldman D, Malloy PJ 1990 Hereditary 1,25-dihydroxyvitamin D resistant rickets: molecular basis and implications

CHAPTER 29 Vitamin D and the Kidney

162.

163.

164.

165.

166. 167.

168.

169.

170. 171. 172. 173.

174.

175.

176. 177.

178.

for the role of 1,25(OH)2D3 in normal physiology. Mol Cell Endocrinol 72:C57–C62. Malloy PJ, Hochberg Z, Tiosano D, Pike JW, Hughes MR, Feldman D 1990 The molecular basis of hereditary 1,25dihydroxyvitamin D3 resistant rickets in seven related families. J Clin Invest 86:2071–2079. Hughes M, Malloy P, Kieback D, et al. 1989 Human vitamin D receptor mutations: identification of molecular defects in hypocalcemic vitamin D resistant rickets. Adv Exp Med Biol 255:491–503. Ritchie HH, Hughes MR, Thompson ET, et al. 1988 An ochre mutation in the vitamin D receptor gene causes hereditary 1,25-dihydroxyvitamin D3-resistant rickets in three families. Proc Natl Acad Sci USA 86:9783–9787. Sone T, Scott RA, Hughes MR, et al. 1989 Mutant vitamin D receptors which confer hereditary resistance to 1,25-dihydroxyvitamin D3 in humans are transcriptionally inactive in vitro. J Biol Chem 264:20230–20234. Hughes MR, Malloy PJ, Kieback DG, et al. 1988 Point mutations in the human vitamin D receptor gene associated with hypocalcemic rickets. Science 242:1702–1705. Stumpf WE, Sar M, Narbaitz R, Reid FA, DeLuca HF, Tanaka Y 1980 Cellular and subcellular localization of 1,25(OH)2-vitamin D3 in rat kidney: Comparison with localization of parathyroid hormone and estradiol. Proc Natl Acad Sci USA 77:1149–1153. Stumpf WE, Sar M, Reid FA, Tanaka Y, DeLuca HF 1979 Target cells for 1,25-dihydroxyvitamin D3 in intestinal tract, stomach, kidney, skin, pituitary, and parathyroid. Science 206:1188–1190. Kawashima H, Kurokawa K 1982 Localization of receptors for 1,25-dihydroxyvitamin D3 along the rat nephron. Direct evidence for presence of the receptors in both proximal and distal nephron. J Biol Chem 257:13428–13432. Liu L, Ng M, Iacopino AM, Dunn ST, Hughes MR, Bourdeau JE 1994 Vitamin D receptor gene expression in mammalian kidney. J Am Soc Nephrol 5:1251–1258. Liu L, Khastgir A, McCauley JM, et al. 1996 RT-PCR microlocalization of mRNAs for calbindin D28k and vitamin D receptor in the murine nephron. Am J Physiol 270:F677–F681. Solvsten H, Fogh K, Svendsen M, et al. 1996 Normal levels of the vitamin D receptor and its message in psoriatic skin. J Investigative Dermatology. Symp Proc 1:28–32. Hoenderop JG, van der Kemp AW, Hartog A, et al. 1999 Molecular identification of the apical Ca2+ channel in 1,25-dihydroxyvitamin D3-responsive epithelia. J Biol Chem 274:8375–8378. Johnson JA, Grande JP, Roche PC, Sweeney WE, Jr., Avner ED, Kumar R 1995 1α,25-dihydroxyvitamin D3 receptor ontogenesis in fetal renal development. Am J Physiol 269:F419–F428. Li YC, Pirro AE, Demay MB 1998 Analysis of vitamin D–dependent calcium-binding protein messenger ribonucleic acid expression in mice lacking the vitamin D receptor. Endocrinology 139:847–851. Kumar R 1996 Abnormalities of the vitamin D receptor in uraemia. Nephrol Dialysis Transplant 11(Suppl 3):6–10. Goff JP, Reinhardt TA, Beckman MJ, Horst RL 1990 Contrasting effects of exogenous 1,25-dihydroxyvitamin D [1,25-(OH)2D] versus endogenous 1,25-(OH)2D, induced by dietary calcium restriction, on vitamin D receptors. Endocrinology 126:1031–1035. Uhland-Smith A, DeLuca HF 1993 The necessity for calcium for increased renal vitamin D receptor in response to 1,25dihydroxyvitamin D. Biochem Biophys Acta 1176:321–326.

533 179. Reinhardt TA, Horst RL 1990 Parathyroid hormone downregulates 1,25-dihydroxyvitamin D receptors (VDR) and VDR messenger ribonucleic acid in vitro and blocks homologous up-regulation of VDR in vivo. Endocrinology 127:942–948. 180. Santiso-Mere D, Sone T, Hilliard GMt, Pike JW, McDonnell DP Positive regulation of the vitamin D receptor by its cognate ligand in heterologous expression systems. Mol Endocrinol 7:833–839. 181. Wiese RJ, Uhland-Smith A, Ross TK, Prahl JM, DeLuca HF. Up-regulation of the vitamin D receptor in response to 1,25dihydroxyvitamin D3 results from ligand-induced stabilization. J Biol Chem 267:20082–20086. 182. Gross M, Kumar R 1990 Physiology and biochemistry of vitamin D–dependent calcium binding proteins. Am J Physiol 259:F195–F209. 183. Desplan C, Heidmann O, Lillie JW, Auffray C, Thomasset M 1983 Sequence of rat intestinal vitamin D–dependent calcium-binding protein derived from a cDNA clone. Evolutionary implications. J Biol Chem 258:13502–13505. 184. Desplan C, Thomasset M, Moukhtar M 1983 Synthesis, molecular cloning, and restriction analysis of DNA complementary to vitamin D–dependent calcium-binding protein mRNA from rat duodenum. J Biol Chem 258:2762–2765. 185. Fullmer CS, Wasserman RH 1987 Chicken intestinal 28-kilodalton calbindin-D: Complete amino acid sequence and structural considerations. Proc Natl Acad Sci USA 84: 4772–4776. 186. Hofmann T, Kawakami M, Hitchman AJ, Harrison JE, Dorrington KJ 1979 The amino acid sequence of porcine intestinal calcium-binding protein. Can J Biochem 57: 737–748. 187. Horrocks WD 1982 Lanthanide ion probes of biomolecular structure. In: Eichorn GL, Marzitti LG (eds) Advances in Inorganic Biochemistry. Elsevier, New York, pp. 201–261. 188. Hunziker W 1986 The 28-kDa vitamin D–dependent calciumbinding protein has a six-domain structure. Proc Natl Acad Sci USA 83:7578–7582. 189. Hunziker W, Schrickel S 1988 Rat brain calbindin D28: Six domain structure and extensive amino acid homology with chicken calbindin D28. Mol Endocrinol 2:465–473. 190. Kumar R, Wieben E, Beecher SJ 1989 The molecular cloning of the complementary deoxyribonucleic acid for bovine vitamin D–dependent calcium-binding protein: structure of the full-length protein and evidence for homologies with other calcium-binding proteins of the troponin-C superfamily of proteins. Mol Endocrinol 3:427–432. 191. Lomri N, Perret C, Gouhier N, Thomasset M 1989 Cloning and analysis of calbindin-D28K cDNA and its expression in the central nervous system. Gene 80:87–98. 192. Parmentier M, Lawson DE, Vassart G 1987 Human 27-kDa calbindin complementary DNA sequence. Evolutionary and functional implications. Eur J Biochem 170:207–215. 193. Takagi T, Nojiri M, Konishi K, Maruyama K, Nonomura Y 1986 Amino acid sequence of vitamin D-dependent calciumbinding protein from bovine cerebellum. FEBS Lett 201: 41–45. 194. Tsarbopoulos A, Gross M, Kumar R, Jardine I Rapid identification of calbindin-D28k cyanogen bromide peptide fragments by plasma desorption mass spectrometry. Biomed Environ Mass Spectrom 18:387–393. 195. Wilson PW, Harding M, Lawson DE 1985 Putative amino acid sequence of chick calcium-binding protein deduced from a complementary DNA sequence. Nucleic Acids Res 13:8867–8881.

534 196. Yamakuni T, Kuwano R, Odani S, Miki N, Yamaguchi Y, Takahashi Y 1986 Nucleotide sequence of cDNA to mRNA for a cerebellar Ca-binding protein, spot 35 protein. Nucleic Acids Res 14:6768. 197. Darwish HM, Krisinger J, Strom M, DeLuca HF 1987 Molecular cloning of the cDNA and chromosomal gene for vitamin D-dependent calcium-binding protein of rat intestine. Proc Natl Acad Sci USA 84:6108–6111. 198. Fullmer CS, Wasserman RH 1981 The amino acid sequence of bovine intestinal calcium-binding protein. J Biol Chem 256:5669–5674. 199. MacManus JP, Watson DC, Yaguchi M 1986 The purification and complete amino acid sequence of the 9000-Mr Ca2+binding protein from rat placenta. Identity with the vitamin D-dependent intestinal Ca2+-binding protein. Biochem J 235:585–595. 200. Szebenyi DM, Obendorf SK, Moffat K 1981 Structure of vitamin D–dependent calcium-binding protein from bovine intestine. Nature 294:327–332. 201. Elms TN, Taylor AN 1987 Calbindin-D28 kappa localization in rat molars during odontogenesis. J Dental Res 66: 1431–1434. 202. Dorrington KJ, Kells DI, Hitchman AJ, Hartison JE, Hofmann T 1978 Spectroscopic studies on the binding of divalent cations to porcine intestinal calcium-binding protein. Can J Biochem 56:492–499. 203. O’Neil JD, Dorrington KJ, Hofmann T 1984 Luminescence and circular-dichroism analysis of terbium binding by pig intestinal calcium-binding protein (relative mass = 9000). Can J Biochem Cell Biol 62:434–442. 204. Chiba K, Ohyashiki T, Mohri T 1983 Quantitative analysis of calcium binding to porcine intestinal calcium-binding protein. J Biochem 93:487–493. 205. Vogel HJ, Drakenberg T, Forsen S, O’Neil JD, Hofmann T 1985 Structural differences in the two calcium binding sites of the porcine intestinal calcium binding protein: a multinuclear NMR study. Biochemistry 24:3870–3876. 206. Shelling JG, Sykes BD 1985 1H nuclear magnetic resonance study of the two calcium-binding sites of porcine intestinal calcium-binding protein. J Biol Chem 260:8342–8347. 207. Bredderman PJ, Wasserman RH 1974 Chemical composition, affinity for calcium, and some related properties of the vitamin D dependent calcium-binding protein. Biochemistry 13:1687–1694. 208. Fullmer CS, Edelstein S, Wasserman RH 1985 Lead-binding properties of intestinal calcium-binding proteins. J Biol Chem 260:6816–6819. 209. Gross MD, Nelsestuen GL, Kumar R 1987 Observations on the binding of lanthanides and calcium to vitamin D-dependent chick intestinal calcium-binding protein. Implications regarding calcium-binding protein function. J Biol Chem 262:6539–6545. 210. Gross MD, Kumar R, Hunziker W 1988 Expression in Escherichia coli of full-length and mutant rat brain calbindin D28. Comparison with the purified native protein. J Biol Chem 263:14426–14432. 211. Veenstra TD, Gross MD, Hunziker W, Kumar R 1995 Identification of metal-binding sites in rat brain calciumbinding protein. J Biol Chem 270:30353–30358. 212. Kumar R, Hunziker W, Gross M, Naylor S, Londowski JM, Schaefer J 1994 The highly efficient production of fulllength and mutant rat brain calcium-binding proteins (calbindins-D28K) in a bacterial expression system. Arch Biochem Biophys 308:311–317.

PETER TEBBEN AND RAJIV KUMAR

213. Wasserman RH, Corradino RA, Fullmer CS, Taylor AN 1974 Some aspects of vitamin D action calcium absorption and the vitamin D–dependent calcium-binding protein. Vitamins Horm 32:299–324. 214. Christakos S, Brunette MG, Norman AW 1981 Localization of immunoreactive vitamin D–dependent calcium binding protein in chick nephron. Endocrinology 109:322–324. 215. Taylor AN, Wasserman RH 1967 Vitamin D3–induced calciumbinding protein: partial purification, electrophoretic visualization, and tissue distribution. Arch Biochem Biophys 119: 536–540. 216. Taylor AN, McIntosh JE, Bourdeau JE 1982 Immunocytochemical localization of vitamin D-dependent calcium-binding protein in renal tubules of rabbit, rat, and chick. Kidney Int 21:765–773. 217. Roth J, Thorens B, Hunziker W, Norman AW, Orci L 1981 Vitamin D–dependent calcium binding protein: immunocytochemical localization in chick kidney. Science 214: 197–200. 218. Roth J, Brown D, Norman AW, Orci L 1982 Localization of the vitamin D–dependent calcium-binding protein in mammalian kidney. Am J Physiol 243:F243–F252. 219. Thomasset M, Parkes CO, Cuisinier-Gleizes P 1982 Rat calcium-binding proteins: distribution, development, and vitamin D dependence. Am J Physiol 243:E483–E488. 220. Hemmingsen C, Staun M, Nielsen PK, Olgaard K 2000 Separate effects of 1,25-dihydroxyvitamin D and calcium on renal calbindin-D28k and intestinal calbindin-D9k. Pharmacol Toxicol 91:111–115. 221. Hoenderop JG, Dardenne O, Van Abel M, et al. 2002 Modulation of renal Ca2+ transport protein genes by dietary Ca2+ and 1,25-dihydroxyvitamin D3 in 25-hydroxyvitamin D3-1α-hydroxylase knockout mice. FASEB J 16: 1398–1406. 222. Li YC, Bolt MJ, Cao LP, Sitrin MD 2001 Effects of vitamin D receptor inactivation on the expression of calbindins and calcium metabolism. Am J Physiol Endocrinol Metab 281:E558–E564. 223. Cao LP, Bolt MJ, Wei M, Sitrin MD, Chun Li Y 2002 Regulation of calbindin-D9k expression by 1,25-dihydroxyvitamin D3 and parathyroid hormone in mouse primary renal tubular cells. Arch Biochem Biophys 400:118–124. 224. Hemmingsen C, Staun M, Lewin E, Nielsen PK, Olgaard K 1996 Effect of parathyroid hormone on renal calbindin-D28k. J Bone Miner Res 11:1086–1093. 225. Bouhtiauy I, Lajeunesse D, Christakos S, Brunette MG 1994 Two vitamin D3–dependent calcium binding proteins increase calcium reabsorption by different mechanisms. I. Effect of CaBP 28K. Kidney Int 45:461–468. 226. Bouhtiauy I, Lajeunesse D, Christakos S, Brunette MG 1994 Two vitamin D3–dependent calcium binding proteins increase calcium reabsorption by different mechanisms. II. Effect of CaBP 9K. Kidney Int 45:469–474. 227. Lee CT, Huynh VM, Lai LW, Lien YH 2002 Cyclosporine A– induced hypercalciuria in calbindin-D28k knockout and wildtype mice. Kidney Int 62:2055–2061. 228. Sooy K, Kohut J, Meyer M, Yeh C, Hou T, Christakos S 1999 Increased urinary calcium excretion in calbindin-D28k knockout mice. J Bone Min Res 14(Suppl. 1):S211. 229. Lutz W, Berndt T, Knox FG, Kumar R 2000 Normal fractional excretion of calcium (Ca) in calbindin D28K null mutant mice. FASEB J 14:A343. 230. Doucet A, Katz AI 1982 High-affinity Ca-Mg-ATPase along the rabbit nephron. Am J Physiol 242:F346–F352.

CHAPTER 29 Vitamin D and the Kidney

231. Borke JL, Caride A, Verma AK, Penniston JT, Kumar R 1990 Cellular and segmental distribution of Ca2+-pump epitopes in rat intestine. Pflüg Arch Eur J Physiol 417:120–122. 232. Wasserman RH, Smith CA, Brindak ME, et al. 1992 Vitamin D and mineral deficiencies increase the plasma membrane calcium pump of chicken intestine. Gastroenterology 102:886–894. 233. Borke JL, Caride A, Verma AK, et al. 1989 Calcium pump epitopes in placental trophoblast basal plasma membranes. Am J Physiol 257:c341–c346. 234. Borke JL, Caride AJ, Yaksh TL, Penniston JT, Kumar R 1989 Cerebrospinal fluid calcium homeostasis: evidence for a plasma membrane Ca2+-pump in mammalian choroid plexus. Brain Res 489:355–360. 235. Wasserman RH, Smith CA, Smith CM, et al. 1991 Immunohistochemical localization of a calcium pump and calbindin-D28k in the oviduct of the laying hen. Histochemistry 96:413–418. 236. Borke JL, Eriksen EF, Minami J, et al. 1988 Epitopes of the human erythrocyte Ca2+-Mg2+ ATPase pump in human osteoblast-like cell plasma membranes. J Clin Endocrinol Metab 67:1299–1304. 237. Kumar R, Haugen JD, Medora M, Yamakawa K, Hruska KA 1993 Detection and partial sequencing of a PMCa2+ pump in the chicken osteoclast. J Bone Min Res 8:S386. 238. Cai Q, Chandler JS, Wasserman RH, Kumar R, Penniston JT 1993 Vitamin D and adaptation to dietary calcium and phosphate deficiencies increase intestinal plasma membrane calcium pump gene expression. Proc Natl Acad Sci USA 90:1345–1349. 239. Wasserman RH, Chandler JS, Meyer SA, et al. 1992 Intestinal calcium transport and calcium extrusion processes at the basolateral membrane. J Nutr 122:662–671. 240. Blankenship KA, Williams JJ, Lawrence MS, McLeish KR, Dean WL, Arthur JM 2001 The calcium-sensing receptor regulates calcium absorption in MDCK cells by inhibition of PMCA. Am J Physiol Renal Fluid & Electrolyte Physiol 280:F815–F822. 241. Hoenderop JG, Nilius B, Bindels RJ 2002 Molecular mechanism of active Ca2+ reabsorption in the distal nephron. Ann Rev Physiol 64:529–549. 242. Peng JB, Hediger MA 2002 A family of calcium-permeable channels in the kidney: distinct roles in renal calcium handling. Curr Opin Nephrol Hypertens 11:555–561. 243. Muller D, Hoenderop JG, Meij IC, et al. 2000 Molecular cloning, tissue distribution, and chromosomal mapping of the human epithelial Ca2+ channel (ECAC1). Genomics 67:48–53. 244. Weber K, Erben RG, Rump A, Adamski J 2001 Gene structure and regulation of the murine epithelial calcium channels ECaC1 and 2. Biochem Biophys Res Commun 289:1287–1294. 245. Muller D, Hoenderop JG, van Os CH, R JMB 2001 The epithelial calcium channel, ECaC1: molecular details of a novel player in renal calcium handling. Nephrol Dialysis Transplant 16:1329–1335. 246. Boyle IT, Gray RW, DeLuca HF 1971 Regulation by calcium of in vivo synthesis of 1,25-dihydroxycholecalciferol and 21,25-dihydroxycholecalciferol. Proc Natl Acad Sci USA 68:2131–2134. 247. Tanaka Y, Castillo L, DeLuca HF 1976 Control of renal vitamin D hydroxylases in birds by sex hormones. Proc Natl Acad Sci USA 73:2701–2705. 248. Tanaka Y, Deluca HF 1973 The control of 25-hydroxyvitamin D metabolism by inorganic phosphorus. Arch Biochem Biophys 154:566–574.

535 249. Iida K, Taniguchi S, Kurokawa K 1993 Distribution of 1,25dihydroxyvitamin D3 receptor and 25-hydroxyvitamin D324-hydroxylase mRNA expression along rat nephron segments. Biochem Biophys Res Commun 194:659–664. 250. Murayama A, Takeyama K, Kitanaka S, et al. 1999 Positive and negative regulations of the renal 25-hydroxyvitamin D3 1α-hydroxylase gene by parathyroid hormone, calcitonin, and 1α,25(OH)2D3 in intact animals. Endocrinology 140:2224–2231. 251. Shinki T, Ueno Y, DeLuca HF, Suda T 1999 Calcitonin is a major regulator for the expression of renal 25-hydroxyvitamin D3-1α-hydroxylase gene in normocalcemic rats. Proc Natl Acad Sci USA 96:8253–8258. 252. Adams ND, Gray RW, Lemann J Jr. 1979 The calciuria of increased fixed acid production in humans: evidence against a role for parathyroid hormone and 1,25(OH)2-vitamin D. Calcif Tissue Int 28:233–238. 253. Bilezikian JP, Canfield RE, Jacobs TP, et al. 1978 Response of 1α,25-dihydroxyvitamin D3 to hypocalcemia in human subjects. N Eng J Med 299:437–441. 254. Boyle IT, Gray RW, Omdahl JL, DeLuca HF 1972 Calcium control of the in vivo biosynthesis of 1,25-dihydroxyvitamin D3: Nicolaysen’s endogenous factor. In: Proceedings of the International Symposium on Endocrinology, Heinemann, London. 255. Drezner MK, Neelon FA, Haussler M, McPherson HT, Lebovitz HE 1976 1,25-Dihydroxycholecalciferol deficiency: The probable cause of hypocalcemia and metabolic bone disease in pseudohypoparathyroidism. J Clin Endocrinol Metab 42:621–628. 256. Fraser DR, Kodicek E 1973 Regulation of 25-hydroxycholecalciferol-1-hydroxylase activity in kidney by parathyroid hormone. Nat New Biol 241:163–166. 257. Garabedian M, Holick MF, Deluca HF, Boyle IT 1972 Control of 25-hydroxycholecalciferol metabolism by parathyroid glands. Proc Natl Acad Sci USA 69:1673–1676. 258. Omdahl JL, Gray RW, Boyle IT, Knutson J, DeLuca HF 1972 Regulation of metabolism of 25-hydroxycholecalciferol by kidney tissue in vitro by dietary calcium. Nat New Biol 237:63–64. 259. Gray RW, Wilz DR, Caldas AE, Lemann J Jr. 1977 The importance of phosphate in regulating plasma 1,25-(OH)2vitamin D levels in humans: Studies in healthy subjects in calcium-stone formers and in patients with primary hyperparathyroidism. J Clin Endocrinol Metab 45:299–306. 260. Yoshida T, Yoshida N, Monkawa T, Hayashi M, Saruta T 2001 Dietary phosphorus deprivation induces 25-hydroxyvitamin D3 1α-hydroxylase gene expression. Endocrinology 142:1720–1726. 261. Zhang MY, Wang X, Wang JT, et al. 2002 Dietary phosphorus transcriptionally regulates 25-hydroxyvitamin D-1αhydroxylase gene expression in the proximal renal tubule. Endocrinology 143:587–595. 262. Kawashima H, Torikai S, Kurokawa K 1981 Calcitonin selectively stimulates 25-hydroxyvitamin D3-1α-hydroxylase in proximal straight tubule of rat kidney. Nature 291:327–329. 263. Bland R, Walker EA, Hughes SV, Stewart PM, Hewison M 1999 Constitutive expression of 25-hydroxyvitamin D3-1αhydroxylase in a transformed human proximal tubule cell line: Evidence for direct regulation of vitamin D metabolism by calcium. Endocrinology 140:2027–2034. 264. Lund B, Sorensen OH, Bishop JE, Norman AW 1980 Stimulation of 1,25-dihydroxyvitamin D production by parathyroid hormone and hypocalcemia in man. J Clin Endocrinol Metab 50:480–484.

536 265. Yoshida N, Yoshida T, Nakamura A, Monkawa T, Hayashi M, Saruta T 1999 Calcitonin induces 25-hydroxyvitamin D3 1α-hydroxylase mRNA expression via protein kinase C pathway in LLC-PK1 cells. J Am Soc Nephrol 10:2474–2479. 266. Adams ND, Gray RW, Lemann J Jr. 1979 The effects of oral CaCO3 loading and dietary calcium deprivation on plasma 1,25-dihydroxyvitamin D concentrations in healthy adults. J Clin Endocrinol Metab 48:1008–1016. 267. Sauveur B, Garabedian M, Fellot C, Mongin P, Balsan S 1977 The effect of induced metabolic acidosis on vitamin D3 metabolism in rachitic chicks. Calcif Tissue Res 23:121–124. 268. Weber HP, Gray RW, Dominguez JH, Lemann J Jr. 1976 The lack of effect of chronic metabolic acidosis on 25-OHvitamin D metabolism and serum parathyroid hormone in humans. J Clin Endocrinol Metab 43:1047–1055. 269. Castillo L, Tanaka Y, DeLuca HF, Sunde ML 1977 The stimulation of 25-hydroxyvitamin D3-1α-hydroxylase by estrogen. Arc Biochem Biophys 179:211–217. 270. Adams ND, Garthwaite TL, Gray RW, Hagen TC, Lemann J Jr. 1979 The interrelationships among prolactin, 1,25-dihydroxyvitamin D, and parathyroid hormone in humans. J Clin Endocrinol Metab 49:628–630. 271. Kumar R, Cohen WR, Epstein FH 1980 Vitamin D and calcium hormones in pregnancy. N Eng J Med 302:1143–1145. 272. Spanos E, Colston KW, Evans IM, Galante LS, Macauley SJ, Macintyre I 1976 Effect of prolactin on vitamin D metabolism. Mol Cell Endocrinol 5:163–167. 273. Eskildsen PC, Lund B, Sorensen OH, Bishop JE, Norman AW 1979 Acromegaly and vitamin D metabolism: Effect of bromocriptine treatment. J Clin Endocrinol Metab 49: 484–486. 274. Gertner JM, Horst RL, Broadus AE, Rasmussen H, Genel M 1979 Parathyroid function and vitamin D metabolism during human growth hormone replacement. J Clin Endocrinol Metab 49:185–188. 275. Menaa C, Vrtovsnik F, Friedlander G, Corvol M, Garabedian M 1995 Insulin-like growth factor I, a unique calciumdependent stimulator of 1,25-dihydroxyvitamin D3 production. Studies in cultured mouse kidney cells. J Biol Chem 270: 25461–25467. 276. Condamine L, Menaa C, Vrtovsnik F, Vztovsnik F, Friedlander G, Garabedian M 1994 Local action of phosphate depletion and insulin-like growth factor 1 on in vitro production of 1,25-dihydroxyvitamin D by cultured mammalian kidney cells. [Erratum appears in J Clin Invest 1995 95(1):following 434]. J Clin Invest 94:1673–1679.

PETER TEBBEN AND RAJIV KUMAR

277. Kumar R, Merimee TJ, Silva P, Epstein FH 1979 The effect of chronic growth hormone excess or deficiency on plasma 1,25-dihydroxyvitamin D levels in man. In: Proceedings of the Fourth Workshop on Vitamin D. De Gruyter, Elmsford, New York. 278. Bianda T, Glatz Y, Bouillon R, Froesch ER, Schmid C 1998 Effects of short-term insulin-like growth factor-I (IGF-I) or growth hormone (GH) treatment on bone metabolism and on production of 1,25-dihydroxycholecalciferol in GH-deficient adults. J Clin Endocrinol Metab 83:81–87. 279. Carre M, Ayigbede O, Miravet L, Rasmussen H 1974 The effect of Prednisolone upon the metabolism and action of 25hydroxy- and 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 71:2996–3000. 280. Favus MJ, Walling MW, Kimberg DV 1973 Effects of 1,25dihydroxycholecalciferol on intestinal calcium transport in cortisone-treated rats. J Clin Invest 52:1680–1685. 281. Feher JJ, Wasserman RH 1979 Intestinal calcium-binding protein and calcium absorption in cortisol-treated chicks: effects of vitamin D3 and 1,25-dihydroxyvitamin D3. Endocrinology 104:547–551. 282. Akeno N, Matsunuma A, Maeda T, Kawane T, Horiuchi N 2000 Regulation of vitamin D-1α-hydroxylase and -24hydroxylase expression by dexamethasone in mouse kidney. J Endocrinol 164:339–348. 283. Bouillon R, Muls E, De Moor P 1980 Influence of thyroid function on the serum concentration of 1,25-dihydroxyvitamin D3. J Clin Endocrinol Metab 51:793–797. 284. Pahuja DN, De Luca HF 1982 Thyroid hormone and vitamin D metabolism in the rat. Arch Biochem Biophys 213:293–298. 285. Jastrup B, Mosekilde L, Melsen F, Lund B, Sorensen OH 1982 Serum levels of vitamin D metabolites and bone remodelling in hyperthyroidism. Metab Clin Exp 31:126–132. 286. Saito H, Kusano K, Kinosaki M, et al. 2003 Human fibroblast growth factor-23 mutants suppress Na+-dependent phosphate co-transport activity and 1α,25-dihydroxyvitamin D3 production. J Biol Chem 278:2206–2211. 287. Shimada T, Mizutani S, Muto T, et al. 2001 Cloning and characterization of FGF23 as a causative factor of tumorinduced osteomalacia [comment]. Proc Natl Acad Sci USA 98:6500–6505. 288. Kumar R, Cohen WR, Silva P, Epstein FH 1979 Elevated 1,25-dihydroxyvitamin D plasma levels in normal human pregnancy and lactation. J Clin Invest 63:342–344. 289. Pike JW, Parker JB, Haussler MR, Boass A, Toverud SV 1979 Dynamic changes in circulating 1,25-dihydroxyvitamin D during reproduction in rats. Science 204:1427–1429.

CHAPTER 30

Vitamin D and the Parathyroids JUSTIN SILVER AND TALLY NAVEH-MANY

Hebrew University Hadassah Medical Center, Jerusalem

Introduction Parathyroid Hormone Biosynthesis Parathyroid Hormone Secretion and the Calcium Sensor Regulation of the Parathyroid Hormone Gene The Parathyroid Hormone Gene Regulation of Parathyroid Hormone Gene Expression by Calcitriol VII. Regulation of the Calcium Receptor by Calcitriol

VIII. Calreticulin and the Action of 1,25(OH)2D3 on the Parathyroid Hormone Gene IX. Parathyroid Hormone Degradation X. Secondary Hyperparathyroidism and Parathyroid Cell Proliferation XI. Conclusions References

I. INTRODUCTION

parathyroid cell. The amino acid sequence has been determined in several species, and there exists a high degree of identity among species, particularly in the amino-terminal region of the molecule. The parathyroids synthesize another protein that is also secreted [3]. This protein, secretory protein I, is identical to chromogranin A isolated from the adrenal medulla, and it is present in other endocrine cells and neoplasms as well. Its function is not known, but it is stored and secreted with PTH despite its differential transcriptional regulation relative to PTH. A 26-kDa N-terminal fragment of chromogranin A (CgA) secreted by bovine parathyroid glands, when added to dispersed, parathyroid cells in primary culture, inhibited the low calciumstimulated secretion of both PTH and CgA, suggesting an autocrine or paracrine regulation of secretion [4].

I. II. III. IV. V. VI.

The action of 1,25(OH)2D3 or its analogs to decrease PTH secretion is now a well-established axiom in clinical medicine for the suppression of the secondary hyperparathyroidism of patients in chronic renal failure. So much so, that it is worthwhile to reflect upon its scientific basis. That is the purpose of the present review. There have been more recent developments spurred by pharmaceutical companies to discover drugs that have more selective actions on the parathyroid. These are discussed in other chapters in this book.

II. PARATHYROID HORMONE BIOSYNTHESIS Parathyroid hormone, a protein of 84 amino acids, is synthesized as a larger precursor, preproparathyroid hormone [1,2]. PreproPTH has a 25-residue “pre” or signal sequence, and a six-residue “pro” sequence. The signal sequence, along with the short pro sequence, functions to direct the protein into the secretory pathway. Like other signal sequences, the pre sequence binds to a signal recognition particle during protein synthesis. The signal recognition particle then delivers the nascent peptide chain to the rough endoplasmic reticulum, where it is threaded through a protein-lined aqueous pore. During this transit, the signal sequence is cleaved off by a signal peptidase, and the pre sequence is rapidly degraded. Because the process of transport and cleavage occurs during protein synthesis, very little intact preproPTH is found within the parathyroid cell. Mature PTH has a molecular mass of approximately 9600 Da and is the only form that is secreted from the VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

III. PARATHYROID HORMONE SECRETION AND THE CALCIUM SENSOR The major regulator of PTH secretion is the concentration of ionized calcium in blood and ECF. Increases in the level of calcium lead to a decrease in PTH secretion. The shape of the dose-response curve is sigmoidal. Such a curve can be defined by four parameters: the maximal secretory rate, the slope of the curve at its midpoint, the level of calcium at the midpoint (often called the set point), and the minimal secretory rate [5]. The parathyroid calcium sensor, the calcium-sensing receptor (CaR), has been characterized and cloned and probably mediates the physiological responses of the cell to calcium [6] (see Chapter 31). The deduced amino acid sequence of the receptor suggests that it spans the plasma membrane seven Copyright © 2005, Elsevier, Inc. All rights reserved.

538 times, like other receptors in the G-protein-linked receptor family. A large extracellular domain, presumed to bind calcium, resembles similar domains in brain metabotropic glutamate receptors as well as bacterial periplasmic proteins designed to bind small ligands, including cations. The most convincing proof of the identity of the parathyroid calcium sensor has been the finding that most patients with familial hyopcalciuric hypercalcemia (FHH), a disease of defective calcium sensing, have a variety of inactivating mutations in the CaR gene [7]. Mice genetically engineered to have only one functioning copy of the CaR gene also have the expected defects in parathyroid calcium sensing [8]. The findings in both patients with FHH and in the genetically engineered mice establish a link between the parathyroid calcium sensor, the CaR, and control of PTH secretion. Vitamin D regulation of the CaR is discussed in Section VII.

IV. REGULATION OF THE PARATHYROID HORMONE GENE

JUSTIN SILVER AND TALLY NAVEH-MANY

which represents the exons, is the mature PTH mRNA, which is then translated into preproPTH. The first intron separates the 5′-untranslated region of the mRNA from the rest of the gene, and the second intron separates most of the sequence encoding the precursorspecific “prepro” region from exon 3, which encodes the mature PTH and the 3′-untranslated region (UTR). The three exons that result are thus roughly divided into functional domains. There is considerable identity among the mammalian PTH genes, which is reflected in an 85% identity between the human and bovine proteins and 75% identity between the human and rat protein. There is less identity in the 3′-noncoding region. The two homologous TATA sequences flanking the human PTH gene direct the synthesis of two human PTH gene transcripts both in normal parathyroid glands and in parathyroid adenomas [11]. The termination codon immediately following the codon for glutamine at position 84 of PTH indicates that there are no additional precursors of PTH with peptide extensions at the carboxyl position.

B. Promoter Sequences The minute-to-minute regulation of PTH blood levels can be explained by the calcium sensor and amplification of this regulation by intracellular degradation of stored hormone. Over a longer time frame, the parathyroid cell regulates the expression of the PTH gene to allow the cell to roughly match the production of PTH to the demand. Regulation of PTH biosynthesis has been studied in intact animals and in primary parathyroid cells in culture. In intact animals, it may be difficult to separate direct effects of a manipulation from indirect responses of the animal [9]. There is no parathyroid cell line, and thus the use of primary cells in culture complements the in vivo studies. Primary cells are unstable, however, and rapidly lose their characteristics of parathyroid cells in culture. Despite these experimental limitations, much has been learned about the regulation of the PTH gene.

V. THE PARATHYROID HORMONE GENE A. Gene Structure The PTH genes from different species that have been cloned all have two introns or intervening sequences and three exons [10]. The primary RNA transcript consists of RNA transcribed from both the introns and exons, and then the RNA sequences derived from the introns are spliced out. The product of this RNA processing,

The regions upstream of the transcribed structural gene determine the tissue specificity and contain the regulatory sequences for the gene. For PTH, this analysis has been hampered by the lack of a parathyroid cell line. Rupp et al. analyzed the human PTH promoter region up to position −805 and identified a number of consensus sequences by computer analysis [12]. These included a sequence resembling the canonical cAMP responsive element 5′-TGACGTCA-3′ at position −81 with a single residue deviation. This element was fused to a reporter gene (chloramphenicol acetyltransferase, CAT) and then transfected into different cell lines. Pharmacological agents that increase cAMP led to an increased expression of the CAT gene, suggesting a functional role for the cAMP-responsive element (CRE). The role of this possible CRE in the context of the PTH gene in the parathyroid gland remains to be established. Demay et al. [13] identified DNA sequences in the human PTH gene that bind the 1,25(OH)2D3 receptor (VDR). Nuclear extracts containing the VDR were examined for binding to sequences in the 5′-flanking region of the human PTH gene. A 25-bp oligonucleotide containing the sequences from −125 to −101 from the start of exon 1 bound nuclear proteins that were recognized by monoclonal antibodies against the VDR. The sequences in this region contained a single copy of a motif (AGGTTCA) that is homologous to the motifs repeated in the up-regulatory VDR response element

539

CHAPTER 30 Vitamin D and the Parathyroids

(VDRE) of osteocalcin. When placed upstream of a heterologous viral promoter, the sequences contained in this 25-bp oligonucleotide mediated transcriptional repression in response to 1,25(OH)2D3 in GH4C1 rat pituitary cells but not in ROS 17/2.8 rat osteosarcoma cells. Therefore, this down-regulatory VDRE differs from up-regulatory VDREs both in sequence composition and in the requirement for particular cellular factors other than the VDR for repressing PTH transcription. Further work is needed to demonstrate that this negative VDRE functions in the context of the PTH gene promoter and to establish whether other VDREs control PTH gene expression. Liu et al. [14] have identified such sequences in the chicken PTH gene and demonstrated their functionality after transfection into the opossum kidney OK cell line.

VI. REGULATION OF PARATHYROID HORMONE GENE EXPRESSION BY CALCITRIOL A. Endocrinological Feedback PTH regulates serum concentrations of calcium and phosphate, which, in turn, regulate the synthesis and secretion of PTH. 1,25(OH)2D3 or calcitriol has independent effects on calcium and phosphate levels, and also participates in a well-defined feedback loop between calcitriol and PTH. PTH increases the renal synthesis of calcitriol. Calcitriol then increases blood calcium largely by increasing the efficiency of intestinal calcium absorption. Calcitriol also potently decreases the transcription of the PTH gene. This action was first demonstrated in vitro in bovine parathyroid cells in primary culture, where calcitriol led to a marked decrease in PTH mRNA levels [15,16] and a consequent decrease in PTH secretion [17–19].

B. In Vivo Studies The physiological relevance of these findings was established by in vivo studies in rats [20]. The localization of the VDR mRNA to the parathyroids was demonstrated by in situ hybridization studies of the thyroparathyroid and duodenum. VDR mRNA was localized to the parathyroids in the same concentration as in the duodenum, calcitriol’s classic target organ (Fig. 1) [21]. Rats injected with amounts of calcitriol that did not increase serum calcium had marked decreases in PTH mRNA levels, reaching <4% of control at 48 hr (Fig. 2). There was also a decrease in

calcitonin mRNA levels in these rats [22]. The effect of 1,25(OH)2D3 on the PTH and calcitonin genes was shown to be transcriptional both in in vivo studies in rats [20] and in in vitro studies with primary cultures of bovine parathyroid cells [23]. Interestingly, in rats given large doses of vitamin D with resultant hypercalcemia there was still a decrease in calcitonin mRNA levels despite the elevated serum calcium which is a secretagogue for calcitonin [24].

C. In Vitro Studies When 684 bp of the 5′-flanking region of the human PTH gene linked to a reporter gene and transfected into a rat pituitary cell line (GH4C1), gene expression was lowered by 1,25(OH)2D3 [25]. These studies suggest that 1,25(OH)2D3 decreases PTH transcription by acting on the 5′-flanking region of the PTH gene. The effect of 1,25(OH)2D3 may involve heterodimerization with the retinoid x acid receptor (RXR) (see Chapter 13). This is because 9-cis-retinoic acid, which binds to RXR, when added to bovine parathyroid cells in primary culture, led to a decrease in PTH mRNA levels [26]. Moreover, combined treatment with 1 × 10−6 M retinoic acid and 1 × 10−8 M 1,25(OH)2D3 more effectively decreased PTH secretion and preproPTH mRNA than did either compound alone [26]. Alternatively, RXRs might synergize with VDRs through actions on distinct sequences.

D. The Vitamin D Response Element (VDRE) 1,25(OH)2D3 negatively regulates expression of the avian PTH (aPTH) gene transcript, and Liu et al. identified a vitamin D response element (VDRE) near the promoter of the aPTH gene [14]. Koszewski et al. converted the negative activity imparted by the aPTH VDRE to a positive transcriptional response through selective mutations introduced into the element [27]. The tested sequences were derived from individual and combined mutations to 2 bp in the 3′-half of the direct repeat element, GGGTCAggaGGGTGT. Cold competition experiments using mutant and wild-type oligonucleotides in the mobility shift assay revealed minor differences in the ability of any of these sequences to compete for binding to a VDR–RXR heterodimer complex composed of recombinant proteins. Ethylation interference footprint analysis for each of the mutants produced unique patterns over the 3′-half-sites that were distinct from the weak, wild-type footprint. Transcriptional outcomes evaluated from a chloramphenicol acetyltransferase reporter construct utilizing the aPTH

540

JUSTIN SILVER AND TALLY NAVEH-MANY

A

B

p

2

1

3

C

D

FIGURE 1

The 1,25(OH)2vitamin D receptor (VDR) is localized to the parathyroid in a similar concentration to that found in the duodenum indicating that the parathyroid is a physiological target organ for 1,25(OH)2D. In situ hybridization with the VDR probe in rat parathyroid-thyroid and duodenum sections. (A1) Parathyroid-thyroid tissue from a control rat. (A2) Parathyroid-thyroid from a 1,25(OH)2D3-treated rat (100 pmol at 24 h). (A3) Duodenum from the 1,25(OH)2D3-treated rat. The white arrows point at the parathyroid glands. (B) A higher power view of A2 that shows the parathyroid gland (p) and thyroid follicles (t). Top figures were photographed under bright-field illumination, whereas bottom figures show dark-field illumination of the same sections. (Reproduced with permission of the American Society of Clinical Investigation [21], by copyright permission of The American Society for Clinical Investigation.) (See color plate.)

200

PTH mRNA and VDR mRNA (% of basal)

200

150

150 VDR

100

100

50

50 PTH

0

0 0

10

20

30 Time (h)

40

50

FIGURE 2 Time course for the effect of 1,25(OH)2D3 on mRNA levels for PTH and the 1,25(OH)2D3 receptor (VDR) in rat thyroparathyroid glands. Rats were injected with either a single dose of 100 pmol 1,25(OH)2D3, or 50 pmol 1,25(OH)2D3 at 0 and 24 hr. The arrow represents the second injection of 1,25(OH)2D3. The data represent the mean ± SE for four rats. (From Naveh-Many et al. [21]. By copyright permission of the American Society for Clinical Investigation.)

promoter found that the individual T→A mutant produced an attenuated negative transcriptional response while the G→C mutant resulted in a reproducibly weak positive transcriptional outcome. The double mutant, however, yielded a fourfold increase in transcription, similar to the sevenfold increase observed from an analogous construct using the human osteocalcin VDRE. UV light cross-linking to gapped oligonucleotides assessed the polarity of heterodimer binding to the wild-type and double mutant sequences and was consistent with the VDR preferentially binding to the 5′-half of both elements. Finally, DNA affinity chromatography was used to immobilize VDR–RXR heterodimer complexes bound to the wild-type and double mutant sequences as bait to identify proteins that may preferentially interact with these DNA-bound heterodimers. This analysis revealed the presence of a p160 protein that specifically interacted with the heterodimer bound to the wild-type VDRE, but was absent from complexes bound to response elements associated with positive transcriptional activity. Thus, the sequence of the individual VDRE appears to play an active role in dictating transcriptional responses that may be mediated by altering

CHAPTER 30 Vitamin D and the Parathyroids

the ability of a VDR-RXR heterodimer to interact with accessory factor proteins. Darwish et al. identified a transcription factor that binds to the promoter region of the human PTH gene adjacent to the negative vitamin D responsive element (VDRE) [28]. Deletion and mutation analysis revealed that the binding site for this factor overlapped with the proximal repeat element of the VDRE. It includes additional nucleotides at the 3′ end of the VDRE. This site has the sequence TTTGAACCTATAGTTGAGAT and a core sequence TGAACCTAT needed for binding of the factor. Experiments with specific anti-VDR antibodies demonstrated that VDR is not found in the factor/DNA complex. However, removing the VDR from the nuclear extract by immunoprecipitation eliminated the binding complex, and the addition of recombinant VDR to the depleted extract did not restore the factor’s ability to bind to the DNA, suggesting that the factor and VDR are closely associated. Transfection experiments with various reporter constructs indicated that the factor is required for the high transcriptional activity of the human PTH gene. This high activity is significantly suppressed by 1,25(OH)2D3. This factor is expressed in several cell types including rat osteoblasts and pituitary. Kimmel-Jehan et al. have shown that VDR-RXR heterodimers induce a DNA bend upon binding to various VDREs by circular permutation and phasing analysis [29]. The VDREs used included the hPTH gene. As shown by circular permutation analysis, VDRRXR induced a distortion in DNA fragments containing various VDREs. The distortions took place with or without a 1,25-(OH)2D3 ligand. The centers of the apparent bend were found in the vicinity of the midpoint of the VDRE. Phasing analysis revealed that VDR-RXR heterodimers induced a directed bend of 26°, not influenced by the presence of hormone. Therefore, similar to other members of the steroid and thyroid nuclear receptor superfamily, VDR-RXR heterodimers induce DNA bending.

E. Regulation of the Vitamin D Receptor A further level at which 1,25(OH)2D3 might regulate the PTH gene would be at the level of the VDR. 1,25(OH)2D3 acts on its target tissues by binding to the VDR, which regulates the transcription of genes with the appropriate recognition sequences. The concentration of the VDR in the 1,25(OH)2D3 target sites could allow a modulation of the 1,25(OH)2D3 effect, with an increase in receptor concentration leading to an amplification of its effect and a decrease in receptor concentration dampening the 1,25(OH)2D3 effect.

541 Naveh-Many et al. [21] injected 1,25(OH)2D3 into rats and measured the levels of the VDR mRNA and PTH mRNA in the parathyro-thyroid tissue. They showed that 1,25(OH)2D3 in physiologically relevant doses led to an increase in VDR mRNA levels in the parathyroid glands in contrast to the decrease in PTH mRNA levels (Fig. 2). This increase in VDR mRNA occurred after a time lag of 6 hr, and a dose response showed a peak at 25 pmol of 1,25(OH)2D3. Weanling rats fed a diet deficient in calcium were markedly hypocalcemic at 3 weeks and had very high serum 1,25(OH)2D3 levels. Despite the chronically high serum 1,25(OH)2D3 levels there was no increase in VDR mRNA levels, and furthermore PTH mRNA levels did not fall but were markedly increased. The low serum calcium may have prevented the increase in parathyroid VDR levels and this may partially explain the suppression of PTH mRNA. Whatever the mechanism, the lack of suppression of PTH synthesis in the setting of hypocalcemia and increased serum 1,25(OH)2D3 is crucial physiologically, because it allows an increase in both PTH and 1,25(OH)2D3 at a time of chronic hypocalcemic stress. Russell et al. [30] studied the parathyroids of chicks with vitamin D deficiency and confirmed that 1,25(OH)2D3 regulates PTH and VDR gene expression in the avian parathyroid gland. The chicks in this study were fed a vitamin D–deficient diet from birth for 21 days and had established secondary hyperparathyroidism. These hypocalcemic chicks were then fed a diet with different calcium contents (0.5, 1.0, and 1.6%) for 6 days. The serum calciums were all still low (5, 6, and 7 mg/dl) with the expected inverse relationship between PTH mRNA and serum calcium. There was also a direct relationship between serum calcium and VDR mRNA levels. This result suggests either that VDR mRNA was not up-regulated in the setting of secondary hyperparathyroidism or that calcium directly regulates the VDR gene. Garfia et al. [31] injected a small dose of 1,25(OH)2D3 to hypercalcemic rats to match the serum 1,25(OH)2D3 levels of hypocalcemic rats. Parathyroid gland VDR mRNA and VDR protein were increased in hypercalcemic rats as compared with hypocalcemic rats. Increasing doses of 1,25(OH)2D3 up-regulated VDR mRNA and VDR only in hypercalcemic rats. Additional experiments showed that the decrease in VDR in hypocalcemic rats prevented the inhibitory effect of 1,25(OH)2D3 on PTH mRNA. They concluded that extracellular Ca regulates VDR expression by parathyroid cells independently of 1,25(OH)2D3 and that by this mechanism hypocalcemia may help prevent the feedback of 1,25(OH)2D3 on the parathyroids. Brown et al. [32] studied vitamin D–deficient rats and confirmed that 1,25(OH)2D3

542 up-regulated the parathyroid VDR mRNA and that in secondary hyperparathyroidism with hypocalcemia the PTH mRNA was up-regulated without change in the VDR mRNA [21]. All these studies show that 1,25(OH)2D3 increases the expression of its receptor’s gene in the parathyroid gland, which would result in increased VDR protein synthesis and increased binding of 1,25(OH)2D3 (Fig. 2). This ligand-dependent receptor up-regulation would lead to an amplified effect of 1,25(OH)2D3 on the PTH gene, and might help explain the dramatic effect of 1,25(OH)2D3 on the PTH gene. Koszewski et al. used interference footprinting protocols to study the interactions of the VDR with either a positive or a negative VDRE [33]. A sequence from the human osteocalcin (hOC) gene was chosen for the prototypical positive VDRE, while an analogous sequence based on the avian parathyroid hormone gene (aPTH) was used as the negative VDRE. Both types of response elements were examined for phosphate backbone contacts, as well as base-specific interactions with guanine and thymine residues. Sources of VDR included partially purified canine intestinal preparations, as well as extracts of recombinant human VDR and RXRα prepared from baculovirus-infected Sf9 insect cells. Cold competition experiments using variable amounts of these oligonucleotides in the mobility shift assay revealed that the hOC element was a fivefold better competitor for heterodimer complex binding than the negative VDRE. Interference footprints revealed extensive strong contacts to the phosphate backbone and individual guanine and thymine nucleotides of the hOC element. The composite hOC footprint was asymmetric for the number and strength of interactions observed over each of the respective direct repeat halfsites. In contrast, the aPTH VDRE footprints revealed fewer points of DNA contact that were limited to the hexanucleotide repeat regions and were strikingly weaker in nature. The alignment of DNA contact points for both elements produced a 5′ stagger that was indicative of successive major groove interactions and consistent with dimer binding. DNA helical representations indicate that the heterodimer contacts to these positive and negative VDREs are substantially different and provide insight into functional aspects of each complex.

F. VDR Gene Deletions To determine what phenotypic abnormalities observed in VDR-ablated mice are secondary to impaired intestinal calcium absorption rather than receptor deficiency, Li et al. normalized mineral ion levels by dietary means [34]. VDR-ablated mice and

JUSTIN SILVER AND TALLY NAVEH-MANY

control littermates were fed a diet rich in calcium lactate that has been shown to prevent secondary hyperparathyroidism in vitamin D–deficient rats (see Chapter 20). This diet normalized growth and random serum ionized calcium levels in the VDR-ablated mice. The correction of ionized calcium levels prevented the development of parathyroid hyperplasia and the increases in PTH mRNA synthesis and in serum PTH levels. VDR-ablated animals fed this diet did not develop rickets or osteomalacia. However, alopecia was still observed in the VDR-ablated mice with normal mineral ions, suggesting that the VDR is required for normal hair growth. This study demonstrates that normalization of mineral ion homeostasis can prevent the development of hyperparathyroidism, osteomalacia, and rickets in the absence of the genomic actions of 1,25(OH)2D3. Van Cromphaut and colleagues [35,36] have also generated mice with deletions of the VDR and showed that the secondary hyperparathyroidism of these VDR-KO mice could be corrected by a highcalcium diet. Hoenderop et al. [37] have presented preliminary findings where they have studied 25(OH)D3-1αhydroxylase knockout mice rescued by a calciumenriched diet. They studied the expression profile of renal genes using microarray chips. 1α-Hydroxylase knockout mice fed a normal Ca2+ diet developed severe hypocalcemia and rickets and died within 3 months. Those mice fed an enriched Ca2+ diet were normocalcemic and not different from wild-type mice. As expected, many genes (±1000) were regulated by the 1α-hydroxylase gene deletion, whereas Ca2+ supplementation revealed ±2000 controlled genes. These latter genes included genes encoding signaling molecules such as the PDZ-domain containing protein channel interacting protein, FK binding protein type 4, and kinases as well as Ca2+ transporting proteins such as the Na+-Ca2+ exchanger, calbindin-D28k, and the Ca2+ sensor calmodulin. The functional significance of these changes in gene expression now need to be determined. The use of calcitriol is limited by its hypercalcemic effect, and therefore a number of calcitriol analogs have been synthesized which are biologically active but are less hypercalcemic than calcitriol (see Chapters 80–88). The ability of calcitriol to decrease PTH gene transcription is used therapeutically in the management of patients with chronic renal failure. Such patients treated with calcitriol in order to prevent the secondary hyperparathyroidism of chronic renal failure (see Chapter 76). The poor response in some patients who do not respond to calcitriol may well result from poor control of serum phosphate, decreased VDR concentration [38], an inhibitory effect of a uremic toxin(s)

CHAPTER 30 Vitamin D and the Parathyroids

on VDR-VDRE binding [39], or tertiary hyperparathyroidism with monoclonal parathyroid tumors [40]. Patel et al. [39] have studied the mechanism of the resistance to the action of calcitriol in chronic renal failure. They used the electrophoretic mobility shift assay to compare the ability of VDRs from normal and renal failure rats to bind to the osteocalcin gene VDRE. VDRs from renal failure rats had only half the DNA binding capacity of VDRs from control rats, despite identical calcitriol binding. Furthermore, incubation of normal VDRs with a uremic plasma ultrafiltrate resulted in a loss of >50% of the binding sites for the osteocalcin VDRE. The inhibitory effect of the uremic ultrafiltrate was due to a specific interaction with the VDR, not RXR. They concluded that an inhibitory effect of a uremic toxin(s) on VDR-VDRE binding could underlie the calcitriol resistance of renal failure.

VII. REGULATION OF THE CALCIUM RECEPTOR BY CALCITRIOL Vitamin D may also amplify its effect on the parathyroid by increasing the activity of the calcium receptor (CaR). Canaff and Hendy showed that in fact there are VDREs in the human CaR’s promoter [41] CaR is expressed in parathyroid chief cells, thyroid C-cells, and cells of the kidney tubule and is essential for maintenance of calcium homeostasis. They showed that parathyroid, thyroid, and kidney CaR mRNA levels increased two-fold at 15 hr after intraperitoneal injection of 1,25(OH)2D3 in rats. Human thyroid C-cell (TT) and kidney proximal tubule cell (HKC) CaR gene transcription increased approximately two-fold at 8 and 12 hr after 1,25(OH)2D3 treatment. The human CaR gene has two promoters yielding alternative transcripts containing either exon 1A or exon 1B 5′-untranslated region sequences that splice to exon 2 some 242 bp before the ATG translation start site. Transcriptional start sites were identified in parathyroid gland and TT cells; that for promoter P1 lies 27 bp downstream of a TATA box, whereas that for promoter P2, which lacks a TATA box, lies in a GC-rich region. In HKC cells, transcriptional activity of a P1 reporter gene construct was 11-fold and of P2 was 33-fold above basal levels. 1,25(OH)2D3 (10−8 M) stimulated P1 activity twofold and P2 activity 2.5-fold. Vitamin D response elements (VDREs), in which half-sites (6 bp) are separated by three nucleotides, were identified in both promoters and shown to confer 1,25(OH)2D3 responsiveness on a heterologous promoter. This responsiveness was lost when the VDREs were mutated. In electrophoretic mobility shift assays with either in vitro transcribed/ translated VDR and RXRα, or HKC nuclear extract,

543 specific protein–DNA complexes were formed in the presence of 1,25(OH)2D3 on oligonucleotides representing the P1 and P2 VDREs. In summary, functional VDREs have been identified in the CaR gene and provide the mechanism whereby 1,25(OH)2D3 upregulates parathyroid, thyroid C-cell, and kidney CaR expression.

VIII. CALRETICULIN AND THE ACTION OF 1,25(OH)2D3 ON THE PARATHYROID HORMONE GENE Another possible level at which 1,25(OH)2D3 might regulate PTH gene expression involves calreticulin. Calreticulin is a calcium binding protein that is present in the endoplasmic reticulum of the cell and also may have a nuclear function. It regulates gene transcription via its ability to bind a protein motif in the DNAbinding domain of nuclear hormone receptors of sterol hormones. It has been shown to prevent 1,25(OH)2D3 binding and action on the osteocalcin gene in vitro [41a]. Sela-Brown et al. showed that calreticulin might inhibit 1,25(OH)2D3 action on the PTH gene as well [42]. Both rat and chicken VDRE sequences of the PTH gene were incubated with recombinant VDR and RXR in a gel retardation assay and showed a clear retarded band. Purified calreticulin inhibited binding of the VDR-RXR complex to the VDREs. This inhibition was due to direct protein–protein interactions between the VDR and calreticulin. To further analyze this effect, opossum kidney (OK) cells were transiently cotransfected with calreticulin expression vectors (sense and antisense) and either rat or chicken PTH gene promoter-CAT constructs [42]. The cells were then assayed for 1,25(OH)2D3-induced CAT gene expression. 1,25(OH)2D3 decreased PTH promoterCAT transcription. Cotransfection with sense calreticulin, which increases calreticulin protein levels, completely inhibited the effect of 1,25(OH)2D3 on the PTH promoters of both rat and chicken. Cotransfection with the antisense calreticulin construct did not interfere with vitamin D’s effect on PTH gene transcription. Sense calreticulin expression had no effect on basal CAT mRNA levels. In order to determine a physiological role for calreticulin in the regulation of the PTH gene, the levels of calreticulin protein were determined in the nuclear fraction of rat parathyroids. The rats were fed either a control diet or a low-calcium diet, which leads to increased PTH mRNA levels despite high serum 1,25(OH)2D3 levels that would be expected to inhibit PTH gene transcription [42]. It was postulated that high calreticulin levels in the nuclear fraction

544 would prevent the effect of 1,25(OH)2D3 on the PTH gene. In fact, the hypocalcemic rats had increased levels of calreticulin protein, as measured by Western blots, in their parathyroid nuclear fraction. This may help explain why hypocalcemia leads to increased PTH gene expression despite high serum 1,25(OH)2D3 levels, and may also be relevant to the refractoriness of the secondary hyperparathyroidism of many chronic renal failure patients to 1,25(OH)2D3 treatment. These studies, therefore, indicate a role for calreticulin in regulating the effect of vitamin D on the PTH gene, and suggest its physiological relevance in chronic renal failure [42].

IX. PARATHYROID HORMONE DEGRADATION A further level of control of serum PTH is at the level of PTH degradation. Preproparathyroid hormone (prepro-PTH) is abundantly synthesized by parathyroid chief cells; yet under normal growth conditions, little or no prepro-PTH can be detected in these cells. The addition of proteasome inhibitors to primary cultures of bovine parathyroid cells caused the accumulation of prepro-PTH and pro-PTH [43]. Proteasome-mediated degradation of PTH precursors therefore may be important in the regulation of the levels of these precursors and hence PTH secretion. However, it is not known whether calcium or vitamin D regulate this process. PTH may be degraded in the parathyroid to carboxy- and amino-terminal fragments in both the parathyroid and in other organs such as the liver and kidney [44]. In the situation of hypercalcemia as much as 90% of the synthesized PTH may be degraded. Enzymes that are involved include furin [45] and protein convertase 1, 2, and 7, which are all expressed in the parathyroid [46]. However, both calcium and 1,25(OH)2D3 did not regulate furin or protein convertase 7 mRNA levels [46].

X. SECONDARY HYPERPARATHYROIDISM AND PARATHYROID CELL PROLIFERATION Chronic changes in the physiological milieu often lead to changes in both parathyroid cell proliferation and PTH gene regulation. In such complicated settings, the regulation of PTH gene expression may well be controlled by mechanisms that differ from those in nonproliferating cells. Further, the effects of change in cell number and activity of individual cells can be

JUSTIN SILVER AND TALLY NAVEH-MANY

complicated and difficult to dissect. Nevertheless, such chronic changes represent commonly observed clinical circumstances of secondary hyperparathyroidism that require examination. The expression and regulation of the PTH gene has been studied in two models of secondary hyperparathyroidism: (1) rats with experimental uremia due to 5/6 nephrectomy and (2) rats with nutritional secondary hyperparathyroidism due to diets deficient in vitamin D and/or calcium. Rats with 5/6 nephrectomy had higher serum creatinine levels and also appreciably higher levels of parathyroid gland PTH mRNA [47]. Their PTH mRNA levels decreased after single injections of 1,25(OH)2D3, a response similar to that of normal rats [47]. Interestingly, the secondary hyperparathyroidism is characterized by an increase in parathyroid gland PTH mRNA but not in VDR mRNA. This suggests that in 5/6 nephrectomy rats there is relatively less VDR mRNA per parathyroid cell, or a relative down-regulation of the VDR, as has been reported in VDR binding studies. The second model of experimental secondary hyperparathyroidism studied was that due to dietary deficiency of vitamin D (−D) and/or calcium (−Ca), as compared to normal vitamin D (ND) and normal calcium (NCa) [48]. These dietary regimes were selected to mimic the secondary hyperparathyroidism in which the stimuli for the production of hyperparathyroidism are the low serum levels of 1,25(OH)2D3 and ionized calcium. Weanling rats were maintained on the diets for 3 weeks and then studied. Rats on diets deficient in both vitamin D and calcium (−D, −Ca) exhibited a 10-fold increase in PTH mRNA as compared to controls (ND, NCa) together with much lower serum calcium levels and also lower serum 1,25(OH)2D3 levels. Calcium deficiency alone (−Ca, ND) led to a fivefold increase in PTH mRNA levels, whereas a diet deficient in vitamin D alone (−D, NCa) led to a twofold increase in PTH mRNA levels. Because renal failure and prolonged changes in blood calcium and 1,25(OH)2D3 can affect both parathyroid cell number and the activity of each parathyroid cell, the change in both these parameters must be assessed in each model in order to understand the various mechanisms of secondary hyperparathyroidism. Parathyroid cell number was determined in thyroparathyroid tissue of normal rats and −D, −Ca rats. To do this, the tissue was enzymatically digested into an isolated cell population, which was then passed through a flow cytometer [fluorescence-activated cell sorter (FACS)] and separated by size into two peaks. The first peak of smaller cells contained parathyroid cells as determined by the presence of PTH mRNA, and the second peak contained thyroid follicular cells and calcitonin-producing cells that hybridized positively

CHAPTER 30 Vitamin D and the Parathyroids

for thyroglobulin mRNA and calcitonin mRNA but not PTH mRNA. There was a 1.6-fold increase in parathyroid cells from the −D, −Ca rats compared to normal rats, and a 10-fold increase in PTH mRNA. Therefore, this model of secondary hyperparathyroidism is characterized by increased gene expression per parathyroid cell, together with a smaller increase in cell number. Further studies by Naveh-Many et al. [49] have clearly demonstrated that hypocalcemia is a stimulus for parathyroid cell proliferation. They studied parathyroid cell proliferation by staining for proliferating cell nuclear antigen (PCNA) and found that a low-calcium diet led to increased levels of PTH mRNA and a 10-fold increase in parathyroid cell proliferation. The secondary hyperparathyroidism of 5/6 nephrectomized rats was characterized by an increase in both PTH mRNA levels and PCNA-positive parathyroid cells. Therefore, both hypocalcemia and uremia induce parathyroid cell proliferation in vivo. The effect of 1,25(OH)2D3 on parathyroid cell proliferation was also studied in this dietary model of secondary hyperparathyroidism. 1,25(OH)2D3 at a dose (25 pmol for 3 days) that lowered PTH mRNA levels had no effect on the number of PCNApositive cells. Higher doses (100 pmol for 7 days) dramatically decreased the number of proliferating cells (unpublished). These findings emphasize the importance of a normal calcium in the prevention of parathyroid cell hyperplasia. The importance of the CaR to parathyroid cell proliferation is also evident in that the calcimimetic NPS R-568 largely prevents the parathyroid cell proliferation in rats with experimental uremia [50,51]. However, the role of the CaR is not that clear in view of the interesting findings of Lewin et al. [52]. Experimental severe secondary hyperparathyroidism is reversed within 1 week after reversal of uremia by an isogenic kidney transplantation in uremic rats. In view of the reports that abnormal PTH secretion in uremia is related to downregulation of CaR [53] and vitamin D receptor (VDR) in the parathyroid glands [54], they studied the expression of CaR and VDR genes after reversal of uremia and hyperparathyroidism in rats given isogenic kidney transplantation. After kidney transplantation into previously uremic rats, the secondary hyperparathyroidism was reversed with the serum PTH levels returning to normal. However, both CaR mRNA and VDR mRNA remained severely reduced (CaR, 39 ± 7%; VDR, 9 ± 3%; P < 0.01) compared with normal rats. In conclusion, circulating plasma PTH levels normalized rapidly after kidney transplantation, despite persisting downregulation of CaR and VDR gene expression. This indicates that up-regulation of CaR mRNA and VDR mRNA is not necessary to induce the rapid normalization of PTH secretion from hyperplastic parathyroid glands.

545 In a separate approach to this question, Imanishi et al. created transgenic mice with the cyclin D1 gene specifically targeted to the parathyroid [55]. In the parathyroids of these rats with hyperparathyroidism there was a down-regulation of the CaR. These results indicate that the changes in the CaR may be secondary to the proliferative state and not causative. Stimulation of the tyrosine kinase pathway may very well be involved in the proliferative response of the parathyroid to hypocalcemia or chronic renal failure. Endothelin-1 (ET-1), a vasoconstrictive peptide, has been shown to act as a mitogen in a variety of cell types. Rat parathyroid cells are reported to synthesize ET-1 and possess its receptors. To test the hypothesis that ET-1 plays a role in parathyroid cell proliferation, Kanesaka et al. tested rats fed a low calcium diet for 8 weeks (low-Ca rats) [56]. The number of the proliferating parathyroid cells, measured by proliferating cell nuclear antigen (PCNA) immunostaining, was significantly increased, with striking immunoreactivity of ET-1 in the low-Ca rats. An endothelin receptor antagonist, bosentan (100 mg/kg.day), prevented any increase in the proliferation of parathyroid cells in the low-Ca rats. These results indicate that ET-1 is involved in parathyroid cell proliferation in vivo and suggest that blocking of ET receptors may become one of the important therapeutic strategies for preventing secondary hyperparathyroidism [56]. Preliminary studies have been reported by Cozzolino et al. [57] with similar results using another specific tyrosine kinase inhibitor that effectively decreased the parathyroid cell proliferation of 5/6 nephrectomized rats. Intriguing insights into the physiology of the parathyroid have been produced from intercrossing mice with homozygous null mutations in the CaR and the PTH gene [58]. Mice homozygous for null mutations in the CaR gene (CaR(−/−)) die shortly after birth because of the effects of severe hyperparathyroidism and hypercalcemia. A wide variety of functions have been attributed to CaR. However, the lethal CaRdeficient phenotype has made it difficult to dissect the direct effect of CaR deficiency from the secondary effects of hyperparathyroidism and hypercalcemia. Kos et al. [58] therefore generated parathyroid hormonedeficient (PTH-deficient) CaR(−/−) mice (Pth(−/−) CaR(−/−)) by intercrossing mice heterozygous for the null CaR allele with mice heterozygous for a null Pth allele. They showed that genetic ablation of PTH is sufficient to rescue the lethal CaR(−/−) phenotype. Pth(−/−)CaR(−/−) mice survive to adulthood with no obvious difference in size or appearance relative to control Pth(−/−) littermates. Histologic examination of most organs did not reveal abnormalities. These Pth(−/−) CaR(−/−) mice exhibited a much wider range of values

546

JUSTIN SILVER AND TALLY NAVEH-MANY

for serum calcium and renal excretion of calcium than we observe in control littermates, despite the absence of any circulating PTH. Therefore, the CaR is necessary for the fine regulation of serum calcium levels and renal calcium excretion independent of its effect on PTH secretion. These results were supported by complementary studies from the laboratory of Quarles. They also used a genetic approach to ablate parathyroid glands and remove the confounding effects of elevated parathyroid hormone (PTH) in CaR-deficient mice [59]. CaR deficiency was transferred onto the glial cells missing 2-deficient (Gcm2-deficient) background by intercrossing CaR- and Gcm2-deficient mice. Gcm2−/− mice fail to develop parathyroid glands. Superimposed Gcm2

low Ca2+ 2+ + Ca

deficiency rescued the perinatal lethality in CaR-deficient mice in association with ablation of the parathyroid glands and correction of the severe hyperparathyroidism. In addition, the double homozygous CaR- and Gcm2deficient mice demonstrated healing of the abnormal mineralization of cartilage and bone associated with CaR deficiency, indicating that rickets and osteomalacia in CaR-deficient mice are not due to an independent function of CaR in bone and cartilage but to the effect of severe hyperparathyroidism in the neonate. Analysis of the skeleton of 6-week-old homozygous CaR- and Gcm2-deficient mice also failed to identify any essential, nonredundant role for CaR in regulating chondrogenesis or osteogenesis. In contrast, concomitant Gcm2 and CaR deficiency failed to rescue the

CaSR PTH

RET

cyclin D1 low Ca2+ high P menin

cell cycle PTH mRNA Rb 5′-UTR

coding

TGFα PTH

endothelin

3′-UTR

An

1,25(OH)2D

+ P

− 1,25(OH)2D

FIGURE 3 Regulation of parathyroid proliferation, gene expression, and secretion. Cyclin D1 driven by the PTH promoter and inactivating mutations of the menin gene are known to cause PT adenomas; germ-line mutations of the latter cause MEN 1. The very rare PT carcinomas show lack of expression of the retinoblastoma protein (pRb). Activating mutations of the RET proto-oncogene result in MEN 2a. A low serum calcium leads to a decreased activation of the CaSR and results in increased PTH secretion (dots), PTH mRNA stability, and PT cell proliferation. A high serum phosphate leads to similar changes in all these parameters. Endothelin and TGFα are increased in the PTs of proliferating PT glands. 1,25(OH)2D3 decreases PTH gene transcription markedly and decreases PT cell proliferation. PTH mRNA stability is regulated by PT cytosolic proteins (trans factors, shown here as small squares) binding to a short defined cis sequence in the PTH mRNA 3′-UTR and preventing degradation by ribonucleases depicted as “Pacmen.” One of these protective proteins is AUF1. In hypocalcemia there is more binding of the trans factors to the cis sequence leading to a more stable transcript. A low serum phosphate leads to much less binding and a rapidly degraded PTH transcript. (Reproduced from [44] with copyright permission of the American Society of Clinical Investigation.)

547

CHAPTER 30 Vitamin D and the Parathyroids

hypocalciuria in CaR-deficient mice, consistent with direct regulation of urinary calcium excretion by CaR in the kidney. Double Gcm2- and CaR-deficient mice therefore provide another model for evaluating the extraparathyroid functions of the CaR [59] and thereby determining direct roles for Ca2+ and the CaR on diverse target organs. In patients with both primary and nodular secondary hyperparathyroidism due to chronic renal failure there is a decrease in VDR mRNA and protein levels [38,60,61]. In hyperparathyroidism there is a decrease in the cyclin kinase inhibitors p21 and p27 with an increase in TGFα in the parathyroids [57,61,62]. Treatment with vitamin D metabolites increase p21 levels and prevent the decrease in TGFα levels and prevent the parathyroid cell proliferation. These results suggest that the balance between the cyclin kinase inhibitors and TGFα plays a role in the pathogenesis of the cell proliferation.

XI. CONCLUSIONS PTH gene expression is powerfully regulated by 1,25(OH)2D3. This is a transcriptional effect and results in a marked decrease in PTH secretion and serum PTH. The effect of 1,25(OH)2D3 on the parathyroid is used in the treatment of many patients in chronic renal failure to prevent their secondary hyperparathyroidism. In addition, the expression of the PTH gene is also regulated by calcium and phosphate. These effects are posttranscriptional. A model of the factors regulating the parathyroid is shown in Fig. 3.

8.

9.

10.

11. 12. 13.

14.

15.

16.

17.

References 18. 1. Potts JTJ, Kronenberg HM, Habener JF, Rich A 1980 Biosynthesis of parathyroid hormone. Ann NY Acad Sci 343: 38–55. 2. Habener JF 1981 Regulation of parathyroid hormone secretion and biosynthesis. Ann Rev Physiol 43:211–223. 3. Cohn DV, Kumarasamy R, Ramp WK 1986 Intracellular processing and secretion of parathyroid gland proteins. Vitam Horm 43:283–316. 4. Drees BM, Hamilton JW 1992 Pancreastatin and bovine parathyroid cell secretion. Bone Miner 17:335–346. 5. Brown EM 1983 Four-parameter model of the sigmoidal relationship between parathyroid hormone release and extracellular calcium concentration in normal and abnormal parathyroid tissue. J Clin Endocrinol Metab 56:572–581. 6. Brown EM, Gamba G, Riccardi D, Lombardi M, Butters R, Kifor O, Sun A, Hediger MA, Lytton J, Hebert J 1993 Cloning and characterization of an extracellular Ca2+ -sensing receptor from bovine parathyroid. Nature 366:575–580. 7. Pollack MR, Brown EM, Chou Y-HW, Hebert SC, Marx SJ, Steinman B, Levi T, Seidman CE, Seidman JG 1993 Mutations

19.

20.

21.

22. 23.

in the human Ca2+-sensing receptor gene cause familial hypocalciuric hypercalcemia and neonatal severe hyperparathyroidism. Cell 75:1297–1303. Ho C, Conner DA, Pollak MR, Ladd DJ, Kifor O, Warren HB, Brown EM, Seidman JG, Seidman CE 1995 A mouse model of human familial hypocalciuric hypercalcemia and neonatal severe hyperparathyroidism. Nat Genet 11:389–394. Silver J, Naveh-Many T, Kronenberg HM 2002 Parathyroid hormone: molecular biology. In: Bilezikian JB, Raisz LG, Rodan GA (eds) Principles of Bone Biology. Academic Press, San Diego, pp. 407–422. Kronenberg HM, Igarashi T, Freeman MW, Okazaki T, Brand SJ, Wirenc KM, Potts JT Jr 1986 Structure and expression of the human parathyroid hormone gene. Recent Prog Horm Res 42:641–663. Igarashi T, Okazaki T, Potter H, Gaz R, Kronenberg HM 1986 Cell-specific expression of the human parathyroid hormone gene in rat pituitary cells. Mol Cell Biol 6: 1830–1833. Rupp E, Mayer H, Wingender E 1990 The promoter of the human parathyroid hormone gene contains a functional cyclic AMP-response element. Nucleic Acids Res 18:5677–5683. Demay MB, Kiernan MS, DeLuca HF, Kronenberg HM 1992 Sequences in the human parathyroid hormone gene that bind the l,25-dihydroxyvitamin D3 receptor and mediate transcriptional repression in response to l,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 89:8097–8101. Liu SM, Koszewski N, Lupez M, Malluche HH, Olivera A, Russell J 1996 Characterization of a response element in the 5′-flanking region of the avian (chicken) parathyroid hormone gene that mediates negative regulation of gene transcription by 1,25-dihydroxyvitamin D3 and binds the vitamin D3 receptor. Mol Endocrinol 10:206–215. Silver J, Russell J, Sherwood LM 1985 Regulation by vitamin D metabolites of messenger ribonucleic acid for preproparathyroid hormone in isolated bovine parathyroid cells. Proc Natl Acad Sci USA 82:4270–4273. Russell J, Silver J, Sherwood LM 1984 The effects of calcium and vitamin D metabolites on cytoplasmic mRNA coding for pre-proparathyroid hormone in isolated parathyroid cells. Trans Assoc Am Physicians 97:296–303. Cantley LK, Russell J, Lettieri D, Sherwood LM 1985 1,25Dihydroxyvitamin D3 suppresses parathyroid hormone secretion from bovine parathyroid cells in tissue culture. Endocrinology 117;2114–2119. Karmali R, Farrow S, Hewison M, Barker S, O’Riordan JL 1989 Effects of 1,25-dihydroxyvitamin D3 and cortisol on bovine and human parathyroid cells. J Endocrinol 123:137–142. Chan YL, McKay C, Dye E, Slatopolsky E 1986 The effect of 1,25 dihydroxycholecalciferol on parathyroid hormone secretion by monolayer cultures of bovine parathyroid cells. Calcif Tissue Int 38:27–32. Silver J, Naveh-Many T, Mayer H, Schmelzer HJ, Popovtzer MM 1986 Regulation by vitamin D metabolites of parathyroid hormone gene transcription in vivo in the rat. J Clin Invest 78:1296–1301. Naveh-Many T, Marx R, Keshet E, Pike JW, Silver J 1990 Regulation of 1,25-dihydroxyvitamin D3 receptor gene expression by 1,25-dihydroxyvitamin D3 in the parathyroid in vivo. J Clin Invest 86:1968–1975. Naveh-Many T, Silver J 1988 Regulation of calcitonin gene transcription by vitamin D metabolites in vivo in the rat. J Clin Invest 81:270–273. Russell J, Lettieri D, Sherwood LM 1986 Suppression by 1,25(OH)2D3 of transcription of the pre-proparathyroid hormone gene. Endocrinology 119:2864–2866.

548 24. Fernandez-Santos JM, Utrilla JC, Conde E, Hevia A, Loda M, Martin-Lacave I 2001 Decrease in calcitonin and parathyroid hormone mRNA levels and hormone secretion under long-term hypervitaminosis D3 in rats. Histol Histopathol 16:407–414. 25. Okazaki T, Igarashi T, Kronenberg HM 1988 5’-Flanking region of the parathyroid hormone gene mediates negative regulation by 1,25-(OH)2 vitamin D3. J Biol Chem 263:2203–2208. 26. MacDonald PN, Ritter C, Brown AJ, Slatopolsky E 1994 Retinoic acid suppresses parathyroid hormone (PTH) secretion and PreproPTH mRNA levels in bovine parathyroid cell culture. J Clin Invest 93:725–730. 27. Koszewski NJ, Ashok S, Russell J 1999 Turning a negative into a positive: vitamin D receptor interactions with the avian parathyroid hormone response element. Mol Endocrinol 13:455–465. 28. Darwish HM, DeLuca HF 1999 Identification of a transcription factor that binds to the promoter region of the human parathyroid hormone gene. Arch Biochem Biophys 365: 123–130. 29. Kimmel-Jehan C, Darwish HM, Strugnell SA, Jehan F, Wiefling B, DeLuca HF 1999 DNA bending is induced by binding of vitamin D receptor-retinoid X receptor heterodimers to vitamin D response elements. J Cell Biochem 74:220–228. 30. Russell J, Bar A, Sherwood LM, Hurwitz S 1993 Interaction between calcium and 1,25-dihydroxyvitamin D3 in the regulation of preproparathyroid hormone and vitamin D receptor messenger ribonucleic acid in avian parathyroids. Endocrinology 132:2639–2644. 31. Garfia B, Canadillas S, Canalejo A, Luque F, Siendones E, Quesada M, Almaden Y, Aguilera-Tejero E, Rodriguez M 2002 Regulation of parathyroid vitamin D receptor expression by extracellular calcium. J Am Soc Nephrol 13:2945–2952. 32. Brown AJ, Zhong M, Finch J, Ritter C, Slatopolsky E 1995 The roles of calcium and 1,25-dihydroxyvitamin D3 in the regulation of vitamin D receptor expression by rat parathyroid glands. Endocrinology 136:1419–1425. 33. Koszewski NJ, Malluche HH, Russell J 2000 Vitamin D receptor interactions with positive and negative DNA response elements: an interference footprint comparison. J Steroid Biochem Mol Biol 72:125–132. 34. Li YC, Amling M, Pirro AE, Priemel M, Meuse J, Baron R, Delling G, Demay MB 1998 Normalization of mineral ion homeostasis by dietary means prevents hyperparathyroidism, rickets, and osteomalacia, but not alopecia in vitamin D receptorablated mice. Endocrinology 139:4391–4396. 35. Van Cromphaut SJ, Dewerchin M, Hoenderop JG, Stockmans I, Van Herck E, Kato S, Bindels RJ, Collen D, Carmeliet P, Bouillon R, Carmeliet G 2001 Duodenal calcium absorption in vitamin D receptor-knockout mice: Functional and molecular aspects. Proc Natl Acad Sci USA 98:13324–13329. 36. Bouillon R, Van Cromphaut S, Carmeliet G 2003 Intestinal calcium absorption: Molecular vitamin D mediated mechanisms. J Cell Biochem 88:332–339. 37. Hoenderop JGJ, Chon H, Gkika D, Bluyssen HAR, Holstege FCP, St-Arnaud R, Braam B, Bindels RJM 2003 Rescue of gene expression by dietary Ca2+ of 25-hydroxyvitamin D3-1αhydroxylase knockout mice. Kidney Int (in press). 38. Fukuda N, Tanaka H, Tominaga Y, Fukagawa M, Kurokawa K, Seino Y 1993 Decreased 1,25-dihydroxyvitamin D3 receptor density is associated with a more severe form of parathyroid hyperplasia in chronic uremic patients. J Clin Invest 92: 1436–1443. 39. Patel SR, Ke HQ, Vanholder R, Koenig R, Hsu CH 1995 Inhibition of calcitriol receptor binding to vitamin D response elements by uremic toxin. J Clin Invest 96:50–59.

JUSTIN SILVER AND TALLY NAVEH-MANY

40. Arnold A, Brown MF, Urena P, Gaz RD, Sarfati E, Drueke TB 1995 Monoclonality of parathyroid tumors in chronic renal failure and in primary parathyroid hyperplasia. J Clin Invest 95:2047–2053. 41. Canaff L, Hendy GN 2002 Human calcium-sensing receptor gene. Vitamin D response elements in promoters P1 and P2 confer transcriptional responsiveness to 1,25-dihydroxyvitamin D. J Biol Chem 277:30337–30350. 41a. Wheeler DG, Horsford J, Michalak M, White JH, Hendy GN 1995 Calreticulin inhibits vitamin D3 signal transduction. Nucleic Acids Res 23:3268–3274. 42. Sela-Brown A, Russell J, Koszewski NJ, Michalak M, Naveh-Many T, Silver J 1998 Calreticulin inhibits vitamin D’s action on the PTH gene in vitro and may prevent vitamin D’s effect in vivo in hypocalcemic rats. Mol Endocrinol 12:1193–1200. 43. Sakwe AM, Engstrom A, Larsson M, Rask L 2002 Biosynthesis and secretion of parathyroid hormone are sensitive to proteasome inhibitors in dispersed bovine parathyroid cells. J Biol Chem 277:17687–17695. 44. Silver J, Kilav R, Naveh-Many T 2002 Mechanisms of secondary hyperparathyroidism. Am J Physiol Renal Physiol 283:F367–F376. 45. Hendy GN, Bennett HP, Gibbs BF, Lazure C, Day R, Seidah NG 1995 Proparathyroid hormone is preferentially cleaved to parathyroid hormone by the prohormone convertase furin. A mass spectrometric study. J Biol Chem 270:9517–9525. 46. Canaff L, Bennett HP, Hou Y, Seidah NG, Hendy GN 1999 Proparathyroid hormone processing by the proprotein convertase-7: Comparison with furin and assessment of modulation of parathyroid convertase messenger ribonucleic acid levels by calcium and 1,25-dihydroxyvitamin D3. Endocrinology 140:3633–3642. 47. Shvil Y, Naveh-Many T, Barach P, Silver J 1990 Regulation of parathyroid cell gene expression in experimental uremia. J Am Soc Nephrol 1:99–104. 48. Naveh-Many T, Silver J 1990 Regulation of parathyroid hormone gene expression by hypocalcemia, hypercalcemia, and vitamin D in the rat. J Clin Invest 86:1313–1319. 49. Naveh-Many T, Rahamimov R, Livni N, Silver J 1995 Parathyroid cell proliferation in normal and chronic renal failure rats: the effects of calcium, phosphate and vitamin D. J Clin Invest 96:1786–1793. 50. Wada M, Furuya Y, Sakiyama J, Kobayashi N, Miyata S, Ishii H, Nagano N 1997 The calcimimetic compound NPS R-568 suppresses parathyroid cell proliferation in rats with renal insufficiency. Control of parathyroid cell growth via a calcium receptor. J Clin Invest 100:2977–2983. 51. Chin J, Miller SC, Wada M, Nagano N, Nemeth EF, Fox J 2000 Activation of the calcium receptor by a calcimimetic compound halts the progression of secondary hyperparathyroidism in uremic rats. J Am Soc Nephrol 11:903–911. 52. Lewin E, Garfia B, Recio FL, Rodriguez M, Olgaard K 2002 Persistent downregulation of calcium-sensing receptor mRNA in rat parathyroids when severe secondary hyperparathyroidism is reversed by an isogenic kidney transplantation. J Am Soc Nephrol 13:2110–2116. 53. Gogusev J, Duchambon P, Hory B, Giovannini M, Goureau Y, Sarfati E, Drueke TB 1997 Depressed expression of calcium receptor in parathyroid gland tissue of patients with hyperparathyroidism. Kidney Int 51:328–336. 54. Drueke TB 1995 The pathogenesis of parathyroid gland hyperplasia in chronic renal failure. Kidney Int 48:259–272. 55. Imanishi Y, Hosokawa Y, Yoshimoto K, Schipani E, Mallya S, Papanikolaou A, Kifor O, Tokura T, Sablosky M,

CHAPTER 30 Vitamin D and the Parathyroids

Ledgard F, Gronowicz G, Wang TC, Schmidt EV, Hall C, Brown EM, Bronson R, Arnold A 2001 Dual abnormalities in cell proliferation and hormone regulation caused by cyclin D1 in a murine model of hyperparathyroidism. J Clin Invest 107: 1093–1102. 56. Kanesaka Y, Tokunaga H, Iwashita K, Fujimura S, Naomi S, Tomita K 2001 Endothelin receptor antagonist prevents parathyroid cell proliferation of low calcium diet-induced hyperparathyroidism in rats. Endocrinology 142:407–413. 57. Cozzolino M, Lu Y, Finch J, Slatopolsky E, Dusso AS 2001 p21WAF1 and TGF-alpha mediate parathyroid growth arrest by vitamin D and high calcium. Kidney Int 60:2109–2117. 58. Kos CH, Karaplis AC, Peng JB, Hediger MA, Goltzman D, Mohammad KS, Guise TA, Pollak MR 2003 The calciumsensing receptor is required for normal calcium homeostasis independent of parathyroid hormone. J Clin Invest 111: 1021–1028.

549 59. Tu Q, Pi M, Karsenty G, Simpson L, Liu S, Quarles LD 2003 Rescue of the skeletal phenotype in CasR-deficient mice by transfer onto the Gcm2 null background. J Clin Invest 111: 1029–1037. 60. Carling T, Rastad J, Szabo E, Westin G, Akerstrom G 2000 Reduced parathyroid vitamin D receptor messenger ribonucleic acid levels in primary and secondary hyperparathyroidism. J Clin Endocrinol Metab 85:2000–2003. 61. Tokumoto M, Tsuruya K, Fukuda K, Kanai H, Kuroki S, Hirakata H 2002 Reduced p21, p27 and vitamin D receptor in the nodular hyperplasia in patients with advanced secondary hyperparathyroidism. Kidney Int 62:1196–1207. 62. Gogusev J, Duchambon P, Stoermann-Chopard C, Giovannini M, Sarfati E, Drueke TB 1996 De novo expression of transforming growth factor-alpha in parathyroid gland tissue of patients with primary or secondary uraemic hyperparathyroidism. Nephrol Dial Transplant 11:2155–2162.

CHAPTER 31

Calcium-Sensing Receptor EDWARD M. BROWN

I. II. III. IV.

Division of Endocrinology, Diabetes and Hypertension, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts

Introduction The CaR: Isolation, Structure, and Intracellular Signaling Role of the CaR in the Parathyroid Role of the CaR in the C-Cell

I. INTRODUCTION In any given person, the extracellular ionized calcium concentration (Ca 2+ o ) is maintained within a very narrow range, which varies by only a few percent over a day, a week, or, for that matter, much of a lifetime [1]. Calcium plays myriad crucial roles both intra- and extracellularly [2]; therefore, it is not surprising that the availability of this ion in the extracellular fluids is so carefully controlled. The extracellular calciumsensing receptor (CaR or CaSR) serves as the body’s thermostat for calcium or “calciostat.” It is a cell surface, G protein–linked receptor [3] that is capable of sensing fluctuations in Ca2+ o from its normal level of only a few percent. It does so primarily through its presence in the chief cells of the parathyroid gland, the thyroidal C-cells, and cells along the renal tubules that are involved in calcium transport [4]. Even minute changes in Ca2+ o modulate the functions of these cells in ways that will restore normocalcemia. The CaR permits extracellular calcium ions, acting through their own cell surface receptor, to serve, in effect, in a hormonelike role as an extracellular first messenger that plays key roles in Ca2+ o homeostasis [5]. This chapter will discuss the structure, function, and biological roles of the CaR, particularly with regard to its central role in mineral ion homeostasis and the interrelationships between the CaR and vitamin D metabolism and action.

II. THE CaR: ISOLATION, STRUCTURE, AND INTRACELLULAR SIGNALING The CaR was the first receptor to be isolated whose principal physiological ligand is an inorganic ion. Studies performed over two decades, investigating the effects of extracellular calcium ions on intracellular second messengers in dispersed parathyroid cells, had indicated that calcium might exert its actions through VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Role of the CaR in the Kidney VI. The CaR in Bone and Intestine VII. Summary References

a cell-surface, G protein–coupled receptor (GPCR) [6–8]. This prediction was confirmed when the receptor was cloned using the technique of expression cloning in Xenopus laevis oocytes [3]. The CaR has three structural domains: a large amino-terminal, extracellular domain (ECD), a seven membrane-spanning “serpentine” motif that is characteristic of the GPCRs, and a long intracellular, carboxyterminal (C)-tail (Fig. 1). The CaR’s ECD is heavily glycosylated, and glycosylation of at least three N-linked glycosylation sites is important for the receptor’s efficient cell surface expression [9]. The active form of the receptor is a dimer [10], which is linked together by two disulfide linkages (at cys129 and cys131) between the ECDs of two CaR monomers [11]. The receptor’s ECD has important determinants for binding calcium ions, although the locations of these sites are currently unknown [12]. Following the binding of calcium to the ECD, the receptor activates the G proteins that enable the CaR to regulate its intracellular signaling pathways (see later discussion). The initiation of signaling requires the binding of G proteins to the receptor’s intracellular loops—particularly the second and third loops—and the proximal part of the C-tail [13,14]. The binding of calcium ions to the receptor’s ECD and potentially to its second extracellular loop [15] presumably produce conformational changes in the transmembrane helices that activate G proteins bound to the receptor’s intracellular domains. The CaR is a member of a subfamily (family C) of the large superfamily of GPCRs—which includes receptors for glutamate, gamma aminobutyric acid (GABA), pheromones, and odorants. All share very large extracellular domains (ECDs) of ~600 amino acids [5]. The recently solved three-dimensional structure of one of the eight GPCRs for glutamate (mGluR1) revealed that its ECD (and probably those of other family C receptors as well) has a bilobed, “venus flytrap” structure [16]. Glutamate binding occurs Copyright © 2005, Elsevier, Inc. All rights reserved.

552

EDWARD M. BROWN

SP NH2

HS

613

670

683

745

770

828

838

635

650

700

725

792

807

862

P P

P

P

Cysteine Conserved

P

N-glycosylation P

PKC site

Acidic HOOC

FIGURE 1 Predicted structure of the human CaR (see text for additional details). SP denotes signal peptide and HS, hydrophobic segment. Modified from Brown EM, Bai M, Pollak M 1997 Familial benign hypocalciuric hypercalcemia and other syndromes of altered responsiveness to extracellular calcium. In: Krane SM, Avioli LV (eds) Metabolic Bone Diseases, 3rd ed. Academic Press, San Diego, pp. 479–499.

in a crevice between the two lobes of each monomer and probably stabilizes the active conformation of the receptor. Elegant studies utilizing molecular modeling and mutagenesis have utilized the mGluR1 model to predict the three-dimensional structure of the CaR and to identify regions of the receptor that are important for dimerization and activation [15,17]. It will be of great interest to eventually identify the precise locations of the CaR’s calcium binding sites using the same approach combined with X-ray crystallography, as the receptor shows marked positive cooperativity, which is an important attribute of its Ca2+ o –sensing capabilities. This cooperativity is the basis for the steepness of the inverse sigmoidal relationship between PTH release and

Ca2+ o that preserves the level of blood calcium within a very narrow range (see Fig. 2 and discussion, later) [18]. There are several lines of evidence supporting the CaR’s role as the “calciostat” for calcium homeostasis, which include experiments-in-nature resulting from mutations in the human CaR gene as well as mouse models in which the receptor has been “knocked out” [4,19]. Persons heterozygous or homozygous for inactivating mutations of the CaR show mild to moderate or moderate to severe hypercalcemia, respectively, owing to “resistance” of the receptor to Ca2+ o [4]. The heterozygous state is the cause of a condition known as familial hypocalciuric hypercalcemia (FHH) or familial benign hypocalciuric hypercalcemia (FBHH) [4,20].

553

CHAPTER 31 Calcium-Sensing Receptor

A

B

Calcium–regulated PTH release from normal parathyoid cells

Calcium–regulated PTH release from normal parathyoid cells Maximum

PTH Release

PTH Release % of maximal

100

50

Slope Midpoint

“Set point”

Minimum 0

2.0

1.0

3.0

[Ca++]

[Ca++], mM

FIGURE 2 (A) Relationship between PTH secretion and extracellular calcium in normal human parathyroid cells. Dispersed parathyroid cells were incubated with the indicated levels of calcium and PTH was determined by radioimmunoassay. Reproduced with permission from Brown EM 1980 Set-point for calcium: its role in normal and abnormal parathyroid secretion. In: Cohn DV, Talmage RV, Matthews JL (eds) Hormonal Control of Calcium Metabolism, Proceedings of the Seventh International Conference on Calcium Regulating Hormones, September 5–9, 1980. International Congress Series No. 511, Excerpta Medica, Amsterdam, pp. 35–43. (B). The four parameters describing the inverse sigmoidal relationship between the extracellular calcium concentration and PTH release in vivo and in vitro: A, maximal secretory rate; B, slope of the curve at the midpoint; C, midpoint or set-point of the curve (the level of calcium producing half of the maximal decrease in secretory rate); and D, minimal secretory rate. Reproduced with permission from Brown EM 1983 Four parameter model of the sigmoidal relationship between parathyroid hormone release and extracellular calcium concentration in normal and abnormal parathyroid tissue. J Clin Endocrinol Metab 56:572–581.

The great majority of persons with FHH have a benign clinical course with none of the usual symptoms and complications of hypercalcemia. In addition to their hypercalcemia, they exhibit inappropriately low or normal levels of urinary calcium excretion given their hypercalcemia—hence the name familial hypocalciuric hypercalcemia [21]. As will be discussed in more detail in the section on the CaR and the kidney, the receptor is expressed in the distal tubule where it inhibits tubular reabsorption of calcium when activated by hypercalcemia. Thus persons with FHH have an impaired capacity to up-regulate renal calcium excretion [22] as well as “resistance” of the parathyroid glands to the usual suppressive effect of hypercalcemia on PTH secretion [23,24]. Homozygous inactivating mutations in the CaR, in contrast, usually produce a condition called neonatal severe hyperparathyroidism (NSHPT), which can be fatal if parathyroidectomy is not carried out urgently in

the immediate postnatal period [4]. The total absence of normal calcium-sensing receptors leads to severe resistance of the parathyroids and kidneys to the usual actions of hypercalcemia on these two tissues. Not surprisingly, mice heterozygous or homozygous for knockout of the receptor display biochemical and phenotypic findings that are similar to those of FHH and NSHPT, respectively [19]. Finally, in contrast to the clinical and biochemical findings in persons with inactivating mutations of the CaR, persons with activating mutations display varying degrees of hypocalcemia, which can be accompanied by hypercalciuria, owing to “oversensitivity” of parathyroid and renal CaRs to Ca2+ o [4,25,26]. Thus these data in humans and mice provide incontrovertible evidence that the calciumsensing receptor plays a central, nonredundant role in maintaining normalcy of Ca2+ o . Although Ca2+ is the CaR’s principal physiological o ligand in vivo, the receptor can be activated by several

554 other ligands, at least two of which—magnesium [3] and certain amino acids [27]—are probably physiologically relevant. Although magnesium is approximately twofold less potent than calcium in modulating the activity of the CaR, and the level of Mg2+ o is lower than that of Ca2+ in the extracellular fluids, persons with o inactivating or activating mutations of the receptor can have increases or decreases in their serum magnesium concentrations, respectively [28]. These alterations in serum magnesium concentration range from those that are within the normal range to those that are frankly hypo- or hypermagnesemic. Thus it is likely that the CaR contributes to “setting” the normal level of extracellular magnesium [28]. Recent studies have shown that certain amino acids, particularly aromatic amino acids, allosterically activate the CaR [27], effectively rendering the receptor more sensitive to any given level of Ca2+ o . Thus, it is possible that the CaR serves a broader role as a “nutrient” receptor, recognizing not only divalent cations but also amino acids. For example, both calcium and aromatic amino acids stimulate gastrin release and acid production in the stomach—effects that could well be mediated by the CaR [29]. There are additional circumstances in which calcium and protein metabolism may be linked in ways that could be mediated by the CaR. A high protein intake engenders hypercalciuria, an action that has traditionally been ascribed to the acid load generated by protein metabolism. Stimulation of renal CaRs by high circulating levels of amino acids, however, could also contribute to the hypercalciuria [30]. Furthermore, a low protein intake in normal individuals, as well as in patients with renal impairment, is accompanied by elevated levels of PTH [31]. This association could be mediated, at least in part, by parathyroid cells sensing a reduction in “nutrient” availability, i.e., the sum of divalent cation and amino acids, and responding with increased PTH secretion. In addition to these endogenous ligands of the CaR, allosteric activators of the receptor, termed “calcimimetics” [32], have been developed, as have CaR antagonists, called “calcilytics” [33]. Phase 3 clinical trials of the calcimimetics have recently been completed, and available data from phase 2 trials suggest that these drugs may eventually provide an effective medical therapy for primary and uremic secondary hyperparathyroidism [34]. Indeed, the calcimimatic, sensipar, recently received FDA approval for the treatment of uremic hyperparathyroidism and parathyroid cancer. A study of 73 hemodialysis patients with uncontrolled secondary hyperparathyroidism, for example, showed that patients treated with a calcimimetic had a statistically significant 33% reduction in serum PTH levels, whereas those treated with placebo had a 3% increase [35]. In addition,

EDWARD M. BROWN

the calcium × phosphate product decreased by 7.9% in the subjects treated with the calcimimetic, while it increased by 11.3% in the controls (P = 0.013). In contrast to calcimimetics, calcilytics provide a means of stimulating endogenous PTH secretion by “tricking” the parathyroid glands into sensing hypocalcemia. A rapidly acting and metabolized calcilytic may provide an alternative to the injection of PTH and its analogs as an anabolic treatment of osteoporosis [33]. The CaR modulates several intracellular signaling systems. It stimulates phospholipases C (PLC), A2 (PLA2) and D (PLD) by coupling to the G proteins, Gq/11 [36]. Stimulation of PLC produces diacylglycerol, which activates protein kinase C (PKC), as well as inositol trisphosphate, which elevates the cytosolic calcium concentration by stimulating the release of calcium from intracellular stores. PLA2 and PLD are stimulated in a PKC-dependent manner [36]; the former generates arachidonic acid, which can be further metabolized to generate multiple messenger molecules, i.e., via the cyclooxygenase and lipoxygenase pathways [37]. The CaR also promotes the influx of extracellular calcium by stimulating Ca2+-permeable ion channels by mechanisms that remain to be elucidated [38]. The receptor also stimulates three mitogen-activated protein kinases (MAPKs)—ERK1/2, p38 MAPK, and JNK [39,40]. All three of these MAPK phosphorylate cytosolic as well as nuclear proteins, and the latter modulate processes such as cell proliferation, differentiation, and apoptosis [41], all of which are known to be regulated by the CaR. Finally, the CaR inhibits adenylate cyclase [42,43]. This inhibition occurs by at least three mechanisms. First, the CaR directly inhibits adenylate cyclase via the inhibitory G protein, Gi. Second, high Ca2+ o stimulates the production of arachidonic acid, which then inhibits adenylate cyclase. Third, a CaR-mediated increase in the cytosolic calcium concentration appears to inhibit calcium-inhibitable isoforms of adenylate cyclase [44]. The first two of these are pertussis toxin sensitive, likely involving Gi, while the third one is pertussis toxin insensitive.

III. ROLE OF THE CaR IN THE PARATHYROID Elevating Ca2+ exerts three key actions on the o parathyroid—reducing PTH secretion, decreasing PTH gene expression, and inhibiting parathyroid cellular proliferation [5]. There is a steep inverse sigmoidal relationship between the circulating level of PTH in vivo or PTH secretion in vitro and the level of Ca2+ o (Fig. 2) [18]. Two important parameters describing this relationship are the midpoint or set point of the curve,

555

CHAPTER 31 Calcium-Sensing Receptor

which contributes to “setting” the level of Ca2+ o , and the slope at the set point, the steepness of which ensures that Ca2+ o is maintained within a narrow range in vivo [18]. The steep inverse sigmoidal relationship between Ca2+ o and PTH secretion is a central element in the homeostatic mechanism in mammals and other tetrapods that maintains near constancy of Ca2+ o (Fig. 3). A reduction in the level of Ca2+ leads to increased PTH secretion, o and this Ca2+ –elevating hormone then acts on target o tissues in ways that will restore Ca2+ to normal. PTH o increases distal tubular reabsorption of calcium [45]— effectively resetting renal tubular calcium reabsorption so as to maintain a higher level of Ca2+ o . PTH also stimulates the 1-hydroxylation of 25-hydroxyvitamin D [46] so as to increase the circulating level of 1,25-dihydroxyvitamin D [1,25(OH)2D3], the body’s second key Ca2+ o -elevating hormone, which then enhances gastrointestinal absorption of calcium and acts, along with PTH, to promote net release of calcium from bone [47]. Thus PTH, through its actions on kidney and bone, not only enhances renal conservation of calcium but also increases the fluxes of calcium into the extracellular fluid from intestine and bone—a concerted series of actions that restores Ca2+ o to normal. The deranged Cao2+–regulated PTH secretion in patients with inactivating or activating mutations of the

Intestine Liver

− −

Bone

+

+ −

Kidney

Ca, PO4

Ca, PO4

ECF Ca++

Vit. D

−

+ 1,25(OH)2D

Ca

−

+ +

+

−

25(OH)D

+

PO4, cAMP

+ PTH

FIGURE 3 The homeostatic system via which Ca 2+ in the ECF o is maintained nearly constant. The solid arrows and lines delineate the actions of PTH and 1,25-dihydroxyvitamin D3 (with the exception of the negative feedback of 1,25(OH)2D3 on PTH secretion, which is shown by a dashed line). The dashed lines and arrows delineate examples of how Ca2+ and phosphate ions directly modulate the functions of target tissues. Plus signs, stimulatory effects; minus signs, inhibitory effects. See text for details. Reproduced with permission from Brown EM, Bai M, Hebert SC 1994 Cloning and characterization of extracellular Ca2+-sensing receptors from parathyroid and kidney: molecular physiology and pathophysiology of Ca2+-sensing. Endocrinologist 4:419–426.

CaR, as well as in mice with knockout of the CaR gene, proves the CaR’s central role in controlling this aspect of parathyroid function [4,19]. Furthermore, the marked parathyroid cellular hyperplasia in infants with NSHPT and mice homozygous for knockout of the CaR gene indicate that the receptor, directly or indirectly, tonically suppresses parathyroid cellular proliferation [19,48]. Preliminary evidence also supports a role for the CaR in mediating the known inhibitory 2+ action of Ca2+ o on PTH gene expression [49]. High Cao also increases the intracellular degradation of PTH, leading to a reduction in the ratio of intact hormone to carboxy-terminal fragments of PTH that are secreted, thereby reducing the amount of bioactive PTH being secreted [50]. This action of high Ca2+ o could be mediated via the CaR, although this possibility has not been formally tested. These effects of high Ca2+ o on parathyroid function provide for a highly orchestrated, temporal and hierarchical control of PTH secretion that permits a graded series of responses, depending on the nature of the hyper- or hypocalcemic stimulus [51]. The most rapid response to hypocalcemia, for example, is increased secretion of preformed stores of PTH, which occurs within seconds [52]. A reduced rate of intracellular degradation of PTH then increases the amount of intact, bioactive PTH available for secretion within about 30 min [53,54]. Increased transcription of the PTH gene and greater stability of preproPTH mRNA augments the amount of PTH synthesized by each parathyroid cell within hours to a day or so [55]. Finally, enhanced parathyroid cellular proliferation occurs within days to weeks or longer and can increase the functional mass of parathyroid tissue enormously [56]. Therefore, depending on the severity and duration of the hypocalcemic stimulus, the parathyroid glands can mount one, several, or all of these responses designed to ensure that sufficient PTH is secreted to normalize Ca2+ o . Although considerable progress has been made in documenting the effects of the CaR on parathyroid function, much remains to learned about the cellular mechanisms through which it exerts these actions (for review, see [51]). Despite several decades of study, the principal intracellular signal transduction pathway(s) through which the CaR regulates PTH secretion and other aspects of parathyroid function remains to be firmly established. In addition to the CaR controlling parathyroid function, several factors modulate CaR gene expression in ways that may be of physiological or pathophysiological relevance. In the chicken but not in the rat parathyroid [57], raising Ca2+ o up-regulates CaR gene expression [58]. If the CaR protein was also up-regulated, which

556

EDWARD M. BROWN

was not addressed in this article, this change could render the CaR more responsive to a given level of Ca2+ o [58]. Administration of 1,25(OH)2D3 in vivo in the rat increases CaR mRNA in the parathyroid [57,59]. The increase in CaR expression in response to increases in Ca2+ o and 1,25(OH)2D3 could provide a mechanism for increasing the activity of the CaR at any given level Ca2+ o and thereby inhibiting PTH secretion under circumstances in which less PTH is needed to maintain 2+ Ca2+ o homeostasis. Interestingly, in the rat high Cao up-regulates expression of the vitamin D receptor (VDR) [57], while vitamin D up-regulates its own receptor [60]. Thus, Ca2+ o and 1,25(OH)2D3 can potentiate their own as well as one another’s actions. Table I summarizes the interactions between calcium/CaR and the VDR. Interleukin (IL)-1β is another factor that up-regulates CaR expression in the parathyroid, with an associated inhibition of PTH release [61]. This IL-1β-induced increase in CaR expression could be a factor contributing to the mild hypocalcemia and inappropriately normal circulating levels of PTH in patients with burn injury [62] or other inflammatory states. Other conditions that are associated with reduced parathyroid CaR expression include primary and uremic secondary hyperparathyroidism in humans [63,64]. This decrease in receptor expression could contribute to the deranged Ca2+ o -regulated PTH secretion in these conditions. Similar to FHH, where the number of normal CaRs is reduced owing to inactivation of one allele of the CaR, reduced CaR expression in hyperparathyroidism could contribute to the increase in the set-point and associated hypercalcemia [18]. In addition, in rats with experimental renal insufficiency as a result of partial nephrectomy, a high phosphorus intake downregulates CaR expression [65]. This alteration in CaR expression is of unknown functional significance, however, since the restoration of normal parathyroid

TABLE I Actions and Interactions of VDR and the CaR in the Parathyroid Biological response CaR expression VDR expression PTH secretion PTH gene expression Cell proliferation aObserved

Action of Cao2+/CaR

Action of vitamin D/VDR

Ref.

⇑a ⇑c ⇓ ⇓b ⇓

⇑ ⇑ ⇓ ⇓ ⇓

58,57 57,60 55 49,55 56

in chicken but not in rat parathyroid gland. in a preliminary study. cAction of high Ca 2+, not proven to be CaR-mediated. o bObserved

function following reinstituting a low-phosphorus diet precedes the associated increase in CaR expression in this animal model.

IV. ROLE OF THE CaR IN THE C-CELL In contrast to the inhibition of PTH secretion by 2+ high Ca2+ o , elevated levels of Cao stimulate secretion of the Ca2+ –lowering hormone, calcitonin (CT), by the o thyroidal C-cells. Calcitonin promotes hypocalcemia by inhibiting bone resorption and, at high doses, stimulating urinary calcium excretion (see Chapter 39). These actions cause substantial hypocalcemia in rodents, but not in adult humans, in whom the hormone contributes little to normal Ca2+ o homeostasis [66]. This hormone has been useful, however, in treating states with elevated bone resorption (e.g., Paget’s disease of bone). Before the cloning of the CaR, the mechanism through which Cao2+ regulates CT secretion was thought to differ in a fundamental way from that in the parathyroid [67]. The availability of DNA and antibody probes for the CaR made it clear, however, that the receptor is expressed in C-cells [68,69]. Subsequent studies have elucidated how the receptor regulates CT secretion in sheep C-cells. The CaR activates calciumand sodium-permeable ion channels, which then produce cellular depolarization and activation of voltage-gated calcium channels. The resultant increase in the cytosolic calcium concentration stimulates CT secretion by classical, calcium-dependent, stimulus–secretion coupling [70]. Administration of vitamin D in the rat increases the expression of CaR mRNA in the thyroid (presumably in C-cells) [71], similar to the result described earlier in the parathyroid cell. Although there are no studies addressing this point, increasing the level of CaR expression could sensitize the response of the C-cell to Ca2+ o , thereby stimulating secretion of this Ca2+ –lowering hormone at a o time when vitamin D is in excess.

V. ROLE OF THE CaR IN THE KIDNEY Calcium exerts diverse actions on the kidney, a number of which are relevant to the physiology and pathophysiology of mineral ion homeostasis [5]. For instance, high Ca2+ o inhibits the 1-hydroxylation of 25-hydroxyvitamin D3 [72], decreases renin secretion, stimulates urinary calcium excretion [73], and diminishes urinary concentrating ability [74]. A number of studies over the past decade have supported the CaR’s roles as mediator of several of these actions. The CaR is expressed in most segments of the nephron [74–76]: It is present in the apical membrane

557

CHAPTER 31 Calcium-Sensing Receptor

of the proximal convoluted (PCT) and straight tubules (PST). In contrast, it is expressed on the basolateral surface of the medullary (MTAL) and cortical thick ascending limbs (CTAL) of Henle’s loop as well as in the macula densa. It resides predominantly on the basolateral surface of the distal convoluted tubule (DCT) and the type A, intercalated cells in the cortical collecting duct (CCD). In the distal portions of the nephron, it is present on the apical membrane of the inner medullary collecting duct (IMCD), the site where vasopressin increases renal tubular reabsorption of water during dehydration. In the proximal tubule, the CaR is a potential mediator of the known, direct inhibitory action of high Ca2+ o on the synthesis of 1,25-dihydroxyvitamin D [72]. There is no direct evidence, however, in support of the receptor’s involvement in this process. The CaR may also regulate the basolateral Na+/K+-ATPase in the PCT [77], which plays an important role in bulk reabsorption of solute in the proximal tubule. This latter action could be a factor contributing to the loss of solute and resultant dehydration that can take place in severely hypercalcemic patients. In the MTAL, the CaR inhibits NaCl reabsorption, thereby diminishing the generation of the hypertonic interstitium needed for vasopressin to maximally stimulate water reabsorption in the IMCD [78]. As will be described in more detail later, the CaR likewise inhibits the action of vasopressin in the IMCD [74], further reducing the ability of the kidney to concentrate the urine. These two actions of the CaR on the renal concentrating mechanism probably account in large measure for the well-documented inhibitory effect of hypercalcemia on urinary concentrating capacity [79]. The high Ca2+ o induced decrease in NaCl reabsorption in the MTAL may be mediated by 20-HETE [80], which is produced as a result of the action of the CYP4A P450-hydroxylase on arachidonic acid (AA) that is liberated by CaRmediated activation of PLA2. An alternative pathway of AA metabolism after its liberation by CaR-evoked stimulation of PLA2 is cyclooxygenase-mediated conversion to PGE2, a known inhibitor of salt reabsorption in the MTAL [81]. In the CTAL, the CaR inhibits the overall activity of the mechanism driving the paracellular reabsorption of NaCl, Ca2+, and Mg2+ (Fig. 4) [80,82]. This system normally operates as follows: the Na+/K+/2Cl cotransporter residing in the apical membrane absorbs Na+, K+, and Cl−. NaCl and Cl− exit the cells via the basolateral membrane, while K+ is recycled into the tubular fluid by an apical K+ channel, thereby generating a lumen positive transepithelial potential. Hormones, such as PTH, which stimulate Ca2+ and Mg2+ reabsorption in the CTAL do so via a cAMP-dependent activation of the Na+/K+/2Cl− cotransporter and the apical recycling

Lumen-positive voltage

+ Urinary space

− Basolateral space Stimulates transport

Na+ 2Cl− K+ H2O

3 P450 4

K+

Hormones cAMP 1

Ca+

Cl− PLA2 2

Ca2+

AA

Ca2+, Mg2+

FIGURE 4 Possible mechanisms through which the CaR controls reabsorption of calcium and magnesium in the CTAL (see text for further details). PTH acts on the CTAL via its basolateral receptor to increase cAMP levels, thereby increasing the overall activity of the Na+/K+/2Cl− cotransporter and recycling K+ channel, which, in turn, increases the lumen positive potential that drives paracellular reabsorption of Ca2+ and Mg2+. Activation of the CaR inhibits this process by inhibiting adenylate cyclase and generating arachidonic acid (AA), which is metabolized to products of the P450 pathway that inhibit the K+ channel and Na+/K+/Cl− cotransporter. Both of these processes diminish the lumen positive potential and pari passu, reabsorption of the divalent cations. Reprinted with permission from Brown EM, Hebert SC 1997 Calcium-receptor regulated parathyroid and renal function. Bone 20:303–309.

K+ channel [82,83]. This increases the lumen-positive potential, thereby promoting reabsorption of NaCl and divalent cations through the paracellular route. The CaR probably diminishes this process by 20-HETEmediated inhibition of the apical K+ channel [80] and by reducing cAMP generation (Fig. 4) [84]. Thus the CaR not only reduces renal calcium reabsorption by CaR-mediated inhibition of PTH secretion but also blocks the effect of PTH on this nephron segment to further reduce renal reabsorption of calcium and magnesium by a direct, CaR-mediated, renal action of Ca2+ o . The CaR’s role in directly regulating renal tubular reabsorption of Ca2+ is strongly supported by the finding that persons with FHH exhibit a marked reduction in their capacity to up-regulate Ca2+ o excretion in the CTAL in the setting of hypercalcemia—even following parathyroidectomy [22]. Vitamin D has been shown in two studies to up-regulate the mRNA encoding the CaR in the kidney [57,59]. The location(s) where this effect takes place is not currently known, but it could conceivably exert important effects on renal function by potentiating the effect of a given level of Ca2+ o on the CaR. In the DCT, reabsorption of calcium and magnesium occurs by the transcellular route [85]. Calcium initially enters the cells via a recently cloned, calcium-permeable channel in the luminal membrane (epithelial calcium

558 channel or ECaC) [86,87] and then exits the basolateral membrane through the Na+/Ca2+-exchanger and/or Ca2+-ATPase. The level of expression of ECaC is upregulated by 1,25(OH)2D3, thereby exerting important control over this rate-limiting step for calcium reabsorption in this nephron segment. High calcium up-regulates the production of calbindin D28K in this nephron segment [88], a biological action that is potentially mediated by the CaR. The biological relevance of this up-regulation of calbindin D28K is not at all clear, since one might have expected that an increase in calbindin D28K would increase calcium reabsorption in the DCT. Calcium can block its own transcellular absorption in MDCK cells—a model of the DCT—by inhibiting the Ca2+-ATPase [89]. In MDCK cells, the CaR activates the inhibitory G proteins Gi2 and Gi3 [90], actions that could contribute to the accompanying reduction in calcium absorption. Moreover, in an immortalized mouse DCT cell line, elevated levels of Ca2+ o diminish 1,25dihydroxyvitamin D-activated magnesium entry, probably via a CaR-mediated mechanism [91]. Further studies are clearly needed to provide more definitive evidence for the involvement of the CaR in the regulation of renal tubular reabsorption in the DCT. In the IMCD, available data implicate the CaR in regulating vasopressin-stimulated water reabsorption in a physiologically important manner [74,92]. Consistent with the presence of the CaR on the apical but not the basolateral membrane of the IMCD, perfusion of the luminal (but not the basolateral) surface of isolated IMCD segments with an elevated level of Ca2+ o reduces vasopressin-stimulated water flow by about 40% [74]. The CaR’s role in this process is supported by the fact that individuals with FHH concentrate their urine more than subjects with primary hyperparathyroidism who have an equivalent degree of hypercalcemia [93]. That is, similar to the resistance of parathyroid secretion and renal calcium reabsorption (most likely in the CTAL) to Ca2+ o in FHH, there also seems to be resistance of the IMCD to the usual inhibitory effect of Ca2+ o on vasopressin-stimulated water flow. Vasopressin increases water reabsorption in the IMCD by stimulating the insertion of aquaporin-2containing vesicles that reside below the apical membrane of the cells into the luminal plasma membrane [93,94]. The hypertonic interstitial fluid surrounding the IMCD then permits passive reabsorption of water. Of interest, the CaR is present along with aquaporin-2 in these vesicles as well as on the apical cell membrane and probably traffics into and out of the plasma membrane along with aquaporin-2 [74]. Since vasopressin produces its effect on aquaporin-2 trafficking by stimulating adenylate cyclase, the CaR could inhibit this process, in part, by diminishing vasopressin-stimulated cAMP accumulation.

EDWARD M. BROWN

What is the significance of the mutually antagonistic effects of vasopressin and high Ca2+ o —presumably acting via the CaR—on vasopressin-enhanced water reabsorption? Renal calcium and water handling have usually been thought to function largely independently of one another under normal physiological conditions. However, there are conditions in which it would be desirable to create a “fail-safe” system that could prevent inordinate urinary concentration when Ca2+ o in the tubular fluid is high [82]. For example, consider the requirement to excrete a calcium load after ingestion of a calcium-rich meal late in the day. The excess calcium would be excreted during the nighttime hours when there was no ingestion of water so as to maintain a sufficiently dilute urine to avoid the possible risk of forming renal stones. By reducing vasopressin-stimulated flow of water, the CaR would, in effect, set an upper limit for the level of Ca2+ that would be reached in the distal nephron. Moreover, if there were habitual intake of excessive amounts of calcium (e.g., in the milk-alkali syndrome) an elevated level of Ca2+ o , via CaR-mediated reduction of NaCl reabsorption in the MTAL [95], would “wash out” the medullary interstitium and further decrease the Ca2+ o concentration that could be achieved in the IMCD due to vasopressin (Fig. 3). Thus coordination of renal calcium and water handling allows for “trade-offs” in the renal handling of calcium and water under physiological circumstances that might otherwise put the nephron in jeopardy.

VI. THE CaR IN BONE AND INTESTINE What is the role of the CaR, if any, in the other two major Ca2+ o –translocating tissues, bone and intestine, that are important contributors to maintaining Ca2+ o homeostasis? High Ca2+ o increases bone formation and diminishes bone resorption in vitro. These actions of Ca2+ o could participate in calcium homeostasis by allowing bone to serve as a “sink” for calcium ions when Ca2+ o is high or as a reservoir for calcium ions when Ca2+ o is low. The role of the CaR, however, if any, in mediating these effects of Ca2+ o remains uncertain. Some authors [96,97], but not others [98], have found the CaR to be present in osteoblast and osteoclast precursors as well as in mature osteoblasts and osteoclasts. Furthermore, the pharmacology for the actions of various divalent and trivalent cations on the functions of cells of both the osteoclast and osteoblast lineages is different in some cases from that expected of the CaR [99–101]. Thus, while there is agreement that Ca2+ o regulates several functions of osteoblasts and osteoclasts and their precursors that participate importantly in bone turnover, identifying the relevant Ca2+ o -sensing mechanism(s) has been elusive.

CHAPTER 31 Calcium-Sensing Receptor

The CaR has been shown to be expressed in epithelial cells along the length of the GI tract that participate in the absorption of dietary calcium, e.g., in the proximal small and large intestines [102,103]. It is likewise expressed in the enteric nervous system, where it potentially could participate in the known inhibitory and stimulatory effects of high and low levels of Ca2+ o , respectively, on gastrointestinal motility. Further work is needed, however, to understand whether the receptor has any physiologically relevant actions on mineral ion absorption from the intestine or other aspects of the GI function that are relevant to mineral ion metabolism.

VII. SUMMARY The CaR is the crucial Ca2+ o sensor in the parathyroid chief cells, C-cells, and cells along the length of the nephron that participate in the control of renal reabsorption of calcium. In response to small changes 2+ in Ca2+ o , the CaR regulates the secretion of the Cao – elevating hormone, PTH, and the Ca2+ -lowering horo mone, CT, in ways that will reestablish normocalcemia. It may also mediate the known, direct inhibitory action of Ca2+ o on the renal synthesis of 1,25(OH)2D3. In addition to modulating the production of these calciotropic hormones, it also exerts direct effects on the kidney, e.g., in the CTAL. Therefore, by diminishing PTH production and enhancing CT secretion as well as by stimulating urinary calcium excretion—both indirectly via PTH and directly via its effects in the CTAL—the CaR permits Ca2+ o to function as the body’s most effective Ca2+ –lowering hormone. In addition to its calciotropic o effects, the CaR likewise integrates renal calcium and water metabolism so as to facilitate the adaptation of free-living terrestrial organisms to their intermittent access to dietary calcium and water.

References 1. Brown EM 1991 Extracellular Ca2+ sensing, regulation of parathyroid cell function, and role of Ca2+ and other ions as extracellular (first) messengers. Physiol Rev 71:371–411. 2. Pozzan T, Rizzuto R, Volpe P, Meldolesi J 1994 Molecular and cellular physiology of intracellular calcium stores. Physiol Rev 74:595–636. 3. Brown EM, Gamba G, Riccardi D, Lombardi M, Butters R, Kifor O, Sun A, Hediger MA, Lytton J, Hebert SC 1993 Cloning and characterization of an extracellular Ca2+-sensing receptor from bovine parathyroid. Nature 366:575–580. 4. Brown EM 1999 Physiology and pathophysiology of the extracellular calcium-sensing receptor. Am J Med 106:238–253. 5. Brown EM MacLeod RJ 2001 Extracellular calcium sensing and extracellular calcium signaling. Physiol Rev 81: 239–297.

559 6. Nemeth E, Scarpa A 1987 Rapid mobilization of cellular Ca2+ in bovine parathyroid cells by external divalent cations. J Biol Chem 202:5188–5196. 7. Shoback DM, Membreno LA, McGhee JG 1988 High calcium and other divalent cations increase inositol trisphosphate in bovine parathyroid cells. Endocrinology 123:382–389. 8. Brown EM, Enyedi P, LeBoff M, Rothberg J, Preston J, Chen C 1987 High extracellular Ca2+ and Mg2+ stimulate accumulation of inositol phosphates in bovine parathyroid cells. FEBS Lett 218:113–118. 9. Ray K, Clapp P, Goldsmith PK, Spiegel AM 1998 Identification of the sites of N-linked glycosylation on the human calcium receptor and assessment of their role in cell surface expression and signal transduction. J Biol Chem 273:34558–34567. 10. Bai M, Trivedi S, Brown EM 1998 Dimerization of the extracellular calcium-sensing receptor (CaR) on the cell surface of CaR-transfected HEK293 cells. J Biol Chem 273:23605–23610. 11. Ray K, Hauschild BC, Steinbach PJ, Goldsmith PK, Hauache O, Spiegel AM 1999 Identification of the cysteine residues in the amino-terminal extracellular domain of the human Ca2+ receptor critical for dimerization. Implications for function of monomeric Ca(2+) receptor. J Biol Chem 274:27642–27650. 12. Hammerland LG, Krapcho KJ, Garrett JE, Alasti N, Hung BC, Simin RT, Levinthal C, Nemeth EF, Fuller FH 1999 Domains determining ligand specificity for Ca2+ receptors. Mol Pharmacol 55:642–648. 13. Chang W, Chen TH, Pratt S, Shoback D 2000 Amino acids in the second and third intracellular loops of the parathyroid Ca2+-sensing receptor mediate efficient coupling to phospholipase C. J Biol Chem 275:19955–19963. 14. Bai M, Trivedi S, Kifor O, Quinn SJ, Brown EM 1999 Intermolecular interactions between dimeric calcium-sensing receptor monomers are important for its normal function. Proc Natl Acad Sci USA 96:2834–2839. 15. Hu J, Reyes-Cruz G, Chen W, Jacobson KA, Spiegel AM. 2002 Identification of acidic residues in the extracellular loops of the seven transmembrane domain of the human Ca2+ receptor critical for response to Ca2+ and a positive allosteric modulator. J Biol Chem 277:46622–46631. 16. Kunishima N, Shimada Y, Tsuji Y, Sato T, Yamamoto M, Kumasaka T, Nakanishi S, Jingami H, Morikawa K 2000 Structural basis of glutamate recognition by a dimeric metabotropic glutamate receptor. Nature 407:971–977. 17. Reyez-Cruz G, Hu J, Goldsmith PK, Steinbach PJ, Spiegel AM 2001 Human Ca(2+) receptor extracellular domain. Analysis of function of lobe I deletion mutants. J Biol Chem 276: 32145–32151. 18. Brown EM 1983 Four parameter model of the sigmoidal relationship between parathyroid hormone release and extracellular calcium concentration in normal and abnormal parathyroid tissue. J Clin Endocrinol Metab 56:572–581. 19. Ho C, Conner DA, Pollak MR, Ladd DJ, Kifor O, Warren HB, Brown EM, Seidman JG, Seidman CE 1995 A mouse model of human familial hypocalciuric hypercalcemia and neonatal severe hyperparathyroidism. Nat Genet 11:389–394. 20. Pollak MR, Brown EM, Chou YH, Hebert SC, Marx SJ, Steinmann B, Levi T, Seidman CE, Seidman JG 1993 Mutations in the human Ca(2+)-sensing receptor gene cause familial hypocalciuric hypercalcemia and neonatal severe hyperparathyroidism. Cell 75:1297–1303. 21. Marx S, Spiegel A, Brown E, Koehler J, Gardner D, Brennan M, Aurbach G 1978 Divalent cation metabolism. Familial hypocalciuric hypercalcemia versus typical primary hyperparathyroidism. Am J Med 65:235–242.

560 22. Attie MF, Gill Jr J, Stock JL, Spiegel AM, Downs Jr RW, Levine MA, Marx SJ 1983 Urinary calcium excretion in familial hypocalciuric hypercalcemia. Persistence of relative hypocalciuria after induction of hypoparathyroidism. J Clin Invest 72:667–676. 23. Khosla S, Ebeling PR, Firek AF, Burritt MM, Kao PC, Heath III H 1993 Calcium infusion suggests a “set-point” abnormality of parathyroid gland function in familial benign hypercalcemia and more complex disturbances in primary hyperparathyroidism. J Clin Endocrinol Metab 76:715–720. 24. Auwerx J, Demedts M, Bouillon R 1984 Altered parathyroid set point to calcium in familial hypocalciuric hypercalcaemia. Acta Endocrinologica (Copenh) 106:215–218. 25. Baron J, Winer KK, Yanovski JA, Cunningham AW, Laue L, Zimmerman D, Cutler GB Jr 1996 Mutations in the Ca(2+)sensing receptor gene cause autosomal dominant and sporadic hypoparathyroidism. Hum Mol Genet 5:601–606. 26. Pollak MR, Brown EM, Estep HL, McLaine PN, Kifor O, Park J, Hebert SC, Seidman CE, Seidman JG 1994 Autosomal dominant hypocalcaemia caused by a Ca(2+)-sensing receptor gene mutation. Nat Genet 8:303–307. 27. Conigrave AD, Quinn SJ, Brown EM 2000 L-Amino acid sensing by the extracellular Ca2+-sensing receptor. Proc Natl Acad Sci USA 97:4814–4819. 28. Strewler GJ 1994 Familial benign hypocalciuric hypercalcemia— from the clinic to the calcium sensor [editorial; comment]. West J Med 160:579–580. 29. McArthur KE, Isenberg JI, Hogan DL, Dreier SJ 1983 Intravenous infusion of L-isomers of phenylalanine and tryptophan stimulate gastric acid secretion at physiologic plasma concentrations in normal subjects and after parietal cell vagotomy. J Clin Invest 71:1254–1262. 30. Conigrave AD, Quinn SJ, Brown EM 2000 Cooperative multimodal sensing and therapeutic implications of the extracellular Ca2+ sensing receptor. Trends Pharmacol Sci 21:401–407. 31. Kerstetter JE, Caseria DM, Mitnick ME, Ellison AF, Gay LF, Liskov TA, Carpenter TO, Insogna KL 1997 Increased circulating concentrations of parathyroid hormone in healthy, young women consuming a protein-restricted diet. Am J Clin Nutr 66:1188–1196. 32. Nemeth EF, Steffey ME, Hammerland LG, Hung BC, Van Wagenen BC, DelMar EG, Balandrin MF 1998 Calcimimetics with potent and selective activity on the parathyroid calcium receptor. Proc Natl Acad Sci USA 95: 4040–4045. 33. Gowen M, Stroup GB, Dodds RA, James IE, Votta BJ, Smith BR, Bhatnagar PK, Lago AM, Callahan JF, DelMar EG, Miller MA, Nemeth EF, Fox J 2000 Antagonizing the parathyroid calcium receptor stimulates parathyroid hormone secretion and bone formation in osteopenic rats. J Clin Invest 105:1595–1604. 34. Nemeth EF, Fox J 1999 Calcimimetic compounds: A direct approach to controlling plasma levels of parathyroid hormone in hyperparathyroidism. Trends Endocrinol Metab 10:66–71. 35. Quarles LD, Sharrard DJ, Adler S, Rosansky SJ, McCary LC, Liu W, Bushinsky DA 2003 The calcimimetic AMG073 as a potential treatment for secondary hyperparathyroidism of endstage renal disease. J Am Soc Nephrol 14:575–583. 36. Kifor O, Diaz R, Butters R, Brown EM 1997 The Ca2+-sensing receptor (CaR) activates phospholipases C, A2, D in bovine parathyroid and CaR-transfected, human embryonic kidney (HEK293) cells. J Bone Miner Res 12:715–725. 37. Wang W, Lu M, Balazy M, Hebert SC 1997 Phospholipase A2 is involved in mediating the effect of extracellular Ca2+ on apical K+ channels in rat TAL. Am J Physiol 273:F421–429.

EDWARD M. BROWN

38. Chang W, Chen TH, Gardner P, Shoback D 1995 Regulation of Ca(2+)-conducting currents in parathyroid cells by extracellular Ca2+ and channel blockers. Am J Physiol 269, E864–877. 39. McNeil SE, Hobson SA, Nipper V, Rodland KD 1998 Functional calcium-sensing receptors in rat fibroblasts are required for activation of SRC kinase and mitogen-activated protein kinase in response to extracellular calcium. J Biol Chem 273:1114–1120. 40. Kifor O, MacLeod RJ, Diaz R, Bai M, Yamaguchi T, Yao T, Kifor I, Brown EM 2001 Regulation of MAP kinase by calciumsensing receptor in bovine parathyroid and CaR-transfected HEK293 cells. Am J Physiol 280:F291–302. 41. Davis RJ 1993 The mitogen-activated protein kinase signal transduction pathway. J Biol Chem 268:14553–14556. 42. Chen C, Barnett J, Congo D, Brown E 1989 Divalent cations suppress 3′,5′-adenosine monophosphate accumulation by stimulating a pertussis toxin-sensitive guanine nucleotide-binding protein in cultured bovine parathyroid cells. Endocrinology 124:233–239. 43. Chang W, Pratt S, Chen TH, Nemeth E, Huang Z, Shoback D 1998 Coupling of calcium receptors to inositol phosphate and cyclic AMP generation in mammalian cells and Xenopus laevis oocytes and immunodetection of receptor protein by regionspecific antipeptide antisera. J Bone Miner Res 13:570–580. 44. de Jesus Ferreira MC, Helies-Toussaint C, Imbert-Teboul M, Bailly C, Verbavatz JM, Bellanger AC, Chabardes D 1998 Co-expression of a Ca2+-inhibitable adenylyl cyclase and of a Ca2+-sensing receptor in the cortical thick ascending limb cell of the rat kidney. Inhibition of hormone-dependent cAMP accumulation by extracellular Ca2+. J Biol Chem 273:15192–15202. 45. Friedman PA, Coutermarsh BA, Kennedy SM, Gesek FA 1996 Parathyroid hormone stimulation of calcium transport is mediated by dual signaling mechanisms involving protein kinase A and protein kinase C. Endocrinology 137:13–20. 46. Fraser DR 1980 Regulation of the metabolism of vitamin D. Physiol Rev 60:551. 47. Henry HL, Norman AW 1984 Vitamin D metabolism and biological actions. Ann Rev Nutr 4:493. 48. Spiegel AM, Harrison HE, Marx SJ, Brown EM, Aurbach GD 1977 Neonatal primary hyperparathyroidism with autosomal dominant inheritance. J Pediatr 90:269–272. 49. Garrett J, Steffey M, Nemeth E 1995 The calcium receptor agonist R-568 suppresses PTH mRNA levels in cultured bovine parathyroid cells. J Bone Miner Res 10 (suppl 1):S387 (Abstract M539. 50. Habener J, B BK, Potts JJ 1975 Calcium-dependent intracellular degradation of parathyroid hormone. A possible mechanism for the regulation of hormone stores. Endocrinology 97:431–441. 51. Diaz R, El-Hajj Fuleihan G, Brown EM 1998 Regulation of parathyroid function. In: Fray JJ, (ed). Handbook of Physiology, Section 7: Endocrinology, Vol. III. Hormonal Regulation of Water and Electrolyte Balance. Oxford University Press, New York, pp. 607–662. 52. Brown EM, Leombruno R, Thatcher J, Burrowes M 1985 The acute secretory response to alterations in the extracellular calcium concentration and dopamine in perifused bovine parathyroid cells. Endocrinology 116:1123–1132. 53. Morrissey JJ, Hamilton JW, MacGregor RR, Cohn DV 1980 The secretion of parathormone fragments 34–84 and 37–84 by dispersed porcine parathyroid cells. Endocrinology 107:164–171. 54. Mayer GP, Keaton JA, Hurst JC, Habener JF 1979 Effects of plasma calcium concentration on the relative proportion of hormone and carboxyl fragments in parathyroid venous blood. Endocrinology 104:1778–1784.

CHAPTER 31 Calcium-Sensing Receptor

55. Silver J, Moallem E, Kilav R, Epstein E, Sela A, Naveh-Many T 1996 New insights into the regulation of parathyroid hormone synthesis and secretion in chronic renal failure. Nephrol Dial Transplant 11:2–5. 56. Silver J, Sela SB, Naveh-Many T 1997 Regulation of parathyroid cell proliferation. Curr Opin Nephrol Hypertens 6:321–326. 57. Brown AJ, Zhong M, Finch J, Ritter C, McCracken R, Morrissey J, Slatopolsky E 1996 Rat calcium-sensing receptor is regulated by vitamin D but not by calcium. Am J Physiol 270:F454–460. 58. Yarden N, Lavelin I, Genina O, Hurwitz S, Diaz R, Brown EM, Pines M 2000 Expression of calcium-sensing receptor gene by avian parathyroid gland in vivo: relationship to plasma calcium. Gen Comp Endocrinol 117:173–181. 59. Canaff L, Hendy GN 2002 Human calcium-sensing receptor gene. Vitamin D response elements in promoters P1 and P2 confer transcriptional responsiveness to 1,25-dihydroxyvitamin D. J Biol Chem 277:30337–30350. 60. Naveh-Many T, Marx R, Keshet E, Pike JW, Silver J 1990 Regulation of 1,25-dihydroxyvitamin D3 receptor gene expression by 1,25-dihydroxyvitamin D3 in the parathyroid gland in vivo. J Clin Invest 86:1969–1975. 61. Nielsen PK, Rasmussen AK, Butters R, Feldt-Rasmussen U, Bendtzen K, Diaz R, Brown EM, Olgaard K 1997 Inhibition of PTH secretion by interleukin-1 beta in bovine parathyroid glands in vitro is associated with an up-regulation of the calcium-sensing receptor mRNA. Biochem Biophys Res Commun 238:880–885. 62. Murphey ED, Chattopadhyay N, Bai M, Kifor O, Harper D, Traber DL, Hawkins HK, Brown EM, Klein GL 2000 Upregulation of the parathyroid calcium-sensing receptor after burn injury in sheep: a potential contributory factor to postburn hypocalcemia. Crit Care Med 28:3885–3890. 63. Gogusev J, Duchambon P, Hory B, Giovannini M, Goureau Y, Sarfati E, Drueke TB 1997 Depressed expression of calcium receptor in parathyroid gland tissue of patients with hyperparathyroidism. Kidney Int 51:328–336. 64. Kifor O, Moore FD, Jr, Wang P, Goldstein M, Vassilev P, Kifor I, Hebert SC, Brown EM 1996 Reduced immunostaining for the extracellular Ca2+-sensing receptor in primary and uremic secondary hyperparathyroidism [see comments]. J Clin Endocrinol Metab 81:1598–1606. 65. Brown AJ, Ritter CS, Finch JL, Slatopolsky EA 1999 Decreased calcium-sensing receptor expression in hyperplastic parathyroid glands of uremic rats: role of dietary phosphate. Kidney Int 55:1284–1292. 66. Austin LA, Heath III H 1981 Calcitonin: Physiology and pathophysiology. N Engl J Med 304:269–278. 67. Eskert R, Scherubl H, Petzelt C, Friedhelm R, Ziegler R 1989 Rhythmic oscillations of cytosolic calcium in rat C-cells. Mol Cell Endocrinol 64:267–270. 68. Freichel M, Zink-Lorenz A, Holloschi A, Hafner M, Flockerzi V, Raue F 1996 Expression of a calcium-sensing receptor in a human medullary thyroid carcinoma cell line and its contribution to calcitonin secretion. Endocrinology 137: 3842–3848. 69. Garrett JE, Tamir H, Kifor O, Simin RT, Rogers KV, Mithal A, Gagel RF, Brown EM 1995 Calcitonin-secreting cells of the thyroid express an extracellular calcium receptor gene. Endocrinology 136:5202–5211. 70. McGehee DS, Aldersberg M, Liu KP, Hsuing S, Heath MJ, Tamir H 1997 Mechanism of extracellular Ca2+ receptor-stimulated hormone release from sheep thyroid parafollicular cells. J Physiol (Lond) 502:31–44.

561 71. Canaff L, Petit JL, Kisiel M, Watson PH, Gascon-Barre M, Hendy GN 2001 Extracellular calcium-sensing receptor is expressed in rat hepatocytes: Coupling to intracellular calcium mobilization and stimulation of bile flow. J Biol Chem. 276:4070–4079. 72. Weisinger JR, Favus MJ, Langman CB, Bushinsky D 1989 Regulation of 1,25-dihydroxyvitamin D3 by calcium in the parathyroidectomized, parathyroid hormone-replete rat. J Bone Miner Res 4:929–935. 73. Quamme GA 1982 Effect of hypercalcemia on renal tubular handling of calcium and magnesium. Can J Physiol Pharmacol 60:1275–1280. 74. Sands JM, Naruse M, Baum M, Jo I, Hebert SC, Brown EM, Harris HW 1997 Apical extracellular calcium/polyvalent cation-sensing receptor regulates vasopressin-elicited water permeability in rat kidney inner medullary collecting duct. J Clin Invest 99:1399–1405. 75. Riccardi D, Hall AE, Chattopadhyay N, Xu JZ, Brown EM, Hebert SC 1998 Localization of the extracellular Ca2+/polyvalent cation-sensing protein in rat kidney. Am J Physiol 274:F611–622. 76. Riccardi D, Lee WS, Lee K, Segre GV, Brown EM, Hebert SC 1996 Localization of the extracellular Ca2+-sensing receptor and PTH/PTHrP receptor in rat kidney. Am J Physiol 271:F951–956. 77. Levi M, Molitoris BA, Burke TJ, Schrier RW, Simon FR 1987 Effects of vitamin D–induced chronic hypercalcemia on rat renal cortical plasma membranes and mitochondria. Am J Physiol 252:F267–275. 78. Hebert SC, Brown EM, Harris HW 1997 Role of the Ca2+sensing receptor in divalent mineral ion homeostasis. J Exp Biol 200:295–302. 79. Levi M, Peterson L, Berl T 1983 Mechanism of concentrating defect in hypercalcemia role of polydipsia and prostaglandins. Kidney Int 23:489–497. 80. Wang WH, Lu M, Hebert SC 1996 Cytochrome P-450 metabolites mediate extracellular Ca2+-induced inhibition of apical K+ channels in the TAL. Am J Physiol 271:C103–111. 81. Wang SJ, Wang WH, McGiff JC, Ferreri NR 2001 CaR-mediated COX-2 expression in primary cultured mTAL cells. Am J Physiol Renal Physiol 281:F658-F664. 82. Hebert SC, Brown EM, Harris HW 1997 Role of the Ca2+sensing receptor in divalent mineral ion homeostasis. J Exp Biol 200:295–302. 83. De Rouffignac C, Di Stefano A, Wittner M, Roinel N, Elalouf JM 1991 Consequences of differential effects of ADH and other peptide hormones on thick ascending limb of mammalian kidney [editorial]. Am J Physiol 260:R1023–1035. 84. Takaichi K, Kurokawa K 1988 Inhibitory guanosine triphosphate-binding protein-mediated regulation of vasopressin action in isolated single medullary tubules of mouse kidney. J Clin Invest 82:1437–1444. 85. Gesek FA, Friedman PA 1992 Mechanism of calcium transport stimulated by chlorothiazide in mouse distal convoluted tubule cells. J Clin Invest 90:429–438. 86. Hoenderop JGJ, Van de Graaf AWCM, Hartog A, van de Graaf SFJ, van Os CH, Willems PHGM, Bindels RJM 1999 Molecular identification of the apical Ca2+ channel in 1,25dihydroxyvitamin D3–responsive epithelia. J Biol Chem 274: 8375–8378. 87. Peng JB, Chen XZ, Berger UV, Vassilev PM, Brown EM, Hediger MA 2000 A rat kidney-specific calcium transporter in the distal nephron. J Biol Chem 275:28186–28194. 88. Clemens T, McGlade S, Garrett K, Craviso G, Hendy G 1989 Extracellular calcium modulates vitamin D–dependent calbindin-D28k gene expression in chick kidney cells. Endocrinology 124:1582–1584.

562 89. Blankenship KA, Williams JJ, Lawrence MS, McLeish KR, Dean WL, Arthur JM 2001 The calcium-sensing receptor regulates calcium absorption in MDCK cells by inhibition of PMCA. Am J Physiol Renal Physiol 280:F815-F822. 90. Arthur JM, Collinsworth GP, Gettys TW, Quarles LD, Raymond JR 1997 Specific coupling of a cation-sensing receptor to G protein alpha-subunits in MDCK cells. Am J Physiol 273:F129–135. 91. Ritchie G, Kerstan D, Dai LJ, Kang HS, Canaff L, Hendy GN, Quamme GA 2001 1,25(OH)2D3 stimulates Mg2+ uptake into MDCT cells: Modulation by extracellular Ca2+ and Mg2+. Am J Physiol Renal Physiol 280:F868–F878. 92. Sands JM, Flores FX, Kato A, Baum MA, Brown EM, Ward DT, Hebert SC, Harris HW 1998 Vasopressin-elicited water and urea permeabilities are altered in IMCD in hypercalcemic rats. Am J Physiol 274:F978–F985. 93. Marx SJ, Attie MF, Stock JL, Spiegel AM, Levine MA 1981 Maximal urine-concentrating ability: Familial hypocalciuric hypercalcemia versus typical primary hyperparathyroidism. J Clin Endocrinol Metab 52:736–740. 94. Robertson GL 1995 Diabetes insipidus. Endocrinol Metab Clin North Am 24:549–572. 95. Wang D, An SJ, Wang WH, McGiff JC, Ferreri NR 2001 CaRmediated COX-2 expression in primary cultured mTAL cells. Am J Physiol Renal Physiol 281:F658–664. 96. Yamaguchi T, Chattopadhyay N, Brown EM 2000 G proteincoupled extracellular Ca2+[Ca(2+) ο ]-sensing receptor(CaR): Roles in

EDWARD M. BROWN

97.

98. 99.

100. 101. 102.

103.

cell signaling and control of diverse cellular functions. Adv Pharmacol 47:209–253. Chang W, Tu C, Chen TH, Komuves L, Oda Y, Pratt S, Miller S, Shoback D 1999 Expression and signal transduction of calcium-sensing receptors in cartilage and bone. Endocrinology 140:5883–5893. Pi M, Hinson TK, Quarles L. 1999 Failure to detect the extracellular calcium-sensing receptor (CasR) in human osteoblast cell lines. J Bone Miner Res 14:1310–1319. Quarles DL, Hartle II JE, Siddhanti SR, Guo R, Hinson TK 1997 A distinct cation-sensing mechanism in MC3T3-E1 osteoblasts functionally related to the calcium receptor. J Bone Miner Res 12:393–402. Quarles LD 1997 Cation-sensing receptors in bone: A novel paradigm for regulating bone remodeling? J Bone Miner Res 12:1971–1974. Zaidi M, Adebanjo OA, Moonga BS, Sun L, Huang CL 1999 Emerging insights into the role of calcium ions in osteoclast regulation. J Bone Miner Res 14:669–674. Chattopadhyay N, Cheng I, Rogers K, Riccardi D, Hall A, Diaz R, Hebert SC, Soybel DI, Brown EM 1998 Identification and localization of extracellular Ca2+-sensing receptor in rat intestine. Am J Physiol 274:G122–130. Cheng I, Klingensmith ME, Chattopadhyay N, Kifor O, Butters RR, Soybel DI, Brown EM 1998 Identification and localization of the extracellular calcium-sensing receptor in human breast. J Clin Endocrinol Metab 83:703–707.

CHAPTER 32

Bone PAULA H. STERN

Department of Molecular Pharmacology Northwestern University Feinberg School of Medicine Chicago, IL 60611

I. Introduction II. Vitamin D Actions on Bone Mineralization III. Vitamin D and Bone Target Genes

IV. Vitamin D Actions on Bone Resorption V. Summary and Conclusions References

I. INTRODUCTION

II. VITAMIN D ACTIONS ON BONE MINERALIZATION

The goal of this chapter is to present an overview of effects of vitamin D and metabolites on bone, focusing on more recent studies, and with an emphasis on in vivo effects. A number of other chapters address aspects of this broad topic in greater detail from different perspectives, and these are noted in the specific sections. Advances in genetic analysis and the generation of knockout mice have supplemented dietary studies and allowed for a more mechanistic understanding of the effects of vitamin D on bone. Gene targeting has allowed the generation of mice that have been valuable models for studies of vitamin D deficiency or excess on bone. Several groups have now generated vitamin D receptor knockout models [1,2]. These mice exhibit low bone mass, hypocalcemia, hypophosphatemia, increased calcitriol, and low 24,25-(OH)2D3, and are a model for hereditary vitamin D–resistant rickets. A detailed description of models of vitamin D insufficiency or excess resulting from manipulation of the vitamin D receptor is discussed in Chapter 20. The current chapter is divided into three major sections. Within each, we consider the impact of new data on a critical question related to vitamin D and bone. In Section II, on mineralization, the issue is whether vitamin D itself is required for mineralization or whether adequate calcium and phosphate are sufficient. In Section III, on effects of vitamin D on specific genes in bone, the question is whether there is a direct effect of the hormone to promote bone formation. In Section IV, on bone resorption, the issue is at what level (physiological or supraphysiological) of vitamin D the hormone causes resorption of bone. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Centuries before vitamin D was recognized as a chemical entity, an anecdotal report, preserved as part of a larger historical work, suggested that the beneficial effects of sunlight on bone may have been recognized. In book III of The Persian Wars (500–600 B.C.), Herodotus notes from observations of dead soldiers on the battlefield that the skulls of the Egyptians were stronger than the skulls of the Persians. The explanation given for this was that the Egyptians had their heads shaved and that the action of the sun caused their skulls to become thick and hard, whereas the Persians had weaker skulls because their heads were covered by turbans [3]. Twenty-five centuries passed before the published description of ultraviolet light as a treatment for rickets in the early 20th century [4]. Studies in the early 1900s also revealed that specific dietary deficiencies of phosphorus and fat-soluble vitamins led to rickets [5–9]. The observations that irradiation of the diet conferred antirachitic properties [10,11] ultimately led to the investigations of vitamin D metabolism [12,13] that have provided a basis for our current understanding of the dietary and metabolic mechanisms of osteomalacia. These advances in biochemistry and metabolism studies have resulted in the introduction of vitamin D metabolites and analogs into the therapeutic armamentarium. These compounds have provided potent and effective agents to treat disorders that lead to inadequate mineralization of bone, including dietary [14] and metabolic disorders of bone, especially bone defects resulting from renal disease [15,16], but also genetic disorders of vitamin D metabolism [17,18]. The vitamin D compounds are also used to suppress parathyroid function to prevent Copyright © 2005, Elsevier, Inc. All rights reserved.

566 the bone loss associated with secondary hyperparathyroidism in renal disease [19,20], and also to prevent bone loss in osteoporosis [21–23]. Vitamin D deficiency results in rickets during development and undermineralized bone or osteomalacia in the adult. The disorders are manifested as an increase in osteoid, as assessed dynamically by measurement of tetracycline labelling [18,24–26]. Effects of vitamin D on rickets and osteomalacia are described in Chapter 63. The major determinant of this undermineralization is a decrease in the concentrations of calcium and phosphate in the serum [27], a consequence of decreased calcium and phosphate absorption from the intestine, the mechanisms of which are discussed in Chapter 24. Strong evidence that the effects of vitamin D on calcium and phosphate absorption are determinant of the effects on mineralization derive from studies of children with hereditary vitamin D–resistant rickets (HVDRR) (Chapter 72) and vitamin D receptor (VDR) knockout mice (Chapter 20). Children with HVDRR treated with intravenous calcium infusions daily for many months show healing of their rickets indicating that normalization of calcium metabolism is sufficient to restore bone and reverse rickets [18]. Similar findings exist in VDR knockout mice [1,28]. The VDR −/− mice exhibit normal growth and development prior to weaning, when they receive a high-calcium diet while nursing. However, once they are weaned, changes become evident [2]. Hypocalcemia and impaired mineralization and growth are seen. A diet high in calcium and lactose rescues the defective mineralization and growth [29]. Dietary phosphorus restriction with normal calcium (0.25% phosphorus, 0.5% calcium) also reversed the defects in growth and mineralization in VDR −/− mice, showing that the proper ratio of the minerals is critical for normal responses [30]. A comparison between two strains of mice, C57BL/6, which have low bone density and C3H/He, which have high bone density, showed that mucosal to serosal calcium transport by duodenal gut sacs was higher in the CeH/He mice [31]. There was no difference in the serum calcitriol levels or in renal 1-α-hydroxylase mRNA between the two strains [31]. Parathyroid hormone (PTH) was able to elicit anabolic effects in vitamin D–deficient rats, provided that the diet was modified to maintain normal calcium and phosphorus in serum [32]. Thus, multiple studies suggest that the indirect effects of vitamin D on intestinal calcium and phosphorus absorption are the critical, if not the sole, determinants of bone mineralization. There is evidence that at least during adolescence the demand for calcium and phosphorus can regulate calcitriol. A study in young females showed that during development there is a positive correlation between

PAULA H. STERN

serum calcitriol and bone mineral accretion in total body and forearm [33]. The effect was postulated to be mediated through an effect of insulin-like growth factor-I on calcitriol synthesis [33]. The mechanism of mineralization of bone matrix is still not fully understood and is discussed in Chapter 27. However, collagen or noncollagenous proteins appear to be involved in the initiation of mineralization in lamellar bone [34], whereas matrix vesicles, similar to those associated with cartilage calcification, may play a role in woven bone [35].

III. VITAMIN D AND BONE TARGET GENES In vivo and in vitro studies show direct effects of vitamin D and calcitriol on important genes and activities in bone. Since many of these responses are associated with the development of a mature osteoblast phenotype, they raise the question as to whether the effect of vitamin D to increase bone mineralization could also exhibit a direct component, possibly facilitating new bone formation, or whether these responses are associated with other effects of vitamin D on the skeleton. Osteoblast genes that respond to vitamin D in vivo include collagen, alkaline phosphatase, osteocalcin, osteopontin, bone sialoprotein, transforming growth factor-β (TGF-β), vascular endothelial growth factor (VEGF), matrix metalloproteinase-9 (MMP-9), β3 integrin, receptor activator of NFκB ligand (RANKL), and osteoprotegerin (OPG), suggesting potential involvement of vitamin D in a range of functional activities in bone. Calcitriol can increase mineralization in later stages of osteoblast cultures, suggesting that there could be a direct effect mediated through changes in gene expression [36]. In vitro studies show many additional target proteins of calcitriol in osteoblasts. Microarray analysis of ROS 18/2.8 cells has identified a number of extracellular matrix and attachment proteins as well as signaling proteins that are affected by either 24-hr treatment with calcitriol or short-term (3-hr) treatment with calcitriol or 25(OH)-16ene-23yne-D3, a calcium-mobilizing analog that did not activate the nuclear receptor [37]. Calcitriol modulates the Runx2/Cbfa1 transcription factor, a regulator of many of the other vitamin D–responsive genes [38,39]. Different effects of calcitriol on Runx2 have been reported, which may relate to the osteoblast model and the incubation time. In MC3T3 and ROS 17/2.8 osteoblasts, Runx2 was downregulated by 24-hr treatment with calcitriol [38], whereas in primary human osteoblasts, the early inhibition was followed by a stimulation after 48 hr [39].

567

CHAPTER 32 Bone

A

9

ng OSTEOCALCIN/well

Control

B

MG-63

PTH (10 nM)

6

FORSKOLIN (10 µM)

3

0

−10 −11 log [1,25-(OH)2D3] (M)

0

−9

280 Control

ng OSTEOCALCIN/well

These studies raise a recurring issue of differences in the responses to calcitriol in different models. The effects of calcitriol on osteoblast gene expression may be determined or regulated by the differentiation state of the cells. Calcitriol stimulates the proliferation of nonadherent rat marrow–derived osteoprogenitor cells [40]. However, calcitriol inhibits proliferation in rat osteoblast cultures and down-regulates collagen synthesis and alkaline phosphatase activity [41]. Other genes are activated, suggesting that calcitriol induces a more differentiated phenotype [42]. TGF-β may play a role in this antiproliferative effect [43]. Skeletal TGF-β is decreased in vitamin D–deficient rats [43]. In human fetal osteoblast cells, calcitriol increased the expression of type I and type II TGF-β receptors [44]. Genes for VEGF and for matrix metalloproteinase-9, which releases active VEGF from the cartilage matrix, are target genes for calcitriol [8,9,45]. The effects of vitamin D on VEGF and MMP-9 may relate functionally to the process of endochondral calcification. Endochondral bone development is impaired in vitamin D deficiency. There is disorganization of the primary spongiosa and elongation of the growth plates. Angiogenesis plays a critical role in endochondral bone development, allowing for introduction of osteoclast and osteoblast precursors [46]. The growth factor, vascular endothelial growth factor (VEGF), is involved in this process, as demonstrated by the impaired vascular invasion and ossification in mice expressing a dominant negative VEGF. Increases in osteocalcin are a well-established response to calcitriol both in vitro and in vivo [47–50]. However, even for this bone matrix protein, differences in response are observed in different models. Species differences have been described in responses of osteocalcin genes from mouse, compared to rat or human after acute treatment in vitro with calcitriol. Although rat and human osteocalcin gene expression increased in response to acute calcitriol exposure [48,49,51,52], mouse osteocalcin did not [52–54]. In contrast, both species showed increased osteopontin gene expression [52]. The species difference in osteocalcin gene response was not observed when the cells were treated with PTH or with chronic calcitriol, both of which increased osteocalcin in all three species. PTH and calcitriol had synergistic effects on osteocalcin production in the rat ROS 17/2.8 osteoblastic cell line, whereas PTH failed to enhance the response to calcitriol in the human MG-63 osteoblastic cell line [55] (Fig. 1). In a study on human bone tissue, the osteocalcin response, and VDR expression, were dependent upon the donor age and the skeletal origin of osteoblasts [56]. Both secreted osteocalcin and osteocalcin mRNA from trabecular bone samples were

240

PTH (10 nM)

200

FORSKOLIN (10 µM)

ROS 17/2.8

##

#

160 120 80 40 0 0

−10 log [1,25-(OH)2D3] (M)

−9

FIGURE 1 Difference in the modulation of the effect of 1,25(OH)2D3 on osteocalcin in human MG-63 osteoblastic cells and rat ROS 17/2.8 osteoblastic cells. Co-incubation with PTH or forskolin potentiated the response in the ROS 17/2.8 cells but not in the MG-63 cells. No effects on proliferation were observed in the cultures. *p < 0.02, **p < 0.002 vs no addition of 1,25-(OH)2D3, PTH, or forskolin; #p < 0.02, ##p < 0.002 for significant interaction between 1,25-(OH)2D3 and PTH or forskolin. From [55], with permission.

higher in younger donors, and in older donors both osteocalcin mRNA and VDR were higher in osteoblasts derived from the knee than in those derived from the hip [56]. Since the function of osteocalcin in bone formation is still unclear, definitive conclusions regarding a direct effect of calcitriol on bone formation through this protein are not yet obvious. Osteopontin synthesis is increased by calcitriol in osteosarcoma cells [57] and bone organ cultures [58], whereas bone sialoprotein is decrease by vitamin D in vitro [58]. In vitamin D–deficient rBSP 2.7 Luc transgenic mice, vitamin D deficiency resulted in decreases in osteopontin mRNA, whereas expression of bone sialoprotein was increased [59]. Thus vitamin D has

568 similar effects on these bone matrix proteins in vitro and in vivo. Osteopontin [60] and β3 integrin [61] are involved in attachment of osteoclasts to bone, and thus the effects on these genes may be related to actions on bone resorption. The role of the increase in bone sialoprotein in the vitamin D–deficient animals is less clear, as the protein has multiple roles, including the stimulation of hydroxyapaptite formation [62], stimulation of angiogenesis [63], and cell attachment [64]. Effects of vitamin D on osteocalcin and osteopontin and matrix Gla protein are discussed in more detail in Chapter 41. Calcitriol decreases procollagen mRNA and collagen synthesis in mature osteoblasts [65]. Although this response on a critical component of the bone matrix appears somewhat paradoxical, it is conceivable that the effect to decrease this early differentiated function could contribute to the recovery from osteomalacia when vitamin D is replaced. Vitamin D actions on collagen are discussed in Chapter 40. Thus the functional consequences of some effects of vitamin D on osteoblast genes are more apparent than others. Effects on RANKL, discussed in the next section, seem clearly related to the actions of vitamin D on bone resorption. In some cases, the roles of the target proteins may be modulatory, relating to differentiative responses that prime the osteoblasts to carry out further responses, or to coordinate the effects on formation and resorption. Although there is increasing knowledge of actions of vitamin D to modulate many genes that are important in normal matrix composition, in cell attachment, and in bone remodeling, a full understanding of the role and importance of these changes in the response of bone to vitamin D is still to be established.

IV. VITAMIN D ACTIONS ON BONE RESORPTION In contrast with the indirect effects on mineralization, the actions of vitamin D on bone resorption are mediated directly on bone. Although the osteoclast is the effector of bone resorption, the target cell of calcitriol action in bone is the osteoblast. Osteoblasts activate osteoclasts through the induction of the membrane-associated protein Receptor Activator of NFκB Ligand (RANKL). RANKL, which is identical to the molecules TRANCE and OPGL, is a member of the tumor necrosis factor-α family. Interaction of RANKL with its receptor RANK on osteoclast precursor cells leads to their differentiation into multinucleated cells capable of resorbing bone [66]. RANKL also promotes the activity and survival of osteoclasts [66].

PAULA H. STERN

RANKL induction is essential for calcitriol-stimulated resorption in vitro, as demonstrated by the finding that a neutralizing anti-RANKL polyclonal antibody inhibited the resorption stimulated by calcitriol, as well as that stimulated by PTH and prostaglandin E2 [67]. The VDR-RXRβ heterodimer is believed to bind to the –937/−922 region of the RANKL gene [68], although this finding is yet to be confirmed. Mutations in this putative vitamin D response element decreased the very modest inductive effect of calcitriol on the RANKL promoter [68]. Coculture studies are consistent with effects of vitamin D on resorption being mediated through increased expression of RANKL in osteoblasts. Combining VDR −/− calvarial osteoblasts with wild-type spleen cells resulted in the failure of osteoclastogenesis, whereas normal osteoclast formation occurred when normal osteoblasts were combined with VDR −/− spleen cells [69]. Effects of PTH or of interleukin-1 to produce tartrate-resistant acid phosphatase (TRAP)-positive multinucleated cells were unaffected when VDR −/− osteoblasts were used, indicating that the mechanisms of these other agents were independent of vitamin D [69]. These findings were consistent with the observation that osteoclasts in VDR −/− mice expressed TRAP and had normal bone resorption [69], since factors other than vitamin D elicited a redundant effect on the pathways. Effects on osteoclastogenesis are discussed in more detail in Chapter 38. The decoy receptor for RANKL, the soluble factor osteoprotegerin (OPG), prevents calcitriol-stimulated bone resorption based on the evidence that it antagonizes the effects of calcitriol on serum calcium, [70,71], osteoclast number [71], and serum cross-linked N-telopeptides [72]. OPG is also modulated by vitamin D analogs; however, divergent effects on OPG have been observed. Calcitriol increased OPG mRNA by 50% in normal trabecular osteoblast cells and increased OPG mRNA 90% and OPG protein 60% in a fetal osteoblast cell line [73]. In another study, the vitamin D analog 2-methylene-19-NOR-(20S)-1,25dihydroxyvitamin D3 (Chapter 87) and calcitriol inhibited OPG production by osteoblastic cells [74]. It has been unclear whether vitamin D-stimulated calcium release from bone is a physiological event, i.e., whether it occurs at physiological concentrations of vitamin D. In vitro studies in bone organ cultures demonstrate that calcitriol stimulates release of calcium from bone at a concentration as low as 10 pM [75] (Fig. 2), which would be at the low end of the physiological range. In vivo studies using tracer kinetics suggested that physiologic replacement doses of vitamin D caused mobilization of calcium from bone [76]. Subsequent studies using other approaches suggest

569

CHAPTER 32 Bone

40 60 80 0 10 8.32 [1,25-(OH)2D] 2.08

20 4.16

40 60 80 8.32

FIGURE 2 Dose-response curve of effects of 1,25-(OH)2D3 on bone resorption in vitro quantified as 45Ca release from prelabeled fetal rat radii and ulnae. A and B are logarithmic plots, respectively. From [75], with permission.

that vitamin D–induced hypercalcemia requires much larger amounts. This was demonstrated by experiments in which a range of calcitriol doses were given to rats maintained on a diet that was vitamin D deficient and low in calcium. These studies demonstrated that mobilization of calcium from bone, as assessed in vivo by serum calcium, required 10–50 times more calcitriol than intestinal absorption of calcium as measured in vitro by gut sac transport [77]. Also, in studies in which low-calcium diet increased expression of mRNA for RANKL and cathepsin K and thyroparathyroidectomy reduced the expression of RANKL and cathepsin K, vitamin D also increased RANKL and cathepsin K; however, a high supraphysiological dose was required [78]. Interestingly, low doses of vitamin D suppressed PTH-stimulated increases in RANKL mRNA and cathepsin K [78]. The antiresorptive effects of vitamin D in vivo are discussed in more detail later. The osteolytic response and hypercalcemic effects of vitamin D becomes increasingly apparent as higher concentrations or metabolites and analogs that bypass critical regulatory steps are used therapeutically. It has been estimated that continued ingestion of 50,000 units or more of vitamin D daily by a person with normal parathyroid function and sensitivity to vitamin D may result in vitamin D intoxication [79]. Impaired mineralization may interfere with resorption. In the rachitic state there are morphological changes in the osteoclasts, with altered ruffled borders and clear zones. The distribution of TRAP-positive cells is modified, probably affecting their ability to carry out normal functions on the matrix [80]. Osteopetrotic rats showed skeletal resistance to vitamin D [81].

p.o.

PTX

2.0

1.5 a

a 1.0

2

20

y1

0 pM ph/culture

da

20

y9

r2 = .986 25

da

30

y3

35

da

40

Although it is clear that vitamin D can stimulate resorption by direct effects on bone, it is also now apparent that vitamin D at physiological doses or vitamin D analogs can inhibit bone resorption in vivo. The effects of vitamin D to inhibit resorption are demonstrated by effects on serum calcium (Fig. 3), osteoclast number [82], and deoxypyridinoline excretion [83]. In ovariectomized mice given alfacalcidol, a prodrug that is metabolized to calcitriol, doses that produced normocalcemia decreased TRAP activity and the numbers of TRAP positive osteoclast precursors [84]. Micro-CT (computerized tomography) dramatically demonstrated the increase in bone mass, which was consistent with histomorphometric measurements of BV/TV, Tb.Th and Tb.Sp [84]. (Abbreviations are defined in the legend to Fig. 4.) Further evidence for the effect of vitamin D to decrease resorption derives from studies in which bone was examined from transgenic mice in which the vitamin D receptor was selectively expressed in mature cells of the osteoblastic lineage by expressing it in an osteocalcin-based vector [85].

y1

B

da

A

Blood Ca2+ (mmol/L)

% bone 45Ca released

45

FIGURE 3 Effect of 1,25-(OH)2D3 and the analog 22-oxo-1,25(OH)2D3 to lower blood ionized calcium in rats with established hypercalcemia PTHrP(1-34) was infused continuously starting on day 3. 1,25-(OH)2D3 (open circles), 22-oxo-1,25-(OH)2D3 (closed circles), or vehicle (x) was given orally. ap < 0.05 vs vehicle-treated group. bp < 0.01 vs. vehicle-treated group. From [82], with permission.

570

PAULA H. STERN

sham

OVX

60

B

E2

0.4

0.1

**

*

40

**

0.08

0.3 +

0.06

+

0.04

20

ALF

E2

**

0

0

sham

0.2

0.1

0.02

0

+

**

Tb.Th (mm)

BV/TV (%)

ALF

Tb.Sp (mm)

A

sham

OVX

ALF OVX

E2

sham

ALF

E2

OVX

FIGURE 4 Results from study showing that the 1,25-(OH)2D3 prodrug 1α(OH)D3 (ALF) increases bone mass in OVX mice. Eight-week-old OVX ddy mice were given 1α(OH)2D3 or estrogen five times a week for 1 month and fifth lumbar vertebrae analyzed by micro-CT. (A) 3D trabecular bone architecture; (B) bone morphometric parameters. *p < 0.05, **p < 0.01 vs. OVX-vehicle group; +p < 0.05 vs. sham group. From [84], with permission. Abbreviations: BV/TV, bone volume/tissue volume; Tb.Th, trabecular thickness; Tb.Sp, trabecular separation; these indices are described in Chapter 59.

These mice demonstrated a 30% reduction in the resorption surface in vertebrae, and an increase in the width and strength of cortical bone due to a doubling of periosteal mineral apposition rate. Multiple mechanisms may be involved in the inhibitory effects of vitamin D on bone resorption in vivo. Inhibition of PTH secretion is likely to be involved. Effects of vitamin D on the parathyroid glands are the topic of Chapter 30. Stimulation of OPG expression could modulate the effects on osteoclastogenesis. Another mechanism for the inhibition of osteoclastogenesis may be a decrease in the pool of osteoclast precursors [84]. Interactions of calcitriol with other hormones on osteoblasts could have modulating effects on bone. Calcitriol inhibited the production of parathyroid hormone related protein (PTHrP) in human primary osteoblast cells [86]. Vitamin D has effects on osteoclasts that could also enhance estrogen effects. In ROS 17/2.8 cells, calcitriol increased expression of estrogen receptors [87], although an opposite effect is found in some other cell systems. In primary osteoblasts, vitamin D increased CYP 19 aromatase [88]. Analogs with enhanced effects on one or more of these responses could produce beneficial effects on

bone without causing hypercalcemia. At the molecular level, mechanisms for differential effects of different analogs could include selectivity at vitamin D response elements, including different cofactor requirements of vitamin D responsive genes in different tissues. A number of chapters (11–18) address the molecular mechanisms of vitamin D action. Pharmacokinetic differences between different vitamin D compounds could also result in an altered balance of effects on different tissues, resulting in greater or lesser effects on bone resorption. Analogs and their different actions are the topic of an entire section (Section VIII, Chapters 80–88).

V. SUMMARY AND CONCLUSIONS Vitamin D has a number of effects on bone, some of which are clearly beneficial and others that could be deleterious to the structure and strength of bone. Insufficient vitamin D results in undermineralized weaker bone that cannot optimally carry out normal functions, and thus vitamin D is beneficial for mineralization. Recent studies support the earlier conclusion

571

CHAPTER 32 Bone

that effects of vitamin D on mineralization are indirect and a consequence of changes in serum calcium and phosphorus. Effects of vitamin D to cause calcium mobilization from bone would generally be considered disadvantageous. Resorptive actions of vitamin D are directly mediated by effects on osteoblasts. Recent studies have revealed that the in vivo effects are biphasic, with inhibition of resorption elicited by lower concentrations and activation of resorption at higher concentrations. In contrast, in vitro studies have failed to show inhibitory effects of vitamin D or calcitriol on resorption. Decreased production of PTH, or increased production of inhibitory proteins, such as OPG by nonbone cells, could play a role in the in vivo antiresorptive actions. A number of osteoblast proteins including matrix proteins are affected by vitamin D and metabolites. Similar responses are generally observed in in vivo and in vitro studies, although there are differences in effects that have been related to stage of osteoblast differentiation, species, and tissue site. Although some of the protein responses have clear functional consequences, other changes are still a matter of speculation and have not been fully integrated into the picture of vitamin D action. In part this is related to the fact that the actions of some of these proteins are still not understood. Thus, there are still many unanswered questions relating to effects of vitamin D on bone, and much need for their elucidation by future research.

9.

10. 11. 12. 13. 14. 15. 16. 17. 18.

19.

20.

References

21.

1. Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997 Targeted ablation of the vitamin D receptor: an animal model of vitamin D-dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94:9831–9835. 2. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Genet 16:391–396. 3. Herodotus, The Persian Wars, Book III. 4. Huldschinsky K 1919 Heilung von Rachitis durch kunstliche Hohensonne. Deut Med Wochenschr 45:712–713. 5. Mellanby E 1918 The part played by an “accessory factor” in the production of experimental rickets. J Physiol 52:liii–lvi. 6. McCollum EV, Simmonds N, Shipley PG, Park EA 1921 The production of rachitis and similar diseases in the rat by deficient diets. J Biol Chem 45:333–341. 7. Sherman HC, Pappenheimer AM 1921 A dietetic production of rickets in rats and its prevention by an inorganic salt. Proc Soc Exptl Biol Med 18:193–197. 8. Shipley PG, Park EA, McCollum EV, Simmons N 1921 Studies on experimental rickets. III. A pathological condition bearing fundamental resemblances to rickets of the human

22.

23. 24. 25. 26. 27. 28.

29.

being resulting from diets low in phosphorus and fat soluble A: The phosphate ion in its prevention. Bull Johns Hopkins Hosp 32:160–166. Shipley PG, Park EA, McCollum EV, Simmons N, Parsons HT 1921 Studies on experimental rickets. II. The effect of cod liver oil administered to rats with experimental rickets. J Biol Chem 45:343–348. Hess AF, Weinstock M 1924 Antirachitic properties imparted to inert fluids and to green vegetables by ultraviolet irradiation. J Biol Chem 62:301–313. Steenbock H, Black A 1924 Fat-soluble vitamins. XVII. The induction of growth-promoting and calcifying properties in a ration by exposure to ultraviolet light. J Biol Chem 61. Holick MF, DeLuca HF 1974 Vitamin D metabolism. Annu Rev Med 25:349–367. DeLuca HF, Zierold C 1998 Mechanisms and functions of vitamin D. Nutr Rev 56:S4–S10; discussion S54–S75. Klein GL, Coburn JW 1984 Metabolic bone disease associated with total parenteral nutrition. Adv Nutr Res 6:67–92. Adams JE 2002 Dialysis bone disease. Semin Dial 15:277–289. Yasuda M, Akiba T, Nihei H 2003 Multicenter clinical trial of 22-oxa-1,25-dihydroxyvitamin D3 for chronic dialysis patients. Am J Kidney Dis 41:S108–S111. Thomas MK, Demay MB 2000 Vitamin D deficiency and disorders of vitamin D metabolism. Endocrinol Metab Clin North Am 29:611–627, viii. Liberman UA 2002 Hereditary deficiencies in vitamin D action. In: Bilezekian JP, Raisz LG, Rodan GA. (eds) Vol. 2. Principles of Bone Biology, Academic Press, San Diego, pp. 1195–1208. Chesney RW, Moorthy AV, Eisman JA, Jax DK, Mazess RB, DeLuca HF 1978 Increased growth after long-term oral 1alpha, 25-vitamin D3 in childhood renal osteodystrophy. New Engl J Med 298:238–242. Andress DL, Norris KC, Coburn JW, Slatopolsky EA, Sherrard DJ 1989 Intravenous calcitriol in the treatment of refractory osteitis fibrosa of chronic renal failure. N Engl J Med 321:274–279. Rodriguez-Martinez MA, Garcia-Cohen EC 2002 Role of Ca2+ and vitamin D in the prevention and treatment of osteoporosis. Pharmacol Ther 93:37–49. Chapuy MC, Arlot ME, Duboeuf F, Brun J, Crouzet B, Arnaud S, Delmas PD, Meunier PJ 1992 Vitamin D3 and calcium to prevent hip fractures in the elderly women. N Engl J Med 327:1637–1642. Fujita T 1992 Vitamin D in the treatment of osteoporosis. Proc Soc Exptl Biol Med 199:394–399. Teitelbaum SL, Bullough PG 1979 The pathophysiology of bone and joint disease. Am J Pathol 96:282–354. Parfitt AM 1983 The physiologic and clinical significance of bone histomorphometric data. In: Recker R (ed) Bone Histomorphometric Data. CRC Press, Boca Raton, FL, p. 143. Recker R, Barger-Lux, HJ, MJ 2002 Transilial bone biopsy. In: Bilezekian JP, Raisz LG, Rodan GA. (eds) Principles of Bone Biology, Vol. 2. Academic Press, San Diego, pp. 1625–1634. Howland J, Kramer B 1922 Factors concerned in the calcification of bone. Trans Am Pediat Soc 34:204–208. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Genet 16:391–396. Amling M, Priemel M, Holzmann T, Chapin K, Rueger JM, Baron R, Demay MB 1999 Rescue of the skeletal phenotype of

572

30.

31.

32.

33. 34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

vitamin D receptor-ablated mice in the setting of normal mineral ion homeostasis: formal histomorphometric and biomechanical analyses. Endocrinology 140:4982–4987. Masuyama R, Nakaya Y, Tanaka S, Tsurukami H, Nakamura T, Watanabe S, Yoshizawa T, Kato S, Suzuki K 2001 Dietary phosphorus restriction reverses the impaired bone mineralization in vitamin D receptor knockout mice. Endocrinology 142:494–497. Armbrecht HJ, Boltz MA, Hodam TL 2002 Differences in intestinal calcium and phosphate transport between low and high bone density mice. Am J Physiol Gastrointest Liver Physiol 282:G130–G136. Toromanoff A, Ammann P, Mosekilde L, Thomsen JS, Riond JL 1997 Parathyroid hormone increases bone formation and improves mineral balance in vitamin D–deficient female rats. Endocrinology 138:2449–2457. Ilich JZ, Badenhop NE, Jelic T, Clairmont AC, Nagode LA, Matkovic V 1997 Calcitriol and bone mass accumulation in females during puberty. Calcif Tissue Int 61:104–107. Landis WJ, Song MJ, Leith A, McEwen L, McEwen BF 1993 Mineral and organic matrix interaction in normally calcifying tendon visualized in three dimensions by high-voltage electron microscopic tomography and graphic image reconstruction. J Struct Biol 110:39–54. Marks JS, Odgren PR 2002 Structure and development of the skeleton. In: Bilezekian JP, Raisz LG, Rodan GA. (eds) Principles of Bone Biology, Vol. 1. Academic Press, San Diego, pp. 3–15. Matsumoto T, Igarashi C, Takeuchi Y, Harada S, Kikuchi T, Yamato H, Ogata E 1991 Stimulation by 1,25-dihydroxyvitamin D3 of in vitro mineralization induced by osteoblast-like MC3T3-E1 cells. Bone 12:27–32. Farach-Carson MC, Xu Y 2002 Microarray detection of gene expression changes induced by 1,25(OH)2D3 and a Ca2+ influxactivating analog in osteoblastic ROS 17/2.8 cells. Steroids 67:467–470. Drissi H, Pouliot A, Koolloos C, Stein JL, Lian JB, Stein GS van Wijnen AJ 2002 1,25-(OH)2-vitamin D3 suppresses the bone-related Runx2/Cbfa1 gene promoter. Exp Cell Res 274:323–333. Viereck V, Siggelkow H, Tauber S, Raddatz D, Schutze N, Hufner M 2002 Differential regulation of Cbfa1/Runx2 and osteocalcin gene expression by vitamin-D3, dexamethasone, and local growth factors in primary human osteoblasts. J Cell Biochem 86:348–356. Rickard DJ, Kazhdan I, Leboy PS 1995 Importance of 1,25dihydroxyvitamin D3 and the nonadherent cells of marrow for osteoblast differentiation from rat marrow stromal cells. Bone 16:671–678. Ishida H, Bellows CG, Aubin JE, Heersche JN 1993 Characterization of the 1,25-(OH)2D3-induced inhibition of bone nodule formation in long-term cultures of fetal rat calvaria cells. Endocrinology 132:61–66. Gurlek A, Kumar R 2001 Regulation of osteoblast growth by interactions between transforming growth factor-beta and 1α,25-dihydroxyvitamin D3. Crit Rev Eukaryot Gene Expr 11:299–317. Finkelman RD, Linkhart TA, Mohan S, Lau KH, Baylink DJ, Bell NH 1991 Vitamin D deficiency causes a selective reduction in deposition of transforming growth factor beta in rat bone: possible mechanism for imparied osteoinduction. Proc Natl Acad Sci 88:3657–3660. Wu Y, Haugen JD, Zinsmeister AR, Kumar R 1997 1α,25Dihydroxyvitamin D3 increases transforming growth factor and transforming growth factor receptor type I and II synthesis

PAULA H. STERN

45.

46.

47.

48. 49.

50.

51.

52. 53.

54.

55.

56.

57.

58.

in human bone cells. Biochem Biophys Res Commun 239: 734–739. Lin R, Amizuka N, Sasaki T, Aarts MM, Ozawa H, Goltzman D, Henderson JE, White JH 2002 1α,25-Dihydroxyvitamin D3 promotes vascularization of the chondro-osseous junction by stimulating expression of vascular endothelial growth factor and matrix metalloproteinase 9. J Bone Miner Res 17: 1604–1612. Poole R 1991 The growth plate: cellular physiology, cartilage assembly and mineralization. In: Hall BK, Newman SA. (eds) Cartilage: Molecular Aspects. CRC Press, Boca Raton, FL, pp. 179–211. Owen TA, Aronow MS, Barone LM, Bettencourt B, Stein GS, Lian JB 1991 Pleiotropic effects of vitamin D on osteoblast gene expression are related to the proliferative and differentiated state of the bone cell phenotype: dependency upon basal levels of gene expression, duration of exposure, and bone matrix competency in normal rat osteoblast cultures. Endocrinology 128:1496–1504. Price PA, Baukol SA 1981 1,25-Dihydroxyvitamin D3 increases serum levels of the vitamin K–dependent bone protein. Biochem Biophys Res Commun 99:928–935. Duda RJ Jr., Kumar R, Nelson KI, Zinsmeister AR, Mann KG, Riggs BL 1987 1,25-Dihydroxyvitamin D stimulation test for osteoblast function in normal and osteoporotic postmenopausal women. J Clin Invest 79:1249–1253. Beresford JN, Gallagher JA, Poser JW, Russell RG 1984 Production of osteocalcin by human bone cells in vitro. Effects of 1,25(OH)2D3, 24,25(OH)2D3, parathyroid hormone, and glucocorticoids. Metabol Bone Dis Rel Res 5:229–234. Yoon KG, Rutledge SJ, Buenaga RF, Rodan GA 1988 Characterization of the rat osteocalcin gene: stimulation of promoter activity by 1,25-dihydroxyvitamin D3. Biochemistry 27:8521–8526. Thomas GP, Bourne A, Eisman JA, Gardiner EM 2000 Species-divergent regulation of human and mouse osteocalcin genes by calciotropic hormones. Exp Cell Res 258:395–402. Clemens TL, Tang H, Maeda S, Kesterson RA, Demayo F, Pike JW, Gundberg CM 1997 Analysis of osteocalcin expression in transgenic mice reveals a species difference in vitamin D regulation of mouse and human osteocalcin genes. J Bone Miner Res 12:1570–1576. Lian JB, Shalhoub V, Aslam F, Frenkel B, Green J, Hamrah M, Stein GS, Stein JL 1997 Species-specific glucocorticoid and 1,25-dihydroxyvitamin D responsiveness in mouse MC3T3-E1 osteoblasts: Dexamethasone inhibits osteoblast differentiation and vitamin D down-regulates osteocalcin gene expression. Endocrinology 138:2117–2127. van Leeuwen JPTM, Birkenhager JC, van den Bemd GCM, Pols HAP 1996 Evidence for coordinated regulation of osteoblast function by 1,25-dihydroxyvitamin D3 and parathryoid hormone. Biochim Biophys Acta 1312:55–62. Martinez P, Moreno I, De Miguel F, Vila V, Esbrit P, Martinez ME 2001 Changes in osteocalcin response to 1,25-dihydroxyvitamin D3 stimulation and basal vitamin D receptor expression in human osteoblastic cells according to donor age and skeletal origin. Bone 29:35–41. Prince CW, Butler WT 1987 1,25-Dihydroxyvitamin D regulates the biosynthesis of osteopontin, a bone-derived cell attachment protein, in clonal osteoblast-like osteosarcoma cells. Coll Relat Res 7:305–313. Chen JT, Sodek J 1996 Regulation of bone sialoprotein and osteopontin mRNA expression by dexamethasone and 1,25-dihydroxyvitamin D3 in rat bone cultures. Connec Tiss Res 34:41–51.

573

CHAPTER 32 Bone

59. Chen JJ, Jin H, Ranly DM, Sodek J, Boyan BD 1999 Altered expression of bone sialoproteins in vitamin D-deficient rBSP2.7Luc transgenic mice. J Bone Miner Res 14:221–229. 60. Reinholt FP, Hultenby K, Oldberg A, Heinegard D 1990 Osteopontin—a possible anchor of osteoclasts to bone. Proc Natl Acad Sci USA 87:4473–4475. 61. Horton MA, Taylor ML, Arnett TR, Helfrich MH 1991 Arg-Gly-Asp (RGD) peptides and the anti-vitronectin receptor antibody 23C6 inhibit dentine resorption and cell spreading by osteoclasts. Exp Cell Res 195:368–375. 62. Reginato AM, Shapiro IM, Lash JW, Jimenez SA 1988 Type X collagen alterations in rachitic chick epiphyseal growth cartilage. J Biol Chem 263:9938–9945. 63. Bellahcene A, Bonjean K, Fohr B, Fedarko NS, Robey FA, Young MF, Fisher LW, Castronovo V 2000 Bone sialoprotein mediates human endothelial cell attachment and migration and promotes angiogenesis. Circ Res 86:885–891. 64. Horton MA, Nesbitt SA, Bennett JH, Stenbeck G 2002 Integrins and other cell surface attachment molecules of bone cells. In: Bilezekian JP, Raisz LG, Rodan GA. (eds) Principles of Bone Biology, Vol. 1. Academic Press, San Diego, pp. 265–286. 65. Rowe DW, Kream BE 1982 Regulation of collagen synthesis in fetal rat calvaria by 1,25-dihyroxyvitamin D3. J Biol Chem 257:8009–8015. 66. Takahashi N, Udagawa N, Takami M, Suda T 2002 Cells of bone: osteoclast generation. In Bilezekian JP, Raisz LG, Rodan GA. (eds) Principles of Bone Biology, Vol. 1. Academic Press, San Diego, pp. 109–126. 67. Tsukii K, Shima N, Mochizuki S, Yamaguchi K, Kinosaki M, Yano K, Shibata O, Udagawa N, Yasuda H, Suda T, Higashio K 1998 Osteoclast differentiation factor mediates an essential signal for bone resorption induced by 1α,25-dihydroxyvitamin D3, prostaglandin E2, or parathyroid hormone in the microenvironment of bone. Biochem Biophys Res Commun 246: 337–341. 68. Kitazawa R, Kitazawa S 2002 Vitamin D3 augments osteoclastogenesis via vitamin D–responsive element of mouse RANKL gene promoter. Biochem Biophys Res Commun 290:650–655. 69. Takeda S, Yoshizawa T, Nagai Y, Yamato H, Fukumoto S, Sekine K, Kato S, Matsumoto T Fujita T 1999 Stimulation of osteoclast formation by 1,25-dihydroxyvitamin D requires its binding to vitamin D receptor (VDR) in osteoblastic cells: Studies using VDR knockout mice. Endocrinology 140: 1005–1008. 70. Yamamoto M, Murakami T, Nishikawa M, Tsuda E, Mochizuki S, Higashio K, Akatsu T, Motoyoshi K, Nagata N 1998 Hypocalcemic effect of osteoclastogenesis inhibitory factor/osteoprotegerin in the thyroparathyroidectomized rat. Endocrinology 139:4012–4015. 71. Morony S, Capparelli C, Lee R, Shimamoto G, Boone T, Lacey DL, Dunstan CR 1999 A chimeric form of osteoprotegerin inhibits hypercalcemia and bone resorption induced by IL-1beta TNF-alpha PTH, PTHrP, and 1,25(OH)2D3. J Bone Miner Res 14:1478–1485. 72. Price PA, June HH, Buckley JR, Williamson MK 2001 Osteoprotegerin inhibits artery calcification induced by warfarin and by vitamin D. Arterioscler Thromb Vasc Biol 21:1610–1616. 73. Hofbauer LC, Dunstan CR, Spelsberg TC, Riggs BL, Khosla S 1998 Osteoprotegerin production by human osteoblast lineage cells is stimulated by vitamin D, bone morphogenetic

74.

75.

76. 77. 78. 79. 80.

81. 82.

83.

84.

85.

86.

87.

88.

protein-2, and cytokines. Biochem Biophys Res Commun 250:776–781. Yamamoto H, Shevde NK, Warrier AC, Plum LA, DeLuca HF, Pike JW 2003 2-Methylene-19-nor-(20S)-1,25-dihydroxyvitamin D3 potently stimulates gene-specific DNA binding of the vitamin D receptor in osteoblasts. J Biol Chem, Epub Jun 9. Stern PH, Phillips TE, Mavreas T 1980 Bioassay of 1,25-dihydroxyvitamin D in human plasma purified by partition, alkaline extraction and high-pressure chromatography. Anal Biochem 102:22–30. Carlsson A 1952 Tracer experiments on the effect of vitamin D on the skeletal metabolism of calcium and phosphorus. Acta Physiol Scand 26:212–220. Suda T, DeLuca HF, Tanaka Y 1970 Biological activity of 25-hydroxyergocalciferol in rats. J Nutr 100:1049–1052. Suda T, Ueno Y, Fujii K, Shinki T 2003 Vitamin D and bone. J Cell Biochem 88:259–266. Marcus R. In: The Pharmacological Basis of Therapeutics. McGraw-Hill 2001 Agents affecting calcification and bone turnover 10th Ed. New York, pp. 1715–1743. Nordahl J, Hollberg K, Mengarelli-Widholm S, Andersson G, Reinholt FP 2000 Morphological and functional features of clasts in low phosphate, vitamin D–deficiency rickets. Calcif Tissue Int 67:400–407. Safadi FF, Hermey DC, Popoff SN, Seifert MF 1999 Skeletal resistance to 1,25-dihydroxyvitamin D3 in osteopetrotic rats. Endocrine 11:309–319. Endo K, Katsumata K, Hirata M, Masaki T, Kubodera N, Nakamura T, Ikeda K, Ogata E 2000 1,25-Dihydroxyvitamin D3 as well as its analogue OCT lower blood calcium through inhibition of bone resorption in hypercalcemic rats with continuous parathyroid hormone-related peptide infusion. J Bone Miner Res 15:175–181. Shiraishi A, Higashi S, Ohkawa H, Kubodera N, Hirasawa T, Ezawa I, Ikeda K, Ogata E 1999 The advantage of alfacalcidol over vitamin D in the treatment of osteoporosis. Calcif Tissue Int 65:311–316. Shibata T, Shira-Ishi A, Sato T, Masaki T, Masuda A, Hishiya A, Ishikura N, Higashi S, Uchida Y, Saito MO, Ito M, Ogata E, Watanabe K, Ikeda K 2002 Vitamin D hormone inhibits osteoclastogenesis in vivo by decreasing the pool of osteoclast precursors in bone marrow. J Bone Miner Res 17:622–629. Gardiner EM, Baldock PA, Thomas GP, Sims NA, Henderson NK, Hollis B, White CP, Sunn KL, Morrison NA, Walsh WR, Eisman JA 2000 Increased formation and decreased resorption of bone in mice with elevated vitamin D receptor in mature cells of the osteoblastic lineage. FASEB J 14:1908–1916. Gomez-Garcia L, Esbrit P, Carreno L, Sabando P, Garcia-Flores M, Martinez ME 2003 Alendronate interacts with the inhibitory effect of 1,25(OH)2D3 on parathyroid hormone-related protein expression in human osteoblastic cells. J Bone Miner Res 18:78–87. Ohsawa M, Mizunuma H, Kagami I, Miyamoto S, Kanuma T, Ibuki Y 2000 Effect of 1,25(OH)2D3 on expression of estrogen receptor-alpha mRNA on rat osteosarcoma cell line (ROS 17/2.8). Life Sci 66:2465–2472. Enjuanes A, Garcia-Giralt N, Supervia A, Nogues X, Mellibovsky L, Carbonell J, Grinberg D, Balcells S,Diez-Perez A 2003 Regulation of CYP19 gene expression in primary human osteoblasts: effects of vitamin D and other treatments. Eur J Endocrinol 148:519–526.

CHAPTER 33

Cartilage and Vitamin D: Genomic and Nongenomic Regulation by 1,25(OH)2D3 and 24,25(OH)2D3 BARBARA D. BOYAN ZVI SCHWARTZ

Department of Biomedical Engineering at Georgia Tech and Emory Uuniversity, Georgia Institute of Technology, Atlanta, Georgia Department of Periodontics, Hebrew University Hadassah Faculty of Dental Medicine, Jerusalem, Israel

I. Chondrogenesis and Endochondral Ossification In Vivo II. Separate Roles for 24,25(OH)2D3 and 1,25(OH)2D3 in Cartilage III. Rapid Actions of Vitamin D and Nongenomic Mechanisms

IV. Physiologic Relevance of Nongenomic Regulation of Matrix Vesicles V. Summary References

I. CHONDROGENESIS AND ENDOCHONDRAL OSSIFICATION IN VIVO

the joints, the cells form what will eventually be articular cartilage. Within the developing embryo, there are many variations on this theme. For example, the mandibular condyle develops a secondary cartilage with a layer of multipotent cells at the outer surface of the cartilage. In postfetal life, these cells can respond to physical forces on the cartilage, enabling considerable modeling to occur. Other cartilages develop with a fibroblastic component, resulting in a broad range of structural properties. Most of the cells in hyaline cartilages, such as the articular cartilages at the ends of long bones, are in a relatively stable state of maturation. However, the cells continue to progress through the lineage, although at a very slow rate. For example, at the interface between articular cartilage and the adjacent subchondral bone, there is a region of cartilage similar to the growth plate in which chondrocytes undergo hypertrophy and mineralize their extracellular matrix. Calcification of the cartilage in this region, termed the “tidemark,” continues throughout the life of the individual, or until the articular cartilage is lost from the joint surface [1]. At the other end of the spectrum is fracture callus and induced bone formation, in which cells traverse the entire lineage relatively rapidly [2–8]. The transition from mesenchymal cells to chondrocytes has been observed in vitro following treatment of neonatal rat muscle cells [9] or chick limb bud cells [10] with bone morphogenetic protein (BMP).

A. The Chondrocyte Lineage Cartilage is a group of tissues produced by chondrocytes that are characterized by a relative lack of vascularity. Cartilage extracellular matrix consists of predominantly type II collagen and proteoglycan, often in the form of proteoglycan aggregate, and the glycosaminoglycan side chains on the proteoglycan core protein are highly sulfated in the mature tissue. In the embryo, cartilage forms from mesoderm and provides the structural framework for a number of other tissues, including bone. Mesenchymal stem cells differentiate into chondroblasts, which produce and maintain the proteoglycan-rich type II collagen matrix. Eventually, the chondrocytes undergo hypertrophy and mineralize their extracellular matrix. Once this has occurred, the tissue is resorbed by osteoclasts, accompanied by vascular invasion of the mineralized regions. Osteoprogenitor cells migrate to the calcified cartilage scaffold and form bone. Ultimately the cartilage within the newly forming bone also resorbs, leaving a marrow cavity. In addition, cartilage at the ends of the bones further differentiates along two divergent pathways. At the juncture of the newly formed bone and the terminal cartilage, the cells develop what will eventually become the growth plate. In the areas that will become VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

576

BARBARA D. BOYAN AND ZVI SCHWARTZ

The growth plate, itself, represents a specialized situation in which the terminal differentiation of the chondrocyte occurs in a linear array so that chondrocytes appear as columns of cells traversing the lineage in clearly demarcated zones of maturation [11,12] (Fig. 1). At one end of the growth plate is the restingzone, also called the reserve zone, in which the cells exhibit a hyaline cartilage-like phenotype. The length of time a cell may remain in the resting-zone is unknown. Regulatory signals stimulate the cells to undergo a proliferative burst, after which they remain in a prehypertrophic state for varying times. Studies using knockout mice for a number of proteins believed to play a role in bone formation have shown that the lesions are frequently expressed most notably in cells of the prehypertrophic zone of cartilage [13,14]. These postproliferative cells are undergoing a major shift in phenotype. Not only do they begin to change shape,

MV

Cell 24,25 LTGFβ

but they also modify their extracellular matrix. It is during the prehypertrophic maturation state that they synthesize the proteins that will make this possible. Once hypertrophy begins, the chondrocytes appear to be in lockstep for their final differentiation along the endochondral lineage [15–19]. Hypertrophic chondrocytes increase markedly in size and must make major changes in the composition of their extracellular matrix to accommodate this [20]. In addition, they must prepare their matrix for mineral deposition. It is perhaps in this aspect of their maturation that the greatest amount of controversy has arisen. As the cells reach the lowest regions of the growth plate, they cease to make cartilage-specific proteins, such as proteoglycan aggregate and type II collagen [18,21–24]. They synthesize type X collagen at this time [25–30]. However, they also begin to make proteins normally associated with bone and tooth mineralization such as osteocalcin,

Growth plate

Zone

Reserve 24,25 BMP2 1,25 BMP4 IGF-1 EGF 24,25 BMP2 BMP4 α2HS IGF-1 bFGF

Proliferation

Prehypertrophic

1,25

TGFβ

TGFβ

24,25 LTGFβ BMP2+4 α2HS PTHrP 1,25 LTGFβ BMP2+4 PTHrP

1,25 BMP2 BMP4 PTHrP

Hypertrophic Calcification

HGF 1,25 PTHrP BMP2+4 IL-1

Calcified cartilage Vascular invasion Bone

Factors in region

FIGURE 1 Schematic drawing of the growth plate showing the local factors and hormones that regulate the chondrocytes at various points in their differentiation. Cells synthesize and secrete a number of cytokines and growth factors, as well as 1,25(OH)2D3 (1,25) and 24,25(OH)2D3 (24,25), as they mature in the endochondral lineage. Although some of these factors have been demonstrated in vivo by immunohistochemistry, others have been inferred from cell culture data in the literature. This list is not intended to be complete, but rather emphasizes the important role of local factors, including vitamin D metabolites, in conveying messages between cells both up and down the growth plate, as well as between cells and their extracellular matrix vesicles (MVs). Our own studies have shown that resting zone cells synthesize latent transforming growth factor-beta-1 (TGF-β1) in vivo and in vitro and incorporate it into their extracellular matrix. MVs isolated from growth-zone chondrocyte cultures (prehypertrophic and hypertrophic zones) can activate TGF-β1 when they are incubated with 1,25(OH)2D3 but not 24,25(OH)2D3. In contrast, MVs produced by resting zone cells do not activate TGF-β1 when treated with either metabolite. BMP, Bone morphogenetic protein; IGF-1, insulin-like growth factor-1; α2HS, α2HSglycoprotein; bFGF, basic fibroblast growth factor; PTHrP, parathyroid hormone-related peptide; IL-1, interleukin-1; HGF, hepatocyte growth factor.

CHAPTER 33 Genomic and Nongenomic Regulation by 1,25(OH)2D3 and 24,25(OH)2D3

osteopontin, and type I collagen [31–33]. It is important to note that these cells are not osteoblasts, just as odontoblasts are not osteoblasts. They are chondrocytes that are mineralizing their matrix, and, to do so, they must make a matrix that can support hydroxyapatite formation. Many of the chondrocytes undergoing hypertrophy are also apoptotic [34]. There have been reports that sister cells may survive the apoptotic process and may go on to become osteoblasts [34,35]. Other investigators have reported that, in the mandibular condyle, some of the cells “trans-differentiate” into osteoblasts. Whether either is the case is difficult to prove, as multipotent cells are present and may undergo differentiation into osteoblasts at this time. This is particularly true in cell culture. Depending on culture conditions, cells may adapt or redifferentiate. Thus, chondrocytes may make type I collagen in culture as an adaptation to the presence of serum or to the process of attachment and spreading. In spite of this, these cells will retain many of the features of their in vivo chondrocyte phenotype. Although there is considerable overlap in the articular cartilage and growth plate chondrocyte phenotypes, the cells do exhibit differences, depending on their geographical location within the tissue and the presence of pathology such as osteoarthritis. Unfortunately, the nature of cell culture has led to a confusing array of findings concerning chondrocyte cell biology. In vivo, articular chondrocytes and other hyaline cartilage cells exist individually or as isolated clones within the extracellular matrix. These cells tend to have a rounded morphology that can be retained in culture through growth in agarose or on type I collagen gels [36–39]. Neither of these materials is physiologically identical to the type II collagen and proteoglycan matrix the cells experience in vivo and, consequently, may impose artifacts on the model. In contrast, growth plate chondrocytes tend to exist in columns and are much more closely packed than the articular cartilage cells. Their phenotype in vivo is distinctly different. They produce a different matrix, they produce extracellular matrix vesicles in much greater numbers, and they are actively undergoing extensive shape changes. Most investigators do not grow these cells in threedimensional cultures [40–48], so direct comparisons with studies on articular chondrocytes are often difficult. In this chapter, we focus on growth plate chondrocytes since this tissue has been most extensively studied with respect to vitamin D. In general, the discussion refers to growth plates typical of postfetal development. Two animal models, chickens and rats, are most often used. Major differences exist between them,

577

and neither model is a precise analog for human endochondral ossification. Most in vivo studies have used the tibial growth plate, although by no means exclusively. The tibia is an example of a primary growth plate. In contrast, the mandibular condyle is a secondary growth plate and, in humans, remains open even in adults. Thus, direct comparisons with the tibial growth plate may not be appropriate. The increased use of cell culture has led investigators to develop the costochondral cartilage of rabbits and rats as a growth plate model. Because these tissues are not exposed to the compressive forces experienced by tibial growth plate, there may be biochemical differences that could affect the generalization of results obtained with these models. A number of studies have used chicken tissue as the model system, and the chick limb bud is a frequently used culture system [49,50]. At the time it is placed in culture, the limb bud contains multipotent mesenchymal cells together with committed osteochondroprogenitor cells. These cells continue along their lineage cascade in culture. In postfetal life, chick tibial growth plate is used because of the extreme sensitivity of chickens to vitamin D [51]. Unfortunately, the complex interdigitation of bone and hypertrophic cartilage can make interpretation of results more difficult; nonetheless, this has been an extremely productive model. The rachitic rat has also been of considerable value to investigators [20,52–54]. To maintain the rat in a rachitic state it is necessary to make it hypophosphatemic as well as vitamin D deficient. In addition, the rat growth plate remains open throughout life, making direct comparisons with the human growth plate more complex. Even so, the rat costochondral cartilage has been an exceptionally valuable model for understanding the role of maturational state on cellular response to vitamin D [47,48].

B. Changes Occurring During Cartilage Maturation The amount of matrix relative to the cell is quite large, and once the cell has synthesized it, management of the matrix must occur by predetermined mechanisms. As the growth plate matures, the relative composition of its extracellular matrix changes, in some cases dramatically [11,12]. How the chondrocyte accomplishes this is of considerable interest. As shown in Fig. 1, growth plate chondrocytes produce factors that may act on the cell in an autocrine manner, on neighboring cells both above and below via paracrine mechanisms, and directly on matrix vesicles in the

578 extracellular matrix. 24,25(OH)2D3 and 1,25(OH)2D3 play important roles in this process. In the resting zone, the matrix produced by the cells is predominantly large proteoglycan aggregate and type II collagen. These cells produce extracellular matrix vesicles in small numbers and incorporate them into the matrix as it is synthesized. The matrix vesicles are enriched in alkaline phosphatase specific activity in comparison with the plasma membrane of the cells [47]. In addition, they contain matrix metalloproteinases capable of metabolizing proteoglycan. As described in greater detail in Section IV of this chapter, the activity of the matrix vesicles is regulated by the cells through signaling molecules secreted by them. Thus, the chondrocyte can modulate the expanse of extracellular matrix in the resting zone though activation of matrix processing enzymes found in the matrix vesicles [55]. In addition, several of the matrix metalloproteinases are stored in the matrix in latent form. These are activated via a cascade of enzyme actions, but the regulatory mechanisms involved are not well understood [56–58]. The nature of the signals that induce resting zone cells to enter into proliferation is also not understood. Controversy still exists as to whether the supply of cells in the proliferating cell zone comes from the resting zone or from multipotent cells that migrate into the tissue. Once proliferation is instituted, the cells undergo a set number of divisions and then enter a prehypertrophic state. The columns of cells that are the hallmark of growth plate anatomy are the result of the directional forces applied to the tissue. When growth plate cartilage is cultured under gravity-free conditions, as in space flight, the architecture of the tissue loses some of its distinctive orientation [59]. Studies using transgenic and knockout mice have showcased the importance of the prehypertrophic state [13,14]. During this phase of growth plate maturation, the cells begin to produce large numbers of matrix vesicles. The levels of alkaline phosphatase specific activity are higher than seen in resting zone matrix vesicles [47], as is the amount of matrix metalloproteinases [55]. The extracellular matrix is predominantly type II collagen and proteoglycan aggregate, but the cells also begin to produce type X collagen [25–27,30]. These cells also synthesize transforming growth factor-β (TGF-β) [60], BMP-2 [61], cartilage-derived growth factors [62], and parathyroid hormone-related peptide (PTHrP) [63]. Thus, disruption of the regulation of this tissue has profound consequences for the ability of the growth plate chondrocyte to undergo hypertrophy. As the cells begin to hypertrophy, the extracellular matrix undergoes extensive and rapid remodeling. Calcium transport from the cells to the matrix is increased, and ultimately hydroxyapatite crystal

BARBARA D. BOYAN AND ZVI SCHWARTZ

formation occurs along the inner leaflet of the matrix vesicle membrane. These 20- to 50-nm-diameter extracellular organelles first appear granular and then swell, and finally, as the crystal grows along the membrane surface, membrane integrity is lost [64]. At the transition from the lower hypertrophic zone to calcified cartilage, the crystals in the erupted matrix vesicles serve as foci for crystal multiplication and growth throughout the matrix. It is important to note that matrix vesicles are in intimate contact with the collagen fibrils [65], such that the mineral deposition occurring in association with the matrix vesicle membranes can contribute to the collagen-associated mineralization also occurring at the base of the hypertrophic cartilage. The crystal formation within the matrix vesicle is accompanied by a series of biochemical changes. Throughout the hypertrophic cartilage, alkaline phosphatase specific activity continues to increase [66]. At the same time phospholipase A2 activity is also increasing [67]. There is a change in the lipid composition of the matrix as well. The amount of calciumphospholipid-phosphate complexes (CPLX) increases [68], as does the content of phosphatidylserine [69], the major phospholipid present in CPLX [70]. The content of lysophospholipids also increases [54], presumably due to the action of phospholipase A2. The relative content of proteolipids, matrix vesicle membrane proteins associated with initial mineral formation, increases [71], indicating that as the nonnucleational lipids are metabolized, those lipids incorporated into the mineral phase of the tissue remain. Once calcification begins, the increases in alkaline phosphatase and phospholipase A2 specific activities and CPLX content are no longer seen [72]. Matrix proteins are also modulated during hypertrophy. To allow for the marked increase in chondrocyte size, there is an extensive retailoring of the matrix. Proteoglycanases degrade the proteoglycan aggregate [24,73] and chondroitinases remove the glycosaminoglycan side chains [23,74,75]. In addition, the cells cease synthesizing type II collagen and begin to secrete type X collagen into the matrix [28,29,76,77]. At the base of the hypertrophic zone, the chondrocytes synthesize and secrete type I collagen, osteonectin, and osteopontin [31–33]. Serum factors are also found in the matrix of the calcified cartilage. Evidence indicates that the chondrocytes may synthesize and secrete some of the proteins themselves. For example, growth plate chondrocytes have been shown to produce α2-HS-glycoprotein [78], hypothesized to be a chemoattractant for osteoclast progenitor cells [79]. The chondrocytes also make hepatocyte growth factor, which promotes vascular invasion [80].

CHAPTER 33 Genomic and Nongenomic Regulation by 1,25(OH)2D3 and 24,25(OH)2D3

C. Regulation by Vitamin D In Vivo Vitamin D has long been associated with cartilage metabolism. The importance of cartilage as a target tissue cannot be underestimated. When vitamin D– replete rats are treated with radiolabeled 25(OH)D3, radiolabeled 1,25(OH)2D3 and 24,25(OH)2D3 accumulate in this tissue. Although circulating levels of these metabolites are in the picomolar range, levels in the growth plate are in the nanomolar range [81], and 1,25(OH)2D3 and 24,25(OH)2D3 accumulate in fracture callus [82]. Moreover, growth plate chondrocytes possess both 1-hydroxylase (Cyp27B1) and 24-hydroxylase (Cyp24) enzymes and produce 1α,25(OH)2D3 and 24R,25(OH)2D3 in a regulated manner at nanomolar concentrations [83–85]. One explanation for the retention of these vitamin D metabolites in the cartilage is that they are bound by cartilage oligomeric matrix protein (COMP), which a structural extracellular matrix protein present in both the growth plate and articular cartilage [86,87]. Although both forms of cartilage have the potential to concentrate vitamin D metabolites, there are tissue-specific differences in response. Receptors for 1α,25(OH)2D3 are found throughout the growth plate [88], but in articular cartilage, these receptors appear to be present only in chondrocytes associated with osteoarthritic or rheumatic [89] lesions, suggesting that they mediate the actions of 1α,25(OH)2D3 on extracellular matrix degradation and calcification. This hypothesis is supported by the observation that VDRs are reduced or absent in tibial dyschondroplasia [90], a condition in which normal chondrocyte hypertrophy and extracellular matrix calcification fail to occur. In addition, recent studies have shown that hyaluronic acid accumulates in the hypertropic zone of growth plates of rachitic animals and this is reversed by treatment with 1,25(OH)2D3 [91]. Moreover, 1,25(OH)2D3 regulates expression of MMP-3 (stromelysin-1) in articular cartilage cells and activity of MMP-3 in growth plate chondrocytes [92–95]. Administration of vitamin D has been used effectively for over 70 years in the treatment of rickets. The effect of vitamin D deficiency on cartilage has been extensively studied. In rickets, the proliferating cell zone usually does not differ much from normal in extent or in arrangement of cells. The primary disturbance caused by vitamin D deficiency is a failure of maturation of prehypertrophic cartilage and reduced mineralization of hypertrophic cartilage matrix. As a result, the prehypertrophic and hypertrophic zones are significantly increased, there is abnormal invasion of metaphyseal blood vessels, and calcification along the longitudinal septa of the cartilage columns is defective.

579

Figure 2 diagrams the changes seen in epiphyseal cartilage in chicks raised for 3 weeks in the absence of ultraviolet light and fed a diet deficient in vitamin D. Crystallographic studies suggest that the mineral crystals that do form in rachitic rats are less mature than those of normal controls [96]. The pathology associated with rickets may reflect lesions in a number of aspects of growth plate development. 1,25(OH)2D3 may regulate the differentiation of postproliferative prehypertrophic cells, such that they do not enter into hypertrophy in a normal manner. However, failure of the cartilage to become calcified is a more important contributor. This may be due in part to a failure of the cells to transport calcium effectively. The distribution of mitochondrial mineral granules in the growth plates of rachitic rats (low phosphate, vitamin D–deficient diet) appears to be modified, as revealed by electron microscopic observation [97]. In control rats, intramitochondrial mineral granules were found in maturing and hypertrophic chondrocytes and not in the chondrocytes at the zone of provisional calcification. In the rachitic rats, intramitochondrial mineral granules were reduced in number and found only in some hypertrophic chondrocytes. The gradient of these intramitochondrial mineral granules was restored by the addition of vitamin D or phosphate [98]. It is clear that the mineralization defect is primarily due to a lack of Ca2+ ions because it can be healed simply by raising the Ca2+ ion content of the serum [99]. The components necessary to support mineral formation are present in rachitic cartilage at least to some extent [100], but it is important to note that the rapid formation of mineral that occurs following vitamin D treatment, while adequate to repair a rachitic lesion, is not necessarily physiological. Vitamin D modulates other aspects of growth plate physiology that may also contribute to the development of rickets and to subsequent regulation of growth plate physiology following treatment. These are outlined next. Lipid metabolism of the growth plate is markedly altered in vitamin D deficiency [71,101]. Treatment of rachitic chicks with vitamin D causes an increase in activity of phospholipase A2 in the growth plate [67]. The relative content of chick growth plate matrix vesicle proteolipids, which are resistant to phospholipase A2 activity, also increases with vitamin D treatment [71]. Other studies show that treatment of rachitic chicks with vitamin D metabolites results in an increase in lysophospholipid content [54]. Moreover, specific classes of phospholipids are affected by specific vitamin D metabolites. A change in the regulation of matrix vesicle enzymes including alkaline phosphatase and the matrix metalloproteinases may also contribute to rickets.

580

BARBARA D. BOYAN AND ZVI SCHWARTZ

Articular cartilage Resting zone cells Proliferating cells Prehypertrophic cells Hypertrophic cells Calcifying cartilage Bone

Epiphyseal plate *

**

*First cut **Second cut

Normal

*

** Rachitic

FIGURE 2 Schematic drawing showing the cellular organization of both the normal and rachitic chick growth plate. Cornish Rock hens were maintained from birth either under normal lighting conditions and chicken diet containing normal levels of calcium and phosphate or in the absence of ultraviolet light and with chicken diet containing low calcium and phosphate. At the age of 3 weeks, the animals were sacrificed, and histologic sections of the long bone epiphyses were prepared, stained with hematoxylin and eosin, and examined. Rachitic cartilage exhibited expanded prehypertrophic and hypertrophic cell zones. The amount of metaphyseal bone was reduced, and there was a widening of the epiphysis. Note the interdigitation of metaphyseal bone spicules with the hypertrophic cartilage, a feature of both normal and rachitic chick growth plates. Enlarged views of the epiphyseal plate appear at bottom.

Expression of alkaline phosphatase mRNA is dependent on 1,25(OH)2D3 [32,106]. Thus, in a rachitic animal, one would expect matrix vesicles that are deficient in this important enzyme to be produced by the chondrocytes, which is in fact the case. Although the precise role of alkaline phosphatase in calcification is not known, its activity is directly correlated with the ability of a tissue to calcify. Under normal conditions, alkaline phosphatase activity increases through the hypertrophic zone [66]. In rachitic animals, activity is markedly suppressed, but treatment with 1,25(OH)2D3 causes a rapid increase in enzyme activity. While it is certainly possible that this is due to new gene transcription, it is also possible that preexisting matrix vesicle alkaline phosphatase is activated by the hormone [107]. Other matrix vesicle enzymes are also regulated by vitamin D. Carbonic anhydrase, which plays a role in maintaining an alkaline pH in the region of crystal growth, is regulated by the hormone [108]. Studies have suggested that both matrix metalloproteinase

activity and protein kinase C (PKC) activity of matrix vesicles are regulated by vitamin D metabolites as well [56–58,109]. Many of these enzymes are Zn-dependent and Zn metabolism is affected by 1,25(OH)2D3 [20]. Proteoglycan degradation is regulated by vitamin D [57,58], but the hormone also modulates growth plate maturation and calcification by regulating proteoglycan synthesis. In vitamin D deficiency in the chick, smaller aggregating proteoglycans are synthesized [110,111]. Restoration of calcium results in the production of larger proteoglycans. These effects, which were shown to be specific for growth cartilages, have been confirmed by others [112]. In healing rickets in the rat, there is a substantial decrease in proteoglycan binding to hyaluronic acid [113]. This suggests that proteoglycan cleavage, which often occurs in the part of the molecule adjacent to the hyaluronic acid binding region and likely mediated by stromelysin [73], is increased in the presence of vitamin D and retarded in rickets.

581

CHAPTER 33 Genomic and Nongenomic Regulation by 1,25(OH)2D3 and 24,25(OH)2D3

Vitamin D deficiency can also alter the rate at which cartilage is resorbed. If resorption of calcified cartilage does not occur, negative feedback through the growth plate to the prehypertrophic and proliferative zones may result. Thus, the absence of hydroxyapatite crystals is important. However, equally important is the absence of molecules that coat the hydroxyapatite crystals and serve as attachment sites for the osteoclasts that will remove the calcified cartilage prior to bone formation. In addition, 1,25(OH)2D3 regulates osteoclast formation, and these cells are needed for proper removal of the calcified cartilage once it forms. Table I provides a summary overview of the actions of vitamin D in each cell zone of the growth plate. There has been some controversy over the role of various vitamin D metabolites in the regulation of growth plate cartilage; however, some of this controversy has resulted from the selection of animal model. For example, the antirachitic activity of vitamin D2 is much less than that of vitamin D3 in the chicken and in a New World monkey, but there is no difference between them in the rat [114]. In addition, in vivo studies have been complicated by the ability of the animal to further metabolize the hormone in the liver and kidney. Various tissues, including growth plate cartilage [115–119], have the ability to metabolize 25-hydroxyvitamin D3 (25OHD3) locally. The hormone 1,25(OH)2D3 has been found to be essential for normal differentiation of cartilage and the prevention of rickets. However, calcification of rachitic cartilage may occur, even in the absence of

1,25(OH)2D3, if the calcium and phosphate content of the extracellular fluid are normal [99,120]. This can be achieved in rachitic chicks either by 1,25(OH)2D3 or even through administration of a high calcium and phosphate diet [121,122]. One of the most elegant experiments demonstrating the importance of 1,25(OH)2D3 in maintaining calcium homeostasis was done using rachitic rats born of rachitic rat mothers [123]. By infusing these rats with calcium, the investigators were able to “heal” the rickets. Similar observations were made by Underwood and DeLuca [12]. Most recently, it was shown that mice lacking a functional vitamin D receptor (VDR−/−) developed rickets [124], but if the mice were raised on a high lactose/high Ca2+ diet, no rickets was evident [99]. Studies such as these suggest that maintenance of the supply of calcium and phosphate is the primary function of vitamin D metabolites in cartilage. It is certainly possible that enough redundancy exists in the growth plate that administration of 1,25(OH)2D3 to rachitic rats can lead to rapid calcification of the matrix. In fact, Howell et al. [100,125] demonstrated that rachitic rat cartilage contains a nucleator capable of promoting calcification in vitro, suggesting that if the calcium and phosphate content of the extracellular fluid were sufficiently high, mineralization in vivo will occur. Once calcification takes place, the resorption of the calcified cartilage can occur, and it may well be that this is critical to the flow of messages up the growth plate to the prehypertrophic zone.

TABLE I Differential Effects of 1α,25(OH)2D3 and 24R,25(OH)2D3 on Rat Costochondral Resting Zone (RC) and Growth Zone (GC) Chondrocytes 1α,25(OH)2D3

24R,25(OH)2D3

Effect

RC

GC

RC

GC

Alkaline phosphatase activity Proteoglycan sulfation Phospholipase C (PLC) activity Phospholipase D (PLD) activity Protein Kinase C (PKC) activity Diacylglycerol (DAG) production Inositol-trisphosphate (IP3) production Mitogen Activated Protein (MAP) kinase activity Phospholipase A2 (PLA2) activity Arachidonic acid release Prostaglandin E2 (PGE2) production Protein Kinase A (PKA) activity Membrane fluidity

— — — — — — —

↑ ↑ ↑ — ↑ ↑ ↑

↑ ↑ — ↑ ↑ ↑ —

— — — — — — —

— — — — — —

↑ ↑ ↑ ↑ ↑ ↑

↑ ↓ ↓ ↓ ↓ ↑

— — — — — ↓

582 Healing of rickets by administration of 1,25(OH)2D3 may only be a part of the vitamin D story. 24,25(OH)2D3 may also play an important role, which is most evident in the resting zone, leading to the hypothesis that 24,25(OH)2D3 is involved in promoting the differentiation of the resting-zone chondrocyte into a more mature phenotype [126]. Local injection of 24,25(OH)2D3 into the upper tibial growth plate [127] or systemic injection of 24,25(OH)2D3 [94] can heal rickets. It is unlikely that this is due to hydroxylation of 24,25(OH)2D3 to 1,24,25(OH)3D3 and subsequent actions of 1,24,25(OH)3D3 on Ca2+ release, as local injection of 1,25(OH)2D3 is relatively ineffective [127]. Rather, it is more likely that 24,25(OH)2D3 acts directly on chondrocytes in the upper growth plate to promote their differentiation along the endochondral lineage. When 24,25(OH)2D3 is injected with 1,25(OH)2D3 into fracture callus, bone repair occurs more rapidly than when 1,25(OH)2D3 is injected alone [82]. The role of 24,25(OH)2D3 in growth plate cartilage development is discussed in greater detail in Section III. However, a few comments should be made concerning the physiology of mice lacking endogenous 24,25(OH)2D3 due to disruption of the gene for 24-hydroxylase [128]. Mice in which the gene for 24-hydroxylase has been knocked out do not express a marked rachitic phenotype. These mice lack a mandible, which is formed in the embryo on Meckel’s cartilage, indicating a cartilage effect. Moreover, they fail to calcify osteoid in certain regions of their bones and there are small but important differences in their growth plates. One reason that these mice are not rachitic in a classic sense is that they have very high levels of circulating 1α,25(OH)2D3 [129]. Thus, their long bone phenotype is more typical of hypervitaminosis D. Only a few studies on the effects of hypervitaminosis D on cartilage and bone in growing animals have been published. Wider epiphyses were reported with excess metaphyseal bone, causing “hypervitaminosis D rickets.” On the other hand, when pregnant rats were treated with high doses of vitamin D2 during days 10–21 of gestation, the fetuses had short bones, mainly due to short diaphyses. The cartilaginous epiphyses were of normal length, but they contained less calcium and atypical chondrocytes [130,131].

II. SEPARATE ROLES FOR 24,25(OH)2D3 AND 1,25(OH)2D3 IN CARTILAGE Numerous studies have shown that 1,25(OH)2D3 regulates the terminal differentiation of hypertrophic cartilage [20,132,133] and maintains the concentration of extracellular Ca2+ so that calcification can occur [120,134]. However, in vivo studies of rachitic rats

BARBARA D. BOYAN AND ZVI SCHWARTZ

in which normal plasma levels of 1,25(OH)2D3 were retained during the first week on a rachitogenic diet showed that mild rachitic lesions occurred anyway [135], suggesting that vitamin D metabolites other than 1,25(OH)2D3 might be involved, although studies of this type do not rule out the fact that low calcium and phosphate availability might be the critical factor. More recent studies (described later in this section) seem to demonstrate a distinct role for 24,25(OH)2D3 in the early stages of endochondral maturation. Confusion over the role of 24,25(OH)2D3 in the growth plate has stemmed in large part from studies published in the 1980s that addressed the role of vitamin D metabolites in the treatment of rickets using a series of novel compounds [136,137]. One vitamin D analog incorporated fluoride in place of the hydroxyl group on C-24. Because this fluorinated analog appeared to have no effect, the investigators concluded that 24,25(OH)2D3 did not possess bioactivity other than as a mechanism to direct excess 25 OHD3 to be excreted. There is an increasing body of data indicating that growth plate cartilage is a target organ for 24,25(OH)2D3. Fine et al. [138] demonstrated that radiolabeled 24,25(OH)2D3 could be found in chick epiphyseal cartilage. Similarly, when vitamin D–replete rats were injected with radiolabeled 25OHD3, radiolabeled 24,25(OH)2D3 was concentrated in growth plate cartilage, but not articular cartilage [139]. Finally, injection of radiolabeled 24,25(OH)2D3 into vitamin D– replete rats also resulted in enhanced uptake of the metabolite by the growth plate [81]. Receptors for both 1,25(OH)2D3 and 24,25(OH)2D3 have been shown by autoradiographic studies in growth plate cartilage [139,140], further supporting the hypothesis that this vitamin D metabolite modulates these cells directly. Although these receptors have never been definitively identified, it appears that regulation of endochondral bone formation involves both 1,25(OH)2D3 and 24,25(OH)2D3. Studies using fetal mouse long bone organ cultures have shown that maintenance of normal hydration, hypertrophy, columnar arrangement, and preservation of mineral in growth plate cartilage requires the presence of both 1,25(OH)2D3 and 24,25(OH)2D3 at physiological concentrations [141]. Similarly, both metabolites are required for maximal production of chondrocalcin (C-propeptide of type II collagen) and calcification of rat [142] and bovine [143] growth plate chondrocyte cultures. The possibility remains that the effects of 24,25(OH)2D3 administration may alter 1,25(OH)2D3 metabolism. The complexities of in vivo homeostasis have made it very difficult to determine if 24R,25(OH)2D3 has a distinct function. Thus, until the advent

CHAPTER 33 Genomic and Nongenomic Regulation by 1,25(OH)2D3 and 24,25(OH)2D3

of cell culture models, it was difficult to identify specific roles for each of the vitamin D metabolites. Studies using chick limb bud cultures provided strong support for the hypothesis that 24R,25(OH)2D3 and 1α,25(OH)2D3 act on separate populations of growth plate cells, but this model, like other three-dimensional cell culture models, consists of a mixed population of chondrocytes at a variety of endochondral maturation states. Among the most completely described cell culture models for examining regulation by these vitamin D metabolites are the rat [47,107] and rabbit [141,144,145] costochondral cartilage systems. Studies using these models show that 24,25(OH)2D3 and 1,25(OH)2D3 exert independent effects on growth plate chondrocytes. For example, 1,25(OH)2D3 stimulates the proliferation of rabbit costal growth plate cells but inhibits proteoglycan synthesis [144]. In contrast, 24,25(OH)2D3 has no effect on proliferation of these cells but stimulates proteoglycan synthesis. Putative receptors for 24,25(OH)2D3 have been localized by autoradiography to the proliferating cells in the growth plate, whereas those for 1,25(OH)2D3 were found in osteoprogenitor cells and osteoblasts [138]. In vitro studies have demonstrated that growth plate chondrocytes exhibit differential responsiveness to 1,25(OH)2D3 and 24,25(OH)2D3. When rabbit growth plate organ cultures were treated with 24,25(OH)2D3, calcium granules were seen in the cytoplasm, but matrix vesicles were devoid of crystals [141]. In contrast, 1,25(OH)2D3 promoted mineral deposition in the matrix. These observations also indicate that 1,25(OH)2D3 and 24,25(OH)2D3 act on different target cells in endochondral bone formation. Using embryonic chick limb bud cells as a model, Boskey et al. [50] found that the less mature cells are affected by 24,25(OH)2D3, whereas the more mature cells are affected by 1,25(OH)2D3. When chick embryo cartilage cells were examined for evidence of 1,25(OH)2D3 receptors, the 1,25(OH)2D3 receptor was found only in growth cartilage cells and not in resting cartilage [145]. Increases in calcification of growth cartilage are paralleled by an increase in 1,25(OH)2D3 receptor levels [146]. The differential response is seen in rats as well. Rat growth plate chondrocytes contain a l,25(OH)2D3-dependent calcium binding protein that is localized in the growth zone of the rat growth plate and is found in the cytoplasm of prehypertrophic cells and in the cytoplasmic processes of hypertrophic chondrocytes [147,148]. Similarly, the protein is found in the growth zone of the chick embryo growth plate [148]. In contrast to this specific effect of 1,25(OH)2D, studies using embryonic chick chondrocytes have demonstrated that 24,25(OH)2D3, but not 1,25(OH)2D3, induces increases in cell proliferation and protein synthesis [107,149].

583

Cell culture studies have confirmed that the response of rat growth plate chondrocytes to vitamin D3 metabolites depends on the zone of maturation from which the cells were originally derived [150]. Rat costochondral resting-zone chondrocytes respond primarily to 24,25(OH)2D3, whereas growth-zone chondrocytes respond primarily to 1,25(OH)2D3. In addition, treatment of resting-zone chondrocytes with 24,25(OH)2D3 for 36 hr causes a change in maturation state, indicating that this hormone has a very specific role in chondrocyte differentiation [126]. The response of each cell to its metabolite includes differential regulation of plasma membrane and matrix vesicle enzyme activity [151] and fluidity [152], collagen and noncollagen protein synthesis and cell proliferation [153], calcium flux [154], phospholipid metabolism [155], and production of vitamin D metabolites [119]. These studies are discussed in greater detail in Section IV. In the rat costochondral cartilage growth plate chondrocyte model, the resting-zone (reserve zone) cartilage is separated by sharp dissection from the adjacent bone and most of the proliferating cell zone of the growth plate cartilage. Similarly, the prehypertrophic and upper hypertrophic zones (growth zone) are separated from the proliferating cell zone and calcified cartilage. Cells in the resting-zone and growth zone are cultured separately, ensuring a clean demarcation of cell maturation. Although articular chondrocytes have a well-known tendency to lose expression of phenotypic markers during long-term culture and subculture [39], the rat costochondral cartilage cells retain their ability to synthesize cartilage-specific proteoglycan and type II collagen through fourth passage. Not only do these cells exhibit differential responses to 24R,25(OH)2D3 and 1α,25(OH)2D3, but they also respond in distinctly different ways to a number of other regulatory agents. This model will be used next to describe the mechanisms of action for each of these vitamin D metabolites.

III. RAPID ACTIONS OF VITAMIN D AND NONGENOMIC MECHANISMS A. Definitions and Models for Studying Rapid Actions In addition to 1,25(OH)2D3 receptors (VDR) [140], putative receptors for 24,25(OH)2D3 have been identified in cartilage [138,139]; therefore, it is probable that many of the effects are via classic VDR pathways, involving changes in gene transcription and mRNA stabilization. This is clearly the case for 1,25(OH)2D3. Chick limb bud mesenchymal cells [156], chick sternal chondrocytes [32], and chick epiphyseal

584 chondrocytes [157] all exhibit l,25(OH)2D3-dependent synthesis of proteins such as alkaline phosphatase. Although less is known concerning the action of 24,25(OH)2D3, its effects on various markers of chondrocyte differentiation strongly suggest that genomic regulatory mechanisms are involved [101,153,158]. This is supported by studies using inhibitors of gene transcription and translation [109]. New evidence has suggested that at least some of the effects of vitamin D metabolites in cartilage occur via membrane-mediated mechanisms, some of which are nongenomic. Many of these effects involve phospholipid metabolism. Early physiological and biochemical studies showed that vitamin D regulated the phospholipid and fatty acid composition of the growth plate during endochondral bone formation [103–105], supported more recent studies on the mechanisms of vitamin D action in this tissue. A nongenomic action of the hormone is one that involves neither new gene transcription nor protein synthesis. Although many rapid effects of vitamin D metabolites may be nongenomic, time course of action is not an a priori proof, as new protein synthesis may occur even in very short times. Examples of nongenomic actions include changes in membrane fluidity [152], rapid turnover of phospholipids [155,159], changes in calcium flux [154,160], and rapid activation of protein kinase C [109]. Even these actions, however, may be downstream from a vitamin D–dependent nuclear event, whether or not the VDR is involved. Thus, the only proofs for nongenomic mechanisms are those that examine responses in the absence of DNA, RNA, or protein synthesis. To definitively prove physiological relevance of rapid and nongenomic actions of vitamin D, it is necessary to demonstrate that they occur in vivo. Few such experiments have been done. However, new molecular techniques are making experiments of this kind more possible. The rat costochondral chondrocyte culture model has been particularly useful for studying the membrane-mediated mechanisms of 1,25(OH)2D3 and 24,25(OH)2D3 action. One reason for this is that both resting-zone and growth-zone chondrocytes produce extracellular matrix vesicles in culture and these organelles can be separated from the cells that produce them by simple biochemical techniques that do not require homogenization or lysis. Because matrix vesicles are exterior to the cell, are isolated right-side-out, and contain no DNA or RNA, any direct effect of the vitamin D metabolites on them is by definition nongenomic. The phospholipid composition [47] and enzyme activity [150,151] of these matrix vesicles depend on the cell of origin. Not only do matrix vesicles produced by growth-zone chondrocytes differ from those produced by resting-zone chondrocytes, but

BARBARA D. BOYAN AND ZVI SCHWARTZ

each type of matrix vesicle differs from the plasma membrane of the cell from which it was derived. Not surprisingly, the basal membrane fluidity of matrix vesicles is distinct from that of the plasma membrane [152], reflecting differences in structure and chemical composition. Thus, we can separate the genomic regulation of the cells resulting in gene transcription, protein synthesis, and new matrix vesicle production, from nongenomic regulation by comparing the effects of 24R,25(OH)2D3 and 1α,25(OH)2D3 on cell cultures with those on isolated plasma membranes and isolated matrix vesicles. Our studies confirmed the hypothesis that 1α,25(OH)2D3 and 24R,25(OH)2D3 target specific subpopulations of growth plate chondrocytes and that both genomic and nongenomic mechanisms are involved. 1α,25(OH)2D3 causes an increase in alkaline phosphatase only in cultures of growth-zone cells, and this effect is targeted to matrix vesicles [161]. In contrast, 24R,25(OH)2D3 affects enzyme activity only in cultures of resting-zone cells and this effect is targeted to matrix vesicles as well. It should be noted, parenthetically, that Hale et al. [157] found that 24,25(OH)2D3 caused an increase in alkaline phosphatase activity in serum-free cultures of chick growth plate chondrocytes at low doses, whereas at higher concentrations enzyme activity was inhibited. Although species differences may be involved, it is more likely that the effects on chick cells represent an initial response in the absence of other serum factors, whereas in the rat costochondral cultures effects of hormone are examined in a background of serum. Indeed, TGF-β1 causes a synergistic increase in matrix vesicle alkaline phosphatase in cultures of rat restingzone cells treated with 24R,25(OH)2D3 [60,162]. 1α,25(OH)2D3 and 24R,25(OH)2D3 also regulated phospholipase A2 specific activity in a cell-specific manner [151]. 1,25(OH)2D3 stimulated phospholipase A2 specific activity in matrix vesicles isolated from growth-zone chondrocyte cultures. In contrast, 24,25(OH)2D3 inhibited phospholipase A2 specific activity in matrix vesicles isolated from resting-zone chondrocyte cultures. Because matrix vesicles are already external to the cells, these data indicate that either the chondrocytes make new matrix vesicles in response to hormone or that the hormones act directly on the matrix vesicles in the matrix. Both mechanisms are involved [107]. 1,25(OH)2D3 stimulates both alkaline phosphatase and phospholipase A2 specific activities when incubated directly with matrix vesicles isolated from growth-zone chondrocytes but has no effect on either enzyme in matrix vesicles isolated from resting-zone cell cultures. In contrast, 24,25(OH)2D3 stimulates alkaline phosphatase and

CHAPTER 33 Genomic and Nongenomic Regulation by 1,25(OH)2D3 and 24,25(OH)2D3

inhibits phospholipase A2 specific activities when incubated directly with matrix vesicles from restingzone cell cultures, but it has no effect on either enzyme in matrix vesicles isolated from growth-zone chondrocyte cultures. These experiments clearly demonstrate that vitamin D metabolites mediate specific effects in the absence of genomic machinery. Moreover, the direct effects are essentially identical to those observed in the intact cultures [163], suggesting that 1,25(OH)2D3 and 24,25(OH)2D3 may operate in vivo, at least in part, through nongenomic pathways.

B. Membrane Fluidity The simplest explanation for how vitamin D metabolites could regulate resting-zone chondrocytes in a differential manner is based on the concept that these lipophilic molecules could alter the fluidity of the membrane in much the same way as is seen in artificial membranes treated with cholesterol [164,165]. The difference in charge density between 1,25(OH)2D3 and 24,25(OH)2D3 would automatically make the metabolites assume different habits in the membrane, with different consequences to its fluid mosaic structure. Furthermore, plasma membranes from resting-zone cells differ from those of growth-zone cells with respect to their lipid composition and enzyme activity, and matrix vesicles produced by these cells differ from their parent plasma membranes and from one another [47]. Not surprisingly, basal fluidity of these membrane preparations differs as well [158]. When resting-zone or growth-zone chondrocyte cultures are treated with 1,25(OH)2D3 or 24,25(OH)2D3, plasma membrane fluidity is changed in a cell maturation–specific manner. 1,25(OH)2D3 increases the fluidity of the growth zone cell membrane but has no effect on resting-zone cell membrane fluidity. 24,25(OH)2D3 decreases the fluidity of the growth-zone cell membrane but increases the fluidity of restingzone cell membranes. Similar effects are seen when these metabolites are incubated directly with matrix vesicles isolated from naive cultures.

585

of radiolabeled arachidonic acid is also cell maturation specific [155,159]. Within 1 min, 1,25(OH)2D3 and 24,25(OH)2D3 begin to modulate arachidonic acid turnover by the cells. 1,25(OH)2D3 stimulates arachidonic acid turnover in growth-zone chondrocytes, but has no effect on resting-zone chondrocytes. In contrast, 24,25(OH)2D3 modulates arachidonic acid turnover in resting-zone chondrocytes. These results suggest that at least part of the effect of the metabolites on cell membranes may be related to changes in phospholipase A2 activity and fatty acid turnover. It is important to note that arachidonic acid incorporation is regulated as well, indicating that the vitamin D metabolites modulate phospholipid synthesis in the growth plate. This effect is rapid and results in a shift in the relative distribution of cellular and matrix vesicle membrane phospholipids [155,159]. As a consequence of altered arachidonic acid release, 1,25(OH)2D3 and 24,25(OH)2D3 regulate prostaglandin production by growth-zone and resting-zone chondrocytes [166]. 1,25(OH)2D3 stimulates production of prostaglandins E1 and E2 (PGE1 and PGE2) from growthzone cells. Basal production of these prostaglandins is higher in resting-zone cell cultures, and when the cultures are treated with 24,25(OH)2D3, production is reduced to levels somewhat similar to those seen in the growth-zone cell cultures. Phospholipase A2 activation is the rate-limiting step in prostaglandin synthesis in response to 1α,25(OH)2D3 [167]. Inhibition of constitutively expressed cyclooxygenase-1 blocks the increase in PGE2 production whereas inhibition of inducible cyclooxygenase-2 has no effect.

D. Calcium Flux One of the most rapid effects of 1,25(OH)2D3 and 24,25(OH)2D3 on the costochondral chondrocytes is to alter Ca2+ flux [154]. Using uptake and release of 45Ca2+ as an indicator, a series of studies showed that the effects of these metabolites were observed within 1 min of exposure. The effects of the metabolites were complex. Although there was crossover in target cell specificity, the kinetics of the response were cell maturation specific.

C. Phospholipid Metabolism A change in membrane fluidity may also be due to rapid retailoring of membrane phospholipids. 1,25(OH)2D3 and 24,25(OH)2D3 differentially regulate phospholipase A2 activity based on the release of radiolabeled arachidonic acid from prelabeled phosphatidylethanolamine [102,151]. The effect of the vitamin D metabolites on general incorporation and release

E. Membrane Signaling One of the most exciting aspects of how 1,25(OH)2D3 and 24,25(OH)2D3 exert their differential effects on chondrocytes has arisen from application of classic peptide membrane receptor technology to the study of lipophilic hormones. These studies indicate

586 that both vitamin D metabolites work through signal transduction mechanisms involving protein kinase C (PKC). 1,25(OH)2D3 activates PKC activity exclusively in growth-zone cells, whereas 24,25(OH)2D3 activates PKC only in resting-zone cells [109]. There are distinct differences in the kinetics of action and the requirement for genomic expression. 1,25(OH)2D3 exerts its effect without gene expression or protein synthesis. The increase is rapid and is maximal at 9 min. In contrast, the increase in PKC activity due to 24,25(OH)2D3 involves new gene expression and new protein synthesis. Increased activity occurs within minutes but the maximal increase is at 90 min. In both instances, the alpha isoform of PKC (PKCα) is activated but there are distinct differences in mechanism [168] (Section III,G). Whereas the effect of 1α,25(OH)2D3 on PKC in growth-zone cells is on existing enzyme, the effect of 24R,25(OH)2D3 on PKC in resting-zone cells includes a rapid effect on existing plasma membrane associated PKC and an increase in new PKC. The vitamin D metabolites also modulate PKC activity in the matrix vesicles. 1α,25(OH)2D3 causes an increase in matrix vesicle PKC activity through production of new organelles whereas 24R,25(OH)2D3 increases PKC activity in matrix vesicles produced by resting-zone cells [158,168]. This new PKC is not PKCα, but PKC-zeta (PKCζ), which is an isoform that does not require Ca2+ and/or phospholipid. 1α,25(OH)2D3 inhibits PKCζ activity when incubated directly with matrix vesicles produced by growth-zone cells, and 24R,25(OH)2D3 inhibits PKCζ when incubated directly with matrix vesicles produced by restingzone cells. This differential distribution of PKC isoforms and their differential regulation by the vitamin D metabolites suggests an explanation for how 1,25(OH)2D3 and 24,25(OH)2D3 may exert one set of effects on the cells and another on the matrix vesicles. Thus, incorporation of PKCζ into matrix vesicles is under genetic control but regulation of preexisting matrix vesicle PKCζ is under nongenomic control. PKC activation is not an end in itself. 1α,25(OH)2D3 and 24R,25(OH)2D3 both mediate their physiological effects on their target cells through PKC-dependent signaling. Some of the PKC-dependent responses are regulated through the ERK1/2 family of mitogenactivated protein kinases (MAPK) [169]. As noted for PKC [109], 1α,25(OH)2D3 exerts its stimulatory effects on MAPK activity in growth-zone cells only and 24R,25(OH)2D3 increases MAPK activity in restingzone cells only. MAPK provides a mechanism by which 1α,25(OH)2D3 and 24R,25(OH)2D3 can modulate gene expression independently of the VDR. It is also possible

BARBARA D. BOYAN AND ZVI SCHWARTZ

that the signaling pathways activated by the vitamin D metabolites may work in concert with the VDR. Recent studies using costochondral resting-zone and growthzone chondrocytes from VDR knockout mice (VDR−/−) [170] show that some physiological responses may be independent of the VDR whereas others require the traditional nuclear receptor. Importantly, the target cell specificity of PKC activation was present in the absence of a functional VDR: 1α,25(OH)2D3 activated PKC in growth-zone cells and 24R,25(OH)2D3 activated PKC in resting-zone cells. The maximal effect of 1α,25(OH)2D3 was at 9 min and occurred by the same mechanism as in rat chondrocytes possessing a functional receptor. The maximal effect of 24R,25(OH)2D3 was at 90 min and occurred by the same mechanism as in rat resting-zone cells. These observations do not rule out the possibility of a truncated form of the VDR playing a role in membrane-mediated signaling, but they do rule out the possibility that the traditional mechanism is involved since the genetic lesion that produced the mice was in the Zn-finger DNA-binding region of the receptor [124]. Moreover, the results indicate that the effect of 24R,25(OH)2D3 is independent of the effect of 1α,25(OH)2D3 and likely involves a different receptor.

F. Membrane Receptors The experiments described above indicate that the direct effects of vitamin D metabolites are likely to occur on the membrane and that receptors other than the traditional VDR are involved. The effects of 1,25(OH)2D3 and 24,25(OH)2D3 are stereospecific; 1α,25(OH)2D3 and 24R,25(OH)2D3 elicit a response whereas 1β,25(OH)2D3 and 24S,25(OH)2D3 do not. Stereospecificity supports a receptor-mediated mechanism rather than a physical change in membrane structure and fluidity, which would be insensitive to differences in enantiomers. Similarly, the effects of a number of 1,25(OH)2D3 analogs with modified A rings and/or CD-ring side chains are all stereospecific [171,172]. Moreover, these analogs have very low affinity for the VDR, suggesting that they are interacting with a different entity or entities. One possibility is the protein ER60. ER60 is the rat homolog of a 1α,25(OH)2D3-binding protein isolated from the basal lateral membranes of chick intestinal epithelium, also termed 1,25-MARRS [173]. Antibodies to ER60 block the effect of 1α,25(OH)2D3 and its analogs on PKC in rat growth-zone chondrocytes, as well as on PKC-dependent responses to the vitamin D metabolite [181,182]. Anti-ER60 antibodies recognize a single 65-kDa band on Western blots of matrix

587

CHAPTER 33 Genomic and Nongenomic Regulation by 1,25(OH)2D3 and 24,25(OH)2D3

vesicles from rat costochondral chondrocytes [175]. This band is not recognized by antibodies to the VDR or by antibodies to annexin II, which has also been suggested to be a membrane receptor for 1α,25(OH)2D3. The anti-ER60 antibodies do not affect any responses to 24R,25(OH)2D3 [174], supporting the hypothesis that the two metabolites work via different receptors. Seo and Norman reported the possibility of a 24,25(OH)2D3 membrane receptor in chick fracture callus, which also supports this hypothesis.

different mechanisms. Phospholipase A2 plays a pivotal role in the differential response of resting-zone cells and growth-zone cells to 1α,25(OH)2D3 and 24R,25(OH)2D3. Each pathway will be described hereafter. 1. 1,25(OH)2D3

The mechanism of rapid action of 1,25(OH)2D3 on growth-zone chondrocytes is shown diagrammatically in Fig. 3. 1,25(OH)2D3 acts via a mechanism involving ER60 to cause a rapid increase in cytosolic phospholipase A2 activity. The increase in phospholipase A2 activity depends upon the presence of phospholipase A2 activating protein (PLAA) (unpublished data). This results in increased fatty acid turnover [155] and, ultimately, to an increase in membrane fluidity [152]. This fluidity change can alter membrane enzyme activity and calcium

G. Mechanism of Action It is now clear that 1α,25(OH)2D3 and 24R,25(OH)2D3 activate PKC in their target cells through distinctly

Membrane effect of 1,25(OH)2D3 in GC cells AA 1,25 PIP2

PLC

Lyso

PLA2

PGE2

t

or

ep

c Re

Gq

PLA2 EP1 Gq

DAG

Tr

an

slo

ca

tio

n

IP3

PKC Ca++ Protein Protein-P Ser/Thr MAP kinase

Nucleus

FIGURE 3 Mechanism for the rapid action of 1,25(OH)2D3 on growth-zone chondrocytes. In growth-zone (GC) cells, 1,25(OH)2D3, acting via a membrane receptor, increases phospholipase A2 (PLA2) activity in the membrane. This results in increased fatty acid turnover and changes in membrane fluidity. Changes in membrane fluidity can alter the activity of enzymes in the membrane and also calcium flux. Stimulation of arachidonic acid (AA) release activates protein kinase C (PKC) directly. Through the action of cyclooxygenase-1, increased production of AA also increases the production of prostaglandins, such as prostaglandin E2 (PGE2), which are potent regulators of chondrocytes. PGE2 binds its EP1 receptor, activating the G-protein pathway via Gq, stimulating adenylate cyclase activity and increasing protein kinase A (PKA) activity. Lyosphospholipids (Lyso) are also produced by the action of PLA2, and they activate phospholipase C (PLC). PLC-beta acts on phosphatidylinositol 4,5-bisphosphate (PIP2) to increase the production of inositol trisphosphate (IP3) and diacylglycerol (DAG). IP3 can stimulate the release of calcium from the endoplasmic reticulum (ER) and into the cell. DAG binds PKC, causing its translocation to the plasma membrane. Once PKC is activated, it can activate a signal transduction cascade, by phosphorylation of serine and threonine residues, leading to mitogen-activated protein kinase (MAPK) activation, phosphorylation of AP-1, and increased transcription from relevant gene promoters.

588

BARBARA D. BOYAN AND ZVI SCHWARTZ

flux [154,160]. Lysophospholipids released as a consequence of phospholipase A2 activity activate phosphatidylinositol-specific phospholipase C (PI-PLC). The specific isoforms of phospholipase C that are responsible are PLC-beta-1 and PLC-beta-3 [176]. Phospholipase C then catalyzes the hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP2), increasing production of phosphoinositol trisphosphate (IP3), leaving diacylglycerol (DAG) associated with the membrane [177]. IP3 stimulates release of Ca2+ from the endoplasmic reticulum (ER) and into the cell from the extracellular fluid. Diacylglycerol binds PKC, causing its translocation to the plasma membrane, and together with Ca2+, activates the enzyme [109]. Stimulation of arachidonic acid release increases local concentrations of this fatty acid. Arachidonic acid by itself can stimulate PKC activity [178,179]. In addition, production of arachidonic

acid is the rate-limiting step in prostaglandin production in response to 1α,25(OH)2D3. Increasing arachidonic acid release results in an increase in PGE2 production, which serves as an autocrine regulator of the chondrocytes, acting through its EP1 receptor to activate PKC [180]. Once PKC is activated, it initiates a signal transduction cascade via phosphorylation of serine and threonine residues, leading to activation of ERK1/2 mitogen-activated protein kinase (MAPK) [169], and phosphorylation of AP-1 sites on the relevant gene promoters. 2. 24,25(OH)2D3

The mechanism of 24,25(OH)2D3 action on restingzone cells is schematically shown in Fig. 4. It is predicated on the observation that 24,25(OH)2D3, acting via a hypothetical membrane receptor, inhibits phospholipase A2 activity [151], an effect that will change

Membrane effect of 24,25(OH)2D3 in RC cells

− −

24,25

+

PLD

r

pto

e ec

R

−

AA

PGE2

PLA2 EP-1

EP-2

+ DAG

+ PKC Protein

+ Protein-P Ser/Thr MAP kinase

Nucleus

FIGURE 4 Mechanism for the rapid action of 24,25(OH)2D3 on resting-zone chondrocytes. In resting-zone chondrocytes (RC cells), 24,25(OH)2D3, acting via a hypothetical receptor, inhibits phospholipase A2 (PLA2) activity. This results in changes in fatty acid turnover, reduced release of arachidonic acid (AA), and production of prostaglandin E2 (PGE2). There is a resultant change in membrane fluidity and calcium flux. 24,25(OH)2D3 increases diacylglycerol (DAG) production via phospholipase D (PLD), which activates protein kinase C (PKC). Inhibition of PLA2 results in increased PKC activity by decreasing release of AA and lysophospholipids, thereby reducing inhibition of PKC by these phospholipid metabolites. Exogenous PGE2 decreases PKC via its EP-1 and EP-2 receptors through a PKA-dependent mechanism. Thus, resulting decreases in prostaglandin production via cyclooxygenase-1 action on reduced pools of AA, increases PKC. PKC may then change cell behavior through phosphorylation of different proteins, ultimately phosphorylating the ERK1/2 family of mitogen-activated protein kinases (MAP kinase), leading to changes in gene expression.

CHAPTER 33 Genomic and Nongenomic Regulation by 1,25(OH)2D3 and 24,25(OH)2D3

fatty acid turnover, release of arachidonic acid, and production of PGE2 [155,166]. There is a resultant change in membrane fluidity [152] and Ca2+ flux. 24,25(OH)2D3 increases diacylglycerol production [177], but does so through a two-step mechanism catalyzed by phospholipase D2 [181]. DAG activates PKC, but PKC is not translocated to the plasma membrane. Although resting-zone cells possess PLCβ1 and PLCβ3, neither enzyme is activated by 24R,25(OH)2D3 and inhibition of PLC activity has no effect on PKC activity in response to 24R,25(OH)2D3 [176]. Interestingly, inhibition of phospholipase A2 results in increased PKC activity [177], and phospholipase A2 activation, or addition of lysophospholipids [176], exogenous arachidonic acid [182], or exogenous PGE2 [183], all inhibit PKC. The effect of PGE2 is via protein kinase A (PKA), which is activated through the EP2 receptor in these cells. These observations suggest that by inhibiting phospholipase A2, even for a short period of time, the inhibitory effect of the enzyme and its products are reduced. Moreover, inhibition of phospholipase A2 reduced levels of lysophospholipid, inhibiting phospholipase C activity. As a result, the phospholipase D–dependent mechanisms dominate in these cells. As noted earlier, PKC activation leads to an increase in the activity of ERK1/2 MAPK [169].

IV. PHYSIOLOGIC RELEVANCE OF NONGENOMIC REGULATION OF MATRIX VESICLES Although the cell can down-regulate undesired nongenomic effects at the plasma membrane, this is more difficult in the matrix. To control events in the matrix, the cell may modulate the rate of production and the chemical composition of matrix vesicles [158,184]. Initially, matrix vesicles are produced under genomic control. 1,25(OH)2D3 and 24,25(OH)2D3 regulate the composition of matrix vesicles through new gene transcription, protein synthesis, and, finally, membrane synthesis. Once matrix vesicles are released into the extracellular matrix, the cells can regulate their maturation through secretion of vitamin D metabolites that act on the matrix vesicle through nongenomic mechanisms. If this is the case, then it is obligatory that the cells produce vitamin D metabolites and that this production be regulated by growth factors and hormones. In fact, chondrocytes have been found to produce both 3H1,25(OH)2D3 and 3H-24,25(OH)2D3 when incubated with 3H-25(OH)D3 [119]. Resting-zone chondrocytes and growth-zone chondrocytes express mRNAs for 1-hydroxylase and 24-hydroxylase and possess both

589

enzyme activities [85]. The expression of the enzymes, their activities, and production of vitamin D metabolites are regulated by 1,25(OH)2D3, 24,25(OH)2D3, TGF-β1, and dexamethasone in a cell maturationspecific manner [83–85,119]. It is likely that systemic 1,25(OH)2D3 and 24,25(OH)2D3 also play roles in this process by providing the conditioning background for cell response, whereas locally secreted metabolites would permit fine-tuning of the matrix vesicles. Matrix vesicles may have multiple functions in the matrix. Those in the lower hypertrophic cell zone of cartilage are probably involved in matrix calcification [185]. In addition, matrix vesicles also appear to be involved in matrix maturation, as they contain matrixprocessing enzymes that degrade proteoglycans [186,187]. The activities of these enzymes are regulated in vitro and in vivo by 1α,25(OH)2D3 and 24R,25(OH)2D3 in a cell-specific manner [94,95]. 1α,25(OH)2D3 increases levels of acid metalloproteinases whereas 24R,25(OH)2D3 decreases levels of these enzymes. In addition to controlling the levels of enzyme activity through genomic regulation, both metabolites can modulate matrix vesicle enzyme activity directly. 24R,25(OH)2D3 stabilizes the matrix vesicle, preventing the release of matrix processing enzymes, whereas 1α,25(OH)2D3 causes a loss of matrix vesicle membrane integrity leading to the release of these enzymes. In addition to their role in calcification, matrix vesicles may play an important role in activation of growth factors present in the extracellular matrix, and this is regulated by the vitamin D metabolites. Growth plate chondrocytes synthesize and store latent TGF-β1 as a macromolecular complex in the extracellular matrix [189]. 24R,25(OH)2D3 modulates the expression of latent TGF-β binding protein-1 by restingzone chondrocytes whereas 1α,25(OH)2D3 modulates expression of the protein as well as deposition of TGF-β1 together with its binding protein in the matrix of growth-zone cells [190]. At the same time, release of latent TGF-β1 and its activation are also controlled by 24R,25(OH)2D3 and 1α,25(OH)2D3. 24R,25(OH)2D3 treatment of matrix vesicles produced by the growth plate cells stabilizes the matrix vesicle membrane, inhibiting the release of matrix vesicle enzymes. In contrast, 1α,25(OH)2D3-treatment of matrix vesicles produced by either resting-zone cells or growth-zone cells activates latent TGF-β1 [188]. Studies examining the direct regulation matrix vesicles produced by growth-zone chondrocytes indicate that 1α,25(OH)2D3 causes the release of MMP-3 (stromelysin-1), which then catalyzes the release of latent TGF-β1 from the extracellular matrix [191] and activates the latent growth factor. These observations suggest that nongenomic regulation of matrix vesicles can result in

590

BARBARA D. BOYAN AND ZVI SCHWARTZ

membrane, where their effects initiate a cascade of biochemical events that lead to maturation of the matrix vesicle, hydroxyapatite crystal formation, degeneration of the integrity of the matrix vesicle membrane, and eventual release of active proteases. The proteases then degrade proteoglycan aggregates in the vicinity of the matrix vesicle, facilitating extracellular matrix calcification. In addition, they may activate latent growth factors that can then act on the cell in an autocrine manner or on adjacent cells via paracrine interactions. Further evidence that 1,25(OH)2D3 and 24,25(OH)2D3 can directly affect proteoglycan degradation and matrix calcification via nongenomic effects on matrix

changes in local growth factor activation. This is a particularly attractive hypothesis in cartilage, where activation of latent growth factors by local decreases in pH (as occurs in osteoclasts) has not been reported. These observations led to the following hypothesis for nongenomic regulation of events in the extracellular matrix (Fig. 5). Chondrocytes produce matrix vesicles under hormonal and growth factor regulation. At the same time, vitamin D metabolites are synthesized by the cells and secreted in response to regulatory factors like 1,25(OH)2D3, 24,25(OH)2D3, TGF-β, or corticosteroids. These factors diffuse into the matrix and interact directly with the plasma membrane. In addition, they also interact with the matrix vesicle

ALPase MMP-3* Chondrocyte 1,25 24,25

P P PA2 PKCζ

Matrix vesicles 1,25 24,25

Mitochondrion

AA ?

Nucleus

PGE2

HA PP PA2

Plasma membrane

Ca++ AA Vitamin D metabolite

?

PGE2

PKCα

? MMP-3 MMP-3*

PA2

? P P

P P

ALPase PKCα = Protien kinase C alpha PKCζ = PKC-zeta AA = Arachidonic acid PGE2 = Prostaglandin E2 PA2 = Phospholipase A2 P = Phosphate

Proteoglycan aggregate 1,25 = 1,25-(OH)2D3 24,25 = 24,25-(OH)2D3 PP MMP-3 = Stromelysin-1 MMP-3* = Phosphorylated MMP-3 ALPase = Alkaline phosphatase HA = Hydroxyapatite

FIGURE 5 Proposed mechanism for the nongenomic regulation of matrix vesicles in the extracellular matrix. Chondrocytes produce matrix vesicles under hormonal and growth factor control. Systemic 1,25(OH)2D3 or 24,25(OH)2D3 interact with classic receptors or stimulate rapid membrane-mediated signal transduction pathways, resulting in new gene expression and, ultimately, new matrix vesicle production. Rapid membrane responses include release of arachidonic acid and prostaglandin E2 (PGE2) production as well as altered calcium flux. At the same time, the cells synthesize vitamin D metabolites. The vitamin D metabolites are secreted into the matrix and interact directly with the plasma membrane, causing PKCα-dependent phosphorylation of matrix metalloproteinase-3 (MMP-3). In addition, they also interact with the membrane of preexisting matrix vesicles, where they initiate a cascade of events leading to matrix vesicle maturation, hydroxyapatite crystal formation, and, in matrix vesicles produced by growthzone chondrocytes, loss of matrix vesicle membrane integrity through stimulation of phospholipases. Once this occurs, active proteinases such as MMP-3 are released. The proteinases degrade proteoglycan aggregates, facilitating matrix calcification. In addition, they may activate latent growth factors. Additional details are in the text.

CHAPTER 33 Genomic and Nongenomic Regulation by 1,25(OH)2D3 and 24,25(OH)2D3

vesicles is now available. When growth-zone chondrocytes are treated with 1,25(OH)2D3, there is an increase in matrix vesicle matrix metalloproteinase (MMP) activity [57,58]. Analysis of the direct effect of 1,25(OH)2D3 on isolated membrane fractions indicates that plasma membrane–associated PKC is increased, resulting in the PKCα-dependent phosphorylation of MMP-3 [56]. In this state, MMP-3 is then packaged into matrix vesicles and released into the matrix. However, no MMP activity is detectable in isolated matrix vesicles unless membrane integrity is lost [55]. When treated with 1,25(OH)2D3, isolated matrix vesicles contain increased phospholipase A2, which destabilizes the matrix vesicle membrane, releasing the MMP into the matrix. At the same time, alkaline phosphatase in the matrix vesicles is elevated in response to 1,25(OH)2D3. We hypothesize that alkaline phosphatase then dephosphorylates MMP-3, resulting in increased MMP activity. As PKCζ activity in the matrix vesicle is decreased by 1,25(OH)2D3 treatment, this isoform of PKC has no effect on matrix vesicle MMP-3, leading to the hypothesis that the enzyme is then fully active in the matrix [56]. When isolated matrix vesicles are incubated in gelatin gels in the presence of proteoglycan, the inhibition of crystal formation normally associated with proteoglycan is lost. Furthermore, treatment of the matrix vesicles with 1,25(OH)2D3 causes an increase in the rate and extent of new crystal formation [187].

V. SUMMARY This chapter has shown that cartilage, much like other tissues, is really a family of tissues spanning a broad spectrum of cell maturation states. In growth plate, a subset of the cartilage phenotype, chondrocytes can be seen at distinct states of maturation in a linear array. Using a variety of in vivo and in vitro assays, investigators have been able to show that the growth plate is sensitive to vitamin D regulation, with 24,25(OH)2D3 affecting less mature cells, particularly those of the resting-zone, and 1,25(OH)2D3 modulating activities in the growth-zone (prehypertrophic and upper hypertrophic) cartilage. Both metabolites exert their effects through genomic mechanisms. However, some of the responses of the cell may involve nongenomic mechanisms. Rapid cell membrane–mediated events may result in secondary genomic responses via protein phosphorylation cascades and MAP kinase; in matrix vesicles, rapid membrane effects may be termed nongenomic because no gene expression or protein synthesis is possible. This is relevant to in vivo regulation of endochondral ossification, since chondrocytes produce and secrete 1,25(OH)2D3 or 24,25(OH)2D3,

591

which then may interact with the matrix vesicles, resulting in modulation of the activity of this extracellular organelle. The consequences of this include activation of latent growth factors, degradation of matrix proteoglycans, and calcium phosphate deposition. Thus, 1,25(OH)2D3 and 24,25(OH)2D3 regulate chondrocyte proliferation, metabolism, differentiation, and maturation, as well as events in the extracellular matrix. The effects are cell maturation dependent and organelle specific, and they may involve both VDR-dependent and VDR-independent genomic as well as nongenomic mechanisms.

References 1. Caplan AI, Boyan BD 1994 Endochondral bone formation: The lineage cascade. In: Hall BK (ed) Bone 8. Mechanisms of Bone Development and Growth. CRC Press, Boca Raton, FL, pp. 1–46. 2. Brighton CT 1984 The biology of fracture repair. In: Murray JA (ed) American Academy of Orthopaedic Surgeons Instructional Course Lectures. Mosby, St. Louis, pp. 60–82. 3. Cornell CN, Lane JM 1992 Newest factors in facture healing. Clin Orthop Relat Res 277:297–311. 4. Nemeth GG, Bolander ME, Martin GR 1988 Growth factors and their role in wound and fracture healing. Prog Clin Biol Res 266:1–17. 5. Joyce ME, Terek RM, Jingushi S, Bolander ME 1990 Role of transforming growth factor-/3 in fracture repair. Ann NY Acad Sci 593:107–123. 6. Reddi AH, Wientroub S, Muthukumaran N 1987 Biologic principles of bone induction. Orthop Clin North Am 18:207–212. 7. Reddi AH, Huggins CB 1972 Biochemical sequences in the transformation of normal fibroblasts in adolescent rats. Proc Natl Acad Sci USA 69:1601–1605. 8. Urist MR, Delange RJ, Finerman GA 1983 Bone cell differentiation and growth factors. Science 220:680–686. 9. Boyan BD, Schwartz Z, Swain LD, Khare AG, Heckman JD, Ramirez V, Peters P, Carnes DL Jr 1992 Initial effects of partially purified bone morphogenetic protein on the expression of glycosaminoglycan, collagen, and alkaline phosphatase in nonunion cell cultures. Clin Orthop 278:286–304. 10. Chen P, Carrington JL, Hammonds RG, Reddi AH 1991 Stimulation of chondrogenesis in limb bud mesoderm cells by recombinant human bone morphogenetic protein 2B (BMP-2B) and modulation by transforming growth factor beta 1 and beta 2. Exp Cell Res 195:509–515. 11. Howell DS, Dean DD 1992 Biology, chemistry and biochemistry of the mammalian growth plate, In: Coe FL, Favus MJ (eds) Disorders of Bone and Mineral Metabolism. Raven, New York, pp. 313–353. 12. Poole AR 1991 The growth plate: Cellular physiology, cartilage assembly and mineralization. In: Hall BK, Newman S (eds) Cartilage: Molecular Aspects. CRC Press, Boca Raton, FL, pp. 179–211. 13. Reddi AH 1994 Bone and cartilage differentiation. Curr Opin Genet Dev 4:737–744. 14. Lanske B, Karaplis AC, Lee K, Luz A, Vortkamp A, Pirro A, Karperien M, Defize LHK, Ho C, Mulligan RC, Abou-Samra AB, Juppner H, Segre GV, Kronenberg HM 1996 PTH/PTHrP

592

15.

16.

17.

18. 19. 20. 21. 22.

23. 24. 25. 26.

27. 28.

29. 30. 31.

32.

receptor in early development and indian hedgehog-regulated bone growth. Science 273:663–666. Hunziker EB, Herrmann W, Schenk RK, Mueller M, Moor H 1984 Cartilage ultrastructure after high pressure freezing, freeze substitution, and low temperature embedding. I. Chondrocyte ultrastructure: Implications for the theories of mineralization and vascular invasion. J Cell Biol 98:267–276. Hunziker EB, Schenk RK, Cruz-Orive LM 1987 Quantitation of chondrocyte performance in growth plate cartilage during longitudinal bone growth. J Bone Joint Surg 69A: 162–173. Carlson CS, Hilley HD, Henrickson CK 1985 Ultrastructure of normal epiphyseal cartilage of the articular-epiphyseal cartilage complex in growing swine. Am J Vet Res 46: 306–313. Howell DS, Carlson L 1968 Alterations in the composition of growth cartilage septa during calcification studied by microscopic x-ray elemental analysis. Exp Cell Res 51:185–195. Buckwalter JA, Mower D, Ungar R. 1986 Morphometric analysis of chondrocyte hypertrophy. J Bone Joint Surg 68A: 243–255. Dean DD, Muniz OE, Berman I, Pita JC, Carreno MR, Woessner JF Jr, Howell DS 1985 Localization of collagenase in the growth plate of rachitic rats. J Clin Invest 76:716–722. Weitzhandler M, Carrino DA, Caplan AI 1988 Proteoglycans synthesized during the cartilage-to-bone transition in developing chick embryos. Bone 9:225–233. Poole AR, Pidoux I, Rosenberg LC 1982 Role of proteoglycans in endochondral ossification: immunofluorescent localization of link protein and proteoglycan monomer in bovine fetal epiphyseal growth plate. J Cell Biol 92:249–260. Buckwalter JA 1983 Proteoglycan structure in calcifying cartilage. Clin Orthop Related Res 172:207–232. Buckwalter JA, Rosenberg LC, Ungar R 1987 Changes in proteoglycan aggregates during cartilage mineralization. Calcif Tissue Int 41:228–236. Schmid TM, Linsenmayer TF 1985 Immunohistochemical localization of short chain cartilage collagen (type X) in avian tissues. J Cell Biol 100:598–605. Stocum DL, Davis RM, Leger M, Conrad HE 1979 Development of the tibiotarsus in the chick embryo: Biosynthetic activities of histologically distinct regions. J Embryol Exp Morphol 54:155–170. Pacifici M, Golden EB, Iwamoto M, Adams SL 1991 Retinoic acid treatment induces type X collagen gene expression in cultured chick chondrocytes. Exp Cell Res 195:38–46. Grant WT, Sussman MD, Balian G 1985 A disulfide-bonded short chain collagen synthesized by degenerative and calcifying zones of bovine growth plate cartilage. J Biol Chem 260: 3798–3803. Kielty CM, Kwan APL, Holmes DF, Schor SL, Grant ME 1985 Type X collagen, a product of hypertrophic chondrocytes. Biochem J 227:545–554. Schmid TM, Linsenmayer TF 1985 Developmental acquisition of type X collagen in the embryonic chick tibiotarsus. Dev Biol 107:373–381. Gerstenfeld LC, Kelly CM, Von Deck M, Lian JB 1990 Effect of 1,25-dihydroxyvitamin D3 on induction of chondrocyte maturation in culture: Extracellular matrix gene expression and morphology. Endocrinology 126:1599–1609. Gerstenfeld LC, Kelly CM, Von Deck M, Lian JB 1990 Comparative morphological and biochemical analysis of hypertrophic, non-hypertrophic and l,25(OH)2D3-treated nonhypertrophic chondrocytes. Connect Tissue Res 24:29–39.

BARBARA D. BOYAN AND ZVI SCHWARTZ

33. Quarto R, Dozin B, Bonaldo P, Cancedda R, Colombatti A 1993 Type IV collagen expression is upregulated in the early events of chondrocyte differentiation. Development 117:245–251. 34. Roach HI, Erenpreisa J, Aigner T 1995 Osteogenic differentiation of hypertrophic chondrocytes involves asymmetric cell divisions and apoptosis. J Cell Biol 131:483–494. 35. Roach HI 1992 Trans-differentiation of hypertrophic chondrocytes into cells capable of producing a mineralized bone matrix. Bone Miner 19:1–20. 36. Aydelotte MB, Raiss RX, Caterson B, Kuettner KE 1992 Influence of interleukin-1 on the morphology and proteoglycan metabolism of cultured bovine articular chondrocytes. Connect Tissue Res 28:143–159. 37. Hauselmann HJ, Aydelotte MB, Schumacher BL, Kuettner KE, Gitelis SH, Thonar EJ 1992 Synthesis and turnover of proteoglycans by human and bovine adult articular chondrocytes cultured in alginate beads. Matrix 12:116–129. 38. McClure J, Bates GP, Rowston H, Grant ME 1988 A comparison of the morphological, histochemical and biochemical features of embryonic chick sternal chondrocytes in vivo with chondrocytes cultured in three-dimensional collagen gels. Bone Miner 3:235–247. 39. Benya PD, Shatter ID 1982 Dedifferentiated chondrocytes reexpress the differentiated collagen phenotype when cultured in agarose gels. Cell 30:215–224. 40. Suzuki F, Takase T, Takigawa M, Uchida A, Shimomura S 1981 Simulation of the initial stage of endochondral ossification: In vitro sequential culture of growth cartilage cells and bone marrow cells. Proc Natl Acad Sci USA 78:2368–2372. 41. Ali SY 1985 Calcification of cartilage. In: Hall BK (ed) Cartilage Structure, Function and Biochemistry. Academic Press, New York, pp. 343–378. 42. Vaananen HK, Morris DC, Anderson HC 1983 Calcification of cartilage matrix in chondrocyte cultures derived from rachitic rat growth plate cartilage. Metab Bone Dis Related Res 5:87–92. 43. Hancock RH, Schwartz Z, Swain LD, Ramirez V, Boyan BD 1990 Effect of dexamethasone on cartilage cell differentiation. J Dent Res 69:293. 44. Schwartz Z, Hancock RH, Dean DD, Brooks BP, Gomez R, Boskey AL, Balian G, Boyan BD 1995 Dexamethasone promotes von Kossa-positive nodule formation and increased alkaline phosphatase activity in costochondral chondrocyte cultures. Endocrine 3:351–360. 45. Nakahara H, Dennis JE, Bruder SP, Haynesworth SE, Lennon DP, Caplan AI 1991 In vitro differentiation of bone and hypertrophic cartilage from periosteal-derived cells. Exp Cell Res 195:492–503. 46. Nakahara H, Watanabe K, Sugrue SP, Olsen BR, Caplan AI 1990 Temporal and spatial distribution of type XII collagen in high cell density culture of periosteal-derived cells. Dev Biol 142:481–485. 47. Boyan BD, Schwartz Z, Swain LD, Carnes DL Jr, Zislis T 1988 Differential expression of phenotype by resting zone and growth region costochondral chondrocytes in vitro. Bone 9:185–194. 48. Boyan BD, Schwartz Z, Swain LD 1992 In vitro studies on the regulation of endochondral ossification by vitamin D. Crit Rev Oral Biol Med 3:15–30. 49. Chen P, Carrington JL, Paralkar VM, Pierce GF, Reddi AH 1992 Chick limb bud mesodermal cell chondrogenesis: Inhibition by isoforms of platelet-derived growth factor and reversal by recombinant bone morphogenetic protein. Exp Cell Res 200:110–117.

CHAPTER 33 Genomic and Nongenomic Regulation by 1,25(OH)2D3 and 24,25(OH)2D3 50. Boskey AL, Stiner D, Doty SB, Binderman I 1991 Requirement of vitamin C for cartilage calcification in a differentiating chick limb-bud mesenchymal cell culture. Bone 12:277–282. 51. O’Keefe RJ, Crabb ID, Puzas JE, Rosier RN 1989 Countercurrent centrifugal elutriation: High-resolution method for the separation of growth-plate chondrocytes. J Bone Joint Surg 71(A):607–620. 52. Dean DD, Muniz OE, Woessner JF Jr, Howell DS 1990 Production of collagenase and tissue inhibitor of metalloproteinases (TIMP) by rat growth plates in culture. Matrix 10:320–330. 53. Atkin I, Dean DD, Muniz OE, Agundez A, Castiglione G, Cohen G, Howell DS, Ornoy A 1992 Enhancement of osteoinduction by vitamin D metabolites in rachitic host rats. J Bone Miner Res 7:863–875. 54. Boskey AL, Wientroub S 1986 The effect of vitamin D deficiency on rat bone lipid composition. Bone 7:277–281. 55. Dean DD, Schwartz Z, Muniz OE, Gomez R, Swain LD, Howell DS, Boyan BD 1992 Matrix vesicles are enriched in metalloproteinases that degrade proteoglycans. Calcif Tissue Int 50:342–349. 56. Schmitz JP, Schwartz Z, Sylvia VL, Dean DD, Calderon F, Boyan BD 1996 Vitamin D3 regulation of stromelysin-1 (MMP-3) in chondrocyte cultures is mediated by protein kinase C. J Cell Physiol 168:570–579. 57. Dean DD, Schwartz Z, Schmitz JP, Muniz OE, Lu Y, Calderon FJ, Howell DS, Boyan BD 1996 Vitamin D regulation of metalloproteinase activity in matrix vesicles. Connect Tissue Res 35:331–336. 58. Dean DD, Schwartz Z, Schmitz JP, Muniz OE, Lu Y, Calderon FJ, Howell DS, Boyan BD 1996 Vitamin D regulation of metalloproteinase activity in matrix vesicles. Connect Tissue Res 35:331–336. 59. Klement BJ, Spooner BS 1994 Pre-metatarsal skeletal development in tissue culture at unit and microgravity. J Exp Zool 269:230–241. 60. Schwartz Z, Bonewald LF, Caulfield K, Brooks BP, Boyan BD 1993 Direct effects of transforming growth factor-/? on chondrocytes are modulated by vitamin D metabolites in a cell maturation-specific manner. Endocrinology 132:1544–1552. 61. Lyons KM, Pelton RW, Hogan BL 1989 Patterns of expression of murine Vgr-1 and BMP-2a RNA suggest that transforming growth factor-beta-like genes coordinately regulate aspects of embryonic development. Genes Dev 3:1657–1668. 62. Chang SC, Hoang B, Thomas JT, Vukicevic S, Luyten FP, Ryba NJ, Kozak CA, Reddi AH, Moos M Jr 1994 Cartilagederived morphogenetic proteins. New members of the transforming growth factor ? superfamily predominantly expressed in long bones during human embryonic development. J Biol Chem 269:28227–28234. 63. Amizuka N, Warshawsky H, Henderson JE, Goltzman D, Karaplis AC 1994 Parathyroid hormone-related peptide-depleted mice show abnormal epiphyseal cartilage development and altered endochondral bone formation. J Cell Biol 126:1611–1623. 64. Schwartz Z, Sela J, Ramirez V, Amir D, Boyan BD 1989 Changes in extracellular matrix vesicles during healing of rat tibial bone: A morphometric and biochemical study. Bone 10:53–60. 65. Wu LN, Genge BR, Lloyd GC, Wuthier RE 1991 Collagenbinding proteins in collagenase-released matrix vesicles from cartilage. Interaction between matrix vesicle proteins and different types of collagen. J Biol Chem 266:1195–1203. 66. Wuthier RE, Register TC 1985 Role of alkaline phosphatase, a polyfunctional enzyme, in mineralizing tissues. In: Butler WT

67. 68. 69.

70. 71. 72. 73.

74.

75.

76. 77. 78.

79. 80.

81. 82.

83.

84.

593

(ed) The Chemistry and Biology of Mineralized Tissues. Ebsco Media, Birmingham, pp. 113–124. Wuthier RE 1973 The role of phospholipids in biological calcification: Distribution of phospholipase activity in calcifying epiphyseal cartilage. Clin Orthop Related Res 90:191–200. Boskey AL, Reddi AH 1983 Changes in lipids during matrixinduced endochondral bone formation. Calcif Tissue Int 35: 549–554. Boskey AL, Posner AS, Lane JM, Goldberg MR, Cordelia DM 1980 Distribution of lipids associated with mineralization in the bovine epiphyseal growth plate. Arch Biochem Biophys 199:305–311. Boskey AL, Posner AS 1976 Extraction of a calciumphospholipid-phosphate complex from bone. Calcif Tissue Int 19:273–283. Boyan B, Ritter NM 1984 Proteolipid relationships in normal and vitamin D deficient chick cartilage. Calcif Tissue Int 36:332–337. Boskey AL 1981 Current concepts of the physiology and biochemistry of calcification. Clin Orthop Related Res 157:225–255. Buckwalter JA, Ehrlich MG, Armstrong AL, Mankin HJ 1987 Electron microscopic analysis of articular cartilage proteoglycan degradation by growth plate enzymes. J Orthop Res 5:128–132. Ingmar B, Wasteson A 1979 Sequential degradation of a chondroitin sulphate trisaccharide by lysosomal enzymes from embryonic-chick epiphyseal cartilage. Biochem J 179: 7–13. Amado R, Ingmar B, Lindahl U, Wasteson A 1974 Depolymerisation and desulphation of chondroitin sulphate by enzymes from embryonic chick cartilage. FEBS Lett 39:49–52. Schmid TM, Conrad HE. 1982 A unique low molecular weight collagen secreted by cultured chick embryo chondrocytes. J Biol Chem 257:12444–12450. Jimenez SA, Yankowski R, Reginato AM 1986 Quantitative analysis of type X collagen biosynthesis by embryonic chick sternal cartilage. Biochem J. 233:357–367. Yang F, Schwartz Z, Swain LD, Lee CC, Bowman BH, Boyan BD 1991 Alpha-2-HS-glycoprotein: Expression in chondrocytes and augmentation of alkaline phosphatase and phospholipase A2 activity. Bone 12:7–15. Malone JD, Teitelbaum SL, Griffin GL, Senior RM, Kahn AJ 1982 Recruitment of osteoclast precursors by purified bone matrix constituents. J Cell Biol 92:227–230. Grumbles RM, Roos BA, Brandi ML, Howard GA, Howell DS 1994 Rachitic growth plate regulation by 1,25(OH)2D3: A putative role for hepatocyte growth factor in angiogenesis. J Bone Miner Res 9(Suppl 1):S373. Seo EG, Schwartz Z, Dean DD, Norman AW, Boyan BD 1996 Preferential accumulation in vivo of 24R,25-dihydroxyvitamin D3 in growth plate cartilage of rats. Endocrine 5:147–155. Kato A, Seo E-G, Einhorn TA, Bishop JE, Norman AW 1998 Studies on 24R,25-dihydroxyvitamin D3: Evidence for a nonnuclear membrane receptor in the chick tibial fracture-healing callus. Bone 23:141–146. Schwartz Z, Brooks BP, Swain LD, Del Toro F, Norman AW, Boyan BD 1992 Production of 1,25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3 by growth zone and resting zone chondrocytes is dependent on cell maturation and is regulated by hormones and growth factors. Endocrinology 130:2495–2504. Pedrozo HA, Boyan BD, Mazock J, Dean DD, Gomez R, Schwartz Z 1999 TGF-β1 regulates 25-hydroxyvitamin D3

594

85.

86.

87.

88.

89.

90.

91. 92.

93.

94.

95.

96. 97. 98. 99.

100.

BARBARA D. BOYAN AND ZVI SCHWARTZ

1α- and 24-hydroxylase activity in cultured growth plate chondrocytes in a maturation-dependent manner. Calcif Tissue Int 64:50–56. Schwartz Z, Pedrozo HA, Sylvia VL, Gomez R, Dean DD, Boyan BD 2001 1α,25-(OH)2D3 regulates 25-hydroxyvitamin D3 24R-hydroxylase activity in growth zone costochondral growth plate chondrocytes via protein kinase C. Calcif Tissue Int 69:365–372. Hecht JT, Deere M, Putnam E, Cole W, Vertel B, Chen H, Lawler J 1998 Characterization of cartilage oligomeric matrix protein (COMP) in human normal and pseudoachondroplasia musculoskeletal tissues. Matrix Biol 17:269–278. Zaucke F, Dinser R, Maurer P, Paulsson M 2001 Cartilage oligomeric matrix protein (COMP) and collagen IX are sensitive markers for the differentiation state of articular primary chondrocytes. Biochem J 358:17–24. Balmain N, Hauchecorne M, Pike JW, Cuisinier-Gleizes P, Matlieu H 1993 Distribution and subcellular immunolocalization of 1,25-dihydroxyvitamin D3 receptors in rat epiphyseal cartilage. Cell Mol Biol 39:339–350. Tetlow LC, Woolley DE 2001 Expression of vitamin D receptors and matrix metalloproteinases in osteoarthritic cartilage and human articular chondrocytes in vitro. Osteoarthritis Cartilage 9:423–431. Berry JL, Farquharson C, Whitehead CC, Mawer EB 1996 Growth plate chondrocycte vitamin D receptor number and affinity are reduced in avian tibial dyschondroplastic lesions. Bone 19:197–203. Genever PG, Dickson IR 1996 Influence of vitamin D status on hyaluronan location in bone. Bone 18:429–435 Schmitz JP, Schwartz Z, Sylvia VL, Dean DD, Calderon F, Boyan BD 1996 Vitamin D3 regulation of stromelysin-1 (MMP-3) in chondrocyte cultures is mediated by protein kinase C. J Cell Physiol 168:570–579. Tetlow LC, Woolley DE 1999 The effects of 1α,25dihydroxyvitamin D930 on matrix metalloproteinase and prostaglandin E2 production by cells of the rheumatoid lesion. Arthritis Res 1:63–70. Dean DD, Schwartz Z, Muniz OE, Carreno MR, Maeda S, Howell DS, Boyan BD 2001 Effect of 1α,25-(OH)2D3 and 24R,25-(OH)2D3 on metalloproteinase activity and cell maturation in growth plate cartilage in vivo. Endocrine 14:311–323. Maeda S, Dean DD, Sylvia VL, Boyan BD, Schwartz Z 2000 Metalloproteinase activity in growth plate chondrocyte cultures is regulated by 1,25-(OH)2D3 and 24,25-(OH)2D3 and mediated through protein kinase C. Matrix Biol 20:87–97. Muller SA, Posner AS, Firschein HE 1966 Effect of vitamin D deficiency on the crystal chemistry of bone mineral. Proc Soc Exp Biol Med 121:844–846. Mathews JL, Martin JH, Sampson HW, Kunin AS, Roan JH 1970 Mitochondrial granules in the normal and rachitic epiphysis. Calcif Tissue Res 5:91–99. Brighton CT, Hunt RM 1978 The role of mitochondria in growth plate calcification as demonstrated in a rachitic model. J Bone Joint Surg Am 60A:630–639. Li YC, Amling M, Pirro AE, Priemel M, Meuse J, Baron R, Delling G, Demay MB 1998 Normalization of mineral ion homeostasis by dietary means prevents hyperparathyroidism, rickets, and osteomalacia, but not alopecia in vitamin D receptor-ablated mice. Endocrinology 139:4391–4396. Howell DS, Pita JC, Alvarez J 1976 Possible role of extracellular matrix vesicles in initial calcification of healing rachitic cartilage. Fed Proc 35:122–126.

101. Boyan BD, Schwartz Z, Carnes DL Jr, Ramirez V 1988 The effects of vitamin D metabolites on the plasma and matrix vesicle membranes of growth and resting cartilage cells in vitro. Endocrinology 122:2851–2860. 102. Swain LD, Schwartz Z, Boyan BD 1992 1,25-(OH)2D3 and 24,25-(OH)2D3 regulation of arachidonic acid turnover in chondrocyte cultures is cell maturation-specific and may involve direct effects on phospholipase A2. Biochim Biophys Acta 1136:45–51. 103. Irving JT, Wuthier RE 1968 Histochemistry and biochemistry of calcification with special reference to the role of lipids. Clin Orthop Related Res 56:237–260. 104. Cruess R, Clark I 1965 Alterations in the lipids of bone caused by hypervitaminosis A & D. Biochem J 96: 262–265. 105. Howell DS, Marquez JF, Pita JC 1965 The nature of phospholipids in normal and rachitic costochondral plates. Arth Rheum 8:1039–1046. 106. Kyeyune-Nyombi E, Lau KH, Baylink DJ, Strong DD 1991 11,25-Dihydroxyvitamin D3 stimulates bone alkaline phosphatase gene transcription and mRNA stabilization in human bone cells. Arch Biochem Biophys 291:316–325. 107. Schwartz Z, Schlader DL, Swain LD, Boyan BD 1988 Direct effects of 1,25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3 on growth zone and resting zone chondrocyte membrane alkaline phosphatase and phospholipase-A2 specific activities. Endocrinology 123:2878–2884. 108. Gay CV, Anderson RE, Leach RM 1985 Activities and distribution of alkaline phosphatase and carbonic anhydrase in the tibial dyschondroplastic lesion and associated growth plate of chicks. Avian Dis 29:812–821. 109. Sylvia VL, Schwartz Z, Schuman L, Morgan RT, Mackey S, Gomez R, Boyan BD 1993 Maturation-dependent regulation of protein kinase C activity by vitamin D3 metabolites in chondrocyte cultures. J Cell Physiol 157:271–278. 110. Roughley PJ, Dickson I 1980 Factors influencing proteoglycan size in rachitic-chick growth cartilage. Biochem J 185: 33–39. 111. Roughley PJ, Dickson IR 1986 A comparison of proteoglycan from chick cartilage of different types and a study of the effect of vitamin D on proteoglycan structure. Connect Tissue Res 14:187–197. 112. Carrino DA, Lidor C, Edelstein S, Caplan AI 1989 Proteoglycan synthesis in vitamin D-deficient cartilage: recovery from vitamin D-deficiency. Connect Tissue Res 19:135–147. 113. Reinholt FP, Engfeldt B, Heinegard D, Hjerpe A 1985 Proteoglycans and glycosaminoglycans of epiphyseal cartilage in florid and healing low phosphate, vitamin D deficiency rickets. Collagen Related Res 5:55–64. 114. Frame B, Parfitt AM 1978 Osteomalacia: Current concepts. Ann Intern Med 89:966–982. 115. Howard GA, Turner RT, Sherrard DJ, Baylink DJ 1981 Human bone cells in culture metabolize 25-hydroxyvitamin D3 to 1,25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3. J Biol Chem 256:7738–7740. 116. Turner RT, Puzas JE, Forte MD, Lester GE, Gray TK, Howard GA, Baylink DJ 1980 In vitro synthesis of lα,25dihydroxy-cholecalciferol and 24,25-dihydroxycholecalciferol by isolated calvarial cells. Proc Natl Acad Sci USA 77:5720–5724. 117. Reichel H, Koeffler HP, Bishop JE, Norman AW 1987 25-Hydroxyvitamin D3 metabolism by lipopolysaccharidestimulated normal human macrophages. J Clin Endocrinol Metab 64:1–9.

CHAPTER 33 Genomic and Nongenomic Regulation by 1,25(OH)2D3 and 24,25(OH)2D3 118. Reichel H, Koeffler HP, Norman AW 1987 25-Hydroxyvitamin D3 metabolism by human T-lymphotropic virustransformed lymphocytes. J Clin Endocrinol Metab 65: 519–525. 119. Schwartz Z, Brooks BP, Swain LD, Del Toro F, Norman AW, Boyan BD 1992 Production of 1,25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3 by growth zone and resting zone chondrocytes is dependent on cell maturation and is regulated by hormones and growth factors. Endocrinology 130:2495–2504. 120. Holtrop ME, Cox KA, Carnes DL, Holick MF 1986 Effects of serum calcium and phosphorus on skeletal mineralization in vitamin D–deficient rats. Am J Physiol 251:E234-E240. 121. Dekel S, Ornoy A, Sekeles E, Noff D, Edelstein S. 1979 Contrasting effects on bone formation and on fracture healing of cholecalciferol and of 1α-hydroxycholecalciferol. Calcif Tissue Int 28:245–251. 122. Ornoy A, Sekeles E, Cohen R, Edelstein S 1979 The role of vitamin D metabolites in calcification of chicken epiphyseal cartilage. In: Norman AW, Schaefer K, Harrath DV, Grigoleit HG, Coburn JW, DeLuca HF (eds) Vitamin D: Basic Research and Its Clinical Application. de Gruyter, Berlin, pp. 363–367. 123. Underwood JL, DeLuca HF 1984 Vitamin D is not directly necessary for bone growth and mineralization. Am J Physiol 246:E493-E498. 124. Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997 Targeted ablation of the vitamin D receptor: A animal model of vitamin D-dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94:9831–9835. 125. Howell DS, Blanco LN, Pita JC 1978 Further characterization of a nucleational agent in hypertrophic cell extracellular cartilage fluid. Metab Bone Dis Related Res 1:155–160. 126. Schwartz Z, Dean DD, Walton JK, Brooks BP, Boyan BD 1995 Treatment of resting zone chondrocytes with 24,25dihydroxyvitamin D3 [24,25(OH)2D3] induces differentiation into a l,25(OH)2D3-responsive phenotype characteristic of growth zone chondrocytes. Endocrinology 136:402–411. 127. Lidor C, Atkin I, Ornoy A, Dekel S, Edelstein S 1987 Healing of rachitic lesions in chicks by 24R,25-dihydroxycholecalciferol administered locally into bone. J Bone Miner Res 2:91–98. 128. St. -Arnaud R 1999 Targeted inactivation of vitamin D hydroxylases in mice. Bone 25:127–129. 129. St.-Arnaud R, Arabian A, Travers R, Barletta F, Raval-Pandya M, Chapin K, Depovere J, Mathieu C, Christakos S, Demay MB, Glorieux FH 2000 Deficient mineralization of intramembranous bone in vitamin D-24hydroxylase-ablated mice is due to elevated 1,25-dihydroxyvitamin D and not to the absence of 24,25-dihydroxyvitamin D. Endocrinology 141:2658–2666. 130. Ornoy A, Menczel J, Nebel L 1968 Alterations in the mineral composition and metabolism of rat fetuses and their placentas induced by maternal hypervitaminosis D2. Isr J Med Sci 4:827–832. 131. Ornoy A, Nebel L, Menczel Y 1969 Impaired osteogenesis of fetal long bones induced by maternal hypervitaminosis D2. Arch Pathol 87:563–571. 132. Kato Y, Shimazu A, Iwamoto M, Nakashima K, Koike T, Suzuki F, Nishii Y, Sato K 1990 Role of 1,25-dihydroxycholecalciferol in growth-plate cartilage: inhibition of terminal differentiation of chondrocytes in vitro and in vivo. Proc Natl Acad Sci USA 17:6522–6526 133. Atkin I, Pita JC, Ornoy A, Agundez A, Castiglione G, Howell DS 1985 Effects of vitamin D metabolites on healing

134.

135.

136.

137.

138. 139. 140.

141.

142.

143. 144.

145.

146.

147.

148.

149.

595

of low phosphate, vitamin D–deficient induced rickets in rats. Bone 6:113–123. Weinstein RS, Underwood JL, Hutson MS, DeLuca HF 1984 Bone histomorphometry in vitamin D–deficient rat infused with calcium and phosphorus. Am J Physiol 246: E499–E505. Harrison JE, Hitchman AJ, Jones G, Tarn CS, Heersche JN 1982 Plasma vitamin D metabolite levels in phosphorus deficient rats during the development of vitamin D deficient rickets. Metab Clin Exp 31:1121–1127. Miller SC, Halloran BP, DeLuca HF, Yamada S, Takayama H, Jee WS 1981 Studies on the role of 24-hydroxylation of vitamin D in the mineralization of cartilage and bone of vitamin D–deficient rats. Calcif Tissue Int 33:489–497. Sai H, Takatsuto S, Ikekawa N, Tanaka Y, Smith C, DeLuca HF 1984 Synthesis and biological activity of (22E,24R)- and (22E,24S)-1α,24-dihydroxy-22-dehydrovitamin D3. Chem Pharm Bull 32:3866–3872. Fine N, Binderman I, Somjen D, Earon Y, Edelstein S, Weisman Y 1985 Autoradiographic localization of 24R,25dihydroxyvitamin D3 in epiphyseal cartilage. Bone 6:99–104. Corvol MT, Ulmann A, Garabedian M 1980 Specific nuclear uptake of 24,25-dihydroxycholecaliferol, a vitamin D3 metabolite biologically active in cartilage. FEBS Lett 116:273–276. Balmain N, Hauchecorne M, Pike JW, Cuisinier-Gleizes P, Matlieu H 1993 Distribution and subcellular immunolocalization of 1,25-dihydroxyvitamin D3 receptors in rat epiphyseal cartilage. Cell Mol Biol 39:339–350. Plachot JJ, DuBois MB, Halpern S, Cournot-Witmer G, Garabedian M, Balsan S 1982 In vitro action of 1,25dihydroxycholecalciferol and 24,25-dihydroxycholecalciferol on matrix organization and mineral distribution in rabbit growth plate. Metab Bone Dis Related Res 4:135–142. Hinek A, Poole AR 1988 The influence of vitamin D metabolites on the calcification of cartilage matrix and the C-propeptide of type II collagen (chondrocalcin). J Bone Miner Res 3:421–429. Hinek A, Reiner A, Poole AR 1987 The calcification of cartilage matrix in chondrocyte culture. Studies of the C-propeptide of type II collagen (chondrocalcin). J Cell Biol 104:1435–1441. Takigawa M, Enomoto M, Shirai E, Nishii Y, Sazuki F 1988 Differential effects of 1α-25-dihydroxycholecalciferol and 24R,25-dihydroxycholecalciferol on the proliferation and the differentiated phenotype of rabbit costal chondrocytes in culture. Endocrinology 122:831–839. Corvol MT, Dumontier MF, Garabedian M, Rappaport R 1978 Vitamin D and cartilage. II. Biological activity of 25-hydroxycholecalciferol and 24,25- and 1,25-dihydroxycholecalciferol in cultured growth plate chondrocytes. Endocrinology 102:1269–1274. Suda S, Takahashi N, Shinki T, Horiuchi N, Yamaguchi S, Yoshiki S, Enomoto S, Suda T 1985 1α,25-Dihydroxyvitamin D3 receptors and their action in embryonic chick chondrocytes. Calcif Tissue Int 37:82–90. Balmain N, Tisserand-Jochem E, Thomasset M, CuisinierGleizes P, Mathieu H 1986 Vitamin D-dependent calciumbinding protein (CaBP-9k) in rat growth cartilage. Histochemistry 84:161–168. Zhou XY, Dempster DW, Marion SL, Pike JW, Haussler MR, Clemens TL. 1986 Bone vitamin D–dependent calciumbinding protein is localized in chondrocytes of growth plate cartilage. Calcif Tissue Int 38:244–247. Somjen D, Kaye AM, Binderman I 1984 24,25-Dihydroxyvitamin D stimulates creatine kinase BB activity in chick cartilage cells in culture. FEBS Lett 167:281–284.

596 150. Boyan BD, Schwartz Z, Carnes DL Jr, Ramirez V 1988 The effects of vitamin D metabolites on the plasma and matrix vesicle membranes of growth and resting cartilage cells in vitro. Endocrinology 122:2851–2860. 151. Schwartz Z, Boyan BD 1988 The effects of vitamin D metabolites on phospholipase A2 activity of growth zone and resting zone cartilage cells in vitro. Endocrinology 122:2191–2198. 152. Swain LD, Schwartz Z, Caulfield K, Brooks BP, Boyan BD 1993 Nongenomic regulation of chondrocyte membrane fluidity by 1,25-(OH)2D3 and 24,25-(OH)2D3 is dependent on cell maturation. Bone 14:609–617. 153. Schwartz Z, Schlader DL, Ramirez V, Kennedy MB, Boyan BD 1989 Effects of vitamin D metabolites on collagen production and cell proliferation of growth zone and resting zone cartilage cells in vitro. J Bone Miner Res 4:199–207. 154. Langston GG, Swain LD, Schwartz Z, Del Toro F, Gomez R, Boyan BD 1990 Effect of 1,25(OH)2D3 and 24,25(OH)2D3 on calcium ion fluxes in costochondral chondrocyte cultures. Calcif Tissue Int 47:230–236. 155. Schwartz Z, Swain LD, Ramirez V, Boyan BD 1990 Regulation of arachidonic acid turnover by 1,25-(OH)2D3 and 24,25-(OH)2D3 in growth zone and resting zone chondrocyte cultures. Biochim Biophys Acta 1027:278–286. 156. Boskey AL, Stiner D, Doty SB, Binderman I, Leboy PS 1992 Studies of mineralization in tissue culture: Optimal conditions for cartilage calcification. Bone Miner 16:11–36. 157. Hale LV, Kemick ML, Wuthier RE 1986 Effect of vitamin D metabolites on the expression of alkaline phosphatase activity by epiphyseal hypertrophic chondrocytes in primary cell culture. J Bone Miner Res 1:489–495. 158. Schwartz Z, Sylvia VL, Larsson D, Nemere I, Casasola D, Dean DD, Boyan BD 2003 1α,25(OH)2D3 regulates chondrocyte matrix vesicle protein kinase C directly via G-protein dependent mechanisms and indirectly via incorporation of PKC during matrix vesicle biogenesis. J Biol Chem 5: 11828–11837. 159. Swain LD, Schwartz Z, Boyan BD 1992 1,25-(OH)2D3 and 24,25-(OH)2D3 regulation of arachidonic acid turnover in chondrocyte cultures is cell maturation-specific and may involve direct effects on phospholipase A2. Biochim Biophys Acta 1136:45–51. 160. Schwartz Z, Langston GG, Swain LD, Boyan BD 1991 Inhibition of 1,25-(OH)2D3- and 24,25-(OH)2D3-dependent stimulation of alkaline phosphatase activity by A23187 suggests a role for calcium in the mechanism of vitamin D regulation of chondrocyte cultures. J Bone Miner Res 6:709–718. 161. Schwartz Z, Knight G, Swain LD, Boyan BD 1988 Localization of vitamin D3–responsive alkaline phosphatase in cultured chondrocytes. J Biol Chem 13:6023–6026. 162. Schwartz Z, Sylvia VL, Dean DD, Boyan BD 1996 The synergistic effect of TGFβ and 24,25-(OH)2D3 on resting zone chondrocytes is metabolite-specific and mediated by PKC. Connect Tissue Res 35:101–106. 163. Boyan BD, Sylvia VL, Dean DD, Schwartz Z 1994 Nongenomic effects of vitamin D. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D: A Pluripotent Steroid Hormone: Structural Studies, Molecular Endocrinology and Clinical Applications. de Gruyter, Berlin, pp. 333–340. 164. Kimelberg HK 1975 Alterations in phospholipiddependent (Na+,K+)-ATPase activity due to lipid fluidity. Effects of cholesterol and Mg2+. Biochim Biophys Acta 413: 143–156.

BARBARA D. BOYAN AND ZVI SCHWARTZ

165. Shinitzky M, Inbar M 1974 Difference in microviscosity induced by different cholesterol levels in the surface membrane lipid layer of normal lymphocytes and malignant lymphoma cells. J Mol Biol 85:603–615. 166. Schwartz Z, Swain LD, Kelly DW, Brooks BP, Boyan BD 1992 Regulation of prostaglandin E2 production by vitamin D metabolites in growth zone and resting zone chondrocyte cultures is dependent on cell maturation. Bone 13:395–401. 167. Sylvia VL, Del Toro F, Dean DD, Hardin RR, Schwartz Z, Boyan BD 2001 Effects of 1α,25-(OH)2D3 on rat growth zone chondrocytes are mediated via cyclooxygenase-1 and phospholipase A2. J Cell Biochem Suppl. 36:32–45. 168. Sylvia VL, Schwartz Z, Ellis EB, Helm SH, Gomez R, Dean DD, Boyan BD 1996 Nongenomic regulation of protein kinase C isoforms by the vitamin D metabolites 1α,25(OH)2D3 and 24R,25-(OH)2D3. J Cell Physiol 167:380–393. 169. Schwartz Z, Ehland H, Sylvia VL, Larsson D, Bingham V, Hardin RR, Lopez D, Dean DD, Boyan BD 2003 1α,25(OH)2D3 and 24R,25(OH)2D3 modulate growth plate chondrocyte physiology via PKC-dependent phosphorylation of ERK1/2 MAP kinase. Endocrinology, 2775–2786. 170. Boyan BD, Sylvia VL, McKinney N, Schwartz Z 2003 Membrane-medicated Actions of vitamin D metabolites are retained in growth plate cartilage cells from vitamin D receptor knockout mice. J Cell Biochem. 171. Boyan BD, Posner GH, Greising DM, White MC, Sylvia VL, Dean DD, Schwartz Z 1997 Hybrid structural analogues of 1,25-(OH)2D3 regulate chondrocyte proliferation and proteoglycan production as well as protein kinase C through a nongenomic pathway. J Cell Biochem 4:457–470. 172. Greising DM, Schwartz Z, Posner GH, Sylvia VL, Dean DD, Boyan BD 1997 A-ring analogues of 1,25(OH)2D3 with low affinity for the vitamin D receptor modulate chondrocytes via membrane effects that race dependent on cell maturation. J Cell Physiol 3:357–367. 173. Berdal A, Mesbah M, Papagerakis P, Nemere I 2003 Putative membrane receptor for 1,25(OH)2 vitamin D3 in human mineralized tissues during prenatal development. Connect Tissue Res 1:136–140. 174. Pedrozo HA, Schwartz Z, Rimes S, Sylvia VL, Nemere I, Posner GH, Dean DD, Boyan BD 1999 Physiological importance of the 1,25-(OH)2D3 membrane receptor and evidence for a membrane receptor specific for 24,25-(OH)2D3. J Bone Miner Res 14:856–867. 175. Nemere I, Schwartz Z, Pedrozo H, Sylvia VL, Dean DD, Boyan BD 1998 Identification of a membrane receptor for 1,25-dihydroxy vitamin D3 which mediates rapid activation of protein kinase C. J Bone Miner Res 13:1353–1359. 176. Schwartz Z, Shaked D, Hardin RR, Gruwell S, Dean DD, Sylvia VL, Boyan BD 2003 1α,25(OH)2D3 causes a rapid increase in phosphatidylinositol-specific PLC-β activity via phospholipase A2 dependent production of lysophospholipid. Steroids 68:423–437. 177. Helm SH, Sylvia VL, Harmon T, Dean DD, Boyan BD, Schwartz Z 1996 24,25-(OH)2D3 regulates protein kinase C through two distinct phospholipid-dependent mechanisms. J Cell Physiol 169:509–521. 178. Luo T, Luo Y, Vallano ML 1995 Arachidonic acid, but not sodium nitroprusside, stimulates presynaptic protein kinase C and phosphorylation of GAP-43 in rat hippocampal slices and synaptosomes. J Neurochem 64:1808–1818. 179. Boyan BD, Sylvia VL, Curry D, Chang Z, Dean DD, Schwartz Z 1998 Arachidonic acid is an autocoid mediator of the differential action of 1,25-(OH)2D3 and 24,25-(OH)2D3 on growth plate chondrocytes. J Cell Physiol 3:516–524.

CHAPTER 33 Genomic and Nongenomic Regulation by 1,25(OH)2D3 and 24,25(OH)2D3 180. Sylvia VL, Del Toro F Jr, Hardin RR, Dean DD, Boyan BD, Schwartz Z 2001 Characterization of PGE2 receptors (EP) and their role as mediators of 1α,25-(OH)2D3 effects on growth zone chondrocytes. J Steroid Biochem Mol Biol 78:261–274. 181. Schwartz Z, Sylvia VL, Luna MH, DeVeau P, Whetstone R, Dean DD, Boyan BD 2001 The effect of 24R,25-(OH)2D3 on protein kinase C activity in chondrocytes is mediated by phospholipase D whereas the effect of 1α,25-(OH)2D3 is mediated by phospholipase C. Steroids 66:683–694. 182. Schwartz Z, Sylvia VL, Curry D, Luna M, Dean DD, Boyan BD 1999 Arachidonic acid directly mediates the rapid effects of 24,25-(OH)2D3 via protein kinase C and indirectly through prostaglandin production in resting zone chondrocytes. Endocrinology 140:2991–3002. 183. Del Toro F, Jr., Sylvia VL, Schubkegel SR, Campos R, Dean DD, Boyan BD, Schwartz Z 2000 Characterization of PGE2 receptors (EP) and their role in 24,25-(OH)2D3-mediated effects on resting zone chondrocytes. J Cell Physiol 182:196–208. 184. Sylvia VL, Schwartz Z, Holmes SC, Dean DD, Boyan BD 1997 24,25-(OH)2D3 regulation of matrix vesicle protein kinase C occurs both during biosynthesis and in the extracellular matrix. Calcif Tissue Int 61:313–321. 185. Sela J, Schwartz Z, Swain LD, Boyan BD 1992 The role of matrix vesicles in calcification. In: Bonucci E (ed) Calcification in Biological Systems. CRC Press, Boca Raton, FL, pp. 73–105.

597

186. Dean DD, Schwartz Z, Muniz OE, Gomez R, Swain LD, Howell DS, Boyan BD 1992 Matrix vesicles contain metalloproteinases that degrade proteoglycans. Bone Miner 17:172–176. 187. Boskey AL, Boyan BD, Doty SB, Feliciano A, Greer K, Weiland D, Swain LD, Schwartz Z 1992 Studies of matrix vesicle-induced mineralization in a gelatin gel. Bone Miner 17:257–262. 188. Boyan BD, Schwartz Z, Park-Snyder S, Dean DD, Yang F, Twardzik D, Bonewald LF 1994 Latent transforming growth factor-β is produced by chondrocytes and activated by extracellular matrix vesicles upon exposure to 1,25(OH)2D3. J Biol Chem 269:28374–28381. 189. Pedrozo HA, Schwartz Z, Gomez R, Ornoy A, Xin-Sheng W, Dallas SL, Bonewald LF, Dean DD, Boyan BD 1998 Growth plate chondrocytes store latent transforming growth factor (TGF)-β1 in their matrix through latent TGF-β1 binding protein-1. 190. Pedrozo HA, Schwartz Z, Mokeyev T, Ornoy A, Xin-Sheng W, Bonewald LF, Dean DD, Boyan BD 1999 Vitamin D3 metabolites regulate LTBP1 and latent TGF-β1 expression and latent TGF-β1 incorporation in the extracellular matrix of chondrocytes. J Cell Biochem 1:151–165. 191. Maeda S, Dean DD, Gay I, Schwartz Z, Boyan BD 2001 Activation of latent transforming growth factor β1 by stromelysin 1 in extracts of growth plate chondrocyte-derived matrix vesicles. J Bone Miner Res 7:1281–1290.

CHAPTER 34

Dento-alveolar Bone Complex and Vitamin D ARIANE BERDAL, ISABELL BAILLEUL ISABELLE BAILLEUL-FORESTIER, JEAN-LUC DAVIDEAU, AND FRÉDÉRIC LÉZOT INSERM E110, Université Paris VII, IFR58, Institut Biomédical des Cordeliers, Paris, France

I. Introduction II. Dental and Periodontal Formation and Functions III. Vitamin D, Dental and Periodontal Development, and Biomineralization

IV. Conclusion References

I. INTRODUCTION

II. DENTAL AND PERIODONTAL FORMATION AND FUNCTIONS

The orofacial skeleton provides an unique opportunity to investigate different systems of elaboration of mineralized tissues under the control of vitamin D. Similar to bone cells, the cells devoted to tooth formation have been shown to express the nuclear vitamin D receptor (VDR) gene and membrane-associated rapid response steroid binding protein (MAARS) (Fig. 1) and to be regulated by the hormone (1,25-dihydroxyvitamin D) throughout their life cycle [1–6]. Several target genes have been identified that play a role jointly in bone and tooth formation at the intial differentiation stage, Msx2 [7,8], in the later deposition and biomineralization of matrix stage, osteocalcin [8,9], osteopontin [10] and bone sialoprotein [10], and two calciproteins, calbindin-D9k and calbindin-D28k [11]. Dental matrices contain several unique proteins that are vitamin D–dependent (amelogenin, enamelin, and amelogenin [5,6]). On the other hand, tooth germ actually provides an exemplary model system for understanding the genetic basis for skeletal patterning. Notably, muscle segment homeobox genes and distalless homeobox genes have been analyzed in detail over the past ten years in mineralized tissues (reviewed in [12] and Fig. 2). The encoded transcription factors are proposed to control the differential activity of osteoblasts depending on the anatomical site [13] by acting on genes such as osteocalcin and others thereby contributing to skeletal morphogenesis [12]. These studies and numerous clinical and experimental data lead to the concept that tooth and periodontium are associated as a physiological entity, termed the dento-alveolar bone complex. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

A. Morphogenesis and Cell Differentiation Odontogenesis involves a sequence of cell communications [14–16] that determines (1) morphogenesis, i.e., acquisition of distinctive morphotypes such as incisors, canines, premolars, and molars inside their site-specific alveolar bone; (2) differentiation of ameloblast, odontoblast, cementoblast, and bone cells; and (3) matrix secretion and biomineralization. Mesenchymal cells are derived from neural crest in a fashion identical to the vast majority of craniofacial bone cells and in contrast to the axial and appendicular skeleton of mesodermal origin [17]. Several signaling cascades have been delineated in tooth germ. Such is the case for BMP4, Shh, and FGF8. Their epithelial production drives nested expression of Msx and Dlx homeogenes inside the dental mesenchyme. The resulting combinatorial expression may trigger site-specific morphogenesis of dental units in rodents [14,15]. Such a function for Msx and Dlx homeogenes is further supported by the phenotype of human mutations, involving the MSX1 gene in tooth agenesis [18] and the DLX3 gene in the trichodento-osseous syndrome [19]. Interestingly, dental phenotypes are associated with major disturbances in craniofacial morphogenesis while the rest of the skeleton is unaffected (reviewed in [20]). Terminal differentiation of ameloblasts [21], odontoblasts [22], and cementoblasts [23] is characterized by cytological and functional modifications. Each cell type shows a distinct phenotype and elaborates a diverse mineralized tissue: enamel, dentin, and cementum. Copyright © 2005, Elsevier, Inc. All rights reserved.

600

ARIANE BERDAL

ET AL .

A

Dentine

Predentine

Am

Odontoblasts

N

B Am

FIGURE 1 Distribution of MARRS protein in human odontoblasts. A 1,25(OH)2D3 membrane-associated rapid response steroid binding protein (MARRS) has been isolated from basolateral membrane of chick intestine. Antibodies raised to the N-terminal sequence by the group of Ilka Nemere enabled functional investigations in cartilage and its demonstration in various species (chicken, rat, carp, and humans) and tissues (enterocytes, brain, kidney, cartilage, bone, and dental cells [4]). This protein has been associated with rapid, nongenomic effects of 1,25(OH)2D3 involving protein kinase C, phospholipase C, AMPc, IP3, and diacylglycerol modulations and calcium fluxes. In dental cells, MARRS protein is exclusively expressed in cells specialized in matrix deposition and mineralization, i.e., ameloblasts and odontoblasts [4]. Such a specificity contrasts with nuclear VDR which is expressed more widely [2]. The ameloblast specificity corresponds with the rapid (1 hour) over-expression of nuclear VDR specifically in ameloblasts [1]. (See color plate).

B. Distinctive Characters of Mineralized Tissues Forming the Dental Crown Tooth crown is composed of a central core of dentin covered by acellular enamel [21]. Dentin shares matrix components with bone and cementum, notably type I collagen and numerous noncollagenous protein species [22]. Dentin morphology is very different from bone and is characterized by the presence of parallel tubules that host the elongated cellular processes of odontoblasts. Unlike the osteoblast, polarized odontoblasts are not embedded inside extracellular matrix but rather arranged in a pseudoepithelial layer. Odontoblasts secrete distinct matrix proteins at various levels of the secretory pole. Thereafter, exported proteins selfassemble, and ordered nucleation and growth of hydroxyapatite crystals occurs abruptly at the mineralization front. Several noncollagenous proteins are proposed to coordinate biomineralization, as they appear to be exported at the level of the mineralization front [22]. Dentin phosphoproteins (DPPs), the major

C

Am

D

PLD

D

FIGURE 2 Dlx2 expression in epithelial dental cells. The data illustrate the morphogenetic specificity of Dlx and Msx homeogenes in mineralized tissues, as reviewed in [12]. Transgenic mice carrying a construct containing 2.7 kb Dlx2 promoter and lacZ reporter gene were used to map Dlx2 expression in dental mineralized tissues of postnatal mice [43,57]. Enamel formation is produced in three sequential steps: (A) a prescretion stage where ameloblasts differentiate; (B) a secretion stage where enamel matrix is deposited and partially mineralized; (C) a maturation stage where ameloblasts resorb matrix proteins and hypermineralization is performed [21]. Matrix protein expression culminates during the secretion phase [55]. Interestingly, Dlx2 showed a reversed expression, suggesting possible interactions between this transcription factor and the amelogenin gene promoter [58] (D) epithelial cells also participate in the formation of cementum and involve Dlx2. (See color plate).

601

CHAPTER 34 Dento-alveolar Bone Complex and Vitamin D

noncollagenous dentin proteins, are characterized by an exceptionally high degree of phosphorylation (45–50% of phosphoserine residues). Several experiments in vitro support the notion that they play a dual role in initiation and control of the final shape of hydroxyapatite crystals [22], depending on their binding to type I collagen. The dentin sialoprotein DSP is the second most abundant dentin matrix noncollagenous protein (5–8%). Similar to bone sialoproteins, DSP is rich in glutamine, aspartic acid, serine, and glycine [24]. It also has a high carbohydrate content [22]. DPPs and DSP may be cleavage products expressed from a single transcript coded by a gene on chromosome 4, termed DSPP gene [24]. Several mutations of DSPP gene have been identified in dentinogenesis imperfecta where major defects in dentin biomineralization occur [25]. The phenotype of DSPP gene null mutation [26] further supports the concept that this gene is instrumental in dentinogenesis. DSP and DPPs are thought to bind to a collagen scaffolding and cooperate in the regulation of hydroxyapatite formation through their high calcium binding properties [22]. The same may be true for all bone noncollagenous proteins also present in dentin (bone sialoprotein, osteopontin, osteonectin, DMP1, MEPE, and BAG75) [27]). The genes of the vast majority of dentin and bone matrix proteins are grouped inside a cluster called SIBLING [28] in the vicinity of two enamel protein genes (amelin/ ameloblastin/sheathlin and enamelin). Ameloblasts produce enamel, which is an exclusive example of epithelium-derived mineralized and acellular

Cementum

Alveolar

Periodontal

bone

ligament

Cementoblasts

Dentine

Odontoblasts

FIGURE 3 In situ hybridization of nuclear VDR is mandibular first molar of 15-day-old rat. The relative expression levels of VDR transcripts are different depending on the cell-type and stage in root formation, as established previously for the tooth crown [2]. Osteoblasts and osteocytes express more VDR than dental cells. (See color plate).

tissue in vertebrates. This superficial hypermineralized barrier contains a packed network of hydroxyapatite crystals assembled in a prismatic structure [21]. In its mature form enamel is almost devoid of organic matrix [29]. In contrast to mesenchymal skeleton formation and biomineralization, which occur concomitantly, amelogenesis is a biphasic process. During the secretion stage, enamel matrix is synthesized and exported while crystals grow along their c-axis. When the full thickness of acellular matrix is reached, the maturation stage involves proteolysis of matrix proteins. Completion of hydroxyapatite crystal growth is performed along the a- and b-axes. Amelogenins constitute 90% of the enamel proteins at the secretion stage. In contrast, mature enamel is nearly devoid of amelogenins, which are cleaved except in pathological conditions, notably in a porcine model of vitamin D pseudodeficiency rickets [30]. Furthermore, they show numerous isoforms, in relation with the existence of alternative transcripts and two distinct proteolysis pathways [29]. Twelve mutations of the chromosome X amelogenin gene have been reported in an enamel-specific genetic disorder: the X-linked amelogenesis imperfecta [31]. A null mutant for amelogenin gene confirms its functional relevance in amelogenesis [32]. Other enamel proteins include amelin/ameloblastin/sheathlin [33] and enamelin [34]. Interestingly, their genes are located on human chromosome 4 inside/in the vicinity of the SIBLING cluster [28] and close to the locus identified for an autosomal form of amelogenesis imperfecta in a Swedish kindred relied to enamelin gene mutation [34]. The dichotomy between dentin- and enamel-specific structural proteins/genes has been questioned by the identification of enamel proteins in dentin and odontoblasts (amelogenins [6], amelin/ameloblastin/sheathlin [35], and DSPP mRNAs in early ameloblasts [36]). These “structural” proteins play a role in cell signaling and presently merge as another biochemical group of bone morphogenetic proteins. Their bone-including capacities were established by in vivo and in vitro experiments [37]. They are proposed to play a role in the epithelialmesenchymal cross-talks that control ameloblast and odontoblast differentiation [38]. Based on these findings [39], enamel proteins are used for clinical purposes to access alveolar bone regeneration.

C. Distinctive Characters of Mineralized Tissues Forming the Dental Root and Periodontium Tooth root is composed of a central core of dentin covered by acellular and cellular cementum [16]. The attachment of tooth to alveolar bone is realized via the

602 periodontal ligament in which fibroblasts produce extrinsic collagen fibers anchored inside cementum and alveolar bone. This dense conjunctive tissue also provides progenitor cementoblasts on one side and osteoblast and osteoclast progenitors for the alveolar bone on the opposite side. In vitro culture of human periodontal ligament cells (HPLCs) allows osteoblastlike differentiation with the expression of alkaline phosphatase, osteocalcin, bone nodule formation [40–42]. HPLCs produce several effectors that are proposed to regulate alveolar bone and cementum cell homeostasis (IGF-1, TGFβ1 [41], and RANK-L and osteoprotegerin [42]). Cementum is not physiologically remodeled, except in primary dentition during tooth replacement [16]. Some epithelial cells are present in the vicinity of the cementum-forming side, producing enamel proteins (ameloblastin/amelin/sheathlin [43]), whose inducing role for osteoblast differentiation is suspected and used in periodontal regeneration [22]. Cementum shares so many matrix components with bone, including the majority of collagens and numerous noncollagenous proteins [44], notably osteopontin and bone sialoprotein [10], that it is in fact rather difficult to specify a cementoblast phenotype. A cementum attachment protein CAP appears to be cementumspecific [44]. Alveolar bone shows an exceptional lability, illustrated by its involution following tooth extraction. The specific physiology of alveolar bone, per se, is not well characterized (reviewed in [45]). Site-specific expression of several transcription factors [46] is proposed to control locally osteoblast function and renewal (reviewed in [12]).

III. VITAMIN D, DENTAL AND PERIODONTAL DEVELOPMENT, AND BIOMINERALIZATION Vitamin D is intimately involved in tooth formation and mineralization. Rachitic teeth constitute a recognizable physio-pathological entity (reviewed in [47]). Indeed, all developmental defects are irreversibly registered inside these nonremodeling tissues. Tooth morphology is abnormal in relation to two distinct effects of vitamin D on odontogenesis: altered morphogenesis and histogenesis. Dental crown shows an undulated enamel–dentin junction and dysmorphology culminating at the cusp tips. (Lezot et al., submitted). The process of mineralized tissue formation is also disturbed. Various experimental studies on rats, dogs, pigs (for review, see [47]) and transgenic VDR −/− mice have shown that, like osteogenesis, bioinactivation of vitamin D affects dentinogenesis, amelogenesis, and cementogenesis [48]. Predentin appears to be widened and mineralization foci

ARIANE BERDAL

ET AL .

FIGURE 4 Dental phenotype of a clinical case of nutritional vitamin D–deficiency rickets. Buccal view of a 7-year-old girl with nutritional rickets diagnosed and treated at 16 months. Major enamel hypoplasia appears on central maxillar and mandibular incisors and lateral mandibular incisors. Cervical enamel is apparently normal, its mineralization provided during the treatment for rickets. (See color plate).

do not coalesce in dentin, inducing the occurence of globular dentin. Enamel dysplasia of various degrees with external pits and grooves as well as enamel hypomaturation and hypomineralization defects are also described. Cementum is hypomineralized [49]. In clinical reports, dental defects have been associated with nutritional and pseudovitamin D deficiency, and also in related disorders such as familial hypophosphatemia. The occurrence of spontaneous periodontal abcess is also reported. The physiopathological understanding of rachitic teeth has been progressively delineated through (1) knowledge of VDR expression and more recently of MAARS protein in the dental cell life cycle, (2) the recognition of dental cells as target cells for 1,25-dihydroxyvitamin D by demonstration of calbindin regulation in ameloblasts and odontoblasts in vivo, (3) the investigation of tooth-specific matrix proteins in various experimental models of vitamin D bioinactivation, (4) the identification of a cross-talk between Msx and Dlx homeogenes and vitamin D regulation during cell communications (which raises the question of vitamin D impact on bone morphogenesis), and (5) the energy merging concept in geriatric dentistry that such dental and periodontal alterations result in periodontal disease and tooth loss.

A. Vitamin D–Dependent Molecules in Murine Epithelial and Mesenchymal Dental Cells A major pitfall in the molecular understanding of vitamin D action on dental cells and dental research in general is the inability to expand dental cells in vitro. Stringent criteria of differentiation limit the validation of experimental model system, i.e., cell elongation, polarization, phenotypic expression of a set of markers, and the formation of spatially organized mineralized tissues. Some studies, however, have been performed with

Proteins

Ameloblasts

DIFFERENTIATION

TRANSITION

Transition G1 to G0

Apoptosis

9 k/28 k

AMEL

mRNAs

9 k/28 k

AMEL

GFE,M

GFE AMEL DSPP

Extracellular matrix GFE,M Proteins mRNAs

MATURATION

SECRETION

DSPP

AMEL

Ca

pH

BIOMINERALIZATION BARRIER GFM SECRETION

PDs OC / ON / BSP

DIFFERENTIATION SECRETION

Proteins mRNAs

Odontoblasts

PDs = Coll. I / DPPs / DSP / DMP1 /

PDs = Coll. I / OC / ON / BSP Osteoblasts

DIFFERENTIATION

FIGURE 5 Schematic representation of relative quantities of mRNAs and proteins shown to be vitamin D–dependent in dental cells. The life cycle of ameloblasts, odontoblasts, and osteoblasts are summarized. The spatial distribution of cell differentiation steps is easily followed in the rodent incisor (reviewed in [21]), aligned along the longitudinal anatomical axis. Such an experimental model enables the study of mRNA and protein expression pattern related to vitamin D and successive stages of enamel, dentine and bone formation. 9k, Calbindin-D9k; 28k, calbindin-D28k; AMEL, amelogenin; DSPP, dentin sialophosphoprotein; GFE, growth factors secreted by epithelial cells; GFM, growth factors secreted by mesenchymal cells; DPPs, dentin phosphoproteins; DSP, dentin sialoproteins which are the cleavage products of the parent DSPP protein; DMP1, dentin matrix protein 1; OC, osteocalcin; ON, osteonectin; BSP, bone sialoprotein. Data were compiled from various sources [1,2,6,11,22,36]. Terminal differentiation of ameloblasts and odontoblasts is associated with the expression of a set of genes including the calbindins and numerous matrix protein genes. Vitamin D affects their expression differentially in ameloblasts and odontoblasts [5,6]. This is associated with impaired biomineralization of dentin and enamel. Our hypothesis is that epithelial mesenchymal exchange of growth factors is maintained by an altered biomineralization process, visualized by sustained enamel proteins expressed in the odontoblasts and dentin protein in the ameloblasts. This situation is not specific for vitamin D action in dental cells but rather is related to hypomineralization.

604 dental clones and primary cultures [41,50–52]. Since many observed effects of vitamin D are suspected to rely on epithelial–mesenchymal cross-talk, such communications are not recapitulated in these clones. Organotypic culture is frequently used in developmental biology [14,15]. But, the small sample size restricts experimental approaches to in situ studies and is not usable for endocrinologic investigations including case-control, dose-effect groups. The experimental strategy set up for the study of vitamin D is therefore based on in vivo sampling, which provided a reasonable amount of biological material from rats and mice [1]. The continuously erupting incisor of the rodent mandible enables microdissection and independent study of epithelial and mesenchymal dental cells. The data show that in vitamin D–deficient rats, a single injection of 1,25(OH)2D3 induces changes in steady-state mRNA levels of both calbindins in enamel and calbindin-D28k only in the dental mesechyme. The concomitant increase of VDR mRNA [2] and protein [1] further supports that variations of calbindin expression are likely related to the genomic action of 1,25(OH)2D3. A significant downregulation of calbinin-D28k in the dental mesenchyme was also obtained in VDR-null mutant VDR −/− hypocalcemic mouse incisor mesenchyme. Dentin and enamel defects are associated with the down-regulation of calbindins in dental cells. But, the exact role of these proteins is still unclear in dental cells.

B. Vitamin D and Dental Crown In contrast, the key role of tissue-specific matrix proteins in dental formation is well established (see section II,B). The dental phenotype of vitamin D deficiency and VDR −/− mice mimics, in part amelogenesis imperfecta where enamel-specific genes are mutated [31,34]. The striking features in rachitic enamel are an increased volume of interprismatic area that results in a reduced size of the prisms and hypomineralization. The prismatic structure of enamel reflects heterogenous orientation of crystals and may be guided by enamel proteins. Amelogenins are proposed to form nanospheres which cooperatively guide calcium and phosphate interactions and coordinate the initiation of apatite crystals formation. The term sheathlin (also amelin/ameloblastin) clearly indicates that this protein is involved in the enamel sheaths of the interprismatic boundary. It may therefore contribute to the second level of enamel architecture, prism morphogenesis. This rachitic morphology was indeed associated with an increased expression level of sheathlin/ameloblastin/ amelin and decreased expression levels of amelogenins in the ameloblasts [5]. The last enamel protein,

ARIANE BERDAL

ET AL .

enamelin, is down-regulated in the same context. The selective impact of vitamin D on enamel-specific mRNAs was confirmed by the induction of these mRNAs after a single injection of 1,25(OH)2D3. Therefore, the morphoregulatory role of vitamin D on enamel is converted with a selective up- and downregulation of enamel-specific gene expression. Rachitic dentin shows widened predentin and uncoalesced foci of mineralization. The candidate gene for dentin defects in DSPP, which does not appear to be modulated by vitamin D [5]. Such cellular data were previously suggested by immunocytochemical findings which show consistent expression of DPP protein in contrast to osteocalcin in vitamin D–deficient rat molars [53]. However, immunolabeling revealed an absence of extracellular DPP in vitamin D–deficient rat dentin, suggesting potential inhibition of DPP exocytosis and/or abnormal binding to collagen or any posttranscriptional event. Interestingly, the expression of enamel proteins in dentin was also affected by vitamin D. The control of amelogenin transcripts by vitamin D was reversed in odontoblasts and ameloblasts, with an overexpression of amelogenins in vitamin D–deficient rat and VDR −/− mice odontoblasts and dentin ([6], Lézot, unpublished data). Such reverse control for amelogenin strongly suggests that vitamin D may indirectly control amelogenin expression via epithelialmesenchymal interactions and further validates an in vivo strategy for vitamin D action on teeth. The regulatory pathways for control of enamel-driven gene promoter activity is partially understood [54]. It has been previously shown that enamel thickness is tightly related to amelogenin gene invalidation experiments [56]. The transcription factor Dlx2 may drive spatial morphoregulation of amelogenin gene expression and enamel thickness, since (1) an inverse relationship between enamel thickness and Dlx2 expression in secretory stage ameloblast has been established [57] and (2) Dlx2 is able to bind the amelogenin promoter and control its activity [58].

C. Vitamin D and Periodontium The role of vitamin D in periodontium has been studied with respect to the impressive defects in dental crown cells associated with rickets. Periodontal abcesses without infectious etiology are associated in some case reports of patients with vitamin D–resistant rickets. Ultrastructural changes in alveolar bone and cementum in rats fed a low-calcium and vitamin D– deficient diet may help to interpret this phenotype [48,59]. Increased resorption is observed in this situation and in VDR −/− hypocalcemic mice (Davideau et al., unpublished). In vitro studies have shown that

605

CHAPTER 34 Dento-alveolar Bone Complex and Vitamin D

periodontal ligament cells may control osteoclast differentiation through a 1,25(OH)2 vitamin controlled D3 balance of RANK-L/osteoprotegerin [42]. Further systematic studies are required. The periodontal consequences of VDR ablation may be the result of impaired intestinal calcium absorption and/or resultant secondary hyperpathyroidism and hypophosphatemia as established in axial bone [48]. Identification of vitamin D target cells in periodontium by in situ hybridization detection of nuclear VDR mRNA reveals that the VDR is strongly expressed in odontoblasts and alveolar bone osteoblasts, and to a lesser extent in periodontal ligament cells. However, root epithelium cell as well as acellular cementum cementoblasts do not appear to express the receptor [2]. MARRS protein has also been shown in alveolar bone [4] but the expression in other periodontal tissues is not established. Analyses of periodontal tissues in VDR −/− hypocalcemic mutant mice support the notion of differential vitamin D responsiveness in periodontal cells. Diverse defects in biomineralization and matrix deposition are observed in alveolar bone, cellular cementum, and root dentin. They may be related to increased bone sialoprotein and decreased osteopontin expression in cementoblasts and endosteal spaces of alveolar bone reported in vitamin D–deficient mice [10]. Furthermore, 1,25(OH)2D3 increases alkaline phosphatase expression in cementoblasts [51]. Finally, recent epidemiologic and genetic studies have highlighted the link between the rate of tooth loss or severity of the periodontitis and the vitamin D status and polymorphism in the VDR gene [60–62]. They suggest further investigation into the molecular mechanisms of vitamin D action during periodontium formation is warranted and support the relevance of vitamin D action in the maintenance of healthy periodontium.

IV. CONCLUSION Dental anomalies are included in fuel descriptions of clinical cases showing calcium and phosphate homeostasis disequilibrium in pediatric medicine and dentistry. Recent data provide some physiopathological interpretation of these observed abnormalities. They include (1) expression of human VDR [63] and MARRS [4] proteins, (2) selective regulation of enamel protein synthesis and affected amelogenesis, (3) activated resorption of alveolar bone/cementum in vitamin bioinactivation and spontaneous abscess, (4) the role of Msx/Dlx homeogenes in cell differentiation and gene regulation. In the adult, the impact of vitamin D on tooth pathology has not been studied. These modest advances, together with clinical findings, suggest that

periodontal disease prevalence and incidence are related to human VDR polymorphism [60,61]. Alveolar bone loss is prevented by nutritional supplementation of vitamin D and calcium [62]. Some experimental data confirm the importance of vitamin D in root formation and resorption. Further exploration is required to delineate the molecular pathways of vitamin D in the numerous periodontal cell types in order to apply preventive management to tooth loss in geriatric medicine and dentistry. From above perspective, dental and bone cells have been analyzed from contrasting points of view. The molecular mechanisms of morphogenesis is relatively easy to delineate in tooth germ when compared to bone where the relative spatial organization of osteoblasts and osteoclasts are complex. However, gene promotes studies of tooth-specific genes has only begun. The spatially and temporally coordinated process of odontogenesis and undissociable periodontal formation may provide good model systems to explore the relationships between genetically and hormonally controlled morphogenesis of the skeleton.

Acknowledgments The collaboration of D. Hotton, P. Papagerakis, M. Mesbah, and M. Oboeuf is gratefully acknowledged.

References 1. Berdal A, Hotton D, Pike JW, Mathieu H, Dupret JM 1993 Cell and stage-specific expression of vitamin D receptor and calbindin-D genes in rat incisor: Regulation by 1,25-dihydroxyvitamin D3. Dev Biol 155:172–179. 2. Davideau JL, Papagerakis P, Hotton D, Lezot F, Berdal A 1996 In situ investigation of vitamin D receptor, alkaline phosphatase, osteocalcin gene expression in oro-facial mineralized tissues. Endocrinology 137:3577–3585. 3. Berdal A, Balmain N, Cuisinier-Gleizes P, Mathieu H 1987 Histology and microradiography of early post-natal molar tooth development in vitamin-D deficient rats. Arch Oral Biol 32:493–498. 4. Mesbah M, Nemere I, Papagerakis P, Nefussi JR, Orestes-Cardoso S, Nessmann C, Berdal A 2002 Expression of a 1,25-dihydroxyvitamin D3 membrane-associated rapidresponse steroid binding protein during human tooth and bone development and biomineralization. J Bone Miner Res 17(9): 1588–1596. 5. Papagerakis P, MacDougall M, Berdal A 2002 Differential epithelial and mesenchymal regulation of tooth-specific matrix proteins expression by 1,25-dihydroxyvitamin D3 in vivo. Connect Tissue Res 43(2–3):372–375. 6. Papagerakis P, MacDougall M, Hotton D, Bailleul-Forestier I, Oboeuf M, Berdal A 2003 Expression of amelogenin in odontoblasts. Bone 32(3):228–240.

606 7. Hodgkinson JE, Davidson CL, Beresford J, Sharpe PT 1993 Expression of human homeobox-containing gene is regulated by l,25(OH)2D3 in bone cells. Biochim Biophys Acta 1174:11–16. 8. Bronckers AL, Price PA, Schrijvers A, Bervoets TJ, Karsenty G 1998. Studies of osteocalcin function in dentin formation in rodent teeth. Eur J Oral Sci 106(3):795–805. 9. Lezot F, Descroix V, Mesbah M, Hotton D, Blin C, Papagerakis P, Mauro N, Kato S, MacDougall M, Sharpe P, Berdal A 2002 Cross-talk between Msx/Dlx homeobox genes and vitamin D during tooth mineralization. Connect Tissue Res 43:509–514. 10. Chen JJ, Jin H, Ranly DM, Sodek J, Boyan BD 1999 Altered expression of bone sialoproteins in vitamin D-deficient rBSP2.7Luc transgenic mice. J Bone Min Res 14(2):221–229. 11. Berdal A, Hotton D, Saffar JL, Thomasset M, Nanci A 1996 Calbindin-D9k and calbindin-D28k expression in rat mineralized tissues in vivo. J Bone Miner Res 11(2):768–779. 12. Wurtz T, Berdal A 2003 Osteoblast precursors at different anatomic sites. Crit Rev Eucaryot Gene Exp, in press. 13. Orestes-Cardoso SM, Nefussi JR, Hotton D, Mesbah M, Orestes-Cardoso MD, Robert B, Berdal A 2001 Postnatal Msx1 expression pattern in craniofacial, axial, and appendicular skeleton of transgenic mice from the first week until the second year. Dev Dyn 221(1):1–13. 14. Thesleff I 2003 Epithelial-mesenchymal signalling regulating tooth morphogenesis. J Cell Sci. 116(Pt 9):1647–1648. 15. Miletich I, Sharpe PT 2003 Normal and abnormal dental development. Hum Mol Genet 12(Suppl 1):R69–73. 16. Ten Cate AR 1997 The development of the periodontium—a largely ectomesenchymally derived unit. Periodontology 13:9–19. 17. Jiang X, Iseki S, Maxson RE, Sucov HM, Morriss-Kay GM 2002 Tissue origins and interactions in the mammalian skull vault. Dev Biol 241:106–116. 18. Van den Boogaard MJ, Dorland M, Beemer FA, van Amstel HK 200 MSX1 mutation is associated with orofacial clefting and tooth agenesis in humans. Nat Genet 24:342–343. 19. Price JA, Bowden DW, Wright JT, Pettenati MJ, Hart TC 1998 Identification of a mutation in DLX3 associated with trichodento-osseous (TDC) syndrome. Hum Mol Genet 7(3):563–569. 20. Francis-West P, Ladher R, Barlow A, Graveson A 1998 Signalling interactions during facial development. Mech Dev 75(1–2):3–28. 21. Nanci A, Smith CE 1992 Development and calcification of enamel. In: Bonucci E (ed) Calcification in Biological Systems. CRC Press, Boca Raton, FL, pp. 313–343. 22. Butler WT, Brunn JC, Qin C, McKee MD 2002 Extracellular matrix proteins and the dynamics of dentin formation. Connect Tissue Res 43(2–3):301–307. 23. Saygin NE, Giannobile WV, Somerman MJ 2000 Molecular and cell biology of cementum. Peridontology 24:73–98. 24. MacDougall M, Simmons D, Luan X, Nydegger J, Feng J, Gu TT 1997 Dentin phosphoprotein and dentin sialoprotein are cleavage products expressed from a single transcript coded by a gene on human chromosome 4. J Biol Chem 272: 836–842. 25. Xiao S, Yu C, Chou X, Yuan W, Wang Y, Bu L, Fu G, Qian M, Yang J, Shi Y, Hu L, Han B, Wang Z, Huang W, Liu J, Chen Z, Zhao G, Kong X 2001 Dentinogenesis imperfecta 1 with or without progressive hearing loss is associated with distinct mutations in DSPP. Nat Genet 27(2):201–204. 26. Sreenath T, Thyagarajan T, Hall B, Longenecker G, D’Souza RN, Hong S, Wright JT, MacDougall M, Sauk JJ, Kulkarni AB

ARIANE BERDAL

27.

28.

29. 30.

31.

32.

33. 34.

35. 36.

37.

38. 39.

40. 41.

42.

ET AL .

2003 Dentin sialophosphoprotein is critical for tooth mineralization. J Dent Res 82(Spec IssB):1868. MacDougall M, Simmons D, Gu TT, Dong J 2002 MEPE/OF45, a new dentin/bone matrix protein and candidate gene for dentin diseases mapping to chromosome 4q21. Connect Tissue Res 43(2–3):320–330. Jain A, Karadag A, Fohr B, Fisher LW, Fedarko NS 2002 Three SIBLINGs (small integrin-binding ligand, N-linked glycoproteins) enhance factor H’s cofactor activity enabling MCP-like cellular evasion of complement-mediated attack. J Biol Chem 227(16):13700–13708. Robinson C, Kirkham J, Brookes SJ, Bonass WA, Shore RC 1995 The chemistry of enamel development. Int J Dev Biol 39:145–152. Limeback H, Schlumbohm C, Sen A, Nikiforuk G 1992 Effects of hypocalcemia/hypophosphatemia on porcine bone and dental hard tissues in an inherited form of type 1 pseudovitamin D deficiency rickets. J Dent Res 71:346–352. Hart PS, Aldred MJ, Crawford PJ, Wright NJ, Hart TC, Wright JT 2002 Amelogenesis imperfecta phenotype-genotype correlations with two amelogenin gene mutations. Arch Oral Biol 47(4):261–265. Gibson CW, Yuan ZA, Hall B, Longenecker G, Chen E, Thyagarajan T, Sreenath T, Wright Jt, Decker S, Piddington R, Harrison G, Kulkarni AB 2001 Amelogenin-deficient mice display an amelogenesis imperfecta phenotype. J Biol Chem 276(34):31871–31875. Cerny R, Slaby I, Hammarstrom L, Wurtz T 1996 A novel gene expressed in rat ameloblasts codes for proteins with cell binding domains. J Bone Miner Res 11:883–891. Rajpar MH, Harley K, Laing C, Davies RM, Dixon MJ 2001 Mutation of the gene encoding the enamel-specific protein, enamelin, causes autosomal-dominant amelogenesis imperfecta. Hum Mol Genet 10(16):1673–1677. Fong CD, Slaby I, Hammarstrom L 1996 Amelin: an enamelrelated protein, transcribed in the cells of epithelial root sheath. J Bone Miner Res 11(7):892–898. MacDougall M, Nydegger J, Gu TT, Simmons D, Luan Z, Cavender A, D’Souza RN 1998 Developmental regulation of dentin sialophosphoprotein during ameloblast differentiation: a potential enamel matrix nucleator. Connect Tissue Res 39(1-3):25–37. Veis A, Tompkins K, Alvares K, Wei K, Wang L, Wang XS, Brownell AG, Jengh Sm, Healy KE 2000 Specific amelogenin gene splice products have signalling effects on cells in culture and in implants in vivo. J Biol Chem 275(52):41263–42172. Veis A 2003 Amelogenin gene splice products: Potential signalling molecules. Cell Mol Life Sci 60(1):38–55. Gestrelius S, Andersson C, Lidstrom D, Hammastrom L, Somerman M 1997 In vitro studies on periodontal ligament cells and enamel matrix derivatives. J Clin Periodontol 24(9 Pt 2): 685–692. Basdra EK, Komposch G 1997 Osteoblast-like properties of human periodontal ligament cells: an in vitro analysis. Eur J Orthod 19(6):615–621. Okubo K, Kobayashi M, Takiguchi T, Takada T, Ohazama A, Okamatsu Y, Hasegawa K 2003 Participation of endogenous IGF-1 and TGF-β1 with enamel derivative-stimulated cell growth in human periodontal ligament cells. J Periodontal Res 38(1):1–9. Hasegawa T, Yoshimura Y, Kikuiri T, Yawaka Y, Takeyama S, Matsumoto A, Oguchi H, Shirakawa T 2002 Expression of receptor activator of NF-kappa B ligand and osteoprotegerin in culture of human periodontal ligament cells. J Periodontal Res 37(6):405–411.

CHAPTER 34 Dento-alveolar Bone Complex and Vitamin D

43. Lezot F, Davideau JL, Thomas B, Sharpe P, Forest N, Berdal A 2000 Epithelial Dlx-2 homeogene expression and cementogenesis. J Histochem Cytochem 48(2):227–284. 44. Pitaru S, Pritzki A, Bar-Kana I, Grosskopf A, Savion N, Narayanan AS 2002 Bone morphologenetic protein 2 induces the expression of cementum attachment protein in human periodontal ligament clones. Connect Tissue Res 43(2–3): 257–264. 45. Sodek J, McKee MD 2000 Molecular and cellular biology of alveolar bone. Periodontology 24:99–126. 46. Orestes-Cardoso SM, Nefussi JR, Hotton D, Mesbah M, Orestes-Cardoso MD, Robert B, Berdal A 2001 Postnatal Msx1 expression pattern in craniofacial, axial, and appendicular skeleton of transgenic mice from the first week until the second year. Dev Dyn 221(1):1–13. 47. Berdal A 1997 Vitamin D action on tooth development and biomineralization. In: Feldman D, Glorieux FH, Pike JW (eds) Vitamin D. Academic Press, New York, p. 423. 48. Amling M, Priemel M, Holsmann T, Chapin K, Rueger JM, Baron R, Demay MB 1999 Rescue of the skeletal phenotype of vitamin D–receptor ablated mice in the setting of normal mineral ion homeostasis: Formal histomorphometric and biomechanical analyses. Endocrinology 140(11):4982–4987. 49. Bielaczyc A, Golebiewska M 1997 Ultrastructural changes of a tooth root in young rats fed a low calcium and vitamin Ddeficient diet. Rocz Akad Med Bialymst 42(Suppl 2):153–158. 50. Blin-Wakkach C, Lezot F, Ghoul-Mazgar S, Hotton D, Monteiro D, Teillaud C, Pibouin L, Orestes-Cardoso S, Papagerakis P, Macdougall M, Robert B, Berdal A 2001 Endogenous Msx1 antisense transcript: in vivo and in vitro evidences, structure, and potential involvement in skeleton development in mammals. Proc Natl Acad Sci 98(13): 7336–7341. 51. Hou LT, Liu CM, Chen YJ, Wong MY, Chen KC, Chen J, Thomas HF 1999 Characterization of dental follicle cells in developing mouse molar. Arch Oral Biol 44(9):759–770. 52. Zhao M, Berry JE, Somerman MJ 2003 Bone morphogenetic protein-2 inhibits differentiation and mineralization of cementoblasts in vitro. J Dent Res 82(1):23–27.

607 53. Berdal A, Gorter de Vries I, Hotton D, Cuisinier-Gleizes P, Mathieu H l991 The cellular and extracellular distribution of osteocalcin and dentin phosphoprotein in teeth of vitamin D– deficient rats. J Biol Bucc 19:45–63. 54. Dhamija S, Krebsbach PH 2001 Role of Cbfa1 in ameloblastin gene transcription. J Biol Chem 276(37):35159–35164. 55. Snead ML, Luo W, Lau EC, Slavkin HC 1988 Spatial- and temporal-restricted pattern for amelogenin gene expression during mouse tooth organogenesis. Development 104(1):77–85. 56. Gibson CW, Yuan ZA, Hall B, Longenecker G, Chen E, Thyagarajan T, Sreenath T, Wright JT, Decker S, Piddington R, Harrison G, Kulkarni AB 2001 Amelogenin-deficient mice display an amelogenesis imperfecta phenotype. J Biol Chem 276(34):31871–31875. 57. Lezot F, Thomas B, Hotton D, Forest N, Orestes-Cardoso S, Robert B, Sharpe P, Berdal A 2000 Biomineralization, lifetime of odontogenic cells and differential expression of the two homeobox genes MSK-1 and DLX-2 in transgenic mice. J Bone Miner Res 15:430 –441. 58. Gibson CW, Greene SC, Yuan ZA, Hotton D, Lezot F, Berdal A 2003 Interaction of Dlx2 transcription factor with upstream regions of amelogenin gene. IADR March 12–15. 59. Kerebel B, Brou E 1981 Effect of vitamin D deficiency and vitamin D3 therapy on the pig cementum. Bull Group Int Rech Sci Stomatol Odontol 24(1):37–45. 60. Sun JL, Meng HX, Cao CF, Tachi Y, Shinohara M, Ueda M, Imai H, Ohura K 2002 Relationship between vitamin D receptor gene polymorphism and periodontotitis. J Periodontal Res 37(4):263–267. 61. Inagaki K, Krall EA, Fleet JC, Garcia RI 2003 Vitamin D receptor alleles, periodontal disease progression, and tooth loss in the VA dental longitudinal study. J Periodontol 74:161–167. 62. Krall EA 2001 The periodontal-systemic connection: Implications for treatment of patients with osteoporosis and periodontal disease. Ann Periodontol 6(1):209–213. 63. Bailleul-Forestier I, Davideau JL, Papagerakis P, Noble I, Nessmann C, Peuchmaur M, Berdal A 1996 Immunolocalization of vitamin D receptor and caldindin-D28k in human tooth germ. Pediatr Res 39(4 Pt 1):636–642.

CHAPTER 35

Vitamin D: Role in Skin and Hair DANIEL D. BIKLE

Endocrine Research Unit, VA Medical Center, UCSF Medical Services, San Francisco, California

I. Introduction II. Cutaneous Production of 1,25-Dihydroxyvitamin D III. Regulation of Keratinocyte Differentiation

IV. Regulation of Hair Follicle Cycling References

I. INTRODUCTION

II. CUTANEOUS PRODUCTION OF 1,25–DIHYDROXYVITAMIN D

1,25-Dihydroxyvitamin D [1,25(OH)2D] and possibly other vitamin D metabolites have functions that extend beyond those of regulating bone mineralization and intestinal calcium transport. The skin is one such tissue where such a broader role is being intensively explored. Besides producing vitamin D, epidermal cells (keratinocytes) make 1,25(OH)2D [1], contain 1,25(OH)2D receptors (VDR) [2–4], and respond to 1,25(OH)2D with changes in proliferation and differentiation [3,5,6]. Calcium is an important modulator of these pathways. Calcium decreases 1,25(OH)2D production and regulates the effects of 1,25(OH)2D on proliferation and differentiation [7]. Calcium is itself an important regulator of keratinocyte proliferation and differentiation [8,9], effects that are in turn modulated by 1,25(OH)2D. Although much of our information about the role of calcium and 1,25(OH)2D comes from in vitro studies, a calcium gradient exists in the epidermis that appears to be an important regulator of proliferation and differentiation in vivo. Animals lacking the ability to produce 1,25(OH)2D have difficulty restoring this gradient if disrupted, and such animals show abnormalities in terminal differentiation of the epidermis even when placed on a high-calcium diet. Animals lacking the vitamin D receptor (VDR) also have an abnormality in epidermal differentiation, but this can be corrected with a highcalcium diet. However, these VDR-null animals have an abnormality in hair follicle cycling not seen in vitamin D–deficient animals or those lacking the capacity to produce 1,25(OH)2D. Exploring the production and action of vitamin D and its metabolites in the epidermis and hair follicle should not only increase our understanding of the mechanisms by which this hormone family regulates keratinocyte function but also lead to advances in our ability to treat diseases of disordered epidermal and hair follicle differentiation including psoriasis and squamous cell carcinoma. The use of calcipotriol for the treatment of psoriasis is the first example of this therapeutic potential. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

A. Metabolism of 25OHD and Identification of the Products Keratinocytes are capable not only of producing vitamin D3 from endogenous sources of 7-dehydrocholesterol (7-DHC) in a regulated fashion (Chapter 3), but of metabolizing the vitamin D via the 25-hydroxylase and 1α-hydroxylase steps to 1,25(OH)2D [1,10–12]. Keratinocytes appear to be the only cell in the body capable of the entire pathway (Fig. 1). The vitamin D-25 hydroxylase in keratinocytes is the same mitochondrial enzyme (CYP27) that converts vitamin D to 25OHD in the liver [13,13a]. Its expression is increased by vitamin D and UVB irradiation [13]. Similarly the 25OHD-1α hydroxylase in the epidermis, which is responsible for 1,25(OH)2D production, is the same enzyme (CYP27B1) as that found in the kidney [14]. Its expression and enzymatic activity are tightly regulated and coupled to the differentiation of these cells. Extrarenal production of 1,25(OH)2D has been demonstrated in both anephric humans [15,16] and pigs [17], although the tissue(s) responsible for the circulating levels of 1,25(OH)2D in anephric animals has not been established. The epidermis is likely to contribute to the total as human keratinocytes rapidly and extensively convert 25OHD to 1,25(OH)2D, although it is not clear how much of the 1,25(OH)2D produced by the epidermis actually enters the circulation. Peak levels of 1,25(OH)2D are reached in the cell within 1 hr after adding 25OHD. By 1 hr 1,25(OH)2D is the main metabolite observed; however, other metabolites appear with continued incubation, many of which represent degradation products of 1,25(OH)2D(10). The apparent Km for the enzyme (25OHD 1α-hydroxylase) metabolizing 25OHD to 1,25(OH)2D is estimated to be 5 × 10−8 M, a value lower than that estimated for the kidney. The production of 1,25(OH)2D by isolated keratinocytes in culture has been confirmed using intact pig skins Copyright © 2005, Elsevier, Inc. All rights reserved.

610

DANIEL D. BIKLE

Vitamin D hydroxylations in keratinocytes Sun Liver

CYP27

7-DHC

D3

CYP27

25(OH)D3 CYP24

CYP27B1 CYP24

25(OH)D3

24,25(OH)2D3 CYP27B1

1,25(OH)2D3 PTH

Keratinocyte TNF

IFN

Parathyroid

Macrophage

FIGURE 1 Regulation of 1,25(OH)2D production in the keratinocyte. 7-Dehydrocholesterol (7-DHC) is converted to vitamin D3 by a photochemical reaction. The vitamin D3 produced is either transported out of the keratinocyte to the liver, where it is converted to 25OHD3, or metabolized directly to 25OHD3 in the keratinocyte by the enzyme CYP27. 25OHD3 is metabolized either to 24,25(OH)2D3 or to 1,25(OH)2D3 by the enzymes CYP 24 and CYP27B1, respectively. Parathyroid hormone (PTH) secreted by the parathyroid gland (PTG) stimulates the production of 1,25(OH)2D3, as does tumor necrosis factor-α (TNF) secreted by keratinocytes and interferon-γ (IFN) secreted by macrophages. 1,25(OH)2D3 promotes its own catabolism by inducing the 24-hydroxylase [also responsible for 24,25(OH)2D3 production] and decreasing IFN secretion by macrophages.

perfused with 25OHD [18]. However, when renal production of 1,25(OH)2D is normal the circulating levels of 1,25(OH)2D are sufficient to limit the contribution from epidermal production. This appears to be due to the induction of 25OHD-24-hydroxylase in the keratinocyte by 1,25(OH)2D, which catabolizes the endogenously produced 1,25(OH)2D before it leaves the cell in which it is produced [19].

B. Hormonal Regulation Both the formation and catabolism of 1,25(OH)2D are under hormonal control. Parathyroid hormone (PTH) (optimal concentration 20 ng/ml) exerts a modest

stimulation of 1,25(OH)2D production, whereas the phosphodiesterase inhibitor isobutylmethylxanthine (IBMX) inhibits its degradation [10]. In combination, PTH and IBMX markedly increase the amount of 1,25(OH)2D that accumulates within the keratinocyte following the addition of 25OHD. These effects are not reproduced by cAMP or its membrane-permeable derivatives, suggesting that the actions of PTH and IBMX may be operating through a mechanism independent of cAMP. The effects of PTH and IBMX are maximal after a 4-hr incubation of cells with these agents before adding 25OHD; that is, the effects are not immediate. In renal cells PTH directs a more acute stimulation of 1,25(OH)2D production [20], and cAMP appears to play a second messenger role [21]. Thus, the regulation

611

CHAPTER 35 Vitamin D: Role in Skin and Hair

C. Effects of Differentiation In Section III,D we review the role of 1,25(OH)2D in promoting differentiation of the keratinocyte. In this

100

10

50

0.2

20

0.1

10

3 5

8 11 15 Days in culture

1α-hydroxylase (---) f mol 1,25(OH)2D3/mg cell protein

20

24-hydroxylase (---) f mol 24,25(OH)2D3/mg cell protein

Transglutaminase(—) f mol3H putrescine/mg cell protein

section, the ability of differentiation to regulate 1,25(OH)2D production is considered. In the presence of adequate levels of calcium, keratinocytes progress from rapidly proliferating cells to cells capable of making cornified envelopes, one of the most distinctive features of epidermal terminal differentiation. The cornified envelope is formed by the cross-linking of precursor molecules such as involucrin and loricrin into an insoluble, durable sheet by the membrane-bound enzyme transglutaminase. As the cells differentiate in culture there is a sequential increase in involucrin, transglutaminase, and cornified envelope formation [4,28]. 25OHD 1α-hydroxylase and 24-hydroxylase also change with differentiation [4] (Fig. 2). Preceding the rise in transglutaminase and involucrin is a rise in 25OHD 1α-hydroxylase activity; the 1α-hydroxylase activity, transglutaminase activity, and involucrin then fall as cornified envelopes appear; the appearance of cornified envelopes and the fall in 1α-hydroxylase activity

Cornified envelope (—) OD340/mg cell protein

of 1,25(OH)2D production in keratinocytes by PTH and cAMP differs from their regulation of 1,25(OH)2D production in renal cells. The mechanism by which PTH stimulates 1,25(OH)2D production in keratinocytes is unclear; PTH receptors have been difficult to detect in normal keratinocytes, and PTH fails to stimulate adenylate cyclase activity in these cells. The hormone 1,25(OH)2D negatively regulates its own levels within the keratinocyte. This negative feedback loop is similar to that observed in the kidney, but it differs from that seen in the macrophage, which lacks this feedback loop. In the keratinocyte, this feedback inhibition is not mediated by an effect on 1,25(OH)2D production but is due solely to stimulation of 1,25(OH)2D catabolism through induction of the enzyme 25OHD 24-hydroxylase that converts 25OHD and 1,25(OH)2D to 24,25(OH)2D and 1,24,25(OH)3D, respectively [14]. The concentrations required to regulate 25OHD metabolism in keratinocytes (EC50 approximately 10−11 M) are free concentrations and are independent of vitamin D binding proteins (DBPs). Addition of serum (which contains DBP) or albumin reduces the free fraction of 1,25(OH)2D and increases the apparent EC50 for total 1,25(OH)2D. However, using direct measurements of the free fraction of 1,25(OH)2D [22], we showed that the EC50 for the free concentration was not altered by serum or albumin. This suggests that DBP does not play an important role in delivering 25OHD to the epidermis, but acts more as a reservoir for circulating vitamin D metabolites. An important difference in the regulation of 25OHD metabolism by 1,25(OH)2D between keratinocytes and renal cells is that the concentration of 1,25(OH)2D required to induce the 24-hydroxylase in renal cells appears to be several orders of magnitude greater than that required for comparable effects in keratinocytes [10,23–25]. Thus, the detectable 1,25(OH)2D production by keratinocytes is exquisitely sensitive to exogenous 1,25(OH)2D. This difference in sensitivity to feedback inhibition by 1,25(OH)2D between keratinocytes and renal cells may account for the observation that following acute nephrectomy extrarenal production of 1,25(OH)2D is very low [26,27]; only with time after renal production has ceased does extrarenal production emerge. A similar observation was made in pig skins perfused with 25OHD; the amount of 1,25(OH)2D produced was initially low but increased after 4–8 hr of perfusion [18].

20

FIGURE 2 Change in 1- and 24-hydroxylase activities in comparison to transglutaminase activity and cornified envelope formation as keratinocytes grow and differentiate in culture. The cells were plated on day 1, and the various measurements were made on the days indicated. Production of 1,25(OH)2D rises ahead of transglutaminase activity as the cells approach confluence, then falls along with transglutaminase activity as cornified envelope formation and 24,25(OH)2D production ensue. Modified from Pillai et al. [4] with permission from J Biol Chem.

612 coincide with a rise in 24-hydroxylase activity [4]. The change in activity reflects a change in expression of the gene, although the means by which the expression of the 1α-hydroxylase is controlled during differentiation has not yet been determined. Growing the cells in 0.1 mM calcium, which retards differentiation [28], permits the cells to maintain higher 1α-hydroxylase activity than when they are grown in 1.2 mM calcium [7], although acute changes in calcium have little effect on 1,25(OH)2D production [10]. These changes in 1α-hydroxylase expression in vitro are consistent with the finding of 1α-hydroxylase in the stratum basale but not suprabasal levels of the epidermis in vivo [29].

D. Regulation by Cytokines Both tumor necrosis factor-α (TNFα) and interferon-γ (IFN-γ) bind to and promote the differentiation of keratinocytes [30,31]. Both cytokines regulate 1,25(OH)2D production by these cells in a manner consistent with their effects on differentiation [7,32] (Fig. 1). Unlike PTH and 1,25(OH)2D, these cytokines must be incubated with the keratinocytes for at least 1 day (not hours) before their effects on 1,25(OH)2D production are observed. These cells are exquisitely sensitive to IFN-γ, with maximal stimulation of 1,25(OH)2D production at concentrations less than 10 pM. Higher concentrations are inhibitory, but such concentrations also profoundly inhibit the proliferation of these cells and limit their ability to differentiate. Keratinocytes grown in 0.1 mM calcium are more sensitive to IFN-γ than cells grown in 1.2 mM calcium [7]. Serum markedly reduces the potency of IFN-γ in this system, for reasons that are obscure. TNFα stimulates 1,25(OH)2D production and transglutaminase activity in preconfluent cells [32], and it can reverse the inhibition seen with the higher concentrations of IFN-γ. The effects of TNFα and IFN-γ are not additive at the lower and stimulatory concentrations of IFN-γ. When TNFα is added after the cells have reached confluence, a time after which 1,25(OH)2D production (and transglutaminase activity) has peaked, TNFα inhibits 1,25(OH)2D production (and transglutaminase activity) even as it stimulates cornified envelope formation. Although IFN-γ is not made in keratinocytes, TNFα is produced by these cells, and its synthesis is stimulated by ultraviolet light [33] and barrier disruption [34]. Thus, environmental perturbations could enhance 1,25(OH)2D production in the skin, and the increased levels of 1,25(OH)2D could play a role in the recovery from UV damage and/or barrier repair [34a].

DANIEL D. BIKLE

E. 1,25(OH)2D Production by Transformed Keratinocytes Keratinocytes from squamous cell carcinomas (SCC) do not differentiate normally in response to calcium [35] or 1,25(OH)2D [36] despite having genes for the differentiation markers that can be induced by serum [20]. Nevertheless, these cells produce 1,25(OH)2D [and 24,25(OH)2D], and in some cases the rates of production are comparable to those of normal keratinocytes [36]. Furthermore, the SCC lines respond to exogenous 1,25(OH)2D with a reduction in 1,25(OH)2D production and an increase in 24,25(OH)2D production, although in some cases the sensitivity of the SCC line to 1,25(OH)2D is less than normal [36]. The levels of the VDR mRNA and protein in SCC are comparable to those in normal keratinocytes [20], suggesting that the reason why 1,25(OH)2D can regulate 25OHD metabolism but not differentiation in SCC lies in other transcription factors required for calcium and 1,25(OH)2D regulation of the differentiation pathway. We [37] have recently demonstrated that the coactivator complex binding to the VDR in SCC and proliferating keratinocytes is the DRIP (vitamin D receptor interacting protein) complex. As keratinocytes differentiate, components of the DRIP complex are no longer produced, whereas SRC 3 (steroid receptor coactivator 3) increases in levels and binding to the VDR. This transition does not take place in SCC. The 24-hydroxylase gene appears to be activated by VDR bound to either DRIP or SRC3, whereas other vitamin D–regulated genes involved with differentiation (e.g., involucrin) prefer VDR bound to SRC3.

F. Clinical Implications The finding of 1,25(OH)2D production by keratinocytes indicates that the skin is at least one source for extrarenal production of this important metabolite. The kidney is the major source, but in anephric individuals the circulating level of 1,25(OH)2D may fall sufficiently low such that the cutaneous degradation of 1,25(OH)2D is no longer induced, allowing more keratinocyte-derived 1,25(OH)2D into the general circulation. Thus, the epidermis may provide 1,25(OH)2D to the body in patients with decreased or absent renal function, accounting for the increase in 1,25(OH)2D levels when such individuals are provided adequate amounts of 25OHD [38]. Although a variety of squamous cell carcinomas are associated with hypercalcemia, this appears to be due to their elaboration of a parathyroid hormone like protein (PTHrP) and not due

613

CHAPTER 35 Vitamin D: Role in Skin and Hair

to uncontrolled 1,25(OH)2D production. Most likely the 1,25(OH)2D produced by keratinocytes serves an autocrine or paracrine function, regulating the proliferation and differentiation of these cells. The ability of the epidermis to produce 1,25(OH)2D from 25OHD and to catabolize the vitamin D metabolites quickly offers several possibilities for the topical administration of vitamin D compounds in the treatment of skin disorders.

III. REGULATION OF KERATINOCYTE DIFFERENTIATION A. Microanatomy of the Epidermis The epidermis is composed of four layers of keratinocytes at different stages of differentiation (Fig. 3). The basal layer (stratum basale) rests on the basal lamina separating the dermis and epidermis. These cells proliferate, providing the cells for the upper differentiating layers. They are large, columnar cells forming intercellular attachments with adjacent cells through desmosomes. An asymmetric distribution of integrins

Intercellular lamellae

Cornified layer

Dense band

Keratohyalin Granular layer Lamellar bodies

Spinous layer

Basal layer

Basal lamina

FIGURE 3

Four principal layers of the epidermis. The basal layer rests on the basal lamina separating the epidermis from the dermis. The cornified layer provides the major barrier to the outside environment. The basal layer provides the cells which differentiate as they pass through the different layers to become corneocytes. Each layer has its distinct appearance, function, and differentiation markers as described in the text.

on their lateral and basal surface may also regulate their attachment to the basal lamina and adjacent cells [39–41]. They contain an extensive keratin network comprising principally keratins K5 (58 kDa) and K14 (50 kDa) [42]. By a process that we are only beginning to understand, cells migrate upward from this basal layer, acquiring the characteristics of a fully differentiated corneocyte, which is eventually sloughed off. The layer above the basal cells is the spinous layer (stratum spinosum). These cells initiate the production of the keratins K1 and K10, which are the keratins characteristic of the more differentiated layers of the epidermis [43]. Cornified envelope precursors such as involucrin [44] also appear in the spinous layer as does the enzyme transglutaminase, responsible for the ε-(γ-glutamyl)lysine cross-linking of these substrates into the insoluble cornified envelope [45]. The keratinocyte contains both a soluble (tissue, TG-C, or type II) and a membrane-bound (particulate, TG-K, or type I) form of transglutaminase. It is the membrane-bound form that correlates with differentiation and is thought to be responsible for the formation of the cornified envelope [45]. The granular layer (stratum granulosum), lying above the spinous layer, is characterized by electron-dense keratohyalin granules. These are of two types [46]. The larger of the two granules contains profilaggrin, the precursor of filaggrin, a protein thought to facilitate the aggregation of keratin filaments [47]. The smaller granule contains loricrin, a major component of the cornified envelope [48]. The granular layer also contains lamellar bodies, lipid-filled structures that fuse with the plasma membrane, divesting their contents into the extracellular space where the lipid contributes to the permeability barrier of skin [49]. As the cells pass from the granular layer to the cornified layer (stratum corneum), they undergo destruction of their organelles with further maturation of the cornified envelope into an insoluble, highly resistant structure surrounding the keratin–filaggrin complex and linked to the extracellular lipid milieu [50]. Calcium forms a steep gradient within the epidermis, with highest concentration in the stratum granulosum [51]. Disruption of the permeability barrier by removing the stratum corneum or extracting its lipids leads to a loss of this calcium gradient [52] resulting in increased lamellar body secretion but reduced expression of the genes for loricrin, profilaggrin, and involucrin [53].

B. Regulators of Growth and Differentiation—General The ability to grow keratinocytes in culture in a manner that permits at least partial differentiation has made

614 it possible to study the regulation of this process [54]. Although this chapter emphasizes the roles of calcium and 1,25(OH)2D in keratinocyte differentiation, a number of hormones, cytokines, and ions are involved. 1. VITAMIN A

Vitamin A and its metabolites and analogs (collectively referred to as retinoids) have long been known to influence epidermal development. Vitamin A deficiency induces squamous metaplasia, causing even normal keratinizing epithelia (e.g., the epidermis) to become hyperkeratotic [55]. In contrast, pharmacological doses of retinoids can induce a mucous metaplasia in what is otherwise a keratinizing epithelium [56]. In culture, it has long been appreciated that retinoids block the terminal differentiation of keratinocytes [57]. Retinoids block the appearance of the suprabasal keratins (Kl and K10) while enhancing the appearance of keratins characteristic of undifferentiated cells [58–60]. Retinoids decrease cornified envelope formation [61] by decreasing type I (or type K) transglutaminase [62] and substrate (involucrin, loricrin) [62,63] levels as well as the expression of filaggrin [64]. These effects are opposite to and antagonize the prodifferentiation actions of 1,25(OH)2D3. Like the actions of 1,25(OH)2D3, many of the actions of retinoic acid (RA) are mediated through changes in gene expression. Two members of the retinoic acid receptor family (RARα and RARγ), whose structures and mechanisms of action are homologous to steroid hormone receptors (including the VDR), have been identified in keratinocytes [65–67]. It was found that epidermal differentiation was blocked in a transgenic mouse with a dominant negative form of RAR in the skin [68]. The RARs are found in transformed as well as normal keratinocytes [69,70], although their expression in SCC may not be normal [69]. Certain retinoic acid metabolites (e.g., 9-cis-retinoic acid) also bind to another family of receptors, the retinoid X receptors (RXRs). The major member of this family in skin is RXRα [67,71]. As will be discussed further in the section on the hair follicle, RXRα is an important partner for VDR in regulating gene expression (also see Chapter 11). An epidermal specific deletion of RXRα results in a phenocopy of the VDR-null mouse with respect to hair follicle cycling and epidermal differentiation [72]. It is likely that RA and 1,25(OH)2D3 interact at the genomic level in their control of keratinocyte differentiation [73–75]. Furthermore, since RARs and VDR each can form heterodimers with RXR [76–79], both proteins may compete for the RXR pool within a given cell. Thus, the antagonism that exists between retinoic acid and 1,25(OH)2D on keratinocyte differentiation may have an explanation at the molecular level.

DANIEL D. BIKLE

2. CYTOKINES AND PEPTIDE GROWTH FACTORS

Keratinocytes produce an array of cytokines and growth factors, many of which have autocrine activity, and respond to still other cytokines and growth factors produced by stromal cells in the dermis [80,81]. Transforming growth factor-α (TGFα) is produced by the keratinocyte and acts through the epidermal growth factor (EGF) receptor to stimulate proliferation and migration [82]. Transforming growth factors-β1 and -β2 are also produced by keratinocytes [83,84], but they inhibit proliferation [85]. TGFβ exerts a number of effects on keratinocytes including reducing the differentiation markers Kl and filaggrin [60], increasing fibronectin, laminin, and α1 type IV collagen [60,86,87], increasing type II transglutaminase without altering type I transglutaminase [88], and decreasing c-myc expression [89]. The c-myc gene has been shown to have a response element for ΤGFβ [89], but whether the other actions of TGFβ are mediated through a similar response element in other genes remains to be demonstrated. Even though TGFα and TGFβ exert opposite effects on keratinocyte proliferation, neither promotes keratinocyte differentiation. Basic fibroblast growth factor (bFGF) is produced by keratinocytes [90] and stimulates their proliferation [91]. A related molecule, keratinocyte growth factor (KGF), is produced by stromal fibroblasts but stimulates the proliferation of keratinocytes [92]. The ability of EGF (or TGFα), bFGF, and KGF to stimulate keratinocyte proliferation is markedly enhanced by the presence of high concentrations of insulin, presumably acting via the IGF-I receptor, suggesting that insulin-like growth factors (IGF-I and -II) are also keratinocyte mitogens [86,91,93,94]. Receptors for IGF-1 on keratinocytes have been found [95]. TNFα is produced by keratinocytes [81] and promotes their differentiation with only a modest antiproliferative effect [32]. IFN-γ, on the other hand, is not made by keratinocytes, but it markedly inhibits their proliferation with little effect on differentiation [7,96]. IFN-γ increases class II antigens (HLA-DR) [30,96] and the intercellular adhesion molecule (ICAM-1) [97] of these cells. Keratinocytes produce platelet-derived growth factors (PDGF) but do not have the PDGF receptor [81]. Parathyroid hormone related peptide (PTHrP) is also produced by keratinocytes [98], but, like PDGF, receptors for PTHrP on normal keratinocytes have been difficult to demonstrate [99,100]. Nevertheless, PTHrP and/or its C-terminal fragments have been reported to inhibit or stimulate keratinocyte proliferation [101,102], and a 7-34 PTH antagonist has been claimed to promote hair growth by reversing the antiproliferative effects of PTHrP on the epidermis [101]. Interleukins 1 (α and β) [103,104], 3 [105], 6 [106],

615

CHAPTER 35 Vitamin D: Role in Skin and Hair

accumulates in large amounts in the cell and the intercellular matrix [51]. This gradient of calcium may provide the driving force for differentiation in intact epidermis [53]. However, most of the information regarding calcium-induced differentiation comes from in vitro studies with cultured keratinocytes. In low calcium-containing medium, keratinocytes proliferate readily but differentiate slowly if at all and remain as a monolayer in culture. On switching the cells to higher calcium concentrations (referred to as the calcium switch), keratinocytes undergo a coordinated set of responses at both the genomic and nongenomic levels that eventuates in a stratified culture in which the cells contain many of the features of the differentiated epithelium. For reasons that are not apparent, mouse keratinocytes are more sensitive to the antiproliferative effects of calcium and require lower concentrations of calcium for differentiation than do human

and 8 [107] are all produced by keratinocytes, as are the colony stimulating factors (CSF) granulocyte– macrophage (GM)-CSF [108], G-CSF, and M-CSF [109]. The roles the interleukins and colony-stimulating factors play in keratinocyte proliferation, migration, and differentiation are uncertain, but acting in a coordinated fashion they probably regulate important aspects of wound healing, inflammation, and cell growth [110] in the epidermis.

C. Calcium-Regulated Differentiation Calcium is the best studied prodifferentiating agent for keratinocytes (Fig. 4). In vivo, a calcium gradient exists in the epidermis such that in the basal and spinous layers calcium is primarily intracellular and in low amounts, but in the upper granular layers calcium

Ca2+

Calcium regulation of differentiation

CaR PIP2

C-SRC

Gq

?

+

PIP3

PLCγ DG + IP3

PI3K

PIP3

PLCβ + PIP2

DG + IP3 PIP2 +

γ

cSR C

C PL

ER

ATP2C1

IP3R

IP3R

ATP2A2

DG +

PI3K +

Ca2+

PIP3 +

Ca2+

IP3 PIP2 Ca2+

Golgi

Differentiation

FIGURE 4 Calcium sensing mechanism of the keratinocyte. The keratinocyte contains a calcium receptor that increases intracellular calcium by activating phospholipase C-β (PLC-β) via a G-protein-coupled process. Calcium also activates PLC-γ via a mechanism requiring c-src and phosphatidyl inositol 3 kinase (PI3K) activity. In addition to the initial effect of calcium to activate these PLCs, calcium also induces their synthesis. Both PLCs hydrolyze phosphatidylinositol bisphosphate (PIP2) to inositol trisphosphate (IP3) and diacylglycerol (DG). IP3 releases calcium (Ca2+) from intracellular stores such as the endoplasmic reticulum (ER) and Golgi by activating its receptor (IP3R), which functions as a calcium channel in these organelles. Besides the IP3R, the ER and Golgi each contain organelle-specific calcium pumps (CaATPases) that pump calcium into these organelles. The Golgi may be at least as critical for calcium signaling in the keratinocyte as the ER.

616 keratinocytes; however, the qualitative effects of calcium on human and murine cells are comparable. 1. CHANGES INDUCED BY THE CALCIUM SWITCH

Within minutes to hours of the calcium switch, morphological changes are apparent, with rapid development of cell-to-cell contact [8], desmosome formation [111], and a realignment of actin and keratin bundles near the cell membrane at the point of intercellular contacts [112]. Desmoplakin (a component of desmosomes), fodrin (an actin and calmodulin binding spectrin-like protein), and calmodulin are redistributed to the membrane shortly after the calcium switch by a mechanism that is blocked by cytochalasin, an agent that disrupts microfilament reorganization [112–114]. These effects do not appear to be under genomic control, although this has not been tested rigorously. Within hours to days of the calcium switch, the cells begin to make involucrin [9,62,115], loricrin [63], transglutaminase [9,62,115], keratins K1 and K10 [116], and filaggrin [116], and they start to form cornified envelopes [9,116]. As evidenced by a rise in mRNA levels for these proteins following the calcium switch [63,115,116], these effects of calcium represent genomic actions. Calcium response regions have been identified in the involucrin [117] and Kl [118] genes. The redistribution of integrin isoforms within days following the calcium switch [39,41,119] may participate in the mechanism by which cells begin to stratify. Calcium-induced increases in TGFβ [83,120] may contribute to the decrease in proliferation that accompanies the calcium switch, but, as discussed earlier, TGFβ does not stimulate differentiation. 2. ROLE OF INTRACELLULAR CALCIUM

The mechanisms by which calcium exerts its effects on keratinocyte differentiation are multiple. The intracellular free calcium ion concentration ([Ca2+]i) increases as keratinocytes differentiate, correlating closely with their ability to form cornified envelopes [6]. Raising the extracellular calcium concentration ([Ca2+]0) increases [Ca2+]i [6,114–118]. This response is saturable [6]. The response of [Ca2+]i to [Ca2+]0 is multiphasic and changes with differentiation. In undifferentiated keratinocytes and in transformed keratinocytes that are unable to differentiate, the switch to higher [Ca2+]0 results in an initial spike of [Ca2+]i that is followed by a plateau level that persists as long as the [Ca2+]0 remains elevated [121–123]. As the cells differentiate, this acute response to [Ca2+]0 is lost [122,123]. Lanthanum, which blocks calcium entry, blocks this response to [Ca2+]0, indicating that much of the rise in [Ca2+]i following a change in [Ca2+]0 is dependent on calcium entry [124], an increase of which has been shown following the

DANIEL D. BIKLE

calcium switch [123,125]. L-type calcium channel blockers do not prevent the rise in calcium uptake [125]. However, several channels have been identified in the keratinocyte membrane that are candidates for mediating calcium induced calcium influx [126–129]. These include calcium activated chloride channels which could hyperpolarize the membrane and increase calcium influx [126,127], nonselective cation channels (NSCC) in which the open time is increased by calcium [128], nicotinic acetylcholine receptors [129], and a recently discovered cGMP gated channel [130] that contains a calcium/calmodulin regulatory element. Furthermore, recent studies from our laboratory (Tu and Bikle, unpublished) suggest that the trp channels 1 and 4 may carry much of this calcium influx and contribute to the store operated channel function in these cells. The prolonged increase in [Ca2+]i after elevation of [Ca2+]0 stands in contrast to the response of [Ca2+]i to ATP [131–133]. ATP increases [Ca2+]i acutely and transiently at all stages of differentiation. The pool from which internal Ca2+ is released by ATP is intracellular and is rapidly depleted by a single dose of ATP, although ionomycin can mobilize additional intracellular calcium. As ATP inhibits rather than promotes keratinocyte differentiation [134], the sustained increase in [Ca2+]i following a [Ca2+]0 increase appears to be essential for the differentiation process. 3. THE CALCIUM RECEPTOR

The acute response of the keratinocyte to calcium resembles that of the parathyroid cell [135], which senses [Ca2+]0 via a seven-transmembrane domain, GTP binding protein-coupled calcium receptor (CaR) [136,137] (see Chapter 31). This receptor was originally discovered in the parathyroid gland, but we have identified the same structure in the keratinocyte [123]. The human CaR cDNA predicts a structure with 1078 amino acids with a calculated molecular weight of 120 kDa. The first 612 residues are predicted to be in the extracellular domain, residues 613–862 in the transmembrane and connecting loops domain, and residues 863–1078 in the C-terminal intracellular tail [137]. However, we [138] also observed that the keratinocyte produces an alternatively spliced variant of the CaR (CaRalt) as it differentiates which lacks exon 5 and so would be missing residues 461–537 in the extracellular domain. As mentioned earlier, keratinocytes lose their ability to sense calcium with differentiation. This change in calcium responsiveness is associated with the switch from the expression of the full-length calcium receptor (CaRfl) to the alternatively spliced form (CaRalt). We [139] discovered that the currently available mouse model in which CaRfl was knocked out by insertion of a neomycin cassette into exon 5 continues

CHAPTER 35 Vitamin D: Role in Skin and Hair

to produce CaRalt. Nevertheless, the epidermis of this mouse is abnormal, containing markedly lower levels of the terminal differentiation markers loricrin and profilaggrin, and the keratinocytes from this mouse fail to respond to calcium with a substantial rise in Cai [139] indicating the requirement for CaRfl in terminal differentiation of mouse epidermis. This does not mean that CaRalt is without function. However, only CaRfl mediates calcium-induced IP3 elevation [139]. Furthermore, when both forms were transfected into keratinocytes cotransfected with an INV promoter/ luciferase gene construct containing the calciumresponse element, only CaRfl enhanced the ability of calcium to stimulate the INV gene [138]. Blocking the production of the CaR (both forms) with an antisense construct in keratinocytes decreased the ability of calcium to raise [Ca2+]i and induce the INV and TG genes [140]. Thus CaRfl is a necessary component of the calcium-sensing apparatus in keratinocytes, but the alternatively spliced form may play an ancillary role. 1,25(OH)2D increases the CaR mRNA levels and prevents their decrease with time [141]. Furthermore, 1,25(OH)2D potentiates the ability of these cells to respond to [Ca2+]0 with a rise in [Ca2+]i [141]. Thus, the CaR appears to be important in mediating the internal Ca2+ response of keratinocytes to [Ca2+]0, and it provides a mechanism by which 1,25(OH)2D3 can regulate calcium-induced epidermal differentiation. 4. PHOSPHOINOSITIDE METABOLISM

The calcium switch stimulates phosphoinositide metabolism, which potentially could provide additional second messengers for mediating its effects on the keratinocyte [142–145]. The main enzymes involved are phospholipase C (PLC) β and γ, which hydrolyze phosphatidyl inositol to the important second messengers inositol trisphosphate (IP3) and diacylglycerol (DG). Both calcium and 1,25(OH)2D induce these enzymes [146,147]. As for the response of [Ca2+]i to [Ca2+]0, the rise in IP3 and DG is both immediate and prolonged following the calcium switch. Other agents such as ATP raise IP3 levels at least as effectively as calcium and yet do not stimulate differentiation [134]. Just as the rise in [Ca2+]i after ATP is transient, so is the rise in IP3. Conceivably, the prolonged rise in [Ca2+]i and IP3 after increases in [Ca2+]0, compared to the transient effects of ATP, contributes to the ability of [Ca2+]0 and not ATP to stimulate differentiation. This prolonged increase in IPs appears to be due to calcium activation of PLC-γ1 [148], although the initial increase in IP3 and [Ca2+]i after the calcium switch appears to be mediated by PLC-β. This extended activation of PLC-γ1 may be mediated by a calcium-induced increase in src family

617 tyrosine kinases (known activators of PLC-γ1) [149] as fyn has been shown to increase after the calcium switch in mouse keratinocytes, and keratinocytes from fyn-deficient mice fail to differentiate normally in response to calcium [150]. Earlier studies from the Krueger group [151] demonstrated that calciuminduced differentiation in human keratinocytes was associated with a rapid increase in c-src activity and a decline in c-yes activity. In human keratinocytes we observed that c-src rather than fyn is increased after the calcium switch, and its inhibition blocks calcium induced differentiation (Xie and Bikle, unpublished). In addition, we have observed an increase in PI3K activity after calcium (as assessed by phosphorylation of the p85α regulatory subunit) which if blocked prevents calcium activation of PLC-γ1. The critical role for PLC-γ1 in mediating calcium-induced differentiation is demonstrated by recent studies [147] in which we blocked PLC-γ1 production with an antisense construct and showed that calcium-induced differentiation was inhibited. Inhibition of PLC-γ1 production does not block the acute rise in [Ca2+]i after the calcium switch, although PLC inhibitors do, indicating that PLC-β is most likely the PLC responsible for the acute response to calcium. Of the two PLC-γ isoforms, 1 and 2, only PLC-γ1 is found in the keratinocyte. The catalytic domains, X and Y, of PLC-γ1 are separated by two SH2 domains and one SH3 domain [149]. The SH2 domains are responsible for binding to phosphotyrosines such as are found on the activated EGFR. The SH3 domain binds to proteins with proline-rich sequences. A pleckstrin homology (PH) domain is found in the N-terminal portion of the molecule, with two half PH domains on either side of the SH2 and SH3 domains. These PH domains enable PLC-γ1 binding to phosphatidylinositols (PIs) in the membrane. A C-terminal C2 domain also enables binding of PLC-γ1 to phospholipids and calcium. Growth factor receptor kinases such as EGFR phosphorylate PLC-γ1 at three tyrosines: 771, 783, and 1254 [152]. Phosphorylation at tyr783 is essential; phosphorylation at tyr1254 maximizes activity, whereas phosphorylation at tyr 771 is dispensable for activity [152] as determined by mutagenesis studies. The SH2 domains are essential for binding to the activated growth factor receptors. However, receptor tyrosine kinases are not the only mechanism for activating PLC-γ1. Src family kinases have also been shown to activate PLC-γ1 in a number of systems [153,154], but it is not clear whether this is a direct effect of these kinases on PLC-γ1. Similarly PI3K activity is associated with PLC-γ1 activation [153–155]. Although PI3K has protein kinase activity (ser/thr kinase), its principal mechanism of action is to phosphorylate membrane phosphatidylinositols in

618 the 3′ position [156]. PLC-γ1 binds to PI3,4,5P3 with high affinity through both its PH domain [157] and its C-terminal SH2 domain [158]. Binding of PLC-γ1 to PI3,4,5P3 stimulates its activity in the absence of tyrosine phosphorylation [159]. PI3Ks are activated by multiple means [156]. The class 1 enzymes are composed of a regulatory and a catalytic subunit. The class 1A enzymes have one of four regulatory subunits: p85α and β, p55, and p50 and three catalytic subunits: p110α, β, δ. The SH2 and SH3 domains of the regulatory units bind them to phosphotyrosines on activated receptor tyrosine kinases or other phosphotyrosine-containing proteins such as the IRS family. Phosphorylation of the regulatory subunit then releases the catalytic subunit, which is now active. Src kinases may activate PI3K by this mechanism. Class 1B PI3K has one known catalytic subunit, p110γ, and one known regulatory subunit, p101. This isoform is activated by the β/γ subunits of G proteins, such that agonists of G protein–coupled receptors including CaR may activate PI3K by this mechanism. The p85α regulatory subunit and both p110α and γ are found in keratinocytes. Inhibition of src kinase activity and PI3K activity each block the ability of calcium to activate PLC-γ1, whereas inhibition of the EGFR fails to do so. Furthermore, when src kinase or PI3K activity is inhibited, calcium fails to induce differentiation. 5. PROTEIN KINASE C

The rise in DG and [Ca2+]i following the calcium switch also results in protein kinase C (PKC) activation. Phorbol esters, which bind to and activate PKC, are well-known tumor promoters in skin [160–163]. However, the initial effects of phorbol esters in vitro are to promote differentiation in cells grown in low calcium [132–134,136–142], effects that are potentiated by calcium [164]. Phorbol esters stimulate PKC, and PKC inhibitors block the ability of both calcium and phorbol esters to promote differentiation [165,166]. Nevertheless, differences between the acute effects of phorbol esters and calcium on differentiation are clear. Phorbol esters block rather than promote calcium induction of Kl and K10 (early differentiation markers), in contrast to their synergistic effects with calcium on the later differentiation markers such as involucrin, loricrin, and filaggrin [167,168]. Calcium does not duplicate all the changes in protein phosphorylation caused by phorbol esters [144,169,170], and phorbol esters, unlike calcium, do not activate the phospholipase C (PLC) pathway [142,148]. Calcium activation of the PLC pathway, by increasing the amount of DG, an endogenous activator of PKC, is a likely means by which calcium stimulates PKC [170], whereas phorbol esters tend to reduce hormonal activation of PLC and the increase in [Ca2+]i

DANIEL D. BIKLE

at least in other cells [171,172]. Thus, it seems certain that although calcium-regulated differentiation involves the PKC pathway, the effects of calcium are not solely due to PKC activation and are not mimicked in their entirety by phorbol esters. Regardless of the fact that PKC activation cannot explain all of the effects of calcium on keratinocyte differentiation, it clearly plays a major role. However, the study of PKC in differentiation is complicated by the large number of isozymes of PKC in the epidermis, most of which are separate gene products and under different modes of regulation and distribution within the epidermis. Mouse keratinocytes contain PKC-α, δ, ε, η, and ζ [173]. Transformed mouse keratinocytes lose their responsiveness to calcium but retain the same complement of isozymes [173,174]. Human keratinocytes contain the same set of isozymes as mouse keratinocytes [175,176]. HaCaT cells (an immortalized human keratinocyte cell line) contain PKC-α, δ, ε, and ζ (no PKCη) [177]. PKC-α is a classic PKC isozyme and is activated by calcium, phorbol esters, and DG. PKC-δ, ε, and η are novel PKCs that are activated by phorbol esters and DG, like the classic PKCs, but are not activated by calcium. PKC-ζ is an atypical PKC that does not respond to calcium or phorbol esters. The cDNAs for these isozymes have been sequenced, and the structural basis for these differences is known [178]. During the first 48 hr after the calcium switch in mouse keratinocytes, the translocation of PKC-α from the cytosol to the membrane paralleled in time the induction of loricrin and profilaggrin better than did the changes in the other isozymes [179,180]. Down-regulation of PKC-α by bryostatin-1 and 12-deoxyphorbol 13-phenylacetate (DPP) also correlated with the inhibition of calcium-induced loricrin and profilaggrin expression better than did the effects of these phorbol esters on the other isozymes [179,181]. Furthermore, blocking the expression of PKC-α with antisense oligonucleotides prevented calcium induction of a number of differentiation markers [179,180]. Thus, PKC-α may be the major isozyme associated with calcium-induced differentiation, although PKC-δ has also been shown to have an important role [182]. The actual mechanism by which PKC induces differentiation is not clear [183–186], although a mechanism involving transcription factors in the Fos and Jun families acting on their AP-1 sites in the promoter regions of the genes involved with differentiation seems likely. PKC activation leads to a rapid increase in expression of c-fos and c-jun [132,133,187] and increased binding to AP-1 sites as assessed by gel retardation studies [188]. However, c-Fos and c-Jun are not the only transcription factors capable of binding to AP-1 sites. Other members of the Fos and Jun families (Fra-1, Fra-2, Jun B, Jun D)

CHAPTER 35 Vitamin D: Role in Skin and Hair

have been found in keratinocytes [189] and are differentially distributed throughout the epidermis. The distal AP-1 site in the involucrin gene promoter region was found to bind Fra-1, Jun B, and Jun D on gel retardation analysis [190]. A dominant negative mutant of c-jun [78], which blocks c-jun/fos-regulated transcriptional activity in AP-1 regulated genes such as prolactin, markedly stimulated transcriptional activity of involucrin gene constructs [168] suggesting that some members of the Fos-Jun family are playing an inhibitory role rather than a stimulatory role in involucrin gene transcription. As both keratin 1 and involucrin contain an AP-1 site within their calcium-responsive regions, a role for members of the Fos and Jun family in calcium-induced differentiation appears likely. As will be discussed later this AP-1 site also is close to the vitamin D response element in the involucrin gene, and when mutated blocks the ability of 1,25(OH)2D to induce this gene [191]. 6. CALMODULIN

Calmodulin also participates in the response of the cell to calcium. Calmodulin levels rise after the calcium switch under circumstances in which proliferation is increased [192,193], and calmodulin antagonists block proliferation [193–195]. Calmodulin and calcium may activate calmodulin/calcium-dependent kinases that could phosphorylate and thus activate critical transcription factors [196]. Calmodulin moves to the membrane along with fodrin [114] and desmoplakin [113] (calmodulin binding proteins) following the calcium switch. Calmodulin antagonists block the development of desmosome formation [197] and cell to cell contacts. At the membrane, calmodulin may participate not only in the formation of intercellular bridges but in the regulation of calcium influx and efflux as well.

D. l,25(OH)2D-Regulated Differentiation The observation that 1,25(OH)2D induces keratinocyte differentiation was first made by Hosomi et al. [3] and provided a rationale for the previous and unexpected finding of 1,25(OH)2D receptors in the skin [2]. As discussed earlier, 1,25(OH)2D is likely to be an autocrine or paracrine factor for epidermal differentiation since it is produced by the keratinocyte, but under normal circumstances keratinocyte production of 1,25(OH)2D does not appear to contribute to circulating levels [1,10]. The receptors for and the production of 1,25(OH)2D vary with differentiation [4,198,199] in a manner that suggests feedback regulation; both are reduced in the later stages of differentiation. 1,25(OH)2D increases involucrin, transglutaminase activity, and

619 cornified envelope formation at subnanomolar concentrations in preconfluent keratinocytes [3,5,6,36,200]. At these concentrations, 1,25(OH)2D has been found to promote proliferation in some studies [201–203], although the antiproliferative actions are most frequently observed, especially when concentrations above 10−9 M are employed. The mechanisms underlying the proliferative actions are not known. The antiproliferative effects are accompanied by a reduction in the mRNA levels for c-myc [204] and increases in the cell cycle inhibitors p21cip and p27kip. Stimulation of differentiation is accompanied by the rise in mRNA and protein levels of involucrin and transglutaminase [115]. Other likely examples of genomic regulation by 1,25(OH)2D include modulation of the skin calcium binding protein [205], induction of the 25OHD 24-hydroxylase [10], decreases in parathyroid hormone-related peptide (PTHrP) [206] and EGF receptor gene expression [204], and increases in TGFβ [207] and plasminogen activator inhibitor [208] gene expression. Where examined, these effects of 1,25(OH)2D can be reproduced by 25OHD [11,36], presumably because of endogenous conversion of 25OHD to 1,25(OH)2D, but are not observed with the biologically inactive β isomer of 1,25(OH)2D [5] (the natural isomer is lα,25(OH)2D). The mechanisms by which 1,25(OH)2D alters keratinocyte differentiation are not fully elucidated (Fig. 5). The VDR is critical for the genomic actions of 1,25(OH)2D; its role, if any, in potential nongenomic actions of 1,25(OH)2D remains to be determined (see Chapter 23). An acute increase in [Ca2+]i associated with an acute increase in phosphoinositide turnover (producing a rise in both IP3 and DG) has been observed in several studies [209–213]. Not all investigators (including ourselves) have been able to reproduce these acute effects of 1,25(OH)2D [123], although a gradual rise in [Ca2+]i and cornified envelope formation is observed [6]. The rise in [Ca2+]i, IP3, and DG is accompanied by translocation of PKC to the membrane [211]. Down-regulation of PKC and inhibition of its activity have been reported to block the ability of 1,25(OH)2D to stimulate cornified envelope formation [211]. However, the role of PKC in mediating or interacting with 1,25(OH)2D in its effects on keratinocyte differentiation remains virtually unexplored. Calcium and 1,25(OH)2D interact in their ability to inhibit proliferation and stimulate involucrin and transglutaminase gene expression [115]. The higher the [Ca2+]0, the more sensitive is the keratinocyte to the antiproliferative effect of 1,25(OH)2D (and vice versa) [200]. The interaction on gene expression is more complex. Both calcium (in the absence of 1,25(OH)2D) and 1,25(OH)2D (at 0.03 mM Ca2+) raise the mRNA levels

620

DANIEL D. BIKLE

Vitamin D regulation of differentiation Ca2+

Ca2+ PIP2

RACK Ca Receptor

GTP

G

PLC

GDP

DG

+ IP3

1,25(OH)2D3

PKC Golgi ER

Ca2+ i

IP3R

Ca2+ i

nucleus AP-1 factors CaRE/AP1

Involucrin Loricrin

TG

CE

FIGURE 5 Regulation by 1,25(OH)2D of keratinocyte differentiation. The calcium receptor and the PLCs are induced by 1,25(OH)2D, which enhances the ability of calcium to raise intracellular calcium, IP3, and diacylglycerol (DG) via the mechanisms described in Fig. 4. The rise in these second messengers leads to protein kinase activation (PKC) in part by translocation of PKCs to their specific membrane receptors (RACK) and opening up of calcium channels in the plasma membrane. 1,25(OH)2D in combination with the increase in intracellular calcium and the AP-1 transcription factors activated by PKC stimulates differentiation by inducing the substrates [e.g., involucrin (Inv) and loricrin] for cornified envelope (CE) formation as well as the enzyme transglutaminase (TG), which cross-links these substrates into the CE.

for involucrin and transglutaminase in a dose-dependent fashion. The stimulation is synergistic at intermediate concentrations of calcium (0.1 mM) and 1,25(OH)2D (10−10 M), but inhibition is observed in combination at higher concentrations. The synergism is more apparent at earlier times after the calcium switch (4 hr) than later (24–72 hr), when increased turnover of the mRNA by the higher combined concentrations of calcium and 1,25(OH)2D becomes dominant. At least one explanation for the synergism in the induction of involucrin is that the calcium response element (CaRE) and vitamin D response element (VDRE) in the involucrin promoter are quite close spatially [191]. Mutations in the AP-1 site within the CaRE block both calcium and 1,25(OH)2D induction of the involucrin gene, but mutations of the VDRE block only its response to 1,25(OH)2D. The molecular mechanisms underlying this complex, doseand time-dependent interaction between calcium and

1,25(OH)2D on keratinocyte differentiation are being actively investigated. The recent availability of mice lacking either the VDR or the 1α-hydroxylase has expanded our understanding of the role of 1,25(OH)2D in epidermal differentiation. Although the most striking feature of the VDR-null mouse is the development of alopecia (also found in many patients with mutations in the VDR— hereditary vitamin D resistance, see Chapters 20 and 72), these mice also exhibit a defect in epidermal differentiation as shown by reduced levels of involucrin, profilaggrin, and loricrin and loss of keratohyalin granules [214]. However, these changes can be reversed when the animals are placed on a high-calcium diet, unlike the abnormalities in the hair follicle (Bikle et al., unpublished). A different phenotype is observed in the 1α-hydroxylase null mouse (see also Chapter 7). These mice also show a reduction in levels of the epidermal

CHAPTER 35 Vitamin D: Role in Skin and Hair

differentiation markers, but unlike the VDR-null animals, a high-calcium diet does not rescue this phenotype [34a]. Furthermore, the 1α-hydroxylase-null animals have a retarded recovery of barrier function when the barrier is disrupted, which on ultrastructural examination is associated with an impaired reestablishment of the calcium gradient in the epidermis [34a]. Moreover, the 1α-hydroxylase-null animals do not have a defect in hair follicle cycling. The difference in phenotypes between these genotypes is surprising and points to the possibility that the 1α-hydroxylase in the epidermis may be doing more than making 1,25(OH)2D, just as the phenotype in the VDR-null animal suggests that the VDR may have functions independent of 1,25(OH)2D.

IV. REGULATION OF HAIR FOLLICLE CYCLING The hair follicle cycle is divided into three main stages: anagen, catagen, and telogen. The duration of these stages varies from location to location on the body and between genders. Furthermore, there are two types of cycles: developmental and postnatal. The first or developmental cycle is initiated during embryogenesis. The follicle develops from specific regions of the epidermis called placodes. The development, number, and placement of these placodes is under the control of a number of factors, none of which are influenced by vitamin D. The follicle is induced to grow by its interaction with specialized mesenchymal cells in the dermis called the dermal papilla. Following this first cycle, which leads to the initial coat of hair, the follicle undergoes repetitive cycling until senescence. Growth of the follicle occurs during anagen. The length of the hair is dependent on the duration of anagen. During this stage the follicle grows through the dermis into the subcutaneous tissue. Following anagen, the follicle enters catagen, during which massive apoptosis occurs primarily in the cells of the proximal follicle (the dermal portion), and the hair shaft produced during anagen is generally shed. The distal portion of the follicle (epidermal portion also giving rise to the sebaceous gland) is less affected. At the end of catagen the follicle enters telogen, the resting phase. A new cycle then begins with anagen. The regulatory elements that control the transition from one stage to the next are poorly understood. Alopecia is a well-known part of the phenotype of many patients with mutations in their vitamin D receptor (VDR) [215,216], a syndrome currently known as hereditary vitamin D resistant rickets (HVDRR) (see Chapter 72). Although not all subjects present with alopecia, those with the more severe disease generally

621 demonstrate this phenotype [216]. In that vitamin D deficiency per se is not associated with alopecia, the explanation for this phenomenon has remained obscure. Exploration of the link between alopecia and VDR received a major boost with the development of the VDR-null mouse by two groups [217,218] (see Chapter 20). These mice develop their first coat of hair normally, but reinitiation of anagen following the first cycle or after depilation is impaired [219]. Reconstitution of the VDR to the VDR-null mouse skin using a tissue-specific promoter reverses the defect in hair growth without reversing the metabolic defects of skeletal growth retardation, hypocalcemia, and rickets otherwise associated with the VDR-null condition [220,221]. On the other hand, correction of the metabolic abnormalities with a high-calcium diet prevents the rickets and hyperparathyroidism but does not prevent the alopecia [222]. The control of hair follicle development and cycling is complex [223,224], and a large number of factors have been implicated in this process including members of the FGF family and their receptors, TGFβ family and their receptors, BMPs and their receptors and antagonists, sonic hedgehog and its receptor patch-1, IGF-I and its receptor and binding proteins, hepatocyte growth factor, hox genes, MSX 1 and 2, PTHrP, prolactin, interleukins, and the various participants in the wnt signaling pathways. Many of these gene products are implicated in placode and/or hair follicle development and so their overexpression or underexpression results in abnormalities in the first or developmental cycle of hair growth, phenotypes different from that seen in the VDR null mouse. However, overexpression of ornithine decarboxylase [225] or spermine/spermidine N′-acetyltransferase [226], conditional skin-specific deletion of RXRα [72], conditional skin-specific mutations of β catenin that do not otherwise block hair follicle development [227,228], and mutations in hairless (hr) [229] all result in a phenotype comparable to that seen in the VDR-null mouse. The hr mutations are of particular interest because both the mouse model [229] and the human disease [230,231] are phenocopies of the VDR-null mouse and human condition, respectively, with respect to the morphologic changes observed in hair cycling. In these models the abnormality leading to alopecia develops during catagen at the end of the first (developmental) cycle. The dissociation of the dermal papilla from the hair bulb by the end of catagen is thought to account for the failure to initiate the subsequent anagen [228,229]. The distal portion of the hair follicle including the sebaceous gland as well as the interfollicular epidermis is little affected [214,227–229], and the large dermal cysts that develop with time contain markers of the differentiated interfollicular

622 epidermis [214,227,228] suggesting their origin from the distal portion of the hair follicle or epidermis. In a clever mixing experiment, Sakai et al. [232] demonstrated that at least in the VDR-null mouse the defect responsible for the hair cycling abnormality lay in the epidermal keratinocytes, and not in the cells within the dermal papilla. A similar conclusion is likely in the hr mouse, since the hr gene is not expressed in the dermal papilla [229], although this has not been formally tested. In a recently completed study (Bikle et al., submitted), we observed that hr expression was increased in the VDR-null mouse, indicating that hr is not downstream of VDR. In determining the possible role that VDR plays in hair follicle cycling, an examination of other genotypes that produce a similar phenotype is useful. The phenotype observed with the RXRα-null epidermis is easy to understand. RXRα is the main RXR in the epidermis. VDR partners with RXR for most of its genomic actions [233], and RXR can promote the nuclear localization of VDR even in the absence of 1,25(OH)2D [234]. Thus the lack of RXRα could result in an ineffective VDR and reproduce the phenotype of the VDR-null mouse. The RXRα-null mouse skin also shares with the VDR-null mouse skin a hyperproliferative response around the dystrophic follicles that has not been observed in hr mice or the polyamine pathway mutants. This difference suggests a role for the VDR and its RXRα partner in controlling cellular proliferation in the epidermis and hair follicle that is independent of hr. A potential role for hr in VDR function in the hair follicle is attractive, but is not established. Hr is expressed in the epidermis, hair follicle, and brain [230]. Its importance in hair follicle cycling is clear [230,235], but its function, like that of VDR in this process, is not. Hr has characteristics of a transcription factor in that it resides in the nucleus; its structure contains a nuclear localization signal, a putative zinc finger, and three LXXLL motifs [236] like that found in coactivators that interact with nuclear hormone receptors such as VDR (see Chapter 16). In the brain, the thyroid receptor (TR) has been suggested as a partner for hr in that TR can bind to hr [237], and the hr gene has a thyroid hormone response element in its promoter [238] making it a potential target for thyroid hormone in the brain. Hr has also been shown to bind to VDR [239]. The increased expression of hr in the VDR null mouse indicates that VDR is not required for hr expression, but this does not rule out an interaction, the absence of which leads to increased hr expression by way of compensation. β-Catenin forms an important part of the wnt signaling pathway. It can function in the cytoplasm linking

DANIEL D. BIKLE

membrane bound cadherins to the cytoskeleton or act as a transcription factor in the nucleus in association with TCF/LEF [240]. Mice in which β-catenin is deleted specifically in the epidermis can show abnormalities either in hair follicle morphogenesis if the deletion occurs during embryogenesis or in hair cycling if deleted after birth [228]. In the latter case, the phenotype resembles that of VDR-null animals in that during the first catagen the hair follicles become dystrophic, the dermal papilla fails to remain associated with the follicle, anagen fails to be reinitiated, and dermal cysts eventually develop. The epidermis remains intact. β-Catenin is found primarily in the cytoplasm of the epidermis but in the nucleus of the cells of the outer root sheath in the hair follicle (Bikle et al., submitted). This location makes it a potential partner for VDR in mediating transcription. The clear requirement for the VDR in hair follicle cycling without a clear requirement for either 1,25(OH)2D or normal levels of calcium indicates that the function of VDR in the hair follicle will be different from its role in other tissues. Exactly what that role is remains to be determined.

References 1. Bikle DD, Nemanic MK, Whitney JO, Elias PW 1986 Neonatal human foreskin keratinocytes produce 1,25-dihydroxyvitamin D3. Biochemistry 25:1545–1548. 2. Stumpf WE, Sar M, Reid FA, Tanaka Y, DeLuca HF 1979 Target cells for 1,25-dihydroxyvitamin D3 in intestinal tract, stomach, kidney, skin, pituitary, and parathyroid. Science 206:1188–1190. 3. Hosomi J, Hosoi J, Abe E, Suda T, Kuroki T 1983 Regulation of terminal differentiation of cultured mouse epidermal cells by 1 alpha,25-dihydroxyvitamin D3. Endocrinology 113: 1950–1957. 4. Pillai S, Bikle DD, Elias PM 1988 1,25-Dihydroxyvitamin D production and receptor binding in human keratinocytes varies with differentiation. J Biol Chem 263:5390–5395. 5. Smith EL, Walworth NC, Holick MF 1986 Effect of 1 alpha, 25-dihydroxyvitamin D3 on the morphologic and biochemical differentiation of cultured human epidermal keratinocytes grown in serum-free conditions. J Invest Dermatol 86: 709–714. 6. Pillai S, Bikle DD 1991 Role of intracellular-free calcium in the cornified envelope formation of keratinocytes: differences in the mode of action of extracellular calcium and 1,25 dihydroxyvitamin D3. J Cell Physiol 146:94–100. 7. Bikle DD, Pillai S, Gee E, Hincenbergs M 1989 Regulation of 1,25-dihydroxyvitamin D production in human keratinocytes by interferon-gamma. Endocrinology 124(2):655–660. 8. Hennings H, Michael D, Cheng C, Steinert P, Holbrook K, Yuspa SH 1980 Calcium regulation of growth and differentiation of mouse epidermal cells in culture. Cell 19: 245–254. 9. Pillai S, Bikle DD, Mancianti ML, Cline P, Hincenbergs M 1990 Calcium regulation of growth and differentiation of

CHAPTER 35 Vitamin D: Role in Skin and Hair

10. 11.

12.

13.

13a. 14.

15.

16.

17. 18. 19.

20.

21.

22. 23. 24.

normal human keratinocytes: modulation of differentiation competence by stages of growth and extracellular calcium. J Cell Physiol 143:294–302. Bikle DD, Nemanic MK, Gee E, Elias P 1986 1,25Dihydroxyvitamin D3 production by human keratinocytes. Kinetics and regulation. J Clin Invest 78:557–566. Matsumoto K, Azuma Y, Kiyoki M, Okumura H, Hashimoto K, Yoshikawa K 1991 Involvement of endogenously produced 1,25-dihydroxyvitamin D-3 in the growth and differentiation of human keratinocytes. Biochim Biophys Acta 1092(3): 311–318. Lehmann B, Genehr T, Knuschke P, Pietzsch J, Meurer M 2001 UVB-induced conversion of 7-dehydrocholesterol to 1α,25-dihydroxyvitamin D3 in an in vitro human skin equivalent model. J Invest Dermatol 117(5):1179–1185. Lehmann B, Tiebel O, Meurer M 1999 Expression of vitamin D3 25-hydroxylase (CYP27) mRNA after induction by vitamin D3 or UVB radiation in keratinocytes of human skin equivalents—a preliminary study. Arch Dermatol Res 291:507–510. Masumoto O, Ohyama Y, Okuda K 1988 Purification and characterization of vitamin D 25-hydroxylase from rat liver mitochondria. J Biol Chem 263:14256–14260. Fu GK, Lin D, Zhang MY, Bikle DD, Shackleton CH, Miller WL, Portale AA 1997 Cloning of human 25-hydroxyvitamin D-1 alpha-hydroxylase and mutations causing vitamin D-dependent rickets type 1. Mol Endocrinol 11: 1961–1970. Barbour GL, Coburn JW, Slatopolsky E, Norman AW, Horst RL 1981 Hypercalcemia in an anephric patient with sarcoidosis: evidence for extrarenal generation of 1,25-dihydroxyvitamin D. N Engl J Med 305:440–443. Lambert PW, Stern PH, Avioli RC, Brackett NC, Turner RT, Greene A, Fu IY, Bell NH 1982 Evidence for extrarenal production of 1α,25-dihydroxyvitamin D in man. J Clin Invest 69(3):722–725. Littledike ET, Horst RL 1982 Metabolism of vitamin D3 in nephrectomized pigs given pharmacological amounts of vitamin D3. Endocrinology 111(6):2008–2013. Bikle DD, Halloran BP, Riviere JE 1994 Production of 1,25 dihydroxyvitamin D3 by perfused pig skin. J Invest Dermatol 102(5):796–798. Xie Z, Munson SJ, Huang N, Portale AA, Miller WL, Bikle DD 2002 The mechanism of 1,25-dihydroxyvitamin D(3) autoregulation in keratinocytes. J Biol Chem 277(40): 36987–36990. Rasmussen H, Wong M, Bikle D, Goodman DB 1972 Hormonal control of the renal conversion of 25-hydroxycholecalciferol to 1,25-dihydroxycholecalciferol. J Clin Invest 51:2502–2504. Rost CR, Bikle DD, Kaplan RA 1981 In vitro stimulation of 25-hydroxycholecalciferol 1 alpha-hydroxylation by parathyroid hormone in chick kidney slices: evidence for a role for adenosine 3′,5′-monophosphate. Endocrinology 108:1002–1026. Bikle DD, Gee E 1989 Free, and not total, 1,25-dihydroxy vitamin D regulates 25-hydroxyvitamin D metabolism by keratinocytes. Endocrinology 124:649–654. Henry HL 1979 Response of chick kidney cell cultures to 1,25-dihydroxyvitamin D3. 467–474. Spanos E, Barrett DI, Chong KT, MacIntyre I 1978 Effect of oestrogen and 1,25-dihydroxycholecalciferol on 25-hydroxycholecalciferol metabolism in primary chick kidney-cell cultures. Biochem J 174(1):231–236.

623 25. Trechsel U, Bonjour JP, Fleisch H 1979 Regulation of the metabolism of 25-hydroxyvitamin D3 in primary cultures of chick kidney cells. J Clin Invest 64(1):206–217. 26. Reeve L, Tanaka Y, DeLuca HF 1983 Studies on the site of 1,25-dihydroxyvitamin D3 synthesis in vivo. J Biol Chem 258(6):3615–3617. 27. Shultz TD, Fox J, Heath H, 3rd, Kumar R 1983 Do tissues other than the kidney produce 1,25-dihydroxyvitamin D3 in vivo? A reexamination. Proc Natl Acad Sci USA 80(6): 1746–1750. 28. Pillai S, Bikle DD, Hincenbergs M, Elias PM 1988 Biochemical and morphological characterization of growth and differentiation of normal human neonatal keratinocytes in a serum-free medium. J Cell Physiol 134(2):229–237. 29. Zehnder D, Bland R, Williams MC, McNinch RW, Howie AJ, Stewart PM, Hewison M 2001 Extrarenal expression of 25-hydroxyvitamin D(3)-1α-hydroxylase. J Clin Endocrinol Metab 86(2):888–894. 30. Morhenn VB, Wood GS 1988 Gamma interferon-induced expression of class II major histocompatibility complex antigens by human keratinocytes. Effects of conditions of culture. Ann N Y Acad Sci 548:321–330. 31. Pillai S, Bikle DD, Eessalu TE, Aggarwal BB, Elias PM 1989 Binding and biological effects of tumor necrosis factor alpha on cultured human neonatal foreskin keratinocytes. J Clin Invest 83(3):816–821. 32. Bikle DD, Pillai S, Gee E, Hincenbergs M 1991 Tumor necrosis factor-alpha regulation of 1,25-dihydroxyvitamin D production by human keratinocytes. Endocrinology 129:33–38. 33. Trefzer U, Brockhaus M, Lotscher H, Parlow F, Budnik A, Grewe M, Christoph H, Kapp A, Schopf E, Luger TA and others 1993 The 55-kD tumor necrosis factor receptor on human keratinocytes is regulated by tumor necrosis factoralpha and by ultraviolet B radiation. J Clin Invest 92(1):462–470. 34. Wood LC, Elias PM, Sequeira-Martin SM, Grunfeld C, Feingold KR 1994 Occlusion lowers cytokine mRNA levels in essential fatty acid-deficient and normal mouse epidermis, but not after acute barrier disruption. J Invest Dermatol 103(6):834–838. 34a. Bikle DD, Chang S, Crumrine D, Elalieh H, Man M-Q, Choi EH, Dardenne O, Xie Z, St. Arnaud R, Feingold K, Elias PM 2004 25-Hydroxyvitamin D 1 alpha-hydroxylase is required for optimal epidermal differentiation and permeability barrier homeostasis. J Invest Dermatol 122:984–992. 35. Rheinwald JG, Beckett MA 1980 Defective terminal differentiation in culture as a consistent and selectable character of malignant human keratinocytes. Cell 22(2 Pt 2):629–632. 36. Bikle DD, Pillai S, Gee E 1991 Squamous carcinoma cell lines produce 1,25 dihydroxyvitamin D, but fail to respond to its prodifferentiating effect. J Invest Dermatol 97(3): 435–441. 37. Bikle DD, Tu CL, Xie Z, Oda Y 2003 Vitamin D regulated keratinocyte differentiation: Role of coactivators. J Cell Biochem 88(2):290–295. 38. Halloran BP, Schaefer P, Lifschitz M, Levens M, Goldsmith RS 1984 Plasma vitamin D metabolite concentrations in chronic renal failure: effect of oral administration of 25-hydroxyvitamin D3. J Clin Endocrinol Metab 59(6):1063–1069. 39. Marchisio PC, Bondanza S, Cremona O, Cancedda R, De Luca M 1991 Polarized expression of integrin receptors (alpha 6 beta 4, alpha 2 beta 1, alpha 3 beta 1, and alpha v beta 5) and their relationship with the cytoskeleton and

624

40.

41. 42. 43. 44.

45.

46.

47. 48.

49.

50. 51. 52.

53.

54. 55. 56. 57. 58.

DANIEL D. BIKLE

basement membrane matrix in cultured human keratinocytes. J Cell Biol 112(4):761–773. Peltonen J, Larjava H, Jaakkola S, Gralnick H, Akiyama SK, Yamada SS, Yamada KM, Uitto J 1989 Localization of integrin receptors for fibronectin, collagen, and laminin in human skin. Variable expression in basal and squamous cell carcinomas. J Clin Invest 84(6):1916–1923. Guo M, Kim LT, Akiyama SK, Gralnick HR, Yamada KM, Grinnell F 1991 Altered processing of integrin receptors during keratinocyte activation. Exp Cell Res 195(2):315–322. Moll R, Franke WW, Schiller DL, Geiger B, Krepler R 1982 The catalog of human cytokeratins: patterns of expression in normal epithelia, tumors and cultured cells. Cell 31(1):11–24. Eichner R, Sun TT, Aebi U 1986 The role of keratin subfamilies and keratin pairs in the formation of human epidermal intermediate filaments. J Cell Biol 102(5):1767–1777. Warhol MJ, Roth J, Lucocq JM, Pinkus GS, Rice RH 1985 Immuno-ultrastructural localization of involucrin in squamous epithelium and cultured keratinocytes. J Histochem Cytochem 33(2):141–149. Thacher SM, Rice RH 1985 Keratinocyte-specific transglutaminase of cultured human epidermal cells: relation to cross-linked envelope formation and terminal differentiation. Cell 40(3):685–695. Steven AC, Bisher ME, Roop DR, Steinert PM 1990 Biosynthetic pathways of filaggrin and loricrin—two major proteins expressed by terminally differentiated epidermal keratinocytes. J Struct Biol 104:150–162. Dale BA, Resing KA, Lonsdale-Eccles JD 1985 Filaggrin: a keratin filament associated protein. Ann NY Acad Sci 455:330–342. Mehrel T, Hohl D, Rothnagel JA, Longley MA, Bundman D, Cheng C, Lichti U, Bisher ME, Steven AC, Steinert PM and others 1990 Identification of a major keratinocyte cell envelope protein, loricrin. Cell 61:1103–1112. Elias PM, Menon GK, Grayson S, Brown BE 1988 Membrane structural alterations in murine stratum corneum: relationship to the localization of polar lipids and phospholipases. J Invest Dermatol 91(1):3–10. Hohl D 1990 Cornified cell envelope. Dermatologica 180:201–211. Menon GK, Grayson S, Elias PM 1985 Ionic calcium reservoirs in mammalian epidermis: Ultrastructural localization by ion-capture cytochemistry. J Invest Dermatol 84(6):508–512. Mauro T, Bench G, Sidderas-Haddad E, Feingold K, Elias P, Cullander C 1998 Acute barrier perturbation abolishes the Ca2+ and K+ gradients in murine epidermis: Quantitative measurement using PIXE. J Invest Dermatol 111:1198–1201. Elias PM, Ahn SK, Denda M, Brown BE, Crumrine D, Kimutai LK, Komuves L, Lee SH, Feingold KR 2002 Modulations in epidermal calcium regulate the expression of differentiation-specific markers. J Invest Dermatol 119: 1128–1136. Rheinwald JG, Green H 1975 Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinizing colonies from single cells. Cell 6(3):331–343. Wobach SB, Howe PR 1925 Tissue changes following deprivation of fat-soluble A vitamin. J Exp Med 43:753–777. Fell HB, Mellanby E 1953 Metaplasia produced in cultures of chick ectoderm by high vitamin A. J Physiol 119:470–488. Wobach SB, Howe PR 1933 Epithelial repair in recovery from vitamin A deficiency. J Exp Med 57:511–526. Fuchs E, Green H 1981 Regulation of terminal differentiation of cultured human keratinocytes by vitamin A. Cell 25(3): 617–625.

59. Gilfix BM, Eckert RL 1985 Coordinate control by vitamin A of keratin gene expression in human keratinocytes. J Biol Chem 260(26):14026–14029. 60. Choi Y, Fuchs E 1990 TGF-beta and retinoic acid: regulators of growth and modifiers of differentiation in human epidermal cells. Cell Regul 1(11):791–809. 61. Kline PR, Rice RH 1983 Modulation of involucrin and envelope competence in human keratinocytes by hydrocortisone, retinylacetate and growth arrest. Cancer Res 43:3203–3207. 62. Rubin AL, Parenteau NL, Rice RH 1989 Coordination of keratinocyte programming in human SCC-13 squamous carcinoma and normal epidermal cells. J Cell Physiol 138(1):208–214. 63. Hohl D, Lichti U, Breitkreutz D, Steinert PM, Roop DR 1991 Transcription of the human loricrin gene in vitro is induced by calcium and cell density and suppressed by retinoic acid. J Invest Dermatol 96:414–418. 64. Fleckman P, Dale BA, Holbrook KA 1985 Profilaggrin, a high-molecular-weight precursor of filaggrin in human epidermis and cultured keratinocytes. J Invest Dermatol 85(6): 507–512. 65. Vollberg TM Sr., Nervi C, George MD, Fujimoto W, Krust A, Jetten AM 1992 Retinoic acid receptors as regulators of human epidermal keratinocyte differentiation. Mol Endocrinol 6(5):667–676. 66. Kastner P, Krust A, Mendelsohn C, Garnier JM, Zelent A, Leroy P, Staub A, Chambon P 1990 Murine isoforms of retinoic acid receptor gamma with specific patterns of expression. Proc Natl Acad Sci USA 87(7):2700–2704. 67. Elder JT, Astrom A, Pettersson U, Tavakkol A, Krust A, Kastner P, Chambon P, Voorhees JJ 1992 Retinoic acid receptors and binding proteins in human skin. J Invest Dermatol 98(6 Suppl):36S–41S. 68. Saitou M, Sugai S, Tanaka T, Shimouchi K, Fuchs E, Narumiya S, Kakizuka A 1995 Inhibition of skin development by targeted expression of a dominant-negative retinoic acid receptor. Nature 374(6518):159–162. 69. Hu L, Crowe DL, Rheinwald JG, Chambon P, Gudas LJ 1991 Abnormal expression of retinoic acid receptors and keratin 19 by human oral and epidermal squamous cell carcinoma cell lines. Cancer Res 51(15):3972–3981. 70. Finzi E, Blake MJ, Celano P, Skouge J, Diwan R 1992 Cellular localization of retinoic acid receptor-gamma expression in normal and neoplastic skin. Am J Pathol 140(6): 1463–1471. 71. Torma H, Rollman O, Vahlquist A 1993 Detection of mRNA transcripts for retinoic acid, vitamin D3, and thyroid hormone (c-erb-A) nuclear receptors in human skin using reverse transcription and polymerase chain reaction. Acta Derm Venereol 73(2):102–107. 72. Li M, Indra AK, Warot X, Brocard J, Messaddeq N, Kato S, Metzger D, Chambon P 2000 Skin abnormalities generated by temporally controlled RXRalpha mutations in mouse epidermis. Nature 407(6804):633–636. 73. Schule R, Umesono K, Mangelsdorf DJ, Bolado J, Pike JW, Evans RM 1990 Jun-Fos and receptors for vitamins A and D recognize a common response element in the human osteocalcin gene. Cell 61(3):497–504. 74. Xie Z, Bikle DD 1997 Cloning of the human phospholipase C-gamma1 promoter and identification of a DR6type vitamin D-responsive element. J Biol Chem 272: 6573–6577. 75. Li XY, Xiao JH, Feng X, Qin L, Voorhees JJ 1997 Retinoid X receptor-specific ligands synergistically upregulate 1,25dihydroxyvitamin D3-dependent transcription in epidermal

CHAPTER 35 Vitamin D: Role in Skin and Hair

76.

77.

78.

79.

80. 81. 82.

83.

84.

85.

86.

87.

88.

89.

90.

keratinocytes in vitro and in vivo. J Invest Dermatol 108(4):506–512. Kliewer SA, Umesono K, Mangelsdorf DJ, Evans RM 1992 Retinoid X receptor interacts with nuclear receptors in retinoic acid, thyroid hormone and vitamin D3 signalling. Nature 355(6359):446–449. Yu VC, Delsert C, Andersen B, Holloway JM, Devary OV, Naar AM, Kim SY, Boutin JM, Glass CK, Rosenfeld MG 1991 RXR beta: a coregulator that enhances binding of retinoic acid, thyroid hormone, and vitamin D receptors to their cognate response elements. Cell 67(6):1251–1266. Forman BM, Casanova J, Raaka BM, Ghysdael J, Samuels HH 1992 Half-site spacing and orientation determines whether thyroid hormone and retinoic acid receptors and related factors bind to DNA response elements as monomers, homodimers, or heterodimers. Mol Endocrinol 6(3): 429–442. Glass CK, Devary OV, Rosenfeld MG 1990 Multiple cell type-specific proteins differentially regulate target sequence recognition by the alpha retinoic acid receptor. Cell 63(4):729–738. McKay IA, Leigh IM 1991 Epidermal cytokines and their roles in cutaneous wound healing. Br J Dermatol 124(6): 513–518. Ansel J, Perry P, Brown J, Damm D, Phan T, Hart C, Luger T, Hefeneider S 1990 Cytokine modulation of keratinocyte cytokines. J Invest Dermatol 94(6 Suppl):101S–107S. Coffey RJ, Jr., Derynck R, Wilcox JN, Bringman TS, Goustin AS, Moses HL, Pittelkow MR 1987 Production and auto-induction of transforming growth factor-alpha in human keratinocytes. Nature 328(6133):817–820. Glick AB, Danielpour D, Morgan D, Sporn MB, Yuspa SH 1990 Induction and autocrine receptor binding of transforming growth factor-beta 2 during terminal differentiation of primary mouse keratinocytes. Mol Endocrinol 4(1):46–52. Coffey RJ, Jr., Sipes NJ, Bascom CC, Graves-Deal R, Pennington CY, Weissman BE, Moses HL 1988 Growth modulation of mouse keratinocytes by transforming growth factors. Cancer Res 48(6):1596–1602. Shipley GD, Pittelkow MR, Wille JJ, Jr., Scott RE, Moses HL 1986 Reversible inhibition of normal human prokeratinocyte proliferation by type beta transforming growth factor-growth inhibitor in serum-free medium. Cancer Res 46(4 Pt 2): 2068–2071. DeLapp NW, Dieckman DK 1990 Effect of basic fibroblast growth factor (bFGF) and insulin-like growth factors type I (IGF-I) and type II (IGF-II) on adult human keratinocyte growth and fibronectin secretion. J Invest Dermatol 94(6):777–780. Vollberg TM, Sr., George MD, Jetten AM 1991 Induction of extracellular matrix gene expression in normal human keratinocytes by transforming growth factor beta is altered by cellular differentiation. Exp Cell Res 193(1):93–100. George MD, Vollberg TM, Floyd EE, Stein JP, Jetten AM 1990 Regulation of transglutaminase type II by transforming growth factor-β1 in normal and transformed human epidermal keratinocytes. J Biol Chem 265(19):11098–11104. Pietenpol JA, Holt JT, Stein RW, Moses HL 1990 Transforming growth factor beta 1 suppression of c-myc gene transcription: role in inhibition of keratinocyte proliferation. Proc Natl Acad Sci USA 87(10):3758–3762. Halaban R, Langdon R, Birchall N, Cuono C, Baird A, Scott G, Moellmann G, McGuire J 1988 Basic fibroblast growth factor from human keratinocytes is a natural mitogen for melanocytes. J Cell Biol 107(4):1611–1619.

625 91. O’Keefe EJ, Chiu ML, Payne RE, Jr. 1988 Stimulation of growth of keratinocytes by basic fibroblast growth factor. J Invest Dermatol 90(5):767–769. 92. Bottaro DP, Rubin JS, Ron D, Finch PW, Florio C, Aaronson SA 1990 Characterization of the receptor for keratinocyte growth factor. Evidence for multiple fibroblast growth factor receptors. J Biol Chem 265(22):12767–12770. 93. Krane JF, Murphy DP, Carter DM, Krueger JG 1991 Synergistic effects of epidermal growth factor (EGF) and insulin-like growth factor I/somatomedin C (IGF-I) on keratinocyte proliferation may be mediated by IGF-I transmodulation of the EGF receptor. J Invest Dermatol 96(4): 419–424. 94. Rubin JS, Osada H, Finch PW, Taylor WG, Rudikoff S, Aaronson SA 1989 Purification and characterization of a newly identified growth factor specific for epithelial cells. Proc Natl Acad Sci USA 86(3):802–806. 95. Misra P, Nickoloff BJ, Morhenn VB, Hintz RL, Rosenfeld RG 1986 Characterization of insulin-like growth factor-I/ somatomedin-C receptors on human keratinocyte monolayers. J Invest Dermatol 87(2):264–267. 96. Nickoloff BJ 1987 Binding of 125I-gamma interferon to cultured human keratinocytes. J Invest Dermatol 89(2): 132–135. 97. Nickoloff BJ, Griffiths CE, Barker JN 1990 The role of adhesion molecules, chemotactic factors, and cytokines in inflammatory and neoplastic skin disease—1990 update. J Invest Dermatol 94(6 Suppl):151S-157S. 98. Merendino JJ, Jr., Insogna KL, Milstone LM, Broadus AE, Stewart AF 1986 A parathyroid hormone-like protein from cultured human keratinocytes. Science 231(4736): 388–390. 99. Hanafin NM, Chen TC, Heinrich G, Segre GV, Holick MF 1995 Cultured human fibroblasts and not cultured human keratinocytes express a PTH/PTHrP receptor mRNA. J Invest Dermatol 105(1):133–137. 100. Orloff JJ, Kats Y, Urena P, Schipani E, Vasavada RC, Philbrick WM, Behal A, Abou-Samra AB, Segre GV, Juppner H 1995 Further evidence for a novel receptor for amino-terminal parathyroid hormone-related protein on keratinocytes and squamous carcinoma cell lines. Endocrinology 136(7):3016–3023. 101. Holick MF, Ray S, Chen TC, Tian X, Persons KS 1994 A parathyroid hormone antagonist stimulates epidermal proliferation and hair growth in mice. Proc Natl Acad Sci USA 91(17):8014–8016. 102. Whitfield JF, Isaacs RJ, Jouishomme H, MacLean S, Chakravarthy BR, Morley P, Barisoni D, Regalia E, Armato U 1996 C-terminal fragment of parathyroid hormone-related protein, PTHrP-(107–111), stimulates membrane-associated protein kinase C activity and modulates the proliferation of human and murine skin keratinocytes. J Cell Physiol 166(1):1–11. 103. Luger TA, Stadler BM, Luger BM, Sztein MB, Schmidt JA, Hawley-Nelson P, Grabner G, Oppenheim JJ 1983 Characteristics of an epidermal cell thymocyte-activating factor (ETAF) produced by human epidermal cells and a human squamous cell carcinoma cell line. J Invest Dermatol 81(3):187–193. 104. Kupper TS, Ballard DW, Chua AO, McGuire JS, Flood PM, Horowitz MC, Langdon R, Lightfoot L, Gubler U 1986 Human keratinocytes contain mRNA indistinguishable from monocyte interleukin 1 alpha and beta mRNA. Keratinocyte epidermal cell-derived thymocyte-activating factor is identical to interleukin 1. J Exp Med 164(6):2095–2100.

626 105. Luger TA, Kock A, Kirnbauer R, Schwarz T, Ansel JC 1988 Keratinocyte-derived interleukin 3. Ann NY Acad Sci 548: 253–261. 106. Grossman RM, Krueger J, Yourish D, Granelli-Piperno A, Murphy DP, May LT, Kupper TS, Sehgal PB, Gottlieb AB 1989 Interleukin 6 is expressed in high levels in psoriatic skin and stimulates proliferation of cultured human keratinocytes. Proc Natl Acad Sci USA 86(16):6367–6371. 107. Larsen CG, Anderson AO, Oppenheim JJ, Matsushima K 1989 Production of interleukin-8 by human dermal fibroblasts and keratinocytes in response to interleukin-1 or tumour necrosis factor. Immunology 68(1):31–36. 108. Kupper TS, Lee F, Coleman D, Chodakewitz J, Flood P, Horowitz M 1988 Keratinocyte derived T-cell growth factor (KTGF) is identical to granulocyte macrophage colony stimulating factor (GM-CSF). J Invest Dermatol 91(2):185–188. 109. Chodakewitz JA, Lacy J, Edwards SE, Birchall N, Coleman DL 1990 Macrophage colony-stimulating factor production by murine and human keratinocytes. Enhancement by bacterial lipopolysaccharide. J Immunol 144(6):2190–2196. 110. Kupper TS, Horowitz M, Birchall N, Mizutani H, Coleman D, McGuire J, Flood P, Dower S, Lee F 1988 Hematopoietic, lymphopoietic, and proinflammatory cytokines produced by human and murine keratinocytes. Ann N Y Acad Sci 548: 262–270. 111. Hennings H, Holbrook KA 1983 Calcium regulation of cell– cell contact and differentiation of epidermal cells in culture. An ultrastructural study. Exp Cell Res 143:127–142. 112. Zamansky GB, Nguyen U, Chou IN 1991 An immunofluorescence study of the calcium-induced coordinated reorganization of microfilaments, keratin intermediate filaments, and microtubules in cultured human epidermal keratinocytes. J Invest Dermatol 97(6):985–994. 113. Inohara S, Tatsumi Y, Cho H, Tanaka Y, Sagami S 1990 Actin filament and desmosome formation in cultured human keratinocytes. Arch Dermatol Res 282(3):210–212. 114. Yoneda K, Fujimoto T, Imanura S, Ogawa K 1990 Fodrin is localized in the cytoplasm of keratinocytes cultured in low calcium medium: Immunoelectron microscopic study. Acta Histochem Cytochem 23:139–148. 115. Su MJ, Bikle DD, Mancianti ML, Pillai S 1994 1,25Dihydroxyvitamin D3 potentiates the keratinocyte response to calcium. J Biol Chem 269:14723–14729. 116. Yuspa SH, Kilkenny AE, Steinert PM, Roop DR 1989 Expression of murine epidermal differentiation markers is tightly regulated by restricted extracellular calcium concentrations in vitro. J Cell Biol 109:1207–1217. 117. Ng DC, Su MJ, Kim R, Bikle DD 1996 Regulation of involucrin gene expression by calcium in normal human keratinocytes. Front Biosci 1:a16-a24. 118. Huff CA, Yuspa SH, Rosenthal D 1993 Identification of control elements 3′ to the human keratin 1 gene that regulate cell type and differentiation-specific expression. J Biol Chem 268(1):377–384. 119. Ryynanen J, Jaakkola S, Engvall E, Peltonen J, Uitto J 1991 Expression of beta 4 integrins in human skin: comparison of epidermal distribution with beta 1-integrin epitopes, and modulation by calcium and vitamin D3 in cultured keratinocytes. J Invest Dermatol 97(3):562–567. 120. Glick AB, Sporn MB, Yuspa SH 1991 Altered regulation of TGF-beta 1 and TGF-alpha in primary keratinocytes and papillomas expressing v-Ha-ras. Mol Carcinog 4(3):210–219. 121. Hennings H, Kruszewski FH, Yuspa SH, Tucker RW 1989 Intracellular calcium alterations in response to increased

DANIEL D. BIKLE

122.

123.

124. 125. 126. 127. 128. 129.

130. 131.

132. 133.

134.

135.

136.

137.

138. 139.

external calcium in normal and neoplastic keratinocytes. Carcinogenesis 10:777–780. Kruszewski FH, Hennings H, Tucker RW, Yuspa SH 1991 Differences in the regulation of intracellular calcium in normal and neoplastic keratinocytes are not caused by ras gene mutations. Cancer Res 51:4206–4212. Bikle DD, Ratnam A, Mauro T, Harris J, Pillai S 1996 Changes in calcium responsiveness and handling during keratinocyte differentiation. Potential role of the calcium receptor. J Clin Invest 97:1085–1093. Pillai S, Bikle DD 1992 Lanthanum influx into cultured human keratinocytes: effect on calcium flux and terminal differentiation. J Cell Physiol 151(3):623–629. Reiss M, Lipsey LR, Zhou ZL 1991 Extracellular calciumdependent regulation of transmembrane calcium fluxes in murine keratinocytes. J Cell Physiol 147:281–291. Galietta LJ, Barone V, De Luca M, Romeo G 1991 Characterization of chloride and cation channels in cultured human keratinocytes. Pflügers Arch 418:18–25. Mauro TM, Pappone PA, Isseroff RR 1990 Extracellular calcium affects the membrane currents of cultured human keratinocytes. J Cell Physiol 143:13–20. Mauro TM, Isseroff RR, Lasarow R, Pappone PA 1993 Ion channels are linked to differentiation in keratinocytes. J Membr Biol 132:201–209. Grando SA, Horton RM, Mauro TM, Kist DA, Lee TX, Dahl MV 1996 Activation of keratinocyte nicotinic cholinergic receptors stimulates calcium influx and enhances cell differentiation. J Invest Dermatol 107:412–418. Oda Y, Timpe LC, McKenzie RC, Sauder DN, Largman C, Mauro T 1997 Alternatively spliced forms of the cGMP-gated channel in human keratinocytes. FEBS Lett 414:140–145. Pillai S, Bikle DD, Mancianti ML, Hincenbergs M 1991 Uncoupling of the calcium-sensing mechanism and differentiation in squamous carcinoma cell lines. Exp Cell Res 192:567–573. Holladay K, Fujiki H, Bowden GT 1992 Okadaic acid induces the expression of both early and secondary response genes in mouse keratinocytes. Mol Carcinog 5(1):16–24. Bollag WB, Xiong Y, Ducote J, Harmon CS 1994 Regulation of fos-lacZ fusion gene expression in primary mouse epidermal keratinocytes isolated from transgenic mice. Biochem J 300(Pt 1):263–270. Pillai S, Bikle DD 1992 Adenosine triphosphate stimulates phosphoinositide metabolism, mobilizes intracellular calcium, and inhibits terminal differentiation of human epidermal keratinocytes. J Clin Invest 90:42–51. Nemeth EF, Scarpa A 1987 Rapid mobilization of cellular Ca2+ in bovine parathyroid cells evoked by extracellular divalent cations. Evidence for a cell surface calcium receptor. J Biol Chem 262(11):5188–5196. Brown EM, Gamba G, Riccardi D, Lombardi M, Butters R, Kifor O, Sun A, Hediger MA, Lytton J, Hebert SC 1993 Cloning and characterization of an extracellular Ca(2+)-sensing receptor from bovine parathyroid. Nature 366(6455):575–580. Garrett JE, Capuano IV, Hammerland LG, Hung BC, Brown EM, Hebert SC, Nemeth EF, Fuller F 1995 Molecular cloning and functional expression of human parathyroid calcium receptor cDNAs. J Biol Chem 270:12919–12925. Oda Y, Tu CL, Pillai S, Bikle DD 1998 The calcium sensing receptor and its alternatively spliced form in keratinocyte differentiation. J Biol Chem 273:23344–23352. Oda Y, Tu CL, Chang W, Crumrine D, Kömüves L, Mauro T, Elias PM, Bikle DD 2000 The calcium sensing receptor and

CHAPTER 35 Vitamin D: Role in Skin and Hair

140. 141. 142.

143.

144.

145.

146.

147. 148.

149. 150. 151.

152.

153.

154. 155.

156.

its alternatively spliced form in murine epidermal differentiation. J Biol Chem 275:1183–1190. Tu CL, Chang W, Bikle DD 2001 The extracellular calciumsensing receptor is required for calcium-induced differentiation in human keratinocytes. J Biol Chem 276:41079–41085. Ratnam AV, Cho JK, Bikle DD 1996 1,25-Dihydroxyvitamin D3 enhances the calcium response of keratinocytes. J Invest Dermatol 106:910. Jaken S, Yuspa SH 1988 Early signals for keratinocyte differentiation: role of Ca2+-mediated inositol lipid metabolism in normal and neoplastic epidermal cells. Carcinogenesis 9:1033–1038. Tang W, Ziboh VA, Isseroff R, Martinez D 1988 Turnover of inositol phospholipids in cultured murine keratinocytes: possible involvement of inositol triphosphate in cellular differentiation. J Invest Dermatol 90:37–43. Moscat J, Fleming TP, Molloy CJ, Lopez-Barahona M, Aaronson SA 1989 The calcium signal for Balb/MK keratinocyte terminal differentiation induces sustained alterations in phosphoinositide metabolism without detectable protein kinase C activation. J Biol Chem 264(19): 11228–11235. Lee E, Yuspa SH 1991 Aluminum fluoride stimulates inositol phosphate metabolism and inhibits expression of differentiation markers in mouse keratinocytes. J Cell Physiol 148: 106–115. Pillai S, Bikle DD, Su MJ, Ratnam A, Abe J 1995 1,25Dihydroxyvitamin D3 upregulates the phosphatidylinositol signaling pathway in human keratinocytes by increasing phospholipase C levels. J Clin Invest 96:602–609. Xie Z, Bikle DD 1999 Phospholipase C-gamma1 is required for calcium-induced keratinocyte differentiation. J Biol Chem 274:20421–20424. Punnonen K, Denning M, Lee E, Li L, Rhee SG, Yuspa SH 1993 Keratinocyte differentiation is associated with changes in the expression and regulation of phospholipase C isoenzymes. J Invest Dermatol 101:719–726. Carpenter G, Ji Q 1999 Phospholipase C-gamma as a signaltransducing element. Exp Cell Res 253:15–24. Calautti E, Missero C, Stein PL, Ezzell RM, Dotto GP 1995 Fyn tyrosine kinase is involved in keratinocyte differentiation control. Genes Dev 9:2279–2291. Zhao Y, Sudol M, Hanafusa H, Krueger J 1992 Increased tyrosine kinase activity of c-Src during calcium-induced keratinocyte differentiation. Proc Natl Acad Sci USA 89: 8298–8302. Kim HK, Kim JW, Zilberstein A, Margolis B, Kim JG, Schlessinger J, Rhee SG 1991 PDGF stimulation of inositol phospholipid hydrolysis requires PLC-gamma 1 phosphorylation on tyrosine residues 783 and 1254. Cell 65:435–441. Bony C, Roche S, Shuichi U, Sasaki T, Crackower MA, Penninger J, Mano H, Puceat M 2001 A specific role of phosphatidylinositol 3-kinase gamma. A regulation of autonomic Ca2+ oscillations in cardiac cells. J Cell Biol 152:717–728. Marrero MB, Venema RC, Ma H, Ling BN, Eaton DC 1998 Erythropoietin receptor-operated Ca2+ channels: Activation by phospholipase C-gamma 1. Kidney Int 53:1259–1268. Barker SA, Lujan D, Wilson BS 1999 Multiple roles for PI 3-kinase in the regulation of PLCgamma activity and Ca2+ mobilization in antigen-stimulated mast cells. J Leukoc Biol 65(3):321–329. Fry MJ 2001 Phosphoinositide 3-kinase signalling in breast cancer: How big a role might it play? Breast Cancer Res 3:304–312.

627 157. Falasca M, Logan SK, Lehto VP, Baccante G, Lemmon MA, Schlessinger J 1998 Activation of phospholipase C gamma by PI 3-kinase-induced PH domain- mediated membrane targeting. EMBO J 17:414–422. 158. Rameh LE, Rhee SG, Spokes K, Kazlauskas A, Cantley LC, Cantley LG 1998 Phosphoinositide 3-kinase regulates phospholipase Cgamma-mediated calcium signaling. J Biol Chem 273:23750–23757. 159. Bae YS, Cantley LG, Chen CS, Kim SR, Kwon KS, Rhee SG 1998 Activation of phospholipase C-gamma by phosphatidylinositol 3,4,5-trisphosphate. J Biol Chem 273:4465–4469. 160. Hawley-Nelson P, Stanley JR, Schmidt J, Gullino M, Yuspa SH 1982 The tumor promoter 12-O-tetradecanoylphorbol13-acetate accelerates keratinocyte differentiation and stimulates growth of an unidentified cell type in cultured human epidermis. Exp Cell Res 137(1):155–167. 161. Yuspa SH, Ben T, Hennings H, Lichti U 1982 Divergent responses in epidermal basal cells exposed to the tumor promoter 12-O-tetradecanoylphorbol-13-acetate. Cancer Res 42:2344–2349. 162. Kitajima Y, Inoue S, Nagao S, Nagata K, Yaoita H, Nozawa Y 1988 Biphasic effects of 12-O-tetradecanoylphorbol-13-acetate on the cell morphology of low calcium-grown human epidermal carcinoma cells: involvement of translocation and down regulation of protein kinase C. Cancer Res 48(4):964–970. 163. Rice RH, Rong XH, Chakravarty R 1988 Suppression of keratinocyte differentiation in SSC-9 human squamous carcinoma cells by benzo[a]pyrene, 12-O-tetradecanoylphorbol-13-acetate and hydroxyurea. Carcinogenesis 9(10):1885–1890. 164. Yuspa SH, Ben T, Hennings H 1983 The induction of epidermal transglutaminase and terminal differentiation by tumor promoters in cultured epidermal cells. Carcinogenesis 4(11): 1413–1418. 165. Dlugosz AA, Yuspa SH 1994 Protein kinase C regulates keratinocyte transglutaminase (TG-K) gene expression in cultured primary mouse epidermal keratinocytes induced to terminally differentiate by calcium. J Invest Dermatol 102: 409–414. 166. Matsui MS, Illarda I, Wang N, DeLeo VA 1993 Protein kinase C agonist and antagonist effects in normal human epidermal keratinocytes. Exp Dermatol 2(6):247–256. 167. Dlugosz AA, Yuspa SH 1993 Coordinate changes in gene expression which mark the spinous to granular cell transition in epidermis are regulated by protein kinase C. J Cell Biol 120(1):217–225. 168. Ng DC, Shafaee S, Lee D, Bikle DD 2000 Requirement of an AP-1 site in the calcium response region of the involucrin promoter. J Biol Chem 275:24080–24088. 169. Filvaroff E, Stern DF, Dotto GP 1990 Tyrosine phosphorylation is an early and specific event involved in primary keratinocyte differentiation. Mol Cell Biol 10(3):1164–1173. 170. Chakravarthy BR, Isaacs RJ, Morley P, Durkin JP, Whitfield JF 1995 Stimulation of protein kinase C during Ca2+-induced keratinocyte differentiation. Selective blockade of MARCKS phosphorylation by calmodulin. J Biol Chem 270(3): 1362–1368. 171. Hepler JR, Earp HS, Harden TK 1988 Long-term phorbol ester treatment down-regulates protein kinase C and sensitizes the phosphoinositide signaling pathway to hormone and growth factor stimulation. Evidence for a role of protein kinase C in agonist-induced desensitization. J Biol Chem 263(16):7610–7619. 172. Kojima I, Shibata H, Ogata E 1986 Phorbol ester inhibits angiotensin-induced activation of phospholipase C in adrenal

628

173.

174.

175.

176.

177.

178.

179.

180.

181.

182.

183.

184. 185.

186.

DANIEL D. BIKLE

glomerulosa cells. Its implication in the sustained action of angiotensin. Biochem J 237(1):253–258. Dlugosz AA, Mischak H, Mushinski JF, Yuspa SH 1992 Transcripts encoding protein kinase C-alpha, -delta, -epsilon, -zeta, and -eta are expressed in basal and differentiating mouse keratinocytes in vitro and exhibit quantitative changes in neoplastic cells. Mol Carcinogenesis 5(4):286–292. Denning MF, Dlugosz AA, Howett MK, Yuspa SH 1993 Expression of an oncogenic rasHa gene in murine keratinocytes induces tyrosine phosphorylation and reduced activity of protein kinase C delta. J Biol Chem 268(35):26079–26081. Reynolds NJ, Baldassare JJ, Henderson PA, Shuler JL, Ballas LM, Burns DJ, Moomaw CR, Fisher GJ 1994 Translocation and downregulation of protein kinase C isoenzymes-alpha and -epsilon by phorbol ester and bryostatin-1 in human keratinocytes and fibroblasts. J Invest Dermatol 103(3):364–369. Fisher GJ, Tavakkol A, Leach K, Burns D, Basta P, Loomis C, Griffiths CE, Cooper KD, Reynolds NJ, Elder JT and others 1993 Differential expression of protein kinase C isoenzymes in normal and psoriatic adult human skin: reduced expression of protein kinase C-beta II in psoriasis. J Invest Dermatol 101(4):553–559. Geiges D, Marks F, Gschwendt M 1995 Loss of protein kinase C delta from human HaCaT keratinocytes upon ras transfection is mediated by TGF alpha. Exp Cell Res 219(1): 299–303. Goodnight J, Mischak H, Mushinski JF 1994 Selective involvement of protein kinase C isozymes in differentiation and neoplastic transformation. Adv Cancer Res 64(1): 159–209. Denning MF, Dlugosz AA, Williams EK, Szallasi Z, Blumberg PM, Yuspa SH 1995 Specific protein kinase C isozymes mediate the induction of keratinocyte differentiation markers by calcium. Cell Growth Differ 6:149–157. Yang LC, Ng DC, Bikle DD 2003 Role of protein kinase C alpha in calcium induced keratinocyte differentiation: defective regulation in squamous cell carcinoma. J Cell Physiol 195(2):249–259. Szallasi Z, Kosa K, Smith CB, Dlugosz AA, Williams EK, Yuspa SH, Blumberg PM 1995 Differential regulation by anti-tumor-promoting 12-deoxyphorbol-13-phenylacetate reveals distinct roles of the classical and novel protein kinase C isozymes in biological responses of primary mouse keratinocytes. Mol Pharmacol 47(2):258–265. Deucher A, Efimova T, Eckert RL 2002 Calciumdependent involucrin expression is inversely regulated by protein kinase C (PKC)alpha and PKCdelta. J Biol Chem 277(19): 17032–17040. Downward J, Waterfield MD, Parker PJ 1985 Autophosphorylation and protein kinase C phosphorylation of the epidermal growth factor receptor. Effect on tyrosine kinase activity and ligand binding affinity. J Biol Chem 260(27): 14538–14546. Schlessinger J 1986 Allosteric regulation of the epidermal growth factor receptor kinase. J Cell Biol 103(6 Pt 1): 2067–2072. Kolch W, Heidecker G, Kochs G, Hummel R, Vahidi H, Mischak H, Finkenzeller G, Marmé D, Rapp UR 1993 Protein kinase C alpha activates RAF-1 by direct phosphorylation, Nature 364(6434):249–252. Goode N, Hughes K, Woodgett JR, Parker PJ 1992 Differential regulation of glycogen synthase kinase-3 beta by protein kinase C isotypes. J Biol Chem 267(24):16878–16882.

187. Yaar M, Gilani A, DiBenedetto PJ, Harkness DD, Gilchrest BA 1993 Gene modulation accompanying differentiation of normal versus malignant keratinocytes. Exp Cell Res 206(2): 235–243. 188. Stanwell C, Denning MF, Rutberg SE, Cheng C, Yuspa SH, Dlugosz AA 1996 Staurosporine induces a sequential program of mouse keratinocyte terminal differentiation through activation of PKC isozymes. J Invest Dermatol 106(3): 482–489. 189. Welter JF, Eckert RL 1995 Differential expression of the fos and jun family members c-fos, fosB, Fra-1, Fra-2, c-jun, junB and junD during human epidermal keratinocyte differentiation. Oncogene 11(12):2681–2687. 190. Welter JF, Crish JF, Agarwal C, Eckert RL 1995 Fos-related antigen (Fra-1), junB, and junD activate human involucrin promoter transcription by binding to proximal and distal AP1 sites to mediate phorbol ester effects on promoter activity. J Biol Chem 270(21):12614–12622. 191. Bikle DD, Ng D, Oda Y, Hanley K, Feingold K, Xie Z 2002 The vitamin D response element of the involucrin gene mediates its regulation by 1,25-dihydroxyvitamin D3. J Invest Dermatol 119(5):1109–1113. 192. Fairley JA, Marcelo CL, Hogan VA, Voorhees JJ 1985 Increased calmodulin levels in psoriasis and low Ca++ regulated mouse epidermal keratinocyte cultures. J Invest Dermatol 84(3):195–198. 193. Al-Ani AM, Messenger AG, Lawry J, Bleehen SS, MacNeil S 1988 Calcium/calmodulin regulation of the proliferation of human epidermal keratinocytes, dermal fibroblasts and mouse B16 melanoma cells in culture. Br J Dermatol 119(3): 295–306. 194. Eichelberg D, Fuchs A 1988 Calmodulin-antagonism inhibits human keratinocyte proliferation. Arch Dermatol Res 280(5):323–324. 195. Grief F, Soroff HS, Albers KM, Taichman LB 1989 The effect of trifluoperazine, a calmodulin antagonist, on the growth of normal and malignant epidermal keratinocytes in culture. Eur J Cancer Clin Oncol 25(1):19–26. 196. Sheng M, Thompson MA, Greenberg ME 1991 CREB: a Ca2+-regulated transcription factor phosphorylated by calmodulin-dependent kinases. Science 252(5011):1427–1430. 197. Sheu HM, Kitajima Y, Yaoita H 1989 Involvement of protein kinase C in translocation of desmoplakins from cytosol to plasma membrane during desmosome formation in human squamous cell carcinoma cells grown in low to normal calcium concentration. Exp Cell Res 185:176–190. 198. Horiuchi N, Clemens TL, Schiller AL, Holick MF 1985 Detection and developmental changes of the 1,25-(OH)2-D3 receptor concentration in mouse skin and intestine. J Invest Dermatol 84(6):461–464. 199. Merke J, Schwittay D, Fürstenberger G, Gross M, Marks F, Ritz E 1985 Demonstration and characterization of 1,25-dihydroxyvitamin D3 receptors in basal cells of epidermis of neonatal and adult mice. Calc Tissue Int 37(3): 257–267. 200. McLane JA, Katz M, Abdelkader N 1990 Effect of 1,25dihydroxyvitamin D3 on human keratinocytes grown under different culture conditions. In Vitro Cell Dev Biol 26: 379–387. 201. Bollag WB, Ducote J, Harmon CS 1995 Biphasic effect of 1,25-dihydroxyvitamin D3 on primary mouse epidermal keratinocyte proliferation. J Cell Physiol 163(2):248–256. 202. Itin PH, Pittelkow MR, Kumar R 1994 Effects of vitamin D metabolites on proliferation and differentiation of cultured

CHAPTER 35 Vitamin D: Role in Skin and Hair

203.

204.

205.

206.

207.

208.

209.

210.

211.

212.

213. 214.

215. 216. 217.

human epidermal keratinocytes grown in serum-free or defined culture medium. Endocrinology 135(5):1793–1798. Gniadecki R 1996 Stimulation versus inhibition of keratinocyte growth by 1,25-dihydroxyvitamin D3: dependence on cell culture conditions. J Invest Dermatol 106(3): 510–516. Matsumoto K, Hashimoto K, Nishida Y, Hashiro M, Yoshikawa K 1990 Growth-inhibitory effects of 1,25dihydroxyvitamin D3 on normal human keratinocytes cultured in serum-free medium. Biochem Biophys Res Commun 166(2):916–923. Rizk-Rabin M, Pavlovitch JH 1988 Effect of vitamin D deficiency and 1,25-dihydroxycholecalciferol treatment on epidermal calcium-binding protein (ECaBP) RNA activity. Mol Cell Endocrinol 60(2–3):145–149. Kremer R, Karaplis AC, Henderson J, Gulliver W, Banville D, Hendy GN, Goltzman D 1991 Regulation of parathyroid hormone-like peptide in cultured normal human keratinocytes. Effect of growth factors and 1,25 dihydroxyvitamin D3 on gene expression and secretion. J Clin Invest 87(3):884–893. Kim HJ, Abdelkader N, Katz M, McLane JA 1992 1,25Dihydroxy-vitamin-D3 enhances antiproliferative effect and transcription of TGF-beta1 on human keratinocytes in culture. J Cell Physiol 151(3):579–587. Iwasha M, Hashimoto K, Hashiro M, Yoshikawa K 1991 Production of plasminogen activator inhibitor-1 (PA-1) by human keratinocytes is induced by 1,25-dihydroxyvitamin D3. J Invest Dermatol 96:558. Tang W, Ziboh VA, Isseroff RR, Martinez D 1987 Novel regulatory actions of 1α,25-dihydroxyvitamin D3 on the metabolism of polyphosphoinositides in murine epidermal keratinocytes. J Cell Physiol 132(1):131–136. McLaughlin JA, Cantley LC, Holick MF 1990 1,25(OH)2D3 increased calcium and phosphatidylinositol metabolism in differentiating cultured human keratinocytes. J Nutr Biochem 1:81–87. Yada Y, Ozeki T, Meguro S, Mori S, Nozawa Y 1989 Signal transduction in the onset of terminal keratinocyte differentiation induced by 1α,25-dihydroxyvitamin D3: Role of protein kinase C translocation. Biochem Biophys Res Commun 163(3):1517–1522. Tang W, Ziboh VA 1991 Agonist/inositol trisphosphateinduced release of calcium from murine keratinocytes: a possible link with keratinocyte differentiation. J Invest Dermatol 96(1):134–138. Bittiner B, Bleehen SS, MacNeil S 1991 1α,25(OH)2 vitamin D3 increases intracellular calcium in human keratinocytes. Br J Dermatol 124(3):230–235. Xie Z, Komuves L, Yu QC, Elalieh H, Ng DC, Leary C, Chang S, Crumrine D, Yoshizawa T, Kato S and others 2002 Lack of the vitamin D receptor is associated with reduced epidermal differentiation and hair follicle growth. J Invest Dermatol 118(1):11–16. Hochberg Z, Gilhar A, Haim S, Friedman-Birnbaum R, Levy J, Benderly A 1985 Calcitriol-resistant rickets with alopecia. Arch Dermatol 121(5):646–647. Marx SJ, Bliziotes MM, Nanes M 1986 Analysis of the relation between alopecia and resistance to 1,25-dihydroxyvitamin D. Clin Endocrinol (Oxf) 25(4):373–381. Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997 Targeted ablation of the vitamin D receptor: an animal model of vitamin D-dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94(18):9831–9835.

629 218. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y and others 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Genet 16(4):391–396. 219. Sakai Y, Demay MB 2000 Evaluation of keratinocyte proliferation and differentiation in vitamin D receptor knockout mice. Endocrinology 141(6):2043–2049. 220. Kong J, Li XJ, Gavin D, Jiang Y, Li YC 2002 Targeted expression of human vitamin D receptor in the skin promotes the initiation of the postnatal hair follicle cycle and rescues the alopecia in vitamin D receptor null mice. J Invest Dermatol 118(4):631–638. 221. Chen CH, Sakai Y, Demay MB 2001 Targeting expression of the human vitamin D receptor to the keratinocytes of vitamin D receptor null mice prevents alopecia. Endocrinology 142(12): 5386–5389. 222. Li YC, Amling M, Pirro AE, Priemel M, Meuse J, Baron R, Delling G, Demay MB 1998 Normalization of mineral ion homeostasis by dietary means prevents hyperparathyroidism, rickets, and osteomalacia, but not alopecia in vitamin D receptor-ablated mice. Endocrinology 139(10): 4391–4396. 223. Millar SE 2002 Molecular mechanisms regulating hair follicle development. J Invest Dermatol 118:216–225. 224. Stenn KS, Paus R 2001 Controls of hair follicle cycling. Physiol Rev 81(1):449–494. 225. Soler AP, Gilliard G, Megosh LC, O’Brien TG 1996 Modulation of murine hair follicle function by alterations in ornithine decarboxylase activity. J Invest Dermatol 106(5): 1108–1113. 226. Pietila M, Parkkinen JJ, Alhonen L, Janne J 2001 Relation of skin polyamines to the hairless phenotype in transgenic mice overexpressing spermidine/spermine N-acetyltransferase. J Invest Dermatol 116(5):801–805. 227. DasGupta R, Rhee H, Fuchs E 2002 A developmental conundrum: a stabilized form of beta-catenin lacking the transcriptional activation domain triggers features of hair cell fate in epidermal cells and epidermal cell fate in hair follicle cells. J Cell Biol 158(2):331–344. 228. Huelsken J, Vogel R, Erdmann B, Cotsarelis G, Birchmeier W 2001 β-Catenin controls hair follicle morphogenesis and stem cell differentiation in the skin. Cell 105(4):533–545. 229. Panteleyev AA, Botchkareva NV, Sundberg JP, Christiano AM, Paus R 1999 The role of the hairless (hr) gene in the regulation of hair follicle catagen transformation. Am J Pathol 155(1): 159–171. 230. Ahmad W, Faiyaz ul Haque M, Brancolini V, Tsou HC, ul Haque S, Lam H, Aita VM, Owen J, deBlaquiere M, Frank J and others 1998 Alopecia universalis associated with a mutation in the human hairless gene. Science 279(5351): 720–724. 231. Miller J, Djabali K, Chen T, Liu Y, Ioffreda M, Lyle S, Christiano AM, Holick M, Cotsarelis G 2001 Atrichia caused by mutations in the vitamin D receptor gene is a phenocopy of generalized atrichia caused by mutations in the hairless gene. J Invest Dermatol 117(3):612–617. 232. Sakai Y, Kishimoto J, Demay MB 2001 Metabolic and cellular analysis of alopecia in vitamin D receptor knockout mice. J Clin Invest 107(8):961–966. 233. Sone T, Kerner S, Pike JW 1991 Vitamin D receptor interaction with specific DNA. Association as a 1,25-dihydroxyvitamin D3-modulated heterodimer. J Biol Chem 266(34): 23296–23305.

630 234. Prufer K, Barsony J 2002 Retinoid X receptor dominates the nuclear import and export of the unliganded vitamin D receptor. Mol Endocrinol 16(8):1738–1751. 235. Panteleyev AA, Paus R, Christiano AM 2000 Patterns of hairless (hr) gene expression in mouse hair follicle morphogenesis and cycling. Am J Pathol 157(4): 1071–1079. 236. Djabali K, Aita VM, Christiano AM 2001 Hairless is translocated to the nucleus via a novel bipartite nuclear localization signal and is associated with the nuclear matrix. J Cell Sci 114(Pt 2):367–376.

DANIEL D. BIKLE

237. Thompson CC, Bottcher MC 1997 The product of a thyroid hormone-responsive gene interacts with thyroid hormone receptors. Proc Natl Acad Sci USA 94(16):8527–8532. 238. Thompson CC 1996 Thyroid hormone-responsive genes in developing cerebellum include a novel synaptotagmin and a hairless homolog. J Neurosci 16(24):7832–7840. 239. Hsieh JC, Sisk JM, Jurutka PW, Haussler CA, Slater SA, Haussler MR, Thompson CC 2003 J Biol Chem 278: 38665–38674. 240. Fuchs E, Merrill BJ, Jamora C, DasGupta R 2001 At the roots of a never-ending cycle. Dev Cell 1(1):13–25.

CHAPTER 36

Regulation of Immune Responses by Vitamin D Receptor Ligands LUCIANO ADORINI

BioXell, Milan, Italy

I. Introduction II. Major Target Cells in Immunoregulation by Vitamin D Receptor Ligands: Dendritic Cells and T Cells III. Possible Mechanisms for the Immunomodulatory Effects of VDR Ligands in Autoimmune Disease Models

IV. Beneficial Effects of VDR Ligands in Allograft Rejection V. Conclusions References

I. INTRODUCTION

development via clonal deletion. Although this is a primary mechanism of self tolerance, it does not completely eliminate T cells specific for self antigens. Selfreactive T cells that have been exported to the periphery can then be functionally inactivated upon antigen recognition in the absence of appropriate costimulatory signals, a process denoted as clonal anergy. Finally, peripheral self-reactive T cells can be suppressed by other T cells. In reality, tolerance is truly redundant, and both deletional and non-deletional mechanisms operate in the thymus and in the periphery [3].

The raison d’être of the immune system is to maintain the biological integrity of the individual. This is accomplished by two integrated levels of immune responses, innate and adaptive. Innate immune responses can be induced in virtually any cell, but they are primarily mediated by specialized cell types, such as macrophages and natural killer cells. Innate immunity is characterized by rapid responses, largely based on the production of proinflammatory mediators, in particular cytokines, chemokines, and reactive oxygen species. This is triggered by recognition of stereotyped patterns conserved in infectious microorganisms via toll-like receptors (TLRs), cell surface molecules able to recognize distinct structural components of pathogens [1]. Activation of signal transduction pathways by TLRs leads to up-regulation of different genes that operate in host defence, including co-stimulatory molecules, cytokines, and chemokines [2]. Shortly afterward, adaptive immunity can be induced. Adaptive immune responses are mediated by cells specialized in antigen presentation, in particular dendritic cells (DCs), and by cells responsible antigen recognition, carried out by T and B lymphocytes. Adaptive immune responses are primarily orchestrated by CD4+ T lymphocytes. To select lymphocytes able to respond to foreign molecules while remaining tolerant to self components, the strategy of the immune system has been to generate a vast repertoire of antigen-specific receptors, distribute it clonally in different lymphocytes, and then eliminate cells capable of recognizing with high affinity self components while permitting the differentiation of T cells potentially able to recognize foreign antigens. However complex as to the mechanisms utilized, the basis for tolerance to self components is relatively simple. First, T cells expressing high-affinity receptors for self antigens can be physically eliminated during thymic VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

A. Selective Immunointervention Failure of tolerance mechanisms may lead to autoimmune diseases and to other immune-mediated pathologies. The progress in understanding the mechanisms of T cell activation and inactivation is currently being translated into strategies able to induce selective immunosuppression to treat different pathological situations, notably autoimmune diseases, as well as allergies and allograft rejection. The medical need for selective immunosuppression is very high, as the available immunosuppressive drugs are substantially inadequate because of limited efficacy, modest selectivity, and considerable toxicity [4]. Key attack points for selective immunointervention have been identified: modulation of antigen recognition, costimulation blockade, induction of regulatory cells, deviation to nonpathogenic or protective responses, neutralization of proinflammatory cytokines, induction or administration of anti-inflammatory cytokines, and modulation of leukocyte trafficking (Table I). Thus, to selectively interfere with the activation of pathogenic T cells, immunosuppressive therapy can be primarily directed to three cellular targets: antigen-presenting cells (APCs), autoreactive T cells, and suppressor/regulatory Copyright © 2005, Elsevier, Inc. All rights reserved.

632

LUCIANO ADORINI

TABLE I Key Attack Points for Selective Immunomodulation Targeting the MHC/TCR complex of pathogenic T cells MHC blockade, deletion, altered peptide ligands Co-receptors (CD4,CD8) Costimulation blockade Inhibition of interaction between CD28-CD80/CD86; CD154-CD40, LIGHT-HVEM; ICOS-ICOSL; CD134-CD134L —up-regulation of negative co-regulators (CD152, PD-1) Immune deviation Skewing to Th2 via APC or direct T-cell modulation Cytokine-based immunointervention Inhibition of proinflammatory cytokines (IL-1, IL-2, IFN-γ, IL-12, TNF-α) Administration of anti-inflammatory cytokines (IL-4, IL-10, IFN-β, TGF-β) Induction of regulatory T cells T cell/TCR peptide vaccination, APC manipulation, cytokines (IL-10, TGF-β) Targeting leukocyte trafficking Adhesion molecules, chemokines, chemokine receptors

T cells, with the common goal to selectively inhibit the activation of pathogenic class II-restricted CD4+ T cells [4]. In autoimmune diseases, pathogenic T cells are usually Th1 cells. CD4+ T cells can be distinguished, TABLE II

based on their pattern of cytokine production, into three major types. Th1 cells are characterized by secretion of interferon-γ (IFN-γ), IL-2, and TNF-β, and they promote cell-mediated immunity able to eliminate intracellular pathogens. Th2 cells selectively produce IL-4 and IL-5 and are involved in the development of humoral immunity protecting against extracellular pathogens. Th0 cells, which could represent either precursors or a terminally differentiated subset, are not restricted in their lymphokine production. A similar distinction applies to CD8+ cells. The development of Th1 and Th2 cells is influenced by several factors, but three are most important: local cytokines, the avidity of ligand–TCR interaction, and the non-MHC genetic polymorphism. Decisive roles in the polarization of T cells are played by IL-12 and IL-4, guiding T cell responses toward the Th1 or Th2 phenotype, respectively [3]. Different forms of immunointervention have been successfully used to prevent and sometimes treat experimental autoimmune diseases and allograft rejection (Table I). Several of these approaches target DCs, aiming at inducing or enhancing tolerogenic properties in this APC type critically involved in modulating T cell responses. A variety of agents, both biologic and pharmacologic, have been shown to promote the intrinsic tolerogenic capacity of DCs [5,6]. Biologics include costimulation-blocking agents, such as antiCD40L and CD152-Ig, and anti-inflammatory cytokines such as IL-10 and TGF-β. Pharmacologic agents include immunosuppressive molecules such as mycophenolate

Phenotypic and Functional Modifications Induced by VDR Ligands in Human Myeloid Dendritic Cells

Phenotype Maturation marker expression CD83 DC-LAMP Antigen uptake Mannose receptor expression Costimulatory molecule expression CD40 CD80 CD86 Inhibitory molecule expression ILT3 ILT4 B7-H1 Chemokine receptor expression CCR7

Effect

Decreased Decreased Increased Decreased Decreased Decreased Increased Unmodified Unmodified Decreased

Compiled from [39,160] and from the author’s unpublished data.

Function Cytokine production IL-10 IL-12 Chemokine production CCL2 CCL17 CCL18 CCL20 CCL22 Apoptosis Maturation-induced T-cell activation Response to alloantigens

Effect

Increased Decreased Increased Decreased Increased Decreased Increased Increased Decreased

CHAPTER 36 Regulation of Immune Responses by Vitamin D Receptor Ligands

mofetil, sirolimus, desoxyspergualin, corticosteroids, and 1,25(OH)2D3.

B. VDR Ligands as Immunoregulatory Agents 1,25(OH)2D3, the activated form of vitamin D, is a secosteroid hormone that has, in addition to its central function in calcium and bone metabolism, important effects on the growth and differentiation of many cell types, and pronounced immunoregulatory properties [7–11]. The biological effects of 1,25(OH)2D3 are mediated by the vitamin D receptor (VDR), a member of the superfamily of nuclear hormone receptors (see Chapters 11–18) [12,13]. Ligand binding induces conformational changes in the VDR, which promote heterodimerization with the retinoid X receptor (RXR) and recruitment of a number of co-repressor and coactivator proteins, including steroid receptor coactivator family members and a multimember coactivator complex, D-receptor interacting proteins (DRIPs). These coactivators induce chromatin remodeling through intrinsic histone-modifying activities and direct recruitment of key transcription initiation components at regulated promoters. Thus, the VDR functions as a ligand-activated transcription factor that binds to specific DNA sequence elements (vitamin D–responsive elements, VDREs) in vitamin D–responsive genes and ultimately influences the rate of RNA polymerase II– mediated transcription [14]. The discovery of VDR expression in most cell types of the immune system [15], in particular in APCs such as macrophages [15] and DCs [16], as well as in both CD4+ and CD8+ T lymphocytes (reviewed in ref. [17]), prompted the investigation of VDR ligands as agents able to modulate T cell responses [18]. Data accumulated in the past few years clearly demonstrate that, in addition to exerting direct effects on T cell activation, VDR ligands markedly modulate the phenotype and function of APCs, and in particular of DCs. In vitro and in vivo experiments have shown that VDR ligands induce DCs to acquire tolerogenic properties that favor the induction of regulatory rather than effector T cells. These intriguing actions of VDR ligands have been demonstrated in several experimental models and could be exploited, in principle, to treat a variety of human autoimmune diseases [19] and other immunomediated pathologies [7–9]. In addition, it is conceivable that 1,25(OH)2D3, which is produced by macrophages [20–22], DCs [23], and T cells [21], could physiologically contribute to regulate innate and adaptive immune responses. This appealing concept, although still speculative, is mostly based on epidemiological data [7–9] and is indirectly supported by the observation

633

that VDR-deficient compared to wild-type mice show hypertrophy of subcutaneous lymph nodes with an increase in mature DCs [24]. Clarification of the physiological role of endogenous VDR ligands in the regulation of immune responses will likely represent a future step of development in this fruitful area of research.

II. MAJOR TARGET CELLS IN IMMUNOREGULATION BY VDR LIGANDS: DENDRITIC CELLS AND T CELLS DCs, a highly specialized APC system critical for the initiation of CD4+ T cell responses, are present, in different stages of maturation, in the circulation as well as in lymphoid and nonlymphoid organs [25]. Immature DCs, such as Langerhans cells in the skin, are found in nonlymphoid tissues, where they exert a sentinel function. After antigen uptake, they migrate through the afferent lymph to T-dependent areas of secondary lymphoid organs where priming of naive T cells may occur. During migration to lymphoid organs, DCs mature into potent APCs by increasing their immunostimulatory properties while decreasing antigencapturing capacity [26]. DCs are heterogeneous in terms not only of maturation state, but also of origin, morphology, phenotype, and function [26,27]. Two distinct DC subsets were originally defined in the human blood based on the expression of CD11c, and they have been subsequently characterized as belonging to the myeloid or lymphoid lineage. Although different denominations have been used, they can be defined as myeloid (M-DCs) and plasmacytoid (P-DCs) DCs [28]. A cell population resembling human P-DCs has also been identified in the mouse [29]. M-DCs are characterized by a monocytic morphology and express myeloid markers such as CD13 and CD33, the β2 integrin CD11c, the inhibitory receptor ILT1, and low levels of the IL-3 receptor α chain CD123. Conversely, P-DCs have a morphology resembling plasma cells, are devoid of myeloid markers, and express high levels of CD4, CD62L, and CD123. M-DCs produce high levels of IL-12, whereas P-DCs high levels of IFN-α [28], cytokines with clearly distinct effects on T cell activation and differentiation. Recently, it has become clear that DCs are not only immunogenic but also tolerogenic, both intrathymically and in the periphery [30]. In particular, immature DCs have been found to have tolerogenic properties and to induce T cells with suppressive activity [31,32]. In contrast, the role of DC subsets in directing the

634

LUCIANO ADORINI

Regulatory/ suppressor T cell High coinhibition Low costimulation

Low coinhibition High costimulation DC

Effector T cell

FIGURE 1 Dendritic cells play a key role in the generation of effector and regulatory T cells. DCs expressing high levels of surface costimulatory molecules, e.g., CD40, CD80, CD86, and secreting IL-12, induce effector T cells, notably Th1 cells. Conversely, DCs expressing low levels of costimulatory molecules, secreting IL-10, and expressing high levels of inhibitory molecules (e.g., ILT3) favor the induction and/or the enhancement of regulatory/ suppressor T cells.

development of T cells with a defined functional role is still unclear, although P-DCs are credited with a higher tolerogenic potential [28]. In any case, DCs expressing low levels of costimulatory molecules, either membrane-bound (e.g., CD40, CD80, CD86) or secreted, such as IL-12, and high levels of inhibitory surface molecules such as ILT3 or secreted molecules such as IL-10, favor the induction of suppressor rather than effector T cells (Fig. 1).

A. Regulatory Effects of VDR Ligands in Dendritic Cells Earlier indications for the capacity of VDR ligands to target APCs [33–35] were corroborated by their ability to inhibit the production of IL-12 [36,37], an APC-derived cytokine critical for Th1 cell development [38]. More recent work has demonstrated that 1,25(OH)2D3 and its analogs have profound effects on the phenotype and function of myeloid DCs (Table II). VDR ligands arrest the differentiation and maturation of DCs, maintaining them in an immature state, as shown by decreased expression of maturation markers and increased antigen uptake [39–44]. Collectively, studies performed either on monocyte-derived DCs from human peripheral blood or on bone-marrow

derived mouse DCs have consistently shown that in vitro treatment of DCs with VDR ligands leads to down-regulated expression of the costimulatory molecules CD40, CD80, CD86, and to markedly decreased IL-12 and enhanced IL-10 production, resulting in inhibition of T-cell activation. The near abrogation of IL-12 production and the strongly enhanced production of IL-10 highlight the important functional effects of 1,25(OH)2D3 and its analogs on DCs and are, at least in part, responsible for the induction of DCs with tolerogenic properties. In addition, DCs treated with VDR ligands up-regulate the expression of ILT3, an inhibitory molecule that has been associated with tolerance induction [45]. 1,25(OH)2D3 utilizes different mechanisms to regulate cytokine production by DCs. IL-12 secretion is inhibited by targeting the NF-κB pathway [37], via NF-κB proteins such as Rel-B and c-Rel [9,46]. Interestingly, antigen-exposed DCs in which Rel-B function is inhibited induce a population of antigenspecific CD4+ cells that regulate immune responses in an IL-10-dependent manner [47]. Suppression of the monocyte recruiter GM-CSF is instead achieved by interaction of ligand-bound VDR monomers with functional repressive complexes in the promoter region of the cytokine [48]. In this case, the VDR–ligand complex acts selectively on the two components required for activation of this promoter/enhancer: it competes with NFAT1 for binding to the composite site, positioning itself adjacent to Jun-Fos on the DNA. Co-occupancy apparently leads to an inhibitory effect on c-Jun transactivation function. These two VDRmediated events effectively block the NFAT1-AP-1 activation complex, resulting in an attenuation of GM-CSF transcription [49]. The prevention of DC differentiation and maturation as well as the modulation of their activation and survival, leading to DCs with tolerogenic phenotype and function that result in T cell hyporesponsiveness, certainly play an important role in the immunoregulatory activities of VDR ligands. These effects are not limited to in vitro activity: 1,25(OH)2D3 and its analogs can also induce DCs with tolerogenic properties in vivo, as demonstrated in models of allograft rejection by oral administration directly to the recipient [50] or by adoptive transfer of in vitro–treated DCs [24]. Tolerogenic DCs induced by a short treatment with 1,25(OH)2D3 are probably responsible for the capacity of this hormone to induce CD4+CD25+ regulatory T cells that are able to mediate transplantation tolerance [50] and arrest the development of autoimmune diabetes [51]. Rag-1-dependent regulatory cells have also been implicated in the prevention of EAE induced by 1,25(OH)2D3,

CHAPTER 36 Regulation of Immune Responses by Vitamin D Receptor Ligands

although no effect on APCs could be demonstrated in this study [52].

B. Effects of VDR Ligands in T Cells As reviewed earlier, VDR ligands modulate DC function, thus shaping T cell activation and development, but they can also have direct effects on T cells. Soon after the discovery of VDR expression in T cells [15,53], 1,25(OH)2D3 was shown to inhibit antigen-induced T cell proliferation [18] and cytokine production [54]. Later studies demonstrated selective inhibition of Th1 cell development [36,55], although it was not clarified how much of this effect could be accounted for by modulation of DC functions. Indeed, several key cytokines in T lymphocytes are direct targets for VDR ligands, in particular Th1-type cytokines such as IL-2 and IFN-γ (Table III). 1,25(OH)2D3 inhibits IL-2 secretion by impairing the transcription factor NF-AT complex formation, because the ligand-bound VDR complex binds to the distal NF-AT binding site of the human IL-2 promoter [56,57]. Another key T cell cytokine, IFN-γ, has been found directly inhibited by 1,25(OH)2D3 through interaction of the ligand-bound VDR complex with a VDRE in the promoter region of the cytokine [58]. Progressive deletion analysis of the IFN-γ promoter revealed that negative regulation by 1,25(OH)2D3 is also exerted at the level of an upstream region containing an enhancer element [58]. However, some in vivo studies have failed to support a direct effect of 1,25(OH)2D3 on IFN-γ production by T cells [59]. VDR ligands are known to control the growth and differentiation of many cell types, using a variety of

TABLE III

Effects of VDR Ligands on T Cells

Effect Inhibition of T cell proliferation Induction of hyporesponsiveness to allo and self antigens Inhibition of IL-2 production Inhibition of IFN-γ production Inhibition of Th1 cell development Variable effects on IL-4 production and deviation to Th2 Increased production of IL-10 Increased expression of CD152 Down-regulation of CD95 expression Enhanced frequency of regulatory T cells

References [18] [39–43] [56,57] [58,65] [36,55] [52,55,59,63–65] [66] [39,50] [60] [50,51,66]

635

different mechanisms [7–9]. 1,25(OH)2D3 inhibits in T cells activation-induced cell death by down-regulating the expression of CD95L, a cell surface molecule that activates apoptosis in CD95 (Fas)-expressing target cells, via repression of CD95L promoter activity through an AF-2-dependent mechanism [60]. Down-regulation of CD95L expression may have functional consequences, because CD95L costimulates the in vivo proliferation of CD8+ T cells [61] and the activated CD95 (Fas) induces DC maturation and a preferential T cell polarization toward the Th1 pathway [62]. This could be one of the mechanisms VDR ligands utilize to arrest indirectly DC maturation, although they directly promote [39], rather than inhibiting, apoptosis in DCs. 1,25(OH)2D3 has been also shown to enhance the development of Th2 cells via a direct effect on naïve CD4+ cells [63]; this could account for the beneficial effect of VDR ligands in the treatment of autoimmune diseases and possibly also allograft rejection. The capacity of 1,25(OH)2D3 to skew T cells towards the Th2 pathway had been previously suggested [59,64], but could not be confirmed by other studies [52,55]. A recent study has actually shown that 1,25(OH)2D3 can inhibit both IFN-γ and IL-4 production in T cells [65]. The inhibition of IL-4 production in naïve T cells does not appear to result from a cell cycle block or from inhibition of Th2 transcription factor expression, but rather from a VDR-induced direct down-regulation of IL-4 transcription. It is puzzling to note that 1,25(OH)2D3 can apparently up-regulate [59,63,64], down-regulate [65], or have no effect [52,55] on IL-4 production, and consequently on Th2 cell development. These disparate results may reflect the different conditions tested, but also illustrate the complex immunoregulatory pathways set in motion by 1,25(OH)2D3. In addition, a novel aspect of the multiple effects of VDR ligands on T cells is provided by the induction of cells with suppressive and regulatory properties. In conclusion, 1,25(OH)2D3 in vivo appears primarily to inhibit Th1 cells and, under appropriate conditions, may favor a deviation to the Th2 pathway. These effects could be, in part, a consequence of direct T cell targeting by 1,25(OH)2D3 and its analogs, but modulation of DC function by VDR ligands certainly plays an important role in shaping the development of T cell responses. Thus, VDR ligands can target T cells both directly and indirectly. The capacity of VDR ligands to target DCs and T cells depends on VDR expression by both cell types and on the presence of common targets in their signal transduction pathways, as exemplified by the ability of VDR ligands to downregulate the nuclear factor NF-κB in DCs [9,37] and in T cells [66].

636 C. Enhancement of Regulatory T Cells by VDR Ligands As discussed earlier, induction of DCs with tolerogenic phenotype and function plays an important role in the immunoregulatory activity of VDR ligands. Tolerogenic DCs induced by a short treatment with 1,25(OH)2D3 or its analogs are likely responsible for the capacity of this hormone to induce CD4+CD25+ suppressor T cells (CD25+Ts) that are able to mediate transplantation tolerance [50] and to arrest the development of autoimmune diabetes [51]. Interest in the role of Ts cells has recently resurged and, among the various populations of T cells, naturally occurring thymic and peripheral CD4+ T cells that coexpress CD25 are currently the most actively investigated [67]. Although several surface molecules expressed by CD25+Ts cells have been suggested to provide key molecular signals for immunosuppression, multiple mechanisms are probably operative. Based on the essential role of cell–cell contact for suppressive activity in vitro, the appropriate localization of CD25+Ts cells could be crucial for their function not only in directing their immunosuppressive activity but also in regulating their homeostasis by guiding them to microenvironmental sources of instruction, survival and/or proliferation signals. CCR4, CCR5, and CCR8, a pattern of chemokine receptors selectively expressed by CD25+Ts cells, could guide them to their cellular targets and control their interaction with APCs and T cells [68]. Two DC subsets, myeloid (M-DCs) and plasmacytoid DCs (P-DCs), have been identified. These subsets are characterized by a distinct expression of pathogenassociated pattern recognition receptors and costimulatory molecules, and by the selective production of immunomodulatory cytokines [28]. We have documented that, in contrast to the high production by circulating human M-DCs, the CCR4 ligands CCL17 and CCL22 are poorly produced by P-DCs [69]. It is noteworthy that blood-borne M-DCs, but not P-DCs, constitutively produce CCL17 and CCL22 ex vivo [69]. This selective constitutive production of CCR4 ligands by immature M-DCs can lead to the preferential attraction of CD25+Ts cells, ultimately favoring tolerance induction. Intriguingly, the production of the CCR4 ligand CCL22 by M-DCs is markedly enhanced by in vitro treatment with VDR ligands (Table II). In contrast, immature P-DCs fail to secrete significant amounts of chemokines targeting any of the receptors so far identified on CD25+Ts cells, arguing against a similar function for these cells. Besides maintaining peripheral immunological tolerance in homeostatic conditions, Ts cells could turn off and limit ongoing

LUCIANO ADORINI

inflammatory responses. Inflammatory signals strongly induce maturation and influx of both M-DCs and P-DCs into secondary lymphoid tissues [28], and maturation of M-DCs and P-DCs enhances their production of several proinflammatory chemokines that can potentially attract different T-cell subsets. Maturing P-DCs, similarly to activated B cells, produce large quantities of the CCR5 ligand CCL4 [69]. Thus, by analogy with the proposed role for CCL4 in CD25+Ts-cell attraction by activated B cells, mature P-DCs could recruit CD25+Ts cells to limit inflammatory responses. Because DCs are pleiotropic modulators of T cell activity, pharmacological agents that manipulate DC function to favor the development of Ts cells could be exploited in the treatment of autoimmune diseases and graft rejection [4]. VDR ligands could be ideally suited for this purpose, as shown by their capacity to enhance CD25+Ts cells and promote tolerance induction in transplantation [50] and autoimmune disease [51] models. In both models, treatment with VDR ligands has a profound effect on the migration of effector T cells, preventing their entry into the pancreatic islets [50,51]. It remains to be seen if these agents can also affect the migration of CD25+Ts cells by regulating their chemokine receptor expression or by modulating chemokine production in target tissues such as pancreatic islets. Our preliminary experiments show evidence for the latter possibility. However, tolerogenic DCs may not always be necessarily involved in the generation of Ts cells by VDR ligands. A combination of 1,25(OH)2D3 and dexamethasone has been shown to induce human and mouse naive CD4+ T cells to differentiate in vitro into Ts cells, even in the absence of APCs [66]. These cells produced IL-10, but not IL-5 or IFN-γ, thus distinguishing them from the previously described Tr1 cells [70]. Upon transfer, the IL-10-producing T cells could prevent central nervous system inflammation, indicating their capacity to exert a suppressive function in vivo [66].

III. POSSIBLE MECHANISMS FOR THE IMMUNOMODULATORY EFFECTS OF VITAMIN D RECEPTOR LIGANDS IN AUTOIMMUNE DISEASE MODELS The immunoregulatory properties of VDR ligands have been studied in different models of autoimmune diseases (Table IV). Notably, 1,25(OH)2D3 and its analogs can prevent systemic lupus erythematosus in MRLlpr/lpr mice [71–73], experimental allergic encephalomyelitis (EAE) [55,74,75], collagen-induced arthritis [76,77], Lyme arthritis [77], inflammatory bowel disease [78],

637

CHAPTER 36 Regulation of Immune Responses by Vitamin D Receptor Ligands

TABLE IV

Effects of VDR Ligand Treatment in Animal Models of Autoimmune Diseases

Experimental models Arthritis Autoimmune diabetes Experimental allergic encephalomyelitis Inflammatory bowel disease Psoriasis Systemic lupus erythematosus

Main effects Decreased incidence and severity of collagen-induced or Lyme arthritis, also when given at disease onset Inhibition of insulitis and reduction of diabetes, even when given after islet infiltration Prevention and treatment of disease, inhibition of relapses Significant amelioration of symptoms, block of disease progression Inhibition of leukocyte activation and amelioration of histological and clinical sign of disease in human psoriatic skin grafts transplanted to SCID mice Inhibition of proteinuria, prevention of skin lesions

and autoimmune diabetes in nonobese diabetic (NOD) mice [51,79,80]. 1,25(OH)2D3 analogs are able not only to prevent but also to treat ongoing autoimmune diseases, as demonstrated by their ability to inhibit type 1 diabetes development in adult NOD mice [51] and the recurrence of autoimmune disease after islet transplantation in the NOD mouse [81], or to ameliorate significantly the chronic-relapsing EAE induced in Biozzi mice by spinal cord homogenate [55]. An important property of 1,25(OH)2D3 and its analogs is their capacity to modulate both APCs and T cells. The induction of tolerogenic DCs, which leads to an enhanced number of CD4+CD25+ regulatory T cells [50,51], renders them appealing for clinical use, especially for the prevention and treatment of autoimmune diseases and graft rejection. In addition, additive and even synergistic effects have been observed between VDR ligands and immunosuppressive agents, such as CsA and sirolimus, in autoimmune diabetes and EAE models [82,83]. Distinct regulatory mechanisms may predominate in different autoimmune disease models, although a common pattern, characterized by inibition of Th1 cell development, has been frequently observed.

References [76,77] [79,80,82,161] [52,55,74,103] [78] [131] [71,73]

with footpad and ankle swelling. Supplementation with 1,25(OH)2D3 of an adequate diet fed to mice infected with B. burgdorferi minimized or prevented these symptoms [77]. The same treatment could also prevent collagen-induced arthritis, and when given to mice with early symptoms prevented the progression to severe arthritis, compared with untreated controls [77]. In a separate study, VDR ligands displayed a similar capacity to prevent and to suppress already established collageninduced arthritis without inducing hypercalcemia [76]. VDR expression by human articular chondrocytes in osteoarthritic cartilage has often been found associated with sites where matrix metalloproteinases (MMPs) expression was prevalent, in contrast to their virtual absence in normal age-matched cartilage [85]. Together with in vitro studies [86], the data suggest that 1,25(OH)2D3 contributes to the regulation of MMPs and PGE2 production by human articular chondrocytes in osteoarthritic cartilage. Coupled to the evidence obtained in animal models, these results suggest that VDR ligands may be able to control, at least in part, RA development.

B. Type 1 Diabetes A. Rheumatoid Arthritis Rheumatoid arthritis (RA) is an immune-mediated disease, with a prominent involvement of Th1 cells [84], characterized by articular inflammation and subsequent tissue damage leading to severe disability and increased mortality. Among the different animal models of RA, two have been used to test the effects of VDR ligands on the course of the disease, namely Lyme arthritis and collagen-induced arthritis in the mouse. Infection of mice with Borrelia burgdorferi, the causative agent of human Lyme arthritis, produces acute arthritic lesions

The nonobese diabetic (NOD) mouse, which spontaneously develops type 1 diabetes with a pathogenesis similar to the human disease, represents a useful model for the study of autoimmune diabetes [87]. Several effector mechanisms leading to specific islet β-cell destruction have been identified, including cytotoxic CD8+ lymphocytes and macrophages [88], both of which are regulated by IL-12-dependent T helper 1 (Th1) cells [89]. The activation of Th1 cells specific for β-cell autoantigens could reflect defective elimination of autoreactive T-cell clones [90], inefficient mechanisms

638 of peripheral tolerance [91], enhanced IL-12 production [92], or impaired suppressive mechanisms [93]. Agents such as 1,25(OH)2D3 and its analogs, able to inhibit in vivo IL-12 production and Th1 development [55] and to enhance CD4+CD25+ regulatory T cells [50], may therefore be beneficial in the treatment of type 1 diabetes. 1,25(OH)2D3 itself reduces the incidence of insulitis [94] and prevents type 1 diabetes development [79], but only when administered to NOD mice starting from 3 weeks of age, before the onset of insulitis. 1,25(OH)2D3 was found ineffective in preventing progression of diabetes in NOD mice when given from 8 weeks of age, when NOD mice present a well-established insulitis [95]. However, a combined treatment of 8-week-old NOD mice with the 1,25(OH)2D3 analog MC 1288 and cyclosporine A reduced the incidence of disease, although neither treatment alone was effective [82]. In contrast, we have identified the 1,25(OH)2D3 analog 1,25-dihydroxy16,23Z-diene-26,27-hexafluoro-19-nor vitamin D3 (Ro 26-2198) that is able, as a monotherapy, to treat ongoing type 1 diabetes in the adult NOD mouse, effectively blocking the disease course [51]. This property is likely due, at least in part, to the increased metabolic stability of this analog against the inactivating C-24 and C-26 hydroxylations, and the C-3 epimerization [96], resulting in a 100-fold more potent immunosuppressive activity compared to 1,25(OH)2D3. A short treatment with nonhypercalcemic doses of Ro 26-2198 inhibits IL-12 production and pancreatic infiltration of Th1 cells while increasing the frequency of CD4+CD25+ regulatory T cells in pancreatic lymph nodes, arresting the immunological progression and preventing the clinical onset of type 1 diabetes in the NOD mouse [51]. Protection from type 1 diabetes was found associated with a selective decrease of Th1 cells in the pancreatic lymph nodes and in the pancreas, without a marked deviation to the Th2 phenotype. The frequency of CD4+CD25+ cells in the pancreatic lymph nodes of Ro 26-2198-treated NOD mice was twofold higher compared to untreated 8-week-old and to age-matched vehicle-treated controls. These cells were anergic as demonstrated by their impaired capacity to proliferate and secrete IFN-γ in response to TCR ligation, inhibited the T cell response to the pancreatic autoantigen IA-2, and delayed disease transfer by pathogenic CD4+CD25− cells [51]. Immature DCs have been shown to induce CD4+ cells with regulatory properties [31], and arrest of DCs at the immature stage induced by Ro 26-2198 treatment could account for the enhanced frequency of CD4+CD25+ cells. CD4+CD25+ regulatory T cells appear to play an important role in controlling the progression of type 1 diabetes in NOD mice, because a low level of CD4+CD25+ T cells correlates with

LUCIANO ADORINI

exacerbation and acceleration of the disease [93]. It is likely that this cell population is more relevant than Th2 cells in disease control, although both could contribute to protection. Indeed, 1,25(OH)2D3 can induce regulatory cells with disease-suppressive activity in the NOD mouse [79] and a disease-preventing 1,25(OH)2D3 analog could deviate pancreas-infiltrating cells to the Th2 phenotype [82]. In addition, the proapoptotic activity of 1,25(OH)2D3 and its analogs can restore the defective sensitivity to apoptosis of NOD lymphocytes [97], leading to a more efficient elimination of potentially dangerous autoimmune effector cells. The increased apoptosis induced by 1,25(OH)2D3 and its analogs in DCs [39] and T cells [97] has been observed after different apoptosis-inducing signals and could help to explain why short-term treatments with these agents afford long-term protection and promote tolerance induction. The observation that ongoing type 1 diabetes in the adult NOD mouse can be arrested by a relatively short course of treatment with a 1,25(OH)2D3 analog [51] suggests that a similar treatment may also inhibit disease progression in prediabetic or newly diagnosed type 1 diabetes patients. Polymorphisms of the vitamin D receptor gene have been associated with type 1 diabetes in different populations [98,99], and epidemiological studies have shown a higher incidence of the disease in northern than in southern latitudes [100], suggesting a possible involvement of a 1,25(OH)2D3 deficiency in the pathogenesis of type 1 diabetes. This is further supported by a large population-based case-control study [101] and by a birth-cohort study [102] showing that the dietary vitamin D supplementation contributes to a significantly decreased risk of type 1 diabetes development.

C. Experimental Allergic Encephalomyelitis Experimental allergic encephalomyelitis (EAE) is considered as a model for multiple sclerosis (MS), and in both diseases Th1-type cells specific for myelin antigens appear to play a pathogenic role [4]. 1,25(OH)2D3 and the nonhypercalcemic analog (5Z,7E,23E,24aE)(1S,3R)-24a,24b-dihomo-9,10-seco-cholesta-5,7, 10(19),23,24a pentaene-1,3,25-triol (Ro 63-2023) have been shown to be selective and potent inhibitors of Th1 development in vitro and in vivo without inducing a deviation to the Th2 phenotype [55]. Administration of 1,25(OH)2D3 or its analog could prevent chronicrelapsing experimental allergic encephalomyelitis (CR-EAE) induced by the MOG peptide 35-55 in Biozzi AB/H mice, and this was associated with a profound reduction of MOG35-55-specific proliferation and Th1 cell development. Importantly, the nonhypercalcaemic analog Ro 63-2023 also provided long-term

CHAPTER 36 Regulation of Immune Responses by Vitamin D Receptor Ligands

protection from EAE relapses induced by immunization with spinal cord homogenate when administered for a short time either at symptom onset or even after the first peak of disease. Neuropathological analysis showed a significant reduction of inflammatory infiltrates, demyelinated areas and axonal loss in brains and spinal cords of treated mice. Thus, inhibition of IL-12-dependent Th1 cell development is associated with effective treatment of CR-EAE, further suggesting the feasibility of this approach in the treatment of multiple sclerosis [4]. These results demonstrate a correlation between the capacity of 1,25(OH)2D3 and the less calcemic analog Ro 63-2023 to inhibit IL-12-dependent Th1 development and to treat EAE, a correlation that was not established by previous studies [59,74,75,103]. Conversely, a systemic increase in the transcripts for TGF-β1 and IL-4 was suggested to be responsible for the capacity of 1,25(OH)2D3 to inhibit EAE [59], in contrast with the results of Mattner et al. [55], demonstrating that 1,25(OH)2D3 is a potent inhibitor of Th1 development and EAE without deviating the response to the Th2 pathway, as well as with the preferential inhibition of Th1 responses by 1,25(OH)2D3 observed by Lemire et al. [36]. The reasons for this discrepancy are not clear, although the different EAE models analyzed could play a role. TGF-β1 [104] and IL-4 [105,106] have been reported to be beneficial in EAE, but this activity has been ascribed to indirect inhibition of encephalitogenic Th1 cells. IL-10 also appears to be critical in the control of pathogenic Th1 responses in EAE [107], and 1,25(OH)2D3 has been shown in vitro to strongly enhance IL-10 production by human DCs [39] and to favor the induction of IL-10-producing regulatory T cells [66]. 1,25(OH)2D3 can cross the intact blood–brain barrier [108] and could therefore directly inhibit CNS APCs, such as microglia, that regulate intracerebral T cell responses [109], or target infiltrating T cells as well as recruited APCs. 1,25(OH)2D3 administration inhibits the expression of inducible nitric oxide synthase in macrophages, activated microglia, and astrocytes during EAE [110], and this could also contribute to amelioration of the disease. Alternatively, the immunomodulatory effects of VDR ligands could be mainly exerted in peripheral lymphoid organs, leading to inhibition of encephalitogenic T cell development.

639

These are chronic recurring illnesses most commonly involving inflammation of the terminal ileum and colon, although they can also affect many sites throughout the alimentary tract. In addition to genetic factors, including also VDR gene polymorphisms [111], the environment contributes to IBD development, and vitamin D may be an important environmental component in this respect. Lower amounts of 1,25(OH)2D3 are synthesized from sunlight exposure in areas in which IBDs occur most often, such as North America and Northern Europe [112], a situation common to other autoimmune diseases [113], in particular type 1 diabetes [100] and multiple sclerosis [114]. Dietary intake of vitamin D is problematic because few foods are naturally rich in vitamin D and weight loss, with consequently reduced vitamin D intake, occurs in the majority of IBD patients. In IBD models, the immune-mediated attack against the gastrointestinal tract has been shown to be mediated by Th1 cells [115], and the production of Th1-type cytokines has also been found associated with human IBDs [116]. Animal models have been developed in which IBD symptoms occur spontaneously, and a wellstudied one is the IL-10 knockout (KO) mouse [117]. In conventional animal facilities, IL-10 KO mice develop enterocolitis within the first 5 to 8 weeks of life, and approximately 30% of these mice die of severe anemia and weight loss [117]. The enterocolitis that develops in IL-10 KO mice is due to an uncontrolled immune response to conventional microbiota, because germfree IL-10 KO mice do not develop disease, and mice raised in specific pathogen–free facilities develop a milder disease [117]. IL-10 KO mice were made vitamin D deficient, vitamin D sufficient, or supplemented with 1,25(OH)2D3 [78]. Vitamin D–deficient, in contrast to vitamin D–sufficient IL-10 KO mice, rapidly developed diarrhea and a severe wasting disease. Administration of 1,25(OH)2D3 significantly ameliorated IBD symptoms in IL-10 KO mice and treatment for as little as 2 weeks blocked disease progression and ameliorated symptoms in mice with already established IBD [78]. This would be consistent with the observation that patients with Crohn’s disease have depressed IL-10 production and respond positively to IL-10 administration [118]. Interestingly, VDR ligands inhibit the proliferation of rectal epithelial cells [119] and of T cells [120] in active ulcerative colitis patients, further suggesting their possible use in the treatment of IBDs.

D. Inflammatory Bowel Disease E. Systemic Lupus Erythematosus Inflammatory bowel diseases (IBDs) are immunemediated diseases of unknown etiology affecting the gastrointestinal tract. At least two distinct forms of IBDs have been defined, ulcerative colitis and Crohn’s disease.

Systemic lupus erythematosus (SLE) is a T-cell dependent antibody-mediated autoimmune disease and the mouse strain MRLlpr/lpr spontaneously develops

640 a SLE-like syndrome sharing many immunological features with human SLE. Administration of VDR ligands significantly prolonged the average life span of MRLlpr/lpr mice and induced a significant reduction in proteinuria, renal arteritis, granuloma formation, and knee joint arthritis [71–73]. In addition, dermatological lesions, such as alopecia, necrosis of the ear, and scab formation, were also completely inhibited by 1,25(OH)2D3 therapy [73]. These data suggest a beneficial role of VDR ligands in the treatment of human SLE. Indeed, VDR ligands can significantly reduce cell proliferation and IgG production, both polyclonal and anti-dsDNA, while enhancing B cell apoptosis in lymphocytes from SLE patients [121]. However, it has also been shown that in (NZBxW)F1 mice, prone to developing SLE, treatment with 1,25(OH)2D3 worsens the disease, possibly explaining how sunlight could be a factor aggravating the course of SLE [122]. These results could be reconciled by the observation that MRLlpr/lpr mice receiving 1,25(OH)2D3 and a diet with a normal/high calcium content (0.87%) showed reduced SLE, whereas the same treatment in MRLlpr/lpr mice on a very low calcium content diet (0.02%) led to accelerated and more severe SLE [7], a situation already noted in EAE [123].

F. Psoriasis Psoriasis is a chronic inflammatory skin disease that affects about 2% of the population. Although the pathogenesis of psoriasis is still incompletely understood, it appears to be primarily a Th1-mediated autoimmune disease involving hyperproliferation of keratinocytes [124]. Given the capacity of VDR ligands to modulate both cell types, their success in treating psoriasis is perhaps not surprising. VDR ligands are currently the mainstay treatment in mild and moderate psoriasis, accounting for about 50% of all drugs used to treat this disease. At present, VDR ligands are used only topically, because a safe analog for systemic use has not yet been developed. In addition to topical calcitriol, calcipotriol and tacalcitol have shown efficacy and safety in extensive controlled studies [125–127]. Mechanistically, the beneficial effects of VDR ligands in psoriasis could reflect inhibition of proliferation and cytokine production by skin-infiltrating T cells [128]. VDR ligands have been shown to increase IL-10 production in psoriatic lesions [129] and to decrease IL-6 and IL-8 secretion by keratinocytes [130]. In addition, 1,25(OH)2D3 but not IL-10 could prevent leukocyte activation and reduce the histological and clinical scores in human psoriatic skin transplanted on to SCID mice [131]. The apoptotic process in psoriatic

LUCIANO ADORINI

lesions has been suggested to be in part regulated by Bcl-xL, and decreasing the expression of Bcl-xL by treatment with VDR ligands might ameliorate psoriatic lesions by contributing to the completion of the apoptotic process [132].

IV. BENEFICIAL EFFECTS OF VDR LIGANDS IN ALLOGRAFT REJECTION Acute allograft rejection is mediated by immunological mechanisms, with APCs, in particular dendritic cells (DCs) and T cells, playing a major role, whereas chronic rejection is mediated by a poorly understood combination of immunological and nonimmunological mechanisms [133]. Current immunosuppressive treatments based on small molecules (cyclosporine A, tacrolimus, sirolimus, mycophenolate mofetil) or on biologicals (anti-CD3, anti-CD52, anti-IL2R) can target quite effectively both T lymphocytes and APCs, but they have also important side effects. Drugs targeting both cell subsets, but devoid of major side effects, would therefore represent a useful addition to the available immunosuppressive agents. In addition, while near optimal control of acute rejection and adequate short-term graft survival have been achieved, problems associated with chronic rejection and long-term immunosuppressive management are rising [134]. Agents able to inhibit chronic rejection, and potentially able to promote transplantation tolerance, would thus fill an important unmet medical need. VDR ligands have pleiotropic immunoregulatory activities that are able to control allograft rejection, as demonstrated in different models of experimental organ transplantation, both acute and chronic (Table V). APCs and T cells can be direct targets of the hormone, leading to the inhibition of pathogenic effector T cells and enhancing the frequency of T cells with suppressive properties, largely via induction of tolerogenic DCs. These immunoregulatory activities, coupled with the absence of major side effects once calcemia is under control, have been translated into effective immunointervention in a variety of graft rejection models, both acute and chronic, showing potential for clinical applications [135,136].

A. Inhibition of Acute Rejection 1,25(OH)2D3 and its analogs can significantly prolong allograft survival in heart [137–139], kidney [140,141], liver [142], pancreatic islets [50,97,143,144], skin [145], and small bowel allografts [138]. In general,

641

CHAPTER 36 Regulation of Immune Responses by Vitamin D Receptor Ligands

TABLE V Effects of VDR Ligand Treatment in Animal Models of Transplantation Transplantated organ

VDR ligand tested

Aorta

MC 1288

Bone marrow

MC 1288

Heart

1,25(OH)2D3 16-ene-1,25(OH)2D3 Paricalcitol MC 1288

Kidney

1,25(OH)2D3 MC 1288

Liver

1,25(OH)2D3

Pancreatic islets

1,25(OH)2D3 KH 1060 MC 1288 TX 527

Skin

1,25(OH)2D3 KH 1060 CB 966

Small bowel

MC 1288

Synopsis of experimental protocol DA aortic allografts into WF rats treated with MC 1288 (0.1 µg/kg/eod) alone or together with CsA (5 mg/kg/d) Lewis bone marrow into BN rats treated with MC 1288 (0.1 µg/kg/eod) alone or together with CsA (5 mg/kg/d) Nonvascularized and vascularized heart allografts in mice or rats. Recipients were treated with VDR ligands alone or combined with low doses of CsA Kidney grafts in high-responder rat strain combinations. Recipients were treated with VDR ligands alone or combined with low doses of CsA

Main effects Reduced proliferation and activation of adventitial T cells induced by MC 1288 alone, decreased intimal hyperplasia when coadministered with CsA Decreased clinical and histological signs of graft-versus-host disease induced by MC 1288, decreased intimal hyperplasia when coadministered with CsA

[146]

[162]

All VDR ligands tested induced marked prolongation of heart allografts, usually superior to that induced by a full dose of CsA

[137–139]

1,25(OH)2D3 significantly prolonged allograft survival, preserved renal creatinine clearance, and decreased proteinuria. In combination with CsA induced inhibition of IL-2 and IL-12 as well as significant up-regulation of IL-10 expression. MC 1288 alone reduced clinical and histological signs of chronic rejection, and combined with CsA reduced the number of acute rejection episodes. Prolonged graft survival by decreasing the severity of acute rejection

[140,141]

ACI vascularized liver into Lewis rats treated with 1,25(OH)2D3 (0.1 or 1 µg/kg/d) alone or together with CsA (2 mg/kg/d) B6 islets transplanted into diabetic Prevention of allogeneic graft BALB/c mice; syngeneic or rejection and induction of xenogeneic islets transplanted tranferable tolerance; prevention into NOD mice. Recipients of autoimmune diabetes recurrence were treated with VDR ligands alone or combined with mycophenolate mofetil, CsA, or IFN-β B6 skin transplanted into Significantly prolonged graft CBA mice treated with survival, most evident with VDR ligands alone or KH 1060. Additive effects combined with CsA when combined with CsA Small bowel transplanted into Reduced amounts of hyaluronan allogenic rats treated with secreted into the intestinal lumen MC 1288 (0.1 µg/kg/d)

these effects have been achieved at the maximum tolerated dose, without inducing hypercalcemia, the major side effect of treatment with VDR ligands. In most experimental models, the acute rejection has been further delayed by combining VDR ligands with

References

[142]

[50,81,143,144]

[145]

[138]

a suboptimal dose of CsA or other immunosuppressive agents (Table V). Although treatment with VDR ligands has consistently shown efficacy in delaying the acute allograft rejection, the effects on chronic rejection are probably the most interesting.

642 B. Inhibition of Chronic Rejection VDR ligands, in association with low doses of cyclosporin A (CsA), can inhibit not only acute but also chronic allograft rejection, as documented by inhibition of adventitial inflammation and intimal hyperplasia in rat aortic allografts [146]. While the prevention of leukocyte infiltration into the adventitia is probably due to the immunomodulatory properties of VDR analogs, the inhibition of intimal cell proliferation, both endothelial and smooth muscle cells, is likely induced by their capacity to regulate cell growth. The 1,25(OH)2D3 analog MC 1288 also reduced clinical and histological signs of chronic graft rejection in rat kidney allografts [140]. The chronic allograft damage index, reflecting the sum of interstitial inflammation and fibrosis, glomerular mesangial matrix increase and sclerosis, vascular intimal proliferation and tubular atrophy, was significantly reduced in recipients treated with MC 1288 and CsA compared to CsA only. Renal graft loss has been found decelerated, in a retrospective study, in patients treated with 1,25(OH)2D3 [147], further suggesting its clinical applicability to inhibit chronic graft rejection.

C. Immunoregulatory Mechanisms Inhibiting Graft Rejection The induction of tolerogenic DCs by VDR ligands, which leads to an enhanced number of CD4+CD25+ regulatory T cells in vivo [50,51], is likely to play an important role in controlling graft rejection, both acute and chronic, and in favoring the establishment of transplantation tolerance. A short treatment with 1,25(OH)2D3 and mycophenolate mofetil, a selective inhibitor of T and B cell proliferation [148] that also modulates APCs [149], induces tolerance to islet allografts associated with an increased frequency of CD4+CD25+ regulatory T cells able to adoptively transfer transplantation tolerance [50]. The induction of tolerogenic DCs could indeed represent a therapeutic strategy promoting tolerance to allografts [150], and the observation that immature myeloid DCs can induce T cell tolerance to specific antigens in human volunteers represents an important proof of concept for this approach [151]. The direct effects of VDR ligands on T cells, in particular the inhibition of IL-2 and IFN-γ production, could also play a role in inhibiting graft rejection. 1,25(OH)2D3 inhibits IL-2 secretion by impairing the formation of the transcription factor complex NF-AT [56,57], and inhibits IFN-γ through interaction of the ligand-bound VDR complex with a VDRE in the promoter region of

LUCIANO ADORINI

the cytokine [58]. In addition, a combination of 1,25(OH)2D3 and low-dose CsA inhibited the expression of IL-2 and IL-12, and increased significantly IL10 expression levels in kidney allografts [141]. It is also possible, as suggested by our preliminary experiments, that VDR ligands may inhibit the production of proinflammatory chemokines by cells of the transplanted organ, thus inhibiting leukocyte recruitment to the graft. Additional mechanisms could rely on the capacity of 1,25(OH)2D3 to significantly reduce bioactive renal TGF-β1 by interacting with Smad proteins, important regulators of TGF-β signal transduction [152]. Since TGF-β has a pronounced pro-fibrotic activity, its decrease in the kidney tissue may inhibit the evolution of chronic rejection in kidney transplants.

D. Combination of VDR Ligands with Immunosuppressive Agents Based on the available evidence of a pro-tolerogenic effect and a reduced incidence of chronic rejection, VDR ligands could be added to standard immunosuppressive regimens in the treatment of allograft rejection. Additive and even synergistic effects have been observed between 1,25(OH)2D3 or its analogs and immunosuppressive agents, in particular CsA, tacrolimus, and sirolimus [83]. These effects have been confirmed in models of graft rejection (Table V), making VDR ligands potentially interesting as dose-reducing agents for conventional immunosuppressive drugs in clinical transplantation. Another positive feature of adding VDR ligands to standard immunosuppressive regimens is their protective effect on bone loss. A rapid bone loss is usually seen after organ transplantation and is enhanced by some immunosuppressive regimens, in particular those based on tacrolimus and steroids [153]. 1,25(OH)2D3 administration has been shown to prevent bone loss in transplanted patients [154–156], although standard prophylactic measures may not always be sufficient to prevent loss of bone mass [157], and 1,25(OH)2D3 analogs with a wider therapeutic window could serve this function also. In addition, 22-oxa-1,25(OH)2D3 (OCT) has been shown to exert an anabolic effect on bone reconstruction by vascularized bone allografts in rats [158], indicating a specific advantage of VDR ligand administration in bone transplantation. Importantly, the use of VDR ligands to control allograft rejection does not appear to increase opportunistic infections [159], a major side effect induced by antirejection drugs, in particular calcineurin inhibitors and glucocorticoids.

643

CHAPTER 36 Regulation of Immune Responses by Vitamin D Receptor Ligands

VDRL

VDRL CD4+ CD25+

IL -12

VDRL

CD40

CD4+

CC

L2 2

DC CD80

IL -2

CD86

M-DC

CTL

IFN -γ

IF Nγ

IL -10

25(OH)2D3

Target cell

-2 IL

IL T3

IFN -γ

Th1

1,25(OH)2D3

MΦ 0 -1 IL

MΦ

Th1 CD4+ CD25+

Th2

VDRL

Th2

-4 IL

VDRL

FIGURE 2 Mechanisms involved in the regulation of immune responses by VDR ligands. VDR ligands (VDRL) can modulate the immune response via several mechanisms in secondary lymphoid organs and in target tissues. In secondary lymphoid organs, 1,25(OH)2D3 inhibits IL-12 production and down-regulates costimulatory molecule expression (CD40, CD80, CD86) expressed by dendritic cells (DCs), while up-regulating their IL-10 production, thus inhibiting the development of Th1 cells and favoring the induction of CD4+CD25+ regulatory T cells and of Th2 cells. VDR ligands can also inhibit the migration of Th1 cells, and they up-regulate ILT3 expression and CCL22 production by myeloid DCs (M-DC), enhancing the recruitment of CD4+CD25+ regulatory T cells and of Th2 cells. In addition, VDR ligands exert direct effects on T cells by inhibiting IL-2 and IFN-γ production. Macrophages (MΦ), as well as DCs and T cells, can synthesize 1,25(OH)2D3 and this may also contribute to the regulation of the local immune response. In target tissues, pathogenic Th1 cells, which can damage target cells via induction of cytotoxic T cells (CTL) and activated macrophages, are reduced in number and their activity is further inhibited by CD4+CD25+ regulatory T cells and by Th2 cells induced by VDR ligands. Arrows indicate stimulation, blunted arrows inhibition, and broken arrows cytotoxicity.

V. CONCLUSIONS VDR ligands, in addition to controlling calcium metabolism and exerting important effects on the growth and differentiation of many cell types, possess pronounced pro-tolerogenic immunoregulatory properties. VDR ligands can act directly on T cells, but DCs appear to be primary targets for their tolerogenic properties. The capacity of VDR ligands to target DCs and T cells is mediated by VDR expression in both cell types and by the presence of common targets in their signal transduction pathways, such as the nuclear factor NF-κB that is down-regulated in APCs and in T cells. VDR ligands can induce in vitro and in vivo tolerogenic DCs able to enhance CD4+CD25+ suppressor T cells that, in turn, inhibit Th1 cell responses. These mechanisms of action can explain some of the immunoregulatory properties of VDR ligands that are potentially relevant for the treatment of Th1-mediated autoimmune diseases and allograft rejection, but may also represent a physiologic element in the regulation of innate and adaptive immune responses (Fig. 2).

A challenge for the future is the development of safe VDR ligands for the systemic treatment of psoriasis, and the translation of orally active VDR ligands to the treatment of other autoimmune diseases and allograft rejection. The accumulating evidence for the multiple immunomodulatory mechanisms regulated by VDR ligands indeed represents a sound basis for a further exploration of their potential in the development of therapies for several immunoemediated disorders.

References 1. Akira S 2003 Mammalian Toll-like receptors. Curr Opin Immunol 15:5–11. 2. Medzhitov R 2001 Toll-like receptors and innate immunity. Nat Rev Immunol 1:135–145. 3. Paul W 2003 Fundamental Immunology, 5th ed. Raven Press, New York. 4. Adorini L 2001 Selective immunointervention in autoimmune diseases: lessons from multiple sclerosis. J Chemother 13:219–234.

644 5. Lagaraine C, Lebranchu Y 2003 Effects of immunosuppressive drugs on dendritic cells and tolerance induction. Transplantation 75:37S–42S. 6. Coates PT, Colvin BL, Kaneko K, Taner T, Thomson AW 2003 Pharmacologic, biologic, and genetic engineering approaches to potentiation of donor-derived dendritic cell tolerogenicity. Transplantation 75:32S–36S. 7. Deluca HF, Cantorna MT 2001 Vitamin D: its role and uses in immunology. FASEB J 15:2579–2585. 8. Mathieu C, Adorini L 2002 The coming of age of 1,25dihydroxyvitamin D3 analogs as immunomodulatory agents. Trends Mol Med 8:174–179. 9. Griffin MD, Xing N, Kumar R 2003 Vitamin D and its analogs as regulators of immune activation and antigen presentation. Annu Rev Nutr 23:117–145. 10. Pinette KV, Yee YK, Amegadzie BY, Nagpal S 2003 Vitamin D receptor as a drug discovery target. Mini Rev Med Chem 3:193–204. 11. Norman AW, Mizwicki MT, Okamura WH 2003 Ligand structure–function relationships in the vitamin D endocrine system from the perspective of drug development (including cancer treatment). Recent Results Cancer Res 164:55–82. 12. Haussler MR, Whitfield GK, Haussler CA, Hsieh JC, Thompson PD, Selznick SH, Dominguez CE, Jurutka PW 1998 The nuclear vitamin D receptor: biological and molecular regulatory properties revealed. J Bone Miner Res 13:325–349. 13. Norman AW, Ishizuka S, Okamura WH 2001 Ligands for the vitamin D endocrine system: different shapes function as agonists and antagonists for genomic and rapid response receptors or as a ligand for the plasma vitamin D binding protein. J Steroid Biochem Mol Biol 76:49–59. 14. Carlberg C, Polly P 1998 Gene regulation by vitamin D3. Crit Rev Eukaryot Gene Expr 8:19–42. 15. Provvedini DM, Tsoukas CD, Deftos LJ, Manolagas SC 1983 1,25-Dihydroxyvitamin D3 receptors in human leukocytes. Science 221:1181–1183. 16. Brennan A, Katz DR, Nunn JD, Barker S, Hewison M, Fraher LJ, O’Riordan JL 1987 Dendritic cells from human tissues express receptors for the immunoregulatory vitamin D3 metabolite, dihydroxycholecalciferol. Immunology 61: 457–461. 17. Veldman CM, Cantorna MT, DeLuca HF 2000 Expression of 1,25-dihydroxyvitamin D3 receptor in the immune system. Arch Biochem Biophys 374:334–338. 18. Bhalla AK, Amento EP, Serog B, Glimcher LH 1984 1,25Dihydroxyvitamin D3 inhibits antigen-induced T cell activation. J Immunol 133:1748–1754. 19. Adorini L 2002 Immunomodulatory effects of vitamin D receptor ligands in autoimmune diseases. Int Immunopharmacol 2:1017–1028. 20. Barnes PF, Modlin RL, Bikle DD, Adams JS 1989 Transpleural gradient of 1,25-dihydroxyvitamin D in tuberculous pleuritis. J Clin Invest 83:1527–1532. 21. Cadranel J, Garabedian M, Milleron B, Guillozo H, Akoun G, Hance AJ 1990 1,25(OH)2D2 production by T lymphocytes and alveolar macrophages recovered by lavage from normocalcemic patients with tuberculosis. J Clin Invest 85:1588–1593. 22. Overbergh L, Decallonne B, Valckx D, Verstuyf A, Depovere J, Laureys J, Rutgeerts O, Saint-Arnaud R, Bouillon R, Mathieu C 2000 Identification and immune regulation of 25-hydroxyvitamin D-1-alpha-hydroxylase in murine macrophages. Clin Exp Immunol 120:139–146. 23. Hewison M, Freeman L, Hughes SV, Evans KN, Bland R, Eliopoulos AG, Kilby MD, Moss PA, Chakraverty R 2003

LUCIANO ADORINI

24.

25. 26. 27. 28. 29.

30.

31.

32. 33.

34.

35.

36. 37.

38. 39.

40.

Differential regulation of vitamin D receptor and its ligand in human monocyte-derived dendritic cells. J Immunol 170:5382–5390. Griffin MD, Lutz W, Phan VA, Bachman LA, McKean DJ, Kumar R 2001 Dendritic cell modulation by 1α,25 dihydroxyvitamin D3 and its analogs: A vitamin D receptor–dependent pathway that promotes a persistent state of immaturity in vitro and in vivo. Proc Natl Acad Sci USA 98:6800–6805. Banchereau J, Steinman RM 1998 Dendritic cells and the control of immunity. Nature 392:245–252. Banchereau J, Briere F, Caux C, Davoust J, Lebecque S, Liu YJ, Pulendran B, Palucka K 2000 Immunobiology of dendritic cells. Ann Rev Immunol 18:767–811. Reid S, Penna G, Adorini L 2000 The control of T cell responses by dendritic cell subsets. Curr Opin Immunol 12:114–121. Shortman K, Liu YJ 2002 Mouse and human dendritic cell subtypes. Nat Rev Immunol 2:151–161. Asselin-Paturel C, Boonstra A, Dalod M, Durand I, Yessaad N, Dezutter-Dambuyant C, Vicari A, O’Garra A, Biron C, Briere F, Trinchieri G 2001 Mouse type I IFN-producing cells are immature APCs with plasmacytoid morphology. Nat Immunol 2:1144–1150. Steinman RM, Hawiger D, Liu K, Bonifaz L, Bonnyay D, Mahnke K, Iyoda T, Ravetch J, Dhodapkar M, Inaba K, Nussenzweig M 2003 Dendritic cell function in vivo during the steady state: a role in peripheral tolerance. Ann NY Acad Sci 987:15–25. Jonuleit H, Schmitt E, Schuler G, Knop J, Enk AH 2000 Induction of interleukin 10-producing, nonproliferating CD4+ T cells with regulatory properties by repetitive stimulation with allogeneic immature human dendritic cells. J Exp Med 192:1213–1222. Lutz MB, Schuler G 2002 Immature, semi-mature and fully mature dendritic cells: Which signals induce tolerance or immunity? Trends Immunol 23:445–449. Tsoukas CD, Watry D, Escobar SS, Provvedini DM, Dinarello CA, Hustmyer FG, Manolagas SC 1989 Inhibition of interleukin-1 production by 1,25-dihydroxyvitamin D3. J Clin Endocrinol Metab 69:127–133. Rigby WF, Waugh M, Graziano RF 1990 Regulation of human monocyte HLA-DR and CD4 antigen expression, and antigen presentation by 1,25-dihydroxyvitamin D3. Blood 76:189–197. Xu H, Soruri A, Gieseler RK, Peters JH 1993 1,25Dihydroxyvitamin D3 exerts opposing effects to IL-4 on MHC class-II antigen expression, accessory activity, and phagocytosis of human monocytes. Scand J Immunol 38:535–540. Lemire JM, Archer DC, Beck L, Spiegelberg HL 1995 Immunosuppressive actions of 1,25-dihydroxyvitamin D3: preferential inhibition of Th1 functions. J Nutr 125:1704S–1708S. D’Ambrosio D, Cippitelli M, Cocciolo MG, Mazzeo D, Di Lucia P, Lang R, Sinigaglia F, Panina-Bordignon P 1998 Inhibition of IL-12 production by 1,25-dihydroxyvitamin D3. Involvement of NF-kappaB downregulation in transcriptional repression of the p40 gene. J Clin Invest 101:252–262. Adorini L 1999 Interleukin-12, a key cytokine in Th1-mediated autoimmune diseases. Cell Mol Life Sci 55:1610–1625. Penna G, Adorini L 2000 1,25-Dihydroxyvitamin D3 inhibits differentiation, maturation, activation and survival of dendritic cells leading to impaired alloreactive T cell activation. J Immunol 164:2405–2411. Piemonti L, Monti P, Sironi M, Fraticelli P, Leone BE, Dal Cin E, Allavena P, Di Carlo V 2000 Vitamin D3 affects differentiation, maturation, and function of human monocyte-derived dendritic cells. J Immunol 164:4443–4451.

CHAPTER 36 Regulation of Immune Responses by Vitamin D Receptor Ligands

41. Griffin MD, Lutz WH, Phan VA, Bachman LA, McKean DJ, Kumar R 2000 Potent inhibition of dendritic cell differentiation and maturation by vitamin D analogs. Biochem Biophys Res Commun 270:701–708. 42. Berer A, Stockl J, Majdic O, Wagner T, Kollars M, Lechner K, Geissler K, Oehler L 2000 1,25-Dihydroxyvitamin D3 inhibits dendritic cell differentiation and maturation in vitro. Exp Hematol 28:575–583. 43. Canning MO, Grotenhuis K, de Wit H, Ruwhof C, Drexhage HA 2001 1-α,25-Dihydroxyvitamin D3 (1,25(OH)2D3) hampers the maturation of fully active immature dendritic cells from monocytes. Eur J Endocrinol 145:351–357. 44. van Halteren AG, van Etten E, de Jong EC, Bouillon R, Roep BO, Mathieu C 2002 Redirection of human autoreactive T-cells upon interaction with dendritic cells modulated by TX527, an analog of 1,25 dihydroxyvitamin D3. Diabetes 51:2119–2125. 45. Chang CC, Ciubotariu R, Manavalan JS, Yuan J, Colovai AI, Piazza F, Lederman S, Colonna M, Cortesini R, Dalla-Favera R, Suciu-Foca N 2002 Tolerization of dendritic cells by T(S) cells: the crucial role of inhibitory receptors ILT3 and ILT4. Nat Immunol 3:237–243. 46. Yu XP, Bellido T, Manolagas SC 1995 Down-regulation of NF-kappa B protein levels in activated human lymphocytes by 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 92:10990–10994. 47. Martin E, O’Sullivan B, Low P, Thomas R 2003 Antigenspecific suppression of a primed immune response by dendritic cells mediated by regulatory T cells secreting interleukin-10. Immunity 18:155–167. 48. Towers TL, Freedman LP 1998 Granulocyte-macrophage colony-stimulating factor gene transcription is directly repressed by the vitamin D3 receptor. Implications for allosteric influences on nuclear receptor structure and function by a DNA element. J Biol Chem 273:10338–10348. 49. Towers TL, Staeva TP, Freedman LP 1999 A two-hit mechanism for vitamin D3–mediated transcriptional repression of the granulocyte-macrophage colony-stimulating factor gene: vitamin D receptor competes for DNA binding with NFAT1 and stabilizes c-Jun. Mol Cell Biol 19:4191–4199. 50. Gregori S, Casorati M, Amuchastegui S, Smiroldo S, Davalli AM, Adorini L 2001 Regulatory T cells induced by 1α,25-dihydroxyvitamin D3 and mycophenolate mofetil treatment mediate transplantation tolerance. J Immunol 167:1945–1953. 51. Gregori G, Giarratana N, Smiroldo S, Uskokovic M, Adorini L 2002 A 1α,25-dihydroxyvitamin D3 analog enhances regulatory T cells and arrests autoimmune diabetes in NOD mice. Diabetes 51:1376–1374. 52. Nashold FE, Hoag KA, Goverman J, Hayes CE 2001 Rag1-dependent cells are necessary for 1,25-dihydroxyvitamin D3 prevention of experimental autoimmune encephalomyelitis. J Neuroimmunol 119:16–29. 53. Bhalla AK, Amento EP, Clemens TL, Holick MF, Krane SM 1983 Specific high-affinity receptors for 1,25-dihydroxyvitamin D3 in human peripheral blood mononuclear cells: presence in monocytes and induction in T lymphocytes following activation. J Clin Endocrinol Metab 57:1308–1310. 54. Rigby WF, Denome S, Fanger MW 1987 Regulation of lymphokine production and human T lymphocyte activation by 1,25-dihydroxyvitamin D3. Specific inhibition at the level of messenger RNA. J Clin Invest 79:1659–1664. 55. Mattner F, Smiroldo S, Galbiati F, Muller M, Di Lucia P, Poliani PL, Martino G, Panina-Bordignon P, Adorini L 2000 Inhibition of Th1 development and treatment of

56.

57.

58. 59.

60.

61. 62.

63.

64.

65.

66.

67. 68. 69.

70.

645

chronic-relapsing experimental allergic encephalomyelitis by a non-hypercalcemic analogue of 1,25-dihydroxyvitamin D3. Eur J Immunol 30:498–508. Alroy I, Towers T, Freedman L 1995 Transcriptional repression of the interleukin-2 gene by vitamin D3: direct inhibition NFATp/AP-1 complex formation by a nuclear hormone receptor. Mol Cell Biol 15:5789–5799. Takeuchi A, Reddy G, Kobayashi T, Okano T, Park J, Sharma S 1998 Nuclear factor of activated T cells (NFAT) as a molecular target for 1α,25-dihydroxyvitamin D3–mediated effects. J Immunol 160:209–218. Cippitelli M, Santoni A 1998 Vitamin D3: a transcriptional modulator of the IFN-γ gene. Eur J Immunol 28: 3017–3030. Cantorna M, Woodward W, Hayes C, DeLuca H 1998 1,25-Dihydroxyvitamin D3 is a positive regulator for the two anti-encephalitogenic cytokines TGF-β1 and IL-4. J Immunol 160:5314–5319. Cippitelli M, Fionda C, Di Bona D, Di Rosa F, Lupo A, Piccoli M, Frati L, Santoni A 2002 Negative regulation of CD95 ligand gene expression by vitamin D3 in T lymphocytes. J Immunol 168:1154–1166. Suzuki I, Martin S, Boursalian TE, Beers C, Fink PJ 2000 Fas ligand costimulates the in vivo proliferation of CD8+ T cells. J Immunol 165:5537–5543. Rescigno M, Piguet V, Valzasina B, Lens S, Zubler R, French L, Kindler V, Tschopp J, Ricciardi-Castagnoli P 2000 Fas engagement induces the maturation of dendritic cells (DCs), the release of interleukin (IL)-1beta, and the production of interferon gamma in the absence of IL-12 during DC-T cell cognate interaction: a new role for Fas ligand in inflammatory responses. J Exp Med 192:1661–1668. Boonstra A, Barrat FJ, Crain C, Heath VL, Savelkoul HF, O’Garra A 2001 1α,25-Dihydroxyvitamin D3 has a direct effect on naive CD4+ T cells to enhance the development of Th2 cells. J Immunol 167:4974–4980. Overbergh L, Decallonne B, Waer M, Rutgeerts O, Valckx D, Casteels KM, Laureys J, Bouillon R, Mathieu C 2000 1α,25Dihydroxyvitamin D3 induces an autoantigen-specific T-helper 1/T-helper 2 immune shift in NOD mice immunized with GAD65 (p524–543). Diabetes 49:1301–1307. Staeva-Vieira TP, Freedman LP 2002 1,25-Dihydroxyvitamin D3 inhibits IFN-gamma and IL-4 levels during in vitro polarization of primary murine CD4+ T cells. J Immunol 168:1181–1189. Barrat FJ, Cua DJ, Boonstra A, Richards DF, Crain C, Savelkoul HF, de Waal-Malefyt R, Coffman RL, Hawrylowicz CM, O’Garra A 2002 In vitro generation of interleukin 10-producing regulatory CD4+ T cells is induced by immunosuppressive drugs and inhibited by T helper type 1 (Th1)- and Th2-inducing cytokines. J Exp Med 195:603–616. Shevach EM 2002 CD4+ CD25+ suppressor T cells: More questions than answers. Nat Rev Immunol 2:389–400. D’Ambrosio D, Sinigaglia F, Adorini L 2003 Special attractions for suppressor T cells. Trends Immunol 24: 122–126. Penna G, Vulcano M, Roncari A, Facchetti F, Sozzani S, Adorini L 2002 Differential chemokine production by myeloid and plasmacytoid dendritic cells. J Immunol 169:6673–6676. Groux H, O’Garra A, Bigler M, Rouleau M, Antonenko S, de Vries JE, Roncarolo MG 1997 A CD4+ T-cell subset inhibits antigen-specific T-cell responses and prevents colitis. Nature 389:737–742.

646 71. Koizumi T, Nakao Y, Matsui T, Nakagawa T, Matsuda S, Komoriya K, Kanai Y, Fujita T 1985 Effects of corticosteroid and 1,24R-dihydroxy-vitamin D3 administration on lymphoproliferation and autoimmune disease in MRL/MPlpr/lpr mice. Int Arch Allergy Appl Immunol 77:396–404. 72. Abe J, Nakamura K, Takita Y, Nakano T, Irie H, Nishii Y 1990 Prevention of immunological disorders in MRL/l mice by a new synthetic analogue of vitamin D3: 22-oxa-1α,25-dihydroxyvitamin D3. J Nutr Sci Vitaminol (Tokyo) 36:21–31. 73. Lemire JM, Ince A, Takashima M 1992 1,25-Dihydroxyvitamin D3 attenuates the expression of experimental murine lupus of MRL/l mice. Autoimmunity 12:143–148. 74. Lemire JM, Archer DC 1991 1,25-Dihydroxyvitamin D3 prevents the in vivo induction of murine experimental autoimmune encephalomyelitis. J Clin Invest 87:1103–1107. 75. Cantorna MT, Hayes CE, DeLuca HF 1996 1,25-Dihydroxyvitamin D3 reversibly blocks the progression of relapsing encephalomyelitis, a model of multiple sclerosis. Proc Natl Acad Sci USA 93:7861–7864. 76. Larsson P, Mattsson L, Klareskog L, Johnsson C 1998 A vitamin D analogue (MC 1288) has immunomodulatory properties and suppresses collagen-induced arthritis (CIA) without causing hypercalcaemia. Clin Exp Immunol 114:277–283. 77. Cantorna MT, Hayes CE, DeLuca HF 1998 1,25Dihydroxycholecalciferol inhibits the progression of arthritis in murine models of human arthritis. J Nutr 128:68–72. 78. Cantorna MT, Munsick C, Bemiss C, Mahon BD 2000 1,25Dihydroxycholecalciferol prevents and ameliorates symptoms of experimental murine inflammatory bowel disease. J Nutr 130:2648–2652. 79. Mathieu C, Waer M, Laureys J, Rutgeerts O, Bouillon R 1994 Prevention of autoimmune diabetes in NOD mice by 1,25 dihydroxyvitamin D3. Diabetologia 37:552–558. 80. Mathieu C, Waer M, Casteels K, Laureys J, Bouillon R 1995 Prevention of type I diabetes in NOD mice by nonhypercalcemic doses of a new structural analog of 1,25-dihydroxyvitamin D3, KH1060. Endocrinology 136:866–872. 81. Casteels K, Waer M, Laureys J, Valckx D, Depovere J, Bouillon R, Mathieu C Prevention of autoimmune destruction of syngeneic islet grafts in spontaneously diabetic nonobese diabetic mice by a combination of a vitamin D3 analog and cyclosporine. Transplantation 65:1225–1232. 82. Casteels KM, Mathieu C, Waer M, Valckx D, Overbergh L, Laureys JM, Bouillon R 1998 Prevention of type I diabetes in nonobese diabetic mice by late intervention with nonhypercalcemic analogs of 1,25-dihydroxyvitamin D3 in combination with a short induction course of cyclosporin A. Endocrinology 139:95–102. 83. van Etten E, Branisteanu DD, Verstuyf A, Waer M, Bouillon R, Mathieu C 2000 Analogs of 1,25-dihydroxyvitamin D3 as dose-reducing agents for classical immunosuppressants. Transplantation 69:1932–1942. 84. Schulze-Koops H, Kalden JR 2001 The balance of Th1/Th2 cytokines in rheumatoid arthritis. Best Pract Res Clin Rheumatol 15:677–691. 85. Tetlow LC, Woolley DE 2001 Expression of vitamin D receptors and matrix metalloproteinases in osteoarthritic cartilage and human articular chondrocytes in vitro. Osteoarthritis Cartilage 9:423–431. 86. Tetlow LC, Woolley DE 1999 The effects of 1α,25-dihydroxyvitamin D3 on matrix metalloproteinase and prostaglandin E2 production by cells of the rheumatoid lesion. Arthritis Res 1:63–70. 87. Atkinson MA, Leiter EH 1999 The NOD mouse model of type 1 diabetes: As good as it gets? Nat Med 5:601–604.

LUCIANO ADORINI

88. Benoist C, Mathis D 1997 Cell death mediators in autoimmune diabetes—no shortage of suspects. Cell 89:1–3. 89. Trembleau S, Penna G, Bosi E, Mortara A, Gately MK, Adorini L 1995 IL-12 administration induces Th1 cells and accelerates autoimmune diabetes in NOD mice. J Exp Med 181:817–821. 90. Ridgway WM, Fasso M, Lanctot A, Garvey C, Fathman CG 1996 Breaking self-tolerance in nonobese diabetic mice. J Exp Med 183:1657–1662. 91. Delovitch TL, Singh B 1997 The non-obese diabetic mouse as a model of autoimmune diabetes: immune dysregulation gets the NOD. Immunity 7:727–738. 92. Adorini L 2001 Interleukin 12 and autoimmune diabetes. Nat Genet 27:131–132. 93. Salomon B, Lenschow DJ, Rhee L, Ashourian N, Singh B, Sharpe A, Bluestone JA 2000 B7/CD28 costimulation is essential for the homeostasis of the CD4+CD25+ immunoregulatory T cells that control autoimmune diabetes. Immunity 12:431–440. 94. Mathieu C, Laureys J, Sobis H, Vandeputte M, Waer M, Bouillon R 1992 1,25-Dihydroxyvitamin D3 prevents insulitis in NOD mice. Diabetes 41:1491–1495. 95. Mathieu C, Casteels K, Boullion R 1997 Vitamin D and diabetes. In: Feldman, Glorieux, Pike (eds) Vitamin D. Academic Press, New York. 96. Uskokovic MR, Norman AW, Manchand PS, Studzinski GP, Campbell MJ, Koeffler HP, Takeuchi A, Siu-Caldera M, Rao DS, Reddy GS 2001 Highly active analogs of 1α,25dihydroxyvitamin D3 that resist metabolism through C-24 oxidation and C-3 epimerization pathways. Steroids 66: 463–471. 97. Casteels KM, Gysemans CA, Waer M, Bouillon R, Laureys JM, Depovere J, Mathieu C 1998 Sex difference in resistance to dexamethasone-induced apoptosis in NOD mice: treatment with 1,25(OH)2D3 restores defect. Diabetes 47: 1033–1037. 98. Pani MA, Knapp M, Donner H, Braun J, Baur MP, Usadel KH, Badenhoop K 2000 Vitamin D receptor allele combinations influence genetic susceptibility to type 1 diabetes in Germans. Diabetes 49:504–507. 99. Chang TJ, Lei HH, Yeh JI, Chiu KC, Lee KC, Chen MC, Tai TY, Chuang LM 2000 Vitamin D receptor gene polymorphisms influence susceptibility to type 1 diabetes mellitus in the Taiwanese population. Clin Endocrinol 52: 575–580. 100. Green A, Gale EA, Patterson CC 1992 Incidence of childhood-onset insulin-dependent diabetes mellitus: The EURODIAB ACE Study. Lancet 339:905–909. 101. Vitamin D supplement in early childhood and risk for Type I (insulin-dependent) diabetes mellitus. The EURODIAB Substudy 2 Study Group 1999 Diabetologia 42:51–54. 102. Hypponen E, Laara E, Reunanen A, Jarvelin MR, Virtanen SM 2001 Intake of vitamin D and risk of type 1 diabetes: a birthcohort study. Lancet 358:1500–1503. 103. Branisteanu D, Waer M, Sobis H, Marcelis S, Vandeputte M, Boullion R 1995 Prevention of murine experimental allergic encephalomyelitis: Cooperative effects of cyclosporine and 1 α,25-(OH)2D3. J Neuroimmunol 61:151–160. 104. Racke M, Dhib-Jalbut S, Cannella B, Alert P, Raine C, McFarlin D 1991 Prevention and treatment of chronic relapsing experimental allergic encephalomyelitis by transforming growth factor-β1. J Immunol 146:3012–3019. 105. Racke MK, Bonomo A, Scott DE, Cannella B, Levine A, Raine CS, Shevach EM, Roecken M 1994 Cytokine-induced immune deviation as a therapy for inflammatory autoimmune disease. J Exp Med 180:1961–1966.

CHAPTER 36 Regulation of Immune Responses by Vitamin D Receptor Ligands

106. Furlan R, Poliani PL, Galbiati F, Bergami A, Grimaldi L, Comi G, Adorini L, Martino G 1998 Central nervous system delivery of interleukin 4 by a nonreplicative herpes simplex type 1 viral vector ameliorates autoimmune demyelination. Hum Gene Ther 9:2605–2617. 107. Bettelli E, Prabhu Das M, Howard ED, Weiner HL, Sobel RA, Kuchroo VK 1998 IL-10 is critical in the regulation of autoimmune encephalomyelitis as demonstrated by studies of IL-10- and IL-4-deficient and transgenic mice. J Immunol 161:3299–3306. 108. Gascon-Barre M, Huet P 1983 Apparent [3H]1,25-dihydroxyvitamin D3 uptake by canine and rodent brain. Am J Physiol 244:266–271. 109. Aloisi F, Ria F, Adorini L 2000 Regulation of T-cell responses by CNS antigen-presenting cells: Different roles for microglia and astrocytes. Immunol Today 21:141–147. 110. Garcion E, Nataf S, Berod A, Darcy F, Brachet P 1997 1,25-Dihydroxyvitamin D3 inhibits the expression of inducible nitric oxide synthase in rat central nervous system during experimental allergic encephalomyelitis. Brain Res 45:255–267. 111. Simmons JD, Mullighan C, Welsh KI, Jewell DP 2000 Vitamin D receptor gene polymorphism: Association with Crohn’s disease susceptibility. Gut 47:211–214. 112. Hendrickson BA, Gokhale R, Cho JH 2002 Clinical aspects and pathophysiology of inflammatory bowel disease. Clin Microbiol Rev 15:79–94. 113. Cantorna MT 2000 Vitamin D and autoimmunity: Is vitamin D status an environmental factor affecting autoimmune disease prevalence? Proc Soc Exp Biol Med 223:230–233. 114. Hayes CE 2000 Vitamin D: a natural inhibitor of multiple sclerosis. Proc Nutr Soc 59:531–535. 115. Bregenholt S, Claesson MH 1998 Increased intracellular Th1 cytokines in scid mice with inflammatory bowel disease. Eur J Immunol 28:379–389. 116. Niessner M, Volk BA 1995 Altered Th1/Th2 cytokine profiles in the intestinal mucosa of patients with inflammatory bowel disease as assessed by quantitative reversed transcribed polymerase chain reaction (RT-PCR). Clin Exp Immunol 101:428–435. 117. Kuhn R, Lohler J, Rennick D, Rajewsky K, Muller W 1993 Interleukin-10-deficient mice develop chronic enterocolitis. Cell 75:263–274. 118. Narula SK, Cutler D, Grint P 1998 Immunomodulation of Crohn’s disease by interleukin-10. Agents Actions Suppl 49:57-65. 119. Thomas MG, Nugent KP, Forbes A, Williamson RC 1994 Calcipotriol inhibits rectal epithelial cell proliferation in ulcerative proctocolitis. Gut 35:1718–1720. 120. Stio M, Bonanomi AG, d’Albasio G, Treves C 2001 Suppressive effect of 1,25-dihydroxyvitamin D3 and its analogues EB 1089 and KH 1060 on T lymphocyte proliferation in active ulcerative colitis. Biochem Pharmacol 61: 365–371. 121. Linker-Israeli M, Elstner E, Klinenberg JR, Wallace DJ, Koeffler HP 2001 Vitamin D3 and its synthetic analogs inhibit the spontaneous in vitro immunoglobulin production by SLE-derived PBMC. Clin Immunol 99:82–93. 122. Vaisberg MW, Kaneno R, Franco MF, Mendes NF 2000 Influence of cholecalciferol (vitamin D3) on the course of experimental systemic lupus erythematosus in F1 (NZBxW) mice. J Clin Lab Anal 14:91–96. 123. Cantorna MT, Humpal-Winter J, DeLuca HF 1999 Dietary calcium is a major factor in 1,25-dihydroxycholecalciferol suppression of experimental autoimmune encephalomyelitis in mice. J Nutr 129:1966–1971.

647

124. Uyemura K, Yamamura M, Fivenson DF, Modlin RL, Nickoloff BJ 1993 The cytokine network in lesional and lesion-free psoriatic skin is characterized by a T-helper type 1 cell-mediated response. J Invest Dermatol 101:701–705. 125. Bourke JF, Iqbal SJ, Hutchinson PE 1997 A randomized double-blind comparison of the effects on systemic calcium homeostasis of topical calcitriol (3 micrograms/g) and calcipotriol (50 micrograms/g) in the treatment of chronic plaque psoriasis vulgaris. Acta Derm Venereol 77: 228–230. 126. Veien NK, Bjerke JR, Rossmann-Ringdahl I, Jakobsen HB 1997 Once daily treatment of psoriasis with tacalcitol compared with twice daily treatment with calcipotriol. A double-blind trial. Br J Dermatol 137:581–586. 127. Ashcroft DM, Po AL, Williams HC, Griffiths CE 2000 Systematic review of comparative efficacy and tolerability of calcipotriol in treating chronic plaque psoriasis. Br Med J 320:963–967. 128. Barna M, Bos JD, Kapsenberg ML, Snijdewint FG 1997 Effect of calcitriol on the production of T-cell-derived cytokines in psoriasis. Br J Dermatol 136:536–541. 129. Kang S, Yi S, Griffiths CE, Fancher L, Hamilton TA, Choi JH 1998 Calcipotriene-induced improvement in psoriasis is associated with reduced interleukin-8 and increased interleukin-10 levels within lesions. Br J Dermatol 138: 77–83. 130. Komine M, Watabe Y, Shimaoka S, Sato F, Kake K, Nishina H, Ohtsuki M, Nakagawa H, Tamaki K 1999 The action of a novel vitamin D3 analogue, OCT, on immunomodulatory function of keratinocytes and lymphocytes. Arch Dermatol Res 291: 500–506. 131. Dam TN, Kang S, Nickoloff BJ, Voorhees JJ 1999 1α,25Dihydroxycholecalciferol and cyclosporine suppress induction and promote resolution of psoriasis in human skin grafts transplanted on to SCID mice. J Invest Dermatol 113:1082–1089. 132. Fukuya Y, Higaki M, Higaki Y, Kawashima M 2002 Effect of vitamin D3 on the increased expression of Bcl-xL in psoriasis. Arch Dermatol Res 293:620–625. 133. Gourishankar S, Halloran P 2002 Late deterioration of organ transplants: A problem in injury and homeostasis. Curr Opin Immunol 14:576. 134. Pascual M, Theruvath T, Kawai T, Tolkoff-Rubin N, Cosimi AB 2002 Strategies to improve long-term outcomes after renal transplantation. N Engl J Med 346:580–590. 135. Adorini L 2002 1,25-Dihydroxyvitamin D3 analogs as potential therapies in transplantation. Curr Opin Investig Drugs 3:1458–1463. 136. Becker BN, Hullett DA, O’Herrin JK, Malin G, Sollinger HW, DeLuca H 2002 Vitamin D as immunomodulatory therapy for kidney transplantation. Transplantation 74: 1204–1206. 137. Lemire JM, Archer DC, Khulkarni A, Ince A, Uskokovic MR, Stepkowski S 1992 Prolongation of the survival of murine cardiac allografts by the vitamin D3 analogue 1,25-dihydroxyδ16-cholecalciferol. Transplantation 54:762–763. 138. Johnsson C, Tufveson G 1994 MC 1288—a vitamin D analogue with immunosuppressive effects on heart and small bowel grafts. Transpl Int 7:392–397. 139. Hullett DA, Cantorna MT, Redaelli C, Humpal-Winter J, Hayes CE, Sollinger HW, Deluca HF 1998 Prolongation of allograft survival by 1,25-dihydroxyvitamin D3. Transplantation 66:824–828. 140. Kallio E, Hayry P, Pakkala S 1996 MC1288, a vitamin D analogue, reduces short- and long-term renal allograft rejection in the rat. Transplant Proc 28:3113.

648 141. Redaelli CA, Wagner M, Gunter-Duwe D, Tian YH, Stahel PF, Mazzucchelli L, Schmid RA, Schilling MK 2002 1α,25-Dihydroxyvitamin D3 shows strong and additive immunomodulatory effects with cyclosporine A in rat renal allotransplants. Kidney Int 61:288–296. 142. Redaelli CA, Wagner M, Tien YH, Mazzucchelli L, Stahel PF, Schilling MK, Dufour JF 2001 1α,25-Dihydroxycholecalciferol reduces rejection and improves survival in rat liver allografts. Hepatology 34:926–934. 143. van Etten E, Gysemans C, Verstuyf A, Bouillon R, Mathieu C 2001 Immunomodulatory properties of a 1,25(OH)2 vitamin D3 analog combined with IFNbeta in an animal model of syngeneic islet transplantation. Transplant Proc 33:2319. 144. Gysemans C, Waer M, Laureys J, Bouillon R, Mathieu C 2001 A combination of KH1060, a vitamin D3 analogue, and cyclosporin prevents early graft failure and prolongs graft survival of xenogeneic islets in nonobese diabetic mice. Transplant Proc 33:2365. 145. Veyron P, Pamphile R, Binderup L, Touraine JL 1993 Two novel vitamin D analogues, KH 1060 and CB 966, prolong skin allograft survival in mice. Transpl Immunol 1:72–76. 146. Raisanen-Sokolowski AK, Pakkala IS, Samila SP, Binderup L, Hayry PJ, Pakkala ST 1997 A vitamin D analog, MC1288, inhibits adventitial inflammation and suppresses intimal lesions in rat aortic allografts. Transplantation 63:936–941. 147. Aschenbrenner J, Hullett D, Heisey D, Pirsch J, Sollinger H, Becker B 2000 1,25-Dihydroxyvitamin D3 (1,25(OH)2D3) has salutary effects on renal allograft function. Transplantation 69:S448. 148. Allison AC, Eugui EM 1996 Purine metabolism and immunosuppressive effects of mycophenolate mofetil (MMF). Clin Transplant 10:77–84. 149. Mehling A, Grabbe S, Voskort M, Schwarz T, Luger TA, Beissert S 2000 Mycophenolate mofetil impairs the maturation and function of murine dendritic cells. J Immunol 165:2374–2381. 150. Lu L, Thomson AW 2002 Manipulation of dendritic cells for tolerance induction in transplantation and autoimmune disease. Transplantation 73:S19–22. 151. Dhodapkar MV, Steinman RM, Krasovsky J, Munz C, Bhardwaj N 2001 Antigen-specific inhibition of effector T cell function in humans after injection of immature dendritic cells. J Exp Med 193:233–238. 152. Aschenbrenner JK, Sollinger HW, Becker BN, Hullett DA 2001 1,25-OH2D3 alters the transforming growth factor beta signaling pathway in renal tissue. J Surg Res 100: 171–175. 153. Cayco AV, Wysolmerski J, Simpson C, Mitnick MA, Gundberg C, Kliger A, Lorber M, Silver D, Basadonna G,

LUCIANO ADORINI

154.

155.

156.

157.

158.

159.

160.

161.

162.

Friedman A, Insogna K, Cruz D, Bia M 2000 Posttransplant bone disease: Evidence for a high bone resorption state. Transplantation 70:1722–1728. Henderson K, Eisman J, Keogh A, MacDonald P, Glanville A, Spratt P, Sambrook P 2001 Protective effect of shortterm calcitriol or cyclical etidronate on bone loss after cardiac or lung transplantation. J Bone Miner Res 16: 565–571. Dhingra RK, Sprague SM, Ojo AO, Leavey SF 2001 Posttransplant bone disease: a case illustrating dramatic improvements in bone density with vitamin D replacement therapy. Transplantation 71:1856–1859. Stempfle HU, Werner C, Siebert U, Assum T, Wehr U, Rambeck WA, Meiser B, Theisen K, Gartner R 2002 The role of tacrolimus (FK506)-based immunosuppression on bone mineral density and bone turnover after cardiac transplantation: A prospective, longitudinal, randomized, double-blind trial with calcitriol. Transplantation 73: 547–552. Massenkeil G, Fiene C, Rosen O, Michael R, Reisinger W, Arnold R 2001 Loss of bone mass and vitamin D deficiency after hematopoietic stem cell transplantation: Standard prophylactic measures fail to prevent osteoporosis. Leukemia 15:1701–1705. Merida L, Shigetomi M, Ihara K, Tsubone T, Ikeda K, Yamaguchi A, Sugiyama T, Kawai S 2002 Effects of vitamin D analog, 22-oxa-1,25-dihydroxyvitamin D3, on bone reconstruction by vascularized bone allograft. Bone 30:422–427. Cantorna MT, Hullett DA, Redaelli C, Brandt CR, Humpal-Winter J, Sollinger HW, Deluca HF 1998 1,25Dihydroxyvitamin D3 prolongs graft survival without compromising host resistance to infection or bone mineral density. Transplantation 66:828–831. Vulcano M, Struyf S, Scapini P, Cassatella M, Bernasconi S, Bonecchi R, Calleri A, Penna G, Adorini L, Luini W, Mantovani A, Van Damme J, Sozzani S 2003 Unique regulation of CCL18 production by maturing dendritic cells. J Immunol 170:3843–3849. Inaba M, Nishizawa Y, Song K, Tanishita H, Okuno S, Miki T, Morii H 1992 Partial protection of 1alpha-hydroxyvitamin D3 against the development of diabetes induced by multiple low-dose streptozotocin injection in CD-1 mice. Metabolism 41:631–635. Pakkala I, Taskinen E, Pakkala S, Raisanen-Sokolowski A 2001 MC1288, a vitamin D analog, prevents acute graft-versushost disease in rat bone marrow transplantation. Bone Marrow Transplant 27:863–867.

CHAPTER 37

Vitamin D and Osteoblasts JANE E. AUBIN

Department of Molecular and Medical Genetics, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada

JOHAN N. M. HEERSCHE

Faculty of Dentistry, University of Toronto, Toronto, Ontario, Canada

I. Introduction II. Effects of 1,25(OH)2D3 on Osteoblasts In Vitro III. Effects of 1,25(OH)2D3 on Bone Apposition Rates In Vivo

IV. Conclusions References

I. INTRODUCTION

inhibition of others, such as type I collagen [28,29] and bone sialoprotein [30–32]. Similarly, 1,25(OH)2D3 was reported to decrease type I collagen and alkaline phosphatase mRNA and protein levels in rat organ cultures [33] and in rat osteosarcoma cells [34]. Whereas these studies were important for establishing the basis for further analysis, the observations were difficult to reconcile with any straightforward model of osteoblast functional status, although it was suggested that at least some of the variable effects might be dependent on the maturational state of the cells used. Many of the early studies relied on analysis of cell lines whose proliferation–differentiation coupling was aberrant, and few experiments were done under conditions in which a complete differentiation sequence could be followed. Better understanding of some of the apparent contradictions and discrepancies has come from several approaches including the use of primary cultures of osteoblastic cells undergoing a developmental sequence in vitro. With these models, it has become clearer that 1,25(OH)2D3 can alter, in a biphasic manner, osteoblastic cells as they undergo a differentiation sequence from early proliferating cell to mature osteoblast. One model that has been useful is that of rat calvaria cell populations. It has been well established that rat calvaria populations contain osteoprogenitor cells that, under appropriate culture conditions (i.e., addition of ascorbic acid and β-glycerophosphate to the culture medium), can differentiate into mature osteoblasts that are capable of forming a mineralized tissue resembling woven bone [35–37] (Fig. 1). In vitro the cells progress through a developmental sequence of proliferation and differentiation with stage-specific expression of growth- and bone-related genes; the sequence has been well characterized by morphological [38], immunohistochemical [36,39,40,42]; and molecular [37,41–44] approaches. The process has been categorized overall as a period of proliferation (~days 1–7), followed by

1,25-Dihydroxyvitamin D3 [1,25(OH)2D3] affects the proliferation and differentiation of numerous cell lineages/cell types. For example, 1,25(OH)2D3 promotes the differentiation of keratinocytes from epidermal precursors [1–4], monocytes-macrophages from myelopoietic progenitors/stem cells and osteoclasts from mononuclear precursors [5,6], and adipocytes in at least some models [7–9] (see Section II,D). It also induces hypertrophy of chondrocytes [10,11]. The ability of 1,25(OH)2D3 to modulate differentiation pathways as well as its ability to alter cell expression through regulation of transcription or through potential nongenomic pathways have been reviewed [12–15], and many aspects are covered elsewhere in this volume. This chapter focuses on the ability of 1,25(OH)2D3 to modulate osteoblast differentiation and activity in vitro and in vivo.

II. EFFECTS OF 1,25(OH)2D3 ON OSTEOBLASTS IN VITRO A. 1,25(OH)2D3 Modifies the Osteoblast Differentiation Pathway In Vitro and Has Selective Effects Depending on the Relative Differentiation Status of the Cells It has been known for some time that osteoblasts possess specific receptors for 1,25(OH)2D3 [16–19], and several early studies of osteoblastic cells in vitro showed that 1,25(OH)2D3 alters expression of osteoblastassociated genes, for example, increasing alkaline phosphatase activity [20–23] and stimulating expression of some of the noncollagenous proteins of bone such as osteopontin [24–26] and osteocalcin [27]. However, concomitant with stimulation of expression of some osteoblast phenotypic markers, there occurs VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

650

JANE E. AUBIN AND JOHAN N. M. HEERSCHE

Other mesenchymal cells including chondrocytes, myoblasts, fibroblasts

Adipocyte progenitor

?

Preadipocyte

Adipocyte

Effects of 1,25(OH)2D3 on adipocyte development

Postulated osteoblast-adipocyte bipotential progenitor

Lining cell

Apoptotic cell Mesenchymal stem cell Effects of 1,25(OH)2D3 on osteoblast development

Early osteoprogenitor

?

Effects of 1,25(OH)2D3 on osteoblast-associated gene expression

Late osteoprogenitor

Preosteoblast

Inhibition of differentiation Runx2

Osteoblast No formation of bone nodules

Osteocyte

Alkaline phosphatase

Type I collagen

Osteopontin

Bone sialoprotein

Osteocalcin

Tβ4

RANKL

Galectin 3 C3

FIGURE 1 The hormone 1,25(OH)2D3 has various effects on cells at different maturational stages during their differentiation from osteoprogenitors to mature osteoblasts, osteocytes, and lining cells. In rat calvaria populations, osteoprogenitor cells are inhibited from undergoing further differentiation, while some osteoblast-associated genes are up-regulated and others are down-regulated in more mature cells in the lineage. 1,25(OH)2D3 also induces adipocyte development from preadipocytes in rat calvaria cell populations. Tβ4, Thymosin β4; C3, third component of complement; RANKL, receptor activator of NF-κB ligand.

multilayering and early nodule formation (~days 7–11) that is characterized by biosynthesis and organization of type I collagen and noncollagenous proteins of bone, growth of nodules (~days 11–15), and finally osteoid mineralization (> day 15) (reviewed in [37,45,46]). Of particular interest is the fact that multiple factors have been found to influence this proliferation–differentiation process, some with effects at multiple stages and often biphasic in nature [47,48]. Some of the most widely studied agents are the glucocorticoids, which stimulate nodule formation [49,50], increase the proliferative lifetime of the bone-forming precursor [51], and induce primitive osteoprogenitors to differentiate that do not do so in its absence [52]. Although some factors appear to be inhibitory or stimulatory throughout the proliferation–differentiation sequence that leads to formation of a bone nodule, others have biphasic effects that depend on the differentiation stage at which cells are treated (reviewed in [48]). For example, IL-1 stimulates nodule formation, possibly through a mitogenic effect on proliferating progenitors, when it is present early, but inhibits nodule formation when present during later sensitive differentiation time windows [53]. 1,25(OH)2D3 belongs within the category of hormones and factors with biphasic

effects and modifies the normal osteoblast developmental pathway and gene expression profiles associated with it. The ability to manipulate the in vitro bone nodule model during precise windows of time corresponding to proliferative stages, early or late events in osteoblast differentiation, and mineralization stages makes it a useful alternative to in vivo studies or most cell lines. In the rat calvaria cell culture model, when 1,25(OH)2D3 treatment is initiated prior to confluence during the proliferative phase, mRNA levels for alkaline phosphatase and type I collagen are suppressed, bone nodules do not form, and the up-regulation of osteopontin and osteocalcin mRNAs that is seen in the normal differentiation process does not occur [54]. In contrast to the inhibitory effects of 1,25(OH)2D3 on osteoblast differentiation when it is added during the proliferative period, acute or continuous treatment of mature (postproliferative) cultures results in up-regulation of at least some osteoblast-associated genes, such as osteopontin and osteocalcin [54], consistent with other observations showing stimulation of the production of these two proteins by 1,25(OH)2D3 in this and other osteoblastic models [24–27] (see also later discussion). Together, these data suggested that 1,25(OH)2D3 may stimulate mature osteoblasts while inhibiting

CHAPTER 37 Vitamin D and Osteoblasts

early osteoid formation or the differentiation of osteoprogenitors. Further evidence for the latter hypothesis was obtained by treating rat calvaria cell cultures either continuously or acutely with pulse treatment of 1,25(OH)2D3 at particular stages during the differentiation process [55]. Continuous treatment of rat calvaria cells with 1,25(OH)2D3 dose-dependently inhibited bone nodule formation, with half-maximal inhibition at 0.06 nM, while concomitantly stimulating rat calvaria cell growth. Notably, inhibition of bone nodule formation occurred when 1,25(OH)2D3 was present during the proliferation phase and up to the early multilayering stage (e.g., up to day 11) even when it was present for as little as 48 hr [55]. Thus, 1,25(OH)2D3 was inhibitory to osteoprogenitor differentiation during proliferation and the earlier stages of differentiation before visible nodule formation occurred, at a time when cell growth was stimulated, and the effect was not reversible on removal of 1,25(OH)2D3 (Fig. 1). Two other vitamin D3 metabolites, 24,25(OH)2D and 1,24,25(OH)3D3, had dose-dependent inhibitory effects similar to those of 1,25(OH)2D3, with effectiveness in the sequence 1,25(OH)2D3 > 1,24,25(OH)3D3 > 24,25(OH)2D3 [55], consistent with the biological efficacy of these metabolites in other systems where they have been compared (e.g., [56–59]). Do other primary osteoblastic cell models behave similarly to the rat calvaria cell model with respect to 1,25(OH)2D3 effects? Interestingly, although some features are the same, others are not, and these are beginning to lead to a consistent picture as more data accumulate on the relative differentiation status of the majority of cells present in each model. Similar to the rat calvaria culture results, addition of 1,25(OH)2D3 to preconfluent/proliferating chick embryonic osteoblastic cultures over a 30-day period resulted in a 2- to 10-fold reduction in collagen, osteopontin, osteocalcin, and alkaline phosphatase content and mineral deposition [60]. However, in contrast to the rat calvaria system, acute treatments for 24 hr during the proliferative phase (day 5), or when cell proliferation had ceased and cultures had started differentiating, or late during the matrix mineralization phase, caused inhibition at the latter two time points but not during the proliferative period. Broess et al. [60] concluded that the inhibitory effects of 1,25(OH)2D3 on the chick cells reflected effects on relatively mature stages and were independent of effects on proliferation, and that the inhibition was of genes associated with the differentiated cells such that matrix deposition and mineralization were inhibited. The authors hypothesized that the lack of correlation between the chick and rat systems at the mature osteoblast stage might reside in the fact that the cells are from bones

651 of different developmental ages, namely, day 17 fetal chick versus day 21 fetal rat. Because the chick calvaria is more highly mineralized at this developmental age than is the rat calvaria at fetal day 21, the cells from the former might be more mature than the latter. Follow-up experiments confirmed that the effects of 1,25(OH)2D3 are dependent on the developmental stage of the osteoblast lineage, and the stimulatory actions of the hormone are targeted to immature osteoblasts, whereas the inhibitory effects of the hormone are on mature osteoblasts in chick [61]. A number of other studies support the idea that the stage of cellular differentiation of different osteoblastic lines and isolated bone-derived populations affects their responsiveness to 1,25(OH)2D3. Regulation by 1,25(OH)2D3 of Runx2, the “master” transcription factor for osteoblast development, also appears to be dependent on the stage of maturation of human osteoblastic cells [62]. Fraser et al. [63] found that different rat osteosarcoma lines were differentially responsive to 1,25(OH)2D3 in terms of osteocalcin or matrix Gla protein; notably, prolonged 1,25(OH)2D3 treatment of ROS 17/2.8 rat osteosarcoma cells resulted in down-regulation of osteocalcin and up-regulation of matrix Gla [64], supporting the notion that, at some differentiation stages, 1,25(OH)2D3 can alter cell expression to a more mature phenotype. Franceschi and colleagues came to similar conclusions by treating the MG-63 human osteosarcoma line with 1,25(OH)2D3; the 1,25(OH)2D3 treatment promoted partial differentiation of the line [65–67]. Evidence that 1,25(OH)2D3 may “push” relatively mature osteoblasts toward an even more mature state comes from human osteoblastic cells also. Innumerable studies have documented the ability of 1,25(OH)2D3 to stimulate expression of osteogenic markers, most notably osteocalcin, which is often undetectable without 1,25(OH)2D3 stimulation, in cells derived from human trabecular bone [23,68–71] (for reviews, see [72–74]). The accumulating evidence suggests that cultures derived from human trabecular bone fragments may be largely representative of more mature cells (see discussion in Marie [73]), consistent with the ability of 1,25(OH)2D3 to stimulate genes such as osteopontin and osteocalcin in mature cells in rat calvaria cultures. However, it is also clear that, concomitant with stimulation of the expression of these markers, the expression of other markers such as bone sialoprotein is inhibited by 1,25(OH)2D3 in several other cell systems (see earlier discussion). This may be hard to reconcile with the suggestion that 1,25(OH)2D3 acts synergistically with dexamethasone to stimulate expression of all osteogenic markers investigated, including alkaline phosphatase, bone sialoprotein, and osteocalcin, in human marrow-derived stromal cells [75].

652 The observation that sequential addition of 1,25(OH)2D3 enhances the actions of dexamethasone in bone marrow cultures and is required for maximum osteocalcin and osteopontin mRNA expression [39,76], however, supports the view that 1,25(OH)2D3 is up-regulating some markers of the mature osteoblast. Rickard et al. [77] found that dexamethasone and 1,25(OH)2D3 did not have the same effects on all osteoblast markers tested in early or more mature colonies and cultures of human bone marrow stromal-derived cells. In particular, 1,25(OH)2D3 stimulated osteocalcin production in both younger and older cultures, and collagen type I especially in older cultures, but was largely without effect on other markers. Combinations of dexamethasone and 1,25(OH)2D3 led to coexpression of osteoblast and adipocyte markers. Rickard et al. [77] discuss various alternative explanations for the discrepancies, including both species differences and developmental stage differences, but more work will be required to correlate the human and the rodent and avian data in the context of multilineage cells and maturational stage of restricted osteoblast progenitors (see also Section II,D). The apparently relatively mature state of human bone-derived cells and the ability of 1,25(OH)2D3 to induce the cells to a more mature state is also supported by other results. Monoclonal antibodies raised against l,25(OH)2D3-treated human trabecular bone osteoblast-like cells were largely directed against epitopes present in end-stage osteoblasts and osteocytes [78]. 1,25(OH)2D3 used alone or in combination with TGFβ in primary cultures of human osteoblastic cells caused the largely spindle-shaped cells to become stellate [79]. When proliferation and expression of collagen type I, alkaline phosphatase, and osteocalcin were analyzed, the authors concluded that TGFβ stimulates matrix synthesis in these human cells and that 1,25(OH)2D3 may push the cells to an end-stage phenotype (lower proliferation, increased osteocalcin). This is consistent with the ideas proposed for rodent and chick cells and other studies in human bonederived cells in which 1,25(OH)2D3 inhibited proliferation and increased alkaline phosphatase and osteocalcin expression [80]. However, the complexity of the results and interpretation is evident when alkaline phosphatase and osteocalcin were colocalized in another study of l,25(OH)2D3-treated human bone-derived cells. Not all cells that made osteocalcin in response to 1,25(OH)2D3 (9%) synthesized alkaline phosphatase (24%) and vice versa, whereas a proportion produced both (12%). Thavarajah et al. concluded that during differentiation in response to 1,25(OH)2D3 human cells ended up with heterogeneous phenotypes [80]. Heterogeneity of mature osteoblasts is now welldocumented in mice, human and rat bones in vivo by in situ hybridization and immunohistochemistry of

JANE E. AUBIN AND JOHAN N. M. HEERSCHE

numerous matrix molecules, hormone receptors including estrogen and parathyroid hormone/parathyroid hormone related protein (PTHR1) receptors, alkaline phosphatase, and other osteoblast-associated markers [81–87] (reviewed in [88]). Heterogeneity has also been found in rat calvaria osteoblasts in vitro by both immunocytochemistry and polymerase chain reaction (PCR) analyses for all markers of mature osteoblasts including alkaline phosphatase, type I collagen, and most strikingly bone sialoprotein, osteocalcin, and osteopontin [39,42,89] (reviewed in [72,88]). The heterogeneity is also seen before and after 1,25(OH)2D3 treatment in nodules in rat calvaria cultures assessed by in situ hybridization [90,91]. Thus, heterogeneity is not a consequence of 1,25(OH)2D3 treatment; rather, a possibly heterogeneous response to 1,25(OH)2D3 must be taken into account in interpreting 1,25(OH)2D3 effects on bone and osteoblastic cells (for further discussion see [88]). Much more must be done as well to dissect the contributions of species differences versus maturational status of the target cells in various and almost always heterogeneous osteoblastic populations in which diverse effects on such master osteoblast regulators as Runx 2 are seen [92]. The concept of a differentiating cell having different windows of responsiveness to the same physiological mediator underscores the need for rigorous determination of cell behavior throughout the differentiation and maturational sequence (see, e.g., [48]). In situ hybridization and immunocytochemistry can be used to further elucidate whether it is specific subpopulations of cells within cultures and within nodules that respond to hormones [90,91]. As discussed previously, in contrast to the inhibitory effects of 1,25(OH)2D3 on osteoblast differentiation when the hormone is added during the proliferative period, acute or continuous treatment to postproliferative mature osteoblasts causes up-regulation of genes such as osteopontin and osteocalcin [54] (Fig. 1). Altered osteocalcin and osteopontin mRNA levels resulting from regulation by acute exposure to 1,25(OH)2D3 were found by in situ hybridization analysis to be restricted to nodule-associated differentiated cells [90,91]. In analyses done with immunocytochemistry in differentiating bone marrow cell cultures, acute exposure to 1,25(OH)2D3 was also found to up-regulate expression of osteocalcin and in addition galectin 3 (i.e., RCC455.4 positive cells [93]), a member of an important class of lectin adhesion-related molecules and found in high abundance in cells also expressing osteocalcin [39]. However, in this case, the up-regulation was evident also in single cells and small groups of cells acutely treated early in the culture period prior to nodule formation; it was concluded that 1,25(OH)2D3 was up-regulating expression in small groups of cells already mature when released from the bone [39].

CHAPTER 37 Vitamin D and Osteoblasts

Consistent with this interpretation, these soon disappeared in vitro, and new osteocalcin positive cells with the concomitant stimulation by 1,25(OH)2D3 appeared later as nodule formation occurred and as seen in the in situ hybridization studies. 1,25(OH)2D3 pulse treatment of nodule-associated cells was found to be accompanied by shape changes (cells become more elongated, flattened, polarized), and it was in the flat cells that osteocalcin was so markedly up-regulated [90]. On the other hand, as already outlined, expression of type I collagen and bone sialoprotein is inhibited by 1,25(OH)2D3 at these cell maturational stages. These changes in morphology and gene expression may relate to bone-resorbing effects of 1,25(OH)2D3 on bone surface lining osteoblasts [90] (see also Section III). Indeed, osteocalcin expression as determined in bone lining cells is observed by in situ hybridization at a significant level in vivo [84,94]. Both Owen et al. [54] and Broess et al. [60] also found that mineralization of osteoid was inhibited by 1,25(OH)2D3 when it was added after deposition of the collagenous matrix, suggesting that the hormone can affect the ability of osteoblasts to mineralize their matrix, perhaps independently of an effect on matrix formation. The effects of 1,25(OH)2D3 on osteoblastic cells indirectly through nonadherent cells in the bone marrow must also be considered. 1,25(OH)2D3 has been found to stimulate stromal cell proliferation and alkaline phosphatase activity in rat bone marrow cultures in a manner dependent on the nonadherent fraction of cells [95]. Although the authors believed they were stimulating the osteoblastic subpopulation of stromal cells, given that it appears that only a proportion of alkaline phosphatase positive colonies are osteoblastic at least under a given set of conditions [96], caution must be used in concluding that the results are specific for osteoblastic cells. In another series of experiments, Long and colleagues have been isolating osteoblast precursors and osteoprogenitors from the nonadherent fraction of human bone marrow. Intriguingly, some nonadherent cells give rise to colonies with cells that express osteocalcin, osteonectin, and bone alkaline phosphatase and respond to 1,25(OH)2D3 [97,98]. Clearly, the widespread presence of the vitamin D receptor (VDR) and its multiple activities in many cell types suggests that many effects of the hormone to alter osteoblast activity may also be mediated via other cell types.

B. 1,25(OH)2D3 Regulates Genes Associated with Osteoblast Proliferation and Differentiation As discussed earlier, 1,25(OH)2D3 up-regulates or down-regulates a wide variety of genes associated with

653 the osteoblast phenotype including Runx2, a master regulator of osteoblast development [99,100]. What are the molecular mediators of these complex and sometimes biphasic biological effects? As discussed elsewhere in this volume, 1,25(OH)2D3 activates multiple signaling pathways in osteoblasts, inducing rapid nongenomic and long-term genomic pathways. Nongenomic pathways involve lipid turnover, activation of Ca2+ channels, and elevation of intracellular Ca2+, all of which occur within seconds of administration of the steroid in vitro. Genomic pathways are mediated by the VDR, a member of the steroid receptor superfamily, and involve transcriptional regulation of target genes. Thus, in at least some cases, 1,25(OH)2D3 effects may be relatively direct on the osteoblast-specific gene. For example, it is known that 1,25(OH)2D3 following its association with the VDR regulates osteocalcin via transcriptional mechanisms; together the hormone and receptor form complexes with other transcription factors, often the retinoid X receptor (RXR), that interact in a sequence-specific manner with promoter regulatory elements (reviewed in [99,101–103]) (see Chapters 13–19). Some other osteoblast regulatory factors may be effective through their ability to regulate VDR levels or activity, for example, tumor necrosis factor-α (TNFα) [104], glucocorticoids [105], and TGF-β [106,107]. However, it is becoming clear that many other transcription factors vary as a function of the differentiation status of osteoblasts, such that changing levels of these may also play a role in the ability of 1,25(OH)2D3 to regulate responsive genes (reviewed in [100,108]). This appears to be true, for example, of Fos-Jun family members, which vary as a function of the rat calvaria cell proliferation–differentiation cycle [109,110]. Interestingly, 1,25(OH)2D3 has been shown to have differential effects on different members of the Fos-Jun family and in some cases to act via regulation of initiation and elongation of transcription (e.g., c-fos) versus at a posttranscriptional level distinct from mRNA stabilization (e.g., c-jun, jun-B) in MC3T3-E1 mouse osteoblast cells [111]. It may also be that one needs to consider another regulatory level, namely, the ability of 1,25(OH)2D3 to modulate binding of other nonreceptor transcription factors to other regulatory sequences; for example, 1,25(OH)2D3 up-regulates the homeodomain protein Msx-2 [112], which is known to be important in skeletal development [113–115], and down-regulates osteocalcin [116–118]. In addition to considering the ability of 1,25(OH)2D3 to regulate genes such as osteocalcin that are associated with differentiated osteoblasts, it will be important to determine how 1,25(OH)2D3 alters the proliferation of precursors and what role this plays in the ability of 1,25(OH)2D3 to block or alter the

654 differentiation status of osteoblastic cells. Observations in other cell types may be enlightening. It has been known for many years that 1,25(OH)2D3 induces myeloid cell lines to differentiate into monocytes– macrophages [5,6]. To isolate target genes of the VDR that initiate the differentiation process, Freedman and colleagues used probes prepared from 1,25(OH)2D3treated or untreated cells to survey a cDNA library from the myelomonocytic U937 cell line [119]. One clone that differentially hybridized is the cyclindependent kinase (Cdk) inhibitor p21WAF1,CIP1, a protein that inhibits the cell cycle by associating with cyclin-Cdk complexes and blocking their function. The p21 mRNA appeared very early (2 hr) after treatment of cells with 1,25(OH)2D3, and the p21 promoter contains a vitamin D response element (VDRE). In studies to confirm that p21 was functionally involved in the 1,25(OH)2D3 regulation of U937 cell differentiation, p21 was transiently overexpressed in these cells, and several monocyte–macrophage markers were up-regulated in the absence of 1,25(OH)2D3. The related Cdk inhibitor p27 had similar activity, suggesting that alterations in cyclin–Cdk complexes induced by 1,25(OH)2D3 can induce the terminal differentiation of this cell type [119]. The widespread presence of VDREs in a number of promoters studied to date means that many genes may be modulated directly by 1,25(OH)2D3 in osteoblastic cells, as the examples already given indicate. It is striking that several genes, only recently identified as osteoblast products and whose functions are not yet fully elucidated in the osteoblast, are also regulated by 1,25(OH)2D3. As mentioned earlier, galectin 3 (by mRNA levels or as immunolabeled with the monoclonal antibody RCC455.4 [93]) is coexpressed at high levels in cells expressing osteocalcin, and both proteins are stimulated by 1,25(OH)2D3 [39]. Importantly, the regulation of galectin 3 expression appears different in osteoblastic cells compared to rat skin fibroblasts, suggesting cell type– specific regulation by hormones such as 1,25(OH)2D3. For example, 1,25(OH)2D3 had no significant effect on and dexamethasone enhanced galectin 3 expression in skin fibroblast cultures, whereas dexamethasone downregulated and 1,25(OH)2D3 up-regulated galectin 3 expression in rat calvaria and ROS17/2.8 cells. This was especially clear at later time points in culture, that is, at times corresponding to late stages in the osteogenic differentiation sequence [93]. Other unexpected molecules have been found to be products of osteoblastic cells and to be regulated by 1,25(OH)2D3. For example, thymosin β4 (Tβ4), a 43-residue peptide member of a family of closely related peptides with unknown function but with actin-binding capacity, was found in a differential hybridization

JANE E. AUBIN AND JOHAN N. M. HEERSCHE

screen of osteoblastic cells and is expressed in ROS 17/2.8 and UMR-106 cells and in neonatal and fetal calvaria of rat. Tβ4 mRNA levels have been reported to be 8- to 10-fold higher in well-differentiated ROS 17/2.8 cells compared to the less differentiated ROS 25/1 line, and 1,25(OH)2D3 down-regulates Tβ4 mRNA expression [120]. In response to 1,25(OH)2D3, the third component of complement (C3) is produced by both the stromal cell line ST2 and primary mouse osteoblastic cells. 1,25(OH)2D3 also stimulates C3 expression in vivo, and C3 has been immunolocalized to periosteal and sutural cells in calvaria and in the tibial metaphyses [121]. An interesting series of experiments suggested that C3 produced by stromal cells in response to 1,25(OH)2D3 may be involved in the differentiation of osteoclasts from their precursors [122]. Of course, the now well-established regulation of expression of RANKL by 1,25(OH)2D3 in stromal/ osteoblastic cells and its functional role in osteoclast formation and activity has become a classic paradigm of 1,25(OH)2D3-regulation of bone resorption ([123]; reviewed in [6,124,125] (see chapter 38).

C. 1,25(OH)2D3 Modifies Osteoblast Responsiveness to Other Hormones and Growth Factors That Influence Osteoblast Differentiation and Activity Among genes regulated by 1,25(OH)2D3 are those for other hormones, growth factors, or their receptors, suggesting that 1,25(OH)2D3 may act in part by modifying responsiveness to other regulators of osteoblast differentiation and activity. Although it is beyond the scope of this chapter to discuss all of the hormone and cytokine networks with which 1,25(OH)2D3 may interact, some examples may be interesting and informative. In UMR106-01 cells, 1,25(OH)2D3 inversely modulates PTH-induced regulator of G protein signaling (RGS)-2, a putative preferential inhibitor of G(q)-mediated phospholipase C activation. Such regulation of RGS-2 may constitute a novel mechanism by which 1,25(OH)2D3 modulates signaling via the PTHR1 and other G protein– coupled receptors in bone [126]. In MC3T3-E1 cells, which express large numbers of IL-4 receptors, 1,25(OH)2D3 augments IL-4 binding, increases receptor abundance (Bmax), and increases receptor mRNA levels, all of which could enhance IL-4 effects, which include increasing cell proliferation and inhibiting alkaline phosphatase activity in these cells [127]. MC3T3-E1 cells also express IL-1 receptor type 1 (by both mRNA and binding studies) and respond to IL-1 through the IL-1 type I receptor to induce IL-6. Accordingly, the ability of 1,25(OH)2D3 to increase the

655

CHAPTER 37 Vitamin D and Osteoblasts

type I receptor in MC3T3-E1 cells may modulate the effect of IL-1 to stimulate IL-6 production [128]. These and other data suggest that 1,25(OH)2D3 could be responsible, in part, for modulating IL-1 effects on the skeleton [129,130]. In other examples, 1,25(OH)2D3 has been shown to increase expression of nerve growth factor (NGF) mRNA in ROS 17/2.8 cell receptor [131] which bind NGF to the apparently low-affinity NGF; interestingly, the regulation is mediated by a 1,25(OH)2D3-induced increase in AP-1 activity and an AP-1 site in the NGF promoter [132]. 1,25(OH)2D3 has been found either to down-regulate [133] or up-regulate [134] specific cell surface endothelin receptors on osteoblastic cells. The opioid gene proenkephalin (PENK), which has been identified in osteoblastic cells and found to be down-regulated as osteoblasts mature in vitro and in vivo, is regulated by a variety of osteotropic hormones including 1,25(OH)2D3 [135]. In human bone marrow stromal cell cultures, 1,25(OH)2D3 appears to potentiate fluoride-mediated anabolic effects such as increased type I collagen production, alkaline phosphatase, and osteocalcin [136]. In the same model, 1,25(OH)2D3 acts synergistically with TGF-β1 in stimulating insulin-like growth factor binding protein 3, which may mediate, in part, the effects of 1,25(OH)2D3 on proliferation and differentiation in these cells [137]. A related issue that should also be mentioned with regard to the effects of 1,25(OH)2D3 on the development and function of skeletal cells in vivo and in vitro is the possible interdependence of the effects of 1,25(OH)2D3, retinoids, and thyroid hormones as a consequence of the heterodimerization capabilities of their receptors with a common member of the family, the retinoid X receptor. Williams et al. [138,139] demonstrated, using a variety of osteoblastic cell lines representing different stages of osteoblastic development, that retinoids were required for expression of responses to 1,25(OH)2D3 and triiodothyronine (T3). The observations by us and others in the rat calvaria bone noduleforming system that retinoids inhibit osteoprogenitor proliferation and differentiation [140] and that T3 has variable effects, depending on the stage of differentiation of the target cells [141,142], suggest that receptor interactions do occur in this and other systems, although a variety of other explanations are possible. 1,25(OH)2D3 is synergistic or antagonistic with retinoic acid with respect to levels of alkaline phosphatase, osteopontin, matrix Gla protein, and type I collagen in the immortalized rat preosteoblast line UMR-201-10B [143]. Evidence that TGFβ interacts with 1,25(OH)2D3 and that the effects are either synergistic (with regard to alkaline phosphatase production [79] and insulin-like growth factor binding protein 3 [137]) or antagonistic (with regard to osteocalcin production [79]) further

illustrate the importance of multiple interactions in studies evaluating the effects of the various vitamin D metabolites.

D. Does 1,25(OH)2D3 Alter Commitment of Progenitor Cells Capable of Forming Fat and Bone? As mentioned in the introduction, 1,25(OH)2D3 is a known regulator of the differentiation and activity of cells of various lineages, including mesenchymal and hemopoietic cells. In some systems, 1,25(OH)2D3 appears to act in synergy with other steroids, including glucocorticoids, to regulate osteoblast differentiation. While investigating the inhibitory effects of vitamin D3 metabolites on bone nodule formation in primary fetal rat calvaria cells [55] discussed earlier, we noted an increased number of adipocyte foci in cultures that had been grown in the presence of 1,25(OH)2D3. Dexamethasone also increases adipocyte number in rat calvaria cultures [7], consistent with other observations that glucocorticoids stimulate the preadipocytes derived from various sources to differentiate into mature adipocytes [144–147]. The reported effects of 1,25(OH)2D3 on adipogenesis are more discrepant. Sato and Hiragun [148] reported that in the mouse preadipocyte cell lines 3T3L1 and ST13, 1,25(OH)2D3 receptors were present in both cell lines at the preadipocyte cell stage, but not in mature adipocytes, and that 1,25(OH)2D3 suppressed the differentiation of preadipocytes into adipocytes. In contrast, Ishida et al. [149] found that 1,25(OH)2D3 decreased cell proliferation and [3H]thymidine uptake in a dose-dependent manner in 3T3L1 cells and increased the number of lipid droplets in the cytoplasm without a significant stimulation in glyceraldehyde-3-phosphate dehydrogenase (G3PDH) activity. This study was in agreement with studies by Vu et al. [8] in the same cell line. It has also been reported, however, that 1,25(OH)2D3 at lower concentration stimulated, but at higher concentration inhibited adipogenesis in 3T3-L1 cells [150,151]. Given the ability of 1,25(OH)2D3 to inhibit adipogenesis when it is added together with dexamethasone in murine calvariae or a murine bone marrow-derived adipocytic clone or primary murine stromal cells [148,152–154], the stimulatory effects observed when rat calvaria cultures are treated with both 1,25(OH)2D3 and dexamethasone are puzzling [7]. Understanding this and other discrepancies noted above will be important. Some, such as differences observed in mouse versus rat calvaria, may be species related, given the observation that concentrations of glucocortcoid that are stimulatory for osteogenesis in rat bone marrow cultures

656 may be inhibitory in mouse marrow cultures [155], that osteogenesis in mouse marrow cultures may not be glucocorticoid dependent [155], and that mouse calvaria cells show a different dose–response profile to glucocorticoids as compared to rat calvaria cells for osteogenic stimulation [156]. Others may reflect differences in primary cultures versus established cell lines, or the presence and types of accessory cells present in heterogeneous primary cultures of calvaria or marrow stroma. Nevertheless, the data on the effects of 1,25(OH)2D3 and dexamethasone on adipocyte formation together with the data showing l,25(OH)2D3-induced inhibition of osteoblastic differentiation in rat calvaria cultures may have relevance to the bone and bone marrow changes that are observed during aging, prolonged immobilzation, glucocorticoid-induced osteoporosis, and postmenopausal osteoporosis in humans, namely, a reduction in bone mass appearing with a concomitant increase in marrow adipose tissue [157–162]. The correlation between bone loss and marrow fat gain is also seen in animal models as well, for example, after, chronic glucocorticoid treatment, low gravity, immobilization, and in the senescence-accelerated mouse model (SAMP6), wherein osteoblastogenesis is decreased and adipogenesis is increased [161,163–165]. These and other observations have led to the hypothesis that 1,25(OH)2D3 may regulate cell fate selection of a mesenchymal stem cell or the commitment and/or differentiation of a bipotential osteoblast–adipocyte progenitor such that fat formation can be induced at the expense of bone formation (Fig. 1) (for reviews, see [37,46,166,167]). In keeping with such a possibility, forced expression of the master adipocyte gene peroxisome proliferator–activated receptor γ2 (PPARγ2) transdifferentiates osteoblastic cells to adipocytes [147], possibility via its ability to down-regulate Runx2 [168]. The difficulty is, however, that inverse relationships between the two phenotypes may result from independent regulatory effects on already committed preadipocytes and committed osteoprogenitors, and most studies have not addressed this possibility rigorously (see commentary [169]). Consistent with this, independent osteoblast and adipocyte lineage cell autonomous effects account for the osteosclerosis concomitant with diminished adipogenesis, for example, in the deltaFosB transgenic mouse model [170]. A replica plating approach was also used recently in the rat calvaria model to show that the frequency of common bipotential progenitors appears to be very low and that the 1,25(OH)2D3-responsive adipocyte progenitors are different from the dexamethasone-responsive adipocyte progenitors, but both are restricted to form adipocytes and not osteoblasts, i.e., the increased adipogenesis seen

JANE E. AUBIN AND JOHAN N. M. HEERSCHE

with 1,25(OH)2D3 and dexamethasone appears not to be a cell fate switch of osteoprogenitors toward adipogenesis [171]. Nevertheless, given the complexities and lack of molecular understanding of the stem and progenitor cell populations in different anatomic sites, and the interactions among 1,25(OH)2D3, other factors, and their respective receptors, it remains of interest to compare the effects of 1,25(OH)2D3 on osteoblast–adipocyte lineage relationships in different precursor cell pools (e.g., the marrow stromal cell versus calvaria- or other bone-derived cell pools) and in different animal models (for review, see [37,46]).

III. EFFECTS OF 1,25(OH)2D3 ON BONE APPOSITION RATES IN VIVO It is clear from the observations in vitro that osteoprogenitors, particularly early osteoprogenitors, are target cells for 1,25(OH)2D3 and 24,25(OH)2D3 action and that the effects of these metabolites on early osteoprogenitor proliferation and differentiation are inhibitory. It is equally clear that another group of target cells for these hormones are mature, functional osteoblasts, and that the effects of 1,25(OH)2D3 on these are stimulatory with regard to osteocalcin and osteopontin secretion but inhibitory with respect to collagen and bone sialoprotein production. The in vivo consequence of this could be that l,25(OH)2D3mediated increased deposition of osteopontin and osteocalcin by secretory osteoblasts approaching the end of their secretory lifetime may render the bone surface in those locations resorbable by osteoclasts, as osteoclasts require the bone matrix to be mineralized in order to be able to resorb it and simultaneously push the osteoblastic cells to differentiate into lining cells (see Section II,A). l,25(OH)2D3-induced inhibition of bone sialoprotein and collagen secretion by early secretory phase osteoblasts could reflect inhibition of matrix deposition by these cells, revert the phenotype to a presecretory cell, and result in the disappearance of osteoblasts from the trabecular bone surface observed in vivo in l,25(OH)2D3-repleted vitamin D– deficient rats [172] (see later discussion). Elucidating the effects of vitamin D metabolites on bone apposition in vivo has been of interest for many years. A series of investigations carried out for the purpose of clarifying the functions of vitamin D metabolites on the linear rate of bone mineral apposition (BMAR) in adult vitamin D–deficient rats [173] suggested that, in the absence of 24,25(OH)2D3, 1,25(OH)2D3 may inhibit mineralization of osteoid. This interpretation was compatible with the observations of Hock et al. [174]

657

CHAPTER 37 Vitamin D and Osteoblasts

that pharmacological concentrations of 1,25(OH)2D3 administered to intact rats increased the amount of osteoid and inhibited mineralization. Repletion of these rats with 200 ng/day of either 25(OH)D3 or 24,25(OH)2D3 for the last 15 days of the 8-week period resulted in a rapid restoration of the BMAR, but treatment with 200 ng/day of 1,25(OH)2D3 had no effect. l,25(OH)2D3-repleted animals exhibited increased osteoclastic activity and decreased numbers of osteoblasts on the trabecular bone surfaces, whereas the 24,25(OH)2D3-repleted animals maintained osteoblastic activity on most of the trabecular bone surfaces without increased osteoclastic activity (C. S. Tarn and J. N. M. Heersche, unpublished). This observation is similar to that of Ono et al. [175], who found that 24,25(OH)2D3 normalized bone formation and resorption in rachitic hypophosphatemic mice, whereas 1,25(OH)2D3 normalized bone formation but also induced excessive stimulation of bone resorption (see also Chapter 38). The fact that all three metabolites of vitamin D tested were capable of reducing the mean osteoid seam width within 15 days of their administration in the vitamin D–restricted rats, whereas only 25OHD3 and 24,25(OH)2D3 stimulated BMAR, points out the complexity of the process of bone formation. Interestingly, the observed effects of 24,25(OH)2D3 and 25OHD3 on the mineralization lag time are in keeping with their effects in stimulating BMAR. However, 1,25(OH)2D3 repletion reduced the lag time but did not affect the BMAR. It is possible that these differences can be accounted for by differences in the effects of these metabolites on organic matrix synthesis. Unfortunately, a systematic evaluation of the effects of vitamin D and its metabolites on organic matrix apposition has not been published to date. Over the past several years, however, use of mouse genetic models has led to an explosion of new information on and challenged some concepts of the physiological functions of vitamin D and its derivatives on osteoblasts and bone (see Chapters 7, 20). For example, in two different VDR-null mouse models [176,177], growth retardation is seen accompanied by progressive hypocalcemia, hypophosphatemia, and compensatory hyperparathyroidism with concomitant severe skeletal defects, including decreased bone mineral density, thinned bone cortex, and widened undermineralized growth plates. Strikingly, however, feeding the mice a rescue diet rich in calcium and phosphorus to normalize serum calcium and PTH levels rescues the bone abnormalities [178]. The data suggest lack of a direct effect of 1,25(OH)2D3 on bone and draw into question the physiological relevance of the vast amounts of data summarized earlier on the apparent direct role of the VDR in osteoblasts.

In a different approach to the question of direct effects of vitamin D on bone, two groups independently also ablated the gene coding the enzyme responsible for the regulated synthesis of 1,25(OH)2D3, 1αOHase [179,180]. The phenotype of the 1αOHase-null mice is similar to that of the VDR knockout mouse, including hypocalcemia, hyperparathyroidism, growth retardation, and osteomalacia, consistent with rickets [179,180]. Treatment of the 1αOHase mutants with 1,25(OH)2D3 rescued the rickets and osteomalacia [181]. The rescue regimen also restored the biomechanical properties of the bone tissue within normal parameters. However, correction of the abnormal mineral homeostasis by feeding with a high-calcium/phosphorus/lactose diet rescued the rickets and osteomalacia but not bone growth; the latter was thought to be due to the inability of passive calcium uptake to meet the high demand for minerals during the period of rapid growth that follows weaning [182]. Finally, a knockout model of 24OHase [183], which metabolizes both the bioactive 1,25(OH)2D3 and its precursor, 25(OH)D3, has allowed the putative role of one of its major metabolites, 24,25(OH)2D3, to be addressed. Loss of 24OHase leads to aberrant intramembranous ossification, which apparently results not from loss of 24,25(OH)2D3 itself but the toxicity of the high 1,25(OH)2D3 levels in the mice. This model also suggests that 24-hydroxylated metabolites are not required for normal intramembranous ossification [183].

IV. CONCLUSIONS The major challenge in the area of regulation of osteoblast differentiation and activity by vitamin D metabolites remains to place findings made in the in vitro models into an in vivo context. Obviously, the new genetically modified mouse models have opened almost as many new questions and issues as have been answered related to the role of vitamin D, its metabolites, the classic VDR, and a putative membrane receptor in osteoblasts and bone. Given the possibility that certain compensatory mechanisms may come into play during development of the various mouse models, more remains to be done with developmentaltime and tissue-specific conditional VDR knockouts in which the skeleton can be assessed in vivo and the genetically altered cells can be probed in ex vivo cultures. The complex interactions between, and interdependence of, the actions of the different metabolites with other endocrine and local regulatory systems discussed in some detail earlier add to the complexity of the issues and will make them more difficult to resolve, but increasingly sophisticated genetic

658 manipulations are also coming into play. Together these approaches should lead to an unambiguous understanding of the direct actions of vitamin D on osteoblasts and bone.

Acknowledgments This work was supported by the Canadian Institutes of Health Research (CIHR MT-12390 to J.E.A. and MT-14655 to J.N.M.H.) and the Stem Cell Network of Centres of Excellence (to J.E.A.).

References 1. Studzinski GP, McLane JA, Uskokovic MR 1993 Signaling pathways for vitamin D-induced differentiation: implications for therapy of proliferative and neoplastic diseases. Crit Rev Eukaryot Gene Expr 3:279–312. 2. Bikle DD, Pillai S 1993 Vitamin D, calcium, and epidermal differentiation. Endocr Rev 14:3–19. 3. Eckert RL, Crish JF, Robinson NA 1997 The epidermal keratinocyte as a model for the study of gene regulation and cell differentiation. Physiol Rev 77:397–424. 4. Bikle DD 2004 Vitamin D regulated keratinocyte differentiation. J Cell Biochem 92:436–444. 5. Suda T, Kobayashi K, Jimi E, Udagawa N, Takahashi N 2001 The molecular basis of osteoclast differentiation and activation. Novartis Found Symp 232:235–247. 6. Suda T, Ueno Y, Fujii K, Shinki T 2003 Vitamin D and bone. J Cell Biochem 88:259–266. 7. Bellows CG, Wang Y-H, Heersche JNM, Aubin JE 1994 1,25 Dihydroxyvitamin D3 stimulates adipocyte differentiation in cultures of fetal rat calvarial cells: Comparison with the effects of dexamethasone. Endocrinol 134:2221–2229. 8. Vu D, Ong JM, Clemens TL, Kern PA 1996 1,25-Dihydroxyvitamin D induces lipoprotein lipase expression in 3T3-L1 cells in association with adipocyte differentiation. Endocrinology 137:1540–1504. 9. Atmani H, Chappard D, Basle MF 2003 Proliferation and differentiation of osteoblasts and adipocytes in rat bone marrow stromal cell cultures: effects of dexamethasone and calcitriol. J Cell Biochem 89:364–372. 10. Gerstenfeld LC, Kelly CM, Von Deck M, Lian JB 1990 Effect of 1,25-dihydroxyvitamin D3 on induction of chondrocyte maturation in culture: extracellular matrix gene expression and morphology. Endocrinology 126:1599–1609. 11. Schwartz Z, Brooks B, Swain L, Del Toro F, Norman A, Boyan B 1992 Production of 1,25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3 by growth zone and resting zone chondrocytes is dependent on cell maturation and is regulated by hormones and growth factors. Endocrinology 130:2495–2504. 12. Lin R, White JH 2004 The pleiotropic actions of vitamin D. Bioessays 26:21–28. 13. Banerjee P, Chatterjee M 2003 Antiproliferative role of vitamin D and its analogs—a brief overview. Mol Cell Biochem 253:247–254. 14. Farach-Carson MC, Nemere I 2003 Membrane receptors for vitamin D steroid hormones: Potential new drug targets. Curr Drug Targets 4:67–76.

JANE E. AUBIN AND JOHAN N. M. HEERSCHE

15. Boyan BD, Sylvia VL, Dean DD, Del Toro F, Schwartz Z 2002 Differential regulation of growth plate chondrocytes by 1α,25-(OH)2D3 and 24R,25-(OH)2D3 involves cell-maturation-specific membrane-receptor-activated phospholipid metabolism. Crit Rev Oral Biol Med 13:143–154. 16. Manolagas SC, Haussler MR, Deftos LJ 1980 1,25-Dihydroxyvitamin D3 receptor-like macromolecule in rat osteogenic sarcoma cell lines. J Biol Chem 255:4414–4417. 17. Chen TL, Feldman D 1981 Regulation of 1,25-dihydroxyvitamin D3 receptors in cultured mouse bone cells. Correlation of receptor concentration with the rate of cell division. J Biol Chem 256:5561–5566. 18. Narbaitz R, Stumpf WE, Sar M, Huang S, DeLuca HF 1983 Autoradiographic localization of target cells for 1 alpha, 25-dihydroxyvitamin D3 in bones from fetal rats. Calcif Tissue Int 35:177–182. 19. Petkovich PM, Heersche JN, Aubin JE, Grigoriadis AE, Jones G 1987 Retinoic acid-induced changes in 1 alpha, 25-dihydroxyvitamin D3 receptor levels in tumor and nontumor cells derived from rat bone. J Natl Cancer Inst 78:265–270. 20. Majeska RJ, Rodan GA 1982 The effect of 1,25(OH)2D3 on alkaline phosphatase in osteoblastic osteosarcoma cells. J Biol Chem 257:3362–3365. 21. Manolagas SC, Burton DW, Deftos LJ 1981 1,25-Dihydroxyvitamin D3 stimulates the alkaline phosphatase activity of osteoblast-like cells. J Biol Chem 256:7115–7117. 22. Kurihara N, Ikeda K, Hakeda Y, Tsunoi M, Maeda N, Kumegawa M 1984 Effect of 1,25-dihydroxyvitamin D3 on alkaline phosphatase activity and collagen synthesis in osteoblastic cells, clone MC3T3-E1. Biochem Biophys Res Commun 119:767–771. 23. Beresford JN, Gallagher JA, Russell RG 1986 1,25Dihydroxyvitamin D3 and human bone-derived cells in vitro: effects on alkaline phosphatase, type I collagen and proliferation. Endocrinology 119:1776–1785. 24. Prince CW, Butler WT 1987 1,25-Dihydroxyvitamin D3 regulates the biosynthesis of osteopontin, a bone-derived cell attachment protein, in clonal osteoblast-like osteosarcoma cells. Coll Relat Res 7:305–313. 25. Chang PL, Prince CW 1991 1α,25-Dihydroxyvitamin D3 stimulates synthesis and secretion of nonphosphorylated osteopontin (secreted phosphoprotein 1) in mouse JB6 epidermal cells. Cancer Res 51:2144–2150. 26. Sodek J, Chen J, Nagata T, Kasugai S, Todescan R, Li J, Kim RH 1995 Regulation of osteopontin expression in osteoblasts. Ann NY Acad Sci 760:223–241. 27. Price PA, Baukol SA 1980 1,25-Dihydroxyvitamin D3 increases synthesis of the vitamin K–dependent bone protein by osteosarcoma cells. J Biol Chem 255:11660–11663. 28. Raisz LG, Maina DM, Gworek SC, Dietrich JW, Canalis EM 1978 Hormonal control of bone collagen synthesis in vitro: Inhibitory effect of 1-hydroxylated vitamin D metabolites. Endocrinology 102:731–735. 29. Kream BE, Rowe D, Smith MD, Maher V, Majeska R 1986 Hormonal regulation of collagen synthesis in a clonal rat osteosarcoma cell line. Endocrinology 119:1922–1928. 30. Oldberg A, Jirskog-Hed B, Axelsson S, Heinegard D 1989 Regulation of bone sialoprotein mRNA by steroid hormones. J Cell Biol 109:3183–3186. 31. Sodek J, Kim RH, Ogata Y, Li J, Yamauchi M, Zhang Q, Freedman LP 1995 Regulation of bone sialoprotein gene transcription by steroid hormones. Connect Tissue Res 32:209–217. 32. van Leeuwen JP, van Driel M, van den Bemd GJ, Pols HA 2001 Vitamin D control of osteoblast function and bone extracellular matrix mineralization. Crit Rev Eukaryot Gene Expr 11:199–226.

CHAPTER 37 Vitamin D and Osteoblasts

33. Canalis E, Lian JB 1985 1,25-Dihydroxyvitamin D3 effects on collagen and DNA synthesis in periosteum and periosteum-free calvaria. Bone 6:457–460. 34. Harrison JR, Petersen DN, Lichtler AC, Mador AT, Rowe DW, Kream BE 1989 1,25-Dihydroxyvitamin D3 inhibits transcription of type I collagen genes in the rat osteosarcoma cell line ROS 17/2.8. Endocrinology 125:327–333. 35. Nefussi JR, Boy-Lefevre ML, Boulekbache H, Forest N 1985 Mineralization in vitro of matrix formed by osteoblasts isolated by collagenase digest. Differentiation 29:160–168. 36. Bellows CG, Aubin JE, Heersche JNM, Antosz ME 1986 Mineralized bone nodules formed in vitro from enzymatically released rat calvaria cell populations. Calcif Tissue Int 38:143–154. 37. Aubin JE, Triffitt J 2002 Mesenchymal stem cells and the osteoblast lineage. In: JP Bilezikian, LG Raisz, GA Rodan (eds) Principles of Bone Biology, 2nd ed. Academic Press, New York, pp. 59–81. 38. Bhargava U, Bar-Lev M, Bellows CG, Aubin JE 1988 Ultrastructural analysis of bone nodules formed in vitro by isolated fetal rat calvaria cells. Bone 9:155–163. 39. Malaval L, Modrowski D, Gupta AK, Modrowski D, Gupta AK, Aubin JE 1994 Cellular expression of bone-related proteins during in vitro osteogenesis in rat bone marrow stromal cell cultures. J Cell Physiol 158:555–572. 40. Malaval L, Liu F, Roche P, Aubin JE 1999 Kinetics of osteoprogenitor proliferation and osteoblast differentiation in vitro. J Cell Biochem 74:616–627. 41. Owen TA, Aronow M, Shalhoub V, Barone LM, Wilming L, Tassinari MS, Kennedy MB, Pockwinse S, Lian JB, Stein GS 1990 Progressive development of the rat osteoblast phenotype in vitro: reciprocal relationships in expression of genes associated with osteoblast proliferation and differentiation during formation of the bone extracellular matrix. J Cell Physiol 143:420–430. 42. Liu F, Malaval L, Gupta A, Aubin JE 1994 Simultaneous detection of multiple bone-related mRNAs and protein expression during osteoblast differentiation: Polymerase chain reaction and immunocytochemical studies at the single cell level. Dev Biol 166:220–234. 43. Liu F, Malaval L, Aubin JE 2003 Global amplification polymerase chain reaction reveals novel transitional stages during osteoprogenitor differentiation. J Cell Sci 116: 1787–1796. 44. Lian JB, Javed A, Zaidi SK, Lengner C, Montecino M, van Wijnen AJ, Stein JL, Stein GS 2004 Regulatory controls for osteoblast growth and differentiation: role of Runx/Cbfa/ AML factors. Crit Rev Eukaryot Gene Expr 14:1–41. 45. Stein GS, Lian JB, Stein JL, van Wijnen AJ, Frenkel B, Montecino M (eds) 1996 Mechanisms Regulating Osteoblast Proliferation and Differentiation. Academic Press, San Diego. 46. Aubin JE, Heersche JNM 2002 Bone cell biology: osteoblasts, osteocytes and osteoclasts. In: FH Glorieux, JM Pettifor, H Jueppner (eds) Pediatric Bone. Academic Press, San Diego, pp. 43–75. 47. Rodan GA 1998 Control of bone formation and resorption: biological and clinical perspective. 25th anniversary issue: New directions and dimensions in cellular biochemistry. J Cell Biochem Suppl 30/31:55–61. 48. Aubin JE 2001 Regulation of osteoblast differentiation and function. Rev Endoc and Metabol Disorders 2:81–94. 49. Bellows CG, Aubin JE, Heersche JNM 1987 Physiological concentrations of glucocorticoids stimulate formation of bone nodules from isolated rat calvaria cells in vitro. Endocrinology 121:1985–1992.

659 50. Bellows CG, Aubin JE 1989 Determination of numbers of osteoprogenitors present in isolated fetal rat calvaria cells in vitro. Dev Biol 133:8–13. 51. Bellows CG, Heersche JNM, Aubin JE 1990 Determination of the capacity for proliferation and differentiation of osteoprogenitor cells in the presence and absence of dexamethasone. Dev Biol 140:132–138. 52. Turksen K, Aubin JE 1991 Positive and negative immunoselection for enrichment of two classes of osteoprogenitor cells. J Cell Biol 114:373–384. 53. Ellies LG, Aubin JE 1990 Temporal sequence of interleukin 1α-mediated stimulation and inhibition of bone formation by isolated fetal rat calvarial cells in vitro. Cytokine 2:430–437. 54. Owen TA, Aronow MA, Barone LM, Bettencourt B, Stein G, Lian JB 1991 Pleiotropic effects of vitamin D on osteoblast gene expression are related to the proliferative and differentiated state of the bone cell phenotype: Dependency upon basal levels of gene expression, duration of exposure and bone matrix competency in normal rat osteoblast cultures. Endocrinology 128:1496–1504. 55. Ishida H, Bellows CG, Aubin JE, Heersche JNM 1993 Characterization of the 1,25-(OH)2D3-induced inhibition of bone nodule formation in long-term cultures of fetal rat calvarial cells. Endocrinology 132:61–66. 56. Spiess YH, Price PA, Deftos JL, Manolagas SC 1986 Phenotype-associated changes in the effects of 1,25-dihydroxyvitamin D3 on alkaline phosphatase and bone GLA-protein of rat osteoblastic cells. Endocrinology 118:1340–1346. 57. Chen TL, Cone CM, Feldman D 1983 Effects of 1α,25-dihydroxyvitamin D3 and glucocorticoids on the growth of rat and mouse osteoblast-like bone cells. Calcif Tissue Int 35:806–811. 58. Harrison JR, Clark NB 1986 Avian medullary bone in organ culture: Effects of vitamin D metabolites on collagen synthesis. Calcif Tissue Int 39:35–43. 59. Bringhurst FR, Potts JT Jr 1982 Effects of vitamin D metabolites and analogs on bone collagen synthesis in vitro. Calcif Tissue Int 34:103–110. 60. Broess M, Riva A, Gerstenfeld LC 1995 Inhibitory effects of 1,25(OH)2 vitamin D3 on collagen type I, osteopontin, and osteocalcin gene expression in chicken osteoblasts. J Cell Biochem 57:440–451. 61. Gerstenfeld LC, Zurakowski D, J.L. S, Nichols DP, Toma CD, Broess M, Bruder SP, Caplan AI 1996 Variable hormone responsiveness of osteoblast populations isoalted at different stages of embryogenesis and its relationship to the osteogenic lineage. Endocrinology 137:3957–3968. 62. Viereck V, Siggelkow H, Tauber S, Raddatz D, Schutze N, Hufner M 2002 Differential regulation of Cbfa1/Runx2 and osteocalcin gene expression by vitamin-D3, dexamethasone, and local growth factors in primary human osteoblasts. J Cell Biochem 86:348–356. 63. Fraser JD, Otawara Y, Price PA 1988 1,25-Dihydroxyvitamin D3 stimulates the synthesis of matrix gamma-carboxyglutamic acid protein by osteosarcoma cells. Mutually exclusive expression of vitamin K–dependent bone proteins by clonal osteoblastic cell lines. J Biol Chem 263:911–916. 64. Fraser JD, Price PA 1990 Induction of matrix Gla protein synthesis during prolonged 1,25-dihydroxyvitamin D3 treatment of osteosarcoma cells. Calcif Tissue Int 46:270–279. 65. Franceschi RT, Young J 1990 Regulation of alkaline phosphatase by 1,25-dihydroxyvitamin D3 and ascorbic acid in bone-derived cells. J Bone Miner Res 5:1157–1167. 66. Franceschi RT, Romano PR, Park KY 1988 Regulation of type I collagen synthesis by 1,25-dihydroxyvitamin D3 in human osteosarcoma cells. J Biol Chem 263:18938–18945.

660 67. Matsumoto T, Igarashi C, Takeuchi Y, Harada S, Kikuchi T, Yamato H, Ogata E 1991 Stimulation by 1,25-dihydroxyvitamin D3 of in vitro mineralization induced by osteoblastlike MC3T3-E1 cells. Bone 12:27–32. 68. Wergedal JE, Baylink DJ 1984 Characterization of cells isolated and cultured from human bone. Proc Soc Exp Biol Med 176:60–69. 69. Auf’mkolk B, Hauschka PV, Schwartz ER 1985 Characterization of human bone cells in culture. Calcif Tissue Int 37:228–235. 70. Robey PG, Termine JD 1985 Human bone cells in vitro. Calcif Tissue Int 37:453–460. 71. Marie PJ, Lomri A, Sabbagh A, Basle M 1989 Culture and behavior of osteoblastic cells isolated from normal trabecular bone surfaces. In Vitro Cell Dev Biol 25:373–380. 72. Rodan GA, Heath JK, Yoon K, Noda M, Rodan SB 1988 Diversity of the osteoblastic phenotype. Ciba Found Symp 136:78–91. 73. Marie PJ 1994 Human osteoblastic cells: a potential tool to assess the etiology of pathologic bone formation. J Bone Miner Res 9:1847–1850. 74. Bodine PV, Komm BS 2002 Tissue culture models for studies of hormone and vitamin action in bone cells. Vitam Horm 64:101–151. 75. Beresford JN, Joyner CJ, Devlin C, Triffitt JT 1994 The effects of dexamethasone and 1,25-dihydroxyvitamin D3 on osteogenic differentiation of human marrow stromal cells in vitro. Arch Oral Biol 39:941–947. 76. Leboy PS, Beresford JN, Devlin C, Owen ME 1991 Dexamethasone induction of osteoblast mRNAs in rat marrow stromal cell cultures. J Cell Physiol 146:370–378. 77. Rickard DJ, Kassem M, Hefferan TE, Sarkar G, Spelsberg TC, Riggs BL 1996 Isolation and characterization of osteoblast precursor cells from human bone marrow. J Bone Miner Res 11:312–324. 78. Walsh S, Dodds RA, James IE, Bradbeer JN, Gowen M 1994 Monoclonal antibodies with selective reactivity against osteoblasts and osteocytes in human bone. J Bone Miner Res 9:1687–1696. 79. Ingram RT, Bonde SK, Riggs BL, Fitzpatrick LA 1994 Effects of transforming growth factor beta (TGF beta) and 1,25 dihydroxyvitamin D3 on the function, cytochemistry and morphology of normal human osteoblast-like cells. Differentiation 55:153–163. 80. Thavarajah M, Evans DB, Kanis JA 1993 Differentiation of heterogeneous phenotypes in human osteoblast cultures in response to 1,25-dihydroxyvitamin D3. Bone 14:763–767. 81. Ninomiya JT, Tracy RP, Calore JD, Gendreau MA, Kelm RJ, Mann KG 1990 Heterogeneity of human bone. J Bone Miner Res 5:933–938. 82. Chen J, Shapiro HS, Wrana JL, Reimers S, Heersche JNM, Sodek J 1991 Localization of bone sialoprotein (BSP) expression to sites of mineralized tissue formation in fetal rat tissues by in situ hybridization. Matrix 11:133–143. 83. Chen J, Singh K, Mukherjee BB, Sodek J 1993 Developmental expression of osteopontin (OPN) mRNA in rat tissues: Evidence for a role for osteopontin in bone formation and resorption. Matrix 13:113–123. 84. Heersche JNM, Reimers S, Wrana JL, Waye MMY, Gupta AK 1992 Changes in expression of α1 type 1 collagen and osteocalcin mRNA in osteoblasts and odontoblasts at different stages of maturity as shown by in situ hybridization. Proc Finn Dent Soc 88:173–182. 85. Ingram RT, Clarke BL, Fisher LW, Fitzpatrick LA 1993 Distribution of noncollagenous proteins in the matrix of adult

JANE E. AUBIN AND JOHAN N. M. HEERSCHE

86.

87. 88. 89. 90.

91.

92.

93. 94.

95.

96. 97. 98.

99.

100.

101.

human bone: evidence of anatomic and functional heterogeneity. J Bone Miner Res 8:1019–1029. Calvi LM, Sims NA, Hunzelman JL, Knight MC, Giovannetti A, Saxton JM, Kronenberg HM, Baron R, Schipani E 2001 Activated parathyroid hormone/parathyroid hormone-related protein receptor in osteoblastic cells differentially affects cortical and trabecular bone. J Clin Invest 107:277–286. Candeliere GA, Liu F, Aubin JE 2001 Individual osteoblasts in the developing calvaria express different gene repertoires. Bone 28:351–361. Aubin JE, Candeliere GA, Bonnelye E 1999 Heterogeneity of the osteoblast phenotype. Endocrinologist 9:25–31. Liu F, Malaval L, Aubin JE 1997 The mature osteoblast phenotype is characterized by extensive plasticity. Exp Cell Res 232:97–105. Pockwinse SM, Stein JL, Lian JB, Stein GS 1995 Developmental stage-specific cellular responses to vitamin D3 and glucocorticoids during differentiation of the osteoblast phenotype: Interrelationships of morphology and gene expression by in situ hybridization. Exp Cell Res 216:244–260. Bellows CG, Reimers SM, Heersche JN 1999 Expression of mRNAs for type-I collagen, bone sialoprotein, osteocalcin, and osteopontin at different stages of osteoblastic differentiation and their regulation by 1,25 dihydroxyvitamin D3. Cell Tissue Res 297:249–259. Prince M, Banerjee C, Javed A, Green J, Lian JB, Stein GS, Bodine PV, Komm BS 2001 Expression and regulation of Runx2/Cbfa1 and osteoblast phenotypic markers during the growth and differentiation of human osteoblasts. J Cell Biochem 80:424–440. Aubin JE, Gupta AK, Bhargava U, Turksen K 1996 Expression and regulation of galectin 3 in rat osteoblastic cells. J Cell Physiol 169:468–480. Weinreb M, Shinar D, Rodan GA 1990 Different pattern of alkaline phosphatase, osteopontin, and osteocalcin expression in developing rat bone visualized by in situ hybridization. J Bone Miner Res 5:831–842. Rickard DJ, Kazhdan I, Leboy PS 1995 Importance of 1,25dihydroxyvitamin D3 and the nonadherent cells of marrow for osteoblast differentiation from rat marrow stromal cells. Bone 16:671–678. Herbertson A, Aubin J 1995 Dexamethasone alters the subpopulation make-up of rat bone marrow stromal cell cultures. J Bone Miner Res 10:285–294. Long MW, Robinson JA, Mann KG 1995 Regulation of human bone marrow-derived osteoprogenitor cells by osteogenic growth factors. J Clin Invest 95:881–887. Eipers PG, Kale S, Taichman RS, Pipia GG, Swords NA, Mann KG, Long MW 2000 Bone marrow accessory cells regulate human bone precursor cell development. Exp Hematol 28:815–825. Lian JB, Stein JL, Stein GS, Montecino M, van Wijnen AJ, Javed A, Gutierrez S 2001 Contributions of nuclear architecture and chromatin to vitamin D-dependent transcriptional control of the rat osteocalcin gene. Steroids 66:159–170. Lian JB, Stein GS 2003 The temporal and spatial subnuclear organization of skeletal gene regulatory machinery: integrating multiple levels of transcriptional control. Calcif Tissue Int 72:631–637. Haussler MR, Jurutka PW, Hsieh JC, Thompson PD, Selznick SH, Haussler CA, Whitfield GK 1995 New understanding of the molecular mechanism of receptor-mediated genomic actions of the vitamin D hormone. Bone 17: 33S–38S.

CHAPTER 37 Vitamin D and Osteoblasts

102. White C, Gardiner E, Eisman J 1998 Tissue specific and vitamin D responsive gene expression in bone. Mol Biol Rep 25:45–61. 103. Lian JB, Stein GS, Stein JL, van Wijnen AJ 1999 Regulated expression of the bone-specific osteocalcin gene by vitamins and hormones. Vitam Horm 55:443–509. 104. Fernandez-Martin JL, Kurian S, Farmer P, Nanes MS 1998 Tumor necrosis factor activates a nuclear inhibitor of vitamin D and retinoid-X receptors. Mol Cell Endocrinol 141:65–72. 105. Godschalk M, Levy JR, Downs RW Jr 1992 Glucocorticoids decrease vitamin D receptor number and gene expression in human osteosarcoma cells. J Bone Miner Res 7:21–27. 106. Staal A, Van Wijnen AJ, Desai RK, Pols HA, Birkenhager JC, Deluca HF, Denhardt DT, Stein JL, Van Leeuwen JP, Stein GS, Lian JB 1996 Antagonistic effects of transforming growth factor-beta on vitamin D3 enhancement of osteocalcin and osteopontin transcription: reduced interactions of vitamin D receptor/retinoid X receptor complexes with vitamin E response elements. Endocrinology 137:2001–2011. 107. Gurlek A, Kumar R 2001 Regulation of osteoblast growth by interactions between transforming growth factor-beta and 1α,25-dihydroxyvitamin D3. Crit Rev Eukaryot Gene Expr 11:299–317. 108. Franceschi RT 2003 Functional cooperativity between osteoblast transcription factors: evidence for the importance of subnuclear macromolecular complexes? Calcif Tissue Int 72:638–642. 109. McCabe LR, Kockx M, Lian J, Stein J, Stein G 1995 Selective expression of fos- and jun-related genes during osteoblast proliferation and differentiation. Exp Cell Res 218:255–262. 110. Sabatakos G, Sims NA, Chen J, Aoki K, Kelz MB, Amling M, Bouali Y, Mukhopadhyay K, Ford K, Nestler EJ, Baron R 2000 Overexpression of DeltaFosB transcription factor(s) increases bone formation and inhibits adipogenesis. Nat Med 6:985–990. 111. Candeliere GA, Prud’homme J, St-Arnaud R 1991 Differential stimulation of fos and jun family members by calcitriol in osteoblastic cells. Mol Endocrinol 5: 1780–1788. 112. Hodgkinson JE, Davidson CL, Beresford J, Sharpe PT 1993 Expression of a human homeobox-containing gene is regulated by 1,25(OH)2D3 in bone cells. Biochim Biophys Acta 1174:11–16. 113. Jabs EW, Muller U, Li X, Ma L, Luo W, Haworth IS, Klisak I, Sparkes R, Warman ML, Mulliken JB, Snead ML, Maxson R 1993 A mutation in the homeodomain of the human Msx-2 gene in a family affected with autosomal dominant craniosynostosis. Cell 785:443–450. 114. Satokata I, Ma L, Ohshima H, Bei M, Woo I, Nishizawa K, Maeda T, Takano Y, Uchiyama M, Heaney S, Peters H, Tang Z, Maxson R, Maas R 2000 Msx2 deficiency in mice causes pleiotropic defects in bone growth and ectodermal organ formation. Nat Genet 24:391–395. 115. Ishii M, Merrill AE, Chan YS, Gitelman I, Rice DP, Sucov HM, Maxson RE Jr, 2003 Msx2 and Twist cooperatively control the development of the neural crest-derived skeletogenic mesenchyme of the murine skull vault. Development 130: 6131–6142. 116. Hoffmann HM, Catron KM, van Wijnen AJ, McCabe LR, Lian JB, Stein GS, Stein JL 1994 Transcriptional control of the tissue-specific, developmentally regulated osteocalcin gene requires a binding motif for the Msx family of homeodomain proteins. Proc Natl Acad Sci USA 91: 12887–128891.

661 117. Towler DA, Rutledge SJ, Rodan GA 1994 Msx-2/Hox 8.1: A transcriptional regulator of the rat osteocalcin promoter. Mol Endocrinol 8:1484–1493. 118. Ryoo HM, Hoffmann HM, Beumer T, Frenkel B, Towler DA, Stein GS, Stein JL, van Wijnen AJ, Lian JB 1997 Stage-specific expression of Dlx-5 during osteoblast differentiation: involvement in regulation of osteocalcin gene expression. Mol Endocrinol 11:1681–1694. 119. Liu M, Lee MH, Cohen M, Bommakanti M, Freedman LP 1996 Transcriptional activation of the Cdk inhibitor p21 by vitamin D3 leads to the induced differentiation of the myelomonocytic cell line U937. Genes Dev 10:142–153. 120. Atkinson MJ, Freeman MW, Kronenberg HM 1990 Thymosin beta 4 is expressed in ROS 17/2.8 osteosarcoma cells in a regulated manner. Mol Endocrinol 4:69–74. 121. Jin CH, Shinki T, Hong MH, Sato T, Yamaguchi A, Ikeda T, Yoshiki S, Abe E, Suda T 1992 1 α,25-dihydroxyvitamin D3 regulates in vivo production of the third component of complement (C3) in bone. Endocrinology 131:2468–2475. 122. Sato T, Abe E, Jin CH, Hong MH, Katagiri T, Kinoshita T, Amizuka N, Ozawa H, Suda T 1993 The biological roles of the third component of complement in osteoclast formation. Endocrinology 133:397–404. 123. Atkins GJ, Kostakis P, Pan B, Farrugia A, Gronthos S, Evdokiou A, Harrison K, Findlay DM, Zannettino AC 2003 RANKL expression is related to the differentiation state of human osteoblasts. J Bone Miner Res 18:1088–1098. 124. Suda T, Takahashi N, Udagawa N, Jimi E, Gillespie MT, Martin TJ 1999 Modulation of osteoclast differentiation and function by the new members of the tumor necrosis factor receptor and ligand families. Endocr Rev 20:345–357. 125. Aubin JE, Bonnelye E 2000 Osteoprotegerin and its ligand: A new paradigm for regulation of osteoclastogenesis and bone resorption. Osteoporos Int 11:905–913. 126. Homme M, Schmitt CP, Himmele R, Hoffmann GF, Mehls O, Schaefer F 2003 Vitamin D and dexamethasone inversely regulate parathyroid hormone-induced regulator of G protein signaling-2 expression in osteoblast-like cells. Endocrinology 144:2496–2504. 127. Lacey DL, Erdmann JM, Tan HL, Ohara J 1993 Murine osteoblast interleukin 4 receptor expression: upregulation by 1,25 dihydroxyvitamin D3. J Cell Biochem 53: 122–134. 128. Lacey DL, Grosso LE, Moser SA, Erdmann J, Tan HL, Pacifici R, Villareal DT 1993 IL-1-induced murine osteoblast IL-6 production is mediated by the type 1 IL-1 receptor and is increased by 1,25 dihydroxyvitamin D3. J Clin Invest 91:1731–1742. 129. Lee SK, Kalinowski J, Jastrzebski S, Lorenzo JA 2002 1,25(OH)2 vitamin D3-stimulated osteoclast formation in spleen-osteoblast cocultures is mediated in part by enhanced IL-1 alpha and receptor activator of NF-κ B ligand production in osteoblasts. J Immunol 169:2374–2380. 130. Kim CH, Kang BS, Lee TK, Park WH, Kim JK, Park YG, Kim HM, Lee YC 2002 IL-1beta regulates cellular proliferation, prostaglandin E2 synthesis, plasminogen activator activity, osteocalcin production, and bone resorptive activity of the mouse calvarial bone cells. Immunopharmacol Immunotoxicol 24:395–407. 131. Jehan F, Naveilhan P, Neveu I, Harvie D, Dicou E, Brachet P, Wion D 1996 Regulation of NGF, BDNF and LNGFR gene expression in ROS 17/2.8 cells. Mol Cell Endocrinol 116:149–156. 132. Veenstra TD, Fahnestock M, Kumar R 1998 An AP-1 site in the nerve growth factor promoter is essential for

662

133.

134.

135. 136.

137.

138.

139.

140.

141.

142.

143. 144.

145. 146.

147.

JANE E. AUBIN AND JOHAN N. M. HEERSCHE

1,25-dihydroxyvitamin D3-mediated nerve growth factor expression in osteoblasts. Biochemistry 37:5988–5994. Nambi P, Wu HL, Lipshutz D, Prabhakar U 1995 Identification and characterization of endothelin receptors on rat osteoblastic osteosarcoma cells: down-regulation by 1,25-dihydroxy-vitamin D3. Mol Pharmacol 47:266–271. Kasperk CH, Borcsok I, Schairer HU, Schneider U, Nawroth PP, Niethard FU, Ziegler R 1997 Endothelin-1 is a potent regulator of human bone cell metabolism in vitro. Calcif Tissue Int 60:368–374. Rosen H, Krichevsky A, Polakiewicz RD, Benzakine S, Bar-Shavit Z 1995 Developmental regulation of proenkephalin gene expression in osteoblasts. Mol Endocrinol 9:1621–1631. Kassem M, Mosekilde L, Eriksen EF 1993 1,25-Dihydroxyvitamin D3 potentiates fluoride-stimulated collagen type I production in cultures of human bone marrow stromal osteoblast-like cells. J Bone Miner Res 8:1453–1458. Kveiborg M, Flyvbjerg A, Kassem M 2002 Synergistic effects of 1,25-dihydroxyvitamin D3 and TGF-beta1 on the production of insulin-like growth factor binding protein 3 in human bone marrow stromal cell cultures. APMIS 110:410–414. Williams GR, Bland R, Sheppard MC 1995 Retinoids modify regulation of endogenous gene expression by vitamin D3 and thyroid hormone in three osteosarcoma cell lines. Endocrinology 136:4304–4314. Bland R, Sammons RL, Sheppard MC, Williams GR 1997 Thyroid hormone, vitamin D and retinoid receptor expression and signalling in primary cultures of rat osteoblast and immortalised osteosarcoma cells. Retinoids modify regulation of endogenous gene expression by vitamin D3 and thyroid hormone in three osteosarcoma cell lines. J Endocrinol 154:63–74. Ohishi K, Nishikawa S, Nagata T, Yamauchi N, Shinohara H, Kido J, Ishida H 1995 Physiological concentrations of retinoic acid suppress the osteoblastic differentiation of fetal rat calvaria cells in vitro. Eur J Endocrinol 133:335–341. Ohishi K, Ishida H, Nagata T, Yamauchi N, Tsurumi C, Nishikawa S, Wakano Y 1994 Thyroid hormone suppresses the differentiation of osteoprogenitor cells to osteoblasts, but enhances functional activities of mature osteoblasts in cultured rat calvaria cells. J Cell Physiol 161:544–552. Ishida H, Bellows CG, Aubin JE, Heersche JNM 1994 Triiodothyronine (T3) and dexamethasone interact to modulate osteoprogenitor cell differentiation in fetal rat calvaria cell cultures. Bone 16:545–549. Choong PF, Martin TJ, Ng KW 1993 Effects of ascorbic acid, calcitriol, and retinoic acid on the differentiation of preosteoblasts. J Orthop Res 11:638–647. Chapman AB, Knight DM, Ringold GM 1985 Glucocorticoid regulation of adipocyte differentiation: Hormonal triggering of the developmental program and induction of a differentiation-dependent gene. J Cell Biol 101:1227–1235. Schiwek DR, Loffler G 1987 Glucocorticoid hormones contribute to the adipogenic activity of human serum. Endocrinology 120:469–474. Deryugina EI, Ratnikov BI, Bourdon MA, Muller-Sieburg CE 1994 Clonal analysis of primary marrow stroma: functional homogeneity in support of lymphoid and myeloid cell lines and identification of positive and negative regulators. Exp Hematol 22:910–918. Nuttall ME, Patton AJ, Olivera DL, Nadeau DP, Gowen M 1998 Human trabecular bone cells are able to express both osteoblastic and adipocytic phenotype: Implications for osteogenic disorders. J Bone Miner Res 13:371–382.

148. Sato M, Hiragun A 1988 Demonstration of 1α,25-dihydroxyvitamin D3 receptor-like molecule in ST 13 and 3T3 L1 preadipocytes and its inhibitory effects on preadipocyte differentiation. J Cell Physiol 135:545–550. 149. Ishida Y, Taniguchi H, Baba S 1988 Possible involvement of 1α,25-dihydroxyvitamin D3 in proliferation and differentiation of 3T3-L1 cells. Biochem Biophys Res Commun 151: 1122–1127. 150. Kawada T, Aoki N, Kamei Y, Maeshige K, Nishiu S, Sugimoto E 1990 Comparative investigation of vitamins and their analogs on terminal differentiation, from preadipocytes to adipocytes, of 3T3-L1 cells. Comp Biochem Physiol A 96:323–326. 151. Kawada T, Kamei Y, Sugimoto E 1996 The possibility of active form of vitamins A and D as suppressors on adipocyte development via ligand-dependent transcriptional regulators. Int J Obes Relat Metab Disord 20:S52–S57. 152. Beresford JN, Bennett JH, Devlin C, Leboy PS, Owen ME 1992 Evidence for an inverse relationship between the differentiation of adipocytic and osteogenic cells in rat bone marrow stromal cell cultures. J Cell Sci 102:341–351. 153. Shionome M, Shinki T, Takahashi N, Hasegawa K, Suda T 1992 1α,25-Dihydroxyvitamin D3 modulation in lipid metabolism in established bone marrow-derived stromal cells, MC3T3-G2/PA6. J Cell Biochem 48:424–430. 154. Kelly KA, Gimble JM 1998 1,25-Dihydroxy vitamin D3 inhibits adipocyte differentiation and gene expression in murine bone marrow stromal cell clones and primary cultures. Endocrinology 139:2622–2628. 155. Falla N, Van Vlassalaer P, Bierkens J, Borremans B, Schoeters G, Van Gorp U 1993 Characterization of a 5-fluorouracil-enriched osteoprogenitor population of the murine bone marrow. Blood 82:3580–3591. 156. Bellows CG, Ciaccia A, Heersche JN 1998 Osteoprogenitor cells in cell populations derived from mouse and rat calvaria differ in their response to corticosterone, cortisol, and cortisone. Bone 23:119–125. 157. Meunier P, Aaron J, Edouard C, Vignon G 1971 Osteoporosis and the replacement of cell populations of the marrow by adipose tissue. A quantitative study of 84 iliac bone biopsies. Clin Orthop 80:147–154. 158. Burkhardt R, Kettner G, Bohm W, Schmidmeier M, Schlag R, Frisch B, Mallmann B, Eisenmenger W, Gilg T 1987 Changes in trabecular bone, hematopoiesis and bone marrow vessels in aplastic anemia, primary osteoporosis, and old age: A comparative histomorphometric study. Bone 8:157–164. 159. Rozman C, Feliu E, Berga L, Reverter JC, Climent C, Ferran MJ 1989 Age-related variations of fat tissue fraction in normal human bone marrow depend both on size and number of adipocytes: A stereological study. Exp Hematol 17:34–37. 160. Minaire P, Edouard C, Arlot M, Meunier PJ 1984 Marrow changes in paraplegic patients. Calcif Tissue Int 36:338–340. 161. Gimble JM, Robinson CE, Kelly KA 1996 The function of adipocytes in the bone marrow stroma: An update. Bone 19:421–428. 162. Justesen J, Stenderup K, Ebbesen EN, Mosekilde L, Steiniche T, Kassem M 2001 Adipocyte tissue volume in bone marrow is increased with aging and in patients with osteoporosis. Biogerontology 2:165–171. 163. Wronski TJ, Morey-Holton E, Jee WS 1981 Skeletal alterations in rats during space flight. Adv Space Res 1:135–140. 164. Martin RB, Zissimos SL 1991 Relationships between marrow fat and bone turnover in ovariectomized and intact rats. Bone 12:123–131.

663

CHAPTER 37 Vitamin D and Osteoblasts

165. Kajkenova O, Lecka-Czernik B, Gubrij I, Hauser SP, Takahashi K, Parfitt AM, Jilka RL, Manolagas SC, Lipschitz DA 1997 Increased adipogenesis and myelopoiesis in the bone marrow of SAMP6, a murine model of defective osteoblastogenesis and low turnover osteopenia. J Bone Miner Res 12:1772–1779. 166. Nuttall ME, Gimble JM 2000 Is there a therapeutic opportunity to either prevent or treat osteopenic disorders by inhibiting marrow adipogenesis? Bone 27:177–184. 167. Gimble JM, Nuttall ME 2004 Bone and fat: old questions, new insights. Endocrine 23:183–188. 168. Lecka-Czernik B, Gubrij I, Moerman EJ, Kajkenova O, Lipschitz DA, Manolagas SC, Jilka RL 1999 Inhibition of Osf2/Cbfa1 expression and terminal osteoblast differentiation by PPARgamma2. J Cell Biochem 74:357–371. 169. Aubin JE 2001 Your partner makes all the difference. In: IBMS–BoneKEy. http://www.bonekey–ibms.org/cgi/content/ full//ibmske;2001013v1. 170. Kveiborg M, Sabatakos G, Chiusaroli R, Wu M, Philbrick WM, Horne WC, Baron R 2004 DeltaFosB induces osteosclerosis and decreases adipogenesis by two independent cellautonomous mechanisms. Mol Cell Biol 24:2820–2830. 171. Bellows CG, Heersche JN 2001 The frequency of common progenitors for adipocytes and osteoblasts and of committed and restricted adipocyte and osteoblast progenitors in fetal rat calvaria cell populations. J Bone Miner Res 16:1983–1993. 172. Tam CS, Jones G, Heersche JN 1981 The effect of vitamin D restriction and repletion on bone apposition in the rat and its dependence on parathyroid hormone. Endocrinology 109:1448–1453. 173. Tam CS, Heersche JN, Jones G, Murray TM, Rasmussen H 1986 The effect of vitamin D on bone in vivo. Endocrinology 118:2217–2224. 174. Hock JM, Gunness-Hey M, Poser J, Olson H, Bell NH, Raisz LG 1986 Stimulation of undermineralized matrix formation by 1,25-dihydroxyvitamin D3 in long bones of rats. Calcif Tissue Int 38:79–86. 175. Ono T, Tanaka H, Yamate T, Nagai Y, Nakamura T, Seino Y 1996 24R,25-Dihydroxyvitamin D3 promotes bone formation without causing excessive resorption in hypophosphatemic mice. Endocrinology 137:2633–2637. 176. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y,

177.

178.

179.

180.

181.

182.

183.

Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Genet 16:391–396. Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997 Targeted ablation of the vitamin D receptor: an animal model of vitamin D-dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94:9831–9835. Li YC, Amling M, Pirro AE, Priemel M, Meuse J, Baron R, Delling G, Demay MB 1998 Normalization of mineral ion homeostasis by dietary means prevents hyperparathyroidism, rickets, and osteomalacia, but not alopecia in vitamin D receptor-ablated mice. Endocrinology 139: 4391–4396. Panda DK, Miao D, Tremblay ML, Sirois J, Farookhi R, Hendy GN, Goltzman D 2001 Targeted ablation of the 25-hydroxyvitamin D 1α-hydroxylase enzyme: Evidence for skeletal, reproductive, and immune dysfunction. Proc Natl Acad Sci USA 98:7498–7503. Dardenne O, Prud’homme J, Arabian A, Glorieux FH, St-Arnaud R 2001 Targeted inactivation of the 25-hydroxyvitamin D3-1α-hydroxylase gene (CYP27B1) creates an animal model of pseudovitamin D–deficiency rickets. Endocrinology 142:3135–3141. Dardenne O, Prudhomme J, Hacking SA, Glorieux FH, St-Arnaud R 2003 Rescue of the pseudo–vitamin D deficiency rickets phenotype of CYP27B1-deficient mice by treatment with 1,25-dihydroxyvitamin D3: Biochemical, histomorphometric, and biomechanical analyses. J Bone Miner Res 18:637–643. Dardenne O, Prud’homme J, Hacking SA, Glorieux FH, St-Arnaud R 2003 Correction of the abnormal mineral ion homeostasis with a high-calcium, high-phosphorus, highlactose diet rescues the PDDR phenotype of mice deficient for the 25-hydroxyvitamin D-1α-hydroxylase (CYP27B1). Bone 32:332–340. St-Arnaud R, Arabian A, Travers R, Barletta F, Raval-Pandya M, Chapin K, Depovere J, Mathieu C, Christakos S, Demay MB, Glorieux FH 2000 Deficient mineralization of intramembranous bone in vitamin D-24-hydroxylase-ablated mice is due to elevated 1,25-dihydroxyvitamin D and not to the absence of 24,25-dihydroxyvitamin D. Endocrinology 141: 2658–2666.

CHAPTER 38

Vitamin D and Osteoclastogenesis HISATAKA YASUDA Center for Experimental Medicine, Institute of Medical Science, University of Tokyo, Tokyo, Japan

KANJI HIGASHIO AND TATSUO SUDA Research Center for Genomic Medicine, Saitama Medical School, Saitama, Japan

I. Introduction II. Discovery of Bone Mineral Mobilization Activity of Vitamin D III. Establishment of a Mouse Coculture System to Recruit Osteoclasts IV. Discovery of Key Factors to Understand the Molecular Mechanism of Osteoclastogenesis

V. Signal Transduction Pathways of the RANKL-RANK System in Osteoclast Development VI. The Biological Relevance of Vitamin D to Osteoclastic Bone Resorption VII. Conclusion References

I. INTRODUCTION

reduced expression of vitamin D target genes. After weaning, however, mutant mice failed to thrive, and alopecia, hypocalcemia, and infertility resulted [4]. (These features of defective VDRs are also observed in type II vitamin D–dependent rickets, see Chapter 72.) Both bone formation and mineralization are severely impaired and are typical features of HVDRR (see Chapter 72). Most of the KO mice died within 15 to 25 weeks after birth due to severe hypocalcemia. Unexpectedly, when these VDR-KO mice were fed a rescue diet containing high calcium, they developed normally even at week 50, but severe alopecia remained [4]. Bone formation and mineralization in the VDR-KO mice were completely reestablished through a diet containing high calcium. From these results, it was concluded that the stimulating effect of 1,25(OH)2D3 on bone formation and mineralization is indirect, occurring through the stimulation of intestinal absorption of calcium by vitamin D.

Bone is a dynamic tissue that is formed and remodeled by continuously occurring bone formation and resorption. An imbalance between bone formation and resorption causes several metabolic bone diseases such as osteoporosis and osteopetrosis. Bone-forming osteoblasts derive from undifferentiated mesenchymal cells, whereas bone-resorbing osteoclasts develop from hemopoietic cells of the monocyte–macrophage lineage. The hemopoietic osteoclast precursor cells differentiate into osteoclasts at bone-resorbing sites under the control of several osteotropic hormones and cytokines. It is well recognized that serum calcium levels are tightly regulated and maintained at 9 to 10 mg/dl in healthy animals and humans [1–3]. Intestine, bone, and kidney are the three major organs involved in this calcium homeostasis. Vitamin D plays a major role in regulating serum calcium homeostasis in concert with parathyroid hormone (PTH) and calcitonin. Most of the biological effects generated by vitamin D are produced by its active metabolite, 1,25(OH)2D3 [1]. Vitamin D receptors (VDRs), which bind 1,25(OH)2D3 specifically, have been reported to be present in these three organs [1]. In bone, VDRs are located preferentially in osteoblasts. In addition, vitamin D–deficient animals and humans exhibit severe rickets and osteomalacia. From these experimental results and clinical evidence, it was postulated that vitamin D directly stimulates bone formation, and in particular, bone mineralization [1,3]. Kato and his associates [4] in Japan first succeeded in generating mice deficient in VDR by gene targeting (see Chapter 20). They showed that in VDR knockout (KO) mice, no appreciable defects were observed in development and growth before weaning, irrespective of the VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

II. DISCOVERY OF BONE MINERAL MOBILIZATION ACTIVITY OF VITAMIN D It appears to be paradoxical, but vitamin D functions in the process of calcium mobilization from calcified bone, making calcium available to the extracellular fluid upon demand by the calcium homeostatic system. This important observation was first reported by Carlsson [5]. He showed that, when hypocalcemic rats maintained on a vitamin D–deficient, low-calcium diet were given 100 IU (2.5 µg) of vitamin D3 orally, their serum calcium was increased from 5 to 8 mg/dl after 3 days (Fig. 1A). Parathyroidectomy (PTX) 2 hr prior to vitamin D3 administration abolished the increase in serum calcium levels (Fig. 1A). Since the diet did not contain any Copyright © 2005, Elsevier, Inc. All rights reserved.

666

HISATAKA YASUDA, KANJI HIGASHIO, AND TATSUO SUDA

A In vivo

B In vitro 25(OH)D3

3.0

10

1,25(OH)2D3 Bone resorption release (T/C ratio)

8 7 6

45Ca

Serum Ca (mg/dl)

9

5 PTX

2.5

2.0

1.5

4 −2 0

1.0 3

9

27

D3 (2.5 µg/head)

81 (h)

PTX

0.01

10−11

0.1

10−10

1 10 100 Concentration (ng/ml) 10−9 10−8 10−7 Concentration (M)

1000

10−6

10−5

FIGURE 1

(A) Discovery of bone mineral mobilization activity of vitamin D in vivo. PTX; parathyroidectomy. Reprinted from Carlsson (1952), with permission. (B) The comparison of the in vitro activity to increase the release of 45Ca from prelabeled bone between 25(OH)D3 and 1,25(OH)2D3. Reprinted from Raisz et al. 1972, Science 175:768–769. Copyright 1972 American Association for the Advancement of Science.

appreciable amounts of calcium, he concluded that vitamin D stimulates mineral moblilization from calcified bone to blood in concert with PTH [5]. The metabolite of vitamin D3 responsible for bone mineral mobilization was 1,25(OH)2D3. Using an in vitro organ culture system, Raisz et al. [6] reported that both 1,25(OH)2D3 and 25-hydroxyvitamin D3 [25(OH)D3] increased the release of 45Ca from prelabeled bone into the culture medium, and that 1,25(OH)2D3 is 80 times more potent than 25(OH)D3 in increasing 45Ca release from prelabeled bone (Fig. 1b) [6]. From these results, they concluded that the metabolite of vitamin D3 that stimulates bone mineral mobilization is indeed 1,25(OH)2D3. Abe et al. [7] discovered the cell differentiation– inducing activity of 1,25(OH)2D3 using mouse and human myeloid leukemia cells. HL-60 is a human promyelocytic leukemia cell line established from a leukemic patient, and the cells can be induced to differentiate into granulocytes by retinoic acid and monocytes– macrophages by 1,25(OH)2D3. 1,25(OH)2D3 was a potent and selective inducer of differentiation of HL-60 cells into macrophages [8]. Furthermore, 1,25(OH)2D3 directly induced fusion of alveolar macrophages at a very high rate [9]. Approximately 80% of the macrophages fused to form multinucleated giant cells by stimulating the differentiation and fusion of macrophages [9]. However, the multinucleated giant cells formed from alveolar macrophages in response to 1,25(OH)2D3 did not satisfy the criteria of osteoclasts.

III. ESTABLISHMENT OF A MOUSE COCULTURE SYSTEM TO RECRUIT OSTEOCLASTS A. A Mouse Coculture System In 1981, Rodan and Martin [10] proposed that osteoblasts or osteoblastic stromal cells may intervene in the process of bone resorption by osteoclasts. Their argument for such a mechanism was based on the observations that first, receptors for most of the boneresorbing hormones and cytokines are localized in osteoblastic cells but not in osteoclasts, and second, the relative binding potencies of these bone-resorbing factors to their receptors in osteoblasts resemble those that induce bone resorption. The same conclusion was reached independently by Chambers [11], who proposed that a factor called osteoclast activating factor (OAF) is produced by osteoblastic cells in response to bone-resorbing hormones and cytokines, and OAF then stimulates osteoclast activation. To examine the possible involvement of osteoblastic cells in osteoclast formation, we established an efficient mouse coculture system to recruit osteoclasts (Fig. 2) [12] based on the concept proposed by Rodan and Martin [10]. In this coculture system, primary osteoblastic cells were isolated from mouse calvaria, and spleen cells isolated from the splenic tissue were used as osteoclast progenitors. When osteoblastic cells alone or spleen cells alone were cultured, no

667

Mouse Calvaria

Osteoblasts

Spleen

Hemopoietic cells

Co-culture for 6–8 days

TRAP-positive MNC formation (cells formed/well)

CHAPTER 38 Vitamin D and Osteoclastogenesis

Osteoblasts

Spleen cells

Osteoblasts + Spleen cells

Membrane filter

400

300

200

100

0 1,25(OH)2D3

−

+

−

+

−

+

−

+

Osteoclasts

FIGURE 2 A mouse coculture system to recruit osteoclasts. Primary osteoblasts from calvaria and/or hemopoetic cells from spleen were cultures for 6–8 days in the presence or absence of 10−8 M 1,25(OH)2D3. Tartrate-resistant acid phosphatase (TRAP)-positive multinucleated cells (MNC), which were identified as osteoclasts, were formed only when osteoblasts and spleen cells were cocultured in the presence of 1,25(OH)2D3.

osteoclasts were formed even in the presence of 1,25(OH)2D3. Multinucleated osteoclasts were formed only when spleen cells and osteoblastic cells were cocultured in the presence of 1,25(OH)2D3. Cell-tocell contact between spleen cells and osteoblastic cells appeared important for osteoclast formation, since no osteoclasts were formed when they were cocultured but separated by a membrane filter. No osteoclasts were formed in the absence of 1,25(OH)2D3 even in the coculture. Considering all this, we hypothesized that the direct contact of spleen cells and osteoblastic cells was essential for osteoclast differentiation [13]. Spleen cells represent osteoclast progenitors, in other words “seeds,” and osteoblastic cells represent the supporting cells to provide a suitable microenvironment for osteoclast formation in bone, in other words “farm” or “soil.”

B. A Hypothetical Factor: Osteoclast Differentiation Factor (ODF) In 1992, we proposed a working hypothesis for osteoclastogenesis based on the extensive studies using the coculture system [13] (Fig. 3). Various bone-resorbing hormones and cytokines including 1,25(OH)2D3, PTH, and interleukin (IL)-11 appeared to act commonly on osteoblastic cells, but not on hemopoietic osteoclast precursors in co-cultures of osteoblastic cells and spleen cells. These bone-resorbing factors were classified into three

categories in terms of their signal transduction pathways: VDR-mediated signals [1,25(OH)2D3], protein kinase Amediated signals [PTH, prostaglandin E2 (PGE2) and IL-1], and gp130-mediated signals [IL-6, IL-11, oncostatin M (OSM), and leukemia inhibitory factor (LIF)]. These three diverse signals appeared to stimulate osteoclast formation independently, since VDR-KO mice and gp130-KO mice possessed osteoclasts in bone tissues in vivo. We proposed that a membrane-bound factor(s), which is commonly induced on osteoblastic cells in response to these bone-resorbing factors, mediates an essential signal to osteoclast progenitors to promote differentiation into mature osteoclasts (Fig. 3). We named the factor “osteoclast differentiation factor” (ODF). ODF appeared to be identical to “stromal cell-derived osteoclast formation activity” (SOFA) proposed by Chambers et al. [14]. Osteoclast progenitors having ODF receptor recognize ODF by cell-to-cell contact and differentiate into osteoclasts. Macrophage colonystimulating factor (M-CSF) produced by osteoblastic cells is also indispensable for both proliferation and differentiation of osteoclast progenitors [15,16]. Yoshida et al. [17] demonstrated that osteopetrotic (op/op) mutant mice with a defect in the development of osteoclasts have a loss-of-function mutation in the coding region of M-CSF gene. Thus, osteoblastic cells are important for osteoclast recruitment in two different ways: one is the production of M-CSF, and the other is the production of a membrane-bound factor such as

668

HISATAKA YASUDA, KANJI HIGASHIO, AND TATSUO SUDA

Hemopoietic stem cells M-CSF

Proliferation and differentiation Monocyte/macrophage lineage cells

1,25(OH)2D3 VDR PHE2 PTH IL-1 1L-6 IL-11 OSM LIF

Differentiation Nucleus

EP4 PTHR1 IL-1R IL-6R IL-11R OSMR LIFR gp130

Osteoclast precursors Commitment of osteoclast differentiation ODF

Mononuclear osteoclasts Fusion and activation

Osteoblasts/stromal cells M-CSF

Osteoclasts

FIGURE 3

A hypothesis for osteoclast development.

ODF commonly induced by several bone-resorbing factors [13,18] (Fig. 3).

IV. DISCOVERY OF KEY FACTORS TO UNDERSTAND THE MOLECULAR MECHANISM OF OSTEOCLASTOGENESIS A. The Discovery of Osteoclastogenesis Inhibitory Factor/Osteoprotegerin (OCIF/OPG) 1. ISOLATION AND MOLECULAR CLONING OF OCIF

In 1997, research groups in Snow Brand Milk Products and Amgen independently discovered an important clue for the identification of ODF. Snow Brand succeeded in purifying a novel factor called osteoclastogenesis inhibitory factor (OCIF) from conditioned medium of human primary fibroblasts (IMR-90) by using bone marrow cells treated with 1,25(OH)2D3 as an assay system [19]. Since fibroblasts are present ubiquitously in the body, it was not envisioned that cells could produce a novel osteoclastogenesis inhibitory factor. We have already reported that IMR-90 cells produce a number of cytokines and growth factors including hepatocyte growth factor (HGF) [20]. We subsequently cloned human OCIF cDNA by polymerase chain reaction (PCR) with primers designed using

the internal amino-acid sequences of protease-digested OCIF [21]. Simonet et al. [22] in Amgen independently cloned the same molecule in an expressed sequence tag cDNA project. They called the factor osteoprotegerin (OPG) because of the findings that transgenic (TG) mice overexpressing this protein exhibited osteopetrosis due to an inhibition of terminal differentiation of osteoclasts. It is surprising that an inhibitor of osteoclastogenesis that had not been postulated before was discovered at the same time by two research groups independently. The American Society for Bone and Mineral Research (ASBMR) President’s Committee on Nomenclature proposed that the name of choice be “osteoprotegerin (OPG)” [23,24]. 2. CHARACTERIZATION OF OPG/OCIF

OPG inhibited all in vitro osteoclast formation elicited through the three distinct signaling pathways stimulated by 1,25(OH)2D3, PTH, and IL-11 [19]. OPG is a heparinbinding basic glycoprotein, and it has been isolated as a monomer with an apparent molecular weight (Mr) of 60 kDa as well as a disulfide-linked homodimer with a Mr of 120 kDa [19]. Nucleotide sequence analysis of OPG cDNA revealed that OPG is a secreted member of the tumor necrosis factor (TNF) receptor family, containing a signal peptide, four cysteine-rich domains (CRDs), and two death domain homologous regions (DDHs), but not a transmembrane (TM) domain (Fig. 4) [21,22]. Analyses of TG mice overexpressing OPG [22] and of animals injected with OPG [21,22] have demonstrated

669

CHAPTER 38 Vitamin D and Osteoclastogenesis

A

SP

B

CRD I 1

22

II 64

III 107

DDH1

IV 144

DDH2

Cys400 Dimer401 361 formation

186 209

Osteoclastogenesis inhibitory activity

Heparin binding Cytotoxicity

FIGURE 4

Structure and functional domains of human OPG. (A) Amino acid sequence of OPG. OPG consists of 401 amino acid residues. An asterisk indicates glutamic acid (E) of the N terminus of mature OPG. Dots indicate cysteines in the N-terminal region. Dotted lines indicate two death domain homologous regions [DDH1 and DDH2]. (B) A representation of functional domains of OPG. A closed box represents the signal peptide (Met1–Gln21). Cysteine-rich domains (CRD I, Glu22–Ala63; II, Pro64–Glu106; III, Cys107–Lys143; IV, Arg144–Ser186) are shown as open boxes, and cysteine residues by vertical lines. CRDs (I–IV) are responsible for osteoclastogenesis inhibitory activity. Striped boxes represent two DDHs [DDH1 (Phe209–Gly286) and DDH2 (His284–Val361)]. A heparin binding site is located in the C-terminal region. Cys400 is responsible for dimer formation. Two DDHs have high potential to mediate apoptosis.

that OPG increases bone mass by suppressing osteoclastic bone resorption. Administration of OPG to normal rats and mice, thyroparathyroidectomized (TPTX) rats, and hypercalcemic nude mice carrying tumors associated with humoral hypercalcemia of malignancy (HHM) similarly decreased serum calcium concentrations rapidly (within 2 hr) [25–28]. Morony et al. [29] also reported that OPG was effective in reducing hypercalcemia induced by various bone-resorbing factors such as 1,25(OH)2D3, PTH, IL-1, and TNF. Several in vivo studies reported that OPG was effective

in reducing and blocking bone destruction and/or tumor burden within the bone induced by breast carcinoma [30] and sarcoma cells [31], as well as in a mouse myeloma model [32]. OPG-KO mice exhibited severe osteoporosis due to enhanced osteoclast formation [33,34]. Destruction of growth plates and lack of trabecular bone with an increase in the number of osteoclasts were also detected in long bones of OPG-KO mice. The strength and mineral density of their bones were greatly decreased. Taken together, these results established

670 the physiological role of OPG as a potent osteoclastogenesis inhibitory factor. Recently, loss-of-function mutations in human OPG gene were reported by two independent groups. Whyte et al. [35] reported that juvenile Paget’s disease could result from a deletion spanning approximately 100 kb in the OPG locus. The deletion included the entire OPG gene, and serum OPG levels in the patients were undetectable. Cundy et al. [36] reported that a 3-bp inframe deletion in exon 3 of the OPG gene caused an idiopathic hyperphosphatasia phenotype. The deletion caused the loss of an aspartate residue at position 182 of OPG, resulting in an inactivating mutation in OPG. Serum OPG levels in patients were comparable to those of normal controls. These results confirm the importance of OPG in bone physiology in humans as well. Yano et al. [37] reported that serum OPG levels tend to increase with age, and the levels were significantly higher in postmenopausal women with osteoporosis than in agematched controls. These results suggest a potential role of OPG in the protection from enhanced or accelerated osteoclastic bone resorption. Analysis of the domain-deletion mutants of OPG revealed that cysteine-rich domains (CRDs) (I–IV), but not death domain homologous (DDH) regions, are essential for the biological activity in vitro [38] (Fig. 4). A heparin-binding site is located in the C-terminal region, and the binding ability of OPG to heparin does not correlate with its osteoclastogenesis inhibitory activity [38]. A cysteine residue at position 400 of OPG is responsible for dimer formation of OPG [38]. Dimerization of OPG is not required for its biological activity in vitro [19,38] and in vivo [25], although the homodimeric form of OPG exerted a more potent hypocalcemic activity in vivo than did the monomeric form [25]. OPG is the first example of two DDHs in a single polypeptide. Organization of the OPG gene in both humans and mice suggests that the tandemly presented DDHs are produced by a duplication of an exon and that there was no potential transmembrane (TM) coding region in the third intron that intervenes between the CRDs and DDHs [39,40]. When an OPG-Fas fusion protein, in which the TM domain of Fas was inserted between the CRDs and DDHs in OPG, was expressed in 293-EBNA human kidney cell line, the fusion protein had the potential to mediate apoptosis [38]. It is curious that a soluble secreted protein, OPG, has two active duplicated DDHs that are not essential for osteoclastogenesis inhibitory activity. Whether the DDHs are physiologically involved in interaction with other proteins or cell death signalings should be elucidated in future experiments.

HISATAKA YASUDA, KANJI HIGASHIO, AND TATSUO SUDA

B. Identification of the Long-Sought-After Ligand “ODF” 1. AN OPG-BINDING PROTEIN AS A POSSIBLE CANDIDATE FOR ODF

A mouse bone marrow stromal cell line, ST2, is known to support osteoclast formation from mouse spleen cells in the presence of 1,25(OH)2D3 and dexamethasone (Dex) [41]. OPG bound to a single class of high-affinity binding sites induced by 1α,25(OH)2D3 and Dex in ST2 cells [21]. When the binding sites on the treated ST2 cells were occupied by OPG, the cells failed to support osteoclast formation from spleen cells. The time course of increase in the binding sites coincided with that of osteoclast formation. These results strongly suggested that the sites are involved in cellto-cell signaling between stromal cells and osteoclast progenitors, and that OPG inhibits osteoclastogenesis by interrupting the signaling through its binding sites. Cross-linking studies using radioactive OPG revealed that a 40-kDa protein induced on the treated ST2 cells binds to OPG [21]. Taken together, these results raised the possibility that the 40-kDa OPG-binding protein is a ligand for OPG, and is identical to ODF. Since members of the TNF receptor family bind to ligands of the TNF family, we assumed that ODF could be a novel member of the TNF ligand family. 2. MOLECULAR CLONING OF ODF

To identify the 40-kDa OPG binding protein, we screened a cDNA expression library of ST2 cells treated with 1,25(OH)2D3 and Dex using radioactive OPG. A cDNA clone encoding 316 amino acid residues (Mr 36-kDa) was isolated (Fig. 5A, B) [42]. Hydropathy analysis showed the absence of a signal sequence and the presence of an internal 24-residue hydrophobic domain, which presumably represents a TM domain. This structure was typical of a type II TM protein with an extracellular C-terminal region. A homology search of the GenBank sequence database revealed that the C-terminal 165 residues of the protein had significant homology to the extracellular domains of the TNF ligand family members. Coimmunoprecipitation of ODF–OPG complex with anti-OPG antibody demonstrated that ODF was a 40-kDa protein [42] that had been observed in the cross-linking studies [21]. As described later, the protein satisfied the major criteria for ODF in view of its biological activity and of the fact that its expression was regulated by bone-resorbing factors. We therefore named the protein ODF. Lacey et al. [43] also cloned the same molecule independently, calling it OPG ligand (OPGL). ODF/OPGL was found to be identical to TNF-related

671

CHAPTER 38 Vitamin D and Osteoclastogenesis

A

B C-terminal 316 316

sRANKL 152 Extracellular domain

76 71

Plasma membrane Cytoplasmic domain

C

48 1 N-terminal

D 2.5 600 2.0 400

1.5 1.0

200 0.5 0

M-CSF OPG

ND − −

0 + −

+ 10

COSRANKL

+ 100

+ −

(ng/ml)

TRAP-positive cells formed/well

3.0 125 I−Calcitonin binding (cpm × 10−3/well; )

TRAP-positive cells formed/well ( )

800

1500

750

0

M-CSF

−

+

−

+

sRANKL

−

−

+

+

COSVec

FIGURE 5 Structure and biological activity of mouse RANKL/ODF. (A) Amino acid sequence of RANKL/ODF. RANKL/ODF consists of 316 amino acid residues. Dotted line indicates the transmembrane domain (TM). Underlines indicate the potential N-glycosylation sites. (B) Schematic structure of RANKL/ODF. RANKL/ODF is a type II transmembrane protein with a short N-terminal cytoplasmic domain (Met1–Arg47), a single TM (Ser48–Phe71; closed box), and a longer C-terminal extracellular domain (Arg72–Asp316). The striped area represents the homologous C-terminal region (Asp152–Asp316) in the TNF ligand family. sRANKL is prepared by fusing the C-terminal region (Asp76–Asp316) of RANKL/ODF to the C-terminal end of thioredoxin. (C) Induction of osteoclasts from spleen cells by the fixed RANKL-expressing COS cells. COS-7 cells transfected with the RANKL expression vector (COSRANKL) or the empty vector (COSVec) were cultured for 2 days on cover slips in 24-well plates, fixed with paraformaldehyde, and washed with phosphate-buffered saline. Mouse spleen cells (7 × 105 cells) were cultured on fixed cells in the presence or absence of 10 ng/ml M-CSF and the indicated concentrations of OPG. After culturing for 6 days, the cells were subjected to TRAP staining (open box), and calcitonin binding (closed box). Data are expressed as the means ± SD of six cultures. ND, not determined. (D) sRANKL induces osteoclasts from spleen cells. Spleen cells were cultured in the presence or absence of M-CSF and/or sRANKL for a week, and then the cells were fixed and stained with TRAP.

672 activation-induced cytokine (TRANCE) [44] and receptor activator of NF-κB ligand (RANKL) [45], which were cloned as factors regulating T-cell and dendritic cell functions. As a standard nomenclature of the same molecule (ODF/OPGL/TRANCE/RANKL), RANKL was proposed by the ASBMR President’s Committee on Nomenclature [23,24]. 3. RANKL/ODF DIFFERENTIATES OSTEOCLAST PROGENITORS TO OSTEOCLASTS

To examine whether RANKL mediates cell-to-cell signals responsible for osteoclastogenesis, we carried out an in vitro osteoclast formation assay by evaluating tartrate-resistant acid phosphatase (TRAP) activity and calcitonin binding, a combination of which is unique to osteoclasts (Fig. 5C) [42]. When COS-7 cells expressing RANKL (COSRANKL) or control COS-7 cells transfected with the empty vector (COSVec) were fixed with paraformaldehyde, then mouse spleen cells were cultured on the fixed cells for 6 days in the presence of M-CSF, TRAP-positive and calcitonin receptor-positive cells appeared on the COSRANKL cells, but not on the COSVec cells. Concurrent addition of OPG to the cultures inhibited the formation of TRAP-positive and calcitonin receptor-positive cells in a dose-dependent manner [42]. These results indicate that RANKL mediates the cell-to-cell signaling essential for osteoclastogenesis. M-CSF was indispensable for the RANKL-mediated osteoclast formation. To further examine the biological effect of RANKL, we produced a genetically engineered soluble RANKL (sRANKL) [42]. sRANKL together with M-CSF induced osteoclasts from spleen cells, and OPG negated the effect of sRANKL (Fig. 5D). Neither osteoblasts/stromal cells nor bone-resorbing factors were required for osteoclast formation. Autoradiography using radioactive calcitonin confirmed the presence of calcitonin receptors on the induced TRAP-positive cells. Furthermore, when these osteoclasts were cultured on dentine slices for 3 days in the presence of sRANKL and M-CSF, numerous resorption pits were formed on the slices. Taken together, these results established that RANKL mediates an essential signal to osteoclast progenitors for their differentiation into active osteoclasts in the presence of M-CSF. Mature monocytes and alveolar macrophages as well as several cell lines of the macrophage lineage can differentiate into osteoclasts, when cocultured with stromal cells in the presence of 1,25(OH)2D3 [14,46]. A macrophage cell line, C7, is also capable of differentiating into osteoclasts in such a coculture system [47]. sRANKL dose-dependently induced the formation of TRAP- and calcitonin receptor-positive multinucleated cells from C7 cells, in the presence of M-CSF, indicating

HISATAKA YASUDA, KANJI HIGASHIO, AND TATSUO SUDA

that RANKL acts directly on osteoclast progenitors [42]. The TRAP-positive multinucleated cells formed from C7 cells also produced numerous resorption pits on dentine slices [42]. To elucidate the mechanism of human osteoclastogenesis, we further examined the effect of sRANKL on human peripheral blood mononuclear cells (PBMCs). Treatment of human PBMCs with mouse sRANKL together with human M-CSF induced osteoclasts, which were TRAP-, vitronectin receptor-, and calcitonin receptor-positive and were capable of resorbing bone [48]. Quinn et al. [49] confirmed that a combination of mouse sRANKL and human M-CSF was sufficient for both human and mouse osteoclast formation in vitro. These results suggest that the model for osteoclastogenesis in mice was also applicable to humans [13,42,50,51]. While M-CSF is indispensable for the RANKL-mediated osteoclast formation, RANKL appears to be the most critical factor for osteoclastogenesis, because it is up-regulated by various bone-resorbing factors [42]. 4. RANKL/ODF TRANSDUCES AN ESSENTIAL SIGNAL FOR BONE RESORPTION IN THE MICROENVIRONMENT OF BONE

To explore the role of RANKL in bone tissues, we examined the effect of sRANKL and OPG on bone resorption in a fetal mouse long bone culture system [52]. sRANKL markedly stimulated 45Ca release from the prelabeled bone in a dose-dependent manner, and OPG abolished the effect of sRANKL, suggesting that RANKL plays a role in osteoclastogenesis in the microenvironment of bone. To further elucidate whether RANKL-mediated signals are involved in bone resorption, we examined the effects of OPG and anti-RANKL polyclonal antibody (Ab) on bone resorption stimulated by various bone-resorbing factors in the organ culture system [52]. IL-1, 1,25(OH)2D3, PGE2, and PTH significantly stimulated 45Ca release from the prelabeled bone. Both OPG and anti-RANKL Ab completely abolished the effect of these stimulators [52]. These results established that RANKL plays a critical role in bone resorption induced by various bone-resorbing factors in the microenvironment of bone. RANKL most likely activates osteoclasts to resorb bone in this system. Taken together with the findings that RANKL expression is up-regulated in osteoblastic cells by various boneresorbing factors such as 1,25(OH)2D3, PGE2, PTH, and IL-11 [42], these results suggest that the stimulation of bone resorption with these factors is mediated through up-regulation of RANKL expression on the membrane of osteoblasts/stromal cells. The hypercalcemic effect of RANKL in mice [43,53], which is the mirror image of the hypocalcemic effect of OPG [25,27], together with

673

CHAPTER 38 Vitamin D and Osteoclastogenesis

the RANKL-mediated bone resorption in the organ cultures, suggests that RANKL also supports the survival and activation of osteoclasts in vivo. These observations also raise the possibility that systemic hormones, such as 1,25(OH)2D3 and PTH, strictly control serum calcium levels through modulation of bone resorption by RANKL and OPG in the microenvironment of bone. Although the administration of sRANKL to mice for 3 days did not increase the number of osteoclasts but activated osteoclasts to resorb bone [43], we recently showed that TG mice overexpressing sRANKL exhibited severe osteoporosis with an increase of osteoclasts [54]. The results indicate that sRANKL induces osteoclast differentiation in vivo as well. RANKL-KO mice exhibit osteopetrosis and do not produce osteoclasts, indicating that RANKL is essential for osteoclast development [55,56]. RANKL-KO mice, however, have normal osteoclast progenitors that can differentiate into osteoclasts when cocultured with normal osteoblasts/stromal cells. The lack of osteoclasts in RANKL-KO mice results from the inability of osteoblasts/stromal cells to support osteoclastogenesis, and not from an intrinsic block in osteoclast development. In addition, RANKL-KO mice completely lack lymph nodes and have a defect in early differentiation of T and B lymphocytes. These results suggest that RANKL is an absolute requirement for osteoclast development and lymph node organogenesis, and that it plays an important role in lymphocyte differentiation as well.

A

C. Identification of a RANKL/ODF Receptor as Receptor Activator of NF-κB (RANK) 1. RANK IS A SIGNALING RECEPTOR FOR RANKL/ODF IN IN VITRO OSTEOCLASTOGENESIS

RANKL directly binds to osteoclast progenitors, suggesting the presence of a membrane-bound receptor for RANKL on the cells [42,43]. Although it was known that RANK, a novel member of TNF receptor family, was a receptor for RANKL in T-cell and dendritic cell interaction [45], the receptor responsible for the RANKL/ ODF-mediated osteoclastogenesis was not identified. Some ligands of the TNF family bind to several receptors of the TNF receptor family. For example, TNF binds to TNFRI and TNFRII, and TRAIL binds to DR4, DR5, DcR1, and DcR2. It was suspected that RANKL might bind to another member of the TNF receptor family, but not to RANK. To understand the RANKL-mediated signal transduction mechanism in osteoclastogenesis, we molecularly cloned the RANKL receptor from a cDNA expression library of mouse osteoclast progenitors by panning [57]. Nucleotide sequence analysis of 11 positive clones revealed that all of them encoded RANK. sRANKL specifically bound to COS cells expressing RANK. A polyclonal Ab against soluble RANK (sRANK) consisting of the extracellular domain of RANK (anti-RANK Ab) induced osteoclastogenesis in the presence of M-CSF (Fig. 6). Namely, anti-RANK Ab mimicked the RANKL function, suggesting that the clustering of RANK was required for

B sRANK

TRAP activity (OD 405 nm)

0.5 OPG

0.4

Anti-RANK Ab

sRANKL

0.3 0.2 0.1

0 Anti-RANK Ab

−

+

−

−

−

−

−

sRANKL

−

−

+

+

+

+

+

sRANK

−

−

−

+

−

−

−

OPG

−

−

−

−

+

−

−

Anti-RANK Fab − − Fab-control

− −

− −

− −

− −

+ −

− +

RANK

Signal

Anti-RANK Fab Osteoclast progenitor

FIGURE 6 Identification of a RANKL receptor as RANK in the in vitro osteoclastogenesis. (A) Mouse spleen cells were cultured in the presence of M-CSF and various factors for a week. Osteoclast formation was shown by TRAP activity measured using TRAP solution assay. Anti-RANK Ab, a polyclonal antibody against sRANK; anti-RANK Fab, Fab fragment of anti-RANK Ab; Fab control, Fab fragment of control Ab. (B) An illustration demonstrating that RANK is a signaling receptor for RANKL in the in vitro osteoclastogenesis. Both sRANKL and anti-RANK Ab bind to RANK to transduce signals in osteoclast progenitors for their differentiation. OPG and sRANK bind to sRANKL and block the sRANKL-mediated osteoclastogenesis. In contrast, anti-RANK Fab binds to RANK and block the interaction between sRANKL and RANK, resulting in inhibition of osteoclastogenesis.

674 the RANKL/RANK-mediated signaling for osteoclastogenesis. In contrast, both sRANK and Fab fragment of anti-RANK Ab (anti-RANK Fab) completely inhibited RANKL-mediated osteoclastogenesis by binding to RANKL and RANK, respectively. While OPG inhibited RANKL-mediated osteoclastogenesis by interrupting the binding of RANKL to RANK, it had no effect on anti RANK-Ab-mediated osteoclastogenesis. Taken together, these results provide the first evidence that RANK is the sole signaling receptor essential for in vitro RANKL-mediated osteoclastogenesis and that OPG acts as a decoy receptor for RANKL to compete against RANK [57]. 2. RANK IS THE SOLE RECEPTOR FOR RANKL IN VIVO

Hsu et al. [58] made TG mice overexpressing sRANK and demonstrated that the mice exhibited osteopetrosis similar to OPG-TG mice. They predicted that RANK was a receptor for RANKL in vivo. Dougall et al. [59] and Li et al. [60] independently reported the evidence that RANK was the receptor for RANKL in vivo by analyzing RANK-KO mice that exhibited almost identical phenotype to that of RANKL-KO mice. RANK-KO mice showed severe osteopetrosis with no osteoclasts. These mice also lacked all lymph nodes and showed defects in early differentiation of B cells. The only difference between RANK-KO mice and RANKL-KO mice was that the former mice showed no defect in T cells and thymus. These results suggest that RANK is the exclusive receptor for RANKL in vivo except for T cell differentiation. The reasons why there is a difference between the two types of KO mice remain to be elucidated in the future, but may be developmental in nature. Gain-of-function mutations in the RANK gene have been identified in patients with familial expansile osteolysis (FEO) and familial Paget’s disease by Hughes et al. [61]. A duplication of 18 or 27 bases in the signal peptide region caused lack of normal cleavage of the signal peptide, resulting in an increase in the RANKmediated NF-κB signaling and osteoclastogenesis. More recently, Whyte and Hughes [62] reported that similar gain-of-function mutation in the RANK gene caused expansile skeletal hyperphosphatasia. These results also confirm the importance of the RANKL/ RANK/OPG system in bone physiology in humans.

D. Nomenclature The nomenclature of the ligand, receptor, and decoy receptor of the newly discovered TNF ligand/receptor family members has been summarized in Fig. 7

HISATAKA YASUDA, KANJI HIGASHIO, AND TATSUO SUDA

according to the recommendation of the ASBMR President’s Committee on Nomenclature [23,24]. RANKL [45], also called ODF [42], OPGL [43], and TRANCE [44], is a new member of the membranebound TNF ligand family (TNFSF11), and it is important for osteoclast development as well as lymphocyte and dendritic cell development. RANK [45], a new member of the membrane-bound TNF receptor family (TNFRSF11A), has been cloned as a receptor for RANKL in immune systems, and it is the signaling receptor for RANKL in osteoclastogenesis as well [57,59,60]. OPG [22], also called OCIF [19,21], is a new soluble member of the TNF receptor family (TNFRSF11B), and it functions as a decoy receptor for RANKL [57]. Figure 8 summarizes the molecular mechanisms of osteoclast formation and activation [21,50,51]. Osteoblasts/stromal cells play an essential role in osteoclastogenesis through the expression of RANKL on the membrane induced by various bone-resorbing factors such as 1,25(OH)2D3, PGE2, PTH, and IL-11. RANKL recognizes osteoclast progenitors which express RANK through a mechanism that involves cell-to-cell contact. M-CSF produced by osteoblasts/ stromal cells is also indispensable for the differentiation of osteoclast progenitors. Osteoclast progenitors differentiate into osteoclasts by binding to RANKL on the osteoblasts/stromal cells. When OPG, a decoy receptor for RANKL, blocks RANKL, osteoclast progenitors expressing RANK are unable to bind to RANKL, leading to inhibition of osteoclast formation. Some factors such as TGFβ that are released from bone by osteoclastic bone resorption may induce OPG to inhibit osteoclastogenesis as a negative feedback loop (see the next section). TNF also induces osteoclastogenesis by a mechanism independent of the RANKL-RANK interaction [63,64] and in concert with RANKL as well [65].

E. Regulation of RANKL and OPG Expression It is now widely accepted that RANKL, RANK, and OPG are three important molecules in osteoclast differentiation and activation in vivo. The next question is how these genes are regulated under physiological and pathological conditions. OPG gene expression in osteoblasts/stromal cells is up-regulated by Ca2+ and is down-regulated by 1,25(OH)2D3 and Dex [21], a combination of which supports osteoclastogenesis when these cells are cocultured with spleen cells [41]. In addition, transforming growth factor β (TGF-β) inhibits osteoclastogenesis through induction of OPG expression by bone marrow stromal cells [66–68].

675

CHAPTER 38 Vitamin D and Osteoclastogenesis

Osteoclast progenitors Osteoclasts

Osteoblasts/stromal cells

OPG

RANKL

RANK

Extracellular

Ligand RANKL

(Receptor activator of NF-κB ligand)

ODF

(Osteoclast differentiation factor)

OPGL

(Osteoprotegerin ligand)

TRANCE

(TNF-related activation-induced cytokine)

TNFSF11

(TNF superfamily 11)

Receptor RANK TNFRSF11A

(Receptor activator of NF-κB) (TNF receptor superfamily 11A)

Decoy receptor OPG

(Osteoprotegerin)

OCIF

(Osteoclastogenesis inhibitory factor)

TNFRSF11B

(TNF receptor superfamily 11B)

FIGURE 7

Nomenclature of the ligand, receptor, and decoy receptor of the newly discovered TNF ligand/receptor family members.

These observations raised the possibility that osteoclastogenesis is regulated by OPG, which is produced as a local factor by osteoblasts/stromal cells in response to osteotropic factors, cytokines, or Ca2+ released at bone-resorbing sites. A report that OPG produced by osteoblasts/stromal cells is important for regulation of osteoclastogenesis [69] strongly supports this possibility. Up-regulation of the RANKL gene expression was observed in ST2 cells treated with 1,25(OH)2D3 and Dex as well as in mouse primary osteoblasts cultured in the presence of such bone resorbing factors as 1,25(OH)2D3, PTH, PGE2, and IL-11 [42]. It is known that RANK expression is regulated, not at the transcriptional level, but at the posttranscriptional level [45]. It has been suggested that the ratio of RANKL to OPG

produced in the microenvironment of bone is important for osteoclastogenesis [70]. Table I summarizes the various bone-resorbing factors that regulate RANKL and OPG expression. They are osteotropic factors (1,25(OH)2D3 [21,70], PTH [21,70,71], PGE2 [21,72], and IL-11 [21,70], IL-6 + soluble IL-6 receptor [sR] [73], OSM [73]), hormones (glucocorticoids [66,74,75] and estrogen [76]), inflammatory cytokines (IL-1 [66,77,78] and TNF [78,79]), factors released from bones by osteoclastic resorption (Ca2+ [21,80], TGF-β [66–68], fibroblast growth factor 2 [FGF-2] [81,82], and BMP-2 [83,84]). Although these factors regulate the RANKL and OPG expression in osteoblastic cells in vitro, the mechanism in which both genes are regulated in vivo is not completely identified.

676

HISATAKA YASUDA, KANJI HIGASHIO, AND TATSUO SUDA

Osteoclast progenitors TNF

Activated osteoclasts

M-CSF Differentiation & activation

c-Fms

RANK sRANKL

OPG RANKL

1,25(OH)2D3 PTH/PGE2/IL-1 IL-6 family

RANKL

TGF-β

Osteoblasts/stromal cells

FIGURE 8 Possible signal transduction pathways regulating osteoclast differentiation and activation.

The master factors regulating the RANKL and OPG expression in bones may be different in each metabolic bone disease, and their identification is needed for clarification of the mechanism of each pathogeny.

F. Clinical Trials Phase I studies of OPG, an Fc-OPG construct (constant fragment of IgG fused to human OPG), suggest that a single subcutaneous dose of OPG is effective in decreasing the levels of bone turnover markers for several days in healthy postmenoposal women [85] TABLE I Effects of Bone-Resorbing Factors on the Expression of RANKL and OPG in Osteoblastic Cells Bone-resorbing factors 1α,25(OH)2D3 PGE2 PTH IL-6+sR OSM IL-11 Glucocorticoid Estrogen IL-1β TNF Ca2+ TGF-β FGF-2 BMP-2

RANKL

OPG

— — — —

Reference [21,70] [21,72] [21,70,71] [73] [73] [21,70] [64,74,75] [76] [66,77,78] [78,79] [21,80] [66–68] [81,82] [83,84]

and patients with breast carcinoma–related bone metastases or multiple myeloma [86]. Peterson et al. [87] have reported that a fully human monoclonal Ab against RANKL rapidly and profoundly suppressed bone resorption in cynomolgus monkeys. These results strongly suggest that inhibitors of RANKL-RANK signaling such as OPG and anti-RANKL Ab are useful and applicable to the treatment of metabolic bone diseases such as osteoporosis, rheumatoid arthritis (RA), and metastatic bone diseases. At present anti-TNF therapy using recombinant soluble receptors (etanercept) and monoclonal Ab (infliximab) to TNF is available for clinical use, and it works for about two out of three adults with RA [88]. This therapy reduces the inflammation associated with RA and may inhibit osteoclastogenesis as well, since TNF was able to induce osteoclastogenesis by a mechanism independent of the RANKL–RANK interaction [63,64] and also in concert with RANKL [65].

V. SIGNAL TRANSDUCTION PATHWAYS OF THE RANKL–RANK SYSTEM IN OSTEOCLAST DEVELOPMENT A. TNF Receptor–Associated Factor (TRAF) Family Proteins TRAF family proteins are adaptor molecules that mediate intracellular signaling of cytokine receptors including the TNF receptor family and Toll/IL-1 receptor family. Six members of the TRAF family have been identified. Although several in vitro studies showed that TRAF1, 2, 3, 5, and 6 bound to the intracellular region of RANK [58,89,90], the most important one is TRAF6. No abnormalities were observed in the bones and

677

CHAPTER 38 Vitamin D and Osteoclastogenesis

osteoclasts in deficient mice of TRAF2, 3, or 5. Naito et al. [91] and Lomaga et al. [92], however, independently reported that TRAF6-KO mice exhibited osteopetrosis with a defect of osteoclast differentiation and function, respectively. In TRAF6-KO mice generated by Naito et al., osteoclasts were totally absent in bone. In contrast, in TRAF6-KO mice generated by Lomaga et al., osteoclasts were present in bone, but they failed to resorb bone. These findings establish the importance of TRAF6 in osteoclast differentiation and function in vivo. Kobayashi et al. [93] investigated the functional domains of TRAF6 by deletion mutant analysis and demonstrated that TRAF6 was composed of at least two domains responsible for osteoclast differentiation and function, respectively. The importance of TRAF6 is also shown by the report by Takayanagi et al. [94], in which interferon-γ (IFN-γ) produced by T cells inhibits osteoclast differentiation by degradation of TRAF6. More recently, Kaji et al. [95] reported that TNF-induced osteoclastogenesis also required TRAF6 using TRAF6KO mice. However, the mechanism is unknown, since TRAF6 does not bind to TNF receptors.

B. A Non-receptor-type Tyrosine Kinase, c-Src It is known that c-Src family proteins play important roles in signal transduction mechanisms involved in the

regulation of cell growth, differentiation, migration, and survival. Interestingly, c-src KO mice exhibited skeletal abnormalities with an osteopetrotic phenotype [96]. In c-src-KO-mice, osteoclast differentiation was normal but the bone-resorbing activity of mature osteoclasts was impaired due to a disorganized ruffled border structure Wong et al. [97] reported that upon RANKL binding, RANK recruited c-Src and TRAF6 and composed a complex of RANK-c-Src-TRAF6, resulting in activation of c-Src, phosphatidylinositol 3′-kinase (PI3K) and Akt/protein kinase B (PKB) sequentially (Fig. 9).

C. Fos Family and Jun-N-Terminal Kinase (JNK)/p38 c-Fos is a member of the Fos family including FosB, Fra-1, and Fra-2. A transcription factor, AP-1, is composed of a member of the Fos family and a member of the Jun family including c-Jun, JunB, and JunD. c-fos-KO mice exhibited severe osteopetrosis with no osteoclasts [98,99]. Matsuo et al. [100] reported that c-fos-KO mice overexpressing Fra-1 rescued osteoclast development. RANKL induced c-Fos expression, then Fra-1 expression in a c-Fos dependent manner. These results establish a link between the RANK signaling and the expression of c-Fos family proteins during osteoclast development. Recently, Takayanagi et al. [101] reported that

Osteoblasts/stromal cells RANKL

Osteoclast progenitors

RANK c-Src TRAF6

c-Fos

PI3K

TRAF2

JNK/p38

PKB

IKKα/IKKβ

Fra-1 AP-1(Fos/Jun) AP1 activation

NF-κB/IκB NF-κB activation

IκB degradation Survival

PU.1, MITF, NFAT2

Differentiation & Activation

FIGURE 9

A model illustrating a mechanism by which osteoblasts/stromal cells regulate osteoclast differentiation and activation.

678 RANKL-induced c-Fos expression also promotes an inhibitor for osteoclastogenesis, IFN-β, thus creating a negative feedback loop that regulates osteoclastogenesis. More recently, the important roles of JNK in the RANKL-induced osteoclastogenesis were shown using JNK1-KO mice [102]. In addition, Matsumoto et al. [103] and Li et al. [104] reported the essential role of p38 activation in response to RANKL in osteoclast differentiation. Thus, c-fos expression as well as JNK/p38 activation in response to RANKL plays critical roles in osteoclast development (Fig. 9).

D. NF-κB NF-κB is a homo- or heterodimeric transcription factor composed of subunits including p50, p52, RelA, RelB, and c-Rel. Franzoso et al. [105] and Iotsova et al. [106] independently generated mice deficient in both p50 and p52 subunits of NF-κB. The double KO mice developed osteopetrosis due to a defect in osteoclast differentiation. Bone marrow transplantation recovered the skeletal abnormality of the KO mice, indicating that the osteopetrotic phenotype is due to a cell-autonomous defect in the development of osteoclasts. We demonstrated that RANKL induced NF-κB activation in osteoclast progenitors and osteoclasts [107]. These results suggest that NF-κB activation plays a crucial role in osteoclast differentiation (Fig. 9). It should be noted, however, that activation of NF-κB is not sufficient for osteoclast differentiation, since other cytokines including IL-1 that activate NF-κB cannot induce osteoclast differentiation in the absence of osteoblasts/stromal cells.

E. Other Transcription Factors Tondravi et al. [108] demonstrated that targeted disruption of PU.1, an ETS domain–containing transcription factor, induced osteopetrosis due to the defect in osteoclast differentiation. No mature macrophages were present in the KO mice, indicating that PU.1 was essential for the differentiation of cells of the monocyte/ macrophage lineage. Microphthalmia transcription factor (MITF) was identified as the gene product for the murine mi locus that causes the microphthalmia mutation. Homozygous mutation (mi/mi) induced severe osteopetrosis in mice, since mononuclear osteoclasts failed to differentiate into multinucleated osteoclasts. It has been reported that MITF regulates the expression of TRAP and cathepsin K genes [109,110]. More recently, Ishida et al. [111] and Takayanagi et al. [112] independently reported the importance of

HISATAKA YASUDA, KANJI HIGASHIO, AND TATSUO SUDA

nuclear factor of activated T cells (NFAT) 2, also called NFATc1, on osteoclast differentiation. cDNA microarray analyses revealed that NFAT2 was induced by sRANKL stimulation in osteoclast progenitors. Ectopic expression of NFAT2 in osteoclast progenitors induced osteoclast differentiation without sRANKL. In addition, NFAT2-deficient embryonic stem cells failed to differentiate into osteoclasts in response to RANKL. These results suggest a pivotal role for NFAT2 in osteoclast differentiation (Fig. 9).

F. Skeletal Phenotype Associated with Alterations of the RANKL–RANK Signaling The molecular mechanism of osteoclast formation and activation shown in Fig. 8 was proposed by in vitro studies, but a number of in vivo studies including administration of recombinant proteins such as OPG and sRANKL and generation of TG and KO mice of the genes related to the RANKL–RANK signaling confirmed that this signaling is essential for osteoclast formation and activation in vivo as well (Table II). In addition, loss-of-function mutations of the OPG gene and gain-of-function mutations of the RANK gene were reported in patients with skeletal abnormalities, suggesting that the mechanism is applicable to humans as well.

VI. THE BIOLOGICAL RELEVANCE OF VITAMIN D TO OSTEOCLASTIC BONE RESORPTION It is likely that 1,25(OH)2D3 is a bone-resorbing hormone, but not a bone-forming hormone at least in vitro. It should be emphasized, however, that the in vivo bone-mobilizing effects of 1,25(OH)2D3 depend on dosage. Figure 10 shows the differences in the dose levels of 1,25(OH)2D3 required for inducing intestinal absorption of calcium and bone mineral mobilization activity [3]. In this experiment, graded doses of 1,25(OH)2D3 were administered to rats fed a low calcium, vitamin D–deficient diet. Intestinal absorption of calcium was determined by the routine everted gut sac method, and bone mobilization activity was monitored by measuring serum calcium levels [113]. Intestinal absorption of calcium was stimulated by as little as 0.1 µg/kg bw of 1,25(OH)2D3, but bone mobilization activity was induced only by 10–50 times higher doses of 1,25(OH)2D3. These results indicate that physiological doses of 1,25(OH)2D3 have a minimal impact on bone mobilization. Only pharmacological or toxic doses of 1,25(OH)2D3 appear to induce bone resorption [3].

679

CHAPTER 38 Vitamin D and Osteoclastogenesis

TABLE II

Skeletal Phenotypes Associated with Alterations of the RANKL–RANK Signaling Signaling and bone resorption

OPG KO [33,34]/mutation [35,36] (loss of function) sRANKL administration [44,53] sRANKL TG [54] RANK mutation [61,62] (gain of function) OPG TG [22] OPG administration [21,22] RANKL KO [55,56] RANK KO [59,60] sRANK TG [58] TRAF6 KO [91,92] c-src KO [96] c-fos KO [98,99] p50/p52 KO [105,106]

(Paget’s disease) Up

Down

B

Intestinal Ca transport

Bone Ca mobilization

0

0.25 (p < 0.01)

1,25(OH)2D3 administered

0 1,25(OH)2D3 administered

µg/kg

(p < 0.05)

Osteopetrosis

or toxic doses of 1,25(OH)2D3 induce bone resorption in vivo, but physiological doses of 1,25(OH)2D3 do not (Fig. 10). To investigate the dose-dependent effects of 1,25(OH)2D3 on bone resorption in more detail, we examined in vivo effects of 1,25(OH)2D3 on the expression of RANKL and OPG mRNA in bone of normal

µg/kg

0.1

Osteoporosis (FEO, Paget’s disease)

A significant question is the relationship between vitamin D and PTH in inducing bone resorption in vivo. Of several systemic hormones and local factors affecting bone remodeling, vitamin D and PTH may be the most important factors for regulating bone formation and resorption. In fact, PTH stimulates both bone formation and resorption in vivo. Also, pharmacological

A

Phenotype

0.1

0.25

1.0

1.0

(p < 0.05)

(p < 0.02)

5.0 (p < 0.05) 2

FIGURE 10

4 6 3 5 Serosal/Mucosal ratio

7

4

7 5 6 Plasma calcium

8 9 (mg/dl)

Dose-response effects of 1,25(OH)2D3 on intestinal calcium transport and bone calcium mobilization activities in vivo. Rats were fed a vitamin D–deficient, low-calcium diet for 3 weeks, then received graded doses of 1,25(OH)2D3. Twenty-four hours later, intestinal calcium transport activity (A) and plasma calcium levels (B) were measured.

680

HISATAKA YASUDA, KANJI HIGASHIO, AND TATSUO SUDA

and TPTX rats. When rats were fed a low-calcium diet for 3 weeks, serum calcium levels were reduced from 10.4 to 7.5 mg/dl within 3 weeks. Concomitantly, they showed much higher expression of RANKL mRNA in bone, compared with rats maintained on an adequate calcium diet. Expression of OPG mRNA in bone did not change appreciably during 1 to 3 weeks of the feeding period. The rise in the expression of bone RANKL mRNA in rats fed a low-calcium diet appeared to be due to hyperparathyroidism induced by low-calcium feeding [114]. Thus, similar experiments were conducted in TPTX rats. TPTX decreased the serum calcium levels from 9.6 to 5.1 mg/dl 24 hr after the surgery (Table III). Concomitantly, TPTX greatly reduced the expression of RANKL mRNA in bone. Again, the OPG mRNA expression was not changed appreciably by TPTX. Daily oral administration of 0.01 or 0.1 µg/kg bw of 1,25(OH)2D3 for 2 weeks had no effect on the expression of RANKL mRNA in bone, but 0.5 µg/kg of daily administration of 1,25(OH)2D3 markedly increased both the expression of bone RANKL mRNA and the serum calcium levels. These results indicate that only pharmacological or toxic doses of 1,25(OH)2D3 induce bone resorption in vivo [114]. To further examine the in vivo relationships between vitamin D and PTH, we examined the effects of 1,25(OH)2D3 on the expression of RANKL mRNA in bone in normocalcemic TPTX rats [114]. In these TPTX rats, serum calcium was maintained nearly normocalcemic by constantly infusing PTH at a rate of 50 ng/hr with an osmotic mini-pump (Table III). These animals had a constant serum PTH level of about 20 pg/ml, which was equivalent to those in normal rats. They were daily given orally 0.01 to 0.5 µg/kg of 1,25(OH)2D3, and gene expression was examined on day 14. Infusion of PTH markedly increased the expression of RANKL mRNA in bone of TPTX rats. To our surprise, daily administration of 0.01 or 0.1 µg/kg of 1,25(OH)2D3 suppressed PTH-induced bone expression of RANKL mRNA. A higher dose of 0.5 µg/kg of TABLE III

1,25(OH)2D3, however, increased both the expression of bone RANKL mRNA and serum calcium levels (Table III). These results suggest that a certain range of physiological doses of 1,25(OH)2D3 suppress RANKL mRNA expression in bone, independent of the suppression of PTH secretion [3,114]. In in vitro conditions, both vitamin D and PTH are capable of inducing bone resorption independently. In contrast, under in vivo conditions, PTH and vitamin D exhibit biological activities in concert with each other in inducing bone resorption. PTH is required for inducing bone-resorbing effects of vitamin D, and vice versa. It is also interesting that some physiological dose levels of 1,25(OH)2D3 are capable of suppressing bone resorption by inhibiting RANKL mRNA expression in bone of normocalcemic TPTX rats constantly infused with PTH. How physiological doses of 1,25(OH)2D3 inhibit PTH-induced bone resorption needs further investigation.

VII. CONCLUSION Physiological doses of 1,25(OH)2D3 do not stimulate bone resorption in vivo, but rather inhibit PTHinduced bone resorption. In order for 1,25(OH)2D3 to induce bone resorption, pharmacological or toxic doses of 1,25(OH)2D3 are required (Fig. 11). Physiological doses of 1,25(OH)2D3 preferentially stimulate intestinal absorption of calcium without inducing bone resorption, which then stimulates bone mineralization (Fig. 11). These results support the concept that physiological doses of vitamin D compounds are useful for the treatment of various metabolic bone diseases such as osteoporosis and secondary hyperparathyroidism. Like other bone-resorbing hormones and cytokines, pharmacological or toxic doses of 1,25(OH)2D3 stimulate bone resorption. RANKL is required for all the steps of osteoclast development: differentiation, fusion, survival, and activation. RANKL mediates signals for osteoclastogenesis through RANK. OPG inhibits the

Dose-Dependent Effects of 1,25(OH)2D3 Orally Administered on Serum Ca and RANKL mRNA Expression in Bone Sham rats

PTH continuously infused (50 ng/hr) 1,25(OH)2D3 orally administered (µg/kg/day) Serum Ca (mg/dl) RANKL mRNA expression in bone

— — 9.6 ±

Adapted from Suda T et al. 2003 Vitamin D and bone J Cell Biochem 88:259–266.

TPTX rats − − 5.1 −

+ − 10.4 ++

+ 0.01 11.1 +

+ 0.1 11.2 +

+ 0.5 14.6 +++

681

CHAPTER 38 Vitamin D and Osteoclastogenesis

1,25(OH)2D3 (Physiological doses)

Intestinal Ca absorption

Suppression of PTH secretion

Minera lization Plasma Ca (9–10 mg/dl)

Bone resorption Intestine Bone 1,25(OH)2D3 (Pharmacological or toxic doses)

Kidney

FIGURE 11 A hypothetical concept of the dose-dependent effects of 1,25(OH)2D3 on intestinal calcium transport and bone resorption activities in vivo.

whole processes of osteoclastogenesis as a decoy receptor by interrupting the binding of RANKL to RANK. The discovery of RANKL, OPG, and RANK opens a new area of research on osteoclast biology. Further studies on these molecules and the RANKL– RANK signal transduction pathways will establish new ways for treating several metabolic bone diseases caused by abnormal osteoclast differentiation and function.

References 1. DeLuca HF 1988 The vitamin D story: a collaborative effort of basic science and clinical medicine. FASEB J 2:224–236. 2. Reichel H, Koeffler HP, Norman AW 1989 The role of the vitamin D endocrine system in health and disease. N Engl J Med 320:980–991. 3. Suda T, Ueno Y, Fujii K, Shinki T 2003 Vitamin D and bone. J Cell Biochem 88:259–266. 4. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Genet 16:391–396. 5. Carlsson A (1952) Tracer experiments on the effects of vitamin D on the skeletal metabolism of calcium and phosphorus. Acta Physiol Scan 26:212–220. 6. Raisz LG, Trummel CL, Holick MF, DeLuca HF 1972 1,25Dihydroxycholecalciferol: a potent stimulator of bone resorption in tissue culture. Science 175:768–769. 7. Abe E, Miyaura C, Sakagami H, Takeda M, Konno K, Yamazaki T, Yoshiki S, Suda T 1981 Differentiation of mouse myeloid leukemia cells induced by 1α,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 78:4990–4994.

8. Tanaka H, Abe E, Miyaura C, Kuribayashi T, Konno K, Nishii Y, Suda T 1982 1α,25-Dihydroxycholecalciferol and a human myeloid leukaemia cell line (HL-60). Biochem J 204: 713–719. 9. Abe E, Miyaura C, Tanaka H, Shiina Y, Kuribayashi T, Suda S, Nishii Y, DeLuca HF, Suda T 1983 1α,25-Dihydroxyvitamin D3 promotes fusion of mouse alveolar macrophages both by a direct mechanism and by a spleen cell-mediated indirect mechanism. Proc Natl Acad Sci USA 80:5583–5587. 10. Rodan GA, Martin TJ 1981 Role of osteoblasts in hormonal control of bone resorption—a hypothesis. Calcif Tissue Int 33:349–351. 11. Chambers TJ 1992 Regulation of osteoclast development and function. In: Rifkin BR, Gay CV eds. Biology and Physiology of the Osteoclast. CRC Press, Boca Raton, FL, pp.105–128. 12. Takahashi N, Akatsu T, Sasaki T, Nicholson GC, Moseley JM, Martin TJ, Suda T 1988 Induction of calcitonin receptors by 1α,25-dihydroxyvitamin D3 in osteoclast-like multinucleated cells formed from mouse bone marrow cells. Endocrinology 123:1504–1510. 13. Suda T, Takahashi N, Martin TJ 1992 Modulation of osteoclast differentiation. Endocr Rev 13:66–80. 14. Chambers TJ, Owens JM, Hattersley G, Jat PS, Noble MD 1993 Generation of osteoclast-inductive and osteoclastogenic cell lines from the H-2KbtsA58 transgenic mouse. Proc Natl Acad Sci USA 90:5578–5582. 15. Takahashi N, Udagawa N, Akatsu T, Tanaka H, Isogai Y, Suda T 1991 Deficiency of osteoclasts in osteopetrotic mice is due to a defect in the local microenvironment provided by osteoblastic cells. Endocrinology 128:1792–1796. 16. Tanaka S, Takahashi N, Udagawa N, Tamura T, Akatsu T, Stanley ER, Kurokawa T, Suda T 1993 Macrophage colonystimulating factor is indispensable for both proliferation and differentiation of osteoclast progenitors. J Clin Invest 91:257–263. 17. Yoshida H, Hayashi S, Kunisada T, Ogawa M, Nishikawa S, Okamura H, Sudo T, Shultz LD 1990 The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature 345:442–444.

682 18. Suda T, Udagawa N, Nakamura I, Miyaura C, Takahashi N 1995 Modulation of osteoclast differentiation by local factors. Bone 17:87S–91S. 19. Tsuda E, Goto M, Mochizuki S, Yano K, Kobayashi F, Morinaga T, Higashio K 1997 Isolation of a novel cytokine from human fibroblasts that specifically inhibits osteoclastogenesis. Biochem Biophys Res Commun 234:137–142. 20. Higashio K, Shima N, Goto M, Itagaki Y, Nagao M, Yasuda H, Morinaga T 1990 Identity of a tumor cytotoxic factor from human fibroblasts and hepatocyte growth factor. Biochem Biophys Res Commun 170:397–404. 21. Yasuda H, Shima N, Nakagawa N, Mochizuki SI, Yano K, Fujise N, Sato Y, Goto M, Yamaguchi K, Kuriyama M, Kanno T, Murakami A, Tsuda E, Morinaga T, Higashio K 1998 Identity of osteoclastogenesis inhibitory factor (OCIF) and osteoprotegerin (OPG): A mechanism by which OPG/ OCIF inhibits osteoclastogenesis in vitro. Endocrinology 139:1329–1337. 22. Simonet WS, Lacey DL, Dunstan CR, Kelley M, Chang MS, Luthy R, Nguyen HQ, Wooden S, Bennett L, Boone T, Shimamoto G, DeRose M, Elliott R, Colombero A, Tan HL, Trail G, Sullivan J, Davy E, Bucay N, Renshaw-Gegg L, Hughes TM, Hill D, Pattison W, Campbell P, Boyle WJ 1997 Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 89:309–319. 23. Riggs BL, Baron R, Boyle WJ, Drezner M, Manolagas S, Martin TJ, Stewart AF, Suda T, Yasuda H, Aubin J, Goltzman D 2000 Proposed standard nomenclature for new tumor necrosis factor family members involved in the regulation of bone resorption. The American Society for Bone and Mineral Research President’s Committee on Nomenclature. J Bone Miner Res 15:2293–2296. 24. Riggs BL, Baron R, Boyle WJ, Drezner M, Manolagas S, Martin TJ, Stewart AF, Suda T, Yasuda H, Aubin J, Goltzman D 2000 Proposed standard nomenclature for new tumor necrosis factor family members involved in the regulation of bone resorption. The American Society for Bone and Mineral Research President’s Committee on Nomenclature. Bone 27:761–764. 25. Tomoyasu A, Goto M, Fujise N, Mochizuki S, Yasuda H, Morinaga T, Tsuda E, Higashio K 1998 Characterization of monomeric and homodimeric forms of osteoclastogenesis inhibitory factor. Biochem Biophys Res Commun 245: 382–387. 26. Akatsu T, Murakami T, Ono K, Nishikawa M, Tsuda E, Mochizuki SI, Fujise N, Higashio K, Motoyoshi K, Yamamoto M, Nagata N 1998 Osteoclastogenesis inhibitory factor exhibits hypocalcemic effects in normal mice and in hypercalcemic nude mice carrying tumors associated with humoral hypercalcemia of malignancy. Bone 23:495–498. 27. Yamamoto M, Murakami T, Nishikawa M, Tsuda E, Mochizuki S, Higashio K, Akatsu T, Motoyoshi K, Nagata N 1998 Hypocalcemic effect of osteoclastogenesis inhibitory factor/osteoprotegerin in the thyroparathyroidectomized rat. Endocrinology 139:4012–4015. 28. Capparelli C, Kostenuik PJ, Morony S, Starnes C, Weimann B, Van G, Scully S, Qi M, Lacey DL, Dunstan CR 2000 Osteoprotegerin prevents and reverses hypercalcemia in a murine model of humoral hypercalcemia of malignancy. Cancer Res 60:783–787. 29. Morony S, Capparelli C, Lee R, Shimamoto G, Boone T, Lacey DL, Dunstan CR 1999 A chimeric form of osteoprotegerin inhibits hypercalcemia and bone resorption induced by IL-1β, TNF-α, PTH, PTHrP, and 1,25(OH)2D3. J Bone Miner Res 14:1478–1485.

HISATAKA YASUDA, KANJI HIGASHIO, AND TATSUO SUDA

30. Morony S, Capparelli C, Sarosi I, Lacey DL, Dunstan CR, Kostenuik PJ 2001 Osteoprotegerin inhibits osteolysis and decreases skeletal tumor burden in syngeneic and nude mouse models of experimental bone metastasis. Cancer Res 61:4432–4436. 31. Honore P, Luger NM, Sabino MA, Schwei MJ, Rogers SD, Mach DB, O’Keefe PF, Ramnaraine ML, Clohisy DR, Mantyh PW 2000 Osteoprotegerin blocks bone cancerinduced skeletal destruction, skeletal pain and pain-related neurochemical reorganization of the spinal cord. Nat Med 6:521–528. 32. Croucher PI, Shipman CM, Lippitt J, Perry M, Asosingh K, Hijzen A, Brabbs AC, van Beek EJ, Holen I, Skerry TM, Dunstan CR, Russell GR, Van Camp B, Vanderkerken K 2001 Osteoprotegerin inhibits the development of osteolytic bone disease in multiple myeloma. Blood 98:3534–3540. 33. Bucay N, Sarosi I, Dunstan CR, Morony S, Tarpley J, Capparelli C, Scully S, Tan HL, Xu W, Lacey DL, Boyle WJ, Simonet WS 1998 Osteoprotegerin-deficient mice develop early onset osteoporosis and arterial calcification. Genes Dev 12:1260–1268. 34. Mizuno A, Amizuka N, Irie K, Murakami A, Fujise N, Kanno T, Sato Y, Nakagawa N, Yasuda H, Mochizuki S, Gomibuchi T, Yano K, Shima N, Washida N, Tsuda E, Morinaga T, Higashio K, Ozawa H 1998 Severe osteoporosis in mice lacking osteoclastogenesis inhibitory factor/osteoprotegerin. Biochem Biophys Res Commun 247:610–615. 35. Whyte MP, Obrecht SE, Finnegan PM, Jones JL, Podgornik MN, McAlister WH, Mumm S 2002 Osteoprotegerin deficiency and juvenile Paget’s disease. N Engl J Med 347: 175–184. 36. Cundy T, Hegde M, Naot D, Chong B, King A, Wallace R, Mulley J, Love DR, Seidel J, Fawkner M, Banovic T, Callon KE, Grey AB, Reid IR, Middleton-Hardie CA, Cornish J 2002 A mutation in the gene TNFRSF11B encoding osteoprotegerin causes an idiopathic hyperphosphatasia phenotype. Hum Mol Genet 11:2119–2127. 37. Yano K, Tsuda E, Washida N, Kobayashi F, Goto M, Harada A, Ikeda K, Higashio K, Yamada Y 1999 Immunological characterization of circulating osteoprotegerin/osteoclastogenesis inhibitory factor: increased serum concentrations in postmenopausal women with osteoporosis. J Bone Miner Res 14:518–527. 38. Yamaguchi K, Kinosaki M, Goto M, Kobayashi F, Tsuda E, Morinaga T, Higashio K 1998 Characterization of structural domains of human osteoclastogenesis inhibitory factor. J Biol Chem 273:5117–5123. 39. Morinaga T, Nakagawa N, Yasuda H, Tsuda E, Higashio K 1998 Cloning and characterization of the gene encoding human osteoprotegerin/osteoclastogenesis-inhibitory factor. Eur J Biochem 254:685–691. 40. Mizuno A, Murakami A, Nakagawa N, Yasuda H, Tsuda E, Morinaga T, Higashio K 1998 Structure of the mouse osteoclastogenesis inhibitory factor (OCIF) gene and its expression in embryogenesis. Gene 215:339–343. 41. Udagawa N, Takahashi N, Akatsu T, Sasaki T, Yamaguchi A, Kodama H, Martin TJ, Suda T 1989 The bone marrowderived stromal cell lines MC3T3-G2/PA6 and ST2 support osteoclast-like cell differentiation in cocultures with mouse spleen cells. Endocrinology 125:1805–1813. 42. Yasuda H, Shima N, Nakagawa N, Yamaguchi K, Kinosaki M, Mochizuki S, Tomoyasu A, Yano K, Goto M, Murakami A, Tsuda E, Morinaga T, Higashio K, Udagawa N, Takahashi N, Suda T 1998 Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is

CHAPTER 38 Vitamin D and Osteoclastogenesis

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

identical to TRANCE/RANKL. Proc Natl Acad Sci USA 95:3597–3602. Lacey DL, Timms E, Tan HL, Kelley MJ, Dunstan CR, Burgess T, Elliott R, Colombero A, Elliott G, Scully S, Hsu H, Sullivan J, Hawkins N, Davy E, Capparelli C, Eli A, Qian YX, Kaufman S, Sarosi I, Shalhoub V, Senaldi G, Guo J, Delaney J, Boyle WJ 1998 Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 93: 165–176. Wong BR, Rho J, Arron J, Robinson E, Orlinick J, Chao M, Kalachikov S, Cayani E, Bartlett FS, 3rd, Frankel WN, Lee SY, Choi Y 1997 TRANCE is a novel ligand of the tumor necrosis factor receptor family that activates c-Jun N-terminal kinase in T cells. J Biol Chem 272:25190–25194. Anderson DM, Maraskovsky E, Billingsley WL, Dougall WC, Tometsko ME, Roux ER, Teepe MC, DuBose RF, Cosman D, Galibert L 1997 A homologue of the TNF receptor and its ligand enhance T-cell growth and dendritic-cell function. Nature 390:175–179. Udagawa N, Takahashi N, Akatsu T, Tanaka H, Sasaki T, Nishihara T, Koga T, Martin TJ, Suda T 1990 Origin of osteoclasts: mature monocytes and macrophages are capable of differentiating into osteoclasts under a suitable microenvironment prepared by bone marrow-derived stromal cells. Proc Natl Acad Sci USA 87:7260–7264. Miyamoto A, Kunisada T, Hemmi H, Yamane T, Yasuda H, Miyake K, Yamazaki H, Hayashi SI 1998 Establishment and characterization of an immortal macrophage-like cell line inducible to differentiate to osteoclasts. Biochem Biophys Res Commun 242:703–709. Matsuzaki K, Udagawa N, Takahashi N, Yamaguchi K, Yasuda H, Shima N, Morinaga T, Toyama Y, Yabe Y, Higashio K, Suda T 1998 Osteoclast differentiation factor (ODF) induces osteoclast-like cell formation in human peripheral blood mononuclear cell cultures. Biochem Biophys Res Commun 246:199–204. Quinn JM, Elliott J, Gillespie MT, Martin TJ 1998 A combination of osteoclast differentiation factor and macrophage-colony stimulating factor is sufficient for both human and mouse osteoclast formation in vitro. Endocrinology 139:4424–4427. Suda T, Takahashi N, Udagawa N, Jimi E, Gillespie MT, Martin TJ 1999 Modulation of osteoclast differentiation and function by the new members of the tumor necrosis factor receptor and ligand families. Endocr Rev 20:345–357. Yasuda H, Shima N, Nakagawa N, Yamaguchi K, Kinosaki M, Goto M, Mochizuki SI, Tsuda E, Morinaga T, Udagawa N, Takahashi N, Suda T, Higashio K 1999 A novel molecular mechanism modulating osteoclast differentiation and function. Bone 25:109–113. Tsukii K, Shima N, Mochizuki S, Yamaguchi K, Kinosaki M, Yano K, Shibata O, Udagawa N, Yasuda H, Suda T, Higashio K 1998 Osteoclast differentiation factor mediates an essential signal for bone resorption induced by 1,25-dihydroxyvitamin D3, prostaglandin E2, or parathyroid hormone in the microenvironment of bone. Biochem Biophys Res Commun 246:337–341. Burgess TL, Qian Y, Kaufman S, Ring BD, Van G, Capparelli C, Kelley M, Hsu H, Boyle WJ, Dunstan CR, Hu S, Lacey DL 1999 The ligand for osteoprotegerin (OPGL) directly activates mature osteoclasts. J Cell Biol 145:527–538. Mizuno A, Kanno T, Hoshi M, Shibata O, Yano K, Fujise N, Kinosaki M, Yamaguchi K, Tsuda E, Murakami A, Yasuda H, Higashio K 2002 Transgenic mice overexpressing soluble osteoclast differentiation factor (sODF) exhibit severe osteoporosis. J Bone Miner Metab 20:337–344.

683 55. Kong YY, Yoshida H, Sarosi I, Tan HL, Timms E, Capparelli C, Morony S, Oliveira-dos-Santos AJ, Van G, Itie A, Khoo W, Wakeham A, Dunstan CR, Lacey DL, Mak TW, Boyle WJ, Penninger JM 1999 OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis. Nature 397:315–323. 56. Kim N, Odgren PR, Kim DK, Marks SC, Jr., Choi Y 2000 Diverse roles of the tumor necrosis factor family member TRANCE in skeletal physiology revealed by TRANCE deficiency and partial rescue by a lymphocyte-expressed TRANCE transgene. Proc Natl Acad Sci USA 97: 10905–10910. 57. Nakagawa N, Kinosaki M, Yamaguchi K, Shima N, Yasuda H, Yano K, Morinaga T, Higashio K 1998 RANK is the essential signaling receptor for osteoclast differentiation factor in osteoclastogenesis. Biochem Biophys Res Commun 253: 395–400. 58. Hsu H, Lacey DL, Dunstan CR, Solovyev I, Colombero A, Timms E, Tan HL, Elliott G, Kelley MJ, Sarosi I, Wang L, Xia XZ, Elliott R, Chiu L, Black T, Scully S, Capparelli C, Morony S, Shimamoto G, Bass MB, Boyle WJ 1999 Tumor necrosis factor receptor family member RANK mediates osteoclast differentiation and activation induced by osteoprotegerin ligand. Proc Natl Acad Sci USA 96:3540–3545. 59. Dougall WC, Glaccum M, Charrier K, Rohrbach K, Brasel K, De Smedt T, Daro E, Smith J, Tometsko ME, Maliszewski CR, Armstrong A, Shen V, Bain S, Cosman D, Anderson D, Morrissey PJ, Peschon JJ, Schuh J 1999 RANK is essential for osteoclast and lymph node development. Genes Dev 13:2412–2424. 60. Li J, Sarosi I, Yan XQ, Morony S, Capparelli C, Tan HL, McCabe S, Elliott R, Scully S, Van G, Kaufman S, Juan SC, Sun Y, Tarpley J, Martin L, Christensen K, McCabe J, Kostenuik P, Hsu H, Fletcher F, Dunstan CR, Lacey DL, Boyle WJ 2000 RANK is the intrinsic hematopoietic cell surface receptor that controls osteoclastogenesis and regulation of bone mass and calcium metabolism. Proc Natl Acad Sci USA 97:1566–1571. 61. Hughes AE, Ralston SH, Marken J, Bell C, MacPherson H, Wallace RG, van Hul W, Whyte MP, Nakatsuka K, Hovy L, Anderson DM 2000 Mutations in TNFRSF11A, affecting the signal peptide of RANK, cause familial expansile osteolysis. Nat Genet 24:45–48. 62. Whyte MP, Hughes AE 2002 Expansile skeletal hyperphosphatasia is caused by a 15-base pair tandem duplication in TNFRSF11A encoding RANK and is allelic to familial expansile osteolysis. J Bone Miner Res 17:26–29. 63. Kobayashi K, Takahashi N, Jimi E, Udagawa N, Takami M, Kotake S, Nakagawa N, Kinosaki M, Yamaguchi K, Shima N, Yasuda H, Morinaga T, Higashio K, Martin TJ, Suda T 2000 Tumor necrosis factor alpha stimulates osteoclast differentiation by a mechanism independent of the ODF/RANKLRANK interaction. J Exp Med 191:275–286. 64. Azuma Y, Kaji K, Katogi R, Takeshita S, Kudo A 2000 Tumor necrosis factor-alpha induces differentiation of and bone resorption by osteoclasts. J Biol Chem 275: 4858–4864. 65. Lam J, Takeshita S, Barker JE, Kanagawa O, Ross FP, Teitelbaum SL 2000 TNF-α induces osteoclastogenesis by direct stimulation of macrophages exposed to permissive levels of RANK ligand. J Clin Invest 106:1481–1488. 66. Murakami T, Yamamoto M, Ono K, Nishikawa M, Nagata N, Motoyoshi K, Akatsu T 1998 Transforming growth factor-β1 increases mRNA levels of osteoclastogenesis inhibitory factor in osteoblastic/stromal cells and inhibits the survival

684

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

HISATAKA YASUDA, KANJI HIGASHIO, AND TATSUO SUDA

of murine osteoclast-like cells. Biochem Biophys Res Commun 252:747–752. Takai H, Kanematsu M, Yano K, Tsuda E, Higashio K, Ikeda K, Watanabe K, Yamada Y 1998 Transforming growth factorbeta stimulates the production of osteoprotegerin/osteoclastogenesis inhibitory factor by bone marrow stromal cells. J Biol Chem 273:27091–27096. Quinn JM, Itoh K, Udagawa N, Hausler K, Yasuda H, Shima N, Mizuno A, Higashio K, Takahashi N, Suda T, Martin TJ, Gillespie MT 2001 Transforming growth factor beta affects osteoclast differentiation via direct and indirect actions. J Bone Miner Res 16:1787–1794. Udagawa N, Takahashi N, Yasuda H, Mizuno A, Itoh K, Ueno Y, Shinki T, Gillespie MT, Martin TJ, Higashio K, Suda T 2000 Osteoprotegerin produced by osteoblasts is an important regulator in osteoclast development and function. Endocrinology 141:3478–3484. Horwood NJ, Elliott J, Martin TJ, Gillespie MT 1998 Osteotropic agents regulate the expression of osteoclast differentiation factor and osteoprotegerin in osteoblastic stromal cells. Endocrinology 139:4743–4746. Lee SK, Lorenzo JA 1999 Parathyroid hormone stimulates TRANCE and inhibits osteoprotegerin messenger ribonucleic acid expression in murine bone marrow cultures: correlation with osteoclast-like cell formation. Endocrinology 140:3552–3561. Brandstrom H, Jonsson KB, Ohlsson C, Vidal O, Ljunghall S, Ljunggren O 1998 Regulation of osteoprotegerin mRNA levels by prostaglandin E2 in human bone marrow stroma cells. Biochem Biophys Res Commun 247:338–341. O’Brien CA, Gubrij I, Lin SC, Saylors RL, Manolagas SC 1999 STAT3 activation in stromal/osteoblastic cells is required for induction of the receptor activator of NF-κB ligand and stimulation of osteoclastogenesis by gp130-utilizing cytokines or interleukin-1 but not 1,25-dihydroxyvitamin D3 or parathyroid hormone. J Biol Chem 274:19301–19308. Vidal NO, Brandstrom H, Jonsson KB, Ohlsson C 1998 Osteoprotegerin mRNA is expressed in primary human osteoblast-like cells: down-regulation by glucocorticoids. J Endocrinol 159:191–195. Hofbauer LC, Gori F, Riggs BL, Lacey DL, Dunstan CR, Spelsberg TC, Khosla S 1999 Stimulation of osteoprotegerin ligand and inhibition of osteoprotegerin production by glucocorticoids in human osteoblastic lineage cells: potential paracrine mechanisms of glucocorticoid-induced osteoporosis. Endocrinology 140:4382–4389. Hofbauer LC, Khosla S, Dunstan CR, Lacey DL, Spelsberg TC, Riggs BL 1999 Estrogen stimulates gene expression and protein production of osteoprotegerin in human osteoblastic cells. Endocrinology 140:4367–4370. Vidal ON, Sjogren K, Eriksson BI, Ljunggren O, Ohlsson C 1998 Osteoprotegerin mRNA is increased by interleukin-1α in the human osteosarcoma cell line MG-63 and in human osteoblast-like cells. Biochem Biophys Res Commun 248:696–700. Hofbauer LC, Lacey DL, Dunstan CR, Spelsberg TC, Riggs BL, Khosla S 1999 Interleukin-1β and tumor necrosis factoralpha, but not interleukin-6, stimulate osteoprotegerin ligand gene expression in human osteoblastic cells. Bone 25:255–259. Brandstrom H, Jonsson KB, Vidal O, Ljunghall S, Ohlsson C, Ljunggren O 1998 Tumor necrosis factor-alpha and -beta upregulate the levels of osteoprotegerin mRNA in human osteosarcoma MG-63 cells. Biochem Biophys Res Commun 248:454–457.

80. Takami M, Takahashi N, Udagawa N, Miyaura C, Suda K, Woo JT, Martin TJ, Nagai K, Suda T 2000 Intracellular calcium and protein kinase C mediate expression of receptor activator of nuclear factor-κB ligand and osteoprotegerin in osteoblasts. Endocrinology 141:4711–4719. 81. Nakagawa N, Yasuda H, Yano K, Mochizuki S, Kobayashi N, Fujimoto H, Shima N, Morinaga T, Chikazu D, Kawaguchi H, Higashio K 1999 Basic fibroblast growth factor induces osteoclast formation by reciprocally regulating the production of osteoclast differentiation factor and osteoclastogenesis inhibitory factor in mouse osteoblastic cells. Biochem Biophys Res Commun 265:158–163. 82. Nakagawa N, Yasuda H, Yano K, Mochizuki S, Kobayashi N, Fujimoto H, Yamaguchi K, Shima N, Morinaga T, Higashio K 1999 Basic fibroblast growth factor inhibits osteoclast formation induced by 1α,25-dihydroxyvitamin D3 through suppressing the production of osteoclast differentiation factor. Biochem Biophys Res Commun 265:45–50. 83. Hofbauer LC, Dunstan CR, Spelsberg TC, Riggs BL, Khosla S 1998 Osteoprotegerin production by human osteoblast lineage cells is stimulated by vitamin D, bone morphogenetic protein-2, and cytokines. Biochem Biophys Res Commun 250:776–781. 84. Koide M, Murase Y, Yamato K, Noguchi T, Okahashi N, Nishihara T 1999 Bone morphogenetic protein-2 enhances osteoclast formation mediated by interleukin-1α through upregulation of osteoclast differentiation factor and cyclooxygenase-2. Biochem Biophys Res Commun 259:97–102. 85. Bekker PJ, Holloway D, Nakanishi A, Arrighi M, Leese PT, Dunstan CR 2001 The effect of a single dose of osteoprotegerin in postmenopausal women. J Bone Miner Res 16:348–360. 86. Body JJ, Greipp P, Coleman RE, Facon T, Geurs F, Fermand JP, Harousseau JL, Lipton A, Mariette X, Williams CD, Nakanishi A, Holloway D, Martin SW, Dunstan CR, Bekker PJ 2003 A Phase I study of AMGN-0007, a recombinant osteoprotegerin construct, in patients with multiple myeloma or breast carcinoma related bone metastases. Cancer 97:887–892. 87. Peterson MC, Stouch BJ, Chen D, Baughman S, M. DA, Dunstan CR, Martin SW 2003 AMG 162, a fully human monoclonal antibody against receptor activator of NF-κB ligand (RANKL), rapidly and profoundly suppresses bone resorption in cynomolgus monkey. Bone 32(Suppl 1):S44. 88. Louie SG, Park B, Yoon H 2003 Biological response modifiers in the management of rheumatoid arthritis. Am J Health Syst Pharm 60:346–355. 89. Galibert L, Tometsko ME, Anderson DM, Cosman D, Dougall WC 1998 The involvement of multiple tumor necrosis factor receptor (TNFR)-associated factors in the signaling mechanisms of receptor activator of NF-κB, a member of the TNFR superfamily. J Biol Chem 273: 34120–34127. 90. Darnay BG, Haridas V, Ni J, Moore PA, Aggarwal BB 1998 Characterization of the intracellular domain of receptor activator of NF-κB (RANK). Interaction with tumor necrosis factor receptor-associated factors and activation of NF-κB and c-Jun N-terminal kinase. J Biol Chem 273: 20551–20555. 91. Naito A, Azuma S, Tanaka S, Miyazaki T, Takaki S, Takatsu K, Nakao K, Nakamura K, Katsuki M, Yamamoto T, Inoue J 1999 Severe osteopetrosis, defective interleukin-1 signaling and lymph node organogenesis in TRAF6-deficient mice. Genes Cells 4:353–362. 92. Lomaga MA, Yeh WC, Sarosi I, Duncan GS, Furlonger C, Ho A, Morony S, Capparelli C, Van G, Kaufman S, van der Heiden A, Itie A, Wakeham A, Khoo W, Sasaki T, Cao Z,

CHAPTER 38 Vitamin D and Osteoclastogenesis

93.

94.

95.

96. 97.

98. 99.

100. 101.

102.

103.

Penninger JM, Paige CJ, Lacey DL, Dunstan CR, Boyle WJ, Goeddel DV, Mak TW 1999 TRAF6 deficiency results in osteopetrosis and defective interleukin-1, CD40, and LPS signaling. Genes Dev 13:1015–1024. Kobayashi N, Kadono Y, Naito A, Matsumoto K, Yamamoto T, Tanaka S, Inoue J 2001 Segregation of TRAF6-mediated signaling pathways clarifies its role in osteoclastogenesis. EMBO J 20:1271–1280. Takayanagi H, Ogasawara K, Hida S, Chiba T, Murata S, Sato K, Takaoka A, Yokochi T, Oda H, Tanaka K, Nakamura K, Taniguchi T 2000 T-cell-mediated regulation of osteoclastogenesis by signaling cross-talk between RANKL and IFN-gamma. Nature 408:600–605. Kaji K, Katogi R, Azuma Y, Naito A, Inoue JI, Kudo A 2001 Tumor necrosis factor alpha-induced osteoclastogenesis requires tumor necrosis factor receptor-associated factor 6. J Bone Miner Res 16:1593–1599. Soriano P, Montgomery C, Geske R, Bradley A 1991 Targeted disruption of the c-src proto-oncogene leads to osteopetrosis in mice. Cell 64:693–702. Wong BR, Besser D, Kim N, Arron JR, Vologodskaia M, Hanafusa H, Choi Y 1999 TRANCE, a TNF family member, activates Akt/PKB through a signaling complex involving TRAF6 and c-Src. Mol Cell 4:1041–1049. Wang ZQ, Ovitt C, Grigoriadis AE, Mohle-Steinlein U, Ruther U, Wagner EF 1992 Bone and haematopoietic defects in mice lacking c-fos. Nature 360:741–745. Grigoriadis AE, Wang ZQ, Cecchini MG, Hofstetter W, Felix R, Fleisch HA, Wagner EF 1994 c-Fos: a key regulator of osteoclast-macrophage lineage determination and bone remodeling. Science 266:443–448. Matsuo K, Owens JM, Tonko M, Elliott C, Chambers TJ, Wagner EF 2000 Fosl1 is a transcriptional target of c-Fos during osteoclast differentiation. Nat Genet 24:184–187. Takayanagi H, Kim S, Matsuo K, Suzuki H, Suzuki T, Sato K, Yokochi T, Oda H, Nakamura K, Ida N, Wagner EF, Taniguchi T 2002 RANKL maintains bone homeostasis through c-Fos-dependent induction of interferon-beta. Nature 416:744–749. David JP, Sabapathy K, Hoffmann O, Idarraga MH, Wagner EF 2002 JNK1 modulates osteoclastogenesis through both c-Jun phosphorylation-dependent and -independent mechanisms. J Cell Sci 115:4317–4325. Matsumoto M, Sudo T, Saito T, Osada H, Tsujimoto M 2000 Involvement of p38 mitogen-activated protein kinase signaling pathway in osteoclastogenesis mediated by receptor activator of NF-κB ligand (RANKL). J Biol Chem 275: 31155–31161.

685 104. Li X, Udagawa N, Itoh K, Suda K, Murase Y, Nishihara T, Suda T, Takahashi N 2002 p38 MAPK-mediated signals are required for inducing osteoclast differentiation but not for osteoclast function. Endocrinology 143:3105–3113. 105. Franzoso G, Carlson L, Xing L, Poljak L, Shores EW, Brown KD, Leonardi A, Tran T, Boyce BF, Siebenlist U 1997 Requirement for NF-κB in osteoclast and B-cell development. Genes Dev 11:3482–3496. 106. Iotsova V, Caamano J, Loy J, Yang Y, Lewin A, Bravo R 1997 Osteopetrosis in mice lacking NF-κB1 and NF-κB2. Nat Med 3:1285–1289. 107. Jimi E, Akiyama S, Tsurukai T, Okahashi N, Kobayashi K, Udagawa N, Nishihara T, Takahashi N, Suda T 1999 Osteoclast differentiation factor acts as a multifunctional regulator in murine osteoclast differentiation and function. J Immunol 163:434–442. 108. Tondravi MM, McKercher SR, Anderson K, Erdmann JM, Quiroz M, Maki R, Teitelbaum SL 1997 Osteopetrosis in mice lacking haematopoietic transcription factor PU.1. Nature 386:81–84. 109. Luchin A, Purdom G, Murphy K, Clark MY, Angel N, Cassady AI, Hume DA, Ostrowski MC 2000 The microphthalmia transcription factor regulates expression of the tartrate-resistant acid phosphatase gene during terminal differentiation of osteoclasts. J Bone Miner Res 15:451–460. 110. Motyckova G, Weilbaecher KN, Horstmann M, Rieman DJ, Fisher DZ, Fisher DE 2001 Linking osteopetrosis and pycnodysostosis: regulation of cathepsin K expression by the microphthalmia transcription factor family. Proc Natl Acad Sci USA 98:5798–5803. 111. Ishida N, Hayashi K, Hoshijima M, Ogawa T, Koga S, Miyatake Y, Kumegawa M, Kimura T, Takeya T 2002 Large scale gene expression analysis of osteoclastogenesis in vitro and elucidation of NFAT2 as a key regulator. J Biol Chem 277:41147–41156. 112. Takayanagi H, Kim S, Koga T, Nishina H, Isshiki M, Yoshida H, Saiura A, Isobe M, Yokochi T, Inoue J, Wagner EF, Mak TW, Kodama T, Taniguchi T 2002 Induction and activation of the transcription factor NFATc1 (NFAT2) integrate RANKL signaling in terminal differentiation of osteoclasts. Dev Cell 3:889–901. 113. Suda T, DeLuca HF, Tanaka Y 1970 Biological activity of 25-hydroxyergocalciferol in rats. J Nutr 100:1049–1052. 114. Ueno Y, Shinki T, Nagai Y, Murayama H, Fujii K, Suda T 2003 1α,25-Dihydroxyvitamin D3 suppresses the expression of RANKL mRNA in bone of thyroparathyroidectomized rats constantly infused with PTH. J Cell Biochem 2003, 90:267–277.

CHAPTER 39

Vitamin D Control of the Calcitonin Gene in Thyroid C Cells ANDREW F. RUSSO ROBERT F. GAGEL

Department of Physiology and Biophysics, University of Iowa, Iowa City, Iowa Section of Endocrinology, University of Texas M.D. Anderson Cancer Center, Houston, Texas

I. Introduction II. Origin of Thyroid C Cells and Their Function in Health and Disease III. Calcitonin and CGRP Production by Normal and Tranformed C Cells IV. Regulation of Calcitonin Levels by Vitamin D

I. INTRODUCTION The parafollicular or C cells of the thyroid gland are a distinctive subset of thyroid cells that have their embryological origin in the neural crest. These cells are located adjacent to the thyroid follicle and represent only ∼1% of the total cells within the thyroid gland. The C cells differ from the more common thyroid follicular cells in several important ways. First, as described below, C cells arise from the neural crest and have neuronal potential. Second, C cells have properties of neuroendocrine cells that are not found in the follicular cells. These properties include the uptake and decarboxylation of amines, the presence of secretory granules that are released directly into blood vessels rather than into the thyroid follicle, and the production of several neuroendocrine peptides including somatostatin, cholecystokinin, bombesin, chromogranin A, and its predominant secretory product, calcitonin (CT). Unlike follicular cells, C cells do not concentrate iodine and do not produce thyroglobulin or thyroid hormone. Third, C cells are scattered and do not directly contact the follicles. The low abundance of C cells within the mammalian thyroid gland makes it difficult to identify these cells without specific immunohistochemical staining techniques for secretory products such as CT. However, the establishment of transformed C cell lines has allowed insights into vitamin D control of CT gene expression. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Regulation of Calcitonin CGRP Gene Transcription VI. Mechanism for Down-Regulation of Calcitonin Gene Transcription by Vitamin D VII. Effects of Vitamin D on Cell Growth VIII. Summary and Future Directions References

II. ORIGIN OF THYROID C CELLS AND THEIR FUNCTION IN HEALTH AND DISEASE A. Embryonic Development of Thyroid C Cells In a classic series of experiments, Le Douarin and colleagues [1] tracked the migration of quail neural crest–derived cells in a chicken model of neural crest development. C cells are derived from vagal neural crest progenitors that give rise to a subset of enteric neurons [1–4]. Similarities between thyroid C cells and serotonergic enteric neurons were first described by Gershon more than 30 years ago using bat and sheep C cell cultures [5]. It has now been shown that rat and human C cells share biochemical and morphological properties with serotonergic neurons that are enhanced in primary cell culture and in C cell tumors [4,6–8]. This neuronal potential makes the C cells an attractive model system for studying serotonergic neurons [9]. After leaving the neural crest, the C cell progenitors migrate to the ultimobranchial body, a discrete structure in fish and birds composed predominantly of CT-producing cells [1]. The migration is thought to be similar in mammalian species, although the C cells are dispersed in the thyroid gland. The regulatory signals that determine the migratory pattern have not been Copyright © 2005, Elsevier, Inc. All rights reserved.

688

ANDREW F. RUSSO AND ROBERT F. GAGEL

fully elucidated, but findings suggest a role for the Hox homeodomain group 3 paralogs [10], the NCAM cell adhesion molecule [11], and the glial cell-derived neurotrophic factor (GDNF) and Ret signaling system [12–15]. GDNF and its receptor complex, Ret and GDNFR-α, form a signaling system important for migration of neural crest cells to form the neural network responsible for normal gastrointestinal motility. The GDNF signaling system includes several components (Fig. 1). The ligand GDNF binds an extracellular lipid-anchored protein called GDNF receptor-α (GDNFR-α). The activated GDNFR-α then drives the dimerization and activation of the transmembrane Ret tyrosine kinase receptor [13–15]. The peptide GDNF is a member of the cysteine knot family of growth factors, which includes transforming growth factor-β (TGFβ), vascular endothelial growth factor, several bone matrix proteins, the platelet-derived growth factors, and the pituitary glycoprotein hormones. The GDNFR-α4 family member has been found to be the predominant GDNFR-α family member expressed in normal and malignant thyroid C cells [16]. Interestingly, this receptor preferentially binds persephin, which is a distinct member of the GDNF neurotrophin family [17]. The overlapping expression of Ret and GDNFR-α4 in C cells and the evidence that Ret mutations transform the C cell make it reasonable to hypothesize a role for

GDNF Mutations in the RET cysteine-rich domain Cadherin activate by enhanced dimerization of RET Codon in the absence of GDNF 609

GDNFR-α

611 Cys-Rich 618 620 634 TM TK Mutations in the tyrosine kinase domain activate without dimerization of RET

FIGURE 1

768 804 918

The RET/GDNFR-α signaling system. Glial cellderived neurotropic factor (GDNF) signals through a novel type of receptor system that combines a classic tyrosine kinase receptor (RET) and an extracellular lipid anchored protein (GDNFR-α). Mutations in a cysteine-rich (Cys-Rich) domain of RET cause multiple endocrine neoplasia type 2A and familial medullary thyroid carcinoma. Mutations of the intracellular tyrosine kinase (TK) domain cause multiple endocrine neoplasia type 2B. Other abbreviations: TM, transmembrane region of RET; Cadherin, a cadherinlike extracellular domain of RET.

this signaling system in the migration and differentiation of C cells.

B. C Cells in Normal Adult Thyroid Gland In the normal mammalian thyroid gland the C cells are distributed throughout the thyroid gland, although in humans the highest concentration is located along a central cephalad–caudal axis in each lobe of the thyroid gland at the junction of the upper one-third and lower two-thirds of the thyroid gland. The C cells have morphological features of neuroendocrine cells with secretory granules that empty into adjacent capillaries.

C. Neoplasia of the C Cell Medullary thyroid carcinoma (MTC) is a malignant neoplasm of the C cells. Hereditary MTC, an autosomal dominant form of thyroid cancer, accounts for 25% of all medullary thyroid carcinoma. Medullary thyroid carcinoma occurs in combination with tumors of the adrenal medulla and parathyroid tumors in a disorder called multiple endocrine neoplasia type 2 (MEN 2) [18,19] or by itself in familial medullary thyroid carcinoma [20]. Mapping studies initiated in the early 1980s localized the causative gene to chromosome 10 and led to the identification in 1993 of c-ret protooncogene mutations as the cause for this neoplasm [21,22]. The most common mutations, accounting for greater than 85% of all mutations found in the hereditary form of MTC, mutate highly conserved cysteines in the extracellular domain of the Ret tyrosine kinase receptor (Fig. 1). A less common mutation, associated with a rare form of MEN 2 (MEN 2B), affects the intracellular tyrosine kinase portion of the Ret receptor [23]. Several lines of evidence support a causative role for these mutations in the genesis of MTC, including genotype/phenotype correlation [24] and the demonstration that transfection of mutant Ret cDNA into NIH-3T3 cells causes transformation [25,26]. These studies have identified two different mechanisms of activation: dimerization of the receptor in the absence of ligand is found with mutations of the cysteine-rich extracellular domain, and mutations of the tyrosine kinase domain cause activation in the absence of either ligand or dimerization (Fig. 1). Sporadic MTC accounts for approximately 75% of all medullary thyroid carcinoma. A somatic mutation of codon 918, the same mutation that causes MEN 2B when present as a germ-line mutation, is found in approximately 25% of all sporadic MTC (Fig. 1) [27].

CHAPTER 39 Vitamin D Control of the Calcitonin Gene in Thyroid C Cells

III. CALCITONIN AND CGRP PRODUCTION BY NORMAL AND TRANSFORMED C CELLS A. Alternative RNA Processing of the CT/CGRP Gene The CT gene (CALC I) has six exons that encode two peptides, CT and α-calcitonin gene-related peptide (CGRP) (Fig. 2). A second gene encodes β-CGRP. This gene does not contain a functional CT exon and less is known about its regulation. For the purposes of this chapter, we will generally refer to α-CGRP simply as CGRP. The primary RNA transcript of the CT gene is processed to produce an mRNA encoding CT by splicing together exons 1 through 4 of the CT gene and utilizing the polyadenylation site immediately downstream of exon 4 [28]. In the C cell, ∼95% of the primary CT/CGRP gene transcript is processed in this way to yield CT. Very little CGRP is produced by the normal C cell. In contrast, the CT/CGRP gene is expressed in neurons to yield ∼ 99% of the primary transcript as CGRP [29,30]. This processing pathway excludes exon 4 to generate a mRNA containing exons 1, 2, 3, 5, and 6 (Fig. 2). Processing of the primary transcript is often significantly altered in MTC, with up to 90% production of CGRP [31–33]. The shift to CGRP production in MTC is consistent with observations that MTC cells have neuronal properties [8,9,33]. In thyroid C cells the primary transcript is processed to include exon 4 and produce a mRNA encoding calcitonin (CT) Intronic regulatory element necessary A(n) for CT-specific splicing

Introns

2

1

Exons 1

2

A 4

3 3

4

A

5 5

6

A(n) In neurons the primary transcript is processed to exclude exon 4 and produce a mRNA encoding calcitonin gene-related peptide (CGRP)

FIGURE 2 Alternative RNA processing of the CT/CGRP gene. The primary transcript can be processed to produce mRNA encoding either calcitonin (CT) or calcitonin gene-related peptide (CGRP). A regulatory element located downstream of exon 4 contains a pyrimidine tract and pseudo 5′ splice site and is required for inclusion and polyadenylation of exon 4. Mutation of this element results in a skip splice with production of mRNA encoding CGRP. Other elements within or preceding exon 4 have been implicated in regulation of this splice choice, but they have not been characterized. A(n), polyA tail; A, polyadenylation site.

689

The mechanism by which the alternative splice is regulated is not yet fully understood [34]. Studies in transgenic mice that were engineered to express the CT/CGRP gene in all tissues suggested that CT mRNA production is the default choice [30]. In these mice, all tissues except the brain and heart expressed CT mRNA. Subsequent mapping studies in various cell lines have revealed several regulatory elements that contribute to enhanced recognition and inclusion of exon 4 [35–39]. The most clearly defined of these is an intron element located approximately 250 nucleotides downstream of the CT exon 4, whose function is essential for CT exon 4 polyadenylation. This novel splicing element contains a 5′ splice site-like sequence [40,41] that interacts with U1 small nuclear ribonucleoproteins, pyrimidine tract binding protein, and accessory splicing factor to form a complex that facilitates polyadenylation of CT exon 4 [42].

B. Transcriptional Regulation of Calcitonin Gene Expression The CT gene is expressed almost exclusively in thyroid C cells and a subset of peripheral and central neurons, along with a scattered population of neuroendocrine cells in the lung, prostate, and pituitary. Transcription of the gene is enhanced by protein kinase A, protein kinase C, and mitogen activated protein kinases (MAP kinases). Transcription is inhibited by the active metabolite of vitamin D, 1,25-dihydroxyvitamin D3 [1,25(OH)2D3], retinoic acid, and 5-HT1 agonists that are commonly used as antimigraine drugs. Glucocorticoids can either activate or repress transcription in a cell-specific manner. Two regions of the CT promoter have been implicated in transcriptional regulation. The first is the neuroendocrine-specific enhancer located approximately 1000 nucleotides upstream of the transcription start site (Fig. 3). The second regulatory region is responsible for cAMP and Rasmediated enhancement and is located approximately 250 nucleotides upstream of the start site (Fig. 4).

IV. REGULATION OF CALCITONIN LEVELS BY VITAMIN D Although early studies of the effect of vitamin D on CT production produced conflicting results, more recent studies defined an inhibitory role of vitamin D on CT gene expression. The studies of Raue and coworkers [43] provided the first clear demonstration that systemic 1,25(OH)2D3 administration causes a reversible decrease in intrathyroidal CT content.

690

ANDREW F. RUSSO AND ROBERT F. GAGEL

Neuroendocrine-specific HLH enhancer −1060 to −905

E1

cAMP responsive enhancer −254 to −88

E2

Ets

−1060

HLH CANNTG −1049/−1041 CACACCTGG

OB2

E3

−905

HLH CANNTG −937/−928 GGCAGCTGTG −916/−907 GGCAGGTGAT

FIGURE 3

Structure of the CT/CGRP neuroendocrine-specific enhancer. The neuroendocrine-specific enhancer is located between −1060 and −905 nucleotides upstream of the transcription start site. The elements El, E2, and E3 are all potential binding sites for helix–loop–helix (HLH) transcription factors. The elements E2 and E3 are absolutely essential for basal and cell-specific transcription. The OB2 element is required in MTC cells for E2 activity. The El, Ets, and other sites contribute only to maximal activity of the enhancer.

Naveh-Many and Silver [44] subsequently showed that doses of 1,25(OH)2D3 that had no effect on the serum calcium concentration lowered CT mRNA to 10% of control within 6 hr of treatment by inhibition of transcription. Although these authors suggest a regulatory loop between CT and 1,25(OH)2D3, the concentration

Neuroendocrine-specific HLH enhancer −1060 to −905

CREB/ATF −254

cAMP responsive enhancer −254 to −88

?

−88

of 1,25(OH)2D3 required for these effects suggests a pharmacological rather than a physiological effect. The conclusion that vitamin D and CT can form a regulatory loop under nonhomeostatic conditions is consistent with evidence that long-term treatment with high levels of vitamin D decreases CT synthesis in vivo [45]. The potential significance of vitamin D regulation of CT levels under times of calcium stress is suggested by studies on individuals with the birth defect Williams syndrome. These individuals have infantile hypercalcemia among other developmental defects [46,47]. Williams children have deficient serum CT levels and abnormalities in their vitamin D metabolism, which might contribute to their hypercalcemia [46]. The deficiency in CT levels is not due to mutations in the CT gene, which suggested that there might be abnormal regulation of the CT gene [48]. It has been found that the chromosomal microdeletion in Williams syndrome removes a gene, Williams syndrome transcription factor (WSTF), that encodes a transcription factor that is required for vitamin D receptor function on some target genes [49]. The authors hypothesize that the lack of WSTF affects vitamin D metabolism, which results in increased 1,25(OH)2D3. While speculative, this elevation in 1,25(OH)2D3 might then account for the lowered CT levels in Williams children. On the flip side of this regulatory loop, the role of CT in regulating 1,25(OH)2D3 synthesis has only recently been resolved. CT can act independently of parathyroid hormone to increase the levels of the biosynthetic enzyme 25-hydroxyvitamin D3-1α-hydroxylase gene in vivo [50,51]. Thus, CT increases 1,25(OH)2D3 levels, while 1,25(OH)2D3 feeds back to inhibit CT production. The exact mechanism by which 1,25(OH)2D3 suppresses CT gene transcription in normal C cells is not known. However, the cells contain detectable levels of the vitamin D receptor (VDR) [52,53], and the inhibition is induced preferentially by the VDR-activating metabolite 1,25(OH)2D3 and not by other metabolites of the vitamin D endocrine system [52]. These findings suggest that repression of the gene is mediated by ligand-dependent transcriptional effects of the VDR and has enabled mapping of the DNA sequences responsible for this effect.

CRE TGACGTCA CRE TGACGTCA OCT ATGCAAAT −253/−246TGACGTCA −169/−162 TGACCTCA −161/−154 ATGCAAAT

FIGURE 4

Structure of the CT/CGRP cAMP responsive enhancer. The cAMP response element is located between −254 and −88 nucelotides upstream of the transcription start site. Several elements are located within this complex element. The CRE, a cAMP-responsive element of this enhancer, is a binding site for CRE binding protein (CREB) and ATF. The CREL/O element contains a transcriptionally active CRE-like motif that does not bind CREB or ATF and an adjacent motif that is homologous to binding sites of the Octamer/POU homeobox transcription factors.

A. Development of a Model System in Which to Study the Effect of 1,25(OH)2D3 on CT Transcription To understand the molecular mechanisms that lead to repression of CT gene expression, it was necessary to identify a cell culture system. The human MTC TT cell line was selected since it expressed the CT gene at a

CHAPTER 39 Vitamin D Control of the Calcitonin Gene in Thyroid C Cells

high level and was repressed by 1,25(OH)2D3 [53]. The rat CA77 MTC line also shows repression of the CT gene by 1,25(OH)2D3 [54,55]. Most, if not all, MTC cell lines are derived from thyroid C cell tumors that have mutations in the RET protooncogene. The TT cell line has a mutation in the RET gene that changes the cysteine at codon 634 to a tryptophan [56,57] (Fig. 1). The TT cell line produces several polypeptide hormones including CT, CGRP, somatostatin, and parathyroid hormone–related protein [58–63]. In the TT cell line, 1,25(OH)2D3 down-regulates transcription of the somatostatin and CT genes [53,63,64] and reverses enhanced expression of the CT gene produced by glucocorticoids [65]. Transcription of the CT gene in the TT cell line is enhanced by glucocorticoids, butyrate, and activation of protein kinases A or C [66–72]. Although the regulation of CT gene expression in this cell line mimics that observed in the normal C cells, there exists the possibility that the transformed phenotype affects the response to sterols.

V. REGULATION OF CALCITONIN CGRP GENE TRANSCRIPTION To understand how 1,25(OH)2D3 regulates CT levels, it is first necessary to describe the mechanisms that control transcription of the CT/CGRP gene. The CT/CGRP gene is regulated by 1,25(OH)2D3 and all other extracellular stimuli exclusively at the transcriptional level. Analyses of the CT/CGRP promoter from either human or rat MTC cell lines have revealed almost identical regulatory elements [73–75]. Therefore this section will describe the promoter from both species, with an emphasis on the human gene. Cloning of the CT/CGRP promoter [76] and detailed analysis of its functional transcription elements revealed that it contains a distal neuroendocrine-specific enhancer [75] and a proximal cAMP and ras-regulated element [72]. All agents that increase or decrease CT/CGRP transcription act through these two elements.

A. Characterization of the cAMP-Induced Enhancer The proximal element of the CT/CGRP gene contains an overlapping set of motifs that are responsive to signal transduction pathways induced by cAMP [72,77] nerve growth factor (NGF) [78], and the activated Ras protein [70,79] (Fig. 4). It contains a perfect cAMP response element (CRE) that functions as a binding site for cAMP response element binding protein

691

(CREB) and related family members. A second element that responds to cAMP contains overlapping motifs: a CRE-like motif flanked by a downstream octamer homeodomain-like binding site that has been termed CREL/O (Fig. 4). Mutation or deletion of the CREB binding site reduces cAMP-induced transcription in transformed C cells but does not abolish it, indicating that the CREL/O element can function without the CREB element. Mutation of the octamer/CRE-like motif, however, seems to diminish the activity of the CREB binding site [72]. The function of these cAMP-responsive elements is cell specific to some extent: a construct containing the CREB binding site is functional in a cAMP-independent manner in HeLa cells, whereas there is little expression in a CT-negative MTC cell line [72]. The CREL/O motif is functional only in the CT-positive MTC cell lines [72].

B. Neuroendocrine-Specific HLH Enhancer Cell-specific transcription of the rat and human CT/CGRP genes is controlled by a distal neuroendocrinespecific enhancer located approximately 1 kb upstream from the transcription start site [73–75a,80]. It is a complex enhancer that contains several HLH motifs and flanking elements that bind cell-specific and noncell-specific transcription factors (Fig. 3). The enhancer appears to be active only in CT-expressing neuroendocrine cell lines derived from MTC or small cell lung carcinoma and in sensory neurons. An important consideration for understanding 1,25(OH)2D3 control of CT levels is that the neuroendocrine-specific enhancer is regulated. Both glucocorticoids and retinoic acid can repress the enhancer [80,81]. Interestingly, repression by 1,25(OH)2D3 [52] and 5-HT1 agonists [82,83] involves both the cAMPresponsive element and the neuroendocrine-specific enhancer. The distal cell-specific enhancer is not regulated by cAMP. However, the magnitude of the cAMPresponse can be enhanced by inclusion of the distal enhancer. These relationships suggest that the two elements function together and that 1,25(OH)2D3 can target this action. The distal neuroendocrine enhancer contains three functional CANNTG motifs (E boxes, termed E1, E2, E3) that bind helix–loop–helix (HLH) proteins [74,75, 80,84,85] (Fig. 3). There are also non-cell-specific enhancer elements that are similar to Ets and SP1 sites. HLH transcription factors regulate cell-specific gene expression in a variety of cell types. The cell-specific HLH transcription factor mammalian achete-scute homolog-1 (MASH-1) was initially considered as a

692

ANDREW F. RUSSO AND ROBERT F. GAGEL

candidate for the cell-specific HLH site since it is expressed in C cell lines [86]. Furthermore, MASH-1 is expressed in normal C cells and is required for development of C cells [4]. However, it is now clear that MASH-1 is not required for expression of the CT/CGRP gene [4,87]. The cell-specificity of the neuroendocrine enhancer has been demonstrated by transgenic and cell culture studies. Transgenic mice with 1.3 or 1.7 kb of 5′ flanking DNA express reporter genes in thyroid C cells and peripheral sensory neurons [74,88]. In cell lines, the E2 HLH site functions autonomously as a cell-specific enhancer [75,80,84,89]. Mutation of the E2 site drastically reduces CT/CGRP promoter activity even in the context of all of the other sites (Fig. 5). The E2 site does not work alone. There is an overlapping octamerbinding element that binds a 68-kDa cell-specific protein that in turn allows synergistic activation of the E2 element in rat and human MTC cell lines [80,89]. The major in vitro HLH binding complex at the E2 site is a heterodimer of the ubiquitously expressed upstream stimulatory factor (USF)-1 and USF-2 proteins [89]. USF and other HLH proteins have been implicated in combinatorial control with other proteins, including homeodomain proteins [90–92]. In neurons, the E2 HLH site alone appears to be necessary and sufficient for enhancer activity [83].

A

C. MAP Kinase Control of CGRP Transcription CT/CGRP gene expression is regulated by factors that activate MAP kinase pathways. The MAP kinases are at the convergence of several signaling cascades that transduce extracellular signals to the nucleus [93]. The first evidence of regulation of the CT/CGRP gene was from studies on Ras, an upstream activator of MAP kinases. Nelkin and colleagues then demonstrated that Ras acts through an element near the CRE that binds a novel zinc-fingered protein [79]. Two physiological stimuli that act through MAP kinases, NGF and nerve depolarization, have been shown to stimulate CT/CGRP enhancer activity. NGF treatment increases CT/CGRP gene expression in postnatal trigeminal and dorsal root ganglia (Fig. 5) [83,94]. In cell lines, NGF acts in a cell-specific manner through the CRE and flanking elements and upstream sequences that include the neuroendocrine enhancer [78]. The role of MAP kinases was demonstrated by a combination of pharmacological inhibitors and the use of constitutively active and dominant negative regulators of the ERK MAP kinase pathway (Fig. 5) [83,95]. Regulation of USF activity at the CT/CGRP E2 site by MAP kinases and other signals is consistent with other genes. Greenberg and colleagues have shown

MEK1 − CRE

HLH CT/CGRP

+ Bam

CRE

HLH mut

−

+ 0

5000 10000 Luciferase activity (light units/20 µg protein)

0

5000 10000 15000 Luciferase activity (light units/20 µg protein)

B Con HLH

CRE NGF NGF + Suma

15000

FIGURE 5 MAP kinase stimulation and sumatriptan repression of the CT/CGRP neuroendocrine HLH enhancer. (A) Cultures of trigeminal ganglia neurons were transfected with the 1250 bp CT/CGRP– luciferase reporter or the 1250 bp reporter with a mutation in the E2 HLH site (Bam insertion). The activity of the 1250-bp promoter, but not the mutant promoter, was increased by cotransfection with a CMVMEK1 expression plasmid. (B) Transfected trigeminal cultures were treated with NGF for 2 hr or pretreated with 10 µM sumatriptan (suma) for 30 min before NGF addition as indicated. Modified from Durham and Russo [83].

CHAPTER 39 Vitamin D Control of the Calcitonin Gene in Thyroid C Cells

that USF is activated by nerve depolarization [96]. Likewise, p38 MAP kinase, BDNF, and TGF-β can regulate gene expression through complex USF sites [97–99]. These results suggest that synergistic interactions between USF and other factors may be a common target for transcriptional regulation by various agents. The regulation of CT/CGRP gene expression by MAP kinases is significant since it may underlie the elevation of CGRP levels following neurogenic inflammation that occurs in several chronic pain syndromes, including arthritis and migraine. Neuronal activity and many inflammatory compounds, including those thought to be released during migraine, are known activators of MAP kinases [101–103]. Clinical studies have documented elevated serum CGRP levels in the jugular outflow vein during migraine [100]. In the neurogenic model of migraine, there is activation of trigeminal nerves that elevates CGRP release. CGRP causes vasodilation and release of inflammatory compounds from surrounding mast cells to further inflame the nerve endings. Although it has not been formally demonstrated in migraineurs, it seems likely that there is increased CGRP synthesis since migraine episodes can last for up to 72 hr [100]. Inflammation of peripheral joints leads to increased CGRP peptide and mRNA levels in the dorsal root ganglia [104]. More recent studies show that the hyperalgesia observed in wild-type CT/CGRP mice following induction of peripheral joint inflammation is markedly reduced in CT/CGRP-α−/− knockout animals [105]. This effect is most likely caused by the lack of CGRP in the dorsal root ganglia. These results implicate CGRP-α in nocioceptive responses. 5-HT1 receptor agonists that are currently used as antimigraine drugs attenuate activation of the CT/ CGRP enhancer by MAP kinases (Fig. 5) [83,95,106]. In patients, these drugs lower CGRP levels and relieve migraine pain. In MTC cells, the 5-HT1 agonists appear to act by increasing the levels of MAP kinase phosphatase-1 (MKP-1). MKP-1 is a dual-specific phosphatase that can inactivate multiple MAP kinases [107]. The angiotensin type 2 and possibly insulin receptors have also been shown to repress MAP kinase activity by elevating MKP-1 [108,109]. An unexpected finding was that activation of 5-HT1 receptors caused a robust and sustained increase in intracellular calcium in MTC cells and neurons that is sufficient to repress the CT/CGRP promoter (Fig. 6) [82,83,110]. The effect of intracellular calcium on gene expression is now recognized as being determined by many parameters, including signal amplitude and duration [111]. A transient increase in intracellular calcium, such as following depolarization, stimulates MAP kinase activation of the CT/CGRP enhancer.

693

Such activation has been widely seen with other MAP kinase–responsive genes [101,102,112]. What is less appreciated is that a prolonged elevation of calcium can have the opposite effect and decrease MAP kinase activity. Meloche and colleagues demonstrated that increased intracellular calcium was necessary and sufficient for induction of MKP-1 expression in a fibroblast cell line [113]. In trigeminal neurons and MTC cells, this prolonged elevation of calcium induces a negative feedback loop due to MKP-1 induction, which leads to repression of the CT/CGRP gene. Thus, there appears to be a dynamic balance between MAP kinases and phosphatases that control the CT/CGRP gene (Fig. 6). Since 1,25(OH)2D3 can repress CT/CGRP gene transcription, it is tempting to speculate that 1,25(OH)2D3 might also down-regulate CGRP levels in the trigeminal ganglia. If so, then 1,25(OH)2D3 might have an unexpected prophylactic value by lowering CGRP levels. This might have potential therapeutic applications for migraine and triptan-overuse syndrome headaches [114].

VI. MECHANISM FOR DOWNREGULATION OF CALCITONIN GENE TRANSCRIPTION BY VITAMIN D A. A Complex Negative VDRE That Involves Both the Neuroendocrine-Specific and cAMP Enhancers How might the vitamin D3 hormone repress the CT/CGRP gene? In general, a common mechanism for gene repression is to counteract stimulation of the gene. This mechanism appears to hold true for the CT/CGRP gene. Treatment with 1,25(OH)2D3 repressed only cAMP-induced, not basal, transcription (Fig. 7). These results suggested that the hormone acts directly on the cAMP-induced enhancer. However, when the cAMP and distal enhancers were examined separately, neither the neuroendocrine enhancer nor the cAMP-induced enhancer were repressed by 1,25(OH)2D3 (Fig. 7). These results suggested the possibility that a vitamin D–induced protein (perhaps the VDR) inhibited the synergism between the two elements rather than acting directly on either. Because such a factor must interact physically with DNA adjacent to one of these enhancers or with one or more of the transcription factors that modulate their activities, it was necessary to map a putative negative response by mutating individual enhancer elements.

694

ANDREW F. RUSSO AND ROBERT F. GAGEL

A

MTC cells

−1 min

+1 min

+20 min

Depolarization ERK Transient Ca2+

B

CT/CGRP Transcription

5-HT1 agonists HLH

Prolonged Ca2+

CRE

MKP-1

MTC cells

−1 min

+1 min

+20 min

FIGURE 6 Differential regulation of MAP kinase activity and CT/CGRP gene transcription by transient and prolonged Ca2+ signals. Intracellular Ca2+ levels in the CA77 MTC cells are shown by Fura-2 fluorescent images. (Top panels) Transient elevation in intracellular Ca2+ after depolarization with 60 mM KCl. The mean change over time is indicated (n = 26 cells). (Bottom panels) Treatment with the 5-HT1 agonist CGS12066 (10 µM) causes a prolonged Ca2+ increase. The mean change over time is indicated (n = 27 cells). A transient Ca2+ elevation stimulates the CT/CGRP enhancer via activation of ERK MAP kinase. Whether the enhancer factors are directly phosphorylated by ERK is not known. In contrast, prolonged elevation of Ca2+ induces expression of MKP-1, which inhibits ERK and decreases CT/CGRP gene transcription. Modified from Durham and Russo [95,109].

In the cAMP response element, three major components were identified: the CRE, the composite CRElike/octamer motif, and the proximal promoter. Internal deletion that removed only the CRE, reduced cAMP-induced transcription by one-half but had no effect on the repression by 1,25(OH)2D3. A replacement of the proximal CT/CGRP promoter with the thymidine kinase promoter showed that the neuroendocrinespecific enhancer, cAMP-induced transcription, and 1,25(OH)2D3 repression of the latter were all unaffected [72,75]. These experiments suggested that in the cAMP-induced enhancer the element important for 1,25(OH)2D3 action was the CREL/O motif.

A similar mapping study was performed within the upstream neuroendocrine-specific enhancer: individual elements were removed, and the effect of the deletion or mutation on basal transcription, cAMPinduced transcription, and repression by 1,25(OH)2D3 was examined (Fig. 8). These mapping experiments showed that E1 did not act synergistically with cAMPinduced transcription and did not contribute to its repression. The E2 element, which was essential for the neuroendocrine-specific constitutive transcription, also contributed to cAMP-induced transcription, but not to l,25(OH)2D3-induced repression. On the other hand, the E3 element was essential for basal and

695

CHAPTER 39 Vitamin D Control of the Calcitonin Gene in Thyroid C Cells

−1460

HLH

cAMP

+91

pCTGH-1 +91

−829

Control 1,25(OH)2D3

pCTGH-15

cAMP −129

+91

1,25(OH)2D3 + cAMP

pCTGH −1333

∆−731/−129

+91

pCTGH-7 0

50

100

150

200

250

300

Promoter activity (% pCTGH-1)

FIGURE 7 Requirement for both the cAMP and neuroendocrine enhancers for 1,25(OH)2D3 inhibition of CT/CGRP gene transcription. Constructs containing either the neuroendocrine-specific enhancer, the cAMP-responsive enhancer, or both were transfected into the human medullary thyroid carcinoma TT cell line. Transiently transfected cells were treated with cAMP, 1,25(OH)2D3, or both. Reporter gene expression (growth hormone) was measured as an indicator of transcriptional activity. Note that basal transcription is similar with or without the cAMP-induced enhancer, but the cAMP-induced transcription is significantly enhanced by the presence of the neuroendocrine enhancer. Modified from Peleg et al. [52].

−1460 pCTGH-1

Pvu II

E-box element

+91

E1+E2+E3

−1078

E1+E2+E3

−1032

E2+E3

−965

E2+E3

−965

∆−920/−829

E2

−935

E3

−829

-

-

−425

-

−129

N.D.

-

−1460

E2+E3 Sal I linker 0

25

50

Repression of promoter activity (% inhibition of cAMP-induced transcription by 1,25(OH)2D3)

FIGURE 8

Mapping the negative VDRE of the CT/CGRP gene. To identify DNA sequences that are required for down-regulation of CT/CGRP gene transcription, studies were focused on the region surrounding the neuroendocrine enhancer. Constructs containing deletions of individual E box elements were transfected into TT cells, and the response of cells to cAMP or a combination of cAMP and 1,25(OH)2D3 was examined. The E3 element of the neuroendocrine-specific enhancer was required for l,25(OH)2D3-induced repression. Modified from Peleg et al. [52].

696

ANDREW F. RUSSO AND ROBERT F. GAGEL

cAMP-induced transcription and repression by 1,25(OH)2D3 (Fig. 8).

B. Proposed Mechanism for Repression of CT/CGRP Gene Transcription by 1,25(OH)2D3 From the mapping experiments described above, we propose a model that involves at least three separate components: a l,25(OH)2D3-recruited factor (perhaps VDR), the E3 HLH transcription factor, and the CREL/O binding protein(s). In our model, the interaction between factors at the E3 and CREL/O sites allows synergistic cAMP-induced transcription. The l,25(OH)2D3-induced factor interferes with the interaction between E3 and CREL/O factors, thus preventing synergism between the two (Fig. 9). In the search for a putative binding site for VDR in the vicinity of CREL/O or E3, one sequence was found that exhibits sequence similarity with the functional VDRE motifs of the osteocalcin, osteopontin, and 24-hydroxylase genes (Table I). This sequence overlaps the binding site for the E3 binding protein(s), suggesting that binding of VDR at this site may have a direct effect on

HLH enhancer factors

cAMP responsive enhancer factors

Enhanced transcription

Vitamin D

Inhibited transcription

FIGURE 9 Model for 1,25(OH)2D3-induced repression of CT/ CGRP gene transcription. Interaction between the neuroendocrinespecific HLH and cAMP-responsive enhancers is necessary for maximal cAMP-induced transcription. A 1,25(OH)2D3-recruited factor may interfere with this synergistic interaction by binding directly to a DNA sequence overlapping the E3 motif or by interacting with the HLH and/or cAMP responsive enhancer binding proteins.

TABLE I Sequence Comparison of Vitamin D– Responsive Elements Promoter

DNA sequence

r24-OH (distal) r24-OH (proximal) rOC hOC mOP Calcitonin (−916/−900)

GGTTCA GCG GGTGCG AGGTGA GTG AGGGCG GGGTGA ATG AGGACA GGGTGA ACG GGGGCA GGTTCA CGA GGTTCA GCAGGTGA TGG ATGGCA

Shown are VDREs from the rat 24-hydroxylase promoter (r24-OH), the rat and human osteocalcin promoters (rOC and hOC), and the mouse osteopontin promoter (mOP). The CT/CGRP sequence implicated in repression by 1,25(OH)2D3 overlaps with the E3 motif (underlined). This sequence resembles the 24-hydroxylase, osteocalcin, and osteopontin gene response elements, which are known binding sites for VDR-retinoid X receptor (RXR) heterodimers.

E3 function. In our studies, however, we were unable to identify significant binding of VDR at this site [72,75]. It is possible that E3 binding proteins are required for anchorage of VDR into this weak binding site, but proving that hypothesis will require purification and characterization of these proteins. Nevertheless, these studies define a complex and interesting transcriptional unit and a potentially novel mechanism by which 1,25(OH)2D3 down-regulates CT/CGRP gene transcription.

VII. EFFECTS OF VITAMIN D ON CELL GROWTH Through their interaction with the VDR and retinoic acid receptor systems, vitamin D and its analogs have been shown to affect differentiation and growth in a variety of cell types. There is a limited and conflicting literature regarding effects in the C cell. Studies in the TT cell line have demonstrated either no effect of 1,25(OH)2D3 [53] or stimulation [115,116] of cell growth. In the studies in which 1,25(OH)2D3 stimulated thymidine incorporation and cell growth, c-myc antisense DNA oligomers abolished the proliferative effect of 1,25(OH)2D3 but not the effect on inhibition of CT/CGRP gene expression [117]. These results suggest the potential for an independent effect of 1,25(OH)2D3 in this cell line, although more investigation is needed. Because of the contradictory effects of 1,25(OH)2D3 on cell growth, it is unclear at this point whether vitamin D analogs will have any role in the treatment of MTC.

CHAPTER 39 Vitamin D Control of the Calcitonin Gene in Thyroid C Cells

VIII. SUMMARY AND FUTURE DIRECTIONS The primary effect of 1,25(OH)2D3 in the C cell is to inhibit transcription of the CT/CGRP gene. The available evidence supports a model in which the VDR interacts with a partial VDRE and interferes with the positive function of several transcription factors, one of which is likely to be a CREB variant. Further definition of this model will require the identification of the specific transcription factors that interact with the upstream neuroendocrine-specific enhancer to regulate CT/CGRP gene transcription. Although these studies point to a clearly defined pathway for vitamin D action in the C cell, it is less clear whether this pathway has regulatory significance in normal physiology. Whereas the involvement of CT in calcium homeostasis was discovered 40 years ago, to a large degree the exact role of CT in normal physiology remains elusive. The development of CT/CGRP-α knockout mice has provided a new system for studying CT in calcium homeostasis and bone formation [118]. We have evidence that the CT/CGRP-α knockout mice display more profound hypercalcemia following treatment with either PTH [119] or 1,25(OH)2D3 (unpublished). Thus, treatment with 1,25(OH)2D3 appears to stimulate a greater increase in serum calcium in mice lacking CT and α-CGRP than in wild-type animals. This suggests a short-term physiologic role for CT and/or CGRP in the prevention of hypercalcemia. Future studies on knockout mice lacking only CT should be informative on the interplay between CT and vitamin D3 in calcium homeostasis. Finally, the speculative possibility that 1,25(OH)2D3 repression of CT/CGRP gene expression may have relevance for the pathological conditions of Williams syndrome and migraine is intriguing. CT is believed to be most important during times of calcium stress, such as during growth, when the balance of calcium levels most requires the actions of calciotropic hormones. The possibility that 1,25(OH)2D3 repression of the CT/CGRP gene may contribute to the infantile hypercalcemia of Williams syndrome deserves further exploration. Likewise, it is tempting to speculate that 1,25(OH)2D3 might be useful for dampening CGRP production as a possible prophylactic treatment for migraine. Further studies will be required to clarify these issues.

References 1. Le Douarin N 1982 The neural crest. In: Developmental and Cell Biology Series, Cambridge Univ. Press, Cambridge 12:1–259.

697

2. Ito K, Sieber-Blum M 1993 Pluripotent and developmentally restricted neural-crest-derived cells in posterior visceral arches. Dev Biol 156:191–200. 3. Payette RF, Bennett GS, Gershon MD 1984 Neurofilament expression in vagal neural crest-derived precursors of enteric neurons. Dev Biol 105:273–287. 4. Lanigan TM, DeRaad S, Russo AF 1998 Requirement of the MASH-1 transcription factor for neuroendocrine differentiation of thyroid C cells. J Neurobiol 34:126–134. 5. Nunez EA, Gershon MD 1972 Synthesis and storage of serotonin by parafollicular (C) cells of the thyroid gland of active, prehibernating and hibernating bats. Endocrinology 90:1008–1024. 6. Barasch JM, Mackey H, Tamir H, Nunez EA, Gershon MD 1987 Induction of a neural phenotype in a serotonergic endocrine cell derived from the neural crest. J Neurosci 7:2874–2883. 7. Tamir H, Liu KP, Payette RF, Hsiung SC, Adlersberg M, Nunez EA, Gershon MD 1989 Human medullary thyroid carcinoma: characterization of the serotonergic and neuronal properties of a neurectodermally derived cell line. J Neurosci 9:1199–1212. 8. Clark MS, Lanigan TM, Page NM, Russo AF 1995 Induction of a serotonergic and neuronal phenotype in thyroid C-cells. J Neurosci 15:6167–6178. 9. Russo AF, Clark MS, Durham PL 1996 Thyroid parafollicular cells. An accessible model for the study of serotonergic neurons. Mol Neurobiol 13:257–76. 10. Manley NR, Capecchi MR 1998 Hox group 3 paralogs regulate the development and migration of the thymus, thyroid, and parathyroid glands. Dev Biol 195:1–15. 11. Nishiyama I, Ogiso M, Oota T, Kimura T, Seki T 1996 Developmental change in expression of highly polysialylated neural cell adhesion molecule in C-cells in rat thyroid gland. Anat Embryol 194:419–426. 12. Schuchardt A, D’Agati V, Larsson-Blomberg L, Costantini F, Pachnis V 1994 Defects in the kidney and enteric nervous system of mice lacking the tyrosine kinase receptor ret. Nature 367:380–383. 13. Trupp M, Ryden M, Jornvall H, Funakoshi H, Timmusk T, Arenas E, Ibanez CF 1995 Peripheral expression and biological activities of GDNF, a new neurotrophic factor for avian and mammalian peripheral neurons. J Cell Biol 130:137–148. 14. Jing S, Wen D, Yu Y, Hoist PL, Luo Y, Fang M, Tamir R, Antonio L, Hu Z, Cupples R, Louis JC, Hu S, Altrock BW, Fox GM 1996 GDNF-induced activation of the Ret protein tyrosine kinase is mediated by GDNFR-α, a novel receptor for GDNF. Cell 85:1113–1124. 15. Durbec P, Marcos-Gutierrez CV, Kilkenny C, Suvanto P, Smith D, Ponder B, Costantini F, Saarma M, Sarlola H, Pachnis V 1996 GDNF signaling through the ret receptor tyrosine kinase. Nature 381:789–793. 16. Lindahl M, Poteryaev D, Yu L, Arumae U, Timmusk T, Bongarzone I, Aiello A, Pierotti MA, Airaksinen MS, Saarma M 2001 Human glial cell line-derived neurotrophic factor receptor alpha 4 is the receptor for persephin and is predominantly expressed in normal and malignant thyroid medullary cells. J Biol Chem 276:9344–9351. 17. Lindahl M, Timmusk T, Rossi J, Saarma M, Airaksinen MS 2000 Expression and alternative splicing of mouse Gfra4 suggest roles in endocrine cell development. Mol Cell Neurosci 15:522–533. 18. Sipple JH 1961 The association of pheochromocytoma with carcinoma of the thyroid gland. Am J Med 31:163–166.

698 19. Steiner AL, Goodman AD, Powers SR 1968 Study of a kindred with pheochromocytoma, medullary carcinoma, hyperparathyroidism and Cushing’s disease: Multiple endocrine neoplasia, type 2. Medicine 47:371–409. 20. Farndon JR, Leight GS, Dilley WG, Baylin SB, Smallridge RC, Harrison TS, Wells SA Jr 1986 Familial medullary thyroid carcinoma without associated endocrinopathies: A distinct clinical entity. Br J Surg 73:278–281. 21. Donis-Keller H, Shenshen D, Chi D, Carlson KM, Toshima K, Lairmore TC, Howe JR, Moley JF, Goodfellow P, Wells SA Jr 1993 Mutations in the RET proto-oncogene are associated with MEN 2A and FMTC. Hum Mol Genet 2:851–856. 22. Mulligan LM, Kwok JBJ, Healey CS, Elsdon MJ, Eng C, Gardner E, Love DR, Mole SE, Moore JK, Papi L, Ponder MA, Telenius H, Tunnacliffe A, Ponder BAJ 1993 Germline mutations of the RET proto-oncogene in multiple endocrine neoplasia type 2A (MEN 2A). Nature 363:458–460. 23. Hofstra RMW, Landsvater RM, Ceccherini I, Stulp RP, Stelwagen T, Luo Y, Pasini B, Hoppener JWM, van AmStel HKP, Romeo G, Lips CJM, Buys CHCM 1994 A mutation in the RET proto-onocogene associated with multiple endocrine neoplasia type 2B and sporadic medullary thyroid carcinoma. Nature 367:375–376. 24. Mulligan LM, Marsh DJ, Robinson BG, Lenoir G, Schuffenecker I, Zedenius J, Lips CJM, Gagel RF, Takai SI, Noll WW, Fink M, Raue F, LaCroix A, Thibodeau SN, Frilling A, Ponder BAJ, Eng C 1995 Genotype–phenotype correlation in multiple endocrine neoplasia type 2: Report of the international RET mutation consortium. J Int Med 238:343–346. 25. Asai N, Iwashita T, Matsuyama M, Takahashi M 1995 Mechanisms of activation of the ret proto-oncogene by multiple endocrine neoplasia 2A mutations. Mol Cell Biol 15: 1613–1619. 26. Santoro M, Carlomagno F, Romano A, Bottaro DP, Dathan NA, Grieco M, Fusco A, Vecchio G, Matoskova B, Kraus MH, Di Fiiore PP 1995 Activation of RET as a dominant transforming gene by germline mutations of MEN 2A and MEN 2B. Science 267:381–383. 27. Wohllk N, Cote GJ, Bugalho MMJ, Ordonez N, Evans DB, Goepfert H, Khorana S, Schultz PS, Richards CS, Gagel RF 1996 Relevance of RET proto-oncogene mutations in sporadic medullary thyroid carcinoma. J Clin Endocrinol Metab 81:3740–3745. 28. Amara SG, Jonas V, Rosenfeld MG, Ong ES, Evans RM 1982 Alternative RNA processing in calcitonin gene expression generates mRNAs encoding different polypeptide products. Nature 298:240–244. 29. Leff SE, Evans RM, Rosenfeld MG 1987 Splice commitment dictates neuron-specific alternative RNA processing in calcitonin/ CGRP gene expression. Cell 48:517–524. 30. Crenshaw EB, Russo AF, Swanson LW, Rosenfeld MG 1987 Neuron-specific alternative RNA processing in transgenic mice expressing a metallothionein–calcitonin fusion gene. Cell 49:389–398. 31. Cote GJ, Gagel RF 1986 Dexamethasone differentially affects the levels of calcitonin and calcitonin gene-related peptide mRNAs expressed in a human medullary thyroid carcinoma cell line. J Biol Chem 261:15524–15528. 32. Russo AF, Nelson C, Roos BA, Rosenfeld MG 1988 Differential regulation of the coexpressed calcitonin/α-CGRP and β-CGRP neuroendocrine genes. J Biol Chem 1263:5–8. 33. Russo AF, Lanigan TM, Sullivan BE 1992 Neuronal properties of a thyroid C-cell line: partial repression by dexamethasone and retinoic acid. Mol Endocrinol 6:207–218.

ANDREW F. RUSSO AND ROBERT F. GAGEL

34. Lou H, Gagel R 1998 Alternative RNA processing—its role in regulating expression of calcitonin/calcitonin gene-related peptide. J Endocrinol 156:401–405. 35. Yeakley JM, Hedjran F, Morfin JP, Merillat N, Rosenfeld MG, Emeson RB 1993 Control of calcitonin/calcitonin gene-related peptide pre-mRNA processing by constitutive intron and exon elements. Mol Cell Biol 13:5999–6011. 36. Emeson RB, Hedjran F, Yeakley JM, Guise JW, Rosenfeld MG 1989 Alternative production of calcitonin and CGRP mRNA is regulated at the calcitonin-specific splice acceptor. Nature 341:76–80. 37. Cote GJ, Stolow DT, Peleg S, Berget SM, Gagel RF 1992 Identification of exon sequences and an exon binding protein involved in alternative RNA splicing of calcitonin/CGRP. Nucleic Acids Res 20:2361–2366. 38. Adema GJ, Bovenberg RA, Jansz HS, Baas PD 1988 Unusual branch point selection involved in splicing of the alternatively processed calcitonin/CGRP-I pre-mRNA. Nucleic Acids Res 16:9513–9526. 39. van Oers CC, Adema GJ, Zandberg H, Moen TC, Baas PD 1994 Two different sequence elements within exon 4 are necessary for calcitonin-specific splicing of the human calcitonin/calcitonin gene-related peptide I pre-mRNA. Mol Cell Biol 14:951–960. 40. Lou H, Yang Y, Cote GJ, Berget SM, Gagel RF 1995 An intron enhancer containing a 5′ splice site sequence in the human calcitonin/calcitonin gene-related peptide gene. Mol Cell Biol 15:7135–7142. 41. Lou H, Cote GJ, Gagel RF 1994 The calcitonin exon and its flanking intronic sequences are sufficient for the regulation of human calcitonin/calcitonin gene-related peptide alternative RNA splicing. Mol Endocrinol 18:1618–1626. 42. Lou H, Gagel RF, Berget SM 1996 An intron enhancer recognized by splicing factors activates polyadenylation. Genes Dev 10:208–219. 43. Raue F, Deutschle I, Kiintzel C, Ziegler R 1984 Reversible diminished calcitonin secretion in the rat during chronic hypercalcemia. Endocrinology 115:2362–2367. 44. Naveh-Many T, Silver J 1988 Regulation of calcitonin gene transcription by vitamin D metabolites in vivo in the rat. J Clin Invest 81:270–273. 45. Fennandez-Santos J, Utrilla J, Conde E, Hevia A, Loda M, Martin-Lacave I 2001 Decrease in calcitonin and parathyroid hormone mRNA levels and hormone secretion under longterm hypervitaminosis D3 in rats. Histol Histopathol 16: 407–414. 46. Jones KL 1990 Williams syndrome: an historical perspective of its evolution, natural history, and etiology. Am J Med Genet (Suppl) 6:89–96. 47. Keating MT 1997 On the trail of genetic culprits in Williams syndrome. Cardiovasc Res 36:134–137. 48. Russo AF, Chamany K, Klemish SW, Hall TM, Murray JC 1991 Characterization of the calcitonin/CGRP gene in Williams syndrome. Am J Med Genet 39:28–33. 49. Kitagawa H, Fujiki R, Yoshimura K, Mezaki Y, Uematsu Y, Matsui D, Ogawa S, Unno K, Okubo M, Tokita A, Nakagawa T, Ito T, Ishimi Y, Nagasawa H, Matsumoto T, Yanagisawa J, Kato S 2003 The chromatin-remodeling complex WINAC targets a nuclear receptor to promoters and is impaired in Williams syndrome. Cell 113:905–917. 50. Shinki T, Ueno Y, DeLuca H, Suda T 1999 Calcitonin is a major regulator for the expression of renal 25-hydroxyvitamin D1α-hydroxylase gene in normocalcemic rats. Proc Natl Acad Sci USA 96:8253–8258.

CHAPTER 39 Vitamin D Control of the Calcitonin Gene in Thyroid C Cells

51. Murayama A, Takeyama K, Kitanaka S, Kodera Y, Kawaguchi Y, Hosoya T, Kato S 1999 Positive and negative regulations of the renal 25-hydroxyvitamin D3 1α-hydroxylase gene by parathyroid hormone, calcitonin, and 1α,25(OH)2D3 in intact animals. Endocrinology 140:2224–2228. 52. Peleg S, Abruzzese RV, Cooper CW, Gagel RF 1993 Downregulation of calcitonin gene transcription by vitamin D requires two widely separated enhancer sequences. Mol Endocrinol 7:999–1008. 53. Cote GJ, Rogers DO, Huang ES, Gagel RF 1987 The effect of 1,25-dihydroxyvitamin D3 treatment on calcitonin and calcitonin gene-related peptide mRNA levels in cultured human thyroid C-cells. Biochem Biophys Res Commun 149: 239–243. 54. Collignon H, Laborie C, Tahri E, el M’Selmi A, Garel JM 1992 Effects of dexamethasone, calcium and 1,25-dihydroxycholecalciferol on calcitonin and calcitonin gene-related peptide mRNA levels from the CA-77 C cell line. Thyroid 2:361–365. 55. Lamari Y, Tahri EH, Collignon H, Garel JM 1994 Steroid hormones and retinoic acid interact in the regulation of calcitonin and calcitonin gene-related peptide secretion and messenger ribonucleic acid levels in CA-77 C cells. Cell Mol Biol 40: 541–550. 56. Cooley LD, Elder FB, Knuth A, Gagel RF 1995 Cytogenetic characterization of three human and three rat medullary thyroid carcinoma cell lines. Cancer Genet Cytogenet 80: 138–149. 57. Carlomagno F, Salvatore D, Santoro M, de Franciscis V, Quadro L, Panariello L, Colantuoni V, Fusco A 1995 Point mutation of the RET proto-oncogene in the TT human medullary thyroid carcinoma cell line. Biochem Biophys Res Commun 207:1022–1028. 58. Cote GJ, Gould JA, Huang SC, Gagel RF 1987 Studies of short-term secretion of peptides produced by alternative RNA processing. Mol Cell Endocrinol 53:211–219. 59. Gagel RF, Palmer WN, Leonhart K, Chan L, Leong SS 1986 Somatostatin production by a human medullary thyroid carcinoma cell line. Endocrinology 118:1643–1651. 60. Berger CL, de Bustros A, Roos BA, Leong SS, Mendelsohn G, Gesell MS, Baylin SB 1984 Human medullary thyroid carcinoma in culture provides a model relating growth dynamics, endocrine cell differentiation, and tumor progression. J Clin Endocrinol Metab 59:338–343. 61. Nelkin BD, Chen KY, deBustros AC, Roos BA, Baylin SB 1989 Changes in calcitonin gene RNA processing during growth of a human medullary thyroid carcinoma cell line. Cancer Res 49:6949–6952. 62. Leong SS, Horoszewicz JS, Shimaoka K, Friedman M, Kawinski E, Song MJ, Ziegel R, Chu TM, Baylin S, Mirand EA 1981 A new cell line for the study of human medullary thyroid carcinoma. In: Advances in Thyroid Neoplasia. Itaha, Rome, Italy, pp. 95–108. 63. Ikeda K, Lu C, Weir EC, Mangin M, Broadus AE 1989 Transcriptional regulation of the parathyroid hormone-related peptide gene by glucocorticoids and vitamin D in a human C-cell line. J Biol Chem 264:15743–15746. 64. Rogers DG, Cote GJ, Huang ES, Gagel RF 1989 1,25Dihydroxy-vitamin D3 silences 3′,5′-cyclic adenosine monophosphate enhancement of somatostatin gene transcription in human thyroid C cells. Mol Endocrinol 3:547–551. 65. Lazaretti CM, Grauer A, Raue F, Ziegler R 1990 25Dihydroxy-vitamin D3 suppresses dexamethasone effects on calcitonin secretion. Mol Cell Endocrinol 71:13–18.

699

66. Muszynski M, Birnbaum RS, Roos BA 1983 Glucocorticoids stimulate the production of preprocalcitonin-derived secretory peptides by a rat medullary thyroid carcinoma cell line. J Biol Chem 258:11678–11683. 67. Nakagawa T, Nelkin BD, Baylin SB, deBustros A 1988 Transcriptional and posttranscriptional modulation of calcitonin gene expression by sodium n-butyrate in cultured human medullary thyroid carcinoma. Cancer Res 48: 2096–2100. 68. deBustros A, Baylin SB, Berger CL, Roos BA, Leong SS, Nelkin BD 1985 Phorbol esters increase calcitonin gene transcription and decrease c-myc mRNA levels in cultured human medullary thyroid carcinoma. J Biol Chem 260: 98–104. 69. Nelkin BD, de Bustros AC, Mabry M, Baylin SB 1989 The molecular biology of medullary thyroid carcinoma. A model for cancer development and progression. JAMA 261: 3130–3135. 70. Nelkin BD, Borges M, Mabry M, Baylin SB 1990 Transcription factor levels in medullary thyroid carcinoma cells differentiated by Harvey ras oncogene: c-jun is increased. Biochem Biophys Res Commun 170:140–146. 71. Nelkin BD, Rosenfeld KI, de Bustros A, Leong SS, Roos BA, Baylin SB 1984 Structure and expression of a gene encoding human calcitonin and calcitonin gene related peptide. Biochem Biophys Res Commun 123:648–655. 72. Monia YT, Peleg S, Gagel RF 1995 Cell type-specific regulation of transcription by cyclic adenosine 3′,5′-monophosphateresponsive elements within the calcitonin promoter. Mol Endocrinol 9:784–793. 73. deBustros A, Lee RY, Compton D, Tseng TY, Baylin SB, Nelkin BD 1990 Differential utilization of calcitonin gene regulatory DNA sequences in cultured lines of medullary thyroid carcinoma and small-cell lung carcinoma. Mol Cell Biol 10:1773–1778. 74. Stolarsky-Fredman LS, Leff SE, Klein ES, Crenshaw III EB, Yeakley J, Rosenfeld MG 1990 A tissue specific enhancer in the rat calcitonin/CGRP gene is active in both neural and endocrine cell types. Mol Endocrinol 4:497–504. 75. Peleg S, Abruzzese RV, Cote GJ, Gagel RF 1990 Transcription of the human calcitonin gene is mediated by a C cell-specific enhancer containing E-box-like elements. Mol Endocrinol 4:1750–1757. 76. Cote GJ, Abruzzese RV, Lips CJM, Gagel RF 1990 Transfection of calcitonin gene regulatory elements into a cell culture model of the C-cell. J Bone Miner Res 5:165–171. 77. deBustros A, Baylin SB, Levine MA, Nelkin BD 1986 Cyclic AMP and phorbol esters separately induce growth inhibition, calcitonin secretion, and calcitonin gene transcription in cultured human medullary thyroid carcinoma. J Biol Chem 261:8036–8041. 78. Watson A, Latchman D 1995 The cyclic AMP response element in the calcitonin/calcitonin gene-related peptide gene promoter is necessary but not sufficient for its activation by nerve growth factor. J Biol Chem 270:9655–9660. 79. Thiagalingam A, deBustros A, Borges M, Jasti R, Compton D, Diamond L, Mabry M, Ball DW, Baylin SB, Nelkin BD 1996 RREB, a novel zinc finger protein, is involved in the differentiation responses to ras in human medullary thyroid carcinoma. Mol Cell Biol 16:5335–5345. 80. Tverberg LA, Russo AF 1993 Regulation of the calcitonin/calcitonin gene-related peptide gene by cell-specific synergy between helix-loop-helix and octamer-binding transcription factors. J Biol Chem 268:15965–15973.

700 81. Lanigan TM, Tverberg LA, Russo AF 1993 Retinoic acid repression of cell-specific helix-loop-helix-octamer activation of the calcitonin/calcitonin gene-related peptide enhancer. Mol Cell Biol 13:6079–6088. 82. Durham PL, Sharma R, Russo AF 1997 Repression of the calcitonin gene-related peptide promoter by 5-HT1 receptor activation. J Neurosci 17:9545–9553. 83. Durham PL, Russo AF 2003 Stimulation of the calcitonin gene-related peptide enhancer by mitogen-activated protein kinases and repression by an antimigraine drug in trigeminal ganglia neurons. J Neurosci 23:807–815. 84. Ball DW, Compton D, Nelkin BD, Baylin SB, deBustros A 1992 Human calcitonin gene regulation by helix-loop-helix reognition sequences. Nucleic Acids Res 20:117–123. 85. Tverberg LA, Russo AF 1992 Cell-specific glucocorticoid repression of calcitonin/calcitonin gene-related peptide transcription. J Biol Chem 267:17567–17573. 86. Ball DW, Azzoli CG, Baylin SB, Chi D, Dou S, DonisKeller H, Cumaraswamy A, Borges M, Nelkin BD 1993 Identification of a human achaete-scute homolog highly expressed in neuroendocrine tumors. Proc Natl Acad Sci USA 90:5648–5652. 87. Guillemot F 1995 Analysis of the role of basic helix-loop-helix transcription factors in the development of neural lineages in the mouse. Biol Cell 84:3–6. 88. Baetscher M, Schmidt E, Shimizu A, Leder P, Fishman MC 1991 SV40 T antigen transforms calcitonin cells of the thyroid but not CGRP-containing neurons in transgenic mice. Oncogene 6:1133–1138. 89. Lanigan TM, Russo AF 1997 Binding of upstream stimulatory factor and a cell-specific activator to the calcitonin/calcitonin gene-related peptide enhancer. J Biol Chem 272:18316–18324. 90. Navankasattusas S, Sawadogo M, Van Bilsen M, Dang CV, Chien KR 1994 The basic helix-loop-helix protein upstream stimulating factor regulates the cardiac ventricular myosin light-chain 2 gene via independent cis regulatory elements. Mol Cell Biol 14:7331–7339. 91. Yoon SO, Chikaraishi DM 1994 Isolation of two E-box binding factors that interact with the rat tyrosine hydroxylase enhancer. J Biol Chem 269:18453–18462. 92. Martin CC, Svitek CA, Oeser JK, Henderson E, Stein R, O’Brien RM 2003 Upstream stimulatory factor (USF) and neurogenic differentiation β-cell E box transactivator 2 (NeuroD/BETA2) contribute to islet-specific glucose-6phosphatase catalytic-subunit-related protein (IGRP) gene expression. Biochem J 371:675–686. 93. Seger R, Krebs EG 1995 The MAPK signaling cascade. FASEB J 9:726–35. 94. Lindsay RM, Harmar AJ 1989 Nerve growth factor regulates expression of neuropeptide genes in adult sensory neurons. Nature 337:362–364. 95. Durham PL, Russo AF 1989 Serotonergic repression of mitogen-activated protein kinase control of the calcitonin generelated peptide enhancer. Molec Endocrinol 12:1002–1009. 96. Chen W, West A, Tao X, Corfas G, Szentirmay M, Sawadogo M, Vinson C, Greenberg M 2003 Upstream stimulatory factors are mediators of Ca+2–responsive transcription in neurons. J Neurosci 23:2572–2581. 97. Riccio A, Pedone PV, Lund LR, Olesen T, Olsen HS, Andreasen PA 1992 Transforming growth factor beta 1responsive element: Closely associated binding sites for USF and CCAAT-binding transcription factor-nuclear factor I in the type 1 plasminogen activator inhibitor gene. Mol Cell Biol 12:1846–1855.

ANDREW F. RUSSO AND ROBERT F. GAGEL

98. Tabuchi A, Sakaya H, Kisukeda T, Fushiki H, Tsuda M 2002 Involvement of an upstream stimulatory factor as well cAMP-responsive element–binding protein in the activation of brain-derived neurotrophic factor gene promoter I. J Biol Chem 277:35920–35931. 99. Galibert MD, Carreira S, Goding CR 2001 The Usf-1 transcription factor is a novel target for the stress-responsive p38 kinase and mediates UV-induced Tyrosinase expression. EMBO J 20:5022–5031. 100. Ferrari MD, Saxena PR 1993 On serotonin and migraine: a clinical and pharmacological review. Cephalalgia 13:151–165. 101. Rosen LB, Ginty DD, Weber MJ, Greenberg ME 1994 Membrane depolarization and calcium influx stimulate MEK and MAP kinase via activation of Ras. Neuron 12: 1207–1221. 102. MacArthur L, Eiden L 1996 Neuropeptide genes: targets of activity-dependent signal transduction. Peptides 17: 721–728. 103. Buzzi MG, Bonamini M, Moskowitz MA 1995 Neurogenic model of migraine. Cephalalgia 15:277–280. 104. Donaldson LF, Harmar AJ, McQueen DS, Seckl JR 1992 Increased expression of preprotachykinin, calcitonin gene-related peptide, but not vasoactive intestinal peptide messenger RNA in dorsal root ganglia during the development of adjuvant monoarthritis in the rat. Mol Brain Res 16:143–149. 105. Zhang L, Hoff AO, Wimalawansa SJ, Cote GJ, Gagel RF, Westlund KN 2001 Arthritic calcitonin/α calcitonin generelated peptide knockout mice have reduced nociceptive hypersensitivity. Pain 89:265–273 106. Durham PL, Russo AF 2002 New insights into the molecular actions of serotonergic antimigraine drugs. Pharmacol Ther 94:77–92. 107. Keyse SM 1995 An emerging family of dual specificity MAP kinase phosphatases. Biochim Biophys Acta 1265:152–160. 108. Horiuchi M, Hayashida W, Kambe T, Yamada T, Dzau VJ 1997 Angiotensin type 2 receptor dephosphorylates Bcl-2 by activating mitogen-activated protein kinase phosphatase-1 and induces apoptosis. J Biol Chem 272:19022–19026. 109. Kusari AB, Byon J, Bandyopadhyay D, Kenner KA, Kusari J 1997 Insulin-induced mitogen-activated protein (MAP) kinase phosphatase-1 (MKP-1) attenuates insulin-stimulated MAP kinase activity: A mechanism for the feedback inhibition of insulin signaling. Mol Endocrinol 11:1532–1543. 110. Durham PL, Russo AF 2000 Differential regulation of mitogenactivated protein kinase-responsive genes by the duration of a calcium signal. Mol Endocrinol 14:1570–1582. 111. Berridge MJ, Bootman MD, Roderick HL 2003 Calcium: Calcium signalling: dynamics, homeostasis and remodelling. Nat Rev Mol Cell Biol 4:517–529. 112. Ghosh A, Greenberg ME 1995 Calcium signaling in neurons: Molecular mechanisms and cellular consequences. Science 268:239–247. 113. Scimeca JC, Servant MJ, Dyer JO, Meloche S 1997 Essential role of calcium in the regulation of MAP kinase phosphatase-1 expression. Oncogene 15:717–725. 114. Limmroth V, Katsarava Z, Fritsche G, Przywara S, Diener HC 2002 Features of medication overuse headache following overuse of different acute headache drugs. Neurology 59: 1011–1014. 115. Lazaretti-Castro M, Grauer A, Baier R, Vieira JG, Ziegler R, Raue F 1995 1,25-Dihydroxyvitamin D3 stimulates growth and inhibits calcitonin secretion in a human C cell carcinoma cell line. Braz J Med Biol Res 28:1013–1018.

CHAPTER 39 Vitamin D Control of the Calcitonin Gene in Thyroid C Cells

116. Baier R, Grauer A, Lazaretti-Castro M, Ziegler R, Raue F 1994 Differential effects of 1,25-dihydroxyvitamin D3 on cell proliferation and calcitonin gene expression. Endocrinology 135:2006–2011. 117. Grauer A, Baier R, Ziegler R, Raue F 1995 Crucial role of c-myc in 1,25(OH)2D3 control of C-cell-carcinoma proliferation. Biochem Biophys Res Commun 213:922–927.

701

118. Zaidi M, Moonga BS, Abe E 2002 Calcitonin and bone formation: a knockout full of surprises. J Clin Invest 110:1769–1771. 119. Hoff AO, Catala-Lehnen P, Thomas PM, Priemel M, Rueger JM, Nasonkin I, Bradley A, Hughes MR, Ordonez N, Cote GJ, Amling M, Robert F, Gagel RF 2002 Increased bone mass is an unexpected phenotype associated with deletion of the calcitonin gene. J Clin Invest 110:1849–1857.

CHAPTER 40

Vitamin D Regulation of Type I Collagen Expression in Bone BARBARA E. KREAM* AND ALEXANDER C. LICHTLER Departments of Medicine* and Genetics and Developmental Biology, The University of Connecticut Health Center, Farmington, Connecticut

I. Introduction II. Regulation of Bone Collagen Synthesis III. Molecular Mechanisms of Regulation

IV. Conclusions and Perspectives References

I. INTRODUCTION

effect of vitamin D on bone mineralization may be due at least in part to its stimulation of intestinal calcium absorption [13]. However, the discovery of high-affinity 1,25(OH)2D3 receptors (VDRs) in cytosolic extracts of embryonic chick and fetal rat calvariae more than two decades ago suggested that 1,25(OH)2D3 also has direct effects on osteoblast function [14,15]. One hallmark effect of 1,25(OH)2D3 action in bone is its ability to stimulate osteoclastic bone resorption [16]. 1,25(OH)2D3 increases osteoclast formation and bone resorption by signaling in cells of the osteoblast lineage [17]. 1,25(OH)2D3 increases the expression of macrophage colony-stimulating factor, which enhances proliferation and differentiation of osteoclast precursors, and receptor activator of nuclear factor-kappaB ligand (RANKL), which increases osteoclast differentiation and survival [17,18]. In addition to its effect on osteoclast formation and resorption, 1,25(OH)2D3 regulates the expression of matrix proteins in osteoblasts, including type I collagen [19], osteocalcin [20], and osteopontin [21]. Type I collagen is the most abundant protein in the body and comprises at least 90% of the organic component of the bone matrix. The biochemistry, molecular biology, and hormonal regulation of collagen genes have been extensively reviewed [22–27]. Type I collagen is produced at high levels by differentiated osteoblasts and is required for the formation of the mineralized bone matrix. Many other cell types synthesize and secrete type I collagen, although in lesser amounts. Each type I collagen molecule consists of three polypeptides: two α1(I) chains and one α2(I) chain. These polypeptides are encoded by separate genes (Col1a1 and Col1a2, respectively) that are expressed in a 2:1 ratio [28]. Collagen synthesis in bone is modulated by a variety of hormones, growth factors, and cytokines, some of which are produced locally by osteoblasts [27,29–31].

Vitamin D has multiple functions in humans and animals [1,2]. It is probably best known as a nutrient required for adequate growth and mineralization of bone. Vitamin D, like parathyroid hormone (PTH), is an important calcium-regulating hormone. PTH is primarily responsible for the acute physiologic maintenance of serum calcium levels. In the presence of prolonged hypocalcemia, PTH increases the renal production of 1,25-dihydroxyvitamin D3 (1,25(OH)2D3), the active metabolite of vitamin D. 1,25(OH)2D3, together with PTH, normalizes serum calcium levels by increasing intestinal calcium transport, bone resorption, and renal calcium reabsorption. To carry out many of its biological actions in its target cells, 1,25(OH)2D3 binds with high affinity to a nuclear receptor, the vitamin D receptor (VDR), which in turn binds to DNA promoter elements and recruits coactivators to regulate gene transcription [3–5]. The mechanisms by which 1,25(OH)2D3 affects bone mineralization are of utmost importance and have been the subject of intense study. Some of the most well recognized features of vitamin D deficiency are undermineralized bone in children (rickets) and adults (osteomalacia), a reduction in bone matrix formation [6] and mineralization [7], and an alteration in the pattern of collagen crosslinking [8]. Calcium deficiency also decreases bone formation and mineralization in rats [9]. Defective bone formation and mineralization in vitamin D–deficient rats can be largely corrected by the administration of calcium and phosphate, suggesting that the trophic effect of vitamin D on the skeleton may be due to its ability to stimulate intestinal calcium absorption [10–12]. Accordingly, preservation of mineral homeostasis in VDR null mice reverses the abnormal skeletal phenotype (including excessive osteoid production) seen in these animals, suggesting that a primary VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

704 Insulin [32], insulin-like growth factor [33], and transforming growth factor-β [34] increase type I collagen synthesis, whereas PTH [35], interleukin-1 [36], and tumor necrosis factor [36] are inhibitory. Glucocorticoids [37] and prostaglandins [38] can be either stimulatory or inhibitory in vitro depending on the model and culture conditions. Because of the critical role of type I collagen in maintaining the structure and function of the skeleton, it is important to understand the mechanisms by which 1,25(OH)2D3 regulates type I collagen expression.

II. REGULATION OF BONE COLLAGEN SYNTHESIS Collagen synthesis in organ cultures of rodent calvariae and cell cultures has been assessed using several different assays [39,40]. In the most widely used assay, calvariae and cells are incubated with tritiated proline for several hours prior to the end of culture. The incorporation of tritiated proline into collagenase-digestible protein (CDP labeling) and noncollagen protein (NCP labeling) is measured in extracts of the cultures using highly purified bacterial collagenase [41]. The percent collagen synthesis is calculated from the CDP and NCP values after correcting for the greater abundance of proline in collagen relative to noncollagen proteins [42]. Collagen production can also be determined by measuring the hydroxyproline content of cell or organ cultures, since hydroxyproline is virtually unique to collagens. These methods do not distinguish between different types of fibrillar collagen. However, the collagen synthesized by bone organ cultures and most osteoblastic cell cultures is largely type I (>95%) so that the CDP labeling value usually reflects type I collagen synthesis. If desired, the production of different collagen types can be distinguished by ion-exchange chromatography and polyacrylamide gel electrophoresis of radiolabeled extracts of cell or organ cultures [43,44]. Type I collagen expression in human cells cultures has also been assessed by measuring secretion of the procollagen I C-terminal propeptide [45,46]. Finally, the use of specific cDNA probes in Northern blotting and allele specific primers in reverse transcriptase–polymerase chain reaction assays can be used to assess collagen mRNA expression in bone models. As discussed later, measurement of the effect of 1,25(OH)2D3 on collagen synthesis and mRNA levels has given comparable results using these different assays. 1,25(OH)2D3 inhibits collagen synthesis in organ cultures of 21-day fetal rat calvariae [19] and neonatal mouse calvariae [47] with little or no effect on noncollagen protein synthesis. Maximal inhibition of collagen

BARBARA E. KREAM AND ALEXANDER C. LICHTLER

synthesis by 1,25(OH)2D3 in rat calvariae (about 50%) occurs at 10 nM [19]. 1,24R,25-(OH)3D3 also inhibits collagen synthesis but is less potent than 1,25(OH)2D3 [19]. 25-(OH)D3 and 24R,25(OH)2D3 do not alter collagen synthesis below 100 nM [19,47]. Vitamin D metabolites inhibit collagen synthesis and stimulate resorption of fetal rat long bones with similar relative potencies that correlate with the affinity of the metabolites for the skeletal VDRs [48]. To determine the cell selectivity of the 1,25(OH)2D3 inhibition of collagen synthesis, organ cultures of fetal rat calvariae were treated with 1,25(OH)2D3 for 22 hr and then radiolabeled with tritiated proline for the final 2 hr of culture. The central bone (mature osteoblasts) was dissected free of the periosteum (less mature osteoprogenitors and fibroblasts) and both compartments were analyzed separately for the incorporation of tritiated proline. 1,25(OH)2D3 decreases collagen synthesis in the central bone but not the periosteum, indicating selectivity of the 1,25(OH)2D3 effect for mature osteoblasts [49, 50]. Using an in vivo protocol in which neonatal rats are given multiple injections of tritiated proline to radiolabel newly synthesized bone matrix, 25 ng of 1,25(OH)2D3 given on days 1, 3, and 5 inhibited bone matrix synthesis as assessed by histomorphometry of autoradiographs of tibia and calvariae [51]. 1,25(OH)2D3 also inhibits collagen production in rat osteoblastic osteosarcoma ROS 17/2.8 cells [52], primary rat [53,54] and mouse osteoblastic cells [55], and an immortalized murine osteoblast cell line (MMB-1) [56]. 1,25(OH)2D3 has a greater inhibitory effect on type I collagen synthesis during log phase growth of primary murine osteoblastic cells than at confluence, perhaps because proliferating cells contained more VDRs [57]. Likewise, 1,25(OH)2D3 inhibition of collagen synthesis is greater in sparse cultures of MMB-1 cells that have higher VDR levels than confluent MMB-1 cells [58]. 1,25(OH)2D3 inhibition of collagen synthesis is equivalent in sparse and confluent rat primary osteoblastic cells [53], but VDR number did not change during growth of the cells [59]. Taken together, these data show that the extent of inhibition of collagen synthesis by 1,25(OH)2D3 is largely determined by the cellular quantity of VDRs. 1,25(OH)2D3 inhibits collagen mRNA levels during the proliferative phase of long-term cultures of rat primary osteoblastic cells [60] and prevents the formation of mineralized bone nodules by these cultures [60,61]. These studies show that 1,25(OH)2D3 inhibits the differentiation of osteoprogenitors that form mineralized nodules in primary rat osteoblastic cell cultures [61]. However, the inhibition of nodule formation by 1,25(OH)2D3 may be secondary to the suppression of type I collagen synthesis in the cultures.

CHAPTER 40 Vitamin D and Collagen Expression

In contrast to the inhibitory effects described earlier, 1,25(OH)2D3 transiently stimulates collagen and noncollagen protein synthesis (about twofold), which peaks at 12–24 hr, in the immortalized murine osteoblastic cell line MC3T3-E1 [62]. In this study, the percent collagen synthesized by the cultures (collagen relative to total protein synthesis) was not reported; as a result, it was not possible to determine the selectivity of the 1,25(OH)2D3 effect for collagen synthesis. 1,25(OH)2D3 also increases collagen expression in the human osteoblastic osteosarcoma cell line MG-63 [45,63] and primary cultures of human osteoblastic cells [64]. However, in other studies, 1,25(OH)2D3 has been shown to decrease the percent collagen synthesis in MC3T3-E1 cells [65,66]. MC3T3-E1 and MG-63 represent preosteoblastic cells that undergo in vitro osteogenic differentiation with ascorbic acid; 1,25(OH)2D3 inhibits cell growth and increases osteocalcin expression and alkaline phosphatase activity in both cell lines. MC3T3-E1 cells, like most immortalized osteoblastic cell lines, display significant phenotypic variation [67]. Therefore some of these discrepant results may be due to variations in the cells used for the experiments. Collectively, these data suggest that 1,25(OH)2D3 may act as a differentiating hormone in early cells of the osteoblast lineage, which results in increased type I collagen expression. In contrast, 1,25(OH)2D3 inhibits type I collagen expression in mature osteoblasts.

III. MOLECULAR MECHANISMS OF REGULATION Initial studies showed that 1,25(OH)2D3 represses collagen synthesis in mature osteoblasts at a pretranslational level [49]. Measurements of procollagen mRNA activity by translation of total RNA in a reticulocyte lysate first showed that 1,25(OH)2D3 inhibited collagen mRNA in the osteoblast-rich central bone but not the periosteum of 21-day fetal rat calvariae [49]. 1,25(OH)2D3 at 10 nM inhibited procollagen mRNA activity at 6 hr; maximal inhibition of about 50% occurred at 24 hr [49]. A single subcutaneous injection of 1,25(OH)2D3 (1.6 ng/g body weight) also decreased procollagen mRNA activity in calvariae [49]. Subsequently, specific cDNA probes were used to show that 1,25(OH)2D3 inhibited Col1a1 mRNA levels in ROS 17/2.8 cells [52] and primary rat [54] and chick calvarial osteoblastic cells [68,69]. Nuclear run-on assays in ROS 17/2.8 cells demonstrate that 1,25(OH)2D3 represses Col1a1 and Col1a2 mRNA levels by a transcriptional mechanism [70]. 1,25(OH)2D3 at 1 and 10 nM decreased the rate of Col1a1 and Col1a2 transcription by

705 about 50%, similar to its effect on collagen synthesis and type I collagen mRNA levels, whereas actin and tubulin transcription were unaffected. 1,25(OH)2D3 repressed Col1a1 and Col1a2 transcription as early as 4 hr with maximal inhibition at 24 hr [70]. DNA motifs that mediate stimulatory effects of 1,25(OH)2D3 on gene expression have been well characterized for several genes [5,71,72]. Vitamin D responsive elements (VDREs) that mediate 1,25(OH)2D3 induction of target genes such as human [73] and rat [74] osteocalcin, mouse osteopontin [75], rat 24-hydroxylase [76], and rat calbindin D-9K [77] contain two perfect or imperfect direct hexameric repeats of the consensus AGGTCA motif separated by three spacer nucleotides [5,71,72]. The consensus VDRE binds a heterodimer of the VDR and the retinoic acid X receptor (RXR) [78]. Negative promoter elements have also been identified. The negative VDRE in the avian PTH promoter is analogous to the consensus VDRE, since it contains two imperfect direct repeats separated by three spacer nucleotides and binds VDR and RXR [79]. In contrast, the negative VDRE in the human PTH gene contains a single AGGTTC motif, and binding of the VDR to this site does not require RXR [80,81]. The negative VDRE of the parathyroid hormone–related protein (PTHrP) gene contains two potential VDREs, one similar to the negative VDRE in the human PTH gene and another identical to the stimulatory VDRE; both motifs bind the VDR [82]. To characterize the regions of the Col1a1 gene that are involved in its repression by 1,25(OH)2D3, we produced a chimeric gene containing a fragment of the rat Col1a1 gene extending from −3518 to +116 bp fused to the chloramphenicol acetyl transferase (CAT) reporter gene termed ColCAT3.6 [83]. 1,25(OH)2D3 inhibited ColCAT3.6 activity in transiently transfected ROS 17/2.8 cells by 50%, similar its effect on the endogenous Col1a1 gene [83]. We then generated a series of ColCAT constructs containing progressive 5′ promoter deletions of the Col1a1 promoter to map 1,25(OH)2D3 response elements [84,85]. In stably transfected cells, 1,25(OH)2D3 inhibited a Col1a1 promoter fragment deleted to −2295 bp (ColCAT2.3) but did not affect a promoter fragment deleted to −1670 bp [85]. These experiments localized an inhibitory 1,25(OH)2D3 element to a region of the Col1a1 promoter from −2295 to −1670 bp. Sequence analysis of the Col1a1 promoter revealed a site between −2240 and −2234 bp that had high homology to both the human and rat osteocalcin VDREs. We hypothesized that the VDR binding to this motif would inhibit Col1a1 transcription. Electrophoretic mobility shift assays using VDR expressed in COS cells or by an adenovirus vector demonstrated that the VDR bound to this sequence in vitro [85].

706 However, deletion of the sequence between −2256 and −2216 bp from the ColCAT3.6 or ColCAT2.3 constructs did not affect the inhibitory effect of 1,25(OH)2D3 on promoter activity [85]. Therefore, 1,25(OH)2D3 does not inhibit Col1a1 transcription in ROS 17/2.8 cells solely via the −2240/−2234 bp site. To determine the effect of 1,25(OH)2D3 on Col1a1 promoter activity in vivo, we produced a series of transgenic mice lines carrying ColCAT constructs [86,87]. 1,25(OH)2D3 inhibited ColCAT3.6 activity in organ cultures of 6- to 8-day-old transgenic mouse calvariae [88]. 1,25(OH)2D3 inhibited CAT mRNA as early as 3 hr, and maximal inhibition of CAT mRNA (50%) was seen at 24 hr. The inhibition of CAT mRNA by 1,25(OH)2D3 was not affected by cycloheximide, suggesting that new protein synthesis is not required for the effect. A series of Col1a1 promoter fragments deleted to −1719 bp were fully inhibited by 1,25(OH)2D3; however, a Col1a1 promoter construct deleted to −1670 could not be analyzed because it did not have detectable basal activity in transgenic calvariae [89]. Subsequently, we showed that the rat Col1a1 promoter contains a homeodomain protein motif immediately downstream from –1683 bp that is required for high levels of promoter expression in osteoblasts in vivo [87]. A similar element is also present in the rat Col1a1 promoter [90]. In organ cultures of transgenic mouse calvariae carrying ColCAT constructs, we found that 1,25(OH)2D3 inhibited CAT activity when the promoter was further deleted to –1683 bp. Moreover, in a transgene having the –1719 bp promoter with a large internal deletion extending from –1284 to –318 bp, the inhibitory action of 1,25(OH)2D3 promoter activity was maintained (A. Ivkovic, A. C. Lichtler, and B. E. Kream, unpublished). Taken together, studies in ROS 17/2.8 cells and transgenic calvariae suggest that down-regulation of the Col1a1 promoter by 1,25(OH)2D3 involves sites located between –1683/–1284 bp or in the proximal promoter downstream from –318 bp. There are no good matches to consensus VDREs within these regions, suggesting several possible mechanisms. For one, 1,25(OH)2D3 repression of Col1a1 could involve binding of the VDR to a novel negative VDRE. Another possibility is that 1,25(OH)2D3 inhibition of Col1a1 expression involves displacement of a stimulatory transcription factor(s) from its cognate DNA binding site, similar to the mechanism by which 1,25(OH)2D3 inhibits the interleukin-2 gene [91]. It is also possible that 1,25(OH)2D3 inhibition of Col1a1 involves interaction of the VDR with other transcription factors rather than binding of the VDR to DNA. Such a mechanism has been described for the inhibition of collagenase expression by glucocorticoids [92,93].

BARBARA E. KREAM AND ALEXANDER C. LICHTLER

Finally, 1,25(OH)2D3 repression of Col1a1 expression could be mediated by an alternative signal transduction pathway. It has been suggested that some biological effects of 1,25(OH)2D3 may be mediated by the protein kinase C (PKC) signaling pathway [72]. We have shown that stimulation of PKC with phorbol myristate acetate inhibits collagen synthesis in fetal rat calvariae [94] and ColCAT3.6 expression in transgenic mouse calvariae [95]. Therefore, 1,25(OH)2D3 activation of the PKC pathway might inhibit Col1a1 expression. This could be mediated by a putative 1,25(OH)2D3 membrane receptor, which activates intracellular signal transduction pathways leading to alteration of gene transcription. Future experiments to identify 1,25(OH)2D3 response elements in the Col1a1 gene will involve the analysis of additional constructs having selected site-directed mutations and internal promoter deletions in cultured osteoblastic cells and transgenic mice.

IV. CONCLUSIONS AND PERSPECTIVES The effect of 1,25(OH)2D3 (inhibitory or stimulatory) may depend in part on in vitro culture conditions such as cell density, the timing and concentration of 1,25(OH)2D3 addition, the presence of ascorbic acid, and the state of maturation of the model. A model has been proposed based on the premise that cells of the osteoblast lineage differ in their response to 1,25(OH)2D3 depending on their state of maturation [96]. 1,25(OH)2D3 stimulates osteoblast markers in immature osteoprogenitor cells (MC3T3-E1 and MG-63 cells) but inhibits these markers in mature osteoblasts (rodent calvarial organ cultures, primary rodent osteoblastic cell cultures, and ROS 17/2.8 cells) [96]. Such a model is consistent with the effects of 1,25(OH)2D3 on bone remodeling during periods of calcium and phosphate deficiency. When serum calcium and phosphate are low, PTH increases the synthesis of 1,25(OH)2D3. Both hormones increase bone resorption to increase the supply of calcium and phosphate for soft tissues. During periods of mineral deficiency, it would be appropriate for 1,25(OH)2D3 to repress collagen synthesis and inhibit the differentiation of late osteoprogenitors as a means of temporarily limiting new bone formation. Such an effect would prevent calcium and phosphate from being redeposited at sites of new osteoid formation. At the same time, 1,25(OH)2D3 may stimulate the differentiation of early osteoprogenitors to differentiation into a new cohort of osteoblasts that would initiate the phase of coupled formation [96].

CHAPTER 40 Vitamin D and Collagen Expression

References 1. Raisz LG 1980 Direct effects of vitamin D and its metabolites on skeletal tissue. Clin Endocrinol Metab 9:27–41. 2. DeLuca HF 1986 The metabolism and functions of vitamin D Adv Exp Med Biol 196:361–375. 3. Haussler MR, Whitfield GK, Haussler CA, Hsieh JC, Thompson PD, Selznick SH, Dominguez CE, Jurutka PW 1998 The nuclear vitamin D receptor: biological and molecular regulatory properties revealed. J Bone Miner Res 13:325–349. 4. Rachez C, Freedman LP 2000 Mechanisms of gene regulation by vitamin D3 receptor: A network of coactivator interactions. Gene 246:9–21. 5. Christakos S 2002 Vitamin D gene regulation. In: Bilezikian J, Raisz LG, Rodan GA (eds) Principles of Bone Biology, Vol. 1. Academic Press, San Diego, pp. 573–586. 6. Canas F, Brand JS, Neuman WF, Terepka AR 1969 Some effects of vitamin D3 on collagen synthesis in rachitic chick cortical bone. Am J Physiol 216:1092–1096. 7. Baylink D, Stauffer M, Wergedal J, Rich C 1970 Formation, mineralization and resorption of bone in vitamin D–deficient rats. J Clin Invest 49:1122–1134. 8. Mechanic GL, Toverud SU, Ramp WK, Gonnerman WA 1975 The effect of vitamin D on the structural crosslinks and maturation of chick bone collagen. Biochim Biophys Acta 393:419–425. 9. Stauffer M, Baylink D, Wergedal J, Rich C 1973 Decreased bone formation, mineralization, and enhanced resorption in calcium-deficient rats. Am J Physiol 225:269–276. 10. Howard GA, Baylink DJ 1980 Matrix formation and osteoid maturation in vitamin D–deficient rats made normocalcemic by dietary means. Mineral Electrolyte Metab 3:44–50. 11. Underwood JL, DeLuca HF 1984 Vitamin D is not directly necessary for bone growth and mineralization. Am J Phsyiol 246:E493–E498. 12. Weinstein RS, Underwood JL, Hutson MS, DeLuca HF 1984 Bone histomorphometry in vitamin D–deficient rats infused with calcium and phosphorus. Am J Physiol 246: E499–E505. 13. Amling M, Priemel M, Holzmann T, Chapin K, Rueger JM, Baron R, Demay MB 1999 Rescue of the skeletal phenotype of vitamin D receptor-ablated mice in the setting of normal mineral ion homeostasis: Formal histomorphometric and biomechanical analyses. Endocrinology 140:4982–4987. 14. Kream BE, Jose M, Yamada S, DeLuca HF 1977 A specific high-affinity binding macromolecule for 1,25-dihydroxyvitamin D3 in fetal bone. Science 197:1086–1088. 15. Manolagas SC, Taylor CM, Anderson DC 1979 Highly specific binding of 1,25-dihydroxycholecalciferol in bone. J Endocrinol 80:35–39. 16. Raisz LG, Trummel CL, Schnoes HK, DeLuca HF 1972 1,25Dihydroxycholecalciferol: A potent stimulator of bone resorption in tissue culture. Science 175:768–769. 17. Suda T, Takahashi N, Martin TJ 1992 Modulation of osteoclast differentiation. Endocrine Rev 13:66–80. 18. Suda T, Takahashi N, Udagawa N, Jimi E, Gillespie MT, Martin TJ 1999 Modulation of osteoclast differentiation and function by the new members of the tumor necrosis factor receptor and ligand families. Endocr Rev 20:345–357. 19. Raisz LG, Maina DM, Gworek SC, Dietrich JW, Canalis EM 1978 Hormonal control of bone collagen synthesis in vitro: Inhibitory effect of 1-hydroxylated vitamin D metabolites. Endocrinology 102:731–735.

707 20. Price PA, Baukol SA 1980 1,25-Dihydroxyvitamin D3 increases synthesis of the vitamin K–dependent bone protein by osteosarcoma cells. J Biol Chem 255:11660–11663. 21. Prince CW, Butler WT 1987 1,25-Dihydroxyvitamin D3 regulates the biosynthesis of osteopontin, a bone-derived cell attachment protein, in clonal osteoblast-like osteosarcoma cells. Coll Relat Res 7:305–313. 22. Raghow R, Thompson JP 1989 Molecular mechanisms of collagen gene expression. Mol Cell Biochem 86:5–18. 23. Vuorio E, de Crombrugghe B 1990 The family of collagen genes. Ann Rev Biochem 59:837–872. 24. Ramirez F, Di Liberto M 1990 Complex and diversified regulatory programs control the expression of vertebrate collagen genes. FASEB J 4:1616–1623. 25. van der Rest M, Garrone R 1991 Collagen family of proteins. FASEB J 5:2814–2823. 26. Rossert J, Terraz C, Dupont S 2000 Regulation of type I collagen genes expression. Nephrol Dial Transplant 15(Suppl 6):66–68. 27. Rossert J, de Crombrugghe B 2002 Type I collagen: Structure, synthesis, and regulation. In Bilezikian J, Raisz LG, Rodan GA (eds) Principles of Bone Biology, Vol. 1. Academic Press, San Diego, pp. 189–210. 28. Vuurst J, Sobel ME, Martin GR 1985 Regulation of type I collagen synthesis: Total proα1(I) and proα2(I) mRNAs are maintained in a 2:1 ratio under varying rates of collagen synthesis. Eur J Biochem 151:449–453. 29. Raisz LG, Kream BE 1983 Regulation of bone formation. N Engl J Med 309:29–35; 83–89. 30. Raisz LG 1988 Bone metabolism and its hormonal regulation. Triangle 27:5–10. 31. Canalis E, McCarthy TL, Centrella M 1991 Growth factors and cytokines in bone cell metabolism. Ann Rev Med 42:17–24. 32. Kream BE, Smith MD, Canalis E, Raisz R 1985 Characterization of the effect of insulin on collagen synthesis in fetal rat bone. Endocrinology 116:296–302. 33. Canalis E 1980 Effect of insulin-like growth factor-I on DNA and protein synthesis in cultured rat calvaria. J Clin Invest 66:709–719. 34. Centrella M, Casinghino S, Ignotz R, McCarthy TL 1992 Multiple regulatory effects by transforming growth factor-beta on type I collagen levels in osteoblast-enriched cultures from fetal rat bone. Endocrinology 131:2863–2872. 35. Kream BE, Rowe DW, Gworek SC, Raisz LG 1980 Parathyroid hormone alters collagen synthesis and procollagen mRNA levels in fetal rat calvaria. Proc Natl Acad Sci USA 77:5654–5658. 36. Harrison JR, Vargas CJ, Petersen DN, Lorenzo JA, Kream BE 1990 Interleukin-1α and phorbol ester inhibition collagen synthesis by a transcriptional mechanism. Mol Endocrinol 4:184–190. 37. Hodge BO, Kream BE 1988 Variable effects of dexamethasone on protein synthesis in clonal rat osteosarcoma cells. Endocrinology 122:2127–2133. 38. Raisz LG, Fall PM 1990 Biphasic effects of prostaglandin E2 on bone formation in cultured fetal rat calvariae: Interaction with cortisol. Endocrinology 126:1654–1659. 39. Stern PH, Raisz LG 1979 Organ culture of bone. In Simmons DJ, Kunin AS (eds) Skeletal Research: An Experimental Approach, Academic Press, New York, pp. 21–59. 40. Gronowicz G, Raisz LG 1996 Bone formation assays. In Raisz LG, Rodan GA, and Bilezikian JP (eds) Principles of Bone Biology. Academic Press, San Diego, pp. 1253–1265. 41. Peterkofsky B, Diegelmann R 1971 Use of a mixture of proteinase-free collagenases for the specific assay of radioactive collagen in the presence of other proteins. Biochemistry 6: 988–994.

708 42. Diegelmann RF, Peterkofsky B 1972 Collagen biosynthesis during connective tissue development in chick embryo. Dev Biol 28:443–453. 43. Sykes BD, Puddle B, Francis M, Smith R 1976 The estimation of two collagens from human dermis by interrupted gel electrophoresis. Biochem Biophys Res Comm 72:1472–1480. 44. Sage H, Bornstein P 1982 Preparation and characterization of procollagens and procollagen-collagen intermediates. In Cunningham LW, Frederiksen DW (eds) Methods in Enzymology, Vol. 82. Academic Press, New York, pp. 96–127. 45. Mahonen A, Jukkola A, Risteli L, Risteli J, Maenpaa PH 1998 Type I procollagen synthesis is regulated by steroids and related hormones in human osteosarcoma cells. J Cell Biochem 68: 151–163. 46. Siggelkow H, Schulz H, Kaesler S, Benzler K, Atkinson MJ, Hufner M 1999 1,25-dihydroxyvitamin-D3 attenuates the confluence-dependent differences in the osteoblast characteristic proteins alkaline phosphatase, procollagen I peptide, and osteocalcin. Calcif Tissue Int 64:414–421. 47. Bringhurst FR, Potts Jr. JT 1982 Effects of vitamin D metabolites and analogs on bone collagen synthesis in vitro. Calcif Tissue Int 34:103–110. 48. Raisz LG, Kream BE, Smith MD, Simmons HA 1980 Comparison of the effects of vitamin D metabolites on collagen synthesis and resorption of fetal rat bone in organ culture. Calcif Tissue Int 32:135–138. 49. Rowe DW, Kream BE 1982 Regulation of collagen synthesis in fetal rat cavlaria by 1,25-dihydroxyvitamin D3. J Biol Chem 257:8009–8015. 50. Canalis E, Lian JB 1985 1,25-Dihydroxyvitamin D3 effects on collagen and DNA synthesis in periosteum and periosteumfree calvaria. Bone 6:457–460. 51. Hock JM, Kream BE, Raisz LG 1982 Autoradiographic study of the effect of 1,25-dihydroxyvitamin D3 on bone matrix synthesis in vitamin D replete rats. Calcif Tissue Int 34:347–351. 52. Kream BE, Rowe D, Smith MD, Maher V, Majeska R 1986 Hormonal regulation of collagen synthesis in a clonal rat osteosarcoma cell line. Endocrinology 119:1922–1928. 53. Chen TL, Hauschka PV, Cabrales S, Feldman D 1986 The effects of 1,25-dihydroxyvitamin D3 and dexamethasone on rat osteoblast-like primary cell cultures: Receptor occupancy and functional expression patterns for three different bioresponses. Endocrinology 118:250–259. 54. Kim HT, Chen TL 1989 1,25-Dihydroxyvitamin D3 interaction with dexamethasone and retinoic acid: effects on procollagen messenger ribonucleic acid levels in rat osteoblast-like cells. Mol Endocrinol 3:97–104. 55. Wong GL, Luben RA, Cohn DV 1977 1,25-Dihydroxycholecalciferol and parathormone: effects on isolated osteoclast-like and osteoblast-like cells. Science 197:663–665. 56. Rosen DM, Luben RA 1983 Multiple hormonal mechanisms for the control of collagen synthesis in an osteoblast-like cell line, MMB-1. Endocrinology 112:992–999. 57. Chen TL, Li JM, Ye TV, Cone CM, Feldman D 1986 Hormonal responses to 1,25-dihydroxyvitamin D3 in cultured mouse osteoblast-like cells—modulation by changes in receptor level. J Cell Physiol 126:21–28. 58. Walters MR, Rosen DM, Norman AW, Luben RA 1982 1,25Dihydroxyvitamin D receptors in an established bone cell lines. Correlation with biochemical responses. J Biol Chem 257:7481–7484. 59. Chen TL, Cone CM, Morey-Holton E, Feldman D 1983 1α,25Dihydroxyvitamin D3 receptors in cultured rat osteoblast-like cells. Glucocorticoid treatment increases receptor content. J Biol Chem 258:4350–4355.

BARBARA E. KREAM AND ALEXANDER C. LICHTLER

60. Owen TA, Aronow MS, Barone LM, Bettencourt B, Stein GS, Lian JB 1991 Pleiotropic effects of vitamin D on osteoblast gene expression are related to the proliferative and differentiated state of the bone cell phenotype: dependency upon basal levels of gene expression, duration of exposure, and bone matrix competency in normal rat osteoblast cultures. Endocrinology 128:1496–1504. 61. Ishida H, Bellows CG, Aubin JE, Heersche JNM 1993 Characterization of the 1,25-(OH)2D3-induced inhibition of bone nodule formation in long-term cultures of fetal rat calvaria cells. Endocrinology 132:61–66. 62. Kurihara N, Ikeda K, Hakeda Y, Tsunoi M, Maeda N, Kumegawa M 1984 Effect of 1,25-dihydroxyvitamin D3 on alkaline phosphatase activity and collagen synthesis in osteoblastic cells, clone MC3T3-E1. Biochem Biophys Res Commun 119:767–771. 63. Franceschi RT, Park K-Y, Romano PR 1988 Regulation of type I collagen synthesis by 1,25-dihydroxyvitamin D3 in human osteosarcoma cells. J Biol Chem 263:18938–18945. 64. Beresford JN, Gallagher JA, Russell RGG 1986 1,25Dihydroxyvitamin D3 and human bone-derived cells in vitro: Effects on alkaline phosphatase, type I collagen and proliferation. Endocrinology 119:1776–1785. 65. Kurihara N, Ishizuka S, Kiyoki M, Haketa Y, Ikeda K, Kumegawa M 1986 Effects of 1,25-dihydroxyvitamin D3 on osteoblastic MC3T3-E1 cells. Endocrinology 118:940–947. 66. Matsumoto T, Igarashi C, Takeuchi Y, Harada S, Kikuchi T, Yamato H, Ogata E 1991 Stimulation by 1,25-dihydroxyvitamin D3 of in vitro mineralization induced by osteoblast-like MC3T3-E1 cells. Bone 12:27–32. 67. Wang D, Christensen K, Chawla K, Xiao G, Krebsbach PH, Franceschi RT 1999 Isolation and characterization of MC3T3-E1 preosteoblast subclones with distinct in vitro and in vivo differentiation/mineralization potential. J Bone Miner Res 14:893–903. 68. Broess M, Riva A, Gerstenfeld LC 1995 Inhibitory effects of 1,25-(OH)2 vitamin D3 on collagen type I, osteopontin and osteocalcin gene expression in chicken embryo fibroblasts. J Cell Biochem 57:440–451. 69. Gerstenfeld LC, Zurakowski D, Schaffer JL, Nichols DP, Toma CD, Broess M, Bruder SP, Caplan AI 1996 Variable hormone responsiveness of osteoblast populations isolated at different stages of embryogenesis and its relationship to the osteogenic lineage. Endocrinology 137:3957–3968. 70. Harrison JR, Petersen DN, Lichtler AC, Mador AT, Rowe DW, Kream BE 1989 1,25-Dihydroxyvitamin D3 inhibits transcription of type I collagen genes in the rat osteosarcoma cell line ROS 17/2.8. Endocrinology 125:327–333. 71. Ozono K, Sone T, Pike W 1991 The genomic mechanism of action of 1,25-dihydroxyvitamin D3. J Bone Miner Res 6: 1021–1027. 72. Darwish HM, DeLuca HF 1996 Recent advances in the molecular biology of vitamin D action. Prog Nucleic Acid Res Mol Biol 53:321–344. 73. Kerner SA, Scott RA, Pike JW 1989 Sequence elements in the human osteocalcin gene confer basal activation and inducible response to hormonal vitamin D3. Proc Natl Acad Sci USA 86:4455–4459. 74. Demay MB, Gerardi JM, DeLuca HF, Kronenberg HM 1990 DNA sequences in the rat osteocalcin gene that bind the 1,25-dihydroxyvitamin D3 receptor and confer responsiveness to 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 87:369–373. 75. Noda M, Vogel RL, Craig AM, Prahl J, DeLuca HF, Denhardt DT 1990 Identification of a DNA sequence responsible for binding of the 1,25-dihydroxyvitamin D3 receptor and 1,25-dihydroxyvitamin D3 enhancement of mouse secreted phosphoprotein 1

709

CHAPTER 40 Vitamin D and Collagen Expression

76.

77. 78.

79.

80.

81.

82.

83.

84.

85.

86.

(Spp-1 or osteopontin) gene expression. Proc Natl Acad Sci USA 87:9995–9999. Ohyama Y, Ozono K, Uchida M, Shinki T, Kato S, Suda T, Yamamoto O, Noshiro M, Kato Y 1994 Identification of a vitamin D–responsive element in the 5′-flanking region of the rat 24-hydroxylase gene. J Biol Chem 269:10545–10550. Darwish HM, DeLuca HF 1992 1,25-Dihydroxyvitamin D3– response element in the 5′-flanking region of the rat calbindin D-9k gene. Proc Natl Acad Sci USA 89:603–607. Towers TL, Luisi BF, Asianov A, Freedman LP 1993 DNA target selectivity by the vitamin D3 receptor: Mechanism of dimer binding to an asymmetric repeat element. Proc Natl Acad Sci USA 90:6310–6314. Liu SM, Koszewski N, Lupez M, Malluche HH, Olivera A, Russell J 1996 Characterization of a response element in the 5′-flanking region of the avian (chicken) PTH gene that mediates negative regulation of gene transcription by 1,25-dihydroxyvitamin D3 and binds the vitamin D3 receptor. Mol Endocrinol 10:206–215. Demay MB, Kiernan MS, DeLuca HF, Kronenberg HM 1992 Sequences in the human parathyroid hormone gene that bind the 1,25-dihydroxyvitamin D3 receptor and mediate transcriptional repression in response to 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 89:8097–8101. Mackey SL, Heymont JL, Kronenberg HM, Demay MB 1996 Vitamin D receptor binding to the negative human prathyroid hormone vitamin D response element does not require the retinoid X receptor. Mol Endocrinol 10:298–305. Falzon M 1996 DNA sequences in the rat parathyroid hormonerelated peptide gene responsible for 1,25-dihydroxyvitamin D3mediated transcriptional repression. Mol Endocrinol 10: 672–681. Lichtler A, Stover ML, Angilly J, Kream B, Rowe DW 1989 Isolation and characterization of the rat α1(I) collagen promoter. Regulation by 1,25-dihydroxyvitamin D. J Biol Chem 264:3072–3077. Pavlin D, Lichtler AC, Bedalov A, Kream BE, Harrison JR, Thomas HF, Gronowicz GA, Clark SH, Woody CO, Rowe DW 1992 Differential utilization of regulatory domains within the α1(I) collagen promoter in osseous and fibroblastic cells. J Cell Biol 116:227–236. Pavlin D, Bedalov A, Kronenberg MS, Kream BE, Rowe DW, Smith CL, Pike JW, Lichtler AC 1994 Analysis of regulatory domains in the COL1A1 gene responsible for 1,25-dihydroxyvitamin D3-mediated transcriptional repression in osteoblastic cells. J Cell Biochem 56:490–501. Bedalov A, Salvatori R, Dodig M, Kronenberg MS, Kapural B, Bogdanovic Z, Kream BE, Woody CO, Clark SH, Mack K,

87.

88.

89.

90.

91.

92.

93.

94. 95.

96.

Rowe DW, Lichtler AC 1995 Regulation of COL1A1 expression in type I collagen producing tissues: identification of a 49 base pair region which is required for transgene expression in bone of transgenic mice. J Bone Miner Res 10:1443–1451. Dodig M, Kronenberg MS, Bedalov A, Kream BE, Gronowicz G, Clark S, Mack K, Liu Y, Maxon R, Pan ZZ, Upholt WB, Rowe DW, Lichtler AC 1996 Identification of a TAATcontaining motif required for high level expression of a COL1A1 promoter in differentiated osteoblasts of transgenic mice. J Biol Chem 271:16422–16429. Bedalov A, Salvatori R, Dodig M, Kapural B, Pavlin D, Kream BE, Clark SH, Woody CO, Rowe DW, Lichtler AC 1998 1,25-Dihydroxyvitamin D3 inhibition of Col1a1 promoter expression in calvariae from neonatal transgenic mice. Biochim Biophys Acta 1398:285–293. Bogdanovic Z, Bedalov A, Krebsbach PH, Woody CO, Clark SH, Thomas HF, Rowe DW, Kream BE, Lichtler AC 1994 Upstream regulatory elements necessary for expression of the rat COL1A1 promoter in transgenic mice. J Bone Miner Res 9:285–292. Rossert JA, Chen SS, Eberspaecher H, Smith CN, De Crombrugghe B 1996 Identification of a minimal sequence of the mouse pro-alpha1(I) collagen promoter that confers high-level osteoblast expression in transgenic mice and that binds a protein selectively present in osteoblasts. Proc Natl Acad Sci USA 93:1027–1031. Alroy I, Tower TL, Freedman LP 1995 Transcriptional repression of the interleukin-2 gene by vitamin D3: Direct inhibition of NFATp/AP-1 complex formation by a nuclear hormone receptor. Mol Cell Biol 15:5789–5799. Schule R, Rangarajan P, Kliewer S, Ransone LJ, Bolado J, Yang N, Verma IM, Evans RM 1990 Functional antagonism between oncoprotein c-Jun and the glucocorticoid receptor. Cell 62:1217–1226. Jonat C, Rahmsdorg HJ, Park, K-K, Cato ABC, Bebel S, Ponta H, Herrlich P 1990 Antitumor promotion and antiinflammation: Downregulation of AP-1 (Fos/Jun) activity by glucocorticoid hormone. Cell 62:1189–1204. Feyen JHM, Petersen DN, Kream BE 1988 Inhibition of bone collagen synthesis by the tumor promoter phorbol 12-myristate 13-acetate. J Bone Miner Res 3:173–179. Bogdanovic Z, Huang YF, Dodig M, Clark SH, Lichtler AC, Kream BE 2000 Parathyroid hormone inhibits collagen synthesis and the activity of rat col1a1 transgenes mainly by a cAMP-mediated pathway in mouse calvariae. J Cell Biochem 77:149–158. Franceschi RT, Young J 1990 Regulation of alkaline phosphatase by 1,25-dihydroxyvitamin D3 and ascorbic acid in bone-derived cells. J Bone Miner Res 5:1157–1167.

CHAPTER 41

Target Genes: Bone Proteins GERALD J. ATKINS AND DAVID M. FINDLAY Hanson Institute, Adelaide, South Australia, Australia; Department of Orthopaedic Surgery and Trauma, University of Adelaide, Adelaide, South Australia, Australia

PAUL H. ANDERSON AND HOWARD A. MORRIS Institute of Medical and Veterinary Science, Adelaide, South Australia, Australia Hanson Institute, Adelaide, South Australia, Australia

I. II. III. IV.

Vitamin D and Skeletal Homeostasis Osteoblast Differentiation and Vitamin D Effects of Vitamin D on Gene Expression during Proliferation Effects of Vitamin D on Gene Expression during Matrix Synthesis

I. VITAMIN D AND SKELETAL HOMEOSTASIS The vitamin D endocrine system plays a primary role in the maintenance of calcium homeostasis. The regulation within narrow limits of the calcium concentration in extracellular fluid (ECF) involves a coordinated response by cells in the intestine, kidney, bone and parathyroid glands [1]. The complex interactions of these tissues, in which circulating 1α,25-dihydroxyvitamin D3 (1,25D) plays an integral role, ensure adequate availability of calcium and phosphate for a number of biological functions including nerve and muscle functions. 1,25D is also essential for the maintenance of a healthy skeleton. Vitamin D deficiency in animals and humans produces defects in bone mineralization, such as rickets and osteomalacia, which are characterized by an increase in osteoid (unmineralized bone matrix) and impaired calcium phosphate deposition [2]. Likewise, ablation of the genes CYP27B1 [3] or VDR [4,5] genes encoding the 25-hydroxyvitamin D-1α-hydroxylase and the 1,25D receptor (VDR), respectively, produced similar defects. VDR-null and CYP27B1-null mice fed a normal chow diet developed significant hypocalcemia, hypophosphatemia, and secondary hyperparathyroidism along with the bone defects characteristic of osteomalacia [4–6]. Mice with either of these gene deletions correct their mineral ion disturbances and secondary hyperparathyroidism when fed a “rescue diet” of high calcium, phosphorus, and lactose. The rescue diet then appears to correct their gross bone mineralization defects with an increase in trabecular bone volume and mechanical properties, as well as normalization of VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Effects of Vitamin D on Gene Expression during Mineralization VI. Osteoblasts as a Source of 1,25D? VII. Concluding Remarks References

bone formation rate and osteoclast surface [6,7]. However, the rescue diet does not restore femoral length in CYP27B1-null mice, which is normalized by 1,25D treatment [8]. These data were reproduced with the “double knockout” VDR-null/CYP27B1-null mice, which also demonstrated that the rescue diet could not restore normal levels of trabecular bone volume after 4 months, or wild-type osteoblast number or osteoclast size [9]. Thus, although the data from the CYP27B1, VDR gene ablation studies suggest that 1,25D may not be essential for bone mineralization when high dietary calcium is available [4–6], there are limitations with interpretation of data from these animals. The rescue diet often improves the skeletal status of the wildtype, control mice [6], and it is not clear whether the CYP27B1-null mice respond exactly like the wild-type mice. Further studies are required, in which these animals are subjected to metabolic stress, such as estrogen deficiency, and also more detailed analyses of the mineral phase and extracellular matrix. In particular, levels of vitamin D–responsive proteins, such as osteocalcin, should be measured with the rescue diet model before the physiological role of 1,25D in the regulation of their expression can be defined. Nonetheless, numerous in vitro studies have shown that 1,25D is capable of regulating both osteoblastic and osteoclastic activity [10,11]. A current view is that 1,25D stimulation of osteoclast-mediated bone resorption is a mechanism by which normocalcemia is maintained. It is therefore interesting to note that in vitamin D deficiency, or in VDR-null mice, hypocalcemia develops despite high levels of serum parathyroid hormone (PTH) [4]. This suggests that both PTH Copyright © 2005, Elsevier, Inc. All rights reserved.

712

PAUL H. ANDERSON

and 1,25D are required in a coordinated response in order to stimulate significant bone resorption and to normalize ECF calcium. Furthermore, evidence that 1,25D can also stimulate mineralization by osteoblasts suggests that 1,25D may be important in the synthetic, as well as the catabolic, phases of the bone remodeling process [12]. In this case, 1,25D-induced osteoclastic activity may provide calcium for the coupled bone mineralization process, rather than only to maintain serum calcium [13]. This review will focus on the role of 1,25D in the expression of bone proteins involved in the synthesis of the organic matrix, with respect to its accompanying effects on osteoblast proliferation and differentiation.

II. OSTEOBLAST DIFFERENTIATION AND VITAMIN D

ET AL .

ultimately mineralized at discrete sites by the incorporation of calcium and phosphate, to form a mature bone matrix (for an excellent review of this topic, see [14]). In vitro studies have shown that 1,25D is capable of regulating osteoblast gene transcription, proliferation, differentiation, and mineralization [15–17]. The genes of matrix proteins such as osteopontin and osteocalcin possess vitamin D–responsive elements (VDREs) within their promoter regions, suggesting a direct action for 1,25D on their expression [18]. Other important bone matrix–associated genes, such as type I collagen and osteonectin, may have nonclassical VDREs in their promoters or be indirectly regulated by 1,25D. A summary of bone proteins regulated by 1,25D is shown in Table I, and the nucleotide sequences of identified VDREs are shown in Table II.

A. An Introduction to Bone Matrix Proteins

B. An Overview of Osteoblast Differentiation

The osteoblast regulates bone synthesis and coordinates its resorption during remodelling in response to a large number of regulatory signals, of which 1,25D appears to be an important and pleiotropic member (Fig. 1). A primary function of the osteoblast is to secrete a specialized organic extracellular matrix, which consists mainly of type I collagen but also a number of noncollagenous proteins, such as osteocalcin, osteopontin, osteonectin, bone sialoprotein-1, and proteoglycans such as versican, and the small chondroitin sulfate proteoglycans, decorin and biglycan. This organic matrix is

Osteoblasts originate from mesenchymal stem cells. The differentiation program of the osteoblast has been described for human, rat, mouse, and chicken [19–23]. Osteoblast lineage cells exhibit a temporal pattern of gene expression reflecting three major periods of cell development: proliferation, matrix maturation, and mineralization, the last as the osteoblast matures into an osteocyte within the newly mineralizing matrix. In the proliferative stage, the osteoblast synthesises a type I collagen-rich matrix, termed osteoid, to support extracellular matrix formation. A number of growth

1α,25(OH)2vitamin D3

Other Regulatory Signals: Thyroid hormones Insulin/IGFs PTH Growth hormone Estrogen TGFβ BMPs Corticosteroids Prostanoids

Osteoblast

Osteoid Integrins Mineralization front

Mineralized matrix

FIGURE 1

Enzymes: Alkaline phosphatase MMPs/collagenases Plasminogen activator

Matrix constituents: Type 1 collagen Osteopontin Bone sialoprotein Thrombospondin Osteonectin Osteocalcin Proteoglycans Growth factors

The regulation of osteogenic osteoblast activity. Osteoblasts respond to a large number of regulatory signals, as shown, by secreting bone matrix and other local regulatory proteins.

713

CHAPTER 41 Target Genes: Bone Proteins

TABLE I Effect of 1,25D on Bone Matrix Protein Expression

cells irrespective of their expression of STRO-1, a marker for immature osteoblast lineage cells, or the bone/liver/ kidney isoform of alkaline phosphatase, a marker for mature osteoblasts.

Bone matrix protein

Type of VDRE present in promoter

Action of 1,25D on gene expression

Type I collagen Bone sialoprotein-1 Osteocalcin Osteonectin Osteopontin Alkaline phosphatase Decorin

Nonclassical Classical Classical Unknown Classical Classical Putative

Up-regulates Species-specific effects Up-regulates Inhibits Up-regulates Up-regulates Unknown

C. A Dual Role for 1,25D in Osteoblast Differentiation

regulatory genes (e.g., c-myc, c-fos, and c-jun), cell cycle progression proteins (e.g., histones and cyclins), and adhesion substrates (e.g., fibronectin) are all expressed and are necessary for the proliferation of the osteoblast. The cessation of proliferation occurs when these genes are down-regulated, and expression of the genes required for the maturation of bone extracellular matrix, for example alkaline phosphatase and a range of bone matrix proteins, is up-regulated. The third phase of osteoblastic differentiation is the mineralization of the extracellular matrix, and the genes for osteopontin and osteocalcin are strongly expressed at this time. Although similar to rodent osteoblast differentiation, demonstrable differences exist for human osteoblasts. Siggelkow and co-workers [20] showed that osteocalcin expression in human osteoblast-like cells was not expressed as a function of their differentiation stage. Similarly, Atkins et al. [17] showed that 1,25D induces osteocalcin expression in normal human osteoblast-like

TABLE II Gene controlled Human osteocalcin Rat osteocalcin Mouse osteocalcin Rat osteopontin (proximal) Rat osteopontin (distal) Mouse osteopontin Pig osteopontin Rat bone sialoprotein Rat RunX2/CBFA1 Human RANKL Mouse RANKL

As mentioned, in vitro evidence suggests that 1,25D exerts effects during both the resorptive and synthetic phases of bone remodeling. In association with other factors including PTH, 1,25D can also indirectly induce osteoclastogenesis by stimulating the differentiation of bone marrow-derived promyelocytes and monocytes to active osteoclasts [24,25]. The tumor necrosis factor (TNF) ligand member, RANKL, itself a 1,25D inducible protein, has been shown to be a critical mediator of 1,25D, PTH, or inflammatory cytokine-induced osteoclastogenesis [26]. Thus, a paradox exists in that 1,25D can potentially induce a proresorptive expression pattern, by increasing expression of RANKL, or a pro-osteogenic pattern in osteoblast lineage cells. One possibility is that the support of osteoclastogenesis and osteogenesis are performed by different types of specialized osteoblasts. A second possibility is that these two diverse osteoblast functions are performed at different stages of osteoblast differentiation, as presented in Fig. 2. Our previous study [17] showed that in primary human osteoblasts, 1,25D induced the expression of RANKL, in phenotypically immature osteoblast precursors, identified by their expression of the marker STRO-1 [27]. However, in phenotypically mature cells, negative for STRO-1 expression, an osteocalcin response predominated [17]. This differential response was not related to

Selected Natural Vitamin D Responsive Elements

Type of regulation

5′-half element

“Spacer”

3′-half element

+ve +ve −ve +ve +ve +ve +ve −ve −ve +ve +ve

GGGTGA GGGTGA GGGCAA AGGTCA AGGTCA GGTTCA GGGTCA AGGGTT AGTACT AGGTCA GAGGTCA

ACG ATG ATG CAC TAT CGA TAT TAT GTG AAG CCT

GGGGCA AGGACA AGGACA AGGGCA GGTTCA GGTTCA GGTTCA AGGTCA AGGTCA ACTACA GGTTCA

714

PAUL H. ANDERSON

Osteoclastogenesis

Osteogenesis C 1,25D

STRO-1+ /AP ±

preOC

RANK

ET AL .

OB OC

RANKL

OCy A 1,25D

Osteoid STRO-1+ /AP ± B 1,25D Mineralized bone Time/Differentiation

FIGURE 2 Differential response of osteoblasts to 1,25D as a function of their differentiation stage during

bone remodeling. (A) 1,25D induces RANKL expression in immature (STRO-1+) osteoblastic stromal cells to initiate osteoclast (OC) recruitment and formation, (B) supports activity of mature OCs, and (C) acts on mature (STRO-1−) OB to promote the osteogenic response and possibly differentiation into bone lining cells or osteocytes (OCy). This model is based on previously published data [17].

levels of VDR expression, or to the overall ability of the cells to respond to 1,25D, evidenced by the expression of other “synthetic phase” 1,25D-responsive genes such as type I collagen and bone sialoprotein-1 (see Section IV of this chapter), which were found to be expressed independently of differentiation stage. Similar results were obtained in mineralizing cultures of primary mouse osteoblasts, where the 1,25D induction of RANKL expression decreased with increasing maturation of the osteoblast [28]. A third and combinatorial model is possible, with the sum of signals received by the osteoblast within a permissive window of its differentiation program determining either an osteoclastogenic or osteogenic response. The foregoing studies imply that the particular cohort of genes expressed in response to 1,25D in the osteoblast is regulated according to their stage of differentiation. Differential responses of osteoblasts to 1,25D may result from different VDR signaling complexes. As detailed elsewhere in this volume, upon ligation of 1,25D with VDR and translocation to the nucleus, the complex forms a heterodimer with the retinoid X receptor (RXR). This induces a VDR conformation that is essential for effective binding to the VDRE. This association serves to recruit nuclear proteins as coactivators or co-repressors, necessary for VDR-mediated transcriptional regulation. In short, the interaction of the 1,25Dbound VDR-RXR complex with nuclear proteins forms a so-called “preinitiation complex,” which regulates the

rate of transcription of the target gene [29,30]. The constitution, and therefore the promoter specificity, of this complex may change with differentiation stage of the cell. It remains to be seen whether other mechanisms of modifying the 1,25D response, such as CpG methylation and inactivation of the VDRE, as has been shown for the RANKL promoter [31], occur for other bone protein genes.

III. EFFECTS OF VITAMIN D ON GENE EXPRESSION DURING PROLIFERATION In general, 1,25D inhibits the proliferation of osteoblasts. This antiproliferative activity is associated with the ability of 1,25D to induce osteoblast differentiation [32,33]. Numerous studies have shown the inhibition of osteoblast proliferation in the human [17,34,35], rat [36,37], and mouse [38–40]. The effects of 1,25D on osteoblast proliferation are, however, dependent on species and maturity of the cell. For example, van den Bemd and co-workers [35] found that in human MG-63 cells, 1,25D could suppress proliferation, whereas in rat osteosarcoma (ROS 17/2.8) cells, 1,25D stimulated growth. Murray and co-workers [37] found that 1,25D inhibited proliferation of the rat osteoblast cell line, G2, and stimulated proliferation in rat osteoblast cell-line, C12. The differences in the effects of 1,25D on osteoblast proliferation are not clear but

715

CHAPTER 41 Target Genes: Bone Proteins

may be due to the differentiation state of the osteoblastlike cell line tested. A further complication of these studies is that immortalized cell lines, in general, proliferate in an uncontrolled fashion, making interpretation of the data difficult. Using carboxyfluorescein succinimidyl ester (CFSE), a fluorescent dye that enables the number of cell divisions to be tracked with respect to other fluorescently tagged proteins [41], Atkins et al. [17] showed that while growth was inhibited overall in a heterogeneous population of normal primary human osteoblast-like cells in the presence of 1,25D, immature cells proliferated more than phenotypically mature cells. This implies that the degree of growth inhibition by 1,25D relates to the inherent growth potential of a particular cell type. Moreover, the effects of 1,25D on osteoblast proliferation appear to be dose-dependent. For example, treatment of human osteoblasts with a low dose of 1,25D (5 × 10−12 M) increased proliferation, whereas a pharmacological dose of 1,25D (5 × 10−6 M) showed decreased proliferation [34]. Type I collagen is expressed in the proliferative stage of osteoblast development and is essential for the tensile strength of bone. Stein and co-workers [21] suggested that the inhibition of type I collagen gene expression prevents subsequent extracellular matrix development. The effect of 1,25D on type I collagen expression during osteoblast proliferation, however, is dependent on the species of osteoblast studied. In human MG-63 osteosarcoma cells, 1,25D stimulated the synthesis of type I collagen [35,42]. In rat and chicken, treatment of osteoblasts with 1,25D reduced type I collagen mRNA transcription and protein synthesis [36,43–46]. The effects of 1,25D on type I collagen synthesis in osteoblasts also appear to be conditional on the differentiation state of the cells. In proliferating rat osteoblasts, acute 1,25D treatment inhibited the high levels of type I collagen expression found at this stage of osteoblast development. During mineralization, however, low basal levels of type I collagen mRNA were stimulated by acute 1,25D treatment and were unaltered by chronic 1,25D treatment [36]. In the mouse, while 1,25D treatment was shown to promote type I collagen breakdown in calvarial osteoblasts [44], it has also been shown to stimulate type I collagen synthesis in earlyphase MC3T3-E1 cells and have no effect in late-phase MC3T3-E1 cells [15].

IV. EFFECTS OF VITAMIN D ON GENE EXPRESSION DURING MATRIX SYNTHESIS Inhibition of osteoblast proliferation can signal their differentiation and induce the up-regulation of the

genes that are expressed only during their progression to the mineralization phase. Osteopontin (OPN), an extracellular glycosylated bone phosphoprotein, is one such gene that, in bone, is secreted by matrix-producing osteoblasts at the mineralization front [36,47]. In human bone marrow cultures, and in MG-63 cells, 1,25D administration was associated with increased levels of OPN mRNA [48]. Cultured rat bone cells and ROS17/2.8 cells were both shown to increase OPN mRNA expression and protein secretion in response to 1,25D administration [49,50]. Low basal levels of OPN mRNA were seen in rat calvarial cultures of intermediate maturity, which were markedly up-regulated by 1,25D [45]. However, 1,25D-mediated stimulation of OPN mRNA in rat calvarial osteoblasts was shown to be far greater in premineralization cells than in mature mineralizing cells, where levels of OPN mRNA were already high [36]. It has been found that the helix–loop–helix-type transcription factor (HES-1) is expressed in osteoblastic cells and is suppressed by 1,25D. Overexpression of HES-1 in ROS17/2.8 cells suppressed the vitamin D– dependent up-regulation of osteopontin gene expression in these cells [51]. TGF-β and PTH were also shown to abrogate 1,25D-mediated induction of OPN in ROS17/2.8 cells [52,53], suggesting that multiple transcription factors and hormones may be involved in regulating 1,25D-mediated up-regulation of OPN activity. Bone sialoprotein (BSP) is largely specific for mineralized tissues and is highly expressed during the initial formation of bone and cementum [54]. The expression of BSP is suppressed by 1,25D treatment in rat calvaria and ROS 17/2.8 cells [55]. A VDRE that is integrated with an inverted TATA box in the rat BSP promoter mediates the suppression of BSP transcription [18,56,57]. In human bone marrow stromal cells, 1,25D treatment alone did not significantly affect the expression of BSP mRNA [48]. However, data from our laboratory demonstrates a positive induction of BSP-1 mRNA by 1,25D in normal human osteoblast-like cells (unpublished data), suggesting further differences between human and rodent responses to 1,25D.

V. EFFECTS OF VITAMIN D ON GENE EXPRESSION DURING MINERALIZATION A. Osteocalcin In the mature osteoblast, 1,25D has been found to down-regulate the expression of BSP and type I

716 collagen [55,58] and increase the expression of OPN and osteocalcin (OCN) [59,60]. OCN, also known as bone Gla protein, produced by osteoblasts has a high affinity for calcium ions of hydroxyapatite and is the most abundant noncollagenous protein in bone [61]. OCN is expressed in postproliferative osteoblasts as well as in osteocytes. Ablation of the OCN gene in mice increases both bone formation and bone mass, although the mechanism for this remains unclear [62]. While the precise function of this protein is not well defined, it has been shown to be a chemotactic factor for osteoclasts and their precursors [63]. OCN has been widely used as a marker of bone formation [64]. It has recently been reported to be released from the matrix in the intact form rather than as fragments during remodeling as a result of osteoclast activity [65]. A number of studies have reported the induction of both OCN mRNA and protein synthesis by 1,25D in human and rat bone cells [34,36,45,66–72], although the pattern of 1,25D-induced expression of the OCN gene seems to depend on the culture system and the stage of maturity of the cells. For example, in human MG-63 cells, 1,25D induction of OCN was highest in subconfluent cultures and decreased in confluent cultures [73]. Similarly, OCN gene expression was found to have a decreased responsiveness to 1,25D in mineralizing human osteoblasts, which was suggested to be due to an accumulation of OCN in the extracellular matrix [74]. There are, however, reports of 1,25D down-regulating OCN expression both in mouse osteoblast cultures [75,76] and in chicken embryonic osteoblasts [46]. The characterization of the OCN gene has identified a number of factors that are potentially involved in the development of the osteoblast phenotype. Besides the identification of a VDRE in the distal promoter of the OCN gene, several other promoter sites have been shown to be crucial in the expression of the OCN and in osteoblastic differentiation. For example, the requirement for the osteoblast transcription factor, core binding factor alpha (CBFA)-1/AML-3, is best demonstrated in the CBFA-1/AML-3-null mutant mouse. These mice died at birth and had major skeletal deformations characterized by disrupted mineralization of osteoblasts [77,78]. Mutations of the three CBFA-1 motifs identified on the osteocalcin promoter were found to lead to abrogation of responsiveness to 1,25D [79]. Another important site identified in the promoter of the OCN gene is the AP-1 site juxtaposed to the VDRE of the OCN gene. While this AP-1 site is essential for 1,25D induction of the OCN gene [80], the suppression of OCN gene expression at the onset of mineralization appears to be related to the interaction

PAUL H. ANDERSON

ET AL .

of specific transcription factors at this AP-1 site. In proliferating osteoblasts, the repression of OCN gene expression is partly due to the expression of c-fos/ c-jun, which binds as a heterodimer to the AP-1 site and blocks the binding of the VDR/RXR complex [19,81]. In contrast, in postproliferative osteoblasts approaching mineralization, the expression of fra-2 results in binding to the AP-1 site, which facilitates VDR/RXR binding to the VDRE and 1,25D-mediated expression of the OCN gene [80,82].

B. Alkaline Phosphatase Alkaline phosphatase activity is important for the mineralization of bone and represents a useful biochemical marker of bone formation [64,83]. Osteoblasts express the bone or tissue nonspecific isoform of alkaline phosphatase (TNAP), which is a glycosylphosphatidylinositol (GPI) anchored cell surface protein [84]. Treatment of rat osteoblast-like cells with 1,25D promoted mineralization, which was associated with high alkaline phosphatase activity [85–87]. While the alkaline phosphatase gene promoter has no classical VDRE, 1,25D was also shown to have a stimulatory effect on alkaline phosphatase mRNA levels, protein synthesis, and activity in human osteoblasts [70,72]. The stage of differentiation of osteoblasts has been shown to determine the response of alkaline phosphatase expression to 1,25D. During the proliferative period of rat osteoblast development, 1,25D inhibited the expression of alkaline phosphatase, whereas during mineralization, 1,25D stimulated alkaline phosphatase mRNA expression [36]. In the mouse, however, 1,25D stimulated alkaline phosphatase activity only in the early phase of osteoblast differentiation and not in the mineralization phase [15].

C. Matrix Gla Protein Matrix Gla protein (MGP), like osteocalcin, requires vitamin K–dependent gamma-carboxylation for its function. MGP has been identified as a calcification inhibitor in cartilage and vasculature since MGP-null mice die soon after birth due to aberrant cartilage and arterial calcification [88]. In both rat UMR 106-01 and ROS17/2.8 cells, 1,25D treatment markedly increased MGP mRNA and protein levels [89,90]. The stimulation of MGP mRNA by 1,25D was shown to be less in the late stages of rat osteoblast differentiation [36] than in earlier stages of osteoblast growth [91].

717

CHAPTER 41 Target Genes: Bone Proteins

VI. OSTEOBLASTS AS A SOURCE OF 1,25D? A question that arises from studies of 1,25D in bone is the source of the hormone in vivo. It is now clear that although 1,25D is mostly expressed in the nephron, there are numerous sites of extrarenal 1,25D synthesis, including the skin, activated monocytes/macrophages, and lymph nodes, among other tissues [92]. It has also been known for more than 20 years that bone is a source of 1,25D [93]. We have demonstrated that normal human osteoblast-like cells abundantly express mRNA encoding the enzyme responsible for its synthesis, CYP27B1 [94]. In addition, we have found CYP27B1 expression in a number of well-characterized human osteosarcoma cell lines, for example MG-63, SAOS-2 and G-292 (authors’ unpublished data). Furthermore, investigation of bone metabolism of 1,25D in an in vivo rat model has found that the level of bone CYP24 mRNA (corresponding to the enzyme that degrades 1,25D) is always directly proportional to the level of bone CYP27B1 mRNA and is not related to circulating 1,25D levels (Anderson and Morris, unpublished data). This implies that, in vivo, local bone production of 1,25D is responsible for regulating osteoblast levels of CYP24 and therefore potentially other 1,25D-responsive genes discussed in this review. These data suggest that 1,25D metabolism may play a more complex and intrinsic role in human osteoblast biology than has been generally recognized. Further characterization of vitamin D metabolism in osteoblasts and how it relates to the known effects of 1,25D on parameters of osteoblast and osteoclast biology, such as those discussed in this review, is an intriguing and exciting area of current research.

VII. CONCLUDING REMARKS Overwhelming clinical evidence suggests that vitamin D is important for calcium/phosphate and skeletal homeostasis. Numerous direct and indirect effects of 1,25D have been demonstrated on a range of critical bone proteins, as discussed in this review, and 1,25D appears to be involved in their regulation at all stages of osteoblast differentiation and indeed, bone remodeling. Future studies will be needed to unravel the complexities surrounding the involvement of 1,25D in the coordination of these processes. However, the evidence for a fundamental and nonredundant role for 1,25D in skeletal biology in vivo, namely, gene-ablated mouse models, is currently lacking. As suggested in Section I, this is in part due to an incomplete analysis of these potentially informative models, rather than a convincing

lack of a biological effect. Additionally, some of the apparently contradictory data surrounding this question arise from the plethora of model systems (immortalized cell lines versus primary cells, the use of different mammalian and other vertebrate species) that have been utilized to investigate this issue, many of which may have significant and confounding limitations. Studies using 1,25D analogs that show potent anabolic effects, for example, the compound 2-methylene-19-nor-(20S)1α,25-dihydroxyvitamin D3 (2-MD) when administered to sham-operated or ovariectomized mice under otherwise normal dietary conditions [95], and other examples discussed elsewhere in this volume, must provide additional evidence for the bone-specific effects and importance of the natural hormone.

References 1. Fukugawa M, Kurokawa K 2002 Calcium homeostasis and imbalance. Nephron 92:41–45. 2. Parfitt AM, Han ZH, Palnitkar S, Rao DS, Shih MS, Nelson D 1997 Effects of ethnicity and age or menopause on osteoblast function, bone mineralization, and osteoid accumulation in iliac bone. J Bone Miner Res 12:1864–1873. 3. Dardenne O, Prud’homme J, Arabian A, Glorieux FH, St-Arnaud R 2001 Targeted inactivation of the 25-hydroxyvitamin D3-1α-hydroxylase gene (CYP27B1) creates an animal model of pseudovitamin D–deficiency rickets. Endocrinology 142:3135–3141. 4. Li YC, Amling M, Pirro AE, Priemel M, Meuse J, Baron R, Delling G, Demay MB 1998 Normalization of mineral ion homeostasis by dietary means prevents hyperparathyroidism, rickets, and osteomalacia, but not alopecia in vitamin D receptorablated mice. Endocrinology 139:4391–4396. 5. Erben RG, Soegiarto DW, Weber K, Zeitz U, Lieberherr M, Gniadecki R, Moller G, Adamski J, Balling R 2002 Deletion of deoxyribonucleic acid binding domain of the vitamin D receptor abrogates genomic and nongenomic functions of vitamin D. Mol Endocrinol 16:1524–1537. 6. Dardenne O, Prud’homme J, Hacking SA, Glorieux FH, St-Arnaud R 2003 Correction of the abnormal mineral ion homeostasis with a high-calcium, high-phosphorus, highlactose diet rescues the PDDR phenotype of mice deficient for the 25-hydroxyvitamin D-1α-hydroxylase (CYP27B1). Bone 32:332–340. 7. Amling M, Priemel M, Holzmann T, Chapin K, Rueger JM, Baron R, Demay MB 1999 Rescue of the skeletal phenotype of vitamin D receptor-ablated mice in the setting of normal mineral ion homeostasis: Formal histomorphometric and biomechanical analyses. Endocrinology 140:4982–4987. 8. Dardenne O, Prud’homme J, Hacking SA, Glorieux FH St-Arnaud R 2003 Rescue of the pseudo-vitamin D deficiency rickets phenotype of CYP27B1-deficient mice by treatment with 1,25-dihydroxyvitamin D3: Biochemical, histomorphometric, and biomechanical analyses. J Bone Miner Res 18:637–643. 9. Panda DK, Miao D, Bolivar I, Li J, Huo R, Hendy GN, Goltzman D 2004 Inactivation of the 25-hydroxyvitamin D1α-hydroxylase and vitamin D receptor demonstrates

718

10. 11. 12. 13. 14.

15.

16.

17.

18. 19.

20.

21. 22. 23.

24.

25. 26.

independent and interdependent effects of calcium and vitamin D on skeletal and mineral homeostasis. J Biol Chem 279: 16754–16766. van Leeuwen JP, van Driel M, van den Bemd GJ, Pols HA 2001 Vitamin D control of osteoblast function and bone extracellular matrix mineralization. Crit Rev Eukaryot Gene Expr 11:199–226. Suda T, Ueno Y, Fujii K, Shinki T 2003 Vitamin D and bone. J Cell Biochem 88:259–266. Langdahl BL, Mortensen L, Vesterby A, Eriksen EF, Charles P 1996 Bone histomorphometry in hypoparathyroid patients treated with vitamin D. Bone 18:103–108. Eriksen EF, Steiniche T, Mosekilde L, Melsen F 1989 Histomorphometric analysis of bone in metabolic bone disease. Endocrinol Metab Clin North Am 18:919–954. Robey GP, Boskey AL Extracellular matrix and biomineralization of bone. In: Favus MJ (ed) Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism, 5th ed. American Society for Bone and Mineral Research, Washington, DC, pp. 38–46. Matsumoto T, Igarashi C, Takeuchi Y, Harada S, Kikuchi T, Yamato H, Ogata E 1991 Stimulation by 1,25-dihydroxyvitamin D3 of in vitro mineralization induced by osteoblastlike MC3T3-E1 cells. Bone 12:27–32. Rickard DJ, Kazhdan I, Leboy PS 1995 Importance of 1,25dihydroxyvitamin D3 and the nonadherent cells of marrow for osteoblast differentiation from rat marrow stromal cells. Bone 16:671–678. Atkins GJ, Kostakis P, Pan B, Farrugia A, Gronthos S, Evdokiou A, Harrison K, Findlay DM, Zannettino AC 2003 RANKL expression is related to the differentiation state of human osteoblasts. J Bone Miner Res 18:1088–1098. Sodek J, Li JJ, Kim RH, Ogata Y, Yamauchi M 1996 Characterization of the bone sialoprotein (BSP) gene promoter. Connect Tissue Res 35:23–31. Owen TA, Bortell R, Yocum SA, Smock SL, Zhang M, Abate C, Shalhoub V, Aronin N, Wright KL, van Wijnen AJ, et al. 1990 Coordinate occupancy of AP-1 sites in the vitamin D-responsive and CCAAT box elements by Fos-Jun in the osteocalcin gene: Model for phenotype suppression of transcription. Proc Natl Acad Sci USA 87:9990–9994. Siggelkow H, Rebenstorff K, Kurre W, Niedhart C, Engel I, Schulz H, Atkinson MJ, Hufner M 1999 Development of the osteoblast phenotype in primary human osteoblasts in culture: Comparison with rat calvarial cells in osteoblast differentiation. J Cell Biochem 75:22–35. Stein GS, Lian JB, Stein JL, Van Wijnen AJ, Montecino M 1996 Transcriptional control of osteoblast growth and differentiation. Physiol Rev 76:593–629. Gerstenfeld LC, Chipman SD, Glowacki J, Lian JB 1987 Expression of differentiated function by mineralizing cultures of chicken osteoblasts. Dev Biol 122:49–60. Choi JY, Lee BH, Song KB, Park RW, Kim IS, Sohn KY, Jo JS, Ryoo HM 1996 Expression patterns of bone-related proteins during osteoblastic differentiation in MC3T3-E1 cells. J Cell Biochem 61:609–618. Abe E, Miyaura C, Sakagami H, Takeda M, Konno K, Yamazaki T, Yoshiki S, Suda T 1981 Differentiation of mouse myeloid leukemia cells induced by 1α,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 78:4990–4994. Tanaka H, Abe E, Miyaura C, Kuribayashi T, Konno K, Nishii Y, Suda T 1982 1α,25-Dihydroxycholecalciferol and a human myeloid leukaemia cell line (HL-60). Biochem J 204:713–719. Hofbauer LC, Khosla S, Dunstan CR, Lacey DL, Boyle WJ, Riggs BL 2000 The roles of osteoprotegerin and osteoprotegerin

PAUL H. ANDERSON

27. 28. 29. 30. 31. 32. 33.

34.

35.

36.

37.

38. 39.

40. 41. 42. 43. 44.

ET AL .

ligand in the paracrine regulation of bone resorption. J Bone Miner Res 15:2–12. Simmons PJ, Torok-Storb B 1991 Identification of stromal cell precursors in human bone marrow by a novel monoclonal antibody, STRO-1. Blood 78:55–62. Thomas GP, Baker SU, Eisman JA, Gardiner EM 2001 Changing RANKL/OPG mRNA expression in differentiating murine primary osteoblasts. J Endocrinol 170:451–460. Chakravarti D, LaMorte VJ, Nelson MC, Nakajima T, Schulman IG, Juguilon H, Montminy M, Evans RM 1996 Role of CBP/P300 in nuclear receptor signalling. Nature 383:99–103. Glass CK, Rose DW, Rosenfeld MG 1997 Nuclear receptor coactivators. Curr Opin Cell Biol 9:222–232. Kitazawa R, Kitazawa S, Maeda S 1999 Promoter structure of mouse RANKL/TRANCE/OPGL/ODF gene. Biochim Biophys Acta 1445:134–141. Stern PH 1990 Vitamin D and bone. Kidney Int Suppl 29:S17–21. Gurlek A, Kumar R 2001 Regulation of osteoblast growth by interactions between transforming growth factor-beta and 1α,25-dihydroxyvitamin D3. Crit Rev Eukaryot Gene Expr 11:299–317. Skjodt H, Gallagher JA, Beresford JN, Couch M, Poser JW, Russell RG 1985 Vitamin D metabolites regulate osteocalcin synthesis and proliferation of human bone cells in vitro. J Endocrinol 105:391–396. van den Bemd GJ, Pols HA, Birkenhager JC, Kleinekoort WM, van Leeuwen JP 1995 Differential effects of 1,25-dihydroxyvitamin D3-analogs on osteoblast-like cells and on in vitro bone resorption. J Steroid Biochem Mol Biol 55:337–346. Owen TA, Aronow MS, Barone LM, Bettencourt B, Stein GS, Lian JB 1991 Pleiotropic effects of vitamin D on osteoblast gene expression are related to the proliferative and differentiated state of the bone cell phenotype: Dependency upon basal levels of gene expression, duration of exposure, and bone matrix competency in normal rat osteoblast cultures. Endocrinology 128:1496–1504. Murray S, Glackin C, Murray E 1993 Variation in 1,25-dihydroxyvitamin D3 regulation of proliferation and alkaline phosphatase activity in late-passage rat osteoblastic cell lines. J Steroid Biochem Mol Biol 46:227–233. Chen TL, Cone CM, Feldman D 1983 Effects of 1α,25-dihydroxyvitamin D3 and glucocorticoids on the growth of rat and mouse osteoblast-like bone cells. Calcif Tissue Int 35:806–811. Kanatani M, Sugimoto T, Fukase M, Chihara K 1993 Effect of 1,25-dihydroxyvitamin D3 on the proliferation of osteoblastic MC3T3-E1 cells by modulating the release of local regulators from monocytes. Biochem Biophys Res Commun 190:529–535. Rubin J, Fan X, Thornton D, Bryant R, Biskobing D 1996 Regulation of murine osteoblast macrophage colony-stimulating factor production by 1,25(OH)2D3. Calcif Tissue Int 59:291–296. Lyons AB, Parish CR 1994 Determination of lymphocyte division by flow cytometry. J Immunol Methods 171:131–137. Franceschi RT, Romano PR, Park KY 1988 Regulation of type I collagen synthesis by 1,25-dihydroxyvitamin D3 in human osteosarcoma cells. J Biol Chem 263:18938–18945. Rowe DW, Kream BE 1982 Regulation of collagen synthesis in fetal rat calvaria by 1,25-dihydroxyvitamin D3. J Biol Chem 257:8009–8015. Thomson BM, Atkinson SJ, Reynolds JJ, Meikle MC 1987 Degradation of type I collagen films by mouse osteoblasts is stimulated by 1,25-dihydroxyvitamin D3 and inhibited by human recombinant TIMP (tissue inhibitor of metalloproteinases). Biochem Biophys Res Commun 148:596–602.

CHAPTER 41 Target Genes: Bone Proteins

45. Bellows CG, Reimers SM, Heersche JN 1999 Expression of mRNAs for type-I collagen, bone sialoprotein, osteocalcin, and osteopontin at different stages of osteoblastic differentiation and their regulation by 1,25-dihydroxyvitamin D3. Cell Tissue Res 297:249–259. 46. Broess M, Riva A, Gerstenfeld LC 1995 Inhibitory effects of 1,25(OH)2 vitamin D3 on collagen type I, osteopontin, and osteocalcin gene expression in chicken osteoblasts. J Cell Biochem 57:440–451. 47. Yoon K, Buenaga R, Rodan GA, Prince CW, Butler WT 1987 Tissue specificity and developmental expression of rat osteopontin. Biochem Biophys Res Commun 148: 1129–1136. 48. Beresford JN, Joyner CJ, Devlin C, Triffitt JT 1994 The effects of dexamethasone and 1,25-dihydroxyvitamin D3 on osteogenic differentiation of human marrow stromal cells in vitro. Arch Oral Biol 39:941–947. 49. Prince CW, Butler WT 1987 1,25-Dihydroxyvitamin D3 regulates the biosynthesis of osteopontin, a bone-derived cell attachment protein, in clonal osteoblast-like osteosarcoma cells. Coll Relat Res 7:305–313. 50. Chen J, Thomas HF, Sodek J, Kim RH, Ogata Y 1996 Regulation of bone sialoprotein and osteopontin mRNA expression by dexamethasone and 1,25-dihydroxyvitamin D3 in rat bone organ cultures. Connect Tissue Res 34:41–51. 51. Matsue M, Kageyama R, Denhardt DT, Noda M 1997 Helix– loop–helix-type transcription factor (HES-1) is expressed in osteoblastic cells, suppressed by 1,25(OH)2 vitamin D3, and modulates 1,25(OH)2 vitamin D3 enhancement of osteopontin gene expression. Bone 20:329–334. 52. Noda M, Rodan GA 1989 Transcriptional regulation of osteopontin production in rat osteoblast-like cells by parathyroid hormone. J Cell Biol 108:713–718. 53. Staal A, Van Wijnen AJ, Desai RK, Pols HA, Birkenhager JC, Deluca HF, Denhardt DT, Stein JL, Van Leeuwen JP, Stein GS, Lian JB 1996 Antagonistic effects of transforming growth factor-beta on vitamin D3 enhancement of osteocalcin and osteopontin transcription: reduced interactions of vitamin D receptor/retinoid X receptor complexes with vitamin E response elements. Endocrinology 137:2001–2011. 54. Ganss B, Kim RH, Sodek J 1999 Bone sialoprotein. Crit Rev Oral Biol Med 10:79–98. 55. Oldberg A, Jirskog-Hed B, Axelsson S, Heinegard D 1989 Regulation of bone sialoprotein mRNA by steroid hormones. J Cell Biol 109:3183–3186. 56. Li JJ, Sodek J 1993 Cloning and characterization of the rat bone sialoprotein gene promoter. Biochem J 289:625–629. 57. Kim RH, Li JJ, Ogata Y, Yamauchi M, Freedman LP 1996 Identification of a vitamin D3-response element that overlaps a unique inverted TATA box in the rat bone sialoprotein gene. Biochem J 318:219–226. 58. Sodek J, Kim RH, Ogata Y, Li J, Yamauchi M, Zhang Q, Freedman LP 1995 Regulation of bone sialoprotein gene transcription by steroid hormones. Connect Tissue Res 32: 209–217. 59. Chang PL, Prince CW 1991 1α,25-Dihydroxyvitamin D3 stimulates synthesis and secretion of nonphosphorylated osteopontin (secreted phosphoprotein 1) in mouse JB6 epidermal cells. Cancer Res 51:2144–2150. 60. Pockwinse SM, Stein JL, Lian JB, Stein GS 1995 Developmental stage-specific cellular responses to vitamin D and glucocorticoids during differentiation of the osteoblast phenotype: Interrelationship of morphology and gene expression by in situ hybridization. Exp Cell Res 216:244–260.

719 61. Price PA, Williamson MK, Lothringer JW 1981 Origin of the vitamin K-dependent bone protein found in plasma and its clearance by kidney and bone. J Biol Chem 256:12760–12766. 62. Ducy P, Desbois C, Boyce B, Pinero G, Story B, Dunstan C, Smith E, Bonadio J, Goldstein S, Gundberg C, Bradley A, Karsenty G 1996 Increased bone formation in osteocalcindeficient mice. Nature 382:448–452. 63. Chenu C, Colucci S, Grano M, Zigrino P, Barattolo R, Zambonin G, Baldini N, Vergnaud P, Delmas PD, Zallone AZ 1994 Osteocalcin induces chemotaxis, secretion of matrix proteins, and calcium-mediated intracellular signaling in human osteoclast-like cells. J Cell Biol 127:1149–1158. 64. Watts NB 1999 Clinical utility of biochemical markers of bone remodeling. Clin Chem 45:1359–1368. 65. Ivaska KK, Hentunen TA, Vaaraniemi J, Ylipahkala H, Pettersson K, Vaananen HK 2004 Release of intact and fragmented osteocalcin molecules from bone matrix during bone resorption in vitro. J Biol Chem 279:18361–18369. 66. Beresford JN, Gallagher JA, Poser JW, Russell RG 1984 Production of osteocalcin by human bone cells in vitro. Effects of 1,25(OH)2D3, 24,25(OH)2D3, parathyroid hormone, and glucocorticoids. Metab Bone Dis Relat Res 5:229–234. 67. Lian J, Stewart C, Puchacz E, Mackowiak S, Shalhoub V, Collart D, Zambetti G, Stein G 1989 Structure of the rat osteocalcin gene and regulation of vitamin D-dependent expression. Proc Natl Acad Sci USA 86:1143–1147. 68. Lajeunesse D, Kiebzak GM, Frondoza C, Sacktor B 1991 Regulation of osteocalcin secretion by human primary bone cells and by the human osteosarcoma cell line MG-63. Bone Miner 14:237–250. 69. Ingram RT, Bonde SK, Riggs BL, Fitzpatrick LA 1994 Effects of transforming growth factor beta (TGF beta) and 1,25-dihydroxyvitamin D3 on the function, cytochemistry and morphology of normal human osteoblast-like cells. Differentiation 55:153–163. 70. Bodine PV, Trailsmith M, Komm BS 1996 Development and characterization of a conditionally transformed adult human osteoblastic cell line. J Bone Miner Res 11:806–819. 71. Mosavin R, Mellon WS 1996 Posttranscriptional regulation of osteocalcin mRNA in clonal osteoblast cells by 1,25-dihydroxyvitamin D3. 574.1905 A67 Arch Biochem Biophys 332:142–152. 72. Hicok KC, Thomas T, Gori F, Rickard DJ, Spelsberg TC, Riggs BL 1998 Development and characterization of conditionally immortalized osteoblast precursor cell lines from human bone marrow stroma. J Bone Miner Res 13:205–217. 73. Lajeunesse D, Frondoza C, Schoffield B, Sacktor B 1990 Osteocalcin secretion by the human osteosarcoma cell line MG-63. J Bone Miner Res 5:915–922. 74. Hosoda K, Kanzaki S, Eguchi H, Kiyoki M, Yamaji T, Koshihara Y, Shiraki M, Seino Y 1993 Secretion of osteocalcin and its propeptide from human osteoblastic cells: Dissociation of the secretory patterns of osteocalcin and its propeptide. J Bone Miner Res 8:553–565. 75. Zhang R, Ducy P, Karsenty G 1997 1,25-Dihydroxyvitamin D3 inhibits osteocalcin expression in mouse through an indirect mechanism. J Biol Chem 272:110–116. 76. Lian JB, Shalhoub V, Aslam F, Frenkel B, Green J, Hamrah M, Stein GS, Stein JL 1997 Species-specific glucocorticoid and 1,25-dihydroxyvitamin D responsiveness in mouse MC3T3-E1 osteoblasts: Dexamethasone inhibits osteoblast differentiation and vitamin D down-regulates osteocalcin gene expression. Endocrinology 138:2117–2127. 77. Komori T, Yagi H, Nomura S, Yamaguchi A, Sasaki K, Deguchi K, Shimizu Y, Bronson RT, Gao YH, Inada M, Sato M,

720

78.

79.

80.

81. 82.

83. 84.

85.

Okamoto R, Kitamura Y, Yoshiki S, Kishimoto T 1997 Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89:755–764. Otto F, Thornell AP, Crompton T, Denzel A, Gilmour KC, Rosewell IR, Stamp GW, Beddington RS, Mundlos S, Olsen BR, Selby PB, Owen MJ 1997 Cbfa1, a candidate gene for cleidocranial dysplasia syndrome, is essential for osteoblast differentiation and bone development. Cell 89:765–771. Javed A, Gutierrez S, Montecino M, van Wijnen AJ, Stein JL, Stein GS, Lian JB 1999 Multiple Cbfa/AML sites in the rat osteocalcin promoter are required for basal and vitamin D– responsive transcription and contribute to chromatin organization. Mol Cell Biol 19:7491–7500. Aslam F, McCabe L, Frenkel B, van Wijnen AJ, Stein GS, Lian JB, Stein JL 1999 AP-1 and vitamin D receptor (VDR) signaling pathways converge at the rat osteocalcin VDR element: Requirement for the internal activating protein-1 site for vitamin D–mediated trans-activation. Endocrinology 140: 63–70. Grigoriadis AE, Schellander K, Wang ZQ, Wagner EF 1993 Osteoblasts are target cells for transformation in c-fos transgenic mice. J Cell Biol 122:685–701. McCabe LR, Banerjee C, Kundu R, Harrison RJ, Dobner PR 1996 Developmental expression and activities of specific fos and jun proteins are functionally related to osteoblast maturation: Role of Fra-2 and Jun D during differentiation. Endocrinology 137:4398–4408. Narisawa S, Frohlander N, Millan JL 1997 Inactivation of two mouse alkaline phosphatase genes and establishment of a model of infantile hypophosphatasia. Dev Dyn 208:432–446. Noda M, Yoon K, Rodan GA, Koppel DE 1987 High lateral mobility of endogenous and transfected alkaline phosphatase: A phosphatidylinositol-anchored membrane protein. J Cell Biol 105:1671–1677. Manolagas SC, Burton DW, Deftos LJ 1981 1,25Dihydroxyvitamin D3 stimulates the alkaline phosphatase activity of osteoblast-like cells. J Biol Chem 256:7115–7117.

PAUL H. ANDERSON

ET AL .

86. Imai K, Neuman MW, Kawase T, Saito S 1992 Calcium in osteoblast-enriched bone cells. Bone 13:217–223. 87. Halstead LR, Scott MJ, Rifas L, Avioli LV 1992 Characterization of osteoblast-like cells from normal adult rat femoral trabecular bone. Calcif Tissue Int 50:93–95. 88. El-Maadawy S, Kaartinen MT, Schinke T, Murshed M, Karsenty G, McKee MD 2003 Cartilage formation and calcification in arteries of mice lacking matrix Gla protein. Connect Tissue Res 44:272–278. 89. Fraser JD, Otawara Y, Price PA 1988 1,25-Dihydroxyvitamin D3 stimulates the synthesis of matrix gamma-carboxyglutamic acid protein by osteosarcoma cells. Mutually exclusive expression of vitamin K–dependent bone proteins by clonal osteoblastic cell lines. J Biol Chem 263:911–916. 90. Fraser JD, Price PA 1990 Induction of matrix Gla protein synthesis during prolonged 1,25-dihydroxyvitamin D3 treatment of osteosarcoma cells. Calcif Tissue Int 46:270–279. 91. Barone LM, Owen TA, Tassinari MS, Bortell R, Stein GS, Lian JB 1991 Developmental expression and hormonal regulation of the rat matrix Gla protein (MGP) gene in chondrogenesis and osteogenesis. J Cell Biochem 46:351–365. 92. Hewison M, Zehnder D, Bland R, Stewart PM 2000 1α-Hydroxylase and the action of vitamin D. J Mol Endocrinol 25:141–148. 93. Turner RT, Howard GA, Puzas JE, Baylink DJ, Knapp DR 1983 Calvarial cells synthesize 1α,25-dihydroxyvitamin D3 from 25-hydroxyvitamin D3. Biochemistry 22:1073–1076. 94. Atkins GJ, Anderson PH, Morris HA, Zannettino ACW, Kostakis P, Findlay DM 2003 The expression of 25-hydroxyvitamin D 24-hydroxylase and 25-hydroxyvitamin D 1αhydroxylase in human osteoblasts. J Bone Miner Res 18(Suppl. 2):M480. 95. Ke HZ, Qi H, Crawford DT, Xu G, Li M, Plum L, Clagett-Dame M, DeLuca HF, Thompson DD, Brown TA 2003 A novel vitamin D analogue restores bone mass and adds extra bone by markedly stimulating bone formation in ovariectomized rats with established osteopenia. J Bone Miner Res 18(Suppl. 2):1153.

CHAPTER 42

The Calbindins: Calbindin-D9K and Calbindin-D28K SYLVIA CHRISTAKOS, YAN LIU, PUNEET DHAWAN, AND XIAORONG PENG Department of Biochemistry and Molecular Biology, University of Medicine and Dentistry of New Jersey, New Jersey Medical School, Newark, New Jersey

I. Introduction and General Considerations II. Localization and Proposed Functional Significance

I. INTRODUCTION AND GENERAL CONSIDERATIONS In the two major target tissues of 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] action, intestine and kidney, one of the most pronounced effects of 1,25(OH)2D3 known is the induction of the calcium binding protein, calbindin, the first identified target of 1,25(OH)2D3 action. There are two major subclasses of calbindin: a protein of approximately 28,000 molecular weight (calbindin-D28K) and a protein of approximately 9,000 molecular weight (calbindin-D9K). Calbindin-D28K is present in highest concentration in avian intestine and in avian and mammalian kidney, brain, and pancreas. Calbindin-D28K has four functional high-affinity calcium binding sites and is highly conserved in evolution. Calbindin-D9K has two calcium binding domains, is present in highest concentration in mammalian intestine, and, unlike calbindin-D28K, is not evolutionarily conserved and has been observed only in mammals. There is no amino acid sequence similarity between calbindin-D9K and calbindin-D28K. The discussion that follows reviews the chemistry, localization, proposed functional significance, and regulation of these calcium binding proteins. In addition, this chapter provides insight into the information obtained by studying these proteins concerning the multiple actions of the vitamin D endocrine system and the basic molecular mechanism of 1,25(OH)2D3 action. Findings indicating that calbindins can be regulated by a number of different hormones and factors are also reviewed. The study of the molecular interactions of several members of the steroid hormone–retinoic acid family as well as the role of signal transduction pathways in the regulation of calbindin-D may be applicable to the regulation of other targets of 1,25(OH)2D3 action. Elucidation of multiple factors and interactions regulating VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

III. Regulation of Calbindin Gene Expression IV. Conclusion References

1,25(OH)2D3 target genes should result in novel insights related to tissue-specific molecular mechanisms involved in calcium homeostasis. One of the most important findings in the vitamin D field has been the discovery by Wasserman and Taylor in 1966 of a 28,000 Mr vitamin D–dependent calcium binding protein in avian intestine [1]. Although previously known as the vitamin D–dependent calcium binding protein (CaBP), in 1985 it became officially known as calbindin-D28K and calbindin-D9K for the 28,000 Mr and the 9000 Mr proteins, respectively [2]. Initially identified in avian intestine [1], calbindin-D28K has since been reported in many other tissues including kidney and bone, and in tissues that are not primary regulators of serum calcium such as pancreas, testes, and brain and in a variety of species [3–9] (see Christakos et al. [9] for review). The importance of the discovery of calbindin-D28K is that key advances in our understanding of the diversity of the vitamin D endocrine system have been made through the study of its tissue distribution and its colocalization with the vitamin D receptor (VDR). In addition, the biosynthesis of calbindin has provided a model for studies that have resulted in an important basic understanding of the molecular mechanism of action of 1,25(OH)2D3 in major target tissues such as intestine and kidney. Chicken and mammalian calbindin-D28K proteins contain 261 amino acid residues, have a molecular weight of approximately 28,000 (30,000 based on amino acid sequence and 28,000 based on migration on sodium dodecyl sulfate–polyacrylamide gels), and are blocked at the amino terminus [9–12]. The mammalian calbindinD28K sequences are 98% similar to one another and 79% similar to chicken calbindin-D28K [9,11,12]. CalbindinD28K is highly conserved in evolution, suggesting an important, fundamental role for calbindin-D28K in mediating intracellular calcium-dependent processes [9]. Copyright © 2005, Elsevier, Inc. All rights reserved.

722

SYLVIA CHRISTAKOS

Unlike calbindin-D28K, calbindin-D9K is observed only in mammals. It has no amino acid sequence homology to calbindn-D28K. It is most abundant in mammalian duodenum, placenta, and uterus [13–16]. It is also present in mammalian yolk sac, lung, bone, and mouse kidney [14,17–23]. Human calbindin-D9K has 79 amino acid residues and a calculated molecular weight of 9015 [24,25]. It is 89% similar to the bovine and porcine sequences and 78% and 77% similar to rat and mouse calbindin-D9K, respectively [24,25]. Calbindin-D28K and calbindin-D9K belong to a family of high-affinity calcium (Ca2+) binding proteins (Kd =10−8 –10−6 M) that contains more than 200 members and is characterized by the EF-hand structural motif [26] (Fig. 1). The EF-hand domain is an octahedral structure consisting of two α helices separated by a 12-aminoacid loop that contains side chain oxygens necessary for orienting the divalent calcium cation [26]. CalbindinD28K contains six EF hands (Fig. 2); however, only four of these actively bind Ca2+ [27,28]. Calbindin-D9K contains two calcium binding sites [29]. Other calcium binding proteins belonging to this family include calmodulin, parvalbumin, troponin C, calretinin, calcineurin, calpain, Spec I, myosin light chains, S100, and recoverin [26,30]. Although the structure of

calbindin-D28K has yet to be elucidated by X-ray crystallography, circular dichroism experiments have shown that calbindin-D28K contains approximately 30% α helix, 20.6% β sheet, and 51% random coil [31]. The three-dimensional structure of calbindin-D9K has been elucidated [32]. Calbindin-D9K has been shown to undergo limited conformational change in the presence or absence of calcium [33]. Both calbindins are heatstable protein and acidic, having a pI value of approximately 5 [34,35]. The calbindins bind other cations in addition to calcium with reduced affinity: Ca2+ > Cd2+ > Sr2+ > Mn2+ > Zn2+ > Ba2+ > Co2+ > Mg2+ [36].

II. LOCALIZATION AND PROPOSED FUNCTIONAL SIGNIFICANCE A. Intestine One of the most pronounced effects of 1,25(OH)2D3 is increased synthesis of intestinal calbindin. Calbindin-D9K in mammalian intestine and calbindin-D28K in avian intestine have been localized primarily in the cytoplasm of absorptive cells [37], which supports the proposed role of calbindin in intestinal calcium

E helix

Ca2+

F helix Ca2+

EF hand

FIGURE 1

ET AL .

EF-hand structural motif (helix–loop–helix). The helices are represented by the extended forefinger and thumb. The clenched middle finger represents the loop that contains the oxygen ligands of the calcium ion. The EF hand is a recurring motif in calbindin and other calcium binding proteins. Reprinted with permission from Stryer L 1995 Biochemistry. Freeman, San Francisco, p. 1064.

723

CHAPTER 42 The Calbindins: Calbindin-D9K and Calbindin-D28K

Ca2+ Binding Domains

α-Helix Region

Ca2+ Binding Domains

FIGURE 2

Position of intervening sequences within the structure of chicken calbindin-D28K. Locations of introns are indicated by circled numbers. Numbers above amino acids indicate codon positions. Invariant Glu/Leu and Gly amino acids are indicated by black circles. Calcium binding domains are separated from the α-helix region by vertical lines. Reprinted with permission from Minghetti et al. [121].

absorption [38–40]. Early studies in chicks established a strong correlation between the level of calbindin and an increase in intestinal Ca2+ transport [41–43]. In the intestine, 1,25(OH)2D3 effects the transfer of Ca2+ across the luminal brush-border membrane, the transfer of calcium through the cell interior, and active calcium extrusion from the basolateral membrane. A vitamin D–inducible apical calcium channel has been identified in 1,25(OH)2D3 responsive epithelia (proximal

duodenum and distal tubule of the kidney), suggesting, for the first time, a mechanism of calcium entry [44–46]. It is thought that calbindin acts to facilitate the diffusion of calcium through the cell interior toward the basolateral membrane [40,42]. Supporting this hypothesis are findings observed in vitamin D receptor knockout mice. In these mice, the major defect that results in rickets is in intestinal calcium absorption [47–49]. The defect in intestinal calcium absorption is accompanied by

724 a 50% reduction in intestinal calbindin-D9K mRNA [50]. The 1,25(OH)2D3 regulation of intestinal calbindin-D9K is also evident in 25-hydroxyvitamin-D3 1α-hydroxylase knockout mice. In these mice, characterized by hypocalcemia, hyperparathyroidism, and skeletal abnormalities characteristic of rickets, intestinal calbindin-D9K mRNA is absent [51]. In addition to facilitated diffusion, it has been suggested that intestinal calbindin may also act as a cytosolic buffer to prevent toxic levels of calcium from accumulating in the intestinal cell during vitamin D–mediated translocation of calcium [40].

B. Kidney Immunocytochemical studies have reported the exclusive localization of calbindin-D28K in the distal nephron (distal convoluted tubule and connecting tubule) in a variety of species including mammals, chickens, and reptiles [3,9,52–54]. Renal calbindin-D28K is localized in the cytosol and the nucleus and is not associated with membranes or filamentous elements. Both calbindin-D28K and calbindin-D9K are localized in mouse distal nephron and perinatal rat distal nephron [23]. Autoradiographic data indicated that the VDR is also predominantly localized in the distal nephron and both calbindins have been reported to be induced by 1,25(OH)2D3 in the kidney [55,56]. Although micropuncture data [57] as well as studies using a mouse distal convoluted tubule cell line [58] have indicated that vitamin D metabolites can enhance calcium transport in the distal nephron, little information is available concerning the exact role of vitamin D–inducible renal calbindins in this process. Transcellular calcium transport in the distal convoluted tubule, similar to transcellular intestinal calcium absorption, involves calcium entry through the apical plasma membrane, diffusion of calcium across the cell, and active extrusion of calcium across the basolateral membrane mediated by a calcium-dependent ATPase [59]. Recent data using apical membrane vesicles reconstituted with calbindinD28K have indicated that this protein can increase the influx of calcium at the apical membrane [60]. Whether calbindin-D28K affects the activity of the recently identified vitamin D–inducible epithelial calcium channel in the distal tubule or other calcium channels previously identified in the distal tubule is not yet known [46,61,62]. It has also been suggested that calbindin-D28K may act to ferry calcium across the cell as in the intestine as well as to buffer calcium, resulting in protection against calcium mediated cell death [63,64]. Calbindin-D9K has been reported to have a different cellular action. Calbindin-D9K has been reported to bind

SYLVIA CHRISTAKOS

ET AL .

calcium and to stimulate ATP dependent extrusion of calcium at the basolateral membrane [65]. The different functions of renal calbindin-D28K and calbindin-D9K suggest different mechanisms that may be involved in the enhancement by 1,25(OH)2D3 of calcium transport in the distal nephron. Recent studies have provided additional evidence for a role of calbindin-D28K in distal tubular calcium reabsorption. Studies using immunosuppressant drugs (CsA and FK-506) that result in nephrotoxicity have noted decreases in calbindin-D28K in the rat coincident with increases in urinary calcium and intratubular calcification [66,67]. The authors suggested that calbindinD28K has a role in calciuria and tubular calcification induced by the immunosuppressant drugs. In addition, calbindin-D28K knockout mice fed a high-calcium diet were found to have significantly increased urinary calcium/creatinine ratio compared to wild-type controls [63,68]. The regulation of renal and intestinal calbindin-D9K was found to be similar in wild-type and knockout mice, indicating that changes in calbindin-D9K were not compensating for the lack of calbindin-D28K and further suggesting different roles for these two vitamin D–dependent calcium binding proteins [63]. Serum calcium was not different in the wild-type and calbindin-D28K knockout mice, suggesting compensatory changes in bone or in intestinal calcium absorption [63,68]. However, it should be noted that mechanisms within the kidney, independent of calbindin-D28K, are also associated with hypercalciuria [69]. In the future it will be of interest to develop calbindin-D9K knockout mice as well as calbindin-D28K and calbindin-D9K double knockout mice in order to provide additional insight into the role of each of these proteins in distal tubule calcium transport.

C. Bone Calbindin-D28K and calbindin-D9K are both present in chondrocytes of growth plate cartilage in rats and calbindin-D28K is present in the growth plate cartilage of chicks [22,70,71]. Although it is not clear whether calbindin is vitamin D dependent in chondrocytes, 1,25(OH)2D3 receptors have been reported in developing chick bone, specifically in dividing chondrocytes [72]. It has been suggested that calbindin may be involved in the movement of calcium in the process of calcification in the chondrocyte [70]. Calbindin-D9K and calbindin-D28K have also been localized to osteoblasts and ameloblasts of rodent teeth, and it has been reported that calbindin-D9K and calbindin-D28K mRNAs are induced by 1,25(OH)2D3 in these cells [20,21].

725

CHAPTER 42 The Calbindins: Calbindin-D9K and Calbindin-D28K

It has also been suggested that elevated expression of calbindin may phenotypically characterize cells that are involved in calcium handling during mineralization [20]. In addition to an association with mineralization, recent evidence has indicated that calbindin-D28K is able to protect against apoptosis of bone cells. Calbindin-D28K was found to protect osteoblastic cells against tumor necrosis factor (TNF) induced apoptosis (Fig. 3) as well as to prevent glucocorticoid-induced apoptosis of osteoblastic and osteocytic cells [73,74]. The protection against both TNF and glucocorticoid induced cell death was found to be at least partially due to the ability of calbindin-D28K to inhibit endogenous caspase 3, a key mediator of apoptosis in response to multiple signals. Calbindin-D28K was found to inhibit caspase 3 but was not cleaved by the caspase. In addition, the inhibition of caspase 3 by calbindin-D28K was reported to be independent of its calcium binding ability. Besides the inhibitor of apoptotic proteins (IAPs), calbindin-D28K is the only other known natural, nononcogenic inhibitor of caspase 3. These findings are novel

FIGURE 3

and have important implications for the prevention of cellular degeneration in bone cells.

D. Pancreas The pancreas was the first nonclassic target tissue in which receptors for 1,25(OH)2D3 were identified [75]. Although 1,25(OH)2D3 has been reported to play a role in insulin secretion, the exact mechanisms remain unclear [76–78]. An early indication that the pancreas may be a target for 1,25(OH)2D3 was the immunocytochemical study of Morrissey et al. [3], which localized calbindin-D28K to the islet. In the chick, calbindin-D28K is detected exclusively in insulin-producing β cells [79] and is responsive to vitamin D [80]. In the rat, however, calbindin-D28K has been reported to be localized in α as well as β cells of the pancreas [81]. Because autoradiographic data have indicated that 1,25(OH)2D3 receptors are localized only in rat β cells [82], and because insulin but not glucagon secretion is affected

Empty vector

Calbindin–D 28K cDNA

Empty vector + TNFα

Calbindin-D28K cDNA + TNFα

Overexpression of calbindin-D28K suppresses nuclear fragmentation of osteoblastic cells induced by TNFα. Cells were transfected with the expression vector pREP4 alone (empty vector) or containing the cDNA for calbindin-D28K (calbindin-D28K) together with an expression vector containing the coding sequence of green fluorescent protein with a nuclear localization sequence. Forty-eight hours after transfection, cells were exposed to 1 nM TNFα for 16 hr. Cells were fixed, mounted, and examined with a Zeiss confocal laser scanning microscope. Note the presence of apoptotic nuclei in the TNFα-treated vector-transfected cells but not in the calbindin-transfected cells similarly treated.

726 in vitamin D–deficient animals [76], studies in the rat suggest that β-cell calbindin may be regulated by 1,25(OH)2D3 while non-β-cell calbindin may be independent of vitamin D. Calbindin-D28K has also been identified in human pancreatic islet cells [83]. Recent studies using pancreatic beta cell lines as well as calbindin-D28K knockout mice have suggested that calbindin-D28K, by regulating intracellular calcium, modulates depolarization-stimulated insulin release [84]. In addition to modulating insulin release, more recent studies have indicated that calbindin-D28K, by buffering calcium, can protect against destruction of beta cells by cytokines by preventing calcium mediated mitochondrial damage and the resultant generation of free radicals [85]. These findings have important therapeutic implications for type 1 diabetes and the prevention of autoimmune destruction of pancreatic β cells.

SYLVIA CHRISTAKOS

ET AL .

uterus, mouse oviduct epithelium and in primary follicles of mouse ovary) [89]. 1,25(OH)2D3 has no effect on calbindin in these tissues. However, calbindin-D9K and calbindin-D28K in rat and chick uterus, respectively, are under the positive control of estradiol [16,90]. In the mouse, calbindin-D28K gene expression is downregulated in the uterus but not in the ovaries and oviduct, suggesting tissue and species specific regulation of calbindin-D28K by estradiol [89]. It has been suggested that transcellular calcium transport in epithelial cells of the uterus and oviduct is facilitated by calbindin [89]. The presence of calbindin in the myometrium suggests the involvement of calbindin in the modulation of intracellular calcium that may alter the frequency and strength of uterine contractions.

G. Nervous Tissue E. Testes Calbindin-D28K has been reported in both chick and rat testes [5,6]. In chick and rat, immunocytochemical studies have revealed that calbindin-D28K is present in spermatogonia and spermatocytes of the seminiferous tubules and some interstitial Leydig cells [5,6]. It has been reported that vitamin D–deficient chicks have significantly (threefold) lower testicular calbindin levels than vitamin D–replete chicks [86]. As calbindinD28K as well as VDR (which is also present in seminiferous tubules) have been shown to correlate with testicular maturation [6,87], the involvement of calbindin-D28K and vitamin D in spermatogenesis and steroidogenesis has been suggested [86].

F. Placenta, Yolk Sac, Egg Shell Gland, and Uterus Calbindin-D9K is present in the placenta and yolk sac of rats and mice [14,17]. Calbindin-D9K is also present in rat endometrium and myometrium [16]. In pregnant rats calbindin-D9K is also expressed in the uterine epithelium [16]. In placenta and yolk sac calbindin-D9K increases at the end of gestation, when there is increased calcium need of the fetus, suggesting a role for calbindinD9K in the transport of calcium to the fetus [14,17]. Unlike calbindin-D9K, calbindin-D28K is not present in rat reproductive tissue. However, calbindin-D28K has been localized in the tubular gland cells of the shell gland in the chick [88], which are involved in calcium secretion during egg-shell formation. Calbindin-D28K is also found in the reproductive tissues of female mice (endometrium and glandular epithelium of mouse

Calbindin-D28K is widely distributed throughout the brain of mammals, avians, reptiles, amphibians, fish, and mollusks [9]. It is present in most neuronal cell groups and fiber tracts and is localized in neuronal elements and some ependymal cells [8,90–92]. In brain, calbindin-D28K is not vitamin D dependent [9]. Neurons containing calbindin-D28K are found in the cerebral cortex in layers 2–4, primarily in pyramidal neurons [8,90–92]. In the hippocampus, both basket cells and pyramidal neurons in CA1 stain positively for calbindin, as do granule cells and fibers in the dentate gyrus [8,90–92]. Purkinje cells of the cerebellum stain most intensely for calbindin-D28K [8,90–92]. It is of interest that the phenotype of the calbindin-D28K knockout mouse is impaired motor coordination [93]. It has been suggested that this phenotype may be the result of abnormal cerebellar activity due to the alteration of synaptically evoked calcium transients in the Purkinje cells in the absence of calbindin [93]. Calbindin-D28K immunoreactive cells are also observed in the hypothalamus, amygdala, pyriform region, and thalamus [8,90–92]. In addition, specific neuronal sensory cells have been shown to contain calbindin-D28K [94–100]. These cell populations include cochlear and vestibular hair cells in the inner ear [94,95], avian basilar papilla [96], cone but not rod photoreceptor cells of avian and mammalian retina [97–99], and conelike, modified photoreceptor cells (pinealocytes) of pineal transducers [100]. The presence of calbindin in specific cells of the sensory pathway suggests the possible involvement of calbindin in mechanisms of signal transduction. In the nervous system it has been suggested that neuronal calbindin, by buffering calcium, can regulate intracellular calcium responses to physiological stimuli and can protect neurons against calcium-mediated

CHAPTER 42 The Calbindins: Calbindin-D9K and Calbindin-D28K

neurotoxicity [101,102]. It has been demonstrated that introduction of exogenous calbindin into sensory neurons can modulate calcium signaling by decreasing the rate of rise of intracellular calcium and by changing the kinetics of decay of the calcium signal [103]. Using adenovirus as an expression vector, overexpression of calbindin-D28K in hippocampal neurons was reported to suppress posttetanic potentiation, possibly by restricting and destabilizing the evoked calcium signal [104]. Calbindin-D28K was also reported to play a role in the control of hypothalamic neuroendocrine neuronal firing patterns [105]. The whole-cell patchclamp method was used to introduce calbindin into rat supraoptic neurons. Calbindin-D28K suppressed Ca2+dependent depolarization afterpotentials and converted phasic into continuous firing [106]. As different firing patterns promote the release of different hypothalamic hormones, it was suggested that calbindin-D28K, by regulating firing patterns, may be involved in the control of hormone secretion from hypothalamic neuroendocrine neurons. These studies [103–105] indicated directly by introduction of exogenous calbindin-D28K via the patch clamp method or by transfection and overexpression that calbindin is an important and effective regulator of calcium-dependent aspects of neuronal function. Correlative evidence between decreases in neuronal calbindin-D28K and neurodegeneration in studies of ischemic injury [106], seizure activity [107,108], and chronic neurodegeneration (Alzheimer’s, Huntington’s, and Parkinson’s diseases) [109–111] have been reported. It has been suggested that decreased calbindin levels may lead to a loss of calcium buffering or intracellular calcium homeostasis, which leads to cytotoxic events associated with neuronal damage and cell death. Direct evidence of a protective role of neuronal calbindinD28K against a variety of insults including exposure to hypoglycemia and IgG from amyotrophic lateral sclerosis patients has been shown in primary cultures of neuronal cells or in neuronal cell lines in which the calbindin-D28K gene has been transfected [112,113]. Expression of calbindin-D28K in neural cells was also found to suppress the proapoptotic actions of mutant presenilin 1 (PS-1), which is causally linked to about 50% of the cases of early-onset familial Alzheimer’s disease [114]. Mutant PS-1 has been reported to sensitize cells to apoptosis induced by amyloid β peptide (Aβ), the major component of plaques in Alzheimer’s disease. Aβ has been reported to damage neurons by a mechanism involving oxidative stress and disruption of calcium homeostasis. Calbindin-D28K protected against the proapoptotic action of mutant PS-1 by attenuating the increase in intracellular calcium and preventing the impairment of mitochondrial function [114].

727

Since calbindin-D28K can prevent neuronal damage in neuropathies, these findings have important therapeutic implications.

H. Calbindin-D28K and Apoptosis Calcium is thought to play an important regulatory function in apoptosis, but the precise mechanism(s) by which calcium promotes cell death is unknown. The first study suggesting that calbindin-D28K plays a protective role in the process of apoptosis used subtraction analysis between the cDNA libraries of two human prostate cell lines [115]. One of the cell lines was androgen independent and the other one androgen dependent, and the results revealed that a hybrid calbindin-D28K gene was specifically expressed in the hormoneindependent cell line. Apoptosis has been observed in prostate cells on androgen depletion. Androgen deprivation of prostate cells triggers an influx of calcium ions into the cells, leading to an increase in intracellular calcium. It was suggested that calbindin-D28K might buffer intracellular calcium and contribute to protection against apoptosis and thus androgen independence in the prostatic cell line. It has been reported that stable transfection and overexpression of calbindin-D28K in lymphocytes protect against apoptosis induced by calcium ionophore, cAMP, and glucocorticoid [116]. A similar protective role for calbindin-D28K has been observed in apoptosissusceptible cells in the central nervous system [112–114] as well as in human embryonic kidney cells (HEK 293), osteoblasts, and pancreatic β cells [64,73,74,85]. These findings indicate that calbindin-D28K has a major role in different cell types in protecting against apoptotic cell death. A further understanding of the mechanisms involved will have important therapeutic implications for the prevention of a number of diseases including osteoporosis and diabetes.

I. Calbindin-D28K Enzyme Activation and Other Potential Targets In addition to its role as an intracellular calcium buffer and in transepithelial calcium transport, there is in vitro evidence that calbindin-D28K may modulate the activities of calmodulin-sensitive enzymes such as calcium-dependent ATPase and phosphodiesterase [117]. There is also evidence for a calcium-dependent, specific interaction of calbindin-D28K with intestinal alkaline phosphatase [118]. In addition recent studies using bacteriophage display have suggested that myo-inositol monophosphatase, a key enzyme in the regulation of

728 the activity of the phosphatidylinositol-signaling pathway, is an activated target of calbindin-D28K [119]. Although the physiological relevance of these findings is not known, these studies suggest that calbindin-D28K may act as an enzyme regulator. In addition, studies in opossum kidney cells indicated that expression of calbindin in these cells resulted in increased phosphate transport that was associated with alterations in the actin cytoskeleton. Pollock and Santiesteban suggested a possible role for calbindin in cytoskeletal reorganization [120].

III. REGULATION OF CALBINDIN GENE EXPRESSION A. Calbindin-D28K 1. GENOMIC ORGANIZATION OF THE CALBINDIN-D28K GENE

The genomic organization of the chicken calbindinD28K gene has been elucidated [121,122] and a partial structure for the human gene was also reported [12,123]. The human gene is located on chromosome 8 [123,124] and is believed to consist of 11 exons, analogous to that demonstrated for the avian gene. Moreover, the total size of the gene is reported to be 18.5 kb in chicken. The protein coding region of the mouse gene shares 77% sequence similarity with the chicken gene [125]. However, no obvious sequence similarity exists between the mammalian and avian promoters except in the region of the TATA box [12,126]. 2. REGULATION BY 1,25(OH)2D3

It is well known that calbindin-D28K in the avian intestine [1,127] and kidney [127] and in the mammalian kidney [128,129] is induced by 1,25(OH)2D3. In chicken, a putative calbindin-D28K vitamin D response element (VDRE) was suggested after computer analysis of the promoter sequence [130], but only a twofold response to 1,25(OH)2D3 was detected in primary kidney cells after transfection with a 2.1-kb segment of the 5′ flanking region of the promoter [131]. A relatively inactive putative VDRE was also reported in the chicken calbindin-D28K promoter by others [132,133] and the response of the mouse calbindin-D28K promoter to 1,25(OH)2D3 is modest [five-fold maximal induction in chloramphenicol acetyltransferase (CAT) activity] [126]. The modest response reflects previous in vivo findings that indicated a small transcriptional response to 1,25(OH)2D3 [124,129]. Similar findings were reported for the in vivo induction of the chick intestinal calbindinD28K gene by 1,25(OH)2D3 [134]. In addition, in VDR

SYLVIA CHRISTAKOS

ET AL .

knockout mice only a small reduction in basal levels of renal calbindin-D28K is observed. However, the response of renal calbindin-D28K in the VDR knockout mouse to 1,25(OH)2D3 is compromised [50]. Also in 25-hydroxyvitamin D3 1α-hydroxylase knockout mice the expression of renal calbindin-D28K is reduced [51]. There is now increasing evidence that the large induction of calbindin-D28K mRNA long after 1,25(OH)2D3 treatment may be due primarily to posttranscriptional mechanisms [124,129,134–136]. Exactly how this action is exerted is not known, but one report suggests that 1,25(OH)2D3 may regulate the expression of an intermediate protein that may be involved in calbindin-D28K mRNA accumulation [136]. These studies suggest that the mechanism of action of 1,25(OH)2D3 on calbindinD28K regulation is more complicated than the conventional hormone receptor–transcriptional activation model, and that this regulation may involve other factors and is mostly posttranscriptional. 3. REGULATION BY OTHER STEROIDS FACTORS

AND

Further studies in intestine and kidney have provided evidence that the calbindin-D28K gene is not exclusively regulated by 1,25(OH)2D3 and that other factors can modulate gene expression. It has been reported that glucocorticoids can inhibit the levels of calbindin-D28K mRNA and protein in intestine of vitamin D–treated chicks, resulting in a comparable decrease in intestinal calcium absorption [137,138]. These findings suggest the involvement of the inhibition of intestinal calbindin in the clinically important hypocalcemic action of glucocorticoids. Adding to the growing body of data suggesting that the regulation of calbindin is more complex than previously thought are studies describing the modulation by calcium of intestinal and renal calbindin-D28K gene expression [135,139,140]. In other tissues where calbindin-D28K is present in significant amounts, for example, in parts of the brain, the regulation of calbindin-D28K appears to be very different from that in the intestine and kidney. In the central nervous system (CNS), 1,25(OH)2D3 has no apparent effect on the levels of calbindin-D28K [9]. Instead, a variety of different factors have been reported to be involved in regulating neuronal calbindin-D28K. Using rat hippocampal cultures, evidence has indicated that neurotropin 3 (NT-3) [141,142], brain-derived neurotropic factor (BDNF) [141,142], fibroblast growth factor (FGF) [141], and tumor necrosis factors (TNFs) [143,144] all can induce calbindin. It has been reported that neurotropic factors may protect against excitotoxic neuronal damage [145]. The induction of calbindin by

729

CHAPTER 42 The Calbindins: Calbindin-D9K and Calbindin-D28K

those factors suggests a role for calbindin-D28K in the process of protection against cytotoxicity. In addition, corticosterone administration in vivo has been reported to increase calbindin-D28K expression in rat hippocampus [146,147]. The glucocorticoid response is specifically localized to the CA1 region [148]. Retinoic acid has also been reported to induce calbindinD28K protein and mRNA in medulloblastoma cells, which express a neuronal phenotype [148]. Furthermore, the content of calbindin-D28K in cultured Purkinje cells can be increased by insulin-like growth factor I (IGF-I) [149]. Similarly IGF-I, as well as insulin, promoted the expression of calbindin-D28K protein in cultured rat embryonic neuronal cells [150]. Thus, neuronal calbindin-D28K can be regulated by steroids as well as by factors that affect signal transduction pathways. These different modes of activation may be important for cell-specific effects of calbindin-D28K. The molecular basis for tissue-specific calbindin-D28K regulation is still not understood, but in vivo experiments using transgenic mice suggest that tissue specificity of calbindin-D28K expression to some degree is controlled by separate elements on the promoter [151]. The transgenic mouse study demonstrated the importance of an in vivo system to investigate the role of sequence elements needed for tissue-specific gene expression and regulation [151]. As discussed previously, calbindin-D28K is also present in the avian egg-shell gland and in the reproductive tissues of female mice. In the avian egg-shell gland, estradiol-17β induces calbindin-D28K in in vivo experiments [90]. In the mouse, calbindin-D28K gene expression was found to be down-regulated by estradiol in the uterus and oviduct [89,152] but up-regulated in the ovaries [152]. Multiple imperfect half-palindromic estrogen responsive elements, which are likely to mediate the estrogen responsiveness of the calbindinD28K gene by estradiol-17β, were present in two regions (−1075/−702 and −175/−78) of the promoter [152]. The complex regulation of calbindin-D28K highlights the fact that while discovered as a vitamin D–regulated gene product, the protein play roles in Ca2+ regulation well beyond that involving vitamin D3.

B. Calbindin-D9K 1. GENOMIC ORGANIZATION OF THE CALBINDIN-D9K GENE

The size of the calbindin-D9K gene is 2.5 kb, and the gene consists of three exons and two introns [153]. The first exon contains the 5′ untranslated region. The second exon codes for the first calcium binding site.

The third exon codes for the second calcium binding site and the 3′ untranslated region. 2. REGULATION OF CALBINDIN-D9K BY 1,25(OH)2D3

Similar to the regulation of avian intestinal calbindinD28K and mammalian renal calbindin-D28K, in vivo findings have indicated that intestinal calbindin-D9K is regulated by 1,25(OH)2D3 by a small, rapid transcriptional stimulation followed by a posttranscriptional effect accounting for a sustained accumulation of mRNA long after cessation of 1,25(OH)2D3 treatment [154]. Unlike renal calbindin-D28K, there is a marked decrease of basal as well as 1,25(OH)2D3 induced levels of intestinal calbindin-D9K mRNA in VDR knockout mice, suggesting that intestinal calbindin-D9K is more sensitive to control by VDR-mediated mechanisms than calbindin-D28K [50]. Recent studies using transgenic mice have shown that the proximal promoter of the calbindin-D9K gene from –117 to +365 and distal element at –3731 to –2894 together but not separately confer the 1,25(OH)2D3 induced transcriptional response. Since this region does not contain a classical VDRE, these findings suggest that the 1,25(OH)2D3 regulation of calbindin-D9K may involve a nonconventional activation pathway. 3. REGULATION OF CALBINDIN-D9K BY OTHER STEROIDS AND FACTORS

In vitro footprinting and gel shift assays suggested that several trans-acting factors other than the VDR, including an ubiquitous factor (NF1), liver-enriched factors (HNF1, C/EBP alpha and beta, and HNF4), and the intestine-specific transcription factor caudal homeobox-2 (Cdx-2), may be important for intestine specific calbindin-D9K gene expression [156]. Further studies using transgenic mice showed that a mutation in the distal Cdx2-binding site of calbindin-D9K promoter dramatically decreased intestinal expression of the calbindin-D9K gene, directly demonstrating the crucial role of Cdx2 for the transcription of this gene in the intestine [157]. With regard to other steroids, besides 1,25(OH)2D3 the expression of calbindin-D9K in the intestine is also regulated by glucocorticoids. Glucocorticoids have been reported to inhibit intestinal calbindin-D9K expression [158,159]. It has been suggested that this decrease may be involved in the reported decrease by glucocorticoids in intestinal calcium absorption. Whether the effect on intestinal calbindin-D9K expression is a primary or a secondary action of glucocorticoids is not yet known. In the uterus, calbindin-D9K is under the control of estrogen but is unaffected by 1,25(OH)2D3. An imperfect

730

SYLVIA CHRISTAKOS

ET AL .

TABLE I Distribution of Calbindin Calbindin-D9K Mammalian intestine [13] Mouse and neonatal rat kidney [23] Rat and mouse yolk sac [14,17] Rat uterus [15,16] Rat and mouse placenta [14] Rat growth cartilage [22] Ameloblasts and osteoblasts of rodent teeth [20,21] Rat lung [18]

Calbindin-D28K Avian intestine [3,34,37] Avian, reptilian, amphibian, and mammalian kidney [3,52–54] Hen egg-shell gland (uterus) [90] Mouse reproductive tissues (uterus, oviduct, ovary) [89] Avian and mammalian beta cells of the pancreas [79,83] Alpha cells of the rat pancreas [81] Rat and chick growth cartilage [22,70,71] Ameloblasts and osteoblasts of rodent teeth; mouse osteoblasts [20,21,73] Brain (avian, reptilian, amphibian, molluskan, fish, and mammalian brain) [9,91,92]

estrogen response element (ERE) that binds the estrogen receptor (ER) has been identified at the border of the first exon and the first intron [160,161]. In vivo experiments using transgenic mice suggest the functionality of this imperfect ERE [162,163].

IV. CONCLUSION We once viewed calbindin-D28K and calbindin-D9K as exclusively vitamin D–dependent proteins. It is now evident that the calbindins are not under the exclusive regulatory control of 1,25(OH)2D3. Calbindin-D28K and calbindin-D9K are present in many different tissues (See Table I) and may serve many different functions. Accordingly, the regulation of these calcium binding proteins is varied and quite complex. In future studies, the generation of calbindin-D9K knockout mice as well as studies using mice in which both calbindins are absent will provide new insight into the mechanism of action of the calbindins, including the role of calbindin in intestinal calcium absorption, in calcium reabsorption in the distal nephron, and in protection against cell death.

References 1. Wasserman RH, Taylor AN 1966 Vitamin D3 induced calcium binding protein in chick intestinal mucosa. Science 152:791–793. 2. Wasserman RH 1985 Nomenclature of the vitamin D induced calcium binding proteins. In: Norman AW, Schaefer K, Grigoleit HG, Herrath DV (eds) Vitamin D: Chemical, Biochemical and Clinical Update, de Gruyter, Berlin, pp. 321–322. 3. Morrissey RL, Bucci TJ, Richard B, Empson RN, Lufkin EG 1975 Calcium binding protein: Its cellular localization in jejunum, kidney and pancreas. Proc Soc Exp Biol Med 149:56–60.

4. Christakos S, Norman AW 1978 A vitamin D-dependent calcium binding protein in bone tissue. Science 202:70–71. 5. Inpanbutr N, Taylor AN 1992 Expression of calbindin-D28K in developing and growing chick testes. Histochemistry 97:335–339. 6. Kagi U, Chafouleas JG, Norman AW, Heizmann CW 1988 Developmental appearance of the Ca2+-binding proteins parvalbumin, calbindin-D28K, S100 proteins and calmodulin during testicular development in the rat. Cell Tissue Res 252:359–365. 7. Taylor AN 1974 Chick brain calcium binding protein: Comparison with intestinal vitamin D–induced calcium binding protein. Arch Biochem Biophys 161:100–108. 8. Jande SS, Maler L, Lawson DEM 1981 Immunohistochemical mapping of vitamin D–induced calcium binding protein in brain. Nature 294:765–767. 9. Christakos S, Gabrielides C, Rhoten WB 1989 Vitamin D– dependent calcium binding proteins: Chemistry, distribution, functional considerations, and molecular biology. Endocr Rev 10:84–107. 10. Fullmer CS, Wasserman RH 1987 Chicken intestinal 28-kilodalton calbindin-D: Complete amino acid sequence and structural considerations. Proc Natl Acad Sci USA 84: 4772–4776. 11. Hunziker W, Schrickel S 1988 Rat brain calbindin-D28K: Six domain structure and extensive amino acid homology with chicken calbindin-D28K. Mol Endocrinol 2:465–473. 12. Parmentier M 1990 Structure of the human cDNAs and genes coding for calbindin-D28K and calretinin. Adv Exp Med Biol 140:21–25. 13. Thomasset M, Parkes CO, Cuisinier-Gleizes P 1982 Rat calcium-binding proteins: distribution, development, and vitamin D dependence. Am J Physiol 243:E483–E488. 14. Bruns ME, Kleeman E, Mills SE, Bruns DE, Herr JC 1985 Immunochemical localization of vitamin D-dependent calcium-binding protein in mouse placenta and yolk sac. Anat Rec 213:514–517, 532–535. 15. Warembourg M, Perret C, Thomasset M 1987 Analysis and in situ detection of cholecalcin messenger RNA (9000 Mr CaBP) in the uterus of the pregnant rat. Cell Tissue Res 247:51–57. 16. Bruns ME, Overpeck JG, Smith GC, Hirsch GN, Mills SE, Bruns DE 1988 Vitamin D–dependent calcium binding protein in rat uterus: differential effects of estrogen, tamoxifen, progesterone, and pregnancy on accumulation and cellular localization. Endocrinology 122:2371–2378.

CHAPTER 42 The Calbindins: Calbindin-D9K and Calbindin-D28K 17. Delorme AV, Cassier P, Geny B, Mathieu H 1983 Immunocytochemical localization of vitamin D–dependent calcium-binding protein in the yolk sac of the rat. Placenta 4:263–270. 18. Dupret JM, L’Horset F, Perret C, Bernaudin JF, Thomasset M 1992 Calbindin-D9K gene expression in the lung of the rat. Absence of regulation by 1,25-dihydroxyvitamin D3 and estrogen. Endocrinology 131:2643–2648. 19. Balmain N 1991 Calbindin-D9K. A vitamin-D-dependent, calcium-binding protein in mineralized tissues. Clin Orthop 265:265–276. 20. Berdal A, Hotton D, Saffar JL, Thomasset M, Nanci A 1996 Calbindin-D9K and calbindin-D28K expression in rat mineralized tissues in vivo. J Bone Miner Res 11:768–779. 21. Berdal A, Hotton D, Pike JW, Mathieu H, Dupret JM 1993 Cell- and stage-specific expression of vitamin D receptor and calbindin genes in rat incisor: regulation by 1,25-dihydroxyvitamin D3. Dev Biol 155:172–179. 22. Balmain N, Tisserand-Jochem E, Thomasset M, CuisinierGleizes P, Mathieu H 1986 Vitamin-D-dependent calciumbinding protein (CaBP-9K) in rat growth cartilage. Histochemistry 84:161–168. 23. Rhoten WB, Bruns ME, Christakos S 1985 Presence and localization of two vitamin D–dependent calcium binding proteins in kidneys of higher vertebrates. Endocrinology 117:674–683. 24. Howard A, Legon S, Spurr NK, Walters JR 1992 Molecular cloning and chromosomal assignment of human calbindinD9K. Biochem Biophys Res Commun 185:663–669. 25. Jeung EB, Krisinger J, Dann JL, Leung PC 1992 Molecular cloning of the full-length cDNA encoding the human calbindinD9K. FEBS Lett 307:224–228. 26. Kawasaki H, Nakayama S, Kretsinger RH 1998 Classification and evolution of EF-hand proteins. Biometals 11:277–295. 27. Hunziker W 1986 The 28-kDa vitamin D-dependent calcium-binding protein has a six-domain structure. Proc Natl Acad Sci USA 83:7578–8582. 28. Akerfeldt KS, Coyne AN, Wilk RR, Thulin E, Linse S 1996 Ca2+-binding stoichiometry of calbindin-D28K as assessed by spectroscopic analyses of synthetic peptide fragments. Biochemistry 35:3662–3669. 29. Nelson MR, Thulin E, Fagan PA, Forsen S, Chazin WJ 2002 The EF-hand domain: A globally cooperative structural unit. Protein Sci 11:198–205. 30. Heizmann CW, Braun K 1992 Changes in Ca2+-binding proteins in human neurodegenerative disorders. Trends Neurosci 15:259–264. 31. Gross MD, Nelsestuen GL, Kumar R 1987 Observations on the binding of lanthanides and calcium to vitamin D–dependent chick intestinal calcium binding protein. J Biol Chem 262: 6539–6545. 32. Maler L, Blankenship J, Rance M, Chazin WJ 2000 Site-site communication in the EF-hand Ca2+-binding protein calbindin D9K. Nat Struct Biol 7:245–250. 33. Skelton NJ, Kordel J, Akke M, Forsen S, Chazin WJ 1994 Signal transduction versus buffering activity in Ca2+-binding proteins. Nat Struct Biol 1:239–245. 34. Wasserman RH, Corradino RA, Taylor AN 1968 Vitamin D dependent calcium binding protein: Purification and some properties. J Biol Chem 243:3978–3986. 35. Kesvatera T, Jonsson B, Thulin E, Linse S 1999 Ionization behavior of acidic residues in calbindin D9K. Proteins 37:106–115. 36. Ingersoll RJ, Wasserman RH 1971 Vitamin D3 induced calcium binding protein: Binding characteristics, conformational effects and other properties. J Biol Chem 246:2808–2814.

731 37. Taylor AN 1981 Immunohistochemical localization of vitamin D–induced calcium binding protein: Relocation of antigen during frozen section processing. J Histochem Cytochem 29:65–73. 38. Feher JJ 1983 Facilitated calcium diffusion by intestinal calcium-binding protein. Am J Physiol 244:C303–C307. 39. Bronner F, Pansu D, Stein WD 1986 An analysis of intestinal calcium transport across the rat intestine. Am J Physiol 150: G561–G569. 40. Wasserman RH, Fullmer CS 1995 Vitamin D and intestinal calcium transport: Facts, speculations and hypotheses. J Nutr 125:1971S–1979S. 41. Taylor AN, Wasserman RH 1969 Correlations between the vitamin D–induced calcium binding protein and intestinal absorption of calcium. Fed Proc 28:1834–1838. 42. Feher JJ, Wasserman RH 1979 Calcium absorption and intestinal calcium-binding protein: Quantitative relationship. Am J Physiol 236:E556–E561. 43. Corradino RA, Fullmer CS, Wasserman RH 1976 Embryonic chick intestine in organ culture: Stimulation of calcium transport by exogenous vitamin D–induced calcium-binding protein. Arch Biochem Biophys 174:738–743. 44. Hoenderop JG, van der Kemp AW, Hartog A, van de Graaf SF, van Os CH, Willems PH, Bindels RJ 1999 Molecular identification of the apical Ca2+ channel in 1,25-dihydroxyvitamin D3–responsive epithelia. J Biol Chem 274:8375–8378. 45. Peng JB, Chen XZ, Berger UV, Vassilev PM, Tsukaguchi H, Brown EM, Hediger MA 1999 Molecular cloning and characterization of a channel-like transporter mediating intestinal calcium absorption. J Biol Chem 274:22739–22746. 46. Peng JB, Chen XZ, Berger UV, Vassilev PM, Brown EM, Hediger MA 2000 A rat kidney-specific calcium transporter in the distal nephron. J Biol Chem 275:28186–28194. 47. Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997 Targeted ablation of the vitamin D receptor: an animal model of vitamin D-dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94:9831–9835. 48. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Genet 16:391–396. 49. Amling M, Priemel M, Holzmann T, Chapin K, Rueger JM, Baron R, Demay MB 1999 Rescue of the skeletal phenotype of vitamin D receptor-ablated mice in the setting of normal mineral ion homeostasis: Formal histomorphometric and biomechanical analyses. Endocrinology 140: 4982–4987. 50. Li YC, Pirro AE, Demay MB 1998 Analysis of vitamin D– dependent calcium-binding protein messenger ribonucleic acid expression in mice lacking the vitamin D receptor. Endocrinology 139:847–851. 51. Panda DK, Miao D, Tremblay ML, Sirois J, Farookhi R, Hendy GN, Goltzman D 2001 Targeted ablation of the 25-hydroxyvitamin D 1α-hydroxylase enzyme: Evidence for skeletal, reproductive, and immune dysfunction. Proc Natl Acad Sci USA 98:7498–7503. 52. Rhoten WB, Christakos S 1981 Immunocytochemical localization of vitamin D–dependent calcium binding protein in mammalian nephron. Endocrinology 109:981–983. 53. Roth J, Thorens B, Hunziker W, Norman AW, Orci L 1981 Vitamin D–dependent calcium binding protein: Immunocytochemical localization in chick kidney. Science 214: 197–200.

732 54. Rhoten WB, Lubit B, Christakos S 1984 Avian and mammalian vitamin D–dependent calcium-binding protein in reptilian nephron. Gen Comp Endocrinol 55:96–103. 55. Stumpf WE, Sar M, Reid FA, Tanaka Y, DeLuca HF 1979 Target cells for 1,25-dihydroxyvitamin D3 in intestinal tract, stomach, skin, pituitary and parathyroid. Science 209: 1188–1190. 56. Stumpf WE, Sar M, Narbauitz R, Reid FA, DeLuca HF, Tanaka Y 1980 Cellular and subcellular localization of 1,25(OH)2D3 in rat kidney: Comparison with localization of parathyroid hormone and estradiol. Proc Natl Acad Sci USA 77:1149–1153. 57. Winaver J, Sylk DB, Robertson JS, Chen TC, Puschett JB 1980 Micropuncture study of the acute renal tubular transport effects of 25-hydroxyvitamin D3 in the dog. Miner Electrolyte Metab 4:178–188. 58. Friedman PA, Gesek FA 1993 Vitamin D accelerates PTHdependent calcium transport in distal convoluted tubule cells. Am J Physiol 265:F300–F308. 59. Friedman PA, Gesek FA 1993 Calcium transport in renal epithelial cells. Am J Physiol 264:F181–F198. 60. Bouhtiauy I, Lajeunesse D, Christakos S, Brunette MG 1994 Two vitamin D3–dependent calcium binding proteins increase calcium reabsorption by different mechanisms I: Effect of CaBP28K. Kidney Int 45:461–468. 61. Hoenderop JG, Muller D, Van Der Kemp AW, Hartog A, Suzuki M, Ishibashi K, Imai M, Sweep F, Willems PH, Van Os CH, Bindels RJ 2001 Calcitriol controls the epithelial calcium channel in kidney. J Am Soc Nephrol 12:1342–1349. 62. Yu AS, Boim M, Hebert SC, Castellano A, Perez-Reyes E, Lytton J 1995 Molecular characterization of renal calcium channel beta-subunit transcripts. Am J Physiol 268:F525–F531. 63. Sooy K, Kohut J, Christakos S 2000 The role of calbindin and 1,25-dihydroxyvitamin D3 in the kidney. Curr Opin Nephrol Hypertens 9:341–347. 64. Turner PR, Mefford S, Christakos S, Nissenson RA 2000 Apoptosis mediated by activation of the G protein-coupled receptor for parathyroid hormone (PTH)/PTH-related protein (PTHrP). Mol Endocrinol 14:241–254. 65. Bouhtiauy I, Lajeunesse D, Christakos S, Brunette MG 1994 Two vitamin D3-dependent calcium binding proteins increase calcium reabsorption by different mechanisms. II. Effect of CaBP 9K. Kidney Int 45:469–474. 66. Grenet O, Bobadilla M, Chibout SD, Steiner S 2000 Evidence for the impairment of the vitamin D activation pathway by cyclosporine A. Biochem Pharmacol 59:267–272. 67. Yang CW, Kim J, Kim YH, Cha JH, Mim SY, Kim YO, Shin YS, Kim YS, Bang BK 1998 Inhibition of calbindin-D28K expression by cyclosporin A in rat kidney: The possible pathogenesis of cyclosporin A-induced hypercalciuria. J Am Soc Nephrol 9:1416–1426. 68. Lee CT, Huynh VM, Lai LW, Lien YH 2002 Cyclosporine A–induced hypercalciuria in calbindin-D28K knockout and wild-type mice. Kidney Int 62:2055–2061. 69. Caride AJ, Chini EN, Penniston JT, Dousa TP 1999 Selective decrease of mRNAs encoding plasma membrane calcium pump isoforms 2 and 3 in rat kidney. Kidney Int 56:1818–1825. 70. Balmain N, Brehier A, Cuisinier-Gleizes P, Mathieu H 1986 Evidence for the presence of calbindin-D28K (CaBP28K) in the tibial growth cartilages of rats. Cell Tissue Res 245:331–335. 71. Zhou XY, Dempster DW, Marion SL, Pike JW, Haussler MR, Clemens TL 1986 Bone vitamin D–dependent calcium binding protein is localized in chondrocytes of growth plate cartilage. Calcif Tissue Int 38:244–247.

SYLVIA CHRISTAKOS

ET AL .

72. Suda S, Takahashi N, Shinki T, Horiuchi N, Yamaguchi A, Yoshiki S, Enomato S, Suda T 1985 1,25-Dihydroxyvitamin D3 receptors and their actions in embryonic chick. Calcif Tissue Int 37:82–90. 73. Bellido T, Huening M, Raval-Pandya M, Manolagas SC, Christakos S 2000 Calbindin-D28K is expressed in osteoblastic cells and suppresses their apoptosis by inhibiting caspase-3 activity. J Biol Chem 275:26328–26332. 74. Liu Y, Porta A, Peng X, Gengaro K, Cunningham EB, Li H, Dominguez LA, Bellido T, Christakos S 2004 Prevention of glucocorticoid induced apoptosis in osteocytes and osteoblasts by calbindin-D28K. J Bone Miner Res 19:479–490. 75. Christakos S, Norman AW 1979 Evidence for a specific high affinity protein for 1,25-dihydroxyvitamin D3 in chick kidney and pancreas. Biochem Biophys Res Commun 89: 56–63. 76. Norman AW, Frankel BJ, Heldt AM, Grodsky GM 1980 Vitamin D deficiency inhibits pancreatic secretion of insulin. Science 209:823–825. 77. Kadowaki S, Norman AW 1985 Demonstration that the vitamin D metabolite 1,25(OH)2D3 and not 24R,25(OH)2vitamin D3 is essential for normal insulin secretion in the perfused rat pancreas. Diabetes 34:315–320. 78. Lee, S, Clark SA, Gill RK, Christakos S 1994 1,25-Dihydroxyvitamin D3 and pancreatic β-cell function: Vitamin D receptors, gene expression and insulin secretion. Endocrinology 134:1602–1610. 79. Roth J, Bonner-Weir S, Norman AW, Orci L 1982 Immunocytochemistry of vitamin D–dependent calcium binding protein in chick pancreas: Exclusive localization in β cells. Endocrinology 111:2216–2218. 80. Kadowaki S, Norman AW 1984 Pancreatic vitamin D– dependent calcium binding protein: Biochemical properties and response to vitamin D. Arch Biochem Biophys 233:228–236. 81. Pochet R, Pipeleers DG, Malaisse WJ 1987 Calbindin-D28K: preferential localization in non-β islet cells of the rat pancreas. Biol Cell 61:155–161. 82. Clark SA, Stumpf WE, Sar M, DeLuca HF, Tanaka Y 1980 Target cells for 1,25-dihydroxyvitamin D3 in the pancreas. Cell Tissue Res 209:515–520. 83. Johnson JA, Grande JP, Roche PC, Kumar R 1994 Immunohistochemical localization of the 1,25(OH)2D3 receptor and calbindin-D28K in human and rat pancreas. Am J Physiol 267:E356–E360. 84. Sooy K, Schermerhorn T, Noda M, Surana M, Rhoten WB, Meyer M, Fleischer N, Sharp GW, Christakos S 1999 Calbindin-D28K controls [Ca2+]i and insulin release. Evidence obtained from calbindin-D28K knockout mice and beta cell lines. J Biol Chem 274:34343–34349. 85. Rabinovitch A, Suarez-Pinzon WL, Sooy K, Strynadka K, Christakos S 2001 Expression of calbindin-D28K in a pancreatic islet beta-cell line protects against cytokine-induced apoptosis and necrosis. Endocrinology 142:3649–3655. 86. Inpanbutr N, Reiswig JD, Bacon WI, Siemens RD, Iacopino AM 1996 Effect of vitamin D on testicular CaBP28K expression and serum testosterone in chickens. Biol Reprod 54:242–248. 87. Walters MR 1984 1,25-Dihydroxyvitamin D3 receptors in seminiferous tubules of the rat testes increase at puberty. Endocrinology 114:2167–2173. 88. Lippiello L, Wasserman RH 1975 Fluorescent antibody localization of the vitamin D dependent calcium binding protein in the oviduct of the laying hen. J Histochem Cytochem 23:111–116.

CHAPTER 42 The Calbindins: Calbindin-D9K and Calbindin-D28K 89. Opperman LA, Saunders TJ, Bruns DE, Boyd JC, Mills SE, Bruns ME 1992 Estrogen inhibits calbindin-D28K expression in mouse uterus. Endocrinology 130:1728–1735. 90. Corradino RA, Smith CA, Krook LP, Fullmer CS 1993 Tissue-specific regulation of shell gland calbindin-D28K biosynthesis by estradiol in precociously matured, vitamin D-depleted chicks. Endocrinology 132:193–198. 91. Roth J, Baeten D, Norman AW, Garcia-Sergura LM 1981 Specific neurons in chick central nervous system stain with an antibody against intestinal vitamin D-dependent calcium binding protein. Brain Res 222:452– 457. 92. Feldman SC, Christakos S 1983 Vitamin D–dependent calcium binding protein in rat brain: Biochemical and immunocytochemical characterization. Endocrinology 112:290–302. 93. Barski JJ, Hartmann J, Rose CR, Hoebeek F, Morl K, NollHussong M, De Zeeuw CI, Konnerth A, Meyer M 2003 Calbindin in cerebellar Purkinje cells is a critical determinant of the precision of motor coordination. J Neurosci 23:3469–3477. 94. Rabie A, Thomasset M, Legrand C 1983 Immunocytochemical detection of a calcium binding protein in the cochlear and vestibular hair cells of the rat. Cell Tissue Res 232:691–696. 95. Legrand C, Brehier A, Clavel MC, Thomasset M, Rabie A 1988 Cholecalcin (28 kDa CaBP) in rat cochlea: Development in normal and hypothyroid animals: An immunocytochemical study. Brain Res 466:121–129. 96. Oberholtzer JC, Buettger C, Summers MC, Matschinsky FM 1988 The 28 kDa calbindin-D is a major calcium binding protein in the basilar papilla of the chick. Proc Natl Acad Sci USA 85:3387–3390. 97. Rabie A, Thomasset M, Parkes CO, Clavel MC 1985 Immunocytochemical detection of 28,000 MW calcium binding protein in horizontal cells of the rat retina. Cell Tissue Res 240:493–496. 98. Schreiner DS, Jande SS, Lawson DEM 1985 Target cells of vitamin D in the retina. Acta Anat 121:153–162. 99. Haley TL, Pochet R, Baizer L, Burton MD, Crabb JW, Parmentier M, Polans AS 1995 Calbindin-D28K immunoreactivity of human cone cells varies with retinal position. Vis Neurosci 12:301–307. 100. Roman A, Brisson P, Pasteels B, Demol S, Pochet R, Collen JP 1988 Pineal–retinal molecular relationships: Immunocytochemical evidence of calbindin-D28K in pineal transducers. Brain Res 442:33–42. 101. Blatow M, Caputi A, Burnashev N, Monyer H, Rozov A 2003 Ca2+ buffer saturation underlies paired pulse facilitation in calbindin-D28K-containing terminals. Neuron 38:79–88. 102. Mattson MP, Rychlick B, Chu C, Christakos S 1991 Evidence for calcium reducing and excitoprotective roles for the calcium binding protein calbindin-D28K in cultured hippocampal neurons. Neuron 6:41–51. 103. Chard PS, Bleakman D, Christakos S, Fullmer CS, Miller RJ 1993 Calcium buffering properties of calbindin-D28K and parvalbumin in rat sensory neurones. J Physiol 472:341–357. 104. Chard PS, Jordan J, Marcuccilli CJ, Miller RJ, Leiden JM, Roos RP, Ghadge GD 1995 Regulation of excitatory transmission at hippocampal synapses by calbindin-D28K. Proc Natl Acad Sci USA 92:5144–5148. 105. Li Z, Decavel C, Hatton GI 1995 Calbindin-D28K: Role in determining intrinsically generated firing patterns in rat supraoptic neurones. J Physiol 4883:601–608. 106. Burke RE, Baimbridge KG 1993 Relative loss of the striatal striosome compartment, defined by calbindin-D28K immunostaining, following developmental hypoxic-ischemic injury. Neuroscience 56:305–315.

733

107. Baimbridge KG, Mody I, Miller JJ 1985 Reduction of rat hippocampal calcium binding protein following commissural, amygdala, septal, perforant path and olfactory bulb kindling. Epilepsia 26:460–465. 108. Sonnenberg J, Frantz GD, Lee S, Heick A, Chu C, Tobin AJ, Christakos S 1991 Calcium binding protein (calbindin-D28K) and glutamate decarboxylase gene expression after kindling induced seizures. Mol Brain Res 9:179–190. 109. Iacopino AM, Christakos S 1990 Specific reduction of neuronal calcium binding protein (calbindin-D28K) gene expression in aging and neurodegenerative diseases. Proc Natl Acad Sci USA 87:4078–4082. 110. Ferrer I, Tunon T, Soriano E, Del Rio A, Iraizoz I, Fonseca M, Guionnet N 1993 Calbindin-D28K immunoreactivity in the temporal neocortex in patients with Alzheimer’s disease. Clin Neuropathol 12:53–58. 111. Geula C, Bu J, Nagykery N, Scinto LF, Chan J, Joseph J, Parker R, Wu CK 2003 Loss of calbindin-D28K from aging human cholinergic basal forebrain: relation to neuronal loss. J Comp Neurol 455:249–259. 112. Meier TJ, Ho DY, Sapolsky RM 1997 Increased expression of calbindin D28K via herpes simplex virus amplicon vector decreases calcium ion mobilization and enhances neuronal survival after hypoglycemic challenge. J Neurochem 69:1039–1047. 113. Ho BK, Alexianu ME, Colom LV, Mohamed AH, Serrano F, Appel SH 1996 Expression of calbindin-D28K in motoneuron hybrid cells after retroviral infection with calbindin-D28K cDNA prevents amyotrophic lateral sclerosis IgG-mediated cytotoxicity. Proc Natl Acad Sci USA 93:6796–6801. 114. Guo Q, Christakos S, Robinson N, Mattson MP 1998 Calbindin-D28K blocks the proapoptotic actions of mutant presenilin 1: reduced oxidative stress and preserved mitochondrial function. Proc Natl Acad Sci USA 95:3227–3232. 115. Liu AY, Abraham BA 1991 Subtractive cloning of a hybrid human endogenous retrovirus and calbindin gene in the prostate cell line PC3. Cancer Res 51:4104–4110. 116. Dowd DR, MacDonald PN, Komm BS, Haussler MR, Miesfeld RL 1992 Stable expression of the calbindin-D28K complementary DNA interferes with the apoptotic pathway in lymphocytes. Endocrinology 6:1843–1848. 117. Reisner PD, Christakos S, Vanaman TC 1992 In vitro enzyme activation with calbindin-D28K the vitamin D dependent 28kDa calcium binding protein. FASEB J 297:127–131. 118. Norman AW, Leathers V 1982 Preparation of a photoaffinity probe for the vitamin D dependent intestinal calcium binding protein: Evidence for a calcium-dependent, specific interaction with intestinal alkaline phosphatase. Biochem Biophys Res Commun 108:220–226. 119. Berggard T, Szczepankiewicz O, Thulin E, Linse S 2002 Myoinositol monophosphatase is an activated target of calbindin-D28K. J Biol Chem 277:41954–41959. 120. Pollock AS, Santiesteban HL 1995 Calbindin expression in renal tubular epithelial cells: Altered sodium phosphate cotransport in association with cytoskeletal rearrangement. J Biol Chem 270:16291–16301. 121. Minghetti PP, Cancela L, Fujisawa Y, Theofan G, Norman AW 1988 Molecular structure of the chicken vitamin D–induced calbindin-D28K gene reveals eleven exons, six Ca2+-binding domains, and numerous promoter regulatory elements. Mol Endocrinol 2:355–367. 122. Wilson PW, Rogers J, Harding M, Pohl V, Pattyn G, Lawson DEM 1988 Structure of chick chromosomal genes for calbindin and calretinin. J Mol Biol 200:615–625.

734 123. Parmentier M, De Vijlder JJM, Muir E, Szpirer C, Islam MQ, Geurts van Kessel A, Lawson DEM, Vassart G 1989 The human calbindin 27 kDa gene: Structure organization of the 5′ and 3′ regions, chromosomal assignment, and restriction fragment length polymorphism. Genomics 4:309–319. 124. Varghese S, Deaven LL, Huang Y-C, Gill RK, Iacopino AM, Christakos S 1989 Transcriptional regulation and chromosomal assignment of the mammalian calbindin-D28K gene. Mol Endocrinol 3:495–502. 125. Wood TL, Kobayashi Y, Franz G, Varghese S, Christakos S, Tobin AJ 1988 Molecular cloning of mammalian 28,000 Mr vitamin D dependent calcium binding protein (calbindin-D28K) in rodent kidney and brain. DNA 7:585–594. 126. Gill RK, Christakos S 1993 Identification of sequence elements in mouse calbindin-D28K gene that confer 1,25-dihydroxyvitamin D3- and butyrate-inducible responses. Proc Natl Acad Sci USA 90:2984–2988. 127. Christakos S, Norman AW 1980 Vitamin D–dependent calcium-binding protein synthesis by chick kidney and duodenal polysomes. Arch Biochem Biophys 203:809–815. 128. Pansini AR, Christakos S 1984 Vitamin D–dependent calcium binding protein in rat kidney: Purification and physicochemical and immunological characterization. J Biol Chem 259:9735–9741. 129. Varghese S, Lee S, Huang Y-C, Christakos S 1988 Analysis of rat vitamin D–dependent calbindin-D28K gene expression. J Biol Chem 263:9776–9784. 130. Minghetti PP, Gibbs PEM, Norman AW 1989 Computer analysis of 1,25-dihydroxyvitamin D3-receptor regulated promoters: Identification of a candidate D3-response element. Biochem Biophys Res Commun 162:869–875. 131. Maiyar AC, Minghetti PP, Norman AW 1991 Transfection of avian vitamin D–dependent calbindin-D28K 5′ flanking promoter sequence in primary chick kidney cells. Mol Cell Endocrinol 78:127–135. 132. Ferrari S, Battini R, Pike WJ 1990 Functional analysis of the promoter region of the gene encoding chicken calbindin-D28K. Adv Exp Med Biol 140:21–25. 133. MacDonald PN, Whitfield GK, Haussler CA, Hocker AM, Haussler MR 1992 Evaluation of a putative vitamin D response element in the avian calcium binding protein gene. DNA Cell Biol 11:377–383. 134. Theofan G, Nguyen AP, Norman AW 1986 Regulation of calbindin-D28K gene expression by 1,25-dihydroxyvitamin D3 is correlated to receptor occupancy. J Biol Chem 261: 16943–16947. 135. Enomoto H, Hendy GN, Andrews GK, Clemens TL 1992 Regulation of avian calbindin-D28K gene expression in primary chick kidney cells: Importance of posttranscriptional mechanisms and calcium ion concentration. Endocrinology 130: 3467–3474. 136. Meyer J, Galligan MA, Jones G, Komm BS, Haussler CA, Haussler MR 1995 l,25(OH)2D3-dependent regulation of calbindin-D28K mRNA requires ongoing protein synthesis in chick duodenal organ culture. J Cell Biochem 58:315–327. 137. Hall AK, Bishop JE, Norman AW 1987 Inhibitory and stimulatory effects of dexamethasone and 1,25-dihydroxyvitamin D3 on chick intestinal calbindin-D28K and its mRNA. Mol Cell Endocrinol 51:25–31. 138. Corradino RA, Fullmer C 1991 Positive cotranscriptional regulation of intestinal calbindin-D28K gene expression by 1,25-dihydroxyvitamin D3 and glucocorticoids. Endocrinology 128:944–950. 139. Theofan G, Kong MW, Hall AK, Norman AW 1987 Expression of calbindin-D28K mRNA as a function of altered

SYLVIA CHRISTAKOS

140. 141. 142. 143.

144.

145. 146. 147.

148. 149.

150.

151. 152. 153.

154.

155.

156.

ET AL .

serum calcium and phosphate levels in vitamin D-replete chick intestine. Mol Cell Endocrinol 54:135–140. Huang Y-C, Christakos S 1988 Modulation of rat calbindinD28K gene expression by 1,25-dihydroxyvitamin D3 and dietary alteration. Mol Endocrinol 2:928–935. Collazo D, Takahashi H, McKay RDG 1992 Cellular targets and trophic functions of neurotrophin-3 in the developing rat hippocampus. Neuron 9:643–656. Ip NY, Li Y, Yancopoulos GD, Lindsay RM 1993 Cultured hippocampal neurons show responses to BDNF, NT-3 and NT-4 but not NGF. J Neurosci 13:3394–3405. Cheng B, Christakos S, Mattson MP 1994 Tumor necrosis factors protect neurons against excitotoxic/metabolic insults and promote maintenance of calcium homeostasis. Neuron 12:139–153. Mattson MP, Cheng B, Baldwin SA, Smith-Swintosky VL, Keller J, Geddes JW, Scheff SW, Christakos S 1995 Brain injury and tumor necrosis factors induce calbindin-D28K in astrocytes: Evidence for a cytoprotective response. J Neurosci Res 42:357–370. Lindvall O, Kokaia Z, Bengzon J, Elmer E, Kokaia M 1994 Neurotrophins and brain insults. Trends Neurosci 17:490–496. Iacopino AM, Christakos S 1990 Corticosterone regulates calbindin-D28K mRNA and protein levels in rat hippocampus. J Biol Chem 265:10177–10180. Krugers HJ, Medema RM, Postema F, Korf J 1995 Regionspecific alterations of calbindin-D28K immunoreactivity in the rat hippocampus following adrenalectomy and corticosterone treatment. Brain Res 696:89–96. Wang Y-Z, Christakos S 1995 Retinoic acid regulates the expression of the calcium binding protein, calbindin-D28K. Mol Endocrinol 9:1510-1521. Nieto-Bona MP, Busiguina S, Torres-Aleman I 1995 Insulinlike growth factor I is an afferent trophic signal that modulates calbindin-D28K in adult Purkinje cells. J Neurosci Res 42:371–376. Yamaguchi T, Keino K, Fukuda J 1995 The effect of insulin and insulin-like growth factor-1 on the expression of calretinin and calbindin-D28K in rat embryonic neurons in culture. Neurochem Int 26:255–262. Pavlou O, Ehlenfeldt R, Horn S, Orr HT 1996 Isolation, characterization and in vivo analysis of the murine calbindin-D28K upstream regulatory region. Mol Brain Res 36:268–279. Gill RK, Christakos S 1995 Regulation by estrogen through the 5′-flanking region of the mouse calbindin-D28K gene. Mol Endocrinol 9:319–326. Perret C, Lomri N, Gouhier N, Auffray C, Thomasset M 1988 The rat vitamin-D-dependent calcium-binding protein (9-kDa CaBP) gene. Complete nucleotide sequence and structural organization. Eur J Biochem 172:43–51. Dupret JM, Brun P, Perret C, Lomri N, Thomasset M, Cuisinier-Gleizes P 1987 Transcriptional and post-transcriptional regulation of vitamin D–dependent calcium-binding protein gene expression in the rat duodenum by 1,25-dihydroxycholecalciferol. J Biol Chem 262:16553–16557. Colnot S, Ovejero C, Romagnolo B, Porteu A, Lacourte P, Thomasset M, Perret C 2000 Transgenic analysis of the response of the rat calbindin-D9K gene to vitamin D. Endocrinology 141:2301–2308. Lambert M, Colnot S, Suh E, L’Horset F, Blin C, Calliot ME, Raymondjean M, Thomasset M, Traber PG, Perret C 1996 Cis-acting elements and transcription factors involved in the intestinal specific expression of the rat calbindin-D9K gene: Binding of the intestine-specific transcription factor Cdx-2 to the TATA box. Eur J Biochem 236:778–788.

CHAPTER 42 The Calbindins: Calbindin-D9K and Calbindin-D28K 157. Colnot S, Romagnolo B, Lambert M, Cluzeaud F, Porteu A, Vandewalle A, Thomasset M, Kahn A, Perret C 1998 Intestinal expression of the calbindin-D9K gene in transgenic mice. Requirement for a Cdx2-binding site in a distal activator region. J Biol Chem 273:31939–31946. 158. Huang YC, Lee S, Stolz R, Gabrielides C, Pansini-Porta A, Bruns ME, Bruns DE, Miffin TE, Pike JW, Christakos S 1989 Effect of hormones and development on the expression of the rat 1,25-dihydroxyvitamin D3 receptor gene. Comparison with calbindin gene expression. J Biol Chem 264: 17454–17461. 159. Li H, Christakos S 1991 Differential regulation by 1,25-dihydroxyvitamin D3 of calbindin-D9K and calbindin-D28K gene expression in mouse kidney. Endocrinology 128:2844–2852. 160. Darwish H, Krisinger J, Furlow JD, Smith C, Murdoch FE, DeLuca HF 1991 An estrogen-responsive element mediates

735

the transcriptional regulation of calbindin-D9K gene in rat uterus. J Biol Chem 266:551–558. 161. L’Horset F, Blin C, Colnot S, Lambert M, Thomasset M, Perret C 1994 Calbindin-D9K gene expression in the uterus: Study of the two messenger ribonucleic acid species and analysis of an imperfect estrogen-responsive element. Endocrinology 134:11–18. 162. Romagnolo B, Cluzeaud F, Lambert M, Colnot S, Porteu A, Molina T, Tomasset M, Vandewalle A, Kahn A, Perret C 1996 Tissue-specific and hormonal regulation of calbindin-D9K fusion genes in transgenic mice. J Biol Chem 271: 16820–16826. 163. Romagnolo B, Molina T, Leroy G, Blin C, Porteux A, Thomasset M, Vandewalle A, Kahn A, Perret C 1996 Estradiol-dependent uterine leiomyomas in transgenic mice. J Clin Invest 98:777–784.

CHAPTER 43

Target Genes: PTHrP DAVID GOLTZMAN AND RICHARD KREMER Department of Medicine, McGill University and McGill University Health Center, Montre´al, Que´bec, Canada

I. II. III. IV.

Introduction PTHrP Gene and Its Products Mechanism of Action of PTHrP Regulation of PTHrP Production

I. INTRODUCTION In this chapter, we first discuss the structure of the parathyroid hormone–related peptide (PTHrP) gene and its products. We next examine the mechanism of action of PTHrP through its endocrine, autocrine, and intracrine pathways. We then describe important biological properties of PTHrP relevant to the development and progression of human cancer. Finally, based on the known biology of PTHrP, we outline the importance of using preclinical animal models to test potential strategies for blocking PTHrP production in cancer with particular emphasis on vitamin D and vitamin D analogs as therapeutic agents.

II. PTHRP GENE AND ITS PRODUCTS The gene encoding human PTHrP is a complex transcriptional unit of approximately 15 kb encoding three potential isoforms through alternative splicing of 139, 141, and 173 amino acids [1] (Fig. 1A). At the amino-terminal end, PTHrP shares a strong homology with parathyroid hormone (PTH) within the first 13 amino acids. Furthermore, human PTH and PTHrP genes are located at similar positions of chromosome 11 and 12, respectively, suggesting that these two genes arose through duplication from a common ancestral gene. There are at least seven exons with exon1 subdivided into Ia, Ib, and Ic [2]. Exon II encodes different 5′ untranslated regions. Exon III and IV encode the preprocoding region and the mature peptide up to amino acid 139, respectively. Exons V, VI, and VII encode the C-terminal amino acids, a stop codon, and 3′ noncoding regions. In contrast, the rat, mouse, and chicken [3–5] PTHrP genes have a much simpler organization with four or five exons encoding a single isoform of VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Therapeutic Strategies to Inhibit PTHrP Production VI. Summary References

141 amino acid. The human gene is regulated by three promoters in the 5′ region [6–8] and potentially by mRNA stability in the 3′ AUUUA-rich untranslated region [9]. PTHrP gene expression is positively or negatively regulated by several factors in its 5′ promoter region [10]. Several growth factors including epidermal growth factor and insulin-like growth factor 1 (IGF1) are stimulatory [10,11], whereas 1,25 dihydroxyvitamin D3 (1,25(OH)2D3) and dexamethasone down-regulate PTHrP expression and secretion in vitro [10,12,13]. The immature forms of PTHrP, the prepro forms, are extended at the amino terminus by 36 amino acids that contains a signal sequence necessary to direct the nascent peptide from the cytosol to the endoplasmic reticulum (ER). This signal sequence is likely cleaved in the ER by signal peptidases. The propeptide is then directed to the Golgi apparatus where it is cleaved and the mature form of PTHrP then stored in secretory granules. These three isoforms are identical up to amino acid 139 but then diverge to encode a unique C-terminal region [14]. The high homology of PTHrP and PTH at the amino-terminal end is responsible for its action on calcium homeostasis including bone resorption, renal calcium reabsorption, and renal phosphate excretion [15]. The first two amino acids are critical for adenylate cyclase stimulation [14]. Although not similar in their primary sequence to PTH, amino acids 14 to 34 are also critical for binding to the PTH/PTHrP receptor [16], suggesting that the tertiary structure contributes to hormone binding. The sequence beyond amino acid 34 is not required for PTHrP activity on calcium homeostasis, but is believed to play a role in other important functions such as growth and differentiation. Indeed, amino acids 35–111 are highly conserved among species, an indication that this region plays a critical role in “other” functions of PTHrP. Beyond amino acid 111, the sequences diverge among species, Copyright © 2005, Elsevier, Inc. All rights reserved.

738

DAVID GOLTZMAN AND RICHARD KREMER

A

Organization of the human PTHrP gene Ia

Ib

Ic

II

III

−36

B

IV

1

V

139

VI

173

VII

141

Structural function relationship Pre-Pro

−36

PTH-Like

1

34

Placental Ca++ transport

37

86

NLS

87 106

Osteostatin

107

139

Unknown function

141

173

FIGURE 1 Organization of the PTHrP gene and structure–function relationship. (A) The boxes represent the exons with exon I subdivided into Ia, Ib, Ic. Hatched boxes represent 5′ untranslated (5′UTR) regions. Solid boxes represent coding sequences. The open boxes represent 3′ untranslated (3′UTR) regions. The solid lines represent alternative splicing. (B) The proposed biologically active domains of human PTHrP are represented. All of PTH-like bioactivity is contained in PTHrP1-34. PTHrP37-86 is thought to play a role in placental calcium transport. The PTHrP nuclear/nucleolar localization sequence (NLS) lies between amino acids 87 and 106. A sequence between amino acids 107 and 137 also called osteostatin is believed to be anabolic for osteoblasts and inhibitory for osteoclasts.

and the human isoform contains a unique sequence from amino acids 141 to 173 whose function has not yet been elucidated. Posttranslational processing also contributes to the generation of additional PTHrP fragments (Fig. 1B) with distinct functions. This processing yields an amino terminal fragment PTHrP 1-36 which can bind the PTH/PTHrP receptor, a mid region fragment (residues 38–86), [17], which plays a role in placental calcium transport [18] and a carboxyterminal fragment also called osteostatin (residues 107–139) which inhibits osteoclastic bone resorption [19] and stimulates osteoblastic bone formation [20]. In addition the human form of PTHrP can be O-glycosylated at the amino terminus [21] and cleaved off at the carboxy terminus to produce fragments [22,23] that accumulate in renal failure.

III. MECHANISM OF ACTION OF PTHRP Both PTH and PTHrP bind to a common seven transmembrane spanning receptor that is linked through G proteins to both adenylate cyclase and phospholipase-C signaling pathways [24]. In addition to the seven transmembrane domains, the PTH/PTHrP receptor contains an extracellular domain of around 180 amino acids and an intracellular carboxyterminal tail of approximately 120 amino acids [25]. The overall organization of the receptor, although similar to those of other G protein–coupled receptors, has a unique gene organization and no sequence homology to other G protein–coupled receptors [26,27]. In addition, the

receptor is able to activate the MAP kinase pathway in a cell-specific fashion [28]. However, similarly to other peptides that bind to cell surface receptors that act through a classical endocrine/paracrine mechanism, it has been proposed that PTHrP can also target the nucleus of the cell using an intracrine signaling pathway [29]. Indeed, the sequence spanning amino acids 88–106 has structural homology with previously described nuclear and nucleolar localization sequences (NLS) [30,31]. Removal of this putative nuclear/nucleolar sequence abolishes intranuclear localization of the prepro PTHrP cDNA transiently expressed in COS-7 cells [29]. Subsequently this nuclear/nucleolar localization has been reported in osteoblasts [29], keratinocytes [32], vascular smooth muscle [33], pancreatic adenocarcinoma [34], melanoma [35], and breast cancer [36]. PTHrP was indeed localized over the dense fibrillar component of the nucleolus [29], suggesting that it could modulate specific nucleolar functions such as transcription/processing of rRNA or ribonucleoprotein complex formation prior to their transport to the cytoplasm. In contrast to this specific localization in COS-7 cells and osteoblasts, PTHrP localization is more diffuse to the nuclear/nucleolar compartments in vascular smooth muscle cells [33] and could therefore regulate other nuclear actions in a cell-specific fashion. It is tempting to speculate that PTHrP localization to the nuclear/nucleolus has specific functions independent of the paracrine/autocrine effects that are secondary to the PTH/PTHrP receptor activation. This hypothesis has been confirmed in at least one cell model [37]. The mechanism by which

739

CHAPTER 43 Target Genes: PTHrP

PTHrP is directed to the nucleus/nucleolus is still elusive and probably not unique. It is presumed that in order to exert its endocrine/paracrine/autocrine function, PTHrP should be internalized, as is the case for other peptides that have receptor-independent nuclear actions such as insulin, prolactin, and epidermal growth factor (EGF) [38]. This hypothesis is supported by studies demonstrating receptor-mediated internalization of PTHrP in cells harboring the osteoblast phenotype [39,40]. However, it has also been reported that PTHrP can be internalized in cells that do not express the classical PTH/PTHrP receptor [41], indicating that an as-yet-unidentified PTHrP receptor recognizing the specific nuclear/nucleolar sequence could be responsible for PTHrP internalization and nuclear/nucleolar targeting. Other possible mechanisms of PTHrP targeting to the nucleus in PTHrP producing cells is the intracrine route that does not utilize receptor/ligand internalization. In this case, PTHrP could be diverted from its secretory pathway from the ER and back to the cytosol by a process called retrograde translocation [42]. Although there is no direct evidence that PTHrP could utilize this pathway to enter the nucleus, indirect evidence indicates that PTHrP is a substrate of the ubiquitin and proteasome-mediated proteolytic system [43] that diverts PTHrP from the ER to the cytosol. However, it has not yet been established that PTHrP could then escape ubiquitination and proteosomal degradation to be redirected to the nucleus. As for other peptides, the nuclear uptake of PTHrP from the cytosol is likely regulated by the cell cycle [44], although only indirect proof of this mechanism has so far been established [32,33]. It has also been suggested that PTHrP can serve as a substrate for CDK2-CDC2 kinase [29], since a consensus motif for CDK2-CDC2 is located immediately upstream of the nuclear/nucleolar localization sequence. Phosphorylation of the NLS can indeed influence nuclear import of other peptides [30] and therefore could potentially modulate PTHrP nuclear import. Importins α and β, also known as kariopherins α and β, play a major role in the nuclear import process of NLS-containing proteins [38]. In the case of PTHrP and in contrast to the majority of other NLS-containing proteins, importin β and not importin α is responsible for nuclear translocation [39]. Indeed, a sequence mapped to amino acids 89–93 in the PHTrP sequence was found to be responsible for interacting with importin β, and residues 380–643 of importin β mediated this interaction. Once translocated to the nucleus/nucleolus compartment, the biological function of PTHrP at these sites remains elusive, but its accumulation in the dense fibrillar component of the nucleolus suggests that it would modulate ribosomal gene transcription or ribosomal synthesis. Additionally, it could affect other nuclear

functions such as DNA replication and/or gene transcription when translocated to the nucleus.

IV. REGULATION OF PTHRP PRODUCTION Traditionally, PTHrP has been regarded as a paraneoplastic hypercalcemic mediator produced by many types of cancer, especially solid tumors [45]. PTHrP is now known to be produced not only by cancer cells but by a wide variety of normal cells (Table I). It has been TABLE I PTHrP Expression in Normal and Cancer Cells and Tissues Normal cells/tissues Adrenal cortex Amnion Aortic smooth muscle Bladder Bone Brain Bronchial epithelium Cardiac muscle Cervical cells Chondrocytes Endometrium Endothelium Epidermis Gonads Kidney Keratinocytes Liver Lung Mammary epithelium Melanocytes Ovaries Pancreatic islet cells Parathyroid glands Placenta Prostate Salivary ducts Small intestine Spleen Stomach mucosa Thyroid Thymus Urothelium Uterus

Cancer cells/tissues Breast Bladder Colon Embryonal carcinoma HTLVI-infected T cells Insulinoma Lung Medullary carcinoma Melanoma Multiple myeloma Oral Osteosarcoma Ovary Prostate Renal Squamous carcinoma Testicular

740

DAVID GOLTZMAN AND RICHARD KREMER

(Fig. 2) and end-stage disease in non-Hodgkin’s lymphoma [53]. One of the major features of PTHrP regulation is its positive regulation by mitogenic stimuli and cytokines such as epidermal growth factor (EGF), insulin-like growth factor I (IGF-1), and TGFβ [10,11,54–56] and its inhibition by 1,25-dihydroxyvitamin D3 [10,57–61]. Additionally, studies have shown that the PTHrP promoter region contains both growth factor responsive sequences and a vitamin D–responsive element (VDRE) [10,62–64]. A variety of other factors have also been shown to be either stimulatory or

proposed that PTHrP functions predominantly as an autocrine/paracrine factor in normal tissues such as keratinocytes [46]. Furthermore, PTHrP produced by cancer cells exhibits growth factor–like properties that may enhance the growth of tumor cells locally in an autocrine fashion [47–49] and subsequently favor tumor progression [50]. Indeed, indirect clinical evidence indicates that PTHrP enhanced the capacity of cancer cells to invade bone [47,51] and that the presence of detectable PTHrP levels in hypercalcemic cancer patients is associated with shorter survival [52]

A

B 1.00

1.00

Pretreatment calcium and PTHrp PTHrP 0, CA ≤ 12 PTHrP > 0 PTHrP 0, CA > 12 PTHrP > 0

10/25 9/16 9/10 22/25

0.75 Proportion surviving

Proportion surviving

0.75

0.50

0.25

Pretreament calcium, PTHrP for Age ≤ 65 PTHrP 0, CA ≤ 12 PTHrP > 0 PTHrP 0, CA > 12 PTHrP > 0

0.50

3/14 6/10 5/6 14/16

0.25

0.00

0.00 0

1

2

3

0

1

Years

C

2

3

Years

1.00 PTHrP and age group PTHrP 0, Age ≤ 65 PTHrP > 0 PTHrP 0, Age > 65 PTHrP > 0

Proportion surviving

0.75

8/20 20/26 11/15 11/15

0.50

0.25

0.00 0

1

2

3

Years

FIGURE 2

(A) Survival in 76 hypercalcemic cancer patients by parathyroid hormone–related peptide (PTHrP) status and pretreatment calcium levels. Numbers shown in the inset are total number of deaths/number of patients at baseline. Number of patients at risk were 40 at 100 days, 22 at 1 year, and 3 at 3 years. (B) Survival in 46 hypercalcemic cancer patients ≤ 65 years old of PTHrP status and pretreatment calcium levels. Number of patients at risk were 26 at 100 days, 13 at 1 year, and 1 at 3 years. (C) Survival in hypercalcemic cancer patients by PTHrP status and age group. Number of patients at risk were 41 at 100 days, 22 at 1 year, and 3 at 3 years. CA ≤ 12, pretreatment serum calcium levels 10.3 to 12 mg/dl; CA > 12, pretreatment serum calcium levels > 12 mg/dl; PTHrP 0, PTHrP not elevated; PTHrP > 0, PTHrP elevated [52].

741

CHAPTER 43 Target Genes: PTHrP

TABLE II Stimulators Serum

Growth factors Epidermal growth factor

IGFI TGFβ IL-6 IL-1β TNFα Ras-RafMAP-Kinase signaling

Calcium Estrogens

Prolactin Calcitonin All-transretinoic acid Phorbol ester Endothelin I

Stimulators and Inhibitors of PTHrP Expression/Production in Normal and Cancer Cells and Tissues Cell/tissue type Human keratinocytes [10] Rat aortic smooth muscle [65] Human squamous carcinoma [55] Rat islet cell line [66] Human squamous carcinoma [54] Human squamous carcinoma [60] Human keratinocytes [10] Human cervical epithelial cells [67] Human mammary epithelial cells [11] Human osteosarcoma cells [70] Human squamous carcinoma [55] Human keratinocyte cell line HaCaT [75] Human mammary epithelial cells [11] Human lung carcinoma [76] Human squamous carcinoma [55] Human uterine cells [56] Human lung carcinoma [76] Human lung carcinoma [76] Human lung carcinoma [76]

Inhibitors 1,25(OH)2D3

EB1089 [1,25(OH)2D3 analog]

22-oxa-1,25(OH)2D3 [1,25(OH)2D3 analog] 9-cis-retinoic acid Glucocorticoids Testosterone Chromogranin A Ras-signaling inhibitors (B1086 and lovastatin)

Cell/tissue type Human medullary carcinoma [57] Human keratinocytes [10] Human keratinocytes [58] Human oral cancer cells [59] Human squamous carcinoma [60] Human T cells transfected with HTLV1 [61] Human melanocytes and melanoma cells [106] Human squamous carcinoma [100] Human squamous carcinoma [60] Human melanoma [106] Human lung cancer cells [64] Human T cells transfected with HTVL1 [61] Human lung cancer cells [64] Human oral cancer cells [59] Human medullary carcinoma [57] Human neuroendocrine cells [71] Rat testicular tumor [13] Human squamous carcinoma [72] Rat Leydig H500 tumor [95]

Human vascular endothelial cells [73] Rat 3T3 fibroblasts transfected with TRP-MET [78] Ras transformed human prostate epithelial cells [77] Human keratinocytes [10] Human squamous carcinoma [54] Rat uterus [105] Rat kidney [68] MCF-7 breast cancer cells [74] Rat mammary gland [107] Human squamous carcinoma [69] Human squamous carcinoma [72] Human cervical epithelial cells [67] Human osteosarcoma cells [70] Rat aortic smooth muscle [65]

inhibitory for PTHrP as indicated in Table II. Since growth factors are generally positive regulators for PTHrP production, it should not be surprising that cancer cells may overexpress PTHrP following activation of the growth factor signaling pathways. The ras-rafMAPKinase activation cascade is activated by EGF and can lead to PTHrP overproduction when constitutively

expressed in cancer cells [77,78]. Additional mechanisms have been proposed to explain PTHrP overexpression in cancer tissues and include silencing by demethylation of specific regulatory sequences of the PTHrP gene [79], gene amplification [80], and resistance to PTHrP-induced inhibition by 1,25(OH)2D3 [54]. This resistance to 1,25(OH)2D3 action was shown to be

742

DAVID GOLTZMAN AND RICHARD KREMER

V. THERAPEUTIC STRATEGIES TO INHIBIT PTHRP PRODUCTION

secondary to phosphorylation of the vitamin D receptor (VDR) heterodimeric partner retinoid X receptor (RXR) at a specific MAP kinase consensus site or SER 160 [81]. Overall, several mechanisms may in theory lead to overproduction of PTHrP by tumor cells, allowing subsequent release into the systemic circulation or enhanced local autocrine/paracrine action. Based on these observations, a number of strategies could be devised to block PTHrP production by cancer cells. A primary target for such strategies is breast cancer, since over 60% of primary breast cancer and over 90% of metastatic breast cancer lesions have been reported to overexpress PTHrP [82–87]. The higher percentage of metastatic lesions overexpressing PTHrP may be explained by a favorable bone microenvironment which produces cytokines such as TGFβ, a known stimulator of PTHrP production [55,56]. Targeting PTHrP in breast cancer has therefore the potential to inhibit mammary tumor growth and its devastating skeletal complications.

Based on our current knowledge of PTHrP biology, several strategies may be devised to block PTHrP production or action by cancer cells. These include PTHrP ablation by immunotherapy or antisense strategies, PTHrP inhibition by altering growth factor production/ signaling or vitamin D therapy (Fig. 3).

A. Immunotherapy Neutralization of the PTHrP molecule using specific antibodies is particularly attractive since it has the potential to work systemically as well as locally to block PTHrP production and action. Previous studies have shown that this strategy worked in an animal model of osteolytic metastases associated with breast cancer [88].

25-OHD3

1,25-(OH)2D3 and analogs

1α-hydroxylase

1,25(OH)2D3

RXR

VDR

RAS

−

VDRE

PTHrP mRNA

PTHrP GFRE GF

MAPK

+ PrePro PTHrP

PTHrP

Antibody

FIGURE 3

Regulation of PTHrP expression/production and potential targets for PTHrP inhibition. Positive regulators of PTHrP include growth factors that can utilize the ras-raf-mitogen-activated protein kinase (MAPK) pathway. Negative regulators include 1,25(OH)2D3 and its analogs. A reduction in PTHrP expression/production is possible through inactivation of the growth factor-MAPK pathway and/or through activation of the VDR/RXR complex by 1,25(OH)2D3, its analogs, or the inactive 1,25(OH)2D3 precursor 25OHD3. Additional strategies could include PTHrP antibody administration (immunotherapy) and gene expression knockout technology (antisense therapy).

743

CHAPTER 43 Target Genes: PTHrP

Mice treated with a PTHrP antibody raised against PTHrP1-34 had fewer osteolytic lesions than control animals. Furthermore, histomorphometric analysis of long bones revealed that this strategy reduced both the number of active osteoclasts at the tumor–bone interface and the tumor burden within bone. These beneficial effects were observed even after the establishment of bone metastatic lesions [89]. In this animal model, osteolytic lesions were not associated with hypercalcemia or detectable PTHrP fragments in the circulation, indicating that the antibody likely had a neutralizing effect by preventing the binding of PTHrP molecules to the PTH/PTHrP receptor within the bone microenvironment. Additional studies indicate that this strategy may also be effective in animal models of PTHrP-induced hypercalcemia demonstrating that systemic infusion of PTHrP antibodies can effectively normalize serum calcium levels [90,91]. These authors used a humanized antibody constructed using the complementary-determining region grafting method. They showed that this humanized antibody inhibited PTHrP-induced cAMP production in rat osteosarcoma ROS17/28-5 cells in vitro and bone resorption markers in vivo. Furthermore, administration of a single dose of this antibody was more effective than bisphosphonates in correcting hypercalcemia and maintaining normal serum calcium levels thereafter. In addition, administration of the humanized antibody but not of bisphosphonates had anticachectic properties and prolonged survival of these animals, suggesting that PTHrP inhibition had additional properties independent of its antihypercalcemic effect. However, the usefulness of this approach in cancer patients remains to be established.

undergo down-regulation of gene expression over time [94]. Consequently, this approach for gene therapy experiments may not yield sufficient effects on gene expression over time to sustain the desired biological effects. In another study, PTHrP inhibition was achieved using a nonretroviral antisense RNA approach and resulted in normalization of serum calcium levels and inhibition of tumor growth in an animal model of malignancy-associated hypercalcemia, the H500 rat Leydig cell tumor model [49]. Despite the potential difficulties in future gene therapy experiments, the antisense RNA approach has the additional advantage over immunotherapy to block both the secretory pathway and the intracrine PTHrP signaling pathway.

C. Inhibition of PTHrP by Interfering with Growth Factor Signaling As indicated earlier, a number of growth factors and cytokines positively regulate PTHrP production (see Table II). Studies on the molecular mechanism of PTHrP induction by growth factors indicate that ras signaling is critical for this effect [77,78]. Constitutive expression of receptor tyrosine kinases and p21ras isoprenylation significantly enhanced PTHrP production indicating that targeting receptor tyrosine kinases and/or ras signaling is a possible strategy to block PTHrP production [78]. Interestingly, lovastatin, a commonly used hypolipidemic agent, is able to block p21 isoprenylation at the inner surface of the plasma membrane and consequently inactivate ras-signaling. In these studies, it was demonstrated that lovastatin blocks PTHrP production [95] in cancer cells in vitro. It has not yet been demonstrated whether this strategy works in vivo.

B. Antisense RNA Technology D. Vitamin D Therapy Antisense RNA inhibits target mRNA sequences by hybridizing and consequently interfering through one of several mechanisms, with the function of the targeted mRNA. This ultimately results in reduced translation of the gene product [92]. This strategy was applied previously to block PTHrP production using two different approaches [46,49]. In one study, a PTHrP cDNA cloned in an antisense orientation in a replication defective retroviral vector [46] was used and resulted in the complete inhibition of PTHrP production in a human keratinocyte cell line. The retrovector used contains the backbone of Moloney murine leukemia virus long terminal repeat promoter (MoMLV-LTR) [93]. However, the MoMLV-LTR and other MoMLV-based vectors that are the most widely used for retroviral-mediated gene transfer experiments are not active in all cell types and

Since the discovery that 1,25(OH)2D3 inhibits PTHrP production in normal human epidermal keratinocytes [10], numerous studies have shown that 1,25(OH)2D3 is a strong inhibitor of PTHrP production in a wide variety of normal and cancer cells (Table II). It was further demonstrated that the mechanism of this inhibitory effect involves vitamin D–responsive elements in the 5′ promoter region of both the rat [10,62,96] and human genes [63]. It is interesting to note that the putative 1,25(OH)2D3 mediated repression sequence of the rat PTHrP promoter overlapped with the growth factor/serum mediated stimulatory region localized between 0.3 and 1.2 kb upstream of the transcriptional start site [10]. Furthermore, although the rat PTHrP VDRE interacts with the VDR/RXR complex like

744 other VDREs [62], the human PTHrP VDRE appears not to interact. The human PTHrP VDRE was shown to recognize VDR but not RXR [63], although studies in human oral cancer cells seem to indicate that suppression of PTHrP expression in this model is mediated through the VDR/RXR complex activation [59]. 1,25(OH)2D3 therapy may therefore represent an alternative option to block PTHrP production in cancer patients. However, since 1,25(OH)2D3 causes hypercalcemia and hypercalciuria at relatively low doses, its applications in malignancy-associated hypercalcemia may be limited. These difficulties could in theory be overcome by synthesizing new vitamin D analogs with lower calcemic activity compared to 1,25(OH)2D3 but with potent anti-tumor effect. Interestingly a number of these analogs were found to have selective properties on growth and differentiation [97–99] and in the hyperproliferative skin disorder psoriasis, due to their rapid metabolic degradation when applied locally to skin lesions. Other analogs with strong antiproliferative and prodifferentiative properties that are administered systemically have also been designed (see Chapters 80–88). One such analog, EB1089, was found to have very low calcemic potency relative to 1,25(OH)2D3 when infused into control animals [100–102]. In vivo, EB1089 is 10–100 times more potent than 1,25(OH)2D3 in inhibiting PTHrP production in cancer cells [60]. Subsequently this analog was used in two preclinical animal models of malignancy-associated hypercalcemia including the rat Leydig tumor H500 implanted into Fisher rats and a xenograft model in which nude mice were implanted with PTHrP-producing HPK1Aras cells. In these models, it was clearly established that EB1089 inhibited PTHrP production by the tumors and its release in the circulation. These actions in turn caused a reduction in hypercalcemia [100,101]. Interestingly, in the H500 Leydig tumor model EB1089 also prolonged survival, an added benefit that may be related to either its antihypercalcemic effect, its direct antitumor effect, or its indirect antitumor effect through PTHrP inhibition. Furthermore, EB1089 demonstrated both preventative and therapeutic potential by either preventing the development of hypercalcemia [100] or by decreasing blood calcium levels once hypercalcemia was achieved [101]. Both strategies represent two distinct but clinically relevant situations. In the first scenario, the analog is administered to a patient with an established diagnosis of cancer but prior to the development of hypercalcemia. In the second scenario, the analog is administered to a cancer patient presenting with hypercalcemia and usually an advanced stage of cancer. In addition to being a potential therapy in malignancyassociated hypercalcemia, vitamin D analogs may also

DAVID GOLTZMAN AND RICHARD KREMER

be useful in targeting PTHrP-producing tumors without hypercalcemia. Indeed, it has been suggested that PTHrP may play an important role in the establishment/ development of osteolytic bone metastasis associated with breast cancer [83,89]. A xenograft model of bone metastasis in which human MDA-MB-231 breast cancer cells are implanted into the left ventricle of nude mice [89,103] was used to test the efficacy of EB1089. The analog was administered in a preventative fashion at the time of tumor implantation [102]. The development of osteolytic bone metastasis was evaluated radiologically, histologically, and by histomorphometry. Kaplan Meier analysis demonstrated that bone lesions detected by radiographs progressed more slowly in EB1089-treated animals and that animal survival time was increased. Furthermore, tumor burden in bone was significantly reduced in EB1089-treated animals. Interestingly, PTHrP is expressed in this MDA-MB231 xenograft model and released by these cancer cells was previously shown to play a causal role in the development of osteolytic bone lesions [89]. Although not reported in the study of El Abdaimi et al. [102], it would be of major interest to know if EB1089 can reduce the production of PTHrP by metastatic tumors in bone in this model. An alternative strategy to block PTHrP production would be to use an inactive vitamin D derivative that could subsequently be metabolized by tumor tissues to active vitamin D metabolites. Such strategy has been used in a xenograft model of ras-transformed keratinocytes transplanted in nude mice [104]. In this study, both alleles of the human 1α-hydroxylase gene were inactivated by homologous recombination (double knockout) in cancer cells and the effect of the inactive precursor 25-hydroxyvitamin D3 (25OHD3) was tested in both wild-type and double knockout cancer cells. In vivo tumor growth was significantly reduced by 25OHD3 in nude mice transplanted with wild-type cells but not in animals transplanted with knockout cells, indicating that 1α-hydroxylase expression by cancer cells was necessary for the growth inhibitory effect of 25OHD3 to occur (Fig. 4). Furthermore, animals remained normocalcemic even when 25OHD3 was administered at doses 100 times higher than 1,25(OH)2D3. In a subsequent study, 25OHD3 was found to be highly effective in reducing PTHrP production by human melanoma cells following its conversion to 1,25(OH)2D3 [50]. In vivo strategies using inactive 1,25(OH)2D3 precursors may therefore be effective in controlling PTHrP overproduction while avoiding the undesired hypercalcemic complications. A potential drawback common to many cancer therapeutic strategies is the possibility of drug resistance. Indeed it has been demonstrated that ras-transformed keratinocytes producing PTHrP

745

CHAPTER 43 Target Genes: PTHrP

Tumor volume (mm3)

A

B

1600

WT (+/+)

1600

DKO (−/−)

1400

1400

Vehicle

1200

25 OHD3

Vehicle

1200

1000

1000

800

800

600

600

400

400

200

200

0

25 OHD3

0 0

10

20

30

40

50

60

0

10

20

30

40

50

60

Time (days)

C

D

Tumor weight (g)

1.5

1.5

WT (+/+)

1

1

0.5

0.5

0 Vehicle

25OHD3

DKO (−/−)

0 Vehicle

25OHD3

In vivo tumor growth kinetics in SCID mice following s.c. injection of 3 × 106 cells in PBS mixed with matrigel (1:1). 25-OHD3 (2000 pM 24 hr) was administered by constant infusion using Alzet osmotic minipumps. SCID mice that received implants of WT control HPK1Aras/pcDNA3 cells (A) or DKO HPK1Aras cells (B). 25-OHD3 administration significantly inhibit tumor growth in animals that received implants of WT-HPK1Aras cells (A and C), but has no effect in animals that received implants of DKO HPK1Aras cells (B and D). Seven and a half weeks after tumor implantation and treatment with 25-OHD3 or vehicle, mice were killed and the weight of tumors in animals implanted with WT (C) or DKO (D) was measured. Data are expressed as means of 12 mice in each group. This experiment was repeated twice. *, significantly different from vehicle-treated animals at each time point; P < 0.05. (From Huang DC, Papavasiliou V, Rhim JS, Horst RL, Kremer R 2002 Mol Cancer Res 1:1–12.)

FIGURE 4

are partially resistant to the inhibitory effect of 1,25(OH)2D3 [54]. It was subsequently demonstrated that ras-induced resistance to vitamin D was a consequence of the activation of the ras-raf-MAPKinase pathway resulting in phosphorylation of the retinoid X receptor on SER260 (a MAPKinase consensus sequence) [81]. Consequently, the efficacy of vitamin D

analogs may be substantially reduced in the commonly occurring ras-transformed tumors and strategies aimed at simultaneously inactivating the ras-raf-MAPKinase signaling pathways and administering the vitamin D analog may therefore be extremely useful. Similarly strategies aimed at inhibiting growth factor signaling may also be used in combination with vitamin D

746 analog therapy to target simultaneously the stimulatory and inhibitory pathways of PTHrP signaling.

VI. SUMMARY Our thorough understanding of PTHrP regulation by both positive and negative regulators has led to the evaluation of new therapeutic options to control the production of this peptide in cancer. Because of the complex nature of PTHrP regulation, it is plausible that combining strategies to target several steps of the regulatory pathway may become a viable therapeutic option in cancer therapy. Vitamin D signaling has emerged as an important potential tool in these strategies by using either more active tissue specific analogs or inactive precursors that could be converted to active forms within the target cancer cells. Additional efforts should be directed to find even more potent vitamin D analogs or precursors that could be tested alone or in combination with other strategies targeting a reduction of PTHrP production.

References 1. Martin TJ, Moseley J, Williams D 1997 Parathyroid hormonerelated protein: hormone and cytokine. J Endocrinol 154: 523–537. 2. Martin TJ, Moseley J, Gillespie MT 1991 Parathyroid hormonerelated protein: biochemistry and molecular biology. Crit Rev Biochem Mol Biol 26:377–395. 3. Karaplis AC, Yasuda J, Hendy GN, Goltzman D, Banville D 1990 Gene-encoding parathyroid hormone-like peptide: nucleotide sequence of the rat gene and comparison with the human homolog. Mol Endocrinol 4:441–446. 4. Mangin M, Ikeda K, Broadus AE 1990 Structure of the mouse gene encoding parathyroid hormone-related peptide. Gene 95:195–202. 5. Thiede MA, Rutledge SJ 1990 Nucleotide sequence of a parathyroid hormone-related peptide expressed by the 10-day chicken embryo. Nucleic Acids Res 18:3062. 6. Mangin M, Ikeda K, Dreyer BE, Broadus AE 1990 Identification of an upstream promoter of the human parathyroid hormone-related peptide gene. Mol Endocrinol 4:851–858. 7. Yasuda T, Banville, D, Hendy GN, Goltzman D 1989 Characterization of the human parathyroid hormone-like peptide gene: Functional and evolutionary aspects. J Biol Chem 13:7720–7725. 8. Vasavada RC, Wysolmerski JJ, Broadus AE, Philbrick WM 1993 Identification and characterization of a GC-rich promoter of the human PTHrP gene. Mol Endocrinol 7:273–282. 9. Moseley JM, Gillespie MT 1995 Parathyroid hormonerelated protein. Crit Rev Clin Lab Sciences 32:299–343. 10. Kremer R, Karaplis AC, Henderson J, Gulliver W, Banville D, Hendy GN, Goiltzman D 1991 Regulation of parathyroid hormone-like peptide in cultured normal human keratinocytes: effect of growth factors and 1,25 dihydroxyvitamin D3 gene expression and secretion. J Clin Invest 87:884–893.

DAVID GOLTZMAN AND RICHARD KREMER

11. Sebag M, Henderson JE, Goltzman D, Kremer R 1994 Regulation of parathyroid hormone-related peptide production in normal human mammary epithelial cells in vitro. Am J Physiol 267(3):C723–730. 12. Glatz JA, Heath JK, Southby J, L’Keefe LM, Kiriyama T, Moseley JM, Martin TJ, Gillespie MT 1994 Dexamethasone regulation of PTHrP expression in a squamous cancer cell line. Mol Cell Endocrinol 101:295–308. 13. Liu B, Goltzman D, Rabbani SA 1993 Regulation of parathyroid hormone related protein production in vitro by the rat hypercalcemic leydig tumor H500. Endocrinology 132: 1658–1664. 14. Orloff JJ, Reddy D, de Papp AE, Yang KH, Soifer NE, Stewart AF 1994 Parathyroid hormone–related protein as a prohormone. Post-translational processing and receptor interactions. Endocr Reviews 15:40–60. 15. Kemp BE, Moseley JM, Rodda CP, Ebeling PR, Wettenhall RE, Stapleton D, Diefenback-Jagger H, Ure F, Michelangeli VP, Simmons HA, et al. 1987 Parathyroid hormone-related protein of malignancy: active synthetic fragments. Science 238: 1568–1570. 16. Stewart AF, Broadus AE 1991 PTHrP: coming of age in the 1990s. J Clin Endocrinol Metab 71:1410–1414. 17. Soifer NE, Dee KE, Insogna KL, Burtis WJ, Matovcik LM, Wu TL, Milstone LM, Broadus AE, Philbrick WM, Stewart AF 1992 PTHrP: Evidence of secretion of a novel mid region fragment by three different cell lines in culture. J Biol Chem 267:18236–18243. 18. Kovacs CS, Lanske B, Hunzelman JL, Guo J, Karaplis AC, Kroninberg HM 1996 Parathyroid hormone related peptide (PTHrP) regulates fetal–placental calcium transport through a receptor distinct from the PTH/PTHrP receptor. Proc Natl Acad Sci USA 93:15233–15238. 19. Cornish J, Callon KE, Nicholson GC, Reid IR 1997 Parathyroid hormone-related protein (107–139) inhibits bone resorption in vivo. Endocrinology 138:1299–1304. 20. Cornish J, Callon KE, Lin C, Xiao C, Moseley JM, Reid IR 1999 Stimulation of osteoblast proliferation by C-terminal fragments of parathyroid hormone-related protein. J Bone Min Res 14:915–922. 21. Wu TL, Soifer NE, Burtis WJ, Milstone LM, Stewart AF 1991 Glycosylation of PTHrP secreted by human epidermal keratinocytes. J Clin Endocrinol Metab 73:1002–1007 22. Burtis WJ, Brady TG, Orloff JJ, Ersbak BJ, Warrell RP Jr, Olson BR, Wu TL, Mitnick ME, Broadus AE, Stewart AF 1990 Immunochemical characterization of circulating parathyroid hormone-related protein in patients with humoral hypercalcemia. N Engl J Med 322:1106–1112. 23. Orloff JJ, Soifer NE, Fodero JP, Dann P, Burtis WJ 1993 Accumulation of carboxyterminal fragments of PTHrP in renal failure. Kidney Int 43:1371–1376. 24. Chorev M, Rosenblatt M 1994 Structure–function analysis of parathyroid hormone and parathyroid hormone-related protein. In: Bilezekian JP, Marcus R, Levine MD (eds) The Parathyroids: Basic and Clinical Concepts. Raven Press, New York, pp. 139–116. 25. Jüppner H 1995 Functional properties of the PTH/PTHrP receptor. Bone (Suppl. 17):39S–42S. 26. Jüppner H 1994 Molecular cloning and characterization of a PTH/PTHrP receptor: a member of an ancient family of G protein coupled receptors. Curr Opin Nephrol Hypertens 3:371–378. 27. Abou-Samra AB, Juppner H, Force T, Freeman MW, Kong XF, Schipani E, Urena P, Richards J, Bonventre JV,

CHAPTER 43 Target Genes: PTHrP

28.

29.

30. 31. 32.

33.

34.

35.

36.

37.

38. 39.

40.

41.

Potts JT Jr, et al. 1992 Expression cloning of a common receptor for PTH and PTHrP from rat osteoblast-like cells: A single receptor stimulates intracellular accumulation of both camp and inositol triphosphates and increases intracellular free calcium. Proc Natl Acad Sci USA 89:2732–2736. Verheijen MH, Defize LH 1997 Parathyroid hormone related peptide activates mitogen-activated protein kinase via a cAMP-mediated pathway independent of Ras. J Biol Chem 272:3423–3429. Henderson JE, Amizuka N, Warshawsky H, Biasotto D, Lanske BM, Goltzman D, Karaplis AC 1995 Nucleolar localization of parathyroid hormone–related peptide enhances survival of chondrocytes under conditions that promote apoptotic cell death. Mol Cell Biol 15:4064–4075. Jans DA, Hubner S 1996 Regulation of protein transport to the nucleus: central role of phosphorylation. Physiol Rev 76:651–685. Jans DA, Xiao CY, Lam MH 2000 Nuclear targeting signal recognition: A key control point in nuclear transport? Bioessays 22:532–544. Lam MH, Olsen SL, Rankin WA, Ho PW, Martin TJ, Gillespie MT, Moseley JM 1997 PTHrP and cell division: expression and localization of PTHrP in a keratinocyte cell line (HaCaT) during the cell cycle. J Cell Physiol 173:433–446. Massfelder T, Dann P, Wu TL, Vasavaba R, Helwig JJ, Stewart AF 1997 Opposing mitogenic and antimitogenic actions of parathyroid hormone related protein in vascular smooth muscle cells: a critical role for nuclear targeting. Proc Natl Acad Sci USA 94:13630–13635. Bouvet M, Nardin SR, Burton DW, Behling C, Carethers JM, Moossa AR, Deftos LJ 2001 Human pancreatic adenocarcinomas express parathyroid hormone-related protein. J Clin Endocrinol Metab 86:310–316. Yeung SC, Eton O, Burton DW, Deftos LJ, VassilopoulouSellin R, Gagel RF 1998 Hypercalcemia due to parathyroid hormone-related protein secretion by melanoma. Horm Res 49:288–291. Bucht E, Rong H, Pernow Y, Nordqvist AC, Eriksson E, Rankin W, von Schoultz EW, Burtis WJ, Granberg B, Falkmer UG, Burton DW, Deftos LJ 1998 Parathyroid hormone-related protein in patients with primary breast cancer and eucalcemia. Cancer Res 58:4113–4116. Henderson JE, He B, Goltzman D, Karaplis AC 1996 Constitutive expression of parathyroid hormone–related peptide (PTHrP) stimulates growth and inhibits differentiation of CFK2 chondrocytes. J Cell Physiol 169:33–41. Jans DA, Hassan G 1998 Nuclear targeting by growth factors, cytokines and their receptors: A role in signaling? Bioessays 20:400–411. Lam MH, Briggs LJ, Hu W, Martin TJ, Gillespie MT, Jans DA 1999 Importin beta recognizes parathyroid hormonerelated protein with high affinity and mediates its nuclear import in the absence of importin alpha. J Biol Chem 274: 7391–7398. Watson PH, Fraher LJ, Natale BV, Kisiel M, Hendy GN, Hodsman AB 2000 Nuclear localization of the type 1 parathyroid hormone/parathyroid hormone-related peptide receptor in MC3T3-E1 cells: Association with serum-induced proliferation. Bone 26:221–225. Aarts MM, Rix A, Guo J, Bringhurst R, Henderson JE 1999 The nucleolar targeting signal of parathyroid hormone–related protein mediates endocytosis and nucleolar translocation. J Bone Min Res 14:1493–1503.

747 42. Kopito RR 1997 ER quality control: The cytoplasmic connection. Cell 88:427–430. 43. Meerovitch K, Wing S, Goltzman D 1997 Pre-proparathyroid hormone-related protein, a secreted peptide is a substrate for the ubiquitin proteolytic system. J Biol Chem 272: 6707–6013. 44. Fujihara SM, Nadler SG 1998 Modulation of nuclear protein import: a novel means of regulating gene expression. Biochem Pharmacol 56:157–161. 45. Stewart AF, Horst R, Deftos LJ, Camden EC, Lang R, Broadus AE 1980 Biochemical evaluation of patients with cancer-associated hypercalcemia. Evidence for humoral and non-humoral groups. N Engl J Med 303:1377–1383. 46. Kaiser SM, Laneuville P, Bernier SM, Rhim JS, Kremer R, Goltzman D 1992 Enhanced growth of a human keratinocyte cell line induced by antisense RNA for parathyroid hormonerelated peptide. J Biol Chem 267:13623–13628. 47. Burton PB, Moniz C, Quirke P, Malik A, Bui TD, Juppner H, Segre GV, Knight DE 1992 Parathyroid hormone-related peptide: expression in fetal and neonatal development. J Pathol 167:291–296. 48. Benitezverguizas J, Esbrit P 1994 Proliferative effect of parathyroid hormone-related protein on the hypercalcemic walker 256 carcinoma cell line. Biochem Biophys Res Commun 198:1281–1289. 49. Rabbani SA, Gladu J, Liu B, Goltzman D 1995 Regulation in vivo of the growth of Leydig cell tumors by antisense ribonucleic acid for parathyroid hormone related peptide. Endocrinology 136:5416–5422. 50. Huang D, Kremer R 2002 Inhibition of metastasis in melanoma following targeted disruption of the PTHrP gene: Enhanced visualization of the invasive process with green fluorescence protein. Proceedings of the 24th Annual Meeting of the ASBMR, San Antonio, TX. 51. Bouizar Z, Spyratos F, Deytieux S, de Vernejoul MC, Jullienne A 1993 Polymerase chain reaction analysis of parathyroid hormone-related protein gene expression in breast cancer patients and occurrence of bone metastases. Cancer Res 53:5076–5078. 52. Truong NU, deB Edwardes MD, Papavasiliou V, Goltzman D, Kremer R 2003 Parathyroid hormone-related peptide and survival of patients with cancer and hypercalcemia. Am J Med 115:115–121. 53. Kremer R, Shustik C, Tabak T, Papavasiliou V, Goltzman D 1996 Parathyroid-hormone-related peptide in hematologic malignancies. Am J Med 100:406–411. 54. Henderson JE, Sebag M, Rhim J, Goltzman D, Kremer R 1991 Dysregulation of parathyroid hormone-like peptide expression and secretion in a keratinocyte model of tumor expression. Cancer Res 51:6521–6528. 55. Merryman JI, De Wille J, Wermeister JR, Capen CL, Rosol TJ 1994 Effects of transforming growth factors on parathyroid hormone–related protein production and ribonucleic acid expression by a squamous carcinoma cell line in vitro. Endocrinology 134:2424–2430. 56. Casey ML, Mike M, Erk A, MacDonald PC 1992 Transforming growth factor β, stimulation of parathyroid hormone related protein expression in human uterine cells in culture; mRNA levels and protein secretion. J Clin Endocrinol Metab 74:950–952. 57. Ikeda K, Lu C, Weir EC, Mangin M, Broadus AE 1989 Transcriptional regulation of the parathyroid hormone related peptide gene by glucocorticoids and vitamin D in a human C-cell line. J Biol Chem 264:15743–15746.

748 58. Sharpe GR, Dillon JP, Durham B, Gallagher JA, Fraser WD 1998 Human keratinocytes express transcripts for three isoforms of parathyroid hormone–related protein (PTHrP), but not for the parathyroid hormone/PRTHrP receptor: Effects of 1,25(OH)2D3. Brit J Dermatol 138:944–951. 59. Abe M, Abeno N, Oshida S, Horiuchi N 1998 Inhibitory effects of 1,25 dihydroxyvitamin D3 and 9-cis-retinoic acid on parathyroid hormone–related protein expression by oral cancer cells (NSC-3). J Endocrinol 156:349–357. 60. Yu J, Papavasiliou V, Rhim J, Goltzman D, Kremer R 1995 Vitamin D analogs: New therapeutic agents for the treatment of squamous cancer and its associated hypercalcemia. Anticancer Drugs 6:101–108. 61. Inoue D, Matsumoto T, Ogata E, Ikeda K 1993 22-Oxacalcitriol, a noncalcemic analogue of calcitriol, suppresses both cell proliferation and parathyroid hormone-related peptide gene expression in human T cell lymphotrophic virus, type I– infected T cells. J Biol Chem 268:16730–16736. 62. Kremer R, Sebag M, Champigny C, Meerovitch K, Hendy GN, White J, Goltzman D 1996 Identification and characterization of 1,25-dihydroxyvitamin D3-responsive repressor sequences in the rat parathyroid hormone-related peptide gene. J Biol Chem 271:16310–16316. 63. Nishishita T, Okazaki T, Ishikawa T, Igarashi T, Hata K, Ogata E, Fujita T 1998 A negative vitamin D response. DNA element in the human parathyroid hormone–related peptide gene binds to vitamin D receptor along with Ku antigen to mediate negative gene regulation by vitamin D. J Biol Chem 273(18):10901–10907. 64. Falzon M, Zhong J 1998 The non-calcemic vitamin D analogs EB1089 and 22-oxacalcitriol suppress serum-induced parathyroid hormone–related peptide gene expression in a lung cancer cell line. Endocrinology 139:1046–1053. 65. Hongo T, Kupfer J, Enomoto H, Sharifi B, Giannella-Neto D, Forrester JS, Singer FR, Goltzman D, Hendy GN, Pirola C, et al. 1991 Abundant expression of parathyroid hormone related protein in primary aortic smooth muscle cells accompanies serum-induced proliferation. J Clin Invest 88:1841–1847. 66. Streutker C, Drucker DJ 1991 Rapid induction of parathyroid hormone-like peptide gene expression by sodium butyrate in a rat islet cell line. Mol Endocrinol 5:703–708. 67. Kremer R, Woodworth CD, Goltzman D 1996 Expression and action of parthyroid hormone-related peptide in human cervical epithelial cells. Am J Physiol 271:C164–C171. 68. Cros M, Silve C, Graulet AM, Morieux C, Urena P, de Vernejoul MC, Bouizar Z 1998 Estrogen stimulates PTHrP but not PTHPTHrP receptor gene expression in the kidney of ovariectomized rat. J Cell Biochem 70:84–93. 69. Chilco PJ, Gerardi JM, Kaczmarczyk SJ, Chu S, Leopold V, Zajac JD 1993 Calcitonin invreases transcription of parathyroid hormone related protein via cAMP. Mol Cell Endocrinol 94:1–7. 70. Rodan SB, Wesolowski G, Ianacone J, Thiede MA, Rodan GA 1989 Production of parathyroid hormone-like peptide in a human osteosarcoma cell line: Stimulation by phorbol esters and epidermal growth factor. J Endocrinol 122:219–227. 71. Lu C, Ikeda K, Deftos LJ, Cazdar AF, Mangin M, Broadus AE 1989 Glucocorticoid regulation of PTHrP gene transcription in a human neuroendocrine cell line. Mol Endocrinol 3: 2034–2040. 72. Deftos LJ, Hogue-Angeletti R, Chalberg C, Su T 1989 PTHrP secretion is stimulated by CT and inhibited by chromogranin A peptides. Endocrinology 125:563–565. 73. Eto M, Akishita M, Ishikawa M, Kozaki K, Yoshizumi M, Hashimoto M, Ako J, Sugimoto N, Nagano K, Sudoh N,

DAVID GOLTZMAN AND RICHARD KREMER

74.

75.

76.

77.

78. 79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

Roba K, Ouchi Y 1998 Cytokine-induced expression of parathyroid hormone–related peptide in cultured human vascular endothelial cells. Biochem Biophys Res Commun 249:339–343. Funk JL, Wei H 1998 Regulation of parathyroid hormone– related protein expression in MCF-7 breast carcinoma cells by estrogen and antiestrogens. Biochem Biophys Res Commun 251:849–854. Heath JK, Southby J, Fukumoto S, O’Keefe LM, Martin TJ, Gillespie MT 1995 Epidermal growth factor-stimulated parathyroid hormone-related protein expression involves increased gene transcription and mRNA stability. Biochem J 307:159–167. Rizzoli R, Feyen JHM, Grau G, Wohlwend A, Sappino AP, Bonjour JP 1994 Regulation of parathyroid hormone-related protein production in a human lung squamous cell carcinoma line. J Endocrinol 143:333–342. Kremer R, Goltzman D, Amizuka N, Webber MM, Rhim JS 1997 ras Activation of human prostate epithelial cells induces overexpression of parathyroid hormone–related peptide. Clin Cancer Res 3:855–859. Aklilu F, Park M, Goltzman D, Rabbani SA 1996 Increased PTHrP production by a tyrosine kinase oncogene, TRP-MET: role of the ras signaling pathway. Am J Physiol 271:E277–E283. Broadus AE, Stewart AF 1994 Parathyroid hormone related protein. Structure, processing and physiological actions. In: Bilezikian JP, Levine MA, Marcus R (eds) The Parathyroids. Raven Press, New York, pp. 259–294. Sidler B, Alpert L, Henderson JE, Deckelbaum R, Amizuka N, Silva E, Goltzman D, Karaplis AC 1996 Amplification of the parathyroid hormone–related peptide (PTHrP) gene amplification in a colonic carcinoma. J Clin Endocrinol 81:2841–2847. Solomon C, White JH, Kremer R 1999 Mitogen-activated protein kinase inhibits 1,25-dihydroxyvitamin D3–dependent signal transduction by phosphorylating human retinoid X receptor alpha. J Clin Invest 103:1729–1735. Powell GJ, Southby J, Danks JA, Stillwell RG, Hayman JA, Henderson MA, Bennett RC, Martin TJ 1991 Localization of parathyroid hormone-related protein in breast cancer metastases: increased incidence in bone compared with other sites. Cancer Res 51:3059–3061. Southby J, Kissin MW, Danks JA, Hayman JA, Moseley JM, Henderson MA, Bennett RC, Martin TJ 1990 Immunohistochemical localization of parathyroid hormone-related protein in human breast cancer. Cancer Res 50:7710–7716. Vargas SJ, Gillespie MT, Powell GJ, Southby J, Danks JA, Moseley JM, Martin TJ 1992 Localization of parathyroid hormone-related protein mRNA expression in breast cancer and metastatic lesions by in situ hybridization. J Bone Miner Res 7:971–979. Bundred NJ, Walder RA, Ratcliffe WA, Wguuck J, Morisson JM, Ratcliffe JG 1992 Parathyroid hormone related protein and skeletal morbidity in breast cancer. Eur J Cancer 28(213):690–692. Kissin MW, Henderson MA, Danks JA, Hayman JA, Bennett RC, Martin TJ 1993 Parathyroid hormone related protein in breast cancers of widely varying prognosis Eur J Surg Oncol 19:134–142. Grill V, Ho P, Body JJ, Johanson N, Lee SC, Krukeja SC, Moseley JM, Martin TJ 1991 Parathyroid hormone-related protein: elevated levels in both humoral hypercalcemia of malignancy and hypercalcemia complicating metastatic breast cancer. J Clin Endocrinol Metab 73:1309–1315. Guise TA 1997 Parathyroid hormone-related protein and bone metastases. Cancer 80:1572–1580.

CHAPTER 43 Target Genes: PTHrP

89. Guise TA, Yin JJ, Taylor SD, Kumagai Y, Dallas M, Boyce BF, Yoneda T, Mundy GR 1996 Evidence for a causal role of parathyroid hormone-related protein in the pathogenesis of human breast cancer mediated osteolysis. J Clin Invest 98: 1544–1549. 90. Sato K 1999 Does PTHrP mediate cancer-associated cachexia as well as humoral hypercalcemia of malignancy? In: 7th Int Forum on Calcified Tissue and Bone Metabolism, pp. 32–37. 91. Sato K, Yamakawa Y, Shimuze K, Satoh T, Mohojomi K, Demura H, Akatsu T, Nagata N, Kasahara T, Ohkawa H, Ohsumi K 1993 Passive immunization with antiparathyroid hormone-related protein antibody markedly prolongs survival time of hypercalcemia nude mice bearing transplanted human PTHrP-producing tumors. J Bone Min Res 8:849–860. 92. Stein CA, Cheng YC 1993 Antisense oligonucleotides as therapeutic agents. Is the bullet really magical? Science 261:1004–1012. 93. Armentano D, Yu SF, Kantoff PW, Von Ruden T, Anderson WF, Gilboa E 1987 Effects of internal viral sequences on the utility of retroviral vectors. J Virol 61:1047–1050. 94. Palmer JD, Rosman GJ, Osborne WRA, Miller AD 1991 Genetically modified skin fibroblasts persist long after transplantation but gradually inactivate introduced genes. Proc Natl Acad Sci USA 88:1330–1334. 95. Aklilu F, Park M, Goltzman D, Rabbani SA 1997 Induction of parathyroid hormone–related peptide by the Ras oncogene: role of Ras farnesylvation inhibitors as potential therapeutic agents for hypercalcemia of malignancy. Cancer Res 57: 4517–4522. 96. Falzon M 1996 DNA sequences in the rat parathyroid hormone-related peptide gene responsible for 1,25-dihydroxyvitamin D3–mediated transcriptional repression. Mol Endocrinol 10:672–681. 97. Binderup L, Brammn E 1988 Effect of a novel vitamin D analog MC903 on cell proliferation and differentiation in vitro and on calcium metabolism in vivo. Biochem Pharmacol 37:889–895. 98. Kissmeyer AM, Binderup L 1991 Calcipotriol (MC903): pharmacokinetics in rats and biological activities of metabolites. A comparative study with 1,25(OH)2D3. Biochem Pharmacol 41:1601–1606.

749 99. Kragballe K, Gjertsen BT, de Hoop D, Karlsmark T, van de Kerkhof PC, Larko O, Nieboer C, Roed-Petersen J, Strand A, Tikjob G 1991 Double-blind, right/left comparison of calcipotriol and betamethasone valerate in treatment of psoriasis vulgaris. Lancet 337:193–196. 100. Haq M, Kremer R, Goltzman D, Rabbani SA 1993 A vitamin D analogue (EB 1089) inhibits parathyroid hormone-related peptide production and prevents the development of malignancy-associated hypercalcemia in vivo. J Clin Invest 91:2416–2422. 101. El Abdaimi K, Papavasiliou V, Rabbani SA, Rhim JS, Goltzman D, Kremer R 1999 Reversal of hypercalcemia with the vitamin D analogue EB 1089 in a human model of squamous cancer. Cancer Res 59:3325–3328. 102. El Abdaimi K, Dion N, Papavasiliou V, Cardinal PE, Binderup L, Goltzman D, Ste-Marie LG, Kremer R 2000 The vitamin D analogue EB 1089 prevents skeletal metastasis and prolongs survival time in nude mice transplanted with human breast cancer cells. Cancer Res 60:4412–4418. 103. Sasaki A, Boyce BF, Story B, Wright KR, Chapman M, Boyce R, Mundy GR, Yoneda T 1995 Bisphosphonate residronate reduces metastatic human breast cancer burden in bone in nude mice. Cancer Res 55:3551–3557. 104. Huang DC, Papavasiliou V, Rhim JS, Horst RL, Kremer R 2002 Targeted disruption of the 25-dihydroxyvitamin D3, 1α-hydroxylase gene in ras-transformed keratinocytes demonstrates that locally produced 1α,25-dihydroxyvitamin D3 suppresses growth and induces differentiation in an autocrine fashion. Mol Cancer Res 1:56–67. 105. Thiede MA, Harm SC, Hasson DM, Gardner RM 1991 In vivo regulation of parathyroid hormone-related peptide messenger ribonucleic acid in the rat uterus by 17β-estradiol. Endocrinology 128:2317–2323. 106. El Abdaimi K, Papavasiliou V, Goltzman D, Kremer R 2000 Expression and regulation of parathyroid hormone-related peptide in normal and malignancy melanocytes. Am J Physiol Cell Physiol 279:C1230–C1238. 107. Thiede MA 1989 The mRNA encoding a parathyroid hormonelike peptide is produced in mammary tissue in response to elevations in serum prolactin. Mol Endocrinol 3:1443–1447. 108. Huang DC, Papavasiliou V, Rhim JS, Horst RL, Kremer R 2002 Mol Cancer Res 1:1–12.

CHAPTER 44

Effects of 1,25-Dihydroxyvitamin D3 on Voltage-Sensitive Calcium Channels in the Vitamin D Endocrine System MARY C. FARACH-CARSON AND JOEL J. BERGH Department of Biological Sciences, University of Delaware

I. Systemic and Intracellular Ca2+ Homeostasis II. Voltage-Sensitive Calcium Channels III. 1,25-Dihydroxyvitamin D3 and Voltage-Sensitive Ca2+ Channels IV. Membrane-Initiated Ca2+ Responses to 1,25(OH)2D3

V. VI. VII. VIII.

I. SYSTEMIC AND INTRACELLULAR Ca2+ HOMEOSTASIS

stores in the endoplasmic reticulum through leak channels and by influx of extracellular Ca2+ through plasma membrane channels that include voltage-sensitive calcium channels (VSCCs), voltage-insensitive calcium channels (VICCs), receptor-operated calcium channels (ROCs), and mechanosensitive divalent cation channels (MDCCs). Calcitropic hormones, including 1,25-dihydroxyvitamin D3 (1,25(OH)2D3), and parathyroid hormone (PTH), and mechanical load modulate the activity of these plasma membrane Ca2+ channels. In osteoblasts, VSCCs serve as key regulators of Ca2+ permeability and are the major Ca2+ channels present in the plasma membrane [3]. Additionally, Ca2+ influx through L-type VSCCs in response to membrane depolarization events maintains the activity of the cAMP- and Ca2+-dependent transcription factor CREB [4]. Evidence suggests that transcriptional activation through CREB is more efficient when the Ca2+ signal is generated through the L-type VSCC than by other channels that permit Ca2+ entry [4,5].

All mammals must maintain plasma Ca2+ concentrations within a narrow homeostatic set point to ensure proper control of Ca2+-regulated cellular function and phenotype. Ca2+ homeostasis involves hormonal regulation by 1,25(OH)2D3 at three major sites, the kidney, intestine, and bone. As serum Ca2+ levels decrease, 1,25(OH)2D3 production increases, facilitating increased Ca2+ absorption in the intestines, decreased Ca2+ excretion in the urine, and a shift in bone remodeling to a state that favors resorption. The skeletal system is the location of roughly 99% of all Ca2+ found in the human body. The skeleton undergoes continuous remodeling, generating a small pool of Ca2+ that is freely exchangeable with the extracellular fluid, and establishing a buffer system to aid in the maintenance of circulating Ca2+ concentrations. Ca2+ levels in cells are kept in a dynamic equilibrium by the activities of channels, pumps, and exchangers in both the plasma membrane and internal Ca2+ storage organelles, including the endoplasmic reticulum, mitochondria, and nucleus. While organelles and intracellular Ca2+-binding proteins can sequester intracellular Ca2+, they only transiently can buffer cytosolic increases. Therefore, the overall maintenance of intracellular Ca2+ levels is maintained by the plasma membrane, which extrudes Ca2+ into the extracellular space using the energy of Ca2+-ATPases and collaborating Na+/Ca2+ exchangers [1,2]. Intracellular Ca2+ levels also are buffered by the uptake of Ca2+ into internal stores, such as the mitochondria and the endoplasmic reticulum. Ca2+ levels in the cytoplasm increase by release of Ca2+ from VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Ca2+ Induced Inactivation of VSCCs Calcium and Transcriptional Responses to 1,25(OH)2D3 Cross-Talk between Membrane and Nuclear Actions Summary and Conclusions References

II. VOLTAGE-SENSITIVE CALCIUM CHANNELS VSCCs are present in all excitable tissues and in most nonexcitable cell types. VSCCs have been identified in several tissues in the vitamin D endocrine system, including the kidney [6], intestine [7], and bone [3,8]. VSCCs mediate the influx of Ca2+ in response to membrane depolarization and regulate numerous intracellular functions, including contraction, secretion, gene transcription, neurotransmitter release, and cellular differentiation. Many of these responses are tuned to Copyright © 2005, Elsevier, Inc. All rights reserved.

752

MARY C. FARACH-CARSON AND JOEL J. BERGH

a specific temporal and spatial pattern of Ca2+ entry [9,10] which must be balanced with the potential toxicity caused by high intracellular levels of Ca2+. The need for regulated patterns of Ca2+ influx is reflected by the presence of multiple unique subtypes of VSCCs, each differing in its kinetics, pharmacology, and tissue distribution. VSCCs were first solubilized and purified from the transverse tubules of skeletal muscle [11]. The purification identified the α1, β, and γ subunits and showed that the α1 and β subunits contain consensus sites for cAMP-dependent phosphorylation [11]. Further biochemical analysis revealed the presence of the α2/δ subunit [12]. Analysis of the protein sequences, hydropathicity, and glycosylation properties of these five subunits produced a model (Fig. 1) composed of the transmembrane α1 subunit in association with the disulfide-linked α2/δ dimer, an intracellular β subunit, and a transmembrane γ subunit [13–15]. The α1 subunit has been the focus of many biological studies because it is the large pore-forming subunit, the site of Ca2+ translocation, and can generate a Ca2+ current in the absence of the other subunits [16]. With a mass of approximately 175 kDa, the α1 subunit is the receptor for three classes of organic Ca2+ channel blockers [17]. Figure 2 shows the general structure of all α1 subunits, consisting of four repeating domains, each containing six hydrophobic transmembrane segments that are embedded in the plasma membrane. The fourth segment of each domain is distinguished by a collection of repeating positively charged amino acid TABLE I α1 Subunit Types and Function Ca2+ channel

Common nomenclature

Ca2+ current type

CaV1.1 CaV1.2

α1S α1C

L L

CaV1.3

α1D

L

CaV1.4 CaV2.1

α1F α1A

L P/Q

CaV2.2

α1B

N

CaV2.3

α1E

R

CaV3.1

α1G

T

CaV3.2

α1H

T

CaV3.3

α1I

T

Primary tissue localization Skeletal muscle Cardiac muscle Neurons, Bone Endocrine cells Neurons Retina Nerve terminal Dendrites Nerve terminal Dendrites Cell bodies Nerve terminals Cardiac muscle Skeletal muscle Neurons Cardiac muscle Neurons Neurons

Ca2+

α2

Extracellular δ γ

Intracellular

α1

β

FIGURE 1

Subunit structure of the L-type VSCC. The L-type VSCC consists of four or five subunits depending on the cell type (figure adapted from [13,14]). The general structure consists of a pore-forming α1 subunit through which Ca2+ travels along the concentration gradient. A glycosylated α2δ subunit is present on the extracellular face and is the result of posttranslational processing of a single gene product. A γ subunit is sometimes present (dotted line). The β subunits are located on the intracellular face and play an important role in modification of channel conductance behavior and in plasma membrane targeting of the VSCC.

residues. With similar topology to Na+ and K+ channels, both the short amino-terminal and long carboxyl-terminal segments of the α1 are located intracellularly where they can be modified during signal transduction [18–21]. Molecular cloning has revealed at least 10 distinct α1 subunits [21,22], but only four are found associated with L-type channel currents [17]. Table I lists each known α1 subunit, the classic nomenclature, the Ca2+ current associated with the subunit, and the tissue where it is primarily expressed. Different gene products code for these four different subtypes of the α1, which are named the α1C, α1D, α1F, and the α1S subtypes. These four subtypes share approximately 80% sequence similarity. When the α1 subunit alone is expressed in Xenopus oocytes, a Ca2+ current is established [2]. Physiological and pharmacological studies demonstrate functional similarities between various α1 subunits,

Domain:

I

II

III

IV Extracellular

NH3+ CO2−

Transmembrane organization of the pore-forming α1 subunit of the L-type VSCC. The L-type VSCC α1 subunit consists of four domains linked by relatively large intracellular loops (adapted from [13,14]). Both the amino and carboxyl termini also are located intracellularly. This orientation presents the cytosolic face of the channel to interact with signaling cascades and permits it to interact with calmodulin (see text).

FIGURE 2

CHAPTER 44 Effects of 1,25-Dihydroxyvitamin D3 on Voltage-Sensitive Calcium Channels

which allow for VSCCs to be classified into high-voltage activated (L-, P/Q-, N, and R-type) and low-voltage activated (T-type) classes [23]. L-type Ca2+ currents are mediated by VSCCs containing α1C, α1D, α1F, and α1S subunits, which have about 80% amino acid identity with one another. α1A, α1B, and α1E (P/Q-, N-type, and R-type VSCC, respectively) make up the remainder of the high-voltage activated VSCC families. The lowvoltage activated T-type VSCCs, composed of α1G, α1H, and α1I, are distantly related to the other known homologs, with less than 25% amino acid sequence identity. The β subunit is a 56-kDa protein that is highly phosphorylated in vitro. Hydrophobicity analysis revealed the lack of a membrane-spanning segment, suggesting it is localized intracellularly. Pulse chase analysis shows that β subunits are posttranscriptionally modified by palmitoylation, which aids in membrane association [24]. In rabbit brain, four distinct β subunits have been identified. When α1 subunits are expressed in Xenopus oocytes, the activation and inactivation kinetics of the Ca2+ channel are abnormally slow compared to measurements performed in native cell preparations [17]. Recombinant coexpression of the α1 subunit with β subunits almost fully restores normal channel currents. This finding supports the idea that the β subunit can regulate channel kinetics, voltage-dependent gating properties, and channel density. The affiliation of different β subunits with the α1 subunit results in calcium channels with different electrophysiological properties [17,24–28]. The α2/δ subunits are disulfide linked and, together, form a 155-kDa protein complex that, after reduction, produces a 125-kDa α2 and 30-kDa δ polypeptide. These subunits are encoded by the same gene and formed by posttranslational processing [29]. The α2 subunit lies extracellularly, while the δ subunit, with a single transmembrane-spanning segment, resides in the plasma membrane. There are five separate mRNA species for these subunits, and more recent studies suggest the existence of two more α2/δ genes. Osteoblastic cells express the α2/δ1 and the α2/δ3 isoforms of the α2/δ subunit [30]. The extracellular α2 subunit facilitates the assembly of α1 at the cell surface, and their ability to modulate α1-induced current is more pronounced if they are coexpressed with the β subunit [13,31–35]. This result shows that, like the β subunit, the α2/δ is capable of regulating current amplitude in L-type Ca2+ channels [17,31,32,36]. The last subunit involved in forming a native L-type VSCC is the γ subunit, whose function is still largely unknown. In skeletal muscle, this highly glycosylated protein exhibits considerable hydrophobicity, suggesting localization to the plasma membrane, and a mass

753

of 35 kDa. In cardiac tissue, purification studies have not yet revealed the presence of a γ subunit, although several protein bands in the range of 120–130 kDa were found. These bands might be complexes of associated auxiliary subunits because a γ subunit-like cDNA has been isolated from cardiac mRNA preparations by PCR cloning [2,17]. Expression of the α1 subunit is sufficient to produce functional channels, but with lower expression and different kinetics and voltage dependence than native channels [16]. Coexpression of the α2/δ subunit confers more normal gating properties and enhances the expression of the channel [29]. Coexpression of β subunits generally increases channel expression, while shifting the voltage dependence of activation and inactivation to more negative membrane potentials and increasing the rate of inactivation [13]. The γ subunit alters peak currents and reduces channel availability by a negative shift of the voltage dependence of inactivation [37].

III. 1,25-DIHYDROXYVITAMIN D3 AND VOLTAGE-SENSITIVE Ca2+ CHANNELS 1,25(OH)2D3 is a critical hormonal modulator of Ca2+ homeostasis and plays an important role in osteoblast function during bone remodeling. Many of its calcemic actions occur in concert with PTH and are sensitive to the cell cycle [38]. Regulation of bone resorption by osteoclasts in response to 1,25(OH)2D3 is mediated through resorptive signals generated by osteoblastic cells. 1,25(OH)2D3 generates biological responses through the regulation of gene transcription and by the commencement of rapid, membrane-initiated events. Long-term (hours to days) treatment with 1,25(OH)2D3 can generate cellular responses by binding the nuclear vitamin D receptor (nVDR) and altering nVDR sensitive gene transcription [39–44]. In mice, 1,25(OH)2D3 stimulates the production of several noncollagenous matrix proteins, including osteopontin [41], and downregulates osteocalcin [42] and parathyroid hormone (PTH) [43,44]. The binding of 1,25(OH)2D3 to its receptor, translocation to the nucleus, interaction with coactivators, and modulation of gene expression is the best characterized cellular response to 1,25(OH)2D3 treatment. However, the initial response after exposure to secosteroid is increasingly appreciated. Similar to observations that other steroid hormones, including estrogen [45] and glucocorticoids [46], have membrane-initiated actions, the role of 1,25(OH)2D3 in membrane events warranted study in bone. Rapid responses to 1,25(OH)2D3 are proposed to be facilitated through the binding of the ligand with a plasma membrane-associated receptor.

754 The rapid actions of 1,25(OH)2D3 have been related to the induction of protein kinase C [47], phospholipase C [48], adenylyl cyclase [49], membrane sphingomyelinases [50], phosphorylation of matrix proteins including OPN [51], and modulation of intracellular Ca2+ levels [52]. Previous reports using microarray analysis demonstrated that treatment of osteoblasts with 1,25(OH)2D3 alters gene expression through nVDR-dependent and independent pathways [53]. Many changes in expression have been observed as early as 3 hr posttreatment for genes that include stress response proteins, transcription factors, and various matrix proteins, which lack a vitamin D response element in their promoters. These observations, along with 1,25(OH)2D3 responsiveness in nVDR-free membrane preparations, suggest the presence of separate nuclear and membrane receptors for 1,25(OH)2D3 both of which can be linked to transcriptional change. A 64.5-kDa protein, called 1,25D3-MARRS (membrane-associated rapid response to steroids) has been identified as a potential membrane receptor for 1,25(OH)2D3 [54–57]. The N-terminal sequence was used to generate a specific antibody that identified a component of the membrane vitamin D response system in other tissues, including chick kidney, and brain [58] and in rat chondrocytes [59]. Using the antibody in a function blocking role inhibited intracellular Ca2+ signaling caused by 1,25(OH)2D3 [55,58–60]. These data suggest that 1,25D3-MARRS is involved in generating rapid signals in response to 1,25(OH)2D3. Interestingly, Ca2+ signals generated by activation of the plasma membrane vitamin D receptor have been associated with changes in gene expression, similar to those seen in response to prostaglandins [47,53,61]. Changes in gene transcription can be observed within 3 hr after addition of 1,25(OH)2D3 or Ca2+ mobilizing vitamin D3 analogs [53], indicating that the rapid transcriptional effects are related to Ca2+ influx rather than to activation of the nVDR. Addition of Ca2+ channel blockers negates many of the rapid actions of 1,25(OH)2D3, including the ability to modify the phosphorylation state of OPN [51], suggesting that the membrane-initiated actions of 1,25(OH)2D3 require the activity of plasma membrane Ca2+ channels.

IV. MEMBRANE-INITIATED Ca2+ RESPONSES TO 1,25(OH)2D3 Application of 1,25(OH)2D3 to target cells results in a rapid and transient increase in intracellular Ca2+ in some cell types [62–64]. The elevation in intracellular Ca2+ is dependent on release from internal stores and influx through the plasma membrane [64].

MARY C. FARACH-CARSON AND JOEL J. BERGH

Low nanomolar concentrations of 1,25(OH)2D3 can elicit a transient local elevation in intracellular Ca2+ through influx of extracellular Ca2+ through the plasma membrane, whereas supraphysiological levels of 1,25(OH)2D3 promote the release of Ca2+ from internal stores and yield measurable Ca2+ transients [62]. Changes in osteoblast Ca2+ permeability in the presence of 1,25(OH)2D3 are regulated by the activity of the plasma membrane vitamin D response system, although the role of the nVDR in this action remains controversial. 1,25(OH)2D3 can trigger a Ca2+ transient in response to 1,25(OH)2D3 treatment; in some reports this occurs even when the nVDR is missing [65]. Recent findings report a loss of rapid responses in knockout mice lacking the nVDR [66], and our laboratory has found a strict correlation between nVDR positive status and rapid responses in more than 30 cell lines tested with various ligand analogs [55,67]. Pharmacological studies provide further insights. The rapid elevation of local intracellular Ca2+ can be mimicked using synthetic analogs of 1,25(OH)2D3. Analog 1,24-dihydroxy-22ene-24-cyclopropyl D3 (termed analog BT) has a high binding affinity for the nVDR and activates genomic signaling pathways that lead to changes in gene transcription of various bone matrix proteins, including osteocalcin, osteopontin, and type-I collagen [67–69]. In contrast to 1,25(OH)2D3, analog BT does not stimulate rapid changes in Ca2+ permeability. Other classes of 1,25(OH)2D3 analogs that lack the 1-α-hydroxyl group, including 25-hydroxy-16-ene-23-yne-D3 (termed analog AT), do not bind to the nVDR, but application of these analogs rapidly increases Ca2+ influx into osteoblastic cells [69]. The use of analogs that can selectively activate the plasma membrane vitamin D response system or the nVDR have provided powerful tools to help identify the specific attributes of short- and longterm cellular responses to 1,25(OH)2D3 treatment [68]. 1,25(OH)2D3 binding to the membrane vitamin D response system activates several signaling pathways. In chick embryogenesis, 1,25(OH)2D3 stimulates VSCC activity through the activation of adenylyl cyclase, leading to the production of cAMP and activation of protein kinase A [70]. The channel activation can be mimicked in the absence of 1,25(OH)2D3 by the addition of forskolin, an adenylyl cyclase activator, and by dibutyryl-cAMP. Electrophysiological studies have shown that the major mechanism for Ca2+ influx into the osteoblast cell is the L-type VSCC, which shows a prolonged open time in the presence of 1,25(OH)2D3 [3] (Fig. 3A). Furthermore, application of the L-type VSCC agonist Bay K 8644 or analog AT increases Ca2+ influx into the osteoblast and shifts the threshold of activation towards the resting potential, an event termed “left shift” [3,71] (Fig. 3B). “Left-shifting” of

755

CHAPTER 44 Effects of 1,25-Dihydroxyvitamin D3 on Voltage-Sensitive Calcium Channels

A

Basal

Interactions of VSCCs and VICCs in Osteoblasts

1,25(OH)2D3

1,25(OH)2D3 “left shift”

+

PTH local depolarization

=

Depolarization response

Change in single channel conductance

B

−40 Vm(mV)

−30

−20

−10

0

10

20

30

40

Left shift −2

VSCC

VICC Ca2+

DHP block −4 Im(×10−10)

FIGURE 3 Effect of 1,25(OH)2D3 on behavior of the L-type

VSCC in osteoblasts. (A) The L-type α1c-containing VSCC in the osteoblast demonstrates a basal activity seen in single-channel recordings that supports basal Ca2+-dependent signaling. Addition of 1,25(OH)2D3 to the cell in continuous recording shows a change in single-channel conductance marked by increased mean open time and long nulls. This occurs without an increase in whole cell inward current (B). The solid dots represent untreated cells; open dots represent cells treated with 5.0 nM 1,25(OH)2D3. The open squares show further increase in left-shift during washout to low nanomolar concentrations. The filled squares represent results when 1 µM BAY K8644, an L-type VSCC agonist, is added. The arrow indicates the direction of the “left shift” indicating a change in the threshold of channel activation toward the resting potential (near −40 in these cells).

the membrane resting potential predicts that plasma membrane VSCCs in osteoblastic cells are more susceptible to opening following stimulation with other hormones acting through ROCs or in response to smaller membrane depolarizations such as may be associated with activation of VICCs or MDCCs. The physiological implication for a “left shift” is evident in the interactions between 1,25(OH)2D3 and PTH. It has long been recognized that PTH can stimulate Ca2+ entry into the osteoblastic cell via influx through a gadolinium-sensitive Ca2+ channel that interacts with neighboring VSCCs [72] (Fig. 4). Pretreatment of osteoblastic cell cultures with low nanomolar concentrations of 1,25(OH)2D3 for 10 min results in an enhancement of PTH-induced Ca2+ influx that is associated with increased bone resorption rates [71,73]. This suggests that 1,25(OH)2D3 serves a priming function to augment PTH-induced Ca2+ influx at the plasma membrane (Fig. 4). Removal of extracellular Ca2+ or application of the L-type VSCC inhibitor nitrendipine, a dihydropyridine (DHP), inhibits the elevation of 1,25(OH)2D3 and PTH induced Ca2+ influx [71], indicating that the enhanced PTH-induced Ca2+ influx depends on the presence and influx of Ca2+ through the L-type VSCC [73].

VSCC Ca2+

Gd 3+block

Ca2+

DHP block

FIGURE 4 Interactions of VSCCs and VICCs in osteoblasts. Full depolarization of VSCCs occurs after osteoblasts are treated first with 1,25(OH)2D3 that “left shifts” the VSCC as shown in Fig. 3, then with a local depolarizing stimulus such as PTH interacting with a VICC. The left-shifted VSCC senses the local depolarization through the VICC and allows large amounts of Ca2+ to enter the cell through the pore (solid dark arrow). As reported earlier [71], the left-shift and full depolarization response through the L-type VSCC can be inhibited by dihydropyridine channel blockers (DHP), whereas the VICC can be inhibited by Gd3+. Under the conditions shown, a Ca2+ transient would occur and involve further Ca2+ release from intracellular stores as discussed in the text.

V. Ca2+-INDUCED INACTIVATION OF VSCCS Osteoblastic cells respond to 1,25(OH)2D3 and express significant amounts of the nVDR [74]. Exposure to physiological levels of 1,25(OH)2D3 produces a rapid change in membrane permeability that precedes any changes in gene expression. In osteoblasts, alteration of membrane Ca2+ permeability is regulated primarily by the activity of the L-type VSCC. Spontaneous and hormonally regulated opening of Ca2+ channels has been observed in resting cells and leads to localized elevations of intracellular Ca2+, sometimes referred to as “sparks” [75], that directly control the release of secretory vesicles and the generation of nuclear-specific Ca2+ signals [5,76]. In cardiac cells, Ca2+ influx rapidly inactivates the Ca2+ current, resulting in the channel current returning to its resting potential during long membrane depolarizations [77,78]. Some time ago, our laboratory demonstrated that inactivation of the L-type VSCC in ROS 17/2.8 cells was faster in 20 mM external Ca2+ and that the steady-state inward current was smaller than in experiments that were performed in 5 mM external Ca2+ [3] (Fig. 5). Ca2+-dependent inactivation was observed when cardiac VSCCs were inserted into lipid bilayers in the absence of cytosolic components and ATP [79], suggesting that Ca2+ is binding to the VSCC directly or to a protein that is coupled with the channel complex. The detection of two exon splice variants of

756

−40

MARY C. FARACH-CARSON AND JOEL J. BERGH

−20

Ca2+-dependent inactivation 40 0 20

60

80

1

2

= 5 mM Ca2+ = 20 mM Ca2+

3

Im

= 20 mM Ba2+

(×10−10A)

FIGURE 5 Ca2+-dependent inactivation of the L-type VSCC in osteoblasts. When Ba2+ is used as the permeant ion (solid dot) a large inward current is seen through α1c. When Ca2+ is placed into the extracellular medium at two concentrations (solid triangle, 5 mM; open circle, 20 mM) rapid channel inactivation occurs (arrow) (adapted from [3]). As discussed in the text, this Ca2+-dependent inactivation involves calmodulin.

the L-type VSCC that have drastically different inactivation rates led to the identification of Ca2+/calmodulin kinase as a regulator of the L-type VSCC [80–82]. The C-terminal end of the rapid inactivating splice variant includes an IQ domain, a well-characterized calmodulin binding site [82]. Two calmodulin mutants, one unable to associate with the VSCC and the other incapable of binding Ca2+, lack the ability to generate the Ca2+dependent inactivation of the L-type VSCC [82–84]. This demonstrates that Ca2+-dependent inactivation of the L-type VSCC requires Ca2+ binding to calmodulin and the subsequent complex must associate with the C-terminal tail of the Ca2+ channel. Calmodulin has been reported to be tethered and localized to the C-terminal tail of the L-type VSCC [85]. This supports the notion that Ca2+ influx across the plasma membrane is a highly regulated localized event under negative feedback regulation rather than a signal for global change in intracellular Ca2+.

VI. CALCIUM AND TRANSCRIPTIONAL RESPONSES TO 1,25(OH)2D3 The classic genomic signals generated by 1,25(OH)2D3 are mediated through binding of the secosteroid–receptor complex to vitamin D response element sequences up-stream of target sequences and the subsequent alteration of gene transcription [86,87] (see Chapter 11–19). In osteoblasts, 48-hr treatment with 1,25(OH)2D3 results in the increase in deposition of osteoid, an extracellular matrix specific to bone cells (see Chapter 37, 41). The increase in deposition is due

to an up-regulation of gene transcription of osteoid proteins, mediated by the actions of the nuclear vitamin D receptor [88]. Activation of this nuclear receptor initiates a signaling cascade that results in alteration of target gene expression. Some genes that are transcriptionally regulated by 1,25(OH)2D3 include alkaline phosphatase, IL-3 receptor, osteopontin, osteocalcin, type I collagen, GM-colony stimulating factor, and PTH. It has been established that depolarization and hormonal stimulated Ca2+ influx into osteoblastic cells is inhibited by the application of dihydropyridines [3], indicating that the L-type VSCCs are involved in this influx of Ca2+. Three of the four known L-type VSCCs, α1C, α1D, and α1S, were found to be expressed in human osteoblastic cells, although only α1C is always present [89]. The identification of these subclasses that are typically seen in excitable tissues, including neurons [90] and skeletal muscle [91], in osteoblastic cells is not surprising, given the fact that they are found in a variety of other nonexcitable tissues and cells including lung [92], kidney [93], pancreas [94], and fibroblasts [95]. Shortterm application of 1,25(OH)2D3 increases the mean open time of the L-type VSCC [3]. In primary osteoblast cultures, 1,25(OH)2D3 application leads to an increase in Ca2+ permeability and, when this is accompanied by release from intracellular stores [62,63], to a subsequent elevation in intracellular Ca2+ levels, resulting in the stimulation of many Ca2+-dependent signaling pathways. If unregulated, continued exposure to 1,25(OH)2D3 could result in sustained influx of intracellular Ca2+ and lead to cell death [76]. It was found that long-term exposure of osteoblasts to low nanomolar concentrations of 1,25(OH)2D3 resulted in down-regulation of α1C mRNA transcript levels, as well as a subsequent decrease in α1C protein expression [96]. Using analog BT, which binds the nuclear vitamin D receptor and does not elicit a plasma-membrane response, it was further demonstrated that the down-regulation of the α1C subunit is due to transcriptional changes mediated through the nuclear receptor and not as a result of a membrane initiated signal [96]. Radioactive Ca2+ influx assays revealed that prolonged exposure to 1,25(OH)2D3 led to a decrease in the amount of Ca2+ that enters the cell through the L-type VSCC, presumably due to the decrease in expression of the α1C subunit. A potential role for down-regulation of the α1C subunit in response to long-term exposure to 1,25(OH)2D3 is to protect the cell from chronic elevations in intracellular Ca2+ that could lead to cell apoptosis. To that end, it has been demonstrated in hippocampal neurons that neuronal vulnerability to excitotoxicity is mediated through Ca2+ influx through the L-type VSCC, and down-regulation of these channels with long-term exposure to 1,25(OH)2D3 leads to increased neuroprotection [97]. Together, these

CHAPTER 44 Effects of 1,25-Dihydroxyvitamin D3 on Voltage-Sensitive Calcium Channels

results suggest that 1,25(OH)2D3 exposure elicits a rapid cellular response, including the activation of various protein kinases, phospholipases, and generation of cAMP, by increasing the ability of Ca2+ to enter the osteoblast through the α1C subunit of the L-type VSCC. Long-term exposure to the secosteroid results in a down-regulation of the α1C subunit in a nuclear receptor–mediated pathway, which results in a diminished ability of the cell to maintain elevated intracellular Ca2+ levels, thus preventing apoptosis and a prolonged cellular response to 1,25(OH)2D3.

757

Finally, in myoblasts, phospholipase c redistribution and activation occurs following rapid 1,25(OH)2D3-induced, Ca2+-dependent signal transduction involving c-Src and PI3K [102]. Taken together, it is clear that the integrated responses of cells to vitamin D hormone represent a continuum of highly regulated responses that often begin with activation of VSCCs and Ca2+ influx, and ultimately lead to changes in gene transcription and cell behavior and phenotype.

VII. SUMMARY AND CONCLUSIONS VII. CROSS-TALK BETWEEN MEMBRANE AND NUCLEAR ACTIONS The genomic and membrane-initiated actions of 1,25(OH)2D3 provide opportunities for cross-talk and feedback loops among the various pathways (recently reviewed in [98]) (see Chapter 23). For example, immediately after 1,25(OH)2D3 treatment, the α1C subunit if the VSCC increases its open time [3] and allows more Ca2+ to enter the cell, especially in sites proximal to the pores. This happens without increasing current amplitude [3]. Local increases in Ca2+ activate calmodulin and can lead to channel inactivation as discussed earlier. Continued presence of the hormone leads to activation of the nuclear receptor and eventual decreases in mRNA levels encoding α1C [96]. The depression in VSCC mRNA and protein levels leads to decreased responsiveness to 1,25(OH)2D3 at the plasma membrane. This mechanism provides cells a means to respond to and then attenuate the rapid response to secosteroid, preventing Ca2+ toxicity while supporting Ca2+-dependent signaling. 1,25(OH)2D3 also interacts with peptide hormones that include PTH, transforming growth factor β (TGFβ), and inflammatory cytokines to modulate cellular responses [98]. Gene expression in many cell types changes in response to 1,25(OH)2D3 treatment alone and as a consequence of cross-talk between 1,25(OH)2D3activated pathways and peptide hormone-activated pathways. As an example, 1,25(OH)2D3 can cause a rapid phosphorylation of serine residues on IκBα in monocytes, which synergize with PKC-dependent signaling pathways to regulate NFκB translocation and signaling [99]. In a similar fashion, Smad proteins conduct signals downstream of TGFβ that mediate cross-talk between TGFβ and 1,25(OH)2D3 signaling in osteoblasts [100]. As another example of cross-talk, PTH treatment of osteoblasts activates PKA, which through phosphorylation modulates VSCC function, and in the presence of 1,25(OH)2D3-activated CaMK alters [Ca2+]i, and regulates secretion of osteoclastic coupling factors [101].

Taken together, it is clear that the integrated responses of cells to vitamin D hormone represent a continuum of highly regulated responses that often begin with activation of VSCCS and Ca2+ influx, and ultimately lead to changes in gene transcription and cell behavior and phenotype. It is impossible to separate the intracellular pathways that involve rapid responses from those that modulate nuclear hormone receptors, for the simple reason that elaborate crosstalk between these response systems is the norm rather than the exception. Activation and inactivation of VSCCs, both low and high voltage channels, as well as regulation of VSCC biosynthesis, assembly and membrance insertion provide first tier controls that fine tune Ca2+-dependent hormone responses of target cells in the vitamin D endocrine system.

Acknowledgments The work in the authors’ laboratory was supported by grants from the NIH/NIDCR. We thank Dr. Errin Lagow and Dr. Dan Carson for their many helpful discussions and proofreading. We acknowledge all of the members of the Carson and Farach-Carson laboratories for their individual contributions to the development of this work over the years, and to our many excellent collaborators. Some of this work was presented in the doctoral thesis of J.J.B.

References 1. Rasmussen H, Barrett P, Zawalich W, Isales C, Stein P, Smallwood J, McCarthy R, Bollag W 1989 Cycling of Ca2+ across the plasma membrane as a mechanism for generating a Ca2+ signal for cell activation. Ann NY Acad Sci 568:73–80. 2. Tsien RW, Tsien RY 1990 Calcium channels, stores, and oscillations. Annu Rev Cell Biol 6:715–760. 3. Caffrey JM, Farach-Carson MC 1989 Vitamin D3 metabolites modulate dihydropyridine-sensitive calcium currents in

758

4. 5. 6.

7.

8. 9. 10. 11.

12.

13. 14. 15.

16.

17.

18. 19.

20.

21.

MARY C. FARACH-CARSON AND JOEL J. BERGH

clonal rat osteosarcoma cells. J Biol Chem 264: 20265–20274. Bading H, Ginty DD, Greenberg ME 1993 Regulation of gene expression in hippocampal neurons by distinct calcium signaling pathways. Science 260:181–186. Catterall WA 1998 Structure and function of neuronal Ca2+ channels and their role in neurotransmitter release. Cell Calcium 24:307–323. Barry EL, Gesek FA, Yu AS, Lytton J, Friedman PA 1998 Distinct calcium channel isoforms mediate parathyroid hormone and chlorothiazide-stimulated calcium entry in transporting epithelial cells. J Membr Biol 161:55–64. Buffenstein R, Sergeev IN, Pettifor JM 1994 Absence of calcitriol-mediated nongenomic actions in isolated intestinal cells of the damara mole-rat (Cryptomys damarensis). Gen Comp Endocrinol 95:25–30. Meszaros JG, Karin NJ, Farach-Carson MC 1996 Voltagesensitive calcium channels in osteoblasts: mediators of plasma membrane signalling events. Connect Tissue Res 35:107–111. Dolmetsch RE, Lewis RS, Goodnow CC, Healy JI 1997 Differential activation of transcription factors induced by Ca2+ response amplitude and duration. Nature 386:855–858. Gu X, Spitzer NC 1995 Distinct aspects of neuronal differentiation encoded by frequency of spontaneous Ca2+ transients. Nature 375:784–787. Curtis BM, Catterall WA 1984 Purification of the calcium antagonist receptor of the voltage-sensitive calcium channel from skeletal muscle transverse tubules. Biochemistry 23: 2113–2118. Hosey MM, Barhanin J, Schmid A, Vandaele S, Ptasienski J, O’Callahan C, Cooper C, Lazdunski M 1987 Photoaffinity labelling and phosphorylation of a 165 kilodalton peptide associated with dihydropyridine and phenylalkylaminesensitive calcium channels. Biochem Biophys Res Commun 147:1137–1145. Catterall WA 2000 Structure and regulation of voltage-gated Ca2+ channels. Annu Rev Cell Dev Biol 16:521–555. Catterall WA 1995 Structure and function of voltage-gated ion channels. Annu Rev Biochem 64:493–531. Takahashi M, Seagar MJ, Jones JF, Reber BF, Catterall WA 1987 Subunit structure of dihydropyridine-sensitive calcium channels from skeletal muscle. Proc Natl Acad Sci USA 84:5478–5482. Perez-Reyes E, Kim HS, Lacerda AE, Horne W, Wei XY, Rampe D, Campbell KP, Brown AM, Birnbaumer L 1989 Induction of calcium currents by the expression of the α1-subunit of the dihydropyridine receptor from skeletal muscle. Nature 340:233–236. Varadi G, Strobeck M, Koch S, Caglioti L, Zucchi C, Palyi G 1999 Molecular elements of ion permeation and selectivity within calcium channels. Crit Rev Biochem Mol Biol 34:181–214. Chesnoy-Marchais D, Fritsch J 1988 Voltage-gated sodium and calcium currents in rat osteoblasts. J Physiol 398:291–311. Chien AJ, Carr KM, Shirokov RE, Rios E, Hosey MM 1996 Identification of palmitoylation sites within the L-type calcium channel β2a subunit and effects on channel function. J Biol Chem 271:26465–26468. Mitterdorfer J, Froschmayr M, Grabner M, Moebius FF, Glossmann H, Striessnig J 1996 Identification of PK-A phosphorylation sites in the carboxyl terminus of L-type calcium channel α1 subunits. Biochemistry 35:9400–9406. Khan I, Ahmad S, Thomas N 1998 Differential translation of the α1 isoforms of L-type calcium channel in rat brain and other tissues. Biochem Mol Biol Int 45:895–904.

22. Randall A, Benham CD 1999 Recent advances in the molecular understanding of voltage-gated Ca2+ channels. Mol Cell Neurosci 14:255–272. 23. Reuter H 1996 Diversity and function of presynaptic calcium channels in the brain. Curr Opin Neurobiol 6:331–337. 24. Catterall WA 1996 Molecular properties of sodium and calcium channels. J Bioenerg Biomembr 28:219–230. 25. McEnery MW, Copeland TD, Vance CL 1998 Altered expression and assembly of N-type calcium channel α1B and β subunits in epileptic lethargic (lh/lh) mouse. J Biol Chem 273:21435–21438. 26. Mermelstein PG, Foehring RC, Tkatch T, Song WJ, Baranauskas G, Surmeier DJ 1999 Properties of Q-type calcium channels in neostriatal and cortical neurons are correlated with β subunit expression. J Neurosci 19:7268–7277. 27. Gao T, Chien AJ, Hosey MM 1999 Complexes of the α1C and β subunits generate the necessary signal for membrane targeting of class C L-type calcium channels. J Biol Chem 274:2137–2144. 28. Walker D, Bichet D, Geib S, Mori E, Cornet V, Snutch TP, Mori Y, De Waard M 1999 A new β subtype-specific interaction in α1A subunit controls P/Q-type Ca2+ channel activation. J Biol Chem 274:12383–12390. 29. Singer D, Biel M, Lotan I, Flockerzi V, Hofmann F, Dascal N 1991 The roles of the subunits in the function of the calcium channel. Science 253:1553–1557. 30. Bergh JJ, Shao Y, Akanbi K, Farach-Carson MC 2003 Rodent osteoblastic cells express voltage-sensitive calcium channels lacking a γ subunit. Calcif Tissue Int, 73:502–510. 31. Dolphin AC, Wyatt CN, Richards J, Beattie RE, Craig P, Lee RH, Cribbs LL, Volsen SG, Perez-Reyes E 1999 The effect of α2-δ and other accessory subunits on expression and properties of the calcium channel α1G. J Physiol 519:35–45. 32. Klugbauer N, Lacinova L, Marais E, Hobom M, Hofmann F 1999 Molecular diversity of the calcium channel α2δ subunit. J Neurosci 19:684–691. 33. Jay SD, Sharp AH, Kahl SD, Vedvick TS, Harpold MM, Campbell KP 1991 Structural characterization of the dihydropyridine-sensitive calcium channel α2δ-subunit and the associated δ peptides. J Biol Chem 266:3287–3293. 34. Gurnett CA, De Waard M, Campbell KP 1996 Dual function of the voltage-dependent Ca2+ channel α2δ subunit in current stimulation and subunit interaction. Neuron 16:431–440. 35. Felix R, Gurnett CA, De Waard M, Campbell KP 1997 Dissection of functional domains of the voltage-dependent Ca2+ channel α2δ subunit. J Neurosci 17:6884–6891. 36. De Jongh KS, Warner C, Catterall WA 1990 Subunits of purified calcium channels α2 and δ are encoded by the same gene. J Biol Chem 265:14738–14741. 37. Wei X, Pan S, Lang W, Kim H, Schneider T, Perez-Reyes E, Birnbaumer L 1995 Molecular determinants of cardiac Ca2+ channel pharmacology. Subunit requirement for the high affinity and allosteric regulation of dihydropyridine binding. J Biol Chem 270:27106–27111. 38. Bizzarri C, Civitelli R 1994 Activation of the Ca2+ message system by parathyroid hormone is dependent on the cell cycle. Endocrinology 134:133–140. 39. Yamamoto T, Ozono K, Shima M, Yamaoka K, Okada S 1998 24R,25-Dihydroxyvitamin D3 increases cyclic GMP contents, leading to an enhancement of osteocalcin synthesis by 1,25-dihydroxyvitamin D3 in cultured human osteoblastic cells. Exp Cell Res 244:71–76. 40. Hannah SS, Norman AW 1994 1α,25(OH)2 vitamin D3regulated expression of the eukaryotic genome. Nutr Rev 52:376–382.

CHAPTER 44 Effects of 1,25-Dihydroxyvitamin D3 on Voltage-Sensitive Calcium Channels 41. Noda M, Vogel RL, Craig AM, Prahl J, DeLuca HF, Denhardt DT 1990 Identification of a DNA sequence responsible for binding of the 1,25-dihydroxyvitamin D3 receptor and 1,25-dihydroxyvitamin D3 enhancement of mouse secreted phosphoprotein 1 (SPP-1 or osteopontin) gene expression. Proc Natl Acad Sci USA 87:9995–9999. 42. Zhang R, Ducy P, Karsenty G 1997 1,25-Dihydroxyvitamin D3 inhibits osteocalcin expression in mouse through an indirect mechanism. J Biol Chem 272:110–116. 43. Russell J, Lettieri D, Sherwood LM 1986 The effects of 1,25dihydroxyvitamin D3 on the synthesis of pre-proparathyroid hormone mRNA. Trans Assoc Am Physicians 99:189–196. 44. Russell J, Lettieri D, Sherwood LM 1986 Suppression by 1,25(OH)2D3 of transcription of the pre-proparathyroid hormone gene. Endocrinology 119:2864–2866. 45. Pappas TC, Gametchu B, Watson CS 1995 Membrane estrogen receptors identified by multiple antibody labeling and impededligand binding. FASEB J 9:404–410. 46. Hua SY, Chen YZ 1989 Membrane receptor-mediated electrophysiological effects of glucocorticoid on mammalian neurons. Endocrinology 124:687–691. 47. van Leeuwen JP, Birkenhager JC, van den Bemd GJ, Buurman CJ, Staal A, Bos MP, Pols HA 1992 Evidence for the functional involvement of protein kinase C in the action of 1,25-dihydroxyvitamin D3 in bone. J Biol Chem 267: 12562–12569. 48. Grosse B, Bourdeau A, Lieberherr M 1993 Oscillations in inositol 1,4,5-trisphosphate and diacylglycerol induced by vitamin D3 metabolites in confluent mouse osteoblasts. J Bone Miner Res 8:1059–1069. 49. Zanello LP, Norman AW 1997 Stimulation by 1α,25(OH)2vitamin D3 of whole cell chloride currents in osteoblastic ROS 17/2.8 cells. A structure–function study. J Biol Chem 272:22617–22622. 50. Liu R, Xu Y, Farach-Carson MC, Vogel JJ, Karin NJ 2000 1,25-Dihydroxyvitamin D3 activates sphingomyelin turnover in ROS17/2.8 osteosarcoma cells without sphingolipidinduced changes in cytosolic Ca2+. Biochem Biophys Res Commun 273: 95–100. 51. Safran JB, Butler WT, Farach-Carson MC 1998 Modulation of osteopontin post-translational state by 1,25-dihydroxyvitamin D3. Dependence on calcium influx. J Biol Chem 273:29935–29941. 52. Long GJ, Rosen JF 1994 Lead perturbs 1,25 dihydroxyvitamin D3 modulation of intracellular calcium metabolism in clonal rat osteoblastic (ROS 17/2.8) cells. Life Sci 54:1395–1402. 53. Farach-Carson MC, Xu Y 2002 Microarray detection of gene expression changes induced by 1,25-dihydroxyvitamin D3 and a calcium influx-activated analog in ROS 17/2.8 cells. Steroids 67:467–470. 54. Nemere I, Dormanen MC, Hammond MW, Okamura WH, Norman AW 1994 Identification of a specific binding protein for 1α,25-dihydroxyvitamin D3 in basal-lateral membranes of chick intestinal epithelium and relationship to transcaltachia. J Biol Chem 269:23750–23756. 55. Farach-Carson MC, Nemere I 2003 Membrane receptors for vitamin D steroid hormones: Potential new drug targets. Curr Drug Targets 4:67–76 56. Norman AW, Song X, Zanello L, Bula C, Okamura WH 1999 Rapid and genomic biological responses are mediated by different shapes of the agonist steroid hormone, 1α,25(OH)2 vitamin D3. Steroids 64:120–128. 57. Mesbah M, Nemere I, Papagerakis P, Nefussi JR, Orestes-Cardoso S, Nessmann C, Berdal A 2002 Expression

58. 59.

60.

61.

62.

63. 64.

65.

66.

67.

68.

69.

70. 71.

72. 73.

759

of a 1,25-dihydroxyvitamin D3 membrane-associated rapidresponse steroid binding protein during human tooth and bone development and biomineralization. J Bone Miner Res 17:1588–15896. Jia Z, Nemere I 1999 Immunochemical studies on the putative plasmalemmal receptor for 1,25-dihydroxyvitamin D3 II. Chick kidney and brain. Steroids 64:541–550. Pedrozo HA, Schwartz Z, Rimes S, Sylvia VL, Nemere I, Posner GH, Dean DD, Boyan BD 1999 Physiological importance of the 1,25(OH)2D3 membrane receptor and evidence for a membrane receptor specific for 24,25(OH)2D3. J Bone Miner Res 14:856–867. Nemere I, Larsson D, Sundell K 2000 A specific binding moiety for 1,25-dihydroxyvitamin D3 in basal lateral membranes of carp enterocytes. Am J Physiol Endocrinol Metab 279:E614–E621. Choi AM, Tucker RW, Carlson SG, Weigand G, Holbrook NJ 1994 Calcium mediates expression of stress-response genes in prostaglandin A2-induced growth arrest. FASEB J 8:1048–1054. Civitelli R, Kim YS, Gunsten SL, Fujimori A, Huskey M, Avioli LV, Hruska KA 1990 Nongenomic activation of the calcium message system by vitamin D metabolites in osteoblast-like cells. Endocrinology 127:2253–2262. Lieberherr M 1987 Effects of vitamin D3 metabolites on cytosolic free calcium in confluent mouse osteoblasts. J Biol Chem 262:13168–13173. Vazquez G, Selles J, de Boland AR, Boland R 1999 Rapid actions of calcitriol and its side chain analogues CB1093 and GS1500 on intracellular calcium levels in skeletal muscle cells: a comparative study. Br J Pharmacol 126:1815–1823. Baran DT, Sorensen AM, Shalhoub V, Owen T, Oberdorf A, Stein G, Lian J 1991 1α,25-dihydroxyvitamin D3 rapidly increases cytosolic calcium in clonal rat osteosarcoma cells lacking the vitamin D receptor. J Bone Miner Res 6:1269–1275. Erben RG, Soegiarto DW, Weber K, Zeitz U, Lieberherr M, Gniadecki R, Moller G, Adamski J, Balling R 2002 Deletion of deoxyribonucleic acid binding domain of the vitamin D receptor abrogates genomic and nongenomic functions of vitamin D. Mol Endocrinol 16:1524–1537. Farach-Carson MC, Sergeev I, Norman AW 1991 Nongenomic actions of 1,25-dihydroxyvitamin D3 in rat osteosarcoma cells: structure-function studies using ligand analogs. Endocrinology 129:1876–1884. Norman AW, Nemere I, Zhou LX, Bishop JE, Lowe KE, Maiyar AC, Collins ED, Taoka T, Sergeev I, Farach-Carson MC 1992 1,25(OH)2-vitamin D3, a steroid hormone that produces biologic effects via both genomic and nongenomic pathways. J Steroid Biochem Mol Biol 41:231–240. Khoury R, Ridall AL, Norman AW, Farach-Carson MC 1994 Target gene activation by 1,25-dihydroxyvitamin D3 in osteosarcoma cells is independent of calcium influx. Endocrinology 135:2446–2453. Norman AW, Wehling M 1999 Overview of the first international meeting on rapid responses to steroid hormones. Steroids 64:3–4. Li W, Duncan RL, Karin NJ, Farach-Carson MC 1997 1,25Dihydroxyvitamin D3 enhances PTH-induced calcium transients in preosteoblasts by activating L-type calcium channels. Am J Physiol 273:E599–E605. Duncan RL, Hruska KA, Misler S 1992 Parathyroid hormone activation of stretch-activated cation channels in osteosarcoma cells (UMR-106.01). FEBS Lett 307:219–223. Li W, Farach-Carson MC 2001 Parathyroid hormone-stimulated resorption in calvaria cultured in serum-free medium is

760

74.

75. 76. 77.

78. 79. 80.

81.

82.

83. 84. 85. 86. 87. 88. 89.

MARY C. FARACH-CARSON AND JOEL J. BERGH

enhanced by the calcium-mobilizing activity of 1,25-dihydroxyvitamin D3. Bone 29:231–235. Dokoh S, Donaldson CA, Haussler MR 1984 Influence of 1,25-dihydroxyvitamin D3 on cultured osteogenic sarcoma cells: correlation with the 1,25-dihydroxyvitamin D3 receptor. Cancer Res 44:2103–2109. Nelson MT, Cheng H, Rubart M, Santana LF, Bonev AD, Knot HJ, Lederer WJ 1995 Relaxation of arterial smooth muscle by calcium sparks. Science 270:633–637. Berridge MJ, Bootman MD, Lipp P 1998 Calcium—a life and death signal. Nature 395:645–648. Lee KS, Marban E, Tsien RW 1985 Inactivation of calcium channels in mammalian heart cells: Joint dependence on membrane potential and intracellular calcium. J Physiol 364:395–411. Nilius B, Benndorf K 1986 Joint voltage- and calciumdependent inactivation of Ca channels in frog atrial myocardium. Biomed Biochim Acta 45:795–811. Haack JA, Rosenberg RL 1994 Calcium-dependent inactivation of L-type calcium channels in planar lipid bilayers. Biophys J 66:1051–1060. Soldatov NM, Oz M, O’Brien KA, Abernethy DR, Morad M 1998 Molecular determinants of L-type Ca2+ channel inactivation. Segment exchange analysis of the carboxyl-terminal cytoplasmic motif encoded by exons 40–42 of the human α1C subunit gene. J Biol Chem 273:957–963. Soldatov NM, Zuhlke RD, Bouron A, Reuter H 1997 Molecular structures involved in L-type calcium channel inactivation. Role of the carboxyl-terminal region encoded by exons 40–42 in α1C subunit in the kinetics and Ca2+ dependence of inactivation. J Biol Chem 272:3560–3566. Zuhlke RD, Bouron A, Soldatov NM, Reuter H 1998 Ca2+ channel sensitivity towards the blocker isradipine is affected by alternative splicing of the human α1C subunit gene. FEBS Lett 427:220–224. Qin N, Olcese R, Bransby M, Lin T, Birnbaumer L 1999 Ca2+-induced inhibition of the cardiac Ca2+ channel depends on calmodulin. Proc Natl Acad Sci USA 96:2435–2438. Zuhlke RD, Pitt GS, Deisseroth K, Tsien RW, Reuter H 1999 Calmodulin supports both inactivation and facilitation of L-type calcium channels. Nature 399:159–162. Pitt GS, Zuhlke RD 2001 Molecular basis of calmodulin tethering and Ca2+-dependent inactivation of L-type Ca2+ channels. J Biol Chem 276:30794–30802. Ozono K, Sone T, Pike JW 1991 The genomic mechanism of action of 1,25-dihydroxyvitamin D3. J Bone Miner Res 6:1021–1027. Haussler MR 1986 Vitamin D receptors: Nature and function. Annu Rev Nutr 6:527–562. Pike JW 1991 Vitamin D3 receptors: Structure and function in transcription. Annu Rev Nutr 11:189–216. Barry EL 2000 Expression of mRNAs for the α1 subunit of voltage-gated calcium channels in human osteoblast-like cell lines and in normal human osteoblasts. Calcif Tissue Int 66:145–50.

90. Hell JW, Westenbroek RE, Warner C, Ahlijanian MK, Prystay W, Gilbert MM, Snutch TP, Catterall WA 1993 Identification and differential subcellular localization of the neuronal class C and class D L-type calcium channel α1 subunits. J Cell Biol 123:949–962. 91. Tanabe T, Takeshima H, Mikami A, Flockerzi V, Takahashi H, Kangawa H, Kojima M, Matsuo H, Hirose T, Numa S 1987 Primary structure of the receptor for calcium channel blockers from skeletal muscle. Nature 328:313–318. 92. Biel M, Ruth P, Bosse E, Hullin R, Stuhmer W, Flockerzi V, Hofmann F 1990 Primary structure and functional expression of a high voltage activated calcium channel from rabbit lung. FEBS Lett 269:409–412. 93. Zhao PL, Wang XT, Zhang XM, Cebotaru V, Cebotaru L, Guo G, Morales M, Guggino SE 2002 Tubular and cellular localization of the cardiac L-type calcium channel in rat kidney. Kidney Int 61:1393–1406. 94. Seino S, Yamada Y, Espinosa R, 3rd, Le Beau MM, Bell GI 1992 Assignment of the gene encoding the α1 subunit of the neuroendocrine/brain-type calcium channel (CACNL1A2) to human chromosome 3, band p14.3. Genomics 13: 1375–1377. 95. Soldatov NM 1992 Molecular diversity of L-type Ca2+ channel transcripts in human fibroblasts. Proc Natl Acad Sci USA 89:4628–4632. 96. Meszaros JG, Karin NJ, Akanbi KA, Farach-Carson MC 1996 Down-regulation of L-type Ca2+ channel transcript levels by 1,25-dihyroxyvitamin D3. Osteoblastic cells express L-type α1C Ca2+ channel isoforms. J Biol Chem 271: 32981–32985. 97. Brewer LD, Thibault V, Chen KC, Langub MC, Landfield PW, Porter NM 2001 Vitamin D hormone confers neuroprotection in parallel with downregulation of L-type calcium channel expression in hippocampal neurons. J Neurosci 21:98–108. 98. Farach-Carson MC, Davis PJ (2003) Steroid hormone interactions with target cells: cross-talk between membrane and nuclear pathways. JPET, in press. 99. Berry DM, Clark CS, Meckling-Gill KA 2002 1α,25Dihydroxyvitamin D3 stimulates phosphorylation of IκBα and synergizes with TPA to induce nuclear translocation of NFκB during monocytic differentiation of NB4 leukemia cells. Exp Cell Res 272:176–184. 100. Gurlek A, Kumar R 2001 Regulation of osteoblast growth by interactions between transforming growth factor-β and 1α,25-dihydroxyvitamin D3. Crit Rev Eukaryot Gene Expr 11:299–317. 101. Bergh JJ, Xu Y, Farach-Carson MC 2004 Osteoprotegerin expression and secretion are regulated by calcium influx through the L-type voltage-sensitive calcium channel. Endocrinology 145:426–436. 102. Buitrago C, Gonzalez Pardo V, de Boland AR 2002 Nongenomic action of 1α,25(OH)2-vitamin D3. Activation of muscle cell PLC γ through the tyrosine kinase c-Src and PtdIns 3-kinase. Eur J Biochem 269:2506–2515.

CHAPTER 45

Vitamin D and the Cellular Response to Oxidative Stress RUTH KOREN AND AMIRAM RAVID Felsenstein Medical Research Center, Beilinson Campus, Rabin Medical Center, Petah Tivka, Israel

I. Reactive Oxygen Species and Redox Homeostasis II. Vitamin D as a Prooxidant III. Vitamin D as an Antioxidant

IV. Discussion References

I. REACTIVE OXYGEN SPECIES AND REDOX HOMEOSTASIS

Traditionally, ROS were considered the unwanted and toxic by-products of living in an aerobic environment. Indeed, when present at vulnerable sites or in excess, ROS can damage the cell and initiate cellular “damage control systems” that arrest cell proliferation, activate repair mechanisms, and induce programmed cell death when the damage is irreparable. Cells exposed to even higher ROS levels cells may undergo necrotic cell death. The severity of cellular damage is determined by the extent of the imbalance between ROS production and antioxidant protection. Oxidative challenge elicits an adaptive response that results in increased cellular repair and antioxidant capacities. Nondamaging smaller perturbations in ROS levels appear to participate in cell signaling associated with fundamental cellular processes such as cell proliferation and differentiation and regulated exocytosis. It is noteworthy that in addition to inducing cell death as a consequence of oxidative damage, ROS participate in signaling pathways that lead to programmed cell death induced by other noxious agents [1–3].

A. Introduction Recent years have witnessed the accumulation of numerous reports on the effects of hormonally active vitamin D derivatives on cellular oxidation–reduction balance. In fact, modulation of this balance may be one of the underlying mechanisms of long-recognized, seemingly unrelated actions of the hormone. Vitamin D derivatives seem to have a prooxidant effect in some systems but an antioxidant one in others. The purpose of this chapter is to present the available evidence for both modes of action and to discuss the possible underlying mechanisms for this complex cross-talk, taking into account current knowledge regarding the regulation of pro- and antioxidant cellular mechanisms.

B. The Janus Face of Reactive Oxygen Species Cells are continuously exposed to reactive oxygen species (ROS) produced intracellularly by normal aerobic metabolism and occasionally to extracellular sources of ROS such as immune cells, radiation, and xenobiotics. ROS are partially reduced compounds of oxygen, which include superoxide (⋅O−2 ), hydrogen peroxide (H2O2), and hydroxyl (⋅OH), peroxyl (ROO⋅), and alkoxyl (RO⋅) radicals. The current state of knowledge regarding the effects of ROS on cellular metabolism and cell fate and the various aspects of cellular ROS handling were reviewed extensively [1–13] and the major insights pertinent to this chapter are outlined hereafter. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

C. The Cellular Redox State The major source of energy required for cellular metabolism in an aerobic environment is derived from the movement of electrons from oxidizable donors to oxygen. Redox couples in cells are responsive to the electron flow. Some of these redox couples are linked and some are independent from other sets if activation energies are high and there are no enzyme systems that can link them kinetically. Redox state is a term originally coined to denote the ratio between the oxidized and reduced forms of an interconvertible redox couple. In recent years, the use of this term was extended to Copyright © 2005, Elsevier, Inc. All rights reserved.

762 describe the overall redox environment of a cell. Schafer and Buettner [7] suggested a more general definition for the redox state: “The redox environment of a linked set of redox couples as found in biological fluids, organelle, cell or tissue is the summation of the products of the reduction potential and reducing capacity of the linked redox couples present.” The reduction potential can be calculated by the Nernst equation and the reducing capacity by the concentration of the reduced species of a redox couple. In practice, it may be difficult or impossible to measure the potential and capacity of all the linked couples within a cell and a representative redox couple is generally used as an indicator of the redox environment. The most commonly employed indicator is the glutathione redox couple (GSSG/2GSH), which provides a very large pool of reducing equivalents and can be considered as the cellular redox buffer. It is now known that control of the intracellular redox environment is vital for proper cellular function. For protection against the constant oxidative challenge, cells have developed defense mechanisms that insure the proper balance between the prooxidant and antioxidant molecules. These antioxidant mechanisms are indispensable for cellular defense against the damage inflicted by ROS and play a crucial role in cellular redox homeostasis and regulation of redox-sensitive metabolic processes.

D. The Generation and Degradation of ROS ROS are generated in the cell both enzymatically, by oxidoreductases and nonenzymatically, as side products of reactions utilizing electron transfer. Mitochondria, the cytochrome P450s and their reductases, and nitric oxide synthases have been implicated in ROS generation. There is no compelling evidence for regulation of ROS production through specific signaling pathways, although it is clearly subject to change according to substrate availability or energy state. An exception to this notion is NADPH oxidase that generates superoxides. The mechanism of triggering and regulation of this enzyme was elucidated in macrophages but it is now recognized that it is also present in other cells. Moreover, it is activated by various agonistic stimuli such as cytokines, growth factors acting via receptor tyrosine kinases, and ligands of Gprotein coupled receptors. Superoxide anion is the main ROS produced in the course of oxidation–reduction reactions in the presence of molecular oxygen. Two molecules of superoxide rapidly dismutate in a reaction accelerated by superoxide dismutase (SOD)

RUTH KOREN AND AMIRAM RAVID

to form hydrogen peroxide and molecular oxygen. Two metal-containing SOD isoenzymes, mitochondrial Mn-SOD and the cytosolic Cu,Zn-SOD, are active in eukaryotic cells. H2O2 reactivity is moderate. It does not react spontaneously with carbon-centered molecules or thiols, but reacts rapidly with thiolates (R-S−) and transition metals. The highly toxic hydroxyl radical can be generated when H2O2 interacts with a reduced transition metal in a Fenton reaction. At least three antioxidant enzyme systems are responsible for the removal of H2O2 from the cellular milieu: catalase, glutathione peroxidases, and peroxiredoxins. Catalase catalyzes the dismutation of H2O2 molecules to water and molecular oxygen. This heme-containing enzyme is predominantly localized in the peroxisomes of mammalian cells. Glutathione peroxidases that contain a selenocysteine are cytosolic enzymes. They catalyze the reduction of H2O2 using reduced glutathione (GSH) as substrate. Glutathione disulfide (GSSG) formed in the course of this reaction is reduced to regenerate GSH by the NADPH dependent flavoenzyme glutathione reductase. Peroxiredoxins are capable of directly reducing H2O2, and the oxidized enzymes formed during the catalytic cycle are reduced by thioredoxin. The ubiquitous thioredoxin system is composed of the antioxidant enzyme pair, thioredoxin and thioredoxin reductase. Thioredoxin is a general disulfide reductant and thioredoxin reductase catalyzes the reduction of the active site disulfide of thioredoxin utilizing NADPH as the source of reductive power. Continuous availability of NADPH is needed to fuel the regeneration of reduced glutathione and thioredoxin and the maintenance of the cellular redox state. The reducing equivalents of NADPH are generated by the flow of carbon through the pentose phosphate pathway while the regulatory enzyme glucose-6phosphate dehydrogenase catalyzes the first and rate limiting step in this pathway.

E. ROS Signaling Various properties of ROS qualify them to act as second messengers in signaling cascades [6–13]: (1) They are enzymatically generated (e.g., by NAD(P)H oxireductases) in response to stimuli such as cytokines and growth factors. (2) Peroxiredoxins maintain enzymatically their basal concentration at a subthreshold level. (3) Their stimulus-elicited rise decays rapidly by glutathione peroxidase and catalase. (4) Their action is specific. ⋅O2− and H2O2 readily react with thiolate (-S−), the ionized form of thiols, forming disulfide and sulfenic acid, but cannot react at a biologically

CHAPTER 45 Vitamin D and the Cellular Response to Oxidative Stress

significant rate with thiols (-SH). Thus, only thiolatecontaining proteins will be affected by ROS directly. However, ROS may mediate the oxidation of thiols in proteins indirectly through disulfide exchange with glutathione disulfide, which is transiently increased during the reduction of hydroperoxides by glutathione peroxidases. (5) Oxidized thiols can be regenerated by GSH, thioredoxin and glutaredoxin, providing a mechanism for reversible inactivation of proteins by ROS. ROS trigger and modulate various cellular signaling pathways and affect transcription mainly via the oxidation of cysteine residues in redox-sensitive proteins. This chemical modification may affect DNA binding or enzymatic activities, the formation or release of protein complexes, or the formation of multimers. ROS were shown to modulate, among others, all the mitogenactivated protein kinases, the phosphatidyl inositol 3-kinase pathway, and the IKK/NFκB signaling pathway. The molecular targets for this action include protein phosphatases, protein kinases, small GTPases, and thioredoxin and glutathione S-transferase Pi (acting as inhibitors of ASK-1 and c-Jun N-terminal kinases, respectively). The modulation of transcription by ROS is due to the activation of signaling systems, to the direct oxidation and inactivation of transcription factors, and to their ability to induce the expression of various transcription factors. Redox-sensitive transcription factors include AP-1, NFκB, p53, Sp1, and nuclear receptors. In this context, it is interesting to note that the vitamin D receptor (VDR) itself is subject to regulation by ROS that oxidize the structural cysteines in its zinc fingers and inhibit its transcriptional activity [14]. The preceding discussion accentuates both the complexity and the redundancy of the cellular networks responsible for maintaining the essential balance between cellular ROS production and degradation. In view of this scenario, it is not unexpected that the action of 1,25(OH)2D3 or other vitamin D3 metabolites is both cell and context dependent. The mode of action of 1,25(OH)2D3 is expected to depend on the identity of the cellular system(s) involved in each case of ROS balance perturbation. It is thus not surprising that the hormone can act as pro- or antioxidant in different cellular systems and even in the same cell under different conditions. In the following paragraphs and Table I we will summarize and discuss the experimental evidence for both activities of vitamin D metabolites. We will first present some cases in which hormonally active vitamin D derivatives either enhance or attenuate the impact of exposure to preformed ROS or ROS generating agents. We will then describe the available direct evidence for vitamin D hormone-mediated modulation of the cellular ROS balance and redox state.

763

Last, we will discuss the possible role of some 1,25(OH)2D3 target genes in mediating either the proor antioxidant action of the hormone.

II. VITAMIN D AS A PROOXIDANT The first evidence for a prooxidant activity of 1,25(OH)2D3 was reported in 1988 by Polla et al. [15]. According to this study, pretreatment of human myelomonocytic leukemia cells with calcitriol increased their susceptibility to the cytotoxic action of H2O2 administered in bolus. Twenty-five years later, this finding was corroborated by a study reporting a similar enhancing activity of calcitriol in the human breast cancer cell line MCF-7 [16]. The cellular effect of ROS produced intracellularly in response to ROS-generating agents may differ from that of ROS applied to cells in bolus. In fact, it was demonstrated that the gene expression profile and intensity were different in cells exposed to H2O2 or to menadione [17], the latter simulating the endogenous cellular generation of ROS in terms of their nature and the site and rate of their production. The quinone moiety of menadione may undergo a oneelectron reduction to the corresponding semiquinone radical by various cellular flavin-centered reductases present in different compartments of the cell. In the presence of oxygen, this radical will form superoxide anions that will be dismutated by SOD to generate H2O2. Despite the difference between the two agents, however, 1,25(OH)2D3 sensitized breast cancer cells to the cytotoxic action of menadione similarly to its action on exogenous H2O2 cytotoxicity [18]. To support the hypothesis that prooxidant action of 1,25(OH)2D3 is the underlying mechanism for the enhanced cytotoxicity, it is necessary to provide evidence that its sensitizing action is specific to ROS-induced cell damage and not to cell damage per se. ROS are also involved in the cytotoxic activity of many natural and pharmacological agents with anticancer activity. Pertinent examples are the immune cytokines tumor necrosis factor (TNF) and interleukin 1 (IL-1), and doxorubicin, the widely used anticancer drug. 1,25(OH)2D3 and other active vitamin D derivatives enhance the cytotoxicity of these agents against breast cancer cells [18–21] and of TNF also against renal cell carcinoma cells [22]. In these experimental systems, cell death is only partly due to ROS and other cytotoxic mechanisms operate as well. By the use of antioxidants such as N-acetylcysteine, GSH, ascorbate, or lipoic acid it was demonstrated that 1,25(OH)2D3 preferentially reinforced the ROS-dependent mechanisms of cell death [18,20,23]. A similar mechanism may also account for

764

RUTH KOREN AND AMIRAM RAVID

TABLE I Cellular Manifestations of the Prooxidant and Antioxidant Activities of Vitamin D Species and cell origin Prooxidant effects

Antioxidant effects

Human myelomonocytic leukemia Human breast cancer Human breast cancer Human breast cancer Human breast cancer Human breast cancer Human renal cell carcinoma Human breast cancer Human renal tubular cells Rat brain substantia nigra Rat brain Rat brain Human myelomonocytic leukemia Human promyelocytic leukemia Human keratinocytes Mouse skin Rat epidermis Human epidermis Chick intestine

Cell line or cell type U937 MCF-7 MCF-7 MCF-7 T-47D MDA-MB-231 MCF-7 SK-RC-29 MCF-7 HK-2 Neurons (in vivo) Mesencephalic neurons (in vitro) Mesencephalic neurons (in vitro) U937 HL-60 HaCaT Keratinocytes (in vivo) Keratinocytes (in vitro) Keratinocytes (in vitro) Epithelium

the findings that 1,25(OH)2D3 sensitized renal tubular cells to iron-mediated toxicity [24], and that both 1,25(OH)2D3 and other active vitamin D derivatives increased the rate of apoptosis of breast cancer cells exposed to ionizing radiation [25]. However, although ROS are implicated in both scenarios of cell death, whether ROS play a role in the enhancing effect of 1,25(OH)2D3 has not been unequivocally established. The notion that 1,25(OH)2D3 precipitates cell death by a prooxidant mechanism is complemented by its lack of effect on ROS-independent modes of cell death, such as those induced by etoposide, interferon α, and cytotoxic lymphocytes [16,18,22]. The antioxidant system protects the cell from oxidative damage inflicted by exposure to excessive ROS levels and also participates in redox homeostasis. We have heretofore discussed the prooxidant role of vitamin D metabolites, manifested as exacerbation of oxidative stress-induced cell death. We will now proceed to describe the effects of the hormone on the cellular redox state. Pretreatment of breast cancer cells with 1,25(OH)2D3 increased the drop in reduced

Outcome of vitamin D action

References

Increased H2O2 cytotoxicity Increased H2O2 cytotoxicity Increased menadione cytotoxicity Increased doxorubicin cytotoxicity Increased doxorubicin cytotoxicity Increased TNF and IL-1 cytotoxicity Increased TNF cytotoxicity Increased ionizing radiation cytotoxicity Increased iron cytotoxicity Decreased 6-OHDA neurotoxicity Decreased 6-OHDA, glutamate and MPP+ neurotoxicity Decreased H2O2 neurotoxicity

[15] [16] [18] [18] [74] [19,20,23] [22] [25] [24] [46] [46,47,48]

Decreased TNF cytotoxicity

[53]

Increased differentiation by antioxidants Decreased H2O2 cytotoxicity Decreased UV-induced sunburn cell formation Decreased UV cytotoxicity

[51,52] [60] [54,55,56]

Decreased UV cytotoxicity

[56]

Calcium absorption

[67,68,69]

[46,47]

[54]

glutathione levels following TNF treatment, indicative of a more oxidized cellular redox environment [20]. More impressively, it turned out that treatment with 1,25(OH)2D3 alone significantly increased the GSSG/ GSH ratio [26]. In other words, the hormone itself can perturb the cellular redox environment and shift it toward a more oxidized state. A shift in the redox state of the major cellular redox buffer, the glutathione system, should be reflected in the potential of other thiol redox couples like those in transcription factors and the active site of enzymes. This surmise was substantiated by the finding that the activity of the redox-sensitive glycolytic enzyme GAPDH was oxidatively inhibited in breast cancer cells treated with 1,25(OH)2D3 in association with an increase of the GSSG/GSH ratio [26]. It is noteworthy that the observed increase in the glutathione redox potential (3–8 mV) is close to the redox change (15 mV) that was shown to abolish the DNA binding capacity of the transcription factors AP-1 and NFκB [27]. Changes in the redox state could also translate into reversible oxidation of cysteines in proteins that determine cell fate. Protein kinases, protein

CHAPTER 45 Vitamin D and the Cellular Response to Oxidative Stress

tyrosine phosphatases, and key components of the apoptotic process, such as the mitochondrial permeability transition pores and caspases, are all subject to redox regulation [28,29]. The increase in the glutathione redox potential may be related to an independent observation of increased ROS levels in breast cancer cells treated with 1,25(OH)2D3 alone under similar conditions [30]. The prooxidant activity of 1,25(OH)2D3 is attributed to its genomic action via the VDR. This may be inferred from the structure–function relationship of vitamin D metabolites and the time required for 1,25(OH)2D3 action [18,19,22,23]. It is thus expected that vitamin D–regulated proteins are involved in the mediation of the prooxidant effects of 1,25(OH)2D3. Two proteins that were shown to be affected by 1,25(OH)2D3 and could be involved in this activity are Cu/Zn-SOD and vitamin D up-regulated protein 1 (VDUP1). As discussed earlier, the cytosolic enzyme Cu/Zn-SOD is one of the major constituents of the cellular defense system against ROS. Treatment of breast cancer cells with 1,25(OH)2D3 was shown to decrease Cu/Zn-SOD gene expression, protein level, and enzyme activity [18]. Decrease in SOD activity may render these cells more vulnerable to oxidative challenge as inferred from previous reports on the effect of SOD overexpression [31,32]. Such a decrease would cause a shift in the balance between .O2− and H2O2. Increased levels of superoxides can, in turn, cause increased oxidative damage due to interaction with NO to form the highly toxic peroxynitrite [33] or to release of transition metal ions from intracellular stores, which supports hydroxyl radical formation via the Fenton reaction [34]. Decrease in SOD activity may also underlie the increased thiol oxidation in redox-sensitive proteins, since .O2− was shown to react with thiolate eight times faster than H2O2 [35]. Chen and DeLuca first described, cloned, and sequenced a novel protein, VDUP1, induced in the promyelocytic leukemia cell line HL-60 in response to 1,25(OH)2D3 treatment [36], but at the time (1994) its function and role remained obscure. Only later, after a series of unrelated studies concerned mainly with the regulation of the thioredoxin system, was it realized that VDUP1 is a thioredoxin binding protein that sequesters thioredoxin and limits its intracellular availability for the redox regulation and antioxidant systems [37]. Thioredoxin is now recognized to be as important for cellular redox homeostasis and redox regulation of proteins as glutathione. Thus, changes in VDUP1 can and do affect the cellular redox environment. Song et al. have shown that 1,25(OH)2D3 up-regulated VDUP1 also in murine melanoma cells [38]. They found that decreasing VDUP1 synthesis by antisense cDNA raised the reducing capacity of thioredoxin and

765

consequently diminished intracellular ROS levels, while up-regulation of the protein by calcitriol, increased ROS levels.

III. VITAMIN D AS AN ANTIOXIDANT Similar to the prooxidant action of 1,25(OH)2D3, evidence for its antioxidant activity relies mainly on the ability of the hormone to limit oxidative damage. It has long been recognized that besides the anticancer activity of the hormone, resulting in induction of apoptosis and inhibition of cell proliferation, it protects some normal cells from death-inducing stimuli. In this context, the neuroprotective action of the hormone has been a subject of numerous studies (see Chapter 100). In this chapter we will focus on the evidence that links neuroprotective effects of 1,25(OH)2D3 to its antioxidant properties. Parkinson’s disease is a chronic neurodegenerative disorder characterized by the selective loss of dopaminergic neurons in the substantia nigra. Autopsy studies backed by experimental models linked the mode of dopaminergic neuron death with excessive production of ROS resulting from dopamine metabolism [39,40]. Glutamate neurotoxicity may also play a critical role in dopaminergic neuron death. Glutamate serves as an excitatory neurotransmitter, but excessive amounts of glutamate cause calcium overload and trigger excessive ROS and nitric oxide production that in turn brings about cell death [41–43]. Parkinsonian symptoms may be induced in experimental animals by treatment with the neurotoxins 6-hydroxydopamine (6-OHDA) or 1-methyl-4-phenylpyridine (MPP+). 6-OHDA is thought to be formed endogenously in Parkinson’s disease patients through dopamine oxidation and to cause dopaminergic cell death via a free radical mechanism [44]. MPP+ accumulates within dopaminergic neurons where it induces a deleterious series of events starting with inhibition of mitochondrial respiration, leading to superoxide formation and energy failure. Again, ROS are thought to be a major mediator of MPP+-induced cell death [40,45]. Pretreatment with 1,25(OH)2D3 reduced neurotoxicity of 6-OHDA in rats [46] and protected cultured neurons against cytotoxicity induced by glutamate [47], 6-OHDA [46], and MPP+ [47,48]. Although suggestive, these findings alone do not provide direct evidence for the antioxidant action of 1,25(OH)2D3. A more direct evidence was obtained by demonstrating a protective effect against cytotoxicity induced by preformed, exogenous ROS such as H2O2 [46,47] or a superoxide generating system (hypoxanthine/xanthine oxidase) [47]. The protective effect was detectable within hours of exposure to the hormone and dependent on protein synthesis, indicating

766 a genomic mode of action of calcitriol. Moreover, 1,25(OH)2D3 reduced intracellular ROS levels formed in response to MPP+ [48] or exogenous H2O2 [47]. A clue to the mechanism of this antioxidant activity is the finding of a significant increase in total glutathione levels in mesencephalic neuron cultures following 1,25(OH)2D3 treatment [48]. Astrocytes play a major role in antioxidative and detoxification processes in the brain. Administration of 1,25(OH)2D3 to rats increased the activity of γ-glutamyl transpeptidase (γ-GT) in astrocytes and pericytes derived from their brains [49]. γ-GT is a membrane-bound enzyme that hydrolyzes extracellular GSH, and thus enables the cellular reutilization of its constituent amino acids to generate intracellular GSH. These findings were verified in primary astrocyte cultures treated with lipopolysaccharide in which 1,25(OH)2D3 treatment markedly increased γ-GT mRNA levels [50]. 1,25(OH)2D3 also brought about an increase in GSH levels in these cells, but this effect may not be entirely attributed to the up-regulation of γ-GT. γ-GT seems to play only a minor role in the increase of glutathione levels in neuronal cultures treated with 1,25(OH)2D3 [48]. However, it is assumed that in vivo astrocytes protect neurons against ROS toxicity through the supply of glutathione precursors. Therefore, up-regulation of γ-GT in astrocytes may well reinforce this cross-talk and contribute to the in vivo antioxidant action of vitamin D metabolites. Another experimental system in which 1,25(OH)2D3 acts as an antioxidant is the promyelocytic leukemia cell line HL-60. A prolonged, 96-hr treatment of these cells with 1,25(OH)2D3 decreased intracellular ROS production both in resting cells and in response to oxidative challenge [51]. Although this finding is consistent with an antioxidant effect of the hormone, it must be borne in mind that 1,25(OH)2D3 induces HL-60 cells to differentiate and that the increased cellular antioxidant capacity may be one manifestation of the more differentiated state. It is noteworthy that wellcharacterized antioxidants such as vitamin E and carnosic acid also promote differentiation of HL-60 cells and that this capacity is greatly enhanced in the presence of low concentrations of 1,25(OH)2D3 [51,52]. This synergism may also be related to the antioxidant capacity of active vitamin D metabolites. In line with the foregoing antioxidant activity of 1,25(OH)2D3 is the report of an inhibitory effect exerted by the hormone on cytotoxicty induced by the intracellular ROS-generating agent TNF in the myelomonocytic leukemia cell line U937 [53]. This protective effect was associated with increased TNF-dependent induction of the antioxidant enzyme Mn-SOD. This finding is in apparent contradiction with the study of Polla et al.

RUTH KOREN AND AMIRAM RAVID

[15] of a 1,25(OH)2D3 dependent enhancement of H2O2 cytotoxicity in this same cell line. These opposite effects of calcitriol may be due to the different types of ROS and antioxidant systems that participate in the cytotoxic processes in these two scenarios. Vitamin D metabolites are known to be potent inducers of keratinocyte differentiation. In recent years, evidence has been brought forward for an additional role of the hormone in the epidermis, the protection of epidermal keratinocytes, both in vivo and in vitro, from UV radiation-induced apoptosis [54–57]. The role of ROS, and particularly H2O2, in mediating the effects of UV on cell fate is well established [58,59]. The possibility that the protective effect of 1,25(OH)2D3 is related to its antioxidant capacity is born out by the finding that 1,25(OH)2D3 also protects keratinocytes from H2O2 cytotoxicity [60]. These protective effects are associated with inhibition of the activation of the stress-activated MAP kinase, c-Jun N-terminal kinase, by H2O2 and UV [57,60]. This action could at least partially account for the protective effect of the hormone as c-Jun N-terminal kinase is known to activate proapoptotic signaling pathways. Metallothioneins (MTs) are ubiquitous sulfhydryl-rich proteins that are readily inducible by heavy metals. One-third of the 61 amino acids of MTs are cysteines and because of this feature, they may serve as expendable targets for oxidants. Although the antioxidant properties of MTs derive mainly from sulfhydryl nucleophilicity, complexation of metals involved in the Fenton reaction can also serve to reduce oxidative stress and ROS damage [61,62]. Indeed, overexpression of MT confers protection of cells against Cu-dependent lipid peroxidation and cytotoxicity [63] and reduces UV-induced apoptosis in keratinocytes both in vivo and in vitro [64]. Moreover, epidermis from MT-null mice is more susceptible to the formation of sunburn cells following exposure to UV [64]. In view of these findings, it seems plausible that up-regulation of epidermal MT gene expression can account, at least partially, for the antioxidant effect of 1,25(OH)2D3 in the skin [54,56]. Treatment with 1,25(OH)2D3 increased MT mRNA levels in cultured keratinocytes but also in liver, kidney, and skin when applied in vivo [65,66]. The effect of the hormone was apparent within 2 hr and maximal after 24 hr and was not dependent on protein synthesis, indicating a direct genomic effect. These findings suggest that 1,25(OH)2D3 may have antioxidant activity in the liver and kidney under conditions when MT levels become rate limiting. It is intriguing that the most studied and the bestcharacterized activity of 1,25(OH)2D3, namely, stimulation of intestinal calcium absorption, may also be related to its antioxidant properties. Depletion of glutathione

767

CHAPTER 45 Vitamin D and the Cellular Response to Oxidative Stress

pools produced a rapid and reversible inhibition of 1,25(OH)2D3-induced calcium transfer from lumen to plasma and of 1,25(OH)2D3-dependent enhancement of the activity of Ca-ATPase in the intestinal basolateral membranes. It has been suggested that this inhibition is due to reversible oxidation of SH groups in the Ca2+ transporter [67]. This surmise is supported by finding that the number of reduced SH groups in brush border membrane preparations prepared from vitamin D-deficient chicks increased twofold by a short-term in vivo treatment with 1,25(OH)2D3 [68,69]. The rapid action of 1,25(OH)2D3 suggests that its antioxidant effect in this system is mediated via a nongenomic mechanism.

IV. DISCUSSION The impressive body of experimental evidence laid out in this chapter attests to the ability of vitamin D derivatives to modulate the cellular systems responsible for redox homeostasis and the response to oxidative stress. Since the cellular redox state and antioxidant systems have important roles in the regulation of cell metabolism and in the determination of cell fate, some effects of 1,25(OH)2D3 such as modulation of cell proliferation, differentiation, and programmed cell death

TABLE II Target molecule Cu,Zn-SOD Mn-SOD γ-Glutamyl transpeptidase Thioredoxin reductase 1 Metallothionein

VDUP1 Glutathione (total)

GSSG/GSH ratio ROS levels

have been attributed to its redox modulating activity. Although a causal relationship between redox modulation by vitamin D and biological outcome was established in some experimental systems [16,18,20,23], most of the available evidence for such an association is correlative in nature. Keeping in mind the genomic mode of action of 1,25(OH)2D3, it is not surprising that efforts have been made to discover vitamin D target genes, the expression of which is up- or down-regulated by the hormone in a way that may account for its pro- or antioxidant actions. The current available knowledge regarding such putative redox regulators is described in the text and summarized in Table II. Some of these genes are directly affected by calcitriol (metallothionein and VDUP1), while the modulation of others requires long incubation times suggesting an indirect action of the hormone. No indication of the existence of one master gene that mediates the redox action of 1,25(OH)2D3 emerges from the available data. However, taking into account the diversity of redox-related genes modulated by 1,25(OH)2D3, it is plausible that the redox activity of the hormone is only one facet of a more general stress response. Transcriptome studies will probably help in answering this question. The functional association between the modulation of specific gene expression and the biological outcome awaits the use of specific inhibitors or gene knockout models.

Effect of Vitamin D on ROS Levels and Redox-Associated Molecules Cell type or line

Direction of effect

Consequence

References

MCF-7 (breast cancer) U937 (myelomonocytic leukemia) Prostate cells (normal and tumor) Astrocytes and pericytes Prostate cells (normal and tumor) Epidermis, keratinocytes Prostate cells (normal) Liver, kidney Prostate cells (tumor) B16F10 (mouse melanoma) Astrocytes Mesencephalic neurons MCF-7 (breast cancer) HL-60 (promyelocytic leukemia) MCF-7 (breast cancer) HL-60 (promyelocytic leukemia) Mesencephalic neurons MCF-7 (breast cancer) B16F10 (mouse melanoma)

Decrease Increase Increase Increase Increase Increase Increase Increase Decrease Increase Increase Increase Decrease Increase Increase Decrease Decrease Increase Increase

Prooxidant Antioxidant Antioxidant Antioxidant Antioxidant Antioxidant Antioxidant Antioxidant Prooxidant Prooxidant Antioxidant Antioxidant Prooxidant Antioxidant Prooxidant Antioxidant Antioxidant Prooxidant Prooxidant

[18] [53] [70] [49,50] [70] [54,56,65] [70] [65,66] [70] [38] [50] [48] [20] [51] [26] [51] [47,48] [30] [38]

768 An intriguing feature of 1,25(OH)2D3 is its ability to exert both prooxidant and antioxidant effects in its target cells. This may occur even in the same cells as exemplified by its prooxidant and antioxidant activities in U937 cells treated with H2O2 or TNF, respectively [15,53]. Some possible explanations for the opposite effects of 1,25(OH)2D3 are: (1) 1,25(OH)2D3 can both increase and decrease the level of different antioxidant proteins in the same cell as was shown in malignant prostate cells (decrease in MT and increase in thioredoxin reductase 1 and Mn-SOD levels) [70]. The outcome of this complex hormonal effect will depend on the role the different antioxidants have in various cellular scenarios. (2) 1,25(OH)2D3 can modulate the level of the same antioxidant in opposite directions in different cell types (e.g., up-regulation of MT in normal prostate cells and down-regulation in malignant prostate cells [70]). (3) The same effect of 1,25(OH)2D3 on a certain antioxidant may have opposite consequences, pro- or antioxidative, depending on the cellular context (e.g., decrease in SOD level may have either prooxidant or antioxidant effects depending on the capacity of the cellular H2O2 degrading systems and increase in γ-GT activity may result in enhanced GSH synthesis on the one hand and increase in H2O2 generation, on the other hand [71]). (4) A primary prooxidant effect of 1,25(OH)2D3 can induce an adaptive response that will increase the antioxidant capacity of the cell. An example that may be relevant in this context is the increase in the cellular content of glucose-6-phosphate dehydrogenase following treatment of breast cancer cells with 1,25(OH)2D3 [26,72]. Glucose-6-phosphate dehydrogenase is the rate-limiting enzyme in the pentose phosphate pathway, which supplies the cell with NADPH, and its up-regulation is part of the adaptive cellular response to oxidative challenge [73]. The above discussion reflects the fact that the study of the effect of vitamin D derivatives on ROS balance and handling and on redox homeostasis is a developing field and that many unexplored areas remain and many questions are still unanswered.

Acknowledgments We would like to acknowledge support of the Israel Science Foundation grants no. 684/931 and 601/99.

References 1. Slater AF, Nobel CS, Orrenius S 1995 The role of intracellular oxidants in apoptosis. Biochim Biophys Acta 1271:59–62.

RUTH KOREN AND AMIRAM RAVID

2. Jacobson MD 1996 Reactive oxygen species and programmed cell death. Trends Biochem Sci 21:83–86. 3. Simon HU, Haj-Yehia A, Levi-Schaffer F 2000 Role of reactive oxygen species (ROS) in apoptosis induction. Apoptosis 5:415–418. 4. Mates M 2000 Effects of antioxidant enzymes in the molecular control of reactive oxygen species toxicology. Toxicology 153:83–104. 5. Nordberg J, Arner ES 2001 Reactive oxygen species, antioxidants, and the mammalian thioredoxin system. Free Radic Biol Med 31:1287–1312. 6. Forman HJ, Torres M 2001 Redox signaling in macrophages. Mol Aspects Med 22:189–216. 7. Schafer FQ, Buettner GR 2001 Redox environment of the cell as viewed through the redox state of the glutathione disulfide/glutathione couple. Free Radic Biol Med 30: 1191–1212. 8. Sen CK, Packer L 1996 Antioxidant and redox regulation of gene transcription. FASEB J 10:709–720. 9. Sun Y, Oberley LW 1996 Redox regulation of transcriptional activators. Free Radic Biol Med 21:335–348. 10. Lander HM 1997 An essential role for free radicals and derived species in signal transduction. FASEB J 11:118–124. 11. Nakamura H, Nakamura K, Yodoi J 1997 Redox regulation of cellular activation. Annu Rev Immunol 15:351–369. 12. Finkel T 1998 Oxygen radicals and signaling. Curr Opin Cell Biol 10:248–253. 13. Kamata H, Hirata H 1999 Redox regulation of cellular signaling. Cell Signal 11:1–14. 14. Kroncke KD, Klotz LO, Suschek CV, Sies H 2002 Comparing nitrosative versus oxidative stress toward zinc finger-dependent transcription. Unique role for NO. J Biol Chem 277: 13294–13301. 15. Polla BS, Bonventre JV, Krane SM 1988 1,25-Dihydroxyvitamin D3 increases the toxicity of hydrogen peroxide in the human monocytic line U937: The role of calcium and heat shock. J Cell Biol 107:373–380. 16. Ravid A, Koren R 2003 The role of reactive oxygen species in the anticancer activity of vitamin D. Recent Results Cancer Res 164:357–367. 17. Chuang YY, Chen Y, Gadisetti, Chandramouli VR, Cook JA, Coffin D, Tsai MH, DeGraff W, Yan H, Zhao S, Russo A, Liu ET, Mitchell JB 2002 Gene expression after treatment with hydrogen peroxide, menadione, or t-butyl hydroperoxide in breast cancer cells. Cancer Res 62:6246–6254. 18. Ravid A, Rocker D, Machlenkin A, Rotem C, Hochman A, Kessler-Icekson G, Liberman UA, Koren R 1999 1,25Dihydroxyvitamin D3 enhances the susceptibility of breast cancer cells to doxorubicin-induced oxidative damage. Cancer Res 59:862–867. 19. Rocker D, Ravid A, LibermanUA, Garach-Jehoshua O, Koren R 1994 1,25-Dihydroxyvitamin D3 potentiates the cytotoxic effect of TNF on human breast cancer cells. Mol Cell Endocrinol 106:157–162. 20. Koren R, Rocker D, Kotestiano O, Liberman UA, Ravid A 2000 Synergistic anticancer activity of 1,25-dihydroxyvitamin D3 and immune cytokines: The involvement of reactive oxygen species. J Steroid Biochem Mol Biol 73:105–112. 21. Colston KW, Hansen CM 2002 Mechanisms implicated in the growth regulatory effects of vitamin D in breast cancer. Endocr Relat Cancer 9:45–59. 22. Yacobi R, Koren R, Liberman UA, Rotem C, Wasserman L, Ravid A 1996 1,25-Dihydroxyvitamin D3 increases the sensitivity of human renal carcinoma cells to tumor necrosis factor

CHAPTER 45 Vitamin D and the Cellular Response to Oxidative Stress

23.

24. 25. 26. 27. 28.

29. 30. 31.

32.

33. 34. 35.

36. 37.

38.

39. 40.

alpha but not to interferon alpha or lymphokine-activated killer cells. J Endocrinol 149:327–333. Weitsman GE, Ravid A, Liberman UA, Koren R 2003 Vitamin D enhances caspase-dependent and -independent TNFα-induced breast cancer cell death: The role of reactive oxygen species and mitochondria. Int J Cancer 106:178–186. Zager RA 1999 Calcitriol directly sensitizes renal tubular cells to ATP-depletion and iron-mediated attack. Am J Pathol 154:1899–1909. Sundaram S, Gewirtz DA 1999 The vitamin D3 analog EB 1089 enhances the response of human breast tumor cells to radiation. Radiat Res 152:479–486. Koren R, Hadari-Naor I, Zuck E, Rotem C, Liberman UA, Ravid A 2001 Vitamin D is a prooxidant in breast cancer cells. Cancer Res 61:1439–1444. Clive DR, Greene JJ 1996 Cooperation of protein disulfide isomerase and redox environment in the regulation of NF-κB and AP1 binding to DNA. Cell Biochem Funct 14:49–55. Broekemeier KM, Klocek CK, Pfeiffer DR 1998 Proton selective substrate of the mitochondrial permeability transition pore: Regulation by the redox state of the electron transport chain. Biochemistry 37:13059–13065. Hampton MB, Fadeel B, Orrenius S 1998 Redox regulation of the caspases during apoptosis. Ann NY Acad Sci 854:328–335. Narvaez CJ, Welsh J 2001 Role of mitochondria and caspases in vitamin D-mediated apoptosis of MCF-7 breast cancer cells. J Biol Chem 276:9101–9107. Doroshow JH, Akman S, Esworthy S, Chu FF, Burke T 1991 Doxorubicin resistance conferred by selective enhancement of intracellular glutathione peroxidase or superoxide dismutase content in human MCF-7 breast cancer cells. Free Radic Res Commun 12–13(Pt. 2):779–781. Manna SK, Zhang HJ, Yan T, Oberley LW, Aggarwal BB 1998 Overexpression of manganese superoxide dismutase suppresses tumor necrosis factor induced apoptosis and activation of nuclear transcription factor-κB and activated protein-1. J Biol Chem 273:13245–13254. Squadrito GL, Pryor WA 1998 Oxidative chemistry of nitric oxide: the roles of superoxide, peroxynitrite, and carbon dioxide. Free Radic Biol Med 25:392–403. Minotti G 1993 Sources and role of iron in lipid peroxidation. Chem Res Toxicol 6:134–146. Barrett WC, DeGnore JP, Konig S, Fales HM, Keng YF, Zhang ZY, Yim MB, Chock PB 1999 Regulation of PTP1B via glutathionylation of the active site cysteine 215. Biochemistry 38:6699–6705. Chen KS, DeLuca HF 1994 Isolation and characterization of a novel cDNA from HL-60 cells treated with 1,25-dihydroxyvitamin D3. Biochim Biophys Acta 1219:26–32. Nishiyama A, Matsui M, Iwata S, Hirota K, Masutani H, Nakamura H, Takagi Y, Sono H, Gon Y, Yodoi J 1999 Identification of thioredoxin-binding protein-2/vitamin D3 up-regulated protein 1 as a negative regulator of thioredoxin function and expression. J Biol Chem 274:21645–21650. Song H, Cho D, Jeon JH, Han SH, Hur DY, Kim YS, Choi I 2003 Vitamin D3 up-regulating protein 1 (VDUP1) antisense DNA regulates tumorigenicity and melanogenesis of murine melanoma cells via regulating the expression of fas ligand and reactive oxygen species. Immunol Lett 86:235–247. Perry TL, Yong VW 1986 Idiopathic Parkinson’s disease, progressive supranuclear palsy and glutathione metabolism in the substantia nigra of patients. Neurosci Lett 67:269–274. Nakamura K, Wang W, Kang UJ 1997 The role of glutathione in dopaminergic neuronal survival. J Neurochem 69:1850–1858.

769

41. Sawada H, Shimohama S, Kawamura T, Akaike A, Kitamura Y Taniguchi T, Kimura J 1996 Mechanism of resistance to NO-induced neurotoxicity in cultured rat dopaminergic neurons. J Neurosci Res 46:509–518. 42. Sawada H, Kawamura T, Shimohama S, Akaike A, Kimura J 1996 Different mechanisms of glutamate-induced neuronal death between dopaminergic and non-dopaminergic neurons in rat mesencephalic culture. J Neurosci Res 43:503–510 43. Sawada H, Shimohama S, Tamura Y, Kawamura T, Akaike A, Kimura J 1996 Methylphenylpyridium ion (MPP+) enhances glutamate-induced cytotoxicity against dopaminergic neurons in cultured rat mesencephalon. J Neurosci Res 43:55–62. 44. Riobo NA, Schopfer FJ, Boveris AD, Cadenas E, Poderoso JJ 2002 The reaction of nitric oxide with 6-hydroxydopamine: implications for Parkinson’s disease. Free Radic Biol Med 32:115–121. 45. Nakai M, Mori A, Watanabe A, Mitsumoto Y 2003 1-Methyl4-phenylpyridinium (MPP+) decreases mitochondrial oxidation–reduction (REDOX) activity and membrane potential (Deltapsi(m)) in rat striatum. Exp Neurol 179:103–110. 46. Wang JY, Wu JN, Cherng TL, Hoffer BJ, Chen HH, Borlongan CV, Wang Y 2001 Vitamin D3 attenuates 6-hydroxydopamine-induced neurotoxicity in rats. Brain Res 904: 67–75. 47. Ibi M, Sawada, H, Nakanishi M, Kume T, Katsuki H, Kaneko S, Shimohama S, Akaike A 2001 Protective effects of 1α,25(OH)2D3 against the neurotoxicity of glutamate and reactive oxygen species in mesencephalic culture. Neuropharmacology 40:761–771. 48. Shinpo K, Kikuchi S, Sasaki H, Moriwaka F, Tashiro K 2000 Effect of 1,25-dihydroxyvitamin D3 on cultured mesencephalic dopaminergic neurons to the combined toxicity caused by L-buthionine sulfoximine and l-methyl-4-phenylpyridine. J Neurosci Res 62:374–382. 49. Garcion E, Thanh XD, Bled F, Teissier E, Dehouck MP, Rigault F, Brachet P, Girault A, Torpier G, and Darcy F 1996 1,25-Dihydroxyvitamin D3 regulates γ1 transpeptidase activity in rat brain. Neurosci Lett 216:183–186. 50. Garcion E, Sindji L, Leblondel G, Brachet P, Darcy F 1999 1,25-Dihydroxyvitamin D3 regulates the synthesis of gammaglutamyl transpeptidase and glutathione levels in rat primary astrocytes. J Neurochem 73:859–866. 51. Danilenko M, Wang Q, Wang X, Levy J, Sharoni Y, Studzinski GP 2003 Carnosic acid potentiates the antioxidant and prodifferentiation effects of 1α,25-dihydroxyvitamin D3 in leukemia cells but does not promote elevation of basal levels of intracellular calcium. Cancer Res 63:1325–1332. 52. Sokoloski JA, Hodnick WF, Mayne ST, Cinquina C, Kim CS, Sartorelli AC 1997 Induction of the differentiation of HL-60 promyelocytic leukemia cells by vitamin E and other antioxidants in combination with low levels of vitamin D3: Possible relationship to NF-kappaB. Leukemia 11:1546–1553. 53. Iwamoto S, Takeda K, Kamijo R, Konno K 1990 Induction of resistance to TNF cytotoxicity and mitochondrial superoxide dismutase on U-937 cells by 1,25-dihydroxyvitamin D3. Biochem Biophys Res Commun 170:73–79. 54. Hanada K, Sawamura D, Nakano H, Hashimoto I 1995 Possible role of 1,25-dihydroxyvitamin D3-induced metallothionein in photoprotection against UVB injury in mouse skin and cultured rat keratinocytes. J Dermatol Sci 9:203–208. 55. Youn JI, Park BS, Chung JH, Lee JH 1997 Photoprotective effect of calcipotriol upon skin photoreaction to UVA and UVB. Photodermatol Photoimmunol Photomed 13: 109–114.

770 56. Lee J, Youn JI 1998 The photoprotective effect of 1,25-dihydroxyvitamin D3 on ultraviolet light B-induced damage in keratinocyte and its mechanism of action. J Dermatol Sci 18:11–18. 57. De Haes P, Garmyn M, Degreef H, Vantieghem K, Bouillon R, Segaert S 2003 1,25-Dihydroxyvitamin D3 inhibits ultraviolet B-induced apoptosis, Jun kinase activation, and interleukin-6 production in primary human keratinocytes. J Cell Biochem 89:663–673. 58. Tyrrell RM, Keyse SM 1990 New trends in photobiology. The interaction of UVA radiation with cultured cells. J Photochem Photobiol B 4:349–361. 59. Black HS 1987 Potential involvement of free radical reactions in ultraviolet light-mediated cutaneous damage. Photochem Photobiol 46:213–221. 60. Ravid A, Rubinstein E, Gamady A, Rotem C, Liberman UA, Koren R 2002 Vitamin D inhibits the activation of stressactivated protein kinases by physiological and environmental stresses in keratinocytes. J Endocrinol 173:525–532. 61. Lazo JS, Kuo SM, Woo ES, Pitt BR 1998 The protein thiol metallothionein as an antioxidant and protectant against antineoplastic drugs. Chem Biol Interact 111–112:255–262. 62. Viarengo A, Burlando B, Ceratto N, Panfoli I 2000 Antioxidant role of metallothioneins: a comparative overview. Cell Mol Biol (Noisy-le-grand) 46:407–417. 63. Fabisiak JP, Pearce LL, Borisenko GG, Tyhurina YY, Tyurin VA, Razzack J, Lazo JS, Pitt BR, Kagan VE 1999 Bifunctional anti/ prooxidant potential of metallothionein: redox signaling of copper binding and release. Antioxid Redox Signal 1:349–364. 64. Hanada K, Sawamura D, Tamai K, Baba T, Hashimoto I, Muramatsu T, Miura N, Naganuma A 1998 Novel function of metallothionein in photoprotection: metallothionein-null mouse exhibits reduced tolerance against ultraviolet B injury in the skin. J Invest Dermatol 111:582–585. 65. Karasawa M, Hosoi J, Hashiba H, Nose K, Tohyama C, Abe E, Suda T, Kuroki T 1987 Regulation of metallothionein gene expression by 1α,25-dihydroxyvitamin D3 in cultured cells and in mice. Proc Natl Acad Sci USA 84:8810–8813.

RUTH KOREN AND AMIRAM RAVID

66. Chou SY, Hannah SS, Lowe KE, Norman AW, Henry HL 1995 Tissue-specific regulation by vitamin D status of nuclear and mitochondrial gene expression in kidney and intestine. Endocrinology 136:5520–5526. 67. Tolosa de Talamoni N, Marchionatti A, Baudino V, Alisio A 1996 Glutathione plays a role in the chick intestinal calcium absorption. Comp Biochem Physiol A. Physiol 115: 127–132. 68. Mykkanen HM, Wasserman RH 1990 Relationship of membrane-bound sulfhydryl groups to vitamin D–stimulated uptake of [75Se]Selenite by the brush border membrane vesicles from chick duodenum. J Nutr 120:882–888. 69. Mykkanen HM, Wasserman RH 1990 Reactivity of sulfhydryl groups in the brush-border membranes of chick duodena is increased by 1,25-dihydroxycholecalciferol. Biochim Biophys Acta 1033:282–286. 70. Krishnan AV, Peehl DM, Feldman D 2003 Inhibition of prostate cancer growth by vitamin D: Regulation of target gene expression. J Cell Biochem 88:363–371. 71. Dominici S, Valentini M, Maellaro E, Del Bello, B, Paolicchi, A, Lorenzini E, Tongiani R, Comporti M, Pompella A 1999 Redox modulation of cell surface protein thiols in U937 lymphoma cells: The role of gamma-glutamyl transpeptidasedependent H2O2 production and S-thiolation. Free Radic Biol Med 27:623–635. 72. Noun A, Garabedian M, Monet JD 1989 Stimulatory effect of 1,25-dihydroxyvitamin D3 on the glucose-6-phosphate dehydrogenase activity in the MCF-7 human breast cancer cell line. Cell Biochem Funct 7:1–6. 73. Ursini MV, Parrella A, Rosa G, Salzano S, Martini G 1997 Enhanced expression of glucose-6-phosphate dehydrogenase in human cells sustaining oxidative stress. Biochem J 323(Pt 3): 801–806. 74. Wang Q, Yang W, Uytingco MS, Christakos S, Wieder R 2000 1,25-Dihydroxyvitamin D3 and all-trans-retinoic acid sensitize breast cancer cells to chemotherapy-induced cell death. Cancer Res 60:2040–2048.

CHAPTER 46

Vitamin D: Role in the Calcium Economy ROBERT P. HEANEY I. II. III. IV.

Creighton University, Omaha, Nebraska

Introduction Overview of the Calcium Economy Calcium Absorptive Input Physiological Sources of Vitamin D Activity

I. INTRODUCTION Vitamin D functions in many body systems, but perhaps the best attested of the nutrient’s actions—and certainly the one most clearly associated with human disease—is its role in transferring calcium (and phosphorus) from ingested food into the body fluids. Calcium, like most divalent cations, is only partially absorbed from the chyme as it travels through the small intestine. This situation creates an opportunity for regulation of absorption, with room both to increase and to decrease calcium extraction efficiency in response to physiological controls. Details of both the many cellular and tissue effects of vitamin D, and of the absorptive process itself, are covered in other chapters in this volume. Here I shall attempt to summarize mainly the meaning and importance of the vitamin D-mediated transfer process from gut to blood and to outline how it fits into the maintenance of the calcium economy. My frame of reference will be the integrated functioning of the intact organism.

V. Optimal Vitamin D Status VI. Summary and Conclusions References

apatite lattice with variable stoichiometry, and embedded in a dense protein matrix. Although cells (osteocytes) ramify throughout bony tissue, the intercellular bony material itself lacks appreciable free water. As a result there is very limited exchange of calcium ions between the bone and the circulating body fluids. Isotopic exchange with tracers injected into the blood is confined to the surface layer of crystals in the bone situated along vascular channels and spaces, and to still incompletely mineralized new forming sites. Taken all together the exchangeable bone calcium moieties amount to only about 25 mmol (1000 mg), or ~0.1% of total skeletal calcium [1]. Moreover, the insolubility of bone mineral is such that, even when there is exchange, there is virtually never net transfer out of bone into the body fluids. Net transfer normally requires formation or resorption of bone tissue.1 A second, biologically critical compartment is intracellular calcium. Here calcium serves as a ubiquitous second messenger, linking signals from outside the cell to the mechanisms constituting the cell’s response. While free calcium ion concentrations in the cytosol

II. OVERVIEW OF THE CALCIUM ECONOMY A. Body Calcium Compartments Body calcium in an adult human amounts to about 15–20 g (0.375–0.5 mol)/kg body weight. This calcium exists in three quite distinct divisions (or compartments). They are distinct because (1) movement of calcium atoms between them is both limited and regulated; and (2) they can and do vary in magnitude independently of one another. The first and most obvious compartment is the calcium in the bones and teeth. Here calcium exists as inorganic mineral crystals, arranged in an imperfect VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

1One possible exception is the calcium carbonate of bone. Carbonate substitutes poorly for phosphate in the apatite lattice, and it is generally considered that, because of different valences and ionic radii for the two ions, carbonate is confined to crystal surfaces. Generally it is assumed that the carbonate is more or less uniformly distributed throughout the bony material. However, bone carbonate is substantially more labile than is bone calcium generally, and it is likely therefore that the carbonate is situated even more superficially than generally presumed, that is, primarily on anatomic bone surfaces, rather than diffusely on crystal surfaces generally. Calcium carbonate might, thus, be a kind of “icing” on the underlying mineralized matrix, sensitive to pH and pCO2 in the extracellular fluid. If this is the case, a limited amount of net movement of calcium into and out of bone would be possible without involving cell-mediated formation or resorption of bone tissue.

Copyright © 2005, Elsevier, Inc. All rights reserved.

774 are typically on the order of 1 × 10–5 mmol [2], total cell calcium is on the order of 0.5–2 mmol/kg tissue [1]. Most of this quantity (~99.99%) is bound to specialized calcium storage proteins (e.g., sequestrin, parvalbumin, calbindin) and located in storage vesicles, typically specialized units of the endoplasmic reticulum. Because the calcium ion is of just the right radius to fit neatly into folds of the peptide chain, and because calcium is capable of forming up to 12 (typically 6–8) coordination bonds with oxygen atoms in the side chains of amino acids projecting off the peptide chain, calcium stabilizes the tertiary structure of many catalytic molecules, thereby activating them. Cytosolic [Ca2+] must be kept very low to prevent constant, uncontrolled activity of the many cell functions in which calcium plays a messenger or activation role. On the other hand, substantial intracellular stores of calcium are necessary because the binding avidity of cytosolic proteins for calcium is so high that, typically, the free path of a calcium ion in the cytosol is only a tiny fraction of a cell diameter [2]. If reservoirs were not diffusely distributed throughout the cytosol, and if extracellular calcium were the only source for this critical second messenger function, activation would be limited to the zone immediately beneath the plasma membrane. Because the cell membrane is relatively impermeable to calcium ions, and because the cytosolic [Ca2+] compartment is so tiny, tracer exchange between ECF calcium and cell stores of calcium is surprisingly slow— typically requiring hours or days to come into tracer equilibrium [1]. Further, cell calcium levels are typically unaffected by acute changes in calcium concentrations in the extracellular fluid (ECF). The third and smallest division of body calcium is the calcium present in the circulating blood and the ECF bathing all the body tissues. This compartment contains typically about 0.4 mmol calcium per kg body weight. Ionized calcium concentration in these fluids is ordinarily about 1.25 mmol/liter (5 mg/dl), a figure that is regulated across the higher vertebrate orders with the same exquisite precision as are, e.g., the concentrations of sodium and potassium. Departures from this level produce well-studied, significant effects on interneuronal signal transmission and on muscular excitability. ECF [Ca2+] is also important for supracellular protein activations such as those in the coagulation cascade. Into and out of this compartment passes all the calcium entering and leaving the body from the outside, as well as entering and leaving bone. These fluxes are summarized schematically in Fig. 1. Together they involve daily quantities amounting to 35–50% of the size of the entire compartment in healthy adults, and to several times that compartment size in infants. Without tight regulation, ECF [Ca2+] would oscillate between

ROBERT P. HEANEY

possibly fatal extremes of hypo- and hypercalcemia as the organism goes from fasting to feeding.

B. Regulation of ECF [Ca2+] A detailed description of the processes involved in regulating ECF [Ca2+] is beyond the purpose of this chapter. To situate vitamin D in this system it is necessary only to note that regulation occurs both by controlling the renal excretory threshold for calcium and by regulating calcium fluxes into and out of the ECF. Two of the transfers out of the ECF illustrated in Fig. 1 drive the system and are essentially unregulated. The other transfers are at least partially responsive and constitute the basis of the regulatory control of ECF [Ca2+]. The unregulated, driving transfers are (1) mineralization of newly deposited bone matrix at bone forming sites and (2) daily obligatory loss out of the body. Vitamin D and its metabolites play critical roles in the two regulated transfers into the ECF—from ingested food and from bone resorption. 1. DRIVING TRANSFERS

a. Mineralization of Bone Mineralization of bone constitutes an unregulated drain because it is a passive phenomenon, lagging several days, and even weeks, behind the osteoblastic cellular activity that initiates the process. The growing mineral deposits extract calcium and phosphorus from blood flowing past the mineralizing site at a decreasing exponential rate for up to 40 weeks after the matrix has been deposited. The only way, in theory, to stop the process is to shut down blood flow to bone. The magnitude of this mineralization drain varies with skeletal remodeling activity and with body size. It amounts to about 0.15 mmol/d/kg body weight in healthy mature adults. It is much higher during growth, of course, and rises once again after mid life [3,4]. b. Obligatory Loss Obligatory loss consists of a combination of cutaneous loss and the fixed components of urinary and endogenous fecal calcium excretion. Cutaneous losses consist not just of sweat calcium but of the calcium contained in shed skin, hair, and nails. As has been noted above, all cells contain substantial amounts of calcium, and their loss from body surfaces inexorably takes calcium with them. Cutaneous losses have been hard to quantify but are estimated to be at least 0.4 mmol/day and more likely closer to 1.5 mmol/day [5,6]. Much higher losses have been reported with vigorous physical activity [7]. The fixed component of endogenous fecal and urinary calcium losses is somewhat more complicated [8]. On average, 3.5–4.0 mmol calcium enters the gut each day

775

CHAPTER 46 Vitamin D: Role in the Calcium Economy

Bone Diet Ca 20.0 Formation 9.0

Resorption 9.0 Skin

3.5 1.5

Extracellular fluid 0.8 25 mmol

5.0

Reabsorbed 197.5

Filtered 200

Fecal Ca 17.7 Urine Ca 3.0

FIGURE 1 Schematic representation of the principal inputs and outputs of the extracellular fluid compartment. Units are in Ca mmol for extracellular fluid and mmol Ca/day for all other entries. (Copyright Robert P. Heaney, 2004. Reproduced with permission.)

from endogenous sources, principally as a component of digestive secretions, but also as the calcium contained in shed mucosal cells (which turn over once every 4–5 days). The precise quantity varies directly with body size and with the amount of food consumed. For reasons not understood, the amount of calcium entering the gut in the digestive secretions varies directly also with phosphorus intake [8]. In any event, this endogenous calcium mixes with food calcium and much of it is subject to absorption, just as is the calcium of food. However, as noted earlier, calcium absorption is always incomplete. Gross absorption averages about 25–30% in healthy adults, and net absorption about 10–11%. Hence much of the secreted calcium is lost in the feces. Moreover, some of the endogenous calcium secretion enters the intestinal stream effectively so low in the gut as to be essentially unreclaimable (e.g., the calcium in colon mucus and in shed colon cells). On a normal diet this distal component has been estimated to be about 0.6 mmol [8]. Typical endogenous fecal calcium values average in the range of 3 mmol/day. Since absorption efficiency in adults is essentially never above 60%,

even on a very low calcium diet (see later discussion), there is an irreducible minimum loss of endogenous calcium through the gut averaging close to 2 mmol/day. If gross absorption is not greater than this figure, the gut becomes a net excretory organ for calcium. The third obligatory loss is through the kidney. It is generally held that renal calcium excretion is controllable, but this is only partly true. PTH certainly regulates tubular calcium reabsorption. However, there is a fixed limit to what that mechanism can accomplish. This limit is itself a function of other variables that are outside the regulatory loop. Best studied of these factors is the renal excretion of sodium, dietary intakes of protein and potassium, net endogenous acid production (NEAP), and absorbed dietary phosphorus. Sodium [9–11], protein [12,13], and NEAP [14] increase urinary calcium loss; phosphorus and potassium decrease it [15,16]. Because sodium and calcium compete for the same reabsorption mechanism in the proximal tubule, the two ions influence one another’s excretion [17]. On average, urine calcium increases by from 0.5 to 1.5 mmol for

776 every 100 mmol of sodium excreted [9,10]. Similarly, urine calcium rises by about 0.25 mmol for every 10 g of protein ingested [12]. The result, for an adult woman ingesting the RDA for protein and the median sodium intake for North Americans, is a level of obligatory urinary loss amounting to about 2 mmol/day. Reducing sodium intake would certainly reduce this obligatory loss. Nevertheless such voluntary dietary change is clearly not a part of any physiological regulatory loop. And thus, to the extent that sodium intake influences obligatory calcium loss, it constitutes a demand to which the calcium homeostatic system must respond. Given typical adult diets in Europe and North America, the sum of these obligatory losses through skin, gut, and kidney is about 5 ± 1 mmol/day, or about onefifth of the total calcium in the ECF. To offset these losses (plus the demands of bone mineralization), the organism regulates countervailing inputs into the ECF from food and bone. It is in these transfers that vitamin D plays its role. The input from the first source is dependent both upon the presence of food in the upper GI tract and the presence of sufficient calcium in that food. Because neither condition can be guaranteed, the second source, bone, is the more reliable and constitutes, in fact, the first line of defense against hypocalcemia.2 2. RESPONSE OF THE SYSTEM TO DEMAND

a. PTH-Mediation of Response Briefly, a fall in ECF [Ca2+] evokes a prompt rise in parathyroid hormone [PTH] release. PTH acts in a classical negative feedback loop to raise the ECF [Ca2+], thereby closing the loop and reducing PTH release. The mechanisms by which PTH raises ECF [Ca2+] illustrate well the complexities of the calcium economy. These mechanisms include: (1) increasing renal phosphate clearance, thereby lowering ECF phosphate levels; (2) increasing renal tubular reabsorption of calcium, thereby allowing system inputs to elevate ECF [Ca2+]; (3) augmenting osteoclast work at existing resorption loci; (4) activating new bone resorption loci; and (5) increasing the activity of the renal 1-α-hydroxylase, thereby increasing serum

2Because it is beyond the scope of this chapter, which is concerned mainly with calcium transfers, I shall not develop further what is actually an even more fundamental mechanism by which PTH regulates ECF [Ca2+], that is, the control of the renal calcium threshold (which is covered in more detail elsewhere [18]). Suffice it to say here only that in the shift from a lower to higher PTH level, renal calcium losses are temporarily curtailed until calcium inputs from bone and gut succeed in elevating [Ca2+]. Then, as the filtered load at the glomerulus rises, calcium excretion returns to its previous level (although renal calcium clearance is now reduced).

ROBERT P. HEANEY

levels of 1,25(OH)2D and augmenting absorption efficiency for ingested food calcium. These five effects reinforce one another in important ways. The earliest effects, occurring within minutes, are a decrease in renal tubular phosphate reabsorption and the resulting fall in serum phosphate. The latter immediately augments existing osteoclastic bone resorption [19] and increases activity of the renal 1-α-hydroxylase [20]. The elevated production of 1,25(OH)2D leads to increased intestinal absorption, elevating ECF [Ca2+] and thereby closing the feedback loop. [1,25(OH)2D also suppresses parathyroid hormone release in its own right.] Finally, 1,25(OH)2D is necessary for efficient osteoclast work. In this last role, it is not known whether variations of 1,25(OH)2D in the physiologic range produce corresponding alterations in bone resorption, and it is a difficult question to study because of the tight regulation of the various components of the system. Nevertheless, it is well established that resorptive work is severely impaired in vitamin D deficiency states (see Fig. 2 and later discussion), and that 1,25(OH)2D in supraphysiologic doses is capable of causing substantial increases in bone resorption. Finally, all of the components of the intestinal calcium transport system, including vitamin D receptors and calbindins, are also found in the distal tubule of the nephron [21]. It is possible, though not yet firmly established, that 1,25(OH)2D may thus enhance recovery of filtered calcium and contribute to the PTH effect of elevating the renal calcium threshold. An example of the integrated response of the system to demand is afforded by what occurs during antler formation in deer in the spring [22]. The calcium demands of antler formation exceed the calcium available in the nutrient-poor late winter foliage. Other things being equal, this drain would lower ECF [Ca2+], but the parathyroid glands, sensing minute reductions in ECF [Ca2+], respond by increasing secretion of PTH, which in turn activates numerous remodeling loci throughout the deer’s skeleton. Since remodeling is asynchronous, with resorption preceding formation at each remodeling locus, the extra burst of newly activated remodeling sites provides a temporary surplus of calcium, which the animal promptly uses to mineralize matrix in the forming antler. A few weeks later, when the skeletal remodeling loci reach their own formation phase, calcium content of the ingested foliage will be higher, and the “mineral debt” created by antler formation will be repaid from ingested greens. b. Vitamin D Deficiency As already noted, vitamin D is essential for the mobilization of calcium from the skeletal calcium reserves. This is well illustrated in the course of nutritional rickets observed in the children of China and Northern Europe before the discovery of

CHAPTER 46 Vitamin D: Role in the Calcium Economy

777

rickets. The alternating bands of high and low density reflect annual periods of greater and lesser vitamin D availability. (Reproduced from A Textbook of X-ray Diagnosis, Vol. 6, 4th edition, 1971, edited by SC Shanks, P Kerley. HK Lewis & Co., Ltd, Toronto. Chapter XLIII, Metabolic and endocrine-induced bone disease by CJ Hodson, p. 661, Fig. 774.)

mineralization of the hypertrophic cartilage zone beneath the growth plate. Then, during the 8 or so months when solar vitamin D was lacking, the rachitic process resumed. The result, visible on X-ray, was a series of bands parallel with the growth plate, with densely mineralized bone tissue alternating with undermineralized layers (Fig. 2). One might have thought that, during the winter, PTH-mediated bone resorption would have attacked bone in the dense layers, making its calcium available to new bone-forming sites (as occurs with antler formation in deer). But in the absence of vitamin D, the bone resorptive process is severely impaired. As a result the rachitic child is unable adequately to mobilize its own calcium reserves. c. Feast and Famine The diets of hominids were high in calcium [24], just as are the diets of contemporary deer and other higher mammals. Foods available to contemporary hunter-gatherers exhibit an annual mean calcium nutrient density of 1.75–2 mmol (70–80 mg)/100 kCal. For individuals of contemporary body size, doing the work of a hunter-gatherer, that value translates to calcium intakes in the range of 50–75 mmol (2000–3000 mg)/day. But, as noted, the environment could not be relied upon to supply calcium-rich food continuously. Periods of fasting, famine, or drought would undoubtedly have threatened hypocalcemia. This fact underscores the importance both of bone as a calcium reserve and of the vitamin D-parathyroid hormone control system, with its ability to release calcium rapidly from bone.3 Contemporary humans in industrialized nations have the same paleolithic physiology as our hominid ancestors. We experience, however, a few crucial differences in external conditions. One is a generally lower exposure to sunlight. Rickets was endemic in Northern Europe in the 19th century, partly because of latitude, partly because of air pollution, and partly because of child labor. Routine use of vitamin D today has all but eliminated that problem in children, but adults, and particularly the elderly, are often vitamin D– insufficient. (The consequences of this low vitamin D exposure are explored elsewhere in this volume.) Another difference is a much reduced nutrient density for calcium in our ingested food. The importance of this latter departure from primitive conditions lies in

vitamin D, and now, unfortunately, occurring once again in exclusively breast-fed children not given vitamin D supplements [23] (see Chapter 49). Prior to routine prophylactic use of vitamin D in countries where rickets was endemic, summer sun exposure produced some vitamin D—enough to allow reasonably normal

3Technically calcium is never actually “released” from bone. (See, however, footnote 1. Rather, a volume of actual bone tissue is resorbed. In the process its calcium is released into the ECF. Thus transfer of calcium out of bone always means some removal of bone tissue. Bone, however, is a very rich source of calcium. A single cubic centimeter of bone contains as much calcium as is contained in the entire circulating blood of an adult human.

FIGURE 2 X-ray of the knee in a child with vitamin D deficiency

778

ROBERT P. HEANEY

the fact that it limits the efficacy of vitamin D in augmenting input from the intestine. This limitation is not widely recognized and rarely is its quantitative impact adequately appreciated. Section III,B of this chapter develops this important component of the calcium economy in greater detail.

these two metabolites to the calcium economy, and of differences in their concentrations in ethnic groups, is still unclear (see later discussion).

3. INDEPENDENCE OF PTH EFFECTOR MECHANISMS

Chapter 24 (by Wasserman) describes the mechanisms of calcium transfer from the intestinal lumen. Here I describe quantitative aspects of that transfer as a part of the integrated calcium economy, and will focus as well on the factors determining the magnitude of the input from the gut into the ECF.

The PTH regulatory system is unique in that the feedback loop operates through three independent effector mechanisms, already described (elevated renal calcium threshold, elevated intestinal calcium absorption, and elevated bone resorption). This seeming redundancy underscores the physiological importance of maintaining constant ECF [Ca2+]. The independence of response of the three PTH effector mechanisms constitutes the substratum for still incompletely explored, but interesting, differences in the way the organism adapts to deficiency and surfeit of calcium. The feedback loop of calcium regulation is typically closed by some variable combination of all three effects. For example, when calcium intake falls, there is an obvious limit to what the absorptive mechanism can yield, forcing higher PTH levels and correspondingly greater renal calcium retention and enhanced net transfer out of bone. Relative differences in the intrinsic responsiveness of those effector organs have important consequences for bone mass [25]. Slightly lower intrinsic responsiveness of osteoclastic resorption to PTH, other things being equal, leads to a slightly higher PTH level, which, in turn, drives all three effector mechanisms slightly harder. The result is that the intestinal absorptive and renal recovery mechanisms for calcium operate at higher efficiency, and bone resorption, at lower efficiency. This leads to more bone and is probably a large part of the explanation for the higher bone mass in persons of black African ancestry. Although there has been some inconsistency in the data reported to date in this regard, the bulk of the evidence points to relative refractoriness to PTHstimulated bone resorption in African Americans [26–32]. For example, Dawson-Hughes et al. [33] showed that blacks exhibited a greater response to dietary calcium reduction than did whites, with larger increments of both PTH and 1,25(OH)2D, indicating a black–white difference in bony response (see Chapter 47). Clear evidence in this regard also comes from the calcium tracer studies of Abrams and his colleagues, showing higher calcium absorption and retention in adolescent black females than in whites at the same pubertal stage [31]. There seem also to be differences in vitamin D metabolism in the two ethnic groups, with blacks showing typically lower serum 25OHD levels and higher 1,25(OH)2D levels than whites [32]. The relative importance of

III. CALCIUM ABSORPTIVE INPUT

A. Location and Timing of Absorption in the Gut As noted in Chapter 24, there is a gradient of concentrations of vitamin D receptors and of calbindin in mucosa along the gut, with highest levels in the duodenum and lowest in the colon mucosa. Accordingly, the avidity (or rate) of active absorption is highest in the duodenum. It is sometimes said that absorption itself is highest there, but this is not correct. That conclusion is based on studies of isolated loops or gut sacs, where movement of the chyme along the intestine cannot occur. Absorbed quantity is the product of absorption rate and residence time; and residence time of the chyme in the duodenum is very brief. Only at very low calcium intakes (or test loads), and with maximal 1,25(OH)2D-stimulated active transport, will it be true that most of the calcium absorbed will be from the duodenum. At more usual intakes, the much longer residence time in the jejunum and ileum means that most of the quantity absorbed occurs from the lower small intestine. The importance of length of exposure to the absorptive surface is reflected in the finding that absorptive efficiency varies directly with mouth-tocecum transit time [34]. Absorption does not occur from the healthy stomach, and thus the beginning of absorption is delayed until gastric emptying begins. This, in turn, is dependent upon the character of the ingested meal or other calcium source. Emptying tends to be most rapid with small fluid ingestates and is slower with solid food and with fat. In healthy individuals ingesting light meals (such as would commonly be employed to test absorption efficiency), calcium absorption is nearly complete by 5 hours after ingestion [35]. Figure 3 presents data on the time course of absorption, using the ratio of the time-dependent apparent absorption fraction to its ultimate value in the individual being tested. As the figure shows, absorption has reached better than 80% of its

779

CHAPTER 46 Vitamin D: Role in the Calcium Economy

0.8

100

*

95

*

*

*

0.7

*

90

Absorption fraction

Calcium absorption (percent completion ± 2 SEM)

105

Late phase absorption (colonic)

85

*

80 75

End of early phase absorption (small intestine)

*significantly diff. from 100%

0 0

10

20 Time (hrs)

30

0.6 0.5 0.4 0.3 0.2 0.1

40

0.0 0

FIGURE 3 Time course of absorption (derived from Barger-Lux

ultimate value by 3 hr after ingestion, and 96% by 7 hr. There is then only a very gradual approach to completion over the next 20 hr. This last component probably reflects a small amount of colonic absorption (or, alternatively, cecal-ileal reflux, with delayed ileal absorption). It should be stressed that the percentage values in Fig. 3 refer to the quantity absorbed, not the quantity ingested. Thus, with typically only 25–30% of a load absorbed (see later discussion), the 4–5% colonic component represents absorption of only about 1% of the ingested load.

B. Absorption as a Function of Intake It has long been recognized that absorption efficiency varies inversely with intake. Figure 4 illustrates this relationship with data obtained from healthy, middle-aged women in whom absorption fraction was measured under controlled metabolic ward conditions and plotted as a function of their ingested intakes [36]. The best fit regression line through the data shows the expected rise in absorption fraction at low calcium intakes. (Note, however, that even at the lowest intakes, predicted mean absorption efficiency is only ~45%.) The higher efficiency at low intakes is traditionally attributed to adaptation, specifically to higher production of 1,25(OH)2D, with a corresponding increase in active calcium absorption. While that explanation is

20 30 Calcium intake (mmol/d)

40

50

FIGURE 4 Fractional absorption plotted as a function of usual calcium intake (in mmol/day) in 525 studies in healthy, middleaged women [36]. The solid line is the least squares regression line derived from a log-log fit. (Copyright Robert P. Heaney, 1989. Reproduced with permission.)

undoubtedly correct, it is also substantially incomplete. This is shown by the data in Fig. 5, which plots nonadaptive absorption fraction as a function of a broad range of calcium load sizes. These studies were performed in women, assigned randomly on any given morning to intake loads spanning a 30-fold range, from 0.4 to 12.5 mmol [37]. Clearly, an inverse relationship

0.9 0.8 0.7 Absorption fraction

et al. [35]). The data plot the percentage completion of absorption (derived from expressing the double-isotope absorption fraction as a ratio of its value at any given time to its final value after 24 hr. Completion of absorption is thus expressed as a value of 100%. As the curve shows, absorption is about 94% complete by 5 hr after oral ingestion. The remaining 6% occurs more slowly and may be presumed to reflect absorption from the colon and/or from ileo-cecal reflux. (Copyright Robert P. Heaney, 1966. Reproduced with permission.)

10

0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.1

1 Calcium load (mmol)

10

FIGURE 5 Fractional absorption in nonadapted healthy women for loads ranging from 0.4 to 12.5 mmol Ca. Error bars are ±2 SEM. The solid line is the least squares regression through the actual data (n = 75), derived from Heaney et al. [37]. The units of the horizontal axis are natural logarithms of load size (in mmol Ca). (Copyright Robert P. Heaney, 1996. Reproduced with permission.)

ROBERT P. HEANEY

0.7

0.6

0.6

0.5

Adapted women

0.4 0.3

Non-adapted women

0.2

9 8 7

0.5

6

0.4

5

0.3

4 3

0.2

2 0.1

0.1

1

0.0

0 0

0.0 0

5

10 15 Calcium intake (mmol)

20

25

FIGURE 6

Combination of the regression lines from Figs. 4 and 5, showing the extent of difference produced by adaptation to the various intakes. (Copyright Robert P. Heaney, 1996. Reproduced with permission.)

is present, just as in the data of Fig. 4. Equally clearly, it cannot be due to adaptation, since the test load was the first exposure these women had to the intake level concerned. Figure 6 plots these two sets of data together and shows that, while both exhibit an inverse relationship between absorption and intake, there is in fact a difference between them, with the adapted women absorbing more at low intakes than the nonadapted (as would be predicted). The zone between the two lines is a semiquantitative expression of the PTHvitamin D-mediated adaptation to the lower intake.4 The most likely explanation for the inverse relationship observed under both sets of conditions is that calcium transfer, whether active or passive, is a slow, inefficient process, with only a limited number of carrier molecules or pores available at any given instant. In the brief interval between exit of a bolus of food from the stomach and the time it reaches the colon, only so many calcium ions can use the available transport. If the number of ions reaching the absorptive site is small, then by numerical necessity the fraction absorbed will be larger than when the number of ions is large. Absorption fraction (or efficiency) is thus a potentially misleading measure (at least if we stop there).

4As Fig. 6 shows, most of the difference occurs at intakes below 500 mg (12.5 mmol). However, the two sets of observations were performed in different groups of women and even the non-adapted set must have had some basal level of 1,25(OH)2D-mediated adaptive absorption. Hence the true extent of adaptive absorption is undoubtedly somewhat greater than indicated solely by the difference between the two lines.

Absorbed calcium (mmol/d)

0.7

Absorption fraction

Absorption fraction

780

10

20

30

40

50

Calcium intake (mmol/d)

FIGURE 7 Fractional absorption and mass absorption for the 525 studies of Fig. 4. The left axis and solid line represent fractional absorption and the right axis and dashed line, mass absorption. (Copyright Robert P. Heaney, 1996. Reproduced with permission.)

It is, however, a necessary starting point because it is the primary datum available from most studies of absorptive physiology. Figure 7 presents the regression line from Fig. 4 and adds a second line representing the actual quantity of calcium absorbed in these same women (i.e., the product of absorption fraction and intake). This variable is obviously the nutritionally relevant one since, in offsetting obligatory losses (or special demands such as antler building or fetal skeletal development), it is a quantity of calcium (not a fraction) that is needed to balance the drains created by calcium leaving the ECF. Figure 7 also illustrates another important aspect of this input to the calcium economy. At low intakes, absorption is quantitatively low, despite being relatively more efficient. A moment’s reflection suffices to show that a large fraction of a small number is, of necessity, a smaller number still. Thus, absorbing even a large fraction of a small intake cannot produce much calcium. The result is that, in the range of intakes commonly encountered among contemporary, industrialized humans, absorptive adaptation (via vitamin D) mitigates the problem created by a low intake, but it does not counterbalance it. A concrete example, employing realistic numbers, will help illustrate this point, and will show additionally how optimal operation of the vitamin D hormonal system is dependent upon—and in fact presumes—the kinds of high calcium intakes found among hunter-gatherer humans and high primates (in whom the system evolved). Contrast how two individuals are able to respond to the increased obligatory loss occasioned by regular daily ingestion of an additional 100 mmol sodium

781

CHAPTER 46 Vitamin D: Role in the Calcium Economy

(approximately the sodium contained in a single fastfood chicken dinner). Assume that one individual is ingesting 5 mmol Ca (200 mg)/day (corresponding to the lower quintile of calcium intakes in U.S. women [38]), and the other, 40 mmol (1600 mg) (approximately the NIH Consensus Conference recommendation [39] for estrogen-deprived, postmenopausal women). Using data from the curve in Fig. 4, the individual with the lower intake absorbs at an efficiency of 44.5% prior to the extra sodium load, and the individual with the higher intake, at 17.8%. (The first, therefore, has a gross absorbed quantity from the diet of 2.2 mmol/day, and the second, 7.1 mmol/day.) The increase in obligatory urinary loss occasioned by the increase in sodium intake will be about 1 mmol/day (see earlier discussion). To offset this loss, the first individual, with the low intake, would have to increase the absorbed quantity to 3.2 mmol/day, which means increasing the already high absorption efficiency by a factor of nearly 1.5 (from 44.5 to 64.5%). By contrast, the individual with the high intake needs to increase only from 17.8 to 20.3%. (These calculations are summarized in Fig. 8.) The adjustment for the individual with a low calcium intake is substantially more than most adults can accomplish, while the second is easily accommodated. The first individual is forced, therefore, to get the needed additional calcium from bone, while the second easily gets it from her food. Thus, while vitamin D plays a critical role in increasing absorptive efficiency in response either to increased losses from the body or to decreased intake, it must be stressed that there is little room in which the PTH–vitamin D endocrine system can operate when

0.8

Added demand: 1 mmol/d Absorption fraction

0.7 0.6

+ 0.20

0.5 0.4 0.3

+ 0.025

0.2 0.1 0.0 0

5

10

15

20

25

30

35

40

45

50

Calcium intake (mmol/d)

FIGURE 8 Changes in absorption efficiency required to offset fully from diet an additional drain of 1 mmol Ca/d, at low and high regions of the curve of Fig. 4. Numbers shown are the required increases in absorption fraction at the respective intakes. (Copyright Robert P. Heaney, 2003. Reproduced with permission.)

calcium intakes are already low. That does not mean that ECF [Ca2+] regulation suffers. The bony calcium reserves are vast—essentially limitless. So long as vitamin D status is above rachitic levels, those reserves will readily be drawn upon to support ECF [Ca2+], using the well-studied mechanisms already described. Naturally, if this drain continues, bone mass will inevitably decline. At high calcium intakes, such as prevailed during hominid evolution, the vitamin D hormonal system not only helps maintain ECF [Ca2+], but total body calcium as well; at low intakes, only the ECF is protected.

C. Partition of Absorption between Active and Passive Mechanisms As described in Chapter 24, absorption occurs both by vitamin D–mediated active transport across the mucosal cells and by passive diffusion around the cells. Is it possible to partition absorption between the active, cellular process and the passive, paracellular process? To a limited extent, the answer is yes. As should be clear from the foregoing, absorption fraction (both passive and active) exhibits an inverse relation to intake or load. Absorption will be high by either mechanism, even approaching 100%, if the load is sufficiently small. But small loads are not nutritionally relevant, no matter how well they are (or are not) absorbed; so in this discussion I shall confine myself to consideration of loads or intakes in the range of (or above) currently recommended intakes. The right end of the regression line in Fig. 4 represents an absorption fraction of approximately 0.15. In work published earlier from our laboratory [40], we extended intakes well above the 2 g (50 mmol) upper limit of Fig. 4, to as high as 8.0 g (200 mmol) Ca/day. Absorption fraction in that study also averaged about 0.15 at these very high intakes, or approximately what we observed at 50 mmol in Fig. 4. The essentially linear character of this absorption across such a broad range of intakes probably reflects the fact that, if passive absorption is due largely to solvent drag, the quantity of calcium absorbed will be a linear function of luminal calcium concentration. It is likely, at habitual intakes in the range of 40–50 mmol Ca/day, that active absorption would be minimal, and it is a virtual certainty that that would be so at supraphysiological intakes of 200 mmol. Studies in patients with end-stage renal disease, with limited ability to synthesize 1,25(OH)2D, also report absorption fractions in the range of 10–20% [41,42]. Taken together, these data indicate that passive absorption is able to extract about 10–15% of the calcium in the ingested

782

ROBERT P. HEANEY

food at nutritionally relevant loads. Variation around that level will presumably relate to inter-individual variations in mucosal mass and in intestinal transit time. What would an absorption efficiency of 15% mean for the calcium economy if passive absorption were the only means for extracting calcium from the diet, as, for example, must be the case with hereditary vitamin D resistant rickets (HVDRR) Type II? Assume an intake of 20 mmol/day. As noted earlier, digestive secretions add 3.5 mmol to the stream, 0.6 mmol of that total too low to be absorbed. Calculation from these data easily shows that, at a 15% absorption efficiency, the gut will be a net excretory organ (if only barely: approximately 0.1 mmol/day net loss). No calcium gain into the body could occur at such an intake. Hence a background, gross absorption of 15% without vitamin D seems entirely compatible with development of severe rickets, either as found in HVDRR or in typical nutritional vitamin D deficiency. Figure 9 systematizes those calculations for varying levels of active absorption and a broad range of calcium intakes, expressing the results in terms of net calcium absorption. It shows graphically, for example, that without active transport it takes an intake of about 26 mmol to ensure zero gut balance, and that an intake of 60 mmol would not suffice to offset extraintestinal losses of 5 mmol through the kidney and skin. By contrast, active absorption of ~16% is sufficient in mature adults

48

Net absorption (mmol/d)

30

40

25

32

20

Needed to offset 5 mmol/d obligatory loss

15 10

24 16 8

5

Active absorption (percent)

35

0

0 −5

0

10

Zero balance across the gut 20 30 40 50 60 Calcium intake (mmol/d)

FIGURE 9 Relationship of vitamin D–mediated, active calcium absorption, calcium intake, and net calcium gain across the gut. Each of the contours represents a different level of active absorption above a baseline passive absorption of 12.5%. (The values along each contour represent the sum total of passive and variable active absorption.) The horizontal dashed lines indicate 0 and 5 mmol/d net absorption, respectively. The former is the value at which the gut switches from a net excretory to a net absorptive mode, and the latter is the value needed to offset typical urinary and cutaneous losses in mature adults. (Copyright Robert P. Heaney, 1999. Reproduced with permission.)

to offset such losses at intakes in the range of current recommendations (25–30 mmol/d).

IV. PHYSIOLOGICAL SOURCES OF VITAMIN D ACTIVITY Thus far I have spoken of vitamin D only generically. As discussed extensively elsewhere in this volume, native vitamin D (calciferol) has very little biological activity in its own right. It is converted in the body into a number of hydroxylated metabolites, the most important of which for the purposes of this chapter are 25OHD3 and 1,25(OH)2D3. It is well established that cholecalciferol is 25-hydroxylated in the liver. The reaction is loosely controlled by circulating levels of 1,25(OH)2D as well as by limited end-product inhibition exerted by 25OHD itself. However, for the most part circulating 25OHD levels are driven mainly by the circulating levels of the precursor, either ergo- or cholecalciferol. In studies performed in my laboratory, serum 25OHD3 rises by approximately 0.7 nmol/L for every µg of oral cholecalciferol after 16 to 20 weeks of daily oral administration [43]. This relationship is consistent with published data from studies of highdose vitamin D treatment and cases of vitamin D intoxication [44,45]. 1,25(OH)2D3 is produced in the kidney, as already described (Chapter 5), under control by serum PTH and serum inorganic phosphate concentrations. It is generally considered to be the biologically active form of the vitamin, responsible for its full spectrum of actions; it is also considered that the precursors (cholecalciferol and 25OHD3) exert activity in their own right only under conditions of intoxication. This conclusion is based mainly upon studies of both binding kinetics of the various metabolites with the vitamin D receptor and of consequent gene expression. If binding affinity for 1,25(OH)2D3 is taken as 1.0, reported values for 25OHD3 are in the range of 1×10–3 to 1×10–4, and for native D, 1×10–6 or lower [46–48]. Accordingly, it is reasoned that normal serum levels of the precursor compounds are too low to exert significant action. However, since serum levels of 25OHD are typically three orders of magnitude higher than those of 1,25(OH)2D, it is not clear, a priori, that 25OHD would be without effect under normal conditions. Protein binding in serum complicates interpretation of these relationships. It is generally presumed that total serum levels are physiologically less meaningful than free serum levels of a metabolite. As discussed in Chapter 8, the vitamin D family of compounds is carried in serum complexed to a D-binding carrier protein [an alpha globulin designated DBP, with a gradient of affinities: highest for 25OHD, lower for native vitamin D3, and

783

CHAPTER 46 Vitamin D: Role in the Calcium Economy

lower still for 1,25(OH)2D]. However, binding to the carrier protein cannot be interpreted without reference to other factors, usually not readily determined in any given situation. These include the relative affinities of the binding protein and the receptor as well as the concentrations of both. Available data on binding at the intact organism level are not sufficient to resolve this issue of hormone dynamics. But such theoretical considerations aside, there is a substantial body of clinical data indicating that 25OHD exerts appreciable biological activity at normal physiological concentrations. Several studies show a surprisingly strong correlation between serum 25OHD and intestinal calcium absorption efficiency in intact humans [34,49–53]. If 25OHD were acting only as a precursor, one should have expected no correlation at all. Where 1,25(OH)2D levels were measured in these studies, the correlation of 1,25(OH)2D with absorption fraction was usually weaker than for 25OHD, or nonsignificant altogether. Clinically, it is widely recognized that 1,25(OH)2D levels are commonly normal in sporadic nutritional rickets, while 25OHD levels are invariably low, and, more to the point, calcium absorption is low. Moreover, Colodro et al. [52], in a dose response study for both metabolites, using calcium absorption in healthy human adults as the endpoint, reported a molar potency for administered 25OHD relative to 1,25(OH)2D of 1:125, not the less than 1:2,000–1:4,000 figure predicted from in vitro receptor binding studies. Barger-Lux et al. [54], in a similar study, found a nearly identical potency ratio (1:100). In this study the rise in absorption fraction produced by oral 25OHD occurred without a detectable change in 1,25(OH)2D level. Taken together these studies provide evidence that some fraction of circulating vitamin D–like activity can be attributed to 25OHD.5 The basis for the interaction of the two vitamin D metabolites in intestinal calcium absorption is unclear. It may be that mucosal cell expression of the 1α-hydroxylase converts circulating 25OHD to 1,25(OH)2D within the cell, thereby augmenting the 1,25(OH)2D levels the receptor actually sees. Alternatively, the two metabolites may act synergistically—1,25(OH)2D in the canonical genomic manner, and 25OHD binding to membrane receptors and opening calcium channels, a process termed “transcaltachia” [57,58] (see Chapter 23). In any event, there can be no question about the potency of 1,25(OH)2D itself. This very active metabolite

5Some of the apparent action of 25OHD in oral dosing studies may be due to a direct effect of the metabolite on the mucosal cell (a first-pass effect). During absorption of 25OHD, mucosal exposure to the agent would be substantially higher than would occur from exposure to plasma 25OHD.

produces a strong enhancement of absorption efficiency when given to intact humans [52–56] with nondeficient levels of serum 25OHD. Transcaltachia, which is a rapid-response, nongenomic action of the D vitamin metabolites, requires occupancy of the nuclear receptor by 1,25(OH)2D [58]. This is shown, for example, in patients with HVDRR type II, who lack functional vitamin D receptors, and who are not able to absorb calcium efficiently despite normal to high circulating levels of both 1,25(OH)2D and 25OHD [59,60]. Thus vitamin D–mediated absorption seems to require both a functioning receptor and some combination of 1,25(OH)2D and 25OHD. Whatever the precise mechanism, the system operates as if there were a floor or background level of absorption in normal individuals that is determined in part by the long-half-life 25OHD, while 1,25(OH)2D produces a quick-acting fine tuning of the system.

V. OPTIMAL VITAMIN D STATUS Optimal vitamin D status can be defined as the daily intake or production of the vitamin (and/or the serum level of 25OHD) that is sufficient so that its availability does not limit any of the metabolic functions dependent upon the vitamin. This notion of limit is illustrated in Fig. 10A. A large body of data indicates that vitamin D-mediated absorption follows a curve such as the one presented in Fig. 10A, rising with intake at levels below the requirement, then flat through a range of sufficiency, then rising again at pharmacologic (or toxic) intakes. Figure 10B shows this behavior as related to actual serum 25OHD concentration, derived from two studies [61,62]. It is likely that the absorptive response, taken in isolation, would be a continuous smooth function of vitamin D dose, such as that indicated by the dashed line in Fig. 10A. However, at the whole-organism level, once calcium absorption is optimal, other factors alter the response to vitamin D exposure. For example, at levels of calcium absorption higher than needed to offset daily losses, PTH levels would drop and 1,25(OH)2D synthesis would fall. Thus, despite a rising solar or oral dose of the precursor vitamin D, absorption plateaus. But when the dose becomes sufficiently high (as in vitamin D intoxication), system controls are saturated and bypassed, and absorption begins to rise once again.6 Intakes below the plateau are clearly insufficient, since they limit

6This second rise in absorption has never been reported for purely solar sources of vitamin D; however, it is well recognized as a component of vitamin D intoxication.

784

ROBERT P. HEANEY

A

B Absorption fraction

Absorption fraction

0.50 0.45 0.40 0.35 0.30 0.25 0.20 0.00 Vitamin D input

0

40

60 80 100 120 Serum 25(OH)D (nmol/L)

140

160

FIGURE 10

(A) Schematic diagram of the likely relationship between absorptive performance and vitamin D input. The dashed line represents the curve in the absence of physiological controls (such as might be found in isolated gut loop studies), while the solid line represents the likely relationship in organisms with functioning control systems. (B) Paired sets of measured absorption values in groups studied at differing values for serum 25OHD. The lines connecting the points denote the pairs of values within each study [61,63]. (Copyright Robert P. Heaney, 1996, 2003. Reproduced with permission.)

absorption anterior to any physiological controls. Intake levels beyond the plateau, by contrast, represent toxicity, i.e., the overriding of physiological controls. (The subject of vitamin D toxicity is discussed in Chapter 78.) Optimal status could thus be defined as an intake or production of the vitamin sufficient to get an individual onto the absorptive plateau. Optimal position on the plateau would depend upon population-level relative risks of toxicity and deficiency. Defining such a level can be approached operationally in several ways. One is by determining the level of vitamin D intake at which calcium absorption does not change further upon giving extra vitamin D at doses within the physiological range [61,63]. Since both PTH levels and PTH-mediated bone resorption are known to rise in the face of vitamin D insufficiency, a complementary approach would be to define the 25OHD level at which the parathyroid response evoked by inadequate absorbed calcium intake is minimized [64,65]. Using the absorptive response to supplemental D [49,61,63,66], such indices of vitamin D status as seasonal variation in serum iPTH and bone remodeling [67–72], or the point at which PTH concentration is minimized [64,65], available evidence converges on a serum value of approximately 80 nmol/liter (32 ng/ml).7 This value is well above the bottom end of the range currently

7The principal exception to this otherwise highly consistent body of data is found mainly in studies from the Netherlands, in which PTH is reported to be minimized below 50 nmol/L [73,74]. The reason for this discrepancy is unclear.

considered normal [~37.5 nmol/liter (15 ng/ml)]. (At the same time it is reassuring to note that a value of 80 nmol/liter is itself well below the mean for individuals with levels of sun exposure such as must have prevailed during hominid evolution [75]). Clinical confirmation of the approximate correctness of this level is found in a recent, large, randomized, controlled trial in which raising serum 25OHD from 53 nmol/liter to 76 nmol/liter resulted in a 33% decrease in typical osteoporotic fractures over 5 years of treatment [62]. Note that the untreated level (53 nmol/liter) was itself well within the usual reference range and might therefore have been considered a “normal” value. The fact that it permitted excess osteoporotic fractures shows clearly that the bottom end of the reference range can no longer be used to determine normality. Since vitamin D is not found in appreciable concentrations in most of the items in the food supply (either primitive or modern), maintenance of optimal vitamin D status requires either sun exposure or, in high northern or southern latitudes, some degree of supplementation/fortification. Such a conclusion has often been uncongenial to traditional nutritionists, who have maintained that humans ought to be able to get all of the nutrients they need from a well-balanced diet. However, it must be stressed that vitamin D is an accidental nutrient, included with the other vitamins by mistake at an early stage of the development of nutritional science. Whatever the merit of the traditional nutritionists’ position, it cannot apply to this essential compound. Now that we understand the situation, we must see that it is certainly no more unnatural to sustain vitamin D status in high latitudes by supplementation

785

CHAPTER 46 Vitamin D: Role in the Calcium Economy

than it is to sustain body temperature there by clothing or shelter. (See also Chapter 61.) Although more work needs to be done to define the bottom end of the acceptable normal range with precision, for now the prudent course would seem to be to aim for a vitamin D intake sufficient to produce a serum 25OHD level of at least 80 nmol/liter (32 ng/ml). As just noted, levels below that point result in calcium absorptive impairment [63] and carry a risk of bone loss and osteoporotic fractures [62]. The 80 nmol/liter level is certainly safe, since healthy college-age adults (with often generous sun exposure) commonly have levels twice that high. Oral cholecalciferol intakes required to produce a level of 80 nmol/liter will depend upon degree of cutaneous synthesis and pretreatment serum 25OHD concentration. As noted earlier, the most direct evidence for dose response in midlife individuals points to a rise of 0.7 nmol/liter for each microgram (40 IU) of daily oral cholecalciferol supplementation [43]. Less direct evidence mostly obtained in the elderly indicates that the rate of response from conditions of more extreme vitamin D depletion could be as much as 1.6–2 nmol/liter per µg of continuous daily oral supplementation. Whatever the actual response rate, it is unlikely to be higher than 2 or much lower than 0.6 mmol/liter/µg/d. Hence a patient with a baseline value of 40 nmol/liter can be estimated to require a daily dose ranging from 800 to 2600 IU to achieve the desired level. It is worth noting that the entirety of this range of required intakes is above the currently recommended intake.

VI. SUMMARY AND CONCLUSIONS The best attested function of vitamin D is the facilitation of transfer of calcium (and phosphorus) into the extracellular fluid from ingested food and from bone. In this capacity, vitamin D functions as a part of a control system that operates to maintain constancy of the calcium ion concentrations in the extracellular fluid against the demands of obligatory excretory losses and skeletal mineralization. In both transfers vitamin D works in concert with parathyroid hormone. Quantitative analysis of the inputs and drains of the calcium economy reveals that, at contemporary calcium intakes, D-mediated absorptive enhancement only partially mitigates the impact of low calcium intake or large calcium losses. However, at intakes closer to those prevailing during hominid evolution, minor shifts in vitamin D-mediated absorption are fully adequate to compensate for stresses on the calcium economy. While 1,25(OH)2D is clearly the most potent form of the vitamin, 25OHD exerts significant vitamin D–like activity in its own right at physiological

serum levels. Optimal vitamin D status is operationally defined as a level of D intake (or production) high enough to ensure that the D-mediated transfers are not limited by D availability. Available data point to a value for serum 25OHD of about 32 ng/ml (80 nmol/liter) as the bottom end of the optimal range. Confirmation of this estimate comes from the demonstration that calcium absorption efficiency is suboptimal below ~80 nmol/liter and that risk of fragility fractures rises as serum 25OHD concentration falls below 80 nmol/liter.

References 1. Heaney RP 1963 Evaluation and interpretation of calcium kinetic data in man. Clin Orthop 31:153–183. 2. Clapham DE 1995 Calcium signaling. Cell 80:259–268. 3. Eastell R, Delmas PD, Hodgson SF, Eriksen EF, Mann KG, Riggs BL 1988 Bone formation rate in older normal women: concurrent assessment with bone histomorphometry, calcium kinetics, and biochemical markers. J Clin Endocrinol Metab 67:741–748. 4. Chapuy MC, Schott AM, Garnero P, Hans D, Delmas PD, Meunier PJ, and EPIDOS study group. 1996 Healthy elderly French women living at home have secondary hyperparathyroidism and high bone turnover in winter. J Clin Endocrinol Metab 81:1129–1133. 5. Charles P 1989 Metabolic bone disease evaluated by a combined calcium balance and tracer kinetic study. Dan Med Bull 36:463–479. 6. Rianon N, Feeback D, Wood R, Driscoll T, Shackelford L, LeBlanc A 2003 Monitoring sweat calcium using skin patches. Calcif Tissue Int 72:694–697. 7. Klesges RC, Ward KD, Shelton ML, Applegate WB, Cantler ED, Palmieri GMA, Harmon K, Davis J 1996 Changes in bone mineral content in male athletes. Mechanisms of action and intervention effects. JAMA 276:226–230. 8. Heaney RP, Recker RR 1994 Determinants of endogenous fecal calcium in healthy women. J Bone Miner Res 9:1621–1627. 9. Nordin BEC, Need AG, Morris HA, Horowitz M 1993 The nature and significance of the relationship between urinary sodium and urinary calcium in women. J Nutr 123:1615–1622. 10. Itoh R, Suyama Y 1996 Sodium excretion in relation to calcium and hydroxyproline excretion in a healthy Japanese population. Am J Clin Nutr 63:735–740. 11. Devine A, Criddle RA, Dick IM, Kerr DA, Prince RL 1995 A longitudinal study of the effect of sodium and calcium intakes on regional bone density in postmenopausal women. Am J Clin Nutr 62:740–745. 12. Heaney RP, Recker RR 1982 Effects of nitrogen, phosphorus, and caffeine on calcium balance in women. J Lab Clin Med 99:46–55. 13. Johnson NE, Alcantara EN, Linkswiler H 1970 Effect of level of protein intake on urinary and fecal calcium and calcium retention of young adult males. J Nutr 100:1425–1430. 14. Sebastian A, Frassetto LA, Sellmeyer DE, Merriam RL, Morris RC Jr 2002 Estimation of the net acid load of the diet of ancestral preagricultural Homo sapiens and their hominid ancestors. Am J Clin Nutr 76:1308–1316. 15. Parfitt AM, Higgins BA, Nassim JR, Collins JA, Hilb A 1964 Metabolic studies in patients with hypercalciuria. Clinical Science 27:463–482.

786 16. New SA, Robins SP, Campbell MK, Martin JC, Garton MJ, Bolton-Smith C, Grubb DA, Lee SJ, Reid DM 2000 Dietary influences on bone mass and bone metabolism: further evidence of a positive link between fruit and vegetable consumption and bone health. Am J Clin Nutr 71:142–151. 17. Walser M 1961 Calcium clearance as a function of sodium clearance in the dog. Am J Physiol 200:769–773. 18. Heaney RP 2003 How does bone support calcium homeostasis? Bone 33:264–268. 19. Raisz LG 1965 Bone resorption in tissue culture. Factors influencing the response of parathyroid hormone. J Clin Invest 44:103–116. 20. Portale AA, Halloran BP, Morris RC Jr 1987 Dietary intake of phosphorus modulates the circardian rhythm in serum concentration of phosphorus. J Clin Invest 80:1147–1154. 21. Feldman D, Malloy PJ, Gross C: Vitamin D 1996 Metabolism and action. In: Marcus R, Feldman D, Kelsey J (eds) Osteoporosis. Academic Press, San Diego, California. pp. 205–225. 22. Banks WJ Jr, Epling GP, Kainer RA, Davis RW 1968 Antler growth and osteoporosis. Anat Rec 162:387–398. 23. Abrams SA 2002 Nutritional rickets: An old disease returns. Nutr Rev 60(4):111–115. 24. Eaton SB, Nelson DA 1991 Calcium in evolutionary perspective. Am J Clin Nutr 54:281S–287S. 25. Heaney RP 1965 A unified concept of osteoporosis. Am J Med 39:877–880. 26. Bell NH, Greene A, Epstein S, Oexmann MJ, Shaw S, Shary J 1985 Evidence for alteration of the vitamin D-endocrine system in blacks. J Clin Invest 76:470–473. 27. Bell NH, Stern PH, Paulson SK 1985 Tight regulation of circulating 1-α,25-dihydroxyvitamin D in black children. N Engl J Med 313:1418. 28. Weinstein RS, Bell NH 1988 Diminished rates of bone formation in normal black adults. N Engl J Med 319:1698–1701. 29. Cosman F, Morgan DC, Nieves JW, Shen V, Luckey MM, Dempster DW, Lindsay R, Parisien M 1997 Resistance to bone resorbing effects of PTH in black women. J Bone Miner Res 12(6):958–966. 30. Aloia JF, Vaswani A, Yeh JK, Flaster E 1996 Risk for osteoporosis in black women. Calcif Tissue Int 59:415–423. 31. Abrams SA, O’Brien KO, Liang LK, Stuff JE 1995 Differences in calcium absorption and kinetics between black and white girls aged 5–16 years. J Bone Miner Res 10:829–833. 32. Heaney RP 2002 Ethnicity, bone status, and the calcium requirement. Nutr Res 22:(1–2):153–178. 33. Dawson-Hughes B, Harris S, Kramich C, Dallal G, Rasmussen HM 1993 Calcium retention and hormone levels in black and white women on high- and low-calcium diets. J Bone Miner Res 8:779–787. 34. Barger-Lux MJ, Heaney RP, Lanspa SJ, Healy JC, DeLuca HF 1995 An investigation of sources of variation in calcium absorption efficiency. J Clin Endocrinol Metab 80:406–411. 35. Barger–Lux MJ, Heaney RP, Recker RR 1989 Time course of calcium absorption in humans: Evidence for a colonic component. Calcif Tissue Int 44:308–311. 36. Heaney RP, Recker RR, Stegman MR, Moy AJ 1989 Calcium absorption in women: relationships to calcium intake, estrogen status, and age. J Bone Miner Res 4:469–475. 37. Heaney RP, Weaver CM, Fitzsimmons ML 1990 The influence of calcium load on absorption fraction. J Bone Miner Res 11:1135–1138. 38. Alaimo K, McDowell MA, Briefel RR, Bischof AM, Caughman CR, Loria CM, Johnson CL 1994 Dietary intake of vitamins, minerals, and fiber of persons 2 months and over in

ROBERT P. HEANEY

39. 40. 41. 42. 43.

44.

45. 46. 47. 48. 49.

50.

51.

52.

53. 54. 55.

56.

the United States: Third National Health and Nutrition Examination Survey, Phase 1, 1988–91. Advance data from vital and health statistics; no. 258. National Center for Health Statistics, Hyattsville, MD. NIH Consensus Conference: Optimal Calcium Intake 1994 JAMA 272:1942–1948. Heaney RP, Saville PD, Recker RR 1975 Calcium absorption as a function of calcium intake. J Lab Clin Med 85:881–890. Recker RR, Saville PD 1971 Calcium absorption in renal failure: its relationship to blood urea nitrogen, dietary calcium intake, time on dialysis, and other variables. J Lab Clin Med 78:380–388. Coburn JW, Koppel MH, Brickman AS, Massry SG 1973 Study of intestinal absorption of calcium in patients with renal failure. Kidney International 3:264–272. Heaney RP, Davies KM, Chen TC, Holick MF, Barger-Lux MJ 2003 Human serum 25-hydroxy-cholecalciferol response to extended oral dosing with cholecalciferol. Am J Clin Nutr 77:204–210. Whyte MP, Haddad JG Jr, Walters DD, Stamp TCB 1979 Vitamin D bioavailability: serum 25-hydroxyvitamin D levels in man after oral, subcutaneous, intramuscular, and intravenous vitamin D administration. J Clin Endocrinol Metab 48:906–911. Byrne PM, Freaney R, McKenna MJ 1995 Vitamin D supplementation in the elderly: review of safety and effectiveness of different regimes. Calcif Tissue Int 56:518–520. Hughes MR, Baylink DJ, Jones PG, Haussler MR 1976 Radioligand receptor assay for 25-hydroxyvitamin D2/D3 and 1-α,25-dihydroxyvitamin D2/D3. J Clin Invest 58:61–70. Brumbaugh PF, Haussler MR 1973 1,25-dihydroxyvitamin D3 receptor: Competitive binding of vitamin D analogs. Life Sci 13:1737–1746. DeLuca HF 1983 The vitamin D–calcium axis. In: Rubin RP, Weiss GB, Putney Jr JW (eds) Calcium in Biological Systems. Plenum Press, New York, pp. 491–511. Francis RM, Peacock M, Storer JH, Davies AEJ, Brown WB, Nordin BEC 1983 Calcium malabsorption in the elderly: the effect of treatment with oral 25-hydroxyvitamin D3. Eur J Clin Invest 3:391–396. Bell NH, Epstein S, Shary J, Greene V, Oexmann MJ, Shaw S 1988 Evidence of probable role for 25-hydroxyvitamin D in the regulation of human calcium metabolism. J Bone Miner Res 3:489–495. Reasner CA II, Dunn JF, Fetchick D, Liel Y, Hollis BW, Epstein S, Shary J, Mundy GR, Bell NH 1990 Alteration of vitamin D metabolism in Mexican-Americans [Letter to the editor]. J Bone Miner Res 5:13–17. Colodro IH, Brickman AS, Coburn JW, Osborn TW, Norman AW 1978 Effect of 25-hydroxyvitamin D3 on intestinal absorption of calcium in normal man and patients with renal failure. Metabolism 27:745–753. Devine A, Wilson SG, Dick IM, Prince RL 2002 Effects of vitamin D metabolites on intestinal calcium absorption and bone turnover in elderly women. Am J Clin Nutr 75:283–288. Barger-Lux MJ, Heaney RP, Dowell S, Bierman J 1996 Relative molar potency of 25-hydroxyvitamin D indicates a major role in calcium absorption. J Bone Miner Res 11:S423. Gallagher JC, Jerpbak CM, Jee WSS, Johnson KA, DeLuca HF, Riggs BL 1982 1,25-Dihydroxyvitamin D3: short- and long-term effects on bone and calcium metabolism in patients with postmenopausal osteoporosis. Proc Natl Acad Sci 79:3325–3329. Dawson-Hughes B, Harris SS, Finneran S, Rasmussen HM 1995 Calcium absorption responses to calcitriol in black and white premenopausal women. J Clin Endocrinol Metab 80:3068–3072.

CHAPTER 46 Vitamin D: Role in the Calcium Economy

57. Norman AW 1990 Intestinal calcium absorption: a vitamin Dhormone-mediated adaptive response. Am J Clin Nutr 51: 290–300. 58. Norman AW, Nemere I, Zhou L-X, Bishop JE, Lowe KE, Maiyar AC, Collins ED, Taoka T, Sergeev I, Farach-Carson MC 1992 1,25(OH)2-vitamin D3, a steroid hormone that produces biologic effects via both genomic and nongenomic pathways. Steroid Biochem Molec Biol 41:231–240. 59. al-Aqeel A, Ozand P, Sobki S, Sewairi W, Marx S 1993 The combined use of intravenous and oral calcium for the treatment of vitamin D dependent rickets type II (VDDRII). Clin Endocrinol Oxf 39:229–237. 60. Simonin G, Chabrol B, Moulene E, Bollini G, Strouc S, Mattei JF, Giraud F 1992 Vitamin D-resistant rickets type II: apropos of 2 cases. Pediatrie-Bucur 47:817–820. 61. Barger-Lux MJ, Heaney RP: Effects of above average summer sun exposure on serum 25-hydroxyvitamin D and calcium absorption. J Clin Endocrinol Metab 87(11): 4952–4956. 62. Trivedi DP, Doll R, Khaw KT 2003 Effect of four monthly oral vitamin D3 (cholecalciferol) supplementation on fractures and mortality in men and women living in the community: randomised double blind controlled trial. Br Med J 326: 469–474. 63. Heaney RP, Dowell MS, Hale CA, Bendich A 2003 Calcium absorption varies within the reference range for serum 25-hydroxyvitamin D. J Am Coll Nutr 22(2):142–146. 64. Chapuy M-C, Preziosi P, Maamer M, Arnaud S, Galan P, Hercberg S, Meunier PJ 1997 Prevalence of vitamin D insufficiency in an adult normal population. Osteoporos Int 7:439–443. 65. Thomas MK, Lloyd-Jones DM, Thadhani RI, Shaw AC, Deraska DJ, Kitch BT, Vamvakas EC, Dick IM, Prince RL, Finkelstein JS 1998 Hypovitaminosis D in medical inpatients. N Engl J Med 338:777–783.

787 66. Krall EA, Dawson–Hughes B 1991 Relation of fractional 47Ca retention to season and rates of bone loss in healthy postmenopausal women. J Bone Miner Res 6:1323–1329. 67. McKenna JM, Freaney R, Meade A, Muldowney FP 1985 Hypovitaminosis D and elevated serum alkaline phosphatase in elderly Irish people. Am J Clin Nutr 41:101–109. 68. Rosen CJ, Morrison A, Zhou H, Storm D, Hunter SJ, Musgrave K, Chen T, Wei W, Holick MF 1994 Elderly women in northern New England exhibit seasonal changes in bone mineral density and calciotropic hormones. Bone Miner 25:83–92. 69. Dawson-Hughes B, Harris SS, Dallal GE 1997 Plasma calcidiol, season, and serum parathyroid hormone concentrations in healthy elderly men and women. Am J Clin Nutr 65:67–71. 70. Salamone LM, Dallal GE, Zantos D, Makrauer F, DawsonHughes B 1993 Contributions of vitamin D intake and seasonal sunlight exposure to plasma 25-hydroxyvitamin D concentration in elderly women. Am J Clin Nutr 58:80–86. 71. Dawson-Hughes B, Dallal GE, Krall EA, Harris S, Sokoll LJ, Falconer G 1991 Effect of vitamin D supplementation on wintertime and overall bone loss in healthy postmenopausal women. Ann Intern Med 115:505–512. 72. Krall EA, Sahyoun N, Tannenbaum S, Dallal GE, Dawson-Hughes B 1989 Effect of vitamin D intake on seasonal variations in parathyroid hormone secretion in postmenopausal women. N Engl J Med 321:1777–1783. 73. Lips P 2001 Vitamin D deficiency and secondary hyperparathyroidism in the elderly: consequences for bone loss and fractures and therapeutic implications. Endocr Rev 22:477–501. 74. Lips P, Duong T, Oleksik A, Black D, Cummings S, Cox D, Nickelsen T 2001 A global study of vitamin D status and parathyroid function in postmenopausal women with osteoporosis: baseline data from the multiple outcomes of raloxifene evaluation clinical trial. J Clin Endocrinol Metab 86:1212–1221. 75. Matsuoka LY, Wortsman J, Hollis BW 1990 Suntanning and cutaneous synthesis of vitamin D3. J Lab Clin Med 116:87–90.

CHAPTER 47

Effects of Race, Geography, Body Habitus, Diet, and Exercise on Vitamin D Metabolism MEHRDAD RAHMANIYAN AND NORMAN H. BELL Department of Medicine, Medical University of South Carolina, Charleston, South Carolina

I. II. III. IV.

Introduction Effects of Race and Geography Effects of Diet Effects of Body Habitus

I. INTRODUCTION Vitamin D metabolism is influenced by a number of factors. These include race, geographic location, diet, body habitus, and exercise. To understand how these factors exert their influence, it is useful to briefly review the metabolism of vitamin D and vitamin D–endocrine system. These subjects are discussed in greater detail in Chapter 2 of this book. A list of populations at risk for developing rickets and osteomalacia and factors involved is shown in Table I. Vitamin D metabolism can be summarized as follows (see Chapter 3 for details). In skin, vitamin D3 is synthesized from dermally produced 7-dehydrocholesterol. Previtamin D3 is converted from 7-dehydrocholesterol by absorption of one photon of ultraviolet sunlight, and the further conversion of previtamin D3 to vitamin D3 is regulated by body heat, a process that takes place over a period of several days and is temperature dependent [1,2]. The vitamin is carried from capillaries in skin by vitamin D–binding protein to the liver, where it is converted to 25-hydroxyvitamin D (25OHD) by vitamin D-25-hydroxylase (CYP2R1), the newly discovered hepatic microsomal enzyme [3,4]. 25OHD is further converted to 1,25-dihydroxyvitamin D [1,25(OH)2D] in the proximal tubule of the kidney by the mitochondrial enzyme 25OHD-lα-hydroxylase (lα-hydroxylase) (CYP27B1). The enzyme is stimulated directly by parathyroid hormone a reaction that is mediated by its messenger cyclic AMP [5–7], and indirectly by growth hormone, through stimulation of insulin-like growth factor-I [9,10] and is inhibited by calcium [8,11,12] and inorganic phosphate [13,14]. In states of vitamin D excess, 25OHD is converted to VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Effects of Exercise VI. Summary References

24,25-dihydroxyvitamin D [24,25(OH)2D] by the mitochondrial enzyme 25OHD-24-hydroxylase (24-hydroxylase) (CYP24A1), which is present in the kidney and other organs [15–17], and less 1,25(OH)2D is produced. Conversely, in states of vitamin D deficiency, less 24,25(OH)2D and more 1,25(OH)2D is produced. The two metabolites undergo additional hydroxylation to form 1,24,25-trihydroxyvitamin D, before being converted to calcitroic acid. This degradative pathway is similar to the classic and alternative pathways that are involved in the transformation of cholesterol to bile acids. CYP24A1 is induced by 1,25(OH)2D by two vitamin D receptor (VDR) response elements in the gene promoter. Thus, 1,25(OH)2D regulates not only its own rate of degradation but that of 25OHD as well [18]. The regulation of degradation of 25OHD by 1,25(OH)2D is underscored by studies in rats which showed that 1,25(OH)2D increases the metabolic clearance rate of 25OHD [19] and by studies in human subjects which showed that the increase in serum 25OHD produced by pharmacologic doses of vitamin D was prevented by the simultaneous administration of 1,25(OH)2D3 [20]. Calcium metabolism is modulated by a negative feedback control system that includes the parathyroids, skeleton, kidneys, and intestine. Serum calcium is kept within a very narrow range by parathyroid hormone (PTH) and 1,25(OH)2D by stimulating osteoclastic bone resorption [21], the reabsorption of calcium by the renal tubules [22], and the intestinal absorption of calcium, a biochemical event mediated by 1,25(OH)2D via the VDR [23,24]. Secretion of PTH by the parathyroid glands is stimulated by phosphate [25] and is inhibited by both calcium and 1,25(OH)2D. Inhibition by calcium is mediated via a calcium-sensing receptor [26] and Copyright © 2005, Elsevier, Inc. All rights reserved.

790

MEHRDAD RAHMANIYAN AND NORMAN H. BELL

TABLE I Causes and Consequences of Vitamin D Deficiency in Various Races and Populations

Population Asian Indians Blacks Caucasians Chinese Egyptians Hispanics Jordanians Libyans Moroccans Pakistanis Polynesians Saudi Arabians Lebanese Turkish Iranians Other Arabs

Increased skin pigment

Diminished exposure to sunshine

Inadequate intake of vitamin D

Rickets

Osteomalacia

+ + − − + + + + + + + + − + + +

+ + − − + − + + + + − + + + + +

+ + + + + − + + + + − + + + + +

+ + + + + − + + + + − + + + + +

+ − − − − − − − − + − − − − − −

Newborn infants of any race or population are prone to develop rickets when breast-fed and kept indoors. Asians and Pakistanis are at risk to develop vitamin D deficiency and osteomalacia, particularly when they reside away from the equator. Despite knowledge of prevention of these diseases, they are widespread throughout the world.

inhibition by 1,25(OH)2D is mediated via the VDR [27,28]. The inhibition of PTH secretion by 1,25(OH)2D in enhanced by up-regulation of the VDR in the parathyroids [29].

II. EFFECTS OF RACE AND GEOGRAPHY A. Blacks 1. VITAMIN D METABOLISM

When blacks live in the Northern or Southern Hemisphere at some distance from the equator, the vitamin D metabolic pathway is altered as a consequence of diminished exposure to sunshine and intensity of sunlight. For example, it was found that male blacks in Zaire had a mean serum 25OHD value of 29 ± 4 pg/ml (72.5 pmol/ml) (± SE) whereas blacks in Belgium had a mean serum 25(OH)D value of 9 ± 1 pg/ml (22.5 pmol/ml). A gradual decline in serum 25OHD over a period of years occurred in black men who moved from Zaire to Belgium [30]. Increased skin pigment and diminished dermal production of vitamin D3 accounts for the reduction in serum 25OHD in blacks [31]. The formation of previtamin D3 from

7-dehydrocholesterol by absorption of photons of ultraviolet light is prevented by their absorption instead by the skin pigment melanin. Compared to Caucasians, decreases in serum 25OHD in African-Americans of about 50% are found in most studies in the United States including newborns, children, and adults [32– 41]. Compared to Caucasians, blacks show decreases in serum vitamin D and 25OHD, modest increases in serum PTH, serum 1,25(OH)2D, and urinary cyclic adenosine 3′,5′-monophosphate, and decreases in urinary calcium [32–42]. The changes are caused by low serum 25OHD as a consequence of vitamin D deficiency and are corrected by treatment with 25OHD3 [40]. Of course, they could also be corrected by treatment with vitamin D. The decrease in urinary calcium results from increased secretion of PTH and increased renal tubular reabsorption of calcium. In this regard, the incidence of calciumcontaining kidney stones in blacks in South Africa is reduced because of a low urinary calcium [42]. There is a seasonal variation in vitamin D metabolism as a consequence of changes in duration of sunshine and intensity of sunlight in all individuals residing in the Northern or Southern Hemisphere, regardless of race or skin color. In the Northern Hemisphere, serum

CHAPTER 47 Effects of Race, Geography, Body Habitus, Diet, and Exercise on Vitamin D Metabolism

25OHD and 1,25(OH)2D are higher in summer than in winter. However, serum 25OHD is lower and serum 1,25(OH)2D is higher in black than in Caucasian men and women regardless of season [37]. Studies of dynamics of secretion of PTH show an exaggerated increased secretion of the hormone in response to induced hypocalcemia and diminished suppression of secretion of the hormone in response to induced hypercalcemia in black compared to white men and women [39]. The demonstration that parathyroid glands are larger in black than in white individuals in postmortem studies in the United States indicates that moderate hypertrophy of the parathyroid glands is associated with long-term increased secretion of PTH [43]. Blacks show a decreased skeletal sensitivity to PTH. This is evident not only by the alteration in vitamin D metabolism system with secondary hyperparathyroidism [32], but by the demonstration that the response of bone markers to infusion of hPTH(1-34) is blunted in premenopausal black compared to premenopausal white women [44]. Blacks retain more calcium than whites through renal conservation and relative preservation of skeletal tissue. These findings may account for the higher bone mass and lower incidence of osteoporosis and fractures in black women [44]. In black and white men and women, 25OHD3 turnover studies showed that reduction in serum 25OHD in blacks probably results from a diminished production rate and not an increased metabolic rate of the metabolite since low values for serum 25OHD were associated with very low values for serum vitamin D [45]. Thus, decreased availability of substrate is the major factor that accounts for low serum 25OHD values in blacks. There was an 8- to 10-fold variation in the production rate of 25OHD in both black and Caucasian men and women. It is not clear whether genetic heterogeneity of CYP2R1 itself, nonspecific hydroxylation by other enzymes, or other factors account for the wide variation. The most recent NHANES (1988–1994) study found that over 40% of African American women of reproductive age had vitamin D deficiency as defined by serum values for 25OHD ≤ 15 ng/ml (37.5 nmol/liter) [38]. In the growing fetus, mothers are the sole source of vitamin D and 25OHD. Vitamin D and 25OHD in maternal milk alone are an inadequate source in newborns and infants, regardless of race. Indeed, concentrations of vitamin D and 25OHD (39 and 310 pg/ml, respectively, 97.5 and 775 pmol/liter) are low even in milk from Caucasian women. Black infants in particular are at risk for developing vitamin D–deficient rickets, especially when they are breast fed [46–50]. Thus, maternal milk provides only a modest portion of nutritional needs for infants and other sources are required [51]. For maintenance of a normal serum 25OHD in both

791

premature and full-term healthy infants, 400 IU (10 µg) of vitamin D per day was found to be adequate [52,53]. A daily supplement of 1000 IU (25 µg) per day is recommended for preterm babies who are breast fed (see Chapter 48). Apparently, there is no maturational delay in the postnatal development of hepatic 25-hydroxylase activity in well babies. 2. BONE MASS

A higher bone mineral density in blacks compared to whites dates from infancy and is present throughout life. In black compared to white prepubertal boys, bone mineral densities of the hip, spine, trochanter, and femoral neck were significantly greater but growth hormone secretion and serum sex hormones were not different [55]. A recent study found that total body bone mineral content is higher in black than in white children of the same age and Tanner stage [53]. Over half of the difference could be accounted for by differences in body size and body composition. In contrast, bone mineral content at the spine was found to be the same in black and white children. One study found no difference in intestinal absorption of calcium and a lower urinary calcium in black compared to white children [37]. However, calcium kinetic studies showed that calcium accretion and intestinal absorption of calcium are higher and bone resorption and urinary calcium are lower in African-American compared to Caucasian girls [42,54]. Whether these differences represent heterogeneity in the populations studied or differences in methodology is not known. Bone mineral density at the lumbar spine, trochanter, and femoral neck are higher in black than in Caucasian women, both premenopausal and postmenopausal, and are higher in black than in white men [56–59]. In addition to growth during childhood and adolescence, the skeleton continuously remodels throughout life (see Chapter 28). Bone remodeling begins with activation of osteoclasts and resorption of small areas called Howship’s lacunae. This is followed by recruitment of osteoblasts or bone-forming cells that repair the resorption site. With aging, skeletal repair is not complete, resulting in bone loss. Generally, the rate of skeletal remodeling is a major determinant of rate of bone loss [60]. Histomorphometric studies showed that the rate of skeletal remodeling in blacks is reduced. As a consequence, this difference could contribute to or be responsible for the greater bone mass in black men and women [60]. In one study, serum 17β-estradiol, a major determinant of growth hormone secretion in both men and women, and growth hormone secretion were 50% higher and bone mineral density of the total body, forearm, trochanter, and femoral neck was significantly higher in the black than in the white men [56].

792 Since estrogen diminishes skeletal remodeling and growth hormone stimulates skeletal growth, racial differences in serum 17β-estradiol and growth hormone secretion in men could be a contributing factor to differences in bone mineral density [56]. In premenopausal women, however, no racial difference in serum 17β-estradiol or growth hormone secretion was found despite a higher bone mineral density of the total body and hip in the black women [59]. Black men and women in the United States are less likely to develop osteoporosis and atraumatic fractures than Caucasians. The rate of fracture for black women is 40% to 60% lower than that for Caucasian women [61–64]. The racial difference in the incidence of fractures is attributed in part to the racial difference in bone mass. In both black and white women, low body weight, a previous stroke, use of aids in walking, and alcohol consumption are risk factors for hip fracture [65,66]. In black women, a shorter hip axis length may protect against fractures [67]. Since the incidence of obesity is twice as high in black than in white women, and since body weight is a major determinant of bone mineral density, greater body weight is a contributing factor to higher bone mineral density in black women. Greater body fat provides cushioning for falls and protects against fractures. Black men and women in South Africa also are less likely to develop osteoporosis and atraumatic fractures than Caucasians [68,69]. In children, no racial difference was found in bone mineral content of the forearm [70]. Bone mineral density of the femoral neck is higher in black than in white women in South Africa, but bone mineral density of the forearm and lumbar spine are not different [71]. Thus, a greater bone mineral density of the femoral neck may account for the lower incidence of femoral fractures in black women, but the racial difference in vertebral fractures must result from factors other than a difference in bone mass. 3. BONE HISTOMORPHOMETRY

Bone formation rate, determined by histomorphometric analysis of biopsies of the iliac crest after double-tetracycline labeling, was found to be lower by about two-thirds in black compared to white men and women in the United States [60]. Static measurements in the two groups were not different. Since the error of histomorphometric measurements is about 10% and racial differences in bone mineral density range from 5% to 10%, these findings are consistent with differences in bone mineral density. Studies with markers of bone turnover remodeling support the histomorphometric findings. For example, serum osteocalcin, a marker of bone formation, is lower in black than in Caucasian subjects [32,72], and urinary hydroxyproline, a marker

MEHRDAD RAHMANIYAN AND NORMAN H. BELL

of bone resorption, is lower in black than in white women both before and after menopause [72]. Urinary N-telopeptides of type I collagen, a highly specific marker of bone resorption, is lower in black than in white men [73]. As noted, decreases in the rate of skeletal remodeling could contribute to the racial difference in bone mineral density. Findings in patients with thyroid disease support this concept. Bone mineral density varies inversely with rates of skeletal remodeling. Remodeling is increased in hyperthyroidism and decreased in hypothyroidism and bone density is decreased and increased, respectively [74]. Inhibition of bone resorption with the bisphosphonate alendronate increases bone mineral density at both the lumbar spine and hip as well in black as it does in Caucasian postmenopausal women [75,76]. In subjects from South Africa, histomorphometric studies of bone biopsies of the iliac crest without doubletetracycline labeling demonstrated thicker trabecular bone, greater osteoid volume, surface, and thickness, and greater erosion surfaces in blacks compared to Caucasians, findings different from those in studies from the United States [77]. The greater values for osteoid volume and erosion in blacks were attributed to higher rates of bone turnover and to the fact that trabecular bone in blacks was renewed more frequently and was therefore less prone to fatigue failure and spontaneous fracture [77].

B. Asian Indians and Pakistanis 1. VITAMIN D METABOLISM

Asian Indians living in South Carolina have a lower serum vitamin D, serum 25(OH)D, urinary calcium, and urinary phosphorus compared with Caucasians, whereas serum PTH and serum 1,25(OH)2D are significantly higher in Asian Indians than in Caucasians [78]. Previous studies in Asian Indians residing in Great Britain showed that, compared to values in Caucasians, serum 25OHD is reduced [79–96], and vitamin D deficiency often leads to rickets in neonates [82,86] and infants [87], as well as in children and adolescents [79,88–94], and to osteomalacia in adults [79,92–96]. Infants of mothers with severe osteomalacia develop rickets. Risk factors for rickets are inadequate exposure to sunlight, malabsorption of vitamin D, and renal and hepatic disorders [97,98]. Rickets and osteomalacia caused by vitamin D deficiency can be treated and prevented by vitamin D in daily doses of 400 to 3000 IU [82,94,96,99,100]. Factors that contribute to vitamin D deficiency in immigrant Asian Indians and Pakistanis include increased skin pigmentation and diminished dermal production of vitamin D3 [31], diminished exposure to

CHAPTER 47 Effects of Race, Geography, Body Habitus, Diet, and Exercise on Vitamin D Metabolism

sunlight in more northern climates [84–101], reduced intake of vitamin D [81,88,92], consumption of nonfortified Chapputi flour that is not enriched with vitamin D [81,88], and vegetarian diets [80,83,96]. Nevertheless, increases in serum vitamin D after exposure to ultraviolet light are comparable in Indians, Pakistanis, and Caucasians [101]. In Pakistanis who live closer to the equator in Pakistan, serum 25OHD values are comparable to those in Caucasians living in Britain [83,102]. Ingestion of vegetarian diets, particularly those without meat, eggs, or fish, and not intake of Chapputi flour, contribute to vitamin D deficiency [103]. However, serum 25OHD also is lower in Asian adults and children who do not consume a vegetarian diet than in Caucasians [80,84]. Because of maternal vitamin D deficiency and lack of supplementation of vitamin D, vitamin D deficiency and rickets were found in Pakistani infants, particularly those who were breast fed [95,104]. 2. BONE MASS

When body mass index and the area over which the X-ray beam during measurement of BMD is projected are taken into account, no difference in bone mass of Asian Indians compared to Caucasians can be demonstrated. Whereas earlier studies demonstrated lower bone densities in Japanese and Chinese, who are smaller compared to Caucasians, the studies did not take body mass index or size into consideration [105]. There is no difference in the incidence of hip fracture in Asian Indians and Caucasians living in England [106]. 3. BONE HISTOMORPHOMETRY

In Asian Indians with clinical and biochemical changes of osteomalacia, biopsies of the iliac crest showed increases in osteoid volume [92]. Since only static histomorphometric measurements were carried out, dynamic measurements after double-tetracycline labeling also should be performed to demonstrate impaired calcification, a sine-qua-non requirement for the diagnosis of osteomalacia.

C. Chinese 1. VITAMIN D METABOLISM

In China, infantile rickets is common. The incidence is highest in breast-fed infants and in infants who are 2 to 4 months of age [107,108]. Rickets sometimes occurs in infants who have a normal serum 25OHD, indicating that other factors may play a role in the pathogenesis of the bone disease in these infants [109]. Thus, normal serum 25OHD values may be found in rachitic patients who had received a dose of vitamin D

793

or have had significant exposure to sunlight before blood sampling. As expected, serum 25OHD was higher in spring, summer, and fall than in winter, serum 25OHD was higher in maternal than cord blood, and serum 25OHD in cord blood correlated with serum 25(OH)D in maternal blood [109]. In a prospective study in two northern and two southern Chinese cities, term infants were treated with 100, 200, or 400 IU vitamin D daily and cord blood at delivery as well as X-rays of the wrist and blood samples were obtained after 6 months of treatment [110]. Half were studied in the fall and half in the spring. The results showed that none of the infants had rickets. Serum 25OHD from cord blood was lower in infants from the north than in those from the south. Serum 25OHD was lower in infants at 6 months of age and was higher with increasing doses of vitamin D. Ossification of the wrist was less common in northern than in southern infants and was more likely to be present in infants born in the fall who had a higher serum 25OHD. In breastfed infants, whereas increased facial exposure to sunshine increased serum 25OHD in some infants, supplementation with vitamin D was required to prevent vitamin D–deficient rickets in the general population [111]. 2. BONE MASS

Chinese women in Beijing, China, have a lower bone mineral density and a modestly lower rate of vertebral fracture than white women in the United States. As in other populations, low bone mineral density and a sedentary life style increase the risk of osteoporosis and fracture [112]. Compared to Caucasians, Asian boys have greater bone mineral content of the lumbar spine at midpuberty and lower whole-body bone mineral content at maturity and Asian girls have lower femoral neck bone mineral content through mid-puberty and lower whole-body bone mineral content in pre-/early puberty. Whole-body bone mineral density and wholebody bone mineral content/height values are significantly lower in mature Asian versus Caucasian males. Differences in bone mineral density and apparent density between Asian and Caucasian subjects are due to differences in body weight and pubertal stage, and, at the femoral neck, to differences in weight-bearing activity [113].

D. Hispanics 1. VITAMIN D METABOLISM

Studies in prepubertal Mexican-American and Caucasian girls in southeastern Texas showed that calcium absorption, urinary calcium excretion, calcium kinetic values, and total body bone mineral content

794 were similar. However, serum PTH was greater and serum 25OHD was lower in Mexican-American girls than in Caucasian girls and serum 1,25(OH)2D was not different in the two groups [114]. Mexican-American men and women have a lower serum vitamin D and 25OHD and higher serum 1,25(OH)2D and serum PTH than Caucasians [115]. In contrast to African Americans [32], urinary calcium, serum osteocalcin, and urinary cyclic adenosine 3′,5′-monophosphate are not different. The low serum vitamin D and 25OHD in MexicanAmericans is attributed to increased skin pigmentation. 2. BONE MASS

Mexican-American women have higher bone mineral density at the proximal femur than Caucasian women, and this difference may account in part for their lower rate of hip fracture. In the United States, Hispanics are intermediate for risk of hip fracture between blacks and whites and Mexican-Americans are at higher risk than Puerto Ricans [115–119].

E. Polynesians 1. VITAMIN D METABOLISM

Serum 25OHD is lower in Polynesians compared to Caucasians. However, serum PTH, 1,25(OH)2D, osteocalcin, alkaline phosphatase, calcitonin, and urinary calcium/creatinine and urinary hydroxyproline/ creatinine ratios are not different [120]. Reduction in serum 25OHD is attributed to increased skin pigmentation. 2. BONE MASS

Bone mineral density of the forearm, lumbar spine, and femoral neck is higher in Polynesian than in Caucasian women even after adjustment for body mass index [121,122]. It is known that a short femoral neck is associated with a low rate of hip fracture [67]. However, the femoral neck is longer in Polynesians than it is in Europeans so that the lower incidence of hip fracture cannot be attributed to a difference in femoral neck length. Either a higher bone mineral density or other more subtle differences in proximal femoral geometry must account for the lower hip fracture incidence [122].

F. Saudi Arabians 1. VITAMIN D METABOLISM

Despite proximity to the equator, vitamin D deficiency due to lack of exposure to sunlight and inadequate

MEHRDAD RAHMANIYAN AND NORMAN H. BELL

intake of vitamin D is the commonest cause of rickets in Saudi Arabia. Rickets also occurs because of inadequate calcium intake and putative vitamin D-25-hydroxylase deficiency [123]. Saudi women are veiled, avoid sunlight, and stay indoors. Consequently, serum 25OHD is often low [124,125]. Vitamin D deficiency is very common in Saudi women who are pregnant, and serum 25OHD is low in cord blood [125]. Vitamin D deficiency is common in Saudi men as well as in Jordanian, Egyptian, and other men living in Saudi Arabia and is attributed to inadequate intake of vitamin D and avoidance of sunlight [126]. Vitamin D deficiency in elderly Saudi men and women is associated with fractures of the femoral neck because of avoidance of sunlight [127]. 2. BONE MASS

Bone mineral density in healthy Saudi women is lower than in American women. This is attributed to a higher number of pregnancies and longer duration of lactation together with prevalent vitamin D deficiency [128]. One study showed a significant correlation between serum 25OHD and back pain. Treatment with vitamin D increased serum 25OHD and alleviated the back pain [129]. Thus, vitamin D deficiency is very common in Saudis of all ages and is a cause of rickets and osteoporosis with fractures.

G. Other Groups In Arabs in the United Arab Emirates (UAE), a correlation between serum 25OHD and biosocial factors including age, education, and living accommodation were found. Serum 25OHD in UAE Arabs and in nonGulf Arabs was significantly lower than in Europeans, whereas serum calcium, phosphorus, alkaline phosphatase, and PTH among the groups were not different [130]. In Lebanese, the degree of vitamin D deficiency lies between that observed in Europe and the United States [131]. A pilot study among Iranian women and their neonates showed that 80% had a low maternal serum 25OHD. Mean maternal plasma calcium and alkaline phosphatase were in the normal range and serum PTH was elevated. The mean cord serum 25OHD was very low. Serum 25OHD in infants of mothers with hypovitaminosis D was almost undetectable [132]. In Turkey, the extent to which clothing causes limited exposure to sunlight was shown to be a major factor in determination of serum 25OHD [133]. In the Netherlands, serum 25OHD was lower and serum PTH was higher in children of Turkish and Moroccan immigrants who have dark skin [134]. In contrast, serum 25OHD in dark-skinned Bedouins of the Negev Desert

CHAPTER 47 Effects of Race, Geography, Body Habitus, Diet, and Exercise on Vitamin D Metabolism

of Israel is not different compared to lighter-skinned Jewish men and women. Presumably a longer duration of exposure to sunshine and greater intensity of sunlight prevented vitamin D deficiency [135]. Biochemical and clinical rickets occur in Libyan infants who are breast fed. The rickets is attributed to inadequate maternal exposure to sunlight and inadequate maternal intake of vitamin D [136]. In Alaska vitamin D deficiency among children less than 2 years is common especially in breast-fed infants [137]. In Australia food is not fortified with vitamin D, and the major source of vitamin D is casual exposure to the sunshine, leading to vitamin D deficiency in adults [138]. Again, serum 25OHD can vary as a consequence of geographical location. Serum values are lower in subjects in southern Argentina (Ushuaia) than in subjects in northern Argentina (Buenos Aires) [139].

III. EFFECTS OF DIET

795

vitamin D [143–147]. Serum vitamin D is low presumably because vitamin D is fat soluble and is stored in fat [148,149]. In obese men and women, compensatory increases in serum PTH, serum 1,25(OH)2D, and urinary cyclic adenosine 3′,5′-monophosphate and decreases in urinary calcium occur [145,147,150]. These changes are reversible with weight loss [150,151]. Treatment with 25OHD3 also corrects these changes: serum 25OHD and urinary calcium are increased and serum 1,25(OH)2D and urinary cyclic adenosine 3′,5′- monophosphate are decreased [147]. After treatment, these changes return to baseline values [147]. On the other hand, in nonobese subjects, administration of 25OHD increases serum 25OHD, and decreases serum 1,25(OH)2D, but does not alter urinary calcium [147]. As noted already, nonobese blacks living in the United States have changes in vitamin D metabolism similar to those of obese Caucasians [32]. These changes are not further altered by obesity [150].

A. Vitamin D Metabolism B. Bone Mass Dietary intake of vitamin D was found to be significantly lower in vegans and lactovegetarians compared with omnivores living in Finland. Throughout the year serum 25OHD was lower and serum PTH was higher in vegans than in omnivores and lactovegetarians. BMD of the lumbar spine was lower in vegans than in omnivores and tended to be lower than in lactovegetarians. Bone mineral density of the neck of femur tended to be lower in vegans than in omnivores and lactovegetarians [140].

Regardless of race, obesity is associated with increases in bone mineral density of the lumbar spine and hip [152,153]. Indeed, body weight is an important determinant of bone mineral density of the lumbar spine, trochanter, and femoral neck in postmenopausal women [154,155]. Radiographic measurements of the metacarpal cortical area showed a higher bone mass in obese compared to nonobese individuals [156]. Also, fat mass is a major determinant of total body bone mineral density in premenopausal and postmenopausal women [152,154,155].

B. Bone Mass In the Netherlands, a high prevalence of rickets was found in infants on macrobiotic diets of unpolished rice, pulses, vegetables with high fiber content, seaweeds, fermented foods, nuts, seeds, and fruits [141] and reduced bone mineral content of the whole body, spine, radius, and hip were found in adolescents who had been fed a macrobiotic diet in early life [142]. Thus, the vitamin D and calcium content of strict vegetarian and macrobiotic diets is inadequate, and the diets need to be fortified to prevent vitamin D deficiency.

IV. EFFECTS OF BODY HABITUS A. Vitamin D Metabolism Obese individuals have a lower serum 25OHD values than nonobese subjects as a result of low serum

C. Bone Histomorphometry Whereas obesity is associated with vitamin D deficiency and a low serum 25(OH)D, bone histomorphometry of the iliac crest is almost always normal. In one study, one of 24 grossly obese individuals was found to have mild osteomalacia and secondary hyperparathyroidism and a second individual had changes associated with increased skeletal remodeling [157]. Morbid obesity is sometimes treated with partial or total biliopancreatic bypass. When bone histomorphometric analysis was performed 1 to 5 years after bypass in 41 patients, all of them had a normal serum 25OHD. However, almost three quarters of them had abnormal bone mineralization, diminished bone formation, and increased bone resorption [158]. Thus, following biliopancreatic bypass for treatment of obesity,

796 osteomalacia is common despite the absence of vitamin D deficiency.

V. EFFECTS OF EXERCISE Weight lifting and other weight-bearing exercises increase bone mass of the lumbar spine, trochanter, femoral neck, and lower extremities in adult and adolescent males [159–162]. Men who had undergone muscle-building exercise for at least 1 year showed increases in serum osteocalcin, an index of bone formation, serum 1,25(OH)2D, and urinary cyclic adenosine 3′,5′-monophosphate compared to men who were sedentary [169]. Serum 1,25(OH)2D and osteocalcin did not correlate with each other. Thus, exercise may increase bone formation, serum 1,25(OH)2D, calcium absorption and bone mineral density. In older individuals, exercise increased total body calcium and bone mineral density of the lumbar spine [163–165]. In aging men, muscle strength of the back correlated with bone mineral density of the lumbar spine and midradius [165]. In postmenopausal women, aerobic, weight-bearing and resistance exercises each increased bone mineral density of the spine and walking increased bone mineral density of the hip [166,167]. In college women, bone mineral density of the lumbar spine was increased in tennis players but not in swimmers, indicating the necessity that the exercise be weight bearing [168].

VI. SUMMARY A number of factors influence vitamin D metabolism. Among these are race, geographic locus, sunlight exposure, diet and cultural traditions, body habitus, and exercise. Dark-skinned individuals who live in temperate regions including blacks, Asian Indians, Pakistanis, Hispanics, and others have diminished production of vitamin D3 in the skin because photons of light energy are absorbed by melanin instead of 7-dehydrocholesterol. Consequently, decreases in serum 25OHD and urinary calcium and increases in circulating PTH occur. Further, vitamin D–deficiency rickets may occur in breast-fed infants of any race if they do not receive supplements of the vitamin and are kept out of sunlight since the contents of vitamin D and 25(OH)D in human milk are inadequate. Infants of dark-skinned mothers are particularly susceptible in this regard. Vitamin D–deficient rickets and osteomalacia occur in dark-skinned Asian Indians and Pakistanis who live in temperate regions. Vitamin D deficiency caused by inadequate dietary vitamin D intake and avoidance of

MEHRDAD RAHMANIYAN AND NORMAN H. BELL

sunlight occurs in Arabs of all ages and the populations from the Middle East and leads to rickets in children and infants and osteoporosis and fractures in adults. Macrobiotic and vegetarian diets are deficient in vitamin D, and infants fed macrobiotic diets are prone to develop rickets. In obese men and women, serum vitamin D is markedly reduced, probably because vitamin D is fat soluble and is stored in fat. Whereas serum 25OHD and urinary calcium are reduced and circulating PTH is increased in obesity, osteomalacia is usually not a clinical problem. Weight-lifting and muscle-building exercise apparently leads to increases in serum 1,25(OH)2D, and urinary calcium, and weightbearing exercise increases bone mineral density and thus may help to prevent osteoporosis and fractures. In summary, vitamin D deficiency is common in peoples from many races throughout the world in both developed and undeveloped countries, even those near the equator. Thus, it is widespread despite the ready availability of assays for serum 25(OH)D to ascertain vitamin D status and the known necessity for supplementary dietary vitamin D and adequate exposure to sunshine, particularly during pregnancy, childhood, adolescence, and aging.

References 1. Holick MF, MacLaughlin JA, Doppelt SH 1981 Regulation of cutaneous previtamin D photosynthesis in man: Skin pigment is not an essential regulator. Science 211:590–593. 2. Holick MF 1981 The cutaneous synthesis of previtamin D3: A unique photo-endocrine system. J Invest Dermatol 77:51–58. 3. Cheng JB, Motola DL, Mangelsdorf DJ, Russell DW 2003 De-orphanization of cytochrome P450 2R1: a microsomal vitamin D-25-hydroxylase. J Biol Chem 278:38084–38093. 4. Cheng JB, Levine MA, Bell NH, Mangelsdorf DJ, Russell DW 2004 Genetic evidence that human CYP2R1 is a key vitamin D 25-hydroxylase. Proc Natl Acad Sci USA 101: 7711–7715. 5. Garabedian M, Holick MF, DeLuca HF, Boyle IT 1972 Control of 25-hydroxycholecalciferol metabolism by parathyroid glands. Proc Natl Acad Sci USA 69:1673–1676. 6. Henry HL 1989 Parathyroid hormone modulation of 25-hydrovitamin D3 metabolism by cultured chick kidney cells is mimicked and enhanced by forskolin. Endocrinology 116:503–510. 7. Horiuchi N, Suda T, Takahashi H, Shimazawa E, Ogata E 1977 In vivo evidence for the intermediary role of 3′,5′cyclic AMP in parathyroid hormone-mediated stimulation of 1,25-dihydroxyvitamin D3 synthesis in rats. Endocrinology 101:969–974. 8. Murayama A, Takeyama K, Kitanaka S, Kodera Y, Kawaguchi Y, Hosoya T, Kato S 1999 Positive and negative regulations of the renal 25-hydroxyvitamin D3 1α-hydroxylase gene by parathyroid hormone, calcitonin, and 1α,25 (OH)2D3 in intact animals. Endocrinology 140:2224–2231. 9. Caverzasio J, Montessuit C, Bonjour JP 1990 Stimulatory effect of insulin-like growth factor-I on renal Pi transport

CHAPTER 47 Effects of Race, Geography, Body Habitus, Diet, and Exercise on Vitamin D Metabolism

10. 11. 12.

13. 14.

15. 16. 17. 18. 19.

20. 21. 22. 23. 24.

25. 26.

27.

28.

and plasma 1,25-dihydroxyvitamin D3. Endocrinology 127: 453–459. Nesbitt T, Drezner MK 1993 Insulin-like growth factor-I regulation of renal 25-hydroxyvitamin D-1α-hydroxylase activity. Endocrinology 132:133–138. Boyle IT, Graw RW, DeLuca HF 1971 Regulation by calcium of in vitro synthesis of 1,25-dihydroxycholecalciferol. Proc Natl Acad Sci USA 68:2131–2134. Bushinsky DA, Riera G, Mavus MJ, Coe FL 1985 Evidence that blood ionized calcium can regulate serum 1,25(OH)2D3 independently of PTH and phosphorus in the rat. J Clin Invest 76:1599–1604. Tanaka Y, DeLuca HF 1973 The control of 25-hydroxyvitamin D metabolism by inorganic phosphorus. Arch Biochem Biophys 154:566–574. Portale AA, Halloran BP, Murphy MM, Morris RC Jr 1986 Oral intake of phosphorus can determine the serum concentration of 1,25-dihydroxyvitamin D by determining its production rate in humans. J Clin Invest 77:7–12. Henry HL 1979 Regulation of the hydroxylation of 25-hydroxyvitamin D3 in vivo and in primary cultures of chick kidney cells. J Biol Chem 254:2722–2729. Trechsel U, Bonjour J-P, Fleisch H 1978 Regulation of the metabolism of 25-hydroxyvitamin D in primary cultures of chick kidney cells. J Clin Invest 64:206–217. Henry HL 1992 Vitamin D hydroxylases. J Cell Biochem 49:4–9. Henry HL 2001 The 25(OH)D3/1α,25(OH)(2)D3-24-hydroxylase: a catabolic biosynthetic enzyme? Steroids 66:391–398. Halloran BP, Bikle DD, Levens MJ, Castro ME, Globu RK, Holton E 1986 Chronic 1,25-dihydroxyvitamin D3 administration in the rat reduces the serum concentration of 25-hydroxyvitamin D by increasing metabolic clearance rate. J Clin Invest 78:622–628. Bell NH, Shaw S, Turner RT 1984 Evidence that 1,25-dihydroxyvitamin D3 inhibits the hepatic production of 25-hydroxyvitamin D in man. J Clin Invest 74:1540–1544. McSheely PM, Bibby NJ 1985 Osteoblastic cells mediate osteoclastic responsiveness to parathyroid hormone. Endocrinology 118:824–828. Agus ZS , Gardner LB, Beck LH, Goldberg M 1973 Effects of parathyroid hormone on renal tubular reabsorption of calcium, sodium and phosphate. Am J Physiol 224:1143–1148. Favus MJ 1985 Factors that influence absorption and secretion of calcium in the small intestine and colon. Am J Physiol 248:G147–G157. Bikle DD, Zolock DT, Munson S 1984 Differential response of duodenal epithelial cells to 1,25-dihydroxyvitamin D3 according to position in the villus: A comparison of calcium uptake, calcium-binding protein, and alkaline phosphatase activity. Endocrinology 115:2077–2084. Silver J, Yalcindag C, Sela-Brown A, Kilav R, Naveh-Many T 1999 Regulation of the parathyroid hormone gene by vitamin D, calcium and phosphate. Kidney Int Suppl 73:S2–S7. Brown EM, Gamba G, Riccardi D, Lombardi M, Butters R, Kifor O, Sun A, Hediger MA, Lytton J, Hebert SC 1993 Cloning and characterization of an extracellular Ca2+-sensing receptor from bovine parathyroid. Nature 366:575–580. Naveh-Many T, Friedlaender MM, Mayer, Silver J 1989 Calcium regulates parathyroid hormone messenger ribonucleic acid (mRNA), but not calcitonin mRNA in vivo in the rat. Dominant role of 1,25-dihydroxyvitamin D. Endocrinology 125:275–280. Silver J, Russell J, Sherwood LM 1985 Regulation by vitamin D metabolites of messenger ribonucleic acid for

29.

30.

31. 32. 33. 34. 35.

36. 37.

38.

39.

40. 41. 42. 43. 44.

45.

46.

797

preproparathyroid hormone in isolated bovine parathyroid cells. Proc Natl Acad Sci USA 82:4270–4273. Naveh-Many T, Marx R, Keshet E, Pike JW, Silver J 1990 Regulation of 1,25-dihydroxyvitamin D3 receptor gene expression by 1,25-dihydroxyvitamin D3 in the parathyroid in vivo. J Clin Invest 86:1968–1975. M’Buyamba-Kabangu JR, Fagard R, Lijnen P, Bouillon R, Lissens W, Amery A 1987 Calcium, vitamin D-endocrine system and parathyroid hormone in black and white males. Calcif Tissue Int 41:70–74. Clemens TL, Henderson SL, Adams JS, Holick MF 1982 Increased skin pigment reduces the capacity of skin to synthesize vitamin D3. Lancet 1:74–76. Bell NH, Greene A, Epstein S, Oexmann MJ, Shaw S, Shary J 1985 Evidence for alteration of the vitamin D-endocrine system in blacks. J Clin Invest 76:470–473. Hollis BW, Pittard WB III 1984 Evaluation of the total fetomaternal vitamin D relationships at term: Evidence for racial differences. J Clin Endocrinol Metab 59:652–657. Luckey MM, Meier DE, Clemens TL 1986 PTH and vitamin D metabolites in premenopausal white and black women. J Bone Miner Res 1(Suppl. 1):381. Dawson-Hughes B, Harris S, Kramich C, Dallal G, Rasmussen HM 1993 Calcium retention and hormone levels in black and white women on high and low calcium diets. J Bone Miner 8:779–787. Katz BS, Jackson GJ, Hollis BW, Bell NH 1992 Diagnostic criteria of vitamin D deficiency. Endocrinologist 3:248–253. Bell NH, Yergey AL, Vieira NE, Oexmann MJ, Shary JR 1993 Demonstration of a difference in urinary calcium, not calcium absorption, in black and white adolescents. J Bone Miner Res 8:1111–1115. Nesby-O’Dell S, Scanlon KS, Cogswell ME, Gillespie C, Hollis BW, Looker CA, Allen C, Doughertly C, Gunter EW, Bowman BA 2002 Hypovitaminosis D prevalence and determinants among African American and white women of reproductive age: Third National Health and Nutrition Examination Survey, 1988–1994. Am J Clin Nutr 76:187–192. Fuleihan GE-H, Gundberg CM, Gleason R, Brown EM, Stromski ME, Grant FD, Conlin PR 1994 Racial differences in parathyroid hormone dynamics. J Clin Endocrinol Metab 79:1642–1647. Bell NH 1995 25-Hydroxyvitamin D3 reverses alteration of the vitamin D-endocrine system in blacks. Am J Med 99: 597–599. Abrams SA, O’Brien KO, Liang LK, Stuff JE 1995 Differences in calcium absorption and kinetics between black and white girls aged 5–16 years. J Bone Miner Res 10:829–833. Modlin M 1967 Urinary calcium in normal adults and in patients with renal stones: An interracial study. Invest Urol 5:49–57. Ghandur-Mnaymneh L, Cassady J, Hajanpour MA, Reiss E 1986 The parathyroid gland in health and disease. Am J Pathol 125:292–299. Cosman F, Morgan DC, Nieves JW, Shen V, Luckey MM, Dempster DW, Lindsay R, Parisien M 1997 Resistance to bone resorbing effects of PTH in black women. J Bone Miner Res 12:958–966. Awumey E, Hollis BW, Bell NH 1996 Low serum 25-hydroxyvitamin D in blacks results from decreased production rate and not increased metabolic clearance rate. J Bone Miner Res 11:8165 (abstract). Bachrach S, Fisher J, Parks JS 1980 An outbreak of vitamin D deficiency rickets in a susceptible population. Pediatrics 64:277–283.

798 47. Edidin DV, Levitsky LL, Schey W, Dumbovic M, Campos A 1980 Resurgence of nutritional rickets associated with breast-feeding and special dietary practices. Pediatrics 65:232–235. 48. Kruger DM, Lyne ED, Kleerekoper M 1987 Vitamin D deficiency rickets. A report of three cases. Clin Orthop 224: 277–283. 49. Key LL 1992 Vitamin D deficiency rickets. Trends Endocrinol Metab 2:81–85. 50. Chang YT, Germain-Lee EL, Doran TF, Migeon CJ, Levine MA, Berkovitz GD 1992 Hypocalcemia in nonwhite breast-fed infants. Clin Pediat 31:695–698. 51. Hollis BW, Roos BA, Draper HH, Lambert PW 1981 Vitamin D and its metabolites in human and bovine milk. J Nutr 111: 1240–1248. 52. Pittard WB, Geddes KM, Husley TC, Hollis BW 1991 How much vitamin D for neonates? Am J Dis Child 145:1147–1149. 53. Hui SL, Dimeglio LA, Longcope C, Peacock M, McClintock R, Perkins AJ, Johnston CC Jr 2003 Difference in bone mass between black and white American children: Attributable to body build, sex hormone levels, or bone turnover? J Clin Endocrinol Metab 88:642–649. 54. Bryant RJ, Wastney ME, Martin BR, Wood O, McCabe GP, Morshidi M, Smith DL, Peacock M, Weaver CM 2003 Racial differences in bone turnover and calcium metabolism in adolescent females. J Clin Endocrinol Metab 88:1043–1047. 55. Wright NM, Papadea N, Veldhuis JD, Bell NH 2002 Growth hormone secretion and bone mineral density in prepubertal black and white boys. Calcif Tissue Int 70:146–152. 56. Wright NM, Renault J, Willi S, Veldhuis JD, Pandey JP, Gordon L, Key LL, Bell NH 1995 Greater secretion of growth hormone in black than in white men: Possible factor in greater bone mineral density—A clinical research center study. J Clin Endocrinol Metab 80:2291–2297. 57. Liel Y, Edwards J, Shary J, Spicer KM, Gordon L, Bell NH 1988 The effects of race and body habitus on bone mineral density of the radius, hip and spine in premenopausal women. J Clin Endocrinol Metab 66:1247–1250. 58. DeSimone DP, Stevens J, Edwards J, Shary J, Gordon L, Bell NH 1989 Influence of body habitus and race on bone mineral density of the midradius, hip and spine in aging women. J Bone Miner Res 4:827–830. 59. Wright NM, Papadea N, Willi S, Veldhuis JD, Pandey JP, Key LL, Bell NH 1996 Demonstration of a lack of racial difference in secretion of growth hormone despite a racial difference in bone mineral density in premenopausal women—A clinical research center study. J Clin Endocrinol Metab 81:1023–1026. 60. Weinstein RS, Bell NH 1988 Diminished rates of bone formation in normal black adults. N Engl J Med 319: 1698–1701. 61. Gyepes M, Melliaz HZ, Katz I 1962 The low incidence of fracture of the hip in the Negro. JAMA 181:1073–1074. 62. Silverman SL, Madison RE 1988 Decreased incidence of hip fracture in hispanics, Asians and Blacks: California hospital discharge data. Am J Public Health 78:1482–1483. 63. Kellie SE, Brody JA 1990 Sex-specific and race-specific hip fracture rates. Am J Public Health 80:326–328. 64. Griffin MR, Ray WA, Fought RL, Melton LJ III 1992 Black-white differences in fracture rates. Am J Epidemiol 136:1378–1385. 65. Pruzansky ME, Turano M, Luckey M, Senie R 1989 Low body weight is a risk factor for hip fracture in both black and white women. J Orthop Res 7:192–197.

MEHRDAD RAHMANIYAN AND NORMAN H. BELL

66. Grisso AJ, Kelsey JL, Strom BL, O’Brien LA, Maislin G, La Pann K, Samelson L, Hoffman S 1994 Risk factors for hip fracture in black women. The Northeast Hip Fracture Study Group. N Engl J Med 330:1555–1559. 67. Cummings SR, Cauley JA, Palermo L, Ross PD, Wasnich RD, Black D, Faulkner KG 1994 Racial differences in hip axis lengths might explain racial differences in rates of hip fracture. Osteoporosis Int 4:226–229. 68. Solomon L 1968 Osteoporosis and fracture of the femoral neck in the South African Bantu. J Bone Joint Surg 50:2–5. 69. Solomon L 1979 Bone density in aging Caucasian and African populations. Lancet 3:1326–1329. 70. Patel DN, Pettifor JM, Becker PJ, Grieve C, Leschner K 1993 The effect of ethnicity on appendicular bone mass in white, coloured and Indian school children. S Afr Med J 83: 847–853. 71. Daniels ED, Pettifor JM, Schnitzler CM, Russel SW, Paten DN 1995 Ethnic differences in bone density in female South African nurses. J Bone Miner Res 10:359–367. 72. Meier De, Luckey MM, Wallenstein S, Lapinski RH, Catherwood B 1992 Racial differences in pre- and postmenopausal bone homeostasis: Association with bone density. J Bone Miner Res 7:1181–1189. 73. Jackson G, Hollis BW, Eyre DR, Baylink DJ, Bell NH 1994 Effects of race and calcium intake on bone markers and calcium metabolism in young adult men. J Bone Miner Res 9:S185 (abstract). 74. Eriksen EF, Mosekilde L, Melsen F 1986 Kinetics of trabecular bone resorption and formation in hypothyroidism: Evidence for a positive balance per remodeling cycle. Bone 7:101–108. 75. Chestnut CH, McClung MR, Ensrud KE, Bell NH, Genant HK, Harris ST, Singer FR, Stock JL, Yood RA, Delmas PD, Pryor-Tillotson S, Santora AC 1995 Alendronate treatment of the postmenopausal osteoporotic woman: Effect of multiple dosages on bone mass and bone remodeling. Am J Med 99:144–152. 76. Liberman UA, Weiss SR, Broil J, Minne HW, Quan H, Bell NH, Rodriquez-Portales J, Downs RW, Dequecker J, Favus M, Capizzi T, Santora II AC, Lombardi A, Shah RV, Hirsch LJ, Karpf DB 1995 Effect of three years treatment with oral alendronate on fracture incidence in women with postmenopausal osteoporosis. N Engl J Med 33: 1437–1443. 77. Schnitzler CM, Pettifor JM, Mesqita JM, Bird MD, Schnaid E, Smyth AE 1990 Histomorphometry of iliac crest bone in 346 normal black and white South African adults. Bone Miner 10:183–199. 78. Awumey EM, Mitra DA, Hollis BW, Kumar R, Bell NH 1998 Vitamin D metabolism is altered in Asian Indians in the southern United States: A clinical research center study. J Clin Endocrinol Metab 83:169–73. 79. Preece MA, Ford JA, Mclntosh WB, Dunnigan MG, Tomlinson S, O’Riordan JLH 1973 Vitamin D deficiency among Asian immigrants to Britain. Lancet 1:907–910. 80. Dent CE, Gupta MM 1975 Plasma 25-hydroxyvitamin-D levels during pregnancy in Caucasians and in vegetarian and non-vegetarian Asians. Lancet 2:1057–1060. 81. Hunt SP, O’Riordan JLH, Windo J, Truswell AS 1976 Vitamin D status in different sub-groups of British Asians. Br Med J 2:1351–1354. 82. Heckmatt JZ, Peacock M, Davies AEJ, McMurray J, Isherwood DM 1979 Plasma 25-hydroxyvitamin D in pregnant Asian women and their babies. Lancet 2:546–548.

CHAPTER 47 Effects of Race, Geography, Body Habitus, Diet, and Exercise on Vitamin D Metabolism

83. Brooke OG, Brown IRF, Cleeve HJW, Sood A 1981 Observations on the vitamin D state of pregnant Asian women in London. Br J Obstet Gynaecol 88:18–26. 84. Ellis G, Woodhead JS, Cooke WT 1977 Serum 25-hydroxyvitamin-D concentrations in adolescent boys. Lancet 1:825–828. 85. O’Hare AE, Uttley WS, Belton NR, Westwood A, Levin SD, Anderson F 1984 Persisting vitamin D deficiency in the Asian adolescent. Arch Dis Child 59:766–770. 86. Ford JA, Davidson DC, Mclntosh WB, Fyfe WM, Dunnigan MG 1973 Neonatal rickets in Asian immigrant population. Br Med J 3:211–212. 87. Arneil GC, Crosbie JC 1963 Infantile rickets returns to Glasgow. Lancet 2:423–425. 88. Goel KM, Sweet EM, Logan RW, Warren JM, Arneil GC, Shanks RA 1976 Florid and subclinical rickets among immigrant children in Glasgow. Lancet 1:1141–1145. 89. Henderson JB, Dunnigan MG, Mclntosh WB, Abdul-Motaal AA, Gettinby G, Glekin BM 1987 The importance of limited exposure to UV radiation and dietary factors in the aetiology of Asian rickets: A risk factor model. Q J Med 63:413–425. 90. Dent CE, Rowe DJF, Round JM, Stamp TC B 1973 Effect of chapattis and ultraviolet irradiation on nutritional rickets in an Indian immigrant. Lancet 1:1282–1284. 91. Wills MR, Day RC, Phillips JB, Bateman EC 1972 Phytic acid and nutritional rickets in immigrants. Lancet 1:771–773. 92. Holmes AM, Enoch BA, Taylor JL, Jones ME 1973 Occult rickets and osteomalacia amongst the Asian immigrant population. Q J Med 42:125–149. 93. Ford JA, Colhoun EM, Mclntosh WB, Dunnigan MG 1972 Rickets and osteomalacia in the Glasgow Pakistani community, 1961–1971. Br Med J 1:677–679. 94. Preece MA, Tomlinson S, Ribot CA, Pietrek J, Korn HT, Davies DM, Ford JA, Dunnigan MG, O’Riordan JLH 1975 Studies of vitamin D deficiency in man. Q J Med 44: 575–589. 95. Finch PJ, Ang L, Colston KW, Nisbet J, Maxwell JD 1992 Blunted seasonal variation in serum 25-hydroxyvitamin D and increased risk of osteomalacia in vegetarian London Asians. Eur J Clin Nutr 46:509–515. 96. Henderson JB, Dunnigan MG, Mclntosh WB, Motaal AA, Hole D 1990 Asian osteomalacia is determined by dietary factors when exposure to ultraviolet radiation is restricted: A risk factor model. Q J Med 76:923–933. 97. Teotia M, Teotia SP 1997 Nutritional and metabolic rickets. Indian J Pediatr 64:153–157. 98. Atiq M, Suria A, Nizami SQ, Ahmed I 1998 Vitamin D status of breastfed Pakistani infants. Acta Paediatr 87:737–740. 99. Pietrek J, Preece MA, Windo J, O’Riordan J, Dunnigan MG, Mclntosh WB, Ford JA 1976 Prevention of vitamin-D deficiency in Asians. Lancet 1:1145–1148. 100. Dunnigan MG, Mclntosh WB, Sutherland GR, Gardee R, Glekin B, Ford JA, Robertson I 1981 Policy for prevention of Asian rickets in Britain: A preliminary assessment of the Glasgow rickets campaign. Br Med J 1:357–360. 101. Lo CW, Paris PW, Holick MF 1986 Indian and Pakistani immigrants have the same capacity as Caucasians to produce vitamin D in response to ultraviolet irradiation. Am J Clin Nutr 44:683–685. 102. Rashid A, Mohammed T, Stephens WP, Warrington S, Berry JL, Mawer EB 1983 Vitamin D state of Asians living in Pakistan. Br Med J 1:182–184. 103. Lamberg B Allardt C, Karkkainen M, Seppanen R, Bistrom H 1993 Low serum 25-hydroxyvitamin D concentrations and

104. 105. 106. 107.

108. 109. 110.

111.

112.

113.

114.

115.

116. 117. 118.

119. 120.

121. 122.

799

secondary hyperparathyroidism in middle-aged strict vegetarians. Am J Clin Nutr 58:684–689. Ahmed I, Atiq M, Iqbal J, Khurshid M, Whittaker P 1995 Vitamin D deficiency rickets in breast-fed infants presenting with hypocalcemic seizures. Acta Paediatr 84:941–942. Cundy C, Cornish J, Evans MC, Gamble G, Stapleton J, Reid IR 1995 Sources of interracial variation in bone mineral density. J Bone Miner Res 10:368–373. Parker M, Anand JK, Myles JW, Lodwick R 1992 Proximal femoral fractures: Prevalance in different racial groups. Eur J Epidemiol 8:730–732. Fong JC, Huong YM, Shen KY, Dewen PL 1980 Prevention and treatment of rickets (report). Zhengzhou, Henan, People’s Republic of China: Henan College of Medicine publication (Chinese), 1–10. Xiao L 1982 Survey of infant rickets. J Chinese Pediatr (Chinese) 20:63–64. Danhui Z, Quingbing Z, Yan X 1990 Serum 25OHD levels in maternal and cord blood in Beijing, China. Acta Paediatr Scand 79:1240–1241. Specker BL, Ho ML, Oestreich A, Yin T-A, Shui Q-M, Chen X-C, Tsang RC 1992 Prospective study of vitamin D supplementation and rickets in China. J Pediatr 120: 733–739. Specker BL, Valanis B, Hertzberg V, Edwards N, Tsang RC 1985 Sunshine exposure and serum 25-hydroxyvitamin D concentrations in breast-fed infants in Beijing, China. J Pediatr 107:928–931. Ling X, Cummings SR, Mingwei Q, Xihe Z, Xioashu C, Nevitt M, Stone K 2000 Vertebral fractures in Beijing, China: the Beijing Osteoporosis Project. J Bone Miner Res 15: 2019–2025. Bhudhikanok GS, Wang MC, Eckert K, Matkin C, Marcus R, Bachrach LK 1996 Differences in bone mineral in young Asian and Caucasian Americans may reflect differences in bone size. J Bone Miner Res 11:1545–1556. Abrams SA, Copeland KC, Gunn SK, Stuff JE, Clarke LL, Ellis KJ 1999 Calcium absorption and kinetics are similar in 7- and 8-year-old Mexican-American and Caucasian girls despite hormonal differences. J Nutr 129:666–671. Reasner III CA, Dunn JF, Fetchick DA, Leil Y, Hollis BW, Epstein S, Shary J, Mundy GR, Bell NH 1990 Alteration of vitamin D metabolism in Mexican-Americans. J Bone Miner Res 5:13–17. Bauer RL, Diehl AK, Barton SA, Brender JA, Deyo RA 1986 Risk of postmenopausal hip fracture in Mexican-American women. Am J Public Health 76:1020–1021. Bauer RL, Deyo RA 1987 Low risk vertebral fracture in Mexican-American women. Arch Intern Med 147:1437–1439. Taaffe DR, Villa ML, Holloway L, Marcus R 2000 Bone mineral density in older non-Hispanic Caucasian and Mexican-American women: Relationship to lean and fat mass. Ann Hum Biol 27:331–344. Lauderdale DS, Jacobsen SJ, Furner SE, Levy PS, Brody JA, Goldberg J 1998 Hip fracture incidence among elderly Hispanics. Am J Public Health 88:1245–1247. Reid IR, Cullen S, Schooler BA, Livingstone NE, Evans MC 1990 Calciotropic hormone levels in Polynesians, evidence against their role in interracial differences in bone mass. J Clin Endocrinol Metab 70:1452–1456. Reid IR, Mackey M, Ibbertson HK 1986 Bone mineral content in Polynesian and white New Zealand women. Br Med J 2:1457–1458. Chin K, Evans MC, Cornish J, Cundy T, Reid IR 1997 Differences in hip axis and femoral neck length in

800

123.

124. 125. 126. 127. 128.

129. 130.

131. 132. 133. 134.

135. 136. 137. 138.

139.

140.

MEHRDAD RAHMANIYAN AND NORMAN H. BELL

premenopausal women of Polynesian, Asian and European origin. Osteoporos Int 7:344–347. Abdullah MA, Salhi HS, Bakry LA, Okamoto E, Abomelha AM, Stevens B, Mousa FM 2002 Adolescent rickets in Saudi Arabia: A rich and sunny country. J Pediatr Endocrinol Metab 15:1017–1025. Fonseca V, Tongia R, El-Hazmi M, Abu-Aisha H 1984 Exposure to sunlight and vitamin D deficiency in Saudi Arabian women. Postgrad Med J 60:589–591. Serinius F, Elidrissy A, Dandona P 1984 Vitamin D nutrition in pregnant women at term and in newly born babies in Saudi Arabia. J Clin Pathol 37:444–447. Sedrani SH 1984 Low 25-hydroxyvitamin D and normal serum calcium concentrations in Saudi Arabia: Riyadh region. Ann Nutr Metab 28:181–185. Al-Arabi KM, Wahab A, Elidrissy TH, Sedrani SH 1984 Is avoidance of sunlight a cause of fractures of the femoral neck in elderly Saudis? Trop Geogr Med 36:273–279. Ghannam NN, Hammami MM, Bakheet SM, Khan BA 1999 Bone mineral density of the spine and femur in healthy Saudi females: Relation to vitamin D status, pregnancy, and lactation. Calcif Tissue Int 65:23–28. Al Faraj S, Al Mutairi K 2003 Vitamin D deficiency and chronic low back pain in Saudi Arabia. Spine 28:177–179. Dawodu A, Absood G, Patel M, Agarwal M, Ezimokhai M, Abdulrazzaq Y, Khalayli G 1998 Biosocial factors affecting vitamin D status of women of childbearing age in the United Arab Emirates. J Biosoc Sci 30:431–437. Gannage-Yared MH, Brax H, Asmar A, Tohme A 1998 [Vitamin D status in aged subjects. Study of a Lebanese population]. Presse Med 27:900–904. Bassir M, Laborie S, Lapillonne A, Claris O, Chappuis MC, Salle BL 2001 Vitamin D deficiency in Iranian mothers and their neonates: A pilot study. Acta Paediatr 90:577–579. Alagol F, Shihadeh Y, Boztepe H, Tanakol R, Yarman S, Azizlerli H, Sandalci O 2000 Sunlight exposure and vitamin D deficiency in Turkish women. J Endocrinol Invest 23:173–177. Meulmeester JF, van den Berg H, Wedel M, Boshuis PG, Hulshof KFAM, Luyken R 1990 Vitamin D status, parathyroid hormone and sunlight in Turkish, Moroccan and Caucasian children in The Netherlands. Eur J Clin Nutr 44:461–470. Shany S, Hirsh J, Berlyne GM 1976 25-Hydroxycholecalciferol levels in Bedouins in the Negev. Am J Clin Nutr 29:1104–1107. Elzouki AY, Markestad T, Elgarrah M, Elhoni N, Aksnes L 1989 Serum concentrations of vitamin D metabolites in rachitic Libyan children. J Pediatr Gastroenterol Nutr 9:507–512. Gessner BD, Plotnik J Muth PT 2003 25-Hydroxyvitamin D levels among healthy children in Alaska. J Pediatr 143: 434–437. Pasco JA Henry MJ, Nicholson GC Sanders KM, Kotowicz MA 2001 Vitamin D status of women in the Geelong Osteoporosis Study: Association with diet and casual exposure to sunlight. Med J Aust 175:401–405. Ladizesky M, Lu Z, Oliveri B, San Roman N, Diaz S, Holick MF, Mautalen C 1995 Solar ultraviolet B radiation and photoproduction of vitamin D3 in central and southern parts of Argentina. J Bone Miner Res 10:545–548. Outila TA, Karkkainen MU, Seppanen RH, Lamberg-Allardt CJ 2000 Dietary intake of vitamin D in premenopausal, healthy vegans was insufficient to maintain concentrations of serum 25-hydroxyvitamin D and intact parathyroid hormone within normal ranges during the winter in Finland. J Am Diet Assoc 100:434–441.

141. Dagnelie PC, Vergote FJVRA, van Staveren WA, van den Berg H, Dingjan PG, Hautvast JGAJ 1990 High prevalence of rickets in infants on macrobiotic diets. Am J Clin Nutr 51:202–208. 142. Parsons TJ, van Dusseldorp M, van der Bliet M, van de Werken K, Schaafsma G, van Staveren WA 1997 Reduced bone mass in Dutch adolescents fed a macrobiotic diet in early life. J Bone Miner Res 12:1486–1494. 143. Compston JE, Vedi S, Ledjer JE, Webb A, Gazet JC, Pilkington TR 1981 Vitamin D status and bone histomorphometry. Am J Clin Nutr 34:2359–2363. 144. Rickers H, Christiansen C, Balslev I, Rodbro P 1984 Impairment of vitamin D metabolism and bone mineral content after intestinal bypass surgery. Scand J Gastroenterol 19:184–189. 145. Bell NH, Epstein S, Greene A, Shary J, Oexmann MJ, Shaw S 1985 Evidence for alteration of the vitamin D-endocrine system in obese subjects. J Clin Invest 76:370–373. 146. Liel Y, Ulmer E, Shary J, Hollis BW, Bell NH 1988 Low circulating vitamin D in obesity. Calcif Tissue Int 43:199–201. 147 Bell NH, Epstein S, Shary J, Greene V, Oexmann MJ, Shaw S 1988 Evidence of a probable role for 25-hydroxyvitamin D in the regulation of calcium metabolism in man. J Bone Miner Res 3:489–495. 148. Rosenstreich SJ, Rich C, Volwiler W 1971 Deposition in and release of vitamin D3 from body fat: Evidence for a storage site in the rat. J Clin Invest 50:679–687. 149. Mawer EB, Backhouse J, Holman CA, Lumb GA, Stanbury SW 1972 The distribution and storage of vitamin D and its metabolites in human tissues. Clin Sci 43:414–431. 150. Epstein S, Bell NH, Shary J, Shaw S, Greene A, Oexmann MJ 1986 Evidence that obesity does not influence the vitamin D– endocrine system in blacks. J Bone Miner Res 1:181–184. 151. Wortsman J, Matsuoka LY, Chen TC, Lu Z, Holick MF 2000 Decreased bioavailability of vitamin D in obesity. Am J Clin Nutr 72:690–693. 152. DeSimone DP, Stevens J, Edwards J, Shary J, Gordon L, Bell NH 1989 Influence of body habitus and race on bone mineral density of the midradius, hip and spine in aging women. J Bone Miner Res 4:827–830. 153. Liel Y, Edwards J, Shary J, Spicer KM, Gordon L, Bell NH 1988 The effects of race and body habitus on bone mineral density of the radius, hip, and spine in premenopausal women. J Clin Endocrinol Metab 66:1247–1250. 154. Reid IR, Ames R, Evans MC, Sharpe S, Gamble F, France JT, Lim TMT, Cundy TF 1992 Determinants of total body and regional bone mineral density in normal postmenopausal women. A key role for fat mass. J Clin Endocrinol Metab 75:45–51. 155. Reid IR, Plank LD, Evans MC 1992 Fat mass is an important determinant of whole body bone density in premenopausal women but not in men. J Clin Endocrinol Metab 75:779–782. 156. Dalen N, Hallberg D, Lamke B 1975 Bone mass in obese subjects. Acta Med Scand 197:353–355. 157. Steiniche T, Vesterby A, Eriksen EF, Melsen F 1986 A histomorphometric determination of iliac bone structure and remodeling in obese subjects. Bone 7:77–82. 158. Compston JE Verdi S, Gianetta E, Watson G, Civalleri D, Scopinaro N 1984 Bone histomorphometry and vitamin D status after biliopancreatic bypass for obesity. Gastroenterology 87:350–356. 159. Nilsson BE, Westlin NE 1971 Bone density in athletes. Clin Orthop Related Res 77:179–182. 160. Granbed H, Jonson R, Hansson T 1987 The loads on the lumbar spine during extreme weight lifting. Spine 12:146–149.

CHAPTER 47 Effects of Race, Geography, Body Habitus, Diet, and Exercise on Vitamin D Metabolism

161. Colletti LA, Edwards J, Gordon L, Shary J, Bell NH 1989 The effects of muscle-building exercise on bone mineral density of the radius, spine, and hip in young men. Calcif Tissue Int 45:12–14. 162. Conroy BP, Kraemer WJ, Maresh CM, Fleck SJ, Stone MH, Fry AC, Miller PD, Kalsky GP 1993 Bone mineral density in elite junior Olympic weightlifters. Med Sci Sports Exercise 25:1103–1109. 163. Aloia JF, Cohn SH, Ostuni JA, Cane R, Ellis K 1978 Prevention of involutional bone loss by exercise. Ann Intern Med 89: 356–358. 164. Krolner B, Toft B, Pors Nielson KS, Tondevold E 1983 Physical exercise as prophylaxis against involutional vertebral bone loss: A controlled trial. Clin Sci 64:541–546. 165. Bevier W, Wiswell R, Pyka G, Kozac K, Newhall K, Marcus R 1989 Relationship of body composition, muscle strength and

166.

167.

168. 169.

801

aerobic capacity to bone density in older men and women. J Bone Miner Res 4:421–432. Bonaiuti D, Shea B, Iovine R, Negrini S, Robinson V, Kemper HC, Wells G, Tugwell P, Cranney A 2002 Exercise for preventing and treating osteoporosis in postmenopausal women. Cochrane Database Syst Rev 3:CD000333. Kelley GA, Kelley KS, Tran ZV 2002 Exercise and lumbar spine bone mineral density in postmenopausal women: A meta-analysis of individual patient data. J Gerontol A Biol Sci Med Sci 57:M599–M604. Jacobson PC, Beaver W, Grubb SA, Taft TN, Talmadge RV 1984 Bone density in women: College athletes and older athletic women. J Orthop Res 2:328–332. Bell NH, Godsen RN, Henry DP, Shary J, Epstein S 1988 The effects of muscle building exercise on vitamin D and mineral metabolism. J Bone Miner Res 3:369–373.

CHAPTER 48

Perinatal Vitamin D Actions NICHOLAS J. BISHOP I. II. III. IV. V. VI.

Academic Department of Child Health, University of Sheffield, United Kingdom

Introduction The Last Trimester of Pregnancy The Normal Term Infant The Term Growth-Retarded Infant The Premature Infant Infants of Diabetic Mothers

I. INTRODUCTION With the cutting of the umbilical cord, the provision of both mineral substrates and the placental factors that had participated in the regulation of skeletal maturation in utero ceases abruptly. The response of the neonate to these sudden changes depends in part on the reserves of calcium, phosphate, and vitamin D laid down during pregnancy. The role of vitamin D in the pregnant mother is dealt with elsewhere in this volume (see Chapter 51). Nevertheless, it is appropriate to include here a summary of the clinical consequences of deficiency or excess in relation to the status of the infant at birth and during the perinatal period. The events of the final trimester are important in establishing good nutritional reserves while maintaining rapid growth. Thus the period covered by this review spans the last 3 months of pregnancy and the first month of extrauterine life for infants born at term. For those born prematurely, the postnatal period covered is the first 3 months. In addition, because of the potential for metabolic bone disease to develop as a result of inadequate calcium and phosphate intake in preterm infants, vitamin D metabolism in this population is considered in the light of mineral substrate provision.

II. THE LAST TRIMESTER OF PREGNANCY A. Malnutrition In countries where vitamin D supplementation of table milk is routine, vitamin D deficiency is unlikely to arise during pregnancy except in recent immigrants with chronic dietary insufficiency of calcium, vitamin D, and other essential nutrients, in groups avoiding dairy products for cultural or dietary reasons (e.g., cow’s milk protein intolerance), and where sunlight exposure is negligible. In malnourished populations, VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

VII. Late Neonatal Hypocalcemia VIII. Recommendations for Vitamin D Intake in the Perinatal Period IX. Summary References

evidence for vitamin D deficiency causing osteomalacia in the mother and abnormal skeletal metabolism in the fetus and infant is strong [1–3]. Infants of severely malnourished mothers may be born with rickets and suffer fractures in the neonatal period. Although much of the literature relating to such infants was written early in the 20th century, recent reports from European centres indicate the need for continued vigilance, particularly in immigrant or refugee populations. Four infants were born with craniotabes to immigrant mothers with osteomalacia in Berlin [2]. The infants exhibited typical biochemical and radiological changes of rickets, which responded well to vitamin D therapy. Observational studies suggest that radiographic bone density is reduced in both malnourished mothers and their infants [3], and this reduction can be ameliorated by calcium supplementation during pregnancy.

B. Vitamin D Deficiency Vitamin D deficiency can of course occur in the absence of malnutrition. Vitamin D nutrition in pregnancy was investigated by Brooke and colleagues in 115 Asian (Pakistani, Hindu Indian, and East African Asian) women living in London, and in 50 of their newborn infants [4]. Maternal serum 25-hydroxyvitamin D (25OHD) concentration at the beginning of the last trimester was 20.2 nmol/liter (8.1 ng/ml), falling to 16.0 nmol/liter (6.4 ng/ml) after delivery. Postpartum, 36% of the women and 32% of the infants had undetectable 25OHD concentrations (less than 3 nmol/liter or 1.2 ng/ml). Alkaline phosphatase bone isoenzyme was elevated (compared with appropriate well-nourished controls) in 20% of the women postpartum, and in 50% of the infants. Five infants developed symptomatic hypocalcemia. Our own recent studies of vitamin D levels in the cord blood of infants born to women pregnant during the spring and early summer months in Sheffield indicate Copyright © 2005, Elsevier, Inc. All rights reserved.

804 widespread vitamin D insufficiency. More than 60% of infants had cord blood levels of vitamin D below 20 nmol/liter. Of the subjects, 90% were white Caucasian (Am J Clin Nutr 2004).

C. Vitamin D Supplementation Maxwell and colleagues studied 126 Asian women whose mean serum 25OHD was 20 nmol/liter (8 ng/ml) at the end of the second trimester. A double-blind study of supplementary vitamin D (1000 IU per day) versus placebo during the third trimester to Asian women living in London was performed. There was increased maternal weight gain (63 versus 46 g/day) and a 50% reduction in the numbers of infants classified as “growth retarded” (born weighing less than 2500 g at term), which closely approached significance at the 5% level, in the supplemented group [5]. Infants in the control group also had larger fontanelles, suggesting delayed ossification. Further follow-up of the cohort was reported to age 1 year, with the observers still blinded to the original randomization. There was increasing divergence of the groups in terms of both weight and length, so that by 1 year of age the infants whose mothers had received the supplemental vitamin D during the third trimester were on average 0.4 kg heavier and 1.6 cm longer than the control group [6]. Marya et al. gave vitamin D 600,000 IU twice (during the seventh and eighth months of pregnancy), or a daily supplement of 1200 IU vitamin D and 375 mg calcium per day throughout the third trimester, or placebo, to Hindu women living in India. The high-dose vitamin D supplement had a greater effect on infant birth weight and cord blood levels of calcium, inorganic phosphate, and alkaline phosphatase activity than the daily supplement, which did not differ greatly in its effects from the placebo; however, there were no data regarding the compliance of the group taking the daily supplement [8]. In well-nourished mothers who do not receive vitamin D supplementation during pregnancy, the situation is less clear. Delvin and colleagues studied the effect of vitamin D supplementation from the end of the first trimester in a group of French mothers [7]. They reported differences (comparing supplemented with unsupplemented controls) in vitamin D metabolite levels at birth in both mothers and their infants. In addition, the postnatal fall in plasma calcium in the infants of unsupplemented mothers was more likely to be associated with symptomatic hypocalcemia. In contrast with the studies of malnourished mothers detailed above, there were no differences in maternal blood calcium or inorganic phosphate concentrations between

NICHOLAS J. BISHOP

groups, and the authors did not report any evidence of active rickets in the infants. There were no differences in birth weight or length between the groups. Further studies undertaken in France by Madelenat and colleagues examined the effect of 80,000 IU of vitamin D given in a single dose at around the end of the second trimester. In this study, which recruited primarily white Caucasian women, 34% of women had a serum 25-hydroxyvitamin D concentration in the osteomalacic range. Following supplementation only one woman continued to have a low 25-hydroxyvitamin D concentration at delivery. The study was undertaken during the winter months. There was no evidence of vitamin D toxicity occurring in the women receiving this dose. Thus, maternal malnutrition with coexisting vitamin D deficiency can result in metabolic bone disease and disturbed calcium and vitamin D metabolism in the neonate. Vitamin D supplementation in the malnourished mother results in improved growth of the child both in terms of birth weight and also subsequent linear growth during infancy. The neonatal metabolic bone disease resulting from maternal malnutrition is amenable, at least in the short term, to standard treatment with vitamin D and calcium supplements. There are no long-term data on the outcome for infants treated in this way. In women with better overall nutritional status, but not receiving routine vitamin D supplementation or consuming fortified foods, administration of vitamin D reduces the incidence of neonatal hypocalcaemia.

III. THE NORMAL TERM INFANT Stearns et al. in the 1930s showed that the rate of linear growth and weight gain in normal infants was related to vitamin D intake. Infants supplemented with 340 as opposed to 135 IU of vitamin D grew more rapidly in both weight and length [9]. Over the years there was a tendency to increase the amount of vitamin D given to infants so that by the 1950s some infants were receiving over 2000 IU per day. Contemporaneously, a number of cases of “idiopathic hypercalcemia” were reported. Hypervitaminosis D resulted in hypercalcemia with polyuria leading to dehydration and its typical consequences, which were documented in some case reports. Of more concern, a recent report suggests that some infants receiving intermittent high-dose vitamin D prophylaxis may go on to develop nephrocalcinosis [10]. Normal term infants born to vitamin D–sufficient mothers have plasma levels of total vitamin D metabolites that correlate closely with those of their mothers [11]. A number of studies have reported that total levels of

805

CHAPTER 48 Perinatal Vitamin D Actions

25OHD and 1,25-dihydroxyvitamin D [1,25(OH)2D], are decreased in cord blood compared with maternal blood at the time of delivery [11–13]; unbound (free) metabolic levels of 25OHD, 24,25(OH)2D and 25,26(OH)2D have been reported as higher in infants’ blood, with free 1,25(OH)2D levels being equal [11]. The first report of 1,25(OH)2D levels in infants born at term indicated that the initially low levels in cord blood rose to normal adult values by 24 hr of age [12]. Longitudinal measurement of vitamin D metabolites in the serum of breast-fed infants (not receiving vitamin D supplements) who were born to vitamin D–replete mothers suggested that depletion of vitamin D stores occurs within 8 weeks of delivery in the majority [11]. In 1963 the American Academy of Pediatrics recommended that, for infants, daily intakes of vitamin D be restricted to 400 IU from all sources [14]. These recommendations are still regarded as appropriate by most pediatricians, although a recent report from the Academy suggested that the recommended dietary allowance (RDA) should be 200 IU per day starting within the first 2 months of life and continuing throughout infancy, childhood, and adolescence [15]. The United Kingdom’s Department of Health recommends 340 IU vitamin D per day to age 6 months [16]. Sufficient vitamin D is available in normal reconstituted infant formulas to meet these recommendations. However, the vitamin D content of human milk is low. Unless the infant is exposed (face and hands) to sunlight for 10 min each day, there is a good case for providing an oral supplement of vitamin D of up to 400 IU per day. The effect of vitamin D supplementation on bone mineralization in wholly breast-fed infants was investigated by Greer et al. [17]. Infants who received 400 IU per day of vitamin D had higher bone mineral content and serum 25OHD levels at age 12 weeks than those not supplemented. The effect on growth and bone mineralization beyond this period remains unknown. There is thus no evidence that increasing the level of vitamin D supplementation beyond 400 IU per day influences either linear growth or bone mass in the immediate postnatal and infant period. Vitamin D supplementation in infancy may however have longer term consequences. Zamora and colleagues have documented a higher bone mineral mass of prepubertal girls who were breast-fed as infants and who received vitamin D supplements [18]. This retrospective study demonstrated increased areal bone mineral density in the femoral neck in the supplemented group. The increase in areal bone mineral density was thought to arise primarily as a result in an increase in bone size. A retrospective birth cohort study by Hyponnen and

colleagues [19] demonstrated an increased incidence of type I diabetes in young adults in Finland who as infants had not received the recommended dose of vitamin D (2000 IU/day) [50]. The potential for early exposure to influence long-term outcomes such as diabetes and osteoporosis is of considerable interest as the incidence of both these diseases is rising rapidly in all populations.

IV. THE TERM GROWTHRETARDED INFANT There is no direct evidence that vitamin D metabolism is altered in the infant who is growth retarded (small for gestational age) due to uteroplacental factors rather than maternal malnutrition. Reduced bone mineral content and reduced serum 1,25(OH)2D and osteocalcin levels have been documented for growth retarded as opposed to term infants, but no difference in 25OHD status was recorded [20]. The authors suggested that reduced uteroplacental blood flow resulted in reduced fetal production of 1,25(OH)2D and hence lower osteocalcin and reduced bone mineral accretion, but reduced transfer of all nutrients including minerals is also likely to have contributed.

V. THE PREMATURE INFANT A. Early Neonatal Hypocalcemia Early neonatal hypocalcemia is a common event occurring in up to 75% preterm infants, chiefly those born with very low birth weight (under 1500 g) [21]. It is usually of short duration and does not express itself clinically in the majority of infants. Immaturity of the vitamin D activation pathway has been suggested as a major underlying factor either alone or in combination with other abnormalities, particularly transient hypoparathyroidism, hypercalcitoninemia, and end-organ resistance to hormonal effects [22]. However, it has been clearly shown that there is an appropriate secretion of parathyroid hormone (PTH) in response to this hypocalcemic stimulus [23,24]. This increase in serum immunoreactive PTH concentration appeared within the initial 24 hr after birth, with levels of both intact PTH (1–84) and the carboxyl-terminal fragment (cPTH) following the same trend [25]. This physiological response to a hypocalcemic stimulus is substantiated by the observation that the increment in immunoreactive PTH levels was blunted when premature infants received calcium infusion; this calcium load buffered the postnatal depression of serium calcium.

806 By day 10 serum levels of PTH (1–84) and cPTH return to euparathyroid values [25,26].

B. Vitamin D in the Neonate

25OHD (nmol/L) ( )

80

350 300

60

250 200

40

150 100

20

50

1,25(OH)2D (pmol/L) ( )

In preterm as in full-term newborns, both total and free 25OHD cord blood levels were lower than in maternal blood and were correlated to those of the mothers [27–29]. Bouillon et al. [30] reported a positive correlation between maternal and cord serum concentrations of both total and free 1,25(OH)2D in premature babies; others found that only those of free 1,25(OH)2D were correlated [27]. This discrepancy could be due to the vitamin D depletion state of the subjects studied [25]. Cord and maternal blood vitamin D binding protein (DBP) levels were also positively related. Hirsfeld and Lunell [31] have excluded the possibility of a placental transfer of this protein by DBP polymorphism analysis; the most likely explanation for this fetomaternal relationship would therefore be common fetal and maternal factors affecting its synthesis. Longitudinal trends in total and free 25OHD and 1,25(OH)2D estimates for preterm infants in the first month of life are shown in Figs. 1 and 2. Many reports have clearly shown that in premature infants, after 28 weeks of gestation, activation of vitamin D is operative as early as 24 hr after birth [23,32–35]. In European countries where dairy products are not enriched with vitamin D, average levels of 25OHD in cord blood are lower than those in North America [23]. Vitamin D supplementation (from 500 to 2000 IU/day) in French preterm infants just after birth improved vitamin D nutritional status as evidenced by rising plasma 25OHD levels. In addition, the administration of vitamin D resulted in an increase in the circulating concentration of 1,25(OH)2D (Fig. 1). By 5 days of

Free 25OHD or 1,25(OH)2D (pmol/L)

NICHOLAS J. BISHOP

3.0

2.0

1.0

25OHD 1,25(OH)2D

0 Cord 0

5

10 15 Days of life

20

25

30

FIGURE 2

Newborn serum free 25OHD and 1,25(OH)2D levels as a function of age. Values are means ± SEM. From Delvin et al. [23] with permission.

age, the plasma levels of 1,25(OH)2D were well above the range observed in reference adolescent groups [33–35]. This sharp elevation was probably linked to hypocalcemia and the concomitant elevated PTH levels. Substrate concentration is a rate limiting factor in the synthesis of 1,25(OH)2D in the presence of hypocalcemia, and thus a strong positive correlation between serum 25OHD and 1,25(OH)2D concentrations was observed during the first 10 days of life over a wide range of 25OHD levels [23,35]. This was well illustrated by the report of Glorieux et al. [23] of twin preterm babies in a controlled study of vitamin D supplementation. Their levels of 25OHD at birth were identical and much higher than those measured in other preterm newborns. In the protocol of early supplementation, one of the twins was assigned to the vitamin D–supplemented group. 1,25(OH)2D increased in both infants in a similar fashion and paralleled the average increase recorded in the supplemented group. This observation emphasizes the importance of maternal 25OHD adequacy during pregnancy and indicates its potential for limitation of 1,25(OH)2D production. Serum bone Gla protein (BGP) values are high at birth (15 + 3 ng/ml); maternal and cord serum BGP levels were not correlated [36,37]. During the first month of life, serum BGP increases and parallels the changes in 1,25(OH)2D but without sustained correlation. These results indicate that serum BGP does not reflect changes in serum 1,25(OH)2D but rather probably the overall rate of bone formation or growth at the tissue level.

0

0 Cord 0

FIGURE 1

5

10 15 Days of life

20

25

30

Newborn serum total 25OHD and 1,25(OH)2D levels as a function of age. Values are means ± SEM. From Delvin et al. [23] with permission.

C. Postnatal Vitamin D Supplementation After the first week of life, in premature infants who received vitamin D, plasma 25OHD remained

CHAPTER 48 Perinatal Vitamin D Actions

constant; 1,25(OH)2D concentration increased up to day 30 with no further change until the end of the first 3 months [35]. The levels of 1,25(OH)2D were more than 2 or 3 times higher than those seen in older children. During this time, there was no significant correlation between vitamin D metabolite concentration and serum calcium and phosphorus levels or calcium and phosphorus intake. The high levels of plasma 1,25(OH)2D beyond the neonatal period may represent a compensatory effect to ensure calcium and phosphorus absorption from the diet at a time where bone demineralization may occur. Osteopenia is seen commonly in premature infants, particularly in those who have received prolonged periods of parenteral feeding or who received a diet insufficient in calcium and phosphate (European formula or human milk). There is now a widespread agreement that deficiency of mineral substrate and not intake and metabolism of vitamin D is the principal etiological factor of osteopenia in low-birth-weight infants [38–40]. Backstrom and colleagues conducted a randomized controlled trial of vitamin D supplementation on bone density and biochemical indices in preterm infants [41]. Thirty-nine infants aged 32 weeks gestation at birth or less received either 200 IU/kilo of vitamin D (maximum 400 IU/day) or 960 IU/day up to age 3 months. Bone mass was determined by dual energy X-ray absorptiometry at the distal end shaft site of the left forearm. At 3 and 6 months corrected age there was no difference between the groups in bone mineral content or areal bone mineral density. The authors concluded that a directly administered vitamin D for preterm infants of 200 IU/kg body weight per day up to a maximum of 400 IU provided biochemical evidence of vitamin D sufficiency, and no functional differences were observed in the study period between this and the higher dose of 960 IU/day.

VI. INFANTS OF DIABETIC MOTHERS Hypocalcemia in infants of diabetic mothers (IDM) has been the subject of a number of studies, and several pathogenic factors have been suggested in this metabolic disorder including hypoparathyroidism, hyperphosphatemia, hypomagnesemia, and defective vitamin D metabolism [42,43]. Hypocalcemia of IDM shares some features with the early neonatal hypocalcemia (ENH) observed in premature babies. Indeed, it appears during the very early hours of life and shows little further change after 24 hr of age. However, it tends to be more severe than ENH and to persist for a longer time. No consistent abnormality in vitamin D metabolism has been observed in IDM, similar to the

807 observations of ENH of premature infants [44]. Neither Salle et al. [44] nor Noguchi et al. [45] detected any major impairment in PTH responsiveness in IDM. Thus, the pathogenesis of hypocalcemia of IDM remains unclear. Possibly, the increased fetal body size of these infants may be responsible for increased calcium needs; the whole body bone mineral content of these babies measured by dual energy X-ray absorptiometry corresponds to that of a newborn baby of the same weight [46]. A prospective controlled study has shown no evidence of hypocalcemia occurring in infants born to women with diet-controlled gestational diabetes [47].

VII. LATE NEONATAL HYPOCALCEMIA Late neonatal hypocalcemia (LNH) is less frequent than ENH and usually brought to attention by the clinical manifestations of tetany and convulsions [48,49]. These are observed from the third or fourth day up to the end of the first month of life. LNH generally affects term infants, but it is also observed in premature infants and IDM in the form of prolonged and severe ENH (see earlier discussion). A now-uncommon cause is the feeding of “doorstep” (unmodified) cow’s milk in which vitamin D content is not controlled. Moreover, its large phosphate content suppresses production of 1,25(OH)2D. Hypocalcemia resulting in convulsions can also occur as a result of osteopetrosis [50], maternal hyperparathyroidism [51], and maternal calcium carbonate consumption during pregnancy [52]; in one case it was reported as a result of a phosphate enema administered to an ex-premature infant at age 6 weeks [53]. Heart failure can occur in some instances and may be misdiagnosed as cardiomyopathy [54]. In late neonatal hypocalcemia, serum PTH levels remain inappropriately low (normal values). Relative hypoparathyroidism thus appears to be the main abnormality but is transient and not due to the absence or hypoplasia of the parathyroid glands as found in the Di George syndrome [55]. Serum 1,25(OH)2D remains normal to moderately elevated, corresponding to what one would expect in response to the low serum PTH levels. This contrasts with the sharp increase in serum 1,25(OH)2D levels observed in infants with ENH [25]. Severe hypocalcemia occurs in perinatal asphyxia when hypoxemia and acidosis persist, despite raised serum PTH levels, suggesting possible end-organ resistance [56] in the context of generally sick or failing cells. Mild LNH requires only a watchful eye, but symptomatic and persistent LNH may require more

808 aggressive intervention. The therapeutic management of LNH based on the pathophysiological findings would be to apply the treatment strategy of hypoparathyroidism. The active form of vitamin D [1,25(OH)2D] may be effective. In most cases the treatment can be discontinued after a few days without relapse of hypocalcemia. In the case of symptomatic hypocalcemia with convulsions, intravenous calcium is recommended (1–2 mmol/kg over 30–60 min, preferably by a central line), within the context of managing the underlying clinical situation.

VIII. RECOMMENDATIONS FOR VITAMIN D INTAKE IN THE PERINATAL PERIOD Recommendations for vitamin D supplementation in pregnancy are currently hard to give since there have been no studies of functional outcomes across a range of populations encompassing both those to be perceived to be at risk, such as mothers with dark skins, reduced sunlight exposure, and vitamin D–deficient diets, and those traditionally perceived to be likely to be vitamin D sufficient, such as white Caucasian mothers with good sunlight exposure and taking foods fortified in some way with vitamin D. The available evidence suggests that 400 units a day during the whole of pregnancy, 1000 units per day over the last 3 months of pregnancy, or up to 100,000 units at monthly intervals during the last 3 months of pregnancy are all well tolerated, safe, and effective methods of delivering vitamin D to pregnant women. The vitamin D requirements of low-birth-weight infants are influenced by the body stores at birth, which in turn are related to the length of gestation and maternal stores. These factors should be taken into consideration when deciding on the policy concerning vitamin D supplementation in each country. The American Academy of Pediatrics recommended that daily intake should be at least 400 IU independently of the vitamin D content of low-birth-weight formula [57]. The European Society of Pediatric Gastroenterology and Nutrition recommended that when low-birth-weight infants are fed human milk they should receive a vitamin D supplement of 1000 IU per day [58]. The work of Backstrom [41] detailed earlier suggests that where maternal vitamin D stores are thought to be normal, 200 IU/kg is likely to be adequate. Formula-fed infants should also be supplemented with vitamin D in order to achieve the same intake as babies receiving breast milk. The American Academy of Pediatrics recently recommended that daily intake should be 200 IU/day, commencing within the first

NICHOLAS J. BISHOP

2 months of life, continuing until adulthood [59]. It should be noted that all the current recommendations are based on biochemical measures of vitamin D sufficiency utilizing a variety of cutoff points for serum 25OHD between 20 and 30 nmol/liter (8–12 ng/ml). Definitive studies relating exposure to functional outcome are more difficult to conduct and are still awaited, in particular for vitamin D supplementation during pregnancy.

IX. SUMMARY Maternal vitamin D intake during the last trimester of pregnancy significantly influences neonatal vitamin D stores and metabolism and may influence growth in infancy. There is no impairment of neonatal vitamin D metabolism consequent on “immaturity,” whatever the gestational age or birth weight of the infant. All mothers should receive an adequate vitamin D intake during the last trimester of pregnancy, and all infants should receive vitamin D in their diet, either as a supplement when the infant is completely breast-fed or as part of a modified cow’s milk–derived formula. There is no place for the use of active vitamin D metabolites in the routine care of healthy infants.

References 1. Coutinho M de L, Dormandy TL, Yudkin S 1968 Maternal malabsorption presenting as rickets. Lancet 1:1048–1052. 2. Park W, Paust H, Kaufmann HJ, Offermann G 1987 Osteomalacia of the mother – rickets of the newborn. Eur J Pediat 146:292–293. 3. Krishnamachari KAVR, Iyengar L 1975 Effect of maternal malnutrition on the bone density of neonates. Am J Clin Nutr 28:482–486. 4. Brooke OG, Brown IRF, Cleeve HJW, Sood A 1980 Observations on the vitamin D state of pregnant Asian women in London. Br J Obstet Gynaecol 88:18–26. 5. Maxwell JD, Ang L, Brooke OG, Brown IR 1981 Vitamin D supplements enhance weight gain and nutritional status in pregnant Asians. Br J Obstet Gynaecol 88:987–991. 6. Brooke OG, Butters F, Wood C 1981 Intrauterine vitamin D nutrition and postnatal growth in Asian infants. Br Med J Clin Res 283:1024. 7. Delvin EE, Salle BL, Glorieux FH, Adeleine P, David LS 1986 Vitamin D supplementation during pregnancy: Effect on neonatal calcium homeostasis. J Pediatr 109:328–334. 8. Marya RK, Rathee S, Lata V, Mudgil S 1981 Effects of vitamin D supplementation in pregnancy. Gynecol Obstet Invest 12:155–161. 9. Stearns G, Jeans PC, Vandecar V 1936 The effect of vitamin D on linear growth in infancy. J Pediatr 9:1–10. 10. Misselwitz J, Hesse V, Markestad T 1990 Nephrocalcinosis, hypercalciuria and elevated serum levels of 1,25-dihydroxyvitamin D in children. Possible link to vitamin D toxicity. Acta Paediatr Scand 79:637–643.

CHAPTER 48 Perinatal Vitamin D Actions

11. Hoogenboezem T, Degenhart J, Munick Keizer-Schrama SM, Bouillon R, Grose WF, Hackeng WH, Visser HK 1989 Vitamin D metabolism in breast-fed infants and their mothers. Pediatr Res 25:623–627. 12. Steichen JJ, Tsang RC, Gratton TL, Hamstra A, DeLuca HF 1980 Vitamin D homeostasis in the perinatal period: 1,25-Dihydroxyvitamin D in maternal, cord, and neonatal blood. N Engl J Med 302:315–319. 13. Delvin EE, Salle BS, Glorieux FH 1991 Vitamin D and calcium homeostasis in pregnancy: Feto-maternal relationships. In: Glorieux FH (ed) Rickets. Vevey/Raven, New York, pp. 91–105. 14. American Academy of Pediatrics Committee on Nutrition 1963 The prophylactic requirements and the toxicity of vitamin D. Pediatrics 31:512. 15. Gartner LM, Greer FR Section on Breastfeeding and Committee on Nutrition 2003 Prevention of rickets and vitamin D deficiency: New guidelines for vitamin D intake. Pediatrics 111: 908–910. 16. HMSO Report Number 41 1991 Dietary and nutrient reference values for food energy and nutrients for the United Kingdom. United Kingdom Department of Health, London. 17. Greer FR, Searcy JE, Levin RS, Steichen JJ, Asch PS, Tsang RC 1981 Bone mineral content and serum 25-hydroxyvitamin D concentrations in breast-fed infants with and without supplemented vitamin D. J Pediatr 98:696–701. 18. Zamora SA, Rizzoli R, Belli DC, Slosman DO, Bonjour JP 1999 Vitamin D supplementation during infancy is associated with higher bone mineral mass in prepubertal girls. J Clin Endocrinol Metab 84:4541–4544. 19. Hypponen E, Laara E, Reunanen A, Jarvelin MR, Virtanen SM 2001 Intake of vitamin D and risk of type 1 diabetes: a birthcohort study. Lancet 358:1500–1503. 20. Namgung R, Tsang RC, Specker BL, Sierra RI, Ho ML 1993 Reduced serum osteocalcin and 1,25-dihydroxyvitamin D concentrations and low bone mineral content in small for gestational age infants: Evidence of decreased bone formation rates. J Pediatr 122:269–275. 21. Rosli A, Fanconi A 1973 Neonatal hypocalcemia. “Early type” in low birth weight newborns. Helv Paediatr Acta 28: 443–457. 22. Tsang RC, Light IJ, Sutherland JM, Kleinman LI 1973 Possible pathogenetic factors in neonatal hypocalcemia of prematurity. The role of gestation, hyperphosphatemia, hypomagnesemia, urinary calcium loss, and parathormone responsiveness. J Pediatr 82:423–429. 23. Glorieux FH, Salle BL, Delvin EE, David LS 1981 Vitamin D metabolism in preterm infants: Serum calcitriol values during the first five days of life. J Pediatr 99:640–643. 24. David L, Salle BL, Chopard JP, Frederich A 1976 Parathyroid function in low birth weight newborns during the first 48 hours of life. In: Stern F-H (ed) Symposium on Intensive Care of the Newborn. Masson, New York, pp. 107–117. 25. Salle BL, Delvin EE, Lapillonne A, Bishop NJ, Glorieux FH. 2000 Perinatal metabolism of vitamin D. Am J Clin Nutr 71(5 Suppl):1317S–1324S. 26. David L, Salle BL, Putet G, Grafmeyer D 1981 Serum immunoreactive calcitonin in low birth weight infants. Description of early changes: Effect of intravenous calcium infusion: relationships with early changes in serum calcium, phosphorus, magnesium, parathyroid hormone, and gastrin levels. Pediatr Res 15:803–808. 27. Delvin EE, Salle BL, Glorieux FH, David LS 1988 Vitamin D metabolism in preterm infants: Effect of a calcium load. Biol Neonate 53:321–326.

809 28. Delvin E, Glorieux F, Salle B, David L, Varenne J 1982 Control of vitamin D metabolism in preterm infants: Fetomaternal relationships. Arch Dis Child 57:754–757. 29. Hillman LS, Haddad JG 1975 Perinatal vitamin D metabolism II. Serial 25-hydroxyvitamin D concentrations in sera of term and preterm infants. J Pediatr 86:928–935. 30. Bouillon R, van Assche FA, van Baelen H, Heyns W, De Moor P 1981 Influence of the vitamin D–binding protein on the serum concentrations of 1,25-dihydroxyvitamin D3: Significance of the free 1,25-dihydroxyvitamin D3 concentration. J Clin Invest 67:589–596. 31. Hirsfeld J, Lunell O 1963 Serum protein synthesis in foetus: Haptoglobins and group-specific components. Nature 196: 1220–1222. 32. Salle BL, Glorieux FH, Delvin EE, David LS, Meunier G 1983 Vitamin D metabolism in preterm infants. Serial serum calcitriol values during the first four days of life. Acta Paediatr Scand 72:203–206. 33. Markestad T, Elzouki A, Legrain M, Ulstein M, Asknes L 1984 Serum concentration of vitamin D metabolites in maternal and umbilical cord blood of Libyan and Norwegian women. Hum Nutr Clin Nutr 38:55–62. 34. Schilling R, Haschke F, Schatten C, Schmid M, Woloszczuk W, Steffan I, Schuster E 1990 High total and free 1,25-dihydroxyvitamin D concentrations in serum of premature infants. Acta Paediatr Scand 79:36–40. 35. Salle BL, Senterre J, Glorieux FH, Delvin EE, Putet G 1987 Vitamin D metabolism in preterm infants. Biol Neonate 52:119–130. 36. Delmas PD, Glorieux FH, Delvin EE, Salle BL, Melki I 1987 Perinatal serum bone Gla-protein and vitamin D metabolites in preterm and full term neonates. J Clin Endocrinol Metab 65:588–591. 37. Pittard WBD, Geddes KM, Hulsey TC, Hollis BW 1992 Osteocalcin, skeletal alkaline phosphatase, and bone mineral content in very low birth weight infants: A longitudinal assessment. Pediat Res 31:181–185. 38. Tsang RC, Demarini S 1995 Rickets and calcium and phosphorus requirements in very low birth weight infants. Monatsschr Kinderkeilkd 43:S125–S129. 39. Shaw NJ, Bishop NJ 1995 Mineral accretion in growing bones— A framework for the future? Arch Dis Child 72:177–179. 40. Lapillonne A, Glorieux FH, Salle BL, Braillon PM, Chambon M, Rigo J, Putet G, Senterre J 1994 Mineral balance and whole body bone mineral content in very low-birth-weight infants. Acta Paediatr 405(Suppl):117–122. 41. Backstrom MC, Maki R, Kuusela AL, Sievanen H, Koivisto AM, Ikonen RS, Kouri T, Maki M 1999 Randomised controlled trial of vitamin D supplementation on bone density and biochemical indices in preterm infants.Arch Dis Child Fetal Neonatal Ed. 80:F161–F166. 42. Tsang RC, Chen IW, Friedman FA, Gigger M, Steichen J, Koffler H, Fenton L, Brown D, Pramanik A, Keenan W, Strub R, Joyce T 1975 Parathyroid function in infants of diabetic mothers. J Pediatr 86:399–404. 43. Bergman L, Kjellmer I, Seltam U 1974 Calcitonin and parathyroid hormone. Relation to early neonatal hypocalcemia in infants of diabetic mothers. Biol Neonate 24:151–160. 44. Salle BL, David L, Glorieux FH, Delvin EE, Louis JJ, Troncy G 1982 Hypocalcemia in infants of diabetic mothers. Studies on circulating calciotropic hormone concentrations. Acta Paediatr Scand 71:573–577. 45. Noguchi A, Eren M, Tsang R 1980 Parathyroid hormone in hypocalcemia and normocalcemic infants of diabetic mothers. J Pediatr 97:112–114.

810 46. Lapillonne A, Guerin S, Braillon P, Claris O, Delmas PD, Salle BL 1997 Diabetes during pregnancy does not alter whole body bone mineral content in infants. J Clin Endocrinol Metab 82:3993–3997. 47. Sarkar S, Watman J, Seigel WM, Schaeffer HA 2003 A prospective controlled study of neonatal morbidities in infants born at 36 weeks or more gestation to women with dietcontrolled gestational diabetes (GDM-class A1). J Perinatol 23:223–228. 48. Balsan S, Alizon M 1968 L’hypoparathyroidie transitoire idiopathique du nourrison. Arch Fr Pediatr 25:1151–1170. 49. Fanconi A, Prader A 1967 Transient congenital idiopathic hypoparathyroidism. Helv Paediatr Acta 22:342–359. 50. Srinivasan M, Abinun M, Cant AJ, Tan K, Oakhill A, Steward CG 2000 Malignant infantile osteopetrosis presenting with neonatal hypocalcaemia. Arch Dis Child Fetal Neonatal Ed 83:F21–F23. 51. Ip P 2003 Neonatal convulsion revealing maternal hyperparathyroidism: An unusual case of late neonatal hypoparathyroidism. Arch Gynecol Obstet. 268:227–229. 52. Robertson WC Jr 2002 Calcium carbonate consumption during pregnancy: An unusual cause of neonatal hypocalcemia. J Child Neurol 17:853–855.

NICHOLAS J. BISHOP

53. Walton DM, Thomas DC, Aly HZ, Short BL 2000 Morbid hypocalcemia associated with phosphate enema in a six-week-old infant. Pediatrics 106:E37. 54. Gulati S, Bajpai A, Juneja R, Kabra M, Bagga A, Kalra V 2001 Hypocalcemic heart failure masquerading as dilated cardiomyopathy. Indian J Pediatr 68:287–90. 55. Perez E, Sullivan KE 2002 Chromosome 22q11.2 deletion syndrome (DiGeorge and velocardiofacial syndromes). Curr Opin Pediatr 14:678–683. 56. Schedewie HK, Odell WD, Fisher DA, Krutzik SR, Dodge M, Cousins L, Fiser WP 1979 Parathormone and perinatal calcium homeostasis. Pediatr Res 13:1–6. 57. American Academy of Pediatrics, Committee on Nutrition 1977 Nutritional needs of low birth weight infants. Pediatrics 60:519–530. 58. European Society of Paediatric Gastroenterology and Nutrition, Committee on Nutrition of the Preterm Infant 1987 Nutrition and feeding of preterm infants. Acta Paediatr Scand 336(Suppl):6. 59. Gartner LM, Greer FR, Section on Breastfeeding and Committee on Nutrition. American Academy of Pediatrics 2003 Prevention of rickets and vitamin D deficiency: new guidelines for vitamin D intake. Pediatrics 111:908–910.

CHAPTER 49

Vitamin D Deficiency and Calcium Absorption during Infancy and Childhood STEVEN A. ABRAMS I. II. III. IV.

USDA/ARS Children’s Nutrition Research Center, Houston, Texas

Introduction Premature Infants Full-Term Infants Toddlers and Prepubertal Children

I. INTRODUCTION Although the importance of providing adequate calcium and vitamin D during childhood and adolescent growth is well known, there remain important gaps in our understanding regarding the process of calcium absorption and utilization in childhood. Certain time periods in development appear to be critical ones in which calcium or vitamin D deficiency can pose very high risks. The first such time periods is in utero, or as commonly reflected in current medical care, the initial months of life of prematurely delivered infants (also see Chapter 48). These infants, especially those < 1.5 kg at birth, are at very high risk for the development of clinical rickets or other manifestations of bone loss. This bone mineral deficiency is primarily related to difficulties in providing, via parenteral or enteral sources, adequate calcium and phosphorus for the extremely rapid bone growth that normally occurs via placental transfer to the fetus [1]. In healthy full-term infants, human milk is recommended as the sole nutritional source for the first 6 months of life [2]. The vitamin D content of human milk is very low and the photoconversion of vitamin D precursors to vitamin D is necessary to obtain adequate vitamin D levels. Since this may not be possible in many infants, it is recommended that all infants receive supplemental vitamin D, either via infant drops or, for infant formula-fed infants, as provided in the formula [3]. Although vitamin D–deficient rickets is a welldescribed problem for older infants and toddlers, extremely little information is available regarding vitamin D requirements and the relationship between vitamin D status and calcium absorption in this age group. The optimal vitamin D intake or concentration to maximize calcium absorption, as well as the lowest VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Calcium Absorption in Adolescents VI. Fortification of Foods with Calcium and Vitamin D for Children VII. Summary and Conclusions References

intake of calcium and vitamin D needed to prevent severe bone loss or rickets, is not well described. Longterm consequences of variations in calcium absorption and bone mineralization in early life are not known. Despite the evidence that most infant formulas provide at least as much, if not more, absorbable calcium than human milk, some data suggest that, in later childhood, the bone mass of infants who are breastfed may be the same or greater than that of formula-fed infants [4,5]. An important issue then is to determine at what age it becomes critical to maximize mineral intake and absorption to lead to an optimal peak bone mass. Some controlled trials have indicated that supplementation of calcium before puberty may be important, whereas others have found benefit in pubertal children [6]. Interpreting such studies is difficult, however, as prestudy calcium intakes are often poorly assessed or controlled and supplemented intakes may exceed the absorptive threshold, leading to little benefit. Because it is the time of most rapid mineralization of the skeleton, efforts to improve peak bone mass during childhood have focused on increasing absorbable calcium intake in adolescents. Efforts in this regard include advocacy campaigns to increase dairy products and considerable efforts to provide fortified foods and beverages to adolescents. Recent research efforts have looked at the effects of race, diet, gender, and other food components on maximizing calcium absorption [6,7]. As part of global efforts to develop nutritional planning, consideration has been given to the role of calcium and vitamin D supplements in food products designed for children in developing countries [8]. Although many of these are countries in which children are exposed to adequate sunshine, they may also have low calcium and vitamin D status due to social or dietary causes. Extending research in calcium absorption to these Copyright © 2005, Elsevier, Inc. All rights reserved.

812

STEVEN A. ABRAMS

populations is important not only to prevent bone loss, but also to understand and prevent an increase in osteoporosis in developing countries in the future.

II. PREMATURE INFANTS Human milk is recognized as the ideal food for virtually all infants [2]. The substantial health and developmental benefits of human milk feeding have been well documented even in the smallest of premature infants. However, for many decades, it has been known that one critical limitation is that the minerals, especially the calcium and phosphorus, in human milk do not meet the needs of rapidly growing premature infants. This and other factors place premature infants at high risk for nutritional rickets (Table I). This can readily be seen in that the fetus accretes 100–120 mg/kg/d of calcium (about 50–60 mg/kg/d of phosphorus) during the third trimester. Since human milk contains about 25 mg/dl of calcium (13 mg/dl of phosphorus) and is usually fed at 150–200 mg/kg/d, even if all of the calcium and phosphorus were to be completely absorbed and retained, it cannot provide more than half of the minerals needed by a 1.0 kg infant to meet the in utero accretion rate [1,9–11]. In reality, calcium absorption in unfortified human milkfed infants is generally about 60%, leading to a net absorption of about 20–30 mg/kg/d or less than 1/3 of the in utero rate [12–14]. The recent addition of various forms of mineral salts and/or mineral fortifiers to human milk and the use of specialized preterm infant formulas with very high calcium content levels have been shown to enhance the amount of calcium and other minerals retained from the diet, to increase bone mineral content of the infants, and to decrease the incidence of osteopenia and frank rickets in preterm infants [1,15,16].

TABLE I High-Risk Criteria for Osteopenia in Premature Infants • • • •

Born at < 27 weeks’ gestation Birth weight of <1000 g Long-term parenteral nutrition Severe bronchopulmonary dysplasia with use of diuretics and fluid restriction • Long-term steroid use • History of necrotizing enterocolitis • High serum alkaline phosphatase activity (>900 IU/liter) and low serum phosphorus (< 5.2 mg/dl)

The bioavailability of the calcium in these fortifiers may be a key aspect of their adequacy. Using a commercially available human milk fortifier, Schanler and Abrams [1] reported that net calcium retention was 104 ± 36 mg/kg/d in premature infants, a value approximating the in utero accretion rate during the third trimester. These retention values are well above those achieved using earlier human milk fortifiers [16] and those more recently reported using a human milk fortifier not used in the United States [17]. That fortifier led to minimally positive calcium retention and therefore little if any benefit to the baby. It is not clear why the bioavailability of the calcium in the fortified feeding was low, but it is likely related either to the form of the calcium or problems with the solubility of the calcium in the fortified milk. Of interest is that the calcium absorption percentages from both fortified human milk and from specialized preterm formula average 50–65% in many studies [13,14]. This percentage absorption is similar to that seen for unfortified human milk in both premature infants and full-term infants. However, in the case of preterm formulas or fortified milk, the calcium intake is about 220 mg/kg/d. This intake vastly exceeds the body-weight adjusted calcium intake from human milk by full-term infants (10 mg/kg/d for a 10-kg infant) [5,12]. This constancy of absorptive fraction in premature infants suggests that much of the calcium absorption by premature infants and newborn full-term infants is not vitamin D–dependent. In a review of more than 100 balance studies, Bronner et al. [13] showed that the calcium absorption fraction varied little with calcium intake in premature infants and thereby suggest that most if not virtually all calcium absorption is vitamin D–independent. Unfortunately, there are no studies of calcium absorption in premature infants over a broad range of calcium intakes to directly determine the effects of calcium intake and vitamin D status on calcium absorption. Such studies would be virtually impossible to do on a practical and ethical basis. Multiple studies have demonstrated that vitamin D intakes of 400 IU/day (or 200 IU/kg up to 400 IU/d) in premature infants leads to adequate vitamin D levels [18–20]. One study demonstrated adequate 25-hydroxyvitamin D concentrations and clinical outcomes with oral vitamin D intakes as low as 160 IU/d [21]. In addition, studies have generally failed to show any clinical benefit of increasing vitamin D intake above 400 IU/day in preterm infants. One study comparing 500 IU to 1000 IU found no short-or long-term benefit (up to age 11) of higher amounts [22]. “There are no data to support the belief that preterm infants

CHAPTER 49 Calcium Absorption in Children

need a disproportionately high vitamin D dose in relation to their weight” [22]. The effects of other formula components on mineral absorption have also been considered. A study using a triple lumen perfusion technique demonstrated that calcium absorption was greater using a solution that included a glucose polymer than one with lactose [23]. As glucose polymers are widely used in preterm formulas, this effect may be clinically important. Altering the fat blend of infant formula to more closely resemble that of human milk may also enhance mineral absorption in premature infants [24,25].

III. FULL-TERM INFANTS A. Calcium Infancy is a time of rapid body growth as infants may triple their birth weight in the first year of life. This rapid body growth is accompanied by comparably rapid bone growth [26]. Remarkably, however, the human milk–fed baby readily mineralizes with all of the calcium coming from mother’s milk in the first 6 months and most of it from human milk during the second 6 months of life [27]. This is possible because of the withdrawal of bone from the maternal skeleton to meet the infant’s needs and the continued high rates of calcium absorption from human milk. Dietary recommendations for calcium intake in infancy are based on the knowledge that calciumdeficiency rickets does not occur in healthy, vitamin D– sufficient, breast-fed infants. Therefore, the calcium intake of the exclusively breast-fed infant, averaging 210 mg/day, was established by the National Academy of Sciences [28], and set as the Adequate Intake for calcium in the first 6 months of life (Table II). In the second 6 months of life, most calcium continues to

TABLE II Calcium Recommendations for Infants and Children in the United States [28] Calcium (mg/d) Infants 0–6 mo Breast-fed Formula-fed Infants 7–12 mo Breast-fed Formula-fed Children 1–3 yrs Children 4–8 yrs

210 315 270 335 500 800

813 come from human milk, but there is some from solid foods. The Adequate Intake for 7- to 12-month-old infants was therefore established as 270 mg/day, which is the sum of the usual intake of calcium from breast milk and solid foods. Infant formulas have a calcium concentration well above that usually found in human milk [28,29]. One justification for providing more calcium in infant formulas than human milk is the belief that calcium is more poorly absorbed from infant formula than from human milk. This perspective is based on studies in which calcium concentrations were much greater in formula than in human milk [5]. These high concentrations may lead to lower fractional calcium absorption in the infants. This inverse relationship was demonstrated by Devizia et al. [30] who reported decreasing fractional absorption in a very small group of infants as formula concentration of calcium increased. Several studies have suggested that in many cases, the fractional absorption of calcium from infant formulas [31–35] is similar to the value for human milk (Table III). Studies of whole-body bone mineral content using DXA support these findings. Calculations both from Fomon and Nelson [5] and from earlier data using metacarpal morphometry [36] suggest a mean calcium accretion rate of approximately 80 mg/day during the first year of life [28]. More recent bone mineral content studies in breast-fed infants have shown slightly higher rates of bone mineralization of approximately 100 mg/d during the first year of life [27]. Although there are few direct comparisons with formula-fed infants, recent studies suggest values of about 150 mg/day in the first 6 months of life [26,37] for formula-fed infants. This rate appears not to change substantially in the second 6 months of life after more solid foods are introduced [37,38]. More data are needed though, especially in the second 6 months of life, to evaluate the relationship between rates of bone mineralization and diet and vitamin D status. With these high absorption fractions, it appears that it is possible to markedly increase total calcium absorption and net calcium retention in infants above that of primarily human milk–fed infants. However, caution should be raised about the benefits of targeting high levels of calcium absorption in infancy, either with high-mineral-containing formulas, or through the use of highly calcium-fortified solid foods for older infants. There are no data to support any long-term benefit by exceeding the calcium absorption or bone mass of breast-fed infants. A study by Jones et al. [4] found a greater bone mineral density at the spine and whole body in 8-year-old children who had been breast-fed

814

STEVEN A. ABRAMS

TABLE III

Calcium Absorption Fraction in Human Milk and Formula-Fed Infants Based on a Reference Intake of 780 ml/day Ca concentration (mg/dl)

Human milk [32] Standard Formula [33] Partially hydrolyzed [31] Human milk [5] Cow milk-based formula [35]

Absorption (%)

Net Ca absorption (mg/d)

61 ± 23 58 ± 13 66 ± 12 58 ± 17 57 ± 15

97 205 220 113 255

25 50 46 25 57

Notes: An estimate was used for endogenous fecal calcium excretion for isotope studies and the net absorption for isotope-based studies was adjusted for this value.

compared to those bottle-fed as infants. This effect was only present for infants who had been breast-fed for at least 3 months. Studies in preterm infants have also failed to show any long-term benefit to greater mineral intake during early infancy [39,40]. This view is supported by animal data in rabbits that do not show any benefit to increasing bone mineral content in early life [41] and is consistent with similar classic data from Gershoff et al. [42].

B. Vitamin D Nutritional rickets in children is described in detail in Chapter 65. In this section, we will consider some specific issues related to the use of vitamin D supplements for infants, especially those in the United States. Human milk is a relatively small source of dietary vitamin D for most infants, usually providing an average of 10–20 IU/day [43,44]. Increased vitamin D in human milk is related to increasing maternal vitamin D status, but the level of supplementation required may be relatively high [3,44]. Infant formulas and cow’s milk are fortified by statute in the United States, although cow’s milk is not fortified with vitamin D in many other countries. Therefore, those at greatest risk of vitamin D deficiency in the United States are infants who are breastfed without adequate sunshine exposure, or those who are weaned to diets containing little vitamin D. In addition to an increased frequency of breast-feeding older infants and toddlers, social conditions and the widespread vigorous use of sun-block have made it more common for infants to receive little sunlight exposure. Because of the resurgence of rickets in the United States [45], it has become clear that policies of selective vitamin D supplementation of high-risk infants are not adequately protective of the entire population at risk.

Therefore, the American Academy of Pediatrics [3] has recommended universal vitamin D supplementation for all infants. For breast-fed infants, the vitamin D is provided as part of a supplement drop; for formula-fed infants it is contained in the formula [Table IV]. Based on the recommendations of the Food and Nutrition Board in the Dietary Reference Intakes as revised in 1997, the recommendation is to provide 200 IU/day to infants beginning by 2 months of age [3,28], Of note is that as vitamin D–deficient rickets is not usually seen in early infancy, it is not crucial to begin supplementation at birth, but as the policy allows, to delay beginning the vitamins for up to 2 months. The choice of 200 IU/day is based on minimal data primarily obtained in a very few infants regarding the amount of vitamin D required to maintain adequate levels [3,28]. Many more data are needed on this topic, however, and given the high level of safety of vitamin D in this dose range, it is possible that higher doses may be needed to optimize vitamin D status [46]. It is very likely, however, that the dose of 200 IU/day is adequate for virtually all infants to prevent overt vitamin D deficiency.

TABLE IV Common Supplemental Vitamin D Sources for Infants and Toddlers Multivitamin dropsa Vitamin D onlyb D-vi-sol (Canada)c Infant formulas Whole milk/juices aUsually

400 IU/ml 8000 IU/ml 400 IU/ml 400 IU/liter 100 IU/240 ml

combined with vitamins A and C. vitamin D available in the United States. Not recommended for routine use. cSingle-source vitamin D (Mead-Johnson, Inc., Evansville, IN) available (June 2003) in Canada but not the United States. bConcentrated

815

CHAPTER 49 Calcium Absorption in Children

There is a substantial gap in data regarding calcium absorption between infants and pubertal children. The calcium adequate intake (AI) of 500 mg/day for children from 12 to 48 months of age was developed based on extrapolation from desirable calcium retention for 4- to 8-year-olds [28]. We have shown that increasing calcium intake via fortified foods in small children leads to increased total calcium absorption [47]. These and similar data demonstrating increased calcium absorption or retention at intakes above 500 mg/day in toddlers imply that the AI for 1- to 3-year-old children may be below the optimal intake level [47–49]. Higher levels may significantly increase calcium absorption and retention without posing any risk to long-term bone development. Few data are available regarding calcium requirements in children prior to puberty. Most of this data is based on balance studies conducted over 50 years ago on diets that are very different from those currently in place [48]. An increase in net calcium absorption when the intake of calcium in 3- to 5-year-old children was increased from 500 to 1200 mg/day has been reported [49]. The benefit was relatively modest, however, and intermediate intake levels, as might more readily be achieved in preschool children, were not evaluated in this study. We compared measures of calcium metabolism in 7- and 8-year-old Mexican-American and non-Hispanic Caucasian girls living in southeastern Texas [50]. Fractional calcium absorption and total body bone mineral content in the girls was not significantly correlated to either PTH or vitamin D levels. We found lower serum 25-hydroxyvitamin D concentrations and higher PTH levels in the Mexican-American girls, but these were not significantly inversely correlated to each other. Seasonal variability was seen for 25-hydroxyvitamin D concentrations in girls of both ethnic groups, but values in all of the girls were >12 ng/ml. Therefore it appears that differences in 25-hydroxyvitamin D and PTH concentrations between MexicanAmerican and Caucasian girls do not have a large effect on calcium absorption in vitamin D–sufficient prepubertal children. The potential benefit to ultimate peak bone mass of increasing calcium intake has been studied in groups of prepubertal children. In one controlled calcium supplementation trial, an increase in bone mass was found when calcium supplements were given to children as young as 6 years of age [51]. However, relatively few children this young were studied, and the duration of effect of this supplementation and its impact on peak

bone mass are uncertain. Further studies are needed to evaluate different levels of calcium intake in this age group. It is also not clear that benefits in bone mass achieved prepuberty will persist through puberty or once high intakes of calcium are stopped.

V. CALCIUM ABSORPTION IN ADOLESCENTS A. Effects of Puberty, Gender, Ethnicity, and Other Genetic Factors Population- and age-related variability in calcium absorption is largely related to factors that are not readily controlled such as pubertal status, ethnicity, and other genetic factors [52–57]. Identification of these factors and understanding their relative contribution to mineralization is an important area of ongoing research. Calcium absorption efficiency increases substantially during puberty and rapidly decreases postpuberty (Fig. 1) [58]. In a longitudinal multiethnic study, we found a significant increase in the utilization of calcium associated during early puberty (Fig. 2) compared with the year prior to the physical changes of puberty. This change was evidenced by increased calcium absorption, and kinetically determined rates of bone calcium deposition [60]. We found that increases in calcium absorption and deposition were associated with maturation of the hypothalamic–pituitary axis as measured by a rise in the gonadotropin-releasing hormone-simulated luteinizing hormone (LH) level. Martin and co-workers [52] have reported the changes

Age-related changes in calcium absorption in girls 600 Calcium absorption, mg/d

IV. TODDLERS AND PREPUBERTAL CHILDREN

500 400 300 200 100 0 5–6

7–8

9 –10 11–12 Age groups, y

13 –14 15 –16

FIGURE 1 The relationship between age and calcium absorption in girls. Maximum rates of absorption are found from 9 to 14 years of age consistent with maximum rates of bone mass accumulation [58].

816

STEVEN A. ABRAMS

B. Effects of Inadequate Calcium and Vitamin D Intake

250 Calcium accretion (mg/d)

Peak accretion rate 200 150 100 50 0

Prepubertal

Early pubertal

FIGURE 2

Increased rates of calcium accretion are seen in girls during early puberty. “Prepubertal” refers to the average rate of calcium accumulation during a 1-year period before the onset of physical changes of puberty and “early puberty” refers to the subsequent year during which pubertal changes were noted to have occurred [59].

in total body bone mineral content in girls using DXA. They found a maximum increment of 260 mg/day at age 12–13 years. These more recent values for peak rates of calcium gain are very similar to the estimates derived from rates of weight change by Leitch and Aitken in the 1950s [61]. There are marked differences in bone mass and the incidence of osteoporosis between African-Americans and Caucasians (see Chapter 47). Several groups have found lower urinary calcium in African-American girls compared to Caucasians [54,56,62]. In addition, it appears that, at similar calcium intakes, AfricanAmerican girls absorb more calcium than do Caucasians [9]. There are no data available for this comparison among males. Numerous recent studies have focused on identifying the mechanisms of the relationship between genetics and osteoporosis by evaluating specific genetic markers and their relationship to bone mass (see Chapter 68). One of the unknown aspects of this important interaction is the relationship between these genetic differences and calcium absorption and excretion. Because of the importance of puberty in bone mass, it is likely that an effect can be seen during pubertal development. We reported [57] a significant relationship between polymorphisms of the vitamin D receptor (VDR) Fok 1 genotype and calcium absorption. Children with the FF genotype absorbed on average 115 mg/day more calcium than those with the ff genotype. Furthermore, we found the FF genotype to be associated with greater bone mineral density in the study subjects.

Extremely low calcium intakes at all ages have been associated with fractures and rickets both in the United States and in other countries, especially Nigeria and South Africa. Much more common in Western countries are intakes of calcium by adolescents far below the recommendation of 1300 mg/day (AI) but high enough to prevent overt clinical deficiency. For example, in girls 14–18 years old, the 10th percentile of usual intakes is 413 mg/day and the 25th percentile is 541 mg/day. This means that nearly 25% of adolescent girls have a daily calcium intake of 40% or less of the recommended amount [28]. Clearly, there is a considerable ability to adapt to these low intakes by increasing fractional absorption. This has been shown in earlier studies by Matkovic and Heaney [48], and in preliminary data from our group [63]. However, net calcium retention remains far below that achieved on more appropriate intakes. There are very few data on the health consequences or effects on calcium physiology of various levels of vitamin D intake or vitamin D status in children. One recent study from Finland indicated that there was markedly improved bone mineralization, especially at the lumbar spine, in girls who had high vitamin D intakes. Those with the highest vitamin D levels showed the most increase longitudinally in bone mineral density at the lumbar spine [64]. This is consistent with findings that vitamin D may be limiting in some populations of adolescents. For example, adolescent boys in France had decreased vitamin D status in winter associated with increased PTH levels [65]. Rickets has been reported in adolescents in Middle Eastern countries who have little sunlight exposure for cultural reasons [66]. Goulding [67] found lower bone mass in girls with distal forearm fractures than in age-matched girls without fractures. A lower calcium intake was reported in the 11- to 15-year-old girls with fractures than in the controls. Wyshak and Frisch [68] similarly reported that high calcium intakes decreased the risk of bone fractures in adolescents. Epidemiological evidence regarding the consequences of low calcium consumption in childhood and adolescence and ultimate adult bone mass has generally shown the expected benefits [69]. However, such data are not conclusive nor do they demonstrate a critical intake level of calcium at which long-term benefit or harm can be derived. Few long-term follow-up studies have been done to evaluate the effects of calcium intake in childhood and adolescence on adult bone mass.

817

CHAPTER 49 Calcium Absorption in Children

C. Effects of Other Factors Including Soda Consumption Wyshak also reported a positive relationship between fractures and carbonated beverage intake [70]. However, this link does not prove a direct cause-and-effect relationship. Some have suggested that the high phosphorus content of the sodas leads to increased calcium losses and lower bone mass. However, there are no prospective data relating bone mass to carbonated beverage intake and no prospective studies of the relationship. Furthermore, it is likely that very high phosphorus intakes are needed to adversely affect calcium metabolism [71]. The amount of phosphorus in sodas is small relative to the total daily intake of most adolescents and therefore it is likely that reasonable soda or carbonated beverage consumption is not a major cause of bone mass loss related to their phosphorus content. However, it remains of concern that excessive intake of some beverage products places adolescents at risk for calcium deficiency probably primarily due to lowered intake of healthier calcium-containing beverages. Further data on the relationship between dietary factors and fractures and bone loss are needed, and at present this relationship remains highly controversial [72,73].

D. Developing Countries In Mexican toddlers, Murphy et al. [74] found relatively high levels of calcium intake (mean 735 ± 199 mg/day). This was far greater than the mean calcium intake of less than 220 mg/day in Egypt and Kenya. The higher intake in Mexico likely is due both to dairy products and lime-treated tortillas in parts of Latin America [74,75]. The bioavailability of the calcium in this diet, especially from the treated tortillas, may be poor, however [76]. These calcium intakes may not occur among poorer populations of Latin America. Wyatt and Tejas [77] have reported large economic differences in calcium intakes in 4- to 6-year-old children in Southern Mexico. In children in the poorest areas, mean calcium intake was only 272 mg/day, primarily coming from corn tortillas. This increased in medium-income families to more acceptable intake levels of 625 mg/day due to the increased availability of dairy products in families of higher socioeconomic status. In Mexico City a mean calcium intake of 516 mg/day was reported for children 1 to 5 years of age [78]. In this group, 25% of the children had calcium intakes of less than 361 mg/day that was associated with higher blood

lead levels (see below). This is in contrast to a 25th percentile value of 599 mg/day for 1- to 3-year-olds (and 649 mg/day for 4- to 8-year-olds) in the United States [28], suggesting a much greater prevalence of very low calcium intakes in Mexico compared to the United States.

VI. FORTIFICATION OF FOODS WITH CALCIUM AND VITAMIN D FOR CHILDREN A. Rationale for Fortification The potential for inadequate intake of calcium and vitamin D among children and adolescents throughout the world is considerable. Among the strategies for resolving this problem are (1) increasing the intake of foods that naturally contain calcium and vitamin D; (2) increasing the absorption of calcium from foods by additional food components or genetic modification (e.g., decreasing oxalate content) which might enhance calcium absorption; (3) universal pill or liquid vitamin and mineral supplementation; or (4) increased mandatory or optional fortification of food sources [79]. Each of these can and will be important in improving bone health in children and adults. Consideration of the use of prebiotics to enhance calcium absorption is provided hereafter. Food fortification, however, is likely to be a key component of any strategy. In considering food fortification with calcium or vitamin D, one must evaluate both the potential benefits of the fortification strategy and the risks related to excess intake from overly zealous fortification. To determine benefit, one must ensure that fortification is necessary based on a low intake by at least one population subgroup, that the nutrient of importance has an important public health need, and that the nutrients being fortified are bioavailable. All of these criteria are easily met for both calcium and vitamin D fortification of food and beverage sources for children and adolescents. Good bioavailability of calciumfortified orange juice has been shown [80]. Limited data indicate relatively good bioavailability of calcium added to bread and grain products [81] and fortified cereal [47]. From a safety perspective, it is worth noting that the upper limit for calcium intake in children over 12 months of age is 2500 mg/day [28]. This intake is well above the 95th percentile of intakes for all population groups and above the 99th percentile of intake for all age and gender groups except adolescent males. Therefore, reasonable calcium fortification of grain

818 products and juices is unlikely to pose a problem for children and adolescents. Traditionally in the United States, relatively few foods and beverages have been vitamin D fortified except for milk. This situation is changing and it is likely that increasing number of commercial products will have added vitamin D. Recently, juice products including orange juice have been allowed to add vitamin D. The safety margin for this is also likely to be very favorable, as the upper limit of 2000 IU/day is far above the intakes of almost anyone who is not taking high-dose medicinal supplements that are not recommended for children [82]. The combination of substantial public need and a high safety margin make calcium and vitamin D appropriate for food and beverage fortification efforts. However, continuing monitoring and education is important to ensure that excess fortification does not occur and that high-dose supplements are not used at the same time as fortificants. Consideration should also be given to the effects of calcium and vitamin D fortification on the status of other minerals, including magnesium and zinc.

B. Enhancers of Calcium Absorption An alternative dietary strategy to enhancing net absorbed calcium is to identify dietary strategies that enhance the calcium bioavailability of the whole diet and which have other health benefits. For example, functional foods including prebiotics such as nondigestible oligosaccharides (NDO) may be of benefit [83,84]. We have completed a study of the effect of 8 g/day of Synergy1 (a NDO composed of a mixture of long and short chain-length molecules) on calcium absorption in young girls (aged 11 to 13.9 years) [85]. Subjects received in random order 8 g/day of NDO (either Synergy1 or oligofructose) and placebo (sucrose), added to a diet providing approximately 1200–1300 mg/day calcium. Calcium absorption was measured after 21 days of adaptation to the NDO or placebo using a stable isotope method. We found a significant increase in calcium absorption while consuming NDOs. Calcium absorption was significantly higher when subjects consumed 8 g/day Synergy1 than when consuming placebo, but no significant benefit was seen from 8 g/day of oligofructose. The increase in calcium absorption (32.3% to 38.2%) represents a relative increase of more than 18%. A change of this magnitude is clinically highly significant. Of importance, this effect was seen at a relatively high calcium intake, whereas simply increasing the amount of calcium in

STEVEN A. ABRAMS

the diet would be unlikely to significantly increase calcium absorption [85,86]. The mechanism by which oligosaccharides might increase calcium absorption is not known. NDOs resist digestion in the human gut, but are fermented to volatile fatty acids in the colon [84]. These fatty acids may have a local effect in the colon by reducing the pH and increasing solubility of mineral in the aqueous phase of the colonic contents permitting higher absorption of minerals in the colon, a site where little calcium absorption normally occurs. Alternatively, the NDO, volatile fatty acids, or some other mediator may have a trophic effect on the gut [46], improving overall “gut health.” Such an effect could increase calcium absorption throughout the length of the gastrointestinal tract. Finally, NDOs may alter the composition of the colonic bacterial flora [83,84,87–89], and this might lead, directly or indirectly, to a change in mineral absorption or overall “gut health.”

VII. SUMMARY AND CONCLUSIONS Calcium and vitamin D are critical nutrients throughout infancy and childhood. Clinically apparent demineralization related to deficiencies can present as rickets, most commonly in premature infants or toddlers. However, calcium intake is especially critical during adolescence. Inadequate intake and absorption of calcium during puberty can lead to a lifelong risk of osteoporosis due to failure to achieve peak bone mass. Currently efforts to enhance the calcium status of children have focused on improving intakes via food fortification strategies and increasing absorption using specific enhancers of absorption. Inadequate calcium and vitamin D status is a global problem with substantial health consequences for children throughout the world.

Acknowledgment This work is a publication of the U.S. Department of Agriculture (USDA)/Agricultural Research Service (ARS) Children’s Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine and Texas Children’s Hospital, Houston, TX. This project has been funded in part with federal funds from the USDA/ARS under Cooperative Agreement number 58-6250-6-001. Contents of this publication do not necessarily reflect the views or policies of the USDA, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.

819

CHAPTER 49 Calcium Absorption in Children

References 1. Schanler RJ, Abrams SA 1995 Postnatal attainment of intrauterine macromineral accretion rates in low birth weight infants fed fortified human milk. J Pediatr 126:441–447. 2. Work Group on Breastfeeding, American Academy of Pediatrics 1997 Breastfeeding and the use of human milk. Pediatrics 100:1035–1039. 3. Gartner LM, Greer FR 2003 Prevention of rickets and vitamin D deficiency: New guidelines for vitamin D intake. Pediatrics 111:908–910. 4. Jones G, Riley M, Dwyer T 2000 Breastfeeding in early life and bone mass in prepubertal children: A longitudinal study. Osteoporos Int 11:146–152. 5. Fomon SJ, Nelson SE 1993 Calcium, phosphorus, magnesium, and sulfur. In: Nutrition of Normal Infants. Mosby-Year Book, St Louis, pp. 192–218. 6. Heaney RP, Abrams SA, Dawson-Hughes B, Looker A, Marcus R, Matkovic V, Weaver C 2000 Peak bone mass. Osteoporos Int 11:985–1009. 7. Baker SS, Cochran WJ, Flores CA, Georgieff MK, Jacobson MS, Jaksic T, Krebs NF 1999 American Academy of Pediatrics. Committee on Nutrition. Calcium requirements of infants, children, and adolescents. Pediatrics 104:1152–1157. 8. Abrams SA, Atkinson S 2003 Calcium, magnesium, phosphorus, and vitamin D fortification of complementary foods. J Nutr 133:29945–29995. 9. Schanler RJ 2001 The use of human milk for premature infants. Pediatr Clin North Am 48:207–219. 10. Greer FR, McCormick A 1988 Improved bone mineralization and growth in premature infants fed fortified own mother’s milk. J Pediatr 112:961–969. 11. Lyon AJ, McIntosh N 1984 Calcium and phosphorus balance in extremely low birthweight infants in the first six weeks of life. Arch Dis Child 59:1145–1150. 12. Schanler RJ, Oh W 1985 Nitrogen and mineral balance in preterm infants fed human milks or formula. J Pediatr Gastroenterol Nutr 4:214–219. 13. Bronner F, Salle BL, Putet G, Rigo J, Senterre J 1992 Net calcium absorption in premature infants: results of 103 metabolic balance studies. Am J Clin Nutr 56:1037–1044. 14. Abrams SA, Esteban NV, Vieira NE, Yergey AL 1991 Dual tracer stable isotopic assessment of calcium absorption and endogenous fecal excretion in low birth weight infants. Pediatr Res 29:615–618. 15. Schanler RJ 1998 The role of human milk fortification for premature infants. Clin Perinatol 25:645–657. 16. Schanler RJ, Abrams SA, Garza C 1988 Mineral balance studies in very low birth weight infants fed human milk. J Pediatr 113: 230–238. 17. Loui A, Raab A, Obladen M, Bratter P 2002 Calcium, phosphorus and magnesium balance: FM 85 fortification of human milk does not meet mineral needs of extremely low birthweight infants. Eur J Clin Nutr 56:228–235. 18. Cooke R, Hollis B, Conner C, Watson D, Werkman S, Chesney R 1990 Vitamin D and mineral metabolism in the very low birth weight infant receiving 400 IU of vitamin D. J Pediatr 116: 423–428. 19. Pittard WB, Geddes KM, Hulsey TC, Hollis BW 1991 How much vitamin D for neonates? Am J Dis Child 145: 1147–1149. 20. Backstrom MC, Maki R, Kuusela AL, Sievanen H, Koivisto AM, Ikonen RS, Kouri T, Maki M 1999 Randomised controlled trial of vitamin D supplementation on bone density and biochemical

21.

22.

23. 24.

25. 26.

27. 28.

29. 30. 31. 32. 33. 34.

35. 36. 37. 38.

indices in preterm infants. Arch Dis Child Fetal Neonatal Ed 80:F161–F166. Koo WW, Krug-Wispe S, Neylan M, Succop P, Oestreich AE, Tsang RC 1995 Effect of three levels of vitamin D intake in preterm infants receiving high mineral-containing milk. J Pediatr Gastroenterol Nutr 21:182–189. Backstrom MC, Maki R, Kuusela AL, Sievanen H, Koivisto AM, Koskinen M, Ikonen RS, Maki M 1999 The long-term effect of early mineral, vitamin D, and breast milk intake on bon mineral status in 9- to 11-year-old children born prematurely. J Pediatr Gastroenterol Nutr 29:575–582. Stathos TH, Shulman RJ, Schanler RJ, Abrams SA 1996 Effects of carbohydrates on calcium absorption in premature infants. Pediatr Res 39:666–670. Carnielli VP, Luijendijk IH, van Goudoever JB, Sulkers EJ, Boerlage AA, Degenhart HJ, Sauer PJ 1995 Feeding premature newborn infants palmitic acid in amounts and stereoisomeric position similar to that of human milk: Effects on fat and mineral balance. Am J Clin Nutr 61:1037–1042. Lucas A, Quinlan P, Abrams S, Ryan S, Lucas PJ 1997 Randomised controlled trial of a synthetic triglyceride milk formula for preterm infants. Arch Dis Child 77:F178–F184. Koo WW, Hammami M, Margeson DP, Nwaesei C, Montalto MB, Lasekan JB 2003 Reduced bone mineralization in infants fed palm olein-containing formula: A randomized, double-blinded, prospective trial. Pediatrics 111:1017–1023. Butte NF, Wong WW, Hopkinson JM, Smith EO, Ellis KJ 2000 Infant feeding mode affects early growth and body composition. Pediatrics 106:1355–1366. Institute of Medicine Food and Nutrition Board’s Standing Committee on the Scientific Evaluation of Dietary Intervals 1997 Calcium. In: Dietary Reference Intervals for Calcium, Phosphorus, Magnesium, Vitamin D and Fluoride. National Academy Press, Washington, DC, pp. 71–146. Life Sciences Research Office (LSRO) report 1998 Assessment of nutrient requirements for infant formulas. J Nutr 128: 2140S–2143S. DeVizia B, Fomon SJ, Nelson SE, Edwards BE, Ziegler EE 1985 Effect of dietary calcium on metabolic balance of normal infants. Pediatr Res 19:800–806. Abrams SA, Griffin IJ, Davila PM 2002 Calcium and zinc absorption from lactose-containing and lactose-free infant formulas. Am J Clin Nutr 76:442–446. Abrams S A, Wen J, Stuff JE 1997 Absorption of calcium, zinc, and iron from breast milk by five- to seven-month-old infants. Pediatr Res 41:384–390. Lifschitz CL, Abrams SA 1998 Addition of rice cereal to formula does not impair mineral bioavailability. J Pediatr Gastroenterol Nutr 26:175–178. Carnielli VP, Luijendijk IH, Van Goudoever JB, Sulkers EJ, Boerlage AA, Degenhart HJ, Sauer PJ 1996 Structural position and amount of palmitic acid in infant formulas: effects on fat, fatty acid, and mineral balance. J Pediatr Gastroenterol Nutr 23:553–560. Nelson SE, Frantz JA, Ziegler EE 1998 Absorption of fat and calcium by infants fed a milk-based formula containing palm olein. J Am Coll Nutr 17:327–332. Garn SM 1972 The course of bone gain and the phases of bone loss. Ortho Clin North Am l3:503–520. Specker BL, Beck A, Kalfwarf H, Ho M 1997 Randomized trial of varying mineral intake on total body bone mineral accretion during the first year of life. Pediatrics 99:E12. Koo WW, Bush AJ, Walters J, Carlson SE 1998 Postnatal development of bone mineral status during infancy. J Am Coll Nutr 17:65–70.

820 39. Schanler RJ, Burns PA, Abrams SA, Garza C 1992 Bone mineralization outcomes in human milk-fed preterm infants. Pediatr Res 31:583–586. 40. Fewtrell MS, Prentice A, Jones SC, Bishop NJ, Stirling D, Buffenstein R, Lunt M, Cole TJ, Lucas A 1999 Bone mineralization and turnover in preterm infants at 8–12 years of age: the effect of early diet. J Bone Miner Res 14:810–820. 41. Gafni RI, McCarthy EF, Hatcher T, Meyers JL, Inoue N, Reddy C, Weise M, Barnes KM, Abad V, Baron J 2002 Recovery from osteoporosis through skeletal growth: early bone mass acquisition has little effect on adult bone density. FASEB J 16:736–738. 42. Gershoff SH, Legg, MA, Hegsted DM 1958 Adaptation to different calcium intakes in dogs. J Nutr 64:303–312. 43. Hollis BW, Roos BA, Draper HH, Lambert PW 1981 Vitamin D and its metabolites in human and bovine milk. J Nutr 111: 1240–1248. 44. Greer FR, Hollis BW, Napoli JL 1984 High concentrations of vitamin D2 in human milk associated with pharmacologic doses of vitamin D2. J Pediatr 105:61–64. 45. Abrams SA 2002 Nutritional rickets: An old disease returns. Nutr Rev 60:111–115. 46. Heaney RP, Weaver CM 2003 Calcium and vitamin D. Endocrinol Metab Clin North Am 32:181–194. 47. Abrams SA, Griffin IJ, Davila P, Liang L 2001 Calcium fortification of breakfast cereal enhances calcium absorption in children without affecting iron absorption. J Pediatr 139: 522–526. 48. Matkovic V, Heaney RP 1992 Calcium balance during human growth: evidence for threshold behavior. Am J Clin Nutr 55:992–996. 49. Ames SK, Gorham BM, Abrams SA 1999 Effects of high vs low calcium intake on calcium absorption and red blood cell iron incorporation by small children. Am J Clin Nutr 70:44–48. 50. Abrams SA, Copeland KC, Gunn SK, Stuff JE, Clark LL, Ellis KJ 1999 Calcium absorption and kinetics are similar in 7- and 8-year-old Mexican-American and Caucasian girls despite hormonal differences. J Nutr 129:666–671. 51. Johnston CC Jr, Miller JZ, Slemenda CW, Reister TK, Hui S, Christian JC, Peacock M 1992 Calcium supplementation and increases in bone mineral density in children. N Engl J Med 327:82–87. 52. Martin AD, Bailey DA, McKay HA, Whiting S 1997 Bone mineral and calcium accretion during puberty. Am J Clin Nutr 66:611–615. 53. Matkovic V, Jelic T, Wardlaw GM, Ilich JZ, Goel PK, Wright JK, Andon MB, Smith KT, Heaney RP 1994 Timing of peak bone mass in Caucasian females and its implication for the prevention of osteoporosis. J Clin Invest 93:799–808. 54. Bryant RJ Wastney ME, Martin BR, Wood O, McCabe GP, Morshidi M, Smith DL, Peacock M, Weaver CM 2003 Racial differences in bone turnover and calcium metabolism in adolescent females. J Clin Endocrinol Metab 88:1043–1047. 55. McKay HA, Bailey DA, Mirwald RL, Davison S, Faulkner RA 1998 Peak bone mineral accrual and age at menarche in adolescent girls: A 6-year longitudinal study. J Pediatrics 133:682–687. 56. Bell NH, Yergey AL, Vieira NE, Oexmann MJ, Shary JR 1993 Demonstration of a difference in urinary calcium, not calcium absorption, in black and white adolescents. J Bone Miner Res 8:1111–1115. 57. Ames SK, Ellis KJ, Gunn SK, Copeland KC, Abrams SA 1999 Vitamin D receptor gene Fok1 polymorphism predicts calcium

STEVEN A. ABRAMS

58. 60.

61. 62. 63. 64.

65.

66.

67.

68. 69. 70. 71. 72. 73. 74.

75.

76. 77. 78.

absorption and bone mineral density in children. J Bone Miner Res 14:740–746. Bronner F, Abrams SA 1998 Development and regulation of calcium metabolism in healthy girls. J Nutr 128:1474–1480. Abrams SA, Copeland KC, Gunn SK, Gundberg CM, Klein KO, and Ellis KJ 2000 Calcium absorption, bone accretion and kinetics increase during early pubertal development in girls. J Clin Endocrinol Metab 85:1805–1808. Leitch I, Aitken FC 1959 The estimation of calcium requirement: A re-examination. Nutr Abst Rev 29:393–409. Abrams SA, O’Brien KO, Liang LK, Stuff JE 1995 Differences in calcium absorption and kinetics between black and white girls age 5–16 years. J Bone Miner Res 10:829–833. Abrams SA, Griffin IJ, Davila P, Liang L, Powledge D 2001 Effects of very low calcium intake on calcium metabolism in pubertal girls. FASEB J 15:A1095. Lehtonen-Veromaa MK, Mottonen TT Nuotio IO, Irjala KM, Leino AE, Viikari JS 2002 Vitamin D and attainment of peak bone mass among peripubertal Finnish girls: A 3-y prospective study. Am J Clin Nutr 76:1446–1453. Guillemant J, Le HT, Maria A, Allemandou A, Peres G, Guillemant S 2001 Wintertime vitamin D deficiency in male adolescents: Effect on parathyroid function and response to vitamin D3 supplements. Osteoporos Int 12:875–879. Abdullah MA, Salhi HS, Bakry LA, Okamoto E, Abomelha AM, Stevens B, Mousa FM 2002 Adolescent rickets in Saudi Arabia: A rich and sunny country. J Pediatr Endocrinol Metab 15:1017–1025. Goulding A, Jones IE, Taylor RW, Manning PJ, Williams SM 2000 More broken bones: A 4-year double cohort study of young girls with and without distal forearm fractures. J Bone Miner Res 15:2011–2018 Wyshak G, Frisch RE 1994 Carbonated beverages, dietary calcium, the dietary calcium/phosphorus ratio, and bone fractures in girls and boys. J Adolesc Health 15:210–215. Teegarden D, Lyle RM, Proulx WR, Johnston CC, Weaver CM 1999 Previous milk consumption is associated with greater bone density in young women. Am J Clin Nutr 69:1014–1017. Wyshak G 2000 Teenaged girls, carbonated beverage consumption, and bone fractures. Arch Pediatr Adolesc Med 154:610–613. Spencer H, Kramer L, Osis D 1988 Do protein and phosphorus cause calcium loss? J Nutr 118:657–660. Calvo MS 2000 Dietary considerations to prevent loss of bone and renal function. Nutrition 16:564–566. Allison DB 2001 Hold the cola alarm. Arch Pediatr Adolesc Med 155:201–202. Murphy SP, Beaton GH, Calloway DH 1992 Estimated mineral intakes of toddlers: Predicted prevalence of inadequacy in village populations in Egypt, Kenya, and Mexico. Am J Clin Nutr 56:565–572. Murphy SP, Calloway DH, Beaton GH 1995 Schoolchildren have similar predicted prevalences of inadequate intakes as toddlers in village populations in Egypt, Kenya, and Mexico. Eur J Clin Nutr 49:647–657. Wyatt CJ, Hernandez ME, Mendez RO 1996 Dialyzable calcium and phosphorus in Mexican diets high in insoluble fiber. J Agric Food Chem 46:4662–4666. Wyatt CJ, Tejas MAT 2000 Nutrient intake and growth of preschool children from different socioeconomic regions in the city of Oaxaca, Mexico. Ann Nutr Metab 44:4–20. Lacasana M, Romiel I, Sanis LH, Palazuelos E, HernandezAvila M 2000 Blood lead levels and calcium intake in Mexico City children under five years of age. Int J Environ Health Res 10:331–340.

CHAPTER 49 Calcium Absorption in Children

79. Calvo MS, Whiting SJ 2003 Prevalence of vitamin D insufficiency in Canada and the United States: Importance to health status and efficacy of current food fortification and dietary supplement use. Nutr Rev 61:107–113. 80. Andon MB, Peacock M, Kanerva RL, De Castro JA 1996 Calcium absorption from apple and orange juice fortified with calcium citrate malate (CCM). J Am Coll Nutr 15:313–316. 81. Martin BR, Weaver CM, Heaney RP, Packard PT, Smith DL 2002 Calcium absorption from three salts and CaSO4-fortified bread in premenopausal women. J Agric Food Chem 50:3874–3876. 82. Heaney RP, Weaver CM 2003 Calcium and vitamin D. Endocrinol Metab Clin North Am 32:181–194. 83. Coudray C, Bellanger J, Castiglia-Delavaud C, Rémésy C, Vermorel M, Rayssignuier Y 1997 Effect of soluble or partly soluble dietary fibres supplementation on absorption and balance of calcium, magnesium, iron and zinc in healthy young men. Eur J Clin Nutr 51:375–380.

821 84. Roberfroid MB 1999 Concepts in functional foods: The case of inulin and oligofructose. J Nutr 129:1398S–1401S. 85. Griffin IJ Davila PM, Abrams SA 2002 Non-digestible oligosaccharides and calcium absorption in girls with adequate calcium intakes. Br J Nutr 287:S187–S191. 86. Griffin IJ, Hicks PMD, Heaney RP, Abrams SA 2003 Enhanced chicory inulin increases calcium absorption mainly in adolescents with lower calcium absorption. Nutr Res 23:901–909. 87. Rémésy C, Levrat MA, Gamet L, Demigne C 1993 Cecal fermentations in rats fed oligosaccharides (inulin) are modulated by dietary calcium level. Am J Physiol 264:G855–G862. 88. Brommage R, Binacua C, Antille S, Carrie AL 1993 Intestinal calcium absorption in rats is stimulated by dietary lactulose and other resistant sugars. J Nutr 123:2186–2194. 89. Coudray C, Fairweather-Tait SJ 1998 Do oligosaccharides affect the intestinal absorption of calcium in humans? Am J Clin Nutr 68:921–923.

CHAPTER 50

Vitamin D Metabolism and Aging BERNARD P. HALLORAN AND ANTHONY A. PORTALE Departments of Medicine and Pediatrics, University of California, and Division of Endocrinology Veterans Affairs Medical Center, San Francisco, California

I. II. III. IV. V.

Introduction Cutaneous Production of Vitamin D Dietary Vitamin D and Intestinal Absorption Synthesis of 25-Hydroxyvitamin D Synthesis and Metabolism of 1,25-Dihydroxyvitamin D

I. INTRODUCTION Aging in the context of this chapter refers to postmaturational aging and not to growth and development. Postmaturational aging is a complex process beginning with senescence at the cellular level. All cells gradually change phenotypically and eventually lose the ability to proliferate as they age. Tissues and organs also age with consequent changes in metabolism and function. And of course organisms age resulting in diminished health and performance. Aging is heterogeneous. Differences in genetic makeup and lifestyle (diet, activity level, environment) influence the progress of senescence. Some people appear to age more rapidly than others. Disease confounds the aging process, and thus the changes observed in an aging population might be a consequence of true aging or the cumulative effect of chronic disease. The changes that occur during aging are often subtle, but their compound effect can be dramatic. Predictably, aging influences vitamin D metabolism, and vitamin D influences the aging process. This chapter deals with how aging affects cutaneous production, dietary availability, metabolism, and action of vitamin D, and how vitamin D influences the progression of aging. An attempt is made to separate the effects of common age-related diseases from the effects of aging per se.

II. CUTANEOUS PRODUCTION OF VITAMIN D The production of vitamin D in the skin is described in Chapter 3. The structure and function of the skin begin to deteriorate during the third decade of life [1]. Epidermal thinning begins around age 20 and tissue loss continues with advancing age [2]. Skin elasticity, keratinocyte number, and cell turnover rate decrease with aging [3–6]. The pattern in gene expression in cultured VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

VI. Tissue Responsiveness and the Role of Vitamin D in the Aging Process VII. Conclusions References

keratinocytes also changes with donor age [7,8]. Barrier function is compromised and total lipid content (including cholesterol) in the stratum corneum is decreased in aged animals [9]. Skin blood flow decreases by nearly 40% between the ages of 20 and 70 years, which reduces dermal clearance of vitamin D [10]. Cutaneous production of vitamin D decreases with advancing age [11–13]. Approximately 80% of the vitamin D formed in the skin is produced in the epidermis, and the amount of precursor to vitamin D3, 7-dehydrocholesterol, is decreased in the epidermis of elderly subjects. Despite a decrease in the number of melanocytes in the skin (normally melanin in the melanocyte acts as a natural sunscreen to reduce production of vitamin D), conversion of 7-dehydrocholesterol to previtamin D3 in human skin samples exposed to ultraviolet radiation is decreased as much as twofold in elderly subjects. Holick et al. [13,14] report that whole body exposure to one minimal erythemal dose of ultraviolet B radiation can increase serum vitamin D to a maximum of 78.1 nmol/liter-ng/ml in young subjects but to a maximum of only 20.8 nmol/liter-ng/ml in elderly subjects 68–80 years of age (Fig. 1). The decrease in cutaneous vitamin D synthesis is presumably a consequence of the decrease in substrate (7-dehydrocholesterol) and reduced dermal clearance of vitamin D. The diminished ability to produce vitamin D in the skin of older subjects is often aggravated by changes in lifestyle and environmental factors. Many older people are homebound or hesitant to venture outdoors, and when exposed to the sun frequently wear potent sunscreens to reduce the risk of skin cancer. Sunlight that has passed through a glass windowpane will not induce the synthesis of vitamin D because of the UV absorbance by the glass [14]. Furthermore, chronic use of sunscreens can dramatically reduce serum vitamin D levels [15]. The elderly, especially those that are Copyright © 2005, Elsevier, Inc. All rights reserved.

Serum concentration of vitamin D (nmol/L)

824

BERNARD P. HALLORAN AND ANTHONY A. PORTALE

100 80 60 Young 40 20 Elderly 0 0

1

2

3

4

5

6

7

Day

FIGURE 1

Effect of whole-body exposure to one minimal erythemal dose of ultraviolet B radiation on the serum concentrations of vitamin D in young (22–30 years) and elderly volunteers. From Holick M [14].

homebound or confined to a nursing home are, therefore, at increased risk of becoming vitamin D deficient, as a result of both diminished efficiency in cutaneous production and reduced effective solar exposure. Gloth et al. [16] report that of 244 elderly men and women over the age of 65 who were deprived of direct sunlight exposure (homebound elderly and nursing home residents) 54% of the community dwellers and 38% of nursing home residents have serum 25OHD levels below 25 nmol/liter (normal range = 25–137 nmol/liter). Obesity is another factor that can contribute to vitamin D deficiency in the elderly. Body mass index (BMI) frequently increases with age; vitamin D is stored in fat tissue, and vitamin D levels have been shown to inversely correlate with BMI after irradiation or oral vitamin D administration [17]. The subject of vitamin D insufficiency in elderly subjects in relationship to osteoporosis is discussed in Chapters 65 and 66, and by Lips [18].

III. DIETARY VITAMIN D AND INTESTINAL ABSORPTION A. Dietary Intake Dietary intake of vitamin D, because of differences in dietary supplementation and food preferences, varies over a broad range in the elderly. The United States Department of Agriculture (USDA) previously established the adult recommended daily allowance (RDA) for vitamin D at 5.0 µg/d or 200 IU/d [19]. Because of the vagaries of cutaneous production of vitamin D, the USDA has abandoned the term RDA in favor of using

the phrase Adequate Intake (AI). The USDA has set the AIs for young adults (< 50 years) at 200 IU/d, middle-aged adults (50–70 years) at 400 IU/d, and elderly adults (> 70 years) at 600 IU/d [20]. The major dietary sources of vitamin D are milk, milk products, butter, fortified margarine, eggs, and some fatty fish. Data from a survey conducted by the National Center for Health Statistics, Centers for Disease Control (National Health and Nutrition Examination Survey — NHANES III, 1988–1994), indicate that the usual adult (all ages) dietary intake of vitamin D in the United States is approximately 5 µg/d and does not change with postmaturational aging [21]. More recent data will be available shortly [22]. Krall et al. [23] in a study of 333 women observed vitamin D intakes ranging from 0.5 to more than 40 µg/d. With advancing age, protein, fat, and total calorie intake decrease [24,25]. Dietary calcium (excluding supplements but combining men and women) decreases from 909 mg/d between 20 and 39 years to 721 mg/d after 60 years [22] and vitamin D intake is tightly coupled to total dietary calcium. Thus, one would predict that vitamin D intake might decrease with age. However, vitamin D supplementation of the normal diet can account for as much as 50% of the total intake of the vitamin [26,27] and supplementation in at least some people increases with age. Sowers et al. [27] report that in 373 women ranging in age from 20 to 80 years the intake of vitamin D from supplements increased from 104 ± 18 IU/d in 25- to 40-year-olds to 202 ± 43 IU/d in 60- to 75-year-olds. Despite such increases, Sharkey et al. [28], who studied a population of cognitively eligible homebound men and women (age 60–85 years) whose meals were delivered to their homes, found that 100% of the men and 99.6% of the women had intakes of vitamin D below the AIs. Although those elderly that are ill or confined indoors often have inadequate total vitamin D intakes [29], the predominance of evidence suggests that in most healthy, free-living elderly the total intake of vitamin D does not change with age [23,25,26,30]. Collectively, direct calculation of total vitamin D intake in elderly populations suggests that overall intake decreases with age in some elderly, especially those that are chronically ill or confined indoors [21,23,27,29], and not in others. Based on the latest Department of Agriculture AIs, which recommend successive increases in dietary vitamin D with aging, this may still be inadequate to meet physiological needs.

B. Intestinal Absorption of Vitamin D Whether intestinal absorption of vitamin D changes with age is controversial. Vitamin D is absorbed in the

CHAPTER 50 Vitamin D Metabolism and Aging

proximal intestine, and in the absence of disease associated with intestinal malabsorption, some studies, but not all [31,32], indicate that vitamin D absorption remains normal in the elderly [33–35,36]. Clemens et al. [33] compared serum vitamin D2 levels in young adults and 25 chronically institutionalized but otherwise healthy elderly adults (mean age 72) after administration of 50,000 IU of vitamin D2. In the absence of gastrointestinal disease, they found no evidence of impaired vitamin D absorption. In contrast, Harris and Dawson-Hughes [36] studied the effect of administering 1800 IU/d of vitamin D2 on plasma concentrations in young and elderly men. The increase in 25-hydroxyvitamin D2 was greater in the young than in the elderly men by nearly twofold. The data suggest that either intestinal absorption or liver hydroxylation of vitamin D is impaired with aging. Importantly, studies to examine whether aging disrupts enterohepatic recirculation of vitamin D have not been performed. The effects of gastrointestinal and hepatobiliary disease on vitamin D absorption are discussed in Chapter 75.

C. Consequences of Vitamin D Insufficiency The consequences of diminished cutaneous production (all elderly), insufficient intake (some elderly), and impaired absorption (some elderly) in the elderly are inadequate vitamin D status. Serum concentrations of vitamin D and 25-hydroxyvitamin D (25OHD) are frequently inadequate to support normal mineral homeostasis [16,18,25,37–45]. The serum concentration of 25OHD in the elderly, a reflection of serum vitamin D levels (see Section IV, on synthesis of 25-hydroxyvitamin D) is related inversely to age and directly to sun exposure. In the presence of disease, malabsorption may contribute to poor vitamin D status. In a study of 433 postmenopausal women, Need et al. [42] showed that serum 25OHD decreased with age and was positively correlated to hours of sunlight. Van der Wielen et al. [38], in a study of 824 elderly people from 11 European countries, report that 36% of men and 47% of women had serum 25OHD levels below 12 nmol/L (the lower range of normal). Serum 25OHD concentrations were directly related to hours of ultraviolet light exposure and factors of physical health status. A study of Fardellone et al. [46] further exemplifies the importance of sun exposure and health status, as well as demonstrates the heterogeneity of aging populations. These investigators studied a group of chronically institutionalized elderly and observed a mean serum 25OHD of only 3.7 nmol/L. Eighty-five percent of the subjects had serum 25OHD levels below 5 nmol/L and 98% had levels below 10 nmol/L (normal

825 range = 10–55 nmol/L). Clearly many older people are vitamin D insufficient or frankly deficient. Disease and medication use frequently exacerbate the problem. Lips [18] illustrates this by reporting a survey of serum 25OHD in elderly subjects by health and residence category. Mild hyperparathyroidism, increased bone turnover, and diminished bone density are associated with moderate to severe vitamin D insufficiency in the elderly. In elderly populations serum 25OHD has been observed to be inversely correlated with both the serum concentration of parathyroid hormone (PTH), markers of bone turnover and bone loss [18,23,42,46–54]. Ooms et al. [48], in a study of 330 healthy women over the age of 70, report that serum 25OHD in 65% of the subjects was less than 12 nmol/L. 25-Hydroxyvitamin D was negatively correlated to serum PTH but only at levels of 25OHD below 10 nmol/L. Bone mineral density was positively correlated with serum 25OHD but only below 12 nmol/L. Furthermore, in a prospective study of 9704 elderly women, Stone et al. [53] report that lower levels of serum 25OHD are associated with increased bone loss in the hip. Severe dietary vitamin D deficiency is associated with increased risk of hip fracture [54]. Supplementation with vitamin D increases serum 25OHD, in most cases decreases serum PTH, and normalizes bone turnover [47,55–57]. Brazier et al. [47] compared vitamin D–sufficient and insufficient elderly men and women before and after treatment of the insufficient group with 800 IU of vitamin D/d. In the vitamin D insufficient subjects, serum PTH and markers of bone turnover were increased prior to treatment but normalized to levels found in age-matched vitamin D– sufficient subjects after treatment. These data suggest that dietary supplementation of elderly men and women with poor or marginal vitamin D status may be beneficial in reducing bone loss associated with aging. Indeed, Dawson-Hughes et al. [58] report that vitamin D supplementation (10 µg/d) to postmenopausal women with an otherwise normal mean intake of 5 µg/d can reduce wintertime bone loss. Further studies by these investigators [59] in 247 healthy ambulatory postmenopausal women consuming an average of 2.5 µg/d of vitamin D showed that supplementation with 17.5 µg/d could reduce the rate of loss of bone mineral density in the femoral neck year-round. Interestingly, the anabolic effects of vitamin D treatment, although associated with an increase in serum 25OHD, are not usually accompanied by an increase in serum 1,25(OH)2D unless the patients are frankly vitamin D deficient [49]. The report by Barger-Lux et al. [60] demonstrating a positive correlation between serum 25OHD (but not 1,25(OH)2D) and calcium absorption in healthy premenopausal

826 women treated with 25OHD is consistent with this idea. These data suggest that 25OHD may be acting directly on the intestine to stimulate calcium absorption (see Chapter 46). Not all individuals benefit from vitamin D supplementation. Orwoll et al. [61] studied normal healthy men ranging in age from 30 to 87 years. Administration of both calcium (1000 mg/d) and vitamin D (25 µg/d) did not affect the normal rate of bone loss. Importantly, however, mean basal dietary calcium and vitamin D in this population was 1159 mg/d and approximately 9 µg/d, respectively. These data are consistent with the findings of Ooms et al. [55] and suggest that elderly populations with adequate dietary calcium and adequate vitamin D status are not likely to improve their mineral balance with additional vitamin D.

IV. SYNTHESIS OF 25-HYDROXYVITAMIN D Synthesis of 25OHD does not appear to be influenced by aging with some exceptions [15,31,36]. The primary site for synthesis of 25OHD is the liver (see Chapter 4). Severe liver disease can lead to a reduction in the synthesis of 25OHD and the serum concentrations of 25OHD, vitamin D binding protein, and total (but not free) 1,25(OH)2D [62,63], and elderly patients with inadequate hepatic function may have a reduction in vitamin D 25-hydroxylase reserve. In the absence of hepatic disease, however, most studies suggest that the activity of the vitamin D-25-hydroxylase is normal in the elderly. Aknes et al. [37] examined the relationships among the serum concentrations of vitamin D, 25OHD, and vitamin D binding protein (DBP) in healthy adults ranging in age from 22 to 96 years. With advancing age, they observed decreases in vitamin D, 25OHD, and DBP but no change or a small increase in the molar ratio of 25OHD to vitamin D. This suggests that hepatic hydroxylation at the 25-position of vitamin D, in the absence of disease, is not impaired by aging. Matsuoka et al. [15] studied the response of serum 25OHD2 to an oral load of vitamin D2 in young and elderly subjects. The increase in serum 25OHD over time was similar between age groups, providing evidence that both intestinal absorption and 25-hydroxylation are normal in healthy elderly people. This contrasts with the findings of Harris et al. [32]. These investigators found that the increase in serum 25OHD in the elderly in response to a 3-week course of oral vitamin D was roughly half that of young subjects, implying either intestinal malabsorption or a defect in 25OHD synthesis.

BERNARD P. HALLORAN AND ANTHONY A. PORTALE

V. SYNTHESIS AND METABOLISM OF 1,25-DIHYDROXYVITAMIN D A consensus regarding the effects of aging on the synthesis and metabolism of 1,25(OH)2D is slowly emerging. The issue is complex because of the heterogeneity of the aging process and the independent effects of aging on renal function, acid–base balance, sex steroid levels, growth hormone status, and other factors, each of which can affect the synthesis and metabolism of 1,25(OH)2D (see Chapter 5).

A. Serum Concentration of 1,25(OH)2D The effect of aging on the serum concentration of 1,25(OH)2D is controversial. With advancing age in men, serum concentrations of 1,25(OH)2D are reported to either decrease [64,65] or remain unchanged [66–68]. In women, serum 1,25(OH)2D is reported to decrease [64,69–71], remain unchanged [67], to increase and then decrease [65], or to decrease and then increase [72]; and in mixed populations to either decrease [73] or remain unchanged [33,74–76]. Findings in aging animals are similar to those in humans. Using the rat as a model of human aging, serum 1,25(OH)2D levels are reported to be decreased [77–80], unchanged [81,82], or increased [83] in aged animals compared to those in younger animals. These apparently discrepant findings result in part from differences in the definition of “old” and differences in the gender, lifestyle, presence of disease, and medication use in the populations studied. Accounting for these differences and better defining the populations studied, has permitted reconciliation of earlier apparent discrepancies. In animal studies of aging, the definition of “old” is critical to their interpretation. The most commonly used animal model for aging studies is the Fischer 344 (F344) rat, which becomes sexually mature at approximately 3 months of age and has a mean lifespan of 24 months. Serum levels of 1,25(OH)2D are higher in young growing F344 rats when compared to levels in mature “adult” animals 3–4 months of age. In many “aging” studies, findings in growing or not yet mature animals have been compared to those in adult animals. Such studies show that serum 1,25(OH)2D levels are lower in adult than in growing animals. However, although such studies have been interpreted to indicate that serum 1,25(OH)2D decreases with age, they do not address the question of whether serum 1,25(OH)2D levels change during postmaturational aging: that is, during the period between 3 and 24 months of age.

827

CHAPTER 50 Vitamin D Metabolism and Aging

Serum concentration of 1,25(OH)2D (pmol/L)

150

100

50

0 MCR of 1,25(OH)2D (mL . s−1 . 70 kg IBW−1)

0.9

0.6

0.3

0.0 90

PR of 1, 25(OH)2D (fmol . s−1 . 70 kg IBW−1)

Nevertheless, even when findings in young adult animals (4–6 months) are compared to those in aged animals (20–28 months) discrepancies remain. The heterogeneity of the aging process and the small number of animals studied in aging studies can account, at least in part, for differences in the findings. Loss of renal function is a major cause of death in the aging F344 rat. Since 1,25(OH)2D is synthesized in the kidney, it is predictable that serum 1,25(OH)2D will be reduced in aged animals with renal insufficiency. However, in aged (24-month) male F344 rats with normal or near-normal glomerular filtration rates, serum concentrations of 1,25(OH)2D are not significantly different from those in young adult (6-month) male F344 rats [82]. It is of interest that in male F344 rats between the ages of 12 and 24 months with normal renal function, the serum concentration of 1,25(OH)2D tends to increase! The trend does not reach statistical significance, but may explain why some investigators have observed an increase in serum 1,25(OH)2D in the aging rat [83]. In humans, renal function deteriorates with postmaturational aging [84–87]. The rate of deterioration varies from subject to subject, and some older people maintain normal or near-normal glomerular filtration rates (GFR) well into their eighth decade of life. In healthy elderly men in whom renal function is normal or near normal, serum 1,25(OH)2D levels are not significantly different from those in young men [68] (Fig. 2). However, in elderly individuals with moderate to severe renal insufficiency (GFR < 70 ml/min), renal synthesis and serum concentrations of 1,25(OH)2D are reduced. Thus, the low serum concentrations of 1,25(OH)2D reported in many elderly populations reflect in large part the age-related fall in GFR [67,68,76]. Moreover, serum vitamin D binding protein (DBP) levels in men do not change with age [73] suggesting that free levels of 1,25(OH)2D do not change. In women, comparison of 1,25(OH)2D concentrations in elderly and young women is confounded by the potential effects of menopause in the elderly, and the observation that serum 1,25(OH)2D changes during the menstrual cycle in younger women. Most data indicate that serum total 1,25(OH)2D levels in healthy women remain constant or gradually increase through the eighth decade [67,88,89]. Although menopause appears to have little effect on the serum concentration of 1,25(OH)2D [72,90,91], a transitory decrease in total serum 1,25(OH)2D has been observed 5–15 years after the menopause, followed by a gradual increase in serum 1,25(OH)2D that correlates with rising serum PTH concentrations [72]. The serum concentration of DBP is reported to be either decreased [73,92,93], normal [37,43], or

60

30

Young men

Old men

0

FIGURE 2 Serum concentration, metabolic clearance rate (MCR), and production rate (PR) of 1,25(OH)2D in healthy young and elderly men after ingesting a constant normal diet for 9 days. The mean of each population is indicated by a dashed line. From Halloran et al. [68].

increased [72] in postmenopausal women. Prince et al. [72] report that in a cross-sectional study of 655 women ranging in age from 35 to 90 years, circulating DBP levels increase after the menopause, then decrease transiently, and finally increase progressively with advancing age, resulting in a progressive decrease in the serum free 1,25(OH)2D index beginning 10–15 years after the menopause. Other studies, however, do not support these findings [92], and thus it is not clear whether the serum concentration of free 1,25(OH)2D changes with age in women.

828

BERNARD P. HALLORAN AND ANTHONY A. PORTALE

B. Kinetics of 1,25(OH)2D Metabolism The serum concentration of 1,25(OH)2D is determined by its production rate (PR) and its metabolic clearance rate (MCR). The MCR of 1,25(OH)2D in the rat is increased by vitamin D deficiency [94] and chronic exogenous administration of 1,25(OH)2D [95], and in the rat [94] and pig [96], by depletion of dietary calcium. The MCR of 1,25(OH)2D is unaffected by changes in dietary phosphorus [96,97], PTH administration [98], and glucocorticoid excess [99]. Loss of renal function decreases the MCR of 1,25(OH)2D in the rat [100]. With advancing age in the F344 male rat with normal or near normal renal function [82], the MCR of 1,25(OH)2D gradually increases (Table I). When compared to values in 6-month-old animals, the MCR is 24% (p < 0.10) higher at 12 months, 30% ( p < 0.05) higher at 18 months and 57% higher ( p < 0.01) at 24 months of age. The PR of 1,25(OH)2D also increases with age in the F344 male rat, and by 24 months of age is 91% (p < 0.01) higher than at 6 months. The increase in 1,25(OH)2D production correlates positively with an increase in serum PTH, presumably reflecting the normal stimulatory effect of PTH on 1α-hydroxylase activity. These data demonstrate that both production and clearance of 1,25(OH)D can increase during postmaturational aging in the rat. They also suggest that in animals with normal GFRs, the kidney remains sensitive to PTH. The kinetics of 1,25(OH)2D metabolism in healthy men [68] and women [88] are unchanged with advancing age. We have shown, in a group of healthy elderly men (mean age 72 years) with normal or near-normal GFRs studied under strictly controlled metabolic conditions, that the PR and MCR of 1,25(OH)2D, as well as its serum concentration, are not different from those in healthy young men (mean age 34 years) [68] (Fig. 2). In both groups of men, the serum 1,25(OH)2D concentrations correlated with values of PR but not of MCR.

TABLE I Effect of Aging on the Serum Concentrations of PTH and 1,25(OH)2D, and the Production and Metabolic Clearance Rates of 1,25(OH)2D in the Rat Age (months) PTH (pg/ml) 1,25(OH)2D (pg/ml) PR (pg/min) MCR (µl/min)

6

12

18

15 ± 2 18 ± 3 22 ± 5a 29 ± 4 25 ± 4 31 ± 3 1.1 ± 0.2 1.1 ± 0.2 1.4 ± 0.1 37 ± 1 46 ± 4 48 ± 4a

24 36 ± 8a 35 ± 4 2.1 ± 0.2a 58 ± 2a

Values are mean ± SEM, n = 6, ap < .05. From Wada et al. [82].

Serum 1,25(OH)2D levels did not correlate with those of 25OHD, suggesting that substrate concentration was not limiting for the production of 1,25(OH)2D in both elderly and young men.

C. Synthesis of 1,25(OH)2D: Trophic Factors 1. SEX STEROIDS

Serum total and free testosterone levels decrease with advancing age in men [101,102], and administration of testosterone to hypogonadal men has been shown to increase modestly both serum total and free 1,25(OH)2D concentrations [103]. These data suggest that the decline in serum testosterone levels with aging may reduce tonic stimulation of the 1α-hydroxylase. Estrogen deficiency leads to a decrease in serum 1,25(OH)2D levels regardless of age [75,104,105]. Estrogen replacement in postmenopausal women can increase both total and free serum 1,25(OH)2D [92,104,106–110], suggesting that with menopause and the accompanying estrogen deficiency, an important trophic factor for the maintenance of serum 1,25(OH)2D is lost in aging women. 2. CALCIUM AND PHOSPHORUS

The morning fasting serum concentration of total calcium in elderly men and women is reported to be either normal [67,69,111], decreased [43,69,111,112], or increased [113,114] when compared with values in younger women. Serum total calcium concentrations can directly influence the serum concentration of 1,25(OH)2D [115], but studies to determine whether aging in humans affects the capacity of calcium to modulate serum 1,25(OH)2D have not been reported. Armbrecht et al. [116] administered a low-calcium diet to young (2-month) and adult (12-month) F344 rats and observed a severalfold increase in serum 1,25(OH)2D concentrations and in renal 1α-hydroxylase mRNA abundance in both groups of animals. However, the levels attained in the adult rats were only 37% and 10%, respectively, of those in the young rats, even though serum levels of PTH were stimulated by the low-calcium diet to a comparable extent in both groups of animals. The effect of postmaturational aging on the response to low-calcium diet was not addressed in these studies. There is general agreement that the serum phosphorus concentration decreases with advancing age in men and remains unchanged in women [67,69]. We showed in healthy young men that moderate and severe restriction of dietary phosphorus can stimulate the serum concentration of 1,25(OH)2D, independently of changes in serum concentrations of either PTH or blood ionized

829

CHAPTER 50 Vitamin D Metabolism and Aging

TABLE II Whole Blood Ca2+ in Healthy Young (39 ±1 yr, n = 13) and Elderly (74 ± 2 yr, n = 9) Men (GFR > 70 ml/min/1.73 m2 ) after 9 Days on a Constant Whole-Food Diet

Whole blood Ca2+ (mg/dl) Morning fasting 24 hr mean aValues

Young men

Elderly men

4.84 ± 0.03a 4.77 ± 0.04

4.84 ± 0.04 4.80 ± 0.04

are mean ± SE. From Portale et al. [121].

TABLE III Serum Concentrations of Phosphorus (24 hr Mean) and 1,25(OH)2D in Healthy Young (29 ± 2 yr, n = 9) and Elderly Men (71 ± 1 yr, n = 7) Consuming Constant Whole-Food Diets Containing 625, 1500, 2300 mg/d/70 kg Diet P (mg/d/70 kg)

Serum P (mg/dl) Young men Elderly men Serum 1,25(OH)2D (pg/ml) Young men Elderly men aValues

650

1500

2300

3.7 ± 0.1a 3.2 ± 0.2

4.2 ± 0.1 3.7 ± 0.2

4.3 ± 0.1 3.8 ± 0.2

43 ± 2 43 ± 2

36 ± 2 33 ± 2

29 ± 2 31 ± 2

are mean ± SE. From Portale et al. [120].

dietary intake of phosphorus, serum 1,25(OH)2D levels in the young and elderly men were virtually identical, but serum phosphorus levels in the elderly were lower than those in the young men (Table III). Across the range of phosphorus intakes, the 24-hr mean serum concentration of phosphorus varied inversely with serum 1,25(OH)2D concentrations in both groups of subjects (r = −0.92, p < 0.0001) (Fig. 3); the slope of

50 Elderly Young

Serum 1,25(OH)2D (pg/ml)

calcium, and these changes in serum 1,25(OH)2D are due to changes in its PR, since its MCR does not change [97,117,118]. Since dietary and serum phosphorus can physiologically regulate 1,25(OH)2D production in humans, it is of interest to determine whether such regulation is affected by aging. Villa et al. [119] report that in postmenopausal women given aluminum hydroxide to reduce intestinal phosphorus absorption, serum phosphorus concentrations decreased by 17% and serum 1,25(OH)2D increased by 38%, without changes in serum total or ionized calcium or PTH concentrations. Although a control group of younger women was not similarly studied, the investigators concluded that older women retain the capacity to increase 1,25(OH)2D concentrations in response to dietary phosphorus restriction. We examined the effects of aging in healthy elderly men on the capacity of phosphorus restriction to stimulate serum 1,25(OH)2D. We first studied healthy young and elderly men with normal or near-normal GFR while receiving a constant whole-food diet containing normal amounts of phosphorus and calcium [120,121]. Both the morning fasting and the 24-hr mean blood concentrations of total and ionized calcium did not differ between young and elderly men (Table II). In contrast, both the morning fasting and the 24-hr mean serum phosphorus concentrations were significantly lower in the elderly men than in the young men. Since a low serum phosphorus concentration would be expected to stimulate 1,25(OH)2D production, mild hypophosphatemia in the elderly would be expected to increase serum levels of 1,25(OH)2D. Given our findings and those of others that serum 1,25(OH)2D is not increased in this population, this suggests that the capacity of phosphorus to regulate renal 1,25(OH)2D production is diminished with age. To investigate this possibility, we then varied dietary phosphorus within its normal range: With restriction of dietary phosphorus from 2300 to 625 mg/d, the magnitude of the increase induced in serum 1,25(OH)2D (47%) in the elderly was virtually the same as that induced in the young men. At each

40

30

3.0

3.5 4.0 24-hr Mean serum phosphorus (mg/dl)

4.5

FIGURE 3 Relationship between concentrations of serum 1,25(OH)2D and 24-hr mean serum phosphorus when dietary phosphorus was normal (squares), then supplemented (triangles), and then restricted (circles) in healthy elderly and young men. Each point represents the mean ± SE of serum 1,25(OH)2D and 24-hr serum phosphorus at steady state. Multiple linear regression analysis indicates that the intercept is significantly lower in elderly men (p < .001). From Portale et al. [120].

830

3. PARATHYROID HORMONE

The serum concentration of parathyroid hormone (PTH) increases progressively with advancing age in men and women [65–68,87,88,111,121–124]. As early as the fifth decade [111], serum PTH concentrations are increased in normal healthy men, and by age 70 are two- to threefold higher than values in young men (30–40 years old) [68,111]. The age-related increase in serum PTH can be attributed, at least in part, to diminishing GFR, but even in healthy elderly men in whom GFR is greater than 70 ml/min, serum PTH is higher than values in young men, albeit within the normal range [120]. The observation that, in aging men, serum PTH concentrations increase but those of 1,25(OH)2D do not suggests that the capacity of PTH to stimulate 1,25(OH)2D production may decrease with advancing age [70,75]. Indeed, Slovik et al. [75] found that intravenous infusion of human PTH(1-34) induced an increase in serum 1,25(OH)2D in healthy young subjects but not in elderly patients with osteoporosis. Tsai et al. [70] reported that the increase in serum 1,25(OH)2D induced by infusion of bovine PTH(1-34) is blunted in elderly postmenopausal women with mild to moderate renal insufficiency when compared to that in healthy young women. These studies, however, were performed in elderly patients with osteoporosis or mild to moderate renal insufficiency, and thus they do not permit separation of the effects of aging from other conditions that may impair the ability of the kidney to respond to PTH. To determine whether aging in normal subjects without osteoporosis or renal insufficiency influences the ability of the kidney to respond to PTH, we administered hPTH(1-34) by intravenous infusion for 24 hr to healthy young and elderly men who were free of conditions known to affect mineral metabolism and in whom the GFR was > 70 ml/min/1.73 m2 [125]. Before administration of PTH, the concentrations of blood ionized calcium and serum 1,25(OH)2D and urinary excretion of calcium and phosphorus were similar in both age groups, but concentrations of serum PTH (+148%), plasma cAMP (+44%), nephrogenous cAMP excretion

(NcAMP) (+56%), and fractional excretion of phosphorus (FEPi) (+44%) were higher in the elderly. With PTH infusion, serum 1,25(OH)2D increased to virtually the same maximum concentration in the young and elderly men (Fig. 4). Urinary cAMP, NcAMP, and FEPi increased, and both the time course and increment were not significantly different between age groups (Fig. 5). TmP/GFR decreased in response to PTH to the same extent in both age groups. These results demonstrate that whereas the time course of the increase in serum 1,25(OH)2D induced by PTH is delayed in the elderly relative to that in the young, maximum renal responsiveness to PTH, in terms of cAMP generation, 1,25(OH)2D production, and phosphaturia, is not impaired in healthy elderly men. These studies raise the question as to why the higher basal serum PTH concentrations in the elderly are not associated with increased serum 1,25(OH)2D concentrations. Such an altered relationship between the serum concentrations of 1,25(OH)2D and those of either PTH or phosphorus may reflect a disorder in the growth hormone (GH) and insulin-like growth factor (IGF) axis in the elderly. Both growth hormone and IGF-I stimulate 1-hydroxylase activity, and an intact

140 hPTH(1-34) Infusion (35U/70kg/hr)

* 120 Serum 1,25(OH)2D (pmol/L)

the relationship was not different between the two groups, but the intercept was lower in the elderly men. These data suggest that the capacity of phosphorus restriction to increase renal production and serum concentration of 1,25(OH)2D is not impaired by aging in healthy men. However, the normal relationship between the serum concentrations of phosphorus and 1,25(OH)2D is altered with advancing age, such that for at any given concentration of serum phosphorus, serum concentrations of 1,25(OH)2D are lower in the elderly.

BERNARD P. HALLORAN AND ANTHONY A. PORTALE

* * * *

100

*

80

0 −4

0 Days

4

8

12 Time (hr)

16

20

24

FIGURE 4 Effect of PTH infusion on the serum concentration

of 1,25(OH)2D in young (39 ± 1, n = 9) (solid line) and elderly (70 ± 1, n = 7) (broken line) men. Values are given for the 2 days before PTH administration and every 4 hr during a 24-hr infusion. The asterisks denote significant differences (p < .05) from baseline (i.e., days 0, −1, −2), using two-way repeated measures ANOVA. From Halloran et al. [125].

831

CHAPTER 50 Vitamin D Metabolism and Aging

Young men NcAMP (nmol/L GFR)

Control

PTH

Elderly men Recovery

Control

PTH

30

Recovery

* ∆ = 12.7 ± 2.0

∆ = 13.3 ± 2.2

*

20

10

0

TmPi (mmol/L GFR)

1.50

Control

PTH

Recovery

Control

PTH

Recovery

1.25 + 1.0

* ∆ = 0.46 ± 0.08

0.75

* ∆ = 0.44 ± 0.06

0.5

FIGURE 5 Effect of PTH infusion and recovery on NcAMP and tubular reabsorptive maximum for phosphorus (TmPi) in young and elderly men. Values represent measurements taken between 0700–0900 hr. Differences are indicated between the control and PTH periods. Using two-way ANOVA, asterisks denote a significant (p < .05) difference from the control period; plus signs denote a significant difference from the control period in young men. From Halloran et al. [125].

GH/IGF-I axis is required for dietary phosphorus restriction to stimulate 1,25(OH)2D production [126–129]. With advancing age, the serum concentrations of GH and IGF-I decrease [130,131], and administration of GH to elderly subjects induces an increase in serum 1,25(OH)2D [132,133]. These observations suggest that the age-related decrease in serum GH and IGF-I may reduce tonic stimulation of the 1α-hydroxylase. If so, the serum concentration of 1,25(OH)2D would be predicted to decrease, resulting in decreased intestinal calcium absorption, hypocalcemia, and stimulation of PTH release. The increase in serum PTH would be expected to stimulate renal synthesis of 1,25(OH)2D, thus restoring tonic stimulation otherwise provided by GH/IGF-I. The mild hyperparathyroidism, consequent hypophosphatemia, and unchanged serum 1,25(OH)2D levels observed in elderly men are consistent with this formulation. Also consistent with this formulation is the observation by Wong et al. that in aging (24 months) [134] and adult (3–4 months) [135] rats,

renal 1α-hydroxylase activity failed to increase with restriction of either dietary calcium or phosphorus, but did increase when dietary restriction was combined with infusion of IGF-I, albeit to levels below those achieved in young (4–6 weeks) rats fed the restricted diets. Whether age-related changes in factors other than GH and IGF-I play a role, remains to be determined.

VI. TISSUE RESPONSIVENESS AND THE ROLE OF VITAMIN D IN THE AGING PROCESS A. Intestinal Calcium Absorption Active intestinal calcium absorption is an excellent marker of tissue responsiveness to vitamin D. Intestinal calcium absorption is reported to decrease with postmaturational aging suggesting that aging may be accompanied by intestinal (and other end organ) resistance to

832

BERNARD P. HALLORAN AND ANTHONY A. PORTALE

1,25(OH)2D [136–144]. In animal studies, evidence strongly supports the contention that calcium absorption decreases with age. Part of the reason for this is the agerelated decrease in circulating 1,25(OH)2D. However, in addition to the contribution associated with the age-related fall in 1,25(OH)2D, most [140,141] but not all [80] investigators find an age-related resistance to the stimulatory effects of 1,25(OH)2D. Wood et al. [140], for example, report that the sensitivity of duodenal calcium transport to the plasma concentration of 1,25(OH)2D is reduced by 54% in aged compared to young rats. These investigators also report that mucosal vitamin D receptor (VDR) levels do not change with age, although other investigators [145–147] find some decrease in VDR number. Binding affinity appears unaffected by age. Other data show that there are also age-related decrements in the intestinal absorptive cell plasma membrane calcium pump [142], calbindin binding protein concentration [140], and ability to translocate protein kinase C to membrane fractions [146] in response to 1,25(OH)2D in aged compared to young animals. In humans, there is near-unanimous agreement that aging is associated with impaired intestinal calcium absorption and reduced sensitivity to 1,25(OH)2D [88,89,139–142]. Eastell et al. [88] measured serum 1,25(OH)2D and intestinal calcium absorption in women 26 to 88 years of age. Calcium absorption did not change with age but serum 1,25(OH)2D gradually increased. They concluded that true calcium absorption in healthy elderly women must be resistant to 1,25(OH)2D. To assess the relationship among calcium absorption, serum 1,25(OH)2D and VDR concentration in the intestinal mucosa, Ebeling et al. [89] studied

44 healthy women between the ages of 20 and 87. Over the age range of the population the serum concentration of 1,25(OH)2D increased by approximately 27%, VDR concentration decreased by 20% (affinity did not change) and intestinal calcium absorption did not change, suggesting that intestinal responsiveness to 1,25(OH)2D declines with age. Despite these observations, Ebeling et al. [148], in a later report, indicate that the responses of serum 1,25(OH)2D and intestinal calcium absorption to restriction of dietary calcium are the same in young (30 ± 1 years) and elderly (74 ± 1 years) women. In a more recent study representing perhaps the most thorough investigation of the age-related changes in intestinal calcium transport in response to 1,25(OH)2D to date, Pattanaungkul et al. [143] measured fractional calcium absorption (FCA), and the serum concentrations of 1,25(OH)2D and the vitamin D binding protein (to calculate the free 1,25(OH)2D index) in young and elderly women in whom they manipulated the circulating level of 1,25(OH)2D over a broad range. The results clearly demonstrate that the absorptive response to a steady-state level of free 1,25(OH)2D (index) is impaired in the elderly (Fig. 6). Vitamin D receptor number is reported to both be decreased [88,89] and remain unchanged [139] with aging. Collectively the animal and human data suggest that the VDR concentration in the intestinal mucosa probably decreases with aging but by a small amount. Nevertheless, there is a marked decrease in intestinal responsiveness to 1,25(OH)2D with age. The mechanisms for this are not clear but appear to involve the calcium transport machinery (Ca-pump, calbindin) of the absorptive cell.

Young 1.2

Elderly 1.2

r = 0.63, p = 0.003

FCA

0.8

FCA

0.8

r = 0.35, p = 0.142

0.4

0

0.4

0

2

4

6

1,25(OH)2D/DBP × 105

FIGURE 6

8

0

0

2

4

6

8

1,25(OH)2D/DBP × 105

Effect of aging on fractional calcium absorption (FCA) in response to the free 1,25(OH)2D index. From Pattanaungkul et al. [143].

833

CHAPTER 50 Vitamin D Metabolism and Aging

B. Bone, Kidney, and Other Tissues Parallel studies in bone comparing responsiveness to 1,25(OH)2D in young and old rodents [147,149,150] and humans [151–155] also show distinct changes with age. Vitamin D receptor concentrations [147] and the number of osteoblasts expressing the VDR [150] both decrease with aging in the rat. Furthermore osteoblasts from old animals are less active and their response to 1,25(OH)2D is altered [149]. Studies with human cells are more extensive. Martinez et al. [154] report lower VDR expression in osteoblasts from older subjects and osteoblasts from old donors are associated with reduced baseline, and impaired osteocalcin [151,152,154,155] and alkaline phosphatase [152,153] response to 1,25(OH)2D (Table IV). In the rat kidney the amount of unoccupied VDR does not appear to change with age [156]. However, binding of the VDR to DNA-cellulose is reduced by nearly 50% in kidney preparations from aged rats. Further, the concentration of the vitamin D–dependent protein, calbindin-D28K, is reported to decrease with advancing age, and the decline has been linked to both lower serum 1,25(OH)2D levels and to impaired responsiveness to 1,25(OH)2D [114]. Since the VDR is found in numerous tissues, perhaps all tissues, and has been shown to stimulate differentiation it is likely that the documented changes in intestinal, bone, and kidney VDR concentration and responsiveness to 1,25(OH)2D reflect a much broader and more generalized change in all tissues. For example, keratinocytes express the VDR and respond to 1,25(OH)2D by altering metabolic activity and differentiation, and mice containing the null allele of VDR (VDR−/−) are more sensitive to chemically induced skin carcinogenesis [157]. Thus, age-related declines in skin VDR levels or cell responsiveness to vitamin D may predispose to neoplasia. Furthermore, hypertension [158], benign prostate hypertrophy and prostate cancer [159], agerelated changes in the sex steroids [160], and calcium TABLE IV Cell Content of Osteocalcin (ng/mg Protein) in Human Osteoblastic Cells after Treatment with 1,25(OH)2D Age (yr) < 50 > 50

Basal 0.7 ± 0.2 0.3 ± 0.2

1,25(OH)2D (10−8 mol/liter) 8.0 ± 6.1a 2.8 ± 1.5a,b

Values are mean ± SEM, n = 6. ap < .05 vs baseline. bp < .05 between 10−8 and 10−6 mol/liter. From Martinez et al. with permission [154].

1,25(OH)2D (10−6 mol/liter) 12.8 ± 7.2a 5.0 ± 2.3a

modulation in brain neurons [161] have all been directly linked to vitamin D. Thus, aging changes vitamin D metabolism, and in turn, alterations in tissue sensitivity to vitamin D are likely to contribute to the aging process.

VII. CONCLUSIONS A complete and accurate description of the effects of aging on vitamin D metabolism cannot yet be made. Many aspects have not been adequately investigated and many of the existing data remain controversial. The following is an attempt to integrate what we think we know combined with a bit of speculation. In the absence of disease and in a free-living population consuming a normal Western diet vitamin D metabolism begins to change around mid-life. Dietary calcium intake and cutaneous production of vitamin D decline, serum growth hormone and IGF-I begin to decrease, and renal function begins to deteriorate. As a consequence of these changes the serum concentration of 25OHD decreases, calcium absorption in the intestine diminishes, and calcium bioavailability to meet serum demands declines. This stimulates PTH secretion [162–164], but because of declining renal function and loss of the normal trophic effects of testosterone, estrogen, and GH/IGF on 1-hydroxylase activity, production of 1,25(OH)2D decreases, remains the same, or increases modestly depending upon the individual. The mild hyperparathyroidism associated with aging stimulates bone turnover. Coupled with the decrease in serum 25OHD and decline in intestinal calcium absorption this shifts the source of calcium for maintenance of the serum pool away from the diet and toward the bone. This shift in mineral balance aggravates the gradual loss of bone associated with aging and contributes to the eventual development of senile osteoporosis. The impact of disease and medication use complicates this enormously. Each individual is different. As we age we become less alike. In addition to the changes in vitamin D bioavailability and metabolism, aging is associated with alterations in tissue responsiveness to vitamin D. These changes not only affect mineral metabolism but also have broad effects on numerous tissues ranging from skin to brain. Thus, aging causes changes in vitamin D metabolism and tissue responsiveness, and these changes in turn are likely to contribute to the aging process. Further studies are clearly needed to better define the effects of aging per se and the confounding effects of disease burden on mineral homeostasis and vitamin D metabolism, and the role of age-related alterations in tissue vitamin D responsiveness to aging.

834

BERNARD P. HALLORAN AND ANTHONY A. PORTALE

References 1. Cerimele D, Celleno L, Serri F 1999 Physiological changes in ageing skin. Brit J Dermatol 122(Suppl 35):13–20. 2. Lavker RM 1979 Structural alterations in exposed and unexposed aged skin. J Invest Dermatol 73:59–66. 3. Goodson WH, Hunt TK 1979 Wound healing and aging. J Invest Dermatol 73:88–91. 4. Daly CH, Odland GF 1979 Age-related changes in the mechanical properties of skin. J Invest Dermatol 73:84–87. 5. Grove GL, Kligman AM 1983 Age-associated changes in human epidermal cell renewal. J Gerontol 38:137–142. 6. Escoffier C, DeRegal J, Rochefort A 1989 Age-related mechanical properties of human skin. J Invest Dermatol 93:353–360. 7. Gilchrest BA, Garmyn M, Yaar M 1994 Aging and photoaging affect gene expression in cultured human keratinocytes. Arch Dermatol 130:82–86. 8. Puschel HU, Chang J, Muller PK, Brinckman J 1995 Attachment of intrinsically and extrinsically aged fibroblasts on collagen and fibronectin. J Photochem Photobiol 27:39–46. 9. Ghadially R, Brown BE, Sequeira-Martin SM, Feingold KR, Elias PM 1995 The aged epidermal permeability barrier. J Clin Invest 95:2281–2290. 10. Tsuchida Y 1993 The effect of aging and arteriosclerosis on human skin blood flow. J Dermatol Sci 5:175–181. 11. MacLaughlin J, Holich MF 1985 Aging decreases the capacity of human skin to produce vitamin D. J Clin Invest 76:1536–1538. 12. Webb AR, Kline L, Holick MF 1988 Influence of season and latitude on the cutaneous synthesis of vitamin D3. J Clin Endocrinol Metab 67:373–378. 13. Holick MF, Matsuoka LY, Wortsman J 1989 Age, vitamin D and solar ultraviolet. Lancet 2:1104–1105. 14. Holick MF 1994 Vitamin D—new horizons for the 21st century. Am J Clin Nutr 60:619–630. 15. Matsuoka LY, Wortsman J, Hanifan N, Holick MF 1988 Chronic sunscreen use decreases circulating concentrations of 25-hydroxvitamin D. Arch Dermatol 124:1802–1804. 16. Gloth FM, Gundberg CM, Hollis BW, Haddad JG, Tobin JD 1995 Vitamin D deficiency in homebound elderly persons. J Am Med Assoc 274:1683–1686. 17. Wortsman J, Masuoka LY, Chen TC, Lu Z, Holick MF 2000 Decreased bioavailability of vitamin D in obesity. Am J Clin Nutr 72:690–693. 18. Lips P 2001 Vitamin D deficiency and secondary hyperparathyroidism in the elderly: consequences for bone loss and fractures, and therapeutic implications. Endo Rev 22:477–501. 19. Recommended Daily Allowances, 10th ed. 1989 Food and Nutrition Board, Commission of Life Sciences, National Research Council, National Academic Press, Washington, DC, pp. 92–98. 20. Holick MF 1998 Vitamin requirements for humans of all ages: New increased requirements for women and men 50 yr and older. Osteop Intern 8:24–29. 21. NHANES III Series 11, No. 245 Dietary Intake of Macronutrients, Micronutrients, and Other Dietary Constituents: United States, 1988–94: http://www.cdc.gov/nchs/data/series/ sr11/sr11 245.pdf 22. Dietary Intake of Ten Key Nutrients for Public Health, 1999–2000. Advance data from vital and health statistics No. 334:http://www.cdc.gov/nchs/data/ad/ad334.pdf 23. Krall EA, Sahyoun N, Tannenbaum S, Dallal GE, Dawson-Hughs B 1989 Effect of vitamin D intake on seasonal

24. 25. 26. 27. 28.

29. 30.

31. 32.

33.

34. 35. 36. 37. 38.

39.

40. 41.

variations in parathyroid hormone secretion in postmenopausal women. N Engl J Med 321:1777–1783. Schock NW 1972 Energy metabolism caloric intake and physical activity of the aging. In: LA Carson (ed) Nutrition in Old Age. Almqvist and Wiksell, Uppsala, Sweden. Elahi VK, Elahi D, Andres R, Tobin JD, Butler MG, Norris AH 1983 A longitudinal study of nutritional intake in men. J Gerontol 38:162–180. Omdahl JL, Garry PJ, Hunsaker LA, Hunt WC, Goodwin JS 1982 Nutritional status in a healthy elderly population: Vitamin D. Am J Clin Nutr 36:1225–1233. Sowers MR, Wallace RB, Hollis BW, Lemke JH 1986 Parameters related to 25OHD levels in a population-based study of women. Am J Clin Nutr 43:621–628. Sharkey JR, Branch LG, Zahoori N, Gueliani C, BushyWhitehead J, Hains PS 2002 Inadequate nutrient intakes among homebound elderly and their correlation with individual characteristics and health-related factors. Am J Clin Nutr 76:1435–1445. Delvin EE, Imbach A, Copti M 1988 Vitamin D nutritional status and related biochemical indices in an autonomous elderly population. Am J Clin Nutr 48:373–378. Payette H, Gray-Donald K 1991 Dietary intake and biochemical indices of nutritional status in an elderly population with estimates of the precision of the 7-day food record. Am J Clin Nutr 54:478–488. Barragry JM, France MW, Corless P, Gupta SP, Switala S, Cohen RD 1978 Intestinal cholecalciferol absorption in elderly and young adults. Clin Sci Mol Med 55:213–220. Harris SS, Dawson-Hughs B, Perrone GA 1999 Plasma 25-hydroxyvitmain D response of young and old men to three weeks of supplementation with 1800 IU/d of vitamin D. Am J Clin Nutr 18:470–474. Clemens TL, Zhou XY, Myles M, Endres D, Lindsay R 1986 Serum vitamin D3 and vitamin D2 concentrations and absorption of vitamin D2 in elderly subjects. J Clin Endocriol Metab 63:656–660. Hollander D, Tarnawski H 1984 Influence of aging on vitamin D absorption and unstirred water layer dimensions in the rat. J Lab Clin Med 103:462–469. Fleming BB, Barrows CH 1982 The influence of aging on intestinal absorption of vitamins A and D by the rat. Exp Gerontol 17:115–120. Harris SS, Dawson-Hughes B 2002 Plasma vitamin D and 25OHD responses of young and old men to supplementation with vitamin D3. J Am Coll Nutr 21:357–362. Aksnes L, Rodland O, Aarskog D 1988 Serum levels of vitamin D3, 25-hydroxyvitamin D3 in elderly and young adults. Bone Miner 3:351–357. Van der Wielen RP, Lowik MR, van den Berg H, de Groot LC, Haller J, Moreiras O, van Staveren WA 1995 Serum vitamin D concentrations among elderly people in Europe. Lancet 346:207–210. McMurtry CT, Young SE, Downs RW, Adler RA 1992 Mild vitamin D deficiency and secondary hyperparathyroidism in nursing home patients receiving adequate dietary vitamin D. J Am Geriat Soc 40:343–347. Weisman Y, Schen RJ, Eisenberg Z, Edelstein S, Harell A 1981 Inadequate status and impaired metabolism of vitamin D in the elderly. Isr J Med Sci 17:19–21. Guillemant J, Oberlin F, Bourgeois P, Guillemant S 1984 Age-related effect of a single oral dose of calcium on parathyroid function: relationship with vitamin status. Am J Clin Nutr 60:403–407.

835

CHAPTER 50 Vitamin D Metabolism and Aging

42. Need AG, Morris HA, Horowitz M, Nordin C 1993 Effects of skin thickness, age, body fat and sunlight on serum 25-hydroxyvitamin D. Am J Clin Nutr 58:882–885. 43. Quesada JM, Coopmans W, Ruiz B, Aljama P, Jans I, Bouillon R 1992 Influence of vitamin D on parathyroid function in the elderly. J Clin Endocriol Metab 75:494–501. 44. McKenna MJ 1992 Differences in vitamin D status between countries in young adults and the elderly. Am J Med 93:69–77. 45. Perry HM, Horowitz M, Morley JE, Patrick P, Vellas B, Baumgartner R, Garry R 1999 Longitudinal changes in serum 25OHD in older people. Metabolism 48:1028–1032. 46. Fardellone P, Sebert JL, Garabedian M, Bellony R, Maamer M, Agbomson F, Brasier M 1995 Prevalence and biological consequences of vitamin D deficiency in elderly institutionalized subjects. Rev Rhum Engl Ed 62:576–581. 47. Brazier M, Kamel S, Maamer M, Agbomson F, Elesper I, Garabedian M, Desmet G, Sebert JL 1995 Markers of bone remodeling in the elderly subject: Effects of vitamin D insufficiency and its correction. J Bone Miner Res 10:1753–1761. 48. Ooms ME, Lipps P, Roos JC, van der Vvijgh WJ, Popp-Snijders C, Bezemer PD, Bouter LM 1995 Vitamin D status and sex hormone binding globulin: Determinants of bone turnover and bone mineral density in elderly women. J Bone Miner Res 10:1177–1184. 49. Guillemant S, Guillemant J, Feteanu D, Sebag-Lanoe R 1989 Effect of vitamin D3 administration on serum 25-hydroxyvitamin D3, 1,25-dihydroxyvitamin D and osteocalcin in vitamin D deficient elderly people. J Ster Biochem 33: 1155–1159. 50. Sakota O, Masud T, San P, Hosking DJ 1999 Vitamin D insufficiency increases bone turnover markers and enhances bone loss at the hip in patients with established vertebral osteoporosis. Clin Endocrinol 51:217–221. 51. Melin AL, Wilske J, Ringertz H, Saaf M 1999 Vitamin D status, parathyroid function and femoral bone density in an elderly Swedish population living at home. Aging 11:200. 52. Rapuri PB, Kinyamu HK, Gallagher JC, Haynatzka V 2002 Seasonal changes in calcitropic hormones, bone markers and bone mineral density in elderly women. J Clin Endocrinol Metab 87:2024–2032. 53. Stone K, Bauer DC, Black DM, Sklarin P, Ensrud KE, Cummings SR 1998 Hormonal predictors of bone loss in elderly women: a prospective study. J Bone Miner Res 13:1167–1174. 54. Myer HE, Hendriksen C, Falch JA, Pedersen JI, Tverdal A 1996 Risk factors for hip fracture in a high incidence area. Osteop Int 5:239–246. 55. Ooms ME, Roos JC, Bezemer PD, van der Vvijgh WJ, Bouter L, Lipps P 1995 Prevention of bone loss by vitamin D supplementation in elderly women: a randomized doubleblind trial. J Clin Endocrinol Metab 80:1052–1058. 56. Himmelstein S, Clemens TL, Rubin A, Lindsay R 1990 Vitamin D supplementation in elderly nursing home residents increases 25OHD but not 1,25(OH)2D. Am J Clin Nutr 52:701–706. 57. Lips P, Wiersinga A, van Ginkel FC 1988 The effect of vitamin D supplementation on vitamin D status and pararthyroid function in elderly subjects. J Clin Endocrinol Metab 67:644–650. 58. Dawson-Hughes B, Dallal GE, Krall EA, Harris S, Sokoll LJ, Falconor G 1991 Effect of vitamin D supplementation on wintertime overall bone loss in healthy postmenopausal women. Ann Intern Med 115:505–512. 59. Dawson-Hughes B, Harris SS, Krall EA, Dallal GE, Falconer G, Green CL 1995 Rates of bone loss in postmenopausal

60. 61.

62.

63.

64. 65. 66. 67. 68.

69.

70. 71. 72.

73. 74. 75. 76. 77.

78.

women randomly assigned to one of two doses of vitamin D. Am J Clin Nutr 61:1140–1145. Barger-Lux MJ, Heaney RP, Lanspa SJ, Healy JC, DeLuca HF 1995 An investigation of sources of variation in calcium absorption efficiency. J Clin Endocrinol Metab 80:406–411. Orwoll ES, Oviatt SK, McClung MR, Deftos LJ, Sexton G 1990 The rate of bone mineral loss in normal men and the effects of calcium and cholecalciferol supplementation. Ann Intern Med 112:29–34. Bouillon R, Auwerx J, Dekeyser L, Fevery J, Lisssns W, De Moor P 1984 Serum vitamin D metabolites and their binding protein in patients with liver cirrhosis. J Clin Endocrinol Metab 59:86–89. Bikle DD, Gee E, Halloran B, Haddad JG 1984 Free 1,25-dihydroxyvitamin D levels in serum from normal subjects, pregnant subjects and subjects with liver disease. J Clin Invest 74:1966–1977. Manolagas S, Culler FL, Howard JE, Brickman AS, Deftos LJ 1983 The cytoreceptor assay for 1,25(OH)2D and its application to clinical studies. J Clin Endocrinol Metab 56:751–760. Epstein S, Bryce G, Hinman JW 1986 The influence of age on bone mineral regulating hormones. Bone 7:421–425. Orwoll ES, Meier D 1986 Alterations in calcium, vitamin D and PTH physiology in normal men with aging. J Clin Endocrinol Metab 63:1262–1269. Sherman SS, Hollis BW, Tobin JD 1990 Vitamin D status and related parameters in a healthy population: the effects of age, sex and season. J Clin Endocrinol Metab 71:405–413. Halloran BP, Portale AA, Lonergan ET, Morris RC 1990 Production and metabolic clearance of 1,25(OH)2D in men: effect of advancing age. J Clin Endocrinol Metab 70: 318–323. Gallagher C, Riggs BL, Eisman JA, Hamstra A, Arnaud SB, DeLuca HF 1979 Intestinal calcium absorption and vitamin D metabolites in normal subjects and osteoporotic patients. J Clin Invest 64:729–736. Tsai KS, Heath H, Kumar R, Riggs BL 1984 Impaired vitamin D metabolism with aging in women. 73:1668–1672. Buchanan JR, Myers CA, Greer RB 1988 Effect of declining renal function on bone density in aging women. Calcif Tissue Int 43:1–6. Prince RL, Dick I, Garcia WP, Retallack RW 1990 The effects of the menopause on calcitriol and PTH: Responses to a low dietary calcium stress test. J Clin Endocrinol Metab 70:1119–1123. Fujisawa Y, Kida K, Matsuda H 1984 Role of change in vitamin D metabolism with age on calcium and phosphorus metabolism in normal human subjects. 59:719–726. Dokoh S, Morita R, Fukunaga M, Yamamoto I, Torizuka K 1978 Competitive protein binding assay for 1,25(OH)2D in human plasma. Endocrinol Jpn 25:431–436. Slovik SM, Adams JS, Neer RM, Holick MF, Potts JT 1981 Deficient production of 1,25(OH)2D in elderly osteoporotic patients. N Engl J Med 305:372–374. Lund B. Sorensen OH, Lund B, Agner E 1982 Serum 1,25(OH)2D in normal subjects and in patients with postmenopausal osteopenia. Horm Metab Res 14:271–274. Armbrecht HJ, Strong R, Boltz M, Rocco D, Wood WG, Richardson A 1988 A modulation of age-related changes in serum 1,25(OH)2D and PTH by dietary restriction in F344 rats. J Nutr 118:1360–1365. Fox J, Mathew B 1991 Heterogeneous response to PTH in aging rats: Evidence for skeletal PTH resistance. Am J Physiol 260:E933–E937.

836 79. Gray RW, Gambert SR 1982 Effect of age on plasma 1,25(OH)2D in the rat. Age 5:54–56. 80. Liang CT, Barns J, Takamoto S, Sacktor B 1989 Effect of age on calcium uptake in isolated duodenal cells: role of 1,25(OH)2D. Endocriology 124:2830–2836. 81. Armbrecht HJ, Forte L, Halloran BP 1989 Effect of age and dietary calcium on renal 25(OH)D metabolism, serum 1,25(OH)2D and PTH. Am J Physiol 246:E266–E270. 82. Wada L, Daly R, Kern D, Halloran B 1992 Kinetics of 1,25-dihydroxyvitamin D metabolism in the aging rat. Am J Physiol 262:E906–E910. 83. Theuns HM, van der Vijgh WJF, Hackeng WHL, Bekker H, Barto R, Lips P, Roholl PJM, Netelenbos JC, Knook DL 1991 Parathyroid hormone and vitamin D metabolites throughout life of two inbred rat strains. In: AW Norman, R Bouillon, M Thomasset (eds) Vitamin D, de Gruyter, New York, pp. 769–770. 84. Schock NW 1945 Inulin, diodrast and urea clearance studies on aged human subjects. Fed Proc 4:65–71. 85. Rowe JW, Anders R, Tobin JR, Norris AH, Schock NW. The effect of age on creatinine clearance in man: a cross-sectional and longitudinal study. J Geront 31:155–163. 86. Lindeman RD, Tobin J, Schock NW 1985 Longitudinal studies on the rate of decline in renal function with age. J Am Geriatr Soc 33:278–285. 87. Marcus R, Madvig P, Young G 1984 Age related changes in PTH and PTH action in normal humans. J Clin Endocrinol Metab 58:223–230. 88. Eastell R, Yergey AL, Vierira NE, Cedel SE, Kumar R, Riggs BL 1991 Interrelationship among vitamin D metabolism, true calcium absorption, parathyroid function and age in women: Evidence of an age-related intestinal resistance to 1,25(OH)2D. J Bone Miner Res 6:125–132. 89. Ebeling PR, Sandgren ME, DiMagno EP, Lane AW, DeLuca HF, Riggs BL 1992 Evidence of an age-related decrease in intestinal responsiveness to vitmain D: Relationship between serum 1,25-dihydroxyvitamin D and intestinal vitamin D receptor concentrations in normal women. J Clin Endocrinol Metab 75:176–182. 90. Hartwell D, Riis BJ, Christiansen C 1990 Changes in vitamin D metabolism during natural and medical menopause. J Clin Endocrinol Metab 71:127–132. 91. Falch JA, Oftebro H, Haug E 1987 Early postmenopausal bone loss is not associated with a decrease in circulating 25OHD, 1,25(OH)2D or vitamin D binding protein. J Clin Endocrinol Metab 64:836–841. 92. Bikle DD, Halloran BP, Harris ST, Portale AA 1992 Progestin antagonism of estrogen stimulated 1,25(OH)2D levels. J Clin Endocrinol Metab 75:519–524. 93. Dick IM, Prince RL, Kelly JJ, Ho KK 1995 Oestrogen effects on calcitriol levels in postmenopausal women: a comparison of oral versus transdermal administration. Clin Endocrinol 43:219–224. 94. Jongen MJ, Bishop JE, Cade C, Norman AW 1987 Effect of dietary calcium, phosphate and vitamin D deprivation on the pharmacokinetics of 1,25(OH)2D in the rat. Horm Metab Res 19:481–485. 95. Halloran BP, Castro ME 1989 Vitamin D kinetics in vivo: Effect of 1,25(OH)2D administration. Am J Physiol 256:E686–E691. 96. Fox J, Ross R 1985 Effect of low phosphorus and low calcium diets on the production and metabolic clearance rates of 1,25(OH)2D in pigs. J Endocrinol 105:169–173. 97. Portale AA, Halloran BP, Murphy MM, Morris RC Jr 1986 Oral intake of phosphorus can determine the serum

BERNARD P. HALLORAN AND ANTHONY A. PORTALE

98. 99.

100. 101.

102. 103. 104.

105.

106.

107.

108. 109.

110.

111. 112. 113.

114.

concentration of 1,25(OH)D2 by determining its production rate in humans. J Clin Invest 77:7–12. Young EW, Hsu CH, Patel S, Simpson RU, Komanicky P 1987 Metabolic degradation and synthesis of calcitriol in spontaneously hypertensive rat. Am J Physiol 252:E778–E782. Seeman E, Kumar R, Hunder GG, Scott M, Heath H, Riggs BL 1980 Production, degradation and circulating levels of 1,25(OH)2D in health and in chronic glucocorticoid excess. J Clin Invest 66:664–669. Hsu CH, Patel S, Young EW, Simpson RU 1987 Production and degradation of calcitriol in renal failure rats. Am J Physiol 253:F1015–F1019. Gray A, Feldman HA, McKinlay JB, Longcope C 1991 Age, disease and changing sex hormone levels in middle-aged men: results of the Massachusettes male aging study. J Clin Endocrinol Metab 73:1016–1025. Vermeulen A 1991 Androgens in the aging male. J Clin Endocrinol Metab 73:221–224. Hagenfeldt Y, Linde K, Sjoberg HE, Zumkeller W, Arver S 1992 Testosterone increases serum 1,25(OH)2D and IGF-I in hypogonadal men. Intern J Androl 15:93–102. Boucher A, D’Amour P, Hamel L 1989 Estrogen replacement decreases the set-point of PTH stimulation by calcium in normal postmenopausal women. J Clin Endocrinol Metab 68:831–836. Scharla SH, Minne HW, Waibel TS 1990 Bone mass reduction after estrogen deprivation by long-acting gonadotropinreleasing hormone agonists and its relation to pretreatment serum concentrations of 1,25(OH)2D. J Clin Endocrinol Metab 70:1055–1061. Marcus R, Villa ML, Cheema M, Cheema C, Newhall K, Holloway L 1992 Effects of conjugated estrogen on the calcitriol response to parathyroid hormone in postmenopausal women. J Clin Endocrin Metab 74:413–418. Packer E, Holloway L, Newhall K, Kanwar G, Butterfield K, Marcus R 1990 Effects of estrogen on daylong circulating calcium,phosphorus,1,25-dihydroxyvitamin D and parathyroid hormone in postmenopausal women. J Bone Miner Res 5:877–884. Stock JL, Coderre JA, Mallette LE 1985 Effects of a short course of estrogen on mineral metabolism in postmenopausal women. J Clin Endocrinol Metab 61:595–600. Gallagher C, Riggs BL, DeLuca HF 1980 Effect of estrogen on calcium absorption and vitamin D metabolites in postmenopausal osteoporosis. J Clin Endocrinol Metab 51: 1359–1364. Selby PL, Peacock M, Barkworth SA, Brown WB, Taylor GA 1985 Early effects of ethinyloestradiol and norethisterone treatment in postmenopausal women. Clin Sci 69:265–271. Minisola S, Pacitti M, Scarda A 1993 Serum ionized calcium, parathyroid hormone and related variables: Effect of age and sex. J Bone Miner 23:183–193. Wiske PS, Epstein S, Bell NH, Queener SF, Edmondson J, Johnston CC 1979 Increases in immunoreactive PTH with age. N Engl J Med 300:1419–1421. Endres DB, Morgan CH, Garry PJ, Omdahl JL 1987 Agerelated changes in serum immunoreactive parathyroid hormone and its biological action in healthy men and women. J Clin Endocrinol Metab 65:724–731. Chen ML, Boltz M, Christakos S, Armbrecht HJ 1992 Agerelated alterations in calbindin-D28K induction by 1,25(OH)2D in primary cultures of rat renal tubule cells. Endocrinology 130:3295–3300.

CHAPTER 50 Vitamin D Metabolism and Aging

115. Hulter HN, Halloran BP, Toto RD, Peterson JC 1985 Long term control of calcitriol concentration in dog and man: The dominant role of plasma calcium concentration in experimental hyperparathyroidism. J Clin Invest 76:695–702. 116. Armbrecht HJ, Hodam TL, Boltz MA, Kumar VB 1999 Capacity of a low calcium diet to induce the renal vitamin D 1α-hydroxylase is decreased in adult rats. Biochem Biophys Res Commun 18:731–734. 117. Portale AA, Halloran BP, Morris RC Jr 1987 Dietary intake of phosphorus modulates the circadian rhythm in serum concentration of phosphorus: Implications for the renal production of 1,25-dihydroxyvitamin D. J Clin Invest 8:1147–1154. 118. Portale AA, Halloran BP, Morris RC Jr 1989 Physiologic regulation of the serum concentration of 1,25-dihydroxyvitamin D by phosphorus in normal men. J Clin Invest 83: 1494–1499. 119. Villa ML, Packer E, Cheema M, Holloway L, Marcus R 1991 Effects of aluminum hydroxide on the parathyroid-vitamin D axis of postmenopausal women. J Clin Endocrinol Metab 73:1256–1261. 120. Portale AA, Halloran BP, Morris RC, Lonergan ET 1996 Effect of aging on the metabolism of phosphorus and 1,25-dihydroxyvitamin D in healthy men. Am J Physiol 270:E483–E490. 121. Portale AA, Lonergan ET, Halloran BP 1997 Aging alters calcium regulated parathyroid hormone release in healthy men. Am J Physiol 272:E139–E146. 122. Young G, Marcus R, Minkoff JR, Kim LY, Segre GV 1987 Age-related rise in parathyroid hormone in man: the use of intact and midmolecule antisera to distinguish hormone secretion from retention. J Bone Miner Res 2:367–373. 123. Forero MS, Klein RF, Nissenson RA, Nelson K, Heath III H, Arnaud CD, Riggs BL 1987 Effect of age on circulating immunoreactive and bioactive parathyroid hormone levels in women. J Bone Miner Res 2:363–368. 124. Ledger GA, Burritt MF, Kao PC, O’Fallon WM, Riggs BL, Khosla S 1994 Abnormalities of PTH secretion in elderly women that are reversible by short term therapy with 1,25(OH)2D. J Clin Endocrinol Metab 79:211–216. 125. Halloran BP, Lonergan ET, Portale AA 1996 Aging and renal responsiveness to parathyroid hormone in healthy men. J Clin Endocrinol Metab 81:2192–2197. 126. Gray RW, Garthwaite TL 1985 Activation of renal 1,25(OH)2D synthesis by phosphate deprivation: evidence for a role for growth hormone. Endocrinology 116:189–193. 127. Gray RW 1987 Evidence that somatomedians mediate the effect of hypophosphatemia to increase serum 1,25(OH)2D levels in rats. Endocrinology 121:504–509. 128. Halloran BP, Spencer EM 1988 Dietary phosphorus and 1,25-dihydroxyvitamin D metabolism: influence of insulinlike growth factor I. Endocrinology 123:1225–1230. 129. Menaa C, Vrtovsnik F, Friedlander G, Corvol M, Garabedian M 1995 Insulin-like growth factor I, a unique calcium-dependent stimulator of 1,25-dihydroxyvitamin D production. J Biol Chem 270:25461–25467. 130. Florini JR, Prinz PN, Vitiello MV, Hintz RL 1985 Somatomedin C levels in healthy young and old men: Relationship to peak and 24-hour integrated levels of growth hormone. J Gerontol 40:2–7. 131. Rudman D 1985 Growth hormone, body composition and aging. J Am Geriatr Soc 33:800–806. 132. Marcus R, Butterfield G, Holloway L, Gilland L, Baylink D, Hintz R, Sherman B 1990 Effects of short term administration of recombinant human growth hormone to elderly people. J Clin Endocrinol Metab 70:519–523.

837 133. Lieberman SA, Holloway L, Marcus R, Hoffman AR 1999 Interactions of growth hormone and parathyroid hormone in renal phosphate, calcium and calcitriol metabolism and bone remodeling in postmenopausal women. J Bone Miner Res:1723–1728. 134. Wong MS, Tembe VA, Favus MJ 2002 Insulin-like growth factor-I stimulates renal 1,25-dihydroxycholecalciferol synthesis in old rats fed a low calcium diet. J Nutr 130:1147–1152. 135. Wong MS, Sriussadaporn S, Tembe VA, Favus MJ 1997 Insulin-like growth factor I increases renal 1,25(OH)2D3 biosynthesis during low-P diet in adult rats. Am J Physiol 272:F698–F703. 136. Francis RM, Peacock M, Taylor GA, Storer JH, Nordin BEC 1984 Calcium malabsorption in elderly women with vertebral fractures: evidence for resistance to the actions of vitamin D metabolites on the bowl. Clin Sci 66:103–107. 137. Avioli LV, McDonald JE, Lee SW 1965 The influence of age on the intestinal absorption of Ca in women and its relation to Ca absorption in postmenopausal osteoporosis. J Clin Invest 44:1960–1967. 138. Bullammore JR, Gallagher JC, Wilkinson JR, Nordin BEC 1970 Effect of age on calcium absorption. Lancet 2:535–537. 139. Heaney RP, Reckler RR, Stegman MR, Moy AJ 1989 Calcium absorption in women: relationships to calcium intake, estrogen status, and age. J Bone Miner Res 4:469–475. 140. Kinyamu HK, Gallagher JC, Prahl JM, Deluca HF, Petranick KM, Lanspa SJ 1997 Association between intestinal vitamin D receptor, calcium absorption, and serum 1,25dihydroxyvitamin D in normal young and elderly women. J Bone Miner Res 12:922–928. 141. Wood RJ, Fleet JC, Cashman K, Bruns ME, Deluca HF 1998 Intestinal calcium absorption in the aged rat: Evidence of intestinal resistance to 1,25(OH)2D. Endocrinology 139: 3843–3848. 142. Armbrecth HJ, Boltz MA, Kumar VB 1999 Intestinal plasma membrane calcium pump protein and its induction by 1,25(OH)2D decrease with age. Am J Physiol 277:G41–G47. 143. Pattanaungkul S, Riggs BL, Yergey AL, Vieira NE, O’Fallon WM, Khosla S 2000 Relationship of intestinal calcium absorption to 1,25-dihydroxyvitamin D levels in young versus elderly women: Evidence for age-related intestinal resistance to 1,25(OH)2D action. J Clin Endocrinol Metab 85:4023–4027. 144. Balogh G, Boland R, de Boland AR 2000 1,25(OH)2D affects the subcellular distribution of protein kinase C isoenzymes in rat duodenum: influence of aging. J Cell Biochem 79: 686–694. 145. Takamoto S, Seino Y, Sacktor B, Liang CT 1990 Effect of age on duodenal 1,25-dihydroxyvitamin D receptors in Wistar rats. Biochim Biophys Acta 1034:22–28. 146. Liang CT, Barns J, Imanaka S, DeLuca HF 1994 Alterations in mRNA expression of duodenal 1,25-dihydroxyvitamin D receptor and vitamin D dependent calcium binding protein in aged Wistar rats. Exp Gerontol 29:179–186. 147. Horst RL, Goff JP, Reinhardt TA 1990 Advancing age results in the reduction of intestinal and bone 1,25-dihydroxyvitamin D receptor. Endocrinology 126:1053–1057. 148. Ebeling PR, Yergey AL, Vierira NE, Burritt MF, O’Fallon WM, Kumar R, Riggs BL 1994 Influence of age on effects of endogenous 1,25(OH)2D on calcium absorption in normal women. Calcif Tissue Res 55:330–334. 149. Wang X, Schwartz Z, Yaffe P, Ornoy A 1999 The expression of TGF-beta and interleukin-1beta mRNA, and the response to 1,25(OH)2D′17 beta-estradiol and testosterone is age

838

150.

151. 152.

153.

154.

155. 156.

BERNARD P. HALLORAN AND ANTHONY A. PORTALE

dependent in primary cultures of mouse-derived osteoblasts in vitro. Endocrine 11:13–22. Duque G, Abdaimi KE, Macoritto M, Miller MM, Kremer R 2002 Estrogens regulate expression and response of 1,25(OH)2D receptors in bone cells: Changes with aging and hormone deprivation. Biochem Biophys Res Commun 299:446–454. Battmann A, Battmann A, Jungt G, Schulz A 1997 Endosteal human bone cells show age-related activity in vitro. Exp Clin Endocrinol Diabetes 105:98–102. Martinez ME, Medina S, Sanchez M, Del Campo MT, Esbrit P, Rodrigo A, Sanchez-Cabezudo MJ, Moreno I, Garces MV, Munuera L 1999 Influence of skeletal site of origin and donor age on 1,25(OH)2D induced response of various osteoblastic markers in human osteoblastic cells. Bone 24:203–209. Katzberg S, Lieberherr M, Ornoy A, Klein BY, Hendel D, Somjen D 1999 Isolation and hormonal responsiveness of primary cultures of human bone-derived cells: Gender and age differences. Bone 25:667–673. Martinez P, Moreno I, De Miguel F, Vila V, Esbrit P, Martinez ME 2001 Changes in osteocalcin response to 1,25(OH)2D stimulation and basal vitamin D receptor expression in human osteoblastic cells according to donor age and skeletal origin. Bone 29:35–41. Klein-Nuland J, Sterck JG, Semeins CM, Lips P, Joldersma M, Baart JA, Burger EH 2002 Donor age and meachnosensitivity of human bone cells. Osteopor Intern 13:137–146. Koszewski NJ, Reinhardt TA, Beitz DC, Horst RL 1990 Developmental changes in rat kidney 1,25(OH)2D receptor. Biochem Biophys Res Commun 170(1):65–72.

157. Zinser GM, Sundberg JP, Welsh J 2002 Vitamin D receptor ablation sensitizes skin to chemically induced tumorigenesis. Carcinogenesis 23:2103–2109. 158. Pfeifer M, Bergerow B, Minne HW, Nachtigall D, Hansen C 2001 Effects of a short term vitamin D and calcium supplementation on blood pressure and PTH levels in elderly women. J Clin Endocrinol Metab 86:1633–1637. 159. Crescioli C, Maggie M, Vannelli GB, Luconi M, Salerno R, Barni T, Gulisano M, Forti G, Serio M 2000 Effect of a vitamin D analogue on keratinocyte growth factor-induced cell proliferation in benign prostate hyperplasia. J Clin Endocrinol Metab 85:2576–2583. 160. Kinuta K, Tanaka H, Moriwake T, Aya K, Kato S, Seino Y 2000 Vitamin D is an important factor in estrogen biosynthesis of both female and male gonads. Endocrinology 141: 1317–1324. 161. Brewer LD, Thibault V, Chen KC, Langub MC, Landfield PW, Porter NM 2001 Vitamin D hormone confers neuro-protection in parallel with down regulation of L-type calcium channel expression in hippocampal neurons. J Neurosci 21:98–108. 162. Dawson-Huges B, Harris SS, Dallal GE 1997 Plasma calcidiol, season and serum PTH concentration in healthy elderly men and women. Am J Clin Nutr 65:67–71. 163. Lips P, Duong T, Oleksik A, Black D, Cummings S, Cox D, Nickelsen T 2001 A global study of vitamin D status and parathyroid function in postmenopausal women with osteoporosis: Baseline data from the multiple raloxifene evaluation clinical trial. J Clin Endocrinol Metab 86:1212–1221. 164. Need AG, Horowitz M, Morris HA, Nordin BC 2000 Vitamin D status: Effects on PTH and 1,25(OH)2D in postmenopausal women. Am J Clin Nutr 71:1577–1581.

CHAPTER 51

Vitamin D Metabolism in Pregnancy and Lactation HEIDI J. KALKWARF Division of General and Community Pediatrics, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio

BONNY L. SPECKER Martin Program in Human Nutrition, South Dakota State University, Brookings, South Dakota

I. Introduction II. Adaptations in Vitamin D and Calcium Metabolism during Pregnancy III. Effects of Low Maternal Vitamin D and Calcium Intake during Pregnancy on the Fetus and Neonate

IV. Adaptations in Vitamin D and Calcium Metabolism during Lactation and after Weaning V. Effects of Low Maternal Vitamin D and Calcium Intakes on Breast Milk Vitamin D and Calcium Concentrations VI. Conclusions References

I. INTRODUCTION

but 250 mg/d during the third trimester [1]. Maternal serum calcium concentrations decrease in the first half of pregnancy, reaching a nadir at mid-gestation due to plasma volume expansion and decreased albumin concentrations [2,3]. Serum concentrations of ionized calcium or calcium adjusted for albumin concentrations show less fluctuation and remain relatively stable throughout pregnancy [2,4,5]. Serum concentrations of intact parathyroid hormone (PTH) have been reported to decrease [2,4,6] or not change [3] over the course of pregnancy. In contrast, circulating concentrations of the active form of vitamin D, 1,25-dihydroxyvitamin D (1,25(OH)2D), are increased during pregnancy. By the second trimester serum concentrations of 1,25(OH)2D increase by 50–100% over pre-pregnant values, and in the third trimester they increase by 100% [3,7] (Fig. 1). The signal to increase 1,25(OH)2D synthesis is not clear, as PTH concentrations are not elevated. Vitamin D binding protein (DBP) concentrations increase in pregnancy possibly due to increased concentrations of estrogen. Although some of the increase in serum 1,25(OH)2D concentrations may be due to the increase in the amount bound to its binding protein, the amount of free 1,25(OH)2D is still elevated [3,8]. Some of the circulating 1,25(OH)2D may be of extrarenal origin as the decidua has been shown to synthesize 1,25(OH)2D [9] (see Chapter 79). Consistent with this is the fact that maternal 1,25(OH)2D concentrations rapidly decrease within a few days after delivery [10]. The increase in 1,25(OH)2D concentrations during pregnancy is accompanied by an increase in intestinal calcium absorption (Fig. 1). Fractional calcium absorption increases by 50–56% over prepregnant

Significant changes in maternal calcium metabolism occur during pregnancy, lactation, and after weaning to provide the calcium needed for fetal bone mineral accretion, for the synthesis of breast milk, and for the restoration of the maternal skeleton. Multiple factors are involved in regulating these processes so that maternal blood calcium concentrations are maintained within a narrow range despite the large changes in calcium fluxes that occur. Primary strategies include changes in the efficiency of absorption of calcium from the intestinal tract, alterations in renal calcium reabsorption and thus urinary calcium excretion, and the flux of calcium in and out of bone. Reproductive hormones have important effects on calcium homeostasis and bone metabolism, and during pregnancy and lactation their actions work in concert with the vitamin D endocrine system to ensure that calcium needs are met for fetal bone mineral accretion, for breast milk production, and to maintain circulating maternal calcium concentrations.

II. ADAPTATIONS IN VITAMIN D AND CALCIUM METABOLISM DURING PREGNANCY A. Vitamin D and Calcium Metabolism Approximately 25 to 30 g of calcium is transferred to the fetal skeleton by the end of pregnancy, the most of which is transferred during the last trimester. The fetus accumulates 2 to 3 mg/d during the first trimester, VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

840 Intestinal calcium absorption 1,25(OH)2D

180

65

140

50

100

35

60

1,25(OH)2D (pmol/liter)

% Calcium absorption

80

HEIDI J. KALKWARF AND BONNY L. SPECKER

20

20 Prepregnant 1st trimester 2nd trimester 3rd trimester

FIGURE 1 Intestinal calcium absorption and serum 1,25(OH)2D concentrations before and during pregnancy. Data from Ritchie et al. [3].

levels in the second trimester and by 54–62% in the third trimester [3,7]. Thus increased maternal intestinal absorption of calcium is an important physiologic adaptation to secure sufficient amounts of this mineral for the fetus. Despite the increased need for calcium, urinary calcium excretion increases by 40–50% over the course of pregnancy. This is most likely due to the marked increase in glomerular filtration rate and increased absorptive load [2,3,11]. Several studies report increased concentrations of biochemical markers of bone turnover during pregnancy. Serum concentrations of biochemical markers of bone formation, namely bone specific alkaline phosphatase and the propeptide of type 1 collagen (PICP), are elevated in the third trimester with a steep peak in the last month of pregnancy [2,6]. It is not clear whether there are changes in the first two trimesters of pregnancy as an increase, decrease, or no change in the concentration of these bone formation markers have been reported [2,6]. Osteocalcin concentrations have been found to decrease [6], decrease then increase [3], or not change during pregnancy [10]. Markers of bone resorption, namely the breakdown products of collagen such as pyridinoline, deoxypyridinoline, and NTx, increase throughout pregnancy reaching a peak at the end of pregnancy [2,3,6]. Although it appears that there is a dissociation of bone resorption and bone formation in the first two trimesters with elevated bone resorption predominating, it is difficult to predict whether there is a net loss of maternal bone during pregnancy using biochemical markers alone. Other factors confound the interpretation of biochemical markers of bone turnover during pregnancy such as whether they are of maternal, placental, or fetal origin, and the effects of pregnancy on the metabolic clearance of these proteins by the liver and kidney. It also has been suggested that

some of the increase in bone turnover markers is due to increased turnover of soft tissue collagen of the uterus and skin [6,12]. Increases in circulating concentrations of insulin-like growth factor-1 (IGF-1) and placental lactogen have been suggested as the possible mechanisms behind increased bone turnover during pregnancy [2,6]. Increases in IGF-1 concentrations precede the increase in bone formation markers, and IGF-1 concentrations correlated more strongly with markers of bone formation than bone resorption. A recent study investigated changes in calcium-regulating hormones and osteoprotegerin (OPG) during pregnancy and found that maternal serum OPG concentrations steadily increased with gestational age [10]. Receptor activator of nuclear factor-κ B ligand (RANKL) is important in osteoclast differentiation [13], and OPG acts as a decoy receptor for RANKL, thereby preventing the differentiation of osteoclast precursors into mature osteoclasts and decreasing bone resorption. The authors speculated that higher OPG concentrations during pregnancy, possibly of placental origin, might play a role in the control of bone metabolism throughout gestation.

B. Changes in Bone Mineral Content and Density during Pregnancy Few studies have included bone mineral density measurements during pregnancy because of the potential risks to the fetus associated with radiation exposure. There is conflicting evidence as to whether there is a net change in bone density during pregnancy. Several different approaches have been used to evaluate the impact of pregnancy on maternal bone. Some longitudinal studies have measured bone density by dual energy X-ray absorptiometry (DXA) before conception and shortly after delivery and have found no significant loss of bone density [3,7,14,15]. However, other studies report losses of 2 to 2.6% at the ultradistal radius [16,17], 2 to 4% at the spine [2,6,18], and 2.4 to 3.6% at the hip [2,19]. In some of these studies bone density was measured as far as 6 weeks postpartum, and significant losses of bone may occur within the immediate postpartum period thereby making it hard to interpret these results. Naylor and co-workers found that the changes in bone density during pregnancy varied according to skeletal site. Bone density at trabecular-rich sites (pelvis and spine) decreased by 3 to 4%, whereas bone density at cortical sites (arms and legs) increased by 2% [6]. Many investigators have investigated changes in bone density by use of ultrasound as it does not involve

CHAPTER 51 Vitamin D Metabolism in Pregnancy and Lactation

radiation exposure. Speed of sound (SOS) and bone ultrasound attenuation (BUA) are strongly correlated with bone density measured at the same skeletal site. Longitudinal studies over the course of pregnancy have documented a decrease in SOS and BUA measured at the os calcis or phalanges in the latter half of pregnancy, particularly in the third trimester [20–23]. These data are consistent with a loss of maternal bone mineral toward the end of pregnancy when the fetus is accreting bone mineral most rapidly. Longitudinal studies of bone density in women who do not lactate after delivery show that bone density at the spine and hip increase by about 2% in the first year postpartum [24–26]. It is possible that this increase in bone may compensate for bone lost during pregnancy. If so, this may explain why studies comparing bone density of women with different pregnancy histories have found no differences in bone density measured many years later [27–30]. Henderson et al. found that even grand multiparous women having six or more pregnancies and long lactations did not have lower bone density of the lumbar spine, femoral neck, or mid-radius than nulliparous women later in life [31]. In summary, several adaptations in the maternal calcium economy occur in order to provide sufficient calcium for fetal bone mineral accretion (Fig. 2). The primary adaptive strategy is an increase in intestinal calcium absorption. Additional calcium may come from demineralization of maternal bone. The increased concentrations of 1,25(OH)2D, estrogen, IGF-1, placental lactogen, and OPG interact to facilitate these changes. Despite the increased need for calcium during pregnancy, urinary calcium losses are increased

Dietary calcium

Intestinal calcium absorption

Fetus

Blood

Urinary calcium

Bone

FIGURE 2 Adaptations in the calcium economy during pregnancy. Solid arrows indicate an increase with arrow thicknesses representing the magnitude of the fluxes.

841

due to increased glomerular filtration rate and an increased absorptive calcium load.

III. EFFECTS OF LOW MATERNAL VITAMIN D AND CALCIUM INTAKE DURING PREGNANCY ON THE FETUS AND NEONATE A. Vitamin D Maternal vitamin D deficiency is more likely to occur in winter months, in countries that do not routinely fortify dairy or other food products with vitamin D, among members of ethnic groups who cover most of their skin, or among individuals with heavily pigmented skin. Few randomized nutritional vitamin D and calcium interventions have been conducted during pregnancy, and the importance of maternal vitamin D and calcium intake is best illustrated from observational studies of women with poor calcium and/or vitamin D intake. These observational studies, along with the results of the few clinical trials that have been conducted, indicate that maternal vitamin D and calcium status are important in neonatal handling of calcium, and possibly in fetal growth and bone maturation and mineralization. Maternal vitamin D deficiency during pregnancy can affect neonatal calcium metabolism. Maternal vitamin D deficiency is associated with secondary hyperparathyroidism, and maternal hyperparathyroidism during pregnancy may lead to neonatal hypocalcemia or tetany [32,33]. In the early 1970s, Purvis and co-workers noted that the occurrence of neonatal tetany among 112 infants was inversely related to the amount of sunlight exposure the mothers had during the last trimester of pregnancy [34]. The authors speculated that the mothers developed hyperparathyroidism secondary to vitamin D deficiency leading to a transitory hypoparathyroidism and hypocalcemia in the neonate. Several investigators subsequently reported that infants of mothers with low vitamin D intake during pregnancy had low serum calcium concentrations in cord blood or during the first week of life [35–37]. Several randomized trials of vitamin D supplementation during pregnancy were later reported. Cockburn and co-workers randomized two obstetric wards, one with 506 women who received 400 IU vitamin D/day from the 12th week of gestation and another with 633 women who did not receive vitamin D [38]. They reported higher maternal, cord, and infant 25-hydroxyvitamin D (25OHD) concentrations with vitamin D supplementation. They also found that the incidence of neonatal hypocalcemia was less with

842

HEIDI J. KALKWARF AND BONNY L. SPECKER

Incidence of hypocalcemia (%)

18 16 14 12 Human milk Formula

10 8 6 4 2 0 Control

Vitamin D

FIGURE 3 Incidence of neonatal hypocalcemia on day 6 by type of feeding (solid bars are formula-fed; open bars are human milk-fed) in infants whose mothers received either no vitamin D (Control) or 400 IU vitamin D/d from 12th week of gestation (Vitamin D). Data from Cockburn et al. [38].

vitamin D supplementation, although this was modified by the infant’s feeding (hypocalcemia greater with formula feeding vs breast-feeding) (Fig. 3) [38]. Several randomized trials of vitamin D supplementation (1000 IU/d) during pregnancy subsequently found that infants of mothers receiving vitamin D had higher serum calcium concentrations within the first week of life than infants of mothers receiving placebo [39]. Brooke and co-workers conducted a randomized, double-blind trial of vitamin D supplementation (1000 IU/d from 28 to 32 weeks gestation) and found that infants of mothers receiving vitamin D had higher serum calcium on days 3 and 6 and a lower incidence of symptomatic hypocalcemia than infants of mothers receiving placebo [39–41]. These studies were completed in populations that are at increased risk for vitamin D deficiency, and the results indicate that adequate maternal vitamin D status during pregnancy may be necessary to ensure appropriate neonatal calcium homeostasis. Maternal vitamin D deficiency during pregnancy also may lead to impaired fetal growth and bone development. The occurrence of vitamin D deficiency is high among Asians from the Indian subcontinent living in Britain [33]. A trial of vitamin D supplementation (1000 IU/d) among pregnant Asian women found that a higher percent of the infants randomized to the placebo group (28.6%) were small-for-gestational-age compared to infants in the supplemented group (15.3%) [39]. Some investigators [37,42], but not all [41], have reported lower birth weights of infants born to mothers with low vs adequate vitamin D status. Decreased skeletal mineralization in utero may be manifested as rickets or osteopenia in the newborn infant. However, fetal or congenital rickets of the newborn are rare.

Case reports of congenital rickets in newborn infants of mothers with severe nutritional osteomalacia associated with vitamin D or calcium deficiency have been reported [43–45]. Reif and co-workers, in a case-control study, reported an association between craniotabes, or delayed ossification of the cranial vertex, and maternal and neonatal 25OHD concentrations. However, these findings have not been replicated in other observational studies or trials [36,39]. Although Brooke and co-workers did not find an association between craniotabes and vitamin D status, they did find that infants of mothers who received placebo had larger fontanelles than infants of mothers supplemented with vitamin D, which is consistent with impaired skull ossification [39]. A study conducted in China also found possible evidence for a relationship between maternal vitamin D deficiency and impaired fetal bone ossification [46]. The presence of wrist ossification centers in neonates was associated with cord serum 25OHD concentrations. A higher rate of ossification centers in newborn infants of mothers with adequate vitamin D status was apparent when compared to infants of mothers with low vitamin D status. Few studies have investigated the role of maternal vitamin D status on infant bone mineralization. Congdon et al. measured forearm BMC using single photon absorptiometry and found that BMC did not differ by history of vitamin D supplementation during pregnancy and was not correlated with cord serum 25OHD concentrations [36]. The majority of the vitamin D supplementation trials reported to date began supplements late in gestation. There are observational studies suggesting that maternal vitamin D status early in gestation may be important in fetal bone development. Seasonal differences in adult bone density have been reported by some [47,48] but not all investigators [49] and may be attributed to seasonal variations in vitamin D status. Studies that have examined seasonal differences in newborn BMC have had conflicting findings. Two studies conducted in the United States found that infants born in the summer have lower BMC compared to infants born in the winter months [50,51]. These findings are opposite to what is seen in adults. However, Namgung and co-workers examined this association in infants born in Korea and also found that winter-born infants had lower BMC than summer-born infants in Korea [52]. One explanation for these contradictory findings is that many United States women take prenatal vitamins containing vitamin D beginning in the second trimester of pregnancy. Thus the observed seasonal effects on infant BMC in the United States may reflect vitamin D status in the first trimester of pregnancy. Because there is minimal fetal calcium accretion in the first trimester,

CHAPTER 51 Vitamin D Metabolism in Pregnancy and Lactation

this would indicate some other function of vitamin D on fetal bone development. In summary, maternal vitamin D status during pregnancy has been shown to be associated with neonatal calcium homeostasis. There are conflicting reports indicating a possible role of maternal vitamin D status in fetal growth and bone development.

B. Calcium Few studies have evaluated the effects of maternal calcium intakes on fetal bone mineral accretion. A study by Raman and co-workers in India found that undernourished pregnant mothers who were supplemented with 300 or 600 mg calcium/d had similar maternal metacarpal bone density compared to mothers not supplemented, but the bone density of their newborns was greater [53]. Similar results have been reported in a large randomized trial of maternal calcium supplementation for the prevention of preeclampsia [54]. A total of 256 women were enrolled in the randomized, double-blind, placebo-controlled trial. Newborn infants of mothers in the lowest quintile of calcium intake (<600 mg/d) who were randomized to calcium supplementation had higher BMC compared to newborns in the lowest quintile whose mothers were randomized to placebo (Fig. 4). There was no difference in neonatal BMC between placebo and supplemented maternal groups in the upper four quintiles of maternal calcium intake. These studies suggest that there is a lower limit to the mother’s calcium regulatory capacity to buffer the fetus from variations in her calcium intake. This intake of approximately 600 mg/d is below the current recommended calcium intakes for pregnant women. 80 70

∗

Placebo Calcium supplemented

Newborn BMC (g)

60 50 40 30 20 10 0 <561

562–775 776–969 970 –1374 >1374 Material dietary calcium intake (mg/d)

FIGURE 4 Effects of maternal calcium intake and calcium supplementation (2 g/d) on newborn total body bone mineral content. Data from Koo et al. [54]. *p < 0.05

843

IV. ADAPTATIONS IN VITAMIN D AND CALCIUM METABOLISM DURING LACTATION AND AFTER WEANING A. Vitamin D and Calcium Metabolism during Lactation Lactating women secrete approximately 200–240 mg of calcium daily in breast milk [55]. Over 6 months of lactation this is equivalent to approximately 6% of her total skeletal calcium reserve. Despite this large transfer of calcium from the maternal circulation, maternal serum calcium concentrations are unchanged [3,56] or slightly elevated [57,58]. There is no increase in PTH concentrations during lactation. In fact, serum concentrations of PTH are lower in lactating as compared to nonlactating women in the first 3 months postpartum [56,59,60]. The lower PTH concentrations during lactation are likely to be a consequence of the rapid bone resorption that occurs especially early in lactation and the resultant increase in serum calcium concentrations. Two potential causes of bone resorption are hypoestrogenemia and elevated circulating concentrations of parathyroid hormone related peptide (PTHrP) [61–63]. Lactation results in prolonged postpartum amenorrhea and hypoestrogenemia due to suppression of the hypothalamic– pituitary–gonadal axis. Hypoestrogenemia is known to result in bone resorption in a variety of clinical and experimental situations. PTHrP also stimulates bone resorption (see Chapter 43). PTHrP is made in the mammary gland and is present in very high concentrations in breast milk [64]. Presumably some of the PTHrP synthesized by the mammary gland may gain access to the maternal circulation. Circulating PTHrP has actions similar to PTH and acts through the PTH receptor [65]. PTHrP is a potent stimulator of bone resorption, and administration of PTHrP results in an immediate increase in serum calcium concentrations [66]. In lactating women, serum concentrations of calcium are more highly correlated with PTHrP than with PTH [62], suggesting that the decrease in PTH also may be secondary to elevated PTHrP concentrations and subsequently increased serum calcium concentrations. Whether or not there is a decrease in urinary calcium excretion during lactation is unclear. Some studies have found a 20 to 50% decrease in urinary calcium excretion in lactating women [3,57,67–70]. However, some of this decrease may be a postpartum phenomenon and not just a result of lactation. Studies that compared urinary calcium excretion in lactating women to that of nonlactating postpartum controls have not found urinary calcium excretion to be lower in lactating women [56,59,60,71].

844

B. Changes in Bone Mineral during Lactation One of the primary changes in calcium homeostasis during lactation is the marked decrease in bone mineral content and density. Decreases of 3% to 9% at the lumbar spine and femoral neck have been reported [3,17,24,25,60,75–78]. The decreases in bone density of the spine and hip occur rapidly within the first 3 to 6 months of lactation, and bone density remains lower with continued lactation [75,76]. The rate of bone loss at these sites in the first 6 months of lactation is significant as it approaches 1% per month. In comparison, menopausal and early postmenopausal women lose bone at the rate of 1% to 2% per year [79]. The amount of bone lost during lactation is variable among women. Women who breast-feed longer, or who have a greater breast milk volume, have greater bone loss compared to women who breast-feed for shorter periods of time [17,55,75,80]. In addition, the length of postpartum amenorrhea is an important determinant of bone loss during lactation. Women who resume menses early have less bone loss than women who have longer periods of postpartum amenorrhea [17,25,76]. Although the length of postpartum amenorrhea and length of lactation are related, some women resume menses while still breastfeeding. Kalkwarf et al. found that the net change in bone density at the lumbar spine at 6 months postpartum was only –1.8% in women who had resumed menses whereas it was –4.4% in women who had not resumed menses despite the fact that both groups were breast-feeding five times

a day [81]. Polatti et al. also found less of a deficit in bone density in lactating women who resumed menses by 5 months postpartum as compared to those were remained amenorrheaic (−3.0% vs −5.8%) [25]. These findings underscore the importance of ovarian hormones in regulating bone loss during lactation. Dietary calcium intake does not appear to affect the amount of bone lost during lactation in women. Bone loss during lactation has been observed in women with high calcium intakes (>1500 mg/d) [17,60,75], and dietary calcium intake has not been shown to be a significant predictor of bone loss during lactation [55,59,75,80]. Furthermore, three randomized calcium supplementation trials have demonstrated that provision of supplemental calcium does not affect bone loss during lactation. Prentice et al. studied 60 lactating women in the Gambia who had a very low calcium intake (274 mg/d). Half of the women received an average of 714 mg of supplemental calcium per day, and half received a placebo. Overall there was a significant loss (1.1%) of bone mineral at the radial shaft by 13 weeks postpartum, but there was no difference in the amount of bone lost between supplemented and unsupplemented lactating women [67]. Kalkwarf et al. randomized 83 lactating women and 81 nonlactating postpartum women whose dietary calcium intake averaged 735 mg/d, to receive a calcium supplement (1 g/d) or placebo for 6 months. There was a small effect (+1.2%) of calcium supplementation on bone density of the lumbar spine when considering all women [24]. However, bone loss did not differ between lactating women who received the calcium supplement and those that received placebo (4.2% vs 4.9%) (Fig. 5). In a supplementation trial conducted in 274 Italian women, Polatti et al. found no difference in the amount of bone lost at the lumbar spine (4.0% vs 4.4%) or the ultradistal radius (2.0% vs 2.2%) between supplemented 8 6 Percent change

Unlike during pregnancy, there is no increase in circulating concentrations of 1,25(OH)2D in lactating as compared to nonlactating postpartum women [56,59]. Commensurate with this finding is that there is no difference in intestinal calcium absorption in lactating as compared to nonlactating women [3,71–73]. Intestinal calcium absorption is increased in lactating rats that are suckling multiple pups, but this does not occur in women nursing one infant. Greer et al. found that women nursing twins had elevated concentrations of PTH and 1,25(OH)2D [74]. Urinary calcium excretion and intestinal calcium absorption were not measured, but presumably elevations in PTH and 1,25(OH)2D concentrations resulted in changes in urinary calcium excretion and intestinal calcium absorption. Although bone demineralization, and not improved efficiency of intestinal calcium absorption, is the primary compensatory response to secure calcium in lactating women, it can be hypothesized that increased absorption efficiency may occur in situations of greater calcium demand such as women nursing multiple infants.

HEIDI J. KALKWARF AND BONNY L. SPECKER

4

Nonlactating groups

2

Calcium

0

Placebo

−2

Lactating groups

−4 −6

Calcium Placebo 0.5

3 Months since delivery

6

FIGURE 5 Effects of lactation and calcium supplementation (1 g/d) on percent change in bone density of the lumbar spine during the first 6 months postpartum. Reproduced from Kalkwarf et al. [24].

845

CHAPTER 51 Vitamin D Metabolism in Pregnancy and Lactation

D. Changes in Bone Mineral after Weaning

Intestinal calcium absorption

Milk

Blood

Urinary calcium

Bone

FIGURE 6

Adaptations in the calcium economy during lactation. Solid arrows indicate an increase with arrow thicknesses representing the magnitude of the fluxes. Dashed arrows indicate a decrease.

(1 g calcium/d) and unsupplemented women over 6 months of lactation [25]. The primary adaptive strategy to secure calcium to support breast milk production is demineralization of maternal bone (Fig. 6). Lactation also may result in urinary calcium conservation, although the results from studies are conflicting as to whether this is a lactation effect or a postpartum effect. Bone loss during lactation is related to postpartum amenorrhea and hypoestrogenemia. Elevated circulating concentrations of PTHrP also may have a role in bone loss during lactation.

C. Vitamin D and Calcium Metabolism after Weaning Additional adjustments in calcium metabolism occur shortly after lactation has stopped as the maternal physiologic system switches from mobilizing calcium from the skeleton and secreting calcium into breast milk to restoring maternal calcium reserves. Some studies have found an increase in serum concentrations of PTH shortly after weaning and decreased urinary calcium excretion, presumably in response to increased PTH concentrations [7,57]. However, these changes have not been found in all studies [56]. Kalkwarf et al. found that serum concentrations of 1,25(OH)2D were higher in women shortly after weaning [72], and this was accompanied by a higher intestinal calcium absorption (37% vs 31%). Ritchie et al. did not find a significant increase in intestinal calcium absorption after weaning, but the smaller sample size in that study may have limited their ability to detect a small increase [3].

Maternal bone density increases rapidly after weaning. Much of the bone density lost during lactation is recovered within the first 6 months after weaning. Laskey and co-workers demonstrated that the resumption of menses was as good a predictor of bone changes as was the length of lactation [76]. Increases in bone density after weaning occur earlier for the spine than for the femoral neck, which may be a consequence of the greater amount of trabecular bone at the spine [17,75,76,78]. Although most studies show a complete recovery of bone density at the spine after weaning, it is not clear that the recovery of bone at the femoral neck is complete, as deficits in bone density at this site were still evident at the end of the study follow-up in most of these studies [17,75,78]. It is possible that a complete recovery of bone may have occurred with a longer follow-up period. Consistent with this are the results of studies in postmenopausal women that have found that lactation history is not a significant predictor of bone density [27–30] and is not associated with increased risk of hip fracture [82–84]. Kalkwarf et al. conducted a calcium supplementation trial in 76 women who had lactated for 6 months and then weaned their infants and 82 nonlactating postpartum controls to determine whether provision of supplemental calcium could enhance bone recovery after weaning [24]. By 6 months after the initiation of weaning (4 months after complete weaning), lactating women who had received 1 g/d of supplemental calcium had a significantly greater increase in spinal BMD compared to women who received the placebo (5.9% vs 4.4%) (Fig. 7). The restoration of bone mass after lactation has ceased is important in maintaining maternal bone health. Lactating groups

8 6 Percent change

Dietary calcium

Weaning

Calcium Placebo

4 Calcium Placebo

2 0

Nonlactating groups

−2 −4 −6

FIGURE 7

6

9 Months since delivery

12

Effects of weaning and calcium supplementation (1 g/d) on percent change in bone density of the lumbar spine during the second 6 months postpartum. Reproduced from Kalkwarf et al. [24].

846

HEIDI J. KALKWARF AND BONNY L. SPECKER

Dietary calcium

Intestinal calcium absorption

Blood

Urinary calcium

Bone

FIGURE 8 Adaptations in the calcium economy after weaning. Solid arrows indicate an increase, with arrow thicknesses representing the magnitude of the fluxes. Dashed arrows indicate a decrease.

Bone mass and density increase in parallel with the return of menstruation and presumably normalization of circulating estrogen concentrations. The recovery of bone mass may be facilitated by an increase in intestinal calcium absorption efficiency and a decrease in urinary calcium excretion, but these alterations have not been found consistently across studies (Fig. 8).

V. EFFECTS OF LOW MATERNAL VITAMIN D AND CALCIUM INTAKES ON BREAST MILK VITAMIN D AND CALCIUM CONCENTRATIONS

infant is dependent upon endogenous synthesis or other dietary sources for vitamin D. Most reported cases of rickets have been of black infants, supporting the premise that persons with dark skin have difficulty synthesizing adequate amounts of vitamin D due to the relative inability of sunlight to penetrate heavily pigmented skin. In addition, the diet of mothers of rachitic infants appears to be low in vitamin D and the mothers may be vitamin D deficient themselves. Although some investigators have found breast milk vitamin D or 25OHD concentrations to be correlated with maternal intake of vitamin D, mothers who consume 600–700 IU vitamin D/d still have low concentrations of vitamin D in breast milk ranging from only 5 to 136 IU/liter. The biological activity of vitamin D in human milk averages 13 IU/liter, while the 25OHD concentrations represent 38 IU/liter. The average biological activity of vitamin D in human milk is less than 50 IU per day, assuming an average intake of 0.75 liter/d [87]. Investigators recently found that consumption of cod liver oil supplements among Icelandic women increased milk vitamin D concentrations, but the milk concentrations were still below the current Nordic recommendations [88]. Ala-Houhala and co-workers from Finland, in a series of vitamin D supplementation trials, found that supplementing lactating mothers with up to 1000 IU vitamin D/d in northern latitudes during winter months increased maternal serum 25OHD concentrations, but did not stabilize infant serum 25OHD concentrations (Fig. 9) [89]. Maternal supplementation with 2000 IU/d was found to normalize infant serum 25OHD concentrations [90]. There were no differences in infant serum calcium or alkaline phosphatase concentrations when mothers were supplemented with

50

A. Vitamin D Infant serum 25-OHD

Vitamin D deficiency leads to rickets in children. Infant formula is routinely fortified with vitamin D, but very low vitamin D concentrations have been found in human milk [85]. Infant serum 25OHD is correlated with maternal vitamin D status early in the neonatal period and is probably a result of placental vitamin D transfer and fetal stores. Beyond the neonatal period, the breast-fed infant’s serum 25OHD concentrations are correlated with neither breast milk vitamin D nor maternal serum 25OHD concentrations [86], and the

40 30

Ala-Houhala, 1985 Ala-Houhala, 1986

20 10

0 Infant: 0 Maternal 0

0 1,000

0 2,000

400 0

1,000 IU/d 0 IU/d

FIGURE 9 Infant serum 25OHD concentrations (ng/ml) after 20 weeks of either infant or maternal vitamin D supplementation. Data from Ala-Houhala et al. [89,90].

847

CHAPTER 51 Vitamin D Metabolism in Pregnancy and Lactation

either 1000 or 2000 IU vitamin D/d or when infants were supplemented with 400 IU/d. Specker and co-workers found based on conservative estimates, that exclusively breast-fed infants residing in Cincinnati could maintain serum 25OHD concentrations above the lower limit of normal (11 ng/ml) with 2 hr of sunshine exposure per week if fully clothed except for the face [86]. The cutoff for defining low 25OHD is based on the concentration at which nutritional rickets has been observed. Other factors such as latitude, season, weather conditions, and use of sunscreens may affect vitamin D status. Large seasonal differences in sunlight exposure and serum 25OHD concentrations over the first year of life have been observed in infants followed longitudinally [91]. These findings indicate that the infant’s sunlight exposure plays a more dominant role in determining his or her vitamin D status than the mother’s vitamin D status or milk vitamin D concentrations.

B. Calcium Two of the randomized trials described earlier that investigated the effect of supplemental calcium on maternal bone changes during lactation also measured milk calcium concentrations. Kalkwarf and co-workers found that women with habitually low calcium intake (<800 mg/d) who were supplemented with calcium (1 g/d) had breast milk calcium concentrations similar to that of women who received the placebo [24] (Fig. 10). Prentice and co-workers also conducted a randomized calcium supplementation trial among Gambian women and found no effect of calcium intake on breast milk calcium concentrations [67].

Breast milk calcium (mg/dL)

35 Calcium supplemented Placebo

30 25 20 15 10 5 0

FIGURE 10

0.5

3 Months postpartum

6

Effects of calcium supplementation (1 g/d) on breast milk calcium concentrations. Data from Kalkwarf et al. [24].

These results are consistent with older observational studies showing that milk calcium concentrations are not associated with maternal calcium intake [92,93].

VI. CONCLUSIONS Multiple changes in the maternal calcium economy occur during pregnancy, lactation, and after weaning to protect maternal calcium concentrations while providing sufficient calcium for fetal bone mineral accretion, breast milk production and maternal bone recovery. The strategies to secure calcium during these physiologic states differ. During pregnancy, the primary strategy to secure additional calcium is by increases in serum concentrations of 1,25(OH)2D and intestinal calcium absorption. During lactation, maternal bone is demineralized, possibly because of the lactation-induced amenorrhea, in order to secure adequate availability of calcium for breast milk production. After weaning and the return of menses, maternal bone density increases. Lactation does not appear to have a long-term negative effect on maternal bone density nor does it increase osteoporotic fracture risk. The maternal calcium regulatory system is able to provide sufficient calcium to the fetus and for breast milk even when calcium intake is low. However, there is some evidence that neonatal calcium homeostasis and fetal bone mineral accretion may be compromised when maternal vitamin D status is low or calcium intake is below 600 mg/d.

References 1. Widdowson EM 1981 Changes in body composition during growth. In: Davis JA, Dobbings J (eds) Scientific Foundations of Paediatrics. William Heinemann Medical Books, London, pp. 330–342. 2. Black A, Topping J, Durham B, Farquharson R, Fraser W 2000 A detailed assessment of alterations in bone turnover, calcium homeostasis, and bone density in normal pregnancy. J Bone Miner Res 15:557–564. 3. Ritchie LD, Fung EB, Halloran BP, Turnlund JR, Van Loan MD, Cann CE, King JC 1998 A longitudinal study of calcium homeostasis during human pregnancy and lactation and after resumption of menses. Am J Clin Nutr 67:693–701. 4. Gallacher SJ, Fraser WD, Owens OJ, Dryburgh FJ, Logue FC, Jenkins A, Kennedy J, Boyle IT 1994 Changes in calciotrophic hormones and biochemical markers of bone turnover in normal human pregnancy. Eur J Endocrinol 131:369–374. 5. Pitkin RM, Reynolds WA, Williams GA, Hargis GK 1979 Calcium metabolism in normal pregnancy: A longitudinal study. Am J Obstet Gynecol 133:781–790. 6. Naylor KE, Iqbal P, Fledelius C, Fraser RB, Eastell R 2000 The effect of pregnancy on bone density and bone turnover. J Bone Miner Res 15:129–137.

848 7. Cross NA, Hillman LS, Allen SH, Krause GF, Vieira NE 1995 Calcium homeostasis and bone metabolism during pregnancy, lactation, and postweaning: A longitudinal study. Am J Clin Nutr 61:514–523. 8. Bikle DD, Gee E, Halloran B, Haddad JG 1984 Free 1,25-Dihydroxyvitamin D levels in serum from normal subjects, pregnant subjects, and subjects with liver disease. J Clin Invest 74:1966–1971. 9. Delvin EE, Arabian A, Glorieux FH, Mamer OA 1985 In vitro metabolism of 25-OH-hydroxycholecalciferol by isolated cells from human decidua. J Clin Endocrinol Metab 60:880–885. 10. Uemura H, Yasui T, Kiyokawa A, Kuwahara A, Ikawa H, Matsuzaki T, Maegawa M, Furumoto H, Irahara M 2002 Serum osteoprotegerin/osteoclastogenesis-inhibitory factor during pregnancy and lactation and the relationship with calciumregulating hormones and bone turnover markers. J Endocrinol 174:353–359. 11. Gertner JM, Coustan DR, Kliger AS, Mallette LE, Ravin N, Broadus AE 1986 Pregnancy as state of physiologic absorptive hypercalciuria. Am J Med 81:451–456. 12. Kent GN, Price RI, Gutteridge DH, Allen JR, Rosman KJ, Smith M, Bhagat CI, Wilson SG, Tetallack RW 1993 Effect of pregnancy and lactation on maternal bone mass and calcium metabolism. Osteoporos Int (Suppl. 1):S44–S47. 13. Martin TJ, Gillespie MT 2001 Receptor activator of nuclear factor kappa B ligand (RANKL): Another link between breast and bone. Trends Endocrinol Metab 12:2–4. 14. Matsumoto I, Kosha S, Noguchi S, Kojima N, Oki T, Douchi T, Nagata Y 1995 Changes of bone mineral density in pregnant and postpartum women. J Obstet Gynaecol 21:419–425. 15. Sowers M, Crutchfield M, Jannausch M, Updike S, Corton G 1991 A prospective evaluation of bone mineral change in pregnancy. Obstet Gynecol 77:841–845. 16. Bjorklund K, Naessen T, Nordstrom ML, Bergstrom S 1999 Pregnancy-related back and pelvic pain and changes in bone density. Acta Obstet Gynecol Scand 78:681–685. 17. Kolthoff N, Eiken P, Kristensen B, Nielsen SP 1998 Bone mineral changes during pregnancy and lactation: A longitudinal cohort study. Clin Sci 94:405–412. 18. Shefras J, Farquharson R 1996 Bone density studies in pregnant women receiving heparin. Eur J Obste Gynecol Reprod Biol 65:171–174. 19. Drinkwater BL, Chesnut CH, III 1991 Bone density changes during pregnancy and lactation in active women: A longitudinal study. Bone Miner 14:153–160. 20. Paparella P, Giorgino R, Maglione A, Lorusso D, Scirpa P, Del Bosco A, Mancuso S 1995 Maternal ultrasound bone density in normal pregnancy. Clin Exp Obstet Gynecol 22:268–278. 21. Gambacciani M, Spinetti A, Gallo R, Cappagli B, Teti GC, Facchini V 1995 Ultrasonographic bone characteristics during normal pregnancy: Longitudinal and cross-sectional evaluation. Am J Obstet Gynecol 173:890–893. 22. Yamaga A, Taga M, Minaguchi H, Sato K 1996 Changes in bone mass as determined by ultrasound and biochemical markers of bone turnover during pregnancy and puerperium: A longitudinal study. J Clin Endocrinol Metab 81:752–756. 23. Aguado F, Revilla M, Hernandez ER, Menendez M, Cortes-Prieto J, Villa LF, Rico H 1998 Ultrasonographic bone velocity in pregnancy: A longitudinal study. Am J Obstet Gynecol 178:1016–1021. 24. Kalkwarf HJ, Specker BL, Bianchi DC, Ranz J, Ho M 1997 The effect of calcium supplementation on bone density during lactation and after weaning. N Engl J Med 337:523–528. 25. Polatti F, Capuzzo E, Viazzo F, Colleoni R, Klersy C 1999 Bone mineral changes during and after lactation. Obstet Gynecol 94:52–56.

HEIDI J. KALKWARF AND BONNY L. SPECKER

26. Hopkinson JM, Butte NF, Ellis K, Smith EO 2000 Lactation delays postpartum bone mineral accretion and temporarily alters its regional distribution in women [In Process Citation]. J Nutr 130:777–783. 27. Kritz-Silverstein D, Barrett-Connor E, Hollenbach KA 1992 Pregnancy and lactation as determinants of bone mineral density in postmenopausal women. Am J Epidemiol 136:1052–1059. 28. Melton LJ, III, Bryant SC, Wahner HW, O’Fallon WM, Malkasian GD, Judd HL, Riggs BL 1993 Influence of breastfeeding and other reproductive factors on bone mass later in life. Osteoporos Int 3:76–83. 29. Fox KM, Magaziner J, Sherwin R, Scott JC, Plato CC, Nevitt M, Cummings S 1993 Reproductive correlates of bone mass in elderly women. J Bone Miner Res 8:901–908. 30. Grainge MJ, Coupland CAC, Cliffe SJ, Chilvers CED, Hosking DJ 2001 Reproductive, menstrual and menopausal factors: which are associated with bone mineral density in early postmenopausal women? Osteopor Int 12:777–787. 31. Henderson PH, 3rd, Sowers M, Kutzko KE, Jannausch ML 2000 Bone mineral density in grand multiparous women with extended lactation. Am J Obstet Gynecol 182:1371–1377. 32. Daaboul J, Sanderson S, Kristensen K, Kitson H 1997 Vitamin D deficiency in pregnant and breast-feeding women and their infants. J Perinatol 17:10–14. 33. Okonofua F, Menon RK, Houlder S, Thomas M, Robinson D, O’Brien S, Dandona P 1986 Parathyroid hormone and neonatal calcium homeostasis: Evidence for secondary hyperparathyroidism in the Asian neonate. Metabolism 35:803–806. 34. Purvis RJ, MacKay GS, Cockburn F, Barrie WJM, Wilkinson EM, Belton NR, Forfar JO 1973 Enamel hypoplasia of the teeth associated with neonatal tetany: A manifestation of maternal vitamin-D deficiency. Lancet (October 13):811–814. 35. Paunier L, Lacourt G, Pilloud P, Schlaeppi P, Sizonenko PC 1978 25-Hydroxyvitamin D and calcium levels in maternal, cord and infant serum in relation to maternal vitamin D intake. Helvetica Paediatr Acta 33:95–103. 36. Congdon P, Horsman A, Kirby PA, Dibble J, Bashir T 1983 Mineral content of the forearms of babies born to Asian and white mothers. Br Med J 286:1234–1235. 37. Marya RK, Rathee S, Lata V, Mudgil S 1981 Effects of vitamin D supplementation in pregnancy. Gynecol Obstet Inves 12:155–161. 38. Cockburn F, Belton NR, Purvis RJ, Giles MM, Brown JK, Turner TL, Wilkinson EM, Forfar JO, Barrie WJM, McKay GS, Pocock SJ 1980 Maternal vitamin D intake and mineral metabolism in mothers and their newborn infants. Br Med J 281:11–14. 39. Brooke OG, Brown IRF, Bone CDM, Carter ND, Cleeve HJW, Maxwell JD, Robinson VP, Winder SM 1980 Vitamin D supplements in pregnant Asian women: Effects on calcium status and fetal growth. Br Med J 280:751–754. 40. Delvin EE, Salle BL, Glorieux FH, Adeleine P, David LS 1986 Vitamin D supplementation during pregnancy: Effect on neonatal calcium homeostasis. J Pediatr 109:328–334. 41. Mallet E, Gugi B, Brunelle P, Henocq A, Basuyua JP, Lemeur H 1986 Vitamin D supplementation in pregnancy: A controlled trial of two methods. Obstet Gynecol 68:300–304. 42. Marya RK, Rathee S, Dua V, Sangwan K 1988 Effect of vitamin D supplementation during pregnancy on foetal growth. Indian J Med Res 88:488–492. 43. Zhou H 1991 Rickets in China. In: Glorieux FH (ed) Raven Press, New York, p. 253. 44. Russell JGB, Hill LF 1974 True fetal rickets. Br Radiol 47:732. 45. Moncrieff M, Fadahunsi TO 1974 Congenital rickets due to maternal vitamin D deficiency. Arch Dis Child 49:810–811.

CHAPTER 51 Vitamin D Metabolism in Pregnancy and Lactation

46. Specker B, Ho M, Oestreich A, Yin T, Shui Q, Chen X, Tsang R 1992 Prospective study of vitamin D supplementation and rickets in China. J Pediatr 120:733–739. 47. Dawson-Hughes B, Harris S 1991 Regional changes in body composition by time of year in healthy postmenopausal women. Am J Clin Nutr 56:307–313. 48. Chapuy MC, Schott AM, Garnero P, Hans D, Delmas PD, Meunier PJ 1996 Healthy elderly French women living at home have secondary hyperparathyroidism and high bone turnover in winter. J Clin Endocrinol Metab 81:1129–1133. 49. Melin A, Wilske J, Ringertz H, Saaf M 2001 Seasonal variations in serum levels of 25-hydroxyvitamin D and parathyroid hormone but no detectable change in femoral neck bone density in an older population with regular outdoor exposure. J Am Geriatr Soc 49:1190–1196. 50. Namgung R, Mimouni F, Campaigne BN, Ho ML, Tsang RC 1992 Low bone mineral content in summer compared with winter-born infants. J Pediatr Gastroenterol Nutr 15:285–288. 51. Namgung R, Tsang RC, Specker BL, Sierra RI, Ho ML 1994 Low bone mineral content and high serum ostocalcin and 1,25dihydroxyvitamin D in summer- versus winter-born newborn infants: An early fetal effect? J Pediatr Gastroenterol Nutr 19:220–227. 52. Namgung R, Tsang RC, Lee C, Han D-G, Ho ML, Sierra RI 1998 Low total body bone mineral content and high bone resorption in Korean winter-born versus summer-born newborn infants. J Pediatr Gastro Nutr 132:421–425. 53. Raman L, Rajalakshmi K, Krishnamachari KAVR, Gowrinath Sastry J 1978 Effect of calcium supplementation to undernourished mothers during pregnancy on the bone density of the neonates. Am J Clin Nutr 31:466–469. 54. Koo W, Walters J, Esterlitz J, Levine R, Bush A, Sibai B 1999 Maternal calcium supplementation and fetal bone mineralization. Obstet Gynecol 94:577–582. 55. Laskey MA, Prentice A, Hanratty LA, Jarjou LMA, Dibba B, Beavan SR, Cole TJ 1998 Bone changes after 3 mo of lactation: Influence of calcium intake, breast-milk output, and vitamin D-receptor genotype. Am J Clin Nutr 67:685–692. 56. Kalkwarf HJ, Specker BL, Ho M 1999 Effects of calcium supplementation on calcium homeostatis and bone turnover in lactating women. J Clin Endocrinol Metab 84:464–470. 57. Kent GN, Price RI, Gutteridge DH, Smith M, Allen JR, Bhagat BH, Barnes MP, Hickling CJ, Retallack RW, Wilson SG, Devlin RD, Davies C, St. John A 1990 Human lactation: Forearm trabecular bone loss, increased bone turnover, and renal conservation of calcium and inorganic phosphate with recovery of bone mass following weaning. J Bone Miner Res 5: 361–369. 58. Specker BL, Tsang RC, Ho ML 1991 Changes in calcium homeostasis over the first year postpartum: Effect of lactation and weaning. Obstet Gynecol 78:56–62. 59. Krebs NF, Reidinger CJ, Robertson AD, Brenner M 1997 Bone mineral density changes during lactation: Maternal, dietary, and biochemical correlates. Am J Clin Nutr 65:1738–1746. 60. Affinito P, Tommaselli GA, Di Carlo C, Guida F, Nappi C. 1996 Changes in bone mineral density and calcium metabolism in breastfeeding women: A one year follow-up study. J Clin Endocrinol Metab 81:2314–2318. 61. Sowers MF, Hollis BW, Shapiro B, Randolph J, Janney CA, Zhang D, Schork MA, Crutchfield M, Stanczyk F, Russell-Aulet M 1996 Elevated parathyroid hormone-related peptide associated with lactation and bone density loss. JAMA 276:549–554. 62. Dobnig H, Kainer F, Stepan V, Winter R, Lipp R, Schaffer M, Kahr A, Nocnik S, Patterer G, Leb G 1995 Elevated parathyroid hormone-related peptide levels after human

63.

64.

65.

66.

67.

68.

69.

70.

71.

72. 73.

74. 75. 76. 77. 78.

849 gestation: relationship to changes in bone and mineral metabolism. J Clin Endocrinol Metab 80:3699–3707. Lippuner K, Zehnder H-J, Casez J-P, Takkinen R, Jaeger P 1996 PTH-related protein is released into the mother’s bloodstream during lactation: Evidence for beneficial effects on maternal calcium-phosphate metabolism. J Bone Miner Res 11:1394–1399. Seki K, Kato T, Sekiya S, Makimura N, Kudoh K, Furuya K, Nagata I 1997 Parathyroid-hormone-related protein in human milk and its relation to milk calcium. Gynecol Obstet Invest 44:102–106. Fraher LJ, Klein K, Marier R, Freeman D, Hendy GN, Goltzman D, Hodsman AB 1995 Comparison of the pharmacokinetics of parenteral parathyroid hormone-(1-34) [PTH-(1-34)] and PTH-related peptide-(1-34) in healthy young humans. J Clin Endocrinol Metab 80:60–64. Fraher LJ, Hodsman AB, Jonas K, Saunders D, Rose CI, Henderson JE, Hendy GN, Goltzman D 1992 A comparison of the in vivo biochemical responses to exogenous parathyroid hormone-(1-34) [PTH-(1-34)] and PTH-related peptide-(1-34) in man. J Clin Endocrinol Metab 75:417–423. Prentice A, Jarjou LMA, Cole TJ, Stirling DM, Dibba B, Fairweather-Tait S 1995 Calcium requirements of lactating Gambian mothers: Effects of a calcium supplement on breast-milk calcium concentration, maternal bone mineral content, and urinary calcium excretion. Am J Clin Nutr 62:58–67. Klein CJ, Moser-Veillon PB, Douglass LW, Ruben KA, Trocki O 1995 A longitudinal study of urinary calcium, magnesium, and zinc excretion in lactating and nonlactating postpartum women. Am J Clin Nutr 61:779–786. Specker BL, Vieira NE, O’Brien KO, Ho ML, Heubi JE, Abrams SA, Yergey AL 1994 Calcium kinetics in lactating women with low and high calcium intakes. Am J Clin Nutr 59:593–599. Donangelo CM, Trugo NMF, Melo GJO, Gomes DD, Henriques C 1996 Calcium homeostasis during pregnancy and lactation in primiparous and multiparous women with subadequate calcium intakes. Nutr Res 16:1631–1640. Moser-Veillon PB, Mangels AR, Vieira NE, Yergey AL, Patterson KY, Hill AD, Veillon C 2001 Calcium fractional absorption and metabolism assessed using stable isotopes differ between postpartum and never pregnant women. J Nutr 131:2295–2299. Kalkwarf HJ, Specker BL, Heubi JE, Vieira NE, Yergey AL 1996 Intestinal calcium absorption of women during lactation and after weaning. Am J Clin Nutr 63:526–531. Fairweather-Tait S, Prentice A, Heumann KG, Jarjou LMA, Stirling DM, Wharf SG, Turnlund JR 1995 Effect of calcium supplements and stage of lactation on the calcium absorption efficiency of lactating women accustomed to low calcium intakes. Am J Clin Nutr 62:1188–1192. Greer FR, Lane J, Ho M 1984 Elevated serum parathyroid hormone, calcitonin, and 1,25-dihydroxyvitamin D in lactating women nursing twins. Am J Clin Nutr 40:562–568. Sowers MF, Corton G, Shapiro B, Jannausch ML, Crutchfield M, Smith ML, Randolph JF, Hollis B 1993 Changes in bone density with lactation. JAMA 269:3130–3135. Laskey M, Prentice A 1999 Bone mineral changes during and after lactation. Obstet Gynecol 94:608–615. Honda A, Kurabayashi T, Yahata T, Tomita M, Takakuwa K, Tanaka K 1998 Lumbar bone mineral density changes during pregnancy and lactation. Int J Gynecol Obstet 63:253–258. Karlsson C, Obrant KJ, Karlsson M 2001 Pregnancy and lactation confer reversible bone loss in humans. Osteoporosis Int 12:828–834.

850 79. Nordin BEC, Need AG, Chatterton BE, Horowitz M, Morris HA 1990 The relative contributions of age and years since menopause to postmenopausal bone loss. J Clin Endocrinol Metab 70:83–88. 80. Little KD, Clapp JF 1998 Self-selected recreational exercise has no impact on early postpartum lactation-induced bone loss. Med Sci Sports Exerc 30:831–836. 81. Kalkwarf HJ, Specker BL 1995 Bone mineral loss during lactation and recovery after weaning. Obstet Gynecol 86:26–32. 82. Michaelsson K, Baron JA, Farahmand BY, Ljunghall S 2001 Influence of parity and lactation on hip fracture risk. Am J Epidemiol 153:1166–1172. 83. Cumming RG, Klineberg RJ 1993 Breastfeeding and other reproductive factors and the risk of hip fractures in elderly women. Int J Epidemiol 22:684–691. 84. Hoffman S, Grisso JA, Kelsey JL, Gammon MD, O’Brien LA 1993 Parity, lactation and hip fracture. Osteoporosis Int 3:171–176. 85. Hollis BW, Roos BA, Draper HH, Lambert PW 1981 Vitamin D and its metabolites in human and bovine milk. J Nutr 111:1240–1248. 86. Specker BL, Valanis B, Hertzberg V, Edwards N, Tsang RC 1985 Sunshine exposure and serum 25-hydroxyvitamin D concentrations in exclusively breast-fed infants. J Pediatr 107:372–376.

HEIDI J. KALKWARF AND BONNY L. SPECKER

87. Specker BL, Tsang RC, Hollis BW 1985 Effect of race and diet on human milk vitamin D and 25-hydroxyvitamin D. Am J Dis Child 139:1134–1137. 88. Olafsdottir AS, Wagner KH, Thorsdottir I, Elmadfa I 2001 Fat-soluble vitamins in the maternal diet, influence of cod liver oil supplementation and impact of maternal diet on human milk composition. Ann Nutr Metab 45:265–272. 89. Ala-Houhala M 1985 25-Hydroxyvitamin D levels during breast-feeding with or without maternal or infantile supplementation of vitamin D. J Pediatr Gastro Nutr 4: 220–226. 90. Ala-Houhala M, Koskinen T, Terho A, Koivula T, Visakorpi J 1986 Maternal compared with vitamin D supplementation. Arch Dis Child 61:1159–1163. 91. Specker BL, Tsang RC 1987 Cyclical serum 25-hydroxyvitamin D concentrations paralleling sunshine exposure in exclusively breast-fed infants. J Pediatr 110:744–747. 92. Moser PB, Reynolds RD, Acharya S, Howard MP, Andon MB 1988 Calcium and magnesium dietary intakes and plasma and milk concentrations of Nepalese lactating women. Am J Clin Nutr 47:735–739. 93. Vaughn LA, Weber CW, Kemberling SR 1979 Longitudinal changes in the mineral content of human milk. Am J Clin Nutr 32:2301–2306.

CHAPTER 52

Vitamin D and Reproductive Organs KEIICHI OZONO, SHIGEO NAKAJIMA TOSHIMI MICHIGAMI

I. II. III. IV. V. VI. VII.

Historical View Chick Embryonic Development and Egg Hatchability Mouse Models for a Lack of Vitamin D Function Fetal Development and Vitamin D Synthesis Calcium Homeostasis in the Fetus and Its Mother Fertility Testis

I. HISTORICAL VIEW Vitamin D is an essential nutrient in human that prevents rickets, and its classical actions are targeted to bone, intestine, and kidney to maintain mineral and bone homeostasis [1]. Besides these classical actions, the natural vitamin D metabolites including 1α,25dihydroxycholecalciferol [1,25(OH)2D3] and 24R,25dihydroxycholecalciferol [24R,25(OH)2D3] are reported to be necessary for embryonic development and normal egg hatchability in white leghorn hens [2]. These interesting reports represent among the first to shed light on nonclassical action of vitamin D such as cell differentiation, proliferation, insulin secretion, and fertility. In mammals, it has been argued that vitamin D itself plays a role in reproductive organs. However, mice in which the vitamin D receptor (VDR) gene has been deleted (VDR-null) show infertility in at least one strain [3]. Moreover, another mouse model for vitamin D resistance in which 1α-hydroxylase was ablated exhibited infertility as well [4]. In humans, osteomalacia may cause distortion of the maternal pelvis and be a risk factor for cephalopelvic disproportion, although no association of vitamin D deficiency with obstructed labor was found in the Karachi study [5]. These findings revealed the significance of the role of vitamin D in reproductive organs in vivo. In this chapter, the roles of vitamin D in reproductive organs are reviewed.

II. CHICK EMBRYONIC DEVELOPMENT AND EGG HATCHABILITY Hen’s eggs have calcium stored in the shell, and in turn calcium precipitation provides the eggshell with strength. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Department of Developmental Medicine (Pediatrics), Osaka University Graduate School of Medicine, Osaka, Japan Department of Environmental Medicine, Osaka Medical Center and Institute for Maternal and Child Health, Osaka, Japan

VIII. IX. X. XI. XII. XIII.

Ovary Uterus Mammary Gland Aromatase Placenta Concluding Remarks References

Before hatching, calcium is mobilized from the eggshell to the embryo. There it supports skeletal development, and the resulting weakening of the eggshell allows the chick to hatch. Immunological analyses demonstrated the presence of the vitamin D receptor (VDR) in the shell gland, suggesting that vitamin D is involved in the control of calcium deposition onto the eggshell [6]. It is also known that the chorioallantoic membrane just inside the eggshell is rich in VDR, especially just before hatching [7]. When a hen is deficient in vitamin D, the eggshell becomes soft due to deficiency of calcium, reducing egg hatchability. Norman et al. reported that both 1,25(OH)2D3 and 24R,25(OH)2D3 were necessary for embryonic development and normal egg hatchability in the hen and in the Japanese quail Coturnix coturnix japonica [2,8]. In their experiments, hatchability equivalent to that of D3repleted hens was obtained only when hens received a combination of 1,25(OH)2D3 and 24R,25(OH)2D3. These results suggest a specific biological role for 24R,25(OH)2D3 in hatchability. In further studies using the two stereoisomers of the vitamin D3 metabolite, namely the naturally occurring 24R,25(OH)2D3 and its unnatural epimer 24S,25(OH)2D3, it was concluded that the former is essential for hatching in vitamin D– deficient hens and Japanese quail [8]. In general, 24R,25(OH)2D3 is believed to be an inactive metabolite of vitamin D3. The compound has weak binding affinity for the VDR and increases serum calcium levels only when administered in large amounts in vitamin D–deficient animals. However, some reports describe specific functions of 24R,25(OH)2D3 in chondrocyte differentiation and bone formation as well as hatchability [9,10]. 1,25(OH)2D3 is able to induce the activity of 24-hydroxylase (this enzyme also converts Copyright © 2005, Elsevier, Inc. All rights reserved.

852

KEIICHI OZONO, SHIGEO NAKAJIMA AND TOSHIMI MICHIGAMI

1,25(OH)2D3) after binding to VDR. In fact, two vitamin D–responsive elements (VDREs) have been identified in the rat and human 24-hydroxylase gene promoter [11]. Because of a reduction in the levels of 24R,25(OH)2D3 in mouse models for vitamin D resistance, the resulting phenotype of these mice does not appear to rule out a specific role for 24R,25(OH)2D3 in vivo. The mechanism of the specific action of 24R,25(OH)2D3, if any, remains to be clarified.

IV. FETAL DEVELOPMENT AND VITAMIN D SYNTHESIS

III. MOUSE MODELS FOR A LACK OF VITAMIN D FUNCTION To fully understand the in vivo function of VDR, a member of the nuclear hormone receptor superfamily, VDR-null mutant mice were generated using established gene engineering technology [3,12] (see Chapter 7). In the mice deficient in VDR, no defects in development and growth were observed before weaning, whereas in the neonate and adult, the mutation resulted in reduced expression of vitamin D target genes. Accordingly, after weaning, the mutant mice failed to thrive and showed alopecia, hypocalcemia, and impaired bone formation. These findings are consistent with typical features of vitamin D–dependent rickets type II in humans. While the lethality among VDR-null mice varies, perhaps due to the difference in calcium contents in water, infertility in the mutant mice was reported in at least one strain [3]. Uterine hypoplasia and impaired folliculogenesis were found in the reproductive organs. These effects will be reviewed later in this chapter. As another form for vitamin D resistance, a mouse deficient in 1α-hydroxylase was developed through targeted ablation of the hormone-binding and hemebinding domains of the 1α-hydroxylase gene [4,13] (see Chapter 7). These mice also developed normally in the womb and as neonates, and only showed hypocalcemia, secondary hyperparathyroidism, retarded growth, and the skeletal abnormalities characteristic of rickets after weaning. These abnormalities are similar to those described in humans with the heritable disease vitamin D–dependent rickets type I [also known as pseudovitamin D–deficiency rickets (PDDR)] (see Chapter 7). Again, female mutant mice were infertile and exhibited uterine hypoplasia and absent corpora lutea. In the mice where vitamin D did not function well, infertility was observed, though not in all mice and not completely, indicating an essential role for vitamin D in fertility. It is not clear whether these effects of vitamin D are direct or involve the maintenance of calcium homeostasis. A discussion on the direct or indirect effect of vitamin D on fertility will be described in the next section.

As shown in the vitamin D–resistance models described earlier, fetal development is not generally altered under conditions of vitamin D deficiency. However, a subtle abnormality in the development of several organs, including reproductive organs, was reported. Kidney is an important organ for vitamin D activation, and its development is related to some extent to the genitals. Immunohistochemical techniques were used to examine the distribution of VDR in developing rat and mouse kidneys and murine metanephric organ culture [14]. This study showed that VDR was present in cells of branching ureteral buds and in the surrounding mesenchyme from gestational day 15, and at later developmental stages in glomerular visceral and parietal epithelial cells and proximal and distal tubules. Expression of the 28-kDa calcium-binding protein (calbindin D28k) gene, a target of vitamin D, was found exclusively in distal tubules from gestational day 19. Hence, calbindin D28k appears later in developing rat and mouse kidney and was distributed differently from that of VDR [14]. We also detected mRNA for VDR in kidneys from gestational day 13.5 using reverse transcription– polymerase chain reaction (RT-PCR) analysis [15]. RT-PCR and whole-mount in situ hybridization were then performed to examine the expression of 1α-hydroxylase and 24-hydroxylase, to investigate the role of VDR in vitamin D metabolism. In the absence of stimulants, the expression of 1α-hydroxylase and 24-hydroxylase was detected from day 13.5 of gestation. Forskolin and 1α,25(OH)2D3 induced the expression of 1α-hydroxylase and 24-hydroxylase, respectively, in a dose- and time-dependent manner. Signals for the expression of either 1α-hydroxylase (Fig. 1) or 24-hydroxylase were detected in kidney explants taken from embryos at 15.5 days of gestation after the appropriate stimulation, but the localization of signals differed between the two enzymes. The expression of both hydroxylases was restricted to the epithelium of developing renal tubules. A similar pattern of the expression of megalin, an endocytotic receptor, to that of 1α-hydroxylase was shown by whole-mount in situ hybridization. These results suggest that the expression of 1α-hydroxylase is induced in epithelium of renal tubules distinct from that of 24-hydroxylase even at the early stage of kidney development and prior to glomerulogenesis. This result is consistent with the idea that the roles of both enzymes in terms of vitamin D metabolism are separate. Megalin was shown to play an essential role in the activation of vitamin D in renal tubular cells by the uptake of 25-hydroxyvitamin D bound to vitamin D

CHAPTER 52 Vitamin D and Reproductive Organs

FIGURE 1

Expression of 1α-hydroxylase during early development. Whole-mount in situ hybridization for 1α-hydroxylase mRNA. Kidney explants obtained from mouse embryo at 15.5 days of gestation were incubated with 10−4 M forskolin for 6 hr and hybridized with the antisense probes for 1α-hydroxylase mRNA. See the detailed materials and methods described in [15].

binding protein based on findings in megalin-knockout mice [16]. Therefore, developmental analyses of the expression sites in kidneys suggest that the function of megalin may couple with that of the 1α-hydroxylase and that the function of VDR may couple with that of the 24-hydroxylase.

V. CALCIUM HOMEOSTASIS IN THE FETUS AND ITS MOTHER Before focusing on the direct or indirect effect of vitamin D on fertility, the characteristic state of calcium and calciotrophic hormones in fetal development, pregnancy, and lactation in the mother will be briefly reviewed [17,18]. During pregnancy and lactation, the level of calcium is dynamically regulated to ensure appropriate development of the skeleton in both the fetus and neonate. The fetus has a relatively high level of serum calcium, which decreases sharply at birth to below normal in infancy, because the active transport of calcium in the placenta is stopped. This decrease in serum calcium induces the secretion of parathyroid hormone (PTH), and PTH in turn increases the serum calcium concentration to the normal level. Active vitamin D is also necessary to maintain calcium and bone metabolism in the fetus and neonate at least in humans. Vitamin D deficiency causes hypocalcemia and rickets even early in the neonatal period in humans [19]. Hypovitaminosis D is sometimes associated

853 with pregnant women even in developed countries and leads to a congenital vitamin D deficiency in the baby. In pregnant women, renal 1,25(OH)2D3 production is stimulated, and there is evidence of 1,25(OH)2D3 production by the decidua/placenta and fetal kidney in vitro [20]. The renal 1α-hydroxylase activity is possibly induced by estrogens and PTH. It is likely that increased serum 1,25(OH)2D3 concentrations increase intestinal calcium absorption during pregnancy. PTH and 1,25(OH)2D3 levels decrease after delivery in the mother, but are increased when lactation is prolonged. The lα-hydroxylase activity may be stimulated by PTH and prolactin during lactation. What is responsible for the active transport of calcium in the placenta is not fully understood. Parathyroid hormone–related protein (PTHrP) is at least one of the factors needed for transport. PTHrP was originally cloned in 1987 as a causative factor of humoral hypercalcemia of malignancy, and named based on its structural similarity to PTH [21]. In fact, PTHrP shares with PTH a membrane-bound receptor (PTH1R) [22]. When secreted in excess, PTHrP causes hypercalcemia. However, PTHrP functions physiologically in the development of cartilage, mammary gland, heart, skin, hair follicles, tooth, pancreas, and kidney as a paracrine factor expressed throughout the body [23,24]. During the course of pregnancy, the serum concentration of PTHrP is elevated, and the elevation is maintained postpartum for several weeks [25]. Previous studies have shown that mice lacking either the PTHrP or the PTH1R gene exhibit severe chondrodysplasia [26]. In addition, in most genetic backgrounds, the PTH1Rnull mice die at midgestation in utero. The cause of death has been reported to be maldevelopment of the heart [27]. No abnormalities were observed in the yolk sac or placenta, implicating the degeneration of the heart as the primary cause of death. In PTH1R-null mice, an increase in placental calcium transfer was observed [27]. Because PTHrP levels are increased in PTH1R-null fetuses, it is presumed that elevated levels of PTHrP contribute to the increased transfer of calcium in the placenta. With its midregion sequence, PTHrP is able to promote active calcium transport in the placenta in the absence of PTH1R [28]. Hence the lack of PTHrP action may indirectly impair the mineralization of the skeleton in the fetus due both to a failure to actively transport calcium in the placenta and to the disruption to chondrocyte differentiation. PTHrP has also been shown to be necessary for the normal development of the mammary gland, and large amounts of PTHrP are found in human milk. As another participant in the active transport of calcium in the placenta, the calcium-sensing receptor (CaSR) has an essential role in monitoring calcium

854

KEIICHI OZONO, SHIGEO NAKAJIMA AND TOSHIMI MICHIGAMI

concentrations [29]. CaSR is expressed in both villous and extravillous regions of the human placenta [30]. CaSR expression was detected in both first-trimester and term placentas. In the villous region of the placenta, CaSR was detected in syncytiotrophoblasts and at lower levels in cytotrophoblasts. Local expression of CaSR in the brush border of syncytiotrophoblasts suggests a role for the maternal Ca concentration in the control of the active transport of calcium between the mother and fetus. In the extravillous region of the placenta, CaSR was detected in cells forming trophoblast columns in anchoring villi, in close proximity to maternal blood vessels, and in transitional cytotrophoblasts, suggesting an important role in controlling calcium levels of both mother and fetus in the development of the placenta and fetus [31]. CaSR may be a key molecule promoting the formation of the placenta because it may be involved in the control of the growth of placental trophoblasts. In addition, PTHrP is responsible for the active calcium transport in the placenta, suggesting that the interaction of PTHrP with CaSR plays an important role in regulating calcium homeostasis in the fetus. CaSR was found originally to regulate the secretion of PTH to control extracellular calcium concentrations in adults, but its role in fetal life remains to be elucidated. CaSR is expressed in the fetal period, but a postnatal increase in the expression of the CaSR gene has been observed in renal tubular cells [32]. A study on CaSR gene knockout mice revealed a role for CaSR in regulating fetal calcium metabolism [33]. Because of the active transport of calcium in the placenta, normal calcium concentrations in fetal blood are increased above the maternal level. The increase in fetal calcium levels depends upon PTHrP as described earlier. However, PTH and 1,25(OH)2D3 also play a role in maintaining serum calcium levels in the fetus. Heterozygous (+/−) and

homozygous (−/−) disruption of CaSR caused a further increase in the fetal calcium level [34]. This increase was modestly blunted or not blunted by concomitant disruption of the PTHrP gene and completely reversed by disruption of the PTH1R gene. These results suggest that PTH is important to the increase in serum calcium levels in these mutant mice. Actually, serum levels of PTH and 1,25(OH)2D3 were substantially increased above the normal low fetal levels by disruption of CaSR. On the other hand, placental calcium transfer was reduced and renal calcium excretion was increased by disruption of CaSR. Thus, increased levels of PTH and 1,25(OH)2D3 did not compensate for placental calcium transfer, suggesting a specific role of PTHrP in the transfer. These in vivo studies indicate that CaSR normally suppresses PTH secretion in the presence of the normal raised (and PTHrP-dependent) fetal calcium level. Disruption of CaSR causes fetal hyperparathyroidism and hypercalcemia, with additional effects on placental calcium transfer [33,35]. These findings indicate that CaSR plays an important role in fetal and neonatal calcium homeostasis by conducting the secretion of PTH.

VI. FERTILITY As described previously, fertility was reduced in vitamin D–deficient animals (Table I). Whether this results from direct or indirect effect of vitamin D remains controversial, however. This issue needs to be considered relative to the difference in species and sex. Initially, several experiments concerning vitamin D deficiency and fertility were done in chickens. For example, chickens were raised from 1 day of age to 8 weeks of age on a vitamin D–deficient diet to induce

TABLE I Infertility in Several Subjects with Defect of Vitamin D Actions Subject Vitamin D–deficient male chicken Vitamin D–deficient female rat Vitamin D–deficient female chicken Vitamin D–deficient male rat VDR-null mouse VDR-null mouse VDR-null mouse 1α-Hydroxylase-null mouse 1α-Hydroxylase-null mouse N.D., not done.

Infertility Infertile Infertile Infertile Reduced Died within 15 weeks Uterine hypoplasia Not mentioned Infertile Infertile (female) Uterine hypoplasia Not mentioned

Ca supplementation

Reference

N.D. Fertile N.D. N.D. N.D.

[36] [37] [38] [39] [3]

N.D. Fertile N.D.

[12] [41] [4]

N.D.

[13]

855

CHAPTER 52 Vitamin D and Reproductive Organs

a vitamin D–deficient state [36]. After 8 weeks, serum vitamin D levels were significantly lower in the chickens fed a vitamin D–deficient diet than those fed a normal diet. In male chickens, calbindin-D28K (CaBP28K) and testosterone levels were reduced in association with the deficiency. The morphology of the seminiferous tubules did not differ between the vitamin D–replete and vitamin D–deficient chickens. Concerning hens with vitamin D deficiency, they may lay fertile eggs but the embryos fail to hatch as described earlier. In vitamin D–deficient female rats, a reduction of overall reproductive capacity was observed. Vitamin D–deficient rats fed a high-calcium, high-phosphorus, 20% lactose diet had normal serum calcium, slightly lower phosphorus, and undetectable 25-hydroxyvitamin D levels [37]. Johnson and DeLuca reported that the decrease in reproductive capacity, previously seen in vitamin D–deficient rats, as indicated by the fertility ratio and pup number per litter, was completely corrected when serum calcium and phosphorus levels were normalized relative to vitamin D–replete rats. Based on these results, it seems likely that the diminished reproductive performance attributed to vitamin D deficiency is the result of hypocalcemia and/or hypophosphatemia caused by the deficiency [37]. In contrast, some studies suggest a specific and direct role of vitamin D in fertility. In one study, both fertility and reproductive capacity of female rats fed on vitamin D–deplete diet were decreased even when serum levels of calcium were not reduced [38]. These results support an essential role of vitamin D in reproduction via mechanisms other than hypocalcemia. Thus, the mechanism of action of 1,25(OH)2D3 separate from an increase in serum calcium levels remains to be determined, although the direct role of vitamin D in fertility remains controversial. Reduced fertility associated with vitamin D deficiency was also reported in male rats. Male weanling rats were fed a vitamin D–deficient or vitamin D-replete diet until maturity, and mated to age-matched, vitamin Dreplete females. Vitamin D–deficient males were found to be capable of reproduction. However, successful matings, i.e., the presence of sperm in the vaginal tract of the female, by vitamin D–deficient males were reduced by 45% when compared to matings by vitamin D–replete males [39]. Fertility (successful pregnancies in spermpositive females) was reduced by 73% in litters from vitamin D–deficient male inseminations when compared to litters from females inseminated by vitamin D–replete males [39]. These results supported the hypothesis that vitamin D and its metabolites are necessary for normal spermatogenesis and sperm function in the male rat. In addition to data based on vitamin D deficiency, the effect of vitamin D on infertility was also investigated

in new mouse models of vitamin D resistance. As described previously, infertility has been reported in VDR-null mutant mice. Hypoplasia of uterus and decreased sperm count associated with sperm motility are observed in female and male mutant mice, respectively. In addition, low activity of aromatase was also shown in these mice. Serum estradiol levels and sperm motility were partially reversed with calcium supplementation [40]. In another study, VDR-null mutant mice fed a normal diet were hypocalcemic and were found to be largely infertile (14% fertility), whereas the fertility level of normocalcemic VDR-null mutant mice and wild-type mice was between 86% and 100%. In these reports, a high-calcium diet led to 100% fertility in the VDRnull mice [41]. Thus, high dietary calcium levels are required for normal reproduction in VDR-null mutant female mice. These results suggest that the defect in reproduction reported previously for VDR-null mutant mice is not the lack of a direct effect of 1,25(OH)2D3 on reproductive function but is the result of hypocalcemia. These results support the idea of an indirect role of vitamin D for reproduction. Similar to observations in VDR-null mice, uterine hypoplasia and decreased ovarian size were also found in a mouse model for the genetic disorder vitamin D–dependent rickets type I in humans where the 1α-hydroxylase activity is impaired [13,42]. Panda et al. reported that folliculogenesis was also abnormal in the mutant mice [4]. Female mutant mice were infertile and exhibited uterine hypoplasia and absent corpora lutea. In addition, the null mutant mice were acyclic and did not ovulate. In a different study, the ovary and uterus of female 1α-hydroxylase-null mutant mice were smaller [4]. Histological analysis of female 1α-hydroxylase-null mutant mice showed a poorly developed endometrium in the hypoplastic uterine at 7 weeks. Ovaries of these mice were smaller than in wild-type mice, ovarian follicles were immature, interstitial tissue was increased, and there were no corpora lutea. This study supports the idea that 1,25(OH)2D3 is important for reproductive organ development and its deficiency results in infertility. However, male reproductive organs were grossly normal in 1α-hydroxylasenull mutant mice. In addition, the abnormality of reproductive organs was more severe in females than males. The reduced activity of aromatase with vitamin D deficiency, which is described in this chapter, may account for the female abnormality in fertility.

VII. TESTIS Since vitamin D–deficient male rats are infertile, vitamin D may exert its effect on the testis and

856

KEIICHI OZONO, SHIGEO NAKAJIMA AND TOSHIMI MICHIGAMI

involve spermatogenesis. In VDR-null mice, decreased sperm count and motility associated with histological abnormalities consisting of dilated lumen of seminiferous tubules and thinner layer of epithelial cells were observed in the male [40]. With respect to the existence of VDR in testis, evidence from autoradiographic studies with 1,25(OH)2D3 labeled with tritium and immunohistochemistry using monoclonal and polyclonal antibodies against VDR shows a broad distribution of VDR in the testes of animals [43]. In rat testes, VDR epitopes were observed in seminiferous tubules, specifically in spermatogonia, Sertoli cells, and spermatocytes, but spermatozoa were faintly stained. Epithelial cells of the epididymis, seminal vesicles, and prostate also expressed VDR epitopes. Biochemical data also revealed the VDR in testis, and its binding affinity for 1,25(OH)2D3 was 1.8 × 10−10 M [44]. These results suggest a nonclassical role for vitamin D in testis other than maintaining calcium and bone homeostasis. The role of vitamin D in testis is also supported by the finding of infertility in vitamin D–deficient animals and VDR or 1α-hydroxylase-deficient mice. The intracellular protein calbindin D28K is a primary target gene for vitamin D. The regulation of calbindin D28K expression by vitamin D has been intensively investigated in chick and rat intestine. Calbindin D28K exists in testis, and the effects of vitamin D deficiency on calbindin D28K and testosterone levels have been investigated in male chickens [36]. The study demonstrated that vitamin D deficiency exerted an effect on calbindin D28K expression in chicken testes. The morphology of the seminiferous tubules did not differ between the vitamin D–replete and vitamin D–deficient chickens. Immunohistochemical analysis revealed that calbindin D28K was present in spermatogonia and spermatocytes of the seminiferous tubules. A few interstitial Leydig cells were positive for calbindin D28K. The amount of calbindin D28K was qualified by radioimmunoassay in the testes and was found to be threefold higher in chickens raised on a normal diet than in those raised on a vitamin D–deficient diet. These results indicate that the decrease in the testicular calbindin D28K concentration might be attributable to vitamin D deficiency despite normal serum testosterone and calcium levels in 8-week-old chickens.

immunoreactivity, while the remaining 16.7% of the normal surface ovarian epithelium was VDR negative [45]. Ovarian calcinoma cells also expressed VDR, and the intensity of VDR immunostaining was significantly increased in ovarian carcinomas as compared to normal ovarian tissue [45]. Concerning the distribution of VDR in ovary, immunostaining for VDR was seen in ovarian follicles, specifically in granulosa cells. Weaker immunostaining of VDR was observed in follicular thecal cells and in the ovarian stroma and germinal epithelium. Corpus luteal cells stained intensely for VDR. Epithelium of fallopian tubes and the uterus also expressed VDR. Hence, both nuclear and cytoplasmic VDR immunostaining was observed in female rat reproductive tissues as detected in testis in male rats. These results suggest that vitamin D may function in the maturation of ovocytes, ovulation, and reproduction. The relationship between ovarian cancer and VDR is another issue to be considered. On analyzing the coexpression of VDR with the proliferation marker Ki-67 or with estrogen and progesterone receptors, no significant correlation was found with ovarian cancer cells [45]. The expression of VDR with ovarian carcinomas seemed to be regulated independently from that of the estrogen receptor or progesterone receptor. VDR expression is generally increased in ovarian carcinomas as compared to normal ovarian tissue, suggesting a role for vitamin D in ovarian cell proliferation [45].

VIII. OVARY Immunohistochemical studies using monoclonal antibody against VDR demonstrated that VDR was expressed in both normal and carcinomatous ovarian tissues. Villena-Heisen and colleagues reported that a total of 83.3% exhibited weak to moderate VDR

IX. UTERUS During decidualization, endometrial cells proliferate rapidly, resulting in an increase in uterine weight. At the same time, endometrial cells differentiate into decidual cells. This process is called a decidual reaction and is usually associated with ovum implantation. Because endometrial cells and decidual cells possess VDR and respond to 1,25(OH)2D3 (induction of 24-hydroxylase), the decidual process may be affected by 1,25(OH)2D3. However, the characteristics of endometrial and decidual cells are distinct. Halhali and colleagues reported that intraluminal injection of female rats with 10–500 ng of 1,25(OH)2D3 on day 5 of pseudopregnancy significantly increased uterine weight and induced a decidual reaction [46]. This effect was observed as early as the third day after the injection. These results suggest that 1,25(OH)2D3 plays a physiological role in the differentiation of endometrial cell into decidual cells, a crucial step in pregnancy. Indeed, although renal tubular cells are the main site of expression of 1alpha-hydroxylase, human decidual cells are also reported to produce 1,25(OH)2D3, particularly at the end of pregnancy. A study using stromal decidual

CHAPTER 52 Vitamin D and Reproductive Organs

cells showed that human uterine cells were capable of synthesizing 1,25(OH)2D3 even during early pregnancy [47]. On the other hand, the synthesis of 1,25(OH)2D3 by endometrial cells is controversial. As the main thesis in the chapter, poor reproductive performance was observed in rats deficient in vitamin D. On the other hand, vitamin D toxicity was also associated with poor reproductive activity reflected in a decrease in the number of matings, implantations, and live births [48]. These changes were reversible, and recovery was observed after treatment with active vitamin D was discontinued. In order to clarify the mechanism of these reversible toxicities, female rats treated with the vitamin D3 metabolite were compared with untreated rats with respect to several reproductive parameters [48]. The estrous cycle was disturbed in vitamin D–treated rats. Also, hypofunctional changes in the corpus luteum in the ovary and the epithelium, endometrium, and uterine gland in the uterus with a decrease in the serum progesterone level were also observed. In addition, vitamin D–induced hypercalcemia decreased calcitonin or PTH levels in serum with morphological changes including atrophy and cyst formation in the parathyroid. However, these changes were reversible, and a recovery was observed after administration of the compound was discontinued. These results indicate that the hypercalcemia caused by 1,25(OH)2D3 disrupts endocrinological homeostasis, which in turn temporarily disrupts the female reproductive system. Cervical cells in the uterus are another possible target of vitamin D, and cervical cell carcinoma is one of the most common cancers in women. The proliferation and differentiation of cervical epithelial cells are mainly regulated by estrogens and progestins. 1,25(OH)2D3 is reported to be an important determinant of the responsiveness to these hormones [49]. 1,25(OH)2D3 induced the expression of insulin-like growth factor-binding protein 3 and inhibited the growth of human ectocervical epithelial cells [49].

X. MAMMARY GLAND Development of the mammary gland occurs predominantly postnatally, and the morphogenesis of mammary gland is achieved through the coordination of signaling networks in both epithelial and stromal cells [50]. While the major proliferative hormones driving pubertal mammary gland development are estrogen and progesterone, studies in transgenic and knockout mice have successfully identified other steroid and peptide hormones that affect development. VDR with an active vitamin D metabolite has been implicated in the control of the differentiation, cell cycle, and apoptosis of

857 mammary cells in culture, but little was known about the physiological relevance of the vitamin D endocrine system in the developing mammary gland in vivo. It was reported that VDR was expressed in epithelial cells of the terminal end bud and subtending ducts, in stromal cells, and in a subset of lymphocytes within the lymph node [50]. In the terminal end bud, a distinct gradient of VDR expression has been observed, with weak staining in proliferative populations and strong staining in differentiated populations. The role of VDR in ductal morphogenesis was examined in VDR-null mice fed high dietary calcium, which normalizes fertility, serum estrogen levels, and neonatal growth [51]. The results of the study indicate that mammary glands from virgin VDR-null mice are heavier and exhibit accelerated growth, as evidenced by higher numbers of terminal end buds, greater ductal outgrowth, and enhanced secondary branch points, compared with glands from age- and weight-matched wild-type mice both in vivo and in organ culture. In addition, glands from VDR-null mice exhibit enhanced growth in response to exogenous estrogen and progesterone compared with glands from wild-type mice. These results provided in vivo evidence that 1,25(OH)2D3 and VDR impact ductal elongation and branching morphogenesis during pubertal development of the mammary gland [51]. Given these results, the vitamin D signaling pathway may participate in negative growth regulation of the mammary gland. 1,25(OH)2D3 interacts with VDR to modulate proliferation and apoptosis in a variety of cell types, including breast cancer cells. It is interesting that mammary glands from VDR-null mice exhibit accelerated growth and branching during puberty, pregnancy, and lactation as compared to wild-type mice [51]. In addition, involution after weaning, a process driven by epithelial cell apoptosis, proceeds at a slower rate in VDR-null mice compared to wild-type mice. These in vivo findings were further evaluated using cells isolated from VDR-null and wild-type mice. In these experiments, the growth of both normal and transformed mammary cells derived from wild-type mice was inhibited by 1,25(OH)2D3. In contrast, cells derived from VDR-null mice did not respond to 1,25(OH)2D3, indicating VDR was responsible for the growth inhibition by 1,25(OH)2D3. In addition to normal development of mammary gland and proliferation of mammary cells, 1,25(OH)2D3 and its various analogs have been shown to inhibit proliferation in human breast cancer cells and in experimental mammary tumors in vivo and in vitro. Epidemiological studies have suggested an association between 1,25(OH)2D3 deficiency and increased risk of various malignancies including cancer of the colon and breast [52]. 1,25(OH)2D3-mediated growth inhibition

858

KEIICHI OZONO, SHIGEO NAKAJIMA AND TOSHIMI MICHIGAMI

was enhanced by the addition of a variety of agents, including steroid hormones, phytoestrogens, and growth factors that up-regulate VDR expression [51]. These results support a role for 1,25(OH)2D3 and VDR in the negative growth regulation of both normal mammary gland and breast cancer cells. Loss of VDR in breast cancer cells, however, does not always lead to increased aggressiveness. Nonetheless, 1,25(OH)2D3 inhibits the growth of most breast cancer cells both in vivo and in vitro. Concerning the mechanisms of growth inhibition of mammary cells and breast cancer cells by 1,25(OH)2D3, an assessment of 1,25(OH)2D3 increased growth arrest and apoptosis was made using cell lines established from DMBA-induced mammary tumors derived from VDR-null and wild-type mice [53]. Zinser and colleagues obtained two VDR-null and two wild-type cell lines, and characterized the growth of these cells in response to 1,25(OH)2D3 [53]. Both wildtype cell lines expressed the VDR protein and were sensitive to growth inhibition by 1,25(OH)2D3 at doses as low as 1 nM. 1,25(OH)2D3 induced G(0)/G(1) arrest and apoptosis in the wild-type cell lines. In contrast, both cell lines from VDR-null mice were completely resistant to 1,25(OH)2D3-mediated growth arrest and apoptosis even when the high concentration of 1,25(OH)2D3 was used. It was confirmed that both cells established from tumors that developed in VDRnull mice lacked VDR mRNA and protein. Therefore, the induction of cell cycle arrest and apoptosis in breast cancer cells by 1,25(OH)2D3 is dependent on the nuclear VDR. Cells lacking VDR remain sensitive to growth arrest mediated by 9-cis-retinoic acid, a ligand for the retinoid X receptor (RXR) that can heterodimerize with VDR. Sensitivity to apoptosis induced by the DNA-damaging agent etoposide was not altered in VDR-null cells, indicating that VDR ablation did not impair apoptotic pathways in general.

tissue is a significant source of estrogen in men and postmenopausal women. The CYP19 gene was cloned in 1989 and found to consist of 10 exons with a variety of tissue-specific promoters [54]. In aromatase-deficient patients, an estrogen deficiency appeared prenatally. The phenotype of these patients indicates the significant role of aromatase for the development and functions of reproductive organs in humans. In men with an aromatase deficiency, male reproductive functions are impaired. Because estrogens play a pivotal role in the control of serum gonadotropin concentrations in the male, male rodents with an aromatase defect actually show impaired sexual behavior and fertility as a consequence of this estrogen defect [56]. In contrast, the patients who had gain-of-function mutations in the CYP19 gene have been described [57]. The patients showed gynecomastia and relatively low levels of testosterone and follicle-stimulating hormone caused by severe estrogen excess. In a 46,XX female with aromatase-deficiency due to molecular defects in the CYP19 (P450arom) gene, a nonadrenal form of pseudohermaphrodism appeared at birth because of the excess of androgen and deficiency of estrogen [58]. The fetal masculinization in this syndrome is likely to be caused by exposure of the female fetus to excessive amounts of testosterone as a result of the defective placental conversion of C19 steroids to estrogens. According to the case report of a female patient with aromatase deficiency, bone aging was delayed. This finding is consistent with the idea that estrogens, in contrast to androgens, are the major sex steroid driving skeletal maturation during puberty. Moreover, estrogen replacement therapy was effective in the growth spurt, breast development, menarche, suppression of gonadotropin levels, and resolution of cysts [58]. Pubertal failure, mild virilization, multicystic ovaries, and hyperstimulation of the ovaries by FSH and LH all result from the inability of the ovary to convert testosterone and androstenedione into estrogens. As described in the first paragraph of this section, the CYP19 gene employs several different promoters (I.1, I.2, I.3, I.4, I.5, I.6, 2a, 1f, and PII) to control expression in a tissue-specific fashion. Although the regulation of aromatase gene expression is complex, one of the main regulators, especially in the ovary, is protein kinase A. It is suggested that a nuclear receptor system comprising the RAR-RXR heterodimer is also involved in the regulation of aromatase activity in MCF-7 breast cancer cells [59]. On the other hand, the PPARgamma ligand troglitazone or RXR ligand LG100268 alone decreased aromatase activity in cultured human ovarian granulosa cells, while combined treatment with troglitazone and LG100268 caused an even greater reduction in this activity. This suggests that the PPARgamma–RXR

XI. AROMATASE Aromatase, the product of the CYP19 gene, is a P-450 containing enzyme that is involved in estrogen biosynthesis [54]. It is a key enzyme that confers both endocrine and paracrine actions of estrogen. In contrast to estrogen, the production of which is limited to the gonads and brain in most vertebrate species, aromatase is expressed in various tissues and cells including adipose stromal cells and syncytiotrophoblasts [55]. In men, estrogen is produced from androgens in Leydig cells of the testis. In women the primary synthetic sites are granulose and luteal cells of the ovary. At the same time, peripheral conversion in adipose

859

CHAPTER 52 Vitamin D and Reproductive Organs

heterodimer system is involved in the regulation of the aromatase gene expression in human granulosa cells. Therefore, RXR is involved in the regulation of the aromatase gene expression in a unique manner in different tissues, i.e., as an RAR–RXR heterodimer in human breast cancer cells, and as a PPARgamma–RXR heterodimer in ovarian granulosa cells. The role of vitamin D in the regulation of estrogen synthesis in gonads was also investigated in a mouse model [40]. VDR-null mutant mice showed gonadal insufficiencies: Uterine hypoplasia and impaired folliculogenesis were observed in the female, and a decreased sperm count and decreased motility with histological abnormality of the testis were observed in the male [40]. Aromatase activity in these mice was also reported to be weak in the ovary, testis, and epididymis at 24%, 58%, and 35% of the wild-type values, respectively. The expression of aromatase itself was also reduced in these organs. Elevated serum levels of LH and FSH revealed hypergonadotropic hypogonadism in these mice. However, in response to supplementation with estradiol, histological abnormalities were normalized in both male and female gonads. Interestingly, calcium supplementation increased aromatase activity and partially corrected the hypogonadism, but LH and FSH levels remained high. Hence, the action of vitamin D on estrogen biosynthesis was partially explained by maintaining calcium homeostasis, although a direct effect of active vitamin D on the expression of the aromatase gene was also apparent [40]. These results are consistent with the idea that vitamin D is essential for full gonadal function in both sexes. As described previously, the tissue-specific expression of the CYP19 gene is regulated by means of tissue-specific promoters through the use of alternative splicing mechanisms. Thus, transcripts containing various 5′-untranslated termini are present in ovary, brain, adipose stromal cells, and placenta [55]. Sun and co-workers reported that VDR is involved in the control of the promoter I.1 of the CYP19 gene [55]. This promoter drives expression of the gene in human placenta and choriocarcinoma cells. Various deletion mutations of the upstream flanking region of exon I.1 were constructed and transfected into human choriocarcinoma (JEG3) cells to examine the region regulated by VDR or retinoic acid receptor (RAR) activity. The longest construct, −924/+10 bp, exhibited the highest level of luciferase reporter gene activity [55]. Interestingly, this activity was induced by vitamin D as well as by LG69 and TTNPB, ligands specific for RXR and RAR respectively. The imperfect palindromic sequence (AGGTCATGCCCC) located at −183 to −172 bp upstream of the transcriptional start site of exon I.1 was found to be important for both basal and retinoid-induced reporter

gene expression. Gel retardation analysis using nuclear extracts of JEG3 cells and the imperfect palindromic sequence as a probe showed the formation of a heterodimer of RXR alpha and VDR. These results suggest that the imperfect palindromic sequence upstream of exon I.1 represents a novel VDRE.

XII. PLACENTA Maternal calcium regulation is adapted to allow fetal growth through enhanced intestinal calcium absorption rather than through the mobilization of maternal skeletal reserves [17]. This adaptive process depends mainly on the two major calciotrophic hormones, PTH and 1,25(OH)2D3, which show quantitative changes from the nonpregnant state. Maternal plasma levels of 1,25(OH)2D3 are elevated during pregnancy in humans and experimental animals [17]. Placenta is thought to be a site for the production of 1,25(OH)2D3 and some reports described 1α-hydroxylase activity here. The expression of the 1α-hydroxylase gene is controversial, however [60]. Nevertheless, in patients with a lack of renal 1α-hydroxylase activity, converting activity was also absent in cells isolated from the decidua of patients with vitamin D-dependency type 1, suggesting that the decidual and renal enzymes are encoded by the same gene [61]. Characterization of VDR in placenta was performed; the molecular weight of the receptor was estimated to be 55 kDa based on gel filtration [62]. The binding affinity of the receptor for 1,25(OH)2D3 was 3.0 × 10−10 M. These results indicate that the properties of the 1,25(OH)2D3 receptor in human placenta are similar to those of the chicken intestinal and human osteoblastic VDR. In contrast to maternal mineral homeostasis, the maintenance of fetal calcium homeostasis depends largely on PTHrP, which regulates active placental calcium transfer and the calcium fluxes across the kidney and bone. The major source of PTHrP is the fetal parathyroid gland and placenta, although some organs including bone and cartilage are able to synthesize PTHrP. The mechanism of active calcium transport by PTHrP remains to be elucidated, although PTHR1 exists in placenta. The relationship between intracellular cAMP and calcium transport needs to be elucidated.

XIII. CONCLUDING REMARKS For complete reproduction, a complex and harmonic regulation of the expression of entire sets of genes is necessary. A copy of the genome is reserved in the gamete, and development begins after the male

860

KEIICHI OZONO, SHIGEO NAKAJIMA AND TOSHIMI MICHIGAMI

and female germ cells are joined. During the process of fetal development, calcium is dynamically mobilized from mother to fetus, and there appears to be a specific system that is involved, for example, in active transport of calcium in the placenta. Vitamin D clearly has a role in reproduction because vitamin D deficiency leads to infertility in chickens and mammals. An indirect effect of vitamin D that is mediated by serum calcium levels appears to play a major part, but any direct effect is still to be considered. The establishment of mouse models for certain types of vitamin D resistance will contribute to the elucidation of the mechanism of infertility under conditions of defective vitamin D action.

12. Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997 Targeted ablation of the vitamin D receptor: an animal model of vitamin D-dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94:9831–9835. 13. Dardenne O, Prud’homme J, Arabian A, Glorieux FH, StArnaud R 2001 Targeted inactivation of the 25-hydroxyvitamin D3-1α-hydroxylase gene (CYP27B1) creates an animal model of pseudovitamin D–deficiency rickets. Endocrinology 142:3135–3141. 14. Johnson JA, Grande JP, Roche PC, Sweeney WE Jr, Avner ED, Kumar R 1995 1α,25-Dihydroxyvitamin D3 receptor ontogenesis in fetal renal development. Am J Physiol 269:F419–F428. 15. Yamagata M, Kimoto A, Michigami T, Nakayama M, Ozono K 2001 Hydroxylases involved in vitamin D metabolism are differentially expressed in murine embryonic kidney: application of whole mount in situ hybridization. Endocrinology 142: 3223–3230. 16. Nykjaer A, Dragun D, Walther D, Vorum H, Jacobsen C, Herz J, Melsen F, Christensen EI, Willnow TE 1999 An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) vitamin D3. Cell 96:507–515. 17. Hosking DJ 1996 Calcium homeostasis in pregnancy. Clin Endocrinol (Oxf) 45:1–6. 18. Verhaeghe J, Bouillon R 1992 Calciotropic hormones during reproduction. J Steroid Biochem Mol Biol 41:469–477. 19. Dawodu A, Agarwal M, Hossain M, Kochiyil J, Zayed R 2003 Hypovitaminosis D and vitamin D deficiency in exclusively breast-feeding infants and their mothers in summer: a justification for vitamin D supplementation of breast-feeding infants. J Pediatr 142:169–173. 20. Stumpf WE, Denny ME 1989 Vitamin D (soltriol), light, and reproduction. Am J Obstet Gynecol 161:1375–1384. 21. Suva LJ, Winslow GA, Wettenhall RE, Hammonds RG, Moseley JM, Diefenbach-Jagger H, Rodda CP, Kemp BE, Rodriguez H, Chen EY, Hudson PJ, Martin TJ, Wood WI 1987 A parathyroid hormone-related protein implicated in malignant hypercalcemia: cloning and expression. Science 237:893–896. 22. Abou-Samra AB, Juppner H, Force T, Freeman MW, Kong XF, Schipani E, Urena P, Richards J, Bonventre JV, Potts JT Jr, Kronenberg HM, Segre GV 1992 Expression cloning of a common receptor for parathyroid hormone and parathyroid hormone-related peptide from rat osteoblast-like cells: a single receptor stimulates intracellular accumulation of both cAMP and inositol trisphosphates and increases intracellular free calcium. Proc Natl Acad Sci USA 89:2732–2736. 23. Orloff JJ, Reddy D, de Papp AE, Yang KH, Soifer NE, Stewart AF 1994 Parathyroid hormone-related protein as a prohormone: posttranslational processing and receptor interactions. Endocr Rev 15:40–60. 24. Karaplis AC 2001 PTHrP: novel roles in skeletal biology. Curr Pharm Des 7:655–670. 25. Ardawi MSM, Nasrat HA, BA’Aqueel HS 1997 Calciumregulating hormones and parathyroid hormone-related peptide in normal human pregnancy and postpartum: a longitudinal study. Eur J Endocrinol 137:402–409. 26. Escande B, Lindner V, Massfelder T, Helwig JJ, Simeoni U 2001 Developmental aspects of parathyroid hormone-related protein biology. Semin Perinatol 25:76–84. 27. Qian J, Colbert MC, Witte D, Kuan CY, Gruenstein E, Osinska H, Lanske B, Kronenberg HM, Clemens TL 2003 Midgestational lethality in mice lacking the parathyroid hormone (PTH)/PTH-related peptide receptor is associated with abrupt cardiomyocyte death. Endocrinology 144: 1053–1061.

References 1. Jones G, Strugnell SA, DeLuca HF 1998 Current understanding of the molecular actions of vitamin D. Physiol Rev 78:1193–1231. 2. Henry HL, Norman AW 1978 Vitamin D: two dihydroxylated metabolites are required for normal chicken egg hatchability. Science 201:835–837. 3. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Genet 16:391–396. 4. Panda DK, Miao D, Tremblay ML, Sirois J, Farookhi R, Hendy GN, Goltzman D 2001 Targeted ablation of the 25hydroxyvitamin D 1α-hydroxylase enzyme: evidence for skeletal, reproductive, and immune dysfunction. Proc Natl Acad Sci USA 98:7498–7503. 5. Brunvand L, Shah SS, Bergstrom S, Haug E 1998 Vitamin D deficiency in pregnancy is not associated with obstructed labor. A study among Pakistani women in Karachi. Acta Obstet Gynecol Scand 77:303–306. 6. Yoshimura Y, Ohira H, Tamura T 1997 Immunocytochemical localization of vitamin D receptors in the shell gland of immature, laying, and molting hens. Gen Comp Endocrinol 108:282–289. 7. Elaroussi MA, Uhland-Smith A, Hellwig W, DeLuca HF 1994 The role of vitamin D in chorioallantoic membrane calcium transport. Biochim Biophys Acta 1192:1–6. 8. Norman AW, Leathers V, Bishop JE 1983 Normal egg hatchability requires the simultaneous administration to the hen of 1α,25-dihydroxycholecalciferol and 24R,25-dihydroxycholecalciferol. J Nutr 113:2505–2515. 9. Boyan BD, Sylvia VL, Curry D, Chang Z, Dean DD, Schwartz Z 1998 Arachidonic acid is an autocoid mediator of the differential action of 1,25-(OH)2D3 and 24,25-(OH)2D3 on growth plate chondrocytes. J Cell Physiol 176:516–524. 10. Nakamura T, Suzuki K, Hirai T, Kurokawa T, Orimo H 1992 Increased bone volume and reduced bone turnover in vitamin D–replete rabbits by the administration of 24R,25-dihydroxyvitamin D3. Bone 13:229–236. 11. Ohyama Y, Ozono K, Uchida M, Shinki T, Kato S, Suda T, Yamamoto O, Noshiro M, Kato Y 1994 Identification of a vitamin D-responsive element in the 5′-flanking region of the rat 25-hydroxyvitamin D3 24-hydroxylase gene. J Biol Chem 269:10545–10550.

861

CHAPTER 52 Vitamin D and Reproductive Organs

28. Kovacs CS, Lanske B, Hunzelman JL, Guo J, Karaplis AC, Kronenberg HM 1996 Parathyroid hormone-related peptide (PTHrP) regulates fetal-placental calcium transport through a receptor distinct from the PTH/PTHrP receptor. Proc Natl Acad Sci USA 93:15233–15238. 29. Brown EM, MacLeod RJ 2001 Extracellular calcium sensing and extracellular calcium signaling. Physiol Rev 81:239–297. 30. Bradbury RA, Cropley J, Kifor O, Lovicu FJ, de Iongh RU, Kable E, Brown EM, Seely EW, Peat BB, Conigrave AD 2002 Localization of the extracellular Ca2+-sensing receptor in the human placenta. Placenta 23:192–200. 31. Kovacs CS, Ho-Pao CL, Hunzelman JL, Lanske B, Fox J, Seidman JG, Seidman CE, Kronenberg HM 1998 Regulation of murine fetal-placental calcium metabolism by the calciumsensing receptor. J Clin Invest 101:2812–2820. 32. Riccardi D, Hall AE, Chattopadhyay N, Xu JZ, Brown EM, Hebert SC 1998 Localization of the extracellular Ca2+/polyvalent cation-sensing protein in rat kidney. Am J Physiol 274:F611–F622. 33. Ho C, Conner DA, Pollak MR, Ladd DJ, Kifor O, Warren HB, Brown EM, Seidman JG, Seidman CE 1995 A mouse model of human familial hypocalciuric hypercalcemia and neonatal severe hyperparathyroidism. Nat Genet 11:389–394. 34. Kovacs CS, Manley NR, Moseley JM, Martin TJ, Kronenberg HM 2001 Fetal parathyroids are not required to maintain placental calcium transport. J Clin Invest 107:1007–1015. 35. Clemens TL, Cormier S, Eichinger A, Endlich K, FiaschiTaesch N, Fischer E, Friedman PA, Karaplis AC, Massfelder T, Rossert J, Schluter KD, Silve C, Stewart AF, Takane K, Helwig JJ 2001 Parathyroid hormone-related protein and its receptors: nuclear functions and roles in the renal and cardiovascular systems, the placental trophoblasts and the pancreatic islets. Br J Pharmacol 134:1113–1136. 36. Inpanbutr N, Reiswig JD, Bacon WL, Slemons RD, Iacopino AM 1996 Effect of vitamin D on testicular CaBP28K expression and serum testosterone in chickens. Biol Reprod 54:242–248. 37. Johnson LE, DeLuca HF 2002 Reproductive defects are corrected in vitamin D-deficient female rats fed a high calcium, phosphorus and lactose diet. J Nutr 132:2270–2273. 38. Hickie JP, Lavigne DM, Woodward WD 1983 Reduced fecundity of vitamin D deficient rats. Comp Biochem Physiol A 74:923–925. 39. Kwiecinski GG, Petrie GI, DeLuca HF 1989 Vitamin D is necessary for reproductive functions of the male rat. J Nutr 119:741–744. 40. Kinuta K, Tanaka H, Moriwake T, Aya K, Kato S, Seino Y 2000 Vitamin D is an important factor in estrogen biosynthesis of both female and male gonads. Endocrinology 141:1317–1324. 41. Johnson LE, DeLuca HF 2001 Vitamin D receptor null mutant mice fed high levels of calcium are fertile. J Nutr 131:1787–1791. 42. Dardenne O, Prud’homme J, Hacking SA, Glorieux FH, St-Arnaud R 2003 Correction of the abnormal mineral ion homeostasis with a high-calcium, high-phosphorus, highlactose diet rescues the PDDR phenotype of mice deficient for the 25-hydroxyvitamin D-1α-hydroxylase (CYP27B1). Bone 32:332–340. 43. Johnson JA, Grande JP, Roche PC, Kumar R 1996 Immunohistochemical detection and distribution of the 1,25dihydroxyvitamin D3 receptor in rat reproductive tissues. Histochem Cell Biol 105:7–15. 44. Habib FK, Maddy SQ, Gelly KJ 1990 Characterisation of receptors for 1,25-dihydroxyvitamin D3 in the human testis. J Steroid Biochem 35:195–199. 45. Villena-Heinsen C, Meyberg R, Axt-Fliedner R, Reitnauer K, Reichrath J, Friedrich M 2002 Immunohistochemical analysis

46. 47.

48. 49.

50. 51.

52.

53. 54. 55.

56.

57.

58.

59. 60.

61.

62.

of 1,25-dihydroxyvitamin-D3-receptors, estrogen and progesterone receptors and Ki-67 in ovarian carcinoma. Anticancer Res 22:2261–2267. Halhali A, Acker GM, Garabedian M 1991 1,25Dihydroxyvitamin D3 induces in vivo the decidualization of rat endometrial cells. J Reprod Fertil 91:59–64. Kachkache M, Rebut-Bonneton C, Demignon J, Cynober E, Garabedian M 1993 Uterine cells other than stromal decidual cells are required for 1,25-dihydroxyvitamin D3 production during early human pregnancy. FEBS Lett 333:83–88. Horii I, Takizawa S, Fujii T 1992 Effect of 1,25-dihydroxyvitamin D3 on the female reproductive system in rats. J Toxicol Sci 17:91–105. Agarwal C, Lambert A, Chandraratna RA, Rorke EA, Eckert RL 1999 Vitamin D regulates human ectocervical epithelial cell proliferation and insulin-like growth factor-binding protein-3 level. Biol Reprod 60:567–572. Zinser G, Packman K, Welsh J 2002 Vitamin D3 receptor ablation alters mammary gland morphogenesis. Development 129: 3067–3076. Welsh J, Wietzke JA, Zinser GM, Smyczek S, Romu S, Tribble E, Welsh JC, Byrne B, Narvaez CJ 2002 Impact of the Vitamin D3 receptor on growth-regulatory pathways in mammary gland and breast cancer. J Steroid Biochem Mol Biol 83:85–92. Garland FC, Garland CF, Gorham ED, Young JF 1990 Geographic variation in breast cancer mortality in the United States: A hypothesis involving exposure to solar radiation. Prev Med 19:614–622. Zinser GM, McEleney K, Welsh J 2003 Characterization of mammary tumor cell lines from wild type and vitamin D3 receptor knockout mice. Mol Cell Endocrinol 28(200):67–80. Meinhardt U, Mullis PE 2002 The aromatase cytochrome P-450 and its clinical impact. Horm Res 57:145–152. Sun T, Zhao Y, Mangelsdorf DJ, Simpson ER 1998 Characterization of a region upstream of exon I.1 of the human CYP19 (aromatase) gene that mediates regulation by retinoids in human choriocarcinoma cells. Endocrinology 139:1684–1691. Rochira V, Balestrieri A, Madeo B, Baraldi E, Faustini-Fustini M, Granata AR, Carani C 2001 Congenital estrogen deficiency: in search of the estrogen role in human male reproduction. Mol Cell Endocrinol 178:107–115. Shozu M, Sebastian S, Takayama K, Hsu WT, Schultz RA, Neely K, Bryant M, Bulun SE 2003 Estrogen excess associated with novel gain-of-function mutations affecting the aromatase gene. N Engl J Med 348:1855–1865. Conte FA, Grumbach MM, Ito Y, Fisher CR, Simpson ER 1994 A syndrome of female pseudohermaphrodism, hypergonadotropic hypogonadism, and multicystic ovaries associated with missense mutations in the gene encoding aromatase (P450arom). J Clin Endocrinol Metab 78:1287–1292. Yanase T, Mu YM, Nishi Y, Goto K, Nomura M, Okabe T, Takayanagi R, Nawata H 2001 Regulation of aromatase by nuclear receptors. J Steroid Biochem Mol Biol 79:187–192. Fu GK, Lin D, Zhang MY, Bikle DD, Shackleton CH, Miller WL, Portale AA 1997 Cloning of human 25-hydroxyvitamin D-1α-hydroxylase and mutations causing vitamin D–dependent rickets type 1. Mol Endocrinol 11:1961–1970. Glorieux FH, Arabian A, Delvin EE 1995 Pseudo-vitamin D deficiency: Absence of 25-hydroxyvitamin D 1α-hydroxylase activity in human placenta decidual cells. J Clin Endocrinol Metab 80:2255–2258. Ross R, Florer J, Halbert K, McIntyre L 1989 Characterization of 1,25-dihydroxyvitamin D3 receptors and in vivo targeting of [3H]-1,25(OH)2D3 in the sheep placenta. Placenta 10:553–567.

CHAPTER 53

Vitamin D Receptor as a Sensor for Toxic Bile Acids DAVID J. MANGELSDORF AND DANIEL L. MOTOLA Howard Hughes Medical Institute, Department of Pharmacology, University of Texas Southwestern Medical Center at Dallas, Texas

I. Introduction II. Bile Acids: Physiologic Roles in Lipid Digestion and Absorption III. The Fate of Bile Acids IV. Nuclear Receptors: Key Regulators of Bile Acid Metabolism

V. Bile Acids, Vitamin D Receptor, and Colon Cancer VI. A Model of Lithocholic Acid Detoxification in the Intestine VII. Perspectives References

I. INTRODUCTION

The rate-limiting step catalyzing this process is modulated by a feedforward loop that controls the conversion of cholesterol into bile acids and a feedback loop that limits bile acid production. Both the feedforward and feedback loops are governed by a unique sensing apparatus that involves several nuclear hormone receptors [3,4] (see section IV). After synthesis by the liver, the majority of bile acids are conjugated to either glycine or taurine before being secreted through the bile canuliculi of the liver for storage in the gallbladder and secretion into the duodenum. Conjugation converts bile acids into a stronger acid and thus at a physiological pH, conjugated bile acids become fully ionized, membrane impermeable, and more water soluble [5,6]. As they reach high concentrations within the intestine they begin to solubilize dietary lipids through the formation of mixed micelles [5,6]. The center of the mixed micelle is essentially a fat globule while the outside is charged because of the ionized sterol nucleus of bile acids. Thus, the mixed micelles are free to associate with the water layer adjacent to the surface mucosa of the intestine to allow for the absorption of dietary lipids, cholesterol, and lipid-soluble vitamins.

In this chapter we highlight recent findings that build on our current understanding of the function of the vitamin D receptor (VDR) within the enterohepatic system. These findings show that in the intestine, VDR acts as a sensor of carcinogenic bile acids and induces their catabolism by up-regulating expression of key detoxifying enzymes [1]. Specifically, VDR binds a toxic secondary bile acid, lithocholic acid (LCA), a bioactive metabolite believed to play a role in the development of colon cancer. Binding of either vitamin D or LCA to VDR leads to receptor activation and induces the detoxification of LCA in the intestine through up-regulating the expression of a cytochrome P450, CYP3A [1]. These findings provide a new understanding of the ability of the enteric system to protect itself against carcinogenic bile acids and underscore the importance of developing new pharmacological therapies that target VDR for the prevention and treatment of colon cancer. A brief review of bile acid physiology along with the details and important implications of this discovery are discussed next.

II. BILE ACIDS: PHYSIOLOGIC ROLES IN LIPID DIGESTION AND ABSORPTION Bile acids are produced by the liver and are secreted into the small intestine where they serve several important functions. Because of their amphipathic properties, bile acids are essential as detergents that solubilize and facilitate absorption of dietary fats. The primary bile acids, cholic acid (CA) and chenodeoxycholic acid (CDCA), are produced in the liver from cholesterol through a series of more than 17 enzymatic steps [2]. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

III. THE FATE OF BILE ACIDS A. Enterohepatic Circulation Approximately 95% of bile acids synthesized in the liver are reclaimed through a process called enterohepatic circulation [5,6]. A small portion of bile acids undergoes passive diffusion and uptake by the proximal intestine. However, the majority of conjugated bile acids, which are hydrophilic, are actively absorbed by the Copyright © 2005, Elsevier, Inc. All rights reserved.

864

DAVID J. MANGELSDORF AND DANIEL L. MOTOLA

allowed to accumulate because of the danger it poses to peripheral tissues, most notably the circulatory system. Thus, conversion of cholesterol into bile acids is a major pathway through which the body excretes excess cholesterol and maintains homeostasis, especially when faced with a large dietary cholesterol load [5,6].

terminal ileum through the help of an apical membrane sodium/bile salt transporter (ABST) [7]. From the enterocytes of the terminal ileum, the bile acids enter the portal venous circulation and return to their site of production in the liver to be secreted once again. This enterohepatic circulation of bile acids helps to maintain a pool of bile acids that can continuously function throughout the day to participate in digestion and absorption of dietary lipids. During the process of circulating between the liver and the intestine, 5% of the bile acid pool is lost with each round. This 5% loss is made up for by the conversion of approximately 500 mg of cholesterol into bile acids each day [2]. Although cholesterol serves several important functions in our bodies, it must not be

B. Production of Secondary Bile Acids As the bile acid pool continually circulates, enteric flora located within the intestine metabolize primary bile acids into the secondary bile acids, deoxycholic acid (DCA) and lithocholic acid (LCA) (Fig. 1) [5,6]. These secondary bile acids result from bacterial

HO

Cholesterol

CYP7A1 CYP8B1 CYP27A1 O OH OH Liver synthesis and conjugation

O

OH

OH Chenodeoxycholic acid

HO

O

S OH

O

S OH

O OH Taurochenodeoxycholic acid

HO

O

O OH

HO

O

N H

OH Taurocholic acid

OH Bacterial modifications

OH

O

N H

Primary bile acids

O

OH Cholic acid

HO

HO

CYP7A1 CYP27A1

Secondary bile acids

OH

Deoxycholic acid

Lithocholic acid

HO

Hydroxylation Detoxification

Sulfation O

O

OH

OH _ HO

FIGURE 1

O3SO

OH

Sulfolithocholic acid Hyodeoxycholic acid

Classical pathway of bile acid synthesis including the production and detoxification of the secondary bile acid lithocholic acid (LCA). Conversion of cholesterol to bile acids occurs exclusively in the liver through a classical and acidic pathway. For a comprehensive review of these two pathways refer to [2]. In the classical pathway, illustrated here, CYP7A1 hydroxylates cholesterol at position 7. Further modifications to the ring structure occur followed by modifications to the side chain, which are carried out by CYP27A1. The primary bile acids, cholic acid (CA) and chenodeoxycholic acid (CDCA), are produced as a result Before being released into the intestine, these bile acids are conjugated to amino acids taurine or glycine. Bacteria within the colon deconjugate and dehydroxylate CA and CDCA into the less polar and more toxic secondary bile acids, deoxycholic acid (DCA) and LCA, respectively. Sulfation or hydroxylation of secondary bile acids, such as LCA, aids in their detoxification and elimination from the body.

865

CHAPTER 53 Vitamin D Receptor and Bile Acids

deconjugation and 7α-dehydroxylation of CA and CDCA, respectively. Both DCA and LCA are much more hydrophobic than their primary bile acid counterparts and therefore more readily concentrate within cells. The detergent like properties and hydrophobicity of DCA and LCA pose a threat to the cell integrity and are believed to cause damage directly to DNA through the formation of adducts and strand breaks and indirectly through the inhibition of DNA repair enzymes [8–10]. These properties alone make bile acids potential carcinogens. Additionally, one of these bile acids, LCA, is associated with liver damage seen in cases of cholestatic liver disease. In addition, administration of LCA has been shown to lead to cholestasis in rodents and high levels of LCA have been observed in patients suffering from chronic cholestatic liver disease [11–13]. Thus, although bile acids serve several important functions, they are also inherently toxic to cells at high levels and therefore must also be maintained at constant homeostatic levels.

IV. NUCLEAR RECEPTORS: KEY REGULATORS OF BILE ACID METABOLISM It should be apparent that very tightly regulated mechanisms of control must exist to maintain both cholesterol and bile acid levels. The molecular mechanisms that govern this regulation have been elucidated within the past few years and have uncovered an important role for nuclear receptors in this process. The following section will briefly describe the role of nuclear receptors in this process and will end with the discussion of how VDR participates in these mechanisms of regulation.

A. Liver-X-Receptor (LXR): An Oxysterol Receptor The nuclear receptors play a major role in the control of cholesterol and bile acid levels. The feedforward control of cholesterol’s conversion into bile acids as well as its excretion into bile is now known to be mediated through the action of two nuclear hormone receptors, the liver X receptors (called LXRα and LXRβ). LXRα is expressed predominately in the liver where it senses the levels of cholesterol through binding of oxysterols (oxidized cholesterol) and in turn modulates the expression of gene networks involved in the catabolism, transport, storage, and absorption of cholesterol [3]. One target of LXRα in mice but not in humans is the cytochrome p450 enzyme CYP7A1, the

rate-limiting enzyme in the conversion of cholesterol into bile acids [2,3]. In both humans and rodents, LXRs also up-regulate the expression of several ATPbinding cassette transporters, which pump cholesterol out of the liver and into the intestine [3]. In the intestine, these same proteins are also regulated by the LXRs to reverse-transport cholesterol out of the enterocytes and thus limit its absorption [3]. When challenged with cholesterol, mice lacking either LXRα alone or both isoforms fail to turn on these pathways and rapidly accumulate toxic levels of cholesterol within the liver [14].

B. Farnesoid X Receptor (FXR): A Bile Acid Receptor In the case of bile acids, feedback mechanisms exist to control their production as well as maintain a normal bile acid pool size and composition. The farnesoid X receptor, FXR, is expressed in the liver and intestine and acts as a receptor for bile acids [15–17]. Bile acid activation of FXR represses the expression of CYP7A1, the rate-limiting enzyme for bile acid synthesis. FXR reduces CYP7A1 gene expression by regulating the expression of a potent repressor called short heterodimer partner (SHP), an unusual orphan receptor that does not possess a DNA binding domain [18,19]. As a result, SHP dimerizes with other nuclear receptors that are bound to the CYP7A1 promoter and dominantly represses transcription, thereby turning off bile acid synthesis. FXR also helps maintain the normal flow of bile acids by modulating the expression of various bile acid transporters and binding proteins in the liver and intestine [20].

C. Pregnane X Receptor (PXR): A Xenobiotic Receptor The pregnane X receptor (PXR) functions primarily in the liver to protect the body against the accumulation of toxic exogenous (xenobiotics) or endogenous (endobiotic) chemicals [21–24]. The large ligand-binding pocket of PXR allows it to bind a structurally diverse group of chemicals that include many prescription drugs, steroids, and over-the-counter herbal remedies [24]. PXR controls the levels of these compounds by up-regulating the gene expression of the enzymes responsible for their metabolism and clearance. These enzymes include the cytochrome p450s, in particular the CYP3A subfamily members, and the sulfotransferases. In cholestatic liver disease, a failure to secrete bile results in the progressive accumulation of bile

866 constituents in the liver causing liver damage and eventual cirrhosis. LCA, as mentioned above, is a highly toxic secondary bile acid metabolite that is elevated in the liver and serum of patients under cholestatic conditions [13]. Recent investigations have now demonstrated that PXR functions as an LCA sensor in the liver so that it can protect the liver against LCA-induced hepatotoxicity [21–24]. At high concentrations (∼100 µM), LCA binds PXR, which is then activated to induce the expression of two genes, N-sulfotransferase (SULT-N), a member of the sulfotransferase gene family, and CYP3A, both of which convert LCA into a more polar and less toxic bile acid that is more easily eliminated from the body [21–24].

V. BILE ACIDS, VITAMIN D RECEPTOR, AND COLON CANCER The most recent finding involving nuclear receptors and the regulation of bile acid metabolism is the surprising role of VDR as an enteric sensor of carcinogenic bile acids. The finding may help explain the protective effects of vitamin D and VDR in preventing colon cancer.

A. High-Fat Diets Increase Risk for Colon Cancer In the United States, colon cancer is the third most common cancer in men and women [25]. Besides genetic predisposition and other environmental risk factors, epidemiological and animal studies have given rise to the prevailing theory that a high dietary intake of fat is a primary contributing factor in the development of colon cancer [26] (see Chapter 91). Associated with this high-fat diet is an increase in the secretion of fecal bile acids, specifically the toxic secondary bile acid, LCA [27]. In animal studies, LCA administration has been shown to induce colon cancer [28,29] and its concentrations have been reported to be elevated in patients with colon cancer [27].

DAVID J. MANGELSDORF AND DANIEL L. MOTOLA

age-adjusted colon cancer mortality rates are highest in the northern parts of the country, especially in the northeast United States where a combination of latitude, climate, and pollution results in a 5-month period without any vitamin D synthesis [30] (see Chapter 90). The geographical differences in fat intake are not likely to account for the differences in colon cancer mortality as the southern United States typically consumes more fat than the northern United States [30]. Furthermore, clinical and laboratory studies have shown that calcium and/or vitamin D supplementation can decrease the incidence of colon cancer associated with a high-fat diet, in high-risk patients, or as a result of LCA administration in rodents [30–32]. These epidemiological and experimental findings suggest that calcium or vitamin D may help protect against the mortality and/or development of colon cancer (also see Chapters 90 and 91). The ability of vitamin D to inhibit growth and induce differentiation and apoptosis in malignant cells is well established [33]. However, the mechanism by which vitamin D prevents colon cancer remains an area of active investigation. A possible explanation for the protective effects of vitamin D comes from recent investigations involving nuclear receptors in the regulation of bile acid metabolism [1]. As was previously mentioned, under conditions of cholestasis the nuclear hormone receptor PXR may be responsible for the detoxification of LCA in the liver through up-regulating the gene expression of CYP3A. However, an additional PXR-independent pathway for LCA detoxification must exist, since PXR-null animals still showed LCA-induced expression of CYP3A [22]. The only other receptor known to interact with primary or secondary bile acids is FXR [15–17] (Section IV, B); however, CYP3A is not known to be an FXR target gene, nor is vitamin D a ligand for FXR or PXR. Interestingly, two reports suggested that CYP3A is a VDR target gene [34,35]. This led to the exploration of the potential role of VDR in bile acid metabolism, specifically of LCA within the intestine [1].

C. VDR as a Bile Acid Receptor B. A Protective Role for Vitamin D In contrast to the positive correlations between high-fat diets, LCA, and colon cancer, the levels of another dietary component, vitamin D, is negatively correlated with the incidence of colon cancer [30]. Epidemiological studies have shown that strikingly similar global geographical patterns exist between colon cancer and rickets. For example, in the United States,

Strong correlative evidence supports the idea that VDR may have evolved the ability to sense bile acids, in particular LCA. Phylogenetic analysis of VDR shows that its primary amino acid sequence is most similar to that of PXR and FXR. The highest identity is found between the DNA-binding and ligand-binding domains of VDR and PXR [36,37]. Interestingly, both PXR and VDR bind to DNA response elements configured as everted repeats spaced by six nucleotides (ER-6)

867

CHAPTER 53 Vitamin D Receptor and Bile Acids

or direct repeats spaced by three nucleotides (DR-3) [33,38] (see Chapter 18). Several experiments were carried out to determine whether or not VDR indeed mediates the PXR-independent LCA-induced expression of CYP3A. The results of these experiments are detailed next. In order to identify the LCA receptor, a panel of enterohepatic nuclear receptors was screened for their ability to bind LCA and induce the expression of CYP3A [1]. The ability of a receptor to bind a ligand and activate gene expression can be measured in vitro using cell-based, ligand-induced reporter activation assays. These results can also be confirmed by using competitive binding assays, which also assess specificity and kinetics of ligand binding. To identify the LCA receptor, an approach was taken based on ligandinduced interaction of a nuclear receptor with its coactivator. The coactivator protein SRC-1 was fused to the DNA binding domain of the yeast transcription factor, GAL4. Each receptor within the panel was also fused to the activation domain of the herpes virus protein,VP16. Human embryonic kidney cells (HEK 293) were transfected with the appropriate plasmids, and each nuclear receptor was tested for its ability to bind LCA, recruit coactivator protein, and activate the expression of a GAL4 responsive reporter gene. At the concentration of LCA tested, 30 µM, only VDR and FXR were activated. Interestingly, a much higher concentration of LCA (≥100 µM) was required for activation of PXR. Further analysis of the ligand specificity of VDR and FXR revealed that each receptor has a distinct specificity profile. VDR was activated by 1,25(OH)2D3, while FXR was not. Similarly, FXR was activated by the primary bile acids, CA and CDCA, while VDR was not. Both receptors bound LCA and its 3-keto metabolite, but only VDR was activated by the 6-keto metabolite of LCA. Dose response curves revealed that VDR was activated by LCA and 3-keto LCA with a median effective concentration (EC50) of 8 µM and 3 µM, respectively. However, FXR required two to three times greater concentrations of LCA and its 3-keto metabolite for activation, while PXR required a concentration at least 10 times greater, when compared to activation of VDR. Taken together, the results from this work suggested that VDR is a more sensitive receptor for LCA and its metabolites than are FXR and PXR, and that the ligand specificity profile of each receptor has allowed each receptor to perform distinct functions. To further confirm the direct interaction of LCA and 3-keto-LCA with VDR, competitive binding experiments were performed using purified recombinant VDR, radiolabeled 1,25(OH)2D3, and increasing concentrations of unlabeled LCA or 3-keto-LCA as competitors. These experiments revealed a direct and specific binding of

LCA and 3-keto-LCA to VDR (inhibition constant Ki = 29 ± 6 µM and 8 ± 3 µM, respectively) [1]. Furthermore, the binding affinity of LCA and 3-keto-LCA for VDR matched closely the dose response data from transactivation experiments and provided convincing evidence that LCA and its metabolites are bona fide ligands for VDR.

D. VDR Regulates CYP3A-Dependent Detoxification of LCA The ability of LCA to induce the expression of CYP3A through VDR was then confirmed using both in vitro and in vivo experiments [1]. Response elements to which VDR binds in its known target genes were found in the promoters of the human, rat, and mouse CYP3A genes. These CYP3A response elements were shown to be bound by VDR/RXR heterodimers using electrophoretic mobility shift assays (EMSA) and were also able to confer VDR-dependent LCA induction of a reporter gene in cell-based transfection assays. These results were confirmed in vivo, as administration of either vitamin D or LCA induced the expression of CYP3A within the intestine and livers of wild-type mice as well as in mice lacking PXR. These experiments provided further evidence confirming the existence of a PXR-independent pathway for CYP3A induction and supported the idea that VDR is a sensitive receptor for LCA and can induce LCA catabolism within the intestine by up-regulating the expression of CYP3A. In addition to CYP3A, other genes involved in LCA metabolism are also induced by VDR (B. Chatterjee, personal communication). One such example is SULT-N, which is a PXR target gene and is important for the clearance of LCA from the liver [23].

VI. A MODEL OF LITHOCHOLIC ACID DETOXIFICATION IN THE INTESTINE The experiments just described provided new insights into the mechanism by which vitamin D might protect against colon cancer and revealed an unexpected role for VDR as a sensor of carcinogenic bile acids. The epidemiological findings of similar geographical distributions of both rickets and colon cancer may therefore be explained under the following model (Fig. 2). Under normal nutritional status, VDR utilizes 1,25(OH)2D3 or LCA to maintain increased levels of CYP3A gene expression. CYP3A converts LCA or its metabolites into less toxic bile acids, thus preventing injury to the intestinal epithelium. However, the risk for developing colon cancer rises significantly when LCA levels are increased as a result of either a sustained high-fat diet

868

DAVID J. MANGELSDORF AND DANIEL L. MOTOLA

A

B Low fat diet normal vitamin D

Vitamin D LCA Catabolized LCA

High fat diet and/or low vitamin D

Excessive level

Normal level CYP3A

CYP3A

N

FIGURE 2 A model for the detoxification of LCA in the intestine. (A) On a low-fat diet, low levels of LCA (open square) and high levels of vitamin D (filled triangle) tip the balance away from colon cancer and toward a normal state. Vitamin D levels are important for maintaining this normal physiological state as 1,25(OH)2D3 can also bind to VDR and activate CYP3A catabolism of LCA. (B). Certain pathological conditions such as sustained high-fat diets or vitamin D deficiency tip the balance toward colon cancer. A sustained high-fat diet overwhelms the capacity of CYP3A to metabolize LCA and excessive levels of LCA “spill over” and cause injury to the colonic epithelium. Alternatively, in conditions of low vitamin D, such as rickets, the balance is similarly shifted toward colon cancer.

or vitamin D deficiencies. Since both of these conditions (i.e., sustained high-fat diets and vitamin D deficiency) are pathophysiologic and beyond what the body has evolved to cope with, the detoxification system cannot keep up with the elevated levels of LCA even though the VDR sensing mechanism is still in place. This results in LCA-induced toxicity to the colon. As predicted by this model, mice lacking VDR have been shown to display enhanced preneoplastic cellular proliferations in the colon [39].

VII. PERSPECTIVES In this chapter we have focused on the recent, and indeed unexpected, finding that VDR can function as both an endocrine receptor (for vitamin D) and a lipid sensing receptor (for LCA). This finding has both evolutionary and pharmacological implications. From an evolutionary perspective, it is interesting to note that VDR exists as one of a number of RXR heterodimer partners that evolved to protect the body against

dietary lipid excess. These receptors (including the LXRs, FXR, and PXR) are all expressed primarily in the enterohepatic system, where they are ideally positioned to sense toxic dietary constituents and eliminate them by activating expression of a metabolic gene network [3]. Further up the evolutionary ladder, a second group of RXR heterodimers has evolved as endocrine receptors (e.g., thyroid hormone and retinoic acid receptors). The fact that VDR has elements of both the endocrine and lipid sensing receptor class lends credence to the hypothesis that the primordial VDR ancestor was likely a bile acid sensor. It is tempting to speculate that from this ancestral receptor, the family expanded to become PXR, FXR, and the LXRs, all of which are phylogenetically related to VDR and share similar sterol-derived ligands. The evolution of VDR from this group of lipid-sensing receptors into an endocrine receptor that regulates calcium homeostasis may not be that surprising if one considers that such a receptor must have some of the same properties as other receptors in this class, such as an abundant enterohepatic expression. The fact that VDR kept its vestigial

CHAPTER 53 Vitamin D Receptor and Bile Acids

bile acid receptor qualities suggests that there was a strong evolutionary drive to keep LCA concentration in check. It will be interesting to test this prediction further by investigating the ligand binding properties of VDR in more primitive species. This work also suggests the importance of this group of receptors as potential therapeutic targets for novel drug discovery. In particular, the work highlighted here should give rise to future studies and efforts aimed at the development of new pharmacological strategies for the prevention of colon cancer. With respect to VDR, the ideal candidate would be a drug that does not carry the unwanted side effects such as hypercalcemia, which are currently seen with many of the VDR agonists currently used to treat psoriasis and osteoporosis. The notion that such selective VDR modulators may be found is strengthened by recent findings showing that most vitamin D analogs bind the ligand-binding pocket of VDR differently than LCA [40]. Consequently, it may be possible to design novel VDR agonists that induce multiple conformations of the VDR’s ligand binding domain, thereby resulting in differential therapeutic actions. The discovery of selective estrogen receptor modulators (SERMs) and elucidation of their mechanisms of action provides an important proof of principle as to how this might actually be accomplished with VDR [41,42].

References 1. Makishima M, Lu TT, Xie W, Whitfield GK, Domoto H, Evans RM, Haussler MR, Mangelsdorf DJ 2002 Vitamin D receptor as an intestinal bile acid sensor. Science 296:1313–1316. 2. Russell DW 2003 The Enzymes, regulation, and genetics of bile acid synthesis. Annu Rev Biochem. 3. Chawla A, Repa JJ, Evans RM, Mangelsdorf DJ 2001 Nuclear receptors and lipid physiology: opening the X-files. Science 294:1866–1870. 4. Lu TT, Repa JJ, Mangelsdorf DJ 2001 Orphan nuclear receptors as eLiXiRs and FiXeRs of sterol metabolism. J Biol Chem 276:37735–37738. 5. Hofmann AF 1999 The continuing importance of bile acids in liver and intestinal disease. Arch Intern Med 159: 2647–2658. 6. Carey MaD, WC 1994 Enterohepatic circulation. In: Arias IM BJ, Fausto N, Jakoby WB, Schachter DA, Sharfritz DA (eds) The Liver: Biology and Pathobiology, 3rd ed. Raven Press, New York, pp. 719–768. 7. Craddock AL, Love MW, Daniel RW, Kirby LC, Walters HC, Wong MH, Dawson PA 1998 Expression and transport properties of the human ileal and renal sodium-dependent bile acid transporter. Am J Physiol 274:G157–G169. 8. Nagengast FM, Grubben MJ, van Munster IP 1995 Role of bile acids in colorectal carcinogenesis. Eur J Cancer 31A: 1067–1070. 9. Hamada K, Umemoto A, Kajikawa A, Seraj MJ, Monden Y 1994 In vitro formation of DNA adducts with bile acids. Carcinogenesis 15:1911–1915.

869 10. Ogawa A, Murate T, Suzuki M, Nimura Y, Yoshida S 1998 Lithocholic acid, a putative tumor promoter, inhibits mammalian DNA polymerase beta. Jpn J Cancer Res 89:1154–1159. 11. Fisher MM, Magnusson R, Miyai K 1971 Bile acid metabolism in mammals. I. Bile acid-induced intrahepatic cholestasis. Lab Invest 25:88–91. 12. Javitt NB 1966 Cholestasis in rats induced by taurolithocholate. Nature 210:1262–1263. 13. Fischer S, Beuers U, Spengler U, Zwiebel FM, Koebe HG 1996 Hepatic levels of bile acids in end-stage chronic cholestatic liver disease. Clin Chim Acta 251:173–186. 14. Peet DJ, Turley SD, Ma W, Janowski BA, Lobaccaro JM, Hammer RE, Mangelsdorf DJ 1998 Cholesterol and bile acid metabolism are impaired in mice lacking the nuclear oxysterol receptor LXR alpha. Cell 93:693–704. 15. Makishima M, Okamoto AY, Repa JJ, Tu H, Learned RM, Luk A, Hull MV, Lustig KD, Mangelsdorf DJ, Shan B 1999 Identification of a nuclear receptor for bile acids. Science 284:1362–1365. 16. Parks DJ, Blanchard SG, Bledsoe RK, Chandra G, Consler TG, Kliewer SA, Stimmel JB, Willson TM, Zavacki AM, Moore DD, Lehmann JM 1999 Bile acids: Natural ligands for an orphan nuclear receptor. Science 284:1365–1368. 17. Wang H, Chen J, Hollister K, Sowers LC, Forman BM 1999 Endogenous bile acids are ligands for the nuclear receptor FXR/BAR. Mol Cell 3:543–553. 18. Lu TT, Makishima M, Repa JJ, Schoonjans K, Kerr TA, Auwerx J, Mangelsdorf DJ 2000 Molecular basis for feedback regulation of bile acid synthesis by nuclear receptors. Mol Cell 6:507–515. 19. Goodwin B, Jones SA, Price RR, Watson MA, McKee DD, Moore LB, Galardi C, Wilson JG, Lewis MC, Roth ME, Maloney PR, Willson TM, Kliewer SA 2000 A regulator cascade of the nuclear receptors FXR, SHP-1, and LRH-1 represses bile acid biosynthesis. Mol Cell 6:517–526. 20. Sinal CJ, Tohkin M, Miyata M, Ward JM, Lambert G, Gonzalez FJ 2000 Targeted disruption of the nuclear receptor FXR/BAR impairs bile acid and lipid homeostasis. Cell 102:731–744. 21. Staudinger JL, Goodwin B, Jones SA, Hawkins-Brown D, MacKenzie KI, LaTour A, Liu Y, Klaassen CD, Brown KK, Reinhard J, Willson TM, Koller BH, Kliewer SA 2001 The nuclear receptor PXR is a lithocholic acid sensor that protects against liver toxicity. Proc Natl Acad Sci USA 98:3369–3374. 22. Xie W, Radominska-Pandya A, Shi Y, Simon CM, Nelson MC, Ong ES, Waxman DJ, Evans RM 2001 An essential role for nuclear receptors SXR/PXR in detoxification of cholestatic bile acids. Proc Natl Acad Sci USA 98:3375–3380. 23. Sonoda J, Xie W, Rosenfeld JM, Barwick JL, Guzelian PS, Evans RM 2002 Regulation of a xenobiotic sulfonation cascade by nuclear pregnane X receptor (PXR). Proc Natl Acad Sci USA 99:13801–13806. 24. Kliewer SA, Goodwin B, Willson TM 2002 The nuclear pregnane X receptor: A key regulator of xenobiotic metabolism. Endocr Rev 23:687–702. 25. Cancer Facts and Figures: American Cancer Society; 2003 [Postscript Document] URL: http://www.cancer.org/docroot/ STT/stt_0.asp 26. Lipkin M, Reddy B, Newmark H, Lamprecht SA 1999 Dietary factors in human colorectal cancer. Annu Rev Nutr 19:545–586. 27. Owen RW, Dodo M, Thompson MH, Hill MJ 1987 Fecal steroids and colorectal cancer. Nutr Cancer 9:73–80. 28. Narisawa T, Magadia NE, Weisburger JH, Wynder EL 1974 Promoting effect of bile acids on colon carcinogenesis after

870

29. 30. 31. 32.

33.

34.

35.

intrarectal instillation of N-methyl-N′-nitro-N-nitrosoguanidine in rats. J Natl Cancer Inst 53:1093–1097. Reddy BS, Watanabe K 1979 Effect of cholesterol metabolites and promoting effect of lithocholic acid in colon carcinogenesis in germ-free and conventional F344 rats. Cancer Res 39:1521–1524. Garland CF, Garland FC, Gorham ED 1999 Calcium and vitamin D. Their potential roles in colon and breast cancer prevention. Ann NY Acad Sci 889:107–119. Pence BC, Buddingh F 1988 Inhibition of dietary fat-promoted colon carcinogenesis in rats by supplemental calcium or vitamin D3. Carcinogenesis 9:187–190. Kawaura A, Tanida N, Sawada K, Oda M, Shimoyama T 1989 Supplemental administration of 1α-hydroxyvitamin D3 inhibits promotion by intrarectal instillation of lithocholic acid in N-methyl-N-nitrosourea-induced colonic tumorigenesis in rats. Carcinogenesis 10:647–649. Haussler MR, Whitfield GK, Haussler CA, Hsieh JC, Thompson PD, Selznick SH, Dominguez CE, Jurutka PW 1998 The nuclear vitamin D receptor: biological and molecular regulatory properties revealed. J Bone Miner Res 13:325–349. Thummel KE, Brimer C, Yasuda K, Thottassery J, Senn T, Lin Y, Ishizuka H, Kharasch E, Schuetz J, Schuetz E 2001 Transcriptional control of intestinal cytochrome P-4503A by 1α,25-dihydroxyvitamin D3. Mol Pharmacol 60:1399–1406. Schmiedlin-Ren P, Thummel KE, Fisher JM, Paine MF, Watkins PB 2001 Induction of CYP3A4 by 1α,25-dihydroxyvitamin D3 is human cell line-specific and is unlikely to involve pregnane X receptor. Drug Metab Dispos 29:1446–1453.

DAVID J. MANGELSDORF AND DANIEL L. MOTOLA

36. Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179. 37. Watkins RE, Wisely GB, Moore LB, Collins JL, Lambert MH, Williams SP, Willson TM, Kliewer SA, Redinbo MR 2001 The human nuclear xenobiotic receptor PXR: structural determinants of directed promiscuity. Science 292:2329–2333. 38. Kliewer SA, Moore JT, Wade L, Staudinger JL, Watson MA, Jones SA, McKee DD, Oliver BB, Willson TM, Zetterstrom RH, Perlmann T, Lehmann JM 1998 An orphan nuclear receptor activated by pregnanes defines a novel steroid signaling pathway. Cell 92:73–82. 39. Kallay E, Pietschmann P, Toyokuni S, Bajna E, Hahn P, Mazzucco K, Bieglmayer C, Kato S, Cross HS 2001 Characterization of a vitamin D receptor knockout mouse as a model of colorectal hyperproliferation and DNA damage. Carcinogenesis 22:1429–1435. 40. Choi M, Yamamoto K, Itoh T, Makishima M, Mangelsdorf DJ, Moras D, DeLuca HF, Yamada S 2003 Interaction between vitamin D receptor and vitamin D ligands. Twodimensional alanine scanning mutational analysis. Chem Biol 10:261–270. 41. Brzozowski AM, Pike AC, Dauter Z, Hubbard RE, Bonn T, Engstrom O, Ohman L, Greene GL, Gustafsson JA, Carlquist M 1997 Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 389:753–758. 42. Shang Y, Brown M 2002 Molecular determinants for the tissue specificity of SERMs. Science 295:2465–2468.

CHAPTER 54

Vitamin D and the Renin–Angiotensin System YAN CHUN LI

Department of Medicine, University of Chicago, Chicago, Illinois

I. Introduction II. The Renin–Angiotensin System III. Sunlight, Vitamin D, and Blood Pressure: Epidemiological and Clinical Observations IV. Vitamin D and Cardiovascular Functions

V. 1,25-Dihydroxyvitamin D3 as a Negative Endocrine Regulator of the Renin–Angiotensin System VI. Vitamin D Analogs as Potential Antihypertensive Agents VII. Conclusion References

I. INTRODUCTION

cells are highly granulated smooth muscle cells located in the media of the afferent arteriole at the vascular pole of the glomerulus, where the afferent arteriole enters, and the efferent arteriole exits, the renal corpuscle (Fig. 2). The main function of renin is to cleave angiotensin I (Ang I), a 10-amino-acid peptide, from angiotensinogen, which is predominantly produced in the liver. Ang I is then converted to an 8-amino-acid peptide, angiotensin II (Ang II), by the angiotensinconverting enzyme (ACE), which primarily resides in the endothelial cells in blood vessels. Further processing of Ang II by animopeptidase A and N produces Ang III and Ang IV. Ang II is the central biological effector of the RAS. Through binding to its receptors, which are G protein-coupled receptors widely distributed and expressed by many cell types [5], Ang II exerts diverse physiological actions that regulate the homeostasis of electrolytes, extracellular volume and blood pressure [6,7]. For instance, Ang II acts on blood vessel smooth

The renin–angiotensin system (RAS) plays a central role in the regulation of blood pressure, extracellular volume, and electrolyte homeostasis. Inappropriate activation of the RAS may lead to hypertension, which is one of the major risk factors for stroke, myocardial infarction, congestive heart failure, progressive atherosclerosis, and renal failure. These are major diseases with high mortality rates in both developing and industrialized countries. However, prevention and therapeutic intervention of hypertension remain a major medical challenge at present, and hypertension prevalence in the United States has increased in the past decade [1]. Our understanding of the renocardiovascular system, including the RAS, will have a direct impact, at least to some degree, on how well we face this challenge. Recent advance has placed the vitamin D endocrine system at an important position in the regulation of the RAS, and this chapter will focus on the relationship between the vitamin D endocrine system and the RAS.

Prorenin

II. THE RENIN–ANGIOTENSIN SYSTEM

Angiotensinogen

Renin

A. An Overview The RAS is a systemic endocrine regulatory cascade (Fig. 1) that involves multiple organs [2], but components of the RAS have also been found inside many tissues such as the brain, the heart, and the kidney, suggesting that it may also function in a paracrine fashion [3]. The exact physiological role of the tissue RAS remains unclear. The first and rate-limiting component of the RAS cascade is renin, an aspartyl protease synthesized and secreted predominantly by the juxtaglomerular (JG) cells in the JG apparatus of the kidney [4]. The JG VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Angiotensin I ACE Angiotensin II

Angiotensin II receptors Thirst

Intestinal Na

ADH

Aldosterone Vasoconstriction

H2O intake Na intake H2O retention Na retention Extracellular volume

Blood pressure

FIGURE 1 The renin–angiotensin system. ADH, antidiuretic hormone, also known as vasopressin; ACE, angiotensin-converting enzyme. Copyright © 2005, Elsevier, Inc. All rights reserved.

872

YAN CHUN LI

muscle cells to cause vasoconstriction; stimulates antidiuretic hormone (ADH, also called vasopressin) production by the central nervous system, which increases water retention in the kidney; stimulates aldosterone synthesis and secretion in the adrenal cortex, which enhances renal sodium reabsorption in the distal tubules; and stimulates the sensation of thirst in the central nervous system and thus increases water intake. Therefore, activation of the renin–angiotensin cascade generally leads to an increase in extracellular volume and blood pressure (Fig. 1), and inappropriate stimulation of the RAS thus induces detrimental effects. Adding a new level of complexity to the RAS, recently a renin receptor has been reported in the heart, kidney, brain, placenta, and liver, which may function to increase the catalytic activity of renin [8], and an ACE homolog, ACE2, has been isolated [9,10] and found to play an essential role in heart functions [11].

B. Control of Renin Synthesis and Secretion Because of its central role in the renin–angiotensin cascade, the biosynthesis and secretion of renin is tightly regulated. Renin is synthesized in the JG cells as a prepropolypeptide precursor during translation in the endoplasmic reticulum. The targeting signal sequence is cleaved during translocation in the endoplasmic reticulum, yielding the inactive prorenin. Prorenin is glycosylated and activated during intracellular transport through the Golgi apparatus, and eventually stored in secretory granules. The prosequence is removed in this process. Renin is secreted from these granules through exocytosis upon stimulation [4,12]. It is well known that renin secretion in the kidney is commonly regulated by a variety of physiological factors, including renal perfusion pressure, renal

Proximal convoluted tubule

Basement membrane

Erythrocytes

Glomerular capillaries

Glomerular basement membrane

Mesangial cells Endothelium

Capsular (parietal epithelium)

Glomerular (visceral epithelium)

Lacis cells Efferent arteriole

JG cells Afferent arteriole

Macula densa

Distal tubule

Smooth muscle cells in media

FIGURE 2 Diagram of the renal corpuscle structure, showing the juxtaglomerular (JG) apparatus. The tubular pole, which is where the proximal tubule leaves the corpuscle, is at the top, and the vascular pole, which is where the afferent arteriole enters and the efferent arteriole leaves the corpuscle, is at the bottom. The granulated renin-producing JG cells are located in the afferent arteriole at the vascular pole and are in close vicinity to the macula densa. (Adapted from Cormack 1984, Introduction to Histology, Fig. 15-5, p. 344, with permission.)

CHAPTER 54 Vitamin D and the Renin–Angiotensin System

873

sympathetic nerve activity, and tubular sodium chloride load [4,6]. The perfusion pressure in the renal artery is the most profound parameter to influence renin secretion: When renal perfusion pressure falls, renin secretion rises; when renal perfusion pressure rises, renin secretion falls. This effect is mediated by a baroreceptor or stretch receptor mechanism in the JG cells [13]. The JG apparatus and the glomerulus have rich sympathetic nerve endings, which express β-adrenergic receptors. Stimulation of renin synthesis and release by renal innervation has been well documented, which is mediated by β-adrenergic receptors and intracellular cyclic AMP [14]. This pathway may exert a tonic stimulatory influence on renin production [15]. Renin secretion is also tightly regulated by the tubular sodium chloride load [16]. There is an inverse relationship between dietary sodium chloride intake and renin secretion. Tubular control of renin release is mediated by the macula densa, which is part of the distal tubule and anatomically in close association with the renin-producing JG cells (Fig. 2). The macula densa senses the sodium chloride load and transduces the signal, possibly via adenosine and ATP, to the JG cells to influence renin production and secretion [12]. At the local level, renin synthesis and release are influenced by a large variety of bioactive molecules. For instance, prostaglandins, nitric oxide, and adrenomedullin are known to stimulate renin secretion, whereas Ang II (as a feedback regulator), endothelin, vasopressin, and adenosine are inhibitors of renin release [4,6,12].

renin secretion, and several cAMP response elements (CRE) have been identified in both murine and human renin gene promoters [12]. However, both CREBdependent and -independent mechanisms may be involved in the cAMP-PKA pathway in human renin promoter activation [20]. Transgenic studies have demonstrated that sequences required for the tissuespecific and development stage–specific expression of the renin gene, as well as for the response to a variety of physiological stimuli, are located within 5 kb of the 5′-flanking region of the murine renin gene [21–23]. In recent years, studies of the renin gene regulation have been greatly facilitated by the establishment of a JG cell-like cell line, namely As4.1, from kidney tumors of SV40 T antigen transgenic mice [23]. The As4.1 cells express a high level of endogenous renin. In the 4.1 kb 5′-flanking region of murine Ren-1c gene, a 223 bp minimal promoter (−117 to +6) and a 242 bp enhancer (−2866 to –2625) at about –2.6 kb upstream of the transcriptional start site have been found to be essential for high-level expression of the renin gene in As4.1 cells [24]. Recent studies have identified an array of transcriptional factors involved in the transcriptional regulation of renin gene expression. These factors include positive regulators such as LXRα, RAR/RXR, CREB/CREM, USF1/USF2, HOX genes, NFI, and SP1/SP3 [25–29], and negative regulators such as NF-Y and Ear-2 [30,31]. Thus, the production of renin is determined by a combined interplay of multiple transcriptional regulators available or activated under a specific physiological condition.

C. Transcriptional Regulation of Renin Gene Expression Renin is encoded by a single gene in humans. In mice, some strains (e.g., C57BL/6) have one renin gene (Ren-1c), whereas others (e.g., DBA/2, J129) contain two renin genes (Ren-1d and Ren-2), which are closely linked and probably resulted from a duplication of the 21 kb Ren-1c-like ancestral gene [17]. The human renin gene and the three mouse renin genes all share the same overall genomic organization (e.g., nine exons and eight introns) and encode highly homologous proteins. For instance, Ren-1 and Ren-2 share 97% amino acid identity [18]. It is believed that the Ren-1 protein is the major source of circulating renin and thus is the major systemic regulator of the renin–angiotensin cascade. Recent studies in transgenic mice demonstrate that the Ren-1d and Ren-2 genes cooperate to preserve the homeostasis of the RAS [19]. Recent studies suggest that expression of the renin gene is regulated by a complex network of transcriptional factors. Cyclic AMP is a major mediator for

III. SUNLIGHT, VITAMIN D, AND BLOOD PRESSURE: EPIDEMIOLOGICAL AND CLINICAL OBSERVATIONS Evidence from epidemiological and clinical studies in recent decades has suggested a connection between the vitamin D endocrine system and blood pressure. As ultraviolet (UV) irradiation is essential for the cutaneous production of vitamin D, circulating vitamin D levels are greatly influenced by geographic locations, seasonal changes, and skin pigmentations. Data obtained from the INTERSALT study centers reveal a linear correlation between the rise in blood pressure or the prevalence of hypertension and the latitudes north or south of the equator [32]. Similarly, data from a national survey in China also show a high-to-low gradient from the north to the south of the country in the prevalence of hypertension and stroke incidence [33]. Seasonal variations in blood pressure have been reported in temperate climates, with blood pressure higher in the winter (low UV irradiation) than in the

874

YAN CHUN LI

SBP (mm Hg)

126.12

124 120.55

121

6

4

2

20

40

60

80

100

120

1,25(OH)2D3 (pg/ml)

FIGURE 4

Inverse relationship between circulating 1,25(OH)2D3 levels and the plasma renin activity in patients with essential hypertension. (Adapted from Resnick et al. 1986, Fig. 2, p. 652, with permission.)

supplementation reduces the blood pressure and plasma renin activity in a patient with pseudohyperparathyroidism and high plasma renin activity [46], and intravenous 1,25(OH)2D3 treatment of hemodialysis patients with secondary hyperparathyroidism significantly reduces the plasma renin and Ang II levels and concomitantly regresses the myocardial hypertrophy [47]. These observations are consistent with a role of the vitamin D endocrine system in the regulation of renocardiovascular functions and blood pressure.

130 127

8 Plasma renin activity (ng/ml/h)

summer (high UV irradiation) [34,35]. Dark skin pigmentation, which affects efficient UV light penetration [36], has also been reported to be associated with higher blood pressure [37,38]. Indeed, UV irradiation has been reported to lower blood pressure in patients with mild essential hypertension [39]. Numerous studies have shown that the serum level of 1,25(OH)2D3 is inversely associated with blood pressure in normotensive and hypertensive subjects (Fig. 3) [40–42]. More interestingly, such inverse relationship has also been reported between circulating 1,25(OH)2D3 levels and plasma renin activity in patients with essential hypertension (Fig. 4) [43]. Vitamin D supplement has been shown to be beneficial to the heart. For instance, in double-blinded, placebo-controlled clinical trials, long-term treatment with 1α-hydroxyvitamin D3 results in a reduction in blood pressure in patients with essential hypertension [44], and short-term vitamin D3 and calcium supplementation reduces blood pressure in elderly women [45]. In clinical cases, it has been reported that 1,25(OH)2D3

120.51

118 115.44 115

<=60

61–70

71–80

>80

82

DBP (mm Hg)

80

79.13

78 76.43

76.38

76.61

61–70

71–80

>80

76 74 72 70 <=60

Serum 1,25(OH)2D3 (pmol/L)

FIGURE 3 Inverse relationship between circulating 1,25(OH)2D3 levels and blood pressure in humans. The graphs show blood pressure levels at different quartiles of serum 1,25(OH)2D3 concentrations. SBP, systolic blood pressure; DBP, diastolic blood pressure. (Adapted from Kristal-Boneh et al. 1997, Fig. 2, p. 1291, with permission.)

IV. VITAMIN D AND CARDIOVASCULAR FUNCTIONS The heart is believed to be a vitamin D target organ, as the vitamin D receptor (VDR) is found expressed in skeletal and cardiac myocytes [48,49]. Early studies, mostly using vitamin D–deficient rats, have showed that 1,25(OH)2D3 may play some roles in the regulation of cardiovascular functions. In rats rendered vitamin D– deficient for 9 weeks, systolic blood pressure as well as cardiac and vascular muscle contractile response are increased [50]. The ratio of heart weight to body weight is also significantly increased in the deficient animals [51]. Simpson and colleagues suggested that the rise in blood pressure and the change in vascular muscle contractile function may be due to hypocalcemia induced by vitamin D-deficiency and thus represent an indirect response to vitamin D, whereas the cardiac hypertrophy and cardiac contractile function may be directly affected by vitamin D, as they cannot be

CHAPTER 54 Vitamin D and the Renin–Angiotensin System

875

prevented by preventing the hypocalcemia with a highcalcium diet in vitamin D-deficient rats [50,52,53]. Further studies demonstrate a significant increase in myocardial collagen in the extracellular space and in V1 myosin isozyme expression in the heart of vitamin D– deficient rats, which might be the basis for abnormal cardiac contractility [51,54]. Interestingly, these vitamin D–deficient rats also show a marked increase in plasma renin activity in both hypocalcemic and normocalcemic states [52]. Therefore, the activation of the RAS may also contribute to the aberrant blood pressure and cardiovascular functions associated with vitamin D deficiency. In primary cultures of neonatal rat cardiac myocytes, 1,25(OH)2D3 has been shown to inhibit ventricular myocyte proliferation [55], and antagonize endothelininduced myocyte hypertrophy [56]. These studies demonstrate that genes associated with myocyte proliferation and hypertrophy, including c-myc, PCNA, and ANP, are suppressed by 1,25(OH)2D3 treatment, suggesting a direct effect of this secosteroid hormone on the cardiac myocytes. However, whether the mechanism of direct regulation of cardiac myocytes by 1,25(OH)2D3 can be applied in intact animals remains to be established.

this hypothesis. By inference from the hypothesis, VDR(−/−) mice are expected to display elevated renin expression. Indeed, we have demonstrated that both renin mRNA and protein levels in the kidney are drastically increased in VDR(−/−) mice. The plasma level of Ang II, which is a downstream product of renin, is also markedly increased in the mutant mice, whereas the expression of angiotensinogen, the substrate of renin, in the liver is the same as in wild-type mice (Fig. 5) [57]. Therefore, the increase in plasma Ang II production appears to be mainly due to an increase in renin activity. As a consequence of the aberrant RAS overstimulation, VDR(−/−) mice develop high blood pressure, cardiac hypertrophy, and an overdrinking behavior, mostly due to the potent vasoconstricting and thirst-inducing effect of Ang II [6,60]. As expected, plasma and urinary aldosterone levels are also markedly increased in VDR(−/−) mice [59]. Urinary volume and urinary salt excretion are also higher, whereas plasma sodium and potassium concentrations remain normal in the mutant mice. Cardiac hypertrophy, likely induced by elevated Ang II production, is reflected by a higher heart-weight-to-body-weight ratio and increased cardiac myocyte size in the left ventricle revealed by histological analyses. Accompanying the cardiac hypertrophy, both cardiac ANP mRNA expression and plasma ANP concentration are increased in VDR(−/−) mice, as a compensatory response of the body. In fact, an increase in ANP expression is the most common feature of cardiac hypertrophy. VDR inactivation alters many physiological parameters; thus it is important to directly correlate the cardiovascular phenotype with the changes in the RAS. In this regard, we have shown that the high blood pressure, cardiac hypertrophy, and increased water intake seen in VDR(−/−) mice can be corrected by treatment with captopril, an ACE inhibitor, or losartan, an Ang II AT1 receptor antagonist. These results confirm that overstimulation of the RAS is mostly responsible for these abnormalities. Because of the critical role of the vitamin D endocrine system in the regulation of calcium homeostasis, inactivation of VDR leads to development of hypocalcemia and secondary hyperparathyroidism [61], which may influence renin production and secretion. To address the contribution of serum calcium or parathyroid hormone (PTH) to renin up-regulation seen in VDR(−/−) mice, we have examined renin expression in 20-day-old VDR(−/−) mice, adult VDR(−/−) mice whose blood calcium levels are normalized by a high calcium–high lactose diet treatment [62], and Gcm2(−/−) mice that lack the parathyroid glands [63]. The 20-day-old VDR(−/−) mice are still normocalcemic, as intestinal calcium absorption is

V. 1,25-DIHYDROXYVITAMIN D3 AS A NEGATIVE ENDOCRINE REGULATOR OF THE RENIN–ANGIOTENSIN SYSTEM A. A Hypothesis Based mostly on the epidemiological and clinical evidence regarding a possible connection between the vitamin D endocrine system and blood pressure, particularly the inverse relationship between the circulating 1,25(OH)2D3 level and plasma renin activity (Fig. 4) [43], we speculate that vitamin D may control blood pressure through regulating the RAS. Specifically, we hypothesize that 1,25(OH)2D3 may function as a negative endocrine regulator of renin gene expression in vivo [57]. Thus, if the hypothesis is correct, disruption of the vitamin D signaling pathway should lead to deregulated elevation of renin expression and consequent stimulation of the RAS, whereas increasing the circulating level of 1,25(OH)2D3 should lead to a suppression of renin production. This hypothesis is strongly supported by a number of in vivo and in vitro studies [57–59].

B. Animal Studies VDR-null mutant mice, which lack VDR-mediated vitamin D signaling, are an ideal animal model to test

876

YAN CHUN LI

+/+

B

−/−

6 Renin mRNA

A

Renin 36B4

5 4 3 2 1 0

+/+

−/−

C −/−

+/+

D

E

Ang II (pg/ml)

600 500

+/+

−/− Angiotensinogen

400 300

36B4

200 100 0

+/+

−/−

FIGURE 5 VDR knockout mice display elevated renin expression and plasma Ang II production. (A) Renin mRNA expression in the kidney. (B) Quantitative results of the Northern blot analyses shown in A. *, P < 0.001 vs +/+ mice. (C) Immunohistochemical staining of the kidney cortex from wild-type and VDR knockout mice with anti-renin antiserum. Arrows indicate the afferent glomerular arterioles in the juxtaglomerular region. (D) Plasma Ang II concentrations in wildtype and VDR knockout mice. *, P < 0.001 vs +/+ mice. (E) Liver angiotensinogen mRNA expression in wild-type and VDR knockout mice. +/+, wild-type mice; −/−, VDR knockout mice; Ang II, angiotensin II. (From Li et al. 2002, with permission.)

independent of vitamin D before weaning. Renin up-regulation is already evident in these young mice, apparently before hypocalcemia develops. Furthermore, in the normocalcemic adult VDR(−/−) mice, renin mRNA level remains elevated, as is the plasma Ang II level. On the other hand, renin expression is normal in the Gcm2(−/−) mice, even though these mutant mice are as hypocalcemic as VDR(−/−) mice [63]. In addition, renin expression remains up-regulated in VDR(−/−) mice whose alopecia is rescued by targeted expression of human VDR in the skin [64]. These data strongly suggest that regulation of renin expression by 1,25(OH)2D3 is independent of calcium metabolism

or alopecia. This conclusion is consistent with previous observations in humans that the inverse relationship between serum 1,25(OH)2D3 levels and plasma renin activity or blood pressure appears independent of the serum calcium level [40,43]. However, the contribution of PTH to the renin up-regulation in VDR(−/−) is less certain yet, because serum PTH starts to rise early in life before hypocalcemia develops and cannot be completely normalized by the dietary intervention, due to the lack of the VDR-mediated vitamin D inhibition of PTH biosynthesis [65]. The hypothesis has been further tested in wild-type mice and in mice lacking 25-hydroxyvitamin D3

CHAPTER 54 Vitamin D and the Renin–Angiotensin System

877

1α-hydroxylase (1αOHase). In wild-type mice rendered vitamin D–deficient by dietary strontium treatment, which inhibits 1,25(OH)2D3 biosynthesis [66], or in 1αOHase(−/−) mice, renin expression in the kidney is markedly up-regulated, as seen in VDR(−/−) mice. On the other hand, in wild-type mice that have received several doses of 1,25(OH)2D3 injection, renin expression is significantly suppressed. Thus, the inhibitory role of 1,25(OH)2D3 in renin biosynthesis is confirmed in other mouse models.

can then be derived from the in vitro studies. In As4.1 cells transiently or stably transfected with human VDR cDNA, treatment with 1,25(OH)2D3 drastically reduces renin mRNA expression in 24 hr in a dosedependent manner [57]. To elucidate the molecular mechanism whereby 1,25(OH)2D3 suppresses renin gene expression, stable hVDR-As4.1 cells are used to analyze the renin gene promoter by luciferase reporter assays. When the cells are transfected with a luciferase reporter plasmid containing the 4.1kb 5′-flanking DNA sequence of the murine Ren-1c gene, 1,25(OH)2D3 treatment markedly reduces the promoter activity, confirming that 1,25(OH)2D3 directly and negatively regulates renin gene transcription by a VDR-mediated mechanism. Through deletion analysis two critical DNA fragments in the Ren-1c gene promoter have been identified. One fragment contains the base pairs from –2720 to –2642, and the other from –117 to +1. The later is the minimal promoter of the Ren-1c gene. These DNA fragments are sufficient to mediate the 1,25(OH)2D3 repression of the promoter activity and are the focus of current investigations aiming at completely elucidating the molecular mechanism underlying the transcriptional repression of renin gene expression by 1,25(OH)2D3.

C. Vitamin D Suppression vs Other Regulatory Mechanisms in Renin Regulation As renin biosynthesis is regulated by multiple physiological factors and pathways, it is important to determine whether 1,25(OH)2D3 regulates renin expression by altering other regulatory mechanisms. To this end, we have examined the response of VDR(−/−) mice to high-salt diet and water deprivation [57]. As in wildtype mice, renin expression in VDR(−/−) mice is markedly suppressed by a diet containing 8% sodium chloride, and is dramatically stimulated by 24-hr dehydration. The same is true for plasma Ang II production. Thus, the mechanisms to sense changes in salt intake and extracellular volume are still functionally intact in VDR(−/−) mice. Moreover, treatment with captopril or losartan, which blocks Ang II signaling, leads to a drastic up-regulation of renin expression in wild-type as well as in VDR(−/−) mice [59], suggesting that the regulatory loop of Ang II feedback inhibition of renin production is also functionally intact in VDR(−/−) mice. In all the above cases VDR(−/−) mice always maintain a significantly higher renin expression than wild-type mice in any circumstance. These observations indicate that, despite a high basal renin synthesis, the basic regulatory mechanisms that control renin production, including the Ang II feedback inhibition and the salt- and volume-sensing mechanisms, are normal in VDR(−/−) mice. Therefore, the sustained renin up-regulation is not mediated by these mechanisms. In another words, vitamin D regulation of renin synthesis is an independent mechanism.

D. Mechanism of Vitamin D Suppression of Renin Gene Expression We have also tested the hypothesis using cell cultures. As4.1 cells maintain a high level of endogenous renin synthesis [23] and thus are very suitable for studying the effect of 1,25(OH)2D3 on renin gene expression. Mechanistic knowledge of the regulation

E. Physiological Implications The finding that 1,25(OH)2D3 regulates the RAS is consistent with the view that the vitamin D endocrine system plays multiple physiological roles. Figure 6 outlines the interaction between the vitamin D endocrine system and the RAS in the regulation of calcium, volume, and blood pressure homeostasis. 1,25(OH)2D3 and PTH are known to be the principal regulators for maintaining the blood calcium concentration. 1,25(OH)2D3 also functions as a negative regulator of renin production and thus of the RAS, which helps explain the inverse relationship between the circulating 1,25(OH)2D3 level and blood pressure reported previously. As some low 1,25(OH)2D3/high renin subjects are able to maintain normal serum calcium levels, the threshold of the circulating 1,25(OH)2D3 level to induce hypocalcemia and hyperreninemia might be different. Under what physiological conditions 1,25(OH)2D3 exerts its inhibitory action on the renin gene remains a matter of speculation. 1,25(OH)2D3 may be one of the general “gate-keepers” to maintain an appropriate level of renin in the body, and/or may act as a counterbalance regulator to antagonize other renin-stimulating factors and prevent the detrimental overstimulation of the RAS. Generally speaking, long-term vitamin D deficiency may increase the risk of high blood pressure

878

YAN CHUN LI

UV light Ca++

PTH

Ca++

Renin

1,25(OH)2D3

Cardiovascular functions

Renal perfusion pressure Tubular sodium chloride load Sympathetic nerve activity

Angiotensin II

Blood pressure

Extracellular volume

FIGURE 6

Interaction between the vitamin D endocrine system and the renin– angiotensin system. 1,25(OH)2D3 feedback regulates the production of parathyroid hormone and suppresses renin biosynthesis. Renin is also feedback-suppressed by Ang II. Ultraviolet light influences blood pressure via 1,25(OH)2D3. 1,25(OH)2D3, PTH, and calcium may also directly affect the cardiovascular functions (dashed lines), as suggested by other studies. PTH, parathyroid hormone. (Adapted from Li 2003, with permission.)

and hypertension, whereas vitamin D supplement may be beneficial to the cardiovascular system.

VI. VITAMIN D ANALOGS AS POTENTIAL ANTIHYPERTENSIVE AGENTS As a major pathogenic contributor to hypertension, the RAS has been an important drug target for therapeutic intervention of hypertension, with ACE inhibitors and Ang II receptor antagonists being among the most popular anti-hypertensive drugs [67]. As high-renin hypertension accounts for 10–20% of the patient population with essential hypertension, specific inhibitors for renin production are of significant therapeutic values. Such inhibitors, in theory, can be used alone or in combination with ACE inhibitors or Ang II receptor antagonists. Patients with high-renin hypertension generally have higher blood pressure [68] and tend to have a more active sympathetic nervous system [69]; thus renin inhibitors may also be used with sympatholytic agents such as the β-blockers. Great efforts have been made in the past to develop angiotensin substrate analogs as renin inhibitors [70–72]. Unfortunately, these peptide renin inhibitors are toxic, and thus are of little use for administration to humans. The finding that 1,25(OH)2D3 suppresses renin biosynthesis has raised the possibility to develop vitamin D analogs into renin inhibitors for therapeutic purposes. With a large number of low calcemic vitamin D analogs synthesized, and some of them already approved for clinical applications [73,74], such a possibility is not unrealistic. In theory, vitamin D analogs

with equal or better potency but less calcemic effects than 1,25(OH)2D3 are good candidates. To search for such candidates, we have set to screen vitamin D analog compounds using the stable hVDRAs4.1 cells by Northern blot and renin promoter luciferase reporter assays. Interestingly, of the nine compounds we initially screened, only the two Gemini compounds, which have double side chains at the carbon 20 position (see Chapter 85) [75], display renin-suppressing activity equal to or better than that of 1,25(OH)2D3, whereas all other vitamin D analogs had little or much less inhibitory activity. Continued screening of 11 more Gemini compounds identified six more candidates. The reason why the double-side-chain Gemini compounds possess more potent activity in renin inhibition than other vitamin D analogs or even 1,25(OH)2D3 itself remains to be explored. The in vivo efficacy of these Gemini compounds is being tested in animals. Preliminary data show that some analogs can indeed significantly inhibit renin mRNA expression in mouse kidneys. Whether the Gemini compounds can reduce blood pressure needs to be further tested in high-renin hypertensive animal models before clinical trials are considered.

VII. CONCLUSION The discovery that 1,25(OH)2D3 functions as a negative endocrine regulator of the RAS reveals a novel physiological function of the vitamin D endocrine system. This finding has great physiological and pharmacological implications for the vitamin D hormone

CHAPTER 54 Vitamin D and the Renin–Angiotensin System

879

and its analogs. Given the importance of this subject, the role of vitamin D in the renocardiovascular system is worth further exploration in future studies. It is also important to elucidate the molecular mechanism underlying vitamin D suppression of renin gene expression. Finally, low-calcemic vitamin D analogs may open a new era for the long-sought therapeutic renin inhibitors and potentially offer a new class of antihypertensive drugs. For diseases such as chronic renal failure and end-stage renal disease, which are commonly associated with cardiovascular problems and secondary hyperparathyroidism, vitamin D analogs with activities to suppress renin and PTH production might offer multiple clinical benefits [74,76,77].

13. Carey RM, McGrath HE, Pentz ES, Gomez RA, Barrett PQ 1997 Biomechanical coupling in renin-releasing cells. J Clin Invest 100:1566–1574. 14. Holmer SR, Kaissling B, Putnik K, Pfeifer M, Kramer BK, Riegger GA, Kurtz A 1997 Beta-adrenergic stimulation of renin expression in vivo. J Hypertens 15:1471–1479. 15. Wagner C, Hinder M, Kramer BK, Kurtz A 1999 Role of renal nerves in the stimulation of the renin system by reduced renal arterial pressure. Hypertension 34:1101–1105. 16. Skott O, Briggs JP 1987 Direct demonstration of macula densa–mediated renin secretion. Science 237:1618–1620. 17. Abel KJ, Gross KW 1990 Physical characterization of genetic rearrangements at the mouse renin loci. Genetics 124:937–947. 18. Sigmund CD, Gross KW 1991 Structure, expression, and regulation of the murine renin genes. Hypertension 18: 446–457. 19. Pentz ES, Lopez ML, Kim HS, Carretero O, Smithies O, Gomez RA 2001 Ren1d and Ren2 cooperate to preserve homeostasis: evidence from mice expressing GFP in place of Ren1d. Physiol Genomics 6:45–55. 20. Ying L, Morris BJ, Sigmund CD 1997 Transactivation of the human renin promoter by the cyclic AMP/protein kinase A pathway is mediated by both cAMP-responsive element binding protein-1 (CREB)-dependent and CREB-independent mechanisms in Calu-6 cells. J Biol Chem 272:2412–2420. 21. Jones CA, Sigmund CD, McGowan RA, Kane-Haas CM, Gross KW 1990 Expression of murine renin genes during fetal development. Mol Endocrinol 4:375–383. 22. Jones CA, Hurley MI, Black TA, Kane CM, Pan L, Pruitt SC, Gross KW 2000 Expression of a renin/GFP transgene in mouse embryonic, extra-embryonic, and adult tissues. Physiol Genomics 4:75–81. 23. Sigmund CD, Okuyama K, Ingelfinger J, Jones CA, Mullins JJ, Kane C, Kim U, Wu CZ, Kenny L, Rustum Y, et al. 1990 Isolation and characterization of renin-expressing cell lines from transgenic mice containing a renin-promoter viral oncogene fusion construct. J Biol Chem 265:19916–19922. 24. Petrovic N, Black TA, Fabian JR, Kane C, Jones CA, Loudon JA, Abonia JP, Sigmund CD, Gross KW 1996 Role of proximal promoter elements in regulation of renin gene transcription. J Biol Chem 271:22499–22505. 25. Tamura K, Chen YE, Horiuchi M, Chen Q, Daviet L, Yang Z, Lopez-Ilasaca M, Mu H, Pratt RE, Dzau VJ 2000 LXRalpha functions as a cAMP-responsive transcriptional regulator of gene expression. Proc Natl Acad Sci USA 97:8513–8518. 26. Shi Q, Gross KW, Sigmund CD 2001 Retinoic acid-mediated activation of the mouse renin enhancer. J Biol Chem 276:3597–3603. 27. Pan L, Black TA, Shi Q, Jones CA, Petrovic N, Loudon J, Kane C, Sigmund CD, Gross KW 2001 Critical roles of a cyclic AMP responsive element and an E-box in regulation of mouse renin gene expression. J Biol Chem 276:45530–45538. 28. Pan L, Xie Y, Black TA, Jones CA, Pruitt SC, Gross KW 2001 An Abd-B class HOX.PBX recognition sequence is required for expression from the mouse Ren-1c gene. J Biol Chem 276:32489–32494. 29. Pan L, Glenn ST, Jones CA, Gronostajski RM, Gross KW 2003 Regulation of renin enhancer activity by nuclear factor I and Sp1/Sp3. Biochim Biophys Acta 1625:280–290. 30. Shi Q, Gross KW, Sigmund CD 2001 NF-Y antagonizes renin enhancer function by blocking stimulatory transcription factors. Hypertension 38:332–336. 31. Liu X, Huang X, Sigmund CD 2003 Identification of a nuclear orphan receptor (Ear2) as a negative regulator of renin gene transcription. Circ Res 92:1033–1040.

References 1. Hajjar I, Kotchen TA 2003 Trends in prevalence, awareness, treatment, and control of hypertension in the United States, 1988–2000. JAMA 290:199–206. 2. Moffett DF, Moffett SB, Schauf CL 1993 Human Physiology. Foundation and Frontiers, 2nd ed. Wm. C. Brown, Dubuque, IA. 3. Lavoie JL, Sigmund CD 2003 Minireview: overview of the renin–angiotensin system—an endocrine and paracrine system. Endocrinology 144:2179–2183. 4. Hackenthal E, Paul M, Ganten D, Taugner R 1990 Morphology, physiology, and molecular biology of renin secretion. Physiol Rev 70:1067–1116. 5. Berry C, Touyz R, Dominiczak AF, Webb RC, Johns DG 2001 Angiotensin receptors: Signaling, vascular pathophysiology, and interactions with ceramide. Am J Physiol Heart Circ Physiol 281:H2337–H2365. 6. Ballermann BJ, Zeidel ML, Gunning ME, Brenner BM 1991 Vasoactive peptides and the kidney. In: Brenner BM, Rector FC (eds) The Kidney, (4th ed). W.B. Saunders, Philadelphia, pp. 510–583. 7. Inagami T 1994 The renin–angiotensin system. Essays Biochem 28:147–164. 8. Nguyen G, Delarue F, Burckle C, Bouzhir L, Giller T, Sraer JD 2002 Pivotal role of the renin/prorenin receptor in angiotensin II production and cellular responses to renin. J Clin Invest 109:1417–1427. 9. Donoghue M, Hsieh F, Baronas E, Godbout K, Gosselin M, Stagliano N, Donovan M, Woolf B, Robison K, Jeyaseelan R, Breitbart RE, Acton S 2000 A novel angiotensin-converting enzyme-related carboxypeptidase (ACE2) converts angiotensin I to angiotensin 1-9. Circ Res 87:E1–E9. 10. Tipnis SR, Hooper NM, Hyde R, Karran E, Christie G, Turner AJ 2000 A human homolog of angiotensin-converting enzyme. Cloning and functional expression as a captopril-insensitive carboxypeptidase. J Biol Chem 275:33238–33243. 11. Crackower MA, Sarao R, Oudit GY, Yagil C, Kozieradzki I, Scanga SE, Oliveira-dos-Santos AJ, da Costa J, Zhang L, Pei Y, Scholey J, Ferrario CM, Manoukian AS, Chappell MC, Backx PH, Yagil Y, Penninger JM 2002 Angiotensin-converting enzyme 2 is an essential regulator of heart function. Nature 417:822–828. 12. Bader M, Ganten D 2000 Regulation of renin: New evidence from cultured cells and genetically modified mice. J Mol Med 78:130–139.

880 32. Rostand SG 1997 Ultraviolet light may contribute to geographic and racial blood pressure differences. Hypertension 30:150–156. 33. He J, Klag MJ, Wu Z, Whelton PK 1995 Stroke in the People’s Republic of China. I. Geographic variations in incidence and risk factors. Stroke 26:2222–2227. 34. Kunes J, Tremblay J, Bellavance F, Hamet P 1991 Influence of environmental temperature on the blood pressure of hypertensive patients in Montreal. Am J Hypertens 4:422–426. 35. Woodhouse PR, Khaw KT, Plummer M 1993 Seasonal variation of blood pressure and its relationship to ambient temperature in an elderly population. J Hypertens 11:1267–1274. 36. Holick MF 1987 Photosynthesis of vitamin D in the skin: Effect of environmental and life-style variables. Fed Proc 46:1876–1882. 37. Harburg E, Gleibermann L, Roeper P, Schork MA, Schull WJ 1978 Skin color, ethnicity, and blood pressure I: Detroit blacks. Am J Public Health 68:1177–1183. 38. Klag MJ, Whelton PK, Coresh J, Grim CE, Kuller LH 1991 The association of skin color with blood pressure in US blacks with low socioeconomic status. JAMA 265:599–602. 39. Krause R, Buhring M, Hopfenmuller W, Holick MF, Sharma AM 1998 Ultraviolet B and blood pressure. Lancet 352:709–710. 40. Kristal-Boneh E, Froom P, Harari G, Ribak J 1997 Association of calcitriol and blood pressure in normotensive men. Hypertension 30:1289–1294. 41. Lind L, Hanni A, Lithell H, Hvarfner A, Sorensen OH, Ljunghall S 1995 Vitamin D is related to blood pressure and other cardiovascular risk factors in middle-aged men. Am J Hypertens 8:894–901. 42. Burgess ED, Hawkins RG, Watanabe M 1990 Interaction of 1,25-dihydroxyvitamin D and plasma renin activity in high renin essential hypertension. Am J Hypertens 3:903–905. 43. Resnick LM, Muller FB, Laragh JH 1986 Calcium-regulating hormones in essential hypertension. Relation to plasma renin activity and sodium metabolism. Ann Intern Med 105:649–654. 44. Lind L, Wengle B, Wide L, Ljunghall S 1989 Reduction of blood pressure during long-term treatment with active vitamin D (alphacalcidol) is dependent on plasma renin activity and calcium status. A double-blind, placebo-controlled study. Am J Hypertens 2:20–25. 45. Pfeifer M, Begerow B, Minne HW, Nachtigall D, Hansen C 2001 Effects of a short-term vitamin D(3) and calcium supplementation on blood pressure and parathyroid hormone levels in elderly women. J Clin Endocrinol Metab 86:1633–1637. 46. Kimura Y, Kawamura M, Owada M, Oshima T, Murooka M, Fujiwara T, Hiramori K 1999 Effectiveness of 1,25-dihydroxyvitamin D supplementation on blood pressure reduction in a pseudohypoparathyroidism patient with high renin activity. Intern Med 38:31–35. 47. Park CW, Oh YS, Shin YS, Kim CM, Kim YS, Kim SY, Choi EJ, Chang YS, Bang BK 1999 Intravenous calcitriol regresses myocardial hypertrophy in hemodialysis patients with secondary hyperparathyroidism. Am J Kidney Dis 33:73–81. 48. Simpson RU, Thomas GA, Arnold AJ 1985 Identification of 1,25-dihydroxyvitamin D3 receptors and activities in muscle. J Biol Chem 260:8882–8891. 49. Fraga C, Blanco M, Vigo E, Segura C, Garcia-Caballero T, Perez-Fernandez R 2002 Ontogenesis of the vitamin D receptor in rat heart. Histochem Cell Biol 117:547–550. 50. Weishaar RE, Simpson RU 1987 Vitamin D3 and cardiovascular function in rats. J Clin Invest 79:1706–1712.

YAN CHUN LI

51. Weishaar RE, Kim SN, Saunders DE, Simpson RU 1990 Involvement of vitamin D3 with cardiovascular function. III. Effects on physical and morphological properties. Am J Physiol 258:E134–E142. 52. Weishaar RE, Simpson RU 1989 The involvement of the endocrine system in regulating cardiovascular function: emphasis on vitamin D3. Endocr Rev 10:351–365. 53. Weishaar RE, Simpson RU 1987 Involvement of vitamin D3 with cardiovascular function. II. Direct and indirect effects. Am J Physiol 253:E675–E683. 54. O’Connell TD, Weishaar RE, Simpson RU 1994 Regulation of myosin isozyme expression by vitamin D3 deficiency and 1,25-dihydroxyvitamin D3 in the rat heart. Endocrinology 134:899–905. 55. O’Connell TD, Berry JE, Jarvis AK, Somerman MJ, Simpson RU 1997 1,25-Dihydroxyvitamin D3 regulation of cardiac myocyte proliferation and hypertrophy. Am J Physiol 272:H1751–H1758. 56. Wu J, Garami M, Cheng T, Gardner DG 1996 1,25(OH)2 vitamin D3, and retinoic acid antagonize endothelin-stimulated hypertrophy of neonatal rat cardiac myocytes. J Clin Invest 97:1577–1588. 57. Li YC, Kong J, Wei M, Chen ZF, Liu SQ, Cao LP 2002 1,25Dihydroxyvitamin D3 is a negative endocrine regulator of the renin-angiotensin system. J Clin Invest 110:229–238. 58. Li YC 2003 Vitamin D regulation of the renin-angiotensin system. J Cell Biochem 88:327–331. 59. Kong J, Li YC 2003 Effect of angiotensin II type I receptor antagonist and angiotensin-converting enzyme inhibitor on vitamin D receptor null mice. Am J Physiol Regul Integr Comp Physiol 285:R255–R261. 60. Fitzsimons JT 1980 Angiotensin stimulation of the central nervous system. Rev Physiol Biochem Pharmacol 87:117–167. 61. Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997 Targeted ablation of the vitamin D receptor: an animal model of vitamin D–dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94:9831–9835. 62. Li YC, Amling M, Pirro AE, Priemel M, Meuse J, Baron R, Delling G, Demay MB 1998 Normalization of mineral ion homeostasis by dietary means prevents hyperparathyroidism, rickets, and osteomalacia, but not alopecia in vitamin D receptorablated mice. Endocrinology 139:4391–4396. 63. Gunther T, Chen ZF, Kim J, Priemel M, Rueger JM, Amling M, Moseley JM, Martin TJ, Anderson DJ, Karsenty G 2000 Genetic ablation of parathyroid glands reveals another source of parathyroid hormone. Nature 406:199–203. 64. Kong J, Li XJ, Gavin D, Jiang Y, Li YC 2002 Targeted expression of human vitamin D receptor in the skin promotes the initiation of postnatal hair follicular cycle and rescue the alopecia in vitamin D receptor null mice. J Invest Dermatol 118:631–638. 65. Silver J, Naveh-many T, Mayer H, Schmeizer HJ, Popovtzer MM 1986 Regulation by vitamin D metabolites of parathyroid hormone gene transcription in vivo in the rat. J Clin Invest 78:1296–1301. 66. Omdahl JL, DeLuca HF 1971 Strontium induced rickets: Metabolic basis. Science 174:949–951. 67. August P 2003 Initial treatment of hypertension. N Engl J Med 348:610–617. 68. Brunner HR, Laragh JH, Baer L, Newton MA, Goodwin FT, Krakoff LR, Bard RH, Buhler FR 1972 Essential hypertension: renin and aldosterone, heart attack and stroke. N Engl J Med 286:441–449. 69. Esler M, Julius S, Zweifler A, Randall O, Harburg E, Gardiner H, DeQuattro V 1997 Mild high-renin essential hypertension. Neurogenic human hypertension? N Engl J Med 296:405–411.

CHAPTER 54 Vitamin D and the Renin–Angiotensin System

881

70. Kokubu T, Ueda E, Fujimoto S, Hiwada K, Kato A 1968 Peptide inhibitors of the renin-angiotensin system. Nature 217:456–457. 71. Poulsen K, Burton J, Haber E 1975 Purification of hog renin by affinity chromatography using the synthetic competitive inhibitor (D-Leu6)octapeptide. Biochim Biophys Acta 400:258–262. 72. Burton J, Cody RJ Jr, Herd JA, Haber E 1980 Specific inhibition of renin by an angiotensinogen analog: studies in sodium depletion and renin-dependent hypertension. Proc Natl Acad Sci USA 77:5476–5479. 73. Brown AJ, Dusso AS, Slatopolsky E 2002 Vitamin D analogues for secondary hyperparathyroidism. Nephrol Dial Transplant 17(Suppl 10):10–19. 74. Malluche HH, Mawad H, Koszewski NJ 2002 Update on vitamin D and its newer analogues: actions and rationale for treatment in chronic renal failure. Kidney Int 62:367–374.

75. Uskokovic MR, Manchand PS, Peleg S, Norman AW 1997 Synthesis and preliminary evaluation of the biological properties of a 1α,25-dihydroxyvitamin D3 analog with two side chains. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D: Chemistry, Biology and Clinical Applications of the Steroid Hormone. University of California, Riverside, pp. 19–21. 76. Park CW, Oh YS, Shin YS, Kim CM, Kim YS, Kim SY, Choi EJ, Chang YS, Bang BK 1999 Intravenous calcitriol regresses myocardial hypertrophy in hemodialysis patients with secondary hyperparathyroidism. Am J Kidney Dis 33:73–81. 77. Rostand SG, Drueke TB 1999 Parathyroid hormone, vitamin D, and cardiovascular disease in chronic renal failure. Kidney Int 56:383–392.

CHAPTER 55

Vitamin D and Muscle RICARDO L. BOLAND

Departamento de Biología, Bioquímica y Farmacia, Universidad Nacional del Sur, (8000), Bahía Blanca, Argentina

I. II. III. IV.

Introduction Vitamin D–Dependent Myopathies Muscle Vitamin D Receptor 1α,25(OH)2D3 Regulation of Calcium Homeostasis in Muscle Cells V. Modulation of Muscle Cell Phosphate Uptake by Vitamin D3 Metabolites

VI. Effect of 1α,25(OH)2D3 on Muscle Cell Proliferation and Differentiation VII. Mechanism of Action of 1α,25(OH)2D3 in Muscle VIII. Summary References

I. INTRODUCTION

the clinical issues of vitamin D action on muscle the reader is referred to Chapter 102 of this book.

Vitamin D plays a major role in the regulation of vertebrate calcium and phosphorus metabolism by acting at the level of intestine, bone, and kidney [1,2]. As discussed in detail in Chapter 2 of this book, vitamin D3 from either dietary sources or endogenous synthesis by ultraviolet irradiation of the skin is transformed into the most active hormonal form, 1α,25-dihydroxyvitamin D3 (1α,25(OH)2D3), by 25-hydroxylation in the liver followed by 1α-hydroxylation of 25-hydroxyvitamin D3 (25(OH)D3) in the kidney. 1α,25(OH)2D3 elicits its effects through two different mechanisms. In addition to regulating gene transcription via its specific intracellular receptor (VDR) like other steroid hormones [3,4], 1α,25(OH)2D3 induces rapid, nontranscriptional responses involving activation of transmembrane signal transduction pathways such as growth factors and peptide hormones [5,6]. In the past few years, 1α,25(OH)2D3 receptors have been identified in a wide range of tissues and cell lines, implying the hormone in effects not directly related to mineral metabolism. Thus, 1α,25(OH)2D3, among various nonclassical actions, may influence the proliferation and differentiation of various cell types, has immunoregulatory properties and modulates insulin secretion and prolactin synthesis [7,8]. Several lines of evidence have demonstrated that skeletal muscle is also a target tissue for vitamin D [9]. This chapter reviews and updates relevant experimental research performed to characterize cellular processes and mechanisms involved in the effects of vitamin D on muscle. These issues will be preceded by a description of essential clinical aspects of vitamin D–dependent myopathies in order to provide a background that will allow to point out the clinical relevance of the basic scientific concepts discussed. For a detailed account of VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

II. VITAMIN D–DEPENDENT MYOPATHIES A. Clinical Background Early clinical observations suggested a relationship between vitamin D and muscle. These studies have shown that a myopathy characterized by muscle weakness, followed by atrophy, is a common symptom in vitamin D deficiency states of various origins [10–13]. The main clinical feature of the myopathy associated with osteomalacia consists of predominantly proximal muscle weakness, which often gives rise to a waddling gait. Muscle wasting without fasciculation or depression of the tendon reflexes may be observed. Plasma muscle enzyme profiles are generally unaltered and only slight nonspecific histopathological abnormalities may be detected [13,14]. Electromyographic evaluation of patients reveals a myopathic pattern as evidenced by a significant reduction in motor unit potential duration and amplitude, and an increased percentage of polyphasicity as compared to controls [11,12]. A similar myopathy has been observed in patients with postgastrectomy vitamin D deficiency, those with idiopathic hypophosphatemic osteomalacia, and in gluten-sensitive enteropathy [13]. Muscle weakness responds to the treatment with vitamin D3 suggesting that the sterol plays an etiological role. Moreover, it has been shown that osteomalacic myopathy and neuropathy are not interrelated [12]. Patients with end-stage renal failure also develop proximal muscle weakness. The myopathy is demonstrable by quantitative electromyography and its histological characterization reveals selective atrophy of type II muscle fibers. Electron microscopy may show in addition Copyright © 2005, Elsevier, Inc. All rights reserved.

884 nonspecific degenerative changes [14]. The myopathy may be considerably improved by the administration of small amounts of 1α,25(OH)2D3 [15,16] and renal transplantation [14], whereas treatment with large doses of vitamin D3 causes only moderate improvement [14]. This suggests that impaired synthesis of 1α,25(OH)2D3 may play a role in the development of muscle weakness in renal diseases. In disagreement with this interpretation, prior work reported a high incidence of a similar myopathy in patients with primary hyperparathyroidism [17]. However, this observation could not be confirmed in two subsequent investigations [10,11]. At that early stage of the research it was proposed that hypophosphatemia was the basic disorder underlying these myopathies as it is a common feature of primary hyperparathyroidism, osteomalacia, and nutritional vitamin D deficiency [18]. Contrary to this proposal it was noted that the myopathy of hypophosphatemic osteomalacia may not resolve with increased phosphate intake, and vitamin D intake is required for recovery. Moreover, the fact that muscle weakness is absent in X-linked hypophosphatemic rickets has been taken as indication that mere hypophosphatemia need not be the etiological factor involved [13]. In addition, 1α,25(OH)2D3 levels are low [19], and it has been pointed out that intracellular phosphate level may remain normal or even higher in spite of severe hypophosphatemia [18].

B. Animal Model Studies The myopathy associated with vitamin D deficiency has been experimentally characterized. First, in electrophysiological studies using an in situ neuromuscular preparation of the soleus, Rodman and Baker [20] found diminished peak tension and prolonged time for recovery halfway to resting tension for single twitches and prolonged relaxation half-life after tetanic contraction in vitamin D–deficient rats. Impaired tension development and slow relaxation after muscle contraction in response to repetitive electrical simulation were also observed in vitamin D–depleted chicks [21]. Interestingly enough, both studies clearly showed that these changes were not related to modifications in blood Ca and P levels and could only be reversed to normal by administration of vitamin D3 [20,21]. These data indicate that vitamin D–dependent muscle weakness may be ascribed to a primary disorder of skeletal muscle function. Studies with animal models have provided information pointing to alterations in muscle calcium metabolism, and to a lesser extent in phosphate metabolism, as etiological factors responsible for vitamin D–dependent myopathies.

RICARDO L. BOLAND

In addition, there is evidence which suggests that impairment of the structure and functional properties of the actomyosin contractile complex may also contribute to muscle dysfunction [9]. The contraction–relaxation cycle is regulated by the concentration of free Ca2+ ions in the sarcoplasm [22–24]. Several data obtained in vivo indicate that vitamin D acts on muscle calcium transport. Ca uptake by the sarcoplasmic reticulum (SR) is diminished in vitamin D–depleted rabbits [25] and chicks [21]. 1α,25(OH)2D3 may be the active metabolite on SR as nephrectomy or strontium feeding (inhibits 1α,25(OH)2D3 synthesis in kidney) in rabbits and chicks, respectively, decreased Ca uptake. Dosage of animals with 1α,25(OH)2D3 reversed these changes [26–29]. 1α,25(OH)2D3 may affect calcium transport across sarcoplasmic reticulum membranes by increasing the number and kinetics of formation of active transport sites of the SR Ca-ATPase. Steady-state levels of phosphoderivative (EP, enzyme phosphate) and phosphorylation velocity of the transport enzyme are decreased in experimental uremia. Administration of 1α,25(OH)2D3 restores both parameters to normal [28,29]. It has been reported that vitamin D3 also affects Ca fluxes through mitochondrial membranes from skeletal muscle in vivo [21]. In addition, an action of vitamin D3 on Ca transport across muscle plasma cell membranes has been shown. Administration of the sterol to vitamin D– deficient chicks markedly increased Ca uptake in subsequently isolated sarcolemma vesicles. This change was accompanied by an increase in Ca-ATPase activity that could in turn be related to increased Vmax and decreased Km [30]. Sarcolemmal Ca transport probably reflects the Ca-ATPase activity of vesicles with insideout membranes as the physiological function of the pump is to extrude Ca2+ from the cells [30]. Desaturation analysis of rachitic chick soleus muscle Ca pools prelabeled with 45Ca suggested that 1α,25(OH)2D3 stimulates Ca efflux from mitochondria and Ca transport (influx and efflux) across plasma cell membranes [31]. The action of 1α,25(OH)2D3 on muscle Ca fluxes may contribute to extracellular calcium homeostasis in conditions of vitamin D deficiency. It has been reported that vitamin D depletion of chicks leads to Ca accumulation in muscle tissue. Treatment with a single dose of 1α,25(OH)2D3 quickly reversed this change. The prompt increase in blood Ca that followed administration of the steroid correlated better with the fall in Ca content of muscle tissue than with stimulation of intestinal Ca absorption [32]. Increased Ca efflux from mitochondria and cytoplasm as revealed by the studies with soleus muscle may provide a mechanism for the 1α,25(OH)2D3 action on Ca mobilization from skeletal muscle of rachitic chicks.

885

CHAPTER 55 Vitamin D and Muscle

An action of vitamin D on muscle phosphate fluxes was first suggested by Birge and Haddad [33], who by administration of vitamin D3 or 25(OH)D3 to vitamin D– deficient, phosphate-depleted rats, detected first a fall in serum P and an increase in 32P uptake by muscle, followed by an increase in muscle ATP and protein synthesis. Nephrectomy did not obliterate these responses, indicating that further conversion of the sterols to 1α,25(OH)2D3 was not required. On the basis of this evidence it was proposed that a reduction in muscle intracellular phosphate may contribute to osteomalacic myopathy [18]. In agreement with these observations, measurement of 32P specific activities in serum and skeletal muscle sarcoplasm and intracellular membranes after in vivo 32P-labeling of vitamin D–deficient and acutely vitamin D3–treated chicks suggested that the sterol stimulates phosphate transport across muscle plasma membranes. Furthermore, prior treatment of vitamin D–depleted chicks with vitamin D3 resulted in a stimulation of Na+ gradient–mediated phosphate transport in isolated sarcolemma vesicles [30]. Studies in vivo have implied vitamin D3 in the synthesis of skeletal muscle contractile proteins. A significant reduction in actomyosin levels has been noted in rats fed a vitamin D–deficient diet [34]. Moreover, treatment of vitamin D–depleted chicks with vitamin D3 caused an increase in actin and troponin C levels [35]. Pointon et al. [36] also observed a reduction in troponin C in rachitic rabbits. The changes observed may also contribute to impaired tension development in vitamin D deficiency. Ca binding to troponin C initiates the mechanism of muscle contraction in vertebrate striated muscle [37]. Vitamin D3 induces, in addition, changes in skeletal muscle phospholipids, which may represent part of the mechanism involved in the effects of the sterol on the transport properties of muscle membranes. Thus, an increase in phosphatidylcholine content at the expense of a proportional decrease in phosphatidylethanolamine levels is observed in sarcolemmal membranes after administration of vitamin D3 to vitamin D–depleted animals [30]. Vitamin D3 also increases in vivo the total phospholipid content of sarcoplasmic reticulum and mitochondria without affecting the relative proportions of phospholipid classes [35]. This may reflect increased availability of sarcoplasmic phosphate due to the action of vitamin D3 metabolites on sarcolemmal phosphate transport [30,33]. The preceding information on vitamin D3 regulation of skeletal muscle ion fluxes, and protein and lipid composition, obtained with animal models, does not necessarily imply a direct action of the sterol on muscle tissue, since the contribution of systemic effects, e.g., changes in blood Ca, P, and PTH levels, cannot be

totally excluded, preventing an accurate description of the mechanisms involved and the identification of active vitamin D3 metabolite(s). However, more recent experimental studies using skeletal muscle cell culture systems (myoblasts/ myotubes) have contributed substantially to identify 1α,25(OH)2D3 as the main biologically active form of vitamin D3 acting in muscle, and to elucidate the mechanisms by which this hormonal metabolite regulates muscle cell Ca2+ and phosphate, proliferation, and differentiation, providing thereby a new basis for the abnormalities in skeletal muscle contractility and growth that occur in states of vitamin D deficit. The following sections will deal with these aspects.

III. MUSCLE VITAMIN D RECEPTOR There is evidence for the presence in muscle cells of an intracellular receptor for 1α,25(OH)2D3, analogous to that found in classical tissues (VDR). Density gradient analysis of [3H]1α,25(OH)2D3 binding by cytosol from primary-cultured chick myoblasts revealed that the hormone specifically binds to a 3.7 S macromolecule, and saturation analysis of the binding showed that the receptor proteins binds 1α,25(OH)2D3 with high affinity (KD 2.46 × 10−10 M) and low capacity (74 fmol/mg protein) [38]. The presence of a specific 1α,25(OH)2D3 binding protein with similar characteristics in murine and human growing myoblasts and myotubes has been reported [39,40]. Additional data on the expression of the VDR in avian muscle cells has been obtained by detection of VDR-mRNA by RT-PCR, Western blot, and immunocytochemical analysis using highly specific antibodies [41,42]. Receptor levels are low in undifferentiated myoblasts and significantly increase during the differentiation process to myotubes [42]. Notably, it has been shown that treatment of chick muscle cells with 1α,25(OH)2D3 promotes tyrosine phosphorylation of the VDR, in agreement with the database for protein consensus motifs indicating the presence within the primary sequence of the chick VDR of a putative tyrosine phosphorylation site corresponding to amino acids 164–170 (KTFDTTY), a region located near the C-terminal end of the receptor DNA binding domain [43]. Similar putative tyrosine phosphorylation sites are also found in human, rat, and mouse VDR sequences (Boland, unpublished observations). All steroid hormone receptors, including the VDR, are phosphorylated and undergo ligand-induced hyperphosphorylation [44]. However, most of the phosphorylated residues identified to date are serines in the N-terminal motif, and tyrosine phosphorylation has only been documented for the estrogen receptor [45].

886 Phosphorylation of serine residues in the human VDR has been involved in 1α,25(OH)2D3-dependent transcriptional activation [46,47]. Recent evidence suggests that phosphorylation of tyrosine residues in the VDR may play a role in hormone modulation of muscle mitogenic tyrosine phosphorylation cascades (Section VI) and Ca2+ influx through store operated channels (Section VII,B).

IV. 1α,25(OH)2D3 REGULATION OF CALCIUM HOMEOSTASIS IN MUSCLE CELLS It is possible that alterations in the mechanisms by which intracellular Ca2+ is regulated in the muscle cell play a major role in the muscle weakness associated with vitamin D deficiency and renal diseases as suggested by the animal model studies described in Section II,B. Cultured avian embryonic skeletal muscle cells have proven to be an adequate model to further characterize 1α,25(OH)2D3 regulation of muscle intracellular Ca2+ homeostasis. Chick myoblasts differentiate in vitro into functionally competent myotubes that are endowed with a VDR and the molecular machinery to respond to 1α,25(OH)2D3 through both the genomic and nongenomic mechanisms that mediate the actions of the hormone (Sections III, VI, and VII). Long-term effects of 1α,25(OH)2D3 on Ca2+ transport in chick skeletal muscle cells have been scarcely studied. Thus, it has been only reported that physiological levels of 1α,25(OH)2D3 increase myoblast 45Ca2+ uptake after 8–24 hr treatment of cultures [9], probably in line with the operation of a nuclear mechanism that involves sterol-induced de novo synthesis in chick muscle of calbindin-D9K (instead of calbindin-D28K as in other avian tissues) and enzymes that convert phosphatidylethanolamine to phosphatidylcholine, according to the information provided in Section VII,A. The nongenomic regulation of myoblast/myotube intracellular Ca2+ by 1α,25(OH)2D3 has also been well characterized. Various aspects of this mechanism have been reproduced using intact differentiated skeletal muscle in vitro. Indeed, studies with isolated chick soleus muscle demonstrated for the first time that 1α,25(OH)2D3 exerts acute effects (1–15 min) on 45Ca2+ uptake, which are not blocked by inhibitors of RNA and protein synthesis but are suppressed by blockers of voltage-dependent calcium channels (VDCC) [48]. Interestingly, this rapid stimulation of skeletal muscle calcium fluxes in response to 1α,25(OH)2D3 has also been observed in vivo [32]. In myoblasts the operation of this mechanism has been firmly established by the fact that 1α,25(OH)2D3–dependent Ca2+ influx could

RICARDO L. BOLAND

be mimicked by the VDCC agonist BAY K8644, and plasma membrane depolarization induced by a high K+ medium. The effects of the combined treatment with 1α,25(OH)2D3 and BAY K8644 or K+ depolarization were not additive. Moreover, the action of the hormone was dependent on extracellular Ca2+ as they were reversibly inhibited by the Ca2+ chelator EGTA. Furthermore, on the basis of their sensitivity to nifedipine and verapamil, the voltage-dependent Ca2+ channels activated in skeletal muscle cells by 1α,25(OH)2D3 were pharmacologically identified as from the L-type [49,50]. Various lines of evidence described in Section VII,B have demonstrated that the hormone rapidly regulates muscle cell Ca2+ influx by G protein–mediated activation of both phospholipase C and adenylyl cyclase, leading to the stimulation of PKC and PKA and activation of VDCC. PKA and PKC modulate the activity of VDCC by phosphorylation, thus increasing the probability of channel opening [51]. 1α,25(OH)2D3 stimulates the phosphorylation of several proteins in muscle cells [50,52]. Identification of these proteins may help to further understand the regulatory action of the hormone on VDCC. Interestingly, both in myoblasts and soleus muscle it has been shown that in response to 1α,25(OH)2D3, calmodulin rapidly translocates from the cytosol to the membrane where it binds a major hormone-dependent phosphoprotein of 28 kDa. There is evidence suggesting that this event mediates the 1α,25(OH)2D3 fast increase in calcium entry through the dihydropyridine-sensitive pathway [50, and references therein]. Spectrofluorometric studies with Fura-2-loaded muscle cells have confirmed the foregoing observations and revealed additional key information related to the regulation of muscle intracellular Ca2+ homeostasis by 1α,25(OH)2D3. The cytosolic Ca2+ response to the hormone involves an initial rapid sterol-induced Ca2+ mobilization from IP3/thapsigargin-sensitive stores followed by cation influx from the extracellular milieu, accounting for a sustained Ca2+ phase that does not return to baseline as long as the cells are exposed to the sterol. This Ca2+ influx was shown to be contributed not only by the well established L-type VDCC-mediated Ca2+ entry but also by a store-operated Ca2+ (SOC; capacitative Ca2+ entry, CCE) channel, therefore introducing a novel aspect into the mechanism of 1α,25(OH)2D3-induced Ca2+ influx across the plasma membrane of muscle cells [53,54]. The effects of the hormone on the profile of changes in intracellular Ca2+ levels are clearly detectable within the 10−12 M–10−8 M concentration range [55]. The SOC influx activated by 1α,25(OH)2D3 was identified by being insensitive to L-type Ca2+ channel antagonists but was fully inhibitable by low micromolar concentrations of

887

CHAPTER 55 Vitamin D and Muscle

La3+ (3 µM) and Ni2+. PI(polyphosphoinositide)-specific PLC blockade prior to 1α,25(OH)2D3 stimulation suppressed both the cytosolic Ca2+ transient and SOC influx. Accordingly, depletion of intracellular Ca2+ stores by thapsigargin reproduced 1α,25(OH)2D3-induced Ca2+ influx, inhibiting any further response to the hormone. Furthermore, 1α,25(OH)2D3 increased the rate of quenching of Fura-2 fluorescence by Mn2+, indicating activation of Mn2+ influx that specifically permeates SOC channels [53,54]. There is evidence on the involvement of protein kinases in the regulation of capacitative calcium influx in muscle cells by 1α,25(OH)2D3. It has been shown that the hormone stimulation of CCE is prevented by inhibitors of PKC (calphostin C, bisindolylmaleimide) and tyrosine kinase (genistein) but unaffected by blockade of the PKA pathway [54]. Of relevance, SOC influx stimulated by 1α,25(OH)2D3 is insensitive to both calmodulin (CAM) antagonists and CAMdependent protein kinase II (CAMKII) inhibitors when added after the IP3-mediated Ca2+ transient but completely abolished when added before it. Moreover, in cells microinjected with antisense oligonucleotides directed against the CAM mRNA the sterol-stimulated SOC influx is reduced up to 60% with respect to uninjected cells [56]. These results suggest that the 1α,25(OH)2D3-induced (IP3-mediated) cytosolic Ca2+ transient is required for CAM activation, which in turn mediates SOC influx in a mechanism that seems to include CAMKII.

V. MODULATION OF MUSCLE CELL PHOSPHATE UPTAKE BY VITAMIN D3 METABOLITES [32P]Phosphate uptake by cultured chick myoblasts is to a great extent Na+-dependent, saturable with respect to phosphate, energy-dependent, and inhibited by ouabain and arsenate, in agreement with the operation of a Na+–phosphate cotransport system in the muscle cell plasma membrane as described for intestine and kidney [57]. Treatment of myoblast cultures with 1α,25(OH)2D3 for 4–24 hr causes a significant dosedependent (10−10–10−7 M) stimulation of phosphate accumulation by the cells, increasing the velocity of phosphate uptake to a greater extent than the total capacity. 25(OH)D3 increases phosphate accumulation by myoblasts roughly to the same extent as 1α,25(OH)2D3, whereas 24,25(OH)2D3 and vitamin D3 are devoid of activity [58,59]. These results are in partial agreement with previous work of Birge [18] showing 25(OH)D3 stimulation of phosphate influx by rat epitrochlear muscle cultures. Our studies also provided

evidence that 1α,25(OH)2D3 affects the Na+-linked component of myoblast phosphate uptake involving de novo RNA and protein synthesis, suggesting that 1α,25(OH)2D3 acts on embryonic muscle phosphate uptake through a receptor-mediated mechanism. Information is lacking on postuptake phosphate metabolism and its possible regulation by vitamin D3 metabolites in muscle cells.

VI. EFFECT OF 1α,25(OH)2D3 ON MUSCLE CELL PROLIFERATION AND DIFFERENTIATION 1α,25(OH)2D3 regulates the proliferation and differentiation of several cell types, including myeloid leukemia and hemopoietic cells [60,61], chondrocytes [62], and keratinocytes [63]. It was proposed that the 1α,25(OH)2D3 receptor complex activates or represses genes related to the cellular cycle [64], for example the protooncogenes c-myc, c-fos, and c-jun [65,66]. In addition, data were obtained indicating that transcriptional regulation of protein kinase C is secondarily involved in the control of c-myc expression by 1α,25(OH)2D3 [66]. There is experimental evidence demonstrating that 1α,25(OH)2D3 also regulates muscle cell proliferation and differentiation, in keeping with the clinical observation of muscle atrophy in states of 1α,25(OH)2D3 deficit (Section II). The action of the hormone on myogenesis has been characterized in chick myoblast cultures. Myoblasts are mononucleated cells that proliferate actively in culture followed by differentiation into multinucleated myotubes expressing phenotypic characteristics of mature muscle fibers [67,68]. Specifically, the morphological profiles of myoblasts cultured for 1–6 days indicate that undifferentiated muscle cells elongate, become aligned, and fuse to form differentiated myotubes as the culture period progresses. Accordingly, in parallel to these changes the rate of DNA synthesis decreases 10-fold during the 1- to 6-day interval and creatine kinase activity augments 10- to 11-fold. In addition, myosin levels markedly increase from 2 to 6 days of culture during myogenesis [69]. Protein kinase C (PKC) is involved in myogenesis, the isoform α playing a pivotal role. Thus, high levels of PKCα are detected in the proliferative stage that markedly diminish as myoblasts differentiate. In addition, down-regulation of PKCα with a phorbol ester inhibits DNA synthesis in dividing myoblasts [69]. Furthermore, specific blockage of the expression of PKCα in myoblasts by using antisense oligodeoxynucleotides (ODNs) results in a significant decrease of culture cell density and DNA synthesis, clearly

888 showing that this isoenzyme is involved in signaling pathways that promote muscle cell proliferation [70]. Using as criteria the just-described changes in morphology and specific biochemical markers of myogenesis, it has been clearly shown that physiological levels of 1α,25(OH)2D3 stimulate muscle cell proliferation and differentiation [31,71–73]. In accord with the involvement of PKC in the regulation of myogenesis, increased activity is observed during 1α,25(OH)2D3 stimulation of myoblast proliferation, whereas inhibition of PKC activity accompanied the effects of the hormone on myoblast differentiation; the specific PKC inhibitor calphostin suppressed hormone potentiation of DNA synthesis in proliferating myoblasts [73,74]. Moreover, the early stimulation of myoblast proliferation mainly correlated to PKCα expression, whereas decreased PKCα levels were observed during the subsequent activation of myoblast differentiation [73]. Recent studies have revealed that activation of tyrosine phosphorylation pathways plays an important role in the mechanism that mediates the effects of 1α,25(OH)2D3 on muscle growth. Tyrosine phosphorylation is a crucial event in signal transduction linked to the mitogen-activated protein kinase (MAPK). Stimulation of the MAPK cascade may occur through activation of receptor tyrosine kinases or G proteincoupled receptors by stimulation of nonreceptor Src kinases or by direct signaling to Raf via PKC. Upon phosphorylation by mitogens, MAPK is translocated from the cytoplasm into the nucleus, which results in the activation or induction of transcription factors leading to the expression of genes involved in the control of cellular growth [75]. 1α,25(OH)2D3 rapidly (within 1 min) promotes tyrosine phosphorylation of MAPK in cultured myoblasts. Other major tyrosine-phosphorylated targets of the hormone in muscle cells are PLCγ, and surprisingly, the c-myc oncoprotein, a novel finding for which there is no information on the signaling component(s) leading to phosphorylation of its tyrosine residues [76]. Fast 1α,25(OH)2D3-dependent increased Src kinase activity has been observed in myoblasts [43]. Preincubation of muscle cells with specific Src inhibitors or their transfection with an antisense ODN against Src mRNA inhibits 1α,25(OH)2D3 activation of MAPK, involving Src as an upstream element that leads to hormone signaling through this cascade [77]. Coimmunoprecipitation analysis have provided evidence that 1α,25(OH)2D3 promotes the formation of complexes between Src and the VDR, and between Src and c-myc, which can be explained by the fact that both the VDR and c-myc behave as 1α,25(OH)2D3-dependent tyrosine phosphorylated proteins (see Section III for the VDR) and may

RICARDO L. BOLAND

interact with Src through the SH2 domain of the latter. Preincubation of myoblasts with a pool of different antisense ODNs against the VDR mRNA (AS-VDR ODNs) significantly reduces Src stimulation, further implying the VDR in hormone activation of Src [78]. However, MAP kinase tyrosine phosphorylation by 1α,25(OH)2D3 is affected to a lesser extent by transfection with AS-VDR ODNs implying that both VDRdependent and VDR-independent signaling mediate hormone stimulation of MAPK [78,79]. In agreement with this interpretation, using intracellular and extracellular Ca2+ mobilizing agents and chelators, as well as specific PKC activators and inhibitors, it has been shown that calcium and protein kinase C are also involved in the stimulation of MAP kinase by 1α,25(OH)2D3 [80]. Recent investigations [81] have established the role of PKC and uncovered other metabolic steps that participate in hormone up-regulation of the MAPK cascade. Thus, 1α,25(OH)2D3 causes a fast significant increase of Raf-1-serine phosphorylation, indicating activation of Raf-1 by the hormone. The PKC inhibitors calphostin C, bisindolylmaleimide I, and Ro 318220 blocked 1α,25(OH)2D3-induced Raf-1 serine phosphorylation, revealing that PKC participates in hormone stimulation of MAPK at the level of Raf-1. Moreover, application of antisense oligonucleotide technology revealed that PKCα specifically mediates this action. In addition, by using a specific Ras peptide inhibitor, Ras has also been involved in 1α,25(OH)2D3 activation of Raf-1. The hormone rapidly induced tyrosine dephosphorylation of RasGTPase-activating protein, suggesting that inhibition of Ras-GTP hydrolysis is part of the mechanism by which 1α,25(OH)2D3 activates Ras in myoblasts [81]. Stimulation of tyrosine phosphorylation cascades by 1α,25(OH)2D3 through the just-described mechanisms causes translocation of MAPK from the cytoplasm to the nucleus in an active phosphorylated form and induces the expression of the growth-related protein c-myc, as the MAPK kinase (MEK) inhibitor PD98059 abolishes stimulation of c-myc synthesis by 1α,25(OH)2D3 [77]. Early studies had already shown that the mitogenic effects of the hormone in myoblasts are correlated to increased c-myc mRNA levels [71]. There is also information available on the signaling cascade leading to PLCγ activation, a relevant event that may account for the activation of PKCα (followed by that of Raf-1) via release of DAG and IP3-mediated Ca2+. Investigations based on the utilization of specific inhibitors and antisense technology have involved Src and phosphatidylinositol 3-kinase (PtdIns3K) in 1α,25(OH)2D3 stimulation of PLCγ tyrosine phosphorylation and its translocation to the cell membrane. Evidence has been obtained indicating that the

889

CHAPTER 55 Vitamin D and Muscle

1α,25(OH)2D3

Plasma membrane

Cγ

PL

y

y Ras GAP

GTP

−

Raf–1

GDP

S

Ras

PI–3K

PKCα

y

c-Src

Shc

Sos

Grb2

GTP GDP S

MEK

y

c-myc

T

VDR

Y

MAPK T

Nuclear membrane

Cellular proliferation c-myc expression

FIGURE 1 Tyrosine phosphorylation cascades involved in the mitogenic effects of 1α,25(OH)2D3 in skeletal muscle cells. PI-3K, phosphatidylinositol 3-kinase; PLCγ, phospholipase Cγ; Ras-GAP, Ras-GTPase-activating protein; MAPK, mitogen activated protein kinase; MEK, MAPK kinase.

hormone increases the physical association of Src and PtdIns3K with PLCγ and induces a Src-dependent tyrosine phosphorylation of the p85 regulatory subunit of PtdIns3K [82]. Altogether, 1α,25(OH)2D3 stimulates in muscle cells an intrincate network of signaling components and interacting pathways (Fig. 1), which provide a mechanism underlying the regulation of muscle cellular growth by the hormone.

VII. MECHANISMS OF ACTION OF 1α,25(OH)2D3 IN MUSCLE A. Genomic Mechanism Congruent with the presence of the VDR in myoblasts and myotubes (Section III), various lines of evidence have shown that 1α,25(OH)2D3 modulates the expression of genes related to the regulation of muscle calcium transport and phospholipid metabolism. The increase in cell 45Ca2+ uptake elicited by the incubation of cultured chick myoblasts/myotubes with the hormone for long treatment intervals (Section III) can be

abolished by actinomycin D and cycloheximide, suggesting that it is dependent on de novo RNA and protein synthesis [83]. Temporally correlated to the 1α,25(OH)2D3-induced changes in Ca2+ transport, an increase in phosphatidylcholine (PC) at the expense of phosphatidylethanolamine (PE) levels is observed in response to the hormone, which is mainly accounted for by up-regulation of PE-N-methyltransferases [83,84]. The effects of 1α,25(OH)2D3 on phospholipid metabolism are also abolished by both actinomycin D and cycloheximide, reflecting the participation of a nuclear mechanism [85,86]. In agreement with these observations, treatment of myoblasts with 1α,25(OH)2D3 for 24 hr results in a marked increase in the incorporation of [3H]leucine into total cell proteins, which is abolished by the addition of cycloheximide or actinomycin D [87]. SDS-PAGE coelectrophoresis of [14C]leucine- and [3H]leucinelabeled proteins from control and 1α,25(OH)2D3-treated chick embryo skeletal muscle myoblasts, respectively, has shown that the hormone preferentially stimulates the synthesis of seven proteins. These myoblast proteins have been partially characterized on the basis of their

890

RICARDO L. BOLAND

molecular weight (9–100 kDa), isoelectric points, Ca2+ binding properties, carbohydrate content, and subcellular localization [87]. Of interest, four of the proteins induced by 1α,25(OH)2D3 bind Ca2+, one of them characterized as a cytosolic 9-kDa macromolecule with an acidic isoelectric point. The 9-kDa myoblast calciumbinding protein (CaBP) comigrates with 125I-labeled rat calbindin-D9K in SDS-PAGE gels, and Western blot analysis with a specific rat intestine calbinding-D9K antibody revealed the presence of an immunoreactive protein of 9 kDa in chick embryo myoblasts treated with 1α,25(OH)2D3 as well as in chick skeletal muscle [88]. 1α,25(OH)2D3 induces a two- to fourfold increase in 9 kDa CaBP mRNA levels after 20–24 hr treatment, coincident with maximum Ca uptake responses generated by the hormone. Northern hybridization analysis using a specific rat calbindin-D9K probe showed that the increase was related to a single 550-nucleotide mRNA species, in accordance with the reported size for rat calbinding-D9K mRNA [88]. However, very low concentrations of mRNA are usually detected, suggesting that the 9-kDa myoblast protein is the stable product of translation of a transient rare mRNA. Further molecular evidence on the expression of calbindin-D9K mRNA in muscle cells as well as in other tissues from the chick has been recently obtained by combined reverse transcription and polymerase chain reaction.

Subcloning and sequencing of a partial 160-bp cDNA PCR product has shown that the cDNA corresponds to calbindin D9K-cDNA [89,90], documenting for the first time the expression of the calbindin-D9K gene in an avian species. Calbindin D-28K was thought heretofore to be the only calcium-binding protein expressed in the chick in response to 1α,25(OH)2D3 [2,7]. An unresolved issue is the identification of the remaining gene products regulated by 1α,25(OH)2D3 in muscle cells, e.g., other Ca2+ binding proteins (17, 40, and 100 kDa; [87]), enzymes of phospholipid metabolism.

B. Nongenomic Signal Transduction Pathways There is a wealth of information showing that 1α,25(OH)2D3 also exerts nongenomic actions at the level of transmembrane second messenger systems, which mediate the fast effects of the hormone on muscle intracellular Ca2+ regulation (Fig. 2). It has been shown that 1α,25(OH)2D3 within the physiological concentration range modifies the activity of myoblast phospholipases in a mode independent of the nucleus. Within seconds to minutes the hormone activates phospholipase C (PLC) generating the second messengers inositol trisphosphate (IP3, a Ca2+ mobilizer)

1,25(OH)2D3

mVDR?

PIP2

PC

mVDR?

PLA2

PLC PLD

DAG

mVDR? AC Gi

IP3 Arachidonic acid

cAMP

ER Ca2+

Ca2+

PKA

VDCC Nucleus

VDCC

SOC

FIGURE 2

Nongenomic signal transduction pathways activated by 1α,25(OH)2D3 in skeletal muscle cells and their relationship to intracellular Ca2+ regulation. mVDR, novel membrane 1α,25(OH)2D3 receptor; PIP2, phosphatidylinositol bisphosphate; PLD, PLC, and PLA2, phospholipases D, C, and A2; AC, adenylyl cyclase; Gi, inhibitory G protein; PC, phosphatidylcholine; ER, endoplasmic reticulum; VDCC, voltage-dependent Ca channel; SOC, store-operated Ca channel.

CHAPTER 55 Vitamin D and Muscle

and diacylglycerol (DAG, a PKC activator) from membrane phosphoinositides [91]. The formation of DAG is biphasic, with the second phase independent of IP3 production and peaking at 5 min. 1α,25(OH)2D3 also stimulates the rapid hydrolysis of phosphatidylcholine (PC) in myoblasts by a phospholipase D (PLD)catalyzed mechanism. PLD activity generates choline and phosphatidic acid, which in turn can be converted to diacylglycerol by the action of a phosphohydrolase, accounting for the second peak of DAG observed in response to the hormone [92]. In addition, 1α,25(OH)2D3 activates phospholipase A2 (PLA2) and the subsequent release of arachidonic acid. The response is rapid (within 1 min) and dose-dependent (10−11–10−7 M) [93]. Rapid modulation of myoblast PLC, PLD, and PLA2 is 1α,25(OH)2D3 specific, as 25(OH)D3 and 24,25(OH)2D3 do not influence enzyme activities. The rapid activation of phospholipases by 1α,25(OH)2D3 involves the participation of guanine nucleotide binding (G) proteins. (AlF4)− and the stable analog GTPγS, which activate G proteins, mimic hormone stimulation of PLA2-mediated arachidonic acid (AA) release from myoblasts prelabeled with [3H]AA, whereas GDPβS and Bordetella pertussis toxin pretreatment abolish 1α,25(OH)2D3-dependent AA release [94]. By using similar experimental approaches, it has been shown that like PLA2, hormone modulation of PLC and PLD is mediated by a pertussis toxin–sensitive GTP-binding protein [95]. In agreement with the 1α,25(OH)2D3-induced generation of DAG via PLC and PLD, it has been reported that the hormone rapidly translocates protein kinase C into the cell membrane and increases its activity in chick muscle soleus muscle in vitro as well as in cultured embryonic muscle cells. In addition, the participation of protein kinase C in the fast 1α,25(OH)2D3 stimulation of 45Ca2+ influx through VDCC channels is supported by experimental evidence obtained with phorbol esters and DAG analogs, which mimic the action of the hormone, as well as PKC inhibitors, which reduce its effects [96,97]. Recent studies have revealed that PKCα is the only PKC isoform activated and translocated from cytosol to the membrane upon 1α,25(OH)2D3 stimulation of muscle cells [98] (Fig. 2, left side). Moreover, transfection of specific antiPKCα antibodies or intranuclear microinjection of antisense oligonucleotides against PKCα mRNA coupled to spectrofluorimetric analysis of changes in intracellular Ca2+ in Fura-2-loaded myoblasts/myotubes shows a marked reduction of hormone-dependent Ca2+ influx [98,99]. There are also data implying the participation of the adenylyl cyclase (AC)/cAMP pathway in the nongenomic mode of action of 1α,25(OH)2D3 in muscle.

891 Physiological concentrations of the hormone elicit very fast (within 30 sec) increases in AC, cAMP levels and PKA activity in both intact differentiated muscle and cultured myoblasts/myotubes. 1α,25(OH)2D3 stimulation of dihydropyridine-sensitive Ca2+ influx is abolished by specific inhibitors of AC and PKA and mimicked by forskolin and dibutyryl cAMP, involving the AC/cAMP/PKA pathway in the hormone nongenomic modulation of VDCC [100,101]. Studies with muscle cells and tissue on the effects of G protein modulators, e.g., fluoride, GTPγS, GDPβS, cholera, and Bordetella pertussis toxins, on 1α,25(OH)2D3-mediated Ca2+ uptake, as well as the observation of hormone-induced decrease of [35S]GTPγS binding to membranes and increased ADP ribosylation of the pertussis toxin–sensitive 41-kDa substrate, led to the proposal that negative regulation of an inhibitory protein coupled to AC is part of the mechanism by which 1α,25(OH)2D3 increases Ca2+ influx through the cAMP-dependent pathway [101,102] (Fig. 2, right side). Direct evidence on the involvement of G proteins in the fast activation of adenylyl cyclase by 1α,25(OH)2D3 has been obtained in experiments in which the effect of the hormone on AC, GTPase, and PKA activities as well as on the phosphorylation of Gαi was studied in membranes from chick skeletal muscle cells [103]. 1α,25(OH)2D3 stimulates AC activity in a dose (0.1–10 nM)- and time (1–5 min)-dependent fashion provided GTP is present in the assay. High-affinity GTPase, related to Gs, is unaffected by the hormone. In the absence of GTP or in the presence of a high concentration of Mn2+, a condition that provides information on adenylyl cyclase activity devoid of G-protein regulation, 1α,25(OH)2D3 effects on AC are abolished. PKA activity is increased in cells pretreated with the hormone. Moreover, immunoprecipitation of Gαi from [32P]-labeled myoblast membranes shows that 1α,25(OH)2D3 increases the phosphorylation of its α subunit. Therefore, these data altogether indicate that in muscle cells 1α,25(OH)2D3 activates adenylyl cyclase by a GTPdependent action implying, as suggested before, amelioration of Gi function by hormone-induced αi phosphorylation [103]. The rapidity with which 1α,25(OH)2D3 activates second messenger systems and Ca2+ channels in muscle cells and the hormone specificity at the physiological vitamin D3 metabolite concentrations suggest that a plasma membrane-bound receptor (mVDR; Fig. 2) may be responsible for the initiation of its effects. The existence of a putative novel cell-surface receptor for 1α,25(OH)2D3 that mediates the nongenomic actions of the hormone has been reported for other target cells. Thus, a 1α,25(OH)2D3 binding protein of 65–66 kDa, different from the nuclear VDR, detected in intestine,

892

RICARDO L. BOLAND

cartilage, kidney, and brain cells, has been functionally related to transcaltachia and rapid activation of PKC [104–106]. It has also been suggested that annexin II may be the membrane receptor that mediates 1α,25(OH)2D3induced rapid increases in cytosolic Ca2+ in rat osteoblast-like cells ROS 24/1, which have very few or undetectable VDR [107]. However, recent observations support the hypothesis that the classic nuclear VDR may be the receptor that mediates, at least in part, the nongenomic effects of 1α,25(OH)2D3 on store-operated Ca2+ (SOC; CCE) influx in muscle. The structural components of SOC channels are the TRP (transient receptor potential; Drosophila melanogaster) proteins, designated as the TRP-canonical (TRPC) subfamily of the larger TRP superfamily of gene products, which function as Ca2+permeable channels mainly regulated by store depletion when expressed in heterologous systems. At present, seven mammalian TRPC proteins are at least known

(TRPC1–TRPC7), which are homologs of the invertebrate counterparts [108]. The involvement of TRPC proteins in SOC entry induced by 1α,25(OH)2D3 in muscle cells has been investigated. Two fragments have been amplified from avian myoblasts by RT-PCR, exhibiting >85% sequence homology with human TRPC3. In agreement with these observations, Northern and Western blots employing TRPC3-probes and antiTRPC3 antibodies, respectively, confirmed endogenous expression of a TRPC3-like protein in muscle cells. As shown in Fig. 3A, transfection of myoblasts with antiTRPC3 antisense ODNs shows reduced CCE induced by 1α,25(OH)2D3. Anti-VDR antisense ODNs also inhibit hormone-dependent SOC Ca2+ influx (Fig. 3A), and coimmunoprecipitation of TRPC3 and VDR is observed (Fig. 3B), suggesting an association between both proteins and a functional role of the receptor in 1α,25(OH)2D3 activation of CCE. Accordingly, it has been shown that short treatment with 1α,25(OH)2D3

A Ca2+ Control

[Ca2+]i (50 nM)

1,25

anti-TRP anti-VDR

1 min 1

B

2

3

4

1

2

3

4

140 kDa 60 kDa

WB: anti-TRP

WB: anti-VDR

FIGURE 3 1α,25(OH)2D3-dependent muscle SOC influx is mediated by TRPC3 proteins and the VDR. (A) Effects of anti-TRPC3 and anti-VDR antisense ODNs on 1α,25(OH)2D3-induced SOC influx, fluorimetrically measured after hormone-elicited store depletion by the Ca2+ readdition protocol [54]. (B) Coimmunoprecipitation of TRPC3 and VDR. Immunoprecipitation was carried out with anti-TRPC3 antibody (right) or with anti-VDR antibody (left) followed by Western blotting with anti-TRPC3 antibody (left) or anti-VDR antibody (right). Lane 1, anti-TRPC3 antibody; lane 2, donkey anti-goat antibody; lane 3, without primary and secondary antibodies; lane 4, cell lysate.

893

CHAPTER 55 Vitamin D and Muscle

A

B 3

1,25

Ca2+ Ratio (340/390)

Ratio (340/390)

2,5

2,5

EGTA

2 1,5 1 0,5

2

EGTA Ca2+

1,5 1

1 min

0

1,25

1 min 0,5

FIGURE 4 Effect of an anti-INAD antibody on 1α,25(OH)2D3-dependent SOC influx in muscle cells. Cultured chick embryo skeletal muscle cells were permeabilized with saponin (50 µg/ml buffer) for 5 min at room temperature (as in [98]), in the presence of either normal rabbit Ig G (A) or an antibody against INAD from Caliphora vicinia (B). After washing, the cells were loaded with Fura-2 and the SOC influx dependent upon store depletion by 1α,25(OH)2D3 [10−9 M] was fluorimetrically measured by the Ca2+ readdition protocol [54].

induces translocation of the VDR from the nucleus to plasma membranes in chick myoblasts/myotubes. This reverse translocation is blocked by colchicine, genistein, or herbimycin, suggesting the involvement of microtubular transport and tyrosine kinase/s in the relocation of the receptor [42]. TRP channels have been shown to be modulated by association of macromolecules integrating signaling supramolecular complexes. The scaffold protein INAD clusters these macromolecules through its PDZ domains [109]. A functional role for a INAD-like protein in hormone activation of CCE is implied by the reduction of 1α,25(OH)2D3-induced Ca2+ influx upon transfection of muscle cells with an anti-INAD antibody (Fig. 4) or microinjection with anti-INAD antisense ODNs (not shown). In addition to TRPC3 and VDR, other components of the putative signaling complex may be calmodulin/CaMKII and Src, which have been shown to interact with TRP proteins (recent unpublished observations) and the VDR [78], respectively, in keeping with the fact that CaM antagonists and inhibitors of CaMKII as well as of tyrosine kinases block 1α,25(OH)2D3-induced SOC influx [54,56]. Figure 5 illustrates the proposed mechanism by which INAD-based signaling complexes with the intervention of the VDR participate in modulation of SOC influx by 1α,25(OH)2D3.

VIII. SUMMARY Muscle weakness and atrophy, and electrophysiologically demonstrable abnormalities in muscle contraction

and relaxation, are observed in vitamin D–deficiency states. These myopathies are independent of changes in blood mineral composition or PTH levels and respond only to vitamin D3 or its metabolites. Various lines of experimental evidence described in this chapter obtained with animal models and cultured embryonic muscle cells (myoblasts), which differentiate in vitro to functionally competent muscle fibers (myotubes), have demonstrated that the hormonal metabolite 1α,25(OH)2D3 is essential for normal homeostasis of intracellular calcium and growth in skeletal muscle and thereby plays an important role in contractility and myogenesis. Other lines of research have implied an action of 25(OH)D3 in muscle phosphate uptake and contractile protein synthesis. 1α,25(OH)2D3 regulates muscle intracellular Ca2+ through both genomic and nongenomic mechanisms. Congruent with the presence of the VDR in myoblasts and myotubes, the hormone induces transcriptionally the synthesis of various Ca2+ binding proteins, one of them identified as calbindin-D9K, and enzymes related to the synthesis of phosphatidylcholine, whose changes correlate with those in 1α,25(OH)2D3-dependent Ca2+ transport. Fast actions of 1α,25(OH)2D3 also exert rapid effects in skeletal muscle initiated in the cell surface by interacting with an as-yet-unidentified hormone plasma membrane receptor, followed by the stimulation of second messenger systems that transmit the signal to the cytoplasm. This mode of action involves the activation of Ca2+ entry through voltage-dependent Ca2+ channels by G-protein-mediated modulation of the adenylyl cyclase/cAMP/PKA and phospholipase

894

RICARDO L. BOLAND

P PLCβ

P

SERCA pump

P

Nucleus

FIGURE 5 Proposed mechanism of action involved in 1α,25(OH)2D3 regulation of SOC influx with participation of the VDR and INAD-based signaling complexes. PM, plasma membrane; PLCβ, phospholipase C β; IP3, inositol trisphosphate; DAG, diacylglycerol; IP3R, IP3 receptor; SERCA, sarcoplasmic–endoplasmic reticulum Ca ATPase; PMCA, plasma membrane Ca ATPase; SOC/CCE, store-operated Ca/capacitative Ca entry channel; VDCC, voltage-dependent Ca channel; CaM, calmodulin; INAD, scaffold protein, TRPC3, C3 isoform of TRP protein.

C/DAG + IP3/PKCα pathways. 1α,25(OH)2D3 also stimulates the release of Ca2+ from intracellular stores and the capacitative influx of the cation through storeoperated calcium (SOC; TRP) channels. Tyrosine kinases (TKs) and the VDR participate in hormone regulation of SOC channels. Accordingly, 1α,25(OH)2D3 induces rapid translocation of the VDR from the nucleus to the plasma membrane, a process that is blocked by TK inhibitors. Of mechanistic relevance, molecular and immunochemical studies and the application of oligonucleotide antisense technology, in conjunction with microspectrofluorimetric analysis, have provided evidence on the participation of signaling supramolecular structures integrated by TRP proteins, calmodulin, Src, VDR, and the scaffold protein INAD. This is an emerging novel concept within the field of nongenomic actions of vitamin D3 actions, which deserve further study. 1α,25(OH)2D3 stimulates proliferation and growth of muscle cells by promoting tyrosine phosphorylation of the MAP kinase cascade, mainly upstream at the level of Src and Raf-1, which requires PKCα, Ca2+, and, in part, the VDR itself. Clearly, significant advances in the characterization of the effects of 1α,25(OH)2D3 on muscle calcium metabolism and growth have been achieved in recent years, which may provide useful insights into the mode

of action of the hormone in skeletal muscle, where it appears to play an important role both in the regulation of cellular Ca2+ and in growth and development.

Acknowledgments The author is indebted to current and previous laboratory colleagues who have contributed during various phases of the investigations reported, in particular Ana R. De Boland, Teresita Bellido, Susana Morelli, Guillermo Vazquez, Maria Julia Marinissen, Claudia Buitrago, Graciela Santillàn, and Daniela Capiati. Work in our laboratory described in this chapter has been supported by the Consejo Nacional de Investigaciones Cientìficas (CONICET), Agencia Nacional de Promociòn Cientìfica y Tecnològica, and Fundaciòn Antorchas, Argentina.

References 1. DeLuca HF 1988 The vitamin D story: a collaborative effort of basic science and clinical medicine. FASEB J 2:224–236. 2. Reichel H, Koeffler H, Norman AW 1989 The role of the vitamin D endocrine system in health and disease. New Engl J Med 320:980–991.

CHAPTER 55 Vitamin D and Muscle

3. Minghetti P, Norman A 1988 1,25(OH)2-vitamin D3 receptors: Gene regulation and genetic circuitry. FASEB J 2:3043–3053. 4. Perlman KA, Kutnerei A, Prahl J, Smith C, Inaba M, Schnoes H, DeLuca HF 1990 24-Homologated 1,25-dihydroxyvitamin D3 compounds: Separation of calcium and differentiation activities. Biochemistry 29:190–196. 5. De Boland A, Nemere I 1992 Rapid actions of vitamin D compounds. J Cell Biochem 49:32–36. 6. Nemere I, Farach-Carson MC 1998 Membrane receptors for steroid hormones: A case for specific cell surface binding sites for vitamin D metabolites and estrogens. Biochim Biophys Res Commun 248:443–449. 7. Walters M 1992 Newly identified functions of the vitamin D endocrine system. Endocr Rev 13:719–764. 8. Walters M 1995 Newly identified functions of the vitamin D endocrine system: Update. Endocr Rev Monogr 4:47–56. 9. Boland R 1986 Role of vitamin D in skeletal muscle function. Endocr Rev 7:434– 448. 10. Smith R, Stern G 1967 Myophathy, osteomalacia and hyperparathyroidism. Brain 90:593–602. 11. Smith R, Stern G 1969 Muscular weakness in osteomalacia and hyperparathyroidism. J Neurol Sci 8:511–520. 12. Skaria J, Katiyar BC, Srivastava TP, Dube B 1975 Myopathy and neuropathy associated with osteomalacia. Acta Neurol Scandinav 51:37–58. 13. Schott GD, Wills MR 1976 Muscle weakness in osteomalacia. Lancet 1:626–629. 14. Floyd M, Ayyar DR, Barwick DD, Hudgson P, Weightman D 1974 Myopathy in chronic renal failure. Q J Med 43:509–523. 15. Henderson RG, Ledingham JGG, Oliver DO, Small DJ, Russel RGG, Smith R, Walton RJ, Preston C 1974 Effects of 1,25-dihydroxy-cholecalciferol on calcium absorption, muscle weakness and bone disease in chronic renal failure. Lancet 1:379–384. 16. Brickman AS, Coburn JW, Massry SG, Norman AW 1974 1,25-Dihydroxy-vitamin D3 in normal man and patients with renal failure. Ann Int Med 80:161–168. 17. Vicale CT 1949 Diagnostic features of muscular syndrome resulting from hyperparathyroidism osteomalacia owing to renal tubular acidosis and perhaps to related disorders of calcium metabolism. Trans Am Neurol Assoc 74:143–147. 18. Birge SJ 1978 Vitamin D, muscle and phosphate homeostasis. Mineral Electrolyte Metab 1:57–64. 19. Haussler MR, Baylink DJ, Hughes MR, Brumbraugh PF, Vergedal JE, Shen FH, Nielsen RL, Counts SJ, Bursak KM, McCain TA 1976 The assay of 1,25-dihydroxy-vitamin D3. Physiologic and pathological modulation of circulating hormone levels. Clin Endocrinol (Oxf) 5(Suppl):151S–165S. 20. Rodman JS, Baker T 1978 Changes in the kinetics of muscle contraction in vitamin D-depleted rats. Kidney Int 13:189–193. 21. Pleasure D, Wyszynski B, Summer A, Schotland D, Feldmann B, Hitz K, Goodman DB, Nugent N 1979 Skeletal muscle calcium metabolism and contractile force in vitamin Ddeficient chicks. J Clin Invest 64:1157–1167. 22. Weber A, Herz R, Reiss I 1964 Role of calcium in contraction and relaxation of muscle. Fed Proc 23:896–200. 23. Ebashi S, Endo M 1968 Calcium ion and muscle contraction. Prog Biophys Mol Biol 18:123–183. 24. Carafoli E, Crompton M 1978 The regulation of intracellular calcium. Curr Top Membr Transp 10:151–216. 25. Curry OB, Bastein JF, Francis MJA, Smith R 1974 Calcium uptake by sarcoplasmic reticulum of muscle from vitamin D– deficient rabbits. Nature 249:83–84.

895 26. Heimberg KW, Matthews C, Ritz E, Augustin J, Hasselbach W 1976 Active Ca transport of sarcoplasmic reticulum during experimental uremia: changes in kinetics and lipid composition. Eur J Biochem 61:207–213. 27. Matthews C, Heimberg KW, Ritz E, Agostini B, Fritzsche J, Hasselbach W 1977 Effect of 1,25-dihydroxy-cholecalciferol on impaired calcium transport by the sarcoplasmic reticulum in experimental uremia. Kidney Int 11:227–235. 28. Boland R, Matthews C, De Boland AR, Ritz E, Hasselbach W 1983 Reversal of decreased phosphorylation of sarcoplasmic reticulum calcium transport ATPase by 1,25-dihydroxycholecalciferol in experimental uremia. Calcif Tissue Int 35:195–201. 29. Boland R, De Boland AR, Ritz E, Hasselbach W 1983 Effect of 1,25-dihydroxy-cholecalciferol on sarcoplasmic reticulum calcium transport in strontium-fed chicks. Calcif Tissue Int 35:190–194. 30. De Boland AR, Gallego S, Boland R 1983 Effects of vitamin D3 on phosphate and calcium transport across and composition of skeletal muscle plasma cell membranes. Biochim Biophys Acta 733:264–273. 31. Giuliani D, Boland R 1984 Effects of vitamin D metabolites on calcium fluxes in intact chicken skeletal muscle and myoblasts cultured in vitro. Calcif Tissue Int 36:200–205. 32. Bauman VK, Valinietse MY, Babarykin DA 1984 Vitamin D3 and 1,25-dihydroxy-vitamin D3 stimulate the skeletal musclecalcium mobilization in rachitic chicks. Arch Biochem Biophys 231:211–216. 33. Birge SJ, Haddad JC 1975 25-Hydroxy-cholecalciferol stimulation of muscle metabolism. J Clin Invest 56:1100–1107. 34. Ströder J, Arensmeyer E 1965 Der Actomyosin-Gehalt der Skeletmuskulatur bei experimenteller Rachitis. Klin Wochenschr 43:1201–1202. 35. De Boland AR, Albornoz LE, Boland R 1983 The effect of cholecalciferol in vivo on proteins and lipids of skeletal muscle from rachitic chicks. Calcif Tissue Int 35:798–805. 36. Pointon JJ, Francis MJO, Smith R 1979 Effects of vitamin D deficieny on sarcoplasmic reticulum function and troponin c concentration of rabbit skeletal muscle. Clin Sci 57:257–263. 37. Zot HB, Potter JD 1984 The role of calcium in the regulation of the sekeletal muscle contraction-relaxation cycle. In: Sigel H (ed) Metal Ions in Biological Systems. Marcel Dekker, New York, pp. 381–402. 38. Boland RL, Norman AW, Ritz E, Hasselbach W 1985 Presence of a 1,25-dihxdroxy-vitamin D3 Receptor in chick skeletal muscle myoblasts. Biochem Biophys Res Commun 128:305–311. 39. Simpson RU, Thomas GA, Arnold AJ 1985 Identification of 1,25-dihydroxyvitamin D3 receptors and activities in muscle. J Biol Chem 260:8882–8891. 40. Costa EM, Blau HM, Feldman D 1986 1,25-Dihydroxyvitamin D3 receptors and hormonal responses in human skeletal muscle cells. Endocrinology 119:2214–2220. 41. Zanello SB, Collins ED, Marinissen MJ, Norman AW, Boland R 1997 Vitamin D receptor expression in chicken muscle tissue and cultured myoblasts. Horm Metab Res 29:231–236. 42. Capiati D, Benassati S, Boland R 2002 1,25(OH)2-Vitamin D3 induces translocation of the vitamin D receptor (VDR) to the plasma membrane in skeletal muscle cells. J Cell Biochem 86:128–135. 43. Buitrago C, Vazquez G, De Boland AR, Boland R 2000 Activation of Src kinase in skeletal muscle cells by 1,25(OH)2vitamin D3 correlates with tyrosine phosphorylation of the

896

44.

45. 46.

47.

48.

49. 50. 51. 52.

53.

54.

55.

56.

57. 58. 59.

RICARDO L. BOLAND

vitamin D receptor (VDR), and VDR–Src interaction. J Cell Biochem 79:274–281. Bodwell JE, Webster JC, Jewell CM, Cidlowski JA, Hu JM, Munck A 1998 Glucocorticoid receptor phosphorylation: Overview, function and cell cycle-dependence. J Steroid Biochem Mol Biol 65:91–99. Kuiper GG, Brinkmann AO 1994 Steroid hormone receptor phosphorylation: Is there a physiological role? Mol Cell Endocrinol 100:103–107. Hsieh JC, Jurutka PW, Galligan MA, Terpening CM, Haussler CA, Samuels DS, Shimizu Y, Shimizu N, Haussler MR 1991 Human Vitamin D receptor is selectively phosphorylated by protein kinase C on serine 51, a residue crucial to its trans-activation function. Proc Natl Acad Sci USA 88:9315–9319. Jurutka PW, Hsieh JC, MacDonald PN, Haussler CA, Haussler MR, Whitfield GK 1993 Phosphorylation of serine 208 in the human Vitamin D receptor. The predominant amino acid phosphorylated by casein kinase II, in vitro, and identification as a significant phosphorylation site in intact cells. J Biol Chem 268:6791–6799. De Boland AR, Boland R 1987 Rapid changes in skeletal muscle calcium uptake induced in vitro by 1,25-dihydroxyvitamin D3 are suppressed by calcium channel blockers. Endocrinology 120:1858–1864. Vazquez G, De Boland AR 1993 Stimulation of dihydropyridine-sensitive Ca2+ influx in cultured myoblasts by 1,25(OH)2-vitamin D3. Biochem Mol Biol Int 31:677–684. De Boland AR, Boland R 1994 Non-genomic signal transduction pathway of vitamin D in muscle. Cell Signal 6:717–724. McDonald TF, Pelzer S, Trautwein W, Pelzer DJ 1994 Regulation and modulation of calcium channels in cardiac, skeletal, and smooth muscle cells. Physiol Rev 74:365–507. Boland R, De Boland AR, Marinissen MJ, Santillan G, Vazquez G, Zanello S 1995 Avian muscle cells as targets for the secosteroid hormone 1,25-dihydroxy-vitamin D3. Mol Cell Endocrinol 114:1–8. Vazquez G, De Boland AR, Boland R 1997 Stimulation of Ca2+ release-activated Ca2+ channels as a potential mechanism involved in non-genomic 1,25(OH)2-vitamin D3induced Ca2+ entry in skeletal muscle cells. Biochem Biophys Res Commun 239:562–565. Vazquez G, De Boland AR, Boland R 1998 1α,25-dihydroxyvitamin D3-induced store-operated Ca2+ influx in skeletal muscle cells. Modulation by phospholipase C, protein kinase C, and tyrosine kinases. J Biol Chem 273:33954–33960. Vazquez G, Selles J, De Boland A, Boland R 1999 Rapid actions of calcitriol and its side chain analogues CB1093 and GS1500 on intracellular calcium levels in skeletal muscle cells: a comparative study. British J Pharmacol 126:1815–1823. Vazquez G, De Boland AR, Boland R 2000 Involvement of calmodulin in 1α,25-dihydroxyvitamin D3 stimulation of store-operated Ca2+ influx in skeletal muscle cells. J Biol Chem 275:16134–16138. Bellido T, Boland R 1987 Phosphate accumulation by muscle in vitro and the influence of vitamin D3 metabolites. Z Naturforsch 42c:237–244. Bellido T, Boland R 1988 Stimulation of myoblast membrane protein synthesis by 25-hydroxy-vitamin D3. Z Naturforsch 44c:807–812. Bellido T, Boland R 1991 Effects of 1,25-dihydroxy-vitamin D3 on phosphate accumulation by myoblasts. Horm Metab Res 23:113–116.

60. Abe E, Miyaura C, Sakagami H, Takeda M, Kormo K, Yamazaki T, Yoshiki S, Suda T 1981 Differentiation of mouse myeloid cells induced by 1alpha,25-dihydroxy-vitamin D3. Proc Natl Acad Sci USA 78:4990–4994. 61. Suda T 1989 The role of 1,25-dihydroxyvitamin D3 in myeloid cell differentiation. Proc Soc Exp Biol Med 191:214–220. 62. Fahrquahrson C, Whitehead C, Rennie J, Loverigde N 1993 In vivo effects of 1,25-dihydroxycholecalciferol on the proliferation and differentiation of avian chondrocytes. J Bone Miner Res 8:1081–1088. 63. Bikle D, Pillai S 1993 Vitamin D, calcium, and epidermal differentiation. Endocr Rev 14:3–19. 64. Simpson R, Hsu T, Begley D, Mitchell B, Alizadeh J 1987 Transcriptional regulation of the c-myc protooncogene by 1,25-dihydroxyvitamin D3 in HL-60 promyelocytic cells. J Biol Chem 262:4104–4108. 65. Tu-Yu A, Curtiss Morris R, Ives H 1993 Differential modulation of fos and jun gene expression by 1,25-dihydroxyvitamin D3. Biochem Biophys Res Commun 193:161–166. 66. Simpson R, Hsu T, Wendt M, Taylor J 1989 1,25-dihydroxyvitamin D3 regulation of c-myc protooncogene transcription. Possible involvement of protein kinase C. J Biol Chem 264:19710–19715. 67. Wakelam MJO 1985 The fusion of myoblasts. Biochem J 228:1–12. 68. O’Neill MC, Stockdale FE 1972 Kinetic analysis of myogenesis in vitro. J Cell Biol 52:52–65. 69. Capiati DA, Limbozzi F, Tellez-Inon MT, Boland R 1999 Evidence on the participation of protein kinase C α in the proliferation of cultured myoblasts. J Cell Biochem 74:292–300. 70. Capiati DA, Vazquez G, Tellez-Inon MT, Boland R 2000 Antisense oligonucleotides targeted against protein kinase C α inhibit proliferation of cultured avian myoblasts. Cell Prolif 33:307–315. 71. Drittanti L, De Boland AR, Boland R 1989 Modulation of DNA synthesis in cultured muscle cells by 1,25-dihydroxyvitamin D3. Biochim Biophys Acta 1014:112–119. 72. Drittanti L, De Boland AR, Boland R 1990 Stimulation of calmodulin synthesis in proliferating myoblasts by 1,25dihydroxy-vitamin D3. Mol Cell Endocrinol 74:143–153. 73. Capiati DA, Tellez-Inon MT, Boland R 1999 Participation of protein kinase C α in 1,25-dihydroxy-vitamin D3 regulation of chick myoblast proliferation and differentiation. Mol Cell Endocrinol 153:39–45. 74. Bellido T, Morelli S, Fernandez LM, Boland R 1993 Evidence for the participation of protein kinase C and 3′,5′-cyclic AMP-dependent protein kinase in the stimulation of muscle cell proliferation by 1,25-dihydroxy-vitamin D3. Mol Cell Endocrinol 90:231–238. 75. Neary JI 1997 MAPK cascades in cell growth and death. News Physiol Sci 12:286–293. 76. Morelli S, Buitrago C, Vazquez G, De Boland AR, Boland R 2000 Involvement of tyrosine kinase activity in 1,25(OH)2vitamin D3 signal transduction in skeletal muscle cells. J Biol Chem 275:3626–3628. 77. Buitrago C, Boland R, De Boland AR 2001 The tyrosine kinase c-Src is required for 1,25(OH)2-vitamin D3 signaling to the nucleus in muscle cells. Biochim Biophys Acta 1541:179–187. 78. Buitrago C, Vazquez G, De Boland AR, Boland R 2001 The vitamin D receptor mediates rapid changes in muscle protein tyrosine phosphorylation induced by 1,25(OH)2D3. Biochem Biophys Res Commun 289:1150–1156.

CHAPTER 55 Vitamin D and Muscle

79. Boland R, De Boland, Buitrago C, Morelli S, Santillan G, Vazquez G, Capiati D, Baldi C 2002 Non-genomic stimulation of tyrosine phosphorylation cascades by 1,25(OH)2D3 by VDR-dependent and -independent mechanisms in muscle cells. Steroids 67:477–482. 80. Morelli S, Buitrago C, Boland R, De Boland AR 2001 The stimulation of MAP kinase by 1,25(OH)2-vitamin D3 in skeletal muscle cells is mediated by protein kinase C and calcium. Mol Cell Endocrinol 173:41–52. 81. Buitrago C, Gonzalez Pardo V, De Boland AR, Boland R 2003 Activation of Raf-1 through Ras and protein kinase Cα mediates 1α,25(OH)2-vitamin D3 regulation of the mitogenactivated protein kinase pathway in muscle cells. J Biol Chem 278:2199–2205. 82. Buitrago C, Gonzalez Pardo V, De Boland AR 2002 Nongenomic action of 1α,25(OH)2-vitamin D3. Activation of muscle cell PLCγ through the tyrosine kinase c-Src and PtdIns 3-kinase. Eur J Biochem 269:2506–2515. 83. De Boland AR, Boland R 1985 Suppression of 1,25-dihydroxy-vitamin D3 calcium transport by protein synthesis inhibitors and changes in phospholipids in skeletal muscle. Biochim Biophys Acta 845:237–241. 84. Bellido T, Drittanti L, De Boland AR, Boland R 1987 The phospholipid and fatty acid composition of skeletal muscle cells during culture in the presence of Vitamin D metabolites. Biochim Biophys Acta 922:162–169. 85. Drittanti L, De Boland AR, Boland R 1987 Changes in muscle lipid metabolism induced in vitro by 1,25-dihydroxyvitamin D3. Biochim Biophys Acta 918:83–92. 86. Drittanti L, De Boland AR, Boland R 1988 Effects of 1,25dihydroxyvitamin D3 on phospholipid metabolism in chick myoblasts. Biochim Biophys Acta 962:1–7. 87. Drittanti L, Boland R, De Boland AR 1989 Induction of specific proteins in cultured skeletal muscle cells by 1,25-dihydroxyvitamin D3. Biochim Biophys Acta 1012: 16–23. 88. Drittanti L, Zanello S, Boland R 1994 Induction of a calbindin-D9K-like protein in avian muscle cells by 1,25dihydroxy-vitamin D3. Biochem Mol Biol 5:859–868. 89. Zanello S, Drittanti L, Norman A, Boland R 1994 Identification of a 9 kDa calcium-binding protein in the chick as calbindin D-9K. In: Norman A, Bouillon R, Thomasset M (eds) Vitamin D. A Pluripotent Steroid Hormone: Structural Studies, Molecular Endocrinology and Clinical Applications. Walter de Gruyter, pp. 418–419. 90. Zanello S, Boland R, Norman A 1995 cDNA sequence identity of a vitamin-dependent calcium-binding protein in the chick to calbindin D-9K. Endocrinology 136:2784–2787. 91. Morelli S, De Boland AR, Boland R 1993 Generation of inositol phosphates, diacylglycerol and calcium fluxes in myoblasts treated with 1,25-dihydroxyvitamin D3. Biochem J 289:675–679. 92. De Boland AR, Morelli S, Boland R 1994 1,25(OH)2vitamin D3 signal transduction in chick myoblasts involves phosphatidylcholine hydrolysis. J Biol Chem 269:8675–8679. 93. De Boland AR, Boland R 1993 1,25-Dihydroxyvitamin D3 induces arachidonate mobilization in embryonic chick myoblasts. Biochim Biophys Acta 1179:98–104. 94. De Boland AR, Morelli S, Boland R 1995 1,25(OH)2vitamin D3 stimulates phospholipase A2 activity via a guanine nucleotide-binding protein in chick myoblasts. Biochim Biophys Acta 1257:274–278.

897 95. Morelli S, Boland R, De Boland AR 1996 1,25(OH)2-vitamin D3 stimulation of phospholipases C and D in muscle cells involves extracellular calcium and a pertussis-sensitive G protein. Mol Cell Endocrinol 122:207–211. 96. Massheimer V, De Boland AR 1992 Modulation of 1,25dihydroxyvitamin D3–dependent Ca2+ uptake in skeletal muscle by protein kinase C. Biochem J 281:349–352. 97. Vazquez G, De Boland AR 1996 Involvement of protein kinase C in the modulation of 1α,25-dihydroxy-vitamin D3induced 45Ca2+ uptake in rat and chick cultured myoblasts. Biochim Biophys Acta 1310:157–162. 98. Capiati DA, Vazquez G, Tellez-Inon MT, Boland R 2001 Role of PKC in 1,25(OH)2-vitamin D3 regulation of intracellular calcium levels during development of skeletal muscle cells in culture. J Cell Biochem 77:200–212. 99. Capiati DA, Vazquez G, Boland R 2001 Protein kinase C modulates the Ca2+ influx phase of the Ca2+ response to 1α,25-dihydroxy-vitamin D3 in skeletal muscle cells. Horm Metab Res 33:201–206. 100. Fernandez LM, Massheimer V, De Boland AR 1990 Cyclic AMP-dependent membrane protein phosphorylation and calmodulin binding are involved in the rapid stimulation of muscle calcium uptake by 1,25-dihydroxyvitamin D3. Calcif Tissue Int 47:314–319. 101. Vazquez G, Boland R, De Boland AR 1995 Modulation by 1,25(OH)2-vitamin D3 of the adenylyl cyclase/cyclic AMP pathway in rat and chick myoblasts. Biochim Biophys Acta 1269:91–97. 102. Boland AR, Flawia M, Coso O, Boland R 1991 A guaninenucleotide protein mediates 1,25-dihydroxy-vitamin D3– dependent rapid stimulation of Ca2+ uptake in skeletal muscle. Biochim Biophys Acta 1094:238–242. 103. Vazquez G, De Boland AR, Boland R 1997 1α,25-(OH)2vitamin D3 stimulates the adenylyl cyclase pathway in muscle cells by a GTP-dependent mechanism which presumably involves phosphorylation of Gαi. Biochem Biophys Res Commun 234:125–128. 104. Nemere I, Dormanen MC, Hammond MW, Okamura WH, Norman AW 1994 Identification of a specific binding protein for 1α,25-dihydroxyvitamin D3 in basal-lateral membranes of chick intestinal epithelium and relationship to transcaltachia. J Biol Chem 269:23750–23576. 105. Nemere I, Schwartz S, Pedrozo H, Sylvia VL, Dean DD, Boyan BD 1998 Identification of a membrane receptor for 1,25-dihydroxyvitamin D3 which mediates rapid activation of protein kinase C. J Bone Min Res 13:1353–1359. 106. Jia Z, Nemere I 1999 Immunochemical studies on the putative plasmalemmal receptor for 1,25-dihydroxyvitamin D3. II. Chick kidney and brain. Steroids 64:541–550. 107. Baran DT, Quail JM, Ray R, Leszyk J, Honeyman T 2000 Annexin II is the membrane receptor that mediates the rapid actions of 1α,25-dihydroxyvitamin D3. J Cell Biochem 78:34–46. 108. Birnbaumer L, Zhu X, Jiang M, Boulay G, Peyton M, Vannier B, Brown D, Platano D, Sadeghi H, Stefani E, Birnbaumer M 1996 On the molecular basis and regulation of cellular capacitative calcium entry: roles for Trp proteins. Proc Natl Acad Sci USA 93:15195–15202. 109. Huber A, Sander P, Bahner M, Paulsen R 1998 The TRP Ca2+ channel assembled in a signaling complex by the PDZ domain protein INAD is phosphorylated through the interaction with protein kinase C (εPKC). FEBS Lett 425:317–322.

CHAPTER 56

Vitamin D and Cardiovascular Medicine DWIGHT A. TOWLER THOMAS L. CLEMENS

Department of Medicine, Division of Bone and Mineral Diseases, Department of Molecular Biology and Pharmacology, Washington University School of Medicine, St. Louis, MO 63110 Department of Cell Biology and Physiology, Department of Internal Medicine, University of Cincinnati School of Medicine, Cincinnati, OH

I. Introduction II. Clinical Evidence for Vitamin D Signaling in Cardiovascular Health III. Indirect Cardiovascular Actions of Vitamin D IV. Direct Actions of Vitamin D in the vasculature

V. Vascular Calcification and Calcitropic Hormones: Cardiovascular Toxicology of Vitamin D VI. Summary and Conclusions References

I. INTRODUCTION

overview of our current knowledge of vitamin D signaling in cardiovascular physiology and toxicology. It is meant to highlight the need for additional contributions by biologists, nutritionists, epidemiologists, and clinicians in order to develop a better understanding of vascular biology and vitamin D signaling—relevant to preserving cardiovascular health and preventing frailty in our aging population.

The role of vitamin D in skeletal physiology and calcium/phosphate homeostasis has been wellrecognized for over 50 years; rachitic musculoskeletal frailty arising from vitamin D deficiency and the clinical response to vitamin D repletion were defining characteristics of this novel fat-soluble agent [1,2]. In recent years, the important roles for vitamin D in epithelial cell cycle physiology and terminal differentiation have become apparent, most saliently in skin [3] and in colonic and prostatic epithelia [4,5]. Perhaps foreseeable from the actions of vitamin D on monocyte proliferation and development ex vivo, in vivo modulation of cell-mediated immunity by vitamin D has been demonstrated via clinical studies of psoriasis, animal models of autoimmune diabetes, and organ transplantation [6–8]. Clinically, concern of cardiovascular toxicity has dominated our thinking of vitamin D in vascular biology; indeed, several well-studied rat models utilize vitamin D toxicity in order to study the pharmacological inhibition of vascular calcification and calciphylaxis [9–13]. Only recently, however, has cardiovascular physiology been tied to vitamin D bioactivity. A woefully small number of key patient-oriented research studies [14,15] and preclinical investigations [16] has adequately addressed the role of vitamin D signaling in cardiovascular health—emphasizing the regulation of hypertension, vascular inflammation, and diabetes risk as immediately germane to atherosclerosis [17,18]. The goal of this chapter is to provide an VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

II. CLINICAL EVIDENCE FOR VITAMIN D SIGNALING IN CARDIOVASCULAR HEALTH A. Epidemiology As detailed elsewhere in this text (Chapters 2 and 3), outside of dietary supplementation, the major source of vitamin D arises from photoconversion of 7-dehydrocholesterol in skin in response to ultraviolet light. Given the tremendous variation in human cutaneous pigment content, several groups have noted the differential effects of diminished sunlight exposure on vitamin D levels in African Americans vs. Caucasians [19] (Chapter 47). Because of the increased risk for hypertension in the former population, Rostand studied the potential correlation of light exposure and vitamin D levels to racial and geographic variations in blood pressure [19]; he found that vitamin D deficiency is indeed a risk factor for hypertension. No intervention study has been performed to date that directly assesses Copyright © 2005, Elsevier, Inc. All rights reserved.

900 the effects of vitamin D nutritional supplementation on blood pressure in African Americans. However, Pfeifer and colleagues recently published a seminal clinical study of calcium with or without vitamin D supplementation on systolic blood pressure (SBP) and diastolic blood pressure (DBP) in elderly, vitamin D–deficient Caucasian women—a population at high risk for osteoporosis and concomitant cardiovascular disease [14]. While calcium (1200 mg) had little, if any, effect on blood pressure, two months of 800 IU of vitamin D with 1200 mg of calcium resulted in an average 9% (13 mm Hg) decrease in SBP (p = 0.02; n = 148). A similar, borderline trend (P = 0.10) was observed for DBP as well, and a modest decrease in basal heart rate was noted, suggesting globally reduced sympathetic tone (not directly assessed). Most importantly, 80% of subjects treated with vitamin D plus calcium experienced a 5 mm Hg decrement in SBP, compared with 40% of calcium treated subjects. The mechanism was unexamined, but may be related to suppressive effects of vitamin D on renin production [16] (vide infra, section III). This past year, Farhleitner and colleagues showed that peripheral arterial vascular disease (PAD) is associated with a high incidence of vitamin D deficiency [15]. They found that a remarkable 71% of individuals with PAD had 25-hydroxyvitamin D levels below 9 ng/ml. Of greater significance, they demonstrated that 40% had secondary hyperparathyroidism, and approximately 10% had frank hypocalcemia—biochemical indices that independently confirmed the high incidence of reduced vitamin D tone in patients with PAD [15]. The data are consistent with vitamin D deficiency arising secondary to the immobilization and isolation that can be associated with severe PAD. However, other studies have shown inverse relationships between 25-hydroxy vitamin D levels and the risk for myocardial infarction [20,21], suggesting a cardiovascular benefit to adequate vitamin D nutrition—or detriment with deficiency. To date, no adequately powered study evaluating the contribution of vitamin D supplementation on primary prevention of either coronary artery disease or PAD has been performed [15]. Of note, a key placebocontrolled arm of the Women’s Health Initiative (WHI) clinical study will assess contributions of vitamin D (400 IU daily) and calcium (1000 mg daily as calcium carbonate) supplementation to bone health and colorectal cancer outcomes in post-menopausal women. Since cardiovascular health endpoints modifiable by hormone replacement therapy are primary endpoints in other arms of this study [22], potential cardiovascular benefits of dietary vitamin D + calcium supplementation may also emerge from analysis of the WHI [23,24] (minimum of 7 years planned follow-up).

DWIGHT A. TOWLER AND THOMAS L. CLEMENS

B. Vitamin D Receptor Clinical Genetics and Cardiovascular Disease The vitamin D receptor (VDR) is a prototypic member of the large family of nuclear receptor transcription factors [25,26]. Like other nuclear receptors, this zinc finger transcription factor regulates gene transcription in heterodimeric VDR complexes via protein-protein and protein-DNA interactions directed by lipophilic ligands (e.g., calcitriol) and Ser/Thr phosphorylation [25,26]. Both genomic and nongenomic responses to vitamin D require the ligand binding domain and zinc fingers of the VDR [27]. A decade ago, a polymorphism in a BsmI cognate was identified in the 3′-UTR of the VDR gene. As compared to transcripts containing this BSM sequence (denoted “b”), transcripts lacking this BSM sequence (denoted “B”) result in higher levels of VDR mRNA accumulation, although this finding is controversial (see Chapter 68) [28,29]. Small studies have provided evidence that the B VDR allele is enriched in patients with calcific aortic stenosis [30]. Moreover, the B allele is associated with acute onset type I diabetes [31], and the VDR BB genotype is associated with a threefold increased risk for type II diabetes with coronary artery disease (CAD) [32]. In a larger study of 3,441 patients referred for angiography, once adjustments are made for traditional risk factors such as diabetes and hypertension, no independent association exists between the VDR BsmI genotype and cardiovascular risk or extent [33]. Thus, genetic contributions of VDR polymorphisms to CAD may be more closely related to modifying the risk for these well-known cardiovascular risk factors. Consistent with this notion, vitamin D supplementation has been shown to decrease the risk for type I diabetes in Finnish children [34] (see Chapter 99 for additional discussion of vitamin D and diabetes). The VDR BB allele is also associated with lower serum calcium concentrations and higher serum phosphate concentrations in patients with chronic renal failure [35]—a population at high risk for cardiovascular disease and calcific vasculopathy. This subset of our population—the hemodialysis patient— has severely perturbed calcium phosphate homeostasis. It is tempting to speculate that VDR genetic polymorphisms exert a greater impact on cardiovascular health in this metabolically stressed milieu, as reflected in PTH responses to calcitriol pharmacotherapy (bb genotype relatively resistant to calcitriol suppression of PTH) [36,37] and cardiovascular mortality risk [38]. The hypophosphatemia and hypocalcemia of severe rickets is an uncommon but reversible cause of congestive heart failure (CHF) [39–41]. Little data exist

CHAPTER 56 Vitamin D and Cardiovascular Medicine

on VDR genotype and CHF, particularly in adults. Recently in a small study (n = 75), Katagiri and colleagues examined relationships between the VDR FokI genotype in women with mild to moderate CHF [42]. While no connection to severity or prevalence of CHF was identified, the fractional excretion of calcium was found to be elevated twofold in CHF patients with the VDR FF allele [42]. Loss of spinal bone mineral density was also greater in VDR FF patients. Calciuresis is coupled with natriuresis; the VDR FF allele correlates positively with furosemide dose and elevated atrial natriuretic peptide concentrations—pharmacological and physiological factors that regulate natriuresis in the CHF patient [42]. While genotype correlation with furosemide dose administered suggests correlation with severity of cardiovascular compromise [42], no direct evidence exists. However, these data do highlight that CHF patients with the VDR FF genotype are at greater risk for negative calcium balance, bone loss, and perhaps fracture.

III. INDIRECT CARDIOVASCULAR ACTIONS OF VITAMIN D A. Vitamin D Regulation of the Renin–Angiotensin-Aldosterone (RAA) Axis As mentioned above, serum vitamin D levels are inversely related to blood pressure, and vitamin D supplementation in borderline deficient women reduces blood pressure. Recently, Li and colleagues (Chapter 54) provided important insights into the mechanism of this physiological response [16]. They demonstrated that activated VDR signaling suppresses the renin gene via ligand-dependent VDR interactions with a negative response element in the renin promoter. Activation of the RAA axis not only increases blood pressure via angiotensin II signaling, but concomitantly drives a myocardial fibrosis response via aldosterone [43]. Given the well-documented cardiovascular benefits of reducing blood pressure and RAA axis activity [43], reduced renin expression may explain, in part, the emerging and desirable cardiovascular actions of adequate vitamin D nutrition [14,19,44].

B. Calcium Phosphate Homeostasis and PTH Secretion As detailed elsewhere in this volume, calcitriol enhances gastrointestinal absorption of calcium and phosphate from dietary sources (Chapter 24). In addition,

901 via the activation of RANKL production from sustentacular bone marrow stromal cells, calcitriol enhances RANKL-dependent differentiation of the bone resorbing osteoclast that also releases calcium and phosphate from skeletal stores (Chapter 38). Via its suppression of PTH induction, vitamin D also enhances renal tubular resorption of phosphate, reducing phosphate loss in the urine [45]. Thus, a major consequence of vitamin D action is the long-term regulation (days to months) of the serum calcium-phosphate product. This has several important indirect cardiovascular consequences relevant to severe vitamin D deficiency and vitamin D toxicity. The phosphate depletion of severe rickets is an uncommon but reversible cause of congestive heart failure (CHF) [39–41]; this highlights the deleterious, indirect consequences of extreme vitamin D malnutrition via reduced myocardial phosphate stores (ATP, creatine phosphate) necessary for normal cardiac contractility. However, in addition to providing adequate mineral substrates for bone deposition at sites of skeletal ossification, an elevated calcium-phosphate product enhances mineral deposition at nonosseous sites [46–48]; a key feature of vitamin D toxicity arises from the dystrophic vascular calcification (vide infra, section V). Thus, at the extremes of vitamin D exposure, perturbations in serum calcium and phosphate indirectly contribute to cardiovascular pathophysiology. It has been recognized for almost 20 years that the PTH gene is a direct target of vitamin D action [49]. As detailed elsewhere in the text (Chapter 30), the production of PTH is under negative regulation by calcitriol [50], and the VDR bb allele has been associated with an increased incidence of primary hyperparathyroidism [51] and impaired PTH suppression by calcitriol [37]. Thus, a subset of vitamin D cardiovascular actions are thus potentially elicited via changes in circulating PTH. Vitamin D deficiency is known to result in secondary hyperparathyroidism, and chronic hyperparathyroidism is associated with hypertension, particularly salient in chronic renal insufficiency [52]. However, the vascular response in X-linked hypophosphatemic rickets—a hereditary disorder characterized by secondary hyperparathyroidism in the absence of vitamin D deficiency (Chapter 69)—provides additional evidence for a role for hyperparathyroidism in vascular tone [53]. Children afflicted with XLH who subsequently develop hypertension have marked secondary hyperparathyroidism and hypertension [53]. Of note, since the latter is often associated with nephrocalcinosis and elevated renin levels, hypertension may occur in response to renal injury [53]. Jorde and colleagues demonstrated that reduced calcium intake at any given level of vitamin D intake is associated with elevated systolic blood pressure, diastolic blood

902

DWIGHT A. TOWLER AND THOMAS L. CLEMENS

pressure, and circulating PTH [54,55]; however, calcium supplementation in patients with secondary hyperparathyroidism does not appear to reduce blood pressure even though PTH levels are reduced [56]. Thus, the relative contributions of vitamin D deficiency to hypertension and ventricular function via the RAA axis (Section III.A), PTH secretion, and direct actions described (Section IV.B) remain to be determined.

cardiovascular disease [65]. The nutritional and pharmacological anti-inflammatory actions of vitamin D— while predicted to reduce cardiovascular risk in part via these mechanisms—have not been adequately studied to either demonstrate or discount potential contributors to cardiovascular health.

C. Vitamin D Regulation of the Coagulation Cascade

A. Calcitriol D Actions in Vascular Smooth Muscle Cells (VSMCs)

Relatively little data are available concerning the actions of vitamin D on the coagulation cascade. In circulating human monocytes and the U937 monocytic cell line [57], calcitriol upregulates thrombomodulin (coactivator of protein C-mediated anticoagulation), and down-regulates the production of tissue factor, which is a key procoagulant that facilitates intrinsic pathway activation during injury and sepsis. Importantly, monocyte procoagulant responses to TNF-alpha were abrogated by calcitriol and synthetic analogs [58]. Consistent with this, calcitriol inhibits disseminated intravascular coagulation (DIC) in response to lipopolysaccharide, but not tissue factor, in a rat model of DIC [59]. Whether vitamin D signaling is of any clinical relevance in the setting of gram-negative sepsis or acute coronary thrombosis is as yet unexamined.

Immunoreactive vitamin D receptors are present in VSMCs [66–68], cardiomyocytes [69–71], and endothelial cells [72]. The direct actions of calcitriol on mesenchymal myofibroblasts and VSMCs are well described. Calcitriol exerts a profound suppressive effect on VSMC and myofibroblast proliferation in response to thrombin [73], but can increase DNA synthesis in serum-free conditions [73] and enhances VSMC migration [66]. Both proliferative and inhibitory activities suggest a potential role for vitamin D in modulating vascular injury responses and neointima formation, but this has not been systematically examined. However, evidence for a role of VSMC vitamin D signaling in calcific vasculopathy of vitamin D toxicity is strong [74]. Vitamin D intoxication induces medial degeneration in both humans and animal models [13,75]. Calcitriol also upregulates the production of alkaline phosphatase in cultured VSMCs [76,77]. This has tremendous functional importance, since alkaline phosphatase—in addition to being an early marker of bone-forming osteoblast differentiation—degrades inorganic pyrophosphate and dephosphorylates osteopontin (OPN) [78]–major physiological inhibitors of heterotopic calcium deposition [79,80]. This is likely secondary to the actions of vitamin D on vascular expression of PTHrP, an autocrine inhibitor of VSMC mineralization, proliferation, and OPN gene expression [76]. Mechanistically, VDR targets a nuclear matrix Ku antigen complex, enhancing stable association of the DNA-dependent protein kinase to the PTHrP promoter with loss of Ku subunit binding and resultant transcriptional suppression [81,82]. The physiologically important roles of adequate vitamin D nutrition in controlling blood pressure, diabetes risk, and inflammatory responses relevant to atherosclerosis are just emerging (Section III). While the consequences of toxic vitamin D exposure of VSMC are well established and being studied, much more research is required to determine the physiological/nutritional

D. Regulation of Inflammation and Cell-Mediated Immunity by Vitamin D Freedman and colleagues using the U937 myelomonocytic cell line highlighted an important role for calcitriol in limiting the proliferative expansion of hematopoietic progenitors [60,61]. Calcitriol enhances expression of p21, a negative regulator of G1 cell-cycle transit. This activity of calcitriol and its analogs has been proved clinically useful in the treatment of psoriasis, a memory T-cell–driven hyperproliferative disorder of skin and joints [62]. Calcitriol inhibits the production of IL-12 [63], an important macrophage-derived inflammatory Th1 cytokine that plays an important role in atherosclerotic progression [64]. The immunomodulatory actions of vitamin D are detailed thoroughly in Chapter 36. Of note, a compelling view of cardiovascular disease in the setting of diabetes, dyslipidemia, and heart transplantation emphasizes the role of vascular inflammation in disease progression [65]; circulating markers of inflammation robustly portend

IV. DIRECT ACTIONS OF VITAMIN D IN THE VASCULATURE

903

CHAPTER 56 Vitamin D and Cardiovascular Medicine

contribution of VSMC VDR signaling in cardiovascular health. Of note, the matrix cytokine OPN [83] is produced by vascular smooth muscle cells and is a direct target of phosphate and glucose induction [84]. OPN enhances the proliferation of VSMCs [85], regulates the size and extent of atheroma formation in animal models of atherosclerosis [86,87], and has the strong potential to promote intimal-medial thickening and calcification in the clinical settings of diabetes, atherosclerosis, and renal insufficiency [85,88–92]. As a Th1-type cytokine [93], OPN plays an important role in enhancing macrophage activation and migration [94] in atherosclerosis [86,87], in part by suppression of the anti-inflammatory cytokine IL-10 and upregulation of IL-12 [86]. However, OPN also can limit VSMC calcification [87] in a phosphorylation-dependent manner [78]. Of note, OPN is directly upregulated by VDR signaling in osteoblasts [95] and in monocyte/macrophages [96,97], but apparently not in vascular smooth muscle cells [76]. The net consequences and relative contributions of OPN cytokine induction vs. the direct, antiproliferative actions of vitamin D in leukocytes and VSMCs have yet to be fully explored.

B. Vitamin D Signaling in Cardiomyocytes As mentioned previously, profound vitamin D deficiency is a recognized reversible cause of congestive heart failure in children, and allelic variations in the VDR gene portend CHF severity. Simpson and colleagues were among the first to identify VDR expression in the cardiomyocytes, and detailed the evolution of cardiac hypertrophy in the vitamin D deficient rat [69,98,99]. A combination of in vivo and in vitro studies indicated that indirect actions (vide supra, section III A) likely contribute. Vitamin D participates in chamber-specific regulation of cardiac gene expression during development [100], and globally inhibits ventricular cardiomyocte myosin heavy chain gene expression (MyHC1, MyHC2, MyHC3) [101,102]. Consistent with this, VDR −/− mice and vitamin D deficient mice exhibit cardiac hypertrophy [16]. Moreover, calcitriol antagonizes cardiomyocyte maturation [101,102] and endothelin mediated hypertrophy in culture [103]. Genomic mechanisms appear to be important for suppression of MyHC3 in ventricular myocardium. A VDR:RXRα heterodimer binds to a negative vitamin D response element (nVDRE) about 780 bp upstream from the transcription initiation site of MyHC3 [100]. Irx4, a cardiac homeodomain protein, is recruited as a negative coregulator to the VDR:RXRα

complex via protein-protein interaction with RXRα [104]. Whether similar complexes assemble on other MyHC genes is unknown. Of note, similar multiprotein VDR repressor complexes have been identified in the nVDRE of the PTHrP promoter [81,82]. The heart is an endocrine organ [105]; atrial cardiomyocytes produce natriuretic peptides that control cardiac development and renal sodium excretion [106]. Gardner and colleagues [107,108] demonstrated that calcitriol suppresses transcription from the atrial natriuretic peptide (ANP) promoter via ligand-dependent assembly of an unusual repressor complex containing the coactivator GRIP1 [109]. Consistent with this, in the clinical setting of CHF, calcitriol and 25-hydroxyvitamin D are inversely correlated with pro-ANP levels (pro-ANP levels determined by rates of production and section) [110]. Potential benefits of manipulating VDR signaling in the setting of acute and chronic CHF have not been systematically examined, but would be predicted to have actions on hypertrophic responses, afterload reduction, cardiac output, and sodium/fluid retention. In addition to the genomic actions of VDR described above in cardiomyocytes, nongenomic actions of vitamin D also occur [111,112], including the rapid activation of protein kinase A, tyrosine phosphorylation, and potentially store-operant calcium mobilization [113]. Boland’s expert contributions are most complete (Chapter 55) [114,115]; while ligand specificity is relaxed, signaling still requires the presence of VDR [114,115]. Drawing upon parallels that exist for nongenomic signaling by other “nuclear” receptors [116], a subpopulation of VDR tethered to a caveolar signaling complex may be responsible for inducing this response [117]. The reader is referred elsewhere in this text for additional detail (Chapters 23 and 55).

C. Endothelial Responses to Vitamin D As compared with actions of vitamin D in other cell types, relatively little information exists concerning the mechanisms whereby VDR ligands regulate endothelial cell (EC) functions. Calcitriol, via its down-regulation of tissue factor (vide supra) in monocytes in response to inflammatory cytokines, inhibits coagulation cascade activation during inflammation. Although actions in umbilical vein ECs appear minimal [57], other ECs have not been examined. In the pulmonary vasculature, ICAM and ELAM expression and neutrophil margination is inhibited by calcitriol [118]. Indeed, evidence has emerged that ECs may express a 1-alpha hydroxylase activity that produces calcitriol as paracrine/intracrine

904 regulator of EC activation to inflammation (Chapter 79) [119]. Additionally, Canfield and colleagues were among the first to demonstrate that calcitriol inhibits bovine aortic EC proliferation in response to vascular endothelial growth factor (VEGF) stimulation [120]. Moreover, calcitriol induced apoptosis of EC undertaking capillary sprouting in culture [120]. Consistent with this observation, in models of tumor angiogenesis, calcitriol and its derivatives function as anti-angiogenic agents, suppressing EC proliferation and differentiation [121,122]. Such actions might be predicted to exert deleterious actions during collateral vessel development in ischemic heart disease. However, angiogenic responses may be tissue specific; calcitriol can also exert pro-angiogenic actions via parenchymal cell-type specific induction of VEGF [123,124]. For example, calcitriol promotes vascularization of the chondro-osseous junction during endochondral bone formation [125]. Calcitriol also upregulates VEGF production in hypertrophic chondrocytes and osteoblasts [126–129] and increases VEGF release in an aortic VSMC cell culture model [130]. Thus, a great deal more is to be learned on the net effects of vitamin D on vascular health as programmed by endothelial components of this complex organ system.

V. VASCULAR CALCIFICATION AND CALCITROPIC HORMONES: CARDIOVASCULAR TOXICOLOGY OF VITAMIN D A. Overview of Vascular Calcification Vascular calcification is a heterogeneous disease that can be usefully categorized into a minimum of three histopathological variants—atherosclerotic intimal calcification, valvular calcification, and medial artery calcification. Calcification of atherosclerotic plaques has been well-known for over a century, first recognized and denoted by Virchow [131]. Initial calcium deposition is probably dystrophic, with calcium phosphate deposition visualized in extracellular matrix fibrils of the intima, fibrinous aggregates, and necrotic cellular debris. However, with progression, active ossification processes resembling bone formation are rapidly recruited [132–134]. Calcification of cardiac valves—most notably the aortic and mitral valves— has been shown to be histologically similar to the process of endochondral and intramembranous ossification [135,136]. In one key study, Kaplan and colleagues identified that 13% of calcified valves have histological evidence of lamellar bone formation [136]. Medial calcification occurs in response to diabetes [137–146]

DWIGHT A. TOWLER AND THOMAS L. CLEMENS

and/or chronic renal insufficiency [89,147]. Unlike the calcification associated with initiation of the atherosclerotic plaque, medial artery calcification initiates via an active osteogenic process that mimics intramembranous bone formation. There is matrix vesicle-dependent accumulation of calcium phosphate [148,149] occurring in immediate juxtaposition to a subset of medial VSMCs called CVCs (calcifying vascular cells) by Demer and colleagues [132,133]. The dysmetabolic state of diabetes and hyperlipidemia upregulates a BMP2–Msx2 regulated osteogenic potential in the aorta [91,132,150]. Data from several labs that pleuripotent microvascular smooth muscle cells—pericytic myofibroblasts— represent the wide-spread vascular osteoprogenitors that can be diverted to the osteogenic lineage [151–154].

B. Regulation of Vascular Calcification by Vitamin D The mechanisms controlling the active and passive components of vascular calcification are poorly understood. Price and colleagues have extensively implemented a vitamin D intoxication/calciphylaxis model to study the pharmacological regulation of vascular calcification in the rat [155–158]. Their data strongly suggest that vascular calcium accumulation is inversely related to net calcium influx into bone [156]. Morii and colleagues identified an important role for paracrine PTHrP signaling in a culture model of vascular calcification [76]. They demonstrated that calcitriol promotes bovine VSMC alkaline phosphatase expression and calcification in part by suppression of endogenous VSMC PTHrP [76]. Moreover, calcitriol-induced alkaline phosphatase expression and VSMC calcification is inhibited in a dose-dependent fashion by PTHrP [76]. Thus, an additional contribution of vitamin D excess to arterial calcification may occur via the downregulation of a paracrine PTHrP signal. Of note, in the LDL receptor −/− mouse, intermittent administration of PTH suppresses vascular calcification while promoting orthotopic mineral deposition (Towler et al, unpublished), consistent with the in vitro results of Morii [76]. Given the results outlined above, it is tempting to speculate that the vasculature has a protective surveillance system for preventing medial calcification proceeding via VSMC PTH/PTHrP receptor signaling— recruited by either pulsatile PTH(1-34) administration (Towler, unpublished observations) or paracrine PTHrP production [76]. In the setting of vitamin D toxicosis [155–158] calcitriol may enhance vascular mineral deposition by suppression of protective PTH and PTHrP signals [76]—in addition to elevating the pro-calcific calcium phosphate product [46] (vide supra).

905

CHAPTER 56 Vitamin D and Cardiovascular Medicine

Demer and colleagues once again published a seminal study demonstrating an inverse relationship between calcitriol levels and coronary calcification as quantified by electron beam computed tomography [159] suggesting a protective effect of calcitriol signaling within the physiological range. This relationship held in both individuals at high risk for coronary disease (patients heterozygous for the LDL receptor deficiency of familial hypercholesterolemia) and in normal individuals [15,159]. Unfortunately, levels of 25-hydroxyvitamin D—the longer lived metabolite that accurately reflects vitamin D nutrition—were not assessed in this study [15,159]. Nevertheless, these data provide tantalizing evidence that arterial vascular disease—whether coronary [159] or peripheral [15]—may be beneficially modulated by vitamin D metabolites.

working model of vitamin D nutrition in cardiovascular health has yet to be established. Mechanism-based preclinical studies that dissect cell-type specific contributions of VDR signaling to cardiovascular biology are required—approachable in genetically manipulatable mouse models (Chapter 20). Such basic research investigations will greatly complement the ongoing clinical studies that explore the utility of VDR ligands in cardiovascular medicine. Additional contributions by biologists, endocrinologists, nutritionists, epidemiologists, and clinicians are sorely needed in order to develop a better understanding of vascular biology and vitamin D nutrition—immediately relevant to promoting the health of our aging population.

References VI. SUMMARY AND CONCLUSIONS

tiv

hy

es

Ri

sk

Ac

op

is

rtr

Ax ab

et

ne

Hy pe

ro

Pr

es

su

re

&

Di

ste

lar

do

cu

Al

as

in-

iov

ns

rd

te d

Ca

gio

loo

es

An

Re

du

ce

sB

ss

in-

re

en

pp Su

sR se

ea cr

Cardiovascular Health

De

Nutrition

t e uc as od at Pr sph sis e at Pho ino lc ph os line Ca Ph ka ial x s Al ard um se oc lci ea My Ca Incr on, m ti ru rP, ca Se TH lcifi s P Ca se C ea SM ular cr In s V asc e V ss re tes pp mo Su Pro

it y

Without a doubt, vitamin D exerts important physiological and pathophysiological actions in the vasculature. A biphasic cardiovascular “dose-response” curve is emerging that highlights the benefits of adequate vitamin D nutrition and the deleterious consequences of vitamin D excess or deficiency (Fig. 1). Both direct and indirect mechanisms contribute to the net vascular responses to vitamin D signaling. Whereas most animal studies have emphasized the role of excessive vitamin D in vascular toxicity, studies in humans have studied consequences of nutritional deficiency on cardiovascular health. Because of the biphasic response curve, the added complexity of both direct vs. indirect mechanisms (for example secondary to changes in calcium homeostasis), and the paucity of “bench-tobedside” research programs in the area, a detailed

Toxicity

Vitamin D Dosage

FIGURE 1

A biphasic vitamin D dose-response curve for cardiovascular health.

1. DeLuca HF 1998 The vitamin D story: a collaborative effort of basic science and clinical medicine. Faseb J 2(3):224–236. 2. Holick MF 2003 Vitamin D: A millenium perspective. J Cell Biochem 88(2):296–307. 3. Smith EL, Pincus SH, Donovan L, Holick MF 1988 A novel approach for the evaluation and treatment of psoriasis. Oral or topical use of 1,25-dihydroxyvitamin D3 can be a safe and effective therapy for psoriasis. J Am Acad Dermatol 19(3):516–528. 4. Fletcher RH, Fairfield KM 2002 Vitamins for chronic disease prevention in adults: clinical applications. Jama 287(23): 3127–3129. 5. Doherty TM, Tang W, Dascalos S, Watson KE, Demer LL, Shavelle RM, Detrano RC 1997 Ethnic origin and serum levels of 1alpha,25-dihydroxyvitamin D3 are independent predictors of coronary calcium mass measured by electronbeam computed tomography. Circulation 96(5):1477–1481. 6. DeLuca HF, Zierold C 1998 Mechanisms and functions of vitamin D. Nutr Rev 56(2 Pt 2):S4–10; discussion S54–S75. 7. Becker BN, Hullett DA, O’Herrin JK, Malin G, Sollinger HW, DeLuca H 2002 Vitamin D as immunomodulatory therapy for kidney transplantation. Transplantation 74(8): 1204–1206. 8. Adorini L, Penna G, Giarratana N, Uskokovic M 2003 Tolerogenic dendritic cells induced by vitamin D receptor ligands enhance regulatory T cells inhibiting allograft rejection and autoimmune diseases. J Cell Biochem 88(2): 227–233. 9. Rosenblum IY, Black HE, Ferrell JF 1977 The effects of various diphosphonates on a rat model of cardiac calciphylaxis. Calcif Tissue Res 23(2):151–159. 10. Block GA 2000 Prevalence and clinical consequences of elevated Ca × P product in hemodialysis patients. Clin Nephrol 54(4):318–324. 11. Cannon EP, Williams BJ 1990 Raised vascular calcium in an animal model: effects on aortic function. Cardiovasc Res 24(1):47–52. 12. Kingma JG Jr., Roy PE 1988 Ultrastructural study of hypervitaminosis D induced arterial calcification in Wistar rats. Artery 16(1):51–61. 13. Niederhoffer N, Bobryshev YV, Lartaud-Idjouadiene I, Giummelly P, Atkinson J 1997 Aortic calcification produced by vitamin D3 plus nicotine. J Vasc Res 34(5):386–398.

906 14. Pfeifer M, Begerow B, Minne HW, Nachtigall D, Hansen C 2001 Effects of a short-term vitamin D3 and calcium supplementation on blood pressure and parathyroid hormone levels in elderly women. J Clin Endocrinol Metab 86(4): 1633–1637. 15. Fahrleitner A, Dobnig H, Obernosterer A, Pilger E, Leb G, Weber K, Kudlacek S, Obermayer-Pietsch BM 2002 Vitamin D deficiency and secondary hyperparathyroidism are common complications in patients with peripheral arterial disease. J Gen Intern Med 17(9):663–669. 16. Li YC, Kong J, Wei M, Chen ZF, Liu SQ, Cao LP 2002 1,25Dihydroxyvitamin D3 is a negative endocrine regulator of the renin-angiotensin system. J Clin Invest 110(2):229–238. 17. Raisanen-Sokolowski AK, Pakkala IS, Samila SP, Binderup L, Hayry PJ, Pakkala ST 1997 A vitamin D analog, MC1288, inhibits adventitial inflammation and suppresses intimal lesions in rat aortic allografts. Transplantation 63(7):936–941. 18. Moreno PR, Purushothaman KR, Fuster V, O’Connor WN 2002 Intimomedial interface damage and adventitial inflammation is increased beneath disrupted atherosclerosis in the aorta: implications for plaque vulnerability. Circulation 105(21):2504–2511. 19. Rostand SG 1997 Ultraviolet light may contribute to geographic and racial blood pressure differences. Hypertension 30(2 Pt 1):150–156. 20. Scragg R, Holdaway I, Jackson R, Lim T 1992 Plasma 25-hydroxyvitamin D3 and its relation to physical activity and other heart disease risk factors in the general population. Ann Epidemiol 2(5):697–703. 21. Scragg R, Jackson R, Holdaway IM, Lim T, Beaglehole R 1990 Myocardial infarction is inversely associated with plasma 25-hydroxyvitamin D3 levels: a community-based study. Int J Epidemiol 19(3):559–563. 22. Wassertheil-Smoller S, Hendrix SL, Limacher M, Heiss G, Kooperberg C, Baird A, Kotchen T, Curb JD, Black H, Rossouw JE, and others 2003 Effect of estrogen plus progestin on stroke in postmenopausal women: the Women’s Health Initiative: a randomized trial. JAMA 289(20):2673–2684. 23. McGowan JA, Pottern L 2000 Commentary on the Women’s Health Initiative. Maturitas 34(2):109–112. 24. Design of the Women’s Health Initiative clinical trial and observational study. The Women’s Health Initiative Study Group 1998 Control Clin Trials 19(1):61–109. 25. MacDonald PN, Baudino TA, Tokumaru H, Dowd DR, Zhang C 2001 Vitamin D receptor and nuclear receptor coactivators: crucial interactions in vitamin D-mediated transcription. Steroids 66(3–5):171–176. 26. Pike JW, Yamamoto H, Shevde NK 2002 Vitamin D receptormediated gene regulation mechanisms and current concepts of vitamin D analog selectivity. Adv Ren Replace Ther 9(3):168–174. 27. Erben RG, Soegiarto DW, Weber K, Zeitz U, Lieberherr M, Gniadecki R, Moller G, Adamski J, Balling R 2002 Deletion of deoxyribonucleic acid binding domain of the vitamin D receptor abrogates genomic and nongenomic functions of vitamin D. Mol Endocrinol 16(7):1524–1537. 28. Morrison NA, Qi JC, Tokita A, Kelly PJ, Crofts L, Nguyen TV, Sambrook PN, Eisman JA 1994 Prediction of bone density from vitamin D receptor alleles. Nature 367(6460):284–287. 29. Riggs BL, Nguyen TV, Melton LJ, 3rd, Morrison NA, O’Fallon WM, Kelly PJ, Egan KS, Sambrook PN, Muhs JM, Eisman JA 1995 The contribution of vitamin D receptor gene alleles to the determination of bone mineral density in normal and osteoporotic women. J Bone Miner Res 10(6):991–996.

DWIGHT A. TOWLER AND THOMAS L. CLEMENS

30. Ortlepp JR, Hoffmann R, Ohme F, Lauscher J, Bleckmann F, Hanrath P 2001 The vitamin D receptor genotype predisposes to the development of calcific aortic valve stenosis. Heart 85(6):635–638. 31. Motohashi Y, Yamada S, Yanagawa T, Maruyama T, Suzuki R, Niino M, Fukazawa T, Kasuga A, Hirose H, Matsubara K, and others 2003 Vitamin D receptor gene polymorphism affects onset pattern of type 1 diabetes. J Clin Endocrinol Metab 88(7):3137–3140. 32. Ortlepp JR, Lauscher J, Hoffmann R, Hanrath P, Joost HG 2001 The vitamin D receptor gene variant is associated with the prevalence of type 2 diabetes mellitus and coronary artery disease. Diabet Med 18(10):842–845. 33. Ortlepp JR, von Korff A, Hanrath P, Zerres K, Hoffmann R 2003 Vitamin D receptor gene polymorphism BsmI is not associated with the prevalence and severity of CAD in a large-scale angiographic cohort of 3441 patients. Eur J Clin Invest 33(2):106–109. 34. Hypponen E, Laara E, Reunanen A, Jarvelin MR, Virtanen SM 2001 Intake of vitamin D and risk of type 1 diabetes: a birthcohort study. Lancet 358(9292):1500–1503. 35. Chudek J, Karkoszka H, Schmidt-Gayk H, Ritz E, Kokot F 2000 Plasma parathyroid hormone, phosphatemia and vitamin D receptor genotype: are they interrelated? J Nephrol 13(1):54–58. 36. Nagaba Y, Heishi M, Tazawa H, Tsukamoto Y, Kobayashi Y 1998 Vitamin D receptor gene polymorphisms affect secondary hyperparathyroidism in hemodialyzed patients. Am J Kidney Dis 32(3):464–469. 37. Marco MP, Martinez I, Betriu A, Craver L, Fibla MJ, Fernandez E 2001 Influence of Bsml vitamin D receptor gene polymorphism on the response to a single bolus of calcitrol in hemodialysis patients. Clin Nephrol 56(2):111–116. 38. Marco MP, Craver L, Betriu A, Fibla J, Fernandez E 2001 Influence of vitamin D receptor gene polymorphisms on mortality risk in hemodialysis patients. Am J Kidney Dis 38(5):965–974. 39. Abdullah M, Bigras JL, McCrindle BW, Mustafa A 1999 Dilated cardiomyopathy as a first sign of nutritional vitamin D deficiency rickets in infancy. Can J Cardiol 15(6): 699–701. 40. Gulati S, Bajpai A, Juneja R, Kabra M, Bagga A, Kalra V 2001 Hypocalcemic heart failure masquerading as dilated cardiomyopathy. Indian J Pediatr 68(3):287–290. 41. Uysal S, Kalayci AG, Baysal K 1999 Cardiac functions in children with vitamin D deficiency rickets. Pediatr Cardiol 20(4):283–286. 42. Nishio K, Mukae S, Aoki S, Itoh S, Konno N, Ozawa K, Satoh R, Katagiri T 2003 Congestive heart failure is associated with the rate of bone loss. J Intern Med 253(4):439–446. 43. Lavoie JL, Sigmund CD 2003 Minireview: overview of the Renin-Angiotensin system—an endocrine and paracrine system. Endocrinology 144(6):2179–2183. 44. Holick MF 2002 Sunlight and vitamin D: both good for cardiovascular health. J Gen Intern Med 17(9):733–735. 45. Silver J, Yalcindag C, Sela-Brown A, Kilav R, Naveh-Many T 1999 Regulation of the parathyroid hormone gene by vitamin D, calcium, and phosphate. Kidney Int Suppl 73:S2–S7. 46. Cozzolino M, Dusso AS, Slatopolsky E 2001 Role of calcium-phosphate product and bone-associated proteins on vascular calcification in renal failure. J Am Soc Nephrol 12(11):2511–2516. 47. Kistler A, Galli B, Horst R, Truitt GA, Uskokovic MR 1989 Effects of vitamin D derivatives on soft tissue calcification in

907

CHAPTER 56 Vitamin D and Cardiovascular Medicine

48.

49. 50.

51.

52. 53.

54. 55. 56. 57.

58.

59.

60.

61.

62. 63.

neonatal and calcium mobilization in adult rats. Arch Toxicol 63(5):394–400. Safadi FF, Thornton P, Magiera H, Hollis BW, Gentile M, Haddad JG, Liebhaber SA, Cooke NE 1999 Osteopathy and resistance to vitamin D toxicity in mice null for vitamin D binding protein. J Clin Invest 103(2):239–251. Russell J, Lettieri D, Sherwood LM 1986 Suppression by 1,25(OH)2D3 of transcription of the preproparathyroid hormone gene. Endocrinology 119(6):2864–2866. Russell J, Ashok S, Koszewski NJ 1999 Vitamin D receptor interactions with the rat parathyroid hormone gene: synergistic effects between two negative vitamin D response elements. J Bone Miner Res 14(11):1828–1837. Carling T, Kindmark A, Hellman P, Lundgren E, Ljunghall S, Rastad J, Akerstrom G, Melhus H 1995 Vitamin D receptor genotypes in primary hyperparathyroidism. Nat Med 1(12): 1309–1311. Rostand SG, Drueke TB 1999 Parathyroid hormone, vitamin D, and cardiovascular disease in chronic renal failure. Kidney Int 56(2):383–392. Alon US, Monzavi R, Lilien M, Rasoulpour M, Geffner ME, Yadin O. Hypertension in hypophosphatemic rickets-role of secondary hyperparathyroidism. Pediatr Nephrol 18(2): 155–158. Jorde R, Sundsfjord J, Haug E, Bonaa KH 2000 Relation between low calcium intake, parathyroid hormone, and blood pressure. Hypertension 35(5):1154–1159. Jorde R, Bonaa KH 2000 Calcium from dairy products, vitamin D intake, and blood pressure: the Tromso Study. Am J Clin Nutr 71(6):1530–1535. Saleh F, Jorde R, Sundsfjord J 2003 Effect of calcium supplementation on blood pressure in patients with secondary hyperparathyroidism. J Endocrinol Invest 26(1):35–41. Koyama T, Shibakura M, Ohsawa M, Kamiyama R, Hirosawa S 1998 Anticoagulant effects of 1alpha,25-dihydroxyvitamin D3 on human myelogenous leukemia cells and monocytes. Blood 92(1):160–167. Ohsawa M, Koyama T, Yamamoto K, Hirosawa S, Kamei S, Kamiyama R 2000 1alpha,25-dihydroxyvitamin D3 and its potent synthetic analogs downregulate tissue factor and upregulate thrombomodulin expression in monocytic cells, counteracting the effects of tumor necrosis factor and oxidized LDL. Circulation 102(23):2867–2872. Asakura H, Aoshima K, Suga Y, Yamazaki M, Morishita E, Saito M, Miyamoto K, Nakao S 2001 Beneficial effect of the active form of vitamin D3 against LPS-induced DIC but not against tissue-factor-induced DIC in rat models. Thromb Haemost 85(2):287–290. Liu M, Lee MH, Cohen M, Bommakanti M, Freedman LP 1996 Transcriptional activation of the Cdk inhibitor p21 by vitamin D3 leads to the induced differentiation of the myelomonocytic cell line U937. Genes Dev 10(2): 142–153. Rots NY, Iavarone A, Bromleigh V, Freedman LP 1999 Induced differentiation of U937 cells by 1,25-dihydroxyvitamin D3 involves cell cycle arrest in G1 that is preceded by a transient proliferative burst and an increase in cyclin expression. Blood 93(8):2721–2729. Scott LJ, Dunn CJ, Goa KL 2001 Calcipotriol ointment. A review of its use in the management of psoriasis. Am J Clin Dermatol 2(2):95–120. D’Ambrosio D, Cippitelli M, Cocciolo MG, Mazzeo D, Di Lucia P, Lang R, Sinigaglia F, Panina-Bordignon P 1998 Inhibition of IL-12 production by 1,25-dihydroxyvitamin D3.

64.

65. 66.

67.

68.

69. 70. 71. 72.

73. 74.

75. 76.

77.

78.

79.

80.

Involvement of NF-kappaB downregulation in transcriptional repression of the p40 gene. J Clin Invest 101(1):252–262. Uyemura K, Demer LL, Castle SC, Jullien D, Berliner JA, Gately MK, Warrier RR, Pham N, Fogelman AM, Modlin RL 1996 Cross-regulatory roles of interleukin (IL)-12 and IL-10 in atherosclerosis. J Clin Invest 97(9):2130–2138. Libby P 2003 Vascular biology of atherosclerosis: overview and state of the art. Am J Cardiol 91(3A):3A–6A. Rebsamen MC, Sun J, Norman AW, Liao JK 2002 1α,25dihydroxyvitamin D3 induces vascular smooth muscle cell migration via activation of phosphatidylinositol 3-kinase. Circ Res 91(1):17–24. Rajasree S, Umashankar PR, Lal AV, Sarma PS, Kartha CC 2002 1,25-dihydroxyvitamin D3 receptor is upregulated in aortic smooth muscle cells during hypervitaminosis D. Life Sci 70(15):1777–1788. Ishibashi K, Evans A, Shingu T, Bian K, Bukoski RD 1995 Differential expression and effect of 1,25-dihydroxyvitamin D3 on myosin in arterial tree of rats. Am J Physiol 269(2 Pt 1): C443–C450. Weishaar RE, Simpson RU 1987 Vitamin D3 and cardiovascular function in rats. J Clin Invest 79(6):1706–1712. Fraga C, Blanco M, Vigo E, Segura C, Garcia-Caballero T, Perez-Fernandez R 2002 Ontogenesis of the vitamin D receptor in rat heart. Histochem Cell Biol 117(6):547–550. O’Connell TD, Simpson RU 1996 Immunochemical identification of the 1,25-dihydroxyvitamin D3 receptor protein in human heart. Cell Biol Int 20(9):621–624. Merke J, Milde P, Lewicka S, Hugel U, Klaus G, Mangelsdorf DJ, Haussler MR, Rauterberg EW, Ritz E 1989 Identification and regulation of 1,25-dihydroxyvitamin D3 receptor activity and biosynthesis of 1,25-dihydroxyvitamin D3. Studies in cultured bovine aortic endothelial cells and human dermal capillaries. J Clin Invest 83(6):1903–1915. Mitsuhashi T, Morris RC Jr., Ives HE 1991 1,25-dihydroxyvitamin D3 modulates growth of vascular smooth muscle cells. J Clin Invest 87(6):1889–1895. Fleckenstein-Grun G, Thimm F, Frey M, Matyas S 1995 Progression and regression by verapamil of vitamin D3– induced calcific medial degeneration in coronary arteries of rats. J Cardiovasc Pharmacol 26(2):207–213. Dahl PR, Winkelmann RK, Connolly SM 1995 The vascular calcification-cutaneous necrosis syndrome. J Am Acad Dermatol 33(1):53–58. Jono S, Nishizawa Y, Shioi A, Morii H 1998 1,25Dihydroxyvitamin D3 increases in vitro vascular calcification by modulating secretion of endogenous parathyroid hormonerelated peptide. Circulation 98(13):1302–1306. Shioi A, Katagi M, Okuno Y, Mori K, Jono S, Koyama H, Nishizawa Y 2002 Induction of bone-type alkaline phosphatase in human vascular smooth muscle cells: roles of tumor necrosis factor-alpha and oncostatin M derived from macrophages. Circ Res 91(1):9–16. Jono S, Peinado C, Giachelli CM 2000 Phosphorylation of osteopontin is required for inhibition of vascular smooth muscle cell calcification. J Biol Chem 275(26): 20197–20203. Hessle L, Johnson KA, Anderson HC, Narisawa S, Sali A, Goding JW, Terkeltaub R, Millan JL 2002 Tissue-nonspecific alkaline phosphatase and plasma cell membrane glycoprotein-1 are central antagonistic regulators of bone mineralization. Proc Natl Acad Sci USA 99(14):9445–9449. Steitz SA, Speer MY, McKee MD, Liaw L, Almeida M, Yang H, Giachelli CM 2002 Osteopontin inhibits mineral deposition

908

81.

82.

83.

84. 85.

86.

87.

88.

89.

90. 91.

92.

93.

94.

95.

DWIGHT A. TOWLER AND THOMAS L. CLEMENS

and promotes regression of ectopic calcification. Am J Pathol 161(6):2035–2046. Okazaki T, Nishimori S, Ogata E, Fujita T 2003 Vitamin D– dependent recruitment of DNA-PK to the chromatinized negative vitamin D response element in the PTHrP gene is required for gene repression by vitamin D. Biochem Biophys Res Commun 304(4):632–637. Nishishita T, Okazaki T, Ishikawa T, Igarashi T, Hata K, Ogata E, Fujita T 1998 A negative vitamin D response DNA element in the human parathyroid hormone-related peptide gene binds to vitamin D receptor along with Ku antigen to mediate negative gene regulation by vitamin D. J Biol Chem 273(18):10901–10907. Denhardt DT, Noda M, O’Regan AW, Pavlin D, Berman JS. Osteopontin as a means to cope with environmental insults: regulation of inflammation, tissue remodeling, and cell survival. J Clin Invest 107(9):1055–1061. Beck GR, Jr., Zerler B, Moran E 2000 Phosphate is a specific signal for induction of osteopontin gene expression. Proc Natl Acad Sci USA 97(15):8352–8357. Isoda K, Nishikawa K, Kamezawa Y, Yoshida M, Kusuhara M, Moroi M, Tada N, Ohsuzu F 2002 Osteopontin plays an important role in the development of medial thickening and neointimal formation. Circ Res 91(1):77–82. Isoda K, Kamezawa Y, Ayaori M, Kusuhara M, Tada N, Ohsuzu F 2003 Osteopontin transgenic mice fed a highcholesterol diet develop early fatty-streak lesions. Circulation 107(5):679–681. Matsui Y, Rittling SR, Okamoto H, Inobe M, Jia N, Shimizu T, Akino M, Sugawara T, Morimoto J, Kimura C, and others 2003 Osteopontin deficiency attenuates atherosclerosis in female apolipoprotein e-deficient mice. Arterioscler Thromb Vasc Biol 23(6):1029–1034. Nitta K, Ishizuka T, Horita S, Hayashi T, Ajiro A, Uchida K, Honda K, Oba T, Kawashima A, Yumura W, and others 2001 Soluble osteopontin and vascular calcification in hemodialysis patients. Nephron 89(4):455–458. Moe SM, O’Neill KD, Duan D, Ahmed S, Chen NX, Leapman SB, Fineberg N, Kopecky K 2002 Medial artery calcification in ESRD patients is associated with deposition of bone matrix proteins. Kidney Int 61(2):638–647. Chen NX, Moe SM 2003 Arterial calcification in diabetes. Curr Diab Rep 3(1):28–32. Towler DA, Bidder M, Latifi T, Coleman T, Semenkovich CF 1998 Diet-induced diabetes activates an osteogenic gene regulatory program in the aortas of low density lipoprotein receptor-deficient mice. J Biol Chem 273(46):30427–30434. Bidder M, Shao JS, Charlton-Kachigian N, Loewy AP, Semenkovich CF, Towler DA 2002 Osteopontin transcription in aortic vascular smooth muscle cells is controlled by glucoseregulated upstream stimulatory factor and activator protein-1 activities. J Biol Chem 277(46):44485–44496. Ashkar S, Weber GF, Panoutsakopoulou V, Sanchirico ME, Jansson M, Zawaideh S, Rittling SR, Denhardt DT, Glimcher MJ, Cantor H 2001 Eta-1 (osteopontin): an early component of type-1 (cell-mediated) immunity. Science 287(5454):860–864. Weber GF, Zawaideh S, Hikita S, Kumar VA, Cantor H, Ashkar S 2002 Phosphorylation–dependent interaction of osteopontin with its receptors regulates macrophage migration and activation. J Leukoc Biol 72(4):752–761. Noda M, Vogel RL, Craig AM, Prahl J, DeLuca HF, Denhardt DT 1990 Identification of a DNA sequence responsible for binding of the 1,25-dihydroxyvitamin D3 receptor

96.

97.

98. 99.

100.

101.

102. 103.

104.

105. 106. 107.

108. 109.

110.

111.

and 1,25-dihydroxyvitamin D3 enhancement of mouse secreted phosphoprotein 1 (SPP-1 or osteopontin) gene expression. Proc Natl Acad Sci USA 87(24):9995–9999. Andersson G, Johansson EK 1996 Adhesion of human myelomonocytic (HL-60) cells induced by 1,25-dihydroxyvitamin D3 and phorbol myristate acetate is dependent on osteopontin synthesis and the alpha v beta 3 integrin. Connect Tissue Res 35(1–4):163–171. Miyazaki Y, Setoguchi M, Yoshida S, Higuchi Y, Akizuki S, Yamamoto S 1990 The mouse osteopontin gene. Expression in monocytic lineages and complete nucleotide sequence. J Biol Chem 265(24):14432–14438. Weishaar RE, Simpson RU 1987 Involvement of vitamin D3 with cardiovascular function. II. Direct and indirect effects. Am J Physiol 253(6 Pt 1):E675–83. Weishaar RE, Kim SN, Saunders DE, Simpson RU 1990 Involvement of vitamin D3 with cardiovascular function. III. Effects on physical and morphological properties. Am J Physiol 258(1 Pt 1):E134–E142. Wang GF, Nikovits W, Jr., Schleinitz M, Stockdale FE 1998 A positive GATA element and a negative vitamin D receptorlike element control atrial chamber-specific expression of a slow myosin heavy-chain gene during cardiac morphogenesis. Mol Cell Biol 18(10):6023–6034. O’Connell TD, Weishaar RE, Simpson RU 1994 Regulation of myosin isozyme expression by vitamin D3 deficiency and 1,25-dihydroxyvitamin D3 in the rat heart. Endocrinology 134(2):899–905. O’Connell TD, Giacherio DA, Jarvis AK, Simpson RU 1995 Inhibition of cardiac myocyte maturation by 1,25-dihydroxyvitamin D3. Endocrinology 136(2):482–488. Wu J, Garami M, Cheng T, Gardner DG 1996 1,25(OH)2 vitamin D3, and retinoic acid antagonize endothelin–stimulated hypertrophy of neonatal rat cardiac myocytes. J Clin Invest 97(7):1577–1588. Wang GF, Nikovits W, Jr., Bao ZZ, Stockdale FE 2001 Irx4 forms an inhibitory complex with the vitamin D and retinoic X receptors to regulate cardiac chamber-specific slow MyHC3 expression. J Biol Chem 276(31):28835–28841. Needleman P, Greenwald JE 1986 Atriopeptin: a cardiac hormone intimately involved in fluid, electrolyte, and bloodpressure homeostasis. N Engl J Med 314(13):828–834. Cameron VA, Ellmers LJ 2003 Minireview: Natriuretic Peptides during Development of the Fetal Heart and Circulation. Endocrinology 144(6):2191–2194. Wu J, Garami M, Cao L, Li Q, Gardner DG 1995 1,25(OH)2D3 suppresses expression and secretion of atrial natriuretic peptide from cardiac myocytes. Am J Physiol 268(6 Pt 1): E1108–E1113. Li Q, Gardner DG 1994 Negative regulation of the human atrial natriuretic peptide gene by 1,25-dihydroxyvitamin D3. J Biol Chem 269(7):4934–4939. Chen S, Cui J, Nakamura K, Ribeiro RC, West BL, Gardner DG 2000 Coactivator-vitamin D receptor interactions mediate inhibition of the atrial natriuretic peptide promoter. J Biol Chem 275(20):15039–15048. Zittermann A, Schleithoff SS, Tenderich G, Berthold HK, Korfer R, Stehle P 2003 Low vitamin D status: a contributing factor in the pathogenesis of congestive heart failure? J Am Coll Cardiol 41(1):105–112. Selles J, Boland R 1991 Evidence on the participation of the 3′,5′-cyclic AMP pathway in the nongenomic action of 1,25dihydroxy-vitamin D3 in cardiac muscle. Mol Cell Endocrinol 82(2–3):229–235.

909

CHAPTER 56 Vitamin D and Cardiovascular Medicine

112. Santillan GE, Boland RL 1998 Studies suggesting the participation of protein kinase A in 1,25(OH)2-vitamin D3–dependent protein phosphorylation in cardiac muscle. J Mol Cell Cardiol 30(2):225–233. 113. Vazquez G, de Boland AR, Boland RL 2000 Involvement of calmodulin in 1alpha,25-dihydroxyvitamin D3 stimulation of store-operated Ca2+ influx in skeletal muscle cells. J Biol Chem 275(21):16134–16138. 114. Boland R, De Boland AR, Buitrago C, Morelli S, Santillan G, Vazquez G, Capiati D, Baldi C 2002 Nongenomic stimulation of tyrosine phosphorylation cascades by 1,25(OH)2D3 by VDR–dependent and –independent mechanisms in muscle cells. Steroids 67(6):477–482. 115. Capiati D, Benassati S, Boland RL 2002 1,25(OH)2-vitamin D3 induces translocation of the vitamin D receptor (VDR) to the plasma membrane in skeletal muscle cells. J Cell Biochem 86(1):128–35. 116. Chambliss KL, Yuhanna IS, Mineo C, Liu P, German Z, Sherman TS, Mendelsohn ME, Anderson RG, Shaul PW 2000 Estrogen receptor alpha and endothelial nitric oxide synthase are organized into a functional signaling module in caveolae. Circ Res 87(11):E44–E52. 117. Norman AW, Olivera CJ, Barreto Silva FR, Bishop JE 2002 A specific binding protein/receptor for 1alpha,25-dihydroxyvitamin D3 is present in an intestinal caveolae membrane fraction. Biochem Biophys Res Commun 298(3):414–419. 118. Chen SF 1995 1alpha,25-Dihydroxyvitamin D3 decreased ICAM-1 and ELAM-1 expressions on pulmonary microvascular endothelial cells and neutrophil motivation. J Steroid Biochem Mol Biol 52(1):67–70. 119. Zehnder D, Bland R, Chana RS, Wheeler DC, Howie AJ, Williams MC, Stewart PM, Hewison M 2002 Synthesis of 1,25-dihydroxyvitamin D3 by human endothelial cells is regulated by inflammatory cytokines: a novel autocrine determinant of vascular cell adhesion. J Am Soc Nephrol 13(3):621–629. 120. Mantell DJ, Owens PE, Bundred NJ, Mawer EB, Canfield AE 2000 1 alpha,25-dihydroxyvitamin D3 inhibits angiogenesis in vitro and in vivo. Circ Res 87(3):214–220. 121. Bernardi RJ, Johnson CS, Modzelewski RA, Trump DL 2002 Antiproliferative effects of 1alpha,25-dihydroxyvitamin D3 and vitamin D analogs on tumor–derived endothelial cells. Endocrinology 143(7):2508–2514. 122. Iseki K, Tatsuta M, Uehara H, Iishi H, Yano H, Sakai N, Ishiguro S 1999 Inhibition of angiogenesis as a mechanism for inhibition by 1alpha-hydroxyvitamin D3 and 1,25-dihydroxyvitamin D3 of colon carcinogenesis induced by azoxymethane in Wistar rats. Int J Cancer 81(5):730–733. 123. Ferrara N 2002 Role of vascular endothelial growth factor in physiologic and pathologic angiogenesis: therapeutic implications. Semin Oncol 29(6 Suppl 16):10–14. 124. Ferrara N, Davis-Smyth T 1997 The biology of vascular endothelial growth factor. Endocr Rev 18(1):4–25. 125. Lin R, Amizuka N, Sasaki T, Aarts MM, Ozawa H, Goltzman D, Henderson JE, White JH 2002 1α,25-dihydroxyvitamin D3 promotes vascularization of the chondro-osseous junction by stimulating expression of vascular endothelial growth factor and matrix metalloproteinase 9. J Bone Miner Res 17(9): 1604–1612. 126. Martinez P, Esbrit P, Rodrigo A, Alvarez-Arroyo MV, Martinez ME 2002 Age-related changes in parathyroid hormone-related protein and vascular endothelial growth factor in human osteoblastic cells. Osteoporos Int 13(11):874–881. 127. Wang DS, Yamazaki K, Nohtomi K, Shizume K, Ohsumi K, Shibuya M, Demura H, Sato K 1996 Increase of vascular

128.

129.

130.

131. 132. 133. 134. 135.

136. 137.

138. 139.

140. 141. 142.

143.

endothelial growth factor mRNA expression by 1,25-dihydroxyvitamin D3 in human osteoblast-like cells. J Bone Miner Res 11(4):472–479. Schlaeppi JM, Gutzwiller S, Finkenzeller G, Fournier B 1997 1,25-Dihydroxyvitamin D3 induces the expression of vascular endothelial growth factor in osteoblastic cells. Endocr Res 23(3):213–229. Wang DS, Miura M, Demura H, Sato K 1997 Anabolic effects of 1,25-dihydroxyvitamin D3 on osteoblasts are enhanced by vascular endothelial growth factor produced by osteoblasts and by growth factors produced by endothelial cells. Endocrinology 138(7):2953–2962. Yamamoto T, Kozawa O, Tanabe K, Akamatsu S, Matsuno H, Dohi S, Hirose H, Uematsu T 2002 1,25-dihydroxyvitamin D3 stimulates vascular endothelial growth factor release in aortic smooth muscle cells: role of p38 mitogen-activated protein kinase. Arch Biochem Biophys 398(1):1–6. Virchow R 1989 Cellular pathology. As based upon physiological and pathological histology. Lecture XVI–Atheromatous affection of arteries. 1858. Nutr Rev 47(1):23–25. Bostrom K, Watson KE, Horn S, Wortham C, Herman IM, Demer LL 1993 Bone morphogenetic protein expression in human atherosclerotic lesions. J Clin Invest 91(4):1800–1809. Bostrom K, Demer LL 2000 Regulatory mechanisms in vascular calcification. Crit Rev Eukaryot Gene Expr 10(2): 151–158. Fitzpatrick LA, Turner RT, Ritman ER 2003 Endochondral bone formation in the heart: a possible mechanism of coronary calcification. Endocrinology 144(6):2214–2219. Srivatsa SS, Harrity PJ, Maercklein PB, Kleppe L, Veinot J, Edwards WD, Johnson CM, Fitzpatrick LA 1997 Increased cellular expression of matrix proteins that regulate mineralization is associated with calcification of native human and porcine xenograft bioprosthetic heart valves. J Clin Invest 99(5):996–1009. Mohler ER, 3rd, Gannon F, Reynolds C, Zimmerman R, Keane MG, Kaplan FS 2001 Bone formation and inflammation in cardiac valves. Circulation 103(11):1522–1528. Mozes G, Keresztury G, Kadar A, Magyar J, Sipos B, Dzsinich S, Gloviczki P 1998 Atherosclerosis in amputated legs of patients with and without diabetes mellitus. Int Angiol 17(4):282–286. Chantelau E, Lee KM, Jungblut R 1997 Distal arterial occlusive disease in diabetes is related to medial arterial calcification. Exp Clin Endocrinol Diabetes 105(Suppl 2):11-13. Lehto S, Niskanen L, Suhonen M, Ronnemaa T, Laakso M 1996 Medial artery calcification. A neglected harbinger of cardiovascular complications in non-insulin-dependent diabetes mellitus. Arterioscler Thromb Vasc Biol 16(8):978–983. Chantelau E, Lee KM, Jungblut R 1995 Association of below-knee atherosclerosis to medial arterial calcification in diabetes mellitus. Diabetes Res Clin Pract 29(3):169–172. Niskanen L, Siitonen O, Suhonen M, Uusitupa MI 1994 Medial artery calcification predicts cardiovascular mortality in patients with NIDDM. Diabetes Care 17(11):1252–1256. Young MJ, Adams JE, Anderson GF, Boulton AJ, Cavanagh PR 1993 Medial arterial calcification in the feet of diabetic patients and matched non-diabetic control subjects. Diabetologia 36(7):615–621. Niskanen LK, Suhonen M, Siitonen O, Lehtinen JM, Uusitupa MI 1990 Aortic and lower limb artery calcification in type 2 (non insulin–dependent) diabetic patients and non diabetic control subjects. A five year follow-up study. Atherosclerosis 84(1):61–71.

910 144. Everhart JE, Pettitt DJ, Knowler WC, Rose FA, Bennett PH 1988 Medial arterial calcification and its association with mortality and complications of diabetes. Diabetologia 31(1):16–23. 145. Nelson RG, Gohdes DM, Everhart JE, Hartner JA, Zwemer FL, Pettitt DJ, Knowler WC 1988 Lower-extremity amputations in NIDDM. 12-yr follow-up study in Pima Indians. Diabetes Care 11(1):8–16. 146. Lithner F, Hietala SO, Steen L 1984 Skeletal lesions and arterial calcifications of the feet in diabetics. Acta Med Scand Suppl 687:47–54. 147. Shioi A, Taniwaki H, Jono S, Okuno Y, Koyama H, Mori K, Nishizawa Y 2001 Monckeberg’s medial sclerosis and inorganic phosphate in uremia. Am J Kidney Dis 38(4 Suppl 1): S47–S49. 148. Tanimura A, McGregor DH, Anderson HC 1986 Calcification in atherosclerosis. I. Human studies. J Exp Pathol 2(4): 261–273. 149. Tanimura A, McGregor DH, Anderson HC 1986 Matrix vesicles in atherosclerotic calcification. Proc Soc Exp Biol Med 172(2):173–177. 150. Tyson KL, Reynolds JL, McNair R, Zhang Q, Weissberg PL, Shanahan CM 2003 Osteo/chondrocytic transcription factors and their target genes exhibit distinct patterns of expression in human arterial calcification. Arterioscler Thromb Vasc Biol 23(3):489–494. 151. Doherty MJ, Ashton BA, Walsh S, Beresford JN, Grant ME, Canfield AE 1998 Vascular pericytes express osteogenic potential in vitro and in vivo. J Bone Miner Res 13(5): 828–838.

DWIGHT A. TOWLER AND THOMAS L. CLEMENS

152. Diaz-Flores L, Gutierrez R, Lopez-Alonso A, Gonzalez R, Varela H 1992 Pericytes as a supplementary source of osteoblasts in periosteal osteogenesis. Clin Orthop 275: 280–286. 153. Brighton CT, Lorich DG, Kupcha R, Reilly TM, Jones AR, Woodbury RA, 2nd 1992 The pericyte as a possible osteoblast progenitor cell. Clin Orthop 275:287–299. 154. Zuk PA, Zhu M, Mizuno H, Huang J, Futrell JW, Katz AJ, Benhaim P, Lorenz HP, Hedrick MH 2001 Multilineage cells from human adipose tissue: implications for cell–based therapies. Tissue Eng 7(2):211–228. 155. Price PA, June HH, Buckley JR, Williamson MK 2001 Osteoprotegerin inhibits artery calcification induced by warfarin and by vitamin D. Arterioscler Thromb Vasc Biol 21(10):1610–1616. 156. Price PA, Faus SA, Williamson MK 2001 Bisphosphonates alendronate and ibandronate inhibit artery calcification at doses comparable to those that inhibit bone resorption. Arterioscler Thromb Vasc Biol 21(5):817–824. 157. Price PA, Buckley JR, Williamson MK 2001 The amino bisphosphonate ibandronate prevents vitamin D toxicity and inhibits vitamin D–induced calcification of arteries, cartilage, lungs, and kidneys in rats. J Nutr 131(11):2910–2915. 158. Price PA, June HH, Buckley JR, Williamson MK 2002 SB 242784, a selective inhibitor of the osteoclastic V-H+ATPase, inhibits arterial calcification in the rat. Circ Res 91(6):547–552. 159. Watson KE, Abrolat ML, Malone LL, Hoeg JM, Doherty T, Detrano R, Demer LL 1997 Active serum vitamin D levels are inversely correlated with coronary calcification. Circulation 96(6):1755–1760.

CHAPTER 57

Approach to the Patient with Metabolic Bone Disease MICHAEL P. WHYTE

Center for Metabolic Bone Disease and Molecular Research, Shriners Hospitals for Children, and Division of Bone and Mineral Diseases, Washington University School of Medicine at Barnes-Jewish Hospital; St. Louis, Missouri

I. Introduction II. Diagnostic Evaluation III. Treatment

IV. Summary References

I. INTRODUCTION

the skeleton. Significantly, a variety of potent hormones and drugs that regulate mineral homeostasis and alter bone remodeling are now available to clinicians [3]. This pharmaceutical armamentarium must be used knowledgeably for efficacy, yet safety. This chapter emphasizes a number of considerations for the approach to the patient with metabolic bone disease, particularly those with disturbances in vitamin D homeostasis. Subsequent chapters in this section of the book discuss in detail the radiology of rickets and osteomalacia (Chapter 60), bone histomorphometry (Chapter 59), measurement of the vitamin D metabolites (Chapter 58), and the pharmacology and therapeutic use of vitamin D preparations (Chapter 61).

Metabolic bone disease traditionally encompasses a considerable number and variety of conditions [1–3]. In fact, the list is rapidly growing as the molecular bases of inherited skeletal syndromes and dysplasias are elucidated using DNA technology [4,5]. Although these disorders are often rare, several are epidemic in various regions of the world (e.g., postmenopausal osteoporosis, vitamin D deficiency rickets). Some can be life-threatening (e.g., severe forms of osteopetrosis and osteogenesis imperfecta); others can be incidental findings (e.g., Paget bone disease and fibrous dysplasia). Patients reflect all ages. Cumulatively, the number of individuals with clinically important metabolic bone disease is significant [1–3]. Diagnosis and treatment of metabolic bone disease can be both intriguing and satisfying, but there are numerous challenges. We now have at our disposal a variety of techniques to image the skeleton and to measure bone mass, assays for many of the factors that condition mineral and skeletal homeostasis as well as biochemical markers of bone turnover, and qualitative and quantitative histopathological methods to directly examine osseous tissue [6]. Furthermore, molecular tests for genetic disorders of the skeleton are becoming increasingly available in commercial and research laboratories [4]. Initial and follow-up patient evaluation benefit greatly from skilled use of these tools. Nevertheless, clinical acumen is more important than ever. Judicious selection from among these advances in technology and circumspect interpretation of the information they provide comes from experience with patients. In fact, successful treatment of metabolic bone disease often requires multidisciplinary medical approaches that may need to be especially broad-based when there is deformity or other structural problems of VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

II. DIAGNOSTIC EVALUATION Patients with metabolic bone disease are often challenging to physicians because many nutritional, environmental, genetic, pharmacological, and toxic factors can disturb mineral metabolism and impact the skeleton [2,3]. This book is testimony to the number and diversity of internal and external perturbations that can impair the biosynthesis, bioactivation, and actions of vitamin D. Patient age is yet another challenge for several reasons. In infants, children, and adolescents, complications of metabolic bone disease can be especially severe and complex because bone growth and modeling are occurring in addition to skeletal remodeling. All three physiological processes can be disturbed in pediatric patients, leading to impairment of growth and alterations in the shape of bones. The outcome is novel physical and radiological findings in children compared to adults. Patient age conditions the pathogenesis and clinical manifestations of these disorders and also provides Copyright © 2005, Elsevier, Inc. All rights reserved.

914 a guide to the etiology. However, diagnoses can be missed or delayed if the diagnostician is unaware that the reference ranges for some of the biochemical parameters of mineral homeostasis (e.g., serum inorganic phosphate, alkaline phosphatase) and all of the markers of skeletal turnover are different for infants and children compared to adults [2,3]. Elderly patients are also uniquely challenging, because they are especially likely to have metabolic bone disease with multifactorial etiology and pathogenesis [1–3,7]. Broad-based medical knowledge is necessary to fully understand the relationships between mineral and skeletal homeostasis and the consequences of specific or complex disturbances [8]. Clinicians who encounter patients with metabolic bone disease routinely need some of the skills of the endocrinologist, nutritionist, nephrologist, geneticist, and often the pediatrician or gerontologist. Familiarity with skeletal radiology and pathology is required. Furthermore, if there is significant bony deformity, the expertise of orthopedics, rheumatology, and rehabilitation medicine will be helpful. Metabolic bone disease is remarkable for the many subspecialties that can contribute to the comprehensive evaluation and effective care of patients. Diagnosis of metabolic bone disease must begin with the acquisition of all of the important information extractable from the medical history and obtainable from a thorough physical examination. This foundation should be supplemented with only the helpful and cost-effective choices from the ever-expanding menu of biochemical tests, as well as the available radiological and histological studies. Most situations will require some biochemical and radiological investigation [2,3]. For the majority of these patients, however, histological assessment before therapy or during follow-up is not necessary. But, when it is, nondecalcified bone processing, staining, and interpretation following tetracycline labeling are often crucial. The importance of the medical history for evaluation of metabolic bone disease cannot be overemphasized. Foremost in diagnosing these disorders is this orderly accumulation of pertinent information directly from patients [9]. The revelations help to guide the physical examination and subsequent laboratory studies and to disclose potentially important medical records. Examples are endless. For deficiency of vitamin D, historical details alone often explain how the patient has come to require supplementation. Prolonged breast feeding, lack of sunlight exposure together with failure to consume foods fortified with vitamin D, use of certain anticonvulsants, or a family history suggestive of a similar heritable disorder exemplify critical information that will be obtained by talking with the patient.

MICHAEL P. WHYTE

For the physician skilled in medical history acquisition concerning metabolic bone disease, it is not unusual for the physical examination and laboratory testing to uphold a diagnosis. Physical examination of the patient with metabolic disease may show findings for or against the diagnosis suspected from the medical history, but the information is also important for revealing deformity or other structural problems of the skeleton requiring attention. Radiological investigation of metabolic bone disease may appropriately be minimal or extensive, but it should be directed by the complete medical history and physical examination. Not uncommonly, the diagnosis is established from characteristic radiographic findings (e.g., Paget bone disease, fibrous dysplasia) [10–12]. If not, x-ray studies often provide an important basis for differential diagnosis (e.g., osteopenia, osteosclerosis, rickets, etc.) or support the diagnostic impression that must instead be confirmed by additional tests (e.g., pseudohypoparathyroidism, mastocytosis, hyperparathyroidism). However, radiological studies are also useful because they may help to assess the severity and evolution of the disorder. In addition, they can reveal skeletal complications not detected by physical examination. Some changes (e.g., physeal widening in rickets, osteolytic lesions in Paget bone disease, etc.) can then be followed to precisely assess responses to medical treatment. Biochemical testing is crucial to characterize any disturbance of mineral homeostasis. Furthermore, biochemical testing provides quantitative information to help guide the intensity of medical treatment and to monitor the patient’s response. Proper selection of laboratory studies at the time of diagnosis is important to completely establish baseline data. Most metabolic bone diseases will be treated medically. If laboratory investigation is incomplete when pharmacological or nutritional intervention begins, an opportunity for diagnosis or for evaluating therapy may be lost. In fact, freezing away some patient serum and an aliquot of urine before medication is prescribed is sometimes worthwhile in case retrospective diagnosis may become necessary. Furthermore, this can help provide especially precise pre- and posttreatment comparisons. Additionally, in select circumstances where a therapeutic challenge is given, preserving pretreatment specimens can be a tactic for saving money. Expensive and low-yield testing can await results of a brief trial of therapy (e.g., suspected environmental vitamin D deficiency rickets). Among the biochemical tests for metabolic bone disease, there is now considerable redundancy (e.g., markers of skeletal turnover) [2,3,6]. Furthermore, some of these assays are useful primarily

915

CHAPTER 57 Approach to the Patient with Metabolic Bone Disease

for research purposes where groups of patients are studied. The utility of a single measurement of some of the markers of bone turnover for an individual patient can be limited because of technical or physiological variability, leading to just modest correlation with pathological processes. Histological study of the skeleton is essential in relatively few clinical circumstances, although it often has considerable research importance. Here too, however, an opportunity for diagnosis and/or baseline assessment may be lost once pharmacologic therapy begins [13].

A. Medical History The complete and accurate medical history and thorough physical examination is the cornerstone for diagnosis, effective and safe treatment, and sound clinical investigation of metabolic bone disease [9].∗ A detailed medical history helps assure that the many adverse external or heritable factors that can affect the patient with metabolic bone disease will be uncovered (Table I). A questionnaire completed by the patient may serve as a beginning, but is hardly a satisfactory substitute. Only by talking with his/her patient will the physician sense how knowledgeable this individual might be, and therefore be able to judge the value of any historical information. Only by talking with the patient will the character of the major signs and symptoms be revealed. Subsequently, the medical history should be reported as a narrative description of the clinical problem. As discussed below, all of the principal elements of the medical history are potentially important for patients with metabolic bone disease. 1. CHIEF COMPLAINT

The chief complaint (or CC) may readily lead to the diagnosis; for example, pain from hip fracture due to osteoporosis or increasing head size due to Paget bone disease. Often, however, the patient’s major concern is more subtle. The above conditions may manifest with

* The more resources we have, and the more complex they are, the greater are the demands upon our clinical skill. These resources are calls upon judgment and not substitutes for it. Do not, therefore, scorn clinical examination; learn it sufficiently to get from it all it holds, and gain in it the confidence it merits. Sir F.M.R. Walshe (1881–1973) Canadian Medical Association Journal 67:395, 1952.

TABLE I Some Potentially Adverse Influences on Mineral and Skeletal Homeostasis Genetic Ethnic background Heritable disorders Lifestyle Inactivity (immobilization) Smoking Alcohol abuse Nutritional Excessive tea drinking (fluorosis) High protein intake Low dietary calcium intake Milk intolerance Vegetarian diet Drugs Anticonvulsants Bisphosphonate excess Chemotherapy Cyclosporin A Diuretics (producing calciuria) Fluoride Glucocorticoids Gonadotropin-releasing hormone (GnRH) agonists or antagonists Heparin Lithium Protease inhibitors Thyroid replacement therapy Vitamin A or D

Medical disorders Acromegaly Anorexia nervosa/Bulimia Celiac disease Cushing syndrome Cystic fibrosis Early menopause Fanconi syndrome Fibrous dysplasia (polyostotic) Gastrointestinal disease Glycogen storage disease Hemochromatosis Hemolytic anemia Hepatitis C infection Hepatobiliary disease Homocystinuria Hyperparathyroidism Hypogonadism Lymphoproliferative disease McCune-Albright syndrome Mastocytosis Multiple myeloma Osteogenesis imperfecta Pancreatic insufficiency Postmenopausal osteoporosis Prolactinoma Renal failure (transplantation) Rheumatologic disorders Secondary amenorrhea Thyrotoxicosis Turner syndrome Type I diabetes

chronic back discomfort or progressive leg deformity, respectively. Such less dramatic difficulties must be recognized not only because they may be pointing to a diagnosis, but because they are objectives of treatment and can help monitor the efficacy of treatment. For example, when vitamin D deficiency causing weakness from myopathy or bone pain from osteomalacia is effectively treated, these difficulties should resolve. Furthermore, no matter how mild or severe the CC, this is the worry for which the patient with metabolic bone disease seeks help.

916

MICHAEL P. WHYTE

TABLE II Metabolic Hypocalcemia (see Table IV) Muscle Asthenia Potbelly with lumbar lordosis Proximal myopathy Waddling gait Denial Caries Delayed eruption Enamel defects

Vitamin D Deficiency: Age-Dependent Signs and Symptoms Skeletal Bone tenderness Cranial sutures widened Craniotabes (skull flattening or asymmetry) Dystocia Flared wrists and ankles Fracture Frontal bossing Harrison’s groove Hypotonia Kyphosis

With disturbances in vitamin D homeostasis, there may be one or more major complaints that can be metabolic or skeletal in origin (Table II). This aspect of the medical history is sometimes particularly challenging, because one problem can be emphasized from among this myriad collection of important symptoms or signs, or because complaints may seem vague or excessive. 2. HISTORY OF PRESENT ILLNESS

Most metabolic bone diseases (including many caused by disturbances in vitamin D homeostasis) are chronic conditions. The detailed history of present illness (or HPI) may therefore be lengthy. Nevertheless, the time invested is crucial because this effort provides the most important historical information. Attention to the details also demonstrates to the patient that the physician understands and cares about his/her illness. Hence, this effort helps secure the patient’s confidence needed for effective treatment. From the HPI, the physician should gather an understanding of the temporal evolution of the disorder— likely essential for accurate and complete diagnosis and effective treatment. Clues predating the symptoms of vitamin D deficiency would be necessary to fully uncover the pathogenesis and etiology. Appreciation of past therapeutic attempts may reveal factors masking a diagnosis. Indeed, the outcome of previous therapy (successful or unsuccessful) may guide diagnosis and future treatment. I find it worthwhile at the outset (for the sake of time and effort, organization, and completeness) to alert the patient that information will be especially valuable if obtained in historical sequence. Patients often require some guidance as this effort begins, but most will then help present their medical history in this

Lax ligaments Limb deformity Listlessness Low back pain Pneumonia Rachitic rosary Rib deformity → respiratory compromise Short stature Sternal indention or protrusion “String-of-pearls” deformity in hands

especially useful way. Tactfully diverting them from excessive or extraneous details usually becomes easy because they now appreciate that the physician is truly concerned and wishes to know, understand, and help them to organize this important information. Nutritional factors could be considered in the HPI. Both mineral metabolism and vitamin D homeostasis are influenced by diet in countries that fortify foods with vitamin D. Strict vegetarians (vegans), who avoid all foods derived from animals, will not benefit from the safety net of vitamin D supplementation of milk products in the United States. Avoidance of dairy foods may also lead to a calcium-deficient diet (Table III). Ovolactovegetarians, however, do drink milk. When the HPI is reported completely, not only will critical clues to disease etiology and pathogenesis emerge, but the physician may also gain important insight for treatment, and here may be a glimpse of the patient’s prognosis. Has this individual been compliant with regard to recommendations for diagnostic studies or medical care; if not, will pharmacologic therapy be safe? Have the manifestations of the disorder been lifelong (suggesting a congenital or genetic problem), or has there been the development of recent symptomatology that should prompt very different diagnostic considerations and interventions? Have complications been substantial, and likely to remain so if medical therapy cannot effect a cure? Only by inquiring about the patient’s illness in detail is the physician likely to learn that previous medical records, radiographs, etc., are available to help avoid expensive duplication of effort, and perhaps to help address important diagnostic and prognostic issues. Finally, by carefully documenting this aspect of a metabolic bone disease, the physician is providing the basis for sound clinical research.

917

CHAPTER 57 Approach to the Patient with Metabolic Bone Disease

TABLE III

Vegetarians

Vegans (strict vegetarians who do not consume dairy products or eggs) Ovolactovegetarians (vegetarians who will eat dairy products and eggs) Buddhists (some sects) Zen (Chinese) Ch′an (Japanese) Muslims (some Islamic sects) Ethnic groups of African, Hispanic, American Indian, Jewish, or Oriental descent (who may have lactose intolerance) Yoga Seventh Day Adventists International Society for Krishna Consciousness Zen Macrobiotic Movement

3. PAST MEDICAL HISTORY

A considerable number and variety of perturbations can cause metabolic bone disease, or can influence the outcome of medical treatment, by impacting on mineral and skeletal homeostasis (Table I). The detailed past medical history (or PMH) will help to disclose these disturbances or factors that can influence or obscure a diagnosis. In the PMH, previous diagnostic studies may be revealed that could prove useful for assessing the metabolic bone disease. Radiographs (e.g., chest x-rays, intravenous pyelogram) taken years ago for other reasons could show whether osteopenia, osteosclerosis, or rachitic change is old or new. Routine biochemical testing reports might help to date the onset of vitamin D deficiency by documenting when serum alkaline phosphatase activity began to rise. Drug history must be carefully assessed here and, if relevant, perhaps detailed instead in the HPI. Many pharmaceuticals can adversely affect the skeleton and disturb mineral or vitamin D homeostasis (e.g., glucocorticoids, certain diuretics, anticonvulsants, etc.) (Chapter 61). A seemingly incidental problem like acne, if overlooked, might fail to reveal tetracycline or retinoid exposure. In fact, these concerns also apply to some over-the-counter preparations (e.g., vitamin A, calcium supplements, antacids). In large amounts, each of these nonprescription items can have important effects on mineral or skeletal homeostasis and cause illness. Nevertheless, they may inadvertently be dismissed during elicitation of a drug history (“Pardon me, doctor. You asked me what medications I was taking.”). Many patients will not consider vitamins, mineral supplements, or antacids to be in this category (Table I).

Furthermore, some medications confound interpretation of biochemical studies aimed at metabolic bone disease because they alter mineral homeostasis (e.g., diuretics that increase or decrease urine calcium levels), elevate serum alkaline phosphatase activity from the liver, etc. Fortunately, in the United States, pharmacological doses of vitamin D require a prescription. Nevertheless, vitamin D intoxication has occurred from excesses inadvertently added to dairy milk (and other errors) that would have gone unrecognized were it not for detailed medical histories. The PMH may also help to predict the nature and frequency of recurrent illness and therefore complications from pharmaceutical treatments. Depending on the patient’s prior medical problems and compliance, will use of vitamin D2 (with its prolonged biological effects) be safer or more risky than a shorter-acting, but more potent, vitamin D metabolite? Such information might help to prevent the abrupt development of hypocalcemia if therapy will be suddenly compromised, or prevent prolonged hypercalcemia if dosing might become excessive. 4. SOCIAL HISTORY

Patient compliance for medical treatment, especially for chronic disorders, is often imperfect. Recognition that an individual will be uncooperative or has health insurance problems or financial difficulties that could affect his/her ability or willingness to undergo diagnostic testing or remain compliant for therapy or follow-up may come in the social history (or SH). Such information can be necessary to formulate not only an effective but also a safe treatment plan, particularly when the disease is severe and/or requires potent medication. The various pharmaceutical preparations of vitamin D, such as 1,25-dihydroxyvitamin D3 [1,25(OH)2D3], 1α-hydroxyvitamin D3 (lαOHD3), dihydrotachysterol (DHT), 25OHD3, and vitamin D2, have very different potencies, biological half-lives, and price (Chapter 61). What will the issues of long-term treatment and drug cost mean for patient compliance, safety, and/or follow-up? The medical literature contains enough case reports of renal failure from vitamin D intoxication when patient monitoring was inadequate. Because vitamin D3 is produced naturally by exposure to ultraviolet light in sunshine, and because nutrition also importantly influences mineral homeostasis, appreciation of climate, clothing, diet, and skin pigmentation is important. Several ethnic, religious, and other groups have vegetarian members who will not consume dairy products for health, spiritual, or ethical reasons (Table III). Recognition in the SH that the patient belongs to one of these populations may disclose a significant factor contributing to his/her metabolic bone disease.

918

MICHAEL P. WHYTE

Physical factors (e.g., exercise and work activities) often impact patients with metabolic bone disease. Indeed, as regimens continue to improve for increasing low skeletal mass, much of what the clinician can do for a child or adult with osteoporosis still comes from cautioning them against potentially traumatic pursuits at play or work. Prevention of falls and proscription against heavy lifting for pediatric or adult forms of osteoporosis are important for minimizing fracture and spinal deformity. In fact, this advice, often guided by history-taking, can sometimes correct vertebral crush deformities in children with brittle bone disease who (unlike adults) can naturally reconstitute their spinal anatomy (Fig. 1). 5. FAMILY HISTORY

The family history (or FH) is often revealing for metabolic bone disease, because many of these conditions are heritable [1–5]. In fact, a correct diagnosis may be disclosed by study of kindred members— familial benign (hypocalciuric) hypercalcemia or osteogenesis imperfecta are good examples. Inborn errors of vitamin D bioactivation or resistance are rare, but they may be uncovered in the FH. Furthermore, significant benefit may come from screening studies to identify, and then to treat or to counsel, other affected relatives who may also represent important clues to the patient’s future complications and prognosis.

To report that the FH is “negative” without first establishing the value of this information is misleading. If the patient is adopted, he/she is unlikely to give relevant data. An understanding of the size of a family is essential before dismissing transmission of a heritable disorder. The patient who is the only child of only children, or from a disrupted family, is not as likely to disclose a heritable disorder as will be the patient from a large, cohesive kindred. This effort can facilitate a diagnosis. Medical records from similarly affected living or deceased family members may establish the diagnosis, and may be an important guide not only concerning prognosis, but also for treatment. 6. REVIEW OF SYSTEMS

Metabolic bone diseases can cause a considerable variety of symptoms. This is especially true for disorders that disturb vitamin D homeostasis and lower extracellular calcium and phosphate levels (Table II). Therefore, a careful “review of systems” may uncover a sufficient number of these problems so that a diagnosis becomes apparent, or a new or additional condition is suspected. Furthermore, this effort provides a baseline from which to judge the impact of subsequent medical therapy. Symptoms that persist after a course of otherwise effective treatment would need further investigation if they were expected to respond.

FIGURE 1 Considerable reconstitution of vertebrae (here, L3–L5) has occurred between ages 14 (left) and 16 (right) years in a boy with idiopathic osteoporosis. He was counseled against lifting and to avoid falls and stopped participating in traumatic exercises in physical education class. No pharmacological intervention had been attempted.

919

CHAPTER 57 Approach to the Patient with Metabolic Bone Disease

B. Physical Examination Many clinical signs as well as significant skeletal deformities can accompany or result from metabolic bone disease—especially in children. Furthermore, not all of these disorders manifest themselves with overt disturbances of hormone or mineral homeostasis (e.g., postmenopausal osteoporosis). Accordingly, physical examination is especially important. Discovery as well as comprehensive treatment of metabolic bone disease often depends on this skill. Occasionally, diagnosis of a metabolic bone disease starts with the identification of one physical finding; for example, blue or gray sclerae (osteogenesis imperfecta), cafe-au-lait spots (McCune-Albright syndrome), tumor (oncogenic rickets/osteomalacia), premature loss of a deciduous tooth (hypophosphatasia), hallux valgus (fibrodysplasia ossificans progressiva), or brachydactyly (pseudohypoparathyroidism, type IA). For some metabolic bone diseases, a constellation of physical findings leads to the diagnosis. Paget disease of bone, when advanced, can feature an enlarging calvarium with bulging temporal arteries, deafness, asymmetrical bowing of the limbs, and localized areas of skeletal warmth. Postmenopausal osteoporosis causes loss of vertebral height with reduced stature leading to a “shortwaisted” appearance, kyphosis or a gibbus (dowager’s hump), a protuberant abdomen (that the patient may mistakenly attribute to weight gain), ribs lowered toward (or in) the pelvic brim, paravertebral muscle spasm, and thin skin (McConkey’s sign) [7]. Unless such physical abnormalities individually or in combination are correctly identified, a diagnosis may be missed. Furthermore, these findings should focus attention on anatomical structures of concern, perhaps requiring treatment (e.g., vertebroplasty for osteoporosis, legbracing for rickets, etc.). With disturbances in vitamin D homeostasis, a plethora of physical findings can develop (Table II). Patient age determines which are likely to be encountered. Low levels or ineffective action of vitamin D can be especially harmful for infants and for children. As discussed below, distinctive physical findings occur in the pediatric age group. Rickets disturbs the most actively growing bones. Because the skull is enlarging especially quickly at birth, craniotabes (flattened posterior skull) is characteristic of essentially congenital disease. A rachitic rosary (enlargement of the costochondral junctions) can appear during the first year of life, when the rib cage forms rapidly. During infancy or childhood, there may be flared wrists and ankles from metaphyseal widening, bony tenderness, and Harrison’s groove (rib cage ridging from

diaphragmatic pull producing a horizontal depression along the lower border of the chest at costal insertions of the diaphragm). Although weight bearing typically bows the lower limbs, especially during the adolescent growth spurt a knock-knee deformity may occur instead. The patient can also have myopathy with reduced muscle strength and tone causing a waddling gait, lax ligaments, indentation of the sternum from forces exerted by the diaphragm and intercostal muscles, delayed eruption of permanent teeth, and enamel defects. In infants, floppiness and hypotonia are characteristic. Rachitic infants and young children commonly are listless and irritable. Bone pain can also occur from fracture and deformity. Hypocalcemia (Chapter 64) can also result from vitamin D deficiency (Chapters 65 and 66), pseudodeficiency (Chapter 71), or resistance (Chapter 72) [14]. Hence, it is important that symptoms of hypocalcemia are elicited during the medical history and physical signs of latent or overt tetany are recognized during the physical examination (Table IV). Hypocalcemia enhances neuromuscular excitability. Consequently, there may be varying degrees (latent or overt) of tetany. Overt tetany usually presents with numbness and tingling

TABLE IV

Signs and Symptoms of Hypocalcemia

Nervous system Increased irritability with latent or overt tetany Mental status changes Seizures Basal ganglia calcification Mental retardation Cardiovascular Prolonged ST-interval with arrhythmia Cardiomyopathy with congestive heart failure Hypotension Other Papilledema Lenticular cataracts Intestinal malabsorption Dysplastic teeth Rickets/osteomalacia Integument changes Joint contractures Vertebral ligament calcification Reprinted with permission from MP Whyte 1993 Hypocalcemia. In: BEC Nordin, AG Need, HA Morris (eds) Metabolic Bone and Stone Disease, 3rd edn. Churchill Livingstone, Edinburgh, pp. 147–162.

920 around the mouth and in the fingertips, and can be followed by muscle spasm in the extremities, face, larynx (causing stridor), and elsewhere, and mental status changes including epileptic seizures. Symptoms and signs may be particularly striking when reductions in extracellular levels of ionized calcium are severe, or when hypocalcemia occurs rapidly. Typically, there is carpopedal spasm manifesting as adduction of the thumb, metacarpophalangeal joint flexion, and interphalangeal joint extension. Latent tetany can be unmasked by eliciting Chvostek’s sign or Trousseau’s sign. Chvostek’s sign is a spasm of the ipsilateral muscles of the face on tapping the facial nerve near its exit from the skull in the region of the parotid gland (just anterior to the ear lobe, below the zygomatic arch). A positive Chvostek’s sign ranges from twitching of the lip at the corner of the mouth, to spasm of all of the facial muscles on the stimulated side. However, slightly positive responses occur in as many as 10 to 15% of healthy adults. Trousseau’s sign is provoked when a sphygmomanometer is inflated on the arm to above the systolic blood pressure for up to 3 minutes [9]. Positive responses consist of carpal spasm with resolution occurring 5 to 10 seconds after the cuff is deflated (i.e., relaxation is not immediate). Both Chvostek’s and Trousseau’s signs can be absent, however, even in severe hypocalcemia. They seem to reflect the rapidity of change in serum calcium levels. Chronic hypocalcemia causes cataracts, dermopathy (Fig. 2), and basal ganglia calcification [14]. Growth rate is an important parameter to follow in infants and children with metabolic bone disease, especially rickets. Improvement or correction of short stature can derive from effective treatment. With it should also come reduction or resolution of skeletal deformity if there is sufficient time for longitudinal growth before

FIGURE 2 Hyperkeratotic dermatosis (shown here, posterior neck) recurs in a 19-year-old black man with pseudohypoparathyroidism, who is poorly compliant with medical therapy. When he stops treatment with calcium and vitamin D, he becomes markedly hypocalcemic, and the hyperkeratotic lesions reappear.

MICHAEL P. WHYTE

physes close at the end of puberty. Height or length is determined with a stadiometer or a tape measure with the patient in bare feet standing or supine, respectively. Weight should also be carefully assessed (and controlled, if excessive). Obesity or inordinate weight gain in girls during late childhood may transiently improve stature, not necessarily reflecting a favorable response to medical therapy. Instead, the physician can mistake the influence of excess weight on accelerating pubertal growth for efficacy of treatment. Here, the growth spurt has merely occurred early, but physes will fuse soon after menarche, negating any improvement in final adult stature. Skeletal deformation can cause much of the morbidity of metabolic bone disease. Bowing of the lower limbs predisposes to osteoarthritis especially in the knees. Prevention, control, or correction of deformities is an important goal of patient care. Without a complete physical examination, such important clinical problems may go unnoticed. Something as inexpensive as a shoe lift can be of considerable benefit, but the correct size and placement must come from accurate evaluation of leg-length inequality if iatrogenic problems are to be avoided. For children with rickets, determination of height and arm span will help to quantitate skeletal deformity as will measurements of the upper and lower segment lengths and calculation of their ratio. Simple measurements—including finger breadth separation of knees or ankles with bowed legs or knock knees, respectively— are useful. With rickets (e.g., X-linked hypophosphatemia), some time may pass before the metabolic bone disease is controlled medically with active metabolites of vitamin D3 and phosphate supplementation. Accordingly, photography, gait analysis, and even videotaping of skeletal deformity may also help assess progression or document response to therapy. A “metabolic myopathy” is a prominent clinical feature of vitamin D deficiency and tumor-induced rickets or osteomalacia. Proximal muscle weakness of the limbs is suspected from a history of difficulty rising from sitting position, negotiating stairs, or combing hair, and can be confirmed by physical examination. Gower’s sign is a simple and excellent way to detect this problem in children who are observed getting up from a seated position on the floor—if they must push up with their hands on their thighs to achieve upright posture, this is a positive test. Other routine assessments of muscle strength should be recorded [9]. In infants and children with rachitic disease, skull shape and calvarial growth should be followed— including recordings of head circumference using standard charts. Early closure of cranial sutures is not uncommon in these disorders [15]. Premature union of

921

CHAPTER 57 Approach to the Patient with Metabolic Bone Disease

sagittal sutures commonly causes a dolichocephalic skull in X-linked hypophosphatemia, but this is usually a cosmetic concern only. In hypophosphatasia, however, there can be either functional or true premature fusion of multiple cranial sutures leading to a scafalocephalic skull, sometimes with raised intracranial pressure. Dystocia from too narrow a birth canal due to pelvic deformity from vitamin D deficiency during childhood was a major cause of puerperal mortality in mothers in the early twentieth century. This deformity should be searched for in pregnant women with a history of rickets. Alopecia is a distinctive clinical finding in some patients with hereditary vitamin D-resistant rickets (vitamin D-dependent rickets, type II) (Chapter 72). However, alopecia or hypotrichosis is also a manifestation of far more prevalent vitamin D-deficiency rickets accompanying malnutrition. Additionally, hypotrichosis occurs in some forms of metaphyseal dysplasia that can be confused clinically and radiographically with rickets [5]. The benign tumors that cause oncogenic (tumorinduced) rickets or osteomalacia are often at least palpable if not visible, but they may be no more than pea-size and are sometimes hidden (Chapter 70). They are frequently found subcutaneously, but can occur anywhere on the body. Some have been discovered intravaginally or in the nasopharynx, and some lie within the skeleton. Because extirpation of these lesions is curative, especially thorough physical examination is essential when this disorder is suspected. If the physician cannot find the tumor on imaging studies (including bone and octreotide scanning), patients should be instructed concerning periodic searches for subcutaneous masses or symptoms from tumor elsewhere. Physical examination yearly is warranted in hopes that previously undetectable lesions have appeared.

TABLE V Biochemical Markers of Bone Remodeling Resorption (osteoclast products) Tartrate-resistant acid phosphatase (serum) Hydroxyproline (urine) Hydroxylysine glycosides (urine) Collagen cross-links (urine and serum) Total pyridinolines (Pyr and/or Dpy)* Free pyridinolines (Pyr and/or Dpy)* Cross-linked N- and C-telopeptides (NTX, CTX) Formation (osteoblast products) Propeptides of type I collagen (serum) C-propeptide N-propeptide Osteocalcin (serum) Alkaline phosphatase (serum) Total activity Bone-specific enzyme *Pyr = pyridinoline; Dpy = deoxypyridinoline.

usually all that are needed. Radiological procedures used for skeletal disease are relatively limited but often give critical information. However, it is wasteful to cover the diagnostic “waterfront” by ordering a “bone battery” of biochemical tests, or a series of low-yield radiological studies. Histopathological assessments are indicated in fewer situations, but may provide definitive or insightful findings.

D. Radiological Examination 1. X-RAY IMAGES

C. Laboratory Testing The medical history and physical examination guide further assessment of the patient with metabolic bone disease by laboratory methods. Beginning the testing without this fundamental information, but instead with biochemical or radiological studies, is not appropriate. Of consternation for some physicians who only occasionally encounter patients with metabolic bone disease is the often bewildering array of expensive assays for factors that condition mineral or skeletal homeostasis as well as the plethora of markers of skeletal apposition or resorption (Table V). Although some biochemical testing is necessary for effective diagnosis or treatment of metabolic bone disease, especially disturbances of vitamin D homeostasis, a few studies are

Radiographs of the skeleton chosen selectively are often crucial for diagnosis and follow-up of patients with metabolic bone disease (Chapter 60) [10–12,16]. However, the “skeletal survey” that examines all bones is an expensive, relatively insensitive, and laborious procedure, causing a not trivial exposure to X-irradiation. Visualization of the shape of the entire skeleton (but using films of just one upper and one lower extremity) is usually indicated to assess a bone dysplasia, but is rarely necessary for evaluation of metabolic bone disease. Instead, the “metabolic bone survey” generally gives the necessary radiographic information for diagnosis and follow-up. Here, one studies the appendicular, as well as the axial skeleton, and therefore delineates both cortical and trabecular bone as well as “yellow” and “red” marrow spaces. The necessary films are

922 a lateral view of the skull and thoracolumbar spine, posteroanterior view of a hand and wrist as well as the chest, and an anteroposterior view of the pelvis and a knee. Chapter 60, as well as several comprehensive texts [10–12,16], and other resources (London Dysmorphology Database, Oxford University Press and POSSUM—Pictures Of Standard Syndromes And Undiagnosed Malformations—The Murdoch Institute for Research Into Birth Defects, Melbourne, Australia), describe the radiographic findings of the metabolic bone diseases and help to distinguish the skeletal dysplasias. When there is rickets, posteroanterior radiographs of the hands and anteroposterior radiographs of the knees will precisely document the presence and degree of physeal and metaphyseal change. Long cassette films of the lower extremities, taken while the patient is standing, help quantitate bowing or knock-knee deformity. Radiographic studies can also provide clues to the particular etiology or pathogenesis of rickets. For those disorders that cause rickets by disturbing vitamin D homeostasis and result in secondary hyperparathyroidism, subperiosteal bone resorption and osteopenia may be noted in addition to growth plate widening and irregular metaphyses (Fig. 3). These findings contrast to most cases of X-linked hypophosphatemia where the skeleton typically has normal or sometimes increased radiodensity and evidence of hyperparathyroidism is generally absent. In hypophosphatasia, there are characteristically peculiar “tongues” of radiolucency that project from the physes into the metaphyses. Rachitic disease of recent onset will symmetrically widen growth plates, whereas long-standing rachitic disease with bony deformity alters the mechanical forces acting on the physes, which in turn become asymmetrically widened. Accordingly, chronic rachitic disease sometimes seems more difficult to diagnose with x-rays (Fig. 4). The rapidity of resolution of rachitic changes on x-ray images also may be of diagnostic significance. With vitamin D deficiency rickets from lack of sunlight exposure, radiographic improvement can occur rapidly (within a few weeks) following a single pharmacological dose of vitamin D. Other forms of rickets, especially those due to renal phosphate wasting, generally take longer (several months or more) to improve or correct with appropriate medical therapy. Sequential radiographs are essential to evaluate the response to therapy for rickets. 2. BONE SCINTIGRAPHY

Bone scintigraphy is an excellent tool for uncovering a variety of abnormalities of the skeleton, but it

MICHAEL P. WHYTE

does not establish a diagnosis [17]. Enhanced radioisotope uptake occurs in areas of increased blood flow to the skeleton, excess osteoid, and accelerated bone formation. Cost-effective assessment starts with the bone scan followed by x-rays of the “hot spots” to guide further study. Bone scanning is generally unnecessary in children with rachitic disease, unless a search for a skeletal source of tumoral rickets is needed when physical examination fails to yield an obvious lesion. In adults, bone scanning can also disclose complications of osteomalacia, including “true” as well as “false” (pseudo) fractures. 3. BONE DENSITOMETRY

Bone mass quantitation is now widely available for clinicians [6]. Dual energy x-ray absorptiometry (DEXA) and quantitative computed tomography (QCT) can give helpful assessments of skeletal mass designated “bone mineral density” (BMD). DEXA or QCT densitometry, however, does not provide a diagnosis. In fact, each of the principal categories of metabolic bone disease (see below)—namely, osteoporosis, osteomalacia, and osteitis fibrosa cystica—can manifest with low BMD. The presence of worrisome BMD revealed by these techniques is merely a point of departure for differential diagnosis. For those rare conditions associated with increased BMD, these tools are similarly helpful. Densitometry has some important technical caveats. When using DEXA for children, it is crucial to understand that the technique generates a so-called areal (two-dimensional) rather than volumetric (threedimensional) assessment of BMD. BMD is measured as gm/cm2 by DEXA and as gm/cm3 by QCT. Accordingly, values for BMD from DEXA are importantly dependent on body size. Children have lower BMD on DEXA compared to adults, but this is largely an artifact of their body size. Small stature individuals, who have otherwise normal skeletons, will seem to have low BMD compared to taller subjects. Hence, the pediatric age group (or individuals who are small for reasons other than metabolic bone disease) may be incorrectly diagnosed with “osteopenia” or “osteoporosis” if this technical phenomenon with DEXA goes unrecognized or uncorrected. QCT can measure BMD with the advantage that it provides a volumetric measurement, and can focus on either cortical or trabecular bone, but with higher exposure to X-irradiation. QCT can also precisely evaluate the anatomy of the skeleton and readily detect extracellular calcification [17]. Quantitation of low BMD in children uses Z-scores (standard deviations from mean values) matched for age and gender and ideally for ethnicity, whereas T-scores (which refer

923

CHAPTER 57 Approach to the Patient with Metabolic Bone Disease

A

B

C

D

E

F

FIGURE 3 Characteristic changes of rickets (growth plate widening and frayed ends of metaphyses) occur in anteroposterior radiographs of the knees of five boys each newly diagnosed with a different type of rachitic disease (A–E). However, as shown, additional information is apparent concerning the pathogenesis and/or etiology. (A) A 1-year-old boy with osteopenia due to secondary hyperparathyroidism from “nutritional” rickets. (B) Similar findings in a 5-year-old boy with osteopenia consistent with documented secondary hyperparathyroidism from anticonvulsant-induced rickets. (C) A 3-year-old boy has normal bone mass and no secondary hyperparathyroidism, consistent with X-linked hypophosphatemia (XLH). (D) A 10-year-old boy has symmetrical widening of the growth plates (seen best in the proximal tibia) but no long bone deformity. Together, these changes suggest recent-onset disease in keeping with suspected tumor-induced rickets. (E) A 10-year-old boy has “tongues” of radiolucency (arrows), that project from physes into metaphyses, characteristic of the childhood form of hypophosphatasia. (F) However, not all disorders that cause metaphyseal irregularity are forms of rickets. This 4-year-old boy, with unremarkable biochemical studies and iliac crest histology following “tetracycline labeling,” has metaphyseal dysplasia.

924

MICHAEL P. WHYTE

TABLE VI

Causes of Chronic Hypophosphatemia

Decreased intestinal absorption Alcohol abuse Antacid abuse Vitamin D deficiency Malabsorption Starvation

FIGURE 4 A 9-year-old girl, who is poorly compliant for medical therapy for X-linked hypophosphatemia (XLH), has lower limb bowing that asymmetrically deforms the growth plates in her knees.

to peak bone mass) are used to assess low bone mass in adults. Software may be FDA-approved for displaying a child’s lumbar spine BMD value, but not for other skeletal regions (e.g., hip). A Web site (http://www-statclass.stanford.edu/pediatric-bones/) provides DEXAbased BMD reference ranges for various skeletal sites in children, but cautions that the information reflects research experience (“not for diagnosis”) and warns about the caveats (see above) for interpreting areal BMD values in children. Interpretation of high BMD values, uses Z-scores for either children or adults [18]. 4. OTHER RADIOLOGICAL PROCEDURES

Magnetic resonance imaging (MRI) is particularly useful for marrow space examination, including delineation of ischemic necrosis of bone. Fat-suppressed imaging showing bone edema may disclose a cause for unexplained bone pain. New applications of MRI include assessment of trabecular bone microanatomy [17]. Ultrasound studies of the skeleton may provide qualitative as well as quantitative information [6].

E. Biochemical Investigation Understandably, circulating calcium levels are closely scrutinized in patients with metabolic bone disease, especially those suspected of having rickets or osteomalacia (Chapters 65 and 66). However, extracellular (serum) phosphate levels may be equally (perhaps more)

Increased urinary losses Renal tubular defects X-linked hypophosphatemia (XLH) Oncogenic rickets or osteomalacia (tumor-induced osteomalacia) Abnormalities of vitamin D metabolism Vitamin D deficiency Vitamin D-dependent rickets Hyperparathyroidism Renal transplantation Alcohol abuse Poorly controlled diabetes mellitus Metabolic or respiratory acidosis Drugs: calcitonin, diuretics, glucocorticoids, bicarbonate Respiratory alkalosis Extracellular fluid volume expansion Severe burns

important when there is defective skeletal mineralization. Chronic hypophosphatemia occurs in a number of conditions associated with rickets or osteomalacia (Table VI). Accordingly, hypophosphatemia is an especially important finding. Clinicians must recognize that the pediatric age group normally has higher fasting blood phosphate levels compared to adults. Serum phosphate levels should be assayed with fasting blood specimens, because food (depending on content) can acutely raise or lower the level. Assay of phosphate levels in a urine specimen can help to distinguish whether hypophosphatemia is due to renal phosphate wasting or to dietary deficiency. Detailed tests of renal phosphate handling using timed collections [e.g., tubular reabsorptive maximum for phosphorus/glomerular filtration rate (TmP/GFR) ratios provide a definitive assessment [2,3]. Disturbances in vitamin D stores or bioactivation commonly result in hypocalcemia, secondary hyperparathyroidism and, consequently, hypophosphatemia. Assay of the serum levels of the major active vitamin D metabolites—25OHD and 1,25(OH)2D—is essential to detect disturbances in vitamin D stores or in vitamin D bioactivation, respectively. The important effects of season on serum 25OHD levels should be considered.

CHAPTER 57 Approach to the Patient with Metabolic Bone Disease

Furthermore, 25OHD is transported in the circulation bound to vitamin D-binding protein (DBP). Accordingly, hypoproteinemia must be recognized before interpreting a serum 25OHD concentration. Low levels of serum 25OHD usually indicate vitamin D deficiency, but this biochemical finding is merely a starting point for differential diagnosis. A considerable TABLE VII

925

number of disorders cause rickets or osteomalacia where serum 25OHD levels are low (Table VII). Assay of serum 1,25(OH)2D concentration is most helpful in exploring

hypercalcemia, but is also necessary to characterize inborn errors of vitamin D bioactivation. Finding a low serum 1,25(OH)2D level also helps in the investigation of oncogenic osteomalacia. However, serum

Causes of Rickets or Osteomalacia

I. Vitamin D deficiency A. Deficient endogenous synthesis 1. Inadequate sunshine 2. Other factors, e.g., genetic, aging, pigmentation B. Dietary 1. Classic “nutritional” 2. Fat-phobic II. Gastrointestinal A. Intestinal 1. Small-bowel diseases with malabsorption, e.g., celiac disease (gluten-sensitive enteropathy) B. Hepatobiliary 1. Cirrhosis 2. Biliary fistula 3. Biliary atresia C. Pancreatic 1. Chronic pancreatic insufficiency III. Disorders of vitamin D bioactivation A. Hereditary 1. Vitamin D dependency, type I (pseudovitamin D deficiency) 2. Vitamin D dependency, type II (hereditary vitamin D-resistant rickets) B. Acquired 1. Anticonvulsant therapy 2. Renal insufficiency (see below) IV. Acidosis A. Distal renal tubular acidosis (classic, type I) 1. Primary (specific etiology not determined) a. Sporadic b. Familial 2. Secondary a. Galactosemia b. Hereditary fructose intolerance with nephrocalcinosis c. Fabry’s disease 3. Hypergammaglobulinemic states 4. Medullary sponge kidney 5. Post renal transplantation

B. Acquired 1. Ureterosigmoidostomy 2. Drug-induced a. Chronic acetazolamide use b. Chronic ammonium chloride use V. Chronic renal failure VI. Phosphate depletion A. Dietary 1. Low phosphate intake 2. Total parenteral nutrition 3. Aluminum hydroxide antacid abuse (or other nonabsorbable hydroxides) B. Impaired renal tubular (? intestinal) phosphate reabsorption 1. Hereditary a. X-linked hypophosphatemic rickets b. Adult-onset vitamin D–resistant hypophosphatemic osteomalacia 2. Acquired a. Sporadic hypophosphatemic osteomalacia (phosphate diabetes) b. Tumor-associated rickets and osteomalacia c. Neurofibromatosis d. McCune-Albright syndrome VII. General renal tubular disorders (Fanconi syndrome) A. Primary renal 1. Idiopathic a. Sporadic b. Familial 2. Associated with systemic metabolic process a. Cystinosis b. Glycogenosis c. Lowe’s syndrome B. Systemic disorder with associated renal disease 1. Hereditary a. Inborn errors (i) Wilson’s disease (ii) Tyrosinemia Continued

926

MICHAEL P. WHYTE

TABLE VII

Causes of Rickets or Osteomalacia—Cont’d

2. Acquired a. Multiple myeloma b. Nephrotic syndrome c. Transplanted kidney 3. Intoxications a. Cadmium b. Lead c. Outdated tetracycline VIII. Primary mineralization defects A. Hereditary 1. Hypophosphatasia B. Acquired 1. Bisphosphonate intoxication 2. Fluorosis

1,25(OH)2D levels should be interpreted mindful that pediatric and adult reference ranges differ, and that low serum phosphate levels are expected to increase 1,25(OH)2D levels in healthy individuals. Hypophosphatemia without physiological elevation in 1,25(OH)2D levels points to a disturbance in renal 1,25(OH)2D biosynthesis.

F. Histopathological Assessment Throughout the past century, clinicians diagnosed and dealt with three major types of metabolic bone disease based on the principal histopathological features and the mineral-to-protein ratios—osteoporosis, osteomalacia, and hyperparathyroidism (osteitis fibrosis cystica). Biopsy of the iliac crest samples cortical as well as trabecular bone and distinguishes especially well these types of metabolic bone disease [13]. Histological examination of the iliac crest must, however, follow in vivo “labeling” of the patient’s skeleton by ingestion of two 3-day courses of tetracycline [6,13] and nondecalcified sectioning and histomorphometric analysis of specimens. However, which of these types of bone disease is present actually provides another starting point for differential diagnosis [2,3,13]. Bone biopsy is not routinely needed to diagnose rachitic disease, which is easily detected by radiological studies of the physes of the wrist and knees together with biochemical testing. In adults with osteomalacia, however, growth plate changes on x-ray imaging have been “lost” as guideposts for diagnosis

IX. States of rapid bone formation with or without a relative defect in bone resorption A. Postoperative hypoparathyroidism with osteitis fibrosa cystica B. Osteopetrosis (“osteopetrorickets”) X. Defective matrix synthesis A. Fibrogenesis imperfecta ossium XI. Miscellaneous A. Magnesium-dependent B. Steroid-sensitive C. Axial osteomalacia

as well as for judging the efficacy of treatment. Thus, bone biopsy may be more helpful in adult patients.

III. TREATMENT Treatment of metabolic bone disease can range from simple (e.g., exposure to sunlight for environmental vitamin D deficiency rickets) to complex (e.g., bone marrow transplantation for the malignant form of osteopetrosis). When there is a disturbance in vitamin D homeostasis, pharmaceuticals are usually needed. However, additional treatment approaches may be required when there is skeletal deformity. Patients who require vitamin D sterols can benefit greatly; but medical care often must be especially skilled. Rickets or osteomalacia can occur from deficiency or impaired bioactivation of vitamin D. Either disturbance will decrease calcium absorption from the gastrointestinal tract, leading to variable degrees of hypocalcemia, secondary hyperparathyroidism, and hypophosphatemia. The hypocalcemia and hypophosphatemia together engender the defective skeletal mineralization (Chapters 65 and 66). Fortunately, it is possible to prevent, cure, or control most aberrations in vitamin D homeostasis. Vitamin D deficiency stemming from socioeconomic factors has relatively uniform etiology and pathogenesis, considerable prevalence, and a long history for mankind that make prevention or treatment well understood and theoretically straightforward. Inborn errors of vitamin D bioactivation can be relatively easy to control, e.g., merely by providing a “replacement” dose of

CHAPTER 57 Approach to the Patient with Metabolic Bone Disease

1,25(OH)2D3 for pseudovitamin D deficiency rickets (vitamin D–dependent rickets, type I), or extremely difficult to treat, as in some patients with hereditary vitamin D resistant rickets (vitamin D-dependent rickets, type II). Here, resistance to 1,25(OH)2D is so great in some cases that it must be “bypassed” therapeutically by providing calcium intravenously in order to heal the rickets. For other clinical situations (especially hepatobiliary, gastrointestinal, or pancreatic disease) leading to vitamin D deficiency, the precise regimen and duration of treatment will vary greatly from patient to patient and depend on measures to control the primary disorder. Hence, an accurate diagnosis, understanding of pathogenesis, and familiarity with the pharmacological armamentarium is essential for successful therapy which may require adjustment from time to time. Now, a considerable number of pharmaceuticals are available for the treatment of metabolic bone disease. In the United States, the three major biological forms of vitamin D [vitamin D, 25OHD, and 1,25(OH)2D] can be prescribed. Also still available is dihydrotachysterol. There are also several preparations that provide inorganic phosphate, and innumerable types of calcium supplementation. Hence, both repletion of vitamin D reserves as well as bypassing steps in vitamin D bioactivation are treatment options. Furthermore, direct (intravenous) administration of calcium is possible when there might be insufficient or delayed action of a vitamin D sterol on the gastrointestinal tract. Importantly, these drugs have different durations of fat and muscle storage and potency. Additionally, the lag times for onset of action and longevity of biological effects vary considerably for these preparations. Finally, the cost of these pharmaceuticals differ greatly (Chapter 61). An accurate diagnosis from among the many disorders that cause rickets or osteomalacia is essential for therapeutic efficacy, safety, and economy. An understanding of the pathophysiology of the aberration in vitamin D homeostasis is required for effective and safe therapy. Use of 1,25(OH)2D3, when there is depletion of body stores of vitamin D, merely circumvents but does not correct the basic problem. Because many disorders disturb vitamin D homeostasis and their severities can differ markedly, each patient and condition must be treated individually. From the above, it is apparent that follow-up of medical treatment for metabolic bone disease—especially rickets and osteomalacia—is essential for many reasons. Only some generalities can be discussed here. For infants, children, and adolescents with rickets, close monitoring of drug therapy is especially important because these individuals are growing. Not only will

927

biochemical and skeletal dynamics change in response to treatment, but there is increasing body size to contend with. Alterations in weight, a growth spurt, or healing of rachitic disease will all likely require changes in dosage. Greater weight means higher doses of vitamin D sterols and/or mineral supplements to control an otherwise static metabolic disturbance. A growth spurt can undo successful therapy if treatment does not increase to keep up with the greater skeletal demands. Conversely, healing of rickets or osteomalacia may require a reduction in dosage, because the skeletal “sump” is now satisfactorily mineralized. It should be clear that individualized follow-up, especially for pediatric patients, is crucial. When managing chronic forms of rickets, help from other medical and surgical (e.g., orthopedic) subspecialties is often necessary, and is optimal when there is significant interdisciplinary exchange. Children with rickets who do not respond to pharmacologic therapy may benefit from bracing, epiphysiodesis (physeal stapling), or osteotomy to improve lower limb deformity. However, these procedures can reflect suboptimal medical therapy and are usually avoidable if dosing, compliance, and follow-up are satisfactory. For some patients, particularly children and adolescents, pill counts to assess compliance can provide important information. Cooperation and monitoring can be improved using pillboxes marked by days of the week. School nurses may help administer patient medications when family life is disrupted. If deformities are not obviously correcting, at least yearly orthopedic evaluation (perhaps biannually during the growth spurt) is advisable. Not all types of rickets or osteomalacia manifest with hypocalcemia or low serum levels of 25OHD or 1,25(OH)2D. Hypophosphatemia, with or without hypocalcemia, can lead to defective skeletal mineralization. Nevertheless, many of these disorders respond to treatment with 1,25(OH)2D3 and mineral supplements [2,3]. Here again, the etiology and pathogenesis of the specific condition must be understood for successful medical management. The most common form of heritable rickets, X-linked hypophosphatemia (XLH), is transmitted as an X-linked dominant trait. The pathogenesis of XLH includes a renal tubular defect engendering loss of phosphate by the kidney [3]. Despite hypophosphatemia, serum 1,25(OH)2D levels are paradoxically normal instead of elevated. More complex proximal renal tubular defects that also feature renal phosphate wasting (Fanconi syndrome) can reflect other inborn errors of metabolism, heritable disorders, or exposure to certain drugs or toxins including heavy metal poisoning (Table VI). Acquired hypophosphatemic rickets

928

MICHAEL P. WHYTE

IV. SUMMARY

FIGURE 5 In X-linked hypophosphatemia (XLH), the extremes of lower limb deformity in children that can be corrected by pharmacological therapy alone are not widely appreciated. Accordingly, close cooperation between the medical and orthopedic disciplines is essential. At age 3 years, a girl with XLH has severe bowing deformity of the lower extremities that can be quantitated with standing, long-cassette radiographs. Compliance for 1,25(OH)2D3 and inorganic phosphate supplementation therapy was excellent, and her deformity is remarkably improved by age 10 years without osteotomy, epiphysiodesis, or lower extremity bracing.

The complex interaction of the many exogenous and endogenous factors that impact vitamin D homeostasis, discussed throughout this book, explain why patients with disturbances in this endocrine process are especially challenging. Nevertheless, these individuals typically benefit greatly from the efforts of physicians who understand such disorders. Demonstration of concern for, and commitment to, the patient begins with a complete medical history and thorough physical examination that wins their confidence and trust. Such rapport will likely be essential for effective management of what will often prove to be a chronic disorder. Physical examination is crucial not only for diagnosis, but to uncover structural skeletal problems. Information gathered by the medical history and physical examination will be the guide to the myriad of biochemical, radiological, and other technologies that can help to establish the etiology and pathogenesis and to set the stage for subsequent treatment and follow-up. Effective therapies are available for derangements of vitamin D homeostasis, but the pharmaceuticals vary significantly in potency, duration of effect, and cost. Once the proper clinical foundation is in place, the physician will usually be gratified by a patient that he/she has greatly helped.

Acknowledgments Supported by Shriners Hospitals for Children and the Clark and Mildred Cox Inherited Metabolic Bone Disease Research Fund.

References

is also a common complication in McCune-Albright syndrome and is characteristic of oncogenic rickets. Treatment with 1,25(OH)2D3 and phosphate is used for the rickets complicating these disorders. Normalization of blood phosphate levels occurs only transiently with oral phosphate supplementation in conditions characterized by renal phosphate wasting. Nevertheless, clinical improvement can be substantial (Fig. 5), without a likely hazardous push to correct phosphate levels measured with fasting blood specimens. Correction of hypophosphatemia should not be considered the objective of therapy, but rather correction of deformity and restoration of an adequate growth rate. Radiographic and other biochemical studies are, instead, useful for judging adequacy and safety or therapy.

1. Avioli LV, Krane SM 1998 Metabolic bone disease and clinically related disorders. Academic Press, San Diego, CA. 2. Coe FL, Favus MJ 2002 Disorders of bone and mineral metabolism. Lippincott, Williams & Wilkins, Philadelphia, PA. 3. Favus MJ 2003 Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism. American Society for Bone and Mineral Research, Washington, DC. 4. Scriver CR, Beaudet AL, Sly WS, Valle D 2001 The metabolic and molecular bases of inherited disease. McGraw-Hill, New York, NY. 5. McKusick V. Online Mendelian Inheritance in Man, OMIM (TM). McKusick-Nathans Institute for Genetic Medicine, Johns Hopkins University (Baltimore, MD) and National Center for Biotechnology Information, National Library of Medicine (Bethesda, MD). (http://www.ncbi.nlm.nih.gov/ entrez/query.fcgi?db=OMIM). 6. Tovey FI, Stamp TCB 1995 The Measurement of Metabolic Bone Disease. Parthenon Publishing Group, New York, NY. 7. Marcus R, Feldman D, Kelsey JL 1996 Osteoporosis. Academic Press, San Diego, CA.

CHAPTER 57 Approach to the Patient with Metabolic Bone Disease

8. Bilezikian JP, Raisz LG, Rodan GA 2002 Principles of bone biology. Academic Press, San Diego, CA. 9. Degowen, Degowen. Diagnostic Examination. 1999 McGrawHill Health Professionals Division New York, NY. 10. Edeiken J, Dalinka MK, Karasick D 1990 Edeiken’s roentgen diagnosis of diseases of bone. Williams & Wilkins, Baltimore, MD. 11. Taybi H, Lachman RS 1996 Radiology of syndromes, metabolic disorders, and skeletal dysplasias. Mosby, St. Louis, MO. 12. Ravell PA 1985 Pathology of Bone. Springer-Verlag, Berlin, Germany. 13. Resnick D 2002 Diagnosis of bone and joint disorders. Saunders, Philadelphia.

929

14. Whyte MP 1993 Hypocalcemia, In: BEC Nordin, AG Need, HA Morris (eds) Metabolic Bone and Stone Disease. Churchill Livingstone, Edinburgh, UK, pp. 147–162. 15. Reilly BJ, Leeming JM, Fraser D 1964 Craniosynostosis in the rachitic spectrum. J Pediatr 64:396–405. 16. Greenfield GB 1990 Radiology of Bone Diseases. Lippincott, Williams & Wilkins, Philadelphia, PA. 17. Resnick D 1996 Bone and Joint Imaging. Saunders, Philadelphia. 18. Whyte MP Misinterpretation of osteodensitometry in high bone mass disease (it isn’t all osteoporasis). Journal of Clinical Densitometry (submitted for publication).

CHAPTER 58

Detection of Vitamin D and Its Major Metabolites* BRUCE W. HOLLIS

I. II. III. IV.

Departments of Pediatrics, Biochemistry, and Molecular Biology, Medical University of South Carolina, Charleston, South Carolina

Introduction Detection of Vitamin D Detection of 25OHD Detection of 24,25(OH)2D

I. INTRODUCTION Vitamin D is a 9,10-seco steroid and is treated as such in the numbering of its carbon skeleton (Fig. 1). Vitamin D occurs in two distinct forms: vitamin D2 and vitamin D3. As shown in Fig. 1, vitamin D3 is a 27-carbon derivative of cholesterol; vitamin D2 is a 28-carbon molecule derived from the plant sterol ergosterol. Besides containing an extra methyl group, vitamin D2 differs from vitamin D3 in that it contains a double bond between carbons 22 and 23. The most important aspects of vitamin D chemistry center on its cis-triene structure. This unique cis-triene structure makes vitamin D and related metabolites susceptible to oxidation, ultraviolet (UV) light-induced conformational changes, heatinduced conformational changes, and attack by free radicals. As a rule, the majority of these transformation products have lower biological activity than vitamin D. It is important to note that, in humans, vitamin D2 and D3 provide similar potency, (although some controversy exists as discussed in Chapter 61), and in this chapter the term vitamin D refers to both compounds. Metabolic activation of vitamin D is achieved through hydroxylation reactions at both carbon 25 of the side chain and, subsequently, carbon 1 of the A ring. Metabolic inactivation of vitamin D takes place primarily through a series of oxidative reactions at carbons 23, 24, and 26 of the side chain of the molecule. These metabolic activations and inactivations are well characterized and result in a plethora of vitamin D metabolites (Fig. 2). Of the compounds shown in Fig. 2, only four, vitamin D, 25-hydroxyvitamin D (25OHD), 24,25dihydroxyvitamin D [24,25(OH)2D], and 1,25-dihydroxyvitamin D [1,25(OH)2D] have been extensively

V. Detection of 1,25(OH)2D VI. Clinical Interpretation and Relevance of Antirachitic Sterol Measurements References

quantitated, and only two of those, namely, 25OHD and 1,25(OH)2D, provide any clinically relevant information. However, the quantitation of vitamin D and 24,25(OH)2D can provide important information in a research environment. Thus, this chapter addresses the quantitation of these four important vitamin D compounds. Further, it is not the intent of this chapter to address the detailed history of vitamin D metabolite analysis, as this can be obtained from previous reviews [1–3]. Rather, the intent of this chapter is to describe how we currently measure vitamin D and its major metabolites in our laboratory, as well as to discuss the appropriate clinical judgments in the selection of a given compound for analysis. The first semiquantitative assay for vitamin D was a bioassay based on the rat-line test [4]. This assay was cumbersome, expensive, and relatively inaccurate. Real progress in vitamin D analysis was not achieved until the advent of high specific activity 3H-labeled vitamin D3 compounds [5]. The introduction of these 28 CH3 22

21

18 CH3 20 12 11 13 9

14

17 16 15

24 23

26 25 CH3 27

8 7 6 5 4

CH2

CH2

10

3 2

1

HO

HO

Vitamin D3

Vitamin D2

*In

the interest of full disclosure, the author wishes to inform the readers that he has been a paid consultant to the DiaSorin Company. VITAMIN D, 2 EDITION FELDMAN, PIKE, AND GLORIEUX ND

FIGURE 1

Molecular structures of vitamins D2 and D3. Copyright © 2005, Elsevier, Inc. All rights reserved.

932

BRUCE W. HOLLIS

DIET

OH

7-dehydrocholesterol hν HO

5(E)-10-nor-10-oxo-D3

O

Vitamin D3

HO

OH

1α,24R(OH)2D3

8α,25(OH)2-9, 10-seco-

OH -4,6,10(19)-choleste O

5(E)-25(OH)-19-nor-10-oxo-D3

25OHD3

trien-3-one

O OH

HO OH OH OH

HO

HO

HO O C O O

HO

25R(OH)-26,23S-peroxylactone D3

OH

O

24S,25(OH)2D3 HO

24,25,26(OH)3D3

HO

OH

HO

OH

25(OH)-24-oxo-D3 HO

O

HO CH2OH

1α,23,25(OH)3 -24-oxo-D3 OH CH2OH

C

HO

OH

OH OH OH 1α,23S,25R,26(OH)3D3

HO

OH

HO

OH OH 1α,25R(OH)2-26,23S-lactol-D3 OH O

OH

HO HO OH HO HO 25R(OH)-26,23S- 25,26,27-trinor- 23(OH)-24-25,26,27- 1,23(OH)2-24,2-tetranor-D 26,27,-tetranor-D -lactone-D3 -24-COOH-D3 3 3

HO 24,25,26,27-tetranor -23-COOH-D3

1α,24S,25(OH)3D3 HO

OH O

O

HO OH 1α,26R,26(OH)3D3 OH

1α,25(OH)2 -24-oxo-D3

OH

23S,25(OH)2-24-oxo-D2

C OH

OH OH

O

OH OH

HO O OH

HO

OH

OH OH

OH OH 1α,238,25(OH)3D3

1α,24R-26(OH)2D3

O OH OH

O C OH O 25R(OH)-26-23S-lacto-D3

O C OH O

OH

OH

HO

OH OH

O H 23,25R-26(OH)3D3

HO 5(E)-24R,25(OH)2-19-nor-10-oxo-D3 OH OH

25S,26(OH)2D3 25R,26(OH)2D3 (mixture1:1)

OH 23,24,25(OH)3O3

1α,25(OH)2D3

OH OH

OH OH 23S,25(OH)2D3

HO

OH

OH

HO

25(OH)-23-dehydro-D3

OH

OH

OH HO

OH

5(E)-25OHD3

OH 24R,25(OH)2D3

OH

HO

OH

HO OH

25(OH)-23-oxo-D3 O

HO

OH

OH

OH

24OHD3

O C CH O HO OH 1α,25R(OH)2-26,23Slactonel-D3

O OH

OH

1α,(OH)-24,25,26,27tetranor-23-COOH-D3 (Calcitroic acid)

FIGURE 2

Summary of metabolic transformations of vitamin D3. From Bouillon R, Okamura WH, Norman AW 1995 Structurefunction relationships in the vitamin D endocrine system. Endocrine Review 16:200–257. © The Endocrine Society.

tracers led to the development of competitive protein binding assays (CPBA) for vitamin D and 25OHD [6,7]. A short time later CPBA for 24,25(OH)2D and radioreceptor assays (RRA) for 1,25(OH)2D were introduced [8,9]. In the late 1970s, high-performance liquid chromatographic (HPLC) analytical procedures for vitamin D and 25OHD were described [10,11]. Subsequently, radioimmunoassay (RIA) techniques began to appear as a means to quantitate 25OHD and 1,25(OH)2D [12,13]. Finally, recent advances in antirachitic sterol analysis have included RIA

coupled with 125I-labeled tracers that require little or no chromatographic treatment of the sample [14,15] and the instrument automation for the direct detection of circulating 25OHD. The assays for vitamin D and its major metabolites that we currently utilize in our laboratory are described in this chapter. These assays are all standalone types of assays as opposed to the multiple-metabolite assays described in years past [16,17]. We chose to do this because seldom in a clinical situation does one require a battery of vitamin D metabolite values. Further, many

933

CHAPTER 58 Detection of Vitamin D and Its Major Metabolites

assays, especially for 25OHD and 1,25(OH)2D, have been optimized as single metabolite procedures [14,15].

II. DETECTION OF VITAMIN D

UV quantitation of vitamins D2 and D3 following nonaqueous reversed phase HPLC provides an accurate, convenient method to measure circulating vitamin D. This method is described here in detail.

A. Background B. Methodology Vitamin D, the parent compound, is by far the most lipophilic of the antirachitic sterols, and for this reason it is the most difficult to quantitate (Table I). The first serious attempt to quantitate vitamin D was performed in 1971 utilizing CPBA [7]. This initial study grossly overestimated the actual amount of circulating vitamin D because of insufficient sample prepurification prior to CPBA. It was later shown that vitamin D could be assessed by CPBA, but only following extensive chromatographic purification of the organic extract, including HPLC [18,19]. The first valid determination of circulating vitamin D was achieved in 1978 by utilizing direct UV detection following a two-step HPLC purification procedure [10]. A short time later, valid CPBA were introduced for the quantitation of circulating vitamin D [18,19]. However, these procedures were cumbersome, as they required extensive sample prepurification prior to CPBA, including HPLC. Vitamin D is also difficult to quantitate because it is the only antirachitic sterol that cannot be extracted from aqueous media utilizing solid-phase extraction techniques [20]. Therefore, unlike its more polar metabolites, vitamin D must be extracted from serum or plasma using liquid-liquid organic extraction techniques. Many of the initial studies used Bligh and Dyer-type total lipid extraction to extract vitamin D from serum samples [7,10,18]. However, these types of extractions remove an extraordinary amount of lipid from the plasma sample. We therefore utilized a more selective organic extraction procedure incorporating methanol-hexane [21]. This extraction method coupled with open cartridge silica chromatography and direct

1. SAMPLE EXTRACTION

A 0.5- to 1-ml of sample serum or plasma is placed into a 13 × 100 mm borosilicate glass culture tube containing 1000 cpm of 3H-vitamin D3 in 25 µl of ethanol to monitor recovery of the endogenous compound through the extraction and chromatographic procedures. Following a 15-min incubation with the tracer, 2 plasma volumes of HPLC-grade methanol are added to each sample. The sample is then vortex-mixed for 1 min followed by the addition of 3 plasma volumes of HPLC-grade hexane. Each tube is capped and vortexmixed for an additional 1 min followed by centrifugation at 1000 g for 10 min. The hexane layer is removed into another 13 × 100 mm culture tube, and the aqueous layer is reextracted in the same fashion. The hexane layers are combined and dried in a heated water bath, 55°C, under N2. The lipid residue is then resuspended in 1 ml of HPLC-grade methylene chloride and capped. 2. SILICA CARTRIDGE CHROMATOGRAPHY

Silica Bond-Elut cartridges (500 mg) and a Vac-Elut cartridge rack were obtained from Varian Instruments (Harbor City, CA). The silica cartridges are washed in order with 5 ml HPLC-grade methanol, 5 ml HPLCgrade isopropanol, and 10 ml HPLC-grade methylene chloride. The sample, in 1 ml of methylene chloride, is then applied to the cartridge and eluted through the cartridge under vacuum into waste. This initial step is followed by 3 ml of 0.2% isopropanol in methylene chloride (discard) and 8 ml of the same solvent (vitamin D). The 8-ml fraction contains vitamins D2 and D3 and is

TABLE I Significant Methods for the Estimation of Vitamin D in Human Seruma Detection method CPBA HPLC CPBA CPBA HPLC

Extraction Methanol-chloroform Methanol-chloroform Ether-methylene chloride Methanol-methylene chloride Methanol-hexane

Preliminary chromatography Silicic acid Preparative HPLC Sephadex LH-20, preparative HPLC Lipidex-5000, preparative HPLC Silica cartridges, Preparation HPLC

Ref. Belsey et al. [7] Jones [10] Horst et al. [19] Hollis et al. [18] Liel et al. [21]

Normal circulating levelsb 24–40 ng/ml 2.2 ± 1.1 ng/ml – 2.3 ± 1.1 ng/ml 9.1 ± 1.0 ng/ml (normal), 1.3 ± 0.1 ng/ml (obese)

a As noted in the text, the Belsey method overestimated the circulating vitamin D levels because of insufficient prepurification prior to competitive protein binding assay (CPBA). b ng/ml × 2.6 = nmol/liter.

934

BRUCE W. HOLLIS

subsequently dried in a heated water bath, 55°C, under N2. The elution profile of vitamin D from the silica cartridge is displayed in Fig. 3. The silica cartridges can be cleaned and regenerated by washing with methanol, isopropanol, and methylene chloride and reused many times. 3. PREPARATIVE NORMAL-PHASE HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY

High-performance liquid chromatography can be performed on any available HPLC system. This normal-phase HPLC step is performed in our laboratory using a 0.4 × 25 cm Zorbax-Sil column packed with 5 µm silica, but any equivalent column could be utilized. The mobile phase comprises hexane-methylene chloride-isopropanol (49.5 : 49.5 : 0.5, v/v) at a flow rate of 2 ml/min. The sample residue from the silica cartridge is dissolved in 150 µl of the mobile phase and injected onto the HPLC column that had been previously calibrated with 10 ng of standard vitamin D3. The elution of vitamins D2 and D3 (they coelute on this system) can be seen in Fig. 4A. The vitamin D fraction is collected in a 12 × 75 mm glass culture tube and dried under N2 at 55°C.

4. QUANTITATIVE REVERSED-PHASE HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY

The final quantitative step is performed using nonaqueous reversed-phase HPLC. The column is a Vydac TP-201-54, 5 µm, wide pore, nonendcapped octadecylsilane (ODS) silica material, 0.4 × 25 cm. This particular column must be used for this procedure to work. The mobile phase comprises acetonitrile-methylene chloride (65:35, v/v) utilizing a flow rate of 1.2 ml/min. This system provides clear resolution of vitamins D2 and D3 (Fig. 4B). This system is calibrated with varying amounts of vitamin D2 and D3 (1–100 ng). The sample residue from normal-phase HPLC is dissolved in 15 µl methylene chloride followed by the addition of 135 µl of acetonitrile and injected onto the HPLC. After elution, final quantitation of vitamin D2 and D3 is by direct UV monitoring of 265 nm. The vitamin D3 portion is collected, dried under N2, and subjected to liquid scintillation counting in order to determine the final recovery of endogenous vitamin D3 from the sample. Calculations are then performed, and the results are reported in as nanograms vitamin D2 and/or D3 per milliliter. A flow diagram of the entire procedure is displayed in Fig. 5.

A

Hexane: ISP

CH2Cl2:ISP 95:5

99.8:0.2

92:8

0.004

D2 + D3

85:15

98.5:1.5

500

3H-1,25-(OH)

0.003

2-D3

300 0.002

Radioactivity (cpm)

500

3H-24,25-(OH)

Optical density 265 nm

100

2-D3

300 100

0.001

0

B

D2 D3

0.004

500 3H-25-OH-D

3

0.003

300 0.002

100

500

3H-Vitamin

0.001

D3

300 0 2

100 0

4

8

12

16

20

24

28

32

36

Elution volume (ml)

FIGURE 3 Elution profiles of radioactive vitamin D3 and its metabolites chromatographed on a silica Bond-Elut (500 mg) cartridge.

FIGURE 4

4

6 8 10 Elution time (min)

12

14

High-performance liquid chromatographic profiles of standard vitamins D2 and D3 on normal-phase (A) and reversedphase (B) systems. Column calibration was achieved by injecting 10 ng of each compound and monitoring optical density at 265 nm. Column conditions are described in the text.

935

CHAPTER 58 Detection of Vitamin D and Its Major Metabolites

0.5 ml Sample or control + 3H-Vitamin D3 Incubate 15 min room temp Extract with methanol : hexane

Centrifuge and remove hexane layer Dry hexane under N2 and resuspend in methylene chloride Apply to silica cartridge and collect vitamin D-containing fraction Dry fraction under N2 and resuspend in normal-phase HPLC mobile phase Apply to normal-phase silica HPLC and collect vitamin D-containing fraction Dry fraction under N2 and resuspend in methylene chloride:acetonitrile Apply to nonaqueous reversed-phase Vydac ODS HPLC and quantitate vitamins D2 and D3 by direct UV absorption. Collect vitamin D3-containing fraction and monitor for 3H-content recovery correction

FIGURE 5

Flow diagram of the HPLC-UV assay for the quantitation of vitamins D2 and D3.

III. DETECTION OF 25OHD A. Background One of the major factors responsible for the explosion of knowledge related to vitamin D metabolism was the introduction of valid CPBA for 25OHD in the early 1970s [6,7] (Table II). One of these assays in particular gained widespread use owing to its relative simplicity and, as a result, has been cited nearly 1000 times [6]. The first assays utilized the vitamin D-binding

TABLE II Detection method CPBA CPBA CPBA HPLC RIA RIA CLIA ang/ml

Significant Methods for the Estimation of 25OHD in Human Serum

Extraction

Preliminary chromatography

Methanol-chloroform Ether Ethanol Methanol-chloroform Acetonitrile Acetonitrile None

Silicic acid Silicic acid None Sephadex LH-20 None None None

× 2.4 = nmol/liter.

protein (DBP) from rat serum as a specific binding agent. These assays all contained some type of organic extraction coupled with sample prepurification by column chromatography, all used 3H-25OHD3 as a tracer, and all required individual sample recovery estimates to account for endogenous losses of 25OHD during the extraction and purification procedures. A nonchromatographic assay for circulating 25OHD was introduced in 1974 [22], but it was never widely accepted because of its nonspecificity and susceptibility to serum lipid interference. Various CPBAs for 25OHD dominated the literature until 1977 when the first valid direct UV quantitative HPLC assay was introduced [11]. 25OHD circulates in the nanogram per milliliter (nanomole/liter) range and thus could be directly quantitated by UV detection following its separation by normal-phase HPLC. Also, HPLC detection provided the advantage of being able to individually quantitate 25OHD2 and 25OHD3. The disadvantages of HPLC quantitation methods are their requirements for expensive equipment and large sample size, cumbersome nature, and the technical expertise to perform this type of analysis. However, HPLC analysis for 25OHD is frequently used in research environments, including our own, and has provided a great deal of significant information. As the clinical demand for circulating 25OHD analysis increased, it was clear that simpler, rapid yet valid assay procedures would be required. To this point, all valid assays required liquid-liquid organic extraction, some sort of chromatographic prepurification, and evaporation of the organic solvents, hardly practical for a clinical chemistry laboratory. Thus, in 1985, the first valid RIA for assessing circulating 25OHD was introduced [13]. This RIA eliminated the need for sample prepurification prior to assay and had no requirement for organic solvent evaporation. However, the method was still based on the use of 3H-25OHD3 as a tracer. This final shortcoming was solved in 1993 when an

Ref. Belsey et al. [7] Hadad and Chyu [6] Belsey et al. [22] Eisman et al. [11] Hollis and Napoli [13] Hollis et al. [14] DiaSorin Corp.

Normal circulating levelsa 18–36 ng/ml 27.3 ± 11.8 ng/ml 20–100 ng/ml 31.9 ± 1.7 ng/ml 25.5 ± 11.8 ng/ml 9.9–41.5 ng/ml 9.5–52.0 ng/ml

936 125I-labeled tracer was developed and incorporated into the RIA for 25OHD [14]. This assay has become the method of choice for assessing 25OHD status and has become the first test for vitamin D approved for clinical diagnosis by the U.S. Food and Drug Administration (FDA). Most recently, DiaSorin Corporation (Stillwater, MN) has introduced an automated, nonextracted chemiluminescent immunoassay (CLIA) for the direct determination of circulating 25(OH)D.

B. HPLC Methodology 1. SAMPLE EXTRACTION

A 0.5-ml sample of serum or plasma is placed into a 12 × 75 mm borosilicate glass culture tube containing 1000 cpm of 3H-25OHD3 in 25 µ of ethanol to monitor recovery of endogenous compound through the extraction and chromatographic procedures. Following a 15-min incubation with the tracer, 1 plasma volume of HPLC-grade acetonitrile is added to each sample. The sample is then vortex-mixed for 1 min followed by centrifugation at 1000 g for 10 min. The supernatant is removed into another 12 × 75 mm culture tube, and 1 plasma volume of 0.4M K2HPO4, pH 10.4, is added. 2. SOLID-PHASE EXTRACTION CHROMATOGRAPHY

C18 silica Sep-Pak cartridges (500 mg) and a SepPak rack were obtained from Waters Associates (Milford, MA). The C18 cartridges are washed in sequence with 5 ml HPLC-grade isopropanol and 5 ml HPLC-grade methanol. The sample is applied to the cartridge and eluted through the cartridge under vacuum into waste. This initial step is followed by 5 ml of 30% water in methanol (discard) and 3 ml of acetonitrile (25OHD). The acetonitrile fraction is dried in a heated water bath, 55°C, under N2. The lipid residue is then suspended in 1 ml of 1.5% isopropanol in hexane and capped. The C18 cartridges can be cleaned and regenerated by washing with 2 ml of methanol and reused many times. 3. SILICA CARTRIDGE CHROMATOGRAPHY

Silica Bond-Elut cartridges (500 mg) and a Vac-Elut cartridge rack were obtained from Varian Instruments. The silica cartridges are washed in order with 5 ml HPLC-grade methanol, 5 ml HPLC-grade isopropanol, and 5 ml HPLC-grade hexane. The sample, in 1 ml of 1.5% isopropanol in hexane, is then applied to the cartridge and eluted through the cartridge under vacuum into waste. This initial elution is followed by 4 ml of 1.5% isopropanol in hexane (discard) and 6 ml of 5% isopropanol in hexane (25OHD). The 6-ml fraction contains 25OHD2 and 25OHD3 and is subsequently dried in

BRUCE W. HOLLIS

a heated water bath, 55°C, under N2. The elution profile of 25OHD3 from the silica cartridge is displayed in Fig. 3. The silica cartridges can be cleaned and reused many times. 4. QUANTITATIVE NORMAL-PHASE HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY

The final quantitative step is performed using normalphase HPLC with a 0.4 × 25 cm Zorbax-Sil column packed with 5 µm spherical silica. The mobile phase is composed of hexane-methylene chloride-isopropanol (50 : 50 : 2.5, v/v) at a flow rate of 2 ml/min. The sample residue from the silica cartridge is dissolved in 150 µl of mobile phase and injected onto the HPLC column previously calibrated with varying amounts of 25OHD2 and 25OHD3 (1–100 ng). This HPLC system provides clear resolution of 25OHD2 and 25OHD3 (Fig. 6A). After elution, final quantitation of 25OHD2 and 25OHD3 is by direct UV monitoring at 265 nm. The 25OHD3 portion is collected, dried under N2, and subjected to liquid scintillation counting in order to determine the final endogenous recovery of 25OHD2 and 25OHD3 from the sample. Calculations are then performed, and the results reported as nanograms 25OHD2 and/or 25OHD3 per milliliter. A flow diagram of the entire procedure is displayed in Fig. 7.

C. RIA Methodology 1. PREPARATION OF ASSAY CALIBRATORS

One of the goals of the RIA procedure for 25OHD was to eliminate the need for individual sample recovery. Another goal was to obtain FDA approval for clinical use of this procedure in the United States. Both of these goals place 25OHD3 in a human serum-based set of assay calibrators. To prepare these calibrators, human serum was “stripped” free of vitamin D metabolites by treatment with activated charcoal. Absence of endogenous 25OHD in the stripped sera was confirmed by direct UV detection of 25OHD in serum following HPLC as described in Section III,B. Subsequently, crystalline 25OHD3 dissolved in absolute ethanol was added to the stripped sera to yield calibrators at concentrations of 0, 5, 12, 40, 100 ng/ml. 2. SAMPLE AND CALIBRATOR EXTRACTION

To extract 25OHD from calibrators and samples, 0.5 ml of acetonitrile is placed into a 12 × 75 mm borosilicate glass tube after which 50 µl of sample or calibrator is dropped through the acetonitrile. After vortex-mixing, the tubes are centrifuged (1000 g, 4°C, 5 min) and 25 µl of supernatant transferred to 12 × 75 mm borosilicate glass tubes and placed on ice.

937

CHAPTER 58 Detection of Vitamin D and Its Major Metabolites

A

0.004 25OHD2

25OHD3

0.5 ml Sample or control + 3H-25OHD3

0.003

Incubate 15 min room temp 0.002

Solid-phase extract using C18 Sep-Pak 0.001

Dry acetonitrile under N2 and resuspend in methylene chloride

0

Optical density 265 nm

B

0.004

Apply to silica cartridge and collect 25OHD-containing fraction

0.003

24,25(OH)2D2

24,25(OH)2D2

Dry fraction under N2 and resuspend in normal-phase HPLC mobile phase

0.002

0.001

Apply to normal-phase silica HPLC and quantitate 25OHD2 and 25OHD3 by direct UV absorption. Collect 25OHD3 containing fraction and monitor for 3H-content for recovery correction

0

C

0.004

FIGURE 7 0.003

1,25(OH)2D2 1,25(OH)2D3

0.002

0.001

0 2

4

6

8

10

12

14

Elution time (min)

FIGURE 6

High-performance liquid chromatographic profiles of standard vitamin D metabolites. (A) 25OHD2 and 25OHD3; (B) 24,25(OH)2D2 and 24,25(OH)2D3; (C) 1,25(OH)2D2 and 1,25(OH)2D3. The HPLC was performed on a normal-phase Zorbax-Sil column, and column calibration was achieved by injecting 10 ng of each compound and monitoring optical density at 265 nm. Column conditions are described in the text.

3. RADIOIMMUNOASSAY

The assay tubes are 12 × 75 mm borosilicate glass tubes containing 25 µl of acetonitrile-extracted calibrators or samples. To each tube add 125I-25OHD derivative (50,000 cpm in 50 µl 1:1 ethanol-10 mM phosphate buffer, pH 7.4) that was synthesized as previously described [14]. Then added to each tube 1.0 ml of primary antibody diluted 1:15,000 in sodium phosphate

Flow diagram of the HPLC-UV assay for the quantitation of 25OHD2 and 25OHD3.

buffer (50 mM, pH 7.4, containing 0.1% swine skin gelatin). Nonspecific binding is estimated using the above buffer minus the antibody. Vortex-mix the contents of the tubes and incubate them for 90 min at 20–25°C. Following this period, add 0.5 ml of a second antibody precipitating complex to each tube, vortex-mix, incubate at 20–25°C for 20 min, and centrifuge (20°C, 2000 g, 20 min). Discard the supernatant and bound the tubes in a gamma well counting system. 25OHD values are calculated directly from the standard curve by the counting system using a smooth-spline method of calculation. The entire 25OHD RIA procedure is displayed in Fig. 8. 4. COMMENTS ON THE 25OHD RIA

This 125I-based RIA is similar to an RIA we introduced several years ago that used 3H-25OHD3 as a tracer [13]. In both cases, antisera were raised against the synthetic vitamin D analog 23,24,25,26,27-pentanor vitamin D-C(22)-carboxylic acid. The syntheses of this analog and its 125I-labeled counterpart have been described in detail [13,14]. Coupling this compound to bovine serum albumin allowed us to generate antibodies that cross-reacted equally with most vitamin D2 and D3

938

BRUCE W. HOLLIS

TABLE III Cross-reactivity of Various Vitamin D Compounds with 25OHD Antiserum and 125I-labeled Vitamin D Derivativea

50 µl Sample, standard or control

500 µl Acetonitrile, 10 min spin

25 µl Extract + 50 µl tracer + 1.0 ml primary antibody

90 min Incubation at room temperature

+ 0.5 ml Precipitating complex

Cross-reactivity (%)b

Steroid Vitamin D2 Vitamin D3 DHT 25(OH)D2 25(OH)D3 25(OH)D3-26,23-lactone 24,25(OH)2D2 24,25(OH)2D3 25,26(OH)2D2 25,26(OH)2D3 1,25(OH)2D2 1,25(OH)2D3

0.8 0.8 < 0.1 100 100 100 100 100 100 100 2.5 2.5

aFrom

Hollis et al. Clin Chem 39:529-533. to displace 50% of the 125I tracer from the 25(OH)D antiserum diluted 15,000-fold. bAbility

20 min Incubation at room temperature 20 min spin

Decant and count

FIGURE 8

a variety of human serum samples (Fig. 9). Further, the present 125I-based RIA was shown to identify vitamin D deficiency in biliary atresia patients as well as vitamin D toxicity in hypoparathyroid patients who were receiving massive vitamin D therapy for the maintenance of plasma calcium (Table IV).

Flow diagram of the direct RIA or the quantitation of

25OHD.

RIA 25OHD determination (ng/ml)

metabolites (Table III). The structures for vitamins D2 and D3 differ only with respect to their side chains (Fig. 1). Because the analog retained the intact structure of vitamin D only up to carbon 22, the structural differences between vitamins D2 and D3 were not involved in the antibody recognition, and antibodies directed against this analog could not discriminate with respect to side chain metabolism of vitamin D. The antibody, however, was specific for the open B-ring cis-triene structure containing a 3β-hydroxyl group that is inherent in all vitamin D compounds. Many vitamin D metabolites other than 25OHD are present in the circulation; however, they contribute only a small percentage (6–7%) to the overall assessment of nutritional vitamin D status as compared with 25OHD [23]. This fact is supported by the comparison of the 25OHD RIA with the UV quantitative HPLC assay for 25OHD described earlier in Section III,B on

RIA = 0.98(HPLC) + 0.01 r2 = 0.98 n = 63

200

150

100

50

0 0

50 100 150 HPLC 25OHD determination (ng/ml)

200

FIGURE 9 25OHD values obtained by the 25OHD RIA (y axis) and by direct UV quantitation of 25OHD following HPLC (x axis).

939

Subject type

n

Normalc Biliary atresia Vitamin D therapyd

36 12 8

Mean

(ng/ml)b

25.7 6.3 145

Range

(ng/ml)b

9.9–41.5 4.3–8.3 92–202

aFrom

Hollis et al. Clin Chem 39:529–533. × 2.496 = nmol/liter. cSamples from subjects in Minnesota in October. dSamples from subjects with hypoparathyroidism or pseudohypoparathyroidism receiving pharmacological doses of vitamin D2. bng/ml

D. Automated Instrumentation CLIA Methodology DiaSorin Corporation (Stillwater, MN) has introduced a method for the direct (no extraction) quantitative determination of 25OHD in serum or plasma utilizing competitive chemiluminescence immunoassay (CLIA). This assay is offered in an automated format using the companies’ LIAISON® platform. Much of the methodology is proprietary, although the assay utilizes a specific antibody to 25OHD that is coated onto magnetic particles (solid phase). The tracer vitamin D is linked to an isoluminol derivative. During the incubation of the sample, 25OHD is dissociated from its binding protein, and competes with the labeled vitamin D for binding sites on the antibody. After the incubation, the unbound material is removed with a wash cycle. Subsequently, the starter reagents are added and a flash chemiluminescent reaction is initiated. The light signal is measured by a photomultiplier as relative light units (RLU) and is inversely proportional to the concentration of 25OHD present in calibrators, controls, and samples. This procedure will assay up to 180 samples/hr with excellent linearity, precision, and sensitivity. Further, this automated assay is in excellent agreement with the widely used RIA (14) in both normal and chronic renal failure patients (Figs. 10 and 11). This assay format would be the method of choice for large throughput clinical laboratories. A similar assay platform has also been introduced by Nichols Institute Diagnostics (San Clemente, CA). This platform is the Nichols ADVANTAGE® system. This instrument is similar to the DiaSorin LIAISON system; however, the assay is very different. The Nichols ADVANTAGE 25OHD assay utilizes the human serum vitamin D–binding protein as a competitive binder instead of an antibody such as that used in the DiaSorin system. The use of the serum binding

160.0 y = 0.98x − 1.5 r = 0.77

140.0 120.0 100.0 80.0 60.0 40.0 20.0 0.0 0.0

20.0

40.0 60.0 80.0 25 OH Vitamin D (RIA) ng/ml

100.0

FIGURE 10

Linear regression analysis of 200 normal clinical samples by 125I-RIA and LIAISON®

protein results in the Nichols assay being less compliant when compared with the standard RIA (Fig. 12). This result is not unexpected since the vitamin D– binding protein is susceptible to nonspecific interferences from unknown substances in serum or plasma [24]. The original version of this method encountered the same problems 30 years ago and was discarded as an invalid procedure [22]. Further, the Nichols system is less sensitive and much less efficient than the comparable DiaSorin system. A direct comparison of the systems is displayed on Table V.

IV. ADDENDUM Recently, two articles have appeared that demonstrate disturbing deficiencies in the Nichols Advantage 25(OH)D assay system. Binkley, et al. [24a] has shown the Advantage assay to perform poorly on clinical samples and advised against its use in clinical settings. A more disturbing deficiency in the Advantage assay is its total inability to detect 25(OH)2 in clinical samples [24b]. This is in direct conflict with the manufacturer’s claim

40.0 Liaison (ng/ml)

TABLE IV Concentration of 25OHD as Determined by 125I-RIA in Various Physiological Statesa

25 OH Vitamin D (Liaison) ng/ml

CHAPTER 58 Detection of Vitamin D and Its Major Metabolites

30.0

y = 0.91x + 0.6 r = 0.82

20.0 10.0 0.0 0.0

10.0

20.0 RIA (ng/ml)

30.0

40.0

FIGURE 11 Linear regression analysis of 118 chronic renal failure samples by 125I-RIA and LIAISON®

25 OH Vitamin D (Advantage) ng/ml

940

BRUCE W. HOLLIS

160.0 140.0

V. DETECTION OF 24,25(OH2)D

y = 1.07x + 11.0 r = 0.59

A. Background

120.0 100.0 80.0 60.0 40.0 20.0 0.0 0.0

20.0 40.0 60.0 80.0 25 OH Vitamin D (RIA) ng/ml

100.0

FIGURE 12

Linear regression analysis of 200 normal clinical samples by 125I-RIA and Advantage®

(Table V) that the assay can identify 25(OH)D2 at 100% efficiency. The seriousness of this problem cannot be overstated in that the Advantage assay system will fail to monitor vitamin D therapy involving Drisdol (vitamin D2) supplementation.

Next to 25OHD, 24,25(OH)2D is quantitatively the most abundant circulating vitamin D metabolite, and, as a result, interest in its circulating levels has persisted. However, to this day, the biological function(s) of 24,25(OH)2D3, if any, remains unresolved. The first assay for 24,25(OH)2D was first reported in 1977 and used CPBA in conjunction with sample prepurification on Sephadex LH-20 [8] (Table VI). However, it was soon discovered that more extensive sample prepurification was required prior to 24,25(OH)2D quantitation owing to substances that interfered in the 24,25(OH)2D CPBA [25]. To further complicate matters, the quantitation of 24,25(OH)2D is especially difficult when both the vitamin D2 and D3 forms are present in the circulation [16]. When both forms of 24,25(OH)2D are present, it is extremely hard to remove other vitamin D metabolites that coelute with 24,25(OH)2D2 and 24,25(OH)2D3 on HPLC

TABLE V Direct Comparison of Liaison® and Advantage® 25OHD Automated Assay Systems Parameter Time to First Result Maximum Throughput Lowest Reportable Value Sensitivity (Analytical) Mean % Recovery Intra-Assay Precision (% CV at 3 concentrations) Inter-Assay Precision (% CV at 3 concentrations) Dilution Linearity Sample Equivalence (Serum vs. Plasma) Cross-Reactivity

Carry-Over Reference Range (2.5th to 97.5th %)

LIAISON® 35 minutes 180 tests/hr 2.6 ng/ml (6.5 nmol/L) < 2.0 ng/ml 105% ± 14% 13.6 ng/ml 10% 24.4 ng/ml 8% 50.7 ng/ml 6% 13.6 ng/ml 19% 24.4 ng/ml 17% 50.7 ng/ml 13% Observed = Expected (1.04) − 0.4; r = 0.95 Plasma = Serum (1.1) – 2.4; r = 0.93 D2 0.0% D3 0.0% 25D2 100.0% 25D3 100.0% 1,25D2 7.1% 1,25D3 21.7% None Median = 25.5 ng/ml Range = 8.6–54.8 ng/ml

ADVANTAGE® 75 minutes 75 tests/hr 7.0 ng/ml (17.5 nmol/l) < 2.0 ng/ml 94% ± 6% 13.3 ng/ml 7% 26.6 ng/ml 4% 55.5 ng/ml 2% 13.3 ng/ml 20% 26.6 ng/ml 16% 55.5 ng/ml 14% Observed = Expected (0.95) + 3.7; r = 0.99 Plasma = Serum (0.96) + 1.3; r = 0.95 D2 0.0% D3 4.5% 25D2 100.0% 25D3 100.0% 1,25D2 0.0% 1,25D3 2.2% None Median = 45.4 ng/ml Range = 20.4–90.2 ng/ml

941

CHAPTER 58 Detection of Vitamin D and Its Major Metabolites

TABLE VI Detection method

Significant Methods for the Estimation of 24,25(OH)2D in Human Serum

Extraction

Preliminary chromatography

CPBA CPBA HPLC RIA

Methanol-methylene chloride Methanol-methylene chloride Methanol-methylene chloride Methanol-methylene chloride

Sephadex LH-20 Sephadex LH-20, preparative HPLC Sephadex LH-20, preparative HPLC Sephadex LH-20, preparative HPLC

CPBA

Solid phase C18OH

Silica cartridges

ang/ml

Ref. Haddad et al. [8] Shepard et al. [25] Dreyer and Goodman [26] Hummer and Christiansen [27] Wei et al. [28]

Normal circulating levelsb 3.7 ± 0.2 ng/ml 3.5 ± 1.4 ng/ml 2.4 ± 1.1 ng/ml 0.1–4.0 ng/ml 3.1 ± 0.7 ng/ml

× 2.4 = nmol/liter.

separation [16]. Further, once 24,25(OH)2D2 and 24,25(OH)2D3 are adequately separated and ready for CPBA, varying affinities of the two metabolites for the DBP require standard curves to be constructed for final quantitation [29]. A report published in 1994 questions the requirement of HPLC prepurification of the serum sample prior to CPBA [28]. However, we are firm believers that in order to perform a valid assay for 24,25(OH)2D, one has to incorporate HPLC prepurification into the assay protocol. We have also chosen to do the final quantitation of 24,25(OH)2D by RIA instead of CPBA. The RIA was chosen because the antibody used is cospecific for the vitamin D2 and D3 forms, and thus only one compound, 24,25(OH)2D3, is required to construct the standard curve (Table III). This procedure is described here in detail.

B. Methodology 1. SAMPLE EXTRACTION

A 0.5-ml sample of serum or plasma is placed into a 12 × 75 mm borosilicate glass culture tube containing 1000 cpm of 3H-24,25(OH)2D3 in 25 µl of ethanol to monitor recovery of endogenous compound through the extraction and chromatographic procedures. Following a 15-min incubation with the tracer, 1 plasma volume of HPLC-grade acetonitrile is added to each sample. The sample is then vortex-mixed for 1 min followed by centrifugation at 1000 g for 10 min. The supernatant is removed into another 12 × 75 mm culture tube, and 1 plasma volume of distilled water is added. 2. SOLID-PHASE EXTRACTION CHROMATOGRAPHY

C18 silica Bond-Elut cartridges (500 mg) and a Vac-Elut cartridge rack were obtained from Varian Instruments. The C18 cartridges are washed in order with 5 ml HPLC-grade isopropanol and 5 ml HPLC-grade methanol. The sample is then applied to the cartridge

and eluted through the cartridge under vacuum into waste. This initial step is followed by 5 ml of 40% water in methanol (discard), 5 ml of 1% methylene chloride in hexane (discard), and 5 ml of 5% isopropanol in hexane [24,25(OH)2D]. The final fraction is dried in a heated water bath, 55°C, under N2. The lipid residue is then resuspended in 150 µl 5% isopropanol in hexane and capped. 3. PREPARATIVE NORMAL-PHASE HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY

The normal-phase HPLC step is performed using a 0.4 × 25 cm Zorbax-Sil column packed with 5 µm silica. The mobile phase is composed of hexane-methylene chloride-isopropanol (80:15:3.5, v/v) at a flow rate of 2 ml/min. The sample residue from the C18 silica cartridge is injected onto the HPLC column that had been previously calibrated with 10 ng of 24,25(OH)2D2 and 24,25(OH)2D3. The elution of these metabolites can be seen in Fig. 6B. The fractions containing 24,25(OH)2D2 and 24,25(OH)2D3 are collected individually in 12 × 75 mm glass tubes and dried under N2 at 55°C. The residue is then redissolved in 500 µl absolute ethanol and capped. 4. RADIOIMMUNOASSAY

The assay tubes are 12 × 75 mm borosilicate glass tubes containing 25 µl of the HPLC-purified extracts in ethanol. The standards for this RIA, which are 24,25(OH)2D3 are placed in 12 × 75 mm tubes in 25 µl ethanol at concentrations between 0 and 200 pg/tube. To each tube add 125I-25OHD derivative (50,000 cpm in 50 µl 1:1 ethanol-10 mM phosphate buffer, pH 7.4) or 3H-25OHD3 (5000 cpm in 25 µl ethanol). Then add to each tube 1.0 ml of primary antibody diluted 1:15,000 in sodium phosphate buffer (50 mM, pH 7.4, containing 0.1% swine skin gelatin). Nonspecific binding is estimated using the above buffer minus the antibody. Vortex-mix the contents of the tubes and incubate them for 90 min at 20–25°C. Following this

942

BRUCE W. HOLLIS

0.5 ml Sample or control + 3H-24,25(OH)2D3 Incubate 15 min room temp

Isolate 24,25(OH)2D by simultaneous extraction and purification using C18support in "phase-switching" mode Dry organics under N2 and resuspend in normalphase HPLC mobile phase Apply to normal-phase silica HPLC and individually collect 24,25(OH)2D3 and 24,25(OH)2D2 Remove a portion for recovery estimation

Dry organics under N2, and resuspend each fraction in ethanol

Quantitate 24,25(OH)2D2 and 24,25(OH)2D3 individually by RIA using antibody cospecific for each compound. 3H or 125I tracers are available for use

FIGURE 13 Flow diagram of the HPLC-RIA assay for the quantitation of 24,25(OH)2D2 and 24,25(OH)2D3.

period, add 0.5 ml of a second antibody precipitating complex to each tube if 125I tracer was used, 0.2 ml of 0.1 M borate buffer containing 1.0% norit A charcoal and 0.1% dextran T-70 if 3H-25(OH)D3 was used, vortex-mix, incubate at 20–25°C for 20 min and centrifuge (20°C, 2000 g, 20 min). In the case of 125I tracer, discard the supernatant and count the tubes in a gamma well counting system. In the case of 3H-25OHD3, remove the supernatant into vials, add scintillation fluid, and monitor for radioactive content in a scintillation counter. 24,25(OH)2D values are calculated from the standard curve in picograms per tube. To convert this value to nanograms per milliliter, correct for dilution used as well as final recovery of 3H-24,25(OH)2D3 added at the beginning of the sample extraction procedure. The entire 24,25(OH)2D RIA procedure is displayed in Fig. 13.

VI. DETECTION OF 1,25(OH)2D A. Background Of all the steroid hormones, 1,25(OH)2D represented the most difficult challenge to the analytical biochemist

with respect to quantitation. 1,25(OH)2D circulates at low picogram per milliliter concentrations (too low for direct UV quantitation), is highly lipophilic, and is relatively unstable, and its precursor, 25OHD, circulates at concentrations in excess of 103 to 104 times that of 1,25(OH)2D. The first RRA for 1,25(OH)2D was introduced in 1974 [9] (Table VII). Although this initial assay was extremely cumbersome, it did provide invaluable information with respect to vitamin D homeostasis. This initial RRA required a 20-ml serum sample, which was extracted using Bligh-Dyer organic extraction. The extract had to be purified by three successive laborious chromatographic systems (there was no HPLC at the time), and chickens had to be sacrificed and vitamin D receptor (VDR) harvested from their intestines at the time of the RRA. By 1976, the volume requirement for this RRA had been reduced to a 5-ml sample and sample prepurification had been modified to include HPLC [30]. However, the sample still had to be extracted using a modified Bligh-Dyer extraction, and then prepurified on Sephadex LH-20, and chicken intestinal VDR was still utilized as a binding agent. In 1978, the first RIA for 1,25(OH)2D was introduced [12]. Although it was an advantage not to have to isolate the intestinal VDR as a binding agent, this RIA was relatively nonspecific, so the cumbersome sample preparative steps were still required. Because of the extreme technical nature of these assays, and the cost of HPLC systems, few laboratories could afford to measure circulating 1,25(OH)2D. Further, because these early techniques were so cumbersome, commercial laboratories did not offer 1,25(OH)2D determinations as a clinical service. This all changed in 1984 with the introduction of a radically new concept for the determination of circulating 1,25(OH)2D [31]. This new RRA utilized solid-phase extraction of 1,25(OH)2D from serum along with silica cartridge purification of 1,25(OH)2D. As a result, the need for HPLC sample prepurification was eliminated. Also, this assay utilized VDR isolated from calf thymus, which proved to be quite stable and thus had to be prepared only periodically. Further, the volume requirement was reduced to 1 ml of serum or plasma. This assay opened the way for any laboratory to measure circulating 1,25(OH)2D. This procedure also resulted in the production of the first commercial kit for 1,25(OH)2D measurement. This RRA was further simplified in 1986 by decreasing the required chromatographic purification steps [32]. Through the mid-1990s, no new advances were reported with respect to the quantitation of circulating 1,25(OH)2D. As good as the calf thymus RRA for 1,25(OH)2D was, it still possessed two serious shortcomings. First, VDR had to be isolated from thymus glands, which

943

CHAPTER 58 Detection of Vitamin D and Its Major Metabolites

TABLE VII Detection method RRA RRA RIA RRA RRA RIA apg/ml

Significant Methods for the Estimation of 1,25(OH)2D in Human Serum

Extraction Methanol-chloroform Methanol-methylene chloride Methanol-chloroform Solid phase C18OH Solid phase C18OH Solid phase C18OH/silica

Preliminary chromatography Silicic acid, Sephadex LH-20, celite Sephadex LH-20, preparative HPLC Sephadex LH-20, preparative HPLC Silica cartridge None None

Ref. Brumbaugh et al. [9] Eisman et al. [30] Clemens et al. [12] Reinhardt et al. [31] Hollis [32] Hollis et al. [15]

Normal circulating levelsb 39 ± 8 pg/ml 29 ± 2 pg/ml 35 pg/ml 37.4 ± 2.2 pg/ml 28.2 ± 11.3 pg/ml 32.2 ± 8.5 pg/ml

× 2.4 = pmol/liter

was still a difficult technique. Second, because the VDR is so specific for its ligand, only 3H-1,25(OH)2D3 could be used as a tracer, eliminating the possibility of using a 125I-based tracer. This is a major handicap, especially for the commercial laboratory. As a result, we have developed and reported in 1996 the first significant advance in 1,25(OH)2D quantification in a decade [15]. This new RIA incorporates an 125I-tracer, as well as standards in an equivalent serum matrix, so individual sample recoveries are no longer required. We describe this new RIA for 1,25(OH)2D along with the standard RRA.

B. RRA Methodology 1. PREPARATION OF CALF THYMUS VDR

Frozen or fresh tissue is processed for VDR as follows (all steps are carried out at 4°C). Thymus tissue is minced with a meat grinder and homogenized (20% w/v) in a buffer containing 50 mM K2HPO4, 5 mM dithiothreitol, 1 mM EDTA, and 400 mM KCl, pH 7.5. The tissue is homogenized using five, 30-sec bursts of a Polytron PT-20 tissue disrupter at a maximum power setting. The homogenate is then centrifuged for 15 min at 20,000 g to remove large particles. The resulting supernatant is centrifuged at 100,000 g for 1 hr, and the “cytosol” (actually a high salt extract that includes nucleus VDR) is collected minus the floating lipid layer. The VDR is then precipitated by the slow addition of solid (NH4)2SO4 to 35% saturation. The cytosol(NH4)2SO4 mixture is stirred for 30 min while maintaining the temperature at 4°C. The mixture is then divided into 15-ml centrifuge tubes and centrifuged at 20,000 g for 20 min. The supernatant is discarded, and tubes are allowed to drain for 5 min. The precipitated VDR is lyophilized and stored under inert gas at −70°C. VDR prepared in this manner is stable for up to 60 hr at room temperature.

2. SAMPLE EXTRACTION

A 1.0 ml sample of serum or plasma is placed into a 12 × 75 mm borosilicate glass culture tube containing 700 cpm of 3H-1,25(OH)2D3 in 25 µl of ethanol to monitor recovery of endogenous compound through the extraction and chromatographic procedures. Following a 15-min incubation with the tracer, 1 ml of HPLC-grade acetonitrile is added to each sample. The sample is then vortex-mixed for 1 min, followed by centrifugation at 1000 g for 10 min. The supernatant is removed into another 12 × 75 mm culture tube, and 1 vol of 0.4 M K2HPO4, pH 10.4, is added followed by vortex-mixing. 3. SOLID-PHASE EXTRACTION AND PURIFICATION CHROMATOGRAPHY

C18OH silica Bond-Elut cartridges (500 mg) and a Vac-Elut cartridge rack were obtained from Varian Instruments. The C18OH cartridges are washed in order with 5 ml HPLC-grade methylene chloride, 5 ml HPLCgrade isopropanol, and 5 ml HPLC-grade methanol. The sample is applied to the cartridge and eluted through the cartridge under vacuum into waste. This initial step is followed by 5 ml of 30% water in methanol (discard), 5 ml of 10% methylene chloride in hexane (discard), 5 ml of 1% isopropanol in hexane (discard), and 5 ml of 3% isopropanol in hexane [1,25(OH)2D] (Fig. 14). This final fraction is dried in a heated water bath, 55°C, under N2. The residue is then suspended in 200 µl absolute ethanol and capped. 4. RADIORECEPTOR ASSAY

Prior to assay, the VDR-containing pellet is reconstituted to its original volume with assay buffer. The assay buffer contains 50 mM K2HPO4, 5 mM dithiothreitol, 1.0 mM EDTA, and 150 mM KCl at pH 7.5. The receptor pellet is dissolved by gentle stirring on ice using a magnetic stir bar. The receptor solution is allowed to mix for 20–30 min. Typically, a small portion of the pellet resists solubilization and is removed by centrifugation

944

BRUCE W. HOLLIS

H2O

100%

100

3H-25(OH)D

Percent of total radioactivity

80 60

CH3OH:H2O

Hexane: CH2CI2

70:30

90:10

Hexane: Isopropanol

99:1

97:3

3

40 20 3H-24,25(OH)

100

2D3

80 60 40 20 100

3H-1,25(OH)

2D3

80 60 40 20 0

2

4

6

8

10 12 14 Elution volume (ml)

16

18

20

22

24

FIGURE 14

Elution of 3H-vitamin D3 and its metabolites from a C18OH Bond-Elut cartridge. From Hollis BW, Clin Chem 31:1815–1819.

at 3000 g for 10 min. The receptor solution is then diluted 1:3–1:9 with assay buffer and kept on ice until use. The correct dilution of receptor used in the assay is determined for each new batch of receptor. At the appropriate dilution for assay use, specific binding in the absence of unlabeled 1,25(OH)2D is 1600–2000 cpm; nonspecific binding is 200–300 cpm. These results assume a specific activity of 130 Ci/mmol for 3H-1,25(OH)2D3 and a 40% counting efficiency for tritium. The assay tubes are 12 × 75 mm borosilicate glass tubes containing 50 µl of C18OH-purified extracts in ethanol. The standards for the assay, 1,25(OH)2D3, are placed in 12 × 75 mm tubes in 50 µl ethanol at concentrations between 1 and 15 pg/tube. Nonspecific binding is estimated by adding 1 ng/tube of 1,25(OH)2D3. To each tube add 0.5 ml of reconstituted thymus cytosol, vortex-mix, and incubate for 1 hr at 15–20°C. Following this initial incubation, each tube receives 3H-1,25(OH) D (5000 cpm in 50 µl ethanol) and the 2 3 incubation proceeds for an additional 1 hr at 15–20°C. Finally, place the assay tubes in an ice bath and add 0.2 ml of 0.1 M borate buffer containing 1.0% norit A charcoal and 0.1% dextran T-70, vortex-mix, incubate 20 min, and centrifuge (4°C, 2000 g, 10 min). Remove the supernatant into vials, add scintillation fluid, and monitor for radioactive content in a scintillation counter. 1,25(OH)2D values are calculated from the standard curve in picograms per tube. To convert this

1.0 ml Sample or control + 3H-1,25(OH)2D3 1.0 ml Acetonitrile, 10 min spin Combine supernatant with 1 vol K2HPO4, pH 10.4

Isolate 1,25(OH)2D by simultaneous extraction and purification Dry organics under N2 and resuspend in ethanol 50 µl Extract + 500 µl thymus receptor preparation 50 µl for recovery estimation

1 hr Incubation at 15–20°C

+ 50 µl 3H-1,25(OH)2D3 1 hr Incubation at 15–20°C + 200 µl Dextran-charcoal solution

Decant and count

20 min Incubation at room temp + 10 min spin

FIGURE 15 Flow diagram of the RRA for the quantitation of 1,25(OH)2D.

945

CHAPTER 58 Detection of Vitamin D and Its Major Metabolites

value to picograms per milliliter, correct for dilution used, as well as final recovery of 3H-1,25(OH)2D3 added at the beginning of the sample extraction procedure. The 1,25(OH)2D RRA procedure is displayed in Fig. 15.

C. RIA Methodology

500 µl Sample, standard or control 500 µl Acetonitrile, 10 min spin

Combine supernatant with 1 vol K2PO4, pH 10.4

1. PREPARATION OF ASSAY CALIBRATORS

As was described for the 25OHD RIA (Section III,C), one of the goals of this RIA procedure was to eliminate the need for individual sample recovery. To prepare the assay calibrators, human serum was stripped free of vitamin D metabolites. The absence of endogenous 1,25(OH)2D in the stripped sera was confirmed by RRA for 1,25(OH)2D as previously described in Section V,B. Subsequently, crystalline 1,25(OH)2D3 dissolved in absolute ethanol was added to the stripped sera to yield calibrators at concentrations of 0, 5, 15, 30, 75, and 200 pg/ml. 2. SAMPLE AND CALIBRATOR EXTRACTION AND PRETREATMENT

The 1,25(OH)2D is extracted from calibrators and samples as follows. First, 0.5 ml of serum of plasma is placed into a 12 × 75 mm borosilicate glass culture tube; 0.5 ml of HPLC-grade acetonitrile is added and vortex-mixed for 1 min followed by centrifugation at 1000 g for 10 min. The supernatant is removed into another 12 × 75 mm culture tube, and 1 vol of 0.4 M K2HPO4, pH 10.4, is added followed by vortex-mixing. 3. SOLID-PHASE EXTRACTION AND SILICA PURIFICATION CHROMATOGRAPHY

C18OH silica “Extra Clean” cartridges (500 mg) and a Vac-Elut cartridge rack were obtained from DiaSorin Corp and Varian Instruments, respectively. The C18OHEC cartridges are washed in order with 5 ml HPCL-grade methylene chloride, 5 ml HPLC-grade isopropanol, and 5 ml HPLC-grade methanol. The sample is applied to the cartridge and eluted through the cartridge under vacuum into waste. This initial step is followed by 5 ml of 30% water in methanol (discard), 5 ml of 10% methylene chloride in hexane (discard), 3 ml of 1% isopropanol in hexane (discard), and 3 ml of 8% isopropanol in hexane [1,25(OH)2D]. This final fraction is dried in a heated water bath, 55°C under N2 or in a rapid vacuum device such as a Labconco RapidVap. The residue is first reconstituted with 50 µl of 95% ethanol with vortex-mixing. Each tube now receives 125 µl of I125-tracer solution with additional vortex-mixing. The sample may be capped and stored at –20°C or one may proceed to finish the assay at this point.

Isolate 1,25(OH)2D by simultaneous extraction and purification

Dry organics under N2 and resuspend in ethanol

75 µl Extract/tracer + 300 µl primary antibody

2 hr Incubation at room temp + 500 µl Precipitating complex

20 min Incubation at room temp + 20 min spin Decant and count

FIGURE 16 Flow diagram of the RIA for the quantitation of 1,25(OH)2D.

4. RADIOIMMUNOASSAY

The assay tubes are 12 × 75 mm borosilicate glass tubes containing 75 µl of the ethanol/tracer-reconstituted extracted calibrators or samples. Then add to each tube 0.35 ml of primary antibody diluted 1:40,000 in sodium phosphate buffer [50 mM, pH 6.2, containing 0.1% swine-skin gelatin and 0.35% polyvinyl alcohol (Mr 13,000–23,000)]. Nonspecific binding is estimated by using the above buffer without the antibody. Vortexmix the contents of the tubes, incubate them for 2 hr at 20–25°C, add 0.5 ml of second antibody precipitating complex, incubate at 20–25°C for 20 minutes, and then centrifuge (20°C, 2000 g, 20 min). Discard the

946

5. COMMENTS ON THE 1,25(OH)2D RIA

Of the procedures developed for determining 1,25(OH)2D status in humans, only a few RRA [31,32] have been able to quantify circulating 1,25(OH)2D without using HPLC for sample prepurification. Many RIA have been published and validated for the quantification of 1,25(OH)2D, but all have included HPLC steps for sample prepurification [12,33,34]. Development of an RIA for quantification of circulating 1,25(OH)2D has been hampered from the beginning by the relatively poor specificity of the antibodies that have been generated. To date, the best antibodies toward 1,25(OH)2D have, at best, a cross-reactivity with the non-1-hydroxylated metabolites of vitamin D of approximately 1%. In comparison, the VDR used in the RRA has a crossreactivity of approximately 0.01% with these more abundant metabolites [31]. Given that the non-1hydroxylated metabolites circulate at concentrations over 1000 times greater than that of 1,25(OH)2D, the magnitude of the problem becomes clear. However, the VDR is so specific that any attempt to introduce a radionuclide such as 125I into 1,25(OH)2D3 erodes the binding between this steroid hormone and the VDR. Therefore, if one wishes to develop 125I-based assays to quantify 1,25(OH)2D, RIA is the only choice. Since the original paper on this assay was published [15], several improvements have been implemented. The major improvement involved the generation of a new antibody. This new antibody has greater sensitivity, specificity as compared to the original antibody [15]. Further, the new antibody expresses 100% crossreactivity between 1,25(OH)2D2 and 1,25(OH)2D3 (Table VIII). This new antibody has allowed the RIA to return to a single column format for sample purification along with the removal of a sample pretreatment step involving NaIO4. This RIA recently received FDA approval for clinical diagnosis in humans and thus far is the only test for 1,25(OH)2D to achieve this status. The concentrations of 1,25(OH)2D as determined in serum from various groups of healthy and pathological subjects (Fig. 17) agree well with values reported in previous studies [31,32]. It is very important to include pathological samples such as those from subjects with biliary atresia and vitamin D toxicity in any assay validation procedure for circulating 1,25(OH)2D. This importance was underlined in our previous report on an unpublished, unvalidated, but commercially available

TABLE VIII Cross-reactivity of Various Vitamin D Compounds with 1,25(OH)2D Antiserum and 125I-labeled 1-Hydroxylated Tracer Cross-reactivity (%)a

Steroid

< 0.001 0.002 0.012 0.003 100 100

Vitamin D3 25(OH)D3 24,25(OH)2D3 25,26(OH)2D3 1,25(OH)2D2 1,25(OH)2D3

aAbility to displace 50% of the 125I tracer from the 1,25(OH) D antiserum 2 diluted 1:40,000.

125I-based RIA for 1,25(OH) D that involves the 2 immunoextraction of 1,25(OH)2D from serum samples and is marketed by IDS Ltd. (Tyne and Wear, UK) [35]. The basis of this kit is selective immunoextraction of 1,25(OH)2D from serum or plasma with a specific monoclonal antibody bound to a solid support. This antibody is directed toward the lα-hydroxylated A ring of 1,25(OH)2D [36]. We concluded that this immunoextraction procedure was highly specific for the lα-hydroxylated forms of vitamin D [35]. However, there was a serious flaw in the assumptions made when this kit was designed: 1,25(OH)2D was the only significant lα-hydroxylated vitamin D metabolite that circulates. Many other lα-hydroxylated metabolites exist in the circulation, including 1,24,25(OH)3D3, 1,25,26(OH)3D3, l,25(OH)2D3-26,23-Lactone, 1,25(OH)2-24-oxo-D3, calcitroic acid, and probably various water-soluble, sidechain conjugates. Some of these compounds are

90 RIA 80 Circulating 1,25(OH)2D (pg/ml)

supernatant and count the tubes in a gamma well counting system. 1,25(OH)2D values are calculated directly from standard curve by the counting system using a smooth-spline method of calculation. The entire 1,25(OH)2D RIA procedure is displayed in Fig. 16.

BRUCE W. HOLLIS

RRA

70 60 50 40 30 20 10 0 Normal subjects

FIGURE 17

Chronic renal failure

Hypoparathyroid

Biliary atresia

Pregnant subjects

Comparison of circulating 1,25(OH)2D measured by RIA and RRA. From Hollis et al. Clin Chem 42:586–592.

947

CHAPTER 58 Detection of Vitamin D and Its Major Metabolites

120 Immobilized 1-OH-specific MAb

Removed

1) 100% 1,25(OH)2D 1) All non-1-hydroxylated metabolites 2) 100% 1,24,25(OH)3D 3) 100% 1,25,26(OH)3D 4) 100% 1,23,25(OH)3D 5) 100% 1,25(OH)2D3-26,23-lactone 6) 100% Calcitroic acid 7) 100% Water-soluble conjugates of calcitroic acid

FIGURE 18

Graphic description of the vitamin D metabolites assayed as “apparent” circulating 1,25(OH)2D by utilizing immunoextraction with a 1-hydroxy-specific monoclonal antibody (MAb) in conjunction with RIA.

bioactive, but most are not, and this assay cannot distinguish among them (Fig. 18). Further, compare how the IDS RIA for 1,25(OH)2D performs outside of normal or chronic renal failure human samples (Fig. 19). This assay appears to be inadequate when presented with selected pathological samples. We have specifically investigated the effects of l,25(OH)2D3-26,23-lactone on the “apparent” 1,25(OH)2D levels using the immunoextraction technique and found it to interfere on an equal molar basis compared with 1,25(OH)2D3 (Fig. 20). We also know that

Circulating 1,25(OH)2D (pg/ml)

400

RRA IDS-RIA

300

1,25(OH)2D assayed (pg/ml)

Measured

100 80 60 40 20 0

0

15 20 50 5 10 1,25(OH)2D3-26,23-lactone added (pg/ml)

FIGURE 20

Effect of exogenously added l,25(OH)2D-26, 23-lactone on the “apparent” serum concentration of 1,25(OH)2D. Concentrations were assessed by RRA (squares), RIA as described in the text (triangles), and IDS immunoextraction RIA (diamonds). From Hollis BW. Clin Chem 41:1313–1314.

l,25(OH)2D3-26,23-lactone is a significant in vivo metabolite in a variety of clinical samples, and we find its concentration to be 0–30% of the respective 1,25(OH)2D concentration. Further, using the RIA based on immunoextraction, we have found “apparent” levels of 1,25(OH)2D to be grossly higher than the actual concentration in vitamin D-intoxicated subjects, hypoparathyroid subjects receiving vitamin D therapy, and biliary atresia patients (Fig. 19). We have also observed some normal samples that displayed 100% elevation from the actual levels. What this assay is recognizing in these samples remains unknown, but it is undoubtedly some lα-hydroxylated metabolite, probably a catabolic product. It is important to note that the RIA described in this chapter, which is based on classic separation procedures, appears to escape this problem of detecting inactive lα-hydroxylated vitamin D metabolites (Figs. 17).

200

VII. CLINICAL INTERPRETATION AND RELEVANCE OF ANTIRACHITIC STEROL MEASUREMENTS

100

A. Vitamin D 0

Normal Chronic Hypo-para- Biliary Calcium Vitamin D thyroid atresia deficient deficient subjects renal rat rat failure

FIGURE 19

Circulating 1,25(OH)2D as determined by the RRA (squares) or IDS immunoextraction RIA (triangles) on a variety of clinical samples. The same samples were compared in each assay. Horizontal lines denote means.

The quantitation of circulating vitamin D is essentially of no clinical importance. The parent compound is a poor indicator of nutritional status because of its short circulating half-life. The circulating levels of vitamin D are also difficult to interpret because the levels are greatly affected by short-term sun exposure

TABLE IX

Relative Circulating Concentrations of Vitamin D, 25OHD, 24,25(OH)2 and 1,25(OH)2 in Various Disease States

Condition

Vitamin Da

25(OH)Db

24,25(OH)2Dc

1,25(OH)2Dd

Nutritional deficiency Hypoparathyroidism Pseudohypoparathyroidism Hyperparathyroidism Tumor-induced osteomalacia Vitamin D-dependent rickets, type I Vitamin D-dependent rickets, type II Sarcoidosis Renal failure Nephrotic syndrome Hypervitaminosis D Cirrhosis Tuberculosis Hodgkin’s disease Lymphoma Wegener’s granulomatosis X-linked hypophosphatemic rickets

Decreased Normal Normal Normal Normal Normal Normal Normal Normal or decreased Decreased Increased Normal or decreased Normal Normal Normal Normal Normal

Decreased Normal Normal Normal Normal Normal Normal Normal Normal or decreased Decreased Increased Normal or decreased Normal Normal Normal Normal Normal

Normal Normal Normal Normal Normal Normal Normal Normal Decreased Decreased Increased Normal or decreased Normal Normal Normal Normal Normal

Increased followed by decrease Decreased Decreased Decreased Decreased Decreased Increased Increased during hypercalcemia Decreased Decreased Normal or decreased Normal or decreased Increased during hypercalcemia Increased during hypercalcemia Increased during hypercalcemia Increased during hypercalcemia Decreased or normal

aNormal

range is 0–30 ng/ml (0–78 nmol/liter) and is extremely variable with respect to sunlight exposure and dietary intake. range is 15–80 ng/ml (37–192 nmol/liter) and is related to season, latitude, and diet. cNormal range is 0.5–4 ng/ml (1.2–9.6 nmol/liter) and is directly related to circulating 25OHD. dNormal range is 20–60 pg/ml (48–144 pmol/liter). bNormal

949

CHAPTER 58 Detection of Vitamin D and Its Major Metabolites

and dietary intake of vitamin D [38,39]. Vitamin D has proved to be useful in assessing intestinal lipid absorptive capacity associated with fat malabsorption syndromes [40,41]. However, this use is more of a research application as opposed to an application used in a clinical diagnosis. Table IX lists a variety of clinical conditions for which the circulating levels of vitamin D have been defined.

endocrine system, including hypoparathyroidism, hyperparathyroidism, and chronic renal failure, the assay of 1,25(OH)2D is a confirmatory test. It is also important to remember that circulating 1,25(OH)2D provides essentially no information with respect to the patient’s nutritional vitamin D status. Thus, circulating 1,25(OH)2D should not be used as an indicator for hypo- or hypervitaminosis D when nutritional factors are suspected (Table IX).

B. 25OHD References Nutritional vitamin D status is defined by the amount of circulating 25OHD [42]. The assessment of circulating 25OHD is thus an important measurement to the clinician. Subnormal circulating levels of 25OHD usually result from inadequate vitamin D intake and/or insufficient sunlight exposure. This combination of events usually puts elderly patients at risk of developing vitamin D deficiency and ensuing secondary hyperparathyroidism, especially if they are homebound [43]. This, in turn, has been shown to result in an increased incidence of hip fractures in the elderly [44]. Other conditions that contribute to nutritional vitamin D deficiency include nephrotic syndrome, chronic renal disease, cirrhosis, and malabsorption syndromes such as biliary atresia (Table IX). Vitamin D intoxication, though rare, still occurs and is most accurately diagnosed by determining circulating 25OHD. Thus, from a clinical standpoint, the determination of circulating 25OHD is the most frequently requested antirachitic sterol measurement.

C. 24,25(OH)2D At the present time, there does not appear to be a compelling reason to measure circulating 24,25(OH)2D in a clinical setting. Even in a research setting, the determination of 24,25(OH)2D is of questionable value as evidenced by the decreased usage of this assay in the literature since the early 1990s.

D. 1,25(OH)2D Circulating 1,25(OH)2D is diagnostic for several clinical conditions, including vitamin D-dependent rickets types I and II, hypercalcemia associated with sarcoidosis, and other hypercalcemic disorders causing increased 1,25(OH)2D levels. These other disorders include tuberculosis, fungal infections, Hodgkin’s disease, lymphoma, and Wegener’s granulomatosis. In all other clinical conditions involving the vitamin D

1. Seamark DA, Trafford DHJ, Makin HLJ 1981 The estimation of vitamin D and its metabolites in human plasma. J Steroid Biochem 14:111–123. 2. Porteous CE, Coldwell RD, Trafford DJH, Makin HLJ 1987 Recent developments in the measurement of vitamin D and its metabolites in human body fluids. J Steroid Biochem 28:785–801. 3. Jones G, Trafford DJH, Makin HLJ, Hollis BW 1992 Vitamin D: Cholecalciferol, ergocalciferol, and hydroxylated metabolites. In: DeLeenheer AP, Lambert WE, Nelis HJ (eds) Modem Chromato-graphic Analysis of Vitamins. Dekker, New York, pp. 73–151. 4. McCollum EV, Simmonds N, Shipley PG, Park EA 1922 Studies on experimental rickets. XVI. A delicate biological test for calcium–depositing substances. J Biol Chem 51:41–49. 5. Suda T, DeLuca HF, Hallick RB 1971 Synthesis of [26 > 273Hj-25-hydroxycholecalciferol. Anal Biochem 43:139–146. 6. Haddad JG, Chyu KJ 1971 Competitive protein-binding radioassay for 25-hydroxycholecalciferol. J Clin Endocrinol Metab 33:992–995. 7. Belsey R, DeLuca HF, Potts JT 1971 Competitive binding assay for vitamin D and 25-OH vitamin D. J Clin Endocrinol Metab 33:554–557. 8. Haddad JG, Min C, Mendelsohn M, Slatopolsky E, Hahn TJ 1977 Competitive protein-binding radioassay of 24,25-dihydroxyvitamin D in sera from normal and anephric subjects. Arch Biochem Biophys 182:390–395. 9. Brumbaugh PF, Haussler DH, Bursac DM, Haussler MR 1974 Filter assay for 1,25-dihydroxyvitamin D3. Utilization of the hormones target tissue chromatin receptor. Biochemistry 13:4091–4097. 10. Jones G 1978 Assay of vitamins D2 and D3, and 25-hydroxyvitamins D2 and D3 in human plasma by high-performance liquid chromatography. Clin Chem 24:287–298. 11. Eisman JA, Shepard RM, DeLuca HF 1977 Determination of 25-hydroxyvitamin D2 and 25-hydroxyvitamin D3 in human plasma using high-pressure liquid chromatography. Anal Biochem 80:298–305. 12. Clemens TL, Hendy GN, Graham RF, Baggiolini EG, Uskokovic MR, O’Riordan JLH 1978 A radioimmunoassay for 1,25-dihydroxyholecaliferol. Clin Sci Mol Med 54:329–332. 13. Hollis BW, Napoli JL 1985 Improved radioimmunoassay for vitamin D and its use in assessing vitamin D status. Clin Chem 31:1815–1819. 14. Hollis BW, Kamerud JQ, Selvaag SR, Lorenz JD, Napoli JL 1993 Determination of vitamin D status by radioimmunoassay with an 125I-labeled tracer. Clin Chem 39:529–533. 15. Hollis BW, Kamerud JQ, Kurkowski A, Beaulieu J, Napoli JL 1996 Quantification of circulating 1,25-dihydroxyvitamin D

950

16.

17.

18. 19. 20.

21. 22. 23. 24. 24a.

24b. 25. 26. 27. 28. 29.

BRUCE W. HOLLIS

by radioimmunoassay with an 125I-labeled tracer. Clin Chem 42:586–592. Horst RL, Littledike ET, Riley JL, Napoli JL 1981 Quantitation of vitamin D and its metabolites and their plasma concentrations in five species of animals. Anal Biochem 116:189–203. Lambert PW, DeOreo PB, Hollis BW, Fu IY, Ginsberg DJ, Roos BA 1981 Concurrent measurement of plasma levels of vitamin D3 and five of its metabolites in normal humans, chronic renal failure patients, and anephric subjects. J Lab Clin Med 98:536–548. Hollis BW, Roos BA, Lambert PW 1981 Vitamin D in plasma: Quantitation by a nonequilibrium ligand binding assay. Steroids 37:609–613. Horst RL, Reinhardt TA, Beitz DC, Littledike ET 1981 A sensitive competitive protein binding assay for vitamin D in plasma. Steroids 37:581–591. Rhodes CJ, Claridge PA, Traffold DJH, Makin KLJ 1983 An evaluation of the use of Sep-Pak Qg cartridges for the extraction of vitamin D3 and some of its metabolites from plasma and urine. J Steroid Biochem 19:1349–1354. Liel Y, Ulmer E, Shary J, Hollis BW, Bell NH 1988 Low circulating vitamin D in obesity. Calcif Tissue Int 43:199–201. Belsey RE, DeLuca HF, Potts JT 1974 A rapid assay for 25-OH-vitamin D3 without preparative chromatography. J Clin Endocrinol Metab 38:1046–1051. Hollis BW, Pittard WB 1984 Evaluation of the total fetomaternal vitamin D relationships at term: Evidence for racial differences. J Clin Endocrinol Metab 59:652–657. Dorantes LM, Arnaud SB, Arnaud CD 1978 Importance of the isolation of 25-hydroxyvitamin D before assay. J Lab Clin Med 91:791–796. Binkley N, Krueger D, Cowgill CS, Plum L, Lake E, Hansen KE, DeLuca HF, Drezner MK 2004 Assay variation confounds the diagnosis of hypovitaminosis D: A cell for standardization. J Clin Endocrinol Metab 89:3152–3157. Hollis BW 2004 Editorial: The determination of circulating 25-hydroxyvitamin D: No easy task. J. Clin Endocrinol Metab 89:3149–3151. Shepard RM, Horst RL, Hamstra AJ, DeLuca HF 1979 Determination of vitamin D and its metabolites in plasma from normal and anephric man. Biochem J 182:55–69. Dreyer BE, Goodman DBP 1981 A simple direct spectrophoto-metric assay for 24,25-dihydroxyvitamin D3. Anal Biochem 114:37–41. Hummer L, Christiansen C 1984 A sensitive and selective radioimmunoassay for serum 24,25-dihydroxycholecaliferol in man. Clin Endocrinol 21:71–79. Wei S, Tanaka H, Kubo T, Ichikawa M, Seino Y 1994 A multiple assay for vitamin D metabolites without high-performance liquid chromatography. Anal Biochem 22:59–365. Jones G, Byrnes B, Palma F, Segev D, Mazur Y 1980 Displacement potency of vitamin D2 analogs in competitive

30.

31.

32. 33. 34. 35. 36.

37. 38.

39.

40.

41.

42. 43. 44.

protein-binding assays for 25-hydroxyvitamin D3, 24,25-dihydroxyvitamin D3 and 1,25-dihydroxyvitamin D3. J Clin Endocrinol Metab 50:773–775. Eisman JA, Hamstra AJ, Kream BE, DeLuca HF 1976 A sensitive, precise, and convenient method for determination of 1,25-dihydroxyvitamin D in human plasma. Arch Biochem Biophys 176:235–243. Reinhardt TA, Horst RL, Orf JW, Hollis BW 1984 A microassay for 1,25-dihydroxyvitamin D not requiring high performance liquid chromatography: Application to clinical studies. J Clin Endocrinol Metab 58:91–98. Hollis BW 1986 Assay of circulating 1,25-dihydroxyvitamin D involving a novel single-cartridge extraction and purification procedure. Clin Chem 32:2060–2063. Bouillon R, De Moor P, Baggiolini EG, Uskokovic MR 1980 A radioimmunoassay for 1,25-dihydroxycholecalciferoI. Clin Chem 26:562–567. Gray TK, McAdoo T, Pool D, Lester GE, Williams ME, Jones G 1981 A modified radioimmunoassay for 1,25-dihydroxychole-calciferol. Clin Chem 27:458–463. Hollis BW 1995 1,25-Dihydroxyvitamin D3-26,23-lactone interferes in determination of 1,25-dihydroxyvitamin D by RIA after immunoextraction. Clin Chem 41:1313–1314. Mawer EB, Berry JL, Bessone J, Shany S, White A 1985. Selection of high-affinity and high specificity monoclonal antibodies for 1,25-dihydroxyvitamin D. Steroids 46:741–754. Bouillon R, Okamura WH, Norman AW 1995 Structurefunction relationships in the vitamin D endocrine system. Endocr Rev 16:200–257. Adams JA, Clemens TL, Parrish JA, Holick MF 1982 Vitamin D synthesis and metabolism after ultraviolet irradiation of normal and vitamin D–deficient subjects. N Engl J Med 306:722–725. Hollis BW, Lowery JW, Pittard WB, Guy DG, Hansen JW 1996 Effect of age on the intestinal absorption of vitamin D3palmitate and nonesterified vitamin D2 in the term human infant. 81:1385–1388. Heubi JE, Hollis BW, Tsang RC 1990 Bone disease in chronic childhood cholestasis II. Better absorption of 25(OH)D than vitamin D in extrahepatic biliary atresia. Pediatr Res 27:26–31. Argao EA, Heubi JE, Hollis BW, Tsang RC 1992 d-alphatocopheryl polyethylene glycol-1000 succinate enhances the absorption of vitamin D in chronic cholestatic liver disease of infancy and childhood. Pediatr Res 31:146–150. Haddad JG, Stamp TCB 1974 Circulating 25-hydroxyvitamin D in man. Am J Med 57:57–62. Gloth MF, Gundberg CM, Hollis BW, Haddad JG, Tobin JD 1995 Vitamin D deficiency in homebound elderly persons. JAMA 274:1683–1686. Chapuy MC, Arlot ME, DuBoeuf F, Brun J, Crouzet B, Arnaud S, Meunier P 1992 Vitamin D3 and calcium to prevent hip fractures in elderly women. N Engl J Med 327:1637–1642.

CHAPTER 59

Bone Histomorphometry JULIET E. COMPSTON

I. II. III. IV. V.

University of Cambridge School of Clinical Medicine, Cambridge, United Kingdom

Introduction Bone Biopsy Histomorphometry Assessment of Mineralization Histological Diagnosis of Osteomalacia

I. INTRODUCTION Bone histomorphometry describes the quantitative assessment of bone remodeling, modeling, and structure. It provides information that is not currently available from other investigative approaches, for example, bone densitometry and biochemical markers of bone turnover. Bone histomorphometry also enables a more precise characterization of disease states and their response to treatment than can be obtained from qualitative examination of bone histology. In the last two decades, there have been significant advances in histomorphometric techniques, most notably the use of computerized rather than manual techniques and the development of sophisticated approaches to the assessment of bone microarchitecture. The application of these techniques has been particularly valuable in determining the cellular pathophysiology of different forms of bone diseases and in defining the mechanisms by which drugs affect bone.

II. BONE BIOPSY A. Procedure The iliac crest is the preferred site for bone biopsy in patients with metabolic bone disease. Most investigators favor the transverse approach, in which a biopsy containing two cortices and intervening cancellous bone is obtained (Fig. 1), in contrast to vertical biopsies, which contain only one cortical plate. In growing individuals, only the transverse approach should be considered, because of the presence of the growth plate along the top of the crest. A number of specially designed trephines are commercially available; ideally for bone histomorphometry, the internal diameter of the specimen should be at least 6 mm. In most cases, the biopsy is performed as an outpatient procedure. For a transiliac biopsy, the patient lies VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

VI. VII. VIII. IX.

Assessment of Bone Turnover Assessment of Remodeling Balance Assessment of Bone Structure Conclusions and Future Developments References

in the supine position, and the specimen is obtained approximately 2 cm below and behind the anterior superior iliac spine. Most operators use a mild sedative, such as midazolam, and some also routinely administer an analgesic by mouth or injection before the procedure. However, if care is taken to ensure adequate local anesthesia during the biopsy, the latter measure is not generally necessary. The biopsy should be performed under sterile conditions. The area around the anterior superior iliac spine is infiltrated with local anesthetic, and the inner and outer periosteum are then anesthetized. The author uses a small trocar and stilette, which is driven with a weighted instrument until it lies just in the outer cortex; local anesthetic is infiltrated under the periosteum, and the trocar and stilette are then advanced through the bone to the inner cortex, where the procedure is repeated. Other investigators anesthetize the inner cortex by introducing a needle through the skin from opposite the biopsy site. A small skin incision is then made, and a hollow cannula with a serrated edge is introduced and placed firmly on the outer periosteum. A smaller, hollow cannula is then inserted through the larger cannula, and the

FIGURE 1

Section of transiliac biopsy obtained with an 8mm internal diameter trephine. The biopsy contains inner and outer cortical plates and intervening cancellous bone. Copyright © 2005, Elsevier, Inc. All rights reserved.

952 biopsy is obtained by advancing the serrated edge of this cannula through the iliac crest until it has reached the outer surface of the inner cortex. The cannula is then withdrawn after rotation through 360° (to ensure that the core of bone has been freed from adjoining tissues), and the biopsy is removed from the cannula using a metal rod. The incision is then sutured and dressed, and the patient is instructed to lie on the side of the biopsy to apply pressure to the site and reduce the risk of bruising. Ideally, this position should be maintained for 2 hr, and thereafter the patient should be advised to rest for 24 hr.

B. Adverse Effects Bone biopsy is safe and generally well tolerated, although there is often some discomfort after the procedure, usually lasting between 24 and 48 hr. The morbidity is low and mainly due to hematoma, which may occasionally be extensive; this is most likely to occur in obese subjects or in patients with bleeding diatheses. Other reported complications include infection, transient femoral nerve palsy, avulsion of the superior ramus of the iliac crest, fracture of the iliac crest, and osteomyelitis. It should be stressed, however, that these are extremely rare, and, overall, the incidence of all complications is less than 1% [1].

JULIET E. COMPSTON

III. HISTOMORPHOMETRY A. Theoretical Considerations Histological sections of bone provide a twodimensional representation of a three-dimensional structure; extrapolation from one to the other requires the application of stereological formulae that are based on the assumptions that sampling is random and unbiased and, for most applications, that the structure is isotropic (i.e., evenly dispersed and randomly orientated in space) [2]. Although these conditions cannot be strictly fulfilled in the case of bone histomorphometry, conversion of histomorphometric indices to three-dimensional units is used by some investigators, whereas others express their data as two-dimensional values. The absolute values generated by these different approaches clearly differ, but in practical terms either is acceptable provided that consistency is maintained. The requirement, for stereological purposes, for bone to be isotropic is clearly not fulfilled in bone in which the macro- and microarchitecture are primarily determined by mechanical forces. However, random and unbiased sampling with respect to isotropy can be achieved using vertical sections, as first described by Baddeley et al. [3] and subsequently applied to bone by Vesterby et al. [4]; the vertical axis of the sections is kept parallel to the axis of the cycloid test system, in which the test lines are defined in relation to the axis (sine-weighted).

C. Indications for Bone Biopsy In clinical practice, bone biopsy is most often performed to exclude or confirm a diagnosis of osteomalacia and to characterize the different forms of renal osteodystrophy. In patients with suspected osteomalacia, the diagnosis may be evident from biochemical and/or radiological abnormalities, but in the absence of these, bone biopsy is required. Chronic renal failure is associated with several types of bone disease, as discussed in Chapter 76; accurate diagnosis is essential to establish the correct treatment. Bone biopsy is also helpful in a number of other, rarer forms of metabolic bone disease, for example, fibrogenesis imperfecta ossium and hypophosphatemic osteomalacia. For diagnostic purposes, qualitative assessment of bone by an experienced histopathologist is sufficient, and histomorphometry is not required. Although bone biopsy is a valuable research tool in osteoporosis, it is of little value diagnostically, mainly because of the heterogeneity of bone loss and the relatively weak correlations between bone mass in iliac crest biopsies and clinically relevant sites such as the spine and femoral neck.

B. Methodology The use of manual techniques using grids and graticules has now been almost entirely replaced by interactive computerized systems. These are faster, more operator-friendly, and possess the ability to perform complex measurements that could not be achieved manually. A number of systems are now available, both commercial and in-house [5,6].

C. Terminology The nomenclature applied to bone histomorphometry has been standardized [7] in an attempt to clarify the cumbersome and sometimes unintelligible terminology used previously. The revised system expresses all data in terms of the source (the structure on which the measurement is made), the measurement, and the referent. The recommended format (including punctuation) is source-measurement/referent, although because only one source is usually used in a particular study, it

953

CHAPTER 59 Bone Histomorphometry

TABLE I Referents Commonly Used in Bone Histomorphometry Referent (3D/2D)

Abbreviation (3D/2D)

Bone surface/perimeter Bone volume/area Tissue volume/area Core volume/area Osteoid surface/perimeter Eroded surface/perimeter Mineralized surface/perimeter Osteoblast surface/perimeter Osteoclast surface/perimeter

BS/B.Pm BV/B.Ar TV/T.Ar CV/C.Ar OS/O.Pm ES/E.Pm Md.S/Md.Pm Ob.S/Ob.Pm Oc.S/Oc.Pm

3D, three dimensions; 2D, two dimensions. Adapted from Parfitt et al. [7], with permission.

is unnecessary to specify this once it has been defined. Area or perimeter (or volume and surface) are used as referents for most measurements (Table I). Histomorphometric data may be described in twodimensional or three-dimensional terms; the system used should be consistent within studies. Primary measurements are referred to as area, perimeter, and width TABLE II

Primary Histomorphometric Indices of Bone Remodeling

Name Bone area Osteoid area Osteoid perimeter Osteoblast perimeter Osteoid width Interstitial width Trabecular width Eroded perimeter Osteoclast perimeter Mineralizing surface Mineral apposition rate Wall width Erosion depth Erosion length Erosion area Cavity number Quiescent perimeter Reversal perimeter

Abbreviation B.Ar/T.Ar O.Ar/T.Ar or O.Ar/B.Ar O.Pm/B.Pm Ob.Pm/B.Pm O.Wi It.Wi Tb.Wi E.Pm/B.Pm Oc.Pm/B.Pm Md.Pm/B.Pm MAR W.Wi E.De E.Le E.Ar N.Cv./B.Pm or N.Cv./T.Ar Q.Pm Rv.Pm

Adapted from Parfitt et al. [7], with permission.

Units % % % % µm µm µm % % % µm/day µm µm µm µm2 no./mm or no./mm2 % %

(two-dimensional nomenclature) or volume, surface, and thickness (three-dimensional nomenclature) (Table II). It should be noted that absolute area and perimeter measured on conventional histological sections have no three-dimensional equivalent but, if three-dimensional nomenclature is adopted, these are referred to as volume and surface although the absolute values are identical. Cortical width and thickness are also numerically equal; however, in other situations, for example, trabecular width, conversion to thickness requires division of width by 4/π (1.273) for isotropic structures or 1.2 for human iliac cancellous bone [2,8]. The units recommended for the revised nomenclature are micrometer and millimeter for length and day and year for time; surface/surface and volume/volume ratios are expressed as percentages, whereas volume/surface ratios are expressed in mm3/mm2. Some of the more commonly used derived histomorphometric indices are shown in Table III. Most histomorphometric studies have been confined to cancellous bone which, because of its high surface to volume ratio, exhibits greater remodeling activity than cortical bone. Although cortical bone predominates throughout the skeleton and is a major determinant of bone strength and fracture risk, it is largely ignored by histomorphometrists. The application of histomorphometric techniques to cortical bone was first described by Frost [9] in the rib and more recently TABLE III

Derived Histomorphometric Indices of Bone Remodeling

Name Adjusted apposition rate Bone formation rate Erosion rate Mineralization lag time Osteoid maturation time Formation period Active formation period Erosion period Reversal period Quiescent period Remodelling period Total period Activation frequency Trabecular separationa Trabecular numbera

Abbreviation Aj.AR BFR/B.Pm BFR/B.Ar ER Mlt Omt FP FP(a+) EP Rv.P QP Rm.P Tt.P Ac.f Tb.Sp Tb.N

Units µm/day µm2/µm/day %/year µm/day day(s) day(s) day(s) day(s) day(s) day(s) day(s) day(s) day(s) /year µm or mm /mm

aMay also be measured directly. Adapted from Parfitt et al. [7], with permission.

954 a detailed analysis of the bone remodeling cycle in human cortical bone [10] and studies of cortical bone structure in the femoral neck and iliac crest [11–14] have been discribed.

D. Limitations of Bone Histomorphometry Certain limitations of bone histomorphometry should be recognized. Some of these are inherent in the restrictions imposed by a single biopsy site and disease heterogeneity, whereas others reflect imperfections in measurement techniques and difficulties in identification of some of the key processes in remodeling; at least some of those in the latter categories may eventually be overcome by methodological improvements in the future. A number of studies have documented the large measurement variance associated with bone histomorphometry, which arises from a number of sources including intra- and interobserver variation, sampling variation, and methodological factors [15–17]. Intra- and interobserver variation reflect the subjective approach to identification of many of the histological features assessed, for example, resorption cavities and osteoid seams. Methodological factors include the criteria used for corticomedullary differentiation, which are often arbitrary, the staining method used, and the magnification at which measurements are made. The technique used for quantitation may also affect the values obtained, as a result of different sampling procedures and variations in the number of sampling points utilized. Many of these sources of variance can be minimized by the standardization of staining, corticomedullary delineation, and magnification and the employment of criteria for identification of osteoid seams, resorption cavities, and newly formed bone structural units. The important issue of how closely bone remodeling in the iliac crest resembles that at other skeletal sites has not been fully resolved. Some metabolic bone disorders, for example, osteomalacia and most forms of renal osteodystrophy, appear to affect the whole skeleton, and in such cases an iliac crest biopsy is representative. In osteoporosis, however, there is clear evidence of disease heterogeneity and it is well documented that bone volume in iliac crest biopsies is a poor indicator of bone loss elsewhere in the skeleton [18]. There is also some evidence for variations in bone turnover at different skeletal sites [19]; however, the demonstration of clear abnormalities of bone remodeling in iliac crest bone obtained from patients with osteoporosis indicates that changes responsible for the disease process are reflected, at least to some extent, in bone from this site. Similarly, changes in bone turnover and remodeling balance have been shown in patients with treated

JULIET E. COMPSTON

osteoporosis and have generally been consistent with the observed changes in bone mass at sites such as the spine and hip. Finally, current histomorphometric techniques are seriously limited by the lack of reliable markers for activation and resorption. Dynamic indices related to these processes are at present calculated from bone formation rates, based on the assumptions that bone resorption and formation are coupled temporally and spatially and that bone remodeling is in a steady state; however, these are unlikely to be tenable in many cases of untreated or treated osteoporosis [20].

IV. ASSESSMENT OF MINERALIZATION A. In vivo Tetracycline Labeling The administration of two time-spaced doses of a tetracycline derivative prior to bone biopsy enables measurement and calculation of dynamic indices of bone formation and, by extrapolation, bone resorption [21]. Such information is not currently obtainable by other means, and tetracycline labeling should therefore be performed whenever possible. Various regimens have been described; most involve a 10–12 day gap between the two doses, bone biopsy being performed 3–5 days after the last dose. The regimen used by the author is as follows: Days 1 and 2 150 mg demeclocycline twice daily 10 days No demeclocycline Days 13 and 14 150 mg demeclocycline twice daily The bone biopsy is performed 3–5 days after day 14. Adverse effects of demeclocycline and related compounds are rare but include diarrhea and other gastrointestinal symptoms. Occasionally, skin rashes occur; these are sometimes severe and may exhibit photosensitivity. There is evidence that different tetracycline compounds differ with respect to their uptake by mineralizing bone. Parfitt et al. [22] reported that demeclocycline labeling resulted in a greater surface extent of fluorescence than oxytetracycline, significantly affecting values for dynamic indices of bone formation. These differences should be borne in mind when comparing data between centers and, in particular, when using control data obtained from other sources.

B. Measurement of Osteoid Osteoid may be distinguished from mineralized bone by several staining procedures, including von Kossa, toluidine blue, Goldner’s trichrome, and solochrome cyanin.

955

CHAPTER 59 Bone Histomorphometry

of osteoid, as osteoid amount tends to be greater in the corticoendosteal region than in pure cancellous bone.

C. Dynamic Indices of Mineralization

FIGURE 2 Toluidine blue-stained section of iliac crest cancellous bone showing mineralized bone (purple/blue) and osteoid (pale blue). The calcification front can be seen as a dark blue line at the interface of the osteoid and mineralized bone (See color plate).

Primary measurements of osteoid include area, perimeter, and seam width; assessment of dynamic indices of mineralization, for example, mineral apposition rate, mineralization lag time, and osteoid maturation rate, requires double tetracycline labeling prior to biopsy [7]. Calcification fronts along osteoid seams can be demonstrated by staining with toluidine blue (Fig. 2; see color insert), Sudan black B, or thionin, but the surface extent assessed in this way does not always correlate well with the mineralizing perimeter as measured using tetracycline uptake [23]. Osteoid seam width may be measured directly or calculated from osteoid area and perimeter. However, when relatively small amounts of osteoid are present, the latter approach is inaccurate because the value for osteoid area, expressed as a percentage of bone area, is influenced not only by the seam width but also by the mineralized bone area [24]. Direct assessment can be made using an eyepiece micrometer and calculating the mean of several equidistant measurements of width along each seam. Measurements of osteoid, in particular its perimeter, are strongly influenced by the magnification used. At high magnifications, it becomes difficult to distinguish osteoid seams from the thin endosteal membrane covering the quiescent bone surface, and for this reason it is preferable to use defined criteria, for example, all seams less than one lamella (3 µm) are excluded. Osteoid measurements may also be affected by the staining procedure used, and poor differentiation of osteoid from mineralized bone in images used for semiautomatic techniques may reduce the accuracy of measurements; thus, values obtained using image analysis are generally lower than those generated by manual measurements [16]. Finally, the delineation of the corticomedullary junction may affect the values obtained for primary measurements

The administration of time-spaced tetracycline labels prior to biopsy enables calculation of the dynamic indices of matrix formation and mineralization that are central to the histomorphometric diagnosis of osteomalacia. Tetracycline binds to calcium and becomes permanently incorporated into the mineralization fronts at sites of active mineralization (Fig. 3; see color insert). In this respect, the timing of the biopsy in relation to the labeling regime is crucial, with a period of 3 to 5 days after the last label enabling deposition of a sufficiently thick layer of new mineral to retain the tetracycline label [25]. The primary histomorphometric indices of bone remodeling are listed in Table II and the derived indices in Table III. Normative histomorphometric data from the author’s laboratory are shown for females (Table IV) and males (Table V). Other groups have also published normative data and in particular normative data for the pediatric age group are now available [25a]. 1. MINERAL APPOSITION RATE

The mineral apposition rate (MAR) is calculated as the distance between two time-spaced tetracycline labels divided by the time between the administration of the labels. Measurements are made from the midpoint of each label at approximately equidistant points along the labeled surface, and the interlabel period is calculated as the number of days between the midpoints of the two labeling periods [26]. MAR is used in the calculation of many derived indices of bone formation,

FIGURE 3

Unstained section of iliac crest cancellous bone viewed by fluorescence microscopy. The tetracycline labels are seen as double yellow fluorescent bands. (See color plate.)

956

JULIET E. COMPSTON

TABLE IV

Normative Histomorphometric Data in Femalesa Age range

Parameter BV/TV (%) OV/BV (%) OS/BS (%) ES/BS (%) Md.S/BS (%) O.Th (µm) W.Th (µm) MAR (µm/day) Mlt (days) BFR (µm3/µm2/day)

19–30 years (n = 5) 25.9(3.1) 2.4(1.4) 13.1(5.9) 2.15(0.36) 9.4(1.8) 5.4(1.9) 45.7(4.9) 0.59(0.06) 12.2(6.2) 0.056(0.013)

31–40 years (n = 6)b 27.7(5.5) 2.7(2.4) 17.7(11.5) 1.84(0.92) 8.1(3.4) 3.9(1.6) 51.2(6.6) 0.60(0.12) 16.5(11.8) 0.060(0.022)

41–50 years (n = 6)c 29.6(2.1) 2.3(1.3) 14.5(7.8) 1.78(1.03) 8.1(4.2) 5.8(1.1) 47.5(4.9) 0.61(0.09) 20.7(8.7) 0.051(0.031)

51–60 years (n = 10)d 23.9(4.5) 3.1(1.9) 21.2(11.3) 1.76(0.83) 13.0(6.7) 5.8(1.6) 36.1(2.9) 0.61(0.10) 21.2(19.6) 0.084(0.045)

61–80 years (n = 6) 19.8(3.9) 4.7(1.9) 35.0(12.1) 1.66(0.66) 14.8(8.1) 5.8(2.3) 32.5(3.6) 0.54(0.07) 29.6(13.5) 0.081(0.043)

aResults are expressed as means, with SD values in parentheses. Data from Vedi et al. [70,71]. Md.S/BS was calculated as the double plus half the single tetracycline-labelled surface. bn = 4 for dynamic variables. cn = 5 for dynamic variables. dn = 9 for dynamic variables.

and accurate measurement is thus of key importance. In the absence of tetracycline uptake, mineral apposition rate and derived indices should be treated as missing data; in biopsies in which only single labels can be detected, the finite lower limit of 0.3 µm/day for mineral apposition rate should be used for the calculation of derived indices [27].

2. ADJUSTABLE APPOSITION RATE

The adjusted apposition rate (Aj.AR) represents the mineral apposition rate averaged over the osteoid surface. In the absence of a mineralization defect, the apposition of matrix and mineral, whilst not synchronous, can be assumed to occur at the same rate, and under such circumstances, the adjusted apposition

TABLE V Normative Histomorphometric Data in Malesa Age range Parameter BV/TV (%) OV/BV (%) OS/BS (%) ES/BS (%) Md.S/BS (%) O.Th (µm) W.Th (µm) MAR (µm/day) Mlt (days) BFR (µm3/µm2/day)

19–30 years (n = 3) 31.3(6.4) 1.6(0.8) 10.0(5.3) 2.84(1.27) 9.9(4.0) 8.3(3.0) 49.7(9.6) 0.67(0.07) 10.5(1.9) 0.066(0.006)

31–40 years (n = 6) 22.2(3.9) 4.1(1.6) 28.3(7.8) 1.69(0.62) 13.5(8.4) 6.1(1.3) 45.9(4.4) 0.60(0.10) 28.5(14.6) 0.079(0.049)

41–50 years (n = 3) 26.9(7.1) 2.8(0.9) 26.3(6.0) 1.68(0.32) 8.7(7.0) 6.9(2.9) 42.8(4.0) 0.55(0.10) 36.1(6.2) 0.047(0.005)

51–60 years (n = 6) 23.0(5.5) 3.1(0.9) 20.0(7.0) 1.77(0.68) 8.8(1.5) 6.2(2.0) 36.9(2.0) 0.57(0.13) 34.7(40.8) 0.051(0.015)

61–80 years (n = 6) 21.4(2.6) 5.6(3.6) 34.8(15.7) 1.91(0.42) 9.2(5.1) 6.6(2.8) 33.4(3.3) 0.53(0.05) 49.4(28.1) 0.045(0.023)

aResults are expressed as means with SD values in parentheses. Data from Vedi et al. [70,71] Md.S/BS was calculated as the double plus half the single tetracycline-labeled surface.

CHAPTER 59 Bone Histomorphometry

957

rate is equivalent to the osteoid or matrix apposition rate. It is calculated as follows: Aj.AR = MAR × Md.Pm/O.Pm. From the above formula, it is clear that Aj.AR is usually less than MAR and cannot exceed it. 3. MINERALIZATION LAG TIME AND OSTEOID MATURATION TIME

The mineralization lag time (Mlt) is the interval between deposition and mineralization of a given amount of osteoid, averaged over the life span of the osteoid seam. It is calculated as Mlt = O.Wi/Aj.AR. The osteoid maturation time (Omt) is the period between the deposition and onset of mineralization of a given amount of osteoid and results from processes such as collagen cross-linking that are necessary before mineralization can proceed. In humans it is usually shorter than Mlt and never exceeds it. It is calculated as follows: Omt = O.Wi/MAR.

V. HISTOLOGICAL DIAGNOSIS OF OSTEOMALACIA

FIGURE 4 (Top) Section of iliac crest stained by the von Kossa technique to demonstrate mineralized bone (black) and osteoid (pink) in a normal subject. (Bottom) Section of iliac crest stained by the von Kossa technique to show osteoid accumulation in a woman with severe privational osteomalacia. (See color plate.)

A. Generalized Osteomalacia Osteomalacia is essentially a histological diagnosis, although biochemical and radiological abnormalities may enable a firm diagnosis to be made without the necessity for histological examination of bone. Nonetheless, osteomalacia may exist in the absence of biochemical and radiological abnormalities [28], and in such cases bone biopsy is the only certain means by which the diagnosis can be established. The cardinal feature of osteomalacia is defective mineralization, which results in accumulation of osteoid with an increase in the width of osteoid seams (Fig. 4; see color insert) and a reduction in the surface extent of osteoid showing tetracycline labeling; there is often an increase in the width of individual tetracycline labels, and the distance between double labels is reduced or undetectable. In severe cases, tetracycline labeling may be absent. Increased bone turnover and erosion depth, due to secondary hyperparathyroidism, are present in the earlier stages of osteomalacia but become less apparent as the mineralized bone surface becomes covered with thick osteoid seams and,

therefore, inaccessible to osteoclastic resorption. In such cases, tunneling resorption may be apparent, and the irregular outline of mineralized bone beneath the thick osteoid seams provides evidence of previous resorption. Para-trabecular fibrosis is also seen in severe cases. In contrast, histological evidence of secondary hyperparathyroidism is absent in untreated hypophosphatemic osteomalacia. In histomorphometric terms, osteomalacia is defined as an increased osteoid seam width and a prolonged mineralization lag time [25]. The criteria for abnormality in these indices depend on the source of the reference data, which may vary as a result of both geographical and methodological factors, but in most centers a mean osteoid seam width greater than 12.5 um and a mineralization lag time in excess of 100 days would be regarded as abnormal. Although the mineral apposition rate is also reduced in osteomalacia, this is not a specific feature because low mineral apposition rates may also result from a reduction in matrix

958 apposition rate as occurs, for example, in postmenopausal osteoporosis, osteogenesis imperfecta, and some forms of secondary osteoporosis. Similarly, the mineralization lag time will be increased in the presence of reduced matrix apposition rate and is therefore not, by itself, pathognomonic of osteomalacia. The distinguishing feature of osteomalacia is that the mineralization lag time is prolonged relative to the adjusted apposition rate, whereas in osteoporosis the reverse is true; this phenomenon accounts for the increase in osteoid seam width that is specific to osteomalacia.

B. Focal Osteomalacia Focal osteomalacia has been described in patients taking bisphosphonate therapy and is characterized by the focal distribution of abnormally thick osteoid seams with impaired mineralization [29,30]. In such cases, the osteoid area and perimeter may be normal, and, because some osteoid seams are of normal width and exhibit normal mineralization, the abnormality may only be detected by examination of the distribution, within single biopsies, of values for osteoid seam width. The significance of these histological changes in terms of fracture risk has not been established; they do not appear, however, to be associated with clinical symptoms or biochemical abnormalities.

VI. ASSESSMENT OF BONE TURNOVER Bone turnover describes the tissue level of bone resorption and formation, a key determinant of which is activation frequency, that is, the probability that a new remodeling cycle will be initiated at any point of the bone surface. The uptake of tetracycline derivatives at sites of actively forming bone enables the rate of mineralization of osteoid seams and the surface extent of bone formation to be assessed and a number of derived indices, including activation frequency, to be calculated. Tetracycline labeling in bone is detected by fluorescence microscopy under blue light.

A. Mineralizing Perimeter or Surface The extent of bone perimeter or surface that exhibits tetracycline fluorescence is an important primary measurement from which many dynamic indices are derived. When a double tetracycline label has been administered, both double and single labels will be seen; this reflects the labeling escape error, caused by initiation of mineralization before the first label or its

JULIET E. COMPSTON

termination between administration of the two labels [31], and probably also the switch from an active to resting state in a minority of osteoid seams (the socalled on/off phenomenon) [2]. Because of the former, the extent of double-labeled surface underestimates the actively mineralizing surface, and the double plus half the single label is therefore used to estimate the mineralizing perimeter. In cases where only single labels can be detected, it has been suggested that the mineralizing perimeter should be expressed as half the singlelabeled perimeter [27]. In the absence of tetracycline administration prior to biopsy, the osteoid perimeter may provide some indication of bone turnover, with increased osteoid perimeter being characteristic of high turnover states. An increase in the extent of perimeter occupied by resorption cavities does not necessarily imply increased bone turnover, however, as these may not represent active resorption but rather reflect failure of formation to occur in previously resorbed cavities.

B. Bone Formation Rates Bone formation rates are usually expressed in terms of the bone perimeter or area. In the former case, either the osteoid perimeter may be used as the referent (adjusted apposition rate) or the total bone perimeter (tissue-based bone formation rate). They are calculated as follows: Adjusted apposition rate = MAR × Md.Pm/O.Pm, Bone formation rate (tissue-based) (BFR/BS) = MAR × Md.Pm/B.Pm. The bone formation rate/bone area (BFR/B.Ar) is calculated as BFR/B.Ar = MAR × Md.Pm × (B.Pm/B.Ar) × 100.

C. Bone Resorption Rates Because of the lack of markers of active resorption analogous to the use of tetracycline to identify actively forming bone, bone resorption rates can only be calculated indirectly, from bone formation rates, based on the assumptions discussed earlier. Erosion rate (ER) is expressed in micrometers per day and calculated as follows: ER = E.De/EP. Calculation of the erosion period (EP) is shown below.

959

CHAPTER 59 Bone Histomorphometry

D. Remodeling Periods The average duration of a single remodeling cycle is described as the remodeling period, which can be further divided into quiescent, erosion, reversal, and formation remodeling periods [2]. The formation period (FP) can be further divided into the active [FP(a+)] and inactive formation period [FP(a−)] [32]; the latter is a measure of the “off-time,” which accounts for the discrepancy between osteoid and mineralizing perimeter after correction for label escape [31]. Formation periods are calculated as follows: FP = W.Wi/Aj.AR FP(a+) = W.Wi/M AR, FP(a−) = FP − FP(a+), where W.Wi (wall width) is the mean width of completed bone structural units. Quiescent, erosion, and reversal periods are calculated as follows: QP = Q.Pm/B.Pm × FP, EP = E.Pm/B.Pm × FP, Rv.P = Rv.Pm/B.Pm × FP. where Rv.Pm = E.Pm − osteoclastic perimeter; Q.Pm = B.Pm − (O.Pm + E.Pm). The mean time between initiation of two successive remodeling cycles at the same site is defined as the total period and calculated as follows: Tt.P = Rm.P + QP.

E. Activation Frequency

previously activated units. Because activation may occur at any site on the bone surface, however, there seems no a priori reason why this should be so, particularly in nonsteady states [20]. In addition, the use of indices of bone formation to calculate activation frequency relies on assumptions about the coupling of bone resorption and formation that are unlikely to be tenable in many disease states.

VII. ASSESSMENT OF REMODELING BALANCE A. Bone Formation Within individual remodeling units, the amount of bone formed is termed the wall width [34]; this is measured as the mean width of completed bone structural units that are identified under polarized light (Fig. 5; see color insert) or by stains such as toluidine blue or thionin [35], which demonstrate the cement line. Completed structural units are identified by the absence of resorption lacunae or osteoid. There is a large variation in reported values for wall width in both normal and osteoporotic subjects [20], reflecting differences in sampling procedures and difficulties in accurate identification of the cement line, which forms the base of the original resorption cavity. Investigation of the effect of a disease or its treatment on wall width (and calculated dynamic indices for which wall width is required) necessitates differentiation of those units formed during the period of observation from those formed prior to this time. This can only be achieved by identification of uncompleted bone structural units, which have a covering of osteoid and hence can be presumed to represent current or recent remodeling activity. Reconstruction of these

Activation frequency (Ac.f) is a key determinant of bone mass in the adult skeleton, and increased activation frequency, resulting in high bone turnover, forms quantitatively the most important mechanism of bone loss in osteoporosis [33]. At present, however, there are no in situ markers of activation, and hence direct assessment of activation frequency cannot be made. Rather, it is calculated as the frequency with which a given site on the bone surface undergoes new remodeling, as follows: Ac.f = 1/Tt.P or (BFR/B.Pm)/W.Wi. These formulae define activation frequency as the reciprocal of the time taken from the initiation of one remodeling cycle to initiation of a new one at that site, thus implying its dependence on the life span of

FIGURE 5 Section of iliac crest biopsy viewed under polarized light to show a completed bone structural unit bounded by the cement line and mineralized bone surface.

960

JULIET E. COMPSTON

forming sites can then be achieved [36]; however, the number of such units that can be identified in any one biopsy is likely to be extremely small and variance correspondingly high.

B. Bone Resorption There are several problems associated with accurate assessment of the amount of bone resorbed during each remodeling cycle. The identification of resorption cavities is often difficult and always to some extent subjective; in addition, it is difficult to identify those cavities in which resorption has been completed. The use of polarized light microscopy to demonstrate cut off lamellae at the edges of the cavity assists recognition [37] (Fig. 6; see color insert), as does the presence of osteoclastlike cells within the cavity. More precise identification of osteoclasts can be achieved by histochemical techniques that demonstrate the presence of tartrate-resistant acid phosphatase, although this is not specific to osteoclasts [38,39]. Finally, it is not usually possible to identify those cavities that have resulted in trabecular perforation. Indirect assessment of erosion depth was first reported by Courpron et al. [40], based on the postulate that the interstitial width, (i.e., the distance between two bone structural units on opposite sides of a trabecula) is inversely proportional to erosion depth. In this model, interstitial width is calculated as the difference between the trabecular width and twice the wall width. However, the relationship between interstitial width and erosion depth is not a simple inverse one, as it is influenced by concomitant changes in wall width and trabecular width [41,42]. Another approach to the direct measurement of erosion depth was reported by Eriksen et al. [43]. In this method, the number of lamellae eroded beneath the bone surface is counted, and cavities are characterized according to the presence of osteoclasts, mononuclear cells, and preosteoblastic cells, these being specifically associated with increasing stages of completion of the resorptive phase. This approach depends critically upon the accurate identification, on morphological grounds, of these different cell types within resorption cavities; even in high quality histological sections this can be difficult, and in the author’s hands 24% of resorption cavities were excluded from measurement because of failure to classify by cell type or inability to define and count the eroded lamellae. This method has not been widely adopted by other groups, although some have used a simplified approach in which the eroded lamellae are counted without subdivision of

FIGURE 6

(Top) Resorption cavity in cancellous iliac crest bone stained by toluidine blue. (Bottom) Same resorption cavity viewed under polarized light. Note the cutoff collagen lamellae at the edges of the cavity. (See color plate.)

cavities by cell type [44]. The latter technique provides an estimate of the mean depth of cavities in all stages of completion and thus underestimates the final resorption depth. The computerized method developed by Garrahan et al. [45] involves reconstruction of the eroded bone surface by a curve-fitting technique (cubic spline) and provides measurements of mean and maximum erosion depth together with the area, surface extent, and number of cavities (Fig. 8). Because all resorption cavities are included in the measurements, the mean values for mean and maximum depth and for area considerably underestimate the final resorption depth and area. This method may also be performed using interactive reconstruction of the eroded bone surface [46]. Another modification is to include for measurement only those cavities that contain a thin layer of osteoid, ensuring that the final resorption depth has been achieved [46]; however, the number of such cavities that can be identified in a single biopsy is usually

961

CHAPTER 59 Bone Histomorphometry

extremely small. Interestingly, the values reported using this approach in normal human biopsies were approximately 17% lower than those obtained in the same biopsies by the technique of counting eroded lamellae and were more consistent with reported agerelated changes in trabecular width [47,48]. Further work is thus required to improve existing techniques for the measurement of erosion depth. The approaches most widely used at present measure the depth of all resorption cavities and therefore underestimate the completed erosion depth, whereas measurements made using Eriksen’s method probably overestimate the true value [49]. Although these limitations preclude accurate assessment of remodeling balance at present, measurement of erosion depth has generated valuable information about the pathophysiology of bone loss in untreated and treated disease states.

VIII. ASSESSMENT OF BONE STRUCTURE The importance of cortical and cancellous bone structure as a determinant of bone strength is well established, and there has been increasing interest in the quantitative assessment of bone microarchitecture. Changes in bone structure have important implications not only for bone strength and fracture risk but also for the timing and efficacy of treatment in osteoporosis. In particular, the potential for anabolic agents to restore cancellous and cortical bone architecture in patients with advanced bone loss is of considerable interest.

A. Structural Determinants of Bone Strength Structural determinants of the mechanical strength of bone include cortical width and porosity and, in cancellous bone, trabecular size, shape, connectivity, and anisotropy. Early approaches to the quantitative assessment of cancellous bone structure were based on direct or indirect measurements of trabecular width, separation, and number [50–54]. Direct measurements of trabecular width can be made using an eyepiece graticule or grid, but nowadays are most commonly performed by using computerized techniques [53,54]. These measurements provide information not only about the mean trabecular width but also about the distributions of trabecular width within individual biopsies. Calculation of trabecular width from area and perimeter measurements may also be performed [55], this approach being based on the assumption that the width of measured structures is small relative to their

length. Calculation of trabecular separation and number from trabecular width and bone area is based on a parallel plate model [55] that may often be inappropriate in cancellous bone. Nevertheless, these approaches have generated useful information and have stimulated the development of more sophisticated techniques for analysis of bone structure. Because all of the structural determinants of cancellous bone strength are three-dimensional characteristics, their assessment on conventional histological sections provides only indirect information about these qualities, and three-dimensional images are required for direct measurement of connectivity, anisotropy, and trabecular size and shape [56]. Although further studies are required to examine the relationship between structural indices obtained using two- and three-dimensional approaches, there are several lines of evidence to support the contention that measurements from two-dimensional sections are representative of three-dimensional structure [57,58].

B. Two-dimensional Approaches 1. STRUT ANALYSIS

Strut analysis is based on the definition of nodes and termini and the topological classification of trabeculae and struts. Garrahan et al. [5] have described a semiautomated procedure in which the binary image of a section is skeletonized and the different strut types are classified as shown in Fig. 7 (see color insert). The total length of each strut type may be expressed as a percentage of the total strut length or in absolute terms. Node-to-node and node-to-loop strut lengths are positively related to connectivity, whereas node-to-terminus and terminus-to-terminus strut lengths are inversely related. Because of the edge effect, termini may be created artefactually, whereas node-to-node and nodeto-loop struts are true indices of connectedness, although their number may be underestimated. Termini may also result from sectioning through a trabecular window in a connected structure. 2. STAR VOLUME

The star volume is defined as the mean volume of solid material or empty space that can be seen unobscured from a point of measurement chosen at random inside the material [59]. Its assessment in histological sections of bone was first reported by Vesterby [60] using the vertical section technique [3] and a cycloid test system. The method may be applied to measurement both of trabecular width (trabecular star volume) and trabecular separation (marrow space star volume),

962

JULIET E. COMPSTON

which a large proportion of the measured intercepts may hit the boundary rather than bone, resulting in underestimation of star volume [61].

Strut analysis

3. TRABECULAR BONE PATTERN FACTOR Nd.Nd

Nd.Tm Nd.Lp

Tm.Tm

FIGURE 7 Diagrammatic representation of different strut types in strut analysis. Cancellous bone is shown in gray. Lines represent the skeletonized axis of the original bone profile. Squares represent nodes (Nd) and termini (Tm) Terminus-to-terminus (Tm.Tm), node-to-loop (Nd.Lp), and node-to-terminus (Nd.Tm) strut types are illustrated. Reprinted from Croucher et al. [61] with permission. (See color plate.)

and theoretically provides an unbiased stereological approach to these indices. The method involves the generation of intercepts from random sampling points, with the cubed length of the intercepts being used in the calculation of star volume (Fig. 8; see color insert). The values generated by marrow star volume measurements are significantly influenced by biopsy size, particularly in poorly interconnected cancellous bone in

Marrow space star volume

Another method is based on the concept that patterns or structures can be defined by the relationship between convex and concave surfaces [62], with convexity indicating poor connectivity and concavity reflecting structural integrity. Using a computer-based system, convexity and concavity are assessed by measurement of the bone perimeter before and after computer-based dilatation of the trabecular surface; whereas thickening of convex structures increases their perimeter, the reverse applies to concave structures. The values obtained for trabecular bone pattern factor, which is calculated as the difference between perimeter measurements before and after dilatation divided by the corresponding difference in area, may be significantly influenced by the computer-based smoothing technique used, the degree of dilatation employed, and the magnification at which the measurements are performed. 4. FRACTAL ANALYSIS

Fractal objects are characterized by scale invariance or self-similarity over a wide range of magnifications so that any one piece of a fractal, if magnified sufficiently, resembles the intact object [63]. Fractal analysis has been described in a number of biological systems, including the bronchial tree and vascular networks; in bone, it has been applied to radiological images and histological sections of bone [64]. The value obtained for the fractal dimension is critically dependent on the magnification used for measurement, and the relationship between fractal dimension and connectivity in cancellous bone has not been established. Multidirectional fractal analysis can be used to assess structural anisotropy [65].

C. Three-dimensional Approaches

V*m.space = 4 × π/3√3/n

FIGURE 8

Binary image used in the assessment of marrow space star volume. The upper active region is defined by the green and blue lines and the lower region by red and blue lines. Blue and yellow squares represent grid points hitting the marrow space. Blue and yellow lines represent grid lines intercepting trabeculae and the edge of the active region. The white arrow shows the direction of the vertical axis. Reprinted from Croucher et al. [61] with permission. (See color plate.)

A number of techniques have been used to generate three-dimensional images of bone. These include reconstruction of serial sections, scanning and stereo microscopy, volumetric, high resolution, and microcomputed tomography, and magnetic resonance imaging [66,67]. The potential for imaging techniques such as magnetic resonance imaging and computed tomography to provide information about bone structure in vivo is an important and active area of current research. At present, such application of these approaches is restricted by limited resolution, partial volume effects, and noise. Nevertheless, such approaches enable direct assessment of connectivity using the Euler number, a topological

963

CHAPTER 59 Bone Histomorphometry

property based on the number of holes and number of connected components in an object [68]. This measurement can also be obtained using the ConnEuler method in which projections are made through parallel thin section pairs, or dissectors, spaced approximately 10–40 apart [69].

8.

9. 10.

IX. CONCLUSIONS AND FUTURE DEVELOPMENTS 11.

Bone histomorphometry is a valuable tool in the assessment of metabolic bone diseases, in terms both of their diagnosis and pathophysiology. It also provides information about the safety and mechanisms of action of pharmacological interventions used in the treatment of bone diseases. In the future, the advances in molecular biological techniques and our knowledge of bone cell biology should enable more accurate identification of the processes of activation and resorption in situ, leading to a better understanding of mechanisms of bone loss and bone gain. In addition, advances in the in vivo assessment of cortical and cancellous bone architecture will provide new insights into structural determinants of bone strength and the ability of anabolic skeletal agents to restore bone architecture in patients with advanced bone loss. Finally, the importance of changes in cortical bone, both in untreated and treated disease, is increasingly recognized and this previously neglected area of bone histomorphometry is likely to be more intensively studied in forthcoming years.

12.

13.

14.

15. 16.

17. 18.

References 1. Rao DS 1983 Practical approach to bone biopsy. In Recker R (ed) Bone Histomorphometry: Techniques and Interpretation. CRC Press, Boca Raton, Florida, pp. 3–11. 2. Parfitt AM 1983 The physiological and clinical significance of bone histomorphometric data, In Recker R (ed) Bone Histomorphometry: Techniques and Interpretations. CRC Press, Boca Raton, Florida, pp. 143–224. 3. Baddeley AJ, Gundersen HJG, Cruz Drive LM 1986 Estimation of surface area from vertical sections. J Microsc 142:259–276. 4. Vesterby A, Kragstrup J, Gundersen HJG, Melsen F 1987 Unbiased stereologic estimation of surface density in bone using vertical sections. Bone 8:13–17. 5. Garrahan NJ, Mellish RWE, Compston JE 1986 A new method for the analysis of two-dimensional trabecular bone structure in human iliac crest biopsies. J Microsc 142:341–349. 6. Compston JE, Garrahan NJ, Croucher PI, Yamaguchi K 1993 Quantitative analysis of trabecular bone structure. Bone 14:187–192. 7. Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H, Meunier PJ, Ott SM, Recker RR 1987 Bone histomorphometry.

19.

20. 21. 22.

23.

24.

Standardization of nomenclature, symbols, and units. J Bone Miner Res 2:595–610. Schwartz MP, Recker RR 1981 Comparison of surface density and volume of human iliac trabecular bone measured directly and by applied sterology. Calcif Tissue Int 33:561–565. Frost HM 1963 Mean formation time of human osteons. Can J Biochem Physiol 41:1307–1310. Agerbaek MO, Eriksen EF, Kragstrup J, Mosekilde LE, Melsen F 1991 A reconstruction of the remodelling cycle in normal human cortical iliac bone. Bone Miner 12: 101–112. Bell NL, Loveridge N, Reeve J, Thomas CD, Feik SA, Clement JG 2001. Super-osteons (remodeling clusters) in the cortex of the femoral shaft: influence of age and gender. Anat Rec 264:378–386. Jordan GR, Loveridge N, Bell KL, Power J, Rushton N, Reeve J 2000 Spatial clustering of remodeling osteons in the femoral neck cortex: a cause of weakness in hip fracture? Bone 26: 305–313. Power J, Noble BS, Loveridge N, Bell KL, Rushton N, Reeve J 2001. Osteocyte lacunar occupancy in the femoral neck cortex: an association with cortical remodeling in hip fracture cases and controls. Calcif Tissue Int 69: 13–19. Vedi S, Bell KL, Loveridge N, Garrahan N, Purdie DW, Compston JE 2003 The effects of hormone replacement therapy on cortical bone in postmenopausal women: a histomorphometric study. Bone 33:330–334. de Vernejoul MC, Kuntz D, Miravet L, Goutalier D, Ryckewaert A 1981 Histomorphometric reproducibility in normal patients. Calcif Tissue Int 33:369–374. Chavassieux PM, Arlot ME, Meunier PJ 1985 Intermethod variation in bone histomorphometry: Comparison between manual and computerized methods applied to iliac bone biopsies. Bone 6:211–219. Wright CDP, Vedi S, Garrahan NJ, Stanton M, Duffy SW, Compston JE 1992 Combined inter-observer and inter-method variation in bone histomorphometry. Bone 13:205–208. Compston JE, Crowe JP, Wells IP, Horton LWL, Hirst D, Merrett AL, Woodhead JS, Williams R 1980 Vitamin D prophylaxis and osteomalacia in chronic cholestatic liver disease. Dig Dis 25:28–32. Eventov I, Frisch B, Cohen Z, Hammel I 1991 Osteopenia, hematopoiesis, and bone remodelling in iliac crest and femoral biopsies: A prospective study of 102 cases of femoral neck fractures. Bone 12:1–6. Compston JE, Croucher PI 1991 Histomorphometric assessment of trabecular bone remodeling in osteoporosis. Bone Miner 14:91–102. Frost HM 1969 Tetracycline-based histological analysis of bone remodelling. Calcif Tissue Int 3:211–237. Parfitt AM, Foldes J, Villanueva AR, Shih MS 1991 Difference in length between demethylchlortetracycline and oxytetracycline: Implications for the interpretation of bone histomorphometric data. Calcif Tissue Int 48: 74–77. Compston JE, Vedi S, Webb A 1985 Relationship between toluidine blue-stained calcification fronts and tetracycline labeled surfaces in normal human iliac crest biopsies. Calcif Tissue Int 37:32–35. Vedi S, Compston JE 1984 Direct and indirect measurements of osteoid seam width in human iliac crest biopsies. Metab Bone Dis Related Res 5:269–274.

964 25. Parfitt AM 1990 Osteomalacia and related disorders. In: Krane SM (ed) Metabolic Bone Disease, 2 Ed., Gnine & Stratton, New York. 25a. Glorieux FH, Travers R, Taylor A, Bowen JR. Rauch F, Norman M, Parfitt AM 2000 Normative data for iliac bone histomorphometry in growing children. Bone 26:103–109. 26. Frost HM 1983 Bone histomorphometry: Analysis of trabecular bone dynamics. In Recker R (ed) Bone Histomorphometry: Techniques and Interpretations. CRC Press, Boca Raton, Florida, pp. 109–131. 27. Foldes J, Shih M-S, Parfitt AM 1990 Frequency distributions of tetracycline-based measurements: Implications for the interpretation of bone formation indices in the absence of double-labeled surfaces. J Bone Miner Res 5:1063–1067. 28. Peach H, Compston JE, Vedi S, Horton LWL 1982 The value of plasma calcium, phosphate and alkaline phosphatase in the diagnosis of histological osteomalacia. J Clin Pathol 35:625–630. 29. Boyce BF, Fogelman I, Ralston S, Johnston E, Ralston S, Boyle IT 1984 Focal osteomalacia due to low-dose diphosphonate therapy in Paget’s disease. Lancet 1:821–824. 30. Adamson BB, Gallacher SJ, Byars J, Ralston SH, Boyle IT, Boyce BF 1993 Mineralization defects with pamidronate therapy for Paget’s disease. Lancet 342:1459–1460. 31. Frost HM 1983 Bone histomorphometry: Choice of marking agent and labeling schedule. In Recker R (ed) Bone Histomorphometry: Techniques and Interpretations. CRC Press, Boca Raton, Florida, pp. 37–51. 32. Arlot M, Edouard C, Meunier PJ, Neer RM, Reeve J 1984 Impaired osteoblast function in osteoporosis: Comparison between calcium balance and dynamic histomorphometry. Br Med J 289:517–520. 33. Frost HM 1985 The pathomechanics of osteoporosis. Clin Orthop Related Res 200:198–225. 34. Lips P, Courpron P, Meunier PJ 1978 Mean wall thickness of trabecular bone packets in the human iliac crest: Changes with age. Calcif Tissue Res 26:13–17. 35. Derkz P, Birkenhager-Frenkel DH 1995 A thionin stain for visualizing bone cells, mineralizing fronts and cement lines in undecalcified bone sections. Biotech Histochem 70:70–74. 36. Steiniche T, Eriksen EF, Kudsk H, Mosekilde L, Melsen F 1992 Reconstruction of the formative site in trabecular bone by a new, quick, and easy method. Bone 13:147–152. 37. Vedi S, Tighe JR, Compston JE 1984 Measurement of total resorption surface in iliac crest trabecular bone in man. Metab Bone Dis Related Res 5:275–280. 38. Burstone MS 1959 Histochemical demonstration of acid phosphatase activity in osteoclasts. J Histochem Cytochem 7:39–41. 39. Evans RA, Dunstan CR, Baylink DJ 1979 Histochemical identification of osteoclasts in undecalcified sections of human bone. Miner Electrolyte Metab 2:179–185. 40. Courpron P, Lepine P, Arlot M, Lips P, Meunier PJ 1980 Mechanisms underlying the reduction with age of the mean wall thickness of trabecular basic structure unit (BSU) in human iliac bone. In Jee WSS, Parfitt AM (eds) Bone histomorphometry, 3rd International Workshop. Armour Montagu, Paris, pp. 323–329. 41. Croucher PI, Mellish RWE, Vedi S, Garrahan NJ, Compston JE 1989 The relationship between resorption depth and mean interstitial bone thickness: Age-related changes in man. Calcif Tissue Int 45:15–19. 42. Parfitt AM, Foldes J 1991 The ambiguity of interstitial bone thickness: A new approach to the mechanism of trabecular thinning. Bone 12:119–122.

JULIET E. COMPSTON

43. Eriksen EF, Gunderson HJG, Melsen F, Mosekilde L 1984 Reconstruction of the resorptive site in iliac trabecular bone; a kinetic model for bone resorption in 20 normal individuals. Metab Bone Dis Related Res 5:235–242. 44. Palle S, Chappard D, Vico L, Riffat G, Alexandre C 1989 Evaluation of the osteoclastic population in iliac crest biopsies from 36 normal subjects: A histoenzymologic and histomorphometric study. J Bone Miner Res 4:501–506. 45. Garrahan NJ, Croucher PI, Compston JE 1990 A computerised technique for the quantitative assessment of resorption cavities in trabecular bone. Bone 11:241–246. 46. Cohen-Solal ME, Shih M-S, Lundy MW, Parfitt AM 1991 A new method for measuring cancellous bone erosion depth: Application to the cellular mechanisms of bone loss in postmenopausal osteoporosis. J Bone Miner Res 6: 1331–1338. 47. Weinstein RS, Hutson MS 1987 Decreased trabecular width and increased trabecular spacing contribute to bone loss with ageing. Bone 8:137–142. 48. Mellish RWE, Garrahan NJ, Compston JE 1989 Age-related changes in trabecular width and spacing in human iliac crest biopsies. Bone Miner 6:331–338. 49. Parfitt AM 1991 Bone remodeling in type 1 osteoporosis (letter). J Bone Miner Res 6:95–97. 50. Wakamatsu E, Sissons HA 1969 The cancellous bone of the iliac crest. Calcif Tissue Res 4:147–161. 51. Whitehouse WJ 1974 The quantitative morphology of anisotropic trabecular bone. J Microsc 101:153–168. 52. Aaron JE, Makins NB, Sagreiya K 1987 The microanatomy of trabecular bone loss in normal aging men and women. Clin Orthop Related Res 215:260–271. 53. Clermonts ECGM, Birkenhager-Frenkel DH 1985 Software for bone histomorphometry by means of a digitizer. Comput Math Prog Biomed 21:185–194. 54. Garrahan NJ, Mellish RWE, Vedi S, Compston JE 1987 Measurement of mean trabecular plate thickness by a new computerized method. Bone 8:227–230. 55. Parfitt AM, Mathews CHE, Villanueva AR, Kleerekoper M, Frame B, Rao DS 1983 Relationship between surface, volume and thickness of iliac trabecular bone in aging and in osteoporosis: Implications for the microanatomic and cellular mechanism of bone loss. J Clin Invest 72:1396–1409. 56. Compston JE 1994 Connectivity of cancellous bone: Assessment and mechanical implications. Bone 15:463–466. 57. Feldkamp LA, Goldstein SA, Parfitt AM, Jesion G, Kleerekoper M 1989 The direct examination of bone architecture in vitro by computed tomography. Bone 4:3–11. 58. Odgaard A, Gundersen HJG 1993 Quantification of connectivity in cancellous bone with special emphasis on 3-D reconstruction. Bone 14:173–182. 59. Serra J 1982 Image Analysis and Mathematical Morphology. Academic Press, London. 60. Vesterby A 1990 Star volume of marrow space and trabeculae in iliac crest: Sampling procedure and correlation to star volume of first lumbar vertebra. Bone 11:149–155. 61. Croucher PI, Garrahan NJ, Compston JE 1996 Assessment of cancellous bone structure: Comparison of strut analysis, trabecular bone pattern factor and marrow space star volume. J Bone Miner Res 11:955–961. 62. Hahn M, Vogel M, Pompesius-Kempa M, Delling G 1992 Trabecular bone pattern factor—A new parameter for simple quantification of bone microarchitecture. Bone 13: 327–330. 63. Mandelbrot BB 1977 Fractals: Form, chance, and dimension. Freeman, San Francisco.

CHAPTER 59 Bone Histomorphometry

64. Weinstein RS, Majumdar S, Genant HK 1992 Fractal geometry applied to the architecture of cancellous bone biopsy specimens. Bone 13:A38. 65. Jacquet G, Ohley WJ, Mont MA, Siffert R, Schmukler R 1990 Measurement of bone structure by fractal dimension. Proc Ann Conf IEEE/EMBS 12:1402–1403. 66. Majumdar S, Genant HK 1995 A review of the recent advances in magnetic resonance imaging in the assessment of osteoporosis. Osteoporosis Int 5:79–92. 67. Genant HK, Engelke K, Fuerst T, Gliier C-C, Grampp S, Harris ST, Jergas M, Lang T, Lu Y, Majumdar S, Mathur A, Takada M 1996 Noninvasive assessment of bone mineral and structure: State of the art. J Bone Miner Res 11:707–730.

965 68. De Hoff RT, Aigeltinger EH, Craig KR 1972 Experimental determination of the topological properties of three-dimensional micro-structures. J Microsc 95:69–91. 69. Gundersen HJG, Boyce RW, Nyengaard JR, Odgaard A 1993 The ConnEulor: Unbiased estimation of connectivity using physical dissectors under projection. Bone 14:217–222. 70. Vedi S, Compston JE, Webb A, Tighe JR 1982 Histomorphometric analysis of bone biopsies from the iliac crest of normal British subjects. Metab Bone Dis Related Res 4:231–236. 71. Vedi S, Compston JE, Webb A, Tighe JR 1983 Histomorphometric analysis of dynamic parameters of trabecular bone formation in the iliac crest of normal British subjects. Metab Bone Related Res 5:69–74.

CHAPTER 60

Radiology of Rickets and Osteomalacia JUDITH E. ADAMS

I. II. III. IV. V.

Clinical Radiology, Imaging Science and Biomedical Engineering, Stopford Building, Oxford Road, The University, Manchester, England, United Kingdom

Introduction and Historical Aspects Vitamin D Deficiency Renal Osteodystrophy Renal Tubular Defects and Hypophosphatemia Acidemia

I. INTRODUCTION AND HISTORICAL ASPECTS Bone resorption is a one-stage process, with osteoclasts resorbing mineral and osteoid together. In contrast, bone formation occurs in two stages: osteoblasts lay down osteoid, which subsequently becomes mineralized. The mineralization of bone matrix depends on the presence of adequate supplies of not only vitamin D, in the form of its active metabolite 1,25-dihydroxyvitamin D [1,25(OH)2D], but also calcium, phosphorus, and alkaline phosphatase, and on a normal pH prevailing in the body environment. If there is a deficiency of these substances for any reason, or if there is severe systemic acidosis, then mineralization of bone will be defective. There will be a qualitative abnormality of bone (in contrast to osteoporosis, which is a quantitative abnormality of bone), with reduction in the mineral/osteoid ratio, resulting in rickets in children and osteomalacia in adults. In the immature skeleton, the radiographic abnormalities predominate at the growing ends of the bones, where endochondral ossification is taking place, giving the classic appearances of rickets. When the skeleton reaches maturity and the process of endochondral ossification has ceased, the defective mineralization of osteoid is evident radiographically as Looser’s zones (pseudofractures, Milkman’s fractures), which are pathognomonic of osteomalacia. Rickets and osteomalacia are therefore synonymous and represent the same disease process, but are the manifestation in either the growing or the mature skeleton. A large number of different diseases can cause the same radiological abnormalities of rickets and osteomalacia [1–3]. Rickets appeared when people began to live in cities during the industrial revolution. The first descriptions of the condition are attributed to Daniel Whistler, who in 1692, while a student at Merton College Oxford, wrote a thesis entitled, “Inaugural Medical Disputation in the VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

VI. VII. VIII. IX.

Differential Diagnoses Vitamin D Intoxication Technical Aspects of Imaging Conclusions References

Disease of English Children which is popularly termed Rickets” for his doctorate at the University of Leyden [4]. Glisson in 1650 wrote a treatise for the Royal College of Physicians on, “De Rachitide sive morbo puerili qui vulgo the Rickets dicitur” [5,6]. In 1666 a postmortem examination was reported by John Locke of a child with rickets who died of pneumonia [7]. The disease was well known in Europe at this time and was regarded as “the English disease.” The origin of the term “rickets” is still debated [8,9]. Fish oil was recognized as a popular cure for “chronic rheumatism” (osteomalacia) and limb deformities (rickets) in children [10]. Robert Darbey, a house surgeon and apothecary to the Manchester Infirmary, was quoted by Percival in 1783 [11] as saying that 50–60 gallons of cod-liver oil were issued every year to patients, usually for the treatment of chronic rheumatism [6]. In 1890 Palm [12] published an essay on the distribution of rickets around the world and noted that it was common in cities where people were deprived of sunlight. He suggested that rickets could be treated with sunlight, but did not proceed to perform the experiment. Others observed that the disease is more common in cities [13]. There was much confusion between conditions that had similar clinical features but different courses of progression and responses to therapies of the day [14,15]. It was the microscopic studies of Pommer in 1885 [16] which distinguished, for the first time, between rickets, achondroplasia, and osteogenesis imperfecta. In 1895 Roentgen discovered X-rays, and it then became possible to display the radiographic features of rickets and osteomalacia. The unraveling of the structure and function of vitamin D and its metabolites during the twentieth century has elucidated the causes for confusion that existed in the past as to the etiology of rickets and the variable response to treatment (see Chapter 1). Copyright © 2005, Elsevier, Inc. All rights reserved.

968 Vitamin D deficiency may occur as a consequence of simple nutritional deficiency (diet, lack of sunlight; see Chapters 65 and 66), due to malabsorption states, chronic liver disease which affects hydroxylation at the 25 position (see Chapter 75), and chronic renal disease in which the active metabolite 1,25(OH)2D is not produced (see Chapter 76). Consequently, a large variety of diseases may result in vitamin D deficiency [17–19] (see Chapter 46). The radiological features of all will be similar, being those of rickets and osteomalacia. This similarity of radiological features but variations in response to treatment contributed to some of the early confusion [2,20,21]. Rickets due to nutritional deprivation was cured by ultraviolet light or physiological doses of vitamin D (400 IU per day), but rickets associated with chronic renal disease was not cured, except if very large pharmacological doses (up to 300,000 IU per day) were used. This lead to the terms “refractory rickets” and “vitamin D resistant rickets” being used for these conditions. In these groups were included the diseases that caused the clinical and radiological features of rickets but were related to phosphate, not vitamin D, deficiency, such as X-linked hypophosphatemia (see Chapter 70) and genetic diseases involving defects in 1-hydroxylase (Chapter 71) and the vitamin D receptor (Chapter 72).

II. VITAMIN D DEFICIENCY The hormone 1,25(OH)2D plays an important role in calcium homeostasis by its actions principally on the bone and intestine, but also on the kidney and parathyroid glands. It promotes the intestinal absorption of calcium and phosphorus from the intestine. On the bone it has two actions; one is to mobilize calcium and phosphorus from the skeleton as required, and the other is to promote maturation and mineralization of the osteoid matrix. Deficiency of vitamin D results in rickets in children and osteomalacia in adults [22]. There are known to be seasonal variations in vitamin D status, with plasma levels being lower in the winter months [23] (see also Chapters 3 and 47). The pathophysiology, clinical descriptions and treatments are discussed elsewhere in this volume.

A. Rickets In the growing skeleton, the effect of vitamin D deficiency and consequent defective mineralization of osteoid is seen at the growing ends of the bones [24–26]. In the early phase, there is widening of the growth plate, which is the radiolucent (unmineralized) gap between the mineralized metaphysis and epiphysis [27,28].

JUDITH E. ADAMS

As the changes become more severe, there is “cupping” of the metaphysis with irregular and poor mineralization (Figs. 1– 4). There is some expansion in width of the metaphysis which results in the apparent soft tissue swelling around the ends of the long bones affected. This produces the expansion at the anterior ends of the ribs referred to as a “rachitic rosary” (Fig. 1C) [29]. There is often a thin “ghost-like” rim of mineralization at the periphery of the metaphysis, since this mineralization occurs by membranous ossification at the periosteum. The margin of outline of the epiphysis also appears indistinct as endochondral ossification at this site is also defective. A method for scoring the severity of rickets has been described (30). The changes are most pronounced at the sites of bone that are growing most actively. These sites, in sequence, are around the knee, the wrist (particularly the ulna, Fig. 3A), the anterior ends of the middle ribs, the proximal femur, and the distal tibia. It is these anatomical sites that should be radiographed if rickets is suspected. As rachitic bone is soft and bends, additional features that develop, once the child begins to walk, are bowing of the legs (genu valgum) or knockknees (genu varum), deformity of the hips (coxa valga or, more usually, coxa vara), in-drawing of the ribs at the insertion of the diaphragm (Harrison’s sulcus), and protrusio acetabulae and tri-radiate deformity of the pelvis. The latter can result in problems with parturition at subsequent pregnancies. Involvement of the bones of the thorax and respiratory tract (larynx and trachea) can result in stridor and respiratory distress [31,32]. Paradoxically, in very severe rickets, where little growth is taking place (i.e., owing to nutritional deprivation or chronic ill health), the features of rickets may not be evident at the pubertal growth plate [33]. In mild vitamin D deficiency, the radiographic features of rickets may only become apparent during the pubertal growth spurt associated with puberty, and then the changes are most prominent at the knee. The radiographic changes may be quite subtle and not involve the entire metaphysis (Fig. 2). The causes, presentation, and features of rickets may vary according to the age of onset [26]. In the rickets of prematurity, there may be little abnormality at the metaphysis, as no skeletal growth is taking place in the premature neonate. However, the bones are osteopenic and prone to fracture. Periosteal reaction is probably due to the accumulation of unmineralised osteoid on the periosteal surface [34]. The cause in preterm infants is an inadequate supply of phosphorus and calcium during periods in the hospital, or those receiving unsupplemented human milk. In young infants, vitamin D levels are closely related to maternal vitamin D status. Although there has been a

969

CHAPTER 60 Radiology of Rickets and Osteomalacia

A

B

C

FIGURE 1

Nutritional vitamin D deficiency rickets in young children. Radiograph of the wrist in a child of approximately 2 years (A) and 3.5 years (B) showing widening of the growth plate with cupping and expansion of the metaphyses, which are poorly mineralized. The expansion results in soft tissue swelling around the ends of the bones. The cortical “tunneling” (subcortical erosion) and hazy trabecular pattern indicate secondary hyperparathyroidism. (C) Chest radiograph in an infant showing the bulbous expansion of the anterior ends of the ribs (arrows) known as a “rachitic rosary.”

A B

C

FIGURE 2

Nutritional rickets in Asian adolescents. (A) Antero-posterior (AP) radiograph of the knee showing widening of the radiolucent growth plate due to poor mineralization and some splaying and cupping of the metaphyses. In other individuals in whom the vitamin D deficiency was mild, the features of rickets only became apparent clinically and radiologically during the growth spurt of puberty in the knee (B) and ankle (C). The changes may be quite subtle and not involve the entire metaphysis, as illustrated. The widened growth plate is evident only at the medial aspect of the metaphyses of the distal femur and proximal tibia (B), and in the lateral aspect of the distal tibia, and in the fibula (C). There is not the cupping, splaying, and irregular and poor mineralization of the metaphyses, which is evident in more florid rickets.

970

JUDITH E. ADAMS

A

B

C

FIGURE 3 Rickets and the effect of treatment. (A) Wrist before treatment. The radiolucent growth plate is increased in width, and there is some cupping of the metaphyses of the distal radius and ulna with irregular and deficient mineralization. Note that the ulnar plate is more severely affected than the radial plate. This is the consequence of the ulna growing in length exclusively from its distal end, the proximal end forming the olecranon. This is the most sensitive site for assessing for the radiographic presence of rickets. (B) Wrist 4 months after treatment with vitamin D. There is healing of the rickets, although the distal segments of the radius and ulna are still reduced in density. This indicates the stage of bone development at which the rickets occurred and the amount of growth that has taken place since then. With time and remodeling, the appearances will become normal. (C) A series of radiographs of the wrist taken over a period of 4 months following treatment with vitamin D; left image: there are the characteristic features of rickets with widening of the growth plates of the radius and ulna and poor mineralization of the metaphyses. Middle image at approximately 6–8 weeks: the radiolucent growth plate is less wide and there is increased mineralization of the metaphyses, but minor abnormalities are still present. Right image: the abnormalities previously present have now disappeared almost completely with healing of the rachitic changes.

971

CHAPTER 60 Radiology of Rickets and Osteomalacia

FIGURE 4 AP radiograph of the legs of a young child with severe rickets. The radiolucent growth plate is increased in width, the metaphyses are splayed and poorly mineralized. There are also Looser’s zones (pseudo fractures) evident as radiolucent lines though the distal fibula bilaterally (arrows). These have some radiodense callus formation, indicating that vitamin D therapy has already commenced. There is also some periosteal reaction evident. This is thought to be caused by the accumulation of unmineralized osteoid on the periosteal surface of bone which lifts the periosteum, stimulating it to mineralize.

decline in the incidence of rickets, with improved social and environmental awareness, vitamin D deficiency remains a significant public health problem in the UK among children of Southeast Asian and Middle Eastern immigrants as a result of a number of factors. These include diminished cutaneous vitamin D synthesis due to migration and residency in more northern latitudes where the sun is lower in the sky, and to voluntary avoidance of sunshine due to religious and cultural practices. There are the additional factors of the rachitic role of vegetarian diets and prolonged breastfeeding without vitamin D supplementation [24–26]. In the newborn and young

infant, softening of skull bones may result in craniotabes [35] and frontal bossing. Depending on the age of onset, there will be effects on the teeth (delay in dental eruption and enamel hypoplasia). If the vitamin D deficiency is treated appropriately, then the radiographic abnormalities of rickets will heal over about 2–3 months (Fig. 3). The radiographic features of healing will lag behind the improvement in biochemical parameters (2 weeks) and clinical symptoms. With treatment, the unmineralized osteoid of the growth plate of the metaphysis and epiphysis will mineralize. This section of abnormal bone may be visible for a period of time and gives some indication as to the age of onset and duration of the period of rickets (Fig. 3). Eventually, this zone becomes indistinguishable from the normal bone with time and remodeling. The zone of provisional calcification that was present at the onset of the disturbance to endochondral ossification may remain (Harris growth arrest line) (Figs. 6 and 12A) as a marker of the age of skeletal maturation at which the rickets occurred [36]. However, this is not specific for rickets and can result from any condition (i.e., a period of ill health, lead poisoning) that inhibits normal endochondral ossification. There is evidence of retarded growth and development in rickets, but this is more marked when the vitamin D deficiency is associated with chronic diseases that reduce calorie intake, general well-being, and activity (i.e., malabsorption, chronic renal disease) than with simple nutritional vitamin D deficiency (J. E. Adams, personal observation). With vitamin D deficiency, there is associated hypocalcemia. To maintain calcium homeostasis, the parathyroid glands are stimulated to secrete parathyroid hormone (PTH). This results in another important feature of vitamin D deficiency rickets [37]. Evidence of secondary hyperparathyroidism, with increased osteoclastic resorption (erosions, bone cysts), is always evident histologically (see Chapter 59), although radiographically evident features are uncommon [38–40], and cystic lesions of bone (brown tumors) are rare [41].

B. Osteomalacia At skeletal maturity, the epiphysis fuses to the metaphysis, and longitudinal bone growth ceases. However, bone turnover continues throughout life to maintain the tensile integrity of the skeleton. Vitamin D deficiency in the adult skeleton results in osteomalacia, the pathognomonic feature of which is the Looser’s zone (pseudofracture, Milkman’s fracture) [42–47]. Looser’s zones are radiolucent areas in the bone that are composed of unmineralized osteoid. They appear as radiolucent lines

972

JUDITH E. ADAMS

A

B

D

C

E

F

G

973

CHAPTER 60 Radiology of Rickets and Osteomalacia

in the bone that are perpendicular to the bone cortex, do not extend across the entire bone shaft, and characteristically have a sclerotic margin (Fig. 5) [48]. They can occur in any bone, but typically are found in the medial portion of the femoral neck, the pubic rami, the lateral border of the scapula, and the ribs. They may involve the first and second ribs, in which traumatic fractures are uncommon and are usually associated with severe trauma. Other less common sites are the metatarsals and metacarpals, the base of the acromium, and the ilium (Figs. 5D, E, G). Again, traumatic fractures of the ilium require very severe trauma, and a lack of history of such severe injury should alert one to the fact that a “fracture” of the ilium may be related to vitamin D deficiency. The etiology of why Looser’s zones occur in the anatomical sites that they do has been much debated [49,50]. At one time, it was thought that they were in the sites of vascular channels, but this theory has been discarded. Their position is most likely to be related to sites of stress in the skeleton. Looser’s zones must be differentiated from insufficiency fractures that occur in osteoporotic bone, particularly in the pubic rami, sacral ala, and calcaneum [51–53]. Insufficiency fractures consist of multiple microfractures and often have florid callus formation, which differentiates them from Looser’s zones [54]. Incremental fractures occur in Paget’s disease of bone and resemble Looser’s zones in appearance (translucent zone, suggesting incomplete fracture, with sclerotic margin), but these tend to occur on the outer (convex) cortex of the bone involved. The typical features of Paget’s disease (sclerotic, disordered trabecular pattern, enlarged bone) serve as distinguishing radiographic features [55]. Complete fractures can occur through Looser’s zones. As in rickets, osteomalacic bone is soft and bends. This is evident by protrusio acetabulae, in which the femoral head deforms the acetabular margin so that the normal “teardrop” outline of the acetabulum is lost (Figs. 5A and B). There may be bowing of the long bones of the legs and tri-radiate deformity of the pelvis (Fig. 6), particularly if the cause of the vitamin D

deficiency has persisted since childhood and has been inadequately treated or untreated. As in rickets, secondary hyperparathyroidism is present and can be manifested radiographically as subperiosteal erosions, particularly in the phalanges, but at other sites also (sacroiliac joints, symphysis pubis, proximal tibiae, outer ends of the clavicle, and “pepper-pot” skull), depending on the intensity of the hyperparathyroidism (Fig. 7). There can also be cortical tunneling and a “hazy” trabecular pattern (Fig. 7B). There may be generalized osteopenia, and vertebral bodies may have concave end plates. This is due to softening of the malacic bone, which is deformed by the cartilagenous intervertebral disc (“codfish” deformity). The etiology of this deformation is different from that which results in endplate irregularity in osteoporosis, in which microfractures occur owing to the bone being brittle rather than soft.

C. Secondary Hyperparathyroidism The most sensitive site for the radiographic features of hyperparathyroidism is the radial sides of the middle phalanges of the second and third fingers. There are characteristic erosions along the cortical surface of these bones (Figs. 7A–C). If there is florid hyperparathyroidism, then cortical erosions may be seen more widely and involve the distal phalanges (Fig. 7A), the outer ends of the clavicle (Fig. 7D), the symphysis pubis, the sacroiliac joints, the upper medial cortex of the tibiae, and the skull vault (“pepper-pot” skull) (Figs. 7E–G). The erosions of hyperparathyroidism in the sacroiliac joints tend to involve the iliac margin of the joint, in contrast to the involvement of both joint surfaces in inflammatory and erosive arthritides (Figs. 7F and 11D). The erosions of hyperparathyroidism in the skull vault (pepper-pot skull) must be differentiated from the “granularity” of the parietal region of the skull vault, which may be a variant of normal. In the former, the margins of the erosions are indistinct; in the latter they

FIGURE 5 Osteomalacia. (A) Nutritional osteomalacia in an Asian. AP radiograph of the right hip. This shows a classic Looser's zone in the medial cortex of the femoral neck. Looser’s zones are generally radiolucent lines that are perpendicular to the bone cortex, do not extend fully across the bone (unless a fracture has occurred through the site), and have sclerotic margins as illustrated. There is also some protrusio acetabulae due to softening of the bone. (B) Pelvis in untreated celiac disease. Bilateral protrusio acetabulae and tri-radiate deformity of the pelvis are apparent. There are Looser's zones through both superior and inferior pubic rami (arrows), with complete fracture through that in the left superior pubic ramus. (C) Nutritional osteomalacia. Chest radiograph left apical area: there is a Looser's zone through the entire width of the left first rib (arrow). Traumatic fractures through the first and second rib normally occur only with severe trauma. (D) Malabsorption osteomalacia. In the pelvis there is an extensive Looser's zone through the left ilium and the right pubic ramus (arrows). There is evidence of pinning of a previous left hip fracture. Severe trauma is required to fracture the ilium. Malabsorption osteomalacia (celiac disease): (E) Chest radiograph with Looser's zones in the lateral border of the right scapula and at the base of the acromium (arrows). (F) Forearm with Looser’s zones with sclerotic margins in both the radius and ulna. Nutritional vitamin D deficiency (G) hand radiograph with Loozer’s zone in the mid shaft of the third metacarpal.

974 A

JUDITH E. ADAMS

B

FIGURE 6 Stigmata of rickets in childhood. (A) Pelvis showing stigmata of past rickets with a tri-radiate deformity of the pelvis and varus deformity of the femoral necks. (B) Bowing of tibia and fibula with evidence of Harris growth arrest lines in the distal shaft of the left tibia (arrows).

are distinct and corticated. Loss of the lamina dura around the teeth does occur but is not specific to hyperparathyroidism, occurring also in other conditions such as Paget’s disease of bone and dental infection. With intense hyperparathyroidism, there is an increase in cortical “tunneling” due to bone resorption and an indistinct, “hazy” trabecular pattern. Erosions may occur along the growth plate and result in displacement of the epiphysis from the metaphysis of the shaft of the bone. This is most likely to occur in association with chronic renal disease (see Section III), since the intensity of the hyperparathyroidism secondary to vitamin D deficiency is related to the duration and severity of the hypocalcemia which acts as the stimulant. Although the changes of rickets occur predominantly at the growth plates, Looser’s zones may also, but rarely, be present in the juvenile skeleton (Fig. 4) [56].

III. RENAL OSTEODYSTROPHY The bone disease that occurs in chronic renal impairment, namely, renal osteodystrophy or uremic

(azo-temic) osteodystrophy, is complex and multifactorial [57–60], and has changed in both clinical and imaging features over the past three decades [60]. Previously they occurred a combination of vitamin D deficiency, which resulted in rickets and osteomalacia, and hypocalcemia [61,62], the latter inducing severe secondary hyperparathyroidism that stimulated osteoclastic resorption of bone [63–65]. This resulted radiographically in subperiosteal erosions, most frequently identified along the radial aspect of the middle phalanx of the second and third fingers (Figs. 7A and C). Because the stimulus to secondary hyperparathyroidism in chronic renal disease was intense and long standing, the skeletal manifestations were often extensive and manifest radiographically not only as sub-periosteal erosions in the hands, but other features also (e.g “pepper pot” skull, brown “tumors” causing bone cysts, osteosclerosis and metastatic calcification). However, these features are now rarely evident on radiographs. This has occurred through the better understanding of vitamin D metabolism and improvements in therapeutic management of patients with chronic renal impairment (calcitriol, 1α vitamin D, renal transplantation,

C

B A

E D

F G

FIGURE 7 Erosions of hyperparathyroidism. (A) Acro-osteolysis and resorption of the distal phalanges that results in pseudo-clubbing. There are also subperiosteal erosions in the radial cortex of the middle phalanges of the second, third, and fourth fingers (renal osteodystrophy). (B) Intracortical tunneling of the cortex of the phalanges, with subperiosteal erosions in the radial cortex of the middle phalanx of the index finger. (C) Extensive subperiosteal erosions along the radial border of the middle phalanx of the third finger. There is adjacent metastatic vascular calcification in the digital artery, indicating phosphate retention and azotemic osteodystrophy. (D) Erosions in the outer end of the clavicle. (E) Lateral skull showing erosions in skull vault giving pepper-pot appearance. (F) Erosions along the iliac margin of the left sacro-iliac joint. (G) Erosions in the syphysis pubis (coronal tomogram).

976

JUDITH E. ADAMS

and dialysis). Metastatic calcification in soft tissues and “adynamic” bone disease (related to aluminum) remain problematic, and new complications have developed as a consequence of treatments (dialysis and transplantation), including amyloid deposition, noninfective sponyloarthropathy, osteonecrosis and osteopenia/ osteoporosis, all of which may have characteristic imaging features [60,66].

A. Hyperparathyroidism 1. SUBPERIOSTEAL EROSIONS

Subperiosteal erosions occur most commonly in the middle phalanges but are also present in the distal phalangeal tufts, causing acro-osteolysis and the clinical sign of “pseudo-clubbing” (Figs. 7A and C). Other sites of erosions include the outer end of the

clavicle (Fig. 7D), the medial aspect of the proximal portion of the tibia, humerus, and femur, and the superior and inferior borders of the ribs [67]. Erosions may also occur adjacent to joints, and consequent damage to the articular subchondral bone can cause symptomatic arthritis [68]. These erosions may simulate those that occur in rheumatoid arthritis (RA) but tend to be a little further from the joint margin and are not generally associated with joint narrowing, periarticular osteopenia, or soft tissue swelling, which are early features of RA [69]. Joints that can be affected include the acromioclavicular, sternoclavicular, sacroiliac, and the symphysis pubis (Figs. 7F and G). In the hand, the distal interphalangeal joints and ulnar aspect of the metacarpophalangeal joints can be involved [70]. Subperiosteal erosions of the phalanges are diagnostic of hyperparathyroidism. In children, erosions can occur in the region

B

A

FIGURE 8 Renal osteodystrophy in an adolescent in the past. (A) Pelvis showing rickets (widened growth plate at femoral metaphyses; varus deformity of femoral necks) and hyperparathryoidism (bone sclerosis and erosions in sacroiliac joints and along proximal femoral metaphyses), resulting in slipped upper femoral epiphysis at right. (B) Anteroposterior view of ankle showing extensive erosions of the metaphyses due to severe secondary hyperparathyroidism. With improved treatment of patients with chronic renal insufficiency (calcitriol, 1α vitamin D, renal transplantation, and dialysis), one should no longer see such cases of rickets and intense secondary hyperparathyroidism related to azotemic osteodystrophy.

977

CHAPTER 60 Radiology of Rickets and Osteomalacia

of the growth plate, causing radiographic abnormalities, which may be mistaken for rickets and which can result in slipped epiphysis and deformity [71,72] (Fig. 8). If the hyperparathyroidism is treated successfully, erosions will fill in, and the cortex will revert to its normal appearance. However, if there has been severe resorption of the distal phalangeal tufts, these cannot reconstitute to their normal shape and may remain shortened and stubby (Fig. 7A). Bone resorption can occur in the regions of insertion of tendons and ligaments, particularly the trochanters, the ischial and humeral tuberosities, the inferior aspect of the calcaneum, and around the elbow. Excessive bone resorption in the skull vault causes the mottled texture of alternating areas of lucency and sclerosis, referred to as a “salt and pepper” or “pepper-pot” skull (Fig. 7E). As the intensity of hyperparathyroidism is now much less intense and longstanding than previously, through the introduction of effective treatments (calcitriol, 1α vitamin D, renal transplantation and dialysis), there may be no features present radiographically, or only subperiosteal erosions in the phalanges. If there are no erosions in this site, then it is unlikely that they will be found in other sites; skeletal surveys previously performed in patients with chronic renal impairment are now inappropriate, particularly as parathyroid hormone levels can be measured directly; a hand radiograph would suffice.

A

B

2. INTRACORTICAL BONE RESORPTION

Intracortical bone resorption is caused by osteoclastic resorption of Haversian canals within the cortex of the bones. Radiographically this causes linear translucencies within the cortex (Fig. 7B). This feature is not specific for hyperparathyroidism and can be found in other disorders in which bone turnover is increased (e.g., Paget’s disease of bone). 3. OSTEOSCLEROSIS

Osteosclerosis occurs uncommonly in primary hyperparathyroidism, but was a common feature of disease secondary to chronic renal disease. Radiographically, the bones appeared increased in density (Fig. 9). This affected particularly the axial skeleton. In the vertebral bodies, the endplates were preferentially involved, giving bands of dense bone adjacent to the end plates with a central band of lower, normal bone density. These alternating bands of normal and sclerotic bone gave a striped pattern described as a “rugger jersey” spine (Figs. 9A and B). The osteosclerosis may be more generalized. It may result from excessive accumulation of poorly mineralized osteoid, which would appear more dense radiographically than normal bone. It was also suggested that it results from an exaggerated osteoblastic response following bone resorption [73].

C

FIGURE 9 Renal osteodystrophy: bone sclerosis of the lateral lumbar spine showing end plate sclerosis, giving a “rugger jersey” appearance of alternating bands of stripes in the lumbar (A) and thoracic (B) spine. (C) Hand radiograph showing Looser's zone at the base of the first metacarpal, indicating osteomalacia, and generalized osteosclerosis due to secondary hyperparathyroidism. With improved management of chronic renal disease, the intensity of secondary hyperparathyroidism is less than in previous decades.

978

JUDITH E. ADAMS

A

B

C

D

FIGURE 10 Brown “tumors” (osteitis fibrosa cystica) in hyperparathyroidism. (A) Oblique radiograph of right lower ribs showing a lytic area expanding the rib (arrow). In renal osteodystrophy: right knee radiograph lateral (B) and antero-posterior (C) projections showing well-defined lytic areas in the distal femur and the proximal fibula. The lesions are causing expansion of the bone with cortical destruction and soft tissue masses, radiological features of aggressive bone lesions. (D) Following parathryoidectomy and regression of the secondary hyperparathyroidism, the destructive lesions have filled in with woven bone and so have increased in density, and the cortex has reconstituted, although there is still expansion of the fibula. Such florid features of hyperparathyroidism are now infrequently seen with improvement in management.

979

CHAPTER 60 Radiology of Rickets and Osteomalacia

4. BROWN TUMORS

Brown tumors represent cavities within the bone in which there has been excessive osteoclastic resorption. Histologically, they are filled with fibrous tissue and osteoclasts, and may undergo necrosis and liquifaction. Radiographically, they appear as radiolucent cysts within the bone. They can occur anywhere in the skeleton and may cause expansion of bones (Fig. 10). They constitute the osteitis fibrosa cystica of hyperparathyroidism first described by Von Recklinghausen. When appropriate treatment was given, these bone cysts would will fill with woven bone and increase in density (Fig. 10D). 5. OSTEOPOROSIS

With excessive bone resorption, combined with defective mineralization, the bones can appear osteopenic (reduced in radiographic density) in some patients. 6. PERIOSTEAL REACTION

Periosteal new bone formation (radiodense line parallel to the petriosteal cortex of a bone) used to be observed in up to 25% of patients with renal osteodystrophy. It occurred most frequently in those with severe bone disease and is thought to be a manifestation of intense hyperparathyroidism. It occurs in the metatarsals, femur, and pelvis and less commonly in the humerus, radius, ulna, tibia, metacarpals, and phalanges [74,75].

B. Metastatic Calcification With the reduced glomerular function of chronic renal failure, there is phosphate retention [76]. This results in an increase in the calcium X phosphate product, and, as a consequence, amorphous calcium phosphate is precipitated in organs and soft tissues [77]. This metastatic calcification can occur in the eyes and skin, causing symptoms of “gritty,” sore eyes and itching, and, in severe cases, calciphylaxis (ischemic necrosis of soft tissue) [78]. Other organs in which calcium is deposited include the heart, stomach, kidneys, lung, and skeletal muscle, where it is rarely detected radiographically. However, radiographic evidence of metastatic calcification is seen in arteries and around the joints (Fig. 11). The periarticular calcification is more common around the large joints (hip, shoulder) (Figs. 11B and C), but can also occur around small joints. The calcification can involve the joint capsule or tendons but more usually lies in the bursae adjacent to joints and bony protuberances (ischial tuberosity). These metastatic

deposits of calcium will cause swelling and may be painful. They may increase rapidly in size and are sometimes erroneously diagnosed as “tumors” on initial clinical examination. Whereas rickets, osteomalacia, and secondary hyperparathyroidism have become much less evident radiographically in patients with chronic renal disease over the past three decades, the prevalence of metastatic calcification has perhaps become more frequent in recent years [79]. This may be due to a combination of excess phosphate not being removed effectively by dialysis, the increased use of calcium carbonate as a phosphate binder, and the occurrence of “adynamic” renal bone disease [80]. This may result in the skeleton being a less effective reservoir for calcium than normal, so that the calcium remains in the extracellular fluid. These calcific masses can regress with appropriate treatment (phosphate binders such as aluminum hydroxide or calcium carbonate) (Figs. 11B, C, and E). Initially, the masses may liquify, and apparent fluid levels are seen on radiographs taken with the patient in the erect position. These fluid levels are the result of the interface between serum and serum plus mineral. There can be complete regression of periarticular calcific masses with appropriate treatment; vascular calcification rarely regresses [81]. The metastatic calcification in end stage renal disease can involve the intimal layer of the coronary arteries. This is common, severe, and significantly associated with ischemic cardiovascular disease. The latter is the etiology of death in half the patients on dialysis. One of the exciting recent developments is the quantitative imaging by electron-beam computed tomography (EBT) of the calcification in coronary arteries and cardiac valves. The calcium in coronary arteries is highly correlated to myocardial infarction and angina in patients on dialysis. Electron beam tomography has the potential to identify those patients at highest risk of cardiovascular morbidity and mortality, and to monitor the change in cardiac calcification with time and the effectiveness of treatment [82,83].

C. Aluminum Toxicity Aluminum toxicity may occur in patients with chronic renal disease due to excessive ingestion of aluminum hydroxide taken to reduce serum phosphate. It also occurs in those patients on hemodialysis in which the dialysate water contains excessive amounts of aluminum [84–86]. The control of aluminum content in dialysate water in recent years has reduced the prevalence of this disorder [87]. Aluminum accumulates

980

JUDITH E. ADAMS

C

B A

D

E

FIGURE 11 Metastatic calcification in renal osteodystrophy. (A) Extensive vascular calcification in vessels of the foot. (B) Extensive “tumor-like” calcification around the right shoulder. There is also bowing of the proximal humerus due to osteomalacic bone softening. (C) Following treatment with a phosphate binder (aluminium hydroxide or calcium carbonate), these calcified masses can liquefy and regress, as illustrated. If the radiographs are taken with the patient in the erect position, there will be seen fluid/fluid mineral interface levels (arrows). (D) Pelvis showing extensive soft tissue calcification around the ischia and left hip. Widening of the sacroiliac joints indicates erosions of hyperparathyroidism. (E) Fingers showing (left) extensive metastatic calcification around the distal phalanx and subperiosteal erosions along the radial cortex of the middle phalanx. (Right) Following therapy with oral phosphate binder, there is reduction in the extent of the metastatic soft tissue calcification. Note the acro-osteolysis due to hyperparathyroidism. Although the features of osteomalacia and secondary hyperparathyroidism are now infrequently seen radiographically with improvement in management of chronic renal impairment, metastatic calcification is still prevalent. This may be due to a combination of ineffective removal of phosphate through dialysis, the use of calcium carbonate as a phosphate binder and “adynamic” bone disease.

at the bone/osteoid interface and inhibits mineralization. This results in rickets and osteomalacia, but the bone also becomes “adynamic,” with reduced osteoid production and turnover. Radiographically this results in reduced bone density and easy fracture [88].

These fractures can occur in unusual sites (second to fourth ribs, odontoid) or have an atypical appearance in long bones [88]. In patients with azotaemic osteodystrophy and aluminium toxicity, there is less osteosclerosis, fewer periosteal erosions and increase

981

CHAPTER 60 Radiology of Rickets and Osteomalacia

in the rate of osteonecrosis following transplantation, than occurs in those individuals with chronic renal failure but without excess aluminium [85].

IV. RENAL TUBULAR DEFECTS AND HYPOPHOSPHATEMIA Inorganic phosphate, glucose, and amino acids are absorbed in the proximal renal tubule; concentration and acidification of the urine in exchange for a fixed base occur at the distal renal tubule. Disorders of the renal tubules may involve either the proximal or distal tubule, or both. They will result in a spectrum of biochemical disturbances that may result in loss of phosphate, glucose, or amino acids alone, or in combination, with additional defects in urine acidification and concentration. These defects may be inherited and present from birth (de Toni-Fanconi syndrome, cystinosis, X-linked hypophosphatemia or XLH) [89] or acquired later in life (e.g., Wilson’s disease, hereditary tyrosinemia, neurofibromatosis, mesenychymal tumors, cadmium poisoning, drug induced [ifosfamide]) [90–92], by tubular function being interfered with either by crystal deposition (e.g., copper, in Wilson’s disease) or a humoral substance, such as is produced by tumors in tumor-induced osteomalacia (TIO) also known as “oncogenic” rickets [93]. It is the renal tubular disorders that cause phosphaturia, which results in rickets and osteomalacia [94]. As the serum calcium is generally normal in these diseases, secondary hyperparathyroidism does not occur (see Chapter 69). Hypophosphatemic rickets has also been described in association with the epidermal naevus syndrome [95].

A. X-linked Hypophosphatemia The genetic disorder XLH is transmitted as an X-linked dominant trait. Sporadic cases through spontaneous mutations also occur. An autosomal dominant variety with variable penetrance has also been described [96]. The incidence is approximately 1 in 25,000, and XLH is now the most common of genetically induced rickets [97–99]. The pathophysiology of XLH and its mode of treatment are discussed in detail in Chapter 69. The disease results through mutations in the PHEX gene (phosphate regulating gene with homologies to endopeptidases on the X chromosome, previously known as PEX), which codes for a type 2 membrane glycoprotein that activates the putative phosphate-regulating hormone known as “phosphatonin” [19,100] and is characterized by lifelong phosphaturia, hypophosphatemia,

and rickets and osteomalacia. Rickets becomes clinically evident around 12 to 18 months of age. The radiological features of XLH are characteristic. Although in the past affected patients were often short in stature with quite marked deformities (bow-legs), improvements in management have reduced these consequences of the disorder. 1. RICKETS

There is widening of the growth plate and defective mineralization of the metaphysis; however the metaphyseal margin tends to be less indistinct than in nutritional rickets, and there is less expansion in the width of the metaphysis (Fig. 12). Changes are most marked at the knee, wrist, ankle, and proximal femur. Healing does occur with appropriate treatment. The growth plates fuse normally at skeletal maturation. The bones are often short and undertubulated (shaft wide in relation to bone length), with bowing of the femur and tibia (Fig. 12D). 2. OSTEOMALACIA

After skeletal maturation, Looser’s zones persist in patients with XLH. These tend to occur in sites different from those in nutritional osteomalacia, and are often present in the outer cortex of the bowed femur (Fig. 12B), although they occur along the medial cortex of the shaft also [101]. Looser’s zones in the ribs and pelvis are rare. They may heal with appropriate treatment, but those that have been present for many years will persist and are presumably filled with fibrous tissue (Figs. 14B and E). In the untreated patient, there is no evidence of hyperparathyroidism as there is no hypocalcemia, in contrast to vitamin D deficiency osteomalacia. However, treatment with large doses of oral phosphate for long periods may induce secondary hyperparathyroidism, which may be exceptionally evident by subperiosteal erosions in the hand [102]. 3. OSTEOSCLEROSIS

Although there is defective mineralization of osteoid in XLH, the bones are commonly increased in density, with a coarse and prominent trabecular pattern [103] (Fig. 12). This is a characteristic feature of the disease and is not related to treatment with vitamin D and phosphate supplements, as it can be present in those who have not received treatment. This bone sclerosis has been shown to involve the petrous bone and structures of the inner ear and may be responsible for the hydropic cochlear pattern of deafness that these patients can develop in later life [104]. Deformity and thickening of the skull and involvement of the skull base can rarely be associated with stenosis of the foramen magnum with a Type 1 Chiari malformation and syringomyelia [105,106].

982 A

JUDITH E. ADAMS

B

C

D

FIGURE 12 Familial, X-linked hypophosphatemic rickets (XLH). (A) Anteroposterior view of ankle showing widened growth plate and Harris growth arrest lines. The appearance of the metaphysis is different from that in nutritional rickets (see Fig. 1). (B) Pelvis showing rachitic changes at the proximal femoral metaphyses, dense bones with a coarse trabecular pattern, and bowing of the femoral shafts with bilateral Looser's zones (through medial cortex right femoral shaft and outer cortex of left femoral shaft [arrows]). (C) Pelvis showing rickets at the proximal femoral metaphyses, dense bones with a coarse trabecular pattern and chronic Looser’s zones in the medial aspect of the proximal shafts of both femora. (D) Femora showing rickets at the proximal and diskal femoral metaphyses, and the femora are bowed and undertubulated with broad shafts. There are Looser's zones through the medial cortex in the proximal shaft of each femur. The bones are also increased in density.

4. ABNORMALITIES OF BONE MODELING

In XLH the bones are often short, with widening of the shaft. The ribs are broad and tend to slope downward more than normal, causing a bell-shaped chest (Fig. 13A). There can be broadening of the distal end of the ulna (Fig. 13B), and often marked bowing of the femur and tibia [107]. 5. EXTRASKELETAL OSSIFICATION

X-linked hypophosphatemia is characterized by an enthesopathy (inflammation in the junctional area between bone and tendon insertion) that heals by ossification of ligament and tendon insertions to bone

in many affected patients [108,109]. This results in new bone formation around the pelvis and spine, with the changes resembling ankylosing spondylitis (Fig. 14). There can be complete ankylosis of the spine, which limits movements. As there is no inflammatory arthritis, the sacro-iliac joints are normal, an important radiographic feature that serves to differentiate this condition from ankylosing spondylitis (Figs. 14A, 15A and B). Ossification can occur in the interosseous membrane of the forearm, forming a synchondrosis between the radius and ulna, and in the leg between the tibia and fibula (Figs. 14D and E). Separate small ossicles can occur around the joints of the hands (Fig. 14C); there

983

CHAPTER 60 Radiology of Rickets and Osteomalacia

A

B

FIGURE 13 Modeling deformities in familial X-linked hypophosphatemic rickets (XLH). (A) Chest radiograph showing that the ribs are broad and slope downward more than normal, giving a rather bell-shaped chest. (B) Wrist showing some bulbous expansion of the distal shaft of the ulna.

is also ossification at tendon insertions in the hands, causing “whiskering” of bone margins. Rarely, spinal cord compression may be caused by a combination of ossification of the ligamentum flavum, thickening of the laminae, and hyperostosis around the apophyseal joints [110] (Fig. 15). It is the ossification of the ligamentum flavum that causes the most significant narrowing of the spinal canal [111,112]. This occurs most commonly in the thoracic spine and generally involves two or three adjacent vertebral segments. Patients may be asymptomatic even with severe spinal canal narrowing, and acute cord compression can be precipitated by quite minor trauma. It is important to be aware of this rare, but recognized, complication of the disease since surgical decompression by laminectomy is curative. The extent of ossification cannot be predicted by the degree of paraspinal or extraskeletal ossification at other sites. Computerized tomography (CT), with its cross-sectional depiction of anatomy, is a useful imaging technique for demonstrating the extent of intraspinal ossification (Figs. 15C–F). Extraskeletal ossification is uncommon in patients with XLH under 40 years of age. 6. OSTEOARTHRITIS

The deformity and bowing of the long bones in XLH cause altered stress through joints, predisposing to degenerative joint disease. This is particularly prominent in the knee joint.

The extent to which the radiographic abnormalities discussed in Section IV,A,1–6 are present varies between affected individuals [113]. In some, all the features are present and so are diagnostic of XLH. In others, there may only be minor abnormalities, and the condition may be overlooked [114].

B. Tumor-Induced (“Oncogenic”) Rickets/Osteomalacia Tumor induced osteomalcia (TIO) or “oncogenic” rickets and osteomalacia were first reported in 1947 [93,115]. There is hypophosphatemia due to excessive urinary phosphate loss, and serum concentrations of 1,25(OH)2D are low or undetectable. The clinical and radiographic features of rickets or osteomalacia can precede identification of the causative tumor by long periods (1–16 years). Bilateral Looser’s zones in the tibia have been described, and can mimic radiographically stress fractures that might occur in athletes in this site [116]. The tumors are usually small, benign, and vascular in origin (hemangiopericytoma) [117,118], but some may be malignant [119] (Fig. 16). Rickets and osteomalacia will heal with surgical removal of the tumor [120,121]. It has now been established that circulating concentrations of fibroblast growth factor 23 (FGF23) are high in tumor induced osteomalacia, and the levels fall after the tumor is removed [122,123]. Often, the tumors are extremely

984 A

C

JUDITH E. ADAMS

B

D

E

FIGURE 14 Enthesopathy and extraskeletal ossification in XLH. (A) Anteroposterior view of pelvis and left hip (B) showing a tri-radiate deformity of the pelvis indicative of rickets in childhood. There is new bone formation around the hip joints and at lesser trochanters. There is a chronic Looser’s zone in the outer cortex of the proximal shaft of the left femur. Note the normal sacroiliac joints which help in distinguishing the etiology of paravertebral ossification, which occurs in XLH, from that which occurs in ankylosing spondylitis. (C) Hands showing ossicles and bony outgrowths related to heads of the metacarpals and joints. (D) Forearm showing synostosis between the radius and ulna due to ossification of the interosseous ligament. (E) Lower leg showing deformity due to bowing and chronic Looser's zones with ossification of the interosseous membrane. Such severe deformity reflects inadequate treatment during childhood.

small and elude detection for many years (Fig. 19). It is important that the patient is vigilant about selfexamination and reports any small palpable lump or skin lesion. More sophisticated imaging (CT, magnetic resonance imaging [MRI] and radionuclide scanning [ocreotide]) may be helpful in localizing more deep-seated lesions [124–128]. The condition is discussed in detail in Chapter 70. If no causal lesion comes to light with thorough imaging and spontaneous remission occurs, the condition described as “pseudo-(tumor-induced) rickets” should be considered, to avoid prolonged medical treatment and futile searches for a neoplasm [129].

V. ACIDEMIA The mineralization of osteoid also requires a normal pH to prevail in the body environment. If there is a

systemic acidosis then the features of rickets or osteomalacia may be evident radiographically. This may occur with ileal replacement of the ureters [130] and with chronic and excessive antacid ingestion [131,132].

VI. DIFFERENTIAL DIAGNOSES Other etiologies may cause radiographic abnormalities of the metaphyses that have to be distinguished from those of rickets. These are the syndromes associated with metaphyseal dysostoses (chondrodysplasias) (e.g. types of Jansen, Schmidt, etc.) (Fig. 17). The metaphyseal changes can vary in severity from mild, such as occur in the Shwachman-Diamond syndrome (Figs. 17A and B), in which there is associated neutropenia and pancreatic insufficiency, to more significant fragmentation of the metaphyses [133–135] (Figs. 17C and D).

A

C

B

D

E

F

FIGURE 15

Changes in XLH. (A) Lateral lumbar spine showing ossification of the paraspinal ligaments and at apophyseal joints. The appearances resemble those of ankylosing spondylitis but can be differentiated from it by the sacroiliac joints being normal and not eroded, as would occur in seronegative arthropathies. (B) Anteroposterior view of lumbar spine showing ossification of paraspinal ligaments resulting in ankylosis but normal sacroiliac joints (arrows). (C–F) Narrowing of the spinal canal may be caused by various factors. (C) Computerized tomography (CT) scan through thoracic spine showing thickening of the laminae and hypertrophy of the apophyseal joints causing narrowing and trefoil deformity of the spinal canal. (D) CT scan through the lower thoracic spine showing severe narrowing of the spinal canal by posterior ossification of the ligamentum flavum. (E) CT thoracic spine of a different patient showing narrowing of the spinal canal by new bone formation lying anterior to the laminae (arrow), severely narrowing the thoracic spinal canal (CT scan). (F) Sagittal reformation of thin (3 mm) contiguous CT sections showing ossification of the ligamentum flavum (arrows), severely reducing the anteroposterior diameter of the spinal canal. D, E, and F reprinted from Adams and Davies [111] with permission.

986

JUDITH E. ADAMS

vomiting due to hypercalcemia (see Chapter 78). The hypercalcemia results in hypercalciuria, nephrocalcinosis, renal impairment, and hypertension. Metastatic calcification and bone sclerosis also occur [141,142] (Fig. 18).

VIII. TECHNICAL ASPECTS OF IMAGING A. Plain Radiographs

FIGURE 16 Tumor-induced or oncogenic osteomalacia in a 46year-old woman who 18 months previously had presented with hypophosphatemic osteomalacia. There is a lytic lesion in the proximal fibula. This was a fibrosarcoma. The osteomalacia healed following surgical excision of the tumor.

In contrast to rickets of vitamin D deficiency, although the metaphyses are fragmented and splayed, they do not exhibit the same cupping and widening of the radiolucent growth plate [136]. Deficiency in alkaline phosphatase will also result in rickets and osteomalacia. This occurs in hypophosphatasia, which is an inborn error of metabolism first reported in 1948 by Rathburn [137]. There is an accumulation of the enzyme substrates, including phosphoethanolamine and of inorganic phosphate, which promotes the development of articular chondrocalcinosis. There are several forms of the disorder, which varies in its severity and age of onset [138].

VII. VITAMIN D INTOXICATION In the past, vitamin D was advocated in the treatment of a great variety of conditions, including tuberculosis (especially lupus vulgaris), sarcoidosis, rheumatoid arthritis, hay fever, chilblains, and asthma. This treatment had no beneficial effect in these conditions, and its use was eventually abandoned, but not before many cases of vitamin D intoxication had been described [139,140]. Although vitamin D intoxication has become less common with the advent of 1,25(OH)2D3 and other active metabolites, the new era of vitamin D usage to treat cancer, psoriasis, and immunological disease (see Section X of this volume) may see a resurgence of interest in vitamin D intoxication. Clinically, the symptoms are of fatigue, malaise, weakness, thirst and polyuria, anorexia, nausea, and

Despite tremendous developments and expansion in the imaging techniques available since the 1970s, plain radiographs remain the principal imaging method in the radiographic diagnosis of metabolic bone disorders, including those involving vitamin D deficiency, rickets, and osteomalacia. When radiographing the hands and feet, image quality must be optimized by using fine grain, single-sided emulsion film and a fine X-ray focal spot (0.6 mm or less). Meticulous attention to detail of the radiographic technique used will enhance the diagnostic features present in the hand radiograph, such as the subperiosteal erosions and intracortical tunneling of hyperparathyroidism. Magnification techniques, either optical or radiographic, can further enhance identification of such diagnostic features of metabolic bone disease. High-resolution radiographs of the torso regions of the body are generally precluded because of the high radiation doses required.

B. Nuclear Medicine In the imaging technique of the skeleton referred to as nuclear medicine, 99mTc-labeled phosphate compounds are administered intravenously [143,144]. They are incorporated into the skeleton, particularly in sites that have either increased vascularity or increased new bone formation. Such areas of increased uptake are evident as “hot spots” on the scan, which is performed, using a gamma camera, 2 hours after administration of the radionuclide. This bone scanning technique is very sensitive to disease in bone, but not specific, in that numerous pathologies may cause hot spots, including infection, Paget’s disease of bone, metastases, and degenerative joint disease (hyperostosis). Radiographs of the relevant anatomical site will help to differentiate these various pathologies. The radionuclide bone scan is sensitive to detecting Looser’s zones that may not be evident radiographically [145–149] (Fig. 19). The areas of increased uptake of label may be bilateral and symmetrical and be present in anatomical sites typical for Looser’s zones (femoral necks, ribs, pubic rami) [150] (Figs. 19B and C). If there is associated secondary hyperparathyroidism, there will be generalized increase in uptake of the

987

CHAPTER 60 Radiology of Rickets and Osteomalacia

A

B

C

D

E

F

FIGURE 17 Differential diagnoses. Metaphyseal dysostosis (chondrodysplasia). The metaphyseal abnormalities may be mild and resemble rickets as evident in the knees (A) and proximal femora (B) in this patient with Shwachman-Diamond syndrome, in which there is associated neutropenia and pancreatic insufficiency. In more severe types, there is greater fragmented mineralization of the metaphyses in the wrists (C) and the hips (D), which do not resemble the radiographic abnormalities of rickets. Hypophosphatasia, which is an inborn error of metabolism in which alkaline phosphatase levels are reduced, also results in rickets, as is evident in the knee (E) of this affected child, and osteomalacia in affected adults (F). As there are no specific therapeutic options, Looser’s zones are chronic, as seen in the neck and shaft of the left femur, and require repeated orthopaedic interventions, such as the intramedullary nailing, which has been undertaken in this case. Chondrocalcinosis may also be evident.

988

JUDITH E. ADAMS

A

B

FIGURE 18 Vitamin D intoxication with associated renal impairment. (A) Pelvis showing bone sclerosis, calcification of vessels in the pelvis, and periosteal reaction around the pelvic brim. There is also calcification of the ilio-lumbar ligaments. (B) Lateral foot showing bone sclerosis and calcification of the ligaments in the foot. There was calcification of ligaments in other sites and of the falx and tentorium in the head.

radionuclide by the skeleton (“super scan”), with elevation of the bone/soft tissue ratio. There will be poor renal uptake of the radionuclide if the cause of the osteomalacia and vitamin D deficiency is chronic renal disease. With appropriate treatment of the osteomalacia, the Looser’s zones will heal, and the hot spots will not be present on subsequent radionuclide scanning. Radionuclide scanning offers a method of monitoring response of osteomalacia to therapy [151,152]; however, clinical symptoms, biochemical parameters, and plain radiographs often suffice for this purpose. As mesenchymal tumors have somatostatin receptors, and may be associated with tumor-induced osteomalacia, Indium-111 labeled pentetreotide has become an important imaging technique to identify the presence and confirm the site of such tumors, which may otherwise prove elusive to clinical localization [153–155] (Fig. 19E). If the radionuclide scan demonstrates an abnormal area of high uptake, then it may additionally be helpful to perform CT of MR targeted at the area of abnormality, since these imaging methods offer higher spatial resolution and superior anatomical detail of the site and size of the lesion (Fig. 19G). Radionuclide scanning in children is of limited value in metabolic bone disorders, because there is high uptake in the normal metaphysis of the growing skeleton. In addition, the examination carries a significant radiation dose to the bone marrow.

C. Other Imaging Techniques Ultrasonography (US), computerized tomography (CT), magnetic resonance imaging (MRI), and angiography generally play little part in the diagnosis and

management of rickets and osteomalacia. The exception to this is their application to the identification and localization of tumors that cause oncogenic rickets. Such tumors are often very small and may be deep seated, so they can be extremely difficult to identify (see Chapter 70). Whole body MR scanning and CT have proved useful in such identification and localization [126,156]. These imaging methods may be particularly applicable after RNS with Indium-111 labeled ocreotide scanning has localized an area of increased uptake, since they can then be targeted at the abnormal site, and will provide superior anatomical localization and spatial resolution than RNS.

D. Bone Mineral Densitometry Methods of bone mineral densitometry (BMD) play an important role in diagnosis of patients with osteoporosis and monitoring the efficacy of treatment [157,158]. The methods available include single energy X-ray absorptiometry (SXA) for forearm (and calcaneus) measurements, dual energy X-ray absorptiometry (DXA) for measurements in the lumbar spine (L1–4), proximal femur (total hip, femoral neck, trochanteric, and Ward’s area) and whole body, quantitative computerized tomography (QCT) for measuring cortical and trabecular bone separately in the lumbar spine and forearm, and broadband ultrasound attenuation (BUA), used to make measurements in the calcaneum. In disease, and treatment, these techniques can provide complementary information because they measure different types of bone in different sites of the skeleton. Axial DXA is the most widely used method in clinical practice [159,160]. The World Health Organization (WHO) has defined

989

CHAPTER 60 Radiology of Rickets and Osteomalacia

A

C

B

D

E

F

FIGURE 19

G

Acquired hypophosphatemic osteomalacia. A 40-year-old man presented with low back and buttock pain. He was thought to have ankylosing spondylitis. (A) Pelvic radiograph, which was considered to be normal at the original hospital of referral. (B) Radionuclide scan of pelvis showing, increased uptake in both femoral necks, which are sites of Looser’s zones. (C) Radionuclide scan of upper torso showing multiple “hot spots” in ribs. The distribution of the hot spots suggests Looser's zones. A radionuclide scan is more sensitive than a radiograph for identifying Looser’s zones. The patient subsequently had a road traffic accident in which he suffered bilateral femoral neck fractures through the Looser’s zones, which in retrospect are subtly present in the original radiograph (A). The patient was found to have hypophosphatemia presumed to be tumor-induced, but no tumor was identified after 15 years despite careful clinical examination and use of other imaging techniques. However, recently he had a radionuclide ocreotide scan, which showed a hot spot in the pelvis (E). The sagittal reformatted crude CT image (F) performed on the gamma camera at the end of the scan indicates that the lesion lies anterior to the bladder. (G) Conventional CT scan through the pelvis confirms a sclerotic lesion in the left superior pubic ramus, adjacent to the symphysis. The etiology of the lesion and whether this is the cause of the oncogenic hypophosphatemic osteomalacia cannot yet be confirmed, as the patient does not wish to have the lesion removed. Figures B, C, E, and F courtesy of Dr Mary Prescott, Consultant in Nuclear Medicine, and Figure G courtesy of Dr Richard Whitehouse, Consultant Radiologist, both at The Royal Infirmary, Manchester, UK. Figure E case report reference 155 (Moran and Paul Int Orthop 2002:26:61–62 with permission).

990

JUDITH E. ADAMS

osteoporosis in terms of bone densitometry as a T score (standard deviation below the mean of peak bone mass in an appropriate sex- and race-match reference group) below –2.5. DXA has advantages (rapid scanning approximately 5 min per site, extremely low radiation doses 2–6 µSv [similar to one day’s natural background radiation] and good reproducibility (CV 1%), but there are also some limitations (artefacts in the lumbar spine from degenerative and hyperostotic changes, particularly in the elderly; size dependency as it is an ‘areal’ density in g/cm2, a particular problem in growing children and patients who are small through disease. QCT is unique amongst methods in providing a true volumetric density (mg/cc), so is not size dependent, and can measure cortical and trabecular bone density separately [161]. SXA measures integral (cortical and trabecular) bone in the forearm, and DXA measures integral bone in the proximal femur, the lumbar spine, and the whole body. In vitamin D deficiency and rickets or osteomalacia, there may be osteopenia; bone desitometry techniques cannot distinguish between reduced BMD being due to osteoporosis (reduced bone mass) and osteomalacia (reduced mineral/osteoid ratio) [162]. If there is secondary hyperparathyroidism, the forearm cortical bone mineral density (BMD) (e.g. distal SXA/DXA) measurement might show the most marked reduction [163]. Where vitamin D deficiency osteomalcia is treated appropriately, there is very rapid increase (+25% or more) in BMD (2–4 weeks) on serial bone densitometry (see Chapter 71).

this acts as a stimulus to secondary hyperparathyroidism, features of which can be identified radiographically (bone erosions). The bone disease of chronic renal impairment (renal osteodystrophy) is complex, being a combination of rickets and osteomalacia [1,25(OH)2D deficiency], hyperparathyroidism (bone erosions, sclerosis), and metastatic calcification (due to phosphate retention). However, the pattern of bone disease in chronic renal impairment has changed over the past 30 years with improved knowledge of vitamin D metabolism, treatments to prevent vitamin D deficiency (calcitriol, 1α vitamin D), transplantation, and dialysis. It is therefore now rare to see florid rickets and osteomalacia radiographically with the associated features of severe and longstanding secondary hyperparathyroidism that was evident in the past. However, metastatic calcification in soft tissues is still witnessed and remains problematic. Plain radiographs remain the most important imaging technique for the diagnosis of metabolic bone disease; radionuclide bone scanning is more sensitive for identifying Looser’s zones. Radionuclide scanning using Indium-111 labeled ocreotide can be useful in localizing tumors associated with tumoral (oncogenic) osteomalacia, as may other cross-sectional imaging methods (ultrasound, computed tomography, magnetic resonance imaging). CT is particularly well suited to demonstrate the intra-spinal ossification which is a rare, but recognized, complication of the enthesopathy associated with X-linked hypophosphatemic osteomalacia.

IX. CONCLUSIONS

References

Mineralization of bone matrix depends on the presence of adequate supplies of not only vitamin D, in the form of its active metabolite 1,25(OH)2D, but also calcium and phosphorus and the presence of alkaline phosphatase and a normal pH. If there is deficiency of these substances for any reason, or if there is severe acidosis, then mineralization of bone will be defective. This results in rickets in childhood and osteomalacia in adults. Radiographically, rickets is evident by bone deformity caused by softening and metaphyseal abnormalities where endochondral ossification is defective. The pathognomonic feature of osteomalacia is the Looser’s zone (pseudofracture). Many different diseases that result in rickets and osteomalacia (vitamin D deficiency, calcium deficiency, hypophosphatemia, hypophosphatasia, and acidemia) may therefore have similar radiographic appearances. There may be distinguishing features, such as bone sclerosis and extraskeletal ossification in XLH. In conditions in which there is hypocalcemia (vitamin D deficiency),

1. Harrison HE, Harrison HC 1975 Rickets then and now. J Pediatr 87:1144. 2. Pitt MJ 1991 Rickets and osteomalacia are still around. Radiol Clin North Am 29:97–118. 3. Turner ML, Dalinka MK 1979 Osteomalacia: Uncommon causes. A J Roentgenol 133:539–540. 4. Smerdon GT 1950 Daniel Whistler and the English Disease. A translation and biographical note. J Hist Med 5:397–415. 5. Glendening L 1942 Source Book of Medical History. Dover, New York, pp. 268–269. 6. Fourman P, Royer 1960 Vitamin D. In: Calcium Metabolism and the Bone. Blackwell, Oxford, and Edinburgh, pp. 104–129. 7. Dewhurst K 1962 Postmortem examination in case of rickets performed by John Locke. Br Med J 2:1466. 8. Hunter R 1972 Rickets, ruckets, rekets, or rackets? Lancet 1:1176–1177. 9. LeVay D 1975 On the derivation of the name “rickets.” Proc R Soc Med 68:46–50. 10. Schutte D 1824 Beobachtungen uber den Nutzen des Berger Leberthrans. Arch Med Erfahrung 2:79–92. 11. Percival T 1783 Observations on the medicinal uses of the oleum jecoris aselli or cod liver oil, in the chronic rheumatism and other painful disorders. Lond Clin Med J 3:392–401.

CHAPTER 60 Radiology of Rickets and Osteomalacia

12. Palm TA 1890 The geographical distribution and aetiology of rickets. Practitioner 45:270–279. 13. Morse JL 1900 The frequency of rickets in infancy in Boston and vicinity. JAMA 34:724. 14. Parsons LG 1927 The bone changes occurring in renal and coeliac infantilism, and their relationship to rickets. Part I. Renal rickets. Arch Dis Child 2:1–25. 15. Parsons LG 1927 The bone changes occurring in renal and coeliac infantilism and their relationship to rickets. Part II. Coeliac rickets. Arch Dis Child 2:198–211. 16. Pommer G 1885 Untersuchungen ober osteomalacie und rachitis. Leipzig. 17. Dent CE 1970 Rickets (and osteomalacia), nutritional, and metabolic (1919–1969). Proc R Soc Med 63:401–408. 18. Hutchison FN, Bell NH 1992 Osteomalacia and rickets. Semin Nephrol 12:127–145. 19. Berry JL, Davies M, Mee AP 2002 Vitamin D metabolism, rickets, and osteomalacia. Sem Musculoskel Radiol 6: 173–181. 20. Pitt MJ 1981 Rachitic and osteomalacia syndromes. Radiol Clin North Am 19:581–598. 21. Smith R 1972 The pathophysiology and management of rickets. Orthop Clin North Am 3:601–621. 22. Stamp TCB, Exton-Smith AN, Richens A 1976 Classical rickets and osteomalacia in Britain. Lancet 2:308. 23. Stamp TCB, Round JM 1974 Seasonal changes in human plasma levels of 25-hydroxyvitamin D. Nature 247:563–565. 24. Mughal Z 2002 Rickets in childhood. Semin Musculoskelet Radiol. 6:183–190. 25. Pettifor JM 2002 Rickets Calcif Tissue Int 70:398–399. 26. Wharton B, Bishop N 2003 Rickets. Lancet 362:1389–1400. 27. Oestreich AE, Ahmad BS 1993 The periphysis and its effect on the metaphysis. II. Application to rickets and other abnormalities. Skel Radiol 22:115–119. 28. Steinbach HL, Noetzli M 1964 Roentgen appearance of the skeleton in osteomalacia and rickets. Am J Roentgenol 91:955–972. 29. Ontell FK, Moore EH, Shephard JA, Shelton DK 1997 The costal cartilages in health and disease. Radiographics 17:571–577. 30. Thatcher TD, Fischer PR, Pettifor JM, Lawson JO, Manaster BJ 2000 Radiographic scoring method for the assessment of the severity of nutritional rickets. J Trop Pediatr 46:132–139. 31. Frankel A, Gruber B, Schey WL 1994 Rickets presenting as stridor and apnea. Ann Otol Rhinol Laryngol 103:905–972. 32. Glasgow JFT, Thomas PS 1977 Rachitic respiratory distress in small preterm infants. Arch Dis Child 52:268–273. 33. Park EA 1954 The influence of severe illness on rickets. Arch Dis Child 29:369–380. 34. Eek S, Gabrielsen LH, Halvosen S 1957 Prematurity and rickets Pediatrics 20:63–77. 35. Pettifor JM, Pentopoulos M, Moodley GP, Isdale JM, Ross FP 1984 Is craniotabes a pathognomonic sign of rickets in 3-month-old infants? S Afr Med J 65:549–551. 36. Harris HA 1933 Rickets. In: Bone Growth in Health and Disease. Oxford Medical Publications, Oxford University Press, London, p. 87. 37. Davies DR, Dent CE, Willcox A 1956 Hyperparathyroidism and steatorrhoea. B Med J 2:1133–1137. 38. Mankin HJ 1974 Rickets, osteomalacia, and renal osteodystrophy—Part I. J Bone Joint Surg (Am) 56:101–128. 39. Mankin HJ 1974 Rickets, osteomalacia, and renal osteodystrophy—Part II. J Bone Joint Surg (Am) 56:352–386. 40. Mankin HJ 1990 Rickets, osteomalacia, and renal osteodystrophy. Orthop Clin North Am 21:81–96.

991 41. Bereket A, Casur Y, Firat P, Yordam N 2000 Brown tumor as a complication of secondary hyperparathyroidism in severe longlasting vitamin D–deficiency rickets. Eur J Paediatr 159:70–73. 42. Looser E 1908 Uber spatrachitis und die Beziehungen zwischen rachitis und osteomalacie. Mitt Grenzgeb Med Chir 18: 678–744. 43. Looser E 1920 Uber spatrachitis und osteomalacie Klinishe ront-genologische und pathologischanatomische Untersuchungen. Drsch Z Chir 152:210–357. 44. Looser E 1920 Uber pathologische Formen von Infraktionen und Callusbildungen bei Rachitis und osteomalakie und anderen knocken Erkrankungen. Zbl Chir 47:1470–1474. 45. Milkman LA 1930 Pseudofractures (hunger osteopathy, late rickets, osteomalacia). Am J Roentgenol 24:29–37. 46. Milkman LA 1934 Multiple spontaneous idiopathic symmetrical fracture. Am J Roentgenol 32:622–634. 47. Reginato AJ, Falasca GF, Pappu R, McKnight B, Agha A 1999 Musculoskeletal manifestations of osteomalacia: Report of 26 cases and literature review. Semin Arthritis Rheum 28:287–303. 48. Camp JD, McCullough JAL 1941 Pseudofractures in diseases affecting the skeletal system. Radiology 36:651–661. 49. LeMay M, Blunt JW 1949 A factor determining the location of pseudofractures in osteomalacia. J Clin Invest 28:521–525. 50. Steinbach HL, Kolb FO, Gilfillan R 1954 A mechanism of the production of pseudofractures in osteomalacia (Milkman’s syndrome). Radiology 62:388–394. 51. Cooper KL 1994 Insufficiency stress fractures. Curr Prob Diagn Radiol 23:29–68. 52. McKenna MJ, Kleerekoper M, Ellis BI, Rao DS, Parfitt AM, Frame B 1987 Atypical insufficiency fractures confused with Looser zones of osteomalacia. Bone 8:71–78. 53. North KAK 1966 Multiple stress fractures simulating osteomalacia. Am J Roentgenol 97:672–675. 54. Quek ST, Peh WCG 2002 Radiology of osteoporosis. Semin Musculoskelet Radiol 6:197–206. 55. Whitehouse RW 2002 Paget’s disease of bone. Semin Musculoskeletal Radiol 6:313–322. 56. Bulloch B, Schubert CJ, Brophy PD, Johnson N, Reed MH, Shapiro RA 2000 Cause and clinical characteristics of rib fractures in infants. Pediatrics 105:E48. 57. Pitt MJ 1995 Rickets and osteomalacia. In: Resnick D (ed) Diagnosis of Bone and Joint Disorders. Saunders, Philadelphia, Pennsylvania, pp. 1885–1920. 58. Shapiro R 1972 Radiologic aspects of renal osteodystrophy. Radiol Clin North Am 10:557–568. 59. Sundaram M 1989 Renal osteodystrophy. Skeletal Radiol 18:415–426. 60. Adams JE 2002 Dialysis bone disease. Semin Dial 15:277–289. 61. McCarthy JT, Kumar R 1986 Behavior of the vitamin D endocrine system in the development of renal osteodystrophy. Semin Nephrol 6:21–30. 62. Peacock M 1978 Renal bone disease. Practitioner 220:913–918. 63. Massry SG, Ritz E 1978 The pathogenesis of secondary hyperparathyroidism of renal failure: Is there a controversy? Arch Intern Med 138:853–856. 64. Norfray J, Calenoff L, DelGreco F, Krumlovsky F 1975 Renal osteodystrophy in patients on hemodialysis as reflected in the bony pelvis. Am J Roentgenol 125:352–358. 65. Parfitt AM 1972 Renal osteodystrophy. Orthop Clin North Am 3:681–698. 66. Theodorou DJ, Theodorou SJ, Resnick D 2002 Imaging in dialysis sponyloarthropathy. Semin Dial 15:290–296. 67. Resnick D, Niwayama G 1976 Subchondral resorption of bone in renal osteodystrophy. Radiology 118:315–321.

992 68. Andresen J, Nielsen HE 1981 Juxta-articular erosions and calcifications in patients with chronic renal failure. Acta Radiol Diagn 22:709–713. 69. Tigges S, Nance EP, Carpenter WA, Erb R 1995 Renal osteodystrophy: imaging findings that mimic those of other diseases. Am J Roentgenol 165:143–148. 70. Resnick DL 1974 Erosive arthritis of the hand and wrist in hyperparathyroidism. Radiology 110:263–269. 71. Goldman AB, Lane JM, Salvati E 1978 Slipped capital femoral epiphyses complicating renal osteodystrophy: A report of three cases. Radiology 126:333–339. 72. Mehls O, Ritz E, Krempien B, Gilli G, Link K, Willich E, Scharer K 1975 Slipped epiphyses in renal osteodystrophy. Arch Dis Child 50:545–554. 73. Wolf HL, Deubo JV 1958 Osteosclerosis in chronic renal disease. Am J Med Sci 235:33–42. 74. Meema HE, Oreopoulos DO, Rabinovich S, Husdan H, Rapoport A 1974 Periosteal new bone formation (periosteal neostosis) in renal osteodystrophy. Radiology 110:513–522. 75. Ritchie WGM, Winney RJ, Davison AM, Robson JS 1975 Periosteal new bone formation developing during haemodialysis for chronic renal failure. Br J Radiol 48:656–661. 76. Slatopolsky E, Rutherford WE, Hruska K, Martin K, Klahr S 1978 How important is phosphate in the pathogenesis of renal osteodystrophy? Arch Intern Med 138:848–852. 77. Parfitt AM 1969 Soft-tissue calcification in uremia. Arch Intern Med 124:544–556. 78. Gipstein RM, Coburn JW, Adams DA, Lee DB, Parsa KP, Sellars A, Suki WN, Massry SG 1976 Calciphylaxix in man: a syndrome of tissue necrosis and vascular calcification in 11 patients with chronic renal failure. Arch Intern Med 136:1273–1280. 79. Weinreich T 1998 Prevention of renal osteodystrophy in peritoneal dialysis. Kidney Int 54:2226–2233. 80. Musci I, Hercz G 1977 Adynamic bone disease: pathogenesis, diagnosis, and clinical relevance. Curr Opin Nephrol Hyperten 6:356–361. 81. Verberckmoes R, Bouillon R, Krempien B 1975 Disappearance of vascular calcification during treatment of renal osteodystrophy—two patients treated with high doses of vitamin D and aluminum hydroxide. Ann Intern Med 82:529–533. 82. Raggi P 2002 Effects of excess calcium load on the cardiovascular system measured with electron beam tomography in end-stage renal failure. Nephrol Dial Transplant 17:332–335. 83. Raggi P, Boulay A, Chasan-Taber S, Amin N, Dillon M, Burke SK, Chertow GM 2002 Cardiac calcification in adult hemodialysis patients. A link between end-stage renal disease and cardiovascular diseases. J Am Coll Cardio 20:695–701. 84. Andress DL, Maloney NA, Coburn JW, Endres DB, Sherrard DJ 1987 Osteomalacia and aplastic bone disease in aluminumrelated osteodystrophy. J Clin Endocrinol Metab 65:11–16. 85. Kriegshauser JS, Swee RG, McCarthy JT, Hauser MF 1987 Aluminum toxicity in patients undergoing dialysis: Radiographic findings and prediction of bone biopsy results. Radiology 164:399–403. 86. Ward MK, Feest TG, Ellis HA, Parkinson IS, Kerr DNS 1978 Osteomalacic dialysis osteodystrophy: Evidence for a waterborne aetiological agent, probably aluminum. Lancet 4:841–845. 87. Smith GD, Winney RJ, McLean A, Robson JS 1987 Aluminum-related osteomalacia: Response to reverse osmosis water treatment. Kidney Int 32:96–101.

JUDITH E. ADAMS

88. Sundaram M, Dessner D, Ballal S 1991 Solitary spontaneous cervical and large bone fracture in aluminum osteodystrophy. Skel Radiol 20:91–94. 89. Schulman JD, Schneider JA 1976 Cystinosis and the Fanconi syndrome. Pediatr Clin North Am 23:779–793. 90. Dubois J, Garel L, Patriquin H, Paradis K, Forget S, Filiatrault D, Grignon A, Russo P, St-Vil D 1996 Imaging features of Type 1 hereditary tyrosinemia: A review of 30 patients. Pediatr Radiol 26:845–851. 91. Takebayashi S, Jimi S, Segawa M, Kiyoshi Y 2000 Cadmium induces osteomalacia mediated by proximal tubular atrophy and disturbance of phosphate reabsorption. A study of 11 autopsies. Pathol Res Pract 196:653–663. 92. Lawson J 2002 Drug-induced metabolic bone disorders. Semin Musculoskel Radiol 6:285–297. 93. Ryan EA, Reiss E 1984 Oncogenous osteomalacia: Review of the world literature of 42 cases. Am J Med 77:501–512. 94. Clarke BL, Wynne AG, Wilson DM, Fitzpatrick LA 1995 Osteomalacia associated with adult Fanconi’s syndrome: Clinical and diagnostic features. Clin Endocrinol 43: 479–490. 95. Olivares JL, Ramos FJ, Carapeto FJ, Bueno M 1999 Epidermal naevus syndrome and hypophosphatemic rickets: Description of a patient with central nervous system anomalies and a review of the literature. Eur J Pediatr 158:103–107. 96. Econs MJ, McEnery PT 1997 Autosomal dominant hypophosphatemic rickets/osteomalacia: Clinical characterization of a novel renal phosphate-wasting disorder. J Clin Endocrinol Metab 82:674–681. 97. Hanna JD, Niimi K, Chan JCM 1991 X-linked hypophosphatemia. Genetic and clinical correlates. Am J Dis Child 145:865–870. 98. Walton J 1976 Familial hypophosphatemic rickets: A delineation of its subdivisions and pathogenesis. Clin Pediatr 15:1007–1012. 99. Weisman Y, Hochberg Z 1994 Genetic rickets and osteomalacia. Curr Ther Endocrinol Metab 5:492–495. 100. Rowe PS 1998 The role of the PHEX (PEX) gene in families with X-linked hypophosphatemic rickets. Curr Opin Nephrol Hypertens 7:367–376. 101. Milgram JW, Compere CL 1981 Hypophosphatemic vitamin D refractory osteomalacia with bilateral femoral pseudofractures. Clin Orthop 160:78–85. 102. Rivkees SA, el-Hajj-Fuleihan G, Brown EM, Crawford JD 1992 Tertiary hyperparathyroidism during high phosphate therapy of familial hypophosphatemic rickets. J Clin Endocrinol Metab 76:1514–1518. 103. Steinbach HL, Kolb FO, Crane JT 1959 Unusual roentgen manifestations of osteomalacia. Am Roentgenol 82:875–886. 104. O’Malley SP, Adams JE, Davies M, Ramsden RT 1988 The petrous temporal bone and deafness in X-linked hypophosphatemic osteomalacia. Clin Radiol 39:528–530. 105. Kuether TA, Piatt JH 1998 Chiari malformation associated with vitamin D–resistant rickets: Case report. Neurosurgery 42:1168–1171. 106. Caldermeyer KS, Boaz JC, Wappner RS, Moran CC, Smith RR, Quets JP 1995 Chiari 1 malformation: Association with hypophosphatemic rickets and MR imaging appearances. Radiology 195:733–738. 107. McAlister WH, Kim GS, Whyte MP 1987 Tibial bowing exacerbated by partial premature epiphyseal closure in sexlinked hypophosphatemic rickets. Radiology 162:461–463. 108. Burnstein MI, Lawson JP, Kottamasu SR, Ellis BI, Micho J 1989 The enthesopathic changes of hypophosphatemic

CHAPTER 60 Radiology of Rickets and Osteomalacia

109.

110.

111. 112. 113.

114. 115. 116. 117. 118. 119. 120.

121. 122. 123.

124. 125. 126. 127.

osteomalacia in adults: Radiologic findings. Am J Roentgenol 153:785–790. Polisson RP, Martinez S, Khoury M, Harrell RM, Lyles KW, Friedman N, Harrelson JM, Reisner E, Drezner MK 1985 Calcification of entheses associated with X-linked hypophosphatemic osteomalacia. N Engl J Med 313:1–6. Cartwright DW, Latham SC, Masel JP, Yelland JDN 1979 Spinal canal stenosis in adults with hypophosphatemic vitamin D–resistant rickets. Aust New Zealand J Med 9:705–708. Adams JE, Davies M 1986 Intra-spinal new bone formation and spinal cord compression in familial hypophosphataemic vitamin D–resistant osteomalacia. Q J Med 61:1117–1129. Dunlop DJ, Stirling AJ 1996 Thoracic cord compression caused by hypophosphatemic rickets: a case report and a review of the world literature. Eur Spine J 5:272–274. Hardy DC, Murphy WA, Siegel BA, Reid IR, Whyte MP 1989 X-linked hypophosphatemia in adults: Prevalence of skeletal radiographic and scintigraphic features. Radiology 171:403–414. Econs MJ, Samsa GP, Monger M, Drezner ML, Fuessner JR 1994 X-linked hypophosphatemic rickets: A disease often unknown to affected patients. Bone Miner 24:17–24. Prader A, Illig R, Uehlinger RE 1959 Rachitis infolge Knochen-tumors. Helv Paediatr Acta 14:554–565. Ohashi K, Ohnishi T, Ishikawa T, Tani H, Uesugi K, Takagi M 1999 Oncogenic osteomalacia presenting as bilateral stress fractures of the tibia Skelet Radiol 28:46–48. Renton P, Shaw DG 1976 Hypophosphatemic osteomalacia secondary to vascular tumors of bone and soft tissue. Skeletal Radiol 1:21–24. Choi JJ, Murphey MD 2000 Angiomatous skeletal lesions. Semin Musculoskel Radiol 4:103–112. Taylor HC, Fallen MD, Velasco ME 1984 Oncogenic osteomalacia and inappropriate antidiuretic hormone secretion due to oat-cell carcinoma. Ann Intern Med 101:786–788. Linovitz RJ, Resnick D, Keissling P, Kondon JJ, Sehler B, Nejdl RJ, Rowe JH, Deftos LJ 1976 Tumor-induced osteomalacia and rickets: A surgically curable syndrome, report of two cases. J Bone Joint Surg (Am) 58:419–423. Pollack JA, Schiller AL, Crawford JD 1973 Rickets and myopathy cured by removal of a nonossifying fibroma of bone. Pediatrics 52:363–371. Wilkins GE, Granleese S, Hegele RG, Holden J, Anderson DW, Bond GP 1995 Oncogenic osteomalacia: Evidence for a humeral phosphaturic factor. J Clin Endocrinol Metab 80:1628–1634. John MR, Wickert H, Zaar K, Jonsson KB, Grauer A, Puppersberger P, Schmidt-Gayk H, Murer H, Zeigler R, Blind E 2001 A case of neuroendocrine oncogenic osteomalacia associated with a PHEX and fibroblast growth factor –23 expressing sinusoidal malignant schwannoma. Bone 29:393–402. Leicht E, Kramann B, Seitz G, Trentz O, Remberger K 1993 Oncogenic osteomalacia: Imaging studies. Bildgebung 60:13–17. Shapiro D, Ben Izhak O, Nachtigal A, Burstein A, Shalom RB, Shagrawi I, Best LA 1995 Tumor, induced osteomalacia. Semin Arthritis Rheum 25:35–46. Edminster KA, Sundaram M 2002 Oncogenic osteomalacia. Semin Musculoskel Radiol 6:191–196. Rhee Y, Lee JD, Shin KH, Lee HC, Huh KB, Lim SK 2001 Oncogenic osteomalacia associated with mesenchymal tumour detected by indium-111 ocreotide scintigraphy. Clin Endocrinol 54:551–554.

993 128. Avila NA, Skarulois M, Rubino DM, Doppman JL 1996 Oncogenic osteomalacia: Lesion detected by MR skeletal survey. Am J Roentgenol 167:343–345. 129. Whyte MP, Podgornik MN, Wollberg VA, Eddy MC, McAlister WH 2001 Pseudo-(tumor-induced) rickets. J Bone Miner Res 16:1564–1571. 130. Salahudeen AK, Elliott RW, Ellis HA 1984 Osteomalacia due to ileal replacement of ureters: A report of 2 cases. J Urol 131:335–337. 131. Boutsen Y, Devogelaer JP, Malghem J, Noel H, Nagant de Deuxchais C 1996 Antacid-induced osteomalacia. Clin Rheumatol 15:75–80. 132. Kassem M, Eriksen EF, Melsen F, Mosekilde L 1999 Antacid-induced osteomalacia: A case report with a histomorphometric analysis. J Intern Med 229:275–279. 133. Berrocal T, Simon MJ, al-Assir I, Prieto C, Pastor I, de Pablo L, Lama R 1995 Shwachman-Diamond syndrome: Clinical, radiological, and sonographic aspects. Pediatr Radiol 25:289–292. 134. Cohen MM 2002 Some chondrodysplasias with short limbs: Molecular perspectives. Am J Med Genet 112:304–313. 135. Cohen MM 2000 Merging the old skeletal biology with the new. II. Molecular aspects of bone formation and growth. J Craniofac Genet Dev Biol 20:94–106. 136. Vohra P 1995 Metaphyseal chondrodysplasia: A differential diagnosis of rickets. Indian J Pediatr 62:131–132. 137. Rathbun JC, MacDonald JW, Robinson HMC, Wanklin JM 1961 Hypophosphatasia: Genetic study. Arch Dis Child 36:540–542. 138. Wedling D, Jeannin-Louys L, Kremer P, Fellmann F, Toussirot E, Mornet E. Adult hypophosphatasia. Current concepts. Joint Bone Spine 68:120–124. 139. Davies M, Adams PH 1978 The continuing risk of vitamin-D intoxication. Lancet 16:621–623. 140. Anning ST, Dawson J, Dolby DE, Ingram JT 1948 The toxic effects of calciferol. Q J Med 17:203–228. 141. Harris PWR 1969 An unusual case of calcinosis due to vitamin D intoxication. Guy’s Hospital Rev 118:553–541. 142. Irnell L 1969 Metastatic calcification of soft tissue on overdosage of vitamin D. Acta Med Scand 185:147–152. 143. Alazraki N 1995 Radionuclide Techniques. In: Resnjck D (Ed) Diagnosis of Bone and Joint Disorders. Saunders, Philadelphia, Pennsylvania, pp. 430–474. 144. Murray IPC 1994 Nuclear medicine in disorders of bone and joints: Growth and Metabolic Disorders. In: Murray IPC and Ell PJ (eds). Nuclear Medicine in Clinical Diagnosis and Treatment. Churchill Livingstone, London, pp. 1035–1040. 145. Lee HK, Sung WW, Solodnik P, Shimshi M 1995 Bone scan in tumor-induced osteomalacia. J Nucl Med 36:247–249. 146. Rosenthall L, Kaye M 1976 Observations in the mechanism of 99mTc-labeled phosphate complex uptake in metabolic bone disease. Semin Nucl Med 6:59–67. 147. Hain SF, Fogelman I 2002 Nuclear medicine studies in metabolic bone disease. Semin Musculoskel Radiol 6:323–329. 148. Mari C, Catafau A, Carrio I 1999 Bone scintigraphy and metabolic disorders. Q J Nucl Med 43:259–267. 149. McAfee JG 1987 Radionuclide imaging in metabolic and systemic skeletal diseases. Semin Nucl Med 17:334–349. 150. Singh BN, Spies SM, Mehta SP, Kesala BA, Quinn JL 1978 Unusual bone scan presentation of osteomalacia: symmetrical uptake—a suggestive sign. Clin Nucl Med 3:292–295. 151. Wu YW, Seto H, Shimizu M, Kageyama M, Watanabe N, Kakishita M 1995 Postgastrectomy osteomalacia with pseudofractures assessed by repeated bone scintigraphy. Ann Nucl Med 9:29–32.

994 152. Meindok H, Rapoport A, Oreopoulos DG, Rabinovich S, Meema HE, Meema S 1985 Quantitative radionuclide scanning in metabolic bone disease. Nucl Med Commun 6:141–148. 153. Nguyen BD, Wang EA 1999 Indium-111 pentetreotide scintigraphy of mesenchymal tumour with oncogenic osteomalacia. Clin Nucl Med 24:130 –131. 154. Jan de Beur SM, Streeten EA, Civelek AC, McCarthy EF, Uribe L, Marx SJ, Onobrakpeya O, Raisz LG, Watts NB, Sharon M, Levine M 2002 Localization of mesenchymal tumors by somatostatin receptor imaging. Lancet 359: 761–761. 155. Moran M, Paul A 2002 Ocreotide scanning in the detection of a mesenchymal tumor in the pubic symphysis causing hypophosphatemic osteomalacia. Int Orthop 26:61–62. 156. Fukumoto S, Takeuchi Y, Nagano A, Fujita T 1999 Diagnostic utility of magnetic resonance imaging skeletal survey in a patient with oncogenic osteomalacia. Bone 25:375–377. 157. Adams JE 1992 Osteoporosis and bone mineral densitometry. Curr Opin Radiol 4:11–19.

JUDITH E. ADAMS

158. Adams JE 1995 Quantitative measurements in osteoporosis. In: Tovey FI and Stamp TCB (eds). The measurement of metabolic bone disease. Parthenon Publishing Group, pp. 107–142. 159. Blake GM, Fogelman I 2002 Dual energy X-ray absorptiometry and its clinical applications. Semin Musculoskel Radiol 6:207–217. 160. Adams JE 2002 Dual-energy X-ray absorptiometry. In: Radiology of Osteoporosis. Grammp S (Ed). Medical Radiology. Springer-Verlag Berlin Heidelberg, pp. 87–100. 161. Guglielmi G, Lang TF 2002 Quantitative tomography. Semin Musculoskel Radiol 6:219–227. 162. Wahner HW 1987 Single- and dual-photon absorptiometry in osteoporosis and osteomalacia. Semin Nucl Med 17: 305–315. 163. Wishart J, Horowitz M, Need A, Nordin BE 1990 Relationship between forearm and vertebral mineral density in postmenopausal women with primary hyperparathyroidism. Arch Intern Med 150:1329–1331.

CHAPTER 61

The Pharmacology of Vitamin D, Including Fortification Strategies REINHOLD VIETH

Department of Laboratory Medicine and Pathobiology, University of Toronto, and Pathology and Laboratory Medicine, Mount Sinai Hospital, Toronto, Canada

I. Introduction II. Indications and Clinical Use: Potential Health Effects of Vitamin D III. Overview of the System of Vitamin D Metabolism, and Its Regulation IV. Dosage Considerations

V. Pharmacokinetic Principles, Volume of Distribution, Turnover and Half-life as It Pertains to Vitamin D VI. Vitamin D Toxicity and Safety Issues VII. Summary References

I. INTRODUCTION

that became our species, Homo sapiens, in our natural, tropical environment. The living conditions of modern humans differ from those that affected our evolution— we avoid exposure of skin to the vitamin D–forming UVB rays of sunshine, and even if we do spend time outdoors, many of us live at latitudes with comparatively little UVB. I contend that we, modern humans, might benefit if we could compensate for the biological consequences of modern life. One such consequence may be an endemic lack of vitamin D that can be corrected by appropriate supplementation. My usual perspective about vitamin D is North American, where vitamin D is primarily regarded as a nutrient. However, in Europe and in most of the rest of the world, the daily use of even 20 mcg (800 IU) of vitamin D is treated as if it were a prescription drug. This drug-oriented perspective has the advantage of imposing a higher expectation on our understanding of the use of vitamin D. Before approving any new drug, government regulators expect to see the answers to relatively standard questions. Pharmaceutical firms need to anticipate these issues as they plan the research necessary for implementation of new products. If vitamin D were a new drug, these questions would include, but are not limited to, the following:

Lack of cholecalciferol (vitamin D) in the diet can cause disease; therefore, vitamin D is a true nutrient in the full sense of the word. The notion that true nutrients may be available only from foods is a misconception; vitamin D, like niacin, is a vitamin that can be acquired without eating it [1]. Vitamin D is readily metabolized to calcidiol [25(OH)D], whose level is the accepted measure of vitamin D nutritional status [2]. 25(OH)D is a prehormone in the same sense that testosterone and T4 are, because like them, it is the circulating, immediate precursor of a signaling molecule [1]. The aim of this chapter is to offer the reader a fresh look at an antiquated nutrient, vitamin D, to address it from the perspective of pharmacology, as if it were a new drug that we need to understand fully so that we can exploit its potential for clinical medicine and to optimize health. Vitamin D differs from other nutrients because we have never had dietary intakes of vitamin D as a reasonable reference point for deciding on how much of this nutrient that people should be consuming. Compared to the 250 mcg (10,000 IU) of vitamin D that adults can obtain by exposing their full skin surface to the sunshine [3], foods contain small amounts of vitamin D (Table I). Our biology was designed by evolution for life in equatorial Africa. It is clear from Table I that foods containing a meaningful amount of vitamin D were not readily available to primates or to early humans. Thus, diet could not have played a role in determining human vitamin D requirements. Requirements for vitamin D were satisfied by the life of the naked ape VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

1a. What is the disease indication for the drug? 1b. What kind of clinical or health effects should we be looking for, based on preclinical animal and laboratory research? 2a. What are the most useful approaches to delivering the drug to people: the vehicle? 2b. What is the appropriate dosage, route of administration, and interval between doses? Copyright © 2005, Elsevier, Inc. All rights reserved.

996

REINHOLD VIETH

TABLE I Vitamin D Content of Selected Food Sources* Amount of Vitamin D per 100 g food, unless specified otherwise under Food Food

Microgram (µg)

According to NIH, USA [139] Cod Liver Oil, per 100 mL Salmon, cooked, 100 g or 31/2 oz Mackerel, cooked, 100 g or 31/2 oz Sardines, canned in oil, drained,100 g or 31/2 oz Eel, cooked, 100 g; 31/2 oz Milk, fortified with vitamin D, (250 mL) 1 cup Margarine, fortified, 1 tablespoon Dry breakfast cereal, if optionally fortified with vitamin D as permitted in the USA, one serving, 190 mL 3/4 cup Liver, beef, cooked, 100 g; 31/2 oz Egg, 1 whole (vitamin D is present in the yolk)

International units (IU)

219 9 9 7 5 2 2 1

9067 360 345 270 200 98 60 50

1 1

30 25

110 120 10 35 33 28 23 18 18 15 15 14 13 11

4400 4800 400 1400 1300 1100 920 720 720 600 600 560 520 440

Meats Beef (lean separated) Beef (total edible) Beef, liver Pork (lean separated) Pork (total edible) Pork, liver Chicken (breast) Chicken liver Turkey

0 0 0 1 1 0 0 0

0 0 0 28 50 0 8 4

Milk products (non-fortified) Cow’s milk Human milk Yogurt

0 0 0

13 13 0

According to A. Takeuchi et al., Japan [140] Fishes Anglerfish (liver only) Skijack (viscera perserves) Skipjack (whole meat) Indo-Pacific marlin Chum salmon Herring Flatfish Bastard halibut (cultured) Bluefin tuna (fatty meat) Grunt Rainbow trout Eel Red sea bream (cultured) Mackerel

*Amounts of Vitamin D shown in this table are per 100 g of the food, unless specified otherwise under food.

(Continued)

997

CHAPTER 61 The Pharmacology of Vitamin D, Including Fortification Strategies

TABLE I Vitamin D Content of Selected Food Sources*—Cont’d Amount of Vitamin D per 100 g food, unless specified otherwise under Food Food

Microgram (µg)

International units (IU)

1 0

24 0

18 3 3 6 0

720 100 120 230 0

400 400 16 4 4 3 1 0

16000 16000 640 160 140 100 50 16

Butter Cheese (cheddar) Eggs Duck egg Japanese quail egg Chicken egg (whole) Chicken egg (yolk) Chicken egg (white) Fungi Woody ear fungus (dried) Silver ear fungus (dried) Shitake (dried) Shimeji Matsutake Commom mushroom Winter fungus Nameko *Amounts of Vitamin D shown in this table are per 100 g of the food, unless specified otherwise under food.

3. What is the desirable target for the plasma concentration; what dose would be needed to attain or ensure this? 4. What, if any, are the biological markers to monitor toxicity, and what are our criteria for determining therapeutic effectiveness? What is the “therapeutic index”, the ratio between toxic vs beneficial dose levels?

and that I advocate higher requirements of vitamin D than most other investigators.

When it comes to plain and simple, nutritional vitamin D (cholecalciferol), the answer to each of these questions is that we have just started to address it in the past decade. Any opinion about vitamin D here is controversial. Lips presents a more conservative perspective on vitamin D requirements elsewhere [4,5] and also see Chapter 62. In an effort to provide some answers to the preceding questions, I will present my personal perspective about the vitamin D system that relates to pharmacological aspects of vitamin D in the adult context. My opinion about vitamin D requirements is not shared by all workers in the field, but I will provide data to support my views. However, the reader is cautioned that differences of opinion exist,

The only officially recognized indications for use of vitamin D in adults are the treatment and prevention of osteomalacia, bone loss, and fractures. Figure 1 summarizes randomized control trials looking at whether vitamin D, with or without calcium, affects risk of nonvertebral fracture. A recent, thorough literature review of vitamin D, 1,25(OH)2D and its analogs is also available, addressing the issue of osteoporosis prevention and treatment [6]. The purpose of Fig. 1 is to show that there has been no evidence that doses of vitamin D less than 800 IU/day are effective in preventing osteoporotic fractures. Although it makes sense conceptually, it is difficult to tell whether or not additional calcium is needed in

II. INDICATIONS AND CLINICAL USE: POTENTIAL HEALTH EFFECTS OF VITAMIN D

998

REINHOLD VIETH

Fracture-prevention studies with vitamin D3

100 800 IU/d

% Fewer fractures(100%× (1−RR))

90

?

80 70

Dawson-Hughes 1997 NEJM 337:670

60 50

Chapuy 1992 NEJM 327:1637 Chapuy 2002 Osteoporosis Int 13:257

40 30

Trivedi 2003 BMJ 326:469

20 10

Lips 1996 Ann Intern Med 124:400 Meyer 2002 J Bone Min Res 17:709

0 0

10 20 30 40 50 60 70 80 90 100 Vitamin D3 mcg/day

FIGURE 1 Summary of randomized-control clinical trials of fracture-prevention using vitamin D, with or without calcium. None of the studies using doses of vitamin D3 providing less than 20 mcg/day was effective in reducing fracture risk [90,137,137]. However, all the studies in which there was a reduction in fracture risk used approximately 20 mcg/day of vitamin D3 [7,10–12,135,137]. This dose includes the background intake; for the work by Dawson-Hughes, background intake was 5 mcg/day [11].

concert with vitamin D, because most studies have combined calcium and vitamin D for comparison to a placebo group receiving neither. There are now two randomized-controlled studies which show that vitamin D3 given by itself in doses of 100,000 IU (2500 µg) orally every 4 months [7], or 150,000 IU (3750 µg) by annual injection [8] reduces the occurrence of fractures. Bone density declines more quickly during winter than during summer [9]. Vitamin D supplements (about 20 mcg (800 IU) per day) combined with calcium, eliminate the faster fall in bone density during winter [9]. Furthermore, three studies show that the combination of calcium and 20 mcg vitamin D daily lower fracture risk in adults older than age 65 [10–12]. Occurrence of fractures is reduced by about a third, even within the first year of these studies, when bone density is not increased by enough to account for the fewer fractures [11]. The explanation for this may be that vitamin D improves muscle strength and balance. This reduces the occurrence of the falls that cause fractures [13–15]. In people younger than age 65, risk of osteoporotic fracture has been difficult to assess. Data from the Nurses Health Study suggest a 37% lower risk of osteoporotic fracture in postmenopausal women younger than 65, if they consume vitamin D in amounts of at least 12.5 mcg/day, compared to women consuming less than

3.5 mcg/day vitamin D [16]. The authors failed to detect any effect of calcium intake, but they suggested that in this cross-sectional study, women with a family history of osteoporosis would have been more likely to take supplemental calcium, confounding a calcium effect. I have a concern that 1,25(OH)2D may be used too aggressively as an alternative to improved vitamin D nutrition in the prevention or treatment of osteoporosis. The point that 1,25(OH)2D has a narrower margin of safety (therapeutic index) than vitamin D has never been raised in analyses comparing them [6,17]. If one cause of osteoporosis is that the vitamin D system is somehow deficient or defective [18], it makes little sense to resort to the use of 1,25(OH)2D. Rickets and osteomalacia may exist despite normal 1,25(OH)2D concentrations. Increases in vitamin D will not increase 1,25(OH)2D levels further [19–22]. As kidney function deteriorates, its endocrine capability also declines, and thus a low serum 1,25(OH)2D level reflects impaired renal function, not poor nutrition [22,23]. Despite many studies looking into the use of 1,25(OH)2D and its analogs to prevent or treat osteoporosis, the review of this topic by Papadimitropoulos concludes that there no reason for anyone to resort to any metabolite other than nutritional vitamin D [6]. I would add that this should be vitamin D3, and at a dose of at least 20 mcg/d.

A. Non-Bone Effects of Vitamin D Vitamin D nutrition probably affects health beyond just bone. The mechanisms involved in mediating the non-classic (i.e. non-bone) effects of vitamin D are probably through 1,25(OH)2D produced locally, using circulating 25(OH)D as the substrate (see Chapter 79). Many tissues possess 25(OH)D-1-alpha-hydroxylase activity, including the skin (basal keratinocytes and hair follicles), lymph nodes (granulomata), pancreas (islets), adrenal medulla, brain, pancreas, and colon [24]. An even wider range of tissues possess receptors for 1,25(OH)2D (VDR) [25]. All of this reveals a system for autocrine or paracrine regulation of tissue processes that involves the local production of 1,25(OH)2D [26]. Sufficient vitamin D nutrition, and hence, appropriate 25(OH)D concentration is essential to this local, paracrine role of 1,25(OH)2D that is not generally reflected in the circulating level of 1,25(OH)2D. The paracrine components of the vitamin D endocrine/paracrine systems account for the many effects of vitamin D nutrition and/or UVB light on health and disease prevention. Vitamin D has been implicated in a wide array of diseases (Table II). While all of the relationships with vitamin D in Table II are statistically significant, most of the evidence for a role of vitamin D is circumstantial.

999

CHAPTER 61 The Pharmacology of Vitamin D, Including Fortification Strategies

TABLE II Diseases and Conditions Known to Be, or Implicated as Being Prevented by Greater Vitamin D Nutrition or Skin UV Exposure Disease Rickets Osteomalacia Osteoporosis Poor calcium absorption Blood-pressure regulation Risk of diabetes Progression osteoarthritis Diminished intra-uterine growth Effects on brain development Resistance to pneumonia Multiple sclerosis, occurrence and progression Prevention of tuberculosis Prevent depression or SAD or improve mood Lessen risk/severity of fibromyalgia Protection against cancers Breast Prostate Large bowel

Type of evidence supporting the association Long established, curative Long established, curative Placebo-controlled, randomized studies that vitamin D prevents loss of bone density, and lessens fracture risk Modest increase in Vitamin D nutrition increases this Epidemiological and randomized interventional data Epidemiological and case-control data Epidemiological, cross-sectional studies Presumed effect Rat experiments Epidemiological association with rickets Epidemiological data, and lab effects on tissue Epidemiological data, and lab effects on tissue Small RCT’s 400 IU/day or 100000 in winter No mood effect of 400 IU/day Cross-sectional study Epidemiological data, and lab effects on tissue Epidemiological, and lab effects on tissue epidemiological and cross-sectional data, based on latitude and serum 25(OH)D

Epidemiological studies show that higher serum 25(OH)D, and/or environmental ultraviolet exposure is associated with lower rates of breast, ovarian, prostate, and colorectal cancers [27–34] (see Chapter 90). More recent statistical analyses also show significant relationships including non-Hodgkin’s lymphoma and cancer of the bladder, esophagus, kidney, lung, pancreas, rectum, stomach and corpus uteri [35]. Multiple sclerosis is more prevalent in populations having lower levels of vitamin D nutrition or ultraviolet exposure [32,36–38], and it has been proposed that vitamin D intake, ranging from 33–95 mcg (1,300 to 3,800 IU) per day, helps prevent the disease [38]. Established osteoarthritis progresses more slowly (is less severe) in adults with higher vitamin D nutritional status, with serum 25(OH)D that exceeds 75 nmol/L [39,40]. The prevalence of hypertension increases with population distance, north or south, from the equator [41]. Blood pressure is lowered in mildly hypertensive patients whose 25(OH)D levels are raised to over 100 nmol/L by tanning [42]. One randomized intervention study showing that vitamin D supplementation at 20 µg/d (800 IU/d) lowers blood pressure in elderly women [43]. The role of vitamin D in regulating the renin-angiotensin system is discussed in

Reference

[7,9–11] [161] [13,14,41,42] [46,141,142] [39,40] [143] [144] [45] [38,145,146] [147,148] [149,150] [151] [152] [153,154] [155] [64,154,156,157] [153,154,158]

Chapter by Li. Vitamin D deficiency may impair immune function in animals [44]. In children there is a strong association between pneumonia and nutritional rickets [45]; however, patients with rickets generally have soft ribs, so that pneumonia could have resulted from a mechanical respiratory inadequacy, not related to immune function. The concept that there is a connection between vitamin D nutrition and immune function is further supported by the apparent protective effect of improved vitamin D nutrition during infancy and childhood against type I diabetes mellitus [46]. The role of vitamin D on the immune system and immune-mediated disease is discussed in Chapter 36, 98, and 99. If any of these nontraditional effects of vitamin D were taken into account, they would result in a substantial upward revision of official recommendations for vitamin D beyond the current Adequate Intake (AI) values [47]. The level of evidence needed to make a health claim that can be sanctioned officially involves more than the circumstantial evidence of laboratory experiments and epidemiology. It requires direct intervention, the controlled administration of the agent to many healthy people, and showing an effect that stands up to statistical analysis. We need randomized intervention

1000

REINHOLD VIETH

trials to take this field beyond preclinical basic research and epidemiological evidence. There are ongoing randomized trials involving vitamin D that relate to cancer, multiple sclerosis, and osteoporosis, but for the most part, they deal with analogs of 1,25(OH)2D, not the nutrient. The nutrient has been very much overlooked for all purposes except rickets, osteomalacia, and osteoporosis. There are three reasons for this. First, the financial incentive lies with the proprietary analogs, driven by private funding that diverts the focus of investigators who are able to do such studies. Second, an optimized dose of vitamin D has never been established for adults. Therefore, “plain” vitamin D sometimes compares poorly with 1,25(OH)2D and its analogs, whose dose is more thoroughly optimized [17], and whose dose is usually designed to be very close to the point where it could cause hypercalcemia. Optimal doses of vitamin D probably vary, depending on the indication, so that one dose may not always be optimal. Third, the official misrepresentation that vitamin D2 and vitamin D3 are equal has resulted in efficacy studies at higher doses that usually involve vitamin D2 because high-dose commercial preparations of vitamin D are comprised of this. One example of this is worth looking at whether

A

Low input of Cholecalciferol from diet or UVB Metabolite “compartment” Vitamin D3 ->

25(OH)D ->

1

4

2

5 3

1,25(OH)2D ->

Within Tissues Possessing 1-OHase

In Plasma

6 24,25(OH)2D and catabolism->

vitamin D2 supplementation might prevent bone loss in steroid-treated patients [48,49]; the effects of vitamin D were marginal, but since plain and simple vitamin D3 was never part of the experimental protocol, the issue remains unresolved. Another example of the unfortunate focus on vitamin D2 instead of the D3 form is the recent Australian study using vitamin D2 at a substantial dose of 250 µg (10,000 IU) weekly, yet producing no significant effect on bone density preservation, and showing essentially no effect on serum 25(OH)D either [50].

III. OVERVIEW OF THE SYSTEM OF VITAMIN D METABOLISM, AND ITS REGULATION Administration of vitamin D is unusual in pharmacology or in endocrinology, because this molecule is two metabolic steps away from the biologically active agent, 1,25(OH)2D. Furthermore, the laboratory test to monitor dose is the concentration of a metabolite, 25(OH)D, and not the compound administered. As a way to provide a conceptual model for metabolic regulation in

B Legend 1 Liver mitochondrial vit D25-hydroxylase 2 Liver microsomal vit D-25hydroxylase 3 Renal 25(OH)D-1hydroxylase 4 Tissue (non-renal) 25(OH)D-1-hydroxylase 5 Renal mitochondrial 25(OH)D-24-hydroxylase 6 Non-renal 1,25(OH)2D-24hydroxylase A “unregulated” step in the flow of metabolism An regulated step in the flow of metabolism 7 Catabolism and excretion

High/normal input of Cholecalciferol from diet or UVB

1 Liver mitochondrial vit D25-hydroxylase

Metabolite “compartment” Vitamin D3 ->

25(OH)D->

1

4

2 Liver microsomal vit D-25hydroxylase

2

5 3

1,25(OH)2D ->

Within Tissues Possessing 1-OHase

In Plasma

6

24,25(OH)2D -> and catabolism

Legend

3 Renal 25(OH)D-1hydroxylase 4 Tissue (non-renal) 25(OH)D-1-hydroxylase 5 Renal mitochondrial 25(OH)D-24-hydroxylase 6 Non-renal 1,25(OH)2D-24hydroxylase

An “unregulated” step in the flow of metabolism Regulated step in the flow of metabolism 7 Catabolism and excretion

FIGURE 2 (A) Metabolism of vitamin D under conditions of low vitamin D supply. The vessels represent metabolic compartments, stages in the metabolism of vitamin D. The height of the shaded portion of each vessel represents the relative concentration of each metabolite indicated in the figure. This figure illustrates the concept that vitamin D metabolism in vivo functions below its Km, i.e., the system behaves according to the first-order reaction kinetics. Just as the flow of water through a hole in a pail reflects the height of water in that pail, the rates of metabolism in the vitamin D system reflect the concentration of precursor at each step. Open passages represent steps in metabolism in which the pertinent enzymes are relatively unregulated. Valves represent steps in metabolism in which there is regulation of flow at the enzyme (this regulation is usually through changes in the amount of enzyme protein in specific tissues, and not allosteric). When vitamin D supplies are low, the flow of 25(OH)D through other potential pathways is compromised to maintain the circulating concentration of 1,25(OH)2D at the level determined by the priority requirements of bone and mineral metabolism. (B) Metabolism of vitamin D under conditions of adequate vitamin D supply. When vitamin D supplies are adequate, the flow of 25(OH)D through other potential pathways, including its utilization by peripheral tissues for paracrine regulation, is no longer compromised. Higher 25(OH)D concentration makes available routes of metabolism other than the one path needed for bone and mineral metabolism. Furthermore, a higher supply of vitamin D leads to an upregulation of 24-hydroxylase and the catabolic pathways associated with it; this accelerates rate of metabolic clearance and metabolite turnover in each compartment.

1001

CHAPTER 61 The Pharmacology of Vitamin D, Including Fortification Strategies

The vitamin D system entails endocrine, paracrine, and autocrine functions that could have never been optimized to cope with the lack of vitamin D created by our modern culture of clothing and sun-avoidance. Inadequate supplies of vitamin D limit the local control that many tissues need so that they can function properly. The question is now being raised, whether the modern, “normal” prevalence of some of the diseases listed in Table II could be reduced substantially if we were to increase our intakes of vitamin D [56–59]. Control of metabolism in the vitamin D endocrine system is very different from the way other steroid hormones are regulated. For conventional steroid hormones, the concentration of substrate (cholesterol) is far higher than the substrate in the vitamin D system. Figure 3 illustrates the effective in-vivo Km of 1-hydroxylase, in relation to the physiological concentration range of its substrate. Our circulating cholesterol concentration is in the order of 5 million nmol/liter; in contrast, 25(OH)D typically circulates at less than 200 nmol/liter. Cholesterol concentration is not a rate-limiting aspect of the body’s capacity to generate steroid hormones; however, 25(OH)D concentration is absolutely ratelimiting for 1,25(OH)2D production. Two ways to increase production of hormone (1,25(OH)2D versus Steroid hormones) (1) More 25(OH)D [= D nutrition] or (2) More enzyme (Vmax) Vmax

Rate of 1,25(OH)2D synthesis

this system, Fig. 2 illustrates the metabolite “compartments” occupied by vitamin D after ingestion or exposure to sunshine. I will show later in this chapter that less than 25 percent of vitamin D that enters the body actually becomes 25(OH)D. More than 75 percent of vitamin D entering the circulation bypasses what we recognize as the vitamin D endocrine system. Instead, most vitamin D entering the circulation is excreted and/or metabolized by other routes not shown here, and most likely, excreted into the bile. Figure 2 consists of two panels to illustrate the metabolic adaptations that exist so that the vitamin D endocrine system can accommodate to a wide range in the substrate concentration. The vitamin D system is optimized to maintain plasma 1,25(OH)2D levels according to the requirements of calcium homeostasis. The earliest compromise to progressive restriction in vitamin D supply is probably a diminished capacity of nonrenal tissues to produce 1,25(OH)2D. Kidney possesses megalin, which facilitates substrate access [51] (see Chapter 10). Since few nonrenal tissues possess megalin, their activity of existing 1-hydroxylase depends primarily on the 25(OH)D concentration [52–55]. This results in a compromise at nonrenal tissues when 25(OH)D levels are low. This is illustrated in Figure 2 by the greater height of one of the valves (number 4 in the figure) representing nonrenal 1-hydroxylase on the “pail” (Fig. 2A vs 2B) that represents the 25(OH)D compartment. If one looks at the system of vitamin D metabolism in Fig. 2 from the perspective of a system designed to adapt to various inputs of the nutrient, it becomes clear that this is a system better designed to cope with an abundance of supply, not a lack of it. First, the flow of vitamin D toward 25(OH)D is remarkably inefficient—most vitamin D entering the body never becomes 25(OH)D. Second, there is no way to correct for deficiency of vitamin D, other than to redirect utilization of 25(OH)D toward 1,25(OH)2D production, which is the pathway most acutely important for life. That is, when supplies of vitamin D are severely restricted, its metabolism is directed only toward the maintenance of calcium homeostasis. To expand on the point that the system of vitamin D metabolism is effectively designed for adjusting for higher inputs, not lower inputs, I offer the example of an air-conditioner system. Air conditioners are designed to compensate for excessive heat, but they are a useless way to compensate for a cold environment. Human vitamin D metabolism was effectively designed through evolution and natural selection for people in an environment without clothing, and living at equatorial latitudes where UVB intensity is always enough to produce a relative abundance of vitamin D. In contrast, most modern humans cover close to 95 percent of skin surface and avoid sunshine.

Physiological range of 25(OH)D

Km

(2) More enzyme (Vmax)

Concentration 0

(1) More substrate

300

nmol/L 5,000,000 (Physiological cholesterol concentration)

Mass action

FIGURE 3 The difference in enzyme kinetics between the vitamin D endocrine system and the substrate supply for conventional steroid hormone systems based on cholesterol. The purpose of this figure is to emphasize that the range of physiologic concentration of 25(OH)D in mammals is less than the Michaelis-Menten constant (Km) of 1-hydroxylase that has been characterized in vitro [138] and in vivo [55]. There are two ways to improve capacity for 1,25(OH)2D production at kidney, and at peripheral tissues: provide more substrate, or increase 1-hydroxylase content of the tissue. This is fundamentally different from the situation relevant to every other part of the endocrine system. No other hormone is so dependent on the arbitrary, external supply of its structural raw material. The concept of a massaction relationship for 1,25(OH)2D production is the basis of the argument that operation of paracrine control systems dependent on vitamin D supply can be improved by improving vitamin D nutrition.

1002 In the acute situation, before adjustments can be made to 24-hydroxylase and catabolic pathways (before the “valves” in Figure 2 can be adjusted), the in vivo production of 1,25(OH)2D is directly proportional to circulating 25(OH)D concentration. In rats, the acute injection of 25(OH)D into the circulation produces a rapid, transient increase in 1,25(OH)2D, directly proportional to the percentage increase in 25(OH)D [55,60]. Since in vivo concentrations of 25(OH)D change slowly, over many months, this first-order relationship between 25(OH)D and 1,25(OH)2D is not normally seen in adults [22]. However, in situations where 1-hydroxylase is tonically stimulated, either because of primary hyperparathyroidism [61] or in granulomatous disease [62,63], modest increases in vitamin D supply will raise plasma 1,25(OH)2D concentration and aggravate hypercalcemia. The model of regulation represented by Figure 2 may help to explain the U-shaped risk curve for prostate cancer vs 25(OH)D for men in Nordic countries. Tuohimaa et al. reported the fascinating observation that in northern countries, the narrow range of 25(OH)D levels between 40–60 nmol/L coincides with the lowest risk of prostate cancer [64]. Although this topic is covered in other chapters of this book (Schwartz and Chen, Chapter 88; Feldman et al., Chapter 93), I propose a hypothesis that a falling 25(OH)D concentration is a non–steady-state situation during which it may not be possible to sustain 1,25(OH)2D at its long-term setpoint at tissues like the prostate that produce autocrine 1,25(OH)2D. High 25(OH)D concentrations may not be problematic per se, but those Nordic men with the highest summertime 25(OH)D levels should be expected to suffer the greatest decline in 25(OH)D during their long winters. This is based on a longitudinal, seasonal study of men in the USA [65]. During winter, steadily falling 25(OH)D concentrations create a need for prostate tissue to continuously increase the ratio of 1-hydroxylase and 24-hydroxylase so that optimal setpoint concentrations of tissue 1,25(OH)2D can be sustained. This hypothesis of a suboptimal setpoint during winter is not unlike what is experienced by a person taking a shower as the supply of hot water is running out; he or she must keep adjusting faucets to maintain water temperature—not a comfortable process. The hypothesis predicts that the U-shaped risk curve for prostate cancer is distinct to high latitudes where winters produce prolonged, gradual declines in 25(OH)D levels.

A. Role of Vitamin D Binding Protein (DBP)

REINHOLD VIETH

carrier proteins. Sex-steroid binding globulin, and glucocorticoid binding globulin each circulate at concentrations of about 150 nmol/L, in the same order of magnitude as their ligands [66]; in contrast, the concentration of vitamin D binding protein is 4700 nmol/L [67]; this represents a 50-fold excess over its vitamin Dderived ligands (see Chapters 8–9). The dynamics of 25(OH)D in tissues are remarkable. Its carrier protein, DBP, is cleared from plasma with a half-life of 1.7 d, which is shorter than the 5-day half life of albumin [68]. Within one hour after injection of radiolabeled DBP, the radiolabel is present in a greater concentration than in plasma, within kidney, liver, skeletal muscle, heart, lung, intestine, testis, and bone [68]. In contrast to DBP, its ligand, 25(OH)D, is cleared slowly from the body, with a half-life of about 10 days in both rabbit [68] and human [69]. The pool of DBP outside plasma is double the size of the intravascular DBP pool, and the molar replacement rate of DBP is reported to be 1,350-fold higher than that of 25(OH)D. The binding of 25(OH)D to DBP does not affect the turnover or tissue uptake of DBP [68]. As a short summary of the preceding, the DBP and/or DBP-25(OH)D complex is removed from plasma by a variety of tissues. The DBP is degraded during this process, and most 25(OH)D released within those tissues is recycled. The molar excess of DBP to 25(OH)D in plasma and the relatively rapid turnover of DBP indicate that a high capacity, high affinity, and dynamic transport mechanism for vitamin D sterols exists in plasma. Although 25(OH)D released into cells because of the metabolic clearance of DBP is recycled, the clearance of DBP provides ready access to vitamin D, 25(OH)D and its metabolites to the liver and kidney. These are the two organs most involved in the clearance of DBP, and the two organs central to the endocrine function of the vitamin D system. Recent new knowledge of the megalin/cubulin system has shed light on the mechanisms of DBPtissue interactions and tissue-specific uptake of DBP (chapters 8–10) [70]. Megalin and cubulin are cellsurface, endocytic receptors, members of the low-density lipoprotein receptor gene family. These help to regulate the concentration of ligands in the extracellular fluids and deliver metabolites to cells in need of these metabolites [71]. Differences in tissue distribution of these cell-surface proteins will affect the accessibility of different tissues to circulating 25(OH)D.

IV. DOSAGE CONSIDERATIONS A. Infants

Differences between steroid hormones and the vitamin D system are amplified further by the large differences in concentrations of their respective plasma

Cholecalciferol, or vitamin D3, given in the form of cod liver oil, has been a folk remedy in northern

CHAPTER 61 The Pharmacology of Vitamin D, Including Fortification Strategies

Europe since the 1700s [72]. Empirically, a small teaspoonful daily was thought to help infants thrive. This arbitrary dose of cod-liver oil has turned out to be a good guess, so far as infants are concerned. The 375 IU (9 µg) of vitamin D3 contained in that teaspoon [73] was confirmed relatively recently as being appropriate for infants [74,75]. A French study utilizing vitamin D2 concluded that neonates might need somewhat more, 1000 IU [76]. If safety of vitamin D during infancy is a concern, it should be kept in mind, that until the late 1960s, the recommended amount of vitamin D for infants in Finland was 2000 IU/day (50 µg/day). A large epidemiologic study suggests that this higher dose lowered risk of juvenile diabetes before age 30 years, by 85% compared to people not receiving vitamin D as infants [46]. Compared to the adult, vitamin D nutrition in the infant and child has been well characterized, and it is the focus of Chapters 48,49,65. There is also an excellent review of the field available by Chesney [77]. The present chapter focuses on the pharmacology of vitamin D in the adult.

B. Adults Until it became clear that vitamin D was important to the health of adults, there was very little thought directed at how much vitamin D adults might need to consume. Until recently, there has been no consensus about what the objective criteria should be for appropriate vitamin D nutrition. In England, an adult recommendation of 2.5 µg (100 IU)/day was established simply because 7 women with severe nutritional osteomalacia showed a response to this amount [78]. Interestingly, the oils of different fish contain different amounts of vitamin D. For example, a teaspoon full of halibut liver oil contains twice as much vitamin D3 as does cod liver oil. If it had been halibut liver oil used in the past, recommendations for vitamin D supplementation could well have been double what they have been through most of the last century. Into the 1960s, the absence of overt rickets or osteomalacia was the only criterion that vitamin D nutrition was adequate [79]. By that same criterion of bone deformity, anthropologists consider vitamin D nutrition to have been a relatively minor problem for ancient populations. The lack of evidence of bone deformity in ancient populations is now explained by the new concept that the lack of vitamin D resulted in a natural selection for white skin color to prevent rickets and osteomalacia within defined environments [80]. Women with osteomalacia would have produced few offspring because rickets and osteomalacia produce a misshapen pelvis. Those women able to produce enough vitamin D to

1003

prevent rickets and osteomalacia would have been the vast majority in any region. Survival depended on adequacy of vitamin D nutrition, and at latitudes away from the equator, natural selection for lighter skin color helped to ensure adequacy for the quality of pelvis needed for vaginal birth. In the 1960s, an expert committee on vitamin D could provide only anecdotal support for “the hypothesis of a small requirement” for vitamin D in adults and recommended one-half the infant dose, to ensure that adults obtain some from the diet [79]. Despite the new knowledge uncovered since that time, dietary vitamin D recommendations for adults have remained very conservative, and still derive from amounts established for neonates. In contrast to the way decisions are made about the dose of any new drug, recommendations for vitamin D have been arbitrary, because there was no firm evidence on which to base decisions. However, even though the evidence about the effects of vitamin D dosages on adult health have become characterized scientifically, those with the final say in setting official nutrient guidelines (not the experts they consulted) continued to focus on lower doses of vitamin D than had been shown effective in the fracture prevention trials discussed previously. The revised recommendations were referred to as the “adequate intake” (AI), because there was no published evidence of efficacy for them [2,47]. The objective measure of vitamin D nutritional status is the 25-hydroxyvitamin D (25(OH)D) concentration in serum or plasma [2]. The consensus on this point has made it possible for researchers to focus on a measurable target when it comes to vitamin D nutrition. Table III summarizes two views of the relationships between long-term vitamin D intakes and the anticipated range of 25(OH)D concentration associated with them. Figure 4 is a dose-response curve to showing the final average 25(OH)D concentrations attained in studies reported in the literature [3,56]. Table IV summarizes incremental responses to different treatment strategies to raise 25(OH)D to steady-state concentrations. Responsiveness to vitamin D administration, as measured by the nmol per liter increase per mcg consumption per day, increases with: a) lower vitamin D dosage, b) lower initial 25(OH)D concentration; c) longer duration of supplementation, suggesting a long half-life and time to plateau. The conventional approach to improving vitamin D nutritional status has been to give either vitamin D3 or vitamin D2 (ergocalciferol). Until recently, availability of 25(OH)D was another option (supply of this product has been discontinued by Organon, NJ, USA). The company’s discontinuation of 25(OH)D may have made sense, because the objective of increasing plasma 25(OH)D concentrations can be almost as easily achieved by providing enough vitamin D3.

1004

REINHOLD VIETH

TABLE III The Clinical Interpretation of Serum 25(OH)D Levels and the Estimated Intakes of Vitamin D Needed to Ensure These Levels (1 µg = 40 IU)

Serum 25(OH)D nmol/L Serum 25(OH)D ng/mL

Deficiency (rickets and osteomalacia)

Insufficiency (increased PTH secretion, osteoporosis)

0–25 0–10

25–40 10–16

Sufficiency 40–100 16–40

Daily intake of Vitamin D3 per day needed to reach the 25(OH)D above: Food and Nutrition Boarda 0 mcg 5–10 mcg (200–400 IU) From the literature reviewedb 0–5 µg 10–15 (200 IU) (400–600 IU)

Desirable (suppress PTH, optimize calcium absorption) 75–160 30–64

5–15 mcg not stated (200–600 IU) 25–100 µg 100–250 µg (1000–4000 IU) (4000–10,000 IU)

Toxic/Therapeutic (might increase urine and serum calcium) >220 >88 ≥95 mcg (3800 IU) >1000 mcg (>40000 IU)

aImplications drawn from current National Academy of Sciences nutritional guidelines that the stated intake will deliver the level of adequacy, i.e. 25(OH)D concentration indicated [2]. The “adequate intake” recommendations for vitamin D vary according to age: adults < 50, 5 mcg/day; 50–70 years, 10 mcg/day; > 70 Years, 15 mcg/day. There is no RDA for vitamin D. bBased on literature [3,5,118,135].

Nonetheless, useful perspectives can be gained from previous experience with 25(OH)D. Barger-Lux and Heaney et al. have shown that as 25(OH)D dosage increases, there is effectively a linear increase in the average 25(OH)D concentration achieved (Table IV).

However, when vitamin D3 is used, the increment in 25(OH)D per mcg per day of vitamin D3 decreases as the dose increases. Since the increase in plasma 25(OH)D concentration per mcg dose is at least four times higher for 25(OH)D

TABLE IV Strategies to Increase Circulating 25(OH)D Concentration in Adults: Effects of Compound, Dose, and Duration1

Compound 25(OH)D3 25(OH)D3 25(OH)D3 Cholecalciferol Cholecalciferol Cholecalciferol Cholecalciferol Cholecalciferol Cholecalciferol Cholecalciferol Cholecalciferol Cholecalciferol Cholecalciferol Cholecalciferol Ergocalciferol

25(OH)D nmol/L increase per µg/day

DOSE µg/day

Duration of dose wks

Absolute increase in 25(OH)D nmol/L

4.1 4.0 3.8

50 10 20

4 4 4

206.4 40 76.1

[119] [119] [119]

1.5 1.4 1.1 1.1 0.8 0.8 0.7 0.6 0.6 0.5 0.5 0.3

15 20 25 21 100 25 138 275 250 100 1250 36

52 8 8 20 52 20 20 20 8 20 8 104

22 27 28.6 23.4 81 19 102.7 169.8 146 51.8 643

[159] [160] [119] [135] [159] [118] [135] [135] [119] [118] [119] [50]

Reference

1The results in this table represent recent work not included in Figure 4. These data were assembled to permit comparison of efficacy dose of different strategies for increasing 25(OH)D concentration. The results are sorted in order of decreasing response to the dose, based on the nmol/L increase in 25(OH)D per mcg/day of oral doses used in these studies. These are studies done during winter and/or comparisons versus parallel control groups.

CHAPTER 61 The Pharmacology of Vitamin D, Including Fortification Strategies

administration than for vitamin D3 administration, we can conclude that fewer than 25 percent of vitamin D molecules ever become 25(OH)D. At least three quarters of the molecules of vitamin D that enter the body are removed by some other fate.

C. The Case Against Ergocalciferol, Vitamin D2 Vitamin D is available in two forms for nutritional supplementation, ergocalciferol (vitamin D2) and cholecalciferol (vitamin D3). Vitamin D2 is manufactured by exposing a fat extract of yeast to UV light. Since no metabolite of vitamin D2 is normally detectable in the blood of humans or primates [81,82], I contend that this should be regarded as a drug, and not a physiological compound. The present discussion focuses on vitamin D3 cholecalciferol, the form of vitamin D naturally present in mammals. Vitamin D3 (from here on, vitamin D) is the more potent form of vitamin D in all primate species and in man [81,82]. Comparisons between the two versions of vitamin D [82], and the meta-analysis of effects on 25(OH)D (Fig. 4) indicate that vitamin D3 is about 4 times as potent as vitamin D2, i.e., 1 µg of D3 = approximately 4 µg of D2.

Serum or plasma 25(OH)D nmol/L

400

1000

Vitamin D intake IU/day 4000 40000

400000

Study group mean data Vit D2-Treated group mean data Individuals, Vit D Hypercalcemia

FIGURE 4

100 1000 Vitamin D intake µg/day

Nonetheless, vitamin D2 continues to be used clinically as if it is equivalent, since official guidelines [2] and pharmacopeas respond slowly to new evidence. The presumption of equivalence is based on 60-year old studies of rickets prevention in infants—evidence recognized as weak, even at the time [73,83]. The older rat data suggesting that vitamin D2 and vitamin D3 were equivalent lose their meaning when it is noted that the rat-line tests last done over 50 years ago were bioassays to establish units for the quantity of vitamin D not readily measured in any other way [84]. For a bioassay yielding “units,” equivalence is not the same thing as equivalence per milligram or per mole. Furthermore, all species tested show differences between the vitamin D2 and vitamin D3 [85,86]. Despite these obvious problems about units, the very conservative approach used by those who frame official statements about nutrients has remained unchanged, that one international unit of vitamin D is equivalent to 25 nanograms of either vitamin D2 or vitamin D3 [2,84]. In Australia, vitamin D3 has never been licensed for use, and the only nutritional form of vitamin D available is vitamin D2. I have summarized the differences between vitamin D2 and vitamin D3 in Table V. Based on the many major differences between the two, it is clear that unless there is some well-characterized reason to favor vitamin D2 (I am not aware of any), all use of vitamin D for nutritional and clinical purposes should in my opinion specify cholecalciferol, vitamin D3.

D. UVB Light on Human Skin as a Dose of Vitamin D

100

10

1005

1E4

Dose-response relationship between daily vitamin D intake and mean 25(OH)D concentration, based on data published in the literature. The solid points show mean results for groups of adults consuming the indicated doses of vitamin D. Results for groups of adults that are unambiguously consuming vitamin D2 are shown by the circled points. Vitamin D3 is about 4 times as potent as vitamin D2, based on tracing the circled points for subjects consuming vitamin D2 back to the trend-line based on vitamin D3. Both axes are log scale. The results, represented by Xs, are for individuals showing the classic hypercalcemic response to toxic levels of prolonged vitamin D consumption. The data used to generate this graph were compiled and published previously [3,56].

In any discussion of vitamin D pharmacology or dosage, it would be a major oversight to ignore the role of sunshine, particularly its UVB component. As described elsewhere in this book (Chapter 3), the synthesis of vitamin D is a self-limiting reaction, reaching an equilibrium after 20–25 min of summer UVB exposure for people with white skin, and producing no net increase in vitamin D production after that [87]. Darker skin requires longer exposure, but the potential yield of vitamin D is the same. Exposure of full skin surface to UVB light, in an amount less than erythemal, is equivalent to a vitamin D consumption of about 250 µg (10,000 IU)/day [88–91]. Lifeguards in the United States and in Israel, as well as farmers in the Caribbean, all exhibit serum 25(OH)D concentrations greater than 100 nmol/L [92–94]. Furthermore, even regular short periods in sun-tan parlors consistently raise serum 25(OH)D well beyond 80 nmol/L [42,95–100]. The highest 25(OH)D concentrations in the groups of adults acquiring vitamin D physiologically (via UV

1006

REINHOLD VIETH

TABLE V

The Case Against Vitamin D2, Compared to Vitamin D3

Vitamin D2 Not detectable in humans or primates unless administered from an external source Vitamin D binding protein has lower affinity for vitamin D2 than for vitamin D3 and its metabolites Generates metabolites for which there is no vitamin D3 equivalent Microsomal 25-hydroxylase does not act on it Per mole of dose, 25(OH)D increases by less than with vitamin D3 The 25(OH)D response to vitamin D2 is less in the elderly than in younger adults All known cases of iatrogenic toxicity with vitamin D involved the vitamin D2 form (albeit, formulations > 25 µg (1000 IU) have usually been vitamin D2) Dose preparations are less stable

exposure) range up to 235 nmol/L [42,92], and none of these studies imply that such 25(OH)D levels have caused hypercalcemia. Since humans evolved as naked apes, whose native habitat was within 30 degrees latitude of the equator, I contend that our genome was selected under conditions of abundant vitamin D supply [3]. As such, it is reasonable to think that the substantially lower levels of 25(OH)D prevalent among modern humans have been accompanied by biological compromises, such as increased PTH secretion [22] and altered cellular metabolism [26]. By now, these compromises may have been detrimental to the health of modern humans for so long, that we are no longer in a position to realize them. Barger-Lux and Heaney studied the effect of sunshine on healthy outdoor workers in the US Midwest, relating it to the vitamin D intakes needed to bring about the 25(OH)D levels observed [65]. They concluded that for these men, the summertime supply of vitamin D from sunshine was approximately 70 µg (2800 IU)/day. This supply during summer did not ensure sufficiency through the winter, when 25(OH)D fell to less than 50 nmol/liter in 3 of 26 subjects and less than 75 nmol/liter in 15 of 26 subjects.

V. PHARMACOKINETIC PRINCIPLES, VOLUME OF DISTRIBUTION, TURNOVER AND HALF-LIFE AS IT PERTAINS TO VITAMIN D A complete understanding of the pharmacokinetics of the vitamin D system has eluded researchers. This is

Vitamin D3 The natural metabolite generated within skin and the oils of fur

Ref. [162] [163]

Substrate for both microsomal and mitochondrial 25-hydroxylases 25(OH)D response to vitamin D3 is the same for young vs older adults All known adult cases of toxicity with vitamin D3 have been unintentional, “industrial” accidents

[164] [165,166] [82] [160,167] [159] [3,110] [121] [109] [71,82]

because of the technical issue of measuring the nanomolar quantities of vitamin D potentially embedded within tissues or excreted in catabolized forms. It is extremely difficult to detect or to measure vitamin D and its metabolites when they exist among great excesses of other lipids. Perhaps the most careful study into the fate of physiological amounts of cholecalciferol was reported by Lawson et al. [101,102]. They exposed rats with shaved skin to ultraviolet light (UVB), and measured vitamin D and 25(OH)D in tissues at various times afterwards. Although adipose tissue concentration of vitamin D was never greater than the plasma concentration, it contained the largest exchangeable pool of vitamin D. Recovery of vitamin D3 in adipose was less than 5 percent of the amount produced within the skin [101], and this low recovery was attributed to vitamin D excretion into the bile. Lawson et al estimated that the volume of distribution of unmetabolized vitamin D3 was approximately four liters per kg (based on concentration decay curves from plasma and total amounts recovered from tissues). Vitamin D is not detectable in the adipose tissue of normal rats [102,103], but with administration of pharmacologic doses [104], or shaving of fur to increase yield fivefold [101], vitamin D is detectable. Brouwer et al estimate the half-life of vitamin D in rat adipose tissue to be 96 days, which is plausible because it compares with the functional halflife of 25(OH)D in humans [3]. In contrast, Lawson et al estimated the vitamin D in rat adipose tissue to have a half-life of 13.8 days [101]. The more rapid half-life reported by Lawson et al was likely due to the younger age of the rats. Pharmacokinetic studies are extremely difficult with vitamin D, since unlike a drug, vitamin D is present in

CHAPTER 61 The Pharmacology of Vitamin D, Including Fortification Strategies

the body naturally. It is impossible to start with completely deprived individuals to do appropriate studies of pharmacokinetics, and the half-life is exceptionally long. Furthermore, the component of nutritional interest is 25(OH)D, a metabolite of either vitamin D2 or vitamin D3. Studies using isotopic techniques show that in humans, molecules of 25(OH)D have a plasma half-life of about 10 days [69,105]. However, a more practical measure of the half-life of 25(OH)D is reflected in the rate at which 25(OH)D concentrations decline upon the sudden elimination of sources of vitamin D (acute deprivation of ultraviolet light). Two studies show that when sailors embark upon two-month-long missions in submarines, the 25(OH)D concentration decreases by approximately 50 percent [3,106,107]. Follow-up of 25(OH)D concentrations in adults who have been intoxicated with vitamin D3 suggest that the functional in vivo half-life is of the order of several months [108–110]. During summer, we can accumulate and store vitamin D well enough so that supplies for vitamin D do not become completely depleted during the winter months. Within three days of a dose of vitamin D, very little of the original vitamin D is detectable in the plasma of rats [111] or humans [112]. Most vitamin D initially entering the circulation appears to be excreted into the bile. The highest total concentrations of vitamin D and its metabolites occur in plasma. However, since plasma represents only 2.5% of body mass, larger pools of vitamin D and 25(OH)D exist in fat and muscle [113–115]. When there is a continuous supply of vitamin D, like the situation for people who regularly expose a large proportion of their skin surface to tropical sunshine, the equilibrium state maintains a balance between vitamin D stored within body compartments and the removal from tissue stores for metabolism and clearance. Under these physiologic, sun-derived circumstances, 25(OH)D concentrations in plasma sustain levels of more than 200 nmol/L [3]. At these levels of vitamin D nutrition, there has never been a concern raised that sudden loss of adipose tissue would either raise 25(OH)D or predispose to vitamin D toxicity. Likewise, despite 70 years of experience with the oral use of vitamin D in amounts that exceed the amounts tenable through sun exposure, there has never been a report of an eventual, sudden excess of vitamin D caused by release from adipose stores because of weight loss.

A. Body Storage of Vitamin D and Inefficient Conversion to 25(OH)D It is thought that since vitamin D is a fat-soluble vitamin, it must show preferential accumulation in

1007

adipose tissue [116,117]. Two studies showed that following a defined dose of vitamin D or sunshine, the rise in 25(OH)D was less for obese individuals than for people who weighed less. These studies did not show that adipose tissue concentrated vitamin D, and they failed to account for the obvious effect of a larger body compartment size, which should produce a lower concentration of anything, regardless of whether adipose plays a role or not. In our study using vitamin D3 doses of 100 µg/day in adults, we found no correlation between weight and serum 25(OH)D [118]. At physiological doses, cholecalciferol (unmetabolized vitamin D3) distributes widely into tissues, not just to adipose, but to skeletal muscle and other organs as well [101,115]. As stated above, turnover of vitamin D stored in tissues produces a long half-life of about two months. The amounts of vitamin D recoverable from tissue stores account for only a fraction of the dose administered [101]. The animal data indicate that more than 75% of the molecules of vitamin D that enter the body are catabolized and excreted without ever being stored in tissues, and without ever becoming 25(OH)D. The human data also support this. In humans, when vitamin D or 25(OH)D are given over the long-term, to achieve an equilibrium concentration of 25(OH)D, it takes more than 4 times as much vitamin D to produce the same 25(OH)D plateau [119]. By definition, at that plateau in 25(OH)D, exchange of vitamin D with tissues is at equilibrium where release of stored vitamin D equals storage of new vitamin D. Still the fourfold difference in efficacy at sustaining 25(OH)D exists when comparing effects of doses of 25(OH)D and vitamin D. That is, the difference in efficacy between 25(OH)D and vitamin D at sustaining plasma 25(OH)D concentrations cannot be explained by deposition of vitamin D into storage sites. The difference in efficacy at sustaining 25(OH)D can only be explained by the loss of vitamin D entering the circulation to fates other than 25-hydroxylation or storage.

VI. VITAMIN D TOXICITY AND SAFETY ISSUES Amounts of vitamin D substantially greater than physiologic amounts >250 µg/day (>10,000 IU/day) are toxic because they saturate circulating vitamin D binding protein (DBP), and they force the percent of vitamin D that is free and unbound to increase [3,120]. At toxic doses, the freely circulating vitamin D, along with its metabolites, accumulate in adipose [104] and muscle [115]. The 100 µg (4000 IU)/day of vitamin D we have used in adults is physiologic and far below what would be needed to change the free fraction of

1008 vitamin D or its circulating metabolites [67]. The average capacity of human plasma DBP to bind vitamin D and its metabolites is 4700 nmol/L [67], and this exceeds by 20 times the physiologic total concentration of its vitamin D–derived ligands. The vast majority of cases of vitamin D intoxication have involved vitamin D2 [3]. The situations involving vitamin D3 to date, have been industrial accidents [109,120,121] or poisonings from an unknown source [110]. In our case, we assayed blood levels of vitamin D and its metabolites by chromatography and found that despite record-high 25(OH)D concentrations in humans (2,400 nmol/liter), they were still small in comparison to a large excess of vitamin D3 (17,000 nmol/liter), suggesting that the capacity of the liver to hydroxylate vitamin D is limited [110]. Like anything that has an effect on living things, vitamin D can be harmful if taken in excess. I contend that the ratio of the physiologically effective dose vs. the toxic level for vitamin D is similar to the safety margin of many other nutrients (including even water). The reason vitamin D has been perceived as toxic was probably because daily ingestion in the milligram (>1000 µg) range has caused harm. In contrast, milligram amounts of other nutrients are benign. Toxicity in normal adults requires intake of more than 1,000 µg (40,000 IU)/day, which reflects amounts of vitamin D that are four times more than the 250 µg (10,000 IU)/day that can be acquired naturally by sunshine [3]. In what I see as an overreaction to the potential for toxicity with vitamin D, the current recommendation of 5 µg (200 IU)/day (called an “Adequate Intake” in North America) for adults under age 50 represents what can only be regarded as a homeopathic dose—about 2% of what adults with white skin can make within 20 min of summer sun. In other words, the fear of vast excess has resulted in physiologically miniscule intake recommendations for adults. Concentrations of 1,25(OH)2D are not increased much by vitamin D intoxication. This reflects the high level of regulation of this hormone via both its synthesis and catabolism. Nonetheless, vitamin D toxicity is probably manifest by the excessive levels of “free” 1,25(OH)2D, displaced from its carrier protein, DBP, by the vast excess of other vitamin D metabolites [122]. This excess was confirmed by studies looking into “free” 1,25(OH)2D concentrations in vitamin D intoxicated individuals [120]. This excess of metabolite over binding capacity was also confirmed by the high total of vitamin D and 25(OH)D concentrations (19,500 nmol/L) in a patient intoxicated after consuming over a million units (>25000 µg) daily for many months [110]. It is also likely that very high levels of 25(OH)D can bind,

REINHOLD VIETH

even if modestly, to the vitamin D receptor to trigger a response as part of the mechanisms for toxicity [123]. We recently reported a safety evaluation of vitamin D3 supplementation of normal adults, involving daily consumption of 100 µg (4,000 IU). Contrary to the hypercalcemia in normal adults reported by Narang [124], and which was used by the Food and Nutrition Board to establish the 50 µg/d (2,000 IU/day) upper limit for vitamin D intake, 100 µg (4000 IU)/day produced no detectable change in serum or urine calcium in more rigorous studies [118,125]. The official safety limit for vitamin D intake without supervision by a physician is referred to as the “upper limit” (UL) [125,126]. This is the amount of vitamin D that the general public can take safely on a long-term basis with no anticipation of harm. Guidelines in both North America [2] and Europe [127] have established the UL as 50 µg (2000 IU)/day. This is a very conservative value that seems to remain constant, despite accumulating evidence to show that higher intakes are safe. The value of 50 µg (2000 IU)/day has remained unchanged since it was mentioned in the 1968 Recommended Dietary Allowance publication as a dose approaching a harmful level [128]. To sustain the very conservative approach of making minimal changes to past recommendations, the only thing to change over the years has been the safety margin applied to the evidence. For example, when the “no observed adverse effect level” (the highest dose shown to have no harmful effect) was 2400 IU/day, based on the Narang study [124], the safety factor applied by the United States food and nutrition Board was 1.2. When subsequent data were published indicating that 4000 IU/day was safe, the safety margin was increased by The European Commission to a value of 2.0 [127]. Recent evidence in men shows that eight weeks of supplementation with 275 µg (12,500 IU]/day of vitamin D does not affect circulating calcium concentration (urine results were not reported) [129]. That is, the dose is non-hypercalcemic, and safe by the safety criterion applied to drug studies of vitamin D analogs [130–132]. Even with the application of a safety factor of 2.75, this would suggest that 100 µg (4000 IU) of vitamin D could be a safe adult UL for vitamin D. I predict that past conservative patterns will remain for officially mandated nutrition guidelines, and that the response to evidence of greater safety of vitamin D will be to adjust safety factor for deriving the UL, so that the UL can stay unchanged at 50 µg (2000 IU)/day until there is solid research showing evidence of a need for higher intakes. The weight of published evidence on toxicity shows that the lowest dose of vitamin D proven to cause

1009

CHAPTER 61 The Pharmacology of Vitamin D, Including Fortification Strategies

hypercalcemia in some healthy adults is 1,000 µg (40,000 IU)/day of the vitamin D2 form [3] (Fig. 4). This translates to 1,000 micrograms, or 1 milligram, taken daily for many months. If a consumer wanted to achieve this toxic dose, he or she would need to take 40 of the 1000 unit pills (the highest dose available in North America without a prescription) every day for many months. One must bear in mind that few studies have looked at effects of high doses of vitamin D on

hypercalciuria, which would be a more sensitive indicator of toxicity [118]. Ten years ago, a dairy in the Boston area, servicing 10,000 households, made prolonged, gross errors in fortifying milk with several milligrams per quart. The case was published quickly [133] and covered by the media. The more rigorous epidemiological follow-up was published later. That showed that the situation contributed to two deaths of susceptible elderly people [121].

TABLE VI Opinions and Best Guesses at the Answers to Pharmacologic Questions that Need to be Addressed in Relation to Vitamin D Nutrition Question 1a. What is the indication for use of the compound? 1b. What kind of clinical or health effects should we be looking for, based on preclinical animal and laboratory research? 2a. What are the most useful approaches to increasing vitamin D status?

2b. The dosage, route of administration, and

2c. Interval between doses?

3. What is the desirable target for the plasma concentration; what dose would be needed to attain or ensure this?

4. What, if any, are the biological markers to monitor toxicity, and what are our criteria for determining therapeutic effectiveness? What is the “therapeutic index”, the ratio between toxic vs beneficial dose levels?

Answer Prevention and cure of rickets and osteomalacia. Fracture prevention; preservation of bone mineral density; normalization of PTH levels Disease prevention: cancer, autoimmune conditions, diabetes, multiple sclerosis, high blood pressure, fibromyalgia, muscle strength, mood. Encouragement to expose a large percent of skin surface to summertime sunshine, 10 min daily for white skin, up to 6 times longer for black skin. Fortification of foods to physiologically meaningful levels of vitamin D, consumption of supplement preparations with vitamin D. Dosage depends on the target concentration of 25(OH)D desired. We can assume a rule of thumb, that a dose of 1 mcg/day vitamin D increases 25(OH)D by 1 nmol/L, after 8 months of use (See Table IV). In non-SI units, this is the equivalent to saying that 100 IU/day increases 25(OH)D by one ng/mL. Oral vitamin D is probably more effective than injection. Since the half-life for decline in 25(OH)D is effectively 2 months, doses of vitamin D could be given monthly (we use weekly in our studies). Less frequent dosing than once every 2 months will generate large fluctuations in 25(OH)D concentrations that may not be desirable, because the enzymes involved in the regulation of 25(OH)D metabolism are functioning in a first-order relationship with substrate. Current consensus points to a goal of ensuring that 25(OH)D levels be higher than 70–100 nmol/L (28–40 ng/mL) [5]. This range reflects average 25(OH)D levels seen in adults taking 25 µg (1,000 IU)/day vitamin D, and who are getting sunshine. To ensure this level for those of the normal population with the weakest response to vitamin D, we need to aim for an average 25(OH)D concentration of about 120 nmol/L (48 ng/mL). This objective requires an intake of 100 µg (4,000 IU)/day for all adults.1 Hypercalcemia is the classic criterion for toxicity of vitamin D, its metabolites, and their analogues. “Non-calcemic” doses are those that do not cause hypercalcemia, and are considered “safe” by conventional criteria. However, the most sensitive clinical index of safety is urine calcium. This is easily monitored by measuring a morning urine calcium/creatinine ratio (normal for this would be mmol/mmol < 1; or in non-SI units, mg/mg < 0.35) [118]. Of greater concern for the long-term use of vitamin D, its metabolites, or analogs, should be the effects on soft-tissue calcification, within aorta, kidney, or other tissues. These effects may be seen radiologically in humans, or by direct measure of tissue calcium in preclinical animal studies.

1Please note that this is not an official recommendation, but rather a scientific opinion offered by the author for research purposes. The dosage of vitamin D required for humans remains a controversial issue.

1010 While hypercalcemia did occur, it was not widespread. By far the most susceptible group to the excess vitamin D was women over age 65 years of age, suggesting that diminished renal function may play a role. The average 25(OH)D concentration of the confirmed cases of vitamin D toxicity was 900 nmol/L (214 ng/mL) [121]; in comparison, physiologically attained 25(OH)D concentrations, obtained through sunshine exposure, can reach 235 nmol/L safely, without hypercalcemia or hypercalciuria. When physiologically higher vitamin D nutrition is associated with hypercalcemia, this reflects aberrant control of 25(OH)D-1-hydroxylase. This would reflect either primary hyperparathyroidism, where PTH continuously stimulates the enzyme in the kidney [61], or granulomatous disease, where peripheral tissue may not have the ability to regulate the 1-hydroxylase that normally serves autocrine/paracrine roles [3,134]. In people with abundant sun exposure (25(OH)D > 150 nmol/L), the presupplement supply of vitamin D could be equivalent to about 100 µg (4000 IU)/day [65]. If such people were to take an additional dosage by mouth of 100 µg/day of vitamin D, this would still be less than the dose of vitamin D shown to be safe in a recent study [135]. Since long-term vitamin D consumption of at least 1000 mcg/d would be needed to cause hypercalcemia, there is a large margin of safety with 100 µg (4000 IU)/day. One concern sometimes expressed, is that if adipose tissue were to break down, a sudden influx of vitamin D from adipose might be toxic [136]. In both rats and cattle, high doses of vitamin D are needed before vitamin D ends up as detectable in adipose tissue [104,115]. Despite being present in “significant” amounts in tissues, storage in tissues in not efficient. As a proportion of what enters the body via the skin or the diet, the amounts of vitamin D stored in adipose are a fraction of the total. In normal humans, adipose tissue content of vitamin D has been reported to be as high as 116 ng/g (approx 5 IU/g adipose) [102]. In cattle intoxicated with 7.5 million IU vitamin D (to cause hypercalcemia, in an experimental process to activate proteases to make beef more tender after slaughter), muscle levels of vitamin D reached 91 ng/g tissue (4 IU/g). The highest tissue level reported in those animals was in the liver, which contained vitamin D at 979 ng/g (39 IU/g) [115]. The point is that while there is “meaningful” storage of vitamin D in tissues, all the evidence indicates that only a fraction of any vitamin D dose ends up in tissues to be withdrawn at later times. If there were a sudden breakdown of 1 Kg of adipose tissue, or liver, this would release as much as 979 µg (39,000 IU) of vitamin D into the body. A toxic excess of vitamin D would require the breakdown daily of

REINHOLD VIETH

one Kg of adipose tissue that had been primed by prior vitamin D intoxication, with daily adipose catabolism to continue for several weeks. When toxic doses of vitamin D are administered, the effect will be manifest during the period of administration. There is no evidence that enough residual vitamin D can be stored in adipose tissue that vitamin D toxicity could possibly arise at some later time, because of weight loss.

VII. SUMMARY To conclude, I return to the pharmacological questions posed at the start of this chapter, and offer Table VI as a way to address the issues, based on the material in this chapter.

References 1. Vieth R 2004 Why “Vitamin D” is not a hormone, and not a synonym for 1,25-dihydroxy-vitamin D, its analogs or deltanoids. J Steroid Biochem Mol Biol 89–90:571–3 2. Standing Committee on the Scientific Evaluation of Dietary Reference Intakes 1997 Dietary reference intakes: calcium, phosphorus, magnesium, vitamin D, and fluoride. National Academy Press. 3. Vieth R 1999 Vitamin D supplementation, 25-hydroxyvitamin D concentrations, and safety. Am J Clin Nutr 69(5): 842–856. 4. Lips P 2004 Which circulating level of 25-hydroxyvitamin D is appropriate? J Steroid Biochem Mol Biol 89–90: 611–4. 5. Dawson-Hughes B, Heaney R, Lips P, Meunier P, Vieth R 2004 Vitamin D Round Table. In: Dawson-Hughes B, Heaney R, Burckhardt P eds. Nutritional Aspects of Osteoporosis. Academic Press, New York. 6. Papadimitropoulos E, Wells G, Shea B et al 2002 Metaanalyses of therapies for postmenopausal osteoporosis. VIII: Meta-analysis of the efficacy of vitamin D treatment in preventing osteoporosis in postmenopausal women. Endocr Rev 23(4):560–569. 7. Trivedi DP, Doll R, Khaw KT 2003 Effect of four monthly oral vitamin D3 (cholecalciferol) supplementation on fractures and mortality in men and women living in the community: randomized double blind controlled trial. BMJ 326:469–475. 8. Heikinheimo RJ, Inkovaara JA, Harju EJ et al. 1992 Annual injection of vitamin D and fractures of aged bones. Calcif Tissue Int 51(2):105–110. 9. Dawson-Hughes B, Dallal GE, Krall EA, Harris S, Sokoll LJ, Falconer G 1991 Effect of vitamin D supplementation on wintertime and overall bone loss in healthy postmenopausal women [see comments]. Annals of Internal Medicine 115:505–512. 10. Chapuy MC, Arlot ME, Duboeuf F et al 2002 Vitamin D3 and calcium to prevent hip fractures in the elderly women. N Engl J Med 327(23):1637–1642. 11. Dawson-Hughes B, Harris SS, Krall EA, Dallal GE 1997 Effect of calcium and vitamin D supplementation on bone

CHAPTER 61 The Pharmacology of Vitamin D, Including Fortification Strategies

12.

13.

14. 15. 16.

17. 18. 19.

20.

21.

22.

23.

24. 25.

26. 27. 28.

density in men and women 65 years of age or older. N Engl J Med 337(10):670–676. Chapuy MC, Pamphile R, Paris E et al 2002 Combined calcium and vitamin D3 supplementation in elderly women: confirmation of reversal of secondary hyperparathyroidism and hip fracture risk: the Decalyos II study. Osteoporos Int 13(3):257–264. Pfeifer M, Begerow B, Minne HW, Abrams C, Nachtigall D, Hansen C 2000 Effects of a short-term vitamin D and calcium supplementation on body sway and secondary hyperparathyroidism in elderly women. J Bone Miner Res 15(6):1113–1118. Bischoff HA, Stahelin HB, Dick W et al 2003 Effects of vitamin D and calcium supplementation on falls: a randomized controlled trial. J Bone Miner Res 18(2):343–351. Janssen HC, Samson MM, Verhaar HJ 2002 Vitamin D deficiency, muscle function, and falls in elderly people. Am J Clin Nutr 75(4):611–615. Feskanich D, Willett WC, Colditz GA 2003 Calcium, vitamin D, milk consumption, and hip fractures: a prospective study among postmenopausal women. Am J Clin Nutr 77(2):504–511. Lau KW, Baylink DJ 1999 Vitamin D therapy of osteoporosis: plain vitamin D therapy versus active vitamin D analog (D-hormone) therapy. Calcif Tissue Int 65(4):295–306. Heaney RP 1999 Lessons for nutritional science from vitamin D [editorial; comment]. Am J Clin Nutr 69(5):825–826. Bouillon RA, Auwerx JH, Lissens WD, Pelemans WK 1987 Vitamin D status in the elderly: seasonal substrate deficiency causes 1,25-dihydroxycholecalciferol deficiency. Am J Clin Nutr 45(4):755–763. Himmelstein S, Clemens TL, Rubin A, Lindsay R 1990 Vitamin D supplementation in elderly nursing home residents increases 25(OH)D but not 1,25(OH)2D. Am J Clin Nutr 52:701–706. Landin-Wilhelmsen K, Wilhelmsen L, Wilske J et al 1995 Sunlight increases serum 25(OH) vitamin D concentration whereas 1,25(OH)2D3 is unaffected. Results from a general population study in Goteborg, Sweden (The WHO MONICA Project). Eur J Clin Nutr 49(6):400–407. Vieth R, Ladak Y, Walfish PG 2003 Age-related changes in the 25-hydroxyvitamin D versus parathyroid hormone relationship suggest a different reason why older adults require more vitamin D. J Clin Endocrinol Metab 88(1):185–191. Ishimura E, Nishizawa Y, Inaba M et al 1999 Serum levels of 1,25-dihydroxyvitamin D, 24,25-dihydroxyvitamin D, and 25-hydroxyvitamin D in nondialyzed patients with chronic renal failure. Kidney Int 55(3):1019–1027. Zehnder D, Bland R, Williams MC et al 2001 Extrarenal Expression of 25-Hydroxyvitamin D3-1alpha-Hydroxylase. J Clin Endocrinol Metab 86(2):888–894. Zineb R, Zhor B, Odile W, Marthe RR 1998 Distinct, tissuespecific regulation of vitamin D receptor in the intestine, kidney, and skin by dietary calcium and vitamin D [In Process Citation]. Endocrinology 139(4):1844–1852. Chen TC, Holick MF 2003 Vitamin D and prostate cancer prevention and treatment. Trends Endocrinol Metab 14(9): 423–430. Lefkowitz ES, Garland CF 1994 Sunlight, vitamin D, and ovarian cancer mortality rates in US women. Int J Epidemiol 23(6):1133–1136. Martinez ME, Giovannucci EL, Colditz GA et al 1996 Calcium, vitamin D, and the occurrence of colorectal cancer among women. J Natl Cancer Inst 88(19):1375–1382.

1011

29. Tangrea J, Helzlsouer K, Pietinen P et al 1997 Serum levels of vitamin D metabolites and the subsequent risk of colon and rectal cancer in Finnish men. Cancer Causes Control 8(4): 615–625. 30. Garland CF, Garland FC, Gorham ED 1991 Can colon cancer incidence and death rates be reduced with calcium and vitamin D? Am J Clin Nutr 54(1 Suppl):193S-201S. 31. Emerson JC, Weiss NS 1992 Colorectal cancer and solar radiation. Cancer Causes & Control 3:95–99. 32. Schwartz GG 1992 Multiple sclerosis and prostate cancer: what do their similar geographies suggest? Neuroepidemiology 11:244–254. 33. Hanchette CL, Schwartz GG 1992 Geographic patterns of prostate cancer mortality. Evidence for a protective effect of ultraviolet radiation. Cancer 70:2861–2869. 34. Ainsleigh HG 1993 Beneficial effects of sun exposure on cancer mortality. [Review]. Preventive Medicine 22: 132–140. 35. Grant WB 2002 An estimate of premature cancer mortality in the U.S. due to inadequate doses of solar ultraviolet-B radiation. Cancer 94(6):1867–1875. 36. Hayes CE 2000 Vitamin D: a natural inhibitor of multiple sclerosis. Proc Nutr Soc 59(4):531–535. 37. McGrath J 2001 Does “imprinting” with low prenatal vitamin D contribute to the risk of various adult disorders? Med Hypotheses 56(3):367–371. 38. Hayes CE, Cantorna MT, DeLuca HF 1997 Vitamin D and multiple sclerosis. Proc Soc Exp Biol Med 216(1):21–27. 39. McAlindon TE, Felson DT, Zhang Y et al 1996 Relation of dietary intake and serum levels of vitamin D to progression of osteoarthritis of the knee among participants in the Framingham Study. Ann Intern Med 125(5):353–359. 40. Lane NE, Gore LR, Cummings SR et al 1999 Serum vitamin D levels and incident changes of radiographic hip osteoarthritis: a longitudinal study. Study of Osteoporotic Fractures Research Group. Arthritis Rheum 42(5):854–860. 41. Rostand SG 1997 Ultraviolet light may contribute to geographic and racial blood pressure differences. Hypertension 30(2 Pt 1):150–156. 42. Krause R, Buhring M, Hopfenmuller W, Holick MF, Sharma AM 1998 Ultraviolet B and blood pressure. Lancet 352(9129):709–710. 43. Pfeifer M, Begerow B, Minne HW, Nachtigall D, Hansen C 2001 Effects of a short-term vitamin D3 and calcium supplementation on blood pressure and parathyroid hormone levels in elderly women. J Clin Endocrinol Metab 86(4): 1633–1637. 44. McMurray DN, Bartow RA, Mintzer CL, HernandezFrontera E 1990 Micronutrient status and immune function in tuberculosis. Annals of the New York Academy of Sciences 587:59–69. 45. Muhe L, Lulseged S, Mason KE, Simoes EA 1997 Casecontrol study of the role of nutritional rickets in the risk of developing pneumonia in Ethiopian children. Lancet 349(9068):1801–1804. 46. Hypponen E, Laara E, Reunanen A, Jarvelin MR, Virtanen SM 2001 Intake of vitamin D and risk of type 1 diabetes: a birth-cohort study. Lancet 358(9292):1500–1503. 47. Vieth R, Fraser D 2002 Vitamin D insufficiency: no recommended dietary allowance exists for this nutrient. CMAJ 166(12):1541–1542. 48. Adachi JD, Bensen WG, Bianchi F et al 1996 Vitamin D and calcium in the prevention of corticosteroid-induced osteoporosis: a 3-year followup. J Rheumatol 23(6):995–1000.

1012 49. Adachi JD, Ioannidis G 1999 Calcium and vitamin D therapy in corticosteroid-induced bone loss: what is the evidence? Calcif Tissue Int 65(4):332–336. 50. Cooper L, Clifton-Bligh PB, Nery ML et al 2003 Vitamin D supplementation and bone mineral density in early postmenopausal women. Am J Clin Nutr 77(5):1324–1329. 51. Moestrup SK, Verroust PJ 2001 Megalin- and cubilinmediated endocytosis of protein-bound vitamins, lipids, and hormones in polarized epithelia. Annu Rev Nutr 21:407–428. 52. Chen TC, Schwartz GG, Burnstein KL, Lokeshwar BL, Holick MF 2000 The in vitro evaluation of 25-hydroxyvitamin D3 and 19-nor-1alpha,25-dihydroxyvitamin D2 as therapeutic agents for prostate cancer [In Process Citation]. Clin Cancer Res 6(3):901–908. 53. Reichel H, Bishop JE, Koeffler HP, Norman AW 1991 Evidence for 1,25-dihydroxyvitamin D3 production by cultured porcine alveolar macrophages. Mol Cell Endocrinol 75(2):163–167. 54. Reichel H, Koeffler HP, Norman AW 1987 25-Hydroxyvitamin D3 metabolism by human T-lymphotropic virustransformed lymphocytes. J Clin Endocrinol Metab 65(3): 519–526. 55. Vieth R, McCarten K, Norwich KH 1990 Role of 25-hydroxyvitamin D3 dose in determining rat 1,25-dihydroxyvitamin D3 production. American Journal of Physiology 258(5 Pt 1): E780–9. 56. Vieth R 2001 Vitamin D nutrition and its potential health benefits for bone, cancer, and other conditions. Journal of Nutrition and Environmental Medicine 11:275–291. 57. Barthel HR, Scharla SH 2003 Benefits beyond the bones— vitamin D against falls, cancer, hypertension and autoimmune diseases. Dtsch Med Wochenschr 128(9):440–446. 58. Holick MF 2003 Vitamin D: A millenium perspective. J Cell Biochem 88(2):296–307. 59. Zittermann A, Schleithoff SS, Tenderich G, Berthold HK, Korfer R, Stehle P 2003 Low vitamin D status: a contributing factor in the pathogenesis of congestive heart failure? J Am Coll Cardiol 41(1):105–112. 60. Vieth R, Milojevic S 1995 Moderate vitamin D3 supplementation lowers serum 1,25-dihydroxy-vitamin D3 in rats. Nutrition Research 15(5):725–731. 61. Vieth R, Bayley TA, Walfish PG, Rosen IB, Pollard A 1991 Relevance of vitamin D metabolite concentrations in supporting the diagnosis of primary hyperparathyroidism. Surgery 110(6):1043–6; discussion 1046–7. 62. Bell NH 1991 Endocrine complications of sarcoidosis. Endocrinology & Metabolism Clinics of North America 20:645–654. 63. Morita R, Yamamoto I, Takada M, Ohnaka Y, Yuu I 1993 Hypervitaminosis D. [Review] [Japanese]. Nippon Rinsho— Japanese Journal of Clinical Medicine 51:984–988. 64. Tuohimaa P, Tenkanen L, Ahonen M et al 2004 Both high and low levels of blood vitamin D are associated with a higher prostate cancer risk: a longitudinal, nested casecontrol study in the Nordic countries. Int J Cancer 108(1): 104–108. 65. Barger-Lux MJ, Heaney RP 2002 Effects of above average summer sun exposure on serum 25-hydroxyvitamin D and calcium absorption. J Clin Endocrinol Metab 87(11): 4952–4956. 66. Dunn JF, Nisula BC, Rodbard D 1981 Transport of steroid hormones: binding of 21 endogenous steroids to both testosterone-binding globulin and corticosteroid-binding globulin in human plasma. J Clin Endocrinol Metab 53(1):58–68.

REINHOLD VIETH 67. Vieth R 1994 Simple method for determining specific binding capacity of vitamin D–binding protein and its use to calculate the concentration of “free” 1,25-dihydroxyvitamin D. Clin Chem 40:435–441. 68. Haddad JG, Fraser DR, Lawson DE 1981 Vitamin D plasma binding protein. Turnover and fate in the rabbit. J Clin Invest 67(5):1550–1560. 69. Vicchio D, Yergey A, O’Brien K, Allen L, Ray R, Holick M 1993 Quantification and kinetics of 25-hydroxyvitamin D3 by isotope dilution liquid chromatography/thermospray mass spectrometry. Biol Mass Spectrom 22(1):53–58. 70. Nykjaer AWTE 2002 The low-density lipoprotein receptor gene family: a cellular Swiss army knife? TRENDS in Cell Biology 12:273–280 71. Nykjaer A, Dragun D, Walther D et al 1999 An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) vitamin D3. Cell 96(4):507–515. 72. Rajakumar K 2003 Vitamin D, cod-liver oil, sunlight, and rickets: a historical perspective. Pediatrics 112(2):e132-e135. 73. Park EA 1940 The therapy of rickets. The J American Medical Association (JAMA) 115(5):370–379. 74. Cooke R, Hollis B, Conner C, Watson D, Werkman S, Chesney R 1990 Vitamin D and mineral metabolism in the very low birth weight infant receiving 400 IU of vitamin D. Journal of Pediatrics 116:423–428. 75. Pittard WB, Geddes KM, Hulsey TC, Hollis BW 1991 How much vitamin D for neonates? Am J Dis Child 145(10): 1147–1149. 76. Zeghoud F, Vervel C, Guillozo H, Walrant-Debray O, Boutignon H, Garabedian M 1997 Subclinical vitamin D deficiency in neonates: definition and response to vitamin D supplements. Am J Clin Nutr 65(3):771–778. 77. Chesney RW 2001 Vitamin D deficiency and rickets. Rev Endocr Metab Disord 2(2):145–151. 78. Dent CE, Smith R 1969 Nutritional osteomalacia. Q J Med 38(150):195–209. 79. Blumberg RW, Forbes GB, Fraser D et al 1963 The prophylactic requirement and the toxicity of vitamin D. Pediatrics 31:512–525. 80. Jablonski NG, Chaplin G 2000 The evolution of human skin coloration. J Hum Evol 39(1):57–106. 81. Marx SJ, Jones G, Weinstein RS, Chrousos GP, Renquist DM 1989 Differences in mineral metabolism among nonhuman primates receiving diets with only vitamin D3 or only vitamin D2. J Clin Endocrinol Metab 69(1282): 1282–1289. 82. Trang H, Cole DE, Rubin LA, Pierratos A, Siu S, Vieth R 1998 Evidence that vitamin D3 increases serum 25-hydroxyvitamin D more efficiently than does vitamin D2. The American Journal of Clinical Nutrition 68:854–848. 83. Bicknell F, Prescott F 1946 Vitamin D. The antirachitic or calcifying vitamin. In: Bicknell F, Prescott F (eds) Vitamins in Medicine. Whitefriars Press, London, pp. 630–707. 84. Norman AW 1972 Problems relating to the definition of an international unit for Vitamin D and its metabolites. J Nutrition 102:1243–1246. 85. Horst RL, Napoli JL, Littledike ET 1982 Discrimination in the metabolism of orally dosed ergocalciferol and cholecalciferol by the pig, rat, and chick. Biochem J 204(1): 185–189. 86. Marx SJ, Jones G, Weinstein RS, Chrousos GP, Renquist DM 1989 Differences in mineral metabolism among nonhuman primates receiving diets with only vitamin D3 or only vitamin D2. J Clin Endocrinol Metab 69(6):1282–1290.

CHAPTER 61 The Pharmacology of Vitamin D, Including Fortification Strategies

87. Webb AR, DeCosta BR, Holick MF 1989 Sunlight regulates the cutaneous production of vitamin D3 by causing its photodegradation. J Clin Endocrinol Metab 68:882–887. 88. Stamp TC 1975 Factors in human vitamin D nutrition and in the production and cure of classical rickets. Proc Nutr Soc 34(2):119–130. 89. Davie MW, Lawson DE, Emberson C, Barnes JL, Roberts GE, Barnes ND 1982 Vitamin D from skin: contribution to vitamin D status compared with oral vitamin D in normal and anticonvulsant-treated subjects. Clin Sci 63(5):461–472. 90. Chel VG, Ooms ME, Popp-Snijders C et al 1998 Ultraviolet irradiation corrects vitamin D deficiency and suppresses secondary hyperparathyroidism in the elderly. J Bone Miner Res 13(8):1238–1242. 91. Holick MF 1995 Environmental factors that influence the cutaneous production of vitamin D. Am J Clin Nutr 61(3 Suppl):638S-645S. 92. Haddock L, Corcino J, Vazquez Md 1982 25(OH)D serum levels in the normal Puerto Rican population and in subjects with tropical sprue and parathyroid disease. Puerto Rico Health Sciences Journal 1:85–91. 93. Haddad JG, Kyung JC 1971 Competitive protein-binding radioassay for 25-hydroxycholecalciferol. Journal of Clinical Endocrinology 33:992–995. 94. Better OS, Shabtai M, Kedar S, Melamud A, Berenheim J, Chaimovitz C 1980 Increased incidence of nephrolithiasis in lifeguards in Israel. In: Massry SG, Ritz E, Jahreis G (eds) Phosphate and Minerals in Health and Disease. Plenum Press, New York, pp. 467–472. 95. Matsuoka LY, Wortsman J, Hollis BW 1990 Suntanning and cutaneous synthesis of vitamin D3. J Lab Clin Med 116(1):87–90. 96. Mawer EB, Berry JL, Sommer-Tsilenis E, Beykirch W, Kuhlwein A, Rohde BT 1984 Ultraviolet irradiation increases serum 1,25-dihydroxyvitamin D in vitamin-D–replete adults. Miner Electrolyte Metab 10(2):117–121. 97. Stamp TC, Haddad JG, Twigg CA 1977 Comparison of oral 25-hydroxycholecalciferol, vitamin D, and ultraviolet light as determinants of circulating 25-hydroxyvitamin D. Lancet 1(8026):1341–1343. 98. Dent CE, Round JM, Rowe DJ, Stamp TC 1973 Effect of chapattis and ultraviolet irradiation on nutritional rickets in an Indian immigrant. Lancet 1(7815):1282–1284. 99. Varghese M, Rodman JS, Williams JJ et al. The effect of ultraviolet B radiation treatments on calcium excretion and vitamin D metabolites in kidney stone formers. Clin Nephrol 31(5):225–231. 100. Falkenbach A, Unkelbach U, Boehm BO et al. 1993 Bone metabolism before and after irradiation with ultraviolet light. Eur J Appl Physiol 66(1):55–59. 101. Lawson DE, Sedrani SH, Douglas J 1986 Interrelationships in rats of tissue pools of cholecalciferol and 25-hydroxycholecalciferol formed in u.v. light. Biochem J 233(2): 535–540. 102. Lawson DE, Douglas J, Lean M, Sedrani S 1986 Estimation of vitamin D3 and 25-hydroxyvitamin D3 in muscle and adipose tissue of rats and man. Clin Chim Acta 157(2):175–181. 103. Vieth R. Reply to FAJ Muskiet et al 2001 Am J Clin Nutr 74(6):863–864. 104. Brouwer DA, van Beek J, Ferwerda H et al 1998 Rat adipose tissue rapidly accumulates and slowly releases an orallyadministered high vitamin D dose. Br J Nutr 79(6):527–532. 105. Clements MR, Davies M, Hayes ME et al 1992 The role of 1,25-dihydroxyvitamin D in the mechanism of acquired vitamin D deficiency. Clin Endocrinol (Oxf) 37(1):17–27.

1013

106. Preece MA, Tomlinson S, Ribot CA et al 1975 Studies of vitamin D deficiency in man. Q J Med 44(176):575–589. 107. Dlugos DJ, Perrotta PL, Horn WG 1995 Effects of the submarine environment on renal-stone risk factors and vitamin D metabolism. Undersea Hyperb Med 22(2):145–152. 108. Adams JS, Lee G 1997 Gains in bone mineral density with resolution of vitamin D intoxication. Ann Intern Med 127(3):203–206. 109. Koutkia P, Chen TC, Holick MF 2002 Vitamin D intoxication associated with an over-the-counter supplement. N Engl J Med 345(1):66–67. 110. Vieth R, Pinto T, Reen BS, Wong MM 2002 Vitamin D poisoning by table sugar. Lancet 359:672. 111. Vieth R, Chan A, Pollard A 1995 125I-RIA kit cannot distinguish vitamin D deficiency as well as a more specific assay for 25-hydroxyvitamin D. Clinical Biochemistry 28:175–179. 112. Stanbury SW, Mawer EB 1980 The metabolism of a physiological dose of radioactive cholecalciferol (vitamin D3) to its hydroxylated metabolites in man. Clin Sci (Colch) 58(6): 523–535. 113. Mawer EB, Backhouse J, Holman CA, Lumb GA, Stanbury SW 1972 The distribution and storage of vitamin D and its metabolites in human tissues. Clin Sci 43(3):413–431. 114. Rosenstreich SJ, Rich C, Volwiler W 1971 Deposition in and release of vitamin D3 from body fat: evidence for a storage site in the rat. J Clin Invest 50(3):679–687. 115. Montgomery JL, Parrish FC, Jr., Beitz DC, Horst RL, Huff-Lonergan EJ, Trenkle AH 2000 The use of vitamin D3 to improve beef tenderness. J Anim Sci 78(10):2615–2621. 116. Arunabh S, Pollack S, Yeh J, Aloia JF 2003 Body fat content and 25-hydroxyvitamin D levels in healthy women. J Clin Endocrinol Metab 88(1):157–161. 117. Wortsman J, Matsuoka LY, Chen TC, Lu Z, Holick MF 2000 Decreased bioavailability of vitamin D in obesity. Am J Clin Nutr 72(3):690–693. 118. Vieth R, Chan PC, MacFarlane GD 2001 Efficacy and safety of vitamin D3 intake exceeding the lowest observed adverse effect level. Am J Clin Nutr 73(2):288–294. 119. Barger-Lux MJ, Heaney RP, Dowell S, Chen TC, Holick MF 1998 Vitamin D and its major metabolites: serum levels after graded oral dosing in healthy men. Osteoporos Int 8(3):222–230. 120. Pettifor JM, Bikle DD, Cavaleros M, Zachen D, Kamdar MC, Ross FP 1995 Serum levels of free 1,25-dihydroxyvitamin D in vitamin D toxicity. Ann Intern Med 122(7):511–513. 121. Blank S, Scanlon KS, Sinks TH, Lett S, Falk H 1995 An outbreak of hypervitaminosis D associated with the overfortification of milk from a home-delivery dairy. Am J Public Health 85(5):656–659. 122. Vieth R 1990 The mechanisms of vitamin D toxicity. Bone & Mineral 11:267–272. 123. Lou YR, Laaksi I, Syvala H et al 2003 25-Hydroxyvitamin D3 is an active hormone in human primary prostatic stromal cells. FASEB J, pp. 124. Narang NK, Gupta RC, Jain MK, Aaronson K 1984 Role of vitamin D in pulmonary tuberculosis. Journal of Association of Physicians of India 32(2):185–186. 125. Munro I 2001 Derivation of tolerable upper intake levels of nutrients. Am J Clin Nutr 74(6):865–867. 126. Hathcock JN 2001 Tolerable upper intake level of vitamin D. Am J Clin Nutr 74(6):865–867. 127. Health & Consumer Protection Directorate-General. Opinion of the Scientific Committee on Food on the Tolerable Upper Intake Level of Vitamin D. European Commission, editor.

1014

128. 129. 130.

131. 132. 133. 134. 135.

136. 137.

138. 139.

140.

141. 142.

143. 144. 145. 146.

http://europa.eu.int/comm/food/fs/sc/scf/out157_en.pdf-Accessed August 11, 2003. 2002. Brussels, Belgium. 1–10–0030. Recommended Dietary Allowances 1968 Seventh Revised Edition ed. Washington, D.C.: National Academy Press. Heaney RP 2003 Quantifying human calcium absorption using pharmacokinetic methods. J Nutr 133(4):1224–1226. Bouillon R, Verstuyf A, Verlinden L, Eelen G, Mathieu C 2003 Prospects for vitamin D receptor modulators as candidate drugs for cancer and (auto)immune diseases. Recent Results Cancer Res 164:353–356. Posner GH 2002 Low-calcemic vitamin D analogs (deltanoids) for human cancer prevention. J Nutr 132(12): 3802S–3803S. van den Bemd GJ, Chang GT 2002 Vitamin D and vitamin D analogs in cancer treatment. Curr Drug Targets 3(1):85–94. Jacobus CH, Holick MF, Shao Q et al 1992 Hypervitaminosis D associated with drinking milk. New England Journal of Medicine 326:1173–1177. Bell NH 1998 Renal and nonrenal 25-hydroxyvitamin D1alpha-hydroxylases and their clinical significance. J Bone Miner Res 13(3):350–353. Heaney RP, Davies KM, Chen TC, Holick MF, Barger-Lux MJ 2003 Human serum 25-hydroxycholecalciferol response to extended oral dosing with cholecalciferol. Am J Clin Nutr 77(1):204–210. Muskiet FA, Dijck-Brouwer DJ, van d, V, Schaafsma A 2001 Do we really need 100 mcg vitamin D/day, and is it safe for all of us? Am J Clin Nutr 74(6):862–863. Lips P, Graafmans WC, Ooms ME, Bezemer PD, Bouter LM 1996 Vitamin D supplementation and fracture incidence in elderly persons. A randomized, placebo-controlled clinical trial. Ann Intern Med 124(4):400–406. Vieth R, Fraser D 1979 Kinetic behavior of 25-hydroxyvitamin D-1-hydroxylase and -24-hydroxylase in rat kidney mitochondria. J Biol Chem 254(24):12455–12460. NIH 2003 Facts about dietary supplements: Vitamin D. Clinical Nutrition Service,Warren Grant Magnuson Clinical Center, National Institute of Health (NIH), Bethesda, MD, in conjunction with the Office of Dietary Supplements (ODS) in the Office of the Director of NIH. National Institute of Health. 12-12-0030. Takeuchi A, Okano T, Ishida Y, Kobayashi T 1995 Effects of dietary vitamin D intake on plasma levels of parathyroid hormone and vitamin D metabolites in healthy Japanese. Miner Electrolyte Metab 21(1-3):217–222. Stene LC, Ulriksen J, Magnus P, Joner G 2000 Use of cod liver oil during pregnancy associated with lower risk of Type I diabetes in the offspring. Diabetologia 43(9):1093–1098. Eva JK 1999 Vitamin D supplement in early childhood and risk for Type I (insulin-dependent) diabetes mellitus. The EURODIAB Substudy 2 Study Group. Diabetologia 42(1):51–54. Fuller K 2000 Lactose, rickets, and the coevolution of genes and culture. Human Ecology 28(3):471–477. Eyles D, Brown J, Mackay-Sim A, McGrath J, Feron F 2003 Vitamin D3 and brain development. Neuroscience 118(3): 641–653. Mahon BD, Bemiss C, Cantorna MT 2001 Altered cytokine profile in patients with multiple sclerosis following vitamin D supplementation. FASEB J 837:4. Embry AF, Snowdon LR, Vieth R 2000 Vitamin D and seasonal fluctuations of gadolinium-enhancing magnetic resonance imaging lesions in multiple sclerosis. Ann Neurol 48(2):271–272.

REINHOLD VIETH

147. Chan TY 2000 Vitamin D deficiency and susceptibility to tuberculosis. Calcif Tissue Int 66(6):476–478. 148. Douglas AS, Ali S, Bakhshi SS 1998 Does vitamin D deficiency account for ethnic differences in tuberculosis seasonality in the UK? Ethn Health 3(4):247–253. 149. Lansdowne AT, Provost SC 1998 Vitamin D3 enhances mood in healthy subjects during winter. Psychopharmacology (Berl) 135(4):319–323. 150. Gloth FM, III, Alam W, Hollis B 1999 Vitamin D vs broad spectrum phototherapy in the treatment of seasonal affective disorder. J Nutr Health Aging 3(1):5–7. 151. Harris S, Dawson-Hughes B 1993 Seasonal mood changes in 250 normal women. Psychiatry Res 49(1):77–87. 152. Al Allaf AW, Mole PA, Paterson CR, Pullar T 2003 Bone health in patients with fibromyalgia. Rheumatology (Oxford) 42:1–5. 153. Garland CF, Garland FC, Gorham ED 1999 Calcium and vitamin D. Their potential roles in colon and breast cancer prevention. Ann N Y Acad Sci 889:107–119. 154. Grant WB 2001 An estimate of excess cancer mortality in the US due to inadequate exposure to solar RV-B radiation (abstract). Photodermatol Photoimmunol Photomed 17:142. 155. Schwartz GG, Wang MH, Zang M, Singh RK, Siegal GP 1997 1α,25-Dihydroxyvitamin D (calcitriol) inhibits the invasiveness of human prostate cancer cells. Cancer Epidemiol Biomarkers Prev 6(9):727–732. 156. Schwartz GG, Whitlatch LW, Chen TC, Lokeshwar BL, Holick MF 1998 Human prostate cells synthesize 1,25-dihydroxyvitamin D3 from 25-hydroxyvitamin D3 [In Process Citation]. Cancer Epidemiol Biomarkers Prev 7(5):391–395. 157. Hsu JY, Feldman D, McNeal JE, Peehl DM 2001 Reduced 1alpha-hydroxylase activity in human prostate cancer cells correlates with decreased susceptibility to 25-hydroxyvitamin D3–induced growth inhibition. Cancer Res 61(7): 2852–2856. 158. Grau MV, Baron JA, Sandler RS et al 2003 Vitamin D, calcium supplementation, and colorectal adenomas: results of a randomized trial. J Natl Cancer Inst 95(23):1765–1771. 159. Vieth R, Dogan M, Cole DEC et al 2003 Vitamin D3 at 90 or 700 mcg weekly for 1 year: responses of 25(OH)D, PTH, urine and plasma calcium. European Calcified Tissue Society, Rome, May, 2003. Calcified Tissue Int 72(4): 377–378 160. Harris S 2002 Can vitamin D supplementation in infancy prevent type 1 diabetes? Nutr Rev 60(4):118–121. 161. Heaney RP, Dowell MS, Hale CA, Bendich A 2003 Calcium absorption varies within the reference range for serum 25-hydroxyvitamin D. J Am Coll Nutr 22:142–6. 162. Marx SJ, Jones G, Weinstein RS, Chrousos GP, Renquist DM 1989 Differences in mineral metabolism among nonhuman primates receiving diets with only vitamin D3 or only vitamin D2. J Clin Endocrinol Metab 69(6):1282–1290. 163. Jones G, Byrnes B, Palma F, Segev D, Mazur Y 1980 Displacement potency of vitamin D2 analogs in competitive protein-binding assays for 25-hydroxyvitamin D3, 24,25dihydroxyvitamin D3, and 1,25-dihydroxyvitamin D3. J Clin Endocrinol Metab 50(4):773–775. 164. Mawer EB, Jones G, Davies M et al 1998 Unique 24-hydroxylated metabolites represent a significant pathway of metabolism of vitamin D2 in humans: 24-hydroxyvitamin D2 and 1,24-dihydroxyvitamin D2 detectable in human serum. J Clin Endocrinol Metab 83(6):2156–2166. 165. Holmberg I, Berlin T, Ewerth S, Bjorkhem I 1986 25-Hydroxylase activity in subcellular fractions from

CHAPTER 61 The Pharmacology of Vitamin D, Including Fortification Strategies

human liver. Evidence for different rates of mitochondrial hydroxylation of vitamin D2 and D3. Scand J Clin Lab Invest 46(8):785–790. 166. Guo YD, Strugnell S, Back DW, Jones G 1993 Transfected human liver cytochrome P-450 hydroxylates vitamin D analogs at different side-chain positions. Proceedings of the

1015

National Academy of Sciences of the United States of America 90:8668–8672. 167. Harris SS, Dawson-Hughes B, Perrone GA 1999 Plasma 25-hydroxyvitamin D responses of younger and older men to three weeks of supplementation with 1800 IU/day of vitamin D. J Am Coll Nutr 18(5):470–474.

CHAPTER 62

How to Define Normal Values for Serum Concentrations of 25-Hydroxyvitamin D? An Overview PAUL LIPS

Department of Endocrinology, VU University Medical Center, Amsterdam, The Netherlands

I. Introduction II. How to Define Normal Values III. Variables Influencing Normal Values of Serum 25(OH)D

IV. Classification of Vitamin D Replete and Deficient States V. Dietary Vitamin D Intake and Recommended Daily Allowances References

I. INTRODUCTION

as in many developing countries [14], or on the high side when a large proportion of the population takes supplements, when food is fortified with vitamin D, or when sunshine is abundant and sunshine exposure is common [15,16]. Traditionally, reference values for serum 25(OH)D have been higher in the U.S. and Australia than in most European countries. Alternatively, reference values will be low in China or Russia [17]. This is caused by differences in sunshine exposure, food fortification with vitamin D, and the use of the vitamin supplements. Another way of defining normal values is by assessing biological or clinical outcomes of vitamin D status. These outcomes include the serum concentration of 1,25(OH)2D, the serum concentration of parathyroid hormone (PTH), bone mineral density (BMD), and the occurrence of bone disease, e.g. osteoporotic fractures or rickets and osteomalacia [3]. These outcomes can be assessed cross-sectionally, as in case-control studies, or longitudinally in prospective epidemiological studies or intervention studies. These methods are summarized in Table I.

Vitamin D deficiency may cause rickets and osteomalacia in the long term [1]. Mild or moderate vitamin D deficiency causes secondary hyperparathyroidism, bone loss, and osteoporosis and has been associated with fractures [2,3]. It has also been related to muscle weakness, colon cancer, and auto-immune diseases, such as diabetes mellitus and multiple sclerosis [4,5]. Vitamin D status is assessed by serum 25(OH)D, which is not the active metabolite [6]. Normal “reference” values have traditionally been derived from population data. However, these depend on sunshine exposure, latitude, and other factors, such as skin pigmentation [7]. The assessment is subject to interlaboratory variation, which may be considerable [8]. The dietary calcium intake influences the consequences of vitamin D deficiency, which become visible earlier when calcium intake is low [9]. During the last decade, it has become general practice to define normal values of serum 25(OH)D with respect to biological endpoints, for example, parathyroid function or bone mineral density [10,11].

II. HOW TO DEFINE NORMAL VALUES

A. Substrate-dependent Synthesis of 1,25(OH)2D

Serum 25(OH)D has been used to assess vitamin D status for more than 20 years, as it is the main circulating metabolite. Reference values have been obtained from population studies, or from presumably healthy subjects, e.g. blood donors [10,12,13]. The problem is that such values may be on the low side when the greater part of the population is vitamin D–deficient,

In the case of vitamin D deficiency, the synthesis of the active metabolite 1,25(OH)2D becomes substratedependent, i.e. dependent on the serum 25(OH)D concentration. This was observed in a Belgian study in nursing home residents. The low serum 25(OH)D in winter caused relatively low serum 1,25(OH)2D in winter and spring (Fig. 1). When the amount of substrate

VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

1020 TABLE I Biological Static and Dynamic Methods to Define the Threshold Between Deficient and Normal Values of Serum 25(OH)D

PAUL LIPS

A

µg/l

25-Hydroxycholecalciferol

p < 0.001

Static methods Positive correlation between serum 1,25(OH)2D and 25(OH)D below threshold serum 25(OH)D. Negative correlation between serum PTH and 25(OH)D below threshold serum 25(OH)D. Positive correlation between bone mineral density (BMD) and serum 25(OH)D below threshold. Association of fractures with low serum 25(OH)D in epidemiological studies. Dynamic methods Substrate dependent increase of serum 1,25(OH)2D following vitamin D supplementation when serum 25(OH)D is below threshold. Decrease of serum PTH after vitamin D supplementation when baseline serum 25(OH)D is below threshold. Increase of BMD after vitamin D supplementation when serum 25(OH)D is below threshold. Decrease of fracture incidence after vitamin D supplementation when baseline serum 25(OH)D is below threshold.

nmol/l 75

30 25 20

50

p < 0.05 15 10

25

5

B

pmol/l 175

ng/l 70

1,25-Dihydroxycholecalciferol NS

125

50

75

30

p < 0.01 10

25

[serum 25(OH)D] increased in spring and summer to 30–40 nmol/l serum, 1,25(OH)2D increased and approached normal values [18]. This substratedependent synthesis of 1,25(OH)2D is also apparent from the positive correlation between serum 25(OH)D and serum 1,25(OH)2D, which was observed in patients with hip fracture and geriatric patients with low serum levels of 25(OH)D. A vitamin D supplementation study (vitamin D3 400 or 800 IU/day) in residents of a home for the elderly and a nursing home showed that serum 1,25(OH)2D increased significantly when baseline serum 25(OH)D was lower than 30 nmol/l, while there was no increase with higher baseline serum 25(OH)D [20]. On the other hand, a study in nursing home residents in the U.S. did not show an increase of serum 1,25(OH)2D, probably because baseline serum 25(OH)D was 45 nmol/l [21]. These studies suggest that serum 25(OH)D should be above 30–40 nmol/l in order to achieve a normal serum 1,25(OH)2D.

B. Secondary Hyperparathyroidism When serum 25(OH)D falls below a certain level, serum 1,25(OH)2D will decrease as its synthesis becomes substrate-dependent. Calcium absorption from the gut will decrease, leading to an increase of PTH secretion. The increase of serum PTH will stimulate the synthesis of 1,25(OH)2D in order to keep calcium

0

2

4

6

8

10

12 Month

FIGURE 1 Seasonal low serum 25(OH)D causes low serum 1,25(OH)2D. When serum 25(OH)D increases in summer, serum 1,25(OH)2D approaches normal values. Reproduced from Bouillon R et al. 1987 Am J Clin Nutr 45:755–763. With permission of the American Society for Clinical Nutrition.

absorption and serum calcium within normal limits [3]. In the end, serum 1,25(OH)2D is maintained within normal limits at the expense of an increase of serum PTH [22]. The increase of serum PTH usually is within normal limits, so the secondary hyperparathyroidism only can be observed at a group level and not in most individuals. As the maintenance of serum 1,25(OH)2D and serum calcium is the outcome of a homeostatic control system, the negative relationship is visible between serum PTH and serum 25(OH)D and not serum 1,25(OH)2D [3]. This was clearly observed in a study of patients with hip fracture and elderly control subjects. While serum 25(OH)D is maximal in summer and reaches its lowest point in late winter or early spring, serum PTH shows an inverse seasonal pattern with maximal levels in winter or early spring and minimal levels in summer or autumn. However, serum 1,25(OH)2D stayed at constant levels throughout the year [22]. A negative correlation has been observed in observational or epidemiological studies in adults, especially elderly people (Fig. 2). The Amsterdam

CHAPTER 62 How to Define Normal Values for Serum Concentrations of 25-Hydroxyvitamin D? An Overview

A

B

PTH(1–84), pmol/l (log-scale) 15 16

12

iPTH (pg/ml)

130

8

1021

p < 0.01

110 90

4

70 55 50 36 30

2

1

10 0 11.6 0

10

20

30

40

50

60

70

80

0

11–

40

60 78

100 120 140 160

180 200

25(OH) D (nmol/l)

90

25(OH) vitamin D (nmol/l)

C

110 100

Parathyroid hormone (pg/ml)

90 80 70 60 50 40 30 20

0–5

6–1

15

20

16–

25

21–

30

26–

>30

25-Hydroxyvitamin D (ng/ml)

FIGURE 2 Negative relationship between serum PTH and serum 25(OH)D with different threshold levels where serum PTH starts to increase. (A) Amsterdam Vitamin D Study, reproduced from Ooms ME 1994 PhD Thesis, Vrije Universiteit, Amsterdam. (B) SUVIMAX study, reproduced with permission from Chapuy MC et al. 1997 Osteoporosis Int 7:439–443. (C) A study in hospital in-patients, reproduced from Thomas MK et al. 1998 N Engl J Med 338:777–783. With permission © Massachusetts Medical Society.

Vitamin D Study, a randomized placebo-controlled study of the effect of vitamin D supplementation on the incidence of hip fractures, showed a negative correlation between serum PTH and serum 25(OH)D when serum 25(OH)D was lower than 25 nmol/l [23] (Fig. 2A). When serum 25(OH)D was higher than

25 nmol/l the correlation was no longer significant. However, the SUVIMAX study, a study in French postmenopausal women, showed an increase of serum PTH when serum 25(OH)D was lower than 78 nmol/l [10] (Fig. 2B). A study in adult patients admitted to a general ward in a U.S. hospital showed an increase of

1022 serum PTH when serum 25(OH) was lower than 75 nmol/l [24] (Fig. 2C). These differences may be partially caused by interlaboratory differences in the assays for 25(OH)D [8,25]. Other important determinants are population characteristics such as age, sex, genetic difference, and dietary calcium intake. Another point is whether a small increase in serum PTH within the normal range should be considered pathological or just a physiological compensatory phenomenon.

PAUL LIPS

BMD left femoral neck, g/cm2 (log-scale) 0.9 0.8 0.7 0.6

0.5

C. Relationship with Bone Mineral Density Vitamin D deficiency is associated with a lower intestinal absorption of calcium. The availability of calcium for mineralization of newly formed bone matrix, the osteoid, may be too low. This causes insufficient secondary mineralization and bone with a low mineral content [26]. When vitamin D deficiency is severe, primary mineralization is also hampered, leading to accumulation of osteoid tissue and rickets or osteomalacia. The secondary hyperparathyroidism causes high bone turnover. Even in mild or moderate vitamin D deficiency, high turnover is associated with relatively young osteons in which secondary mineralization is not yet completed. The accumulation of osteoid and young incompletely mineralized bone is associated with low bone mineral density (BMD) when measured by densitometric techniques such as DXA [26]. In some cross-sectional studies, a relationship was observed between BMD and serum 25(OH)D. In 330 elderly women participating in the Amsterdam Vitamin D Study, a positive relationship was observed between baseline serum 25(OH)D and BMD of the femoral neck [11]. This relationship was significant when serum 25(OH)D was lower than 30 nmol/l (Fig. 3). The regression indicated that BMD could be 5 or 10% lower when serum 25(OH)D was 20 or 10 nmol/l, respectively. A positive correlation between serum 25(OH)D and BMD of the hip was also observed in studies from the U.K. and from New Zealand [27,28]. Baseline data of the MORE study, a placebo-controlled study in more than 6000 postmenopausal women, showed a relationship between serum 25(OH)D and BMD of the trochanter [16]. Subjects with a serum 25(OH)D lower than 25 nmol/l had a BMD 4% lower than those with serum 25(OH)D higher than 25 nmol/l. There was no difference between the groups with regard to BMD of the femoral neck or lumbar spine. Data on the effect of vitamin D treatment on BMD are scarce. In the Amsterdam Vitamin D Study, treatment with vitamin D3 400 IU/day increased the BMD of the femoral neck by 2.2% after 2 years in comparison with the placebo group [29]. In the French Decalyos Study,

0.4

0

10

20

30 40 50 60 25 (OH) vitamin D (nmol/l)

70

80

FIGURE 3 Relationship between BMD of the femoral neck and serum 25(OH)D. Significant correlation (p < 0.001) when serum 25(OH)D < 30 nmol/l. When serum 25(OH)D > 30 nmol/l the correlation was not significant. Reproduced from Ooms ME et al. 1995 J Bone Miner Res 10:1177–1184. With permission of the American Society for Bone and Mineral Research.

the BMD increase in the femoral neck was 6% in the group treated with vitamin D3 800 IU/day and calcium 1200 mg/day in comparison with the control group that received double placebo [30]. However, the treatment effect in the former group may be due to calcium as well as vitamin D. The increase in BMD was not linked to a low baseline serum 25(OH)D in these studies.

D. Relationship with Fractures The question of whether vitamin D deficiency is a risk factor for fractures has been assessed in large prospective epidemiological studies, such as the Study on Osteoporotic Fractures (SOF). In this study of 9704 elderly women, vitamin D deficiency, defined as serum 25(OH)D lower than 47 nmol/l, was not associated with an increased risk for hip or vertebral fracture [31]. However, a low serum 1,25(OH)2D (≤57 pmol/l) increased the risk for hip fracture (RR 2.1 adjusted for age and weight). A high serum PTH was not a risk factor in this study. The Longitudinal Aging Study Amsterdam also assessed the effect of a low serum 25(OH)D on fracture risk [32]. A low serum 25(OH)D (<30 nmol/l) was associated with an increased fracture risk (RR 1.6 adjusted for age and sex). Fracture risk according to fracture type was not studied because of limited follow-up. A Norwegian study has assessed the effect of vitamin D intake on hip fracture risk. A low intake (<100 IU/day) was associated with an increased hip fracture risk, but serum 25(OH)D was not measured in

1023

CHAPTER 62 How to Define Normal Values for Serum Concentrations of 25-Hydroxyvitamin D? An Overview

this study [33]. The incidence of hip fractures is much higher in Northern Europe than in Western or Southern Europe [34]. This suggests a possible involvement of vitamin D deficiency as a risk factor. However, the Euronut Seneca Study and the MORE Study showed that serum 25(OH)D was higher in Northern European countries than in Southern Europe [13,16]. This may be due to sun-seeking behavior in northern climates and sun-avoiding behavior in sunny countries. The consumption of fatty fish and the use of vitamin D supplements may also explain this gradient of higher serum 25(OH)D with more northern latitudes.

25(OH)D (nmol/l) 100 r=0.69 80

60

40

r=0.72

r=0.84

20

0

III. VARIABLES INFLUENCING NORMAL VALUES OF SERUM 25(OH)D A. Comparability of Assays for Serum 25(OH)D The observed differences in threshold values for serum 25(OH)D may be explained in part by differences in assays for serum 25(OH)D (see Chapter 58). Interlaboratory comparison studies have demonstrated a great variability in assay results [8,25]. Most assays are able to discriminate between low and average serum 25(OH)D values. However, it is difficult to compare studies from different countries and to establish internationally validated threshold levels for discrimination of vitamin D–deficient states. Most assays use an extraction step followed by competitive protein binding assay (CPB) or radioimmunoassay (RIA). The gold standard may be purification by high performance liquid chromatography (HPLC) followed by CPB or RIA [8]. A recent interlaboratory comparison showed large differences in serum 25(OH)D between Lyon using CPB and Amsterdam using HPLC followed by CPB [25]. The results on the same serum samples were 85% higher in Lyon than in Amsterdam (Fig. 4, Table II). This cross-calibration was used to correct serum values in order to compare the results of two large prospective intervention studies [30,35]. Following correction, it could be concluded that the elderly in Lyon were more vitamin D–deficient than those in Amsterdam and that the values of the vitamin D treatment were very similar in both studies. An interlaboratory comparison of four laboratories using a CPB and one laboratory using a RIA for serum 25(OH)D showed that the highest laboratory produced 38% higher serum 25(OH)D values than the lowest laboratory [25]. This is an important observation when considering a threshold for vitamin D insufficiency of 37.5 nmol/l (15 ng/ml) or 50 nmol/l (20 ng/ml), which may be similar or very different according to the assay used.

CPB

RIA

HPLC

Lyon

Lyon

Amsterdam

FIGURE 4 Interlaboratory comparison between assays for serum 25(OH)D from Lyon and Amsterdam. Reproduced with permission from Lips P et al. 1999 Osteoporosis Int 9:394–397.

B. Influence of Calcium Intake A low calcium intake is associated with a higher serum PTH than a high calcium intake. Serum PTH decreases within one hour after oral calcium load [36]. When the mean calcium intake was increased from 800 mg to 2400 mg per day in postmenopausal women, serum PTH decreased about 30% during 24 hours [37]. A low calcium intake may also influence vitamin D metabolism because PTH stimulates the renal hydroxylation of 25(OH)D into 1,25(OH)2D.

TABLE II Mean Serum 25(OH)D in Lyon (Decalyos Study) and Amsterdam (Amsterdam Vitamin D Study) Before and After Supplementation with Placebo or Vitamin D1

Placebo groups Baseline 1 year Vitamin D groups Baseline 1 year

Lyon before correction

Lyon after correction

32.5 25.0

17.9 13.7

26.0 nmol/l 23.0 nmol/l

21.6 56.7

27.0 nmol/l 62.0 nmol/l

40 105

Amsterdam

1The values for Lyon were corrected to those of Amsterdam using a correction factor of 0.54 derived from the study in ref. 25 (Fig. 4). Other data are derived from ref 29 and 30.

1024 Rats on a low calcium intake had a higher serum PTH and serum 1,25(OH)2D than rats with a high calcium intake [38]. This was associated with an increased metabolic clearance rate of 25(OH)D. This was confirmed by a clinical study of patients with primary hyperparathyroidism or secondary hyperparathyroidism following gastrectomy [9]. A high serum 1,25(OH)2D was associated with a low half life of 25(OH)D. Thus, a low calcium intake may aggravate vitamin D deficiency while a high calcium intake may have a vitamin D sparing effect. Calcium intake may influence the degree of secondary hyperparathyroidism associated with vitamin D deficiency and thus influence the threshold serum 25(OH)D where serum PTH starts to increase. This may partially explain why this threshold is lower in countries where calcium intake is high such as The Netherlands, than in countries such as France where calcium intake is low.

C. Other Variables Influencing Serum PTH Serum PTH is also influenced by renal function, the use of diuretics, and estrogen deficiency. The glomerular filtration rate slowly decreases with age from about 125 ml/min at age 20 to 60 ml/min at age 80. This is accompanied by a gradual increase of serum PTH, which is positively correlated with serum creatinine [39,40]. The increase of serum PTH may be caused by slight phosphate retention and lower synthesis of 1,25(OH)2D. The loop diuretic furosemide increases calcium excretion, decreases ionized calcium by inducing alkalosis and thereby increases serum PTH [41,42]. The thiazide diuretics increase calcium reabsorption, but paradoxically may also increase serum PTH [43]. The effect of estrogen on serum PTH is complex. The age-related increase in serum PTH did not occur in postmenopausal women on estrogen replacement therapy [44]. Estrogen also interacts with the secondary hyperparathyroidism following vitamin D deficiency. In the Amsterdam Vitamin D Study, the serum concentration of sex hormone binding globulin (SHBG) interacted with the negative relationship between serum PTH and serum 25(OH)D. Serum SHBG correlates negatively with the free estrogen concentration. Mean serum PTH was high in vitamin D–deficient elderly with high serum SHBG, and it was normal with a low serum SHBG [11]. This suggests that estrogen protects against secondary hyperparathyroidism caused by vitamin D deficiency. Treatment with vitamin D had a greater effect on BMD of the femoral neck when serum SHBG was high than when serum SHBG was low [29]. This supports the hypothesis that a higher free estrogen level protects

PAUL LIPS

against the consequences of vitamin D deficiency and secondary hyperparathyroidism.

IV. CLASSIFICATION OF VITAMIN D REPLETE AND DEFICIENT STATES It may be clear from the above, that vitamin D status is the complex result of sunshine exposure, latitude, sunscreen use, clothing, skin pigmentation, vitamin D intake with fatty fish, dairy products, fortified foods, and vitamin D supplement use [3,7]. In addition, the consequences of vitamin D deficiency depend on calcium intake and on other variables that influence parathyroid function, such as the decrease of renal function with age, low estrogen status, and the use of loop diuretics such as furosemide [3]. In addition, the reported serum 25(OH)D levels for sufficiency or deficiency depend on interlaboratory variation, which may explain unexpected findings [25]. As discussed above, the serum PTH concentration starts to rise when serum 25(OH)D falls below 25 nmol/l, 50 nmol/l or higher up to 78 nmol/l in various surveys [10,11,16]. The consequences of vitamin D deficiency may be arranged in three stages (Table III). Early, mild vitamin D deficiency or insufficiency causes secondary hyperparathyroidism, high turnover, and bone loss. Moderate vitamin D deficiency is associated with secondary hyperparathyroidism and slower secondary mineralization, leading to bone with a lower degree of mineralization. The third stage, severe vitamin D deficiency causes a disturbance of primary mineralization, leading to the accumulation of osteoid tissue, the hallmark of osteomalacia in adults, and accumulation of nonmineralized hypertrophic cartilage at the epiphysial zones characterizing rickets in children (see also Chapter 63).

A. Mild Vitamin D Deficiency Worldwide During the last decade, it has been suggested that the required normal range of serum 25(OH)D level should be raised because it became clear that the point where serum PTH starts to rise is not as low as previously assumed [3]. In the same period, investigators started to realize that mild vitamin D deficiency or insufficiency or inadequacy was more common than anticipated. In this chapter, mild vitamin D deficiency is defined as a serum 25(OH)D between 25 and 50 nmol/l [3]. While the consequences on an individual basis may be mild, a somewhat greater bone loss and a slightly increased fracture risk, the consequences on a population scale may be more important because

CHAPTER 62 How to Define Normal Values for Serum Concentrations of 25-Hydroxyvitamin D? An Overview

Table III Vitamin D status

Serum 25(OH)D (nmol/l)

Vitamin D replete Mild deficiency (insufficiency) Moderate deficiency Severe deficiency

Classification of Vitamin D Deficient States

Serum 25(OH)D ng/ml

Serum PTH increase

>50 25–50

>20 10–20

0% 5–15%

Normal Normal/high turnover

— Increased fracture risk

12.5–25 <12.5

5–10 <5

15–30% >30 %

High turnover Incipient or overt osteomalacia

Increased fracture risk Bone pain, fractures, Looser zones

of the large number of people who are vitamin D insufficient. In addition, other diseases that might be due to vitamin D insufficiency or disturbances in vitamin D metabolism become more important [3]. Several investigators have attempted to define the problem on a worldwide scale. McKenna and colleagues identified 117 studies on vitamin D status in young adults and/or elderly [15]. These studies mainly came from North America, Scandinavia, and Europe. It appeared from these studies that mild vitamin D deficiency was more common in Western and Southern Europe than in Scandinavia or North America. While vitamin D deficiency was very common in the elderly in winter and spring in Europe and North America, vitamin D deficiency was not rare in young adults in Europe during winter. A similar trend was observed in another review with higher serum 25(OH)D levels in North America and Australia than in Europe and Asia. Serum 25(OH)D was lower in institutionalized elderly than in independent living young and older

A

Bone histology

Consequences

adults (Fig. 5). A limitation of these reviews is the probably large interlaboratory variation in the assays for serum 25(OH)D. The Euronut Seneca study in Europe and the MORE study, which was done globally, used a central laboratory facility [13,16]. Serum 25(OH)D varied in the Seneca Study between 22 nmol/l in Greece and 46 nmol/l in Scandinavia. There was a very significant positive relationship between serum 25(OH)D and latitude with the highest levels in Northern and the lowest levels in Southern Europe. Similar observations were made in the MORE study, a double blind study on the effect of raloxifene in women with postmenopausal osteoporosis [16]. In this study, a similar positive relationship was observed between serum 25(OH)D and northern latitude in Europe with levels ranging from 50–60 nmol/l in Southern Europe to 70–90 nmol/l in Northern Europe. There are several explanations for this south-north gradient, the inverse of what should be expected. An explanation might be that people in northern countries like direct sunshine and sunbathing,

B 100

100

90

90

Serum 25(OH)D (nmol/l)

Serum 25(OH)D (nmol/l)

1025

80 70 60 50 40 30

80 70 60 50 40 30

20

20

10

10

0

0 NW Europe

FIGURE 5

Middle & S. Europe

Middle East

Asia

Australia N.Zealand

USA, Canada

Adults, Independent Outpatients, Inpatients, Hip fracture postmenop. elderly home for geriatric p., patients women elderly nursing home

Mean values of serum 25(OH)D from 43 studies according to geographical region (A) or to subject/patient/residence category (B). Reproduced from Lips P 2001 Endocr Rev 22:477–501. With permission of Endocrine Society.

1026 whereas people living in sunny countries tend to avoid the sun. The light skin in Northern Europe favors vitamin D synthesis, while the synthesis is less in a more pigmented skin in Mediterranean countries. In addition, consumption of fatty fish and the use of vitamin supplements may be more widespread in Northern Europe. In the Seneca Study, vitamin D deficiency (defined as serum 25(OH)D < 30 nmol/l) was observed in 47% of the participants. In the MORE Study, a serum 25(OH)D < 50 nmol/l was found in 39.3% of the participants in Southern Europe and in 12.6% of those in Northern Europe. A sunny climate is no guarantee for an adequate vitamin D status as follows from recent studies in adults from Lebanon [45], Ethiopia [14], and Italy [46].

B. Risk Groups Practicing physicians should be aware of risk factors for vitamin D insufficiency or deficiency. Risk factors relating to vitamin D production include low sunshine exposure, sunscreen use, highly pigmented skin, and clothes covering most parts of the body [7] (see Chapter 47). Risk factors relating to vitamin D metabolism are low calcium intake and medication such as antiepileptics, that increase vitamin D catabolism [47] (see Chapter 74). Using these risk factors, risk groups can be defined. Older people often are immobile and do not go outside. Persons with skin conditions, e.g. skin cancer, should not stay outside in direct sunshine. People with a dark skin need much more sunshine exposure to synthesize a similar quantity of vitamin D than people with a light skin (see Chapter 3). Cultural and religious customs may determine clothing habits and may restrict sunshine exposure. Very low serum 25(OH)D levels have been reported from Saudi Arabia, Lebanon, and Ethiopia [14,45,48]. Immigrants from North Africa, the Middle East, and India often have very low serum 25(OH)D levels [49,50]. Calcium intake is low in people with intolerance for dairy products due to lactase deficiency or following gastrectomy. Patients who are on chronic anti-epileptic medication may also carry a high risk for vitamin D deficiency [47]. These groups need special attention from general practitioners and public health care.

V. DIETARY VITAMIN D INTAKE AND RECOMMENDED DAILY ALLOWANCES The circulating 25(OH)D originates for the greater part from cutaneous synthesis following sunshine

PAUL LIPS

exposure and for a small part from dietary intake. Dietary intake becomes more important when sunshine exposure is low, as is common in elderly people and non-western immigrants [19,51]. However, most diets only contain small amounts of vitamin D3 in dairy products and eggs. Only fatty fish such as herring, mackerel, and halibut contain considerable amounts of vitamin D [19]. An important dietary source may be vitamin D–fortified products. Milk fortified with 400 IU of vitamin D3 per quart or liter is common in the U.S. In many European countries, however, only margarine is fortified with vitamin D3 3 IU/g. Typical dietary intakes of vitamin D are more than 200 IU/day in the U.S. and about 100 IU/day in many European countries [19,52]. A dietary vitamin D intake of 400 IU/day may be attained in the U.S. when sufficient fortified dairy products are used. In most Europeans, high dietary intakes of vitamin D can only be obtained with regular consumption of fatty fish. Currently, dietary recommendations are given as adequate intakes. These intakes have been defined in the U.S. to be 200 IU/day in adults until age 50 years, 400 IU/day in adults from 51 to 70 years, and 600 IU/day in persons older than 70 years [53]. The European Community has recommended a vitamin D3 intake of 0 to 400 IU/day for adults from 18 to 44 years, depending on sunshine exposure, being high in active people and low in the housebound [54]. The recommended intake for older persons (65 years and older) has been defined as 400 IU/day. It is difficult to assure these intakes with a normal diet unless a considerable amount of fatty fish or vitamin D3 fortified products is consumed. Therefore, most risk groups have to rely on vitamin D supplements.

A. Consequences for Public Health The above mentioned risk groups, older persons, immigrants, and persons with dark skin, are of considerable size and comprise a large part of the population for which vitamin D supplementation may be considered. A daily supplement of 400–600 IU is very effective but impractical. A larger supplement once per week, once per month, or once per three months may be equally effective and easier to distribute [55,56]. Dietary advice may be effective when fatty fish is recommended three times per week, but this is not feasible for many people. Fortification of dairy products with vitamin D3 is more attractive [57] as these products are widely used and these products usually contain a lot of calcium. Milk is an excellent carrier for vitamin D fortification, and it also is an important source of protein.

CHAPTER 62 How to Define Normal Values for Serum Concentrations of 25-Hydroxyvitamin D? An Overview

References 1. Frame B, Parfitt AM 1978 Osteomalacia: current concepts. Ann Intern Med 89:966–982. 2. Chalmers J, Barclay A, Davison AM, Macleod DAD, Williams DA 1969 Quantitative measurements of osteoid in health and disease. Clin Orthop 63:196–209. 3. Lips P 2001 Vitamin D deficiency and secondary hyperparathyroidism in the elderly: consequences for bone loss and fractures and therapeutic implications. Endocrine Rev 22:477–501. 4. Walters MR 1992 Newly identified actions of the vitamin D endocrine system. Endocr Rev 13:719–764. 5. Bouillon R, Okamura WH, Norman AW 1995 Structure-function relationships in the vitamin D endocrine system. Endocr Rev 16:200–257. 6. Reichel H, Koeffler HP, Norman AW 1989 The role of the vitamin D endocrine system in health and disease. N Engl J Med 20:980–991. 7. Holick MF 1995 Environmental factors that influence the cutaneous production of vitamin D. Am J Clin Nutr 1(Suppl 3): 638S–645S. 8. Jongen MJM, Ginkel v FC, Vijgh vd WJF, Kuipers S, Netelenbos JC, Lips P 1984 An international comparison of vitamin D metabolite measurements. Clin Chem 30:399–403. 9. Davies M, Heys SE, Selby PL, Berry JL, Mawer EB 1997 Increased catabolism of 25-hydroxyvitamin D in patients with partial gastrectomy and elevated 1,25-dihydroxyvitamin D levels. Implications for metabolic bone disease. J Clin Endocrinol Metab 82:209–212. 10. Chapuy MC, Preziosi P, Maamer M, Arnaud S, Galan P, Hercberg S, Meunier PJ 1997 Prevalence of vitamin D insufficiency in an adult normal population. Osteoporos Int 7:439–443. 11. Ooms ME, Lips P, Roos JC, van der Vijgh WJF, Popp-Snijders C, Bezemer PD, Bouter LM 1995 Vitamin D status and sex hormone binding globulin: determinants of bone turnover and bone mineral density in elderly women. J Bone Miner Res 10:1177–1184. 12. Netelenbos JC, Jongen MJM, van der Vijgh WJF, Lips P, van Ginkel FC 1985 Vitamin D status in urinary calcium stone formation. Arch Intern Med 145:681–685. 13. Wielen vd RPJ, Lowik MRH, Berg vd H, Groot de LCPGM, Haller J, Moreiras O, Staveren v WA 1995 Serum vitamin D concentrations among elderly people in Europe. Lancet 346:207–210. 14. Feleke Y, Abdulkadir J, Mshana R, Mekbib TA, Brunvand L, Berg JP, Falch JA 1994 Low levels of serum calcidiol in an African population compared to a North European population. Eur J Endocrinol 141:358–360. 15. McKenna M 1992 Differences in vitamin D status between countries in young adults and the elderly. Am J Med 93:69–77. 16. Lips P, Duong T, Oleksik A, Black D, Cummings S, Cox D, Nickelsen T 2001 A global study of vitamin D status and parathyroid function in postmenopausal women with osteoporosis: Baseline data from the Multiple Outcomes of Raloxifene Evaluation Clinical Trial. J Clin Endocrinol Metab 86:1212–1221. 17. Zhao XH 1992 Nutritional situation of Beijing residents. Southeast Asian J Trop Med Pub Health 23(Suppl 3):65–68. 18. Bouillon RA, Auwerx JH, Lissens WD, Pelemans WK 1987 Vitamin D status in the elderly: seasonal substrate deficiency causes 1,25-dihydroxycholecalciferol deficiency. Am J Clin Nutr 45:755–763.

1027

19. Lips P, van Ginkel FC, Jongen MJM, Rubertus A, van der Vijgh WJF and Netelenbos JC 1987 Determinants of vitamin D status in patients with hip fracture and elderly control subjects. Am J Clin Nutr 46:1005–1010. 20. Lips P, Wiersinga A, van Ginkel FC, Jongen MJM, Netelenbos JC, Hackeng WHL, Delmas PD and van der Vijgh WJF 1988 The effect of vitamin D supplementation on vitamin D status and parathyroid function in elderly subjects. J Clin Endocrinol Metab 67:644–650. 21. Himmelstein S, Clemens TL, Rubin A, Lindsay R 1990 Vitamin D supplementation in elderly nursing home residents increases 25-OHD but not 1,25(OH)2D. Am J Clin Nutr 52:701–706. 22. Lips P, Hackeng WHL, Jongen MJM, van Ginkel FC, Netelenbos JC 1983 Seasonal variation in serum concentrations of parathyroid hormone in elderly people. J Clin Endocrinol Metab 57:204–206. 23. Ooms ME 1994 Osteoporosis in elderly women: vitamin D deficiency and other risk factors. PhD Thesis, Vrije Universiteit, Amsterdam. 24. Thomas MK, Lloyd-Jones DM, Thadhani RI, Shaw AC, Deraska DJ, Kitch BT, Vamvakas EC, Dick IM, Prince RL, Finkelstein JS 1998 Hypovitaminosis D in medical inpatients. N Engl J Med 338:777–783. 25. Lips P, Chapuy MC, Dawson-Hughes B, Pols HAP, Holick MF 1999 An international comparison of serum 25-hydroxyvitamin D measurements. Osteoporos Int 9:394–397. 26. Parfitt AM 1980 Morphologic basis of bone mineral measurements: transient and steady state effects of treatment in osteoporosis. Miner Electrolyte Metab 4:273–287. 27. Khaw KT, Sneyd MJ, Compston J 1992 Bone density, parathyroid hormone and 25-hydroxyvitamin D concentrations in middle-aged women. BMJ 305:273–277. 28. McAuley KA, Jones S, Lewis-Barned NJ, Manning P, Goulding A 1997 Low vitamin D status is common among elderly Dunedin women. N Z Med J 110:275–277. 29. Ooms ME, Roos JC, Bezemer PD, van der Vijgh WJF, Bouter LM, Lips P 1995 Prevention of bone loss by vitamin D supplementation in elderly women: a randomized double-blind trial. J Clin Endocrinol Metab 80:1052–1058. 30. Chapuy MC, Arlot ME, Duboeuf F, Brun J, Crouzet B, Arnaud S, Delmas PD, Meunier PJ 1992 Vitamin D3 and calcium to prevent hip fractures in elderly women. N Engl J Med 327: 1637–1642. 31. Cummings SR, Browner WS, Bauer D, Stone K, Ensrud K, Jamal S, Ettinger B 1998 Endogenous hormones and the risk of hip and vertebral fractures among older women. N Engl J Med 339:733–738. 32. Lips P, Pluijm SMF, Popp-Snijders C, Smit JH 2001 Vitamin D status, sex hormone binding globulin, IGF-1 and markers of bone turnover as determinants of bone mass and fractures in the Longitudinal Aging Study Amsterdam. J Bone Miner Res 16(Suppl 1):S166. 33. Meyer HE, Henriksen C, Falch JA, Pedersen JI, Tverdal A 1995 Risk factors for hip fracture in a high incidence area: A case–control study from Oslo, Norway. Osteoporos Int 5:239–246. 34. Johnell O, Gullberg B, Allender E, Kanis JA 1992 The apparent incidence of hip fracture in Europe: a study of national register sources. MEDOS Study Group. Osteoporosis Int 2:298–302. 35. Lips P, Graafmans WC, Ooms ME, Bezemer PD, Bouter LM 1996 Vitamin D supplementation and fracture incidence in elderly persons. A randomized, placebo-controlled clinical trial. Ann Intern Med 124:400–406.

1028 36. Lips P, Netelenbos JC, van Doorn L, Hackeng WH, Lips CJM 1991 Stimulation and suppression of intact parathyroid hormone (PTH1-84) in normal subjects and hyperparathyroid patients. Clin Endocrinol 35:35–40. 37. McKane R, Khosla S, Egan KS, Robins SP, Burritt MF, Riggs BL 1996 Role of calcium intake in modulating agerelated increases in parathyroid function and bone resorption. J Clin Endocrinol Metab 81:1699–1703. 38. Clements MR, Johnson L, Fraser DR A new mechanism for induced vitamin D deficiency in calcium deprivation. Nature 325:62–65. 39. Wiske PS, Epstein S, Bell NH, Queener SF, Edmondson J, Johnston C 1997 Increases in immunoreactive parathyroid hormone with age. N Engl J Med 300:1419–1421. 40. Marcus R, Madvig P, Young G 1984 Age-related changes in parathyroid hormone and parathyroid hormone action in normal humans. J Clin Endocrinol Metab 58:223–230. 41. Gabow PA, Hanson TJ, Popovtzer MM, Schrier RW 1977 Furosemide-induced reduction in ionized calcium in hypoparathyroid patients. Ann Intern Med 86:579–581. 42. Stein MS, Scherer SC, Walton SL, Gilbert RE, Ebeling PR, Flicker L, Wark JD 1996 Risk factors for secondary hyperparathyroidism in a nursing home population. Clin Endocrinol (Oxf) 44:375–383. 43. Rejnmark L, Vestergaard P, Heickendorff L, Andreasen F, Mosekilde L 2001 Effects of thiazide- and loop-diuretics, alone or in combination, on calcitropic hormones and biochemical bone markers: a randomized controlled study. J Intern Med 250:144–53. 44. Khosla S, Atkinson EJ, Melton SLJ, Riggs BL 1997 Effects of age and estrogen status on serum parathyroid hormone levels and biochemical markers of bone turnover in women: a population–based study. J Clin Endocrinol Metab 82:1522–1527. 45. Gannage-Yared MH, Chemali R, Yaacoub N, Halaby G 2000 Hypovitaminosis D in a sunny country: relation to lifestyle and biochemical markers. J Bone Miner Res 15:1856–1862. 46. Bettica P, Bevilacqua M, Vago T, Norbiato G 1999 High prevalence of hypovitaminosis D among free-living postmenopausal

PAUL LIPS

47. 48. 49. 50. 51.

52. 53. 54. 55.

56.

57.

women referred to an osteoporosis outpatient clinic in Northern Italy for initial screening. Osteoporos Int 9:226–229. Hahn TJ, Birge SJ, Scharp CR, Avioli LV 1972 Phenobarbitalinduced alterations in vitamin D metabolism. J Clin Invest 51:741–748. Sedrani SH, Elidrissy AWTH, El Arabi KM 1983 Sunlight and vitamin D status in normal Saudi subjects. Am J Clin Nutr 38:129–132. Preece MA, Ford JA, McIntosh WB, Dunnigan MG, Tomlinson S, O’Riordan JLH 1973 Vitamin D deficiency among Asian immigrants to Britain. Lancet 1: 907–910. Grootjans-Geerts I 2001 Hypovitaminosis D: een versluierde diagnose. Ned Tijdschr Geneeskd 145:2057–2060. Glerup H, Mikkelsen K, Poulsen L, Hass E, Overbeek S, Andersen H 2000 Hypovitaminosis D myopathy without biochemical signs of osteomalacic bone involvement. Calcif Tissue Int 66:419–424. Omdahl JL, Garry PJ, Hunsaker LA, Hunt WC, Goodwin JS 1982 Nutritional status in a healthy elderly population: vitamin D. Am J Clin Nutr 36:1225–1233. Holick MF 1998 Vitamin D requirements for humans of all ages: new increased requirements for women and men 50 years and older. Osteoporos Int 8(Suppl 2):S24–S29. European Commission. Report on osteoporosis in the European Community: action for prevention. European Commission DG V Directorate for Public Health, Luxembourg 1998. Weisman Y, Schen RJ, Eisenberg Z, Amarilio N, Graff E, Edelstein-Singer M, Goldray D, Harell A 1986 Single oral high-dose vitamin D3 prophylaxis in the elderly. J Am Geriatr Soc 34:515–518. Khaw KT, Scragg R, Murphy S 1994 Single-dose cholecalciferol suppresses the winter increase in parathyroid hormone concentrations in healthy older men and women. Am J Clin Nutr 59:1040–1044. Keane EM, Rochfort A, Cox J, McGovern D, Coakley D, Walsh JB 1992 Vitamin-D–fortified liquid milk—a highly effective method of vitamin D administration for house-bound and institutionalized elderly. Gerontology 38:280–284.

CHAPTER 63

Vitamin D and the Pathogenesis of Rickets and Osteomalacia A. MICHAEL PARFITT

Division of Endocrinology and Center for Osteoporosis and Metabolic Bone Disease, University of Arkansas for Medical Sciences, Little Rock, Arkansas

I. Introduction II. Morphologic and Biochemical Aspects of Mineralization III. Processes Leading to Accumulation of Unmineralized Tissue IV. Evolution of Vitamin D Related Bone Disease

V. Aspects of Vitamin D Metabolism Relevant to Rickets and Osteomalacia VI. Vitamin D and the Pathogenesis of Impaired Mineralization References

I. INTRODUCTION

II. MORPHOLOGIC AND BIOCHEMICAL ASPECTS OF MINERALIZATION

The vitamin D field has become so diverse and so complex that many forget how it all started—it was the study of rickets that led to the discovery of vitamin D. Despite the multiplicity of effects on nontraditional target tissues, the principal function of vitamin D and its derivatives, in humans and most other mammals, is still to facilitate the processes and mechanisms that are necessary to prevent rickets and its adult counterpart osteomalacia. These diseases are both consequences of defective mineralization, but within different tissues; the mineralization of growth plate cartilage and of bone have many features in common, but there are also important differences. Vitamin D deficiency may be broadly classified as extrinsic, due to some combination of nutritional deficiency and inadequate exposure to sunlight, and intrinsic, due to some combination of impaired absorption and accelerated catabolism of vitamin D metabolites. The relative importance of these mechanisms may be different in different parts of the world, and different in children and adults. In both rickets and osteomalacia, there may be hypophosphatemia, hypocalcemia, and secondary hyperparathyroidism, but their temporal relationships to one another and to the events in bone may be different. A major unsolved problem is whether changes in the composition of the blood are sufficient to account for the effects of vitamin D deficiency on cartilage and bone, or whether one or more of the metabolites of vitamin D has actions on skeletal cells that promote mineral deposition. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

The process whereby ions in solution are transformed into a solid phase falls within the domain of physical chemistry, but skeletal mineralization is also a biological process that is controlled with regard to its location, timing, rate, and relationship to cells and to extracellular connective tissue matrices. A comprehensive theory of mineralization must be consistent with the laws of chemistry and physics but must also account for its morphologic features. Disregard of these features led early students of bone, such as Franklin McLean, to believe that the matrix became mineralized as soon as it was formed, and that the presence of any unmineralized matrix was pathological [1]. This belief matched the notion that biological mineralization was nothing more than the precipitation, within the appropriate matrix, of crystals from a supersaturated solution, and that only the composition of the solution determined whether mineralization occurred [2]. The invariable existence of a significant amount of unmineralized bone matrix, or osteoid, in mammalian bone was first demonstrated by Lacroix and students in dogs and cats [3], and soon after confirmed in human subjects by Frost and Villanueva [4]. Microscopic examination of undecalcified sections of bone obtained after double tetracycline labeling (Chapter 59) allows the process of mineralization to be observed in situ, with preservation of its spatial relationships to the bone and the cells and introduction of the dimension of time [5,6]. Tetracycline labeling has Copyright © 2005, Elsevier, Inc. All rights reserved.

1030 been applied much less frequently to the study of mineralization in cartilage than in bone, because the rate of advance of the mineralization front and the consequent distance between the two labels is driven mainly by the rate of longitudinal growth, which reflects the rate of chondrocyte proliferation [7]. However, two important features apply to mineralization in both tissues: spatial localization and a measurable time delay between the synthesis of matrix and the deposition of mineral within it. In cartilage, the label is an aggregate of discrete patches, each corresponding to a single longitudinal septum, that form a band about 50 µm in width, extending all the way across the growth plate [8], and the average time delay is about 24 hr. In bone, the label is continuous, more sharply demarcated, and only 2 to 3 µm in width, and the average time delay is about 2 weeks. In cartilage, the delay may reflect changes in gene expression in the chondrocytes [9], but in bone the delay reflects the need for extracellular changes in the matrix, collectively referred to as maturation, to occur before mineralization can begin. These changes include completion of cross-linking between collagen fibrils [10,11] and the development of precise orientation, conformation, and aggregation of a variety of noncollagenous proteins and proteoglycans [12,13]. In culture, maturation does not occur in the absence of viable cells [14]. Bone mineral consists of Ca2+, PO43−, OH−, and CO2− ions, arranged in space in accordance with the crystal lattice structure of hydroxyapatite [15–18] (Chapter 27). The composition is indeterminate because some of the constituent ions can be replaced by other ions of similar radius, and at some lattice points calcium ions can be missing altogether [19]. The mineralizing potential of extracellular fluid (ECF) depends on the free ionic activity, or effective concentration (denoted a), of Ca2+, HPO42−, and H+ ions. The activity coefficients relating effective to actual ionic concentrations depend mainly on pH, temperature, and total ionic strength, which are all fairly constant in ECF, but the coefficients are lower and more variable for divalent than for univalent ions [15]. For both Ca and P, ionic concentrations differ from the total concentrations normally measured because of protein binding and ion complexing, which are also affected by pH. The often calculated total plasma calcium × phosphate product, although meaningless in terms of physical chemistry, bears a rough empirical relationship to the true thermodynamic activity product [aCa2+] [aHPO42−]. Mineralization is a phase transformation, not a chemical reaction [17], but it is more likely that the complex structural order of hydroxyapatite is attained in steps rather than all at once [16]. At sites of mineralization, the successive addition of Ca2+ and HPO42− ions

A. MICHAEL PARFITT

present in ECF, and simultaneous removal of protons, generates a series of compounds beginning with secondary calcium phosphate or brushite (CaHPO42H2O), the first solid phase to be formed, and ending with hydroxyapatite [Ca10(PO4)6(OH)2] [6]. The relevant activity products in ECF are in the region of metastability with respect to bone mineral, being undersaturated with respect to brushite but supersaturated with respect to hydroxyapatite [16]. Mechanisms to accomplish initial mineral deposition include concentration gradients between mineralizing and nonmineralizing sites maintained by cells and by the ion binding and releasing properties of a variety of macromolecules synthesized by cells [13,15,16,18], sequestration and subsequent release of calcium by mitochondria [20], and heterogeneous nucleation by outside agents or substances [17]. Alkaline phosphatase is essential for normal mineralization [21], but its function remains unknown. Mechanisms to restrain the growth of hydroxyapatite crystals include the precise spatial relationships between mineral and matrix [13], the presence at critical locations of chelators of calcium [13] and inhibitors of mineralization such as pyrophosphate [22], albumin [23], and decorin [24], and the cellular and biochemical characteristics of the quiescent bone surface, which is the site of reversible mineral exchange with systemic ECF [19]. Within this general framework, two types of mineralization can be recognized [25] (Chapter 27). In growth plate cartilage and woven bone, which are temporary structures destined soon to be removed, the matrix is loosely textured and the collagen fibrils are small, immature, and disordered. Mineral is deposited in the form of approximately spherical clusters of randomly oriented crystals of varying size. The clusters, termed calcospherulites, are spatially associated with matrix vesicles, which are small membrane-bound particles that are derived by an unknown mechanism from chondrocytes. The vesicles are abundant and appear to be the only structures available for nucleation [26]. Additional, more active roles in promoting mineralization that have been proposed [26–28] must be reconciled with their distribution, with highest density in the resting and hypertrophic zones and lowest density in the proliferative and calcifying zones [29]. In lamellar bone, which is invariably formed in apposition to an existing surface, the matrix is compact in texture, matrix vesicles are infrequent or absent, and the collagen fibrils are long and highly ordered. The mineral crystals are aligned with their long axis parallel to the collagen fibrils and are initially deposited within the hole zones by heterogeneous nucleation, but longer and wider crystals are subsequently formed on and between the fibrils [18,25]. The differences between

CHAPTER 63 Vitamin D and the Pathogenesis of Rickets and Osteomalacia

these two types of mineralization explain why rickets and osteomalacia can under some circumstances vary independently in their severity and response to treatment [30].

III. PROCESSES LEADING TO ACCUMULATION OF UNMINERALIZED TISSUE A. Growth Plate Cartilage and Osteoid in the Growing Skeleton The process of endochondral ossification is described in Chapters 27 and 33. The width of the epiphyseal growth plate depends on the rate of longitudinal growth, which is determined by the rate of new chondrocyte production [31] and the life span for completion of maturation prior to initial calcification. For example, in 5-week-old rats a width of about 600 µm corresponds to a growth rate of 330 µm/day and a life span of about 1.8 days [26], and in 10-week-old rats a width of 350 µm corresponds to a growth rate of 180 µm/day and a life span of about 1.9 days [32]. These and other data indicate that although longitudinal growth slows progressively with increasing age, growth plate life span remains approximately constant. In experimental rickets in rats, even though longitudinal growth is reduced threefold, growth plate width increases about fourfold in 6 weeks or by about 27 µm/day because of an increase in life span of at least twelvefold [32]; in severe rickets, in the absence of treatment, growth plate life span is limited only by the age of the animal. Evidently, the characteristic increase in growth plate width occurs despite a reduction in the rate of growth and is due entirely to a profound delay in mineralization, the pathogenesis of which is discussed later. There is also structural disorganization of the metaphysis, partly due to mechanical effects [33] and partly to a profoundly altered pattern of vascular invasion [34]. During the transformation of calcified cartilage to primary and then secondary spongiosa there is extensive deposition of osteoid. The kinetics of its production, life span, and mineralization have never been studied by tetracycline labeling, but accumulation of osteoid contributes to the microscopic characteristics of the rachitic metaphysis. A possible source of confusion must be addressed at this point. The term “rickets” is commonly applied to the totality of skeletal abnormalities associated with defective mineralization in the growing skeleton, but it is more accurate to restrict the term to changes in the growth plate and adjacent metaphysis. In the vertebral bodies and ilium, cancellous

1031

bone tissue occupies all the available space enclosed by the cortices and does not undergo removal, but is remodeled in a manner similar to the adult skeleton. During both intramembranous ossification [35] and long bone growth, osteoid is formed beneath the periosteum and mineralized bone is removed on the inner surface. When mineralization is defective, the accumulation of osteoid at sites other than the growing metaphysis should be referred to as osteomalacia, not as rickets. Thus, impaired mineralization leads to both rickets and osteomalacia in the growing skeleton but only to osteomalacia in the mature skeleton. The kinetics of osteoid production and mineralization in the growing skeleton have been studied most thoroughly in the rat tibia; several important observations were made [36]. First, although osteoid seams are generally thinner in rats than in larger animals, they are invariably present at sites of bone formation; because of the temporal separation between matrix apposition and mineralization, there is also spatial separation. Second, in 3-week-old rats, osteoid is found underneath the entire circumference of the periosteum; because its extent cannot change except as a result of growth, a significant increase in osteoid accumulation can occur only if the thickness of the seam increases. This contrasts with the mature skeleton, in which osteoid almost always increases in surface extent before it increases in thickness. Third, analogous to the regulation of growth plate width, osteoid thickness depends on the rate of matrix apposition, which corresponds to the rate of lateral (and hence longitudinal) growth [37], and on the delay before the onset of mineralization that is imposed by osteoid maturation, known as the mineralization lag time (Mlt), which corresponds to the growth plate life span. Like growth plate width, osteoid thickness declines with increasing age because of a decline in matrix apposition rate with a relatively constant lag time.

B. Life History of Individual Osteoid Seams in the Adult Skeleton The formation of each new bone structural unit (osteon or hemiosteon) begins at the cement surface, a thin layer of lowly mineralized collagen-poor but glycoprotein-rich connective tissue [38] that is laid down on the floor of the resorption cavity at the end of the reversal phase of each remodeling cycle, represented in two-dimensional histological sections by the cement line, which remains in the same location, separating new bone from old. The boundary between mineralized and unmineralized bone, referred to as the osteoidbone interface, is the location of the mineralization

1032

A. MICHAEL PARFITT

front, which normally moves away from the cement line during bone formation. Its rate of advance must be distinguished from the rate at which mineralization proceeds after it is initiated [39]. In an individual moiety of bone matrix, mineral accumulation as a function of time is a continuous process that is conveniently subdivided into two stages. There is an early rapid increase to about 75–80% of maximum within the first few days, referred to as primary mineralization. It involves multiplication in the number of crystals, occurs close to the osteoblast, and may be influenced by its function. A much slower increase to about 95% of maximum or more, over many months or even years, is referred to as secondary mineralization. It involves slow growth in the size of crystals, with displacement of water, occurs remote in both time and space from

the osteoblast, and is presumably governed entirely by physicochemical factors [39]. A team of osteoblasts assembles on the cement surface and begins to deposit a layer of bone matrix referred to as an osteoid seam, which in standard histological sections appears in cortical bone as a ring, and in cancellous bone as a crescent tapering at each end. Each seam has a measurable life span, during which characteristic changes occur in the morphological features and function of the osteoblasts, and in the thickness of the seam (Fig. 1). Matrix apposition is most rapid (2.0 to 3.0 µm/day) at the outset, and the seam reaches a maximum thickness of approximately 15 to 20 µm after about 10–15 days, just before mineralization begins. Mineral apposition is also most rapid at the outset (1.0 to 1.5 µm/day), and thereafter is always

FP

40 Mlt

AR

x-

M

30

O.Th W.Th AR

20

M

Distance from cement line (µm)

50

MB.Th

10

0 0

20

40

60

80

100

120

Time from onset of matrix synthesis (days)

FIGURE 1 Model of bone formation, with growth curves for matrix apposition above and mineral apposition below, showing distances from the cement line as functions of time at a single cross-sectional location of a representative basic metabolic unit (BMU). At any distance from the cement line, the horizontal distance between the lines is the instantaneous mineralization lag time (Mlt) at that distance. At any time, the vertical distance between the lines is the instantaneous osteoid thickness (O.Th) at that time. At any point the slopes of the lines (tangents) represent instantaneous apposition rates for matrix (Mx.AR) or mineral (MAR). For example, at t = 30 days, the instantaneous values are 20 µm for mineralized bone thickness (MB.Th), 16 µm for osteoid thickness, 0.5 µm/day for matrix osteoid apposition rate, and 0.8 µm/day for mineral apposition rate; the matrix deposited at that time will have a mineralization lag time of 26 days. Formation period (FP) is counted from the onset of matrix synthesis to the completion of mineralization, and in this example is 120 days, at which time completed wall thickness (W.Th) equals 50 µm. It is evident that the total area between the curves is given by FP × mean O.Th and by W.Th × mean Mlt, so that these expressions are equal. Furthermore, it follows that O.Th = Mlt × Mx.AR. Reprinted from Parfitt [5], in Chemistry and Biology of Mineralized Tissues, 1992, pp. 465–474, with kind permission from Elsevier Science.

1033

CHAPTER 63 Vitamin D and the Pathogenesis of Rickets and Osteomalacia

more rapid than matrix apposition so that osteoid seam thickness declines. Both matrix and mineral apposition progressively slow with time, as the osteoblasts become flatter and more extended in shape. About 80 days after the onset of mineralization when the bone surface has returned to its previous location about 50 µm from the cement line, matrix synthesis stops. The osteoid seam thickness has by now fallen to about 6 µm, and for a further 30 days mineral apposition continues at a progressively declining rate that is too slow for tetracycline fixation to occur [40]. Eventually, the osteoid seam disappears, because all the new matrix has become mineralized. The osteoblasts have now completed their histological transformation into lining cells, and construction of the new BSU at that cross-sectional location is finished. A key quantity in understanding the mechanisms of osteoid accumulation and the pathogenesis of osteomalacia is the mineralization lag time (Mlt); this was defined earlier for the rat, but its method of calculation and significance are somewhat different in the adult human skeleton. In the rat, periosteal bone formation is continuous, the entire bone-forming surface is labeled with tetracycline, there is no need to distinguish between instantaneous and mean values, and the best estimate of the matrix apposition rate is the mineral apposition rate (MAR) so that Mlt = O.Th/MAR (where

O.Th is osteoid thickness). In humans, bone formation is cyclical (Chapter 28), tetracycline fixation does not occur during terminal mineralization, and it is important to distinguish between instantaneous and mean values. Osteoid thickness at any distance from the cement line is the product of the instantaneous matrix apposition rate (Mx.AR) and the instantaneous Mlt (Fig. 1). Instantaneous values are provided only by complete remodeling sequence reconstruction [41], and in practice only mean values are obtained. The best estimate of the mean matrix apposition rate is the mineral apposition rate averaged over the entire osteoid surface, referred to as the adjusted apposition rate (Aj.AR), so that mean Mlt = mean O.Th/Aj.AR [39]. In the rat, Mlt is identical with the osteoid maturation time (Omt), but in humans this may be true only for the initial Mlt, which is usually about 10 days (Fig. 1). The cause of the subsequent increase in Mlt to about 30 days, which has no counterpart in the rat, is unknown. One possibility is that the time required for matrix maturation increases with the age of the osteoblast, and changes in matrix apposition rate and lag time together determine the progress of mineralization (Fig. 2A). In this case, lag time would be an independent variable that remained identical with maturation time, and the changes in mineral apposition rate would follow automatically. Alternatively, the rates of

B

Distance (µm)

A

0

40

80

120

0

40

80

120

Time (days)

FIGURE 2 Two models of the relationship between matrix apposition, mineral apposition, and mineralization lag time. (A) Matrix apposition rate and mineralization lag time are separately and independently regulated as functions of osteoblast age, and the rate of mineral apposition changes as an automatic consequence. (B) Rates of matrix and mineral apposition are separately and independently regulated as functions of time, and the lag time changes as an automatic consequence. In both cases, the genuine independent variables are depicted by solid lines and the automatically determined variables by dashed lines. Reprinted from Parfitt [5], in Chemistry and Biology of Mineralized Tissues, 1992, pp. 465–474, with kind permission from Elsevier Science.

1034

A. MICHAEL PARFITT

matrix and mineral apposition could be separately and independently regulated as functions of osteoblast age (Fig. 2B). For example, there could be a decline in the supply of mineral, since the net inward calcium flux characteristic of osteoblasts must at some point change to the outward calcium gradient without net flux characteristic of lining cells [19]. In this case, lag time would not be an independent variable but would progressively exceed maturation time. Fortunately, this uncertainty does not detract from the usefulness of Mlt in the understanding of histomorphometric data and of the mechanisms of osteoid accumulation [5,36]. To avoid the inconvenience of an infinite value (MAR = 0), it is sometimes useful to calculate the reciprocal of Mlt as the osteoid mineralization rate (OMR): OMR (%/d) = 100/Mlt (d)

(1)

C. Osteoid Indices and the Recognition of Impaired Mineralization Osteoid accumulation is assessed by three independent measurements (Chapter 59) that are related as follows [42] and defined below and in Fig. 1: OV/BV (%) = O.Th (mm) × OS/BS (%) × BS/BV (mm2/mm3)

(2)

A fall in trabecular thickness, as occurs to a modest extent in aging and in osteoporosis, will increase BS/BV and so increase OV/BV even if surface and width are unchanged; for the most accurate interpretation, OV/BV should be corrected to the expected trabecular thickness for age and sex. Each of the three indices is related differently to the underlying kinetic determinants. Osteoid thickness has already been discussed and is given by O.Th (µm) = Aj.AR (µm/day) × Mlt (days)

(3)

Osteoid surface per unit of bone surface (OS/BS) is determined entirely by the mean osteoid seam life span or formation period (FP) and by the average frequency with which new osteoid appears at any point on the bone surface, which in the steady state is the same as the frequency of remodeling activation (Ac.f): OS/BS (%) = FP (years) × Ac.f (year−1) × 100

(4)

Osteoid volume is determined entirely by the fractional rate of bone turnover, which is the same as the volume-based bone formation rate (BFR/BV), and by

the mean life span of an individual moiety of osteoid, which is the same as the mineralization lag time: OV/BV (%) = BFR/BV (%/year) × Mlt (years) (5) Because in the steady state bone turnover is determined entirely by the frequency of remodeling activation and the surface to volume ratio, and because FP is inversely proportional to Aj.AR [39], each of the three static indices of osteoid accumulation is determined by a different pair of the same three kinetic indices [39]. Notably, osteoid volume is independent of matrix (or mineral) apposition rate, which in the steady state affects surface and thickness equally in opposite directions. Although a reduction in mineral apposition rate is frequently taken to indicate defective mineralization, it is evident from Fig. 1 that matrix and mineral apposition are closely coupled and that the mean mineral apposition rate can never exceed the mean matrix apposition rate. Consequently, a reduction in the mean rate of matrix apposition inevitably leads to, and is much the most common cause of, a reduction in the mean rate of mineral apposition. Both in normal subjects and in patients with any metabolic bone disease except osteomalacia, there is a significant positive correlation between mean osteoid thickness and mean adjusted apposition rate, with broadly similar slopes (b) and intercepts (a) of the regression lines [42]. Although such a relationship is to be expected, it has an unanticipated consequence for the interpretation of the mineralization lag time, as we can write O.Th = b (Aj.AR) + a.

(6)

If this is combined with Eq. (3), we obtain Mlt = b + a/Aj.AR

(7)

Because of this relationship, which defines a rectangular hyperbola, when the matrix apposition rate falls, the mineralization lag time increases [6]. Another way of arriving at the same conclusion is to consider the effect of prolongation of FP, which is also an inevitable consequence of a reduction in matrix apposition rate. From Fig. 1 it is clear that FP × mean O.Th = W.Th × mean Mlt

(8)

Because O.Th has a minimum value [the intercept in Eq. (6)] and W.Th (the mean thickness of a completed BSU) is effectively constant in the short term, an increase in FP must be accompanied by an increase in Mlt. It follows from this reasoning that neither a reduction in Aj.AR nor an increase in Mlt indicate that

1035

CHAPTER 63 Vitamin D and the Pathogenesis of Rickets and Osteomalacia

(Aj.AR; Fig. 3B). When Aj.AR is above 0.1 µm/day, there is the usual positive relationship between these variables; below 0.1 µm/day further decrements in Aj.AR are associated not with a fall in osteoid thickness as in all other situations, but with a progressive increase, limited only by the normal thickness of new matrix or W.Th [44]. This reversal is the cardinal kinetic characteristic of defective mineralization; a similar hyperbolic relationship is found between osteoid thickness and the fraction of osteoid surface undergoing mineralization [45]. On the basis of these relationships, the author defines osteomalacia by a combination of mean mineralization lag time more than 100 days (OMR < 1/d) and mean osteoid thickness above the upper 95% confidence limit predicted by the regression of osteoid thickness on Aj.AR in normal and osteoporotic subjects (Fig. 3B), or, more simply, above an absolute value of 12.5 µm (corrected for section obliquity). To distinguish generalized osteomalacia from focal and atypical variant forms (6) needs an additional criterion of OV/BV greater than 10%. Diagnosis can be simplified by combining the different measurements into a single mineralization index (MI, 46), which in arbitrary units is 0 to 15 in normal subjects and greater than 30 in patients fulfilling the other criteria for osteomalacia. Patients with HVO who

mineralization is defective unless they are accompanied by an increase in O.Th.

IV. EVOLUTION OF VITAMIN D RELATED BONE DISEASE A. Histological Evolution and the Kinetic Definition of Osteomalacia In most patients, osteomalacia is preceded for many years by clinically silent secondary hyperparathyroidism that accelerates the irreversible age-related loss of cortical bone [43]. Exposition of this concept is aided by using the term “hypovitaminosis D osteopathy” (HVO) to encompass the totality of osseous complications of deficiency or altered metabolism of vitamin D [6]. There is usually no relationship between osteoid thickness and osteoid surface, but in HVO there is a hyperbolic relationship between these variables [6,42] (Fig. 3A). This indicates that osteoid surface increases first and that osteoid thickness increases only slightly until OS/BS exceeds 70%, after which further increases in osteoid volume are due mainly to increasing thickness. In the same patients, osteoid thickness shows a more complex relationship to adjusted apposition rate 60

A

B

M

lt

=

10

0d

50

40 O.Th 30 (µm) 20

10

0 0

20

40

60

OS/BS (%)

80

100

0

0.1

0.2

0.3

0.4

0.5

Aj.AR (µm/d)

FIGURE 3 Relationship of osteoid thickness (O.Th) to osteoid surface (OS/BS) (A) and adjusted apposition rate (Aj.AR) (B). In normal subjects and in patients with osteoporosis, there is no relationship between osteoid thickness and surface (A), but there is a significant positive relationship between osteoid thickness and adjusted apposition rate (B). During the development of osteomalacia there is a direct hyperbolic relationship between osteoid thickness and surface (A) and an inverse hyperbolic relationship between osteoid thickness and adjusted apposition rate (B); the usual ranges of values in established osteomalacia are shaded. The oblique line through the origin in (B) corresponds to a mineralization lag time of 100 days. Reprinted with permission from Parfitt AM. The physiologic and pathogenetic significance of bone histomorphometric data. In: Coe FL, Favus MJ (eds.), Disorders of Bone and Mineral Metabolism. 2nd Edition. 2002. Lippincott Williams and Wilkins, Philadelphia, pp. 469–485.

1036

A. MICHAEL PARFITT

do not meet these criteria have increased volume and surface but not thickness of osteoid, increased bone formation rate (BFR), the normal positive relationship between O.Th and Aj.AR, increased osteoclast indices [6], and low forearm bone density due to cortical thinning [47], resembling in every respect the histologic and densitometric features of primary hyperparathyroidism. This analysis identifies the earliest stage of HVO (HVOi), when osteoid accumulation is due mainly to increased frequency of remodeling activation and bone turnover, before the emergence of a significant mineralization defect, as being due to secondary hyperparathyroidism [47] (Chapter 30). This concept has been sharpened by the MI [46], which permits HVOi to be subdivided into HVOia in which MI is normal (0–15), and HVOib, in which MI is moderately increased [15–30], but BFR is still high. Patients with HVO who meet the criteria for defective mineralization are further subdivided into those who retain some tetracycline double labels (HVOii) and those with no double labels (HVOiii). Pertinent histologic data in the four subgroups of HVO are given in Table I. A different perspective on the fundamental nature of osteomalacia can be gained from the model of osteoid seam life span (Fig. 4). In every other condition, all matrix formed eventually mineralizes; the slopes of the curves representing matrix and mineral apposition, although initially divergent, eventually converge, and the loop formed by these curves ultimately closes. In contrast, in untreated osteomalacia some matrix remains permanently unmineralized, the slopes of matrix and mineral apposition remain divergent, and the loop never closes. The model also illuminates the difference between HVOii, in which mineralization ceases prematurely, so

that the earliest formed matrix at each forming site becomes mineralized but the later formed matrix does not, and HVOiii, in which mineralization never starts, so that none of the matrix formed becomes mineralized. As all patients with HVOiii at the time of biopsy have likely been through the stage of HVOii, they show a mixture of the two types of osteoid seam depicted in Fig. 4. Both thickness and volume of osteoid are significantly greater in HVOiii than in HVOii [Table I], but even in the most severe cases individual values for mean osteoid thickness fall within the reference range for mean wall thickness [44].

B. Biochemical Evolution of Rickets and Osteomalacia The cardinal metabolic consequence of vitamin D deficiency is reduced net intestinal absorption of calcium [6,45] (see also Chapter 24). Fecal calcium excretion is close to and can even exceed dietary intake, but urinary calcium is low and calcium balance rarely more negative than −100 mg/day [48]. There is an equimolar deficit in net absorption of inorganic phosphate, but the relative change is much smaller. According to the usual interpretation, calcium malabsorption leads in sequence to a fall in plasma calcium, secondary hyperparathyroidism, reduced renal tubular reabsorption of phosphate, hypophosphatemia, and reduction in calcium × phosphate product, which falls even further with the advent of more severe hypocalcemia. Eventually, deposition of mineral in osteoid is impaired because the supply of the relevant ions is reduced, and the alkaline phosphatase activity then

TABLE I Histology of Bone Mineralization Variable n OS/BS (%) O.Th (µm)a OV/BV (%) BFR/BSb Mlt (d)c OMR (%/d) MId

Normal

HVOia

HVOib

HVOii

HVOiii

143 18.0 (9.5) 9.0 (1.7) 1.25 (0.74) 13.8 (9.7) 36.8 (1.73) 3.06 (1.23) 8.0 (3.3)

18 37.4 (16.7) 8.4(2.4) 6.2(3.4) 34.2(26.3) 39.8(1.62) 2.89(1.41) 9.3(3.8)

8 57.5(18.8) 10.6(1.9) 10.8(3.8) 26.7(22.3) 113(2.10) 1.24(0.59) 20.0(3.0)

13 87.0(5.9) 20.6(8.1) 28.8(8.6) 12.4(8.4) 355(2.2) 0.213(0.169) 54.8(18.1)

29 90.3(8.4) 26.5(10.8) 40.8(17.0) 0 8 0 75.1(28.5)

Histologic indices of osteoid accumulation and their determinants in normal subjects and in patients with different degrees of hypovitaminosis D osteopathy (HVO), as defined in the text. Effects of sex, race, and age are small in relation to disease-related changes. Data expressed as mean (SD) but for Mlt expressed as geometric mean and multiplicative SD. aCorrected for section obliquity. bUnits are µm3/µm2/y. cCalculated after log transformation. dArbitrary units [46].

1037

CHAPTER 63 Vitamin D and the Pathogenesis of Rickets and Osteomalacia

Matrix Mlt

Mineral

Normal

Distance (µm) 60 40 O. Th 20

Normal W. Th

0

(Mild)

60 40

(Severe)

Osteomalacia

20 0 60 40 20 0 100

200

300 400 Time (days)

500

600

FIGURE 4 Kinetics of matrix and mineral apposition in osteomalacia. Evolution of bone formation at a single cross-sectional location. Each graph is constructed in a manner similar to Fig. 1, but with altered time scale. Mlt, Mineralization lag time; O.Th, osteoid thickness; W.Th, wall thickness. Note that in mild osteomalacia (HVOii) mineralization is delayed in onset, retarded in rate, and premature in termination, whereas in severe osteomalacia (HVOiii) no mineralization occurs at all. Reprinted from Parfitt AM 1990 Bone-forming cells in clinical conditions. In: Hall BK (ed.), Bone: A Treatise, Vol. 1, The Osteoblast and Osteocyte, 1990, p. 395. Copyright CRC Press, Boca Raton, FL.

rises. This traditional scheme requires considerable modification with regard to the differences related to age of onset, the order in which the changes occur, their pathophysiology, and their diagnostic significance. Many years ago, three stages were recognized in simple vitamin D deficiency in infants [49]. In Stage I there was hypocalcemia and normal plasma phosphate, in Stage II plasma calcium rose to normal, plasma phosphate fell below normal, and alkaline phosphatase increased modestly, and in Stage III plasma calcium and phosphate were both reduced, with no further change in alkaline phosphatase. However, in current practice there are many exceptions to this sequence [50] (Chapter 65). Early hypocalcemia is rarely observed in older children [50,51] but is quite common during adolescence [52], although very uncommon in adults [53] (Table I). The ability to release calcium from bone may be compromised during periods of rapid

growth, but in infants the abnormality is associated with a delayed increase in serum parathyroid hormone (PTH) [54], normal skeletal and renal tubular responsiveness to exogenous PTH [49], and spontaneous improvement; in adolescents and adults there is an appropriate increase in PTH levels but impaired renal tubular as well as skeletal responsiveness to PTH [52,55], and the abnormality persists in the absence of treatment. Further discussion of this acquired form of pseudohypoparathyroidism is beyond the scope of this chapter, except for two points pertinent to subsequent discussion. First, plasma calcium is determined by the homeostatic system at quiescent bone surfaces [19]. This system is independent of remodeling and is regulated jointly by PTH and one or more metabolites of vitamin D [15]. Second, deficiency of vitamin D causes hypocalcemia mainly by loss of its effects on bone.

1038

A. MICHAEL PARFITT

TABLE II

Biochemical Evolution of Hypovitaminosis D Osteopathy (HVO)a

Normalb (n = 23) Age (years) Plasma calcidiol (ng/ml) Plasma calcitriol (pg/ml) Plasma calciumc (md/dl) Plasma phosphate (mg/dl) Plasma Ca × P [(mg/dl)2] Alkaline phosphatase (IU) NcAMPd (nmol/dl GF)

60.3 ± 1.3 23.7 ± 3.0 40.8 ± 6.9 9.64 ± 0.08 3.47 ± 0.08 33.5 ± 0.7 82.8 ± 3.9 2.04 ± 0.26

HVOi (n = 26) 57.2 ±2.1 6.0 ± 0.5 46.0 ± 4.7(11) 9.12 ± 0.11+ 3.36 ± 0.11 30.7 ± 1.1* 132 ± 7.2‡ 4.01 + 0.34

HVOii (n = 11)

HVOiii (n = 28)

50.5 ± 6.3 6.8 ± 0.9(10) 39.1 ± 3.7(8) 7.95 ± 0.39+ 2.91 ± 0.23+ 23.3 ± 1.6‡ 201 ± 31.1 6.62 ± 0.79

58.1 ± 2.4 4.1 ± 0.6(18)* 21.7 ± 3.2(10)+ 8.02 ± 0.17 2.64 ± 0.13 21.2 ± 1.2 284 ± 24.2* 5.94 ± 0.61

a Stages are defined in the text. Number of analyses are shown in parentheses when less than number of subjects. Data are means ± SE. Significance levels are shown for differences in mean values from the column immediately to the left: *, p < 0.05; +, p < 0.01; ‡ p < 0.001. Data reprinted from Parfitt [6]. bVolunteers for bone biopsy. c Corrected for albumin. dN cAMP, Nephrogenous cyclic AMP, an in vivo bioassay for circulating PTH, which is given in units of nanomoles per deciliter of glomerular filtrate.

In most adults with HVOi, the mean plasma calcium is slightly reduced, but the individual values are almost always normal (Table II). PTH secretion is increased as shown both by radioimmunoassay [54,56] and by excretion of nephrogenous cyclic AMP (NcAMP) (Table II). Although mean tubular reabsorptive maximum for phosphorus divided by the glomerular filtration rate (TmPi/GFR) and plasma phosphate are both slightly reduced, individual values are usually normal. Twenty-four hour urinary calcium excretion and fasting urinary calcium/creatinine are often but not invariably reduced, and a moderate elevation of alkaline phosphatase is the most consistent abnormality. In extrinsic vitamin D depletion, the plasma calcidiol level at which abnormal mineral metabolism can first be detected in an individual is usually below 5 ng/ml [45,52], but in subjects with values between 5 and 10 ng/ml there is a slight but statistically significant depression of mean plasma calcium and phosphate and urinary calcium and elevation of PTH [56], and most patients with histologically verified HVOi have calcidiol values in this range (Table II). In intrinsic vitamin D depletion, the complete biochemical, histological, and bone densitometric syndrome of HVOi can occur at plasma calcidiol levels between 10 and 20 ng/ml [47,53], presumably because there is an independent mechanism for calcium malabsorption and consequent secondary hyperparathyroidism that is unrelated to vitamin D but accounts for the normal mean level of plasma calcitriol. The progression of HVO through Stages II and III is similar to the progression of infantile rickets through Stages II and III, but it occurs over a much longer time scale and differs somewhat in detail. In general all the biochemical abnormalities become more severe.

Plasma and urinary calcium, TmPi/GFR, and plasma phosphate levels become lower, and PTH, NcAMP, and alkaline phosphatase levels higher (Table II), but there are many individual exceptions. As in rickets, patients with severe hyperparathyroidism may suffer impaired tubular reabsorption of bicarbonate and amino acids as well as phosphate, resembling proximal renal tubular acidosis or the Fanconi syndrome [6,45], except for increased rather than decreased tubular reabsorption of calcium. Hypophosphatemia is adequately explained by increased PTH secretion without the need to postulate an additional effect of vitamin D metabolite deficiency. Indeed, for the same increase in NcAMP, TmPi/GFR is higher in secondary than in primary hyperparathyroidism because of the independent effect of plasma calcium to decrease phosphate reabsorption [53,57]. The mean calcidiol level is not significantly lower in HVOii than in HVOi, but it does fall further in HVOiii (Table II). In contrast, the calcitriol levels can be normal in Stage II and do not become consistently subnormal until Stage III. Others have also found both normal and subnormal calcitriol levels in osteomalacia, although not classifying their cases in the same manner [6]. In summary, in HVOi there are characteristically no symptoms until a fracture occurs, which is why the existence of this intermediary stage was unrecognized for so long. The only biochemical abnormality that would be revealed by routine screening is a raised plasma alkaline phosphatase. Both fasting and 24-hr urinary calcium excretion are usually reduced. Skeletal radiographs are either normal or show only nonspecific osteopenia, but bone densitometry reveals that age-related loss of bone is accelerated, especially appendicular

CHAPTER 63 Vitamin D and the Pathogenesis of Rickets and Osteomalacia

cortical bone but also axial trabecular bone, with a corresponding increase in fracture risk [47,53]. Plasma calcidiol is usually but not invariably low. There is both biochemical and histological evidence of secondary hyperparathyroidism and of increased bone turnover. Defective mineralization is either absent (HVOia) or no more severe than in primary hyperparathyroidism (HVOib). Paradoxically, the deficit in forearm bone density is greater in HVOi than in HVOii and HVOiii despite less severe hyperparathyroidism. This can only be explained by slower progression and consequently longer duration of accelerated cortical bone loss and increased fracture risk [6]. It is likely that some patients remain arrested indefinitely at this stage, a few progress to more severe hyperparathyroidism with radiographic osteitis fibrosa, but others eventually develop the complete clinical, biochemical, radiographic, and histological syndrome of osteomalacia. However, it may reasonably be assumed that all patients in Stages II or III at the time of diagnosis traveled earlier through Stage I.

V. ASPECTS OF VITAMIN D METABOLISM RELEVANT TO RICKETS AND OSTEOMALACIA Vitamin D metabolism can be affected at one of six levels [6]. Identification of the level is important in planning treatment, although the summation of independent factors at several levels may be needed to produce clinical effects, and some diseases affect more than one level. Each level is associated with a characteristic profile of vitamin D metabolite concentrations in blood, but these must be interpreted with caution because changes in vitamin D binding protein (DBP) can alter total concentrations without altering free concentrations [6] (Chapter 8). Body stores of vitamin D, located mainly in fat and to a lesser extent in muscle, are derived either from the photochemical production in skin of cholecalciferol or from dietary intake and intestinal absorption of either chole- or ergocalciferol. Although the former source is more physiological [58], with current lifestyles the latter is equally important. The distinction was made earlier between extrinsic vitamin D depletion, due to some combination of reduced skin synthesis and reduced intake, and intrinsic depletion, due to intestinal malabsorption of vitamin D, augmented by an additional mechanism for increased fecal loss of vitamin D [6]. Two infrequently emphasized features of normal vitamin D metabolism are relevant to the pathogenesis of rickets and osteomalacia: first, its wastefulness in normal circumstances [52] and second, its susceptibility to disruption by calcium deficiency [59,60].

1039

Metabolic pathways leading from calciferol to calcitriol have been extensively investigated but are preferentially followed only when body stores are greatly depleted. Normally, about 70% of the daily supply of both calciferol and calcidiol is converted to more polar metabolites of low or absent biological activity that undergo biliary and eventually fecal excretion [52], so that only about 10% of available calciferol is used for calcitriol production. The proportion following the alternate pathways can decrease to very low levels when necessary, but it increases to 90% for calciferol and 99% for calcidiol in vitamin D treated hypoparathyroidism [61]. Despite their quantitative importance in overall vitamin D economy, not much is known about either the metabolites formed or their mechanism of production, and even less about how distribution between different pathways is regulated (Chapter 2). Because of the usual wide margin of safety, malabsorption of dietary vitamin D is rarely of sufficient severity to be the sole mechanism responsible for vitamin D depletion. The first additional mechanism to be proposed was interruption of a conservative enterohepatic circulation of calcidiol [6]. This proposal accounted for depletion of vitamin D of dermal as well as dietary origin, but the magnitude of this pathway in human subjects, if it occurs at all, is much too small to fulfill its postulated role [52,62] (Chapter 75). It now seems much more likely that the additional mechanism is accelerated catabolism of calcidiol in the liver initiated by calcium deprivation [63], whether due to dietary deficiency or intestinal malabsorption. It is because of calcium deprivation that plasma calcitriol levels are increased in the earlier stages of intrinsic HVOi [64], as they are also in patients treated with anticonvulsants [52], although this is not evident from Table II, which reports only data from patients who had bone biopsy. Accelerated calcidiol catabolism is mediated by secondary hyperparathyroidism, either as a direct effect of increased circulating PTH [59,63] or (more likely) as an indirect effect due to increased serum levels of calcitriol [59,65,66]. This mechanism explains the occurrence of vitamin D deficiency in geographic regions with high sun exposure but low calcium intake [59], and it contributes to vitamin D deficiency in a variety of gastrointestinal disorders [60,66] (Chapter 75). The hepatic 25-hydroxylation of calciferol to calcidiol provides the principal transport form of vitamin D and an additional component of body stores, located mainly in muscle [6]. This process is impaired in cirrhosis of the liver, but rarely to a level that causes osteomalacia (unless there is also malabsorption, as in biliary cirrhosis) because the liver has such a large reserve capacity [52]. Significant calcidiol deficiency that is not due to depletion of its precursor is most

1040 commonly the result of increased catabolism to biologically inactive metabolites from drug-induced enzyme induction (Chapter 74) or from stimulation of existing enzymes by calcitriol or PTH [65] (Chapter 75). Loss of calcidiol bound to protein (both DBP and albumin) occurs in the nephrotic syndrome and leads to secondary hyperparathyroidism and osteoid accumulation in the absence of impaired renal function [6]. A similar mechanism operates during peritoneal dialysis, and urinary loss of calcidiol is also increased in patients with biliary cirrhosis (Chapter 75). Calcitriol deficiency with normal body stores of vitamin D is most commonly the result of chronic renal failure (Chapter 76) but can also be due to a genetic defect in renal 1α-hydroxylation, referred to as pseudo vitamin D deficiency rickets (PVDR), hereditary hypocalcemia or vitamin D dependency type I (Chapter 71). Plasma calcitriol levels are reduced by magnesium depletion [6], but osteomalacia as a consequence has not been demonstrated. Calcitriol synthesis is impaired by deficiency of PTH (Chapter 64), but it is doubtful whether this causes osteomalacia, possibly because bone turnover is so low. However, there is one adequately documented case of osteomalacia due to pseudohypoparathyroidism with secondary hyperparathyroidism [67]. Very low plasma calcitriol levels are found during prolonged total parenteral nutrition, but they have not been clearly related to the presence or type of metabolic bone disease [6]. Finally, calcitriol may be ineffective because of one of several kinds of defect in its receptors (VDR), referred to as hereditary vitamin D-resistant rickets (HVDRR) or vitamin D dependency rickets type II (Chapter 72).

VI. VITAMIN D AND THE PATHOGENESIS OF IMPAIRED MINERALIZATION The role of vitamin D in sustaining normal mineralization has given rise to two related controversies. First, is calcitriol the only metabolite of physiological importance, other than as a precursor [68], or must some other metabolite also be considered [69]? Second, is the action of vitamin D on bone mediated solely by changes in the calcium and phosphate concentrations in ECF [68], or does it influence mineralization more directly [48,52]? In both cases the contestants have often failed to recognize the difference between an essential function that confers an absolute requirement and a contributory function that confers only a relative requirement. Concerning the first controversy, calcitriol alone can correct the clinical, biochemical, radiographic, and histological effects of vitamin D deficiency both in humans [70,71] and in the rat [72,73].

A. MICHAEL PARFITT

Claims to the contrary [69,74] reflect the inability of intermittent oral administration to sustain an adequate blood level [72]. No other metabolite is essential; the impairment of mineralization during intramembranous ossification that was previously attributed to deficiency of 24 hydroxy-calcidiol [75] was shown to be the result of very high circulating levels of calcitriol [70]. This counterintuitive effect has been known for some time [77], but its mechanism is still unknown. Nevertheless, it is possible that 24-hydroxy calcidiol contributes to processes such as bone embryonic development and maturation of growth plate chondrocytes [75] (Chapter 33). Concerning the second controversy, there is evidence that the effects of vitamin D deficiency in humans can be corrected by giving enough calcium and phosphate intravenously [78], and that defects in calcitriol receptor binding can also be bypassed by providing sufficient mineral substrate [79–81]. The strength of the evidence is examined later in Section VI,B. In completely vitamin D–deficient rats, maintenance of normal plasma calcium and phosphate levels maintained normal growth, as well as normal growth plate and bone mineralization [82–83]. In clinical studies, the determination that radiographic and histological abnormalities are due to defective mineralization rather than to secondary hyperparathyroidism is not as straightforward as is often assumed, but if the conclusions of these studies are taken at face value [78–81], then vitamin D is clearly not essential for mineralization. Nevertheless, one or more metabolites of vitamin D could have direct actions on those cells that contribute to the process in normal circumstances [6,52,74].

A. The Effects of Vitamin D Are Not Mediated Solely by Circulating Calcitriol In patients with histologically verified osteomalacia or with radiographically unambiguous rickets, plasma calcitriol concentrations can be within the appropriate reference ranges [45,84–87] (Table II). The levels are indeed inappropriately low for the degrees of PTH hypersecretion and hypophosphatemia [86], as is indicated by the very high levels attained during the early stages of treatment with vitamin D [71,87,88], but the lack of target cell responses to a concentration of calcitriol that is normally adequate requires explanation. In adults with osteomalacia, both biochemical and histological indices of vitamin D depletion appear to correlate better with either the sum of calcidiol (in ng/ml) and calcitriol (in pg/ml) concentrations, or with calcidiol alone, than with calcitriol alone [45,53]. This suggests that calcidiol, or some other metabolite for which calcidiol is a precursor, such as 24-hydroxycalcidiol,

1041

CHAPTER 63 Vitamin D and the Pathogenesis of Rickets and Osteomalacia

might function as an agonist for calcitriol, or might have its own effects mediated by a different receptor. In infants with untreated rickets, the plasma Ca × P product correlated with the plasma concentration of calcitriol and not calcidiol, although a higher than normal calcitriol level was needed to maintain a normal product [88]. In this study the plasma level of 24-hydroxycalcidiol was often undetectable; this compound might function in a permissive manner, so that a fall in its concentration below a critical level would increase the need for calcitriol, but without dose-related effects above the critical level. Higher than normal calcitriol levels could be needed to correct hypocalcemia and to restore normal mineralization when the calcitriol responsive cells are separated from the mineralized bone by a much wider than normal layer of uncalcified osteoid through which the mineral ions must travel [5,6,52] (Fig. 5). However, no similar reason is evident for the failure of intestinal mucosal cells to accomplish normal calcium transport, at least in patients who do not have an independent cause for impaired calcium absorption, such as intestinal

Mineralization: H+

Ca+

disease (Chapter 75) or anticonvulsant administration [89]. Theories that ascribe all manifestations of rickets and osteomalacia to deficiency of circulating calcitriol alone may be able to account for its persistence, but they have much greater difficulty accounting for its initiation. At the onset of HVO in patients with extrinsic vitamin D deficiency, what causes calcium malabsorption and a small but significant fall in plasma calcium when plasma calcitriol is maintained at a normal (or even increased) level by secondary hyperparathyroidism [90,91]? A similar argument applies to the increased vitamin D requirement of primary hyperparathyroidism, due to accelerated calcidiol catabolism in the presence of increased plasma calcitriol levels [92]. During the evolution of HVO, there is an early fall in plasma concentrations of both calcidiol [47,53,84,91,93] (Table I) and 24-hydroxycalcidiol [74,85,88]. Calcium absorption and retention in bone are increased in humans by pharmacological doses of 24-hydroxycalcidiol [69], but it is not known whether such effects occur at physiological blood levels. Calcidiol binds to intestinal receptors for calcitriol, but

Systemic or local control? HPO42−

25HCC

1,25DHCC

Ca

P

Capillary

Osteoblast 1,25 DHCC Osteoid

Osteocyte M. front interface

Mineralized bone

FIGURE 5 Biochemical and morphological approaches to mineralization. At left are shown the directional movements of ions between a blood vessel above and the bone below without reference to intervening structures. These structures are depicted diagrammatically at right. 25HCC, 25-Hydroxycholecalciferol (calcidiol); 1,25DHCC, 1,25-dihydroxycholecalciferol (calcitriol); M., mineralization; interface, boundary between osteoid and mineralized bone. The osteoblasts and osteocytes can be influenced by circulating levels of calcium, phosphate, and calcitriol, and also by locally produced calcitriol, either autocrine or paracrine. Reprinted from Parfitt [5] 1992 in Chemistry and Biology of Mineralized Tissues, pp. 465–474, with kind permission from Elsevier Science.

1042 with approximately 500- to 1000-fold lower affinity (Chapter 11), although calcidiol is only 100 times less effective than calcitriol in promoting bone resorption in vitro [94]. Seemingly, these differences in activity could be offset by the much higher total plasma concentration of calcidiol, but there is only a tenfold difference in free concentrations, based on estimates of the association constants for binding to the same circulating protein [95]. Consequently, a fall in plasma calcidiol level below normal could not significantly modify total receptor occupancy in the target cells that respond to circulating calcitriol, although some more complex effect on receptor function remains possible [96]. A more promising approach to the clinical paradox is the possibility that one or more dihydroxylated metabolites are produced locally in target tissues, as is strongly suggested by studies with isolated bone (97,98) and intestinal cells [99,100], by the in vivo intestinal response to a pharmacological oral dose of calcidiol [101], and by the in vitro effects of calcidiol to increase bone resorption [102] and intestinal calcium transport [103] in isolated tissues. These studies were carried out when calcidiol was believed to be the active form of vitamin D, and were forgotten (or suppressed) when this belief was superseded, but the results can only be explained by the local conversion of calcidiol to calcitriol. If bone cell and intestinal cell 1a-hydroxylases were less influenced by PTH and phosphate than is the renal 1a-hydroxylase, as is generally the case for extrarenal calcitriol production [104, Chapter 5], local production of calcitriol would be more substrate dependent than circulating calcitriol and would be impaired sooner by a fall in plasma calcidiol concentration below normal. This mechanism would also account for the improvement in bone mineralization and intestinal calcium absorption brought about by calcidiol administration in patients with chronic renal failure [105], and for the much greater relative therapeutic potency of dihydrotachysterol in hypoparathyroidism than in osteomalacia [61]. Similar considerations would apply to local production of 24-hydroxycalcidiol [99]. It seems reasonable to speculate that circulating calcitriol is most important for the regulation of calcium homeostasis, locally produced calcitriol (and possibly 24-hydroxycalcidiol) are most important for the regulation of bone remodeling and mineralization, and both circulating and locally produced metabolites are important for the regulation of calcium absorption.

B. Evidence for Direct as Well as Indirect Effects of Vitamin D on Bone It has been known since the early 1920s that active infantile rickets is associated with low plasma phosphate

A. MICHAEL PARFITT

(less than 3.0 mg/dl) and low total plasma Ca × P product [less than 30 (mg/dl)2] and healing rickets with increases in these values [106,107]. Freshly harvested rachitic rat growth plate cartilage will mineralize in normal rat serum and in aqueous solutions with the same pH and Ca × P product [108,109]. Calcification occurs in the same region as the cartilage as in vivo, and fails to occur if the viability of the tissue is compromised. The relationship of rickets to the total plasma Ca × P product has been confirmed many times [6,45], and the relationship of the product to the thermodynamic activity product for various solid phases has been analyzed [109]. The same relationship holds in rickets complicating osteopetrosis, in which low plasma levels of calcium and/or phosphate are due to increased mineral sequestration in bone [110]. The relationship is disturbed in patients with renal failure [111], possibly due to the presence of mineralization inhibitors, such as magnesium [112], or the effect of metabolic acidosis on the activity coefficient of PO43− [109]. The data clearly established the importance of plasma composition, which is disturbed indirectly by vitamin D deficiency [109], but do not rule out additional direct local effects. Such effects of calcitriol have been demonstrated in matrix vesicles, including increases in the activity of alkaline phosphatase and metalloproteinase [113]. In older children and adults, the relationship between plasma composition and the state of mineralization is less consistent [6,48,74], being clearly evident in some series of patients [114–116] but not in others [117–119]. There are significant correlations between plasma phosphate and adjusted apposition rate, between plasma calcium and mean osteoid thickness [120] and between plasma Ca × P product and MI [46], but their magnitude is too low for useful prediction in individual patients. Also the high unexplained variance leaves plenty of room for other factors [46]. Calculation of an ion product more clearly related to the physical chemistry of bone mineral may remove some discrepancies [109], but many remain. A more serious flaw in this line of reasoning is that single measurements in the fasting state, as in normal clinical practice, do not adequately represent body fluid composition because of the substantial circadian variation [19,121]. Nevertheless, it seems unlikely that such variation could account for the absence of osteomalacia in some patients with a degree of persistent hypophosphatemia that in other patients would be regarded as a sufficient explanation for their osteomalacia [120,122]. Even in the rat, a species in which mineralization probably depends more closely on plasma composition than in humans, healing of rickets can be detected radiographically in response to vitamin D administration while the Ca × P product is still subnormal [73].

CHAPTER 63 Vitamin D and the Pathogenesis of Rickets and Osteomalacia

It was previously demonstrated that the early effects of vitamin D deficiency in adults are due entirely to secondary hyperparathyroidism in the presence of normal bone mineralization; the correction of such effects by calcium has no bearing on the issue in question. Neither does the cure by calcium administration of clinical and radiographic appearances that resemble rickets but are due to severe dietary calcium deficiency [123]. Radiographic evidence for impaired mineralization is often ambiguous. Looser’s zones can occur in the absence of osteomalacia and heal spontaneously [124]. Metaphyseal erosions in the absence of increased growth plate width are due to osteitis fibrosa, but they have often been attributed erroneously to rickets [125]. Problems in the histological recognition of defective mineralization were outlined earlier and are discussed in more detail elsewhere [6]. In the previously cited clinical studies [78–81], the authors demonstrated little awareness of the difficulties just mentioned, and the evidence that the lesions corrected by mineral administration alone were the result of defective mineralization rather than of secondary hyperparathyroidism was inconclusive. Of six cases included in the four reports, three had metaphyseal erosions with normal growth plate width, wrongly reported as rickets [79,80], three had qualitative bone histology only [78,79], and two had no bone biopsy at all [79,80]. Only in one case was there convincing histological evidence of osteomalacia, which healed with prolonged calcium infusion [81]. Studies in the rat have provided stronger evidence that mineral alone can be effective in vivo in the absence of vitamin D. Restoration of normocalcemia in vitamin D–deficient rats by a high calcium diet corrected both abnormal bone enzyme activity [126] and defective osteoid maturation [127]. Complete vitamin D deficiency was not demonstrated in these experiments, but in 25-day-old rats weaned from vitamin D–deficient mothers, and maintained without access to ultraviolet light or dietary vitamin D, plasma levels of calcitriol were undetectable [82]. Calcium chloride and buffered sodium phosphate infused into separate jugular veins, in amounts sufficient to maintain the same plasma calcium and phosphate levels as in vitamin D–replete rats, maintained normal growth plate width and normal tetracycline based indices of mineralization [83]. This experiment demonstrated conclusively that vitamin D was not essential for mineralization but did not exclude a contributory role for vitamin D under normal conditions. First, as previously discussed, there is only a very approximate relationship between total plasma levels and thermodynamic activity products. More importantly, the plasma levels were measured only once every 3 days at an unspecified time; because of circadian variation [121] the mean levels could have

1043

been higher in the infused than in the vitamin D– replete animals. Osteoid surface and volume were significantly lower in the infused rats [83], consistent with oversuppression of PTH secretion. Consequently, it remains possible that higher mean plasma levels of calcium and phosphate are needed to sustain normal mineralization in the absence than in the presence of vitamin D. The persistence in early osteomalacia of some doubly labeled surfaces with normal or only moderately reduced rates of mineral apposition [6] indicates that mineralization can proceed at the beginning of the osteoid seam life span, although it ceases prematurely (Fig. 4). Mineralizing and nonmineralizing osteoid seams are often close together, sometimes even in direct continuity, and they are exposed to the same microcirculation, so that the difference between them cannot be explained in terms of chemical change alone. However, at doubly labeled seams, a higher proportion of the surface is lined by osteoblasts [44,128], suggesting that these cells, possibly in conjunction with the osteocytes derived from them lying within the osteoid [129], are able to promote mineralization in the face of a moderate reduction in plasma ion product, but do so for a shorter time than normal in vitamin D depletion. When this function is lost, mineralization ceases even though matrix apposition continues slowly and the osteoid seam gets progressively thicker. In more severe osteomalacia, osteoblasts are fewer or absent altogether [130], mineralization never begins, and double labels are not found (Fig. 4). A similar relationship is observed during treatment: the recovery of mineralization in response to calcitriol administration, indicated by double labeling, occurs preferentially at surfaces where new osteoblasts have appeared [131,132]. The bone histological data in patients with osteomalacia strongly suggest that deficiency of calcitriol (and possibly also other metabolites) impairs some function of the osteoblast that favors mineralization. This proposal is consistent with the presence in osteoblasts of calcitriol receptors (VDR) [133], the autoradiographic localization of labeled calcitriol in osteoblast nuclei [134], the in vitro stimulation by calcitriol of several actions of osteoblasts including production of alkaline phosphatase [135] and osteocalcin [136], and sodium independent phosphate transport [137], the in vivo enhancement by calcitriol of mineral apposition rate in young mice [138], and the morphological changes induced by calcitriol in the cells lining quiescent bone surfaces [139] that are of osteoblast lineage [19]. A local cellular effect of one or more vitamin D metabolites would also account for the abnormalities in collagen cross-linking and other changes in bone matrix maturation and composition [36,140,141], and the changes in intermediary

1044 metabolism in cartilage cells [142,143], that have been found in vitamin D deficiency, although it is less clear that these are the result of a direct rather than an indirect effect of vitamin D on osteoblast and chondrocyte function. Finally, the proposal also accounts for the greater ability of vitamin D than calcium carbonate to improve bone mineralization in chronic renal failure, despite equivalent changes in total Ca × P product [144], on the assumption that bone cells can make calcitriol from its precursor. The usual approach to mineralization has been to study the physical chemistry of the solution in which the ions originate, and the events taking place in bone, and to largely ignore what happens in between [5]. But the circulating ions have to traverse a rather complex pathway before they arrive at the site of mineralization (Fig. 5). Having left the capillary and diffused through marrow connective tissue, they must pass through a layer of osteoblasts and a layer of osteoid before they can reach the site of mineral disposition. Osteoblasts on the surface, osteocytes within the osteoid, and osteocytes within mineralized bone are joined by a communicating network of cellular processes within the canaliculi. Very little is known about how mineral ions actually travel through this complex structure, but it would be surprising if cellular transport mechanisms of some kind were not involved in the movement of ions from the extracellular fluid to the site of mineral deposition. Indeed, it seems likely that calcitriol could stimulate the inward transport of calcium and/or phosphate ions through or between cells at sites of mineralization [5], consistent with its known effects on the cells of the intestinal mucosa (Chapter 24) and the renal tubule (Chapter 29). The osteoblast thus influences mineralization in two ways, by its effects on matrix maturation [14] and by its effects on mineral transport [52]. Furthermore, calcium and phosphate ions must be regarded not just as substrates for apatite formation but as part of the environment of the cell that is involved in their transport, as osteoblast function is influenced by the circulating and presumably also local levels of calcium [145] and phosphate [52,120]. The concept that mineralization normally depends both on the availability of substrate ions via the circulation and on the activity of osteoblasts and chondrocytes, and that vitamin D influences both of these processes, although by no means rigorously established, enables all the apparently conflicting data, laboratory and clinical, to be reconciled. The concept has the additional merit of providing a basis for unifying the pathogenesis of all major forms of osteomalacia, as hereditary or acquired defects in phosphate transport across the renal tubular epithelium, whether intrinsic or due to humoral factors,

A. MICHAEL PARFITT

could plausibly be accompanied by similar defects in transport across the quasi-epithelium that covers all bone surfaces [5,146].

References 1. McLean FC, Urist MR 1955 Bone: An Introduction to the Physiology of Skeletal Tissue. Univ. of Chicago Press, Chicago. 2. Robison R 1932 The Significance of Phosphoric Esters in Metabolism. New York Univ. Press, New York. 3. Lacroix P 1960 45Ca autoradiography in the study of bone tissue. In: Rodahl K, Nicholson JT, Brown EM (eds) Bone as a Tissue. McGraw-Hill, New York, pp. 262–279. 4. Frost HM, Villanueva AR 1960 Observations on osteoid seams. Henry Ford Hospital Med Bull 8:212–219. 5. Parfitt AM 1992 Human bone mineralization studied by in vivo tetracycline labeling: Application to the pathophysiology of osteomalacia. In Slavkin M, Price P (eds) Chemistry and Biology of Mineralized Tissues. Excerpta Medica, Amsterdam, pp. 465–474. 6. Parfitt AM 1997 Osteomalacia and related disorders. In: Avioli LV, Krane SM (eds) Metabolic Bone Disease and Clinically Related Disorders, 3rd Ed. Academic Press Inc., San Diego, pp. 645–662. 7. Hansson LI, Menander-Sellman K, Stenstrom A, Thorngren KG 1972 Rate of normal longitudinal bone growth in the rat. Calcif Tissue Res 10:238–251. 8. Thorngren KG, Hansson LI, Menander-Sellman K, Stenstrom A 1973 Effect of hypophysectomy on longitudinal bone growth in the rat. Calcif Tissue Res 11:281–300. 9. Caplan AI, Boyan BD 1994 Endochondral bone formation: The lineage cascade. In Hall BK (ed) Bone: A Treatise, Volume 8, Mechanisms of Bone Development and Growth. CRC Press, Boca Raton, Florida, pp. 1–46. 10. Yamauchi M, Chandler GS, Katz EP 1992 Collagen crosslinking and mineralization. In: Slavkin M, Price P (eds) Chemistry and Biology of Mineralized Tissues. Excerpta Medica, Amsterdam, pp. 39–46. 11. Gerstenfeld LC, Riva A, Hodgens K, Eyre DR, Landis WJ 1993 Post-translational control of collagen fibrillogenesis in mineralizing cultures of chick osteoblasts. J Bone Miner Res 8:1031–1043. 12. Boskey AL 1990 Mineral-matrix interactions in bone and cartilage. Clin Orthop Related Res 281:244–274. 13. Gorkale JA, Robey PG, Boskey AL 2001 The biochemistry of bone. In: Marcus R, Feldman D, Kelsey J (eds) Osteoporosis. 2nd Ed. Academic Press, San Diego, pp. 107–188. 14. Marsh ME, Munne AM, Vogel JJ, Cui Y, Franceschi RT 1995 Mineralization of Bone-like Extracellular Matrix in the Absence of Functional Osteoblasts. J Bone Min Res 10:1635–1643. 15. Parfitt AM, Kleerekoper M 1980 The divalent ion homeostatic system: Physiology and metabolism of calcium, phosphorus, magnesium, and bone. In: Maxwell M, Kleeman CR (eds) Clinical Disorders of Fluid and Electrolyte Metabolism, 3rd Ed. McGraw-Hill, New York, pp. 269–398. 16. Neuman WF 1980 Bone material and calcification mechanisms. In: Urist MR (ed) Fundamental and Clinical Bone Physiology. Lippincott, Philadelphia, Pennsylvania, pp. 83–107. 17. Glimcher MJ 1992 The nature of the mineral component of bone and the mechanism of calcification. In: Coe FL,

CHAPTER 63 Vitamin D and the Pathogenesis of Rickets and Osteomalacia

18. 19.

20. 21.

22. 23. 24.

25. 26.

27.

28. 29.

30.

31.

32. 33. 34. 35.

Favus MJ (eds) Disorders of Bone and Mineral Metabolism. Raven, New York, pp. 265–286. Glowacki J 2002 Mechanisms of Biomineralization. In: Disorders of Bone and Mineral Metabolism 2nd Ed. FL Coe, MJ Favus (eds) Lippincott Williams & Wilkins, pp. 227–234. Parfitt AM 1993 Calcium homeostasis. In: Mundy GR, Martin TJ (eds) Handbook of Experimental Pharmacology, Volume 107, Physiology and Pharmacology of Bone. SpringerVerlag, Heidelberg, pp. 1–65. Brighton CT, Hunt RM 1978 The role of mitochondria in growth plate calcification as demonstrated in a rachitic model. J Bone Joint Surg 60-A:630–639. Fedde KN, Blair L, Silverstein J, Coburn SP, Ryan LM, Weinstein RS, Waymire K, Narisawa S, Millan JL, MacGregor GR, Whyte MP 1999 Alkaline phosphatase knock-out mice recapitulate the metabolic and skeletal defects of infantile hypophosphatasia. J Bone Min Res. 14:2015–2026. Alcock NW, Shils ME 1968 Association of inorganic pyrophosphatase activity with normal calcification of rat costal cartilage in vivo. Biochem J 112:505–510. Meyer JL, Fleisch H 1984 Determination of calcium phosphate inhibitor activity; critical assessment of the methodology. Miner Electrolyte Metab 10:249–258. Hoshi K, Kemmotsu S, Takeuchi Y, Amizuka N, Ozawa H 1999 The Primary Calcification in Bones Follows Removal of Decorin and Fusion of Collagen Fibrils. J Bone Min Res 14:273–280. Christoffersen J, Landis WJ 1991 A contribution with review to the description of mineralization of bone and other calcified tissues in vivo. Anat Rec 230:435–450. Schenk RK, Hunziker EB 1991 Growth plate: Histophysiology, cell, and matrix turnover. In: Glorieux FH (ed) Rickets, Nestle Nutrition Workshop Series, Vol. 21. Raven, New York, pp. 63–78. Anderson HC, Morris DC 1993 Mineralization. In: Mundy GR, Martin TJ (eds) Handbook of Experimental Pharmacology, Volume 107, Physiology and Pharmacology of Bone. SpringerVerlag, Heidelberg, pp. 267–298. Boskey AL, Boyan BD, Schwartz Z 1997 Matrix Vesicles Promote Mineralization in a Gelatin Gel. Calcif Tissue Int, 60:309–315. Reinholt, Engfeldt B, Hjerpe A, Jansson K 1982 Stereological studies on the epiphyseal growth plate with special reference to the distribution of matrix vesicles. J Ultrastruct Res 80:270–279. Glorieux FH, Marie PJ, Pettifor JM, Delvin EE 1980 Bone response to phosphate salts, ergocalciferol, and calcitriol in hypophosphatemic vitamin D resistant rickets. N Engl J Med 303:1023–1031. Kember NF, Kirkwood JK 1991 Cell kinetics and the study of longitudinal bone growth: A perspective. In: Dixon AD, Sarnat BG, Hoyte DA (eds) Fundamentals of Bone Growth: Methodology and Applications. CRC Press, Boca Raton, Florida, pp. 153–162. Dodds GS, Cameron HC 1934 Studies on experimental rickets in rats. I. Structural modifications of the epiphyseal cartilages in the tibia and other bones. Am J Anat 55:135–165. Park EA 1939 Observations on the pathology of rickets with particular reference to the changes at the cartilage-shaft junctions of the growing bones. Bull NY Acad Med 15:495–543. Bicknell F, Prescott F 1953 The Vitamins in Medicine. Heinemann, London, pp. 529–533. Bailie JM, Irving JT 1955 Development and healing of rickets in intramembranous bone. Acta Med Sci 152(Suppl. 306):1–14.

1045

36. Baylink D, Stauffer M, Wergedal J, Rich C 1970 Formation, mineralization, and resorption of bone in vitamin D–deficient rats. J Clin Invest 49:1122–1134. 37. Parfitt AM 1997 Genetic effects on bone mass and turnoverrelevance to black/white differences. J Amer Coll Nutr 16:325–333. 38. Schaffler MB, Burr DB, Frederickson RG 1987 Morphology of the osteonal cement line in human bone. Anat Rec 217:223–228. 39. Parfitt AM 1992 The physiologic and pathogenetic significance of bone histomorphometric data. In: FL Coe, MJ Favus, (eds) Disorders of Bone and Mineral Metabolism. 2nd Edition. 2002. Lippincott Williams and Wilkins. Philadelphia, pp. 469–485. 40. Parfitt AM, Foldes J, Villanueva AR, Shih MS 1991 The difference in label length between demethylchlortetracycline and oxytetracycline: Implications for the interpretation of bone histomorphometry. Calcif Tissue Int 48:74–77. 41. Eriksen EF 1986 Normal and pathological remodeling of human trabecular bone: Three-dimensional reconstruction of the remodeling sequence in normals and in metabolic bone disease. Endocr Rev 4:379–408. 42. Parfitt AM 1984 The cellular mechanisms of osteoid accumulation in metabolic bone disease. In: Mineral Metabolism Research in Italy, Vol. 4. Wichtig Editore, Milano, pp. 3–9. 43. Parfitt AM 1986 Accelerated cortical bone loss: Primary and secondary hyperparathyroidism. In: Uhthoff H (ed) Current Concepts of Bone Fragility. Springer-Verlag, New York, pp. 279–285. 44. Meunier PJ, Van Linthoudt D, Edouard C, Charhon S, Arlot M 1981 Histological analysis of the mechanisms underlying pathogenesis and healing of osteomalacia. Proceedings, Sixteenth European Calcified Tissue Symposium. Abstracts 33:771–774. 45. Peacock M 1993 Osteomalacia and rickets. In: Nordin BEC, Need AG, Morris HA (eds), Metabolic Bone and Stone Disease, 3rd Ed. Churchill Livingstone, London, pp. 83–118. 46. Parfitt AM, Qiu S, Rao DS. The Mineralization Index: A new approach to the histomorphometric appraisal of osteomalacia. (in press). 47. Parfitt AM 2001 The Evolution of Vitamin D related Bone Disease: The Importance of an Early Stage of Increased Bone Turnover Without Impaired Mineralization. In: Nutritional Aspects of Osteoporosis, P Burkhardt, R Heaney, B DawsonHughes (eds). Academic Press, San Diego, pp. 197–208. 48. Stanbury SW 1980 Vitamin D and calcium metabolism. In: Norman AW (ed) Vitamin D. Molecular Biology and Clinical Nutrition. Dekker, New York, pp. 251–319. 49. Fraser D, Kooh SW, Scriver CR 1967 Hyperparathyroidism as the cause of hyperaminoaciduria and phosphaturia in human vitamin D deficiency. Pediatr Res 1:425–435. 50. David L 1991 Common vitamin D–deficiency rickets. In: Glorieux FH (ed) Rickets, Nestle Nutrition Workshop Series, Vol 21. Raven, New York, pp. 107–122. 51. Vainsel M, Manderlier T, Corvilain J, Vis HL 1974 Study of the secondary hyperparathyroidism in vitamin D–deficiency rickets. I. Aspects of mineral metabolism. Biomedicine 21:368–371. 52. Stanbury SW, Mawer EB 1990 Metabolic Disturbances in Acquired Osteomalacia. In: Cohen RD, Lewis B, Alberti KGMM, Denman AM (eds) The Metabolic and Molecular Basis of Acquired Disease. Bailliere Tindall, London, pp. 1717–1782. 53. Rao DS, Villanueva A, Mathews M, Pumo B, Frame B, Kleerekoper M, Parfitt AM 1983 Histologic evolution of

1046

54.

55.

56.

57.

58. 59. 60. 61. 62. 63. 64.

65.

66.

67.

68. 69.

vitamin depletion in patients with intestinal malabsorption or dietary deficiency. In: Frame B, Potts JT Jr (eds) Clinical Disorders of Bone and Mineral Metabolism. Excerpta Medica, Amsterdam, pp. 224–226. Arnaud CD 1991 Parathyroid hormone and its role in the pathophysiology of the common forms of rickets and osteomalacia. In: Glorieux FH (ed) Rickets, Nestle Nutrition Workshop Series, Vol. 21. Raven, New York, pp. 47–61. Rao DS, Parfitt AM, Kleerekoper M, Pumo BS, Frame B 1985 Dissociation between the effects of endogenous parathyroid hormone on cAMP generation and on phosphate reabsorption in hypocalcemia due to vitamin D depletion: An acquired disorder resembling pseudohypoparathyroidism type II. J Clin Endocrinol Metab 61:285–290. Brazier M, Kamel S, Maamer M, Agbomson F, Elesper I, Garabedian M, Desmet G, Sebert JL 1995 Markers of bone remodeling in the elderly subject: Effects of vitamin D insufficiency and its correction. J Bone Miner Res 10:1753–1761. Parfitt AM, Kleerekoper M, Cruz C 1986 Reduced phosphate reabsorption unrelated to parathyroid hormone after renal transplantation: Implications for the pathogenesis of hyperparathyroidism in chronic renal failure. Miner Electrolyte Metab 12:356–362. Fraser DR 1983 The physiological economy of vitamin D. Lancet 1:969–972. Fraser DR 1991 Physiology of vitamin D and calcium homeostasis. In: Glorieux FH (ed) Rickets, Nestle Nutrition Workshop Series, Vol 21. Raven, New York, pp. 23–34. Mawer EF, Davies M 2001 Vitamin D nutrition and bone disease in adults. Reviews Endocrine & Metabolic Disorders. 2:153–164. Parfitt AM 1978 Adult hypoparathyroidism: Treatment with calcifediol. Arch Intern Med 138:874–881. Clements MR, Chalmers TM, Fraser DR 1984 Enterohepatic circulation of vitamin D: A reappraisal of the hypothesis. Lancet 1:1376–1379. Clements MR, Johnson L, Fraser DR 1987 A new mechanism for induced vitamin D deficiency in calcium deprivation. Nature 325:62–65. Bisballe S, Eriksen EF, Melsen F, Mosekilde L, Sorensen OH, Hessov I 1991 Osteopenia and osteomalacia after gastrectomy: Interrelations between biochemical markers of bone remodeling, vitamin D metabolites, and bone histomorphometry. Gut 32:1303–1307. Clements MR, Davies M, Hayes ME, Hickey CD, Lumb GA, Mawer EB, Adams PH 1992 The role of 1,25-dihydroxyvitamin D in the mechanism of acquired vitamin D deficiency. Clin Endocrinol 37:17–27. Davies M, Heys SE, Selby PL, Berry JL, Mawer EB 1997 Increased catabolism of 25-hydroxyvitamin D in patients with partial gastrectomy and elevated 1,25-dihydroxyvitamin D levels. Implications for metabolic bone disease. J Clin Endocrinol 82:209–212. Epstein S, Meunier PJ, Lambert PW, Stern PH, Bell NH 1993 1,25-Dihydroxyvitamin D3 corrects osteomalacia in hypoparathyroidism and pseudohypoparathyroidism. Acata Endocrinol 103:241–247. Brommage R, DeLuca HF 1985 Evidence that 1,25-dihydroxyvitamin D3 is the physiologically active metabolite of vitamin D3. Endocr Rev 6:491–511. Norman AW, Roth J, Orci L 1982 The vitamin D endocrine system: Steroid metabolism, hormone receptors, and biological response (calcium binding proteins). Endocr Rev 3:331–366.

A. MICHAEL PARFITT

70. Nagant de Deuxchaisnes C, Rombouts-Lindemans C, Huaux JP, Withofs H, Meersseman F 1979 Healing of vitamin D– deficient osteomalacia by the administration of 1,25(OH)2D3. In: MacIntyre I, Szelke M (eds) Molecular Endocrinology, pp. 375–404. 71. Papapoulos SE, Clements TL, Fraher LJ, Gleed J, O’Riordan JLH 1980 Metabolites of vitamin D3 in human vitamin deficiency: Effect of vitamin D3 or 1,25-dihydroxycholecalciferol. Lancet 2:612–615. 72. Parfitt AM, Mathews CHE, Brommage R, Jarnagin K, DeLuca HF 1984 Calcitriol but no other metabolite of vitamin D is essential for normal bone growth and development in the rat. J Clin Invest 73:576–586. 73. Lund B, Charles P, Egsmose C, Lund BJ, Melson F, Mosekilde L, Storm T, Sondergard H, Thode J, Sorensen OH 1985 Changes in vitamin D metabolites and bone histology in rats during recovery from rickets. Calcif Tissue Int 37:478–483. 74. Rasmussen H, Baron R, Broadus A, DeFronzo R, Lang R, Horst R 1980 1,25(OH)2D3 is not the only D metabolite involved in the pathogenesis of osteomalacia. Am J Med 69:360–368. 75. St Amaud R, Glorieux FH 1998 Editorial: 24, 25-dihydroxyvitamin D-Active Metabolite or Inactive Catabolite? Endocrinol 39:3371–3374. 76. St. Amaud R, Arabian A, Travers R, Barletta F, Raval-Pandya M, Chapin K, Depovere J, Mathieu C, Christakos S, Demay MB, Glorieux FH 2000 Deficient mineralization of intramembranous bone in vitamin D-24-hydroxylase–ablated mice is due to elevated 1,25-dihydroxyvitamin D and not to the absence of 24, 25-dihydroxyvitamin D. Endocrinology 141: 2658–2666. 77. Wronski TJ, Halloran BP, Bikle DD, Globus RK, Morey-Holton ER 1986 Chronic administration of 1,25dihydroxyvitamin D3: increased bone but impaired mineralization. Endocrinology 119:2580–2585. 78. Popovtzer MM, Mathay R, Alfrey AC, Block M, Beck P, Miles J, Reeve EB 1973 Vitamin D deficiency osteomalacia— Healing of the bone disease in the absence of vitamin D with intravenous calcium and phosphorus infusions. In: Frame B, Parfitt AM, Duncan H (eds) Clinical Aspects of Metabolic Bone Disease. Excerpta Medica, Amsterdam, pp. 382–387. 79. Sakati N, Woodhouse NJY, Niles N, Harfl H, de Grange DA, Marx S 1986 Hereditary resistance to 1,25-dihydroxyvitamin D: Clinical and radiological improvement during high-dose oral calcium therapy. Horm Res 24:280–287. 80. Weisman Y, Bab I, Gazit D, Spirer Z, Jaffe M, Hochberg Z 1987 Long-term intracaval calcium infusion therapy in endorgan resistance to 1,25-dihydroxyvitamin D. Am J Med 83:984–990. 81. Balsan S, Garabedian M, Larchet M, Gorski AM, Cournot G, Tau C, Bourdeau A, Silve C, Ricour C 1986 Long-term nocturnal calcium infusions can cure rickets and promote normal mineralization in hereditary resistance to 1,25-dihydroxyvitamin D. J Clin Invest 77:1661–1667. 82. Underwood JR, DeLuca H 1984 Vitamin D is not directly necessary for bone growth and mineralization. Am J Physiol 246:493–498. 83. Weinstein RS, Underwood JL, Hutson MS, DeLuca HF 1984 Bone histomorphometry in vitamin D-deficient rats infused with calcium and phosphorus. Am J Physiol 246: E499–E505. 84. Compston JE, Vedi S, Merrett AL, Clemens TL, O’Riordan JLH, Woodhead JS 1981 Privational and malabsorption

CHAPTER 63 Vitamin D and the Pathogenesis of Rickets and Osteomalacia

85. 86.

87.

88.

89. 90.

91. 92.

93.

94. 95. 96. 97.

98.

99.

100.

101.

metabolic bone disease: Plasma vitamin D metabolite concentrations and their relationship to quantitative bone histology. Metab Bone Dis Related Res 3:165–170. Kashiwa H, Nishi Y, Usui T, Seino Y 1981 A case of rickets with normal serum level of 1,25-(OH)2D and low 25-OHD. Hiroshima J Med Sci 30:61–63. Chesney RW, Zimmerman J, Hamstra A, DeLuca MF, Mazess RB 1981 Vitamin D metabolite concentrations in vitamin D deficiency. Are calcitriol levels normal? Am J Dis Children 135:1025–1028. Markestad T, Halvorsen S, Halvorsen KS, et al. 1984 Plasma concentrations of vitamin D metabolites before and during treatment of vitamin D deficiency rickets in children. Acta Paediatr Scand 73:225–231. Stanbury SW, Taylor CM, Lumb GA, Mawer B, Berry J, Hann J, Wallace J 1981 Formation of vitamin D metabolites following correction of human vitamin D deficiency. Miner Electrolyte Metab 5:212–227. Wahl TO, Gobuty AH, Lukert BP 1981 Long-term anticonvulsant therapy and intestinal calcium absorption. Clin Pharmacol Ther 30:506–516. Demiaux B, Arlot ME, Chapuy M-C, Meunier PJ, Delmas PD 1992 Serum osteocalcin is increased in patients with osteomalacia: Correlations with biochemical and histomorphometric findings. J Clin Endocrinol Metab 74: 1146–1151. Stanbury SW 1981 Vitamin D and hyperparathyroidism. J R Coll Phys Lond 15:205–217. Clements MR, Davies M, Fraser DR, Lumb GA, Mawer EB, Adams PH 1987 Metabolic inactivation of vitamin D is enhanced in primary hyperparathyroidism. Clin Sci 73:659–664. Stamp TCB, Walker PG, Perry W, Jenkins MW 1980 Nutritional osteomalacia and late rickets in greater London 1974–1979: Clinical and metabolic studies in 45 patients. Clin Endocrinol Metab 9:81–105. Raisz LG, Trummel CL, Holick MF, DeLuca HF 1972 1,25Dihydroxycholecalciferol: A potent stimulator of bone resorption in tissue culture. Science 175:768–869. Bouillon R, Van Baelen H 1981 Transport of vitamin D: Significance of free and total concentrations of the vitamin D metabolites. Calcif Tissue Int 33:451–453. Wilhelm F, Norman AW 1984 Cooperativity in the binding of 1,25-dihydroxyvitamin D3 to the chick intestinal receptor. FEBS Lett 170:239–242. Howard GA, Turner RT, Sherrard DJ, Baylink DJ 1981 Human bone cells in culture metabolize 25-hydroxyvitamin D3 to 1,25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3. J Biol Chem 256:7738–7740. Ichikawa F, Sato K, Nanjo M, Nishii Y, Shinki T, Takahashi N, Suda T 1995 Mouse primary osteoblasts express vitamin D3 25-hydroxylase mRNA and convert 1 α-hydroxyvitamin D3 into 1α,25-dihydroxyvitamin D3. Bone. 16:129–135. Puzas JE, Turner RT, Howard GA, Baylink DJ 1983 Cells isolated from embryonic intestine synthesize 1,25-dihydroxyvitamin-D3 and 24,25-dihydroxyvitamin-D3 in culture. Endocrinology 112:378–380. Takezawa K, Moorthy B, Mandel ML, Garancis JC, Ghazarian JG 1990 Antigenic and catalytic disparity in the distribution of cytochrome P-450-dependent 25-hydroxyvitamin D3-1 α- and 24-hydroxylases. Histochemistry 95: 37–42. McDonald GB, Lau K-HW, Schy AL, Wergedal JE, Baylink DJ 1985 Intestinal metabolism and portal venous transport of

102. 103. 104. 105.

106. 107. 108. 109. 110. 111. 112.

113.

114.

115. 116. 117.

118. 119.

120.

1047

1,25(OH)2D3, 25(OH)D3, and vitamin D3 in the rat. Am J Physiol 248:G633–G638. Trummel CL, Raisz LG, Blunt JW, DeLuca HF 1969 25-Hydroxycholecalciferol: Stimulation of bone resorption in tissue culture. Science 163:1450–1451. Olson EB, DeLuca HF 1969 25-Hydroxycholecalciferol: Direct effect on calcium transport. Science 165:405–407. Parfitt AM, Gallagher JC, Heaney RP, Johnston CC, Neer R, Whedon GD 1982 Vitamin D and bone health in the elderly. Am J Clin Nutr 36:1014–1031. Eastwood JB, Stamp TCB, DeWardener HE, Bordier PJ, Arnaud CD 1976 The effect of 25-hydroxy vitamin D3 in the osteomalacia of chronic renal failure. Clin Sci Mol Med 52:499–508. Howland J, Kramer B 1921 Calcium and phosphorus in the serum in relation to rickets. Am J Dis Chldren 22:105–119. Howland J, Kramer B 1923 A study of the calcium and inorganic phosphorus of the serum in relation to rickets and tetany. Monatschrift fur Kinderheilkd 25:279–293. Shipley PG, Kramer B, Howland J 1926 Studies upon calcification in vitro. Biochem J 20:379–387. Nordin BEC, Smith DA 1967 Pathogenesis and treatment of osteomalacia. In: Hioco DJ (ed) L’Osteomalacie. Masson & Cie, Paris, pp. 374–399. Kaplan FS, August CS, Fallon MD, Gannon F, Haddad JG 1993 Osteopetrorickets—The paradox of plenty. Pathophysiology and treatment. Clin Orthop Related Res 294:64–78. Stanbury SW 1962 Osteomalacia, Schweiz Med Wochenschr 29:883–892. Yendt ER, Connor TB, Howard JE 1955 In vitro calcification of rachitic rat cartilage in normal and pathological human sera with some observations on the pathogenesis of renal rickets. Bull Johns Hopkins Hospital 96/97:1–19. Dean DD, Boyan BD, Muniz OE, Howell DS, Schwartz Z 1996 Vitamin D Metabolites Regulate Matrix Vesicle Metalloproteinase content in a Cell Maturation-Dependent Manner. Calci Tiss Internat. 59:109–116. Bordier PH, Hioco D, Roquier, Hepner GW, Thompson GR 1969 Effects of intravenous vitamin D on bone and phosphate metabolism in osteomalacia. Calcif Tissue Res 4:78–83. Bordier P, Pechet MM, Hesse R, Marie P, Rasmussen H 1974 Response of adult patients with osteomalacia to treatment with crystalline 1α-hydroxy vitamin D3. N Engl J Med 291:866–871. Bordier P, Rasmussen H, Marie P, Miravet L, Gueris J, Ryckwaert A 1978 Vitamin D metabolites and bone mineralization in man. J Clin Endocrinol Metab 46:284–294. Compston JE, Horton LWL, Thompson RPH 1979 Treatment of osteomalacia associated with primary biliary cirrhosis with parenteral vitamin D2 or oral 25-hydroxyvitamin D3. Gut 20:133–136. Compston JE, Crowe JP, Horton LWL 1979 Treatment of osteomalacia associated with primary biliary cirrhosis with oral 1α-hydroxy vitamin D3. Br Med J 309–312. Compston JE, Horton LWL, Laker MF, Merrett AL, Woodhead JS, Gazet J-C, Pilkington TRE 1980 Treatment of bone disease after jejunoileal bypass for obesity with oral 1α-hydroxyvitamin D3. Gut 21:669–674. Parfitt AM, Villanueva AR 1982 Hypophosphatemia and osteoblast function in human bone disease. In: Massry SG, Letteri JM, Ritz E (eds) Proceedings, 5th International Workshop on Phosphate and Other Minerals. Regulation of Phosphate and Mineral Metabolism. Adv Exp Med Biol 151:209–216.

1048 121. Markowitz ME, Rosen JF, Laxminarayan S, Mizruchi M 1984 Circadian rhythms of blood minerals during adolescence. Pediatr Res 18:456–462. 122. De Vernejoul MC, Marie PJ, Miravet L, Ryckewaert A 1983 Chronic hypophosphatemia without osteomalacia. In: Frame B, Potts JT Jr (eds) Clinical Disorders of Bone and Mineral Metabolism. Excerpta Medica, Amsterdam, pp. 232–236. 123. Oginni LM, Sharp CA, Worsfold M, Badru OS, Davie JWJ 1999 Healing of rickets after calcium supplementation. Lancet 353:296. 124. McKenna MJ, Kleerekoper M, Ellis BI, Dao BS, Parfitt AM, Frame B 1987 Atypical insufficiency fractures confused with Looser zones of osteomalacia. Bone 8:71–78. 125. Parfitt AM 1977 The clinical and radiographic manifestations of renal osteodystrophy. In: David DG (ed) Perspectives in Hypertension and Nephrology: Calcium Metabolism in Renal Failure and Nephrolithiasis. Wiley, New York, pp. 145–195. 126. Wergedal JE, Baylink JE 1971 Factors affecting bone enzymatic activity in vitamin D–deficient rats. Am J Physiol 220:406–409. 127. Howard GA, Baylink DJ 1980 Matrix formation and osteoid maturation in vitamin D–deficient rats made normocalcemic by dietary means. Miner Electrolyte Metab 3:44–50. 128. Sebert JL, Meunier PJ 1984 Role physiopathologique de la vitamine D et de ses metabolites dans l’osteomalacie. In: Bouillon R, Boudailliez B, Marie A, et al. (eds) Vitamine D et Maladies des Os et du Metabolisme Mineral. Masson, Paris, pp. 109–145. 129. Bordier PJ, Marie P, Miravet L, et al. 1976 Morphological and morphometrical characteristics of the mineralization front. A vitamin D regulated sequence of the bone remodeling. In: Meunier PJ (ed) Bone Histomorphometry. Second International Workshop. Armour Montagu, Paris, pp. 335–354. 130. Weinstein RS 2002 Clinical use of bone biopsy. In: Coe FL, Favus MJ (eds) Disorders of Bone and Mineral Metabolism. 2nd Ed. Lippincott, Williams & Wilkins. Philadelphia, pp. 449–468. 131. Meunier PJ, Edouard C, Arlot M, et al. 1979 Effects of 1,25dihydroxyvitamin D on bone mineralization. In: Maclntyre I, Szelke M (eds) Molecular Endocrinology. Elsevier/NorthHolland, Amsterdam, pp. 283–292. 132. Marie PJ, Glorieux FH 1981 Histomorphometric study of bone remodeling in hypophosphatemic vitamin D–resistant rickets. Metab Bone Dis Related Res 3:31–38.

A. MICHAEL PARFITT

133. Manolagas SC, Haussler MR, Deftos LJ 1980 1,25Dihydroxyvitamin D3 receptor-like macromolecule in rat osteogenic sarcoma cell lines. J Biol Chem 255:4414–4417. 134. Stumpf WE, Ssar M, DeLuca HF 1981 Sites of action of 1,25(OH)2 vitamin D3 identified by thaw-mount autoradiography. In: Cohn DV, Talmage RV, Matthews JL (eds) Hormonal Control of Calcium Metabolism. Excerpta Medica, Amsterdam, pp. 222–229. 135. Manolagas SC, Burton DW, Deftos LJ 1981 1,25Dihydroxyvitamin D3 stimulates the alkaline phosphatase activity of osteoblastlike cells. J Biol Chem 256:7115–7117. 136. Lian JB, Coutts M, Canalis E 1985 Studies of hormonal regulation of osteocalcin synthesis in cultured fetal rat calvariae. J Biol Chem 260:8706–8710. 137. Veldman CM, Schläpper I, Schmid Ch 1997 1α,25-Dihydroxyvitamin D3 stimulates sodium-dependent phosphate transport in osteoblast-like cells. Bone 21:41–47. 138. Marie PJ, Hott M, Garba M-T 1985 Contrasting effects of 1,25-dihydroxyvitamin D3 on bone matrix and mineral appositional rates in the mouse. Metabolism 34:777–783. 139. Krempien B, Klimpel F 1980 Action of 1,25-dihydroxycholecalciferol on cartilage mineralization and on endosteal lining cells of bone. Virch Arch [A] 388:335–347. 140. Stern PH 1980 The D vitamins and bone. Pharmacol Rev 32:47–80. 141. Dickson IR, Roughley PJ 1993 The effects of vitamin D deficiency on proteoglycan and hyaluronate constituents of chick bone. Biochim Biophys Acta 1181:15–22. 142. Tulpule PG, Patwardhan VN 1954 Mode of action of vitamin D. The effect of vitamin D deficiency on the rate of anaerobic glycolysis and pyruvate oxidation by epiphyseal cartilage. Biochem J 58:61–65. 143. Klein GL, Simmons DJ 1993 Nutritional rickets: Thoughts about pathogenesis. Ann Med 25:379–384. 144. Eastwood JB, Bordier PJ, Clarkson EM, Tun Chot S, De Wardener HE 1973 The contrasting effects on bone histology of vitamin D and of calcium carbonate in the osteomalacia of chronic renal failure. Clin Sci Mol Med 47:23–42. 145. Stauffer M, Baylink D, Wergedal J, Rich C 1975 Decreased bone formation, mineralization, and enhanced resorption in calcium-deficient rats. Am J Physiol 225:269–276. 146. Ecarot B, Glorieux FH, Desbarats M, Travers R, Labelle L 1992 Defective bone formation by Hyp mouse bone cells transplanted into normal mice: Evidence in favor of an intrinsic osteoblast defect. J Bone Miner Res 7:215–220.

CHAPTER 64

The Hypocalcemic Disorders: Differential Diagnosis and Therapeutic Use of Vitamin D THOMAS O. CARPENTER

Department of Pediatrics, Yale University School of Medicine, New Haven, Connecticut

KARL L. INSOGNA

Department of Medicine, Yale University School of Medicine, New Haven, Connecticut

I. Physiology II. Differential Diagnosis of Hypocalcemia

I. PHYSIOLOGY A. Hypocalcemia and Its Manifestations Hypocalcemia refers to an abnormally low concentration of ionized calcium in extracellular fluid, almost invariably sampled from the bloodstream. Manifestations of hypocalcemia are related to increased neuromuscular irritability [1]. Tetany is the classic sign of hypocalcemia, yet it is variable in presentation. Paresthesias often occur first around the mouth or in the fingertips and may progress to overt spasm of the muscles of the face and extremities, the latter typified by carpopedal spasm. More subtle presentations have included complaints of writer’s cramp or generalized stiffness. Children with tetanic laryngospasm due to hypocalcemia have been mistakenly diagnosed with croup [2]. Infants are more likely than adults to present with jitteriness or twitching, which can progress to overt tonic-clonic seizure activity. Lethargy and cyanosis have also been described in this age group. The term latent tetany refers to signs elicitable with provocative stimuli such as ischemia (Trousseau test) or percussion (e.g., of the facial nerve to elicit Chvostek’s sign). Neither the degree of hypocalcemia nor the rapidity with which it develops necessarily correlate with clinical manifestations. Hypomagnesemia or hyperkalemia may present with similar findings, which can be exacerbated in the setting of hypocalcemia. Conversely, hypermagnesemia or hypokalemia can mask symptoms in a hypocalcemic individual. Abnormalities of repolarization of cardiac musculature may result in a prolonged Q-T interval on the electrocardiogram (EKG). The Q-T interval corrected for heart rate [Q-TC, which equals Q-T/(R-R interval)1/2], VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

III. Therapy for Hypocalcemia References

is normally less than 0.40 ± 0.04 sec. This abnormality is not always present during hypocalcemia, and it may also be seen in hypokalemia. Cardiac failure may occur in the setting of hypocalcemia [3]. Papilledema has also been attributed to hypocalcemia [4]. Chronic hypocalcemia caused by deficient calcium intake during periods of significant skeletal growth may result in rickets and osteomalacia (see Chapter 63). Severe osteoporosis and dental abnormalities have also been reported in long-standing untreated hypoparathyroidism [5,6]. A mineralization defect has been described in hypoparathyroidism; however, these skeletal consequences appear to be more prevalent in the setting of endemic calcium deficiency, where secondary hyperparathyroidism is evident. Basal ganglia calcifications are typical findings in long-standing hypoparathyroidism as well [7]. Abnormalities in the integument including dry skin, coarse hair, and a form of psoriasis that responds to normalization of the serum calcium concentration [8] have all been described in states of long-standing hypocalcemia. Regulatory mechanisms maintain the concentration of ionized calcium within a remarkably narrow range of 4.48 to 5.28 mg/dl in whole blood [9]. The ionized fraction of total serum calcium is generally estimated to be approximately 50%, with the remainder of the total serum calcium being bound to serum proteins, most notably albumin, and to a lesser extent complexed with anions, such as citrate or sulfate. Only the ionized fraction of total serum calcium is physiologically important, and it is this component that is regulated on a minuteto-minute basis. Although it is possible to measure ionized calcium routinely in large clinical laboratories, the specimen Copyright © 2005, Elsevier, Inc. All rights reserved.

1050 must usually be obtained anaerobically and analyzed promptly. Therefore, total serum calcium is often used as an indirect assessment of the ionized calcium fraction. A decrease in serum protein concentrations (particularly albumin) often results in reduced total serum calcium concentrations, with preservation of a normal concentration of ionized calcium. Patients in whom this occurs will be asymptomatic, displaying none of the signs or symptoms of hypocalcemia. These findings are often present in patients with nephrotic syndrome, chronic illness, malnutrition, cirrhosis, and volume over-expansion. A number of clinical guidelines have been suggested that correct for the effects of decreased serum albumin on total serum calcium concentration. One commonly cited rule of thumb is to add 0.8 mg/dl to the total serum calcium for every 1 g/dl decline in serum albumin below 4.0 g/dl. However, these estimates have been shown to be somewhat inaccurate under many circumstances, and it is preferable to directly determine the ionized calcium concentration in the setting where the total serum calcium measure is not representative of the ionized calcium measure [10].

B. Role of Parathyroid Hormone in the Acute Defense of Ionized Serum Calcium Concentration Parathyroid hormone (PTH) is secreted by the parathyroid glands in response to a fall in ionized serum calcium concentration [9]. The relationship between decrements in ionized calcium within the physiological range and increments in PTH secretion is quite steep, permitting rapid and substantial changes in PTH secretion in response to minor fluctuations in ionized calcium [9]. The details of this response have been elucidated with the cloning of the seven transmembrane domain, G–protein-coupled calcium sensor, which is expressed in the parathyroid glands as well as in a variety of other tissues [11] (see Chapter 31). A rise in ionized calcium suppresses PTH secretion by activating this receptor [11]. Parathyroid hormone acts to regulate ionized calcium through its effects in three principal target tissues, bone, kidney, and intestine (see Chapter 30). The cellular mechanisms of action of PTH have been clarified by the cloning of the PTH receptor, also a seven transmembranedomain, G–protein-coupled receptor [12]. Downstream signaling from the PTH receptor involves activation of both protein kinase A- and protein kinase C-dependent pathways [13–15]. Parathyroid hormone acts to increase bone resorption, liberating calcium from the mineralized matrix of

THOMAS O. CARPENTER AND KARL L. INSOGNA

bone and thereby increasing the ionized calcium concentration of the extracellular fluid [9]. Details of the cellular mechanism by which this occurs are incompletely understood. The principal target cell for PTH in bone appears to be the osteoblast or osteoblast-like stromal cell, rather than the resorbing cell itself, the osteoclast [16,17]. In response to PTH, osteoblasts or osteoblastlike stromal cells release locally active cytokines that appear to increase the number and activity of osteoclasts [18]. As the resorptive response to PTH is quite rapid, the early effects of the hormone are likely mediated by activation of existing osteoclasts. A miscible pool of incompletely mineralized calcium at the endosteal surface of bone may be the most readily accessible source of calcium liberated in response to this action. It has been speculated that osteocytes may mediate release of calcium from this pool [19]. These effects are evident in 6–12 hr [20]. The renal effects of PTH to defend serum calcium occur within minutes (see Chapter 76). PTH increases calcium reabsorption in the distal tubule. This effect is greatest in the distal convoluted tubule, where a sodium/calcium exchanger is regulated by PTH [21]. In the proximal renal tubule, PTH acts via a cAMPdependent mechanism to decrease phosphate reabsorption, resulting in increased phosphaturia. PTH effects this change by prompting the removal of sodium/phosphate cotransporters from the renal tubular apical membrane [22]. These two effects both serve to acutely increase serum calcium; one by causing less calcium to be excreted by the kidney, the other by lowering circulating concentrations of phosphate which favors an increase in ionized calcium. The third site of action of PTH in the defense of serum calcium is at the intestine. This is an indirect effect, described in detail below, and is a consequence of the ability of PTH to stimulate the renal production of 1,25-dihydroxyvitamin D [1,25(OH)2D].

C. Vitamin D in the Long-term Maintenance of Eucalcemia Long-term eucalcemia is maintained, in large part, via the vitamin D endocrine system. This system operates in the classic manner of a steroid hormone, resulting in de novo protein synthesis directed by vitamin D responsive genes [23] as discussed in detail in Section II of this volume. As noted above, acute changes in serum ionized calcium levels are sensed by G–protein-coupled calcium-sensing receptors located within the parathyroid cell membrane [11]. PTH acts rapidly to correct a fall in serum calcium, and a sustained increase in PTH also stimulates production of 1,25(OH)2D, which

CHAPTER 64 The Hypocalcemic Disorders: Differential Diagnosis and Therapeutic Use of Vitamin D

enhances intestinal calcium absorption. PTH mediates this change by increasing the activity of the renal 25-hydroxyvitamin D (25OHD) 1-hydroxylase enzyme complex, located in the inner mitochondrial membrane of renal tubular cells. A number of physiological studies have demonstrated increased enzyme activity in animals that were administered PTH [24,25] and decreased activity following parathyroidectomy [26]. The mechanism is discussed in Chapters 5 and 29. PTH also acutely regulates the 1-hydroxylase enzyme complex by altering the phosphorylation state of the associated ferredoxin molecule [27]. Direct stimulation of 1-hydroxylase activity in the absence of PTH can also occur because a low serum calcium level enhances enzyme activity in parathyroidectomized rats [28,29] and in hypoparathyroid humans [5]. Increased synthesis of 1,25(OH)2D results in greater circulating levels of the metabolite, which gain access to specific vitamin D receptors (VDR). The hormonereceptor complex then binds to vitamin D response elements (VDRE) in the regulatory regions of target genes (Chapters 11–14). Of importance to long-term control of calcium homeostasis is the induction by 1,25(OH)2D of the intestinal 9-kDa calcium binding protein, calbindin-D9k (Chapter 42), which is thought to play a role in vitamin D–mediated increases in calcium absorption in the jejunum and duodenum [30] (see also Chapters 24 and 25). Induction of calbindin takes hours to days and is more sustained than the acute compensatory changes that occur with the initial rise in PTH in response to hypocalcemia. These features define a classic feedback loop suitable for long-term calcium homeostasis. More recently rapid, nongenomic actions of 1,25(OH)2D mediating calcium transport across intestinal mucosa have been described [31] (see Chapter 23). The importance of this system is emphasized by clinical observations in patients with severe vitamin D deficiency (see below). During vitamin D deprivation, the initial decline in ionized serum calcium results in secondary hyperparathyroidism, which maximizes 1,25(OH)2D production and allows for maintenance of eucalcemia in the early stages. Eventually, this compensatory mechanism fails, and intestinal calcium absorption is sufficiently compromised that frank hypocalcemia develops. This may be compounded by an induced resistance to PTH seen in severe hypocalcemic or vitamin D–deficient states [1]. In children with hereditary resistance to vitamin D (HVDRR; see Chapter 72) caused by mutations in the VDR, the compensatory changes described previously are interrupted by defective VDR and the inability of 1,25(OH)2D to signal to the nucleus [32]. Such patients can have severe hypocalcemia leading to convulsions, coma, and death.

1051

The vitamin D system has further complexities that are currently not well understood. For example, some children with vitamin D deficiency (defined by low 25OHD levels) may manifest hypocalcemia even when circulating 1,25(OH)2D levels are actually elevated. Possible explanations for this phenomenon include decrements in expression of calbindin during hypocalcemia, or a requirement for other circulating vitamin D metabolites which are not present. This situation is discussed further in Chapter 31. Whether changes in the concentration of VDR levels play a role in this “resistant” state is not clear, as receptor levels have been reported to increase, decrease, or not change with manipulation of the ambient calcium concentration [33–35] (see Chapters 12 and 78). Finally, the system must return to basal levels of function after calcium availability is restored. The absence of this selfregulating feature would result in hypercalcemia.

D. Biochemical Changes Induced by Hypocalcemia As noted above, the immediate consequence of hypocalcemia is secretion of PTH. In addition to increasing serum calcium levels, PTH stimulates renal phosphate (Pi) excretion. The fall in serum phosphate may, however, be compensated by sufficient mobilization of phosphate (as well as calcium) from bone, so that circulating phosphate remains largely unchanged. The principle of mass action is thought to maintain the stability of the Ca × Pi ion product in the blood. As a consequence, local concentrations of the two major mineral components (Ca and Pi) of hydroxyapatite (HA) are able to influence the rate of movement in and out of the mineral phase of bone: [Ca] + [Pi]

[HA].

Thus, a fall in ionized Ca would favor an increase in serum phosphate concentration. Given all of these influences, sustained hypocalcemia usually results in a biochemical picture of secondary hyperparathyroidism, elevated 1,25(OH)2D levels, and variable changes in serum phosphate. If hypocalcemia develops in the setting of diminished or absent PTH function, serum phosphate is usually elevated, owing to increased renal phosphate reabsorption. In this instance, treatment with 1,25(OH)2D would also increase serum phosphate, since this metabolite enhances intestinal phosphate absorption. The effect of hypocalcemia on circulating vitamin D metabolites is complex. An increase in the biosynthesis of 1,25(OH)2D occurs, as reviewed above. This is largely secondary to the induced increase in

1052

THOMAS O. CARPENTER AND KARL L. INSOGNA

Vit D Milk

25(OH)D

1,25(OH)2D 1,25(OH)2D Ca Absorption

Low Ca Intake

[Ca]i

PTH

[PO4]

Bone mineralization

FIGURE 1 Mechanisms of the development of osteomalacia and rickets. Deficiency of vitamin D intake and/or limited ultraviolet light exposure lead to limited vitamin D stores as reflected by a decreased circulating 25OH D level. Reduced availability of this substrate is presumed to limit 1,25(OH)2D production, resulting in impaired intestinal calcium absorption. Calcium availability for skeletal mineralization is subsequently compromised, and secondary hyperparathyroidism, with concomitant hypophosphatemia occur. Restricted dietary calcium intake can also result in a similar pathophysiology. Increased turnover of vitamin D in the calcium deficient state may result in a greater risk of vitamin D insufficiency. The paradox of elevated levels of 1,25 dihydroxyvitamin D in these disorders is well recognized.

circulating PTH, but can be a direct consequence of the fall in calcium ion concentration. It has been determined that calcium deprivation results in a general increase in turnover of the parent vitamin D metabolite, 25OHD, such that vitamin D stores are depleted at a more rapid rate than normal [36]. The clinical implication of this finding is that susceptibility to vitamin D deficiency may be greater when concomitant calcium deprivation is present (see Fig. 1).

PTH resistance as a consequence of PTH/PTH-related peptide (PTHrP) receptor or postreceptor defects, or (3) in the setting of normal or increased PTH activity and normal PTH receptor function. Thus, although many of the etiologies relate to abnormalities in PTH, they are very relevant to this book because vitamin D metabolism is always involved and vitamin D is the cornerstone of therapy.

II. DIFFERENTIAL DIAGNOSIS OF HYPOCALCEMIA

A variety of congenital or acquired disorders can lead to developmental failure of the parathyroid glands, failure of functional hormone production, or destruction of the glands. These are all present as hypocalcemia, usually with attendant hyperphosphatemia and undetectable or inappropriately low levels of circulating PTH. a. Failure of Organogenesis: DiGeorge Sequence DiGeorge sequence is an uncommon developmental

A. Classification A rapid increase in PTH serves as the major defense against acute hypocalcemia. We therefore classify these disorders as those which manifest hypocalcemia (1) due to abnormalities of PTH availability, (2) due to

1. HYPOCALCEMIA DUE TO ABNORMALITIES OF PTH AVAILABILITY

CHAPTER 64 The Hypocalcemic Disorders: Differential Diagnosis and Therapeutic Use of Vitamin D

disorder that affects the third and fourth branchial clefts and results in dysgenesis of the thymus and the parathyroid glands [37]. Tetany and seizures are common features of the early course of infants with DiGeorge sequence. However, abnormalities in T-cell function with subsequent increased risk for infection often become a major feature of this disorder later in life [38]. The gene(s) responsible for the cardiac, thymic, and parathyroid features of the DiGeorge phenotype is TBX1 [39]. Microdeletions of chromosome 22 are found in many patients with this disorder. b. Idiopathic Hypoparathyroidism As discussed in Sections II.A.1.c-e, a variety of causes of inherited idiopathic hypoparathyroidism have now been delineated, including one family where the abnormality maps to the X chromosome. Linkage analysis has identified the Xq26-X27 region as the probable site for the molecular abnormality, which presents with failure of parathyroid gland development [40]. c. Molecular Abnormalities in the PTH Gene Parathyroid hormone is secreted and synthesized by a classic secretory pathway. The initial translation product is a prepropeptide, which requires cleavage of the amino-terminal pre and pro sequences before secretion. A family with autosomal dominant inheritance of hypoparathyroidism has been reported in which a missense mutation (Cys-18 → Arg) results in an abnormal signal sequence and diminished uptake of preproPTH into the endoplasmic reticulum [41]. Another family with recessively inherited hypoparathyroidism has been reported in which the prepro sequence is deleted by a splicing mutation [42]. d. Molecular Abnormalities in the Calcium-Sensing Receptor Gene The gene for the calcium-sensing receptor has been mapped to chromosome 3 [43]. As mentioned in Section I.B, ionized calcium is a ligand for this receptor, and receptor occupancy suppresses PTH secretion (see Chapter 31). Numerous individuals and families with a variety of activating mutations in this receptor have now been reported; the associated condition is referred to as autosomal dominant familial hypocalcemia [44]. One such kindred with a peculiar predisposition to nephrocalcinosis and renal insufficiency has been described [45]. e. Autoimmune Polyglandular Syndrome Type 1 An autoimmune disorder termed autoimmune polyglandular syndrome type 1 is characterized by early development of hypoparathyroidism in association with Addison’s disease and mucocutaneous candidiasis. The majority of affected individuals will manifest hypocalcemia by the age of 10 [46]. In addition to Addison’s disease, one-third of the patients will develop other endocrine disorders, diabetes mellitus, pernicious anemia, or premature ovarian failure [46]. This disorder is now

1053

known to be represented by mutations in a gene encoding (AIRE) an autoimmune regulatory protein containing a zinc finger motif, and is a candidate transcription factor [47]. f. Postsurgery Reduction in PTH Given the close anatomic relationship of the parathyroid glands to the thyroid, complete or near complete extirpation of the thyroid gland as part of the management of either Graves’ disease or thyroid cancer can be complicated by destruction or vascular compromise of parathyroid tissue and varying degrees of hypoparathyroidism. This should be a rare complication of thyroid surgery, and with experienced surgeons occurs with a frequency less than 10%. Even when destruction of the parathyroid glands does not occur following neck surgery, so-called stunned parathyroids with transient declines of approximately 1 mg/dl in total serum calcium are often observed in the first 24 to 48 hr postoperatively. This is presumably due to transient vascular or mechanical damage to the glands. Considerable variability in the degree of hypoparathyroidism following neck surgery occurs, ranging from asymptomatic reduction in parathyroid reserve to frank tetany, requiring chronic therapy with vitamin D (calcitriol) and calcium. g. Infiltrative Diseases and Deposition of Heavy Metals Although uncommon, malignant metastasis to the parathyroid glands with hypoparathyroidism has been reported, usually with breast cancer [48]. It has been postulated that granulomatous involvement of the parathyroids in sarcoidosis can lead to hypoparathyroidism [49]. Patients with transfusion-dependent thalassemia can develop hypoparathyroidism due to hemochromatosis secondary to deposition of iron in the glands [50]. In Wilson’s disease hypoparathyroidism can occur, presumably because of copper deposition [51]. Finally, impaired parathyroid reserve has been reported in diabetic patients with uremia [52]. h. Radiation Although the parathyroid glands are quite resistant to radiation, hypoparathyroidism following radioactive iodine treatment for hyperthyroidism has been described [53]. i. Functional Defects in PTH Secretion Magnesium is an important cofactor for parathyroid hormone secretion, apparently required for release of the stored hormone from secretory granules. In severe cases of hypomagnesemia, usually with circulating levels below 1 mg/dl, suppressed parathyroid secretion can occur [54]. This can be seen in the settings of chronic gastrointestinal disease, nutritional deficiency especially in alcoholics, or therapy with cis-platinum. Resistance to the action of PTH at the level of bone and kidney may also contribute to the hypocalcemia seen in the setting of magnesium deficiency. Replenishment of magnesium stores promptly restores parathyroid function to normal.

1054 Transient hypocalcemia in neonates has been reported to be associated with maternal hyperparathyroidism. 2. HYPOCALCEMIA DUE TO RESISTANCE ACTIONS OF PTH

TO THE

Several disorders of PTH action have hypocalcemia as their principal manifestation. a. Pseudohypoparathyroidism Peripheral tissue insensitivity or resistance to PTH is classically termed pseudohypoparathyroidism (PHP) [55]. The characteristic biochemical manifestations of PHP are hypocalcemia and hyperphosphatemia, as in hypoparathyroidism; however, circulating levels of PTH are elevated, rather than low or undetectable. The renal tubule is the primary site of PTH resistance, although variable degrees of skeletal resistance, depending on treatment status, have also been reported [56]. However, if the skeletal response is unimpaired, lesions characteristic of hyperparathyroidism, including osteitis fibrosa cystica, can develop. PTH stimulates renal cAMP production, and levels of cAMP increase in the urine following administration of the hormone. A direct correlation has been demonstrated between the degree of PTH resistance (as assessed by the magnitude of the change in cAMP excretion or renal phosphate threshold) and the ambient circulating PTH level [57]. The renal cAMP response is the basis of a diagnostic test that allows partial classification of this heterogeneous group of disorders. Individuals with PHP that demonstrate a blunted urinary cAMP response have PHP type I. Those that generate a normal cAMP response have PHP type II. PHP type I has been further characterized into types Ia and Ib. Type Ia describes those individuals with the Albright’s hereditary osteodystrophy (AHO) phenotype, which includes short stature and large frame, broad facies, and shortened fourth metacarpals. Soft tissue calcifications and multiple endocrine abnormalities are often present. These individuals often have a mutation in the α subunit of the stimulatory guanine nucleotide binding regulatory protein, Gs [58]. This regulatory protein couples membrane receptors to adenylate cyclase, thereby regulating receptor-dependent cAMP production. The presence of Gs in various cell types accounts for the generalized hormone resistance that may occur. For example, affected patients often have elevated thyrotropin (TSH) levels with a compensated euthyroid state. Variable degrees of gonadotropin, antidiuretic hormone (ADH), adrenocorticotropin (ACTH), and glucagon resistance have been described. Type Ib PHP is manifest primarily by PTH resistance, and the AHO phenotype is not present. It has been speculated

THOMAS O. CARPENTER AND KARL L. INSOGNA

that this phenotype may be due to tissue specific imprinting of the Gs alpha gene, or tissue specific splice variants that variably alter the function of the protein in different tissues [59]. A diagnosis of type II PHP appears to describe a variety of defects distal to cAMP generation in the cascade of hormone action. There is no distinct phenotype, although various autoimmune findings have been described in some patients. Finally, others have suggested that a circulating PTH inhibitor may play a role in the pathogenesis of PHP and have suggested that this inhibitor may be generated by parathyroid tissue itself [60]. Resistance to PTH has also been described in hypomagnesemia, as described below. b. Hypomagnesemia Magnesium deficiency can interfere with parathyroid secretion and function [54]. Serum magnesium levels are usually moderately to severely depressed (below the range of 1.0–1.4 mg/dl) before this occurs. Despite hypocalcemia, PTH levels may be inappropriately low or only modestly elevated, and tetany refractory to calcium supplementation can ensue. Hypomagnesemia, per se, may cause tetanic symptoms, although concomitant hypocalcemia is more common. Insufficient PTH secretion is the most widely accepted cause of refractory hypocalcemia in magnesium deficiency [54], although it has been suggested that resistance to the calcemic actions of PTH and vitamin D may play a role as well [61]. Impairment of vitamin D synthesis may also be at work [62]. As PTH stimulates conversion of 25OHD to 1,25(OH)2D, hypoparathyroidism may result in low circulating 1,25(OH)2D levels, further compromising the body’s defense against hypocalcemia [63]. Whether target tissue resistance to infused PTH occurs in magnesium deficiency remains controversial, and it has been suggested that this apparent resistance may simply reflect differences in the basal levels of circulating PTH [64]. To further complicate matters, generalized malnutrition including vitamin D deficiency is often present in hypomagnesemic patients [65]. Hypomagnesemia has been associated with alcohol abuse and may result from inherited disorders of magnesium excretion and/or absorption [66]. It can also be induced by the renal tubular effects of several drugs, including amphotericin B, aminoglycoside antibiotics, chemotherapeutic agents (particularly cis-platinum), diuretics, and cyclosporin. 3. HYPOCALCEMIA IN THE SETTING OF NORMAL OR INCREASED PTH ACTIVITY AND NORMAL PTH RECEPTOR FUNCTION

Despite normal PTH function and downstream signaling from its receptor, hypocalcemia can still occur

CHAPTER 64 The Hypocalcemic Disorders: Differential Diagnosis and Therapeutic Use of Vitamin D

due to disturbances in skeletal homeostasis, vitamin D metabolism, and a variety of medical illnesses. a. Neonatal Hypocalcemia The newborn infant undergoes an acute transition to independently regulated mineral homeostasis at parturition (see also Chapter 48). The maternal source of calcium is eliminated, and the infant’s circulating calcium level transiently decreases, with recovery occurring by the third day post-partum. Infants of diabetic or preeclamptic mothers and infants who suffer perinatal asphyxia or other fetal complications may experience an exaggerated fall in serum calcium with a delayed recovery phase. Management with intravenous calcium supplementation is required in the event of symptoms or severely low serum calcium levels. This condition is referred to as early neonatal hypocalcemia, and is usually transient. It may be associated with transient hypomagnesemia. Hypocalcemia presenting at 5–10 days of life is referred to as late neonatal hypocalcemia. This presentation is more typical of term infants after enteral feeding has been established. Infants of hyperparathyroid mothers may present with symptomatic hypocalcemia within this period, but have also been reported to present as late as 1 year of life. Familial forms of hypoparathyroidism may present as either “early” or “late” hypocalcemia. Mild to moderate neonatal hypocalcemia commonly occurs in patients with congenital heart disease (apart from those defects common in the DiGeorge sequence) [67] and in some cases can be attributed to transient impairment of parathyroid function. b. Hypocalcemia Due to Vitamin D Malnutrition (see Fig. 1). Vitamin D synthesis in the skin requires adequate exposure to ultraviolet light. Thus, vitamin D deficiency is uncommon in settings where sunlight exposure is abundant. In extremes of latitude (e.g., northern climates in North America), and where industrial pollution can interfere with transmission of UV light, normal vitamin D status is dependent on adequate dietary vitamin D intake. Supplementation of milk products with vitamin D has significantly reduced the incidence of vitamin D deficiency in North America. Despite these measures, certain populations are at risk for development of vitamin D deficiency, and severe hypocalcemia may be a presenting manifestation of the disorder (see also Chapters 47, 61, 62, and 65). A convergence of various risk factors for development of vitamin D deficiency occurs in breast-fed infants in the first 18 months of life. Presentation appears to be most common during the winter or early spring in northern U.S. cities. The limited direct sunlight exposure during the winter season is a major factor. Black children are at greater risk due to the greater quanta

1055

of UV light required to penetrate melanin in the pigmented dermis and induce previtamin D formation [68]. Breast milk contains only small amounts of vitamin D, even when the mother is receiving pharmacological doses of the vitamin (Chapter 51). Others have pointed out that dietary practices, including vegetarianism and high grain intake, may place infants at greater risk for the development of this condition [69] (Chapter 47). Another group at high risk for development of vitamin D deficiency is found at the other extreme of life, the elderly, because of general nutritional compromise and limited sunlight exposure (see Chapters 50 and 66). Biochemical findings in these conditions vary with the severity or duration of deficiency. Most agree that serum calcium levels in moderate to severe vitamin D deficiency are often normal, compensated by secondary elevations in PTH [70]. In severe vitamin D deficiency, however, overt hypocalcemia is usually manifest, despite elevated circulating PTH. Serum phosphate levels tend to be slightly low or normal. Circulating alkaline phosphatase activity of bone origin is usually markedly elevated in children and can be elevated in adults. The best available test to assess total body vitamin D status is the level of circulating 25OHD (see Chapter 58). Levels of this metabolite are low in vitamin D deficiency but may rise to normal with recent ingestion of vitamin D or significant sunlight exposure, whereas bone symptoms such as pain, and signs such as leg bowing, persist. In children, radiographs of rachitic extremities at the time of sampling may reveal hyperdense lines of remineralization at the physes, consistent with recent exposure to vitamin D, despite the presence of overt physical findings (see Chapter 60). Circulating 1,25(OH)2D levels may be low, normal, or elevated during vitamin D deficiency. This may appear paradoxical, but it should be recognized that 1,25(OH)2D circulates in 1000-fold lower concentrations than 25OHD. Furthermore, in the setting of vitamin D deficiency, production of 1,25(OH)2D is maximized. Thus, efficient conversion of small amounts of newly ingested or synthesized 25OHD may markedly increase the circulating 1,25(OH)2D concentration. Perhaps a more intriguing paradox in this setting is the continued malabsorption of calcium despite normal concentrations of 1,25(OH)2D. The skeletal consequence of isolated severe vitamin D deficiency in children is rickets, a disorder of the epiphyseal growth plate. The defective mineralization processes ultimately result in malalignment deformities of the long bones. In adult bone, vitamin D deficiency causes osteomalacia, which is characterized histomorphometrically by excess undermineralized

1056 osteoid and a markedly delayed mineralization rate (see Chapter 63). Adults with osteomalacia may suffer painful pseudofractures, particularly in weight-bearing long bones. c. Hypocalcemia Due to Vitamin D Malabsorption Because vitamin D is a fat-soluble vitamin, generalized fat malabsorption may result in vitamin D deficiency. Gastrointestinal diseases such as Crohn’s disease, celiac sprue, and pancreatic insufficiency can be accompanied by hypocalcemia due to vitamin D malabsorption [71] (see also Chapter 75). We have also encountered children presenting with vitamin D deficiency rickets who have ultimately been diagnosed with cystic fibrosis and fat malabsorption. In addition, interruption of the enterohepatic circulation of both 25OHD and 1,25(OH)2D may lower body vitamin D stores. It is also possible that the diseased bowel may not be able to respond to 1,25(OH)2D. Mild hypocalcemia and secondary hyperparathyroidism is also seen in cholestatic liver diseases such as primary biliary cirrhosis [71]. Circulating levels of 25OHD are reduced in this setting owing to impaired hydroxylation of vitamin D in the liver and also because of intestinal malabsorption of vitamin D. d. Hypocalcemia Due to 1-Hydroxylase Deficiency Impaired metabolism of 25OHD to 1,25(OH)2D is an autosomal recessive condition, in which hypocalcemia and severe rachitic abnormalities occur [72] (see also Chapter 71). The disorder (also termed pseudo-vitamin D-deficiency rickets or vitamin D–dependent rickets, type 1) is inherited in an autosomal recessive manner and is characterized by biochemical features similar to those of vitamin D–deficiency rickets, with the exceptions that circulating 25OHD levels are normal and circulating 1,25(OH)2D levels are low. Mutations in the gene encoding the ferrodoxin-binding component of the mitochondrial P450 enzyme, 25-hydroxyvitamin D 1 alpha hydroxylase (CYP27B1) have been shown to cause this condition [73]. Restoration of eucalcemia and correction of rickets is attainable with physiological doses of 1,25(OH)2D3. e. Hypocalcemia Due to Hereditary Resistance to 1,25(OH)2D A defect in target tissue responsivity to 1,25(OH)2D was clinically described shortly after the capacity to measure circulating 1,25(OH)2D became available [32]. Patients with hypocalcemia caused by hereditary resistance to 1,25(OH)2D have severe manifestations of vitamin D–deficiency rickets; however, serum 25OHD concentrations are normal, and 1,25(OH)2D levels are usually elevated. This disorder is inherited in an autosomal recessive manner. Additional features in many patients include alopecia totalis and oligodontia. The disease is variably responsive to large doses of 1,25(OH)2D3 and oral calcium therapy. In the

THOMAS O. CARPENTER AND KARL L. INSOGNA

most resistant cases, however, long-term parenteral calcium infusions can normalize the serum chemistries and cure the skeletal lesions [74]. The positive therapeutic response to parenteral calcium suggests that mediation of calcium absorption at the intestine is the critical systemic action for 1,25(OH)2D. Several defects in the coding region of the vitamin D receptor (VDR) that impair or prevent either hormone or DNA binding have been described in these patients. Reduced expression of the VDR has also been described. This condition is quite rare but serves as an interesting experiment of nature in which the receptormediated function of 1,25(OH)2D3 is specifically ablated (see also Chapter 72). f. Hypocalcemia Due to Dietary Calcium Deficiency Although uncommon, extremely low calcium intakes have been reported to be associated with mild hypocalcemia. Nigerian and South African children with calcium intakes of 150 mg/day or less were found to have decreased serum calcium values, secondary hyperparathyroidism, and rickets [75–77]. The children were not vitamin D–deficient, and their biochemical abnormalities and bone disease responded to treatment with calcium alone. Our group has recently demonstrated that this disorder may be more common in North American infants than previously expected. A review of nutritional rickets in New Haven, Connecticut revealed that 50% of cases had normal circulating values of 25-OHD, and some were even on vitamin supplementation. Increasing dietary calcium resulted in radiographic and biochemical improvement. This phenomenon appears to occur after children have been weaned to diets with little to no dairy product content, with fluids consisting mostly of juices and soft drinks [78]. g. Hypocalcemia Induced by Hyperphosphatemia Since the 1930s, it has been appreciated that oral or parenteral phosphate can induce a decline in serum calcium concentrations. Herbert et al. have demonstrated that phosphate infusions lower serum calcium in both the presence and absence of parathyroid glands [79]. Moreover, they reported that the changes in peak urinary calcium excretion during phosphate administration are not sufficient to account for the fall in the serum calcium. The theory they advanced remains the best explanation available for this phenomenon and centers on the hypothesis that the calcium × phosphate molar product, when exceeded, leads to spontaneous precipitation of calcium salts in soft tissues. The Ca × P product, when estimated from total serum ion concentrations (as mg/dl), is normally taken to be <60 in adults or <80 in small children. Hyperphosphatemia sufficient to cause hypocalcemia is usually abrupt in onset and severe in magnitude. Typical clinical settings include (1) excessive enteral or parental phosphate administration, (2) the tumor lysis

CHAPTER 64 The Hypocalcemic Disorders: Differential Diagnosis and Therapeutic Use of Vitamin D

syndrome, and (3) rhabdomyolysis-induced acute renal failure. Hypocalcemia induced by either oral or parental phosphate administration is often associated with soft tissue calcification. Such ectopic calcification has been observed during the treatment of hypophosphatemia due to either diabetic ketoacidosis or acute alcoholism. Adults receiving phosphate-containing enemas and infants fed “humanized” cow milk rich in phosphate may also become hypocalcemic [80,81]. Under most circumstances discontinuation of exogenous phosphate intake leads to prompt return of the serum calcium level to normal. Hypocalcemia in the setting of massive tumor lysis results from the release of intracellular phosphate as a consequence of chemotherapy-induced cell death, usually during the treatment of rapidly proliferating neoplasms [82]. The hypocalcemia may continue beyond the period of hyperphosphatemia and appears to be aggravated by suppressed 1,25(OH)2D levels [83]. The use of phosphate binding antacids, oral calcium, and, in severe cases, 1,25(OH)2D3 may help to correct the serum calcium level. Rhabdomyolysis-induced acute renal failure occurs with trauma and drug or alcohol abuse. Marked hypocalcemia can occur in the early oliguric phase, and moderate to severe hypercalcemia in the subsequent polyuric phase. Llach et al. have described hyperphosphatemia and suppressed serum 1,25(OH)2D levels during the initial hypocalcemia, suggesting a mechanism similar to that seen in the tumor lysis syndrome [84]. The appearance of hypercalcemia and high serum 1,25(OH)2D levels during the diuretic phase may result from rapid development of secondary hyperparathyroidism during the initial hypocalcemic period. Treatment includes restriction of phosphate intake and efforts to prevent hypocalcemia during the early stages of the disease. h. Hypocalcemia Due to Accelerated Skeletal Mineralization Bone remodeling is a controlled process of tissue renewal which, in healthy individuals, results in closely matched rates of bone resorption and formation (see Chapter 28). If skeletal mineralization exceeds the rate of bone resorption, hypocalcemia can occur. One setting in which this can be observed is following surgical correction of primary or tertiary hyperparathyroidism. The abrupt cessation of PTH-mediated osteoclastic bone resorption with concomitant rapid remineralization of an undermineralized skeleton can lead to “hungry bone syndrome” with severe, even lifethreatening hypocalcemia [85]. Postoperative treatment should be instituted when the serum calcium level falls below 8.0 mg/dl, using oral or parenteral calcium supplements and if necessary 1,25(OH)2D3. In general, this condition resolves over the course of several days,

1057

although distinguishing it from permanent postoperative hypoparathyroidism can be difficult and requires gradual discontinuation of supportive therapy with careful monitoring. Hypocalcemia also may occur in patients with bony metastases that induce bone formation, as with prostatic and breast cancer [86]. Finally, institution of therapy for vitamin D deficiency osteomalacia or rickets can sometimes lead to a fall in serum calcium associated with rapid mineralization of previously unmineralized osteoid [87]. This is self-limited and can usually be prevented with supplemental calcium. i. Medical Illness Hypocalcemia in the setting of renal failure results from hyperphosphatemia due to reduced renal phosphate clearance by the failing kidney and is complicated by impaired biosynthesis of 1,25(OH)2D [88] (see Chapter 76). Hypocalcemia and tetany were first reported in patients with pancreatitis in the early 1940s [89]. Pancreatic lipase released from the damaged gland is believed to liberate free fatty acids that chelate calcium, thereby removing it from the extracellular fluid [90]. Hypomagnesemia resulting from poor oral intake, alcohol use, or vomiting may contribute to the hypocalcemia. Hypocalcemia in the setting of pancreatitis often suggests a poor clinical course. Treatment consists of parental calcium and magnesium when indicated. Hypocalcemia may also occur in patients with acute sepsis. In one series, 20% of such patients evidenced reductions in ionized serum calcium [91]. Hypocalcemia in this series was associated with a poor prognosis (50% mortality, compared to 30% in eucalcemic patients). This phenomenon is most often reported with gramnegative sepsis but has occurred in toxic shock syndrome caused by staphylococcal infection [92]. The pathophysiology of hypocalcemia in these two settings is unknown. Finally, it has been suggested that parathyroid gland reserve is subnormal in patients with AIDS, although hypocalcemia is not a prominent feature of that disorder [93]. j. Medications A variety of medications have been reported to decrease serum ionized calcium concentration. Many of these drugs are used to treat hypercalcemia and/or excessive bone resorption, and hypocalcemia results from their overzealous use. Thus, mithramycin, calcitonin, and the bisphosphonates can all cause hypocalcemia. In susceptible individuals, prolonged therapy with diphenylhydantoin or phenobarbital can lead to hypocalcemia, owing in part to enhanced catabolism of vitamin D metabolites [94] (Chapter 74). Citrated blood products, particularly when used for large volume transfusions or plasma plasmapheresis, can cause hypocalcemia [95]. Radiocontrast agents that contain EDTA (ethylenediaminetetraacetic acid) can

1058 also induce falls in serum ionized calcium levels [96]. Finally, foscarnet (trisodium phosphonoformate), used in the treatment of patients with AIDS, has been reported to cause a decline in ionized serum calcium, perhaps through complexing extracellular calcium [97].

III. THERAPY FOR HYPOCALCEMIA A. Acute Management 1. NEWBORNS

It may be necessary to treat early neonatal hypocalcemia when the circulating concentration of total serum is less than 5–6 mg/dl in premature infants, and less than 6–7 mg/dl in term infants. Appropriate emergency therapy of acute symptomatic hypocalcemia consists of a slow intravenous infusion (<1 ml/min) of calcium gluconate in a 10% (w/v) solution. The calcium gluconate salt consists of 9% elemental calcium. A well-functioning indwelling intravascular catheter should be used, to avoid extravasation. Calcium should never be administered intramuscularly because of local tissue toxicity. It is important to perform cardiac monitoring and careful observation during acute infusions. A total infusion of 1–3 ml will usually arrest convulsions, and no more than 2 mg of elemental calcium per kilogram body weight should be given as a single dose. Such bolus infusions may be repeated up to 4 times in a 24-hr period. If severe hypocalcemia persists, however, it is generally more effective to use a long-term calcium gluconate infusion, such that 20–50 mg of elemental calcium per kilogram body weight is infused over an entire 24-hr period. Calcium chloride is more irritating than calcium gluconate and is not the preferred salt for infusion. Neither bicarbonate nor phosphate should be coinfused with calcium in order to prevent precipitation of their respective calcium salts, either in the infusion line or in the vein. 2. ADULTS

In adults, emergency management consists of 10–20 ml of 10% calcium gluconate infused over a 10- to 15-min period. In the longer term one can dilute 10 ampules of calcium gluconate in 1 liter of 5% dextrose and, beginning at a rate of 50 ml/hr, titrate the rate to maintain the serum calcium in the low normal range. Finally, in the setting of acute exacerbations of calcium malabsorption, as may typically occur in patients with autoimmune hypoparathyroidism with associated gastrointestinal disorders, nocturnal nasogastric supplementation with calcium carbonate or calcium gluconate has been employed, providing up to 20 mg of elemental calcium per kilogram body weight

THOMAS O. CARPENTER AND KARL L. INSOGNA

per 8 hr, as necessary, until the underlying intestinal disturbance has resolved. 3. ROLE OF MAGNESIUM SUPPLEMENTATION

In the setting of hypomagnesemia, magnesium therapy may be required to restore PTH secretion and peripheral activity. Prior to administration of magnesium salts, assessment of renal function and urinary output should be performed. Magnesium treatment in infancy consists of 5–10 mg of elemental magnesium (Mg) per kilogram body weight. Although magnesium may be given intramuscularly, the intravenous route is preferred. Magnesium sulfate septahydrate (MgSO4 • 7H2O) is available as a 50% solution, containing 48 Mg/ml of elemental magnesium. These small volumes may be further diluted, but they should be infused slowly; the dose may be repeated every 12–24 hr. In older individuals, up to 2.4 mg of elemental Mg per kilogram body weight can be given over a 10–min period (to a maximum of 180 mg). Others prefer a continuous infusion of 576 mg of elemental magnesium over 24 hr. The length of therapy must be individualized, and maintenance with oral magnesium salts should be implemented in cases where ongoing hypomagnesemia is anticipated. Magnesium levels should be monitored to avoid toxicity. Deep tendon reflexes can be examined, and therapy should be halted if they diminish. As with calcium therapy, cardiac monitoring should be performed and therapy stopped if EKG changes occur. Intravenous calcium gluconate is a useful antidote for magnesium intoxication, and should be available at the bedside.

B. Long-term Treatment Many of the causes of hypocalcemia discussed previously are corrected by treating the underlying disorder (e.g., vitamin D deficiency, tumor lysis syndrome). Relatively few of these disorders require maintenance therapy for hypocalcemia, and of those, the most important are hypoparathyroidism and pseudohypoparathyroidism. Vitamin D–resistant states, although rare, comprise a third group of patients that require long-term treatment. Hypoparathyroidism, whether primary or secondary to trauma or surgery, is the most frequently encountered condition that requires chronic therapy to maintain eucalcemia. In these individuals, the goal is to maintain serum calcium in the low normal range (8.5 to 9.2 mg/dl as measured by atomic absorption spectrophotometry). This will reduce the likelihood of symptoms such as circumoral tingling, signs such as carpopedal spasm, as

CHAPTER 64 The Hypocalcemic Disorders: Differential Diagnosis and Therapeutic Use of Vitamin D

well as more long-term complications such as cataracts. Long-term treatment with PTH is not yet practical although short-term therapy has been successful [98]. There is no single best way to achieve stable eucalcemia, although the combination of a vitamin D metabolite with calcium supplements is generally preferred. A wide variety of preparations of both are available (see Tables I and II). Because of the prolonged toxicity that occurs with excessive ingestion of either ergocalciferol (vitamin D2) or 25OHD, we generally prefer to use rapid acting preparations of vitamin D for the treatment of this disorder. Toxicity, when it occurs with these preparations, corrects more rapidly with discontinuation of the drug. Calcitriol [1,25(OH)2D3] and dihydrotachysterol (DHT) are two preparations suited to this purpose. Both are fully active in vivo. In general, it is best to aim for a stable dose of one of these agents and to further regulate the serum calcium by adjusting the intake of supplemental calcium, rather than by making repeated changes in vitamin D metabolite therapy. We use calcitriol in the majority of

1059

our patients. The dose of calcitriol can range from as little as 0.25 up to 2.0 ug/day [99,100]. We have estimated the biological half-life of the drug as 12–14 hr. Hypercalcemia, when it develops during therapy with calcitriol, usually resolves within 3–4 days after discontinuing the drug, although we have had patients in whom it has taken over 1 week for serum calcium to normalize. In addition to a high calcium diet, calcium supplements are important for the treatment of hypoparathyroidism. Doses of 1000–2000 mg/day of calcium may be necessary. Preparations including the carbonate, citrate, lactate, gluconate, and glucobionate salts are suitable for this purpose (Table I). We prefer calcium carbonate because it is inexpensive, well-tolerated, and easily acquired. In some cases of hypoparathyroidism, a thiazide diuretic may be useful in augmenting serum calcium levels and reducing the hypercalciuria that can occur with the institution of treatment. Several vitamin D analogs have been developed for the treatment of secondary hyperparathyroidism in the setting of renal failure

TABLE I Oral Calcium Preparationsa Drug Calcium Carbonate Calcium carbonate (available in generic brands)

Os-Cal 500® Caltrate 600® Tums® (Regular, EX, Ultra) Alka-Mints® Viactiv® Calcium citrate Calcium citrate (available in generic brands)

Dosage form

Various formulations, including suspension, tablets, chewable tablets (depending on manufacturer) Tablet Tablet Chewable tablet Chewable tablet Chewable

Elemental calcium (mg/tablet)

Cost per 1000 mg of elemental calcium*

100 mg/ml (susp), 260 mg (tab), 500 mg (tab) (depends on manufacturer)

Cost will vary by manufacturer (generally lower than name brands)

500 mg 600 mg 200 mg, 300 mg, 400 mg 340 mg 500 mg

$0.18 $0.16 $0.30, $0.12, $0.18 $0.12 $0.26

260 mg, 500 mg (depends on manufacturer)

Cost will vary by manufacturer (generally lower than name brands)

Citracal®

Various formulations including tablets, effervescent tablets, oral suspension (depending on manufacturer) Tablet

200 mg

$0.13

Calcium glubionate Neo-Calglucon®

Syrup

115 mg/5 ml

$1.74

*Retail cost will vary between retail pharmacies. References –Lacy CF, Armstrong LL, Goldman MP, Lance LL. Drug Information Handbook. Lexi-Comp. 2003;11th ed. 220–23. –Kastrup EK ed. Drug Facts and Comparisons. St. Louis: Facts & Comparisons; August 2003. –Walgreens Pharmacy. www.walgreens.com. Accessed on July 31, 2003. a Adapted with permission from Carpenter T 1996 Rickets. In: Berg F, Inglefinger J, Wald E (eds) Gellis and Kagan’s Current Pediatric Therapy, 15th ed. Saunders, Philadelphia, pp. 363–367.

1060

THOMAS O. CARPENTER AND KARL L. INSOGNA

TABLE II

Vitamin D and Related Agentsa,b

Name

Formulation

Vitamin D (calciferol) Drisdol®

Solution: 8000 IU/ml Tablet: 25,000 IU 50,000 IU Dihydrotachysterol (DHT) (Hytakerol®) Solution: 0.2 mg/5ml Tablets: 0.125 mg 0.2 mg 0.4 mg 1,25 dihydroxyvitamin D (calcitriol) Rocaltrol® 0.25 µg capsule 0.50 µg capsule 1.0 µg/ml (oral solution) Calcijex® solution: ampules for IV use containing solutions with 1 or 2 µg/ml of drug 1 µg vitamin D = 40 IU

Typical dosec 2000 IU/day 1 tablet/day 1 tablet/day 0.5 mg/day 0.5 mg/day 0.5 mg/day 0.4 mg/day 0.5 µg/day 0.5 µg/day 0.5 µg/day

a Adapted with permission from Carpenter T 1996 Rickets. In: Berg F, Ingelfinger J, Wald E (eds) Gellis and Kagan’s Current Pediatric Therapy, 15th ed. Saunders, Philadelphia, pp. 363–367. b Daily dose may vary significantly, depending on condition.

(see Chapter 76) [101]. These analogs are relatively more selective inhibitors of parathyroid proliferation than calcitriol. The limited calcemic activity of these compounds renders them less useful in the management of hypocalcemia. Magnesium deficiency can occur in patients with hypoparathyroidism, most often secondary to steatorrhea, which is seen in the autoimmune forms of this disorder [102,103]. This may render a patient relatively resistant to therapy, and therefore magnesium deficiency should be considered in individuals whose therapeutic requirements unexpectedly increase. In pseudohypoparathyroidism, the therapeutic approach is similar to that in primary hypoparathyroidism, the principal difference being that hypercalciuria is less of an issue, and it is generally easier to maintain eucalcemia in these individuals. Untreated patients with pseudohypoparathyroidism may have variable defects in mineralization and initially may require high dose therapy to achieve eucalcemia as their bones remineralize. Requirements will drop as the bone lesion heals, often heralded by a fall in serum alkaline phosphatase and a rise in serum calcium levels. Individuals with 1-hydroxylase deficiency have a defect in the ability to generate 1,25(OH)2D from the precursor metabolite 25OHD. In this disorder, eucalcemia can be achieved by supplying 1,25(OH)2D3 in physiological dosages [72]. In contrast, hereditary resistance to 1,25(OH)2D represents a spectrum of resistance to therapy, with some individuals responding to doses of calcitriol in the usual therapeutic range and others resistant to even massive doses of the drug [32]. As noted above, chronic therapy with parenteral infusions

of calcium have resulted in improvement of the rickets and normalization of all serum biochemical parameters (see Chapter 72) [74]. Although as discussed above, recently developed analogs of vitamin D have not been generally recommended for therapy of hypocalcemia, two novel analogs (20-Epi-1,25(OH)2D3 and JK-1626-2) have specifically been found to be efficacious in the treatment of cases of hereditary resistance to vitamin D caused by mutations in the ligand-binding domain of VDR [104]. A variety of stresses such as trauma, infection, and pregnancy can increase the therapeutic requirements of patients with chronic hypocalcemia, and the clinician should be alert to this possibility.

Acknowledgments Dr. Carpenter is supported by a grant from the National Institutes of Health (HD1288). Dr. Insogna is supported by grants from the NIH (AR39571) and both are supported an NIH Core Center Grant (AR 46032).

References 1. Harrison H, Harrison H 1979 Hypocalcemia states. In: Disorders of Calcium and Phosphate Metabolism in Childhood and Adolescence. Saunders, Philadelphia, Pennsylvania, pp. 47–99. 2. Sharief N, Matthew DJ, Dillon MJ 1991 Hypocalcaemic stridor in children. How often is it missed? Clin Pediatr 30:51–52.

CHAPTER 64 The Hypocalcemic Disorders: Differential Diagnosis and Therapeutic Use of Vitamin D

3. Wong C, Lau C, Cheng C, Leung W, Freedman B 1990 Hypocalcemic myocardial dysfunction: Short- and long-term improvement with calcium replacement. Am Heart J 120:381–386. 4. Alpan G, Glick B, Peleg O, Eyal F 1991 Pseudotumor cerebri and coma in vitamin D–dependent rickets. Clin Pediatr 20:254–256. 5. Carpenter TO, Insogna KL, Boulware SD, Mitnick MA 1990 Vitamin D metabolism in chronic childhood hypoparathyroidism: Evidence for a direct regulatory effect of calcium. J Pediatr 116:252–257. 6. Nikiforuk G, Fraser D 1979 Etiology of enamel hypoplasia and interglobular dentin: The roles of hypocalcemia and hypophosphatemia. Metab Bone Dis Related Res 2:17–23. 7. Ilium F, Dupont E 1985 Prevalence of CT-detected calcification in the basal ganglia in idiopathic hypoparathyroidism and pseudo-hypoparathyroidism. Neuroradiology 27:32–37. 8. Stewart AF, Battaglini-Sabetta J, Milstone LM 1984 Hypocalcemia-induced psoriasis of von Zumbusch: New experience with an old syndrome. Ann Intern Med 100:677–680. 9. Brown EM 1991 Extracellular Ca2+ sensing, regulation of parathyroid cell function, and role of Ca2+ and other ions as extracellular (first) messengers. Physiol Rev 71:371–411. 10. Ladenson J, Lewis J, Boyd J 1978 Failure of total calcium corrected for protein, albumin, and pH to correctly assess free calcium status. J Clin Endocrinol Metab 46:986–993. 11. Chattopadhyay N, Mithal A, Brown EM 1996 The calciumsensing receptor: A window into the physiology and pathophysiology of mineral ion metabolism. Endocr Rev 17:289–307. 12. Juppner H, Abou-Samra AB, Freeman MV, Kong X, Schipani E, Richards J, Kolakawski L, Hock J, Kronenberg H, Serge G 1991 G protein-linked receptor for parathyroid hormone and parathyroid hormone-related peptide. Science 254:1024–1026. 13. Livesey SA, Kemp BE, Re C, Partridge NC, Martin T 1982 Selective hormonal activation of cyclic AMP-dependent protein-kinase isoenzymes in normal and malignant osteoblasts. J Biol Chem 257:14983–14987. 14. Hruska KA, Moskowitz D, Esbrit P, Civitelli R, Westbrook S, Huskey M 1987 Stimulation of inositol triphosphate and diacyl-glycerol production in renal tubular cells by parathyroid hormone. J Clin Invest 79:230–239. 15. Abou-Samra A, Juppner H, Force T, Freeman M, Knog X, Schipani E, Urena P, Richards J, Bonventre J, Potts J, Kronenberg H, Segre G 1992 Expression cloning of a common receptor for parathyroid hormone and parathyroid hormone-related peptide from rat osteoblast-like cells: A single receptor stimulates intracellular accumulation of both cAMP and inositol triphosphates and increases intracellular free calcium. Proc Natl Acad Sci USA 89:2732–2736. 16. Lee K, Deeds D, Bond A, Juppner H, Abou-Samra A-B, Segre G 1993 In situ localization of PTH/PTHrP receptor mRNA in the bone of fetal and young rats. Bone 14:341–345. 17. Rouleau M, Mitchell J, Goltzman D 1990 Characterization of the major parathyroid hormone target cell in the endosteal metaphysis of rat long bones. J Bone Miner Res 5:1043–1053. 18. McSheehy P, Chambers T 1986 Osteoblast-like cells in the presence of parathyroid hormone release soluble factor that stimulates osteoclastic bone resorption. Endocrinology 119:1654–1659. 19. Talmage R, Doppelt S, Fondren F 1976 An interpretation of acute changes in plasma 45Ca following parathyroid hormone administration to thyroparathyroidectomized rats. Calcif Tissue Res 22:117–128.

1061

20. King G, Holtrop M, Raisz L 1978 The relation of ultrastructural changes in osteoclasts to resorption in bone cultures stimulated with parathyroid hormone. Metab Bone Dis Related Res 1:67–74. 21. Friedman PA, Gesek FA 1993 Calcium transport in renal epithelial cells. Am J Physiol (Renal, Fluid Electrolyte Physiol) 33:F181–F198. 22. Letscher M, Kaissling B, Biber J, Murer H, Kempson ST, Levi M 1996 Regulation of rat renal Na/Pi cotransporter by parathyroid hormone: Immunohistochemistry. Kidney Int 49:1010–1018. 23. Minghetti PP, Norman AW 1988 l,25(OH)2-Vitamin D3 receptors: Gene regulation and genetic circuitry. FASEB J 2:3043–3053. 24. Armbrecht J, Wongsurawat N, Zenser TV, Davis BB 1982 Differential effects of parathyroid hormone on the renal 25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3 production of young and adult rats. Endocrinology 111: 1339–1344. 25. Walker AT, Stewart AF, Korn EA, Shiratori T, Mitnick MA, Carpenter TO 1989 Effect of parathyroid hormone-like peptides on 25-hydroxyvitamin D-1a-hydroxylase activity in rodents. Am J Physiol 258:E297–E303. 26. Booth BE, Tsai HC, Morris C 1977 Parathyroidectomy reduces 25-hydroxyvitamin D3 1a-hydroxylase activity in the hypocalcemic vitamin D–deficient chick. J Clin Invest 60:1314–1320. 27. Siegel N, Wongsurawat N, Armbrecht J 1986 Parathyroid hormone stimulates dephosphorylation of the renoredoxin component of the 25-hydroxyvitamin D3-1a-hydroxylase from rat renal cortex. J Biol Chem 261:16998–17003. 28. Matsumoto T, Ikeda K, Morita K, Fukumoto S, Takahashi H, Ogata E 1987 Blood Ca2+ modulates responsiveness of renal 25(OH)D3-1a-hydroxylase to PTH in rats. Am J Physiol (Endocrinol Metab 16) 253:E503–E507. 29. Bushinsky DA, Riera GS, Favus MJ, Coe FL 1985 Evidence that blood ionized calcium can regulate serum 1,25(OH)2D3 independently of parathyroid hormone and phosphorus in the rat. J Clin Invest 76:1599–1604. 30. Kumar R 1991 Vitamin D and calcium transport. Kidney Int 40:1177–1189. 31. Norman AW, Song X, Zanello L, Bula C, Okamura WH 1999 Rapid and genomic biological responses are mediated by different shapes of the agonist steroid hormone, 1alpha,25(OH)2vitamin D3. Steroids 64:120–128. 32. Liberman UA, Marx SJ 1996 Vitamin D–dependent rickets. In: Favus M (ed) Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism, 3rd Ed. Raven, New York, pp. 311–316. 33. Favus MJ, Mangelsdorf DJ, Tembe V, Coe BJ, Haussler MR 1988 Evidence for in vivo upregulation of the intestinal vitamin D receptor during dietary calcium restriction in the rat. J Clin Invest 82:218–224. 34. van Leeuwen JPTM, Birkenhager JC, Buurman CJ, Semite JP, Pols HAP 1990 Functional involvement of calcium in the homologous up-regulation of the 1,25-dihydroxyvitamin D3 receptor in osteoblast-like cells. FEBS Lett 270: 165–167. 35. Sandgren ME, DeLuca HF 1990 Serum calcium and vitamin D regulate 1,25-dihydroxyvitamin D3 receptor concentration in rat kidney in vivo. Proc Natl Acad Sci USA 87: 4312–4314. 36. Clements MR, Johnson L, Fraser DR 1987 A new mechanism for induced vitamin D deficiency in calcium deprivation. Nature 325:62–65.

1062 37. Goltzman D, Cole D 1996 Hypoparathyroidism. In: Favus M (ed) Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism, 3rd Ed. Raven, New York, pp. 220–223. 38. Conley M, Beckwith J, Mancer J, Tenckhoff L 1979 The spectrum of DiGeorge syndrome. J Pediatr 94:883–890. 39. Yagi H, Furutani Y, Hamada, Sasaki T, Asakawa S, Minoshima S, Ichida F, Joo K, Kimura M, Imamura S, Kamatani N, Momma K, Takao A, Nakazawa M, Shimizu N, Matsuoka R 2003 Role of TBX1 in human del22q11.2 syndrome. Lancet 362:1366–1373. 40. Thakker R, Davies K, Whyte M, Wooding C, O’Riordan J 1990 Mapping the gene causing X-linked recessive idiopathic hypoparathyroidism to Xq26-Xq27 by linkage studies. J Clin Invest 86:40–45. 41. Arnold A, Horst S, Gardella T, Baba H, Levine M, Kronenberg H 1990 Mutation of the signal peptide-encoding region of the preproparathyroid hormone gene in familial hypoparathyroidism. J Clin Invest 86:1084–1087. 42. Parkinson D, Thakker R 1992 A donor splice site mutation in the parathyroid gene is associated with autosomal recessive hypoparathyroidism. Nature Genet 1:149–152. 43. Chou Y, Brown E, Levi T, Crowe G, Atkinson A, Arnqvist H, Toss G, Fuleihan G, Seidman J, Seidman C 1992 The gene responsible for familial hypocalciuric hypercalcemia maps to chromosome 3 in four unrelated families. Nature Genet 1:295–300. 44. Pollak MR, Brown EM, Step L, McLaine PN, Kifor O, Park J, Hebert SC, Seidman CE, Seidman JG 1994 Autosomal dominant hypocalcaemia caused by a Ca2+-sensing receptor gene mutation. Nature Genet 8:303–307. 45. Pearce S, Williamson C, Kifor O, Bai M, Coulthard M, Davies M, Lewis-Earned N, McCredie D, Powell H, KendallTaylor P, Brown E, Thakker R 1996 A familial syndrome of hypocalcemia with hypercalciuria due to mutations in the calcium-sensing receptor. N Engl J Med 335:1115–1122. 46. Neufeld M, MacLaren N, Blizzard R 1981 Two types of autoimmune Addison’s disease associated with different polyglandular autoimmune (PGA) syndromes. Medicine 60:355–362. 47. Nagamine K, Peterson P, Scott HS, Kudoh J, Minoshima S, Heino M, Krohn KJ, Lalioti MD, Mullis PE, Antonarakis SE, Kawasaki K, Asakawa S, Ito F, Shimizu N 1997 Positional cloning of the APECED gene. Nature Genetics. 17:393–398. 48. Horwitz C, Myers W, Foote F 1972 Secondary malignant tumors of the parathyroid glands. Report of two cases with associated hypoparathyroidism. Am J Med 52:797–808. 49. Dill J 1983 Hypoparathyroidism in sarcoidosis. South Med J 76:414. 50. Brezis M, Shalev O, Leibel B, Bernheim J, Ben-Ishay D 1980 Phosphorus retention and hypoparathyroidism associated with transfusional iron overload in thalassaemia. Miner Electrolyte Metab 4:57. 51. Carpenter TO, Carnes DL Jr, Anast CS 1983 Hypoparathyroidism in Wilson’s disease. N Engl J Med 309:873–877. 52. Heidbreder E, Gotz R, Schafferhans K, Heidland A 1986 Diminished parathyroid gland responsiveness to hypocalcemia in diabetic patients with uremia. Nephron 42:285–289. 53. Burch WM, Posillico JT 1983 Hypoparathyroidism after 131I therapy with subsequent return of parathyroid function. J Clin Endocrinol Metab 57:398–401. 54. Anast C, Mohs J, Kalpan S, Burns P 1972 Evidence for parathyroid failure in magnesium deficiency. Science 177:606–608. 55. Albright F, Burnett CH, Smith PH, Parson W 1942 Pseudohypoparathyroidism. An example of “Seabright-Bantam syndrome.” Endocrinology 30:922–932.

THOMAS O. CARPENTER AND KARL L. INSOGNA

56. Kruse K, Kracht U, Wohlfart K, Kruse U 1989 Biochemical markers of bone turnover, intact serum parathyroid hormone and renal calcium excretion in patients with pseudohypoparathyroidism and hypoparathyroidism before and during vitamin D treatment. Eur J Pediatr 148:535–539. 57. Stone M, Hosking D, Garcia-Himmelstine C, White D, Rosenblum D, Worth H 1993 The renal response to exogenous parathyroid hormone in treated pseudohypoparathyroidism Bone 14:727–735. 58. Patten JL, Johns DR, Valle D, Eil C, Gruppuso P, Steele G, Smallwood PM, Levine MA 1990 Mutation in the gene encoding the stimulatory G protein of adenylate cyclase in Albright’s hereditary osteodystrophy. N Engl J Med 322:1412–1419. 59. Weinstein LS, Yu S, Warner DR, Liu J 2001 Endocrine manifestations of stimulatory G protein alpha-subunit mutations and the role of genomic imprinting. Endocrine Reviews 22:675–705. 60. Loveridge N, Tschopp FT, Born W, Devogelaer JP, de Deux-chaisnes CN, Fischer JA 1986 Separation of inhibitory activity from biologically active parathyroid hormone in patients with pseudohypoparathyroidism type I. Biochim Biophys Acta 889:117–122. 61. Rude RK, Oldham SB, Singer FR 1976 Functional hypoparathyroidism and parathyroid hormone end-organ resistance in human magnesium deficiency. Clin Endocrinol 5:209–224. 62. Carpenter T 1988 Disturbances of vitamin D metabolism and action during clinical and experimental magnesium deficiency. Magnesium Res 1:131–139. 63. Rude RK, Adams JS, Ryzen E, Endres DB, Miimi H, Horst RI, Haddad JG, Singer FR 1985 Low serum concentrations of 1,25-dihydroxyvitamin D in human magnesium deficiency. J Clin Endocrinol Metab 61:933–940. 64. Allgrove J, Adami S, Fraher L, Reuben A, O’Riordan JLH 1984 Hypomagnesaemia: Studies of parathyroid hormone secretion and function. Clin Endocrinol 21:435–449. 65. Fuss M, Bergmann P, Bergans A, Bagon J, Cogan E, Pepersack T, van Gossum M, Corvilain J 1989 Correction of low circulating levels of 1,25-dihydroxyvitamin D by 25-hydroxyvitamin D during reversal of hypomagnesaemia. Clin Endocrinol 31:31–38. 66. Anast CS, Gardner DW 1981 Magnesium metabolism. In Bronner F, Coburn J (eds) Disorders of Mineral Metabolism, Vol. 3. Academic Press, New York, pp. 423–506. 67. Robertie PG, Butterworth JF, Prielipp RC, Tucker WY, Zaloga GP1992 Parathyroid hormone responses to marked hypocalcemia in infants and young children undergoing repair of congenital heart disease. J Am Collect Cardiol 20:672–677. 68. Clemens TL, Henderson SL, Adams JS, Holick MF 1982 Increased skin pigment reduces the capacity of skin to synthesize vitamin D3. Lancet 1:74–76. 69. Dagnelie PC, Vergote FJVRA, van Staveren WA, van den Berg H, Dingjan PG, Hautvast JGAJ 1990 High prevalence of rickets in infants on macrobiotic diets. Am J Clin Nutr 51:202–208. 70. Kruse K 1995 Pathophysiology of calcium metabolism in children with vitamin D–deficiency rickets. J Pediatr 126:736–741. 71. Kumar R 1983 Hepatic and intestinal osteodystrophy and the hepatobiliary metabolism of vitamin D. Ann Intern Med 98:662–663. 72. Balsan S 1991 Hereditary pseudo-deficiency rickets or vitamin D–dependency type I. In: Glorieux FH (ed) Rickets

CHAPTER 64 The Hypocalcemic Disorders: Differential Diagnosis and Therapeutic Use of Vitamin D

73.

74.

75.

76. 77. 78.

79. 80. 81.

82. 83. 84. 85. 86. 87.

88. 89. 90.

(Nestle Nutrition Workshop Series, Vol. 21). Raven, New York, pp. 55–165. Wang X, Zhang MYH, Miller WL, Portale AA 2002 Novel gene mutations in patients with 1α-hydroxylase deficiency that confer partial enzyme activity in vitro. J Clin Endocrinol Metab 87:2424–2430. Balsan S, Garabedian M, Larchet M, Gorski A, Coumot G, Tau C, Bourdeau A, Silve C, Ricour C 1986 Long-term nocturnal calcium infusions can cure rickets and promote normal mineralization in hereditary resistance to 1,25-dihydroxyvitamin D. J Clin Invest 77:1661–1667. Pettifor JM, Ross FP, Travers R, Glorieux FH, DeLuca HF 1981 Dietary calcium deficiency: A syndrome associated with bone deformities and elevated serum 1,25-dihydroxyvitamin D concentration. Metab Bone Dis Related Res 2:301–305. Pettifor JM, Ross P, Wang J, Moodley GP, Couper-Smith J 1978 Rickets in children of rural origin in South Africa: Is low dietary calcium a factor? J Pediatr 92:320–324. Marie P, Pettifor J, Ross F, Glorieux F 1982 Histological osteomalacia due to dietary calcium deficiency in children. N Engl J Med 307:584–588. DeLucia MC, Mitnick ME, Carpenter TO 2003 Nutritional rickets with normal circulating 25-hydroxyvitamin D: a call for re-examining the role of dietary calcium intake in North American children. J Clin Endocrinol Metab 88:3539–3545. Herbert L, Lemann J, Petersen J, Lennon E 1966 Studies of the mechanism by which phosphate infusion lowers serum calcium concentration. J Clin Invest 45:1886–1894. Chernow B, Rainey T, Georges L, O’Brian J 1981 Iatrogenic hyperphosphatemia: A metabolic consideration in critical care medicine. Crit Care Med 9:772–774. Venkataraman P, Tsang R, Greer F, Noguchi A, Laskarzewski P, Steichen J 1985 Late infantile tetany and secondary hyperparathyroidism in infants fed humanized cow milk formula. Am J Dis Children 139:664–668. Arrambide K, Toto R 1993 Tumor lysis syndrome. Semin Nephrol 13:273–280. Dunlay R, Camp M, Allon M, Fanti P, Malluche H, Llach F 1989 Calcitriol in prolonged hypocalcemia due to the tumor lysis syndrome. Ann Intern Med 110:162–164. Llach F, Felsenfeld A, Haussler M 1981 The pathophysiology of altered calcium metabolism in rhabdomyolysisinduced acute renal failure. N Engl J Med 305:117–123. Brasier A, Nussbaum S 1988 Hungry bone syndrome: Clinical and biochemical predictors of its occurrence after parathyroid surgery. Am J Med 84:654–660. Abramson EC, Gajardo H, Kukreja SC 1990 Hypocalcemia in cancer. Bone Miner 10:161–169. Anast C, Carpenter T, Key L 1990 Metabolic bone disorders in children. In: Avioli L, Krane S (eds) Metabolic Bone Disease and Related Research. Saunders, Philadelphia, Pennsylvania, pp. 850–887. Llach F, Bover J 1996 Renal Osteodystrophy. In: Brenner B (ed) The Kidney, 5th Ed. Saunders, Philadelphia, Pennsylvania, pp. 2187–2273. Edmondson H, Berne C 1944 Calcium changes in acute pancreatic necrosis. Surg Gynecol Obstet 79:240–244. Stewart AF, Longo W, Kreutter D, Jacob R, Burtis WJ 1986 Hypocalcemia due to calcium soap formation in a patient with a pancreatic fistula. N Engl J Med 315:496–498.

1063

91. Zaloga GP, Chernow B 1987 The multifactorial basis for hypocalcemia during sepsis. Studies of the parathyroid hormone-vitamin D axis. Ann Intern Med 107:36–41. 92. Chesney RW, McCarron DM, Haddad JG, Hawker CD, DiBella FP, Chesney PJ, Davis JP 1983 Pathogenic mechanisms of the hypocalcemia of the staphylococcal toxic-shock syndrome. J Lab Clin Med 101:576–585. 93. Jaeger P, Otto S, Speck RF, Villiger L, Horber FF, Casez J-P, Takkinen R 1994 Altered parathyroid gland function in severely immunocompromised patients infected with human immunodeficiency virus. J Clin Endocrinol Metab 79:1701–1705. 94. Weinstein R, Bryce G, Sappington L, King K, Gallagher B 1984 Decreased serum ionized calcium and normal vitamin D metabolite levels with anticonvulsant drug treatment. J Clin Endocrinol Metab 58:1003–1009. 95. Tofalletti J, Nissenson RA, Endres D, McGarry E, Mogollon G 1985 Influence of continuous infusion of citrate on responses of immunoreactive PTH, calcium, magnesium components, and other electrolytes in normal adults during plasmapheresis. J Clin Endocrinol Metab 60: 874–879. 96. Mallette LE, Gomez LS1982 Systemic hypocalcemia after clinical injection of radiographic contrast media: Amelioration by omission of calcium chelating agents. Radiology 147:677–679. 97. Jacobson MA, Gambertoglio JG, Aweeka FT, Causey DM, Portale AA 1991 Foscarnet-induced hypocalcemia and effects of foscarnet on calcium metabolism. J Clin Endocrinol Metab 72:1130–1135. 98. Winer K, Yanovski J, Cutler G 1996 Synthetic human parathyroid hormone 1-34 vs calcitriol and calcium in the treatment of hypoparathyroidism: Results of a short-term randomized crossover trial. JAMA 276:631–636. 99. Russell R, Smith R, Walton R, Preston C, Basson R, Henderson R 1976 1,25-Dihydroxycholecalciferol and 1a-hydroxycholecalciferol in hypoparathyroidism. Lancet 2:14–17 (July 6). 100. Neer R, Holick M, DeLuca H, Potts J 1975 Effects of 1a-hydroxyvitamin D3 on calcium and phosphorus metabolism in hypoparathyroidism. Metabolism 24:1403–1413. 101. Slatopolsky E, Dusso A, Brown AJ 2002 Control of uremic bone disease: role of vitamin D analogs. Kidney Int 61(Suppl 80): S143–S148. 102. Rosier A, Rabinowitz D 1973 Magnesium-induced reversal of vitamin D–resistance in hypoparathyroidism. Lancet 1:803–804 (April 4). 103. Ahonen P, Myllarniemi S, Sipila I, Perheentupa J 1990 Clinical variation of autoimmune polyendocrinopathycandidiasis-ectodermal dystrophy (APECED) in a series of 68 patients. N Engl J Med 322:1829–1836. 104. Gardezi SA, Nguyen C, Malloy PJ, Posner GH, Feldman D, Peleg S 2001 A rationale for treatment of hereditary vitamin D– resistant rickets with analogs of 1 alpha,25-dihydroxyvitamin D3. J Biol Chem 276:29148–29156. 105. Lacy CF, Armstrong LL, Goldman MP, Lance LL. Drug Information Handbook. Lexi-Comp. 2003;11th ed. 220–23. 106. Kastrup EK ed. Drug Facts and Comparisons. St. Louis: Facts & Comparisons; August 2003. 107. Walgreens Pharmacy. www.walgreens.com. Accessed on July 31, 2003.

CHAPTER 65

Vitamin D Deficiency and Nutritional Rickets in Children JOHN M. PETTIFOR

MRC Mineral Metabolism Research Unit, Department of Pediatrics, University of the Witwatersrand and Chris Hani Baragwanath Hospital, P O Bertsham 2013, South Africa

I. Introduction II. Historical Perspective III. The Epidemiology of Vitamin D Deficiency and Nutritional Rickets IV. Clinical Presentation V. Biochemical Abnormalities

VI. Radiologic Changes VII. Treatment and Prevention VIII. Dietary Calcium Deficiency IX. The Pathogenetic Spectrum of Nutritional Rickets X. Conclusions References

I. INTRODUCTION

was known in Europe as “the English disease,” was more common in the cities than in rural areas. Prior to the industrial revolution, it was associated with affluence, as the children of well-to-do families were often completely covered by clothing and were kept indoors. With the migration of large numbers of people from rural to urban areas at the time of the industrial revolution, the disease became associated with poverty and overcrowding in the developing urban slums. A number of studies in the late 19th and early 20th centuries documented the almost universal prevalence of rickets in young children in cities in northern Europe (for example in Glasgow [1] and Vienna [2]). However, with the realization of the importance of ultraviolet light in preventing nutritional rickets and the discovery and isolation of vitamin D in the first quarter of the 20th century [3] (see Chapter 1), program were introduced to prevent vitamin D deficiency. In the United Kingdom, a number of foods were fortified with vitamin D during the World War II. This led to a rapid reduction in the number of children diagnosed with rickets, but in the following years the incidence of idiopathic hypercalcemia rose in infants, which at the time was thought to be due to uncontrolled fortification of various foods (especially milk and cereals) leading to daily intakes of 100 µg or more [4]. As a result, the fortification of foods and the use of vitamin D supplements fell into disrepute, and the prevalence of vitamin D deficiency and nutritional rickets has increased, particularly among the immigrant Asian population. In the United States, the universal fortification of milk with vitamin D at 400 IU/quart from the 1930s has almost eradicated nutritional rickets except in families who exclude milk from their diets [5]. However, as had

Rickets is a clinical syndrome that presents in children as a result of a failure of or delay in mineralization of the growth plate of growing bones. There are numerous different causes, the majority of which can be grouped into three major categories: those which primarily result in a failure to maintain normal calcium homeostasis; those which primarily affect phosphate homeostasis; and those which directly inhibit the mineralization process. Globally, rickets due to nutritional causes (which fall into the calciopenic group) remains the most frequent form of the disease seen. However, in a number of industrialized countries, such as the U.S., the genetic forms of hypophosphatemic rickets are now probably more prevalent than the nutritional causes outside the neonatal period, as a result of the fortification of foods with vitamin D and the use of vitamin D supplements in at-risk groups. Nevertheless, the last decade has seen a resurgence of nutritional rickets in minority communities in a number of developed countries.

II. HISTORICAL PERSPECTIVE Although nutritional rickets is often considered to be a disease of industrialization, descriptions of rickets have been attributed to both Homer (900 BC) and Soranus Ephesius (130 AD). More recently, attention was drawn to rickets by Daniel Whistler in 1645, and five years later Francis Glisson (1650) provided a classic description of the disease. It was described as a disease that occurred in young children, produced severe deformities, and was often fatal. The condition, which VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

1066 occurred in the United Kingdom, a few cases of vitamin D toxicity have been reported as a result of the lack of monitoring of the fortification process [6]. In central Europe, rickets has been effectively prevented in infants and young children by the intermittent administration (every three to five months for the first two years of life) of high doses of vitamin D (“stosstherapie”) [7].

III. THE EPIDEMIOLOGY OF VITAMIN D DEFICIENCY AND NUTRITIONAL RICKETS Vitamin D deficiency is a prerequisite for the development of nutritional rickets in the majority of children. Thus, the disease is typically associated with a lack of ultraviolet light exposure or dietary vitamin D. As commonly ingested foods are generally deficient in vitamin D (the exceptions being oily fishes or fortified foods), the normal diet contributes little to the vitamin D status of an individual [8], so adequate skin exposure to ultraviolet radiation is essential for the prevention of rickets in most situations [9] (see Chapter 3). Consequently, rickets occurs most frequently in infants before they are able to walk and get outside, in children living in countries at the extremes of latitude, or in communities in which social custom prevents adequate sunlight exposure through excessive skin coverage by clothes or through the practice of purdah. Vitamin D deficiency rickets is most prevalent in children under two years of age, with a peak incidence between 3 and 18 months [10,11]. The disease is uncommon in infants under three months of age because 25OHD readily crosses the placenta [12,13], thus providing the newborn infant with some protection against vitamin D deficiency [14] (see Chapter 51). Because 25OHD is not the major storage form of vitamin D and has a turnover time of three to four weeks, serum levels fall rapidly after birth unless additional sources of vitamin D are obtained by the young infant [15]. Neonatal or congenital rickets has been described in infants born to mothers who are themselves vitamin D deficient [16–21], and hypocalcemia is a common finding in neonates born to vitamin D–deficient mothers. In a number of studies vitamin D deficiency rickets has been noted to occur more commonly in boys than girls [22–24], however the mechanism for this remains unclear. It has been suggested that vitamin D deficiency rickets might be an hereditary disease, which manifests itself only under adverse circumstances [23,25]. In a study of infants with rickets and their parents [25], urinary excretion of α-amino

JOHN M. PETTIFOR

acids was increased in one-third of the infants a long time after the rickets had healed, many of the parents had increased amino acid and phosphorus excretion, and a good correlation was found between the excretion of individual amino acids by an infant and its parents. The authors suggested that these findings indicate a genetic factor playing a role in predisposing a child to rickets; however, the mode of inheritance is unclear. In the early literature, breast feeding was reported to be protective against rickets [26]. More recently however, it has been described as a risk factor for the development of rickets [5,27–29]. In recent years, specifically designed breast milk substitutes have replaced natural cow’s milk as the major source of nutrients for the non breast-fed infant. This alteration in feeding patterns may account for the apparent change in risk associated with breast-feeding for several reasons; first, breast milk substitutes are fortified with vitamin D at 400 IU/liter while natural cow’s milk contains little vitamin D [30]; second, the calcium:phosphorus ratio in breast milk substitutes (ratio ~2:1) is more appropriate than that in cow’s milk (ratio ~1:1) for optimizing intestinal calcium absorption; and third, breast milk usually contains only small quantities of vitamin D or its metabolites (between 20–65 IU/liter) [30,31]. However, there is evidence that vitamin D metabolites may cross into breast milk from the mother in sufficient quantities to maintain normal serum concentrations of 25OHD in the suckling infant if the mother receives vitamin D supplements in high doses (~2000 IU/day) [15,32]. In the breast-fed infant not receiving vitamin D supplements, the maintenance of an adequate vitamin D status is dependent mainly on the infant’s exposure to ultraviolet light [33,34]. Specker and coworkers [33,34] have shown a marked seasonal variation in serum 25OHD concentrations in breast-fed infants, which is dependent on the time spent outdoors and on the extent of skin exposed to sunlight (see Chapter 51). They have estimated that an infant in Cincinnati (latitude 39° 09′ N) needs to be outdoors for either 20 minutes a week in a diaper only or for two hours a week fully clothed but without a hat to maintain normal circulating concentrations of 25OHD [33]. Seasonal variations in serum 25OHD concentrations have also been documented in a number of countries in older children and adults [35–37], and these variations appear to correlate with the amount of ultraviolet light reaching the earth [38]. These observations highlight the importance of the photo-biosynthesis of vitamin D3 in the skin to prevent vitamin D deficiency and thus rickets in many populations in the world. In a number of countries such as Turkey [39], Saudi Arabia [29], India [40], China [21] (including Tibet [41]), Algeria [37],

1067

CHAPTER 65 Nutritional Rickets in Children

Iran [10], Kuwait [42], Nigeria [43], Ethiopia [44] and in others in the tropics and subtropics [45–47] rickets remains a problem despite generally good daily hours of sunshine. A number of factors contribute to the persistence of the problem in these areas; these include overcrowding and poverty, atmospheric pollution [48], purdah, lack of access to sunlight, a lack of vitamin D fortified foods or regular vitamin D supplements, and diets that are low in calcium and high in inhibitors of calcium absorption. This subject is dealt with in more detail in Chapters 47 and 62. In the U.S., despite the almost complete eradication of vitamin D deficiency among Caucasian children, several studies have highlighted the resurgence of the problem in specific groups [5,27,28,49–53], namely vegans and children on macrobiotic diets, children who are breast-fed for prolonged periods, and black children [54]. It is suggested that the combination of decreased vitamin D3 formation in the dark skin, extensive skin coverage by clothing, low dietary vitamin D intakes because of the lack of dairy products, and the generally low dietary calcium intakes associated with vegetarian diets all contribute to an increased risk for vitamin D deficiency in these groups. A similar pattern has also been documented in a number of European countries [55–59], and in Australia and New Zealand [60–62]. Perhaps the most intensively investigated community has been the Asian population in Great Britain because of the high prevalence of vitamin D deficiency and “Asian rickets” in children of all ages [63] and adults [64,65]. The age distribution of Asian children with rickets is described as being biphasic, with one peak in the classical age group of vitamin D deficiency (9–36 months) and the other related to the pubertal growth spurt [1,66]. Since the initial descriptions of the resurgence of rickets in the United Kingdom in the early 1960s, numerous studies have been undertaken to determine why Asians are predisposed to the problem when other immigrants such as West Indians are not. Among the hypotheses put forward are simple vitamin D deficiency due to the dark skin and lack of skin surface exposed to sunlight [67,68], low calcium diets associated with vegetarianism [69], and impaired intestinal calcium absorption associated with high phytate diets [70]. A unifying hypothesis, proposed by Clements [55], suggests that in a situation of relative vitamin D insufficiency, the low dietary calcium and high phytate content of the typical Asian vegetarian diet leads to mild secondary hyperparathyroidism and a resultant increase in the catabolism of vitamin D. The progressive decline in vitamin D status culminates in the development of rickets (see Section VIII of this chapter).

IV. CLINICAL PRESENTATION The majority of clinical signs in children with rickets results from the effects of vitamin D deficiency on the mineralization process at the growth plate or on calcium homeostasis. Fraser and coworkers [24] have described three stages in the progression of vitamin D deficiency. Stage I is characterised by hypocalcemia with clinical signs related to the presence of hypocalcemia, in stage II the clinical features of impaired bone mineralization become apparent, and in stage III signs of both hypocalcemia and severe rickets are present. This division of the progression of vitamin D deficiency rickets is conceptually useful, but there is considerable clinical overlap between the various stages. The early clinical manifestations of vitamin D deficiency (stage I) are related to hypocalcemia and are more commonly seen in young infants (less than 6 months of age). They may present with convulsions [71,72], apnoeic episodes [73] or tetany with no clinical signs of rickets. Few children present clinically in stage I as the majority who later present with rickets pass through this phase without developing symptomatic hypocalcemia. Pseudotumor cerebri [74] and cataracts, probably due to hypocalcemia, have been reported in a young infant with rickets [75]. It has been suggested that symptomatic hypocalcemia in infants with vitamin D deficiency might be precipitated by an acute illness [76], in which there is a release of intracellular phosphate [77]. As the deficiency progresses, the classical features of rickets become apparent. Typically, the infant or young child presents with a delay in motor milestones, hypotonia, and progressive deformities of the long bones. The deformities are most noticeable at the distal forearm with enlargement of the wrist and bowing of the distal radius and ulna, and in the legs with progressive lateral bowing of the femur and tibia. The site and type of deformity are dependent on the age of the child and the weight bearing patterns in the limbs. Thus, in the small infant, deformities of the forearms and anterior bowing of the distal tibias are more common, while in the toddler who has started to walk an exaggeration of the normal physiological bowing of the legs (genu varum) is characteristic. In the older child, valgus deformities of the legs or a windswept deformity (valgus deformity of one leg and varus deformity of the other) may be apparent. The characteristic feature in the ribs is enlargement of the costochondral junctions leading to visible beading along the anterolateral aspects of the chest (the rachitic rosary). In the infant or young child with severe rickets, the muscular pull of the diaphragmatic attachments to the lower ribs

1068

FIGURE 1

A young infant with vitamin D deficiency rickets, presenting with respiratory distress. The child shows the characteristic deformities of the chest associated with severe rickets. The lateral diameter of the chest is reduced and bilateral Harrison’s sulci are present. The abdomen has a protuberant appearance.

results in the development of Harrison’s sulcus (Fig. 1). The negative intrapleural pressure associated with breathing may result in narrowing of the lateral diameter of the chest (the violin case deformity) with consequent severe respiratory embarrassment. Increased sweating has also been described in young infants and probably relates to the increased work of breathing due to the decreased compliance associated with the excessively malleable ribs. In premature infants with rickets (which may also be due to dietary phosphorus deficiency), fractures of the ribs may be the first clinical sign to draw attention to the problem [78]. Other skeletal abnormalities include a delay in the closure of the fontanelles, parietal and frontal bossing, and the presence of craniotabes [79]. Although craniotabes

JOHN M. PETTIFOR

is often considered to be pathognomonic of rickets, it may be a normal finding in normal young infants [80,81]. Craniosynostosis, involving the coronal or multiple sutures, has been described in approximately 25% of patients, who were followed up after having suffered from vitamin D deficiency rickets [82]. The development of craniosynostosis appears to be related to the degree of severity of the rickets and thus to the severity of the mineralization defect, and inversely to the age of onset of the rickets. A delay in tooth eruption is a feature of rickets in the young child and enamel hypoplasia of teeth may occur if rickets develops prior to the completion of enamel deposition. The latter has been reported in the primary dentition of infants born to mothers who are vitamin D deficient [83], and is seen in the secondary dentition of children who have suffered from rickets during early childhood. Hypotonia, decreased activity, and a protuberant abdomen are characteristic features of advanced vitamin D deficiency rickets in the infant and young child. These signs are probably analogous to the proximal muscle weakness described in vitamin D deficient adolescents and adults [84]. In this situation, deep tendon reflexes are retained and may be brisk. The pathogenesis of the myopathy is thought to be due primarily to vitamin D deficiency, rather than hypophosphatemia [85] (see Chapters 55 and 102). Dilated cardiomyopathy and cardiac failure [86,87] have also been described in young infants with vitamin D deficiency. The mechanism is thought to be due to the effect of hypocalcemia on cardiac muscle function, rather than a direct effect of hypovitaminosis D [88]. Infants and young children with rickets are prone to an increased number and severity of infections [39]. Although the increase in respiratory infections may be explained on the thoracic cage abnormalities (softening of the ribs, the enlarged costochondral junctions, and the decreased thoracic movement due to muscle weakness), other reasons for the increase in diarrheal disease must be sought. The now well-documented role of 1,25(OH)2D in modulating immune function [89,90] may contribute to the observed increase in infections (see Chapter 35). Impaired phagocytosis [91] and neutrophil motility [92] have been described in children with vitamin D deficiency rickets. A possibly associated abnormality is anemia, thrombocytopenia, leucocytosis, myelocytosis, erythroblastosis, myelofibrosis [93], myeloid metaplasia, and hepatosplenomegaly (von Jacksch-Luzet syndrome) [94], which has been described in infants with rickets [95,96]. Although the exact pathogenetic mechanisms for this syndrome are unclear, vitamin D deficiency has been implicated based on the clinical observation that

CHAPTER 65 Nutritional Rickets in Children

vitamin D therapy cures the condition and on experimental evidence showing that 1,25(OH)2D has antiproliferative activity on myeloid leukaemia cell lines [97].

V. BIOCHEMICAL ABNORMALITIES The hallmark of vitamin D deficiency is a low circulating level of 25OHD. In children, a normal range of approximately 12 to 50 ng/ml (30–125 nmol/liter) has been found in the majority of studies [33,98,99] conducted in communities in which vitamin D deficiency rickets is uncommon. However, the normal range is dependent on the vitamin D and calcium contents of the diet and on the ultraviolet light exposure of the skin. In a number of studies a marked seasonal variation in levels has been recorded [34,35,100], reflecting in part the seasonal changes in the amount of ultraviolet light reaching the earth. In countries at high latitude where foods are not vitamin D fortified, serum 25OHD concentrations in some “normal” children may be in the range documented in symptomatic children with vitamin D deficiency [35,63]. Thus, the development of symptoms depends on the duration and severity of low 25OHD concentrations and on the ability of the kidney to achieve adequate 1,25(OH)2D concentrations in the face of decreased substrate for the gastrointestinal tract to maintain calcium absorption at a level appropriate to meet the demands of the growing child. In children with 25OHD concentrations within the normal reference range, there is no correlation between serum 25OHD and 1,25(OH)2D concentrations. However, once 25OHD levels fall below ∼12 ng/ml (30 nmol/liter), 1,25(OH)2D concentrations correlate with those of 25OHD [101,102]. In the majority of studies in which 25OHD values have been measured in children with rickets, concentrations have been found to be less than 4–5 ng/ml (10–12.5 nmol/liter) in most patients [63,103,104], although other workers have found higher values [105–107]. The classical biochemical changes in vitamin D– deficient children who have radiological changes of rickets are a combination of hypocalcemia, hypophosphatemia, and elevated alkaline phosphatase and parathyroid hormone concentrations. In the early phase of vitamin D deficiency before the development of radiological signs (stage I), hypocalcemia may be the only biochemical abnormality [24]. Acute illness may precipitate hypocalcemia in the vitamin D–depleted infant through the sudden increase in serum phosphorus concentrations [77]. The biochemical picture in stage I rickets may be confused with that of pseudohypoparathyroidism [108], as serum hypocalcemia, hyperphosphatemia, and normal alkaline phosphatase may

1069 be found. As the disease progresses, secondary hyperparathyroidism in response to the hypocalcemia induces a partial correction of the low serum calcium concentration, which may return to levels within the normal range, and increases phosphate excretion by the kidney resulting in hypophosphatemia (stage II) [109]. At this stage, serum alkaline phosphatase concentrations are usually elevated and other renal manifestations of secondary hyperparathyroidism, such as increased cyclic AMP excretion, generalized aminoaciduria, impaired acid excretion, and decreased urinary calcium excretion, are found [110]. In stage III of the disease, the radiological features are more severe, hypocalcemia once again becomes apparent, and alkaline phosphatase concentrations rise further [24]. The elegant studies conducted by Fraser and coworkers [24] before the availability of immunoassays for the measurement of serum parathyroid hormone (PTH) concentrations, suggested that in stage I vitamin D deficiency serum concentrations of PTH are normal as serum phosphorus values and urinary amino acid excretion are within the normal range. Their patients only had radiologic evidence of calvarial demineralization without other bone changes of rickets. More recent data support this conclusion as normal PTH concentrations have been reported in the early hypocalcemic phase of symptomatic vitamin D deficiency [71]. However, Kruse [111] found elevated PTH values and increased urinary cyclic AMP excretion in children with stage I rickets. This discrepancy can possibly be explained by the fact that the patients in the latter study might represent a slightly later stage of vitamin D deficiency than those in the other studies as the children were selected on the presence of radiologic changes. Evidence of end-organ resistance to PTH has been found in the young children with both mild and more severe radiological rickets [111,112]. In the study by Kruse [111] the children with mild rickets remained normophosphatemic and had normal renal handling of phosphate (TmP/GFR) despite elevated PTH concentrations and increased urinary cyclic AMP excretion. Similar indirect evidence of PTH resistance (hypocalcemia, normophosphatemia, and a decrease in the phosphate excretion index) was noted by Taitz and de Lacy [112] in infants with more severe radiologic rickets. Resistance to PTH has also been described in hypocalcemic adolescents with mild rickets [113]. Usually, however, as the severity of the rickets increases (stages II and III), so PTH values rise further and renal hyporesponsiveness is overcome [111]. Thus, hypophosphatemia and a decrease in TmP/GFR become hallmarks of the disease. Markers of bone turnover are typically elevated in nutritional rickets in response to the development of secondary hyperparathyroidism. Urinary hydroxyproline

1070 excretion may be within the normal range in stage I rickets, but is elevated in patients with radiologic rickets [111], and an increase in serum concentrations of bone resorption markers have been reported in children with untreated rickets [114,115]. Similarly, serum alkaline phosphatase values may be normal in stage I of vitamin D deficiency, but rise with the degree of severity of the radiologic changes. Bone turnover markers (especially those of bone resorption) rise in the first 2–3 weeks of treatment, and then fall progressively to normal values over a period of 4–6 weeks [115]. Of all the readily available biochemical tests that might be deranged in nutritional rickets, alkaline phosphatase has been used most frequently as a screening test. However, although alkaline phosphatase is elevated in the vast majority of children with radiological changes, it lacks specificity [45,81,116]. Further, the degree of elevation of serum concentrations does not necessarily correlate with the radiological severity of the bone disease [45]. Whether or not the measurement of bone specific alkaline phosphatase in patients with suspected rickets will be of greater sensitivity and specificity is unclear at present [117]. Recently, in a small study of rachitic subjects, it has been suggested that the measurement of deoxypyridinoline in a first morning void urine sample might be a useful indicator of rickets, values being significantly higher in patients than in age matched controls [118]. Osteocalcin is a noncollagenous bone matrix protein that binds to hydroxyapatite and is secreted by osteoblasts during mineralization [119]. Serum concentrations are higher in children than adults and peak during the pubertal growth spurt [120]. In the few children with untreated vitamin D deficiency rickets, in whom serum osteocalcin concentrations have been measured, values have been reported to be low [115] or normal [121], and may rise rapidly on therapy to supranormal concentrations [122]. A Nigerian study [114] of 12 rachitic children found slightly elevated serum osteocalcin concentrations compared to values in agematched controls, however it was suggested that the children might have suffered from dietary calcium deficiency rather than vitamin D deficiency. In patients with vitamin D deficiency, serum 1,25(OH)2D concentrations have been reported to be low, normal, or even elevated [101,104,107,111,123,124], while 24,25(OH)2D values are low or undetectable [101,104,107,124,125]. Kruse [111] found that 1,25(OH)2D values were higher in children with stage II rickets than in those with either stage I or stage III rickets. The finding of normal or elevated levels of 1,25(OH)2D in vitamin D deficiency–rickets has led some researchers to conclude that other vitamin D metabolites, such as 24,25(OH)2D, are necessary for

JOHN M. PETTIFOR

the maintenance of normal calcium homeostasis [123,126,127]. Others have suggested that although concentrations are within the normal range, they are inappropriately low for the degree of hyperparathyroidism [111,124]. As discussed later, the latter hypothesis is more likely. A possible pathophysiological progression of vitamin D deficiency rickets in children may be described as follows [111,128]. As the child becomes progressively vitamin D–depleted, a stage is reached when the serum 25OHD concentration falls below that required to maintain a serum 1,25(OH)2D level necessary for normal calcium homeostasis. The resultant hypocalcemia (stage I rickets) leads to secondary hyperparathyroidism, which through the stimulation of 1α-hydroxlase, increases 1,25(OH)2D production despite falling 25OHD concentrations. In concert with PTH, 1,25(OH)2D increases bone resorption and intestinal calcium absorption, thus returning serum calcium concentrations towards normal (stage II rickets). The presence of hypophosphatemia at this stage is probably responsible for the mineralization defect and the development of radiologic rickets. It is during this phase that serum 1,25(OH)2D concentrations may be elevated [111]. A possible explanation for the failure of the elevated 1,25(OH)2D levels to reduce the hyperparathyroidism and heal the bone disease at this stage is that they are not high enough to meet the increased calcium requirements associated with the generalized mineralization defect and increased bone turnover. Support for this hypothesis come from data which show that 1,25(OH)2D concentrations rise to considerably higher levels (3–5 times normal) during the healing process even when only small doses of vitamin D are provided [101,111] and that intestinal calcium absorption may reach ~80% of dietary calcium intake during this phase [101]. As 25OHD concentrations fall further, 1,25(OH)2D levels once again fall, despite persistent hyperparathyroidism, because of the lack of substrate. Hypocalcemia again becomes apparent as intestinal calcium absorption falls and calcium mobilization from bone decreases due to the lack of 1,25(OH)2D, which has a permissive action on bone resorption by PTH [129,130]. The combination of both hypocalcemia and hypophosphatemia increases the severity of the bone disease (stage III).

VI. RADIOLOGIC CHANGES The typical radiologic changes associated with vitamin D deficiency rickets have been well described and are discussed in Chapter 60. Stage I rickets characteristically

CHAPTER 65 Nutritional Rickets in Children

shows few radiologic signs, although demineralization of the calvarium and loss of definition of the skull sutures have been described [24], but these signs are difficult to quantify. The changes of rickets are best visualized at the growth plate of rapidly growing bones. In the upper limbs, the distal ulna is the site that may show best the early signs of impaired mineralization. In the older child, the metaphyses around the knees become more useful. The early signs of rickets include widening of the epiphyseal plate and a loss of definition of the provisional zone of calcification at the metaphysis [131]. As the disease progresses, the disorganization of the growth plate becomes more apparent with cupping, splaying, spur formation, and stippling [77,132] (Fig. 2). The appearance of epiphyses may be delayed or they appear small, osteopenic and ill-defined. The shafts of the long bones show features of both hyperparathyroidism and osteomalacia. Osteopenia is a characteristic feature which in the so called “atrophic” form of the disease may be very severe [45]. The cortices become thin and may show periosteal new bone formation, although this is more frequently seen during healing. The trabecular pattern is reduced and appears coarse. Deformities of the shafts of the long bones are typically present and in severe rickets, pathological fractures and Looser’s zones may be noted. In vitamin D deficiency rickets, features of hyperparathyroidism, such as subperiosteal erosions,

1071 are uncommon. However, loss of the lamina dura round the teeth is frequently seen. Enlargement and splaying of the costochondral junctions on the lateral radiographs of the chest have been used as a sign of rickets; however, in one study, mild changes were found to be unreliable as their presence did not correlate with serum 25OHD concentrations or with other features of rickets at the distal radius and ulna [133]. Rickets during adolescence may be difficult to detect using the conventional radiographic sites of the wrist and knees as the epiphyseal plates narrow and epiphyses fuse. A radiograph of the pelvis may be useful in this situation as the secondary iliac and ischial ossification centres may be abnormally wide [134]. These centers appear at puberty and normally unite with the rest of the bone between the 15th and 25th years of age. The sign of early healing of rickets is described as broadened bands of increased density replacing the normal sharp metaphyseal lines (Fig. 2). The demarcation of the broad bands on the diaphyseal side of the shaft may be poorly defined [131]. Healing in more severe cases of rickets may first appear as bands of mineralization occurring distal to and separated from the irregular and frayed metaphyses. There is then gradual filling in of the demineralized area proximal to the initial band of mineralization with remodelling and

FIGURE 2 The radiographic features of vitamin D deficiency rickets at the wrist. Left panel: untreated vitamin D deficiency showing underdevelopment of the epiphyses, widening of the epiphyseal plates, splaying and irregularity of the metaphyses and loss of the provisional zones of calcification. The shafts show coarsening of the trabecular pattern and loss of the normal cortical definition. Middle panel: Response after three months of vitamin D therapy. The metaphyses show clear signs of healing with dense bands of calcification at the distal ends of the metaphyses, narrowing of the epiphyseal plates, and more clearly defined epiphyses. The trabecular pattern still appears coarse but shows improvement. Right panel: Six months after starting vitamin D therapy. The radiographic changes of rickets have disappeared. The epiphyses, epiphyseal plates, metaphyses, and trabecular structure are normal. (Reproduced with permission [132].)

1072 the development of a normal trabecular pattern. Periosteal new bone formation may be seen which gradually becomes incorporated into the cortices of the long bones.

VII. TREATMENT AND PREVENTION A. Treatment Vitamin D deficiency rickets can be effectively treated by the oral administration of small doses of either vitamin D2 or D3, provided there is no evidence of gastrointestinal malabsorption. Stanbury et al. [101] showed that an oral vitamin D dose of between 200 to 450 IU/day produced a rise in serum 1,25(OH)2D concentrations to normal values within 1–3 days. The latter climbed to reach a peak some five times the normal mean after one to three weeks, despite serum 25OHD values remaining less than 10 ng/ml (25 nmol/l). Spontaneous improvement in the biochemical features of rickets has been reported to occur in children with biochemical abnormalities during the summer months, associated with a rise in serum 25OHD values due presumably to increased ultraviolet light exposure [103]. More generally, however, doses of vitamin D between 5,000 and 15,000 IU/day for three to four weeks are used in the management of rickets. Normalization of serum calcium and phosphorus concentrations occur within 1 and 3 weeks [111], although serum alkaline phosphatase concentrations and urinary hydroxyproline excretion remain elevated for several months. Despite the return to normal of serum PTH, calcium, and phosphorus values within three weeks, serum 1,25(OH)2D concentrations may remain elevated for up to 10 weeks [107,111]. Serum 24,24(OH)2D values, which are often undetectable in the untreated patient, rise with the progressive increase in serum 25OHD concentrations during treatment [107]. Lower doses of vitamin D (1000–2000 IU/day) do produce healing but the response is less rapid. In Central Europe, a single dose of 600,000 IU vitamin D (either orally or intramuscularly) has been found to be effective, resulting in a rapid improvement in biochemical abnormalities within a few days and radiologic evidence of healing within two weeks [77,135]. A sustained drop in serum alkaline phosphatase is seen within 6 to 12 weeks [135]. Single dose therapy has an advantage over smaller daily doses as it avoids the problem of compliance, which was thought to be responsible for the lack of response in 40% of children with vitamin D– deficiency rickets in a study conducted in Kuwait [136]. A recent article has raised concern about the use of 600,000 IU vitamin D in the treatment of rickets as hypercalcemia was reported in a small number of infants a month after having received the vitamin D

JOHN M. PETTIFOR

dose [137]. These authors suggest that a dose of 150,000 IU is equally effective as the larger dose in the management of the disease without running the risk of hypercalcemia. Besides ensuring an adequate vitamin D intake, the calcium content of the diet should be optimized (between 600 and 1000 mg/day) during the initial stages of management. This is particularly true for children who are on vegetarian or low calcium containing diets [138] and for those who are severely hypocalcemic. In symptomatic patients, a single dose of calcium gluconate (1–2 ml/kg of a 10% solution) may be given slowly intravenously and the diet supplemented with 10% calcium gluconate (5 ml/kg/day in divided doses).

B. Prevention As discussed in section III of this chapter, vitamin D deficiency rickets remains a problem in a number of atrisk groups despite readily available methods of preventing the disease. A number of studies in several countries have been conducted prospectively in breastfed infants to assess vitamin D status. Several have shown a fall in serum 25OHD concentrations in those infants who were not vitamin D–supplemented, to levels in the vitamin D–deficient range [15,139,140], although this is not a universal finding [141,142]. Further, a number of studies have highlighted the high prevalence of vitamin D deficiency in mothers during pregnancy and lactation, which exacerbates the severity and onset of vitamin D deficiency in their offspring [14,65,143,144]. Preventive strategies should be directed not only at breast-fed infants but also at pregnant and breast-feeding women [145]. Both North America [146] and the United Kingdom [147] recommend dietary intakes of vitamin D of between 200 and 400 IU/day for pregnant and lactating women to ensure adequate circulating 25OHD levels. Although at normal circulating maternal 25OHD concentrations, the vitamin D content of breast milk is limited (see Section III), there is evidence that maternal supplementation with vitamin D at 2000 IU/day may increase breast-milk vitamin D concentrations sufficiently to maintain the infant’s 25OHD within the normal range. The North American and United Kingdom groups recommend dietary intakes of between 200 and 350 IU vitamin D for the breast-fed infant [146,147]. The American Academy of Pediatrics in its latest recommendations suggest that infants less than 6 months of age should be kept out of direct sunlight, that children’s activities should minimize sunlight exposure, and that sunscreens should be used because of the indirect

CHAPTER 65 Nutritional Rickets in Children

evidence that early exposure to sunlight might determine the risk of skin cancer in later life [148]. These recommendations make it imperative that if the above guidelines are followed, supplemental vitamin D (200–400 IU/day) should be provided to all breast-fed and weaned infants ingesting less than 500 ml of infant milk formula/day [148–151]. In a prospective study conducted in China on infants from birth to 6 months of age, it was concluded that supplemental vitamin D at a dose of 400 IU/day produced more normal circulating 25OHD concentrations than did either 100 or 200 IU/day [152], however, in a small number of infants even 400 IU/day did not maintain 25OHD levels above 11 ng/ml (27.5 nmol/liter). Nevertheless, no radiologic evidence of rickets was found in any of the infants in the three groups at six months of age. The use of 400 IU vitamin D daily to prevent vitamin D deficiency in at-risk infants is supported by a study from Turkey, in which it was found that no rickets occurred in infants receiving the supplement compared to a prevalence of 3.8% in those that did not [39]. Infants fed milk formulas or cow’s milk fortified with vitamin D do not require vitamin D supplements, as their intake of milk generally provides sufficient vitamin D to prevent deficiency [141]. As discussed earlier, high single dose therapy (stosstherapie) has been used with success in the treatment of vitamin D deficiency rickets in a number of countries. A similar dose has also been used on a regular intermittent basis of every 3 to 5 months for the first 18 months of life as a means of prevention of vitamin D deficiency. Little data is available on the efficacy of such prophylaxis. However, in a study to assess the effect of these high doses of vitamin D (600,000 IU) on calcium and vitamin D metabolism in infants [7], it was found that serum 25OHD concentrations reached very high levels two weeks after each administration, but that these had returned to normal prior to the next dose. 1,25(OH)2D generally remained within the normal range, but 34% of infants were hypercalcemic at some stage during the study. These results led the authors to conclude that the dosage regimen as used during the study was excessive and unsafe [7]. Following these results, a study to assess the efficacy of a single dose of vitamin D (600,000 IU or 15 mg) at 15 days of life, compared to 200,000 IU (5 mg) at birth or 100,000 IU (2.5 mg) at birth and three monthly for nine months was undertaken [153]. Two weeks after the initial administration, 28 of 30 infants in the 15 mg group had serum 25OHD concentrations above the upper limit of normal (mean ± SD for the group; 307 ± 160 nmol/liter) compared to 58% (150 ± 55 nmol/liter) in the 5 mg group and 23% (92 ± 42 nmol/liter) in the 2.5 mg group. At 6 months of age,

1073 50% of the infants who had received 15 mg at birth, still had elevated 25OHD concentrations, while in the 5 mg group none had elevated levels. In the group receiving 2.5 mg every three months, serum 25OHD values were in the normal range on each occasion prior to receiving the next dose. Although hypercalcemia was not detected in any of the infants, serum calcium concentrations were higher in the 15 mg group two weeks after receiving the dose than in the other two groups. The authors concluded that intermittent doses of 15 mg vitamin D during the first year of life are excessive, and that 5 mg every six months or even better 2.5 mg every three months are more suitable for the prevention of vitamin D deficiency in at-risk infants. Vitamin D supplementation should be considered for all breast-fed infants living in temperate climates until they are ambulatory and are able to play outside [154]. Even in countries closer to the equator, where sunlight exposure should not be a problem, social customs may place the mother and infant at risk from vitamin D deficiency. In such situations (e.g., the Middle East and in Muslim communities in North Africa) vitamin D supplementation may also be necessary to reduce the high prevalence of vitamin D deficiency [144,155]. In a number of countries, vitamin D deficiency is not just a disease of breast-fed infants and their mothers. Rickets has been described in adolescents of Indian and Pakistani descent in the United Kingdom and in the Middle East [156,157], while hypovitaminosis D has been reported in adolescents in a number of European countries [158–161], India [162,163] and China [164]. Furthermore there is an increasing awareness of the high prevalence of what is considered to be vitamin D insufficiency in many elderly subjects in Europe and North America [165]. With the widespread nature of vitamin D deficiency in many countries, vitamin D supplementation is unlikely to be an effective means of combating the disease on a community basis, so food fortification should be considered as a possible solution. Although the untargeted fortification of foods other than milk and infant milk formulas has been used in the past as a means of addressing the high prevalence of vitamin D deficiency in countries where the risk of vitamin D deficiency is high, the problems experienced in the United Kingdom after World War II have led to it falling into disfavor (see Section II of this chapter). More recently, the use of targeted food fortification has been studied in the Asian community in Great Britain as a means of reducing the high prevalence of vitamin D deficiency in both adults and children in that community [166]. In a small pilot study, it was found that the fortification of chapatti flour at a level of 6000 IU/kg produced a sustained and significant rise in serum

1074 25OHD concentrations to values within the normal range over a six-month period comparable to that achieved by a weekly dose of 3000 IU vitamin D. Over the six-month period, serum calcium and phosphorus values rose, and the number of subjects with biochemical abnormalities suggestive of rickets fell. The authors conclude that fortification of chapatti flour is a cheap and effective method of preventing vitamin D deficiency in the Asian community in Britain, and has the advantage over daily or intermittent vitamin D supplementation as the compliance utilizing the latter form of prevention is often poor. Nevertheless, food fortification remains an emotional public issue. In the U.S. not only have there been isolated reports of vitamin D toxicity related to inadequate monitoring of the fortification process, but underfortification is also a problem. Holick [9] reports that in his study fewer than 30% of milk samples from all sections of the U.S. and British Columbia contained the specified amount, and that 14% to 21% of skim milk samples contained no detectable vitamin D.

VIII. DIETARY CALCIUM DEFICIENCY Conventional wisdom has been that nutritional rickets is primarily due to vitamin D deficiency, although dietary calcium intake modulates the severity and rapidity of onset of the disease [167,168]. However, over the last three decades evidence has been accumulating that implicates low dietary calcium intakes as a cause of rickets in the face of serum 25OHD concentrations within the normal reference range. The subject is also discussed in Chapter 64. Isolated case reports of rickets developing in infants and toddlers, who were placed on very low calcium diets, have been published [169–171]. Their clinical and biochemical presentations were very similar to those of infants with vitamin D deficiency, however in three of the five infants, serum 25OHD and 1,25(OH)2D values were reported to be greater than 9 ng/ml (22.5 nmol/liter) and 118 pg/ml (295 pmol/liter), respectively. In none of the five infants was a therapeutic trial of calcium supplementation of the diet alone tried; however, the clinical and biochemical presentation suggested to the authors that dietary calcium deficiency was the primary factor responsible for the development of rickets. More convincing evidence of dietary calcium deficiency as a cause for rickets in children comes from studies in South Africa [172,173], Nigeria [174–177] and India [162], and possibly Bangladesh [178], where the staple diets of children are characteristically low in calcium because of the lack of readily available dairy products and the low/extra calcium content of the

JOHN M. PETTIFOR

cereals (maize [corn], cassava, yam, rice, and plantain) [42,179–181]. In the South African children, dietary calcium intakes have been estimated to be between 90 and 300 mg/day in those children suffering from rickets compared to between 200 and 500 mg/day in age matched controls [179], while in the Nigerian children, both patients and controls had similar but very low calcium intakes (200 mg/day) [182]. In South Africa, the children typically come from rural areas and present with signs and symptoms of rickets between the ages of 4 and 15 years [181], while in Nigeria they present younger, between 1 and 9 years of age [42]. In the South African series, half the children presented with knock-knees, while the others presented with either bow-legs or wind-swept deformities (Fig. 3). Bow-legs were more common in the Nigerian children, probably reflecting their earlier age of presentation [182]. Unlike vitamin D deficiency, symptoms of muscle weakness are characteristically absent in older children with dietary calcium deficiency. Radiologically, the features are typical of calciopenic rickets with osteopenia and features of hyperparathyroidism being frequent findings (Fig. 4). The severity of the metaphyseal changes is variable. Older children (teenagers) may have no radiologic changes of rickets, despite features of osteomalacia on the iliac crest bone biopsy [183]. Younger children may show evidence of only minor degrees of impaired endochondral calcification, while in others the metaphyseal changes may be quite marked. In a Nigerian study, the degree of severity of radiologic rickets was correlated with serum alkaline phosphatase values [184]. The biochemical features are similar to those of other causes of calciopenic rickets. Hypocalcemia, low urinary calcium excretion, and elevated serum PTH and alkaline phosphatase concentrations are characteristic, while serum phosphorus values are variable and often within the reference range for age [42,180,181]. Serum 25OHD values are normal (mean 16.4 ng/ml and 14.4 ng/ml in the South African and Nigerian children respectively) and 1,25(OH)2D concentrations are elevated [175,176,185]. In a Nigerian study [42] and in the South African children [121], serum osteocalcin levels are similar to those of nonrachitic controls in the majority of patients, although another report from Nigeria found slightly higher levels in rachitic patients than controls [114]. The finding of normophosphatemia and normal renal handling of phosphorus (TmP/GFR) suggests that a peripheral resistance to PTH might be prevalent in this form of rickets. Iliac crest bone biopsies reveal evidence of osteomalacia and hyperparathyroidism in those children who have radiologic features of rickets [173], while in the teenagers without radiologic changes but lower

CHAPTER 65 Nutritional Rickets in Children

1075

FIGURE 4 The radiographic features of dietary calcium deficiency rickets in the lower limbs of a child. The long bones are osteopenic with deformities characteristic of long-standing rickets. The metaphyses show evidence of impaired mineralization and growth arrest lines. FIGURE 3 The clinical presentation of children with dietary calcium deficiency. The deformities are typically more severe in the legs with a predominance of knock-knees or windswept deformities. Upper limb deformities are usually mild if present at all. (Reproduced with permission [181])

limb deformities, the histologic picture varies from that of decreased bone volume, through features of hyperparathyroidism, to frank osteomalacia and hyperparathyroidism [183]. In both the Nigerian and South African studies, clinical, biochemical, and radiologic healing has been achieved through increasing the calcium intake of the children to between 800 and 1500 mg/day without the administration of vitamin D supplements [42,172].

More recently, a study using a calcium supplement of only 350 mg/day, reported complete healing within six months [177]. In a randomised controlled trial, calcium supplements alone or calcium and vitamin D together were equally effective in healing the bone disease and were significantly better than vitamin D therapy alone [175]. In the majority of the South African children, orthopedic corrective surgery has been necessary to correct the deformities of the legs once biochemical and radiologic healing has occurred. This has not been the pattern in the younger Nigerian children with rickets, who have shown remarkable remodelling

1076 and straightening of deformities without orthopedic surgical intervention [177]. The data available from epidemiologic studies conducted in a rural area in South Africa in which a number of the affected children live, suggest that asymptomatic dietary calcium deficiency is prevalent in schoolchildren living in the area. Some 13% of children between the ages of 7 and 12 years were hypocalcemic, 41.5% had elevated alkaline phosphatase concentrations, and 76% had low urinary calcium excretion [186]. It is unclear whether these children have long term sequelae as a result of the poor calcium intakes. However, studies do indicate that asymptomatic children with biochemical abnormalities living in the rural community have lower appendicular bone mass than those with normal biochemistries [179] and that children in the community as a whole have lower appendicular bone mass than their urban peers [187]. Although dietary calcium intakes in children with biochemical changes suggestive of dietary calcium deficiency are very low, it is unclear what role the high phytate or oxalate contents of the diet play in aggravating the symptoms. Nevertheless, biochemical improvement can be achieved by supplementing the children with 500 mg calcium daily [188]. The finding of similarly low dietary calcium intakes in patients with rickets and age matched controls in Nigeria is intriguing [182], as it suggests that other factors besides low dietary calcium intakes might influence the development of rickets in affected children. Such factors might include differing amounts of inhibitors of calcium absorption in the diet, differing growth rates and therefore calcium requirements in the children, or genetic differences that make the rachitic children less able to adapt to low dietary calcium intakes than control subjects. A number of these factors are currently under investigation. A small study has found that there are significant differences in the frequency of vitamin D polymorphisms between affected and control children, but the significance of these findings is unclear at present [189].

IX. THE PATHOGENETIC SPECTRUM OF NUTRITIONAL RICKETS Nutritional rickets has been viewed for some time as being due to an inadequate supply of vitamin D through either an inadequate dietary intake or insufficient skin exposure to ultraviolet radiation, or more recently due to low dietary calcium intake in the face of a normal vitamin D status. However, these pathogenetic concepts are too simplistic. Early studies by Mellanby [190] had shown the effect of cereals in exacerbating the clinical development of vitamin D

JOHN M. PETTIFOR

deficiency rickets in dogs. More recently, studies in baboons have confirmed these findings [170]. The resurgence of rickets and osteomalacia in the Asian community in Great Britain has provided the impetus for detailed studies into the pathogenesis of vitamin D deficiency, and bone disease in that community. Although vitamin D deficiency as assessed by circulating 25OHD concentrations, is the hallmark of the disease in Asians [63,191,192], the mechanisms for the low vitamin D status and the high prevalence of rickets were unclear. It is apparent that the majority of Asians in Britain do not spend less time outdoors than their Caucasian counterparts [193]. Further, although they have darker skins than Caucasians, which might reduce the amount of vitamin D formed in response to sunlight exposure, West Indians living in Britain have even darker skins, yet very few cases of rickets have been described in this ethnic group [55]. Within the Asian community, studies have highlighted the findings that risk factors for the disease include: living at high latitude, Hindu religion, immigration from East Africa, vegetarianism, high fiber diets, and the consumption of chapatti [64,69,70]. The association with vegetarianism, high fiber diets and cereals of high extraction suggests that dietary factors play a role. Support for this comes from two studies that have documented healing of rickets on removing chapattis from the diet [194,195], although this is not a universal finding [68]. Over the past fifteen years, research has shown that both high fiber diets and intestinal malabsorption reduce the serum half-life of 25OHD by approximately one-third [196,197]. Further, experiments in rats have demonstrated that an elevation in serum 1,25(OH)2D concentrations, either by exogenous administration or endogenously through a low calcium diet, increases the metabolic clearance rate of 25OHD without altering its rate of production [198–200]. The fall in serum 25OHD levels could be accounted for by an increase in polar metabolites appearing in the feces. Similar findings have been reported from studies in man [201,202]. Conversely, increasing the calcium content of the diet has been shown to increase serum 25OHD and decrease serum 1,25(OH)2D concentrations [203]. These studies convincingly show that dietary calcium and phytate content influence the catabolism of 25OHD through altering serum 1,25(OH)2D concentrations. In the light of the above studies, Clements [55] has proposed that the low dietary calcium and high phytate diet of the Asian population in Britain increases vitamin D catabolism and vitamin D requirements. In the face of a marginal vitamin D status due to living at high latitude and the low dietary vitamin D content of the diet, the increased catabolism is sufficient to precipitate vitamin D deficiency and clinical rickets and osteomalacia.

1077

CHAPTER 65 Nutritional Rickets in Children

dietary calcium intakes were responsible for rickets in young children while vitamin D deficiency played a major role in adolescents [162].

Thus, nutritional rickets has a spectrum of pathogenetic mechanisms ranging from pure vitamin D deficiency associated with adequate calcium intakes, as might occur in the breast-fed infant, at one end of the spectrum, to pure dietary calcium deficiency with an adequate vitamin D status, as documented in Nigerian and South African rural children, at the other end of the spectrum [204]. In between these two extremes lies the situation exemplified by the Asian community in Britain, where both poor calcium intakes or absorption and marginal vitamin D status combine to lead to frank vitamin D deficiency and rickets (Fig. 5). It is likely that the high prevalence of rickets in vegetarian or immigrant children reported from the U.S. [27,28], Norway [57], Holland [58,59], and a number of tropical and subtropical countries [45] might be due to a mechanism similar to that in the Asian community, while osteomalacia in Bedouin adults in the Middle East reflects mainly dietary calcium deficiency [205]. A number of recent studies have highlighted the complex interaction between vitamin D and calcium intakes in the pathogenesis of nutritional rickets in children. A review of 43 patients diagnosed as having nutritional rickets in New Haven, Connecticut, found low 25OHD levels in only 22%, and the majority of infants had been weaned onto diets with minimal dairy content [51]. The authors concluded that low dietary calcium intakes probably played a major role in the pathogenesis of the disease. Similar findings are reported from India, where it is suggested that low

X. CONCLUSIONS Despite readily accessible and effective means to eradicate rickets globally, the disease remains a major public health problem in many countries, not only in temperate regions of the world but also in tropical and subtropical countries. In many developed countries, the promotion of exclusive breast-feeding during the first six months of life and the concerns about the long-term effect of sunlight exposure during this period have exacerbated the risks of vitamin D deficiency in the young infant. In some subtropical countries, social customs play an important role in preventing adequate vitamin D status not only in the young infant but also in the pregnant and lactating mother. In a number of developing countries, low dietary calcium intakes appear to play a major role in the pathogenesis of rickets in older children. Recent studies have helped to provide an all embracing concept of the interaction of vitamin D and calcium intakes in the pathogenesis of rickets. There remains a need for international agencies to place the eradication of vitamin D deficiency among young children in many parts of the world as a priority. Nutritional rickets not only leads to an increase in infant mortality but also has serious long-term health sequelae.

Lack of UV Light Inadequate dietary vitamin D

25OHD catabolism

Low 25OHD Low 1,25(OH)2D

1,25(OH)2D

Impaired calcium absorption Dietary calcium deficiency

High dietary phytate/ low dietary calcium

Indequate calcium absorption for requirement of growing child

Serum ionized calcium

PTH Serum phosphate

Impaired mineralization

RICKETS

FIGURE 5

The spectrum of nutritional rickets. At either ends of the pathogenetic spectrum are vitamin D deficiency and dietary calcium deficiency. In between lie combinations in varying degrees of relative vitamin D insufficiency and decreased dietary calcium content or bioavailability.

1078 References 1. Arneil GC 1975 Nutritional rickets in children in Glasgow. Proc Nutr Soc 34:101–109. 2. Chick DH 1976 Study of rickets in Vienna 1919–1922. Med Hist 20:41–51. 3. Rafter GW 1987 Elmer McCollum and the disappearance of rickets. Perspect Biol Med 30:527–534. 4. Stephen JML 1975 Epidemiological and dietary aspects of rickets and osteomalacia. Proc Nutr Soc 34:131–138. 5. Rudolf M, Arulanantham K, Greenstein RM 1980 Unsuspected nutritional rickets. Pediatrics 66:72–76. 6. Jacobus CH, Holick MF, Shao Q, Chen TC, Holm IA, Kolodny JM, Fuleihan GE-H, Seeley EW 1992 Hypervitaminosis D associated with drinking milk. N Engl J Med 326:1173–1177. 7. Markestad T, Hesse V, Siebenhuner M, Jahreis G, Aksnes L, Plenert W, Aarskog D 1987 Intermittent high-dose vitamin D prophylaxis during infancy: effect on vitamin D metabolites, calcium, and phosphorus. Am J Clin Nutr 46:652–658. 8. Portale AA, Halloran BP, Harris ST, Bikle DD, Morris RC 1992 Metabolic acidosis reverses the increase in serum 1,25(OH)2D in phosphorus-restricted normal men. Am J Physiol 263:E1164–E1170. 9. Holick MF 1995 Environmental factors that influence the cutaneous production of vitamin D. Am J Clin Nutr 61: 638S–645S. 10. Salimpour R 1975 Rickets in Tehran. Arch Dis Child 50: 63–66. 11. Al Hag AI, Karrar ZA 1995 Nutritional vitamin D deficiency rickets in Sudanese children. Ann Trop Paediatr 15: 69–76. 12. Hillman LS, Haddad JG 1974 Human perinatal vitamin D metabolism 1:25-hydroxyvitamin D in maternal and cord blood. J Pediatr 84:742–749. 13. Gertner JM, Glassman MS, Coutsan DR, Goodman DBP 1980 Fetomaternal vitamin D relationships at term. J Pediatr 97:637–640. 14. Andiran N, Yordam N, Ozon A 2002 Risk factors for vitamin D deficiency in breast-fed newborns and their mothers. Nutrition 18:47–50. 15. Rothberg AD, Pettifor JM, Cohen DF, Sonnendecker EWW, Ross FP 1982 Maternal-infant vitamin D relationships during breast-feeding. J Pediatr 101:500–503. 16. Maxwell JP 1934 Further studies in adult rickets (osteomalacia) and fetal rickets. Proc Roy Soc Med 28:265–300. 17. Anatoliotaki M, Tsilimigaki A, Tsekoura T, Schinaki A, Stefanaki S, Nikolaidou P 2003 Congenital rickets due to maternal vitamin D deficiency in a sunny island of Greece. Acta Paediatr 92:389–391. 18. Mohapatra A, Sankaranarayanan K, Kadam SS, Binoy S, Kanbur WA, Mondkar JA 2003 Congenital rickets. J Trop Pediatr 49:126–127. 19. Maiyegun SO, Malek AH, Devarajan LV, Dahniya MH 2002 Severe congenital rickets secondary to maternal hypovitaminosis D: a case report. Ann Trop Paediatr 22:191–195. 20. Moncrieff M, Fadahunsi TO 1974 Congenital rickets due to maternal vitamin D deficiency. Arch Dis Child 49:810–811. 21. Zhou H 1991 Rickets in China. In: Glorieux FH (ed), Rickets. Nestec, Vevey: Raven Press, New York, pp. 253–261. 22. Chen Y 1994 Prematurity as a predictor of rickets in Shanghai infants. Public Health 108:333–339. 23. Lapatsanis P, Deliyanni V, Doxiadis S 1968 Vitamin D deficiency rickets in Greece. J Pediatr 73:195–202.

JOHN M. PETTIFOR

24. Fraser D, Kooh SW, Scriver CR 1967 Hyperparathyroidism as the cause of hyperaminoaciduria and phosphaturia in human vitamin D deficiency. Pediat Res 1:425–435. 25. Doxiadis S, Angelis C, Karatzas P, Vrettos C, Lapatsanis P 1976 Genetic aspects of nutritional rickets. Arch Dis Child 51:83–90. 26. Dancaster CP, Jackson WPU 1961 Studies in rickets in the Cape Peninsula II. Aetiology. S Afr Med J 35:890–894. 27. Edidin DV, Levitsky LL, Schey W, Dumbuvic N, Campos A 1980 Resurgence of nutritional rickets associated with breastfeeding and special dietary practices. Pediatrics 65:232–235. 28. Bachrach S, Fisher J, Parks JS 1979 An outbreak of vitamin D deficiency rickets in a susceptible population. Pediatrics 64: 871–877. 29. Elidrissy ATH, Sedrani SH, Lawson DEM 1984 Vitamin D deficiency in mothers of rachitic infants. Calcif Tissue Int 36:266–268. 30. Hollis BW, Roos BA, Draper HH, Lambert PW 1981 Vitamin D and its metabolites in human and bovine milk. J Nutr 111: 1240–1248. 31. Specker BL, Tsang RC, Hollis BW 1985 Effect of race and diet on human-milk vitamin D and 25-hydroxyvitiamin D. Am J Dis Child 139:1134–1137. 32. Ala-Houhala M, Koskinen T, Terho A, Kiovula T, Visakorpi J 1986 Maternal compared with infant vitamin D supplementation. Arch Dis Child 61:1159–1163. 33. Specker BL, Valanis B, Hertzberg V, Edwards N, Tsang RC 1985 Sunshine exposure and serum 25-hydroxyvitamin D concentrations in exclusively breast-fed infants. J Pediatr 107:372–376. 34. Specker BL, Tsang RC 1987 Cyclical serum 25-hydroxyvitamin D concentrations paralleling sunshine exposure in exclusively breast-fed infants. J Pediatr 110:744–747. 35. Olivieri MB, Ladizesky M, Mautalen CA, Alonso A, Martinez L 1993 Seasonal variations of 25 hydroxyvitamin D and parathyroid hormone in Ushuaia (Argentina), the southernmost city of the world. Bone Miner 20:99–108. 36. Harris SS, Dawson-Hughes B 1998 Seasonal changes in plasma 25-hydroxyvitamin D concentrations of young American black and white women. Am J Clin Nutr 67: 1232–1236. 37. Garabedian M, Ben-Mekhbi H 1991 Is vitamin D–deficiency rickets a public health problem in France and Algeria? In: Glorieux FH (ed) Rickets. Nestec, Vevey; Raven Press, New York, pp. 215–221. 38. Ladizesky M, Lu Z, Oliveri B, Roman NS, Diaz S, Holick MF, Mautalen C 1995 Solar ultraviolet B radiation and photoproduction of vitamin D3 in central and southern areas of Argentina. J Bone Miner Res 10:545–549. 39. Beser E, Cakmakci T 1994 Factors affecting the morbidity of vitamin D deficiency rickets and primary protection. East Afr Med J 71:358–362. 40. Ghai OP, Koul PB 1991 Rickets in India. In: Glorieux FH (ed) Rickets. Nestec, Vevey: Raven Press, New York, pp. 247–252. 41. Harris NS, Crawford PB, Yangzom Y, Pinzo L, Gyaltsen P, Hudes M 2001 Nutritional and health status of Tibetan children living at high altitudes. N Engl J Med 344:341–347. 42. Okonofua F, Gill DS, Alabi ZO, Thomas M, Bell JL, Dandona P 1991 Rickets in Nigerian children: a consequence of calcium malnutrition. Metabolism 40:209–213. 43. Laditan AAO, Adeniyi A 1975 Rickets in Nigerian children: Response to vitamin D. J Trop Med Hyg 78:206–209. 44. Lulseged S, Fitwi G 1999 Vitamin D deficiency rickets: socio-demographic and clinical risk factors in children seen

CHAPTER 65 Nutritional Rickets in Children

45. 46. 47. 48.

49. 50. 51.

52. 53. 54.

55. 56. 57. 58.

59.

60. 61. 62. 63.

64. 65.

at a referral hospital in Addis Ababa. East Afr Med J 76:457–461. Bhattacharyya AK 1992 Nutritional rickets in the tropics. In: Simopoulos AP (ed) Nutritional triggers for health and in disease. Karger, Basel, pp. 140–197. Karrar ZA 1998 Vitamin D deficiency rickets in developing countries. Ann Trop Paediatr 18(Suppl):S89–S92. Bereket A 2003 Rickets in developing countries. Endocr Dev 6:220–232. Agarwal KS, Mughal MZ, Upadhyay P, Berry JL, Mawer EB, Puliyel JM 2002 The impact of atmospheric pollution on vitamin D status of infants and toddlers in Delhi, India. Arch Dis Child 87:111–113. Finberg L 1979 Human choice, vegetable deficiencies, and vegetarian rickets. Am J Dis Child 133:129. Chesney RW 2002 Rickets: the third wave. Clin Pediatr (Phila) 41:137–139. DeLucia MC, Mitnick ME, Carpenter TO 2003 Nutritional rickets with normal circulating 25-hydroxyvitamin D: a call for reexamining the role of dietary calcium intake in North American infants. J Clin Endocrinol Metab 88:3539–3545. McCaffree J 2001 Rickets on the rise. J Am Diet Assoc 101:16–17. Dwyer JT, Dietz WH, Hass G, Suskind R 1979 Risk of nutritional rickets among vegetarian children. Am J Dis Child 133:134–140. Kreiter SR, Schwartz RP, Kirkman HN Jr, Charlton PA, Calikoglu AS, Davenport ML 2000 Nutritional rickets in African American breast-fed infants. J Pediatr 137: 153–157. Clements MR 1989 The problem of rickets in UK Asians. J Hum Nutr Diet 2:105–116. Pedersen P, Michaelsen KF, Molgaard C 2003 Children with nutritional rickets referred to hospitals in Copenhagen during a 10-year period. Acta Paediatr 92:87–90. Hellebostad M, Markestad T, Halvorsen KS 1985 Vitamin D deficiency rickets and vitamin B12 deficiency in vegetarian children. Acta Paediatr Scand 74:191–195. Meulmeester JF, van den Berg H, Wedel M, Boshuis PG, Hulshof KFAM, Luyken R 1990 Vitamin D status, parathyroid hormone, and sunlight in Turkish, Moroccan, and Caucasian children in The Netherlands. Eur J Clin Nutr 44:461–470. Dagnelie PC, Vergote FJVRA, van Staveren WA, van den Berg H, Dingjan PG, Hautvast JGAJ 1990 High prevalence of rickets in infants on macrobiotic diets. Am J Clin Nutr 51:202–208. Nozza JM, Rodda CP 2001 Vitamin D deficiency in mothers of infants with rickets. Med J Aust 175:253–255. Mason RS, Diamond TH 2001 Vitamin D deficiency and multicultural Australia. Med J Aust 175:236–237. Blok BH, Grant CC, McNeil AR, Reid IR 2000 Characteristics of children with florid vitamin D deficient rickets in the Auckland region in 1998. NZ Med J 113:374–376. Ford JA, McIntosh WB, Butterfield R, Preece MA, Pietrek J, Arrow-Smith WA, Arthurton MW, Turner W, O’Riordan JLH, Dunnigan MG 1976 Clinical and subclinical vitamin D deficiency in Bradford children. Arch Dis Child 51:939–943. Finch PJ, Ang L, Eastwood JB, Maxwell JD 1992 Clinical and histological spectrum of osteomalacia among Asians in South London. Quart J Med 83:439–448. Datta S, Alfaham M, Davies DP, Dunstan F, Woodhead S, Evans J, Richards B 2002 Vitamin D deficiency in pregnant women from a non-European ethnic minority population— an interventional study. Brit J Obstet Gynaec 109:905–908.

1079 66. Moncrieff MW, Lunt HRW, Arthur LJH 1973 Nutritional rickets at puberty. Arch Dis Child 48:221–224. 67. Hodgkin P, Kay GH, Hine PM, Lumb GA, Stanbury SW 1973 Vitamin-D deficiency in Asians at home and in Britain. Lancet ii:167–172. 68. Dent CE, Round JM, Rowe DJF, Stamp TCB 1973 Effect of chapattis and ultraviolet irradiation on nutritional rickets in an Indian immigrant. Lancet i:1282–1284. 69. Henderson JB, Dunnigan MG, McIntosh WB, Abdul-Motaal A, Hole D 1990 Asian osteomalacia is determined by dietary factors when exposure to ultraviolet radiation is restricted: a risk factor model. Quart J Med 76:923–933. 70. Henderson JB, Dunnigan MG, McIntosh WB, Abdul-Motaal AA, Gettinby G, Glekin BM 1987 The importance of limited exposure to ultraviolet radiation and dietary factors in the aetiology of Asian rickets: a risk-factor model. Quart J Med 63:413–425. 71. Bonnici F 1978 Functional hypoparathyroidism in infantile hypocalcaemic stage I vitamin D deficiency rickets. S Afr Med J 54:611–612. 72. Johnson GH, Willis F 2003 Seizures as the presenting feature of rickets in an infant. Med J Aust 178: 467–468. 73. Buchanan N, Pettifor JM, Cane RD, Bill PLA 1978 Infantile apnoea due to profound hypocalcaemia associated with vitamin D deficiency. S Afr Med J 53:766–767. 74. Salaria M, Poddar B, Parmar V 2001 Rickets presenting as pseudotumour cerebri and seizures. Indian J Pediatr 68:181. 75. Hochman HI, Mejlszenkier JD 1977 Cataracts and pseudotumor cerebri in an infant with vitamin D–deficiency rickets. J Pediatr 90:252–254. 76. Park EA 1954 The influence of severe illness on rickets. Arch Dis Child 29:369–380. 77. Harrison HE, Harrison HC 1979 Rickets and osteomalacia. In: Disorders of calcium and phosphate metabolism in childhood and adolescence. W.B. Saunders, Philadelphia, pp. 141–256. 78. Geggel RL, Pereira GR, Spackman TJ 1978 Fractured ribs: Unusual presentation of rickets in premature infants. J Pediatr 93:680–682. 79. Opie WH, Muller CJB, Kamfer H 1975 The diagnosis of vitamin D deficiency rickets. Pediat Radiol 3:105–110. 80. Pettifor JM, Pentopoulos M, Moodley GP, Isdale JM, Ross FP 1984 Is craniotabes a pathognomonic sign of rickets in 3-month-old infants? S Afr Med J 65:549–551. 81. Dancaster CP, Jackson WPU 1960 Studies in rickets in the Cape Peninsula 1. Cranial softening in a colored population and its relationship to the radiological and biochemical changes of rickets. S Afr Med J 34:776–780. 82. Reilly BJ, Leeming JM, Fraser D 1964 Craniosynostosis in the rachitic spectrum. J Pediatr 64:396–405. 83. Purvis RJ, Barrie MWJ, MacKay GS, Wilkinson EM, Cockburn F, Belton NR, Forfar JO 1973 Enamel hypoplasia of the teeth associated with neonatal tetany: a manifestation of maternal vitamin–D deficiency. Lancet ii:811–814. 84. Schott GD, Wills MR 1976 Muscle weakness in osteomalacia. Lancet i:626–629. 85. Glerup H, Mikkelsen K, Poulsen L, Hass E, Overbeck S, Andersen H, Charles P, Eriksen EF 2000 Hypovitaminosis D myopathy without biochemical signs of osteomalacic bone involvement. Calcif Tissue Int 66:419–424. 86. Price DI, Stanford LC, Braden DS, Ebeid MR, Smith JC 2003 Hypocalcemic rickets: An unusual cause of dilated cardiomyopathy. Pediatr Cardiol 24:510–512.

1080 87. Olgun H, Ceviz N, Ozkan B 2003 A case of dilated cardiomyopathy due to nutritional vitamin D deficiency rickets. Turk J Pediatr 45:152–154. 88. Uysal S, Kalayci AG, Baysal K 1999 Cardiac functions in children with vitamin D deficiency rickets. Pediatr Cardiol 20:283–286. 89. Manolagas SC, Yu XP, Girasole G, Bellido T 1994 Vitamin D and the hematolymphopoietic tissue: a 1994 update. Semin Nephrol 14:129–143. 90. DeLuca HF, Cantorna MT 2001 Vitamin D: its role and uses in immunology. FASEB J 15:2579–2585. 91. Stroder J, Kasal P 1970 Evaluation of phagocytosis in rickets. Acta Paediatr Scand 59:288–292. 92. Lorente F, Fontan G, Jara P, Casas C, Garcia-Rodriguez MC, Ojeda JA 1976 Defective neutrophil motility in hypovitaminosis D rickets. Acta Paediatr Scand 65:695–699. 93. Atiq M, Fadoo Z, Naz F, Khurshid M 1999 Myelofibrosis in severe vitamin D deficiency rickets. J Pak Med Assoc 49: 174–177. 94. Gruner BA, DeNapoli TS, Elshihabi S, Britton HA, Langevin AM, Thomas PJ, Weitman SD 2003 Anemia and hepatosplenomegaly as presenting features in a child with rickets and secondary myelofibrosis. J Pediatr Hematol Oncol 25:813–815. 95. Yetgin S, Ozoylu S 1982 Myeloid metaplasia in vitamin D deficiency rickets. Scand J Haematol 28:180–185. 96. David L 1991 Common vitamin D-deficiency rickets. In: Glorieux FH (ed) Rickets. Nestec, Vevey: Raven Press, New York, pp. 107–122. 97. Suda T 1987 Cellular mechanisms of fusion of hemopoietic cells induced by 1α,25 dihydroxyvitamin D3. In: Cohn, DV, Martin TJ, Meunier P J (eds) Calcium regulation and bone metabolism: basic and clinical aspects. Excerpta Medica, Amsterdam, pp. 363–370. 98. Haddad JG, Chyu KJ 1971 Competitive protein binding radioassay for 25-hydroxycholecalciferol. J Clin Endocrinol 33:992–995. 99. Pettifor JM, Ross FP, Moodley GP, Margo G 1978 Serum calcium, magnesium, phosphorus, alkaline phosphatase and 25-hydroxyvitamin D concentrations in a paediatric population. S Afr Med J 53:751–754. 100. McLaughlin M, Raggatt PR, Fairney A, Brown DJ, Lester E, Wills MR 1974 Seasonal variation in serum 25-hydroxycholecalciferol in healthy people. Lancet i:536–538. 101. Stanbury SW, Taylor CM, Lumb GA, Mawer EB, Berry J, Hann J, Wallace J 1981 Formation of vitamin D metabolites following correction of human vitamin D deficiency: observations in patients with nutritional osteomalacia. Miner Electrolyte Metab 5:212–227. 102. Mawer EB, Backhouse J, Hill LF, Lumb GA, De Silva P, Taylor CM, Stanbury SW 1975 Vitamin D metabolism and parathyroid function in man. Clin Sci Mol Med 48:349–365. 103. Gupta MM, Round JM, Stamp TCB 1974 Spontaneous cure of vitamin-D deficency in Asians during summer in Britain. Lancet i:586–588. 104. Garabedian M, Vainsel M, Mallet E, Guillozo H, Toppet M, Grimberg R, NGuyen TM, Balsan S 1983 Circulating vitamin D metabolite concentrations in children with nutritional rickets. J Pediatr 103:381–386. 105. Arnaud SB, Stickler GB, Haworth JC 1976 Serum 25-hydroxyvitamin D in infantile rickets. Pediatrics 57:221–225. 106. Goel KM, Sweet EM, Logan RW, Warren JM, Arneil GC, Shanks RA 1976 Florid and subclinical rickets among immigrant children in Glasgow. Lancet i:1141–1145.

JOHN M. PETTIFOR

107. Markestad T, Halvorsen S, Seeger Halvorsen K, Aksnes L, Aarskog D 1984 Plasma concentrations of vitamin D metabolites before and during treatment of vitamin D deficiency rickets in children. Acta Paediatr Scand 73:225–231. 108. Srivastava T, Alon US 2002 Stage I vitamin D-deficiency rickets mimicking pseudohypoparathyroidism type II. Clin Pediatr (Phila) 41:263–268. 109. Taitz LS and de Lacy CD 1962 Parathyroid function in vitamin D deficiency rickets 1. Phosphorus excretion index in vitamin D deficiency rickets in South African bantu infants. Pediatrics 30:875–883. 110. Muldowney FP, Freaney R, McGeeney D 1968 Renal tubular acidosis and amino-aciduria in osteomalacia of dietary or intestinal origin. Quart J Med 37:517–539. 111. Kruse K 1995 Pathophysiology of calcium metabolism in children with vitamin D-deficiency rickets. J Pediatr 126: 736–741. 112. Taitz LS, de Lacy CD 1962 Parathyroid function in vitamin D deficiency rickets II. The relationship of parathyroid function to bone changes and incidence of tetany in vitamin D deficiency rickets in South African bantu infants. Pediatrics 30:884–892. 113. Stanbury SW, Torkington P, Lumb GA, Adams PH, De Silva P, Taylor CM 1975 Asian rickets and osteomalacia: patterns of parathyroid response in vitamin D deficiency. Proc Nutr Soc 34:111–117. 114. Scariano JK, Walter EA, Glew RH, Hollis BW, Henry A, Ocheke I, Isichei CO 1995 Serum levels of the pyridinoline crosslinked carboxyterminal telopeptide of type I collagen (ICTP) and osteocalcin in rachitic children in Nigeria. Clin Biochem 28:541–545. 115. Baroncelli GI, Bertelloni S, Ceccarelli C, Amato V, Saggese G 2000 Bone turnover in children with vitamin D deficiency rickets before and during treatment. Acta Paediatrica 89: 513–518. 116. Editorial 1971 Diagnosis of nutritional rickets. Lancet ii:28–29. 117. Nawawi H, Girgis SI 2002 Serum levels of bone-specific alkaline phosphatase and procollagen type I carboxyterminal peptide in vitamin D deficiency. Southeast Asian J Trop Med Public Health 33(Suppl 2):124–130. 118. Soylu H, Aras S, Kutlu NO, Egri M, Sazak S 2001 Urinary free deoxypyridinoline assessment in recognition of rickets. J Trop Pediatr 47:186–187. 119. Calvo S, Eyre DR, Gundberg CM 1996 Molecular basis and clinical application of biological markers of bone turnover. Endocr Rev 17:333–368. 120. Cole DEC, Carpenter TO, Gundberg CM 1985 Serum osteocalcin concentrations in children with metabolic bone disease. J Pediatr 106:770–776. 121. Daniels ED, Pettifor JM, Moodley GP 2000 Serum osteocalcin has limited usefulness as a diagnostic marker for rickets. Eur J Pediatr 159:730–733. 122. Greig F, Casas J, Castells S 1989 Changes in plasma osteocalcin concentrations during treatment of rickets. J Pediatr 114:820–823. 123. Eastwood JB, de Wardener HE, Gray RW, Lemann JR Jr 1979 Normal plasma-1,25-(OH)2-vitamin–D concentrations in nutritional osteomalacia. Lancet i:1377–1378. 124. Chesney RW, Zimmerman J, Hamstra A, DeLuca HF, Mazess RB 1981 Vitamin D metabolite concentrations in vitamin D deficiency. Am J Dis Child 135:1025–1028. 125. NGuyen TM, Guillozo H, Garabedian M, Mallet E, Balsan S 1979 Serum concentrations of 24,25-dihydroxyvitamin D in

1081

CHAPTER 65 Nutritional Rickets in Children

126. 127.

128.

129. 130. 131. 132.

133. 134. 135. 136.

137.

138.

139.

140.

141. 142. 143. 144.

normal children and in children with rickets. Pediat Res 13:973–976. Rosen JF, Chesney RW 1983 Circulating calcitriol concentrations in health and disease. J Pediatr 103:1–17. Rasmussen H, Baron R, Broadus A, DeFronzo R, Lang R, Horst R 1980 1,25(OH)2D3 is not the only D metabolite involved in the pathogenesis of osteomalacia. Am J Med 69:360–368. Arnaud CD 1991 Parathyroid hormone and its role in the pathophysiology of the common forms of rickets and osteomalacia. In: Glorieux FH (ed) Rickets. Nestec, Vevey: Raven Press, New York, pp. 47–61. Rasmussen H, De Luca H, Arnaud CD, Hawker C, Von Stedingk M 1963 The relationship between vitamin D and parathyroid hormone. J Clin Invest 42:1940–1946. Arnaud CD, Rasmussen H, Anast C 1966 Further studies on the interrelationship between parathyroid hormone and vitamin D. J Clin Invest 45:1955–1964. Richards IDG, Sweet EM, Arneil GC 1968 Infantile rickets persists in Glasgow. Lancet i:803–805. Pettifor JM 1991 Calcium, phosphorus, and vitamin D. In: Ballabriga A, Brusner O, Dobbing J, Gracey M, Senterre J (eds) Clinical nutrition of the young child. Nestec, Vevey: Raven Press, New York, pp. 497–516. Pettifor JM, Isdale JM, Sahakian J, Hansen JDL 1980 Diagnosis of subclinical rickets. Arch Dis Child 55:155–157. Hunter GJ, Schneidau A, Hunter JV, Chapman M 1984 Rickets in adolescence. Clin Radiol 35:419–421. Shah BR and Finberg L 1994 Single-day therapy for nutritional vitamin D–deficiency rickets: a preferred method. J Pediatr 125:487–490. Lubani MM, Al-Shab TS, Al-Saleh QA, Sharda DC, Quattawi SA, Ahmed SAH, Moussa MA, Reavey PC 1989 Vitamin D–deficiency rickets in Kuwait: the prevalence of a preventable disease. Ann Trop Paediatr 3:134–139. Cesur Y, Caksen H, Gundem A, Kirimi E, Odabas D 2003 Comparison of low and high dose of vitamin D treatment in nutritional vitamin D deficiency rickets. J Pediatr Endocrinol Metab 16:1105–1109. Kutluk G, Cetinkaya F, Basak M 2002 Comparisons of oral calcium, high dose vitamin D, and a combination of these in the treatment of nutritional rickets in children. J Trop Pediatr 48:351–353. Greer FR, Searcy JE, Levin RS, Steichen JJ, Steichen-Asche PS, Tsang RC 1982 Bone mineral content and serum 25-hydroxyvitamin D concentrations in breast-fed infants with or without supplemental vitamin D: One-year follow-up. J Pediatr 100:919–922. Ala-Houhala M 1985 25-hydroxyvitamin D levels during breast-feeding with or without maternal or infantile supplementation of vitamin D. J Pediatr Gastroenterol Nutr 4: 220–226. Roberts CC, Chan GM, Folland D, Rayburn C, Jackson R 1981 Adequate bone mineralization in breast-fed infants. J Pediatr 99:192–196. Birkbeck JA, Scott HF 1980 25-hydroxycholecalciferol serum levels in breast-fed infants. Arch Dis Child 50:691–695. Bassir M, Laborie S, Lapillonne A, Claris O, Chappuis MC, Salle BL 2001 Vitamin D deficiency in Iranian mothers and their neonates: a pilot study. Acta Paediatr 90:577–579. Dawodu A, Agarwal M, Hossain M, Kochiyil J, Zayed R 2003 Hypovitaminosis D and vitamin D deficiency in exclusively breast-feeding infants and their mothers in

145. 146.

147. 148. 149. 150. 151. 152. 153.

154. 155. 156. 157. 158. 159.

160.

161.

162.

163.

summer: a justification for vitamin D supplementation of breast-feeding infants. J Pediatr 142:169–173. Wharton B, Bishop N 2003 Rickets. Lancet 362:1389–1400. Standing Committee on the Scientific Evaluation of Dietary Reference Intakes 1997 Dietary reference intakes for calcium, phosphorus, magnesium, vitamin D, fluoride. National Academy Press, Washington. Department of Health 1998 Nutrition and bone health: with particular reference to calcium and vitamin D. Her Majesty’s Stationery Office, London, p. 115. Gartner LM, Greer FR 2003 Prevention of rickets and vitamin D deficiency: new guidelines for vitamin D intake. Pediatrics 111:908–910. Finberg L 1981 Human milk feeding and vitamin D supplementation. J Pediatr 99:228–229. Ozsoylu S, Hansaoglu A 1981 25-hydroxycholecalciferol levels in breast-fed infants. Arch Dis Child 56:318. Editorial 1973 The need for vitamin-D supplements. Lancet i:1097–1098. Specker BL, Ho ML, Oestreich A, Yin T, Shui Q, Chen X, Tsang RC 1992 Prospective study of vitamin D supplementation and rickets in China. J Pediatr 120:733–739. Zeghoud F, Ben-Mekhbi H, Djeghri N, Garabedian M 1994 Vitamin D prophylaxis during infancy: comparison of the long-term effects of three intermittent doses (15, 5, or 2.5 mg) on 25-hydroxyvitamin D concentrations. Am J Clin Nutr 60:393–396. Chesney RW 2003 Rickets: An old form for a new century. Pediatr Int 45:509–511. Atiq M, Suria A, Nizami SQ, Ahmed I 1998 Vitamin D status of breastfed Pakistani infants. Acta Paediatr 87:737–740. Al Jurayyan NA, El Desouki ME, Al Herbish AS, Al Mazyad AS, Al Qhtani MM 2002 Nutritional rickets and osteomalacia in school children and adolescents. Saudi Med J 23:182–185. Narchi H 2000 Case-control study of diet and sun exposure in adolescents with symptomatic rickets. Ann Trop Paediatr 20:217–221. Scharla SH 1998 Prevalence of subclinical vitamin D deficiency in different European countries. Osteoporos Int (Suppl 8):S7–S12. Cheng S, Tylavsky F, Kroger H, Karkkainen M, Lyytikainen A, Koistinen A, Mahonen A, Alen M, Halleen J, Vaananen K, Lamberg-Allardt C 2003 Association of low 25-hydroxyvitamin D concentrations with elevated parathyroid hormone concentrations and low cortical bone density in early pubertal and prepubertal Finnish girls. Am J Clin Nutr 78: 485–492. Outila TA, Karkkainen MU, Lamberg-Allardt CJ 2001 Vitamin D status affects serum parathyroid hormone concentrations during winter in female adolescents: associations with forearm bone mineral density. Am J Clin Nutr 74:206–210. Guillemant J, Le HT, Maria A, Allemandou A, Peres G, Guillemant S 2001 Wintertime vitamin D deficiency in male adolescents: effect on parathyroid function and response to vitamin D3 supplements. Osteoporos Int 12:875–879. Balasubramanian K, Rajeswari J, Gulab, Govil YC, Agarwal AK, Kumar A, Bhatia V 2003 Varying role of vitamin D deficiency in the etiology of rickets in young children vs. adolescents in northern India. J Trop Pediatr 49:201–206. Rajeswari J, Balasubramanian K, Bhatia V, Sharma VP, Agarwal AK 2003 Aetiology and clinical profile of osteomalacia in adolescent girls in northern India. Natl Med J India 16:139–142.

1082 164. Du X, Greenfield H, Fraser DR, Ge K, Trube A, Wang Y 2001 Vitamin D deficiency and associated factors in adolescent girls in Beijing. Am J Clin Nutr 74:494–500. 165. Calvo MS, Whiting SJ 2003 Prevalence of vitamin D insufficiency in Canada and the United States: importance to health status and efficacy of current food fortification and dietary supplement use. Nutr Rev 61:107–113. 166. Pietrek J, Preece MA, Windo J, O’Riordan JLH, Dunnigan MG, McIntosh WB, Ford JA 1976 Prevention of vitamin-D deficiency in Asians. Lancet i:1145–1148. 167. Walker ARP 1953 Does a low intake of calcium cause or promote the development of rickets? Am J Clin Nutr 3:114–120. 168. Irwin MI, Kienholz EW 1973 A conspectus of research on calcium requirements of man. J Nutr 103:1020–1095. 169. Maltz HE, Fish MB, Holliday MA 1970 Calcium deficiency rickets and the renal response to calcium infusion. Pediatrics 46:865–870. 170. Sly MR, van der Walt WH, Du Bruyn D, Pettifor JM, Marie PJ 1984 Exacerbation of rickets and osteomalacia by maize: a study of bone histomorphometry and composition in young baboons. Calcif Tissue Int 36:370–379. 171. Proesman W, Legius E, Eggermont E 1988 Rickets due to calcium deficiency. Mary Ann Liebert, New York, p. 15. 172. Pettifor JM, Ross P, Wang J, Moodley G, Couper-Smith J 1978 Rickets in children of rural origin in South Africa: is low dietary calcium a factor? J Pediatr 92:320–324. 173. Marie PJ, Pettifor JM, Ross FP, Glorieux FH 1982 Histological osteomalacia due to dietary calcium deficiency in children. N Engl J Med 307:584–588. 174. Thacher TD, Ighogboja SI, Fischer PR 1997 Rickets without vitamin D deficiency in Nigerian children. Ambulatory Child Health 3:56–64. 175. Thacher TD, Fischer PR, Pettifor JM, Lawson JO, Isichei CO, Reading JC, Chan GM 1999 A comparison of calcium, vitamin D, or both for nutritional rickets in Nigerian children. N Engl J Med 341:563–568. 176. Oginni LM, Worsfold M, Oyelami OA, Sharp CA, Powell DE, Davie MW 1996 Etiology of rickets in Nigerian children. J Pediatr 128:692–694. 177. Oginni LM, Sharp CA, Badru OS, Risteli J, Davie MW, Worsfold M 2003 Radiological and biochemical resolution of nutritional rickets with calcium. Arch Dis Child 88:812–817. 178. Fischer PR, Rahman A, Cimma JP, Kyaw-Myint TO, Kabir AR, Talukder K, Hassan N, Manaster BJ, Staab DB, Duxbury JM, Welch RM, Meisner CA, Haque S, Combs GF Jr 1999 Nutritional rickets without vitamin D deficiency in Bangladesh. J Trop Pediatr 45:291–293. 179. Eyberg C, Pettifor JM, Moodley G 1986 Dietary calcium intake in rural black South African children. The relationship between calcium intake and calcium nutritional status. Hum Nutr Clin Nutr 40C:69–74. 180. Bhimma R, Pettifor JM, Coovadia HM, Moodley M, Adhikari M 1995 Rickets in black children beyond infancy in Natal. S Afr Med J 85:668–672. 181. Pettifor JM 1991 Dietary calcium deficiency. In: Glorieux FH (ed) Rickets. Nestec, Vevey: Raven Press, New York, pp. 123–143. 182. Thacher TD, Fischer PR, Pettifor JM, Lawson JO, Isichei C, Chan GM 2000 Case-control study of factors associated with nutritional rickets in Nigerian children. J Pediatr 137: 367–373.

JOHN M. PETTIFOR

183. Schnitzler CM, Pettifor JM, Patel D, Mesquita JM, Moodley GP, Zachen D 1994 Metabolic bone disease in black teenagers with genu valgum or varum without radiologic rickets: a bone histomorphometric study. J Bone Miner Res 9:479–486. 184. Thacher TD, Fischer PR, Pettifor JM, Lawson JO, Manaster BJ, Reading JC 2000 Radiographic scoring method for the assessment of the severity of nutritional rickets. Journal of Tropical Pediatrics 46:132–139. 185. Pettifor JM, Ross FP, Travers R, Glorieux FH, DeLuca HF 1981 Dietary calcium deficiency: a syndrome associated with bone deformities and elevated serum 1,25-dihydroxyvitamin D concentrations. Metab Bone Rel Res 2:301–305. 186. Pettifor JM, Ross FP, Moodley GP, Shuenyane E 1979 Calcium deficiency in rural black children in South Africa— a comparison between rural and urban communities. Am J Clin Nutr 32:2477–2483. 187. Pettifor JM, Moodley GP 1997 Appendicular bone mass in children with a high prevalence of low dietary calcium intakes. J Bone Miner Res, 12:1824–1832. 188. Pettifor JM, Ross FP, Moodley GP, Shuenyane E 1981 The effect of dietary calcium supplementation on serum calcium, phosphorus, and alkaline phosphatase concentrations in a rural black population. Am J Clin Nutr 34:2187–2191. 189. Fischer PR, Thacher TD, Pettifor JM, Jorde LB, Eccleshall TR, Feldman D 2000 Vitamin D receptor polymorphisms and nutritional rickets in Nigerian children. J Bone Miner Res 15:2206–2210. 190. Mellanby E 1919 An experimental investigation on rickets. Lancet i:407–412. 191. Iqbal SJ, Kaddam I, Wassif W, Nichol F, Walls J 1994 Continuing clinically severe vitamin D deficiency in Asians in the UK (Leicester). Postgrad Med J 70:708–714. 192. Preece MA, Ford JA, McIntosh WB, Dunnigan MG, Tomlinson S, O’Riordan JL H 1973 Vitamin-D deficiency among Asian immigrants to Britain. Lancet i:907–910. 193. Dunnigan MG, McIntosh WB, Ford JA 1976 Rickets in Asian immigrants. Lancet i:1346. 194. Wills MR, Day RC, Phillips JB, Bateman EC 1972 Phytic acid and nutritional rickets in immigrants. Lancet i:771–773. 195. Ford JA, Colhoun EM, McIntosh WB, Dunnigan MG 1972 Biochemical response of late rickets and osteomalacia to a chupatty-free diet. Br Med J 3:446–447. 196. Batchelor AJ, Compston JE 1983 Reduced plasma half-life of radio-labelled 25-hydroxyvitamin D3 in subjects receiving a high-fiber diet. Br J Nutr 49:213–216. 197. Batchelor AJ, Watson G, Compston JE 1982 Changes in plasma half-life and clearance of 3H-25-hydroxyvitamin D3 in patients with intestinal malabsorption. Gut 23: 1068–1071. 198. Clements MR, Johnson L, Fraser DR 1987 A new mechanism for induced vitamin D deficiency in calcium deprivation. Nature 325:62–65. 199. Halloran BP, Castro ME 1989 Vitamin D kinetics in vivo: effect of 1,25-dihydroxyvitamin D administration. Am J Physiol 256:E686–E691. 200. Halloran BP, Bikle DD, Levens MJ, Castro ME, Globus RK, Holton E 1986 Chronic 1,25-dihydroxyvitamin D3 administration in the rat reduces serum concentration of 25-hydroxyvitamin D by increasing metabolic clearance rate. J Clin Invest 78:622–628.

CHAPTER 65 Nutritional Rickets in Children

201. Clements MR, Davies M, Fraser DR, Lumb GA, Mawer B, Adams PH 1987 Metabolic inactivation of vitamin D is enhanced in primary hyperparathyroidism. Clin Sci 73: 659–664. 202. Clements MR, Davies M, Hayes ME, Hickey CD, Lumb GA, Mawer EB, Adams PH 1992 The role of 1,25-dihydroxyvitamin D in the mechanism of acquired vitamin D deficiency. Clin Endocrinol 37:17–27.

1083 203. Berlin T, Bjorkhem I 1988 Effect of calcium intake on serum levels of 25-hydroxyvitamin D3. Eur J Clin Invest 18: 52–55. 204. Pettifor JM 1994 Privational rickets: a modern perspective. J Roy Soc Med 87:723–725. 205. Shany S, Hirsh J, Berlyne GM 1976 25-Hydroxycholecalciferol levels in bedouins in the Negev. Am J Clin Nutr 29:1104–1107.

CHAPTER 66

Vitamin D Insufficiency in Adults and the Elderly PIERRE J. MEUNIER AND MARIE-CLAIRE CHAPUY INSERM Unit 403, Faculty Laennec and Department of Rheumatology and Bone Disease, Edouard Herriot Hospital, Lyon, France

I. II. III. IV.

Introduction Definition of Vitamin D Deficiency and Insufficiency Determinants of Vitamin D Insufficiency Consequences of Low Vitamin D Status

I. INTRODUCTION In contrast to the extensive attention paid to vitamin D as a component of nutritional health in infants and children, for whom vitamin D deficiency was synonymous with rickets, the relationship of vitamin D to the nutritional health of adults or elderly subjects had been largely ignored for a long time. Studies on the vitamin D status of the elderly began only in the mid-1970s when assays for serum 25-hydroxyvitamin D (25OHD), the best barometer of vitamin D status, became available. The initial studies were carried out in Europe, and it was not until 1982 that studies on the nutritional status of the elderly living in the United States included data on the vitamin D status of the aged population [1]. With the major changes in demography that occurred since the 1960s leading to increased life expectancy of the population, vitamin D insufficiency represents a timely and common health problem in older people. Vitamin D insufficiency is frequently associated with abnormal bone metabolism including secondary hyperparathyroidism, which induces an increase in bone turnover and bone loss, particularly in cortical bone. This leads to an increased risk of fracture, especially of hip fracture, in those subjects with vitamin D insufficiency. The purpose of this chapter is to define this state of vitamin D insufficiency, discuss its prevalence among adults and older subjects, and analyze its causes and its consequences on parathyroid function, bone, and muscle. At the end of the chapter, the utility of increasing vitamin D intake, not only in the elderly but also in adults, is discussed. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Prevalence of Vitamin D Insufficiency VI. Preventive Measures: Correction of Low Vitamin D Status VII. Conclusions References

II. DEFINITION OF VITAMIN D DEFICIENCY AND INSUFFICIENCY At the outset, it is essential to clarify the definition of vitamin D “insufficiency.” Some confusion exists in the literature between the terms vitamin D “insufficiency” and “deficiency” or depletion. The term “insufficient” is defined as lacking in something necessary for completeness, and “deplete” is defined as empty [2]. “Repletion,” or “sufficiency” on the other hand, connote fullness. In this chapter, we define a subject with reduced vitamin D as vitamin D insufficient, and a subject severely lacking in vitamin D as vitamin D deplete or deficient. Peacock et al., in 1985, have been among the first authors to propose definitions of vitamin D deficiency and insufficiency based on serum 25OHD concentrations [3]. According to these authors, vitamin D deficiency occurs with serum 25OHD levels ranging from 0 to 4 ng/ml (0–10 nmol/liter) with evident secondary hyperparathyroidism and malabsorption of calcium, leading to the histological evidence of frank osteomalacia. In vitamin D insufficiency, that is, 25OHD levels ranging from 4 to 20 ng/ml (10–50 nmol/liter), there is mild hyperparathyroidism, suboptimal calcium absorption, high bone turnover, and reduced bone density. In vitamin D sufficiency the 25OHD levels range from 20 to 80 ng/ml (50 to 200 nmol/liter), and there is no disturbance in calcium homeostasis and bone metabolism. Peacock et al. have further defined vitamin D sufficiency by examining the serum 1,25-dihydroxyvitamin D [1,25(OH)2D] response to treatment with 25OHD3 in various groups of patients and healthy subjects. Vitamin D sufficiency [i.e., no change in serum 1,25(OH)2D levels] Copyright © 2005, Elsevier, Inc. All rights reserved.

1086 occurred at a serum 25OHD level in excess of 20 ng/ml (50 nmol/liter), which is above the lower limit of the classical normal range for 25OHD values. Other investigators have used the parathyroid hormone (PTH) level as an index of vitamin D repletion, on the basis of the rationale that vitamin D insufficiency results in decreased calcium absorption, a subtle decline in blood ionized calcium, and consequently an increase in PTH values [4–6]. For Gloth and coworkers [7,8] and Webb et al. [9], the lowest acceptable limit for 25OHD levels was 37 nmol/liter (15 ng/ml). Later, in 1998, Lips et al. pointed out that the 25(OH)D level associated with maximal suppression of PTH was 30 nmol/L (12 ng/ml) [5], but more recently the results of several studies have shown that the serum 25(OH)D level corresponding to the lower limit of vitamin D sufficiency was much higher than this “classical” threshold of 30 nmol/L. These studies have placed the estimate at 75 to 80 nmol/L (Chapuy et al., 1997 [10]), 65 to 75 nmol/L (Thomas et al., 1998 [11]), 50 nmol/L (Malabanan et al., 1998 [12]), 82 nmol/L (Krall et al., 1989 [6]), 99 nmol/L (Dawson-Hughes et al., 1997 [13]), 75 nmol/L (Tangpricha et al., 2002 [14]) and 100 nmol/L (Vieth et al., 2003 [15]). Thus, the estimates of 25 (OH)D required for maximal PTH suppression vary widely from 30 to 100 nmol/L, and there is a cluster of estimates in the 75 to 80 nmol/L range. However, serum 25OHD concentration varies with country, season, and sunshine exposure. The mean values of serum 25OHD are higher in the United States than in northwestern European countries. This may result in different thresholds for diagnosing insufficiency versus sufficiency. As noted above, substrate-dependent synthesis is another way to diagnose a vitamin D–deficient state; when vitamin D therapy results in an increase in the serum 1,25(OH)2D concentration, vitamin D insufficiency is probable [3,16]. This approach might be complicated by an increase in PTH values secondary to low 25OHD values. Also, the comparability of the cutoffs proposed by different authors may vary because different methods have been used for 25OHD assays (see Chapter 58). As the diagnosis of hypervitaminosis D is based on a blood test, whether very low or “undetectable” levels of vitamin D represent deficiency or depletion will depend in part on the sensitivity of the assay method. Assays using extraction and purification give lower results than those without a preparative chromatographic step. In the future, new radioimmunoassays, which give a higher correlation with the high-performance liquid chromatography (HPLC) method [17], certainly will be used more frequently and will allow better comparison of the data from different studies.

PIERRE J. MEUNIER AND MARIE-CLAIRE CHAPUY

On the other hand, because the major biological effects of vitamin D are mediated by 1,25(OH)2D, many researchers would probably agree that vitamin D deficiency should entail 1,25(OH)2D deficiency. However, low 1,25(OH)2D may not always be associated with low 25OHD, and normal values of 1,25(OH)2D can exist in the presence of low 25OHD levels combined with an increased PTH concentration. The determination of circulating 1,25(OH)2D concentrations has very limited clinical utility except in the diagnosis of some conditions involving the vitamin D endocrine system such as vitamin D–dependent rickets types I (PDDR) and II (HVDRR) and some hypercalcemic states caused by increased levels of 1,25(OH)2D. Even if 1,25(OH)2D is the active form of vitamin D, its measurement provides little information in many disorders and no information at all on the nutritional status of vitamin D [18]. Returning to the question of terminology, it appears that the term “deficiency” used in many studies or reports is heterogeneous and includes several different states. We believe that the term “insufficiency” is wellsuited to define the state of hypovitaminosis D that induces other abnormalities of bone metabolism and changes in biochemical indices, but this is not yet accepted by all investigators. The reappraisal at a higher level of the definition threshold for vitamin D insufficiency has two practical consequences: first, vitamin D insufficiency is much more common than formerly believed and this is relevant to larger possibilities of vitamin D supplements for preventing bone loss and fractures, particularly in elderly people; second, it is important to ensure that the serum 25 OHD level obtained after vitamin D supplementation reaches this new threshold. Most studies have shown that an intake of 800 IU/day (20 µg) is needed to attain the desired 25 OHD level (also see Chapters 61 and 62).

III. DETERMINANTS OF VITAMIN D INSUFFICIENCY The vitamin D status of a subject is derived mainly from cutaneous synthesis initiated by solar irradiation of the skin and also from dietary intake. A reduction of one or both sources unavoidably leads to vitamin D insufficiency. Some of these issues are also discussed in Chapter 46.

A. Sunlight Deprivation The body stores of vitamin D are mainly dependent on the cutaneous synthesis of vitamin D3, but this

1087

CHAPTER 66 Vitamin D Insufficiency in Adults and the Elderly

B. Low Vitamin D Intake The natural sources of dietary vitamin D are fatty fishes, fish liver oil, and to a lesser extent eggs. Regional differences in vitamin D intake are important. Mean vitamin D intake per day is lower in Europe compared with both North America and Scandinavia. In North America, milk is fortified with 400 lU/quart of either vitamin D2 or vitamin D3, but studies have revealed

10

Young adults

Recommended intake

Elderly

8 Oral vitamin D intake (µ/day)

synthesis is in turn dependent on several factors such as the duration of sunlight exposure, the latitude of the country, the season, the time of day, and atmospheric conditions (see Chapter 3). Holick found that in Boston (42° N), exposure to sunlight on cloudless days between the months of November and February for up to 5 hr did not result in any significant production of vitamin D [19]. At high latitudes, the stores of vitamin D are mainly generated during the summer months. However, the penetration of effective ultraviolet rays (285 to 310 nm) into the cutaneous layers is modified by the type of clothing, the blockage of effective rays by window glass, and the capacity of skin to produce vitamin D. Generally, on going outdoors, healthy elderly subjects take protective action to reduce sunlight exposure either by use of clothing and sunscreens or simply by just avoiding direct sun. Even in a sunny country, Lebanon, a severe hypovitaminosis D has been shown in 31% of a population of 316 volunteers, more prevalent in women (41.5%), particularly in the veiled ones (62%) [20]. Recommendations for reducing the risk of skin cancer have heightened the situation. Aging has a dramatic impact on the skin: after the age of 20, skin thickness decreases linearly with age, and the capacity to produce previtamin D3 is reduced. The increase in 25OHD values after the same amount of simulated sunlight was 3 times higher in younger subjects aged 22–30 years than in elderly subjects aged 62–80 years [21]. This is due in part to the age-related decline in skin thickness [22]. However, it has been shown that irradiation with UV-B in the very elderly for a few minutes per day leads to adequate improvement of the vitamin D status. It is as effective as oral vitamin D3 (400 IU/day) in increasing serum 25 OHD and suppressing secondary hyperparathyroidism [23]. Even in young adults, the seasonal variation of 25OHD concentrations might be explained by sun deprivation during winter months, when hypovitaminosis D is more frequent. Elderly institutionalized people often are unable to go outdoors because many are infirm or sick; thus, their vitamin D status depends exclusively on vitamin D supplied by food.

6

4

2

0 N Amer

Scand

Europe

FIGURE 1

Mean of the average vitamin D intake in young adults and elderly from studies collated according to geographic region. The recommended intake is 10 µg/day. From McKenna [24] with permission.

that upward of 70% of all milk samples tested did not contain this amount. In Scandinavia (as in Japan), people have a substantial amount of fish (salmon, mackerel, herring) as a part of their diet. In the review by McKenna [24], the mean vitamin D intake was found to be 2.5 ± 1.3 µg/day in Europe, 6.2 ± 2.4 µg/day in North America, and 5.2 ± 2.0 µg/day in Scandinavia. In all studies combined, no differences were found between the vitamin D intake of young adults and the elderly (Fig. 1). The primacy of oral intake over sunlight exposure in maintaining vitamin D stores during the wintertime, especially in elderly subjects residing in Western Europe, and even in North American healthy postmenopausal women, has been demonstrated [6]. The current U.S. recommended dietary allowance of vitamin D for adults is 200 IU or 5 µg. From many studies, it appears that the need is greater and may approach 400 IU or 10 µg per day in order to maintain adequate 25OHD concentrations in adults living in high latitude countries. It is likely that the true vitamin D requirement, in the absence of any exposure to sunlight, is closer to 600 IU or 15 µg/day [19]. In elderly people who are confined indoors and who get no direct sunlight exposure, the requirement is increased to 800 IU/day. In addition, some studies, but not all [25,26], found that

1088 intestinal vitamin D absorption was diminished in the elderly by as much as 40% compared to a younger population [27,28]. We have shown that an oral vitamin D supplementation with 800 IU or 20 µg given every day is safe and has proved its efficacy to maintain adequate 25OHD concentrations [29].

PIERRE J. MEUNIER AND MARIE-CLAIRE CHAPUY

Low vitamin D intake Reduced skin synthesis

+

Renal dysfunction

Reduced 25OHD Reduced 1,25(OH)2D

Reduced calcium absorption

C. Medical Causes

Hypocalcemia

IV. CONSEQUENCES OF LOW VITAMIN D STATUS A. Biochemical Consequences: Increased Serum PTH As the synthesis of the active metabolite is substrate dependent, vitamin D insufficiency will lead to a small decrease in 1,25(OH)2D levels. The lowest value of serum 25OHD necessary to maintain the normal synthesis of 1,25(OH)2D is not known. In clinical studies, this threshold value was found to be lower than 12 ng/ml (30 nmol/liter) by Lips et al. [5] and 15 ng/ml (37.5 nmol/liter) by Bouillon et al. [16]. The subsequent subtle decrease in 1,25(OH)2D levels leads to decreased serum calcium and induces an increase in serum PTH levels. This mechanism was suggested in 1983 by Riggs and Melton as a pathophysiological factor for so-called “type II osteoporosis” [31] (Fig. 2) (see Chapter 67). This significant increase in serum PTH with age in subjects living either at home or in an institution has been reported in many studies [32–35].

Secondary hyperparathyroidism Increased bone remodeling Increased bone loss Increased risk of fragility fractures

FIGURE 2 Pathophysiological mechanism of senile secondary hyperparathyroidism. Although increased PTH will stimulate 1,25(OH)2D synthesis to correct hypocalcemia, this action is not efficient due to the decrease in the substrate (25OHD) concentration for 1α-hydroxylase activity.

It has been found to be partly related to renal dysfunction, age, and/or increasing years since menopause, and mainly to a fall in 25OHD levels [4,5,36] (Fig. 3) (see also Chapter 50). Renal insufficiency denoted by a declining glomerular filtration rate appears to be a cause [37], but not the main cause of the age-related rise in serum PTH [38]. Vitamin D insufficiency seems to be a significant determinant of senile secondary hyperparathyroidism. We have found that the main determinants for PTH values in free-living elderly women were, at equal weight, age and 25OHD values and, to a lesser degree, 2.6

n : 300 r = 0.33 p = 0.0001

2.2 log PTH

Vitamin D insufficiency also may be due to malabsorption of vitamin D (Chapter 75) or to drugs known to alter the metabolism of vitamin D (Chapter 74). Vitamin D is a fat-soluble vitamin, and disorders that lead to steatorrhea are associated with vitamin D insufficiency. This is seen in subjects with hepatobiliary and gastrointestinal disorders, such as celiac disease, cystic fibrosis, chronic pancreatitis, and partial or total gastrectomy. In such subjects, the enteropatic circulation of 25OHD may be also disordered [30]. Vitamin D insufficiency may occur in patients taking anticonvulsant therapy, such as phenytoins, phenobarbital, and glutethimide (see Chapter 74). In these patients, the 25OHD concentrations are decreased; however, the mechanisms involved in this reduction are uncertain. One possibility is that anticonvulsants induce hepatic microsomal enzymes, which metabolize vitamin D to biologically inactive degradation products. Another possible mechanism may be drug interference in the synthesis of 1,25(OH)2D.

1.8

1.4

0 0

20

40

60

80

100

s 25OHD (ng/ml)

FIGURE 3 Inverse relationship between serum 25OHD and log PTH values in 300 elderly women living in institutions.

CHAPTER 66 Vitamin D Insufficiency in Adults and the Elderly

creatinine clearance [39]. When all these variables were adjusted in a multiple regression model, age and 25OHD were still significant predictors of PTH, but not creatinine clearance [39]. Ooms et al. [40] have found that only 25OHD values below a threshold of 10 ng/ml (25 nmol/liter) were related to PTH. Below this threshold, PTH increased 14.1% for every 4 ng/ml (10 nmol/liter) drop in serum 25OHD. Notwithstanding the fact that in several studies vitamin D only accounted for less than 10% of the variance in PTH, it must be emphasized that senile secondary hyperparathyroidism can be reversed by vitamin D supplements [4,5,29,41]. However, the association between 25OHD and PTH is likely to be influenced by the levels of calcium intake, as calcium supplementation was also demonstrated to be able to lower PTH levels in elderly subjects [42].

B. Impact on Bone Severe and prolonged deficiency in vitamin D is associated with osteomalacia characterized by defective mineralized bone and reduced bone strength (Chapter 63). On the other hand, vitamin D insufficiency that is not severe enough to cause osteomalacia may nevertheless contribute to hip fracture risk in the elderly by decreasing calcium absorption and increasing PTH secretion, leading to increased bone turnover and bone loss, particularly in cortical bone (Fig. 2) [1,31,43,44]. This concept seems to be confirmed by the higher PTH levels found in patients with hip fractures than in matched controls [45,46]. It is also supported by histomorphometric studies that have shown an increase in resorption parameter values (i.e., reduced thickness of iliac cortices and increased number of osteoclasts per square millimeter of bone section) in patients with vertebral fractures [43,45]. These bone changes have been confirmed by Okano et al. [47]. On the other hand, overt osteomalacia, as determined by both increased thickness of osteoid seams and decrease in the calcification rate measured through tetracycline double labeling, is very rare and has been found in fewer than 10% of patients with hip fracture [43,44,48]. The histological picture of the first stage of vitamin D–deficient osteopathy (vitamin D insufficiency) is caused by secondary hyperparathyroidism and cannot be distinguished from hyperparathyroid bone disease on histological grounds [49] (also see Chapters 59 and 63). Several studies have shown a relationship between femoral neck bone mineral density (BMD), vitamin D insufficiency, and secondary hyperparathyroidism not only in the elderly [40,50,51], but also in middle-aged women [52,53]. Martinez et al. found an association

1089 between low femoral BMD and low 25OHD levels in normal women older than 65 [54]. For Villareal et al., there was a relationship between vertebral bone density assessed by quantitative computerized tomography and PTH values only in subjects with low 25OHD values (<15 ng/ ml or 38 nmol/liter) [55]. Ooms et al. [40] found similar results for the BMD of the hip, the best fit being obtained with a threshold value of 25OHD at 12 ng/ml (30 nmol/liter). The femoral BMD was 5% higher for every 4 ng/ml (10 nmol/liter) increase in 25OHD up to the threshold. In this study, serum PTH was negatively related to BMD at all measurements’ sites, with the correlation coefficient ranging from −0.19 for the distal radius to -0.27 for the left femoral neck. In contrast, at a serum 25OHD level of 4 ng/ml (10 nmol/liter), the BMD of the femoral neck was reduced by 9.3%, which is 0.6 SD below the average BMD for an adequate vitamin D status (i.e., above 12 ng/ml or 30 nmol/liter). According to the data of Cummings et al. [56] and Ooms et al. [40], a 25OHD level of 4 ng/ml (10 nmol/liter) results in a relative risk of hip fracture of 1.8. In a German population, Scharla et al., found a borderline significant positive correlation between femoral neck BMD and 25OHD values in men older than 70 years (r = 0.34, p < 0.03). In women between 50 and 70 years (r = 0.36, p < 0.02), there was no association between vertebral BMD and 25OHD values [57]. In a cross-sectional study in the United Kingdom, a positive relationship between serum 25OHD values and BMD of the lumbar spine, the femoral neck, and greater trochanter was observed in a group of 138 middle-aged women volunteers (45–65 years) [52]. In a prospective study, Dawson-Hughes et al. [41] have shown that vitamin D insufficiency noted in winter contributes to spinal bone loss in healthy postmenopausal women (mean age 61 years), which can be reduced by daily supplements with 10 µg of vitamin D. Several studies have shown correlations between bone biochemical markers and the kinetic and histomorphometric evaluation of bone formation and resorption, and it has been possible to demonstrate the impact of 25OHD insufficiency on bone status using the biochemical markers of bone turnover. Brazier, Kamel, and coworkers [58,59] have reported a two- to three-fold elevation of resorption estimated by excretion of pyridinoline crosslinks in elderly subjects with vitamin D insufficiency and secondary hyperparathyroidism, as compared with vitamin D–sufficient elderly. In free-living healthy elderly women, we have found that the mean values of bone alkaline phosphatase, osteocalcin, and collagen C-telopeptide (crosslaps) were significantly increased as compared with the results obtained in young women [39] (Table I). For these

1090

PIERRE J. MEUNIER AND MARIE-CLAIRE CHAPUY

TABLE I Comparison of Biochemical Markers of Bone Remodeling in Elderly and Young Women during the Winter Marker

Elderly women living at home (EPIDOS study, n = 405)

Young women (OFELY study, n = 54)

15.2 ± 6.2 (a) 24.9 ± 9.6 (a) 0.36 ± 0.22 (b) 29 ± 12 (a) 311 ± 168 (a)

8.5 ± 2.6 14.9 ± 4.4 0.26 ± 0.15 19 ± 7 186 ± 108

Bone alkaline phosphatase (µg/ml) Osteocalcin (ng/ml) Urinary calcium (mmol/nmol creatinine) Urinary hydroxyproline (µmol/mmol creatinine) Crosslaps* (µg/mmol creatinine)

Data are expressed as means ± SD. Lowercase letters show results of statistical analyses: (a), significantly different from young women < p = 0.0001; (b), significantly different from young women p = 0.001. From Chapuy et al. [39]. * Collagen C-telopeptide.

markers, we found significant positive correlations with PTH values and negative correlations with hip BMD. In these healthy ambulatory women, we did not find a correlation between 25OHD and PTH with BMD values as reported in some studies [52,54,55], but other researchers have also been unable to show a significant correlation between BMD and PTH values [51,60]. Thus, if an increase in PTH levels secondary to vitamin D insufficiency is one certain cause of age-related bone loss and fragility, this suggests that other unidentified factors may play a major role in the increase in bone turnover. Ooms et al. [61] found that low serum 25OHD concentrations are associated with higher PTH and osteocalcin and lower BMD of the hip, but also that when high sex hormone-binding globulin (inverse measure of estrogen activity) is combined with vitamin D insufficiency, the secondary hyperparathyroidism is more severe. This suggests that low estrogen activity causes decreased sensitivity of the gut to 1,25(OH)2D [62], leading to higher serum PTH levels and increasing the impact of vitamin D insufficiency. The observation, in several studies, of a seasonal variation in bone mineral density in normal subjects provides indirect evidence that relatively small changes in vitamin D status may have significant effects on bone mass. The demonstration that late wintertime bone loss could be prevented by small increases in vitamin D intake [41] provides a powerful argument in favor of the hypothesis that vitamin D status contributes to increased bone turnover and cortical bone loss (see also Chapter 63).

C. Other Effects: Muscle Weakness Hypovitaminosis D with or without osteomalacia has been associated with muscle weakness, limb pain, and impaired muscle function. This might be explained by the fact that vitamin D receptors (VDR) exist in

muscle, or it may be secondary to the mineral abnormality (see Chapters 55 and 102). Birge and Haddad have shown that 25OHD directly influences intracellular accumulation of phosphate by muscle and offered this as an important role in the maintenance of muscle metabolism and function [63]. A syndrome of hyperesthesia has been described in association with hypovitaminosis D in five homebound elderly subjects by Gloth et al. [64], but no systematic studies of pain and muscle strength have been performed in an older population before and after treatment of vitamin D insufficiency [65]. Pain and weakness can lead to functional disability that may prevent a person from venturing outdoors, and this, in turn, exacerbates a poor vitamin D status by decreasing exposure to sunlight. Corless et al., in a multicenter study, were unable to demonstrate an effect of vitamin D supplements on the functional activity levels in elderly subjects [66]. On the other hand, Sorensen et al. have shown that treatment with lα-hydroxyvitamin D (1αOHD) was followed by a reduction in the time to dress [67]. This functional disability associated with vitamin D insufficiency may increase the incidence of falls and consequently the risk of hip fracture. Until relatively recently, studies on the prevention of senile osteoporosis and fractures with vitamin D supplements have failed to mention the incidence of falls [29,68,69], but a recent study run by Bischoff et al. [70] in 122 elderly institutionalized women has shown a reduction of 49% of the risk of falling in the group receiving 800 IU of vitamin D3 and 1200 mg/day of calcium compared to the group receiving only calcium.

V. PREVALENCE OF VITAMIN D INSUFFICIENCY In analyzing the prevalence of vitamin D insufficiency in populations, it is necessary to divide the

1091

CHAPTER 66 Vitamin D Insufficiency in Adults and the Elderly

Winter

Spr/Fall

Summer

120 100 25OHD nmol/liter

subjects according to age, country of residence, and season in which the study was performed (see also Chapters 47 and 62). To facilitate the analysis of vitamin D status, countries were grouped according to geographic regions: North America, Scandinavia, Europe, and others. Elderly subjects were subdivided into healthy elderly persons living at home and institutionalized subjects. In addition, the seasonal variations in serum 25OHD levels were taken into account. To analyze the values reported in the literature, we have largely relied on the review by McKenna [24] in which 117 studies of vitamin D status from 27 regions published between 1971 and 1990 were analyzed. In some groups of subjects, and especially in healthy elderly, the number of studies is very small.

80 60 40 20

A. Vitamin D Insufficiency in Adults (30–70 years)

0 N Amer

In North America where dairy products are fortified with vitamin D, vitamin D insufficiency in young adults is rare [71–73]. In the study by Sherman et al. [35], fewer than 1% of men and 4% of women had 25OHD values lower than 14 ng/ml (35 nmol/liter), and these low values were found in May–June, when mean values are at their nadir. In Scandinavian countries, where vitamin D intake is high, vitamin D insufficiency was noted in 4 to 9% of young adults during winter and in up to 5% during summer [74]. In Ireland [75], Italy [76,77], and England [24], 15 to 40% of young adults have hypovitaminosis D during the winter (Fig. 4). In a cross-sectional survey in a large Swiss population (3276 subjects aged 25–74 years) between November 1988 and June 1989, 6% of subjects had 25OHD values lower than 8 ng/ml (20 nmol/liter), and 34% had values lower than 15 ng/ml (37.5 nmol/liter) [78]. In subjects younger than 65 years, there was a small but clear cyclical seasonal variation with a nadir in February (median 41 nmol/liter i.e., 16.4 ng/ml) and a zenith in June (median 55 nmol/liter i.e., 22 ng/ml). In contrast with previous studies which reported more important differences between young and elderly adults, in this population-based cohort, there was only a small and nonsignificant decrease in 25OHD values between the ages of 24 and 75 years. This may be because the upper age limit was only 75 years. In a study that we performed on a large adult French population from the SUVIMAX project (Supplementation with Vitamins Minerals and Antioxidants), 25OHD and PTH status were measured in 1584 healthy adults living in 16 large towns between November 1994 and April 1995 (772 men aged 45–60 years and 812 women aged 35–60 years) [79] (Table II).

Scand

Europe

FIGURE 4

Mean of average serum 25OHD levels in young adults reported during the winter, spring and fall combined, and summer from studies collated according to geographic region. From McKenna [24] with permission.

The SUVIMAX project is a large international epidemiologic study to assess the effects of vitamins, minerals, and antioxidants in the prevention of global and specific mortality and morbidity due to cardiovascular diseases, cancers, and cataracts. In this adult population, 14% of subjects had 25OHD lower than 12 ng/ml (30 nmol/liter). There was no age effect, but a sex difference between 25OHD levels was apparent (vitamin D insufficiency was present in 15.5% of women and only in 12.4% of men; p < 0.04). Major differences were observed between different regions within the country. In addition, there was a significant negative correlation between 25OHD and PTH values for the whole population (r = 0.20, p < 0.001). The mean PTH levels were 48.2 ± 22.0 pg/ml for subjects with 25OHD concentrations at or below 30 ng/ml and 37.9 ± 14.2 pg/ml for the others (p < 0.01). This study demonstrates that vitamin D insufficiency is present in a substantial percentage of the general adult urban population in France, a finding that has been confirmed by two other studies [80,81]. In low latitude countries, hypovitaminosis D is also present in adults. For example, in Saudi Arabia up to 40% of native inhabitants and immigrants from Africa have hypovitaminosis D in winter [82]. In healthy young adult Japanese, the frequency of hypovitaminosis D reaches 5% in males and 25% in females [83]. Actually, it appears that the vitamin D status of the young adult

1092

PIERRE J. MEUNIER AND MARIE-CLAIRE CHAPUY

TABLE II

Region North Center Northeast Northwest Paris Rhone-Alpes Mediterranean coat South Southwest

Prevalence of Vitamin D Insufficiency in the General Adult Population Living in 16 Large French Townsa n 200 99 199 300 98 200 299 89 100

Mean serum 25OHD (nmol/liter)b

Vitamin D insufficiencyc (% hypovitaminosis D)

Mean serum PTH (pg/ml)d

29 27 18 14 13 9 7 6 0

42 ± 15 40 ± 15 42 ± 16 38 ± 17 46 ± 24 40 + 15 35 ± 13 40 ± 11 37 ± 11

43 ± 21 49 ± 25 52 ± 26 58 ± 29 59 + 25 62 ± 27 68 ± 27 81 ± 27 94 ± 38

aA

total of 1584 subjects were studied: 772 males from 45 to 60 years old and 812 females from 35 to 60 years old. From Chapuy et al. [79]. convert nmol/liter to ng/ml 25OHD, divide by 2.5. cVitamin D insufficiency is deemed as a serum 25OHD level below 30 nmol/liter. dPTH normal range is 55 pg/ml. bTo

population needs to receive more attention, especially in winter, in countries where foods are not fortified. In the studies referenced by the McKenna analysis [24], the time of the year at which the measurements was performed is generally recorded, so as to permit a comparison of results with reference to the season of study [72,84,85]. In the studies performed earlier, the season and therefore the individual exposure to sunlight are not always reported.

Town-specific mean 25OHD concentrations range from 10 to 24 ng/ml (25–59 nmol/liter) for men and from 8 to 19 ng/ml (21–48 nmol/liter) for women. Overall 36% of men and 47% of women had 25OHD concentrations below 12 ng/ml (30 nmol/liter), the lowest concentrations being found surprisingly in southern European towns in France, Spain, and Italy. The low

B. Vitamin D Insufficiency in the Elderly (70–90 years)

Spr/Fall

Summer

80

60 25OHD nmol/liter

Many studies, primarily those from Europe, have indicated that low levels of vitamin D may be more prevalent in older persons even if they do not appear related to the aging process per se (Fig. 5). In the healthy elderly population in North America and Scandinavia, nearly 25% of subjects had low values in winter but less than 5% had low levels throughout the year [25,86–90]. So, in these countries, even if the vitamin D intake is adequate and equal or greater than the recommended dietary allowance, the fraction of the vitamin D pool due to sunlight exposure is very important. As was found in normal adults, in elderly subjects there is a marked seasonal variation in 25OHD levels. In Europe, the frequency of vitamin D insufficiency in winter ranges from 8 to 60% [16,24,84,91–94]. The Euronut Seneca study has evaluated the 25OHD concentrations in 824 free-living elderly people from 16 towns (latitudes between 35° and 61° N) in 11 European countries between December 1988 and March 1989 [95].

Winter

40

20

0 N Amer

FIGURE 5

Scand

Europe

Mean of average serum 25OHD levels in healthy elderly reported during winter, spring and fall combined, and summer from studies collated according to geographic region. From McKenna [24] with permission.

1093

CHAPTER 66 Vitamin D Insufficiency in Adults and the Elderly

TABLE III

Number of subjects Age (years) Calcium intake (mg/day) Serum calcium 25OHD PTH (pg/ml)

Comparison of Biochemical Values in Elderly and Young Women during the Winter* Elderly women living at home

Young (EPIDOS)

440 80 ± 3 569 ± 338 2.35 ± 0.07 42.5 ± 25.0 63 ± 28

59 87 ± 6 (a) – 2.22 ± 0.09 (a) 15.5 ± 6.5 (c) 76 ± 49 (d)

Elderly (OFELY) 54 34 ± 5 (a) 810 ± 280 (b) 2.35 ± 0.07 (mmol/liter) 62.0 ± 40.0 (b) (nmol/liter) 43 ± 15 (c)

*Data are expressed as means ± SD. Lowercase letters show results of statistical analyses: (a), p < 0.0001 compared to EPIDOS and OFELY studies; (b), p < 0.0001 compared to EPIDOS study; (c), p < 0.001 compared to EPIDOS study; (d), p < 0.05 compared to EPIDOS study. From Chapuy et al. [39].

have low vitamin D status, with 25OHD levels at or below 10 ng/ml (25 nmol/liter) [97–99]. In Europe, vitamin D insufficiency in elderly homebound subjects ranged from 70 to 100% [24]. The European 25OHD values in institutionalized subjects are significantly lower than both Scandinavian and North American values (Fig. 6). Only a few studies were conducted during specific seasons, but one would not expect much seasonal change as these elderly subjects have little or no exposure to sunlight [100]. The mean European values, based on year-round determinations of 25OHD and calculated from values reported in 16 European studies [24], is lower than 8 ng/ml (20 nmol/liter).

80

25OHD nmol/liter

25OHD values in this study were largely explained by attitudes toward sunlight exposure. We have studied [39] the vitamin D status of 440 healthy, free-living elderly women aged 75–90 years during winter and living in five French cities whose latitude varies from 49°9′ to 43°6′ N. The mean 25OHD level was not different among the five cities. However, in all of the cities, the mean 25OHD level was significantly lower than the mean level found in young healthy women from the OFELY cohort study that were recruited during the same winter period (42.5 ± 25 versus 62.0 ± 40.0 nmol/liter). (OFELY is a prospective study of the boneloss determinants in women aged 30–95 years randomly selected from a large insurance company in Lyon, France.) Nevertheless, the mean values of the free-living elderly women were significantly higher (p = 0.03) than the winter values obtained in 59 institutionalized elderly women (42.5 ± 25 versus 15.5 ± 6.5 nmol/liter) (Table III). Among the institutionalized elderly women, 39% exhibited vitamin D insufficiency (25OHD values < 12 ng/ml, i.e., 30 nmol/liter) and only 16% had normal 25OHD values greater than or equal to 25 ng/ml (62.5 nmol/liter), the mean value of young women (Table II). Vitamin D insufficiency was associated in these healthy women with biochemical indices of secondary hyperparathyroidism and increased bone turnover. In comparison to the few studies of free-living people, the vitamin D status of homebound subjects has been extensively studied. In the review by McKenna [24], the prevalence of vitamin D insufficiency in the institutionalized elderly population varies from 3 to 28% in North America. This prevalence reaches 54% in the study by Gloth et al. [96], who compared homebound elderly in community versus nursing homes. Despite apparently adequate vitamin D intake (200 to 400 IU/ day or 5 to 10 µg/day), several studies have shown that between 30 and 50% of older homebound subjects

40

20

0 N Amer

FIGURE 6

Scand

Europe

Mean of average serum 25OHD levels in institutionalized elderly reported throughout the year from studies collated according to geographic region. From McKenna [24] with permission.

1094 In our DECALYOS I study, we have measured the 25OHD values at baseline in 280 very elderly women (84 ± 6 years) who were ambulatory but lived in nursing homes. (The DECALYOS study was the prospective study undertaken to determine the effects of vitamin D and calcium supplements on hip fracture incidence [29] in Lyon, France.) The mean 25OHD values were 14 ± 10 ng/ml (35 ± 25 nmol/liter), and 44% of subjects had 25OHD values lower than or equal to 12 ng/ml (30 nmol/liter). In another sample of elderly institutionalized women recruited for the recent DECALYOS II study [101], we found 66% of women with both an inadequate calcium intake (< 800 mg/day) and low vitamin D status (serum 25OHD <12 ng/ml). This suggests a valid rationale for combined calcium and vitamin D supplementation. Similarly, in the Dutch study of Ooms et al. [61], 65% of the 330 healthy women, residents of home apartments for the elderly, had 25OHD values below 12 ng/ml (30 nmol/liter), and in 34% the 25OHD level was below 8 ng/ml (20 nmol/liter). In winter 83% and in summer 50% of the subjects had 25OHD levels that were below 12 ng/ml (30 nmol/liter). The median levels of 25OHD were significantly higher for inhabitants of apartments for the elderly than for residents of homes for the elderly (11.6 and 8.8 pg/ml, respectively, i.e., 29 and 22 nmol/liter; p < 0.0001).

VI. PREVENTIVE MEASURES: CORRECTION OF LOW VITAMIN D STATUS The effects on bone turnover of vitamin D insufficiency and secondary hyperparathyroidism have suggested that correction of subclinical hypovitaminosis D in the elderly may have a beneficial effect on the secretion of PTH, bone loss, and consequently the risk of fracture in elderly subjects.

A. Effects on Secondary Hyperparathyroidism In the late 1980s, in a six-month trial of supplementation with vitamin D2 (800 IU or 20 µg/day) and calcium (1 g/day) given to elderly subjects with vitamin D insufficiency living in an institution, we were able to reduce serum PTH concentrations by more than 30% with a parallel normalization of 25OHD concentrations [4]. With only 400 IU/day (10 µg) of vitamin D3, Ooms et al. restored to normal 25OHD concentrations and decreased the mean PTH level by about 6 to 15% in an elderly Dutch population [40]. These two studies demonstrated that supplementation with low doses of

PIERRE J. MEUNIER AND MARIE-CLAIRE CHAPUY

vitamin D (400 to 800 IU/day or 10 to 20 µg/day) leads to adequate improvement of vitamin D status and parathyroid function with a greater effect of the 800 IU daily dose. There does not appear to be a need for a higher dose, as has been suggested by others [1]. In contrast, vitamin D supplementation with 10 µg of vitamin D of nursing home residents in the United States having 25OHD levels of 16 ng/ml or greater (>40 nmol/liter) [102], or of about 50 nmol/liter [103] did not significantly decrease PTH levels.

B. Effects on Bone Mass By increasing the vitamin D intake from 100 to 500 IU daily, Dawson-Hughes et al. [41] were able to significantly reduce the late wintertime bone loss and to improve the net bone density of the spine. Among normal, middle-aged women, Khaw et al. found a significant direct relationship between serum 25OHD and bone mass of the spine and femoral neck, with an inverse correlation between bone density and PTH concentration, which is consistent with effects of small changes in vitamin D status on bone mass [52]. The use of 15,000 IU or 375 µg of oral vitamin D2 weekly has also been shown to reduce metacarpal bone loss in normal women aged 65 to 74 years living in the community [104]. In our DECALYOS I study [29], the elderly women treated daily for 18 months with 800 IU or 20 µg of vitamin D3 and 1.2 g of calcium showed an increase of 2.7% in the BMD of the proximal femur. During the same period, the women in the placebo group showed a decrease of femoral BMD equal to 4.6%. In DECALYOS II study, femoral neck BMD decreased in the placebo group (−2.4% per year) while remaining unchanged in women treated with 800 IU of vitamin D3 and 1200 mg/day of calcium (+0.3% per year). No significant difference between groups was found for changes in distal radius BMD [101]. In the Dutch study of Ooms et al., a daily treatment of elderly women (mean calcium intake 859 mg/daily) with 400 IU or 10 µg of vitamin D3 over a two-year period was associated with an increase in femoral neck BMD equal to 2.3% in comparison with the placebo group. In these women at baseline, hip BMD was positively correlated with the serum 25OHD concentration below a threshold level (serum 25OHD < 12 ng/ml or 30 mmol/liter [40]).

C. Effects on Fracture Rate The effects of vitamin D supplements on bone mass suggested that correction of vitamin D insufficiency in the elderly with low doses of vitamin D may have

1095

CHAPTER 66 Vitamin D Insufficiency in Adults and the Elderly

beneficial effects on fracture incidence [105]. This was the aim of several studies. In our DECALYOS I study, 3270 healthy ambulatory women (mean age 84 ± 6 years) living in a nursing home received daily either 800 IU or 20 µg of vitamin D3 plus 1.2 g of elemental calcium or a double placebo. After 18 months of follow-up, analysis showed that there was a 43% reduction in hip fractures (p < 0.05) and a 32% reduction in all nonvertebral fractures (p = 0.015) in the treated group. The results of the intention-to-treat analysis were also similar: 80 hip fractures in the vitamin D3–calcium group and 110 in the placebo group (27% reduction; p < 0.01). At the same time, serum PTH decreased by 46% (p < 0.001) and the serum 25OHD level increased by 160% (p < 0.001) without change in the 1,25(OH)2D levels. After a further 18 months of treatment for 1404 women, the beneficial effects of the treatment on nonvertebral fractures was confirmed. At the end of 36 months of follow-up in the intention-to-treat analysis, there were 17.2% fewer nonvertebral fractures (255 versus 308, p < 0.02) and 23% fewer hip fractures (137 versus 178, p < 0.02) in the vitamin D3–calcium group (Fig. 7). The probability of hip fractures was decreased (odds ratio 0.73, CI 0.62 to 0.84), as was that of all nonvertebral fractures (odds ratio 0.72, CI 0.60 to 0.84) [29,68] (Fig. 8). This study has pointed out the importance of vitamin D insufficiency as a major determinant of senile secondary hyperparathyroidism and bone loss (see also Chapter 78). However, the study did not permit the elucidation of the relative importance of calcium

versus vitamin D. Also, the possible effect of vitamin D supplements on the incidence of falls was not studied. In the DECALYOS II study performed in 583 ambulatory institutionalized women (mean age 85.2 years), the relative risk (RR) of hip fracture in the placebo group compared with the active treatment (800 UI vitamin D3 and 1200 mg calcium/day) was 1.69 (CI 0.96 to 3.0), which is similar to that found in DECALYOS I (RR = 1.7; CI 1.0 to 2.8) [101]. In the study of Heikinheimo et al. [106], 799 elderly men and women, living either in residential care or in their own home, were followed for 2 to 5 years. In those treated with an annual injection of 150,000 to 300,000 IU of vitamin D2, there was a significant reduction in the number of fractures in the upper limbs and ribs but no significant reduction in hip fractures. The reduction in hip fracture incidence reached 22% (9.4% in the control group and 7.3% in the intervention group), which is similar to the reduction found in our DECALYOS I study (23%). One reason for the lack of significance was possibly the smaller sample size. The prevention of fracture with vitamin D alone found by Heikinheimo might suggest that the results obtained with the association of calcium and vitamin D would be primarily due to vitamin D [106,107]. In a prospective double blind trial of 2578 men and women (mean age 80 ± 6 years) living either at home or in an institution, Lips et al. studied the effect of a daily supplement with 400 IU of vitamin D3 [108]. After one year of vitamin D supplementation in a subgroup of

Total number of fractures

Subjects with at least one fracture

400

400

p < 0.02*

368

p < 0.02*

308

301 300

300 255 −18.2% 184

200

200 −17.2%

138 100

D3Ca All fractures

Placebo

255

100

−25%

Placebo

178

D3Ca

Hip fractures

−23%

Placebo

D3Ca All fractures

Placebo

D3Ca

Hip fractures

*log rank

FIGURE 7 Effects of vitamin D3 and calcium supplementation during three years on numbers of fractures in elderly women (DECALYOS I study). One group received placebo, and the other group (D3Ca) received 800 IU of vitamin D3 plus 1.2 g elemental calcium [29].

1096

PIERRE J. MEUNIER AND MARIE-CLAIRE CHAPUY

A Cumulative probability of fracture

Placebo Ca + D3

0.20 0.15

p < 0.01 0.10 0.05

Cumulative probability of fracture

B

0.25

0.16 Placebo 0.12

Ca + D3

0.08

p < 0.02 0.04

0

0 0

6

12

18

24

30

36

Months

0

6

12

18

24

30

36

Months

FIGURE 8

Cumulative probability of hip fractures (A) and all fractures (B) in the placebo and vitamin D3–calcium groups (DECALYOS I study). From Meunier et al. [50] with permission.

women [40], the PTH level was reduced by only 6% from baseline, and after two years the BMD of the femoral neck had increased 2.2% in the vitamin D group compared to the placebo group. Nevertheless, the vitamin D supplementation did not significantly decrease the incidence of hip fracture after a maximum treatment period of 3.5 years. Similarly, in the recent study reported by Meyer et al. [103] and performed in residents of 51 Norwegian nursing homes (mean age 85.0 years), an intervention with 400 IU of vitamin D3 per day over two years had no effect on the incidence of hip fracture or all nonvertebral fractures. The differences observed in the results of the French and the Dutch studies may be explained by the lower dietary calcium intake in France than in the Netherlands and the use of calcium in the French study. In addition, the participants in the French study were four years older, less active at baseline, and were all residents of nursing homes. Hip fracture incidence was lower in the Dutch study (29 per 1000 versus 40 per 1000), so its power may have been insufficient to demonstrate significant differences in the fracture rates. Nevertheless, the increase in 25OHD levels and more importantly the reduction in PTH secretion were much lower in the Dutch and the Norwegian studies, raising the possibility that the daily dose of 400 IU of vitamin D3 was suboptimal (minus 39% in DECALYOS I study—confirmed in DECALYOS II study—versus minus 6% in the Dutch study and no change in the Norwegian study after one year of treatment). These results indicate that a larger treatment effect was obtained in the French studies [68,101] where 800 IU/day of vitamin D had been given. In addition to these four prospective studies on the preventive effect of vitamin D on hip fracture

incidence, Banstam and Kanis in a preliminary retrospective report from the MEDOS (MEDOS: Mediterranean Osteoporosis Study) study did not find that the use of vitamin D was associated with a significant decrease in the risk of hip fracture, as was the use of calcium, estrogen, or calcitonins [109]. When the data were reanalyzed including not all the hip fractures but only the low energy fractures, the use of vitamin D supplement was associated with a 26% (but nonsignificant) decrease in the risk of hip fracture [93]. In addition, the risk reduction was influenced by age and body mass index (BMI). For women 80 years or older, the reduction in hip fracture risk for vitamin D users was 37% ( p = 0.04), and for these women with a BMI less than 20 kg/m2, the use of vitamin D was associated with a marked and significant reduction in hip fracture risk of 55% ( p = 0.01). In this study, vitamin D was taken for a time ranging from 1 to 20 years, and the doses used were not known. From all these studies, it appears that vitamin D supplements are undoubtedly useful in the prevention of hip fracture, but there are still two critical questions that need to be answered. What is the optimal dose, and who should receive supplements?

VII. CONCLUSIONS There is now convincing evidence that there are several forms of low vitamin D states that induce different forms of bone disease. Severe and prolonged “deficiency” of vitamin D, which is no longer very common, is associated with osteomalacia. In contrast, less severe vitamin D depletion, which might be called vitamin D “insufficiency,” is very common in elderly subjects

1097

CHAPTER 66 Vitamin D Insufficiency in Adults and the Elderly

living in institutions or at home but also appears not to be rare in healthy adults, particularly in winter. A progressive state of vitamin D insufficiency per se does not imply that bone disease is present but does identify a high risk status [1]. Vitamin D insufficiency will ultimately lead to a state of vitamin D deficiency. However, before overt clinical symptoms of osteomalacia become apparent, vitamin D insufficiency may give rise to problems at the bone level as a consequence of secondary hyperparathyroidism, resulting in increased bone turnover, bone loss, and risk of fractures. The observation of seasonal variation in bone mineral density in normal subjects provides evidence that relatively small changes in vitamin D status may have significant effects on bone mass. In agreement with several authors, we propose that vitamin D insufficiency might be redefined by a 25OHD level equal to or under 30 ng/ml (75 nmol/liter), this value being required for maximal PTH suppression. In elderly subjects it has been shown that 800 IU of vitamin D3 are usually needed to reach this optimal level of serum 25OHD. In many countries, natural dietary sources do not readily permit an intake of the recommended amounts of vitamin D; thus, the available mode of prevention of vitamin D insufficiency might be increased exposure to sunlight, fortification of foodstuffs, and oral or injectable vitamin D supplementation. Because the prevalence of vitamin D insufficiency is very high among elderly subjects, an increase in vitamin D intake becomes essential, especially for those living in an institution. Promoting sun exposure is difficult to realize, and placing ultraviolet lamps in the living room would probably not be efficient enough. Supplementation by vitamin D preparations seems to be the best solution. This supplementation should be higher than 400 lU/day in the healthy elderly to increase 25OHD levels, because this dose was not sufficient for prevention of hip fractures in elderly subjects [103,108]. However, 800 IU of vitamin D3 with 1.2 g of calcium was able to decrease by 25% the hip fractures’ incidence of institutionalized elderly women in three years and to maintain normal 25OHD and PTH concentrations. As it appears that vitamin D supplementation is effective and safe in preventing vitamin D insufficiency, daily low dose treatment (800 IU/daily) may be the best regimen. However, an intermittent high dose (100,000 IU) given orally or by injection every six months may be an effective alternative [94]. In contrast, very high single doses, like 600,000 IU twice a year, should be used cautiously because of the risk of inducing transient hypervitaminosis D and side effects due to hypercalcemia, including increased calcification rate of vascular atheromatous plaques.

References 1. Parfitt MA, Gallagher JC, Heaney RP, Johnston CC, Neer R, Whedon GD 1982 Vitamin D and bone health in the elderly. Am J Clin Nutr 36:1014–1031. 2. Walters MR, Kollenkirchen U, Fox J 1992 What is vitamin D deficiency? Proc Soc Exp Biol Med 199:385–393. 3. Peacock M, Selby PL, Francis RM, Brown WB, Hordon L 1985 Vitamin D deficiency, insufficiency, sufficiency and intoxication. What do they mean? In: Norman A, Schaefer K, Grigolett MG, Herrath DV (eds) Sixth Workshop on Vitamin D. De Gruyter, Berlin and New York, pp. 569–570. 4. Chapuy MC, Chapuy P, Meunier PJ 1987 Effect of calcium and vitamin D supplements on calcium metabolism in the elderly. Am J Clin Nutr 46:324–328. 5. Lips P, Wierzinga A, Van Ginkel FC, Jongen MJM, Netelenbos JC 1988 The effect of vitamin D supplementation on vitamin D status and parathyroid function in elderly subjects. J Clin Endocrinol Metab 67:644–650. 6. Krall EA, Sahyoun N, Tannebaum S, Dallal GE, Dawson-Hughes B 1989 Effect of vitamin D intake on seasonal variations in parathyroid hormone secretion in postmenopausal women. N Engl J Med 321:1777. 7. Gloth FM, Gundberg CM, Hollis BW, Haddad JG, Tobin JD 1995 Vitamin D deficiency in homebound elderly persons. JAMA 274:1683–1686. 8. Gloth FM, Tobin JD 1995 Vitamin D deficiency in older people. J Am Geriatr Soc 43:822–828. 9. Webb AR, Pilbeam C, Hanafin N, Holick MF 1990 An evaluation of the relative contributions of exposure to sunlight and of diet to the circulating concentrations of 25-hydroxyvitamin D in an elderly nursing home population in Boston. Am J Clin Nutr 51:1075–1081. 10. Chapuy MC, Preziosi P, Maamer M, Arnaud S, Galan P, Herckberg S, Meunier PJ 1997 Prevalence of vitamin D insufficiency in an adult normal population. Osteoporos Int 7:439–443. 11. Thomas MK, Lloyd-Jones DM, Thadhani RI, Shaw AC, Deraska DJ, Kitch BT, Vamvakas EC, Dick IM, Prince RL, Finkelstein JS 1998 Hypovitaminosis D in medical inpatients. N Engl J Med 338:777–783. 12. Malabanan AO, Veronikis IE, Holick MF 1998 Redefining vitamin D insufficiency. Lancet 351:805–806. 13. Dawson-Hughes B, Harris SS, Dallal GE 1997 Plasma calcidiol, season and serum parathyroid hormone concentrations in healthy elderly men and women. Am J Clin Nutr 65:67–71. 14. Tangpricha V, Pearce EN, Chen TC, Holick MF 2002 Vitamin D insufficiency among free-living healthy young adults. Am J Med 112:659–6620. 15. Vieth R, Ladak Y, Walfish PG 2003 Age-related changes in the 25-hydroxyvitamin D versus parathyroid hormone relationship suggest a different reason why older adults require more vitamin D. J Clin Endocrinol Metab 88:185–191. 16. Bouillon RA, Auwerx JD, Lissens WD, Pelemans WK 1987 Vitamin D status in the elderly, seasonal substrate deficiency causes 1,25-dihydroxycholecalciferol deficiency. Am J Clin Nutr 45:755–763. 17. Lips P, Chapuy MC, Dawson-Hughes B, Pols HAP, Holick MF 1999 An international comparison of serum 25-hydroxyvitamin D measurements. Osteoporos Int 9:394–397. 18. Hollis B 1996 Assessment of vitamin D nutritional and hormonal status: What to measure and how to do it. Calcif Tissue Int 58:4–5.

1098 19. Holick MF, McCollum Award lecture 1994 Vitamin D—New horizons for the 21st century. Am J Clin Nutr 60:619–630. 20. Gannagé-Yared MH, Chemali R, Yaacoub N, Halaby G 2000 Hypovitaminosis D in a sunny country: relation to lifestyle and bone markers. J Bone Miner Res 15:1856–1862. 21. Holick MF, Matsuoka LY, Wortsman J 1989 Age, vitamin D, and solar ultraviolet. Lancet 1:1104–1105. 22. Need AG, Morris HA, Horowitz M, Nordin BEC 1993 Effects of skin thickness, age, body fat, and sunlight on serum 25-hydroxyvitamin D. Am J Clin Nutr 58:882–885. 23. Chel VGM, Ooms ME, Popp-Snijders C, Pavel S, Schothorst AA, Meulemans CCE, Lips P 1998 Ultraviolet irradiation corrects vitamin D deficiency and suppresses secondary hyperparathyroidism in the elderly. J Bone Miner Res 13:1238–1242. 24. McKenna MJ 1992 Differences in vitamin D status between countries in young adults and in elderly. Am J Med 93:69–77. 25. Clemens TL, Zhou X, Myles M, Endres D, Lindsay R 1986 Serum vitamin D2 and vitamin D3 metabolite concentrations and absorption of vitamin D2 in elderly subjects. J Clin Endocrinol Metab 63:656–660. 26. Holick MF 1986 Vitamin D requirements for the elderly. Clin Nutr 5:121–129. 27. Weisman Y, Schen RJ, Einsenberg Z, Edelstein S, Harell A 1981 Inadequate status and impaired metabolism of vitamin D in elderly. J Med Sci 17:19–21. 28. Baragry JM, France MW, Corles D, Gupta SP, Switala S, Boucher BJ, Cohen RD 1978 Intestinal cholecalciferol absorption in the elderly and in younger adults. Clin Sci Mol Med 55:213–220. 29. Chapuy MC, Arlot ME, Duboeuf F, Brun J, Crouzet B, Arnaud S, Delmas PD, Meunier PJ 1992 Vitamin D3 and calcium to prevent hip fractures in elderly women. N Engl J Med 237:1637–1642. 30. Mundy GR 1989 In: Calcium Homeostasis: Hypercalcemia and Hypocalcemia. Martin Dunitz, London, pp. 185–186. 31. Riggs BL, Melton LJ 1983 Evidence of two distinct syndromes of involutional osteoporosis. Am J Med 75:899–901. 32. Chapuy MC, Durr F, Chapuy P 1983 Age-related changes in parathyroid hormone and 25-hydroxycholecalciferol levels. J Gerontol 38:19–22. 33. Forero MS, Klein RF, Nissenson RA, Nelson K, Heath H, Arnaud CD, Riggs BL 1987 Effect of age on circulating immunoreactive and bioactive parathyroid hormone levels in women. J Bone Miner Res 2:363–366. 34. Epstein S, Bryce G, Hinman JW, Miller ON, Riggs BL, Hui SL, Johnston CC 1986 The influence of age on bone mineral regulating hormones. Bone 7:421–425. 35. Sherman SS, Hollis BW, Tobin JD 1990 Vitamin D status and related parameters in a healthy population: The effects of age, sex, and season. J Clin Endocrinol Metab 71:405–413. 36. Prince RL, Dick I, Devine A, Price RI, Gutteridge DH, Kerr D, Criddle A, Garcia-Webb P, St John A 1995 The effects of menopause and age on calciotropic hormones: A cross sectional study on healthy women aged 35 to 90. J Bone Miner Res 10:835–842. 37. Freaney R, McBrinn Y, Me Kenna MJ 1993 Secondary hyperparathyroidism in elderly people: Combined effect of renal insufficiency and vitamin D deficiency. Am J Clin Nutr 58:187–191. 38. Young G, Marcus R, Minkoff JR, Kirn LY, Segre GV 1987 Age-related rise in parathyroid hormone in men: The use of intact and midmolecule antisera to distinguish hormone secretion from retention. J Bone Miner Res 2:367–374.

PIERRE J. MEUNIER AND MARIE-CLAIRE CHAPUY

39. Chapuy MC, Schott AM, Garnero P, Hans D, Delmas PD, Meunier PJ 1996 Healthy elderly French women living at home have secondary hyperparathyroidism and high bone turnover in winter. J Clin Endocrinol Metab 81: 1129–1133. 40. Ooms ME, Roos JC, Bezemer PD, Van Der Vijch WJF, Bouter LM, Lips P 1995 Prevention of bone loss by vitamin D supplementation in elderly women: A randomized double blind trial. J Clin Endocrinol Metab 80:1052–1058. 41. Dawson-Hughes B, Dallal GE, Krall EA, Harris S, Sokoll LJ, Falconer G 1991 Effect of vitamin D supplementation on overall bone loss in healthy postmenopausal women. Ann Intern Med 115:505–512. 42. Kochersberger G, Westlund R, Lyles KW 1990 Calcium supplementation lowers serum parathyroid hormone levels in elderly subjects. J Gerontol 45:M159–M162. 43. Lips P, Netelendos C, Jongen MJM, Van Ginkel FC, Altuis AL, Vanschaik CL, Vandervijch WJF, Vermeiden JPW, Van Der Meer C 1982 Histomorphometric profile and vitamin D status in patients with femoral neck fracture. Metab Bone Dis Related Res 4:85–93. 44. Lips P 2001 Vitamin D deficiency and secondary hyperparathyroidism in the elderly: consequences for bone loss and fractures and therapeutic implications. Endocrine Reviews 22:477–501. 45. Johnston CC, Norton J, Khairi MRA, Kernek C, Edouard C, Meunier PJ 1985 Heterogeneity of fracture syndromes in postmenopausal women. J Clin Endocrinol Metab 61:551–556. 46. Compston JE, Silver AC, Croucher PI, Brown RC, Woodhead JS 1989 Elevated serum intact parathyroid hormone levels in elderly patients with hip fracture. Clin Endocrinol 31:667–672. 47. Okano T, Yamamoto K, Hagino H, Hishimoto H 1992 Iliac bone histomorphometry in Japanese women with hip fracture. In: Bone Morphometry, Sixth International Congress, Lexington, 4–9 October, Abstract 99. 48. Benhamou CL, Chappard D, Gauvain JB 1991 Hyperparathyroidism in proximal femur fractures: Biological and histomorphometric study in 21 patients over 75 years old. Clin Rheumatol 10:144–160. 49. Rao DS, Villanueva A, Mathews SM 1983 Histologic evaluation of vitamin D depletion in patients with intestinal malabsorption or dietary deficiency. In: Frame B, Potts J Jr (eds) Clinical Disorders of Bone and Mineral Metabolism. Excerpta Medica, Amsterdam, pp. 224–226. 50. Meunier PJ, Chapuy MC, Arlot ME, Delmas PD, Duboeuf F 1994 Can we stop bone loss and prevent hip fractures in the elderly? Osteoporosis Int 4(Suppl 1):S71–S76. 51. Rosen CJ, Morrison A, Zhou H, Storm D, Hunter SJ, Musgrave K, Chen T, Wei W, Holick MF 1994 Elderly women in northern New England exhibit seasonal changes in bone mineral density and calciotropic hormones. Bone Miner 2:83–92. 52. Khaw KT, Sheyd MJ, Compston J 1992 Bone density, parathyroid hormone, and 25-hydroxyvitamin D concentrations in middle-aged women. Br Med J 305:273–277. 53. Lukert B, Higgins J, Stoskopf M 1992 Menopausal bone loss is partially regulated by dietary intake of vitamin D. Calcif Tissue Int 51:173–179. 54. Martinez ME, Delcampo MJ, Sanchez-Cabezudo MJ, Garcia JA, Sanchez-Calvin MT, Torrijos A, Coya J 1994 Relations between calcidiol serum levels and bone mineral density in postmenopausal women with low bone density. Calcif Tissue Int 55:253–256.

CHAPTER 66 Vitamin D Insufficiency in Adults and the Elderly

55. Villareal DT, Civitelli R, Chines A, Avioli LV 1991 Subclinical vitamin D deficiency in postmenopausal women with low vertebral bone mass. J Clin Endocrinol Metab 72:628–634. 56. Cummings SR, Black DM, Nevitt MC, Browner W, Cauley J, Ensrud K, Genant HK, Palermo L, Scott J, Vogt TM 1993 Bone density at various sites for prediction of hip fractures. Lancet 341:72–75. 57. Scharla SH, Scheidt-Nave C, Leidig G, Seibel M, Ziegler R 1994 Association between serum 25-hydroxyvitamin D and bone mineral density in a normal population sample in Germany. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D: A Pluripotent Steroid Hormone: Structural Studies, Molecular Endocrinology and Clinical Applications. de Gruyter, Berlin, p. 863. 58. Brazier M, Kamel S, Maamer M, Agbomson F, Elseper I, Garabedian M, Desmet G, Sebert JL 1995 Markers of bone remodeling in the elderly subjects: Effects of vitamin D insufficiency and its correction. J Bone Miner Res 10:1753–1761. 59. Kamel S, Brazier M, Picar C, Boitte F, Samson L, Desmet G, Sebert JL 1994 Urinary excretion of pyridinoline crosslinks measured by immunoassay and HPLC techniques in normal subjects and elderly patients with vitamin D deficiency. Bone Miner 26:197–208. 60. Dawson Hughes B, Harris SS, Krall EA, Dallal GE, Falconer G, Green CL 1995 Roles of bone loss in postmenopausal women randomly assigned to one or two dosages of vitamin D. Am J Clin Nutr 61:1140–1145. 61. Ooms ME, Lips P, Roos JC, Vandervijch WJF, PoppSnijdeers C, Bezemer PD, Bouter LM 1995 Vitamin D status and sex hormone binding globulin: Determinants of bone turnover and bone mineral density in elderly women. J Bone Miner Res 10:1177–1184. 62. Gennari C, Agnusdei D, Nardi P, Civitelli R 1990 Estrogen preserves a normal intestinal response to 1,25-dihydroxyvitamin D3 in oophorectomized women. J Clin Endocrinol Metab 71:1288–1293. 63. Birge SJ, Haddad JC 1975 25-Hydroxycholecalciferol stimulation of muscle metabolism. J Clin Invest 56:1100–1107. 64. Gloth FM, Lindsay JM, Zelesnick LB, Greenough WB 1991 Can vitamin D deficiency produce an unusual pain syndrome. Arch Intern Med 151:1662–1664. 65. Gloth FM, Smith CE, Hollis BW, Tobin JD 1995 Functional improvement with vitamin D replenishment in a cohort of frail vitamin D–deficient older people. J Am Geriatr 43:1269–1271. 66. Corless D, Dawson D, Fraser F, Ellis M, Evans SJ, Perry JD, Reisner C, Silver CP, Beer M, Boucher BJ 1985 Do vitamin D supplements improve the physical capabilities of elderly hospital patients. Age Aging 14:76–84. 67. Sorensen OH, Lund BI, Saltin B 1979 Myopathy in bone loss ageing: Improvement by treatment with 1α-hydroxycholecalciferol and calcium. Clin Sci 56:157–161. 68. Chapuy MC, Arlot ME, Delmas PD, Meunier PJ 1994 Effect of calcium and cholecalciferol treatment for three years on hip fractures in elderly women. Br Med J 308:1081–1082. 69. Tilyard MW, Spears GFS, Thomson J, Dovey S 1992 Treatment of postmenopausal osteoporosis with calcitriol or calcium. N Engl J Med 326:357–362. 70. Bischoff HA, Stahelin HB, Dick W, Akos R, Knecht M, Salis C, Nebiker M, Theiler R, Pfeifer M, Begerow B, Lew RA, Conzelmann M 2003 Effects of Vitamin D and calcium supplements on falls: a randomized controlled trial. J Bone Miner Res 18:343–351.

1099 71. Haddad JG, Hahn TJ 1973 Natural and synthetic sources of circulating 25-hydroxyvitamin D in man. Nature 244:515–517. 72. Styrd RP, Gilbertson TJ, Brunden MN 1979 A seasonal variation study of 25-hydroxyvitamin D3 serum levels in normal humans. J Clin Endocrinol Metab 48:771–775. 73. Chesney RW, Rosen JF, Hamstra AJ, Smith C, Mahaffey K, DeLuca HF 1981 Absence of seasonal variation in serum concentrations of 1,25-dihydroxyvitamin D despite a rise in 25-hydroxyvitamin D in summer. J Clin Endocrinol Metab 53:139–142. 74. Lund B, Sorenson OH 1979 Measurement of 25-hydroxyvitamin D in serum and its relation to sunshine, age, and vitamin D intake in the Danish population. Scand J Clin Lab Invest 39:23–30. 75. Vik T, Try K, Stromme JH 1980 The vitamin D of man at 70° north. Scand J Clin Lab Invest 40:227–232. 76. Benucci A, Tommasi M, Fantappie B, Scardigli S, Ottanelli S, Pratesi E, Romano S 1993 Serum 25-hydroxyvitamin D levels in normal subjects: Seasonal variation and relationship with parathyroid hormone and osteocalcin. J Nucl Biol Med 37:77–82. 77. Carnevale V, Modoni S, Pileri S, Di Giorgio A, Chiodini I, Minisola S, Vieth R, Scillitani A 2001 Longitudinal evaluation of vitamin D status in healthy subjects from Southern Italy: seasonal and gender differences Osteoporos Int 12:1026–1030. 78. Burnand B, Slovtskis D, Gianoli F, Cornuz J, Rickenback M, Paccaud F, Burckhardt P 1992 Serum 25-hydroxyvitamin D: distribution and determinants in the Swiss population. Am J Clin Nutr 56:537–542. 79. Chapuy MC, Preciozi P, Maamer M, Arnaud S, Galan P, Hercberg S, Meunier PJ 1997 Prevalence of hypovitaminosis D in the femoral adult French population. Osteoporos Int 7:439–443. 80. Ribot C, Pouilles JM, Tremollieres F 1996 Vitamin D status in a large cohort of early postmenopausal French women. Osteoporosis Int 6(Suppl. l):abstract PSU218. 81. Baudoin C, Aquino D, Charlier C, Cohen Solal M, Gueris J, Devernejoul MC 1996 Vitamin D deficiency in postmenopausal women in region of Paris. Osteoporosis Int 6(Suppl. l):abstract PSU281. 82. Sedrani SH, Elidrisy AWTH, El Arabi KM 1983 Sunlight and vitamin D status in normal Saudi subjects. Am J Clin Nutr 38:129–132. 83. Takeuchi A, Okano T, Ishida Y, Kovayashi T 1995 Dietary vitamin D intake and vitamin D nutritional status in the healthy young and institutionalized elderly people. In: Burckhardt P Heaney RP (eds) Nutritional Aspects of Osteoporosis. Ares Serono Symposia Publications, Rome, pp. 351–355. 84. Stamp TCB, Round JM 1974 Seasonal changes in human plasma levels of 25-hydroxyvitamin. Nature 247:563–565. 85. Poskitt EME, Cole TJ, Lawson DEM 1979 Diet, sunlight, and 25-hydroxyvitamin D in healthy children and adults. Br Med J 1:221–223. 86. Delvin EE, Imbach A, Copti M 1988 Vitamin D nutritional status and related biochemical indices in an autonomous elderly population. Am J Clin Nutr 48:373–378. 87. Von Knorring J, Slatis P, Weber TH, Helenius T 1982 Serum levels of 25-hydroxyvitamin D, 24,25-dihydroxyvitamin D and parathyroid hormone in patients with femoral neck fracture in southern Finland. Clin Endocrinol 17:189–194. 88. Omdahl JL, Garry PJ, Hunsaker LA, Hunt WC, Goodwin JS 1982 Nutritional status in a healthy elderly population: Vitamin D. Am J Clin Nutr 36:1125–1133.

1100 89. Lukert BP, Carey M, McCarty B, Tiemann S, Goodnight L, Helm M, Hassanein R, Stevenson S, Stoskopf M, Doolan L 1987 Influence of nutritional factors on calcium regulating hormone and bone mass. Calcif Tissue Int 40:119–125. 90. Schmidt-Gayk H, Goosen J, Lendle F, Seidel D 1997 Serum 25-hydroxycholecalciferol in myocardial infarction. Atherosclerosis 26:55–58. 91. Rapin CH, Lagier R, Boivin G, Jung A, McGee W 1982 Biochemical findings in blood of aged patients with femoral neck fractures: A contribution to the detection of occult osteomalacia. Calcif Tissue Int 34:465–469. 92. Newton HMV, Sheltawy M, Hay AWM, Morgan B 1985 The relations between vitamin D2 and D3 in elderly women in Great Britain. Am J Clin Nutr 41:760–764. 93. Ranstam J, Kanis JA 1995 Influence of age and body mass on the effects of vitamin D on hip fractures risk. Osteoporosis Int 5:450–454. 94. Byrne PM, Preatney R, McKenna MJ 1995 Vitamin D supplementation in the elderly: Review of safety and effectiveness of different regimes. Calcif Tissue Int 56:518–520. 95. Van Der Wielen RP, Lowik MRH, Van Der Berg H, Degroot L, Haller J, Moreras O, Vanstaveren WA 1995 Serum 25OHD concentrations among elderly people in Europe. Lancet 346:207–210. 96. Gloth MD, Tobin JD, Smith CE, Hollis BW, Gunberg CM 1993 The prevalence of vitamin D deficiency in homebound elderly: Community vs nursing home. J Am Geriatr Soc 41(Suppl.):17–22. 97. Gloth FM, Tobin JD, Shermann SS, Hollis BW 1991 Is the recommended daily allowance for vitamin D too low in the homebound elderly? J Am Geriatr Soc 39:137–141. 98. McMurtry CT, Yound SE, Downs RW, Adler RA 1992 Mild vitamin D deficiency and secondary hyperparathyroidism in nursing home patients receiving adequate dietary vitamin D. J Am Geriatr Soc 40:343–347. 99. O’Dowd KJ, Clemens TL, Kelsey JL, Lindsay R 1993 Exogenous calciferol (vitamin D) and vitamin D endocrine status among elderly nursing home residents in the New York City area. J Am Geriatr Soc 41:414–421.

PIERRE J. MEUNIER AND MARIE-CLAIRE CHAPUY

100. Davies M, Mawer EB, Hann JT, Taylor JL 1986 Seasonal changes in the biochemical indices of vitamin D deficiency in the elderly: A comparison of people in residential homes, long stay ward, and attending a day hospital. Age Aging 15:77–83. 101. Chapuy MC, Pamphile R, Paris E, Kempf C, Schlichting M, Arnaud S, Garnero P, Meunier PJ 2002 Combined calcium and vitamin D3 supplementation in elderly women: confirmation of reversal of secondary hyperparathyroidism and hip fracture risk: the Decalyos II study. Osteoporos Int 13:257–264. 102. Himmelstein S, Clemens TL, Rubin A, Lindsay R 1990 Vitamin D supplementation in elderly nursing home residents increases 25OHD but not 1,25(OH)2D. Am J Clin Nutr 52:701–706. 103. Meyer HE, Smedshaug GB, Kvaavik E, Falch JA, Tverdal A, Pedersen JI 2002 Can vitamin D supplementation reduce the risk of fracture in the elderly? A randomized controlled trial. J Bone Min Res 17:709–715. 104. Nordin BEC, Baker MR, Horsman A, Peacock M 1985 A prospective trial on the effect of vitamin D supplementation on metacarpal bone loss in elderly women. Am J Clin Nutr 42:470–747. 105. Compston JE 1995 The role of vitamin D and calcium supplementation in the prevention of osteoporotic fractures in the elderly. Clin Endocrinol 43:393–405. 106. Heikinheimo RJ, Inkovaara JA, Harjv EJ, Haavisto MV, Kaarela RH, Kajata JM, Kokko AML, Kokko LA, Rajala SA 1992 Annual injections of vitamin D and fractures of aged bone. Calcif Tissue Int 51:105–110. 107. Torgeson D, Campbell M 1994 Vitamin D alone may be helpful. Br Med J 309:193 (letter). 108. Lips P, Graafmans WC, Ooms ME, Bezemer PD, Bouter LM 1996 Vitamin D supplementation and fracture incidence in elderly persons. Ann Intern Med 124:400–406. 109. Kanis JA, Johnell O, Gullberg B, Allander E, Dilsen G, Gennari C, Lopez-Vaz AA, Lyritis JP, Mazuoli G, Miravet L, Passeri M, Cano RP, Rapado A, Ribot C 1992 Evidence for the efficacy of bone active drugs in the prevention of hip fracture. Br Med J 305:1124–1128.

CHAPTER 67

Vitamin D and Osteoporosis RICHARD EASTELL AND B. LAWRENCE RIGGS Bone Metabolism Group, Division of Clinical Sciences (North), University of Sheffield, Sheffield, England and the Department of Endocrinology and Metabolism, Mayo Clinic and Mayo Foundation, Rochester, Minnesota, USA

I. II. III. IV. V. VI.

Introduction Effect of Age on Levels of 25OHD Effect of Age on Levels of 1,25(OH)2D The Effect of Age on Calcium Intake and Absorption Age and Parathyroid Hormone Interrelationships between Age-related Bone Loss, Estrogen, PTH, and Vitamin D Metabolism

VII. Summary of Changes in Vitamin D with Aging VIII. Treatment of Established Osteoporosis with Vitamin D IX. Conclusions References

Involutional osteoporosis begins in middle life and becomes progressively more common with advancing age. It is clinically manifest as fractures. It is likely that changes in calcium homeostasis contribute substantially to the pathogenesis of osteoporosis. A proposed model for these interactions on bone loss with aging is shown in Figure 3.

United States [13]. This figure is as high as 75% for the housebound elderly [18]. In Europe, latitude is not a major determinant of 25OHD status. Indeed, people living in southern Europe have lower levels of 25OHD mainly as a result of their attitudes to sunshine exposure, such as the use of long-sleeved clothes [19]. This threshold figure of 30 nmol/L may be too conservative (see Section IX) and thus vitamin D deficiency may be more prevalent than these figures suggest (see Chapter 61).

II. EFFECT OF AGE ON LEVELS OF 25OHD

A. Mechanisms for the Age-related Decrease in 25OHD

Most studies have shown that plasma levels of 25OHD decrease with age by about 50% in both men and women [1–14] (Fig. 1). The importance of this decline in 25OHD to the pathogenesis of osteoporosis has been a major topic of study in the osteoporosis field over the past few years [15]. Lips [16] has proposed that the lower limit of reference range for 25OHD in the summer is taken as 30 nmol/l (see Chapter 62), based on the following arguments:

There are several possible mechanisms for the agerelated decrease in plasma 25OHD, and these will be considered in turn.

I. INTRODUCTION

1. This is the lower limit of the reference range established for the summer based on blood donors. 2. This is the level of 25 OHD above which 1,25(OH)2D is not substrate-dependent. 3. This is the level of 25OHD above which BMD does not correlate with 25OHD [17]. 4. This is the level of 25OHD above which further treatment with vitamin D will not suppress PTH any further. Based on this figure, 25 to 50% of the elderly are vitamin D–deficient in either Europe [18,19] or the VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

1. DIETARY INTAKE OF VITAMIN D

Most elderly subjects do not consume the Recommended Daily Allowance for vitamin D, but this is also true for younger subjects [20]. McKenna [20] found no evidence for differences in vitamin D intake between young adults and the elderly in any geographical region, although average intake varied greatly between countries. 2. VITAMIN D ABSORPTION AND C25 HYDROXYLATION

Evidence relating to the effect of age on vitamin D absorption is conflicting. Barraguy et al. [3] administered 3H-vitamin D to young and elderly subjects and found that absorption was greater in young subjects (13.2% vs. 7.6% in 6 hr). However, administration of vitamin D2 results in similar increments in plasma 25OHD in young and elderly subjects [21,22]. One explanation for these apparently conflicting results is Copyright © 2005, Elsevier, Inc. All rights reserved.

1102

RICHARD EASTELL AND B. LAWRENCE RIGGS

Serum 25(OH)D, ng/mL

80 ∆ −48% r = 0.41, P < 0.001

60

vitamin D was reported as 21 to 27 days [3] and did not increase with age. Clemens et al. reported similar findings [21].

III. EFFECT OF AGE ON LEVELS OF 1,25(OH)2D

40

20

0 30

40

50

60 Age, yrs

70

80

90

FIGURE 1 Influence of age on the serum concentration of 25OHD in a sample of healthy women [from Tsai et al. [13], with permission].

that vitamin D absorption might be decreased in the elderly, but that C25 hydroxylase activity could be increased (possibly as a result of low levels of 1,25(OH)2D in the elderly [23]). 3. EFFECT OF AGE ON PRODUCTION OF VITAMIN D IN SKIN

Exposure to sunshine is often less in the elderly than in the young [2,6]. It has been suggested that widespread use of sunscreens to prevent skin cancer may make the elderly more susceptible to vitamin D deficiency [24], although it is not clear whether sunscreen use increases with age. Evidence regarding the effect of age on the capacity of human skin to produce vitamin D remains controversial (see Chapter 50). MacLaughlin and Holick reported a decrease in 7-dehydrocholesterol (provitamin D) in skin biopsy specimens with aging [25]. Furthermore, conversion of 7-dehydrocholesterol by ultraviolet light was decreased in skin from elderly subjects. Exposure of subjects to ultraviolet light results in an increase in circulating levels of 25OHD, and this response has been reported to be the same in young and old subjects [26] or fourfold greater in young subjects [27]. It has been suggested that the age-related decrease in skinfold thickness might be related to declining levels of 25OHD [28]. 4. EFFECT OF AGE ON CLEARANCE OF 25OHD

The effect of age on the metabolic clearance rate of 25OHD has not been studied directly, but the half-life of plasma 25OHD after administration of 3H-labeled

Plasma levels of 1,25(OH)2D probably decrease with age in most populations, especially after age 65 [5,7,12, 21,29–33]. However, many studies have reported unchanged or increasing levels over a wide age range in studies of both men and women [9,11,34–39], while others have reported increasing levels from age 35 up to age 65, followed by a decrease [37,38,40,41]. Aksnes et al. found that 1,25(OH)2D showed no age-related decrease in active healthy subjects living at home, but decreased significantly in hospitalized geriatric patients not receiving Vitamin D [1]. These conflicting findings suggest that more than one mechanism could account for the age-related changes in 1,25(OH)2D, and that the importance of these mechanisms could vary in different populations.

A. Mechanisms for Age-related Changes in 1,25(OH)2D The age-related decline in levels of 1,25(OH)2D could be due to a change in the level of vitamin D–binding protein (DBP), 25OHD substrate deficiency, estrogen deficiency, decreased renal 1α-hydroxylase activity, or increased metabolic clearance of 1,25(OH)2D. These possible mechanisms will be considered in turn. 1. CHANGES IN VITAMIN D–BINDING PROTEIN

Interpretation of 1,25(OH)2D levels is complicated by the effect of vitamin D–binding protein (DBP), which may also modulate its action. Less than 1% of circulating 1,25(OH)2D is free, and the non DBPbound fraction may correspond better to its biological action under certain circumstances [42]. Changes in vitamin D–binding protein (DBP) can cause profound changes in plasma 1,25(OH)2D levels, but recent experiments in DBP knockout mice have questioned whether levels of DBP can affect the action of 1,25(OH)2D, even though the levels of total vitamin D metabolites are decreased [43] (see Chapter 9). The increases found in pregnancy appear to represent, at least in part, increases in serum free serum 1,25(OH)2D [44]. Moreover, a change in DBP concentration is unlikely to account for a decrease in 1,25(OH)2D with aging, because DBP concentrations have been reported to show no significant change with age [1,30], or a minimal

1103

CHAPTER 67 Vitamin D and Osteoporosis

decrease [12], or even an increase, and when a free calcitriol index is calculated this clearly decreases with age in postmenopausal women [45]. 2. 25OHD SUBSTRATE DEFICIENCY

A marked reduction in the level of 25OHD is likely to be a rate-limiting factor for 1,25(OH)2D production [16]. However, in most elderly populations, the level of 25OHD is probably not rate limiting. This is likely to account for the different effects of vitamin D supplementation in the elderly, with increased levels of 1,25(OH)2D reported in some elderly populations [18,46], but not in others [47,48]. Lips et al. [18] found that the increase in 1,25(OH)2D after vitamin D supplementation was inversely related to the initial 25OHD concentration. Some studies have shown a decrease in 1,25(OH)2D with aging, even though the level of 25OHD did not decline [5,33]. 3. ESTROGEN DEFICIENCY AND 1,25(OH)2D

The contribution of the menopause and estrogen deficiency to the age-related decline in 1,25(OH)2D is uncertain. Sowers et al. [32] found that the age-related decrease in 1,25(OH)2D was largely accounted for by menopausal status, but other studies have found that the menopause has no effect on calculated free 1,25(OH)2D [49,50]. Similarly, some studies have shown that postmenopausal estrogen administration increases total 1,25(OH)2D [51,52], as well as calculated free 1,25(OH)2D [51] while others have found that estrogen administration results in a transient increase in 1,25(OH)2D and DBP, with no change in calculated free 1,25(OH)2D [53]. 4. RENAL 1α-HYDROXYLASE ACTIVITY

In subjects without marked 25OHD substrate deficiency, the most likely cause for the decrease in plasma 1,25(OH)2D levels with aging is decreased renal 1αhydroxylase activity. Renal 1α-hydroxylase activity is increased by PTH, growth hormone, hypocalcemia and hypophosphatemia, and decreased by 1,25(OH)2D [54–56]. lα-hydroxylase activity is also affected by the amount of functioning renal tissue [57]. In early chronic renal failure, the glomerular filtration rate (GFR) determined by radioisotopic DTPA (diethylene triamine penta-acetic acid) clearance is positively correlated with 1,25(OH)2D [57], and these changes are first detectable when renal function is only slightly impaired (GFR < 70 ml/min). There are at least three possible reasons for reduced renal lα-hydroxylase activity with aging. First, aging is associated with a decline in renal mass, and an association between impaired renal function and 1,25(OH)2D levels with normal aging have been reported in some studies [33,58], but not in others [37,59,60].

Secondly, the rise in 1,25(OH)2D in response to PTH infusion is blunted with aging [13,33,61]. This refractoriness to PTH with aging does not appear to be due to estrogen deprivation [50,52]. Studies of 1,25(OH)2D production in renal slices from rats of different ages have indicated that the refractoriness to PTH may be a specific defect, and that the 1,25(OH)2D response to calcitonin is preserved [62]. Finally, growth hormone stimulates renal 1,25(OH)2D production [56,63,64], and growth hormone secretion decreases with age [65,66]. In one cross-sectional study [12], serum IGF-I was found to be the most important determinant of the fall in 1,25(OH)2D with age. Although lα-hydroxylase activity is decreased by rising phosphate concentrations [54], this is not likely to account for an age-related decline in 1,25(OH)2D because serum phosphate tends to decline with age [30,37,67–69]. 5. CLEARANCE OF 1,25(OH)2D

Studies in rats indicate that the age-related decrease in 1,25(OH)2D may be due to increased metabolic clearance of 1,25(OH)2D [70,71]. However, Eastell et al. found that the clearance of infused [3H]1,25(OH)2D tended to decrease with age in healthy women (age range 26 to 88 years), and that this was associated with an age-related increase in serum levels of 1,25(OH)2D [40]. In another study, Halloran et al. found that clearance of 1,25(OH)2D did not vary with age in men, but in this study subjects were selected so as not to show the usual age-related decrease in creatinine clearance [72].

IV. THE EFFECT OF AGE ON CALCIUM INTAKE AND ABSORPTION Net calcium absorption is determined both by dietary calcium intake and the efficiency of calcium absorption. Cross-sectional studies in the United States have shown that calcium intake declines by about 10% in men and women between 35 and 75 years of age [73]. This decrease is due to a reduction in overall caloric intake as the calcium density of the diet does not decline with age [73]. Calcium is absorbed from the intestine by an active 1,25(OH)2D dependent process, and by passive vitamin D–independent mechanisms [74]. The vitamin D– dependent mechanisms are saturated at low intake, and differences in fractional calcium absorption due to variations in the level of 1,25(OH)2D may only be evident when calcium intake is low [74]. Several studies have shown that the efficiency of intestinal calcium absorption decreases with age. This decrease is found by calcium balance studies [75,76],

1104 by jejunal perfusion studies [77], and by radiocalcium absorption tests [5,75,78–80]. Nevertheless, there is some evidence that the actual amount of calcium absorbed from the habitual diet might not vary markedly with age. Eastell et al. [40] found that when true fractional calcium absorption (TFCA) was measured by tracing the habitual diet over 24 hours, true calcium absorption (TFCA multiplied by dietary calcium) did not decline with age. Ebeling et al. [36] reported similar findings using the same technique employing stable isotopes of calcium. In a longitudinal study, Heaney et al. [75] measured fractional calcium absorption by balance and double-tracer methods from a diet constructed to match each subject’s current dietary calcium intake. Although fractional calcium absorption clearly decreased in women who went through the menopause between their first and second study, fractional absorption decreased only slightly (about 0.2% per year) in women who experienced no change in estrogen status. Furthermore, in this study, calcium intake increased slightly with aging, and therefore the decline in net calcium absorption is likely to have been minimal. These differences may relate to methodological factors. The interpretation and relevance of these tests depend on the test calcium load and the length of time over which measurements are made. Calcium absorption tests that use small fixed amounts of calcium carrier measure predominantly active calcium absorption. It is likely that active, but not passive calcium absorption, decreases with age, and in general the lower the amount of fixed calcium carrier, the greater the apparent agerelated decrease is in calcium absorption. Using the small intestine perfusion technique, Ireland and Fordtran [77] found that calcium absorption in elderly subjects on a high calcium diet (2000 mg/day) was not different from that in younger subjects, but it was lower in elderly subjects when both groups were studied on a low calcium diet (300 mg/day). Restriction of measurements to the early time period after radiotracer ingestion (<6 hours) may also overestimate the effect of age on calcium absorption. At this time, part of the ingested dose is still in the large intestine, and isotopic equilibrium is not achieved until about 24 hours [81]. Colonic absorption probably accounts for about 5% calcium absorption in healthy subjects, and this proportion is likely to increase in subjects in whom absorptive efficiency is low [81]. Calcium absorption tests which limit blood sampling to within one hour of tracer ingestion [34,82] may also overestimate the effect of aging because gastric emptying may be considerably delayed in the elderly [83]. Calcium absorption tests in which calcium tracer is administered without food are also likely to exaggerate the effect of age on calcium absorption.

RICHARD EASTELL AND B. LAWRENCE RIGGS

Gastric acidity is markedly reduced in a substantial proportion of elderly subjects [84]. In subjects with achlorhydria Recker et al. found that calcium absorption was very low when administered while fasting, but was normal in the presence of food [85]. In summary, there is good evidence to suggest that calcium absorption is somewhat less efficient in the elderly. Elderly subjects show poor adaptation to variations in calcium intake and calcium balance in the elderly is likely to be more sensitive to calcium intake than in young adults. However, the apparent effect of age may be exaggerated when samples are collected in the early time-period after tracer ingestion, when the carrier dose is low, and when calcium is administered without food.

A. Mechanisms for Decreased Calcium Absorption with Aging 1. DECREASED 1,25(OH)2D

When subjects with widely varying levels of 1,25(OH)2D are studied, calcium absorption and 1,25(OH)2D are highly correlated. This correlation is greatest when subjects are fed low calcium meals [74]. However, in population studies, a decrease in the level of 1,25(OH)2D with age is not a universal finding (see earlier), and several studies have not found a relationship between calcium absorption and 1,25(OH)2D [34,40,86], unless they have included a wide range of vitamin D status (see below). 2. INTESTINAL 1,25(OH)2D RESISTANCE

Another factor that could be related to a decrease in calcium absorption with age is a decrease in intestinal responsiveness to 1,25(OH)2D. In rat studies, calcium uptake by isolated duodenal cells, the number of intestinal receptors for 1,25(OH)2D, the calbindin response to 1,25(OH)2D, and the effect of 1,25(OH)2D therapy on calcium absorption have been reported to decline with age, but there is disagreement about the relative importance of low serum 1,25(OH)2D, and 1,25(OH)2D resistance in this model [87–90]. In human studies, there are three main lines of evidence supporting a possible role of intestinal 1,25(OH)2D resistance. First, some studies have shown an age-related increase in the level of 1,25(OH)2D in association with unchanged calcium absorption [34,91]. Second, direct measurements of tissue 1,25(OH)2D receptor (VDR) content in the duodenal mucosa have shown an age-related decrease of about 30% between age 20 and age 80 [34], although this has not been found by all investigators [92]. Third, the positive relationship between fractional calcium absorption and an index of free 1,25(OH)2D

1105

CHAPTER 67 Vitamin D and Osteoporosis

Young 1.2

Elderly 1.2

r = 0.63, P = 0.003

r = 0.35, P = 0.142

0.8

FCA

FCA

0.8

0.4

0

0.4

0

2

4

6

8

1,25(OH)2D/DBP × 105

0

0

2

4

6

8

1,25(OH)2D/DBP × 105

FIGURE 2 Relationship between fractional calcium absorption (assessed by double isotope method with 100 mg calcium carrier) and the serum free 1,25(OH)2D index (1,25(OH)2D/Vitamin D binding protein) in 20 young adult women and in 19 elderly normal women. Experimental protocol induced levels of serum free 1,25(OH)2D index that ranged from below to above normal. Note that the significant correlations present in the young adult women were lost in the elderly women. From Pattanaungkul et al. [39], with permission.

levels (total 1,25(OH)2D/DBP) found in young women is not found in the elderly [39,93]. This was studied over a wide range of vitamin D levels, induced by altering the dietary calcium or by administering oral 1,25(OH)2D (Fig. 2). However, the response to an increase in endogenous 1,25(OH)2D induced by a low calcium diet was not affected by age [36]. This study measured the important variable, true fractional calcium absorption, however, this measures both active and passive absorption. It is possible that had a lower calcium carrier been used, so as to trace active transport, that a blunted response to 1,25(OH)2D might have been observed. An important recent observation is that common allelic variants in the gene encoding the vitamin D receptor might predict up to 75% of the genetic effect on bone mineral density in healthy adults, although this effect is likely to be much smaller as evaluated by a meta-analysis [94] (see Chapter 68). Women with the low bone mineral density genotype (BB) also appear to have higher bone turnover than women with the BB genotype. These differences could relate to decreased VDR expression in the BB genotype, with consequent effects on intestinal calcium absorption, PTH secretion, and bone mineralization. However, no changes in VDR

expression have been found with the common VDR genotype variations studied so far. 3. ESTROGEN DEFICIENCY AND CALCIUM ABSORPTION

The menopause is likely to account for a substantial proportion of the age-related decrease in calcium absorption in women. In a longitudinal study, Heaney et al. [75] found that the age-related decline in fractional absorption was considerably larger in women who went through the menopause between studies, compared to subjects who had no change in estrogen status. Estrogen administration increases intestinal calcium absorption [95] in postmenopausal osteoporotic women and maintains normal calcium absorption and responsiveness to 1,25(OH)2D in premenopausal women undergoing elective ovariectomy [96]. Also, estrogen may have a direct effect on uptake of calcium by duodenal cells [97]. 4. CALCIUM ABSORPTION AND THE AGING STOMACH

Atrophic gastritis and reduced gastric acid production is common among the elderly, with an estimated prevalence of 20 to 50% in the sixth and seventh decades of life [84]. Calcium bioavailability is probably not affected by high gastric pH per se, although this is controversial [85,98–100]. Although atrophic

1106 gastritis in the elderly is rarely severe enough to cause pernicious anemia, patients with pernicious anemia do appear to have reduced bone mineral density [101] and are at increased risk of osteoporotic fracture [102], even though calcium absorption from food is usually unimpaired [85,101]. It has been suggested that bone loss in pernicious anemia could be due to decreased secretion of a bone-stimulating factor produced by the gastric mucosa [102]. These findings raise the intriguing possibility that changes in the aging stomach could be associated with bone loss independently of any effect on calcium absorption. Partial gastrectomy is also associated with bone mineral loss, and it has been suggested that this could be related to reduced postprandial calcitonin secretion [103].

B. Which Form of Vitamin D Is the Major Determinant of Calcium Absorption? The most potent natural metabolite of vitamin D is 1,25(OH)2D, and its potency ratio to 25OHD is approximately 2000:1. However, the circulating level of 25OHD is almost 1000 times higher than 1,25(OH)2D. It has been calculated by Barger-Lux et al. [104] that 25OHD contributes about 25% of the circulating vitamin D activity. Colodro et al. [105] tested the effect on calcium absorption in renal failure of different metabolites of vitamin D and calculated that the potency ratio in humans may be closer to 400:1. This would indicate that 25OHD contributes up to 90% of the circulating vitamin D activity. This would explain why several groups have found stronger relationships between 25OHD and calcium absorption fraction than between 1,25(OH)2D and calcium absorption fraction [104,106]. However, these studies were done before it was recognized that 1α-hydroxylase is present in many target tissues, including intestine, and that local conversion of 25OHD to 1,25(OH)2D may confound these calculations.

V. AGE AND PARATHYROID HORMONE A. Effect of Age on Parathyroid Function The principal immunoreactive form of PTH in the plasma is heterogeneous C-terminal fragments, especially when renal function is impaired. Several crosssectional studies using fragment-detecting assays showed increases in the concentration of PTH with age [11,12,30,68,107–112]. These studies are difficult to interpret, because renal function declines with age, resulting in increased retention of C-terminal fragments. However, many more recent studies have confirmed an age-related increase in intact PTH in women, which is probably progressive after the menopause

RICHARD EASTELL AND B. LAWRENCE RIGGS

[12,37,82,91,111,113–117], although a few studies have not found this to be the case [118,119]. Serum intact PTH also increases with age in men [120,121]. In concurrent studies, the age-related increase in carboxyl-terminal PTH is about fourfold higher than the increase in intact PTH [12]. The effects of age on the level of intact PTH are modest, with average reported increases of about 10% per decade after the menopause. The rise in intact PTH is probably similar in men [37]. Bioactive PTH measured using an adenylate cyclase bioassay also appears to increase with age [122]. Moreover, there is evidence of increased PTH activity with age, based on excretion of cyclic AMP [11,109–111], and a decrease in the theoretical renal phosphate threshold, TmP/GFR [107,109,111], between the second and the ninth decade of life. This decrease in TmP/GFR is the likely cause of the decreased serum phosphorus discussed previously. It may be important to consider the pattern of PTH release when considering the effect of age on PTH secretion. PTH secretion shows a marked circadian rhythm, and the form of this rhythm may alter with aging [123]. Pulsatility of PTH secretion has also been shown in several studies using frequent sampling techniques [124–126], and it has been postulated that the pattern of pulsatility may govern the biological action of PTH. The pulse amplitude of PTH secretion may be markedly reduced in osteoporosis [124]. The effect of aging on the pattern of PTH release has not been studied, but this is clearly of some interest.

B. Mechanisms for Increased Intact PTH with Aging Theoretically, increased PTH levels could be due to hypocalcemia, a greater parathyroid cell mass (parathyroid hyperplasia), a shift in the set-point of calcium stimulated PTH secretion, or decreased PTH clearance. 1,25(OH)2D can alter PTH secretion by altering the set-point [127–130], as well as the maximum secretory capacity [131]. It is possible that subtle hypocalcemia is an important cause of age-related changes in PTH secretion. Total and ionized calcium in serum decrease slightly with age in most studies (see earlier). Dawson-Hughes et al. showed that healthy postmenopausal women with levels of serum ionized calcium in the lowest quintile had significantly higher levels of intact PTH than subjects with ionized calcium in the highest quintile [132]. A similar inverse relationship between ionized calcium and intact PTH has been shown by other investigators [12,133]. In rats, however, the age-related increase in PTH does not appear to be related to low plasma calcium [71,134].

CHAPTER 67 Vitamin D and Osteoporosis

Aging could also be associated with an increase in the set-point for suppression of PTH secretion by calcium (implying higher PTH secretion at any given concentration of ionized calcium). A shift in set-point could be a primary effect of aging on the sensitivity of the parathyroid gland to calcium, or could be secondary to other factors, such as low 1,25(OH)2D or decreased parathyroid responsiveness to 1,25(OH)2D. However, Landin-Wilhelmsen et al. [7] reported a positive correlation between 1,25(OH)2D and PTH, and this would argue against a decrease in 1,25(OH)2D as a cause of the increase in PTH. They did report a negative correlation between PTH and 25OHD (as did Hegarty et al. [135]), and this again raises the question of the relative contribution of 25OHD to the biologicial activity of vitamin D (see above). Studies in rats have shown that PTH secretion is higher at any given level of calcium in aged animals in vivo [136,137] and in vitro [138]. However, in elderly human subjects, the set-point does not appear to change with aging, but basal secretion and maximum secretory capacity are increased [131]. This suggests that the rise in PTH with aging is due to parathyroid hyperplasia, possibly secondary to chronic hypocalcemia, rather than an altered set-point. Furthermore, in this study the magnitude of the decrease in basal and maximal PTH secretion after 1,25(OH)2D therapy was similar in young and elderly subjects, suggesting that parathyroid responsiveness to 1,25(OH)2D is preserved in the elderly. A number of studies have shown a significant inverse relationship between levels of intact PTH and 1,25(OH)2D in elderly subjects [12,48], but this may be secondary to impaired calcium absorption rather than an effect of low 1,25(OH)2D itself. Treatment of elderly subjects with vitamin D results in a reduction in the level of PTH [18,139,140]. A possible relationship between declining renal function and increase in intact PTH has been explored in several studies. Some studies have shown that the level of intact PTH is independently related to declining renal function [60,116]. In contrast, other studies in rats [71,116,134] and humans [12,59] have shown that declining renal function is not independently associated with the age-related increase in intact PTH. There is no evidence for a change in the clearance of intact PTH with age in man, and studies in rats have suggest that a decline in PTH clearance cannot account for high levels of PTH with advancing age [134].

VI. INTERRELATIONSHIPS BETWEEN AGE-RELATED BONE LOSS, ESTROGEN, PTH, AND VITAMIN D METABOLISM The age-related increase in PTH may play a pivotal role in age-related bone loss [141]. In this model,

1107 decreased net intestinal calcium absorption (resulting from dietary calcium deficiency and reduced levels of 1,25(OH)2D) results in mild hypocalcemia and secondary hyperparathyroidism. PTH excess results in increased bone resorption and net bone loss (see Figure 3). Although it is clear that PTH and vitamin D metabolism alter with aging, the relationship between these changes and bone loss is not certain. Inappropriately elevated PTH in primary hyperparathyroidism is associated with reduced bone density [142,143] but it is uncertain whether mild asymptomatic hyperparathyroidism is associated with an increased risk of fracture or progressive bone loss [144–146]. There are conflicting data on whether the age-related increase in PTH or decrease in 25OHD and 1,25(OH)2D are causally related to the age-related increases in bone turnover and decreases in BMD. Population-based studies from Rochester, MN have shown that serum PTH is positively correlated with bone turnover markers and negatively correlated with BMD at several sites [120,147]. Moreover, when young adult premenopausal women and elderly postmenopausal women were compared, equalization of serum PTH levels by calcium infusion lead to an equalization of bone turnover markers, suggesting a causal relationship [148]. Others have shown a relationship between BMD and the level of 1,25(OH)2D [32,149], PTH and 25OHD [149]. Ooms et al. [17] reported that 25OHD levels did correlate with hip BMD below the threshold of 30 nmol/l [17]. In one study, PTH did relate to rates of bone loss from the forearm in premenopausal women [150]. Others have failed to find significant correlations between serum PTH or vitamin D metabolites and age-related changes in bone turnover [117], bone mineral density [13,37,59,114], or bone loss [151]. These differences may be explained in part by the use of samples that were not population-based, the selection of healthy volunteers, or the lack of statistical power. In contrast to the above findings, some studies have shown a relationship between bone mineral density and the level of 1,25(OH)2D. It is likely that differences in net calcium absorption account for only a small proportion (possibly about 10%) of the variance in bone density in the elderly population [152]. Pharmacological doses of calcium may reduce the rate of bone loss in elderly subjects [153,154], especially in subjects with a low habitual calcium intake. However, most of the variance in the rate of bone loss is likely to be independent of intestinal calcium supply. Orwoll et al. found that healthy men (30 to 87 years old) show substantial bone loss in the radius (1.0%/year) and spine (2.3%/year), which is not altered by calcium supplementation (1000 mg/day) or vitamin D (25 mg/day) [155].

1108

RICHARD EASTELL AND B. LAWRENCE RIGGS

Decreased bone formation (cellular level)

Intestinal resistance to 1,25(OH)2D

Decreased 25OHD 1α-hydroxylase activity

Decreased 25(OHD)

Decreased 1,25(OH)2D production

Decreased calcium absorption

Secondary hyperparathyroidism

Bone loss

FIGURE 3 A model for the proposed changes in calcium homeostasis and bone turnover with age.

Devine et al. reported that calcium absorption measured using the single-tracer stable strontium method did not correlate with bone density in women more than 10 years past the menopause [86]. Riggs et al. [156,157] have proposed that the increase in PTH with aging may be a consequence of estrogen deficiency. The strongest evidence for this is the effect of estrogen on PTH and bone resorption in older women—both are decreased [158] (Table I). This could be explained by an effect of estrogen on the intestine and kidney. The intestine has been shown to contain functional estrogen receptors [159] and estrogen increases intestinal calcium absorption both in rodents [159] and in humans [96,160]. This effect is a consequence of upregulation of the epithelial calcium channel (CaT1) in the duodenum, and the mechanism is independent of vitamin D [161]. In women who had been ovariectomized six months earlier, the responsiveness of calcium absorption to short-term treatment with 1,25(OH)2D was blunted, and this was restored by estrogen replacement [96].

TABLE I Comparison of 30 Young Premenopausal Women (Pre M), 30 Untreated Postmenopausal Women (Post M), and 30 (Post M) Women Receiving Long-Term Estrogen Therapy (after McKane et al. [158]) Variables Number Age, years Estrogen status Serum PTH, pmol/L Urinary NTX, nmol/mmol creatinine

Pre M

Post M

30 32 Replete 2.7 29

30 74 Untreated 3.6** 43***

Post M + estrogen 30 74 ERT 2.5 25

**, p <0.01, ***, p<0.001, comparison of Pre M and Post M. There were no differences between Pre M and Post M + estrogen. ERT, estrogen replacement therapy.

1109

CHAPTER 67 Vitamin D and Osteoporosis

Estrogen treatment in elderly postmenopausal women increases renal tubular reabsorption of calcium [162].

VII. SUMMARY OF CHANGES IN VITAMIN D WITH AGING Thus, with age there are a number of changes in vitamin D and its actions. 1. There is a decrease in 25OHD that probably results from decreased UV light exposure, decreased effect of UV light on skin synthesis of vitamin D, and decreased absorption of vitamin D from the diet. 2. There is a decrease in 1,25(OH)2D after the age of 65 years that is partly a result of the decrease in the substrate, 25OHD, and partly a result of the decrease of renal lα-hydroxylase activity. 3. There is a decrease in active calcium absorption that results from the decrease in vitamin D metabolites (25OHD and 1,25(OH)2D), from intestinal resistance to the action of vitamin D, and to estrogen deficiency. 4. There is an increase in PTH secretion. There are further changes in patients with osteoporosis-related fractures, and these indicate different pathogenetic factors for these fracture syndromes, supporting the concept of type I and type II osteoporosis. In type I osteoporosis, the major fractures are in the forearm and vertebra, and postmenopausal women

under the age of 75 years are affected. In type II osteoporosis the major fractures are of the proximal forearm and vertebra, and both men and women over the age of 70 years are affected. 1. In type I osteoporosis, there is a decrease in both calcium absorption and 1,25(OH)2D, but the PTH levels are lower than expected, indicating that the primary defect is likely to be an increase in net bone resorption (see Fig. 4). 2. In type II osteoporosis, there is an increase in PTH and a decrease in 25OHD that is greater than these changes that occur in all subjects with aging. This supports the importance of borderline vitamin D deficiency as an important cause of hip fracture.

VIII. TREATMENT OF ESTABLISHED OSTEOPOROSIS WITH VITAMIN D A. Rationale and General Principles A distinction must be made between the use of physiological replacement dosages of vitamin D to treat nutritional deficiency and the use of therapeutic dosages of vitamin D or natural vitamin D metabolites to treat osteoporosis when vitamin D stores are normal. Physiologic replacement of vitamin D can be achieved with small dosages of 1000 IU per day or less. Because vitamin D plays a key role in the regulation of calcium metabolism and in the maintenance of bone mass,

Oestrogen deficiency

Direct renal effects

Skeletal effects

Direct intestinal effects

Bone loss

Other factors

Decreased PTH secretion

Decreased 1,25(OH)2D production

Decreased calcium excretion

Increased calcium excretion

FIGURE 4 A model for the proposed changes in calcium homeostasis and bone turnover in type I osteoporosis

1110 deficiency states should always be searched for and, when present, corrected with relatively small dosages of vitamin D. As reviewed earlier in this chapter, there is substantial evidence that many patients with osteoporosis have an impairment in the metabolism of vitamin D to its physiologically active metabolite, 1,25(OH)2D, which contributes to their negative calcium metabolism and bone loss. This impairment can be overcome by using large dosages of vitamin D (10,000 IU to 25,000 IU per day) or by small dosages of 1,25(OH)2D3. Both vitamin D and its main circulating metabolite, 25OHD are stored in muscle and fat and over time large amounts can be retained. Because of these large stores, severe hypercalcemia and hypercalciuria due to vitamin D intoxication can occur relatively suddenly when pharmacologic dosages of vitamin D are used long-term and, when it does occur, it may last for weeks or even months. In contrast, 1,25(OH)2D does not have significant longterm storage and if hypercalcemia or hypercalciuria occur, they rapidly resolve over a few days. Because of this and because physiological or near physiological dosages are effective in increasing calcium absorption [163], 1,25(OH)2D3 is preferred over pharmacologic dosages of vitamin D in the treatment of osteoporosis. Calcium absorption is impaired in both postmenopausal and age-related osteoporosis, and it can be corrected by therapy with 1,25(OH)2D3. If increasing calcium absorption were the only effect of 1,25(OH)2D3 in the treatment of osteoporosis, large dosages of oral calcium might effectively substitute for it. However, as reviewed elsewhere in this book (Section III), 1,25(OH)2D3 also enhances renal calcium conservation, increases the differentiation of bone cells, acts directly on muscle, and has many other beneficial effects. Finally, from a therapeutic standpoint, it should be recognized that the effects of 1,25(OH)2D3 on calcium metabolism and bone turnover are triphasic. In dosages below 0.5 µg per day, the drug fails to increase calcium absorption in many subjects [163], and thus, these dosages should be considered to be incompletely effective. In dosages above 0.75 µg per day, some subjects will have increases in urine and serum calcium levels, particularly if the intake of calcium is not restricted, and there also may be evidence of increased bone resorption mediated by stimulation of osteoclasts by 1,25(OH)2D3. In dosages above 1.0 µg per day, these adverse effects are much more common. The most favorable results from 1,25(OH)2D3 treatment of established osteoporosis have been achieved using dosages in the intermediate range of 0.5 µg per day to 0.75 µg per day. Thus, the optimal therapeutic range appears to be quite narrow and may vary among patients. This restricted therapeutic range constitutes one of the

RICHARD EASTELL AND B. LAWRENCE RIGGS

major limitations in the use of this drug for treatment. Differences in dosages employed in different clinical trials may explain some of the inconsistent and conflicting therapeutic results reported in the scientific literature. 1. CORRECTION OF NUTRITIONAL VITAMIN D DEFICIENCY

Although vitamin D deficiency classically is associated with osteomalacia, milder deficiency states can lead to osteoporosis [164]. As reviewed elsewhere in this book (Chapters 61 and 62), states of mild vitamin D deficiency may be more common than has been believed previously, particularly among the elderly who often have poorer diets and have decreased solar exposure. This is particularly likely to occur in those counties that do not supplement dairy products with vitamin D, in countries with more northerly latitudes, and during the winter months. There is considerable uncertainty about the level of 25OHD that merits vitamin D replacement. Malabanan et al. [165] treated 35 patients for 8 weeks with 50,000 IU vitamin D orally per week. They stratified the baseline 25OHD to find out if there is some threshold above which there is no benefit, as assessed by suppression of PTH. They identified 50 nmol/L as such a threshold. Jesudason et al. [166] identified a similar threshold (60 nmol/L) by examining the level of 25OHD below which bone turnover markers (and PTH) increased in older women attending a metabolic bone center. Supplementation with vitamin D resulted in increases in calcium absorption up to a level of 25OHD of 80 nmol/L [167]. Thus, the threshold of 30 nmol/L mentioned earlier in the chapter may need to be revised upwards [15]. Chapuy et al. [140] randomly treated 3,270 elderly French women who were confined to either nursing homes or apartments for the elderly with either 800 IU of vitamin D and 500 mg of elemental calcium per day or double placebo. After 18 months, the numbers of hip fractures were 43 percent lower and nonvertebral fractures were 32% lower and serum parathyroid hormone was 44% lower, all statistically significant changes (see Chapter 66). Heinkinheimo et al. [168] treated or followed without treatment a group of 621 elderly (>85 yrs of age), free living or institutionalized men and women with annual intramuscular injections of ergocalciferol over a four-year period. The treatment group had a slightly, but significantly lower (16%) fracture rate than did the control group (22%). Ooms et al. [169] treated 348 elderly Dutch women with 400 IU per day of vitamin D or placebo and found that there were small, but significant decreases in serum PTH and increases in femoral neck density. Lips et al. [170] treated 2,578 elderly (>80 yrs of age) women and men

CHAPTER 67 Vitamin D and Osteoporosis

who were living independently or in nursing homes with 400 IU of placebo. Over 3.5 years of observation, there were no differences in the number of hip or total fractures. The major difference between this study and that of the French study was that the subjects in Chapuy’s study had lower levels of serum 25(OH)D and received a higher dosage of vitamin D and a calcium supplement. Trivedi et al. [171] recently reported that oral vitamin D given 100,000 IU orally every four months to men and women over 65 years for 5 years reduced the risk of fractures. Dawson-Hughes et al. [172] treated men and women over the age of 65 years with calcium (500 mg/day) and vitamin D (700 IU/day) for 3 years. There was prevention of bone loss and a reduction in the risk of nonvertebral fractures. From these and other studies, it appears that many elderly persons have varying degrees of vitamin D deficiency and that correction of this with small replacement dosages is beneficial. Lips et al.[173] found that an average of 300 IU per day of vitamin D was sufficient to correct vitamin D deficiency in elderly women. However, because of evidence that the elderly may absorb vitamin D less well than the young and because low dosages are safe and inexpensive, it would seem prudent to ensure that elderly individuals take 800 IU per day of vitamin D and have a calcium intake of 1,000 mg per day or more. 2. TREATMENT WITH 1,25(OH)2D3

Despite more than 15 years of clinical trials, 1,25(OH)2D3 still has not been approved for the treatment of established osteoporosis in the United States. It is approved, however, in the United Kingdom, Australia, Italy, Japan, New Zealand, as well as in 16 other countries. However, 1,25(OH)2D3 has been approved in most countries for treatment of hypocalcemia and renal osteodystrophy, and thus is available to physicians for “off label” usage.

B. Efficacy in Postmenopausal Osteoporosis Only randomized, prospective clinical trials with objective endpoints will be reviewed here. The earliest evaluation of 1,25(OH)2D3 in postmenopausal osteoporotic women was reported by Gallagher et al. [174] who compared the effect of six months treatment with 0.5 µg per day of 1,25(OH)2D3 or placebo. As compared with either baseline measurements or with changes in the placebo group, they found that 1,25(OH)2D3 significantly increased calcium absorption, decreased bone resorption rate, and improved calcium balance. The most systematic study reported thus far was a large

1111 three-center study comparing the effects of 1,25(OH)2D3 on the rate of change in bone density in postmenopausal osteoporotic women. All patients had their calcium intake adjusted to 1000 mg per day. Unfortunately, the protocol required that the dosage of 1,25(OH)2D3 be increased until toxicity occurred and then reduced to a dosage that did not produce hypercalcemia or hypercalciuria. In the centers of Aloia et al. [175] and Gallagher et al. [176], there were significant increases in bone density, whereas in the center of Ott and Chesnut [177], there were no significant differences from the placebo group. However, Ott and Chesnut enrolled patients with milder disease (the mean number of vertebral fractures in their study was 1.2 versus 2.9 and 4.1 in the other two studies), and they had reduced the mean daily dosage of 1,25(OH)2D3 to a much lower dosage (0.43 µg versus 0.80 µg and 0.67 µg in the other two centers). Subsequently, Ott and Chesnut [177] reanalyzed their data and found that those subjects who received 0.5 µg per day or more responded by increases in bone density, whereas those with lower dosages did not. Orimo et al. [178] found in 80 postmenopausal osteoporotic Japanese women that one year of treatment with 1α-hydroxyvitamin D [1α(OH)D] increased bone density by 1.8% at the lumbar spine and by 4.6% at the femoral trochanter as compared with the placebo group, whereas there was no significant change at the femoral neck. Christiansen et al. [179] failed to demonstrate a significant effect of a small dosage of 0.25 µg per day of 1,25(OH)2D3 on retarding bone loss in perimenopausal normal women. However, Need et al. [180] found that treatment of postmenopausal osteoporosis for 15 months with a dosage of 0.25 µg per day combined with a calcium supplement of 1000 mg per day reduced urine hydroxyproline excretion and increased bone density as compared with a control group. Some studies have reported that 1,25(OH)2D3 treatment decreases the occurrence of vertebral fractures. In a two-center double-blind study over three years, 1,25(OH)2D3 resulted in a 65% reduction in vertebral fractures, although only the first year was placebocontrolled [181]. Tilyard et al. [182] compared the effect of 1,25(OH)2D3 treatment in postmenopausal osteoporotic women with a control group receiving only calcium supplementation. At the end of three years, they found a significant 63 percent decrease in vertebral fractures compared to the control group. However, the data were only significant when the results of the first year and subjects having more than five fractures were excluded. Also, inexplicably, the vertebral fracture rate increased in the group receiving calcium supplementation. Finally, Orimo et al. [178] found that one year’s treatment of 80 Japanese women

1112 with 1α(OH)D3 reduced significantly the vertebral fracture rate by 83% as compared with placebo. Thus far, there has been no controlled study of the effect of 1,25(OH)2D3 in the treatment of elderly women with age-related (type II) osteoporosis. Gallagher et al. [160] recently have reported the results of a three-year double-blind prospective controlled trial in 489 elderly women randomized in a factorial design to treatment with placebo, 1,25(OH)2D3 (0.5 µg/day), estrogen replacement therapy (ERT) or ERT, plus 1,25(OH)2D3. In compliant subjects, they found that the 1,25(OH)2D3 treatment group had a greater increase in BMD (∼2%, P<0.005) at various scanning sites than the placebo group, although the group receiving ERT had a substantially greater increase in BMD. However, the group receiving both 1,25(OH)2D3 and ERT had the greatest increase in BMD. Interestingly, the two groups receiving 1,25(OH)2D3 had 50% fewer fractures than the two groups who did not receive 1,25(OH)2D3. Because of the factorial nature of the design, the relative effects of 1,25(OH)2D3 and estrogen on BMD could be calculated. Using this approach, the estimated proportion of the bone effect of estrogen that was mediated by 1,25(OH)2D3 was 34.5% for the mean of four scanning sites, whereas the proportion of the bone effect of 1,25(OH)2D3 that was independent of estrogen was estimated to be 18.3%. Although this study provides useful data, a major caveat is that subjects were excluded who had BMD values below the age-adjusted normal range. In previous studies, we have found that osteopenic and osteoporotic patients have the greatest abnormalities of calcium absorption and vitamin D metabolism [5,183]. Thus, these must be considered to be minimal estimates, and the true impact of primary abnormalities of vitamin D metabolism and action is likely to be larger. There have been few studies comparing vitamin D preparations with bisphosphonates for the treatment of osteoporosis. Shiraki et al. [184] studied 210 patients with osteoporosis and compared alendronate (5 mg/day) with alfacalcidol (1 µg/day) for 48 weeks and found significantly greater increases in BMD with alendronate than with alfacalcidol.

C. Meta-analysis of the Clinical Trials of Vitamin D Papadimitropoulos et al. [185] performed a metaanalysis of standard vitamin D (vitamin D2 and D3, and 25OHD) and “active” vitamin D [1,25(OH)2D3 and 1-alpha(OH)D] for the period 1966 to 1999 using predefined criteria. They found eight suitable studies for vertebral fracture and six suitable studies for

RICHARD EASTELL AND B. LAWRENCE RIGGS

nonvertebral fracture, and the results are shown in Table II. These show a decrease in the risk of vertebral fracture and possibly nonvertebral fracture, although most of the studies of vertebral fractures were with “active” vitamin D, and most of those of nonvertebral fracture were with standard vitamin D. They reported more consistent effects of “active” vitamin D on bone mineral density than vitamin D. However, not all the studies quoted above were captured by this meta-analysis [171] and so we look forward to an update. They found that patients were more likely to discontinue vitamin D than placebo as a result of adverse effects or abnormal laboratory tests (presumably raised serum calcium).

D. Safety This remains a concern because of the relatively narrow therapeutic range for this drug. With dosages higher than 0.75 µg per day, hypercalciuria and mild hypercalcemia may occur in a minority of patients, particularly in those with unrestricted calcium intake. Thus far, significant deterioration of renal function has not been reported in subjects receiving recommended dosages, and the occurrence of kidney stones is rare. With the recommended dosages of 0.5 to 0.75 µg per day and with calcium intake restricted to 1000 mg per day, the drug appears to be quite safe. Thus, Tilyard [182] found no evidence of hypercalcemia, hypercalciuria, or nephrocalcinosis after 528 patient-years of observation. Nonetheless, monitoring of serum and urine calcium values no less than yearly is recommended. 1. CALCIUM AND VITAMIN D AS SUPPLEMENTS TO OTHER TREATMENTS

Almost all of the Phase III clinical trials for the treatment of postmenopausal osteoporosis included a calcium supplement and most included vitamin D [186]. However, when these treatments were approved by the regulatory authorities, the only recommendation made was that the diet should contain sufficient calcium and vitamin D. There is thus an important question as to whether the calcium and vitamin D is simply a placebo in these studies, and whether women with osteoporosis commonly attain adequate intakes of calcium and vitamin D without supplementation. There have been many clinical trials of drugs for postmenopausal osteoporosis. A summary of the major clinical trials is included in Table III. It can be seen that the usual dose of calcium supplementation was between 500 and 1000 mg/day and that of vitamin D between 250 and 600 IU/day. These amounts of vitamin D would

1113

CHAPTER 67 Vitamin D and Osteoporosis

TABLE II Meta-analysis of Fracture Trials with Vitamin D [185] Vertebral fracture, relative risk (95% confidence interval)

Reference

a. Standard Vitamin D (vitamin D or 25-hydroxyvitamin D) Baeksgaard [197] 0.33 (0.01 to 8.05) Chapuy [140] Lips [170] Dawson-Hughes [172] Pooled estimate b. “Active” Vitamin D (1,25(OH)2D3 and 1-α-OHD3) Gallagher [198] 0.90 (0.42 to 1.89) Orimo [178] 0.37 (0.09 to 1.44) Ott [177] 1.4 (0.59 to 3.62) Tilyard [182] 0.43 (0.31 to 0.61) Guesens [199] 0.88 (0.43 to 1.80) Orimo [200] 0.46 (0.31 to 0.69) Caniggia [201] 0.20 (0.01 to 3.54) Pooled 0.64 (0.44 to 0.92)

be expected to increase serum 25-hydroxyvitamin D (25-OHD) by between 6 and 15 nmol/L [187]. The recommendation in the Fracture Intervention Trial (FIT) study was that a food frequency questionnaire be used to evaluate dietary calcium [188]. If the dietary calcium was below 1000 mg/day, then the patient should be given 500 mg/day calcium per day. About 82% of women with osteoporosis failed to achieve this level. In the Vertebral Efficacy with Risedronate Therapy (VERT) studies, the recommendation was to measure 25OHD and if the value was below 40 nmol/L to supplement with vitamin D (up to 500 IU/day). About 36% of patients in the VERT multinational study required supplementation [189]. This threshold is quite low—it is becoming accepted that a threshold of

Nonvertebral fracture, relative risk (95% confidence interval)

0.75 (0.61 to 0.91) 1.04 (0.77 to 1.41) 0.45 (0.22 to 0.91) 0.78 (0.55 to 1.09)

1.10 (0.02 to 2.0) 2.20 (0.52 to 9.24) 0.50 (0.25 to 1.00)

0.87 (0.29 to 2.59)

FIT 1 and 2 [188,202] VERT MN and NA [189,203] PROOF [204] MORE [205]

Antiresorptive Drug Alendronate Risedronate Calcitonin Raloxifene

180 3270 1916 213 5399

50 80 86 622 32 86 14 970

80 nmol/L is desirable to prevent the biochemical changes of subclinical osteomalacia (high parathyroid hormone and bone turnover markers) [187]. Thus, it is likely that most people receiving treatment for osteoporosis are in need of supplementation with calcium and vitamin D if the conditions under which these trials were conducted are to be reproduced.

E. Rationale for Supplementation There is a positive calcium balance after starting therapy with antiresorptive therapy. The total body bone mineral content increases by 1% at one year after treatment with alendronate plus calcium (500 mg/day) [190].

TABLE III The Level of Calcium and Vitamin D Supplementation in Several Large Clinical Trials of Postmenopausal Osteoporosis Name of trial

n

Calcium, mg/day 500* 1000 1000 500

*If dietary calcium less than 1000 mg/day (in 82% of subjects in FIT 1) ** If 25OHD less than 40 nmol/L (in 36% of subjects in VERT-MN)

Vitamin D, IU/day 250 Up to 500** 400 Up to 600

1114 This is equivalent to a positive balance of about 20 mg/day. In women given 500 mg/day of calcium alone, there was 0.5% decrease in total body mineral content, and this would be equivalent to a negative calcium balance of about 10 mg/day, so the extra calcium requirement resulting from alendronate 10 mg/day administration would be 30 mg/day. This could be met by an adaptation of the body by an increase in the fraction of dietary calcium absorbed [191]. However, this mechanism is less effective in the elderly. It could be met by the calcium supplement—if about 10% were absorbed, then an extra 50 mg/day of calcium would be available to support the positive calcium balance. Vitamin D deficiency prevents the antiresorptive effect of bisphosphonates and impairs muscle strength. Koster et al. [192] recruited 28 osteopenic patients with vitamin D deficiency (25OHD < 40 nmol/L) and compared their BMD response to one year of cyclical etidronate with a control group with normal 25OHD. The control group showed an increase in BMD of 2 to 6% in response to cyclic etidronate, depending on the site of measurement. The vitamin D–deficient group had little or no improvement in the BMD, and further deterioration of the vitamin D deficiency, as indicated by a decrease in serum calcium and an increase in parathyroid hormone and osteocalcin. It is likely that the vitamin D deficiency prevented the increase in calcitriol and hence calcium absorption in response to the etidronate therapy. There is some evidence that vitamin D therapy might reduce the risk of falls in postmenopausal women (see Chapter 102). Pfeifer et al. [193] reported that 148 women with low 25OHD (<50 nmol/L) were given a calcium supplement (1000 mg/day) and randomized to vitamin D (800 IU/day) or placebo for one year. The vitamin D resulted in decreased serum PTH, a decrease in body sway, and in the average number of falls per subject. Bischoff et al. [194] reported that 122 elderly women were given a calcium supplement (1200 mg/day) and randomized to vitamin D (800 IU/day) or placebo for 3 months. The vitamin D resulted in a decrease in the average number of falls per subject (by 49%), and an improvement in musculoskeletal function. These studies both assessed the number of falls and not the number of subjects who fell and this is a limitation. Nonetheless, this reduction in the risk of falling could translate into a decrease in the risk of fracture. Thus, almost all of the clinical trials of treatments for osteoporosis included calcium and/or vitamin D as supplements. These supplements are more effective than placebo and act in their own right to reduce the risk of fracture. It is likely, although not proven, that they add to the antifracture effects of the antiresorptive drugs with which they are given. It would appear that the dose that

RICHARD EASTELL AND B. LAWRENCE RIGGS

would help reduce fractures and strengthen muscle would be 1000 mg of calcium and 800 IU of vitamin D per day. This type of supplementation is now commonly included in guidelines for the management of osteoporosis [195].

IX. CONCLUSIONS Although data in the literature are conflicting, the large majority of studies indicate that 1,25(OH)2D3 is moderately effective in reducing bone loss and possibly also in reducing vertebral fracture occurrence in women with type I osteoporosis. Efficacy seems to be dose-related with the best results occurring in patients treated with 0.5 to 0.75 µg per day. The drug appears to be quite safe in this dosage range, although restriction of calcium intake and monitoring serum and urine values for calcium is recommended. The osteoporotic women who are most likely to benefit from 1,25(OH)2D3 are those with impaired calcium absorption. Unfortunately, it is not practical to measure calcium absorption directly. However, Riggs and Nelson [183] correlated calcium absorption and urinary calcium excretion and found that most osteoporotic women who had intestinal calcium malabsorption on a normal calcium diet had values for urine calcium excretion below 100 mg per day. Calcium and vitamin D should be considered standard adjuncts to treatment with antiresorptive therapy, such as bisphosphonates. Many of the elderly patients with osteoporosis are vitamin D–deficient and should be treated with small dosages of vitamin D. For the remainder, treatment with small dosages of calcitriol, 0.25 µg per day combined with 1000 mg per day of elemental calcium seems rational because of the documented decrease in serum 1,25(OH)2D levels in many patients. However, the value and safety of this drug in these elderly patients clearly needs better documentation. The other area that requires further development is the use of synthetic metabolites of vitamin D. Shevde et al. [196] have described the effects of 2-methylene19-nor-(205)-1,25(OH)2D3 in rats. There is a potent effect on osteoblasts and a large increase in bone mineral density with little change in calcium absorption (see Chapter 87). These selective vitamin D receptor modulators may prove to be useful therapies for the treatment of osteoporosis.

References 1. Aksnes L, Rodland O, Odegaard OR, Bakke KJ, Aarskog D 1989 Serum levels of vitamin D metabolites in the elderly. Acta Endocrinol 127:27–33.

CHAPTER 67 Vitamin D and Osteoporosis

2. Baker MR, Peacock M, Nordin BEC 1980 The decline in vitamin D status with age. Age & Aging 9:249–252. 3. Barraguy JM, France MW, Corless D, Gupta SP, Switala S, Boucher BJ et al. 1978 Intestinal cholecalciferol absorption in the elderly and in younger adults. Clin Sci Mol Biol 55: 213–220. 4. Chapuy MC, Durr F, Chapuy P 1983 Age-related changes in parathyroid hormone and 25 hydroxycholecalciferol levels. J Gerontol 38:19–22. 5. Gallagher JC, Riggs BL, Eisman J, Hamstra A, Arnaud SB, DeLuca HF 1979 Intestinal calcium absorption and serum vitamin D metabolites in normal subjects and osteoporotic patients. J Clin Invest 64:729–736. 6. Guggenheim K, Kravitz M, Tal R, Kaufmann NA 1979 Biochemical parameters of vitamin D nutrition in old people in Jerusalem. Nutr Metab 23:172–178. 7. Landin-Wilhelmsen K, Wilhalmsen L, Wilske J, Lappas G, Rosen T, Lindstedt G et al. 1995 Sunlight increases serum 24(OH) vitamin D concentration whereas 1,25(OH)2D3 is unaffected. Eur J Clin Nutr 49:400–407. 8. Lore F, Di Cairano G, Signorini AM, Caniggia A 1981 Serum levels of 25-hydroxyvitamin D in postmenopausal osteoporosis. Calcif Tiss Int 33:467–470. 9. Lund B, Sorensen OH 1979 Measurement of 25-hydroxyvitamin D in serum and its relation to sunshine, age, and vitamin D. Scand J Clin Lab Invest 39:23–30. 10. Omdahl JL, Garry PJ, Hunsaker LA, Hunt WC, Goodwin JS 1982 Nutritional status in a healthy elderly population: vitamin D. Am J Clin Nutr 36:1225–1233. 11. Orwoll ES, Meier DE 1986 Alterations in calcium, vitamin D, and parathyroid hormone physiology in normal men with aging: relationship to the development of senile osteopenia. J Clin Endocrinol Metab 63:1262–1269. 12. Quesada JM, Coopmans W, Ruiz B, Aljama P, Jans I, Bouillon R 1992 Influence of vitamin D on parathyroid function in the elderly. J Clin Endocrinol Metab 75: 494–501. 13. Tsai KS, Wahner HW, Offord KP, Melton LJI, Kumar R, Riggs BL 1987 Effect of aging on vitamin D stores and bone density in women. Calcif Tissue Int 40:241–243. 14. Weisman Y, Schen RJ, Eisenberg Z, Edelstein S, Harell A 1981 Inadequate status and impaired metabolism of vitamin D in the elderly. Isr J Med Sci 17:19–21. 15. Heaney RP 2003 Vitamin D, nutritional deficiency, and the medical paradigm. J Clin Endocrinol Metab 88(11):5107–5108. 16. Lips P 1996 Vitamin D deficiency and osteoporosis: the role of vitamin D deficiency and treatment with vitamin D and analogues in the prevention of osteoporosis-related fractures. Eur J Clin Invest 26:436–442. 17. Ooms MEW, Lips P, Roos JC, van der Vijgh WJF, PoppSnijders C, Bezemer PD et al. 1995 Vitamin D status and sex hormone binding globulin: determinants of bone turnover and bone mineral density in elderly women. J Bone Miner Res 10:1177–1184. 18. Lips P, Wiersinga A, Vanginkel FC, Jongen MJM, Netelenbos JC, Hackeng WHL et al. 1988 The effect of vitamin D supplementation on vitamin D status and parathyroid function in elderly subjects. J Clin Endocrinol Metab 67:644–650. 19. Van Der Wielaen RPJ, Lowik MRH, Van Der Berg H, de Groot LCPGM, Haller J, Moreiral O et al. 1995 Serum vitamin D concentrations among elderly people in Europe. Lancet 346:207–210. 20. McKenna MJ 1992 Differences in vitamin D status between countries in young adults and the elderly. Am J Med 93:69–77.

1115 21. Clemens TL, Zhou X-Y, Myles M, Endres D, Lindsay R 1986 Serum vitamin D2 and vitamin D3 metabolite concentrations and absorption of vitamin D2 in elderly subjects. J Clin Endocrinol Metab 63:656–660. 22. Somerville PJ, Lien JWK, Kaye M 1977 The calcium and vitamin D status in an elderly female population and their response to administered supplemental vitamin D3. J Gerontol 32:659–663. 23. Bell NH 1985 Vitamin D endocrine system. J Clin Invest 76:1–6. 24. Mutsuoka LY, Wortsman J, Hanifan N, Holick MF 1988 Chronic sunscreen use decreases circulating concentrations of 25-hydroxyvitamin D. Arch Dertmatol 124:1802–1804. 25. Maclaughlin J, Holick MF 1985 Aging decreases the capacity of human skin to produce vitamin D3. J Clin Invest 76: 1536–1538. 26. Davie M, Lawson, DEM 1980 Assessment of plasma 25hydroxyvitamin D response to ultraviolet irradiation over a controlled area in young and elderly subjects. Clin Sci 58: 235–242. 27. Holick MF, Matsuoka LY, Wortsman J 1989 Age, vitamin D and solar ultraviolet. Lancet 2:1104–1105. 28. Need AG, Morris HA, Horowitz M, Nordin BEC 1992 Skinfold thickness, age, and vitamin D. J Bone Miner Res 7(Supplement):165. 29. Dandona P, Menon RK, Shenoy R, Houlder S, Thomas M, Mallinson, WJW 1986 Low 1,25-dihydroxyvitamin D, secondary hyperparathyroidism, and normal osteocalcin in elderly subjects. J Clin Endocrinol Metab 63:459–462. 30. Fujisawa Y, Kida K, Matsuda H 1984 Role of change in vitamin D metabolism with age in calcium and phosphorus metabolism in normal human subjects. J Clin Endocrinol Metab 59:719–726. 31. Manolagas SC, Howard J, Culler F, Catherwood BD, Deftos LJ 1982 Cytoreceptor assay for 1,25(OH)2D: a simple, rapid, and reliable test for clinical application. Clin Res 30:527A. 32. Sowers MR, Wallace RB, Hollis BW 1990 The relationship of 1,25-dihydroxyvitamin D and radial bone mass. Bone & Mineral 10:139–148. 33. Tsai K-S, Heath HI, Kumar R, Riggs BL 1984 Impaired vitamin D metabolism with aging in women. Possible role in pathogenesis of senile osteoporosis. J Clin Invest 73: 1668–1672. 34. Ebeling PR, Sandgren ME, Dimagno EP, Lane AW, DeLuca HF, Riggs BL 1992 Evidence of an age-related decrease in intestinal responsiveness to vitamin D: relationship between serum 1,25 dihydroxyvitamin D3 and intestinal vitamin D receptor concentrations in normal women. J Clin Endocrinol Metab 75:176–182. 35. Dokoh S, Morita R, Fukunaga M, Yamamoto I, Torizuka K 1978 Competitive protein binding assay for 1,25-dihydroxyvitamin D in human plasma. Endocrinol Japan 25:431–436. 36. Ebeling PR, Yergey AL, Vieira NE, Burritt MF, O’Fallon WM, Kumar R et al. 1994 Influence of age on effects of endogenous 1,25-dihydroxyvitamin D on calcium absorption in normal women. Calcif Tiss Int 55:330–334. 37. Sherman SS, Hollis BW, Tobin JD 1990 Vitamin D status and related parameters in a healthy population: the effects of age, sex, and season. J Clin Endocrinol Metab 71 No 2:405–413. 38. Kinyamu HK, Gallagher JC, Balhorn KE, Petranick KM, Rafferty KA 1997 Serum vitamin D metabolites and calcium absorption in normal young and elderly free-living women and in women living in nursing homes. Am J Clin Nutr 65(3):790–797.

1116 39. Pattanaungkul S, Riggs BL, Yergey AL, Vieira NE, O’Fallon WM, Khosla S 2000 Relationship of intestinal calcium absorption to 1,25-dihydroxyvitamin D [1,25(OH)2D] levels in young versus elderly women: evidence for age-related intestinal resistance to 1,25(OH)2D action. J Clin Endocrinol Metab 85(11):4023–4027. 40. Eastell R, Yergey AL, Vieira NE, Cedel SL, Kumar R, Riggs BL 1991 Interrelationship among vitamin D metabolism, true calcium absorption, parathyroid function, and age in women: evidence of an age-related intestinal resistance to 1,25-dihydroxyvitamin D action. J Bone Miner Res 6(2):125–132. 41. Epstein S, Bryce G, Hinman JW, Miller ON, Riggs BL, Hui SL et al. 1986 The influence of age on bone mineral regulating hormones. Bone 7:421–425. 42. Vargas S, Bouillon R, Van Baelen H, Raisz LG 1990 Effects of vitamin D-binding protein on bone resorption stimulated by 1,25-dihydroxyvitamin D3. Calcif Tissue Int 47:164–168. 43. Safadi FF, Thornton P, Magiera H, Hollis BW, Gentile M, Haddad JG et al. 1999 Osteopathy and resistance to vitamin D toxicity in mice null for vitamin D–binding protein. J Clin Invest 103(2):239–251. 44. Bikle DD, Gee E, Halloran B, Haddad JG 1984 Free 1,25dihydroxyvitamin D levels in serum from normal subjects, pregnant subjects, and subjects with liver disease. J Clin Invest 74:1966–1971. 45. Prince RL, Dick I, Devine A, Price RI, Gutteridge DH, Kerr D et al. 1995 The effects of menopause and age on calciotropic hormones: a cross-sectional study of 655 healthy women aged 35 to 90. J Bone Miner Res 10:835–842. 46. Bouillon R, Auwerx J, Dekeyser L, Fevery J, Lissens W, DeMoor P 1984 Serum vitamin D metabolites and their binding protein in patients with liver cirrhosis. J Clin Endocrinol Metab 59:86–89. 47. Fraher LJ, Caveney AN, Hodsman AB 1990 Vitamin D status of elderly institutionalized Canadians: effect of season and oral vitamin D2 on circulating 1,2(OH)2D. J Bone Miner Res 5(Supplement 2):133. 48. Himmelstein S, Clemens TL, Rubin A, Lindsay R 1990 Vitamin D supplementation in elderly nursing home residents increases 25(OH)D but not 1,25(OH)2D. Am J Clin Nutr 52:701–706. 49. Falch JA, Oftebro H, Hang E 1987 Early postmenopausal bone loss is not associated with a decrease in circulating levels of 25-hydroxyvitamin D, 1,25-dihydroxyvitamin D, or vitamin D–binding protein. J Clin Endocrinol Metab 64:836–841. 50. Prince RL, Dick I, Garcia-Webb P, Retallack RW 1990 The effects of the menopause on calcitriol and parthyroid hormone: responses to a low dietary calcium stress test. J Clin Endocrinol Metab 70:1119–1123. 51. Cheema C, Grant BF, Marcus R 1989 Effects of estrogen on circulating “free” and total 1,25-dihydroxyvitamin D and on the parathyroid vitamin D axis in postmenopausal women. J Clin Invest 83:537–542. 52. Marcus R, Villa ML, Cheema M, Cheema C, Newhall K, Holloway L 1992 Effects of conjugated estrogen on the calcitriol response to parathyroid hormone in postmenopausal women. J Clin Endocrinol Metab 74:413–418. 53. Hartwell D, Hassager C, Overgaard K, Riis BJ, Podenphant J, Christiansen C 1990 Vitamin D metabolism in osteoporotic women during treatment with estrogen, an anabolic steroid, or calcitonin. Acta Endocrinol 122:715–721. 54. Kumar R, Cahan DH, Madias NE, Harrington JT, Kurtin P, Dawsonhughes BF 1991 Vitamin D and calcium transport. Kidney Int 40:1177–1189.

RICHARD EASTELL AND B. LAWRENCE RIGGS

55. Reichel H, Koeffler P, Norman AW 1989 The role of the vitamin D endocrine system in health and disease. New Eng J Med 320:980–991. 56. Spencer EM, Halloran B 1992 The dependence of PTH’s action on the renal 1-hydroxylase on insulin-like growth factor-1. J Bone Miner Res 7(Supplement 1):172. 57. St John A, Thomas MB, Davies CP, Mullan B, Dick I, Hutchison B et al. 1992 Determinants of intact parathyroid hormone and free 1,25-dihydroxyvitamin D levels in mild and moderate renal failure. Nephron 61:422–427. 58. Francis RM, Peacock M, Barkworth SA 1984 Renal impairment and its effects on calcium metabolism in elderly women. Age & Aging 13:14–20. 59. Sherman SS, Tobin JD, Hollis BW, Gundberg CM, Roy TA, Plato CC 1992 Biochemical parameters associated with low bone density in healthy men and women. J Bone Miner Res 7:1123–1130. 60. Webb AR, Pilbeam C, Hanafin N, Holick MF 1990 An evaluation of the relative contributions of exposure to sunlight and of diet to the circulating concentrations of 25-hydroxyvitamin D in an elderly nursing home population in Boston. Am J Clin Nutr 51:1075–1081. 61. Slovik DM, Adams JS, Neer RM, Holick MF, Potts JT 1981 Deficient production of 1,25-dihydroxyvitamin D in elderly osteoporotic patients. New Eng J Med 305:372–374. 62. Armbrecht HJ, Wongsurawat N, Paschal R 1987 Effect of age on renal responsiveness to parathyroid hormone and calcitonin in rats. J Endocrinol 114:173–178. 63. Bouillon R. Growth hormone and bone. Horm Res 36:49–55. 64. Spencer EM, Tobiassen O 1981 The mechanism of the action of growth hormone on vitamin D metabolism in the rat. Endocrinol 108:1064–1070. 65. Rudman D, Rao UMP 1991 The hypothalamic-growth hormone-Somatomedin C Axis: the effect of aging. In: Morley JE, editor. Endocrinology and metabolism in the elderly. Blackwell USA, 1991:35–55. 66. Zadik Z, Chalew SA, McCarter RJ, Meistas M, Kowarski AA 1985 The influence of age on the 24-hour integrated concentration of growth hormone in normal individuals. J Clin Endocrinol Metab 60:513–516. 67. Tietz NW, Shuey DF, Wekstein DR 1992 Laboratory values in fit aging individuals—sexagenarians through centenarians. Clin Chem 38(6):1167–1185. 68. Wiske PS, Epstein S, Bell NH, Queener SF, Edmondson J, Johnston CC 1979 Increases in immunoreactive parathyroid hormone with age. New Eng J Med 300:1419–1421. 69. Yendt ER, Cohanim M, Rosenberg GM 1986 Reduced serum calcium and inorganic phosphate levels in normal elderly women. J Gerontol 41:325–330. 70. Johnson JA, Goff JP, Beitz RL, Horst RL, Reinhardt TA 1992 Expression of 1,25 hydroxyvitamin D-24-hydroxylase changes with aging. J Bone Miner Res 7(Supplement 1): 150. 71. Wada L, Daly R, Kern D, Halloran B 1992 Kinetics of 1,25Dihydroxyvitamin D metabolism in the aging rat. Am J Physiol 262:E906–E910. 72. Halloran BP, Portale AA, Lonergan ET, Morris RC Jr 1990 Production and metabolic clearance of 1,25-dihydroxyvitamin D in men: effect of advancing age. J Clin Endocrinol Metab 70 No 2:318–323. 73. Carroll MD, Abraham S, Dresser CM 1983 Dietary intake source data: United States, 1976–80. Vital and health statistics. DHHS Publ 231(Series 11 (PHS)):83–1681. 74. Sheikh MS, Schiller LR, Fordtran JS 1990 In vivo intestinal absorption of calcium in humans. Miner Electrolyte Metab 16:130–146.

CHAPTER 67 Vitamin D and Osteoporosis

75. Heaney RP, Recker RR, Stegman MR, Moy AJ 1989 Calcium absorption in women: Relationships to calcium intake, estrogen status, and age. J Bone Miner Res 4:469–475. 76. Nordin BEC, Wilkinson R, Marshall DH, Gallagher JC, Williams A, Peacock M 1976 Calcium absorption in the elderly. Calcif Tiss Int 21:442–451. 77. Ireland P, Fordtran JS 1973 Effect of dietary calcium and age on jejunal calcium absorption in humans studied by intestinal perfusion. J Clin Invest 52:2672–2681. 78. Alevizaki CC, Ikkos DG, Singhelakis P 1973 Progressive decrease of true intestinal calcium absorption with age in normal man. J Nucl Med 14:760–762. 79. Avioli LV, McDonald JE, Lee SW 1965 The influence of age on the intestinal absorption of 47-Ca in women and its relation to 47-Ca absorption in postmenopausal osteoporosis. J Clin Invest 44:1960–1967. 80. Bullamore JR, Gallagher JC, Wilkinson R, Nordin BEC, Marshall DH 1970 Effect of age on calcium absorption. Lancet 2:535–537. 81. Barger-Lux MJ, Heaney RP, Recker RR 1989 Time course of calcium absorption in humans: Evidence for a colonic component. Calcif Tissue Int 44:308–311. 82. Chan ELP, Lau E, Shek CC, MacDonald D, Woo J, Leung PC et al. 1992 Age-related changes in bone density, serum parathyroid hormone, calcium absorption and other indices of bone metabolism in Chinese women. Clin Endocrinol 36: 375–381. 83. Evans MA, Triggs EJ, Cheung M, Creasey H, Paull PD 1981 Gastric emptying rate in the elderly: implications for drug therapy. J Am Geriatr Soc 29:201–205. 84. Russell RM 1992 Changes in gastrointestinal function attributed to aging. Am J Clin Nutr 55:1203S–1207S. 85. Recker RR 1985 Calcium absorption and achlorhydria. N Engl J Med 313:70–73. 86. Devine A, Prince RL, Kerr DA, Dick IM, Criddle A, Kent GN et al. 1993 Correlates of intestinal calcium absorption in women 10 years past the menopause. Calcif Tissue Int 52:358–360. 87. Armbrecht HJ, Bolz M, Strong R, Richardson A, Bruns MEH, Christakos S 1989 Expression of calbindin D decreases with age in intestine and kidney. Endocrinol 125:2950–2956. 88. Horst RL, Goff JP, Reinhardt TA 1990 Advancing age results in reduction of intestinal and bone 1,25 dihydroxyvitamin D receptor. Endocrinol 126:1053–1057. 89. Liang CT, Barnes J, Takamato S, Sacktor B 1989 Effect of age on calcium uptake in isolated duodenum cells: role of 1,25-dihydroxyvitamin D3. Endocrinol 124:2830–2836. 90. Takamoto S, Seino Y, Sacktor B, Liang CT 1990 Effect of age on duodenal 1,25 dihydroxyvitamin D3 receptors in Wistar rats. Biochem Biophys Acta 1034:22–28. 91. Lissner L, Bengtsson C, Hansson T 1991 Bone mineral content in relation to lactation history in pre- and postmenopausal women. Calcif Tiss Int 48:319–325. 92. Kinyamu HK, Gallagher JC, Prahl JM, DeLuca HF, Petranick KM, Lanspa SJ 1997 Association between intestinal vitamin D receptor, calcium absorption, and serum 1,25 dihydroxyvitamin D in normal young and elderly women. J Bone Miner Res 12(6):922–928. 93. Barger-Lux MJ, Heaney RP, Hayes J, DeLuca HF, Johnson ML, Gong G 1995 Vitamin D receptor gene polymorphism, bone mass, body size, and vitamin D receptor density. Calcif Tissue Int 57(2):161–162. 94. Cooper GS, Umbach DM 1996 Are vitamin D receptor polymorphisms associated with bone mineral density? A meta-analysis. J Bone Miner Res 11(12):1841–1849.

1117 95. Gallagher JC, Riggs BL, DeLuca HF 1980 Effect of estrogen on calcium absorption and serum vitamin D metabolism in postmenopausal osteoporosis. J Clin Endocrinol Metab 51:1359–1364. 96. Gennari C, Agnusdei D, Nardi P, Civitelli R 1990 Estrogen preserves a normal intestinal responsiveness to 1,25dihydroxyvitamin D3 in oophorectomized women. J Clin Endocrinol Metab 71:1288–1293. 97. Arjmandi BH, Salih MA, Herbert DC, Sims SH, Kalu DN 1993 Evidence for estrogen receptor-linked calcium transport in the intestine. Bone Miner 21:63–74. 98. Bo-Linn GW, Davis GR, Baddrus DJ, Morawski SG, Ana CS, Fordtran JS 1984 An evaluation of the importance of gastric acid secretion in the absorption of dietary calcium. J Clin Invest 73:640–647. 99. Eastell R, Vieira NE, Yergey AL, Wahner HW, Silverstein MN, Kumar R et al. 1992 Pernicious anaemia as a risk factor for osteoporosis. Clin Sci 82:681–685. 100. Knox TA, Kassarjian Z, Dawsonhughes B, Golner BB, Dallal GE, Arora S et al. 1991 Calcium absorption in elderly subjects on high-fiber and low-fiber diets—effect of gastric acidity. Am J Clin Nutr 53:1480–1486. 101. Eastell R, Vieira NE, Yergey AL, Wahner HW, Silverstein MN, Kumar R et al. 1992 Pernicious anemia as a risk factor for osteoporosis. Clin Sci (Colch) 82(6):681–685. 102. Goerss JB, Kim CH, Atkinson EJ, Eastell R, O’Fallon WM, Melton LJ3 1992 Risk of fractures in patients with pernicious anemia. J Bone Miner Res 7(5):573–579. 103. Filipponi P, Gregorio F, Cristallini S, Mannarelli C, Blass A, Scarponi AM et al. 1990 Partial gastrectomy and mineral metabolism: effects on gastrin-calcitonin release. Bone Miner 11(2):199–208. 104. Barger-Lux MJ, Heaney RP, Lanspa SJ, Healy JC, DeLuca HF 1995 An investigation of sources of variation in calcium absorption efficiency. J Clin Endocrinol Metab 80: 406–411. 105. Colodro IH, Brickman AS, Coburn JW, Osborn TW, Norman AW 1978 Effect of 25-hydroxyvitamin D3 on intestinal absorption of calcium in normal man and patients with renal failure. Metabolism 27:745–753. 106. Francis RM, Peacock M, Storer JH, Davies AEJ, Brown WB, Nordin BEC 1983 Calcium malabsorption in the elderly: the effect of treatment with oral 25-hydroxyvitamin D3. Eur J Clin Invest 13:391–396. 107. Berlyne GM, Ben-Ari J, Kushelevsky A, Idelman A, Galinsky D, Hirsch M et al. 1975 The aetiology of senile osteoporosis: secondary hyperparathyroidism due to renal failure. Quart J Med 64:505–521. 108. Delmas P, Stenner D, Wahner HW, Mann KG, Riggs BL 1983 Increase in serum bone gamma-carboxyglutamic acid protein with aging in women: implications for the mechanism of age-related bone loss. J Clin Invest 71:1316–1321. 109. Insogna KL, Lewis AM, Lipinski BA, Byrant C, Baran DT 1981 Effect of age on serum immunoreactive parathyroid hormone and its biological effects. J Clin Endocrinol Metab 53:1072–1075. 110. Kotowicz MA, Melton LJ, Cedel SL, O’Fallon WM, Riggs BL 1990 Effect of age on variables relating to calcium and phosphorus metabolism in women. J Bone Miner Res 5:345–352. 111. Marcus R, Madvig P, Young G 1984 Age-related changes in parathyroid hormone and parathyroid hormone action in normal humans. J Clin Endocrinol Metab 58:223–230. 112. Van Breukelen FJM, Bijvoet OLM, Van Oosterom AT 1979 Inhibition of osteolytic bone lesions by (3-amino-1-hydroxypropylidene)-1,1-bisphosphonate (APD). Lancet I:803–805.

1118 113. Eastell R, Peel NFA, Hannon RA, Blumsohn A, Price A, Colwell A et al. 1993 The effect of age on bone collagen turnover as assessed by pyridinium crosslinks and procollagen I c-terminal peptide. Osteoporosis International 3(Suppl 1):S100–S101. 114. Flicker L, Lichtenstein M, Colman P, Buirski G, Kaymakei B, Hopper JL et al. 1992 The effect of aging on intact PTH and bone density in women. J Am Geriatr Soc 40:1135–1138. 115. Klee GG, Preissner CM, Schryver PG, Taylor RL, Kao PC 1992 Multisite immunochemiluminometric assay for simultaneously measuring whole-molecule and amino-terminal fragments of human parathyrin. Clin Chem 38:628–635. 116. Yendt ER, Cohanim M, Jarzylo S, Jones G, Rosenberg G 1991 Bone mass is related to creatinine clearance in normal elderly women. J Bone Miner Res 6:1043–1050. 117. Young G, Marcus R, Minkoff JR, Kim LY, Sergre GV 1987 Age-related rise in parathyroid hormone in man—the use of intact and midmolecule antisera to distinguish hormone secretion from retention. Bone Mineral Res 2:367–374. 118. Bouillon R, Coopmans W, Degroote DEH, Radoux D, Ellard PH 1990 Immunoradiometric assay of parathyrin with polyclonal and monoclonal region-specific antibodies. Clin Chem 36:271–276. 119. Sokoll LJ, Morrow FD, Quirbach DM, Dawson-Hughes B 1988 Intact parathyrin in postmenopausal women. Clin Chem 34:407–410. 120. Khosla S, Melton LJ, III, Atkinson EJ, O’Fallon WM, Klee GG, Riggs BL 1998 Relationship of serum sex steroid levels and bone turnover markers with bone mineral density in men and women: a key role for bioavailable estrogen. J Clin Endocrinol Metab 83(7):2266–2274. 121. Fatayerji D, Eastell R 1999 Age-related changes in bone turnover in men. J Bone Mineral Res 14(7):1203–1210. 122. Forero MS, Klein RF, Nissenson RA, Nelson K, Heath HI, Arnaud CD et al. 1987 Effect of age on circulating immunoreactive and bioactive parathyroid hormone levels in women. J Bone Miner Res 2:363–366. 123. Eastell R, Simmons PS, Colwell A, Assiri AM, Burritt MF, Russell RG et al. 1992 Nyctohemeral changes in bone turnover assessed by serum bone gla-protein concentration and urinary deoxypyridinoline excretion: effects of growth and aging. Clin Sci (Colch) 83(3):375–382. 124. Harms HM, Prank K, Brosa U, Schlinke E, Neubauer O, Brabant G et al. 1992 Classification of dynamical diseases by new mathematical tools: Application of multidimensional phase space analyses to the pulsatile secretion of parathyroid hormone. Eur J Clin Invest 22:371–377. 125. Kitamura N, Shigeno C, Shiomi K 1990 Episodic fluctuation in serum intact parathyroid hormone concentration in men. J Clin Endocrinol Metab 70:252–263. 126. Samuels MH, Veldhuis J, Cawley C, Urban RJ, Luther M, Bauer R et al. 1993 Pulsatile secretion of parathyroid hormone in normal young subjects—assessment by deconvolution analysis. J Clin Endocrinol Metab 77:399–403. 127. Cantley LK, Russell J, Lettieri D, Sherwood LM 1985 1,25dihydroxyvitamin D3 suppresses parathyroid hormone secretion from bovine parathyroid cells in tissue culture. Endocrinol 117:2114–2119. 128. Gruson M, Demignon J, Del Pinto Montes J, Miravet L 1982 Comparative effects of some hydroxylated vitamin D metabolites on parathyrin secretion by dispersed rat parathyroid cells in vitro. Steroids 40:275–285. 129. Russell J, Lettieri D, Sherwood LM 1986 Suppression by 1,25(OH)2D3 of transcription of the preproparathyroid hormone gene. Endocrinol 119:2864–2866.

RICHARD EASTELL AND B. LAWRENCE RIGGS

130. Silver J, Naveh-Many T, Mayer H, Schmelzer HJ, Popoutzer MM 1986 Regulation by vitamin D metabolites of parathyroid hormone gene transcription in vivo in the rat. J Clin Invest 78:1296–1301. 131. Ledger GA, Burritt MF, Kao PC, O’Fallon WM, Riggs BL, Khosla S 1994 Abnormalities of parathyroid hormone secretion in elderly women that are reversible by short-term therapy with 1,25-dihydroxyvitamin D3. J Clin Endocrinol Metab 79:211–216. 132. Dawson-Hughes B, Harris S, Dallal GE 1991 Serum ionized calcium, as well as phosphorus and parathyroid hormone, is associated with the plasma 1,25-dihydroxyvitamin D3 concentration in normal postmenopausal women. J Bone Miner Res 6:461–468. 133. Sorva A, Risteli J, Risteli L, Valimaki M, Tilvis R 1991 Effects of vitamin-D and calcium on markers of bone metabolism in geriatric patients with low serum 25Hydroxyvitamin-D levels. Calcif Tissue Int 49:S88–S89. 134. Fox J, Mathew MB 1991 Heterogeneous response to PTH in aging rats—evidence for skeletal PTH resistance. Am J Physiol 260:E933–E937. 135. Hegarty V, Woodhouse P, Khaw KT 1994 Seasonal variation in 25-hydroxyvitamin D and parathyroid hormone concentrations in healthy elderly women. Age Aging 23:478–482. 136. Fox J 1991 Regulation of parathyroid-hormone secretion by plasma calcium in aging rats. Am J Physiol 260:E220–E225. 137. Uden P, Halloran B, Daly R, Duh QY, Clark O 1992 Set-point for parathyroid hormone release increases with postmaturational aging in the rat. Endocrinology 131:2251–2256. 138. Wongsurawat N, Armbrecht HJ 1987 Comparison of calcium effect on in vitro calcitonin and parathyroid hormone release by young and aged thyroparathyroid glands. Exp Gerontol 22:263–269. 139. Chapuy MC, Chapuy P, Meunier PJ 1987 Calcium and vitamin D supplements: effects on calcium-metabolism in elderly people. Am J Clin Nutr 46:324–328. 140. Chapuy MC, Arlot ME, Duboeuf F, Brun J, Crouzet B, Arnaud S et al. 1992 Vitamin D3 and calcium to prevent hip fractures in elderly women. New Eng J Med 327:1637–1642. 141. Gallagher JC 1992 Pathophysiology of osteoporosis. Semin Nephrol 12:109–115. 142. Eastell R, Kennedy NSJ, Smith MA, Tothill P, Edwards CRW 1986 Changes in total body calcium following surgery for primary hyperparathyroidism. Bone 7:269–272. 143. Martin P, Bergmann P, Gillet C, Fuss M, Kinnaert P, Corvilain J et al. 1986 Partially reversible osteopenia after surgery for primary hyperparathyroidism. Arch Int Med 146:689–691. 144. Melton LJ, Atkinson EJ, Ofallon WM, Heath H 1992 Risk of age-related fractures in patients with primary hyperparathyroidism. Arch Intern Med 152:2269–2273. 145. Heath HI 1991 Clinical spectrum of primary hyperparathyroidism: evolution with changes in medical practice and technology. J Bone Miner Res 6(Supplement 2):63–70. 146. Parfitt MA, Rao DS, Kleerekoper M 1991 Asymptomatic primary hyperparathyroidism discovered by multichannel biochemical screening: clinical course and considerations bearing on the need for surgical intervention. J Bone Miner Res 6(Supplement 2):97–101. 147. Delmas PD, Stenner D, Wahner H, Mann KG, Riggs BL 1983 Increase in serum bone gamma-carboxyglutamic acid protein with aging in women. J Clin Invest 71:1316–1321. 148. Ledger GA, Burritt MF, Kao PC, O’Fallon WM, Riggs BL, Khosla S 1995 Role of parathyroid hormone in mediating nocturnal and age-related increases in bone resorption. J Clin Endocrinol Metab 80(11):3304–3310.

1119

CHAPTER 67 Vitamin D and Osteoporosis

149. Martinez ME, Del Campo MT, Sanchez-Cabezudo MJ 1994 Relations between calcidiol serum levels and bone mineral density in postmenopausal women with low bone density. Calcif Tissue Int 55:253–256. 150. Lukert B, Higgins J, Stoskopf M 1992 Menopausal bone loss is partially regulated by dietary intake of vitamin D. Calcif Tissue Int 51:173–179. 151. Kleerekoper M, Wilson P, Peterson E, Nelson DA 1992 Acute parathyroid hormone response to intramuscular calcitonin may predict femoral neck bone loss in older postmenopausal women. Journal of Bone and Mineral Research 7[Suppl 1], S312. Ref Type: Abstract. 152. Avioli LV, Heaney RP 1991 Calcium intake and bone health. Calcif Tissue Int 48:221–223. 153. Cumming RC 1990 Calcium intake and bone mass: A quantitative review of the evidence. Calcif Tissue Int 47: 194–201. 154. Dawson-Hughes B, Dallal GE, Drall EA, Sadowski L, Sahyoun N, Tannerbaum S 1990 A controlled trial of the effect of calcium supplementation on bone density in postmenopausal women. New Eng J Med 323(13):878–883. 155. Orwoll EW, Oviatt SK, McClung MR, Deftos LJ, Sexton G 1990 The rate of bone mineral loss in normal men and the effects of calcium and cholecalciferol supplementation. Ann Int Med 112:29–34. 156. Riggs BL, Khosla S, Melton LJ III 1998 A unitary model for involutional osteoporosis: estrogen deficiency causes both type I and type II osteoporosis in postmenopausal women and contributes to bone loss in aging men [see comments]. J Bone Miner Res 13(5):763–773. 157. Riggs BL, Khosla S, Melton LJ, III 2002 Sex steroids and the construction and conservation of the adult skeleton. Endocr Rev 23(3):279–302. 158. McKane WR, Khosla S, Risteli J, Robins SP, Muhs JM, Riggs BL 1997 Role of estrogen deficiency in pathogenesis of secondary hyperparathyroidism and increased bone resorption in elderly women. Proc Assoc Am Physicians 109(2):174–180. 159. Thomas ML, Xu X, Norfleet AM, Watson CS 1993 The presence of functional estrogen receptors in intestinal epithelial cells. Endocrinol 132(1):426–430. 160. Gallagher JC, Fowler SE, Detter JR, Sherman SS 2001 Combination treatment with estrogen and calcitriol in the prevention of age-related bone loss. J Clin Endocrinol Metab 86(8):3618–3628. 161. Van Cromphaut SJ, Rummens K, Stockmans I, Van Herck E, Dijcks FA, Ederveen AG et al. 2003 Intestinal calcium transporter genes are upregulated by estrogens and the reproductive cycle through vitamin D receptor-independent mechanisms. J Bone Miner Res 18(10):1725–1736. 162. McKane WR, Khosla S, Burritt MF, Kao PC, Wilson DM, Ory SJ et al. 1995 Mechanism of renal calcium conservation with estrogen replacement therapy in women in early postmenopause—a clinical research center study. J Clin Endocrinol Metab 80(12):3458–3464. 163. Klein RG, Arnaud SB, Gallagher JC, DeLuca HF, Riggs BL 1977 Intestinal calcium absorption in exogenous hypercortisonism. J Clin Invest 60:253–259. 164. Parfitt AM, Gallagher JC, Heaney RP, Johnston CC, Neer R, Whedon GD 1992 Vitamin D and bone health in the elderly. Am J Clin Nutr 36:1014–1031. 165. Malabanan A, Veronikis IE, Holick MF 1995 Redefining vitamin D insufficiency. Lancet 351(9105):805–806. 166. Jesudason D, Need AG, Horowitz M, O’Loughlin PD, Morris HA, Nordin BE 2002 Relationship between serum

167. 168.

169.

170.

171.

172.

173.

174.

175. 176.

177. 178.

179.

180.

181.

182.

25-hydroxyvitamin D and bone resorption markers in vitamin D insufficiency. Bone 31(5):626–630. Heaney RP, Dowell MS, Hale CA, Bendich A 2003 Calcium absorption varies within the reference range for serum 25hydroxyvitamin D. J Am Coll Nutr 22(2):142–146. Heikinheimo RJ, Inkovaara JA, Harju EJ, Haavisto MV, Kaarela RH, Kataja JM et al. 1992 Annual injection of vitamin D and fractures of aged bones. Calcif Tissue Int 51(2):105–110. Ooms ME, Roos JC, Bezemer PD, van der Vijgh WJF, Bouter LM, Lips P 1995 Prevention of bone loss by vitamin D supplementation in elderly women: A randomized doubleblind trial. J Clin Endocrinol Metab 80:1052–1058. Lips P, Graafmans WC, Ooms ME, Bezemer PD, Bouter LM 1996 Vitamin D supplementation and fracture incidence in elderly persons. A randomised, placebo-controlled clinical trial. Ann Int Med 124:400–406. Trivedi DP, Doll R, Khaw KT 2003 Effect of four monthly oral vitamin D3 (cholecalciferol) supplementation on fractures and mortality in men and women living in the community: randomized double-blind controlled trial. BMJ 326 (7387):469. Dawson-Hughes B, Harris SS, Krall EA, Dallal GE 1997 Effect of calcium and vitamin D supplementation on bone density in men and women 65 years of age or older. N Engl J Med 337(10):670–676. Lips P, van Ginkel FC, Jongen MJM, Rubertus F, van der Vijgh WJF, Netelenbos JC 1987 Determinants of vitamin D status in patients with hip fracture and in elderly control subjects. Am J Clin Nutr 46:1005–1010. Gallagher JC, Jerpbak CM, Jee WSS, Johnson KA, DeLuca HF, Riggs BL 1982 1,25-dihydroxyvitamin D3: short- and long-term effects on bone and calcium metabolism in patients with postmenopausal osteoporosis. Proc Natl Acad Sci USA 79:3325–3329. Aloia JF, Vaswani A, Yeh JK, Ellis K, Yasumura A, Cohn SH 1988 Calcitriol in the treatment of postmenopausal osteoporosis. Am J Med 84:401–408. Gallagher JC, Riggs BL 1990 Action of 1,25-dihydroxyvitamin D3 on calcium balance and bone turnover and its effect on vertebral fracture rate. Metabolism 39(4 Suppl 1): 30–34. Ott SM, Chesnut CH 1989 Calcitriol treatment is not effective in postmenopausal osteoporosis. Ann Int Med 110:267–274. Orimo H, Shiraki M, Hayashi Y, Hoshino T, Onaya T, Miyazaki S et al. 1994 Effects of 1-hydroxyvitamin D3 on lumbar bone mineral density and vertebral fractures in patients with postmenopausal osteoporosis. Calcif Tissue Int 54:370–376. Christiansen C, Christiansen MS, Rodbro P, Hagen C, Transbol I 1981 Effect of 1,25-dihydroxyvitamin D3 in itself or combined with hormone treatment in preventing postmenopausal osteoporosis. Eur J Clin Invest 11:305–309. Need AG, Nordin BEC, Horwitz M, Morris HA 1990 Calcium and calcitriol therapy in osteoporotic postmenopausal women with impaired calcium absorption. Metabolism 39:53–54. Gallagher JC, Riggs BL, Recker RR, Goldgar D 1989 The effect of calcitriol on patients with postmenopausal osteoporosis with special reference to fracture frequency. Proc Soc Exp Biol Med 191:287–292. Tilyard MW, Spears GF, Thomson J, Dovey S 1992 Treatment of postmenopausal osteoporosis with calcitriol or calcium [see comments]. N Engl J Med 326(6):357–362.

1120 183. Riggs BL, Nelson KI 1985 Effect of long term treatment with calcitriol on calcium absorption and mineral metabolism in postmenopausal osteoporosis. J Clin Endocrinol Metab 61:457–461. 184. Shiraki M, Kushida K, Fukunaga M, Kishimoto H, Taga M, Nakamura T et al. 1999 A double-masked multicenter comparative study between alendronate and alfacalcidol in Japanese patients with osteoporosis. The Alendronate Phase III Osteoporosis Treatment Research Group. Osteoporos Int 10(3):183–192. 185. Papadimitropoulos E, Wells G, Shea B, Gillespie W, Weaver B, Zytaruk N et al. 2002 Meta-analyses of therapies for postmenopausal osteoporosis. VIII: Meta-analysis of the efficacy of vitamin D treatment in preventing osteoporosis in postmenopausal women. Endocr Rev 23(4):560–569. 186. Eastell R 1998 Treatment of postmenopausal osteoporosis. N Engl J Med 338(11):736–746. 187. Heaney RP, Weaver CM 2003 Calcium and vitamin D. Endocrinol Metab Clin North Am 32(1):181–viii. 188. Black DM, Cummings SR, Karpf DB, Cauley JA, Thompson DE, Nevitt MC et al. 1996 Randomized trial of effect of alendronate on risk of fracture in women with existing vertebral fractures. Fracture Intervention Trial Research Group. Lancet 348(9041):1535–1541. 189. Reginster J, Minne HW, Sorensen OH, Hooper M, Roux C, Brandi ML et al. 2000 Randomized trial of the effects of risedronate on vertebral fractures in women with established postmenopausal osteoporosis. Vertebral Efficacy with Risedronate Therapy (VERT) Study Group. Osteoporosis International 11(1):83–91. 190. Liberman UA, Weiss SR, Broll J, Minne HW et al. 1995 Effect of oral alendronate on bone mineral density and the incidence of fractures in postmenopausal osteoporosis. New Eng J Med 333(22):1437–1443. 191. Adami S, Frijlink WB, Bijvoet OL, O’Riordan JL, Clemens TL, Papapoulos SE 1982 Regulation of calcium absorption by 1,25,dihydroxy-vitamin D—studies of the effects of a bisphosphonate treatment. Calcif Tissue Int 34(4):317–320. 192. Koster JC, Hackeng WH, Mulder H 1996 Diminished effect of etidronate in vitamin D deficient osteopenic postmenopausal women. European Journal of Clinical Pharmacology 51(2):145–147. 193. Pfeifer M, Begerow B, Minne HW, Abrams C, Nachtigall D, Hansen C 2000 Effects of a short-term vitamin D and calcium supplementation on body sway and secondary hyperparathyroidism in elderly women. J Bone Miner Res 15(6):1113–1118. 194. Bischoff HA, Stahelin HB, Dick W, Akos R, Knecht M, Salis C et al. 2003 Effects of vitamin D and calcium supplementation on falls: a randomized controlled trial. J Bone Miner Res 18(2):343–351. 195. Hodgson SF, Watts NB, Bilezikian JP, Clarke BL, Gray TK, Harris DW et al. 2001 American Association of Clinical

RICHARD EASTELL AND B. LAWRENCE RIGGS

196.

197. 198.

199.

200.

201.

202.

203.

204.

205.

Endocrinologists 2001 Medical Guidelines for Clinical Practice for the Prevention and Management of Postmenopausal Osteoporosis. Endocr Pract 7(4):293–312. Shevde NK, Plum LA, Clagett-Dame M, Yamamoto H, Pike JW, DeLuca HF 2002 A potent analog of 1alpha,25-dihydroxyvitamin D3 selectively induces bone formation. Proc Natl Acad Sci USA 99(21):13487–13491. Baeksgaard L, Andersen KP, Hyldstrup L 1998 Calcium and vitamin D supplementation increases spinal BMD in healthy, postmenopausal women. Osteoporos Int 8(3):255–260. Gallagher JC, Goldgar D 1990 Treatment of postmenopausal osteoporosis with high doses of synthetic calcitriol. A randomized controlled study. Ann Intern Med 113(9): 649–655. Guesens P, Dequecker J 1986 Long-term effect of nandrolone decanoate, 1-alpha hydroxyvitamin D3, or intermittent calcium infusion therapy on bone mineral content, bone remodelling and fracture rate in symptomatic osteoporosis: a double blind controlled study. Bone and Mineral 1:347–357. Orimo H, Shiraki M, Hayashi T, Nakamura T 1987 Reduced occurrence of vertebral crush fractures in senile osteoporosis treated with 1-alpha(OH)-vitamin D3. Bone and Mineral 47:52. Caniggia A, Delling G, Nuti R, Lore F, Vattimo A 1984 Clinical, biochemical and histological results of a double-blind trial with 1,25-dihydroxyvitamin D3, estradiol, and placebo in post-menopausal osteoporosis. Acta Vitaminol Enzymol 6:117–128. Cummings SR, Black DM, Thompson DE, Applegate WB, Barrett-Connor E, Musliner TA et al. 1998 Effect of alendronate on risk of fracture in women with low bone density but without vertebral fractures: results from the Fracture Intervention Trial. JAMA 280(24):2077–2082. Harris ST, Watts NB, Genant HK, McKeever CD, Hangartner T, Keller M et al. 1999 Effects of risedronate treatment on vertebral and nonvertebral fractures in women with postmenopausal osteoporosis: a randomized controlled trial. Vertebral Efficacy with Risedronate Therapy (VERT) Study Group. JAMA 282(14):1344–1352. Chesnut CH, III, Silverman S, Andriano K, Genant H, Gimona A, Harris S et al. 2000 A randomized trial of nasal spray salmon calcitonin in postmenopausal women with established osteoporosis: the prevent recurrence of osteoporotic fractures study. PROOF Study Group. Am J Med 109(4):267–276. Ettinger B, Black DM, Mitlak BH, Knickerbocker RK, Nickelsen T, Genant HK et al. 1999 Reduction of vertebral fracture risk in postmenopausal women with osteoporosis treated with raloxifene: results from a three-year randomized clinical trial. Multiple Outcomes of Raloxifene Evaluation (MORE) Investigators. JAMA 282(7):637–645.

CHAPTER 68

Genetic Vitamin D Receptor Polymorphisms and Risk of Disease ANDRÉ G. UITTERLINDEN, YUE FANG, JOYCE B.J. VAN MEURS, AND HUIBERT A.P. POLS Genetic Laboratory, Department of Internal Medicine, Er asmus Medical Centre, NL-3015 GD Rotterdam, The Netherlands

I. Introduction II. Structure and Polymorphism of the VDR Gene III. Association Analysis in Disease States

IV. Conclusions References

I. Introduction

importance over genome search approaches using linkage analysis [2,3]. The interpretation of polymorphic variations in the VDR gene is severely hindered by the fact that until now only a few polymorphisms in this large gene have been studied, and that most of these are anonymous restriction fragment length polymorphisms (RFLP). One expects them to be linked to truly functional polymorphisms elsewhere in the VDR gene (or in nearby gene(s)), which can then explain the associations observed. Thus, to understand the mechanisms underlying the associations one has to analyze the genomic organization of the VDR locus, to identify which genes are present in the chromosomal area, to categorize all relevant VDR polymorphisms, to determine the haplotypes across the gene, to determine their relationship with the RFLP markers used so far, and finally to perform association analyses with relevant phenotypic endpoints such as disease. Below, we present a more detailed description of the genomic organization of the VDR gene, including discussion on polymorphisms, linkage disequilibrium, and haplotypes. We then describe association studies of VDR polymorphisms in relation to different diseases. Historically speaking, studies of VDR polymorphisms in relation to bone endpoints, including osteoporosis in particular, have received most attention while the analysis of VDR polymorphisms in relation to other diseases, including breast and prostate cancer and immune-related disorders, has reached the literature somewhat later on. This allows studies on associations with bone endpoints to be compared to a certain extent and to illustrate some of the difficulties in interpreting the results. This is much less possible for VDR polymorphism studies in relation to other disease endpoints, although similar interpretation problems exist. Essentially, these interpretation problems

The secosteroid hormone vitamin D, its receptor (VDR), and the metabolizing enzymes involved in the formation of the biologically active form of the hormone, acting together, are major players in the vitamin D endocrine system. This system plays an important role in skeletal metabolism, including intestinal calcium absorption, but has also been shown to play an important role in other metabolic pathways, such as those involved in the immune response and cancer [1]. In the immune system, for example, vitamin D promotes monocyte differentiation and inhibits lymphocyte proliferation and secretion of cytokines, such as IL2, interferon-γ, and IL12. In several different types of cancer cells vitamin D has been shown to have antiproliferative effects. These aspects of the vitamin D endocrine system are extensively discussed elsewhere in this volume. At the same time it is also widely known that large interindividual differences exist. One approach to understand interindividual differences in the vitamin D endocrine system is to study the influence of variations in the DNA sequence of important proteins of this system. For example, deleterious mutations in the VDR gene cause 1,25-dihydroxyvitamin D resistant rickets, a rare monogenetic disease (Chapter 72). More subtle sequence variations (polymorphisms) in the VDR gene occur much more frequently in the population, but they have not been systematically analyzed and their effects on VDR function are poorly understood. Their influence on the vitamin D endocrine system is currently under scrutiny in relation to a number of so-called complex diseases and traits, such as osteoporosis. This so-called candidate gene approach in the genetic dissection of complex traits is currently gaining increased VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

1122

ANDRÉ G. UITTERLINDEN, YUE FANG, JOYCE B.J. VAN MEURS, AND HUIBERT A.P. POLS

find their origin in the lack of knowledge of which polymorphisms are present in the VDR gene area and not knowing what the functional relevance is of these polymorphisms. Therefore, most attention in this review will be focused on these aspects rather than providing an exhaustive review of all the studies that have found their way into the literature on VDR polymorphisms and association with one or other disease endpoint.

II. STRUCTURE AND POLYMORPHISM OF THE VDR GENE A. Genomic Structure of the 12q13 Locus After the cloning of the human VDR cDNA in 1988 by Baker et al. [4], it took almost 10 years before major parts of the genomic structure of the human VDR gene became clear as described by Miyamoto et al. [5]. All of this happened, of course, before the Human Genome Project began to bear fruit in the form of easily accessible databases where genomic sequences could be found. Yet, these databases are still not complete and for particular genes efforts have to be made to determine their genomic structure. The location of the VDR gene on the physical map of chromosome 12 was elucidated first roughly by linkage mapping by Labuda et al. [6] and later on somewhat more refined by Fluorescent in Situ Hybridization (FISH) and radiation hybrid mapping by Taymans et al. [7]. However, these studies defined the position of the VDR gene in very general terms with a resolution of >100 kb and this is insufficient for understanding the role of VDR polymorphisms in disease. The major reason to know the exact gene content and distances of this gene area comes from genetic association studies. Since mostly anonymous polymorphisms have been studied so far, one of the explanations for the associations observed is that the effect is not due to the VDR gene itself, but rather the association is explained by another, nearby gene. It is important to know which genes are also present in the area, how close they are to the VDR gene, and how likely it is that these other genes play a role in explaining the associations. Therefore, we have extended these physical mapping studies of the VDR gene to create a high-resolution physical map of the chromosomal 12q13 region in which the VDR gene is located. For this we applied thorough analysis of all available databases with genomic sequences and pulsed field gel electrophoresis to determine the exact location and order of the VDR gene and its neighboring genes (see Figs. 1–3). Our analysis of the genomic organization of the VDR locus at chromosome 12q13.1 has shown that the VDR gene itself is quite large (just over 100 kb; see Fig. 2;

van Meurs et al., unpublished), has an extensive promoter region capable of generating multiple tissue-specific transcripts (8; see also Chapters 11–12 for more detailed information on this) and lies just downstream from the COL2A1 gene [9,10]. In Fig. 1 a summary is presented of the current mapping information of the VDR-COL2A1 region. On the genetic map [11] both genes are mapped around position 63 cM, between the markers D12S85 and D12S368. The radiation hybrid map [12] confirms the close conjunction of the two genes, since both genes are mapped close to the 210 centiRod position. The STS-based contig map [13] shows that VDR and COL2A1 are located on two separate PAC clones (P1057I20 and P228P16, respectively (see Fig. 1). Figure 2 shows a compilation of all the information from the databases and our own results with respect to the VDR-COL2A1 locus combined with information of the genomic organization of the VDR gene. Recently, most of this information can also be found in the draft of the human genome [14], but until now, a gap in the sequence was still present between COL2A1 and VDR. To elucidate the distance between the two PAC clones, P1057I20 and P228P16 were used as probes in a fiber-FISH experiment (see Fig. 3). The distance between the two PAC clones was estimated to be 50 kb. The distance between the two genes was subsequently estimated to be 30–60 kb, depending on the size of the first intron of VDR between exon 1f and 1e. Completion of the physical map was achieved by sequencing 2.5 kb of sequence upstream of exon 1f, which overlapped with the 3′-end of PAC clone P228P16. Based on this, we calculated the total sequence length between the end of exon 52 of COL2A1 and the beginning of VDR exon 1f to be 30.3 kb. Figure 2 shows the completed physical map of the region surrounding VDR. Besides already identified genes (phosphofructo kinase (PFKM), sentrin/SUMO specific protease (SENP1), and histone deacetylase (HDAC7), also some so far unknown genes were identified (information available at http://compbio.ornl.gov/channel/). From these results, it appears that, in the absence of evidence for the role of the other genes in the area in, e.g., bone metabolism, the VDR gene is the obvious candidate gene to explain the associations found. The next step is to determine the polymorphisms in the relevant areas and determine how they interact with each other in genetic terms and in functional terms.

B. VDR Polymorphisms Several genes in the human genome have been analyzed in detail for the occurrence of genetic variations which occur at a frequency of > 1 % in the population

1123

CHAPTER 68 Genetic Vitamin D Receptor Polymorphisms and Risk of Disease

Chromosome 12

Genetic map (cM)

Radiation hybrid map (cR) 180 cR

50 cM

Contig map

D12S331

Genomic PAC clones

D12S85

D12S87/333

200 cR

D12S331 D12S1301

WI-16433

WI-16433 SSRP129C LAMR1

PP11

60 cM D12S85

D12S85 COL2A1/VDR

13.1

210 cR

13.3

WI-14839 P228P16

VDR/PP11 COL2A1 WI-14839 PFKM

D12S368

VDR

WI-18050 WI-6377

70 cM D12S325

COL2A1 P1057I20 PFKM

D12S355 D12S83

D12S339

220 cR

WI-18050

D12S1702

80 cM D12S368

WI-6377

FIGURE 1 Genomic mapping of the VDR gene to chromosome 12q13.1. VDR and COL2A1 are located closely together and are shown amidst different types of genetic DNA markers on maps of increasing resolution going from left to right. From left to right is displayed: a karyotype with a high resolution banding pattern, the genetic map based on anonymous DNA markers, the more detailed radiation hybrid genetic map, the physical contig map and finally the rough map position of two PAC clones, used for the fiber-FISH experiment shown in Fig. 3.

(polymorphisms), such as the lipoprotein lipase gene [15] and the ACE gene [16]. From these and other analyses we now know that human polymorphisms occur on average at 1 out of every 300 bp. Some areas carry less polymorphism, such as the protein encoding regions

(exons), where the occurrence can be as low as 1 in every 2000 bp of exonic sequence [17–19]. Information on the existence of VDR polymorphisms so far has come from analysis of only limited areas in the gene and by using rather insensitive techniques

Centromere

Telomere PAC 228P16 0

30

PFK 0

VATPase

50

PAC 1057I20

60

SENP1

100

90

120

VDR

COL2A1

150

200

250

150

Kbp

HDAC7

300

350

400 Kbp

FIGURE 2 Genomic structure of the VDR-COL2A1 locus on chromosome 12q13.1 and detailed physical map position of the PAC clones used for the Fiber-FISH experiments shown in Figure 3. The arrow indicates direction of transcription of the VDR gene. PFK = Phosphofructokinase; VATPase = Vacuolar ATP-ase; SENP1 = sentrin/SUMO-specific protease; HDAC7 = Histone Deacetylase 7.

1124

ANDRÉ G. UITTERLINDEN, YUE FANG, JOYCE B.J. VAN MEURS, AND HUIBERT A.P. POLS

COL2A1

P228P16 193 kb VDR

± 50 kb

P1057I20 122 kb

FIGURE 3 Fiber-FISH analysis of the VDR-COL2A1 locus. PAC-clone P1057120 was labeled with Texas red and P228P16 with FITC green and were used as probes in a fiber-FISH (fluorescent in situ hybridization) experiment. In this picture fluorescent signals are detected along a single DNA molecule. The known length of the PAC clone inserts (1057120 = 122 kb and P228P16 = 193 kb) allows to estimate the physical distance between them, i.e., in this case about 50 kb.

to find polymorphisms, such as screening with different restriction enzymes for polymorphic banding patterns in Southern blot hybridization experiments. Examples of this include the ApaI- [20], EcoRV- [21], BsmI- [21], TaqI- [22], and Tru9I- [23] restriction fragment length polymorphisms (RFLPs) discovered at the 3′ end of the VDR gene by this approach in the early 1990s. In Figs. 4 and 5, a number of the currently known VDR polymorphisms are depicted. A special case in this respect is presented by the discovery of the so-called FokI RFLP. Upon comparison of the original Baker sequence of the VDR cDNA [4], two potential translation initiation start sites (ATG) were observed and subsequent sequence comparisons have shown that a T to C polymorphism exists (ATG to ACG) at the first potential start site [24–26]. This polymorphism, also referred to as the Start Codon Polymorphism or SCP, was later defined by using the FokI restriction enzyme in an RFLP test [27]. Thus, two protein variants can exist corresponding to the two available start sites: a long version of the VDR protein

(the T-allele detected as the “f” allele; also referred to as the M1 form, i.e., the methionine at first position) and a protein shortened by three amino acids (the C-allele detected as the “F” allele; also referred to as the M4 form, i.e., the methionine at fourth position). Up until now, this is the only known polymorphism that alters the VDR protein. A more informative approach to find polymorphisms is to simply determine the basepair sequence of the same part of the VDR sequence in a number of different individuals. This was applied by Morrison et al. for the 3.2 kb 3′ untranslated region (UTR) of the gene when searching for additional polymorphisms and possible functional variations to explain an association of BMD differences with the (supposedly nonfunctional) BsmI RFLP [22]. When they sequenced two subjects who were homozygous for the most frequent BsmI-ApaI-TaqI haplotypes (see below), i.e., BAt-BAt and baT-baT, they reported 13 polymorphisms in 3.2 kb, corresponding to the expected 1 in 300 bp to be variant. Among the sequence differences they reported

1125

CHAPTER 68 Genetic Vitamin D Receptor Polymorphisms and Risk of Disease

100 kb

5′ Promoter

Coding exons

3′ Regulatory

1

f

e ad

b

c2

FokI* C/T

Cdx2 G/A

3

7 8 9 3′-UTR

45 6

Ins/delG

C/T

RFLP: BsmI Tru9I bp: A/G G/A

UTR polymorphisms

EcoRV ApaI TaqI* G/A G/T T/C

FIGURE 4 Exon-intron structure of the VDR gene and position of known polymorphisms. * indicates that these polymorphisms are in the coding sequence. See text for further details on the polymorphisms. The 3′UTR polymorphisms are shown in more detail in Figure 5.

and 3 were rare in the eight subjects they analyzed [28]. Surprisingly, only two polymorphisms were found in both the sequence analyses by Durrin et al. [28] and Morrison et al. [22]. When considering the methods used and the number of subjects analyzed, it seems likely that the polymorphisms reported by Durrin et al. do not contain sequence errors and seem

was a polyA-tract with a varying number of A-residues with alleles that vary in length between 12–18 adenosines. Durrin et al. expanded this approach and sequenced the 3′UTR in five subjects homozygous for the baT-haplotype and three subjects homozygous for the BAt-haplotype and in total identified seven polymorphisms (see Fig. 5) of which 4 were common

bp 1285

2085

TGA exon 9

2885

DE-I

C/A

+/−T*

3685

Alu-like

DE-II

+/−AGCCC T/C

C/A

A/T

4485

DE-III

A/G

Poly[A] n = 13–24* Haplotype 1 = baT C

A 18–24 insT

C

Haplotype 2 = BAt A

A

A 13 –17 delT

A

T

FIGURE 5 Structure and position of polymorphisms in the 3′UTR of the VDR gene based on Morrison et al. [22] and Durrin et al. [28]. The bp numbering is according to Baker et al. [4]. TGA indicates the stop codon in exon 9 where the 3′UTR starts. DE I–III refer to the so-called destabilizing elements as identified by Durrin et al. Below, the two most frequent BsmI-ApaI-TaqI haplotypes across this part of the gene are depicted, i.e., haplotype 1 (= baT) and haplotype 2 (= BAt), and the alleles of the 3′UTR polymorphisms to which they are linked based on eight subjects analyzed. The * indicates the polymorphisms that were identified in both studies.

1126

ANDRÉ G. UITTERLINDEN, YUE FANG, JOYCE B.J. VAN MEURS, AND HUIBERT A.P. POLS

to be more accurate. However, the number of subjects analyzed is still limited, and they were highly selected, so it is likely that more as yet uncovered polymorphisms exist in the 3.2 kb UTR (see below). A similar sequence comparison approach was followed by Brown et al. when they analyzed the coding region of the VDR gene in 59 parathyroid tumors to find mutations [29]. Apart from the previously reported TaqI- and FokI-polymorphisms, they reported no polymorphisms in the coding region and found two intronic polymorphisms near exon 2 and 8 (shown in Fig. 4). Another VDR polymorphism that was found through sequence analysis of a targeted area is the so-called Cdx2 polymorphism. Arai and colleagues reported a G to A sequence variation among Japanese women in (what they thought was) the VDR 1a promoter when characterizing this promoter area of the VDR gene [30]. The G to A polymorphism is in a binding site for an intestinal-specific transcription factor, called Cdx2 [31]. Recently, our laboratory has shown the Cdx2 polymorphism to be in the VDR 1e promoter (see Fig. 4), to be present among Caucasians as well as other race groups, and we have developed an allele-specific genotyping assay for it [32]. Currently, it is possible to identify potential polymorphisms through bio-informatic approaches by mining of databases such as the NCBI SNP consortium (dbSNP; http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=snp) or the Celera database (http://cds.celera.com). Indeed, many polymorphisms are usually reported in such databases for any gene of interest, but their validity has to be judged very carefully. For example, for the area which we analyzed in detail for the VDR gene, only 40% of the polymorphisms currently reported in these databases turn out to be real (Fang and van Meurs, unpublished). Although we can expect the databases to improve over time, for polymorphism finding in the near future it might still be necessary to generate reliable sequence information from multiple individuals. We have recently conducted a more comprehensive study of the VDR gene and found 63 polymorphisms across 22 kb of sequenced parts the VDR gene (Fang et al., manuscript in preparation). Apart from the known polymorphisms shown in Figs. 4 and 5, also a number of novel polymorphisms were found in the promoter areas including in and around exons 1f-1c, in and around exons 2–9 and in the 3′UTR. The majority of polymorphisms we found to be in regulatory areas rather than in coding exons. In general, this does not seem to be unexpected since variation in the actual encoded protein itself, such as minor changes in the amino acid sequence, might result in drastic changes in function,

such as affinity for the ligand and binding to DNA. More likely, one can assume that polymorphic variation and, thus, population variance, can exist in areas that regulate the level of expression of this gene, such as the 5′ promoter area and the 3′UTR region. Once we know which polymorphisms are present in a certain candidate gene area, it is important to understand how they relate to each other, i.e., in a genetic sense as well as in a biologically functional sense. Genetically, we need to understand which alleles are linked to each other by studying the linkage disequilibrium between polymorphisms. Functionally, we have to determine how certain combinations of alleles across such a candidate gene might augment or diminish certain effects on gene function.

C. Linkage Disequilibrium and Haplotypes Linkage Disequilibrium (LD) measures describe the association (or co-occurrence) of alleles of adjacent polymorphisms with each other [33]. This means in practice that one polymorphism can predict the other adjacent “linked” one because very little recombination has occurred between them over the time of evolution and population history. High levels of LD in a certain area will coincide with a limited number of “haplotypes” in that area. Haplotypes are blocks of linked alleles of adjacent polymorphisms, whereby the length of such a block coincides with the strength of LD across the area. By analysis of polymorphisms across 51 autosomal areas in the human genome, it has become evident that indeed such haplotype blocks exist [34]. The haplotype block size can vary between 5 kb to >50 kb with an average around 10–20 kb. That means that frequent haplotype alleles can be found that encompass the polymorphic variation in such areas. In practice, this also means that relatively few polymorphisms have to be genotyped to “cover” the variance in a certain area. Therefore, a massive effort is currently under way to determine a haplotype map of the human genome [33,34]. It follows that the LD and haplotype structure of a certain candidate gene, such as the VDR, is important for association analyses to understand how the polymorphic variation in such a gene can contribute to risk of disease and population variance of certain phenotypes of interest. When a certain allele of a polymorphism has been found to be associated with—say—risk of fracture, it follows that this association might be explained by the effect of this particular allele, but also—because of LD and the haplotype structure—by one or more other alleles that happen to be linked to this allele

CHAPTER 68 Genetic Vitamin D Receptor Polymorphisms and Risk of Disease

within the haplotype. Once we know which haplotype carries this risk allele, we can determine by cell biological and molecular biological functional analyses which of the variants on that haplotype allele truly cause this effect. Based on some of the known polymorphisms shown in Fig. 4, several studies have analyzed the extent of Linkage Disequilibrium (LD) across the VDR gene. Since these analyses have used only a small number of polymorphisms, accurate information on LD and haplotypes so far has been very limited. Nevertheless, strong LD at the 3′ end of the gene has been observed for the BsmI-, ApaI-, EcoRV-, and TaqI-RFLPs [21,22]. This information was taken a step further by our laboratory to describe molecular haplotypes for these RFLPs [35]. We identified five haplotype alleles in a large Caucasian population of which haplotype 1 (baT; 48%) and 2 (BAt; 40%) were the most frequent and corresponded to the ones identified by Morrison et al. [22]. In line with this, strong LD was also observed between the BsmI RFLP and the polyA variable number of tandem repeats (VNTR) in the 3′UTR [36]. This latter polymorphism has at least 12 different alleles (in 627 subjects analyzed in [36] across five ethnic groups) but the allele size distribution of the poly-A VNTR essentially follows a bimodal distribution. This pattern is such that this marker can be characterized as bi-allelic and that subjects can be classified as having alleles with short or long polyA stretches. Ingles et al. reported strong linkage between the “b” allele and a long poly-A stretch and the “B” allele and a short poly-A stretch. Combined with the results of Morrison et al. [22], Durrin et al. [28], and our results [35], it follows that the Bsm-Apa-Taq haplotype 1 (baT) is linked to a large number of As in the poly-A VNTR (n = 18−24, Long or L alleles), while haplotype 2 (BAt) is linked to a smaller number of As (n = 13−17, short or S alleles). See Fig. 5. Interestingly, several studies have shown that the FokI polymorphism showed no linkage to any of the other VDR polymorphisms. We have recently determined the LD pattern across the VDR gene using the polymorphisms we have recently discovered (see above; Fang et al., manuscript in preparation). For this we determined the genotype for 41 polymorphisms across the VDR gene in a group of 235 middle-aged Caucasians and then analyzed their association or co-occurrence by calculating D′ values using the PHASE program [37]. By plotting these pair wise measures of LD (so-called D′ values) between each of the polymorphisms using the GOLD program [38], a graphical display of LD across a certain area is obtained (shown in Fig. 6). This shows that at least 4–5 areas of high LD with an average size of

1127

10–20 kb can be recognized in the 100 kb of genomic sequence that encompasses the VDR gene, using these 41 polymorphisms. This is very much in line with what has been found in relation to LD blocks elsewhere in the human genome [33,34]. Importantly, this also predicts that in these regions of high LD, a limited number of frequent haplotypes will encapsulate all the polymorphic variation in that area. For the 3′ region we already know that this is the case because the haplotypes we previously identified in this region using the BsmI-ApaI-TaqI RFLPs [35] correspond closely to the haplotypes predicted when using all available polymorphisms in this area (see Figs. 5 and 6; Fang et al., manuscript in preparation). For the other areas of high LD, we are currently determining the haplotype structure and composition.

D. Ethnic Variation in Polymorphisms VDR polymorphisms have been identified and analyzed so far mostly in Caucasians and to a lesser extent in other ethnic groups. For example, the Cdx2 polymorphism was discovered in Japanese [30] and has only recently been analyzed in Caucasians [32]. For the most widely studied VDR polymorphisms, sometimes substantial differences have been noted between races and/or ethnic groups (see Table I and ref. 39). For example, the f allele of the FokI RFLP, corresponding with the 427 aa long VDR protein variant, occurs with lower frequency in Africans when compared to Caucasians and Asians, while the B allele of the BsmI RFLP has a lower frequency in Asians compared to Caucasians and Africans. Similarly, Ingles et al. showed differences between ethnic groups for the polyA VNTR in the 3′UTR [36]. We have found the frequency of the Cdx2 A allele to vary widely across different ethnic groups with the A allele being lowest in Caucasians, at about 19% population frequency, and highest (74%) in Africans [32]. Why do we see such differences? In general, all polymorphisms start as mutations which occur perhaps due to a DNA damage event, and then can grow in frequency in the population and become true polymorphisms. Thus, allele frequency differences between ethnic groups most likely result from evolutionary processes and population genetic behavior. The same holds true for the LD between the polymorphisms and the haplotype structure. In Table I the frequencies are presented of the Bsm-Apa-Taq haplotypes, which we have recently determined in different ethnic groups (Fang et al., manuscript in preparation). They show complex patterns across ethnic groups, which cannot

1128

ANDRÉ G. UITTERLINDEN, YUE FANG, JOYCE B.J. VAN MEURS, AND HUIBERT A.P. POLS

3′UTR

E2–E9

1c

1e–1b

1f

41 40 39 38 37 36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1

1.00

Taq Apa Bsm Fok

D′

Cdx-2

0.00 Cdx-2 1

2

3 4 5

6

7 8

Fok

Bsm

Apa Taq

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

FIGURE 6

Graphic display of a measure of strength of Linkage Disequilibrium (LD), the D′ value, across the VDR gene in Caucasians (Fang et al., unpublished). Shown are pairwise D′ values between 41 SNPs across the VDR gene, plotted at equal distance from each other. On the left and at the bottom, the position of some known VDR-SNPs is presented. The color bar indicates the strength of D′, with red representing very strong LD and blue representing very weak LD. Analysis is based on genotypes generated for 41 SNPs in 235 Caucasian subjects using the PHASE and GOLD programs [37,38]. (See color plate.)

TABLE I Comparison of VDR Allele Frequencies Across the Three Major Ethnic Groups for the Most Widely Studied Polymorphisms* Ethnic group VDR polymorphism Individual polymorphisms Cdx2 FokI BsmI ApaI TaqI polyA

Minor allele**

Caucasian (%)

Asian (%)

African (%)

A f B A T Short n = 13−17

19 34 42 44 43 41

43 51 7 74 8 12

74 24 36 31 31 29

baT: haplotype 1 BAt: haplotype 2 bAT: haplotype 3

48 39 11

75 7 17

26 16 59

Bsm-Apa-Taq haplotypes

The data are allele frequencies based on the total number of chromosomes. Haplotypes of very low frequency are omitted. *Data are from various sources [20–28,32,35,36] and from Fang et al. (manuscript in preparation). **Minor allele in Caucasians refers to the less common allele; see text for details on definition of alleles.

CHAPTER 68 Genetic Vitamin D Receptor Polymorphisms and Risk of Disease

be derived from the frequency differences of the individual composite polymorphisms. This is probably due to the existence of different sets of alleles linked together, to form haplotypes across the 3′ end of the VDR gene. In other words, the haplotype “1” (or baT) in Caucasians is not defining the same set of linked alleles as the haplotype 1 in Asians or Africans. The set of haplotypes will reflect the out-of-Africa theory describing the origin of human populations across the globe, and perhaps also be due to gene-environment interactions in which certain variants might have had survival and/or reproductive advantages. One can assume that relatively “old” polymorphisms show little variation between different ethnic groups, whereas relatively new polymorphisms might display large differences. In this respect, for example the Cdx2 polymorphism appears relatively new and the FokI RFLP rather old. For the interpretation of association studies, we can assume that individual functional polymorphisms might have the same functional effect in different ethnic groups, because the physiological role of the vitamin D endocrine system will not be vastly different between ethnic groups. However, the individual polymorphisms will explain more or less of the population variance given their difference in frequency between these ethnic groups. This is also the basis for interpretation of ecological studies where allele frequencies across (ethnic) groups are correlated with the different incidence of disease/phenotype between such groups. For nonfunctional or anonymous polymorphisms, the situation is different because here we rely on the LD that is detected by the polymorphism to explain its association with a disease. And as shown for the BsmApa-Taq haplotypes in Table I, the frequencies of these “marker haplotypes” (consisting of nonfunctional polymorphisms which are markers for truly functional alleles elsewhere) are very different between ethnic groups. More importantly, the particular alleles that will be linked in the individual haplotypes is likely to be very different. This means that haplotype 1 (baT) in Caucasians is not describing the same set of linked alleles as in Asians or Africans. Thus, if haplotype 1 is found associated in Caucasians and also in Africans, this can be due to linkage with a totally different functional allele. Alternatively, if an association of such a marker-haplotype is seen in Caucasians but, for example, not in Asians, this can be explained because the LD between the marker and the functional allele is different between these groups. It is therefore essential to elucidate the genetic structure of the linked alleles in such haplotype groups across the complete VDR gene in the different ethnic groups, if we want to compare genetic association results between these groups.

1129

As indicated before, we are currently performing such a study (Fang et al., manuscript in preparation). At this time it is difficult to fully understand the consequences of ethnic allele variation, because of the vastly different environmental factors between such ethnic groups, such as diet and exercise. In addition, there are very different “genetic backgrounds” (i.e., the remainder of the genome and genetic variation therein) between ethnic groups in which these VDR polymorphisms interact with each other and with other genetic variants of other genes. In understanding this it is therefore important to start by defining the haplotype structure of VDR polymorphisms in different ethnic groups. This will reveal which VDR alleles occur together on certain haplotypes and what their frequency is in different ethnic groups. Subsequently, together with the information on differences in environmental factors between ethnic groups, we can start to understand the relevant gene-environment interactions.

E. Functionality of Polymorphisms The interpretation of the association studies using VDR polymorphisms is severely hindered by the fact that most of the polymorphisms used are anonymous, i.e., have an unknown functional effect. The likely explanation for any observed association is then to assume the presence of a truly functional sequence variation elsewhere in the gene which is—to a certain extent—in linkage with an allele of the anonymous polymorphism used. As can be understood from the complex organization of the VDR gene (see Fig. 4), the identification of these functional polymorphisms in the VDR gene is a challenging enterprise. While these results are still eagerly awaited, several investigators have—nevertheless—over the past years analyzed multiple bio-response parameters using the anonymous polymorphisms, including the BsmI RFLP, and Bsm-Apa-Taq haplotypes, and the polyA VNTR in the 3′UTR. These studies include in vitro cell biological and molecular biological studies, but also in vivo measurements of biochemical markers and response to treatments with vitamin D, calcium, and even HRT or bisphosphonates. In Fig. 7 a schematic representation is shown of how functionality of polymorphisms can be tested at different levels, i.e., at the mRNA level, the protein level, cell level, etc. In the optimal situation, the polymorphism is expected to show similar directions of the allelic effects at all these levels and, allow a connection of mechanisms at the molecular level with epidemiological findings at the population level. For the Cdx2 and the FokI polymorphism, this has now been established to a large extent, while for the

1130

ANDRÉ G. UITTERLINDEN, YUE FANG, JOYCE B.J. VAN MEURS, AND HUIBERT A.P. POLS

BsmI-ApaI-TaqI RFLPs and the polyA VNTR polymorphism, this is less certain and controversy still exists. Although one might argue that it is useless to perform any functional study with an anonymous polymorphism, results from such studies might still be informative because of the (strong) LD that is present between the marker polymorphisms and the truly functional polymorphism. An overview of the functional studies for each of these VDR polymorphisms is presented in Table II and these are discussed below. 1. CDX2 POLYMORPHISM

The Cdx2 polymorphism has been well-characterized by the studies of Arai et al. and Yamamoto et al. [30,31]. The G to A polymorphism is located in the Cdx2 binding site in the 1e promoter region, and this site is suggested to play an important role in intestinalspecific transcription of the VDR gene. As the intestine is the site where the calcium absorption predominantly takes place, the Cdx2 site is thought to influence the vitamin D regulation of calcium absorption. The A-allele has been demonstrated to be more “active” than the G-allele by binding the Cdx2 transcription factor more strongly, and by having more transcriptional activity [30]. Thus, the A-allele is thought to cause increased VDR expression in the intestine and, thereby, can increase the transcription of calcium transport proteins such as calbindin-9K and -28K, TRPV5, TRPV6 (Chapters 24 and 25). This could enhance the intestinal absorption of calcium and result in increased BMD. Although this increased BMD has indeed been demonstrated for Japanese women who carry the A-allele [30], this was not found in Caucasian women [32]. Yet, the A-allele of this polymorphism was indeed found to be associated with decreased fracture risk (as would be expected from having an increased BMD) in a large study of Caucasian women, but independently of BMD [32]. Therefore, although the functionality of this polymorphism has indeed been convincingly demonstrated, the exact mechanism whereby the A-allele would confer lower risk for fracture has not been elucidated yet and requires further study. 2. FOKI POLYMORPHISM

From the genetic perspective, it is important to note that the FokI RFLP can be considered an independent marker in the VDR gene since there is no LD with any of the other VDR polymorphisms and the LD area surrounding this polymorphism seems to be very small (<2 kb; see Fig. 6). Therefore, LD with another polymorphism is not a likely explanation for the associations observed with this polymorphism and so functional studies should be focused on the polymorphism itself.

In a study by Arai et al. [27], evidence for the functionality of the FokI polymorphism was obtained. Results from transcriptional activation studies in transfected HeLa cells using a reporter construct under the control of a short portion of the rat 24-hydroxylase gene promoter region (−291–+9) containing a VDRE, suggested the short 424 amino acid VDR protein variant (corresponding with the C-allele or “big F” allele) to be more active than the long 427 aa variant, with a 1.7-fold difference between the two variants. Gross et al. [40] were not able to confirm these results. They also analyzed FokI allelic differences in the transcription activation characteristics of the VDR protein, but now in COS7 cells and using the rat 24 hydroxylase promoter although containing a slightly larger region of the promoter (−1399–+76). In addition, they analyzed VDREs from the human osteocalcin gene and the rat osteopontin gene, but could not see FokI allelic differences in these systems. Yet, the authors noted that it might be difficult to observe the relatively small effects of this polymorphism in these test systems. Moreover, small differences in cell type, promoter area, and gene-specificity of the VDRE might be crucial to see an allelic difference or not. Jurutka et al. [42] demonstrated the 424 aa VDR variant to interact more efficiently with the transcription factor TFIIB, using reporter constructs containing 1100 bp of a rat osteocalcin promoter in COS7-, HeLa, and ROS2/3 cells. The authors concluded the 424 aa short VDR variant to represent a more transcriptionally potent VDR protein. This notion was corroborated by the same authors in an analysis of 20 fibroblast cell lines of different endogenous FokI VDR genotype using a reporter construct containing four copies of the rat osteocalcin gene [43]. Results from our own laboratory [44] seem to confirm the higher activity of the 424 aa short VDR variant while using a different readout of functionality. We tested capacity for growth inhibition by vitamin D in PBMCs of different genotype for the FokI RFLP in cells derived from 72 postmenopausal women. We observed that the PBMCs carrying the “F”-allele (corresponding to the 424 aa short variant) had a lower ED50 and, thus, had a more active VDR variant in inhibiting the (PHA induced) cell growth [44]. In conclusion, we can state that the FokI polymorphism seems to be functional and that the 424 aa VDR variant (F-allele) is somewhat more active than the 427 aa variant (f-allele) in terms of its transactivation capacity as a transcription factor. There might be a genespecific effect in that some promoter areas of vitamin D target genes might be more sensitive to this VDR genotype-dependent difference in activity, while others may not. Together with cell type specific interaction

TABLE IIA Functional Effects of VDR Polymorphisms: In vitro Studies Allelic differences Polymorphism Cdx2; A to G

Study [Reference] Arai et al., 2001 [30,31]

FokI; Arai et al., 1997 [27] C (F; 424 aa; M4) Gross et al., 1998 [40] to T (f; 427 aa; M1) Correa et al., 1999 [41] Jurutka et al., 2000 [42]

Cell type

Durrin et al., 1999 [28] Crofts et al., 1996 [46]

Gross et al., 1998 [47]

Transactivation by VDR

Cell growth inhibition

A >G

–

A>G

–

Hela cell line GMK-Cos7 Human fibroblasts Parathyroid adenoma’s COS7, Hela, ROS 2/3

– No Difference No Difference No Difference

– – – –

424 aa > 427 aa No Difference No Difference –

– –

No Difference

–

424 aa > 427 aa

–

–

–

424 aa > 427 aa

–

–

–

–

424 aa > 427 aa

424 aa > 427 aa

–

–

–

BAt> baT

–

–

–

BAt > baT No Difference baT > BAt No Difference (ba)T > (BA)t

– No Difference baT > BAt No Difference –

– – – – –

– – – – –

No Difference

–

No Difference

–

Human fibroblasts (n = 20) Collin et al., 2000 [44] Human PBMC (n = 72) Ogunkolade et al. [211a] Human PBMC (n = 41) Morrison et al., 1994 [22] GMK Cos-7

Beaumont et al., 1998 [45]

VDR mRNA stability

Caco-2 intestinal cell line

Whitfield et al., 2001 [43]

Bsm, Apa, Taq, UTR (polyA VNTR)

VDR expression

Rat ROS 17/2.8 GMK Cos-7 Human osteoblast Mouse NIH3T3 cells Human fibroblasts (n = 9), PBL (n = 13) Human fibroblasts (n = 9)

Mechanism The G allele has diminished binding of the Cdx2 transcription factor

– The 424 aa VDR variant (M4) interacts more efficiently with transcription factor TFIIB

mRNA stability of BAt is larger than of baT

Continued

TABLE IIA Functional Effects of VDR Polymorphisms: In vitro Studies—Cont’d Allelic differences Polymorphism Bsm, Apa, Taq, UTR (Cont’d)

Study [Reference] Whitfield et al., 2001 [43] Yamagata et al., 1999 [48] Mocharla et al., 1997 [49] Collin et al., 2000 [44] Verbeek et al., 1997 [50]

Carling et al., 1997, 1998 [51,52] Ohtera et al., 2001 [53] Ogunkolade et al. [211a]

GMK, Green Monkey Kidney. PBMC, Peripheral blood mononuclear cells. PBL, Peripheral blood lymphocytes. –, Not analyzed.

Cell type Human fibroblasts (n = 11) Human PBMC (n=24; Japanese) Human PBMC (n =38) Human PBMC (n = 72) Human PBL, leukemia cell line, prostate cell line Human parathyroid adenoma’s (n = 42) Human osteoblasts (n = 18; Japanese) Human PBMC (n = 41)

VDR expression

VDR mRNA stability

Transactivation by VDR

Cell growth inhibition

–

–

–

(BA)t > (ba)T

–

L>S (baT > BAt) –

–

No Difference

–

–

–

–

–

–

No Difference

(ba)T > (BA)t

No Difference

–

–

BAt > baT

–

–

BAt > baT

–

–

baT > BAt

–

baT > BAt

Mechanism

TABLE IIB Polymorphism

Functional Effects of VDR Polymorphisms: In vivo Studies*

Study [Reference]

Marker

FokI

Ogunkolade et al. [211a]

Insulin secretion after OGTT** (n = 143 Bangladeshi)

Bsm, Apa, EcoRV Bsm Bsm Bsm Bsm Apa Taq Bsm Bsm Bsm Apa Taq Bsm Apa Taq Bsm Bsm Bsm Apa Taq

Morrison et al., 1992 [21] Morrison et al., 1994 [22] Kroger et al., 1995 [56] Fleet et al., 1995 [57] Garnero et al., 1995 [58] Krall et al., 1995 [59] Howard et al., 1995 [60] Tokita et al., 1996 [61] Tsai et al., 1996 [62] Mocharla et al., 1997 [49] Rauch et al., 1997 [63] McClure et al., 1997 [64]

Osteocalcin (n = 91) Calcitriol ((n = 117) Osteocalcin, ICTP (n = 23) Osteocalcin, calcitriol (n = 154) Osteocalcin, 25(OH), etc. (n = 189) 1,25(OH)2D (n = 229) Osteocalcin, calcitriol, ICTP (n = 21) Osteocalcin, BAP, 1,25(OH)2D (n = 159; Japanese) Osteocalcin, BAP, PICP, NTX (n = 268; Chinese) Osteocalcin, calcitriol (n = 38) Osteocalcin, AlkPhosp, PICP (n = 50) Osteocalcin, calcitriol, PTH (n = 103)

Bsm Bsm Bsm, FokI Bsm Bsm Apa Taq

Graafmans et al., 1997 [65] Hansen et al., 1998 [66] Willing et al., 1998, 1999 [67,68] Ferrari et al. 1999 [69] Bell et al., 2001 [70]

Bsm Apa Taq

Hitman et al., 1998 [211,250]

Osteocalcin, calcitriol Osteocalcin, 25(OH), BAP (n = 200) Osteocalcin, PTH, 25(OH) (n = 372; n = 261) Osteocalcin, PTH (n = 72 young men) Osteocalcin, PTH, 1,25(OH)2D (n = 39 African men, n = 44 Caucasian men) Insulin secretion after OGTT** (n = 143 Bangladeshi)

Allelic effect: serum concentration F > f for insulin secretion index (in 25(OH)D deficient subjects) BA(t) > ba(T) B(At) > b(aT) b(aT) > B(At) No Difference B(At) > b(aT): NS trend*** No Difference B(At) > b(aT) BAt > baT No Difference No Difference B(At) > b(aT): NS trend BAt > baT: NS trend for PTH , calcitriol baT > BAt: NS trend for osteocalcin No Difference No Difference No Difference B(At) > b(aT) (b)a(T) > (B)A(t) for PICP BAt > baT for insulin secretion index (in 25(OH)D deficient subjects)

*Studies used either one or all of the 3′ polymorphisms, i.e., BsmI, ApaI, EcoRV, and TaqI. To relate these results to what is described in this chapter for sequence variation at the 3′end of the VDR gene, the genotypes and allelic effects are presented as haplotypes 1 (baT) or 2 (BAt). Letters in brackets refer to polymorphisms not actuallly tested in the study but inferred by this author based on VDR haplotype structure in the 3′ area. All studies are in Caucasians unless stated otherwise. ** OGTT = Oral glucose tolerance test. *** NS trend: non–significant trend.

TABLE IIC

Functional Effects of VDR Polymorphisms: In vivo Studies of Response to Treatment* Response

Polymorphism

Study [Reference]

Treatment

Serum markers

BMD

FokI Bsm Bsm Apa Taq Bsm Bsm Bsm Apa Taq Bsm Taq Bsm

Kurabayashi et al., 1999 [73] Howard et al., 1995 [60] Matsuyama et al., 1995 [71] Graafmans et al., 1997 [65] Krall et al., 1995 [59] Deng et al., 1998 [72] Kurabayashi et al., 1999 [73] Ho et al., 1999 [74] Marc et al., 1999 [75]

1 year HRT (n = 82) 7 days calcitriol (n = 21) 12 month 1αOHD3 (n = 115) 2 year calcitriol (n = 81) 2 year calcium (n = 229) 3.5 year HRT (n = 108) 1 year HRT (n = 82) >1 year corticosteroids (n = 263) 1 year etidronate (n = 24)

– b(aT) > B(At) – No Difference – – – – –

No Difference – baT > BAt B(At) > b(aT) B(At) > b(aT) B(aT) > B(At) (ba)T > (BA)t No Difference B(At) > b(aT)

*Studies used either one or all of the 3′ polymorphisms, i.e., BsmI, ApaI, EcoRV, and TaqI. To relate these results to what is described in this chapter for sequence variation at the 3′end of the VDR gene, the genotypes and allelic effects are presented as haplotypes 1 (baT) or 2 (BAt). Letters in brackets refer to polymorphisms not actuallly tested in the study but inferred by this author based on VDR haplotype structure in the 3′ area. All studies are in Caucasians unless stated otherwise. ** OGTT = Oral glucose tolerance test. *** NS trend: non–significant trend.

CHAPTER 68 Genetic Vitamin D Receptor Polymorphisms and Risk of Disease

VDR DNA polymorphism

Organizational level

“Read-out” of functionality

mRNA

– Level, stability, splicing/isoforms

Protein

– Level, stability, isoforms, protein-protein

Cells

– E.g., transcriptional activity (OC-VDRE) – E.g., cell growth inhibition (PBMC)

Humans

– Serum parameters: e.g.,osteocalcin – Intestinal calcium absorption – BMD – Intervention: e.g., vitamin D3

Association with disease

FIGURE 7

Schematic depiction of the different organizational levels in physiology which determines the relation between a DNA polymorphism and the association with an endpoint in an epidemiological study, which are at the same time the levels at which functionality of VDR polymorphisms can be determined. Such functionality can be established using different types of read-outs of test systems indicated on the right of the arrow. OC VDRE = osteocalcin vitamin D responsive element. PBMC = peripheral blood mononuclear cells.

with co-transcription factors, this might result in a cell type-specific and organ-specific expression of the genotype-dependent differences. 3. BSM-APA-TAQ AND 3′UTR POLYMORPHISMS

Most efforts to identify functional sequence variations in the VDR gene have focused on the 3′ regulatory region because this is close to the anonymous markers used mostly up until now in association studies (see Figs. 4 and 5). While the BsmI, ApaI, and TaqI RFLPs are located near the 3′ end of the gene, the LD extends into the 3′ regulatory region containing the UTR. We have already discussed the fact that the 3′UTR of the VDR gene contains many polymorphisms and, thus, through strong LD, these other polymorphisms might explain associations observed with Bsm-Apa and/or Taq RFLPs. The 3′UTR of genes is known to be involved in regulation of gene expression, especially through regulation of mRNA stability. This finding includes the steroid receptor genes which contain extensive 3′UTRs, such as the glucocorticoid receptor α [54]. For the latter receptor, polymorphisms in the 3′UTR have been described in the so-called AUUUA-motifs which influence the mRNA stability [55].

1135

Morrison and colleagues provided evidence of differential luciferase activity for the two 3′UTR-variants that are linked to the two most frequent haplotypes, i.e., “baT” (haplotype 1 according to ref. 35) and “BAt” (haplotype 2). Durrin and colleagues have shown that certain parts of the UTR, so-called destabilizing elements, are involved in determining stability of the VDR-mRNA [28]. However, the UTRs linked to “baT” and “BAt” haplotype were not found by them to differ with respect to mRNA stability [28]. Furthermore, heterologous constructs (human VDR-UTR sequences coupled to a rabbit β-globin gene) and cell types (mouse NIH3T3 cells) were used to test for functionality. Especially, since it is known that 3′-UTRs display celltype specific effects on mRNA stability, this could be important in demonstrating functionality of sequence variations in the UTR. Although it is assumed that mRNA stability differences might underly the allelic differences, alternative explanations should still be considered. Other studies have analyzed differences in expression levels according to the 3′ polymorphisms, and although there is a tendency for the BAt haplotype to display overall somewhat higher levels of mRNA expression than baT, the results have not been consistent (see Table IIA). Recently, Whitfield and colleagues demonstrated functional significance of the translation initiation codon polymorphism (detected as FokI RFLP) and the poly(A) stretch in the 3′UTR [43]. In a series of 20 fibroblast cell lines of different VDR genotype, the relative transcription efficiency was measured of the endogenous VDR-protein. The VDRs differed by genotype at both the FokI RFLP (F and f alleles) and the poly(A) stretch with Long (L) and Short (S) alleles. The endogenous VDR protein is then acting as a transcription factor for a 1,25-dihydroxyvitamin D3responsive reporter gene (containing the rat osteocalcin gene VDRE), which is transfected in the cell. This study provided evidence for so-called high VDR activity (the “FL” genotype) and low VDR activity ( the “fS” genotype). One of the possible explanations mentioned included differences in translational activity (rather than mRNA stability) of the different mRNA-3′UTR variants. However, further research is necessary to prove that assumption. In any case, this study also illustrated the importance of analyzing multiple polymorphisms in the VDR gene in relation to each other (as is illustrated in Fig. 8). At another level, the responses by VDR genotype have been analyzed as differences in serum markers (Table IIB). VDR is then thought to act on vitamin D responsive genes, e.g., through gene-specific VDREs, which results in certain protein/protein fragments being secreted into the circulation. These 17 studies

1136

ANDRÉ G. UITTERLINDEN, YUE FANG, JOYCE B.J. VAN MEURS, AND HUIBERT A.P. POLS

VDR polymorphisms 5′ Promoter: Cdx2-polymorphism: production of mRNA: A > G allele (e.g. in intestinal cell)

3′UTR: Coding region: Bsm-ApaTaq haplotypes: FokI RFLP: F = C ~ 424 aa ~ M(et) 4: “more active” Haplotype1 = baT= polyA L (“more active”) Haplotype 2 = BAt = polyA S isoform f = T ~ 427 aa ~ M(et) 1 coding Promotor 3′UTR

Gene-SNPs: Polymorphisms: Genotypes:

Possible alleles : A or G F or f G/A

Haplotype alleles:

-VDR mRNA -VDR protein

F/f

1/2

1=

G

f

2

2=

A

F

1

ff FFFFFFFF

Haplotype 1 or 2 G/A

F/f

1/2

3=

G

f

1

4=

A

F

2

Target cell: (e.g. intestinal cell)

fffff FFFFF

Functionality:

More of a “more active” VDR (=F)

Phenotype effect:

Calcium absorption increased

More of a “less active” VDR (=f)

Calcium absorption reduced

FIGURE 8 The importance of gene-wide haplotypes in the VDR gene. Three adjacent SNPs in different parts of the VDR gene are shown for two individuals (A and B indicated at the bottom). The subjects A and B have identical genotypes, i.e., they are both heterozygous for all three SNPs. However, they have different haplotypecombinations: 1+2 for subjects A and 3 +4 for subjects B. The promoter area regulates production of mRNA while the 3′UTR is involved in stability/degradation of mRNA and their interaction/combined effects regulates the net availability of the mRNA for translation into the VDR protein. In this case, the example is shown for the Cdx-2 promoter polymorphism which has two alleles A and G, of which the A allele is the more active variant in intestinal cells. For the 3′UTR the two different variants, haplotype 1 and haplotype 2, are presented consisting of haplotypes of the Bsm-Apa-Taq RFLPs. The haplotype 1 is supposedly the more active/less unstable 3′UTR resulting in more VDR protein being produced. The VDR protein can occur in two variants: “little f ” (less active, M1, 427 aa) and “big F” (more active, M4, 423 aa) and both individual A and B are heterozygous for this polymorphism. The result of the particular haplotype combinations is that individual A has less of the “risk” VDR protein, i.e., the little f variant (M1, 427 aa long), than individual B in the target cell. This could not have been predicted by analyzing single SNPs and/or only looking at genotypes of individual SNPs, but is only evident upon analysis of the gene-wide haplotypes.

include several different serum markers thought to be vitamin D specific, such as osteocalcin, PTH, and bone-formation—and resorption markers (see Table IIB for references). In particular osteocalcin has been analyzed because this is a highly vitamin D responsive gene, and it is frequently measured in clinical practice to monitor bone metabolism (for bone formation activity). Although 7 (out of 17) studies were not able to detect genotype-dependent differences in serum levels of osteocalcin (or other bone markers), 8 studies reported individuals with the BAt haplotype to have higher osteocalcin levels than those with the baT haplotype, while 3 (out of 17) studies observed an opposite trend.

A similar pattern can be distinguished when we analyze the studies which determined the in vivo response to treatment by VDR genotype (Table IIC). Studies included the analysis of response to treatment with calcitriol, calcium, corticosteroids, HRT, and etidronate and responses mostly involved measuring changes in BMD. Four out of 9 studies reported the response for BAt to be better than for baT. However, 3 out of 9 showed the opposite effect, while 3 showed no effect. Complicating factors to interpret these studies are, of course, the different endpoints being measured and the different polymorphisms in usually (and understandably) small studies most likely lacking power to demonstrate subtle effects.

CHAPTER 68 Genetic Vitamin D Receptor Polymorphisms and Risk of Disease

Thus, when we examine Tables IIA–C, the picture is still complicated but there seems to be a trend for the BAt-haplotype (linked to Short polyA VNTR alleles in the 3′UTR) to display somewhat better responses than the baT-haplotype (linked to Long polyA VNTR alleles). Together with the in vitro studies there is therefore some argument to state that it might indeed be haplotype 2 (BAt) which—in general—confers a better response than haplotype 1 (baT). It is tempting to speculate that perhaps this is due to a slightly better mRNA stability and half life. This would theoretically result in higher numbers of VDR being present in the target cell giving this target cell a better response to vitamin D. In view of what has been discussed previously, it is not very surprising that these “functional” studies have shown alleles of only few polymorphisms being consistently associated with all of the different parameters of functionality such as displayed in Fig. 7. This holds true for the FokI polymorphism and the Cdx2 polymorphism. Yet, the studies of the 3′ polymorphisms have been hampered. Major caveats applied to these studies are: (a) the use of the anonymous rather than functional polymorphisms to group subjects and cells by genotype, and (b) the use of different types of bioresponses and different cell types and cell culture conditions in which the vitamin D response might not be evident under the conditions of the experiment. Therefore, the identification of truly functional polymorphisms in the 3′UTR and the use of different welldefined cell types will help in clarifying the molecular mechanisms underlying the associations observed especially for the 3′ polymorphisms such as BsmI, ApaI, and TaqI. In addition, it is very important to analyze all known VDR polymorphisms and their inter-relationships in such experiments since they will interact with each other to determine VDR expression and activity (see Fig. 8). This was indicated by the work of Whitfield et al. [43], and this is further illustrated in Fig. 8 where the interaction of promoter-, coding-, and 3′UTR-polymorphisms is highlighted. In a normally active cell, certain promoter polymorphisms will “join forces” with certain 3′UTR polymorphisms in regulating the amount of VDR mRNA being available in a certain target cell. Together, they determine the expression of the known Fokl variants, F and f, which are the (functionally different) VDR proteins that act, e.g., on VDREs to activate certain vitamin D responsive genes to be expressed. In the example, subjects A and B have identical genotypes at three polymorphisms (the Cdx2, the FokI, and the Bsm-Apa-Taq 3′UTR polymorphisms), but only differ in their particular combinations of alleles on chromosomes (i.e., their haplotypes). This ultimately results in less “high activity” VDR proteins (i.e., having the

1137

“F”-allele) being expressed in cells of subject B, which is then expected to display lower responses to vitamin D. In this case, the Cdx2 promoter variant is involved and differences between subjects are then expected at the level of calcium absorption in intestinal cells. However, when only one of the polymorphisms would have been tested in cell biological experiments, this would not have been detected (unless very large genotype groups would have been analyzed to detect the individual subtle differences). Moreover, if only the three individual polymorphisms would have been analyzed and the haplotypes would not have been taken into account, these effects would not have been noticed. Thus, not controlling for the underlying complexities in VDR polymorphisms, i.e., by not analyzing multiple polymorphisms and analyzing their haplotypes, can help to explain contradictory results from in vitro and in vivo functional experiments.

III. ASSOCIATION ANALYSIS IN DISEASE STATES As many chapters in this book illustrate, the vitamin D endocrine system has been shown to be involved in a number of endocrine pathways related to calcium metabolism, immune-modulation, regulation of cell growth and differentiation (of keratinocytes, osteoblasts, cancer cells, T-cells), etc. [1]. Thus, for a pleiotropic “master” gene such as the VDR, one can expect to find associations of this gene with multiple traits and disease phenotypes. Indeed, VDR polymorphisms have been found associated with a number of different phenotypes of which several are supported by multiple independent and large studies reporting similar associations. However, inconsistencies have also been noted between these studies as well. This can be due to a number of factors and they are briefly summarized below. 1. Studies are mostly small and therefore of limited statistical power in order to detect differences of small effect size. This leads to spurious results occurring by chance rather than due to biological effects. It is important to realize that the effects of the VDR gene are expected to be small and, thus, will require large studies to demonstrate them. This holds true for clinical studies designed to find association with disease, as well as for molecular biological studies aiming to demonstrate molecular and cellular effects. In addition, investigators sometimes are too strict in claiming there is NO effect simply because the magic p < 0.05 is not reached and instead, for example, the p value = 0.08. This could mean that in that particular study there is a real

1138

ANDRÉ G. UITTERLINDEN, YUE FANG, JOYCE B.J. VAN MEURS, AND HUIBERT A.P. POLS

effect to be seen, but it simply needs more power to be detected. Thus, ultimately, meta-analyses (combining the results of a large number of individual association studies) will be crucial to estimate the effect size of a particular (set of) polymorphism(s). We will here refer to a number of excellent reviews on the topic of meta-analyses and their application in the field of complex genetics [76–78] for further information. 2. The polymorphisms that have been examined in the association studies are anonymous, i.e., they have no known function. This means that the polymorphism itself cannot explain the association due to some differential functional effect of the alleles but rather, that it merely serves as a marker for a truly functional polymorphism located elsewhere, but relatively close by, in the VDR gene. As a result different alleles can be reported to be the risk allele. A good example of this are the association studies (still) using the individual BsmI-, ApaI-, or TaqIRFLP. The phenomenon underlying this is called Linkage Disequilibrium (LD), and it is described under Section II.C. Analysis of the LD over this 5.5 kb region at the 3′UTR of the VDR gene in different ethnic population groups indicated that the LD differed among populations [28]. A single RFLP, such as the BsmI RFLP which is the most frequently used in association studies of the VDR gene, is therefore not a good marker for the LD with other sequence variations and, thus, the use of the BsmI RFLP might contribute to heterogeneity among association studies. We have now established the LD and haplotype structure of the VDR gene and defined the borders of the genomic area to look for functional polymorphisms once an association has been observed. LD is different between populations and between ethnic groups and, thus, the linkage between the marker allele and the truly functional allele can be different, leading to allelic heterogeneity in the associations. 3. Polymorphisms within a gene are interacting with each other (intragenic interaction) to determine how that gene functions in a given cell and/or subject. This is illustrated for the VDR gene in Fig. 8. The 5′ promoter and 3′UTR polymorphisms determine how much of a given VDR mRNA will be expressed in which cell, while the coding variations determine how well that VDR protein does its job. When only one polymorphism is selected for study, this intragenic interaction is ignored. This can result in unrecognized heterogeneity between experiments/ studies when different 5′ promoter variants and 3′UTR variants are interacting (see Fig. 8). Yet, if studies are of sufficient size, this type of intragenic interaction heterogeneity might be averaged out

across the genotype groups being compared and the effect of an individual polymorphism can still be distinguished. 4. The misconception is that different endpoints (phenotypes) in a complex disease are expected to show similar, if not identical associations, for a given polymorphism. If, for example, an association with fracture risk has been observed, this does not necessarily imply that associations with BMD will be seen as well. Low BMD is only one of several risk factors for fracture. Also, the effect size of a polymorphism on BMD can be so minimal that it will be very difficult to demonstrate an effect on fracture risk: a decrease of 1 standard deviation (SD) in BMD is necessary to see a 2.5-fold increase in fracture risk, while the likely biological effects of individual polymorphisms will be about 0.1–0.3 SD. The associations of VDR polymorphisms with osteoporosis could be explained by a number of processes in which vitamin D plays a role (through the VDR) and which are all important for osteoporosis and fracture risk, including calcium absorption, osteoblast regulation, BMD, and muscle control. Similar arguments can be made for risk factors for other diseases (Kellgren score and joint replacement in osteoarthritis; hypertension and myocardial infarction in cardiovascular disease; insulin secretion and glucose level in diabetes; etc.). 5. The potential confounding effects, which arise from the pleiotropic effects of the vitamin D endocrine system and the VDR as a central player in this, can influence the associations observed. For example, VDR gene variants can influence calcium metabolism through differential absorption in the intestine and, at the same time, influence bone turnover. However, in the same individuals, the occurrence of osteophytosis (as a characteristic of osteoarthritis) can be influenced by the VDR gene variants, together with multiple other factors resulting in a net effect on BMD measured at a certain site, at a certain age, and in a subject with a certain dietary intake of calcium and with a certain disease history. Only with carefully designed studies that control for such factors and are of sufficient power, can such intricate interacting effects be disentangled. Below, a brief review of association studies of VDR polymorphisms in relation to several disease states is presented. The discussion does not seek to be complete, but rather to illustrate the pleiotropic character of the vitamin D endocrine system, and in particular the effect of VDR gene polymorphisms. Several excellent reviews have appeared elsewhere which can be used as further reference on this topic [e.g., 39]. In Table III

TABLE III

Pleiotropic Effects of VDR Alleles*

Cdx2 Number of subjects in association study

Number of studies

Positive

FokI

Negative

Overall allele effect***

Number of studies

Number of subjects in association study

BsmI-ApaI-TaqI-polyA

Positive

Negative

Overall allele effect***

Number of studies

Number of subjects in association study Positive Negative

Overall allele effect***

PHENOTYPE/DISEASE Calcium handling Intestinal calcium absorption Calcium stone formation

–

–

–

–

1

72

–

F>f

9

435

375

baT > BAt

–

–

–

–

–

–

–

–

5

721

283

baT > BAt

Bone Osteocalcin BMD Anthropometry Muscle strength Fracture protection

– 2 – – 1

– 56 – – 2848

– 2848 – – –

– A>G – – A>G

7 14 1 – 2

– 1581 159 – 564

1103 1151 – – 263

– F>f F>f – F>f

15

1136

9 1 10

708 Meta** 1463 264 2073

209 – 1664

BAt > baT BAt > baT baT >BAt baT > BAt baT > BAt

–

–

–

1

160

–

F>f

9

1832

1634

BAt > baT

–

–

–

–

4

237

– – – –

– – – –

4 3 1 –

2013 467 – –

10 2 3 1

3691 Meta** 2594 799 ? 306

2537

– – – –

f>F No effect F= f F>f F>f –

19

– – – –

895 Meta** – 1847 424 –

1759 59 1495 –

BAt > baT No effect BAt > baT BAt > baT BAt > baT BAt > baT

–

–

–

–

1

120

–

F>f

7

1443

370

BAt > baT

–

–

–

–

–

–

–

–

4

528

175

baT > BAt

Osteoarthritis Protection for knee OA/disc degeneration Cancer/proliferative disease Prostate cancer protection Prostate cancer protection** Breast cancer protection Colon cancer protection Melanoma protection Renal cell carcinoma protection Hyperparathyroidism protection Psoriasis protection ? = Inconclusive.

Continued

Table III

Pleiotropic Effects of VDR Alleles*—Cont’d

Cdx2

Immuno-modulation Diabetes mellitus type1 protection Diabetes mellitus type 2 protection Sarcoidosis protection Multiple sclerosis protection Rheumatoid arthritis Crohn’s disease protection Grave’s disease protection Addison’s disease protection Infection protection (TBC, leprosy mycobacterium, hepatitis B, HIV) Periodontal disease protection Cardiovascular disease Hypertension protection Atherosclerosis protection Congestive heart failure protection Myocardial infarction protection Total numbers Overall effect

Number of subjects in association study

Number of studies

Positive

–

FokI

Negative

Overall allele effect***

Number of studies

–

–

–

–

–

–

– –

– –

– – – – –

Number of subjects in association study

BsmI-ApaI-TaqI-polyA

Positive

Negative

Overall allele effect***

6

1344

842

F>f

–

–

–

–

– –

– –

–

–

– – – – –

– – – – –

– – – – –

1 1 1 1 4

–

–

–

–

– –

– –

– –

–

–

– 3

Number of subjects in association study Positive Negative

Overall allele effect***

11

3138

314

BAt > baT

–

7

4171

189

BAt > baT

–

–

3 1

206 172

265 –

? BAt > baT

– – 281 315 611

102 567 – – 464

– – F>F F>f F>f

3 2 1 1 6

120 874 375 315 2961

470 – – – 464

? baT > BAt baT > BAt baT > BAt BAt > baT

–

–

–

–

2

309

–

?

– –

– –

– –

– –

– –

3 3

1523 241

247 3441

? ?

–

–

–

–

–

–

1

–

88

–

–

–

–

–

–

–

–

1

1978

–

BAt > baT

2904

2848

453

8373

7209

149

32,532

16,074

Risk allele:

A>G G

Risk allele:

F>f f

Number of studies

Risk allele: ?

? 9 × baT 14 × BAt

*See text for further details on the associations. ** Meta-analysis has been performed; see text for further details. ***For measurements of quantitative traits (such as osteocalcin level or BMD), e.g., F > f means that the mean value for the measurement was higher in the F-group than in the f-group. For dichotomous traits such as disease we present the extent of protection against the disease (i.e., the inverse of the risk for disease). So, in that case F > f means that the risk for disease was higher in the f group. ? = Inconclusive.

CHAPTER 68 Genetic Vitamin D Receptor Polymorphisms and Risk of Disease

the results of the association studies are summarized for the three most widely studied polymorphisms. So far, the vast majority of association studies have been performed in Caucasians (including European and American Whites, Hispanics, and Arabics), while occasional reports have also appeared on studies in Asian populations (including Japanese, Chinese, and Korean subjects) and in African populations (i.e., African-Americans). Although, as discussed above (3 and 4), ethnic variation is an important source of heterogeneity between association studies, and we have not presented them separately here, because statistical power is usually far too low to draw definite conclusions on ethnic-specific effects of polymorphisms. However, once sufficient power is obtained in this respect, it will be necessary to interpret association studies separately for each ethnic group (see Section 3 below).

A. Osteoporosis A plausible line of reasoning to implicate the vitamin D endocrine system in osteoporosis and fracture suggests that vitamin D regulates absorption of calcium from the intestine, thereby influencing bone mineral density (BMD), which in turn is an important determinant of the risk of osteoporotic fracture. However, apart from BMD, other risk factors for osteoporosis, such as anthropometry (bone size) and muscle control, are also under the influence of the vitamin D endocrine system (see Chapter 102). After the initial observation by Morrison and colleagues on the association between the VDR BsmI RFLP and BMD [22], some of these other risk factors for osteoporosis have been analyzed in association studies, and they are discussed below. However, only a few association analyses have focused on the clinically most relevant endpoint of osteoporosis: fractures. Most attention has gone to BMD while the number of studies on calcium absorption, anthropometry, and muscle strength have been rather limited. This reflects the widespread availability of BMD data, the most widely used diagnostic criterion for osteoporosis. 1. INTESTINAL CALCIUM ABSORPTION

Several studies indicated that the relationship between VDR polymorphisms and BMD is modified by dietary calcium intake. In particular, Krall and colleagues [59] demonstrated that the association between the VDR 3′ polymorphisms and rates of bone loss became more evident at low calcium intake (i.e., mean < 400 mg/day). In addition, Kiel and colleagues [79] demonstrated that the relation between calcium intake and BMD was dependent on VDR genotype (for the

1141

3′ BsmI RFLP). Taken together these studies suggest a gene-environment interaction in the relation between dietary calcium intake and BMD mediated by VDR genotype. Furthermore, these data suggest that the intestinal absorption of calcium is dependent on VDR genotype. Such a relationship was indeed demonstrated in five studies (with a total of 435 subjects) using the laborious isotope tests of intestinal absorption [80–84]. Four of these showed the B(At) haplotype allele to be associated with decreased calcium absorption compared to the b(aT) haplotype allele [80–82]. In line with this, in a study of 84 Thai women (in whom the frequency of the BsmI RFLP B-allele is lower (B < 10%) than in Caucasians; B = 42%), the b-allele was found to have higher urinary calcium excretion, which the authors believed to be due to higher intestinal calcium absorption [83]. One study of 72 children (7–12 yrs) showed the F-allele of the FokI RFLP (demonstrated to be the “more active” VDR; see above) to be associated with higher calcium absorption than the f-allele [84]. Nevertheless, these findings are not universal since four studies (with a total of 375 subjects) could not find any significant association between VDR genotype and calcium absorption [85–88]. Another VDR polymorphism of great interest in this respect is the Cdx2-polymorphism in the 1e/1a promoter area of the VDR gene (see also Section II,E,1 and references 30–32). This polymorphism is a G to A substitution in a binding site for the intestinal-specific transcription factor Cdx2 and has been shown to have differential binding affinity for Cdx2 (A > G; [31]). It is now thought that this polymorphism leads to different levels of expression of the VDR in the intestine, whereby the A-allele has higher expression levels than the G-allele. VDR regulates the expression of several calcium transport-related genes (e.g., calbindin and calcium transporters ECAC1 and ECAC2), which thereby would lead to higher calcium-absorption for the A-allele compared to the G-allele. This sequence of VDR mediated actions in the intestine is thought to ultimately result in increased BMD [31] and lower risk for fracture [32]. While several aspects of this hypothesis have indeed been shown to be the case, the direct influence of this polymorphism on the process of vitamin D–regulated calcium absorption in intestinal cells, has not been demonstrated yet. The above discussion indicates that probably there is a modest contribution of the 3′ polymorphisms (defined as the BsmI-ApaI-TaqI haplotypes) and the FokI polymorphism to explain some of the interindividual variability in calcium absorption. These effects may be more pronounced in subjects with low dietary calcium intake (< 500 mg/day). Adding the likely (but so far undisclosed) effect of the Cdx2 polymorphism on

1142

ANDRÉ G. UITTERLINDEN, YUE FANG, JOYCE B.J. VAN MEURS, AND HUIBERT A.P. POLS

calcium absorption, the mechanisms shown in Fig. 8 act together to determine the contribution of genetic variations in the complete VDR gene to intestinal calcium absorption. 2. CALCIUM STONE FORMATION

Idiopathic calcium nephrolithiasis, or calcium stone formation, is related to the excretion of excessive amounts of calcium in the urine, or idiopathic hypercalciuria. Hypercalciuria is frequently associated with increased intestinal absorption of calcium, which in turn is regulated by the vitamin D endocrine system (Chapter 77). This rationale explains why several studies have analyzed the relation between the VDR gene and calcium stone formation. A study to implicate the VDR gene in calcium stone formation was based on linkage analysis in FrenchCanadian sib-pairs, collected from renal stone clinics on the basis of the presence of at least one calcium stone episode [192]. This study demonstrated suggestive evidence for linkage of this trait to markers in and near the VDR gene, but VDR polymorphisms were not assessed in this study. This linkage analysis approach can implicate a gene in a certain phenotype. Following this study, six association studies have now appeared [193–198]. Four studies with a total of 721 Caucasian subjects showed the baT haplotype allele to be associated with either: (a) increased urinary calcium excretion [193], or (b) more aggressive stone disease [194,196], or (c) more hypocitruria, a risk factor for calcium nephrolithiasis [195]. One study with 214 Chinese subjects could not find evidence for a different distribution of BsmI RFLP alleles between stone formers and controls [197]. The initial study to assess the role of VDR polymorphisms in hypercalciuria did not observe a different distribution of BsmI alleles between 33 hypercalciuric patients and 36 race/age-matched controls [198]. Together with the previous studies on calcium absorption (see Section III,A,1 above) and urinary calcium excretion [83], these data suggest that the VDR baT haplotype allele is associated with increased calcium absorption and higher urinary calcium excretion, resulting in a somewhat increased risk for calcium stone formation. In view of the frequently observed decreased BMD in hypercalciuric patients, this also suggests a possible mechanism explaining the associations of VDR genotype with BMD. 3. BONE MINERAL DENSITY

In the initial studies on VDR polymorphisms, Morrison et al. reported that the BsmI RFLP in the last intron of the VDR gene was related to serum osteocalcin concentration [21]. These authors subsequently found the BsmI RFLP to be associated with differences in BMD

in a twin study and in postmenopausal women [22]. Although the initial observations on the twin study have been withdrawn [89], in the following years dozens of papers were published analyzing the same RFLP in relation to BMD. Some of these confirmed the initial observation, while others could not find an association or found another allele associated. In the largest single cohort study published so far and which analyzed 1782 Dutch elderly men and women, no effect of single RFLPs on BMD was observed, but a small effect was detected employing haplotypes constructed of the three adjacent 3′ RFLPs [35]. A meta-analysis of 29 studies (excluding the Dutch cohort) on the relationship of VDR genotype with BMD [90] concluded that VDR BsmI RFLP genotype is associated with BMD in elderly subjects but with only 1–2% difference between extreme genotypes with the B allele being associated with 1–2% lower BMD. In addition, Gong and colleagues analyzed 75 articles and abstracts on the relation between VDR genotype and BMD [91], and in particular tested whether either of the b-, a-, T-, or F-alleles were associated with increased BMD, BMC, cortical area, thickness and/or diameter, higher response to vitamin D treatment, increased calcium absorption, and/or low bone turnover rate, bone loss, and frequency of fractures. They concluded that increased BMD is indeed associated with the b/a/T/F-allele, especially in females before the menopause thus contributing to peak bone mass. However, the effect on BMD of the 3′VDR polymorphisms by themselves is small and, thus, difficult to consistently detect. When analyzing individual effects of the FokI RFLP on BMD, the “f” allele (corresponding to the 427 aa longer VDR protein with somewhat lower activity) has been found associated with a lower BMD in nine study populations [26,27,69,84,92–96] with a total of 1581 subjects, mostly women. Five other studies, however, with a total of 1151 subjects, have not detected a significant effect on BMD [68,97–100]. Taken together, the results indicate that there probably is a real effect of the FokI RFLP on BMD and that the effect size of this polymorphism is probably similar to what is seen for the 3′ polymorphisms (1–3%). The FokI RFLP is not in linkage disequilibrium with the 3′ polymorphisms (see above) and can therefore not “explain” the association results of the BsmI-, ApaI-, and TaqI polymorphisms. Rather, it should be treated as a separate marker with individual effects. In line with what is shown in Fig. 8, these polymorphisms are likely to have intragenic interaction resulting in combined effects, but this has not been extensively analyzed so far. However, suggestive evidence for such intragenic interaction between the FokI RFLP and the 3′BsmI, ApaI, and TaqI RFLPs has been obtained by Ferrari

CHAPTER 68 Genetic Vitamin D Receptor Polymorphisms and Risk of Disease

and colleagues with respect to effects on BMD [93]. Also see the work of Haussler and colleagues with respect to functional molecular and cellular effects of combined VDR polymorphisms discussed in Chapter 13 and Section II,E,2. The recently described G to A sequence variation in the Cdx-2 binding element just upstream of exon 1e/1a has also been found to be associated with BMD (see Section II,E,1). Arai and colleagues reported that the G-allele had a decreased transactivation capacity and was associated with a 10% decrease in lumbar spine BMD in 58 postmenopausal Japanese women, but not in premenopausal women. More recently, we have extended the association analysis of this polymorphism with osteoporosis to a large group of >2200 postmenopausal Caucasian women [32], but we could not observe Cdx2 genotype-related differences in BMD. Perhaps this is due to the relatively high dietary calcium intake in these subjects and/or differences in vitamin D serum levels between (ethnic) populations. Further research is necessary to understand these differences between studies. 4. ANTHROPOMETRY

Bone size is an important determinant of bone strength, and thus, a risk factor for osteoporosis. Vitamin D is known to regulate the proliferation, differentiation, and maturation of cells responsible for skeletal growth, i.e., the chondrocytes of the epiphysial growth plate, and osteoblasts (see Chapter 33). Hence, genetic variations in the VDR gene might contribute to interindividual differences in bone dimensions, growth and skeletal size characteristics, expressed as difference in height/stature, vertebral area, or femur shaft diameter. Indeed, six studies (with a total of 1463 subjects) have shown associations between the 3′ end polymorphisms and anthropometric measurements [101–106]. While most studies found the b-allele (in the baT haplotype) to be associated with increased height, two studies found the opposite allele associated [107,108], while one study could not find a significant association [109]. So far, only one study has analyzed the FokI RFLP in relation to height in 159 Japanese young adults, with the f-allele being associated with a lower height [110]. Altogether, the data indicate several polymorphisms across the VDR gene to be associated with anthropometric differences. 5. MUSCLE STRENGTH

VDRs are expressed in striated muscles, and decreased serum 25(OH)D3 has been found to correlate with decreased muscle strength (see Chapters 55 and 102). However, only one study has analyzed the association between VDR polymorphisms and muscle

1143

strength [111]. In 264 nonobese (BMI < 30) Caucasian women (>70 yrs), the b(aT) allele was found associated with increased muscle strength measured as quadriceps and grip strength. The association of the VDR Bsm RFLP with muscle strength explained the association with femoral neck BMD also observed in these (nonobese) women. BMD is known to be correlated with muscle strength. Together, these associations suggest that VDR polymorphisms might play a role in determining muscle strength, and, through variation in fall risk, might influence the risk of osteoporotic fracture rather than by influencing BMD. However, studies that include assessments of muscle strength together with incidence of fractures and measurements of BMD are necessary to prove this assumption. This hypothesis is further discussed in Chapter 102. 6. FRACTURES

After the initial findings of Morrison et al. [22,89], it was thought that the VDR genotype-dependent differences in BMD would translate into VDR genotype-dependent differences in fracture risk. This has initiated both ecological as well as epidemiological studies on the relation between VDR genotype and fracture risk. An ecological analysis of 14 studies suggested that higher population frequencies of the (BA)t allele were correlated with higher age-adjusted hip fracture rates [112,113]. Eleven epidemiological studies on the relation of VDR polymorphisms and fracture risk have appeared. Six studies of a total of 2073 Caucasian men and women [114–119] suggested that the VDR genotype was associated with increased fracture risk, including both hip and vertebral fractures. This effect was mostly independent of VDR genotype-related differences in BMD [118], reflecting the relatively small effect of VDR genotype on BMD (see also the results of the meta-analysis on this relationship). Although five of the studies indicated the B-allele (as present in the BAt haplotype allele) was associated with increased fracture risk, one large study of 1004 women [118] showed the baT haplotype to be the risk allele. Four other studies with a total of 1664 Caucasian women did not find a relationship between fracture risk and VDR genotype as defined by the Bsm, Apa, and/or Taq polymorphism [120–123]. The FokI RFLP has been analyzed in two studies of which one reported an increased fracture risk for the f-allele seen in 564 postmenopausal women [124] while the other study did not see an effect in 263 elderly men and women [117]. These results might indicate a small risk associated with the f-allele (corresponding to the 427 aa longer VDR protein with a somewhat lower transactivation activity (see Section II,E,2).

1144

ANDRÉ G. UITTERLINDEN, YUE FANG, JOYCE B.J. VAN MEURS, AND HUIBERT A.P. POLS

The Cdx2-polymorphism has been analyzed in only one study cohort, a large group of 2848 Caucasian women from the Rotterdam Study [32]. In this cohort, we found the A-allele (18% frequency in Caucasians) to be associated with a protective effect on fracture risk, and so the G-variant (82% population frequency in Caucasians) was associated with increased fracture risk in Caucasians. Although this polymorphism was reported to be associated with BMD differences in a small cohort of Japanese postmenopausal women (see above), we have not seen BMD differences related to this polymorphism in our cohort of Caucasian women. The lack of an effect could be due to interaction with the relatively high calcium intake in this group of Caucasian women, and/or differences in circulating vitamin D levels between Caucasian and Asian subjects, but this requires further study. When we view the data on Cdx2, FokI, and the BsmApa-Taq polymorphisms together, these data suggest that there probably is a VDR genotype-dependent increased fracture risk. Again, intragenic interaction mechanisms such as depicted in Fig. 8 might underly this effect. It is also clear that the effect on fracture risk is caused by other factors in addition to the small effects on BMD. In view of what has been discussed above, the genotype-dependent differences in fracture risk could be explained, in part by VDR genotypedependent differences in bone geometry, or muscle strength, or combinations of these. Studies controlling for all of these factors will be necessary to disentangle these effects and determine the combined effects of the VDR polymorphisms. Only one meta-analysis has been done [90] but others, using more subjects and focusing on endpoints other than just BMD, are eagerly awaited.

B. Osteoarthritis The radiographic “Kellgren score” used to diagnose osteoarthritis (OA) is composed of asessment of osteophytosis (bony outgrowths) as well as of cartilage damage by measuring joint space narrowing (JSN). The vitamin D endocrine system has been implicated in the etiology of osteoarthritis/osteoarthrosis by demonstration of expression of the VDR in chondrocytes, which produce the cartilage and are also present in osteophytes [125,126]. In addition, observations of low vitamin D levels in subjects with radiographic osteoarthritis of the knee lended further support to the notion that vitamin D can play a role in the etiology of OA [127] and that perhaps polymorphisms in the VDR play a role in explaining some of the interindividual variability of risk for OA. Five studies with a total of 1832 men and women found indeed an association

between the VDR Bsm-, Apa-, or Taq-polymorphism with aspects of osteoarthritis, assessed either as radiographically determined OA (ROA) of the knee, hip dysplasia, or spinal/intervertebral disc degeneration [128–132]. However, four other studies involving 1634 subjects did not find significant evidence for such an association [133–136]. One study that also analyzed the FokI polymorphism in 160 twin males, found evidence to suggest that the f allele (corresponding to the 427 aa longer VDR protein with a somewhat lower activity) was associated with increased risk of intervertebral disc degeneration [132]. Interestingly, also here allelic heterogeneity and controversy exists for the Bsm-Apa-Taq polymorphisms in that two studies found the “baT” haplotype to be associated with increased risk for OA, while three other studies found the “BAt” haplotype to be the risk allele. A problem with interpreting the studies of VDR polymorphisms and OA is the diversity in OA phenotypes. This ranges from subphenotypes in the Kellgren score (osteophytosis and JSN) to the clinically relevant phenotype requiring joint replacement. Similar to the situation for osteoporosis, we have to realize that finding an association with the Kellgren score does not necessarily mean that the association will translate to risk for joint replacement. Joint replacement is dependent on many other risk factors as well, such as joint pain, and it is well-known that the Kellgren score does not correlate well with joint pain. An interesting observation in this line of research is that the VDR gene happens to be located very near the COL2A1 gene (see Fig. 2), which encodes the major constituent of cartilage, i.e., the collagen type 2 protein. Thus, one possible explanation for the associations found for the VDR gene with OA is that these were explained by LD of the VDR (marker) polymorphisms with (functional) polymorphisms in the COL2A1 gene. By analyzing polymorphisms in both genes in relation to OA in the same population as was used to find the VDR associations, we were able to exclude this explanation [137]. Rather, it seems that both genes are associated with OA, but with different aspects of the disease. The association with VDR polymorphisms seems to be driven by osteophytosis, the bony characteristic of OA, while the association with COL2A1 seems to be driven by JSN, which reflects cartilage loss. Also in view of the function of these gene products, such associations would make sense. It is interesting to note that osteophytosis is also regarded as a loss of growth inhibition for cells present in the subchondral bone [126]. In this respect, the involvement of the VDR in osteophytosis could be very similar to the involvement of the VDR in cancer (see below) and in fact might reflect a similar mechanism

CHAPTER 68 Genetic Vitamin D Receptor Polymorphisms and Risk of Disease

(variation in the control of cell growth by VDR) underlying the association. Taken together, both the 3′ polymorphisms, as well as the FokI polymorphism, seem to be associated with OA. However, similar to what we discussed for the other disease phenotypes, meta-analyses will be necessary to determine the true effect size of the individual polymorphisms, while intragenic interaction of the different polymorphisms will have to be taken into account.

C. Cancer and Hyperproliferative Disease The role of vitamin D in cancer etiology has been widely studied and is the subject of separate chapters in this book (see all chapters in Section IX). It is therefore not surprising that the role of VDR polymorphisms in relation to interindividual variability of risk for different types of cancers has been studied, including prostate cancer, breast cancer, melanoma, colon cancer, lung cancer, renal cell carcinoma, and some hyperplastic syndromes including hyperparathyroidism. 1. PROSTATE CANCER

Because of striking relationships between the incidence of prostate cancer and sunlight/UV exposure (expressed as latitude on earth), suggesting a relationship with vitamin D production, the study of VDR polymorphisms in prostate cancers has received the most attention with now 21 studies having analyzed 7360 subjects (see Table III), including mostly Caucasian men but also African (n = 325) and Japanese (n = 1902) subjects. Nine studies reported an association between the BsmApaTaq polymorphisms and prostate cancer risk, seven of which identified the baT haplotype as the risk allele [138–144] and two studies the BAt haplotype allele [145,146]. Ten studies did not find a significant association between these polymorphisms and prostate cancer [147–156]. Four studies including 1132 subjects analyzed the FokI polymorphisms, two of which found the f-allele (corresponding to the 427 aa longer VDR protein with a somewhat lower activity) to be associated with decreased risk of prostate cancer or decreased aggressiveness [156,157], while one identified the f-allele as a risk allele [158]. One study did not find a significant association with FokI [146]. Several indications for gene-environment interactions have been observed, including augmentation of the association in subjects with low vitamin D levels [140]. A formal meta-analysis of 17 studies, including Caucasian, Asian, and African study populations, could not find significant evidence for any of the four VDR polymorphisms analyzed, i.e., FokI, BsmI, TaqI, and polyA, to be a major determinant of susceptibility to prostate

1145

cancer [158a]. This meta-analysis did not address haplotypes of the 3′ polymorphisms, while some weak and nonsignificant trends corresponding to a slight protective effect of the B(A)t-haplotype allele could be observed. Although such an effect would correspond to what is noted in Table III for the effect of the 3′ VDR polymorphisms on prostate cancer, the robust design of this meta-analysis excludes the polymorphisms analyzed to have large effects, if any, on prostate cancer. 2. BREAST CANCER

Numerous studies have appeared showing similar patterns for breast cancer as for prostate cancer. Eight studies including 2594 subjects, mostly Caucasian women, observed an association of the Bsm-, Apa-, or Taq-polymorphisms with increased risk for breast cancer, defined as either sporadic breast cancers, breast cancer progression, or risk for (bone) metastasis. Again, there is some evidence for allelic heterogeneity with five studies with 1719 women, identifying the baT haplotype allele as the risk allele [159–163], and three studies, with 875 women, identifying the BAt haplotype as the risk allele [164–166]. Two large studies of a total of 1759 Caucasian women [167,168] did not find evidence for significant association, although one study suggested interaction with hormone replacement therapy (HRT), with the baT haplotype increasing breast cancer risk only in HRT users [168]. Somewhat surprisingly, four large studies with a total of 2013 women and derived from the same study populations in which the 3′VDR polymorphisms were analyzed have not found any significant evidence to suggest the FokI polymorphism modifies the risk for breast cancer [160,162–164]. 3. COLON CANCER

Two studies on the relation between Bsm-Apa-Taq polymorphisms and colorectal carcinoma have appeared. One large study involving 799 subjects showed the baT haplotype allele to be associated with increased risk for colon cancer [169], while a smaller study of 59 subjects could not find a significant association [170]. Of particular interest was the observation that the risk was increased in subjects with lower dietary vitamin D intake, suggesting a gene-environment interaction [169]. Four studies on the FokI polymorphism and colorectal cancer have appeared with a total of 2341 subjects [171–174]. Two studies (with a total of 1847 subjects) found an association albeit that allelic heterogeneity was evident: one study including 767 Caucasian subjects found the F-allele to be the risk allele [171], while a study with 1107 Chinese subjects found the f-allele to be the risk allele [172]. A study with 467 Caucasian

1146

ANDRÉ G. UITTERLINDEN, YUE FANG, JOYCE B.J. VAN MEURS, AND HUIBERT A.P. POLS

subjects found no significant association between the FokI polymorphisms and risk for colorectal cancer [173]. 4. MELANOMA AND BASAL CELL CARCINOMA

Three studies have analyzed VDR polymorphisms in relation to malignant melanoma and basal cell carcinoma (BCC). One study observed an increased risk for BCC [175] while two others, involving 1495 subjects, could not find significant evidence to support this [176,177]. Perhaps this is due to the choice of a particular subphenotype, the multiple presentation phenotype, in the latter association study [177]. One study analyzing 424 subjects [176] observed the f-allele of the FokI polymorphism to be associated with increased risk of malignant melanoma in a hospital-based case-control study. In addition, they found evidence for intragenic interaction between the FokI and Bsm-Apa-Taq polymorphisms in that subjects homozygous for the f-t combination of alleles had the highest risk for increased Breslow thickness, a measure to express thickness of the tumors [176].

differentiation of cultured keratinocytes (see Chapters 35 and 101). Four studies have analyzed the association of VDR polymorphisms with the risk of psoriasis [186–188]. Three studies (of Korean and Japanese subjects) observed an association of the Bsm-Apa-Taq polymorphisms, but with evidence for allelic heterogeneity. Two studies, with a total of 344 subjects, found the BAt allele associated with increased risk [186,187], while one study with 184 subjects found the baT allele to be the risk allele [188]. One study of 175 Caucasians could not find evidence for association with the BsmI polymorphism [189]. Four studies have addressed the VDR genotypedependent differences in response to treatment of psoriasis with calcipotriol cream, a potent vitamin D analog used for topical treatment of the disease. Three studies of 166 patients could not find significant evidence for a VDR genotype-dependent difference in response to calcipotriol [189–191], while a recent study analyzing the FokI polymorphism found the f-allele to be increased in frequency among nonresponders [188].

5. RENAL CELL CARCINOMA

A single study reported an association of the baT haplotype allele with increased risk for the aggressive renal cell carcinoma, observed in 306 Japanese subjects [178]. 6. HYPERPARATHYROIDISM

Vitamin D has a strong influence on parathyroid hormone (PTH) secretion and proliferation of parathyroid cells (see Chapters 30, 76, 78, and 103). Several studies have suggested that VDR polymorphisms influence the risk of primary and secondary hyperparathyroidism. Both diseases are characterized by benign parathyroid adenomas or parathyroid hyperplasia, and are accompanied by substantial elevations in the secretion of PTH. While four studies, with a total of 1443 subjects, have observed the baT haplotype allele to be associated with increased risk for primary [179–181], as well as secondary hyperparathyroidism [182], three other studies, with a total of 370 subjects, could not find significant evidence to support this [183–185]. Only one study analyzed the FokI polymorphism [184] and found in 120 women that the f-allele was associated with increased risk for primary hyperparathyroidism. 7. PSORIASIS

Psoriasis is a common skin disease characterized by hyperproliferation of keratinocytes and inflammation. It has also immunological elements and could be discussed under Immune Related Diseases. VDRs have been demonstrated in keratinocytes and vitamin D is known to inhibit proliferation and to induce terminal

D. Immune-related Diseases 1. DIABETES

Vitamin D has important immunomodulatory properties (see Chapters 36, 98, and 99). 1,25(OH)2D3 inhibits T-cell proliferation and can suppress both TNFα as well as IL-1 production. The vitamin D endocrine system can inhibit pancreatic insulin secretion while in the NOD mouse administration of 1,25(OH)2D3 can prevent the development of insulin dependent diabetes mellitus (IDDM), as well as the autoimmune insulinitis. Furthermore, vitamin D levels were found to be reduced in subjects at risk for development of noninsulin dependent diabetes mellitus (NIDDM). Thus, the vitamin D endocrine system can play a role in the etiology of diabetes type 1 (IDDM) as well as type 2 (NIDDM). This explains why the role of VDR polymorphisms was analyzed in both type 1 (T1DM) and type 2 diabetes mellitus (T2DM). Studies involving mostly Caucasians, but also Indian and Japanese subjects have been performed. Both linkage analysis of families and populationbased association studies with VDR polymorphisms have been done. Three Transmission Disequilbrium Testing (TDT) studies with a total of 449 families (corresponding to roughly 1500 subjects) have found initial evidence to suggest that VDR polymorphisms influence risk for T1DM [199–201]. Two studies found the baT haplotype allele associated with increased risk for T1DM [199,200], while one study found the BAt haplotype allele as the risk allele [201]. The involvement

CHAPTER 68 Genetic Vitamin D Receptor Polymorphisms and Risk of Disease

of VDR polymorphisms in T1DM etiology was soon confirmed by association studies [202–207]. Similarly, three studies with a total of 747 subjects found the baT allele as the risk allele for T1DM [202–204], while three other studies encompassing 891 subjects, identified the BAt as the risk allele [205–207]. Two studies could not find evidence for association of the BsmI, ApaI, or TaqI polymorphisms with T1DM [208,209]. In partly the same study populations, the contribution of the FokI polymorphism to the risk for T1DM has been analyzed. Three studies, including one TDT analysis, with a total of 1344 subjects demonstrated significant association between the F-allele and increased risk for T1DM [200,209,210], while three other studies, also including one TDT analysis, found no significant evidence for a contribution of the FokI polymorphism to a risk of T1DM [201,203,204]. The mechanism whereby the association of VDR polymorphisms with T1DM is explained remains unclear. The selective beta-cell destruction seen in T1D is caused by a T-cell mediated autoimmune process. Vitamin D has been shown to inhibit T-cell activation both in vitro and in vivo and inhibits the secretion of IL-1, IL-2, IL-6, IL-12, TNFα, and IFNγ. These cytokines play an important role in the development of T-cells, which are thought to be involved in the pathogenesis of several chronic inflammatory autoimmune diseases. VDR genotype-dependent differences in regulation of cytokine production and/or cell growth inhibition could therefore be the basis for the observed association of VDR polymorphisms with T1DM. However, more molecular and cell biological experiments will be neccesary to prove this. To assess the involvement of VDR polymorphisms in type 2 diabetes mellitus (T2DM) seven studies have been performed on a total of 4360 subjects, mostly in Caucasians although the initial study was done in Bangladeshi Asians. Four studies with a total of 2339 subjects reported the baT allele to be associated with increased risk for T2DM [211–214], sometimes expressed as decreased insulin secretion [211,211a] or an increased risk for obesity in T2DM subjects [213], while two other studies involving 1832 subjects reported the BAt allele to be the risk allele [215,216], sometimes expressed as higher fasting glucose levels [216]. One study (of 189 subjects) could not find a significant association [217]. The mechanism whereby VDR polymorphism associations with T2DM were explained remains unclear. It could involve direct regulation of insulin secretion [211,211a] or perhaps inhibition of growth of pancreatic beta-cells. As explained earlier in this chapter, LD with another gene is a very unlikely explanation.

1147

2. OTHER AUTOIMMUNE DISORDERS

In line with what has been observed for diabetes, other autoimmune disorders also have been reported to be associated with altered vitamin D levels and VDR polymorphisms, including sarcoidosis, multiple sclerosis, Crohn’s disease, Graves disease, and Addison’s disease. (See also Chapter 98.) For sarcoidosis, a systematic granulomatous disorder of unknown etiology, three studies were reported of which one found the BAt haplotype to be associated with increased risk [218], while two other studies could not find an association [219,220]. Multiple sclerosis (MS) is an autoimmune disorder directed against the myelin sheath around axons in the central nervous system. For multiple sclerosis, two studies on the relation with VDR polymorphisms have appeared that describe the same association in the same population, i.e., the baT haplotype was associated with increased risk for multiple sclerosis in 172 Japanese subjects [221,222]. However, a linkage study of 187 Canadian families encompassing 236 sibling pairs analyzed the ApaI and TaqI polymorphisms in the VDR gene and a microsatellite (D12S85) near the VDR gene, by Transmission Disequilibrium Testing (TDT), and could not find evidence for linkage or association of the VDR gene with MS [222a]. In addition, this later study was unable to demonstrate linkage or association with MS for polymorphisms in other genes of the vitamin D endocrine pathway, i.e., the gene encoding the vitamin D–binding protein (4q12) and the gene encoding the 1α-hydroxylase (12q13). While TDT is rather insensitive to detect the more subtle effects of polymorphisms, this study excludes major effects of variations in these vitamin D–related genes to play a role in the etiology of MS. Rheumatoid arthritis (RA) is a systemic inflammatory autoimmune disease of unknown etiology manifested by inflammation of joints and resulting in progressive joint destruction. Three studies analyzed the distribution of VDR polymorphisms between RA cases and controls with a total of 590 subjects, of which 368 were Korean [222 b,c,d]. The study of a Spanish population observed a slightly earlier onset of RA in the B(A)t-haplotype carriers [222b], the two other studies could not find association for the 3′ polymorphisms with RA. Only the German study analyzed the contribution of the FokI polymorphism but could not find no evidence for an association with RA [222d]. For Crohn’s disease [Inflammatory Bowel Disease or IBD], which is characterized by aberrant regulation of mucosal immune response, two studies appeared with a total of 874 subjects which both found the BAt haplotype allele to be associated with increased risk for Crohn’s disease [223,224]. For the FokI polymorphism no association was observed [223].

1148

ANDRÉ G. UITTERLINDEN, YUE FANG, JOYCE B.J. VAN MEURS, AND HUIBERT A.P. POLS

Graves’ disease (GD) is an autoimmune thyroid disease in which TSH receptor autoantibodies cause hyperthyroidism. Two studies of the relation between VDR polymorphisms and GD of a total of 656 Japanese subjects found an association, one with the BAt haplotype allele [225] and one with the F allele of the FokI polymorphism [226]. Finally, Addison’s disease (AD) is an uncommon disorder that results from the T-cell mediated destruction of adrenocortical cells and shares a number of genetic susceptibility markers with T1DM, GD, and Hashimoto’s thyroiditis, including the 12q12 region where the VDR is located. Only one study on the relation between VDR polymorphisms and AD has been conducted and found the fallele of the FokI polymorphism and the BAthaplotype allele of the Bsm-Apa-Taq RFLPs to be associated with increased risk for AD [227]. 3. SUSCEPTIBILITY TO INFECTION

Susceptibility to infectious disease caused by bacteria and/or virus is influenced by environmental factors and by genetic factors in the bacterial/viral agents (determining its ability to infect). Also important are host factors that determine the susceptibility of that host for infection by the infectious agents. In view of what is known about the role of vitamin D in the immune response (see above and Chapters 36 and 98) and what has been described above on the VDR role in autoimmune disorders, it is not very surprising that the role of VDR polymorphisms has also been examined in relation to susceptibility to a number of different infectious diseases. These include tuberculosis [228,229,232,233], leprosy [230], Mycobacterium malmoense [231], as well as the hepatitis B and HIV virus [232,234]. The results are summarized in Table III and similar to what has been seen before, the results vary in the particular allele being associated (baT or BAt and F or f as the risk allele) and some studies not showing an association. In line with this, associations also have been reported between VDR Bsm-Apa-Taq alleles and periodontal disease as well as the intra-oral inflammation seen after bacterial infection [235,236].

E. Cardiovascular Disease A number of studies have analyzed serum vitamin D levels and VDR genotypes in relation to heart disease. The observed relationships suggest an involvement of the vitamin D endocrine system in the etiology of several aspects of heart disease (see also Chapters 54 and 56). Two studies of a total of 1523 subjects observed BsmApa-Taq haplotypes to be associated with increased risk of high blood pressure or hypertension, albeit with

evidence for allelic heterogeneity with one study finding the baT as the risk allele [237], while another study found the BAt allele as the risk allele [238]. One study (in 247 Japanese) could not find evidence for an association of BsmI polymorphism with hypertension, although the B-allele/BAt haplotype was associated with decreased serum calcium levels [239]. In view of the involvement of vitamin D in calcium handling and the occurrence of calcification processes (including expression of many bone genes such as osteocalcin) at atherosclerotic plaques, the involvement of VDR genotype in atherosclerosis has been investigated. Two studies with a total of 241 subjects found association, again with evidence for allelic heterogeneity. One study observed the baT as the risk allele [240], while another study of 200 subjects found the BAt haplotype as the risk allele [241]. However, one large study of 3441 subjects could not find an association [242]. Congestive heart failure (CHF) is a cardiac dysfunction syndrome characterized by a reduced left ventricular ejection fraction, with muscle weakness and fatigue as major symptoms. An alteration in intracellular calcium handling seems to play a role in the impaired contractility of the myocardium. Several studies have found vitamin D deficiency in CHF patients. VDR is expressed in cardiac muscle and is known to activate calcium channels in cardiac muscle cells. However, one study of 88 CHF patients and controls could not find evidence for an association of VDR genotype with CHF [243]. Finally, we have observed in a large sample of 1978 men and women from the Rotterdam Study increased risk for myocardial infarction (MI) for carriers of the baT haplotype. Although the overall effect was modest (only 20% more risk), the relationship was modified by dietary calcium intake with those with high dietary calcium intake (>1200 mg/day) having three- to fourfold increased risk for MI [244]. The relationship seemed to be independent of known risk factors such as atherosclerosis and hypertension. In view of the above associations, and the clear involvement of the vitamin D endocrine system in cardiac function, it seems probable that VDR genotype influences risk of heart disease, but the mechanism whereby this occurs remains unclear and, thus, more studies are necessary.

IV. CONCLUSIONS When we survey the data in Tables II and III, it seems that the different VDR polymorphisms studied until now (Cdx2, FokI, and the 3′ Bsm-Apa-Taq RFLPs) are associated with differences in biological responses and risk of disease. The findings are, however, far from

CHAPTER 68 Genetic Vitamin D Receptor Polymorphisms and Risk of Disease

universal as is clear from the fact that the studies finding positive associations are mostly balanced by those that are not. More formal meta-analyses will be necessary to deal with this and to assess the magnitude of effects, but they are likely to be small. The first metaanalyses, on BMD, indeed demonstrated this to be the case. This is not surprising because most of the time we are looking at complex traits and diseases in which multiple genes play a role, of which the VDR gene is likely to be just one of many. A major limitation of the association studies using VDR polymorphisms in relation to disease endpoints has been the limited number of polymorphisms that have been analyzed and, thus, the lack of control of intragenic interactions between polymorphisms. In addition, the lack of statistical power of most studies to detect the expected subtle effects and misconceptions about how such small biological effects could be translated to the risk of disease has led to a number of controversies in the field. Interactions among genes and interactions with environmental factors also play a role in the action of this pleiotropic steroid hormone receptor transcription factor. For example, dietary Ca-intake is known to differ substantially between countries and populations, while circulating serum vitamin D levels, which are determined by several metabolizing enzymes, also differ between populations. Consequently, gene-gene and gene-environment interactions can then differ between different populations and will seriously affect interpretation of association results. Only very large studies that can control for these factors are able to overcome this difficulty. It is likely that still more polymorphisms, including functional ones, will be discovered in the complex promoter region of the VDR gene, and large population studies will be necessary to document the LD over the region and to evaluate the associations with relevant disease endpoints. In particular, studies should be undertaken in which the VDR gene is systematically scanned for sequence variations, such as has been done for other candidate genes. Haplotype analyses should be used to identify groups of SNPs linked together and, thus, simplify the association analyses and increase understanding of the associations observed. Until clearly functional polymorphisms are identified in the VDR gene, interpretation of meta-analyses to evaluate consistency of associations and estimate effective size of a polymorphism will be cumbersome. This is especially true for the 3′ polymorphisms where we still do not know which polymorphism(s) in linkage with these variations are driving certain associations. For the Cdx2 promoter polymorphism evidence seems convincing to conclude this to be a functional polymorphism while more association studies are required to

1149

demonstrate its effects, particularly in intestinal calcium metabolism. Also for the FokI polymorphism, evidence seems convincing to conclude this is a functional polymorphism. Taken together, it is clear that multiple polymorphic variations exist in the VDR gene that could each have different types of consequences (as is illustrated in Fig. 8). Thus, 5′ promoter variations can affect mRNA expression patterns and VDR levels while 3′ UTR sequence variations can affect the mRNA stability and/or protein translation efficiency. In combination, these genotypic differences are likely to affect the VDR protein concentration and/or function, depending on the cell type, developmental stage, and activation status. In summary, one can conclude that VDR gene variants seem to influence a number of biological endpoints. Yet, the associations have different magnitudes with BMD probably being one of the weaker effects. In different study populations, different alleles of the anonymous RFLPs can be found associated with the same endpoint. This probably reflects that linkage disequilibrium, between the anonymous marker alleles and the causative alleles in the VDR gene, is likely to be different between populations. Finding functional sequence variants that matter, establishing the phase of alleles across the entire VDR gene, and defining haplotype patterns is therefore required to put the associations observed with VDR gene polymorphisms in biological perspective. Meta-analyses of association studies are the way forward to determine the effect size of the small but true effects on disease risk.

References 1. Haussler MR, Whitfield GK, Haussler CA, Hsieh J-C, Thompson PD, Selznick SH, Encinas Dominguez C, Jurutka PW 1998 The nuclear vitamin D receptor: biological and molecular regulatory properties revealed. J Bone Miner Res 13:325–349. 2. Lander ES, Schork NJ 1994 Genetic dissection of complex traits. Science 265:2037–2048. 3. Risch N, Merikangas K 1996 The future of genetic studies of complex human diseases. Science 273:1516–1517. 4. Baker AR, McDonell DP, Hughes M, Crisp TM, Mangelsdorf DJ, Haussler MR, Pike JW, Shine J, O’Malley BW 1988 Cloning and expression of full-length cDNA encoding human vitamin D receptor. Proc Natl Acad Sci USA 85: 3294–3298. 5. Miyamoto K-I, Kesterson RA, Yamamoto H, Taketani Y, Nishiwaki E, Tatsumi S, Inoue Y, Morita K, Takeda E, Pike JW 1997 Structural organization of the human vitamin D receptor chromosomal gene and its promotor. Mol Endocrinol 11:1165–1179. 6. Labuda M, Fujiwara TM, Ross MV, Morgan K, Garcia-Heras J, Ledbetter D, Hughes MR, Glorieux FH 1992 Two hereditary

1150

7.

8.

9.

10. 11. 12. 13. 14. 15.

16. 17.

18.

19.

20. 21.

22. 23.

ANDRÉ G. UITTERLINDEN, YUE FANG, JOYCE B.J. VAN MEURS, AND HUIBERT A.P. POLS

defects related to vitamin D metabolism map to the same region of human chromsome 12q13-14. J Bone Miner Res 7:1447–1453. Taymans SE, Pack S, Pak E, Orban Z, Barsony J, Zhuang Z, Stratakis CA The human vitamin D receptor gene (VDR) is localized to region 12cen-q12 by Fluorescent In Situ Hybridization and Radiation Hybrid Mapping: Genetic and Physical VDR map. J Bone Miner Res 14:1163–1166. Crofts LA, Hancock MS, Morrison NA, Eisman JA 1998 Multiple promotors direct the tissue-specific expression of novel N-terminal variant human vitamin D receptor gene transcripts. Proc Natl Acad Sci USA 95:10529–10534. Takahashi E, Hori T, Sutherland GR 1990 Mapping of the human type II collagen gene (COL2A1) proximal to fra(12) (q13.1) by nonisotopic in situ hybridization. Cytogenet Cell Genet 54:84–85. Huang MC, Seyer JM, Thompson JP, Spinella DG, Cheah KS, Kang AH 1991 Genomic organization of the human procollagen alpha 1(II) collagen gene. Eur J Biochem 195:593–600. http://www.marshfieldclinic.org/research/genetics/default http://www.ncbi.nlm.nih.gov/cgi-bin/Entrez/maps.cgi?org = hum&chr =12 http://sequence.aecom.yu.edu/chr12 http://genome.ucsc.edu Nickerson DA, Taylor SL, Weiss KM, Clark AG, Hutchinson RG, Stengard J, Salomaa V, Vartiainen E, Boerwinkle E, Sing CF 1998 DNA sequence diversity in a 9.7-kb region of the human lipoprotein lipase gene. Nature Genet 19:233–240. Rieder MJ, Taylor SL, Clark AG, Nickerson DA 1999 Sequence variation in the human angiotensin converting enzyme. Nature Genet 22:59–62. Cambien F, Poirier O, Nicaud V, Herrman S, Mallet C, Ricard S, Behague I, Hallet V, Blanc H, Loukaci V, Thillet J, Evans A, Ruidavets J-B, Arveiler D, Luc G, Tiret L 1999 Sequence diversity in 36 candidate genes for cardiovascular disorders. Am J Hum Genet 65:183–191. Cargill M, Altshuler D, Ireland J, Sklar P, Ardlie K, Patil N, Lane CR, Lim EP, Kalyanaraman N, Nemesh J, Ziaugra L, Friedland L, Rolfe A, Warrington J, Lipshutz R, Daley GQ, Lander ES 1999 Characterization of single-nucleotide polymorphisms in coding regions of human genes. Nature Genet 22:231–238. Wang DG , Fan J-B, Siao C-J, Berno A, Young P, Sapolsky R, Ghandour G, Perkins N, Winchester E, Spencer J, Kruglyak L, Stein L, Hsie L, Topaloglou T, Hubbell E, Robinson E, Mittmann M, Morris MS, Shen N, Kilburn D, Rioux J, Nusbaum C, Rozen S, Hudson TJ, Lipshutz R, Chee M, Lander ES 1998 Large-scale identification, mapping, and genotyping of single-nucleotide polymorphisms in the human genome. Science 280:1077–1082. Faraco JH, Morrison NA, Baker A, Shine J, Frossard PM 1989 ApaI dimorphism at the human vitamin D receptor gene locus. Nucl Acids Res 17:2450. Morrison NA, Yeoman R, Kelly PJ, Eisman JA 1992 Contribution of trans-acting factor alleles to normal physiological variability: vitamin D receptor gene polymorphisms and circulating osteocalcin. Proc Natl Acad Sci USA 89:6665–6669. Morrison NA, Qi JJC, Tokita A, Kelly PJ, Crofts L, Nguyen TV, Sambrook PN, Eisman JA 1994 Prediction of bone density from vitamin D receptor alleles. Nature 367:284–287. Ye W-Z, Reis AF, Velho G 2000 Identification of a novel Tru9I polymorphism in the human vitamin D receptor gene. J Hum Genet 45:56–57.

24. Saijo T, Ito M, Takeda E, Huq AH, Naito E, Yokota I, Sone T, Pike JW, Kuroda Y 1991 A unique mutation in the vitamin D receptor gene in three Japanese patients with vitamin D– dependent rickets type II: Utility of single strand conformation polymorphism analysis for heterozygous carrier detection. Am J Hum Genet 49:668–673. 25. Sturzenbecker L, Scardaville B, Kratzeisen C, Katz M, Abarzua P, McLane J 1997 Isolation and analysis of cDNA encoding a naturally-occurring truncated form of the human vitamin D receptor. In: Bouillon R, Norman AW, Thomasset M (eds) Vitamin D: a pluripotent steroid hormone: structural studies, molecular endocrinology, and clinical applications. W De Gruyter, New York, NY, USA, pp. 253–257. 26. Gross C, Eccleshall TR, Malloy PJ, Villa ML, Marcus R, Feldman D 1996 The presence of a polymorphism at the translation initiation site of the vitamin D receptor gene is associated with low bone mineral density in postmenopausal Mexican-American women. J Bone Miner Res 11:1850–1855. 27. Arai H, Miyamoto K-I, Taketani Y, Yamamoto H, Iemori Y, Morita K, Tonai T, Nishisho T, Mori S, Takeda E 1997 A vitamin D receptor polymorphism in the translation initiation codon: effect on protein activity and relation to bone mineral density in Japanese women. J Bone Miner Res 12:915–921. 28. Durrin LK, Haile RW, Ingles SA, Coetzee GA 1999 Vitamin D receptor 3′-untranslated region polymorphisms: lack of effect on mRNA stability. Biochim Biophys Acta 1453:311–320. 29. Brown SB, Brierley TT, Palanisamy N, Salusky IB, Goodman W, Brandi ML, Drueke TB, Sarfati E, Urena P, Chaganti RSK, Pike JW, Arnold A 2000 Vitamin D receptor as a candidate tumor-suppressor gene in severe hyperparathyroidism of uremia. J Clin Endocrinol Metab 85:868–872. 30. Arai H, Miyamoto KI, Yoshida M, Yamamoto H, Taketani Y, Morita K, Kubota M, Yoshida S, Ikeda M, Watabe F, Kanemasa Y, Takeda E 2001 The polymorphism in the caudal-related homeodomain protein Cdx-2 binding element in the human vitamin D receptor gene. J Bone Miner Res 16:1256–1264. 31. Yamamoto H, Miyamoto K-I, Bailing L, Taketani Y, Kitano M, Inoue Y, Morita K, Pike JW, Takeda E 1999 The caudalrelated homeodomain protein Cdx-2 regulates vitamin D receptor gene expression in the small intestine. J Bone Miner Res 14:240–247. 32. Fang Y, van Meurs JB, Bergink AP, Hofman A, van Duijn CM, van Leeuwen JP, Pols HA, Uitterlinden AG 2003 Cdx-2 polymorphism in the promoter region of the human vitamin D receptor gene determines susceptibility to fracture in the elderly. J Bone Miner Res 18:1632–1641. 33. Wall JD, Pritchard JK 2003 Haplotype blocks and linkage disequilibrium in the human genome. Nature Reviews Genetics 4:587–597. 34. Gabriel SB, Schaffner SF, Nguyen H, Moore JM, Roy J, Blumenstiel B, Higgins J, DeFelice M, Lochner A, Faggart M, Liu-Codero SN, Rotimi C, Adeyemo A, Cooper R, Ward R, Lander ES, Daly MJ, Altshuler D 2002 The structure of haplotype blocks in the human genome. Science 296:2225–2229. 35. Uitterlinden AG, Pols HAP, Burger H, Huang Q, van Daele PLA, van Duijn CM, Hofman A, Birkenhäger JC, van Leeuwen JPTM 1996 A large-scale population-based study of the association of vitamin D receptor gene polymorphisms with bone mineral density. J Bone Miner Res 11:1242–1248. 36. Ingles SA, Haile RW, Henderson BE, Kolonel LN, Nakaichi G, Shi C-Y, Yu MC, Ross RK, Coetzee GA 1997 Strength of linkage disequilibrium between two vitamin D receptor markers in five ethnic groups: implications for association studies. Cancer Epidemiol Biomark Prevent 6:93–98.

CHAPTER 68 Genetic Vitamin D Receptor Polymorphisms and Risk of Disease

37. http://archimedes.well.ox.ac.uk/pise/PHASE 38. http://www.sph.umich.edu/csg/abecasis/GOLD 39. Zmuda JM, Cauley JA, Ferrell RE 2000 Molecular epidemiology of vitamin D receptor gene variants. Epidemiol Rev 22:203–217. 40. Gross C, Krishnan AV, Malloy PJ, Ross Eccleshall T, Zhao X-Y, Feldman D 1998 The vitamin D receptor gene start codon polymorphism: a functional analysis of FokI variants. J Bone Miner Res 13:1691–1699. 41. Correa P, Rastad J, Schwarz P, Westin G, Kindmark A, Lundgren E, Akerstrom G, Carling T 1999 The vitamin D receptor (VDR) start codon polymorphism in primary hyperparathyroidism and parathyroid VDR messenger ribonucleic acid levels. J Clin Endocrinol Metab 84:1690–1694. 42. Jurutka PW, Remus LS, Whitfield GK, Thompson PD, Hsieh JC, Zitzer H, Tavakkoli P, Galligan MA, Dang HTL, Haussler CA, Haussler MR 2000 The polymorphic N terminus in human vitamin D receptor isoforms influences transcriptional activity by modulating interaction with transcription factor IIB. Mol Endocrinol 14:401–420. 43. Whitfield GK, Remus LS, Jurutka PW, Zitzer H, Oza AK, Dang HTL, Haussler CA, Galligan MA, Thatcher ML, Dominguez CE, Haussler MR 2001 Functionally relevant polymorphisms in the human nucelar vitamin D receptor gene. Mol Cell Endocrinol 177:145–159. 44. Collin EM, Weel AEAM, Uitterlinden AG, Buurman CJ, Birkenhager JC, Pols HAP, van Leeuwen JPTM 2000 Consequences of vitamin D receptor gene polymorphisms for growth inhibition of cultured human peripheral blood mononuclear cells by 1,25-dihydroxyvitamin D3. Clin Endocrinol 52:211–216. 45. Beaumont M, Bennett AJ, White DA, Hosking DJ 1998 Allelic differences in the 3′ untranslated region of the vitamin D receptor gene affect mRNA levels in bone cells. Osteop Int 8:37 (abstract P081). 46. Crofts LA, Morrison NA, Dudman N, Eisman JA 1996 Differential expression of VDR gene alleles. J Bone Miner Res 11:S208 (abstract S473). 47. Gross C, Musiol IM, Eccleshall TR, Malloy PJ, Feldman D 1998 Vitamin D receptor gene polymorphisms: analysis of ligand binding and hormone responsiveness in cultured skin fibroblasts. Biochem Biophys Res Comm 242:467–473. 48. Yamagata M, Nakajima S, Tokita A, Sakai N, Yanagihara I, Yabuta K, Ozono K 1999 Analysis of stable levels of messenger RNA derived from different polymorphic alleles in the vitamin D receptor gene. J Bone Miner Metab 17: 164–170. 49. Mocharla H, Butch AW, Pappas AA, Flick JT, Weinstein RS, de Togni P, Jilka RL, Roberson PK, Parfitt AM, Manolagas SC 1997 Quantification of vitamin D receptor mRNA by competitive polymerase chain reaction in PBMC: lack of correspondence with common allelic variants. J Bone Miner Res 12:726–733. 50. Verbeek W, Gombart AF, Shiohara M, Campbell M, Koeffler HP 1997 Vitamin D receptor: no evidence for allelespecific mRNA stability in cells which are heterozygous for the TaqI restriction enzyme polymorphism. Biochem Biophys Res Comm 238:77–80. 51. Carling T, Ridefelt P, Hellman P, Rastad J, Akerstrom G 1997 Vitamin D receptor polymorphisms correlate to parathyroid cell function in primary hyperparathyroidism. J Clin Endocrinol Metab 82:1772–1775. 52. Carling T, Rastad J, Akerstrom G, Westin G 1998 Vitamin D receptor (VDR) and parathyroid hormone messenger ribonucleic acid levels correspond to polymorphic VDR allelels

53. 54. 55.

56. 57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

1151

in human parathyroid tumors. J Clin Endocrinol Metab 83: 2255–2259. Ohtera K, Ishii S, Matsuyama T 2001 Influence of the vitamin D receptor alleles on human osteoblast-like cells. J Bone Joint Surg 83-B:134–138. Decker CJ, Parker R 1995 Diversity of cytoplasmic functions for the 3′ untranslated region of eukaryotic transcripts. Curr Opin Cell Biol 7:386–392. Schaaf MJM, Cidlowski JA 2002 AUUUA motifs in the 3′UTR of human glucocorticoid receptor alpha and beta mRNA destabilize mRNA and decrease receptor protein expression. Steroids 67:627–636. Kroger H, Mahonen A, Ryhanen S, Turunen A-M, Alhava E, Maenpaa P 1995 Vitamin D receptor genotypes and bone mineral density. Lancet 345:1238. Fleet JC, Harris SS, Wood RJ, Dawson-Hughes B 1995 The BsmI vitamin D receptor restriction fragment length polymorphism (BB) predicts low bone density in premenopausal black and white women. J Bone Miner Res 10: 985–990. Garnero P, Borel O, Sornay-Rendu E, Delmas PD 1995 Vitamin D receptor gene polymorphisms do not predict bone turnover and bone mass in healthy premenopausal women. J Bone Miner Res 10:1283–1288. Krall EA, Parry P, Lichter JB, Dawson-Hughes B 1995 Vitamin D receptor alleles and rates of bone loss: influence of years since menopause and calcium intake. J Bone Miner Res 10:978–984. Howard G, Nguyen T, Morrison N, Watanabe T, Sambrook P, Eisman J, Kelly PJ 1995 Genetic influences on bone density: physiological correlates of vitamin D receptor gene alleles in premenopausal women. J Clin Endocrinol Metab 80: 2800–2805. Tokita A, Matsumoto H, Morrison NA, Tawa T, Miura Y, Fukamauchi K, Mitsuhashi N, Irimoto M, Yamamori S, Miura M, Watanabe T, Kuwabara Y, Yabuta K, Eisman JA 1996 Vitamin D receptor allelels, bone mineral density, and turnover in premenopausal Japanese women. J Bone Miner Res 11:1003–1009. Tsai KS, Hsu SHJ, Cheng WC, Chen CK, Chieng PU, Pan WH 1996 Bone mineral density and bone markers in relation to vitamin D receptor gene polymorphisms in Chinese men and women. Bone 19:513–518. Rauch F, Radermacher A, Danz A, Schiedermaier U, Golucke A, Michalk D, Schonau E 1997 Vitamin D receptor genotypes and changes of bone density in physically active German women with high calcium intake. Exp Clin Endocrinol Diabetes 105:103–108. McClure L, Eccleshall TR, Gross C, Villa ML, Lin N, Ramaswamy V, Kohlmeier L, Kelsey JL, Marcus R, Feldman D 1997 Vitamin D receptor polymorphisms, bone mineral density, and bone metabolism in postmenopausal MexicanAmerican women. J Bone Miner Res 12:234–240. Graafmans WC, Lips P, Ooms ME, van Leeuwen JPTM, Pols HAP, Uitterlinden AG 1997 The effect of vitamin D supplementation on the bone mineral density of the femoral neck is associated with vitamin D receptor genotype. J Bone Miner Res 12:1241–1245. Hansen TS, Abrahamsen B, Henriksen FL, Hermann AP, Jensen LB, Horder M, Gram J 1998 Vitamin D receptor alleles do not predict bone mineral density or bone loss in Danish perimenopausal women. Bone 22:571–575. Willing M, Sowers M, Aron D, Clark MK, Burns T, Bunten C, Crutchfield M, D’Ágostino D, Jannausch M 1998 Bone mineral density and its change in white women: estrogen and

1152

68.

69.

70.

71. 72.

73.

74. 75. 76. 77. 78. 79.

80.

81.

82.

83.

ANDRÉ G. UITTERLINDEN, YUE FANG, JOYCE B.J. VAN MEURS, AND HUIBERT A.P. POLS

vitamin D receptor genotypes and their interaction. J Bone Miner Res 13:695–705. Sowers M, Willing M, Burns T, Deschenes S, Hollis B, Crutchfield M, Jannausch M 1999 Genetic markers, bone mineral density, and serum osteocalcin levels. J Bone Miner Res 14:1411–1419. Ferrari S, Manen D, Bonjour J-P, Slosman D, Rizzoli R 1999 Bone mineral mass and calcium and phosphate metabolism in young men: relationships with vitamin D receptor alleleic polymorphisms. J Clin Endocrinol Metab 84:2043–2048. Bell NH, Morrison NA, Nguyen TV, Eisman JA, Hollis BW 2001 ApaI polymorphisms of the vitamin D receptor predict bone density of the lumbar spine and not racial difference in bone density in young men. J Lab Clin Med 137:133–140. Matsuyama T, Ishii S, Tokita A, Yabuta K, Yamamori S, Morrison NA, Eisman JA 1995 Vitamin D receptor genotypes and bone mineral density. Lancet 345:1238–1239. Deng HW, Li J, Li J-L, Johnson M, Gong G, Davis KM, Recker RR 1998 Change of bone mass in postmenopausal Caucasian women with and without hormone replacement therapy is associated with vitamin D receptor and estrogen receptor genotypes. Hum Genet 103:576–585. Kurabayashi T, Tomita M, Matsushita H, Yahata T, Honda A, Takakuwa K, Tanaka K 1999 Association of vitamin D and estrogen receptor gene polymorphism with the effect of hormone replacement therapy on bone mineral density in Japanese women. Am J Obstet Gynecol 180:1115–1120. Ho YV, Briganti EM, Duan Y, Buchanan R, Hall S, Seeman E 1999 Polymorphisms of the vitamin D receptor gene and corticoid-related osteoporosis. Osteop Int 9:134–138. Marc J, Prezelj J, Komel R, Kocijancic A 1999 VDR genotype and response to etidronate therapy in late postmenopausal women. Osteop Int 10:303–306. Ioannidis JP, Lau J 1999 Pooling research results: benefits and limitations of meta-analysis. Jt Comm J Qual Improv 25:462–469. Ioannidis JP, Trikalinos TA, Ntzani EE, ContopoulosIoannidis DG 2001 Replication validity of genetic association studies. Nat Genet 29:306–309. Ioannidis JP 2003 Genetic associations: false or true? Trends Mol Med 9:135–138. Kiel DP, Myers RH, Cupples LA, Kong XF, Zhu XH, Ordovas J, Schaeffer EJ, Felson DT, Rush D, Wilson PW, Eisman JA, Holick MF 1997 The BsmI vitamin D receptor restriction fragment length polymorphism (bb) influences the effect of calcium intake on bone mineral density. J Bone Miner Res 12:1049–1057. Dawson-Hughes B, Harris SS, Finneran S 1995 Calcium absorption on high and low calcium intakes in relation to vitamin D receptor genotype. J Clin Endocrinol Metab 80:3657–3661. Gennari L, Becherini L, Masi L, Gonelli S, Cepollaro C, Martini S, Mansani R, Brandi ML 1997 Vitamin D receptor genotypes and intestinal calcium absorption in postmenopausal women. Calc Tissue Int 61:460–463. Wishart JM, Horowitz M, Need AG, Scopacasa F, Morris HA, Clifton PM, Nordin BEC 1997 Relations between calcium intake, calcitriol, polymorphisms of the vitamin D receptor gene, and calcium absorption in premenopausal women. Am J Clin Nutr 65:798–802. Ongphiphadhanakul B, Rajatanavin R, Chanprasertyothin S, Chailurkit L, Piaseu N, Teerarungsikul K, Sirisriro R, Komindr S, Puavilai G 1997 Vitamin D receptor gene polymorphism is associated with urinary calcium excretion but

84.

85.

86.

87.

88.

89. 90. 91.

92.

93.

94.

95.

96.

97.

not with bone mineral density in postmenopausal women. J Endocrinol Invest 20:592–596. Ames SK, Ellis KJ, Gunn SK, Copeland KC, Abrams SA 1999 Vitamin D receptor gene FokI polymorphism predicts calcium absorption and bone mineral density in children. J Bone Miner Res 14:740–746. Francis RM, Harrington F, Turner E, Papiha SS, Datta HK 1997 Vitamin D receptor gene polymorphism in men and its effect on bone density and calcium absorption. Clin Endocrinol 46:83–86. Kinyamu HK, Gallagher JC, Knezetic JA, DeLuca HF, Prahl JM, Lanspra SJ 1997 Effect of vitamin D receptor genotypes on calcium absorption, duodenal vitamin D receptor concentration, and serum 1,25 dihydroxyvitamin D levels in normal women. Calc Tissue Int 60:491–495. Wolf RL, Cauley JA, Baker CE, Ferrell RE, Charron M, Cagiulla AW, Salamone LM, Heany RP, Kuller LH 2000 Factors associated with calcium absorption efficiency in preand perimenopausal women. Am J Nutr 72:466–471. Laaksonen M, Karkkainen M, Outila T, Vanninen T, Ray C, Lamberg-Allardt C 2002 Vitamin D receptor gene BsmIpolymorphism in Finnish premenopausal and postmenopausal women: its association with bone mineral density, markers of bone turnover, and intestinal calcium absorption, with adjustment for lifestyle factors. J Bone Miner Metab 20:383–390. Morrison NA, Qi JJC, Tokita A, Kelly PJ, Crofts L, Nguyen TV, Sambrook PN, Eisman JA 1997 Prediction of bone density from vitamin D receptor alleles (correction). Nature 387:106. Cooper GS, Umbach DM 1996 Are vitamin D receptor polymorphisms associated with bone mineral density? A metaanalysis. J Bone Miner Res 11:1841–1849. Gong G, Stern HS, Cheng S-C, Fong N, Mordeson J, Deng H-W, Recker RR 1999 The association of bone mineral density with vitamin D receptor gene polymorphisms. Osteop Int 9:55–64. Harris SS, Eccleshall TR, Gross C, Dawson-Hughes B, Feldman D 1997 The vitamin D receptor start codon polymorphism (FokI) and bone mineral density in premenopausal American black and white women. J Bone Miner Res 12:1043–1048. Ferrari S, Rizzoli R, Manen D, Slosman D, Bonjour JP 1998 Vitamin D receptor gene start codon polymorphism (FokI) and bone mineral density: interaction with age, dietary calcium, and 3′ end region polymorphisms. J Bone Miner Res 13:925–930. Gennari L, Becherini L, Mansani R, Masi L, Falchetti A, Morelli A, Colli E, Gonelli S, Cepollaro C, Brandi ML 1999 FokI polymorphism at translation initiation site of the vitamin D receptor gene predicts bone mineral density and vertebral fractures in postmenopausal Italian women. J Bone Miner Res 14:1379–1386. Lucotte G, Mercier G, Burckel A 1999 The vitamin D receptor FokI start codon polymorphism and bone mineral density in osteoporotic postmenopausal French women. Clin Genet 56:221–224. Kanan RM, Varanasi SS, Francis RM, Parker L, Datta HK 2000 Vitamin D receptor gene start codon polymorphism (FokI) and bone mineral density in healthy male subjects. Clin Endocrinol 53:93–98. Eccleshall TR, Garnero P, Gross C, Delmas PD, Feldman D 1998 Lack of correlation between start codon polymorphism of the vitamin D receptor gene and bone mineral density in premenopausal French women: the OFELY study. J Bone Miner Res 13:31–35.

CHAPTER 68 Genetic Vitamin D Receptor Polymorphisms and Risk of Disease

98. Zmuda JM, Cauley JA, Danielson ME, Theobald TM, Ferell RE 1999 Vitamin D receptor translation initiation codon polymorphism and markers of osteoporotic risk in older AfricanAmerican women. Osteop Int 9:214–219. 99. Cheng WC, Tsai KS 1999 The vitamin D receptor start codon polymorphism (FokI) and bone mineral density in premenopausal women in Taiwan. Osteop Int 9:545–549. 100. Wynne F, Drummond F, O’Sullivan K, Daly M, Shanahan F, Molloy MG, Quane KA 2002 Investigation of the genetic influence of the OPG, VDR (FokI), and COLIA1 Sp1 polymorphisms on BMD in the Irish population. Calc Tissue Int 71:26–35. 101. Heany RP, Barger-Lux MJ, Davies KM, Ryan RA, Johnson ML, Gong G 1997 Bone dimensional change with age: interactions of genetic, hormonal, and body size variables. Osteop Int 7:426–431. 102. Tao C, Yu T, Garnett S, Briody J, Knight J, Woodhead H, Cowell CT 1998 Vitamin D receptor alleles predict growth and bone density in girls. Arch Dis Child 79:488–494. 103. Suarez F, Rossignol C, Garabedian M 1998 Interactive effects of estradiol and vitamin D receptor gene polymorphisms as a possible determinant of growth in male and female infants. J Clin Endocrinol Metab 83:3563–3568. 104. Lorentzon M, Lorentzon R, Nordstrom P 2000 Vitamin D receptor gene polymorphism is associated with birth height, growth to adolescence, and adult stature in healthy Caucasian men: a cross-sectional and longitudinal study. J Clin Endocrinol Metab 85:1666–1670. 105. Dennison EM, Arden NK, Keen RW, Syddall H, Day INM, Spector TD, Cooper C 2001 Birthweight, vitamin D receptor genotype, and the programming of osteoporosis. Pediatr Perinat Epidemiol 15:211–219. 106. van der Sluis IM, de Muinck Keizer-Schrama SMPF, Krenning EP, Pols HAP, Uitterlinden AG 2003 Vitamin D receptor gene polymorphisms predict height and bone size, rather than bone density in children and young adults. Calc Tissue Int. 73:332–338. 107. Keen RW, Egger P, Fall C, Major PJ, Lanchbury JS, Spector TD, Cooper C 1997 Polymorphisms of the vitamin D receptor, infant growth, and adult bone mass. Calc Tissue Int 60:233–235. 108. Suarez F, Zegoud F, Rossignol C, Walrant O, Garabedian M 1997 Association between vitamin D receptor gene polymorphisms and sex-dependent growth during the first two years of life. J Clin Endocrinol Metab 82:2966–2970. 109. Baroncelli GI, Federico G, Bertelloni S, Ceccarelli C, Cupelli D, Saggese G 1999 Vitamin D receptor genotype does not predict bone mineral density, bone turnover, and growth in prepubertal children. Horm Res 51:150–156. 110. Minamitani K, Takahashi Y, Minagawa M, Yasuda T, Niimi H 1998 Difference in height associated with a translation start site polymorphism in the vitamin D receptor gene. Ped Res 44:628–632. 111. Geussens P, VandeVyver C, VanHoof J, Cassiman J-J, Boonen S, Raus J 1997 Quadriceps and grips strength are related to vitamin D receptor genotype in elderly nonobese women. J Bone Miner Res 12:2082–2088. 112. Beaven S, Prentice A, Yan L, Dibba B, Ralston S 1996 Differences in vitamin D receptor genotype and geographical variation in osteoporosis. Lancet 348:136–137. 113. Young RP, Lau EM, Birjandi Z, Critchley JA, Woo J 1996 Interethnic differences in hip fracture rate and the vitamin D receptor polymorphism. Lancet 348:688–689. 114. Berg JP, Falch JA, Haug E 1996 Fracture rate, pre- and postmenopausal bone loss are not associated with vitamin D

115.

116.

117.

118.

119.

120.

121.

122.

123.

124.

125.

126. 127.

128.

1153

receptor genotype in a high-endemic area of osteoporosis. Eur J Endocrinol 135:96–100. Feskanich D, Hunter DJ, Willett WC, Hankinson SE, Hollis BW, Hough HL, Kelsey KT, Colditz GA 1998 Vitamin D receptor genotype and the risk of bone fractures in women. Epidemiol 9:535–539. Gomez C, Naves ML, Barrios Y, Diaz JB, Fernandez JL, Salido E, Torres A, Cannata JB 1999 Vitamin D receptor gene polymorphisms, bone mass, bone loss, and prevalence of vertebral fracture: differences in postmenopausal women and men. Osteop Int 10:175–182. Langdahl BL, Gravholt CH, Brixen K, Eriksen EF 2000 Polymorphisms in the vitamin D receptor gene and bone mass, bone turnover, and osteoporotic fractures. Eur J Clin Invest 30:608–617. Uitterlinden AG, Weel AEAM, Burger H, Fang Y, van Duijn CM, Hofman A, van Leeuwen JPTM, Pols HAP 2001 Interaction between the vitamin D receptor gene and collagen type Iα1 gene in susceptibility for fracture. J Bone Miner Res 16:379–385. White CP, Nguyen TV, Jones G, Morrison NA, Gardiner EM, Kelly PJ, Sambrook PN 1994 Vitamin D receptor alleles predict osteoporotic fracture risk. J Bone Miner Res S263 (abstract). Houston LA, Grant SFA, Reid DM, Ralston SR 1996 Vitamin D receptor polymorphism, bone mineral density, and osteoporotic vertebral fracture: studies in a UK population. Bone 18:249–252. Ramalho AC, Lazaretti-Castro M, Hauache O, Kasamatsu T, Brandao C, Reis AF, Takata E, Cafalli F, Tavares F, Gimeno SG, Vieira JG 1998 Fractures of the proximal femur: correlation with vitamin D receptor gene polymorphism. Braz J Med Biol Res 31:921–927. Ensrud KE, Stone K, Cauley JA, White C, Zmuda JM, Nguyen TV, Eisman JA, Cummings SR 1999 Vitamin D receptor gene polymorphisms and the risk of fractures in older women. For the Study of Osteoporotic Fractures Research Group. J Bone Miner Res 14:1637–1645. Valimaki S, Tahtela R, Kainulainen K, Laitinen K, Loyttyniemi E, Sulkava R, Valimaki M, Kontula K 2001 Relation of collagen type I alpha 1 (COLIA1) and vitamin D receptor genotypes to bone mass, turnover, and fractures in early postmenopausal women and to hip fractures in elderly people. Eur J Intern Med 12:48–56. Gennari L, Becherinni L, Mansani R, Masi L, Falchetti A, Morelli A, Colli E, Gonelli S, Cepollaro C, Brandi ML 1999 FokI polymorphism at translation initiation site of the vitamin D receptor gene predicts bone mineral density and vertebral fractures in postmenopausal Italian women. J Bone Miner Res 14:1379–1386. Balmain N, Hauche corne M, Pike JW, Cuisinier-Gleizes P, Mathieu P 1993 Distribution and subcellular immunolocalization of the 1,25 dihydroxy vitamin D3 receptors in rat epiphysial cartilage. Cell Mol Biol 39:339–350. Dodds RA, Gowen M 1994 The growing osteophyte: a model system for the study of human bone development and remodelling in situ. J Histotechnol 17:37–45. McAlindon TE, Felson DT, Zhang Y, Hannan MT, Aliabadi P, Weissman B, Rush D, Wilson PWF, Jacques P 1996 Relation of dietary intake and serum levels of vitamin D to progression of osteoarthritis of the knee among participants in the Framingham Study. Ann Intern Med 125:353–359. Uitterlinden AG, Burger H, Huang Q, Odding E, van Duijn CM, Hofman A, Birkenhager JC, van Leeuwen JPTM, Pols HAP

1154

129.

130. 131. 132.

133.

134.

135.

136.

137.

138. 139.

140.

141.

142.

143.

ANDRÉ G. UITTERLINDEN, YUE FANG, JOYCE B.J. VAN MEURS, AND HUIBERT A.P. POLS

1997 Vitamin D receptor genotype is associated with radiographic osteoarthritis at the knee. J Clin Invest 100:259–263. Keen RW, Hart DJ, Lanchbury JS, Spector TD 1997 Association of early osteoarthritis of the knee with a Taq I polymorphism of the vitamin D receptor gene. Arthr Rheum 40:1444–1449. Granchi D, Stea S, Sudanese A, Toni A, Baldini N, Giunti A 2002 Association of two gene polymorphisms with osteoarthritis secondary to hip dysplasia. Clin Orthop 403:108–117. Jones G, White C, Sambrook P, Eisman J 1998 Allelic variation in the vitamin D receptor, lifestyle factors and lumbar spinal degenerative disease. Ann Rheum Dis 57:94–99. Videman T, Leppavuori J, Kaprio J, Battie M, Gibbons LE, Peltonen L, Koskenvuo M 1998 Intragenic polymorphisms of the vitamin D receptor gene associated with intervertebral disc degeneration. Spine 23:2477–2485. Aerssens J, Dequeker J, Peeters J, Breemans S, Boonen S 1998 Lack of association between osteoarthritis of the hip and gene polymorphisms of VDR, COLIA1, and COL2A1 in postmenopausal women. Arthr Rheum 41:1946–1950. Huang J, Ushiyama T, Inoue K, Kawasaki T, Hukuda S 2000 Vitamin D receptor gene polymorphisms and osteoarthritis of the hand, hip, knee: a case-control study in Japan. Rheumatology 39:79–84. Loughlin J, Sinsheimer JS, Mustafa Z, Carr AJ, Clipsham K, Bloomfield VA, Chitnavis J, Bailey A, Sykes B, Chapman K 2000 Association of the vitamin D receptor gene, the type I collagen gene COLIA1, and the estrogen receptor gene in idiopathic osteoarthritis. J Rheum 27:779–784. Baldwin CT, Cupples LA, Joost O, Demissie S, Chaisson C, McAlindon T, Myers RH, Felson D 2002 Absence of linkage or association for osteoarthritis with the vitamin D receptor/type II collagen locus: the Framingham Osteoarthritis Study. J Rheum 29:161–165. Uitterlinden AG, Burger H, van Duijn CM, Huang Q, Hofman A, Birkenhager JC, van Leeuwen JPTM, Pols HAP 2000 Adjacent genes for COL2A1 and the vitamin D receptor are associated with separate features of radiographic osteoarthritis of the knee. Arthr Rheum 43:1456–1464. Taylor JA, Hirvonen A, Watson M, Pittman G, Mohler JL, Bell DA 1996 Association of prostate cancer with vitamin D receptor gene polymorphism. Cancer Res 56:4108–4110. Ingles SA, Ross RK, Yu MC, Irvine RA, La Pera G, Haile RW, Coetzee GA 1997 Association of prostate cancer risk with genetic polymorphisms in vitamin D receptor and androgen receptor. J Natl Cancer Inst 89:166–170. Ma J, Stampfer MJ, Gann PH, Hough HL, Giovannucci E, Kelsey KT, Hennekens CH, Hunter DJ 1998 Vitamin D receptor polymorphisms, circulating vitamin D metabolites, and risk of prostate cancer in United States physicians. Cancer Epidemiol Biomarkers Prev 7:385–390. Habuchi T, Suzuki T, Sasaki R, Wang L, Sato K, Satoh S, Akao T, Tsuchiya N, Shimoda N, Wada Y, Koizumi A, Chihara J, Ogawa O, Kato T 2000 Association of vitamin D receptor gene polymorphism with prostate cancer and benign prostatic hyperplasia in a Japanese population. Cancer Res 60:305–308. Medeiros R, Morais A, Vasconcelos A, Costa S, Pinto D, Oliveira J, Lopes C 2002 The role of vitamin D receptor gene polymorphisms in the susceptibility to prostate cancer of a southern European population. J Hum Genet 47: 413–418. Hamasaki T, Inatomi H, Katoh T, Ikuyama T, Matsumoto T 2002 Significance of vitamin D receptor gene polymorphism

144.

145.

146.

147. 148. 149.

150. 151.

152.

153.

154.

155.

156.

157.

158.

for risk and disease severity of prostate cancer and benign prostatic hyperplasia in Japanese. Urol Int 68:226–231. Hamasaki T, Inatomi H, Katoh T, Ikuyama T, Matsumoto T 2001 Clinical and pathological significance of vitamin D receptor gene polymorphism for prostate cancer, which is associated with a higher mortality in Japanese. Endocr J 48:543–549. Ingles SA, Coetzee GA, Ross RK, Henderson BE, Kolonel LN, Crocitto L, Wang W, Haile RW 1998 Association of prostate cancer with vitamin D receptor haplotypes in AfricanAmericans. Cancer Res 58:1620–1623. Correa-Cerro L, Berthon P, Haussler J, Bochum S, Drelon E, Mangin P, Fournier G, Paiss T, Cussenot O, Vogel W 1999 Vitamin D receptor polymorphisms as markers in prostate cancer. Hum Genet 105:281–287. Kibel AS, Isaacs SD, Isaacs WB, Bova GS 1998 Vitamin D receptor polymorphisms and lethal prostate cancer. J Urol 160:1405–1409. Furuya Y, Akakura K, Masai M, Ito H 1999 Vitamin D receptor gene polymorphism in Japanese patients with prostate cancer. Endocr J 46:467–470. Watanabe M, Fukutome K, Murata M, Uemura H, Kubota Y, Kawamura J, Yatani R 1999 Significance of vitamin D receptor gene polymorphism for prostate cancer risk in Japanese. Anticancer Res 19:4511–4514. Blazer DG, III, Umbach DM, Bostick RM, Taylor JA 2000 Vitamin D receptor polymorphisms and prostate cancer. Mol Carcinog 27:18–23. Bousema JT, Bussemakers MJ, van Houwelingen KP, Debruyne FM, Verbeek AL, de la Rosette JJ, Kiemeney LA 2000 Polymorphisms in the vitamin D receptor gene and the androgen receptor gene and the risk of benign prostatic hyperplasia. Eur Urol 37:234–238. Schatzl G, Gsur A, Bernhofer G, Haidinger G, Hinteregger S, Vutuc C, Haitel A, Micksche M, Marberger M, Madersbacher S 2001 Association of vitamin D receptor and 17 hydroxylase gene polymorphisms with benign prostatic hyperplasia and benign enlargement. Urology 57:567–572. Suzuki K, Matsui H, Ohtake N, Nakata S, Takei T, Koike H, Nakazato H, Okugi H, Hasumi M, Fukabori Y, Kurokawa K, Yamanaka H 2003 Vitamin D receptor gene polymorphism in familial prostate cancer in a Japanese population. Int J Urol 10:261–266. Tayeb MT, Clark C, Haites NE, Sharp L, Murray GI, McLeod HL 2003 CYP3A4 and VDR gene polymorphisms and the risk of prostate cancer in men with benign prostate hyperplasia. Br J Cancer 88:928–932. Gsur A, Madersbacher S, Haidinger G, Schatzl G, Marberger M, Vutuc C, Micksche M 2002 Vitamin D receptor gene polymorphism and prostate cancer risk. Prostate 51:30–34. Chokkalingam AP, McGlynn KA, Gao YT, Pollak M, Deng J, Sesterhenn IA, Mostofi FK, Fraumeni JF, Hsing AW 2001 Vitamin D receptor gene polymorphisms, insulin-like growth factors, and prostate cancer risk: a population-based, casecontrol study in China. Cancer Res 61:4333–4336. Xu Y, Shibata A, McNeal JE, Stamey TA, Feldman D, Peehl DM 2003 Vitamin D receptor start codon polymorphism (FokI) and prostate cancer progression. Cancer Epidemiol Biomarkers Prev 12:23–27. Luscombe CJ, French ME, Liu S, Saxby MF, Jones PW, Fryer AA, Strange RC 2001 Outcome in prostate cancer associations with skin type and polymorphism in pigmentationrelated genes. Carcinogenesis 22:1343–1347.

CHAPTER 68 Genetic Vitamin D Receptor Polymorphisms and Risk of Disease

158a. Ntais C, Polycarpou A, Ioannidis JPA 2003 Vitamin D receptor polymorphisms and risk of prostate cancer: A meta-analysis. Cancer Epidemiol Biomark Prevent 12: 1395–1402. 159. Ruggiero M, Pacini S, Aterini S, Fallai C, Ruggiero C, Pacini P 1998 Vitamin D receptor gene polymorphism is associated with metastatic breast cancer. Oncol Res 10:43–46. 160. Curran JE, Vaughan T, Lea RA, Weinstein SR, Morrison NA, Griffiths LR 1999 Association of a vitamin D receptor polymorphism with sporadic breast cancer development. Int J Cancer 83:723–726. 161. Lundin AC, Soderkvist P, Eriksson B, Bergman-Jungestrom M, Wingren S 1999 Association of breast cancer progression with a vitamin D receptor gene polymorphism. Southeast Sweden Breast Cancer Group. Cancer Res 59:2332–2334. 162. Bretherton-Watt D, Given-Wilson R, Mansi JL, Thomas V, Carter N, Colston KW 2001 Vitamin D receptor gene polymorphisms are associated with breast cancer risk in a UK Caucasian population. Br J Cancer 85:171–175. 163. Guy M, Lowe LC, Bretherton-Watt D, Mansi JL, Colston KW 2003 Approaches to evaluating the association of vitamin D receptor polymorphisms with breast cancer risk. Recent Results Cancer Res 164:43–54. 164. Ingles SA, Garcia DG, Wang W, Nieters A, Henderson BE, Kolonel LN, Haile RW, Coetzee GA 2000 Vitamin D receptor genotype and breast cancer in Latinas (United States). Cancer Causes Control 11:25–30. 165. Hou MF, Tien YC, Lin GT, Chen CJ, Liu CS, Lin SY, Huang TJ 2002 Association of vitamin D receptor gene polymorphism with sporadic breast cancer in Taiwanese patients. Breast Cancer Res Treat 74:1–7. 166. Schondorf T, Eisberg C, Wassmer G, Warm M, Becker M, Rein DT, Gohring UJ 2003 Association of the vitamin D receptor genotype with bone metastases in breast cancer patients. Oncology 64:154–159. 167. Dunning AM, McBride S, Gregory J, Durocher F, Foster NA, Healey CS, Smith N, Pharoah PD, Luben RN, Easton DF, Ponder BA 1999 No association between androgen or vitamin D receptor gene polymorphisms and risk of breast cancer. Carcinogenesis 20:2131–2135. 168. Newcomb PA, Kim H, Trentham-Dietz, Farin F, Hunter D, Egan KM 2002 Vitamin D receptor polymorphism and breast cancer risk. Cancer Epidemiol Biomark Prev 11:1503–1504. 169. Kim HS, Newcomb PA, Ulrich CM, Keener CL, Bigler J, Farin FM, Bostick RM, Potter JD 2001 Vitamin D receptor polymorphism and the risk of colorectal adenomas: evidence of interaction with dietary vitamin D and calcium. Cancer Epidemiol Biomarkers Prev 10:869–874. 170. Speer G, Dworak O, Cseh K, Bori Z, Salamon D, Torok I, Winkler G, Vargha P, Nagy Z, Takacs I, Kucsera M, Lakatos P 2000 Vitamin D receptor gene BsmI polymorphism correlates with erbB-2/HER-2 expression in human rectal cancer. Oncology 58:242–247. 171. Ingles SA, Wang J, Coetzee GA, Lee ER, Frankl HD, Haile RW 2001 Vitamin D receptor polymorphisms and risk of colorectal adenomas (United States). Cancer Causes Control 12:607–614. 172. Wong HL, Seow A, Arakawa K, Lee HP, Yu MC, Ingles SA 2003 Vitamin D receptor start codon polymorphism and colorectal cancer risk: effect modification by dietary calcium and fat in Singapore Chinese. Carcinogenesis 24:1091–1095.

1155

173. Peters U, McGlynn KA, Chatterjee N, Gunter E, GarciaClosas M, Rothman N, Sinha R 2001 Vitamin D, calcium, and vitamin D receptor polymorphism in colorectal adenomas. Cancer Epidemiol Biomarkers Prev 10:1267–1274. 174. Slatter ML, Yakumo K, Hoffman M, Neuhausen S 2001 Variants of the VDR gene and risk of colon cancer (United States). Cancer Causes Control 12:359–364. 175. Ramachandran S, Fryer AA, Lovatt TJ, Smith AG, Lear JT, Jones PW, Stange RC 2003 Combined effects of gender, skin type, and polymorphic genes on clinical phenotype: use of rate of increase in numbers of basal cell carcinoma as a model system. Cancer Lett 189:175–181. 176. Hutchinson PE, Osborne JE, Lear JT, Smith AG, Bowers PW, Morris PN, Jones PW, York C, Strange RC, Fryer AA 2000 Vitamin D receptor polymorphisms are associated with altered prognosis in patients with malignant melanoma. Clin Cancer Res 6:498–504. 177. Ramachandran S, Fryer AA, Smith AG, Lear JT, Bowers B, Hartland AJ, Whiteside JR, Jones PW, Strange RC 2001 Basal cell carcinomas: association of allelic variants with a high-risk subgroup of patients with the multiple presentation phenotype. Pharmacogenetics 11:247–254. 178. Ikuyama T, Hamasaki T, Inatomi H, Katoh T, Muratani T, Matsumoto T 2002 Association of vitamin D receptor gene polymorphism with renal cell carcinoma in Japanese. Endocr J 49:433–438. 179. Carling T, Kindmark A, Hellman P, Lundgren E, Ljunghall S, Rastad J, Akerstrom G, Melhus H 1995 Vitamin D receptor genotypes in primary hyperparathyroidism. Nat Med 1: 1309–1311. 180. Carling T, Kindmark A, Hellman P, Holmberg L, Akerstrom G, Rastad J 1997 Vitamin D receptor alleles b, a, and T: risk factors for sporadic primary hyperparathyroidism (HPT), but not HPT of uremia or MEN1. Biochem Biophys Res Commun 231:329–332. 181. Carling T, Ridefelt P, Hellman P, Juhlin C, Lundgren E, Akerstrom G, Rastad J 1998 Vitamin D receptor gene polymorphism and parathyroid calcium sensor protein (CAS/gp330) expression in primary hyperparathyroidism. World J Surg 22:700–706. 182. Nagaba Y, Heishi M, Tazawa H, Tsukamoto Y, Kobayashi Y 1998 Vitamin D receptor gene polymorphisms affect secondary hyperparathyroidism in hemodialyzed patients. Am J Kidney Dis 32:464–469. 183. Menarguez J, Goicoechea M, Crist bal E, Arribas B, Martinez ME, Alcazar JA, Carri n R, Polo JR 1999 Lack of relationship betwen BsmI vitamin D receptor polymorphism and primary hyperparathyroidism in a Spanish female population. Calc Tissue Int 65:214–216. 184. Sosa M, Torres A, Martin N, Salido E, Liminana JM, Barrios Y, De Miguel E, Betancor P 2000 The distribution of two different vitamin D receptor polymorphisms (BsmI and start codon) in primary hyperparathyroidism. J Intern Med 247:124–130. 185. Pacheo D, Menarguez J, Cristobal E, Arribas B, Alcazar JA, Carrion R, Polo JR 2000 BsmI vitamin D receptor polymorphism and pathogenesis of parathyroid adenoma. Med Sci Monit 6:658–660. 186. Park B-S, Park J-S, Lee D-Y, Youn J-I, Kim I-G 1999 Vitamin D receptor polymorphism is associated with psoriasis. J Invest Derm 112:113–116. 187. Okita H, Ohtsuka T, Yamakage A, Yamazaki S 2002 Polymorphism of the vitamin D(3) receptor in patients with psoriasis. Arch Dermatol Res 294:159–162.

1156

ANDRÉ G. UITTERLINDEN, YUE FANG, JOYCE B.J. VAN MEURS, AND HUIBERT A.P. POLS

188. Saeki H, Asano N, Tsunemi Y, Takekoshi T, Kishimoto M, Mitsui H, Tada Y, Torii H, Komine M, Asahina A, Tamaki K 2002 Polymorphisms of the vitamin D receptor gene in Japanese patients with psoriasis vulgaris. J Derm Sci 30:167–171. 189. Mee JB, Cork MJ 1998 Vitamin D receptor polymorphism and calcipotriol response in patients with psoriasis. J Invest Derm 110:301–302. 190. Kontula K, Valimaki S, Kainulainen K, Viitanen A-M, Keski-Oja J 1997 Vitamin D receptor polymorphism and treatment of psoriasis with calcipotriol. Br J Derm 136:977–978. 191. Lee DY, Park BS, Choi KH, Jeon JH, Cho KH, Song KY, Kim IG, Youn JI 2002 Vitamin D receptor genotypes are not associated with clinical response to calcipotriol in Korean psoriasis patients. Arch Dermatol Res 294:1–5. 192. Scott P, Ouimet D, Valiquette L, Guay G, Proulx Y, Trouvé M-L, Gagnon B, Bonnardeaux A 1999 Suggestive evidence for a susceptibility gene near the vitamin D receptor locus in idiopathic calcium stone formation. J Am Soc Nephrol 10:1007–1013. 193. Ruggiero M, Pacini S, Amato M, Aterini S, Chiarugi V 1999 Association between vitamin D receptor gene polymorphism and nephrolithiasis. Miner Electrolyte Metab 25:185–190. 194. Jackman SV, Kibel AS, Ovuworie CA, Moore RG, Kavoussi LR, Jarrett TW 1999 Familial calcium stone disease: TaqI polymorphism and the vitamin D receptor. J Endourol 13:313–316. 195. Mosetti G, Vuotto P, Rendina D, Numis FG, Viceconti R, Girodano F, Cioffi M, Scopacasa F, Nunziata V 2003 Association between vitamin D receptor gene polymorphisms and tubular citrate handling in calcium nephrolithiasis. J Intern Med 253:194–200. 196. Ozkaya O, Soylemezoglu M, Gonen S, Buyan N, Hasanoglu E 2003 Polymorphisms in the vitamin D receptor gene and the risk of calcium nephrolithiasis in children. Eur Urol 44:150–154. 197. Chen WC, Chen HY, Hsu CD, Wu JY, Tsai FJ 2001 No association of vitamin D receptor gene BsmI polymorphisms with calcium oxalate stone formation. Mol Urol 5:7–10. 198. Zerwekh JE, Hughes MR, Reed BY, Breslau NA, Heller HJ, Lemke M, Nasonkin I, Pak CY 1995 Evidence for normal vitamin D receptor messenger ribonucleic acid and genotype in absorptive hypercalciuria. J Clin Endocrinol Metab 80:2960–2965. 199. McDermott MF, Ramachandran A, Ogunkolade BW, Aganna E, Curtis D, Boucher BJ, Snehalatha C, Hitman GA 1997 Allelic variation in the vitamin D receptor influences susceptibility to IDDM in Indian Asians. Diabetologica 40:971–975. 200. Guja C, Marshall S, Welsh K, Merriman M, Smith A, Todd JA, Ionescu-Tirgoviste C 2002 The study of CTLA-4 and vitamin D receptor polymorphisms in the Romanian type 1 diabetes population. J Cell Mol Med 6:75–81. 201. Pani MA, Knapp M, Donner H, Braun J, Baur J, Baur MP, Usadel KH, Badenhoop K 2000 Vitamin D receptor allele combinations influence genetic susceptibility to type 1 diabetes in Germans. Diabetes 49:504–507. 202. Chang TJ, Lei HH, Yeh JI, Chiu KC, Lee KC, Chen MC, Tai TY, Chuang LM 2000 Vitamin D receptor gene polymorphisms influence susceptibility to type 1 diabetes mellitus in the Taiwanese population. Clin Endocrinol 52:575–580. 203. Fassbender WJ, Goertz B, Weismuller K, Steinhauer B, Stracke H, Auch D, Linn T, Bretzel RG 2002 VDR gene polymorphisms are overrepresented in German patients with type 1 diabetes compared to healthy controls without effect on biochemical parameters of bone metabolism. Horm Metab Res 34:330–337.

204. Gyorffy B, Vasarhelyi B, Krikovszky D, Madacsy L, Tordai A, Tulassay T, Szabo A 2002 Gender-specific association of vitamin D receptor polymorphism combinations with type 1 diabetes mellitus. Eur J Endocrinol 147:803–808. 205. Taverna MJ, Sola A, Guyot-Argenton C, Pacher N, Bruzzo F, Slama G, Reach G, Selam JL 2002 Taq I polymorphism of the vitamin D receptor and risk of severe retinopathy. Diabetologica 45:436–442. 206. Skrabic V, Zemunik T, Situm M, Terzic J 2003 Viatmin D receptor polymorphism and susceptibility to type 1 diabetes in the Dalmatian population. Diabetes Res Clin Prac 59: 31–35. 207. Motohashi Y, Yamada S, Yanagawa T, Maruyama T, Suzuki R, Niino M, Fukazawa T, Kasuga A, Hirose H, Matsubara K, Shimada A, Saruta T 2003 Vitamin D receptor gene polymorphism affects onset pattern of type 1 diabetes. J Clin Endocrinol Metab 88:3137–3140. 208. Aterini S, Pacini S, Amato M, Ruggiero M 2000 Vitamin D receptor gene polymorphism and diabetes mellitus prevalence in hemodialysis patients. Nephron 84:186. 209. Yokota I, Satomura S, Kitamura S, Taki Y, Naito E, Ito M, Nishisho K, Kuroda Y 2002 Association between vitamin D receptor genotype and age of onset in juvenile Japanese patients with type 1 diabetes. Diabetes Care 25:1244. 210. Ban Y, Taniyama M, Yanagawa T, Yamada S, Maruyama T, Kasuga A, Ban Y 2001 Vitamin D receptor initiation codon polymorphism influences genetic susceptibility to type 1 diabetes mellitus in the Japanese population. BMC Medical Genetics 2:7 (http://www.biomedcentral.com/14712350/2/7). 211. Hitman GA, Mannan N, McDermott MF, Aganna E, Ogunkolade BW, Hales CN, Boucher BJ 1998 Vitamin D receptor gene polymorphisms influence insulin secretion in Bangladeshi Asians. Diabetes 47:688–690. 211a. Ogunkolade BW, Boucher BJ, Prahl JM, Bustin SA, Burrin JM, Noonan K, North BV, Mannan N, McDermott MF, DeLuca HF, Hitman GA 2002 Vitamin D receptor (VDR) mRNA and VDR protein levels in relation to vitamin D status, insulin secretory capacity, and VDR genotype in Bangladeshi Asians. Diabetes 51:2294–2300. 212. Speer G, Cseh K, Winkler G, Vargha P, Braun E, Takacs I, Lakatos P 2001 Vitamin D and estrogen receptor gene polymorphisms in type 2 diabetes mellitus and in android type obesity. Eur J Endocrinol 144:385–389. 213. Ye WZ, Reis AF, Dubois-Laforgue D, Bellane-Chantelot C, Timsit J, Velho G 2001 Vitamin D receptor gene polymorphisms are associated with obesity in type 2 diabetic subjects with early age of onset. Eur J Endocrinol 145: 181–186. 214. Oh JY, Barett-Connor E 2002 Association between vitamin D receptor polymorphism and type 2 diabetes or metabolic syndrome in community dwelling older adults: the Ranch Bernardo Study. Metabolism 51:356–359. 215. Ortlepp JR, Lauscher J, Hoffman R, Hanrath P, Joost HG 2001 The vitamin D receptor gene variant is associated with the prevalence of type 2 diabetes mellitus and coronary artery disease. Diabet Med 18:842–845. 216. Ortlepp JR, Metrikat J, Albrecht M, von Korff A, Hanrath P, Hoffman R 2003 The vitamin D receptor gene variant and physical activity predicts fasting glucose levels in healthy young men. Diabet Med 20:451–454. 217. Boullu-Sanchis S, Lepretre F, Hedelin G, Donnet JP, Schaffer P, Froguel P, Pinget M 1999 Type 2 diabetes mellitus: association study of five candidate genes in an Indian population of Guadeloupe, genetic contribution of FAB2 polymorphism. Diabetes Metab 25:150–156.

CHAPTER 68 Genetic Vitamin D Receptor Polymorphisms and Risk of Disease

218. Niimi T, Tomita H, Sato S, Kawaguchi H, Akita K, Maeda H, Sugiura Y, Ueda R 1999 Vitamin D receptor gene polymorphisms in patients with sarcoidosis. Am J Respir Crit Care Med 160:1107–1109. 219. Guleva I, Seitzer U 2000 Vitamin D receptor gene polymorphism in patients with sarcoidosis. Am J Respir Crit Care Med 162:760–761. 220. Niimi T, Tomita H, Sato S, Akita K, Maeda H, Kawaguchi H, Mori T, Sugiura Y, Yoshinouchi T, Ueda R 2000 Vitamin D receptor gene polymorphism and calcium metabolism in sarcoidosis patients. Sarcoidosis Vasc Diffuse Ling Dis 17:266–269. 221. Fukazawa T, Yabe I, Kikuchi S, Sasaki H, Hamada T, Miyasaka K, Tashiro K 1999 Association of vitamin D receptor gene polymorphism with multiple sclerosis in Japanese. J Neurol Sci 166:47–52. 222. Niino M, Fukazawa T, Yabe I, Kikuchi S, Sasaki H, Tashiro K 2000 Vitamin D receptor gene polymorphism in multiple sclerosis and the association with HLA class II alleles. J Neurol Sci 177:65–71. 222a. Steckley JL, Dyment DA, Sadovnick AD, Risch N, Hayes C, Ebers GC, and the Canadian Collaborative Study Group 2000 Genetic analysis of vitamin D related genes in Canadian multiple sclerosis patients. Neurology 54:729–732. 222b. Garcia-Lozano JR, Gonzalez-Escribano MF, Valenzuela A, Garcia A, Nunez-Roldan A 2001 Association of vitamin D receptor genotypes with early onset rheumatoid arthritis. Eur J Immunogenet 28:89–93. 222c. Lee CK, Hong JS, Cho YS, Yoo B, Kim GS, Moon HB 2001 Lack of relationship between vitamin D receptor polymorphism and bone erosion in rheumatoid arthritis. J Korean Med Sci 16:188–192. 222d. Goertz B, Fassbender WJ, Williams JC, Marzeion AM, Bretzel RG, Stracke H, Berliner MN 2003 Vitamin D receptor genotypes are not associated with rheumatoid arthritis or biochemical parameters of bone turnover in German RA patients. Clin Exp Rheumatol 21:333–339. 223. Simmons JD, Mullighan C, Welsh KI, Jewell DP 2000 Vitamin D receptor gene polymorphism: association with Crohns’ disease susceptibility. Gut 47:211–214. 224. Martin K, Radlmayr M, Borchers R, Heinzlmann M, Folwaczny C 2002 Candidate genes co-localized to linkage regions in inflammatory bowel disease. Digestion 66:121–126. 225. Ban Y, Taniyama M, Ban Y 2000 Vitamin D receptor gene polymorphism is associated with Graves’ disease in the Japanese population. J Clin Endocrinol Metab 85:4639–4643. 226. Ban Y, Ban Y, Taniyama M, Katagiri T 2000 Vitamin D receptor initiation codon polymorphism in Japanese patients with Graves’ disease. Thyroid 10:475–480. 227. Pani MA, Seissler J, Usadel KH, Badenhoop K 2002 Vitamin D receptor genotype is associated with Adison’s disease. Eur J Endocrinol 147:635–640. 228. Liu W, Zhang CY, Wu XM, Tian L, Li CZ, Zhao QM, Zhang PH, Yang SM, Yang H, Zhang XT, Cao WC 2003 A case-control study on the vitamin D receptor gene polymorphisms and susceptibility to pulmonary tuberculosis. Zhonghua Liu Xing Bing Xue Za Zhi 24:389–392. 229. Delgado JC, Baena A, Thim S, Goldfeld AE 2002 Ethnicspecific genetic associations with pulmonary tuberculosis. J Infect Dis 186:1463–1468. 230. Roy S, Frodsham A, Saha B, Hazra SK, Mascie-Taylor CGN, Hill AVS 1999 Association of vitamin D receptor genotype with Leprosy type. J Infect Dis 179:187–191.

1157

231. Gelder CM, Hart KW, Williams OM, Lyons E, Welsh KI, Campbell IA, Marshall SE 2000 Vitamin D receptor gene polymorphisms and susceptibility to Mycobacterium malmoense pulmonary disease. J Infect Dis 181:2099–2102. 232. Bellamy R, Ruwende C, Corrah T, McAdam KPWJ, Thursz M, Whittle HC, Hill AVS 1999 Tuberculosis and chronic hepatitis B virus infection in Africans and variation in the vitamin D receptor gene. J Infect Dis 179:721–724. 233. Wilkinson RJ, Llewelyn M, Toosi Z, Patel P, Pasvol G, Lalvani A, Wright D, Latif M, Davidson RN 2000 Influence of vitamin D deficiency and vitamin D receptor polymorphisms on tuberculosis among Gujarati Asians in west London: a case-control study. Lancet 355:618–621. 234. Barber Y, Rubio C, Fernandez E, Rubio M, Fibla J 2001 Host genetic background at CCR5 Chemkine Recptor and vitamin D receptor loci and human immunodeficiency and (HIV) type 1 disease progression and HIV-seropositive injection drug users. J Infect Dis 184:1279–1288. 235. Henning BJ, Parkhill JM, Chapple IL, Heasman PA, Taylor JJ 1999 Association of vitamin D receptor gene polymorphisms with localized early-onset periodontal diseases. J Perdiodontol 70:1032–1038. 236. Tachi Y, Shimpuku H, Nosaka Y, Kawamura T, Shinohara M, Ueda M, Imai H, Ohura K 2003 Vitamin D receptor gene polymorphisms is associated with chronic periodontitis. Life Sci 73:3313–3321. 237. Muray S, Parisi E, Cardus A, Craver L, Fernandez E 2003 Influence of vitamin D receptor gene polymorphisms and 25-hydroxyvitamin D on blood pressure in apparently healthy subjects. J Hypertens 21:2069–2075. 238. Lee BK, Lee GS, Stewart WF, Ahn KD, Simon D, Kelsey KT, Todd AC, Schwartz BS 2001 Associations of blood pressure and hypertension with lead dose measures and polymorphisms in the vitamin D receptor and delta-aminovulinic acid dehydratase genes. Environ Health Perspect 109: 383–389. 239. Nakano Y, Oshima T, Sasaki S, Yamaoka K, Matsumoto T, Hirao H, Ozono R, Matsuura H, Kajiyama G, Kambe M 2000 Vitamin D receptor gene polymorphism is associated with serum total and ionized calcium concentration. J Mol Med 78:575–579. 240. Van Schooten FJ, Hirvonen A, Maas LM, de Mol BA, Kleinjans JC, Bell DA, Durrer JD 1998 Putative susceptibility markers of coronary artery disease: association between VDR genotype, smoking, and aromatic DNA adduct levels in human right atrial tissue. FASEBJ 12:1409–1417. 241. Ortlepp JR, Hoffman R, Ohme F, Lauscher J, Bleckmann F, Hanrath P 2001 The vitamin D receptor genotype predisposes to the development of calcific aortic valve stenosis. Heart 85:635–638. 242. Ortlepp JR, von Korff A, Hanrath P, Zerres K, Hoffman R 2003 Vitamin D receptor gene polymorphism BsmI is not associated with the prevalence and severity of CAD in a large-scale angiographic cohort of 3441 patients. Eur J Clin Invest 33:106–109. 243. Zittermann A, Schulze Schleithoff S, Tenderich G, Berthold HK, Korfer R, Stehle P 2003 Low vitamin D status: a contributing factor in the pathogenesis of congestive heart failure JACC 41:105–112. 244. Uitterlinden AG, Burger H, Witteman JCM, van Leeuwen JPTM, Pols HAP 1998 Genetic relation between osteoporosis and cardiovascular disease: vitamin D receptor polymorphism predicts myocardial infarction. Osteop Int 8: (abstract OR12).

CHAPTER 69

Clinical Disorders of Phosphate Homeostasis MARC K. DREZNER

Department of Medicine, University of Wisconsin and Geriatrics Research Education and Clinical Center, William H. Middleton Veterans Administration Medical Center, Madison, Wisconsin

I. Introduction II. Disorders of Phosphate Homeostasis

I. Introduction Extensive studies over the past several decades have established that phosphate homeostasis and vitamin D metabolism are reciprocally regulated. As discussed in Chapters 26 and 29, calcitriol promotes phosphate absorption from the intestine, mobilization from bone, and reabsorption in the renal tubule [1]. In turn, phosphate depletion or hypophosphatemia stimulates renal production of 1,25(OH)2D (calcitriol), while phosphate overload or hyperphosphatemia inhibits renal 25(OH)D1α-hydroxylase activity [2]. Since phosphorus is one of the most abundant constituents of all tissues, disturbances in phosphate homeostasis can affect almost any organ system [3]. Indeed, a deficiency or excess of this mineral can have profound effects on a variety of tissues, which include consequences of hypophosphatemia, such as osteomalacia, rickets, red cell dysfunction, rhabdomyolysis, metabolic acidosis and cardiomyopathy, and of hyperphosphatemia, such as soft tissue calcification, hypocalcemia, tetany, and secondary hyperparathyroidism. In many cases, the inter-relationship between phosphate homeostasis and vitamin D metabolism precludes establishing whether the consequences of hypo- or hyperphosphatemia are singularly related to this abnormality or are modified by changes in calcitriol production. Occasionally, however, discrimination between these possibilities has been achieved by evaluation of the therapeutic response to phosphate supplementation or depletion. Such studies indicate that few of the phosphate homeostatic disorders respond adequately to therapeutically induced alterations in phosphate alone, but do regress upon coincident modification of the vitamin D status. The detailed control mechanisms that regulate phosphate homeostasis and vitamin D metabolism in intestine and kidney are reviewed in Chapters 24 and 29 and overall physiology in Chapter 26. Following is a summary of important elements of the phosphate homeostatic schema and the VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

III. Disorders Related to An Altered Phosphate Load References

regulation of vitamin D metabolism that pertain to an understanding of the diseases described in the remainder of this chapter and in Chapter 70.

A. Regulation of Phosphate Homeostasis The kidney is the major arbiter of extracellular phosphate (Pi) homeostasis and plays a key role in bone mineralization and growth. Most of the filtered phosphorus is reabsorbed in the proximal tubule, with approximately 60% of the filtered load reclaimed in the proximal convoluted tubule and 15–20% in the proximal straight tubule [4]. In addition, a small but variable portion (<10%) of filtered phosphorus is reabsorbed in more distal segments of the nephron. Transepithelial phosphorus transport is effectively unidirectional and includes uptake at the brush border membrane of the renal tubule cell, translocation across the cell and efflux at the basolateral membrane [5]. Apical sodium-dependent phosphate (Na/Pi) cotransport across the luminal (brush border) membrane is rate limiting and the target for physiological/pathophysiological alterations. The uptake at this site is mediated by Na+-dependent phosphate transporters that reside in the brush border membrane and depend on the Na+,K+ATPase to maintain the Na+ gradient that drives the transport system [6]. In contrast, basolateral Pi-transport systems are not well defined. Efflux of Pi across the basolateral membrane may involve an anion exchange mechanism and/or a “Pi leak” to complete transcellular reabsorptive flux, and a Na+-dependent Pi uptake mechanism to guarantee Pi uptake from the interstitium if apical influx is insufficient to maintain cellular metabolism. Three distinct and unrelated families of mammalian Na/Pi-cotransporters have been identified, types I, II, and III. All three types are expressed in proximal tubular cells and have the capacity to induce an increase in Na-dependent Pi uptake in heterologous expression Copyright © 2005, Elsevier, Inc. All rights reserved.

1160

MARC K. DREZNER

systems [7]. Although the type I Na/Pi cotransporter is localized in the brush border membrane of proximal tubular cells [8], this transporter does not contribute significantly to regulation of proximal tubular Pi flux. Indeed heterologous expression studies indicate that the type I-mediated Na/Pi cotransport does not have the characteristics and regulatory features of brush border membrane Na/Pi cotransport. Rather, these studies suggest that the type I cotransporter has a channel function, mediating the flux of chloride and organic anions [7,9,10]. In contrast, the type II cotransporter, located predominantly in the apical membrane of the cells in the S1 segments of the proximal tubule in kidneys of young (IIc) and adult animals (IIa) [11], plays a key role in renal Pi handling. Indeed, studies in adult animals support the vital function of the type IIa receptor in several ways: 1) exclusive proximal tubular brush border membrane localization of this receptor [12]; 2) loss of 70–80% of brush border membrane Na/Pi cotransport upon disruption of the gene encoding the type IIa cotransporter [13]; and 3) correlation of type IIa protein abundance in brush border membranes with Na/Pi cotransport activity under a variety of physiologic/ pathophysiological conditions [7]. The gene for the type II cotransporter has been mapped to chromosome 5q35. The type III Na/Pi cotransporters are cell-surface viral receptors [14], and they appear to exhibit ubiquitous renal and extrarenal expression. Although type III mRNA expression is detected in all nephron segments [6], studies to localize the type III protein in apical or

T3

Low Pi acute +

chronic +

+

NPT2 Gene

NPT2 mRNA

−

basolateral cell membranes are lacking. However, the type III Na/Pi cotransporters may be responsible for basolateral Pi influx in all tubular cells to maintain cell metabolism, as well as in proximal tubular cells under conditions of limited apical influx. Proximal tubular Na/Pi cotransport is regulated by a variety of hormones/metabolic factors, which elicit an increase or decrease in Pi reabsorption through alteration of the type IIa cotransporter (NPT2a) protein availability (Fig. 1). Thus, parathyroid hormone (PTH)dependent inhibition of Na+-Pi cotransport depends upon internalization of the cell surface NPT2a protein [15]. Likewise, the acute phase of renal adaptation to phosphate restriction is associated with an increase in NPT2a protein, which is rapidly reversed by a high Pi diet [16]. Not surprisingly, NPT2a protein availability is also central to modulation of the abnormal renal phosphate transport underlying a large number of phosphate homeostatic disorders. These changes in Na/Pi cotransport and NPT2a protein expression occur predominantly in the absence of variations in NPT2a mRNA levels. Thus, rapid increments or decrements in Na/Pi cotransport follow either membrane insertion of NPT2a protein, which may be preceded by de novo synthesis of the protein, or to membrane retrieval of NPT2a protein, followed by lysosomal degradation [7]. Although such membrane trafficking of NPT2a protein is an unequivocally central element in the regulation of Na/Pi cotransport, changes in NPT2a mRNA levels do occur after prolonged treatment with 1,25(OH)2D, PTH or thyroid hormone, following chronic changes in dietary Pi and in disorders of Pi homeostasis, such as

−

Intracellular NPT2 protein

chronic −

acute +

HYP (Gy) EFG, Dexamethasone

FIGURE 1

BBM NPT2 protein

PTH, High Pi

Hormonal and metabolic regulation of NPT2a gene expression and protein production in the kidney. Solid lines represent pathways of regulation that are supported by available data, while dashed lines depict instances where direct evidence for a regulatory mechanism is not available. BBM, brush border membrane; EGF, epidermal growth factor; NPT2a, Na+-dependent phosphate cotransporter; Hyp (Gy), murine homologues of X-linked hypophosphatemic rickets (XLH); T3, triiodothyronine. Adapted from Tenenhouse [4].

CHAPTER 69 Clinical Disorders of Phosphate Homeostasis

X-linked hypophosphatemia (XLH). Thus, a secondary mechanism operates under select conditions to regulate Na/Pi cotransport (Fig. 1). The intracellular signaling mechanisms involved in insertion/retrieval of NPT2a protein remain incompletely understood. However, several studies have identified numerous signals for NPT2a internalization. PTH effects on NPT2a protein are initiated by binding to receptors that activate protein kinase C on apical membranes and protein kinase A and C on basolateral membranes [17]. In contrast, atrial natriuretic protein induced internalization of NPT2a protein is mediated by protein kinase G activation [18]. Recent investigations suggest that these different signaling pathways converge on the extracellular signal-regulated kinase/ mitogen-activated protein kinase (ERK/MAPK) pathway to internalize the NPT2a protein [19]. In concert with these observations, additional reports indicate that FGF-23, a putative phosphatonin, which may contribute to renal Pi wasting in a variety of genetic and acquired human disorders (see below), inhibits NPT2a-mediated Na/Pi cotransport via activation of MAPK [20]. However, the downstream targets for ERK/MAPK remain unknown, as changes in the phosphorylation state of the NPT2a transporter have not been demonstrated in response to ERK/MAPK modulated hormonal/ metabolic regulation of renal Pi transport [21]. In any case, the signaling process regulating internalization/retrieval of NPT2a is insufficient to explain the integrated regulation of Pi homeostasis at the kidney. Rather it appears that protein/protein interactions may also participate in regulation of Pi reabsorption. In this regard, NaPi-Cap 1 and NHERF-1, brush border membrane proteins, may interact with NPT2a via one of multiple PDZ-domains, resulting in release of the protein from an apical scaffold, thereby permitting internalization [22,23]. Clearly, further studies are necessary to determine the interrelated mechanisms regulating the many processes that affect the abundance of NPT2a in brush border membranes.

B. Phosphate Dependent Modulation of Vitamin D Metabolism The production of 1,25(OH)2D, the active metabolite of vitamin D, is under stringent control (see Chapter 5). Indeed, 1α-hydroxylation represents the most important regulatory mechanism in the metabolism of vitamin D [24]. In normal adults, serum 1,25(OH)2D concentrations change little in response to repeated dosing with vitamin D, and remain normal, or even decline, in vitamin D intoxication. Phosphorus is one of the three major factors that regulate the activity of

1161 the enzyme [25]. In this regard, in rats, mice, and humans, phosphate depletion and resultant hypophosphatemia stimulate 25-hydroxyvitamin D-1α-hydroxylase activity and increase the serum 1,25(OH)2D concentration, whereas phosphate loading and consequent hyperphosphatemia inhibit formation of this metabolite. The mechanism whereby phosphorus modulates this adaptive effect remains unknown. Fukase et al. [26] reported that phosphorus may have a direct effect on the kidney and several studies indicate that the effect is, in fact, independent of PTH. In contrast, alternative evidence suggests that phosphorus regulation of 1,25(OH)2D production may depend on growth hormone [27] or insulin-like growth factor I [28]. Regardless of the mechanism, the potential role of phosphorus in regulating vitamin D metabolism is central to understanding and appropriately treating many of the clinical disorders of phosphate homeostasis. As will become evident in the remainder of this chapter, however, the aberrant regulation of vitamin D metabolism encountered in many of these diseases is considered paradoxical. Thus, in virtually all disorders secondary to abnormal renal phosphate transport, hypophosphatemia or hyperphosphatemia is perplexingly associated with decreased or increased serum 1,25(OH)2D levels, respectively. These observations suggest that the effect of phosphorus on 25-hydroxyvitamin D1α-hydroxylase activity may, in fact, occur indirectly and secondary to alterations in renal phosphate transport systems. In this regard, phosphate depletion and hypophosphatemia and phosphate loading and hyperphosphatemia are associated with compensatory changes in renal phosphate transport that may mediate changes in 1,25(OH)2D production. In accord, the prevailing serum calcitriol levels in normals and patients with disorders of renal phosphate transport, XLH, tumorinduced osteomalacia (TIO) and tumoral calcinosis (TC), display a highly significant positive correlation with the renal TmP/GFR [29], supporting the possibility that renal tubular reabsorption of phosphate may be a major determinant of renal 1,25(OH)2D production (Fig. 2). Of course, these data do not establish that alterations in 25-hydroxyvitamin D-1α-hydroxylase activity, in response to phosphate depletion or loading, are dependent upon renal phosphate transport. However, the recent observations that phosphate depletion does not increase 1,25(OH)2D production in mice subjected to treatment with phosphonoformic acid, which precludes a compensatory alteration in TmP/GFR, suggests an important role for the renal phosphate transport system in modulating phosphorus mediated effects on 1α-hydroxylase activity under all conditions (Fig. 3). Additionally, preliminary studies have documented that mice with targeted disruption of the NHERF-1

1162

MARC K. DREZNER

84

Plasma 1,25(OH)2 (pg/ml)

72 60 XLH TIO Normals Tumoral calcinosis

48

While these data indicate that the renal production of 1,25(OH)2D and the renal tubular reabsorption of phosphate are linked, the precise mechanism(s) underlying this association remain uncertain. However, it is clear that understanding the pathophysiology of, and defining appropriate treatment regimens for, many clinical disorders of phosphate homeostasis require investigation of the vitamin D regulatory system.

36 r = 0.85 p < 0.001

24

II. DISORDERS OF PHOSPHATE HOMEOSTASIS

12

0

1

2

3

4 7 5 6 TmP/GFR (mg/dl)

8

9

10

FIGURE 2 Correlation between renal TmP/GFR and plasma 1,25(OH)2D levels in normals and patients with XLH, TIO, and tumoral calcinosis. Significant linear correlation is evident, suggesting a potential relationship between renal phosphate transport and the prevailing plasma levels of active vitamin D. TIO, tumorinduced osteomalacia. From Drezner [29].

gene, which singularly promotes internalization of proximal tubule Npt2a, manifest ~50% decreased apical membrane localized Npt2a and consequent renal P wasting and hypophosphatemia [22], and paradoxically exhibit inappropriately normal serum 1,25(OH)2D levels, similar to hyp-mice with comparable Npt2a deficiency. In contrast, Tenenhouse et al. [30] concluded from studies in the Npt2a−/−-mouse that Npt2a and renal P transport do not influence 25(OH)D-1α-hydroxylase activity. However, these data may have limited applicability to understanding the role of renal phosphate transport on vitamin D metabolism, since it is well known that compensatory developmental changes often confound studies in knockout mice with complete absence of a protein function. In this regard, several investigators [11,31] recently reported that the sodium-dependent P cotransporter, Npt2c, normally expressed in murine kidneys of young animals, has sustained activity in adult Npt2a−/−-mice, which is likely responsible for the “inappropriate” residual renal P transport in these mutants and the resultant limited hypophosphatemia. The absence of similar Npt2c expression in the kidneys of adult normal and NHERF knockout mice [32] substantiates that enhanced Npt2c expression in adult Npt2a−/−-mice may be a unique compensatory mechanism, which limits the applicability of studies in this model to conclusions regarding regulation of vitamin D metabolism.

The variety of diseases, therapeutic agents, and physiological states that affect phosphate homeostasis are numerous and reflect a diverse pathophysiology. Indeed, rational choice of an appropriate treatment for many of these disorders depends on determining the precise cause for the abnormality. In general, defects in phosphate homeostasis result from impaired renal tubular phosphate reabsorption and a consequent change in the TmP/GFR, an altered phosphate load due to varied intake, or abnormal gastrointestinal absorption or translocation of phosphorus between the extracellular fluid and tissues (Table I). The disorders of renal phosphate transport are the most common of these diseases and have been intensively studied. Indeed, investigation of these conditions has provided new insight into the reciprocal regulation of phosphate homeostasis and vitamin D metabolism. In the remainder of this chapter and in Chapter 70, several clinical states that represent disorders of phosphate homeostasis will be discussed and the potential role of the vitamin D endocrine system in their pathogenesis and phenotypic presentation highlighted.

A. Disorders of Renal Phosphate Transport: Hypophosphatemic Diseases The disorders of renal phosphate transport, which lead to phosphate wasting and hypophosphatemia, are by far the most common disturbances of phosphate homeostasis. However, the pathophysiological basis of these disorders has remained elusive, largely because the hormonal/metabolic control of phosphate homeostasis at the kidneys and in bone is not completely understood. In this regard, the function of the PTH-vitamin D axis is not sufficient to explain the physiological complexity of systemic phosphate homeostasis. Nevertheless, the PTH-vitamin D independent mechanism(s) by which phosphate excess or phosphate depletion influence the net rate of proximal tubule phosphate reabsorption and bone mineralization remains unknown. Recently, significant new information has emerged from studies of the hypophosphatemic disorders, Tumor-induced

1163

CHAPTER 69 Clinical Disorders of Phosphate Homeostasis

A

Renal 25(OH)D-1-hydroxylase activity

10

mg/dl

8

*

*

6 4 2 0

Control Dietary Control Renal −P −P

1,25-Dihydroxyvitamin D produced (fmoles/mg kidney/min)

Serum phosphorus concentration

*

14 12 10 8 6 4 2 0

Control Dietary Control Renal −P −P

*Significantly different from control at p = 0.05

B

Serum phosphorus concentration

Renal 25(OH)D-1-hydroxylase activity

10

mg/dl

**

**

6 4 2

0 0

+ 0

+ +

**Significantly different from respective control (0,0) at p = 0.01

1,25 OH2d Produced (fmoles/mg kidney/min)

8

0 Dietary- P PFA- P

**

12 10 8 6 4 2

0 Dietary- P PFA- P

0 0

+ 0

+ +

PFA, Phosphonofomic acid -P, Phosphate depletion

FIGURE 3 A. While hypophosphatemia resulting from phosphate restriction increases renal 25(OH)D-1α-hydroxylase activity, hypophosphatemia of similar magnitude secondary to phosphonoformic acid (PFA) mediated renal phosphate wasting has no effect on enzyme function. These data suggest that hypophosphatemia is not the proximal trigger mechanism for 1,25(OH)2D production. B. In similar experiments the effects of phosphate depletion and consequent hypophosphatemia on renal 25(OH)D-1α-hydroxylase activity are blocked by the coincident administration of phosphonoformic acid, which prevents the expected compensatory change in renal phosphate flux attendant upon the depletion. These observations reaffirm the potentially important role of renal phosphate transport in modulating the effects of phosphate on for 1,25(OH)2D production.

osteomalacia (TIO), X-linked hypophosphatemic rickets/osteomalacia (XLH) and Autosomal dominant hypophosphatemic rickets (ADHR), which brings new light to the mechanisms potentially regulating phosphate homeostasis and bone mineralization. These investigations have putatively identified circulating factors that alter renal tubular phosphate reabsorption and bone mineralization. Indeed, this work has lent credence to the possibility that a hormone or family of hormones, called phosphatonin(s) and/or minhibin(s) exist, which regulate renal phosphate transport and bone mineralization and

participate in the pathophysiological cascade of events underlying many of the renal phosphate wasting disorders. In fact, current theories have emerged, which suggest that a common metabolic pathway underlies many of these hypophosphatemic diseases. 1. THE COMMON PATHWAY UNDERLYING THE PATHOGENESIS OF TIO, XLH, AND ADHR

As detailed below and in Chapter 70, XLH and ADHR are disorders characterized by hypophosphatemia, due to impaired renal tubular reabsorption

1164

MARC K. DREZNER

TABLE I Disorders of Phosphate Homeostasis Abnormal Renal Phosphate Transport Hypophosphatemic syndromes Genetic Diseases X-Linked hypophosphatemia (XLH) Hereditary hypophosphatemic rickets with hypercalciuria (HHRH) Tumor-induced osteomalacia (TIO) Autosomal dominant hypophosphatemic rickets (ADHR) Autosomal recessive hypophosphatemic rickets (ARHR) McCune Albright Syndrome (MAS) Fanconi syndrome, type I (FS I) Familial Idiopathic Cystinosis (Lignac-Fanconi Disease) Oculocerebrorenal (Lowe) syndrome Glycogen storage disease Wilson disease Galactosemia Tyrosinemia Hereditary Fructose Intolerance Neurofibromatosis Linear nevus sebaceous syndrome Fanconi syndrome, type II (FS II) Acquired Disorders Tumor-Induced Osteomalacia Mesenchymal, epidermal and endodermal tumors Light chain nephropathy Renal transplantation

of phosphate, inappropriately normal or decreased production and serum levels of 1,25(OH)2D, and defective cartilage and bone mineralization. In contrast, TIO and McCune-Albright syndrome (MAS) (caused by activating mutations of Gsα) are hypophosphatemic disorders caused by production of a factor(s) by tumors or fibrous dysplastic bone cells, respectively, which leads to phenotypic features similar to those of the hereditary phosphate wasting diseases. Based on the shared phenotype of these disorders, several groups have postulated the existence of a unique circulating protein(s), common to each of these diseases, which inhibits sodium-dependent phosphate reabsorption by the renal proximal tubule by PTH distinct mechanisms, impairs bone and cartilage mineralization, and counters hypophosphatemia-mediated increments in the renal production of 1,25(OH)2D. Initial evidence lending credence to the possibility that there is such a common metabolic pathway underlying these diseases derived from studies of patients

Multiple Myeloma Cadmium intoxication Lead intoxication Tetracycline (outdated) administration Hyperphosphatemic syndromes Tumoral calcinosis Altered Phosphate Load Hypophosphatemic syndromes Decreased Phosphate Availability Phosphate deprivation Gastrointestinal malabsorption Transcellular Shift of Phosphate Alkalosis Glucose Administration Combined Mechanisms Alcoholism Burns Nutritional recovery syndrome Diabetic ketoacidosis Hyperphosphatemic syndromes Vitamin D intoxication Rhabdomyolysis Cytotoxic therapy Malignant hyperthermia

with TIO. By culturing a tumor associated with this disease, Cai et al. [33] demonstrated that supernatants of such tumor cells maintained in culture contained a factor(s) that inhibited sodium-dependent phosphate reabsorption in renal cultured epithelia. The effects of this factor(s) were specific, cyclic AMP independent and distinct from those of PTH. These data, as well as subsequent reports from other groups [34–36], indicated the existence of a novel substance, putatively named phosphatonin [37], that alters phosphate reabsorption in the renal tubule. Further investigation, utilizing serial analysis of gene expression and array profiling to identify overexpressed genes in tumors from patients with TIO, confirmed enhanced production of several genes, including those that encode fibroblast growth factor (FGF) 23, secreted frizzlerelated protein (sFRP)-4, and matrix extracellular phosphoglycoprotein (MEPE) [38–42]. The possibility that these observations linked the pathogenesis of TIO to that of ADHR was suggested

CHAPTER 69 Clinical Disorders of Phosphate Homeostasis

by the seminal discovery that FGF 23 has a pathogenetic role in patients with ADHR [43,44]. Patients with this syndrome have missense mutations in the FGF 23 gene at the 176-RXXR-179 site (R176Q, R179W, and R179Q), rendering the FGF 23 protein resistant to cleavage/hydrolysis and inactivation by proteolytic enzymes [43,45]. The resultant increase in circulating levels of cleavage-resistant FGF 23 was presumed responsible for phosphaturia and abnormal bone mineralization due to direct actions of this hormone on the renal proximal tubule and the osteoblasts. Since tumors in patients with TIO overexpress FGF 23 and mutations in the FGF 23 gene underlie ADHR, the possibility that FGF 23 is the presumed phosphatonin operative in these diseases seemed plausible. Indeed, this possibility was supported by a series of observations, which established the biological activity of FGF 23, an approximately 26-kDA circulating protein, consisting of an N-terminal FGF homology domain and a novel 71-amino acid C-terminus of uncertain function. In this regard, Shimada et al. [41] reported that administration of biosynthetically prepared full-length FGF 23 to mice resulted in phosphaturia but did not lead to upregulation of 1,25(OH)2D production. Bai et al. [46] also discovered that intact FGF 23 has enhanced in vivo biological potency, and tumors derived from cells that overexpressed FGF 23 caused a more severe osteomalacia and rickets in nude mice than observed due to hypophosphatemia alone, suggesting a direct effect of FGF 23 on cartilage and bone. In concert, Bowe et al. [39] found that FGF 23 inhibited sodium-dependent phosphate transport in cultured renal epithelia by decreasing the expression of the NPT2a mRNA and protein. Conversely, FGF 23-null mice exhibit elevated serum phosphorus levels and increased 1,25(OH)2D production, confirming an essential and nonredundant role of FGF 23 in regulation of phosphate homeostasis [47], while FGF 23 transgenic mice exhibit a phenotype consistent with a severe form of ADHR [48]. In addition, several groups have found that circulating FGF 23 levels are elevated in the majority of patients with TIO and those with ADHR, as well as those with MAS, who manifest hypophosphatemia [49–51]. Studies of XLH have further reinforced the possibility that FGF 23 is the phosphatonin, providing a common link to the pathogenesis of the phosphate wasting diseases. Early studies by Meyer et al. [52], testing parabiosis between hyp- and normal mice, suggested the presence of a circulating factor that causes hypophosphatemia in the mouse model of XLH. In concert, Lajeunesse et al. [53] discovered the presence of a serum factor in hyp-mice that inhibits phosphate transport in renal epithelia, and others reported that a

1165 similar factor is elaborated by hyp-mouse osteoblasts, which not only inhibits renal phosphate transport but impairs osteoblast mineralization [54,55]. Further, renal cross-transplantation studies confirmed the presence of a phosphate transport inhibitory factor in hyp-mice [56]. With the discovery that mutations of the PHEX gene underlie XLH (see below), several observations suggested the likely interrelated events linking the HYP phenotype to FGF 23. Most notably, the absence of PHEX in the kidneys of hyp-mice [57–63] indicated that the gene mutation must indirectly regulate the expression of NPT2/Npt2 in renal tubular cells. In addition, recognition that PHEX is one of six members of the M13 family of zinc-dependent type II cell surface membrane metalloproteinases [64] suggested that PHEX/Phex most likely inactivates a novel phosphaturic/minhibin hormone. Thus, the most plausible pathophysiological basis for XLH is: 1) an inactivating mutation of PHEX produces inadequate amounts of the PHEX endopeptidase; 2) a resultant ineffective or inadequate degradation/ inactivation of phosphatonin/minhibin occurs, causing elevated circulating levels of bioactive protein; and 3) a consequent repressed expression of NPT2 manifests, resulting in renal P wasting and hypophosphatemia, and unknown hormone provoked changes in the osteoblast retard mineralization. This formulation has been supported, and the role of FGF 23 as the phosphatonin apparently confirmed by the observations that: 1) FGF 23 is possibly a substrate for PHEX [39]; 2) circulating FGF 23 levels are elevated in hyp-mice [65] and in a subset of patients with XLH [49,50]; 3) and antibody mediated neutralization of FGF 23 ameliorates hypophosphatemia and rickets in hyp-mice [65]. Collectively, the aforementioned observations have formed the basis for an enzyme/substrate model for the common pathogenesis of XLH, TIO, and ADHR. According to this model, FGF 23 is phosphatonin(/minhibin), the hormone that inhibits sodiumdependent phosphate uptake in the renal proximal tubule and impairs bone mineralization. This model presumes that only full-length FGF 23 is phosphaturic and impairs mineralization, and it postulates that the cell surface enzyme PHEX degrades FGF 23 into inactive fragments. Further, the model predicts that FGF 23 is increased in ADHR because of mutations in FGF 23 that render it resistant to PHEX-dependent cleavage, in XLH because inactivating mutations of PHEX prevent the normal degradation of FGF 23, and in TIO because overproduction of FGF 23 overwhelms the inactivation capacity of degradative mechanisms. Although this model has support, especially for ADHR and possibly TIO, there are a number of inconsistencies and unexplained observations, which raise concerns about whether this simple PHEX-FGF 23 hypothesis

1166 is correct. First and foremost, several investigators have identified other potential phosphatonins/minhibins, which may be operative in TIO and XLH. Second, it has not been established that FGF 23 is unequivocally a substrate for PHEX. Third, the biololgical activites of FGF 23, which have been documented, are incomplete and, more importantly, do not explain the panorama of phenotypic abnormalities, common to the hypophosphatemic diseases. a. The Family of Phosphatonins/Minhibins The most compelling observation, which challenges the PHEX-FGF 23 hypothesis, is the existence of additional phosphatonin/minhibin-like molecules. Thus, while it is tempting to speculate that TIO is due solely to excessive production of FGF 23, this is not likely the case since tumors overexpress other molecules, including sFRP-4 and MEPE [42] that exhibit the characteristics of a phosphatonin/minhibin. In this regard, although the previously conceived information about secreted frizzle related proteins, including their localization at the plasma membrane and/or in the ECM and their dependence on Wnt-dependent signaling, created an initial bias that sFRP-4 is an unlikely candidate for a phosphatonin, recent investigations by Berndt et al. [66] have required re-evaluation of this preconception. Indeed, they have provided compelling evidence that sFRP-4 has unmistakable characteristics of a phosphatonin. Thus, recombinant sFRP-4 inhibits sodiumdependent phosphate transport in cultured opossum renal epithelial cells, indicating a possible direct action of sFRP-4 on proximal tubular phosphate transport. Further, the systemic administration of recombinant sFRP-4 caused phosphaturia (and hypophosphatemia) in normal rats without stimulating renal 25-hydroxyvitamin D-1α-hydroxylase activity. Moreover, the effects on renal epithelia occurred through a mechanism that involved antagonism of Wnt-dependent β-catenin pathways. And finally, Berndt et al. [66] extended these observations by showing that sFRP-4 is present in normal human serum and in the serum of patients with TIO. Similar experiments have established MEPE as a phosphatonin or more likely a minhibin. In this regard, extensive experiments have documented that MEPE is exclusively expressed in osteoblasts and osteocytes in rodents [67–72] and in abundance in human bone, as well as in human brain, albeit in lesser amounts [67,70–72]. Moreover, MEPE-expression occurs in all tumors from patients with TIO and is notably absent in nonphosphaturic tumors [38,39,41,67]. Further, Rowe et al. [73] have demonstrated that MEPE inhibits phosphate transport in renal epithelia in vitro and when administered in vivo to rats results in significant dose dependent phosphaturia and hypophosphatemia. However, a recent

MARC K. DREZNER

report of the MEPE null-mutant mouse phenotype showed that the MEPE-deficient mice did not have abnormalities of serum phosphorus, as might be expected if MEPE played an important role in phosphate homeostasis [72]. Moreover, Liu et al. [74] reported that transfer of MEPE deficiency onto the hyp-mouse background failed to rescue the hypophosphatemia in the mutants, suggesting that MEPE may not cause phosphaturia in the setting of inactivating Phex mutations. Thus, it is possible that MEPE may serve only as a minhibin and not a phosphatonin. In accord with this possibility, MEPE-deficient mice exhibit increased bone formation and mineralization, consistent with the possibility that MEPE, under physiological conditions, plays an inhibitory role in bone mineralization and formation. Moreover, Rowe et al. [73] documented that MEPE dose-dependently inhibited BMP2 mediated mineralization of a murine osteoblast cell-line (2T3) in vitro. This activity was localized to a cathepsin-B released carboxy-terminal MEPE-peptide, in accord with the multifunctional nature of MEPE, which harbors distinct domains. Further, MEPE transcripts are increased in poorly mineralizing bone derived from hyp-mice [70], consistent with a role for MEPE in the mineralization defect in XLH. In addition, the targeted overexpression of PHEX to osteoblasts is associated with reductions in MEPE expression and normalization of osteoblastmediated mineralization ex vivo [69]. Finally, consistent with the enzyme/substrate model for the common pathogenesis of the hypophosphatemic disorders, studies suggest that Phex may modify the hydrolysis of MEPE by inhibiting cathepsin B-dependent cleavage of the protein, which may inplicate MEPE in the local regulation of mineralization through Phex-dependent mechanisms [68]. b. The Role of FGF 23 as a PHEX Substrate A variety of additional observations likewise challenge the PHEX-FGF 23 hypothesis. Most notably, recent investigations provide compelling evidence that FGF 23 is not a substrate for PHEX. The initial studies of Bowe et al. [39], which identified FGF 23 as a substrate for PHEX, provided fundamental justification of the enzyme/substrate model. However, documentation that recombinant PHEX may cleave FGF 23 at the RXXR motif or a nearby site has not been confirmed by other investigations [75,76]. Indeed, additional studies suggest that the initial report of PHEX-dependent metabolism of FGF 23 may have represented degradation secondary to furin-like proteolytic enzymes, contaminating the reticulolysate preparation used [40,41], an hypothesis confirmed by the observed degradation of recombinant FGF 23 after incubation with reticulolysates alone, as well as with reticulolysates expressing either the inactive 3′ truncated Phex or the full-length wild-type Phex [77].

1167

CHAPTER 69 Clinical Disorders of Phosphate Homeostasis

Moreover, the inability to demonstrate Phex-dependent cleavage of intact FGF 23 or its N- and C-terminal fragments using active recombinant Phex and FGF 23 proteins expressed in mammalian cells or proteins synthesized in vitro confirmed that FGF 23 is not a substrate for Phex [77]. Such failure to demonstrate that FGF 23 is a substrate for Phex occurred despite the observations that full-length FGF 23 contains putative cleavage sites for Phex hydrolysis [76]. Thus, the mere presence of consensus cleavage sites for an enzyme in a protein does not necessarily indicate that such a site is available to the enzyme. Indeed, the inability of Phex to hydrolyze intact FGF 23 may indicate a constraint placed by incompatible three-dimensional structure that limits enzyme substrate interactions. Regardless, the available data do not substantiate the enzyme/ substrate model of common pathogenesis for the hypophosphatemic disorders. In this regard, although the genetic mutation underlying ADHR undoubtedly limits proteolytic degradation of FGF 23, thereby increasing the circulating concentration of a biologically active molecule, the enzyme normally responsible for the degradation remains questionable. In addition, the failure of Phex to degrade FGF 23 discredits the hypothesis that failure of mutated Phex to degrade FGF 23 accounts for the increase of this phosphaturic hormone in XLH. Nevertheless, the circulating concentration of FGF 23 is increased in hyp-mice and a subset of patients with XLH, suggesting that an alternative mechanism enhances hormone production or impairs hormone degradation in this disease state. In this regard, Liu et al. [77] recently reported that inactivating mutations of Phex resulted in increased expression of FGF 23 transcripts in the bone and cultured osteoblasts of the hyp-mouse, indicating that Phex may somehow regulate FGF 23 biosynthesis. Indeed, in hyp-mice lacking a functional Phex increased levels of fgf 23 transcripts were observed in mandible, calvaria, and diaphysis of the long bone. Moreover, cultured osteoblasts from hyp-mice likewise exhibited increased fgf 23 transcripts. In concert with these observations, recent investigations in patients with the McCune-Albright syndrome have identified FGF 23 production by fibrous displasia osteoprogenitors and normal bone forming cells in vitro [51,78,79]. While these data suggest that bone is the source of increased FGF 23 in XLH, several observations challenge this interpretation. First, the FGF 23 produced by the bone in hyp-mice, according to Liu et al. [77], is of relatively small magnitude and may have no physiological effect. Second, other studies have failed to confirm the presence of FGF 23 in normal bone and/or osteoblasts [80]. Third, no data exist providing a mechanism by which inactive Phex

results in increased FGF 23 production. Fourth, the presence of increased protease activity in the hyp-mouse bone matrix, secondary to inactivated Phex [69], raises significant doubt that osteoblast-secreted full-length FGF 23 would reach the circulation. And finally, transgenic overexpression of Phex in the hyp-mouse, which should limit the increased production of fgf 23 transcripts and the resultant circulating levels of this protein, according to the hypothesis advanced by Liu et al. [81], fails to normalize renal phosphate transport. As a consequence, it appears that the mechanism underlying the purported elevation of FGF 23 levels in XLH remains controversial, precluding progress towards ascertainment of a common pathogenesis for the hypophosphatemic disorders. c. The Biological Activities of the Phosphatonins/ Minhibins As related above, multiple studies have linked the biological activities of the putative phosphatonins/minhibins to the phenotype of the hypophosphatemic diseases. However, in each case, the testing is incomplete or the documented biological activity does not explain a phenotypic characteristic manifest as part of the ADHR, TIO, or XLH syndrome. For example, as shown in Table II, studies do not yet exist that demonstrate the in vitro inhibitory activity of either FGF 23 or sFRP-4 on osteoblast mineralization. TABLE II Activity of Candidate Phosphatonin/ Minhibin Proteins FGF 23 Bone Inhibits mineralization In vitro In vivo Kidney Inhibits Na+-Pi transport In vitro In vivo Decreases Npt2a transcription In vitro In vivo Vitamin D metabolism Alters 25(OH)D-1α-hydroxylase mRNA In vitro In vivo Alters 25(OH)D-1α-hydroxylase protein In vitro In vivo

sFRP4

MEPE

? +

? ?

+ ?

+ +

+ +

+ +

+ +

+ +

– –

⇓ ⇓

– –

– ⇑

? ?

? ?

? ?

1168 Yet, studies in hyp-mice and/or transgenic FGF 23 mice clearly indicate that the mineralization defect in the hypophosphatemic disorders exceeds that expected from hypophosphatemia alone and does not normalize with phosphate therapy. Hence, an effect of minhibin on osteoblast mineralization function seems certain. Interestingly, such an in vitro effect has been documented for MEPE but no in vivo data are available yet. In contrast, testing of the putative phosphatonins on renal sodium-dependent phosphate transport is relatively complete and supports the possible role of each of these compounds in the pathogenesis of the hypophosphatemic diseases. However, as noted previously, various in vivo animal models do not uniformly substantiate such a role for either MEPE or FGF 23. Moreover, presumed restoration of normal FGF 23 activity in hyp-mice by a variety of techniques fails to normalize renal phosphate transport and hypophosphatemia [65,69,81,82]. Interestingly, the anticipated effects of the phosphatonins/minhibins on vitamin D metabolism are controversial. Although phosphate mediated enhancement of calcitriol production and serum levels is uniformly inhibited in ADHR, TIO, and XLH, the cause of this abnormality in each disorder is less certain. Since regulation of calcitriol production occurs primarily at the transcription level, several groups have anticipated that a phosphatonin/minhibin would inhibit renal 25(OH)D1α-hydroxylase transcription. In accord, current studies indicate that FGF 23 fulfills this criterion. However, investigations in the hyp-mouse indicate that renal 25(OH)D-1α-hydroxylase mRNA is, in fact, elevated under basal conditions and following PTH stimulation [83]. Therefore, the inhibition of calcitriol production occurs at the translational level [84]. As shown in Table II, available data have not linked any of the phosphatonins/minhibins to this phenotypic characteristic of XLH. These observations clearly indicate that further studies are essential in order to identify which of the putative phosphatonin/minhibin proteins have biological properties consistent with those anticipated in the various hypophosphatemic diseases. Although a common pathogenetic mechanism for these diseases has not been unequivocally defined, progress in our understanding of the various disorders of phosphate homeostasis, which have been identified, is truly remarkable. Consideration of these diseases in the remainder of this chapter and in Chapter 70 will highlight many of these advances. 2. X-LINKED HYPOPHOSPHATEMIA (XLH)

XLH is the prototypic renal phosphate wasting disorder, characterized in general by progressively severe skeletal abnormalities and growth retardation.

MARC K. DREZNER

The syndrome occurs as an X-linked dominant disorder with complete penetrance of a renal tubular abnormality resulting in phosphate wasting and consequent hypophosphatemia (Table III). The clinical expression of the disease is widely variable, ranging from a mild abnormality, the apparent isolated occurrence of hypophosphatemia, to severe rickets and/or osteomalacia [85]. In children, the most common clinically evident manifestations include short stature and limb deformities. This height deficiency is more evident in the lower extremities, since they represent the fastest growing segment before puberty. In contrast, upper segment growth is generally less affected. The majority of children with the disease exhibit enlargement of the wrists and/or knees secondary to rickets, as well as bowing of the lower extremities. Additional signs of the disease may include late dentition, tooth abscesses secondary to poor mineralization of the interglobular dentine, and premature cranial synostosis. Many of these features do not become apparent until age 6 to 12 months or older [86]. In spite of marked variability in the clinical presentation, bone biopsies in affected children and adults invariably reveal low turnover osteomalacia without osteopenia. The severity of the bone disorder has no relationship to sex, the extent of the biochemical abnormalities, or the degree of the clinical disability. In untreated youths and adults, the serum 25(OH)D levels are normal and the concentration of 1,25(OH)2D is in the low-normal range [87–89]. The paradoxical occurrence of hypophosphatemia and normal serum calcitriol levels is due to aberrant regulation of renal 25(OH)D-1α-hydroxylase activity. Studies in hyp- and gy-mice, the murine homologues of the human disease, have established that defective regulation is confined to the enzyme localized in the proximal convoluted tubule, the site of the abnormal phosphate transport [90–93]. a. Pathophysiology Investigators generally agree that the primary inborn error in XLH results in an expressed abnormality of the renal proximal tubule that impairs Pi reabsorption. This defect has been indirectly identified in affected patients and directly demonstrated in the brush border membranes of the proximal nephron in hyp-mice. Until recently, whether this renal abnormality is primary or secondary to the elaboration of a humoral factor has been controversial. In this regard, demonstration that renal tubule cells from hyp-mice maintained in primary culture exhibit a persistent defect in renal Pi transport [94,95], likely due to decreased expression of NPT2a mRNA and immunoreactive protein [96–98], supported the presence of a primary renal abnormality. In contrast, transfer of the defect in renal Pi transport to normal and/or parathyroidectomized normal mice parabiosed to hyp-mice implicated a humoral factor in the pathogenesis of the

1169

CHAPTER 69 Clinical Disorders of Phosphate Homeostasis

TABLE III. Biochemical/Genetic Characteristics of the Prototypic Phosphopenic Disorders in Man XLH

TIO

ADHR

ARHR

HHRH

FS

TC

I

II

P Homeostasis Serum P Renal TmP/GFR GI P Absorption FGF-23

⇓ ⇓ ⇓ N or ⇑

⇓ ⇓ ⇓ ⇑

⇓ ⇓ ⇓ ⇑

⇓ ⇓ ⇑ ?

⇓ ⇓ ⇑ ?

⇓ ⇓ ⇓ ?

⇓ ⇓ ⇑ ?

⇑ ⇑ ⇑ ?

Ca Homeostasis Serum Ca Urine Ca Nephrolithiasis GI Ca Absorption Serum PTH

N ⇓ – ⇓ N

N ⇓ – ⇓ N

N ⇓ – ⇓ N

N ⇑ + ⇑ N

N ⇑ – ⇑ N

N ⇓ – ⇑ N

N ⇑ – ⇑ N

N ⇑ – ⇑ N

Vitamin D Metabolism 25(OH)D 1,25(OH)2D

N N/⇓

N ⇓

N N/⇓

N ⇑

N ⇑

N N/⇓

N ⇑

N ⇑

Bone Metabolism Serum Alk Phos Serum NPT

N/⇑ N

N/⇑ N

N/⇑ N

N/⇑ N

N/⇑ N

N/⇑ N

N/⇑ N

N ?

+ X-linked dominant PHEX

– –

+ Autosomal Recessive ?

Variable Variable

+

+

? ?

Variable ?

Genetics Familial Transmission Abnormal Gene

–

+ Autosomal dominant FGF-23

+ Autosomal Recessive CLCN5

Variable

XLH, X–linked hypophosphatemia; TIO, tumor-induced osteomalacia; HHRH, hereditary hypophosphatemic rickets with hypercalciuria; FS I, Fanconi syndrome, type I; FS II, Fanconi syndrome, type II; ADHR, autosomal dominant hypophosphatemic rickets; ARHR, autosomal recessive hypophosphatemic rickets; TC, tumoral calcinosis. N, normal; ⇓, decreased; ⇑, increased. Modified from Econs et al. [202].

disease [99,100]. Current studies, however, have provided compelling evidence that the defect in renal Pi transport in XLH is secondary to the effects of a circulating hormone or metabolic factor. In this regard, immortalized cell cultures from the renal tubules of hyp- and gy-mice exhibit normal Na+-phosphate transport, suggesting that the paradoxical effects observed in primary cultures may represent the effects of impressed memory and not an intrinsic abnormality [101,102]. Moreover, the report that cross-transplantation of kidneys in normal and hyp-mice results in neither transfer of the mutant phenotype or its correction unequivocally established the humoral basis for XLH [56]. Subsequent efforts, which resulted in localization of the gene encoding the Na+-phosphate co-transporter

to chromosome 5, further substantiated the conclusion that the renal defect in brush-border membrane phosphate transport is not intrinsic to the kidney in XLH [103]. While these data establish the presence of a humoral abnormality in XLH, the identity of the putative factor and the spectrum of its activity have not been definitively elucidated. Nevertheless, several investigators have identified and characterized the biological activities of a variety of phosphaturic factors (inhibitors of Na+-dependent phosphate transport) and mineralization inhibitory factors, which may play a role in the pathogenesis of XLH (see above). Moreover, several reports have documented production of phosphaturic and mineralization inhibitory factors by hyp-mouse osteoblasts maintained in culture [55,67,69,77,101].

1170 Therefore, as noted above, these studies argue that a circulating factor(s), phosphatonin(s)/minhibin(s), plays an important role in the pathophysiological cascade responsible for X-linked hypophosphatemia. b. Genetic defect Efforts to better understand XLH have led to identification of the genetic defect underlying this disease. In 1986 Read et al. [104] and Machler et al. [105] reported linkage of the DNA probes DXS41 and DXS43, which had been previously mapped to Xp22.31-p21.3, to the HYP gene locus. In subsequent studies Thakker et al. [106,107] reported linkage to the HYP locus of additional polymorphic DNA, DXS197, and DXS207 and, using multipoint mapping techniques, determined the most likely order of the markers as Xpter-DXS85-(DXS43/DXS197)-HYP-DXS41-Xcen and Xpter-DXS43-HYP-(DXS207/DXS41)-Xcen, respectively. The relatively small number of informative pedigrees available for these studies prevented definitive determination of the genetic map along the Xp22-p21 region of the X-chromosome and only allowed identification of flanking markers for the HYP locus 20 centimorgans (cM) apart. More recently, the independent and collaborative efforts of the HYP consortium resulted in the study of 13 multigenerational pedigrees and consequent refined mapping of the Xp22.1-p21 region of the X chromosome, identification of tightly linked flanking markers for the HYP locus, construction of a YAC contig spanning the HYP gene region, and eventual cloning and identification of the disease gene as PHEX, a Phosphate regulating gene with homologies to Endopeptidases located on the X-chromosome. In brief, these studies ascertained a locus order on Xp22.1 of: Xcen-DXS451-(DXS41/DXS92)-DXS274DXS1052-DXS1683-PHYP-DXS7474-DXS365(DXS443/DXS3424)-DXS257-(GLR/DXS43)-DXS3 15-Xtel Moreover, the physical distance between the flanking markers, DXS1683 and DXS7474, was determined as 350kb and their location on a single YAC ascertained. Subsequently, a cosmid contig spanning the HYP gene region was constructed and efforts directed at discovery of deletions within the HYP region. Identification of several such deletions permitted characterization of cDNA clones that mapped to cosmid fragments in the vicinity of the deletions. Database searches with these cDNAs detected homologies at the peptide level to a family of endopeptidase genes that includes neutral endopeptidase (NEP), endothelin-converting enzyme-1 (ECE-1), and the Kell antigen. These efforts clearly established PHEX as the candidate gene responsible for XLH [64,108–112]. Identification of the gene associated with XLH as PHEX [64] has facilitated efforts to better understand this disease. The gene codes for a 749-amino acid

MARC K. DREZNER

protein, consisting of three domains: 1) a small aminoterminal intracellular tail; 2) a single, short transmembrane peptide; and 3) a large carboxyterminal extracellular peptide, which, typical of zinc metalloproteases [113], has 10 conserved cysteine residues and a HEXXH pentapeptide motif. The homology of PHEX with metalloproteases resulted in inclusion of this protein in the M13 family of membrane-bound metalloproteases, also known as neutral endopeptidases [114–116]. M13 family members, including neutral endopeptidase 24.11 (NEP), endothelin-converting enzymes 1 and 2, the Kell blood group antigen, neprilysin-like peptide (NL1), and endothelin converting enzyme-like 1 [113,115,117–123], degrade or activate a variety of peptide hormones. Preservation in the PHEX structure of catalytic glutamate and histidine residues (equivalent to Glu648 and His711 of NL1) argues for similar protease activity, as does alignment of PHEX mutations with regions required for peptidase activity in NL1 [124]. Further, like other neutral endopeptidases, immunofluorescent studies reveal a cell-surface location for PHEX in an orientation consistent with a type II integral membrane glycoprotein [124]. In any case, cloning the PHEX gene led relatively rapidly to cloning the homologous murine Phex gene and identification of the mutations in the murine homologues of XLH, the hyp- and gy-mice [58,125,126]. Unlike 97% of known genes, neither the human nor murine gene has a Kozak sequence, a purine at the –3 position before the ATG initiation sequence [125–127]. Since such genes are often post-transcriptionally regulated, this anomaly may impact understanding the hormonal and metabolic regulation of PHEX/Phex. Many investigators [57–64,127–135] have used Phex/ Phex localization and mutation detection to help formulate the pathogenetic scheme for XLH discussed above. Investigation of murine tissues and cell cultures revealed that Phex is predominantly expressed in bones and teeth [57,58,60–62,126], while mRNA, protein or both have also been found in lung, brain, muscle, gonads, skin, and parathyroid glands [124,136]. Experiments in neonatal and adult mice further documented that the cellular locations of Phex in bone and teeth are the osteoblast/osteocyte and the odontoblast/ ameloblast, respectively, while subcellular locations are the cell surface membrane, the endoplasmic reticulum, and the Golgi compartment. Notably Phex expression is absent in the visceral abdominal organs, including the kidney, liver hepatocytes, and intestine, and in cardiac and skeletal muscle. In any case, the ontogeny of Phex expression reveals that the protein is expressed in osteoblasts at both primary and secondary ossification centers, suggesting a possible role in mineralization in vivo.

CHAPTER 69 Clinical Disorders of Phosphate Homeostasis

PHEX mutations consisting of deletions, insertions, and duplications, as well as splice site, nonsense and missense mutations, have been documented in >160 patients with XLH [64,127–134] and are scattered throughout exons 2–22, which encode the 749-amino acid extracellular protein domain. In addition, a single mutation within the 5′ untranslated region has been identified [130]. Although these mutations invariably cause loss of function, the mechanism by which such loss of activity occurs is unclear. However, preliminary data indicate that missense mutations interfere with protein trafficking, resulting in protein sequestration in the endoplasmic reticulum [133]. Until recently PHEX coding region mutations had not been detected in ~35% of patients. Accordingly, Christie et al. [134] explored whether such subjects have intronic mutations that result in mRNA splicing abnormalities. They found in one patient a unique mutation in intron 7 that created a novel donor splice site, which interacts with 3 naturally occurring acceptor splice sites, leading to the incorporation of 3 pseudoexons in PHEX transcripts. Translation of these pseudoexons results in the inclusion of missense amino acids into the PHEX protein or a truncated protein, lacking 5 of 10 conserved cysteine residues and the pentapeptide zinc-binding motif. These observations suggest that intron mutations may represent a proportion of the gene abnormalities undiscovered in one-third of patients with XLH. In order to confirm the possibility that diminished PHEX/Phex expression in osteoblasts initiates the cascade of events responsible for the pathogenesis of XLH, several investigators have used targeted overexpression of Phex in attempts to normalize osteoblast mineralization, in vitro, and rescue the Hyp phenotype in vivo. Surprisingly, however, these studies [81,137] revealed that restoration of Phex expression and enzymatic activity to immortalized hyp-mouse osteoblasts, by retroviral mediated transduction, does not restore their capacity to mineralize extracellular matrix in vitro, under conditions supporting normal mineralization. Moreover, in complementary studies Liu et al. [81] and Bai et al. [69] found transgenic hyp-mice (Osc-PhexHyp; pOb2.3[ColIaI]-Phex-Hyp), despite expressing abundant Phex mRNA and enzyme activity in mature osteoblasts and osteocytes, exhibited hypophosphatemia and persistently abnormal vitamin D metabolism. In the setting of P depletion, although exhibiting a modest improvement in bone mineralization, the transgenic mice maintained histological evidence of osteomalacia, similar to that in nontransgenic hyp-mice. These observations are consistent with several possibilities, acknowledged by Liu et al. [81] and Bai et al. [69]. First, despite theoretical evidence to the contrary (see above), extraosseous Phex expression may play an

1171 important role in the modulation of phosphatonin activity. In support of this option, Miyamura et al. [138] using syngeneic bone marrow transplantation, were able to partly reverse the biochemical abnormalities in hyp-mice with an engraftment that was not restricted to cells of the osteoblast lineage, but included donor cells to alternate tissues, in many of which PHEX transcripts have been detected [58–61]. Second, the temporal and developmental expression of the Osc and pOb2.3 promoterdriven Phex expression may not mimic the endogenous regulation of Phex. In this regard, the transgenic animals may experience PHEX expression later than, or in osteoblast-related cell subpopulations different from those in normal animals. In fact, neither promoter is expressed in the preosteoblast and the osteocalcin promoter appears at least four days later than PHEX in normally developing osteoblasts [69,81]. Thus, lack of Phex activity early in osteoblast development (in preosteoblasts or preceding osteocalcin expression in osteoblasts) may result in failure to alter an otherwise immutable osteoblast dysfunction, the continued presence of which contributes to the impairment of mineralization. In accord, later expression of PHEX may not rescue the phenotype, as the immutable change is refractory to endopeptidase activity. However, Erben et al. [82] recently reported that ubiquitous overexpression of Phex under the control of the β-actin promoter in two different mouse lines markedly improved BMD but did not completely resolve the bone mineralization defect and failed to alter the abnormal phosphate homeostasis, raising significant further questions regarding the interaction between the PHEX gene defect and phenotypic expression of the disease. Regardless, in no way do these observations exclude a role in the mineralization process for Phex expression in the mature osteoblast. Indeed, recent studies support the possibility that abnormal mineralization in hyp-mouse osteoblasts is due to a combination of Phex inactivity in the mature cell and Phex refractory abnormal P transport in the osteoblast, possibly due to gene inactivity early in cell development [139]. In such circumstances, the rescue of the Hyp phenotype by Phex expression in mature osteoblasts may be limited, in part, by phosphate availability. Hence, hypophosphatemia per se may contribute to the severity of the mineralization defect in hyp-mice. The plausibility of this possibility is reinforced by the discovery that P deficiency results in rickets in developing animals (and humans) [140] and the recent observations, which document that hyp-mouse neonates, exposed to normal phosphorus levels during embryonic development, manifest at birth apparently normal endochondral mineralization and only mildly abnormal mineralization in metaphyseal bone [32]. In any case, the inconsistent results of

1172 efforts to rescue the HYP phenotype clearly indicate that salient elements of the complex pathogenesis underlying XLH remain unknown. c. Pathogenesis In spite of the remarkable advances that have been made in understanding the genetic abnormality and pathophysiology of XLH, the detailed pathogenetic mechanism underlying this disease remains unknown (see above). While as related previously, the identity of phosphatonin remains uncertain, it is still tempting to speculate that the PHEX gene product acts directly or indirectly on a phosphaturic factor that regulates renal phosphate handling. Moreover, precedent for such activity exists as neutral endopeptidases inactivate enkephalins and atrial natriuretic peptide [141,142]. However, the data from parabiotic studies of normal and hyp-mice argue strongly that extracellular degradation of the phosphaturic factor does not occur. Indeed, such activity would preclude transfer of the hyp-mouse phenotype to parabiosed normals. Alternatively, the PHEX gene product may function intracellularly to inactivate phosphatonin. In this regard, Jalal et al. [143] recently reported the internalization of NEP and a potential role for this enzyme in intracellular metabolism. Less likely, the PHEX gene product may enzymatically activate a protein, which suppresses production of phosphatonin. While this is consistent with all previous data, it is a complex process and requires production of PHEX, phosphatonin, and the suppressor protein in the same cell in order to accommodate the data from the parabiotic studies. Nevertheless, in accord with this possibility, Mari et al. [144] have reported that the NEP on human T-cells may be involved in the production of lymphokines through the processing of an activating factor at the surface of the lymphocyte. In any of these cases, a defect in the PHEX gene will result in overproduction and circulation of phosphatonin and consequent inhibition of renal Na+-phosphate transport, the likely scenario in the pathogenesis of XLH. Although such overproduction of phosphatonin is a favored hypothesis, as noted previously, it is possible that XLH results from the inability of mutant PEX to activate a phosphate conserving hormone. However, the only known phosphate-conserving hormone, stanniocalcin, is synthesized in active form within the kidney and has little known bioactivity in man. These features strongly mitigate against a role for stanniocalcin in the pathogenesis of XLH. The coexistence of osteoblast defects in XLH further confounds understanding the pathophysiology of this disorder. Elegant experiments, which documented the abnormal mineralization of periostea and osteoblasts of hyp-mice following transplantation into the muscle of normal mice, provide clear evidence for intrinsic defects

MARC K. DREZNER

in the bone of mutants [145]. Indeed, proof of specific osteoblast abnormalities has been provided by studies that show decreased phosphorylation of osteopontin and increased osteocalcin levels in cells from hyp-mice [146]. Based on these observations, it is tempting to speculate that co-ordinate PHEX expression and phosphatonin production in osteoblasts may impart innate functional abnormalities to these cells in X-linked hypophosphatemia [147]. Indeed, the phosphatonin, likely processed by the PHEX gene product, may have multiple activities beyond regulation of renal phosphate transport, which may include modulation of vitamin D metabolism and osteoid mineralization, defects that characterize XLH. In concert with this possibility, PHEX expression correlates temporally with osteoblast mediated mineralization in vitro. In any case, it is evident that further information is requisite to enhance our understanding of the pathogenesis of XLH and, in turn, regulation of phosphate homeostasis, osteoblast function, vitamin D metabolism, and osteoid mineralization. Such data are not only critical to understanding the pathogenesis of XLH and the regulation of mineral homeostasis, but may have significant impact upon determination of optimal treatment strategies for many of the vitamin D–resistant diseases. d. Treatment In past years, physicians employed pharmacologic doses of vitamin D as the cornerstone for treatment of XLH. However, long-term observations indicate that this therapy fails to cure the disease and poses the serious problem of recurrent vitamin D intoxication and renal damage. More recently, choice of therapy for this disease has been remarkably influenced by the increased understanding of the pathophysiological factors, which affect phenotypic expression of the disorder. Thus, current treatment strategies for children directly address the combined calcitriol and phosphorus deficiency characteristic of the disease. Generally, the regimen includes a period of titration to achieve a maximum dose of calcitriol, 1-3 µg/d in two divided doses and phosphorus, 1-4 g/d in 4-5 divided doses [148,149]. Such combined therapy often improves growth velocity, normalizes lower extremity deformities, and induces healing of the attendant bone disease. Refractoriness to the growth-promoting effects of treatment, however, is often encountered particularly in youths presenting at <5th percentile in height [150]. For that reason, the use of recombinant growth hormone as an additional treatment component has been advocated recently. Definite positive effects have been observed in young patients with XLH [151,152]. Of course, treatment involves a significant risk of toxicity that is generally expressed as abnormalities of calcium homeostasis, most notably secondary hyperparathyroidism that may become autonomous and require surgery.

CHAPTER 69 Clinical Disorders of Phosphate Homeostasis

Detrimental effects on renal function secondary to abnormalities such as nephrocalcinosis are also possible. Thus, frequent monitoring of the phosphate and calcitriol dosage during growth is mandatory. Therapy in adults is reserved for episodes of intractable bone pain and refractory nonunion bone fractures. 3. HEREDITARY HYPOPHOSPHATEMIC RICKETS WITH HYPERCALCIURIA (HHRH)

This rare genetic disease is marked by hypophosphatemic rickets with hypercalciuria [153]. The cardinal biochemical features of the disorder include hypophosphatemia due to increased renal phosphate clearance and normocalcemia. In contrast to other diseases in which renal phosphate transport is limited, patients with HHRH exhibit increased 1,25(OH)2D production [153–157] (Table III). The resultant elevated serum calcitriol levels enhance the gastrointestinal calcium absorption, which in turn increases the filtered renal calcium load and inhibits parathyroid secretion. These events cause the characteristic hypercalciuria observed in affected patients. The clinical expression of the disease is heterogeneous. In general, children become symptomatic between the ages of six months and seven years. Initial symptoms consist of bone pain or deformities of the lower limbs (or both), which progressively interferes with gait and physical activity. The bone deformities vary from genu varum or genu valgum to anterior external bowing of the femur and coxa vara. Additional features at presentation include short stature with disproportionately short lower limbs, muscle weakness, and radiological signs of rickets or osteomalacia (or both). These various symptoms and signs may exist separately or in combination and may be present in a mild or severe form. A large number of apparently unaffected relatives of patients with HHRH exhibit an additional mode of disease expression [157]. These subjects, although without evidence of bone disease, manifest idiopathic hypercalciuria (IH), most evident in post-prandial periods, as well as a pattern of biochemical abnormalities similar to those of children with rickets and osteomalacia. Quantitatively, however, the abnormalities are milder, and the relevant biochemical values intermediate between those observed in family members with HHRH and those in normal relatives. The absence of bone disease in these patients may be explained by relatively mild hypophosphatemia compared to the severe phosphate depletion evidenced in patients with the full spectrum of the disorder [157]. Only four unrelated kindreds with HHRH have been described, including an extended family of Bedouin origin that includes 13 patients with HHRH and 42 with

1173 hypercalciuria; a smaller kindred of oriental Jewish origin with 5 affected members; and a family of Yemenite Jewish origin that includes 2 patients with HHRH and 2 with hypercalciuria [155]. However, a phenotypically similar disorder, childhood idiopathic hypercalciuria with bone lesions (rickets) and stunted linear growth has been independently recognized. The similarity of this syndrome to HHRH suggests that they may be one and the same pathologic condition. Moreover, several patients with a sporadic occurrence of HHRH have been recently recognized. Studies are generally incomplete, however, and the presence of hypercalciuria in relatives has not been excluded. a. Pathophysiology Liberman, Tieder, and associates [153,157,158] have presented data that indicate that the primary inborn error underlying this disorder is an expressed abnormality in the renal proximal tubule, which impairs phosphate reabsorption. They propose that this pivotal defect in turn stimulates renal 25(OH)D1α-hydroxylase, thus promoting the production of 1,25(OH)2D and increasing its serum and tissue levels. As a result intestinal calcium and phosphorus absorption is augmented and the renal filtered calcium load consequently increased. The enhanced intestinal calcium absorption also suppresses parathyroid function. In addition, prolonged hypophosphatemia diminishes osteoid mineralization, resulting in rickets and/or osteomalacia. The proposal that abnormal phosphate transport results in increased calcitriol production remains untested. Indeed, the elevation of 1,25(OH)2D in patients with HHRH is a unique phenotypic manifestation of the disease that distinguishes it from other disorders in which abnormal phosphate transport is likewise manifest. Such heterogeneity in the phenotype of these disorders suggests that disease at variable anatomical sites along the proximal convoluted tubule or involving compensatory enhancement of sodium-dependent phosphate cotransporters, such as that in the Npt2a−/− mouse [11,31], while uniformly impairing phosphate transport, may not necessarily inhibit 25(OH)D-1α-hydroxylase activity. b. Genetic Defect Autosomal recessive transmission of HHRH seems consistent with the inheritance pattern in the described kindreds. However, if HHRH is an autosomal recessive disease and individuals with IH are heterozygous for the mutant allele, IH must be an incompletely penetrant trait because not all obligate heterozygotes manifest hypercalciuria. Alternatively, it has been suggested that HHRH and IH could be the result of mutations in two different genes [157]. The strongest support against this hypothesis is that when individuals with HHRH are treated with oral Pi, both the hypophosphatemia and the hypercalciuria are corrected. Nevertheless, the variability in clinical

1174 presentation in the various described kindred suggests that mutations in different genes may be involved. Regardless, the search for a specific gene defect underlying this disease has been unrewarding. Indeed, efforts to implicate an abnormality in the NPT2a gene in patients with HHRH have been unsuccessful [159]. c. Treatment In accord with the hypothesis that a singular defect in renal phosphate–transport underlies HHRH, patients have been treated successfully with high dose phosphorus (1–2.5 g/d in 5 divided doses) alone. Within several weeks after initiation of therapy, bone pain disappears and muscular strength improves substantially. Moreover, the majority of treated subjects exhibit accelerated linear growth and radiological signs of rickets are completely absent within 4–9 months. Concordantly, serum phosphorus values increase towards normal, the 1,25(OH)2D concentration decreases, and alkaline phosphatase activity declines to normal. Despite this favorable response, limited investigation indicates that the osteomalacic component of the bone disease does not exhibit normalization. Further studies will be necessary, therefore, to determine if phosphorus alone will be sufficient treatment for this rare disorder. In any case, administration of phosphorus in patients with this disease does not result in the same spectrum of complications encountered upon its use in other disorders. Most notably, nephrocalcinosis, a common complication in treated patients with XLH, occurs infrequently in subjects with HHRH. In fact, the rare occurrence of this complication is associated with a history of vitamin D intoxication prior to initiation of treatment with phosphorus. Similarly, the development of secondary hyperparathyroidism in treated patients with HHRH has not been reported, although expectation of this complication is high since oral administration of phosphate does diminish the circulating calcium concentration and, in turn, stimulates parathyroid function. 4. FANCONI SYNDROME (FS)

Rickets and osteomalacia are frequently associated with Fanconi syndrome, a disorder characterized by hyperphosphaturia and consequent hypophosphatemia, hyperaminoaciduria, renal glycosuria, albuminuria, and proximal renal tubular acidosis [160–163]. Damage to the renal proximal tubule, secondary to genetic disease or environmental toxins, represents the common underlying mechanism of this disease [164]. Resultant dysfunction causes renal wasting of those substances primarily reabsorbed at the proximal tubule. The inherited form may occur in isolation (in the absence of any other metabolic disease) or secondary to various primary Mendelian diseases. The associated bone disease in this disorder is likely secondary to hypophosphatemia and/or acidosis, abnormalities that occur in association

MARC K. DREZNER

with aberrantly (Fanconi syndrome, type I) or normally regulated (Fanconi syndrome, type II) vitamin D metabolism. a. Fanconi Syndrome, Type I Renal phosphate wasting and hypophosphatemia are the hallmark abnormalities of this disease, which resembles in many respects the more common genetic disease, XLH. In this regard, occurrence of abnormal bone mineralization appears dependent upon the prevailing renal phosphate wasting and resultant hypophosphatemia. Indeed, disease subtypes in which isolated wasting of amino acids, glucose, or potassium occur, are not associated with rickets and/or osteomalacia. Further, in the majority of patients studied, affected subjects exhibit abnormal vitamin D metabolism, characterized by serum 1,25(OH)2D levels that are overtly decreased or abnormally low relative to the prevailing serum phosphorus concentration [165–167]. Although the aberrantly regulated calcitriol biosynthesis may be due to the abnormal renal phosphate transport, as suspected in patients with XLH, proximal tubule damage and acidosis may play important roles. A notable difference between this syndrome and XLH is a prevailing acidosis, which may contribute to the bone disease. In this regard, several studies indicate that acidosis may exert multiple deleterious effects on bone. Such negative sequellae are related to the loss of bone calcium that occurs secondary to calcium release for use in buffering [160,168]. Alternatively, several investigators [169,170] have reported that acidosis may impair bone mineralization secondary to direct inhibition of renal 25(OH)D-1α−hydroxylase activity. Others dispute these findings and claim that acidosis does not cause rickets or osteomalacia in the absence of hypophosphatemia. Indeed, Brenner et al. [171] reported that the rachitic/osteomalacic component of this disorder occurs only in patients with type 2 renal tubular acidosis and phosphate wasting. In contrast, those with type 1 and 4 renal tubular acidosis displayed no evidence of abnormal bone mineralization. Thus, the interplay of acidosis and phosphate depletion on bone mineralization in this disorder remains poorly understood. Most likely, however, hypophosphatemia and abnormally regulated vitamin D metabolism are the primary factors underlying rickets and osteomalacia in this form of Fanconi Syndrome. b. Fanconi’s Syndrome, Type II Tieder et al. [172] have described two siblings (from a consanguineous mating) who presented with classic characteristics of Fanconi syndrome, including renal phosphate wasting, glycosuria, generalized aminoaciduria, and increased urinary uric acid excretion. However, these patients had appropriately elevated (relative to the decreased serum phosphorus concentration) serum 1,25(OH)2D

CHAPTER 69 Clinical Disorders of Phosphate Homeostasis

levels and consequent hypercalciuria despite normal serum parathyroid hormone levels and cyclic AMP excretion. Moreover, treatment with phosphate reduced the serum calcitriol in these patients into the normal range and normalized the urinary calcium excretion. In many regards, this syndrome resembles HHRH and represents a variant of Fanconi syndrome, referred to as type II disease. The bone disease in affected subjects is likely due to the effects of hypophosphatemia. In any case, the existence of this variant form of disease is probably the result of renal damage to a unique segment of the proximal tubule or involvement of a different mechanism at the same site [172]. Further studies will be necessary to distinguish these possibilities. c. Treatment Ideally, treatment of the bone disease in this disorder should offset the pathophysiological defect influencing proximal renal tubular function. In many cases, however, the primary abnormality remains unknown. Moreover, efforts to decrease tissue levels of causal toxic metabolites by dietary (such as in fructose intolerance) or pharmacological means (such as in cystinosis and Wilson syndrome) have met with variable success. Indeed, no evidence exists that indicates if the proximal tubule damage is reversible upon relief of an acute toxicity. Regardless, in instances when specific therapies are not available or do not lead to normalization of the primary defect, therapy must be directed at raising the serum phosphorus concentration, replacing calcitriol (in type I disease) and reversing the associated acidosis. However, use of phosphorus and calcitriol in this disease has been limited. In general, such replacement therapy leads to substantial improvement or resolution of the bone disease [173]. Unfortunately, growth and developmental abnormalities, more likely associated with the underlying genetic or acquired disease, remain substantially impaired [173]. More efficacious therapy, therefore, is dependent upon future research into the causes of the multiple disorders that underlie this syndrome. 5. X-LINKED RECESSIVE HYPOPHOSPHATEMIA (XLRH)

The initial description of X-linked recessive hypophosphatemic rickets involved a family in which males presented with rickets or osteomalacia, hypophosphatemia, and a reduced renal threshhold for phosphate reabsorption. In contrast to patients with XLH, affected subjects exhibited hypercalciuria, elevated serum 1,25(OH)2D levels (Table I), and proteinuria of up to 3 g/day. Patients also developed nephrolithiasis and nephrocalcinosis with progressive renal failure in early adulthood. Female carriers in the family were not hypophosphatemic and lacked any biochemical abnormalities other than hypercalciuria. Three related syndromes have been reported independently: X-linked

1175 recessive nephrolithiasis with renal failure, Dent’s disease, and low-molecular-weight proteinuria with hypercalciuria and nephrocalcinosis. These syndromes differ in degree from each other, but common themes include proximal tubular reabsorptive failure, nephrolithiasis, nephrocalcinosis, progressive renal insufficiency, and, in some cases, rickets or osteomalacia. Identification of mutations in the voltage-gated chloride-channel gene CLCN5 in all four syndromes has established that they are phenotypic variants of a single disease and are not separate entities [174,175]. However, the varied manifestations that may be associated with mutations in this gene, particularly the presence of hypophosphatemia and rickets/osteomalacia, underscore that environmental differences, diet, and/or modifying genetic backgrounds may influence phenotypic expression of the disease.

B. Disorders of Renal Phosphate Transport: Hyperphosphatemic Diseases 1. (HYPERPHOSPHATEMIC) TUMORAL CALCINOSIS (TC)

Tumoral calcinosis is a rare genetic disease characterized by periarticular cystic and solid tumorous calcifications. Most patients in North America with this disorder are black and about one-third of the cases are familial. There is no gender preference. Biochemical markers of the disorder include hyperphosphatemia and an elevated serum 1,25(OH)2D concentration in many patients. Using these criteria, evidence has been presented for autosomal recessive inheritance of this syndrome. However, an abnormality of dentition, marked by short bulbous roots, pulp stones, and radicular dentin deposited in swirls, is a phenotypic marker of the disease that is variably expressed [176]. Thus, this disorder may have multiple formes frustes that could complicate genetic analysis. Indeed, using the dental lesion as well as the more classic biochemical and clinical hallmarks of the disease, an autosomal dominant pattern of transmission has been documented [176]. The hyperphosphatemia characteristic of the disease results from an increase in capacity of renal tubular phosphate reabsorption secondary to an unknown defect [176,177]. Hypocalcemia is not a consequence of this abnormality, however, and the serum parathyroid hormone concentration is normal. Moreover, the phosphaturic and urinary cAMP responses to parathyroid hormone are not disturbed. Thus, the defect does not represent renal insensitivity to a hormone, or hypoparathyroidism. Rather, the basis of the disease is probably an innate or hormone/metabolic factor mediated abnormality of the renal tubule that enhances phosphate reabsorption. The identification of the phosphatonin family of

1176 hormones and documentation of their biological activities suggest that an inactivating mutation of a phosphatonin may underlie tumoral calcinosis. Indeed, the presence of an elevated serum phosphorus concentration and increased 1,25(OH)2D production in the FGF 23 knockout mouse supports this possibility [47]. Interestingly, affected patients manifest increased circulating 1,25(OH)2D levels despite hyperphosphatemia, underlining the fact that it is TmP/GFR rather than the serum phosphorus concentration that controls 25(OH)D-1α-hydroxylase activity (Fig. 2). Undoubtedly, the calcific tumors result from the elevated calcium-phosphorus product. The observation that long-term phosphorus depletion alone [178] or in association with administration of aluminum hydroxide [179] or acetazolamide, a phosphaturic agent [180], leads to resolution of the tumor masses supports this possibility. Furthermore, reduction of phosphate levels in extracellular fluid helps prevent reformation of mineral deposits [178]. In addition, preliminary studies indicate that calcitonin therapy may also be efficacious by enhancing phosphaturia [181]. An acquired form of this disease is rarely seen in patients with end-stage renal failure. Affected patients manifest hyperphosphatemia in association with either: 1) an inappropriately elevated calcitriol level for the degree of renal failure, hyperparathyroidism or hyperphosphatemia; or 2) long-term treatment with calcium carbonate, calcitriol, or high calcium-content dialysates. Again, calcific tumors likely result from an elevated calcium-phosphorus product. Indeed, complete remission of the tumors occurs with treatment with vinpocetine, a mineral scavenger drug.

III. DISORDERS RELATED TO AN ALTERED PHOSPHATE LOAD A. Decreased Phosphate Load 1. PHOSPHATE DEPRIVATION

Hypophosphatemia and phosphate depletion due to inadequate dietary intake are rare. With a decline in ingested phosphate, the renal TmP increases and urinary phosphate excretion decreases [182]. In addition, gastrointestinal phosphate secretion gradually lessens. However, severe dietary deprivation (less than 100 mg/d) leads to a prolonged period of negative phosphate balance and total body depletion. Affected females may display hypophosphatemia (1.4 to 2.5 mg/dL); interestingly, in contrast, males generally do not manifest a decreased serum phosphate concentration in response to dietary deprivation. Nevertheless, attempts to maintain phosphate homeostasis in both sexes include suppression of the serum PTH concentration and increased

MARC K. DREZNER

1,25(OH)2D production. Thus, hypercalciuria may be associated with the syndrome. Whether the efficiency of calcitriol responsiveness or differential effects on end organs, such as the gastrointestinal tract or bone, underlies the noted gender difference remains unclear. Total starvation does not cause hypophosphatemia. The catabolic effects of total food deprivation result in the release of phosphate from intracellular stores, which compensates for the negative phosphorus balance. However, refeeding of the starved person will result in hypophosphatemia if phosphate deprivation is maintained. 2. GASTROINTESTINAL MALABSORPTION

Gastrointestinal absorption of phosphorus may be decreased with the use of aluminum- or magnesiumcontaining antacids; prolonged use of these drugs in large amounts has been associated with hypophosphatemia and a negative phosphorus balance [183]. Long-term reduction of the serum phosphorus concentration owing to chronic, excessive use of antacids leads to frank osteomalacia and myopathy. The osteomalacia results as a direct consequence of the phosphate depletion and in spite of normal vitamin D stores and an increased serum 1,25(OH)2D concentration, which occurs in response to the hypophosphatemia. In contrast, the pathophysiology of variably occurring osteomalacia secondary to gastrointestinal diseases that cause steatorrhea or rapid transit time (e.g., Crohn’s disease, postgastrectomy states, and intestinal fistulas) is significantly different [184]. In this case, mild to moderate hypophosphatemia occurs due to vitamin D malabsorption/deficiency and resultant secondary hyperparathyroidism and renal phosphate wasting. Further, the relation between the metabolic bone disease, vitamin D deficiency, and hypophosphatemia is complex and likely involves the influence of phosphopenia and vitamin D–dependent calciopenia on bone mineralization. However, the impact of vitamin D deficiency may be overriding since osteomalacia may occur in the absence of hypophosphatemia. Regardless, in the presence of hypophosphatemia (and/or phosphate depletion), an elevated serum calcitriol level is part of this syndrome and has no apparent effect on the evolution of the bone disease.

B. Transcellular Shift In a large proportion of clinically important cases of hypophosphatemia, a sudden shift of phosphorus from the extracellular to the intracellular compartment is responsible for the decline of the serum phosphate concentration. This ion movement occurs in response

1177

CHAPTER 69 Clinical Disorders of Phosphate Homeostasis

to naturally occurring disturbances and after the administration of certain compounds. 1. ALKALOSIS

Alkalosis secondary to intense hyperventilation may depress serum phosphate levels to less than 1 mg/dL [185]. A similar degree of alkalemia owing to excess bicarbonate also causes hypophosphatemia, but of a much lesser magnitude (2.5 to 3.5 mg/dL). The disparity between the effects of a respiratory and metabolic alkalosis is related to the more pronounced intracellular alkalosis that occurs during hyperventilation. The phosphate shift to the intracellular compartment results from the utilization attendant on glucose phosphorylation, a process stimulated by a pH-dependent activation of phosphofructokinase. 2. GLUCOSE ADMINISTRATION

The administration of glucose and insulin often results in moderate hypophosphatemia [186]. Endogenous or exogenous insulin increases the cellular uptake not only of glucose, but also of phosphorus. The most responsive cells are those of the liver and skeletal muscle. The decline of the serum phosphate concentration generally does not exceed 0.5 mg/dL. A lesser decrease is manifest in patients with type II diabetes mellitus and insulin resistance or those with a disease causing a diminished skeletal mass. The administration of fructose and glycerol similarly reduces the serum phosphorus concentration. In contrast to glucose, fructose administration may be associated with more pronounced hypophosphatemia; the more striking effect is due to unregulated uptake by the liver.

C. Combined Mechanisms There are special clinical situations in which an altered phosphate and consequent hypophosphatemia result from both a transcellular shift of phosphorus and phosphate deprivation or renal phosphate wasting. These disorders represent some of the more common and profound causes of a decreased serum phosphorus concentration.

hyperventilation that occurs in patients with cirrhosis or during alcohol withdrawal [186]. Moreover, many alcoholic patients are hypomagnesemic, which potentiates renal phosphate wasting by an unclear mechanism. 2. BURNS

Within several days after sustaining an extensive burn, patients often manifest severe hypophosphatemia. The initial insult induces a transient retention of salt and water. When the fluid is mobilized, significant urinary phosphorus loss ensues. Coupled with the shift of phosphorus to the intracellular compartment, which occurs secondary to hyperventilation, and the anabolic state, profound hypophosphatemia may result. 3. NUTRITIONAL RECOVERY SYNDROME

Refeeding of starved individuals or maintaining nutritional support by parenteral nutrition or by tube feeding, without adequate phosphorus supplementation, may also cause hypophosphatemia [187]. A prerequisite for the decreased serum phosphate concentration in affected patients is that their cells must be capable of an anabolic response. As new proteins are synthesized and glucose is transported intracellularly, phosphate demand depletes reserves. Several days are generally required after the initiation of refeeding in order to establish an anabolic condition. In patients receiving total parenteral nutrition, serum phosphate levels may be further depressed if sepsis supervenes and a respiratory alkalosis develops. 4. DIABETIC KETOACIDOSIS

Poor control of blood glucose and consequent glycosuria, polyuria, and ketoacidosis invariably cause renal phosphate wasting [186]. The concomitant volume contraction may yield a normal serum phosphate concentration. However, with insulin therapy, the administration of fluids, and correction of the acidosis, serum and urine phosphate fall precipitously. The resultant hypophosphatemia may contribute to insulin resistance and slow the resolution of the ketoacidosis. Hence, the administration of phosphate may improve the capacity to metabolize glucose and facilitate recovery.

1. ALCOHOLISM

Alcoholic patients frequently enter the hospital with hypophosphatemia. However, many do not exhibit a decreased serum phosphate concentration until several days have elapsed and refeeding has begun. The multiple factors that underlie the hypophosphatemia include poor dietary intake, use of phosphate binders to treat gastritis, excessive urinary losses of phosphorus, and shift of phosphorus from the extracellular to the intracellular compartment, owing to glucose administration and/or

D. Increased Phosphate Load 1. VITAMIN D INTOXICATION

An increase of the phosphate load from exogenous sources generally does not cause hyperphosphatemia because the excessive phosphorus is excreted by the kidney. However, an increased serum phosphate concentration may occur in vitamin D intoxication when the gastrointestinal absorption of phosphate is markedly

1178 enhanced. Increased phosphate mobilization from bone and a reduction of GFR, secondary to hypercalcemia and/or nephrocalcinosis, may also contribute to the evolution of the hyperphosphatemia. The chronic ingestion of large doses of vitamin D, in excess of 100,000 IU/day, is generally required to cause intoxication. Suspected hypervitaminosis D may be investigated using specific assays, which can document excessive amounts of vitamin D and its metabolites in the circulation (see Chapter 58). 2. RHABDOMYOLYSIS

Because muscle contains a large amount of phosphate, necrosis of muscle tissue may acutely increase the endogenous phosphate load and result in hyperphosphatemia. Such muscle necrosis (rhabdomyolysis) may complicate heat stroke, acute arterial occlusion, hyperosmolar nonketotic coma, trauma, toxic agents such as ethanol and heroin, and idiopathic paroxysmal myoglobinuria [188,189]. Muscle biopsy often reveals myolytic denervation, and as a consequence, acute renal failure caused by myoglobin excretion frequently complicates the clinical presentation and contributes to the hyperphosphatemia. However, an elevated serum phosphate concentration may precede evidence of renal failure, or occur in its complete absence when rhabdomyolysis is present. The diagnosis is confirmed by elevated serum creatine phosphokinase, uric acid, and lactate dehydrogenase concentrations, and the demonstration of heme-positive urine in the absence of red blood cells. Therapy is directed at the underlying disorder with maintenance of the extracellular volume to avoid volume depletion and alkalinization of the urine to prevent uric acid accumulation and consequent acute tubular necrosis. 3. CYTOTOXIC THERAPY

Cytotoxic therapy often causes cell destruction and liberation of phosphorus into the circulation [190]. The lysis of tumor cells begins within one to two days after initiating treatment, and is followed quickly by an elevation of the serum phosphate concentration. Hyperphosphatemia supervenes, however, only when the treated malignancies have a large tumor burden, rapid cell turnover, and substantial intracellular phosphorus content. Such malignancies include lymphoblastic leukemia, various types of lymphoma, and acute myeloproliferative syndromes. Hyperkalemia and hyperuricemia also occur. Indeed, uric acid nephropathy may cause renal insufficiency that predisposes to further phosphate retention and worsening of the hyperphosphatemia. 4. MALIGNANT HYPERTHERMIA

Malignant hyperthermia is a rare familial syndrome characterized by an abrupt rise in body temperature

MARC K. DREZNER

during the course of anesthesia [191]. The disease appears to be autosomal dominant in transmission, and an elevated serum creatine phosphokinase concentration is found in otherwise normal family members. Hyperphosphatemia results from shifts of phosphate from muscle cells to the extracellular pool. A high mortality rate accompanies the syndrome.

E. Clinical Signs and Symptoms of Abnormal Serum Phosphorus in Diseases Caused by an Altered Phosphate Load As related above, a wide variety of diseases and syndromes with varying clinical manifestations have the characteristic biochemical abnormalities of hyperphosphatemia or hypophosphatemia. A unique complex of disturbances often is directly related to the abnormal phosphate homeostasis. The recognition of these signs and symptoms may lead to appropriate biochemical testing, the diagnosis of an unsuspected disease, and initiation of lifesaving or curative treatment. 1. HYPOPHOSPHATEMIA

A low serum phosphorus level is associated with symptoms only if there is concomitant phosphate depletion. The presence of phosphate deficiency, however, may cause widespread disturbances. This is not surprising, since severe hypophosphatemia causes two critical abnormalities that impact on virtually all organ systems. First, a deficiency of 2,3-diphosphoglycerate (2,3-DPG) occurs in red cells, which is associated with an increased affinity of hemoglobin for oxygen and, therefore, tissue hypoxia. Second, there is a decline of tissue ATP content and a concomitant decrease in the availability of energy-rich phosphate compounds that are essential for cell function [192,193]. The major clinical syndromes resulting from these abnormalities include nervous system dysfunction, anorexia, nausea, vomiting, ileus, muscle weakness, cardiomyopathy, respiratory insufficiency, hemolytic anemia, and impaired leukocyte and platelet function. Additionally, phosphate deficiency causes osteomalacia and bone pain, clinical sequelae that are probably independent of the aforementioned abnormalities. Central nervous system dysfunction has been well characterized in severe hypophosphatemia, especially in patients receiving total parenteral nutrition for diseases causing severe weight loss. A sequence of symptoms compatible with a metabolic encephalopathy usually begins one or more weeks after the initiation of therapy with solutions that contain glucose and amino acids, but lack adequate phosphorus supplementation to prevent hypophosphatemia. The onset of dysfunction is marked by irritability, muscle weakness, numbness,

1179

CHAPTER 69 Clinical Disorders of Phosphate Homeostasis

and paresthesias, with progression to dysarthria, confusion, obtundation, coma, and death [194]. These patients have a profoundly diminished red cell 2,3-DPG content. Both biochemical abnormalities and clinical symptoms improve as patients receive phosphorus supplementation. Peripheral neuropathies, Guillain-Barre-like paralysis, hyporeflexia, intention tremor, and ballismus have also been described with hypophosphatemia and phosphate depletion. The effects of hypophosphatemia on muscle depend on the severity and chronicity of the deficiency. Chronic phosphorus deficiency results in a proximal myopathy with striking atrophy and weakness. Osteomalacia frequently accompanies the myopathy, so patients complain of pain in weight-bearing bones. Normal values for serum creatine phosphokinase and aldolase activities are characteristically present. Rhabdomyolysis does not occur with chronic phosphate depletion. In contrast, acute hypophosphatemia can lead to rhabdomyolysis with muscle weakness and pain. Most cases occur in chronic alcoholics or patients receiving total parenteral nutrition. In both groups of patients, muscle pain, swelling, and stiffness occur three to eight days after the initiation of therapies that do not contain adequate amounts of phosphorus. Muscle paralysis and diaphragmatic failure may occur. Studies of muscle tissue from chronically phosphate-depleted dogs made acutely hypophosphatemic show a decrement in cellular content of phosphorus, ATP, and adenosine diphosphate (ADP). Rhabdomyolysis occurred in these muscle fibers. The laboratory findings in patients with hypophosphatemic myopathy and with rhabdomyolysis include elevated serum creatine phosphokinase levels; however, serum phosphate levels may become normal if enough necrosis has occurred with subsequent phosphorus release. Also, renal failure and hypocalcemia can be associated with the syndrome. Myocardial performance can be abnormal at serum phosphate levels of 0.7 to 1.4 mg/dL. This occurs when ATP depletion causes impairment of the actin-myosin interaction, the calcium pump of the sarcoplasma, and the sodium-potassium pump of the cell membrane [195]. The net result is reduced stroke work and cardiac output, which may progress to congestive heart failure. These problems are reversible with phosphate replacement. Respiratory failure can occur owing to failure of diaphragmatic contraction in hypophosphatemic patients. When serum phosphate levels are raised, diaphragmatic contractility improves. The postulated mechanism for the respiratory failure is muscle weakness secondary to inadequate levels of ATP and decreased glycolysis as the result of phosphate depletion [196]. Two disturbances of red cell function may occur secondary to phosphorus deficiency. First, as intraerythrocyte ATP production is decreased, the erythrocyte

cell membrane becomes rigid, which can cause hemolysis [197]. This is rare and is usually seen in septic, uremic, acidotic, or alcoholic patients, when serum phosphate levels are less than 0.5 mg/dL. Second, the limited production of 2,3-DPG causes a leftward shift of the oxyhemoglobin curve and impairs the release of oxygen to peripheral tissues. Such a consequence of chronic hypophosphatemia has been documented in children with XLH and proposed as one factor underlying retarded statural growth [198]. Leukocyte dysfunction, which complicates phosphate deficiency, includes decreased chemotaxis, phagocytosis, and bactericidal activity [199]. These abnormalities increase the host susceptibility to infection. The mechanism by which hypophosphatemia impairs the various activities of the leukocyte probably is related to impairment of ATP synthesis. Decreased availability of energy impairs microtubules that regulate the mechanical properties of leukocytes and limit the rate of synthesis of organic phosphate compounds that are necessary for endocytosis. Abnormal platelet survival, causing thrombocytopenia, profuse gastrointestinal bleeding, and cutaneous bleeding, has also been described in association with phosphate depletion in animal studies. Despite these abnormalities, there is little evidence that hypophosphatemia is a primary cause of hemorrhage in humans. Perhaps the most consistent abnormalities associated with phosphate depletion are those on bone. Acute phosphate depletion induces dissolution of apatite crystal from the osseous matrix. This effect may be due to 1,25(OH)2D3, which is increased in response to phosphate depletion in both animals and humans. More prolonged hypophosphatemia leads to rickets and osteomalacia. This complication is a common feature of phosphate depletion. However, the ultimate cause is variable. While simple phosphate depletion alone may underlie the genesis of the abnormal mineralization, in many disorders the defect is secondary to phosphate depletion and commensurate 1,25(OH)2D3 deficiency. Thus, treatment of this complication may often require combination therapy, phosphate supplements, and calcitriol. 2. HYPERPHOSPHATEMIA

Hypocalcemia and consequent tetany are the most serious clinical sequelae of hyperphosphatemia [200]. The decreased serum calcium concentration results from the deposition of calcium phosphate salts in soft tissue, a process that may lead to symptomatic ectopic calcification. The dystrophic calcification is frequently seen in acute and chronic renal failure, hypoparathyroidism, pseudohypoparathyroidism, and tumoral calcinosis. Indeed, deposition of calcium-phosphate complexes in the kidney may predispose to acute

1180 renal failure. When the calcium-phosphate product exceeds 70, the probability that soft tissue calcification will occur increases sharply. In addition, local factors, such as tissue pH and injury (e.g. necrotic or hypoxic tissue), may predispose to precipitation of the calciumphosphate salts. In chronic renal failure, calcification occurs in arteries, muscle tissue, periarticular spaces, the myocardial conduction system, lungs, and the kidney. Affected patients may also have ocular calcification, causing the “red eye” syndrome of uremia and subcutaneous calcification, which also plays a role in uremic pruritus. On the other hand, a predisposition to calcification of periarticular surfaces of the hips, elbows, shoulders, and other large joints occurs in tumoral calcinosis. In some disease states, hyperphosphatemia may also play an important role in the development of secondary hyperparathyroidism [201]. A decrement in the serum calcium concentration secondary to hyperphosphatemia stimulates the release of PTH. Furthermore, hyperphosphatemia decreases the activity of 25(OH)D-1α-hydroxylase. The consequent diminished production of 1,25(OH)2D3 impairs the gastrointestinal absorption of calcium and induces skeletal resistance to PTH, influences that augment the development of hyperparathyroidism. Thus, hyperphosphatemia triggers a cascade of events that impact on calcium homeostasis at multiple sites. The prevention of secondary hyperparathyroidism, metabolic bone disease, and soft tissue and vascular calcification in affected patients, therefore, depends on ultimately controlling the serum phosphate concentration.

F. Treatment of Abnormal Serum Phosphorus in Diseases Caused by an Abnormal Phosphate Load Treatment of the myriad of diseases that characteristically display hyperphosphatemia or hypophosphatemia depends on determining the mechanism underlying their pathogenesis. The cause can almost always be ascertained by assessment of the clinical setting, determination of renal function, measurement of urinary phosphate excretion, and analysis of arterial carbon dioxide tension and pH. Therapy is aimed at correcting both the serum phosphate concentration and associated complications. The treatment of phosphate depletion depends on replacing body phosphorus stores. Preventive measures, however, will preclude the onset of phosphate depletion. Thus, appropriate monitoring of patients taking large doses of aluminum-containing antacids and provision of phosphorus intravenously to patients

MARC K. DREZNER

with diabetic ketoacidosis will preserve phosphate stores. Alternatively, the treatment of established phosphate depletion may require 2.5 to 3.7 g of phosphate daily, preferably administered orally in four equally divided doses. Providing K-Phos Neutral tablets, which contain 250 mg of elemental phosphorus per tablet, will fulfill this goal. If oral therapy is not tolerated and the serum phosphate shows a downward trend approaching dangerous levels (<1.2 mg/dL), intravenous phosphate supplementation at a dose of 10 mg/kg body weight/day may be administered. Such therapy should be discontinued when the serum phosphate reaches values >2.0 mg/dL. However, therapy is not required for many of the conditions resulting in phosphate depletion. Only when the consequences of severe depletion are manifest should treatment be initiated. Theoretically, decreasing the TmP, increasing the GFR, or diminishing the phosphate load (exogenous or endogenous) may decrease the elevated serum phosphate. There are no generally available pharmacologic means of acutely altering the GFR or reducing the TmP. However, chronic use of drugs, such as acetazolamide, which decreases TmP and induces phosphaturia, is effective as ancillary treatment of disorders such as tumoral calcinosis. Nevertheless, regulation of hyperphosphatemia is most often achieved by reducing the renal phosphate load. In tumoral calcinosis and chronic renal failure, such an effect is obtained by restricting the dietary phosphate intake or by administering phosphate binders such as calcium carbonate or aluminum hydroxide. Alternative strategies for management of load-dependent hyperphosphatemia include the administration of intravenous calcium or intravenous glucose and insulin. The consequence of such intervention is sequestration of phosphate in bone or soft tissues. Dialysis can also be used for the acute management of load-dependent disorders, or the chronic maintenance of phosphate overload such as those that complicate chronic renal failure.

References 1. Reichel H, Koeffler HP, Norman AW 1989 The role of the vitamin D endocrine system in health and disease. N Engl J Med 320:980–991. 2. Fraser DR 1980 Regulation of metabolism of vitamin D. Physiol Rev 60:551–613. 3. Yanagawa N, Nakhoul F, Kurokawa K, Lee DBN 1994 Physiology of phosphorus metabolism. In: Narins RG (ed.) Clinical Disorders of Fluid and Electrolyte Metabolism, 5th Edition, McGraw Hill, Inc., New York, pp. 307–371. 4. Tenenhouse HS 1997 Cellular and molecular mechanisms of renal P transport. J Bone Miner Res 12:159–164. 5. Murer H, Biber J 1996 Molecular mechanisms of renal apical Na/phosphate cotransport. Annu Rev Physiol 58:607–618.

1181

CHAPTER 69 Clinical Disorders of Phosphate Homeostasis

6. Tenenhouse HS, Roy S, Martel J, Gauthier C 1998 Differential expression, abundance, and regulation of Na+phosphate cotransporter genes in murine kidney. Am J Physiol 275(4 Pt 2):F527–F534. 7. Murer H, Hernanco N, Forster I, Biber J 2000 Proximal tubular P reabsorption: Molecular Mechanisms. Physiol Rev 80(4):1373–1409. 8. Biber J, Custer M, Werner A, Kaissling B, Murer H 1993 Localization of Na-Pi, a Na/Pi cotransporter in rabbit kidney proximal tubules. II. Localization by immunohistochemistry. Pflugers Arch 424:210–215. 9. Busch AE, Schuster A, Waldegger S, Wagner CA, Zempel G, Broer S, Biber J, Murer H, Lang F 1993 Expression of a renal type I sodium/phosphate transporter (NaPi-I) induces a conductance in Zenopus oocytes permeable for organic and inorganic anions. Proc Natl Acad Sci USA 93:5347–5351. 10. Yabuuchi H, Tamai I, Morita K, Kouda T, Miyamoto K, Takeda E, Tsuji A 1998 Hepatic sinusoidal membrane transport of anionic drugs mediated by anion transporter NptI. Journal of Pharmacology and Experimental Therapeutics 286:1391–1396. 11. Segawa H, Kaneko I, Takahashi A, Kuwahata M, Ito M, Ohkido I, Tatsumi S, Miyamoto K-i 2002 Growth-related renal type II Na/Pi cotransporter. J Biol Chem 277(22): 19665–19672. 12. Custer M, Lotscher M, Biber J, Murer H, Kaissling B 1994 Expression of Na-Pi cotransport in rat kidney: Localization by RT-PCR and immunohistochemistry. Am J Physiol 266:F767–F774. 13. Beck L, Karaplis AC, Amizuka N, Hewson AS, Ozawa H, Tenenhouse HS 1998 Targeted inactivation of Npt2 in mice leads to severe renal phosphate wasting, hypercalciuria, and skeletal abnormalities. Proc Natl Acad Sci USA 95(9):5372–7. 14. Kavanaugh MP, Miller DG, Zhang W, Law W, Kozak SL, Kabat D, Miller AD 1994 Cell-surface receptors for gibbon ape leukemia virus and amphotropic murine retrovirus are inducible sodium-dependent P symporters. Proc Natl Acad Sci USA 91:7071–7075. 15. Kempson SA, Lotscher M, Kaissling B, Biber J, Murer H, Levi M 1995 Parathyroid hormone action on phosphate transporter mRNA and protein in rat renal proximal tubules. Am J Physiol 268:F784–F791. 16. Levi M 2000 Npt2 is the major mediator of renal phosphate transport. Am J Dis Kidney 36:1276–1280. 17. Traebert M, Volkl H, Biber J, Murer H, Kaissling B 2000 Luminal and contraluminal action of 1-34 and 3-34 PTH peptides on renal type IIa Na-Pi cotransporter. Am J Physiol 278:F792–F798. 18. Bacic D, Hernando N, Traebert M, Lederer E, Volkl H, Biber J, Kaissling B, Murer H 2001 Regulation of the renal type IIa Na/Pi cotransporter by cGMP. Pflugers Arch 443:306–313. 19. Bacic D, Schulz N, Biber J, Kaissling B, Murer H, Wagner CA 2003 Involvement of the MAPK-kinase pathway in the PTH-mediated regulation of the proximal tubule type IIa Na(+)/P(i) cotransporter in mouse kidney. Pflugers Arch 446(1):52–60. 20. Yamashita T, Konishi M, Miyake A, Inui K, Itoh N 2002 Fibroblast growth factor (FGF)-23 inhibits renal phosphate reabsorption by activation of the mitogen-activated protein kinase pathway. J Biol Chem 277(31):28265–70. 21. Jankowski M, Hilfiker H, Biber J, Murer H 2001 The opossum kidney cell type IIa Na/P(i) cotransporter is a phosphoprotein. Kidney Blood Press Res 24:1–4. 22. Shenolikar S, Voltz JW, Minkoff CM, Wade JB, Weinman EJ 2002 Targeted disruption of the mouse NHERF-1 gene

23.

24. 25. 26.

27.

28.

29.

30.

31. 32. 33.

34. 35.

36.

37. 38.

39.

40.

promotes internalization of proximal tubule sodiumphosphate cotransporter type IIa and renal phosphate wasting. Proc Natl Acad Sci USA 99(17):11470–11475. Hernando N, Deliot N, Gisler SM, Lederer E, Weinman EJ, Biber J, Murer H 2002 PDZ-domain interactions and apical expression of type IIa Na/Pi cotransporters. Proc Natl Acad Sci USA 99:11957–11962. Bell NH 1988 Vitamin D metabolism in health and disease. Henry Ford Hosp Med J 36:40–52. Tanaka Y, DeLuca HF 1973 The control of 25-hydroxyvitamin D metabolism by inorganic phosphorus. Arch Biochem Biophys 154:566–574. Fukase M, Birge S, Rifas L, Avioli LV, Chase LR 1982 Regulation of 25 hydroxyvitamin D3 1-hydroxylase in serumfree monolayer culture of mouse kidney. Endocrinology 110:1073–1075. Gray RW, Garthwaite TL 1985 Activation of renal 1,25-dihyroxyvitamin D3 synthesis by phosphate deprivation: Evidence for a role for growth hormone. Endocrinology 116: 189–193. Gray RW, Lemann JJ 1985 Vitamin D metabolism: the renalpituitary axis. In: Norman AW, Schaefer K, Grigoleit H-G, Herrath Dv (eds.) Vitamin D—chemical, biochemical, and clinical update. de Gruyter: Berlin. Drezner MK 1987 Understanding the pathogenesis of Xlinked hypophosphatemic rickets: A requisite for successful therapy. In: Zackson DA (ed.) CPC series: Cases in metabolic bone disease. Triclinica Communications, New York, pp 1–11. Tenenhouse HS, Martel J, Gauthier C, Zhang MY, Portale AA 2001 Renal expression of the sodium/phosphate cotransporter gene, Npt2, is not required for regulation of renal 1 alpha-hydroxylase by phosphate. Endocrinology 142(3):1124–9. Spitzer A, Barac-Nieto M 2001 Ontogeny of renal phosphate transport and the process of growth. Pediatr Nephrol 16(9):763–71. Drezner MK 2002 Unpublished Observations. Cai Q, Hodgson SF, Kao PC, Lennon VA, Klee GG, Zinsmiester AR, Kumar R 1994 Brief report: inhibition of renal phosphate transport by a tumor product in a patient with oncogenic osteomalacia. N Engl J Med 330:1645–1649. Nelson AE 1996 Characteristics of tumor cell bioactivity in oncogenic osteomalacia. Mol Cell Biol 124:17–23. Wilkins GE, Granleese S, Hegele RG, Holden J, Anderson DW, Bondy GP 1995 Oncogenic osteomalacia: evidence for a humoral phosphaturic factor. J Clin Endocrinol Metab 80(5):1628–1634. Nelson AE, Mason RS, Hogan JJ, Diamond T, Robinson BG 1998 Tumor expression studies indicate that HEM-1 is unlikely to be the active factor in oncogenic osteomalacia. Bone 23(6):549–553. Econs MJ, Drezner MK 1994 Tumor-induced osteomalacia— unveiling a new hormone. N Engl J Med 330(23):1679–1681. Jan de Beur SM, Finnegan RB, Vassiliadis J, Cook B, Barberio D, Estes S, Manavalan P, Petroziello J, Madden S, Cho JY, Kumar R, Levine MA, Schiavi SC 2002 Tumors associated with oncogenic osteomalacia express markers of bone and mineral metabolism. J Bone Miner Res 17:1102–1110. Bowe AE, Finnegan R, Jan de Beur SM, Cho J, Levine MA, Kumar R, Schiavi SC 2001 FGF-23 inhibits renal tubular phosphate transport and is a PHEX substrate. Biochem Biophys Res Commun 284(4):977–81. White KE, Jonsson KB, Carn G, Hampson G, Spector TD, Mannstadt M, Lorenz-Depiereux B, Miyauchi A, Yang IM,

1182

41.

42.

43. 44.

45.

46.

47.

48.

49.

50.

51.

52. 53. 54.

Ljunggren O, Meitinger T, Strom TM, Juppner H, Econs MJ 2001 The autosomal dominant hypophosphatemic rickets (ADHR) gene is a secreted polypeptide overexpressed by tumors that cause phosphate wasting. J Clin Endocrinol Metab 86(2):497–500. Shimada T, Mizutani S, Muto T, Yoneya T, Hino R, Takeda S, Takeuchi Y, Fujita T, Fukumoto S, Yamashita T 2001 Cloning and characterization of FGF23 as a causative factor of tumorinduced osteomalacia. Proc Natl Acad Sci USA 98(11): 6500–6505. De Beur SM, Finnegan RB, Vassiliadis J, Cook B, Barberio D, Estes S, Manavalan P, Petroziello J, Madden SL, Cho JY, Kumar R, Levine MA, Schiavi SC 2002 Tumors associated with oncogenic osteomalacia express genes important in bone and mineral metabolism. J Bone Miner Res 17(6):1102–1110. 2000 Autosomal dominant hypophosphataemic rickets is associated with mutations in FGF23. The ADHR Consortium. Nat Genet 26(3):345–8. Econs MJ, McEnery PT 1997 Autosomal dominant hypophosphatemic rickets/osteomalacia: clinical characterization of a novel renal phosphate-wasting disorder. J Clin Endocrinol Metab 82(2):674–81. White KE, Carn G, Lorenz-Depiereux B, Benet-Pages A, Strom TM, Econs MJ 2001 Autosomal-dominant hypophosphatemic rickets (ADHR) mutations stabilize FGF-23. Kidney International 60(6):2079–86. Bai XY, Miao D, Goltzman D, Karaplis AC 2003 The autosomal dominant hypophosphatemic rickets R176Q mutation in fibroblast growth factor 23 resists proteolytic cleavage and enhances in vivo biological potency. J Biol Chem 278(11): 9843–9849. Shimada T, Yoneya T, Hino R, Takeuchi Y, Fukumoto S, Tamashita T 2001 Transgenic mice expressing fibroblast growth factor 23 demonstrate hypophosphatemia with low serum 1,25-dihydroxyvitamin D and rickets/osteomalacia. J Bone Miner Res 16(Suppl 1):S151. Bai XY, Miao DS, Li JR, Goltzman D, Karaplis AC 2003 Transgenic mice overexpressing human FGF-23 (R176Q) recapitulate a severe form of the ADHR phenotype. J Bone Miner Res 18:S24 (1089). Jonsson KB, Zahradnik R, Larsson T, White KE, Hampson G, Miyauchi A, Econs M, Lavigne J, Jüppner H 2002 FGF-23 is a circulating factor that is elevated in oncogenic osteomalacia and X-linked hypophosphatemic rickets. N Engl J Med 348: 1656–1663. Yamazaki Y, Okazaki R, Shibata M, Hasegawa Y, Satoh K, Tajima T, Takeuchi Y, Fujita T, Nakahara K, Yamashita T, Fukumoto S 2002 Increased circulatory level of biologically active full-length FGF-23 in rickets/osteomalacia. J Clin Endocrinol Metab 87:4957–4960. Riminucci M, Collins MT, Fedarko NS, Cherman N, Corsi A, White KE, Waguespack S, Gupta A, Hannon T, Econs MJ, Bianco P, Gehron Robey P 2003 FGF-23 in fibrous dysplasia of bone and its relationship to renal phosphate wasting. J Clin Invest 112(5):683–692. Meyer RA, Jr., Meyer MH, Gray RW 1989 Parabiosis suggests a humoral factor is involved in X-linked hypophosphatemia in mice. J Bone Miner Res 4(4):493–500. Lajeunesse D, Meyer RA, Jr., Hamel L 1996 Direct demonstration of a humorally-mediated inhibition of renal phosphate transport in the Hyp mouse. Kidney Int 50(5):1531–8. Nesbitt T, Fujiwara I, Thomas R, Xiao ZS, Quarles LD, Drezner MK 1999 Coordinated maturational regulation of PHEX and renal phosphate transport inhibitory activity: evidence for the pathophysiological role of PHEX in X-linked hypophosphatemia. J Bone Miner Res 14(12):2027–35.

MARC K. DREZNER

55. Xiao ZS, Crenshaw M, Guo R, Nesbitt T, Drezner MK, Quarles LD 1998 Intrinsic mineralization defect in Hyp mouse osteoblasts. Am J Physiol 275(4 Pt 1):E700–8. 56. Nesbitt T, Coffman TM, Griffiths R, Drezner MK 1992 Crosstransplantation of kidneys in normal and Hyp mice. Evidence that the Hyp mouse phenotype is unrelated to an intrinsic renal defect. J Clin Invest 89(5):1453–9. 57. Thompson DL, Sabbagh Y, Tenenhouse HS, Roche PC, Drezner MK, Salisbury JL, Grande JP, Poeschla EM, Kumar R 2002 Ontogeny of Phex/PHEX protein expression in mouse embryo and subcellular localization in osteoblasts. J Bone Miner Res 17(2):311–20. 58. Beck L, Soumounou Y, Martel J, Krishnamurthy G, Gauthier C, Goodyer CG, Tenenhouse HS 1997 Pex/PEX tissue distribution and evidence for a deletion in the 3′ region of the Pex gene in X-linked hypophosphatemic mice. J Clin Invest 99(6):1200–9. 59. Meyer MH, Meyer RA, Jr. 2000 MRNA expression of Phex in mice and rats: the effect of low phosphate diet. Endocrine 13(1):81–7. 60. Zoidis E, Zapf J, Schmid C 2000 Phex cDNA cloning from rat bone and studies on phex mRNA expression: tissue-specificity, age-dependency, and regulation by insulin-like growth factor (IGF) I in vivo. Mol Cell Endocrinol 168(1–2):41–51. 61. Ruchon AF, Tenenhouse HS, Marcinkiewicz M, Siegfried G, Aubin JE, DesGroseillers L, Crine P, Boileau G 2000 Developmental expression and tissue distribution of Phex protein: effect of the Hyp mutation and relationship to bone markers. J Bone Miner Res 15(8):1440–50. 62. Ruchon AF, Marcinkiewicz M, Siegfried G, Tenenhouse HS, DesGroseillers L, Crine P, Boileau G 1998 Pex mRNA is localized in developing mouse osteoblasts and odontoblasts. J Histochem Cytochem 46(4):459–68. 63. Grieff M, Mumm S, Waeltz P, Mazzarella R, Whyte MP, Thakker RV, Schlessinger D 1997 Expression and cloning of the human X-linked hypophosphatemia gene cDNA. Biochem Biophys Res Commun 231(3):635–9. 64. 1995 A gene (PEX) with homologies to endopeptidases is mutated in patients with X-linked hypophosphatemic rickets. The HYP Consortium. Nat Genet 11(2):130–6. 65. Aono Y, Shimada T, Yamazaki Y, Hino R, Takeuchi Y, Fujita T, Fukumoto S, Nagano N, Wada M, Yamashita T 2003 The neutralization of FGF-23 ameliorates hypophosphatemia and rickets in hyp-mice. J Bone Miner Res 18(Suppl 2):S16 (1056). 66. Berndt T, Craig TA, Bowe AE, Vassiliadis J, Reczek D, Finnegan R, Jan de Beur SM, Schiavi SC, Kumar R 2003 Secreted frizzled-related protein 4 is a potent tumor-derived phosphaturic agent. J Clin Invest 112:785–794. 67. Rowe PS, de Zoysa PA, Dong R, Wang HR, White KE, Econs MJ, Oudet CL 2000 MEPE, a new gene expressed in bone marrow and tumors causing osteomalacia. Genomics 67:54–68. 68. Guo R, Rowe PSN, Liu S, Simpson LG, Xiao ZS, Quarles LD 2002 Inhibition of MEPE cleavage by Phex. Biochem Biophys Res Commun 297:38–45. 69. Bai X, Miao D, Panda D, Grady S, McKee MD, Goltzman D, Karaplis AC 2002 Partial Rescue of the Hyp Phenotype by Osteoblast-Targeted PHEX (Phosphate-Regulating Gene with Homologies to Endopeptidases on the X Chromosome) Expression. Mol Endocrinol 16(12):2913–2925. 70. Argiro L, Desbarats M, Glorieux FH, Ecarot B 2001 Mepe, the gene encoding a tumor-secreted protein in oncogenic hypophosphatemic osteomalacia, is expressed in bone. Genomics 74(3):342–351. 71. Petersen DN, Tkalcevic GT, Mansolf AL, Rivera-Gonzalez R, Brown TA 2000 Identification of osteoblast/osteocyte factor 45 (OF45), a bone-specific cDNA encoding an RGD-containing

CHAPTER 69 Clinical Disorders of Phosphate Homeostasis

72.

73.

74.

75. 76.

77. 78.

79.

80. 81. 82.

83.

84.

85.

86.

protein that is highly expressed in osteoblasts and osteocytes. J Biol Chem 275:36172–36180. Gowen LC, Petersen DN, Mansolf AL, Qi H, Stock JL, Tkalcevic GT, Simmons HA, Crawford DT, Chidsey-Frink KL, Ke HZ, McNeish JD, Brown TA 2003 Targeted disruption of the osteoblast/osteocyte factor 45 gene (OF45) results in increased bone formation and bone mass. J Biol Chem 278(3): 1998–2007. Rowe PSN, Kumagai Y, Gutierrez G, Garrett IR, Blacher R, Rosen D, Chen D, Drezner MK, Quarles LD, Mundy GR 2003 MEPE regulates bone mineralization and phosphate transport: PHEX and the MEPE ASARM-peptide. Bone: In press. Liu S, Brown TA, Xiao Z, Guo R, Quarles LD 2003 Deletion of Mepe in hyp-mice fails to correct hypophosphatemia but partially rescues abnormal mineralization. J Bone Miner Res 18:In press. Guo R, Liu S, Spurney RF, Quarles LD 2001 Analysis of recombinant Phex: an endopeptidase in search of a substrate. Am J Physiol Endocrinol Metab 281(4):E837–E847. Campos M, Couture C, Hirata IY, Juliano MA, Loisel TP, Crine P, Juliano L, Boileau G, Carmona AK 2003 Human recombinant endopeptidase PHEX has a strict S1′ specificity for acidic residues and cleaves peptides derived from fibroblast growth factor-23 and matrix extracellular phosphoglycoprotein. Biochem J 373(Pt 1):271–279. Liu S, Guo R, Simpson LG, Xiao ZS, Burnham CE, Quarles LD 2003 Regulation of FGF23 expression but not degradation by phex. J Biol Chem 278(39):37419–37426. Collins MT, Chebli C, Jones J, Kushner H, Consugar M, Rinaldo P, Wientroub S, Bianco P, Robey PG 2001 Renal phosphate wasting in fibrous dysplasia of bone is part of a generalized renal tubular dysfunction similar to that seen in tumor-induced osteomalacia. Journal of Bone & Mineral Research 16(5):806–13. Yamamoto T, Miyamoto KI, Ozono K, Taketani Y, Katai K, Miyauchi A, Shima M, Yoshikawa H, Yoh K, Takeda E, Okada S 2001 Hypophosphatemic rickets accompanying McCune-Albright syndrome: evidence that a humoral factor causes hypophosphatemia. Journal of Bone & Mineral Metabolism 19(5):287–95. Guo R, Liu S, Spurney RF, Quarles LD 2001 Analysis of recombinant Phex: an endopeptidase in search of a substrate. Am J Physiol Endocrinol Metab 281:E837-E847. Liu S, Guo R, Tu Q, Quarles LD 2002 Overexpression of Phex in osteoblasts fails to rescue the Hyp mouse phenotype. Journal of Biological Chemistry 277(5):3686–97. Erben RG, Mayer D, Weber K, Johnson T, Nonsson K, Juppner H, Lanske B 2003 Ubiquitous overexpression of Phex does not fully rescue the Hyp mouse phenotype. J Bone Miner Res 18:S22 (1080). Fujiwara I, Thomas R, Horst RL, Aravindan RG, Drezner MK 2003 Aberrant post-transcriptional regulation of renal 25-hydroxyvitamin D-1a-hydroxylase activity in the hypmouse. J Bone Miner Res 18:434–442. Yuan B, Xing Y, Veber R, Drezner MK 2003 Abnormal translational regulation of renal 25-hydroxyvitamin D-1α-hydroxylase activity in the hyp-mouse. J Bone Miner Res 18(Suppl 2): S282–S283 (SU427). Lobaugh B, Burch WM, Jr, Drezner MK 1984 Abnormalities of vitamin D metabolism and action in the vitamin D resistant rachitic and osteomalacic diseases. In: Kumar R (ed.) Vitamin D. Martinus Nijhoff, Boston, pp 665–720. Harrison HE, Harrison HC, Lifshitz F, Johnson AD 1966 Growth disturbance in hereditary hypophosphatemia. Am J Dis Child 112:290–297.

1183 87. Haddad JG, Chyu KJ, Hahn TJ, Stamp TCB 1973 Serum concentrations of 25-hydroxyvitamin D in sex-linked hypophosphatemic vitamin D–resistant rickets. J Lab Clin Med 81:22–27. 88. Delvin EE, Glorieux FH 1981 Serum 1,25-dihydroxyvitamin D concentration in hypophosphatemic vitamin D–resistant rickets. Calcif Tissue Int 33(173–175). 89. Lyles KW, Clark AG, Drezner MK 1982 Serum 1,25-dihydroxyvitamin D levels in subjects with X-linked hypophosphatemic rickets and osteomalacia. Calcif Tissue Int 34(2): 125–30. 90. Lobaugh B, Drezner MK 1983 Abnormal regulation of renal 25-hydroxyvitamin D-1alpha-hydroxylase activity in the Xlinked hypophosphatemic mouse. J Clin Invest 71(2):400–3. 91. Nesbitt T, Drezner MK, Lobaugh B 1986 Abnormal parathyroid hormone stimulation of 25-hydroxyvitamin D-1 alphahydroxylase activity in the hypophosphatemic mouse. Evidence for a generalized defect of vitamin D metabolism. J Clin Invest 77(1):181–7. 92. Nesbitt T, Lobaugh B, Drezner MK 1987 Calcitonin stimulation of renal 25-hydroxyvitamin D-1 alpha-hydroxylase activity in hypophosphatemic mice. Evidence that the regulation of calcitriol production is not universally abnormal in X-linked hypophosphatemia. J Clin Invest 79(1):15–9. 93. Nesbitt T, Drezner MK 1990 Abnormal parathyroid hormone-related peptide stimulation of renal 25-hydroxyvitamin D-1-hydroxylase in Hyp mice: evidence for a generalized defect of enzyme activity in the proximal convoluted tubule. Endocrinology 127(2):843–8. 94. Bell CL, Tenenhouse HS, Scriver CR 1988 Primary cultures of renal epithelial cells from X-linked hypophosphatemic (Hyp) mice express defects in phosphate transport and vitamin D metabolism. Am J Hum Genet 43(3):293–303. 95. Dobre CV, Alvarez UM, Hruska KA 1990 Primary culture of hypophosphatemic proximal tubule cells express defective adaptation to P. J Bone Miner Res 5(Suppl1):S205. 96. Collins JF, Ghishan FK 1994 Molecular cloning, functional expression, tissue distribution, and in situ hybridization of the renal sodium phosphate (Na+/P(i)) transporter in the control and hypophosphatemic mouse. Faseb J 8(11): 862–8. 97. Tenenhouse HS, Werner A, Biber J, Ma S, Martel J, Roy S, Murer H 1994 Renal Na+-phosphate cotransport in murine X-linked hypophosphatemic rickets: molecular characterization. J Clin Invest 93:671–676. 98. Tenenhouse HS, Martel J, Biber J, Murer H 1995 Effect of P(i) restriction on renal Na(+)-P(i) cotransporter mRNA and immunoreactive protein in X-linked Hyp mice. Am J Physiol 268(6 Pt 2):F1062–F1069. 99. Meyer RA, Jr., Meyer MH, Gray RW 1989 Parabiosis suggests a humoral factor is involved in X-linked hypophosphatemia in mice. J Bone Miner Res 4:493–500. 100. Meyer RA, Jr., Tenenhouse HS, Meyer MH, Klugerman AH 1989 The renal phosphate transport defect in normal mice parabiosed to X-linked hypophosphatemic mice persists after parathyroidectomy. J Bone Miner Res 4:523–532. 101. Nesbitt T, Econs MJ, Byun JK, Martel J, Tenenhouse HS, Drezner MK 1995 Phosphate transport in immortalized cell cultures from the renal proximal tubule of normal and Hyp mice: evidence that the HYP gene locus product is an extrarenal factor. J Bone Miner Res 10(9):1327–33. 102. Nesbitt T, Byun JK, Drezner MK 1996 Normal phosphate transport in cells from the S2 and S3 segments of Hypmouse proximal renal tubules. Endocrinology 137(3):943–8. 103. Kos CH, Tihy F, Econs MJ, Murer H, Lemieux N, Tenenhouse HS 1994 Localization of a renal sodium P

1184

104.

105.

106.

107.

108.

109.

110.

111.

112.

113. 114.

115. 116.

117.

cotransporter gene to human chromosome 5q35. Genomics 19:176–177. Read AP, Thakker RV, Davies KE, Mountford RC, Brenton DP, Davies M, Glorieux F, Harris R, Hendy GN, King A, McGlade S, Peacock CJ, Smith R, O’Riordan JLH 1986 Mapping of human X-linked hypophosphataemic rickets by multilocus linkage analysis. Hum Genet 73(3):267–70. Machler M, Frey D, Gal A, Orth U, Wienker TF, Fanconi A, Schmid W 1986 X-linked dominant hypophosphatemia is closely linked to DNA markers DXS41 and DXS43 at Xp22. Hum Genet 73(3):271–5. Thakker RV, Read AP, Davies KE, Whyte MP, Weksberg R, Glorieux F, Davies M, Mountford RC, Harris R, King A, et al. 1987 Bridging markers defining the map position of X-linked hypophosphataemic rickets. J Med Genet 24(12): 756–60. Thakker RV, Davies KE, Read AP, Tippett P, Wooding C, Flint T, Wood S, Kruse TA, Whyte MP, O’Riordan JLH 1990 Linkage analysis of two cloned DNA sequences, DXS197 and DXS207, in hypophosphatemic rickets families. Genomics 8(2):189–193. Econs MJ, Fain PR, Norman M, Speer MC, Pericak-Vance MA, Becker PA, Barker DF, Taylor A, Drezner MK 1993 Flanking markers define the X-linked hypophosphatemic rickets gene locus. J Bone Miner Res 8(9):1149–52. Econs MJ, Rowe PS, Francis F, Barker DF, Speer MC, Norman M, Fain PR, Weissenbach J, Read A, Davis KE, Becker PA, Lehrach H, O’Riordan J, Drezner MK 1994 Fine structure mapping of the human X-linked hypophosphatemic rickets gene locus. J Clin Endocrinol Metab 79(5):1351–4. Econs MJ, Francis F, Rowe PSN, Speer MC, Pericak-Vance MA, Becker PA, O’Riordan J, Lehrach H, Drezner MK 1994 Dinucleotide repeat polymorphism of the microsatellite repeat probe, A0563. Hum Mol Genet 3:680. Francis F, Rowe PS, Econs MJ, See CG, Benham F, O’Riordan JL, Drezner MK, Hamvas RM, Lehrach H 1994 A YAC contig spanning the hypophosphatemic rickets disease gene (HYP) candidate region. Genomics 21(1):229–37. Rowe PS, Goulding JN, Francis F, Oudet C, Econs MJ, Hanauer A, Lehrach H, Read AP, Mountford RC, Summerfield T, Weissenbach J, Fraser W, Drezner MK, Davies KE, O’Riordan JL 1996 The gene for X-linked hypophosphataemic rickets maps to a 200-300kb region in Xp22.1, and is located on a single YAC containing a putative vitamin D response element (VDRE). Hum Genet 97(3):345–52. Turner AJ, Tanzawa K 1997 Mammalian membrane metallopeptidases: NEP, ECE, KELL, and PEX. Faseb J 11(5): 355–64. Emoto N, Yanagisawa M 1995 Endothelian-converting enzyme-2 is a membrane-bound phosphramidon-sensitive metalloprotease with acidic pH optimum. J Biol Chem 270:15262–15268. Lee S, Zambas ED, Marsh WL, Redman CM 1991 Molecular cloning and primary structure of Kell blood group protein. Proc Natl Acad Sci USA 88:6353–6357. Kiryu-Seo S, Sasaki M, Yokohama H, Nakagomi S, Hirayama T, Aoki S, Wada K, Kiyama H 2000 Damageinduced neuronal endopeptidase (DINE) is a unique meallopeptidase expressed in response to neuronal damage and activates superoxide scavenge. Proc Natl Acad Sci USA 97:4345–4350. Shimada K, Takahashi M, Turner AJ, Tanzawa K 1996 Rat endothelin-converting enzyme-1 forms a dimer through

MARC K. DREZNER

118.

119.

120.

121.

122.

123.

124.

125.

126.

127.

128.

129. 130.

Cys412 with a similar catalytic mechanism and a distinct substrate-binding mechanism compared with neutral endopeptidase-24.11. J Biochem 315:863–867. D’Adamio L, Shipp MA, Masteller EL, Reinherz EL 1989 Organization of the gene encoding common acute lymphoblastic leukemia antigen (neutral endopeptidase 24.11): multiple miniexons and separate 5′ untranslated regions. Proc Natl Acad Sci USA 86:7103–7107. Xu D, Emoto N, Giald A, Slaughter C, Kaw S, deWit D, Yanagisawa M 1994 ECE-1: a membrane-bound metalloprotease that catalyzes the proteolytic activation of big endothelin-1. Cell 78:473–485. Russell F, Davenport A 1999 Evidence for intracellular endothelin-converting enzyme-2 expression in cultured human vascular endothelial cells. Circulation Research 84:891–896. Valdenaire O, Richards JG, Faull RLM, Schweizer A 1999 XCE, a new member of the endothelin-converting enzyme and neutral endopeptidase family, is preferentailly expressed in the CNS. Brain Res Mol Brain Res 64:211–221. Ghaddar G, Ruchon AF, Carpentier M, Marcinkiewicz M, Seidah NG, Crine P, Desgroseillers L, Boileau G 2000 Molecular cloning and biochemical characterization of a new mouse testis soluble-zinc-metallopeptidase of the neprilysin family. Biochem J 347(Pt 2):419–29. Valdenaire O, Rohrbacher E, Langeveld A, Schweizer A, Meijers C 2000 Organization and chromosomal localization of human ECEL1 (XCE) gene encoding a zinc metallopeptidase involved in the nervous control of respiration. Biochemistry Journal 346:611–616. Lipman ML, Panda D, Bennett HP, Henderson JE, Shane E, Shen Y, Goltzman D, Karaplis AC 1998 Cloning of human PEX cDNA. Expression, subcellular localization, and endopeptidase activity. J Biol Chem 273(22):13729–37. Strom TM, Francis F, Lorenz B, Boddrich A, Econs MJ, Lehrach H, Meitinger T 1997 Pex gene deletions in Gy and Hyp mice provide mouse models for X-linked hypophosphatemia. Hum Mol Genet 6(2):165–71. Du L, Desbarats M, Viel J, Glorieux FH, Cawthorn C, Ecarot B 1996 cDNA cloning of the murine Pex gene implicated in X-linked hypophosphatemia and evidence for expression in bone. Genomics 36(1):22–8. Francis F, Strom TM, Hennig S, Boddrich A, Lorenz B, Brandau O, Mohnike KL, Cagnoli M, Steffens C, Klages S, Borzym K, Pohl T, Oudet C, Econs MJ, Rowe PS, Reinhardt R, Meitinger T, Lehrach H 1997 Genomic organization of the human PEX gene mutated in X-linked dominant hypophosphatemic rickets. Genome Res 7(6):573–85. Rowe PS, Oudet CL, Francis F, Sinding C, Pannetier S, Econs MJ, Strom TM, Meitinger T, Garabedian M, David A, Macher MA, Questiaux E, Popowska E, Pronicka E, Read AP, Mokrzycki A, Glorieux FH, Drezner MK, Hanauer A, Lehrach H, Goulding JN, O’Riordan JL 1997 Distribution of mutations in the PEX gene in families with X-linked hypophosphataemic rickets (HYP). Hum Mol Genet 6(4): 539–49. Holm IA, Huang X, Kunkel LM 1997 Mutational analysis of the PEX gene in patients with X-linked hypophosphatemic rickets. Am J Hum Genet 60(4):790–7. Dixon PH, Christie PT, Wooding C, Trump D, Grieff M, Holm I, Gertner JM, Schmidtke J, Shah B, Shaw N, Smith C, Tau C, Schlessinger D, Whyte MP, Thakker RV 1998 Mutational analysis of PHEX gene in X-linked hypophosphatemia. J Clin Endocrinol Metab 83(10):3615–23.

CHAPTER 69 Clinical Disorders of Phosphate Homeostasis

131. Filisetti D, Ostermann G, von Bredow M, Strom T, Filler G, Ehrich J, Pannetier S, Garnier JM, Rowe P, Francis F, Julienne A, Hanauer A, Econs MJ, Oudet C 1999 Nonrandom distribution of mutations in the PHEX gene, and underdetected missense mutations at nonconserved residues. Eur J Hum Genet 7(5):615–9. 132. Tyynismaa H, Kaitila I, Nanto-Salonen K, Ala-Houhala M, Alitalo T 2000 Identification of fifteen novel PHEX gene mutations in Finnish patients with hypophosphatemic rickets. Hum Mutat 15(4):383–4. 133. Sabbagh Y, Boileau G, DesGroseillers B, Tenenhouse HS 2001 Turnover and rescue of mutant PHEX proteins sequestered in the endoplasmic reticulum. J Bone Miner Res 16(Suppl 1):S227. 134. Christie PT, Harding B, Nesbit MA, Whyte MP, Thakker RV 2001 X-linked hypophosphatemia attributable to pseudoexons of the PHEX gene. J Clin Endocrinol Metab 86(8): 3840–4. 135. Econs MJ, Friedman NE, Rowe PS, Speer MC, Francis F, Strom TM, Oudet C, Smith JA, Ninomiya JT, Lee BE, Bergen H 1998 A PHEX gene mutation is responsible for adult-onset vitamin D–resistant hypophosphatemic osteomalacia: evidence that the disorder is not a distinct entity from Xlinked hypophosphatemic rickets. J Clin Endocrinol Metab 83(10):3459–62. 136. Blydt-Hansen TD, Tenenhouse HS, Goodyer P 1999 PHEX expression in parathyroid gland and parathyroid hormone dysregulation in X-linked hypophosphatemia. Pediatr Nephrol 13(7):607–11. 137. Sabbagh Y, Londowski JM, Mathieson D, Gauthier C, Boileau G, Tenenhouse HS, Poeschia EM, Kumar R 2000 Stable expression of PHEX in hypophosphatemic (Hyp) mouse osteoblasts using a viral vector partially corrects the mutant cell phenotype: implications for gene therapy. Journal of the American Society of Nephrology 11:413A. 138. Miyamura T, Tanaka H, Inoue M, Ichinose Y, Seino Y 2000 The effects of bone marrow transplantation on X-linked hypophosphatemic mice. J Bone Miner Res 15(8): 1451–8. 139. Bhargava A, Xing Y, Drezner MK 2003 Abnormal phosphate transport in hyp-mouse osteoblasts: a possible factor in the bone mineralization defect. J Bone Miner Res In press. 140. Kumar R, Riggs BL 1980 Pathologic physiology of bone. In: Urist MR (ed.) Fundamental and Clinical Bone Physiology. J.B. Lippincott Company: Philadelphia, pp 394–405. 141. Florentin D, Sassi A, Roques BP 1984 A highly sensitive fluorimetric assay for “enkephalinase,” a neutral metalloendopeptidase that releases Tyr-Gly-Gly from enkephalins. Anal Biochem 141:62–69. 142. Koehn JA, Norman JA, Jones BN, LeSueur L, Sakane Y, Ghai RD 1987 Degradation of atrial natriuretic factor by kidney cortex membranes. J Biol Chem 262:11623–11627. 143. Jalal F, Lemay G, Zollinger M, Berthelot A, Boileau G, Crine P 1991 Neutral endopeptidase, a major brush border protein of the kidney proximal nephron, is directly targeted to the apical domain when expressed in Madrin-Darbey kidney cells. J Biol Chem 266:19826–19857. 144. Mari B, Checler F, Ponzio G, Peyron JF, Manie S, Farahifar D, Rossi B, Auberger P 1992 Jurkat T cells express a functional neutral endopeptidase activity (CALLA) involved in T cell activation. EMBO J 11:3875–3885. 145. Marie PJ, Glorieux FH 1983 Relation between hypomineralized periosteocytic lesions and bone mineralization in vitamin D–resistant rickets. Calcif Tissue Int 35:443–448.

1185 146. Rifas L, Cheng S, Halstead LR, Gupta A, Hruska KA, Avioli LV 1997 Skeletal case in kinase activity defect in the HYP mouse. Calcif Tissue Int 61(3):256–9. 147. Guo R, Quarles LD 1997 Cloning and sequencing of human PEX from a bone cDNA library: evidence for its developmental stage-specific regulation in osteoblasts. J Bone Miner Res 12(7):1009–17. 148. Friedman NE, M.K. D 1991 Genetic Osteomalacia. In: Bardin CW (ed.) Current Therapy in Endocrinology and Metabolism, 4th Edition. BC Decker, Inc. Philadelphia, pp 421–428. 149. Glorieux FH, Marie PJ, Pettifor JM, Delvin EE 1980 Bone response to phosphate salts, ergocalciferol, and calcitrol in hypophosphatemic vitamin D–resistant rickets. N Engl J Med 303:1023–1031. 150. Friedman NE, Lobaugh B, Drezner MK 1993 Effects of calcitriol and phosphorus therapy on the growth of patients with X-linked hypophosphatemia. J Clin Endocrinol Metab 76(4):839–44. 151. Saggerve G, Baronelli G, Butelloni S, Perri G 1995 Long-term growth hormone treatment in children with renal hypophosphatemic rickets: effects on growth, mineral metabolism, and bone density. J Pediatr 127:395–402. 152. Saggese G, Baroncelli GI, Barsanti S 1998 [Growth hormone treatment of familial hypophosphatemic rickets]. Arch Pediatr 5(Suppl 4):360S–363S. 153. Tieder M, Modai D, Samuel R, Arie R, Halabi A, Bab I, Gabizon D, Liberman UA 1985 Hereditary hypophosphatemic rickets with hypercalciuria. N Engl J Med 312:611–617. 154. Sermet-Gaudelus I, Garabedian M, Dechaux M, Lenoir G, Rey J, M. T 2001 Hereditary hypophosphatemic rickets with hypercalciuria: report of a new kindred. Nephron 88:83–86. 155. Tieder M, Arie R, Bab I, Maor J, Liberman, UA 1992 A new kindred with hereditary hypophosphatemic rickets with hypercalciuria: implications for correct diagnosis and treatment. Nephron 62:176–181. 156. Gazit D, Tieder M, Liberman UA, Passi-Even L, IA. B 1991 Osteomalacia in hereditary hypophosphatemic rickets with hypercalciuria: a correlative clinical-histomorphometric study. J Clin Endocrinol Metab 72:229–235. 157. Tieder M, Modai D, Shaked U, Samuel R, Arie R, Halabe A, Maor J, Weissgarten J, Averbukh Z, Cohen N 1987 “Idiopathic” hypercalciuria and hereditary hypophosphatemic rickets. Two phenotypical expressions of a common genetic defect. N Engl J Med 316:125–129. 158. Liberman UA 1988 Inborn errors in vitamin D metabolism— Their contribution to the understanding of vitamin D metabolism. In: Norman AW, Schaefer K, Grigoleit H-G, Herrath D (eds.) Vitamin D molecular, cellular, and clinical endocrinology. Walter de Gruyter: Berlin, pp 935–947. 159. Jones A, Tzenova J, Frappier D, Crumley M, Roslin N, Kos C, Tieder M, Langman C, Proesmans W, Carpenter T, Rice A, Anderson D, Morgan K, Fujiwara T, H. T 2001 Hereditary hypophosphatemic rickets with hypercalciuria is not caused by mutations in the Na/Pi cotransporter NPT2 gene. J Am Soc Nephrol 12:507–514. 160. Chan JCM, Alon U 1985 Tubular disorders of acid-base and phosphate metabolism. Nephron 40:257–279. 161. Chesney RW 1990 Fanconi syndrome and renal tubular acidosis. In: Favus MJ (ed.) Primer on Metabolic Bone Diseases and Disorders of Mineral Metabolism, 1st ed. Am Soc Bone Miner Res, Kelseyville, CA. 162. De Toni G 1933 Remarks on the relations between renal rickets (renal dwarfism) and renal diabetes. Acta Paediatr Scand 16:479–484.

1186 163. McCune DJ, Mason HH, Clarke HT 1943 Intractable hypophosphatemic rickets with renal glycosuria and acidosis (the Fanconi syndrome). Am J Dis Child 65:81–146. 164. Bergeron M, Gougoux A, Vinay P 1995 The Renal Fanconi Syndrome. In: Scriver CR, Beaudet AL, Sly WS, Valle D (eds.) The metabolic and molecular bases of inherited disease, 7th ed. McGraw-Hill, New York, pp 3691–3704. 165. Chesney RW, Rosen JF, Hamstra AJ, DeLuca HF 1980 Serum 1,25-dihydroxyvitamin D levels in normal children and in vitamin D disorders. Am J Dis Child 134:135–139. 166. Steinherz R, Chesney RW, Schulman JD, DeLuca HF, Phelps M 1983 Circulating vitamin D metabolites in nehphropathic cystinosis. J Pediatr 102(592–294). 167. Chesney RW, Kaplan BS, Phelps M, DeLuca HF 1984 Renal tubular acidosis does not alter circulating values of calcitriol. J Pediatr 104:51–55. 168. Chevalier RL 1983 Hypercalciuria in a child with primary Fanconi syndrome and hearing loss. Int J Pediatr Nephrol 4:53–57. 169. Lee SW, Russell J, Avioli LV 1977 25-hydroxycholecalciferol to 1,25-dihydroxycholecalciferol: conversion impaired by systemic metabolic acidosis. Science 195:994–996. 170. Brewer ED, Tsai HC, Szeto KS, Morris RC 1977 Maleic acid induced impaired conversion of 25(OH)D3 to 1,25(OH)2D3: implications for Fanconi’s syndrome. Kidney Int 12: 244–252. 171. Brenner RJ, Spring DB, Sebastian A, McSherry EM, Genant HK, Palubinskas AJ, Morris RC 1982 Incidence of radiographically evident bone disease, nephrocalcinosis and nephrolithiasis in various types of renal tubular acidosis. N Engl J Med 307:217–221. 172. Tieder M, Arie R, Modai D, Samuel R, Weissgarten J, Liberman UA 1988 Elevated serum 1,25-dihydroxyvitamin D concentrations in siblings with primary Fanconi’s syndrome. N Engl J Med 319:845–849. 173. Schneider JA, Schulman JD 1983 Cystinosis. In: Stanbury JB, Wyngaarden JB, Fredrickson DS, Goldstein JL, Brown MS (eds.) The Metabolic Basis of Inherited Disease, 5th ed. McGraw-Hill: New York, pp 1844–1866. 174. Scheinman SJ, Pook MA, Wooding C, Pang JT, Frymoyer PA, Thakker RV 1997 Mapping the gene causing X-linked recessive nephrolithiasis to Xp11.22 by linkage studies. J Clin Invest 91:2351–2357. 175. Scheinman SJ 1998 X-linked hypercalciuric nephrolithiasis: clinical syndromes and chloride channel mutations. Kidney Int 53:3–17. 176. Lyles KW, Burkes EJ, Ellis GJ, Lucas KJ, Dolan EA, Drezner MK 1985 Genetic transmission of tumoral calcinosis: Autosomal dominant with variable clinical expressivity. J Clin Endocrinol Metab 60:1093–1096. 177. Prince MJ, Schaefer PC, Godsmith RS, Chausmer AB 1982 Hyperphosphatemic tumoral calcinosis: Association with elevation of serum 1,25, dihydroxycholecalciferol concentrations. Ann Inter Med 96:586–591. 178. Mozaffarian G, Lafferty FW, Pearson OH 1972 Treatment of tumoral calcinosis with phosphorus deprivation. Ann Inter Med 77:741–745. 179. Gregosiewicz A, Warda E 1989 Tumoral calcinosis: Successful medical treatment. J Bone Joint Surg Am 71A:1244–1249. 180. Yamaguchi T, Sugimoto T, Imai Y, Fukase M, Fujita T, Chihara K 1995 Successful treatment of hyperphosphatemic tumoral calcinosis with long-term acetazolamide. Bone 16:247S–250S.

MARC K. DREZNER

181. Salvi A, Cerudelli B, Cimino A, Zuccato F, Giustina G 1983 Phosphaturic action of calcitonin in pseudotumoral calcinosis. Horm Metab Res 15:260. 182. Levine BS, Ho LD, Pasiecznik K, Coburn JV 1986 Renal adaptation to phosphorus deprivation. J Bone Miner Res 1:33–40. 183. Lotz M, Zisman E, Bartter FC 1968 Evidence for a phosphorus-depletion syndrome. N Engl J Med 278:409–415. 184. Baker LRI, Ackrill P, Cattell WR, Stamp TC, Watson L 1974 Iatrogenic osteomalacia and myopathy due to phosphate depletion. British Medical Journal 3:150–152. 185. Mostellar ME, Tuttle EP 1964 Effects of alkalosis on plasma concentration and urinary excretion of inorganic phosphate in man. J Clin Invest 43:138–145. 186. Knochel JP 1977 The pathophysiology and clinical characteristics of severe hypophosphatemia. Arch Int Med 137: 203–220. 187. Sheldon GF, Grzyb S 1975 Phosphate depletion and repletion: relation to parenteral nutrition and oxygen transport. Ann Surg 182:683–689. 188. Grossman RA, Hamilton RW, Morse BM, Penn AS, Goldberg M 1974 Nontraumatic rhabdomyolysis and acute renal failure. N Engl J Med 291:807–811. 189. Koffler A, Fnedler RM, Massry SG 1976 Acute renal failure due to nontraumatic rhabdomyolysis. Ann Inter Med 85: 23–28. 190. Zusman J, Brown DM, Nesbitt ME 1973 Hyperphosphatemia, hyperphosphaturia, and hypocalcemia in acute Iymphoblastic leukemia. N Engl J Med 289:1335–1337. 191. Denborough MA, Forster JFA, Hudson MC, Carter NG, Zapf P 1970 Biochemical changes in malignant hyperpyrexia. Lancet 1:1137–1138. 192. Duhm J 1971 2,3-DPG-induced displacements of the oxyhemoglobin dissociation curve of blood: mechanisms and consequences. Adv Exp Med Biol 37A:179–186. 193. Travis SF, Sugerman HJ, Ruberg RL, Dudrick SJ, DelivoriaPapadopoulos M, Miller LD, Oski FA 1977 Alterations of red cell glycolytic intermediates and oxygen transport as a consequence of hypophosphatemia in patients receiving intravenous hyperalimentation. N Engl J Med 297:901–904. 194. Parfitt AM, Kleerekoper M 1980 Clinical disorders of calcium, phosphorus and magnesium metabolism. In: Maxwell MH, Kleeman CR (eds.) Clinical disorders of fluid and electrolyte metabolism. McGraw-Hill: New York, pp 947–1151. 195. O’Connor LR, Wheeler WS, Bethune JE 1977 Effect of hypophosphatemia on myocardial performance in man. N Engl J Med 297:901–903. 196. Newman JH, Neff TA, Ziporen P 1977 Acute respiratory failure associated with hypophosphatemia. N Engl J Med 296:1101–1103. 197. Klock JC, Williams HE, Mentzer WK 1974 Hemolytic anemia and somatic cell dysfunction in severe hypophosphatemia. Arch Intern Med 134:360–364. 198. Glorieux FH, Scriver CR, Reade TM, Goldman H, Roseborough A 1972 Use of phosphate and vitamin D to prevent dwarfism and rickets in X-linked hypophosphatemia. N Engl J Med 287(10):481–7. 199. Craddock PR, Yawota Y, Van Santen L, Gilberstadt S, Silivis S, Jacob HS 1974 Acquired phagocyte dysfunction: a complication of the hypophosphatemia of parenteral hyperalimentation. N Engl J Med 290:1403–1407. 200. Herbert LA, Lemann J, Peterson JR, Lennon EJ 1966 Studies of the mechanism by which phosphate infusion lowers serum calcium concentration. J Clin Invest 45:1886–1891.

CHAPTER 69 Clinical Disorders of Phosphate Homeostasis

201. Sinha TK, Allen DO, Queener SF, Bell NH, Larson S, McClintock R 1977 Effects of acetazolamide on the renal excretion of phosphate in hyperparathyroidism and pseudohypoparathyroidism. J Lab Clin Med 89: 1188–1197.

1187 202. Econs MJ, Drezner MK 1992 Bone Disease Resulting from Inherited Disorders of Renal Tubule Transport and Vitamin D Metabolism. In: Coe FL, Favus MJ (eds.) Disorders of Bone and Mineral Metabolism. Raven Press, Ltd: New York, pp 935–950.

CHAPTER 70

Disorders of Phosphate Metabolism: Autosomal Dominant Hypophosphatemic Rickets, Tumor Induced Osteomalacia, Fibrous Dysplasia, and the Pathophysiological Relevance of FGF23 MICHAEL J. ECONS

I. II. III. IV. V.

Departments of Medicine and Medical and Molecular Genetics Indiana University School of Medicine, Indianapolis, IN, 46202

Introduction ADHR Tumor Induced Osteomalacia Fibrous Dysplasia The Role of FGF23 in XLH

I. INTRODUCTION Disorders of phosphate homeostasis are not only clinically interesting, but recent data concerning these disorders sheds light on maintenance of normal phosphate homeostasis. Chapter 69 summarized recent work in X-linked hypophosphatemic rickets. This chapter will review the disorders autosomal dominant hypophosphatemic rickets (ADHR) and tumor induced osteomalacia (TIO) and the pathogenesis of hypophosphatemia in fibrous dysplasia. The relevance of FGF23 in the pathogenesis of these disorders, as well as XLH, will also be reviewed.

II. ADHR A. Clinical Features Autosomal dominant hypophosphatemic rickets (ADHR, MIM#193100) is characterized by isolated renal phosphate wasting and inappropriately normal concentrations of calcitriol. Patients frequently present with bone pain, rickets, and tooth abscesses. Bianchine et al. described a small ADHR family [1]. The father, who was markedly affected, had isolated renal phosphate wasting, short stature, and lower extremity deformity VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

VI. FGF23 in Health and Its Potential Role in Maintenance of Normal Phosphate and Vitamin D Homeostasis VII. Summary References

from rickets. He had two affected daughters and one affected son. These investigators reported that the father had a marked tendency towards fracture. Otherwise, the clinical course in these individuals appeared to be similar to that of XLH patients. However, the family was too small to fully appreciate the manifestations of the disorder and its incomplete penetrance. We evaluated a large ADHR kindred with many affected individuals [2]. This kindred provided us with an opportunity to explore the phenotypic variability of this disease in a large number of individuals with the same mutation. There was no evidence of genetic anticipation or imprinting. In contrast to XLH (see Chapter 69), ADHR displays incomplete penetrance and delayed onset of penetrance. The family contains two subgroups of affected individuals. One subgroup consists of patients who presented during childhood with phosphate wasting, rickets, and lower extremity deformity in a pattern similar to the classic presentation of XLH. The second group consists of individuals who presented as adolescents or adults. These individuals complained of bone pain, fatigue, weakness, and insufficiency fractures, but did not have lower extremity deformities [2]. In some cases, patients were wheelchair or bed bound from bone pain and weakness. Their clinical presentations were essentially identical to patients who present Copyright © 2005, Elsevier, Inc. All rights reserved.

1190 with tumor induced osteomalacia (TIO, see below), although none of the ADHR patients were ever found to develop tumors. Of note, all of the individuals who presented with delayed onset of penetrance are female. In subsequent observations, all new patients that have presented with delayed onset of ADHR have also been women (unpublished observations). Furthermore, two male patients presented in childhood with renal phosphate wasting and radiographic evidence of rickets and were treated for several years. They were subsequently taken off treatment for several years. On subsequent reevaluation, they were found to be clinically asymptomatic and have normal serum phosphorus and TMP/ GFR determinations [2]. In addition to the patients noted above, we found unaffected individuals who are carriers for the ADHR mutation [2]. Thus, the clinical manifestations of ADHR are even more variable than those observed in XLH.

B. Identification of the FGF23 Gene To identify the gene responsible for ADHR, we performed a genome wide linkage scan in a large ADHR kindred. Analysis of these results demonstrated that the gene was located on chromosome 12p13 [3]. Further fine mapping experiments limited the critical region to approximately 1.5 Mb. Using a variety of exon prediction programs, other informatics based approaches, and RT-PCR, we identified several novel and known genes in the critical region. We found mutations in four ADHR families in a novel gene, subsequently named FGF23 [4]. Mutation detection studies in the four ADHR kindreds revealed three missense changes affecting two arginines, residing three amino acids apart. Two unrelated families had the same change, R176Q (527G>A). One family had an R179W (535C>T) change and another had an R179Q (536G >A) substitution [4]. The mutations segregated with the disease in each family and were not found in normal individuals. In addition to the ADHR families, we performed mutation detection in index cases from a family with Hypophosphatemic Bone Disease [5], two families with tumoral calcinosis, a family with hypophosphatemia and multiple congenital abnormalities, as well as in 18 hypophosphatemia patients without PHEX mutations. Sequencing of the entire coding region, 880 bp upstream of the initiation codon, 1873 bp downstream of the coding sequence, and both of the predicted polyadenylation sites did not reveal mutations in any of these diseases except the ADHR patients. FGF23 is a 251 amino acid protein and shares greatest similarity with FGF19 and FGF21. Although FGF receptor mutations are known to cause human diseases,

MICHAEL J. ECONS

until recently, when FGF14 were found to cause cerebral axatia [6], FGF23 was the only FGF implicated in human disease. The first 24 amino acids are the signal peptide and in vitro studies demonstrate that FGF23 is rapidly secreted [4,7]. To determine the mechanism of how missense mutations in FGF23 cause ADHR, we performed transient transfections with the native FGF23 cDNA, as well as cDNAs carrying the three disease causing mutations. Antibodies directed toward the C-terminal portion of FGF23 revealed that the native FGF23 protein resolved as 32 kD and 12 kD species in conditioned media; however, the three mutated proteins were detected only as the 32 kD band. An N-terminal FLAG-tagged native FGF23 resolved as two bands of 36 kD (32 kD+FLAG) and 26 kD when detected with a FLAG antibody, whereas the R176Q mutant resolved primarily as the 36 kD protein species (Fig. 1) [8]. This work has subsequently been confirmed by several investigators [9,10]. Cleavage of FGF23 was not enhanced by extracellular incubation of FGF23 with HEK293 cells, and native and mutant FGF23s bound heparin [8]. Therefore, FGF23 proteins containing the ADHR mutations are secreted and produce polypeptides less sensitive to protease cleavage than wild type FGF23. Thus, the ADHR mutations protect FGF23 from proteolysis, thereby potentially elevating circulating concentrations of intact FGF23 and leading to phosphate wasting in ADHR patients. FGF23 is expressed at low levels in normal tissues. Originally, northern blots from multiple tissues failed to reveal expression. However, initial RT-PCR from RNA from a variety of tissues indicates that FGF23 is expressed in heart, liver, and thyroid/parathyroid [4], and additional reports indicate that the gene is also expressed in thymus [11]. Recent studies [12] indicate that FGF23 is expressed in bone in bone marrow stromal cells, osteoblasts, and osteocytes. Expression appears to be most prominent in active bone, such as during fracture repair.

III. TUMOR INDUCED OSTEOMALACIA Tumor induced osteomalacia (TIO), also called oncogenic hypophosphatemic osteomalacia, is an uncommon disorder that results from tumor secretion of a substance or substances, often referred to as phosphatonin(s) [13]. The clinical picture is somewhat dependent on age of presentation, but patients frequently present with proximal muscle weakness, bone pain, and fatigue. In some cases the weakness can be quite profound as the disease progresses. Children who have open epiphyses will have rickets with resulting

1191 F R LAG 17 6Q

F FG LAG F2 3

CHAPTER 70 ADHR, TIO, Fibrous Dysplasia, and FGF23

−208 −131 −96

−43.9 36−

−35

26− −17.8

−8.3

FGF-23 Protein:

Anti-FGF23 Epitope RHTR S

FLAG Epitope

FIGURE 1 Western analysis demonstrating that an N-terminal FLAG-tagged native FGF23 with two bands (intact and cleaved fragment) while the R176Q mutant resolved primarily as the full length 36 kD protein species. A schematic showing the relative positions of the FLAG and FGF-23 antibody epitopes, as well as the cleavage site is shown below the Western analysis. From reference 8 with permission.

lower extremity deformities. Adults often develop fractures and pseudofractures. Of note, the clinical picture in adults is essentially identical to that seen in ADHR patients with adult onset of disease. Unfortunately, delayed diagnosis of TIO is common, and patients may become bedridden due to fractures and muscle weakness by the time the diagnosis is established [14]. Hypophosphatemia results from renal phosphate wasting as evidenced by a reduced TMP/GFR and calcitriol concentration is either inappropriately normal or low. Compared with XLH and ADHR, where calcitriol levels tend to be inappropriately normal, calcitriol levels in TIO are more often below the normal range. A wide variety of tumors have been implicated in causing the disease [15], although the tumors are frequently of mesenchymal origin. Recent studies [16] demonstrate that misclassification of these tumors is common. Upon reevaluation by a pathologist familiar with TIO, 24 of 29 tumors were classified as phosphaturic mesenchymal tumor (mixed connective tissue variant) (PMTMCT) [16].

Tumor removal, if the tumor can be found, results in complete remission of the phosphate wasting, vitamin D abnormalities, and osteomalacia. Unfortunately, even tumors that were originally classified as nonmalignant can reoccur, sometimes many years after successful removal of the original tumor [17]. Although there has been some success in locating these tumors with octreatide scans [18] and other functional imaging modalities, the tumors are frequently small and very difficult to locate. Many tumors are not found, and patients with presumed TIO frequently require treatment with high dose calcitriol and phosphate, using an identical protocol to that used for XLH patients (see Chapter 69). In light of the clinical similarity between ADHR and tumor induced osteomalacia (TIO), we examined tumor tissue for FGF23 expression. These studies demonstrated that tumors that cause renal phosphate wasting markedly overexpress FGF23 [7]. The level of FGF23 expression in these tumors is several orders of

1192 magnitude higher than in normal tissues. These results have been confirmed by other investigators [19] in subsequent studies. Shimada et al. [19] also found that FGF23 was highly expressed in tumors that cause phosphate wasting. They found that injection of FGF23 into mice resulted in renal phosphate wasting. Furthermore, mice implanted with FGF23 overexpressing cells manifested renal phosphate wasting, inappropriately low calcitriol concentrations, and rickets [19]. Essentially, these mice are models of tumor induced osteomalacia. In subsequent studies [20], these investigators made FGF23 transgenic mice. FGF23 transgenic mice manifest renal phosphate wasting with reduced expression of the sodium dependent phosphate cotransporter, Npt2a. Calcitriol concentrations were markedly reduced, and the transgenic mice had osteomalacia [20]. Although current data implicate FGF23 as a “phosphatonin,” FGF23 is not the only gene that is overexpressed in these tumors. Jan de Beur et al. performed serial analysis of gene expression (SAGE) and found many genes that were differentially expressed between TIO tumors and other tumors, but matrix extracellular phosphoglycoprotein (MEPE), frizzled related protein 4 (FRP4), and FGF23 were prominently overexpressed in tumors that caused phosphate wasting [21]. Followup studies by several investigators are in progress, however, some data is currently available. MEPE is highly expressed in TIO tumors. However, the MEPE knockout mouse does not have a defect in phosphate homeostasis, but does have increased bone formation and increased mineralization, indicating that MEPE likely has a role in these processes [22]. Of note, implantation of MEPE overexpressing cells into nude mice did not result in defects in phosphate or vitamin D homeostasis [19]. However, MEPE is overexpressed in bone from the Hyp mouse compared to controls, indicating that it may have a role in defective bone mineralization in XLH [11]. Recently, Berndt et al. examined FRP4 as a candidate “phosphatonin”[23]. They reported that FRP4 inhibits phosphate transport in opossum kidney cells. Infusing FRP4 into rats results in phosphaturia without altering calcitriol concentrations. However, this group found that serum FRP4 concentrations in a TIO patient were normal. Thus, additional studies are necessary to fully elucidate the role of FRP4 in the pathogenesis of TIO.

IV. FIBROUS DYSPLASIA Fibrous dysplasia can be a component of McCune-Albright syndrome (characterized by fibrous dysplasia, hyperpigmented café-au-lait skin lesions, and hyperfunctioning endocrinopathies), or it can occur

MICHAEL J. ECONS

alone [24]. Weinstein and coworkers have shown that fibrous dysplasia results from activating somatic mutations of GNAS, which codes for the alpha subunit of the stimulatory G protein, Gs [25]. These lesions result from excessive proliferation and abnormal differentiation of mesenchymal osteoprogenitor cells. Fibrous dysplasia, particularly when associated with the McCunne Albright syndrome, can be accompanied by renal phosphate wasting and inappropriately normal calcitriol concentrations. Studies by Collins et al. [26] demonstrated that phosphate wasting occurred in approximately 50% of their patient population. Analysis of patients with hypophosphatemia due to fibrous dysplasia demonstrated that FGF23 concentrations were markedly increased compared to patients with fibrous dysplasia without phosphate wasting and normal individuals [12]. Moreover, FGF23 concentrations correlated with disease burden. Thus, patients with extensive disease had higher concentrations of FGF23 and more phosphate wasting than patients with limited disease. In accord with this clinical finding, in situ hybridization data demonstrated that FGF23 is expressed in the lesions in both the fibrous component of the lesion (mostly made up of immature osteogenic cells) and in the bone cells associated with abnormal bone trabeculae [12]. To determine if FGF23 expression was limited to abnormal osteogenic cells Riminucci et al. performed in situ hybridization in normal bone [12]. FGF23 was weakly expressed in inactive osetoblasts, and osteocytes, but there was more robust expression in osteoprogenetor cells, osteoblasts, and osteocytes in a fracture callous. In light of the fact that mutant and nonmutant cells both made FGF23, we compared FGF23 expression in cell cultures of both mutant and normal bone marrow stromal cells. Under the conditions examined, mutant stromal cells in culture did not produce more FGF23 than normal cells under the same conditions [12]. In light of this observation, the reader may question why there is a correlation between FGF23 concentrations and disease burden in these patients. Perhaps the most tenable hypothesis is that normal bone cells are appropriately regulated by whatever homeostatic mechanisms regulate FGF23 concentrations. Cells in fibrous dysplasia lesions may not respond appropriately to signals to decrease production of FGF23. When disease burden rises to a critical level, production of FGF23 exceeds what is needed for normal phosphate and vitamin D homeostasis and phosphate wasting, inhibition of calcitriol production, and enhanced calcitriol degradation ensues. This hypothesis is particularly enticing when one considers preliminary data that suggests a role for FGF23 in normal phosphate and vitamin D homeostasis (see below).

CHAPTER 70 ADHR, TIO, Fibrous Dysplasia, and FGF23

V. THE ROLE OF FGF23 IN XLH XLH is extensively covered in the previous chapter. Therefore, this chapter will briefly focus on the potential role of FGF23 in the pathogenesis of XLH. The potential roles of MEPE and FRP4 in the pathogenesis of XLH are also covered in Chapter 69 and will not be discussed further. As noted in the previous chapter, XLH is an X linked dominant disorder that results from inactivating mutations in the PHEX gene, which codes for a member of the M13 endopeptidase family [27]. Over 170 different mutations have been described that give rise to the disease (www.phexdb.mcgill.ca). XLH is characterized biochemically by renal phosphate wasting and inappropriately normal calcitriol concentrations. Parabiosis experiments [28] and renal crosstransplantation experiments [29] in the Hyp mouse implicate a circulating factor in the pathogenesis of the disease. In light of these findings, several investigators have measured FGF23 concentrations in XLH patients using assays that detect the C terminal portion of the molecule [30,31] or an assay with antibodies to both N and C termini [32]. Results indicate that most, but not all, XLH patients have increased FGF23 concentrations, compared to normal individuals. In those XLH patients with normal FGF23 levels, this concentration may be inappropriately normal given the low serum phosphorus concentrations. Of note, Hyp mice have an approximately tenfold increase in FGF23 concentrations compared to littermates [33]. Aono et al. studied the affects of anti FGF23 neutralizing antibodies in Hyp mice [33]. A single injection of neutralizing antibody caused dose dependent elevations of serum phosphate and calcitriol concentrations. Furthermore, a four-week trial of FGF23 neutralizing antibody led to a dose dependent improvement in growth and bone histomorphometric parameters, indicating that FGF23 may be at the center of the pathogenesis of XLH [33]. One issue that has not been adequately explained is why inactivating mutations in PHEX result in a dominant inheritance pattern. Most disorders that result from inactivating mutations in a gene that codes for an enzyme have a recessive inheritance pattern. Indeed, if FGF23 was a PHEX substrate one would expect XLH to be a recessive disorder and the parabiosis experiments performed by Meyer et al. [28] to result in rescue of the Hyp phenotype, rather than the normal mouse manifesting renal phosphate wasting, which is what was observed. Current data indicate that intact FGF23 is not a PHEX substrate [11,34]. Furthermore, recent data by Liu et at. [11] indicate that Hyp mice have markedly increased production of FGF23 mRNA in bone (MEPE is also overexpressed in Hyp mouse bone). Since PHEX and FGF23 are expressed by the

1193 same cells, the absence of functional PHEX protein in the cell likely leads to increased FGF23 production. These data provide a potential mechanism for why XLH is an X linked dominant disorder. X chromosome inactivation is random in female patients with XLH [35] and each cell in an affected female either has a normal or mutant copy of the PHEX gene. Thus, the possibility exists that those cells that have the mutant PHEX could secrete increased amounts of FGF23, resulting in increased circulating FGF23 concentrations. Clearly, this hypothesis must be adequately tested, but additional data may shed light on a question that has puzzled researchers since the PHEX gene was originally described.

VI. FGF23 IN HEALTH AND ITS POTENTIAL ROLE IN MAINTENANCE OF NORMAL PHOSPHATE AND VITAMIN D HOMEOSTASIS Currently, there is inadequate data to definitively determine what role, if any, FGF23 plays in maintenance of normal serum phosphate and calcitriol concentrations. FGF23 concentrations do not vary with age, making it unlikely that FGF23 is the primary mechanism by which children maintain a higher serum phosphate concentration than adults [30] (and Econs unpublished results). However, this likely means that the set point for FGF23 is at a higher level of phosphate in children than adults. The first, and most striking, evidence that FGF23 may play a role in vitamin D and phosphate homeostasis is the phenotype of the FGF23 knock-out mouse [36]. Heterozygous FGF23 knock-out mice are reported to be indistinguishable from normal. The homozygous knock-out mice appear normal at birth, but manifest increased serum phosphorus and calcitriol concentration compared to controls by the tenth day of life. Additionally, these mice display growth retardation and shortened life span with none of the knock-outs surviving past 12 weeks [36]. No data is currently available to see if these mice could be rescued by a low phosphate diet. Further data supporting the potential role of FGF23 in normal phosphate homeostasis is provided by Yamashita et al. [37], who measured FGF23 concentrations in rats fed different amounts of dietary phosphate. As would be predicted if FGF23 played a role in phosphate homeostasis, rats fed a high phosphate diet had higher FGF23 concentrations than those fed normal diets. Moreover, rats fed low phosphate diets displayed suppressed FGF23 concentrations [37]. These studies are supported by a study by Ferrari and colleagues [38] who gave young

1194 men normal, low, and high phosphate diets. Dietary phosphate restriction with addition of a phosphate binder decreased FGF23 concentrations by 29 ± 6%, and high phosphate diets increased FGF23 concentrations by 31 ± 9%. While the above data support the notion that FGF23 plays a role in normal phosphate homeostasis, there are no data regarding the rapidity of the response of FGF23 to changes in serum phosphate levels or dietary manipulations. Therefore, substantial work is needed to further investigate the role of FGF23 in normal phosphate homeostasis.

VII. SUMMARY Over the past few years there has been tremendous progress in our understanding of a variety of phosphate wasting disorders. The field continues to evolve at a very rapid rate. The exact roles of FGF23 and other candidate “phosphatonins” are not completely elucidated. However, a few points should be emphasized: 1) XLH, ADHR, TIO, and fibrous dysplasia all manifest renal phosphate wasting and inappropriately normal or low calcitriol concentrations; 2) missense mutations that protect FGF23 from proteolytic degradation cause ADHR; 3) FGF23, MEPE, and FRP4 are overexpressed in tumors that cause TIO; 4) transgenic mice that overexpress FGF23 display phosphate wasting and abnormalities in vitamin D metabolism; 5) injection of FGF23 into mice results in phosphate wasting and abnormalities of vitamin D metabolism; 6) implantation of CHO cells that express FGF23 into nude mice results in a phenotype that replicates that seen in TIO; 7) the FGF23 knock-out mouse manifests hyperphosphatemia and increased calcitriol concentrations; 8) most XLH patients have increased concentrations of FGF23 and normal levels in the remaining patients may be inappropriately normal; and 9) injection of neutralizing antibody against FGF23 into Hyp mice results in correction of the biochemical defect in a dose dependent fashion. Thus, although much remains to be done, work over the past few years has identified new genes that play roles in the pathophysiology of a variety of disorders of phosphate and vitamin D metabolism. The exact role of these genes and their role in normal phosphate and vitamin D homeostasis will require additional study.

Acknowledgements Work performed in the author’s laboratory was supported by NIH grants R01AR42228 and K24AR02095.

MICHAEL J. ECONS

References 1. Bianchine JW, Stambler AA, Harrison HE 1971 Familial hypophosphatemic rickets showing autosomal dominant inheritance. Birth Defects: Original Article Series 7:287–295. 2. Econs MJ, McEnery PT 1997 Autosomal dominant hypophosphatemic rickets/osteomalacia: clinical characterization of a novel renal phosphate wasting disorder. J Clin Endocrinol Metab 82:674–681. 3. Econs MJ, McEnery PT, Lennon F, Speer MC 1997 Autosomal dominant hypophosphatemic rickets is linked to chromosome 12p13. J Clin Invest 100:2653–2657. 4. The ADHR Consortium 2000 Autosomal dominant hypophosphatemic rickets is associated with mutations in FGF23. Nature Genetics 26:345–348. 5. Scriver CR, MacDonald W, Reade T, Glorieux RH, Nogrady B 1977 Hypophosphatemic nonrachitic bone disease: an entity distinct from X-linked hypophosphatemia in the renal defect, bone involvement, and inheritance. American Journal of Medical Genetics 1:101–117. 6. Van Swieten JC, Brusse E, De Graaf BM, et al. 2003 A mutation in the fibroblast growth factor 14 gene is associated with autosomal dominant cerebral ataxia. American Journal of Human Genetics 72:191–199. 7. White KE, Jonsson KB, Carn G, et al. 2001 The autosomal dominant hypophosphatemic rickets (ADHR) gene is a secreted polypeptide overexpressed by tumors that cause phosphate wasting. J Clin Endocrinol Metab 86(2):497–500. 8. White KE, Carn G, Lorenz-Depiereux B, Benet-Pages A, Strom TM, Econs MJ 2001 Autosomal dominant hypophosphatemic rickets (ADHR) mutations stabilize FGF-23. Kidney International 60:2079–2086. 9. Bai XY, Miao D, Goltzman D, Karaplis AC 2003 The autosomal dominant hypophosphatemic rickets R176Q mutation in fibroblast growth factor 23 resists proteolytic cleavage and enhances in vivo biological potency. J Bio Chem 278:9843–9849. 10. Shimada T, Muto T, Urakawa I, et al. 2002 Mutant FGF-23, responsible for autosomal dominant hypophosphatemic rickets, is resistant to proteolytic cleavage and causes hypophosphatemia in vivo. Endocrinology 143:3179–3182. 11. Liu SGR, Simpson JG, Ziao ZS, Burnham CE, Quarles LD 2003 Regulation of fibroblastic growth factor 23 expression but not degradation by PHEX. J Bio Chem 278:37419–37426. 12. Riminucci M, Collins MT, Fedarko NS, et al. 2003 FGF-23 in fibrous dysplasia of bone and its relationship to renal phosphate wasting. J Clin Invest 112:683–692. 13. Econs MJ, Drezner MK 1994 Tumor-induced osteomalacia— unveiling a new hormone. N Engl J Med 330:1679–1681. 14. Ryan EA, Reiss E 1984 Oncogenous osteomalacia: Review of the world literature of 42 cases and report of two new cases. American Journal of Medicine 77:501–512. 15. Drezner MK 1996) Tumor-induced rickets and osteomalacia. In: Favus MJ (ed) Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism, 3 ed. Lippincott-Raven, Philadelphia, pp. 319–325. 16. Folpe A, Fanburg-Smith JC., Billings SD, Bisceglia M, Bertoni F, Cho JY, Econs MJ, Inwards CY, JandeBeur SM, Mentzel T, Montgomery E, Michael M, Miettinen M, Reith JD, O’Connell JX, Rosenberg AE, Rubin BP, Sweet DE, Vinh TN, World LE, Wehrli BM, White KE, Zaino RJ, Weiss SW 2003 Most osteomalacia-associated mesencyhmal tumors are a single histopathologic entity: an analysis of 32 cases and a comprehensive review of the literature. Am J Surg Path. In Press.

1195

CHAPTER 70 ADHR, TIO, Fibrous Dysplasia, and FGF23

17. Cai Q, Hodgson SF, Kao PC, et al. 1994 Brief report: Inhibition of renal phosphate transport by a tumor product in a patient with oncogenic osteomalacia. N Engl J Med 330:1645–1649. 18. Seufert J, Ebert K, Muller J, et al. 2001 Octreotide therapy for tumor-induced osteomalacia. N Engl J Med 345:1883–1888. 19. Shimada T, Mizutani S, Muto T, et al. 2001 Cloning and characterization of FGF23 as a causative factor of tumor-induced osteomalacia. Proceedings of the National Academy of Sciences 98:6494–6499. 20. Shimada T, Yoneya T, Hino R, Takeuchi Y, Fukumoto S, Yamashita T 2001 Transgenic mice expressing fibroblast growth factor 23 (FGF23) demonstrate hypophosphatemia with low serum 1,25-dihydroxyvitamin D [1,25(OH)2D] and rickets/osteomalacia. Journal of Bone and Mineral Research 16 Suppl.1:S151. 21. Jan De Beur SM, Finnegan RB, Vassiliadis J, et al. 2002 Tumors associated with oncogenic osteomalacia express genes important in bone and mineral metabolism. Journal of Bone & Mineral Research 17:1102–1110. 22. Gowen LC, Petersen DN, Mansolf AL, et al. 2003 targeted disruption of the osteoblast/osteocyte factor 45 gene (OF45) results in increased bone formation and bone mass. Journal of Biological Chemistry 278:1998–2007. 23. Berndt T, Craig T, Bowe A, et al. 2003 Secreted frizzled related protein 4 is a potent tumor-derived phosphaturic agent. J Clin Invest 112:642–646. 24. Weinstein LS 2000 Fibrous dysplasia and the McCune-Albright syndrome. Humana Press: Totowa NJ. 25. Weinstein LS, Shenker A, Gejman PV, Merino MJ, Friedman E, Spiegel AM 1991 Activating mutations of the stimulatory G protein in the McCune-Albright syndrome. [Comment]. N Eng J Med 325:1688–1695. 26. Collins MT, Chebli C, Jones J, et al. 2001 Renal phosphate wasting in fibrous dysplasia of bone is part of a generalized renal tubular dysfunction similar to that seen in tumor-induced osteomalacia. J Bone Miner Res 16:806–813. 27. The Hyp Consortium 1995 A gene (PEX) with homologies to endopeptidases is mutated in patients with X-linked hypophosphatemic rickets. Nature Genetics 11:130–136. 28. Meyer RA, Jr., Meyer MH, Gray RW 1989 Parabiosis suggests a humoral factor is involved in X-linked hypophosphatemia in mice. J Bone Miner Res 4:493–500. 29. Nesbitt T, Coffman TM, Griffiths R, Drezner MK 1992 Crosstransplantation of kidneys in normal and Hyp mice.

30.

31. 32.

33.

34.

35.

36.

37.

38.

Evidence that the Hyp mouse phenotype is unrelated to an intrinsic renal defect. Journal of Clinical Investigation 89:1453–1459. Jonsson KBZR, Larsson T, White KE, Toshitsugu S, Imanishi Y, Tamamoto T, Hampson G, Miyauchi A, Econs MJ, Lavigne J, Juppner H 2003 Fibroblast growth factor 23 in oncogenic osteomalacia and X-linked hypophosphatemia. New England Journal of Medicine 348(17):1656–1663. Weber T, Liu S, Indridason OS, Quarles LD 2003 Serum FGF23 levels in normal and disordered phosphorous homeostasis. J Bone Miner Res 18:1227–1234. Yamazaki Y, Okazaki R, Shibata M, et al. 2002 Increased circulatory level of biologically active full-length FGF-23 in patients with hypophosphatemic rickets/osteomalacia. J Clin Endocrinol Metab 87:4957–4960. Aono Y, Shimada T, Yamazaki Y, Hino R, Takeuchi Y, Fujita T, Fukumoto S, Nagano N, Wada M, Yamashita T 2003 The neutralization of FGF-23 ameliorates hypophosphatemia and rickets in Hyp mice. American Society for Bone and Mineral Research: Minneapolis, MN, p. S16. Campos MCC, Hirata IY, Juliano MA, Loisel TP, Crine P, Juliano L, Boileau G, Carmona CK 2003 Human recombinant endopeptidase PHEX has a strict S1, specificity for acidic residues and cleaves peptides derived from fibroblast growth factor-23 and matrix extracellular phosphoglycoprotein. Biochem J. 373:271–279. Orstavik KH, Orstavik RE, Halse J, Knudtzon J 1996 X chromosome inactivation pattern in female carriers of X linked hypophosphataemic rickets. J Medical Genetics 33:700–703. Shimada TKM, Hasegawa H, Yamazaki T, Ohguma A, Takeuchi Y, Fujita T, Fukumoto S, Tomzuka K, Yamashita T 2002 Targeted ablation of FGF-23 causes hyperphosphatemia, increased 1,25-dihydroxyvitamin D level and severe growth retardation. J Bone and Miner Res: S168. Yamashita T, Hasegawa H, Yamazaki Y, Kawata T, Urakawa I, Shimada T, Takeuchi Y, Fujita T, Fukumoto S, Nagano N 2002 Involvement of FGF-23 in abnormal vitamin D and mineral metabolism associated with renal insufficiency. Journal of American Society of Nephrology 13:577A. Ferrari S, Bonjour J, Rizzoli R 2003 Evidence for a physiological role of FGF-23 in the regulation of renal phosphate reabsorption and plasma calcitriol in healthy humans. American Society of Bone and Mineral Research Annual Meeting: Minneapolis, MN, p. S24.

CHAPTER 71

Vitamin D Pseudodeficiency FRANCIS H. GLORIEUX RENÉ ST-ARNAUD

I. II. III. IV. V.

Genetics Unit, Shriners Hospital for Children, and Departments of Surgery, Pediatrics, and Human Genetics, McGill University, Montréal, Québec, Canada Genetics Unit, Shriners Hospital for Children, and Departments of Surgery and Human Genetics, McGill University, Montréal, Québec, Canada

Introduction Clinical Manifestations Biochemical Findings Placenta Studies Genetic Studies

I. INTRODUCTION Following the description by Albright et al. in 1937 [1] of “rickets resistant to vitamin D,” a number of observations were published [2,3] that indicated that there was a variant of resistant rickets which differed from the classic hypophosphatemic type (X-linked hypophosphatemic rickets, or XLH; see Chapter 69) by its clinical and biological symptoms and response to therapy. It was indeed shown by Prader et al. [4] that this form of rickets had an early onset (within the first year of life), contrary to the XLH type. The disease symptoms also included the development of profound hypocalcemia, tooth enamel hypoplasia, and a response to daily administration of large amounts of vitamin D. In view of the latter, the term “vitamin D dependency” was proposed to describe the new syndrome [5]. In 1973, when calcitriol [1,25(OH)2D3] became available as a therapeutic agent, it was demonstrated that this rare form of rickets responded to physiological amounts of calcitriol [6]. It was then recognized that this disease was due to an inborn error of metabolism involving the defective conversion of calcidiol (25OHD3) to calcitriol (Fig. 1). For this reason, we feel it more appropriate to return to the original terminology of Prader and use the term pseudovitamin D-deficiency rickets (PDDR) to describe this form of rickets. In 1978, another inborn error of vitamin D metabolism was recognized in which a clinical picture of pseudovitamin D–deficiency developed despite high circulating concentrations of endogenously produced calcitriol [7]. In some pedigrees, the phenotype is compounded by the presence of complete alopecia [8]. This second variety of pseudodeficiency has been termed vitamin D–dependency type II, pseudovitamin D–deficiency type II, calcitriol-resistant rickets, VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

VI. Molecular Defect VII. Treatment VIII. Conclusion References

hypocalcemic vitamin D–resistant rickets, and hereditary 1,25-dihydroxyvitamin D–resistant rickets (HVDRR). The latter term is favored by Malloy, Pike, and Feldman, who discuss it in detail in Chapter 72. HVDRR is caused by a spectrum of mutations affecting the vitamin D receptor (VDR) in target tissues causing true resistance to calcitriol action. The human VDR, a 50-kDa protein, belongs to the steroid-thyroidretinoic acid receptor superfamily of genes [9]. It comprises at least two functional domains, a ligand binding domain and a DNA binding domain (Fig. 1). Mutations affecting both have been found in HVDRR families (see Chapter 72).

II. CLINICAL MANIFESTATIONS The clinical course of PDDR is similar to that of nutritional rickets due to simple vitamin D deficiency. The patients are healthy at birth. The first symptoms usually appear within the first year of life. Hypotonia, muscle weakness (proximal myopathy), and growth retardation are common manifestations. Motor problems translate into regression in head control and the ability to stand. In some patients, the initial event is convulsions or tetany. Pathological fractures may occur. A history of adequate mineral and vitamin D intake, without evidence of intestinal malabsorption, is a constant finding. Infant death by hypocalcemia or pulmonary infections was not infrequent in the past when the diagnosis was either missed (confused with a neurological or respiratory condition) or made too late. Physical examination reveals a small, hypotonic child with features similar to those found in patients with vitamin D–deficiency rickets. There is a wide anterior fontanel with frontal bossing and frequent Copyright © 2005, Elsevier, Inc. All rights reserved.

1198

FRANCIS H. GLORIEUX AND RENÉ ST-ARNAUD

Liver: Vitamin D 25-hydroxylase

Cytoplasm

25(OH)D3

VDR

VDR

PDDR

HO

HO

DBD LBD

HO

HO

DNA

HO

RXR Transcription

1,25(OH)2D3 HO

Translation

Kidney: 25-hydroxyvitamin D 1α-hydroxylase

mRNA

Nucleus

FIGURE 1 Schematic representation of the main steps of the vitamin D biosynthetic pathway, where genetic aberrations may lead to rickets and osteomalacia. The renal defect in PDDR is indicated by the break in the 1,25(OH)2D3 arrow arising in the kidney. The mutation leads to insufficient synthesis of 1,25(OH)2D3. The left part of the figure represents a target cell where schematic coupling of the ligand to its receptor (VDR) takes place in the cytosol or, more likely, in the nucleus. The VDR then heterodimerizes with the RXR receptor. For ease of presentation, the RXR ligand (9-cis retinoic acid) is not depicted. The complex then binds to DNA to regulate transcription. Various mutations affecting either one of the two VDR domains cause hereditary vitamin D–resistant rickets (HVDRR) (see Chapter 72).

craniotabes (easy depression of the softened parietooccipital area). Tooth eruption is delayed, and erupted teeth show evidence of enamel hypoplasia. A rachitic rosary is either visible or palpable. In the appendicular skeleton, enlargement of the metaphyseal areas is more evident in the wrists and ankles, and there is a variable degree of deformity (bowing) of long bone diaphyses. The Chvostek sign (twitching of the upper lip on light finger tapping of the facial nerve) reflects nerve irritability, a consequence of a rapid drop in serum calcium. Radiological examination of the skeleton reveals diffuse osteopenia and the classic metaphyseal changes of vitamin D deficiency. There is fraying, cupping, widening, and fuzziness of the zone of provisional calcification immediately under the growth plate. These changes are seen better and detected earlier in the most active growth plates, namely, the distal ulna and femur and the proximal and distal tibia. Changes in the diaphyses may not be evident when metaphyseal changes are first detected. However, they will appear a few weeks later as rarefaction, coarse trabeculation, cortical thinning, and subperiosteal erosion (see Chapter 60). The latter reflects the increased resorption induced by secondary hyperparathyroidism.

III. BIOCHEMICAL FINDINGS Hypocalcemia is the cardinal feature in PDDR. Serum calcium concentration will drop below 2 mmol/liter (8 mg/dl). This, particularly if the decrease is rapid, may give rise to tetany and convulsions, which may occur prior to any radiological evidence of rickets. Persistent hypocalcemia triggers secondary hyperparathyroidism and hyperaminoaciduria [10]. Urinary calcium content is low, whereas fecal calcium is high, reflecting impaired intestinal calcium absorption. Increased urinary cAMP is not a consistent finding, and normal values have been observed in PDDR patients with high circulating parathyroid hormone (PTH) levels [11]. Serum phosphate concentration may be normal or low. Hypophosphatemia, when present, is usually of a lesser degree than in XLH. It is the result of both impairment of intestinal absorption and increased urinary loss induced by secondary hyperparathyroidism. Serum alkaline phosphatase activity is consistently elevated (over 300 IU/liter). Its increase often precedes the appearance of clinical symptoms. The calcemic response to PTH is usually but not necessarily absent [12]. Studies of circulating vitamin D metabolites have provided a key insight into the pathogenesis of PDDR.

1199

CHAPTER 71 Vitamin D Pseudodeficiency

Serum levels of 25OHD are normal in untreated patients and elevated in patients receiving large daily amounts of vitamin D [11]. These results indicate that intestinal absorption of vitamin D and its hydroxylation in the liver are not impaired in PDDR. Circulating levels of 1,25(OH)2D are low in untreated patients [11–13]. This is evident immediately after birth, months before any clinical evidence of rickets develops. Even when patients are treated with large doses of vitamin D, causing major increases in the circulating levels of calcidiol, calcitriol levels do not reach the normal range (Fig. 2). This clearly identifies defective activity of the 25OHD 1α-hydroxylase enzyme as the basic abnormality in PDDR and differentiates it from HVDRR (see Chapter 72). Although 1,25(OH)2D serum levels are low, they are not undetectable. This finding, coupled with the observation that serum levels of 1,25(OH)2D are positively correlated to the serum concentrations of 25OHD in PDDR patients (either untreated or treated with large amounts of vitamin D) suggests that the renal 1α-hydroxylase is not totally absent in PDDR. Thus, the mutation probably affects the structural integrity of the enzyme, resulting in a modification of its kinetics [11]. Balsan and associates [14] have reported on normal calcitriol levels in untreated PDDR

100 200

150 60

100 40

50

20

Serum 1,25(OH)2D3 (pmol/l)

Serum 1,25(OH)2D3 (pg/ml)

80

IV. PLACENTA STUDIES In 1979, Weisman et al. [18] demonstrated that, besides the mammalian kidney, human placenta decidua was a major site of 1,25(OH)2D synthesis. This was further substantiated by Delvin et al. [19], who also demonstrated that the involved enzyme was regulated by feedback mechanisms [20]. Over the years, in the cohort of our patients successfully treated with replacement therapy (see below), several have reached adulthood and became pregnant. At delivery, decidual cells were harvested from the placentas of these PDDR patients and were studied to evaluate their ability to hydroxylate 25OHD at the lα position. As shown in Fig. 3, we demonstrated that decidual cells from women with PDDR lack that function, making them likely targets for the mutation [21]. The physiological importance of this defect particularly with regard to fetal development is unclear. Replacement therapy with calcitriol was maintained throughout pregnancy, and patients had uneventful pregnancies and delivered healthy normocalcemic babies. The placenta thus represents a unique, albeit rare, source of mutant cells for further characterization of the PDDR mutation.

V. GENETIC STUDIES 0

0 Control

Untreated +D2

+1,25(OH)2D3

PDDR

FIGURE 2

patients. Such values, however, should be considered as inappropriate in the face of rickets, hypocalcemia, and secondary hyperparathyroidism. These differences may also reflect genetic heterogeneity among PDDR pedigrees that will only be resolved at the molecular level. Circulating levels of 24,25(OH)2D are normal in PDDR patients and are highly correlated with those of 25OHD, indicating a fully functional 24-hydroxylase enzyme [15,16]. These findings, as well as the observation that modulation of the expression of the 25OHD 24-hydroxylase is regulated independently from that of the 1α-hydroxylase [17], strongly suggest that the two renal hydroxylases are distinct gene products (see Chapters 5 and 6).

Serum calcitriol concentrations in control children and in PDDR patients either untreated or treated with high doses of vitamin D (+D2) or calcitriol [1,25(OH)2D3]. The data scatter in the latter group reflects both dosage and the variable length of time between drug administration and blood sampling.

Pseudovitamin D–deficiency rickets is inherited as a simple autosomal recessive trait [5]. No phenotypic abnormalities have been observed in presumed obligate heterozygotes [10]. Although quite rare, PDDR is present with unusual frequency in a subset of the French-Canadian population [22]. With the cooperation of the several large families under our care, we set out to map the PDDR locus by using DNA markers and linkage analysis to approach the primary defect in PDDR.

1200

FRANCIS H. GLORIEUX AND RENÉ ST-ARNAUD

Decidual cell 1α-Hydroxylase activity in control & PDDR patients 1,25(OH)2D3

1α-Hydroxylase activity (dpm/fraction)

8,000

Control PDDR

Fetal membranes

6,000

4,000

Decidua • MEM-Eagle's salts • HEPES (10 mM, pH 7.4) • Streptomycin/penicillin • Collagenase, 3 hrs, 37°C • Percoll gradient Decidual cells

24,25(OH)2D3

2,000

0 0

5

10

15

20

25

30

Fraction number

FIGURE 3 High-performance liquid chromatography elution patterns of the radioactivity extracted from medium and cells isolated from the decidua of one control and one PDDR patient. The elution positions of the two vitamin D dihydroxylated metabolites are indicated.

The studied kindreds included 17 affected individuals and 59 healthy relatives, of whom 17 were obligate heterozygotes (having affected progeny). It was found that the mutated gene was linked to polymorphic RFLP (restriction fragment length polymorphism) markers in the region of band 14 of the long arm of chromosome 12 [12ql4] [23]. Multipoint linkage analysis and studies of haplotypes (groups of tightly linked markers segregating together over the generations) and recombinants strongly suggest the localization of the PDDR locus between COL2A1 (coding for the α1 chain of type II collagen) and a cluster of three anonymous probes [D12S14, D12S17, and D12S6], which segregate as a three-marker haplotype. Linkage disequilibrium (i.e., combinations of closely linked genes occurring more often than expected with random distribution) has been observed between the PDDR locus and the three-marker haplotype in the group of kindreds studied [24]. The finding supports the notion of a founder effect that had taken place in the second half of the seventeenth century (about 12 generations ago). This is consistent with the present-day prevalence of 1 in 2400 births and carrier rate of 1 in 26 individuals in Northeastern Quebec [25]. The VDR gene has also been assigned to chromosome 12 by Southern blot analysis of a panel of humanChinese hamster cell hybrid DNAs. Using in situ hybridization, the VDR was found to map to the same 12ql2-14 region where PDDR was localized [26]. Because the VDR cDNA exhibits an Apa1 dimorphism [27], it was used as a RFLP marker in linkage analyses of samples from 21 of our PDDR families. The PDDR and VDR

loci are located in close proximity to the markers COL2A1, ELA (elastase), and D12S15. It is likely that the genetic distance between the two genes involved in the control of vitamin D activity is in the range of a few centimorgans, which, in physical terms, may correspond to 1–10 megabases. We find, at present, no specific reason for this proximity, but its functional significance may be established in the future [26].

VI. MOLECULAR DEFECT Remarkable progress was recently made in the understanding of the molecular etiology of PDDR through the cloning of the cDNA encoding the 1αhydroxylase enzyme, from rat [28,29] and mouse [30] kidney, and human keratinocytes and kidney [31]. The human gene was also cloned, sequenced, and mapped to chromosome 12q13.1–13.3 by fluorescence in situ hybridization [30,32,33], consistent with the earlier mapping of the disease by linkage analysis. The definite proof that mutations in the 1α-hydroxylase gene are responsible for the PDDR phenotype comes from the identification of such mutations in PDDR patients and obligate carriers. The first report was by Fu et al. [34] in 1997; since then additional mutations in various ethnic groups have been published [32,35]. To date, 31 different mutations have been observed in PDDR patients and their parents [31–33,35–39]. They are dispersed throughout the 1α-hydroxylase sequence, affecting all exons and two intervening sequences (Fig 4). All patients have mutations on both alleles, but some

1201

CHAPTER 71 Vitamin D Pseudodeficiency

gggcg → cttcgg

D164N

IVS3 + 1g → a

S323Y

958∆G Q65H 212∆G

P143L

1921∆G

E189L E189G

R107H G125E

1984∆C

IVS2 + 1g → a

T321R

W241X

W433X R389G R389H R389C

R453C R429P

P382S

R335P

7bp DUPL. 2bp DUPL. V478G P497R

T409I L343F

FIGURE 4

Mutations detected in PDDR patients and obligate heterozygotes. A schematic representation of the 1α-hydroxylase gene is shown. The dark-shaded boxes correspond to the nine exons of the gene, and the light-shaded boxes at either ends represent the 5′- and 3′nontranslated regions. Mutations are presented above and below the gene map. Numbers refer to amino acid residues. ∆: deletion; gggcg → cttcg: deletion of gggcg and substitution of cttcg beginning at nucleotide 897 in exon 2; IVS2 or IVS3 + 1g → a: splice site mutation in intron (intervening sequence) 2 or 3; 7 or 2bp.DUPL.: 7 or 2 basepairs duplication.

harbor a compound heterozygosity (a different mutation on each allele). The mutations detected at the highest frequency are 958∆G, common among French Canadian patients because of a founder effect [24,33,37], and a mutation in exon 8 that causes a 7 basepairs duplication. The latter alters the downstream reading frame [33,37,38]. An important aspect of the identification of mutations in the 1α-hydroxylase gene is to correlate genotype and phenotype, i.e. the severity of the disease and the circulating levels of 1,25(OH)2D. In most cases, although levels are low, they are detectable, suggesting some degree of residual 1α-hydroxylase activity. This could result from the mutations affecting the structural integrity of the enzyme and hence its kinetics. Such an effect could derive from missense mutations, but not from frameshift (deletions, inversions, and duplications) or nonsense mutations that would result in complete inactivity of the mutant 1α-hydroxylase. An animal model of PDDR was engineered independently by two laboratories using targeted inactivation of the gene of interest in mice [39,40]. The engineered mutation is transmitted with the expected mendelian ratio. A detailed description of the model is presented in Chapter 7.

dose, probably for the lifetime of the patient (Table I). Under such treatment, circulating levels of 25OHD increase sharply, with only minor changes in the levels of 1,25(OH)2D (Fig. 2). It is likely that massive concentrations of 25OHD are able to bind to VDR and induce the response of the target organs to normalize calcium homeostasis. However, because such therapy leads to progressive accumulation of vitamin D in fat and muscle tissues, adjustment in case of overdose is difficult and slow to come into effect. Furthermore, the therapeutic doses are close to the toxic doses and place the patient at risk for nephrocalcinosis and impaired renal function. There have been reports on the use of 25OHD3 as a therapeutic agent in PDDR [41]. The doses used are smaller than those of vitamin D (Table I) and induce a similar response. The action of 25OHD3 is likely to be similar to the one of vitamin D itself, by maintaining high serum concentrations of 25OHD. The low availability and high cost of such a preparation have discouraged its widespread use as a long-term therapy for PDDR.

TABLE I Vitamin D Dosage Requirements of PDDR Patients Dosage (µg/day)

VII. TREATMENT Vitamin D2, given at an appropriate daily dose, can be used to treat PDDR. The biochemical and clinical abnormalities regress and normal linear growth is restored. The dose of vitamin D2 to heal the bone disease may be as high as 2.5 mg (100,000 IU) per day. This dose can be reduced by half or more to a maintenance

Compound Vitamin D 25OHD3 1αOHD3 1,25(OH)2D3

To heal rickets 1000–2500 250–1000 2–5 1–3

Maintenance 500–1250 100–500 1–2 0.25–1

1202

FRANCIS H. GLORIEUX AND RENÉ ST-ARNAUD

The treatment of choice is replacement therapy with calcitriol. Before this compound became available from commercial sources, several investigators used the monohydroxylated analog 1αOHD3, which requires only liver hydroxylation at the 25 position (a step not affected by the PDDR mutation) to fully mimic 1,25(OH)2D3 [42]. The response is rapid with healing of rickets in 7–9 weeks, requiring a daily dosage of 2–5 µg. The maintenance dose is about half the initial dose (Table I). Withdrawal induces a reappearance of symptoms within 3 weeks. Thus, long-term compliance is a more important consideration than in the case of vitamin D treatment. On a weight basis, 1αOHD3 is about half as potent as 1,25(OH)2D [42], nullifying any possible economic advantage in favor of the monohydroxylated form. The reason for this difference in potency has not been investigated, but may be related to a difference in intestinal absorption or to a variable degree of 25-hydroxylation of 1αOHD3. Replacement therapy with calcitriol results in rapid and complete correction of the abnormal phenotype,

eliminating hypocalcemia, secondary hyperparathyroidism, and radiographic evidence of rickets (Figs. 5 and 6). The restoration of bone mineral content is equally rapid (Fig. 5), and histological evidence of healing has been documented [43]. Severe tooth enamel hypoplasia is a common complication of PDDR, which is only partially corrected if treatment, as it is usually the case, is started around 12–15 months of age when permanent tooth enamel has already started to develop (Fig. 7). The calcitriol regimen calls for an initial dose of 1–3 µg/day continued until bone is healed, and is followed by a maintenance dose of 0.25-1 µg/day (Table I) to be continued probably throughout life. An important component of treatment is to ensure adequate calcium intake during the bone healing phase. Dietary sources are supplemented to ensure a daily supply of around 1 g of elemental calcium. Needs are monitored by frequent (bimonthly) assessment of urinary calcium excretion. The latter can easily be assessed by measuring calcium and creatinine in an aliquot of the second

A.D. 25

2.4

PTH (pmol/l)

Ca (mmol)

2.8

2.0 1.6

10

0 Aug

Sep

Oct

Nov

Dec

Jan

Feb

Mar

3000

Aug

Sep

Oct

Nov

Dec

Jan

Feb

Mar

Aug

Sep

Oct

Nov

Dec

Jan

Feb

Mar

Dec

Jan

Feb

Mar

0 BMD (Z score)

P’ase Alc (U/L)

15

5

1.2

2000

1000

−2 −4 −6 −8

0

1.0

Sep

Oct

Nov

Dec

Jan

Feb

Mar

Ca (300 mg/d)

0.5

Rocaltrol Rx

0.0 Aug

FIGURE 5

Sep

Oct

Nov

Dec

Jan

Feb

Mar

Rocaltrol (µg/d)

Aug Rocaltrol (µg/d)

20

1.0

Ca (300 mg/d)

0.5 Rocaltrol Rx

0.0 Aug

Sep

Oct

Nov

Biochemical response to treatment in a 16-month-old boy with PDDR treated with calcitriol (Rocaltrol) and calcium supplements. There was a progressive correction of the hypocalcemia (Ca) and secondary hyperparathyroidism (PTH) with concomittant decrease in alkaline phosphatase activity (P’ase Alc). Correction of the osteopenia followed the same pattern [BMD is bone mineral density of the lumbar spine by dual energy X-ray absorptiometry (DXA)]; the Z score is based on standard deviations from the mean BMD in age-matched controls.

1203

CHAPTER 71 Vitamin D Pseudodeficiency

FIGURE 7 Permanent incisors of a 9-year-old patient with PDDR in whom calcitriol treatment was initiated at age 14 months. The part of the enamel that was formed before treatment remains hypoplastic. Subsequent to treatment, normal enamel was produced.

FIGURE 6 Radiographs of the right wrist (upper panel) and knee (lower panel) of a patient with PDDR (same as in Fig. 5). (Left) Before treatment; (right) after only 3 weeks of treatment, healing of rickets is well under way.

void of the morning by a fasting patient. Normal values for the calcium/creatinine ratio are <0.35 (mg/mg) or <1.1 (mmol/mmol). In the untreated state, calciuria is very low. It will stay low as long as bone is rapidly remineralizing. An increase in calciuria is the most sensitive index of efficient therapy. Hypercalciuria (which precedes hypercalcemia by weeks) calls for reducing calcitriol progressively to the maintenance dosage. Once the latter is established, assessment of calciuria every 3 months is sufficient to keep control of the treatment. Requirements have been remarkably stable in our cohort of 32 patients treated for up to 21 years. During normal pregnancy, calcitriol circulating levels steadily increase to about twice the control values [44]. This adaptation to the specific needs of pregnancy can be mimicked in pregnant patients with PDDR by increasing the daily calcitriol dose during the second half of pregnancy. In seven such patients, we increased the calcitriol dose by 50–100% of the maintenance dose. All women gave birth to normal

(obligate heterozygote) babies. The maintenance dose was progressively reestablished after delivery (F. H. Glorieux, unpublished data). Hypercalciuria is not infrequent during treatment with calcitriol, particularly during the first year of administration, because changes in urinary calcium excretion are used to adjust the daily calcitriol dose. High levels of calcium excretion may amplify the pattern of calcium deposition in the normal kidney [45] and generate echodense images of the renal pyramids. This is referred to as nephrocalcinosis [46] and has been detected in several patients with PDDR treated with either vitamin D2 or calcitriol [47]. As it may reflect a potential nephrotoxic effect of calcitriol, we have now included in our treatment protocol an annual renal ultrasonography study and evaluation of the creatinine clearance. A positive ultrasound was observed in 10 of 20 patients with PDDR treated for a mean of 8.1 years. The intensity of the images did not change with time. Two patients have shown a decrease in creatinine clearance. However, both had a history of vitamin D intoxication prior to calcitriol therapy [48]. Thus, duration of administration and dosage of the compounds used for treatment will influence the development of renal medullary changes. Frequent renal imaging and assessment of renal function are therefore essential.

VIII. CONCLUSION Pseudovitamin D–deficiency rickets is a rare condition inherited as an autosomal recessive trait that results in an inadequate synthesis of calcitriol that

1204 compromises intestinal calcium absorption and bone mineralization. The majority of the cases described are part of large kindreds from Northeastern Québec in Canada. Extensive genetic studies of those families have allowed an assignment of the PDDR locus to the long arm of chromosome 12 in close vicinity to the VDR gene. With the cloning of the 1α-hydroxylase gene came the characterization of an array of mutations that all lead to severe alteration of the 1α-hydroxylase activity. Replacement therapy with small daily doses of calcitriol is the treatment of choice in PDDR. It is highly efficient, removing this condition from the list of lethal mutations. It should probably be continued throughout life. Because of the potential nephrotoxicity of this treatment, regular monitoring of kidney function is mandatory. PDDR was the first described inborn error of vitamin D metabolism. Through the complete unraveling of its molecular defect, it has contributed in a major way to our understanding of vitamin D biology. Because of the easiness and efficacy of the replacement therapy, it is unlikely that any form of gene-based therapy will be considered anytime soon.

References 1. Albright F, Butler AM, Bloomberg E 1937 Rickets resistant to vitamin D therapy. Am J Dis Children 54:529–547. 2. Royer P 1960. Etude sur les rachitismes vitamino-resistants hypophosphatemiques idiopathiques. Acta Clin Belg 15:499–517. 3. Fraser D, Salter RB (1958) The diagnosis and management of various types of rickets. Pediatr Clin North Am 5:417–441. 4. Prader A, Illig R, Heierli E 1961 Eine besondere form des primare vitamin D–resistenten rachitis mit hypocalcamie und autosomaldominanten Erbgang: Die hereditare PseudoMangelrachitis. Helv Paediatr Acta 16:452–468. 5. Scriver CR 1970 Vitamin D dependency. Pediatrics 45:361–363. 6. Fraser D, Kooh SW, Kind HP, Rollick MF, Tanaka Y, DeLuca HF 1973 Pathogenesis of hereditary vitamin D–dependent rickets. An inborn error of vitamin D metabolism involving defective conversion of 25-hydroxyvitamin D to 1α,25-dihydroxyvitamin D. N Engl J Med 289:817–822. 7. Marx SJ, Spiegel AM, Brown EM, Gardner DG, Downs RW, Attie M, Hamstra AJ, DeLuca HF 1978 A familial syndrome of decrease in sensitivity to 1,25-dihydroxyvitamin D, J Clin Endocrinol Metab 47:1303–1310. 8. Rosen JF, Fleischman AR, Finberg L, Hamstra AJ, DeLuca HF 1979 Rickets with alopecia: An inborn error of vitamin D metabolism. J Pediatr 94:729–735. 9. Baker AR, McDonnell DP, Hughes M, Crisp TM, Mangelsdorf DJ, Haussler MR, Pike JW, Shine J, O’Malley BW 1988 Cloning and expression of full length cDNA encoding human vitamin D receptor. Proc Natl Acad Sci USA 85: 3294–3298. 10. Arnaud C, Maijer R, Reade TM, Scriver CR, Whelan DT 1970 Vitamin D dependency: An inherited postnatal syndrome with secondary hyperparathyroidism. Pediatrics 46:871–880. 11. Delvin EE, Glorieux FH, Marie PJ, Pettifor JM 1981 Vitamin D– dependency: Replacement therapy with calcitriol. J Pediatr 99:26–34.

FRANCIS H. GLORIEUX AND RENÉ ST-ARNAUD

12. Rosen JF, Finberg L 1972 Vitamin D–dependent rickets: Actions of parathyroid hormone and 25-hydroxycholecalciferol. Pediatr Res 6:552–562. 13. Scriver CR, Reade TM, Hamstra AJ, DeLuca HF 1978 Serum 1,25-dihydroxyvitamin D levels in normal subjects and in patients with hereditary rickets or bone disease. N Engl J Med 299:976–979. 14. Balsan S Garabedian M, Sorgniard R, Holick MF, DeLuca HF 1975 1,25-Dihydroxyvitamin D3 and 1,α-hydroxyvitamin D3 in children: Biologic and therapeutic effects in nutritional rickets and different types of vitamin D resistance. Pediatr Res 9:593–599. 15. Glorieux FH, Delvin EE 1991 Pseudo-vitamin D–deficiency rickets. In: Vitamin D: Regulation, Structure-Function Analysis and Clinical Application, de Gruyter: Berlin and New York, pp. 238–245. 16. Mandla S, Jones G, Tenenhouse HS 1992 Normal 24-hydroxylation of vitamin D metabolites in patients with vitamin D– dependency rickets type I. Structural implications for vitamin D hydroxylases. J Clin Endocrinol Metab 74:814–820. 17. Arabian A, Grover J, Barre MG, Delvin EE 1993 Rat kidney 25-hydroxyvitamin D3 1α- and 24-hydroxylases: Evidence for two distinct gene products. J Steroid Biochem Mol Biol 45:513–516. 18. Weisman Y, Harell A, Edelstein S, David M, Spirer Z, Golander A 1979 lα,25-Dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3 in vitro synthesis by human decidua and placenta. Nature 281:317–319. 19. Delvin EE, Arabian A, Glorieux FH, Mamer OA 1985 In vitro metabolism of 25-hydroxycholecalciferol by isolated cells from human decidua. J Clin Endocrinol Metab 601:880–885. 20. Delvin EE, Arabian A 1987 Kinetics and regulation of 25-hydroxycholecalciferol 1α-hydroxylase from cells isolated from human term decidua. Eur J Biochem 163:659–662. 21. Glorieux FH, Arabian A, Delvin EE 1995 Pseudo-vitamin D deficiency: Absence of 25-hydroxyvitamin D 1α-hydroxylase activity in human placenta decidual cells. J Clin Endocrinol Metab 80:2255–2258. 22. Bouchard G, Laberge C, Scriver CR, Glorieux F, Declos M, Bergeron L, Larochelle J, Mortezai S 1984 Etude demographique et genealogique de deux maladies hereditaires au Saguenay. Cahiers Quebecois de Demographic 13:117–137. 23. Labuda M, Morgan K, Glorieux FH 1990 Mapping autosomal recessive vitamin D–dependency type I to chromosome 12ql4 by linkage analysis. Am J Hum Genet 47:28–36. 24. Labuda M, Labuda D, Korab-Laskowska M, Cole DEC, Zietkiewicz E, Weissenbach J, Popwska E, Pronicka E, Root AW Glorieux FH 1996 Linkage disequilibrium analysis in young populations: Pseudovitamin D–deficiency rickets (PDDR) and the founder effect in French Canadians. Am J Hum Genet 59:633–643. 25. De Braekeleer M. 1991 Hereditary disorders in Saguenay-LacSt-Jean (Quebec, Canada). Hum Hered 41:141–146. 26. Labuda M, Fujiwara TM, Ross MV, Morgan K, Garcia-Heras J, Ledbetter DH, Hughes MR, Glorieux FH 1992 Two hereditary defects related to vitamin D metabolism map to the same region of human chromosome 12ql3-14. J Bone Miner Res 7:1447–1453. 27. Faraco JH, Morrison NA, Baker A, Shine J, Frossard PM 1989 Apal dimorphism at the human vitamin D receptor gene locus Nucleic Acids Res 17:2150. 28. St-Arnaud R, Messerlian S, Moir JM, Omdahl JL, Glorieux FH 1997: The 25-hydroxyvitamin D 1-alpha-hydroxylase gene maps to the pseudovitamin D–deficiency rickets (PDDR) disease locus. J Bone Miner Res 12:1552–1559.

CHAPTER 71 Vitamin D Pseudodeficiency

29. Shinki T, Shimada H, Wakino S, et al. 1997 Cloning and expression of rat 25-hydroxyvitamin D3-1alpha-hydroxylase cDNA. Proc Natl Acad Sci USA 94:12920–12925. 30. Takeyama K, Kitanaka S, Sato T, et al. 1997 25-Hydroxyvitamin D3 1alpha-hydroxylase and vitamin D synthesis. Science 277:1827–1830. 31. Fu GK, Lin D, Zhang MY, et al. 1997 Cloning of human 25-hydroxyvitamin D-1 alpha-hydroxylase and mutations causing vitamin D–dependent rickets type 1. Mol Endocrinol 11:1961–1970. 32. Kitanaka S, Takeyama K, Murayama A, et al. 1998 Inactivating mutations in the 25-hydroxyvitamin D3 1alphahydroxylase gene in patients with pseudovitamin D–deficiency rickets. N Engl J Med 338:653–661. 33. Yoshida T, Monkawa T, Tenenhouse HS, et al. 1998 Two novel 1alpha-hydroxylase mutations in French-Canadians with vitamin D–dependency rickets type I. Kidney Int 54:1437–1443. 34. Fu GK, Portale AA, Miller WL 1997 Complete structure of the human gene for the vitamin D 1alpha-hydroxylase, P450c1alpha. DNA Cell Biol 16:1499–1507. 35. Kitanaka S, Murayama A, Sakaki T, et al. 1999 No enzyme activity of 25-hydroxyvitamin D3 1alpha-hydroxylase gene product in pseudovitamin D–deficiency rickets, including that with mild clinical manifestation. J Clin Endocrinol Metab 84:4111–4117. 36. Wang X, Zhang MY, Miller WL, Portale AA 2002 Novel gene mutations in patients with 1alpha-hydroxylase deficiency that confer partial enzyme activity in vitro. J Clin Endocrinol Metab 87:2424–2430. 37. Wang JT, Lin CJ, Burridge SM, et al. 1998 Genetics of vitamin D 1alpha-hydroxylase deficiency in 17 families. Am J Hum Genet 63:1694–1702. 38. Smith SJ, Rucka AK, Berry JL, et al. 1999 Novel mutations in the 1alpha-hydroxylase (P450c1) gene in three families with pseudovitamin D–deficiency rickets resulting in loss of functional enzyme activity in blood-derived macrophages. J Bone Miner Res 14:730–739.

1205 39. Dardenne O, Prud’homme J, Arabian A, Glorieux FH, St-Arnaud R 2001 Targeted inactivation of the 25-hydroxyvitamin D(3)-1(alpha)-hydroxylase gene (CYP27B1) creates an animal model of pseudovitamin D–deficiency rickets. Endocrinology 142:3135–3141. 40. Panda DK, Miao D, Tremblay ML, et al. 2001 Targeted ablation of the 25-hydroxyvitamin D 1alpha-hydroxylase enzyme: evidence for skeletal, reproductive, and immune dysfunction. Proc Natl Acad Sci USA 98:7498–7503. 41. Balsan S, Garabedian M, Lieberherr M, Gueris J, Ulmann A 1979 Serum 1,25-dihydroxyvitamin D concentrations in two different types of pseudo-deficiency rickets. In Norman AW, Schaefer K, Herrath DV, Grigoleit H-G, Coburn JW, DeLuca HF, Mawer EB, Suda T (eds) Vitamin D: Basic Research and Its Clinical Application, de Gruyter: Berlin and New York, pp. 1143–1149. 42. Reade TM, Scriver CR, Glorieux FH, Nogrady B, Delvin E, Poirier R, Holick MF, DeLuca HF 1975 Response to crystalline la-hydroxyvitamin D3 in vitamin D–dependency. Pediatr Res 9:593–599. 43. Delvin EE, Glorieux FH, Marie PJ, Pettifor JM 1981 Vitamin D– dependency: Replacement therapy with calcitriol. J Pediatr 99:26–34. 44. Delvin EE, Salle BL, Glorieux FH, Adeleine P, David LS 1986 Vitamin D supplementation during pregnancy: Effect on neonatal calcium homeostasis. J Pediatr 109:328–334. 45. Anderson L, McDonald JR 1946 The origin, frequency, and significance of microscopic calculi in the kidney. Surg Gynecol Obstet 82:275–282. 46. Alon U, Brewer WH, Chan JCM 1983 Nephrocalcinosis: Detection by ultrasonography. Pediatrics 71:970–973. 47. Goodyear PR, Kronick JB, Jequier S, Reade TM, Seriver CR 1987 Nephrocalcinosis and its relationship to treatment of hereditary rickets. J Pediatr 11:700–704. 48. Glorieux FH 1990 Calcitriol treatment in vitamin D–dependent and vitamin D–resistant rickets. Metabolism 39:(Supp 1): 10–12.

CHAPTER 72

Hereditary 1,25-Dihydroxyvitamin D– Resistant Rickets PETER J. MALLOY J. WESLEY PIKE DAVID FELDMAN

I. II. III. IV. V.

Division of Endocrinology, Gerontology, and Metabolism, Stanford University School of Medicine, Stanford, California Department of Biochemistry, University of Wisconsin-Madison, Madison, Wisconsin Division of Endocrinology, Gerontology, and Metabolism, Stanford University School of Medicine, Stanford, California

Introduction The Clinical Features of HVDRR Mechanism of 1,25(OH)2D Action Cellular Basis of HVDRR Molecular Basis for HVDRR

I. INTRODUCTION Vitamin D, the primary regulator of calcium homeostasis in the body, is particularly important in skeletal development and in bone mineralization. The active form of vitamin D, 1α,25-dihydroxyvitamin D, [1,25(OH)2D3], functions by binding with high affinity to specific vitamin D receptors (VDR). The VDR is a member of the steroid-thyroid-retinoid receptor gene superfamily of nuclear transcription factors that regulate the expression of specific target genes in response to hormone binding. Hereditary vitamin D–resistant rickets (HVDRR) is a rare genetic disease that is due to a generalized resistance to 1,25(OH)2D3 [1–3]. HVDRR is caused by heterogeneous mutations in the VDR gene that alter the function of the receptor ultimately leading to complete or partial resistance to 1,25(OH)2D3. In this chapter on HVDRR, we describe the clinical manifestations of the disease and discuss the genetic defects in the VDR underlying the molecular basis for HVDRR. Over the years, a number of different names have been used to describe the condition caused by 1,25(OH)2D resistance. In addition to HVDRR, the disease has been referred to as vitamin D–dependent rickets type II (VDDR-II), pseudovitamin D deficiency type II (PDDR II), calcitriol-resistant rickets (CRR), and hypocalcemic vitamin D–resistant rickets (HVDRR). We use the term hereditary 1,25-dihydroxyvitamin D– resistant rickets (HVDRR) since this disease is now known to be caused by genetic defects in the VDR that VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

VI. Therapy of HVDRR VII. Alopecia VIII. Concluding Remarks References

lead to resistance to the action of 1,25(OH)2D3 [3]. The term HVDRR provides a better description of the disease as it is based on genetic defects that produce resistance; it is not dependent on 1,25(OH)2D but is due to resistance to 1,25(OH)2D3 action. In the Online Mendelian Inheritance in Man Web site (http:// www.ncbi.nlm.nih.gov/omim/), this disease is referred to as vitamin D–resistant rickets with end-organ unresponsiveness to 1,25-dihydroxycholecalciferol, ricketsalopecia syndrome, VDDR II with alopecia, and hypocalcemic vitamin D–resistant rickets. The notion that diseases could be due to hormone resistance emerged in 1937 when Albright et al. [4] described a patient with rickets who had normal serum calcium levels but low phosphate levels. The patient was treated with abnormally high doses of vitamin D and responded to the therapy. Keen observation of this case led the authors to suggest that the cause of the condition was due to end-organ resistance to vitamin D, and thus the concept of hormone resistance evolved. The patient they described appears to have had what is now known as X-linked hypophosphatemic rickets (XLH, described in Chapter 69). Twenty-four years later, Prader et al. [5] reported on two patients with rickets who were hypocalcemic and hypophosphatemic. These patients also responded to high doses of vitamin D and they referred to this condition as vitamin D–dependent rickets type I (VDDR-I). The cause of rickets in these individuals is due to an inborn error in the conversion of vitamin D to the hormonally active form 1,25(OH)2D. VDDR I arises from mutations in Copyright © 2005, Elsevier, Inc. All rights reserved.

1208

PETER J. MALLOY, J. WESLEY PIKE, AND DAVID FELDMAN

the gene encoding 25-hydroxy-1α-hydroxylase (1αOHase) [6], the enzyme that converts 25-hydroxyvitamin D3 (25(OH)D3) to 1,25(OH)2D3 [7,8]. This disease is described in Chapter 71. In 1978, the first cases of HVDRR were reported by Brooks et al. [9] and Marx et al. [10]. The patient in the Brooks et al. study exhibited hypocalcemia, hypophosphatemia, and secondary hyperparathyroidism. The clinical findings were similar to patients with VDDR I except that the patient had markedly increased serum levels of 1,25(OH)2D. Brooks et al. [9] postulated that the rickets was due to an end-organ resistance to 1,25(OH)22D3, and they named the syndrome vitamin D–dependent rickets, type II (VDDR II). Marx et al. [10] reported similar findings in two children and also suggested that the disease was due to end-organ resistance to 1,25(OH)2D. Since these initial studies there have been many reports of patients with apparent target organ resistance to 1,25(OH)2D [3]. In this chapter, we will review the clinical features and the genetic basis underlying the disease.

II. THE CLINICAL FEATURES OF HVDRR A. Clinical and Biochemical Findings HVDRR is manifested by a constellation of signs and symptoms caused by a generalized resistance to 1,25(OH)2D. The major feature of HVDRR is rickets that is due to defective mineralization of newly forming bone and pre-osseous cartilage. In HVDRR the rickets is generally displayed early, usually within months of birth. The rickets is usually severe and affected children suffer from bone pain, muscle weakness, and hypotonia. In the worst cases, convulsions due to the hypocalcemia have occurred. Children are often growth-retarded, and they frequently develop severe dental caries or exhibit enamel hypoplasia of the teeth [11–17]. Some infants have died from pneumonia as a result of poor respiratory movement due to severe rickets of the chest wall [12,15,18]. In many cases, children with HVDRR have sparse body hair and some have total scalp and body alopecia including eyebrows and in some cases eyelashes (Fig. 1). Alopecia will be discussed in more detail below. An example of the typical serum biochemistry levels found in HVDRR cases is shown in Table I. The abnormalities include low serum concentrations of calcium and phosphate and elevated serum alkaline phosphatase activity. The hypocalcemia leads to secondary hyperparathyroidism. The elevated parathyroid hormone (PTH) level then contributes to the

FIGURE 1

Children with HVDRR and alopecia. Reprinted with permission from The Journal of Pediatrics, JF Rosen, AR Fleischman, L Fineberg, A Hamstra, and HF DeLuca. Rickets with alopecia: An inborn error of vitamin D metabolism. 1979; 94:729–735.

hypophosphatemia. These clinical and biochemical findings are also common to patients with 1α-hydroxylase deficiency (Table II) as described in Chapter 71. On the other hand, in HVDRR patients the serum 25(OH)D values are normal and the 1,25(OH)2D levels are elevated. This singular feature distinguishes HVDRR from 1α-hydroxylase deficiency (VDDR-I) where the serum 1,25(OH)2D values are depressed or absent. When analyzed, the 24,25(OH)2D levels have been normal or low [12,15,19–25]. Patients with 1α-hydroxylase deficiency (VDDR I) can be successfully treated with physiologic doses of calcitriol that circumvent the 1α-hydroxylase deficiency and restore the circulating 1,25(OH)2D levels to normal. In contrast, patients with HVDRR do not respond to physiologic doses of calcitriol and most patients are resistant to even extreme supra-physiologic doses of all forms of vitamin D therapy (Table II).

1209

CHAPTER 72 Hereditary 1,25-Dihydroxyvitamin D–Resistant Rickets

TABLE I Biochemical Profile of a Patient with HVDRR on Therapy Biochemical marker Calcium (mmol/liter) Phosphate (mmol/liter) ALP (IU/liter) 25(OH)D (nmol/liter) 1,25(OH)2D (pmol/liter) PTH (pmol/liter)

Normal values

Referral values

40 daysa

80 daysb

100 days

2.2–2.6 1.4–2.2 145–320 25–85 40–105 <8

1.86 1.0 3056 30 521 –

1.77 1.0 3991 37.4 953 34.2

1.80 1.0 3800 250 1830 69.9

1.71 0.9 3609 211 1560 64.5

aTreatment:

250 mg elemental calcium 4 times per day and 0.5 mg calcitriol (Rocaltrol) twice per day and 20,000 IU vitamin D 3 daily. 250 mg elemental calcium 4 times per day and 5 mg calcitriol twice per day. ALP, alkaline phosphatase. Adapted from Zhu et al.: J Bone Miner Res 13:259–264, 1998 with permission of the American Society for Bone and Mineral Research. bTreatment:

HVDRR is inherited as autosomal recessive disease. The recessive nature of the disease is evident from the patient’s parents and siblings who are heterozygous for the genetic trait, but show no symptoms of the disease and have normal bone development. In most cases, consanguinity in the family lineage can be found and intermarriage is highly associated with the disease. Males and females are equally affected and often a family has several affected children [26].

B. Pathophysiology In HVDRR, the intestine, and other target organs, including bone, the parathyroid glands, and kidneys, are resistant to 1,25(OH)2D action. Without vitamin D action, the intestine becomes less efficient in promoting

TABLE II

A Comparison of 1α-hydroxylase Deficiency and HVDRR

Feature Gene mutated Autosomal recessive Manifested at early age Rickets Hypocalcemia Alopecia PTH 25(OH)D levels 1,25(OH)2D levels Response to physiological doses of 1,25(OH)2D3 a1α-Hydroxylase

1α-Hydroxylase deficiencya

HVDRR

CYP27B1 Yes Yes Yes Yes No Elevated Normal Low Yes

VDR Yes Yes Yes Yes Sometimes Elevated Normal Elevated No

deficiency is also known as VDDR I or PDDR.

calcium and phosphate absorption into the circulation. It is now well established that the biological actions of 1,25(OH)2D are mediated by the VDR, a nuclear transcription factor that regulates gene expression in 1,25(OH)2D–responsive cells (see Chapter 13). As will be discussed in detail below, the hallmark of the HVDRR syndrome is resistance to 1,25(OH)2D action. It is now clear that the usual cause of HVDRR is due to mutations in the VDR that render the receptor non– functional or less functional than the wild-type VDR. The primary biological process attributed to vitamin D is maintenance of calcium and bone homeostasis. 1,25(OH)2D is essential for promoting the transport of calcium and phosphate across the small intestine and into the circulation. Adequate delivery of calcium and phosphate to the bone is essential for the normal mineralization of bone. Approximately half of the total calcium absorption by the intestine is attributed to 1,25(OH)2D action while passive absorption accounts for the remaining half [27,28]. Since vitamin D regulates the translocation of calcium, conditions that adversely affect the 1,25(OH)2D action pathway cause a decrease in mineral transport leading to hypocalcemia. The hypocalcemia and the resistance of the parathyroid gland to suppression by 1,25(OH)2D because of defective VDR within the gland, in turn, results in secondary hyperparathyroidism. The increase in circulating 1,25(OH)2D levels are due to an increase in renal 1αhydroxylase activity caused by both elevated PTH and hypophosphatemia to up-regulate 1α-hydroxylase gene expression, as well as failure of elevated 1,25(OH)2D to suppress 1α-hydroxylase. The hypophosphatemia results from the elevated PTH down-regulating the Na/P cotransporter and/or by the loss of a functional VDR in the kidney, as well as decreased intestinal absorption (see Chapters 26, 29, and 69). The calcium and phosphate deficiencies compromise normal bone mineralization leading to rickets in children and osteomalacia in adults.

1210 C. Alopecia Alopecia totalis (sometimes called atrichia) is a clinical feature that is found in many patients with HVDRR (Fig. 1). Some patients have sparse body hair and some exhibit total scalp and body alopecia [21,29,30]. Children with extreme alopecia often lack eyebrows and in some cases eyelashes. Hair loss may be evident at birth or occurs during the first few months of life. An analysis of HVDRR patients shows that there is some correlation between the severity of rickets and the presence of alopecia [30]. Patients with alopecia are generally more resistant to calcitriol therapy than those without alopecia. In families with a prior history of the disease, the absence of scalp hair in newborns provides initial diagnostic evidence for HVDRR. The mechanism causing alopecia is unknown but VDRs are present in the hair follicle [31,32]. Skin biopsy has revealed apparently normal follicles with no hair. The lack of 1,25(OH)2D action during a critical stage of hair follicle development is the suspected cause of alopecia. It is interesting to note that alopecia is not associated with other diseases of vitamin D deficiency. Alopecia is also discussed in Chapters 20 and 35.

PETER J. MALLOY, J. WESLEY PIKE, AND DAVID FELDMAN

VDRs have also been found in hematolymphopoietic cells and 1,25(OH)2D has been shown to regulate cell differentiation and the production of interleukins and cytokines [42] (see Chapter 36). Neutrophils isolated from HVDRR patients exhibit only minor aberrations in their fungicidal activity [43], and HVDRR patients have no clinically apparent immunologic defects. In the light of the diverse actions of 1,25(OH)2D demonstrated in many tissues, the absence of related findings in children with HVDRR suggests that the pleiotropic responses regulated by 1,25(OH)2D in nonosteogenic tissues are redundant and that other factors or compensatory mechanisms subsume the role of vitamin D in such a way that abnormalities are not clinically manifested. This possibility does not necessarily minimize the contribution of 1,25(OH)2D to these systems under normal physiologic conditions. Similarly, the VDR knockout mouse displays the same phenotypic and physiologic patterns as patients with HVDRR [44,45]. The VDR knockout mouse model can be used to analyze the abnormalities caused by the loss of VDR action in detail that is not possible in the HVDRR patients (see Chapter 20).

III. MECHANISM OF 1,25(OH)2D ACTION D. Other Aspects of HVDRR

A. The Vitamin D Receptor

As mentioned above and discussed extensively in this volume, in addition to maintaining calcium homeostasis, 1,25(OH)2D has been shown to regulate a number of biological processes in many tissues [33–39]. Although there are multiple pleiotropic tissue responses regulated by 1,25(OH)2D, children with HVDRR appear relatively normal except for the constellation of features that relate to their calcium deficiency, rickets, and alopecia. VDRs have been found in endocrine glands such as pituitary, pancreas, parathyroid, gonads, and placenta, and 1,25(OH)2D3 regulates hormone synthesis and secretion from these glands [33–38]. Hochberg et al. [40] examined insulin, thyrotropin (TSH), prolactin (PRL), growth hormone (GH), and testosterone levels in sera from patients with HVDRR and found no abnormalities in hormone secretion. Furthermore, Even et al. [41] showed that urinary cAMP and excretion of potassium, phosphorus, and bicarbonate were normal in HVDRR patients following a PTH challenge. However, PTH failed to decrease urinary calcium and sodium excretion in these patients to the extent found in the control patients. This suggests that 1,25(OH)2D may selectively modulate the renal response to PTH and facilitate the PTH-induced reabsorption of calcium and sodium [41].

The VDR is a ligand-activated nuclear transcription factor and a member of the steroid-thyroid-retinoid receptor gene superfamily [46]. The VDR is similar in overall structure to the other members of the steroid-thyroid-retinoid receptor superfamily, having a highly conserved DNA-binding domain (DBD) and a more variable ligand-binding domain (LBD) (see Chapter 11). The VDR gene is located on chromosome 12 and is composed of 14 or more exons (Fig. 2). Exons 2 and 3 encode the DBD and exons 4 through 9 encode the LBD. The human VDR contains either 424 or 427 amino acids due a polymorphism in a transcription start site. Differential use of promoters and alternative splicing may lead to even longer forms (see Chapter 12). 1. DNA BINDING DOMAIN (DBD)

The DBD extends from amino acids residues 24–90 at the N-terminus of the VDR (Fig. 2). The DBD folds into two loops or modules of 12–13 amino acids each. Each module contains four cysteine residues that coordinate the binding of one zinc atom forming a “zinc-finger” structure. The two zinc modules of the VDR are not topologically equivalent and serve different functions within the protein [47]. In the first zinc finger

1211

CHAPTER 72 Hereditary 1,25-Dihydroxyvitamin D–Resistant Rickets

VDR Chromosome 12

Exons 1F 1A 1B

ATG 1C 2

TGA 3

45 6

9

7 8

1E 1D

DNA ~75kbp

F H G T F 70 D A N N K 30 R P Box A T R D Box D I M R R G H T C C D C C 40 G ++ V ++ Q E N zn zn G G F A P C C C C I K RSM T 60 R R G F 80 F V P K R KA L F 50 L K R C V D I G M MK E F 20

Protein ~48kda AF-2 H1

H3

H4 H5 S1-3H6 H7 H8

E1

DNA binding domain

H9

H10 H11 H12

β-turn

Ligand binding domain

FIGURE 2 Arrangement of the chromosomal gene and domains of the VDR. The structural organization of the human VDR gene which spans approximately 75 kilobases of DNA is shown [187]. The location of the start (ATG) and termination (TGA) codons are indicated. The exons encoding the various domains and structural motifs in the VDR protein are shown. The VDR DNA binding domain is comprised of two zinc finger modules each of which contains 4 invariant cysteine residues that function to coordinate a single zinc atom. Two α-helices (helix A and B) shaded in the diagram are located on the carboxy terminal side of each zinc module. Amino acid residues essential to functional interaction of these α-helices with either DNA or with RXR are boxed and designated the P-box and D-box, respectively. In the ligand-binding domain the position of the α-helices (H1–H12) and β-turns (S1–3) are shown as shaded and hatched boxes respectively. The E1 and AF-2 regions are indicated.

module (the most N-terminal), an α-helix known as the P-box (residues 42–46), functions to direct specific DNA-binding in the major groove of the DNA-binding site. In the second zinc finger module, an α-helix known as the D-box (residues 61–65) serves as a dimerization interface for interaction with the retinoid X receptor (RXR) DNA-binding domain [48,49]. Immediately downstream of the second zinc finger (aa residues 90–101) lies an extended α-helix known as the T-box. This T-box region also likely interacts with RXR and makes minor groove contacts with nucleotides in the DNA half-sites of VDREs (see Chapters 13 and 14). As will be discussed below, mutations in critical amino acids within both zinc finger modules render the VDR nonfunctional and cause HVDRR, presumably by interfering with VDR binding to DNA. 2. LIGAND BINDING DOMAIN (LBD)

The structure of the VDR LBD that stretches over two-thirds of the protein from amino acids 123 to 427 has been determined by X-ray crystallography [50] (see Chapter 15). The VDR LBD is formed by 12

α-helices (H1–H12) and 3 β-sheets (S1–S3) (Fig. 2). The LBD also has a variable length region between helix H1 and helix H3 (loop 1–3). Conserved residues located in a 34 amino acid cluster from the C-terminus of helix H3 to the middle of helix H5 form the hydrophobic core. These residues hold together H3, H4, H5, H8, and H9, and the interhelical loops 3–4 and 8–9. Residues in H1, H3, H5, β-turn, loop 6–7, H11, loop 11–12 and H12 form the framework for a 3-dimensional ligand binding pocket. Helix H12 is thought to form a retractable lid that traps and holds the ligand in position. Once inside the binding pocket, the ligand makes contact with specific amino acid residues of the various α-helices that line the face of the pocket thereby transmitting its signal (see Chapter 15). Ligand binding causes a conformational change in the VDR that allows the receptor to form protein heterodimers with RXR. VDR elements involved in RXR heterodimerization are located within the LBD, and include H9, H10, and an E1 domain that overlaps H4 and H5 (Fig. 2). Two regions of the VDR LBD are essential for its transactivation function. These regions include the AF-2 domain that

1212 encompasses H12 (residues 416–424) and the region between amino acids 232–272 encompassing H3 and H4 [50]. Repositioning of helix H12 after ligand binding is critical to the formation of a hydrophobic cleft that can accommodate the LxxLL motif of coactivators and allow them to bind to the receptor (see Chapter 16). Ligand binding also results in exposure of other regions of the VDR that act to recruit coactivator proteins or facilitate contact with proteins associated with the core transcriptional machinery, such as TFIIB or the TAFs [51,52]. The coactivators, such as SRC-1, are active in modifying chromatin [53–55] (see Chapter 19). Mutations in the VDR LBD may cause HVDRR through a number of mechanisms. They may completely prevent ligand binding or reduce its affinity for 1,25(OH)2D3. Alternatively, mutations may alter VDR conformation compromising its ability to heterodimerize with RXR, bind to DNA, or interact with coactivators.

B. Regulation of Gene Expression by 1,25-Dihydroxyvitamin D The mechanism of action of vitamin D is detailed in multiple chapters in Section II of this book. Thus, we will briefly review only those aspects of vitamin D action particularly relevant to understanding HVDRR and the mutations responsible for causing this syndrome. The biological actions of 1,25(OH)2D in tissues and cells are orchestrated through complex changes in gene expression [34,56–58]. These changes lead to cell-specific alterations in the level of proteins directly responsible for a myriad of differentiated cell functions, as well as in proteins that function as transcription factors or as signaling molecules to regulate secondary and tertiary levels of gene expression [59]. In the latter case, these molecules may function directly within the cell or indirectly via additional cellular signaling pathways in either autocrine or paracrine fashion. As indicated earlier, most, if not all, of the molecular actions of 1,25(OH)2D in the nucleus are mediated by the VDR. A simplified model of 1,25(OH)2D activated gene transcription is shown in Fig. 3. A more comprehensive analysis of 1,25(OH)2D action is presented in Chapter 13. In brief, after synthesis in the kidney, 1,25(OH)2D circulates in the blood mostly bound to DBP and perhaps other carriers with a small fraction of hormone in the free state (see Chapter 8). The free, fat-soluble hormone is believed to enter target cells through the lipid bilayer of the cell membrane, although additional complex mechanisms may play a role (see Chapter 10). Once inside the cell, 1,25(OH)2D3 activates the VDR, prompting translocation to the nucleus, whereupon a

PETER J. MALLOY, J. WESLEY PIKE, AND DAVID FELDMAN

series of changes occur that enable the VDR to activate gene transcription [60]. The mechanism of action is described in great detail in Section II of this book. One event that happens following 1,25(OH)2D–binding is that the VDR is posttranslationally modified by phosphorylation [61]. Phosphorylation may increase transactivation, however, its precise role in the activation of the VDR is currently unclear. Ligand binding promotes VDR-RXR heterodimerization and DNA binding. The VDR-RXR heterodimer complex binds with high affinity through their DBDs to vitamin D response elements (VDREs) located in the promoter region of target genes [57]. The typical VDRE contains two imperfect hexanucleotides arranged as direct repeats separated by a 3 nucleotide base spacer. In the LBD, 1,25(OH)2D3 makes contact with specific amino acid residues lining the ligandbinding pocket [50]. 1,25(OH)2D3 binding triggers helix H12 to fold back upon the LBD enclosing the ligand. A coactivator-binding cleft is formed by the repositioning of helix H12 together with helix H3, allowing the recruitment of coactivators and other transcription factors [62]. The VDR-complex together with the general transcription apparatus drives the transcription of 1,25(OH)2D-responsive genes that ultimately determine the cellular response to the hormone. The proof that the cause of HVDRR was defective regulation of gene expression by VDR was initially developed through studies of HVDRR where natural mutations in the VDR gene prevented 1,25(OH)2D3 induction of target genes, such as 24-hydroxylase [20,22,63]. Further studies demonstrated that 1,25(OH)2D3 could regulate promoter activity and that specific DNA sequences within the promoter were required for recognition by the VDR-RXR heterodimer. Direct regulation of gene expression by 1,25(OH)2D3 has been demonstrated using the human [64–66] and rat [67–69] osteocalcin gene, the osteopontin gene [70], the calbindin genes [71,72], and the 25hydroxyvitamin D3 24-hydroxylase genes [73–75]. HVDRR mutant VDRs were eventually shown to be incapable of activating such promoter constructs both supporting the critical role of functional VDR in transactivation as well as defining the defect causing HVDRR [76–78].

IV. CELLULAR BASIS OF HVDRR A. Initial Studies Using Cultured Skin Fibroblasts The syndrome of HVDRR was first recognized as an entity in 1978–79 [9–11,79]. At the present time, more than 100 patients with vitamin D resistance

1213

CHAPTER 72 Hereditary 1,25-Dihydroxyvitamin D–Resistant Rickets

Mechanism of action Cytoplasm O H

O H

O H

Ligand-binding CH2 O H

H O

H O

CH2 O H

CH2 O H

H O

VDR

O H

H O

RXR O H

H O

CH2 O H

DBP Hetrodimerization

CH2 O H

1,25-Dihydroxyvitamin D3

O H

RXR

Nuclear translocation Nucleus DNA-binding

H O

O H

VDR

g160 Coactivator recruitment

DRIPs

O H CH O 2 H

H O

Z Zn

GTA

mRNA

ZZn 3

VDRE Response

Protein

AAA

FIGURE 3

Mechanism of 1,25-dihydroxyvitamin D3 action. 1,25-dihydroxyvitamin D3 circulates in the blood mainly bound to the vitamin D–binding protein (DBP) with a small amount in equilibrium with the free or unbound state. The free lipophilic ligand diffuses through the lipid bilayer of the cell membrane and binds with high affinity to the VDR. In the nucleus the ligand-bound VDR heterodimerizes with RXR and binds to VDREs via the two zinc finger modules of the DNA–binding domain of the receptors. Ligand binding also induces changes in the VDR that allows it to recruit coactivators. The VDR-RXR–coactivator complex interacts with the general transcription apparatus (GTA) and initiates gene transcription. The physiologic response to the hormone is manifested by the newly synthesized proteins that then elicit intracellular or extracellular activities.

have been described [3]. A summary of these cases is shown in Table III. Throughout this chapter, the HVDRR cases are denoted by a family number, e.g. F1, F2, etc. Studies to elucidate the nature of the defect in HVDRR cases began soon after receptors for 1,25(OH)2D3 were found in skin [31,80–82]. Feldman et al. [81] demonstrated that the VDR was present in fresh and cultured human foreskin as well as in cultured keratinocytes and dermal fibroblasts grown from adult skin biopsies. The studies to unravel the cause of the disease began in earnest in 1981 when Eil et al. [83]

showed in cultured skin fibroblasts that the cause of the cellular defect in patients (families F1 and F3) with HVDRR was due to the defective nuclear uptake of 1,25(OH)2D3. The following year, Feldman et al. [20] analyzed the VDR in cultured skin fibroblasts from two siblings with HVDRR (F11). In this study, they demonstrated that cytosolic extracts of cultured fibroblasts from the HVDRR patients had undetectable levels of [3H]1,25(OH)2D3 binding. Furthermore, when cultured fibroblasts from normal subjects were treated with 1,25(OH)2D3, they were able to demonstrate an increase in 24-hydroxylase activity, a well-characterized

TABLE III Familya F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 F11

F12 F13 F14 F15 F16 F17 F18 F19 F20 F21 F22 F23 F24 F25 F26 F27

Patient name/ descriptionb IIB, patient 1, 1a IIC, patient 2, 1b Patient Patient 1, patient 2a, 2a Patient 2, patient 2b, 2b Patient Patient, patient 3, 3, kindred 3, P3 K.N. Patient Patient M.A., kindred 6, patient 6, 6 Patient I.H., A1, patient 2, I.K., case 1 R.K., patient 1, A2, case 2 Patient 4 Patient 5 Patient Patient A, patient 5 Patient B, patient 4 B patient S.H., patient 3, case 3, C1 R.H., patient 4, case 4, C2 D1 D2 Kindred 7, patient 7, 7, P7 I.S., patient Patient Patient Patient 1, N.D. Patient 1 Patient 2 Patient 3

Ethnic origin

Compilation of HVDRR Cases

Consanguinity

Onset age

No No No Yes Yes No Yes

20 months 5 months 15 yr 1 yr 1 yr 2 yr 10 months

Arab

Yes No ? No

15 months 18 months 45 yr 1 yr

Arab

No Yes

12 yr 1 yr

Yes Yes Yes Yes Yes Yes ? No No Yes Yes

<1 yr

Japanese

Arab Haitian

Kuwait Hispanic Saudi Arab Japanese Japanese Japanese

Yes No Yes Yes Yes No No

? 5 months 2 yr 19 months 8 months 6 months 1 month ? ? 1 yr 1 yr 4 months 2 months 2 months 2 months

Male

Female

1 1 1 1 1 1 1 1 1 1 1 1 1 1

1 2 1

2 1

1 1 3

1 1 1 1 1 1 1

Total affected

Alopecia

2

No

1 2

No Yes

1 1

No Yes

1 1 1 1

Yes Yes ? Yes

1 2

No Yes

1 1 1 1 4 2 1 2

Yes Yes Yes No Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes

3 2 1 1 1 1 2 1 1

1,25D binding

VDR mutation

+

+ + +

Ile314Ser Arg80Gln

−

−

Tyr295stop

− ± −

Tyr295stop

+

[10,83,95] [10,83] [9] [11,83,86,133] [11,83,86,95] [79,85,140] [12,86,95,110] [96,160] [13] [161] [19,86]

−

− ± − − − + + + − ± + −

References

Tyr295stop Tyr295stop Arg73Gln Arg73Gln Arg 80Gln Arg274Leu Gly46Asp

Arg50Gln

[14] [20,21,63,129] [21,29,63,129] [86,87] [86,87] [15] [15] [21] [21,29,63,77] [21,29,63,77] [22,76] [22,76] [95,110,133] [18,132,139] [96] [117,118] [97,100] [23] [23,111] [23]

TABLE III Patient name/ descriptionb

Familya F28 F29 F30 F31

F41 F42

Patient 2, M.T. Patient, line 10 Line 15 G1 G2 Line 11, patient 1 Line 11b, patient 2 Patient 1, patient 1a, patient 4a Patient 2, patient 1b, patient 4b E1 F1 H1 J1 K1 L1 Ro-VDR, brother Al-VDR, sister Ab-VDR Patient

F43 F44 F45

Child Patient II, case 2 Propositus

F46 F47 F48 F49

Line 14 Patient I J.K. N1 N2 Patient

F32 F33

F34 F35 F36 F37 F38 F39 F40

F50 F51

Patient Sister

Ethnic origin

Consanguinity

Persian-Jewish Saudi Saudi Arab Turkish Japanese

Arab Arab Arab Arab Arab Arab

Yes Yes

Compilation of HVDRR Cases—Cont’d Onset age

Male

9 months 13 months

Yes Yes Yes Yes Yes

5 weeks

Yes

16 months

16 months

Yes Yes No No No Yes

Total affected

Alopecia

1 1

1 2

Yes Yes

1 1

2

Female

1 1 1

2 2

1 1 1 1 1 1 1 1

1 1 1 1 1 1 2

1 1

Tunisian JapaneseBrazilian Moroccan Mauritius English Tunisian-Jewish Greek Turkish

Yes Yes Yes No No Yes Yes No Yes Yes

16 months 7 months 6 months 9 months 3 months

1 1

1 1

1 1 1

1

2 1 1 2

1 1 1 1 1

1 2

1

1,25D binding

VDR mutation

References

Yes Yes Yes Yes Yes

− − − + + − − +

Tyr295stop Tyr295stop Gly33Asp Gly33Asp Gln152stop Gln152stop Arg50Gln

[97,100] [17,130,133] [130] [76,89] [76,89] [132,133,162] [130,162] [111,112]

Yes

+

Arg50Gln

[111,112]

Yes Yes Yes Yes Yes Yes No No Yes

− − − − −

Tyr295stop Tyr295stop Tyr295stop Tyr295stop Tyr295stop

−

Exon 7–9 deletion Cys190Trp Lys45Glu His35Gln

[26,77] [26] [26,77] [26] [26] [26] [101] [101] [101] [138]

Yes Yes

+ +

Yes Yes Yes Yes Yes Yes

− + + + +

No No

+ +

Arg73stop Phe47Ile None Arg80Gln Arg80Gln Eliminate splice His305Gln His305Gln

[138] [114,116] [24] [130] [114,115] [151,152] [25] [25] [137] [141,142] [141,142] Continued

TABLE III Patient name/ descriptionb

Familya F52 F53 F54 F55 F56 F57

F60 F61 F62

Patient 2 Patient Patient Patient 1, B.G. Patient 2, A.H. patient 3, A.J. patient 4, U.A. patient patient 1 patient 2 patient patient patient

F63 F64 F65

patient patient FC

F58 F59

a Number b Name

Ethnic origin

Consanguinity

Compilation of HVDRR Cases—Cont’d Onset age

Male

Female 1

French-Canadian Brazilian Greek German Indian Indian Hmong Algerian Algerian Iranian

Yes No Yes Yes Yes Yes Yes Yes

Caucasian

No

Saudi Saudi Arab

Yes Yes Yes

2 yr 7 months 15 months 2.5 yr 4 months 5 months

1 1 1 1 1

1

Total affected

Alopecia

1 1 2

2 1

1 1

2 1

1

20 months

1

1 1

1

2

3

1

assigned to family for citation. or description of propositus used in references. In some cases multiple designations were used for the same patient.

1,25D binding

Yes Yes Yes Yes Yes Yes Yes Yes No No No Yes Yes

+ − nd − − + + + + + nd

No Yes Yes

+ -

VDR mutation Arg391Cys Arg30stop Arg30stop Arg73stop Create splice Gln259Pro Gln259Pro Phe251Cys Trp286Arg Trp286Arg Glu420Lys Gln317stop Glu329Lys/ 366delC Ile268Thr Arg30stop Tyr295stop

References [140] [134] [135] [131] [131] [131] [131] [146] [144] [144] [149] [136] [150] [145] Unpublished Unpublished

CHAPTER 72 Hereditary 1,25-Dihydroxyvitamin D–Resistant Rickets

marker of 1,25(OH)2D3 responsiveness. However, the patients’ fibroblasts failed to induce 24-hydroxylase activity following hormone treatment, demonstrating that they were resistant to high concentrations of 1,25(OH)2D3. Subsequently, a number of other HVDRR cases were examined using cultured skin fibroblasts [15,84–86] or cells derived from bone [87]. Some patients’ fibroblasts lacked specific [3H]1,25(OH)2D3 binding [15,84–87], similar to the cases reported by Feldman et al. [20]. On the other hand, some fibroblasts exhibited normal [3H]1,25(OH)2D3 binding but were unresponsive to 1,25(OH)2D3 treatment [15,22, 85,87–89]. The conclusion reached was that the HVDRR syndrome was caused by cellular resistance to 1,25(OH)2D3 action and was due to at least two types of defects in the VDR–one that impairs ligand binding, and one that retains ligand binding, but causes resistance to 1,25(OH)2D3 by a defect downstream of binding. Griffin and Zerwekh [85] and Liberman et al. [86,87] also used 24-hydroxylase activity to demonstrate 1,25(OH)2D3 resistance. Clemens et al. [84], on the other hand, showed that fibroblasts from HVDRR patients were not growth-arrested following hormone treatment in contrast to fibroblasts from healthy individuals that were growth-arrested. These early observations showed that cells from HVDRR patients were resistant to 1,25(OH)2D3, and that there were a variety of abnormalities in the VDR. As the number of reports on HVDRR increased, the heterogeneous nature of the defects in the VDR became more apparent. Hochberg et al. [21,29] reported the clinical findings in four patients from two unrelated families of Arab origin (F11,F18) who exhibited HVDRR and alopecia. A follow-up study by Chen et al. [63] showed that fibroblasts from three of these patients and a patient from an unrelated family from Germany (F17) had no [3H]1,25(OH)2D3 binding and 1,25(OH)2D3 treatment failed to induce 24-hydroxylase activity. Pike et al. [90] used a radioligand immunoassay [91] and a monoclonal antibody to the chick VDR [92–94], to demonstrate the presence of an immunoreactive protein in cell extracts from fibroblasts of HVDRR patients that exhibited no ligand binding. The authors speculated that the lack of [3H]1,25(OH)2D3 binding in these patients was not due to defective synthesis of the VDR protein, but was due to defects in the VDR LBD that prevented ligand binding [90]. Liberman et al. [88] described four cases (F1,F3,F5,F20) with normal ligand binding and 1,25(OH)2D3 resistance. Two of the cases (F5, F20) exhibited VDRs with a low affinity for DNA similar to the F19 and F31 families. Gamblin et al. [95] examined 1,25(OH)2D3 induction of 24-hydroxylase activity

1217

in F5 and F20 fibroblasts and demonstrated complete hormone resistance. In the other cases, F1 and F3, Liberman et al. [88] demonstrated that the VDRs had a reduced ability to localize to the nucleus despite showing a normal affinity for DNA. Gamblin et al. [95] further showed that the F1 and F3 fibroblasts exhibited 24-hydroxylase activity when exposed to high concentrations of 1,25(OH)2D3. Patients F1 and F3, whose fibroblasts showed a response to high concentrations of 1,25(OH)2D3 in vitro, also exhibited a calcemic response to high doses of calciferols in vivo. Castells et al. [96] also reported on a patient (F22) who had sparse hair, rickets, and high circulating 1,25(OH)2D3 levels. Studies of the VDR from the patient’s fibroblasts showed that the receptor had a decreased affinity for [3H]1,25(OH)2D3. The patient showed a marked improvement after treatment with extremely high doses of 1,25(OH)2D3, apparently overcoming the low affinity binding abnormality in the VDR. Based on the concept that VDR binding to DNA is essential for activity, Hirst et al. [22] showed that defective DNA binding could be the cause of resistance in cases that had normal ligand binding. In a study of a family from Haiti (F19) with two sisters with HVDRR and an unaffected sister, they showed that the fibroblasts from the affected individuals had normal [3H] 1,25(OH)2D3-binding but were resistant to 1,25(OH)2D3 treatment. The fibroblasts from the two sisters with HVDRR, despite normal [3H]1,25(OH)2D3 binding, were unresponsive to 1,25(OH)2D3 treatment. The authors further demonstrated that the VDR from the patient’s fibroblasts exhibited a significant decrease in affinity for general DNA. Using calf thymus DNAcellulose chromatography, the VDR from the unaffected sister bound strongly to DNA requiring high concentrations of KCl for elution (170–173 mM KCl), whereas the patient’s VDR bound weakly to DNA and eluted at lower concentrations of KCl (105–109 mM). A subsequent study by Malloy et al. [89] demonstrated a similar DNA binding defect in the VDR from HVDRR patients (F31) who had normal [3H]1,25(OH)2D3 binding. DNA-cellulose chromatography clearly revealed that the patient’s VDR had a low affinity for DNA. Furthermore, the authors showed that the cells from the parents had two forms of the VDR, one with a high affinity for DNA (eluting at 200 mM KC1) and the other with a low affinity for DNA (eluting at 100 mM KCl). This was the first clear evidence of heterozygosity in parents of HVDRR children. It was suspected that the defects in these cases would likely be due to point mutations in the DBD [22,89], that later proved to be correct [76].

1218 B. Studies in Other Cell Types In addition to studying cultured skin fibroblasts, a number of other cell types have been used to study HVDRR cases. These include peripheral mononuclear cells [97], phytohemagglutinin (PHA)-stimulated lymphocytes [98,99], myeloid progenitor cells [100], Epstein-Barr virus (EBV) immortalized B lymphoblasts [26,76,77,89], and HTLV-1 virus immortalized T lymphoblasts [101]. It is interesting to note that although EBV immortalized B lymphoblasts from normal subjects express wild-type VDR, they nevertheless fail to induce 24-hydroxylase activity or to show inhibition of cell growth in response to 1,25(OH)2D3 [26]. On the other hand, PHA-stimulated lymphocytes and HTLV-1 immortalized T lymphoblasts from normal subjects do respond to 1,25(OH)2D3 [101,102]. Using PHA-stimulated lymphocytes, Takeda et al. [99] showed that HVDRR can rapidly be diagnosed by the failure of 1,25(OH)2D3 to inhibit DNA synthesis or induce 24-hydroxylase activity [98,99]. Takeda et al. [99] also showed that PHA-stimulated lymphocytes from parents of children with HVDRR express intermediate levels of 24-hydroxylase in response to 1,25(OH)2D3, whereas other studies have failed to demonstrate defects in cells from parents or unaffected siblings.

V. MOLECULAR BASIS FOR HVDRR A. First Description of a Genetic Defect in the Nuclear Receptor Superfamily The biochemical and cellular data obtained from the earlier studies of HVDRR patients provided a framework to begin the search for the molecular cause of this disease. Investigations to determine the specific mutations in the VDR that cause HVDRR began shortly after the human VDR cDNA sequence was elucidated by Baker et al. [103]. During this same time period, the polymerase chain reaction (PCR) technique [104] was developed that provided a method to rapidly amplify genes for sequence analysis. The amino acid sequence of the VDR cDNA suggested the presence of highly conserved zinc-finger structures that were thought to be involved in DNA binding. The initial focus was HVDRR patients who had normal [3H]1,25(OH)2D3 binding but abnormal binding to DNA, since it was suspected that this defect would arise from mutations in the VDR DBD. In 1988, Hughes et al. [76] used PCR to amplify exons of the VDR gene from DNA isolated from the F19 and F31 families that were defective in DNA binding [22]. The patients in the F31 family were shown to have a unique G to A single base change in exon 3 that encodes the second zinc module

PETER J. MALLOY, J. WESLEY PIKE, AND DAVID FELDMAN

of the DBD. This missense mutation replaced arginine with glutamine at amino acid residue 73 (Arg73Gln) (Fig. 4). In the F19 family a G to A transition was identified in exon 2 that encodes the first zinc module of the DBD. This missense mutation changed glycine to aspartic acid at amino acid residue 33 in the first zinc finger module (Gly33Asp) (Fig. 4). Only the mutant sequences were found in the children with HVDRR. Their parents had a normal and a mutant sequence, demonstrating the genetic transmission and recessive nature of the disease. The study by Hughes et al. [76] was the first description of a genetic defect in any member of the steroid-thyroid-retinoid receptor gene superfamily. Mutations have now been found in many of the classical receptors including thyroid receptor (TR) [105], androgen receptor (AR) [106], estrogen receptor (ER) [107], glucocorticoid receptor (GR) [108], and mineralocorticoid receptor (MR) [109]. To determine whether the missense mutations found in these families were the cause of 1,25(OH)2D3 resistance in the HVDRR patients, the Arg73Gln and the Gly33Asp mutations were generated in the wild-type VDR cDNA by site-directed mutagenesis [78]. The recreated mutant VDRs were expressed in COS-1 cells and exhibited normal [3H]1,25(OH)2D3 binding and weak binding to DNA similar to the VDR in the patient’s fibroblasts. In addition, Sone et al. [78] demonstrated that the mutant VDRs were transcriptionally inactive in cotransfection experiments in CV-1 cells. Using an osteocalcin-CAT reporter plasmid, they showed that CAT activity could be induced by the wild-type VDR but not by the Arg73Gln or the Gly33Asp mutant VDRs [78]. These data proved that the missense mutations caused the VDR to be transcriptionally inactive and were the cause of 1,25(OH)2D3 resistance in the patients. Since the original report by Hughes et al. [76], a number of VDR mutations have been identified in patients with HVDRR. Over 100 cases of HVDRR have been recorded, and a number of these have been analyzed at the biochemical and molecular level [3] (Table III). Several genetic abnormalities have been found in the VDR gene, mainly missense and nonsense mutations, but also splice site mutations and a gene deletion. A description of these mutations and the consequences of the abnormality in the VDR will be discussed below.

B. Mutations in the VDR DNA Binding Domain (DBD) 1. DESCRIPTION OF DBD MUTATIONS

Since the initial report by Hughes et al. [76], a number of mutations have been identified in the VDR DBD.

1219

CHAPTER 72 Hereditary 1,25-Dihydroxyvitamin D–Resistant Rickets

Mutations in the VDR DNA binding domain

70 30

D Box P Box 40

90

20

FIGURE 4

Model of the DNA-binding domain of the VDR and location of mutations causing HVDRR. The two zinc finger modules and the amino acid composition of the DBD are shown. Conserved amino acids are depicted as shaded circles. Numbers specify amino acid number.

The location of these mutations within the DBD is illustrated schematically in Fig. 4. Sone et al. [110] examined the VDR from two unrelated patients (F5 and F20) previously shown to exhibit a ligand-binding positive and low affinity DNA-binding phenotype by Liberman et al. [86,88]. In both patients, a G to A missense mutation was identified in exon 3. The mutation changed arginine to glutamine at amino acid 80 in the second zinc-finger module (Arg80Gln). The recreated Arg80Gln mutant receptor had a high affinity for [3H]1,25(OH)2D3, bound weakly to DNA, and was unable to activate gene transcription from a reporter plasmid, demonstrating that this molecular defect is the cause of HVDRR in these cases [110]. Malloy et al. [25] also identified the same Arg80Gln mutation in two siblings with HVDRR (F49). The F49 family and the families (F5, F20) described by Sone et al. [110] both had origins in North Africa, however, no genetic relationship between these families could be established. Saijo et al. [111] described a DBD mutation in three HVDRR patients from two unrelated families (F26 and F33) of Japanese origin. Earlier investigations showed that fibroblasts from the patients had normal [3H]1,25(OH)2D3 binding but the VDR exhibited abnormal nuclear binding [98,112,113]. A G to A mutation was found in exon 3 that converted arginine to glutamine at amino acid 50 in the at the base of the first zinc finger module (Arg50Gln). The children’s parents were identified as carriers of both the normal and mutant alleles using single strand conformational polymorphism (SSCP) [111]. No data were provided to confirm the functional consequences of this defect on the VDR.

A missense mutation in the first zinc module of the VDR DBD that changed a histidine to glutamine at amino acid 35 (His35Gln) was identified by Yagi et al. [24]. The mutation replaced a positively charged amino acid with a neutral one. The VDR from the patient’s cells (F45) exhibited normal 1,25(OH)2D3 binding but the receptors exhibited low affinity binding to DNA. The patient’s fibroblasts were transiently transfected with a VDRE-CAT reporter plasmid to test for 1,25(OH)2D3 responsiveness. The transformed cells were unable to induce gene transcription. However, when the patient’s fibroblasts were cotransformed with the wild-type VDR cDNA and the reporter plasmid, the cells acquired the ability to respond to hormone. Two mutations in the VDR DBD were reported by Rut et al. [114]. One patient (F47) described previously by Lin and Uttley [115] had an A to G mutation in exon 2. The missense mutation resulted in lysine being replaced by glutamic acid at amino acid 45 (Lys45Glu). In the same study, Rut et al. [114] examined the VDR gene in a patient (F44) described previously by Simonin et al. [116]. They identified a unique T to C base change in exon 2 that resulted in phenylalanine being replaced by isoleucine at amino acid 47 (Phe47Ile). The recreated mutant VDRs exhibited normal ligand binding, but were transcriptionally inactive [114]. Lin et al. [117] examined the VDR gene for mutations in a patient (F23) with HVDRR previously described by Sakati et al. [118]. A unique G to A base change was found in exon 2. The mutation resulted in a glycine being changed to an aspartic acid at amino acid 46 (Gly46Asp). The recreated Gly46Asp mutant

1220 VDR exhibited the characteristics of a DBD mutation in that the mutant receptor bound [3H]1,25(OH)2D3 normally, but displayed a reduced affinity for DNA. The mutant receptor was also shown to be transcriptionally inactive in reporter gene assays. The authors used PCR and a restriction fragment length polymorphism (RFLP) generated by the mutation to demonstrate that the patient was homozygous for the mutation and the patient’s father was a carrier of the mutant allele. In contrast to the other DBD mutations described above, the mutation at Gly46 occurs in an amino acid that is not well conserved in the steroid-thyroid-retinoid receptor superfamily. However, Gly46 is conserved among receptors that form heterodimers with RXR proteins such as TR and RAR. 2. STRUCTURAL ANALYSIS OF DBD MUTATIONS

As of this writing, the crystal structure of the VDR DBD has not been reported. However, crystallographic studies of the GR [119], RXR, and TR [120] DBD structures have been elucidated and, based on these studies, one can extrapolate the alterations created by the mutations to the VDR DBD [121,122]. Crystallographic analyses of the GR demonstrate that amino acids 457–469 (corresponding to residues 38–50 in the VDR) form an α-helix which joins the two zinc finger modules. This α-helix packs perpendicularly with a second α-helix at the base of the second zinc finger. Together, the hydrophobic residues of these two α-helices comprise the hydrophobic core of the DBD. Lys45Glu and Gly46Asp mutations are located in the P-box (aa residues 42–46), a region of the receptor likely important in contacting the DNA bases and determining the specificity of the receptor for specific VDREs (Fig. 4). Rut et al. [114] proposed that the Lys45Glu mutation would disturb the hydrogen bonding between Lys45 and a guanine nucleoside in the VDRE half-site. The conversion of glycine to aspartic acid, a bulky, charged amino acid, probably leads to unfavorable electrostatic interactions with the negatively-charged phosphate backbone of the DNA helix that may prevent the receptor from contacting specific nucleotide bases in the VDRE. Alternatively, the Gly46Asp mutation may eliminate the ability of the VDR to specifically recognize VDREs [117]. Similarly, the Gly33Asp mutation is expected to repel the negatively-charged phosphate backbone due to the negatively-charged aspartic acid [114]. On the other hand, the substitution of glycine for histidine in the His35Gly mutation most likely eliminates a hydrogen bond donated from the positively-charged histidine to the phosphate of a guanine nucleoside in the VDRE [114]. In the Phe47Ile mutation, the loss of the phenylalanine ring structure may disrupt the integrity of the hydrophobic core of the

PETER J. MALLOY, J. WESLEY PIKE, AND DAVID FELDMAN

DBD. The mutation may obstruct the formation of the proposed α-helical structure at the base of the first zinc finger such that the VDR could not bind normally to its VDRE [114]. It is interesting to note that four of the DBD mutations, Lys45Glu, Gly46Asp, Phe47Ile, and Arg50Gln, occur in a LysXxxPhePhe[Lys/Arg]Arg sequence motif, which has been identified as a binding site for calreticulin [123–125]. Calreticulin binds to the VDR and cotransfection of calreticulin expression plasmids with a VDRE/RARE-luciferase reporter construct that causes a decrease in the reporter gene activation by VDR [124]. Since calreticulin may modulate VDR transactivation, disruption of the calreticulin binding site could lead to a decrease in VDR function. The effect of these mutations on calreticulin actions on the VDR has not been investigated.

C. Mutations Causing Premature Termination of the VDR 1. STOP MUTATIONS

The first molecular analysis of three HVDRR cases (F18, F34, F36) that exhibited no [3H]1,25(OH)2D3 binding was reported by Ritchie et al. [77]. A single unique C to A base change was found in exon 8 that changed the codon for tyrosine (TAC) to an ochre termination codon (TAA) (Tyr295stop) (Fig. 5). The location of the premature stop at amino acid 295 truncates 132 aa of the carboxy terminus of the VDR, which results in the deletion of a major portion of the LBD, thereby creating the ligand-binding negative phenotype. The recreated mutant VDR with a molecular size of 32,000 daltons was unable to bind [3H]1,25(OH)2D3 and failed to activate gene transcription, demonstrating that this mutation was the cause of the hormone resistant state in the three patients. The Tyr295stop mutation was the first stop mutation identified in the VDR. The location of this mutation and other mutations that cause premature termination of the VDR is shown in Fig. 5 and summarized in Table III. The three families studied by Ritchie et al. [77] and four additional families (F35, F37, F38, and F39) that comprise a large kindred where consanguineous marriages were common was analyzed by Malloy et al. [26] (Table III). A total of 8 children from this kindred exhibited HVDRR with alopecia. All of the affected children were homozygous for the Tyr295stop ochre mutation, and their parents were heterozygous as determined by analysis of a Rsa I restriction fragment length polymorphism (RFLP) created by the mutation [26]. Interestingly, the 32,000 molecular weight truncated protein that is predicted to be produced by this mutation

1221

CHAPTER 72 Hereditary 1,25-Dihydroxyvitamin D–Resistant Rickets

Stop mutations Arg30X

T A 30R D G C V G C I VP R 20

G

F

H

Tyr295X

F N A M

70 D N K T I R DC

T C 40 G E N F G P C C KG T 60 F F R RS M K R KA L F 50

Arg73X

GIn152X

GIn317X

R R H C Q A C 80 R L K R C V D I G M M KE F

AF-2

H3 H4H5 S1-3H6 H7 H8

H1

H9

H10 H11 H12

β-turn

E1

Splicing mutations and base deletion Exons 1F 1A 1B

ATG 1C 2

TGA 3

45 6

7 8

9

1E 1D GTCAGT → GTGAGT 366delC GTAAGT → GTAACT

FIGURE 5 Mutations that cause premature termination of the VDR. The location of the nonsense mutations causing HVDRR are illustrated in the VDR protein. The location of the splice site mutations and the single base deletion are depicted in the VDR gene.

could not be detected in extracts of cultured dermal fibroblasts using Western blots. In addition, in all but one case, the VDR mRNA was undetectable on Northern blot using RNA isolated from the cultured fibroblasts or EBV-transformed lymphoblasts from the patients. The absence of mRNA transcripts has been reported in other genetic diseases where a premature stop mutation has been found [126–128] and, in this case, apparently accounts for the absence of the mutant VDR protein. Since the reports by Ritchie et al. [77] and Malloy et al. [26], a number of other premature stop mutations have been identified in the VDR from patients with HVDRR [3]. The locations of these mutations are shown in Fig. 5. Family F11 described in earlier papers [20,21,29,63] had two affected children with HVDRR who exhibited the ligand-binding negative phenotype. This family of Christian Arabs live in the same town as the extended kindred (F18, F34–F39) described above who are Muslim Arabs. Although there is no known genetic relationship between family F11 and the large kindred, the Tyr295stop mutation is the cause of HVDRR in this family as well [129]. The Tyr295stop mutation was also identified by Wiese et al. [130] in two related patients (F29, F30) from Saudi Arabia who were previously studied by Bliziotes et al. [17]. These patients are apparently unrelated to the other families with the same mutation. It is not clear whether Tyr295stop mutation represents a mutational “hot spot” in the VDR gene or whether these cases all descended from a common founder mutation. A patient with HVDRR from a Moroccan family (F46) was examined for mutations in the VDR gene

by Wiese et al. [130]. At the cellular level, no [3H]1,25(OH)2D3 binding was detected in the patient’s cells. The authors discovered an opal mutation (CGA to TGA) in which a C to T substitution introduced a premature stop codon at amino acid 73 (Fig. 5). The Arg73stop mutation truncates the receptor in the middle of the second zinc finger module, resulting in the production of a 72 amino acid polypeptide. No evidence for the presence of a truncated VDR in the patient’s cells was demonstrated since both the LBD and monoclonal antibody binding sites were deleted in the mutant protein. Interestingly, the Arg73stop mutation, occurs in the same codon that gives rise to the Arg73Gln mutation, but at a different nucleotide base [76]. In the Arg73stop mutation, CGA is mutated to TGA, while in the Arg73Gln mutation the CGA is mutated to CAA. The Arg73stop mutation has also been identified in a young boy (F56) from Greece who had HVDRR with alopecia [131]. A premature stop codon caused by an amber mutation (CAG to TAG) in exon 4 was found in a Turkish patient (F32) with HVDRR by Kristjansson et al. [132]. Previous studies using fibroblasts from this patient had demonstrated an absence of ligand binding and 1,25(OH)2D3 responses [133]. This premature stop mutation occurs at Gln152 and deletes 306 amino acids of the VDR (Fig. 5). As expected, the Gln152stop mutant VDR was unresponsive to 1,25(OH)2D3 in gene activation assays. The Gln152stop mutation was also identified by Wiese et al. [130] in a HVDRR patient (F32) previously reported by Barsony et al. [133]. Studies of a young boy (F53) of French-Canadian origin with HVDRR and alopecia has been reported

1222 by Zhu et al. [134]. The patient’s fibroblasts lacked specific [3H]1,25(OH)2D3 binding and failed to exhibit 24-hydroxylase mRNA induction after treatment with up to 100 nM 1,25(OH)2D3. Northern blotting showed that the cells expressed a normal-sized VDR mRNA, but Western blotting failed to detect any protein. A C to T base substitution was located in exon 2, which changed the codon for arginine (CAG) at amino acid 30 to an opal stop codon (TAG) (Arg30stop) (Fig. 5). The 29 amino acid polypeptide represents the shortest truncated protein produced by a premature stop mutation in the VDR. The mutation eliminated 398 amino acids including the LBD, the monoclonal antibody epitope, the second zinc finger module, and a portion of the first zinc finger module. The same Arg30stop mutation was also identified in two children with HVDRR from a family (F54) living in Brazil [135]. One child died at 4 yr of age due to cadiorespiratory insufficiency. Interestingly, the parents, who were first cousins, were phenotypically normal, but had slightly elevated levels of serum 1,25(OH)2D. The mean value for the father was 73 pg/ml and for the mother 93 pg/ml (normal range 20–80 pg/ml). The mildly elevated 1,25(OH)2D values raise the possibility of mild vitamin D resistance in the heterozygotic parents, a finding that has not been documented previously in other parents of HVDRR children. An Iranian girl (F61) whose cells were unresponsive to high doses of 1,25(OH)2D3 and exhibited no [3H]1,25(OH)2D3 binding was shown to have a mutation in exon 8 that changed the codon for glutamine to a termination codon (Gln317stop) (Fig. 5) [136]. 2. SPLICE SITE MUTATIONS

Hawa et al. [137] examined the VDR in a young Greek girl (F50) with HVDRR and alopecia. Using RT-PCR and DNA sequencing, they showed that the patient’s RNA sequence diverged from the wild-type sequence at nucleotide 147. The sequence from exon 3 that encodes the second zinc finger module was deleted, and the sequence that followed was from exon 4. Sequence analysis of the VDR chromosomal gene found no mutations in the exons, however, a G to C base change was found in the 5′ end of intron E (Fig. 5). This single nucleotide change converts the wild-type sequence from GTGAGT to GTGACT and eliminates the 5′ donor splice site (consensus sequence: GT(A/G)(A/T)G(T/A/C)). The loss of the 5′ donor splice site caused exon 3 to be skipped in the processing of the VDR transcript and introduced a reading frameshift that resulted in a premature stop codon in exon 4. The mutant protein contains 92 amino acids of the wild-type sequence plus 6 amino acids due to the frameshift (Glu92fs) (Fig. 5). The shortened VDR had no [3H]1,25(OH)2D3 binding and failed to induce 24-hydroxylase activity.

PETER J. MALLOY, J. WESLEY PIKE, AND DAVID FELDMAN

A splice site mutation was also identified in a German patient (F56) with HVDRR and alopecia [131]. Studies of the patient’s fibroblasts showed absent [3H]1,25(OH)2D3 binding and failure to induce 24hydroxylase activity with 1,25(OH)2D3 treatment. In this case, a cryptic 5′ donor splice site was generated in exon 6. The mutation in this case, a C to G transition, changed the sequence from GTCAGT to GTGAGT. This single base change did not alter the amino acid coding sequence in exon 6, but introduced a splice site that could be recognized by the spliceosome complex during RNA processing. As a result, the mutation caused a 56 bp deletion in exon 6, which led to a frameshift 15 bases into exon 7. The mutant protein contains 233 amino acids of the wild-type sequence and an additional 4 amino acids due to the frameshift (Leu233fs) (Fig. 5). The mutation caused the truncation of 194 amino acids of the VDR leading to a loss of 1,25(OH)2D3 binding and hormone responsiveness. 3. GENE DELETION

There has been one report of patient (F42) with a major structural defect in the VDR gene that was the cause of HVDRR [138]. The defect, found by PCR and Southern blotting, was a deletion in the VDR gene that eliminated exons 7, 8, and 9. This is the only case thus far reported in which a partial deletion in the VDR gene has been shown to cause HVDRR. All of the nonsense mutations and the splice site mutations that lead to frameshifts, as well as the gene deletion, result in truncated VDRs. Although some of the mutants VDRs may have intact DBDs, the loss of the LBD and therefore its ability to bind 1,25(OH)2D3, associate with RXR and interact with coactivators makes the receptors nonfunctional and causes complete hormone resistance.

D. Mutations in the VDR Ligand-Binding Domain (LBD) 1. MUTATIONS THAT AFFECT 1,25(OH)2D3 BINDING

The first missense mutation found in the VDR LBD that resulted in an amino acid substitution was described by Rut et al. [139] and Kristjansson et al. [132]. The patient from Kuwait (F21) had HVDRR, but did not have alopecia. Preliminary studies by Fraher et al. [18] on the patient’s fibroblasts showed absent [3H]1,25(OH)2D3 binding. However, a later study by Rut et al. [139] showed that the fibroblasts contained normal amounts of [3H]1,25(OH)2D3 binding, but the affinity of the receptor for 1,25(OH)2D3 was significantly reduced (Kd = 1 × 10−9M) compared to normal controls

1223

CHAPTER 72 Hereditary 1,25-Dihydroxyvitamin D–Resistant Rickets

(Kd = 0.7 × 10−10M). The patient’s fibroblasts failed to induce 24-hydroxylase activity when treated with high doses of 1,25(OH)2D3 [18, 139]. Molecular analysis of the VDR gene identified a unique G to T missense mutation in exon 7 [132, 139]. This mutation resulted in replacement of a positively-charged arginine residue by a neutral-charged leucine at amino acid 274 (Arg274Leu) (Fig. 6). The Arg274Leu mutation is located in helix H5. In transactivation assays, the recreated Arg274Leu mutant VDR was relatively resistant to vitamin D, requiring approximately 1000-fold more 1,25(OH)2D3 to activate gene transcription than the wild-type receptor [132]. Although the resistance caused by the defective VDR could be overcome by treating with high concentrations of 1,25(OH)2D3 in vitro, the patient failed to respond to massive does of the hormone and eventually died of pneumonia. Whitfield et al. [140] analyzed the VDR from a girl (F4) who had the classic symptoms of HVDRR, but without alopecia. The patient’s fibroblasts were originally examined by Griffin and Zerwekh [85] who showed that the cells had normal 1,25(OH)2D3 binding but had defective induction of 24-hydroxylase activity. Sequencing of the VDR gene uncovered a T to G substitution in exon 8 that changed isoleucine to serine at amino acid 314 (Ile314Ser) (Fig. 6). The Ile314Ser mutation occurs in H7. The mutation causes a subtle defect in heterodimerization with RXR and decreased response to 1,25(OH)2D3 in transactivation assays. This patient showed a nearly complete cure when treated with pharmacological doses of 25-hydroxyvitamin D3. A missense mutation in the VDR LDB was described by Malloy et al. [141]. The patient suffered from HVDRR and two other rare genetic disorders, congenital generalized lipoatrophic diabetes (Berardinelli-Seip syndrome) and persistent müllerian duct syndrome [142]. The patient, a Turkish boy (F51), had rickets

and high 1,25(OH)2D3 levels, but did not have alopecia. He was treated with extremely high doses of calcitriol (Rocaltrol 12.5 µg/day), which eventually normalized his serum calcium and ultimately improved his rickets. However, the child died of apparently unrelated problems. The patient’s fibroblasts expressed normal VDR levels, but the affinity for 1,25(OH)2D3 was decreased by about twofold. The patient’s fibroblasts required approximately fivefold more 1,25(OH)2D3 to induce 24-hydroxylase mRNA compared to control cells. Sequence analysis of the VDR gene uncovered a C to G missense mutation in exon 8. This mutation leads to replacement of histidine by glutamine at amino acid 305 (His305Gln) (Fig. 6). The His305Gln mutation occurs in the interhelical loop between H6-H7. Interestingly, [3H]1,25(OH)2D3-binding studies of the reconstructed mutant protein demonstrated an eight-fold lower affinity for 1,25(OH)2D3 compared to the wild-type VDR when the assays were performed at 24°C (vs two-fold at 0°C). In gene transactivation assays, the His305Gln mutant VDR required five-fold more 1,25(OH)2D3 to achieve the same level of activity as the wild-type VDR. RFLP analysis with AlwN I showed that the patient and a sibling with HVDRR were homozygous for the mutation and that their parents were heterozygous. The boy’s sister, who also had HVDRR and the same mutation in the VDR, did not exhibit the other genetic defects. No explanation was forthcoming for the presence of three genetic defects in a single individual. It is unclear how the His305Gln mutation in the VDR is related, if at all, to the two other genetic abnormalities that were present in this child. The congenital total lipodystrophy has been shown to be caused by a mutation in the seipin gene [143] on chromosome 11, not chromosome 12 where the VDR is located. A genetic cause for the persistent müllerian duct syndrome in this patient has not been found.

Phe251Cys GIn259Pro Ile268Thr Arg274Leu Trp286Arg His305GIn Ile314Ser

Cys190Trp

Arg391Cys

Glu329Lys

H1

H3

H4 H5 S1−3H6 H7 H8 E1

FIGURE 6

β-turns

H9

Glu420Lys

H10 H11 H12

AF-2

Schematic illustration of the ligand-binding domain of the VDR and location of amino acid substitutions causing HVDRR. The α-helices (H1-H12) of the VDR LBD are depicted as shaded rectangles and the β-turns are drawn as a hatched rectangle. The loops connecting the helices are drawn as solid lines. The E1 and AF-2 regions are shown below the α-helices.

1224 An Algerian boy and his younger sister (F59) both with HVDRR without alopecia were shown to have a T to C mutation in exon 7 that changed tryptophan to arginine at amino acid 286 (Trp286Arg) (Fig. 6) [144]. Trp286 is the only tryptophan in the VDR. The patient’s fibroblasts expressed a normal size VDR protein and normal length VDR mRNA but no specific [3H]1,25(OH)2D3 binding was observed, and the cells were totally unresponsive to 1,25(OH)2D3 treatment. The Trp286Arg mutation is located in the β1 sheet of the three-stranded β sheet between helices H5-H6. A young Saudi Arabian girl with HVDRR, but without alopecia, was homozygous for a unique T to C mutation in exon 7 of the VDR gene that changed the codon for isoleucine to threonine at amino acid 268 (Ile268Thr) [145]. Based on the crystallographic studies of the VDR LBD, Ile268 located in helix H5 directly interacts with 1,25(OH)2D3. The Ile268Thr mutant VDR exhibited a ~tenfold lower affinity for [3H]1,25(OH)2D3 compared to the WT VDR consistent with its interaction with the ligand. However, in transactivation assays, the Ile268Thr mutant required ~100fold higher concentrations of 1,25(OH)2D3 to stimulate gene transcription compared to the WT VDR. The Ile268Thr mutant also exhibited a marked decrease in RXR binding compared to the WT VDR. Ile268 is also involved in the hydrophobic stabilization of helix H12. Consistent with this activity, the Ile268Thr mutant required ~100-fold more 1,25(OH)2D3 to promote binding to the coactivators SRC-1 and DRIP205. These cumulative defects cause 1,25(OH)2D3 resistance and results in the syndrome of HVDRR in the patient [145]. A missense mutation was also described in the VDR LBD by Thompson et al. [138]. The HVDRR patient (F43) was shown to have a mutation in exon 5. In this case, a cysteine was changed to tryptophan at amino acid 190 (Cys190Trp) (Fig. 6). The Cys190Trp mutation occurs in the loop between H1 and H3. Further details about this case were not included in this preliminary report, however, deletion of this region of the VDR LBD has been shown to have no effect on ligand binding or transactivation. Also, we have recreated Cys190Trp mutation and have shown that it does not affect transactivation (Malloy unpublished). These data suggest that the Cys190Trp mutation was unlikely to be the sole cause of HVDRR in this case. a. Structural Analysis of LBD Mutations That Affect 1,25(OH)2D3 Binding Since the crystal structure of the holo-VDR has been reported, a more definitive explanation of the effects of the LBD mutations on the VDR is possible. In one patient (F21), Arg274 was mutated to Leu (Arg274Leu) and resulted in a 1,000-fold decrease in ligand binding affinity [132]. This can now be

PETER J. MALLOY, J. WESLEY PIKE, AND DAVID FELDMAN

explained by the fact that Arg274 is a contact point for the 1-hydroxyl group of 1,25(OH)2D3 [50]. Thus, the mutation to leucine alters the contact point and lowers the binding affinity of the VDR for 1,25(OH)2D3. In a different patient (F51), His305 was mutated to Gln (His305Gln) that lowered the binding affinity for 1,25(OH)2D3 by about five–tenfold [141]. This can now be explained by the fact that His305 is a contact point for the 25-hydroxyl group of 1,25(OH)2D3 [50]. Again, the mutation to glutamine alters the contact point and lowers the binding affinity of the VDR for 1,25(OH)2D3. In a third patient (F59), Trp286 was mutated to Arg (Trp286Arg) that totally abolished ligand binding [144]. Trp286 makes contact with the α face of the C ring in 1,25(OH)2D3, and is involved in forming the hydrophobic channel where the conjugated triene connecting the A and the C rings fits [50]. The Trp286Arg mutation alters the contact point with the ligand and causes resistance to 1,25(OH)2D3 [144]. It is evident from these data that the disruption of a ligand contact point can be the basis for HVDRR (see Chapter 15). 2. MUTATIONS THAT AFFECT VDR-RXR HETERODIMERIZATION

As mentioned above, the VDR requires heterodimerization with RXR for activity. Disruption of this protein: protein interaction can thereby cause 1,25(OH)2D3 resistance. The first patient found to have a mutation in the VDR that disrupted VDR-RXR heterodimerization was described by Whitfield et al. [140]. The patient was a young girl (F52) who had HVDRR and alopecia. Sequencing showed that the patient had a C to T base change in exon 9 that converted an arginine to cysteine at amino acid 391 (Arg391Cys) (Fig. 6). The Arg391Cys mutation is located in helix H10. Ligand binding was unaffected but transactivation of a reporter gene by the Arg391Cys mutant VDR required higher than normal concentrations of 1,25(OH)2D3 for activity. However, when RXR was co-transfected in the assays, the transactivation activity could be restored to normal levels. VDR-RXR interactions and VDRE binding were further examined using electrophoretic mobility shift assays (EMSA). The Arg391Cys mutant exhibited a lower capacity for forming a VDR-RXR-VDRE complex than the wild-type VDR. The Arg391Cys mutation thus reduces the interaction between the VDR and RXR. By increasing the RXR protein concentration, the affinity defect could be overcome, and the transactivation rescued. This study demonstrated the importance of both 1,25(OH)2D3 binding and RXR heterodimerization in VDR mediated gene transactivation. This was the first report of a mutation in the VDR that interfered with RXR binding and caused HVDRR.

CHAPTER 72 Hereditary 1,25-Dihydroxyvitamin D–Resistant Rickets

Two siblings, a brother and sister from India (F57) that had HVDRR with alopecia, were studied by Cockerill et al. [131]. The children’s parents were first cousins. Using cultured fibroblasts, [3H]1,25(OH)2D3 binding was found to be normal but 1,25(OH)2D3 induction of 24-hydroxylase activity was absent. DNA sequence analysis revealed a single base alteration in exon 7 that changed glutamine to proline at amino acid 259 (Gln259Pro) (Fig. 6). The Gln259Pro mutation occurs in helix H4. Although Gln259Pro had no apparent affect on ligand binding there was evidence of impaired VDR-RXR-VDRE formation. Whereas the wild-type VDR formed two complexes (complex A and complex B) in the EMSA, the Gln259Pro mutant VDR showed a reduction in the formation of complex B and an enhancement in complex A. The authors speculated that the Gln259Pro mutation in the VDR affected protein: protein interactions, possibly by increasing the affinity of the receptor for an unidentified protein. In transactivation assays, the recreated Gln259Pro mutant VDR was functionally inactive. Malloy et al. [146] examined the VDR in a young Hmong boy (F58) with HVDRR and alopecia. Analyses of the VDR demonstrated a normal size receptor on Western blots, however, the amount of VDR was decreased compared to normal. Northern blot analysis of 24-hydroxylase mRNA induction showed that the patient’s fibroblasts were approximately 1,000-fold less responsive to 1,25(OH)2D3 than control fibroblasts confirming target organ resistance. A unique missense mutation was found in exon 6 that changed a phenylalanine to cysteine at amino acid 251 (Phe251Cys) (Fig. 6). In [3H]1,25(OH)2D3 binding experiments, the recreated Phe251Cys mutant VDR exhibited a lower affinity for the ligand than wild-type VDR when assayed at 24°C. Using GST pull-down assays and yeast two-hybrid constructs, the Phe251Cys mutant VDR was shown to have reduced capacity to bind RXR. In transactivation assays, cotransfection of RXR partially restored the activity of the mutant receptor. The Phe251Cys mutation occurs in the E1 region (aa 244–263) of the VDR LBD. The E1 region overlaps the C-terminal portion of helix H3, loop 3–4 and the N-terminal portion of helix H4. This structural motif is highly conserved throughout the nuclear receptor superfamily. A cluster of hydrophobic amino acids within the E1 region is critical to the threedimensional folding and formation of the ligand-binding pocket [147]. At the center of this region is an invariant aromatic phenylalanine residue which corresponds to Phe251 in the VDR. Since Phe251 is in such a critical site in the LBD, replacing the aromatic amino acid phenylalanine with a small hydrophilic cysteine

1225

residue likely disrupts the ligand-binding pocket of the VDR and interferes with the fundamental conformation required for optimal function. From the crystallographic studies, Phe251 does not form a direct contact point with the bound ligand. Indeed, at 4°C normal ligand binding was observed. However, at elevated temperatures, the binding affinity was severely decreased, suggesting that the Phe251Cys mutation disrupted the folding of the ligand-binding pocket [146]. a. Structural Analysis of VDR-RXR Heterodimerization Mutations Although the VDR-RXR dimer crystallographic studies have not been performed as of this writing, studies of RXRα, RAR-RXRα, and PPARγRXRα, show that the dimer interface is formed from helix H9 and helix H10 and the interhelical loops between H7–H8 and H8–H9 [50]. The Arg391Cys mutation occurs in helix H10 [140]. The mutation lowers the binding affinity for RXR causing 1,25(OH)2D3 resistance. However, in transactivation assays, the 1,25(OH)2D3 resistance could be overcome by overexpression of RXR [140]. The Phe251Cys mutation reduces the affinity of the VDR for 1,25(OH)2D3 and alters RXR binding [146]. The Phe251Cys mutation is located in the E1 region in the interhelical loop between H3–H4. This region of the VDR is not part of the dimer interface; however, the loop does appear to be positioned beneath the interhelical loop between H8–H9 and helix H9. The mutation apparently disrupts the formation of the dimer interface of the receptor and results in a decreased ability to heterodimerize with RXRα. The Gln259Pro mutation is also located in the E1 region in helix H4. The Gln259Pro mutant did not affect 1,25(OH)2D3 binding but exhibited an abnormality in forming a complex on VDREs in EMSA [131]. A study by Whitfield et al. [148] has shown that a Gln259Gly mutant VDR binds ligand normally, but is defective in its ability to form a complex with RXR. They also showed that addition of increasing amounts of RXR could restore the transactivation capacity of the Gln259Gly mutant to near wild-type levels. These single amino acid changes demonstrate the interdependency of ligand binding and heterodimer binding for transactivation. 3. MUTATIONS THAT AFFECT COACTIVATOR BINDING

As noted above, the VDR also must recruit coactivators for transcriptional activity. It is now clear from crystallographic studies of the VDR and other members of the steroid receptor superfamily that repositioning of helix H12 is an essential event that occurs as a consequence of ligand binding and is necessary for

1226 transactivation [50]. The repositioning of helix 12 leads to the formation of the hydrophobic cleft critical for coactivator binding. Therefore, disruption of this surface interface may cause hormone resistance and HVDRR. A study by Malloy et al. [149] examined the VDR in a young boy (F60) with HVDRR. The patient did not have alopecia. The patient’s fibroblasts exhibited normal ligand binding, but the cells were totally resistant to 1,25(OH)2D3. A novel missense mutation was found in exon 9 that changed a glutamic acid to lysine at amino acid 420 (Glu420Lys) (Fig. 6). The Glu420Lys mutation is located in helix H12. The recreated Glu420Lys mutant VDR showed no defects in VDRRXR heterodimerization or binding to VDREs. However, the mutation prevented the coactivator SRC-1 and DRIP205 from binding to the VDR. In transactivation assays, cotransfection of SRC-1 failed to restore transactivation by the mutant VDR. This case represents the first description of a naturally occurring mutation in the VDR that disrupts coactivator interaction and causes HVDRR [149]. The polar interactions that stabilize the repositioning of helix H12 involve a conserved salt-bridge between Lys264 in helix H4 and Glu420 in helix H12 and a hydrogen bond between Ser235 in helix H3 and Thr415 in helix H12. The Glu420Lys mutation prevents the correct repositioning of helix H12 after binding the ligand. The substitution of the negatively-charged glutamic acid (Glu420) with a positively-charged lysine residue (Lys420) would prevent the polar interaction with the positively-charged lysine (Lys264) salt bridge partner. A charge clamp formed by Lys246, Lys264, and Glu420 enables the VDR to recruit and bind coactivators through their LxxLL motifs. The Lys246 in helix 3 and Glu420 in helix H12 are thought to be indispensable for binding the LxxLL peptide on the coactivator. The Glu420Lys mutation disrupts coactivator binding and causes the hormone resistance seen in the patient. 4. OTHER MUTATIONS IN THE VDR LBD

There has been one report of a patient (F62) with HVDRR and alopecia who had two mutations in the VDR gene [150]. One mutation in exon 8 changed a glutamic acid to lysine at amino acid 329 (Glu329Lys) (Fig. 6) and the second mutation was a single base deletion of a cytosine at nucleotide 366 (366delC) (Fig. 5). The deletion of the cytosine results in a shift in the reading frame in exon 4 and leads to a premature termination signal in the VDR message. The premature termination signal truncates most of the LBD. The affect of the Glu329Lys mutation on VDR function was not reported.

PETER J. MALLOY, J. WESLEY PIKE, AND DAVID FELDMAN

E. HVDRR without Mutations in the VDR Since the initial description of HVDRR as a genetic disorder, mutations in the VDR gene were suspected as the likely cause of 1,25(OH)2D3 resistance. Although the VDR is the principle determinant in the 1,25(OH)2D3 action pathway, it is possible that target organ resistance to 1,25(OH)2D3 may result from mutations in other proteins that are essential to the transactivation process. The following cases highlight this possibility. Hewison et al. [151] has described a case of HVDRR in which a mutation could not be found in the VDR. The patient (F48), a young girl of English descent, exhibited all of the hallmarks of HVDRR including alopecia. The patient’s fibroblasts expressed a normal-sized VDR mRNA and exhibited normal [3H]1,25(OH)2D3 binding. However, no induction of 24-hydroxylase activity was observed when the fibroblasts were treated with up to 1 µM 1,25(OH)2D3. Although the fibroblasts were clearly resistant to 1,25(OH)2D3, no mutations were found within the coding region of the VDR gene. The patient’s VDR mRNA was reverse transcribed and amplified by PCR and then expressed in CV-1 cells. In transactivation assays, the patient’s VDR exhibited a normal transactivation response to 1,25(OH)2D3. These results clearly demonstrated that the patient’s VDR was normal. The authors suggested that the tissue resistance was not due to a defect in the VDR and that the hormone resistance causing HVDRR was most likely the result of a mutation in an essential protein that participates in the 1,25(OH)2D3 action pathway. In a follow-up study of this interesting case, Chen et al. [152] proposed that the cause of 1,25(OH)2D3 resistance was due to the abnormal expression of a VDRE-interacting hormone response element-binding protein. This protein is a member of the heterogeneous nuclear ribonucleoprotein (hnRNP) family that binds to double-stranded DNA and modulates RNA processing. In New World primates the hormone response element-binding proteins have been shown to act as dominant negatives and are believed to cause target organ resistance to adrenal, gonadal, and vitamin D sterol/steroid hormones [153–158]. The fact that the patient had HVDRR without a mutation in the VDR highlights the importance of the complex machinery involved in VDR transactivation and that mutations in the hormone signaling pathway may result in 1,25(OH)2D3 resistance. In Cauca, Columbia, more than 200 patients have been diagnosed with a disease that somewhat resembles HVDRR without alopecia [159]. The patients exhibit lower limb deformities due to rickets but are

CHAPTER 72 Hereditary 1,25-Dihydroxyvitamin D–Resistant Rickets

otherwise in good physical condition. Rickets limited to the lower extremities, as in these cases, have not been reported in other HVDRR families. The affected individuals have serum calcium levels that are within the normal range, but their serum 1,25(OH)2D levels are unusually high suggesting target-organ resistance. Fibroblasts from two of the most severely affected patients showed a low induction of 24-hydroxylase activity by 1,25(OH)2D3. However, no mutations were found in the VDR gene. Since the cause of vitamin D resistance in this instance was not due to mutations in the VDR and a functional response to 1,25(OH)2D3 was demonstrated, it has not been clearly ascertained whether this entity is a variant of HVDRR. The high prevalence of the disease in the population and the localized distribution of the rickets raises the possibility of an environmental cause. These cases support the concept that target organ resistance to 1,25(OH)2D3 may be due to mechanisms other than mutations in the VDR.

VI. THERAPY OF HVDRR A. Vitamin D As detailed in the preceding sections, HVDRR is almost always caused by heterogeneous mutations in the VDR that result in partial or total resistance to 1,25(OH)2D3. Partially or totally inactive VDRs reduce calcium absorption from the intestine to the circulation and result in hypocalcemia. Low serum calcium levels lead to a decrease in bone mineralization and cause rickets. In order to cure rickets, calcium levels must be normalized. A number of therapies using calcium and active vitamin D metabolites aimed at increasing the serum calcium levels have been tried and the responses have varied widely. For the most part, it was thought that patients with HVDRR without alopecia were better responders to treatment with vitamin D preparations than those patients with alopecia [30]. In several of the earlier cases that were reported, patients without alopecia showed improvement both clinically and radiologically to the administration of pharmacological doses of vitamin D ranging from 5000 to 40,000 IU/day [9,10,79]; 20 to 200 µg/day of 25(OH)D3; and 17–20 µg/day of 1,25(OH)2D3 [10]. Of the patients with HVDRR without alopecia, a few have had their VDR analyzed at the molecular level. Patient (F51) with the His305Gln mutation in the VDR LBD a contact point for the 25-hydroxyl group of 1,25(OH)2D3 responded to 12.5 µg/day calcitriol [141, 142]. The treatment overcame the low affinity-binding defect and achieved adequate VDR occupancy to mediate normal 1,25(OH)2D3

1227

responses. Patient (F4) with the Ile314Ser mutation in the VDR LBD was treated with 1 mg/day of vitamin D2 from age 2 to age 18 [79]. At age 20 following an uneventful pregnancy, the patient developed hypocalcemia and was treated successfully with 50 µg/day of 25(OH)D3. On the other hand, patient F21 with the Arg274Leu mutation in the VDR LBD, a contact point for the 1-hydroxyl group of 1,25(OH)2D3, was unresponsive to treatment with 600,000 IU vitamin D; up to 24 µg/day of 1,25(OH)2D3 (calcitriol); and 12 µg/day 1α (OH)D3. The patient later died of pneumonia [18]. Fibroblasts from this patient exhibited no specific [3H]1,25(OH)2D3 binding and were unresponsive to hormone treatment. However, the recreated Arg274Leu mutant VDR did exhibit transactivation activity at high doses of 1,25(OH)2D3 [132]. In general, HVDRR patients with alopecia are more resistant to treatment with vitamin D metabolites. However, a small number of these patients have been treated successfully using vitamin D. Two patients showed signs of improvement when given vitamin D or 1α(OH)D3 [14,160] and one patient responded to 25(OH)D as well as 1α(OH)D3 [15]. 1α(OH)D3 and 1,25(OH)2D3 also were effective treatments in other cases [22,23,96,112,161], including patients with the Arg50Gln and Arg73Gln mutations [76, 111]. Two siblings (F32), with the Glu152stop mutation, showed no increase in serum calcium during high dose vitamin D treatment despite raising their circulating 1,25(OH)2D levels to more than 100 times the mean normal range. However, notwithstanding their low serum calcium concentrations, healing of rickets and suppression of PTH was evident [162]. In one case, where vitamin D and 1,25(OH)2D3 therapies failed, the patient responded to oral phosphorous treatment [11]. The molecular cause of HVDRR in this case has not been elucidated. In many cases when patients fail to respond to 1,25(OH)2D3, intensive calcium therapy is used as described below.

B. Calcium The most significant development in the treatment of HVDRR was first reported by Balsan et al. [163]. In their insightful study, they used long term intravenous (IV) calcium infusions to successfully treat a child with HVDRR who had failed prior treatments with large doses of vitamin D derivatives and/or oral calcium [15]. This novel therapy bypassed the calcium absorption defect in the intestine caused by the mutant VDR. The patient was infused with high IV doses of calcium during the nocturnal hours over a 9-month period. Clinical improvement including relief of bone

1228 pain was observed within the first 2 weeks of therapy. Within 7 months, the child gained both weight and height. Eventually, the serum calcium normalized, the secondary hyperparathyroidism was reversed, and the rickets ultimately resolved as assessed by X-ray and bone biopsy. However, when the IV infusions were discontinued, the rickets returned. Several other groups have reported using IV calcium infusions as a therapy for HVDRR [17,164,165]. Weisman et al. [164] treated two patients with IV calcium and showed a decrease in their serum alkaline phosphatase activity and an increase in their serum calcium and phosphate over a one-year period [164]. X-ray analysis showed resolution of the rickets with the appearance of normal mineralization of bone. In some cases, after radiological healing of the rickets has been achieved with IV calcium infusions, high dose oral calcium therapy has been shown to be effective in maintaining normal serum calcium concentrations [165]. For those HVDRR children who do not respond to high dose calcitriol, this two-step protocol is initiated at an early age. Administration of oral calcium salts to restore serum calcium also has been used as a therapy for HVDRR patients. In a study by Sakati et al. [118], a patient (F23) who failed to respond to calciferols received 3–4 grams of elemental calcium orally per day and showed clinical improvement during 4 months of therapy. The patient cells were later shown to be totally resistant to 1,25(OH)2D3, due to a Gly46Asp mutation in the VDR DBD [117]. Although many vitamin D actions have been described on bone cells (see Chapter 32 and 37), the administration of calcium alone is sufficient to reverse the rickets. The correction of the hyperparathyroidism by IV calcium and return of phosphate to normal suggests that the hypophosphatemia was due to the secondary hyperparathyroidism. In the VDR knockout mice, a high calcium rescue diet also corrects the bone deformities (See Chapter 20). Only the alopecia is not reversed. These findings suggest that the critical defect in defective receptors is intestinal absorption of calcium.

C. Prenatal Diagnosis Pregnant women from families with a history of HVDRR can be screened for mutations in the VDR. Cultured cells from chorionic villus samples or amniotic fluid have been used to determine whether the fetus has HVDRR using [3H]1,25(OH)2D3 binding assays and induction of 24-hydroxylase activity. A prenatal diagnosis of HVDRR can also determined by examining DNA from chorionic villus samples for

PETER J. MALLOY, J. WESLEY PIKE, AND DAVID FELDMAN

RFLPs generated by known mutations in the VDR gene [166,167].

D. Spontaneous Healing An interesting dilemma regarding HVDRR is that several patients have had a spontaneous improvement in their disease [21,22,63]. Spontaneous healing of rickets usually happens between 7–15 years of age and is not necessarily associated with the time of puberty. Sometimes, the spontaneous recovery occurs after the patient has undergone a relatively ineffective longterm treatment with vitamin D metabolites and mineral replacement. The healing process arises spontaneously and does not appear to be related to the treatment. In some patients, spontaneous improvement occurred after treatment was discontinued [22]. The patients appear to remain eucalcemic without therapy and show no evidence of osteomalacia or rickets. Interestingly, cultured fibroblasts obtained from a skin biopsy of a patient taken after spontaneously healing of the rickets had occurred were still resistant to 1,25(OH)2D3 [22]. Spontaneous improvement has been observed in patients [21,63] with the Tyr295stop mutation [25,77] and in patients [22] with the Arg73Gln mutation [76]. Despite the patient’s improvement in their hypocalcemia and rickets, alopecia remained [21,22,63]. It is not uncommon for children to “outgrow” genetic diseases and, perhaps after skeletal growth has been completed, the body can more easily compensate for the defective VDR gene by other mechanisms. Many patients can be maintained on oral calcium; however, occasionally some patients fall back to hypocalcemia due to decreased compliance or lack of tolerance to the high doses of calcium. IV calcium therapy is then reinstituted. After IV therapy normalizes the rickets and metabolic abnormalities, some patients can be switched to oral calcium. The hypocalcemia and secondary hyperparathyroidism can often be controlled with high dose oral calcium. A number of observations about the patients have been obtained in personal communications with Dr. Zeev Hochberg, the physician caring for several patients with the Tyr295stop mutation. As the patients get older they appear normal on physical exam and X-ray. Bone mineral density is reduced but slowly improving over time. When calcium treatment is commenced at an early age, their eventual height is normal. Although the females have normal menstruation, their reproductive function is unknown at this time. Immunologically, they appear to experience no more infections than normal. No cases of cancer have been observed.

CHAPTER 72 Hereditary 1,25-Dihydroxyvitamin D–Resistant Rickets

E. Future Therapy Using Vitamin D Analogs Vitamin D analogs have been proposed as a potential therapy for patients with HVDRR, especially those with mutations in the VDR LBD [168–171]. The use of vitamin D analogs is based on the rationale that they bind to the VDR at different amino acids than 1,25(OH)2D3. Using cultured fibroblasts from patients and in vitro transactivation assays, the vitamin D analogs 20-epi1,25(OH)2D3 and 1β-hydroxymethyl-3-epi-16-ene26α,27α-bishomo-25(OH)D3 were shown to partially or completely restore the responsiveness to the Arg274Leu and His305Gln mutant VDRs, but were less effective in activating the Phe251Cys mutant [168]. These results suggest that amino acids that are involved in ligand binding rather than amino acids that are involved in heterodimerization or coactivator binding are more likely to respond to analogs. Future therapy using vitamin D analogs could be based on a rationale drug design for individual patients with specific types of VDR mutations.

VII. ALOPECIA The role of vitamin D in the regulation of hair growth is still unclear. However, analysis of patients with HVDRR and VDR knockout mice has revealed a number of interesting facts concerning alopecia (see Chapters 20 and 35). First, alopecia is not found in patients with 1α-hydroxylase deficiency (VDDR I) and other forms of vitamin D deficiency. In the 1α-hydroxylase knockout mouse model, abnormalities develop in skeletal, reproductive and immune function [172]. However, the 1α-hydroxylase knockout mice do not develop alopecia. These findings suggest that 1,25(OH)2D3 itself is not required for hair development. On the other hand, VDR knockout mice develop alopecia, indicating that the VDR is essential for hair growth [44,45]. Furthermore, targeting of the WT VDR to keratinocytes of the VDR knockout mouse prevents alopecia [173]. These findings raise the question of how vitamin D and the VDR are involved in regulating hair growth. Patients with DBD mutations and premature stop mutations all have alopecia and are totally hormone resistant. Alopecia remains unchanged even after patients undergo successful therapy or show spontaneous improvement in rickets. The patients that did not develop alopecia all had missense mutations in the VDR LBD. Some of these mutations affect ligand binding (Arg274Leu, His305Gln, Ile314Ser, and Trp286Arg). Three of these amino acids Arg274, His305, Trp286 are contact points for 1,25(OH)2D3. Since the patients with the His305Gln and Ile314Ser mutations were

1229

somewhat responsive to vitamin D therapy, it is reasonable to speculate that the limited VDR function may have prevented the development of alopecia after birth. On the other hand, the patients with the Arg274Leu and Trp286Arg mutations that severely diminished or abolished [3H]1,25(OH)2D3 binding were totally resistant to 1,25(OH)2D3, yet did not have alopecia. In contrast, in patients with LBD mutations that reduced heterodimerization with RXRα (Phe251Cys, Gln259Pro, and Arg391Cys), but had little or no effect on ligand binding, alopecia was present. Although these mutations caused 1,25(OH)2D3 resistance in the patient, the function of the mutant VDRs could be restored in vitro by supraphysiological doses of the hormone and addition of excess RXR. Perhaps most interestingly was the patient with the Glu420Lys mutation that prevents coactivator binding but not ligand binding or RXR heterodimerization. The patient did not have alopecia, yet was clearly resistant to high doses of hormone. Based on these findings, one can conclude that mutations that affect DNA binding, VDR-RXR heterodimerization, or that truncate the LBD are linked to alopecia while mutations that affect ligand binding or coactivator binding are not. This suggests that VDR-RXR heterodimerization and DNA binding are critical for VDR function in hair development. These findings also suggest that ligand binding and coactivator binding are not essential functions of the VDR for hair growth and/or to prevent alopecia. A role for RXR is clearly demonstrated since targeted inactivation of RXRα in keratinocytes also causes alopecia [174]. The alopecia associated with HVDRR is clinically and pathologically indistinguishable from the generalized atrichia with papules found in patients with mutations in the hairless (hr) gene [150,175]. Like the VDR, HR is a zinc finger protein suggesting that it interacts with DNA. The hr gene is expressed in many tissues, especially in the skin and brain [176]. HR has recently been shown to interact with the VDR, and coexpression of HR and VDR blocks VDR mediated transactivation [177]. The data on the VDR mutations, combined with the findings in the 1α-hydroxylase knockout mouse model, suggests that the role of the VDR in the hair cycle is to repress the expression of some gene(s) in a ligandindependent manner. This activity requires RXR heterodimerization and DNA binding, but not interaction with coactivators. HR may also be required as a corepressor for this negative regulatory activity of the VDR. The VDR is a negative regulator of a number of genes, and the loss of a negative regulatory activity by the unliganded VDR could potentially lead to the derepression of those genes that could ultimately lead to alopecia. Since the VDR is a negative regulator of the PTHrP gene [178] and overexpression of PTHrP

1230 has been shown to cause alopecia and is involved in hair cycle regulation [179,180], dysregulation of the PTHrP gene may be a likely candidate in the development of alopecia in HVDRR.

VIII. CONCLUDING REMARKS HVDRR is a rare recessive genetic disorder caused by heterogeneous mutations in the VDR that result in end-organ resistance to 1,25(OH)2D3 action. The major manifestation of the defective VDR on the vitamin D endocrine system is to decrease intestinal calcium and phosphate absorption that results in decreased bone mineralization and rickets. Secondary hyperparathyroidism results as a consequence of the low serum calcium. The classical role of 1,25(OH)2D is to regulate mineral homeostasis, achieved through its coordinated actions on intestine, kidney, bone, and parathyroid gland [181,182]. The VDR is also expressed in a wide variety of tissues, including kidney, skin, liver, pancreas, muscle, breast, prostate, adrenal, thyroid, and cells of mesenchymal or hematopoietic origin [31,32,35,36,80,183,184]. From the ubiquitous nature of the VDR, it appears that the role of the VDR in cellular function is not homeostatic but rather pleiotropic. The expanded scope of vitamin D actions include stimulation of differentiation, inhibition of cell proliferation, and suppression of the immune response [35,36,39,184]. In addition, the regulation of cellular proliferation and differentiation by 1,25(OH)2D is a common feature in many tissues examined, and it is likely that this regulatory feature is a fundamental component of all biological responses to 1,25(OH)2D. Notwithstanding the complexity and diversity of biological responses elicited by 1,25(OH)2D, the profound skeletal abnormalities demonstrated in patients with HVDRR emphasizes the fundamental and essential role of 1,25(OH)2D in calcium homeostasis. It is interesting to note that despite the many pleiotropic processes regulated by 1,25(OH)2D3, children with HVDRR exhibit only symptoms that relate to their calcium deficiency and/or alopecia. After treatment, they appear normal in all respects except for the alopecia. Analysis of HVDRR patients provides many interesting insights into vitamin D physiology and the role of the VDR in mediating 1,25(OH)2D3 action. Since 1978, more than 60 families with HVDRR have been reported. A number of cases have been examined for mutations in the VDR. However, since some of these earlier cases of HVDRR presented late in life, it is possible that they may have been due to other conditions unrelated to genetic defects in the VDR or to some form of acquired resistance to 1,25(OH)2D3. Presently, eight missense mutations have been identified

PETER J. MALLOY, J. WESLEY PIKE, AND DAVID FELDMAN

in the VDR DBD. All of the DBD mutations prevent the VDR from binding to DNA, causing total 1,25(OH)2D3 resistance even though 1,25(OH)2D3 binding is normal. In the LBD, 10 missense mutations have been identified. These mutations disrupt ligand binding, VDR-RXR heterodimerization, or modify coactivatorbinding sites. They result in partial or total hormone resistance. In addition, premature termination codons and splice site mutations truncate the VDR and cause total 1,25(OH)2D3 resistance. The successful use of IV calcium infusion or oral calcium therapy for HVDRR raises interesting questions about the role of vitamin D in bone homeostasis. Correction of hypocalcemia and secondary hyperparathyroidism leads to healing of the rickets as assessed by X-ray and bone biopsy. However, careful analysis of bone histomorphometry has not been accomplished in HVDRR cases. Recent investigations comparing the 1αhydroxylase knockout mice (1α(OH)ase−/−) with the VDR knockout mice (VDR−/−) has revealed interesting differences in phenotypes [185]. Defects of a subtle nature were found in bone histomorphometry suggesting incomplete normalization of osteoblastic function. Growth plates were more widened and disorganized in the 1α(OH)ase−/− mice, while the VDR−/− mice were slightly thickened and more organized. However, when the animals were fed a rescue diet of 2% calcium and 20% lactose, the cartilaginous growth plate was normalized in the VDR−/− mice, but remained larger and more disorganized in the 1α(OH)ase−/− mice. The parathyroid glands were enlarged in both mutants, but fed the rescue diet, the parathyroid gland size normalized in the VDR−/− mice, but remained increased in the 1α(OH)ase−/− mice [185]. Thus, although there are many well-defined actions of vitamin D on osteoblasts, the response to normalization of serum calcium in HVDRR patients and VDR−/− mice suggests that 1,25(OH)2D3 action on osteoblasts is not essential to form bone, although the bone is not completely normal. The implication is that 1,25(OH)2D3 action on bone is mainly due to its effects on intestinal mineral absorption to provide calcium and phosphate for bone formation. The same conclusion was reached by Underwood and DeLuca [186], who showed that rickets could be prevented in vitamin D–deficient rats by calcium and phosphate infusions. Also, normalizing serum calcium by IV infusion is sufficient to suppress PTH overproduction in HVDRR children. This suggests that the hypophosphatemia in these patients is mainly the result of secondary hyperparathyroidism and not inadequate intestinal phosphate absorption. The findings also suggest that 1,25(OH)2D3 action is not essential to suppress PTH in secondary hyperparathyroidism.

CHAPTER 72 Hereditary 1,25-Dihydroxyvitamin D–Resistant Rickets

The mechanism for the spontaneous improvement exhibited by some HVDRR children as they get older is an interesting dilemma. One hypothesis, that explains the normalization of the 1,25(OH)2D3 endocrine system in the face of inactive VDRs, is that some other transcription factor can substitute for the defective vitamin D system. Possibly RAR, RXR, or TR can substitute for a nonfunctional VDR and activate the appropriate target genes to reverse the hypocalcemia and restore the bones to normal. In this context, Whitfield et al. [140] have shown in vitro that addition of RXR can rescue mutant VDR with defects in the heterodimerization domain and restore hormone responsiveness. Another possibility is that a diet sufficient in calcium may bypass the need for vitamin D action when the demand for calcium for bone growth is diminished at an older age. In conclusion, the biochemical and genetic analysis of the VDR in HVDRR patients has yielded important insights into the structure and function of the receptor in mediating 1,25(OH)2D3 action. Similarly, studies of the affected children with HVDRR continues to provide further insight into the biological role of 1,25(OH)2D3 in vivo. A concerted investigative approach of HVDRR at the clinical, cellular, and molecular level has proven exceedingly valuable in understanding the mechanism of action of 1,25(OH)2D3 and improving the diagnostic and clinical management of this rare genetic disease.

9.

10.

11. 12.

13. 14.

15.

16. 17.

References 1. Feldman D, Malloy PJ 1990 Hereditary 1,25-dihydroxyvitamin D–resistant rickets: molecular basis and implications for the role of 1,25(OH)2D3 in normal physiology. Mol Cell Endocrinol 72:C57–62. 2. Hughes MR, Malloy PJ, O’Malley BW, Pike JW, Feldman D 1991 Genetic defects of the 1,25-dihydroxyvitamin D3 receptor. J Recept Res 11:699–716. 3. Malloy PJ, Pike JW, Feldman D 1999 The vitamin D receptor and the syndrome of hereditary 1,25-dihydroxyvitamin D– resistant rickets. Endocr Rev 20:156–188. 4. Albright F, Butler AM, Bloomberg E 1937 Rickets resistant to vitamin D therapy. Am J Dis Child 54:531–547. 5. Prader VA, Illig R, Heierli E 1961 Eine besondere form der primaren vitamin D–resistenten rachitis mit hypocalcamie und autosomal-dominantem erbgang: die hereditare pseudomangelrachitis. Helvetica Paediatrica Acta 16:452–468. 6. Takeyama K, Kitanaka S, Sato T, Kobori M, Yanagisawa J, Kato S 1997 25-Hydroxyvitamin D3 1alpha-hydroxylase and vitamin D synthesis. Science 277:1827–1830. 7. Fu GK, Lin D, Zhang MY, Bikle DD, Shackleton CH, Miller WL, Portale AA 1997 Cloning of human 25-hydroxyvitamin D-1 alpha-hydroxylase and mutations causing vitamin D–dependent rickets type 1. Mol Endocrinol 11: 1961–1970. 8. Kitanaka S, Takeyama K, Murayama A, Sato T, Okumura K, Nogami M, Hasegawa Y, Niimi H, Yanagisawa J, Tanaka T,

18.

19.

20.

21. 22.

23.

1231 Kato S 1998 Inactivating mutations in the 25-hydroxyvitamin D3 1(alpha)-hydroxylase gene in patients with pseudovitamin D–deficiency rickets. N Engl J Med 338: 653–661. Brooks MH, Bell NH, Love L, Stern PH, Orfei E, Queener SF, Hamstra AJ, DeLuca HF 1978 Vitamin D–dependent rickets type II. Resistance of target organs to 1,25-dihydroxyvitamin D. N Engl J Med 298:996–999. Marx SJ, Spiegel AM, Brown EM, Gardner DG, Downs RW, Jr., Attie M, Hamstra AJ, DeLuca HF 1978 A familial syndrome of decrease in sensitivity to 1,25-dihydroxyvitamin D. J Clin Endocrinol Metab 47:1303–1310. Rosen JF, Fleischman AR, Finberg L, Hamstra A, DeLuca HF 1979 Rickets with alopecia: an inborn error of vitamin D metabolism. J Pediatr 94:729–735. Liberman UA, Samuel R, Halabe A, Kauli R, Edelstein S, Weisman Y, Papapoulos SE, Clemens TL, Fraher LJ, O’Riordan JL 1980 End-organ resistance to 1,25-dihydroxycholecalciferol. Lancet 1:504–506. Sockalosky JJ, Ulstrom RA, DeLuca HF, Brown DM 1980 Vitamin D–resistant rickets: end-organ unresponsiveness to 1,25(OH)2D3. J Pediatr 96:701–703. Kudoh T, Kumagai T, Uetsuji N, Tsugawa S, Oyanagi K, Chiba Y, Minami R, Nakao T 1981 Vitamin D–dependent rickets: decreased sensitivity to 1,25-dihydroxyvitamin D. Eur J Pediatr 137:307–311. Balsan S, Garabedian M, Liberman UA, Eil C, Bourdeau A, Guillozo H, Grimberg R, Le Deunff MJ, Lieberherr M, Guimbaud P, Broyer M, Marx SJ 1983 Rickets and alopecia with resistance to 1,25-dihydroxyvitamin D: two different clinical courses with two different cellular defects. J Clin Endocrinol Metab 57:803–811. Laufer D, Benderly A, Hochberg Z 1987 Dental pathology in calcitirol resistant rickets. J Oral Med 42:272–275. Bliziotes M, Yergey AL, Nanes MS, Muenzer J, Begley MG, Viera NE, Kher KK, Brandi ML, Marx SJ 1988 Absent intestinal response to calciferols in hereditary resistance to 1,25-dihydroxyvitamin D: documentation and effective therapy with high dose intravenous calcium infusions. J Clin Endocrinol Metab 66:294–300. Fraher LJ, Karmali R, Hinde FR, Hendy GN, Jani H, Nicholson L, Grant D, O’Riordan JL 1986 Vitamin D–dependent rickets type II: extreme end organ resistance to 1,25dihydroxy vitamin D3 in a patient without alopecia. Eur J Pediatr 145:389–395. Beer S, Tieder M, Kohelet D, Liberman OA, Vure E, BarJoseph G, Gabizon D, Borochowitz ZU, Varon W, Modai D 1981 Vitamin D–resistant rickets with alopecia: A form of end organ resistance to 1,25-dihydroxyvitamin D. Clin Endocrinol 14:395–402. Feldman D, Chen T, Cone C, Hirst M, Shani S, Benderli A, Hochberg Z 1982 Vitamin D–resistant rickets with alopecia: cultured skin fibroblasts exhibit defective cytoplasmic receptors and unresponsiveness to 1,25(OH)2D3. J Clin Endocrinol Metab 55:1020–1022. Hochberg Z, Benderli A, Levy J, Vardi P, Weisman Y, Chen T, Feldman D 1984 1,25-Dihydroxyvitamin D resistance, rickets, and alopecia. Am J Med 77:805–811. Hirst MA, Hochman HI, Feldman D 1985 Vitamin D resistance and alopecia: a kindred with normal 1,25dihydroxyvitamin D binding, but decreased receptor affinity for deoxyribonucleic acid. J Clin Endocrinol Metab 60: 490–495. Takeda E, Kuroda Y, Saijo T, Naito E, Kobashi H, Yokota I, Miyao M 1987 1 alpha-hydroxyvitamin D3 treatment of three

1232

24.

25.

26.

27. 28. 29. 30. 31.

32.

33. 34. 35. 36. 37. 38. 39.

40.

41.

42.

patients with 1,25-dihydroxyvitamin D–receptor-defect rickets and alopecia. Pediatrics 80:97–101. Yagi H, Ozono K, Miyake H, Nagashima K, Kuroume T, Pike JW 1993 A new point mutation in the deoxyribonucleic acid-binding domain of the vitamin D–receptor in a kindred with hereditary 1,25-dihydroxyvitamin D-resistant rickets. J Clin Endocrinol Metab 76:509–512. Malloy PJ, Weisman Y, Feldman D 1994 Hereditary 1 alpha,25-dihydroxyvitamin D–resistant rickets resulting from a mutation in the vitamin D receptor deoxyribonucleic acid-binding domain. J Clin Endocrinol Metab 78:313–316. Malloy PJ, Hochberg Z, Tiosano D, Pike JW, Hughes MR, Feldman D 1990 The molecular basis of hereditary 1,25-dihydroxyvitamin D3–resistant rickets in seven related families. J Clin Invest 86:2071–2079. Heaney RP 1997 Vitamin D: role in the calcium economy. In: Feldman D, Glorieux F, Pike JW (eds) Vitamin D. Academic Press: San Diego, pp 485–497. Heaney RP, Barger-Lux MJ, Dowell MS, Chen TC, Holick MF 1997 Calcium absorptive effects of vitamin D and its major metabolites. J Clin Endocrinol Metab 82:4111–4116. Hochberg Z, Gilhar A, Haim S, Friedman-Birnbaum R, Levy J, Benderly A 1985 Calcitriol-resistant rickets with alopecia. Arch Dermatol 121:646–647. Marx SJ, Bliziotes MM, Nanes M 1986 Analysis of the relation between alopecia and resistance to 1,25-dihydroxyvitamin D. Clin Endocrinol 25:373–381. Stumpf WE, Sar M, Reid FA, Tanaka Y, DeLuca HF 1979 Target cells for 1,25-dihydroxyvitamin D3 in intestinal tract, stomach, kidney, skin, pituitary, and parathyroid. Science 206:1188–1190. Berger U, Wilson P, McClelland RA, Colston K, Haussler MR, Pike JW, Coombes RC 1988 Immunocytochemical detection of 1,25-dihydroxyvitamin–D receptors in normal human tissues. J Clin Endocrinol Metab 67:607–613. Reichel H, Koeffler HP, Norman AW 1989 The role of the vitamin D endocrine system in health and disease. N Engl J Med 320:980–991. Pike JW 1991 Vitamin D3 receptors: structure and function in transcription. Ann Rev Nutr 11:189–216. Walters MR 1992 Newly identified actions of the vitamin D endocrine system. Endocrine Rev 13:719–764. Bikle DD 1992 Clinical counterpoint: Vitamin D: new actions, new analogs, new therapeutic potential. Endocrine Rev 13:765–784. Darwish H, DeLuca HF 1993 Vitamin D–regulated gene expression. Crit Rev Eukaryot Gene Expr 3:89–116. MacDonald PN, Dowd DR, Haussler MR 1994 New insight into the structure and functions of the vitamin D receptor. Semin Nephrol 14:101–118. Feldman D, Malloy PJ, Gross C 2001 Vitamin D: biology, action, and clinical implications. In: Marcus R, Feldman D, Kelsey J (eds) Osteoporosis. Academic Press: San Diego, pp 257–303. Hochberg Z, Borochowitz Z, Benderli A, Vardi P, Oren S, Spirer Z, Heyman I, Weisman Y 1985 Does 1,25-dihydroxyvitamin D participate in the regulation of hormone release from endocrine glands? J Clin Endocrinol Metab 60: 57–61. Even L, Weisman Y, Goldray D, Hochberg Z 1996 Selective modulation by vitamin D of renal response to parathyroid hormone: a study in calcitriol-resistant rickets. J Clin Endocrinol Metab 81:2836–2840. Manolagas SC, Yu XP, Girasole G, Bellido T 1994 Vitamin D and the hematolymphopoietic tissue: a 1994 update. Semin Nephrol 14:129–143.

PETER J. MALLOY, J. WESLEY PIKE, AND DAVID FELDMAN

43. Etzioni A, Hochberg Z, Pollak S, Meshulam T, Zakut V, Tzehoval E, Keisari Y, Aviram I, Spirer Z, Benderly A, Weisman Y 1989 Defective leukocyte fungicidal activity in end-organ resistance to 1,25-dihydroxyvitamin D. Pediatr Res 25:276–279. 44. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia, and growth retardation after weaning. Nat Genet 16:391–396. 45. Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997 Targeted ablation of the vitamin D receptor: an animal model of vitamin D–dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94:9831–9835. 46. Mangelsdorf DJ, Thummel C, Beato M, Herrlich P, Schutz G, Umesono K, Blumberg B, Kastner P, Mark M, Chambon P, et al. 1995 The nuclear receptor superfamily: the second decade. Cell 83:835–839. 47. Berg JM 1988 Proposed structure for the zinc-binding domains from transcription factor IIIA and related proteins. Proc Natl Acad Sci USA 85:99–102. 48. Mader S, Kumar V, de Verneuil H, Chambon P 1989 Three amino acids of the oestrogen receptor are essential to its ability to distinguish an oestrogen from a glucocorticoid-responsive element. Nature 338:271–274. 49. Umesono K, Evans RM 1989 Determinants of target gene specificity for steroid/thyroid hormone receptors. Cell 57:1139–1146. 50. Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179. 51. Blanco JC, Wang IM, Tsai SY, Tsai MJ, O’Malley BW, Jurutka PW, Haussler MR, Ozato K 1995 Transcription factor TFIIB and the vitamin D receptor cooperatively activate ligand-dependent transcription. Proc Natl Acad Sci USA 92:1535–1539. 52. MacDonald PN, Sherman DR, Dowd DR, Jefcoat SC, Jr., DeLisle RK 1995 The vitamin D receptor interacts with general transcription factor IIB. J Biol Chem 270:4748–4752. 53. Onate SA, Tsai SY, Tsai MJ, O’Malley BW 1995 Sequence and characterization of a coactivator for the steroid hormone receptor superfamily. Science 270:1354–1357. 54. Masuyama H, Brownfield CM, St-Arnaud R, MacDonald PN 1997 Evidence for ligand-dependent intramolecular folding of the AF-2 domain in vitamin D receptor-activated transcription and coactivator interaction. Mol Endocrinol 11:1507–1517. 55. Gill RK, Atkins LM, Hollis BW, Bell NH 1998 Mapping the domains of the interaction of the vitamin D receptor and steroid receptor coactivator-1. Mol Endocrinol 12:57–65. 56. DeLuca HF, Zierold C 1998 Mechanisms and functions of vitamin D. Nutr Rev 56:S4–10; discussion S54–75. 57. Haussler MR, Whitfield GK, Haussler CA, Hsieh J, Thompson PD, Selznick SH, Dominguez CE, Jurutka PW 1998 The nuclear vitamin D receptor: biological and molecular regulatory properties revealed. J Bone Min Res 13:325–349. 58. Rachez C, Freedman LP 2000 Mechanisms of gene regulation by vitamin D3 receptor: a network of coactivator interactions. Gene 246:9–21. 59. Minghetti PP, Norman AW 1988 1,25(OH)2-vitamin D3 receptors: gene regulation and genetic circuitry. FASEB J 2: 3043–3053. 60. Racz A, Barsony J 1999 Hormone-dependent translocation of vitamin D receptors is linked to transactivation. J Biol Chem 274:19352–19360.

CHAPTER 72 Hereditary 1,25-Dihydroxyvitamin D–Resistant Rickets

61. Jurutka PW, Hsieh JC, MacDonald PN, Terpening CM, Haussler CA, Haussler MR, Whitfield GK 1993 Phosphorylation of serine 208 in the human vitamin D receptor. The predominant amino acid phosphorylated by casein kinase II, in vitro, and identification as a significant phosphorylation site in intact cells. J Biol Chem 268:6791–6799. 62. Freedman LP 1999 Increasing the complexity of coactivation in nuclear receptor signaling. Cell 97:5–8. 63. Chen TL, Hirst MA, Cone CM, Hochberg Z, Tietze HU, Feldman D 1984 1,25-dihydroxyvitamin D resistance, rickets, and alopecia: analysis of receptors and bioresponse in cultured fibroblasts from patients and parents. J Clin Endocrinol Metab 59:383–388. 64. Kerner SA, Scott RA, Pike JW 1989 Sequence elements in the human osteocalcin gene confer basal activation and inducible response to hormonal vitamin D3. Proc Natl Acad Sci USA 86:4455–4459. 65. Morrison NA, Shine J, Fragonas JC, Verkest V, McMenemy ML, Eisman JA 1989 1,25-dihydroxyvitamin D–responsive element and glucocorticoid repression in the osteocalcin gene. Science 246:1158–1161. 66. Ozono K, Liao J, Kerner SA, Scott RA, Pike JW 1990 The vitamin D–responsive element in the human osteocalcin gene. Association with a nuclear proto-oncogene enhancer. J Biol Chem 265:21881–21888. 67. Lian J, Stewart C, Puchacz E, Mackowiak S, Shalhoub V, Collart D, Zambetti G, Stein G 1989 Structure of the rat osteocalcin gene and regulation of vitamin D–dependent expression. Proc Natl Acad Sci USA 86:1143–1147. 68. Demay MB, Gerardi JM, DeLuca HF, Kronenberg HM 1990 DNA sequences in the rat osteocalcin gene that bind the 1,25-dihydroxyvitamin D3 receptor and confer responsiveness to 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 87:369–373. 69. Terpening CM, Haussler CA, Jurutka PW, Galligan MA, Komm BS, Haussler MR 1991 The vitamin D–responsive element in the rat bone Gla protein gene is an imperfect direct repeat that cooperates with other cis-elements in 1,25dihydroxyvitamin D3-mediated transcriptional activation. Mol Endocrinol 5:373–385. 70. Noda M, Vogel RL, Craig AM, Prahl J, DeLuca HF, Denhardt DT 1990 Identification of a DNA sequence responsible for binding of the 1,25-dihydroxyvitamin D3 receptor and 1,25-dihydroxyvitamin D3 enhancement of mouse secreted phosphoprotein 1 (SPP-1 or osteopontin) gene expression. Proc Natl Acad Sci USA 87:9995–9999. 71. Darwish HM, DeLuca HF 1992 Identification of a 1,25-dihydroxyvitamin D3–response element in the 5′-flanking region of the rat calbindin D-9k gene. Proc Natl Acad Sci USA 89:603–607. 72. Gill RK, Christakos S 1993 Identification of sequence elements in mouse calbindin-D28k gene that confer 1,25dihydroxyvitamin D3- and butyrate-inducible responses. Proc Natl Acad Sci USA 90:2984–2988. 73. Ohyama Y, Ozono K, Uchida M, Shinki T, Kato S, Suda T, Yamamoto O, Noshiro M, Kato Y 1994 Identification of a vitamin D–responsive element in the 5′-flanking region of the rat 25-hydroxyvitamin D3 24-hydroxylase gene. J Biol Chem 269:10545–10550. 74. Chen KS, DeLuca HF 1995 Cloning of the human 1 alpha,25-dihydroxyvitamin D3 24-hydroxylase gene promoter and identification of two vitamin D–responsive elements. Biochim Biophys Acta 1263:1–9. 75. Zierold C, Darwish HM, DeLuca HF 1995 Two vitamin D response elements function in the rat 1,25-dihydroxyvitamin D 24-hydroxylase promoter. J Biol Chem 270:1675–1678.

1233

76. Hughes MR, Malloy PJ, Kieback DG, Kesterson RA, Pike JW, Feldman D, O’Malley BW 1988 Point mutations in the human vitamin D receptor gene associated with hypocalcemic rickets. Science 242:1702–1705. 77. Ritchie HH, Hughes MR, Thompson ET, Malloy PJ, Hochberg Z, Feldman D, Pike JW, O’Malley BW 1989 An ochre mutation in the vitamin D receptor gene causes hereditary 1,25-dihydroxyvitamin D3–resistant rickets in three families. Proc Natl Acad Sci USA 86:9783–9787. 78. Sone T, Scott RA, Hughes MR, Malloy PJ, Feldman D, O’Malley BW, Pike JW 1989 Mutant vitamin D receptors which confer hereditary resistance to 1,25-dihydroxyvitamin D3 in humans are transcriptionally inactive in vitro. J Biol Chem 264:20230–20234. 79. Zerwekh JE, Glass K, Jowsey J, Pak CY 1979 An unique form of osteomalacia associated with end organ refractoriness to 1,25-dihydroxyvitamin D and apparent defective synthesis of 25-hydroxyvitamin D. J Clin Endocrinol Metab 49:171–175. 80. Colston K, Hirst M, Feldman D 1980 Organ distribution of the cytoplasmic 1,25-dihydroxycholecalciferol receptor in various mouse tissues. Endocrinology 107:1916–1922. 81. Feldman D, Chen T, Hirst M, Colston K, Karasek M, Cone C 1980 Demonstration of 1,25-dihydroxyvitamin D3 receptors in human skin biopsies. J Clin Endocrinol Metab 51:1463–1465. 82. Simpson RU, DeLuca HF 1980 Characterization of a receptor-like protein for 1,25-dihydroxyvitamin D3 in rat skin. Proc Natl Acad Sci USA 77:5822–5826. 83. Eil C, Liberman UA, Rosen JF, Marx SJ 1981 A cellular defect in hereditary vitamin-D–dependent rickets type II: defective nuclear uptake of 1,25-dihydroxyvitamin D in cultured skin fibroblasts. N Engl J Med 304:1588–1591. 84. Clemens TL, Adams JS, Horiuchi N, Gilchrest BA, Cho H, Tsuchiya Y, Matsuo N, Suda T, Holick MF 1983 Interaction of 1,25-dihydroxyvitamin-D3 with keratinocytes and fibroblasts from skin of normal subjects and a subject with vitamin D– dependent rickets, type II: A model for study of the mode of action of 1,25-dihydroxyvitamin D3. J Clin Endocrinol Metab 56:824–830. 85. Griffin JE, Zerwekh JE 1983 Impaired stimulation of 25-hydroxyvitamin D-24-hydroxylase in fibroblasts from a patient with vitamin D–dependent rickets, type II. A form of receptor-positive resistance to 1,25-dihydroxyvitamin D3. J Clin Invest 72:1190–1199. 86. Liberman UA, Eil C, Marx SJ 1983 Resistance to 1,25(OH)2D3: Association with heterogeneous defects in cultured skin fibroblasts. J Clin Invest 71:192–200. 87. Liberman UA, Eil C, Holst P, Rosen JF, Marx SJ 1983 Hereditary resistance to 1,25-dihydroxyvitamin D: defective function of receptors for 1,25-dihydroxyvitamin D in cells cultured from bone. J Clin Endocrinol Metab 57: 958–962. 88. Liberman UA, Eil C, Marx SJ 1986 Receptor-positive hereditary resistance to 1,25-dihydroxyvitamin D: chromatography of receptor complexes on deoxyribonucleic acid-cellulose shows two classes of mutation. J Clin Endocrinol Metab 62:122–126. 89. Malloy PJ, Hochberg Z, Pike JW, Feldman D 1989 Abnormal binding of vitamin D receptors to deoxyribonucleic acid in a kindred with vitamin D–dependent rickets, type II. J Clin Endocrinol Metab 68:263–269. 90. Pike JW, Dokoh S, Haussler MR, Liberman UA, Marx SJ, Eil C 1984 Vitamin D3–resistant fibroblasts have immunoassayable 1,25-dihydroxyvitamin D3 receptors. Science 224: 879–881.

1234 91. Dokoh S, Haussler MR, Pike JW 1984 Development of a radioligand immunoassay for 1,25-dihydroxycholecalciferol receptors utilizing monoclonal antibody. Biochem J 221:129–136. 92. Pike JW, Donaldson CA, Marion SL, Haussler MR 1982 Development of hybridomas secreting monoclonal antibodies to the chicken intestinal 1 alpha,25-dihydroxyvitamin D3 receptor. Proc Natl Acad Sci USA 79:7719–7723. 93. Pike JW, Marion SL, Donaldson CA, Haussler MR 1983 Serum and monoclonal antibodies against the chick intestinal receptor for 1,25-dihydroxyvitamin D3. Generation by a preparation enriched in a 64,000-dalton protein. J Biol Chem 258:1289–1296. 94. Pike JW 1984 Monoclonal antibodies to chick intestinal receptors for 1,25-dihydroxyvitamin D3. Interaction and effects of binding on receptor function. J Biol Chem 259:1167–1173. 95. Gamblin GT, Liberman UA, Eil C, Downs RWJ, Degrange DA, Marx SJ 1985 Vitamin D–dependent rickets type II: Defective induction of 25-hydroxyvitamin D3-24-hydroxylase by 1,25-dihydroxyvitamin D3 in cultured skin fibroblasts. J Clin Invest 75:954–960. 96. Castells S, Greig F, Fusi MA, Finberg L, Yasumura S, Liberman UA, Eil C, Marx SJ 1986 Severely deficient binding of 1,25-dihydroxyvitamin D to its receptors in a patient responsive to high doses of this hormone. J Clin Endocrinol Metab 63:252–256. 97. Koren R, Ravid A, Liberman UA, Hochberg Z, Weisman Y, Novogrodsky A 1985 Defective binding and function of 1,25-dihydroxyvitamin D3 receptors in peripheral mononuclear cells of patients with end-organ resistance to 1,25-dihydroxyvitamin D. J Clin Invest 76:2012–2015. 98. Takeda E, Kuroda Y, Saijo T, Toshima K, Naito E, Kobashi H, Iwakuni Y, Miyao M 1986 Rapid diagnosis of vitamin D– dependent rickets type II by use of phytohemagglutinin-stimulated lymphocytes. Clin Chim Acta 155:245–250. 99. Takeda E, Yokota I, Ito M, Kobashi H, Saijo T, Kuroda Y 1990 25-Hydroxyvitamin D-24-hydroxylase in phytohemagglutinin-stimulated lymphocytes: intermediate bioresponse to 1,25-dihydroxyvitamin D3 of cells from parents of patients with vitamin D–dependent rickets type II. J Clin Endocrinol Metab 70:1068–1074. 100. Nagler A, Merchav S, Fabian I, Tatarsky I, Hochberg Z 1987 Myeloid progenitors from the bone marrow of patients with vitamin D–resistant rickets (type II) fail to respond to 1,25(OH)2D3. Brit J Haematol 67:267–271. 101. Koeffler HP, Bishop JE, Reichel H, Singer F, Nagler A, Tobler A, Walka M, Norman AW 1990 Lymphocyte cell lines from vitamin D–dependent rickets type II show functional defects in the 1 alpha,25-dihydroxyvitamin D3 receptor. Mol Cell Endocrinol 70:1–11. 102. Takeda E, Yokota I, Saijo T, Kawakami I, Ito M, Kuroda Y 1990 Effect of long-term treatment with massive doses of 1 alpha-hydroxyvitamin D3 on calcium-phosphate balance in patients with vitamin D–dependent rickets type II. Acta Paediatr Jpn 32:39–43. 103. Baker AR, McDonnell DP, Hughes M, Crisp TM, Mangelsdorf DJ, Haussler MR, Pike JW, Shine J, O’Malley BW 1988 Cloning and expression of full-length cDNA encoding human vitamin D receptor. Proc Natl Acad Sci USA 85:3294–3298. 104. Saiki RK, Gelfand DH, Stoffel S, Scharf SJ, Higuchi R, Horn GT, Mullis KB, Erlich HA 1988 Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239:487–491.

PETER J. MALLOY, J. WESLEY PIKE, AND DAVID FELDMAN

105. Refetoff S, Weiss RE, Usala SJ 1993 The syndromes of resistance to thyroid hormone. Endocrine Rev 14: 348–399. 106. Quigley CA, De Bellis A, Marschke KB, el-Awady MK, Wilson EM, French FS 1995 Androgen receptor defects: historical, clinical, and molecular perspectives. Endocrine Rev 16:271–321. 107. Smith EP, Boyd J, Frank GR, Takahashi H, Cohen RM, Specker B, Williams TC, Lubahn DB, Korach KS 1994 Estrogen resistance caused by a mutation in the estrogenreceptor gene in a man. N Engl J Med 331:1056–1061. 108. Werner S, Bronnegard M 1996 Molecular basis of glucocorticoid-resistant syndromes. Steroids 61:216–221. 109. Geller DS, Rodriguez-Soriano J, Vallo Boado A, Schifter S, Bayer M, Chang SS, Lifton RP 1998 Mutations in the mineralocorticoid receptor gene cause autosomal dominant pseudohypoaldosteronism type I. Nat Genet 19:279–281. 110. Sone T, Marx SJ, Liberman UA, Pike JW 1990 A unique point mutation in the human vitamin D receptor chromosomal gene confers hereditary resistance to 1,25-dihydroxyvitamin D3. Mol Endocrinol 4:623–631. 111. Saijo T, Ito M, Takeda E, Huq AH, Naito E, Yokota I, Sone T, Pike JW, Kuroda Y 1991 A unique mutation in the vitamin D receptor gene in three Japanese patients with vitamin D–dependent rickets type II: utility of single-strand conformation polymorphism analysis for heterozygous carrier detection. Am J Hum Genet 49:668–673. 112. Takeda E, Yokota I, Kawakami I, Hashimoto T, Kuroda Y, Arase S 1989 Two siblings with vitamin D–dependent rickets type II: no recurrence of rickets for 14 years after cessation of therapy. Eur J Pediatr 149:54–57. 113. Yokota I, Takeda E, Ito M, Kobashi H, Saijo T, Kuroda Y 1991 Clinical and biochemical findings in parents of children with vitamin D–dependent rickets Type II. J Inherit Metab Dis 14:231–240. 114. Rut AR, Hewison M, Kristjansson K, Luisi B, Hughes MR, O’Riordan JL 1994 Two mutations causing vitamin D– resistant rickets: modelling on the basis of steroid hormone receptor DNA-binding domain crystal structures. Clin Endocrinol 41:581–590. 115. Lin JP, Uttley WS 1993 Intra-atrial calcium infusions, growth, and development in end organ resistance to vitamin D. Arch Dis Child 69:689–692. 116. Simonin G, Chabrol B, Moulene E, Bollini G, Strouc S, Mattei JF, Giraud F 1992 Vitamin D–resistant rickets type II: apropos of 2 cases. Pediatrie 47:817–820. 117. Lin NU-T, Malloy PJ, Sakati N, Al-Ashwal A, Feldman D 1996 A novel mutation in the deoxyribonucleic acid-binding domain of the vitamin D receptor gene causes hereditary 1,25-dihydroxyvitamin D–resistant rickets. J Clin Endocrinol Metab 81:2564–2569. 118. Sakati N, Woodhouse NJY, Niles N, Harfi H, de Grange DA, Marx S 1986 Hereditary resistance to 1,25-dihydroxyvitamin D: clinical and radiological improvement during high-dose oral calcium therapy. Hormone Res 24:280–287. 119. Luisi BF, Xu WX, Otinowski Z, Freedman LP, Yamamoto KR, Sigler PB 1991 Crystallographic analysis of the interaction of the glucocorticoid receptor with DNA. Nature 352: 497–505. 120. Rastinejad F, Perlmann T, Evans RM, Sigler PB 1995 Structural determinants of nuclear receptor assembly on DNA direct repeats. Nature 375:203–211. 121. Freedman LP, Luisi BF 1993 On the mechanism of DNA binding by nuclear hormone receptors: a structural and functional perspective. J Cell Biochem 51:140–150.

CHAPTER 72 Hereditary 1,25-Dihydroxyvitamin D–Resistant Rickets

122. Luisi BF, Schwabe JW, Freedman LP 1994 The steroid/ nuclear receptors: from three-dimensional structure to complex function. Vitam Horm 49:1–47. 123. Burns K, Duggan B, Atkinson EA, Famulski KS, Nemer M, Bleackley RC, Michalak M 1994 Modulation of gene expression by calreticulin binding to the glucocorticoid receptor. Nature 367:476–480. 124. Cao X, Teitelbaum SL, Dedhar S, Zhang L, Ross FP 1994 Retinoic acid and calreticulin, a novel ER-derived transcription factor inhibits 1,25(OH)2D3-induced gene transcription. J Bone Miner Res 9:S145. 125. Dedhar S, Rennie PS, Shago M, Leung Hagesteijn C-Y, Yang H, Filmus J, Hawley RG, Bruchovsky N, Cheng H, Matusik RJ, Giguere V 1994 Inhibition of nuclear hormone receptor activity by calreticulin. Nature 367:480–483. 126. Orkin SH 1984 The mutation and polymorphism of the human ß-globin gene and its surrounding DNA. Ann Rev Genet 18:131–171. 127. Neufeld EF 1989 Natural history and inherited disorders of a lysosomal enzyme, β-hexosaminidase. J Biol Chem 264:10927–10930. 128. Sicinski P, Geng Y, Ryder-Cook AS, Barnard EA, Darlison MG, Barnard PJ 1989 The molecular basis of muscular dystrophy in the mdx mouse: a point mutation. Science 244: 1578–1580. 129. Malloy PJ, Hughes MR, Pike JW, Feldman D 1991 Vitamin D receptor mutations and hereditary 1,25-dihydroxyvitamin D– resistant rickets. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D: gene regulation, structure-function analysis, and clinical application. Eighth workshop on vitamin D. Walter de Gruyter: New York, pp 116–124. 130. Wiese RJ, Goto H, Prahl JM, Marx SJ, Thomas M, al-Aqeel A, DeLuca HF 1993 Vitamin D-dependency rickets type II: truncated vitamin D–receptor in three kindreds. Mol Cell Endocrinol 90:197–201. 131. Cockerill FJ, Hawa NS, Yousaf N, Hewison M, O’Riordan JL, Farrow SM 1997 Mutations in the vitamin D receptor gene in three kindreds associated with hereditary vitamin D–resistant rickets. J Clin Endocrinol Metab 82:3156–3160. 132. Kristjansson K, Rut AR, Hewison M, O’Riordan JL, Hughes MR 1993 Two mutations in the hormone binding domain of the vitamin D receptor cause tissue resistance to 1,25 dihydroxyvitamin D3. J Clin Invest 92:12–16. 133. Barsony J, McKoy W, DeGrange DA, Liberman UA, Marx SJ 1989 Selective expression of a normal action of the 1,25dihydroxyvitamin D3 receptor in human skin fibroblasts with hereditary severe defects in multiple actions of that receptor. J Clin Invest 83:2093–2101. 134. Zhu WJ, Malloy PJ, Delvin E, Chabot G, Feldman D 1998 Hereditary 1,25-dihydroxyvitamin D–resistant rickets due to an opal mutation causing premature termination of the vitamin D receptor. J Bone Miner Res 13:259–264. 135. Mechica JB, Leite MO, Mendonca BB, Frazzatto ES, Borelli A, Latronico AC 1997 A novel nonsense mutation in the first zinc finger of the vitamin D receptor causing hereditary 1,25dihydroxyvitamin D3–resistant rickets. J Clin Endocrinol Metab 82:3892–3894. 136. Malloy PJ, Zhu W, Bouillon R, Feldman D 2002 A novel nonsense mutation in the ligand binding domain of the vitamin D receptor causes hereditary 1,25-dihydroxyvitamin D– resistant rickets. Mol Genet Metab 77:314–318. 137. Hawa NS, Cockerill FJ, Vadher S, Hewison M, Rut AR, Pike JW, O’Riordan JL, Farrow SM 1996 Identification of a novel mutation in hereditary vitamin D–resistant rickets causing exon skipping. Clin Endocrinol 45:85–92.

1235

138. Thompson E, Kristjansson K, Hughes M Molecular scanning methods for mutation detection: application to the 1,25-dihydroxyvitamin D receptor. Eighth workshop on Vitamin D, Paris, France, 1991, p 6. 139. Rut AR, Hewison M, Rowe P, Hughes M, Grant D, O’Riordan JLH 1991 A novel mutation in the steroid binding region of the vitamin D receptor (VDR) gene in hereditary vitamin D–resistant rickets (HVDRR). In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D: gene regulation, structurefunction analysis, and clinical application. Eighth workshop on vitamin D. Walter de Gruyter: New York, pp 94–95. 140. Whitfield GK, Selznick SH, Haussler CA, Hsieh JC, Galligan MA, Jurutka PW, Thompson PD, Lee SM, Zerwekh JE, Haussler MR 1996 Vitamin D receptors from patients with resistance to 1,25-dihydroxyvitamin D3: point mutations confer reduced transactivation in response to ligand and impaired interaction with the retinoid X receptor heterodimeric partner. Mol Endocrinol 10:1617–1631. 141. Malloy PJ, Eccleshall TR, Gross C, Van Maldergem L, Bouillon R, Feldman D 1997 Hereditary vitamin D–resistant rickets caused by a novel mutation in the vitamin D receptor that results in decreased affinity for hormone and cellular hyporesponsiveness. J Clin Invest 99:297–304. 142. Van Maldergem L, Bachy A, Feldman D, Bouillon R, Maassen J, Dreyer M, Rey R, Holm C, Gillerot Y 1996 Syndrome of lipoatrophic diabetes, vitamin D–resistant rickets, and persistent Müllerian ducts in a Turkish boy born to consanguineous parents. Am J Med Genet 64:506–513. 143. Magre J, Delepine M, Khallouf E, Gedde-Dahl T, Jr., Van Maldergem L, Sobel E, Papp J, Meier M, Megarbane A, Bachy A, Verloes A, d’Abronzo FH, Seemanova E, Assan R, Baudic N, Bourut C, Czernichow P, Huet F, Grigorescu F, de Kerdanet M, Lacombe D, Labrune P, Lanza M, Loret H, Matsuda F, Navarro J, Nivelon-Chevalier A, Polak M, Robert JJ, Tric P, Tubiana-Rufi N, Vigouroux C, Weissenbach J, Savasta S, Maassen JA, Trygstad O, Bogalho P, Freitas P, Medina JL, Bonnicci F, Joffe BI, Loyson G, Panz VR, Raal FJ, O’Rahilly S, Stephenson T, Kahn CR, Lathrop M, Capeau J 2001 Identification of the gene altered in Berardinelli-Seip congenital lipodystrophy on chromosome 11q13. Nat Genet 28:365–370. 144. Nguyen TM, Adiceam P, Kottler ML, Guillozo H, RizkRabin M, Brouillard F, Lagier P, Palix C, Garnier JM, Garabedian M 2002 Tryptophan missense mutation in the ligand-binding domain of the vitamin D receptor causes severe resistance to 1,25-dihydroxyvitamin D. J Bone Miner Res 17:1728–1737. 145. Malloy PJ, Xu R, Peng L, Peleg S, Ashwal A, Feldman D 2004 Heredity 1,25-dihydroxyvitamin D resistant rickets due to a mutation causing multiple defects in vitamin D receptor function. Endocrinology, in press. 146. Malloy PJ, Zhu W, Zhao XY, Pehling GB, Feldman D 2001 A novel inborn error in the ligand-binding domain of the vitamin D receptor causes hereditary vitamin D–resistant rickets. Mol Genet Metab 73:138–148. 147. Wurtz JM, Bourguet W, Renaud JP, Vivat V, Chambon P, Moras D, Gronemeyer H 1996 A canonical structure for the ligand-binding domain of nuclear receptors. Nat Struct Biol 3:87–94. 148. Whitfield GK, Hsieh JC, Nakajima S, MacDonald PN, Thompson PD, Jurutka PW, Haussler CA, Haussler MR 1995 A highly conserved region in the hormone-binding domain of the human vitamin D receptor contains residues vital for heterodimerization with retinoid X receptor and for transcriptional activation. Mol Endocrinol 9:1166–1179.

1236 149. Malloy PJ, Xu R, Peng L, Clark PA, Feldman D 2002 A novel mutation in helix 12 of the vitamin D receptor impairs coactivator interaction and causes hereditary 1,25-dihydroxyvitamin D–resistant rickets without alopecia. Mol Endocrinol 16:2538–2546. 150. Miller J, Djabali K, Chen T, Liu Y, Ioffreda M, Lyle S, Christiano AM, Holick M, Cotsarelis G 2001 Atrichia caused by mutations in the vitamin D receptor gene is a phenocopy of generalized atrichia caused by mutations in the hairless gene. J Invest Dermatol 117:612–617. 151. Hewison M, Rut AR, Kristjansson K, Walker RE, Dillon MJ, Hughes MR, O’Riordan JL 1993 Tissue resistance to 1,25dihydroxyvitamin D without a mutation of the vitamin D receptor gene. Clin Endocrinol 39:663–670. 152. Chen H, Hewison M, Hu B, Adams JS 2003 Heterogeneous nuclear ribonucleoprotein (hnRNP) binding to hormone response elements: A cause of vitamin D resistance. Proc Natl Acad Sci USA 100:6109–6114. 153. Chrousos GP, Renquist D, Brandon D, Eil C, Pugeat M, Vigersky R, Cutler GB, Jr., Loriaux DL, Lipsett MB 1982 Glucocorticoid hormone resistance during primate evolution: receptor-mediated mechanisms. Proc Natl Acad Sci USA 79:2036–2040. 154. Chrousos GP, Loriaux DL, Brandon D, Shull J, Renquist D, Hogan W, Tomita M, Lipsett MB 1984 Adaptation of the mineralocorticoid target tissues to the high circulating cortisol and progesterone plasma levels in the squirrel monkey. Endocrinology 115:25–32. 155. Chrousos GP, Brandon D, Renquist DM, Tomita M, Johnson E, Loriaux DL, Lipsett MB 1984 Uterine estrogen and progesterone receptors in an estrogen- and progesterone“resistant” primate. J Clin Endocrinol Metab 58:516–520. 156. Adams JS, Gacad MA, Baker AJ, Kheun G, Rude RK 1985 Diminished internalization and action of 1,25-dihydroxyvitamin D3 in dermal fibroblasts cultured from New World primates. Endocrinology 116:2523–2527. 157. Takahashi N, Suda S, Shinki T, Horiuchi N, Shiina Y, Tanioka Y, Koizumi H, Suda T 1985 The mechanism of end-organ resistance to 1 alpha,25-dihydroxycholecalciferol in the common marmoset. Biochem J 227:555–563. 158. Reynolds PD, Ruan Y, Smith DF, Scammell JG 1999 Glucocorticoid resistance in the squirrel monkey is associated with overexpression of the immunophilin FKBP51. J Clin Endocrinol Metab 84:663–669. 159. Giraldo A, Pino W, Garcia-Ramirez LF, Pineda M, Iglesias A 1995 Vitamin D–dependent rickets type II and normal vitamin D receptor cDNA sequence. A cluster in a rural area of Cauca, Colombia, with more than 200 affected children. Clin Genet 48:57–65. 160. Tsuchiya Y, Matsuo N, Cho H, Kumagai M, Yasaka A, Suda T, Orimo H, Shiraki M 1980 An unusual form of vitamin D– dependent rickets in a child: alopecia and marked end-organ hyposensitivity to biologically active vitamin D. J Clin Endocrinol Metab 51:685–690. 161. Fujita T, Nomura M, Okajima S, Furuya H 1980 Adult-onset vitamin D–resistant osteomalacia with the unresponsiveness to parathyroid hormone. J Clin Endocrinol Metab 50: 927–931. 162. Kruse K, Feldmann E 1995 Healing of rickets during vitamin D therapy despite defective vitamin D receptors in two siblings with vitamin D–dependent rickets type II. J Pediatr 126:145–148. 163. Balsan S, Garabedian M, Larchet M, Gorski AM, Cournot G, Tau C, Bourdeau A, Silve C, Ricour C 1986 Long-term

PETER J. MALLOY, J. WESLEY PIKE, AND DAVID FELDMAN

164.

165. 166.

167.

168.

169.

170.

171.

172.

173.

174.

175.

176.

177.

nocturnal calcium infusions can cure rickets and promote normal mineralization in hereditary resistance to 1,25-dihydroxyvitamin D. J Clin Invest 77:1661–1667. Weisman Y, Bab I, Gazit D, Spirer Z, Jaffe M, Hochberg Z 1987 Long-term intracaval calcium infusion therapy in end-organ resistance to 1,25-dihydroxyvitamin D. Am J Med 83:984–990. Hochberg Z, Tiosano D, Even L 1992 Calcium therapy for calcitriol-resistant rickets. J Pediatr 121:803–808. Weisman Y, Jaccard N, Legum C, Spirer Z, Yedwab G, Even L, Edelstein S, Kaye AM, Hochberg Z 1990 Prenatal diagnosis of vitamin D–dependent rickets, type II: response to 1,25dihydroxyvitamin D in amniotic fluid cells and fetal tissues. J Clin Endocrinol Metab 71:937–943. Weisman Y, Malloy PJ, Krishnan AV, Jaccard N, Feldman D, Hochberg Z Prenatal diagnosis of calcitriol resistant rickets (CRR) by 1,25(OH)2D3 binding, 24-hydroxylase induction and RFLP analysis. Ninth workshop on Vitamin D, Orlando, 1994, p 106. Gardezi SA, Nguyen C, Malloy PJ, Posner GH, Feldman D, Peleg S 2001 A rationale for treatment of hereditary vitamin D– resistant rickets with analogs of 1 alpha,25-dihydroxyvitamin D3. J Biol Chem 276:29148–29156. Swann SL, Bergh J, Farach-Carson MC, Ocasio CA, Koh JT 2002 Structure-based design of selective agonists for a rickets-associated mutant of the vitamin D receptor. J Am Chem Soc 124:13795–13805. Swann SL, Bergh JJ, Farach-Carson MC, Koh JT 2002 Rational design of vitamin D3 analogues which selectively restore activity to a vitamin D receptor mutant associated with rickets. Org Lett 4:3863–3866. Kittaka A, Kurihara M, Peleg S, Suhara Y, Takayama H 2003 2alpha-(3-Hydroxypropyl)- and 2alpha-(3-Hydroxypropoxy)1alpha,25-dihydroxyvitamin D(3) Accessible to vitamin D receptor mutant related to hereditary vitamin D–resistant rickets. Chem Pharm Bull (Tokyo) 51:357–358. Panda DK, Miao D, Tremblay ML, Sirois J, Farookhi R, Hendy GN, Goltzman D 2001 Targeted ablation of the 25-hydroxyvitamin D 1alpha-hydroxylase enzyme: evidence for skeletal, reproductive, and immune dysfunction. Proc Natl Acad Sci USA 98:7498–7503. Chen CH, Sakai Y, Demay MB 2001 Targeting expression of the human vitamin D receptor to the keratinocytes of vitamin D receptor null mice prevents alopecia. Endocrinology 142:5386–5389. Li M, Chiba H, Warot X, Messaddeq N, Gerard C, Chambon P, Metzger D 2001 RXR-alpha ablation in skin keratinocytes results in alopecia and epidermal alterations. Development 128:675–688. Ahmad W, Faiyaz ul Haque M, Brancolini V, Tsou HC, ul Haque S, Lam H, Aita VM, Owen J, deBlaquiere M, Frank J, Cserhalmi-Friedman PB, Leask A, McGrath JA, Peacocke M, Ahmad M, Ott J, Christiano AM 1998 Alopecia universalis associated with a mutation in the human hairless gene. Science 279:720–724. Cichon S, Anker M, Vogt IR, Rohleder H, Putzstuck M, Hillmer A, Farooq SA, Al-Dhafri KS, Ahmad M, Haque S, Rietschel M, Propping P, Kruse R, Nothen MM 1998 Cloning, genomic organization, alternative transcripts, and mutational analysis of the gene responsible for autosomal recessive universal congenital alopecia. Hum Mol Genet 7:1671–1679. Hsieh JC, Sisk JM, Jurutka PW, Haussler CA, Slater SA, Haussler MR, Thompson CC 2003 Physical and functional

CHAPTER 72 Hereditary 1,25-Dihydroxyvitamin D–Resistant Rickets

178.

179.

180.

181. 182.

interaction between the vitamin D receptor and hairless corepressor, two proteins required for hair cycling. J Biol Chem 7:7. Ikeda K, Lu C, Weir EC, Mangin M, Broadus AE 1989 Transcriptional regulation of the parathyroid hormonerelated peptide gene by glucocorticoids and vitamin D in a human C-cell line. J Biol Chem 264:15743–15746. Wysolmerski JJ, Broadus AE, Zhou J, Fuchs E, Milstone LM, Philbrick WM 1994 Overexpression of parathyroid hormonerelated protein in the skin of transgenic mice interferes with hair follicle development. Proc Natl Acad Sci USA 91:1133–1137. Cho YM, Woodard GL, Dunbar M, Gocken T, Jimenez JA, Foley J 2003 Hair-cycle–dependent expression of parathyroid hormone-related protein and its type I receptor: evidence for regulation at the anagen to catagen transition. J Invest Dermatol 120:715–727. Haussler MR, McCain TA 1977 Basic and clinical concepts related to vitamin D metabolism and action. N Engl J Med 297:1041–1050. DeLuca HF 1979 The vitamin D system in the regulation of calcium and phosphorus metabolism. Nutr Rev 37:161–193.

1237

183. Clemens TL, Garrett KP, Zhou XY, Pike JW, Haussler MR, Dempster DW 1988 Immunocytochemical localization of the 1,25-dihydroxyvitamin D3 receptor in target cells. Endocrinology 122:1224–1230. 184. Skowronski R, Peehl D, Feldman D 1993 Vitamin D and prostate cancer: 1,25-dihydroxyvitamin D3 receptors and actions in prostate cancer cell lines. Endocrinology 132: 1952–1960. 185. Panda DK, Miao D, Bolivar I, Li J, Huo R, Hendy GN, Goltzman D 2004 Inactivation of the 25-hydroxyvitamin D 1 alpha-hydroxylase and vitamin D receptor demonstrates independent and interdependent effects of calcium and vitamin D on skeletal and mineral homeostasis. J Biol Chem 279:16754–16766. 186. Underwood JL, DeLuca HF 1984 Vitamin D is not directly necessary for bone growth and mineralization. Am J Physiol 246:E493–498. 187. Miyamoto K, Kesterson RA, Yamamoto H, Taketani Y, Nishiwaki E, Tatsumi S, Inoue Y, Morita K, Takeda E, Pike JW 1997 Structural organization of the human vitamin D receptor chromosomal gene and its promoter. Mol Endocrinol 11:1165–1179.

CHAPTER 73

Glucocorticoids and Vitamin D PHILIP SAMBROOK

Institute of Bone & Joint Research, University of Sydney, Sydney, Australia

I. Introduction II. Steroid Receptors and Actions in Bone III. Effect of Glucocorticoids on Vitamin D Metabolism

IV. Vitamin D As a Treatment for GIO V. Summary References

I. INTRODUCTION

II. STEROID RECEPTORS AND ACTIONS IN BONE

This chapter gives a general overview of the current understanding of glucocorticoid induced osteoporosis (GIO) with special reference to vitamin D metabolism. This chapter also includes a review of steroid hormone receptor expression in bone cells with specific reference to glucocorticoid and vitamin D hormones and their interaction. Clinical aspects of GIO and the role of vitamin D metabolites as therapeutic agents in GIO are also discussed. Bone loss resulting from long-term glucocorticoid therapy is a common clinical problem. A number of different glucocorticoid mediated effects are responsible for GIO, including 1) direct inhibitory effects of glucocorticoids upon osteoblast, osteocyte, and osteoclast function leading to reduced bone remodeling and diminished repair of microdamage in bone as well as enhanced osteoblast and osteocyte apoptosis; 2) antagonism by glucocorticoids of gonadal function and inhibition of the osteoanabolic action of sex steroids; 3) increased renal excretion and reduced intestinal absorption of calcium leading to negative calcium balance that can promote secondary hyperparathyroidism; and 4) potential effects on vitamin D metabolism [1–3]. From a mechanistic point of view, all of these effects have long been considered to be mediated at the molecular level exclusively by genomic actions. However, there is now increasing interest in the existence of rapid glucocorticoid effects that are incompatible with this classical mode of action [1]. These rapid effects, termed nongenomic effects, appear to be mediated by glucocorticoid interactions with biological membranes, either through binding to membrane receptors or by physicochemical interactions [1]. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

A. Receptors The skeleton is a complex tissue and hormonal control of bone remodeling is elaborate (see Chapter 28). The important role that steroid hormones play in bone cell development and in the maintenance of normal bone architecture is well established, but it is only relatively recently that it has become possible to describe the precise mechanism of action of steroid hormones including glucocorticoids and related hormones such as vitamin D, all of which act via structurally homologous nuclear receptors that form part of the steroid/ thyroid receptor superfamily (described in Section II of this book). The action of all these hormones are mediated by hormone binding to these nuclear receptors, which act as ligand-dependent transcription factors to either activate or repress target gene expression [4]. Thus, for glucocorticoids, it is currently believed that most of their biological activities are mediated via binding to the glucocorticoid receptor (GR). By this classic genomic mechanism, lipophilic glucocorticoid passes across the cell membrane, attaches to the cytosolic GR and after dimerization, the GR binds to conserved sequence motifs (glucocorticoid response elements or GREs) to positively or negatively regulate specific gene transcription [5]. However, it is also now recognized that some biological activities of glucocorticoids may be mediated via other transcription factors, such as AP-1 and NFkB, independent of GR binding to DNA but dependent upon interaction with these factors [5]. Similarly, the diverse biological activities of vitamin D which include effects on bone, cancer, and immune cells are mediated by binding of the ligand, calcitriol, Copyright © 2005, Elsevier, Inc. All rights reserved.

1240 or 1,25-dihydroxyvitamin D3 (1,25(OH)2D) to the ubiquitously expressed vitamin D receptor (VDR). The VDR functions as a heterodimer with retinoid X receptors (RXR) to bind vitamin D response elements (VDREs) within gene promoters and interact with a number of cofactors that are critical for transcriptional activation. It appears that both GR and VDR share a large number of coactivators and corepressors, raising the possibility that shared components for certain functions could be a limiting effect. In humans, alternative splicing of GR mRNA produces two similar isoforms, GRα and GRβ. However GRβ does not bind ligand and, although it may act as a dominant negative regulator of GRα activity, most studies consequently have focused on the GRα isoform [4]. In terms of glucocorticoid effects on bone cells, cytosolic binding studies have shown specific glucocorticoid binding sites predominantly in cells of osteoblastic lineage. This glucocorticoid binding occurs in animal and human osteoblasts and human osteosarcoma cells [6–14]. The effects of glucocorticoid on bone cells appear to be species specific. For example, in rats glucocorticoids increase differentiation of osteoblasts and osteoblasts progenitors [15–17]. The increase in osteoblast differentiation is associated with induction of osteoblast marker genes such as alkaline phosphatase, osteocalcin, osteopontin, and bone sialoprotein [18–20]. In mice, however, the principal effect of glucocorticoids is to stimulate bone resorption and osteoclast formation [21,22]. In osteoblast cultures in human cells, glucocorticoids appear necessary for osteoblast differentiation and may induce some osteoblast marker genes such as alkaline phosphatase [22–25], while inhibiting others such as osteocalcin [22–25], although these effects may differ depending upon the age and stage of differentiation of the cultured cells [25,26].

B. Calcitriol Vitamin D also has predominant effects on osteoblasts and modulates the expression of a number of osteoblastic genes including alkaline phosphatase, osteopontin, and osteocalcin [27–32], as well as inhibiting collagen synthesis [33–35]. However, the osteoblastic response to 1,25(OH)2D appears to vary in relation to the stage of osteoblast development [26] (see Chapters 32 and 37). Since osteocalcin, a protein produced by osteoblasts, is induced by 1,25(OH)2D (see Chapter 41) but suppressed by glucocorticoids, study of this gene has provided many insights into the interactions of vitamin D and glucocorticoids at the genomic level. For example, transcription of the rat osteocalcin gene is controlled

PHILIP SAMBROOK

by basal and hormone response elements located in proximal and distal sites, respectively [36]. 1,25(OH)2D acts on the osteocalcin gene via the distal VDRE within the promoter at nucleotides −465 to −437. This VDRE functions as an enhancer and cannot induce transcription but requires basal expression. Morrison et al. [37] have shown that glucocorticoids repress both 1,25(OH)2D induction and basal activity of the osteocalcin promoter through a region distinct from the VDRE. Indeed, the rat and human osteocalcin promoters contain multiple GREs [38–41]. In the rat osteocalcin gene, GREs in the distal (nucleotides −697 to −683) and proximal promoter regions (−16 to −1) bind GR and suppress 1,25(OH)2D induced transcription of osteocalcin [41]. Shalhoub et al. [20] have shown the influence of glucocorticoids on osteocalcin transcription is dependent on the stage of osteoblast maturation. In proliferating early stage osteoblasts, dexamethasone increased osteocalcin transcriptional rates, although not to the extent of vitamin D, and only in cells continuously treated with dexamethasone for long periods. In mature cells, acute treatment with dexamethasone resulted in decreased transcription and mRNA levels. The results indicate that transcriptional control of basal and hormone-regulated osteocalcin expression predominates in immature osteoblasts prior to matrix mineralization. However, in mature osteoblasts, osteocalcin expression was controlled primarily by post-transcriptional mechanisms reflected by elevated mRNA levels with a decline in transcription. Vitamin D, alone or in combination with dexamethasone, was a significant factor contributing to mRNA stabilization in mature osteoblasts with a mineralized extracellular matrix. Transcriptional modifications in response to dexamethasone were reflected by quantitative differences between proliferating and mature osteoblasts in the formation of GR binding complexes at the proximal osteocalcin GRE. Both VDR and GR basal mRNA levels were significantly higher in mature osteoblasts than in early stage bone cells. Dexamethasone significantly increased VDR transcription on day 7 (proliferation stage) and day 20 (differentiated osteoblasts) in fetal rat calvariae, but this resulted in a small increase in VDR mRNA accumulation on day 7 and a decrease on day 20. These results suggest there are developmental, stage-specific effects of steroid hormone on transcriptional regulation of bone-expressed genes, and inverse relationships between levels of transcription and cellular representation of mRNA with osteocalcin message stabilized in mature osteoblasts. Further studies by the same group [42] were consistent with selective influences of 1,25(OH)2D and glucocorticoids as a function of osteoblast maturation and species-specific responsiveness of mouse bone-expressed genes, to steroid

CHAPTER 73 Glucocorticoids and Vitamin D

hormones during osteoblast differentiation. These findings were in contrast to findings from other osteoblast culture systems. Binding of the VDR by 1,25(OH)2D induces conformational changes in the receptor that enable it to interact with several types of cofactors that are necessary for transcriptional activation (see Chapter 16). It has been shown that osteocalcin gene activation by 1,25(OH)2D at the VDRE region is accompanied by changes in chromatin structure and that such chromatin remodeling is a prerequisite for transcription [43]. Indeed, unlike the GR, the VDR-RXR appears unable to bind its target sequence until nucleosomal remodeling occurs, to allow occupancy of binding sites in the distal region of the osteocalcin gene promoter by the regulatory factors responsible for 1,25(OH)2D dependent enhancement of transcription. The extent to which such transcriptional coactivators are involved in modulating tissue-specific functions of the VDR is unclear. Recent studies by Issa et al. [44] have investigated the role of p160 coactivators in regulating VDR function and interaction with the heterodimeric partner of VDR, RXR. Two p160 coactivators, glucocorticoid receptor interacting protein-1 (GRIP1) and receptor associated coactivator-3 (RAC3) [45,46], appear to interact directly with the VDR, but only in the presence of the ligand. Deletional analyses of VDR suggest that GRIP1 and RAC3 required an intact VDR activation function domain (AF-2) for efficient interaction. Co-expression studies indicated that both GRIP1 and RAC3 co-assemble with the VDR to form an active transcriptional complex. They also form ternary complexes with VDR homodimers and VDR: RXRα heterodimers. Consistent with a role in modulating VDR function in bone, GRIP1 potentiated transactivation of the osteocalcin promoter, whereas RAC3 enhanced VDR activation indirectly through RXR. These data suggest different p160 coactivators regulate VDR function via different mechanisms and that the VDR recruits different coactivators depending on specific gene and cellular contexts.

III. EFFECT OF GLUCOCORTICOIDS ON VITAMIN D METABOLISM The metabolism of serum 25-hydroxy vitamin D (25OHD) occurs mainly in the kidney where it is converted to 1,25(OH)2D by the enzyme 1-α-hydroxylase and to 24,25-dihydroxyvitamin D by the enzyme vitamin D-24-hydroxylase (see Chapters 5 and 6). It has been suggested that chronic glucocorticoid therapy can increase renal expression of vitamin D-24-hydroxylase and decrease expression or renal 1-α-hydroxylase [47].

1241 Kurahashi et al. [48] investigated the mechanisms of this increase in UMR-106 osteoblast-like cells and found dexamethasone dose dependently increased 24-hydroxylase mRNA expression and enzymatic activity in the presence of 1,25(OH)2D. The mechanism of this effect appeared to involve activation of the AP-1 site by increased c-fos protein. However, the significance of this observation is uncertain since changes in circulating vitamin D metabolites with glucocorticoid therapy are not clearly established. As noted above, although multiple different glucocorticoid mediated effects are responsible for GIO, consistent findings from many studies of increased renal calcium excretion and reduced intestinal absorption of calcium, which could lead to secondary hyperparathyroidism, have lead to numerous studies to look for alterations in circulating vitamin D metabolites in patients receiving exogenous glucocorticoids or with Cushing’s syndrome. However, just as the published data looking for increased serum PTH concentrations with glucocorticoids are conflicting, there is even less evidence that changes in vitamin D metabolism are involved in the pathophysiology of GIO. At the same time it must be acknowledged that interpretation of the published data is difficult because many of the studies have been cross-sectional in nature with small samples, although some prospective studies have been performed (Table I). Long-term excess glucocorticoids have been reported to produce varied effects on vitamin D metabolites such as 25OHD or 1,25(OH)2D, including reductions [49–51], no change [52–55], or small increases [56–59]. Chesney et al. [49] reported reductions in serum 1,25(OH)2D in 22 children receiving long-term glucocorticoid treatment for various glomerular diseases, including nephrotic syndrome (mean + SD serum 1,25(OH)2D: 20 + 4 pg/ml vs. 53 + 5 pg/ml in controls, p < 0.005). Moreover, the reduction in concentration of serum 1,25(OH)2D correlated with the dose of corticosteroid administered and with reduction in forearm bone mineral content. In contrast, in 10 children with chronic glomerulonephritis not treated with glucocorticoids, who had similar serum creatinine to those children treated with glucocorticoids, serum 1,25(OH)2D concentrations were similar to those in 18 healthy controls, indicating that glomerular renal disease per se did not account for the observed reduction. In a subsequent study, Chesney et al. [50] measured vitamin D metabolites in 8 children with chronic glomerulonephritis not treated with prednisone (group I), 9 nonedematous children with nephrotic syndrome treated with prednisone for more than 18 months (group 2), and in 5 children with nephrotic edema also treated with prednisone (group 3). Reductions in

1242

PHILIP SAMBROOK

TABLE I Studies of Serum Vitamin D Metabolites in Glucocorticoid-Induced Osteoporosis Findings Reference Chesney [49] Chesney [50] Klein [61] Seeman [52] Slovik [53] Cannigia [62] Hahn [56] Hahn [60] Findling [55] Morris [51] Prummel [52] Cosman [59]

Type

Disease population

Sample size

25(OH)D

1,25(OH)2D

Cross-sectional Cross-sectional Cross-sectional Cross-sectional Cross-sectional Cross-sectional Prospective Prospective Prospective Cross-sectional Prospective Prospective

Renal, pediatric Renal, pediatric Rheumatic Mixed Asthma Not stated Normal Rheumatic Cushing’s Adult, miscellaneous Graves disease Multiple sclerosis

22 21 27 14 48 15 12 17 7 60 10 56

N L if edema L L L-N L-N N N N N N

L L – N – – small H – N L-N N H

L = low; N = normal; H = high or elevated.

serum calcium, albumin, and 25(OH)D were found in group 3 only, whereas both group 2 and group 3 patients showed reduced values of 1,25(OH)2D (p < 0.001 vs. group I or controls). It was concluded that chronic glucocorticoid administration in children with glomerulonephritis and minimally impaired renal function (group 2) was associated with a reduction in the circulating level of 1,25(OH)2D, since children with comparable type and degree of renal disease but nonglucocorticoid treatment (group I) had normal 1,25(OH)2D values. Children with nephrotic edema (group 3) had greater reduction of 1,25(OH)2D values, as well as lower 25(OH)D values and serum calcium values, considered possibly related to a urinary loss of vitamin D–binding protein. No changes in PTH were evident in either glucocorticoid-treated or edematous patients, suggesting that the acute elevation in PTH seen after prednisone treatment is an acute phenomenon. Morris et al. [51] examined serum 1,25(OH)2D levels in 60 postmenopausal women on glucocorticoid therapy (29 with and 31 without vertebral compression fractures), and they compared these results with those from 31 normal age-matched postmenopausal women. Serum 1,25(OH)2D levels were slightly reduced in both glucocorticoid treated groups (mean + SE: 80 + 8.4 and 92 + 7.9 pmol/l respectively) compared to normal subjects (107 + 7.3 pmol/l), but these differences were not significant. Seeman et al. [52] studied circulating levels of vitamin D metabolites in 6 patients with endogenous Cushing’s syndrome and 8 patients treated with

prednisone (mean dose 50 mg/day, range 30–60 mg/day) for one month for various connective tissue diseases. Comparing “euglucocorticoid” levels (i.e., after surgical correction in the Cushing’s patients or before treatment in the prednisone therapy group) with the hyperglucocorticoid state (i.e., before surgery or after prednisone therapy, respectively), they observed a nonsignificant reduction in serum 1,25(OH)2D (mean + SD: 32 + 8 vs. 23 + 6 pg/ml), but significantly lower serum 25(OH)D (22 + 2 vs. 18 + 2 ng/ml) in the hyperglucocorticoid state. Kinetic studies using tritiated 1,25(OH)2D in 10 hyperglucocorticoid patients and 14 normal controls revealed no evidence for altered production or degradation. Slovik et al. [53] observed that mean serum 25(OH)D and PTH levels in 48 adult asthmatic patients on chronic glucocorticoid therapy were not significantly different from a disease control group of 12 asthmatic patients not on glucocorticoids, but nine such patients had abnormally low 25(OH)D levels. Hahn et al. [56] studied 12 normal adults treated with prednisone 20 mg daily for 2 weeks. Serum 25(OH)D did not change significantly but serum 1,25(OH)2D rose significantly as did serum PTH, but the latter rise was not significant. In another prospective study of the effect of administration of 25(OH)D in GIO (discussed below), Hahn et al. [59] observed no difference in baseline serum 25(OH)D in 17 patients treated with chronic glucocorticoids (dose greater than 7.5 mg prednisone equivalent for at least 18 months) compared to 15 normal subjects.

1243

CHAPTER 73 Glucocorticoids and Vitamin D

IV. VITAMIN D AS A TREATMENT FOR GIO Even though the literature is controversial as to whether circulating metabolites of vitamin D are affected by glucocorticoid therapy, the original rationale for use of vitamin D in GIO was based upon the demonstration of impaired calcium absorption in glucocorticoid treated subjects and the assumption that this could be

(1–84)PTH (pmol/L)

6.0 5.5 5.0 4.5 4.0 3.5 3.0

1,25(OH)2D (pmol/L)

125 110 95 80 65 50

Phosphorus (mmol/L)

1.2 1.1 1.0 0.9 0.8 0.7 180 TRAP (nkat/L)

165 150 135 120 105

90 0.6 0.5 BGP (nmol/L)

Thus, cross-sectional studies have generally reported no change in vitamin D metabolites in response to glucocorticoid therapy. Findling et al. [55] studied vitamin D metabolites in 7 patients with spontaneous ACTH-dependent Cushing’s syndrome. Remission of hypercortisolism resulted in a significant increase in tubular reabsorption of phosphate and serum phosphate. Serum PTH levels were normal during Cushing’s syndrome, but fell significantly after remission. Plasma 25(OH)D and 1,25(OH)2D did not differ from measurements in 97 normal subjects. After treatment, serum 25(OH)D did not change, but serum 1,25(OH)2D fell (mean 44 to 22 pg/ml, p < 0.02) and was inversely correlated with serum phosphate (r = 0.59; p < 0.01), but did not correlate with serum PTH. It was concluded that impairment of intestinal calcium absorption in Cushing’s syndrome could not be attributed to any decrease in the circulating levels of 1,25(OH)2D. In a small study, Prummel et al. [54] prospectively measured biochemical markers of bone turnover in 10 euthyroid patients with Graves ophthalmopathy treated with tapering prednisone (initial dose 60 mg/day) over 19 weeks. There were no significant changes in serum 25(OH)D or 1,25(OH)2D, although serum intact PTH fell slightly. In another larger prospective study, Cosman et al. [59] studied various biochemical markers of bone turnover in a larger population of 56 patients with multiple sclerosis, also treated with tapering corticosteroids (initially intravenous methylprednisolone 1 g daily for 10 days, then 500 mg/day for 2 days, 250 mg/day for 2 days followed by oral prednisone 80 mg/day reducing in dose over 28 days). During and after treatment, there were no changes in serum 25(OH)D, but serum 1,25(OH)2D increased and serum phosphate decreased within 3 days of commencing corticosteroid. Serum PTH increased later to a peak at 2 weeks and serum 1,25(OH)2D subsequently declined (Fig. 1). The early changes in serum 1,25(OH)2D were considered to represent a direct effect of glucocorticoids on the kidney, rather than being secondary to the decrease in serum phosphate, with the later decline secondary to the change in PTH.

0.4 0.3 0.2 0.1 0.0

#

B

#

0.5

2

1

3

4–6

>10

Weeks during steroid administration 1.0

IV Solumedrol (g)

FIGURE 1

.50 .25

80

60 40 2010 5

Oral prednisone (mg)

Effects of intravenous followed by oral glucocorticoid over 10 weeks on serum indices of mineral and skeletal metabolism. Reproduced from J Bone Miner Res 1994, 9: 1097–1105 with permission of the American Society for Bone and Mineral Research.

1244 reversed by vitamin D [60–63]. Therapeutic studies of vitamin D have examined both changes in calcium absorption and bone mass or density. Some of the earlier studies have been performed with simple vitamin D or calciferols; however, the active metabolites, calcitriol and alfacalcidol (1α-hydroxy vitamin D), have also been studied. Recent interest has also focused on the nonskeletal benefits of vitamin D (such as reducing the risk of falls; see Chapter 102) as a further mechanism of reduced fracture risk, although this has not been specifically studied in GIO.

A. Calcium Absorption Klein et al. [61] compared fractional calcium absorption in 27 patients receiving prednisone compared to 27 age- and sex-matched controls. In patients receiving high doses of prednisone (15–100 mg/day) calcium absorption and serum 25OHD were decreased. However, in patients receiving low doses (8–10 mg/day), or high doses (30–100 mg) on an alternate day schedule, both of these parameters were normal. Calcium absorption correlated inversely with daily prednisone dosage. Administration of 0.4 µg of calcitriol daily for 7 days in 5 patients led to an increase in calcium absorption. A study by Hahn et al. [60] examined the effect of treatment with calcidiol (25 hydroxyvitamin D) 40 µg/day plus calcium in 17 patients with GIO compared to 15 controls. The glucocorticoid group had reduced calcium absorption and increased serum PTH, but similar serum 25(OH)D levels and reduced forearm bone mass compared to controls. Treatment increased calcium absorption by 46%. Caniggia et al. [62] reported that low to normal serum 25(OH)D in patients treated with chronic glucocorticoids and impaired calcium absorption could be reversed by treatment with either calcidiol or calcitriol. Braun et al. [64] performed a double-blind placebo controlled study of alfacalcidol 2 µg daily for 6 months in 14 patients receiving long-term glucocorticoid therapy. Treatment with alfacalcidol increased calcium absorption at 3 and 6 months and, on repeat iliac crest biopsy, showed a decrease in active resorption and a trend for increased trabecular bone volume, suggesting no suppression of bone formation was occurring. Morris et al. [51] examined the relation between calcium absorption and serum 1,25(OH)2D levels in a set of 60 postmenopausal women on glucocorticoid therapy (29 with and 31 without vertebral compression fractures) and compared these results with those from 31 normal postmenopausal women age-matched women. Calcium absorption was reduced in glucocorticoid treated patients and shown to be linearly related

PHILIP SAMBROOK

to serum 1,25(OH)2D in both glucocorticoid treated groups and in the glucocorticoid set as a whole. However only about one-third of the impairment in calcium absorption was accounted for by serum 1,25(OH)2D levels. This apparent resistance to the intestinal action of 1,25(OH)2D was quantified by a Z score, which expressed the difference between the measured calcium absorption and that predicted from the 1,25(OH)2D level. The calcium absorption was significantly reduced in the fracture group by –0.52 standard deviation. Despite the consistency in findings about impaired calcium absorption, its mechanism remains unclear. Studies of changes in intestinal 1,25(OH)2D receptor expression have yielded conflicting results. Dexamethasone administration in male rats caused a reduction in receptor number in jejunal villous cells; however, administration of prednisone in dogs increased duodenal concentrations of a 1,25(OH)2D specific binding protein [65,66]. Moreover, no changes in plasma 1,25(OH)2D or in renal hydroxylase mRNA abundance or enzyme activity occur in response to dexamethasone [47]. Although the demonstration of impaired calcium absorption with glucocorticoids provided a rationale for use of vitamin D in GIO, other data from animal and human studies also suggested a role for vitamin D as a therapeutic agent. In a rat model, when calcitriol was coadministered with glucocorticoids, suppression of serum osteocalcin was delayed, and effects of glucocorticoids on osteoid volume, tetracycline labeled surface, and trabecular bone volume, assessed histomorphometrically, were prevented [67]. In clinical studies, coadministration of calcitriol with prednisone produced less suppression of serum osteocalcin than administration of prednisone alone. Nielsen et al. [68] treated 7 healthy volunteers with prednisone—10 mg daily for 2 days with or without 2 µg calcitriol in a cross-over design. Prednisone inhibited and reversed the nocturnal rise in serum osteocalcin, whereas after calcitriol plus prednisone, the time course of serum osteocalcin almost paralleled the placebo (see Fig. 2). Lems et al. [69] examined the effect of low dose prednisone (10 mg/day) and the possible preventative effects of calcitriol on bone metabolism in 8 healthy, young male volunteers. The study consisted of four observation periods: in the first period, prednisone was administered for one week; in the second, third, and fourth periods, calcium (500 mg/day), calcitriol (0.5 µg bd), and calcium in combination with calcitriol respectively, were added to prednisone. Treatment with prednisone alone led to a decrease in osteocalcin and a (nonsignificant) increase in PTH, but PTH decreased during cotreatment with calcitriol (−16%) and calcium plus calcitriol (−44%; p < 0.01). It was concluded that the increase in PTH during prednisone

1245

CHAPTER 73 Glucocorticoids and Vitamin D

A 30

Serum osteocalcin (ng/ml)

26

22

18

14

10 2000 h

0400 h

1200 h

(time, clock hour)

B 8

∆Serum osteocalcin (ng/ml)

6

4 2 0 −2 −4 −6

2000 h

0400 h

1200 h

(time, clock hour)

FIGURE 2 Mean serum osteocalcin levels against time in five normal subjects (A) mean curves using raw data; (B) mean smoothed (moving average technique) data transformed by subtracting the 2000 h values of each day from all subsequent times: (❍) placebo; (•) 2 µg 1,25(OH)2D3 + placebo; (∆) 2 µg 1,25(OH)2D3 + 10 mg prednisone; and (▲) 10 mg prednisone. Reproduced from J Bone Miner Res 1991, 5:435–441 with permission of the American Society for Bone and Mineral Research.

could be prevented by taking calcitriol combined with calcium supplementation.

B. Bone Mass These observations led inevitably to studies of the effects of vitamin D on bone mass. Vitamin D in

various formulations, usually in combination with calcium, has been studied as a treatment for GIO in studies dating from the 1970s. These studies can be broadly divided into primary prevention (in patients commencing glucocorticoid therapy) or treatment (in patients on chronic glucocorticoids) (Table II). Hahn et al. [70] examined the effect of treatment with calcium 500 mg/day and vitamin D 50,000 units per week on bone mass in patients on chronic glucocorticoid therapy. A significant increase in forearm bone mass was observed with vitamin D treatment, but the study was not randomized. Since bone mass was only assessed in the radius (as the measurement of spinal bone density was not available at that time), the clinical relevance of these findings to other skeletal sites is unclear. Another study by Hahn et al. [60] examined the effect of treatment with calcidiol (40 µg/day) plus calcium in 17 patients receiving chronic glucocorticoids compared to 15 controls. Treatment with vitamin D improved bone mass over 12 months by 13.2 % at the metaphyseal site and 2.1% at the diaphyseal site. Bijlsma et al. [71] studied the effect of calcium 500 mg daily versus calcium plus a vitamin D preparation (dihydrotachysterol 4000 IU on alternate days) in 21 patients on long-term glucocorticoid therapy (mean daily prednisone dose 14 mg/day) for 2 years. A small increase in lumbar spine BMD was noted in both groups, but there was no significant difference between groups. Adachi et al. [72] compared combination calcium plus vitamin D (1000 mg daily plus 50,000 units weekly, respectively) against placebo over 3 years in 62 patients initiating corticosteroids (i.e., primary prevention). Bone loss at the lumbar spine was reduced by treatment with calcium/vitamin D, but the difference was not significantly different from the placebo (Fig. 3). However, a secondary prevention study by Buckley et al. [73], in 65 patients receiving chronic low-dose corticosteroids for rheumatoid arthritis observed an annual spinal loss of 2.0% in placebo-treated patients compared to 0.7% gain in calcium/vitamin D3-treated patients (1000 mg plus 500 IU/day, respectively). As the patients were receiving a chronic low dose of corticosteroids, the BMD rise may have been explicable as a “remodeling transient”, and the results may not necessarily be applicable to patients commencing corticosteroids (i.e., primary prevention) or those treated with higher doses. Sambrook et al. [74] examined the effect of 12 months of calcium, calcitriol, or calcitonin in 103 patients starting corticosteroids. Patients treated with calcium lost bone rapidly at the lumbar spine (−4.3% in the first year), whereas patients treated with either calcitriol alone (mean dose 0.6 µg/day) or calcitriol

1246

PHILIP SAMBROOK

TABLE II

Trials of Vitamin D Metabolites on Bone Mineral Density BMD effect size (% change vs. control)*

Reference

Prevention/ treatment

Type

Hahn [70] Hahn [60] Dykman [78] Biljsma [71] Sambrook [74] Adachi [72] Buckley [73] Reginster [75] Lambrindouki [79] Ringe [80] Sambrook [82] Ringe [81]

Nonrandomized, open Nonrandomized, open Randomized,DB Nonrandomized, open Randomized,DB Randomized Randomized Randomized,DB Randomized,DB Nonrandomized, open Randomized, open Randomized, open

Treatment Treatment Treatment Treatment Prevention Prevention Treatment Prevention Treatment Treatment Mixed Treatment

Agent

Sample size

Vit D 50,000 IU/wk Calcidiol 40 µg/d Calcitriol 0.4 µg/d Vit D 2000 IU/d Calcitriol 0.75 µg/d Vit D 50,000 U/wk Vit D 500 IU/d Alfacalcidol 1 µg/d Calcitriol 0.5 µg/d Alfacalcidol 1 µg/d* Calcitriol 0.5 µg/d* Alfacalcidol 1 µg/d*

Forearm

Spine

4.4 5.8 NS** – – – – – – – – –

– – – 5.4 3.0 −0.69 2.65 6.06 1.6 2.0 −0.2 3.2

26 32 23 21 103 62 65 145 81 85 198 204

* vs simple vitamin D. Transplant studies not included as immunosuppressives other than glucocorticoids included. ** Small increase both groups, % difference not stated.

plus calcitonin lost at a much reduced rate (−1.3% and −0.2% per year, respectively (Fig. 4). Both groups were significantly different from the calcium group. Approximately 25% of patients developed mild hypercalcemia, probably related to coadministration of calcium, which settled with a reduction in the calcitriol dosage. A randomized double-blind controlled trial in 145 patients starting corticosteroids compared 1 µg/day

B

Intention–to–treat 4

4

0

0

−4

−4

−8

−8

−12

0

12

24 Months

FIGURE 3

36

−12

BMD L2–L4 % Change from baseline

BMD L2–L4 % Change from baseline

A

of alfacalcidol with calcium [75]. After 12 months, the change in spinal bone density with alfacalcidol was +0.4% compared to –5.7% with calcium. Hypercalcemia occurred in only 6.7% of alfacalcidol-treated patients. A recent randomized, double-blind prospective trial found that prophylactic treatment with calcitriol in 66 patients undergoing cardiac or single lung transplantation was able to prevent or markedly reduce bone

Prednisone treated to 35 months 4

4

0

0

−4

−4

−8

−8

−12

0

12

24

36

−12

Months

Lumbar spine BMD, percentage change from baseline (A) intention to treat analysis (B) prednisone treated to 36 months. (A) Intention to treat analysis (B) prednisone treated to 36 months; ■ Placebo treatment; ▲ Vitamin D and calcium treatment group. Reproduced from J Rheumatol 1996, 23:995–1000 with permission.

1247

CHAPTER 73 Glucocorticoids and Vitamin D

1

Lumbar spine

0 −1 −2 −3 −4

Percent change

−5

1

Femoral neck

0 −1 −2 −3 −4

Distal radius 2 1 0 −1 −2 −3 −4

Year 1

Year 2 Group 1 Group 2 Group 3

FIGURE 4

Mean bone mineral density of the lumbar spine, femoral neck, and distal radius as the percent change per year in corticosteroid treated patients. Reproduced from New Engl J Med 1993, 328:1747–1752 with permission. Group 1 received calcitriol, calcitonin, and calcium; group 2, calcitriol and calcium; and group 3, calcium alone. After one year, with regard to bone loss in the lumbar spine, P = 0.0035 for the overall difference between groups; P = 0.0001 for the difference between groups 1 and 3; P = 0.026 for the difference between groups 2 and 3; P = 0.94 for the difference between groups 1 and 2. After two years, P = 0.044 for the overall difference between groups; P = 0.17 for the difference between groups 1 and 3, P = 0.94 for the difference between groups 2 and 3, and P = 0.014 for the difference between groups 1 and 2. The one-year results represented 92 patients, and the two-year results 64 patients.

loss over 24 months [76]. Similarly, Neuhaus et al. [77] have also reported that calcitriol was effective in increasing spine and femoral BMD in patients after liver transplantation. However, the mechanisms of post transplant bone loss are likely to reflect not just the effects of glucocorticoids but also other immunosuppressive agents, such as cyclosporine (see Chapter 74) and so the findings of these two studies should be interpreted in this context.

Not all studies with active metabolites have shown positive effects on bone density. Dykman et al. [78] studied 23 rheumatic disease patients with GIO in an 18-month double-blind, randomized study to assess the effect of oral calcium and calcitriol (mean dose 0.4 µg/day) or calcium versus placebo. Intestinal calcium absorption was increased and serum PTH levels were suppressed by calcitriol; however, no significant gain in forearm bone mass occurred, and fractures were as frequent in both groups. In the calcitriol group, histomorphometric analysis of iliac crest biopsy specimens demonstrated a decrease in osteoclasts/mm2 of trabecular bone (p < 0.05) and parameters of osteoblastic activity (p < 0.05), indicating that calcitriol reduced both bone resorption and formation. Lambrinoudaki et al. [79] studied the effect of calcitriol in GIO in a double-blind, placebo-controlled study of 81 premenopausal women with systemic lupus erythematosus on chronic steroid therapy (mean cumulative prednisone dose of 28 g). They were randomly allocated to three groups: Group 1: 0.5 µg calcitriol and 1200 mg calcium daily; Group 2: 1,200 mg calcium and placebo calcitriol; and Group 3: both placebo calcitriol and placebo calcium. At the end of two years, patients in the calcitriol group exhibited a significant increase of 2.1% in BMD at the lumbar spine compared to baseline value (p < 0.05). This change was not significantly different from the respective change in either calcium or placebo group (0.4% and 0.3%, respectively). No significant changes were observed in any treatment group in BMD at the hip or radius. It was concluded that premenopausal women with lupus taking prolonged glucocorticoid therapy had lower bone density, but showed no significant bone loss over the two-year study period. The beneficial effect of calcitriol treatment in these premenopausal women was small when it was instituted late in the course of glucocorticoid therapy.

C. Efficacy of Different Vitamin D Preparations Whether plain vitamin D is less efficacious than active metabolites in GIO is unclear, but three recent studies have addressed this question. Ringe et al. [80] evaluated the efficacy of alfacalcidol compared with simple vitamin D in patients on chronic corticosteroids. Eighty-five patients on long-term corticosteroid therapy were allocated to either 1 µg alfacalcidol plus calcium 500 mg daily or 1000 IU vitamin D3 plus 500 mg calcium in a nonrandomized parallel group study. The two groups were similar in age, sex, underlying diseases, initial BMD (lumbar spine: mean T-score −3.28 and −3.25, respectively), and rates of vertebral and

1248 nonvertebral fractures. During the three-year study, a small but significant increase was seen in lumbar spine BMD in the alfacalcidol group (+2.0%, p < 0.0001) with no significant changes at the femoral neck. In the vitamin D group, there were no significant changes at either site. More recently, Ringe et al. [81] have repeated this study with the same treatment regimen in a larger study sample and as a randomized open label trial. In 204 patients on chronic glucocorticoids over three years, significant increases in BMD were seen at lumbar spine (+ 2.4%) and the femoral neck (+1.2%) in the alfacalcidol group with no significant changes in the plain vitamin D group at either site (−0.7% and + 0.7%, respectively). By the end of the study, 16 new vertebral fractures had occurred in 10 patients of the alfacalcidol group and 25 in 25 patients of the vitamin D group. These studies suggest alfacalcidol is superior to simple vitamin D in the treatment of established glucocorticoids osteoporosis. In contrast, Sambrook [82] compared treatment with calcitriol with ergocalciferol or alendronate in 198 patients commencing or already being treated with chronic glucocorticoids. Patients were randomized to one of three groups: calcitriol 0.5 to 0.75 µg/day; simple vitamin D (ergocalciferol 30,000 IU weekly) plus calcium carbonate (600 mg daily); or alendronate 10 mg/day plus calcium carbonate (600 mg daily). Over two years, mean lumbar BMD change was +5.9% with alendronate, −0.5% with ergocalciferol, and –0.7% with calcitriol (p < 0.001). At the femoral neck, there was no significant difference in BMD change between the treatments over two years, alendronate (+0.9%), ergocalciferol (−2.2%), and calcitriol (−3.2%). The calcitriol group was treated with a higher cumulative glucocorticoid dose, but after adjustment, no significant difference was seen between calcitriol or ergocalciferol in the prevention of bone loss, but both were inferior to alendronate.

V. SUMMARY To summarize, evidence from all these randomized trials taken together suggests that patients receiving glucocorticoids, who are at risk of rapid bone loss and consequent fracture, should be actively considered for prophylactic measures that include a vitamin D metabolite [83]. However, based upon the available evidence (not reviewed here), first-line therapy would be a bisphosphonate, with vitamin D as adjunctive therapy or second-line therapy. Moreover, although vitamin D metabolites appear to confer an additional therapeutic effect in GIO, it remains unclear whether the active metabolites are superior to simple vitamin D in this context and further studies are required.

PHILIP SAMBROOK

References 1. Patschan D, Loddenkemper K, Buttgereit F 2001 Molecular mechanisms of glucocorticoid-induced osteoporosis. Bone 29:498–505. 2. Sambrook P, Lane NE 2001 Corticosteroid osteoporosis, Balliere’s best practice and research. Clin Rheumatol 15:3, 401–413. 3. Weinstein RS, Jilka RL, Parfitt AF, Manolagas SC 1998 Inhibition of osteoblastogenesis and promotion of apoptosis of osteoblasts and osteocytes by glucocorticoids. J Clin Invest 102:272–282. 4. Bland R 2000 Steroid hormone receptor expression and action in bone. Clin Science 98:217–240. 5. Karin M 1998 New twists in gene regulation by glucocorticoid receptor: is DNA binding dispensable? Cell 15:93, 487–490. 6. Chen TL, Aronow L, Feldman D 1997 Glucocorticoid receptors and inhibition of bone cell growth in primary culture. Endocrinology 100:619–628. 7. Manolagas SC, Anderson DC 1978 Detection of highaffinity glucocorticoid binding in rat bone. J Endocrinol 76: 379–380. 8. Chen TL, Feldman D 1979 Glucocorticoid receptors and actions in subpopulations of cultured rat bone cells. J Clin Invest 63:750–757. 9. Haussler MR, Manolagas SC, Deftos LJ 1980 Glucocorticoid receptor in clonal osteosarcoma cell lines: a novel system for investigating bone active hormones. Biochem Biophys Res Commun 94:373–380. 10. Masuyama A, Ouchi Y, Sato F, Hosoi T, Nakamura T, Orimo H 1992 Characteristics of steroid hormone receptors in cultured MC3T3-E1 osteoblastic cells and effect of steroid hormones on cell proliferation. Calcif Tissue Int 51:376–381. 11. Suzuki S, Koga Takaoka K, Ono K, Sato B 1993 Effects of retinoic acid on steroid and vitamin D3 receptors in cultured mouse osteosarcoma cells. Bone 14:7–12. 12. Liesegang P, Romalo G, Sudmann M, Wolf L, Schweikert HU 1994 Human osteoblast-like cells contain specific, saturable, high-affinity glucocorticoid, androgen, oestrogen, and 1alpha, 25-hydroxycholecalciferol receptors. J Androl 15:194–199. 13. Song LN 1994 Effects of retinoic acid and dexamethasone on proliferation, differentiation, and glucocorticoid receptor expression in cultured human osteosarcoma cells. Oncol Res 6:111–118. 14. Bland R, Worker CA, Noble BS, et al. 1999 Characterisation of 1 alpha-hydroxysteroid dehydrogenase activity and corticosteroid receptor expression in human osteosarcoma cell lines. J Endocrinol 161:455–464. 15. Bellows CG, Ciaccia A, Heersche JNM 1998 Osteoprogenitor cells in cell populations derived from mouse and rat calvaria differ in the response to corticosterone, cortisol and cortisone. Bone 23:119–125. 16. Boden SD, Hair G, Titus L, Racine M, McCuaig K, Wozney JM, Nanes MS 1997 Glucocorticoid-induced differentiation of fetal rat calvarial osteoblasts is mediated by bone morphogenetic protein 6. Endocrinol 138:2820–2828. 17. Ishida Y, Heersche JM 1997 Progesterone stimulates proliferation and differentiation of osteoprogenitor cells in bone cell populations derived from adult female but not from adult male rats. Bone 20:17–25. 18. Green E, Todd B, Heath D 1990 Mechanism of glucocorticoid regulation of alkaline phosphatase gene expression in osteoblast-like cells. Eur J Biochem 188:147–153.

1249

CHAPTER 73 Glucocorticoids and Vitamin D

19. Ogata T, Yamauchi M, Kim RH, Li JJ, Freedman LP, Sodek J 1995 Glucocorticoid regulation of bone sialoprotein (BSP) gene-expression—identification of a glucocorticoid response element in the bone sialoprotein gene promoter. Eur J Biochem 230:183–192. 20. Shalhoub V, Aslam F, Breen E, van Wijnen A, Bortell R, Stein GS, Stein JL, Lian JB 1998 Multiple levels of steroid hormone-dependent control of osteocalcin during osteoblast differentiation: glucocorticoid regulation of basal and vitamin D– stimulated gene expression. J Cell Biochem 69:154–68. 21. Conaway HH, Grigorie D, Lerner UH 1996 Stimulation of neonatal mouse calvarial bone resorption by the glucocorticoids hydrocortisone and dexamethasone. J Bone Miner Res 11:1419–1429. 22. Kaji H, Sugimoto T, Kanatani M, Nishiyama K, Chihara K 1997 Dexamethasone stimulates osteoclast-like cell formation by directly acting on hemopoietic blast cells and enhances osteoclast-like cell formation stimulated by parathyroid hormone and prostaglandin E2. J Bone Miner Res 12: 734–741. 23. Wong MM, Rao LG, Ly H, Hamilton L, Tong J, Sturtridge W, McBroom R, Aubin JE, Murray TM 1990 Long-term effects of physiologic concentrations of dexamethasone in human bonederived cells. J Bone Miner Res 5:803–813. 24. Subramaniam M, Colvard D, Keeting PE, Rasmussen K, Riggs BL, Spelsberg TC 1992 Glucocorticoid regulation of alkaline phosphatase, osteocalcin, and protooncogenes and normal human osteoblast-like cells. J Cell Biochem 50: 411–424. 25. Sutherl MS, Rao LG, Muzaffar SA, Wylie JN, Wong MM, McBroom RJ, Murray TM 1995 Age-dependent expression of osteoblastic phenotypic markers in normal human osteoblasts cultured long-term in the presence of dexamethasone. Osteop Inter 5:335–343. 26. Pockwinse SM, Stein JL, Lian JB, Stein GS 1995 Developmental stage-specific cellular responses to vitamin D glucocorticoids during differentiation of the osteoblast phenotype: interelationship of morphology and gene expression by in situ hybridization. Exp Cell Res 216:244–260. 27. Skjodt H, Gallagher JA, Beresford JN, Couch M, Poser JW, Russell RGG 1985 Vitamin D metabolites regulate osteocalcin synthesis and proliferation of human bone cells in vitro. J Endocrinol 105:391–396. 28. Demay MM, Gerardi JM, DeLuca HF, Kronenberg HM 1990 DNA sequences in the rat osteocalcin gene that bond the 1,25-dihydroxyvitamin D3 receptor and confer responsiveness to 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci (USA) 87: 369–373. 29. Williams GR, Bland R, Sheppard MC 1995 Retinoids modify regulation of endogenous gene expression by vitamin D3 and thyroid hormone in three osteosarcoma cell lines. Endocrinology 136:4304–4314. 30. Noda M, Vogel RL, Craig AM, Prahl J, DeLuca HF, Denhardt DT 1990 Identification of a DNA sequence responsible for binding of the 1,25-dihydroxyvitamin D3 receptor and 1,25-dihydroxyvitamin d3 enhancement of mouse secreted phosphoprotein 1 (spp-1 or osteopontin) gene expression. Proc Natl Acad Sci (USA) 87:9995–9999. 31. Pols HAP, Schilte HP, Herrmann-Erlee NMP, Visser TJ, Birkenhager JC 1986 The effect of 1,25-dihydroxyvitamin D3 on growth, alkaline phosphatase, and adenylate cyclase of rate osteoblast-like cells. Bone Miner 1:397–405. 32. Owen TA, Arnow MS, Barone LM, Bettencourt B, Stein GS, Lian JB 1991 Pleiotropic effects of vitamin D on osteoblast

33. 34. 35.

36.

37.

38.

39. 40.

41.

42.

43.

44.

45.

46.

gene expression are related to the proliferative and differentiated state of the bone cell phenotype: dependency upon basal levels of gene expression, duration of exposure, and bone matrix. Endocrinol 128:1496–1504. Rowe DW, Kream BE 1982 Regulation of collagen synthesis in rat calvaria by 1,25-dihydroxyvitamin D3. J Biol Chem 257:8009–8015. Kream BE, Rowe D, Smith MD, Maher V, Majeska R 1986 Hormonal regulation of collagen synthesis in a clonal rat osteosarcoma cell line. Endocrinol 119:1922–1928. Harrison JR, Peterson DN, Lichtler AC, Mador AT, Rowe DW, Kream BE 1989 1,25-dihydroxyvitamin D3-inhibits transcription of type 1 collagen genes in the rat osteosarcoma cell line ROS 17/2.8. Endocrinol 125:327–333. Montecino M, Pockwinse S, Lian J, Stein G, Stein J 1994 DNase I hypersensitive sites in promoter elements associated with basal and vitamin D–dependent transcription of the bonespecific osteocalcin gene. Biochem 33:348–353. Morrison NA, Shine J, Fragonas JC, Verkest V, McMenemy ML, Eisman JA 1989 1,25-dihydroxyvitamin D–responsive element and glucocorticoid repression in the osteocalcin gene. Science 246:1158–61. Stromstedt PE, Poellinger L, Gustafsson JA, Carlstedt-Duke J 1991 The glucocorticoid receptor binds to a sequence overlapping the TATA box of the human osteocalcin promoter: a potential mechanism for negative regulation. Mol Cell Biol 11(6): 3379–3383. Morrison N, Eisman J 1993 Role of the negative glucocorticoid regulatory element in glucocorticoid repression of the human osteocalcin promoter. J Bone Min Res 8(8):969–975. Aslam F, Shalhoub V, van Wijnen AJ, Banerjee C, Bortell R, Shakoori AR, Litwack G, Stein JL, Stein GS, Lian JB 1995 Contributions of distal and proximal promoter elements to glucocorticoid regulation of osteocalcin gene transcription. Molecular Endocrinology 9(6):679–90. Heinrichs AA, Bortell R, Rahman S, Stein JL, Alnemri ES, Litwack G, Lian JB, Stein GS 1993 Identification of multiple glucocorticoid receptor-binding sites in the rat osteocalcin gene promoter. Biochemistry 32(42):11436–11444. Lian JB, Shalhoub V, Aslam F, Frenkel B, Green J, Hamrah M, Stein GS, Stein JL 1997 Species-specific glucocorticoid and 1,25-dihydroxyvitamin D responsiveness in mouse MC3T3-E1 osteoblasts: dexamethasone inhibits osteoblast differentiation and vitamin D downregulates osteocalcin gene expression. Endocrinol 138: 2117–2127. Paredes R, Gutierrez J, Gutierrez S, Allison L, Puchi M, Imschenetzky M, van Wijnen A, Lian J, Stein G, Stein J, Montecino M 2002 Interaction of the 1alpha,25-dihydroxyvitamin D3 receptor at the distal promoter region of the bonespecific osteocalcin gene requires nucleosomal remodeling. Biochemical J 363:667–676. Issa LL, Leong GM, Barry JB, Sutherland RL, Eisman JA 2001 Glucocorticoid receptor-interacting protein-1 and receptorassociated coactivator-3 differentially interact with the vitamin D receptor (VDR) and regulate VDR-retinoid X receptor transcriptional cross-talk. Endocrinol 142:1606–1615. Hong H, Kohli K, Garabedian MJ, Stallcup MR 1997 GRIP1, a transcriptional coactivator for the AF-2 transactivation domain of steroid, thyroid, retinoid, and vitamin D receptors. Molec Cell Biol 17:2735–2744. Li H, Gomes PJ, Chen JD 1997 RAC3, a steroid/nuclear receptorassociated coactivator, that is related to SRC-1 and TIF2. Proc Natl Acad Sci USA 94:8479–8484.

1250 47. Akeno N, Matsunuma A, Maeda T, Kawane T, Horiuchi N 2000 Regulation of vitamin D-1alpha-hydroxylase and -24hydroxylase expression by dexamethasone in mouse kidney. J Endocrinol 164:339–348. 48. Kurahashil I, Matsunuma A, Kawane T, Abe M, Horiuchi N 2002 Dexamethasone enhances vitamin D-24-hydroxylase expression in osteoblastic (UMR-106) and renal (LLC-PK1) cells treated with 1alpha,25-dihydroxyvitamin D3. Endocrine 17:109–118. 49. Chesney RW, Hamstra A, Rose P, DeLuca HF 1984 Vitamin D and parathyroid hormone status in children with the nephrotic syndrome and chronic mild glomerulonephritis. Inter J Pediat Neph 5:1–4. 50. Chesney RW, Mazess RB, Hamstra AJ, DeLuca HF, O’Reagan S 1978 Reduction of serum-1, 25-dihydroxyvitamin-D3 in children receiving glucocorticoids. Lancet 2:1123–1125. 51. Morris HA, Need AG, O’Loughlin PD, Horowitz M, Bridges A, Nordin BE 1990 Malabsorption of calcium in corticosteroidinduced osteoporosis. Calcif Tiss Int 46:305–308. 52. Seeman E, Kumar R, Hunder GG, Scott M, Heath H, 3rd, Riggs BL 1980 Production, degradation, and circulating levels of 1,25-dihydroxyvitamin D in health and in chronic glucocorticoid excess. J Clin Invest 66:664–669. 53. Slovik DM, Neer RM, Ohman JL, Lowell FC, Clark MB, Segre GV, Potts JT Jr 1980 Parathyroid hormone and 25-hydroxyvitamin D levels in glucocorticoid-treated patients. Clin Endocrinol 12:243–248. 54. Prummel MF, Wiersinga WM, Lips P, Sanders GTB, Sauerwein HP 1991 The course of biochemical parameters of bone turnover during treatment with corticosteroids. J Clin Endocrinol Metab 72:382–367. 55. Findling JW, Adams ND, Lemann J Jr, Gray RW, Thomas CJ, Tyrrell JB 1982 Vitamin D metabolites and parathyroid hormone in Cushing’s syndrome: relationship to calcium and phosphorus homeostasis. J Clin Endoc Metab 54: 1039–1044. 56. Hahn TJ, Halstead LR, Baran DT 1981 Effects of short-term glucocorticoid administration on intestinal calcium absorption and circulating vitamin D metabolite concentrations in man. J Clin Endoc Metab 1:111–115. 57. Braun JJ, Juttmann JR, Visser TJ, Birkenhager JC 1982 Shortterm effect of prednisone on serum 1,25-dihydroxyvitamin D in normal individuals and in hyper- and hyperparathyroidism. Clin Endocrinol 17:21–28. 58. Bikle DD, Halloran B, Fong L, Steinbach L, Shellito J 1993 Elevated 1,25-dihydroxyvitamin D levels in patients with chronic obstructive pulmonary disease treated with prednisone. J Clin Endocr Metab 76:456–461. 59. Cosman F, Nieves J, Herbert J, Shen V, Lindsay R 1994 Highdose glucocorticoids in multiple sclerosis patients exert direct effects on the kidney and skeleton. J Bone Miner Res 9:1097–1105. 60. Hahn TJ, Halstead LR, Teitelbaum SL, Hahn BH 1979 Altered mineral metabolism in glucocorticoid-induced osteopenia. Effect of 25-hydroxyvitamin D administration. J Clin Invest 64:655–665. 61. Klein RG, Arnaud SB, Gallagher JC, De Luca HF, Riggs BL 1977 Intestinal calcium absorption in exogenous hypercortisolism, Role of 25-hydroxyvitamin D and corticosteroid dose. J Clin Invest 60:253–259. 62. Caniggia A, Nuti R, Lore F, Vattimo A 1981 Pathophysiology of the adverse effects of glucoactive corticosteroids on calcium metabolism in man. J Ster Biochem 15:153–161. 63. Colette C, Monnier L, Pares Herbute N, Blotman F, Mirouze J 1987 Calcium absorption in corticoid-treated subjects—effects

PHILIP SAMBROOK

64.

65. 66.

67.

68.

69.

70. 71.

72.

73.

74.

75. 76.

77. 78.

79.

of a single oral dose of calcitriol. Hormone Metab Res 19:335–338. Braun JJ, Birkenhager-Frenkel DH, Rietveld AH, Juttmann JR, Visser TJ, Birkenhager JC 1983 Influence of 1 alpha-(OH)D3 administration on bone and bone mineral metabolism in patients on chronic glucocorticoid treatment; a double-blind controlled study. Clin Endocrinol 19:265–273. Chan SD, Chiu DK, Atkins D 1984 Mechanism of the regulation of the 1alpha,25-dihydroxyvitamin D3 receptor in the rat jejunum by glucocorticoids. J Endocrinol 103:295–300. Korkor AB, Kuchibotla J, Arrieh M, Gray RW, Gleason WA Jr. 1985 The effects of chronic prednisone administration on intestinal receptors for 1,25-dihydroxyvitamin D3 in the dog. Endocrinol 117:2267–2273. Jowell PS, Epstein S, Fallon MD, Reinhardt TA, Ismail F 1987 1,25-Dihydroxyvitamin D3 modulates glucocorticoid-induced alteration in serum bone Gla protein and bone histomorphometry. Endocrinol 120:531–536. Nielsen HK, Brixen K, Mustapha K, Mosekilde L 1991 Acute effect of 1,25-dihydroxyvitamin D3, prednisone, and 1,25 dihydroxyvitamin D3 plus prednisone on serum osteocalcin in normal individuals. J Bone Miner Res 6:435–441. Lems WF, Van Veen GJ, Gerrits MI, Jacobs JW, Houben HH, Van Rijn HJ, Bijlsma JW 1998 Effect of low-dose prednisone (with calcium and calcitriol supplementation) on calcium and bone metabolism in healthy volunteers. Brit J Rheumatol 37:27–33. Hahn TJ, Hahn BH 1976 Osteopenia in subjects with rheumatic diseases, principles of diagnosis and therapy. Semin Arthritis Rheum 6:165–188. Bijlsma JW, Raymakers JA, Mosch C, Hoekstra A, Derksen RH, Baart de la Faille H, Duursma SA 1988 Effect of oral calcium and vitamin D on glucocorticoid-induced osteopenia. Clin Exp Rheumatol 6:113–119. Adachi JD, Bensen WG, Bianchi F, Cividino A, Pillersdorf S, Sebaldt RJ, Tugwell P, Gordon M, Steele M, Webber C, Goldsmith CH 1996 Vitamin D and calcium in the prevention of corticosteroid-induced osteoporosis: a three-year follow-up study. J Rheumatol 23:995–1000. Buckley LM, Leib ES, Cartularo KS, Vacek PM, Cooper SM 1996 Calcium and vitamin D3 supplementation prevents bone loss in the spine secondary to low-dose corticosteroids in patients with rheumatoid arthritis. Ann Intern Med 125:961–968. Sambrook PN, Birmingham J, Kelly PJ, Kempler S, Pocock NA, Eisman JA 1993 Prevention of corticosteroid osteoporosis; a comparison of calcium, calcitriol, and calcitonin. New Engl J Med 328:1747–1752. Reginster JY, Kuntz D, Verdickt W, Wouters M, Guillevin L, Menkes CJ, Nielsen K 1999 Prophylactic use of alfacalcidol in corticosteroid-induced osteoporosis. Osteop Inter 9:75–81. Sambrook PN, Henderson NK, Keogh A, MacDonald P, Glanville A, Spratt P, Bergin P, Ebeling P, Eisman JA 2000 Effect of calcitriol on bone loss after cardiac or lung transplantation. J Bone Miner Res 15:1818–1824. Neuhaus R, Kubo A, Lohmann R, Rayes N, Hierholzer J, Neuhaus P 1999 Calcitriol in prevention and therapy of osteoporosis after liver transplantation. Transplant Proc 31:472–473. Dykman TR, Haralson KM, Gluck OS, Murphy WA, Teitelbaum SL, Hahn TJ, Hahn BH 1984 Effect of oral 1,25dihydroxyvitamin D and calcium on glucocorticoid-induced osteopenia in patients with rheumatic diseases. Arthritis Rheum 27:1336–1343. Lambrinoudaki I, Chan DT, Lau CS, Wong RW, Yeung SS, Kung AW 2000 Effect of calcitriol on bone mineral density in premenopausal Chinese women taking chronic steroid therapy.

CHAPTER 73 Glucocorticoids and Vitamin D

A randomized, double-blind, placebo-controlled study. J Rheumatol 27:1759–1765. 80. Ringe JD, Coster A, Meng T, Schacht E, Umbach R 1999 Treatment of glucocorticoid-induced osteoporosis with alfacalcidol/calcium versus vitamin D/calcium. Calcif Tiss Int 65:337–340. 81. Ringe JD, Dorst A, Faber H, Schacht E 2003 Treatment of established glucocorticoid-induced osteoporosis with alfacalcidol or plain vitamin D. Calcif Tiss Int 72:4, Abs 037.

1251 82. Sambrook PN, Kotowicz M, Nash P, Styles CB, Naganathan V, Henderson-Briffa KN, Eisman JA, Nicholson GC 2003 Prevention and treatment of glucocorticoid-induced osteoporosis: a comparison of calcitriol, vitamin D plus calcium and alendronate plus calcium. J Bone Miner Res 18:919–924. 83. Amin S, Lavalley MP, Simms RW, Felson DT 2002 The comparative efficacy of drug therapies used for the management of corticosteroid induced osteoporosis: a meta regression. J Bone Miner Res 17:1512–1526.

CHAPTER 74

Drug and Hormone Effects on Vitamin D Metabolism SOL EPSTEIN AND ADINA E. SCHNEIDER Division of Endocrinology, Diabetes, and Bone Diseases, Department of Medicine, Mount Sinai School of Medicine, New York, NY

I. Introduction II. Hormone Effects on Vitamin D Metabolism III. Drug Effects on Vitamin D Metabolism

IV. Conclusion References

I. INTRODUCTION

vitamin D metabolism. (For a more extensive review, refer to Chapter 78.) A large body of evidence has established PTH as one of the main regulating hormones of vitamin D metabolism. Early rodent and chicken work demonstrated that PTH was necessary for the stimulation of 1α-hydroxylase and synthesis of 1,25(OH)2D3 [1–7]. Similarly, most in vitro studies have shown increased production 1,25(OH)2D3 [8–10], stimulation of 1α-hydroxylase, and inhibition of 24-hydroxylase in the presence of PTH [11–15]. The mechanism by which PTH enhances renal 1α-hydroxylase and inhibits 24-hydroxylase is most probably via second messengers. PTH increases cyclic AMP (cAMP) production in proximal tubules [16–18] and in renal slices from adult rats treated with PTH [12,13,15]. Further evidence for a role of cAMP emerged following studies by Horiuchi and co-workers [19]. They found that an infusion of cAMP and dibutyryl cAMP into vitamin D–deficient rats increases the conversion of tritiated 25-hydroxyvitamin D3 (25OHD3) to 1,25(OH)2D3 [7,19]. Furthermore, cAMP [8,9,14] and forskolin [20,21], a direct activator of adenylate cyclase which increases intracellular cAMP levels [22], have been shown, in vitro, to enhance the production of 1,25(OH)2D3 and inhibit 24-hydroxylase activity [7]. The application of molecular technology has led to important mechanistic insights of PTH action. The promotor region of the gene for 1α-hydroxylase has been shown to contain cAMP-response elements which respond to PTH [23,24]. PTH has also been shown to decrease 24-hydroxylase activity through suppression of cytochrome P450 24-hydroxylase gene expression [25], thus strengthening the association of PTH, cAMP, and 1,25(OH)2D3. Another mechanism of down-regulation of 24-hydroxylase appears to be by altering the stability of its mRNA [26]. Along with cAMP, investigators have demonstrated that the phospholipase C/protein kinase C (PKC) second messenger system may also mediate PTH stimulation

This chapter will discuss the effects that drugs and hormones have on vitamin D metabolism. The level of active vitamin D metabolite and the activity of the renal hydroxylase enzymes that are responsible for regulating its production are governed by a variety of hormones and cations. Parathyroid hormone (PTH), phosphate, and calcium are the principal controlling factors of the renal hydroxylases (discussed in Chapter 5), yet several other hormones may also play a role in the metabolism of vitamin D. Among those discussed here are parathyroid hormone-related protein (PTHrp), calcitonin, prolactin, growth hormone and insulin-like growth factor (IGF), sex steroids (estrogen, testosterone, and progesterone), insulin, thyroid hormone, prostaglandin, interferon-γ, and tumor necrosis factor-α. The findings are summarized in Table I. Along with the endogenous regulators of the vitamin D endocrine system, some exogenous ones are also capable of exerting an effect on the Vitamin D endocrine system. Drugs are being increasingly recognized in this regard. Three that have received the most attention are the anticonvulsants, ethanol, and corticosteroids, but many others have also been implicated. The mechanisms by which these drugs affect vitamin D metabolism are discussed in detail and the findings summarized in Table II (see Section III below). It should be realized that the introduction of new drugs and molecular technology may expand this area considerably.

II. HORMONE EFFECTS ON VITAMIN D METABOLISM A. Parathyroid Hormone and Parathyroid Hormone-Related Protein This section summarizes the relationship between exogenous PTH or stimulated PTH secretion and VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

1254

SOL EPSTEIN AND ADINA E. SCHNEIDER

TABLE I Hormone Effects on Vitamin D Metabolism Hormone PTH

PTHrP

Calcitonin

Growth Hormone

IGF-1 Prolactin

Insulin

Estrogen

Testosterone

Progesterone

Thyroid Hormone

Prostaglandins Tumor Necrosis Factor α Interferon γ

Study In vitro Animal Human In vitro Animal Human In vitro Animal Human In vitro Animal Human In vitro Animal In vitro Bird Mammal Human In vitro Animal Human In vitro Animal Human In vitro Animal Human In vitro Animal Human In vitro Animal Human In vitro Animal Human In vitro In vitro Animal

25OHD3

1,25OH2D3

24,25OH2D3

⇑ ⇑ ⇑

⇑ ⇑ ⇑

⇑ ⇑ ⇑ ⇔ ⇔

⇓⇑ ⇑ ⇑⇔

⇔

⇔ ⇑ ⇑⇓

⇓ ⇑⇔ ⇓

⇑ ⇑

⇔

⇑ ⇑ ⇑⇔ ⇔⇑

⇔ ⇔

⇑ ⇑ ⇔⇑

⇓ ⇓ ⇔⇓

⇔ ⇔

⇑⇓⇔ ⇑⇔ ⇑⇔

⇓⇑⇔

⇔ ⇔

⇑ ⇔⇑ ⇑

⇓

⇑ ⇔

⇑ ⇔ ⇔⇑

⇑

⇑ ⇔ ⇓

⇓ ⇓ ⇓

⇑ ⇑ ⇑

⇑ ⇑⇔ ⇑

⇓

⇑ ⇑ ⇔

Symbols denote effects on serum levels of vitamin D metabolites: ⇑, increased; ⇓, decreased; ⇔ unchanged. See text for details and references.

of 1,25(OH)2D3 secretion by mammalian renal proximal tubules in vitro [27,28]. This occurs at concentrations of PTH that are insufficient to raise cAMP content [27]. It is conceivable that PTH activation of 1,25(OH)2D3 may involve both signaling pathways, with the PKC pathway being responsive to lower concentrations of PTH [27,29]. Clinically, patients with primary hyperparathyroidism have mean plasma concentrations of 1,25(OH)2D3 that are significantly increased in approximately one-third

of cases versus controls [30–33]. In patients with Paget’s bone disease, 1,25(OH)2D3 elevation in response to mithramycin-induced hypocalcemia also corresponds to an increase in PTH secretion [34]. In addition, treatment with 1,25(OH)2D3 will correct the hypocalcemia found in hypoparathyroid patients [35–37]. Moreover, patients with pseudohypoparathyroidism, a disease characterized by metabolic unresponsiveness to PTH, have low circulating levels of 1,25(OH)2D3 [38]. 1,25(OH)2D3 is also widely used to

1255

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

TABLE II Drug Anticonvulsants Corticosteroids Ethanol Ketoconazole Statins Cholestyramine Fibric Acid Ezetimibe Bisphosponates

Thiazide diuretics Calcium channel blockers Heparin Cimetidine Aluminum Oral Parenteral Antituberculous agents Caffeine Theophylline Immunosuppressants Flouride Olestra Orlistat Lithium

Drug Effect on Vitamin D Metabolism Study

25OHD3

1,25(OH)2D3

24,25(OH)2D3

Animal Human Animal Human Animal Human In vitro Human Human Animal Human Human Animal In vitro Animal Human Human Animal Human Animal Human Animal Human Animal Human Human Human In vitro Animal In vitro Animal Animal Human Animal Human Animal Human Human Human

⇔ ⇓⇔

⇔ ⇑⇔⇓

⇓

⇔ ⇓⇔

⇓⇔ ⇓⇔⇑

⇔ ⇔

⇑ ⇓⇔

⇓ ⇓⇔

⇔

⇓ ⇓

⇔

⇔

⇓ ⇓⇔

⇔

⇓

⇑

⇓

⇔

⇔⇓

⇓ ⇑ ⇑⇓

⇓

⇑

⇓

⇑

⇔

⇓ ⇔

⇔

⇓ ⇓

⇔

⇔

⇔

⇓ ⇔⇓ ⇔ ⇔

⇔⇓ ⇓ ⇑

⇔⇓

⇔⇓ ⇓ ⇔⇑

⇓

⇓ ⇔

⇑

⇔

⇔⇑⇓ ⇔

⇑

⇔ ⇔

⇔ ⇔

⇔ ⇔ ⇔⇓ ⇔

⇔⇓

Symbols denote effects on serum levels of vitamin D metabolites: ⇑, increased; ⇓, decreased; ⇔ unchanged. See text for details and references.

decrease PTH levels in uremic patients with secondary hyperparathyroidism [39]. Another major influence on vitamin D metabolism is PTHrP (see also Chapter 43). PTHrP is a protein that is produced by a number of solid tumors, especially squamous and renal cell carcinomas, and is thought to be responsible for many cases of hypercalcemia of malignancy. However, it is also seen in the normal

physiological situation [40]. PTHrP binds to the classic type I PTH/PTHrP receptor, which is expressed in many tissues, such as bone and lung, as well as to alternative type II receptors in several nonclassic PTH target tissues, such as keratinocytes and squamous carcinoma cell lines [41]. PTHrP has a similar activity to PTH [42,43]; however, unlike in primary hyperparathyroidism where patients have normal or elevated

1256 1,25(OH)2D3 levels (see above), 1,25(OH)2D3 is not elevated and may be decreased in patients with hypercalcemia of malignancy and increased urinary cAMP excretion [44,45]. Thus, the hypercalcemia associated with humoral hypercalcemia of malignancy (HHM) is thought to be caused by increased renal calcium reabsorption [46,47] and/or increased bone resorption [33,48], rather than increased intestinal calcium absorption. Unlike the human scenario of HHM, rodent studies have shown an elevation in serum calcium and 1,25(OH)2D3 levels following tumor transplantation [49] and infusion of synthesized N-terminal fragments of PTHrP [50,51]. Walker et al. [51] also demonstrated direct stimulation of 1α-hydroxylase in rodent kidney slices in vitro by PTHrP (l–36). Conversely, Michigami et al. recently described an animal model of HHM in which serum 1,25(OH)2D3 levels were markedly reduced in the setting of severe hypercalcemia [52]. In this model, PTHrP-producing infantile fibrosarcomas were innoculated into nude rats. Administration of a bisphosphonate to these rats normalized their serum calcium and increased their 1,25(OH)2D3 levels. Moreover, 1α-hydroxylase activity was decreased in the hypercalcemic rats and rose after treatment with a bisphosphonate. In contrast, administration of a neutralizing antibody to PTHrP led to a reduction in serum calcium without an increase in 1,25(OH)2D3 levels [52]. These results suggest that PTHrP stimulates 1α-hydroxylase in this animal model. Because of the differences observed between HHM and infusion of synthesized PTHrP, additional factors, other than PTHrP, may be involved in the inhibition of renal 1,25(OH)2D3 production associated with HHM [53]. It has been suggested that hypercalcemia itself may be the cause of low 1,25(OH)2D3 levels [54]. However, Nakayama et al. [33] reported that low 1,25(OH)2D3 levels were demonstrated at any serum calcium level in patients with hypercalcemia of malignancy compared to patients with primary hyperparathyroidism. This suggests that the reduction in serum 1,25(OH)2D3 observed in these patients cannot be explained solely by an elevation in serum calcium [33]. As in animals, a study by Everhart-Caye et al. [55] revealed a doserelated increase in renal production of 1,25(OH)2D3 following a 6-hr infusion of human PTHrP (1–36) into healthy subjects. This confirms a previous report by Fraher and co-workers [56], who observed a similar increase in 1,25(OH)2D3 production following human PTHrP (1–34) infusion in healthy subjects. Although the in vivo human studies of Fraher and EverhartCaye do indeed demonstrate PTHrP stimulation of 1,25(OH)2D3 production, they were conducted in normal healthy subjects and do not mirror the clinical situation as seen in hypercalcemia of malignancy, in

SOL EPSTEIN AND ADINA E. SCHNEIDER

which a host of factors may be at play (as discussed below). They were also short-term infusion studies, and it is quite possible that equivalent findings may not be produced with longer-term studies. More recently, Horowitz et al. [57] directly compared the effects of 48-hour infusions of human PTHrP (1–36) versus human PTH (1–34) in healthy subjects. Although the calcemic, renal handling of calcium and phosphaturic effects were similar, PTH was significantly more effective at stimulating renal 1,25(OH)2D3 than PTHrP [57]. These findings, however, do not explain the low 1,25(OH)2D3 levels seen in HHM. Possible explanations for the disparate effect of PTHrP in hypercalcemia of malignancy are that other regions, or alternatively the full length of the PTHrP (1–141) molecule, may act differently from the foreshortened synthetic molecules administered in previous studies [56]. This may modify or impair the capacity of the N-terminal moiety of PTHrP to stimulate 1α-hydroxylase, as might other factors also produced by tumors, such as the cytokines, interleukin-1 (IL-1), and tumor necrosis factor, and/or growth factors [58]. The other clinical scenario in which PTHrP may play an important role is during lactation. PTHrP may regulate mammary blood flow, and is a potential mediator of the changes in calcium metabolism seen during lactation [59]. Mather et al. [60] reported a case of a woman with hypoparathyroidism whose need for calcium and calcitriol supplementation abated during lactation Her 1,25(OH)2D3 levels remained in the normal range, despite a cessation of supplemental calcitriol and a lack of endogenous PTH. This was associated with a rise in PTHrP and suggests that PTHrP was able to stimulate production of 1,25(OH)2D3 in this setting. This lends further support to the hypothesis that the decrease in 1,25(OH)2D3 seen in HHM is related to other tumor-related factors. In summary, available evidence, both in vitro and in vivo in animal models and humans, clearly defines the role of PTH as an activator of 1,25(OH)2D3 synthesis and an inhibitor of 24,25(OH)2D3 production. Amino-terminal fragments of PTHrP also stimulate 1,25(OH)2D3 production experimentally, both in vitro and in vivo in animals and humans. However, it is as yet unclear as to why patients with HHM and elevated PTHrP have reduced 1,25(OH)2D3.

B. Calcitonin The role of calcitonin in vitamin D metabolism has been thoroughly investigated in both in vivo and in vitro studies (see Chapter 39). Initial in vivo studies demonstrated a pronounced increase in 1,25OH2D3

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

levels and decrease in 24,25(OH)2D3 levels after the administration of synthetic salmon calcitonin to vitamin D–deficient rats [61]. Subsequent studies attempted to clarify if this effect was independent of PTH. Lorenc and co-workers [62] found that following thyroparathyroidectomy (TPTX), the calcitonin effect was eliminated. In contrast, subsequent studies involving vitamin D–deficient [63] and vitamin D–replete [64] TPTX rats demonstrated increased production of 1,25(OH)2D3 following calcitonin administration. Furthermore, the actions of PTH and calcitonin were additive, suggesting independent effects [63]. This effect was seen in rats fed regular diets as well as calcium-free diets. There were no significant changes in 25OHD3 levels [64]. However, Shinki et al. [65] demonstrated that calcitonin administered to sham and TPTX normocalcemic rats caused an increase in the expression of renal 25-hydroxylase and increased the conversion of 25OHD3 to l,25(OH)2D3. This effect was not seen in hypocalcemic rats whose 25-hydroxylase increased in response to PTH injection [65]. Calcitonin also appears to act as a negative regulator of intestinal 24-hydroxylase. The inhibition of intestinal 24-hydroxylase activity and expression may spare 1,25(OH)2D3 from deactivation and thereby result in enhanced l,25(OH)2D3 mediated activity [66]. The majority of in vitro data have supported the positive effect of calcitonin on 1,25OH2D3 levels [9,67,68]. Similar to in vivo data, 1α-hydroxylase was stimulated by calcitonin in vitamin D–deficient rat kidneys, postTPTX [68]. This effect was independent of the adenylate cyclase system which, as mentioned earlier, is believed to be the mechanism of PTH stimulation of 1α-hydroxylase [19,63]. These findings suggest that calcitonin acts independently of PTH. In human studies, patients with medullary carcinoma of the thyroid and excess calcitonin had serum 1,25(OH)2D3 levels that were elevated [69]. However, therapeutic use of calcitonin has produced mixed results. Injectable [70] and nasal calcitonin [62,63] treatment daily for postmenopausal osteoporosis produce either no changes [70–72] or an increase in 1,25(OH)2D3 levels at 6 months, which returned to normal at 1 year, with PTH and 25OHD3 levels being unchanged during the study period [73]. No changes in 1,25(OH)2D3 levels were observed in patients with Paget’s disease of bone treated for 3 months with calcitonin, although in these patients 24,25(OH)2D3 was elevated, most probably due to decreased consumption by osteoclasts and osteoblasts [74]. Other studies involving the treatment of Paget’s disease of bone with nasal calcitonin did show a rise in 1,25(OH)2D3 after 18 months duration, yet longer term therapy had no effect. Levels of other vitamin D

1257

metabolites such as 24,25(OH)2D3 and 25OHD3 were unchanged [75]. In summary, in vivo and most in vitro work suggest that calcitonin plays an important role in stimulating 1,25(OH)2D3 production. The less than convincing evidence seen in clinical studies may be related to the doses used or the route of calcitonin administration.

C. Growth Hormone and Insulin-like Growth Factor The recent advent of recombinant growth hormone (GH) has spurred an increased interest into the role of this hormone and its intermediary, insulin like growth factor (IGF-1), in bone metabolism. One of the many functions of GH is to stimulate intestinal calcium absorption, which has been shown in both rats [76] and humans [77,78]. Although the exact mechanism of this effect is unknown, it may be a l,25(OH)2D3 dependent phenomenon. Hypophysectomy and subsequent GH deficiency in rats was shown to reduce the level of 1,25(OH)2D3 and to increase the conversion of 25OHD3 to 24,25(OH)2D3 [79–82]. GH replacement led to the restoration of normal levels of vitamin D metabolites [79–82]. No increase in the metabolic clearance or increase in tissue metabolism of 1,25(OH)2D3 occurred; thus, the most likely explanation is that hypophysectomy decreases the renal 1α-hydroxylase activity [81]. Furthermore, hypophysectomized rats fed a low phosphate diet and replaced with rat GH had increased levels of 1,25(OH)2D3, as a result of stimulation of 1α-hydroxylase activity [83]. In other animal work, exogenous porcine GH also increased 1,25(OH)2D3 in intact pigs [84]. Of note, a more recent study in rats found that hypophysectomy also modulated the regulation of 24-hydroxylase by phosphate [84]. Clinical studies have yielded conflicting results. Acromegalic subjects, with endogenous GH excess, have increased levels of 1,25(OH)2D3 [32,84–86], which are reduced by treatment with bromocriptine, which decreases GH [84–86]. Short-term recombinant GH therapy in healthy young [87] or elderly people [88] caused significant increases in 1,25(OH)2D3, and increased vitamin D levels in response to phosphate depletion were found to be dependent on the presence of GH [89]. However, other studies, have found that chronic GH therapy does not seem to cause a rise in 1,25(OH)2D3 levels [90,91]. GH replacement given to children with GH deficiency did not alter 25OHD3, 1,25(OH)2D3, or PTH levels [92]. Interestingly, a recent study in children with end-stage renal disease found that higher doses of calcitriol were needed in those children treated with GH [93].

1258 If GH does, in fact, increase 1,25(OH)2D3 levels in humans, the possible mechanism remains controversial. Although Marcus et al. [88] found that the increase is mediated by an increase in PTH, other studies have not arrived at the same conclusion [91,94]. More recent studies continue to find the effect to be independent of PTH. Wei et al. [95] reported that children treated with GH experienced an increase in 1,25(OH)2D3 after one month of therapy, which subsequently decreased at 3 months. 24,25-Dihydroxyvitamin D decreased at both one and three months and returned to baseline at 6 months. No change in 25OHD3 was seen. IFG-1 levels increased in this study, while PTH levels declined, suggesting the effect of GH on vitamin D metabolism may be mediated by IGF-1 [95]. A small, randomized crossover study of IGF-1 versus GH found that IGF-1 increased the free calcitriol index, while calcium, phosphate, and PTH levels were unchanged [96]. Similarly, Wright et al. [97] found that GH’s positive effect on 1,25(OH)2D3 levels was independent of PTH and likely mediated by IGF-1. In vitro evidence supports the above clinical findings and suggests that the effects of GH in vivo on 1,25(OH)2D3 production are indirect, as GH fails to stimulate 1,25(OH)2D3 production in chick renal preparations [98]. Insulin-like growth factor type I (IGF-I) receptors are present on the basolateral membrane of renal proximal tubules [99], and low concentrations of exogenous IGF-I enhance 1,25(OH)2D3 synthesis when phosphate concentrations are low [100]. In vivo animal studies also show that the GH-dependent increases in serum 1,25(OH)2D3 levels induced by dietary phosphate restriction may be mediated by IGF-I, as administration of IGF-I to hypophysectomized rats increases serum 1,25(OH)2D3 to approximately the same degree as GH [101,102]. In addition, Gray [103] showed that serum concentrations of 1,25(OH)2D3 are directly related to the serum levels of IGF-I in rats fed a low phosphate diet. IGF-1 was also shown to stimulate renal 1α-hydroxylase activity in a time- and dosedependent manner in weanling mice that were phosphate depleted [104], but this was also demonstrated to be independent of changes in serum calcium or phosphate [104]. Currently, no data suggest that IGF-II affects vitamin D metabolism. The majority of in vivo data seem to suggest that GH, probably via IGF-I and especially during hypophosphatemia, is an important regulator of serum 1,25(OH)2D3 levels. Furthermore, it is possible that hormone- and phosphate-dependent enzyme stimulation occur by different mechanisms. Data from human clinical studies are more conflicting but increasingly support the theory that GH, via IGF-I, regulates serum 1,25(OH)2D3. This may partially explain the

SOL EPSTEIN AND ADINA E. SCHNEIDER

positive effects of GH on bone mass in patients with GH deficiency.

D. Prolactin Initial work involving prolactin (PRL) as a possible stimulator of 1,25(OH)2D3 synthesis came from in vitro studies of chick cell cultures. Spanos et al. [105–107] and Bickle et al. [98] found that ovine PRL could stimulate 1α-hydroxylase. Unlike the case in birds, however, in mammals, fish, and amphibians PRL does not seem to play a significant role in vitamin D metabolism [79,83,108,109]. In certain physiological situations, such as lactation, however, PRL may play a role in vitamin D metabolism. In rats, PRL stimulates intestinal transport of calcium [110], especially during lactation, when increased serum levels of 1,25(OH)2D3 are found [111,112]. Suppression of PRL by bromocriptine decreases 1,25(OH)2D3 levels in lactating rats, but the drug has no effect in nonlactating controls [113]. Increased vitamin D levels are also found in women during pregnancy and lactation [32,114,115]. The negative calcium balance and transient decrease in bone mineral density observed in lacatating women may be related to increased 1,25(OH)2D3. A study of rural Mexican women found that women who were lactating had higher 1,25(OH)2D3 than age matched nonlactating women [116]. In support of these studies, a patient with hypoparathyroidism was found to require less exogenous vitamin D supplementation during lactation than at other times [117]. Conversely, a study of postpartum women, followed for 18 months, failed to find a relationship between vitamin D levels and serum prolactin, estradiol, lactation status, or PTHrP [118]. Another study of pregnant women did not find a relationship between PRL levels and 1,25(OH)2D3 [119]. The rise in PRL associated with lactation may, however, lead to a release of PTHrP. Other than in lactating females, PRL most probably has no effect on plasma 1,25(OH)2D3 levels or calcium metabolism, as patients with hyperprolactinemia due to functioning pituitary adenomas had 25OHD3 and 1,25(OH)2D3 levels in the normal range versus age-matched controls, as well as similar serum calcium, phosphate, and PTH levels [32,120–124]. Although PRL seems to play a role in vitamin D metabolism in birds, this effect is likely species-specific. The majority of animal and clinical data suggest its positive effect on 1,25(OH)2D3 production is limited to states of physiological elevation in PRL, such as lactation. For a more detailed discussion of the effects of lactation on vitamin D metabolism the reader is referred to Chapter 51.

1259

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

E. Insulin This section will be limited to the effect of insulin on vitamin D metabolism. The immunomodulatory role of vitamin D and the effect of vitamin D on insulin release will not be addressed (see Chapter 99). Patients with insulin-dependent diabetes mellitus suffer from a number of disturbances in bone mineral metabolism, including alterations in vitamin D metabolism [125]. Work in streptozotocin- and alloxan-induced diabetic rats by Schneider and co-workers [126,127] first suggested a role for insulin in the production of 1,25(OH)2D3. This experimental diabetic model is associated with a reduction in duodenal calcium absorption, calcium binding protein, and total and ionized calcium levels [126,127]. Treatment with 1,25(OH)2D3 but not 25OHD3 corrects subnormal calcium absorption in these rats [128], as does insulin replacement [129]. These data suggest that a lack of insulin in diabetes impairs 1α-hydroxylation. Further support of this concept emerged after finding that 1,25(OH)2D3 levels in streptozotocin-induced diabetic rats were depressed to one-eighth the level in control rats and were returned to control values with insulin treatment [130]. No changes in serum 25OHD3 were found, suggesting either decreased 1α-hydroxylation of 25OHD3 or increased catabolism of 1,25(OH)2D3. It is possible that a change in PTH is responsible for decreased conversion of 25OHD3. It has been shown that PTH levels increase in diabetic rats, probably secondary to calcium malabsorption resulting from decreased 1,25(OH)2D3 [131–134]; however, some studies have reported low PTH levels in streptozotocin-induced diabetic rats [135]. To further localize the site at which insulin is proposed to act, Spencer et al. [136] studied the conversion of 3H-25OHD to 3H-1,25(OH) D , as well as the metabolic 3 2 3 clearance of 3H-1,25(OH)2D3 in control, streptozotocin-induced diabetic and insulin-treated streptozotocin-diabetic rats. The results showed that the metabolic clearance was not increased in diabetic rats and that the in vivo conversion of 3H-25OHD3 to 3H-1,25(OH) D was reduced by 60% in diabetic rats, 2 3 normalizing with insulin therapy. No intrinsic intestinal mucosal defect in the incorporation of 3H-1,25(OH)2D3 was evident [136]. These results further support the notion that a lack of insulin impairs 1α-hydroxylation. The osteopenia seen in streptozotocin-induced diabetic rats [134,135] is also significantly attenuated by treatment with 25OHD3, suggesting that vitamin D deficiency contributes to this bone loss [137]. Other studies in diabetic animals have demonstrated a decrease in 1α-hydroxylase activity and an increase in 24-hydroxylase activity [138,139]. Another study in the diabetic animal model found that insulin deficiency

directly inhibited 25-hydroxylase activity despite finding normal levels of serum 25OHD3 [140]. Insulin may also play an important role in the stimulation of renal 1,25(OH)2D3 synthesis in response to phosphate deprivation. In streptozotocin-diabetic rats, 1,25(OH)2D3 increased only slightly following phosphate deprivation, as compared with a marked response in rats replaced with insulin [141]. Decreased serum vitamin D–binding protein (DBP) has also been reported in experimentally induced diabetic rats, resulting in decreased total 1,25(OH)2D3 and normal free 1,25(OH)2D3 [142]. In vitro evidence suggests a permissive role of insulin in vitamin D metabolism. Primary cultures of chick kidney cells in serum-free medium respond to PTH with increased production of 1,25(OH)2D3 only when exposed to insulin [143]. Confirming in vitro work, Wongsurawat et al. [133], using a renal slice technique, showed reduced 1,25(OH)2D3 and increased 24,25(OH)2D3 levels in streptozotocin-induced diabetic rats, which reversed with insulin therapy. PTH levels were found to be higher in diabetic rats relative to controls [133]. Renal resistance to PTH has been suggested as the mechanism by which 1α-hydroxylase is depressed in these rats [144]. However Wongsurawat et al. showed a normal cAMP response to PTH in streptozotocin-induced diabetes [133], which makes this theory unlikely. There is also in vitro data to suggest that insulin increases the capacity of PTH to increase in the activity of 24-hydroxylase [145]. Clinical studies have shown decreased 1,25(OH)2D3 concentrations with increased 24,25(OH)2D3 and normal 25OHD3 levels in insulin-dependent diabetic children [146,147] and in poorly controlled African diabetics [148]. However, no abnormalities of calcium or vitamin D metabolism have been found in adult insulin-dependent diabetics [149]. In summary, diabetic animals show abnormalities in vitamin D metabolism with depressed levels of 1,25(OH)2D3, increased 24,25(OH)2D3, and normal levels of 25OHD3. In human studies findings have been inconsistent, which may be due to variations among the study populations. Alternatively, the animal model of diabetes may differ from diabetes in the clinical setting with regard to changes in vitamin D.

F. Sex Steroids 1. ESTRADIOL

The possible role of estrogen in vitamin D metabolism was first demonstrated in avian studies involving egg-laying Japanese quail [150]. Increased estradiol, either via ovulation [150] or exogenous

1260 administration [151], resulted in enhanced production of 1,25(OH)2D3. In females 24,25(OH)2D3 was stimulated, but in males the concentration of this metabolite was reduced. Further studies supported the notion that estradiol might be a regulator of vitamin D. Castillo et al. [152] administered 5 mg of estradiol to mature male quail and found that it markedly stimulated 1α-hydroxylase and suppressed 24-hydroxylase activity 24 hours after administration. In castrated male chickens, estradiol injections stimulated 1,25(OH)2D3 production in vitro, but only in the presence of testosterone or progesterone [153]. If all three hormones were present, they acted in a synergistic manner to stimulate 1,25(OH)2D3 synthesis [153]. Similarly, the injection of stilbestrol, an estrogenic drug, into immature male chickens stimulated 1α-hydroxylase and suppressed 24,25(OH)2D3 production in chick kidney homogenates [154]. However, there have been some contradictory in vitro studies that suggest that estradiol may not exert a stimulatory effect on the hydroxylase enzymes [155,156]. Thus, the in vivo stimulation noted in the above studies may occur via indirect means, possibly via PTH stimulation, as discussed below. In mammalian studies, female rats treated with estradiol benzoate daily for 8 days were found to have increased 1,25(OH)2D3 concentrations in plasma, gut mucosa, and kidneys [157]. Others have not found a rise in 1,25(OH)2D3 levels in rats treated with estradiol, but have recorded increased in vivo intestinal absorption of calcium [158–160]. In keeping with these findings, another study in rats found that estradiol’s ability to increase intestinal calcium absorption was independent of 1,25(OH)2D3 [161]. A study by Criddle et al. [162] suggests the decrease in renal calcium excretion associated with estrogen replacement is independent of 1,25(OH)2D3 and may be dependent on increased expression of calbindin D28k protein, located in the distal renal tubal. Furthermore, Liel et al. [163] found that the administration of estrogen to ovariectomized rats was found to increase intestinal calcium absorption by increasing the activity of duodenal vitamin D receptors (VDRs), rather than increasing levels 1,25(OH)2D3. However, in a study by Ash and Goldin [164], both young and old ovariectomized rats administered 3H-25OHD3 had reduced 3H-1,25(OH) D production, which was increased with 2 3 estradiol replacement. In the same study, parathyroidectomy eliminated estradiol’s therapeutic effect on 3H-1,25(OH) D recovery, which implicates PTH. 2 3 Estradiol may either act directly on the parathyroid gland or act by decreasing bone resorption and lowering serum calcium, thus triggering PTH secretion [164]. Despite these data, these findings have not been consistently demonstrated in ovariectomized rats by

SOL EPSTEIN AND ADINA E. SCHNEIDER

others [158,165]. Of note, recent in vivo data also demonstrate that estrogen increases the expression of the vitamin D receptor (VDR) in bone [166]. Clinically, plasma 1,25(OH)2D3 levels are elevated in human pregnancy and remain high postpartum in lactating women [114,115] (see Chapter 51). Levels of 25OHD3 in pregnant women are similar to those of controls [167]. However, another study did find a 39% increase in 25OHD3 levels in premenopausal women receiving oral contraceptives compared with nonusers [168]. Despite an expected increase in DBP, free 1,25(OH)2D3 levels are also elevated in the pregnant state [169]. Low [170,171] and normal levels [172] of 1,25(OH)2D3 have been reported in early postmenopausal women. The increased bone resorption associated with the postmenopaual period leads to a slight elevation of serum calcium, which decreases PTH secretion, and subsequently reduces renal 1α-hydroxylase activation and 1,25(OH)2D3 production. This ultimately results in decreased calcium absorption and negative calcium balance [173] (see Chapter 67). The fall in calcium absorption seen during this period, however, appears to be only partially explained by this fall in 1,25(OH)2D3 levels [174]. An alternative explanation for the decline in calcitriol is that estrogen deficiency may alter the responsivness of PTH to changes in calcium. However, in a study of 16 women who were rendered estrogen deficient via administration of a GnRH analogue, no change in the ability of PTH to respond to changes in calcium or stimulate 1,25(OH)2D3 was observed [175]. Serum levels of 25OHD3 have been reported as unchanged [170,172] or increased [176] in various studies. Recent studies in postmenopausal women have demonstrated increased 1,25(OH)2D3 levels after both short-term and long-term estrogen therapy [171,177–179]. In women with postmenopausal osteoporosis, estrogen replacement results in decreased bone resorption that lowers serum calcium, which subsequently stimulates PTH and renal 1α-hydroxylase leading to increased 1,25(OH)2D3 production. The latter increases intestinal calcium absorption [180]. However, Stock et al. [181], using sensitive assays for PTH, failed to show a rise in PTH levels accompanying the elevated 1,25(OH)2D3 in postmenopausal women treated with estradiol. In fact 2 weeks of treatment with estradiol decreased PTH levels. This suggests that the estrogen effect on vitamin D metabolism may not only be secondary to a change in PTH [181]. Estradiol-induced elevation of 1,25(OH)2D3 may occur as a result of increased DBP levels. A higher concentration of DBP would increase total 1,25(OH)2D3, but leave free 1,25(OH)2D3 unchanged [182,183]. However, as in pregnancy, free levels of 1,25(OH)2D3

1261

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

were shown to be elevated in response to estradiol treatment [184]. Although the evidence for a direct effect of estradiol on vitamin D metabolism in mammals is inconclusive, the majority of studies have found a positive effect of estrogen on 1,25(OH)2D3 levels. This effect may be indirect as estrogen alters intestinal calcium absorption, bone resorption, and PTH levels which increases, in turn, 1,25(OH)2D3 levels. 2. TESTOSTERONE

Evidence for a role for testosterone in vitamin D metabolism was first demonstrated in avian studies. An interdependence and synergism between testosterone and estradiol were observed during the in vitro stimulation of 1α-hydroxylase in castrated male chickens [153]. Castillo et al. [152] also demonstrated that testosterone alone administered to the male quail suppressed 24-hydroxylase but produced little change in 1α-hydroxylase. Again the data among species are not uniform. Hypoandrogenemia as a result of orchiectomy in rats has produced mixed findings, with either no change in total or free vitamin D metabolites [185] or decreased 1,25(OH)2D3 and DBP with normal free 1,25(OH)2D3 levels [186,187]. Castrated male guinea pigs demonstrated a 50% decline in 1α-hydroxylase activity, which was reversed with testosterone replacement. Ovariectomized female guinea pigs also responded to testosterone therapy with a 50% increase in 1α-hydroxylase activity. Both groups, however, had similar serum levels of 1,25(OH)2D3, DBP, and free 1,25(OH)2D3 levels versus controls [188]. Of note, testosterone administration to sexually immature, vitamin D3 replete male chicks was shown to decrease circulating levels of 1,25(OH)2D3, while intestine and bone concentrations were significantly increased [189]. There are limited data on the effect of testosterone on 25OHD3 levels. One study demonstrated an elevation in serum 25OHD3 levels following androgen treatment in ultraviolet irradiated rats compared to those that did not receive testosterone [190]. Studies in humans have been focused on hypogonadal men. Hagenfeldt et al. [191] studied hypogonadal men before and after treatment with testosterone enanthate every 3–4 weeks, for varying lengths of time. They found that basal serum 1,25(OH)2D3, DBP and free 1,25(OH)2D3 concentrations were similar to those of controls; however, testosterone treatment still increased total 1,25(OH)2D3 and free 1,25(OH)2D3 significantly. In contrast, Morley et al. [192] looked at the effects of testosterone replacement therapy in elderly hypogonadal males (mean age 77.6 ± 2.3 years) and found no effect on PTH or serum vitamin D metabolites pre- or posttreatment. Studies in hypogonadal men with osteoporosis have found both normal [193,194]

and low [195] serum 1,25(OH)2D3 levels. In the latter study by Francis et al. [195], testosterone replacement therapy increased both total and free 1,25(OH)2D3 levels. However, patients treated with orchidectomy for prostate cancer were found to have no changes in total 1,25(OH)2D3 or DBP concentrations [196]. In studies involving pubertal boys, results have been equivocal. Krabbe et al. [197] found no changes in the serum levels of vitamin D metabolites before and after peak pubertal testosterone surges, whereas Aksnes et al. [198] did find an increase in vitamin D. It is interesting to note that androgens have been shown to regulate vitamin D receptors in epithelial and stromal cells of the human prostate cells [199]. Moreover, genetic polymorphisms in the vitamin D receptor have been shown to be associated with prostate cancer [200]. Please refer to Chapters 94 and 68, respectively, for detailed discussions of these topics. In summary, it is unlikely that testosterone is a major controlling factor in vitamin D metabolism, but it may play a minor role in overall vitamin D homeostasis. Hypogonadal men, especially those at risk for osteoporosis, should be monitored for vitamin D deficiency and receive supplementation if necessary. 3. PROGESTERONE

Work by Tanaka et al. [153] demonstrated that in castrated male chickens progesterone, like testosterone, supported the stimulation of 1α-hydroxylase by estradiol. As previously mentioned, further marked stimulation of 1α-hydroxylase activity occurred with combined progesterone, testosterone, and estradiol treatment [153], demonstrating pronounced synergy among the sex steroids. In Japanese female quail treated with progesterone in vitro, renal production of 1,25(OH)2D3 was stimulated; however, this was significantly less than that of estradiol [151]. In the same study, immature male quail treated with progesterone had increased 24,25(OH)2D3 production [151]. Similar findings were recorded by Castillo et al. [152]. Unlike in birds, treatment of ovariectomized rats with progesterone led to an increase in 25OHD3, while 1,25(OH)2D3 levels were similar to controls [159]. In postmenopausal women, progesterone in combination with estrogen was found to lower estradiol-stimulated increases in total and free vitamin D levels [201], and norethisterone treatment caused a slight decrease in free and total vitamin D [183]. No effects on vitamin D metabolites following medroxyprogesterone therapy were seen in male patients treated for glucocorticoidinduced osteoporosis [202]. Although the role of progesterone in vitamin D metabolism has been less extensively investigated than

1262 the other sex steroids, the available literature suggests that it likely has a minor function, if any.

G. Thyroid Hormone The effect of thyroid hormone on vitamin D metabolism is of particular interest given the concern over the impact of this hormone on bone metabolism in general. In an extensive review on bone and mineral metabolism in thyroid disease, Auwerx and Bouillon [203] describe the changes encountered in this disease. In hyperthyroidism, excess thyroid hormone stimulates bone resorption [203–206], which increases serum calcium and phosphate concentration with resultant suppression of PTH [205,207–209] and a decrease in 1,25(OH)2D3 production. This leads to lower intestinal calcium absorption [210,211], which Peerenboom et al. [211] had earlier demonstrated to be reversible after treatment of the thyroid abnormality. Another possible factor contributing to low plasma 1,25(OH)2D3 levels found in hyperthyroidism is that of enhanced metabolic clearance of 1,25(OH)2D3. Karsenty et al. [212] studied seven hyperthyroid patients and found increased 1,25(OH)2D3 clearance after administration of tritiated 1,25(OH)2D3. Hyperthyroidism has been reported to be associated with either unchanged [211,213–215] or decreased [216,217] levels of 25OHD3. A recent study found that 68% of men and 29% of women who were undergoing subtotal thyroidectomy for Grave’s disease had vitamin D deficiency, defined as 25OHD3 < 25 nmol/l [218]. In fact, despite normal bone mass and markers of bone turnover, 25OHD3 and 1,25(OH)2D3 levels have been shown to be persistently reduced in euthyroid patients who were 6 years post treatment of hyperthyroidism [219]. Dietary vitamin D intake and exposure to sunlight may contribute to these differences in 25OHD3. Circulating levels of 24,25(OH)2D3 are generally increased in hyperthyroid patients [214,215]. Hypothyroidism is associated with decreased bone turnover [220] and low serum calcium levels that activate PTH, which, in turn, enhances 1α-hydroxylase activity [209,213,221]. Finally, serum-binding protein, DBP, does not appear to be affected by thyroid status [213]. Similar to human hyperthyroidism, daily injections of L-thyroxine to rats decreased 1,25(OH)2D3 and increased 24,25(OH)2D3 and 25OHD3 levels [222]. Hypothyroidism in the rat induced by the drug propylthiouracil, however, produces confusing results in that it is associated with low calcium and phosphate levels and decreased 1,25(OH)2D3 concentration [223].

SOL EPSTEIN AND ADINA E. SCHNEIDER

There is some in vitro evidence to suggest that thyroid hormone directly affects renal 25OHD3 metabolism. Kano and Jones [224] found that thyroxine, triiodothyronine, and thyrotropin (TSH) decreased 1,25(OH)2D3 synthesis in perfused rat kidneys from vitamin D–deplete rats, whereas 24,25(OH)2D3 synthesis was increased in kidneys from vitamin D–replete rats. Miller and Ghazarian [225] also demonstrated a 50% reduction in 1α-hydroxylase activity in vitamin D– deficient chicks and stimulation of 24-hydroxylase activity in vitamin D–replete chicks following thyroxine administration. The majority of evidence suggests that thyroid hormone indirectly affects vitamin D metabolism via alterations in serum calcium, phosphate, and PTH. A direct action of thyroid hormone on renal hydroxylase activity, however, has been suggested by in vitro work. Again, the clinical significance in vitamin D–replete states is unknown.

H. Prostaglandins The effect of prostaglandins on vitamin D metabolism has been investigated in vitro and in vivo. Work by Trechsel et al. [226] in primary chick kidney cell cultures found that the addition of prostaglandin E2 (PGE2) and prostaglandin F2a (PGF2a) stimulated 1α-hydroxylase activity in a dose-dependent manner. PGE2 also significantly decreased 24-hydroxylase activity [226]. A proposed mechanism of action is that prostaglandins act through an increase in cAMP [226,227], but as discussed below this has not been proven. Further in vitro work by Wark et al. [228] in isolated renal tubules, prepared from vitamin D–deficient chicks, showed that the addition of PGE2 to the tubule incubation medium in the presence of 1,25(OH)2D3 increased 1,25(OH)2D3 production. Acetylsalicylic acid, which inhibits prostaglandin synthesis, decreased the prostaglandin content of the tubule medium and subsequently inhibited 1,25(OH)2D3 production. Frusemide raised prostaglandin content in a dosedependent manner, which led to a significant increase in 1,25(OH)2D3 production and decreased 24,25(OH)2D3 synthesis [228]. Kurose et al. [229] found similar results in vitro. In vivo work involving vitamin D–deficient TPTX rats by Yamada et al. [230] found that, after an intraarterial infusion of PGE2, 1,25(OH)2D3 production was significantly stimulated. No changes in plasma calcium or phosphate levels or urinary cAMP excretion were observed, suggesting that the effects of prostaglandins are independent of the cAMP system [230].

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

In a further rat study, the effects of PGE2 on the actions of PTH and calcitonin and the conversion of 3H-25OHD to 3H-1,25(OH) D were investigated. 3 2 3 PGE2 inhibited calcitonin stimulation of 1,25(OH)2D3 but had no effect on PTH-stimulated 1,25(OH)2D3 production. This suggests that PGE2 may modulate the actions of calcitonin, but not those of PTH on 1α-hydroxylase [231]. Reduced plasma levels of 25OHD3 and 1,25(OH)2D3 were also found following the administration of indomethacin, a potent inhibitor of prostaglandin synthesis, to pregnant rabbits [232]. In contrast to the above findings, a lack of effect of prostaglandin in vivo has been shown by Katz et al. [233], who administered PGE2 by subcutaneous injection daily for 3 weeks to rats. They found no change in 1,25(OH)2D3 or PTH levels, although a significant increase in bone mass in PGE2-treated rats was demonstrated [233]. Furthermore, long-term subcutaneous administration of indomethacin failed to alter 1,25(OH)2D3 or affect histomorphometric indices of bone formation and resorption [234]. Limited clinical data are available to assess the role of PGE2 in humans. In one study, elevated levels of PGE2 were believed to be the cause of hypercalcemia and enhanced 1α-hydroxylase activity in children with Bartter’s syndrome [235]. Another study of children with idiopathic hypercalcuria found a positive correlation between 1,25(OH)2D3 and prostaglandin E2 activity [236]. Overall, although it would seem that prostaglandins stimulate 1α-hydroxylase in vitro, variable results in vivo create doubt over the exact contribution that prostaglandins make in vitamin D metabolism. Again, the problem of which particular animal species is used may confound the results.

I. Tumor Necrosis Factor-α and Interferon-γ The majority of information regarding the effects of particular cytokines on vitamin D metabolism has been derived from in vitro studies. Pryke et al. [237] examined the effects of tumor necrosis factor-α (TNFα) on 1α-hydroxylase activity in cultured alveolar macrophages. Incubation for 6 days with 50 IU TNFα resulted in an average fourfold increase in 1,25(OH)2D3 production. An increase in 1α-hydroxylase activity was maximally reached at 72 hours. This finding may account for the “spontaneous” 1α-hydroxylase activity of sarcoid macrophages encountered in sarcoidosis [237] (see Chapter 79). A further study by Bickle et al. [238] provided evidence that TNFα stimulates 1,25(OH)2D3 production in human keratinocytes. In preconfluent

1263

cells, TNFα stimulates 1,25(OH)2D3 production; however, this ceased once the keratinocytes achieved confluence [238]. TNFα has also been shown to induce 1α-hydroxylase activity in human endothelial cells [239]. No effect of TNFα has been shown on bone cell lines, but TNFα was shown to inhibit vitamin D–receptor number in osteoblastic cells [240]. TNFα, via activation of NF-kappaB, has also been shown to decrease osteoblast transcriptional responsiveness to 1,25(OH)2D3 [241]. This may be one of the mechanisms by which TNFα contributes to bone loss in disease states such as postmenopausal osteoporosis and inflammatory arthritis. However the predominant role of TNFα may be independent of 1,25(OH)2D3 and result from osteoclastic effects on bone resorption. There are also some data on interferon-γ (IFN-γ) and its role in vitamin D metabolism. Cultured normal human pulmonary alveolar macrophages in the presence of IFN-γ increased 1,25(OH)2D3 production in a dose-dependent manner [242]. Bickle et al. [243] provided confirmatory findings showing that IFN-γ stimulated preconfluent keratinocytes to produce 1,25(OH)2D3. Bone marrow-derived macrophages were also demonstrated to respond to IFN-γ with enhanced 1,25(OH)2D3 production [244]. Because lung T lymphocytes from sarcoidosis patients produce IFN-γ, this may play a role in extrarenal 1,25(OH)2D3 production in vivo in sarcoidosis [245]. Little is known about the in vivo effect of IFN-γ on vitamin D metabolism. Mann et al. [246] studied the effect of IFN-γ on bone mineral metabolism in rats and reported no changes in serum ionized calcium, PTH, or 1,25(OH)2D3 levels, whereas IFN-γ had a significant osteopenic effect. Human studies on the effects of these cytokines are lacking, and the underlying disease for which these factors are therapeutically administered may influence the results (e.g., chronic active hepatitis and malignancy). In summary, aside from the possibility that TNFα and IFN-γ may play a role in extrarenal 1,25(OH)2D3 production in certain disease states, like sarcoidosis, it is at present unclear if TNFα and IFN-γ physiologically influence vitamin D metabolism. The reader should refer to Chapter 79 for further information on extra-renal 1α-hydroxylase activity in diseases such as sarcoidosis.

III. DRUG EFFECTS ON VITAMIN D METABOLISM The effects of drugs on vitamin D metabolism are summarized in Table II.

1264 A. Anticonvulsants Anticonvulsants have long been recognized to cause a number of alterations in bone mineral metabolism. In 1968 Kruse [247] first reported osteomalacia resulting from the use of anticonvulsants, and since then there have been numerous documentations of this finding [248–250]. Despite this, the effect of these agents on vitamin D metabolism remains controversial. Numerous animal and clinical studies have yielded conflicting results. Animal studies have demonstrated enhanced metabolism [251] and biliary excretion [252] of vitamin D. Although Hahn et al. [253] showed an initial increase in 25OHD3 after phenobarbital treatment in the rat, this was followed by a subsequent decline in levels. However, Ohta et al. [254] reported no effect of low-dose phenytoin treatment on serum 25OHD3 or 1,25(OH)2D3 under conditions where the rats were fed a vitamin D–supplemented diet. More recent animal data revealed that administration of phenytoin to growing rats for 5 weeks led to a significant decrease in osteocalcin levels, while there were not significant changes in serum calcium, pyridinoline, 25OHD3 or PTH compared with vehicle-treated rats [255]. This same study showed decreased trabecular bone volume and trabecular thickness, without significant change in osteoid thickness in phenytoin-treated animals. This may be modified by the administation of vitamin D and/or analogs [255]. A number of biochemical changes have been demonstrated in humans after anticonvulsant use. Serum calcium is reduced [249,256–258], which leads to the development of secondary hyperparathyroidism [249,258,264,266]. However, serum phosphate levels have been reported to be unchanged [249,258,262, 264,266]. Levels of 1,25(OH)2D3 have been reported to be high [267,273,274], normal [265,267,269], or low [266], while 24,25(OH)2D3 levels have been shown to be decreased [274,275] following long-term anticonvulsant use. An increase in 1,25(OH)2D3 concentration may be secondary to increased PTH secretion. Levels of DBP are reported as unchanged [266], and intestinal absorption of vitamin D is not altered by anticonvulsant drugs [249]. The overwhelming majority of reports demonstrate low serum levels of 25OHD3 [249,258–268]; however, there are some studies in which 25OHD3 levels are unchanged [250,269–272]. These conflicting data have been thought to be due to differences in study design and in particular, differences in the ambulatory status of subjects [276]. One might hypothesize that the variation in vitamin D levels may be more related to exposure to ultraviolet light than to the direct effect

SOL EPSTEIN AND ADINA E. SCHNEIDER

of these agents on vitamin D metabolism. To control for this potential important confounder, more recent studies have attempted to characterize the skeletal health of ambulatory patients. A recent study of 30 ambulatory adult patients treated with phenytoin, carbamazepine or valproate found significantly lower 25OHD3 levels than age- and sex-matched controls [277]. The decrease in 25OHD3 was independent of anticonvulsant type. In contrast, a study of 18 ambulatory pediatric patients on valproate or carbamezapine found normal levels of vitamin D and PTH [278]. Another study of 60 pediatric patients receiving carbamazapine also found evidence of normal vitamin D metabolism [279]. Anticonvulsants are unlikely to have uniform effects on vitamin D metabolism. Phenobarbital, phenytoin, carbamazepine, and primidone are well-known inducers of hepatic cytochrome P450 enzymes [280–282]. The observed reduction in 25OHD3 levels with these agents is thought to arise from their enhancing the hepatic breakdown of vitamin D into inactive polar metabolites other than 25OHD3 [252,260,291]. Hahn et al. [260] demonstrated, both in vivo in humans and in vitro in rat liver, that phenobarbital stimulated the conversion of vitamin D to more polar metabolites. Serum 25OHD3 levels are clearly decreased in patients receiving phenobarbital [259–262,268] and phenytoin [259,261,262,265,268,274,283]. Conflicting evidence, however, exits regarding the effect of carbamazepine on 25OHD3 levels. Some reports show no change [276,284], and some show decreased levels [259,285,286] with carbamazepine. The newer anticonvulsants such as valproate, lamotrigine, clonazapam, gabapentin, topamirate, and ethosuximide are noninducers of cytochrome P450 enzymes. No reduction of 25OHD3 concentration has been reported with the use of sodium valproate [259,261]. Yet, osteoporosis has been reported in children on chronic valproic acid therapy [287,288], and long-term use of valproate and lamotragine has been shown to be associated with reduced bone formation [289]. Of note, a recent study of 71 patients on chronic anticonvulsant therapy found no significant difference in 25OHD3 levels between patients on inducers versus noninducers; despite this, patients on inducers did have lower BMD than patients on noninducers [290]. Various factors that have been shown to be associated with more severe changes in vitamin D metabolism include: polytherapy [256,259,263,268], larger total daily dose [256,262,265,268], duration of therapy [262], and female sex [266]. Treatment of anticonvulsant-induced alterations of vitamin D with replacement therapy of 400–4000 IU/day of vitamin D3 has been shown to be effective in normalizing parameters of mineral metabolism and improving bone

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

mass [249,292]. In addition, treatment of anticonvulsant osteomalacia with small doses of oral 25OHD3 has also proved effective [263]. In summary, the overall results demonstrate that most anticonvulsants increase the metabolism of 25OHD3, and therefore it is recommended that vitamin D metabolites be consistently monitored. For patients on anticonvulsant therapy for longer than 6 months, the prophylactic use of vitamin D should be considered. Those especially at risk include long-term institutionalized patients, those with reduced ultraviolet light exposure, those with poor dietary intake of vitamin D, and those receiving long-term treatment with multiple anticonvulsant drugs [292].

B. Corticosteroids Corticosteroid use is among the most important causes of secondary osteoporosis (see Chapter 73). Pharmacological levels of corticosteroids impair the intestinal absorption of calcium [293–296] and induce hypercalciuria [297,298]. Both lead to a state of secondary hyperparathyroidism [298–301]. Hyperparathyroidism, however, does not appear to explain the observed effects of corticosteroids on the skeleton. The histology of bone exposed to glucocorticoids reveals decreased bone remodeling, while that of bone exposed to hyperparathyroidism shows increased bone remodeling [302]. The evidence against PTH having a significant putative role in the development of glucocorticoid-induced osteoporosis is detailed in a recent review by Rubin and Bilezekian [303]. In this section, we focus on potential actions of corticosteroids on vitamin D metabolism. There are multiple other pathways postulated for the effect of these agents on bone, most notably the decrease in number and function of osteoblasts, i.e. increased apoptosis, changes in synthesis and binding of skeletal growth factors, and enhanced activity of 11-beta hydroxysteroid dehydrogenase type 1 in osteoblasts [304]. Levels of 25OHD3 were initially shown to be low in glucocorticoid-treated men. Avioli and co-workers [305] reported evidence of impaired hepatic conversion of vitamin D to 25OHD3. Furthermore, Klein et al. [295] demonstrated low 25OHD3 levels following high dose prednisone therapy, although in patients receiving low or alternate day doses 25OHD3 levels were unchanged. Similar findings were reported by Seeman et al. [299] in patients receiving high dose glucocorticoids for the treatment of connective tissue disorders. These findings are plausible in light of the fact that corticosteroids are known to induce hepatic microsomal oxidase enzymes [280], and hence chronic glucocorticoid use could

1265

produce effects similar to chronic anticonvulsant use (see Section III.A above). However, other studies have failed to show a decrease in 25OHD3 concentrations. No differences in serum 25OHD3 levels have been shown in patients with corticosteroid-induced osteopenia [300,306,307] or in healthy [295] or diseased subjects treated with corticosteroids [308–310]. Patients treated with inhaled steroids for asthma also fail to demonstrate any alterations in 25OHD3, 1,25(OH)2D3, or PTH levels [311,312]. Furthermore, patients with Cushing’s disease were also found to have normal levels of 25OHD3 [313,314]. Similarly, in animal studies, no effect on the conversion of vitamin D to 25OHD3 or 25OHD3 to 1,25(OH)2D3 has been reported [296,315,316]. Carre et al. [317] suggested that use of prednisolone may enhance the inactivation of 1,25(OH)2D3 to more polar biologically inactive metabolites at the tissue level in rats. The same study showed no alteration in the conversion of 3H-25OHD3 to 3H-1,25(OH)2D3 or 3H-24,25(OH) D . This is in accordance with the find2 3 ings of Favus et al. [316], who demonstrated normal conversion of tritiated 25OHD3 to 1,25(OH)2D3 and normal subcellular localization of 1,25(OH)2D3 in the intestinal mucosa of glucocorticoid-treated rats. In children treated with glucocorticoids for glomerulonephritis, 1,25(OH)2D3 concentrations have been shown to be reduced [318]. However, these children had proteinuria, which could decrease the levels of DBP. Adolescents with systemic lupus erythematosus also had low 1,25(OH)2D3 levels following treatment with glucocorticoids [319]. However, the effect of corticosteroids on 1,25(OH)2D3 levels remains controversial. Seeman et al. [299] reported no changes in 1,25(OH)2D3 levels in 14 patients with either endogenous or exogenous glucocorticoid excess. They also showed no differences in the production or metabolic clearance rate of 1,25(OH)2D3 [299]. Others have confirmed these findings of unchanged 1,25(OH)2D3 levels [293,307–309,315]. However, certain studies have shown increased 1,25(OH)2D3 values following subacute prednisone use [294,320–322], and in patients with Cushing’s disease 1,25(OH)2D3 levels are in the normal range but decrease following remission. This suggests that higher, though normal, values of 1,25(OH)2D3 are present in the untreated state [313]. Increased levels of 1,25(OH)2D3 may be due to the state of secondary hyperparathyroidism produced by glucocorticoids. Cortisol has been shown to stimulate PTH secretion by rat parathyroid glands in vitro [323]. In vivo this has been demonstrated both in rats [324] and in humans following acute and chronic administration of glucocorticoids [298–301]. Elevated PTH is not a consistent finding, however, as normal PTH

1266 levels have been reported in both endogenous [313] as well as exogenous [294,299,307–309] glucocorticoid excess. Studies have also demonstrated no differences in levels of DBP [293] or concentrations of 24,25(OH)2D3 [294,307–309]. The exact effect that corticosteroids have on vitamin D metabolism remains controversial. The conflicting evidence may reflect the differences in experimental conditions including variations in dosages, dietary intake, and/or sunlight exposure, and very importantly the effect of the underlying disease for which the glucocorticoids were given. Another explanation might be variation in genetic susceptibility to the effects of glucocorticoids among individuals. The current literature, however, does not support an association between vitamin D receptor gene polymorphisms and corticosteroid-induced osteoporosis [325]. Please refer to Chapter 73 for a detailed discussion of glucocorticoids and vitamin D.

C. Ethanol It is well known that chronic alchoholics suffer from bone disease, and abnormalities in vitamin D metabolism have been well described. Despite this, the most common histomorphometric abnormality seen in these patients is osteoporosis, rather than osteomalacia [326]. Serum levels of 25OHD3 have been reported to be low [327–333] or normal [334–336] among alcoholic patients. Serum 1,25(OH)2D3 has also been shown to be reduced [332,337,338] or unchanged [334]. This seems to be due to a combination of factors. First, many alcoholics have an inadequate dietary supply of vitamin D. A cross-sectional study of 181 male alcoholics revealed a small but significant association between bone mass and 25OHD3; however, a marked association between nutritional status and bone mass was observed [339]. Second, many have insufficient exposure to the ultraviolet light of the sun for adequate synthesis of vitamin D. Malabsorption [340] and increased biliary excretion of 25OHD3 [341] are other possible factors involved. However, intestinal absorption of vitamin D has been shown to be normal in patients with alcoholic liver disease [336,342]. Induction by alcohol of the cytochrome P450 system may also occur with subsequent increase in the degradation of vitamin D metabolites in the liver [343]. Ethanol may also inhibit hydroxylase activity in the kidney [344] or liver. However, hepatic hydroxylation was normal in cirrhotic alcoholics [336,343,345]. Decreased DBP may lead to low levels of vitamin D metabolites in patients with cirrhotic liver disease [346,347]. It is important to note that acute ethanol

SOL EPSTEIN AND ADINA E. SCHNEIDER

ingestion has not been shown to cause changes in vitamin D metabolism [348]. One might hypothesize that PTH mediates the effect of ethanol on vitamin D metabolism. However, human and animal data on the effects of ethanol on PTH are conflicting. Acute ethanol loading in rats has produced both elevated [349] and suppressed [350] serum levels of PTH, whereas alcohol has been shown to stimulate PTH release from bovine parathyroid cells in vitro [351]. In humans, acute alcohol ingestion produces unchanged [352] or increased PTH [351] levels. Serum levels of PTH in chronic alcoholics have been normal [345,353–355], decreased [339], or increased [328,332,334,356], the latter probably secondary to diminished intestinal absorption of calcium [357–360]. Increased PTH secretion leads to inhibition of tubular reabsorption of phosphate and lower serum phosphate levels [361]. However, 1,25(OH)2D3 levels are not increased despite increased PTH and low phosphate, which suggests a direct inhibition of 1α-hydroxylase by ethanol [362]. The decreased PTH levels reported in some human studies may be related to hypomagnesemia [363]. In contrast to human studies, ethanol in rats seems to increase the production of 25OHD3 by the liver and lowers 1,25(OH)2D3 levels [364–366]. This is due to the presence of both mitochondrial and microsomal hydroxylases in the rat, as opposed to only mitochondrial enzymes in humans [367,368]. Ethanol inhibits mitochondrial enzymes, but induces microsomal ones. This might lead to increased 25OHD3 in rats by microsomal induction and to decreased 25OHD3 in humans by mitochondrial suppression [369]. Thus, the rat does not appear to be an appropriate model for assessing changes in vitamin D metabolism induced by ethanol. In summary, in chronic alcoholics, the circulating concentrations of vitamin D metabolites likely depend on the overall nutritional and health status of the individual [370]. A well-functioning alcoholic with satisfactory dietary ingestion and adequate sunlight exposure might be expected to have normal levels of circulating 25OHD3 or 1,25(OH)2D3. In contrast, nonfunctioning alcoholics with poor nutrition and reduced sunlight exposure are likely to have low 25OHD3 and 1,25(OH)2D3 levels [370].

D. Ketoconazole The antifungal agent ketoconazole has been shown to inhibit cytochrome P450-dependent enzymes [371], and thus might be expected to alter vitamin D metabolism. In in vitro data demonstrated a dosedependent reduction in 24-hydroxylase activity [372].

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

Additional in vitro data revealed that both ketoconazole and a similar antifungal, miconazole, behave as competitive inhibitors of 1α-hydroxylation of 25OHD3 [373]. Clinically, a reduction in 1,25(OH)2D3 was demonstrated in healthy men treated for 1 week with ketoconazole [374]. No changes in 25OHD3, PTH, or serum calcium or phosphate levels were shown, suggesting a direct inhibitory effect of the drug on renal 1α-hydroxylase activity [374]. Subsequently, Glass and Eil [375] treated patients with primary hyperparathyroidism and hypercalcemia for one week with ketoconazole. The treatment produced a reduction in 1,25(OH)2D3 levels. Serum total calcium but not serum ionized calcium levels fell, with no changes in 25OHD3, PTH, or serum phosphate [375]. A slightly longer study of two weeks duration, also in hyperparathyroid patients, confirmed low 1,25(OH)2D3 levels but also demonstrated a nonsignificant fall in 25OHD3 [376]. This raises the possibility that the 25-hydroxylase enzyme is also inhibited, as it is a cytochrome P450-dependent enzyme. Perhaps studies of a longer duration will demonstrate significant falls in 25OHD3 as well. Finally, ketoconazole has been shown to be effective in decreasing serum 1,25(OH)2D3 and calcium concentrations, both in cultured pulmonary alveolar macrophages taken from patients with sarcoidosis [377] and in vivo in sarcoid patients [377,378]. This suggests that ketoconazole also inhibits the extrarenal production of 1,25(OH)2D3 known to occur in sarcoidosis (see Chapter 79). Current data on the effects of ketoconazole on vitamin D metabolism suggest that it decreases 1,25(OH)2D3 levels by directly inhibiting 1α-hydroxylase activity. The effects of ketoconazole on 25-hydroxylase activity remain to be determined. During treatment with ketoconazole for any of its indications, susceptible individuals should be monitored and supplemented with vitamin D, if necessary. Of note, combination therapy with ketoconazole and vitamin D analogs enhances the inhibitory effects of vitamin D on prostate cancer cell growth [379]. See Chapter 94 for a discussion of vitamin D and prostate cancer.

E. Hypolipidemics Drugs that inhibit 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMG-CoA reductase) are the most commonly prescribed agents for the treatment hypercholesterolemia. Because HMG-CoA reductase inhibitors (or STATINs, as they are also known) are potent inhibitors of cholesterol synthesis, they may impact vitamin D production because cholesterol is the vitamin D3 precursor. Ismail et al. [380] studied 40 hypercholesterolemic patients treated for 24 weeks

1267

with one of the selective HMG-CoA reductase inhibitors, pravastatin (40–80 mg daily). Results showed that although levels of total and low density lipoprotein (LDL) cholesterol were significantly reduced, no changes in 25OHD3, 1,25(OH)2D3, or PTH levels were observed [380]. The capacity of the skin to synthesize vitamin D3 after ultraviolet light exposure also showed no changes after three months of pravastatin therapy [381]. Lack of effect of HMG-CoA reductase inhibitors on vitamin D metabolism has been confirmed by others [382], and indeed combination therapy using both pravastatin (40 mg/day) and the bile acid sequestrant cholestyramine (24 g/day) failed to affect levels of vitamin D metabolites or PTH [380]. Although statins have a theoretic potential to alter vitamin D metabolism, the literature does not seem to show that these agents reduce or increase vitamin D levels. This is intriguing in light of recent literature suggesting that statins may have an anabolic effect on bone [383] and may be protective against fracture [384]. Absorption of vitamin D from the gut requires the presence of bile acid. Moreover, vitamin D is excreted in the bile [385], and some degree of enterohepatic circulation of vitamin D occurs [386,387]. Thus, it stands to reason that cholestyramine might adversely affect vitamin D absorption. There have been reports in rats of decreased vitamin D absorption following treatment with cholestyramine [388]. Moreover, there are isolated case reports of osteomalacia associated with long-term cholestyramine use, which were attributed to cholestyramine-induced vitamin D deficiency [389,390]. Compston and Thompson [391] also reported decreased levels of 25OHD3 and reduced intestinal absorption of vitamin D in patients with primary biliary cirrhosis treated with cholestyramine for greater than two years. However, a large double-blind randomized trial of patients treated with cholestyramine (24 g/day) for four months showed similar levels of vitamin D metabolites, PTH, and serum calcium and phosphate compared with placebo-treated patients [392]. Ismail et al. [392] showed equivalent findings after six months of therapy [392], as did long-term studies using another bile acid sequestrant, colestipol, for the treatment of children with familial hypercholesterolemia [393,394]. Ezetimibe, a novel cholesterol absorption inhibitor, has been shown not to inhibit the absorption of vitamin D across the intestinal wall in rodents [295]. The effect of this agent on vitamin D absorption, calcium metabolism, or bone mass has not been reported in humans. A single report of the effect of fibrates on vitamin D metabolism showed a decline in 25OHD3 and a rise in 1,25(OH)2D3 levels [396]. It seems, therefore, that vitamin D is unaffected by the statin drugs and short-term cholestyramine treatment.

1268 Long-term cholestyramine therapy may affect 25OHD3, and levels should be routinely monitored in susceptible individuals and vitamin D supplementation provided, if necessary. The effect of cholestyramine on bone mass and fracture has not been investigated. Research into the effects of ezetimibe on vitamin D metabolism in humans is warranted.

F. Bisphosphonates Bisphosphonates are widely used for the treatment of osteoporosis, Paget’s bone disease, and hypercalcemia of malignancy. By inhibiting bone resorption [397,398], the agents cause an increase in the calcium mineral content of bone and a decrease in serum calcium levels [399]. This fall in serum calcium results in the stimulation of PTH secretion. PTH reduces the renal tubular reabsorption of phosphate, which reduces serum phosphate levels. Both increased PTH and low phosphate levels can lead to increased 1,25(OH)2D3 production [400–402]. The first generation bisphosphonate, ethane1-hydroxy-l,1-diphosphate (EHDP), also known as etidronate, was initially shown to cause a reduction of 1,25(OH)2D3 levels at high doses in vivo [403,404]. This effect was probably an indirect one as EHDP failed to stimulate or inhibit 1α- or 24-hydroxylase in primary chick kidney cell cultures [10]. Later, evidence emerged to suggest that bisphosphonates indirectly stimulate 1α-hydroxylase at least in part by some unknown humoral factor. An experimental bisphosphonate compound, YM175, was shown in TPTX and sham-operated rats to increase 1,25(OH)2D3 and 1α-hydroxylase in a dose-dependent manner; however, no increase in 1α-hydroxylase activity was demonstrated in vitro [405]. In clinical studies, Paget’s disease patients treated with etidronate had higher 1,25(OH)2D3 levels without significant changes in serum PTH, ionized calcium, or phosphate [66,406]. Lawson-Matthew et al. [407] found increased 1,25(OH)2D3 concentrations following shortterm oral etidronate treatment but decreased 1,25(OH)2D3 levels after high-dose intravenous etidronate. The effects of etidronate are thus most probably doserelated. In rats low-dose etidronate inhibits bone resorption and calcium release from bone and stimulates dietary calcium absorption. The lower calcium concentration, in turn, probably either directly or indirectly, via PTH, mediates 1α-hydroxylase activation. However, high-dose etidronate inhibits bone formation as well as resorption, and decreased calcium absorption from the gut is found [399]. Paget’s disease patients treated with another bisphosphonate, aminohydroxy-propylidene bisphosphonate (APD), known as pamidronate, also

SOL EPSTEIN AND ADINA E. SCHNEIDER

had elevated 1,25(OH)2D3 levels following shortterm intravenous [408,409] and oral [410] administration. PTH rose to twice pretreatment levels in a response to a fall in ionized calcium levels [408,410]. 24,25(OH)2D3 concentration declined, but 25OHD3 remained unchanged [408]. Concentrations of 1,25(OH)2D3 [411,412], as well as PTH [412–414], were also elevated after treatment with pamidronate in patients with tumor-associated hypercalcemia. The third generation bisphosphonate 4-aminohydroxybutylidene-l,1-bisphosphonate, or alendronate, is a 100- to 500-fold more potent inhibitor of bone resorption than is etidronate [415]. Postmenopausal women treated with alendronate showed initial rises in 1,25(OH)2D3 and PTH that normalized after chronic administration of the drug [401,416,417], probably because of inhibition of bone resorption and decreased serum calcium concentrations. 1,25(OH)2D3 and PTH levels also rose following intravenous alendronate infusion for the treatment of hypercalcemia of malignancy [412]. Bisphosphonates also appear to have an effect on 25(OH)D3 levels. A randomized controlled trial of 20 breast cancer patients treated with clodronate versus calcitonin revealed decreased 25(OH)D3 levels in addition to decreased calcium levels and increased PTH and 1,25(OH)2D3 levels in the clodronate treated subjects [418]. In general, bisphosphonates do appear to influence vitamin D metabolism. The most common finding is an elevation in 1,25(OH)2D3 levels, which is most likely due to secondary alterations in serum-ionized calcium, phosphate, and PTH levels. These alterations generally appear to be short-term, but may theoretically also confer some initial benefit in regard to an anabolic action of the elevated PTH levels. It is advisable to rule out vitamin D deficiency prior to initiation of bisphosphonate therapy and to supplement all patients with calcium and vitamin D.

G. Thiazide Diuretics Clinically, the use of thiazide diuretics has been associated with favorable effects on bone mineral density [419–421] and hip fracture rate [423–425]. These beneficial effects may be related to its ability to decrease PTH-stimulated bone resorption and decrease bone remodeling [422]. Thiazide use leads to a number of alterations in bone mineral metabolism. Thiazides are well known to decrease urinary calcium excretion [420,425–427], with resultant increases in serum calcium concentration [426,428]. This leads to reduced PTH levels [420,422], which in turn decrease 1,25(OH)2D3

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

synthesis [420,422,429,430] and leads to a reduction in intestinal calcium absorption [420]. Clinical studies on the effect of hydrochlorothiazide on vitamin D metabolism have shown a consistent decrease in 1,25OH2D3 levels. The administration of 50 mg/day of hydrochlorothiazide to postmenopausal women significantly decreased 1,25(OH)2D3 and increased 25OHD3 and 24,25(OH)2D3 levels [420]. Similar findings of low 1,25(OH)2D3 levels were reported by Sowers and co-workers [429]. Riis and Christiansen [431] conducted a double-blind, long-term controlled trial in early postmenopausal women and demonstrated a trend toward lower 1,25(OH)2D3 with a significantly elevated 24,25(OH)2D3 concentration [431]. No evidence of a direct effect on the synthesis or degradation of 1,25(OH)2D3 by thiazides is currently available. It thus seems that changes in vitamin D metabolites found after the use of thiazide diuretics are most likely secondary to alterations in serum calcium concentrations and PTH levels.

H. Calcium Channel Blockers Calcium channel blockers, of which nifedipine, verapamil, and diltiazem are the most well recognized and frequently prescribed, are chemically dissimilar. As such they have different effects on PTH secretion. In vitro experiments show that verapamil both increases and decreases PTH secretion, depending on extracellular calcium concentrations [432–434]. Diltiazem inhibited PTH release in bovine parathyroid cells by 40% and in human parathyroid cells by 20% [435]. A similar inhibition of PTH secretion was seen with nitrendipine, an analog of nifedipine [436]. It has been suggested that in certain circumstances calcium channel blockers can act as calcium agonists, thus leading to an increase rather than a decrease in intracellular calcium concentration with associated inhibition of PTH secretion [435]. Verapamil has been reported to inhibit PTH secretion from rat parathyroid glands in vitro [437] and from goat parathyroid glands perfused in vivo [434]. In conflicting in vivo animal studies, verapamil stimulated PTH secretion in rats [438,439], whereas 1,25(OH)2D3 levels were decreased [438]. Clinical studies, however, have shown no change in PTH levels using either verapamil [440] or nifedipine [441]. Only a short-term (three-day) study with diltiazem reported decreased PTH levels with normal ionized calcium and phosphate concentrations [435]. Studies evaluating vitamin D metabolites showed that 16 weeks of diltiazem administration had no effect on either PTH or 1,25(OH)2D3 levels [442]. Long-term use

1269

of nifedipine also failed to alter PTH or 25OHD3 levels or to affect serum parameters of bone turnover or bone mineral density in a group of males treated with calcium channel blockers for coronary heart disease [433]. Finally, calcium channel blockers are known inhibitors of hepatic microsomal cytochrome P450-dependent enzymes, and thus have the potential to cause alterations in vitamin D metabolites [434]. Despite possible interactions with PTH release, all past and recent clinical data have failed to demonstrate any effect on vitamin D metabolites, and thus calcium channel blockers cannot be considered to have a major influence on vitamin D metabolism.

I. Heparin Long-term use of heparin has been associated with the development of osteopenia in humans [435–438] and in rats [439–441]. Chronic heparin use is most frequently encountered in pregnant women with venous thrombosis. Pronounced bone loss and low 1,25(OH)2D3 levels have been seen in such patients, while serum 25OHD3 and 24,25(OH)2D3 levels and calcium and phosphate concentrations have been unchanged [442,443]. Heparin decreases bone formation in cultured fetal rat calvaria [444,445] and stimulates bone resorption by increasing the number and activity of osteoclasts in vitro [446]. Mutoh et al. [439] treated four-week-old vitamin D–deficient rats with heparin (2000 lU/day). Significant bone loss developed after two weeks, which peaked at four weeks. No change in serum total or ionized calcium was observed, but a significant elevation of serum PTH was seen. Furthermore, 1,25(OH)2D3 levels were decreased by 54% versus controls [439]. Although in the Mutoh et al. study no change in serum ionized calcium was demonstrated, this is presumed to be the mechanism by which heparin increases serum PTH. Heparin has been reported to have a high affinity for calcium ions [447]. This may lead to lower calcium levels, which would stimulate PTH release. However, this is a doubtful mechanism for heparin-induced osteopenia, as calcium salts of heparin are as effective in inducing osteoporosis as the corresponding sodium salts [448]. The reasons for the low 1,25(OH)2D3 values with heparin are entirely speculative, but may involve direct inhibition of the 1α-hydroxylase system, with the low 1,25(OH)2D3 levels in turn influencing receptors on the parathyroid gland, which induces PTH stimulation. The use of low molecular weight heparins (LMWH) in the treatment of thrombotic disorders has steadily grown over the past decade. The long-term effect of

1270 these agents on bone is yet to be determined. In vitro data suggest that LMWHs inhibit osteoblast growth [449]. Delayed fracture repair has also been reported in rabbits treated with the LMWH, enoxaparin [450]. A prospective study of 16 pregnant women who received 19–32 weeks of enoxaparin revealed a significant decrease in bone mineral density of the proximal femur at six months postpartum [451]. The effect of LMWH on vitamin D metabolism has not been studied.

SOL EPSTEIN AND ADINA E. SCHNEIDER

ranitidine, have less effect on hepatic drug metabolism, [461] and thus are less likely to interfere with vitamin D metabolism. No studies employing ranitidine or other H2 receptor antagonists and their effects on vitamin D metabolism are currently available. Data on the effect of cimetidine on bone density and fracture incidence has either not been explored or is unknown.

K. Aluminum J. Cimetidine Cimetidine, a histamine H2 receptor antagonist, is frequently used for the treatment of peptic ulcer disease. In vitro studies have demonstrated histamine H2 receptors in both normal and adenomatous parathyroid gland tissue [452], and stimulation of these receptors by histamine increases PTH release [453]. Cimetidine has been shown to decrease serum PTH in patients with either parathyroid adenoma [454,455] or secondary hyperparathyroidism due to chronic renal insufficiency [456,457]. However, there have been equivocal findings regarding the effects on serum calcium accompanying the changes in PTH, with both low [454] and unchanged levels [455,457,458] being found. Cimetidine was also shown to significantly decrease net intestinal calcium transport either secondary to its effect on PTH or via changes in vitamin D [459]. Cimetidine is an inhibitor of microsomal drug metabolism [460,461], and thus one might expect this agent to inhibit hepatic vitamin D 25-hydroxylase, a cytochrome P450-dependent enzyme. Animal studies have shown a dose-dependent decrease in 25-hydroxylase activity in the presence of increasing concentrations of cimetidine [462]. Cimetidine was also shown to reduce 25OHD3 levels in hens [463]. In humans, short-term use of cimetidine (800 mg/day for four weeks) did not decrease the level of 25OHD3 but prevented the expected seasonal rise in 25OHD3 [404]. After cessation of cimetidine therapy, levels rose significantly. Levels of 1,25(OH)2D3 and 24,25(OH)2D3 were not affected, and serum calcium and phosphate concentrations remained normal [464]. A review of the literature up till the present shows that cimetidine has multiple effects on calciotropic hormones. PTH levels are decreased, and 25OHD3 is reduced by inhibition of hepatic 25-hydroxylase activity. Despite low PTH levels, however, no changes in serum 1,25(OH)2D3 or serum calcium or phosphate concentrations are evident. Monitoring of 25OHD3 levels is indicated only in susceptible individuals, such as those with hepatic insufficiency, poor nutrition, or the elderly. Other H2 receptor antagonists, such as

Chronic use of dialysate or total parenteral nutrition (TPN) containing aluminum is known to cause metabolic bone disease. Aluminum containing TPN has also been associated with low levels of 1,25(OH)2D3 [466]. Casein hydrolysate, which is used as a protein source in some TPN solutions, was identified as containing substantial amounts of aluminum [467]. These patients also develop a low-turnover osteomalacia with aluminum accumulation in bone [468]. Patients undergoing hemodialysis with water containing high levels of aluminum also have a high incidence of aluminum bone disease [469,470]. Aluminum content in bone is elevated and correlates positively with the development of osteomalacia [471–475]. Patients with renal failure are also susceptible to aluminum accumulation in bone [474,476]. There is some evidence that aluminum may act indirectly on bone by suppressing PTH release. Hemodialysis patients with osteomalacia have demonstrated lower mean PTH values than hemodialysis patients with normal 25OHD3 levels [473,475]. In rats, aluminum was shown to accumulate in parathyroid tissue [477], and reports show that aluminum impairs PTH release in vitro [478]. This may lead to low 1,25(OH)2D3 production. High-dose aluminum injections have also caused osteomalacia [474,479] and lowered PTH levels in rats [479]. However, in another similar experiment in rats using equivalent doses of aluminum, no skeletal changes or alterations in serum vitamin D metabolites were observed [480]. In dogs, serum calcium increased and serum phosphate and PTH did not change significantly following five weeks of parenteral aluminum administration daily [481]. 25OHD3 was normal, but a marked decline in 1,25(OH)2D3 was demonstrated. Renal function also declined, and this may account for the changes in 1,25(OH)2D3. However, the reduction in 1,25(OH)2D3 occurred prior to the appearance of renal impairment, suggesting a direct inhibitory effect on the synthesis of 1,25(OH)2D3 [482]. Opposite effects on vitamin D metabolites seem to occur following the oral ingestion of aluminum hydroxide as an antacid. Increased values of 1,25(OH)2D3 may occur with hypophosphatemia induced by oral

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

aluminum salts, which in turn increases 1α-hydroxylase activity in the kidney [483]. Aluminum is absorbed from the gut and deposits in bone in patients both with [484,485] and without [486] renal impairment. Daylong intake by a group of postmenopausal women resulted in a fall in serum phosphate, which correlated significantly with a rise in 1,25(OH)2D3 levels. Total and ionized calcium and PTH levels were unchanged [487]. This study was very short in duration, however, and longer studies would be required to assess the effects of oral aluminum on vitamin D metabolism. Nevertheless, chronic use of high doses of aluminum containing antacids has been reported to induce osteomalacia and rickets [488–431].

L. Antituberculous Agents Anecdotal case reports of rifampicin (rifampin)induced osteomalacia [492] led to further investigation of the effect of antituberculous agents on vitamin D metabolism. Studies by Brodie and colleagues [493] initially suggested that the frequently-used first-line antituberculous drugs rifampicin and isoniazid, alone or in combination, could affect vitamin D metabolism. In the first study, short-term use of rifampicin (600 mg/day for two weeks) in healthy subjects reduced plasma 25OHD3 levels by as much as 70%, whereas 1,25(OH)2D3 and PTH remained unchanged [493]. Rifampicin is a known hepatic enzyme-inducer [494], and the enzyme 25-hydroxylase is a cytochrome P450dependent enzyme located in the liver [495]. It is most likely, therefore, that the decreased levels of 25OHD3 represent increased hepatic metabolism of 25OHD3 by 25-hydroxylase [493]. Short-term use of isoniazid (300 mg/day) in healthy subjects also produced a decrease in 25OHD3 and 1,25OH2D3 levels, accompanied by a fall in serum calcium and phosphate and rise in PTH [344]. Inhibition by isoniazid of hepatic enzyme activity [496] could explain the decreased levels of 25OHD3 and 1,25(OH)2D3, as hepatic 25-hydroxylase and renal 1α-hydroxylase are both cytochrome P450-dependent enzyme systems [497]. A further study using both rifampicin and isoniazid also led to decreased 25OHD3 and 1,25(OH)2D3 levels together with raised PTH [498]. Similar short-term effects of rifampicin and isoniazid have been reported by others [499,500]. Despite evidence of short-term derangements in vitamin D metabolism, it seems that long-term studies reveal no significant effects [501,502]. One study reported that treatment of tuberculous patients with both rifampicin and isoniazid for nine months produced no significant alterations in 25OHD3 and l,25(OH)2D3

1271

levels [502]. A more recent study found that tuberculous patients had low baseline 25OHD3, and PTH levels and high urinary calcium and 1,25(OH)2D3 levels, presumably due to granulomatous synthesis of 1α-hydroxylase. After treatment with isoniazid and rifampin for nine months, these parameters normalized [503]. Thus, overall it seems that only in the short-term are patients treated with these agents susceptible to vitamin D deficiency. Patients in developing countries with insufficient vitamin D intake and others at risk for vitamin D deficiency may require vitamin D supplementation, but it is unlikely that antituberculous drugs contribute to the development of overt metabolic bone disease. In fact, as suggested by Martinez et al. [503], long-term antituberculous therapy may lead to a correction of the abnormalities of vitamin D and calcium metabolism seen in these patients as it treats the underlying condition.

M. Caffeine The first mention of a potential effect of caffeine on bone metabolism is found in a study by Daniell [504], who noted a higher caffeine intake in osteoporotic patients compared to age-matched controls. Heaney and Recker [505], in studies of perimenopausal women, were the first to show a negative calcium balance in association with caffeine intake. Additional controlled trials of caffeine use indicate that in individuals ingesting inadequate dietary calcium, caffeine ingestion leads to a small negative calcium balance via a small but significant decrease in calcium absorption efficiency [506]. Barger-Lux and Heaney have estimated that this small effect on calcium balance can be offset by the addition of 1–2 tablespoons of milk per cup of coffee [507]. The more recent literature examining the effects of caffeine on the skeleton is conflicting. The Framingham Osteoporosis Study failed to show an affect of caffeine intake on bone mass in elderly men and women [508]. Rico et al. failed to show a relationship between caffeine intake on bone mass as measured by quantitiative phalangeal bone ultrasound in 93 healthy postmenopausal women. [509]. Similarly, Lloyd et al. [510] did not find an association between dietary caffeine intake and total body or hip BMD. However, another study found that elderly women who consumed > 300 mg/d of caffeine had higher rates of bone loss than women with an intake of < 300 mg/d [511]. These conflicting reports may be due to differences in determination of caffeine intake and methods of assessing bone density [509]. Caffeine has been reported to enhance hepatic microsomal drug metabolism in rats and mice [512,513].

1272 An inhibitory effect on the conversion of 25OHD3 to 1,25(OH)2D3 was reported in isolated renal tubules from vitamin D–deficient chicks [514]. Studies in rats following chronic caffeine administration showed normal serum calcium but increased urinary excretion, and intestinal endogenous excretion of calcium, as well, increased intestinal absorption of calcium [515]. Yeh and Aloia [516] studied serial changes of serum calcium, PTH, 1,25(OH)2D3, and calcium balance in young and old adult rats following daily caffeine administration for four weeks. In young rats, urinary calcium excretion increased and serum calcium decreased initially, but then returned to control levels. Serum PTH and 1,25(OH)2D3 increased after two weeks, and intestinal absorption of calcium remained unchanged. In adult rats similar changes occurred except that 1,25(OH)2D3 levels were similar to those in controls [516]. One drawback to this study is that the caffeine content that these rats received is the equivalent of 16 cups of coffee each day, which would be rather excessive for human consumption. A further study demonstrated no effect on bone histomorphometry after administration of caffeine to rats [517]. Thus, caffeine seems to have some effect on calcium homeostasis in humans and animals. Its impact is likely more substantial in patients who are already at nutritional risk, such as the elderly. A clear effect on vitamin D metabolism has not been demonstrated.

N. Theophylline Theophylline, a once common treatment for reactive airway disease, has been shown to have an impact on vitamin D metabolism. An in vitro study showed an inhibitory effect of theophylline on the conversion of 25OHD3 to 1,25(OH)2D3 [514]. This occurred despite an increase in renal tubule cAMP levels, which have been shown to enhance 1,25(OH)2D3 formation in vivo [19]. Furthermore, there are reports of enhanced 24-hydroxylase activity by aminophylline in normal birds [518] and rats [519]. In rat studies, the effect of long-term constant subcutaneous theophylline infusion was assessed. Increased urinary calcium excretion was demonstrated together with a reduction in total body calcium. Serum 25OHD3 was decreased, but no changes in 1,25(OH)2D3 or PTH levels were observed [520]. One possible explanation for the effects on 25OHD3 is that theophylline, an agent that has been previously demonstrated to induce hepatic microsomal enzymes [513,521], enhances conversion of 25OHD3 to other metabolites. In humans, theophylline increased phosphate and calcium urinary excretion in healthy males [522].

SOL EPSTEIN AND ADINA E. SCHNEIDER

McPherson et al. [523] found that patients hospitalized for theophylline toxicity had hypercalcemia that normalized after theophylline was discontinued. PTH levels were unchanged, which implies that theophylline may act by enhancing the action of available PTH. Although animal data suggest that theophylline may alter vitamin D metabolism and clinical studies demonstrate a change in calcium balance, a direct effect on vitamin D has not been demonstrated in humans.

O. Immunosuppressants Potent immunosuppressants, which are widely used in the management of transplant patients and patients suffering from autoimmune conditions, have altered the prognosis of these disorders. This section will not address the role of vitamin D as an immune modulator (see Chapter 36) or its use in immune related disorders (see Chapters 98 and 99). The T-cell-specific immunosuppressant cyclosporin A (CsA) produces a high turnover osteopenia in the rat [524–526]. In both the rat and mouse, CsA has been shown to stimulate 1,25(OH)2D3 production in the absence of any changes in serum-ionized calcium, phosphate, or PTH levels [525–527]. A significant increase in 24-hydroxylase activity has been seen in kidney homogenates from rats following 14 days of oral CsA treatment (15 mg/kg). Furthermore, a significant increase in 1α-hydroxylase activity was demonstrated in mice treated with 30–50 mg/kg CsA for three days, which was shown to be due to renal and not extrarenal stimulation of 1α-hydroxylase [527]. There was no evidence of increased metabolic clearance of 1,25(OH)2D3 in CsA-treated rats [527]. Young rats responded to CsA with greater 1,25(OH)2D3 production than older ones [528], and 1,25(OH)2D3 levels normalized following cessation of CsA treatment [526]. Grenet et al. [529] recently reported that CaA induced an 85% decrease in the vitamin D–binding proteins, calbindin –D28k, a 40% decrease in the vitamin D receptor and a 69% decrease in the 24-hydroxylase enzyme, while 1,25(OH)2D3 was increased. These findings provide further support for the evidence that CsA significantly impairs the vitamin D activation pathway. Studies in Rowett-nude T-cell-deficient rats also showed a similar increase in 1,25(OH)2D3 production in response to treatment with CsA. This suggests that CsA affects vitamin D independent of the presence of T lymphocytes [530]. Cyclosporin G (CsG), an equipotent immunosuppressive molecular analog of CsA that is less nephrotoxic, [531,532] also increased 1,25(OH)2D3 levels in rats, independent of changes in

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

ionized calcium or PTH levels, but to a lesser degree than CsA [533–535]. The fungal macrolide tacrolimus (FK506) is another potent immunosuppressive that has a similar action to CsA [536]. In the rat, FK506 appears as deleterious as CsA, in that it too produces severe high turnover osteopenia; however, unlike CsA, it has not been shown to have an effect on 1,25(OH)2D3 production [533,535]. Azathioprine, a thioguanine derivative of mercaptopurine, acts as a purine antagonist and is an effective antiproliferative agent. It, too, failed to influence 1,25(OH)2D3, PTH, or ionized calcium levels in the rat [537]. Finally, administration to rats of another immunosuppressant, rapamycin (sirolimus), caused either no change [538] or low 1,25(OH)2D3 levels [539] with normal PTH or phosphate and ionized calcium concentrations. No reports of effects on vitamin D by any of the other immunosuppressants are known. Calcineurin inhibitors are most widely used in the posttransplantation setting and thus clinical data on their effect on vitamin D metabolism are difficult to interpret, as the underlying disease process and renal impairment induced by these agents interfere with vitamin D metabolism. Moreover, multiple drugs including corticosteroids, which inhibit 1,25(OH)2D3, are given concurrently. A recent prospective study of 57 patients post kidney transplant found a high prevalence osteomalacia on bone biopsy despite normal levels of calcidiol and calcitriol. The authors theorize that these findings suggest a state of vitamin D resistance [540]. In addition, posttransplantation, the use of CsA [541–545], or azathioprine [543–548] does not appear to result in any changes in 1,25(OH)2D3 levels compared to control subjects. These agents have also been evaluated in the nontransplant setting. Long-term CsA treatment of patients with multiple sclerosis revealed similar 1,25(OH)2D3, 25OHD3, ionized calcium, and phosphate levels compared to those treated with azathioprine, although the latter group had lower PTH levels [549]. Patients with primary biliary cirrhosis treated for one year with CsA also showed no evidence of any changes in vitamin D metabolites or PTH levels [550]. Thus, it would seem that only experimentally, in rats and mice, does CsA affect vitamin D and that immunosuppressants, collectively, do not significantly affect vitamin D metabolism in the clinical setting. Nevertheless, the use of calcitriol has been shown to be beneficial in patients receiving these agents. A recent prospective study of the effects of calcitriol in 53 patients post heart transplantation on high-dose tacrolimus, found, at 12 months, that femoral neck BMD stabilized and lumbar spine BMD increased

1273

significantly in patients treated with calcitriol, while femoral neck BMD decreased and spine BMD stabilized in placebo-treated subjects. Both groups received calcium supplementation and treatment of hypogonadism [551]. Thus, because of the overall effect on bone loss, it is advocated that patients treated with calcineurin inhibitors be supplemented with vitamin D analogs and oral daily calcium [552,553].

P. Fluoride Fluoride is no longer used as a treatment for osteoporosis; however, it is commonly found in drinking water at varying concentrations. Although studies in rats have found reduced calcium and phosphate levels when fluoride was added to drinking water at the 2.0 mM level, no changes in either serum 25OHD3 or 1,25(OH)2D3 concentrations were reported [554]. In human studies, no changes in serum PTH, 1,25(OH)2D3, 25OHD3, or calcium and phosphate levels were found after one to two years of fluoride treatment for osteoporosis in both men and women [555,556]. Thus, these data suggest that fluoride does not interfere with vitamin D metabolism.

Q. Olestra Olestra, formerly known as sucrose polyester (SPE), is a nonabsorbable mixture of hexa-, hepta-, and octa-carbon fatty acid esters of sucrose. It is an edible material that can be incorporated into the diet as a fat substitute. Olestra has physical properties similar to those of conventional dietary fats [557,558]; however, it is not absorbed [559] or hydrolyzed by gastric lipases [560]. As dietary vitamin D is absorbed from the intestine in association with dietary fats [561] and has an enterohepatic circulation, these processes may be altered by the presence of nondigestible lipid. Although there have been reports of reduced absorption of the fat-soluble vitamins A and E [557,558,562,563], all human studies have thus far shown no significant effects on serum 25OHD levels [562–565] or dietary vitamin D absorption [564]. A recent post-marketing surveillance study did not show an association between olestra consumption and concentrations of fat-soluble vitamins [566]. A 20-month feeding study in dogs also showed no effects on vitamin D status following Olestra ingestion [567]. Since photoinduced cutaneous synthesis of vitamin D is the major factor determining vitamin D status [568–570], Olestra ingestion would not be expected to adversely affect vitamin D nutritional status.

1274 R. Orlistat Orlistat is a gastric and pancreatic lipase inhibitor, which is widely used in the management of obesity. Given its inhibition of absorption of lipids from the gastrointestinal tract, it has the potential to inhibit the absorption of fat soluble vitamins. A small prospective study found a significant decrease in vitamin D levels after one month of therapy with orlistat in adolescents, despite the use of a multivitamin [571]. However, a randomized, double-blinded study found no significant difference in vitamin D levels between orlistat users and controls [572]. Additional studies examining the effect of this agent on vitamin D metabolism are warranted. In the meantime, it may be prudent to monitor vitamin D levels in patients using this drug.

S. Lithium Lithium is a monovalent cation and is widely used in the management of bipolar affective disorders. It is well known to have an array of endocrine-related side effects, including the alteration of systemic calcium metabolism. In humans, lithium has been shown to lower serum phosphate [573–575] and reduce urinary calcium excretion [575–580]. This occurs as a result of increased tubular reabsorption of calcium, and it results in hypercalcemia [576–578,581–585]. Increases in PTH levels and parathyroid volume have been attributed to lithium therapy [574,575,577,581–585,587]. Drug withdrawal reverses these effects [506]. In vitro evidence has shown that lithium stimulates the release of PTH from human parathyroid tissue [512]. Parathyroid adenoma has been reported in several patients with lithium-induced hyperparathyroidism [588], although another possible mechanism by which lithium may elevate PTH is secondary to lithium-induced nephropathy [589,590]. However, serum creatinine levels were found to be normal in a patient with increased PTH [506]. Renal tubular acidosis has also been associated with lithium use [586]. There is limited knowledge about the effects of lithium on vitamin D as most metabolic studies following lithium administration in humans and animals have not measured vitamin D metabolites. One study found elevated PTH levels, yet normal 1,25(OH)2D3 levels after either short-term (mean, 1.7 months) or longterm (mean, 103 months) lithium carbonate administration [508]. Another study of 10 patients treated for one month with lithium carbonate noted elevated serum PTH and reduced 1,25(OH)2D3 levels, although 25OHD3 and serum calcium levels remained unchanged [591]. This is surprising as the elevation of

SOL EPSTEIN AND ADINA E. SCHNEIDER

PTH levels would be expected to increase rather than decrease 1,25(OH)2D3 levels. The authors hypothesize that lithium may act by inhibiting renal 1α-hydroxylase. The sample size and mean serum lithium levels were similar in both of the above studies; thus, the conflicting observations may relate to differences in the patient populations. Although lithium use can affect serum calcium, phosphate, and PTH levels, a direct effect on vitamin D metabolism has not been demonstrated thus far.

IV. CONCLUSION The effects of various endogenous and exogenous hormones and drugs on the metabolism of vitamin D are complex. PTH and PTHrP are clearly potent stimulators of 1,25(OH)2D3 production. Past and current literature suggest that growth hormone, via IGF-1, and estrogen increase production of 1,25(OH)2D3. Longterm use of most anticonvulsants leads to increased 25OHD3 catabolism which is likely the main cause of the low bone mass observed in these patients. A strong association exists between the use of corticosteroids and immunosuppressants and the development of significant bone loss; however, the effect of these agents on vitamin D metabolism in humans is less clear. Although statins do not appear to effect vitamin D metabolism, long term use of cholestyramine may decrease 25OHD3 levels. Parenteral aluminum appears to decrease 1,25(OH)2D3 levels, while its oral use results in hypophosphatemia and increased 1,25(OH)2D3 levels. The evidence on the effect of ethanol and commonly-used agents such as heparin, lithium, theophylline, and cimetidine on vitamin D metabolism is inconclusive. More careful screening of susceptible individuals, such as those with poor nutritional status or who lack exposure to ultraviolet light, is likely indicated. Further attention must be directed towards new drugs, such as ezetimibe and low molecular weight heparin, which have the potential to influence vitamin D absorption and metabolism, respectively.

References 1. Garabedian M, Holick MF, DeLuca HF, Boyle IT 1976 Control of 25-hydroxycholecalciferol metabolism by parathyroid glands. Proc Natl Acad Sci USA 69:1673–1676. 2. Fraser DR, Kodicek E 1973 Regulation of 25-hydroxycholecalciferol-1-hydroxylase activity in kidney by parathyroid hormone. Nature (New Biol) 241:163–166. 3. Henry HL, Midgett RJ, Norman AW 1974 Regulation of 25-hydroxyvitamin D3-l-hydroxylase in vivo. J Biol Chem 249:7584–7592.

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

4. Henry HL 1979 Regulation of the hydroxylation of 25-hydroxyvitamin D3 in vivo and in primary cultures of chick kidney cells. J Biol Chem 254:2722–2729. 5. Booth BE, Tsai HC, Morris RC Jr 1977 Parathyroidectomy reduces 25-hydroxyvitamin D3-1a-hydroxylase activity in the hypocalcemic vitamin D–deficient chick. J Clin Invest 60:1314–1320. 6. Mitlak BH, Williams DC, Bryant HU, Paul DC, Neer RM 1992 Intermittent administration of bovine PTH(l-34) increases serum 1,25-dihydroxyvitamin D concentrations and spinal bone density in senile (23 month) rats. J Bone Miner Res 7:479–484. 7. Shigematsu T, Horiuchi N, Ogura Y, Miyahara T, Suda T 1986 Human parathyroid hormone inhibits renal 24-hydroxylase activity of 25-hydroxyvitamin D3 by a mechanism involving adenosine 3′,5′-monophosphate in rats. Endocrinology 118:1583–1589. 8. Rasmussen H, Wong M, Bikle D, Goodman DBP 1972 Hormonal control of the renal conversion of 25-hydroxycholecalciferol to 1,25-dihydroxycholecalciferol. J Clin Invest 51:2502–2504. 9. Larkins RG, MacAuley SJ, Rapoport A, Mertin TJ, Tulloch BR, Byfield PGH, Matthews EW, Maclntyre I 1974 Effects of nucleotides, hormones, ions, and 1,25-dihydroxycholecalciferol on 1,25-dihydroxycholecalciferol production in isolated chick renal tubules. Clin Sci Mol Med 46:569–582. 10. Trechsel U, Bonjour J-P, Fleisch H 1979 Regulation of the metabolism of 25-hydroxyvitamin D3 in primary cultures of chick kidney cells. J Clin Invest 64:206–217. 11. Juan D, DeLuca HF 1977 The regulation of 24,25-dihydroxyvitamin D3 production in cultures of monkey kidney cells. Endocrinology 101:1184–1193. 12. Armbrecht HJ, Wongsurawat N, Zenser T, Davis BB 1982 Differential effects of parathyroid hormone on the renal 1,25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3 production of young and adult rats. Endocrinology 111: 1339–1344. 13. Armbrecht HJ, Wongsurawat N, Zenser TV, Davis BB 1984 Effect of PTH and 1,25(OH)2D3 on renal 25(OH)D3 metabolism, adenylate cyclase, and protein kinase. Am J Physiol 246:E102–E107. 14. Rost CR, Bickle DD, Kaplan RA 1981 In vitro stimulation of 25-hydroxycholecalciferol 1a-hydroxylation by parathyroid hormone in chick kidney slices: Evidence for a role for adenosine 3′,5′-monophosphate. Endocrinology 108: 1002–1006. 15. Fukase M, Birge SJ Jr, Rifas L, Avioli LV, Chase LR 1982 Regulation of 25 hydroxyvitamin D3 1-hydroxylase in serumfree monolayer culture of mouse kidney. Endocrinology 110:1073–1075. 16. Murer H, Werner A, Reshkin S, Wuarin F, Biber J 1991 Cellular mechanisms in proximal tubular reabsorption of phosphate. Am J Physiol 260:C885–C899. 17. Ro H-K, Tembe V, Krug T, Yang P-YJ, Bushinsky DA, Favus MJ 1990 Acidosis inhibits 1,25(OH)2D3 but not cAMP production in response to parathyroid hormone in the rat. J Bone Miner Res 5:273–278. 18. Chase LR, Aurbach GD 1967 Parathyroid function and the renal excretion of 3′,5′-adenylic acid. Proc Natl Acad Sci 58:518–525. 19. Horiuchi N, Suda T, Takahashi H, Shimazawa E, Ogata E 1977 In vivo evidence for the intermediary role of 3′,5′-cyclic AMP in parathyroid hormone-induced stimulation of 1a,25dihydroxyvitamin D3 synthesis in rats. Endocrinology 101:969–974.

1275

20. Armbrecht HJ, Forte LR, Wongsurawat N, Zenser TV, Davis BB 1984 Forskolin increases 1,25-dihydroxyvitamin D3 production by rat renal slices in vitro. Endocrinology 114: 644–649. 21. Henry HL 1985 Parathyroid hormone modulation of 25-hydroxyvitamin D3 metabolism by cultured chick kidney cells is mimicked and enhanced by forskolin. Endocrinology 116:503–510. 22. Seamon KB, Daly JW 1982 Forskolin: A unique diterpene activator of cyclic AMP generating systems. J Cyclic Nucleotide Res 71:201. 23. Kong XF, Zhu KH, Peo YL, Jackson DM, Holcih MF 1999 Molecular cloning, characterization, and promotor analysis of human 25-hydroxyvitamin D3-1-alpha-hydroxylase gene. PNAS 36:6988–6993. 24. Murayama A, Takeyama K, Kitanaka S, Kodera Y, Kawaguchi Y, Hosoya T, Kato S 1999 Positive and negative regulations of the renal 25-hydroxyvitamin D3 1-alphahydroxylase gene by parathyroid hormone, calcitonin and 1 alpha,25(OH)2D3 in intact animals. Endocrinology 140:2224–2231. 25. Shinki T, Jin CH, Nishimura A, Nagai Y, Ohyama Y, Noshiro M, Okuda K, Suda T 1992 Parathyroid hormone inhibits 25-hydroxyvitamin D3-24-hydroxylase mRNA expression stimulated by 1a,25-dihydroxyvitamin D3 in rat kidney but not in intestine. J Biol Chem 267:13757–13762. 26. Zierold C, Mings JA, DeLuca HF 2001 Parathyroid hormone regulates 25-hydroxyvitamin D3-24-hydroxylase mRNA by altering its stability. PNAS 98:13572–13576. 27. Janulis M, Tembe V, Favus MJ 1992 Role of protein kinase C in parathyroid hormone stimulation of renal 1,25-dihydroxyvitamin D3 secretion. J Clin Invest 90:2278–2283. 28. Ro HK, Tembe V, Favus MJ 1992 Evidence that activation of protein kinase-C can stimulate 1,25-dihydroxyvitamin D3 secretion by rat proximal tubules. Endocrinology 131: 1424–1428. 29. Friedlander J, Janulis M, Tembe V, Ro HK, Wong M-S, Favus MJ 1994 Loss of parathyroid hormone-stimulated 1,25dihydroxyvitamin D3 production in aging does not involve protein kinase A or C pathways. J Bone Miner Res 9:339–345. 30. Kaplan RK, Haussler MR, Deftos LJ, Bone H, Pak CYC 1977 The role of 1a,25-dihydroxyvitamin D in the mediation of intestinal hyperabsorption of calcium in primary hyperparathyroidism and absorptive hypercalciuria. J Clin Invest 59:756–760. 31. Mawer EB, Backhouse J, Hill LF, Lumb GA, De Silva P, Taylor CM, Stanbury SW 1975 Vitamin D metabolism and parathyroid function in man. Clin Sci Mol Med 48:349–365. 32. Brown DJ, Spanos E, Maclntyre I 1980 Role of pituitary hormone in regulating renal vitamin D metabolism in man. Br Med J 1:277–278. 33. Nakayama K, Fukumoto S, Takeda S, Takeuchi Y, Ishikawa T, Miura M, Hata K, Hane M, Tamura Y, Tanaka Y, Kitaoka M, Obara T, Ogata E, Matsumoto T 1996 Differences in bone and vitamin D metabolism between primary hyperparathyroidism and malignancy-associated hypercalcemia. J Clin Endocrinol Metab 81:607–611. 34. Bilezikian JP, Canfield RE, Jacobs TP, Polay JS, D’Adamo AP, Eisman JA, DeLuca HF 1978 Response of 1a,25-dihydroxyvitamin D3 to hypocalcemia in human subjects. N Engl J Med 299:437–441. 35. Rosen JF, Fleischman AR, Finberg L, Eisman J, DeLuca HF 1977 1,25-Dihydroxycholecalciferol: Its use in the long-term management of idiopathic hypoparathyroidism in children. J Clin Endocrinol Metab 45:457–468.

1276 36. Kooh SW, Fraser D, DeLuca HF, Holick MF, Belsey RE, Clark MB, Murray TM 1975 Treatment of hypoparathyroidism and pseudohypoparathyroidism with metabolites of vitamin D: Evidence for impaired conversion 25-hydroxyvitamin D to 1a,25-dihydroxyvitamin D. N Engl J Med 293:840–844. 37. Davies M, Taylor CM, Hill LF, Stanbury SW 1977 1,25Dihydroxycholecalciferol in hypoparathyroidism. Lancet 1:55–59. 38. Drezner MK, Neelon FA, Haussler M, McPherson HT, Lebovitz HE 1976 1,25-Dihydroxycholecalciferol deficiency: The probable cause of hypocalcemia and metabolic bone disease in pseudohypoparathyroidism. J Clin Endocrinol Metab 42:621–628. 39. Brancaccio D, Cozzolino M, Gorio A, Di Giulio AM, Gallieni M 2002 Bone disease in uremic patients: advances in PTH suppression. J Nephrol 15(Suppl 6):S86–S93. 40. Bilezikian JP 1990 Parathyroid hormone-related peptide in sickness and in health. N Engl J Med 322:1151–1153. 41. Orloff JJ, Kats Y, Urena P, Schipani E, Vasavada RC, Philbrick WM, Behal A, Abou-Samra A-B, Segre GV, Juppner H 1995 Further evidence for a novel receptor for amino-terminal parathyroid hormone-related protein on keratinocytes and squamous carcinoma cell lines. Endocrinology 136:3016–3023. 42. Strewler GJ, Stern PH, Jacobs JW, Eveloff J, Klein RF, Leung SC, Rosenblatt M, Nissenson RA 1987 Parathyroid hormone-like protein from human renal carcinoma cells. Structural and functional homology with parathyroid hormone. J Clin Invest 80:1803–1807. 43. Yates AJP, Gutierrez GE, Smolens P, Travis PS, Katz MS, Auf-demorte TB, Boyce BF, Hymer TK, Poser JW, Mundy GR 1988 Effects of a synthetic peptide of a parathyroid hormonerelated protein on calcium homeostasis, renal tubular calcium reabsorption, and bone metabolism in vivo and in vitro in rodents. J Clin Invest 81:932–938. 44. Stewart AF, Horst R, Deftos LJ, Cadman EC, Lang R, Broadus AE 1980 Biochemical evaluation of patients with cancer-associated hypercalcemia. N Engl J Med 303: 1377–1383. 45. Ralston SH, Cowan RA, Robertson RA, Gardner MD, Boyle IT 1984 Circulating vitamin D metabolites and hypercalcemia of malignancy. Acta Endocrinol 106:556–563. 46. Syed MA, Horowitz MJ, Tedesco MB, Garcia-Ocana A, Wisniewksi SR, Stewart AF 2001 Parathyroid hormone related protein (1-36) stimulates renal tubular reabsorption in normal human volunteers: implications for the pathogenesis of humoral hypercalcemia of malignancy. J Clin Endocrinol Metab 55:219–227. 47. Horowitz MJ, Tedesco MB, Sereika SM, Hollis BW, GarciaOcana A, Stewart AF 2003 Direct comparison of sustained infusion of human parathyroid-related protein versus hPTH on serum calcium, plasma 1,25 dihydroxyvitamin D concentration and calcium excretion in healthy volunteers. J Clin Endocrinol Metab 88:1603–1609. 48. Stewler GJ 2000 The physiology of parathyroid hormonerelated protein. N Engl J Med 342:177–185. 49. Insogna KL, Stewart AF, Vignery AMC, Weir EC, Namnum PA, Baron RE, Kirkwood JM, Deftos LM, Broadus AE 1984 Biochemical and histomorphometric characterization of a rat model for humoral hypercalcemia of malignancy. Endocrinology 114:888–896. 50. Horiuchi N, Caulfield MP, Fisher JE, Goldman ME, McKee RL, Reagan JE, Levy JJ, Nutt RF, Rodan SB, Schofield TL,

SOL EPSTEIN AND ADINA E. SCHNEIDER

51.

52.

53.

54.

55.

56.

57.

58. 59. 60.

61. 62. 63.

64.

Clemens TL, Rosenblatt M 1987 Similarity of synthetic peptide from human tumor to parathyroid hormone in vivo and in vitro. Science 238:1566–1568. Walker AT, Stewart AF, Korn EA, Shiratori T, Mitnick MA, Carpenter TO 1990 Effect of parathyroid hormone-like peptides on 25-hydroxyvitamin D-1a-hydroxylase activity in rodents. Am J Physiol 258:E297–E303. Michigami T, Yamato H, Suzuki H, Nagai-Itagaki Y, Sato K 2001 Conflicting actions of parathyroid hormone-related protein and serum calcium as regulators of 25-hydroxyvitamin D3-1α-hydroxylase expression in a nude rat model of humoral hypercalcemia of malignancy. J Endocrinol 171:249–257. Fukumoto S, Matsumoto T, Yamoto H, Kawashima H, Ueyama Y, Tamaoki N, Ogata E 1989 Suppression of serum 1,25-dihydroxyvitamin D in humoral hypercalcemia of malignancy is caused by an elaboration of a factor that inhibits renal 1,25-dihydroxyvitamin D3 production. Endocrinology 124:2057–2062. Schilling T, Pecherstorfer M, Blind E, Leidig G, Ziegler R, Raue F 1993 Parathyroid hormone-related protein (PTHrP) does not regulate 1,25-dihydroxyvitamin D serum levels in hypercalcemia of malignancy. J Clin Endocrinol Metab 76:801–803. Everhart-Caye M, Inzucchi SE, Guinness-Henry J, Mitnick MA, Stewart AF 1996 Parathyroid hormone (PTH)related protein(l-36) is equipotent to PTH(l-34) in humans. J Clin Endocrinol Metab 81:199–208. Fraher LJ, Hodsman AB, Jonas K, Saunders D, Rose CI, Henderson JE, Hendy GN, Goltzman D 1992 A comparison of the in vivo biochemical responses to exogenous parathyroid hormone-(1-34) [PTH-(1-34)] and PTHrelated peptide-(l-34) in man. J Clin Endocrinol Metab 75: 417–423. Horowitz MJ, Tedesco MB, Sereika SM, Hollis BW, GarciaOcana A, Stewart AF 2003 Direct comparison of sustained infusion of human parathyroid-related protein versus hPTH on serum calcium, plasma 1,25 dihydroxyvitmain D concentration and calcium excretion in healthy volunteers. J Clin Endocrinol Metab 88:1603–1609. Mundy GR 1988 Hypercalcemia of malignancy revisited. J Clin Invest 82:1–6. Kovac CS, Kronenberg HM 1997 Maternal-fetal calcium and bone metabolism during pregnancy, puerperium, and lactation. Endocr Rev 18:832–872. Mather KJ, Chik CL, Corenblum B 1999 Maintenance of serum calcium by parathyroid-related peptide during lactation in a hypoparathyroid patient. J Clin Endocr Metab 84:424–426. Galante L, Colston KW, Macauley SJ, Maclntyre I 1972 Effect of calcitonin on vitamin D metabolism. Nature 238:271–273. Lorenc R, Tanaka Y, DeLuca HF, Jones G 1977 Lack of effect of calcitonin on the regulation of vitamin D metabolism in the rat. Endocrinology 100:468–472. Horiuchi N, Takahashi H, Matsumoto T, Takahashi N, Shimazawa E, Suda T, Ogata E 1979 Salmon calcitonininduced stimulation of 1a,25-dihydroxycholecalciferol synthesis in rats involving a mechanism independent of adenosine 3′,5′-cyclic monophosphate. Biochem J 184:269–275. Jaeger P, Jones W, Clemens TL, Hayslett JP 1986 Evidence that calcitonin stimulates 1,25-dihydroxyvitamin D production and intestinal absorption of calcium in vivo. J Clin Invest 78:456–461.

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

65. Shinki T, Ueno Y, DeLuca HF, Suda T 1999 Calcitonin is a major regulator for the expression of renal 25-hydroxyvitamin D3-1alpha-hydroxylase gene in normocalcemic rats. Proc Natl Acad Sci USA 96:8253–8258. 66. Beckman MJ, Goff JP, Reinhardt TA, Beitz DC, Horst RL 1994 In vivo regulation of rat intestinal 24-hydroxylase: Potential new role of calcitonin. Endocrinology 135: 1951–1955. 67. Armbrecht HJ, Wongsurawat N, Paschal RE 1987 Effect of age on renal responsiveness to parathyroid hormone and calcitonin in rats. J Endocrinol 114:173–178. 68. Kawashima H, Torikai S, Kurokawa K 1981 Calcitonin selectively stimulates 25-hydroxyvitamin D3-1a-hydroxylase in proximal straight tubule of rat kidney. Nature 291: 327–329. 69. Emmertsen K, Melsen F, Mosekilde L, Lund B, Lund B, Sorensen OH, Nielsen HE, Solling H, Hansen HH 1981 Altered vitamin D metabolism and bone remodeling in patients with medullary thyroid carcinoma and hypercalcitonemia. Metab Bone Dis Related Res 4:17–23. 70. Aloia JF, Vaswani A, Kapoor A, Yeh JK, Cohn SH 1985 Treatment of osteoporosis with calcitonin, with and without growth hormone. Metabolism 34:124–129. 71. Hartwell D, Hassager C, Overgaard K, Riis BJ, Podenphant J, Christiansen C 1990 Vitamin D metabolism in osteoporotic women during treatment with estrogen, and anabolic steroid, or calcitonin. Acta Endocrinol 122:715–721. 72. Thamsborg G, Jensen JEB, Kollerup G, Hauge EM, Melsen F, Sorensen OH 1996 Effect of nasal salmon calcitonin on bone remodeling and bone mass in postmenopausal women. Bone 18:207–212. 73. Thamsborg G, Storm TL, Daugaard H, Schiffer S, Sorensen OH 1991 Circulating levels of calciotropic hormones during treatment with nasal salmon calcitonin. Acta Endocrinol 125:127–131. 74. Nunziata V, Giannattasio R, Di Giovanni G, Lettera AM, Nunziata CA 1993 Vitamin D status in Paget’s bone disease. Effect of calcitonin therapy. Clin Orthop Related Res 293:366–371. 75. Devlin RD, Gutteridge DH, Prince RL, Retallack RW, Worth GK 1990 Alterations in vitamin D metabolites during treatment of Paget’s disease of bone with calcitonin or etidronate. J Bone Miner Res 5:1121–1126. 76. Mainoya JR 1975 Effects of bovine growth hormone, human placental lactogen, and ovine prolactin on intestinal fluid and ion transport in the rat. Endocrinology 96:1165. 77. Henneman PH, Forbes AP, Moldawer M, Dimpsey EF, Carrol EL 1962 Effects of human growth hormone in man. J Clin Invest 39:1223. 78. Finkelstein JD, Schachter D 1962 Active transport of calcium by intestine: Effects of hypophysectomy and growth hormone. Am J Physiol 203:873. 79. Spanos E, Barret D, Maclntyre I, Pike JW, Safilian EF, Haussler MR 1978 Effect of growth hormone on vitamin D metabolism. Nature 273:246–247. 80. Fontaine O, Pavlovitch H, Balsan S 1978 25-Hydroxycholecalciferol metabolism in hypophysectomized rats. Endocrinology 102:1822–1826. 81. Spencer EM, Tobiassen O 1981 The mechanism of the action of growth hormone on vitamin D metabolism in the rat. Endocrinology 108:1064–1070. 82. Pahuja DN, DeLuca HF 1981 Role of the hypophysis in the regulation of vitamin D metabolism. Mol Cell Endocrinol 23:345–350.

1277

83. Gray RW, Garthwaite TL 1985 Activation of renal 1,25-dihydroxyvitamin D3 synthesis by phosphate deprivation: Evidence for a role for growth hormone. Endocrinology 116:189–193. 84. Denis I, Thomasset M, Pointillart A 1994 Influence of exogenous porcine growth hormone on vitamin D metabolism and calcium and phosphorus absorption in intact pigs. Calcif Tissue Int 54:489–492. 84. Wu S, Grieff M, Brown AJ 1997 Regulation of renal vitamin D-24-hydroxylase by phosphate: effects of hypophysectomy, growth hormone, and insulin-like growth factor I. Biochem Biophys Res Commun; 233:813–817. 85. Eskildsen PC, Lund B, Sorensen OH, Lund B, Bishop JE, Norman AW 1979 Acromegaly and vitamin D metabolism: Effect of bromocriptine treatment. Endocrinology 49:484–486. 85. Lund B, Eskildsen PC, Lund B, Norman AW, Sorensen OH 1981 Calcium and vitamin D metabolism in acromegaly. Acta Endocrinol 96:444–450. 86. Bijlsma JWJ, Nortier JWR, Duursma SA, Croughs RJM, Bosch R, Thijssen JHH 1983 Changes in bone metabolism during treatment of acromegaly. Acta Endocrinol 104:153–159. 87. Brixen K, Nielsen HK, Bouillon R, Flyvberg A, Mosekilde L 1992 Effects of short-term growth hormone treatment on PTH, calcitriol, thyroid hormones, insulin, and glucagon. Acta Endocrinol 127:331–336. 88. Marcus R, Butterfield G, Holloway L, Gilliland L, Baylink DJ, Hintz RL, Sherman BM 1990 Effects of short-term administration of recombinant human growth hormone to elderly people. J Clin Endocrinol Metab 70:519–527. 89. Harbison MD, Gertner JM 1990 Permissive action of growth hormone on the renal response to dietary phosphorus deprivation. J Clin Endocrinol Metab 70:1035–1040. 90. Chipman JJ, Zerwekh J, Nicar M, Marks J, Pak CYC 1980 Effect of growth hormone administration: Reciprocal changes in serum 1a,25-dihydroxyvitamin D and intestinal calcium absorption. J Clin Endocrinol Metab 51:321–324. 91. Burstein S, Chen IW, Tsang RC 1983 Effects of growth hormone replacement therapy on 1,25-dihydroxyvitamin D and calcium metabolism. J Clin Endocrinol Metab 56:1246–1251. 92. Gertner JM, Horst RL, Broadus AE, Rasmussen H, Genel M 1979 Parathyroid function and vitamin D metabolism during human growth hormone replacement. J Clin Endocrinol Metab 49:185–188. 93. Panczyk-Tomaszewska M, Ziolkowska H, Debinski A, Sieniawska M 2000 Vitamin D metabolite requirements in dialysed children receiving recombinant human growth hormone. Nephrol Dial Transplant 15:375–378. 94. Beshyah SA, Thomas E, Kyd P, Sharp P, Fairney A, Johnson DG 1994 The effect of growth hormone replacement therapy in hypopituitary adults on calcium and bone metabolism. Clin Endocrinol 40:383–391. 95. Wei S, Tanaka H, Kubo T, Ono T, Kanzaki S, Seino Y 1997 Growth hormone increases 1,25 dihydroxyvitamin D levels and decreases 24,25-dihyroxyvitamin D levels in children with growth hormone deficiency. Eur J Endocrinol 136:30–32. 96. Bianda T, Hussain MA, Glatz Y, Bouillon R, Froesch ER, Schmid C 1997 Effects of short-term insulin-like growth factor–1 or growth hormone on bone turnover, renal phosphate reabsorption, and 1,25 dihydroxyvitamin D3 production in healthy man. J Intern Med 241:143–150. 97. Wright NM, Papadea N, Wentz B, Hollis B, Bell NH 1997 Increased serum 1,24 dihydroxyvitamin D after growth hormone administration is not parathyroid hormone medicated. Calcif Tissue Int 61:101–103.

1278 98. Bickle DD, Spencer EM, Burke WH, Rost CR 1980 Prolactin but not growth hormone stimulates 1,25-dihydroxyvitamin D3 production by chick renal preparations in vitro. Endocrinology 107:81–84. 99. Hammerman MR, Gavin III JR 1986 Binding of IGF-I and IGF-I-stimulated phosphorylation in canine renal basolateral membrane. Am J Physiol 251:E32–E41. 100. Condamine L, Vztovsnik F, Friedlander G, Menaa C, Garabedian M 1994 Local action of phosphate depletion and insulin-like growth factor 1 on in vitro production of 1,25-dihydroxyvitamin D by cultured mammalian kidney cells. J Clin Invest 94:1673–1679. 101. Halloran BP, Spencer EM 1988 Dietary phosphate and 1,25-dihydroxyvitamin D metabolism: Influence of insulinlike growth factor 1. Endocrinology 123:1225–1229. 102. Caverzasio J, Motessuit C, Bonjour JP 1990 Stimulatory effect of insulin-like growth factor-1 on renal Pi transport and plasma 1,25-dihydroxyvitamin D3. Endocrinology 127: 453–459. 103. Gray RW 1987 Evidence that somatomedins mediate the effect of hypophosphatemia to increase serum 1,25-dihydroxyvitamin D3 levels in rats. Endocrinology 121:504–512. 104. Nesbitt T, Drezner MK 1993 Insulin-like growth factor-1 regulation of renal 25-hydroxyvitamin D-1-hydroxylase activity. Endocrinology 132:133–138. 105. Spanos E, Pike JW, Haussler MR, Colston KW, Evans IMA, Goldner AM, McCain TA, Maclntyre I 1976 Circulating 1a,25-dihydroxyvitamin D in the chicken: Enhancement by injection of prolactin and during egg laying. Life Sci 19:1751–1756. 106. Spanos E, Brown DJ, Stevenson JC, Maclntyre I 1981 Stimulation of 1,25-dihydroxycholecalciferol production by prolactin and related peptides in intact renal cell preparations in vitro. Biochim Biophys Acta 672:7–15. 107. Spanos E, Colston KW, Evans IMS, Galante LS, Macauley SJ, Maclntyre I 1976 Effect of prolactin on vitamin D metabolism. Mol Cell Endocrinol 5:163–167. 108. Matsumoto T, Horiuchi N, Suda T, Takahashi H, Shimazawa E, Ogata E 1979 Failure to demonstrate stimulatory effect of prolactin on vitamin D metabolism in vitamin-D–deficient rats. Metabolism 28:925–927. 109. Kenny AD, Baksi SN, Galli-Galardo SM 1977 Vitamin D metabolism in amphibia and fish. Fed Proc 36:1097. 110. Mainoya JR 1975 Further studies on the action of prolactin on fluid and ion absorption by the rat jejunum. Endocrinology 96:1158–1164. 111. Reichlin S 1974 Neuroendocrinology: Lactation. In: William RH (ed) Textbook of Endocrinology. Saunders: Philadelphia, Pennsylvania, pp. 800–801. 112. Boass A, Toverud SA, McCain TA, Pike JW, Haussler MR 1977 Elevated serum levels of 1a,25-dihydroxycholecalciferol in lactating rats. Nature 267:630–632. 113. Maclntyre I, Colston KW, Szelke M, Spanos E 1978 A survey of the hormonal factors that control calcium metabolism. Ann NY Acad Sci 307:345–355. 114. Kumar R, Cohen WR, Silva P, Epstein FH 1979 Elevated 1,25-dihydroxyvitamin D plasma levels in normal human pregnancy and lactation. J Clin Invest 63:342–344. 115. Lund B, Seines A 1979 Plasma 1,25-dihydroxyvitamin D levels in pregnancy and lactation. Acta Endocrinol 92:330–335. 116. DeSantiago S, Alonso L, Halhali A, Larrea F, Isiard F, Bourges H 2002 Negative calcium balance during lactation in rural Mexican women. Am J Clin Nutr 76: 845–851.

SOL EPSTEIN AND ADINA E. SCHNEIDER

117. Cundy T, Raining SA, Guilland-Cumming DF, Butler J, Kanis JA 1987 Remission of hypoparathyroidism during lactation: Evidence for a physiological role for prolactin in the regulation of vitamin D metabolism. Clin Endocrinol 26:667–674. 118. Sowers M, Zhang D, Hollis BW, Shapiro B, Janney CA, Crutchfield M, Schork MA, Stanczyk F, Randolph J 1998 Role of calcitropic hormones in calcium mobilization of lactation. Am J Clin Nutr 67:284–291. 119. Ardawi MS, Nasrat HA, BA’Aqueel HS 1997 Calciumregulating hormones and parathyroid-related peptide in normal human pregnancy and postpartum. A longitudinal study. Eur J Endocrinol 137:402–409. 120. Kumar R, Abboud CF, Riggs BL 1980 The effect of elevated prolactin levels on plasma 1,25-dihydroxyvitamin D and intestinal absorption of calcium. Mayo Clin Proc 55:51–53. 121. Adams ND, Garthwaite TL, Gray RW, Hagen TC, Lemann J Jr 1979 The interrelationships among prolactin, 1,25-dihydroxyvitamin D, and parathyroid hormone in humans. J Clin Endocrinol Metab 49:628–630. 122. Klibanski A, Neer R, Beitins I, Ridgeway E, McArthur J 1981 Decreased bone density in hyperprolactinemic women. N Engl J Med 303:1511. 123. Schlechte JA, Sherman B, Martin R 1983 Bone density in amenorrheic women with and without hyperprolactinemia. J Clin Endocrinol Metab 56:1120–1123. 124. Koppelman MCS, Kurtz DW, Morrish KA, Bou E, Susser JK, Shapiro JR, Loriaux DL 1984 Vertebral body bone mineral content in hyperprolactinemic women. J Clin Endocrinol Metab; 59:1050–1053. 125. Bouillon R 1991 Diabetic bone disease. Calcif Tissue Int 49:155–160. 126. Schneider LE, Schedl HP 1972 Diabetes and intestinal calcium absorption in the rat. Am J Physiol 223:1319–1323. 127. Schneider LE, Wilson HD, Schedl HP 1974 Effects of alloxan diabetes on duodenal calcium-binding protein in the rat. Am J Physiol 227:832–838. 128. Schneider LE, Omdahl EJ, Schedl HP 1976 Effects of vitamin D and its metabolites on calcium transport in the diabetic rat. Endocrinology 99:793–799. 129. Schneider LE, Nowosielski LM, Schedl HP 1977 Insulintreatment of diabetic rats: Effects on duodenal calcium absorption. Endocrinology 100:67–73. 130. Schneider LE, Schedl HP, McCain T, Haussler MR 1977 Experimental diabetes reduces circulating 1,25-dihydroxyvitamin D in the rat. Science 196:1452–1454. 131. Schneider LE, Hargis EK, Schedl HP, Williams GA 1974 Parathyroid function in the alloxan diabetic rat. Endocrinology 95:749. 132. Schedl HP, Heath H, Wenger J 1978 Serum calcitonin and parathyroid hormone in experimental diabetes: Effects of insulin treatment. Endocrinology 103:1368–1373. 133. Wongsurawat N, Armbrecht HJ, Zenser TV, Davis BB, Thomas ML, Forte LR 1983 1,25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3 production by isolated renal slices is modulated by diabetes and insulin in the rat. Diabetes 32:302–306. 134. Glajchen N, Epstein S, Thomas S, Fallen M, Chakrabarti S 1988 Bone mineral metabolism in experimental diabetes mellitus: Osteocalcin as a measure of bone remodeling. Endocrinology 123:290–295. 135. Shires R, Teitelbaum SL, Bergfeld MA, Fallen MD, Slatopolsky E, Avioli LV 1981 The effect of streptozotocininduced chronic diabetes mellitus on bone and mineral homeostasis in the rat. J Clin Lab Med 97:231–240.

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

136. Spencer EM, Khalil M, Tobiassen O 1980 Experimental diabetes in the rat causes an insulin-reversible decrease in renal 25-hydroxyvitamin D3-1a-hydroxylase activity. Endocrinology 107:300–305. 137. Takeshita N, Yoshino T, Mutoh S, Yamaguchi I 1994 Possible involvement of vitamin D3-deficiency and relatively enhanced bone resorption in the development of bone loss in streptozotocin-induced diabetic rats. Life Sci 55:291–299. 138. Hough S, Fausto A, Sonn Y, Dong Jo OK, Birge SJ, Avioli LV 1983 Vitamin D metabolism in the chronic streptozotocininduced diabetic rat. Endocrinology 113:790–796. 139. Epstein S, Takizawa M, Stein B, Katz IA, Joffe II, Romero DF, Liang XG, Ke HZ, Jee WSS, Jacobs TW, Berlin J 1994 The effect of cyclosporin A on bone mineral metabolism in experimental diabetes mellitus in the rat. J Bone Miner Res 9:557–566. 140. Sulimovici S, Roginsky MS 1980 Hepatic metabolism of vitamin D3 in streptozotocin-induced diabetic rat. Acta Endocrinol 93:346–350. 141. Matsumoto T, Kawanobe Y, Ezawa I, Shibuya N, Hata K, Ogata E 1986 Role of insulin in the increase in serum 1,25-dihydroxyvitamin D concentrations in response to phosphorus deprivation in streptozotocin-induced diabetic rats. Endocrinology 118:1440–1444. 142. Nyomba B, Bouillon R, Lissens W, Van Baelen H, De Moor P 1985 1,25-Dihydroxyvitamin D and vitamin D–binding protein are both decreased in streptozotocin-diabetic rats. Endocrinology 116:2483–2488. 143. Henry HL 1981 Insulin permits parathyroid hormone stimulation of 1,25-dihydroxyvitamin D3 production in cultured kidney cells. Endocrinology 108:733–735. 144. Sulimovici S, Roginsky MS, Susser F 1981 Nephrogenous cyclic AMP in streptozotocin-induced diabetic rat. Biochem Biophys Res Commun 100:471–477. 145. Ambrecht HJ, Wongsurawat VJ, Hodam TL, Wongsurawat N 1996 Insulin markedly potentiates the capacity of parathyroid hormone to increase expression of 25-hydroxyvitamin D3 24-hydroxylase in rat osteoblastic cells in the presence of 1,25 dihydroxyvitamin D3. FEBS Lett 393:77–80. 146. Gertner J, Horst R, Tamborlane W 1979 Mineral metabolism and vitamin D status in juvenile diabetics: Changes following normalization of plasma glucose with a portable infusion pump. Diabetes 28:354. 147. Frazer TE, White NH, Hough S, Santiago JV, McGee BR, Bryce G, Mallon J, Avioli LV 1981 Alterations in circulating vitamin D metabolites in the young insulin–dependent diabetic. J Clin Endocrinol Metab 53:1154–1159. 148. Nyomba BL, Bouillon R, Bidingija M, Kandjingu K, De Moor P 1986 Vitamin D metabolites and their binding protein in adult diabetic patients. Diabetes 35:911–915. 149. Heath III H, Lambert PW, Service FJ, Arnaud SB 1979 Calcium homeostasis in diabetes mellitus. J Clin Endocrinol Metab 49:462–466. 150. Kenny AD 1976 Vitamin D metabolism: Physiological regulation in egg-laying Japanese quail. Am J Physiol 230:1609–1615. 151. Baksi SN, Kenny AD 1977 Vitamin D3 metabolism in immature Japanese quail: Effect of ovarian hormones. Endocrinology 101:1216–1220. 152. Castillo L, Tanaka Y, DeLuca HF, Sunde ML 1977 The stimulation of 25-hydroxyvitamin D3-1-a-hydroxylase by estrogen. Arch Biochem Biophys 179:211–217. 153. Tanaka Y, Castillo L, Wineland MJ, DeLuca HF 1978 Synergistic effect of progesterone, testosterone, and estradiol

154.

155. 156. 157. 158. 159.

160. 161.

162. 163.

164. 165.

166.

167. 168. 169.

170.

1279 in the stimulation of chick renal 25-hydroxyvitamin D3-1ahydroxylase. Endocrinology 103:2035–2039. Pike JW, Spanos E, Colston KW, Maclntyre I, Haussler MR 1978 Influence of estrogen on renal vitamin D hydroxylases and serum 1a,25(OH)2D3 in chicks. Am J Physiol 235: E338–E343. Trechsel U, Bonjour J-P, Fleisch H 1979 Regulation of the metabolism of 25-hydroxyvitamin D3 in primary cultures of chick kidney cells. J Clin Invest 64:206–217. Henry HL 1981 25(OH)D3 metabolism in kidney cell cultures: Lack of direct effect of estradiol. Am J Physiol 240:E119–E124. Baksi SN, Kenny AD 1978 Does estradiol stimulate in vivo production of 1,25-dihydroxyvitamin D3 in the rat. Life Sci 22:787–792. Kalu DN, Liu CC, Hardin RR, Hollis BW 1989 The aged rat model of ovarian hormone deficiency bone loss. Endocrinology 124:7–16. Kalu DN, Salerno E, Liu CC, Echon R, Ray M, Garza-Zapata M, Hollis BW 1991 A comparative study of the actions of tamoxifen, estrogen, and progesterone in the oophorectomized rat. Bone Miner 15:109–124. Arjmandi BH, Hollis BW, Kalu DN 1994 In vivo effect of 17/3-estradiol on intestinal calcium absorption in rats. Bone Miner 26:181–189. Colin EM, Van Den Bemd GJ, Van Aken M Christakos S, De Jonge HR, Deluca HF, Prahl JM, Birkenhager JC, Buurman CJ, Pols HA, Van Leeuwen JP 1999 Evidence for involvement of 17 beta-estradiol in intestinal calcium absorption independent of 1,25 dihydroxyvitamin D3 level in the rat. J Bone Miner Res 14:57–64. Criddle RA, Zheng MH, Dick IM, Callus B, Prince RL 1997 Estrogen responsiveness of renal calbindin-D28k gene expression in rat kidney. J Cell Biochem 65:340–348. Liel Y, Shany S, Smirnoff P, Schwartz B 1999 Estrogen increases 1,25 dihydroxyvitamin D receptors expression and bioresponse in the rat duodenal mucosa. Endocrinology 140: 280–285. Ash SL, Goldin BR 1988 Effects of age and estrogen on renal vitamin D metabolism in the female rat. Am J Clin Nutr 47:694–699. Ismail F, Epstein S, Fallon MD, Thomas SB, Reinhardt TA 1988 Serum bone gla protein and the vitamin D endocrine system in the oophorectomized rat. Endocrinology 122:624–630. Duque G, Abdaimi KE, Macoritto M, Miller MM, Kremer R 2002 Estrogens regulate expression and response of 1,25dihydroxyvitamin D3 receptors in bone cells. Changes with aging and hormone deprivation. Biochem Biophys Res Commun 299:446–454. Turton CWG, Stamp TCB, Stanley P, Maxwell JD 1977 Altered vitamin D metabolism in pregnancy. Lancet 1:222–224. Harris SS, Dawson-Hughes B 1998 The association of oral contraceptive use with plasma 25-hydroxyvitamin D levels. J Am Coll Nutr 17:282–284. Bickle DD, Gee E, Halloran B, Haddad JG 1984 Free 1,25dihydroxyvitamin D levels in serum from normal subjects, pregnant subjects, and subjects with liver disease. J Clin Invest 74:1966–1971. Gallagher JC, Riggs BL, Eisman J, Hamstra A, Arnaud SB, DeLuca HF 1979 Intestinal calcium absorption and serum vitamin D metabolites in normal subjects and osteoporotic patients. J Clin Invest 64:729–736.

1280 171. Lund B, Sorensen OH, Lund B, Agner E 1982 Serum 1,25dihydroxyvitamin D in normal subjects and in patients with post-menopausal osteopenia. Influence of age, renal function, and oestrogen therapy. Horm Metab Res 14:271–274. 172. Falch JA, Oftebro H, Haug E 1987 Early postmenopausal bone loss is not associated with a decrease in circulating levels of 25-hydroxyvitamin D, 1,25-dihydroxyvitamin D, or vitamin D–binding protein. J Clin Endocrinol Metab 64:836–841. 173. Gallagher JC 1990 The pathogenesis of osteoporosis. Bone Miner 9:215–227. 174. Wishart JM, Scopacasa F, Horowitz M, Morris HA, Need AG, Clifton PM, Nordin BE 2000 Effect of perimenopause on calcium absorption: a longitudinal study. Climacteric 3:102–108. 175. Finkelstein JS, Scheonfeld DA 1999 Effects of gonadal suppression on the regulation of parathyroid hormone and 1,25 dihydroxyvitamin D secretion in women. J Clin Endocrinol Metab 84:2151–2161. 176. Lore F, Di Cairano G, Signorini AM, Caniggia A 1981 Serum levels of 25-hydroxyvitamin D in postmenopausal osteoporosis. Calcif Tissue Int 33:467–470. 177. van Hoof HJC, van der Mooren MJ, Swinkels LMJW, Rolland R, Benraad THJ 1994 Hormone replacement therapy increases serum 1,25-dihydroxyvitamin D: A two-year prospective study. Calcif Tissue Int 55:417–419. 178. van Hoff HJ, van der Mooren MJ, Swinkels LM, Sweep CG, Merkus JM, Benraad TJ 1999 Female sex hormone replacement therapy increases free 1,25 dihydroxyvitamin D3: a one-year perspective study. Clin Endocrinol 50:511–516. 179. Jeikkinen A, Parviainen MT, Tuppurainen MT, Niskanen L, Komulainen MH, Saarikoski S 1998 Effects of postmenopausal hormone replacement therapy with and without vitamin D3 on circulating levels of 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D. Calcif Tissue Int 62:26–30. 180. Gallagher JC, Riggs BL, DeLuca HF 1980 Effect of estrogen on calcium absorption and serum vitamin D metabolites in postmenopausal osteoporosis. J Clin Endocrinol Metab 51:1359–1364. 181. Stock JL, Coderre JA, Mallette LE 1985 Effects of short course of estrogen on mineral metabolism in postmenopausal women. J Clin Endocrinol Metab 61:595–600. 182. Bouillon R, Van Assche FA, Van Baelen H, Heyns W, De Moor P 1981 Influence of the vitamin D–binding protein on the serum concentration of 1,25-dihydroxyvitamin D3. Significance of the free 1,25-dihydroxyvitamin D3 concentration. J Clin Invest 67:589–596. 183. Selby PL, Peacock M, Barkworth SA, Brown WB, Taylor GA 1985 Early effects of ethinyloestradiol and norethisterone treatment in postmenopausal women on bone resorption and calcium hormones. Clin Sci 69:265–271. 184. Cheesma C, Grant BF, Marcus R 1989 Effects of estrogen on circulating “free” and total 1,25-dihydroxyvitamin D and on the parathyroid-vitamin D axis in postmenopausal women. J Clin Invest 83:537–542. 185. Vanderschueren D, Van Herck E, Suiker AMH, Visser WJ, Schot LPC, Bouillon R 1992 Bone and mineral metabolism in aged male rats: Short- and long-term effects of androgen deficiency. Endocrinology 130(No. 5):2906–2916. 186. Bouillon R, Vandoren G, Van Baelen H, De Moor P 1978 Immunochemical measurement of the vitamin D-binding protein in rat serum. Endocrinology 102:1710–1715. 187. Nyomba BL, Bouillon R, DeMoor P 1987 Evidence for an interaction of insulin and sex steroids in the regulation of vitamin D metabolism in the rat. J Endocrinol 115:295–301.

SOL EPSTEIN AND ADINA E. SCHNEIDER

188. Hagenfeldt Y, Eriksson H, Bjorkhem I 1989 Stimulatory effect of testosterone on renal 25-hydroxyvitamin D3 1a-hydroxylase in guinea pig. Biochim Biophys Acta 1002:84–88. 189. Otremski I, Lev-Ran M, Salama R, Edelstein S 1997 The metabolism of vitamin D3 in response to testosterone. Calcif Tissue Int 60:485–487. 190. Ohata M, Sakagami Y, Fujita T 1977 Elevation of serum 25-hydroxycalciferol levels in androgen-treated and ultravioletirradiated rats. Endocrinol Jpn 24:519–521. 191. Hagenfeldt Y, Linde K, Sjoberg HE, Zumkeller W, Arver S 1992 Testosterone increases serum 1,25-dihydroxyvitamin D and insulin-like growth factor-1 in hypogonadal men. Int J Androl 15:93–102. 192. Morley JE, Perry III HM, Kaiser FE, Kraenzle D, Jensen J, Houston K, Mattammal M, Perry HM Jr 1993 Effects of testosterone replacement therapy in old hypogonadal males: A preliminary study. J Am Geriatr Soc 41:149–152. 193. Jackson JA, Kleerekoper M, Parfitt M, Sudhaker Rao D, Villanueva AR, Frame B 1987 Bone histomorphometry in hypogonadal and eugonadal men with spinal osteoporosis. J Clin Endocrinol Metab 65:53–58. 194. Finkelstein JS, Klibanski A, Neer RM, Doppelt SH, Rosenthal DI, Segre GV, Crowley WFJ 1989 Increase in bone density during treatment of men with idiopathic hypogonadotropic hypogonadism. J Clin Endocrinol Metab 69(No. 4):776–783. 195. Francis RM, Peacock M, Aaron JE, Selby PL, Taylor GA, Thompson J, Marshall DH, Horsman A 1986 Osteoporosis in hypogonadal men: Role of decreased plasma 1,25dihydroxyvitamin D, calcium malabsorption, and low bone formation. Bone 7:261–268. 196. Hagenfeldt Y, Carlstrom K, Berlin T, Stege R 1991 Effects of orchidectomy and different modes of high-dose estrogen treatment on circulating “free” and total 1,25-dihydroxyvitamin D in patients with prostatic cancer. J Steroid Biochem Mol Biol 39:155–159. 197. Krabbe S, Hummer L, Christiansen C 1986 Serum levels of vitamin D metabolites and testosterone in male puberty. J Clin Endocrinol Metab 62:503–507. 198. Aksnes L, Aarskog D 1982 Plasma concentrations of vitamin D metabolites in puberty: Effect of sexual maturation and implications for growth. J Clin Endocrinol Metab 55:94–101. 199. Leman ES, DeMiguel F, Gao AC, Getzenberg RH 2003 Regulation of androgen and vitamin D receptors by 1,25dihydroxyvitamin D3 in human prostate epithelia and stromal cell. J Urol 170: 235–240. 200. Ingles SA, Ross RK, Yu MC, Irvine RA, Pera G, Haile RW 1997 Association of prostate cancer risk with genetic polymorphisms of the vitamin D receptor and androgen receptor. J Natl Cancer Inst 89:166–170. 201. Bickle DD, Halloran BP, Harris ST, Potale AA 1992 Progestin antagonism of estrogen stimulated 1,25-dihydroxyvitamin D levels. J Clin Endocrinol Metab 72:519–523. 202. Grecu EO, Simmons R, Baylink DJ, Haloran BP, Spencer ME 1991 Effects of medroxyprogesterone acetate on some parameters of calcium metabolism in patients with glucocorticoidinduced osteoporosis. Bone Miner 13:153–161. 203. Auwerx J, Bouillon R 1986 Mineral and bone metabolism in thyroid disease: A review. Q J Med 60:737–752. 204. Mundy GR, Shapiro JL, Bandelin JG, Canalis EM, Raisz LG 1976 Direct stimulation of bone resorption by thyroid hormones. J Clin Invest 58:529–534. 205. Mosekilde L, Melsen F, Bagger JP, Myhre-Jensen O, Sorensen NS 1977 Bone changes in hyperthyroidism: Interrelationships between bone morphometry, thyroid

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

206. 207.

208. 209. 210. 211.

212. 213. 214. 215. 216. 217. 218.

219.

220. 221. 222.

223. 224. 225.

function, and calcium-phosphorus metabolism. Acta Endocrinol 85:515–525. Bayley TA, Harrison JE, McNeill KG, Mernagh JR 1980 Effect of thyrotoxicosis and its treatment on bone mineral and muscle mass. J Clin Endocrinol Metab 50:916–922. Mosekilde L, Christensen MS 1977 Decreased parathyroid function in hyperthyroidism: Interrelationships between serum parathyroid hormone, calcium-phosphorus metabolism, and thyroid function. Acta Endocrinol 84:566–575. Burman KD, Monchik JM, Earll JM, Wartofsky L 1976 Ionized and total serum calcium and parathyroid hormone in hyperthyroidism. Ann Int Med 84:668–671. Bouillon R, De Moor P 1973 Parathyroid function in patients with hyper- or hypothyroidism. J Clin Endocrinol Metab 38:999–1004. Shafer RB, Gregory DH 1972 Calcium malabsorption in hyperthyroidism. Gastroenterology 63:235–239. Peerenboom H, Keck E, Kruskemper HL, Strohmeyer G 1984 The defect of intestinal calcium transport in hyperthyroidism and its response to therapy. J Clin Endocrinol Metab 59:936–940. Karsenty G, Bouchard P, Ulmann A, Schaison G 1985 Elevated metabolic clearance rate of l,25-dihydroxyvitamin D3 in hyperthyroidism. Acta Endocrinol 110:70–74. Bouillon R, Muls E, De Moor P 1980 Influence of thyroid function on the serum concentration of 1,25-dihydroxyvitamin D3. J Clin Endocrinol Metab 51:793–797. Jastrup B, Mosekilde L, Melsen F, Lund B, Lund B, Sorensen OH 1982 Serum levels of vitamin D metabolites and bone remodeling in hyperthyroidism. Metabolism 31:126–132. Macfarlane IA, Mawer EB, Berry J, Hann J 1982 Vitamin D metabolism in hyperthyroidism. Clin Endocrinol 17: 51–59. Velentzas C, Oreopoulos DG, From G, Porret B, Rapoport A 1977 Vitamin D levels in thyrotoxicosis. Lancet 1:370–371. Mosekilde L, Lund B, Sorensen OH, Christensen MS, Melsen F 1977 Serum-25-hydroxycholecalciferol in hyperthyroidism. Lancet 1:806–807. Yamashita H, Noguchi S, Murakami T, et al. 2000 Calcium and its regulating hormones in patients with Graves disease: sex differences and relation to postoperative tetany. Eur J Surg 166: 924–8. Langdahl BL, Loft AG, Eriksen EF, Mosekilde L, Charles P 1996 Bone mass, bone turnover, calcium homeostasis, and body composition in surgically and radioiodine-treated former hyperthyroid patients. Thyroid 6:169–175. Krane SM, Brownell GL, Stanbury JB, Corrigan H 1956 The effect of thyroid disease on calcium metabolism in man. J Clin Invest 35:874–897. Bijlsma JWJ, Duursma SA, Roelofs JMM, der Kinderen PJ 1983 Thyroid function and bone turnover. Acta Endocrinol 104:42–49. Weisman Y, Eisenberg Z, Lubelski R, Spirer Z, Edelstein S, Hare 11 A 1981 Decreased 1,25-dihydroxycholecalciferol and increased 25-hydroxy- and 24,25-dihydroxycholecalciferol in tissues of rats treated with thyroxine. Calcif Tissue Int 33:445–447. Pahuja DN, De Luca HF 1982 Thyroid hormone and vitamin D metabolism in the rat. Arch Biochem Biophys 213:293–298. Kano K, Jones G 1984 Direct in vitro effect of thyroid hormones on 25-hydroxyvitamin D3 metabolism in the perfused rat kidney. Endocrinology 114:330–336. Miller ML, Ghazarian JG 1981 Differential response of kidney mitochondrial calcium-regulating mixed functional oxidase to thyrotoxicosis. J Biol Chem 256:5643–5645.

1281

226. Trechsel U, Taylor CM, Bonjour J-P, Fleisch H 1980 Influence of prostaglandins and of cyclic nucleotides on the metabolism of 25-hydroxyvitamin D3 in primary chick kidney cell culture. Biochem Biophys Res Commun 93: 1210–1216. 227. Biddulph DM, Currie MG, Wrenn RW 1979 Effects of interactions of parathyroid hormone and prostaglandins on adenosine 3′,5′-monophosphate concentrations in isolated renal tubules. Endocrinology 104:1164–1171. 228. Wark JD, Larkins RG, Eisman JA, Wilson KR 1981 Regulation of 25-hydroxyvitamin D-1a-hydroxylase in chick-isolated renal tubules: Effects of prostaglandin E2, frusemide and acetylsalicyclic acid. Clin Sci 61:53–59. 229. Kurose H, Sonn YM, Jafari A, Birge SJ, Avioli LV 1985 Effects of prostaglandin E2 and indomethacin on 25-hydroxyvitamin D3-1a-hydroxylase activity in isolated kidney cells of normal and streptozotocin-induced diabetic rats. Calcif Tissue Int 37:625–629. 230. Yamada M, Matsumoto T, Takahashi N, Suda T, Ogata E 1983 Stimulatory effect of prostaglandin E2 on 1a,25-dihydroxyvitamin D3 synthesis in rats. Biochem J 216:237–240. 231. Yamada M, Matsumoto T, Su K-W, Ogata E 1985 Inhibition of prostaglandin E2 of renal effects of calcitonin in rats. Endocrinology 116:693–697. 232. Sedrani SH, El-Banna AA 1987 Effect of indomethacin on plasma levels of vitamin D metabolites, oestradiol, and progesterone in rabbits during early pregnancy. Comp Biochem Physiol 87A:635–639. 233. Katz IA, Jee WSS, Joffe II, Stein B, Takizawa M, Jacobs TW, Setterberg R, Lin BY, Tang LY, Ke HZ, Zeng QQ, Berlin JA, Epstein S 1992 Prostaglandin E2 alleviates cyclosporin Ainduced bone loss in the rat. J Bone Miner Res 7:1191–1200. 234. Boiskin I, Epstein S, Ismail F, Fallen MD, Levy W 1988 Long-term administration of prostaglandin inhibitors in vivo fail to influence cartilage and bone mineral metabolism in the rat. Bone Miner 4:27–36. 235. de Rovetto CR, Welch TR, Hug G, Clark KE, Bergstrom W 1989 Hypercalciuria with Bartter syndrome: Evidence for an abnormality of vitamin D metabolism. J Pediatr 115:397–404. 236. Hasanoglu A, Ercan ZS, Buyan N, Memioglu N, Hasanoglu E 1997 Parathyroid, 1,25 dihydroxyvitamin D and prostaglandin E correlation in children with idiopathich hypercalcuria. Prostaglandins Leukot Essent Fatty Acids 56:235–237. 237. Pryke AM, Duggan C, White CP, Posen S, Mason RS 1990 Tumor necrosis factor-alpha induces vitamin D-1-hydroxylase activity in normal human alveolar macrophages. J Cell Physiol 142:652–656. 238. Bickle DD, Pillai S, Gee E, Hincenbergs M 1991 Tumor necrosis factor-a regulation of 1,25-dihydroxyvitamin D production by human keratinocytes. Endocrinology 129: 33–38. 239. Zehnder D, Bland R, Chana RS, Wheeler DC, Howie AJ, Williams M, Stewart PM, Hewison M 2002 Synthesis of 1,25 dihydroxyvitamin D3 by human endothelial cells is regulated by inflammatory cytokines: an novel autocrine determinant of vascular cell adhension. J Am Soc Nephrol 13:621–629. 240. Mayur N, Lewis S, Catherwood BD, Nanes MS 1993 Tumor necrosis factor a decreases 1,25-dihydroxyvitamin D3 receptors in osteoblastic ROS 17/2.8 cells. J Bone Miner Res 8:997–1003. 241. Farmer PK, He X, Schmitz ML, Rubin J, Nanes MS 2000 Inhibitory effect of NF-κB on 1,25-dihydroxyvitamin D3 and retinoid X receptor function. Am J Physiol Endocrinol Metab 279:E213–E220.

1282 242. Koeffler HP, Reichel H, Bishop JE, Norman AW 1989 Gamma Interferon stimulates production of 1,25-dihydroxyvitamin D3 by normal human macrophages. Biochem Biophys Res Commun 127:596–603. 243. Bickle DD, Pillai S, Gee E, Hincenbergs M 1989 Regulation of 1,25-dihydroxyvitamin D production in human keratinocytes by interferon-y. Endocrinology 124:655–660. 244. Reichel H, Koeffler HP, Norman AW 1987 Synthesis in vitro of 1,25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3 by interferon-y-stimulated normal human bone marrow and alveolar macrophages. J Biol Chem 262:10931–10937. 245. Robinson BWS, McLemore TL, Crystal RG 1985 Gamma interferon is spontaneously released by alveolar macrophages and lung T lymphocytes in patients with pulmonary sarcoidosis. J Clin Invest 75:1488–1495. 246. Mann GN, Jacobs TW, Buchinsky FJ, Armstrong EC, Li M, Ke HZ, Ma YF, Jee WSS, Epstein S 1994 Interferon-gamma causes loss of bone volume in vivo and fails to ameliorate cyclosporin A-induced osteopenia. Endocrinology 135: 1077–1083. 247. Kruse R 1968 Osteopathien bei antiepileptisher langzeittherapie. I fonatsschrift fur Kinderheilkd 116:378–381. 248. Ashworth B, Horn DB 1977 Evidence of osteomalacia in an outpatient group of adult epileptics. Epilepsia 18:37–43. 249. Hahn TJ, Halstead LR 1979 Anticonvulsant drug-induced osteomalacia: Alterations in mineral metabolism and response to vitamin D3 administration. Calcif Tissue Int 27:13–18. 250. Lau KHW, Nakade O, Barr B, Taylor AK, Hochin K, Baylink DJ 1995 Phenytoin increases markers of osteogenesis for the human species in vitro and in vivo. J Clin Endocrinol Metab 80:2347–2353. 251. Hahn TJ, Scharp CR, Avioli LV 1974 Effect of phenobarbital administration on the subcellular distribution of vitamin D3-3H in rat liver. Endocrinology 94:1489–1495. 252. Silver J, Neale G, Thompson GR 1974 Effect of phenobarbitone treatment on vitamin D metabolism in mammals. Clin Sci Mol Med 46:433–448. 253. Hahn TJ, Halstead LR, Scharp CR, Haddad JG Jr 1975 Enhanced biotransformation and biologic efficacy of vitamin D following phenobarbital administration in the rat. Clin Res 23:111A. 254. Ohta T, Wergedal JE, Gruber HE, Baylink DJ, Lau KHW 1995 Low-dose phenytoin is an osteogenic agent in the rat. Calcif Tissue Int 56:42–48. 255. Onodera K, Takahashi A, Mayanagi H, Wakabayashi H, Kamei J, Shi H 2001 Phenytoin-induced bone loss and its prevention with alfacalcidiol or calcitriol in growing rats. Calcif Tissue Int 69:109–116. 256. Richens A, Rowe DJF 1970 Disturbance of calcium metabolism by anticonvulsant drugs. Br Med J 4:73–76. 257. Hunter J, Maxwell JD, Stewart DA, Parsons V, Williams R 1971 Altered calcium metabolism in epileptic children on anticonvulsants. Br Med J 4:202–204. 258. Weinstein RS, Bryce GF, Sappington LJ, King DW, Gallagher BB 1984 Decreased serum-ionized calcium and normal vitamin D metabolite levels with anticonvulsant drug treatment. J Clin Endocrinol Metab 58:1003–1009. 259. Gough H, Goggin T, Bissessar A, Baker M, Crowley M, Callaghan N 1986 A comparative study of the relative influence of different anticonvulsant drugs, UV exposure, and diet on vitamin D and calcium metabolism in out-patients with epilepsy. Q J Med 59:569–577. 260. Hahn TJ, Birge SJ, Scharp CR, Avioli LV 1972 Phenobarbitalinduced alterations in vitamin D metabolism. J Clin Invest 51:741–748.

SOL EPSTEIN AND ADINA E. SCHNEIDER

261. Davie MWJ, Emberson CE, Lawson DEM, Roberts GE, Barnes JLC, Barnes ND, Heeley AF 1983 Low plasma 25-hydroxyvitamin D and serum calcium levels in institutionalized epileptic subjects: Associated risk factors, consequences, and response to treatment with vitamin D. Q J Med 205:79–91. 262. Hahn TJ, Hendin BA, Scharp CR, Haddad JG 1972 Effect of chronic anticonvulsant therapy on serum 25-hydroxycalciferol levels in adults. N Engl J Med 287:900–904. 263. Stamp TCB, Round JM, Rowe DJF, Haddad JG 1972 Plasma levels and therapeutic effect of 25-hydroxycholecalciferol in epileptic patients taking anticonvulsant drugs. Br Med J 4:9–12. 264. Bouillon R, Reynaert J, Claes JH, Lissens W, De Moor P 1975 The effect of anticonvulsant therapy on serum levels of 25-hydroxyvitamin D, calcium, and parathyroid hormone. J Clin Endocrinol Metab 41:1130–1135. 265. Gascon-Barre M, Villeneuve J-P, Lebrun L-H 1984 Effect of increasing doses of phenytoin on the plasma 25-hydroxyvitamin D and 25-dihydroxyvitamin D and 1,25-dihydroxyvitamin D concentrations. J Am College Nutr 3:45–50. 266. Valimaki M, Tiihonen M, Laitinen K, Tahtela R, Karkkainen M, Lamberg-Allardt C, Makela P, Tunninen R 1994 Bone mineral density measured by dual-energy X-ray absorptiometry and novel markers of bone formation and resorption in patients on antiepileptic drugs. J Bone Miner Res 9:631–637. 267. Jubiz W, Haussler MR, McCain TA, Tolman KG 1977 Plasma 1,25-dihydroxyvitamin D levels in patients receiving anticonvulsant drugs. J Clin Endocrinol Metab 44:617–621. 268. Hahn TJ, Hendin BA, Scharp CR, Boisseau VC, Haddad JG Jr 1975 Serum 25-hydroxycalciferol levels and bone mass in children on chronic anticonvulsant therapy. N Engl J Med 292:550–554. 269. Takeshita N, Seino Y, Ishida H, Seino Y, Tanaka H, Tsutsumi C, Ogata K, Kiyohara K, Kato H, Nozawa M, Akiyama Y, Kara K, Imura H 1989 Increasing circulating levels of y-carboxyglutamic acid-containing protein and decreased bone mass in children on anticonvulsant therapy. Calcif Tissue Int 44:80–85. 270. Wark JD, Larkins RG, Perry-Keene D, Peter CT, Ross DL, Sloman JG 1979 Chronic diphenylhydantoin therapy does not reduce plasma 25-hydroxyvitamin D. Clin Endocrinol 11:267–274. 271. Tjellesen L, Nilas L, Christiansen C 1983 Does carbamazepine cause disturbances in calcium metabolism in epileptic patients? Acta Neurol Scand 68:13–19. 272. Filardi S, Guerreiro CA, Magna LA, Marques Neto JF 2000 Bone mineral density, vitamin D, and anticonvulsant therapy. Arq Neuropsiquiatr 58:616–620. 273. Bell RD, Pak CYC, Zerwekh J, Barilla DE, Vasko M 1979 Effect of phenytoin on bone and vitamin D metabolism. Ann Neurol 5:374–378. 274. Zerwekh JE, Homan R, Tindall R, Pak CYC 1982 Decreased 24,25-dihydroxyvitamin D concentration during long-term, anticonvulsant therapy in adult epileptics. Ann Neurol 12:184–186. 275. Weisman Y, Fattal A, Eisenberg Z, Harel S, Spirer Z, Harell A 1979 Decreased serum 24,25-dihydroxy-vitamin D concentrations in children receiving chronic anticonvulsant therapy. Br Med J 2:521–523. 276. Baer MT, Kozlowski BW, Blyler EM, Trahms Cm, Taylor ML 1997 Vitamin D, calcium, and bone status in children with developmental delay in relation to anticonvulsant use and ambulatory status. Am J Clin Nutr 65: 1042–1051.

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

277. Pedrera JD, Canal ML Carvajal J, Postigo S, Villa LF, Hernandez ER, Rico H 2000 Influence of vitamin D administration on bone ultrasound measurements in patients on anticonvulsant therapy. Euro J Clin Investigat 30: 895–905. 278. Tsukahara H, Kimura K, Todoroki Y, Ohshima Y, Hiraoka M, Shigematsu Y, Tsukahara Y, Miura M, Mayumi M 2002 Bone mineral status in ambulatory pediatric patients on long-term antiepileptic drug therapy. Pediatrics Int 44:247. 279. Verrotti A, Greco R, Latini G, Morgese G, Chiarelli F 2002 Increased bone turnover in prebupertal, pubertal, and postpubertal patients receiving carbamazepine. Epilepsia 43:1488–1492. 280. Conney AH 1967 Pharmacological implications of microsomal enzyme induction. Pharmacol Rev 19:317–366. 281. Eichelbaum M, Ekbom K, Bertilsson L, Ringer-Berger VA, Rane A 1975 Plasma kinetics of carbamazepine and its epoxide metabolite in man after single and multiple doses. Eur J Clin Pharmacol 8:337–341. 282. Perucca E 1978 Clinical consequences of microsomal enzyme-induction by antiepileptic drugs. Pharmacol Ther 2:285–314. 283. Mosekilde L, Christensen MS, Lund B, Sorensen OH, Melsen F 1977 The interrelationships between serum 25-hydroxycholecalciferol, serum parathyroid hormone, and bone changes in anticonvulsant osteomalacia. Acta Endocrinol 84:559–565. 284. Nielsen HE, Melsen F, Lund B, Sorensen OH, Mosekilde L 1983 Bone histomorphometry, vitamin D metabolites, and calcium phosphate metabolism in anticonvulsant treatment with carbamazepine. Calcif Tissue Int 35(Suppl.):224. 285. Hoikka V, Alhava EM, Karjalainen P, Keranen T, Savolainen KE, Riekkinen P, Korhonen R 1984 Carbamazepine and bone mineral metabolism. Acta Neurol Scand 69:77–80. 286. Feldkamp J, Becker A, Witte OW, Scharff D, Scherbaum WA 2000 Long-term anticonvulsant therapy leads to low bone mineral density—evidence for direct drug effects of phenytoin and carbamazepine on human osteoblast-like cells. Exp Clin Endocrinol Diabetes 108(1):37–43. 287. Kroger H, Lotaniemi A, Vaino P, Alhava E 1992 Bone densitometry of the spine and femur in children by dual X-ray absorptiometry. Bone Miner 17:75–85. 288. Sheth RD, Wesolowski CA, Jacob JC, Penney S, Hobbs GR, Riggs JE, Bodensteiner JB 1995 Effect of carbamazepine and valproate on bone mineral density. J Pediatr 127:256–262. 289. Guo CY, Ronen GM, Atkinson SA 2001 Long-term valproate and lamotrigine treatment may be a marker for reduced growth and bone mass in children with epilepsy. Epilepsia 42:1141–1147. 290. Farhat G, Yamout B, Mikati MA, Demirjian S, Sawaya R, El-Hajj Fuleihan G 2002 Effect of antiepileptic drug on bone density in ambulatory patients. Neurology 58:1348–1353. 291. Dent CE, Richens A, Rowe DJF, Stamp TCB 1970 Osteomalacia with long-term anticonvulsant therapy in epilepsy. Br Med J 4:69–72. 292. Collins N, Maher J, Cole M, Baker M, Callaghan N 1991 A prospective study to evaluate the dose of vitamin D required to correct low 25-hydroxyvitamin D levels, calcium, and alkaline phosphatase in patients at risk of developing antiepileptic drug-induced osteomalacia. Q J Med 286:113–122. 293. Morris HA, Need AG, O’Loughlin PD, Horowitz M, Bridges A, Nordin BEC 1990 Malabsorption of calcium in corticosteroid-induced osteoporosis. Calcif Tissue Int 46:305–308. 294. Hahn TJ, Halstead LR, Baran DT 1981 Effects of shortterm glucocorticoid administration on intestinal calcium

295. 296. 297.

298.

299.

300. 301. 302.

303.

304. 305. 306. 307.

308.

309. 310.

311.

312. 313.

1283 absorption and circulating vitamin D metabolite concentrations in man. J Clin Endocrinol Metab 52:111–115. Klein RG, Arnaud SB, Gallagher JC, DeLuca HF, Riggs BL 1977 Intestinal calcium absorption in exogenous hypercortisonism. J Clin Invest 60:253–259. Kimberg DV, Baerg RD, Gershon E, Graudusius RT 1971 Effect of cortisone treatment on the active transport of calcium by the small intestine. J Clin Invest 50:1309–1321. Adams JS, Wahl TO, Lukert BP 1981 Effect of hydrochlorothiazide and dietary sodium restriction on calcium metabolism in corticosteroid treated patients. Metabolism 30:217–221. Suzuki Y, Ichikawa Y, Saito E, Homma M 1983 Importance of increased urinary calcium excretion in the development of secondary hyperparathyroidism of patients under glucocorticoid therapy. Metabolism 32:151–156. Seeman E, Kumar R, Hunder G, Scott M, Heath III H, Riggs BL 1980 Production, degradation, and circulating levels of 1,25-dihydroxyvitamin D in health and in glucocorticoid excess. J Clin Invest 66:664–669. Hahn TJ, Halstead LR, Teitelbaum SL, Hahn BH 1979 Altered mineral metabolism in glucocorticoid-induced osteopenia. J Clin Invest 64:655. Lukert BP, Adams JS 1976 Calcium and phosphorus homeostasis in man. Arch Intern Med 136:1249–1253. Carbonare LD, Arlot ME, Chavassieux PM, et al. 2001 Comparison of trabecular bone microarchitecture and remodeling in glucorticoid-induced and postmenopausal osteoporosis. J Bone Miner Res 16:97–103. Rubin M, Bilezikian JP 2002 The role of parathyroid hormone in the pathogensis of glucocorticoid-induced osteoporosis: A re-examination of the evidence. J Clin Endocrinol Metab 87:4022–4041. Canalis E 2003 Mechanisms of glucocorticoid-induced osteoporosis. Current opinions in rheumatology 15:454–457. Avioli LV, Birge SJ, Lee SW 1968 Effects of prednisone on vitamin D metabolism in man. Clin Endocrinol 28:1341–1346. Hahn TJ, Halstead LR, Haddad JG Jr 1977 Serum 25-hydroxyvitamin D concentrations in patients receiving chronic corticosteroid therapy. J Clin Lab Med 90:399–404. LoCascio V, Bonucci E, Imbimbo B, Ballanti P, Adami S, Milani S, Tartarotti D, DellaRocca C 1990 Bone loss in response to long-term glucocorticoid therapy. Bone Miner 8:39–51. Prummel MF, Wiersinga WM, Lips P, Sanders GTP, Sauerwein HP 1991 The course of biochemical parameters of bone turnover during treatment with corticosteroids. J Clin Endocrinol Metab 72:382–386. Zerwekh JE, Emkey RD, Harris ED Jr 1984 Low-dose prednisone therapy in rheumatoid arthritis: Effect on vitamin D metabolism. Arth Rheum 27:1050–1052. Slovik DM, Neer RM, Ohman JL, Lowell FC, Clark MB, Segre GV, Potts JT Jr 1980 Parathyroid hormone and 25-hydroxyvitamin D levels in glucocorticoid-treated patients. Clin Endocrinol 12:243–248. Hodsman AB, Toogood JH, Jennings B, Fraher LJ, Baskerville JC 1991 Differential effects of inhaled budesonide and oral prednisolone on serum osteocalcin. J Clin Endocrinol Metab 72:530–540. Jennings BH, Andersson KE, Johansson SA 1991 The assessment of the systemic effects of inhaled glucocorticosteroids. Eur J Clin Pharmacol 41:11–16. Findling JW, Adams ND, Lemann J Jr, Gray RW, Thomas CJ, Tyrrell JB 1982 Vitamin D metabolites and parathyroid hormone in Cushing’s syndrome: Relationship to calcium

1284

314. 315.

316.

317.

318.

319.

320.

321. 322.

323. 324.

325. 326.

327.

328. 329.

330.

and phosphorus homeostasis. J Clin Endocrinol Metab 54:1039–1044. Aloia JF, Roginsky M, Ellis K, Shukla K, Chon S 1974 Skeletal metabolism and body composition in Cushing’s syndrome. J Clin Endocrinol Metab 39:981–985. Jowell PS, Epstein S, Fallen MD, Reinhardt TA, Ismail F 1987 1,25-Dihydroxyvitamin D3 modulates glucocorticoidinduced alteration in serum bone gla protein and bone histomorphometry. Endocrinology 120:531–536. Favus MJ, Kimberg DV, Millar GN, Gershon E 1973 Effects of cortisone administration on the metabolism and localization of 25-hydroxycholecalciferol in the rat. J Clin Invest 52:1328–1335. Carre M, Ayigbede O, Miravet L, Rasmussen H 1974 The effect of prednisolone upon the metabolism and action of 25-hydroxy- and 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 71:2996–3000. Chesney RW, Hamstra AJ, Mazess RB, DeLuca HF, O’Reagan S 1978 Reduction of serum-l,25-dihydroxyvitamin-D3 in children receiving glucocorticoids. Lancet 2:1123–1125. O’Regan S, Chesney RW, Hamstra A, Eisman JA, O’Gorman AM, DeLuca HF 1979 Reduced serum 1,25-(OH)2 vitamin D3 levels in prednisone-treated adolescents with systemic lupus erythematosus. Acta Paediatr Scand 68:109–111. Bikle DD, Halloran B, Fong L, Steinbach L, Shellito J 1993 Elevated 1,25-dihydroxyvitamin D levels in patients with chronic obstructive pulmonary disease treated with prednisone. J Clin Endocrinol Metab 76:456–461. Lukert BP, Stanbury SW, Mawer EB 1973 Vitamin D and intestinal transport of calcium: Effects of prednisolone. Endocrinology 93:718–722. Braun JJ, Juttmann JR, Visser TJ, Birkenhager JC 1982 Short-term effect of prednisone on serum 1,25-dihydroxyvitamin D in normal individuals and in hyper- and hypoparathyroidism. Clin Endocrinol 17:21–28. Au WYW 1976 Cortisol stimulation of parathyroid hormone secretion by rat parathyroid glands in organ culture. Science 193:1015–1017. Williams GA, Peterson WC, Bowser EN, Henderson WJ, Hargis GK, Martinez NJ 1974 Interrelationship of parathyroid and adrenocortical function in calcium homeostasis in the rat. Endocrinology 95:707–712. Ho YV, Briganti EM, Duan Y, Buchanan R, Hall S, Seeman E 1999 Polymorphism of the vitamin D receptor gene and corticosteroid-related osteoporosis. Osteoporos Int 9:134–138. Lindholm J, Steiniche T, Rasmussen E, Thamsborg G, Nielson IO 1991 Bone disorder in men with chronic alchoholism: a reversible disease. J Clin Endocrinol Metab 73:118–124. Bjorneboe GEA, Johnsen J, Bjorneboe A, Rousseau B, Pedersen JI, Norum KR, Morland J, Drevon CA 1986 Effect of alcohol consumption on serum concentration of 25-hydroxyvitamin D3, retinol, and retinol-binding protein. Am J Clin Nutr 44:678–682. Feitelberg S, Epstein S, Ismail F, D’Amanda C 1987 Deranged bone mineral metabolism in chronic alcoholism. Metabolism 36:322–326. Mobarhan SA, Russell RM, Recker RR, Posner DB, Iber FL, Miller P 1984 Metabolic bone disease in alcoholic cirrhosis: A comparison of the effect of vitamin D2, 25-hydroxyvitamin D, or supportive treatment. Hepatology 4:266–273. Bjorneboe GA, Johnsen J, Bjorneboe A, Morland J, Drevon C 1987 Effect of heavy alcohol consumption on serum

SOL EPSTEIN AND ADINA E. SCHNEIDER

331.

332. 333. 334.

335.

336.

337.

338.

339.

340. 341. 342. 343. 344.

345.

346.

347.

concentrations of fat-soluble vitamins and selenium. Alcohol and Alcoholism (Suppl. 1):533–537. Peris P, Pares A, Guanabens N, Del Rio L, Pons F, Jesus M, De Osaba M, Monegal A, Caballeria J, Rodes J, MunozGomez J 1994 Bone mass improves in alcoholics after two years of abstinence. J Bone Miner Res 9:1607–1612. Lalor BC, France MW, Powell D, Adams PH, Counihan TB 1986 Bone and mineral metabolism and chronic alcohol abuse. Q J Med 59:497–511. Verbanck M, Verbanck J, Brauman J, Mullier JP 1977 Bone histology and 25-OH vitamin D levels in alcoholics without cirrhosis. Calcif Tissue Res 22(Suppl.):538–541. Laitinen K, Lamberg-Allardt C, Tunninen R, Karonen S-L, Ylikahri R, Valimaki M 1991 Effects of three weeks’ moderate alcohol intake on bone and mineral metabolism in normal men. Bone Miner 13:139–151. Pepersack T, Fuss M, Otero J, Bergmann P, Valsamis J, Corvilain J 1992 Longitudinal study of bone metabolism after ethanol withdrawal in alcoholic patients. J Bone Miner Res 7:383–387. Barragry JM, Long RG, France MW, Wills MR, Boucher BJ, Sherlock S 1979 Intestinal absorption of cholecalciferol in alcoholic liver disease and primary biliary cirrhosis. Gut 20:559–564. Lindholm J, Steiniche T, Rasmussen E, Thamsborg G, Nielsen IO, Brockstedt-Rasmussen H, Storm T, Hyldstrup L, Schou C 1991 Bone disorder in men with chronic alcoholism: A reversible disease? J Clin Endocrinol Metab 73:118–124. Bjorneboe GE, Bjorneboe A, Johnsen J, Skylv N, Oftebro N, Gautvik KM, Hoiseth A, Morland J, Drevon CA 1988 Calcium status and calcium-regulating hormones in alcoholics. Alcohol Clin Exp Res 12:229–232. Santolaria F, Gonzalez-Reimers E, Perez-Manzano JL, Milena A, Gomez-Rodriguez MA, Gonzalez-Diaz A, de la Vega MJ, Martinez-Riera A 2000 Osteopenia assessed by body composition analysis is related to malnutrition in alcoholic patients. Alcohol 22:147–157. Meyer M, Wechsler S, Shibolet S, Jedwab M, Harell A, Edelsteim S 1978 Malabsorption of vitamin D in man and rat with liver cirrhosis. J Mol Med 3:29–37. Gascon-Barre M, Joly J-G 1981 The biliary excretion of [3H]-25-hydroxyvitamin D3 following chronic ethanol administration in the rat. Life Sci 28:279–286. Lund B, Sorensen OH, Hilden M, Lund B 1977 The hepatic conversion of vitamin D in alcoholics with varying degrees of liver affection. Acta Med Scand 202:221–224. Avioli LV, Lee SW, McDonald JE, Lund J, DeLuca HF 1967 Metabolism of D3-3H in human subjects. Distribution to blood, bile, feces, and urine. J Clin Invest 46:983–992. Brodie MJ, Boobis AR, Hillyard CJ, Abeyasekera G, Maclntyre I, Park BK 1981 Effect of isoniazid on vitamin D metabolism and hepatic monooxygenase activity. Clin Pharmacol Ther 30:363–367. Posner DB, Russell RM, Absood S, Connor TB, Davis C, Martin L, Williams JB, Norris AH, Merchant C 1978 Effective 25-hydroxylation of vitamin D2 in alcoholic cirrhosis. Gastroenterology 74:866–870. Bickle DD, Halloran BP, Gee E, Ryzen E, Haddad JG 1986 Free 25-hydroxyvitamin D levels are normal in subjects with liver disease and reduced total 25-hydroxyvitamin D levels. J Clin Invest 78:748–752. Bouillon R, Auwerx J, Dekeyser L, Fevery J, Lissens W, De Moor P 1984 Serum vitamin D metabolites and their

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

348.

349.

350. 351.

352. 353.

354.

355. 356. 357. 358.

359. 360. 361. 362. 363. 364.

365.

366.

binding protein in patients with liver cirrhosis. J Clin Endocrinol Metab 59:86–89. Laitinen K, Lamberg C, Tunninen R, Karonen SL, Ylikahri R, Valimaki M 1991 Effects of three weeks of moderate alchohol intake on bone and mineral metabolism in normal men. Bone Miner 13:139–151. Shah JH, Bowser EN, Hargis GK, Wongsurawat N, Banerjee P, Henederson WJ, Williams GA 1978 Effect of ethanol on parathyroid hormone secretion in the rat. Metabolism 27: 257–260. Channard J, Lacour B, Drueke T, Brunois JP, Ruiz JC 1980 Effect of acute ethanol loading on parathyroid gland secretion in the rat. Adv Exp Med Biol 128:495–504. Williams GA, Bowser EN, Hargis GK, Kukreja SC, Shah JH, Vora NM, Henderson WJ 1978 Effect of ethanol on parathyroid hormone and calcitonin secretion in man. Proc Soc Exp Biol Med 159:187–191. Ljunghall S, Lundin L, Wide L 1985 Acute effects of ethanol intake on the serum concentrations of parathyroid hormone, calcium, and phosphate. Exp Clin Endocrinol 85:365–368. Bjorneboe GE, Bjorneboe A, Johnsen J, Skylv N, Oftebro H, Gautvik KM, Hoiseth A, Morland J, Drevon CA 1988 Calcium status and calcium-regulating hormones in alcoholics. Alcoholism 12:229–232. Laitinen K, Valimaki M, Lamberg-Allardt C, Kivisaari L, Lalla M, Karkkainen M, Ylikahri R 1990 Deranged vitamin D metabolism but normal bone mineral density in Finnish noncirrhotic male alcoholics. Alcoholism 14:551–556. Crilly RG, Anderson C, Hogan D, Delaquerriere-Richardson L 1988 Bone histomorphometry, bone mass, and related parameters in alcoholic males. Calcif Tissue Int 43:269–276. Bickle DD, Genant HK, Cann C, Recker RR, Halloran BP, Strewler GJ 1985 Bone disease in alcohol abuse. Ann Intern Med 103:42–48. Krawitt EL 1975 Effect of ethanol ingestion on duodenal calcium transport. J Clin Lab Med 85:665–671. Vodoz JF, Luisier M, Donath A, Courvoisier B, Garcia B 1977 Diminution de l’absorption intestinale de 47-calcium dans l’alcoolisme chronique. Schweiz Med Wochenschr 107:1525–1529. Avery DH, Overall JE, Calil HM, Hollister LE 1983 Plasma calcium and phosphate during alcohol intoxication. Alcoholics versus nonalcoholics. J Stud Alcohol 44:205–214. Krawitt EL 1973 Ethanol inhibits intestinal calcium transport in rats. Nature 243:88–89. Territo MC, Tanaka KR 1974 Hypophosphatemia in chronic alcoholism. Arch Intern Med 134:445–447. Kent JC, Devlin RD, Gutteridge DH, Retallack RW 1979 Effect of alcohol on renal vitamin D metabolism in chickens. Biochem Biophys Res Commun 89:155–161. Rude RK, Oldham SB, Sharp CF Jr, Singer FR 1978 Parathyroid hormone secretion in magnesium deficiency. J Clin Endocrinol Metab 78:748–52. Gascon-Barre M 1982 Plasma 25-hydroxyvitamin D3 response to pharmacological dose of vitamin D3 or 25-hydroxyvitamin D3 during chronic ethanol administration in the rat. Horm Metab Res 14:332–333. Turner RT, Aloia RC, Segel LD, Hannon KS, Bell NH 1988 Chronic alcohol treatment results in disturbed vitamin D metabolism and skeletal abnormalities in the rat. Alcohol Clin Exp Res 12:159–162. Gascon-Barre M 1982 Interrelationships between vitamin D3 and 25-hydroxyvitamin D3 during chronic ethanol administration in the rat. Metabolism 31:67–72.

1285

367. Saarem K, Bergseth S, Oftebro H, Pedersen JI 1984 Subcellular localization of vitamin D3 25-hydroxylase in human liver. J Biol Chem 259:10936–10940. 368. Saarem K, Pedersen JI 1985 25-Hydroxylation of 1-hydroxyvitamin D3 in rat and human liver. Biochim Biophys Acta 840:117–126. 369. Laitinen K (ed) 1993 Alcohol and Bone. Yliopistopaino, Helsinki. 370. Gascon-Barre M 1985 Influence of chronic ethanol consumption on the metabolism and action of vitamin D. J Am College Nutr 4:565–574. 371. Feldman D 1986 Ketoconazole and other imidizole derivatives as inhibitors of steroidogenesis. Endocr Rev 7:409–420. 372. Loose DS, Kan PB, Hirst MA, Marcus RA, Feldman D 1983 Ketoconazole blocks adrenal steroidogenesis by inhibiting cytochrome P450-dependent enzymes. J Clin Invest 71:1495–1499. 373. Henry HL 1985 Effect of ketoconazole and miconazole on 25-hydroxyvitamin D3 metabolism by cultured chick kidney cells. J Steroid Biochem Mol Biol 23:991–994. 374. Glass AR, Eil C 1986 Ketoconazol-induced reduction in serum 1,25-dihydroxyvitamin D. J Clin Endocrinol Metab 63(No. 3):766–769. 375. Glass AR, Eil C 1988 Ketoconazole-induced reduction in serum 1,25-dihydroxyvitamin D and total serum calcium in hypercalcemic rats. J Clin Endocrinol Metab 66:934–938. 376. Riancho JA, Amado JA, Freijanes J, Otero M, Gonzalez Marcias J 1989 Ketoconazole and vitamin D metabolism in hyperparathyroidism. Horm Metab Res 21:51. 377. Adams JS, Sharma OP, Diz MM, Endres DB 1990 Ketoconazole decreases the serum 1,25-dihydroxyvitamin D and calcium concentration in sarcoidosis-associated hypercalcemia. J Clin Endocrinol Metab 70:1090–1095. 378. Glass AR, Cerletty JM, Elliott W, Lemann J Jr, Gray RW, Eil C 1990 Ketoconazole reduces elevated serum levels of 1,25-dihydroxyvitamin D in hypercalcemic sarcoidosis. J Endocrinol Invest 13:407–413. 379. Peehl DM, Seto E, Hsu JY, Feldman D 2002 Preclinical activity of ketoconazole in combination with calcitriol or the vitamin D analogue EB 1089 in prostate cancer cells. J Urol 168:1583–1588. 380. Ismail F, Corder CN, Epstein S, Barbi G, Thomas S 1990 Effects of pravastatin and cholestyramine on circulating levels of parathyroid hormone and vitamin D metabolites. Clin Ther 12:427–430. 381. Dobs A, Levine MA, Margolis S 1991 Effects of pravastatin, a new HMG-CoA reductase inhibitor, on vitamin D synthesis in man. Metabolism 40:524–528. 382. Montagnani M, Lore F, Di Cairano G, Gonnelli S, Ciuoli C, Montagnani A, Gennari C 1994 Effects of pravastatin treatment on vitamin D metabolites. Clin Ther 16:824–829. 383. Mundy G, Garrett R, Harris S, Chan J, Chen D, Rossini G, Boyce B, Gutierrez G 1999 Stimulation of bone formation in vitro and in rodents by statins. Science 286:1946–1949. 384. Meier Cr, Schlienger RG, Kraenzlin ME, Schlegel B, Jick H 2000 HMG-CoA reductase inhibitors and the risk of fracture. JAMA 282:3205–3210. Wang PS, Solomon DH, Mogun H, Avorn J 2000 HMG-CoA reductase inhibitors and the risk of hip fracture in elderly patients. JAMA 283:3211–3216. 385. Gascon-Barre M 1982 Biliary excretion of [3H]-25-hydroxyvitamin D3 in the vitamin D-depleted rat. Am J Physiol 242:G522–G532. 386. Clements MR, Chalmers TM, Fraser DR 1984 Enterohepatic circulation of vitamin D: A repraisal of the hypothesis. Lancet 1:1376–1379.

1286 387. Kumar R, Nagubandi S, Mattox VR, Londowski JM 1980 Enterohepatic physiology of 1,25-dihydroxyvitamin D3. J Clin Invest 65:277–284. 388. Thompson WG, Thompson GR 1969 Effect of cholestyramine on the absorption of vitamin D3 and calcium. Gut 10:717–722. 389. Heaton KW, Lever JV, Barnard D 1972 Osteomalacia associated with cholestyramine therapy for posileectomy diarrhea. Gastroenterology 62:642–646. 390. Compston JE, Horton LWL 1978 Oral 25-hydroxyvitamin D3 in treatment of osteomalacia associated with ileal resection and cholestyramine therapy. Gastroenterology 74:900–902. 391. Compston JE, Thompson RPH 1977 Intestinal absorption of 25-hydroxyvitamin D and osteomalacia in primary biliary cirrhosis. Lancet 1:721–724. 392. Hoogwerf BJ, Hibbard DM, Hunninghake DB 1992 Effects of long-term cholestyramine administration on vitamin D and parathormone levels in middle-aged men with hypercholesterolemia. J Clin Lab Med 119:407–411. 393. Schwarz KB, Goldstein PD, Witztum JL, Schonfeld G 1980 Fat-soluble vitamin concentrations in hypercholesterolemic children treated with colestipol. Pediatrics 65:243–250. 394. Tsang RC, Roginsky MS, Mellies M, Glueck CJ, Kashyap ML 1977 Plasma 25 OH-vitamin D: Familial hypercholesterolemic children on colestipol resin. Clin Res 35:567A. 395. van Heek M, Farley C, Compton DS, Hoos L, Davis HR 2001 Ezetimibe selectively inhibits intestinal cholesterol absorption in rodents in the presence and absence of exocrine pancreatic function. Br J Pharmacol 134:409–417. 396. Wilczek H, Sobra J, Ceska R, Justova V, Prochazkova R, Kvasilova M, Juzova Z 1993 Lecba fibraty a metabolismus vitaminu D. Casopis Lekaru Ceskych 132:630–632. 397. Fleisch H, Russell RGG, Francis MD 1969 Diphosphonates inhibit hydroxyapatite dissolution in vitro and bone resorption in tissue culture and in vivo. Science 165:1262–1264. 398. Russell RG, Muhlbauer RC, Bisaz S, Williams DA, Fleisch H 1970 The influence of pyrophosphate, condensed phosphates, phosphonates, and other phosphate compounds on the dissolution of hydroxyapatite in vitro and on bone resorption induced by parathyroid hormone in tissue culture and in thyroparathyroidectomized rats. Calcif Tissue Res 6:183–196. 399. Gasser AB, Morgan DB, Fleisch H, Richelle LJ 1972 The influence of two diphosphonates on calcium metabolism in the rat. Clin Sci 43:31–45. 400. Kanis JA, Gertz BJ, Singer F, Ortolani S 1995 Rationale for the use of alendronate in osteoporosis. Osteoporosis Int 5:1–13. 401. Rossini M, Gatti D, Zamberlan N, Braga V, Dorizzi R, Adami S 1994 Long-term effects of a treatment course with oral alendronate on postmenopausal osteoporosis. J Bone Miner Res 9(11):1833–1837. 402. O’Doherty DP, Bickerstaff DR, McCloskey EV, Hamdy NAT, Beneton MNC, Harris S, Mian M, Kanis JA 1990 Treatment of Paget’s disease of bone with aminohydroxybutylidene bisphosphonate. J Bone Miner Res 5:483–491. 403. Trechsel U, Taylor CM, Eisman JA, Bonjour JP, Fleisch H 1981 Plasma levels of vitamin D metabolites in diphosphonatetreated rats. Clin Sci 61:471–476. 404. Hill LF, Lumb GA, Mawer EB, Stanbury SW 1973 Indirect inhibition of biosynthesis of 1,25-dihydroxycholecalciferol in rats treated with diphosphonate. Clin Sci 44:335–347. 405. Nagao Y, Ishitobi Y, Kinoshita H, Fukushima S, Kawashima H 1991 YM175, a new bisphosphonate, increases serum 1,25dihydroxyvitamin D in rats via stimulating renal 1-hydroxylase activity. Biochem Biophys Res Commun 180:1172–1178.

SOL EPSTEIN AND ADINA E. SCHNEIDER

406. Ralston SH, Boyce BF, Cowan RA, Fogelman I, Smith ML, Jenkins A, Boyle IT 1987 The effect of 1a-hydroxyvitamin D3 on the mineralization defect in disodium-etidronate-treated Paget’s disease—A double-blind, randomized clinical study. J Bone Miner Res 2:5–12. 407. Lawson-Matthew PJ, Guilland-Cummings DF, Yates AJP, Russell RGG, Kanis JA 1988 Contrasting effects of intravenous and oral etidronate on vitamin D metabolism. Clin Sci 74:101–106. 408. Devlin RD, Retallack RW, Fenton AJ, Grill V, Gutteridge DH, Kent GN, Prince RL, Worth GK 1994 Long-term elevation of 1,25 dihydroxyvitamin D after short-term intravenous administration of pamidronate (aminohydroxypropylidene bisphosphonate, APD) in Paget’s disease of bone. J Bone Miner Res 9:81–85. 409. Papapoulos SE, Frolich M, Mudde AH, Harinck HIJ, Berg HVD, Bijvoet OLM 1987 Serum osteocalcin in Paget’s disease of bone: Basal concentrations and response to bisphosphonate treatment. J Clin Endocrinol Metab 65:89–94. 410. Adami S, Frijlink WB, Bijvoet OLM, O’Riordan JLH, Clemens TL, Papapoulos SE 1982 Regulation of calcium absorption by 1,25-dihydroxyvitamin D—Studies of the effects of a bisphosphonate treatment. Calcif Tissue Int 34: 317–320. 411. Body JJ, Magritte A, Seraj F, Sculier JP, Borkowski A 1989 Aminohydroxypropylidene bisphosphonate (APD) treatment for tumor-associated hypercalcemia: A randomized comparison between a three-day treatment and single 24-hour infusions. J Bone Miner Res 4:923–928. 412. Budayr AA, Zysset E, Jenzer A, Thiebaud D, Ammann P, Rizzoli R, Jaquet-Muller F, Bonjour JP, Gertz B, Burckhardt P, Halloran BP, Nissenson RA, Strewler GJ 1994 Effects of treatment of malignancy-associated hypercalcemia on serum parathyroid hormone-related protein. J Bone Miner Res 9:521–526. 413. Grill V, Murray RML, Ho PWM, Santamaria JD, Pitt P, Potts C, Jerums G, Martin TJ 1992 Circulating PTH and PTHrP levels before and after treatment of tumor-induced hypercalcemia with pamidronate disodium (APD). J Clin Endocrinol Metab 74:1468–1470. 414. Fraser WD, Logue FC, Gallacher SJ, O’Reilly DSJ, Beastall GH, Ralston SH, Boyle IT 1991 Direct and indirect assessment of the parathyroid hormone response to pamidronate therapy in Paget’s disease of bone and hypercalcemia of malignancy. Bone Miner 12:113–121. 415. Sahni M, Guenther HL, Fleisch H, Collin P, Martin TJ 1993 Bisphosphonates act on rat bone resorption through the mediation of osteblasts. J Clin Invest 91:2004–2011. 416. Chesnut CH, McClung MR, Ensrud KE, Bell NH, Genant HK, Harris ST, Singer FR, Stock JL, Yood RA, Delmas PD, Kher U, Pryor-Tillotson S, Santora II AC 1995 Alendronate treatment of the postmenopausal osteoporotic woman: Effect of multiple dosages on bone mass and bone remodeling. Am J Med 99:144–152. 417. Harris ST, Gertz BJ, Genant HK, Eyre DR, Survill TT, Ventura JN, DeBrock J, Ricerca E, Chesnut III CH 1993 The effect of short-term treatment with alendronate on vertebral density and biochemical markers of bone remodeling in early postmenopausal women. J Clin Endocrinol Metab 76 (No. 6):1399–1406. 418. Martinez ME, Pastrana P, Snachez-Cabezudo MJ, Jariego D, Del Campo MT 1997 Effect of clodronate on calcidiol serum levels in women with breast cancer. Calcif Tissue In 61:148–150.

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

419. Wasnich RD, Benfante RJ, Yano K, Heilbrun L, Vogel JM 1983 Thiazide effect on the mineral content of bone. N Engl J Med 309:344–347. 420. Sakhaee K, Nicar MJ, Glass K, Zerwekh JE, Pak CYC 1984 Reduction in intestinal calcium absorption by hydrochorothiazide in postmenopausal osteoporosis. J Clin Endocrinol Metab 59:1037–1043. 420. Wasnich R, Davis J, Ross P, Vogel J 1990 Effect of thiazide on rates of bone mineral loss: A longitudinal study. Br Med J 301:1303–1305. 421. Dawson-Hughes B, Harris S 1993 Thiazides and seasonal bone changes in healthy postmenopausal women. Bone Miner 21:41–51. 422. LaCroix AZ, Wienphal J, White LR, Wallace RB, Scherr PA, George LK, Cornoni-Huntley J, Ostfeld AM 1990 Thiazide diuretic agents and the incidence of hip fractures. N Engl J Med 322:286–290. 423. Ray WA, Downey W, Griffin MR, Melton III LJ 1989 Longterm use of thiazide diuretics and risk of hip fracture. Lancet 1:687–690. 424. Felson DT, Sloutskis D, Anderson JJ, Anthony JM, Kiel DP 1991 Thiazide diuretics and the risk of hip fracture. JAMA 265:370–373. 425. Middler S, Pak CYC, Murad F, Bartter C 1973 Thiazide diuretics and calcium metabolism. Metabolism 22:139–146. 426. Ljunghall S, Backman U, Danielson BG, Fellstrom B, Johansson G, Wikstrom B 1981 Calcium and magnesium metabolism during long-term treatment with thiazides. Scand J Urol Nephrol 15:257–262. 427. Lamberg B-A, Kuhlback B 1959 Effect of chorothiazide and hydrochlorothiazide on the excretion of calcium in urine. Scand J Clin Lab Invest 11:351–357. 428. Stote RM, Smith LH, Wilson DM, Dube WJ, Goldsmith RS, Arnaud CD 1972 Hydrochorothiazide effects on serum calcium and immunoreactive parathyroid hormone concentrations. Ann Intern Med 77:587–591. 429. Sowers MR, Wallace RB, Hollis BW 1990 The relationship of 1,25-dihydroxyvitamin D and radial bone mass. Bone Miner 10:139–148. 430. Zerwekh JE, Pak CYC 1980 Selective effects of thiazide therapy on serum 1a,25-dihydroxyvitamin D and intestinal calcium absorption in renal and absorptive hypercalciurias. Metabolism 29:13–17. 431. Riis B, Christiansen C 1985 Actions of thiazides on vitamin D metabolism: A controlled therapeutic trial in normal women early in the postmenopause. Metabolism 34:421–424. 432. Ramp WK, Cooper CW, Ross III AJ, Wells SA Jr 1979 Effects of calcium and cyclic nucleotides on rat calcitonin and parathyroid hormone secretion. Mol Cell Endocrinol 14:205–215. 433. Larsson R, Akerstrom G, Gylfe E, Johansson H, Ljunghall S, Rastad J, Wallfelt C 1985 Paradoxical effects of K+ and D-600 on parathyroid hormone secretion and cytoplasmic Ca2+ in normal bovine and pathological human parathyroid cells. Biochim Biophys Acta 847:263–269. 434. Hove K, Sand O 1981 Evidence for a function of calcium influx in the stimulation of hormone release from the parathyroid gland in the goat. Acta Physiol Scand 113:37–43. 435. Seely EW, LeBoff S, Brown EM, Chen C, Posillico JT, Hollenberg NK, Williams GH 1989 The calcium channel blocker diltiazem lowers serum parathyroid hormone levels in vivo and in vitro. J Clin Endocrinol Metab 68:1007–1012. 436. Cooper CW, Borosky SA, Farrell PE, Steinsland S 1986 Effects of the calcium channel activator BAY-K-8644 on

437.

438. 439. 440.

441.

442.

433. 434. 435. 436. 437. 438. 439. 440. 441. 442. 443. 444. 445. 446.

1287 in vitro secretion of calcitonin and parathyroid hormone. Endocrinology 118:545–549. Ross III AJ, Cooper CW, Ramp WK, Wells Jr SA 1979 Concurrent secretion of calcitonin and parathyroid hormone in vitro: Effects of drugs that alter cellular calcium transport. Surg Forum 30:102–104. Fox J, Della-Santina CP 1989 Oral verapamil and calcium and vitamin D metabolism in rats: Effect of dietary calcium. Am J Physiol 257:E632–E638. Bogin E, Chagnac A, Juppner H, Levi J 1987 Effect of verapamil on plasma parathyroid hormone. J Clin Chem Clin Biochem 25:83–85. Frishman WH, Klein NA, Charlap S, Klein P, Norsratian F, Strom JA, Sherwood LM 1984 Comparative effects of verapamil and propanolol on parathyroid hormone and serum calcium concentration. In: Packer M, Frishman WH (eds) Calcium Channel Antagonists in Cardiovascular Disease. Appleton-Century-Crofts: Norwalk, Connecticut, pp. 1445–1449. Gozzelino G, Fubini A, Isaia GG, Soagliotti G, Gamna G 1981 Evaluation of calcium metabolism and pituitary gonadotropic secretion during treatment with nifedipine. G Ital Cardiol 11:1445–1449. Townsend R, Dipette DJ, Evans RR, Davis WR, Green A, Graham GA, Wallace JM, Holland OB 1990 Effects of calcium channel blockade on calcium homeostasis in mild to moderate essential hypertension. Am J Med Sci 300: 133–137. Albers MM, Johnson W, Vivian V, Jackson RD 1991 Chronic use of the calcium channel blocker nifedipine has no significant effect on bone metabolism in men. Bone 12:39–42. Renton KW 1985 Inhibition of hepatic microsomal drug metabolism by the calcium channel blockers diltiazem and verapamil. Biochem Pharmacol 34:2549–2553. Avioli LV 1975 Heparin-induced osteopenia: An appraisal. Adv Exp Med Biol 52:375–387. Sackler JP, Liu L 1973 Heparin-induced osteoporosis. Br J Radiol 46:548–550. Wise PH, Hall AJ 1980 Heparin-induced osteopenia in pregnancy. Br Med J 2:110–111. Griffiths HT, Liu DTY 1984 Severe heparin osteoporosis in pregnancy. Postgrad Med J 60:424–425. Mutoh S, Takeshita N, Yoshino T, Yamaguchi I 1993 Characterization of heparin-induced osteopenia in rats. Endocrinology 133:2743–2748. Matzsch T, Bergqvist D, Hedner U, Nilsson B, Ostergaard P 1986 Heparin-induced osteoporosis in rats. Thromb Haemost 56:293–294. Monreal M, Vinas L, Monreal L, Lavin S, Lafoz E, Angles AM 1990 Heparin-related osteoporosis in rats. Haemostasis 20:204–207. Aarskog D, Aksnes L, Lehmann V 1980 Low 1,25-dihydroxyvitamin D in heparin-induced osteopenia. Lancet 2:650–651. Haram K, Hervig T, Thordarson H, Aksnes L 1993 Osteopenia caused by heparin treatment in pregnancy. Acta Obstet Gynecol Scand 72:674–675. Hurley MM, Gronowicz G, Kream BE, Raisz LG 1990 Effect of heparin on bone formation in cultured fetal rat calvaria. Calcif Tissue Int 46:183–188. Hurley MM, Kream BE, Raisz LG 1990 Structural determinants of the capacity of heparin to inhibit collagen synthesis in 21-day fetal rat calvariae. J Bone Miner Res 5:1127–1133. Chowdhury MH, Hamada C, Dempster DW 1992 Effects of heparin on osteoclast activity. J Bone Miner Res 7:771–777.

1288 447. Hahnemann S 1965 Heparin and osteoporosis. Lancet 2:855–856. 448. Ellis HA 1965 Effects of long-term administration to animals of dextran sulfate. J Pathol Bacteriol 89:437–460. 449. Kock HJ, Handschin AE 2002 Clin Appl Thromb Hemost 8:251–255. 450. Street J, McGrath M, O’Regan K, Wakai A, McGuinness A, Redmond P 2000 Clin Orthopaed and Related Res 381:278–289. 451. Casele HL, Laifer SA 2000 J Matern Fetal Med 9:122–125. 452. Williams GA, Longley RS, Bowser EN, Hargis GK, Kukreja SC, Vora NM, Johnson PA, Jackson BL, Kawahara WJ, Henderson WJ 1981 Parathyroid hormone secretion in normal man and in primary hyperparathyroidism: Role of histamine H2 receptors. J Clin Endocrinol Metab 52:122–127. 453. Abboud HE, Zimmerman D, Edis AJ, Dousa TP 1980 Histamine and parathyroid adenoma: Effect on cyclic AMP accumulation. Clin Res 28:515A. 454. Sherwood JK, Ackroyd FW, Garcia M 1980 Effect of cimetidine on circulating parathyroid hormone in primary hyperparathyroidism. Lancet 1:616–620. 455. Wiske PS, Epstein S, Norton JA Jr, Bell NH, Johnston CC Jr 1983 The effects of intravenous and oral cimetidine in primary hyperparathyroidism. Horm Metab Res 15:245–248. 456. Jacob AI, Lanier D Jr, Canterbury J, Bourgoignie JJ 1980 Reduction by cimetidine of serum parathyroid hormone levels in uremic patients. N Engl J Med 302:671–674. 457. Beehler CJ, Beckner JR, Rosenquist RC, Shankel SW 1980 Parathyroid hormone suppression by cimetidine in uremic patients. Ann Intern Med 93:840–841. 458. Palmer FJ, Sawyers TM, Wierzbinski SJ 1980 Cimetidine and hyperparathyroidism. N Engl J Med 302:692. 459. Ghishan FK, Walker F, Meneely R, Patwardhan R, Speeg KV Jr 1981 Intestinal calcium transport: Effect of cimetidine. J Nutr 111:2157–2161. 460. Speeg KV Jr, Patwardhan RV, Avant GR, Mitchell MC, Schenker S 1982 Inhibition of microsomal drug metabolism by histamine H2-receptor antagonists studied in vivo and in vitro in rodents. Gastroenterology 82:89–96. 461. Henry DA, Macdonald IA, Kitchingman G, Bell GD, Langman MJS 1980 Cimetidine and ranitidine: Comparison of effects on hepatic drug metabolism. Br Med J 281:775–777. 462. Bengoa JM, Bolt MJ, Rosenberg IH 1984 Hepatic vitamin D 25-hydroxylase inhibition by cimetidine and isoniazid. J Clin Lab Med 104:546–552. 463. Wyatt C, Jensen LS, Rowland III GN 1990 Effect of cimetidine on eggshell quality and plasma 25-hydroxycholecalciferol in laying hens. Poultry Sci 69:1892–1899. 464. Odes HS, Fraser GM, Krugliak P, Lamprecht SA, Shany S 1980 Effect of cimetidine on hepatic vitamin D metabolism in humans. Digestion 46:61–64. 466. Klein GL, Horst RL, Norman AW, Ament ME, Slatopolsky E, Coburn JW 1981 Reduced serum levels of la,25-dihydroxyvitamin D during long-term total parenteral nutrition. Ann Intern Med 94:638–643. 467. Klein GL, Alfrey AC, Miller NL, Sherrard DJ, Hazlet TK, Ament ME, Coburn JW 1982 Aluminum loading during parenteral nutrition. Am J Clin Nutr 35:1425–1429. 468. Ott SM, Maloney NA, Klein GL, Alfrey AC, Ament ME, Coburn JW, Sherrard DJ 1983 Aluminum is associated with low bone formation in patients receiving chronic parenteral nutrition. Ann Intern Med 98:910–914. 469. Parkinson IS, Feest TG, Ward MK, Fawcett RWP, Kerr DNS 1979 Fracturing dialysis osteodystrophy and dialysis encephalopathy. Lancet 1:406–409.

SOL EPSTEIN AND ADINA E. SCHNEIDER

470. Pierides AM, Edwards WG Jr, Cullum UX Jr, McCall JT, Ellis HA 1980 Hemodialysis encephalopathy with osteomalacic fractures and muscle weakness. Kidney Int 18:115–124. 471. Ott SM, Maloney NA, Coburn JW, Alfrey AC, Sherrard DJ 1982 The prevalence of bone aluminum deposition in renal osteodystrophy and its relation to the response to calcitriol therapy. N Engl J Med 307:709–713. 472. Maloney NA, Ott SM, Alfrey AC, Miller NL, Coburn JW, Sherrard DJ 1982 Histological quantitation of aluminum in iliac bone from patients with renal failure. J Clin Lab Med 99:206–216. 473. Cournot-Witmer G, Zingraff J, Plachot JJ, Escaig F, Lefevre R, Boumati P, Bourdeau A, Garabedian M, Galle P, Bourdon R, Drueke T, Balsan S 1981 Aluminum localization in bone from hemodialyzed patients: Relationship to matrix mineralization. Kidney Int 20:375–385. 474. Ellis HA, McCarthy JH, Herrington J 1979 Bone aluminum in haemodialysed patients and in rats injected with aluminium chloride: Relationship to impaired bone mineralization. J Clin Pathol 32:832–844. 475. Hodsman AB, Sherrard DJ, Alfrey AC, Ott S, Brickman AS, Miller NS, Maloney NA, Coburn JW 1982 Bone aluminum and histomorphometric features of renal osteodystrophy. J Clin Endocrinol Metab 54:539–546. 476. Alfrey AC, Hegg A, Craswell P 1980 Metabolism and toxicity of aluminum in renal failure. Am J Clin Nutr 33:1509–1516. 477. Cann CE, Prussin SG, Gordan GS 1979 Aluminum uptake by the parathyroid glands. J Clin Endocrinol Metab 49:543–545. 478. Morrissey J, Rothstein M, Mayor G, Slatopolsky E 1983 Suppression of parathyroid hormone secretion by aluminum. Kidney Int 23:699–704. 479. Robertson JA, Felsenfeld AJ, Haygood CC, Wilson P, Clarke C, Llach F 1983 Animal model of aluminum-induced osteomalacia: Role of chronic renal failure. Kidney Int 23:327–335. 480. Chan YL, Alfrey AC, Posen S, Lissner D, Hills E, Dunstan CR, Evans RA 1983 Effect of aluminum on normal and uremic rats: Tissue distribution, vitamin D metabolites, and quantitative bone histology. Calcif Tissue Int 35:344–351. 481. Henry DA, Goodman WG, Nudelman RK, DiDomenico NC, Alfrey AC, Slatopolsky E, Stanley TM, Coburn JW 1984 Parenteral aluminum administration in the dog: I. Plasma kinetics, tissue levels, calcium metabolism, and parathyroid hormone. Kidney Int 25:362–369. 482. Goodman WG, Henry DA, Horst R, Nudelman RK, Alfrey AC, Coburn JW 1984 Parenteral aluminum administration in the dog: II. Induction of osteomalacia and effect on vitamin D metabolism. Kidney Int 25:370–375. 483. DeLuca HF 1994 Third F Raymond Keating Jr Memorial symposium parathyroid hormone, calcitonin, and vitamin D. Clinical consideration. II. Vitamin D—1973. Am J Med 57:1–12. 484. Berlyne GM, Pest D, Ben-Ari J, Weinberger J, Stern M, Gilmore GR, Levine R 1970 Hyperaluminaemia from aluminum resins in renal failure. Lancet 2:494–496. 485. Clarkson EM, Luck VA, Hynson WV, Bailey RR, Eastwood JB, Woodhead JS, Clements VR, O’Riordan JLH, De Wardener HE 1972 The effect of aluminum hydroxide on calcium, phosphorus, and aluminum balances, the serum parathyroid hormone concentration and the aluminum content of bone in patients with chronic renal failure. Clin Sci 43:519–531. 486. Recker RR, Blotcky AJ, Leffler JA, Rack EP 1977 Evidence for aluminum absorption from the gastrointestinal tract and bone deposition by aluminum carbonate ingestion with normal renal function. J Clin Lab Med 90:810–815.

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

487. Villa ML, Packer E, Cheema M, Holloway L, Marcus R 1991 Effects of aluminum hydroxide on the parathyroid vitamin D axis of postmenopausal women. J Clin Endocrinol Metab 73:1256–1261. 488. Bloom WL, Flinchum D 1960 Osteomalacia and pseudofractures caused by ingestion of aluminum hydroxide. JAMA 174:1227–1230. 489. Chines A, Pacifi R 1990 Antacid and sulcrafate-induced hypophosphatemic osteomalacia. A case report and review of the literature. Calcif Tissue Int 47:291–295. 490. Woodson GC 1998 An interesting case of osteomalacia due to antacid use associated with stainable bone aluminum in a patient with normal renal function. Bone 22:695–698. 491. Pattaragarn A, Alon US 2001 Antacid-induced rickets in infancy. Clin Pediatr 40:389–393. 492. Shah SC, Sharma RK, Chitle H, Chitle AR 1981 Rifampicininduced osteomalacia. Tubercle 62:207–209. 493. Brodie MJ, Boobis AR, Dollery CT, Hillyard CJ, Brown DJ, Maclntyre I, Park BK 1980 Rifampicin and vitamin D metabolism. Clin Pharmacol Ther 27:810–814. 494. Onhaus E, Park BK 1979 Measurement of urinary 6-/3hydroxycortisol excretion as an in vivo parameter in the clinical assessment of the microsomal enzyme-inducing capacity of antipyrine, phenobarbitone, and rifampicin. Eur J Clin Pharmacol 15:139–145. 495. Madhok TC, Schnoes HK, DeLuca HF 1978 Incorporation of oxygen-18 into the 25-position of cholecalciferol by hepatic cholecalciferol 25-hydroxylase. Biochem J 175:479–482. 496. Kutt H, Verebely K, McDowell F 1968 Inhibition of diphenylhydantoin metabolism in rats and in rat liver microsomes by antitubercular drugs. Neurology 18:706–710. 497. Ghazarian JG, DeLuca HF 1977 Kidney microsomal metabolism of 25-hydroxyvitamin D3. Biochem Biophys Res Commun 75:550–555. 498. Brodie MJ, Boobis AR, Hillyard CJ, Abeyasekera G, Stevenson JC, Maclntyre I, Park BK 1982 Effect of rifampicin and isoniazid on vitamin D metabolism. Clin Pharmacol Ther 32:525–530. 499. Toppet M, Vainsel M, Vertongen F, Fuss M, Cantraine F 1988 Evolution sequentielle des metabolites de la vitamine D sous isoniazide et rifampicine. Arch Francaises Pediatr 45:145–148. 500. Saggese G, Cesaretti G, Bertelloni S, Morganti E, Bottone E 1985 Isoniazid and vitamin D metabolism. In: Norman AW, Schaefer K, Grigoleit H-G, Herrath DV (eds) Vitamin D: Chemical, Biochemical, and Clinical Update. de Gruyter: Berlin, pp. 1123–1124. 501. Perry W, Brown J, Erooga MA, Stamp TCB 1982 Calcium metabolism during rifampicin and isoniazid therapy for tuberculosis. J R Soc Med 75:533–536. 502. Williams SE, Wardman AG, Taylor GA, Peacock M, Cooke NJ 1985 Long-term study of the effect of rifampicin and isoniazid on vitamin D metabolism. Tubercle 66:49–54. 503. Martinez ME, Gonzalez J, Sanchez MJ, Pena JM, Vasquez JJ 1996 Remission of hypercalcuria in patients with tuberculosis after treatment. Calcif Tissue Int 59:17–20. 504. Daniell HW 1976 Osteoporosis of the slender smoker. Arch Intern Med 136:298–304. 505. Heany RP, Recker RR 1982 Effects of nitrogen, phosphorus, and caffeine on calcium balance in women. J Clin Lab Med 99:46–55. 506. Heaney RP 2002 Effects of caffeine on bone and calcium economy. Food and Chem Toxicol 40:1263–1270. 507. Barger-Lux MJ, Heaney RP 1995 Caffeine and the calcium economy revisited. Osteopor Internat 5:97–102.

1289

508. Hannan MT, Felson DT, Dawson-Hughes B, Tucker KL, Cupples LA, Wilson PW, Kiel DP 2000 Risk factors of longitudinal bone loss in elderly men and women. The Framingham Osteoporosis Study. J Bone Miner Res 14:710–720. 509. Rico H, Canal ML, Manas P, Lavado JM, Costa C, Pedrera JD 2002 Nutrition 18:189–193. 510. Lloyd T, Rollings N, Eggli DF, Kieselhorst K, Chinchilli VM 1997 Dietary caffeine intake and bone status of postmenopausal women. Am J Clin Nutr 65:1826–1830. 511. Rapuri PR, Gallagher JC, Kinyamu HK, Ryschon KL 2001 Caffeine intake increases the rate of bone loss in elderly women and interacts with the vitamin D receptor genotypes. Am J Clin Nutr 74:694–700. 512. Mitoma C, Lombrozo L, LeValley SE, Dehn F 1969 Nature of the effect of caffeine on the drug metabolizing enzymes. Arch Biochem Biophys 134:434–441. 513. Thithapandha A, Chaturapit S, Limlomwongse L, Sobhon P 1974 The effects of xanthines on mouse liver cell. Arch Biochem Biophys 161:178–186. 514. Taft JL, French M, Danks JA, Larkins RG 1984 Opposing actions of methylxanthines and dibutyryl cAMP on 1,25 dihydroxyvitamin D3 production and calcium fluxes in isolated chick renal tubules. Biochem Biophys Res Commun 121:355–363. 515. Yeh JK, Aloia JF, Semla HM, Chen SY 1986 Influence of injected caffeine on the metabolism of calcium and the retention and excretion of sodium, potassium, phosphorus, magnesium, zinc, and copper in rats. J Nutr 116:273–280. 516. Yeh JK, Aloia JF 1986 Differential effects of caffeine administration on calcium and vitamin D metabolism in young and adult rats. J Bone Miner Res 1:251–258. 517. Glajchen N, Ismail F, Epstein S, Jowell PS, Fallen M 1988 The effect of chronic caffeine administration on serum markers of bone mineral metabolism and bone histomorphometry in the rat. Calcif Tissue Int 43:277–280. 518. Kulkowski JA, Chan T, Martinez J, Ghazarian JG 1979 Modulation of 25-hydroxyvitamin D3-24-hydroxylase by aminophylline: A cytochrome P-450 monooxygenase system. Biochem Biophys Res Commun 90:50–57. 519. Pedersen JI, Shobaki HH, Holmberg I, Bergseth S, Bjorkhem I 1983 25-Hydroxyvitamin D3-24-hydroxylase in rat kidney mitochondria. J Biol Chem 258:742–746. 520. Fortenbery EJ, McDermott MT, Duncan WE 1999 Effect of theophylline on calcium metabolism and circulating vitamin D metabolites. J Bone Miner Res 5:321–324. 521. Lohmann SM, Miech RP 1976 Theophylline metabolism by the rat liver microsomal system. J Pharmacol Exp Ther 196:213–225. 522. Colin A, Kraiem Z, Kahana KL, Hochberg Z 1984 Effects of theophylline on urinary excretion of cyclic AMP, calcium, and phosphorus. Miner Electrolyte Metab 10:359–361. 523. McPherson ML, Prince SR, Atamer ER, Maxwell DB, Ross-Clunis H, Estep HL 1986 Theophylline-induced hypercalcemia. Ann Intern Med 105:52–54. 524. Movsowitz C, Epstein S, Fallen M, Ismail F, Thomas S 1988 Cyclosporin A in vivo produces severe osteopenia in the rat: Effect of dose and duration of administration. Endocrinology 123:2571–2577. 525. Movsowitz C, Epstein S, Ismail F, Fallon M, Thomas S 1989 Cyclosporin A in the oophorectomized rat: Unexpected severe bone resorption. J Bone Miner Res 4:393–398. 526. Schlosberg M, Movsowitz C, Epstein S, Fallon MD, Thomas S 1989 The effect of cyclosporin A administration and its

1290

527.

528.

529.

530.

531. 532. 533.

534.

535.

536. 537.

538.

539.

540.

541.

542.

543.

withdrawal on bone mineral metabolism in the rat. Endocrinology 124:2179–2184. Stein B, Halloran BP, Reinhardt G, Engstrom GW, Bales CW, Drezner MK, Currie KL, Takizawa M, Adams JS, Epstein S 1991 Cyclosporin-A increases synthesis of 1,25-dihydroxyvitamin D3 in the rat and mouse. Endocrinology 128: 1369–1373. Katz I, Li M, Joffe I, Stein B, Jacobs T, Liang X, Ke H, Jee W, Epstein S 1994 The influence of age on cyclosporin A-induced alterations in bone mineral metabolism in the rat in vivo. J Bone Miner Res 9:59–67. Grenet O, Bobadilla M, Salah-Dine C, Steiner S 2000 Evidence for the impairment of the vitamin D activation pathway by cyclosporine A. Biochemical Pharmacology 59:267–272. Buchinsky FJ, Ma YF, Mann G, Rucinski B, Bryer HP, Romero DF, Jee WSS, Epstein S 1996 T-lymphocytes play a critical role in the development of cyclosporine-induced osteopenia. Endocrinology 137:2278–2285. Rooth P, Dawidson I, Diller K, Clothier N 1988 In vivo fluorescence microscopy reveals cyclosporine G to be less nephrotoxic than cyclosporine A. Trans Proc 20:707–709. Tejani A, Lancman I, Pomarantz A, Khawar M, Chen C 1988 Nephrotoxicity of cyclosporine A and cyclosporine G in a rat model. Transplantation 45:184–187. Jacobs TW, Katz IA, Joffe II, Stein B, Takizawa M, Epstein S 1991 The effect of FK 506, cyclosporine A, and cyclosporine G on serum 1,25-dihydroxyvitamin D3 levels. Trans Proc 23: 3188–3189. Stein B, Takizawa M, Schlosberg M, Movsowitz C, Fallon M, Berlin JA, Epstein S 1992 Evidence that cyclosporine G is less deleterious to rat bone in vivo than cyclosporine A. Transplantation 53:628–632. Cvetkovic M, Mann GN, Romero DF, Liang X, Ma YF, Jee WSS, Epstein S 1994 The deleterious effects of long-term cyclosporin A, cyclosporin G, and FK506 on bone mineral metabolism in vivo. Transplantation 57:1231–1237. Morris R 1994 Modes of action of FK506, cyclosporin A, and rapamycin. Trans Proc 26:3272–3275. Bryer HP, Isserow JA, Armstrong EC, Mann GN, Rucinski B, Buchinsky FJ, Romero DF, Epstein S 1995 Azathioprine alone in bone sparing and does not alter cyclosporin Ainduced osteopenia in the rat. J Bone Miner Res 10:132–138. Romero DF, Buchinsky FJ, Rucinski B, Cvetkovic M, Bryer HP, Liang XG, Ma YF, Jee SS, Epstein S 1995 Rapamycin: A bone sparing immunosuppressant? J Bone Miner Res 10:760–768. Joffe I, Katz I, Sehgal S, Bex F, Kharode Y, Tamasi J, Epstein S 1993 Lack of change of cancellous bone volume with shortterm use of the new immunosuppressant rapamycin in rats. Calcif Tissue Int 53:45–52. Monier-Faugere MC, Mawad H, QI Q, Friedler RM, Malluche HH 2000 High prevalence of low bone turnover and occurrence of osteomalacia after kidney transplantation. J Am Soc Nephrol 11:1093–1099. Saha HHT, Salmela KT, Ahonen PJ, Pietila KO, Morsky PJ, Mustonen JT, Lalla MLT, Pasternak AI 1994 Sequential changes in vitamin D and calcium metabolism after successful renal transplantation. Scand J Urol Nephrol 28:21–27. Riancho JA, de Francisco ALM, del Arco C, Amado JA, Cotorruelo JG, Arias M, Gonzalez-Marcias J 1988 Serum levels of 1,25-dihydroxyvitamin D after renal transplantation. Miner Electrolyte Metab 14:332–337. Valero MA, Loinaz C, Larrodera L, Leon M, Moreno E, Hawkins F 1995 Calcitonin and bisphosphonates treatment

SOL EPSTEIN AND ADINA E. SCHNEIDER

544. 545.

546. 547.

548.

549.

550.

551.

552. 553. 554.

555.

556.

557.

558. 559.

560.

in bone loss after liver transplantation. Calcif Tissue Int 57:15–19. McDonald JA, Dunstan CR, Dilworth P, Sherbon K, Ross Sheil AG, Evans RA, McCaughan GW 1991 Bone loss after liver transplantation. Hepatology 14:613–619. Shane E, Rivas M, Staron RB, Silverberg SJ, Seibel MJ, Kuiper J, Mancini D, Addresso V, Michler RE, Factor-Litvak P 1996 Fracture after cardiac transplantation: A prospective longitudinal study. J Clin Endocrinol Metab 81:1740–1746. Cundy T, Kanis JA, Heynen G, Morris PJ, Oliver DO 1983 Calcium metabolism and hyperparathyroidism after renal transplantation. Q J Med 205:67–78. Felsenfeld AJ, Gutman RA, Drezner M, Llach F 1986 Hypophosphatemia in long-term renal transplant recipients: Effects on bone histology and 1,25-dihydroxycholecalciferol. Miner Electrolyte Metab 12:333–341. Sakhee K, Brinker K, Helderman JH, Bengfort JL, Nicar MJ, Hull AR, Pak CYC 1985 Disturbances in mineral metabolism after successful renal transplantation. Miner Electrolyte Metab 11:167–172. Reichel H, Griibinger A, Knehans A, Kiihn K, Schmidt-Gayk H, Ritz E 1992 Long-term therapy with cyclosporin A does not influence serum concentrations of vitamin D metabolites in patients with multiple sclerosis. Clin Invest 70:595–599. Hanley DA, Ayer LM, Gundberg CM, Minuk GY 1991 Parameters of calcium metabolism during a pilot study of cyclosporin A in patients with symptomatic primary biliary cirrhosis. Clin Invets Med 14:282–287. Stempfle H, Werner C, Siebert U, Assum T, et al. 2002 The role of tacrolimus-based immunosuppression on bone density and bone turnover after cardiac transplantation: a prospective, longitudinal, randomized, double-blind trial with calcitriol. Transplantation 73:547–552. Epstein S, Shane E 1996 Transplantation osteoporosis. In: R Marcus, D Feldman, J Kelsey (eds) Osteoporosis Academic Press: New York, pp. 947–957. Shane E, Epstein S 1994 Immunosuppressive therapy and the skeleton. Trends Endocrinol Metab 4:169–175. Turner RT, Francis R, Brown D, Garand J, Hannon KS, Bell NH 1989 The effects of fluoride on bone and implant histomorphometry in growing rats. J Bone Miner Res 4: 477–484. Manzke E, Rawley R, Vose G, Roginsky M, Rader JI, Baylink DJ 1977 Effect of fluoride therapy on nondialyzable urinary hydroxyproline, serum alkaline phosphatase, parathyroid hormone, and 25-hydroxyvitamin D. Metabolism 26:1005–1010. Dure-Smith BA, Parley SM, Linkhart SG, Parley JR, Baylink DJ 1996 Calcium deficiency in fluoride-treated osteoporotic patients despite calcium supplementation. J Clin Endocrinol Metab 81:269–275. Fallat RW, Glueck CJ, Lutmer R, Mattson FH 1976 Shortterm study of sucrose polyester a nonabsorbable fa-like material as a dietary agent for lowering plasma cholesterol. Am J Clin Nutr 29:1204–1215. Grouse JR, Grundy SM 1979 Effects of sucrose polyester on cholesterol metabolism in man. Metabolism 28:994–1000. Mattson FH, Volpenhein RA 1972 Rate and extent of absorption of the fatty acids of fully esterified glycerol, erythritol, xylitol, and sucrose as measured in thoracic duct cannulated rats. J Nutr 102:1177–1180. Mattson FH, Volpenhein RA 1987 Hydrolysis of fully esterified alcohols containing from one to eight hydroxyl groups by the lipolytic enzymes of rat pancreatic juice. J Lipid Res 13:325–328.

CHAPTER 74 Drug and Hormone Effects on Vitamin D Metabolism

561. Kuksis A 1987 Absorption of fat-soluble vitamins In: Kuksis A (ed) Fat Absorption, Vol 2. CRC Press: Boca Raton, Florida, pp. 65–86. 562. Lueck CJ, Hastings MM, Alien C, Hogg E, Baehler E, Gartside PS, Phillips D, Jones M, Hollenboch EJ, Braun B, Anastasia JV 1982 Sucrose polyester and covert caloric dilution. Am J Clin Nutr 35:1352–1359. 563. Mellies MJ, Vitale C, Jandacek RJ, Lamkin GE, Glueck CJ 1985 The substitution of sucrose polyester for dietary fat in obese, hypercholestrolemic outpatients. Am J Clin Nutr 41:1–12. 564. Jones DY, Miller KW, Koonsvitsky BP, Ebert ML, Lin PYT, Jones MB, DeLuca HF 1991 Serum 25-hydroxyvitamin D concentrations of free-living subjects consuming olestra. Am J Clin Nutr 53:1281–1287. 565. Mellies MJ, Jandacek RJ, Taulbee JD, Tweksbury MB, Lamkin G, Baehler L, King P, Boggs D, Goldman S, Gouge A, Tsang R, Glueck CJ 1983 A double-blind, placebo-controlled study of sucrose polyester in hypercholesterolemic outpatients. Am J Clin Nutr 37:339–346. 566. Thornquist MD, Kristal AR, Patterson RE, Neuhouser ML, Rock CL, Neumark-Sztainer D, Cheskin LJ 2000 J Nutr. 130:1711–1718. 567. Miller KW, Wood FE, Stuard SB, Alden CL 1991 A 20-month olestra feeding study in dogs. Food Chem Toxic 29:427–435. 568. Lawson DEM, Paul AA, Black AE, Cole TJ, Mandal AR, Davie M 1979 Relative contributions of diet and sunlight to vitamin D state in the elderly. Br Med J 2:303–305. 569. Poskitt EME, Cole TJ, Lawson DEM 1979 Diet, sunlight, and 25-hydroxyvitamin D in healthy children and adults. Br Med J 1:221–223. 570. Haddad JG, Hahn TJ 1973 Natural and synthetic sources of circulating 25-hydroxyvitamin D in man. Nature 244: 515–516. 571. McDuffie JR, Calis KA, Booth SL, Waifo GI, Yanovski JA 2002 Effects of orlistat on fat soluble vitamins in obese adolescents. Pharmacotherapy 814–822. 572. Gotfredsen A, Westergren Hendel H, Anderson T 2001 Influence of orlistat on bone turnover and body composition. Int J Obes Relat Metab Disord 25:1154–1160. 573. Mellerup ET, Lauritsen B, Dam H, Rafaelson OJ 1976 Lithium effects on diurnal rhythms of calcium, magnesium, and phosphate metabolism in manic-melancholic disorder. Acta Psychiat Scand 53:360–370. 574. Davis BM, Pfefferbaum A, Krutzik S, Davis KL 1981 Lithium’s effect on parathyroid hormone. Am J Psychiatry 138:489–492. 575. Plenge P, Rafaelsen OJ 1982 Lithium effects on calcium, magnesium, and phosphate in man: Effects on balance, bone

576.

577. 578. 579. 580. 581. 582. 583. 584. 585.

586. 587.

588. 589. 590. 591.

1291 mineral content, faecal, and urinary excretion. Acta Psychiat Scand 66:361–373. Mallette LE, Khouri K, Zengotita H, Hollis BW, Malini S 1989 Lithium treatment increases intact and midregion parathyroid hormone and parathyroid volume. J Clin Endocrinol Metab 68:654–660. Nielsen JL, Christiansen MS, Pedersen EB, Darling S, Amdisen A 1977 Parathyroid hormone in serum during lithium therapy. Scand J Clin Lab Invest 37:369–372. Miller PD, Dubovsky SL, McDonald KM, Arnaud C, Schrier RW 1978 Hypocalciuric effect of lithium on man. Miner Electrolyte Metab 1:3–11. Bjorum N, Hornum I, Mellerup ET, Plenge PK, Rafaelsen OJ 1975 Lithium, calcium, and phosphate. Lancet 1:1243. Birnbaum J, Klandorf H, Giuliano A, Van Herle A 1988 Lithium stimulates the release of human parathyroid hormone in vitro. J Clin Endocrinol Metab 66:1187–1191. Christiansen C, Baastrup PC, Transbol I 1976 Lithium, hypercalcemia, hypermagnesaemia, and hyperparathyroidism. Lancet 1:969. Shen F-H, Sherrard DJ 1982 Lithium-induced hyperparathyroidism: An alteration of the “set point.” Ann Intern Med 96:63–65. Christiansen C, Baastrup PC, Lindgreen P, Transol I 1978 Endocrine effects of lithium: II. Primary hyperparathyroidism. Acta Endocrinol 88:528–534. Christiansen C, Baastrup P, Transol I 1980 Development of primary hyperparathyroidism during lithium therapy. Neuropsychobiology 6:280–283. Nordenstrom J, Elvius M, Bagedahl-Strindlund M, Zhao B, Torring O 1994 Biochemical hyperparathyroidism and bone mineral status in patients treated long-term with lithium. Metabolism 43:1562–1567. Perez GO, Oster JR, Vaavioucle CA 1975 Incomplete syndrome of renal tubular acidosis induced by lithium carbonate. J Clin Lab Med 86:386–394. Nordenstrom J, Strigard K, Perbeck L, Willems J, BagedahlStrindlund M, Linder J 1992 Hyperparathyroidism associated with treatment of manic-depressive disorders by lithium. Eur J Surg 158:207–211. Mallette LE, Eichhorn E 1986 Effects of lithium carbonate on human calcium metabolism. Arch Intern Med 146:770–776. Hestbech J, Hansen HE, Amdisen A, Olsen S 1977 Chronic renal lesions following long-term treatment with lithium. Kidney Int 12:205–213 Hansen HE, Hestbech J, Sorensen JL, Norgaard K, Heiskov J, Amidisen A 1979 Chronic interstitial nephropathy in patients on long-term lithium treatment. Q J Med 192:577–591. Rosenblatt S, Chanley JD, Segal RL 1989 The effect of lithium on vitamin D metabolism. Biol Psychiatry 26:206–208.

CHAPTER 75

Bone Disorders Associated with Gastrointestinal and Hepatobiliary Disease MICHAEL DAVIES, JACQUELINE L. BERRY AND ANDREW P. MEE University of Manchester, Vitamin D Research Group, Department of Medicine, Manchester Royal Infirmary, Manchester M13 9WL, UK

I. II. III. IV.

Introduction Metabolic Disturbances in Gastrointestinal Disease Acquired Bone Disease in Gastrointestinal Disorders Gastrointestinal Conditions Associated with Bone Disorders

I. INTRODUCTION The major bone diseases associated with gastrointestinal and liver disease are osteomalacia, caused by severe and prolonged deficiency of vitamin D, which may more correctly be described as vitamin D depletion, and, more commonly, osteoporosis in which decreased bone density may be partially attributable to low calcium intake or absorption and excessive bone resorption, which may be mediated by raised PTH. This latter state may arise as a result of an insufficiency (as opposed to depletion) of vitamin D or of calcium. The gastrointestinal tract plays a primary role in the biological activity of vitamin D, so it is not surprising that diseases of the liver and intestine have consequences for vitamin D metabolism and function that may result in bone disease. The liver is the site of the first stage of the metabolic activation of vitamin D, namely the introduction of a hydroxyl group at C25, and is also important in the elimination of vitamin D metabolites in the bile (see Chapter 4). Furthermore, the liver is also the site of synthesis of the plasma vitamin D–binding protein (DBP, Chapters 8 and 9). Most importantly, the mucosa of the small intestine is a major target organ for the active metabolite of vitamin D, 1,25-dihydroxyvitamin D (1,25(OH)2D), which promotes the absorption of calcium from the diet. This is effected by genomic reactions in which 1,25(OH)2D, binding to its specific intranuclear receptor mediates the transcription of genes for proteins involved in calcium transport; recent evidence also supports the concept of a rapid nongenomic action on calcium flux involving a membrane receptor (see Chapters 11, 23, 24, 42). VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Liver Disease VI. Summary References

II. METABOLIC DISTURBANCES IN GASTROINTESTINAL DISEASE A. Development of Vitamin D Deficiency and Depletion 1. CONTROLLING FACTORS IN THE SUPPLY OF VITAMIN D

The physiological source of vitamin D is solar irradiation of the precursor 7-dehydrocholesterol in skin to give vitamin D3, but both vitamin D2 and D3 can also be supplied in the diet. There is presently no agreed-upon definition of a sufficient level of vitamin D, in terms of serum 25-hydroxyvitamin D (25OHD) concentration (see discussion in Chapters 46, 61, and 62). However, the idea that an adequate vitamin D status can be defined as the absence of osteomalacia (see Section IIA1b) has been largely rejected. It is now evident that there is an intermediate state in which vitamin D levels are insufficient, but not frankly depleted. Such a state is associated with increased levels of parathyroid hormone (PTH), and is characterized by increased synthesis of 1,25(OH)2D, presumably driven by PTH, when a dose of vitamin D is given [1,2]. Because the suppression of PTH is seen as beneficial for bone, many now regard serum 25OHD concentrations of 50–75 nmol/l [3,4] or even higher as desirable [5], since at these concentrations PTH approaches a minimum in relation with 25OHD. A recent paper by Veith et al. [6], however, found no such plateau in PTH with increasing 25OHD, but demonstrated that, irrespective of age, PTH concentration decreased as 25OHD increased, although the decline in PTH was more gradual in the older Copyright © 2005, Elsevier, Inc. All rights reserved.

1294

MICHAEL DAVIES, JACQUELINE L. BERRY, AND ANDREW P. MEE

patients for the same level of 25OHD. They suggested that to ensure that all adults have 25OHD levels of at least 40 nmol/l they should consume, at least, 20–25 µg (800–1000IU) of vitamin D daily. This is consistent with the findings of Glerup et al. [7] in their study of Danish women, who also proposed that adults with limited sunlight exposure should receive 800–1000 IU/day. a. The Importance of the Intestine Gastrointestinal and hepatobiliary diseases are characterized by general problems with absorption of lipids, whether because of poorly functioning enterocytes, a reduction in their number, or lack of biliary secretion; as vitamin D is a lipid soluble molecule, malabsorption of dietary vitamin D may then contribute to vitamin D deficiency [8–10]. However, for people in those parts of the world where there is little fortification of foodstuffs with the vitamin, endogenous synthesis of vitamin D in the skin, as a result of solar exposure, is the major source of the vitamin [11,12] (also see Chapters 3 & 47). In this case, the diet supplies only a fraction, often less than 25–30%, of the daily requirement (see Fig. 1). Thus, in normal circumstances, partial impairment of dietary absorption would not be expected to have a major effect on vitamin D status. However, ill health, such as may be encountered in patients with serious gastrointestinal or liver disease, may lead to a reluctance or inability to spend time outdoors, thus decreasing the endogenous synthesis of vitamin D. When the opportunity to make vitamin D in the skin is minimal, provision of the vitamin in the diet, however little, becomes crucial [13], and any impairment of absorption is more likely to have a significant effect. Factors affecting the elimination of vitamin D by the liver may

Number of subjects

30

20

also play a critical role in the face of gastrointestinal and hepatobiliary disease and may contribute to a state of “acquired vitamin D deficiency,” see Section IIA2c. b. The Significance of Changes in 1,25-dihydroxyvitamin D Concentration Vitamin D deficiency has been defined in various ways, the most extreme of which is the development of osteomalacia or rickets. An earlier indication is the measurement of serum 25OHD levels; concentrations associated with the clinical signs of deficiency are usually associated with levels below 12–20 nmol/l [14,15]. Also, biochemical indices such as the plasma concentrations of calcium, inorganic phosphate, parathyroid hormone, alkaline phosphatase, or osteocalcin are sometimes used. Ultimately, vitamin D deficiency must be characterized by the inability to synthesize sufficient 1,25(OH)2D to enable adequate absorption of calcium from the intestine or to stimulate osteoblastic activity, and this state can then be defined as vitamin D depletion. During the development of vitamin D deficiency, whether arising from a lack of solar exposure, an inadequate dietary supply, or intestinal malabsorption, the concept of a normal range for the serum 1,25(OH)2D concentration becomes invalid. The process can be envisaged as a progressive descending spiral in which low 25OHD results in a reduction of 1,25(OH)2D synthesis, thus impairing calcium absorption (Fig. 2). The tendency for calcium to fall stimulates PTH secretion by the parathyroid glands, and the trophic action of PTH on the renal 1α-hydroxylase increases 1,25(OH)2D synthesis, thus temporarily increasing the efficiency of intestinal calcium absorption. However, increased 1,25(OH)2D synthesis depletes still further a diminishing supply of 25OHD, and the process will continue, (in the absence of a new supply of vitamin D) until there is insufficient substrate to provide the level of 1,25(OH)2D required for adequate absorption of calcium, despite an increasing degree of secondary hyperparathyroidism. There is also evidence (Section IIA2c) that a raised concentration of 1,25(OH)2D may itself influence the process, by shortening the serum half-life of 25OHD, thus accelerating the onset of vitamin D deficiency. 2. THE ROLE OF THE LIVER

10

0

.

.

. 100 .

.

. 200 .

.

.

300 . >400

Vitamin D (IU/day)

FIGURE 1

Daily dietary intake of vitamin D in a sample of Caucasians (light bars) and Asian immigrants (dark bars) living in northwest England.

The liver plays two major roles in vitamin D metabolism; 25-hydroxylation of vitamin D occurs in the endoplasmic reticulum of hepatocytes, and the liver is a major excretory organ, both of vitamin D and 25OHD (when these are present in excessive amounts), but mainly of water soluble compounds, principally glucuronide conjugates, all of which are present in bile [16–19]. There is a minimal effect on 25-hydroxylation in the liver if hepatocellular function is maintained, (Section V), and impairment of 25-hydroxylation is not

CHAPTER 75 Bone Disorders Associated with Gastrointestinal and Hepatobiliary Disease

1295

VITAMIN D SUFFICIENCY

Lowered concentration of plasma concentration of 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D

Vitamin D supply diminished by lack of dietary vitamin, malabsorption of dietary vitamin or decreased exposure to sunlight

Decreased intestinal absorption of calcium Increased synthesis and secretion of PTH

Further decrease in 25hydroxyvitamin D through increased turnover and increased consumption

Failure to absorb calcium Calcium resorbed from bone Bone unable to mineralise

Increased synthesis of 1,25-dihydroxyvitamin D corrects defective calcium absorption; gives normocalcaemia at expense of PTH VITAMIN D INSUFFIENCY Insufficient 25-hydroxyvitamin D to form adequate 1,25-dihydroxy vitamin D

VITAMIN D DEPLETION Osteomalacia and rickets

FIGURE 2

Spiral of developing vitamin D deficiency showing progression through a state of vitamin D insufficiency, characterized by secondary hyperparathyroidism and increased synthesis of 1,25-dihydroxyvitamin D.

believed to make a major contribution to the loss of vitamin D activity until the later stages of disease. a. Biliary Secretion of Vitamin D Metabolites In normal subjects, the biliary-fecal route appears to be the major excretory pathway for vitamin D, but precise chemical identification of many of the excretory products is still awaited. The recognition of the importance of the renal 24-hydroxylase in producing metabolites such as 24,25-dihydroxyvitamin D (24,25(OH)2D), which can then undergo sidechain cleavage and form carboxylic acids similar to bile acids [20,21], raises the question as to whether urine may be a major excretory route for water soluble vitamin D metabolites. However, the results of tracer experiments in humans show in general, preferential excretion via the intestinal tract, and hence do not support this argument [16–19]. Some experiments in rats seem to indicate considerable urinary excretion in this species [22], but in humans the balance of evidence is that the fecal excretion is much more important, except in the particular case of biliary obstruction in which metabolites

normally excreted by the biliary-fecal route are instead passed through the kidneys, see Section V,B. b. The Enterohepatic Circulation of Vitamin D and Its Metabolites The polar vitamin D derivatives in bile probably have no significant biological activity and only insignificant amounts of unchanged vitamin D or 25OHD are usually present [16–19]. Nevertheless, the idea that biologically active metabolites of vitamin D might enter a conservative enterohepatic circulation has proved an attractive one to attempt to explain the acquired vitamin D deficiency associated with various gastrointestinal diseases in which lipid absorption is impaired. It has been suggested that interruption of such a circulation would explain the wastage of vitamin D that occurs in conditions such as celiac disease and gastric or intestinal resection. However, most studies that appear to support such a hypothesis have used an intravenous bolus dose of labeled vitamin D or 25OHD [23]. In these circumstances, especially if a nonphysiological vehicle has been used, the labeled compound may be seen as a foreign compound and

cleared rapidly by the liver, mainly as hydrophilic metabolites, but also, immediately after injection, in the form of unchanged sterol. It has been argued that vitamin D, absorbed from the diet and entering the circulation rapidly in the form of chylomicrons, constitutes a nonphysiological presentation to the liver. The relative inefficiency of oral vitamin D, as opposed to that synthesized in the skin and transported to the liver on DBP, has been attributed to such a mechanism [24]. In a study often quoted to support the enterohepatic circulation hypothesis, a high proportion of injected radioactivity given as 25OHD3 was recovered in the bile, and was assumed still to be present in that form [23]. In fact, by comparison with other studies, it is most likely that, after the initial equilibration period when the 25OHD would have become bound to DBP, the label would be largely in the form of polar derivatives. Experiments in rats have demonstrated that if biliary excretion products of isotopically labeled vitamin D are collected and given to other animals, they can be absorbed and re-excreted in the bile [25]. There is little evidence, however, that these metabolites have biological activity; any conjugates would need to be cleaved enzymatically to release the vitamin D moiety, which is itself likely to be a catabolic product with low activity. It is, therefore, unlikely that any enterohepatic circulation of vitamin D metabolites that does occur is of a conservative nature or of physiological significance. Indeed, biliary excretion of vitamin D metabolites appears to be dose-dependent, increasing both in humans and rats when vitamin D is plentiful, and being suppressed in vitamin D deficiency [18,25]. An alternative hypothesis for the significance of the biliary secretion of vitamin D compounds is that this mechanism enables the body to dispose of a highly potent biological compound if present in excess amounts, and may be seen as a form of detoxification, rather than an attempt to conserve a scarce resource [24–26]. The potential importance of bile acids as ligands for the VDR is discussed in Chapter 53. c. The Development of Acquired Vitamin D Deficiency The concept of interruption of an enterohepatic circulation of vitamin D metabolites is also inadequate to explain the development of vitamin D deficiency in patients with gastrointestinal and hepatobiliary disease, see above [19,24]. In addition to patients with fat malabsorption, various other conditions are characterized by wasting of vitamin D. These include primary hyperparathyroidism, anticonvulsant therapy, and high cereal and high fiber diets. A theory has been proposed by Clements et al. [27] in an attempt to explain this phenomenon. In the rat, these authors showed that withdrawal of calcium from the diet increased the

MICHAEL DAVIES, JACQUELINE L. BERRY, AND ANDREW P. MEE

plasma clearance of 25OHD; a similar effect was observed when phenobarbitone was administered to animals on a normal calcium diet, and also when fiber was added to the diet. In addition, the plasma half-life of 25OHD was decreased by secondary hyperparathyroidism. Parallel studies on human subjects have shown that both primary and secondary hyperparathyroidism can increase the catabolism of 25OHD3 [28,29]. In all these studies, there was a strong inverse relationship between the plasma half-life of 25OHD and the prevailing concentration of 1,25(OH)2D, (see Fig. 3). In those conditions in which 1,25(OH)2D was raised, 3H-25OHD was removed from the circulation more rapidly, and there was a corresponding increase in fecal radioactivity [28]. The reasons for these phenomena are not fully understood, although it has been postulated that 1,25(OH)2D may stimulate, perhaps by a receptor-mediated mechanism, the hepatic conjugation reactions that render vitamin D and 25OHD water soluble [29]. In addition, 1,25(OH)2D is known to stimulate the renal 24-hydroxylase, which initiates the sidechain cleavage cascade, though the subsequent metabolic fate of these compounds in humans is unclear. The relationship between 1,25(OH)2D level and the turnover of 25OHD is obtained in a variety of gastrointestinal conditions in which, as a result of problems with calcium absorption leading to secondary hyperparathyroidism, serum 1,25(OH)2D is raised (see Sections IIB and III). The phenomenon of increased loss of plasma 25OHD in patients, in whom either dietary intake of the vitamin or solar exposure is inadequate, could help to explain the prevalence of vitamin D deficiency in gastrointestinal disease.

t1/2 3H-25-hydroxyvitamin D (days)

1296

45 40 35 30 25 20 15 10 5 0 0

50 100 Plasma 1,25-dihydroxyvitamin D (pg/ml)

150

FIGURE 3 Strong inverse relationship between the plasma halflife of 25-hydroxyvitamin D and the prevailing concentration of 1,25-dihydroxyvitamin D, rs = −0.706, P < 0.001. Data from studies reported by Clements et al. [28,29].

CHAPTER 75 Bone Disorders Associated with Gastrointestinal and Hepatobiliary Disease

B. Malabsorption of Calcium The development of osteomalacia in patients with steatorrhea is usually attributed to malabsorption of dietary vitamin D, but the degree of malabsorption is rarely severe enough to account for the development of vitamin D deficiency in populations with a high oral intake of the vitamin. Since the small intestine is a target organ for vitamin D, it is possible that the functional activity of 1,25(OH)2D may be compromised (Chapter 24). Colston et al. [30] concluded that, although VDR were abundant in crypt cells from patients with celiac disease, malabsorption of calcium resulted from the loss of vitamin D–regulated proteins and enzymes located in the most mature enterocytes of the mid and tip villous regions. Absorption of dietary calcium has been found to be deficient in various types of gastrointestinal disease, including celiac disease, and Crohn’s disease, in which fat malabsorption may lead to the formation of calcium soaps. The phenomenon may also be observed following gastrectomy and intestinal bypass surgery, where fat malabsorption is not a problem, see Section III. The resulting tendency to hypocalcemia with the induction of secondary hyperparathyroidism has been shown in many of these conditions to lead to raised levels of 1,25(OH)2D, the condition which is known to lead to wastage of 25OHD and the development of vitamin D deficiency, as discussed previously in Section IIA2c. Even with an adequate supply of vitamin D, malabsorption of calcium can lead to secondary hyperparathyroidism; vitamin D deficiency will not occur, but the skeleton may be subject to the effects of high levels of PTH for many years. It is recognized that PTH excess is detrimental to the skeleton, especially at cortical sites. Any increase in bone turnover will also exacerbate the remodeling imbalances that accompany the bone loss associated with aging. Raised levels of 1,25(OH)2D also stimulate bone resorption and may be deleterious to the skeleton. These phenomena may explain why the principal bone disease in patients with gastrointestinal disease is osteoporosis.

III. ACQUIRED BONE DISEASE IN GASTROINTESTINAL DISORDERS

1297

and easy fracture. Varying degrees of perturbation of calcium and vitamin D metabolism occur in patients with gastrointestinal disease, and the extent, duration, and severity of these derangements are important in determining the nature of any underlying bone disease. Since the major source of vitamin D is cutaneous synthesis rather than the diet (Section IIA1), adequate solar exposure will protect against vitamin D deficiency.

A. The Role of Secondary Hyperparathyroidism It is now well recognized that certain diseases of the gastrointestinal tract are accompanied by secondary hyperparathyroidism with an increase in the circulating concentration of 1,25(OH)2D. It is assumed that these changes are an adaptation to the calcium malabsorption that has been documented in many diseases affecting the gut. Secondary hyperparathyroidism may be successful in correcting calcium malabsorption provided there is a sufficient supply of vitamin D. However, if there is an increase in the catabolism of vitamin D (Section IIA2c), the body may become depleted of vitamin D, the amount of 1,25(OH)2D will be insufficient for normal calcium absorption, and so secondary hyperparathyroidism will be intensified (Fig. 2). This leads to an increase in bone resorption and an increase in remodeling imbalance. In addition, the effects of excess PTH on the kidney will produce phosphate wastage and a reduction in the serum phosphate concentration. These changes, if present for a prolonged period of time, result in the development of osteomalacia due to insufficient 1,25(OH)2D and mineral for normal bone formation. If the supply of vitamin D from diet or solar exposure is adequate for the increased demands resulting from calcium malabsorption and secondary hyperparathyroidism, then osteomalacia will not occur, but the skeleton will be subjected to the effects of excess PTH perhaps over many years. This can lead to loss of bony tissue, particularly at cortical sites with the development of osteoporosis and increased tendency to fracture.

B. The Nature of the Bone Disease Osteomalacia has been associated with diseases of the gastrointestinal tract for many years, but the pathophysiological mechanisms responsible for this bone disease have only recently been recognized. Furthermore, the mechanisms that lead to osteomalacia may also lead to loss of skeletal tissue, osteoporosis,

While osteomalacia may occur in patients with gastrointestinal disease, osteoporosis is far more common; this may be explained in part by the state of vitamin D nutrition in patients with gut disease. Where the amount of solar exposure is limited by geographical

1298 latitude, then vitamin D deficiency and osteomalacia are more likely to be seen. Following intestinal bypass surgery for obesity, it has been estimated that 12% of patients in Europe and 4% in the United States of America develop osteomalacia [31]. This reinforces the importance in Northern Europe, including the United Kingdom, of extrinsic (privational) vitamin D deficiency in determining the development of osteomalacia in intestinal [32] and hepatic [33] disorders. The frequency of osteomalacia in gastrointestinal and hepatobiliary disease has previously been greatly overestimated. This has resulted from misinterpretation of both biochemical and histological findings. Although a raised serum alkaline phosphatase, often accompanied by serum calcium values in the lower part of the normal range, may be an early sign of underlying osteomalacia, the same biochemical changes are to be found in other clinical situations and may not be due to disturbances in calcium and vitamin D metabolism. Both hepatic and intestinal isoforms of alkaline phosphatase may be increased in certain diseases, and to be certain that a raised level of alkaline phosphatase arises from bone, it is necessary to use a bone-specific assay. Hyperosteoidosis due to secondary hyperparathyroidism has been misinterpreted as osteomalacia, as have been measurements of increased osteoid volume. Clinically, skeletal pain in a patient with primary biliary cirrhosis is more likely to arise from multiple osteoporotic related fractures than from osteomalacia. Osteoporosis is therefore by far the most common metabolic bone disease complicating disorders of the liver and gastrointestinal tract. There is a high incidence of fractures (especially vertebral) in patients with chronic liver disease or a past history of partial gastrectomy. Bone densitometry has demonstrated a significant reduction in bone mineral content in both the axial and appendicular skeleton, even in asymptomatic patients.

C. The Role of Factors Other Than Vitamin D and Calcium Gastrointestinal diseases may result in bone disease by mechanisms other than disturbances in calcium and vitamin D metabolism. Protein deficiency is common in some diseases of the liver and small bowel and may adversely affect bone [34]. Hypoalbuminemia can depress osteoblast function, and other nutrients not immediately considered relevant to the skeleton may be important in bone cell function [35]. Deficiencies of other vitamins, including vitamins A, C, and K may affect bone, although their significance in intestinal disease is not fully established. Vitamin C is a co-factor

MICHAEL DAVIES, JACQUELINE L. BERRY, AND ANDREW P. MEE

for prolyl hydroxylase, which is necessary for the formation of stable collagen polymers. And two proteins, bone Gla protein (osteocalcin) and matrix Gla protein are generated by vitamin K–dependent enzymes. Deficiencies in vitamin A levels have been demonstrated in one third of patients with primary biliary cirrhosis [10].

D. The Development of Metabolic Bone Disease in Gastrointestinal Disease In many situations, however, an absolute or relative deficiency of calcium or vitamin D will be pivotal to the development of metabolic bone disease. Rao et al. [36] studied the histological evolution of vitamin D deficiency in patients with intestinal malabsorption and a plasma 25OHD of less than 25 nmol/l. In early vitamin D depletion, there is a lack of clinical symptoms, although osteopenia may be present with histological evidence of hyperparathyroidism; there may also be an increased risk of fracture. As the severity increases, mineralization becomes more defective, bone formation rate declines, osteoid tissue increases in both thickness and surface extent, and changes of hyperparathyroidism become more severe [36]. Bone disease resulting from intestinal disease will display features of osteopenia, hyperosteoidosis, osteitis fibrosa, and osteomalacia. The extent and degree of these changes will be determined by the duration and severity of vitamin D deficiency, calcium malabsorption, and accompanying secondary hyperparathyroidism. Additional factors such as the use of corticosteroids or parenteral nutrition (see Section IV,F) may also affect bone. Since the mucosal barrier is often compromised in small bowel disease, it may also be possible for unspecified nutritional or perhaps toxic substances to affect bone. Histomorphometric analysis of transiliac bone biopsy material has revealed several differing abnormalities, some of which represent the evolution of osteomalacia [31] (see Chapter 63). The first stage in the development of osteomalacia is secondary hyperparathyroidism where the osteoid surface and volume are increased, but osteoid thickness and mineralization lag times are normal. Some patients with vitamin D deficiency show reduced adjusted appositional rate and prolongation of the mineralization lag time. This latter condition is described as hypovitaminosis D osteopathy type I (HVO-I) by Parfitt [31], who divides osteomalacia into HVO-II and HVO-III when osteoid thickness exceeds 15 µm, and mineralization lag time exceeds 100 days. In HVO-III there is no mineralization occurring (using double tetracycline labels), while in HVO-II, mineralization still occurs. Atypical and focal

CHAPTER 75 Bone Disorders Associated with Gastrointestinal and Hepatobiliary Disease

osteomalacias are also seen occasionally. Florid hyperparathyroidism can result in high turnover osteoporosis, but low turnover osteoporosis is perhaps the most common lesion seen. Osteoid thickness is normal or reduced, bone formation and appositional rates are reduced, and there is little or no evidence of hyperparathyroidism. Low turnover osteoporosis is often associated with general undernutrition and protein malnutrition.

IV. GASTROINTESTINAL CONDITIONS ASSOCIATED WITH BONE DISORDERS The various pathophysiological mechanisms resulting in metabolic bone disease have been discussed earlier. Whether any gastroenterological disturbance produces bone disease will depend upon the extent to which calcium and vitamin D metabolism are disturbed. Thus, disturbances of colonic function are unlikely to affect calcium metabolism, but since inflammatory bowel disease of the colon is commonly treated by systemic steroids, treatment of colonic disease may impact upon the skeleton (see Chapters 73 and 74).

A. Post Gastrectomy Bone Disease 1. THE PATTERN OF BONE DISEASE

a. The Incidence of Osteomalacia The changing patterns in both incidence and medical management of peptic ulcer disease has meant that the problem of post gastrectomy bone disease is in decline, and this trend is likely to continue as the need for stomach resection is reduced. However, surgery to alleviate problems with obesity can also lead to problems with bone disease [37]. Osteopenia or osteoporosis is far more common than frank osteomalacia. The overall reported incidence of osteomalacia varies in different centers, and this is in part explained by the differing criteria used for diagnosing osteomalacia. In a survey of 1228 patients following partial gastrectomy, Paterson et al. [38] found only 6 cases of osteomalacia using clinical, biochemical, and histological criteria. Tovey et al. [39] found osteomalacia in 10 of 227 postgastrectomy subjects, but bone biopsies were performed in only 15 of the patients. When more extensive use is made of bone histology, the incidence of diagnosed osteomalacia tends to increase, and in some series is over 20% [40,41]. Despite the numerous explanations for postgastrectomy bone disease, one unanswered question is why the problem of osteomalacia is greater in women when 80% of gastrectomy patients are men. However, bone disease related to gastrectomy does not develop for several years after surgery, and

1299

since many of those affected are middle-aged or elderly, it may not be easy to uncouple the effects of the menopause and the aging process upon the skeleton from those of gastric surgery. b. Vitamin D Metabolism Osteomalacia responds to small doses of vitamin D [42], and vitamin D absorption is normal in the absence of steatorrhea and only reduced by 40% of intake in the most severe states of malabsorption [43]. The etiology of vitamin D deficiency has therefore never been clear. However, it has been shown that some subjects with gastrectomy show a reduced half-life of 25OHD. These patients had evidence of secondary hyperparathyroidism with increased serum 1,25(OH)2D levels. Lowering of the hormone levels by large calcium supplements was accompanied by a prolongation of the half-life of 25OHD [44]. Other workers have also shown evidence of hyperparathyroidism in gastrectomized patients [45,46]. Nilas et al. [47] showed reduced serum 25OHD levels, increased 1,25(OH)2D levels, reduced calcium absorption and osteopenia in subjects several years after gastrectomy. Rümenapf et al. [48,49], using dynamic bone histomorphometry, found high turnover osteopenia using a rat model of total gastrectomy. Animals had high 1,25(OH)2D levels, but normal serum calcium and PTH. Using minipigs, Maier et al. [50] showed the animals to have low 25OHD, a reduced serum calcium, and secondary hyperparathyroidism. Other potential factors in producing abnormalities in calcium metabolism in this group are the reduced food intake resulting from loss of stomach area, reduced acid secretion, which is important for calcium absorption, poor admixture of food with digestive juices, and intestinal hurry. However, long term treatment with H2-receptor antagonists does not appear to affect the skeleton adversely [51]. c. Bone Turnover Vertebral deformity and fracture are more common in patients with a past history of gastrectomy [52–54], when compared with controls. Rao et al. [45] found a past history of gastrectomy in 5% of patients seen with vertebral fracture compared with 1% of controls. Histologically, the bone shows thin but extensive osteoid seams, a low appositional rate resulting from decreased collagen synthesis, and a low bone formation rate. Evidence of secondary hyperparathyroidism may be present and Parfitt et al. argue that these changes are the result of accelerated bone turnover with net loss of bone. Impaired recruitment and activity of osteoblasts then results in defective bone repair, predisposing to fracture [35,55]. 2. CLINICAL FEATURES OF POSTGASTRECTOMY BONE DISEASE

Osteoporosis is usually without symptoms until a fracture occurs. In postgastrectomy patients, there are

1300 often symptoms from disturbed bowel function, for example loose motions and steatorrhea, and these should be taken as indications for assessing the patient for bone disease. Multiple vertebral fractures lead to loss of height, thoracic kyphosis, and chronic back pain, which is difficult to control. Not surprisingly, accompanying depression is common, and may exacerbate the pain. Osteomalacia presents insidiously with vague bone pain and muscle weakness, particularly affecting the proximal muscles. As the bone disease progresses, the patient becomes increasingly incapacitated by bone pains and muscle weakness. Walking becomes labored, the patient developing a waddling gait and often having to climb stairs “crab-like,” holding the stair rail with both hands. In the most extreme cases, pain around joints from an enthesopathy, due to the effects of hyperparathyroidism, may simulate active arthritis. Fractures may occur, particularly in chronic cases, and especially where pseudofractures have been present; a frequent site is the femoral neck. Insufficiency fractures from osteoporosis can mimic Loosers zones radiologically [56], but the clinical picture is one of acute pain in relation to the fracture, rather than the history of chronic vague bone pain that characterizes osteomalacia. 3. BIOCHEMISTRY

Serum biochemistry in cases of osteoporosis is normal, although the alkaline phosphatase may be slightly increased, especially if there has been a recent fracture. Evidence for secondary hyperparathyroidism may be present together with raised 1,25(OH)2D. Bone densitometry will show a reduced bone mass that may be more marked at cortical sites in the forearm and hip than in the spine. In osteomalacia there is usually a reduction in serum phosphate with a low normal or frankly reduced serum calcium and elevated alkaline phosphatase. If measured, PTH will be found to be increased, 25OHD reduced, and 1,25(OH)2 low, normal, or increased, depending upon the amount of vitamin D recently provided from the diet or by solar exposure. 4. MANAGEMENT

The management aim for patients at risk should be to ensure normal plasma biochemistry by providing sufficient calcium and vitamin D as a dietary supplement, and thus suppress any increase in PTH and ensure normal 25OHD levels. When osteomalacia is present, oral vitamin D supplements can be given as a large single bolus using doses of several hundred thousand units, or much smaller doses of 1000–4000 IU (25–100 µg) daily. Patients with steatorrhea will require larger oral doses, or parenteral vitamin D, and should be monitored by measuring serum 25OHD regularly, until it can be shown that normal 25OHD values are being achieved.

MICHAEL DAVIES, JACQUELINE L. BERRY, AND ANDREW P. MEE

In the light of difficulties with calcium absorption, a large calcium supplement (1–2 g of elemental calcium) should also be given long term. Patients without overt bone disease should have blood assayed for PTH and 25OHD. If there is any evidence of secondary hyperparathyroidism, sufficient oral calcium should be given to suppress the elevated PTH. Vitamin D should also be given if the serum 25OHD is low. Bone densitometry should be performed and if evidence of osteoporosis is found, then appropriate antiresorptive therapy, such as bisphosphonate treatment, should be considered.

B. Celiac Disease 1. CLINICAL FEATURES

Celiac disease is an inflammatory condition of the small intestine triggered by ingesting gluten, present in wheat, rye, or barley. The development of sensitive and specific serological screening tests such as antiendomysial and tissue transglutaminase antibodies [57] has resulted in an increase in the estimated prevalence of celiac disease from an often quoted 1 in 2000–4000 of the population to a more recent study where 0.4% of healthy blood donors had positive serology results for celiac disease [58]. The prevalence varies in different ethnic groups, and is particularly common in those of Irish descent. While reduced bone mass is found in a significant proportion of patients compared with ageand sex-matched controls [59–63], symptomatic bone disease is uncommon in celiac disease. Using antiendomysial antibodies in asymptomatic osteoporotic subjects, Lindh et al. [64] showed that the incidence of celiac disease was tenfold higher than in the normal population. In adults on treatment for celiac disease, osteopenia is twice that expected for a normal population with a reduction in bone density of between 7 and 13% [65,66]. Until recently, no data existed to show an increase of fracture in celiac patients, but in a crosssectional case control study by Vazquez et al. [67], a quarter of 165 patients with celiac disease had a history of 1 to 5 peripheral fractures compared with 8% of an age- and sex-matched control group. The fractures occurred before the diagnosis of celiac disease, or in those subjects who did not comply with a gluten-free diet. The findings of Selby et al. [68], that cortical bone is more affected than trabecular bone almost certainly because of the effects of excess parathyroid hormone upon the skeleton, helps to explain the findings of Vazquez et al. [67]. More recently, Valdimarsson et al. have confirmed these findings [69]. In children, institution of treatment with a gluten-free diet results in a normal bone mass when they become teenagers [70,71]. In a small group of prepubertal

CHAPTER 75 Bone Disorders Associated with Gastrointestinal and Hepatobiliary Disease

children with celiac disease, the addition of calcium and vitamin D to gluten-free diet showed increments in BMD greater than in an age- and sex-matched control population, but the Z scores of the celiac group remained lower than the control group after two years of treatment [72]. These observations in children, adolescents, and adults indicate that there may be an irreversible element to the bone loss (or reduced accrual of bone), which may be a function of the duration of untreated celiac disease, a condition that is much more common than previously considered. An adult diagnosed with celiac disease may have had the problem since childhood and failed to achieve adequate peak bone mass, whereas the developing child can fully develop the skeleton if the gut is returned to normality. Occult celiac disease with no overt signs or symptoms of intestinal malabsorption may occasionally present as an osteomalacic syndrome [73,74], and small bowel biopsy should be considered in the assessment of all patients with osteomalacia in whom the cause is unclear. Alternatively, gastrointestinal symptoms may be so florid as to obscure the symptoms of an underlying osteomalacia. As celiac disease affects the duodenum and jejunum more severely than the ileum, malabsorption of calcium and vitamin D are common [8,75]. However, not all patients with celiac disease malabsorb vitamin D; in a small study, the celiac disease patient with the most severe malabsorption of vitamin D had the highest serum level of 25OHD, whereas absorption of vitamin D was normal in two patients who showed some degree of vitamin D deficiency [43]. These findings re-emphasize the importance of cutaneous synthesis of vitamin D. 2. DEVELOPMENT OF VITAMIN D DEFICIENCY

Evidence of poor vitamin D nutrition can be found in untreated celiac disease. Dibble et al. [76] found two untreated patients with low serum 25OHD levels (<12 nmol/l) but normal values in 12 subjects successfully treated with a gluten-free diet. Ferretti et al. [57] and Arnaud et al. [77] also found low 25OHD values in celiac patients. In the small series reported by Melvin et al. [75], 5 of 9 patients had histological osteomalacia (this study was performed before measurements of vitamin D metabolites were possible). The mechanisms leading to vitamin D deficiency have previously been attributed to interruption of a conservative enterohepatic circulation of 25OHD [23], but this concept has been largely discarded, see Section IIA2b. A patient with celiac disease, secondary hyperparathyroidism, and elevated serum 1,25(OH)2D was described in a study by Clements et al. [29].

1301

The half-life of radio-labeled 25OHD3 was shortened in the hyperparathyroid state compared with the euparathyroid state, which was achieved by using large supplements of calcium. In a recent study by Selby et al. [68] and Valdimarsson et al. [69], a high proportion of patients with treated celiac disease showed evidence of secondary hyperparathyroidism, suggesting that there may be a continuing problem with calcium absorption despite a gluten-free diet. Support for these last two observations is also forthcoming from the recent papers of Corazza et al. [78] and Ferretti et al. [63]. These workers found evidence of low bone mass in celiac patients, which was improved but not completely corrected by a gluten-free diet. Many untreated patients have secondary hyperparathyroidism with significantly raised serum 1,25(OH)2D levels and significantly lower 25OHD levels in the untreated state, compared with the treated condition. 3. MANAGEMENT

The treatment of choice is a gluten-free diet, but not all patients will be able to adhere to such a strict regime. Some individuals either overtly or covertly eat a diet containing gluten, which may explain the findings of Selby et al. above [68]. Additional vitamin D will cure any osteomalacia, and there is probably a good case for giving long-term calcium supplements of 1–2 g per day to guard against the effects of occult secondary hyperparathyroidism on the skeleton.

C. Pancreatic Disease Pancreatic insufficiency is not normally associated with metabolic bone disease, unless complicated by other conditions such as cystic fibrosis or alcoholism. This paradox implies that the disturbances to mineral and bone metabolism seen in other forms of bowel and liver disease have little to do with steatorrhea. Low 25OHD levels have been reported in cystic fibrosis [79,80], and both osteoporosis and osteomalacia have been described in this condition [81,82]. Meredith and Rosenberg [9] have questioned the rarity of bone disease associated with pancreatic steatorrhea, and they consider that osteomalacia may be more common than previously believed. Patients should receive pancreatic supplements and, if serum 25OHD is shown to be low, oral vitamin D at 800 IU/day. If, however, steatorrhoea remains a problem, larger oral doses will be necessary. Serum 25OHD levels should be measured every 1–2 months if daily doses in excess of 10,000 IU/day (0.25 mg) are used. Occasionally, parenteral vitamin D 100,000 IU per month may prove necessary.

1302 D. Inflammatory Bowel Disease 1. CLINICAL AND BIOCHEMICAL FEATURES

Of the two main inflammatory bowel diseases, bone disease is more often seen in Crohn’s disease than in ulcerative colitis. Since the colon is principally an organ for conservation of salt and water, disturbance of its function does not normally affect the skeleton. However, loss of bone and osteoporosis may result from the use of corticosteroids to control disease (see Chapter 73). Malabsorption of vitamin D and 25OHD have been documented in Crohn’s disease [9,43], and calcium is also malabsorbed but perhaps less often than previously considered. In one study [83], only 4 of 31 patients with active Crohn’s disease were found to have reduced net absorption of calcium and negative calcium balance and 2 of the 4 had undergone ileal resection. Driscoll et al. [84] studied 82 patients with Crohn’s disease from 9 of whom transiliac crest bone biopsies were taken. Six of these patients were found to have osteomalacia; 3 had a repeat biopsy, which showed improvement following treatment with vitamin D. There was a high incidence of vitamin D deficiency as assessed by serum 25OHD, 65% having a low 25OHD; in 25% of cases, serum 25OHD was below 25 nmol/l. It was in this lowest group where the cases of osteomalacia occurred. The lowest 25OHD values were found in those patients who had had a previous ileal resection. Nine of 25 patients with Crohn’s disease, assessed by Compston et al. [85] had increased osteoid on bone biopsy. More recently, vitamin D status was assessed in 112 patients ranging from 5–22 years of age [86]. The overall prevalence of hypovitaminosis in this group of patients was only 16%, however hypovitaminosis was much more common in winter (31%), in African Americans (56%), in patients with disease confined to the upper GI tract (44%), and in patients with a greater lifelong exposure to corticosteroids [86]. Osteoporosis is more common than osteomalacia. Compston et al. [87] found a reduced bone mass in 30% of an unselected group of patients with Crohn’s disease. The pathogenesis of osteomalacia may be similar to that in celiac disease, but no reports have shown good evidence of secondary hyperparathyroidism with high serum concentrations of 1,25(OH)2D as has been seen in celiac disease. The fact that calcium absorption is often normal [83] does not exclude excess parathyroid activity because normal calcium absorption may only be maintained at the expense of increased parathyroid drive. However, 1,25(OH)2D induced hepatic wastage of 25OHD seems less common in patients with Crohn’s disease. Glucocorticoids, which are used frequently to control inflammatory bowel disease, can both impair calcium

MICHAEL DAVIES, JACQUELINE L. BERRY, AND ANDREW P. MEE

absorption and reduce osteoblast function thus exacerbating those processes known to produce bone loss. Protein and calorie malnutrition, which are seen in severe cases of inflammatory bowel disease, will also adversely affect the skeleton [88]. There is increasing interest in factors of the inflammatory process as possible mediators of bone resorption and osteoblast function. While the nature of these factors is unknown, a candidate cytokine is interleukin 6 (IL-6). Pollak et al. [89] showed an increase in serum IL-6 in osteoporotic patients with inflammatory bowel disease, with normal BMD in patients with normal IL-6 levels. Finally, Hewison et al. [90] have produced data suggesting that ectopic production of 1,25(OH)2D by activated macrophages in Crohn’s granuloma may be detrimental to the skeleton. These authors showed a negative correlation in serum 1,25(OH)2D levels and lumbar spine BMD, which was independent of corticosteroid use. Serum 1,25(OH)2D values were elevated (> 60 pg/ml) in 42% of 138 patients with Crohn’s disease without evidence of secondary hyperparathyroidism. Despite the known beneficial effects of corticosteroids on suppressing extra-renal 1α-hydroxylase (Chapter 79), the Crohn’s patients receiving high doses of steroid had similar 1,25(OH)2D levels compared with Crohn’s patients receiving low doses of corticosteroid. 2. MANAGEMENT

Patients with inflammatory bowel disease should be managed in a similar way to patients with partial gastrectomy. Osteoporosis may pose a problem for management using oral bisphosphonates because of their poor absorption and tendency to cause bowel upset. Intravenous bisphosphonates or nasal calcitonin can be used as an alternative to oral antiresorptive therapy. There is also a possible role for 1,25(OH)2D or analog in the treatment of inflammatory bowel disease itself. It has been shown that isolated T lymphocytes from patients with ulcerative colitis are particularly sensitive to cyclosporin A, and that there is a synergistic effect on proliferation when the cyclosporin was given in combination with 1,25(OH)2D and several analogs [91], suggesting that this combination therapy may be more useful in patients than cyclosporin A alone. (Cyclosporin is further discussed in Chapter 74.)

E. Jejuno-ileal bypass 1. CLINICAL FEATURES

Intestinal bypass surgery for obesity became popular during the 1970s but was followed by a considerable mortality and morbidity, including a high incidence of

CHAPTER 75 Bone Disorders Associated with Gastrointestinal and Hepatobiliary Disease

skeletal complications. Metabolic bone disease developed after a variable number of years in a significant proportion of patients following bypass surgery for obesity, however, obesity per se is known to be sometimes associated with an increase in PTH [92]. Dano and Christiansen [93] found the bone mineral content to be reduced compared with controls in a group of 37 obese patients before undergoing bypass surgery. Surgery, resulting in severe intestinal malabsorption, may exacerbate pre-existing, and perhaps subtle, abnormalities of calcium metabolism. In about half the patients, there is a fall in serum calcium and magnesium after surgery, and although serum 25OHD levels fall, 1,25(OH)2D levels are maintained [94–96]. Osteomalacia is seen on bone biopsy in up to 60% of patients and in some, 1,25(OH)2D levels are low and PTH values are raised [35,37,94,97]. Profound hypocalcemia may occur secondary to the hypoparathyroidism of magnesium deficiency when osteomalacia is believed to be less severe [35]. Osteomalacia resulting from bypass surgery is seen more frequently in Europe than in America. The etiology of osteomalacia in bypass patients has been attributed to malabsorption of vitamin D, but the functional abnormalities produced by intestinal bypass surgery should favor hepatic wastage of 25OHD, although this hypothesis has not been tested. Osteopenia is also found following surgery with loss of bone in both the appendicular and axial skeleton [95]. 2. MANAGEMENT

Vitamin D and 1α-hydroxyvitamin D have been used with success to treat osteomalacia, but the native vitamin should be adequate to correct any deficiency. The parenteral route may be necessary in view of loss of functional bowel. Liver disease may follow this type of operation, but impaired hepatic hydroxylation of vitamin D is rarely a problem (see Section V). Treatment failures have responded to antibiotics used to eradicate bacterial overgrowth in the bypass segment [98].

F. Total Parenteral Nutrition (TPN) 1. CLINICAL FEATURES

TPN is used for those individuals in whom disease of the bowel is so severe that oral feeding cannot achieve adequate nutrition. Most patients commencing TPN do so after years of chronic bowel disease, and therefore are likely to have pre-existing bone disease. This section is primarily concerned with the effects of TPN per se on the bone. TPN may not provide all the nutrients or micronutrients required and historically calcium and phosphate were omitted from solutions, resulting in rickets

1303

in some children [99]. However, the major iatrogenic bone lesion resulted from the contamination of casein hydrolysate by aluminum at concentrations of up to 1 mg/l [100]. Serum samples from these patients contained large amounts, and bone contained more than 30 times the normal level of aluminum [99]. It is not surprising, therefore, that bone biopsies in most of these patients showed signs of osteomalacia. Clinically, the problem presented with peri-articular bone pains, especially in the legs, back, and ribs, and was only improved by cessation of TPN [101]. The disease has largely disappeared following the substitution of casein by amino acids [102]. Several biochemical abnormalities were found, but these may have been related to the amount of protein present in the feed. Hypercalciuria, low PTH, and low 1,25(OH)2D values returned to normal when purified amino acids were introduced. However, hypercalciuria persists if the amino acid content is kept too high [103]. The bone disease seen in these patients is mainly osteopenia [104,105], probably related at least in part to bone changes before TPN was commenced. 2. MANAGEMENT

Bone density, biochemistry, and bone histology should be evaluated at the onset of TPN and any hyperparathyroidism, vitamin D deficiency, or osteomalacia corrected by the appropriate use of calcium and vitamin D. If only osteoporosis is present, parenteral bisphosphonate treatment should be considered, and the response should be monitored by bone densitometry. Whenever possible, a return to enteral feeding should be encouraged. In cases of severe aluminum poisoning, chelation with desferoxamine, as used in dialysis patients, may be effective.

V. LIVER DISEASE A. Hepatic Osteodystrophy While both osteomalacia and osteopenia contribute to the syndrome of “hepatic osteodystrophy,” the condition is characterized by reduced bone mass leading to severe osteoporosis and fractures. Osteomalacia occurs more commonly in the presence of severe longterm cholestasis, as in primary biliary cirrhosis (PBC), but even in PBC, it is rare and its incidence appears largely restricted to Northern Europe [106–108]. Earlier studies in which a high incidence of osteomalacia in bone biopsies was reported (e.g. 22 out of 32 patients, mainly with cholestatic disease) used diagnostic methods, which have now been superseded [109,110]. Compston et al. [111], using more stringent criteria including the presence of calcification fronts,

1304 reported a much lower incidence; only 4 out of 32 patients had evidence of osteomalacia. A similarly low incidence was recently reported by Crosbie et al. [112]; only 2 of 27 patients with chronic liver disease showed biochemical evidence of osteomalacia. A rather higher incidence was reported by Dibble et al. [33], who found osteomalacia in 9 out of 29 patients; of these 9, 5 had PBC, 2 had chronic active hepatitis, and one each had sclerosing cholangitis and alcoholic cirrhosis. Indeed, using his strict criteria for diagnosing osteomalacia, Parfitt [31] has argued that only eight histologically proven cases of hepatic osteomalacia have been published [33,111,113–116]. Many other reported cases have used invalid methods involving decalcified bone [110,117] or not excluding HVOI (Section III). By such strict criteria, osteomalacia is most uncommon, but when present is associated with severe vitamin D deficiency and secondary hyperparathyroidism [32,105]. Conflicting reports have appeared on the levels of PTH in severe liver disease; an earlier study [118], claiming a defect in hepatic cleavage of PTH, was challenged by Klein et al. [119] who failed to find hyperparathyroidism using a range of different assays for the hormone. The latter authors concluded that raised PTH was unlikely to play a part in hepatic osteodystrophy. Dibble et al. [33] found evidence of slightly increased PTH in advanced cirrhosis, but osteomalacia was only found together with chronic cholestasis and vitamin D deficiency. The central role of the liver in maintaining a balance between the synthesis of 25OHD, eliminating excess vitamin and its metabolites, and thus regulating plasma levels of 25OHD is discussed elsewhere (Chapter 4). It had been supposed that osteomalacia in liver disease could be explained by impairment of hepatic 25-hydroxylation, and low plasma levels of 25OHD have been reported in many types of liver disease, including alcoholic hepatitis and cirrhosis, lupoid and cryptogenic cirrhosis, and primary biliary cirrhosis [117]. It is now recognized that where hepatocellular function is maintained, 25-hydroxylation of vitamin D remains normal until a very late stage of disease, and that malabsorption of vitamin D as a result of steatorrhea is a more likely cause of deficiency. Where steatorrhea is present, absorption of calcium may well be impaired, leading to osteoporosis. Malabsorption of both vitamin D and calcium may be compounded by the use of bile sequestration agents [120] (see Chapter 74). Osteoporosis is the most common bone problem seen in chronic liver disease [121] and it may be exacerbated by drugs used to control the disease, such as steroids in chronic active hepatitis. The most common histomorphometric parameter is not that of increased bone turnover, as is often associated with intestinal

MICHAEL DAVIES, JACQUELINE L. BERRY, AND ANDREW P. MEE

disease, but a low turnover state with low values for osteoid volume and surface extent. Appositional rate is also low when tetracycline labeling is used [108,121]. Although eroded surfaces without osteoclasts may be increased, this may reflect delayed formation rather than increased resorption [31]. This low turnover state reflects depressed osteoblastic function, which may be part of the chronic debilitating nature of liver disease, or may be due to the persistence of toxins normally removed by the liver. The histological features are, however, similar to those seen in patients after jejunoileal bypass.

B. Primary Biliary Cirrhosis 1. CLINICAL FEATURES

Vitamin D deficiency has been frequently reported in PBC, at times associated with osteomalacia, and characterized by low plasma levels of 25OHD [117,122]. The explanation is probably poor intestinal absorption [123] from a diet containing little vitamin D; in addition, there may be problems with endogenous synthesis of the vitamin. Sunshine may be deliberately avoided for cosmetic reasons or because of itching and discomfort, and it is possible that UV irradiation is less effective because of the pigmented skin. Claims that the osteomalacia of PBC was refractory to treatment with vitamin D were based upon early studies, the results of which may have been misinterpreted. Reports claiming that parenteral vitamin D therapy was ineffective in PBC may be explained by the use of intramuscular oily preparations of the vitamin that have been shown to have low bioavailability [124]. Similarly, the availability of vitamin D from oral preparations may depend on the vehicle. Absorption from tablets tends to be poor, especially in the presence of fat malabsorption, whereas absorption from an ethanol solution given in milk is relatively efficient [112]. The deficiency can be corrected by a modest increase in the oral supply of vitamin D or by giving UV irradiation [106,113]. Successful treatment has also been given using 25OHD, which being less lipophilic, is absorbed more efficiently than vitamin D [43,125]. Osteomalacia, as noted above (Section VA), is rare, and may be easily missed because of the severity of symptoms of the liver disease. Changes in alkaline phosphatase, phosphate, and calcium may reflect more the liver disease than changes in the skeleton. The clinician needs to be alert to the possibility of vitamin D deficiency and to identify patients at risk by measuring serum 25OHD. Bone pain from rib and vertebral fractures is more likely to result from osteoporosis, and the prevention and treatment of this disease is far more difficult than the prevention or treatment of osteomalacia.

CHAPTER 75 Bone Disorders Associated with Gastrointestinal and Hepatobiliary Disease

2. VITAMIN D METABOLISM

Given an adequate supply of vitamin D, the patient with PBC can metabolize the vitamin normally [126]. Reported low levels of 25OHD and 1,25(OH)2D in PBC [116,119,126] seem to be related to malabsorption of the vitamin, rather than to any intrinsic defect in metabolism. Subnormal concentrations of vitamin D metabolites are associated with established osteomalacia, and provide an adequate explanation for the development of the disease. There is little evidence for raised levels of either 1,25(OH)2D or PTH in PBC, and so the conditions favoring the 1,25(OH)2D–induced increased turnover of 25OHD are not present. Where there is considerable biliary obstruction, water-soluble vitamin D metabolites appear in unusually large amounts in urine [106, 123, 126], see Fig. 4. Jung et al. [106] found urinary excretion of metabolites

A Urinary radioactivity (% dose)

35 30 25 20 15 10 5 0 Control

PBC

B Urinary radioactivity (% dose)

35 30 25 20 15 10 5 0 0

50

100

150

200

250

300

350

Bilirubin (µmol/L)

FIGURE 4 Urinary excretion of vitamin D metabolites in patients with primary biliary cirrhosis (PBC) showing (A) percentage dose excreted compared to control subjects and (B) the strong relationship to the degree of biliary obstruction as indicated by plasma bilirubin, Kendall’s coefficient of concordance τ = 0.788, P = 0.001. Data taken from Mawer et al. [126].

1305

after administering vitamin D but not after 25OHD, and concluded that the water soluble metabolites had arisen from vitamin D but not via the 25-hydroxylation pathway. It is doubtful whether this urinary excretion contributes quantitatively to vitamin D deficiency in PBC, since these metabolites would otherwise be excreted via the fecal route and do not appear to constitute an increased net loss. 3. MANAGEMENT

Oral doses of vitamin D have been found to be effective in treating the osteomalacia in PBC patients. Biochemistry and calcium balance were normalized, and bone histology improved after treatment [113]. The precise dose of vitamin D was selected in this study by first measuring the absorption of a 3H-labeled test dose and calculating the administered dose to fit the patient’s needs. The greatest degree of malabsorption was 26% of the administered dose. Doses of 400–4000 IU in solution per day were used, but, in the absence of information on absorption, a dose of 2000 IU per day should be adequate for most patients. Treatment with UV irradiation is also effective [106,113]. Tablets of vitamin D are less well absorbed than solutions, and treatment with high dose tablets needs to be monitored. If, after a trial of 10,000 IU per day serum 25OHD does not rise towards 50 nmol/l, then regular monthly or three-monthly injections of 300,000 IU (7.5 mg) should be tried and the response assessed by measuring serum 25OHD. If a response greater than 125 nmol/l is produced, then the dose should be reduced. If the extent of steatorrhea or cholestasis increase, then the oral dose may need to be increased. At no time should a patient receive more than 10,000 IU per day by mouth without checking 25OHD levels, as eventually intoxication may result. There is no need to use 25OHD or 1α−hydroxylated derivatives, although these are effective in curing bone disease. Osteoporosis is a difficult problem to address, and any treatment is experimental. Currently, treatment with bisphosphonates or hormone replacement therapy may be considered for those with a low, or falling bone mineral density, but given that the state is one of low turnover, it is difficult to envisage much benefit accruing from antiresorptive treatment. It is possible that fluoride, or parathyroid hormone, which stimulate bone formation, might be beneficial, but this awaits a detailed clinical trial. Despite beneficial effects on the underlying liver disease, treatment with ursodeoxycholic acid conferred no benefits on the skeleton in a study of 88 patients with PBC [127]. Supplementation with calcium was superior to treatment with calcitonin in maintaining

bone mass, albeit transiently, in a study of 25 women with PBC-associated osteoporosis [128]. The anabolic effects of 1–34 parathyroid hormone recently licensed for postmenopausal osteoporosis would seem to the authors to be the most promising clinical trial for the low turnover osteoporosis of PBC.

C. Alcoholic Liver Disease

MICHAEL DAVIES, JACQUELINE L. BERRY, AND ANDREW P. MEE

1.6 25-Hydroxylation index

1306

1.2

0.8

0.4

1. CLINICAL FEATURES

Osteoporosis is the more frequent bony abnormality in patients with alcoholic liver disease, as in PBC, but osteomalacia has been reported occasionally [129–131]. The osteoporosis is characterized by a high incidence of fractures [129–131], but these do not follow a classic pattern. In a recent series, 27 of 76 chronic alcoholic male patients had vertebral fractures, but in only 5 of these was bone mineral density below the fracture threshold [132]. In evaluating fractures, it is important to distinguish between bone disease, which arises directly from the cirrhosis, ethanol associated osteopenia [133,134], and increased trauma (including seizures), resulting from the lifestyle of some alcoholic patients [131,132]. A study of the histomorphometry of alcoholinduced bone disease established major changes, even though chronic liver damage was not severe [135]. A significant decrease in bone volume was recorded together with increased resorption surfaces and increased osteoclast number. Low assayed levels of plasma 25OHD have been reported frequently in patients with alcoholic cirrhosis. The development of vitamin D deficiency in alcoholic liver disease probably has causes in common with the same problem in PBC, namely malabsorption of vitamin D (in patients with cholestasis) and lack of exposure to sunlight. Poor nutrition in alcoholic patients may also reduce the intake of vitamin D. 2. VITAMIN D METABOLISM

In contrast to PBC, abnormal hepatocellular function in alcoholic liver disease may be associated with disturbances of vitamin D metabolism. Plasma levels of 25OHD were shown to correlate with the degree of hepatic dysfunction, as measured by the antipyrine breath test [136], and defective synthesis of 25OHD was demonstrated by Jung et al. [137]. In a subsequent study, the impairment was shown to be correlated to plasma levels of bilirubin and prothrombin [118] (see Fig. 5). In this study, synthesis of 1,25(OH)2D was also shown to be defective in three of the patients studied, and it was attributed to the low level of precursor 25OHD and to poor renal function.

0.0 100

101 102 Bilirubin (µmol/L)

103

FIGURE 5

Impaired 25-hydroxylation of vitamin D in patients with alcoholic liver disease related to the plasma level of bilirubin, Kendall’s coefficient of concordance τ = −0.788, P < 0.02. Data taken from Mawer et al. [126].

Assayed concentrations of 1,25(OH)2D were low in a minority of alcoholic patients reported by Lalor et al. [130], but despite this evidence for defective vitamin D metabolism, osteomalacia is a comparatively rare component of alcoholic liver disease. 3. MANAGEMENT

Alcohol intake should cease, though this is a difficult goal to achieve. Vitamin D and calcium supplements should be given to those with evidence of low serum 25OHD and a poor (< 1g) intake of calcium. This simple maneuver will protect the skeleton from osteomalacia and reduce the tendency toward increased bone resorption. If alcohol continues to be consumed to the extent where amnesia is a problem, then compliance with the rigors of bisphosphonate therapy is unlikely. However, reformed alcoholics with osteoporosis seem as likely to respond favorably to antiresorptive therapy as any other group. Monitoring of bone mineral density at the start of therapy with follow-up after 12–18 months will give an indication as to whether treatment should continue long-term.

VI. SUMMARY It should not be surprising that metabolic bone disease arises when there is chronic disease affecting the gastrointestinal tract. The mineral phase of the skeleton is maintained by the absorption of calcium and other elements from the diet. It has sometimes been difficult to disentangle the effects of the disease itself on the skeleton from any effects, which might be

CHAPTER 75 Bone Disorders Associated with Gastrointestinal and Hepatobiliary Disease

exerted by the treatment given for that disease, but understanding of these factors is improving. Interpretation of the pathophysiological mechanisms involved has sometimes been clouded by misinterpretation of data, and there are still many problem areas. The mechanisms that result in low turnover osteoporosis are not fully understood, nor why states of secondary hyperparathyroidism should result in vitamin D wastage. It seems clear, however, that osteomalacia results primarily from vitamin D deficiency although calcium deficiency may also cause rickets and osteomalacia (see Chapters 64 and 65). The predisposition to osteomalacia may be present in some individuals before development of the gastrointestinal disease, but it can be reversed by appropriate, sometimes parenteral, treatment with the native vitamin. Likewise, HVOI may be managed by large calcium supplements, and perhaps vitamin D, to suppress secondary hyperparathyroidism. It seems that individuals with chronic debilitating bowel or liver disease are programmed to develop osteoporosis through poorly understood disturbances of osteoblast function, which may include effects from interleukins and other cytokines, produced by the inflammatory process itself. Research into this aspect of bone disease needs to be actively pursued, as the lives of many patients with advanced disease can now be prolonged by artificial feeding or by organ transplantation; these procedures themselves may introduce compounding factors into the genesis of metabolic bone disease.

References 1. Peacock M, Selby PL, Francis RM, Brown WB, Hordon L 1985 Vitamin D deficiency, insufficiency, and intoxication. What do they mean? In: Norman AW, Schaefer K, Grigoleit H-G, Herrath D (eds) Vitamin D, Chemical, Biochemical, and Clinical Update. de Gruyter: Berlin, 569–570. 2. Krall EA, Sahyoun N., Tannenbaum S, Dallal GE, DawsonHughes B 1989 Effect of vitamin D intake on seasonal variations in parathyroid hormone secretion in postmenopausal women. New Engl J Med 321:1777–1783. 3. Dawson-Hughes B, Harris SS, Krall EA, Dallal GE 1997 Effect of calcium and vitamin D supplementation on bone density in men and women 65 years of age or older. N Engl J Med 327:670–676. 4. Malabanan A, Veronikis IE Holick MF 1998 Redefining vitamin D insufficiency. Lancet 351:805–806. 5. Chapuy MC, Preziosi P, Maamer M, Arnaud S, Galan P, Hercberg S, Meunier PJ 1997 Prevalence of vitamin D insufficiency in an adult normal population. Osteoporosis Int 7:439–443. 6. Veith R, Ladak Y, Walfish PG 2003 Age-related changes in the 25-hydroxyvitamin D versus parathyroid hormone relationship suggests a different reason why older adults require more vitamin D. J Clin Endocrinol Metab 88:185–191.

1307

7. Glerup H, Mikkelsen K, Poulsen L, Hass E, Overbeck O, Thomsen J, Charles P, Eriksen EF 2000 Commonly recommended daily intake of vitamin D is not sufficient if sunlight exposure is limited. J Int Med 247:260–268. 8. Thompson GR, Lewis B, Booth CC 1966 Absorption of vitamin D3-3H in control subjects and patients with intestinal malabsorption. J Clin Invest 45:94–102. 9. Meredith SC, Rosenberg IH 1980 Gastrointestinal-hepatic disorders and osteomalacia. In: Avioli LV, Raisz LG (eds) Clinics in Endocrinology and Metabolism 9. WB Saunders and Co Ltd: London, pp. 131–150. 10. Phillips JR, Angulo P, Petterson T, Lindor KD 2001 Fatsoluble vitamin levels in patients with primary biliary cirrhosis. Am J Gastroenterol 96:2745–2750. 11. Poskitt EME, Cole TJ, Lawson DEM 1979 Diet, sunlight, and 25(OH)D in healthy children and adults. Br Med J i:221–223. 12. Stamp TCB, Haddad JG, Twigg CA 1977 Comparison of oral 25-hydroxycholecalciferol, vitamin D, and ultraviolet light as determinants of circulating 25-hydroxyvitamin D. Lancet i:1341–1343. 13. Davies M, Mawer EB, Hann JT, Taylor JL 1986 Seasonal changes in the biochemical indices of vitamin D-deficiency in the elderly. A comparison of people in residential homes, long-stay wards, and attending a day hospital. Age and Aging 15:77–83. 14. Mawer EB 1980 Clinical implications of measurements of circulating vitamin D metabolites. In: Avioli LV, Raisz LG, (eds) Clinics in Endocrinology and Metabolism. WB Saunders & Co Ltd: London 9:131–150. 15. Berry JL, Davies M, Mee AP 2002 Vitamin D metabolism, rickets, and osteomalacia. Sem Musculoskel Radiol 6:173–181. 16. Avioli LV, Lee SW, McDonald JE, Lund J, De Luca HF 1967 Metabolism of vitamin D3-3H in human subjects distribution in blood, bile, feces, and urine. J Clin Invest 46: 983–992. 17. De Luca HF, Schnoes HK 1983 Vitamin D: recent advances. Ann Rev Biochem 52:411–439. 18. Mawer EB, Backhouse J, Holman CA, Lumb GA, Stanbury SW 1972 The distribution and storage of vitamin D and its metabolites in human tissues. Clin Sci 43:413–431. 19. Clements MR, Chalmers TM, Fraser DR 1984 Enterohepatic circulation of vitamin D: a reappraisal of the hypothesis. Lancet i:1376–1379. 20. Makin G, Lotues D, Buford V, Ray R, Jones G 1989 Target cell metabolism of 1,25-dihydroxyvitamin D3 to calcitroic acid. Biochem J 262:173–180. 21. Jones G, Kung M, Kano K 1983 The isolation and identification of the new metabolites of 25-hydroxyvitamin D3 produced in the kidney. J Biol Chem 258:12920–12928. 22. Bolt MJG, Jensen WE, Sitrin MD 1992 Metabolism of 25-hydroxyvitamin D3 in rats: low calcium diet vs. calcitriol infusion. Am J Physiol (Endocrinol Metab) 262(25): E359–E367. 23. Arnaud SB, Goldsmith RS, Lambert PW, Go VLW 1975 25-hydroxyvitamin D3: Evidence of an enterohepatic circulation in man. Proc Soc Exp Biol 149:570–572. 24. Fraser DR 1983 The physiological economy of vitamin D. Lancet i:969–972. 25. Gascon-Barre M 1986 Is there any physiological significance to the enterohepatic circulation of vitamin D sterols. J Am Coll Nutr 5:317–324.

1308 26. Mawer EB 1979 The role of the liver in the control of vitamin D metabolism. In: Norman AW, (ed) Vitamin D, Basic Research and its Clinical Application. De Gruyter: Berlin, 533–561. 27. Clements MR, Johnson L, Fraser DR 1987 A new mechanism for induced vitamin D deficiency in calcium deprivation. Nature 325:62–65. 28. Clements MR, Davies M, Fraser DR, Lumb GA, Mawer EB, Adams PH 1987 Metabolic inactivation of vitamin D is enhanced in primary hyperparathyroidism. Clin Sci 73:659–664. 29. Clements MR, Davies M, Hayes ME, Hickey CD, Lumb GA, Mawer EB, Adams PH 1992 The role of 1,25-dihydroxyvitamin D in the mechanism of acquired vitamin D deficiency. Clin Endocrinol 3 7:17–27. 30. Colston KW, Mackay AG, Finlayson C, Wu JCY, Maxwell JD 1994 Localization of vitamin D receptor in normal human duodenum and in patients with celiac disease. Gut 35:1219–1225. 31. Parfitt AM 1990 Osteomalacia and related disorders. In: Avioli LV, Krane SM (eds) Metabolic Bone Disease and Clinically Related Disorders. WB Saunders and Co: Philadelphia, London, Toronto, Montreal, Sydney, Tokyo, 329–396. 32. Pittet PG, Davies M, Lawson DEM 1979 Role of nutrition in the development of osteomalacia in the elderly. Nutr Metab 23:109–116. 33. Dibble JB, Sheridan P, Hampshire R, Hardy GJ, Losowsky MS 1982 Osteomalacia, vitamin D deficiency, and cholestasis in chronic liver disease. Quart J Med 51:89–103. 34. Parfitt AM 1983 Dietary risk factors for age-related bone loss and fractures. Lancet ii:1181–1184. 35. Parfitt AM, Podenphant J, Villanueva AR, Frame B 1985 Metabolic bone disease with and without osteomalacia after intestinal bypass surgery. A bone histomorphometric study. Bone 6:211–220. 36. Rao DS, Villanueva AR, Mathews M, Pumo B, Kleerekoper M, Parfitt AM 1983 Histologic evolution of vitamin D–depletion in patients with intestinal malabsorption or dietary deficiency. In: Frame B, Potts Jnr JT (eds) Clinical Disorders of Bone and Mineral Metabolism. Excerpta Medica: Amsterdam, 224–226. 37. Goldner WS, O’Dorisio TM, Dillon JS, Mason EE 2002 Severe metabolic bone disease as a long-term complication of obesity surgery. Obesity Surgery 12:685–692. 38. Paterson CR, Woods CG, Pulvertaft CN, Fourman P 1965 Search for osteomalacia in 1228 patients after gastrectomy and other operations on the stomach. Lancet ii:1085–1088. 39. Tovey FI, Karamanolis DG, Godfrey J, Clark CG 1985 Postgastrectomy nutrition: methods of outpatient screening for early osteomalacia. Hum Nutr Clin Nutr 39C:439–446. 40. Eddy RL 1971 Metabolic bone disease after gastrectomy. Am J Med 50:442–449. 41. Deller DJ, Begley MD, Edwards RG, Addison M 1964 Metabolic effects of partial gastrectomy with special reference to calcium and folic acid. Gut 5:218–225. 42. Morgan DB, Hunt G, Paterson CR 1970 The osteomalacia syndrome after stomach operations. Quart J Med 39:395–410. 43. Davies M, Mawer EB, Krawitt EL 1980 Comparative absorption of vitamin D3 and 25 hydroxyvitamin D3 in intestinal disease. Gut 21:287–292. 44. Davies M, Heys SE, Selby PS, Berry JL, Mawer EB 1997 Increased catabolism of 25-hydroxyvitamin D in patients with partial gastrectomy and elevated 1,25-dihydroxyvitamin D

MICHAEL DAVIES, JACQUELINE L. BERRY, AND ANDREW P. MEE

45.

46.

47. 48.

49.

50. 51. 52. 53. 54.

55.

56. 57.

58. 59. 60. 61.

levels. Implications for metabolic bone disease. J Clin Endocrinol Metab 82:209–212. Rao RS, Kleerekoper M, Rogers M, Frame B, Parfitt AM 1984 Is gastrectomy a risk factor for osteoporosis? In: Christiansen C, Arnaud CD, Nordin BEC, Parfitt AM, Peck WA, Riggs BL (eds) Osteoporosis: Proceedings of the Copenhagen International Symposium on Osteoporosis. Aalborg Stiftsborgtrykkeri: Copenhagen, pp. 775–777. Bisballe S, Eriksen FF, Melsen F, Mosekilde L, Sorensen O, Hessov I 1991 Osteopenia and osteomalacia after gastrectomy: interrelations between biochemical markers of bone remodeling, vitamin D metabolites, and bone histomorphometry. Gut 32:1303–1307. Nilas L, Christiansen C, Christiansen J 1985 Regulation of vitamin D and calcium metabolism after gastrectomy. Gut 26:252–257. Rümenapf G, Schwille PO, Erben RG, Schreiber M, Fries W, Schmiedl A, Hohenberger W 1997 Osteopenia following total gastrectomy in the rat: state of mineral metabolism and bone histomorphometry. Eur J Surg Res 29:209–221. Rümenapf G, Schwille PO, Erben RG, Schreiber M, Berge B, Fries W, Schmiedl A, Koroma S, Hohenberger W 1998 Gastric fundectomy in the rat: effects on mineral and bone metabolism, with emphasis on the gastrin-calcitonin-parathyroid hormonevitamin D axis. Calcif Tissue Int 62:433–441. Maier GW, Kreis ME, Zittel TT, Becker HD 1997 Calcium regulation and bone mass loss after total gastrectomy in pigs. Ann Surg 225:181–192. Adachi Y, Shiota E, Matsumata T, Iso Y, Yoh R, Kitano S 1998 Bone mineral density in patients taking H2-receptor antagonist. Calcif Tissue Int 62:283–285. Deller DJ, Begley MD 1983 Calcium metabolism and the bones after partial gastrectomy. Clinical features and radiology of bones. Aust Ann Med 12:282–294. Nilsson BE, Westlin NE 1971 The fracture incidence after gastrectomy. Acta Chir Scand 137:533–534. Mellstrom D, Johansson C, Johnell O, Lindstedt G, Lundberg PA, Obrant K, Schoon I, Toss G, Ytterberg B 1993 Osteoporosis, metabolic aberrations, and increased risk of vertebral fractures after partial gastrectomy. Calcif Tissue Int 53:370–377. Parfitt AM, Mathews CHE, Villanueva AR, Rao DS, Rogers M, Kleerekoper M, Frame B 1983 Microstructural and cellular basis of age-related bone loss and osteoporosis. In: Frame B, Potts Jnr JT (eds) Clinical Disorders of Bone and Mineral Metabolism. Excerpta Medica: Amsterdam, 328–332. McKenna MJ, Kleerekoper M, Ellis BI, Rao DS, Parfitt AM, Frame B 1987 Atypical insufficiency fractures confused with Looser zones of osteomalacia. Bone 8:71–78. Troncone R, Maurano F, Rossi M, Micillo M, Greco L, Auricchio R, Salerno G, Salvatore F, Sacchetti L 1999 IgA antibodies to tissue transglutaminase: an effective diagnostic test for celiac disease. J Pediatr 134:166–171. Trier JS 1998 Diagnosis of celiac sprue. Gastroenterology 115:211–216. Bode S, Hassager C, Gudmand-Hoyer E, Christiansen C 1991 Body composition and calcium metabolism in adulttreated celiac disease. Gut 32:1342–1345. Caraceni MP, Molteni N, Bardella MT, Ortolani S, Gandolini GG, Bianchi P 1988 Bone and mineral metabolism in adult celiac disease. Am J Gastroenterol 83:274–277. Pistorius LR, Sweidan WH, Purdie DW, Stee SA, Howey S, Bennett JR, Sutton DR 1995 Celiac disease and bone mineral density in adult female patients. Gut 37:639–642.

CHAPTER 75 Bone Disorders Associated with Gastrointestinal and Hepatobiliary Disease

62. Walters JRF 1994 Bone mineral density in celiac disease. Gut 35:150–151. 63. Ferretti J, Mazure R, Tanoue P, Marino A, Cointry G, Vazquez H, Niveloni S, Pedreira S, Maurino E, Zanchetta J, Bai JC 2003 Analysis of the structure and strength of bones in celiac disease patients. Am J Gastroenterol 98: 382–390. 64. Lindh E, Ljunghall S, Larsson K, Lavo B 1992 Screening for antibodies against gliadin in patients with osteoporosis. J Intern Med 231:403–406. 65. McFarlane J, Bhalla A, Morgan L, Reeves D, Robertson DAF 1992 Osteoporosis: a frequent finding in treated adult celiac disease. Gut 33:S48. 66. Butcher GP, Banks LM, Walters JRF 1992 Reduced bone mineral density in celiac disease—the need for bone densitometry estimations. Gut 33:S54. 67. Vazquez H, Mazure R, Gonzalez D, Flores D, Pedreira S, Niveloni S, Smecuol E, Maurino E, Bai JC 2000 Risk of fractures in celiac disease patients: a cross-sectional, casecontrol study. Am J Gastroenterol 95:183–189. 68. Selby PL, Davies M, Adams JE, Mawer EB 1999 Bone loss in celiac disease is related to secondary hyperparathyroidism. J Bone Min Res 14:652–657. 69. Valdimarsson T, Toss G, Löfman O, Ström M 2000 Threeyears follow-up of bone density in adult celiac disease: significance of secondary hyperparathyroidism. Scand J Gastroenterol 35:274–280. 70. Molteni N, Caraceni MP, Bardella MT, Ortolani S, Gandolini GG, Bianchi P 1990 Bone mineral density in adult celiac patients and the effect of gluten-free diet from childhood. Am J Gastroenterol 85:51–53. 71. Bianchi ML, Bardella MT 2002 Bone and celiac disease. Calcif Tissue Int 71:465–471. 72. Muzzo S, Burrows R, Burgueno M, Rios G, Bergenfreid C, Chavez E, Leiva L 2000 Effect of calcium and vitamin D supplementation on bone mineral density of celiac children. Nutr Res 20:1241–1247. 73. Moss AJ, Waterhouse C, Terry R 1965 Gluten-sensitive enteropathy with osteomalacia but without steatorrhoea. N Engl J Med 272:825–830. 74. Hajjar ET, Vincenti F, Salti IS 1974 Gluten-induced enteropathy. Arch Intern Med 134:565–566. 75. Melvin KEW, Hepner GW, Bordier P, Neale G, Joplin GF 1970 Calcium metabolism and bone pathology in adult celiac disease. Quart J Med 39:83–113. 76. Dibble JB, Sheridan P, Losowsky MS 1984 A survey of vitamin D deficiency in gastrointestinal and liver disorders. Quart J Med 209:119–134. 77. Arnaud SB, Newcomer AD, Dickson ER, Arnaud CD, Go VLW 1976 Altered mineral metabolism in patients with gastrointestinal (GI) diseases. Gastroenterology 70:860. 78. Corazza GR, Di Sario A, Cecchetti L, Tarozzi C, Corrao G, Bernardi M, Gasbarrini G 1995 Bone mass and metabolism in patients with celiac disease. Gastroenterology 109: 122–128. 79. Hubbard VS, Farrell PM, de Sant’Agnese PA 1979 25-hydroxycholecalciferol levels in patients with cystic fibrosis. J Pediatr 94:84–86. 80. Donovan DS Jr, Papadopulos A, Staron RB, Addesso V, Schulman L, McGregor C, Cosman F, Lindsay RL, Shane E 1998 Bone mass and vitamin D deficiency in adults with advanced cystic fibrosis lung disease. Am J Respir Crit Care Med 158:1892–1899. 81. Hahn TJ, Squires AE, Halstead LR, Strominger DB 1979 Reduced serum 25-hydroxyvitamin D concentration and

82.

83. 84. 85. 86.

87. 88.

89.

90.

91.

92. 93. 94.

95.

96.

97.

98.

1309

disordered mineral metabolism in patients with cystic fibrosis. J Pediatr 94:38–42. Haworth CS, Selby PL, Webb AK, Dodd ME, Musson H, Niven RMcL, Economou G, Horrocks AW, Freemont AJ, Mawer EB, Adams JE 1999 Low bone mineral density in adults with cystic fibrosis. Thorax 54:961–967. Krawitt EL, Beeken WL, Janney CD 1976 Calcium absorption in Crohn’s disease. Gastroenterology 71:251–254. Driscoll RH, Meredith SC, Sitrin M, Rosenberg IH 1982 Vitamin D deficiency and bone disease in patients with Crohn’s. Gastroenterology 83:1252–1258. Compston JE, Ayers AB, Horton LW, Tighe JR, Creamer B 1978 Osteomalacia after small intestinal resection. Lancet i:9–12. Sentongo TA, Semaeo EJ, Stettler N, Piccoli DA, Stallings VA, Zemel BS 2002 Vitamin D status in children, adolescents, and young adults with Crohn’s disease. Am J Clin Nutr 76: 1077–1081. Compston JE, Judd D, Crawley EO, Evans WD, Evans C, Church HA, Reid EM, Rhodes J 1987 Osteoporosis in patients with inflammatory bowel disease. Gut 28:410–415. Geinoz G, Rapin CH, Rizzoli R, Kraemer R, Buchs B, Slosman D, Michel JP, Bonjour JP 1993 Relationship between bone mineral density and dietary intakes in the elderly. Osteoporosis Int 3:242–248. Pollak RD, Karmeli F, Eliakim R, Ackerman Z, Tabb K, Rachmilevitz D 1998 Femoral neck osteopenia in patients with inflammatory bowel disease. Am J Gastroenterol 93:1483–1490. Hewison M, Abreu MT, Kantorovitch V, Vasiliauskas EA, Gruntmanis U, Matuk R, Daigle K, Chen S, Zehnder D, Lin Y-C, Yang H, Adams JS 2003 Elevated 1,25-dihydroxyvitamin D in patients with Crohn’s disease: A novel risk factor for low bone mineral density. J Bone Min Res 18(Suppl 2):417. Stio M, Treves C, Celli A, Tarantion O, d’Albasio G, Bonanomi AG 2002 Synergistic inhibitory effect of ciclosporine A and vitamin D derivatives on T-lymphocyte proliferation in active ulcerative colitis. Am J Gastroenterol 97:679–689. Atkinson RL, Dahms WT, Bray GA, Schwartz AA 1978 Parathyroid hormone in obesity: effects of intestinal bypass. Miner Electrolyte Metab 1:315–320. Dano P, Christiansen C 1978 Calcium malabsorption and absence of bone decalcification following intestinal shunt operation for obesity. Scand J Gastroenterol 13:81–85. Teitelbaum SL, Halverson JD, Bates M, Wise L, Haddad JG 1977 Abnormalities of circulating 25-OH vitamin D after jejuno-ileal bypass for obesity and evidence of an adaptive response. Ann Int Med 856:288–293. Rickers H, Christiansen C, Balsev I, Rodbro P 1984 Impairment of vitamin D metabolism and bone mineral content after intestinal bypass for obesity. Scand J Gastroenterol 19:184–189. Hey H, Stokholm KH, Lund BJ, Lund BI, Sorensen OH 1982 Vitamin D deficiency in obese patients and changes in circulating vitamin D metabolites following jejunoileal bypass. Int J Obes 6:473–479. Mosekilde L, Melsen F, Hessov I, Christensen MS, Lund BJ, Lund BI, Sorensen OH 1980 Low serum levels of 1,25 dihydroxyvitamin D and histomorphometric evidence of osteomalacia after intestinal bypass for obesity. Gut 21:624–631. Compston JE, Horton LWL, Laker MF, Merrett AL, Woodhead JS, Gazet J-C, Pilkington TRE 1980 Treatment of bone disease after jejunoileal bypass for obesity with oral 1-hydroxyvitamin D3. Gut 21:669–674.

1310 99. Klein GL, Chesney RW 1986 Metabolic bone disease associated with total parenteral nutrition. In: Lebenthal E (ed). Total parenteral nutrition: Indication, utilization, complications and pathophysiological considerations. Raven Press: New York, 431–442. 100. Klein GL, Alfrey AC, Miller ML, Sherrard DJ, Hazlet TK, Ament ME, Coburn JW 1982 Aluminum loading during total parenteral nutrition. Am J Clin Nutr 35:1425–1429. 101. Klein GL, Targoff CM, Ament ME, Sherrard DJ, Bluestone R, Young JH, Norman AW, Coburn JW 1980 Bone disease associated with total parenteral nutrition. Lancet ii: 1041–1044. 102. Vargas JH, Klein GL, Ament ME, Ott SM, Sherrard DJ, Horst RL, Berquist WE, Alfrey AC, Slatopolsky E, Coburn JW 1988 Metabolic bone disease of total parenteral nutrition: course after changing from casein to amino acids in parenteral solutions with reduced aluminum content. Am J Clin Nutr 48:1070–1078. 103. Bengoa JM, Sitrin MD, Wood RJ, Rosenberg IH 1983 Amino acid-induced hypercalciuria in patients on total parenteral nutrition. Am J Clin Nutr 38:264–269. 104. Lipkin EW, Ott SM, Klein GL 1987 Heterogeneity of bone histology in parenteral nutrition patients. Am J Clin Nutr 46:673–680. 105. Shike M, Shils ME, Heller A, Alcock N, Vigorita V, Brockman R, Holick MF, Lane J, Flombaum C 1986 Bone disease in prolonged parenteral nutrition: osteopenia without mineralization defect. Am J Clin Nutr 44:89–98. 106. Jung RT, Davie M, Siklos P, Chalmers TM, Lawson DEM 1979 Vitamin D metabolism in acute and chronic cholestasis. Gut 20:840–847. 107. Recker RR, Maddrey W, Herlong F, Sorrell M, Russell R 1983 Primary biliary cirrhosis and alcoholic cirrhosis as examples of chronic liver disease associated with bone disease. In: Frame B, Potts JT Jr (eds) Clinical Disorders of Bone and Mineral Metabolism, Excerpta Medica: International Congress Series, 617:227–230. 108. Stellon AJ, Webb A, Compston J, Williams R 1987 Low bone turnover state in primary biliary cirrhosis. Hepatology 7:137–142. 109. Atkinson M, Nordin BEC, Sherlock S 1956 Malabsorption and bone disease in prolonged obstructive jaundice. Quart J Med 25:299–312. 110. Long RG, Varghese Z, Meinhard EA, Skinner RK, Wills MR, Sherlock S 1978 Parenteral 1,25 dihydroxycholecalciferol in hepatic osteomalacia. Br Med J i:75–77. 111. Compston JE, Crowe JP, Wells IP, Horton LWL, Hirst D, Merrett AL, Woodhead JS, Williams R 1980 Vitamin D prophylaxis and osteomalacia in chronic cholestatic liver disease. Dig Dis Sci 25:28–32. 112. Crosbie OM, Freaney R, McKenna MJ, Hegarty JE 1999 Bone density, vitamin D status, and disordered bone remodeling in end-stage chronic liver disease. Calcif Tissue Int 64:295–300. 113. Davies M, Mawer EB, Klass HJ, Lumb GA, Berry JL, Warnes TW 1983 Vitamin D deficiency osteomalacia and primary biliary cirrhosis, the response to orally administered vitamin D3. Dig Dis Sci 28:145–153. 114. Compston JE, Horton LWL, Thompson RPH 1979 Treatment of osteomalacia associated with PBC with parenteral vitamin D2 or oral 25 hydroxyvitamin D3. Gut 20:133–136. 115. Compston JE, Crowe JP, Horton LWL 1979 Treatment of osteomalacia associated with primary biliary cirrhosis with oral 1 alpha hydroxyvitamin D3. Br Med J 2:309.

MICHAEL DAVIES, JACQUELINE L. BERRY, AND ANDREW P. MEE

116. Danielson A, Lorentzon R, Larsson S-E 1982 Normal hepatic vitamin D metabolism in icteric primary biliary cirrhosis associated with pronounced vitamin D deficiency symptoms. Hepatogastroenterology 29:6–8. 117. Long RG, Skinner RK, Wills MR, Sherlock S 1976 Serum 25-Hydroxy-Vitamin-D in untreated parenchymal and cholestatic liver disease. Lancet ii:650–652. 118. Atkinson MJ, Vido I, Keck E, Hesch RD 1983 Hepatic osteodystrophy in primary biliary cirrhosis: a possible defect in Kupffer cell mediated cleavage of parathyroid hormone. Clin Endocrinol 118:21–28. 119. Klein GL, Endres DB, Colonna II JD, Berquist WE, Goldstein LI, Busuttil RW, Deftos LJ 1989 Absence of hyperparathyroidism in severe liver disease. Calcif Tissue Int 44:330–334. 120. Thompson WG, Thompson GR 1969 Effect of cholestyramine on the absorption of vitamin D3 and calcium. Gut 10:717–722. 121. Hodgson SF, Dickson ER, Eastell R, Eriksen EF, Bryant SC, Riggs BL 1993 Rates of cancellous bone remodeling and turnover in osteopenia associated with primary biliary cirrhosis. Bone 14:819–827. 122. Wagonfield JB, Nemchausky BA, Bolt M, Horst JV, Boyer JH, Rosenberg IH 1976 Comparison of vitamin D and 25-hydroxyvitamin D in the therapy of primary biliary cirrhosis. Lancet ii:391–394. 123. Krawitt EL, Grundman MJ, Mawer EB 1977 Absorption hydroxylation and excretion of vitamin D3 in primary biliary cirrhosis. Lancet ii:1246–1249. 124. Davies M, Mawer EB 1979 The absorption and metabolism of vitamin D3 from subcutaneous and intramuscular injection sites. In: Norman AW (ed) Vitamin D: Basic Research and Its Clinical Application. De Gruyter: Berlin, pp. 609–612. 125. Reed JS, Meredith SC, Nemohausky BA, Rosenberg IH, Boyer JL 1980 Bone disease in primary biliary cirrhosis: reversal of osteomalacia by 25-hydroxyvitamin D. Gastroenterology 78:512–517. 126. Mawer EB, Klass HJ, Warnes TW, Berry JL 1985 Metabolism of vitamin D in patients with primary biliary cirrhosis and alcoholic liver disease. Clin Sci 69: 561–570. 127. Lindor KD, James CH, Crippin JS, Jorgensen RA, Dickson ER 1995 Bone disease in primary biliary cirrhosis: does urodeoxycholic acid make a difference? Hepatology 21:389–392. 128. Camisasca M, Crosignani A, Battezzati PM, Albisetti W, Grandinetti G, Pietrogrande L, Biffi A, Zuin M, Podda M 1994 Parenteral calcitonin for metabolic bone disease associated with primary biliary cirrhosis. Hepatology 20: 633–637. 129. Nilsson BE 1970 Conditions contributing to fracture of the femoral neck. Acta Chir Scand 136:383–384. 130. Lalor BC, France MW, Powell D, Adams PH, Counihan TB 1986 Bone and mineral metabolism and chronic alcohol abuse. Quart J Med 229:497–511. 131. Israel Y, Orrego H, Holt S, Macdonald DW, Meema HE 1980 Identification of alcohol abuse: thoracic fractures on routine X-rays as indicators of alcoholism. Alcohol Clin Exp Res 420–422. 132. Peris P, Guanabens N, Pares A, Pons F, de Rio L, Monegal A, Suris X, Caballeria J, Rodes J, Munoz-Gomez J 1995 Vertebral fractures and osteopenia in chronic alcoholic patients. Calcif Tiss Int 57:111–114.

CHAPTER 75 Bone Disorders Associated with Gastrointestinal and Hepatobiliary Disease

133. Diamond T, Stiel D, Lunzer M, Wilkinson M, Posen S 1989 Ethanol reduces bone formation and may cause osteoporosis. Am J Med 86:282–288. 134. Crilly RG, Anderson C, Hogan D, Delaquerriere-Richardson L 1988 Bone histomorphometry, bone mass, and related parameters in alcoholic males. Calcif Tissue Int 43:269–276. 135. Diez A, Puig J, Serrano S, Marinoso M-L, Bosch J, Marrugat J, Mellibovsky L, Nogues X, Knobel H, Aubia J 1994

1311

Alcohol-induced bone disease in the absence of severe chronic liver damage. J Bone Min Res 9:825. 136. Hepner GW, Roginsky M, Moo HF 1986 Abnormal vitamin D metabolism in patients with cirrhosis. Dig Dis 21: 527–532. 137. Jung RT, Davie M, Hunter JO, Chalmers TM, Lawson DEM 1978 Abnormal vitamin D metabolism in cirrhosis. Gut 19:290–293.

CHAPTER 76

Vitamin D and Renal Failure ADRIANA S. DUSSO, ALEX J. BROWN, AND EDUARDO A. SLATOPOLSKY Renal Division, Washington University School of Medicine, St. Louis, Missouri

I. II. III. IV.

Introduction Alterations in Vitamin D Bioactivation to 1,25(OH)2D Alterations in 1,25(OH)2D/VDR Action Tissue Specific Effects of Low Calcitriol and Abnormal VDR Function

I. INTRODUCTION The kidney is a central component of the powerful endocrine system evolved to maintain extracellular calcium within narrow limits, a process vital for normal cellular physiology as well as skeletal integrity. Fig. 1A summarizes the main secretory control and hormonal interactions between the kidney, the parathyroid glands, intestine, and bone responsible for the maintenance of normal calcium homeostasis, and the alterations in the functions of this system induced by kidney disease Fig. 1B shows the changes found in kidney disease and will be discussed further below in Section IV. Kidney disease not only impairs the renal handling of calcium and phosphate ions, but also the endocrine capacity of the kidney to secrete the potent calcitropic hormone 1,25-dihydroxyvitamin D [1,25(OH)2D or calcitriol] [1]. The immediate consequence of calcitriol deficiency is decreased intestinal calcium absorption. The resulting hypocalcemia, along with the low circulating calcitriol levels, enhances the synthesis and secretion of parathyroid hormone (PTH) and eventually induces hyperplastic parathyroid-cell growth in an effort to re-establish calcium balance [2]. High circulating levels of PTH cause bone loss, osteitis fibrosa, and various degrees of skeletal abnormalities known as renal osteodystrophy. High serum levels of PTH also cause systemic toxicities, including cardiovascular, endocrine, nervous, immunologic, and cutaneous dysfunctions that markedly increase morbidity and mortality rates in renal failure patients [2]. Calcitriol not only has a critical role in the maintenance of calcium homeostasis and serum PTH levels, but also has potent effects in the differentiation and function in many cell types including skin, pancreas, and muscle, as well as hematopoietic, immune, and nervous systems [1]. In fact, epidemiological studies suggest an association between calcitriol deficiency VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Vitamin D Therapy in Chronic Renal Failure VI. Summary References

and hypertension, diabetes, as well as immunological and neuromuscular abnormalities, all of which are common disorders in patients with advanced kidney disease. This chapter presents the current understanding of the molecular mechanisms leading to calcitriol deficiency and resistance to calcitriol action in chronic renal failure. The pathophysiological implications of these defects is examined in classical vitamin D target organs (intestine, parathyroid glands, bone, and the kidney), as well as in tissues unrelated to calcium homeostasis, which, as mentioned, compromise the well being of patients with kidney disease. The last section, on vitamin D therapy in predialysis and patients undergoing hemodialysis, includes an overview of the treatment of renal osteodystrophy with special emphasis on recent clinical trials with calcitriol analogs, as well as the 2003 recommendations from the United States National Kidney Foundation.

II. ALTERATIONS IN VITAMIN D BIOACTIVATION TO 1,25(OH)2D A. Decreased Renal Mass The kidney proximal convoluted tubule is the principal site for the final and most critical step in vitamin D biological activation: 1α-hydroxylation of 25-hydroxyvitamin D (25OHD) to the potent calcitropic-steroid hormone 1,25-dihydroxyvitamin D (1,25(OH)2D or calcitriol) [1]. To maintain extracellular calcium within narrow limits, renal 1α-hydroxylase, a cytochromeP450 mixed-function oxidase, is tightly regulated. PTH, hypocalcemia, and hypophosphatemia are the major inducers, whereas hyperphosphatemia, hypercalcemia, and calcitriol, the enzyme product, repress its activity (see Chapters 5 and 7). Copyright © 2005, Elsevier, Inc. All rights reserved.

1314

ADRIANA S. DUSSO, ALEX J. BROWN, AND EDUARDO A. SLATOPOLSKY

CaSR VDR

CaSR

+ _ PTH

PTH +

ECF +

+

VDR _ ECF +

+

25D 1-Hydroxylase

25D 1-Hydroxylase

Ca Ca +

1,25D

1,25D

+

+ +

+

Normal

Kidney disease

FIGURE 1 Left panel. Calcium fluxes and hormonal interactions between the kidney, bone, intestine, and parathyroid glands responsible for the maintenance of extracellular calcium homeostasis. Right panel. Alterations to this powerful endocrine system induced by chronic kidney disease. (See text for details.)

calcitriol production in CKD. This section presents the mechanisms underlying the abnormalities in renal calcitriol synthesis induced by CKD.

50 40 Calcitriol, pg/ml

Martinez et al. (see Figure 2) [3], as well as other investigators, showed that, in the course of chronic kidney disease (CKD), the levels of calcitriol in blood remain in the normal range until the glomerular filtration rate (GFR) falls below 50% of normal [4–7]. Some investigators, however, have described patients whose plasma levels of calcitriol were below normal with creatinine clearances between 50 and 80 ml per minute [8]. Importantly, even normal levels of calcitriol should be considered abnormally low for the elevated circulating concentrations of PTH at these early stages of kidney disease. From these findings, it can be inferred that, in addition to the progressive reduction in functional renal mass, other abnormalities impair calcitriol production by the remnant 1α-hydroxylase. These abnormalities include impaired delivery of its substrate, 25OHD, to mitochondrialrenal 1α-hydroxylase, as well as a blunted induction of enzymatic activity in response to PTH. Direct inhibition of the activity of remnant renal 1α-hydroxylase by hyperphosphatemia, acidosis, and/or the accumulation of uremic toxins is an additional contributor to reduced

30 20 10 0

105

95

85

75

65 55 45 GFR, ml/min

35

25

15

FIGURE 2 Correlation between serum calcitriol and glomerular filtration rates (GFR) in 165 patients with different degrees of CKD. * indicates GFR levels at which the decreases in calcitriol reached statistical significance. Adapted from Martinez et al. [3].

CHAPTER 76 Vitamin D and Renal Failure

B. Reduced Substrate Availability to Renal 1α-hydroyxlase In patients with CKD with a GFR below 25 ml/min and therefore very limiting renal-1α-hydroxylase activity, serum calcitriol levels are low only in the presence of normal serum 25OHD concentrations [9]. However, serum levels of calcitriol were normalized in these severely uremic patients by increasing the serum concentrations of the substrate, 25OHD, to supraphysiological levels through oral supplementation [9]. Furthermore, a strong correlation was found between serum levels of substrate and product of renal 1αhydroxylase. This association, which does not occur in individuals with normal kidney function, is believed to result from an impairment in 25OHD availability to renal 1α-hydroxylase in severe renal failure. Studies by Nykjaer and collaborators in the megalinnull mice [10] provided new insights into the mechanisms responsible for impaired substrate availability and the strong correlation between serum levels of 25OHD and calcitriol in CKD (see Chapter 10). The megalinnull mice challenged the concept that renal 25OHD uptake by proximal tubular cells occurs through simple diffusion of the sterol through the basolateral membrane upon dissociation from its main carrier in circulation (the vitamin D binding protein, DBP). In fact, simple diffusion of 25OHD through the cell membrane cannot explain the impaired substrate availability to mitochondrial 1α-hydroxylase that is present in severe uremia. Instead, it was shown that the 25OHD/DBP complex in the circulation is filtered through the glomerulus and endocytosed into the proximal tubular cell via the apical-membrane receptor, megalin [10], a member of the LDL receptor superfamily. In CKD, the lower the GFR, the lower the amount of filtered 25OHD/DBP complex, which in turn, limits the amount of intracellular 25OHD available for conversion to calcitriol by renal 1α-hydroxylase. 25OHD supplementation increases the proportion of 25OHD/ DBP-complex in the blood with the consequent enhancement of the amount of 25OHD filtered and the intracellular levels available for bioactivation to calcitriol by the renal 1α-hydroxylase. However, as discussed below, 25OHD supplementation in severe renal failure could also enhance circulating calcitriol by increasing substrate availability to nonrenal 1α-hydroxylase. Importantly, in rats, renal megalin-mRNA levels progressively decrease by two weeks after the induction of renal failure [11]. In CKD, reduced renal-megalin levels would worsen the already abnormal uptake of 25OHD by the proximal convoluted tubules caused by the reduced glomerular filtration of the 25OHD/DBPcomplex. The demonstration that calcitriol up-regulates

1315 megalin expression in renal cells in culture [12] raises the possibility that the low circulating calcitriol in renal failure may constitute an additional contributor to reduced megalin expression in cells of the proximal convoluted tubules. Abnormal 25OHD delivery to extrarenal sources of calcitriol also occurs in kidney disease [13]. 1α-hydroxylase expression in numerous nonrenal cells has been well documented (see Chapter 79). The contribution of nonrenal 1α-hydroxylases to systemic calcitriol was conclusively demonstrated in bilaterally nephrectomized patients undergoing hemodialysis [14]. Importantly, in anephric humans, a strong correlation exists between serum levels of calcitriol and its precursor 25OHD, similar to that previously described in severely uremic patients (GFR below 25 ml/min). These findings led us to an evaluation of the impact of kidney disease on substrate availability to monocyte/macrophage 1αhydroxylase, the same enzyme as renal 1α-hydroxylase, though more readily accessible. Peripheral blood monocytes from hemodialysis patients elicit a markedly impaired uptake of 25OHD compared to that in monocytes from normal individuals [13]. This defective uptake could be corrected by normalizing the low serum calcitriol levels of hemodialysis patients through intravenous calcitriol supplementation for 15 days. Neither the mechanisms mediating impaired25OHD uptake nor those involved in its correction by calcitriol treatment are known. Calcitriol was shown to modulate LDL-receptor expression and function in the human monocytic cell line HL60 [15,16]. A similar mechanism could partially account for calcitriolinduced correction of the impaired uptake of 25OHD by peripheral blood monocytes from hemodialysis patients.

C. Blunted Induction of Renal-1α-hydroxylase by PTH- and Calcium PTH is a potent stimulator of calcitriol synthesis. Patients with hypoparathyroidism have low calcitriol levels despite persistent hypocalcemia [17]. Furthermore, parathyroidectomy severely blunts the induction of renal 1α-hydroxylase by hypocalcemia [18]. As mentioned earlier in this section, with the elevations in serum PTH occurring at early stages of renal disease, normal calcitriol levels in patients with GFR between 50 and 80 ml/min reflect abnormalities in PTH induction of renal calcitriol synthesis. In fact, Ritz and collaborators demonstrated an impaired increase in serum calcitriol in response to pharmacological doses of PTH in patients with a GFR of 70 ml/min and normal calcitriol levels [19], compared to individuals with normal

1316

ADRIANA S. DUSSO, ALEX J. BROWN, AND EDUARDO A. SLATOPOLSKY

kidney function. Mechanistically, PTH is known to activate 1α-hydroxylase-gene transcription through a cAMP-mediated mechanism [20–22]. Acidosis, a feature commonly present in renal failure, could contribute to impaired PTH induction of calcitriol production in these patients. In the dog, acidosis blunts the action of PTH on 1α-hydroyxlase of the proximal convoluted tubule. This effect can be overcome by cAMP, and it is unrelated to a loss of renal PTH receptors. Taken together, these findings suggest that acidosis induces an altered coupling of PTH receptor with adenylate cyclase [23] that blunts PTH enhancement of calcitriol production. Phosphate retention, a common feature in renal failure and a potent inhibitor of renal 1α-hydroxylase [24], could also account for counteracting the stimulatory actions of PTH on calcitriol production. Since acidosis also blunts the response to PTH on phosphate reabsorption [25], it could affect PTH induction of renal 1α-hydroxylase indirectly through concurrent effects on phosphate homeostasis. Similar to the impaired induction of 1α-hydroxylase activity by PTH in CKD, a blunted increase in serum calcitriol in response to calcium restriction was reported by Prince and collaborators [26].

D. Direct Inhibition of 1α-hydroxylase Expression and Activity by Hyperphosphatemia, Acidosis, and Accumulation of Uremic Toxins In individuals with normal kidney function, phosphate restriction increases renal calcitriol production and serum calcitriol levels, despite a decrease in serum PTH [27]. See Chapter 26 for a discussion of phosphate. Interestingly, whereas hypophysectomy blocks the stimulating action of dietary phosphate restriction on 1α-hydroxylase activity, injection of growth hormone or IGF-I to hypophysectomized rats restores the induction of renal calcitriol production in response to a low dietary phosphate intake [28]. Recent studies in intact mice and rats demonstrated that phosphate restriction not only modulates the activity but also mRNA levels of renal 1α-hydroxylase [29,30]. Furthermore, they supported the involvement of a growth hormone mediated mechanism for low P induction of renal-1α-hydroxylase mRNA levels. Kidney disease also appears to blunt growth hormone/ low P-signaling pathway, mediating the induction of renal 1α-hydroxylase mRNA levels and activity. There are no increases in serum calcitriol in response to P restriction in patients with end-stage renal disease [31] or in severely uremic dogs [32]. The possibility also

exists that the induction by phosphorus restriction of the very low amounts of remnant 1α-hydroxylase is insufficient to result in measurable increases in serum calcitriol levels. Transepithelial inorganic phosphate transport by the renal tubule was suggested as the mechanism underlying low phosphate induction of renal-1α-hydroxylase. However, studies in the mice null for the phosphateregulated renal Na/P cotransporter 2 (NPT2a) revealed that intact renal-Na/P co-transport is not required for the regulation of 1α-hydroxylase mRNA levels and activity by dietary phosphate restriction [33]. In contrast to the stimulatory effects of phosphate restriction, oral phosphate supplementation reduces serum calcitriol concentrations in patients with idiopathic hypercalciuria or in children with moderate renal insufficiency [34]. As mentioned in the prior section, hyperphosphatemia could partially account for the blunted response to PTH induction of renal calcitriol production. Also, metabolic acidosis, through alterations in phosphate homeostasis or by inducing increases in ionized calcium, could affect renal 1α-hydroxylase activity. However, the actual contribution of acidosis to modulate renal 1α-hydroxylase is controversial. Acidosis has been shown to decrease, increase, or not to change serum calcitriol levels [35–39]. Accumulation of uremic toxins in CKD could account for reduced renal-calcitriol synthesis. In normal rats, the infusion of uremic-plasma ultrafiltrate markedly decreases calcitriol production rates. Low molecular weight compounds in uremic plasma ultrafiltrate were identified in vitro as the inhibitors of 1α-hydroxylase activity [40]. The same authors have also shown that in partially nephrectomized rats, high dietary protein intake, which increases the production of uremic toxins, suppresses calcitriol synthesis. Taken together these studies suggest that, in CKD, there is an active suppression of the reserve for synthetic calcitriol capacity in the remnant kidney. The actual contribution of impaired substrate availability and/or direct inhibition of transcriptional and/or posttranscriptional regulatory mechanisms on 1α-hydroxylase-expression and activity is unclear at present.

E. Abnormal Calcitriol Catabolism Serum calcitriol levels are tightly regulated through simultaneous control of synthetic and catabolic rates [1]. The ubiquitously distributed enzyme 24-hydroxylase catalyzes most steps in the major pathway responsible for calcitriol metabolic inactivation. 24-Hydroxylasegene expression is highly inducible by calcitriol (see Chapters 6 and 7). Calcitriol induction of its own

1317

CHAPTER 76 Vitamin D and Renal Failure

catabolism provides a fine-tuning for both calcitriol control of its own levels in circulation and the magnitude of the response to calcitriol in target organs [1]. In fact, in normal individuals, the metabolic clearance rate of calcitriol is accelerated when its production rate increases [41]. In contrast to calcitriol stimulation of its own degradation, PTH down-regulates 24-hydroxylase activity in normal individuals [42]. This reduction in calcitriol catabolism could partially account for PTH up-regulation of serum calcitriol levels and the subsequent increases in intestinal calcium absorption. From the low circulating levels of calcitriol and the high PTH in CKD, reduced catabolic rates of calcitriol were expected. However, decreased, unaltered, and increased rates of calcitriol catabolism were reported in CKD. Hsu and collaborators reported a decreased calcitriol metabolic clearance rate in patients with CKD [43] and in the rat model of experimental renal insufficiency [44]. This reduction was considered a compensatory mechanism to maintain normal serum calcitriol levels at early stages of kidney disease. In contrast to these findings, studies of calcitriol metabolic clearance rate in dogs with mild to severe CKD demonstrated no changes in the rate of calcitriol catabolism with the progression of kidney disease. The changes in serum levels of calcitriol reflected the decrease in production rates with the progression of kidney disease [45]. In relation to enhanced calcitriol catabolism in CKD, it is possible that a mechanism similar to that blunting PTH induction of 1α-hydroxylase impairs PTH reduction of 24-hydroxylase. In fact, enhanced rather than reduced intestinal 24-hydroxylase was found in renal failure [46]. This increase in intestinal calcitriol degradation could partially account for the blunted response to calcitriol treatment to increase intestinal calcium absorption. Metabolic acidosis, commonly present in CKD, was also shown to increase 24-hydroxylase activity [47]. Independent of the controversial data on changes in the rate of calcitriol catabolism induced by kidney disease, it is obvious from the low serum calcitriol levels that reduced production rate is the main contributor to calcitriol deficiency.

III. ALTERATIONS IN 1,25(OH)2D/VDR ACTION Most, if not all, of calcitriol biological actions are mediated by a high affinity receptor, the vitamin D receptor (VDR), which acts as a ligand-activated

transcription factor (see Chapters 11, 13, 16, and 17). In kidney disease, in addition to the decreases in calcitriol synthesis described in the previous section, there is compelling evidence of resistance to calcitriol actions. Decreased VDR levels in target tissues and abnormalities in calcitriol/VDR regulation of the expression of vitamin D responsive genes have been observed in both patients and experimental animals with kidney disease. This section presents the current understanding of the mechanisms underlying both defects, as well as the potential contribution of VDR polymorphisms in the dialysis population to VDR expression and activity.

A. Defective Homologous Up-regulation of VDR The best-known regulator of VDR expression in target tissues is calcitriol itself (see Chapters 11, 12, and 15). Calcitriol up-regulation of VDR expression involves dual mechanisms: It increases VDR mRNA levels and/or VDR protein stability [48,49]. The latter is the most common process in calcitriol up-regulation of VDR content. Calcitriol binding to the VDR prevents receptor degradation by the proteasome complex, thus enhancing the half-life of the ligand-bound VDR compared to that of the unliganded VDR [49]. In renal failure, because serum calcitriol levels are low, a decrease in VDR content is predictable. In fact, calcitriol-binding studies in the parathyroid glands of humans [50,51], rats [52], and dogs [53] support a reduction in VDR number in uremia without changes in the affinity of the receptor for calcitriol. Immunohistochemical studies in the parathyroid glands of uremic patients corroborated these results and also demonstrated that the lowest VDR content was found in areas of nodular growth, the most aggressive form of parathyroid hyperplasia in severe kidney disease [50]. The latter finding suggests that the mitogenic signals that trigger parathyroid hyperplasia could also down-regulate VDR expression. A contribution of calcitriol deficiency to reduced VDR content was supported by studies in parathyroid glands from normal and uremic rats (see Fig. 3), which demonstrated a strong correlation between serum levels of calcitriol and parathyroid VDR content [54]. This association also suggested the potential of calcitriol therapy to correct the reduced parathyroid VDR content in renal failure patients. In fact, Fig. 4 shows that in uremic rats, the reduced parathyroid VDR content could be normalized by administration of either calcitriol or its less-calcemic analog 22-oxa-calcitriol [54]. Controversial reports exist on calcitriol binding to and regulation of intestinal VDR. Increased and reduced VDR binding was found in uremic rats

1318

ADRIANA S. DUSSO, ALEX J. BROWN, AND EDUARDO A. SLATOPOLSKY

1,25(OH)2D3–Binding (fmol/mg prot)

800 700 600 500 400 300 200

r = 0.849; p < 0.001

100 0 0

20

10

30

40

50

60

Serum 1,25(OH)2D3 (ng/ml)

FIGURE 3

Correlation between parathyroid VDR content and serum calcitriol levels in uremic and control animals. Adapted from Denda et al. [54].

compared to normal animals. In normal rats, calcitriol has no effect on VDR mRNA levels but markedly increases VDR protein by 12 to 24 h [55]. In contrast, in uremic rats, there is a blunted response to calcitriol. Despite an increase in VDR mRNA levels by 5 h, there is only a mild increase in VDR protein after 12 to 24 h of calcitriol exposure [55]. High PTH levels or hypocalcemia in uremic rats could impair calcitriol up-regulation of intestinal VDR since induction of hyperparathyroidism by calcium restriction was shown to prevent calcitriol up-regulation of intestinal VDR [56]. In contrast to these reports, Patel and co-workers demonstrated similar increases in VDR binding after 18 h exposure to calcitriol that were preceded by transient increases in VDR mRNA [57]. In the kidney, vitamin D or calcitriol supplementation to rats with a normal calcium intake increases VDR

VDR binding (fmol/mg protein)

1000 800

*

*

*

*

600 400 200 0

0 ng 0 ng Vehicle Normal

6 ng 2 ng 1,25(OH)2D3

8 ng

50 ng OCT

Uremic

FIGURE 4 Calcitriol [1,25(OH)2D3] and 22-oxacalcitriol [OCT] supplementation corrected the reduced-VDR content in the parathyroid glands of uremic rats (0 ng) to the levels of rats with normal renal function (Normal). Adapted from Denda et al. [54].

content up to five fold [58–60]. While Brown and collaborators reported a mild induction of kidney VDR mRNA by calcitriol in rats fed a high calcium diet [61], other investigators found no induction of renal VDR mRNA by calcitriol [49,62]. Controversial reports also exist demonstrating either reduced or normal VDR levels in peripheral blood monocytes from hemodialysis patients compared to monocytes from normal individuals [13,63].

B. Potential Contribution of VDR Polymorphisms to Reduced-VDR Expression and Function The human VDR is encoded by a gene, which localizes in chromosome 12. The gene has 11 exons, and the first three are variably present in VDR mRNAs. There are normal genetic variants or polymorphisms of the human VDR [64,65]. This topic is the focus of Chapters 68 and 12. The most frequently studied of these polymorphisms are located at the 3′ untranslated region in the intron separating exon VIII and IX and were defined by the restriction enzymes BmsI, ApaI, and TaqI. None of these variants affects either the expression or the translated VDR protein. Linked to these polymorphisms is a microsatelite poly A repeat of variable length (long or short), which may affect mRNA stability. A most recent polymorphism to be studied, FokI, is not linked to the others and results in a three amino acid shorter VDR with higher biological activity. Controversial reports exist on the association between the frequency of certain alleles and changes in bone density [66], the propensity to hyperparathyroidism [67,68], and resistance to vitamin D therapy [69]. In patients with renal failure from Western populations, higher PTH levels were reported to associate

1319

CHAPTER 76 Vitamin D and Renal Failure

with the bb genotype [70]. However, a faster bone mineral loss is seen in the BB group [71]. Also patients in the BB group can remain longer in hemodialysis before they need parathyroidectomy [72]. In Japanese patients, high PTH associates with the bb genotype only in one group of patients [73], whereas no association with the bb, but with the aa genotype was also reported [74]. Bone density after renal transplantation was also lower in the bb group as a result of the impact of this genotype on the severity of secondary hyperparathyroidism [75]. Moreover, hemodialysis patients in the aa-group showed higher sensitivity to parathyroid responsiveness to changes in [Ca++] than those in the AA or Aa groups with repercussions on the onset and progression of secondary hyperparathyroidism [76]. The cause of these associations is still unclear because, as mentioned at the beginning of this section, the encoded VDR protein remains unchanged, and there are no changes in VDR mRNA stability to affect VDR expression. No association has been reported between these polymorphisms and accelerated VDR-targeting for proteosomal degradation. As mentioned, important limitations of all of these epidemiological studies are that correlations were sought between a single specific polymorphism or between the Bsm-Apa-Taq linkage group and the physiological parameter of interest [77]. More importantly, most studies lack analysis of the direct influence of allelic variation on VDR protein expression or activity. These limitations leave open the possibility that the observed correlations might be due to another nearby site or even to a different gene. In a recent study, the functional significance of two unlinked-human VDR-gene polymorphisms [at a FokI restriction site (F/f) in exon II and a singlet A repeat (L/S)] was examined simultaneously [77]. Higher activity of the F and L biallelic forms was found. Statistical significance emerged when genotypes at both sites were considered simultaneously [77]. This study suggests that simultaneous reassessment of F/f and L/S genotypes in the hemodialysis population could provide insights into the actual contribution of VDR polymorphisms to abnormal parathyroid function, important for prognosis and therapy. The same study raised the possibility that polymorphisms in the 5′ region of the VDR gene might affect the activity of the VDR promoters [78], leading to the expression of altered quantities of VDR protein. Recently, a polymorphism was described in a binding site for Cdx-2, a homeodomain protein involved in intestine-specific VDR expression [79]. In fact, in a large cohort of Japanese women, the A allele at the Cdx-2 locus, correlated with higher bone mineral density in the lumbar spine consistent with a slightly greater activity of a VDR promoter

construct [80]. This polymorphism also should be considered in epidemiological studies relating VDRmediated intestinal absorption of Ca and P, as they could impact both bone mineral density and parathyroid function in the hemodialysis patient population.

C. Additional Mechanisms for Calcitriol Resistance in CKD The calcitriol synthesized in the kidney by mitochondrial 1α-hydroxylase is transported in the blood by carrier proteins. Vitamin D–binding protein (DBP) is the main carrier. However, calcitriol also binds albumin and lipoproteins [1]. Recently, reports showing that the free form of calcitriol triggers biological responses after entering target cells by simple diffusion have been challenged. The demonstration that 25OHD uptake by renal proximal tubular cells occurs through megalinmediated endocytosis [10], raised the possibility that calcitriol entrance to target cells could also involve a megalin- or LDL receptor-mediated endocytosis. Megalin is expressed in several epithelial cell types [81], including parathyroid cells [82] that respond to calcitriol. Interestingly, the LDL receptor was shown to mediate calcitriol uptake by human fibroblasts [83]. Therefore, uremia-induced reduction in megalin expression could constitute an additional mechanism for calcitriol resistance independent of the abnormalities described in VDR content or function. Once inside the target cell, calcitriol binding to the VDR activates the receptor to translocate from the cytosol to the nucleus where it heterodimerizes with its partner, the retinoid X receptor, RXR. The VDR/RXR complex binds specific sequences in the promoter region of target genes, called vitamin D response elements (VDRE), and recruits basal transcription factors and co-regulator molecules to either increase or suppress the rate of gene transcription by RNA-polymerase II [1]. Numerous genes, transcriptionally induced or suppressed by the calcitriol/VDR-complex, are relevant for the efficacy of calcitriol therapy in renal failure. The main biological action resulting from calcitriol regulation of the expression of these genes is a tight control of calcium homeostasis by the parathyroid glands, bone, intestine, and the kidney. Specifically, these genes affect calcitriol/VDR regulation of the rates of calcitriol synthesis and catabolism, suppression of PTH synthesis, and induction of the expression of the calcium-sensing receptor (CaSR) [1,84]. Calcitriol also controls the expression of genes unrelated to calcium homeostasis, such as those involved in the inhibition of the renin/angiotensin system [85], modulation of immune responses, and suppression of

1320

ADRIANA S. DUSSO, ALEX J. BROWN, AND EDUARDO A. SLATOPOLSKY

transcription. Figure 6 lists uremia-induced mechanisms underlying the reduced RXR/VDR heterodimerization, as well as the impaired binding of the VDR/RXR heterodimer to DNA with the consequent decrease in calcitriol/VDR regulation of gene expression. Studies in unilaterally nephrectomized rats demonstrated a reduction in the content of a 50 kDa RXR isoform in cell extracts from the remnant kidney. This decrease in RXR results in a reduction of the binding of the endogenous VDR/RXR heterodimer to the VDRE of the mouse osteopontin promoter [86]. A similar reduction of RXR content in the parathyroid glands of these rats could explain their enhanced serum PTH levels in the absence of hypocalcemia or hypophosphatemia [86], since RXR mediates both VDR and retinoic acid suppression of PTH mRNA levels and protein expression [87]. The contribution of the accumulation of uremic toxins to calcitriol resistance has been well documented [88]. Ultrafiltrate from uremic plasma causes a dose-dependent inhibition of VDR/RXR binding to VDRE and calcitriol/VDR-transactivating function [89]. Increases in nuclear calreticulin also modulate calcitriol/VDR action. Calreticulin is a cytosolic protein that binds integrins in the plasma membrane

cell proliferation [1]. Several mechanisms have been identified as responsible for the reduced efficacy of calcitriol to control the expression of these genes in renal failure. The remainder of this section addresses the pathophysiological implications of the abnormal calcitriol/VDR transcriptional activity in renal failure patients. The magnitude of calcitriol/VDR induction or suppression of gene transcription is determined mainly by the intracellular levels of both calcitriol and the VDR, and both are reduced in renal failure. Thus, reduced VDR levels, which lead to decreased formation of the calcitriol/VDR complex, are major contributors to the resistance to calcitriol therapy in advanced kidney disease. However, abnormalities in steps downstream from ligand binding to the VDR were also demonstrated in studies comparing calcitriol action in peripheral blood monocytes from normal individuals and hemodialysis patients. Figure 5 shows that in the presence of a similar VDR content, the binding of endogenous VDR/RXR complex to DNA is markedy impaired in peripheral blood monocytes from hemodialysis patients compared to that in monocytes from normal individuals, thus leading to an 80% inhibition of the ability of exogenous calcitriol to induce 24-hydroxylase gene

VDR 1,25D

Kidney disease Disease

Ligand binding -complexformation formation Reduced 1,25D/VDR-complex 1,25D VDR Reduced VDR/RXR heterodimerization RXR Impaired VDR/RXR binding to DNA Uremic toxins Uremic toxins Calreticulin Calreticulin Squelching Squelching (VDR-unrelated VDR RXR pathways)

VDR Nucleus

Heterodimerization RXR

VDR DNA binding RNA Pol II

VDRE Transactivation/transrepression Transactivation / transrepression RNA Pol II mRNA

B

CoReg VDR

RXR

VDRE

FIGURE 5

Calcitriol (1,25D)/VDR-transcriptional control of the expression of vitamin D responsive genes involves ligand binding to VDR; VDR heterodimerization with RXR; DNA binding of the VDR/RXR complex to the VDRE, and recruitment of basal transcription factors (B), co-regulator molecules (Co-reg), and RNA polymerase II (RNA Pol II) to activate or repress gene transcription. Kidney disease induces several mechanisms (listed on the left) responsible for impairing critical steps (indicated by blue arrows) in calcitriol/VDR transcriptional activity. See CD-ROM for color.

1321

CHAPTER 76 Vitamin D and Renal Failure

1.2

0.8

0.4

12

24-OHlase/GAPDH mRNA

VDR/RXR binding to VDRE

1,25D binding to VDR (fmol/ug DNA)

1.2

0.8

0.4

0

0 N

U

8

4

0 N

U

N

U

FIGURE 6 Uremia-induced impairment in calcitriol/VDR-mediated transcription unrelated to reduced-VDR content. Despite a similar VDR content [specific calcitriol (1,25D)-binding to VDR, left panel] between peripheral blood monocytes from hemodialysis patients (U; n =10) and normal individuals (N; n =10), the binding of the endogenous RXR/VDR complex to the VDRE of the human 24-hydroxylase promoter (middle panel) is markedly impaired in monocytes from hemodialysis patients. Also, a reduced induction of 24-hydroyxlase mRNA levels by 0.24 nM calcitriol (right panel) occurs in monocytes from hemodialysis patients.

and the DNA–binding domain of nuclear receptors, including the VDR, thus interfering with receptormediated transactivation [90]. Hypocalcemia, commonly present in renal failure and caused by either low serum calcitriol levels or hyperphosphatemia, enhances nuclear levels of parathyroid calreticulin. In vitro studies demonstrated that increases in nuclear calreticulin inhibit VDR/RXR binding to VDRE in a dose dependent manner and totally abolish calcitriol suppression of PTH gene transcription [90]. However, there is no report demonstrating actual increases in the nuclear levels of calreticulin induced by CKD in the parathyroid glands of experimental animals that could account for the resistance to calcitriol treatment. Impaired binding of the VDR/RXR complex to DNA and a concomitant reduction in calcitriol/ VDR-mediated transcription also occur as a result of activation of VDR-unrelated pathways. In human monocytes and macrophages, cytokine activation markedly inhibits calcitriol/VDR-mediated gene transcription. Specifically, activation by the cytokine gamma interferon of its signaling molecule, Stat1, induces physical interactions between Stat1 and the DNA-binding domain of the VDR that impair VDR/ RXR binding to VDRE and consequently, calcitriol transactivation of the 24-hydroxylase and osteocalcin genes [91]. The higher levels of inflammatory

cytokines after hemodialysis could contribute to vitamin D resistance. Little is known at present on how renal failure affects the last and most critical step in calcitriol/VDRmediated transactivation or transrepression of target genes. Binding of the VDR/RXR heterodimer to the VDRE of genes induced by vitamin D begins the recruitment of co-activator molecules that act synergistically with the VDR to markedly amplify calcitriol/VDR-mediated transcription [92–94]. The complex interactions of the VDR/RXR with VDRE and co-regulators of VDR-transactivation or transrepression of vitamin D responsive genes raises the possibility that, in uremia, vitamin D resistance could also result from decreased expression of essential coactivator or co-repressor molecules, or from defective recruitment of these molecules by the VDR. Megalin was shown to modulate VDR-mediated transactivation through sequestration of a component of the VDR-transcriptional complex [95]. Thus, the reduction in megalin expression induced by kidney disease could partially account for abnormalities in calcitriol/VDR transcriptional activity. Uremiainduced activation of VDR-unrelated signaling pathways may also interfere with the recruitment by the VDR of co-regulator molecules to the transcription pre-initiation complex.

1322

ADRIANA S. DUSSO, ALEX J. BROWN, AND EDUARDO A. SLATOPOLSKY

The section below describes tissue specific consequences of calcitriol deficiency and vitamin D resistance that contribute to bone disease and systemic toxicity.

IV. TISSUE SPECIFIC EFFECTS OF LOW CALCITRIOL AND ABNORMAL VDR FUNCTION A. Classical Calcitriol Responsive Organs Renal osteodystrophy is the term used to describe the many different patterns of skeletal abnormalities that occur in the course of chronic kidney disease. Alterations in divalent ion, PTH, and vitamin D metabolism play an important part in the pathogenesis and maintenance of skeletal disorders, such as hyperparathyroid bone disease. This section addresses the specific contribution of low serum calcitriol and/or resistance to calcitriol/VDR action in intestine, parathyroid glands, bone, and kidney to alter the powerful calcium-control system causing renal osteodystrophy (see Fig. 1B). 1. INTESTINE

The most critical calcitriol action in vivo is the stimulation of the small intestine to absorb calcium and phosphate, as conclusively demonstrated by recent studies in the VDR null mice [96] (see Chapters 20 and 25). Calcitriol induces active cellular calcium uptake and transport mechanisms (see Chapters 24, 25, and 26). The active-calcium uptake requires the epithelial calcium channel ECaC2 or TRPV6 and, to a lesser extent, ECaC1 (TRPV5). Thereafter, calbindin 9K transports calcium across the cell, and the final delivery to the bloodstream involves the plasma membrane calcium pump. The initial active calcium uptake is highly dependent on vitamin D and the rate-limiting step in intestinal calcium absorption [97]. Studies in the VDR KO mice showed reduced expression of both channels [97,98]. Furthermore, the mRNA levels for both channels are up-regulated upon calcitriol supplementation [99]. Importantly, differential modulation of the expression of these channels, namely up-regulation by calcitriol but not by its analog, 19nor-1,25(OH)2D2, could partially explain the less calcemic properties of the latter [100]. In view of the low serum calcitriol concentrations, reduced intestinal calcium absorption is expected in patients with end-stage renal disease [101,102] and in animal models of experimental renal failure [103,104]. In fact, calcitriol treatment increases intestinal calcium transport in these patients [105]. However, low serum calcitriol levels can only partially account for the

reduction in calcium absorption associated with CKD. A blunted calcitriol stimulation of intestinal calcium absorption occurs in patients with kidney disease and in experimental animals with renal failure compared to normal controls [106]. This finding suggests the existence of abnormalities in either intestinal VDR expression or function. There are no comparative studies on intestinal VDR content in normal individuals and patients with CKD. We also lack epidemiological studies addressing the distribution in the hemodialysis population of the 5′ polymorphism affecting Cdx-2 binding and, consequently, an intestine-specific reduction in VDR expression [79]. Although the contribution of reduced VDR to intestinal calcitriol resistance cannot be completely ruled out, the increases in calcium absorption following dialysis [107,108] suggests the involvement of steps in calcitriol action downstream of reduced intestinal calcitriol/VDR-complex formation. It is unclear whether the increase in calcium absorption after dialysis could be accounted for by the removal of uremic toxins that affect VDR function, or other changes such as a decrease in serum phosphate or volume depletion [108]. In CKD, there is a greater induction of intestinal 24-hydroxylase by calcitriol [46]. This suggests that induced degradation of calcitriol could contribute to the blunted response to the sterol to increase intestinal calcium absorption (see Chapter 24). As with intestinal calcium absorption, phosphate absorption is decreased in renal failure and could be enhanced by administration of calcitriol of 1alpha-hydroxy vitamin D3 (see Chapter 26). 2. PARATHYROID GLANDS

Nearly all patients with end-stage renal disease develop secondary hyperparathyroidism. This condition is characterized by parathyroid hyperplasia and increased synthesis and secretion of parathyroid hormone. High circulating levels of PTH cause osteitis fibrosa, bone loss, and systemic toxicities, including cardiovascular complications, all of which markedly increase morbidity and mortality in hemodialysis patients. Hypocalcemia, hyperphosphatemia due to phosphate retention, and calcitriol deficiency are the three main causes of secondary hyperparathyroidism [109]. Hyperphosphatemia and calcitriol deficiency also enhance parathyroid function indirectly by lowering serum calcium [109]. The ability of calcitriol to inhibit PTH synthesis and arrest parathyroid cell growth in vivo and in vitro has been known for many years [1]. The mechanisms mediating calcitriol-transcriptional repression of the PTH gene are well characterized and described in Chapters 30 and 31. In contrast, direct characterization of the pathogenic mechanisms underlying both induction

1323

CHAPTER 76 Vitamin D and Renal Failure

of parathyroid cell proliferation by kidney disease and its suppression by calcitriol has been difficult because of a lack of an appropriate parathyroid cell line and the rapid dedifferentiation of primary cultures of hyperplastic parathyroid cells. Studies in our laboratory showed that enhanced parathyroid co-expression of the growth promoter, transforming growth factor-α (TGFα) and its receptor, the epidermal growth factor receptor (EGFR), is a major

TGFα

EGFR

LP

HP + D

HP

PCNA

mitogenic signal in experimental renal failure [110]. Similar to secondary hyperparathyroidism in humans [111], parathyroid levels of TGFα are higher in uremic rats than in normal controls. Furthermore, when uremia-induced hyperplasia is worsened by either a high phosphate [110] or a low calcium intake [112], the increases in parathyroid TGFα and EGFR expression correlate directly with high proliferating activity and gland enlargement (see Fig. 7). More importantly, highly

7 days after 5/6NX

FIGURE 7

Enhanced co-expression of TGFα and EGFR contributes to parathyroid hyperplasia in CKD. Parathyroid expression of the potent TGFα/EGFR-growth loop correlates directly with both high proliferating activity [Proliferating Cell Nuclear Antigen (PCNA) expression] and enhanced parathyroid gland size. Representative immunohistochemical staining of PCNA, TGFα and EGFR in parathyroid glands from 5/6 nephrectomized rats fed either a high phosphate (HP: 0.9% P per g of diet) or a low phosphate diet (LP: 0.2% P per g of diet). Whereas, high dietary phosphate worsens uremia-induced parathyroid hyperplasia, phosphate restriction prevents uremia-induced parathyroid hyperplasia. The antiproliferative properties of prophylactic calcitriol administration (D; 4 ng of calcitriol, i.p, every other day for one week) are associated with the prevention of the increases in TGFα and EGFR induced by uremia and high dietary P.

1324

ADRIANA S. DUSSO, ALEX J. BROWN, AND EDUARDO A. SLATOPOLSKY

P < 0.01

PCNA/Area (relative to LP)

8 6

P < 0.01

4 2 0

U-HP + DMSO n = 15

U-HP + AG1478 n = 12

U-LP + DMSO n = 11

FIGURE 8 Enhanced parathyroid TGFα/EGFR co-expression is a major contributor to parathyroid hyperplasia in early renal failure. Prophylactic administration of the highly specific EGFR tyrosinekinase inhibitor, AG1478, prevents the high parathyroid-proliferating activity [Nuclei staining positive for PCNA/Gland Area (PCNA/ Area)] induced by uremia (u) and high dietary P (HP: 1.2% P per g of diet), by 7 days after the onset of renal failure by 5/6 nephrectomy. The potency of anti-EGFR therapy to inhibit PCNA is not down to the level of P restriction (LP: 0.2 % P per g of diet; arbitrarily assigned a value of 1).

specific inhibitors of EGFR-tyrosine kinase, which abolish growth signals from ligand-activated EGFR, prevent uremia- and high phosphate-induced parathyroid cell proliferation (see Fig. 8) [113]. Calcitriol treatment effectively controls parathyroid hyperplasia in early renal failure and also in established secondary hyperparathyroidism. In early stages of kidney disease in rats, prophylactic vitamin D administration [calcitriol or its analog 19-nor-1,25dihydroxyvitamin D2 (19-nor)] counteracts parathyroid hyperplasia through dual mechanisms: Prevention of uremia-induced increases in parathyroid TGFα and EGFR expression, which occur within the first week after inducing kidney failure by 5/6 nephrectomy (see Fig. 7, bottom panels) [112], and induction of the expression of the cell-cycle inhibitor p21 [112]. The efficacy of calcitriol and/or analog therapy [19-nor and 22-oxacalcitriol (OCT)] in preventing further enlargement of the parathyroid gland in established secondary hyperparathyroidism (high TGFα/ EGFR overexpression) suggested that calcitriol antiproliferative properties could involve down-regulation of the potent mitogenic signals emerging from TGFα/ EGFR overexpression. In fact, in normal and carcinogenic cell lines, overexpressing EGFR, the potent vitamin D–antiproliferative actions involve inhibition of EGFR-growth promoting signals from the plasma membrane, as well as EGFR-transactivation of the cyclin D1 gene [114]. Increased expression of parathyroid cyclin D1 is a common feature in secondary hyperparathyroidism in humans [115]. The time of exposure

to calcitriol required to suppress parathyroid expression of TGFα and EGFR in vivo and/or TGFα/EGFR growth signals in vitro [114] suggests the involvement of the VDR rather than rapid, nongenomic calcitriol actions. It is unclear at present whether these novel antiproliferative properties of the calcitriol/VDR complex involve direct transcriptional regulation. In addition to inhibiting the expression and signaling of the TGFα/EGFR-growth loop, calcitriol antiproliferative actions in hyperplastic parathyroid glands involve induction of the cyclin-dependent kinase inhibitors p21 and p27 [112,116]. A strong, direct correlation exists between parathyroid VDR levels and p21 and p27 content in human secondary hyperparathyroidism. The p21 gene is under direct transcriptional induction by the calcitriol/VDR complex [117]. Calcitriol reduction of c-myc expression was also postulated as an underlying mechanism for calcitriol antiproliferative properties in secondary hyperparathyroidism in humans [118]. As renal disease progresses, the low serum calcitriol levels lead to a reduction in parathyroid VDR, thus rendering the parathyroid gland more resistant to the suppression of cell growth and PTH synthesis in response to calcitriol therapy. The ability of calcitriol or its less calcemic analogs to increase parathyroid VDR levels in uremic rats [54] indicates that calcitriol deficiency is, in itself, a determinant of parathyroid resistance in advanced kidney disease. In addition to direct suppression of PTH gene expression, calcitriol appears to be critical for the response of the parathyroid gland to calcium. In rats, parathyroid levels of the calcium sensing receptor (CaSR) mRNA were decreased by 40% by vitamin D deficiency [119] and enhanced by calcitriol treatment in a time and dose-dependent manner. See Chapter 31 for discussion of CaSR. Functional vitamin D responsive elements were identified in both promoters of the human CaSR gene [84]. In advanced renal failure, a strong association exists between defective levels of parathyroid CaSR and low VDR in areas of high proliferative activity [120]. Up-regulation of parathyroid CaSR by calcitriol treatment could explain the decrease in the set point for PTH suppression by calcium [121], and also, the higher levels of the CaSR in surgically removed parathyroid glands from patients receiving calcitriol compared to those from untreated patients [122]. Figure 9 summarizes the multiple direct effects of low serum calcitriol on the parathyroid glands leading to parathyroid hyperplasia and secondary hyperparathyroidism. It is clear from these actions of calcitriol that early interventions with low doses of calcitriol or its less calcemic analogs should prevent the decreases in VDR and CaSR responsible for the reduced sensitivity

1325

CHAPTER 76 Vitamin D and Renal Failure

1,25D

VDR

1,25D/VDR Suppression of PTH gene

Induction of CaSR gene CaSR Ca inhibition of PTHgene expression

Suppression of mitogenic signals

Induction of antimitogenic signals

TGFα/EGFR

p21 p27 anti-EGFR

Set Point

PTH

Hyperplastic growth

FIGURE 9 Role of calcitriol deficiency in the pathogenesis of secondary hyperparathyroidism in chronic kidney disease.

of hyperplastic parathyroid glands to control PTH synthesis and cell growth in response to calcium, calcitriol, or analog therapy. Low-dose calcitriol administration at early stages in CKD should prevent the onset of the most aggressive forms of secondary hyperparathyroidism. 3. BONE

Vitamin D is essential for the development and maintenance of a mineralized skeleton (see Chapters 32, 37, and 40). However, studies in the VDR knock-out mice [96] and children with hereditary Vitamin D resistant rickets (Chapter 72) demonstrated that vitamin D is not mandatory for the ossification process. Calcitriol induces bone mineralization by increasing intestinal absorption and therefore serum levels of calcium and phosphate. In fact, only a fraction of renal failure patients show evidence of defective mineralization, despite the low serum calcitriol levels [123]. In renal failure, calcitriol deficiency causes decreased intestinal calcium absorption and hypocalcemia, a potent stimulus for parathyroid gland hyperplasia and consequently, increases in circulating PTH. High serum PTH levels are the main determinant of osteoclastogenesis and osteoclast activation, causing osteitis fibrosa and bone loss in patients with CKD. The direct effects of calcitriol deficiency leading to hyperparathyroidism and renal osteodystrophy were extensively discussed in the previous section. From these pathogenic mechanisms, it is clear why treatment with calcitriol or its less calcemic analogs is the therapy of choice to control the skeletal abnormalities caused by hyperparathyroidism in advance kidney disease. An important consideration in the use

of calcitriol/analog therapy is to avoid oversuppression of PTH with the consequent decrease in bone turnover to abnormally low levels that cause adynamic bone disease. The recommendations for therapy by the National Kidney Foundation of the USA are specified in the last section of this chapter. 4. KIDNEY

Vitamin D actions in the kidney are the focus of Chapter 29. Because of their major impact on serum calcitriol levels and calcium homeostasis, the most important actions of calcitriol in the kidney are the suppression of 1α-hydroxylase and the stimulation of 24-hydroxylase. In addition to decreasing 1α-hydroxylase activity by reducing serum PTH, calcitriol induction of the expression of the CaSR in the proximal tubules could also mediate the sensitivity of the renal 1α-hydroxylase to calcium. Calcitriol involvement in the renal handling of calcium and phosphate continues to be controversial due to the simultaneous effects of calcitriol on intestinal calcium and phosphate absorption, which affects the filtered load of both ions, and on serum PTH. Calcitriol enhances renal calcium reabsorption and calbindin expression, and accelerates PTH-dependent calcium transport in the distal tubule [124], the kidney site with the highest VDR content and the determinant of the final excretion of calcium into the urine. The epithelial calcium channel (ECaC or LVRT5) is an important target in calcitriol-mediated calcium reabsorption. Several putative VDR binding sites have been located in the human promoter of the renal epithelial calcium channel. More importantly for the calcitriol deficiency that associates with kidney disease, decreases in circulating

1326

ADRIANA S. DUSSO, ALEX J. BROWN, AND EDUARDO A. SLATOPOLSKY

levels of calcitriol resulted in a marked decline in the expression of the channel at the protein and mRNA levels [125].

B. Nonclassical Calcitriol Responsive Systems Hypertension, diabetes, immunological disorders, as well as neuromuscular defects, are important co-morbid conditions in CKD. This section addresses the potential contribution of calcitriol deficiency and/or resistance to calcitriol to the onset of these pathologic features in renal failure patients. It is important to emphasize that all of these pathological processes are multifactorial since no obvious hematopoietic, immunological, or neurological abnormalities were found in the VDR null mice. 1. RENIN/ANGIOTENSIN SYSTEM

Hypertension is a common complication in CKD. The renin/angiotensin system plays a central role in the regulation of blood pressure, electrolyte, and volume homeostasis. Several epidemiological and clinical studies suggested an association between inadequate sunlight exposure or low-serum calcitriol with high blood pressure and/or high plasma renin activity [85,126]. As fully described in Chapter 54, calcitriol acts as a negative endocrine regulator of the renin/angiotensin system. In the VDR null mice, marked increases in renin expression and plasma-angiotensin II production caused hypertension, cardiac hypertrophy, and increased water intake. Further support for an association between calcitriol deficiency and hypertension came from studies in the wild type mice. Whereas inhibition of calcitriol synthesis also led to an increase in renin expression, calcitriol administration suppressed renin production through a VDR-mediated mechanism unrelated to changes in serum calcium. 2. PANCREAS

Vitamin D deficiency is associated with impaired glucose-mediated insulin secretion that can be reversed by calcitriol repletion [127]. Calcitriol, through a VDRmediated modulation of calbindin expression, controls intracellular calcium flux in the islet cells, which in turn affect insulin release [128]. In CKD, there is abnormal insulin secretion, a blunted response of the pancreatic β cell to glucose challenge, and insulin resistance [129–131]. Calcitriol deficiency mediates the abnormal regulation of insulin secretion independently of alterations in VDR levels in pancreatic β cells. Also, calcitriol administration corrects the

abnormal insulin secretion independently of changes in serum levels of calcium or PTH [132,133]. In experimental animals, vitamin D deficiency is also associated with an earlier and more aggressive onset of diabetes (see Chapter 99), a finding that raises the possibility that prolonged calcitriol deficiency in renal failure could accelerate the development of diabetes in kidney disease patients. 3. IMMUNE SYSTEM

Calcitriol is an important immunomodulatory steroid, as extensively reviewed in Chapters 36, 98, and 99. Calcitriol modulates the function of antigenpresenting cells and T lymphocytes. The immune system functions abnormally in CKD— increased cases of infections, decreased response to vaccines, and reduced skin and homograft survival being noted [134]. Serum from CKD patients has been shown to inhibit cell-mediated immune responses in vitro, including lymphocyte blastogenesis, interferon production by normal lymphocytes, and monocyte phagocytosis [134]. The contribution of calcitriol deficiency to abnormal immune cell function was suggested by the alterations in the immune system in vitamin D–deficient rickets. The decreased neutrophil phagocytosis [135] and decreased polymorphonuclear leukocyte and macrophage activity [136] in these patients can be corrected by treatment of cells with 1,25(OH)2D3 in vitro [135,136]. In CKD, calcitriol therapy greatly improved the low lymphocyte counts [137,138], the chemotactic response of polymorphonuclear leukocytes [139], and defects in T-cell-mediated immune responses. It also increased the ratio of T helper to T suppressor cells [140] and enhanced the depressed natural killer cell activity [141]. Mitogen activation of lymphocytes from hemodialysis patients is attenuated, but can be restored by prior treatment of the patients with 1α−hydroxyvitamin D3, a precursor of calcitriol [142]. The production of calcitriol by activated macrophages was postulated as an autocrine/paracrine system that inhibits further T-cell activation and lymphokine production, thus preventing a potentially self-destructive response [135]. As mentioned in the section on the abnormal delivery of substrate to renal and extrarenal 1α-hydroxylases, in CKD, 25OHD uptake was markedly impaired in peripheral blood monocytes from hemodialysis patients and could be normalized by calcitriol treatment. Abnormalities in local calcitriol synthesis induced by calcitriol deficiency could contribute to the abnormal T-cell function in renal failure. In fact, calcitriol blocks the proliferation of activated T-cells in vitro [143] and normalizes the decreased

CHAPTER 76 Vitamin D and Renal Failure

production of interleukin-2, a T-cell growth factor, by lymphocytes from dialysis [144], an effect independent of changes in serum calcium, phosphate, and PTH levels. 4. MUSCLE

Skeletal muscle weakness and atrophy, with electrophysiological abnormalities in muscle contraction and relaxation occur in vitamin D deficiency (see Chapter 102), in calcitriol deficiency due to CKD, and with the prolonged use of anticonvulsant drugs that decrease serum 25OHD levels. Although these defects were originally attributed to low calcium, there is evidence of direct calcitriol action on skeletal muscle [145]. In the heart, calcitriol controls hypertrophy in cardiac myocytes [146] and the synthesis and release of atrial natriuretic factor [147] (see Chapter 55). In end-stage renal disease, therapy with 25OHD or calcitriol improves the left ventricular function in patients with cardiomyopathies and the skeletal muscle weakness. The mechanisms involved are unclear. In vitro, vitamin D analogs elicit a differential potency to regulate muscle-cell metabolism and growth [148], which suggests their therapeutic potential in ameliorating the CKD-associated myopathies. 5. NERVOUS SYSTEM

Calcitriol actions in the nervous system include induction of VDR content (VDR is expressed in the brain and on several regions of the central and peripheral nervous system) [149], as well as increases in the conductance velocity of motor neurons, and induction of the synthesis of neurotrophic factors, such as nerve growth factors and neurotrophins, that prevent the loss of injured neurons [150,151]. Calcitriol also enhances the expression of glial cell line-derived neurotrophic factor, a potential candidate to treat Parkinson’s disease [152]. The impact of calcitriol deficiency in CKD on VDR expression and the neurological abnormalities in these patients is unclear at present. See Chapter 100 for a discussion of Vitamin D action in the central nervous system.

V. VITAMIN D THERAPY IN CHRONIC RENAL FAILURE The ultimate goal of calcitriol therapy is the treatment of osteitis fibrosa, the most common form of renal osteodystrophy, resulting from sustained secondary hyperparathyroidism. Although most patients maintained on chronic dialysis have very low levels of calcitriol, the majority of them do not show on bone

1327 biopsy the presence of osteomalacia secondary to vitamin D deficiency. Thus, as discussed in this section, not all patients with advanced renal insufficiency or even those maintained on chronic hemodialysis or peritoneal dialysis require calcitriol treatment. The therapeutic approaches differ for patients before and after they enter a dialysis program and thus are discussed separately.

A. Prevention and Treatment of Secondary Hyperparathyroidism in Patients with Chronic Renal Failure before Treatment with Dialysis (Chronic Kidney Disease Stages 3 and 4) As described above, the serum levels of calcitriol decrease with the progression of renal disease, and the majority of patients have low levels when the GFR is less than 50 ml/min. Therefore, it would seem appropriate to replace this hormone. However, the use of calcitriol in patients with moderate to advanced CRF is not completely free of side effects, and many physicians do not prescribe calcitriol until there are overt manifestations of secondary hyperparathyroidism and bone disease. Although the concern for potential aggravation of renal insufficiency is understandable, careful administration of calcitriol has been beneficial for the majority of patients. Baker and collaborators [153] studied 16 patients with CKD (creatinine clearance 20–59 ml/min). The patients received either calcitriol at a dose 0.25 to 0.5 µg/daily or placebo. Bone biopsies were performed before entrance into the study and after 12 months of treatment. Bone histology was abnormal in all patients. Calcitriol treatment was associated with a significant fall in serum phosphorus and alkaline phosphatase concentrations, as well as with histological evidence of improvement of hyperparathyroid effects in bone. Over the 12 months of study, there was no significant deterioriation of renal function attributable to the treatment. It was recommended that long-term calcitriol administration in patients with moderate renal failure should be given only to those patients that have high levels of PTH, and that these patients should be followed closely by their physicians. Hypercalcemia arising from calcitriol treatment could further aggravate the abnormality in renal function. Thus, control of serum calcium and phosphorus and frequent monitoring of PTH levels are imperative in order to prevent potential side effects induced by administration of calcitriol.

1328

ADRIANA S. DUSSO, ALEX J. BROWN, AND EDUARDO A. SLATOPOLSKY

Nordal and collaborators [154] studied 13 patients with moderate to terminal renal failure with low doses (average 0.36 ug/day) of calcitriol up to the time of renal transplantation. All patients who started calcitriol treatment and had a creatinine clearance about 30 ml/min had normal bone histology at the time of renal transplantation. This was not observed, however, when calcitriol treatment was started with a creatinine clearance below 30 ml/min. The study suggests that the full benefit of calcitriol at the bone level is obtained only if prophylactic administration is started early in the course of renal failure. It is important to emphasize that these investigators used a small dose of calcitriol. It is possible that such a low dose may not have been sufficient to control secondary hyperparathyroidism and osteitis fibrosa in patients with advanced renal insufficiency. To obtain further information, Hamdy and collaborators [155] treated a large number of patients with 1α-hydroxyvitamin D3 [1αOHD3; alfa-calcidol] in a double-blind, prospective, randomised, placebocontrolled study. They studied 176 patients with mild to advanced CKD (creatinine clearance 15–50 ml/min) and with no clinical, biochemical, or radiographic evidence of bone disease. Two years of 1αOHD3 therapy was initiated at a dose of 0.25 µg per day and was titrated according to serum calcium concentration. A total of 132 patients had histological evidence of bone disease at the start of the study; 89 patients received 1αOHD3, and 87 control patients received placebo. After treatment, PTH concentrations had increased by 126% in controls and had not changed in patients given 1αOHD3 (p < 0.001). Although hypercalcemic episodes occurred in 10 patients, they resolved rapidly after the dose of 1αOHD3 was decreased. Histological indices of high bone turnover significantly improved in patients given 1αOHD3 and significantly deteriorated in controls. There was no difference in the rate of progression of renal failure between the two groups. The investigators concluded that early administration of 1αOHD3 can safely and beneficially alter the natural course of renal bone disease in patients with mild to advanced renal failure. Until recently, there was not a uniform approach for the treatment of mineral and bone metabolism in patients with different degrees of renal insufficiency, and precise recommendations had not been developed. The United States National Kidney Foundation (NKF) has now established a classification for stages of CKD ranging from insignificant to end-stage renal disease. CKD Stage 1 includes GFR > 90 ml/min/ 1.73 m2; Stage 2, GFR from 60 to 89; Stage 3, from 30 to 59; Stage 4, from 29 to 15; Stage 5, GFR < 15 or dialysis. Since secondary hyperparathyroidism and alterations in bone and mineral metabolism are early

manifestations of chronic renal failure, the NKF established the Kidney Disease Outcomes Quality Interactive K/DOQI. These recommendations indicate that treatment for the alterations mentioned above should start early in patients with CKD Stage 3 (GFR from 30 to 59 ml/min/1.73 m2). At this stage of kidney disease, if the intact PTH is greater than 70 pg/ml after correction of serum phosphorus (less than 4.6 mg/dl) and serum calcium (9.5 mg/dl), patients should receive oral calcitriol at the dose of 0.25 µg/day or alfacalcidol at the dose of 0.25 to 0.5 µg/day or doxercalciferol at the dose of 2.5 µg three times per week. It is critical that the patients have levels of 25OHD greater than 30 ng/ml. If the levels are lower, ergocalciferol should be prescribed to the patients. Usually, 50,000 IU every two weeks for one month, and then once a month for six months. Physicians should monitor the levels of 25OHD and adjust the dose accordingly. For patients with CKD Stage 4 (GFR: 15–29 ml/min /1.73 m2), similar recommendations apply. However, treatment with calcitriol or analogs should be instituted if the PTH is greater than 100 pg/ml. Regardless of the way the patient is treated, serum calcium and phosphorus levels should be measured frequently to prevent metastatic calcification, nephrocalcinosis, and acceleration of renal disease. The levels of PTH should be kept between 40–70 pg/ml for CKD Stage 3 and 70–100 pg/ml for patients with CKD Stage 4. Lower values for serum PTH may predispose the patients to develop adynamic bone disease. On the other hand, if the levels of PTH are allowed to increase to higher levels, the patients may develop severe osteitis fibrosa.

B. Treatment of Secondary Hyperparathyroidism in Patients with Chronic Renal Failure Maintained on Hemodialysis (CKD Stage 5) Despite dietary phosphate restriction, the use of phosphate binders, the choice of appropriate levels of calcium in the dialysate, and adequate intake of dietary calcium, a significant number of uremic patients still develop features of osteitis fibrosa. The knowledge of the major pathogenic role of calcitriol deficiency to renal osteodystrophy has created interest in the use of calcitriol in such patients. When patients show evidence of secondary hyperparathyroidism (e.g., bone erosions, high intact PTH levels (greater than 300 pg/ml), and increased alkaline phosphatase), adequate treatment with calcitriol often leads to improvement. Results of numerous studies indicate the efficacy of calcitriol treatment in patients with symptomatic renal osteodystrophy. These clinical

CHAPTER 76 Vitamin D and Renal Failure

evaluations have shown an improvement in muscle strength and bone pain. In addition, biochemical markers, such as plasma alkaline phosphatase, have decreased along with a fall in the levels of serum PTH. Bone histology has shown a decrease in marrow fibrosis and other features of secondary hyperparathyroidism, such as increased bone resorption and number of osteoclasts. The dose of oral calcitriol utilized in these trials has varied from 0.25 to 1.0 µg/day, and the major side effect of such treatment is the appearance of hypercalcemia. Hypercalcemia may occur only after many weeks or months of treatment, or it may appear sooner in patients with aluminum-induced osteomalacia or in adynamicbone disease. There is substantial degradation of calcitriol in the intestine, and it is possible that the oral administration of the vitamin D sterols augments calcium absorption very effectively; however, the delivery of calcitriol to other target organs may be substantially less. In fact, studies by Maung and collaborators [156] have demonstrated that oral administration of 1alphavitamin D2 is more calcemic and phosphatemic than the intravenous route. Slatopolsky and collaborators [157] studied the effects of intravenous administration of calcitriol in patients maintained on chronic hemodialysis. Twenty patients were given calcitriol intravenously at the end of each dialysis. The dose was initially 0.5 µg and was gradually increased to a maximum of 4.0 µg per dialysis. Calcitriol was discontinued after eight weeks of treatment, and blood samples were obtained for an additional three weeks. In all patients, there was a substantial decrease in the levels of PTH during the period of intravenous calcitriol treatment, with a mean decrement of 71 ± 3.2%. After calcitriol was discontinued, PTH increased rapidly in all patients. There was a significant correlation between the increase in ionized calcium and the decrease in PTH level, showing a crucial role for calcium in the suppression of PTH. However, in addition to the calcemic effect, calcitriol directly modified the secretion of PTH. The decrease in PTH levels was observed before there was an increase in ionized calcium. The effects of oral and intravenous administration of calcitriol were compared, and it was shown that the intravenous calcitriol had a greater suppressive effect on the release of PTH than when calcitriol was administered by the oral route. The intravenous administration of calcitriol may allow a greater delivery to peripheral tissues such as the parathyroid glands and thereby generate greater expression of biological effects at these sites. Similar results were found by Andress and collaborators [158]; they studied 12 patients on hemodialysis who were not responding to oral calcitriol and were being considered

1329 for parathyroidectomy at the time. All patients exhibited baseline bone formation rates that were above normal, and the rates fell by a mean of 59% during treatment. The results indicate that intravenous administration of calcitriol is also effective in ameliorating osteitis fibrosa in patients who have moderate to severe secondary hyperparathyroidism. As mentioned before, oral calcitriol failed to suppress the secretion of PTH adequately in these patients. Parathyroidectomy, originally considered for these patients, became unnecessary due to the implementation of intravenous calcitriol. Since the original study, more than 75 reports on the effect of intravenous calcitriol therapy in hemodialyzed CKD patients have been published, including information on more than 1000 patients. In several studies 1αOHD3 was administered. Interestingly, Delmez and collaborators [121], in a study of the effect of calcitriol in a group of patients maintained on hemodialysis, found significant decreases in the levels of serum PTH and an improvement in the calcium set point for PTH secretion after two weeks of intravenous calcitriol therapy (Fig. 10). Similar results were found by Dunlay et al. [159] in a group of nine patients maintained on hemodialysis. The patients received 2 µg intravenously after each dialysis. Intravenous calcitriol resulted in a significant decrease within two weeks and continued decrease of the high serum PTH levels by the end of 10 weeks. Although these investigators did not find a change in the calcium set point, they found a shift in the calcium-PTH sigmoidal curve toward normal. Thus, both studies [121,159] show an increase in sensitivity of the parathyroid glands to serum calcium after the administration of intravenous calcitriol. In other countries, owing to an inability to obtain the intravenous preparation of calcitriol, several clinical trials investigated the use of high doses of calcitriol given orally in an intermittent manner (oral pulse therapy). Several studies in adults and children with renal failure have been published. Most of the results show significant improvement in the suppression of PTH [160,161]. However, in many cases this approach had to be discontinued because of the development of hypercalcemia and hyperphosphatemia. Quarles et al. [162] published the results of a prospective trial of oral pulse versus intravenous calcitriol in the treatment of secondary hyperparathyroidism in hemodialysis patients. These investigators found that episodes of hypercalcemia and hyperphosphatemia were similar in both treatment groups and limited the dose of calcitriol that could be administered. In this study, they found that intermittent calcitriol therapy, regardless of the route of administration, was poorly tolerated, failed to correct parathyroid gland size and functional abnormalities, and had a limited ability to sustain serum

1330

ADRIANA S. DUSSO, ALEX J. BROWN, AND EDUARDO A. SLATOPOLSKY

350

Control +1,25(OH)2D3 = Set point

325

N–PTH (pq/ml)

300

275

Shift 250

225

200

175

0 4.6

5.0

5.4 5.6 ICa (mq/dl)

5.8

6.2

FIGURE 10

Effects of intravenous calcitriol on PTH secretion during calcium infusion in a representative patient. During the control infusion (•) the set point (x) of serum ionized calcium (ICa) was 5.04 mg/dl. After two weeks of intravenous calcitriol ( ), the PTH levels decreased, despite a lower ICa value, and the set point decreased to 4.64 mg/dl. From Delmez et al. [121].

1200

1.6

1000

1.5

800

1.4 1.3

600

1.2 400 1.1 200 0

Ca2+, mmol/liter

indicated that, in some resistant patients, higher doses of calcitriol (4 to 6 µg/treatment) were able to reduce the levels of PTH. Llach and collaborators [164] also demonstrated the importance in dosing intravenous calcitriol in dialysis

i-PTH, pg/ml

PTH reductions in end-stage renal failure in patients with severe hyperparathyroidism. Unfortunately, the development of severe hyperphosphatemia in many patients interferes with the beneficial effects of calcitriol. On the other hand, Cannella et al. [163] found significant healing of secondary hyperparathyroidism in chronic hemodialysis patients treated with long-term intravenous calcitriol. These investigators followed a group of patients for approximately 35 weeks; the initial dose of calcitriol was 30 ng/kg of body weight intravenously three times a week after each dialysis. The mean pretreatment serum PTH concentration was 966 ± 160 pg/ml, and the values decreased significantly by the first week and fell by an average of 80% by week 35 (Fig. 11). The ionized calcium concentration was 4.76 ± 0.4 mg/dl and rose slightly to 5.36 mg/dl by the fourth week. There were significant decreases in all bone-morphometric indices of secondary hyperparathyroidism. These investigators clearly demonstrated, contrary to the studies of Quarles and collaborators [162], that intravenous calcitriol is very effective in suppressing secondary hyperparathyroidism. It is of utmost importance to emphasize that Cannella and collaborators [163] were able to control the levels of serum phosphorus, allowing them to provide calcitriol on a more sustained basis. They also

0 −10 −5

0.9 0

5 10 15 20 Time, weeks

25 30 35

FIGURE 11 Weekly values (means ± SE) for plasma concentrations of intact PTH ( ) and ionized calcium (•) in 8 hemodialysis patients, before and after the start (arrow) of intravenous calcitriol therapy (30 ng/kg body weight thrice weekly for 8 months). Asterisks indicate significant results versus time 0: *, p < 0.05; **, p < 0.01. Dashed area is the normal PTH range. Used with permission from Kidney International, vol 46, p. 1124, 1994.

1331

CHAPTER 76 Vitamin D and Renal Failure

patients with severe hyperparathyroidism. They studied 10 patients with severe hyperparathyroidism (PTH > 1200 pg/ml and serum phosphorus < 6.5 mg/dl). Ten patients with a mean PTH of 1826 ± 146 pg/ml, were treated for a minimum of 48 weeks with an intravenous dose of calcitriol commensurate with the levels of PTH. The initial calcitriol dose had to be increased in seven patients. The mean maximum dose of calcitriol was 3.8 µg three times a week. The authors concluded that patients with severe hyperparathyroidism respond well to intravenous calcitriol; however, the dose of this vitamin D metabolite should be adjusted according to PTH levels, and hyperphosphatemia should be kept under control. Thus, it seems that although there is no agreement with regard to the route and frequency of administration, the oral pulse and intravenous administration are the most accepted therapies. However, the meta-analysis of four trials that compared intermittent intravenous calcitriol with oral calcitriol, in randomized controlled studies [165,166] or cross-over trials [167,168], indicated that intravenous therapy was more effective than oral treatment (either daily or “pulse” treatment) for the suppression of intact PTH levels. Beneficial results also have been observed in patients maintained on continuous ambulatory peritoneal dialysis. Salusky et al. [169] studied the pharmokinetics of calcitriol in continuous ambulatory and cycling peritoneal dialysis patients. The kinetics of calcitriol was evaluated after a single dose of 60 ng/kg body weight (equivalent to 4.2 ug for a 70-kg man) given orally, intravenously, or intraperitoneally in six patients. The area under the curve for the increment of serum calcitriol concentration above baseline levels for the 24 hr after a single dose of calcitriol was 62% greater following intravenous injection (2340 ± 115 pg/ml) than after either oral (1442 ± 191 pg/ml) or intraperitoneal (1562 ± 195 pg/ml) administration. These investigators, using a radioisotope tracer of calcitriol, found that 30 to 40% of the hormone adheres to plastic components of the peritoneal dialysate delivery system. By modifying the technique of intraperitoneal calcitriol administration, the authors found that they could obtain a dosage effect comparable to intravenous administration. Thus, it would seem that intraperitoneal administration of calcitriol also is very effective in the control of secondary hyperparathyroidism if precautions to prevent adherence to plastic are taken. It is recommended now that in patients with CKD Stage 5 (GFR < 15 ml/min/1.73 m2) or dialysis, serum PTH be determined at least every three months and serum calcium and phosphorus once a month. The ideal serum intact PTH levels should be 150 to 300 pg/ml. Patients with serum intact PTH greater than 300 pg/ml should receive calcitriol or vitamin D analogs, providing

that the serum phosphorus is less than 5.5 mg/dl and the serum calcium less than 9.8 mg/dl. Ideally, the serum phosphorus should be 3.5 to 5.5 mg/dl, and the Ca × P product less than 55 mg2 × dl2. The administration of calcitriol or its analogs to patients with CKD Stage 5 depends on the levels of circulating intact PTH. For patients with circulating levels of intact PTH of 300–600 pg/ml, the dose of intravenous calcitriol should be 0.5 to 1.5 µg I.V. × 3 per week. For those patients with more severe secondary hyperparathyroidism (PTH 600–1200), the dose of calcitriol should be 2 to 4 µg I.V. three times per week.

C. Use of New Less Calcemic Analogs of Calcitriol In an effort to utilize the actions of vitamin D on the parathyroid gland and minimize the toxicities of such therapy, structural alterations of the vitamin D molecule were undertaken to try to develop vitamin D analogs that may retain the effects on the parathyroid glands but have a lesser effect on the calcium and phosphate metabolism [170]. These analogs would be relatively selective in suppressing parathyroid hyperfunction and therefore, more useful therapeutic agents. This subject is extensively discussed in Chapters 80–88. Currently, there is experimental and clinical evidence for the efficacy of four of such vitamin D analogs, which have been approved for the treatment of secondary hyperparathyroidism. Two of these analogs have been developed in Japan, 22-oxacalcitriol and falecalcitriol and the other two in the United States, 19-nor-1,25(OH)2D2 and 1-alpha hydroxy D2. 22-Oxacalcitriol (OCT) differs from calcitriol by an oxygen substitution at position 22 (see Chapter 86). This structure modification appears to reduce the affinity of OCT for the vitamin D receptor, as well as for DBP. The decreased affinity for DBP results in rapid clearance from the circulation, and this may be a mechanism that accounts for low calcemic and phosphatemic effect of OCT [171]. Falecalcitriol is an analog in which the hydrogens of carbons 26 and 27 have been substituted by fluorine atoms. This vitamin D analog has greater activity than calcitriol and is considerably more calcemic and more potent in calcifying epiphyseal cartilage in rats [172]. The increased potency is likely due to a decreased catabolism of this sterol. In patients with chronic renal failure, falecalcitriol was effective in decreasing PTH and appeared to be somewhat more effective than alfacalcidol in suppressing secondary hyperparathyroidism [173]. In the United States, 19-nor-1,25(OH)2D2 (paricalcitol) has been released into the market with the name

ADRIANA S. DUSSO, ALEX J. BROWN, AND EDUARDO A. SLATOPOLSKY

of Zemplar®. This vitamin D analog lacks the carbon at position 19. Zemplar® has been studied extensively and demonstrated to suppress PTH secretion in vitro as potently as calcitriol. Studies in experimental animals have shown that 19-nor-1,25(OH)2D2 is effective in suppressing PTH levels with less hypercalcemia and hyperphosphatemia than that observed with calcitriol treatment. Indeed 19-nor-1,25(OH)2D2 is approximately 10 times less active than calcitriol in mobilizing calcium and phosphate from bone [174]. This vitamin D analog is in widespread clinical use in patients on hemodialysis in the United States, and has been demonstrated to be effective in suppressing PTH levels. Thus, while 3 times more 19-nor-1,25(OH)2D2 than calcitriol is required to achieve equivalent suppression of PTH in animals, studies in patients indicate that a ratio of 3 to 4 is required [175–177]. Similarly, while paricalcitol is 10 times less calcemic and phosphatemic than calcitriol in animals studies, in patients with end-stage renal disease on a low calcium diet, at least 8 times more paricalcitol is required to achieve a similar increment in serum calcium, presumably representing calcium mobilized from bone [178]. Sprague and collaborators demonstrated less severe hyperphosphatemia in patients treated with paricalcitol compared to calcitriol [176]. Recent studies indicate a 16% decrease in mortality in a retrospective study over a period of three years, in a large group of patients treated with paricalcitol (29,025) when compared to those receiving calcitriol (38,378) [179]. Moreover, the two-year survival rate among patients who switched from calcitriol to paricalcitriol was 73% as compared to 64% among those who switched from paricalcitol to calcitriol. The exact mechanism for this effect is not clear at the present time. Another analog of vitamin D is 1-alpha hydroxy D2, commercially known as Hectorol®. This analog is used in the United States for the treatment of secondary hyperparathyroidism. This compound is considered a prohormone since it lacks the 25-hydroxyl group, and it is 25-hydroxylated in the liver to 1,25(OH)2D2. Comparative studies in normal and uremic animals (see Fig. 12) have shown [180] that 1-alpha hydroxy D2 is more hypercalcemic and hyperphosphatemic than 19-nor-1,25(OH)2D2. Further studies in patients are necessary to corroborate this initial experimental observation.

VI. SUMMARY PTH and calcitriol are the major factors responsible for maintaining extracellular calcium homeostasis

125 Ca × P product (mg2/dl2)

1332

100 75 50 25 0

Vehicle

50 ng

100 ng 250 ng

19-norD2

50 ng

100 ng

250 ng

1α-OHD2

FIGURE 12 Effects of vehicle, 19-nor-1,25(OH)2D2 [19-nor; 50, 100, or 250 ng; n = 10] and 1α(OH)D2 [1α(OH)D2; 50, 100, or 250 ng; n = 9] on the Ca × P product in uremic rats. Rats were treated three times a week for two weeks. * and ** indicate p < 0.01 or p < 0.001 versus control rats. Adapted from Slatopolsky et al. [180].

within narrow limits despite the large bidirectional fluxes of calcium across the intestine, bone, and especially the kidney. In chronic kidney disease, the progressive reduction in kidney function not only causes defective renal handling of calcium and phosphorus, but a decrease in renal calcitriol synthesis, which is proportional to the reduction in functional renal mass. In addition to reduced renal 1α-hydroxylase, the enzyme synthesizing calcitriol, several mechanisms contribute to worsen renal calcitriol synthesis. These mechanisms include: impaired substrate availability to renal 1α-hydroxylase; inhibition of renal 1α-hydroxylase activity by hyperphosphatemia, metabolic acidosis and accumulation of uremic toxins; and a blunted response to PTH induction of 1α-hydroxylase activity. The low serum calcitriol levels result in a marked reduction in intestinal calcium absorption with concomitant hypocalcemia, as well as a proportional reduction in VDR levels in critical targets, such as the parathyroid glands. Chronic kidney disease also impairs the activity of the calcitriol/VDR-complex as a transcriptional regulator of the expression of calcitriol responsive genes. Two mechanisms were identified: A reduction in the cellular levels of the VDR-transcriptional partner, the retinoidX-receptor, RXR, and impaired VDR/RXR-DNA– binding interactions. Accumulation of uremic toxins; hypocalcemia- and/or hyperphosphatemia-induced increases in nuclear calreticulin; and uremia-induced activation of transcription factors from VDR-unrelated pathways reduce VDR binding to vitamin D responsive elements on the DNA.

1333

CHAPTER 76 Vitamin D and Renal Failure

In the parathyroid glands, the reduction in calcitriol/ VDR expression and transcriptional activity results in a defective inhibition of PTH synthesis as well as the potent mitogenic signals emerging from overexpression of TGFα and EGFR; impaired induction of the antiproliferative molecules p21 and p27; and defective induction of the CaSR. These defects cause parathyroid hyperplasia, high serum PTH, and reduced sensitivity of the parathyroid gland to suppress growth and PTH secretion in response to calcitriol or calcium. High serum PTH levels lead to osteitis fibrosa and bone loss and systemic toxicities, all of which increase the morbidity and mortality in patients with CKD. Early therapeutic interventions with calcitriol are recommended to delay the onset of calcitriol resistance by preventing the decreases in parathyroid-VDR and CaSR content. In established secondary hyperparathyroidism, the new less-calcemic vitamin D analogs are the treatment of choice. Four of these analogs are available in the USA and Japan. Although not all analog formulations are equally effective in controlling hypercalcemia and hyperphosphatemia in patients with advanced kidney disease, they offer a wider therapeutic window to counteract vitamin D resistance without causing adynamic bone disease. In the case of paricalcitol, there is a survival advantage over exclusive calcitriol therapy. The 2003 recommendations by the National Kidney Foundation of the USA provide optimal dosage of calcitriol and its less calcemic analogs for the different stages of CKD, as well as the corrections in the therapeutic approach based on a close control of serum PTH, P and Ca levels to maximize the efficacy of treatment avoiding adynamic bone disease and the risk of vascular calcifications.

7.

8. 9.

10.

11.

12. 13.

14. 15.

16.

17.

References 1. Brown AJ, Dusso A, Slatopolsky E 1999 Vitamin D. Am J Physiol 277(2 Pt 2):F157–75. 2. Slatopolsky E, Brown A, Dusso A 1999 Pathogenesis of secondary hyperparathyroidism. Kidney Int Suppl 73:S14–S19. 3. Martinez I, Saracho R, Montenegro J, Llach F 1996 A deficit of calcitriol synthesis may not be the initial factor in the pathogenesis of secondary hyperparathyroidism. Nephrol Dial Transplant 11(Suppl 3):22–8. 4. Mason RS, Lissner D, Wilkinson M, Posen S 1980 Vitamin D metabolites and their relationship to azotaemic osteodystrophy. Clin Endocrinol (Oxf) 13(4):375–85. 5. Christiansen C, Christensen MS, Melsen F, Rodbro P, DeLuca HF 1981 Mineral metabolism in chronic renal failure with special reference to serum concentrations of 1,25(OH)2D and 24,25(OH)2D. Clin Nephrol 15(1):18–22. 6. Juttmann JR, Buurman CJ, De Kam E, Visser TJ, Birkenhager JC 1981 Serum concentrations of metabolites of

18. 19.

20. 21.

22.

vitamin D in patients with chronic renal failure (CRF). Consequences for the treatment with 1-alpha-hydroxyderivatives. Clin Endocrinol (Oxf) 14(3):225–36. Tessitore N, Venturi A, Adami S, Roncari C, Rugiu C, Corgnati A, Bonucci E, Maschio G 1987 Relationship between serum vitamin D metabolites and dietary intake of phosphate in patients with early renal failure. Miner Electrolyte Metab 13(1):38–44. Wilson L, Felsenfeld A, Drezner MK, Llach F 1985 Altered divalent ion metabolism in early renal failure: role of 1,25(OH)2D. Kidney Int 27(3):565–73. Halloran BP, Schaefer P, Lifschitz M, Levens M, Goldsmith RS 1984 Plasma vitamin D metabolite concentrations in chronic renal failure: effect of oral administration of 25-hydroxyvitamin D3. J Clin Endocrinol Metab 59(6):1063–9. Nykjaer A, Dragun D, Walther D, Vorum H, Jacobsen C, Herz J, Melsen F, Christensen EI, Willnow TE 1999 An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) vitamin D3. Cell 96(4):507–15. Takemoto F, Shinki T, Yokoyama K, Inokami T, Hara S, Yamada A, Kurokawa K, Uchida S 2003 Gene expression of vitamin D hydroxylase and megalin in the remnant kidney of nephrectomized rats. Kidney Int 64(2):414–20. Liu W, Yu WR, Carling T, Juhlin C, Rastad J, Ridefelt P, Akerstrom G, Hellman P 1998 Regulation of gp330/megalin expression by vitamins A and D. Eur J Clin Invest 28(2):100–7. Gallieni M, Kamimura S, Ahmed A, Bravo E, Delmez J, Slatopolsky E, Dusso A 1995 Kinetics of monocyte 1 alphahydroxylase in renal failure. Am J Physiol 268(4 Pt 2): F746–53. Dusso A, Lopez-Hilker S, Rapp N, Slatopolsky E 1988 Extra-renal production of calcitriol in chronic renal failure. Kidney Int 34(3):368–75. Kohno M, Takahashi S, Oida K, Suzuki J, Tamai T, Yamamoto T, Nakai T 1997 1 alpha,25-dihydroxyvitamin D3 induces very low-density lipoprotein receptor mRNA expression in HL-60 cells in association with monocytic differentiation. Atherosclerosis 133(1):45–9. Jouni ZE, Winzerling JJ, McNamara DJ 1995 1,25Dihydroxyvitamin D3–induced HL-60 macrophages: regulation of cholesterol and LDL metabolism. Atherosclerosis 117(1):125–38. Haussler MR, Baylink DJ, Hughes MR, Brumbaugh PF, Wergedal JE, Shen FH, Nielsen RL, Counts SJ, Bursac KM, McCain TA 1976 The assay of 1alpha,25-dihydroxyvitamin D3: physiologic and pathologic modulation of circulating hormone levels. Clin Endocrinol (Oxf) 5 (Suppl):151S–65S. Garabedian M, Holick MF, Deluca HF, Boyle IT 1972 Control of 25-hydroxycholecalciferol metabolism by parathyroid glands. Proc Natl Acad Sci USA 69(7):1673–6. Ritz E, Seidel A, Ramisch H, Szabo A, Bouillon R 1991 Attenuated rise of 1,25 (OH)2 vitamin D3 in response to parathyroid hormone in patients with incipient renal failure. Nephron 57(3):314–18. Henry HL, Norman AW 1984 Vitamin D: metabolism and biological actions. Annu Rev Nutr 4:493–520. Murayama A, Takeyama K, Kitanaka S, Kodera Y, Hosoya T, Kato S 1998 The promoter of the human 25-hydroxyvitamin D3 1 alpha-hydroxylase gene confers positive and negative responsiveness to PTH, calcitonin, and 1 alpha,25(OH)2D3. Biochem Biophys Res Commun 249(1):11–16. Brenza HL, Kimmel-Jehan C, Jehan F, Shinki T, Wakino S, Anazawa H, Suda T, DeLuca HF 1998 Parathyroid hormone

1334

23.

24.

25. 26. 27.

28. 29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

ADRIANA S. DUSSO, ALEX J. BROWN, AND EDUARDO A. SLATOPOLSKY

activation of the 25-hydroxyvitamin D3-1alpha-hydroxylase gene promoter. Proc Natl Acad Sci USA 95(4):1387–91. Bellorin-Font E, Humpierres J, Weisinger JR, Milanes CL, Sylva V, Paz-Martinez V 1985 Effect of metabolic acidosis on the PTH receptor-adenylate cyclase system of canine kidney. Am J Physiol 249(4 Pt 2):F566–72. Shinki T, Shimada H, Wakino S, Anazawa H, Hayashi M, Saruta T, DeLuca HF, Suda T 1997 Cloning and expression of rat 25-hydroxyvitamin D3-1alpha-hydroxylase cDNA. Proc Natl Acad Sci USA 94(24):12920–5. Beck N, Kim HP, Kim KS 1975 Effect of metabolic acidosis on renal action of parathyroid hormone. Am J Physiol 228(5):1483–88. Prince RL, Hutchison BG, Kent JC, Kent GN, Retallack RW 1988 Calcitriol deficiency with retained synthetic reserve in chronic renal failure. Kidney Int 33(3):722–8. Portale AA, Halloran BP, Morris RC Jr 1989 Physiologic regulation of the serum concentration of 1,25-dihydroxyvitamin D by phosphorus in normal men. J Clin Invest 83(5):1494–9. Gray RW 1981 Control of plasma 1,25-(OH)2-vitamin D concentrations by calcium and phosphorus in the rat: effects of hypophysectomy. Calcif Tissue Int 33(5):485–8. Zhang MY, Wang X, Wang JT, Compagnone NA, Mellon SH, Olson JL, Tenenhouse HS, Miller WL, Portale AA 2002 Dietary phosphorus transcriptionally regulates 25OHD-1alphahydroxylase gene expression in the proximal renal tubule. Endocrinology 143(2):587–95. Yoshida T, Yoshida N, Monkawa T, Hayashi M, Saruta T 2001 Dietary phosphorus deprivation induces 25-hydroxyvitamin D3 1alpha-hydroxylase gene expression. Endocrinology 142(5):1720–6. Lucas PA, Brown RC, Woodhead JS, Coles GA 1986 1,25dihydroxycholecalciferol and parathyroid hormone in advanced chronic renal failure: effects of simultaneous protein and phosphorus restriction. Clin Nephrol 25(1):7–10. Lopez-Hilker S, Dusso AS, Rapp NS, Martin KJ, Slatopolsky E 1990 Phosphorus restriction reverses hyperparathyroidism in uremia independent of changes in calcium and calcitriol. Am J Physiol 259(3 Pt 2):F432–7. Tenenhouse HS, Martel J, Gauthier C, Zhang MY, Portale AA 2001 Renal expression of the sodium/phosphate cotransporter gene, Npt2, is not required for regulation of renal 1 alphahydroxylase by phosphate. Endocrinology 142(3):1124–9. Portale AA, Booth BE, Halloran BP, Morris RC Jr 1984 Effect of dietary phosphorus on circulating concentrations of 1,25-dihydroxyvitamin D and immunoreactive parathyroid hormone in children with moderate renal insufficiency. J Clin Invest 73(6):1580–9. Gafter U, Kraut JA, Lee DB, Silis V, Walling MW, Kurokawa K, Haussler MR, Coburn JW 1980 Effect of metabolic acidosis in intestinal absorption of calcium and phosphorus. Am J Physiol 239(6):G480–4. Bushinsky DA, Favus MJ, Schneider AB, Sen PK, Sherwood LM, Coe FL 1982 Effects of metabolic acidosis on PTH and 1,25(OH)2D3 response to low calcium diet. Am J Physiol 243(6):F570–5. Kraut JA, Gordon EM, Ransom JC, Horst R, Slatopolsky E, Coburn JW, Kurokawa K 1983 Effect of chronic metabolic acidosis on vitamin D metabolism in humans. Kidney Int 24(5):644–8. Bushinsky DA, Riera GS, Favus MJ, Coe FL 1985 Response of serum 1,25(OH)2D3 to variation of ionized calcium during chronic acidosis. Am J Physiol 249(3 Pt 2):F361–5.

39. Langman CB, Bushinsky DA, Favus MJ, Coe FL 1986 Ca and P regulation of 1,25(OH)2D3 synthesis by vitamin D– replete rat tubules during acidosis. Am J Physiol 251(5 Pt 2): F911–8. 40. Hsu CH, Patel S 1992 Uremic plasma contains factors inhibiting 1 alpha-hydroxylase activity. J Am Soc Nephrol 3(4):947–52. 41. Maierhofer WJ, Gray RW, Adams ND, Smith GA, Lemann J Jr 1981 Synthesis and metabolic clearance of 1,25-dihydroxyvitamin D as determinants of serum concentrations: a comparison of two methods. J Clin Endocrinol Metab 53(3):472–5. 42. Henry HL 1981 Insulin permits parathyroid hormone stimulation of 1,25-dihydroxyvitamin D3 production in cultured kidney cells. Endocrinology 108(2):733–5. 43. Hsu CH, Patel S, Buchsbaum BL 1991 Calcitriol metabolism in patients with chronic renal failure. Am J Kidney Dis 17(2):185–90. 44. Hsu CH, Patel S, Young EW, Simpson RU 1987 Production and degradation of calcitriol in renal failure rats. Am J Physiol 253(5 Pt 2):F1015–9. 45. Dusso A, Lopez-Hilker S, Lewis-Finch J, Grooms P, Brown A, Martin K, Slatopolsky E 1989 Metabolic clearance rate and production rate of calcitriol in uremia. Kidney Int 35(3):860–4. 46. Koyama H, Inaba M, Nishizawa Y, Ishimura E, Imanishi Y, Hini M, Furuyama T, Takagi H, Morii H 1994 Potentiated 1,25(OH)2D3-induced 24-hydroxylase gene expression in uremic rat intestine. Am J Physiol 267(6 Pt 2):F926–30. 47. Lee SW, Russell J, Avioli LV 1977 25-hydroxycholecalciferol to 1,25-dihydroxycholecalciferol: conversion impaired by systemic metabolic acidosis. Science 195(4282):994–6. 48. Costa EM, Feldman D 1986 Homologous up-regulation of the 1,25 (OH)2 vitamin D3 receptor in rats. Biochem Biophys Res Commun 137(2):742–7. 49. Wiese RJ, Uhland-Smith A, Ross TK, Prahl JM, DeLuca HF 1992 Up-regulation of the vitamin D receptor in response to 1,25-dihydroxyvitamin D3 results from ligand-induced stabilization. J Biol Chem 267(28):20082–6. 50. Fukuda N, Tanaka H, Tominaga Y, Fukagawa M, Kurokawa K, Seino Y 1993 Decreased 1,25-dihydroxyvitamin D3 receptor density is associated with a more severe form of parathyroid hyperplasia in chronic uremic patients. J Clin Invest 92(3): 1436–43. 51. Korkor AB 1987 Reduced binding of [3H]1,25-dihydroxyvitamin D3 in the parathyroid glands of patients with renal failure. N Engl J Med 316(25):1573–7. 52. Merke J, Hugel U, Zlotkowski A, Szabo A, Bommer J, Mall G, Ritz E 1987 Diminished parathyroid 1,25(OH)2D3 receptors in experimental uremia. Kidney Int 32(3):350–3. 53. Brown AJ, Dusso A, Lopez-Hilker S, Lewis-Finch J, Grooms P, Slatopolsky E 1989 1,25-(OH)2D receptors are decreased in parathyroid glands from chronically uremic dogs. Kidney Int 35(1):19–23. 54. Denda M, Finch J, Brown AJ, Nishii Y, Kubodera N, Slatopolsky E 1996 1,25-dihydroxyvitamin D3 and 22-oxacalcitriol prevent the decrease in vitamin D receptor content in the parathyroid glands of uremic rats. Kidney Int 50(1):34–9. 55. Koyama H, Nishizawa Y, Inaba M, Hino M, Prahl JM, DeLuca HF, Morii H 1994 Impaired homologous up-regulation of vitamin D receptor in rats with chronic renal failure. Am J Physiol 266(5 Pt 2):F706–12. 56. Reinhardt TA, Horst RL 1990 Parathyroid hormone downregulates 1,25-dihydroxyvitamin D receptors (VDR) and VDR messenger ribonucleic acid in vitro and blocks homologous up-regulation of VDR in vivo. Endocrinology 127(2):942–8.

CHAPTER 76 Vitamin D and Renal Failure

57. Patel SR, Ke HQ, Hsu CH 1994 Regulation of calcitriol receptor and its mRNA in normal and renal failure rats. Kidney Int 45(4):1020–27. 58. Goff JP, Reinhardt TA, Beckman MJ, Horst RL 1990 Contrasting effects of exogenous 1,25-dihydroxyvitamin D [1,25-(OH)2D] versus endogenous 1,25-(OH)2D, induced by dietary calcium restriction, on vitamin D receptors. Endocrinology 126(2):1031–35. 59. Sandgren ME, DeLuca HF 1990 Serum calcium and vitamin D regulate 1,25-dihydroxyvitamin D3 receptor concentration in rat kidney in vivo. Proc Natl Acad Sci USA 87(11): 4312–14. 60. Uhland-Smith A, DeLuca HF 1993 The necessity for calcium for increased renal vitamin D receptor in response to 1,25-dihydroxyvitamin D. Biochim Biophys Acta 1176(3): 321–6. 61. Brown AJ, Zhong M, Finch J, Ritter C, Slatopolsky E 1995 The roles of calcium and 1,25-dihydroxyvitamin D3 in the regulation of vitamin D receptor expression by rat parathyroid glands. Endocrinology 136(4):1419–25. 62. Huang YC, Lee S, Stolz R, Gabrielides C, Pansini-Porta A, Bruns ME, Bruns DE, Miffin TE, Pike JW, Christakos S 1989 Effect of hormones and development on the expression of the rat 1,25-dihydroxyvitamin D3 receptor gene. Comparison with calbindin gene expression. J Biol Chem 264(29):17454–61. 63. Martinez J, Olmos JM, de Francisco AL, Amado JA, Riancho JA, Gonzalez-Macias J 1994 1,25-Dihydroxyvitamin D3 receptors in peripheral blood mononuclear cells from patients with primary and secondary hyperparathyroidism. Bone Miner 27(1):25–32. 64. Morrison NA, Qi JC, Tokita A, Kelly PJ, Crofts L, Nguyen TV, Sambrook PN, Eisman JA 1994 Prediction of bone density from vitamin D receptor alleles. Nature 367(6460):284–7. 65. Koshiyama H, Sone T, Nakao K 1995 Vitamin D–receptor gene polymorphism and bone loss. Lancet 345(8955):990–1. 66. Eisman JA 1999 Genetics of osteoporosis. Endocr Rev 20(6):788–804. 67. Gomez Alonso C, Naves Diaz ML, Diaz-Corte C, Fernandez Martin JL, Cannata Andia JB 1998 Vitamin D receptor gene (VDR) polymorphisms: effect on bone mass, bone loss, and parathyroid hormone regulation. Nephrol Dial Transplant 13(Suppl 3):73–7. 68. Carling T, Kindmark A, Hellman P, Lundgren E, Ljunghall S, Rastad J, Akerstrom G, Melhus H 1995 Vitamin D receptor genotypes in primary hyperparathyroidism. Nat Med 1(12):1309–11. 69. Kontula K, Valimaki S, Kainulainen K, Viitanen AM, Keski-Oja J 1997 Vitamin D receptor polymorphism and treatment of psoriasis with calcipotriol. Br J Dermatol 136(6): 977–8. 70. Fernandez E, Fibla J, Betriu A, Piulats JM, Almirall J, Montoliu J 1997 Association between vitamin D receptor gene polymorphism and relative hypoparathyroidism in patients with chronic renal failure. J Am Soc Nephrol 8(10):1546–52. 71. Akiba T, Ando R, Kurihara S, Heishi M, Tazawa H, Marumo F 1997 Is the bone mass of hemodialysis patients genetically determined? Kidney Int Suppl 62:S69–71. 72. Borras M, Torregrossa V, Oliveras A, Almirall J, Ma Paz M, Betriu A, Martin M, Muray S, Fibla J, Fernandez E 2003 BB genotype of the vitamin D receptor gene polymorphism postpones parathyroidectomy in hemodialysis patients. J Nephrol 16(1):116–20.

1335 73. Nagaba Y, Heishi M, Tazawa H, Tsukamoto Y, Kobayashi Y 1998 Vitamin D receptor gene polymorphisms affect secondary hyperparathyroidism in hemodialyzed patients. Am J Kidney Dis 32(3):464–9. 74. Yokoyama K, Shigematsu T, Tsukada T, Ogura Y, Takemoto F, Hara S, Yamada A, Kawaguchi Y, Hosoya T 1998 Apa I polymorphism in the vitamin D receptor gene may affect the parathyroid response in Japanese with end-stage renal disease. Kidney Int 53(2):454–8. 75. Giannini S, D’Angelo A, Nobile M, Carraro G, Rigotti P, Silva-Netto F, Pavan S, Marchini F, Zaninotto M, Dalla Carbonare L, Sartori L, Crepaldi G 2002 The effects of vitamin D receptor polymorphism on secondary hyperparathyroidism and bone density after renal transplantation. J Bone Miner Res 17(10):1768–73. 76. Yokoyama K, Shigematsu T, Kagami S, Tsukada T, Arai T, Hara S, Yamada A, Kawaguchi Y, Hosoya T 2001 Vitamin D receptor gene polymorphism detected by digestion with Apa I influences the parathyroid response to extracellular calcium in Japanese chronic dialysis patients. Nephron 89(3):315–20. 77. Whitfield GK, Remus LS, Jurutka PW, Zitzer H, Oza AK, Dang HT, Haussler CA, Galligan MA, Thatcher ML, Encinas Dominguez C, Haussler MR 2001 Functionally relevant polymorphisms in the human nuclear vitamin D receptor gene. Mol Cell Endocrinol 177(1–2):145–9. 78. Crofts LA, Hancock MS, Morrison NA, Eisman JA 1998 Multiple promoters direct the tissue-specific expression of novel N-terminal variant human vitamin D receptor gene transcripts. Proc Natl Acad Sci USA 95(18):10529–34. 79. Yamamoto H, Miyamoto K, Li B, Taketani Y, Kitano M, Inoue Y, Morita K, Pike JW, Takeda E 1999 The caudalrelated homeodomain protein Cdx-2 regulates vitamin D receptor gene expression in the small intestine. J Bone Miner Res 14(2):240–7. 80. Arai H, Miyamoto KI, Yoshida M, Yamamoto H, Taketani Y, Morita K, Kubota M, Yoshida S, Ikeda M, Watabe F, Kanemasa Y, Takeda E 2001 The polymorphism in the caudal-related homeodomain protein Cdx-2 binding element in the human vitamin D receptor gene. J Bone Miner Res 16(7):1256–64. 81. Verroust PJ, Christensen EI 2002 Megalin and cubilin—the story of two multipurpose receptors unfolds. Nephrol Dial Transplant 17(11):1867–71. 82. Segersten U, Correa P, Hewison M, Hellman P, Dralle H, Carling T, Akerstrom G, Westin G 2002 25OHD(3)-1alphahydroxylase expression in normal and pathological parathyroid glands. J Clin Endocrinol Metab 87(6):2967–72. 83. Teramoto T, Endo K, Ikeda K, Kubodera N, Kinoshita M, Yamanaka M, Ogata E 1995 Binding of vitamin D to low-density-lipoprotein (LDL) and LDL receptor-mediated pathway into cells. Biochem Biophys Res Commun 215(1): 199–204. 84. Canaff L, Hendy GN 2002 Human calcium-sensing receptor gene. Vitamin D response elements in promoters P1 and P2 confer transcriptional responsiveness to 1,25-dihydroxyvitamin D. J Biol Chem 277(33):30337–50. 85. Li YC, Kong J, Wei M, Chen ZF, Liu SQ, Cao LP 2002 1,25Dihydroxyvitamin D(3) is a negative endocrine regulator of the renin-angiotensin system. J Clin Invest 110(2):229–38. 86. Sawaya BP, Koszewski NJ, Qi Q, Langub MC, MonierFaugere MC, Malluche HH 1997 Secondary hyperparathyroidism and vitamin D receptor binding to vitamin D response elements in rats with incipient renal failure. J Am Soc Nephrol 8(2):271–78.

1336

ADRIANA S. DUSSO, ALEX J. BROWN, AND EDUARDO A. SLATOPOLSKY

87. MacDonald PN, Ritter C, Brown AJ, Slatopolsky E 1994 Retinoic acid suppresses parathyroid hormone (PTH) secretion and PreproPTH mRNA levels in bovine parathyroid cell culture. J Clin Invest 93(2):725–30. 88. Hsu CH, Patel SR 1995 Altered vitamin D metabolism and receptor interaction with the target genes in renal failure: calcitriol receptor interaction with its target gene in renal failure. Curr Opin Nephrol Hypertens 4(4):302–6. 89. Patel SR, Ke HQ, Vanholder R, Koenig RJ, Hsu CH 1995 Inhibition of calcitriol receptor binding to vitamin D response elements by uremic toxins. J Clin Invest 96(1):50–9. 90. Sela-Brown A, Russell J, Koszewski NJ, Michalak M, Naveh-Many T, Silver J 1998 Calreticulin inhibits vitamin D’s action on the PTH gene in vitro and may prevent vitamin D’s effect in vivo in hypocalcemic rats. Mol Endocrinol 12(8):1193–200. 91. Vidal M, Ramana CV, Dusso AS 2002 Stat1-vitamin D receptor interactions antagonize 1,25-dihydroxyvitamin D transcriptional activity and enhance stat1-mediated transcription. Mol Cell Biol 22(8):2777–87. 92. Jurutka PW, Whitfield GK, Hsieh JC, Thompson PD, Haussler CA, Haussler MR 2001 Molecular nature of the vitamin D receptor and its role in regulation of gene expression. Rev Endocr Metab Disord 2(2):203–16. 93. Rachez C, Freedman LP 2001 Mediator complexes and transcription. Curr Opin Cell Biol 13(3):274–80. 94. Rachez C, Freedman LP 2000 Mechanisms of gene regulation by vitamin D(3) receptor: a network of coactivator interactions. Gene 246(1–2):9–21. 95. May P, Bock HH, Herz J 2003 Integration of endocytosis and signal transduction by lipoprotein receptors. Sci STKE 2003(176):PE12. 96. Li YC, Amling M, Pirro AE, Priemel M, Meuse J, Baron R, Delling G, Demay MB 1998 Normalization of mineral ion homeostasis by dietary means prevents hyperparathyroidism, rickets, and osteomalacia, but not alopecia in vitamin D receptor-ablated mice. Endocrinology 139(10):4391–96. 97. Van Cromphaut SJ, Dewerchin M, Hoenderop JG, Stockmans I, Van Herck E, Kato S, Bindels RJ, Collen D, Carmeliet P, Bouillon R, Carmeliet G 2001 Duodenal calcium absorption in vitamin D receptor knockout mice: functional and molecular aspects. Proc Natl Acad Sci USA 98(23):13324–29. 98. Bouillon R, Van Cromphaut S, Carmeliet G 2003 Intestinal calcium absorption: Molecular vitamin D–mediated mechanisms. J Cell Biochem 88(2):332–9. 99. van Abel M, Hoenderop JG, van der Kemp AW, van Leeuwen JP, Bindels RJ 2003 Regulation of the epithelial Ca2+ channels in small intestine as studied by quantitative mRNA detection. Am J Physiol Gastrointest Liver Physiol 285(1):G78–85. 100. Brown AJ, Finch J, Slatopolsky E 2002 Differential effects of 19-nor-1,25-dihydroxyvitamin D(2) and 1,25-dihydroxyvitamin D(3) on intestinal calcium and phosphate transport. J Lab Clin Med 139(5):279–84. 101. Ogg CS 1968 The intestinal absorption of 47Ca by patients in chronic renal failure. Clin Sci 34(3):467–71. 102. Recker RR, Saville PD 1971 Calcium absorption in renal failure: its relationship to blood urea nitrogen, dietary calcium intake, time on dialysis, and other variables. J Lab Clin Med 78(3):380–8. 103. Wong RG, Norman AW, Reddy CR, Coburn JW 1972 Biologic effects of 1,25-dihydroxycholecalciferol (a highly active vitamin D metabolite) in acutely uremic rats. J Clin Invest 51(5):1287–91.

104. Hartenbower DL, Coburn JW, Reddy CR, Norman AW 1974 Calciferol metabolism and intestinal calcium transport in the chick with reduced renal function. J Lab Clin Med 83(1):38–45. 105. Brickman AS, Coburn JW, Norman AW 1972 Action of 1,25-dihydroxycholecalciferol, a potent, kidney-produced metabolite of vitamin D, in uremic man. N Engl J Med 287(18):891–95. 106. Walling MW, Kimberg DV, Wasserman RH, Feinberg RR 1976 Duodenal active transport of calcium and phosphate in vitamin D–deficient rats: effects of nephrectomy, cestrum diurnum, and 1alpha,25-dihydroxyvitamin D3. Endocrinology 98(5):1130–34. 107. Juttmann JR, Hagenouw-Taal JC, Lameyer LD, Ruis AM, Birkenhager JC 1978 Intestinal calcium absorption, serum phosphate, and parathyroid hormone in patients with chronic renal failure and osteodystrophy before and during hemodialysis. Calcif Tissue Res 26(2):119–26. 108. Chanard JM, Drueke T, Zingraff J, Man NK, Russo-Marie F, Funck-Brentano JL 1976 Effects of haemodialysis on fractional intestinal absorption of calcium in uremia. Eur J Clin Invest 6(3):261–4. 109. Slatopolsky E, Brown A, Dusso A 2001 Role of phosphorus in the pathogenesis of secondary hyperparathyroidism. Am J Kidney Dis 37(1 Suppl 2):S54–7. 110. Dusso AS, Pavlopoulos T, Naumovich L, Lu Y, Finch J, Brown AJ, Morrissey J, Slatopolsky E 2001 p21(WAF1) and transforming growth factor-alpha mediate dietary phosphate regulation of parathyroid cell growth. Kidney Int 59(3):855–65. 111. Gogusev J, Duchambon P, Stoermann-Chopard C, Giovannini M, Sarfati E, Drueke TB 1996 De novo expression of transforming growth-factor alpha in parathyroid gland tissue of patients with primary or secondary uremic hyperparathyroidism. Nephrol Dial Transplant 11(11):2155–62. 112. Cozzolino M, Lu Y, Finch J, Slatopolsky E, Dusso AS 2001 p21WAF1 and TGF-alpha mediate parathyroid growth arrest by vitamin D and high calcium. Kidney Int 60(6):2109–17. 113. Cozzolino MLY, Slatopolsky E, Dusso A 2002 Specific inhibiton of EGFR-signaling prevents high phosphorusinduced parathyroid hyperplasia in renal failure. J Am Soc Nephrol 13:193A. 114. Cordero JB, Cozzolino M, Lu Y, Vidal M, Slatopolsky E, Stahl PD, Barbieri MA, Dusso A 2002 1,25-Dihydroxyvitamin D down-regulates cell membrane growth—and nuclear growth—promoting signals by the epidermal growth factor receptor. J Biol Chem 277(41):38965–71. 115. Tominaga Y, Tsuzuki T, Uchida K, Haba T, Otsuka S, Ichimori T, Yamada K, Numano M, Tanaka Y, Takagi H 1999 Expression of PRAD1/cyclin D1, retinoblastoma gene products, and Ki67 in parathyroid hyperplasia caused by chronic renal failure versus primary adenoma. Kidney Int 55(4):1375–83. 116. Tokumoto M, Tsuruya K, Fukuda K, Kanai H, Kuroki S, Hirakata H 2002 Reduced p21, p27, and vitamin D receptor in the nodular hyperplasia in patients with advanced secondary hyperparathyroidism. Kidney Int 62(4):1196–207. 117. Liu M, Lee MH, Cohen M, Bommakanti M, Freedman LP 1996 Transcriptional activation of the Cdk inhibitor p21 by vitamin D3 leads to the induced differentiation of the myelomonocytic cell line U937. Genes Dev 10(2):142–53. 118. Kremer R, Bolivar I, Goltzman D, Hendy GN 1989 Influence of calcium and 1,25-dihydroxycholecalciferol on proliferation and proto-oncogene expression in primary cultures of bovine parathyroid cells. Endocrinology 125(2):935–41.

CHAPTER 76 Vitamin D and Renal Failure

119. Brown AJ, Zhong M, Finch J, Ritter C, McCracken R, Morrissey J, Slatopolsky E 1996 Rat calcium-sensing receptor is regulated by vitamin D but not by calcium. Am J Physiol 270(3 Pt 2):F454–60. 120. Kifor O, Moore FD Jr Wang P, Goldstein M, Vassilev P, Kifor I, Hebert SC, Brown EM 1996 Reduced immunostaining for the extracellular Ca2+-sensing receptor in primary and uremic secondary hyperparathyroidism. J Clin Endocrinol Metab 81(4):1598–06. 121. Delmez JA, Tindira C, Grooms P, Dusso A, Windus DW, Slatopolsky E 1989 Parathyroid hormone suppression by intravenous 1,25-dihydroxyvitamin D. A role for increased sensitivity to calcium. J Clin Invest 83(4):1349–55. 122. Shiraishi K, Tsuchida M, Wada T, Yoshihiro S, Takai K, Suga A, Kaneda Y, Naito K 2001 22-Oxacalcitriol upregulates p21(WAF1/Cip1) in human parathyroid glands. A preliminary report. Am J Nephrol 21(6):507–11. 123. Malluche HH, Ritz E, Lange HP, Kutschera L, Hodgson M, Seiffert U, Schoeppe W 1976 Bone histology in incipient and advanced renal failure. Kidney Int 9(4):355–62. 124. Friedman PA, Gesek FA 1993 Vitamin D3 accelerates PTHdependent calcium transport in distal convoluted tubule cells. Am J Physiol 265(2 Pt 2):F300–08. 125. Hoenderop JG, Muller D, Van Der Kemp AW, Hartog A, Suzuki M, Ishibashi K, Imai M, Sweep F, Willems PH, Van Os CH, Bindels RJ 2001 Calcitriol controls the epithelial calcium channel in kidney. J Am Soc Nephrol 12(7):1342–9. 126. Li YC 2003 Vitamin D regulation of the renin-angiotensin system. J Cell Biochem 88(2):327–31. 127. Chertow BS, Sivitz WI, Baranetsky NG, Clark SA, Waite A, Deluca HF 1983 Cellular mechanisms of insulin release: the effects of vitamin D deficiency and repletion on rat insulin secretion. Endocrinology 113(4):1511–8. 128. Christakos S, Friedlander EJ, Frandsen BR, Norman AW 1979 Studies on the mode of action of calciferol. XIII. Development of a radioimmunoassay for vitamin D–dependent chick intestinal calcium-binding protein and tissue distribution. Endocrinology 104(5):1495–503. 129. Lowrie EG, Soeldner JS, Hampers CL, Merrill JP 1970 Glucose metabolism and insulin secretion in uremic, prediabetic, and normal subjects. J Lab Clin Med 76(4):603–15. 130. Allegra V, Mengozzi G, Martimbianco L, Vasile A 1990 Glucose-induced insulin secretion in uremia: effects of aminophylline infusion and glucose loads. Kidney Int 38(6):1146–50. 131. Alvestrand A, Mujagic M, Wajngot A, Efendic S 1989 Glucose intolerance in uremic patients: the relative contributions of impaired beta-cell function and insulin resistance. Clin Nephrol 31(4):175–83. 132. Quesada JM, Martin-Malo A, Santiago J, Hervas F, Martinez ME, Castillo D, Barrio V, Aljama P 1990 Effect of calcitriol on insulin secretion in uremia. Nephrol Dial Transplant 5(12):1013–7. 133. Allegra V, Luisetto G, Mengozzi G, Martimbianco L, Vasile A 1994 Glucose-induced insulin secretion in uremia: role of 1 alpha,25(HO)2-vitamin D3. Nephron 68(1):41–7. 134. Goldblum SE, Reed WP 1980 Host defenses and immunologic alterations associated with chronic hemodialysis. Ann Intern Med 93(4):597–613. 135. Reichel H, Koeffler HP, Norman AW 1989 The role of the vitamin D endocrine system in health and disease. N Engl J Med 320(15):980–91. 136. Bar-Shavit Z, Noff D, Edelstein S, Meyer M, Shibolet S, Goldman R 1981 1,25-dihydroxyvitamin D3 and the regulation of macrophage function. Calcif Tissue Int 33(6):673–6.

1337 137. Park K, Ha SK, Han DS 1988 Studies on lymphocyte subpopulations and cell-mediated immunity in patients with chronic renal failure. Yonsei Med J 29(2):109–16. 138. Lin CY, Huang TP 1988 Serial cell-mediated immunological changes in terminal uremic patients on continuous ambulatory peritoneal dialysis therapy. Am J Nephrol 8(5):355–62. 139. Venezio FR, Kozeny GA, DiVincenzo CA, Hano JE 1988 Effects of 1,25 dihydroxyvitamin D3 on leukocyte function in patients receiving chronic hemodialysis. J Infect Dis 158(5):1102–5. 140. Bargman JM, Silverman ED, Klein MH 1989 Effect of 1,25dihydroxyvitamin D3 in vivo on circulating T lymphocytes. Miner Electrolyte Metab 15(6):359–64. 141. Quesada JM, Serrano I, Borrego F, Martin A, Pena J, Solana R 1995 Calcitriol effect on natural killer cells from hemodialyzed and normal subjects. Calcif Tissue Int 56(2):113–17. 142. Tabata T, Suzuki R, Kikunami K, Matsushita Y, Inoue T, Okamoto T, Miki T, Nishizawa Y, Morii H 1986 The effect of 1 alpha-hydroxyvitamin D3 on cell-mediated immunity in hemodialyzed patients. J Clin Endocrinol Metab 63(5): 1218–21. 143. Lemire JM, Adams JS, Kermani-Arab V, Bakke AC, Sakai R, Jordan SC 1985 1,25-Dihydroxyvitamin D3 suppresses human T helper/inducer lymphocyte activity in vitro. J Immunol 134(5):3032–5. 144. Tabata T, Shoji T, Kikunami K, Matsushita Y, Inoue T, Tanaka S, Hino M, Miki T, Nishizawa Y, Morii H 1988 In vivo effect of 1 alpha-hydroxyvitamin D3 on interleukin-2 production in hemodialysis patients. Nephron 50(4):295–298. 145. Boland R 1986 Role of vitamin D in skeletal muscle function. Endocr Rev 7(4):434–48. 146. Wu J, Garami M, Cheng T, Gardner DG 1996 1,25(OH)2 vitamin D3 and retinoic acid antagonize endothelin-stimulated hypertrophy of neonatal rat cardiac myocytes. J Clin Invest 97(7):1577–88. 147. Wu J, Garami M, Cao L, Li Q, Gardner DG 1995 1,25(OH)2D3 suppresses expression and secretion of atrial natriuretic peptide from cardiac myocytes. Am J Physiol 268(6 Pt 1):E1108–13. 148. Selles J, Massheimer V, Santillan G, Marinissen MJ, Boland R 1997 Effects of calcitriol and its analogues, calcipotriol (MC 903) and 20-epi-1alpha,25-dihydroxyvitamin D3 (MC 1288), on calcium influx and DNA synthesis in cultured muscle cells. Biochem Pharmacol 53(12):1807–1014. 149. Brown J, Bianco JI, McGrath JJ, Eyles DW 2003 1,25-dihydroxyvitamin D3 induces nerve growth factor, promotes neurite outgrowth, and inhibits mitosis in embryonic rat hippocampal neurons. Neurosci Lett 343(2):139–43. 150. Cai Q, Tapper DN, Gilmour RF Jr deTalamoni N, Aloia RC, Wasserman RH 1994 Modulation of the excitability of avian peripheral nerves by vitamin D: relation to calbindin-D28k, calcium status, and lipid composition. Cell Calcium 15(5):401–10. 151. Cantorna MT, Hayes CE, DeLuca HF 1996 1,25-Dihydroxyvitamin D3 reversibly blocks the progression of relapsing encephalomyelitis, a model of multiple sclerosis. Proc Natl Acad Sci U S A 93(15):7861–4. 152. Sanchez B, Lopez-Martin E, Segura C, Labandeira-Garcia JL, Perez-Fernandez R 2002 1,25-Dihydroxyvitamin D(3) increases striatal GDNF mRNA and protein expression in adult rats. Brain Res Mol Brain Res 108(1–2):143–6. 153. Baker LR, Abrams L, Roe CJ, Faugere MC, Fanti P, Subayti Y, Malluche HH 1989 1,25(OH)2D3 administration in moderate renal failure: a prospective double-blind trial. Kidney Int 35(2):661–9.

1338

ADRIANA S. DUSSO, ALEX J. BROWN, AND EDUARDO A. SLATOPOLSKY

154. Nordal KP, Dahl E, Halse J, Attramadal A, Flatmark A 1995 Long-term, low-dose calcitriol treatment in predialysis chronic renal failure: can it prevent hyperparathyroid bone disease? Nephrol Dial Transplant 10(2):203–6. 155. Hamdy NA, Kanis JA, Beneton MN, Brown CB, Juttmann JR, Jordans JG, Josse S, Meyrier A, Lins RL, Fairey IT 1995 Effect of alfacalcidol on natural course of renal bone disease in mild to moderate renal failure. Bmj 310(6976):358–63. 156. Maung HM, Elangovan L, Frazao JM, Bower JD, Kelley BJ, Acchiardo SR, Rodriguez HJ, Norris KC, Sigala JF, Rutkowski M, Robertson JA, Goodman WG, Levine BS, Chesney RW, Mazess RB, Kyllo DM, Douglass LL, Bishop CW, Coburn JW 2001 Efficacy and side effects of intermittent intravenous and oral doxercalciferol (1alphahydroxyvitamin D(2)) in dialysis patients with secondary hyperparathyroidism: a sequential comparison. Am J Kidney Dis 37(3):532–43. 157. Slatopolsky E, Weerts C, Thielan J, Horst R, Harter H, Martin KJ 1984 Marked suppression of secondary hyperparathyroidism by intravenous administration of 1,25-dihydroxycholecalciferol in uremic patients. J Clin Invest 74(6):2136–43. 158. Andress DL, Norris KC, Coburn JW, Slatopolsky EA, Sherrard DJ 1989 Intravenous calcitriol in the treatment of refractory osteitis fibrosa of chronic renal failure. N Engl J Med 321(5):274–9. 159. Dunlay R, Rodriguez M, Felsenfeld AJ, Llach F 1989 Direct inhibitory effect of calcitriol on parathyroid function (sigmoidal curve) in dialysis. Kidney Int 36(6):1093–8. 160. Tsukamoto Y, Nomura M, Takahashi Y, Takagi Y, Yoshida A, Nagaoka T, Togashi K, Kikawada R, Marumo F 1991 The “oral 1,25-dihydroxyvitamin D3 pulse therapy” in hemodialysis patients with severe secondary hyperparathyroidism. Nephron 57(1):23–8. 161. Tsukamoto Y, Mariyo R, Nomura Y, Sato N, Faugere MC, Malluche HH 1993 Long-term effect of oral calcitriol pulse therapy on bone in hemodialysis patients. Bone 14(3):421–5. 162. Quarles LD, Yohay DA, Carroll BA, Spritzer CE, Minda SA, Bartholomay D, Lobaugh BA 1994 Prospective trial of pulse oral versus intravenous calcitriol treatment of hyperparathyroidism in ESRD. Kidney Int 45(6):1710–21. 163. Cannella G, Bonucci E, Rolla D, Ballanti P, Moriero E, De Grandi R, Augeri C, Claudiani F, Di Maio G 1994 Evidence of healing of secondary hyperparathyroidism in chronically hemodialyzed uremic patients treated with longterm intravenous calcitriol. Kidney Int 46(4):1124–32. 164. Llach F, Hervas J, Cerezo S 1995 The importance of dosing intravenous calcitriol in dialysis patients with severe hyperparathyroidism. Am J Kidney Dis 26(5):845–51. 165. Bacchini G, Fabrizi F, Pontoriero G, Marcelli D, Di Filippo S, Locatelli F 1997 “Pulse oral” versus intravenous calcitriol therapy in chronic hemodialysis patients. A prospective and randomized study. Nephron 77(3):267–72. 166. Indridason OS, Quarles LD 2000 Comparison of treatments for mild secondary hyperparathyroidism in hemodialysis patients. Durham Renal Osteodystrophy Study Group. Kidney Int 57(1):282–92.

167. Fischer ER, Harris DC 1993 Comparison of intermittent oral and intravenous calcitriol in hemodialysis patients with secondary hyperparathyroidism. Clin Nephrol 40(4): 216–20. 168. Liou HH, Chiang SS, Huang TP, Shieh SD, Akmal M 1994 Comparative effect of oral or intravenous calcitriol on secondary hyperparathyroidism in chronic hemodialysis patients. Miner Electrolyte Metab 20(3):97–102. 169. Salusky IB, Goodman WG, Horst R, Segre GV, Kim L, Norris KC, Adams JS, Holloway M, Fine RN, Coburn JW 1990 Pharmacokinetics of calcitriol in continuous ambulatory and cycling peritoneal dialysis patients. Am J Kidney Dis 16(2):126–32. 170. Slatopolsky E, Brown AJ 2002 Vitamin D analogs for the treatment of secondary hyperparathyroidism. Blood Purif 20(1):109–12. 171. Dusso AS, Negrea L, Gunawardhana S, Lopez-Hilker S, Finch J, Mori T, Nishii Y, Slatopolsky E, Brown AJ 1991 On the mechanisms for the selective action of vitamin D analogs. Endocrinology 128(4):1687–92. 172. Tanaka Y, DeLuca HF, Kobayashi Y, Ikekawa N 1984 26,26,26,27,27,27-hexafluoro-1,25-dihydroxyvitamin D3: a highly potent, long-lasting analog of 1,25-dihydroxyvitamin D3. Arch Biochem Biophys 229(1):348–54. 173. Akiba T, Marumo F, Owada A, Kurihara S, Inoue A, Chida Y, Ando R, Shinoda T, Ishida Y, Ohashi Y 1998 Controlled trial of falecalcitriol versus alfacalcidol in suppression of parathyroid hormone in hemodialysis patients with secondary hyperparathyroidism. Am J Kidney Dis 32(2):238–46. 174. Finch JL, Brown AJ, Slatopolsky E 1999 Differential effects of 1,25-dihydroxyvitamin D3 and 19-nor-1,25-dihydroxyvitamin D2 on calcium and phosphorus resorption in bone. J Am Soc Nephrol 10(5):980–5. 175. Llach F, Yudd M 2001 Paricalcitol in dialysis patients with calcitriol-resistant secondary hyperparathyroidism. Am J Kidney Dis 38(5 Suppl 5):S45–50. 176. Sprague SM, Lerma E, McCormmick D, Abraham M, Batlle D 2001 Suppression of parathyroid hormone secretion in hemodialysis patients: comparison of paricalcitol with calcitriol. Am J Kidney Dis 38(5 Suppl 5):S51–56. 177. Martin KJ, Gonzalez EA, Gellens ME, Hamm LL, Abboud H, Lindberg J 1998 Therapy of secondary hyperparathyroidism with 19-nor-1alpha,25-dihydroxyvitamin D2. Am J Kidney Dis 32(2 Suppl 2):S61–6. 178. Coyne DW, Grieff M, Ahya SN, Giles K, Norwood K, Slatopolsky E 2002 Differential effects of acute administration of 19-nor-1,25-dihydroxyvitamin D2 and 1,25-dihydroxyvitamin D3 on serum calcium and phosphorus in hemodialysis patients. Am J Kidney Dis 40(6):1283–8. 179. Teng M, Wolf M, Lowrie E, Ofsthun N, Lazarus JM, Thadhani R 2003 Survival of patients undergoing hemodialysis with paricalcitol or calcitriol therapy. N Engl J Med 349(5):446–56. 180. Slatopolsky E, Cozzolino M, Finch JL 2002 Differential effects of 19-nor-1,25-(OH)(2)D(2) and 1alpha-hydroxyvitamin D(2) on calcium and phosphorus in normal and uremic rats. Kidney Int 62(4):1277–84.

CHAPTER 77

Idiopathic Hypercalciuria and Nephrolithiasis MURRAY J. FAVUS FREDRIC L. COE

Section of Endocrinology, The University of Chicago Pritzker School of Medicine, Chicago, Illinois Nephrology Section, The University of Chicago Pritzker School of Medicine, Chicago, Illinois

I. Introduction II. Idiopathic Hypercalciuria III. Genetic Hypercalciuric Rats IV. Current View of Human Genetic Hypercalciuria

V. Therapeutics of Idiopathic Hypercalciuria and Effects on Calcium Metabolism VI. Risk of Stone Formation Using Vitamin D Analogs References

I. INTRODUCTION

administration of small doses of 1,25(OH)2D3 to healthy volunteers. An animal model of genetic hypercalciuria has been developed in Sprague-Dawley rats by breeding hypercalciuric male and female animals. The hypercalciuria in genetic hypercalciuric stone forming (GHS) rats is due to an increase in intestinal Ca absorption and bone resorption and decreased renal Ca reabsorption. Elevated vitamin D receptor (VDR) content in intestinal mucosa, renal tubules, and bone cells strongly supports the concept that hypercalciuria is a state of vitamin D receptor-mediated excess. A post-transcriptional dysregulation of VDR is suggested by decreased VDR mRNA and increased accumulation of normal VDR protein that has normal binding affinity for 1,25(OH)2D3. The nature of the genetic defect in GHS rats and in human IH that permit hypercalciuria remains unknown.

This chapter focuses on idiopathic hypercalciuria (IH) as a major cause of hypercalciuria and nephrolithiasis and the potential role of vitamin D. Less frequent causes of hypercalcemia and hypercalciuria may also promote renal stone formation and are discussed in Chapters 78 and 79. IH is found in 5 to 7% of the adult population, is responsible for 50% of calcium oxalate nephrolithiasis, and is the most common 1,25-dihydroxyvitamin D [1,25(OH)2D] excess state. Idiopathic hypercalciuria is characterized by normocalcemia in the absence of known systemic causes of hypercalciuria. Increased intestinal calcium (Ca) absorption is almost always increased, and serum 1,25(OH)2D levels are elevated in one-third to one-half of patients. Serum parathyroid hormone (PTH) levels are elevated in less than 5%. The pathogenesis of IH is unknown but several models have been offered from observations in patients including: a primary increase in intestinal Ca absorption; a primary overproduction of 1,25(OH)2D; and a primary renal tubular Ca transport defect or “renal leak” of Ca. Evidence for each model can be found in some patients with IH, suggesting the disorder may be heterogeneous. There is also evidence that IH is a state of 1,25(OH)2D excess. As already mentioned one-third to one-half of IH patients have elevated serum 1,25(OH)2D levels. The remaining 50% with normal serum 1,25(OH)2D levels cannot be distinguished from those with elevated levels because intestinal Ca absorption is just as high, and negative Ca balance may develop during low Ca intake. Of interest is the observation that all of the changes in Ca metabolism characteristic of IH can be induced by the VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

II. IDIOPATHIC HYPERCALCIURIA A. Overview Hypercalciuria is common among patients with Ca oxalate nephrolithiasis and is thought to contribute to stone formation by increasing the state of urine supersaturation with respect to Ca and oxalate. Flocks [1] first commented on the frequency of hypercalciuria among patients with Ca stones; however, it was not until the mid-1950s that Albright and Henneman [2,3] defined the condition of IH as hypercalciuria with normal serum Ca, no systemic illness, and no clinical skeletal disease. The definition of hypercalciuria is arbitrary and based on the distribution of urine Ca excretion values among unselected populations of healthy men Copyright © 2005, Elsevier, Inc. All rights reserved.

1340

MURRAY J. FAVUS AND FREDRIC L. COE

TABLE I Causes of Normocalcemic Hypercalciuria

for either sex [6]. Using these definitions, hypercalciuria is found in about 50% of calcium oxalate stone formers [6,7] and is the most common cause of normocalcemic hypercalciuric stone formation [7]. The diagnosis of IH requires the exclusion of the known causes of normocalcemic hypercalciuria (see Table I). Surveys of stone formers attending kidney stone clinics report a high proportion with kidney stone formation among first-degree relatives [8,9]. A genetic basis of IH was further suggested by subsequent surveys [10–12] that revealed a strong familial occurrence of IH with high rates of vertical and horizontal penetrance (see Fig. 1) consistent with an autosomal dominant mode of inheritance. IH also occurs in children with the same frequency of occurrence as in adults [13]. That hypercalciuria can have a genetic origin has been clearly demonstrated by breeding experiments in which the offspring of spontaneously hypercalciuric male and female Sprague-Dawley rats are intensely hypercalciuric [14–16]. Other human hypercalciuric genetic disorders have been described, but they differ from IH in

Paget’s disease Sarcoidosis Hyperthyroidism Renal tubular acidosis Cushing’s syndrome Immobilization Malignant tumor Furosemide administration

and women in Western countries [4,5]. The distribution of urine Ca forms a continuum that is clustered about a mean with a long tail of higher values. IH patients are those whose urine Ca exceeds the arbitrary upper limit of normal, which is most commonly defined as greater than 300 mg/24 hr for men, greater than 250 mg/24 hr for women, or greater than 4 mg/kg body weight or 140 mg Ca per gram urine creatinine

Family 3

Family 1 S S

S

S

S

S

S

S

S

S

S

∗ ∗

∗

S

∗

∗ Family 4

Family 2

Family 5 S

S

S

S

S

S S

∗ Family 6

∗

∗

S

∗

Family 7

∗

Family 8 S

S

∗

Family 9 S

S ∗

∗

∗

∗

∗ ∗

FIGURE 1

S

S ∗

S

S

∗

∗

Family pedigrees of nine probands with idiopathic hypercalciuria. Solid circles and solid squares are females and males with hypercalciuria; S is stone formation; * indicates children (younger than age 20). Arrows indicate probands from each family. Dashed symbols are relatives who were not studied. Hypercalciuria occurred in 11 of 24 siblings, 7 of 16 offspring, and 1 of 3 parents of probands. Reprinted by permission of The New England Journal of Medicine (Coe FL, Parks JH, Moore ES. Familial idiopathic hypercalciuria N Engl J Med 300:337–340). Copyright 1979, Massachusetts Medical Society.

1341

CHAPTER 77 Idiopathic Hypercalciuria and Nephrolithiasis

B. Pathogenesis of Human Idiopathic Hypercalciuria 1. RENAL HISTOPATHOLOGY IN CALCIUM OXALATE NEPHROLITHIASIS

Interstitial crystal deposition at or near the tips of papillae are found in 100% of kidneys of Ca oxalate stone formers who have IH and no systemic cause of hypercalciuria or other cause of stone formation (Table I). Less frequently (43%), nonstone formers may have such papillary depositions [20]. These lesions first described by Randall [21] have recently been found to be composed of calcium phosphate (apatite) and contain no Ca oxalate [22]. The plaques originate in the basement membrane of the thin loops of Henle and spread from there through the interstitium to just beneath the urothelium. There is no Ca phosphate or Ca oxalate crystal depositions within the renal tubule lumen. Rather, the Ca phosphate plaque may serve as a site of heterogeneous nucleation of Ca oxalate crystals that subsequently grow and form Ca oxalate kidney stones [22,23]. The role of hypercalciuria in the development of the Ca phosphate interstitial lesions of Randall’s plaques remains unknown; however, the Ca phosphate plaques are rather specific for Ca oxalate stone formers, as the interstitial lesions are absent in intestinal bypass patients who form Ca oxalate stones [22]. 2. INCREASED INTESTINAL CALCIUM ABSORPTION

Normally, the quantity of Ca absorbed is determined by dietary Ca intake and the efficiency of intestinal Ca absorption [24]. Absorption of Ca across the intestine is the sum of two transepithelial transport processes: a nonsaturable paracellular pathway and a saturable, cellular active transport system [25,26] (also see Chapters 24, 25). Absorption via the paracellular path is diffusional and driven by the lumen-to-blood Ca gradient [24]. The cellular pathway is vitamin D–dependent and is

regulated by the ambient concentration of 1,25(OH)2D. Thus, intraluminal Ca concentration and tissue 1,25(OH)2D levels are the driving forces for Ca translocation via the paracellular and cellular pathways, respectively. Increased intestinal Ca absorption has been found in most patients with IH [27–35]. Using either a single oral dose of Ca isotope to measure fecal isotope excretion or double Ca isotope administration in which the intravenous dose adjusts for isotope distribution, IH patients were shown to have an increase in the Ca absorptive flux (Fig. 2). External Ca balance studies conducted while IH patients and normal nonstone formers ingested diets containing comparable amounts of Ca show net intestinal Ca absorption rates to be greater in IH patients [36]. Biopsies of proximal intestine obtained following oral Ca isotopic administration reveal increased mucosal accumulation of isotope compared to normocalciuric nonstoneformers [37]. Thus, by all techniques used, IH is characterized by increased intestinal Ca absorption. 3. ELEVATED 1,25(OH)2D

Kaplan and colleagues [33] first reported elevated serum levels of 1,25(OH)2D in a group of patients with IH. Subsequently, others have confirmed that, on average, serum 1,25(OH)2D levels are higher in IH (Fig. 3). Increased in vivo conversion of tritiated 25-hydroxyvitamin D3 (3H-25OHD3) to 3H-1,25(OH)2D3 with normal metabolic clearance [38] in a group of IH patients with elevated serum 1,25(OH)2D levels indicate that the increase in serum 1,25(OH)2D levels in some IH patients is the result of increased production. Of note is the considerable overlap of serum 1,25(OH)2D levels between IH patients and nonstone formers (Fig. 3). Normal IH

100 % Calcium absorbed

having either a renal phosphate leak [17] that may lead to rickets (Chapter 69), renal tubular acidosis [18], or X-linked recessive stone formation with early renal failure [19]. Idiopathic hypercalciuria is a common disorder that affects 5 to 7% of otherwise healthy men and women [4]. If 50% of stone formers have IH [6,7], and the frequency of stone disease among adults is 0.5%, then 80 to 90% of IH is asymptomatic and never associated with stone formation. The increased frequency of osteopenia (see Section B.4) suggests that hypercalciuria may be an important pathogenetic factor for development of low bone mass even among those who do not form stones.

80 60 40 20 0 Birge

Wills

Pak

Kaplan

Shen

FIGURE 2 Intestinal Ca absorption in healthy volunteers and patients with IH. Absorption rates are expressed as percentages of dietary Ca absorbed as calculated from the appearance of Ca isotopes in blood or fecal collections. Values are means (horizontal bar) ± 2 standard deviations. Names indicate references: Birge [28]; Wills [29]; Pak [32]; Kaplan [33]; and Shen [34].

1342

MURRAY J. FAVUS AND FREDRIC L. COE

Normal

Serum 1,25(OH)2D3 pg/ml

100

IH

80

60

40

20

Shen Kaplan

Gray Coe Insogna VanDenBerg Breslau

FIGURE 3 Plot of means ± 2 SD of serum 1,25(OH)2D in IH patients and nonstone formers. Horizontal bar is mean of group. Names indicate references: Kaplan [33]; Shen [34]; Insogna [38]; Coe [40]; Gray [42]; Van Den Berg [63]; and Breslau [64].

Thus, for about 50% of IH patients, increased intestinal Ca transport may be caused by increased circulating 1,25(OH)2D. For the remainder, other mechanisms of increased Ca absorption must be considered. The mechanism whereby 1,25(OH)2D production is increased in IH is unknown. The major regulators of renal proximal tubule mitochondrial 25-hydroxyvitamin D 1-hydroxylase (1-hydroxylase) activity include PTH, phosphate depletion, and insulin-like growth factor-I (IGF-I) (see Chapter 5). However, only 5% of IH patients have elevated circulating PTH levels [32,39], and urinary cAMP levels, a surrogate measure of PTH, are also normal in most patients [32,40,41]. Mild hypophosphatemia with reduced renal tubular phosphate reabsorption has been described in as many as one-third of IH patients [33,34,39,42]. A strong inverse association between serum 1,25(OH)2D levels and renal tubular phosphate reabsorption has been reported [34,41]. As elevated PTH or hypophosphatemia accompany elevated serum 1,25(OH)2D in only a minority of patients; the cause of increased serum 1,25(OH)2D in most patients with IH remains unknown. Detailed studies of IGF-I have not been performed. The recent description of sequences of mutations in the q23.3-q24 region of the first chromosome in three kindred with absorptive IH [43] involves a region containing a gene that is analogous with the rat soluble adenylate cyclase gene. This first description of specific

base pair substitutions suggests the possibility of a gene defect associated with IH that may involve altered receptor signaling. Whether this mutation alters functions related to the regulation of the 1-hydroxylase remains to be determined. However, caution has been expressed in accepting this report as conclusively demonstrating that the substitutions or a mutation of this gene causes IH [44]. Serum 1,25(OH)2D values in normals and IH patients overlap extensively in each series reported (Fig. 3). Kaplan et al. [33] found that in patients with absorptive IH [defined as normal fasting urine Ca and normal or elevated serum 1,25(OH)2D, intestinal Ca absorption measured by fecal excretion of orally administered 47Ca was increased out of proportion to the simultaneously measured serum 1,25(OH)2D concentration (Fig. 4B). In contrast, a strong positive correlation between intestinal Ca absorption and serum 1,25(OH)2D is found in normal volunteers, normocalciuric stone formers, patients with primary hyperparathyroidism, and IH patients with fasting hypercalciuria (Fig. 4A). The high intestinal Ca absorption rates with normal or elevated serum 1,25(OH)2D levels suggest that the pathogenesis of IH is heterogeneous, with at least one phenotype resulting from 1,25(OH)2D overproduction. 4. DECREASED RENAL CALCIUM REABSORPTION

A defect in the tubular reabsorption of Ca, a so-called renal leak of Ca, has been postulated as a cause of hypercalciuria in IH. Two reports [45,46] found a greater fraction of filtered Ca excreted in the urine of IH patients compared to nonstone formers. The values were calculated from inulin clearance or creatinine clearance and used blood ionized Ca as an estimate of ultrafilterable Ca. Although urinary sodium (Na) excretion is a major determinant of Ca excretion in normal and IH patients, there is no evidence that patients overingest or overexcrete Na. Hydrochlorothiazide and acetazolamide increase urine Ca, Na, and magnesium (Mg) excretion in IH compared to normals [47], suggesting a generalized defect in proximal tubule electrolyte and water transport in IH patients. The basis for the abnormal renal transport is not known, but increased activity of the erythrocyte plasma membrane Ca2+, Mg2+-ATPase in IH patients and correlation of enzyme activity with urine Ca excretion in families with IH [48] suggests a more widespread genetic defect in monovalent and divalent ion transport. 5. LOW BONE MASS

Abnormal skeletal metabolism in IH has been demonstrated by low bone mineral density of the distal radius [49,50] and lumbar spine [51–53] and by lower skeletal Ca content by neutron activation analysis [54].

1343

CHAPTER 77 Idiopathic Hypercalciuria and Nephrolithiasis

1.0

A Fractional Ca absorption (α)

Fractional Ca absorption (α)

1.0 0.8

0.6

0.4 PHPT RH NN Control

0.2

0

20

40 60 80 100 Plasma 1α,25(OH)2D pg/ml

120

B

0.8

0.6

0.4

0.2

0 20

100 40 60 80 Plasma 1α,25(OH)2D pg/ml

120

FIGURE 4 Relationship of calcium absorption to 1,25(OH)2D levels. (A) Fractional intestinal absorption of oral 47Ca versus serum 1,25(OH)2D level in normal controls (open circles), normocalciuric stone formers (NN, filled circles), IH stone formers who have fasting hypercalciuria and elevated PTH (RH, filled squares), and patients with primary hyperparathyroidism (PHPT, open squares). (B) Fractional calcium absorption of IH patients with absorptive hypercalciuria (normal fasting urine Ca) superimposed on the 95% confidence limits for the relationship in controls. Reproduced from The Journal of Clinical Investigation, 1977, Vol. 59, pp. 756–760 [33], by copyright permission of The American Society for Clinical Investigation.

Reports differ as to possible pathogenesis, with low bone density found only in those with renal leak hypercalciuria in one study [50], and low bone density in those with absorptive hypercalciuria in another study [52]. Information on bone dynamics is limited to one early study in which 47Ca labeling was interpreted as increased bone turnover, with bone resorption and formation both increased [55]. Two studies of bone histology showed reduced bone apposition rate, delayed mineralization of osteoid seams, and prolonged mineralization lag time and formation period [56,57]. These observations suggest defective mineralization, which may be caused by hypophosphatemia in some patients. The observations are also consistent with a defect in osteoblastic function. Measurements of biochemical markers of bone turnover reveal increased urine hydroxyproline excretion in unselected IH patients [58] and increased serum osteocalcin in IH patients with renal but not absorptive hypercalciuria [59]. Whether the low bone density is a result of the lifelong hypercalciuria, habitual low Ca intake, or a genetic defect in osteoblast function independent of urine Ca excretion remains to be determined. In a study of 59 subjects from 11 families with at least one member a hypercalciuric Ca oxalate stoneformer [60], lumbar spine and femoral neck bone density Z scores varied inversely with urine Ca and urine ammonium in the stoneformers but not in the nonstone formers. There were no correlations of Z score for bone turnover markers or serum 1,25(OH)2D levels. Ca consumption was lower in stoneformers, suggesting that the admonition to ingest a low Ca diet to avoid more stones, in fact predisposes to bone loss.

The well-documented low bone mass in IH patients is associated with increased fracture risk [61]. Reduction of urine Ca during thiazide therapy has been studied in a small number of IH patients and found to be effective in improving bone mass (see Section V.B below). 6. PROPOSED PATHOGENETIC MODELS HYPERCALCIURIA

OF IDIOPATHIC

On the basis of the consistent increase in intestinal Ca absorption, normal or elevated serum 1,25(OH)2D levels, and normal or elevated fasting urinary Ca, Pak and colleagues [62] separated IH into three groups: absorptive, renal, and resorptive. In the first, primary intestinal Ca hyperabsorption (Fig. 5A) would transiently raise postprandial serum Ca above normal and increase ultrafilterable Ca. Postprandial hypercalcemia would transiently suppress PTH secretion, resulting in reduced tubular Ca reabsorption and hypercalciuria. In the second, a primary renal tubular leak of Ca (Fig. 5B) would cause hypercalciuria and a transient reduction in serum Ca. Secondary hyperparathyroidism would normalize serum Ca and increase proximal tubule 1,25(OH)2D synthesis, which would stimulate intestinal Ca absorption. PTH secretion would then decline to the extent that serum Ca is normalized. This scenario predicts that serum 1,25(OH)2D would be elevated in renal IH and normal or elevated in absorptive IH [63,64]. A third possibility is based on a primary overproduction of 1,25(OH)2D3 that increases intestinal Ca absorption and bone resorption (Fig. 5C) while PTH remains normal and fasting urine Ca excretion may be normal or elevated.

1344

MURRAY J. FAVUS AND FREDRIC L. COE

A

B

Intestinal Ca absorption

Renal tubular Ca transport Hypercalciuria

Transient

Serum Ca Serum Ca

Parathyroid hormone Serum PTH Normal Serum calcitriol

Tubular Ca reabsorption

Serum calcitriol

Intestinal Ca absorption

Hypercalciuria

C

Primary renal calcitriol overproduction

Serum calcitriol

Target tissue calcitriol/vitamin D receptor complex

Intestinal Ca absorption

Bone resorption

?

Renal tubular Ca transport

Hypercalciuria

FIGURE 5 Three proposed models of IH. (A) Absorptive IH with primary intestinal overabsorption, postprandial hypercalcemia, suppressed PTH, and normal fasting urine Ca. Serum 1,25(OH)2D is normal. (B) Primary renal tubular leak of Ca leads to a transient decrease in serum Ca and elevated PTH with secondary increases in serum 1,25(OH)2D and intestinal Ca absorption. Fasting urine Ca is elevated. (C) Primary overproduction of 1,25(OH)2D increases serum 1,25(OH)2D and stimulates intestinal Ca absorption and bone resorption. Serum PTH is normal or decreased, and fasting urine Ca is normal or elevated. 7. TESTS OF THE MODELS

Knowledge of the pathophysiology of IH is fundamental to developing rational therapy for the prevention of recurrent kidney stones. If the model of primary intestinal overabsorption were correct, then dietary Ca restriction would reduce the amount of Ca absorbed and Ca excreted in the urine without altering bone mass. If a renal leak of Ca were the primary event, or if urinary Ca originates from bone rather than diet, then restricting dietary Ca will have little effect on urinary Ca excretion, while worsening Ca balance and promoting bone loss. Testing certain predictions has assessed the accuracy of the absorptive and renal models. a. Fasting Serum PTH, Urine Ca If repeated episodes of postprandial hypercalcemia suppress PTH secretion sufficiently to cause chronic hypoparathyroidism, fasting serum PTH and urine cAMP would be low and fasting urine Ca elevated. Transient suppression of PTH

would permit normal serum PTH, urine cAMP, and fasting urine Ca. In contrast, renal IH requires increased PTH and urine cAMP and increased fasting urine Ca [65]. As existing PTH radioimmunoassays do not differentiate normal from low values, most IH patients have been found to have PTH levels in the normal range. Although normal fasting urine Ca is not unusual among IH patients, only 5% have elevated PTH and, therefore, fail to meet the criteria for renal IH [32,62]. Thus, a primary renal leak of Ca with a secondary increase in PTH cannot account for fasting hypercalciuria in a majority of patients. About 24% of patients meet the criteria of absorptive hypercalciuria by reducing urine Ca during fasting to maintain neutral Ca balance [63]. b. External Ca Balance The relationship between net intestinal Ca absorption and 24-hr urine Ca excretion calculated from 6-day balance studies is different

CHAPTER 77 Idiopathic Hypercalciuria and Nephrolithiasis

Urinary calcium, mg/day

600

500

400

300

200

100

−200

−100

0

100

200

300

400

500

600

Net intestinal calcium absorption, mg/day

FIGURE 6 Urinary Ca excretion as a function of net intestinal Ca absorption. Data are derived from 6-day external mineral balance studies. Solid lines indicate the 95% confidence limits about the mean regression line derived from the data on 195 adult nonstone formers. Individual balance studies performed on 51 patients with IH are shown as open circles. The dashed line represents equivalent rates of urinary Ca excretion and net intestinal Ca absorption (the line of identity). Normal values are from [47,58,68,69]. Values from patients are from References [47,57,71–74], and J. Lemann (personal communication, 1992). Adapted from Asplin et al. [95].

in IH patients compared to normal subjects (Fig. 6) [66–72]. In nonstone formers, urinary Ca excretion is positively correlated with net absorption, and overall Ca balance is positive when net absorption is greater than 200 mg per 24 hr (see 95% confidence limits calculated from balance studies on normal subjects in Fig. 6). Net Ca absorption tends to be greater in IH patients, and for every level of net absorption, 24-hr urine Ca excretion is higher in the patients compared to healthy subjects. In IH patients, a greater portion of absorbed Ca is excreted in the urine. In normal subjects, net Ca absorption exceeds urine Ca excretion, and balance is positive when net absorption exceeds 200 mg/24 hr. In contrast, almost 50% of the IH patients have urine Ca excretion in excess of net absorption and are in negative Ca balance, even when allowance is made for some variability in the balance data (± 50 mg). Thus, at all levels of net Ca absorption, negative Ca balance (above the zero balance or above the line of identity) is common in IH patients but not in healthy subjects. Negative Ca balance in the presence of adequate Ca intake is incompatible with a primary hyperabsorption of dietary Ca or absorptive hypercalciuria and cannot, by itself, account for the hypercalciuria.

1345 c. Urine Ca and Ca Balance during Low Ca Diet The hypothesis of primary intestinal Ca overabsorption predicts that dietary Ca restriction would reduce the amount of Ca absorbed and would therefore reduce urinary Ca excretion. Like normal subjects, IH patients would be in positive or neutral Ca balance when net absorption is above 200 mg/24 hr (Fig. 6). A low Ca diet would reduce urine Ca excretion through an increase in PTH secretion, which would promote distal tubular Ca reabsorption. In contrast, patients with a primary renal Ca leak would be unable to conserve urine Ca at any level of Ca intake and would maintain an excessive or inappropriately high urine Ca excretion even during low Ca diet. As a result, Ca balance during low Ca diet would shift from positive or neutral to negative or become more negative. Serum PTH would be expected to increase to high levels during a low Ca diet. To test whether the responses of IH patients fit these predictions, Coe et al. [40] fed a low Ca diet (2 mg/kg/ day) to nine normal volunteers and 26 unselected IH stone formers. After 10 days on the diet, urine Ca excretion decreased to 2.0 mg/kg body weight or less in both patients and controls, but 17 of the 26 IH patients (Fig. 7) showed values greater than the highest value in normal controls. In patients, urine Ca excretion (CaE) ranged from normal to persistently high levels. It exceeded Ca intake (CaI) (Fig. 7, CaI − CaE) in 11 of the 26 patients and none of the nonstone-forming controls. Thus, almost 50% of the patients had more Ca in the urine than what was provided by the diet and were clearly in negative Ca balance. The results indicate that a chronic low Ca diet may be detrimental for some patients, as the inability to conserve urine Ca would eventually lead to clinically detectable bone loss. The data also suggest that some patients with IH may have diet-dependent hypercalciuria, whereas others have diet-independent hypercalciuria. The two proposed mechanisms cannot be readily distinguished by any clear discontinuity in the distribution of urine Ca values, and serum PTH and 1,25(OH)2D levels do not predict the urine Ca responses during a low Ca diet. Patients with the highest urine Ca and most extreme negative Ca balance had serum PTH and 1,25 (OH)2D levels that were not different from patients who conserved Ca to the levels found in normal subjects. d. Role of 1,25(OH)2D Excess The majority of patients are classified as having absorptive hypercalciuria [62,65], yet negative Ca balance during low Ca diet [40] without elevated PTH, or 1,25(OH)2D is not predicted by the absorptive model (Fig. 5A). Patients who meet the criteria of renal hypercalciuria tend to

1346

MURRAY J. FAVUS AND FREDRIC L. COE 3

2

2

1

1

0

0

−1

−1

−2

−2

−3

−3

3

2

2

1

1

5

5

4

4

3

3

2

2

1

1

0

0 Normals

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

3

1 2 3 4 5 6 7 8 9

UCa (mg/kg/24 hr)

Cal (mg/kg/24 hr)

Cal - CaE (mg/kg/24 hr)

3

Patients

FIGURE 7 Calcium intake and urinary excretion in patients with IH and normal subjects. Urine Ca excretion (UCa) and Ca intake (Cal) are during a low Ca diet. Mean Ca intakes for patients and controls (2.29 ± 0.15 versus 2.31 ± 0.05 mg/kg/day) were not different. Mean urine excretion rates during low Ca intake and values of Cal−CaE (an index of Ca balance) differed significantly between normals and IH patients. Subjects and patients are arranged in ascending order of urinary Ca excretion. Reprinted by permission of the publisher from Coe et al. [40]. “Effects of low-calcium diet on urine calcium excretion, parathyroid function and serum 1,25(OH)2D3 levels in patients with idiopathic hypercalciuria and in normal subjects,” American Journal of Medicine, Vol. 72, pp. 25–32. Copyright 1982 by Excerpta Medica Inc.

have higher serum 1,25(OH)2D levels, but only a small portion have elevated PTH levels. Further, serum 1,25(OH)2D levels do not predict whether patients will be classified as absorptive or renal, and at least onethird of patients have normal serum 1,25(OH)2D levels despite intestinal Ca hyperabsorption. For them, the mechanism of intestinal Ca hyperabsorption remains unexplained.

The model of primary vitamin D excess (Fig. 5C) is supported by elevated 1,25(OH)2D production rates and enhanced biological actions of 1,25(OH)2D, including increased intestinal Ca absorption and bone resorption. Creation of a mild form of 1,25(OH)2D excess was achieved by the administration of pharmacological doses of 1,25(OH)2D3 (3.0 ug/day) to healthy men for 10 days while Ca intake varied from low (160 mg) to normal (372 mg) or high (880 mg) [73–75]. Increased urine Ca excretion and net intestinal Ca absorption led to negative Ca balance as calculated from 6-day metabolic balance studies (Fig. 8). Dietary Ca strongly influenced the response to 1,25(OH)2D3, as Ca balance was more negative during low Ca intake, and the increase in urine Ca resulted primarily from accelerated bone resorption. At low-normal or normal Ca intake, 1,25(OH)2D3 administration increased urine Ca and net intestinal Ca absorption, and Ca balance remained neutral. During normal Ca diet, 1,25(OH)2D3 administration maintained neutral or positive Ca balance. Thus, 3 ug/day of 1,25(OH)2D3, which was insufficient to cause hypercalcemia, during the 10-day study has profound effects on intestinal Ca absorption, urine Ca excretion, and Ca balance. Further, 1,25(OH)2D3 administration caused negative Ca balance only during low Ca intake. Thus, l,25(OH)2D3 induced changes in Ca balance in normal subjects similar to that observed in IH patients on comparable levels of Ca intake. In other experiments, ketoconazole administration to IH patients inhibited renal 1,25(OH)2D biosynthesis [64] and decreased serum 1,25(OH)2D levels, intestinal Ca absorption, and urine Ca excretion. The results of the effects of 1,25(OH)2D3 treatment and the response to ketoconazole provide further support for a primary 1,25(OH)2D excess in at least some patients with IH. The nature of the disordered regulation of renal 1,25(OH)2D production or action remains to be determined, as neither elevated PTH nor hypophosphatemia were present in responders or were absent in nonresponders to ketoconazole.

III. GENETIC HYPERCALCIURIC RATS Tests of the absorptive, renal, and vitamin D excess models of IH have been complicated by difficulty in controlling for potential variables such as inheritance and environmental factors that may influence dietary patterns. The availability of an animal model of IH would permit the testing of the three hypotheses under conditions that exclude genetic and dietary influences. The strong familial occurrence of IH in humans and the high frequency of elevated urine Ca in adult men

1347

CHAPTER 77 Idiopathic Hypercalciuria and Nephrolithiasis

Normal calcium diet 880 mg/day Low calcium diet 160 mg/day

Low normal calcium diet 372 mg/day

640

640

480

480

320

320

160

160

0

0

−160

−160

−320

−320 Net absorption

Urine calcium

Calcium balance

Net absorption

Urine calcium

Calcium balance

Net absorption

Urine calcium

Calcium balance

FIGURE 8 Intestinal Ca absorption, urine Ca excretion, and Ca balance in normal men receiving 1,25(OH)2D3 (hatched bars) or controls (open bars) at varying levels of dietary Ca. Values (mg/24 hr) are means ± SEM for 6 men per group. For Ca balance, values above the horizontal line indicate positive balance and those below the line, negative balance. Data from Maierhofer et al. [76] and Adams et al. [77]. Reprinted with permission from Coe FL, Parks JH 1988 Nephrolithiasis: Pathogenesis and Treatment, Second Edition, Year Book Publishers: Chicago 1988 [75].

and women suggested that spontaneous hypercalciuria might also be found in animals.

A. Establishment of a Colony of Genetically Hypercalciuric Rats The distribution of urine Ca excretion in a population of male Sprague-Dawley rats fed a normal Ca diet (0.8% Ca) was similar to that found in a population of healthy humans in that it followed a nonGaussian distribution, with values clustering about the mean and a long tail of higher values [14]. Using an arbitrary definition of hypercalciuria as urine Ca greater than two standard deviations above the mean value, about 5 to 10% of male and female rats were hypercalciuric. Mating males and females with the most severe hypercalciuria resulted in offspring with hypercalciuria. The most hypercalciuric offspring were used for repeated matings, leading to a colony with hypercalciuria that has increased in intensity and frequency with each successive generation [15]. By the twentieth generation,

over 95% of males and females were hypercalciuric. By the fortieth generation, mean urine Ca excretion was 7.0 ± 0.3 mg/24 hr compared to the stable mean excretion of less than 0.75 mg/24 hr by wild-type rats [77]. Hypercalciuria is lifelong and may be detected as soon as the animals are weaned (about 50 g body weight). Weight and growth of the hypercalciuric rats have been comparable to wild-type Sprague-Dawley rats obtained from the same supplier that provided the original spontaneously hypercalciuric animals. No anatomical or structural abnormalities have been identified; however, by 18 weeks of age 100% of the animals have grossly evident Ca-containing kidney stones in the upper and lower urinary tracts [78]. No stones are found in the kidney or urinary tract of wild-type rats.

B. Serum and Urine Chemistries Serum Ca and Mg are within the normal range in the genetic hypercalciuric stone-forming (GHS) male and female rats [15]. Serum phosphate is lower in

1348

MURRAY J. FAVUS AND FREDRIC L. COE

C. Mineral Balance Six-day external balance studies performed while the animals were fed a normal Ca diet showed the animals to be in positive balance for Ca, Mg, and phosphorus [15] with greater net Ca absorption in GHS rats. The GHS rats maintained positive Ca balance because the increased urine Ca excretion was matched by a greater net intestinal Ca absorption.

D. Intestinal Calcium Transport To investigate the mechanism of the increased Ca absorption, segments of duodenum were mounted in vitro in modified Ussing chambers, and transepithelial bidirectional fluxes of Ca were measured in the absence of electrochemical gradients [22]. Under these conditions, [15], duodenal segments from GHS rats had a fivefold increase in the mucosal-to-serosal (absorptive) transepithelial flux of Ca (Jms), whereas the secretory flux of Ca from serosa to mucosa (Jsm) was only mildly increased compared to wild type (Table II). As Ca Jms was 10 to 12 times higher than Ca Jsm, changes in Jsm had a nonsignificant effect on net Ca absorption.

E. Serum 1,25(OH)2D Circulating 1,25(OH)2D levels were lower in the fourth generation GHS rats; however, the differences TABLE II

In Vitro Bidirectional Duodenal Calcium Active Transport*

Flux

NM

GHM

NF

GHF

Jms Jsm Jnet

51 ± 12 11 ± 2 40 ± 11

264 ± 27 19 ± 2 245 ± 28

29 ± 9 14 ± 2 14 ± 8

258 ± 40 23 ± 2 235 ± 40

*Values are means ± SE for 5 to 11 rats per group. NM and NF are normocalciuric (wild-type) male and female rats, respectively. GHM and GHF are genetic hypercalciuric male and female rats, respectively. Jms and Jsm are mucosal-to-serosal and serosal-to-mucosal fluxes of Ca, respectively. Jnet is net Ca absorption, where Jnet = Jms − Jsm. Adapted from Li et al. [16] and reproduced with permission from The Journal of Clinical Investigation, 1993, Vol. 91, pp. 661–667, by copyright permission of The American Society for Clinical Investigation.

240 200 Duodenal calcium Jnet (nmol/cm2/h)

female rats, and there is no difference between GHS and wild-type males and females. Serum PTH levels in GHS rats are not different from controls. Urine volumes are greater in the GHS rats.

160 120 80 40 0 −40 0

20

40 60 80 Serum 1,25(OH)2D (pg/ml)

100

120

FIGURE 9 Duodenal Ca net flux (Jnet) as a function of serum 1,25(OH)2D for hypercalciuric and normocalciuric male (open and filled squares, respectively) and female (open and filled circles, respectively) rats. Jnet and serum 1,25(OH)2D were correlated for male and female normocalciuric rats (r = 0.789, n = 12, p < 0.001, solid line) and for male and female GHS rats (r = 0.500, n = 17, p < 0.03, dotted line). The regressions were different (F ratio = 5.469, p < 0.015). Reproduced from The Journal of Clinical Investigation, 1988, Vol. 82, pp. 1585–1591 [15], by copyright permission of The American Society for Clinical Investigation.

disappeared by the tenth generation [at 190 g, mean ± SD serum 1,25(OH)2D was 135 ± 12 versus 174 ± 19 pg/ml, nonsignificant], and no subsequent differences in serum 1,25(OH)2D levels have been observed [16]. In vitro duodenal net flux (Jnet, equal to Jms − Jsm) for Ca was positively correlated with serum 1,25(OH)2D in normocalciuric and GHS male and female rats (Fig. 9). However, the regression coefficients were different for the wild-type and GHS rats, with the latter having a steeper slope. The greater Ca Jnet in GHS rats with serum 1,25(OH)2D levels comparable to the wild-type rats strongly suggests that duodenal Ca-transporting cells in GH rats are more sensitive to 1,25(OH)2D.

F. Role of the Vitamin D Receptor The increased intestinal Ca transport and normal serum 1,25(OH)2D levels in GHS rats suggested either that Ca transport was being stimulated by an unidentified, vitamin D–independent process or that 1,25(OH)2D action was being amplified at the level of vitamin D target tissues. As 1,25(OH)2D stimulates Ca transport by binding to the vitamin D receptor (VDR) to up-regulate vitamin D–dependent genes that encode for proteins involved in transepithelial Ca transport, and because the biological actions of 1,25(OH)2D are directly related to the tissue VDR content [79–81], VDR binding in intestinal epithelial cells was measured. Duodenal cytosolic

1349

CHAPTER 77 Idiopathic Hypercalciuria and Nephrolithiasis

400

VDR (fmol/mg)

300

200

100

0 0.0

0.5

1.0 1.5 2.0 1,25(OH)2D3 (nM)

2.5

3.0

FIGURE 10

Specific binding of 3H-1,25(OH)2D3 to duodenal cytosolic fractions (VDR) prepared from GHS rats (filled circles) and wild-type controls (open circles) while fed a normal Ca diet. Values are means ± SEM for four observations per concentration point. *, p < 0.05; **, p < 0.01; ***, p < 0.005 vs controls. Reproduced from The Journal of Clinical Investigation, 1993, Vol. 91, pp. 661–667 [16], by copyright permission of The American Society for Clinical Investigation.

fractions prepared in high potassium buffer from male GHS rats bound more 3H-1,25(OH)2D3 than comparable fractions from wild-type control rats [16] (Fig. 10). Cytosolic fractions from kidney cortex and from splenic monocytes also exhibited greater specific binding of 1,25(OH)2D3. Scatchard analysis of the specific binding curves revealed a single class of VDR binding sites in tissues from both wild-type and GHS rats. The number of VDR binding sites in GHS rat duodenal cells was double that found in cells from wild-type rats (536 ± 73 versus 243 ± 42 fmol/mg protein; n = 8 and n = 14; p < 0.001), with comparable affinity of the receptor for its ligand (0.33 ± 0.01 versus 0.49 ± 0.01 nM; nonsignificant). A twofold increase in VDR binding sites was also found in GHS rat renal cortical homogenates [16]. Using Western blotting, homogenates of duodenal mucosa from GHS rats contained a band at 50 kDa that comigrated with duodenal extracts from wild-type rats and with recombinant human VDR. The bands from the GHS rat tissues were more intense compared to controls, confirming that the increase in specific 3H1,25(OH)2D3 binding was due to an increase in VDR protein. Northern analysis of RNA extracts from GHS and wild-type rat tissues revealed a single species of VDR mRNA at 4.4 kb with no difference in migration between the two groups [16]. Duodenal extracts from GHS rats contained less VDR mRNA than controls. Estimates of duodenal cell transcription rates using standard nuclear run-on assays found no clear difference

between GHS rats and controls [16]. The in vivo halflife of the VDR mRNA in GHS rat duodenum was comparable to that of controls (6 hr). Administration of a small dose of 1,25(OH)2D3 (30 ng as a single dose) resulted in a significant elevation of VDR message and prolongation of message half-life in GHS rats but not controls [82]. Thus, in GHS rat intestine, the increased VDR level is not due to an increase in VDR gene transcription. The data are consistent with either an increase in VDR mRNA translation efficiency or changes that result in a prolongation of the VDR half-life. The increased accumulation of the vitamin D–dependent calbindin-D9K found in GHS rat duodenum [16] is evidence that the increased level of VDR is functional and that the increased Ca transport is likely a vitamin D– mediated process. Major questions remain as to the genetic basis of the increased VDR activity. However, sequence of a cDNA prepared from GHS rat duodenal VDR mRNA failed to reveal any difference compared to the sequence of VDR cDNA prepared from wild-type duodenum. The cumulative evidence suggests that the primary genetic defect does not directly involve the VDR gene.

G. Increased Bone Resorption In vitro studies of bone resorption using neonatal calvariae from normal and GHS rats show that Ca efflux (a measure of bone resorption) increases in a dosedependent manner in the presence of 1,25(OH)2D3 or PTH [83]. The dose-response curve is much steeper for 1,25(OH)2D3 in calvariae from GHS rats, whereas the dose-response curves for PTH-stimulated Ca efflux are not different between control and GHS calvariae. Western blotting showed a fourfold increase in VDR protein from GHS neonatal rat calvariae [83]. Thus, the increase in target tissue VDR exerts biological actions that increase l,25(OH)2D3–dependent bone resorption, which likely contributes to the hypercalciuria.

H. Response to Low Calcium Diet To test whether the hypercalciuria in GHS rats is the result of a primary overabsorption of dietary Ca, GHS and wild-type control rats were fed diets either normal (0.6% Ca) or low (0.02% Ca) with respect to Ca. During the low Ca diet, urine Ca excretion decreased in both groups (Fig. 11); however, urine Ca remained higher in GHS rats and resulted in negative Ca balance [84]. The inability of GHS rats to conserve Ca during low Ca intake excludes overabsorption of dietary Ca as the sole cause of hypercalciuria in GHS rats.

1350

MURRAY J. FAVUS AND FREDRIC L. COE

VDR to compensate for urinary Ca losses has not been excluded. Further information is required regarding the renal handling of Ca in GHS rats and whether the GHS genotype results in a primary defect in renal Ca transport.

Urine calcium excretion (mg/day)

6 5 4 3

GHS – NCD GHS – LCD Ctl – NCD Ctl – LCD

2

IV. CURRENT VIEW OF HUMAN GENETIC HYPERCALCIURIA

1 0 2

4

6

8

10

12 Day

14

16

18

20

22

FIGURE 11 Daily urine Ca excretion in nineteenth-generation GHS rats (open symbols) or wild-type control rats (filled symbols) fed a normal Ca diet (NCD, 0.6% Ca, triangles) during days 1–10 followed by either continuation of the NCD (triangles) or feeding of a low Ca diet (LCD, 0.02% Ca, circles). Rats were pair-fed to 13 g of diet per day. Reprinted with permission from Kim et al. [86].

I. Summary of Pathogenesis in the Genetic Hypercalciuric Rat Figure 12 summarizes current knowledge of the pathogenesis of hypercalciuria in the GHS rats. Breeding by selection for hypercalciuria has emphasized a trait in the offspring that likely involves the expression of several genes for full phenotypic expression. To date, none of the genes has been identified. Studies implicate the increased VDR concentration as part of the primary event(s) and a cause of the hypercalciuria; however, a secondary adaptive increase in

Striking similarities in Ca metabolism between GHS rats, IH patients, and human volunteers treated with 1,25(OH)2D3 (Table III) strongly support a primary role of excess 1,25(OH)2D biological action in the pathogenesis of human IH. When deprived of dietary Ca, few patients conserve Ca to the extent that normals do (Fig. 7). The renal IH model predicts ongoing urinary losses of Ca independent of Ca intake, and negative Ca balance during a low calcium diet. However, most patients have normal, not elevated PTH, as renal IH would require. Therefore, the absorptive and renal models of hypercalciuria cannot explain the response of most patients to a low Ca diet. In nonstone formers, 1,25(OH)2D3 administration changes urine Ca and Ca balance to those observed in a majority of IH patients who have either normal or elevated serum 1,25(OH)2D levels. For some patients, elevated serum 1,25(OH)2D3 increases in intestinal Ca hyperabsorption and urine Ca excretion, and causes negative Ca balance during low Ca intake. The source of 1,25(OH)2D excess is more elusive in patients with normal serum 1,25(OH)2D levels. They may be more similar to the GHS rats in that both have normal serum 1,25(OH)2D, increased

Breeding for hypercalciuria

Genomic events

Vitamin D receptor

Intestinal Ca absorption

Bone resorption

? Renal tubular Ca transport

Hypercalciuria

FIGURE 12 Proposed series of events that result from breeding selection for hypercalciuric rats. The renal handling of Ca by GHS rats and the role of increased VDR content, if any, in the transport process remain unknown.

1351

CHAPTER 77 Idiopathic Hypercalciuria and Nephrolithiasis

TABLE III

Pathophysiology of Genetic Hypercalciuria

Parameter Serum Ca Serum phosphate Serum 1,25(OH)2D Urinary Ca on NCD Urinary Ca on LCD Intestinal Ca absorption Ca balance on NCD Ca balance on LCD

Human

Human

N N I I I I Pos-N N-Neg

N N-D N-I I N-I I N-Neg N-Neg

GHS rats N N N I I I Pos Neg

Values for human controls are responses to treatment with 3 ug 1,25(OH)2D3 daily for 7 days compared to pretreatment. GHS, genetic hypercalciuric stone-forming; NCD, normal Ca diet; LCD, low Ca diet; N, normal; I, increased; D, decreased; Pos, positive; Neg, negative.

intestinal absorption and bone resorption during a low Ca diet, and low bone density. Whether these changes in human IH are due to increased intestinal, renal, and bone cell VDR content that can amplify the biological actions of normal circulating 1,25(OH)2D levels remains to be determined.

V. THERAPEUTICS OF IDIOPATHIC HYPERCALCIURIA AND EFFECTS ON CALCIUM METABOLISM A. Dietary Calcium Restriction Hypercalciuria promotes urine calcium oxalate supersaturation and increases spontaneous crystal formation [85]. The goals of preventive therapy are to reduce Ca oxalate supersaturation by increasing urine volume and decreasing urine Ca excretion. If the pathophysiological role of 1,25(OH)2D excess or VDR excess is borne out, then ideal therapy may eventually include either a specific 1,25(OH)2D antagonist or an inhibitor of VDR function. In the absence of such agents, therapies will continue to concentrate on lowering urine Ca through indirect means. Since the description of IH, dietary Ca restriction has been recommended to lower urine Ca. Dietary Ca restriction or the use of Ca-binding resin to prevent absorption [86] could be efficacious for patients with primary intestinal Ca hyperabsorption (absorptive hypercalciuria). However, it appears that many patients develop negative Ca balance during low Ca intake. For them, chronic dietary Ca restriction and negative Ca balance would eventually cause bone loss, osteoporosis, and increased fracture risk. Reports of lower bone

density in IH patients suggest that Ca restriction may only worsen the existing reduction in bone mass. Therefore, treatment with Ca restriction requires knowledge that the patient will normally conserve urine Ca and not develop negative Ca balance.

B. Thiazides Thiazide and the related chlorthalidone diuretics reduce urine Ca excretion by inducing a NaCl diuresis, which causes volume contraction and decreased Ca delivery to the distal tubule segments [87]. These agents also stimulate distal tubule Ca reabsorption through a direct interaction with the tubule cells [86–88]. Thiazides may decrease or have no effect [31,35,89] on intestinal Ca transport in IH patients, and serum 1,25(OH)2D and PTH levels are not changed by thiazide. In one study, IH patients treated with chlorthalidone for six months improved Ca balance to or toward positive by decreasing both urine Ca and intestinal Ca absorption [90], with urine Ca declining to a greater extent than intestinal absorption. The epidemiological studies suggesting that chronic thiazide therapy reduces fracture risk [91,92] may result from druginduced improvement in Ca balance [89] and reduced bone turnover and improved mineralization [59]. The effects of thiazide on urine Ca and bone metabolism are accompanied by a decrease in new Ca stone formation compared to placebo controls [93]. The beneficial effect of thiazide is evident during the second and third year of therapy, when stone recurrence is reduced by about 50 to 80%. The reduction in new stone formation is due to a decrease in urine Ca oxalate supersaturation, as urine Ca declines while oxalate is unchanged. As thiazides can reduce urine Ca excretion and stone formation rates in all forms of IH [94], knowledge of the pathogenesis of IH in each patient may not be required prior to selecting thiazide therapy.

VI. RISK OF STONE FORMATION USING VITAMIN D ANALOGS A growing research interest in the cell differentiation and immune modulator effects of vitamin D and analogs may result in their use in a variety of disorders [95–97] (also see chapters in Sections VIII and IX). However, the development of hypercalciuria and hypercalcemia may limit the use of the naturally occurring vitamin D metabolites, as well as synthetic analogs [98,99]. While some vitamin D analogs are reported to have little or no hypercalcemic action, hypercalcemia and hypercalciuria may appear at higher doses through

1352 the classic vitamin D actions on intestine, kidney, and bone [97,98]. Low Ca diets have had only modest beneficial effects to limit hypercalciuria and hypercalcemia and could promote bone loss. It remains to be determined whether the newer vitamin D analogs with less calcemic activity will, in practice, cause less calciuria and a lower risk of kidney stone formation. Until such actions of the vitamin D analogs are known, standard approaches to minimize stone formation should be followed. These include (1) assuring sufficient fluid intake to maintain at least 1.5 liters urine output per day; (2) if necessary, increasing urine citrate excretion to normal in those with low citrate [85]; and (3) discontinuing or reducing treatment if significant hypercalciuria develops. The addition of a thiazide may avoid or minimize hypercalciuria, but hypercalcemia may occur because of thiazide-induced Ca retention.

VII. SUMMARY Idiopathic hypercalciuria is the most common cause of Ca oxalate kidney stone formation, and is the most common cause of 1,25(OH)2D3 excess. Several models of hypercalciuria incorporate intestinal Ca hyperabsorption, increased bone resorption, and decreased renal tubule Ca reabsorption, all of which can be accounted for by the elevated serum 1,25(OH)2D3 found in about 50% of patients. There is evidence for a defect in the regulation of the 1-hydroxylase, but the nature of the dysregulation remains unknown. In those patients with normal serum 1,25(OH)2D3 levels, the possibility of vitamin D receptor excess as found in GHS rats, offers a testable hypothesis.

MURRAY J. FAVUS AND FREDRIC L. COE

8. 9. 10. 11. 12. 13. 14. 15. 16.

17.

18. 19. 20. 21. 22.

References 1. Flocks RH 1939 Calcium and phosphorus excretion in the urine of patients with renal or ureteral calculi. JAMA 13:1466–1471. 2. Albright F, Henneman P, Benedict PH, Forbes HR 1953 Idiopathic hypercalciuria. A preliminary report. Proc R Soc Med Lond (Biol) 46:1077. 3. Henneman PH, Benedict PH, Forbes AP Dudley HR 1958 Idiopathic hypercalciuria. N Engl J Med 259:802–807. 4. Hodgkinson A, Pyrah LN 1958 The urinary excretion of calcium and inorganic phosphate in 344 patients with calcium stones of renal origin. Br J Surg 46:10–18. 5. Robertson WG, Morgan DB 1972 The distribution of urinary calcium excretions in normal persons and stone-formers. Clin Chim Acta 37:503–508. 6. Coe FL, Parks JH, Asplin JR 1992 The pathogenesis and treatment of kidney stones. N Engl J Med 327:1141–1152. 7. Coe FL 1977 Treated and untreated recurrent calcium nephrolithiasis in patients with idiopathic hypercalciuria,

23. 24.

25. 26. 27. 28.

hyperuricosuria, or no metabolic disorder. Ann Intern Med 87:404–410. McGeown MG 1960 Heredity in renal stone disease. Clin Sci 19:465–471. Resnick M, Pridgen DB, Goodman HO 1968 Genetic predisposition of calcium oxalate renal calculi. N Engl J Med 278:1313–1318. Coe FL, Parks JH, Moore ES 1979 Familial idiopathic hypercalciuria. N Engl J Med 300:337–340. Pak CYC, McGuire J, Peterson R, Britton F, Harrod MJ 1981 Familial absorptive hypercalciuria in a large kindred. J Urol 126:717–719. Mehes K, Szelid Z 1980 Autosomal dominant inheritance of hypercalciuria. Eur J Pediatr 133:239–242. Moore ES, Coe FL, McMann BJ, Favus MJ 1978 Idiopathic hypercalciuria in children: Prevalence and metabolic characteristics. J Pediatr 92:906–910. Coe FL, Favus MJ 1981 Hypercalciuric states. Miner Electrolyte Metab 5:183–200. Bushinsky DA, Favus MJ 1988 Mechanism of hypercalciuria in genetic hypercalciuric rats. Inherited defect in intestinal calcium transport. J Clin Invest 82:1585–1591. Li X-Q, Tembe V, Horwitz GM, Bushinsky DA, Favus MJ 1993 Increased intestinal vitamin D receptor in genetic hypercalciuric rats: A cause of intestinal calcium hyperabsorption. J Clin Invest 91:661–667. Broadus AE, Insogna KL, Lang R, Mallette LE, Oren DA, Gertner JM, Kliger AS, Ellison AF 1984 A consideration of the hormonal basis and phosphate leak hypothesis of absorptive hypercalciuria. J Clin Endocrinol Metab 58:161–169. Buckalew VM, Purvis ML, Shulman MG, Herndon CN, Rudman D 1974 Hereditary renal tubular acidosis. Medicine 53:229–254. Frymoyer PA, Scheinman SJ, Dunham PB, Jones DB, Hueber P, Schroeder ET 1991 X-linked recessive nephrolithiasis with renal failure. N Engl J Med 325:681–686. Low RK, Stoller ML 1997 Endoscopic mapping of renal papillae for Randall’s plaques in patients with urinary stone disease. J Urol 158:2062–2064. Randall A 1937 The origin and growth of renal calculi. Ann Surg 105:1009–1027. Evan AP, Lingeman JE, Coe FL, Parks JH, Bledsoe SB, Shao Y, Sommer AJ, Paterson RF, Kuo RL, Grynpas M 2003 Randall’s plaque of patients with nephrolithiasis begins in basement membranes of thin loops of Henle. J Clin Invest 111:607–616. Bushinsky DA 2003 Nephrolithiasis: site of the initial solid phase. J Clin Invest 111:602–605. Klugman VA, Favus MJ 1996 Intestinal absorption of calcium, magnesium, and phosphorus. In: Coe FL, Favus MJ, Pak CYC, Parks JH, Preminger GM (eds) Kidney Stones: Medical and Surgical Management, 1st Ed. Lippincott-Raven: Philadelphia, Pennsylvania, pp. 210–221. Bronner F, Pansu D, Stein WD 1986 An analysis of intestinal calcium transport across the rat intestine. Am J Physiol 250:G561–G569. Favus MJ 1985 Factors that influence absorption and secretion of calcium in the small intestine and colon. Am J Physiol 248:G147–G157. Caniggia A, Gennari C, Cesari L 1965 Intestinal absorption of 45Ca in stone-forming patients. Br Med J 1:427–429. Birge SJ, Peck WA, Berman M, Wheadon GD 1969 Study of calcium absorption in man: A kinetic analysis and physiologic model. J Clin Invest 48:1705–1713.

CHAPTER 77 Idiopathic Hypercalciuria and Nephrolithiasis

29. Wills MR, Zisman E, Wortsman J, Evens RG, Pak CYC, Bartter FC 1970 The measurement of intestinal calcium absorption in nephrolithiasis. Clin Sci 39:95–106. 30. Pak CYC, East DA, Sanzenbacher LJ, Delea CS, Bartter FC 1972 Gastrointestinal calcium absorption in nephrolithiasis. J Clin Endocrinol Metab 35:261–270. 31. Ehrig U, Harrison JE, Wilson DR 1974 Effect of long-term thiazide therapy on intestinal calcium absorption in patients with recurrent renal calculi. Metabolism 23:139–149. 32. Pak CYC, Ohata M, Lawrence EC, Snyder W 1974 The hypercalciurias: Causes, parathyroid functions, and diagnostic criteria. J Clin Invest 54:387–400. 33. Kaplan RA, Haussler MR, Deftos LJ, Bone H, Pak CYC 1977 The role of 1,25-dihydroxyvitamin D in the mediation of intestinal hyperabsorption of calcium in primary hyperparathyroidism and absorptive hypercalciuria. J Clin Invest 59:756–760. 34. Shen FH, Baylink DJ, Nielsen RL, Sherrard DJ, Ivey JL, Haussler MR 1977 Increased serum 1,25-dihydroxyvitamin D in idiopathic hypercalciuria. J Lab Clin Med 90:955–962. 35. Barilla DE, Tolentino R, Kaplan RA, Pak CYC 1978 Selective effects of thiazide on intestinal absorption of calcium in absorptive and renal hypercalciurias. Metabolism 27:125–131. 36. Lemann J Jr 1992 Pathogenesis of idiopathic hypercalciuria and nephrolithiasis. In: Coe FL, Favus MJ (eds) Disorders of Bone and Mineral Metabolism, 1st Ed. Raven: New York, pp. 685–706. 37. Duncombe VM, Watts RWE, Peters TJ 1984 Studies on intestinal calcium absorption in patients with idiopathic hypercalciuria. Q J Med 209:69–79. 38. Insogna KL, Broadus AE, Dreyer BE, Ellison AF, Gertner JM 1985 Elevated production rate of 1,25-dihydroxyvitamin D in patients with absorptive hypercalciuria. J Clin Endocrinol Metab 61:490–495. 39. Bataille P, Bouillon R, Fournier A, Renaud H, Gueris J, Idrissi A 1987 Increased plasma concentrations of total and free 1,25(OH)2D3 in calcium stone formers with idiopathic hypercalciuria. Contrib Nephrol 58:137–142. 40. Coe FL, Favus MJ, Crockett T, Strauss LM, Parks JH, Porat A, Gantt CL, Sherwood LM 1982 Effects of low-calcium diet on urine calcium excretion, parathyroid function, and serum 1,25(OH)2D3 levels in patients with idiopathic hypercalciuria and in normal subjects. Am J Med 72:25–32. 41. Broadus AE, Insogna KL, Lang R, Ellison AF, Dreyer BE 1984 Evidence for disordered control of 1,25-dihydroxyvitamin D production in absorptive hypercalciuria. N Engl J Med 311:73–80. 42. Gray RW, Wilz DR, Caldas AE, Lemann J Jr 1977 The importance of phosphate in regulating plasma l,25(OH)2 vitamin D levels in humans: Studies in healthy subjects, in calcium stone formers, and in patients with primary hyperparathyroidism. J Clin Endocrinol Metab 45:299–306. 43. Reed BY, Gitomer WL, Heller HJ, Hsu MC, Lemke M, Padalino P, Pak CYC 2002 Identification and characterization of a gene with base substitutions associated with the absorptive hypercalciuria phenotype and low spinal bone density. J Clin Endocrinol Metab 87:1476–1485. 44. Econs MJ, Foroud T 2002 Editorial: The genetics of absorptive hypercalciuria—A note of caution. J Clin Endocrinol Metab 87:1473–1475. 45. Edwards NA, Hodgkinson A 1965 Metabolic studies in patients with idiopathic hypercalciuria. Clin Sci 29:143–157. 46. Peacock M, Nordin BEC 1968 Tubular reabsorption of calcium in normal and hypercalciuric subjects. J Clin Pathol 21:353–358.

1353 47. Sutton RAL, Walker VR 1980 Responses to hydrochlorothiazide and acetazolamide in patients with calcium stones. N Engl J Med 302:709–713. 48. Bianchi G, Vezzoli G, Cusi D, Cova T, Elli A, Soldati L, Tripodi G, Surian M, Ottaviano E, Rigatti P 1988 Abnormal red-cell calcium pump in patients with idiopathic hypercalciuria. N Engl J Med 319:897–901. 49. Alhava EM, Juuti M, Karjalainen P 1976 Bone mineral density in patients with urolithiasis. Scand J Urol Nephrol 10:154–156. 50. Lawoyin S, Sismilich S, Browne R, Pak CYC 1979 Bone mineral content in patients with calcium urolithiasis. Metabolism 28:1250–1254. 51. Borgi L, Meschi T, Guerra A, Maninetti L, Pedrazzoni M, Macato A, Vescovi P, Novarini A 1991 Vertebral mineral content in diet-dependent and diet-independent hypercalciuria. J Urol 146:1334–1338. 52. Bataille P, Achard JM, Fournier A, Boudailliez B, Westeel PF, Laval Jeantet MAL, Bouillon R, Sebert JL 1991 Diet, vitamin D, and vertebral mineral density in hypercalciuric calcium stone formers. Kidney Int 39:1193–1205. 53. Pietschmann F, Breslau NA, Pak CYC 1992 Reduced vertebral bone density in hypercalciuric nephrolithiasis. J Bone Miner Res 7:1383–1388. 54. Barkin J, Wilson DR, Manuel MA, Arnold B, Murray T, Harrison J 1985 Bone mineral content in idiopathic calcium nephrolithiasis. Miner Electrolyte Metab 11:19–24. 55. Liberman UA, Sperling O, Atsmon A, Frank M, Modan M, de-Vries A 1968 Metabolic and calcium kinetic studies in idiopathic hypercalciuria. J Clin Invest 47:2580–2590. 56. Malluche HH, Tschoepe W, Ritz E, Meyer-Sabelle W, Massry SG 1980 Abnormal bone histology in idiopathic hypercalciuria. J Clin Endocrinol Metab 50:654–658. 57. Steiniche T, Mosekilde L, Christensen MS, Melsen F 1989 A histomorphometric determination of iliac bone remodeling in patients with recurrent renal stone formation and idiopathic hypercalciuria. APMIS 97:309–316. 58. Sutton RAL, Walker VR 1986 Bone resorption and hypercalciuria in calcium stone formers. Metabolism 35:465–488. 59. Urivetzky M, Anna PS, Smith AD 1988 Plasma osteocalcin levels in stone disease. A potential aid in the differential diagnosis of calcium nephrolithiasis. J Urol 139:12–14. 60. Asplin JR, Bauer KA, Kinder J, Muller G, Coe BJ, Parks JH, Coe FL 2003 Bone mineral density and urine calcium excretion among subjects with and without nephrolithiasis. Kidney Internat 63:662–669. 61. Melton LJ III, Crowson CS, Khosla S, Wilson DM, O’Fallon WM 1998 Fracture risk among patients with urolithiasis: A population-based cohort study. Kidney Int 53: 459–464. 62. Pak CYC, Kaplan R, Bone H, Townsend J, Waters O 1975 A simple test for the diagnosis of absorptive, resorptive, and renal hypercalciurias. N Engl J Med 292:497–500. 63. Van Den Berg CJ, Kumar R, Wilson DM, Heath III H, Smith LH 1980 Orthophosphate therapy decreases urinary calcium excretion and serum 1,25-dihydroxyvitamin D concentrations in idiopathic hypercalciuria. J Clin Endocrinol Metab 51:998–1001. 64. Breslau NA, Preminger GM, Adams BV, Otey J, Pak CYC 1992 Use of ketoconazole to probe the pathogenetic importance of 1,25-dihydroxyvitamin D in absorptive hypercalciuria. J Clin Endocrinol Metab 75:1446–1452. 65. Pak CYC, Britton F, Peterson R, Ward D, Northcutt C, Breslau NA, McGuire J, Sakahee K, Bush S, Nicar M, Norman DA,

1354

66. 67. 68. 69.

70. 71. 72. 73. 74.

75.

76. 77.

78. 79. 80.

81. 82.

Peters P 1980 Ambulatory evaluation of nephrolithiasis: Classification, clinical presentation and diagnostic criteria. Am J Med 69:19–30. Knapp EL 1943 Studies on the urinary excretion of calcium. Ph.D. Thesis, Department of Chemistry, State University of Iowa, Ames. Lafferty FW, Pearson OH 1963 Skeletal, intestinal, and renal calcium dynamics in hyperparathyroidism. J Clin Endocrinol Metab 23:891–902. Nassim JR, Higgins BA 1965 Control of idiopathic hypercalciuria. Br Med J 1:675–681. Jackson WPU, Dancaster C 1959 A consideration of the hypercalciuria in sarcoidosis, idiopathic hypercalciuria, and that produced by vitamin D. A new suggestion regarding calcium metabolism. J Clin Endocrinol Metab 19:658–680. Harrison AR 1959 Some results of metabolic investigations in cases of renal stone. Br J Urol 31:398. Dent CE, Harper CM, Parfitt AM 1964 The effect of cellulose phosphate on calcium metabolism in patients with hypercalciuria. Clin Sci 27:417–425. Parfitt AM, Higgins BA, Nassim JR, Collins JA, Hilb A 1964 Metabolic studies in patients with hypercalciuria. Clin Sci 27:463–482. Coe FL, Parks JH 1988 Nephrolithiasis: Pathogenesis and Treatment, 2nd Ed. Year Book Publishers: Chicago, p. 113. Maierhofer WJ, Lemann J Jr, Gray RW, Cheung HS 1984 Dietary calcium and serum 1,25-(OH)2 vitamin D concentrations as determinants of calcium balance in healthy men. Kidney Int 26:752–759. Adams ND, Gray RW, Lemann J Jr 1979 The effects of oral CaCO3 loading and dietary calcium deprivation on plasma 1,25-dihydroxyvitamin D concentration in healthy adults. J Clin Endocrinol Metab 48:1008–1016. Adams ND, Gray RW, Lemann J Jr, Cheung HS 1982 Effects of calcitriol administration on calcium metabolism in healthy men. Kidney Int 21:90–97. Bashir MA, Nakagawa Y, Riordon D, Coe FL, Bushinsky DA 1995 Increased dietary oxalate does not increase urinary calcium oxalate oversaturation in hypercalciuric rats. J Am Soc Nephrol 6:943 (abstract). Bushinsky DA, Nilsson EL, Nakagawa Y, Coe FL 1995 Stone formation in genetic hypercalciuric rats. Kidney Int 48:1705–1713. Costa EM, Hirst MA, Feldman D 1985 Regulation of 1,25dihydroxyvitamin D3 receptor by vitamin D analogs in cultured mammalian cells. Endocrinology 117:2203–2210. Pols HAP, Birkenhager JC, Schlite JP, Visser TJ 1988 Evidence that self-induced metabolism of 1,25-dihydroxyvitamin D3 limits the homologous up-regulation of its receptor in rat osteosarcoma cells. Biochim Biophys Acta 970:122–129. Reinhardt TA, Horst RL 1989 Self-induction of 1,25dihydroxyvitamin D3 metabolism limits receptor occupancy and target tissue responsiveness. J Biol Chem 264:15917–15921. Yao J, Kathpalia P, Bushinsky DA, Favus MJ 1998 Hyperresponsiveness of vitamin D receptor gene expression to 1,25-dihydroxyvitamin D3. A new characteristic of genetic hypercalciuric stone-forming rats. J Clin Invest 101:2223–2232.

MURRAY J. FAVUS AND FREDRIC L. COE

83. Krieger NS, Stathopoulos VM, Bushinsky DA 1996 Increased sensitivity to 1,25(OH)2D3 in bone from genetic hypercalciuric rats. Am J Physiol 271:C130–C135. 84. Kirn M, Sessler NE, Tembe V, Favus MJ, Bushinsky DA 1993 Response of genetic hypercalciuric rats to a low calcium diet. Kidney Int 43:189–196. 85. Parks JH, Coe FL 1996 Pathogenesis and treatment of calcium stones. Semin Nephrol 16:398–411. 86. Wilson DR, Strauss AL, Manuel MA 1984 Comparison of medical treatments for the prevention of recurrent calcium nephrolithiasis. Urol Res 12:39–40. 87. Edwards BR, Baer PG, Sutton RA, Dirks JH 1973 Micropuncture study of diuretic effects on sodium and calcium reabsorption in the dog nephron. J Clin Invest 52:2418–2427. 88. Costanzo LS, Windhager EE 1978 Calcium and sodium transport by the distal convoluted tubule of the rat. Am J Physiol 235:F492–F506. 89. Zerwekh JE, Pak CYC 1980 Selective effects of thiazide therapy on serum l-alpha,25-dihydroxyvitamin D and intestinal calcium absorption in renal and absorptive hypercalciurias. Metabolism 29:13–17. 90. Coe FL, Parks JP, Bushinsky DA, Langman CB, Favus MJ 1988 Chlorthalidone promotes mineral retention in patients with idiopathic hypercalciuria. Kidney Int 33:1140–1146. 91. Wasnich RD, Benfante RJ, Yano K, Heilbrun L, Vogel JM 1983 Thiazide effect on the mineral content of bone. N Engl J Med 309:344–347. 92. LaCroix AZ, Wienpahl J, White LR, Wallace RB, Scherr PA, George LK 1990 Thiazide diuretic agents and the incidence of hip fracture. N Engl J Med 322:286–290. 93. Asplin JR, Favus MJ, Coe FL 1996 Nephrolithiasis. In: Brenner BR (ed) The Kidney, 5th Ed. Saunders: Philadelphia, Pennsylvania, pp. 1893–1935. 94. Ohkawa M, Tokunga S, Nakashima T, Orito M, Hisazumi H 1992 Thiazide treatment for calcium nephrolithiasis in patients with idiopathic hypercalciuria. Br J Urol 69:571–576. 95. Cheskis B, Lemon BD, Uskokovic M, Lomedico PT, Freedman LP 1995 Vitamin D3 retinoid X receptor dimerization, DNA binding, and transactivation are differentially affected by analogs of 1,25-dihydroxyvitamin D3. Mol Endocrinol 9:1814–1824. 96. Skowronski RJ, Peehl DM, Feldman D 1995 Actions of vitamin D3, analogs on human prostate cancer cell lines: Comparison with 1,25-dihydroxyvitamin D3. Endocrinology 136:20–26. 97. Fleet JC, Bradley J, Reddy GS, Ray R, Wood RJ 1996 1 alpha,25-(OH)2 vitamin D analogs with minimal in vivo calcemic activity can stimulate significant transepithelial calcium transport and mRNA expression in vitro. Arch Biochem Biophys 329:228–234. 98. Naveh-Many T, Silver J 1993 Effects of calcitriol, 22-oxacalcitriol, and calcipotriol on serum calcium and parathyroid hormone gene expression. Endocrinology 133:2724–2728. 99. Brown AJ, Finch J, Grieff M, Ritter C, Kubodera N, Nishii Y, Slatopolsky E 1993 The mechanism for the disparate actions of calcitriol and 22-oxacaltriol in the intestine. Endocrinology 133:1158–1164.

CHAPTER 78

Hypercalcemia Due to Vitamin D Toxicity MISHAELA R. RUBIN AND SUSAN THYS-JACOBS Department of Medicine, College of Physicians and Surgeons, Columbia University, New York, New York

FREDRIECH K. W. CHAN Department of Medicine, Queen Elizabeth Hospital, Hong Kong

LILIA M. C. KOBERLE Health Sciences Department, Federal University, Sao Carlos, Brazil

JOHN P. BILEZIKIAN Department of Medicine and Department of Pharmacology, College of Physicians and Surgeons, Columbia University, New York, New York

I. II. III. IV. V.

Introduction Forms of Exogenous Vitamin D Toxicity Forms of Endogenous Vitamin D Toxicity Mechanisms of Vitamin D Toxicity Clinical Manifestations

I. INTRODUCTION Vitamin D toxicity is not a common cause of hypercalcemia. In the differential diagnosis of hypercalcemia, it is often buried amid a long list of other more and less common causes (Table I). Among the more common causes, primary hyperparathyroidism and hypercalcemia of malignancy are the principal etiologies. Together, primary hyperparathyroidism and hypercalcemia of malignancy constitute the overwhelming majority of causes of hypercalcemia. They are in fact so common that the practical issue in the diagnosis of a hypercalcemic individual is to distinguish between these two etiologies first and not to consider other etiologies until these two have been ruled out. Patients with primary hyperparathyroidism tend to be asymptomatic, whereas patients with hypercalcemia of malignancy tend to be ill. The diagnosis of primary hyperparathyroidism is established by an elevated concentration of parathyroid hormone (PTH), an association that is made in over 90% of patients with primary hyperparathyroidism. In contrast, patients with hypercalcemia of malignancy, including those whose hypercalcemia is due to the elaboration of parathyroid hormone-related protein (PTHrP), show levels of PTH that are typically suppressed. If the PTH level is suppressed, the diagnosis of primary hyperparathyroidism is ruled out. The diagnosis of malignancy, however, is VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

VI. VII. VIII. IX.

Diagnosis of Vitamin D Toxicity Treatment of Vitamin D Toxicity Evidence for Benefits of Higher Vitamin D Levels Summary and Conclusions References

not necessarily ruled in. Certainly, if a malignancy is detected that is classically associated with hypercalcemia, such as squamous cell carcinoma of the lung, the etiology becomes clear. However, the longer list of other causes of hypercalcemia is also associated, with rare exceptions, with reduced levels of PTH. This situation, namely, elevated serum calcium concentration with reduced or undetectable levels of PTH, is seen in the various forms of vitamin D toxicity. If primary hyperparathyroidism is ruled out and malignancy is not apparent, the likelihood of vitamin D toxicity looms as an important possible etiology of hypercalcemia. In that long list of other causes, vitamin D toxicity now becomes a major diagnostic consideration (Table I). This chapter reviews the various forms of vitamin D toxicity, mechanisms of hypercalcemia due to vitamin D toxicity, clinical manifestations, diagnosis, and management.

II. FORMS OF EXOGENOUS VITAMIN D TOXICITY Vitamin D toxicity can be life threatening and associated with high morbidity, if not identified quickly. Hypervitaminosis D with hypercalcemia may be secondary to excessive intake of parent vitamin D, its metabolites 25-hydroxyvitamin D (25OHD), 1,25dihydroxyvitamin D [1,25(OH)2D], or vitamin D Copyright © 2005, Elsevier, Inc. All rights reserved.

1356 TABLE I Differential Diagnosis of Hypercalcemia Primary hyperparathyroidism Sporadic (adenoma, hyperplasia, or carcinoma) Familial Isolated Cystic Multiple endocrine neoplasia type I or II Malignancy Parathyroid hormone-related protein Excess production of 1,25-dihydroxyvitamin D Other factors (cytokines, growth factors) Disorders of vitamin D Exogenous vitamin D toxicity—parent D compound, 25OHD,1,25(OH)2D Endogenous production of 25-hydroxyvitamin D (Williams syndrome) Endogenous production of 1,25-dihydroxyvitamin D Granulomatous diseases a. Sarcoidosis b. Tuberculosis c. Histoplasmosis d. Coccidioidomycosis e. Leprosy f. Others Lymphoma Nonparathyroid endocrine disorders Thyrotoxicosis Pheochromocytoma Acute adrenal insufficiency Vasoactive intestinal polypeptide hormone-producing tumor (VIPoma) Medications Thiazide diuretics Lithium Estrogens/antiestrogens, testosterone in breast cancer Milk-alkali syndrome Vitamin A toxicity Familial hypocalciuric hypercalcemia Immobilization Parenteral nutrition Aluminum excess Acute and chronic renal disease

analogs; to increased production of 25OHD or 1,25(OH)2D from exogenous substrate; and even to topical applications of potent vitamin D analogs.

A. Vitamin D and 25-Hydroxyvitamin D Toxicity The most common etiology of vitamin D toxicity is inadvertent or improper oral use of pharmaceutical preparations. Excessive ingestion of vitamin D (usually greater than 10,000 IU daily) can cause vitamin D intoxication that is recognized by markedly elevated levels of 25OHD (usually >150 ng/ml) in association with levels of 1,25(OH)2D that are only slightly

MISHAELA R. RUBIN

ET AL .

elevated. Hyperphosphatemia typically accompanies the hypercalcemia [1–3]. The hyperphosphatemia can be a clue to the etiology of the hypercalcemia as due to vitamin D toxicity. The usual setting of vitamin D toxicity is in its use as a therapy for the hypophosphatemic disorders: hypoparathyroidism, pseudohypoparathyroidism, osteomalacia, renal failure, or osteoporosis. Ingestion of excessive quantities of 25OHD, 1-alpha-hydroxyvitamin D, 1,25(OH)2D, dihydrotachysterol, or exuberant use of the topical calcipotriene (Dovonex) for psoriasis can cause vitamin D intoxication [4]. Health conscious adults have been reported to ingest large doses of megavitamins from over the counter supplements, in amounts that may exceed 2 million IU of vitamin D daily [5]. Cancer patients, in particular, have been observed to consume excess nutritional supplements such as calcium, vitamin D, and shark cartilage [6]. Excessive sunlight exposure can raise serum concentrations of 25OHD to as high as 79 ng/ml (normal range 9–52 ng/ml), but there is no evidence that sunlight exposure alone can result in vitamin D toxicity and hypercalcemia in normal individuals [7]. Hypercalcemia associated with granulomatous diseases, such as sarcoidosis, can be worsened by excessive sunlight exposure. Natural foods, in general, other than fatty fish, eggs, milk, and liver do not contain much vitamin D. Hypervitaminosis D has been associated with drinking milk when erroneously fortified with massive concentrations of vitamin D. One investigation of eight patients manifesting symptoms of nausea, vomiting, weight loss, hyperirritability, or failure-to-thrive revealed markedly elevated mean concentrations of 25OHD of 293 ± 174 ng/ml (nl: 9–52 ng/ml) [3]. Analysis of the milk production facility at the local dairy revealed excessive vitamin D fortification of milk with up to 245,840 IU per liter (232,565 IU of vitamin D3 per quart). Usual fortification of milk in the United States is 400 IU per quart. Milk is not fortified with vitamin D in other parts of the world. Generally, milk is the only dairy product that is fortified with vitamin D in the United States. In addition to milk, vitamin D fortification of natural foods includes certain breakfast cereals, pasta, baked goods, fats, and recently orange juice oils [8]. There is no documentation that excessive ingestion of any of these other fortified foods has ever resulted in vitamin D toxicity. However, industrial contamination of table sugar with vitamin D3 and consequent severe vitamin D toxicity (25OHD 1555 nmol/L, nl: 20–80 nmol/l) has been reported [9]. Vitamin D2 and vitamin D3, although used interchangeably in the treatment of metabolic bone diseases, may differ in toxic potential at higher doses. In general, vitamin D3 appears to be somewhat more toxic than D2. Investigations in rats, sheep, pigs, horses, and primates support differences in metabolic clearance

1357

CHAPTER 78 Hypercalcemia Due to Vitamin D Toxicity

rates and in toxicity between the two vitamin D compounds [10]. In horses, vitamin D2 has a lower toxicity compared to vitamin D3 [11]. Massive doses of vitamin D3 administered to Old World primates can cause toxicity and death, whereas equivalent doses of vitamin D2 are better tolerated [12]. In human subjects, it has been shown that vitamin D3 increases 25OHD levels 1.7 times more than the equivalent dose of vitamin D2 [13]. The current, officially recommended dietary allowance (RDA) for vitamin D is 400 IU per day, but many authoritative bodies are calling for increases in the requirements to 600 units in individuals over the age of 70 [14]. Although this chapter concerns itself with vitamin D toxicity, the reason for the trend to increase recommendations for vitamin D intake is the large numbers of free-living adults who are being shown worldwide to have vitamin D deficiency (see Chapters 61–62) [15,16]. The smallest dose of parent vitamin D in adults that can produce toxicity and hypercalcemia is not known, but is clearly much higher than the RDA [17]. The threshold for vitamin D toxicity was evaluated in a study in which 61 subjects were randomized to 1000 or 4000 IU vitamin D3 daily for 2–5 months. Increases in 25OHD were greater in the higher dose than lower dose group (96.4 ± 14.6 vs. 68.7 ± 16.9 nmol/L), but remained within the physiologic range, leading the authors to conclude that as much as 4000 IU daily was a safe dose [18]. On the other hand, in infants, daily dosages of 2000 IU or less have been associated with hypercalcemia and nephrocalcinosis [19]. Intermittent oral dosages of 15 mg or 600,000 IU to infants to prevent vitamin deficiency have been shown to be excessive during the first year of life, resulting in transient hypercalcemia and vitamin D overload [20,21,22]. Lower amounts of 5 mg (200,000 IU) every 6 months or 2.5 mg (100,000 IU) every 3 months appear to be safer and to provide better protection in high risk infants. In adults, doses of greater than 40,000–60,000 IU per day, as commonly used in the treatment of hypoparathyroidism, can be associated with significant toxicity. Individuals manifest wide variations both in their response to hypercalcemic doses of vitamin D and in the duration of the effect. This variation in individual responsiveness might reflect differences in intestinal absorption and vitamin D metabolism, in the concentration of free vitamin D metabolites, in the rate of degradation of the metabolites and conversion to inactive metabolites, and in the capacity of storage sites for 25OHD [17]. Factors that enhance susceptibility to vitamin D toxicity and hypercalcemia include increased dietary calcium intake, reduced renal function, co-administration of vitamin A, and granulomatous disorders such as sarcoidosis that render subjects more sensitive to vitamin D (see Chapter 79) [2]. Hypercalciuria in hypervitaminosis D usually presents much earlier than

hypercalcemia, but it is easily missed for the obvious reason that hypercalciuria is not routinely measured in the absence of renal symptomatology.

B. 1,25-Dihydroxyvitamin D Toxicity The greater potency of 1,25(OH)2D3 and its direct actions on target tissues have resulted in its increased use for a variety of metabolic bone diseases [23]. Its ability to inhibit PTH synthesis and secretion has also made 1,25(OH)2D3 and its analogs useful agents in patients with renal osteodystrophy and secondary hyperparathyroidism. Most recently, 1,25(OH)2D3 has been found to inhibit the growth of human cancer cells in vitro [24] (see below, and Section VIII of this book). As 1,25(OH)2D3 is increasingly recognized for its antiproliferative, prodifferentiating, and immunomodulatory actions, its potential therapeutic use is expanding [25–27]. Thus, considerable attention has focused on possible toxic effects of 1,25(OH)2D3 not usually associated with the parent vitamin D compound. The incidence of hypercalcemia and hypercalciuria with 1,25(OH)2D3 use has been reported as very high, with one review citing complications in two-thirds of treated patients [28]. The mechanisms of the hypercalcemia are increased intestinal absorption and potentiation of osteoclastic activity. Dosages of 1,25(OH)2D3 above 0.75 µg/day have been associated with toxicity, whereas dosages at or below 0.5 µg/day rarely result in toxicity. One investigation showed that over 90% of patients on doses of 1,25(OH)2D3 between 1.0 and 2.0 µg/day became hypercalcemic, and all had hypercalciuria when calcium intake was set at 1000 mg per day [29]. Accelerated deterioration of renal function was recorded in a number of reports in patients with renal insufficiency receiving 1,25(OH)2D3 therapy [30]. Compared to oral therapy, intravenous administration of 1,25(OH)2D3 to renal dialysis patients induces hypercalcemia less frequently, with a smaller increment in the serum calcium concentration and a more effective reduction of PTH levels [31]. Other studies, however, suggest that intermittent oral pulse administration of 1,25(OH)2D3 may be effective, though not as effective as intravenous 1,25(OH)2D3, in suppressing PTH in uremic patients with secondary hyperparathyroidism [32–34] (see Chapter 76 for further discussion).

C. Toxicity Due to Synthetic Analogs In one investigation, oral pulse therapy with 1α-hydroxyvitamin D3 (1αOHD3) resulted in a rapid control of secondary hyperparathyroidism without causing hypercalcemia or hyperphosphatemia [35].

1358 However, 1α-OHD3 may harbor potential calcemic effects similar to 1,25(OH)2D3 in the treatment of renal osteodystrophy. Crocker et al. [36] investigated the comparative toxicity of vitamin D, 1α-OHD3, and 1,25(OH)2D3 in weanling male mice at three different doses over a four-week period. 1α-OHD3 appeared to be more toxic in the high dose group only, with significantly higher serum calcium levels, higher urinary calcium excretion, and severe nephrocalcinosis [36]. 1α-OHD3 has been described as less potent than 1,25(OH)2D3 at low doses but equipotent at doses greater than 2.0 µg/day. At the higher doses, there is a delayed onset of action and a prolonged half-life, suggesting a potential for cumulative toxicity in renal insufficiency [37,38]. The potential for hypercalcemia, hypercalciuria, and soft tissue calcifications limits the clinical usefulness of 1α-OHD3. Mortensen and colleagues compared the toxicity of both 1α-OHD3 and 1,25(OH)2D3 in rats fed standard or low calcium diets. High doses of either compound resulted in severe hypercalcemia, with retarded growth, nephrosis, and structural bone changes in the rats fed the standard diet. On the low calcium diet, however, slight hypercalcemia occurred, but without growth retardation or bone changes. There was minimal effect on the kidney. Calcium restriction again proved effective in protecting the animals against the toxic effects of the vitamin D analogs. Animals fed the low calcium diet tolerated 1α-OHD3 at dose levels up to 10 times higher than rats on the standard diets [39]. In human subjects, 1α-OHD3 causes toxic effects at doses above 1.0 µg/day, but doses of 0.5 to 1.0 µg/day appear to be safe. Because of the relatively narrow therapeutic window of vitamin D3 compounds, a synthetic analog of vitamin D2, 1α-OHD2 (doxercalciferol) was developed with the concept that the window of therapeutic efficacy to toxicity would be wider. In postmenopausal osteopenic women, doses of doxercalciferol ranging from 1.0 to 5.0 µg/day were administered to 15 subjects. There was no evidence of vitamin D toxicity manifesting as either hypercalciuria or hypercalcemia, whereas significant therapeutic effects on osteoblastic activity were demonstrated [40]. Similar to the 1α-OHD3, doxercalciferol requires obligatory hepatic 25-hydroxylation for activation. However, doxercalciferol is able to activate its catabolic pathway via hepatic 24-hydroxylation with a lower potential for toxicity [41]. These investigations on synthetic analogs seem to confirm earlier reports that vitamin D2 compounds, in general, are as efficacious and somewhat better tolerated than D3 compounds. Because of our understanding of the nonclassic target tissue effects of vitamin D in the modulation of hormones and cytokines, and in the regulation of cellular differentiation and proliferation, newer clinical uses have been developed (see Section IX of this book).

MISHAELA R. RUBIN

ET AL .

The clinical applications of these newer properties of vitamin D, however, have also been tempered by the potential for complications, such as hypercalcemia and hypercalciuria, prompting the development of other analogs to distinguish even better calcemic from antiproliferative effects [42] (see also Section VIII of this book). Depending on the chemical modification of the basic structure of vitamin D, some analogs do demonstrate reduced calcemic activity, but others have been developed with increased calcemic activity owing to enhanced intestinal calcium absorption and bone mineral mobilization. Fluorination of C-24, C-26, or C-27 apparently results in markedly increased calcemic activity resulting from reduced enzymatic degradation of the side chain. Calcemic potency of 1,25(OH)2D3 and its analogs can be also enhanced at least two- to fivefold by epimerization at the C-20 site [43]. The vitamin D analogs in use for secondary hyperparathyroidism in the United States include doxercalciferol, 22-oxacalcitriol (1,25-dihydroxy22-oxavitamin D3) and paricalcitol (19-nor-1,25dihydroxyvitamin D2). Each analog retains suppressive action on PTH and parathyroid gland growth, but has less calcemic and phosphatemic activity than calcitriol. It is unclear how the analogs compare to each other, although in rats, paricalcitol is less calcemic and phosphatemic than doxercalciferol [44]. Overall, the effect of vitamin D analogs to minimize the calcium-phosphate product might reduce vascular calcification [45] and mortality in the renal failure population [46]. Of additional potential importance is the decreased likelihood of low bone turnover, or adynamic bone disease, with the use of these agents [47–48]. The mechanism for the differential actions of vitamin D analogs is not completely understood. Oxacalcitriol, for example, has a low affinity for vitamin D–binding protein, so more of the drug circulates in the free form, allowing it to be more rapidly metabolized than calcitriol [49]. This leads to a shorter half-life, which could explain the small and transient stimulation of intestinal calcium absorption. It does not, however, seem to account for the prolonged inhibition of PTH release (see Chapter 86). Other vitamin D analogs, such as topical calcipotriol (MC903), have proved very effective in the treatment of psoriasis (see Chapter 101). Because of its low absorption rate and rapid degradation, calcipotriol is believed to have negligible effects on systemic calcium homeostasis when administered topically. However, isolated cases of hypercalcemia and hypercalciuria have been reported, even in patients taking recommended doses [50]. In one investigation, Bourke and colleagues noted suppression of serum PTH concentrations in all patients within two weeks of treatment with calcipotriol. Mean serum and urine calcium

CHAPTER 78 Hypercalcemia Due to Vitamin D Toxicity

levels increased during treatment and fell following withdrawal [51]. The authors concluded that although this particular synthetic analog alters serum and urinary calcium with a dose-dependent effect on systemic calcium homeostasis, it is well tolerated and effective for mild to moderate chronic plaque psoriasis. However, it is potentially hazardous in extensive, unstable, exfoliative disease [52].

III. FORMS OF ENDOGENOUS VITAMIN D TOXICITY A. Endogenous Production of 25-Hydroxyvitamin D Hypervitaminosis D with hypercalcemia is rarely due to endogenous dysregulation of vitamin D metabolites as seen in Williams syndrome [53]. Williams syndrome, an idiopathic infantile form of hypercalcemia, is associated with late psychomotor development, selective mental deficiency, and supravalvular aortic stenosis [54]. The hypercalcemia has been reported to range widely from 12 to 19 mg/dl but usually subsides by 4 years of age. One report suggests an exaggerated production of 25OHD with small doses of vitamin D as a possible etiology of the hypervitaminosis D [53].

B. Production of 1,25-Dihydroxyvitamin D 1. GRANULOMATOUS DISEASES

In contrast to the megadosages of vitamin D that are usually required to produce vitamin D toxicity, patients with granulomatous diseases can develop hypercalcemia rather easily without excessive intake of exogenous vitamin D. They are said to be hypersensitive to vitamin D. The etiology of the vitamin D toxicity in this syndrome is due to poorly regulated extrarenal synthesis of 1,25(OH)2D by the granulomatous tissue itself (as described in detail in Chapter 79). In contrast to the various presentations of vitamin D toxicity described earlier, the responsible metabolite in granulomatous disease is quite different. In the case of vitamin D toxicity due to overdosage of vitamin D or 25OHD, 25OHD is the active metabolite; renal production of 1,25(OH)2D in this setting is highly regulated and not excessively high. In granulomatous tissue, however, 1,25(OH)2D formation is not subject to control by any recognized regulators, such as PTH, phosphorus, or calcium. Thus, this syndrome is due to ectopic production of 1,25(OH)2D by the granulomatous tissue itself. The mechanisms by which hypercalcemia occurs, however, are similar to all other vitamin D toxic

1359 states, namely, increased intestinal calcium absorption and enhanced osteoclastic bone resorption [55,56]. Many studies have led to greater understanding of the pathophysiology and immunological features associated with this syndrome. a. Sarcoidosis Abnormalities in calcium metabolism have long been noted in patients with sarcoidosis [57]. Sarcoidosis is also the most common granulomatous disease associated with hypercalcemia. Approximately 10% of patients with sarcoidosis will develop hypercalcemia, and as many as 50% will experience hypercalciuria at some time during the course of the disease [58]. Hypercalciuria is invariably present when patients develop hypercalcemia. In the 1950s, studies had already revealed similarities between hypercalcemia of sarcoidosis and the hypercalcemia of vitamin D toxicity, namely, increased intestinal absorption of calcium, hypercalciuria, and therapeutic efficacy of glucocorticoids [59,60]. The major distinguishing feature was in the amount of vitamin D associated with the hypercalcemia and/or hypercalciuria. Seasonal variation of the serum calcium level in sarcoidosis was correlated with availability of sunlight as a source of vitamin D [61]. In the late 1970s, two independent groups showed that the vitamin D-like principle that appeared to be responsible in sarcoidosis was, in fact, the active metabolite of vitamin D, 1,25(OH)2D3 [56,62]. Ectopic production of 1,25(OH)2D3 was confirmed by demonstrating high circulating concentrations of 1,25(OH)2D3 in anephric patients with sarcoidosis on hemodialysis during hypercalcemic episodes [63,64]. This observation showed unequivocally that the kidney, usually the sole source of 1,25(OH)2D3 in nonpregnant individuals, could not be the source of 1,25(OH)2D3 in these patients. The serum calcium and 1,25(OH)2D3 levels were positively correlated with indices of disease activity [65–67], namely, the extent of granuloma formation and the angiotensin-converting enzyme level. It was subsequently shown that the granulomatous tissue was, in fact, the site of 1,25(OH)2D3 production. The lα-hydroxylase enzyme responsible for formation of 1,25(OH)2D3 was present in lymph node homogenates [68]. Moreover, pulmonary alveolar macrophages [69] could be shown to catalyze the formation of an 3H-labeled 25OHD3 metabolite. This metabolite was definitively identified as 1,25(OH)2D3 by high-performance liquid chromatography (HPLC), by the chick intestinal receptor assay for 1,25(OH)2D3, by UV spectroscopy, and by mass spectrometry [70]. The production of mRNA for 1α-hydroxyalse is markedly increased in alveolar macrophages isolated from hypercalcemic patients with sarcoid [71]. Importantly, control of the macrophage 1α-hydroxylase enzyme differs

1360 from that of the renal 1α-hydroxylase. The renal 1α-hydroxylase is regulated at the level of transcription by calciotropic hormones, and is exquisitely autoregulated by 1,25(OH)2D3 itself [72]. In contrast, the macrophage 1α-hydroxylase mRNA expression is potently stimulated by inflammatory agents, such as γ-interferon [73], and shows no feedback control in response to 1,25(OH)2D3 [74]. Communication between signaling pathways of γ-interferon and the vitamin D receptor has recently been reported [75]. These mechanisms account for the uncontrolled synthesis of 1,25(OH)2D3 and the characteristic finding of increased sensitivity to vitamin D in these patients [76], so that patients even without major increases in 1,25(OH)2D3 can become hypercalemic. Conversely, abnormal 1,25(OH)2D3 metabolism has been described in some patients with sarcoidosis who are normocalciuric and normocalcemic [77]. Another property of the macrophage 1α-hydroxylase enzyme is that it is inhibited in a dose-dependent fashion by dexamethasone and chloroquine that do not influence the renal 1α-hydroxylase enzyme that catalyzes synthesis of 1,25(OH)2D3 [78]. These in vitro observations have direct clinical relevance. There are several mechanisms by which calcium metabolism is disturbed in sarcoidosis [79]. First, 1,25(OH)2D3 causes hypercalcemia, in part, by stimulating intestinal calcium absorption. A low calcium diet [80,81], alone or in association with cellulose phosphate [82], was found to normalize the calcium level in some patients with sarcoidosis. Second, 1,25(OH)2D3 directly stimulates osteoclastic-mediated bone resorption; skeletal granulomas are not required for this effect [83–85]. The increased flux of calcium into the extracellular space by these gastrointestinal and skeletal mechanisms, aided by suppression of PTH [62–64], leads to hypercalciuria. Chronic hypercalciuria favors nephrocalcinosis and renal stone formation [86]. When the kidneys are unable to excrete the calcium presented to them, because of either declining renal function, enhanced bone resorption, a sudden influx of dietary calcium, dehydration, or any combination of these events, hypercalcemia ensues [87]. Granulomatous production of PTHrP may also play a role in abnormal calcium metabolism [88], where TNFα and interleukin-6, produced by macrophages, increase PTHrP gene expression. PTHrP was reported in one series to be present in 85% of biopsies of granulomatous tissue from patients with sarcoidosis [88]. b. Tuberculosis Longitudinal studies from the United States [89] and India [90] suggested that 16 to 28% of patients with tuberculosis develop hypercalcemia. However, in these early studies, vitamin D supplements were employed, increasing the risk and severity

MISHAELA R. RUBIN

ET AL .

of hypercalcemia. A similar study from Greece [91] reported a figure as high as 48% when serum calcium was corrected to a normal albumin level. Other studies from the United Kingdom [92], Belgium [93], Hong Kong [94], and Malaysia [95] have shown a much lower prevalence of hypercalcemia, in the range of 0 to 2.3%. It is likely that hypercalcemia is not as common in tuberculosis as was previously thought [96]. This discrepancy might be attributable to regional differences in calcium and vitamin D intake, which can unmask hypercalcemia [97], along with increased sun exposure. Reports of high circulating levels of 1,25(OH)2D3 in three anephric patients with tuberculosis support an extrarenal source of the active vitamin D metabolite [98,99]. Positive correlation of the albumin-adjusted calcium level with the radiographic extent of the disease has been shown [94]. Hypercalcemia in tuberculosis may occur weeks to months after starting antituberculosis chemotherapy [89,90]. Thus, the hypercalcemia is not related to the presence of viable acid-fast bacilli, but rather to the granulomatous process and associated reactions. As with sarcoidosis, hypercalcemia in tuberculosis can be controlled by administration of glucocorticoids [100]. In patients with tuberculous pleuritis, the mean free 1,25(OH)2D3 concentration in pleural fluid was selectively concentrated by 5.3-fold over that in serum [101]. Positive correlation between the concentrations of substrate (25OHD3) and product [1,25(OH)2D3] in pleural fluid supported the idea that 1,25(OH)2D3 was produced locally by activated inflammatory cells in or adjacent to the pleural space. The pleural fluid was found to have high concentrations of γ-interferon, a cytokine known to stimulate activated macrophages in vitro to synthesize 1,25(OH)2D3 [102]. Cells obtained from bronchoalveolar lavage in patients with tuberculosis were also found to synthesize 1,25(OH)2D3 in vitro. An important source of the active vitamin D metabolite appears to be the CD8 + T lymphocytes at the granulomatous sites [103]. If one wonders etiologically about the production of 1,25(OH)2D3 under these circumstances, the immunomodulatory functions of 1,25(OH)2D3 acting as a beneficial local paracrine factor could be pertinent (see Chapter 79). Viewed in this context, hypercalcemia occurs when 1,25(OH)2D3 is produced in such quantities that it gains entry into the circulation. Hypercalcemia in tuberculosis is usually mild and asymptomatic. Besides glucocorticoids, ketoconazole administration has been associated with a rapid decline in 1,25(OH)2D3 and normalization of serum calcium levels [104]. Long-term antituberculosis therapy with isoniazid and rifampin can also be effective in treating the hypercalcemia by controlling the disease.

CHAPTER 78 Hypercalcemia Due to Vitamin D Toxicity

c. Other Granulomatous Diseases Besides the more detailed studies of hypercalcemia in tuberculosis, hypercalcemia has also been reported in other infectious diseases, including leprosy [105], coccidioidomycosis [106], histoplasmosis [107], candidiasis [108], catscratch disease [109], and pneumocystis carinii pneumonia [110]. Noninfectious associations, aside from sarcoidosis, have been reported with eosinophilic granuloma [111], berylliosis [112], silicone-induced granuloma [113], paraffin-induced granulomatosis [114], Wegener’s granulomatosis [115], Langerhans’ cell granulomatosis [116], Crohn’s disease [117], and infantile fat necrosis [118]. The mechanism of increased production of the active vitamin D metabolite is believed to be shared by all of these granulomatous disorders. Spontaneous idiopathic excess production of calcitriol in the absence of granulomatous disease has also been reported in patients with elevated angiotensin converting enzyme levels [119], presumably with increased calcitriol production by macrophages. A possible role for 1,25(OH)2D3 in these granulomatous disorders as noted above includes immunomodulatory features, which are discussed in Chapters 36 and 98. 2. LYMPHOMA

Hypercalcemia has been reported to occur in 5% [120] and 15% [121] of patients with Hodgkin’s disease and non-Hodgkin’s lymphoma (NHL), respectively. Up to 80% of patients with human T-cell leukemia virus type 1 (HTLV-l)-associated adult T-cell lymphoma/ leukemia (ATLL) will develop hypercalcemia [122]. As is the case with other malignancies, hypercalcemia is a poor prognostic feature in lymphoma [123], adding substantially to morbidity and mortality. The humoral mediators of hypercalcemia in lymphoma are multiple and heterogeneous. However, evidence has shown 1,25(OH)2D3 to be an important factor in many cases. Hodgkin’s disease is most consistently associated with 1,25(OH)2D3 when hypercalcemia develops. Since the first report of hypercalcemia complicating Hodgkin’s disease in 1956 [124], more than 60 cases have been described. In a retrospective review of the literature [125], 84% of patients had a peak serum calcium above 12 mg/dl, 74% of the patients had Ann Arbor stage III or IV disease, and 68% were symptomatic with night sweats, fever, and weight loss. Only 3 of 23 patients had radiological evidence of lytic bone lesions. In 17 hypercalcemic patients, all but one patient had an elevated 1,25(OH)2D3 level. There is no evidence to implicate parathyroid hormone-related peptide (PTHrP) as a mediator of hypercalcemia in Hodgkin’s disease. Two patients with Hodgkin’s disease [126,127] were reported to have intermittent hypercalcemia

1361 during two consecutive summers or on vitamin D challenge. There was a close association between hypercalcemia and the abnormally raised 1,25(OH)2D3 level, but serum 25OHD3 was within the normal range. These observations support the idea that the mechanism of the hypercalcemia in Hodgkin’s disease is similar to that of the granulomatous diseases, namely, production by the lymphomatous tissue of 1,25(OH)2D3. A number of cases of 1,25(OH)2D3-induced hypercalcemia in non-Hodgkin’s lymphoma have been described. Most patients had bulky or advanced stage disease, but no clinically or radiographically evident bone lesions. In one case, the 1,25(OH)2D3-mediated hypercalcemia was associated with transformation from a chronic lymphocytic leukemia to an aggressive high-grade non-Hodgkin’s lymphoma [128]. Data supporting extrarenal synthesis of 1,25(OH)2D3 are the presence of severe renal failure in a number of instances [129,130]; the demonstration of in vitro conversion of 25OHD3 to 1,25(OH)2D3 by excised lymph node homogenates [131]; the prompt decline of 1,25(OH)2D3 levels to normal after excision of an isolated splenic lymphoma [132] and a primary ovarian lymphoma [133]; and sensitivity to glucocorticoid suppression [130]. Five of ten patients with either AIDS or non-AIDS associated non-Hodgkin’s lymphoma and hypercalcemia had frankly elevated serum 1,25(OH)2D3 concentrations [134]. Other malignant lymphoproliferative diseases associated with 1,25(OH)2D3–mediated hypercalcemia include lymphomatoid granulomatosis [135], dysgerminoma [136], and an inflammatory myofibroblastic tumor [137]. In a prospective study by Seymour et al. [138], a control group was composed of 16 patients with hypercalcemia and multiple myeloma. Using the mean serum 1,25(OH)2D3 level of the control patients plus 3 standard deviations, the investigators defined the upper limit of expected serum 1,25(OH)2D3 during hypercalcemia as 42 pg/ml, well below the upper limit of 76 pg/ml for the normocalcemic reference range. Thus, the typical hypercalcemic patient, if represented by this cohort of patients with multiple myeloma, shows a lower range of normal for 1,25(OH)2D3 concentration. Of the 22 hypercalcemic patients with non-Hodgkin’s lymphoma, 12 (55%) had elevated serum 1,25(OH)2D3 levels. Moreover, the serum levels of corrected calcium and 1,25(OH)2D3 were strongly correlated with one another. Even in the normocalcemic group with non-Hodgkin’s lymphoma, 71% were hypercalciuric and 18% had elevated serum 1,25(OH)2D3 levels. The precise cell type responsible for the extrarenal synthesis of 1,25(OH)2D3 in lymphoma remains to be established. There are two possibilities. One is the tumor-infiltrating reactive macrophage, recognized by

1362 a “starry-sky” appearance [139] in intermediate and high-grade lymphomas, in which hypercalcemia is also most common. Alternatively, it may be that a particular clone of the malignant lymphoma cell synthesizes 1,25(OH)2D3 [140]. Recent immunohistochemical analysis of the enzyme 1α-hydroxylase in a B-cell lymphoma associated with hypercalcemia and raised circulating levels of 1,25(OH)2D3 suggests that the tumor itself is not a source of the steroid hormone [141]. Rather, macrophages adjacent to the tumor are likely to be the major site of ectopic 1,25(OH)2D3 synthesis [141]. 1,25-Dihydroxyvitamin D is only one cause of hypercalcemia in lymphoma. About half of the patients with non-Hodgkin’s lymphoma and hypercalcemia have suppressed 1,25(OH)2D3 levels. Additional circulating or local osteolytic factors are likely to be involved. Two of 22 patients in the study by Seymour et al. had elevated PTHrP levels. A few other cases of hypercalcemia in non-Hodgkin’s lymphoma associated with elevated levels of PTHrP have been reported [142–144]. Cytokines such as interleukin-1, tumor necrosis factor-α (TNFα), and transforming growth factor (TGFβ) may also play a role in the pathogenesis of lymphomaassociated hypercalcemia. Although HTLV-1-transformed lymphocytes were shown in vitro to possess the capacity to convert 25OHD3 to 1,25(OH)2D3 [145], most studies have shown reduced 1,25(OH)2D3 levels in hypercalcemia associated with HTLV-1-related adult T-cell leukemia/ lymphoma [146,147]. PTHrP is most strongly implicated as the major mediator in this syndrome [148]. PTHrP messenger RNA has been demonstrated in HTLV-1-infected T cells [149] and tumor cells from adult T-cell lymphoma/leukemia (ATLL) patients with hypercalcemia [150]. Nevertheless, there are two welldocumented instances of elevated 1,25(OH)2D3 levels in ATLL [122,130]. In the first case, a PTHrP level was not available. In the second case, concomitant elevation of 1,25(OH)2D3 and PTHrP was shown, suggesting the possibility of increased renal 1α-hydroxylase activity secondary to PTHrP. Alternatively, the tissue could be the site of both PTHrP and 1,25(OH)2D3 formation. Most patients with hypercalcemia due to classic squamous cell carcinoma have elevated PTHrP levels and either suppressed or normal 1,25(OH)2D3 levels.

IV. MECHANISMS OF VITAMIN D TOXICITY A. General Mechanisms Vitamin D toxicity may occur in patients due to any one of the three forms of vitamin D, namely, the vitamin D parent compound, 25OHD, or 1,25(OH)2D.

MISHAELA R. RUBIN

ET AL .

Multiple factors may influence susceptibility to vitamin D toxicity and include the concentration of the vitamin D metabolite itself, vitamin D receptor (VDR) number, activity of lα-hydroxylase, the metabolic degradation pathway, and the capacity of the vitamin D–binding protein (DBP). Vitamin D2 or D3 toxicity is more difficult to manage than toxicity due to its metabolites 25OHD or 1,25(OH)2D. In part, this is due to the extensive lipid solubility of the parent compound in liver, muscle, and fat tissues and corresponding large storage capacity. As a result, the half-life of vitamin D ranges from 20 days to months. In contrast, the biological half-life of the less lipophilic compound 25OHD is shorter, approximately 15 days [151]. The biological half-life of the least lipophilic compound 1,25(OH)2D, is much shorter, approximately 15 hr [152]. In general, duration of toxicity is related to the half-life of the vitamin D compound. Thus, the hypercalcemia of parent vitamin D overdose can last for as long as 18 months, long after dosing is discontinued, because of its slow release from fat deposits. Over-dosage of 25OHD can persist for weeks also, but excessive 1,25(OH)2D toxicity is more rapidly reversed because 1,25(OH)2D is not stored in appreciable amounts in the body [66]. The toxicity of either parent vitamin D or 25OHD is due to 25OHD. In an investigation examining the concentrations of vitamin D3 and its metabolites in the rat as influenced by various intakes of vitamin D3 or 25OHD, Shepard and DeLuca found that large intakes of vitamin D3, ranging from 0.65 to 6500 nmol/day, resulted in excessive concentrations of vitamin D3 and 25OHD3 but not in 1,25(OH)2D3 (Table II) [153]. Similarly, increased dosages of 25OHD3 ranging from 0.46 to 4600 nmol/day resulted in excessive amounts of 25OHD3, but not of vitamin D3 or 1,25(OH)2D3 (Table III). In the setting of toxicity due to overadministration of 1,25(OH)2D3, the active metabolite itself is responsible for the hypercalcemia [154]. Unlike 1,25(OH)2D whose production is tightly regulated in the kidney, the production of 25OHD is not tightly controlled by the liver. The high capacity for 25-hydroxylation of vitamin D in the liver as well as poor regulation at this site allows for massive amounts of 25OHD to be generated from large amounts of vitamin D. Thus, excessive concentrations of 25OHD are typically measured in vitamin D toxicity. As would be expected, PTH levels are suppressed in this form of hypercalcemia.

B. Role of Vitamin D Receptor (VDR) in Vitamin D Toxicity Various investigations have helped to shed light on the interrelationship among vitamin D metabolites, the

1363

CHAPTER 78 Hypercalcemia Due to Vitamin D Toxicity

TABLE II Plasma Concentrations of Vitamin D3 and Metabolites in Rats Given Various Amounts of Vitamin D3a Amount (nmol/day)

Vitamin D3 (ng/ml)

25OHD3 (ng/ml)

0.65 6.5 65 650 6500

11.3 ± 6.1 110 ± 43 368 ± 121 1339 ± 329c 3108

2.3 ± 1.9 14.7 ± 8.6 74.2 ± 14.5 643 ± 93b 1111

Lactone (ng/ml) < 0.06 0.35 ± 0.12 10.3 ± 3.9 64.5 ± 19.1c 43.6

24,25(OH)2D3 (ng/ml)

25,26(OH)2D3 (ng/ml)

1,25(OH)2D3 (pg/ml)

Plasma calcium (mg/100 ml)

0.56 ± 0.13 3.98 ± 1.90 25.5 ± 5.2 73.5 ± 29.6d 86.5

< 0.2 0.20 ± 0.36 7.60 ± 2.78 16.4 ± 4.7c 8.4

80 ± 60 77 ± 64 88 ± 9 51 ± 11c 37

9.0 ± 0.1 9.4 ± 0.4 9.7 ± 0.3 12.4 ± l.06 13.8

a Rats were orally dosed daily for 14 days with indicated amounts of vitamin D . Data are means of 5 rats ± SD. Reprinted with permission from 3 Shepard RM, DeLuca HF 1980. Arch Biochem Biophys 202:43–50. b Differs from control group (0.65 nmol/day) and from group receiving 65 nmol/day at p < 0.001. c Differs from group receiving 65 nmol/day at p < 0.001. d Differs from control group (0.65 nmol/day) at p < 0.001 and from group receiving 65 nmol/day at p < 0.010.

VDR, and PTH in vitamin D toxicity. The biologically active form of vitamin D, 1,25(OH)2D, as is typical of other steroid hormones, binds to a specific intracellular receptor protein (VDR) within its target tissues. The hormone-VDR complex then triggers subsequent transcriptional events by binding to DNA elements. Regulation of cellular VDR numbers is believed to be an important mechanism by which cellular responsiveness to 1,25(OH)2D is modulated, because the biological activity of 1,25(OH)2D is proportional both to tissue VDR number and concentration of 1,25(OH)2D (see Section II of this book for a detailed discussion). Increased VDR concentrations imply enhanced tissue responsiveness to 1,25-dihydroxyvitamin D, whereas decreased receptor numbers indicate reduced tissue responsiveness. Several investigations have suggested that exogenous 1,25(OH)2D3 can lead to homologous

TABLE III Amount (nmol/day) 0.46 4.6 46 460 4600

up-regulation of VDR in vitro and in vivo, in contrast to endogenous production of 1,25(OH)2D3. In vitro and in vivo administration of 1,25(OH)2D3 to rats has been shown to increase VDR content. In vitro exposure of human skin fibroblasts and osteosarcoma cells to 1,25(OH)2D3 has been shown to result in a three- to fivefold increase in VDR number [155]. Similarly, in vivo studies have shown increased VDR with exogenous administration of 1,25(OH)2D3. Costa and Feldman administered 1500 pmol/kg of 1,25(OH)2D3 daily to rats and found a 30% increase in intestinal VDR and a threefold increase in renal VDR concentration [156]. Reinhart et al. infused rats with 250 pmol/kg of 1,25(OH)2D3 daily for six days and noted a 22% increase in VDR levels in the intestine and a 37% increase in bone [157]. Goff and colleagues infused 36 ng of 1,25(OH)2D3 to rats over seven days and found

Plasma Concentrations of Vitamin D3 and Metabolites in Rats Given Various Amounts of 25OHD3a Vitamin D3 (ng/ml) < 0.5 < 0.5 < 0.5 < 0.5 < 0.5

25OHD3 (ng/ml) 6.2 ± 2.3 56.3 ± 11.4 199 ± 24 436 ± 58 688 ± 145b

Lactone (ng/ml) 0.31 ± 0.05 3.02 ± 0.63 32.5 ± 8.5 118 ± 26 110 ± 38c

24,25(OH)2D3 (ng/ml)

25,26(OH)2D3 (ng/ml)

2.29 ± 0.54 11.7 ± 3.3 57.3 ± 19.5 170 ± 22 214 ± 117d

< 0.2 < 0.2 1.19 ± 0.49 4.02 ± 1.02 6.31 ± 1.79e

1,25(OH)2D3 (pg/ml) 187 ± 72 192 ± 65 82 ± 29 33 ± 8 22 ± 1f

Plasma calcium (mg/100 ml) 9.8 ± 0.5 9.3 ± 0.5 10.2 ± 0.4 9.7 ± 0.4 14.0 ± 0.5*

a Rats were orally dosed daily for 14 days with indicated amounts of 25OHD . Data are means of 5 rats ± SD. Reprinted with permission from Shepard RM, 3 DeLuca HF 1980 Arch Biochem Biophys 202:43–50. b Differs from control group (0.46 nmol/day) at p < 0.001 and from group receiving 460 nmol/day at p < 0.010. c Differs from control group (0.46 nmol/day) at p < 0.001. d Differs from control group (0.46 nmol/day) at p < 0.005. e Differs from group receiving 460 nmol/day at p < 0.005. f Differs from control group (0.46 nmol/day) at p < 0.001 and from group receiving 460 nmol/day at p < 0.050. *Differs from control group (0.46 nmol/day) and from group receiving 460 nmol/day at p < 0.001.

1364

MISHAELA R. RUBIN

600

Control

p < 0.01a

Vitamin-D2-treated Vitamin-D3-treated

500

p < 0.05 Unoccupied receptor (fmol/mg protein)

a 1.5-fold increase in duodenal VDR content and a three-fold increase in renal VDR content [158]. Goff et al. [158] also demonstrated that endogenously produced 1,25(OH)2D3 has a different effect than exogenous administration of 1,25(OH)2D3 on tissue VDR content. Rats fed a calcium-restricted diet resulting in “nutritional” hyperparathyroidism achieved a similar increase in endogenous 1,25(OH)2D3 concentration as rats administered exogenous 1,25(OH)2D3. However, calcium-restricted rats failed to up-regulate VDR content in the duodenum or kidney, presumably a consequence of the negative control of VDR by PTH [159]. This point has at least conceptual relevance in the case of vitamin D toxicity. Rather than downregulation occurring during hypervitaminosis D, which is a more typical regulatory and protective event to limit tissue responsiveness, exposure of cells to exogenous 1,25(OH)2D results in enhanced responsiveness by virtue of up-regulation. Such a mechanism would be of particular clinical relevance if the toxicity were due to overexposure of 1,25(OH)2D. Moreover, in this setting, the associated suppression of PTH would prevent the regulatory mechanism from being operative. Evidence suggests that in parent vitamin D toxicity, target tissues are responding to high concentrations of 25OHD, not 1,25(OH)2D. Concentrations of 1,25(OH)2D are typically only slightly increased, if at all. The hypercalcemia is due to the effects of pharmacologically high levels of 25OHD, even though in physiological settings, 25OHD has little potency. At high concentrations, 25OHD can compete for binding at VDR sites, and thereby produce biological effects similar to those of 1,25(OH)2D on intestine and bone [160]. Beckman and colleagues [161] suggested, furthermore, that hypervitaminosis D, like excessive 1,25(OH)2D, is associated with homologous up-regulation of intestinal VDR. Their investigation demonstrated that supraphysiological amounts of vitamin D2 or vitamin D3 administered to rats at doses of 25,000 IU

ET AL .

400

300

200

100

0

Day 6 of treatment

FIGURE 1 Intestinal VDR in rats treated six days with 25,000 IU/ day of either vitamin D2 or vitamin D3 relative to the response in age-matched controls. aResponse in vitamin D3-treated rats significantly different from that in vitamin D2-treated rats (p < 0.05). Reprinted with permission from Beckman MJ, et al. [161].

daily for six days resulted in increasing plasma 25OHD concentrations with significant up-regulation of intestinal VDR concentration and hypercalcemia. Plasma 1,25(OH)2D levels were not altered substantially (see Table IV and Fig. 1). A comparison between hypervitaminosis D3 and D2 was also made [161]. No differences in 25OHD and plasma calcium concentrations were noted between either preparations. Concentrations of 25OHD in each case were markedly higher than the control group. The concentration of 1,25(OH)2D was observed to be only slightly greater in the vitamin D3-treated group than the vitamin D2-treated group. Because the 25OHD

TABLE IV Changes in Body Weight, Plasma Calcium, and Plasma Vitamin D Metabolites in Rats Treated for Six Days with Either 25,000 IU/day of Vitamin D2 or Vitamin D3a Group Control Vitamin D2-treated Vitamin D3-treated

Body weight (g) 251 ± 5 230 ± 17 201 ± 18*

Plasma calcium (mg/dl) 9.5 ± 0.7 11.8 ± 0.6c 12.0 ± 0.9c

25OHD (ng/ml)

1,25(OH)2D (pg/ml)

20 ± 2 466 ± 36a 506 ± 67b

112 ± 11 123 ± 12 150 ± 8c,d

represent means ± SE. Reprinted with permission from Beckman MJ et al,1990. Biochem Biophys Res Commun 169:910–915. difference at p < 0.01 of the treated groups relative to the control group. c Significant difference at p < 0.05 of the treated group relative to the control group. d Statistical significance between the D - and D -treated groups (n = 6). 2 3 a Data

b Significant

1365

CHAPTER 78 Hypercalcemia Due to Vitamin D Toxicity

concentrations were elevated twenty- to twenty-five-fold, whereas 1,25(OH)2D showed only minimal increases, the biochemical and clinical changes associated with parent vitamin D toxicity were attributed to 25OHD. The data provided further support for the importance of 25OHD as the major toxic metabolite in vitamin D–associated hypercalcemia, as well as for the importance of increased intestinal VDR in the pathophysiological process that leads to enhanced effects of this metabolite.

C. Control of Renal 1α-Hydroxylase in Vitamin D Toxicity Some investigators have suggested that toxic effects of excessive concentrations of 25OHD may result from PTH suppression and down-regulation of 1α-hydroxylase with increased concentrations of 25OHD. PTH and 1,25(OH)2D have known reciprocal actions on 1α-hydroxylase and 24-hydroxylase activities. PTH stimulates 1α-hydroxylase activity and down-regulates 24-hydroxylase activity; 1,25(OH)2D, on the other hand, down-regulates 1α-hydroxylase activity and stimulates 24-hydroxylase activity. Beckman and colleagues [162] studied the effects of an excess of vitamin D3 and dietary calcium restriction on tissue 1α-hydroxylase and 24-hydroxylase activity in rats. Four groups of rats with different dietary calcium and vitamin D3 concentrations were studied (normal calcium, NC; low calcium, LC; and the excess vitamin D groups with normal or low calcium, NCT and LCT). The data showed that in the setting of a calciumrestricted diet, a nutritional hyperparathyroidism ensued (Table V). Under conditions of excess vitamin D3 at doses of 75,000 IU per week and on a calcium-restricted diet, elevations in PTH facilitated the elimination of

25OHD3 through its metabolism to 1,25(OH)2D3 and or degradation to 24,25(OH)2D3. The elevation in PTH was accompanied by increased activation of renal 1α-hydroxylase activity, lower concentrations of 25OHD3, increased activation of intestinal 24-hydroxylase activity, and lower renal VDR content compared to the normal calcium group (Table VI). In contrast, the normal calcium diet in the vitamin D3 excess group contributed to the toxicity by virtue of suppressed PTH concentrations resulting in down-regulation of renal 1α-hydroxylase and decreased 24-hydroxylase activity, and, thus, higher 25OHD3 concentrations. On the other hand, dietary calcium restriction in the setting of vitamin D3 excess seemed to be protective, providing less biological stimulation due to higher PTH concentrations with reduced VDR, increased activation of both 1α-hydroxylase and 24-hydroxylase activities, greater reductions in 25OHD3 concentrations, and lower concentrations of total calcium resulting in a less toxic state. So the low calcium diet protects, not only by contributing to less hypercalcemia, but also by facilitating metabolic pathways of vitamin D inactivation.

D. Inhibition of the Catabolic Pathway of 24-Hydroxylase Others have proposed that inhibition of the enzymes that degrade the vitamin D metabolites may have a role in the pathogenesis of hypervitaminosis D. 1,25(OH)2D is a known regulator of its own catabolism and an inhibitor of its synthesis. In the kidney, intestine and other targets 1,25(OH)2D induces the enzyme 24-hydroxylase. This enzyme initiates a catabolic cascade that ultimately causes side chain oxidation, cleavage, and metabolic elimination of both 1,25(OH)2D and 25OHD, and it accounts for 35–40% of the

TABLE V Changes in Plasma Calcium, Phosphorus, PTH, and 1,25(OH)2D3 Concentrations in Response to Dietary Calcium Restriction and Vitamin D3 Excessa Treatmentb NC NCT LC LCT

Calcium (mg/dl)

Phosphorus (mg/dl)

11.2 ± 0.1 14.6 ± 0.3c 9.1 ± 0.1c 9.7 ± 0.4d,e

9.3 ± 0.7 9.5 ± 0.5 8.9 ± 0.5 9.0 ± 0.7

25OHD3 (ng/ml) 15.2 ± 1.7 443 ± 43c <1.0c 244 ± 17c,e

PTH (pg/ml) 48.0 ± 2.0 44.0 ± 3.0 162.0 ± 10.0c 154.0 ± 19.0c,e

1,25(OH)2D3 (pg/ml) 116.0 ± 7.0 48.0 ± 8.0 615.0 ± 110.0c 99.0 ± 12.0

aValues are means ± SE (n = 6). Significant differences were measured by Tukey’s multiple range test. Reprinted from Beckman MJ et al., 1995. Arch Biochem Biophys 319:535–539 with permission. bNC, 1.0–1.2% Ca, normal D ; NCT, 1.0–1.2% Ca, excess D ; LC, 0.02% Ca, normal D ; LCT, 0.02% Ca, excess D . 3 3 3 3 cSignificant difference at p < 0.001 versus NC group. dSignificant difference at p < 0.01 versus NC group. eSignificant difference at p < 0.001 versus NCT group.

1366

MISHAELA R. RUBIN

ET AL .

TABLE VI Changes in Renal VDR and Intestinal 25OHD3-Hydroxylases in Response to Dietary Calcium Restriction and Vitamin D3 Excessa Treatment

Renal VDR (fmol/mg protein) 200 ± 14 541 ± 21c 115 ± 8c 97 ± 6c,e

NC NCT LC LCT

Renal 1α-hydroxylase (pg/hr • mg tissue) 1.5 ± 0.2 1.1 ± 0.1 28.7 ± 5.6b,c 3.6 ± 0.4d,e

Renal 24-hydroxylase (pg/min • mg tissue)

Intestinal 24-hydroxylase (pg/min • mg tissue)

6±1 276 ± 15c,e ND* ND

7±1 19 ± 7 80 ± 16d 174 ± 33c,e

a Values are means ± SE (n = 6–10). Groups NC, NCT, LC, and LCT are defined in Table V. Significant differences were measured by Tukey’s multiple range test. Reprinted with permission from Beckman MJ 1995, et al. Arch Biochem Biophys 319:535–539. b ND, not detected. c Significant difference at p < 0.01 versus NC group. d Significant difference at p < 0.05 versus NC group. e Significant difference at p < 0.01 versus NCT group.

catabolism of 1,25(OH)2D [163]. The remainder of the metabolic degradation is due to other side chain oxidations and biliary clearance. Reinhart and Horst [164] initially proposed that blunting of the catabolic pathway of 1,25(OH)2D3 with high concentrations of 24,25(OH)2D3 in rat cells would competitively inhibit further inactivation of 1,25(OH)2D3, resulting in an accumulation of 1,25(OH)2D3 and toxicity. Clinical investigations of the down-regulation of rat intestinal 24-hydroxylase and its inhibition by calcitonin may help to elucidate a role of this hormone in potentiating the toxicity of vitamin D. 24-Hydroxylation is important in the inactivation of both 1,25(OH)2D3 and 25OHD3, and in the kidney is largely regulated inversely by 1α-hydroxylation [165]. In a study examining the effects of dietary calcium and vitamin D status on the regulation of intestinal 24-hydroxylase enzyme and mRNA expression, rats were fed normal or low calcium diets with variable amounts of vitamin D [166]. Half of the rats on the normal and low calcium

diets were administered pharmacological doses of vitamin D3 (25,000 IU three times weekly). Excess vitamin D3 resulted in significant elevations in plasma 25OHD3 in both calcium groups (LCT and NCT), with a much larger increase noted in the normal calcium group (NCT). Hypercalcemia was most severe in the NCT group, whereas rats in the low calcium and vitamin D3 excess group (LCT) had plasma calcium levels similar to the NC group (see Table VII). Because the NCT was accompanied by an increased calcitonin concentration compared to the LCT, the authors suggested that the increased calcitonin in the NCT group may have suppressed 24-hydroxylase activity, with resultant higher 25OHD3 and calcium concentrations [166]. This concept was further supported when rats, subjected to thyroparathyroidectomy (TPTX), which eliminated endogenous calcitonin (Table VIII), were found to have higher concentrations of 24-hydroxylase activity than the NCT group. Through inhibition of intestinal 24-hydroxylase activity, calcitonin could be associated

TABLE VII Changes in Plasma Calcium, 25OHD, 1,25(OH)2D, and Calcitonin Concentrations in Response to Dietary Calcium Restriction and Vitamin D3 Excessa Treatment NC NCT LC LCT

Calcium (mg/dl) 10.7 ± 0.2 12.6 ± 0.3b 9.7 ± 0.2b 10.5 ± 0.2c

25OHD (ng/ml) 108 ± 16 1812 ± 165* 47 ± 10c 1130 ± 62b,c

1,25(OH)2D (pg/ml) 106 ± 17 96 ± 12 459 ± 70c 188 ± 41

Calcitonin (pg/ml) 21.2 ± 1.1 36.1 ± 2.5b 15.9 ± 0.9 19.4 ± 2.3c

a Values are means ± SEM (n = 6). Groups NC, NCT, LC, and LCT are defined in Table V. Significant differences were measured by Tukey’s multiple range test after analysis of variance. Reprinted from Beckman et al. [166] with permission. b p < 0.05 versus NC group. c p < 0.05 versus NCT group.

1367

CHAPTER 78 Hypercalcemia Due to Vitamin D Toxicity

TABLE VIII Intestinal 24-Hydroxylase, Plasma Calcium, and 1,25(OH)2D in Response to Vitamin D Excess and Thyroparathyroidectomya Treatmentb NCT NCT/TPTX NCT/CT NCT/TPTX/CT

Calcium (mg/dl)

1,25(OH)2D (pg/ml)

12.9 ± 0.4 13.8 ± 0.3 9.7 ± 0.6c 10.3 ± 0.4c

86 ± 16 78 ± 9 80 ± 13 82 ± 11

24-Hydroxylase (pg/min • mg protein) 6.7 ± 0.5 14.1 ± 0.2c 5.3 ± 0.2 3.8 ± 0.2c,d

a Reprinted

from Beckman et al. [166] with permission. Controls given excess vitamin D3; NCT/TPTX, NCT animals that underwent TPTX; NCT/CT, NCT animals treated with 100 IU calcitonin 4 hr before death; NCT/TPTX/CT, NCT/TPTX animals treated with 100 IU calcitonin 4 hr before death. c Different from NCT (p < 0.001). d Different from NCT/TPTX (p < 0.001). b NCT,

with reduced turnover and catabolism of 25OHD3, thereby potentiating its toxicity. Thus, increased expression of 24-hydroxylase activity in cases of pharmacological amounts of 25OHD3 may be an important mechanism to counteract vitamin D toxicity. A key role for 24-hydroxylase in preventing the development of vitamin D toxicosis was found in a recent animal study [167]. Growing dogs given 135-fold vitamin D3 supplementation actually had a decrease in plasma 1,25(OH)2D3 levels by 40% as compared to controls, despite an increase in 1,25(OH)2D3 production. This was attributed to an upgraded catabolism of 1,25(OH)2D3 by 24-hydroxylase, as evidenced by increased gene expression of renal and intestinal 24-hydroxylase, thus providing an efficient hormonal counteraction [167].

E. Vitamin D Binding–Protein and the Level of Free Metabolite in Vitamin D Toxicity The vitamin D–binding protein (DBP) is a specific transport protein that binds large quantities of the circulating vitamin D metabolites (see Chapter 8). Similar to the situation for other steroid hormones, fatsoluble compounds, and thyroid hormones, only a small fraction of the metabolites circulate free in plasma. The binding affinity of the protein for the vitamin D metabolites is moderate and the capacity is great (only 5% of binding sites on DBP are normally occupied). In addition, the various metabolites have different binding affinities for the protein, in the following sequence: 25OHD > 24,25(OH)2D > 1,25(OH)2D [168]. Of note is the fact that the potent metabolite 1,25(OH)2D has the least affinity for DBP, but the highest affinity for the intracellular VDR that triggers subsequent transcriptional events. Therefore, freeing

bound 1,25(OH)2D metabolite from DBP could promote its entry into various tissues and promote biological activity [169]. In states of vitamin D toxicity, the presence of elevated free 1,25(OH)2D levels despite normal total 1,25(OH)2D levels suggests that 1,25(OH)2D is displaced from DBP by 25OHD, resulting in a rise of serum free calcitriol [170]. Evidence indicates that the biologically active form of the vitamin D steroid hormone is the free hormone that is accessible to cells [171]. Because of technical difficulties in measuring the free hormone, the determination of vitamin D status involves a measurement combining free vitamin D and DBP concentrations. In normal individuals, 85% of the total 1,25(OH)2D is bound to DBP, 15% is bound to albumin, and 0.4% is free [172]. However, under conditions of altered or reduced albumin and DBP concentrations, as in liver or kidney disease, the free hormone may provide different information compared to the total measured concentration of vitamin D. Theoretically, total hormone concentration in such settings may erroneously suggest deficiency of vitamin D with needless institution of replacement therapy. Bikle and colleagues noted that subjects with liver disease have reduced DBP concentrations with low total 1,25(OH)2D and 25OHD levels, whereas free forms are normal [173,174]. Similarly, in certain forms of renal disease, the concentrations of DBP and vitamin D metabolites are reduced, thus measurements of total hormone may provide an inaccurate reflection of vitamin D status. Koenig et al. [175] investigated free and total 1,25(OH)2D concentrations in subjects with renal disease. Patients with nephrotic syndrome, with varying degrees of renal failure, and on chronic hemodialysis and peritoneal dialysis were studied. The serum concentrations of total and free 1,25(OH)2D correlated well with one another in the patients with renal failure

1368 and those undergoing hemodialysis. The concentrations of DBP and 25OHD, thus, were unaffected by renal function. The concentrations of total 1,25(OH)2D accurately reflected free 1,25(OH)2D in patients with varying degrees of renal failure when DBP levels remained normal. However, this did not hold true for the subjects with nephrotic syndrome or those on chronic peritoneal dialysis, who lost DBP and bound vitamin D metabolites into the urine or peritoneal fluid, respectively, with a rise in the percentage free 1,25(OH)2D (also megalin may play a role as discussed in Chapter 10). Measurement of free metabolite in these particular patients may be important to avoid vitamin D toxicity when supplementation is instituted. Thus, in this context, the binding proteins of the vitamin D metabolites not only serve a transport function, but also may provide a buffering mechanism to protect against toxicity [176].

V. CLINICAL MANIFESTATIONS The clinical manifestations of vitamin D toxicity resulting from hypercalcemia reflect the essential role of calcium in many tissues and targets, including bone, the cardiovascular system, nerves, and cellular enzymes. Initial signs and symptoms of hypervitaminosis D may be similar to other hypercalcemic states and include generalized weakness and fatigue. Central nervous system features may include confusion, difficulty in concentration, drowsiness, apathy, and coma [177]. Neuropsychiatric symptoms include depression and psychosis, which resolve following improvement of the hypercalcemia. Hypercalcemia can affect the gastrointestinal tract and cause anorexia, nausea, vomiting, and constipation. It can induce hypergastrinemia, but only in men does it appear to be associated with peptic ulcer disease. There is no evidence that peptic ulcers are more common in any other form of hypercalcemia. Rarely, pancreatitis may be a presentation of either acute or chronic hypercalcemia. In the heart, hypercalcemia may shorten the repolarization phase of conduction reducing the Q-T interval on the electrocardiogram (EKG). EKG changes in vitamin D toxicity have been mistaken for myocardial ischemia [178]. A more accurate EKG indication of the level of hypercalcemia is the Q-T interval corrected for rate. Bradyarrhythmias and first degree heart block have been described, but are rare. Hypercalcemia may potentiate the action of digitalis on the heart [179]. Kidney function is affected because high concentrations of calcium alter the action of vasopressin on the renal tubules. The net result is reduced urinary

MISHAELA R. RUBIN

ET AL .

concentrating ability and a form of nephrogenic diabetes insipidus. This usually presents as polyuria, but rarely is the volume as high as that associated with central diabetes insipidus. Symptoms may include polydipsia, which is an expected consequence of polyuria. Hypercalciuria is one of the earliest signs of vitamin D toxicity and precedes the occurrence of hypercalcemia. The initial hypercalciuria may be ameliorated as renal failure progresses because of reduced calcium clearance. The pathophysiology of hypercalcemia can be rapidly worsened when dehydration develops. When reduced renal blood flow occurs, less calcium is presented to the renal glomerulus, and hypercalcemia can rapidly progress. Renal impairment from the hypercalcemia is reversible if of short duration. Chronic, uncontrolled hypercalcemia can lead to deposition of calcium phosphate salts in the kidney and permanent damage with eventual nephrocalcinosis. In an investigation of vitamin D–induced nephrocalcinosis, Scarpelli and colleagues [180] noted that cell damage, specifically in mitochondria, preceded intracellular calcium deposition. The hypercalcemia induced in rats by excessive vitamin D administration caused mitochondrial swelling, cell injury, and subsequent calcification. Ectopic soft tissue calcification can be a particular problem in hypervitaminosis D. The tendency towards soft tissue calcification is compounded by the combination of hypercalcemia and hyperphosphatemia, often exceeding the solubility product of the two ions [181–183]. In rats exposed to excessive vitamin D, Hass and colleagues demonstrated that the pathological processes of vitamin D toxicity were related to dosage, length of time between doses, and duration of exposure [184]. For rats subjected to sublethal doses, generalized calcinosis was seen after only eight days, when a total of 300,000 units of ergosterol was administered. Pathologically, bones appeared more brittle than normal, with increased cortical bone resorption, increased numbers of osteoclasts, and reduced numbers of osteoblasts. Abnormal calcium deposits were noted in the aorta and its major branches, heart, kidney, muscle, and respiratory tract. The earliest evidence of hypervitaminosis D was in the proximal aorta. Muscle tissue was the least resistant to calcification, with the order of decreasing susceptibility being smooth muscle > cardiac muscle > skeletal muscle [185]. The liver, brain, and pituitary were not affected by high doses of vitamin D. Permanent dental changes have also been reported with hypervitaminosis D, including enamel hypoplasia and focal pulp calcification [186]. Bone mineral density can be decreased due to excessive bone resorption [182,187], changes which can be reversed when vitamin D levels return to normal [188].

1369

CHAPTER 78 Hypercalcemia Due to Vitamin D Toxicity

VI. Diagnosis of Vitamin D Toxicity With modern assays for calciotropic hormones, PTH, 25OHD, and 1,25(OH)2D (see Chapter 58), one can readily differentiate vitamin D metabolite-mediated hypercalcemia from other causes of hypercalcemia. The circulating intact PTH level, preferably measured by the two-site immunoradiometric assay (IRMA) or immunochemiluminometric assay (ICMA), should be suppressed in virtually all hypercalcemic disorders with the exception of primary hyperparathyroidism, familial hypocalciuric hypercalcemia, administration of lithium or thiazides, and renal failure. Although patients with malignancy-associated hypercalcemia tend to have a higher serum calcium concentration than those with other causes of hypercalcemia, diminished glomerular filtration rate and subsequent reduction in renal calcium excretion can dramatically increase the serum calcium level in any hypercalcemic patient. In contrast to the low serum phosphorus level in patients with hypercalcemia due to PTH or PTHrP, the serum phosphorus level is at the upper limit of normal or frankly elevated in patients with vitamin D metabolite-mediated hypercalcemia. This is due to increased intestinal absorption and reduced renal clearance of phosphate. An elevated 25OHD concentration with normal 1,25(OH)2D level is indicative of toxicity with exogenously administered vitamin D or 25OHD. The serum 1,25(OH)2D level may be normally increased in patients with primary hyperparathyroidism due to the induction of renal 1α-hydroxylase by PTH. Abnormally high 1,25(OH)2D levels, in the setting of suppressed PTH and hypercalcemia, indicate dysregulated production of 1,25(OH)2D due to either granulomatous diseases, lymphoma, or toxicity with exogenous 1,25(OH)2D or lα-OHD. In cases of hypercalcemia due to PTHrP or local osteolytic factors, the serum 1,25(OH)2D concentration is usually suppressed. In patients with hypercalcemia due to toxicity with other vitamin D analogs such as dihydrotachysterol (DHT) [189] and calcipotriol, the active metabolites may not be recognized by the conventional competitive protein binding assays for 1,25(OH)2D. The diagnosis of vitamin D toxicity can be made on clinical grounds. Detailed clinical and drug history are of paramount importance in order to make an early diagnosis. Most patients who are suffering from vitamin D toxicity are taking vitamin D for osteoporosis, hypoparathyroidism, pseudohypoparathyroidism, hypophosphatemia, osteomalacia, or renal osteodystrophy in excessive dosages or at too frequent dosing intervals. Therefore, one should have a high index of suspicion in patients who are being treated with pharmacological dosages of vitamin D or its metabolites.

Patients with granulomatous diseases or lymphoma usually have widespread active disease when hypercalcemia develops. In such cases, the diagnosis is obvious at the time of presentation. However, exceptions do exist. In patients with unexplained hypercalcemia, if the 1,25(OH)2D level is elevated and other more easily identifiable causes for this elevation such as primary hyperparathyroidism, pregnancy, and exogenous toxicity (by history) are excluded, measurement of angiotensin converting enzyme level and a systemic search for lymph node enlargement, pulmonary, renal, hepatosplenic, ocular, central nervous system, and bone marrow granulomas or lymphoma should be made.

VII. Treatment of Vitamin D Toxicity Dietary calcium and vitamin D restriction and avoidance of exposure to sunlight and other ultraviolet light sources should be advised to patients at high risk to develop vitamin D metabolite-mediated hypercalcemia. Those at risk include patients with granulomatous diseases and lymphoma whose disease is widespread and active and patients who are already hypercalciuric. Daily dietary calcium intake should be minimized to approximately 400 mg or less in these patients. Any use of vitamin D supplements should be discontinued. The patient should be encouraged to use sunscreen [sun protection factor (SPF) >15] as much as possible when out of doors. The calcium level should be monitored closely in patients who have a previous history of hypercalcemia or hypercalciuria, or who have recently taken diets enriched in vitamin D and calcium, or who have a recent history of excessive sunlight exposure. A reduction in oxalate intake may also be advisable, so as to prevent an increase in oxalate absorption and hyperoxaluria, which may increase the risk of kidney stone formation, despite a reduction in urinary calcium excretion [190]. When hypercalcemia develops, the aforementioned preventive measures will help to ameliorate the severity of hypercalcemia. General measures in those who are symptomatic include hydration with normal saline and the judicious use of a loop diuretic, like furosemide. Specific inhibitors of bone resorption, such as bisphosphonates [182,187] and calcitonin, can be helpful. Recently, a 3-month-old infant with vitamin D intoxication due to oversupplementation (serum calcium 18.5 mg/dl and 25OHD 360 ng/ml) was treated with alendronate (5–10 mg/d) for 18 days with resolution of the hypercalcemia [22]. Glucocorticoids have proved to be particularly effective in vitamin D intoxication, granulomatous diseases, and lymphoma (see also Chapter 73). The precise

1370

VIII. EVIDENCE FOR BENEFITS OF HIGHER VITAMIN D LEVELS A. Bone Health Recent evidence indicates that the accepted threshold of vitamin D sufficiency might not be satisfactory, a shift in paradigm which could conceivably lead towards more zealous vitamin D replacement. Most of the daily vitamin D requirements in a healthy individual (3000–5000 IU) are met with cutaneously synthesized accumulations from solar sources during the preceding summer [199]. However, summer sun exposure is probably not sufficient. In 26 healthy men,

Cumulative probability of fracture

mechanism of action of glucocorticoids in calcium homeostasis is not known. Nonetheless, they are useful because they (1) directly inhibit gastrointestinal absorption of calcium by decreasing the synthesis of calcium-binding protein (calbindin-D) and decreasing active transcellular transport [191], (2) increase urinary excretion of calcium [192], and (3) may alter hepatic vitamin D metabolism to favor the production of inactive vitamin D metabolites, resulting in lower concentrations of 25OHD [193]. Evidence also suggests that they may increase the degradation of 1,25(OH)2D at the receptor sites [194]. Glucocorticoids may also limit osteoclastic bone resorption [195]. Institution of glucocorticoid therapy results in prompt decline of the circulating 1,25(OH)2D concentrations within 3 to 4 days [66]. Patients with nonhematological malignancies and those with primary hyperparathyroidism do not usually respond to glucocorticoids. Aminoquinolones (chloroquine and hydroxychloroquine) are also capable of reducing the 1,25(OH)2D and calcium concentrations in patients with sarcoidosis [196]. The theoretical advantage of aminoquinolones over glucocorticoids is that correction of the 1,25(OH)2D should result in rapid recovery of at least some of the bone density lost to the disease [188]. In lymphoma cells, however, aminoquinolones do not have the same regulatory effects on the excess 1,25(OH)2D as they do in granulomatous disease. In the presence of lymphoma, it is preferable to use steroid-containing antitumor regimens [198]. Owing to the limited experience with aminoquinolone drugs as antihypercalcemic agents and their potential side effects, they should be reserved for patients in whom steroid therapy is unsuccessful or specifically contraindicated. Ketoconazole, an antifungal agent, in high dosages can inhibit the mitochondrial cytochrome P450-linked 25OHD 1α-hydroxylase irrespective of whether it is renal [189] or extrarenal as in sarcoidosis [197] and tuberculosis [104].

MISHAELA R. RUBIN

ET AL .

0.14 Vitamin D 0.12 Placebo 0.10 0.08 0.06 0.04 0.02 0

0

10

20

30

40

50

60

70 Months

FIGURE 2 Cumulative probability of any first fracture in a randomized controlled trial according to treatment with vitamin D (n =1345) or placebo (n = 1341), based on Cox regression; difference between two groups, p = 0.04. Reproduced with permission from Trivedi, et al. [201].

25OHD levels went from 122 nmol/L in late summer to 74 nmol/L in late winter [200]. Furthermore, 25OHD levels that are well above the bottom end of the conventional reference range are probably, in fact, not optimal. Heaney and colleagues recently demonstrated that 25OHD needs to be at least 80 nmol/L (32.4 ng/ml) to maximize intestinal calcium absorption [14]. This finding was validated by the subsequent results of a large randomized controlled trial (n = 2686), in which treatment with vitamin D in a dose sufficient to raise serum 25OHD from 53 to 74 nmol/L decreased fracture risk at hip, forearm, or spine by 33% ( p = 0.02) (see Fig. 2) [201]. Precisely how much more vitamin D supplementation might be necessary has not been determined, although some estimates can be made. It is known that an eight week course of additional vitamin D at 400 IU daily will raise 25OHD by 11 nmol/L [1] and that maintenance of a normal PTH in the absence of sun exposure requires 1000 IU of daily vitamin D [202]. It thus remains to be seen whether the recommended daily allowance (RDA) for vitamin D will be increased and whether this will enhance the potential for vitamin D toxicity (see also Chapters 61 and 62).

B. Cellular Health Once thought to exert its effects solely on bone, kidney, and intestine, 1,25(OH)2D and its synthetic analogs are increasingly recognized to possess a wider variety of noncalcemic roles, including antiproliferative, prodifferentiative, and immunomodulatory actions. It has now been ascertained that prostate, colon, skin, and

1371

CHAPTER 78 Hypercalcemia Due to Vitamin D Toxicity

osteoblasts can express 1α-hydroxylase and synthesize 1,25(OH)2D locally [24, 203]. An example of the clinical ramifications of the nonclassic actions of 1,25(OH)2D is the efficacy of calcipotriene in treating psoriasis [4]. Another novel potential action of vitamin D might be to increase bone formation. A highly potent 1,25(OH)2D analog (2MD) was recently shown to have an anabolic effect in the rodent skeleton [204] (Chapter 87). The noncalcemic actions of 1,25(OH)2D, like the calcemic ones, appear to be mediated through the VDR. One constraint in the use of vitamin D for cellular health has been the dose-limiting hypercalcemia, although this has been partially circumvented with the use of the newer synthetic vitamin D analogs. For example, paricalcitol, a less-calcemic vitamin D analog, has been found in vitro to inhibit proliferation of myeloid leukemia, myeloma, and colon cancer cells [27]. As with 1,25(OH)2D, the VDR controls most of the effects of the synthetic analogs on proliferation, apoptosis, differentiation, and angiogenesis [205]. The antiproliferative effects of 1,25(OH)2D have been demonstrated directly in the prostate and colon and indirectly in the parathyroid. Human prostate cancer cells contain receptors for 1,25(OH)2D and respond to vitamin D in vitro with increases in differentiation and apoptosis and decreases in proliferation, invasiveness, and metastases [26]. Epidemiologically, an association has been observed between decreased sun exposure or vitamin D deficiency and an increased risk of prostate cancer at an earlier age [26]. In a small clinical trial, 22 patients with prostate cancer recurrence were treated with calcitriol 0.5 µg/kg once weekly for 10 months with only transient hypercalcemia. This strategy of intermittent dosing apparently allows very high doses of calcitriol to be administered without hypercalcemic side-effects, although the primary efficacy endpoint of a 50% reduction in PSA was not achieved [206]. In the colon, similar in vitro evidence indicates that cultured transformed colon cancer cells can convert 25OHD to 1,25(OH)2D [24]. mRNA for 1αhydroxylase has been identified in normal colon tissue and in malignant and adjacent normal colon tissue [24]. As in prostate cancer, epidemiologic data suggest that the risk of dying from colorectal cancer is highest in areas with the least amount of sunlight (see Chapter 90). Finally, recent evidence suggests that local production of 1,25(OH)2D regulates parathyroid cell growth and differentiation. The production of 1αhydroxylase has been detected in parathyroid tissue, but at higher levels in adenomas and hyperplastic tissue. This implies that in addition to feedback control by circulating 1,25(OH)2D levels, parathyroid cells may also be influenced by local 1α-hydroxylase activity with possible growth controlling effects [207].

1,25(OH)2D is known to exert a potent immunomodulatory effect on activated human lymphocytes in vitro. In fact, it has been proposed that 1,25(OH)2D produced by macrophages in granulomatous disease exerts a paracrine immunoinhibitory effect on neighboring, activated lymphocytes to slow an otherwise overly exuberant immune response that may be detrimental to the host [208]. The physiological significance of this has been highlighted by the recent development of 1α-hydroxylase knockout mouse models [209,210], which present with multiple enlarged lymph nodes (see Chapter 67). An additional immunomodulatory action of vitamin D is inhibition of the autoimmune reaction targeted towards the β cells of the pancreas. In nonobese diabetic (NOD) mice, a murine model of human type I diabetes mellitus, 1,25(OH)2D pretreatment decreased the incidence of type 1 diabetes [211] (see Chapter 99). More recently, a vitamin D–sufficient status alone was shown to confer partial protection against the development of type I diabetes mellitus in NOD mice [212]. These observations appear to have direct clinical relevance. The risk of type 1 diabetes mellitus was reduced by 80% in children treated with 2000 IU vitamin D daily after age 1 [25]. These noncalcemic actions of vitamin D thus have many potential pharmacologic applications; whether this will enhance the potential for vitamin D toxicity remains to be seen.

VIII. SUMMARY AND CONCLUSIONS Vitamin D toxicity is not a common cause of hypercalcemia, but it can be life threatening if not identified quickly. The major causes of hypercalcemia are primary hyperparathyroidism and malignancy. If these two etiologies are excluded, vitamin D toxicity becomes an important diagnostic consideration. There are many forms of exogenous and endogenous vitamin D toxicity. Inadvertent excessive use of pharmaceutical preparations is the most common etiology of exogenous toxicity. Excessive amounts of the parent compound, vitamin D, can be most difficult to manage as compared to toxicity due to the metabolites 25OHD or 1,25(OH)2D. Extensive lipid solubility of vitamin D accounts for its extraordinary half-life and tendency for prolonged hypercalcemia. New clinical applications of 1,25(OH)2D and its synthetic analogs have been accompanied by the increased potential for toxicity. Endogenous etiologies may result from ectopic production of 1,25(OH)2D in granulomatous diseases, such as sarcoidosis and tuberculosis, or in lymphoma. Many different mechanisms have been proposed to account for vitamin D toxicity, including the vitamin D metabolite itself, VDR number, activity of 1α-hydroxylase,

1372 inhibition of vitamin D metabolism, and the capacity of DBP. Mounting evidence that higher levels of vitamin D may have beneficial effects on bone and cellular health may predispose to enhanced administration of vitamin D in the future and thereby increased frequency of vitamin D toxicity.

Acknowledgment This review was facilitated, in part, by support from a grant from the National Institutes of Health (DK 32333).

References 1. Barger-Lux MJ, Heaney RP, Dowell S, Chen TC, Holick MF 1998 Vitamin D and its major metabolites: serum levels after graded oral dosing in healthy men. Osteoporos Int 8(3): 222–230. 2. Vieth R 1990 The mechanisms of vitamin D toxicity. Bone Miner 11(3):267–272. 3. Jacobus CH, Holick MF, Shao Q, Chen TC, Holm IA, Kolodny JM, et al. 1992 Hypervitaminosis D associated with drinking milk. N Engl J Med 326(18):1173–1177. 4. Holick MF 1998 Clinical efficacy of 1,25-dihydroxyvitamin D3 and its analogs in the treatment of psoriasis. Retinoids 14:12–17. 5. Koutkia P, Chen TC, Holick MF 2001 Vitamin D intoxication associated with an over-the-counter supplement. N Engl J Med 345(1):66–67. 6. Lagman R, Walsh D 2003 Dangerous nutrition? Calcium, vitamin D, shark cartilage nutritional supplements and cancerrelated hypercalcemia. Support Care Cancer 11(4):232–235. 7. Haddad JG, Chyu KJ 1971 Competitive protein-binding radioassay for 25-hydroxycholecalciferol. J Clin Endocrinol Metab 33(6):992–995. 8. Tangpricha V, Koutkia P, Rieke SM, Chen TC, Perez AA, Holick MF 2003 Fortification of orange juice with vitamin D: a novel approach for enhancing vitamin D nutritional health. Am J Clin Nutr 77(6):1478–1483. 9. Vieth, R, Pinto TR, Reen BS, Wong MM 2002 Vitamin D poisoning by table sugar. Lancet 359(9307):672. 10. Horst RL, Littledike ET, Riley JL, Napoli JL 1981 Quantitation of vitamin D and its metabolites and their plasma concentrations in five species of animals. Anal Biochem 116(1):189–203. 11. Harrington DD, Page EH 1983 Acute vitamin D3 toxicosis in horses: case reports and experimental studies of the comparative toxicity of vitamins D2 and D3. J Am Vet Med Assoc 182(12):1358–1369. 12. Hunt RD, Garcia FG, Walsh RJ 1972 A comparison of the toxicity of ergocalciferol and cholecalciferol in rhesus monkeys (Macaca mulatta). J Nutr 102(8):975–986. 13. Trang HM, Cole DE, Rubin LA, Pierratos A, Siu S, Vieth R 1998 Evidence that vitamin D3 increases serum 25-hydroxyvitamin D more efficiently than does vitamin D2. Am J Clin Nutr 68(4):854–858. 14. Heaney RP, Dowell MS, Hale CA, Bendich A 2003 Calcium absorption varies within the reference range for serum 25-hydroxyvitamin D. J Am Coll Nutr 22(2):142–146.

MISHAELA R. RUBIN

ET AL .

15. Passeri G, Pini G, Troiano L, Vescovini R, Sansoni P, Passeri M, et al. 2003 Low vitamin D status, high bone turnover, and bone fractures in centenarians. J Clin Endocrinol Metab 88(11):5109–5115. 16. Heaney RP, 2003 Vitamin D, nutritional deficiency, and the medical paradigm. J Clin Endocrinol Metab 88(11): 5107–5108. 17. Parfitt AM, Gallagher JC, Heaney RP, Johnston CC, Neer R, Whedon GD 1982 Vitamin D and bone health in the elderly. Am J Clin Nutr 36(5 Suppl):1014–1031. 18. Vieth R, Chan PC, MacFarlane GD 2001 Efficacy and safety of vitamin D3 intake exceeding the lowest observed adverse effect level. Am J Clin Nutr 73(2):288–294. 19. Administration, F.a.D, Over the Counter Report. 1979, Federal Register, p. 16164–16169. 20. Markestad T, Hesse V, Siebenhuner M, Jahreis G, Aksnes L, Plenert W, et al. 1987 Intermittent high-dose vitamin D prophylaxis during infancy: effect on vitamin D metabolites, calcium, and phosphorus. Am J Clin Nutr 46(4):652–658. 21. Zeghoud F, Ben-Mekhbi H, Djeghri N, Garabedian M 1994 Vitamin D prophylaxis during infancy: comparison of the long-term effects of three intermittent doses (15, 5, or 2.5 mg) on 25-hydroxyvitamin D concentrations. Am J Clin Nutr 60(3):393–396. 22. Bereket A, Erdogan T 2003 Oral bisphosphonate therapy for vitamin D intoxication of the infant. Pediatrics 111(4 Pt 1): 899–901. 23. Tilyard MW, Spears GF, Thomson J, Dovey S 1992 Treatment of postmenopausal osteoporosis with calcitriol or calcium. N Engl J Med 326(6):357–362. 24. Tangpricha V, Flanagan JN, Whitlatch LW, Tseng CC, Chen TC, Holt PR, et al. 2001 25-hydroxyvitamin D-1α-hydroxylase in normal and malignant colon tissue. Lancet 357(9269): 1673–1674. 25. Hypponen E, Laara E, Reunanen A, Jarvelin MR, Virtanen SM 2001 Intake of vitamin D and risk of type 1 diabetes: a birthcohort study. Lancet 358(9292):1500–1503. 26. Chen TC, Holick MF 2003 Vitamin D and prostate cancer prevention and treatment. Trends Endocrinol Metab 14(9): 423–430. 27. Kumagai T, O’Kelly J, Said JW, Koeffler HP 2003 Vitamin D2 analog 19-nor-1,25-dihydroxyvitamin D2: antitumor activity against leukemia, myeloma, and colon cancer cells. J Natl Cancer Inst 95(12):896–905. 28. Massry SG, Goldstein DA 1979 Is calcitriol [1,25(OH)2D3] harmful to renal function? JAMA 242(17):1875–1876. 29. Aloia JF, Vaswani A, Yeh JK, Ellis K, Yasumura S, Cohn SH 1988 Calcitriol in the treatment of postmenopausal osteoporosis. Am J Med 84(3 Pt 1):401–408. 30. Christiansen C, Rodbro P, Christensen MS, Hartnack B, Transbol I 1978 Deterioration of renal function during treatment of chronic renal failure with 1,25-dihydroxycholecalciferol. Lancet 2(8092 Pt 1):700–703. 31. Slatopolsky E, Weerts C, Thielan J, Horst R, Harter H, Martin KJ 1984 Marked suppression of secondary hyperparathyroidism by intravenous administration of 1,25-dihydroxy-cholecalciferol in uremic patients. J Clin Invest 74(6):2136–2143. 32. Caravaca F, Cubero JJ, Jimenez F, Lopez JM, Aparicio A, Cid MC, et al. 1995 Effect of the mode of calcitriol administration on PTH-ionized calcium relationship in uremic patients with secondary hyperparathyroidism. Nephrol Dial Transplant 10(5):665–670. 33. Monier Faugere MC, Malluche HH 1994 Calcitriol pulse therapy in patients with end-stage renal failure. Curr Opin Nephrol Hypertens 3(6):615–619.

CHAPTER 78 Hypercalcemia Due to Vitamin D Toxicity

34. Juergensen P, Santacroce S, Mooraki A, Cooper K, Finkelstein FO, Kliger AS 1994 Pulse oral calcitriol to treat hyperparathyroidism in 43 CAPD patients. Adv Perit Dial 10:259–260. 35. Feinstein S, Algur N, Drukker A 1994 Oral pulse therapy with vitamin D3 for control of secondary hyperparathyroidism. Pediatr Nephrol 8(6):724–726. 36. Crocker JF, Muhtadie SF, Hamilton DC, Cole DE 1985 The comparative toxicity of vitamin D metabolites in the weanling mouse. Toxicol Appl Pharmacol 80(1):119–126. 37. Brickman AS, Coburn JW, Friedman GR, Okamura WH, Massry SG, Norman AW 1976 Comparison of effects of 1 alphahydroxy-vitamin D3 and 1,25-dihydroxy-vitamin D3 in man. J Clin Invest 57(6):1540–1547. 38. Kanis JA, Russell RG, Smith R 1977 Physiological and therapeutic differences between vitamin D, its metabolites, and analogs. Clin Endocrinol (Oxf) 7(Suppl.):191s–201s. 39. Mortensen JT, Brinck P, Binderup L 1993 Toxicity of vitamin D analogs in rats fed diets with standard or low calcium contents. Pharmacol Toxicol 72(2):124–127. 40. Gallagher JC, Bishop CW, Knutson JC, Mazess RB, DeLuca HF 1994 Effects of increasing doses of 1α-hydroxyvitamin D2 on calcium homeostasis in postmenopausal osteopenic women. J Bone Miner Res 9(5):607–614. 41. Sjoden, G, Smith C, Lindgren U, DeLuca HF 1985 1 alphahydroxyvitamin D2 is less toxic than 1α-hydroxyvitamin D3 in the rat. Proc Soc Exp Biol Med 178(3):432–436. 42. Pols HA, Birkenhager JC, van Leeuwen JP 1994 Vitamin D analogs: from molecule to clinical application. Clin Endocrinol (Oxf) 40(3):285–292. 43. Bouillon R, Okamura WH, Norman AW 1995 Structurefunction relationships in the vitamin D endocrine system. Endocr Rev 16(2):200–257. 44. Slatopolsky, E, Finch J, Brown A 2003 New vitamin D analogs. Kidney Int 85(Suppl.):83–87. 45. Salusky IB, Goodman WG 2002 Cardiovascular calcification in end-stage renal disease. Nephrol Dial Transplant 17(2):336–339. 46. Teng M, Wolf M, Lowrie E, Ofsthun N, Lazarus JM, Thadhani R 2003 Survival of patients undergoing hemodialysis with paricalcitol or calcitriol therapy. N Engl J Med 349(5):446–456. 47. Malluche HH, Monier Faugere MC, Koszewski NJ 2002 Use and indication of vitamin D and vitamin D analogs in patients with renal bone disease. Nephrol Dial Transplant 17 10(Suppl.):6–9. 48. Cannata Andia JB, 2000 Adynamic bone and chronic renal failure: an overview. Am J Med Sci 320(2):81–84. 49. Brown AJ, Finch J, Grieff M, Ritter C, Kubodera N, Nishii Y, et al. 1993 The mechanism for the disparate actions of calcitriol and 22-oxacalcitriol in the intestine. Endocrinology 133(3):1158–1164. 50. Hardman KA, Heath DA, Nelson HM 1993 Hypercalcaemia associated with calcipotriol (Dovonex) treatment. Bmj 306(6882):896. 51. Bourke JF, Berth-Jones J, Mumford R, Iqbal SJ, Hutchinson PE 1994 High dose topical calcipotriol consistently reduces serum parathyroid hormone levels. Clin Endocrinol (Oxf) 41(3): 295–297. 52. Dwyer C, Chapman RS 1991 Calcipotriol and hypercalcemia. Lancet 338(8769):764–765. 53. Taylor AB, Stern PH, Bell NH 1982 Abnormal regulation of circulating 25-hydroxyvitamin D in the Williams syndrome. N Engl J Med 306(16):972–975. 54. Ewart AK, Morris CA, Atkinson D, Jin W, Sternes K, Spallone P, et al. 1993 Hemizygosity at the elastin locus in a developmental disorder, Williams syndrome. Nat Genet 5(1):11–16.

1373 55. Omdahl, J, Holick M, Suda T, Tanaka Y, DeLuca HF 1971 Biological activity of 1,25-dihydroxycholecalciferol. Biochemistry 10(15):2935–2940. 56. Papapoulos SE, Clemens TL, Fraher LJ, Lewin IG, Sandler LM, O’Riordan JL 1979 1,25-dihydroxycholecalciferol in the pathogenesis of the hypercalcemia of sarcoidosis. Lancet 1(8117): 627–630. 57. Harrel G, Fisher S 1939 Blood chemical changes in Boeck’s sarcoid with particular reference to protein, calcium, and phosphate values. J Clin Invest 18:687–693. 58. Studdy PR, Bird R, Neville E, James DG 1980 Biochemical findings in sarcoidosis. J Clin Pathol 33(6):528–533. 59. Anderson J, Harper C, Dent CE, Philpot GR 1954 Effect of cortisone on calcium metabolism in sarcoidosis with hypercalcemia; possibly antagonistic actions of cortisone and vitamin D. Lancet 267(6841):720–724. 60. Albright F, Carroll EL Dempsey EF, Henneman PH 1956 The cause of hypercalcuria in sarcoid and its treatment with cortisone and sodium phytate. J Clin Invest 35(11):1229–1242. 61. Taylor RL, Lynch HJ, Jr., and Wysor WG, Jr. 1963 Seasonal influence of sunlight on the hypercalcemia of sarcoidosis. Am J Med 34:221–227. 62. Bell NH, Stern PH, Pantzer E, Sinha TK, DeLuca HF 1979 Evidence that increased circulating 1α,25-dihydroxyvitamin D is the probable cause for abnormal calcium metabolism in sarcoidosis. J Clin Invest 64(1):218–225. 63. Barbour GL, Coburn JW, Slatopolsky E, Norman AW, Horst RL 1981 Hypercalcemia in an anephric patient with sarcoidosis: evidence for extrarenal generation of 1,25-dihydroxyvitamin D. N Engl J Med 305(8):440–443. 64. Maesaka JK, Batuman V, Pablo NC, Shakamuri S 1982 Elevated 1,25-dihydroxyvitamin D levels: occurrence with sarcoidosis with end-stage renal disease. Arch Intern Med 142(6):1206–1207. 65. Adams JS, Gacad MA, Anders A, Endres DB, Sharma OP 1986 Biochemical indicators of disordered vitamin D and calcium homeostasis in sarcoidosis. Sarcoidosis 3(1):1–6. 66. Adams JS, 1989 Vitamin D metabolite-mediated hypercalcemia. Endocrinol Metab Clin North Am 18(3):765–778. 67. Meyrier A, Valeyre D, Bouillon R, Paillard F, Battesti JP, Georges R 1985 Resorptive versus absorptive hypercalciuria in sarcoidosis: correlations with 25-hydroxy vitamin D3 and 1,25-dihydroxy vitamin D3 and parameters of disease activity. Q J Med 54(215):269–281. 68. Mason RS, Frankel T, Chan YL, Lissner D, Posen S 1984 Vitamin D conversion by sarcoid lymph node homogenate. Ann Intern Med 100(1):59–61. 69. Adams JS, Sharma OP, Gacad MA, Singer FR 1983 Metabolism of 25-hydroxyvitamin D3 by cultured pulmonary alveolar macrophages in sarcoidosis. J Clin Invest 72(5): 1856–1860. 70. Adams JS, Singer FR, Gacad MA, Sharma OP, Hayes MJ, Vouros P, et al. 1985 Isolation and structural identification of 1,25-dihydroxyvitamin D3 produced by cultured alveolar macrophages in sarcoidosis. J Clin Endocrinol Metab 60(5): 960–966. 71. Inui N, Murayama A, Sasaki S, Suda T, Chida K, Kato S, et al. 2001 Correlation between 25-hydroxyvitamin D3 1αhydroxylase gene expression in alveolar macrophages and the activity of sarcoidosis. Am J Med 110(9):687–693. 72. Kong XF, Zhu XH, Pei YL, Jackson DM, Holick MF 1999 Molecular cloning, characterization, and promoter analysis of the human 25-hydroxyvitamin D3–1α-hydroxylase gene. Proc Natl Acad Sci USA 96(12):6988–6993. 73. Dusso AS, Kamimura S, Gallieni M, Zhong M, Negrea L, Shapiro S, et al. 1997 Gamma-interferon-induced resistance

1374

74. 75.

76. 77.

78.

79. 80.

81. 82. 83. 84. 85. 86. 87.

88. 89. 90. 91. 92.

to 1,25-(OH)2D3 in human monocytes and macrophages: a mechanism for the hypercalcemia of various granulomatoses. J Clin Endocrinol Metab 82(7):2222–2232. Monkawa T, Yoshida T, Hayashi M, Saruta T 2000 Identification of 25-hydroxyvitamin D3 1α-hydroxylase gene expression in macrophages. Kidney Int 58(2):559–568. Vidal M, Ramana CV, Dusso AS 2002 Stat1-vitamin D receptor interactions antagonize 1,25-dihydroxyvitamin D transcriptional activity and enhance stat1-mediated transcription. Mol Cell Biol 22(8):2777–2787. Bell NH, Gill JR Jr. Bartter FC 1964 On the abnormal calcium absorption in sarcoidosis. Evidence for increased sensitivity to vitamin D. Am J Med 36:500–513. Basile JN, Liel Y, Shary J, Bell NH 1993 Increased calcium intake does not suppress circulating 1,25-dihydroxyvitamin D in normocalcemic patients with sarcoidosis. J Clin Invest 91(4):1396–1398. Reichel H, Koeffler HP, Barbers R, Norman AW 1987 Regulation of 1,25-dihydroxyvitamin D3 production by cultured alveolar macrophages from normal human donors and from patients with pulmonary sarcoidosis. J Clin Endocrinol Metab 65(6):1201–1209. Fuss M, Pepersack T, Gillet C, Karmali R, Corvilain J 1992 Calcium and vitamin D metabolism in granulomatous diseases. Clin Rheumatol 11(1):28–36. Jackson WP, Dancaster C 1959 A consideration of the hypercalciuria in sarcoidosis, idiopathic hypercalciuria, and that produced by vitamin D; a new suggestion regarding calcium metabolism. J Clin Endocrinol Metab 19(6):658–680. Bell NH, Bartter FC 1967 Studies of 47-Ca metabolism in sarcoidosis: evidence for increased sensitivity of bone to vitamin D. Acta Endocrinol (Copenh) 54(1):173–180. Dwarakanathan A, Ryan WG 1987 Hypercalcemia of sarcoidosis treated with cellulose sodium phosphate. Bone Miner 2(4):333–336. Insogna KL, Dreyer BE, Mitnick M, Ellison AF, Broadus AE 1988 Enhanced production rate of 1,25-dihydroxyvitamin D in sarcoidosis. J Clin Endocrinol Metab 66(1):72–75. Fallon MD, Perry HM 3rd, Teitelbaum SL 1981 Skeletal sarcoidosis with osteopenia. Metab Bone Dis Relat Res 3(3):171–4. Alexandre C, Chappard D, Vergnon JM, Emonot A, Riffat G 1987 The bone in noncorticoid-treated sarcoidosis. A histomorphometric study. Rev Rhum Mal Osteoartic 54(2):159–162. Scholz DA, Keating FR, Jr. 1956 Renal insufficiency, renal calculi, and nephrocalcinosis in sarcoidosis; report of eight cases. Am J Med 21(1):75–84. Sandler LM, Winearls CG, Fraher LJ, Clemens TL, Smith R, JL O’Riordan 1984 Studies of the hypercalcaemia of sarcoidosis: effect of steroids and exogenous vitamin D3 on the circulating concentrations of 1,25-dihydroxy vitamin D3. Q J Med 53(210):165–180. Zeimer HJ, Greenaway TM, Slavin J, Hards DK, Zhou H, Doery JC, et al. 1998 Parathyroid-hormone-related protein in sarcoidosis. Am J Pathol 152(1):17–21. Abbasi AA, Chemplavil JK, Farah S, Muller BF, Arnstein AR 1979 Hypercalcemia in active pulmonary tuberculosis. Ann Intern Med 90(3):324–328. Sharma SC, 1981 Serum calcium in pulmonary tuberculosis. Postgrad Med J 57(673):694–696. Kitrou MP, Phytou Pallikari A, Tzannes SE, Virvidakis K, Mountokalakis TD 1982 Hypercalcemia in active pulmonary tuberculosis. Ann Intern Med 96(2):255. Davies PD, Brown RC, Woodhead JS 1985 Serum concentrations of vitamin D metabolites in untreated tuberculosis. Thorax 40(3):187–190.

MISHAELA R. RUBIN

ET AL .

93. Fuss M, Karmali R, Pepersack T, Bergans A, Dierckx P, Prigogine T, et al. 1988 Are tuberculous patients at a great risk from hypercalcemia? Q J Med 69(259):869–878. 94. Chan TY, Chan CH, Shek CC, Davies PD 1992 Hypercalcemia in active pulmonary tuberculosis and its occurrence in relation to the radiographic extent of disease. Southeast Asian J Trop Med Public Health 23(4):702–704. 95. Tan TT, Lee BC, Khalid BA 1993 Low incidence of hypercalcaemia in tuberculosis in Malaysia. J Trop Med Hyg 96(6):349–351. 96. Kelestimur F, Guven M, Ozesmi M, Pasaoglu H 1996 Does tuberculosis really cause hypercalcemia? J Endocrinol Invest 19(10):678–681. 97. Chan TY 1997 Differences in vitamin D status and calcium intake: possible explanations for the regional variations in the prevalence of hypercalcemia in tuberculosis. Calcif Tissue Int 60(1):91–93. 98. Felsenfeld AJ, Drezner MK, Llach F 1986 Hypercalcemia and elevated calcitriol in a maintenance dialysis patient with tuberculosis. Arch Intern Med 146(10):1941–1945. 99. Peces R, Alvarez J 1987 Hypercalcemia and elevated 1,25(OH)2D3 levels in a dialysis patient with disseminated tuberculosis. Nephron 46(4):377–379. 100. Shai F, Baker RK, Addrizzo JR, Wallach S 1972 Hypercalcemia in mycobacterial infection. J Clin Endocrinol Metab 34(2):251–256. 101. Barnes PF, Modlin RL, Bikle DD, Adams JS 1989 Transpleural gradient of 1,25-dihydroxyvitamin D in tuberculous pleuritis. J Clin Invest 83(5):1527–1532. 102. Adams JS, Modlin RL, Diz MM, Barnes PF 1989 Potentiation of the macrophage 25-hydroxyvitamin D-1-hydroxylation reaction by human tuberculous pleural effusion fluid. J Clin Endocrinol Metab 69(2):457–460. 103. Cadranel J, Garabedian M, Milleron B, Guillozo H, Akoun G, Hance AJ 1990 1,25(OH)2D2 production by T lymphocytes and alveolar macrophages recovered by lavage from normocalcemic patients with tuberculosis. J Clin Invest 85(5):1588–1593. 104. Saggese G, Bertelloni S, Baroncelli GI, Di Nero G 1993 Ketoconazole decreases the serum-ionized calcium and 1,25-dihydroxyvitamin D levels in tuberculosis-associated hypercalcemia. Am J Dis Child 147(3):270–273. 105. Ryzen E, Singer FR 1985 Hypercalcemia in leprosy. Arch Intern Med 145(7):1305–6. 106. Lee JC, Catanzaro A, Parthemore JG, Roach B, Deftos LJ 1977 Hypercalcemia in disseminated coccidioidomycosis. N Engl J Med 297(8):431–433. 107. Murray JJ, Heim CR 1985 Hypercalcemia in disseminated histoplasmosis. Aggravation by vitamin D. Am J Med 78(5):881–884. 108. Kantarjian HM, Saad MF, Estey EH, Sellin RV, Samaan NA 1983 Hypercalcemia in disseminated candidiasis. Am J Med 74(4):721–724. 109. Bosch X 1998 Hypercalcemia due to endogenous overproduction of active vitamin D in identical twins with catscratch disease. Jama 279(7):532–534. 110. Ahmed B, Jaspan JB 1993 Case report: hypercalcemia in a patient with AIDS and pneumocystis carinii pneumonia. Am J Med Sci 306(5):313–316. 111. Jurney TH 1984 Hypercalcemia in a patient with eosinophilic granuloma. Am J Med 76(3):527–528. 112. Stoeckle JD, Hardy HL, Weber AL 1969 Chronic beryllium disease. Long-term follow-up of sixty cases and selective review of the literature. Am J Med 46(4):545–561. 113. Kozeny GA, Barbato AL, Bansal VK, Vertuno LL, Hano JE 1984 Hypercalcemia associated with silicone-induced granulomas. N Engl J Med 311(17):1103–1105.

1375

CHAPTER 78 Hypercalcemia Due to Vitamin D Toxicity

114. Albitar S, Genin R, Fen-Chong M, Schohn D, Riviere JP, Serveaux MO, et al. 1997 Multisystem granulomatous injuries 28 years after paraffin injections. Nephrol Dial Transplant 12(9):1974–1976. 115. Edelson GW, Talpos GB, Bone HG 3rd 1993, Hypercalcemia associated with Wegener’s granulomatosis and hyperparathyroidism: etiology and management. Am J Nephrol 13(4):275–277. 116. Al-Ali H, Yabis AA, Issa E, Salem Z, Tawil A, Khoury N, et al. 2002 Hypercalcemia in Langerhans’ cell granulomatosis with elevated 1,25 dihydroxyvitamin D (calcitriol) level. Bone 30(1):331–334. 117. Bosch X 1998 Hypercalcemia due to endogenous overproduction of 1,25-dihydroxyvitamin D in Crohn’s disease. Gastroenterology 114(5):1061–1065. 118. Cook JS, Stone MS, Hansen JR 1992 Hypercalcemia in association with subcutaneous fat necrosis of the newborn: studies of calcium-regulating hormones. Pediatrics 90(1 Pt 1): 93–96. 119. Evron E, Goland S, von der Walde J, Schattner A, Sthoeger ZM 1997 Idiopathic calcitriol-induced hypercalcemia. A new disease entity? Arch Intern Med 157(18):2142–2145. 120. Burt ME, Brennan MF 1980 Incidence of hypercalcemia and malignant neoplasm. Arch Surg 115(6):704–7. 121. Vassilopoulou-Sellin R, Newman BM, Taylor SH, Guinee VF 1993 Incidence of hypercalcemia in patients with malignancy referred to a comprehensive cancer center. Cancer 71(4):1309–1312. 122. Johnston SR, Hammond PJ 1992 Elevated serum parathyroid hormone-related protein and 1,25-dihydroxycholecalciferol in hypercalcemia associated with adult T-cell leukemialymphoma. Postgrad Med J 68(803):753–755. 123. Major prognostic factors of patients with adult T-cell leukemia-lymphoma: a cooperative study. Lymphoma Study Group (1984–1987). Leuk Res 15(2–3):81–90. 124. Gellhorn, A, Plimpton CH 1956 Hypercalcemia in malignant disease without evidence of bone destruction. Am J Med 21(5):750–759. 125. Seymour JF, Gagel RF 1993 Calcitriol: the major humoral mediator of hypercalcemia in Hodgkin’s disease and nonHodgkin’s lymphomas. Blood 82(5):1383–1394. 126. Davies M, Hayes ME, Yin JA, Berry JL, Mawer EB 1994 Abnormal synthesis of 1,25-dihydroxyvitamin D in patients with malignant lymphoma. J Clin Endocrinol Metab 78(5):1202–1207. 127. Karmali R, Barker S, Hewison M, Fraher L, Katz DR, O’Riordan J.L 1990 Intermittent hypercalcaemia and vitamin D sensitivity in Hodgkin’s disease. Postgrad Med J 66(779): 757–760. 128. Freeman NJ, Holik D 2003 Uncommon syndromes and treatment manifestations of malignancy: Case 3. Richter’s syndrome heralded by refractory hypercalcemia. J Clin Oncol 21(1):170–172. 129. Walker IR, 1974 Lymphoma with hypercalcemia. Can Med Assoc J 111(9):928–930. 130. Breslau NA, McGuire JL, Zerwekh JE, Frenkel EP, Pak CY 1984 Hypercalcemia associated with increased serum calcitriol levels in three patients with lymphoma. Ann Intern Med 100(1):1–6. 131. Mudde AH, van den Berg H, Boshuis PG, Breedveld FC, Markusse HM, Kluin PM, et al. 1987 Ectopic production of 1,25-dihydroxyvitamin D by B-cell lymphoma as a cause of hypercalcemia. Cancer 59(9):1543–1546. 132. Rosenthal N, Insogna KL, Godsall JW, Smaldone L, Waldron JA, Stewart AF 1985 Elevations in circulating

133. 134.

135.

136. 137.

138.

139.

140. 141.

142. 143.

144.

145.

146. 147.

148. 149.

1,25-dihydroxyvitamin D in three patients with lymphomaassociated hypercalcemia. J Clin Endocrinol Metab 60(1): 29–33. Rizvi AA, Bowman MA, Vaughters RB, 3rd, Isales C, Mulloy AL 2003 Primary ovarian lymphoma manifesting with severe hypercalcemia. Endocr Pract 9(5):389–393. Adams JS, Fernandez M, Gacad MA, Gill PS, Endres DB, Rasheed S, et al. 1989 Vitamin D metabolite-mediated hypercalcemia and hypercalciuria patients with AIDS- and non-AIDS-associated lymphoma. Blood 73(1):235–239. Scheinman SJ, Kelberman MW, Tatum AH, Zamkoff KW 1991 Hypercalcemia with excess serum 1,25 dihydroxyvitamin D in lymphomatoid granulomatosis/angiocentric lymphoma. Am J Med Sci 301(3):178–181. Grote TH, Hainsworth JD 1987 Hypercalcemia and elevated serum calcitriol in a patient with seminoma. Arch Intern Med 147(12):2212–2213. Ogose A, Kawashima H, Morita O, Hotta T, Umezu H, Endo N 2003 Increase in serum 1,25-dihydroxyvitamin D and hypercalcemia in a patient with inflammatory myofibroblastic tumor. J Clin Pathol 56(4):310–312. Seymour JF, Gagel RF, Hagemeister FB, Dimopoulos MA, Cabanillas F 1994 Calcitriol production in hypercalcemic and normocalcemic patients with non-Hodgkin’s lymphoma. Ann Intern Med 121(9):633–640. National Cancer Institute sponsored study of classifications of non-Hodgkin’s lymphomas: summary and description of a working formulation for clinical usage. The Non-Hodgkin’s Lymphoma Pathologic Classification Project. Cancer 49(10): 2112–2135. Cox M, Haddad JG 1994 Lymphoma, hypercalcemia, and the sunshine vitamin. Ann Intern Med 121(9):709–712. Hewison M, Kantorovich V, Liker HR AJ, Van Herle, Cohan P, Zehnder D, et al. 2003 Vitamin D-mediated hypercalcemia in lymphoma: evidence for hormone production by tumor-adjacent macrophages. J Bone Miner Res 18(3): 579–582. Kremer R, Shustik C, Tabak T, Papavasiliou V, Goltzman D 1996 Parathyroid hormone-related peptide in hematologic malignancies. Am J Med 100(4):406–411. Dodwell DJ, Abbas SK, Morton AR, Howell A 1991 Parathyroid hormone-related protein (50-69) and response to pamidronate therapy for tumor-induced hypercalcemia. Eur J Cancer 27(12):1629–1633. Wada S, Kitamura H, Matsuura Y, Katayama Y, Ohkawa H, Kugai N, et al. 1992 Parathyroid hormone-related protein as a cause of hypercalcemia in a B-cell type malignant lymphoma. Intern Med 31(8):968–972. Fetchick DA, Bertolini DR, Sarin PS, Weintraub ST, Mundy GR, Dunn JF 1986 Production of 1,25-dihydroxyvitamin D3 by human T-cell lymphotrophic virus I-transformed lymphocytes. J Clin Invest 78(2):592–596. Dodd RC, Winkler CF, Williams ME, Bunn PA, Gray TK 1986 Calcitriol levels in hypercalcemic patients with adult Tcell lymphoma. Arch Intern Med 146(10):1971–1972. Fukumoto S, Matsumoto T, Ikeda K, Yamashita T, Watanabe T, Yamaguchi K, et al. 1988 Clinical evaluation of calcium metabolism in adult T-cell leukemia/lymphoma. Arch Intern Med 148(4):921–925. Senba M, Kawai K 1992 Hypercalcemia and production of parathyroid hormone-like protein in adult T-cell leukemialymphoma. Eur J Haematol 48(5):278–279. Motokura T, Fukumoto S, Takahashi S, Watanabe T, Matsumoto T, Igarashi T, et al. 1988 Expression of parathyroid hormone-related protein in a human T-cell lymphotrophic

1376

150.

151. 152.

153.

154. 155.

156. 157.

158.

159.

160. 161.

162.

163. 164.

165.

virus type I-infected T-cell line. Biochem Biophys Res Commun 154(3):1182–1188. Moseley JM, Danks JA, Grill V, Lister TA, Horton MA 1991 Immunocytochemical demonstration of PTHrP protein in neoplastic tissue of HTLV-1 positive human adult T-cell leukemia/lymphoma: implications for the mechanism of hypercalcemia. Br J Cancer 64(4):745–748. Haddad JG Jr, Rojanasathit S 1976 Acute administration of 25-hydroxycholecalciferol in man. J Clin Endocrinol Metab 42(2):284–290. Kawakami, M, Imawari M, Goodman DS 1979 Quantitative studies of the interaction of cholecalciferol (vitamin D3) and its metabolites with different genetic variants of the serumbinding protein for these sterols. Biochem J 179(2):413–423. Shepard R, DeLuca HF 1980 Plasma concentrations of vitamin D3 and its metabolites in the rat as influenced by vitamin D3 or 25-hydroxyvitamin D3 intakes. Arch Biochem Biophys 202:43–53. Vieth R, McCarten K, Norwich KH 1990 Role of 25-hydroxyvitamin D3 dose in determining rat 1,25-dihydroxyvitamin D3 production. Am J Physiol 258(5 Pt 1): E780–9. Reinhardt TA, Horst RL 1989 Ketoconazole inhibits selfinduced metabolism of 1,25-dihydroxyvitamin D3 and amplifies 1,25-dihydroxyvitamin D3 receptor up-regulation in rat osteosarcoma cells. Arch Biochem Biophys 272(2):459–465. Costa EM, Feldman D 1986 Homologous up-regulation of the 1,25 (OH)2 vitamin D3 receptor in rats. Biochem Biophys Res Commun 137(2):742–747. Reinhardt TA, Horst R, Engstrom G, Atkins K 1998 Ketoconazole potentiates 1,25(OH)2D-directed up-regulation of 1,25(OH)2D receptros in rat intestine and bone. Seventh Workshop on Vitamin D, In: Norman AW, et al. (eds) Vitamin D: molecular, cellular and clinical endocrinology, de Gruyter: Berlin, p. 233. Goff JP, Reinhardt TA, Beckman MJ, Horst RL 1990 Contrasting effects of exogenous 1,25-dihydroxyvitamin D [1,25-(OH)2D] versus endogenous 1,25-(OH)2D, induced by dietary calcium restriction, on vitamin D receptors. Endocrinology 126(2):1031–1035. Reinhardt TA, Horst R 1990 PTH down-regulates 1,25(OH)2D receptors and VDR mRNA in vitro and blocks homologous up-regulation of VDR in vitro. Endocrinology 127:942–948. Haussler MR, Cordy PE 1982 Metabolites and analogs of vitamin D. Which for what? JAMA 247(6):841–844. Beckman MJ, Horst RL, Reinhardt TA, Beitz DC 1990 Up-regulation of the intestinal 1,25-dihydroxyvitamin D receptor during hypervitaminosis D: a comparison between vitamin D2 and vitamin D3. Biochem Biophys Res Commun 169(3):910–915. Beckman MJ, Johnson JA, Goff JP, Reinhardt TA, Beitz DC, Horst RL 1995 The role of dietary calcium in the physiology of vitamin D toxicity: excess dietary vitamin D3 blunts parathyroid hormone induction of kidney 1-hydroxylase. Arch Biochem Biophys 319(2):535–539. Haussler MR, 1986 Vitamin D receptors: nature and function. Annu Rev Nutr 6:527–562. Reinhardt TA, Horst RL 1989 Self-induction of 1,25dihydroxyvitamin D3 metabolism limits receptor occupancy and target tissue responsiveness. J Biol Chem 264(27): 15917–15921. Shigematsu, T, Horiuchi N, Ogura Y, Miyahara T, Suda T 1986 Human parathyroid hormone inhibits renal 24-hydroxylase activity of 25-hydroxyvitamin D3 by a mechanism

MISHAELA R. RUBIN

166.

167.

168. 169. 170. 171.

172.

173.

174.

175. 176. 177. 178.

179. 180. 181. 182. 183.

ET AL .

involving adenosine 3′,5′-monophosphate in rats. Endocrinology 118(4):1583–1589. Beckman MJ, Goff JP, Reinhardt TA, Beitz DC 1994 and Horst RL, In vivo regulation of rat intestinal 24-hydroxylase: potential new role of calcitonin. Endocrinology 135(5): 1951–1955. Tryfonidou MA, Oosterlaken-Dijksterhuis MA, Mol JA, van den Ingh T.S, van den Brom WE, Hazewinkel HA 2003 24-Hydroxylase: potential key regulator in hypervitaminosis D3 in growing dogs. Am J Physiol Endocrinol Metab 284(3): E505–13. Mallon JP, Matuszewski D, Sheppard H 1980 Binding specificity of the rat serum vitamin D transport protein. J Steroid Biochem 13(4):409–413. Adams JS, 1984 Specific internalization of 1,25-dihydroxyvitamin D3 by cultured intestinal epithelial cells. J Steroid Biochem 20(4A):857–862. Pettifor JM, Bikle DD, Cavaleros M, Zachen D, Kamdar MC, Ross FP 1995 Serum levels of free 1,25-dihydroxyvitamin D in vitamin D toxicity. Ann Intern Med 122(7):511–513. Bouillon, RFA, Van Assche H, Van Baelen, Heyns W, De Moor P 1981 Influence of the vitamin D–binding protein on the serum concentration of 1,25-dihydroxyvitamin D3. Significance of the free 1,25-dihydroxyvitamin D3 concentration. J Clin Invest 67(3):589–596. Bikle DD, Siiteri PK, Ryzen E, Haddad JG 1985 Serum protein-binding of 1,25-dihydroxyvitamin D: a reevaluation by direct measurement of free metabolite levels. J Clin Endocrinol Metab 61(5):969–975. Bikle DD, Gee E, Halloran B, Haddad JG 1984 Free 1,25dihydroxyvitamin D levels in serum from normal subjects, pregnant subjects, and subjects with liver disease. J Clin Invest 74(6):1966–1971. Bikle DD, Halloran BP, Gee E, Ryzen E, Haddad JG 1986 Free 25-hydroxyvitamin D levels are normal in subjects with liver disease and reduced total 25-hydroxyvitamin D levels. J Clin Invest 78(3):748–752. Koenig KG, Lindberg JS, Zerwekh JE, Padalino PK, Cushner HM, Copley JB 1992 Free and total 1,25-dihydroxyvitamin D levels in subjects with renal disease. Kidney Int 41(1):161–5. Mendel CM, 1989 The free hormone hypothesis: a physiologically-based mathematical model. Endocr Rev 10(3): 232–274. Edelson GW, Kleerekoper M 1995 Hypercalcemic crisis. Med Clin North Am 79(1):79–92. Ashizawa, N, Arakawa S, Koide Y, Toda G, Seto S, Yano K 2003 Hypercalcemia due to vitamin D intoxication with clinical features mimicking acute myocardial infarction. Intern Med 42(4):340–344. Nussbaum SR, 1993 Pathophysiology and management of severe hypercalcemia. Endocrinol Metab Clin North Am 22(2):343–362. Scarpelli DG, Tremblay G, Pearse AG 1960 A comparative cytochemical and cytologic study of vitamin D–induced nephrocalcinosis. Am J Pathol 36:331–353. Shetty KR, Ajlouni K, Rosenfled PS, Hagen TC 1975 Protracted vitamin D intoxication. Arch Intern Med 135(7): 986–988. Rizzoli, R, Stoermann C, Ammann P, Bonjour JP 1994 Hypercalcemia and hyperosteolysis in vitamin D intoxication: effects of clodronate therapy. Bone 15(2):193–198. Hefti, E, Trechsel U, Fleisch H, Bonjour JP 1983 Nature of calcemic effect of 1,25-dihydroxyvitamin D3 in experimental hypoparathyroidism. Am J Physiol 244(4): E313–E316.

1377

CHAPTER 78 Hypercalcemia Due to Vitamin D Toxicity

184. Hass GM, Trueheart RE, Taylor CB, Stumpe M 1958 An experimental histologic study of hypervitaminosis D. Am J Pathol 34(3):395–431. 185. Swierczynski, J, Nagel G, Zydowo MM 1987 Calcium content in some organs of rats treated with a toxic calciol dosis. Pharmacology 34(1):57–60. 186. Giunta JL, Dental changes in hypervitaminosis D 1998 Oral Surg Oral Med Oral Pathol Oral Radiol Endod 85(4):410–413. 187. Selby PL, Davies M, Marks JS, Mawer EB 1995 Vitamin D intoxication causes hypercalcemia by increased bone resorption which responds to pamidronate. Clin Endocrinol (Oxf) 43(5):531–536. 188. Adams JS, Lee G 1997 Gains in bone mineral density with resolution of vitamin D intoxication. Ann Intern Med 127(3):203–206. 189. Glass AR, Eil C 1988 Ketoconazole-induced reduction in serum 1,25-dihydroxyvitamin D and total serum calcium in hypercalcemic patients. J Clin Endocrinol Metab 66(5): 934–938. 190. Kogan BA, Konnak JW, Lau K 1982 Marked hyperoxaluria in sarcoidosis during orthophosphate therapy. J Urol 127(2): 339–340. 191. Feher JJ, Wasserman RH 1979 Intestinal calcium-binding protein and calcium absorption in cortisol-treated chicks: effects of vitamin D3 and 1,25-dihydroxyvitamin D3. Endocrinology 104(2):547–551. 192. Suzuki, Y, Ichikawa Y, Saito E, Homma M 1983 Importance of increased urinary calcium excretion in the development of secondary hyperparathyroidism of patients under glucocorticoid therapy. Metabolism 32(2):151–156. 193. Lukert BP, Raisz LG 1990 Glucocorticoid-induced osteoporosis: pathogenesis and management. Ann Intern Med 112(5):352–364. 194. Carre, M, Ayigbede O, Miravet L, Rasmussen H 1974 The effect of prednisolone upon the metabolism and action of 25-hydroxy- and 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 71(8):2996–3000. 195. Bilezikian JP, 1989 Etiologies and therapy of hypercalcemia. Endocrinol Metab Clin North Am 18(2):389–414. 196. O’Leary TJ, Jones G, Yip A, Lohnes D, Cohanim M, Yendt ER 1986 The effects of chloroquine on serum 1,25dihydroxyvitamin D and calcium metabolism in sarcoidosis. N Engl J Med 315(12):727–730. 197. Adams JS, Sharma OP, Diz MM, Endres DB 1990 Ketoconazole decreases the serum 1,25-dihydroxyvitamin D and calcium concentration in sarcoidosis-associated hypercalcemia. J Clin Endocrinol Metab 70(4):1090–5. 198. Adams JS, Kantorovich V 1999 Inability of short-term, lowdose hydroxychloroquine to resolve vitamin D–mediated hypercalcemia in patients with B-cell lymphoma. J Clin Endocrinol Metab 84(2):799–801. 199. Heaney RP, Davies KM, Chen TC, Holick MF, Barger-Lux MJ 2003 Human serum 25-hydroxycholecalciferol response

200. 201.

202.

203.

204.

205. 206.

207.

208. 209.

210.

211. 212.

to extended oral dosing with cholecalciferol. Am J Clin Nutr 77(1):204–210. Barger-Lux MJ, Heaney RP 2002 Effects of above average summer sun exposure on serum 25-hydroxyvitamin D and calcium absorption. J Clin Endocrinol Metab 87(11):4952–4956. Trivedi DP, Doll R, Khaw KT 2003 Effect of four monthly oral vitamin D3 (cholecalciferol) supplementation on fractures and mortality in men and women living in the community: randomized double-blind controlled trial. BMJ 326(7387):469. Dawson-Hughes B, Dallal GE, Krall EA, Harris S, Sokoll LJ, Falconer G 1991 Effect of vitamin D supplementation on wintertime and overall bone loss in healthy postmenopausal women. Ann Intern Med 115(7):505–512. Schwartz GG, Whitlatch LW, Chen TC, Lokeshwar BL, Holick MF 1998 Human prostate cells synthesize 1,25-dihydroxyvitamin D3 from 25-hydroxyvitamin D3. Cancer Epidemiol Biomarkers Prev 7(5):391–395. Shevde NK, Plum LA, Clagett-Dame M, Yamamoto H, Pike JW, DeLuca HF 2002 A potent analog of 1α,25-dihydroxyvitamin D3 selectively induces bone formation. Proc Natl Acad Sci USA 99(21):13487–13491. Guyton KZ, Kensler TW, Posner GH 2003 Vitamin D and vitamin D analogs as cancer chemopreventive agents. Nutr Rev 61(7):227–238. Beer TM, Lemmon D, Lowe BA, Henner WD 2003 Highdose weekly oral calcitriol in patients with a rising PSA after prostatectomy or radiation for prostate carcinoma. Cancer 97(5):1217–1224. Segersten, U, Correa P, Hewison M, Hellman P, Dralle H, Carling T, et al. 2002 25-hydroxyvitamin D3-1α-hydroxylase expression in normal and pathological parathyroid glands. J Clin Endocrinol Metab 87(6):2967–2972. Lemire JM, 1995 Immunomodulatory actions of 1,25-dihydroxyvitamin D3. J Steroid Biochem Mol Biol 53(1–6): 599–602. Panda DK, Miao D, Tremblay ML, Sirois J, Farookhi R, Hendy GN, et al. 2001 Targeted ablation of the 25-hydroxyvitamin D 1α-hydroxylase enzyme: evidence for skeletal, reproductive, and immune dysfunction. Proc Natl Acad Sci USA 98(13):7498–7503. Dardenne, O, Prud’homme J, Arabian A, Glorieux FH, St Arnaud R 2001 Targeted inactivation of the 25-hydroxyvitamin D3-1α-hydroxylase gene (CYP27B1) creates an animal model of pseudovitamin D–deficiency rickets. Endocrinology 142(7):3135–3141. Mathieu, C, Waer M, Laureys J, Rutgeerts O, Bouillon R 1994 Prevention of autoimmune diabetes in NOD mice by 1,25 dihydroxyvitamin D3. Diabetologia 37(6):552–558. Zella JB, McCary LC, DeLuca HF 2003 Oral administration of 1,25-dihydroxyvitamin D3 completely protects NOD mice from insulin-dependent diabetes mellitus. Arch Biochem Biophys 417(1):77–80.

CHAPTER 79

Extra-renal 1α-Hydroxylase Activity and Human Disease MARTIN HEWISON JOHN S. ADAMS

Division of Medical Sciences, The University of Birmingham, Queen Elizabeth Medical Centre, Birmingham B15 2TH, UK Division of Endocrinology, Metabolism and Diabetes, Cedars-Sinai Medical Center, Los Angeles, USA

I. Introduction II. Vitamin D and Granuloma-Forming Disease: Historical Perspective III. Pathophysiology of Disordered Calcium Balance in Sarcoidosis: A Model for the Extra-renal Production of an Active Vitamin D Metabolite in Human Disease IV. Local Immunoregulatory Effects of Active Vitamin D Metabolites

V. Human Diseases Associated with the Extra-renal Overproduction of Active Vitamin D Metabolites VI. Diagnosis, Prevention, and Treatment of the Patient with Endogenous Vitamin D Intoxication References

I. INTRODUCTION

immunological targets for active vitamin D metabolites and propose a model in which macrophage-derived vitamin D metabolites play a role in the modulation of the local immune responses. The fifth section will provide a comprehensive review of the various human diseases proposed to be associated with the overproduction of active vitamin D metabolites from an extra-renal source. The sixth and final section of this chapter will address the clinical aspects of disordered extra-renal 1α-hydroxylase; this will include a discussion of the diagnosis, treatment, and prevention of hypercalcemia and hypercalciuria in the patient with endogenous vitamin D intoxication.

The period of time following the publication of the first edition of this book has witnessed some remarkable advances in our understanding of the enzyme 25-hydroxyvitamin D-1α-hydroxylase (1α-hydroxylase), including the cloning of the gene for this enzyme (CYP1α or CYP27b), the development of knockout animal models, and vastly improved tools for analysis of tissue-specific expression of CYP1α. In this chapter we have incorporated these new developments into the framework of the original chapter on the pathophysiology of dysregulated vitamin D metabolism associated with granuloma-forming and malignant lymphoproliferative disorders. We have placed the seminal observations of extra-renal 1α-hydroxylase activity in diseases, such as sarcoidosis, alongside the current studies that have highlighted a much wider tissue distribution of the enzyme, including epithelial cells. The fundamental structure of the chapter has been retained, but additional sections have been added. After this brief introduction (first section), the second section of the chapter will review the historical aspects of extra-renal synthesis of 1,25-dihydroxyvitamin D (1,25(OH)2D) associated with inflammatory disease, including recent studies that have expanded the pathological relevance of extra-renal 1α-hydroxylase. The third section of the chapter will describe what we know about the mechanics and regulation of the vitamin D metabolizing enzymes present in inflammatory cells. The fourth section will recapitulate the potential VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

II. VITAMIN D AND GRANULOMAFORMING DISEASE: HISTORICAL PERSPECTIVE A. Evidence of Endogenous Vitamin D Intoxication Associated with Sarcoidosis A pathophysiological relationship between vitamin D and sarcoidosis was first recognized by Harrell and Fisher in 1939 [1]. Among the six hypercalcemic patients in their initial report, one was observed to experience a steep rise in the serum calcium concentration following ingestion of cod liver oil known to be enriched in vitamin D. Almost two decades passed before Henneman et al. [2] demonstrated in 1956 that the hypercalcemic syndrome of sarcoidosis, characterized Copyright © 2005, Elsevier, Inc. All rights reserved.

1380 by increased intestinal calcium absorption and bone resorption, was remarkably similar to that of exogenous vitamin D intoxication and was treatable by the administration of glucocorticoids. This is summarized in Section IV of this chapter and is documented in greater detail in Chapters 5, 7, and 93–96. In 1963, Taylor and coworkers [3] performed the first, large-scale seasonal evaluation of serum calcium levels in patients with sarcoidosis. They found that there was a significant increase in the mean serum calcium concentration in 345 patients with sarcoidosis from winter to summer, but no such change in over 12,000 control subjects. This was the first evidence that there was an association between enhanced vitamin D synthesis, known to occur principally during the summer months, and the blood level of calcium in patients with sarcoidosis. This observation was prospectively confirmed by Dent [4], who was able to increase the serum calcium concentration in patients with active sarcoidosis upon exposure to whole body ultraviolet radiation. The Dent study also helped validate the earlier work of Hendrix [5] who achieved resolution of hypercalcemia and hypercalciuria in two patients with sarcoidosis by institution of vitamin D–deficient diets and environmental sunlight deprivation.

B. Evidence for the Extra-renal Overproduction of an Active Vitamin D Metabolite The above mentioned studies led Bell et al. [6] to propose in 1964 that development of a clinical abnormality in calcium balance in patients with active sarcoidosis resulted from an increase in target organ responsiveness to vitamin D. This view persisted for more than a decade. However, after the discovery of 1,25(OH)2D as the active vitamin D hormone [7–9] and the development of sensitive and specific assays for the hormone in blood [10–12], investigators were quick to determine that the hypercalcemia of sarcoidosis was the result of an increase in the circulating concentrations of a vitamin D metabolite that interacted with the vitamin D receptor (VDR) [13–16]. The fact that a vitamin D hormone was made outside the kidney in hypercalcemic patients with sarcoidosis was first discovered by Barbour and colleagues in 1981 [17]. These investigators reported high concentrations of a vitamin D metabolite detected as 1,25(OH)2D in the circulation of a hypercalcemic, anephric patient with active sarcoidosis. Two years later, Adams et al. [18] determined the macrophage to be the extra-renal source of the active vitamin D metabolite. Unequivocal structural characterization of the metabolite as 1,25(OH)2D was obtained by these same investigators in 1985 [19].

MARTIN HEWISON AND JOHN S. ADAMS

C. Cloning of the CYP1α Gene Provides New Perspectives for Extra-renal Synthesis of 1,25(OH)2D The original studies describing synthesis of 1,25(OH)2D by activated macrophages and the potential consequences of this with respect to inflammatory diseases such as sarcoidosis has stimulated a much broader appreciation of extra-renal activation of vitamin D. This is summarized in section IV of this chapter and is documented in greater detail in Chapters 7, 93, 94, and 95. Further investigation of macrophage 1α-hydroxylase activity also highlighted several crucial differences between the activity of the enzyme in these cells when compared to its classical renal counterpart. For example, the macrophage 1α-hydroxylase is not subject to the exquisite autoregulation characteristic of its kidney counterpart, raising the possibility that renal and extra-renal synthesis of 1,25(OH)2D is catalyzed by distinct enzymes. This and other mechanistic features of extra-renal 1α-hydroxylase are discussed in section III of this chapter. The most significant contributing factor to our current understanding of extra-renal 1,25(OH)2D production has been the cloning of the gene for 1α-hydroxylase (CYP1α). After initial isolation of the mouse gene (Cyp1α) from renal tissue [20], it is notable that the human homolog (CYP1α) was cloned from keratinocytes, a well-established extra-renal site for 1,25(OH)2D production [21]. That this gene was identical to that in the kidney strongly supported the notion of a single but differentially regulated 1α-hydroxylase protein in renal and extra-renal tissues. Further support for this postulate was provided by Mawer and colleagues who showed that macrophages from patients harboring mutations in the CYP1α gene had impaired levels of 1,25(OH)2D production similar to that observed in the renal enzyme [22]. The availability of sequence information has also facilitated the development of specific antisera and probes for 1α-hydroxylase. This has further emphasized the widespread tissue distribution of 1α-hydroxylase, but has also helped to confirm the identity between the renal and extra-renal enzymes [23,24]. Advances in our understanding of extra-renal 1α-hydroxylase have also led to its implication in diseases beyond the original observation of abnormal synthesis of 1,25(OH)2D in some patients with sarcoidosis. A key development has been the expression and function of 1α-hydroxylase in breast, prostate, and colon cancer, and this is discussed in greater detail in Chapters 93–95. In the remainder of this chapter, we will focus on the established link between extra-renal 1α-hydroxylase and granulomatous diseases.

CHAPTER 79 Extra-renal 1α-Hydroxylase Activity and Human Disease

III. PATHOPHYSIOLOGY OF DISORDERED CALCIUM BALANCE IN SARCOIDOSIS: A MODEL FOR THE EXTRA-RENAL PRODUCTION OF AN ACTIVE VITAMIN D METABOLITE IN HUMAN DISEASE A. Clinical Evidence for Dysregulated Overproduction of the Vitamin D Hormone As has been described in detail in earlier chapters, the synthesis of 1,25(OH)2D by the renal 1α-hydroxylase is normally strictly regulated with levels of the hormone product being some 1000-fold less plentiful in the circulation than that of the principal substrate for the enzyme, 25-hydroxyvitamin D (25OHD). Hormone synthesis in the kidney is stimulated by an increase in the serum parathyroid hormone (PTH) concentration, a decrease in the serum phosphate concentration, and a decrease in the activity of the competing vitamin D 24-hydroxylase (24-hydroxylase). Synthesis of 1,25(OH)2D is inhibited by a decrease in the circulating PTH levels, increased serum phosphate, and increased 24-hydroxylase activity. There are now at least four clear lines of clinical evidence to indicate that endogenous 1,25(OH)2D production in hypercalcemic/ hypercalciuric patients with sarcoidosis is dysregulated and not bound by the same set of endocrine factors known to regulate 1,25(OH)2D synthesis in the kidney [25]. First, hypercalcemic patients with sarcoidosis possess a frankly high or inappropriately elevated serum 1,25(OH)2D concentration, although their serum PTH level is suppressed and their serum phosphate concentration is relatively elevated [26,27]. If 1,25(OH)2D synthesis were under the regulation of PTH, phosphate, and 1,25(OH)2D itself, then 1,25(OH)2D concentrations in such patients should be low. Second, the serum 1,25(OH)2D concentration in patients with active sarcoidosis is very sensitive to an increase in available substrate [28], while the serum 1,25(OH)2D level in normal individuals is not influenced by small or even moderate increments in the circulating 25OHD concentration. Clinically, this aspect of dysregulation is manifest by the long-recognized association of the appearance of hypercalciuria and/or hypercalciuria in sarcoidosis patients in the summer months or following holidays to geographic locations at lower latitudes than those at which the patient normally resides [28,29]. This link between an increase in cutaneous vitamin D synthesis and the development of a clinical abnormality in calcium balance can be replicated by the oral administration of vitamin D [15,27,30] to such patients.

1381

It can also be substantiated on a biochemical basis by demonstration of a positive correlation between the serum 25OHD and 1,25(OH)2D concentrations in patients with active sarcoidosis, but not in normal human subjects [31]. Third, the rate of endogenous 1,25(OH)2D production, which is significantly increased in patients with sarcoidosis [32], is unusually sensitive to inhibition by factors (i.e. drugs) that do not influence the renal 1α-hydroxylase at the same doses. Anti-inflammatory concentrations of glucocorticoids have long been recognized as effective combatants of sarcoidosis-associated hypercalcemia and have also been shown to dramatically lower elevated 1,25(OH)2D levels [30,31,33]. On the other hand, administration of the same glucocorticoid doses to patients without sarcoidosis is not associated with a clinically relevant reduction in the serum 1,25(OH)2D or calcium concentration. Chloroquine and its hydroxylated analog, hydroxychloroquine, are other examples of pharmaceutical agents that appear to act preferentially on the extra-renal vitamin D-1α-hydroxylase reaction, which is active in patients with sarcoidosis [34–36]. Fourth, the serum calcium and 1,25(OH)2D concentrations are positively correlated to indices of disease activity in patients with sarcoidosis [37–39]; patients with widespread disease and high angiotensinconverting enzyme (ACE) activity are more likely to be hypercalciuric or frankly hypercalcemic.

B. Correlates In Vitro for Dysregulated 1,25(OH)2D Production In Vivo Investigators have now generated a substantial body of experimental data from cells, including inflammatory cells harvested directly from patients with sarcoidosis, to indicate that the dysregulated vitamin D hormone synthesis in sarcoidosis is probably not due to expression of a 1α-hydroxylase that is different from the renal enzyme, but rather to expression of the authentic 1αhydroxylase in a macrophage, not a kidney cell [22]. In fact, each of the above mentioned pieces of clinical evidence for dysregulated vitamin D hormone production in this disease can be borne out in vitro in cells from patients with this disease [40]. 1. SUBCELLULAR LOCALIZATION, SUBSTRATE SELECTIVITY, AND KINETICS OF THE MACROPHAGE VITAMIN D-1-HYDROXYLASE

As is the case with the 1α-hydroxylase of renal origin, the macrophage enzyme is a mitochondrial mixed function oxidase with detectable cytochrome P450 activity [41] (see Fig. 1). Like the renal 1α-hydroxylase reconstituted from mitochondrial extracts, the presence

1382

MARTIN HEWISON AND JOHN S. ADAMS

Mitochondrial membrane

NADPH

FP

e−

Fdxred

1,25-D

CYP1α

1α-hydroxylase

e−

NADP+

e−

FPred

Fdx

e−

CYP1α red

_ O2

25-D

+ NO

FIGURE 1

Mitochondrial electron transport associated with the vitamin D-1α-hydroxylase reaction: a model for interaction with nitric oxide (NO). NADPH supplies the electron transport chain of accessory proteins associated with 1α-hydroxylase, consisting of a flavoprotein reductase (FP), a ferredoxin (Fdx), and the 1α-hydroxylase cytochrome P450 (CYP1α). A stimulatory effect on the enzyme may also be mediated by relatively low intracellular NO levels. An electron (e−) generated from NO is donated to oxidized NADP, thus forming NADPH. On the other hand, the inhibitory effect on the enzyme which occurs at relatively high NO levels in the cell results from competition with O2 binding to the P450 heme group, inhibiting the enzyme.

of a flavoprotein, ferredoxin reductase, an electron source, and molecular oxygen (O2) are all required for electron transfer to the cytochrome P450 and for the insertion of an oxygen atom in the substrate [41]. Also, like the renal 1α-hydroxylase, we now know the macrophage 1α-hydroxylase is inhibited by the napthoquinones, molecules which compete with reductase for donated electrons, and by the imidazoles, compounds which compete with the enzyme for receipt of O2 [42]. Similar to the 1α-hydroxylase isolated from the mitochondria of proximal renal tubular epithelial cells, the macrophage enzyme requires a secosterol (vitamin D sterol molecule with an open B-ring) as substrate [43]. Also similar to the renal 1α-hydroxylase, the macrophage enzyme has a particular affinity for secosterols bearing a carbon-25 hydroxy group as is encountered in the two preferred substrates for this enzyme, 25OHD and 24,25-dihydroxyvitamin D (24,25(OH)2D) [43,44]; the calculated Km (affinity) of the 1α-hydroxylase in pulmonary alveolar macrophages derived directly from patients with active sarcoidosis is in the range of 50–100 nM for these two substrates [43,44]. The availability of cDNA sequences for 1α-hydroxylase expression studies has shed more light on the catalytic properties of the enzyme (20,21,45,46) but, as yet, has failed to provide a clear mechanism for the differential regulation of 1,25(OH)2D production in renal and extra-renal tissues. Some of the potential explanations for this are discussed in the following sections. 2. MACROPHAGES LACK RESPONSIVENESS TO PTH, CALCIUM, PHOSPHATE

In vivo there appear to be three major regulators of the renal 1α-hydroxylase—the serum concentration

of calcium, parathyroid hormone, and phosphate [47] (left panel, Fig. 2). Hypocalcemia enhances the activity of the renal 1α-hydroxylase, but much of this stimulatory effect may be indirectly mediated through parathyroid hormone (PTH). Any decrease in the serum calcium concentration below normal is a stimulus for increased secretion of PTH [48] which, in turn, is a direct stimulator of the renal 1α-hydroxylase [49]. Recent promoter-reporter analyses have shown that both PTH and calcitonin stimulate transactivation of 1α-hydroxylase [50,51], although other studies have suggested that PTH can also effect changes in 1,25(OH)2D production via transient alteration in the phosphorylation status of the ferredoxin, which contributes electrons to 1α-hydroxylase [52]. A change in the serum phosphate concentration is the other major regulator of renal 1,25(OH)2D production; in adult humans, dietary phosphorus restriction causes an increase in circulating concentrations of 1,25(OH)2D to 80% above control values, an increase not due to accelerated metabolic clearance of this hormone [53]. Dietary phosphorous supplementation will have the opposite effect. Although the mechanism by which a drop in the serum phosphate level will increase renal 1,25(OH)2D production remains uncertain [54], there is no doubt that there exists a concerted, cooperative attempt of the calcium-phosphorous-PTH axis in man to regulate the conversion of 25OHD to 1,25(OH)2D in the kidney. For example, a drop in the serum calcium concentration will be immediately registered by the parathyroid cell calcium receptor, which will release its inhibition on PTH production and secretion. An increase in the circulating PTH will directly stimulate the renal 1α-hydroxylase, while a PTH-mediated phosphaturic response and a subsequent decrement in

1383

CHAPTER 79 Extra-renal 1α-Hydroxylase Activity and Human Disease

Kidney

Macrophage

24,25-D & 1,24,25-D

+

+

25-D

PTH

X X

_

1,25-D

[P]

[Ca]

1,25-D

25-D

X

X

PTH

[P]

[Ca]

FIGURE 2

Model distinguishing the regulation of 1α-hydroxylase in the proximal renal tubular epithelial cell of the kidney (left panel) and in the granuloma-forming disease-activated macrophage (right panel). In the kidney, the enzymatic conversion of substrate 25-hydroxyvitamin D (25-D) to product 1,25-dihydroxyvitamin D (1,25-D) is subject to negative feedback control with down-regulation of enzyme activity under the influence of 1) a calcium-mediated decrease in the circulating parathyroid (PTH); 2) a 1,25-D-mediated increase in the serum phosphate [P] level; and 3) a 1,25-D–mediated increase in vitamin D 24-hyroxylase activity (24,25-dihydroxyvitamin D (24,25-D) and 1,24,25-trihydroxyvitamin D (1,24,25-D) production). The macrophage lacks responsiveness to changes in the extracellular PTH and [P] and harbors little or no detectable vitamin D-24-hydroxylase (lack of regulation designated X).

the serum phosphate level will indirectly promote 1,25(OH)2D production. The macrophage 1α-hyroxylase, on the other hand, is immune to the stimulatory effects of PTH and phosphate [42,55] (right panel, Fig. 2). The macrophage plasma membrane is not enriched with PTH receptors [56], and there is no evidence that any PTH receptors which are resident in the macrophage membrane are responsive to PTH or PTHrP in terms of stimulating the protein kinase signaling pathways that are associated with stimulation of the renal 1α-hydroxylase. Similarly, the macrophage enzyme appears to be uninfluenced by changes in the extracellular phosphate concentration [42]. Moreover, exposure of activated macrophages expressing 1α-hydroxylase to a calcium ionophore stimulates the hydroxylation reaction [57], while increasing the extracellular calcium concentration has the opposite, inhibitory effect on the renal 1α-hydroxylase [58]. These observations appear to confirm the fact that the three most important extracellular signaling systems for the renal 1α-hydroxylase are not heeded by the macrophage enzyme and provide an explanation for why 1,25(OH)2D production by the macrophage in diseases like sarcoidosis is not subject to negative feedback control as reflected by a drop in the serum PTH concentration and an increase in the circulating calcium and phosphate level. Furthermore, with

the possible exception of insulin-like growth factor-1 (IGF-1) [59], there is no evidence that the macrophage 1α-hydroxylation reaction is influenced by any of the other endocrine factors, including estrogen, prolactin, and growth hormone, purported to increase the renal production of 1,25(OH)2D [60–63]. By contrast, macrophage 1α-hydroxylase activity is potently inhibited by anti-inflammatory agents, such as glucocorticoids, which have little or no effect on the renal enzyme. In vivo, this is likely to be due in part to the effects of glucocorticoids on macrophage differentiation and apoptosis. However, studies in vitro suggest that there is also direct inhibition of macrophage 1α-hydroxylase activity by glucocorticoids [40].

3. Macrophages Lack 1,25(OH)2D-directed 24-hydroxylase Activity The other major contributor to the circulating 1,25(OH)2D level is the activity of 24-hydroxylase. Like the 1α-hydroxylase, 24-hydroxylase is a hemebinding mitochondrial enzyme requiring NADPH, molecular oxygen, and magnesium ions [52,64]. The cDNA and gene sequences for human, rat, and chicken 24-hydroxylase, now referred to as CYP24/Cyp24, were cloned several years prior to CYP1α [65–67].

1384 As depicted in Fig. 2 (left panel), expression of CYP24 is stimulated in kidney cells by 1,25(OH)2D, especially if the protein kinase C (PKC) pathway is also upregulated [68–70]. PTH appears to exert an opposite, inhibitory effect on CYP24 gene transcription and 24,25(OH)2D synthesis [71]. There is dual impact of this mitochondrial, cytochrome P450-linked enzyme system on vitamin D and calcium balance in adult animals, including man. Because it is coexpressed in the kidney along with the 1α-hydroxylase, the first point of impact is on regulation of substrate 25OHD available to the 1α-hydroxylase. Like the 1α-hydroxylase, the 24-hydroxylase exhibits a preference for 25hydroxylated secosterol substrates [72]. Although its affinity for 25OHD is reported to be somewhat less than that of renal 1α-hydroxylase, its capacity for substrate is substantially greater [44]. Hence, when up-regulated under the influence of circulating or locallyproduced 1,25(OH)2D or diminished serum PTH levels, the 24-hydroxylase has the capacity to compete with 1α-hydroxylase for substrate 25OHD. Under physiological conditions, this state of competitive substrate deprivation for the 1α-hydroxylase will persist until the serum calcium and PTH concentration are normalized. The second point of impact of the vitamin D-24hydroxylase on the circulating 1,25(OH)2D concentration is at the level of catabolism of 1,25(OH)2D itself. Although both 25OHD and 1,25(OH)2D are metabolized by 24-hydroxylase [73], current data strongly suggest that the latter is the preferred substrate [64]. Considering the fact that the 24-hydroxylase is the initial step in the conversion of 1,25(OH)2D to nonbiologically-active, water-soluble, excretable metabolites of the hormone, up-regulation of this enzyme will contribute to the lowering of 1,25(OH)2D hormone levels. In contrast to precursor monocytic cells, the macrophage lacks detectable 24-hydroxylase activity (Fig. 2, right panel) [43]. Therefore, unlike the renal tubular epithelial cells and indeed other epithelial cells [74], macrophages do not possess the capability of shunting substrate 25OHD or the 1α-hydroxylase product 1,25(OH)2D down the catabolic 24-hydroxylase pathway. The net result is dysregulated overproduction of 1,25(OH)2D by the macrophage, escape of the hormone into the general circulation, and the eventual development of hypercalcemia. 4. MACROPHAGE 1α-HYDROXYLASE EXHIBITS RESPONSIVENESS TO IMMUNE CELL REGULATORS

The lack of negative feedback control on the 1α-hydroxylase expressed in the macrophage as just described can account for the failure to appropriately inhibit 1,25(OH)2D synthesis in inflammatory diseases like sarcoidosis. However, this does not adequately

MARTIN HEWISON AND JOHN S. ADAMS

explain the fact that 1,25(OH)2D production rates are increased well above normal in patients with these diseases at a time when the renal 1α-hydroxylase is inhibited. This observation suggests that there must be an alternative, “nonclassical” set of factors that stimulate the synthesis of 1,25(OH)2D by the macrophage but not by the kidney. Clinical observations from a number of investigative groups around the world indicate that sarcoidosis patients with diffuse, infiltrative pulmonary disease are at greater risk to develop dysregulated vitamin D metabolism. Cultured pulmonary alveolar macrophages (PAM) from such patients were more likely to synthesize more 1,25(OH)2D in vitro on a per cell basis than PAM from a host with less intense or no alveolitis [75,76]. These results led to the conclusion that the specific activity of the 1α-hydroxylase reaction in macrophages from patients with active pulmonary sarcoidosis was regulated by endogenously-synthesized factors, which also modulated the intensity of the host immune response. Of the various bioactive cytokines concentrated in the alveolar space of patients with active sarcoidosis [77,78], IFNγ was found to be the principal cytokine stimulator of the sarcoid macrophage 1α-hydroxylation reaction [77]; by itself at maximally effective concentrations in vitro, IFNγ increased basal hydroxylase activity over four-fold. However, it is now clear that other immunomodulators are also able to stimulate macrophage 1α-hydroxylase, including other cytokines such as tumor necrosis factor α (TNFα) [43,79] and interleukin-2 (IL-2) [55], as well as pathogen-associated peptides such as bacterial lipopolysaccharide (LPS) [42,55]. As all of these factors have been implicated in the maturation of macrophage responsiveness within the innate immune system, it seems likely that up-regulated 1α-hydroxylase activity is a common feature of activated macrophages. However, two key questions remain: first, is there a specific mechanism involved in up-regulation of macrophage 1α-hydroxylase activity; second, why is macrophage 1α-hydroxylase activity pathologically elevated in patients with inflammatory diseases, such as sarcoidosis? These issues are addressed in the following sections. a. Cytokines Despite the advent of gene sequence information for CYP1α and in particular promoter analyses for classical 1α-hydroxylase regulators such as PTH, our current understanding of the molecular mechanisms involved in regulating macrophage 1α-hydroxylase activity is still poor. It seems likely that several pathways are involved—for example, IFNγ signals via Janus Kinase 1 (JAK1) and JAK2 with subsequent phosphorylation of signal transducers and activators of transcription alpha (STAT alpha) and subsequent transregulation of target genes via cis-acting

1385

CHAPTER 79 Extra-renal 1α-Hydroxylase Activity and Human Disease

promoter elements [80]. However, the JAK/STAT pathway is essential for the effects of many cytokines and growth factors, including some members of the interleukin family (e.g. IL-2 and IL-6) [81]. The JAK/ STAT system may also interact with other signaling pathways, including p38 mitogen-activated protein kinase (MAP kinase) and nuclear factor-κB (NF-κB) [82,83]. The net effect of this extensive cross-talk means that a variety of factors, including the cytokines outlined above together with appropriate growth factors such as granulocyte-macrophage colonystimulating factor (GM-CSF), may be able to stimulate 1α-hydroxylase in macrophages. Disappointingly, the availability of CYP1α gene sequence information and promoter-reporter constructs has shed no further light on the nuclear transactivation factors that mediate cytokine-induced synthesis of 1,25(OH)2D. Nevertheless, the CYP1α promoter region includes putative AP-1 and NF-κB binding sites, which are potential targets for cytokine-regulation of 1α-hydroxylase [50,51]. Alternatively signaling via cytokines such as IFNγ may lead to the activation of other, calcium-dependent pathways in the macrophage, specifically the protein kinase C (PKC) [84] and phospholipase A2 (PLA2) pathways [85,86]. Because the macrophage 1α-hydroxylase was not influenced by attempts to directly stimulate or inhibit PKC, attention has focused on the PLA2 pathway and the endogenous arachidonic acid metabolic cascade as the signal transduction pathway of most influence over the macrophage enzyme. Further dissection of the intracellular arachidonate metabolic pathway in this cell demonstrated that signal transduction through the 5-lipoxygenase pathway, specifically with the generation of leukotriene C4 (LTC4), was most critical to an increase in 1,25(OH)2D synthesis [87]. These studies were extended to investigate another compound with potential actions in the PLA2arachidonic acid pathway, the 4-amino quinoline derivative chloroquine. 1,25(OH)2D synthesis by macrophages was completely inhibited by exposure to 10−6 M chloroquine in vitro [36]. Furthermore, this effect is independent of chloroquine’s apparent ability to alter the pH of intracellular organelles. When given orally to a hypercalcemic patient with sarcoidosis, chloroquine [34,36] or its analog hydroxychloroquine [35] can effectively reduce the serum 1,25(OH)2D and calcium concentration within a matter of 36 hours. b. Lipopolysaccharide (LPS) LPS is a bioactive lipid extractable from the cell wall of infectious microorganisms, including the mycobacterium. On macrophages, LPS interacts with a complex that includes the cell surface CD14 receptor and toll-like receptor 4 (TLR4), together with the accessory proteins MD-2 and MyD88 [88]. TLR4 is one of ten TLR proteins,

similar to the TNF receptor family, that function as pathogen-recognition receptors and which signal via NF-κB and p38 MAP kinase [89]. TLR4/CD14 is strongly expressed on cells from the immune system, including macrophages and dendritic cells (DCs), and LPS is a potent inducer of 1α-hydroxylase in human monocyte/macrophage-like cells [42,55] (Fig. 3). TLRs are also expressed by epithelial cells at “barrier” sites, including the skin, lungs, gastrointestinal tract, and distal nephron. Here, as with macrophages, TLR expression is able to support LPS inducibility of 1,25(OH)2D production. In recent studies, Hewison and co-workers demonstrated the presence of CD14 and TLR4 on cortical collecting duct HCD cells, but not proximal tubule HKC-8 cells [90]. As a consequence, HCD cells showed potent induction of 1α-hydroxylase activity in response to both PTH and LPS, while HKC-8 cells responded to PTH alone. Thus, it seems likely that TLR expression and signaling acts as a pivotal mechanism in regulating extra-renal 1α-hydroxylase activity. In fact, the most reproducibly effective stimulation of the macrophage 1α-hydroxylase

Macrophage CD14 TNFα IL-1β MD-2 TLR4 LPS hsp70 IFNγ

CD14 IFNr

VDR

iNOS CYP1α

IDBP

1,25-D

NO + e− O2 P450 1α 25-D IDBP 25-D

FIGURE 3 Proposed mechanism for the amplification of macrophage 1α-hydroxylase. Macrophage stimulatory agents lipopolysaccharide (LPS), interferon-γ (IFNγ), and heat-shock proteins (hsps) with their respective cell-surface receptor molecules (LPS and hsps: CD14, toll-like receptor 4 (TLR4), MD-2 complex. IFNγ: IFN receptor (IFNr). This leads to: 1) up-regulation of CYP1α mRNA expression and increased levels of 1α-hydroxylase protein (P4501α); 2) up-regulation of iNOS expression. Nitric oxide (NO) is synthesized from an extracellular source of molecular oxygen (O2) and L-arginine (L-arg). NO can serve as an electrondonating source for enzymatic conversion of 25-hydroxyvitamin D (25-D) to 1,25-dihydroxyvitamin D (1,25-D). 3) increased conversion of 25-D to 1,25-D. The transport of 25-D to mitochondrial 1αhydroxylase and 1,25-D to nucleus is facilitated by hsp70-like intracellular vitamin D–binding proteins (IDBP). NO and 1,25-D act in an intracrine fashion to up-regulate expression of the cytokines interleukin-1β (IL-1ß), tumor necrosis factor α (TNFα), and the LPS receptor molecules; TNF and CD14 promote intracrine stimulation of NO and 1,25-D synthesis.

1386 in vitro is achieved by coexposure of macrophages to IFNγ and LPS [55]. LPS and IFNγ commonly activate different signal transduction pathways, but as outlined previously, there is potential for cross-talk between these pathways, which may have a significant impact on transactivation of CYP1α. Notably, IFNγ and LPS are also the two most effective stimulators of nitric oxide (NO) synthesis in macrophages, and this led Adams and co-workers to hypothesize that production of NO and 1,25(OH)2D in macrophage-like cells may be functionally linked [91–93]. c. Nitric Oxide (NO) The generation of NO in the macrophage is under the control of the enzyme inducible nitric oxide synthase (iNOS) [94]. In contrast to the more stringently regulated, constitutively expressed isoforms of the enzyme (cNOS) that are localized to endothelial cells and neurons, are regulated by calcium, and are capable of producing only modest amounts of NO, the calcium-independent iNOS remains tonically active when “induced” and is capable of generating large quantities of NO in and around the cell [95]. It is therefore interesting that two of the major stimulators of the human macrophage 1α-hydroxylase, IFNγ and LPS, are also key transcriptional regulators of the iNOS gene [96,97], which is itself a cytochrome P450-linked oxidase [98]. These observations coupled with the fact that NO has established inhibitory effects on other cytochrome P450s [99,100] suggested a possible link with the enzymes involved in vitamin D metabolism. Data indicate that NO may act as an alternative to NADPH as a source of unpaired electrons for the 1α-hydroxylase reaction in macrophages [91,92]. However, as the amount of NO generated inside the macrophage continues to increase, there is a reflex decrease in 1,25(OH)2D production [93], suggesting that there is a built-in limit on the ability of the cell to produce active vitamin D. This inhibitory effect of NO on the macrophage 1α-hydroxylase is almost certainly due to competition of NO with oxygen for binding to the heme center of the enzyme. A similar effect has been very recently demonstrated for a number of heme-containing enzymes [99,100], including those involved in steroid hormone metabolism [101]. d. Other Potential Regulators of the Macrophage 1α-hydroxylase Another potential autoregulator of macrophage 1,25(OH)2D synthesis is the stressinduced heat shock-70 (hsp70) family of proteins [102,103]. These proteins are ubiquitously distributed in the nuclear, cytoplasmic, mitochondrial, and endoplasm reticular compartments of eukaryotic cells. Hsp70s were first recognized as heat-shock-responsive ATP-binding proteins with ATPase activity [104]. Stress-induced proteins have a well-established link

MARTIN HEWISON AND JOHN S. ADAMS

with immune responses and inflammatory disease [104,105]. This stems first from the fact that peptides bound or linked to HSPs can elicit potent antigenspecific immunity [106]. In addition, proteins such as hsp70 are able to stimulate cells of the innate immune system directly, and thus act as “danger”-signaling molecules in a similar fashion to LPS [107]. These observations are supported by studies which have shown that TLRs act as exogenous and enodogenous signal transduction pathways for hsp responses [108,109]. A specific link between hsp/TLR signaling and vitamin D metabolism has yet to be studied, but it seems likely that hsps will act as important regulators of the macrophage 1α-hydroxylase. This may be due in part to TLR/NF-κB/p38 MAP kinase-mediated transcriptional regulation as detailed in section 4.b. However, hsps are also characterized functionally by their ability to bind and release hydrophobic segments of an unfolded polypeptide chain in an ATP-hydrolytic reaction cycle [103]. This so-called “chaperone” function of hsp70 is critical for a number of intracellular protein-protein interactions [105], and in the targeting and translocation of molecules across the endoplasmic and mitochondrial membrane [105,110–112]. In macrophage-like cells, hsp70 expression is known to be induced by both physical (i.e. heat) and cytokine (i.e. IFN) stimuli, but its expression is also dramatically enhanced by 1,25(OH)2D [113]. Furthermore, in a series of studies we have shown that proteins from the hsp70 family have a high capacity for intracellular binding of 25-hydroxylated vitamin D metabolites [114]. These hsp70 homologs now termed intracellular vitamin D–binding proteins (IDBPs) were originally identified as vitamin D–resistant cells from New World primates [115]. Initially, these proteins were thought to be the functional basis for the end-organ resistance to 1,25(OH)2D in New World primates. However, subsequent data have shown that the IDBPs are in fact potent activators of 1,25(OH)2D–induced gene transactivation [116]. Rather, the underlying cause of vitamin D resistance in New World primates has now been shown to be due to constitutive overexpression of a vitamin D response element-binding protein (VDRE-BP) from the heterogeneous nuclear ribonucleoprotein in the A family (see Chapter 21), with the IDBPs acting as a putative compensatory mechanism by acting as intracellular chaperones for 1,25(OH)2D [117]. Crucially, cells overexpressing IDBPs also showed increased synthesis of 1,25(OH)2D, indicating that their chaperone activity was not restricted to increased VDR-mediated transactivation [116]. Thus, it is possible that up-regulated IDBP expression in disease-activated macrophages is another key factor in the increased synthesis of 1,25(OH)2D by these cells. Because of its

1387

CHAPTER 79 Extra-renal 1α-Hydroxylase Activity and Human Disease

capacity to bind 25OHD as well as 1,25(OH)2D, hsp70/IDBP may serve to concentrate, on a relatively low-affinity, high-capacity binding protein, substrate for the 1α-hydroxylase from the general circulation. In fact, by virtue of their organelle-targeting sequences, hsp70s or related molecules may be critical in the directed translocation of 25OHD to the inner mitochondrial membrane where the 1α-hydroxylase actually resides. In this respect, IDBPs may provide a crucial link between the intracellular organelles and endocytic proteins, such as megalin, which are known to act as an interface with serum-bound steroid hormones [118]. Megalin is strongly expressed in the proximal renal tubules where it plays a pivotal role in transporting 25OHD [119], but its role in directing macrophage metabolism of vitamin D remains unclear. The actions of megalin are discussed in greater detail in Chapter 10.

IV. LOCAL IMMUNOREGULATORY EFFECTS OF ACTIVE VITAMIN D METABOLITES A. Intracrine/Autocrine Action on the Monocyte/Macrophage The immunomodulatory properties of 1,25(OH)2D and synthetic analogs of vitamin D are discussed extensively in Chapter 36, which also outlines potential therapeutic applications for auto-immune disease and host-graft rejection. However, a key question remains as to the normal physiological function of locally synthesized 1,25(OH)2D. Furthermore, what is the

Paracrine Action of 1,25-D

+ 1-OHase

25-D

Endocrine Action of 1,25-D Hypercalcemia =“tip of the iceburg”

+ 1,25-D

MACROPHAGE

1-OHase

25-D

1,25-D

1,25-D

antigen

+ LYMPHOCYTES lymphokines

Effect endocrine

MACROPHAGE

+ IL-1, TNF

Target Tissues bone, gut

IL-1, TNF

+

lymphocyte

paracrine

macrophage

intracrine

antigen − LYMPHOCYTES

increasing 1,25-D production

Autocrine Action of 1,25-D

purpose of 1,25(OH)2D production in patients with granuloma-forming disease like sarcoidosis? Is local synthesis of the hormone by macrophages beneficial or detrimental to the host? These are important questions that investigators in a number of centers around the world have been addressing since the 1980s when it became known that the activated, circulating monocytes and tissue macrophages expressed the VDR [120,121]. Expression of the receptor for the active vitamin D hormone indicated that the macrophage could actually be a target for the 1,25(OH)2D that the cell itself was making. Indeed, investigators have suggested that 1,25(OH)2D has the potential to interact with the monocyte/macrophage in either an intracrine or autocrine mode [122,123] (left panel, Fig. 4). For example, incubation with a VDR-saturating concentration of 1,25(OH)2D increases IL-1ß expression by eightfold and decreases by 1000-fold the concentration of stimulator lipopolysaccharide (LPS) required to achieve maximal IL-1ß gene expression [124]. This extraordinary priming effect of the 1,25(OH)2D for LPS stimulation of the IL-1 gene can also be observed for another monokine gene product, TNF [125] and is due to 1,25(OH)2D–mediated induction of the gene for CD14, which acts as an accessory receptor with TLR4. The multiplicity and complexity of actions of 1,25(OH)2D on the macrophage are detailed in Chapter 36. Suffice it to say that it is now widely accepted that the actions of hormone are directed toward stimulation of macrophage function. For example, 1,25(OH)2D is known to enhance giant cell formation [126], monokine production [127–129], and cytotoxic function [130,131]. Conversely, 1,25(OH)2D acts as a

lymphokines

FIGURE 4 Schemes for local production and action of macrophage-derived 1,25-dihydroxyvitamin D (1,25-D). In an autocrine or intracrine mode (left panel), 1,25-D promotes antigen handling and monokine production. In a paracrine mode (middle panel), 1,25-D acts in a negative feedback fashion to “brake” what may turn out to be an overexuberant lymphocyte response to presented antigen and local monokines. If the immune response and 1,25-D production are persistent, then the hormone can escape the local inflammatory microenvironment and act in an endocrine mode (right panel) to alter host calcium balance.

1388 potent suppressor of antigen presentation by both macrophages [132] and other professional antigen presenting cells (APCs), such as dendritic cells (DCs) [133]. The latter has attracted considerable recent attention because of the link between vitamin D responsiveness in DCs and macrophages. Specifically, the ability of 1,25(OH)2D and vitamin D analogs to inhibit DC function appears to be dependent on the suppression of DC differentiation, thereby maintaining DCs in an immature, immune tolerant state [134]. Importantly, studies in vitro indicate that this is accompanied by a reinduction of the macrophage marker CD14 [135,136], suggesting that the suppressive effects of 1,25(OH)2D on DCs are counterbalanced by enhanced macrophage development/function. In this way, 1,25(OH)2D may play an important role in modulating the balance between innate (macrophage/phagocytic) and acquired (DC/APC) immune responses. The significance of this has been further underlined through recent studies by us and others who have shown that monocyte-derived DCs express the same 1α-hydroxylase as renal cells and macrophages and are able to synthesize significant levels of 1,25(OH)2D [137,138]. As a result, DC differentiation and function was potently suppressed by 25OHD, as well as 1,25(OH)2D, suggesting that this may be the pivotal mechanism linking vitamin D status with normal immune function. This process would also be consistent with the overproduction of 1,25(OH)2D observed in diseases such as sarcoidosis. Specifically, the enhanced localized synthesis of 1,25(OH)2D by DCs following an immune challenge would further stimulate tissue macrophage population. This, in turn, may lead to even higher levels of 1,25(OH)2D production and in some cases potential spill-over into systemic levels of the hormone. These functional consequences of hormone action indicate that elaboration of 1,25(OH)2D by activated macrophages and DCs in human diseases like sarcoidosis is important in modulation of the local cellular immune response to the granuloma-causing antigen, promoting antigen processing, containment, and destruction. The system is designed for maximal efficiency in that the hormone interacts with the VDR in the same cell in which the hormone is made. Thus, under normal circumstances, relatively high intracellular and local concentrations of hormone can be achieved to modulate macrophage/DC action without having a generalized, endocrine effect on the host (see Fig. 4).

B. Paracrine Action on Lymphocyte As outlined above, vitamin D is a potent autocrine/ intracrine regulator of both innate and acquired

MARTIN HEWISON AND JOHN S. ADAMS

immune reponses. The effects of 1,25(OH)2D on acquired, lymphocyte-directed immune reponses are due, at least in part, to indirect actions via the suppression of antigen presentation by macrophages and DCs. However, there is also a breadth of evidence for direct effects of 1,25(OH)2D on lymphocytes [139–142]. Specifically, if lymphokine stimulation of macrophage 1,25(OH)2D synthesis were persistent because of difficulty in macrophage-mediated elimination of the offending antigen, then one might conceive of a situation in which the lipid soluble vitamin D hormone escapes the confines of the macrophage (middle panel, Fig. 4). Once outside of the macrophage, 1,25(OH)2D would be free to interact in a paracrine fashion with antigen-activated T- and B-lymphocytes in the local inflammatory microenvironment. The many reported actions of active vitamin D metabolites on cells of the T-lymphocyte (T-cell) lineage are also described in Chapter 36. In general and in contrast to stimulatory effects of the hormone on monocyte-macrophage cells, 1,25(OH)2D and most of the nonhypercalcemic analogs of vitamin D will inhibit T-cell responsiveness to mitogen or antigen challenge. Interaction of the VDR with its cognate ligand in activated lymphocytes and natural killer (NK) cells inhibits cellular proliferation [143,144], generally decreases lymphokine production [143,145], and inhibits T-cell-directed B-cell immunoglobulin synthesis [146] and delayed-type hypersensitivity reactions [147]. More recent studies have shown that 1,25(OH)2D and vitamin D analogs also influence the nature of T-cell responses by promoting specific T-cell subgroups. Initial studies indicated that 1,25(OH)2D preferentially enhanced a shift from potentially damaging cellular-based T-cell responses involving type 1 helper T-cells (Th1) to a more benign humoral-based immunity via Th2 cells [148–150]. However, more recent work has highlighted an exciting alternative role for 1,25(OH)2D as a potent stimulator of suppressor T-cells also termed T-regulatory cells [151–153]. In view of the importance of T-regulatory cells in directing immune tolerance, the implications for the effects of 1,25(OH)2D on these cells are considerable both in terms of its therapeutic potential and as a basis for a clearer role for vitamin D in normal immune responses. These and other issues are discussed in much greater detail in Chapter 36. In summary it is now clear that immunomodulatory actions of either exogenously added or locally synthesized 1,25(OH)2D can be broadly divided into either autocrine/intracrine activation of macrophage function or paracrine suppression of lymphocyte function. We have postulated that this apparent paradox in immunoactions of the hormone, to dampen lymphocyte activity

CHAPTER 79 Extra-renal 1α-Hydroxylase Activity and Human Disease

while stimulating monocyte/macrophage function, is designed to maximize the ability of the host to combat and contain the granuloma-causing antigen, while controlling the potentially self-destructive lymphocytic response to that offending antigen. In other words, in order to prevent “overstimulation” of lymphocytes by monokines elaborated at the site of inflammation, some hormone produced by the macrophage will escape the confines of the cell in which it was made, will interact with neighboring, VDR-expressing, activated lymphocytes, and will tend to restrain what might be an otherwise overzealous, self-destructive T-cell and B-cell response to the offending antigen. As depicted in the right panel of Fig. 4, only at times of heightened immunoreactivity (i.e. extraordinary disease activity) does monokine production escape the confines of the site of inflammation and spill over into the general circulation causing elevated 1,25(OH)2D concentrations. This model would indicate that the endocrine actions of a locally produced vitamin D metabolite that escapes the inflammatory microenvironment is the exception rather than the rule. It also suggests that 1,25(OH)2D is by design an immunomodulatory cytokine in these lymphoproliferative diseases and not a hormone meant to modulate calcium homeostasis in the host. The overall importance of vitamin D as part of the normal immunomodulatory machinery is still open to some debate. Studies in vivo using animals [154–157] and humans [158,159] have highlighted significant immune abnormalities associated with vitamin D deficiency. Likewise, vitamin D supplementation has been shown to have positive effects on immune responses [160]. The extent to which this is mediated via macrophage 1α-hydroxylase and the contribution of this mechanism to the effects of UV exposure on diseases such as TB [161] are still under scrutiny and are discussed in much greater detail in Chapter 36. However, further light has been shed on this issue following the recent development of VDR [162] and 1α-hydroxylase [163] knockout mice. Although initial analysis of the VDR knockout mouse revealed minimal changes in the immune function [164], subsequent studies have documented dysregulated T-helper cell function in these animals [165]. In a similar fashion both VDR and 1α-hydroxylase knockout mice have characteristic lymph nodes dysplasia that is consistent with abnormal DC function [134,163]. These observations have underlined the potential importance of vitamin D metabolism and signaling as a regulator of immune responses and in future studies it will be interesting to assess the way these animals respond to specific inflammatory diseases.

1389

C. Accumulation of 1,25(OH)2D at Sites of Inflammation 1. GRANULOMA-FORMING DISEASES

Although much effort has been expended to ascertain the immunomodulatory potential of active vitamin D metabolites in human disease, considerably less is known about the immunoactions of these endogenously synthesized molecules in vivo in man and animals. If 1,25(OH)2D is truly a naturally occurring “cytokine,” then one should be able to document accumulation of the metabolite at sites of inflammation and show that the inflammatory cells at this site are under the influence of the locally-produced vitamin D metabolite. This was first accomplished by Barnes et al. [166], not in sarcoidosis, but in the infectious granuloma-forming disease, tuberculosis. They determined that the pleural space in nonhypercalcemic/calciuric patients infected with mycobacterium tuberculosis was one such site of 1,25(OH)2D accumulation. They detected a steep gradient for free, biologically-active 1,25(OH)2D across the visceral pleura in patients’ tuberculous effusions (but not in patients with nontuberculous effusions), showed that PPD-reacting T-cell clones from these patients expressed the VDR, and determined that the stimulated proliferation of these T-cell clones was susceptible to 1,25(OH)2D–mediated inhibition. They also showed that the pleural fluid of these patients contained an IFN-like peptide that stimulated the synthesis of 1,25(OH)2D by heterologous sarcoid macrophages [167]. Collectively, these data supported the idea put forward by Rook and colleagues [168] that there exists in the pleural microenvironment of patients with active pulmonary tuberculosis a system whereby: 1) mycobacterium-activated macrophages are stimulated to make 1,25(OH)2D; 2) this synthetic reaction is supported by proliferating and lymphokine- (i.e. IFNγ) producing lymphocytes in the local site of inflammation; and 3) the local accumulation of lymphokines, in turn, acts to further augment the local production of 1,25(OH)2D by the macrophage (see Fig. 4). Investigators [169] have viewed this sort of positive feedback effect of IFNγ on macrophage 1,25(OH)2D production in vivo as an efficient mechanism for dealing with antigens, like myco-bacteria, the “sarcoid antigen,” or certain viruses that are difficult for the host to irradicate. 2. OTHER INFLAMMATORY DISEASE STATES

Mawer and colleagues [170,171] have demonstrated substrate-dependent accumulation of 1,25(OH)2D in the synovial fluid of patients with “inflammatory arthritis,” including subjects with rheumatoid arthritis. These investigators speculate that the local increase in 1,25(OH)2D synthesis may contribute to periarticular

1390

MARTIN HEWISON AND JOHN S. ADAMS

bone loss in such individuals. A positive tissue-toserum gradient for 1,25(OH)2D has also been suggested in peritoneal dialysis patients, particularly when afflicted with peritonitis [164,165]; peritoneal macrophages from such patients have been shown to metabolize 25OHD to 1,25(OH)2D in vitro [172,173].

V. HUMAN DISEASES ASSOCIATED WITH THE EXTRA-RENAL OVERPRODUCTION OF ACTIVE VITAMIN D METABOLITES (see Table I) A. Granuloma-Forming Diseases 1. SARCOIDOSIS

Sarcoidosis is the human disease most commonly complicated by endogenous vitamin D intoxication [1–6,14–19,174–177]. In the most expansive studies published on the topic, roughly 10% of patients with sarcoidosis will develop hypercalcemia [177], and up to 50% will suffer from hypercalciuria [177] at some time during the course of their disease. In their retrospective, worldwide review of serum calcium concentrations in 3,676 patients with sarcoidosis, James et al. [177] recorded an 11% incidence of hypercalcemia (serum calcium ≥ 10.5 mg/dL). Studdy et al. [176]

TABLE I Human Disease Associated with 1,25-Dihydroxyvitamin D-mediated Hypercalcemia/Hypercalciuria Granuloma-forming diseases Infectious Tuberculosis Leprosy Candidiasis Crytococcosis Histoplasmosis Coccidioidomycosis AIDS-related pneumocystis Noninfectious Sarcoidosis Silicone-induced granulomatosis Eosinophilic granuloma Wegener’s granulomatosis Langerhans cell histiocytosis Berylliosis Infantile fat necrosis Crohn’s disease Malignant lymphoproliferative disease Hodgkin’s disease Non-Hodgkin’s lymphoma Dysgerminoma/seminoma

[182–191] [192] [193] [194] [195] [196] [197] [14–19] [198] [199] [200] [201] [202] [203] [204,205] [209–211, 213] [212, 214–216] [217]

studied 547 patients with biopsy-proven sarcoidosis in Great Britain and found hypercalcemia to be 38% more frequent in men than women and more common among Caucasians than individuals of West Indian descent. Although not systematically studied, the frequency of hypercalcemia among patients with sarcoidosis tends to be consistently higher in North America than in Northern Europe [178]. This is perhaps due to the lower latitude and more direct sunlight exposure in the United States. Although fractional intestinal calcium absorption may be increased under the influence of 1,25-(OH)2D and fractional urinary calcium excretion may be decreased in patients with renal insufficiency [38], the principal source of calcium which accumulates in the circulation in this disease is the skeleton. This fact is perhaps most strongly confirmed by the observations of Rizatto et al. [179] who documented in serial fashion a significant decrease in bone mineral density in a group of patients with chronic active sarcoidosis in whom anti-inflammatory agents, including glucocorticoids, were not used in management compared to age- and sex-matched control subjects. This fact is confirmed by the long-standing observations that hypercalcemia persists in patients with active sarcoidosis in absence of ingested calcium and may be contributed to by increased bone resorption [33]. The proximal cause of bone loss is increased osteoclastmediated bone resorption [180] and does not require the presence of extensive granulomata in the bone [181]. These observations suggest that an osteoclast activating factor (OAF) exists in this disease. One such bonafide OAF is of course already known, namely 1,25(OH)2D. 2. TUBERCULOSIS

Of the other human granuloma-forming diseases reported to be associated with vitamin D metabolitemediated hypercalcemia, tuberculosis is the most commonly reported aside from sarcoidosis. Hypercalcemia has been recognized as a complication of infection with mycobacterium tuberculosis for over eight decades [182]. That this disturbance in calcium balance is caused by the extra-renal overproduction of an active vitamin D metabolite was confirmed by investigators in the mid 1980s [183,184]. As is the case with sarcoidosis, the circulating vitamin D metabolite causing hypercalcemia 1) appears to be 1,25(OH)2D [185,186]; 2) is synthesized by disease-activated macrophages [187,188], 3) is abnormally responsive to small changes in the serum concentration of substrate 25OHD [189], and 4) is reducible under the influence of glucocorticoid in vivo [190,191]. The prevalence of hypercalcemia

CHAPTER 79 Extra-renal 1α-Hydroxylase Activity and Human Disease

in patients may be as high as 26% [182] and may be even higher, particularly in the era of AIDS, because of frequent association of hyopalbuminemia (i.e. from malnutrition) in patients with tuberculosis. The source of 1,25(OH)2D in this disease is, as it is in all of the other granuloma-forming diseases, extra-renal [183] most likely arising from the macrophage [187]. 3. OTHER INFECTIOUS DISEASES

Hypercalciuria or overt hypercalcemia has also been observed in a number of infectious diseases, most characterized by widespread granuloma formation and macrophage proliferation in infected tissue. Included among these diseases are leprosy [192], disseminated candidiasis [193], crytococcosis [194], histoplasmosis [195], and coccidioidomycosis [196]. Hypercalcemia in most of these conditions has been documented to be associated with inappropriately elevated serum concentrations of 1,25(OH)2D. The true prevalence and incidence of hypercalcemia and hypercalciuria in patients with these diseases in unknown. However, it is likely that this complication of dysregulated vitamin D metabolism and action associated with these diseases will increase in frequency as the number of immunocompromised patients, especially those with AIDS, increases worldwide. For example, hypercalcemia in association with elevated circulating levels of 1,25(OH)2D has been reported in an AIDS patient with pneumocystis [197]; both serum calcium and 1,25(OH)2D concentrations dropped in this patient with successful treatment of his opportunistic infection. 4. NONINFECTIOUS GRANULOMA-FORMING DISEASES

The syndrome of extra-renal overproduction of 1,25(OH)2D has also been documented in adult patients with widespread silicone-induced granulomata [198], eosinophilic granuloma [199], Wegener’s granulomatosis [200], and Langerhans cell histiocytosis [201]. Although the active vitamin D metabolite was not measured, dysregulated calcium balance in the granuloma-forming pulmonary disease berylliosis is also attributed to the extra-renal production of 1,25(OH)2D [202]. In addition, 1,25(OH)2D–mediated hypercalcemia has been observed in newborn infants suffering from massive subcutaneous fat necrosis [203]; this is a transient disorder associated with birth trauma and characterized histopathologically by the proliferation of “foreign body-type” giant cells around cholesterolshaped crystals in necrotizing, subcutaneous adipose tissue. Finally, there are also reports of elevated serum levels of 1,25(OH)2D and associated hypercalcemia in patients with inflammatory bowel disease [204]. The possible impact of extra-renal 1α-hydroxylase in this

1391

clinical situation is an exciting new development, in part because of the prevalence of Crohn’s disease particularly in developed countries [205], but also because of several recent reports which have documented expression of 1α-hydroxylase along the gastrointestinal tract [24,206–208].

B. Malignant Lymphoproliferative Disorders By the 1980s, data was accumulating to suggest that a vitamin D–mediated disturbance in calcium metabolism was not confined to patients with granuloma-forming diseases and could also be observed in patients with lymphoproliferative neoplasms [209–212]. More recent reports [213,214] indicate that the extra-renal overproduction of 1,25(OH)2D is the most common cause of hypercalciuria and hypercalcemia in patients with non-Hodgkin and Hodgkin lymphoma, especially in patients with B-cell neoplasms, whether or not the tumor is associated with AIDS in the patient [212]. In fact, in the Seymour study [206] 71% of normocalcemic patients with non-Hodgkin lymphoma had hypercalciuria (fractional urinary calcium excretion >0.15 mg/dL glomerular filtrate) and most of these had serum 1,25(OH)2D levels that were above the mid range of normal or frankly elevated. As is the situation with hypercalciuric/calcemic patients with sarcoidosis or other granuloma-forming disease and elevated circulating 1,25(OH)2D levels, the serum concentrations of PTH are suppressed and PTHrP normal (i.e. not elevated) in lymphoma patients, indicative of the state of dysregulated overproduction of the active vitamin D hormone. Results of clinical studies of hypercalcemic patients with lymphoma pre and post successful anti-tumor therapy [212–215] are most compatible with either the tumor being an immediate source of an active vitamin D metabolite or the source of a soluble factor (i.e. peptide), which stimulates the production of 1,25(OH)2D in the kidney or in other inflammatory cells. Recent data from our group suggest that the latter is the case. Specifically, we carried out extensive analysis of a patient with hypercalcemia and raised circulating levels of 1,25(OH)2D associated with a splenic B-cell lymphoma [216]. The abnormalities in serum 1,25(OH)2D and calcium were corrected following resection of the spleen, and subsequent immunonhistochemical analysis of this tissue revealed increased expression of 1α-hydroxylase in macrophages adjacent to the tumor, but not in the tumor itself. This raises further potentially important questions: 1) what is the nature of the tumor-derived factor that is able to stimulate macrophage 1α-hydroxylase?

1392 2) Is macrophage-derived 1,25(OH)2D a contributing factor to the hypercalcemia associated with other types of tumors? In seeking to answer the latter question, current studies within our group have focused on the expression and function of macrophage 1α-hydroxylase in dysgerminomas, gonadal tumors that are associated with granulomata and which have shown previously to be linked to hypercalcemia [217].

C. Non-Granuloma-forming Conditions As outlined in section II.C, cloning of the gene for 1α-hydroxylase (CYP1α) has enabled a much more comprehensive appraisal of the tissue distribution of this enzyme than was previously available. Indeed, the human cDNA for CYP1α was cloned from keratinocytes, a well-established extra-renal source of 1,25(OH)2D [218,219]. Studies have defined both the renal and extra-renal tissues that express 1α-hydroxylase [23,24]. Specific sites of interest include skin [24,218,219], prostate [220], placenta [24,221,222] (one of the first tissues to show extra-renal synthesis of 1,25(OH)2D [223]), parathyroids [224], vasculature [225,226], gut [24,206–208], brain [24], and pancreas [24]. The precise function of 1α-hydroxylase expression at these sites is currently the focus of considerable attention. In tissues such as the prostate [220,227], colon, [206–208], and parathyroids [224,228], the enzyme has been postulated to fulfill an antitumor function by increasing local concentrations of antiproliferative 1,25(OH)2D, and this is discussed in greater detail in Chapters 93–96. However, in many extrarenal tissues particularly the pancreas, gut, vasculature, and brain, it is probable that 1α-hydroxylase will fulfill a function which is more akin to the macrophage, namely as a generator of immunomodulatory 1,25(OH)2D. For example, the most potent activator of 1α-hydroxylase in cultured human endothelial cells [225], as well as epithelial cells from the distal nephron [90], is LPS, with both cell types, like macrophages, showing strong expression of CD14/TLR4. Thus, the function of extra-renal 1α-hydroxylase may be far more diverse than originally thought: putative roles in preclampsia [221], implantation [222], and vascular disease [225] have been proposed in addition to its link with common cancers [206–208,220,227]. Nevertheless, at the present moment in time the most well-documented pathological paradigm for extra-renal 1α-hydroxylase remains the overproduction of 1,25(OH)2D associated with granulomatous disease. The clinical management of this is therefore discussed in greater detail in the following sections.

MARTIN HEWISON AND JOHN S. ADAMS

VI. DIAGNOSIS, PREVENTION, AND TREATMENT OF THE PATIENT WITH ENDOGENOUS VITAMIN D INTOXICATION A. Diagnosis The diagnosis of so-called “endogenous” vitamin D intoxication is made when the following three criteria are met. First is the presence of hypercalciuria and/or hypercalcemia in a patient with an inappropriately elevated serum 1,25(OH)2D (i.e. the serum 1,25(OH)2D concentration is not suppressed below 20 pg/ml). Second is the presence in the serum of an appropriately suppressed PTH level if the patient’s free (ionized) serum calcium concentration is high; this is evidence that the calcium sensing receptor in the plasma membrane of the host’s parathyroid cell is normally operative. This distinguishes the patient with primary hyperparathyroidism and elevated 1,25(OH)2D levels, in whom the calcium-sensing receptor signal transduction pathway to control PTH synthesis and release is disrupted, and from the individual with endogenous vitamin D intoxication and elevated 1,25(OH)2D levels. The other major exception here is the patient with absorptive hypercalciuria who possesses, as a primary or secondary abnormality, an inappropriately elevated circulating 1,25(OH)2D concentration [229]. Third is the exclusion of exogenous vitamin D intoxication arising from the oral or parental administration of an active vitamin D metabolite or the substrate for endogenous synthesis of an active vitamin D metabolite. The most common cause of exogenous vitamin D intoxication occurs with the ingestion or injection of large doses of vitamin D2 (ergocalciferol) or vitamin D3 (cholecalciferol), and can usually be detected by measuring a frankly elevated serum 25OHD level; most, if not all, currently available serum assays for 25OHD do not distinguish 25OHD2 from 25OHD3 [230]. Exogenous vitamin D intoxication may occur in patients taking too much 1-α-OHD, dihydrotachysterol (DHT), or 1,25(OH)2D itself; the two former compounds undergo 25-hydroxylation in the host hepatocyte. In these instances, the 1,25(OH)2D, not the 25OHD, concentration will be elevated, making the distinction from endogenous vitamin D intoxication from the extra-renal overproduction of 1,25(OH)2D impossible on strictly biochemical grounds. In these situations, a complete knowledge of the medications to which the patient has access is critical for making the correct diagnosis. Examples of such patients would be those receiving relatively large amounts of vitamin D, a vitamin D metabolite, or a vitamin D analog (i.e. patients with hypoparathyroidism, renal failure, and

CHAPTER 79 Extra-renal 1α-Hydroxylase Activity and Human Disease

psoriasis, respectively). Most of the newer vitamin D analogs currently in clinical use [231] will not be measured efficiently in the serum 1,25(OH)2D assay, so awareness of the use of these kinds of topically- and orally-administered agents is of particular importance to the diagnosing clinician.

B. Early Detection and Prevention of Hypercalciuria/Hypercalcemia 1. IDENTIFYING PATIENTS AT RISK

Considering the fact that the means of specifically inhibiting the production of active metabolites of vitamin D or of blocking the response of cells to active vitamin D derivatives is not yet available, the best way to treat vitamin D–mediated abnormalities in calcium balance is to prevent their occurrence. The first step is to identify patients at risk. This encompasses primarily patients with granuloma-forming disease as well as patients with malignant lymphoproliferative disorders, especially B-cell and Hodgkin’s lymphoma. Disordered calcium balance in these groups of patients results from the endogenous and dysregulated overproduction of 1,25(OH)2D by inflammatory cells. Production of the offending vitamin D metabolite is, in turn, directly related to the amount of substrate 25OHD available to the macrophage 1α-hydroxylase (see Fig. 2) as well as to the severity and activity of the underlying disease. In terms of sarcoidosis, for example, patients at risk would be those with: 1) widespread, active disease; 2) a previous history of hypercalciuria or hypercalcemia; 3) a diet enriched in vitamin D and/or calcium; 4) a recent history of sunlight exposure or treatment with vitamin D; and 5) an intercurrent condition, or medicinal treatment of an intercurrent condition, that increases bone resorption or decreases the glomerular filtration rate. 2. SCREENING PATIENTS AT RISK

Since hypercalciuria almost always precedes the development of overt hypercalcemia in this set of disorders, patients at risk should be checked for the presence of occult hypercalciuria. This is best accomplished by a fasting two-hour urine collection for calcium and creatinine. If the calcium:creatinine ratio (gm:gm) is not abnormally high (<0.16), then a 24-hour urine collection for the fractional calcium excretion rate is necessary to establish hypercalciuria; presumably due to increased bone resorption, many patients with vitamin D–mediated hypercalciuria will be hypercalciuric even in the absence of recent food (calcium) ingestion. If screening is to be done only on an annual basis, then the late summer or early autumn

1393

when 25OHD levels are usually at their peak is the best time [76]. 3. PREVENTION OF HYPERCALCEMIA

Prevention of an overt disorder in calcium balance is obviously preferable to dealing with hypercalcemia. This is especially true for patients 1) in whom one episode of hypercalciuria or hypercalcemia has already been documented and 2) are taking supplemental vitamin D and calcium preparations for another medical indication (i.e. osteoporosis). Prevention is best achieved by monitoring the serum calcium and urinary calcium excretion rate on a regular basis. If either is high, then the serum concentration of 25OHD and 1,25(OH)2D level should be obtained; the former should be evaluated to rule out the existence or coexistence of exogenous vitamin D intoxication. For patients determined to be at risk from the appropriate monitoring analyses, measures to prevent worsening hypercalciuria anf frank hypercalcemia should be instituted. These measures should include: 1) the use of UVB-absorbing sunscreens on exposed body parts when they anticipate being out-ofdoors for periods in excess of 20–30 minutes; 2) caution against ingestion of vitamin and food supplements containing ≥400 IU vitamin D; 3) education on the vitamin D and calcium content of foods, vitamin supplements, and medicinal agents like antacids; 4) caution against the regular ingestion of elemental calcium in excess of 1000 mg daily; and 5) education regarding the earliest signs of hypercalciuria (i.e. nocturia).

C. Treatment of Hypercalciuria and Hypercalcemia Because they reside on the same pathophysiological spectrum, the treatment aim of normalization of the urinary calcium excretion rate is the same whether the patient is hypercalciuric or frankly hypercalcemic. There are three general therapeutic goals. First is reduction in the serum concentration of the offending vitamin D metabolite or derivative. In patients with exogenous vitamin D intoxication, this is usually accomplished by cessation or a reduction in the dose of the vitamin D preparation being used by that patient. Remembering that 25OHD has a serum half-life of months, vitamin D–intoxicated patients may require as much as a year off therapy. In patients suffering from endogenous intoxication with 1,25(OH)2D made by inflammatory cells, reduction in the serum 1,25(OH)2D level can be most reliably achieved by treatment with anti-inflammatory doses of glucocorticoid (adult dose of 40 mg prednisone or equivalent per day). At these doses, steroid therapy should have little effect on the

1394 renal 1α-hydroxylase, so there is little concern for inducing hypocalcemia with glucocorticoid administration. In patients with extra-renal production of the hormone, steroid therapy should result in a drop in the serum 1,25(OH)2D concentration within a matter of 3–4 days, followed shortly thereafter by a decrease in the filtered load of calcium and urinary calcium excretion rate provided that the patient’s glomerular filtration rate is maintained. In patients who fail glucocorticoids or in whom glucocorticoids are contraindicated, treatment with the 4-aminoquinoline class of drugs like chloroquine (250 mg twice daily) or hydroxychloroquine (up to 400 mg daily) may be effective [34–36]. A less desirable therapeutic alternative is the cytochrome P450 inhibitor ketoconazole [232]; it will effectively reduce the serum 1,25(OH)2D concentration [233–235], but the therapeutic margin of safety is narrow; doses of the drug that inhibit the macrophage P450 system are very close to those that will also inhibit endogenous glucocorticoid and sex steroid production. The second goal of therapy is to limit the actions of the vitamin D derivative at its target tissues, the gut and bone. A reduction in intestinal calcium absorption is best accomplished by elimination of as much calcium as possible from the diet. Such conservative measures are rarely effective in patients with active, widespread disease, so glucocorticoid administration may also be required to block vitamin D–mediated calcium absorption and bone resorption [236]. Although not well studied, other skeletal antiresorptive agents, like calcitonin and the bisphosphonates, do not appear to be particularly effective in blocking active vitamin D metabolitemediated bone resorption. The third goal of therapy is to enhance urinary calcium excretion to a point where the filtered load of calcium is insufficient to cause either hypercalcemia or hypercalciuria. This can be achieved by maintenance of the vascular volume (glomerular filtration rate) and urinary flow rate, and if needed, by the use of a “loop” diuretic, like furosemide, to inhibit calcium reabsorption from the urine. The effects of successfully reducing the serum 1,25(OH)2D concentration and managing hypercalcemia/hypercalciuria on the patient’s skeleton longterm are not known. There is preliminary evidence that successful treatment of exogenous vitamin D intoxication may result in a transient increase in bone mineral density [237].

References 1. Harrell GT, Fisher S 1939 Blood chemical changes in Boeck’s sarcoid. J Clin Invest 18:687–693.

MARTIN HEWISON AND JOHN S. ADAMS 2. Henneman PH, Dempsey EF, Carrol EJ, Albright F 1956 The causes of hypercalcemia in sarcoid and its treatment with cortisone. J Clin Invest 35:1229–1242. 3. Taylor RL, Lynch HJ Jr, Wysor WG 1963 Seasonal influence of sunlight on the hypercalcemia of sarcoidosis. Am J Med 35:67–89. 4. Dent CE, Flynn FV, Nabarro JDN 1953 Hypercalcemia and impairment of renal function in generalized sarcoidosis. Br Med J 2:808–810. 5. Hendrix JZ 1963 The remission of hypercalcemia and hypercalciuria in systemic sarcoidosis by vitamin D depletion. Clin Res 11:220–225. 6. Bell NH, Gill JR Jr, Barter FC 1964 Abnormal calcium absorption in sarcoidosis: evidence for increased sensitivity to vitamin D. Am J Med 36:500–513. 7. Holick MF, Schnoes THK, DeLuca HF, Suda T, Cousins RJ 1992 Isolation and identification of 1,25-dihydroxycholecalciferol. A metabolite of vitamin D active in intestine. J NIH Res 4:88–96. 8. Bell NH 1985 Vitamin D endocrine system. J Clin Invest 76:1–6. 9. Hughes MF, Baylink DJ, Jones PG, Haussler MR 1976 Radioligand receptor assay for 25-hydroxyvitamin D2/D3 and 1α,25-dihydroxyvitamin D2/D3. J Clin Invest 58:61–70. 10. Clemens TL, Hendy GN, Graham RF 1978 A radioimmunoassay for 1,25-dihydroxycholecalciferol. Clin Sci Mol Med 54:329–332. 11. Gray TK, McAdoo T 1979 Radioimmunoassay for 1,25-dihydroxyvitamin D3. In: Norman AW, Schaefer K, von-Herrath D (eds) Vitamin D Basic Research and Its Clinical Application. Berlin: de Gruyter, pp. 763–767. 12. Bouillon R, DeMoor P, Baggiolini EG, Uskokovic MR 1980 A radioimmunoassay for 1,25-dihydroxycholecalciferol. Clin Chem 26:562–567. 13. Holick MF 1990 The use and interpretation of assays for vitamin D and its metabolites. J Nutr 120:1464–1469. 14. Bell NH, Stern PH, Pantzer E, Sinha TK, DeLuca 1999 Evidence that increased circulating 1,25-dihydroxyvitamin D is the probable cause for abnormal calcium metabolism in sarcoidosis. J Clin Invest 64:218–225. 15. Papapoulos SE, Clemens TL, Fraher LJ, Lewin IG, Sandler LM, O’Riordan JL 1979 1,25-Dihydroxycholecalciferol in the pathogenesis of the hypercalcemia of sarcoid. Lancet 1: 627–630. 16. Stern PH, Olazabal J, Bell NH 1980 Evidence for abnormal regulation of circulating 1,25-dihydroxyvitamin D in patients with sarcoidosis. J Clin Invest 66:852–855. 17. Barbour GL, Coburn JW, Slatopolsky E, Norman AW, Horst RL 1981 Hypercalcemia in an anephric patient with sarcoidosis. N Engl J Med 305:440–443. 18. Adams JS, Sharma OP, Gacad MA, Singer FR 1983 Metabolism of 25-hydroxyvitamin D3 by cultured alveolar macrophages in sarcoidosis. J Clin Invest 72:1856–1860. 19. Adams JS, Singer FR, Gacad MA, Sharma OP, Hayes MJ, Vouros P, Holick MF 1985 Isolation and structural identification of 1,25-dihydroxyvitamin D3 produced by cultured alveolar macrophages in sarcoidosis. J Clin Endocrinol Metab 60: 960–966. 20. Takeyama K, Kitanaka S, Sato T, Kobori M, Yanagisawa J, Kato S 1997 25-Hydroxyvitamin D3 1α-hydroxylase and vitamin D synthesis. Science 277:1827–1830. 21. Fu GK, Lin D, Zhang MY, Bikle DD, Shackleton CH, Miller WL, Portale AA 1997 Cloning of human 25-hydroxyvitamin D-1α-hydroxylase and mutations causing vitamin D–dependent rickets type 1. Mol Endocrinol 11:1961–1970.

CHAPTER 79 Extra-renal 1α-Hydroxylase Activity and Human Disease

22. Smith SJ, Rucka AK, Berry JL, Davies M, Mylchreest S, Paterson CR, Heath DA, Tassabehji M, Read AP, Mee AP, Mawer EB 1999 Novel mutations in the 1α-hydroxylase (P450c1) gene in three families with pseudovitamin D– deficiency rickets resulting in loss of functional enzyme activity in blood-derived macrophages. J Bone Miner Res 14:730–739. 23. Zehnder D, Bland R, Walker EA, Bradwell AR, Howie AJ, Hewison M, Stewart PM Expression of 25-hydroxyvitamin D3–1α-hydroxylase in the human kidney. J Am Soc Nephrol 10:2465–2473. 24. Zehnder D, Bland R, Williams MC, McNinch RW, Howie AJ, Stewart PM, Hewison M 2001 Extra-renal expression of 25-hydroxyvitamin D3-1α-hydroxylase. J Clin Endo Metab 86:888–894. 25. Bell NH 1991 Endocrine complications of sarcoidosis. Endo Metab Clin North Am 20:645–654. 26. Basile JN, Leil Y, Shary J, Bell NH 1993 Increased calcium intake does not suppress circulating 1,25-dihydroxyvitamin D in normocalcemic patients with sarcoidosis. J Clin Invest 91: 1396–1398. 27. Sandler LM, Winearls CG, Fraher LJ, Clemens TL, Smith R, O’Riordan JLH 1984 Studies of the hypercalcemia of sarcoidosis. Quart J Med 53:165–180. 28. Papapoulos SE, Clemens TL, Fraher LJ 1979 Dihydroxycholecalciferol in the pathogenesis of the hypercalcemia of sarcoid. Lancet 1:627–630. 29. Cronin CC, Dinneen SF, O’Mahony MS, Bredin CP, O’Sullivan DJ 1990 Precipitation of hypercalcemia in sarcoidosis by foreign sun holidays: report of four cases. Postgrad Med J 66:307–309. 30. Insogna KL, Dreyer BE, Mitnich M, Ellison AF, Broadus A 1988 Enhanced production of 1,25-dihydroxyvitamin D in sarcoidosis. J Clin Endocrinol Metab 66:72–75. 31. Stern PH, De Olazabal J, Bell NH 1980 Evidence for abnormal regulation of circulating 1α,25-dihydroxyvitamin D in patients with sarcoidosis and normal calcium metabolism. J Clin Invest 66:852–855. 32. Shulman LE, Schoenrich E, Harvey A 1952 The effects of adrenocorticotropic hormone (ACTH) and cortisone on sarcoidosis. Bull John Hopkins Hosp 91:371–415. 33. Anderson J, Dent CE, Harper C, Philpot GR 1954 Effect of cortisone on calcium metabolism in sarcoidosis with hypercalcemia. Lancet 2:720–724. 34. O’Leary TJ, Jones G, Yip A, Lohnes D, Cohanim M, Yendt ER 1986 The effects of chloroquine on serum 1,25-dihydroxyvitamin D and calcium metabolism in sarcoidosis. N Engl J Med 315:727–730. 35. Barre PE, Gascon-Barre M, Meakins JL, Goltzman D 1987 Hydroxychloroquine treatment of hypercalcemia in a patient with sarcoidosis. Am J Med 82:1259–1262. 36. Adams JS, Diz MM, Sharma OP 1989 Effective reduction in the serum 1,25-dihydroxyvitamin D and calcium concentration in sarcoidosis-associated hypercalcemia with short-course chloroquine therapy. Ann Int Med 111:437–438. 37. Singer FR, Adams JS 1986 Abnormal calcium homeostasis in sarcoidosis. N Engl J Med 315:755–756. 38. Meyrier A, Valeyre D, Bouillon R, Paillard F, Battesti JP, Georges R 1985 Resorptive versus absorptive hypercalciuria in sarcoidosis. Quart J Med 54:269–281. 39. Adams JS, Gacad MA, Anders A, Endres DB, Sharma DP 1986 Biochemical indicators of disordered vitamin D and calcium homeostasis in sarcoidosis. Sarcoidosis 3:1–6. 40. Adams JS, Gacad MA 1985 Characterization of 1α-hydroxylation of vitamin D3 sterols by cultured macrophages from patients with sarcoidosis. J Exp Med 161:755–765.

1395

41. Shany S, Adams JS 1993 Subcellular localization of the 25-hydroxyvitamin D3-1-hydroxylase and partial purification from the chick myelomonocytic cell line HD-11. J Bone Min Res 8:269–276. 42. Adams JS, Ren S-Y, Arbelle JE, Horiuchi N, Gray RW, Clemens TL, Shany S 1994 Regulated production and intracrine action of 1,25-dihydroxyvitamin D3 in chick myelomonocytic cell line HD-11. Endocrinology 134:2567–2573. 43. Reichel H, Koeffler HP, Norman AW 1987 Synthesis in vitro of 1,25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3 by interferon-gamma-stimulated normal human bone marrow and alveolar macrophages. J Biol Chem 262:10931–10937. 44. Fraser DR 1980 Regulation of the metabolism of vitamin D. Physiol Rev 60:551–613. 45. Nakamura Y, Eto TA, Taniguchi T, Miyamoto K, Nagatomo J, Shiotsuki H, Sueta H, Higashi S, Okuda KI, Setoguchi T 1997 Purification and characterization of 25-hydroxyvitamin D3 1α-hydroxylase from rat kidney mitochondria. FEBS Lett 419:45–48. 46. Sakaki T, Sawada N, Nonaka Y, Ohyama Y, Inouye K 1999 Metabolic studies using recombinant Escherichia coli cells producing rat mitochondrial CYP24. CYP24 can convert 1α,25-dihydroxyvitamin D3 to calcitroic acid. Eur J Biochem 262:43–48. 47. Reichel H, Koeffler HP, Norman AW 1989 The role of the vitamin D endocrine system in health and disease. N Engl J Med 320:980–991. 48. Taylor R 1994 A new receptor for calcium ions. J NIH Res 6:25–27. 49. Garabedian M, Holick MF, DeLuca HF, Boyle IT 1972 Control of 25-hydroxy-cholecalciferol metabolism by parathyroid glands. Proc Natl Acad Sci 69:1973–1976. 50. Murayama A, Takeyama K, Kitanaka S, Kodera Y, Hosoya T, Kato S 1998 The promoter of the human 25-hydroxyvitamin D3 1alpha-hydroxylase gene confers positive and negative responsiveness to PTH, calcitonin, and 1α,25(OH)2D3. Biochem Biophys Res Commun 249:11–16. 51. Kong XF, Zhu XH, Pei YL, Jackson DM, Holick MF 1999 Molecular cloning, characterization, and promoter analysis of the human 25-hydroxyvitamin D3-1α-hydroxylase gene. Proc Natl Acad Sci USA 96:6988–6993. 52. Henry HL 1992 Vitamin D hydroxylases. J Cell Biochem 49:4–9. 53. Portale AA, Halloran BP, Murphy MM, Morris RC, Jr. 1986 Oral intake of phosphorus can determine the serum concentration of 1,25-dihydroxyvitamin D by determining its production rate in humans. J Clin Invest 77:7–12. 54. Tenenhouse HS, Martel J, Gauthier C, Zhang MY, Portale AA 2001 Renal expression of the sodium/phosphate cotransporter gene, Npt2, is not required for regulation of renal 1 alphahydroxylase by phosphate. Endocrinology 142:1124–1129. 55. Reichel H, Koeffler HP, Barbers R, Norman AW 1987 Regulation of 1,25-dihydroxyvitamin D3 production by cultured alveolar macrophages from normal human donors and patients with pulmonary sarcoidosis. J Clin Endocrinol Metab 65:1201–1209. 56. Hakeda Y, Hiura K, Sato T, Okazaki R, Matsumoto T, Ogata E, Ishitani R, Kumegawa M 1989 Existence of parathyroid hormone-binding sites on murine hemopoietic blast cells. Biochem Biophys Res Commun 163:1481–1486. 57. Yuan JY, Freemont AJ, Mawer EB, Hayes ME 1992 Regulation of 1α,25-dihydroxyvitamin D3 synthesis in macrophages from arthritic joints by phorbol ester, dibutyrylcAMP and calcium ionophore (A23187). FEBS Lett 311: 71–74.

1396 58. Bland R, Walker EA, Hughes SV, Stewart PM, Hewison M 1999 Constitutive expression of 25-hydroxyvitamin D3-1αhydroxylase in a transformed human proximal tubule cell line: evidence for direct regulation of vitamin D metabolism by calcium. Endocrinology 140:2027–2034. 59. Nesbitt T, Drezner MK 1993 Insulin-like growth factor-1 regulation of renal 25-hydroxyvitamin D-1α-hydroxylase activity. Endocrinology 132:133–138. 60. Henry HH 1981 25(OH)D3 metabolism in kidney cell cultures: lack of a direct effect of estradiol. Am J Physiol 240:E119–E124. 61. Adams ND, Garthwaite TL, Gray RW 1979 The interrelationship among prolactin, 1,25-dihydroxyvitamin D, and parathyroid hormone in humans. J Clin Endocrinol Metab 49:628–630. 62. Kumar R, Merimee TJ, Silva P, Epstein FH 1979 The effect of chronic excess or deficiency of growth hormone on plasma 1,25-dihydroxyvitamin D levels in man. In: Norman AW, Schaefer K, von Herrath D (eds.) Vitamin D, Basic Research and Its Clinical Application. Berlin: de Gruyter pp.1005–1009. 63. Brixen K, Nielsen HK, Bouillon R, Flyvbjerg A, Mosekilde L 1992 Effects of short-term growth hormone treatment on PTH, calcitriol, thyroid hormones, insulin, and glucagon. Acta Endocrinol 127:331–336. 64. Omdahl JL, Bobrovnikova EA, Choe S, Dwivedi PP, May BK 2001 Overview of regulatory cytochrome P450 enzymes of the vitamin D pathway. Steroids 66:381–389. 65. Ohyama Y, Noshiro M, Okuda K 1991 Cloning and expression of cDNA encoding 25-hydroxyvitamin D3 24-hydroxylase. FEBS Letter 278:195–198. 66. Chen K, Goto H, DeLuca HF 1992 Isolation and expression of human 1,25-dihydroxyvitamin D3 24-hydroxylase cDNA. J Bone Mineral Res 7:S148. 67. Chen KS, Prahl JM, DeLuca HF 1993 Isolation and expression of human 1,25-dihydroxyvitamin D3-24-hydroxylase. Proc Natl Acad Sci 90:4543–4547. 68. Ismail R, Elaroussi MA, DeLuca HF 1993 Regulation of chicken kidney vitamin D3 24-hydroxylase mRNA by 1,25dihydoxyvitamin D3 and parathyroid hormone. J Bone Min Res 8:S208. 69. Uchida M, Shinki T, Ohyama Y, Noshiro M, Okda K, Suda T 1993 Protein kinase C up-regulates 1α,25-dihydroxyvitamin D3 induced expression of the 24-hydroxylase gene. J Bone Min Res 8:S171. 70. Chen ML, Boltz MA, Armbrecht HJ 1993 Effects of 1,25dihydroxyvitamin D3 and phorbol ester on 25-hydroxyvitamin D3-24-hydroxylase cytochrome P-450 messenger ribonucleic acid levels in primary cultures of rat renal cells. Endocrinology 132:1782–1788. 71. Zierold C, Mings JA, DeLuca HF 2003 Regulation of 25-hydroxyvitamin D3-24-hydroxylase mRNA by 1,25-dihydroxyvitamin D3 and parathyroid hormone. J Cell Biochem 88:234–237. 72. Henry HL 1979 Regulation of the hydroxylation of 25-hydroxyvitamin D3 in vivo and in primary cultures of chick kidney cells. J Biol Chem 254:2722–2729. 73. Inouye K, Sakaki T 2001 Enzymatic studies on the key enzymes of vitamin D metabolism; 1α-hydroxylase (CYP27B1) and 24-hydroxylase (CYP24). Biotechnol Annu Rev 7:179–194. 74. Xie Z, Munson SJ, Huang N, Portale AA, Miller WL, Bikle DD 2002 The mechanism of 1,25-dihydroxyvitamin D3 autoregulation in keratinocytes. J Biol Chem 277:36987–36990. 75. Adams JS, Gacad MA, Singer FR, Sharma OP 1986 Production of 1,25-dihydroxyvitamin D3 by pulmonary alveolar macrophages from patients with sarcoidosis. NY Acad Sci 465:587–594. 76. Adams JS 1992 Hypercalcemia and hypercalciuria. Sem Resp Med 13:402–410.

MARTIN HEWISON AND JOHN S. ADAMS

77. Adams JS, Modlin RL, Diz MM, Barnes PF 1989 Potentiation of the macrophage 25-hydroxyvitamin D-1α-hydroxylation reaction by human tuberculous pleural effusion fluid. J Clin Endocrinol Metab 69:457–460. 78. Hunninghake GW 1986 Role of alveolar macrophage and lung T-cell derived mediators in pulmonary sarcoidosis. Ann NY Acad Sci 465:82–90. 79. Pryke AM, Duggan C, White CP, Posen S, Mason RS 1990 Tumor necrosis factor-alpha induces vitamin D-1-hydroxylase activity in normal human alveolar macrophages. J Cell Physiol 142:652–656. 80. Platanias LC, Fish EN 1999 Signaling pathways activated by interferons. Exp Hematol 27:1583–1592. 81. Imada K, Leonard WJ 2000 The Jak-STAT pathway. Mol Immunol 37:1–11. 82. Ishihara K, Hirano T 2002 Molecular basis of the cell specificity of cytokine action. Biochim Biophys Acta 1592:281–296. 83. Xaus J, Comalada M, Valledor AF, Cardo M, Herrero C, Soler C, Lloberas J, Celada A 2001 Molecular mechanisms involved in macrophage survival, proliferation, activation, or apoptosis. Immunobiology 204:543–550. 84. Celada A, Schreiber RD 1996 Role of protein kinase C and intracellular calcium mobilization in the induction of macrophage tumoricidal activity by interferon-gamma. J Immunol 137:2373. 85. Wightman PD, Humes JL, Davies P, Bonney RJ 1981 Identification of two phospholipase A2 activities in resident mouse peritoneal macrophages. Biochem J 195:427. 86. Wightman PD, Dahlgren M, Bonney RS 1982 Protein kinase activation of phospholipase A2 in sonicates of mouse peritoneal macrophages. J Biol Chem 257:6650. 87. Adams JS, Gacad MA, Diz MM, Nadler JL 1990 A role for endogenous arachidonate metabolites in the regulated expression of the 25-hydroxyvitamin D-1-hydroxylation reaction in cultured alveolar macrophages from patients with sarcoidosis. J Clin Endocrinol Metab 70:595–600. 88. Sabroe I, Jones EC, Usher LR, Whyte MK, Dower SK 2002 Toll-like receptor (TLR)2 and TLR4 in human peripheral blood granulocytes: a critical role for monocytes in leukocyte lipopolysaccharide responses. J Immunol 168:4701–4710. 89. Barton GM, Medzhitov R 2002 Toll-like receptors and their ligands. Curr Top Microbiol Immunol 270:81–92. 90. Bland R, Zehnder D, Hughes SV, Ronco PM, Stewart PM, Hewison M 2001 Regulation of vitamin D-1α-hydroxylase in a human cortical collecting duct cell line. Kidney International 60:1277–1286. 91. Adams JS, Ren SY, Arbelle JE, Clemens TL, Shany S 1994 A role for nitric oxide in the regulated expression of the 25-hydroxyvitamin D-1-hydroxylation reaction in the chick myelomonocytic cell line HD-11. Endocrinology 134:499–502. 92. Adams JS, Ren S-Y, Arbelle J, Shany S, Gacad MA 1995 Coordinate regulation of nitric oxide and 1,25-dihydroxyvitamin D production in the avian myelomonocytic cell line HD-11. Endocrinology 136:2262–2269. 93. Adams JS, Ren S-Y 1996 Autoregulation of 1,25-dihydroxyvitamin D synthesis in macrophage mitochondria by nitric oxide. Endocrinology 137:4514–4517. 94. Marletta MA 1994 Nitric oxide synthase: aspects concerning structure and catalysis. Cell 78:927–930. 95. Nathan C, Xie Q-W 1994 Nitric oxide synthases: Roles, tolls, and controls. Cell 78:915–918. 96. Salkowski CA, Detore G, McNally R, van Rooijen N, Vogel SN 1997 Regulation of inducible nitric oxide synthase messenger RNA expression and nitric oxide production by lipopolysaccharide in vivo: the roles of macrophages, endogenous IFN-gamma, and TNF receptor-1-mediated signaling. J Immunol 158:905–912.

CHAPTER 79 Extra-renal 1α-Hydroxylase Activity and Human Disease

97. Alley EW, Murphy WJ, Russell SW 1995 A classical enhancer element responsive to both lipopolysaccharide and interferon-gamma augments induction of the iNOS gene in mouse macrophages. Gene 158:247–251. 98. White KA, Marletta MA 1992 Nitric oxide synthase is a cytochrome P-450 type hemoprotein. Biochemistry 31: 6627–6631. 99. Khatsenko OG, Gross SS, Rifkind AB, Vane JR 1993 Nitric oxide is a mediator of the decrease in cytochrome P450dependent metabolism caused by immunostimulants. Proc Natl Acad Sci 90:11147–11151. 100. Stadler J, Trockfeld J, Schmalix WA, Brill T, Siewert JR, Greim H, Doehmer J 1994 Inhibition of cytochromes P450-1A by nitric oxide. Proc Natl Acad Sci 91:3559–3563. 101. Van Voorhis BJ, Dunn MS, Snyder GD, Weiner CP 1994 Nitric oxide: An autocine regulator of human granulosaluteal cell steroidogenesis. Endocrinology 135:1799–1806. 102. Hartl FU 1996 Molecular chaperones in cellular protein folding. Nature 381:571–580. 103. Rutherford SL, Zuker CS 1994 Protein folding and the regulation of signaling pathways. Cell 79:1129–1132. 104. Minowada G, Weich WJ 1995 Clinical implications of the stress response. J Clin Invest 95:3–12. 105. Moseley P 2000 Stress proteins and the immune response. Immunopharmacol 48:299–302. 106. More S, Breloer M, Fleischer B, von Bonin A 1999 Activation of cytotoxic T-cells in vitro by recombinant gp96 fusion proteins irrespective of the “fused” antigenic peptide sequence. Immunol Lett 69:275–282. 107. Wallin RP, Lundqvist A, More SH, von Bonin A, Kiessling R, Ljunggren HG 2002 Heat-shock proteins as activators of the innate immune system. Trends Immunol 23:130–135. 108. Vabulas RM, Ahmad-Nejad P, Ghose S, Kirschning CJ, Issels RD, Wagner H 2002 HSP70 as endogenous stimulus of the Toll/interleukin-1 receptor signal pathway. J Biol Chem 277:15107–15112. 109. Asea A, Rehli M, Kabingu E, Boch JA, Bare O, Auron PE, Stevenson MA, Calderwood SK 2002 Novel signal transduction pathway utilized by extracellular HSP70: role of toll-like receptor (TLR) 2 and TLR4. J Biol Chem 277:15028–15034. 110. Berthold J, Bauer MF, Schneider H-C, Klaus C, Dietmeier K, Neupert W, Brunner M 1995 The MIM complex mediates preprotein translocation across the mitochondrial inner membrane and couples it to the mt-Hsp70/ATP driving system. Cell 81:1085–1093. 111. Brodsky JL 1996 Post-translational protein translocation: not all hsc70s are created equal. Trends Biochem Sci 21:122–126. 112. Voos W, Martin H, Krimmer T, Pfanner N 1999 Mechanisms of protein translocation into mitochondria. Biochim Biophys Acta 1422:235–254. 113. Polla BS, Healy AM, Wojno WC, Krane SM 1987 Hormone 1α,25-dihydroxyvitamin D3 modulates heat shock response in monocytes. Am J Physiol 252:C640–C649. 114. Gacad MA, Adams JS 1998 Proteins in the heat shock-70 family specifically bind 25-hydroxyvitamin D3 and 17-betaestradiol. J Clin Endocrinol Metab 83:1264–1267. 115. Gacad MA, Chen H, Arbelle JE, LeBon T, Adams JS 1997 Functional characterization and purification of an intracellular vitamin D–binding protein in vitamin D–resistant new world primate cells. Amino acid sequence homology with proteins in the hsp-70 family. J Biol Chem 272:8433–8440. 116. Wu S, Ren S, Chen H, Chun RF, Gacad MA, Adams JS 2000 Intracellular vitamin D–binding proteins: novel facilitators of vitamin D–directed transactivation. Mol Endocrinol 14: 1387–1397.

1397

117. Adams JS, Chen H, Chun RF, Nguyen L, Wu S, Ren SY, Barsony J, Gacad MA 2003 Novel regulators of vitamin D action and metabolism: Lessons learned at the Los Angeles zoo. J Cell Biochem 88:308–314. 118. Nykjaer A, Dragun D, Walther D, Vorum H, Jacobsen C, Herz J, Melsen F, Christensen EI, Willnow TE 1999 An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) vitamin D3. Cell 96:507–515. 119. Willnow TE, Nykjaer A 2002 Pathways for kidney-specific uptake of the steroid hormone 25-hydroxyvitamin D3. Curr Opin Lipidol 13:255–260. 120. Provvedini DM, Tsoukas CD, Deftos LJ, Manolagas SC 1983 1,25-dihydroxyvitamin D3 receptors in human leukocytes. Science 221:1181–1182. 121. Bhalla AK, Amento EP, Clemens TL, Holick MF, Krane SM 1983 Specific high-affinity receptors for 1,25-dihydroxyvitamin D3 in human peripheral blood mononuclear cells. J Clin Endocrinol Metab 57:1308–1310. 122. Rigby W 1988 The immunobiology of vitamin D. Immunol Today 9:54–58. 123. Hewison M 1992 Vitamin D and the immune system. J Endocrinology 132:173–175. 124. Fagan DL, Prehn JL, Jordan SC, Adams JS 1991 The human myelomonocytic cell line U937 as a model for studying alterations in monokine gene expression by 1,25-dihydroxyvitamin D. Mol Endocrinol 5:179–186. 125. Prehn JL, Fagan DL, Jordan SC, Adams JS 1992 Potentiation of lipopolysaccharide-induced tumor necrosis factor-alpha expression by 1,25-dihydroxyvitamin D3. Blood 80:2811–2816. 126. Bar-Shavit Z, Teitelbaum SL, Reitsma P, Hall A, Pegg LE, Trial J, Kahn AJ 1983 Induction of monocytic differentiation and bone resorption by 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 80:5907–5910. 127. Bhalla AK, Amento EP, Krane SM 1986 Differential effects of 1,25-dihydroxyvitamin D3 on human lymphocytes and monocyte/macrophages: inhibition of interleukin-2 and augmentation of interleukin-1 production. Cell Immunol 98: 311–322. 128. Ucla C, Roux-Lombard P, Dayer J-M, Mach B 1990 IFNγ drastically modifies the regulation of IL-1 genes by endotoxin in U937 cells. J Clin Invest 85:185–191. 129. Morel PA, Manolagas SC, Provvedini DM, Wegmann DR, Chiller JM 1986 Interferon-gamma-induced IA expression in WEHI-3 cells is enhanced by the presence of 1,25-dihydroxyvitamin D3. J Immunol 136:2181–2186. 130. Rook GAW, Steele J, Fraher L, Barker S, Karmali R, O’Riordan J, Stanford J 1986 Vitamin D3, gamma-interferon, and control of proliferation of mycobacterium tuberculosis by human monocytes. Immunology 57:159–163. 131. Cohen MS, Mesler DE, Snipes RG, Gray TK 1986 1,25dihydroxyvitamin D3 activates secretion of hydrogen peroxide by human monocytes. J Immunol 136:1049–1053. 132. Rigby WF, Waugh M, Graziano RF 1990 Regulation of human monocyte HLA-DR and CD4 antigen expression and antigen presentation by 1,25-dihydroxyvitamin D3. Blood 76:189–197. 133. Penna G, Adorini L 2000 1α,25-dihydroxyvitamin D3 inhibits differentiation, maturation, activation, and survival of dendritic cells leading to impaired alloreactive T-cell activation. J Immunol 164:2405–2411. 134. Griffin MD, Lutz W, Phan VA, Bachman LA, McKean DJ, Kumar R 2001 Dendritic cell modulation by 1α,25-dihydroxyvitamin D3 and its analogs: a vitamin D receptordependent pathway that promotes a persistent state of immaturity in vitro and in vivo. Proc Natl Acad Sci USA 98:6800–6805.

1398 135. Piemonti L, Monti P, Sironi M, Fraticelli P, Leone BE, Dal Cin E, Allavena P, Di Carlo V 2000 Vitamin D3 affects differentiation, maturation, and function of human monocytederived dendritic cells. J Immunol 164:4443–4451. 136. Berer A, Stockl J, Majdic O, Wagner T, Kollars M, Lechner K, Geissler K, Oehler L 2000 1,25-Dihydroxyvitamin D3 inhibits dendritic cell differentiation and maturation in vitro. Exp Hematol 28:575–583. 137. Hewison M, Freeman L, Hughes SV, Evans KN, Bland R, Eliopoulos AG, Kilby MD, Moss PA, Chakraverty R 2003 Differential regulation of vitamin D receptor and its ligand in human monocyte-derived dendritic cells. J Immunol 170: 5382–5890. 138. Fritsche J, Mondal K, Ehrnsperger A, Andreesen R, Kreutz M 2003 Regulation of 25-hydroxyvitamin D3-1α-hydroxylase and production of 1α,25-dihydroxyvitamin D3 by human dendritic cells. Blood. 139. Deluca HF, Cantorna MT 2001 Vitamin D: its role and uses in immunology. FASEB J 15:2579–2585. 140. Lemire JM 1992 Immunomodulatory role of 1,25-dihydroxyvitamin D3. J Cell Biochem 49:26–31. 141. Lemire JM, Adams JS, Kermani-Arab V, Bakke AC, Sakai R, Jordan SC 1985 1,25-dihydroxyvitamin D suppresses human T helper-inducer lymphocyte activity in vitro. J Immunol 134:219–224. 142. Tsoukas CD, Provvedini DM, Manolagas SC 1984 1,25dihydroxyvitamin D3, a novel immunoregulatory hormone. Science 224:1438–1440. 143. Nun JD, Katz DR, Barker S, Fraher LJ, Hewison M, Hendy GN, O’Riordan JL 1986 Regulation of human tonsillar T-cell proliferation by the active metabolite of vitamin D3. Immunol 59:479–484. 144. Rigby WF, Noelle RJ, Krause K, Fanger MW 1985 The effects of 1,25-dihydroxyvitamin D3 on human T lymphocyte activation and proliferation. J Immunol 135:2279–2286. 145. Rigby WFC, Denome S, Fanger MW 1987 Regulation of lymphokine production and human T lymphocyte activation by 1,25-(OH)2D3. J Clin Invest 79:1659–1664. 146. Lemire JM, Adams JS, Sakai R, Jordan SC 1984 1,25dihydroxyvitamin D suppresses proliferation and immunoglobulin production by normal human peripheral blood mononuclear cells. J Clin Invest 74:857–861. 147. Lemire JM, Adams JS 1992 1,25-dihydroxyvitamin D inhibits delayed-type hypersensitivity mediated by T-cell clones inducing experimental auto-immune encephalomyelitis. J Bone Min Res 7:171–178. 148. Boonstra A, Barrat FJ, Crain C, Heath VL, Savelkoul HF, O’Garra A 2001 1α,25-dihydroxyvitamin D3 has a direct effect on naive CD4(+) T cells to enhance the development of Th2 cells. J Immunol 167:4974–4980. 149. Lemire JM, Archer DC, Beck L, Spiegelberg HL 1995 Immunosuppressive actions of 1,25-dihydroxyvitamin D3: preferential inhibition of Th1 functions. J Nutr 125(6 Suppl): 1704S–1708S. 150. Rook GA, Hernandez-Pando R, Lightman SL 1994 Hormones, peripherally activated prohormones, and regulation of the Th1/Th2 balance. Immunol Today 15:301–303. 151. Barrat FJ, Cua DJ, Boonstra A, Richards DF, Crain C, Savelkoul HF, de Waal-Malefyt R, Coffman RL, Hawrylowicz CM, O’Garra A 2002 In vitro generation of interleukin 10producing regulatory CD4(+) T-cells is induced by immunosuppressive drugs and inhibited by T helper type 1 (Th1)- and Th2-inducing cytokines. J Exp Med 195:603–616. 152. Gregori S, Casorati M, Amuchastegui S, Smiroldo S, Davalli AM, Adorini L 2001 Regulatory T-cells induced by

MARTIN HEWISON AND JOHN S. ADAMS

153. 154. 155. 156. 157. 158.

159. 160. 161.

162.

163.

164.

165.

166. 167.

168.

1α,25-dihydroxyvitamin D3 and mycophenolate mofetil treatment mediate transplantation tolerance. J Immunol 167: 1945–1953. Adorini L 2003 Tolerogenic dendritic cells induced by vitamin D receptor ligands enhance regulatory T-cells inhibiting autoimmune diabetes. Ann N Y Acad Sci 987:258–252. Aslam SM, Garlich JD, Qureshi MA 1998 Vitamin D deficiency alters the immune responses of broiler chicks. Poult Sci 77:842–849. Yang S, Smith C, Prahl JM, Luo X, DeLuca HF 1993 Vitamin D deficiency suppresses cell-mediated immunity in vivo. Arch Biochem Biophys 303:98–106. McMurray DN, Bartow RA, Mintzer CL, HernandezFrontera E 1990 Micronutrient status and immune function in tuberculosis. Ann N Y Acad Sci 587:59–69. Wientroub S, Winter CC, Wahl SM, Wahl LM 1989 Effect of vitamin D deficiency on macrophage and lymphocyte function in the rat. Calcif Tissue Int 44:125–130. Mariani E, Ravaglia G, Forti P, Meneghetti A, Tarozzi A, Maioli F, Boschi F, Pratelli L, Pizzoferrato A, Piras F, Facchini A 1999 Vitamin D, thyroid hormones, and muscle mass influence natural killer (NK) innate immunity in healthy nonagenarians and centenarians. Clin Exp Immunol 116:19–27. Yener E, Coker C, Cura A, Keskinoglu A, Mir S 1995 Lymphocyte subpopulations in children with vitamin D– deficient rickets. Acta Paediatr Jpn 37:500–502. The EURODIAB Substudy 2 Study Group 1999 Vitamin D supplement in early childhood and risk for Type I (insulindependent) diabetes mellitus. Diabetologia 42:51–54. Jeevan A, Kripke ML 1989 Effect of a single exposure to ultraviolet radiation on Mycobacterium bovis bacillus Calmette-Guerin infection in mice. J Immunol 143: 2837–2843. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia, and growth retardation after weaning. Nat Genet 16:391–396. Panda DK, Miao D, Tremblay ML, Sirois J, Farookhi R, Hendy GN, Goltzman D 2001 Targeted ablation of the 25-hydroxyvitamin D 1α-hydroxylase enzyme: evidence for skeletal, reproductive, and immune dysfunction. Proc Natl Acad Sci U S A 98:7498–7503. Mathieu C, Van Etten E, Gysemans C, Decallonne B, Kato S, Laureys J, Depovere J, Valckx D, Verstuyf A, Bouillon R 2001 In vitro and in vivo analysis of the immune system of vitamin D receptor knockout mice. J Bone Miner Res 16:2057–2065. O’Kelly J, Hisatake J, Hisatake Y, Bishop J, Norman A, Koeffler HP 2002 Normal myelopoiesis but abnormal T lymphocyte responses in vitamin D receptor knockout mice. J Clin Invest 109:1091–1099. Barnes PF, Modlin RL, Bikle DD, Adams JS 1989 Transpleural gradient of 1,25-dihydroxyvitamin D in tuberculous pleuritis. J Clin Invest 83:1527–1532. Adams JS, Modlin RL, Diz MM, Barnes PF 1989 Potentiation of the macrophage 25-hydroxyvitamin D-1-hydroxylation reaction by human tuberculous pleural effusion fluid. J Clin Endocrinol Metab 69:457–460. Rook GAW, Taverne J, Leveton C, Steele J 1987 The role of gamma-interferon, vitamin D3 metabolites, and tumor necrosis factor in the pathogenesis of tuberculosis. Immunology 62:229–234.

CHAPTER 79 Extra-renal 1α-Hydroxylase Activity and Human Disease

169. Rook GAW 1988 The role of vitamin D in tuberculosis. Am Rev Respir Dis 138:768–770. 170. Mawer EB, Hayes ME, Still PE, Davies M, Lumb GA, Palit J, Holt PJL 1991 Evidence for nonrenal synthesis of 1,25dihydroxyvitamin D in patients with inflammatory arthritis. J Bone Min Res 6:733–739. 171. Hayes ME, Bayley D, Still P, Palit J, Denton J, Freemont AJ, Cooper RG, Mawer EB 1992 Differential metabolism of 25hydroxyvitamin D3 by cultured synovial fluid macrophages and fibroblast-like cells from patients with arthritis. Ann Rheum Dis 51:220–226. 172. Hayes ME, O’Donoghue DJ, Ballardie FW, Mawer EB 1987 Peritonitis induces the synthesis of 1α,25-dihydroxyvitamin D3 in macrophages from CAPD patients. FEBS 220:307–310. 173. Shany S, Rapoport J, Zuili I, Gavriel A, Lavi N, Chaimovitz C 1991 Metabolism of 25-OH-vitamin D3 by peritoneal macrophages from CAPD patients. Kid Int 39:1005–1011. 174. Barnard J, Newman LS 2001 Sarcoidosis: immunology, rheumatic involvement, and therapeutics. Curr Opin Rheumatol 13:84–91. 175. Sharma OP 1996 Vitamin D, calcium, and sarcoidosis. Chest 109:535–539. 176. Studdy PR, Bird R, Neville E, James DG 1980 Biochemical findings in sarcoidosis. J Clin Pathol 33:528–533. 177. James DG, Neville E, Siltzbach LE, et al. 1976 A worldwide review of sarcoidosis. Ann N Y Acad Sci 278:321–334. 178. James DG, Siltzbach LE, Sharma OP, Carstairs LS 1969 A tale of two cities: a comparison of sarcoidosis in London and New York. Arch Intern Med 123:187–191. 179. Rizzato G, Montemurro L, Fraioli P 1992 Bone mineral content in sarcoidosis. Semin Resp Med 13:411–423. 180. Vergnon GM, Chappard D, Mounier D, et al. 1988 Phosphocalcic metabolism, bone quantitative histomorphometry, and clinical activity in 10 cases of sarcoidosis. In: Grassi C, Rizzato G, Pozzi E (eds.) Sarcoidosis and other granulomatous disorders. Elsevier: Amsterdam pp. 499–502. 181. Fallon MD, Perry HM III, Teitelbaum SL 1981 Skeletal sarcoidosis with osteopenia. Metab Bone Dis Res 3:171–174. 182. Need AG, Phillips PJ 1980 Hypercalcemia associated with tuberculosis. Br Med J 280:831. 183. Felsenfeld AJ, Drezner MK, Llach F 1945 Hypercalcemia and elevated calcitriol in a maintenance dialysis patient with tuberculosis. Arch Intern Med 146:1941–1945. 184. Gkonos PJ, London R, Hendler ED 1984 Hypercalcemia and elevated 1,25-dihydroxyvitamin D levels in a patient with end-stage renal disease and active tuberculosis. N Engl J Med 311:1683–1685. 185. Epstein S, Stern PH, Bell NH, Dowdeswell I, Turner RT Evidence for abnormal regulation of circulating 1α,25-dihydroxyvitamin D in patients with pulmonary tuberculosis and normal calcium metabolism. Calcif Tissue Int 36:541–544. 186. Bell NH, Shary J, Shaw S, Turner RT 1985 Hypercalcemia associated with increased circulating 1,25-dihydroxyvitamin D in a patient with pulmonary tuberculosis. Calcif Tissue Int 37:588–591. 187. Cadranel J, Garabedian M, Milleron B, Guillozo H, Akoun G, Hance AJ 1990 1,25(OH)2D3 production by T lymphocytes and alveolar macrophages recovered by lavage from normocalcemic patients with tuberculosis. J Clin Invest 85: 1588–1593. 188. Cadranel JL, Garabedian M, Milleron B, Guillozzo H, Valeyre D, Paillard F, Akoun G, Hance AJ 1994 Vitamin D metabolism by alveolar immune cells in tuberculosis: correlation with calcium metabolism and clinical manifestations. Eur Res J 7:1103–1110.

1399

189. Isaacs RD, Nicholson GI, Holdaway IM. Miliary tuberculosis with hypercalcemia and raised vitamin D concentrations. Thorax 42:555–556. 190. Shai F, Baker RK, Addrizzo JR, Wallach S 1972 Hypercalcemia in mycobacterial infection. J Clin Endocrinol Metab 34:251–256. 191. Braman SS, Goldman AL, Schwarz MI 1973 Steriodresponsive hypercalcemia in disseminated bone tuberculosis. Arch Intern Med 90:327–328. 192. Hoffman VH, Korzeniowski OM 1986 Leprosy, hypercalcemia, and elevated serum calcitriol levels. Ann Intern Med 105:890–891. 193. Ryzen E, Rea TH, Singer FR 1988 Hypercalcemia and abnormal 1,25-dihydroxyvitamin D concentrations in leprosy. Am J Med 84:325–329. 194. Kantarijian HM, Saad MF, Estey EH, Sellin RV, Samaan NA 1983 Hypercalcemia in disseminated candidiasis. Am J Med 74:72l–724. 195. Walker JV, Baran D, Yakub YN, Freeman RB 1977 Histoplasmosis with hypercalcemia, renal failure, and papillary necrosis. Confusion with sarcoidosis. JAMA 237: 1350–1352. 196. Parker MS, Dokoh S, Woolfenden JM, Buchsbaum HW 1984 Hypercalcemia in coccidioidomycosis. Am J Med 76: 341–343. 197. Ahmed B, Jaspan JB 1993 Case report: hypercalcemia in a patient with AIDS and Pneumocystis carinii pneumonia. Am J Med Sci 306:313–316. 198. Kozeny G, Barbato A, Bansal VK, Vertuno LL, Hano JE 1984 Hypercalcemia associated with silicone-induced granulomas. N Engl J Med 311:1103–1105. 199. Jurney TH 1984 Hypercalcemia in a patient with eosinophilic granuloma. Am J Med 76:527–528. 200. Edelson GW, Talpos GB, Bone HG III 1993 Hypercalcemia associated with Wegener’s granulomatosis and hyperparathyroidism: etiology and management. Am J Nephrol 13: 275–277. 201. Al-Ali H, Yabis AA, Issa E, Salem Z, Tawil A, Khoury N, Fuleihan Gel-H 2002 Hypercalcemia in Langerhans’ cell granulomatosis with elevated 1,25 dihydroxyvitamin D (calcitriol) level. Bone 30:331–334. 202. Stoeckle JD, Hardy HL, Weber AL 1969 Chronic beryllium disease. Long-term follow-up of 60 cases and selective review of the literature. Am J Med 46:545–561. 203. Cook JS, Stone MS, Hansen JR 1992 Hypercalcemia in association with subcutaneous fat necrosis of the newborn: studies of calcium-regulating hormones. Ped 90:93–96. 204. Bosch X 1998 Hypercalcemia due to endogenous overproduction of 1,25-dihydroxyvitamin D in Crohn’s disease. Gastroenterol 114:1061–1065. 205. Andres PG, Friedman LS 1999 Epidemiology and the natural course of inflammatory bowel disease. Gastroenterol Clin North Am 28:255–281. 206. Tangpricha V, Flanagan JN, Whitlatch LW, Tseng CC, Chen TC, Holt PR, Lipkin MS, Holick MF 2001 25-hydroxyvitamin D-1α-hydroxylase in normal and malignant colon tissue. Lancet 357:1673–1674. 207. Bareis P, Bises G, Bischof MG, Cross HS, Peterlik M 2001 25-hydroxy-vitamin D metabolism in human colon cancer cells during tumor progression. Biochem Biophys Res Commun 285:1012–1017. 208. Ogunkolade BW, Boucher BJ, Fairclough PD, Hitman GA, Dorudi S, Jenkins PJ, Bustin SA 2002 Expression of 25-hydroxyvitamin D-1α-hydroxylase mRNA in individuals with colorectal cancer. Lancet 359:1831–1832.

1400 209. Zaloga GP, Eil C, Medbery CA 1985 Humoral hypercalcemia in Hodgkin’s disease. Arch Intern Med 145:155–157. 210. Rosenthal N, Insogna KL, Godsall JW 1985 Elevations in circulating 1,25-dihydroxyvitamin D3 in three patients with lymphoma-associated hypercalcemia. J Clin Endocrinol Metab 60:29–33. 211. Davies M, Mawer EB, Hayes ME, Lumb GA 1985 Abnormal vitamin D metabolism in Hodgkin’s lymphoma. Lancet 1: 1186–1188. 212. Adams JS, Fernandez M, Endres DB, Gill PS, Rasheed S, Singer FR 1979 Hypercalcemia, hypercalciuria, and elevated serum 1,25-dihydroxyvitamin D concentrations in patients with AIDS- and non-AIDS-associated lymphoma. Blood 73:235–239. 213. Seymour JF, Gagel RF 1993 Calcitriol: The major humoral mediator of hypercalcemia in Hodgkin’s lymphomas. Blood 82:1383–1394. 214. Seymour JF, Gagel RF, Hagemeister FB, Dimopoulos MA, Cabanillas F 1994 Calcitriol production in hypercalcemic and normocalcemic patients with non-Hodgkin’s lymphoma. Ann Intern Med 121:633–640. 215. Davies M, Hayes ME, Liu Yin JA, Berry JL, Mawer EB 1994 Abnormal synthesis of 1,25-dihydroxyvitamin D in patients with malignant lymphoma. J Clin Endocrinol Metab 78: 1202–1207. 216. Hewison M, Kantorovich V, Liker HR, Van Herle AJ, Cohan P, Zehnder D, Adams JS 2003 Vitamin D–mediated hypercalcemia in lymphoma: evidence for hormone production by tumor-adjacent macrophages. J Bone Miner Res 18:579–582. 217. Grote TH, Hainsworth JD 1987 Hypercalcemia and elevated serum calcitriol in a patient with seminoma. Arch Intern Med 147:2212–2213. 218. Bikle DD, Pillai S, Gee E, Hincenbergs M 1991 Tumor necrosis factor-α regulation of 1,25-dihydroxyvitamin D production by human keratinocytes. Endocrinology 129:33–38. 219. Bikle DD, Halloran BP, Riviere JE 1994 Production of 1,25 dihydroxyvitamin D3 by perfused pig skin. J Invest Dermatol 102:796–798. 220. Schwartz GG, Whitlatch LW, Chen TC, Lokeshwar BL, Holick MF 1998 Human prostate cells synthesize 1,25dihydroxyvitamin D3 from 25-hydroxyvitamin D3. Cancer Epidemiol Biomarkers Prev 7:391–395. 221. Zehnder D, Evans KN, Kilby MD, Bulmer JN, Innes BA, Stewart PM, Hewison M 2002 The ontogeny of 25-hydroxyvitamin D3 1α-hydroxylase expression in human placenta and decidua. Am J Pathol 161:105–114. 222. Diaz L, Sanchez I, Avila E, Halhali A, Vilchis F, Larrea F 2000 Identification of a 25-hydroxyvitamin D3 1α-hydroxylase gene transcription product in cultures of human syncytiotrophoblast cells. J Clin Endocrinol Metab 85:2543–2549. 223. Weisman Y, Harell A, Edelstein S, David M, Spirer Z, Golander A 1979 1α,25-dihydroxyvitamin D3 and 24,25dihydroxyvitamin D3 in vitro synthesis by human decidua and placenta. Nature 281:317–319.

MARTIN HEWISON AND JOHN S. ADAMS

224. Segersten U, Correa P, Hewison M, Hellman P, Dralle H, Carling T, Akerstrom G, Westin G 2002 25-hydroxyvitamin D3–1α-hydroxylase expression in normal and pathological parathyroid glands. J Clin Endocrinol Metab 87:2967–2972. 225. Zehnder D, Bland R, Chana RS, Wheeler DC, Howie AJ, Williams MC, Stewart PM, Hewison M 2002 Synthesis of 1,25-dihydroxyvitamin D3 by human endothelial cells is regulated by inflammatory cytokines: a novel autocrine determinant of vascular cell adhesion. J Am Soc Nephrol 13:621–629. 226. Merke J, Milde P, Lewicka S, Hugel U, Klaus G, Mangelsdorf DJ, Haussler MR, Rauterberg EW, Ritz E 1989 Identification and regulation of 1,25-dihydroxyvitamin D3 receptor activity and biosynthesis of 1,25-dihydroxyvitamin D3. Studies in cultured bovine aortic endothelial cells and human dermal capillaries. J Clin Invest 83:1903–1915. 227. Hsu JY, Feldman D, McNeal JE, Peehl DM 2001 Reduced 1alpha-hydroxylase activity in human prostate cancer cells correlates with decreased susceptibility to 25-hydroxyvitamin D3–induced growth inhibition. Cancer Res 61:2852–2856. 228. Correa P, Segersten U, Hellman P, Akerstrom G, Westin G 2002 Increased 25-hydroxyvitamin D3 1α-hydroxylase and reduced 25-hydroxyvitamin D3 24-hydroxylase expression in parathyroid tumors—new prospects for treatment of hyperparathyroidism with vitamin D. J Clin Endocrinol Metab 87:5826–5829. 229. Holick MF, Adams JS 1990 Vitamin D metabolism and biological function. In: Avioli LV, Krane SM (eds.) Metabolic Bone Disease and Clinically Related Disorders. W. B. Saunders: Philadelphia, PA pp 155–195. 230. Adams JS 1996 Hypercalcemia due to granuloma-forming disorders. In: Favus MJ, Christakos S (eds.) Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism. Lippincott-Raven: Philadelphia, PA pp. 206–209. 231. Bouillon R, Okamura WH, Norman AW 1995 Structurefunction relationships in the vitamin D endocrine system. Endocr Rev 16:200–256. 232. Feldman D 1986 Imidazole derivatives as inhibitors of steroidogensis. Endocr Rev 7:409–430. 233. Glass AR, Eil C 1986 Ketoconazole-induced reduction in serum 1,25-dihydroxyvitamin D. J Clin Endocrinol Metab 63:766–769. 234. Glass AR, Eil C 1988 Ketoconazole-induced reduction in serum 1,25-(OH)2D3 and total serum calcium in hypercalcemic patients. J Clin Endocrinol Metab 66:934–938. 235. Saggese G, Bertelloni S, Baroncelli GI, DiNero, G 1993 Ketoconazole decreases the serum ionized calcium and 1,25dihydroxyvitamin D levels in tuberculosis-associated hypercalcemia. AJDC 147:270–273. 236. Pont A, Williams PL, Loose DS, Feldman D, Reitz RE, Bochra C, Stevens DA 1982 Ketoconazole blocks adrenal steroid synthesis. Ann Int Med 97:370–372. 237. Adams JS, Lee G 1997 Recovery of bone mineral density following exogenous vitamin D intoxication. Ann Int Med 127:203–206.

Vitamin D Analogs: An Overview J. WESLEY PIKE

The classical function of 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) is to regulate calcium and phosphorus homeostasis in vertebrate organisms. This regulation is achieved through direct actions by the hormone on the intestine, kidney, and bone, and through its ability to regulate inversely the production of PTH from the parathyroid glands. Intestinal absorption of calcium from the diet and conservation of calcium at the level of the kidney are considered paramount to maintaining blood levels of calcium and phosphorus in the long term, although bone provides an immediate and readily available emergency source, particularly when the dietary mineral content is deficient. These fundamental actions provided the early impetus for the use of vitamin D and its derivatives in a variety of clinical settings of deranged mineral metabolism, many of which are discussed in earlier sections of this book. More recently, it has become clear that 1,25(OH)2D3 exerts additional biologic actions on a wide range of cell and tissue types, primarily as a regulator of growth, differentiation, and general cell function. These actions of vitamin D are also discussed throughout the book from both basic science as well as clinical perspectives. Importantly, these highly pleiotropic actions of 1,25(OH)2D3 have prompted the therapeutic use of this hormone for a wide range of clinical indications. They have also provided the rationale for the development of a broad range of vitamin D analogs with selective biological activities. It is universally accepted that the primary side effect of 1,25(OH)2D3 is its tendency to raise serum calcium levels and to cause hypercalcemia and hypercalciuria. This effect leads to renal stones, soft tissue calcification and can even be lethal when extreme. Unfortunately, this toxic side effect has narrowed the use of natural vitamin D metabolites for both mineralrelated as well as novel indications, many of which are described in the following three sections of this monograph. This reality stimulated a search for alternative approaches that could reduce or alleviate these side effects, and led to the identification by Chugai Pharmaceuticals of novel analogs of vitamin D (see Chapter 86) that only weakly stimulate bone calcium mobilization or intestinal calcium absorption (defined here as analogs with low calcemic potential or LCP). This exciting discovery spurred the development of VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

several pharmaceutical programs, particularly those at Roche and Leo Pharmaceutical Companies (see Chapters 84 and 85), aimed at synthesizing additional vitamin D analogs with even more selective properties. Interestingly, while considerable success has been achieved in the development of new analogs with LCP, neither the structural rules for preparing such analogs nor the mechanisms whereby these compounds exert their selective actions are completely understood. Unfortunately, the intense focus on analogs with LCP has limited efforts to develop additional novel ways of circumventing 1,25(OH)2D3 induced hypercalcemia. Accordingly, it has been only recently that specific alternative approaches such as selective and intermittent treatment regimes and combination therapies have been considered. These approaches in conjunction with new and perhaps more selective vitamin D analogs could provide the means of treating a wide variety of cancers for which therapies are currently unavailable. The natural actions of 1,25(OH)2D3 in calcium and phosphorus homeostasis have been studied for many decades. Nevertheless, it is perhaps most vividly highlighted in the human syndromes of 25-hydroxyvitamin D3-1α-hydroxylase (1-OHase) deficiency (Chapter 71) and vitamin D receptor dysfunction (hereditary 1,25(OH)2D3 resistant rickets or HVDRR, see Chapter 72). Importantly, the effects of these genetic alterations in humans have been recapitulated most recently in mice using homologous recombination techniques (see Chapters 7 and 20). Despite the utility of the mouse models for in-depth experimentation, however, it is interesting to note that several of the most important principles of vitamin D actions have been illuminated through the study of the human syndromes. For example, the skeletal abnormalities observed as a consequence of HVDRR in humans are rescued via administration of high levels of calcium and phosphorus. A similar observation has been made in the VDR null mouse. This early discovery in humans and subsequently in mice emphasizes the indirect role of vitamin D in both skeletal formation and mineralization while at the same time highlighting its direct role in the intestine and perhaps kidney. Whether all skeletal abnormalities are fully ameliorated under high dietary calcium and phosphorus conditions is presently the subject of much debate. It seems likely that the VDR Copyright © 2005, Elsevier, Inc. All rights reserved.

1404 null mice strain will provide the means to answer this important question. While high calcium and phosphorus levels rescue many of the phenotypic defects that arise in both the human syndromes and in the two animal models, it is unable to rescue all of them. One of particular interest is a hair follicle abnormality that results in alopecia. Surprisingly, while fully penetrant in VDR-null mice, this defect is seen only in a subset of patients with HVDRR and is not seen in 1-OHase deficiency in either humans or mice. These studies have stimulated a working hypothesis that alopecia results only from mutations in the VDR locus that fully compromise the expression of the VDR itself (see Chapters 13 and 72). If true, it would identify a new and potentially unique role for the VDR in hair production that is either independent of 1,25(OH)2D3 or dependent upon a novel and as yet uncharacterized ligand. Collectively, these findings highlight both the complex interrelationship that exists between calcium and phosphorus homeostasis and vitamin D and how clinical disease can call attention to the most intricate of molecular details. Why consider these issues in the context of sections on vitamin D analogs and their possible indications? It is clear from the above that the biologic effects of

J. WESLEY PIKE

1,25(OH)2D3 can be direct or indirect through various combinations of regulators. Keratinocyte differentiation, as an example, is regulated by both 1,25(OH)2D3 and local calcium concentrations, the latter operating through the modulation of specific transcription factors that act in concert with the VDR (see Chapter 35). Other biological activities, in contrast, arise directly as a result of either 1,25(OH)2D3 or calcium action. Of note are recent studies in the NOD mouse, a model for human type 1 diabetes and in experimental autoimmune encephalomyelitis (EAE), a model for human multiple sclerosis (MS). Interestingly, 1,25(OH)2D3 suppresses the development of diabetes in the NOD mouse even in the absence of hypercalcemia (see Chapters 98 and 99). Based upon this finding, a rationale for the prevention or treatment of clinical type 1 diabetes with new vitamin D analogs with LCP seems attractive. On the other hand, intervention in MS, for example, where both hypercalcemia and vitamin D may play independent roles, could be more problematic. Clearly, delineating the specific roles of vitamin D in both basic biological processes and various disease states will be a prerequisite to the successful application of vitamin D analogs for effective therapeutic indications.

CHAPTER 80

Overview: Rational Design of 1α,25-Dihydroxyvitamin D3 Analogs (Deltanoids) GARY H. POSNER AND MEHMET KAHRAMAN Department of Chemistry, School of Arts and Sciences, The Johns Hopkins University, Baltimore, Maryland

I. Introduction II. Rationale Based on Metabolism III. Rationale Based on Molecular Biology

IV. Rationale Based on Organic Chemistry V. Conclusions References

I. INTRODUCTION

following this one provide more detailed discussions of the design rationale and of the therapeutic value of several U.S. government-approved deltanoid drugs, as well as of deltanoid drug candidates that are now in human clinical trials.

Although the natural hormone 1α,25-dihydroxyvitamin D3 [1,25(OH)2D3, calcitriol, Rocaltrol (Chart I) can inhibit the growth of various human cancers and can regulate the human immune system [1], when used at supraphysiological levels this hormone can produce serious side effects such as severe hypercalcemia and soft tissue calcifications [2]. Therefore, the fundamental challenge to organic and medicinal chemists working on designing analogs of 1,25(OH)2D3 as new drugs is to incorporate structural changes and functional group modifications leading to new deltanoids (vitamin D analogs) that are still efficacious, but safer and more selective than the natural hormone (see Chapters 82–87). Such new deltanoids often serve also as sensitive probes of the fundamental molecular biology underlying a deltanoid’s mechanism of action and its biological activity profile. With several thousand deltanoids having been synthesized and evaluated [3], two generalizations among many are as follows: 1) it is still not possible to predict reliably what specific new structural or functional group change will produce a desirably efficacious but low-calcemic deltanoid and 2) progress has been made by both trial and error and by rational design. This chapter will focus on three important rationales that have guided, and continue to guide, design of new deltanoids as drugs (Chart I) and as drug candidates (Chart II): metabolism, molecular biology, and organic chemistry. Generalized deltanoid Structure 1 (chemical structures are indicated in the text by bold numerals) summarizes the most successful kind of structural modifications of 1,25(OH)2D3 that have been accomplished already. The chapters immediately VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

II. RATIONALE BASED ON METABOLISM A. Prodrugs Biosynthesis of the hormone 1,25(OH)2D3 from 7-dehydrocholesterol [4] is outlined in Scheme 1 and is fully discussed in Chapters 2–5. Some of the first deltanoid drugs were designed based on this biosynthetic route. For example, 25(OH)D3 is the immediate precursor to 1,25(OH)2D3 on this biochemical path. Because 25(OH)D3 is less calcemic than the hormone 1,25(OH)2D3 [5], this prodrug Calderol (Chart I) has been used in humans to treat renal failure and osteoporosis; its slow and steady enzymatic conversion in the kidney into low levels of 1,25(OH)2D3 [6,7] gives it a good balance of efficacy and safety (i.e. a good therapeutic index). Likewise, the clinically-used drug 1(OH)D3 (One-Alpha, Chart I) is a precursor requiring enzymatic C-25 hydroxylation to produce low and constant amounts of the natural hormone 1,25(OH)2D3 [8]. Similarly, the drug Hectorol (Chart I) requires liver enzymepromoted C-25 hydroxylation to form 1,25(OH)2D2 (the hormonal form of vitamin D2) [9]. The more recent drug Zemplar (Chart I) is the low-calcemic 19-nor version of 1,25(OH)2D2 [10]. Also, 1(OH)D5 is being developed as an enzymatically activated prodrug for 1,25(OH)2D5 [11]. Copyright © 2005, Elsevier, Inc. All rights reserved.

1406

GARY H. POSNER AND MEHMET KAHRAMAN

Chart I: DELTANOIDS IN USE AS DRUGS Indications Deltanoid

Deltanoid OH

H H

HO

Secondary hyperparathyroidism

Renal failure Rocaltrol (Hoffmann-La Roche) Osteoporosis

OH

OH

H

H

Renal failure

H

HO

HO

Renal failure One-Alpha (Leo Pharmaceutical) Osteoporosis OH

Secondary hyperparathyroidism

H

Secondary hyperparathyroidism

H

Osteoporosis Calderol (Organon)

OH H

H

HO

Indications

H H H

Hectorol (Bone Care International) OH

HO

Zemplar (Abbott Laboratories)

OH

OH

OH

Psoriasis

Psoriasis 24

24

H

H

H

H

H

HO

Secondary hyperparathyroidism

Tacalcitol (Teijin)

Dovonex (Leo Pharmaceutical) HO

OH

OH

O H

OH

H

Secondary hyperparathyroidism Psoriasis

Maxacalcitol (Chugai OCT) HO

OH

B. Catabolism Inhibitors The major catabolism of 1,25(OH)2D3 involves enzyme-mediated side chain hydroxylation and, ultimately, side chain fragmentation producing therapeutically inactive calcitroic acid (Scheme 2 and Chapter 81). If such catabolism could be slowed or avoided by small structural and/or functional group changes to the

natural hormone, then perhaps very small doses of such new deltanoids could be used therapeutically without causing hypercalcemia [12]. This rationale has led successfully to a variety of new deltanoids in human clinical trials (Chart II) and to others as promising potential drug candidates. Structural changes that inhibit deltanoid catabolism can be divided into two main categories: remote and nearby.

1407

CHAPTER 80 Overview: Rational Design of 1α,25-Dihydroxyvitamin D3 Analogs (Deltanoids)

Chart II: DELTANOIDS IN CLINICAL TRIALS Deltanoid

Indications

Deltanoid

Indications CF3

Osteoporosis

OH

OH

H

CF3 H

H

Ro 26-9228 (Hoffmann-La Roche)

HO

HO

OH

Pancreas tumor

1. REMOTE STRUCTURAL CHANGES

Two publications appearing simultaneously in 1993 were the first reports that remote structural/functional changes can influence the rate of side-chain catabolic oxidation [13,14]. Hoffmann-La Roche’s 16-en-24-oxo1,25(OH)2D3 (2) resists enzymatic 23-hydroxylation [13], and Johns Hopkins’ homolog 1-CH2OH-25(OH)2D3 3 resists 24-hydroxylation in human leukemic cells under the conditions in which 1,25(OH)2D3 is easily 24-hydroxylated [14]. These two reports were the first indications that the presence of the 16-ene modification, as in deltanoid 2, and the presence of a very remote Drug Design Alterations

π-bonds/heteroatoms extra carbons

}

20-epi

Side Chain 20

C, D-Ring

X

24

X

16

OH

X=Me; CF3; Et

16-ene H Y

HO

H

HO

OH O

OH

3 2 1

Osteoporosis

Colon tumor Seocalcitol (Leo EB 1089)

A Ring

OH

H

Breast tumor

H

Seco-B-Ring

OH

Neoplasm

H

HO

Osteoporosis Falecalcitriol (Taisho/Sumitomo)

F

22 24

Secondary hyperparathyroidism

Y

Y, Y = CH2; H, H (19-nor)

Z

Z = 1α-OH; 1β-OH; 1β-CH2OH OR; =CH2 1

STRUCTURE 1

ED-71 (Chugai) OH

extra methylene group at the 1-position, as in deltanoid 3, could alter the ligand-enzyme interaction sufficiently to cause a significant slowing in the rate of the enzymatic side chain oxidation. 2. NEARBY STRUCTURAL CHANGES

Because catabolism of the hormone 1,25(OH)2D3 occurs mainly toward the terminus of the side chain (e.g. at C-24 and C-26), retarding such enzymatic hydroxylation and thus prolonging biological half-life has been achieved effectively by incorporating side chain structural changes that operate via steric and electronic effects. Chart II shows Ro 26-9228 [15] and Leo EB 1089 [16], both low-calcemic selective agonists now in human clinical trials, in which unsaturation at C-24 makes the sp2-hybridized C-24 vinyl-C-H bond stronger and thus less easily oxidized than the normal sp3-hybridized C-24-H bond [17]. Both Ro 26-9228 and Leo EB 1089 also incorporate an enlarged and therefore sterically protective diethyl carbinol environment near C-24, relative to natural 1,25(OH)2D3 having a smaller dimethyl carbinol environment near C-24. Also in Chart II, Falecalcitriol incorporates two side chain CF3 groups that prevent enzymatic C-26 oxidation. Replacing a C-H group by a C-F group, often a miniscule change in a large deltanoid, leads to decreased catabolism because a C-F bond is much stronger and, therefore, less easily broken (i.e. oxidized) than a C-H bond [18]. Thus, in the early 1980s, 24F21,25(OH)2D3 (4) and related 24-fluorinated deltanoids were prepared [19,20]. Although C-24 catabolic hydroxylation was indeed prevented in these compounds, they

1408

GARY H. POSNER AND MEHMET KAHRAMAN

25

25

H

H

H

hv

∆

1

1

7

1

H

H HO 3β

HO 3β

7-Dehydrocholesterol

Previtamin D3

H

Vitamin D3 (s-cis conformer)

H

1. DBP

H

1,25(OH)2D3

25

OH

H

1. DBP

H 6

2. Liver

1α-Hydroxylase 1α OH

6

25

OH

2. Kidney

HO 3β

H 7

HO 3β

25

H

25

7

25-Hydroxylase

HO

HO 3β Vitamin D3

25(OH)D3

(s-trans conformer)

SCHEME 1 Biosynthesis of 1α,25-Dihydroxyvitamin D3

were disappointingly similar to 1,25(OH)2D3 in terms of high calcemic activity [19,20]. In sharp contrast, hybrid deltanoid Hopkins QW 1624F2-2 5 [21], blocked toward C-24 hydroxylation by two fluorine atoms but carrying a calcemia-lowering 1-CH2OH group, is a selective agonist that inhibits mouse skin tumorigenesis without causing hypercalcemia or animal weight loss over four months of treatment, whereas 1,25(OH)2D3 is lethal under these conditions [22]. Large scale synthesis of 24-difluorinated Hopkins QW 1624F2-2 5 has been completed under the auspices of the NIH RAPID program, and thus this hybrid deltanoid is now available to the scientific community for further study as a molecular probe and as a drug candidate. C. ACTIVE METABOLITES

When studying the biological profile of any new compounds, it is prudent to check whether any metabolites are physiologically active. Two examples in the vitamin D field are especially noteworthy. Deltanoid 16-ene-24-oxo-1,25(OH)2D3 (2), a C-24 oxidation metabolite of synthetic 16-ene-1,25(OH)2D3, has some promising therapeutic potential [23]. Also, 3-epimerization has been documented as an important and apparently common metabolic process leading to A-ring cis-oriented 1α,3α-diol deltanoids, many of which are physiologically active [24].

III. RATIONALE BASED ON MOLECULAR BIOLOGY Rational design of new deltanoids that selectively modulate the nuclear vitamin D receptor (VDR) [25] has been enlightened by X-ray crystal structures of the VDR complexed to 1,25(OH)2D2 and of the VDR complexed also to some other superagonist deltanoids [26–29]. Among the important learnings from these crystal structures are the following: 1) the side chain tertiary OH group of 1,25(OH)2D3 is bound between H-bond acceptor His 305 and H-bond donor His 397; 2) the ligand occupies only 56% of the VDR-binding domain cavity (vs. 66% occupation by retinoic acid); and 3) the ligand-binding domain (LBD) cavity is substantially larger (697 Å3) than in the related hormone receptor RXRγ (412 Å3). Consistent with this relatively large LBD cavity of the VDR, diverse conformationally flexible and yet biologically potent 19-nor deltanoids lacking a D-ring (6), or a C-ring (7), or both C- and D-rings (8) have been reported [30–33]. Impressively, despite the absence of both C- and D-rings, deltanoid 8 has been selected by Hoffmann-La Roche for development as a low-calcemic oral antipsoriatic drug candidate [33]. Based on the reported crystal structure of 1,25(OH)2D3 in the LBD of VDR [26–28], computeraided molecular modeling and conformational analysis

1409

CHAPTER 80 Overview: Rational Design of 1α,25-Dihydroxyvitamin D3 Analogs (Deltanoids)

OH 25

24

OH

H

OH

H

H

H 24-hydroxylase

HO 3β

HO

1α OH

OH

1,24,25(OH)3D3

1,25(OH)2D3

24-oxidation O

O

23

H

OH

24

OH

H

H

OH

H 23-oxidation

OH

HO

HO

OH

COOH H

H

HO 3β

1

OH

calcitroic acid

SCHEME 2 Major Catabolism of 1α,25-Dihydroxyvitamin D3

have assisted rational design of deltanoids that bind to the LBD and that are potent and selective. Among the successful deltanoids designed in this way are the following: 1α-OCH2Ar deltanoid 9 that selectively restores activity to the VDR mutant associated with rickets [34]; nonsteroidal symmetrical bisphenol derivative 10 (see Chapter 88) that also is a selective

agonist for a rickets-associated mutant of the VDR [35]; 22-ethyl deltanoid 11 that is 100 times more potent than the natural hormone in cell-differentiating activity [36–38] and that has a low energy side chain conformation occupying the “active space region” [37–40]; and arocalciferol 12, one of the first examples of a deltanoid with a conformationally restricted side chain [41].

1410

GARY H. POSNER AND MEHMET KAHRAMAN

O 24 23

24

OH

OH

H

16

H

H

3

HO

1

HO

OH

CH2OH

2

3

STRUCTURES 1 AND 2

F

F

F

24

F 24

OH

OH

H

16

H

H

3

HO

OH

1

HO

24F2-1,25(OH)2D3(4)

CH2OH

Hopkins QW1624F2-2(5)

STRUCTURES 4 AND 5

F3C

OH

CF3 OH CF3 H

OH

HO 6

HO

OH

7

STRUCTURES 6 AND 8

HO

OH 8

OH CF3

1411

CHAPTER 80 Overview: Rational Design of 1α,25-Dihydroxyvitamin D3 Analogs (Deltanoids)

O OH

H

O

H

HO

O O

OH

O 9

10 Et OH 22

OH

H

H

H

HO

H

OH

HO 11

OH 12

O S

O

24

16

H

HO

OH JK-1624SO2 13

STRUCTURES 9–13

Also, conceptually new 16-ene-24-sulfone deltanoid 13 [42], a transcriptionally potent but low-calcemic deltanoid, docks in the mutant VDR LBD with a sulfone oxygen atom close to both His 305 (2.78 Å) and to His 397 (2.19 Å) [43]; this proximity of the sulfone oxygen to both His 305 and His 397 supports the original design rationale expecting that the sulfone group would act as an H-bond acceptor [42]. DeLuca’s group was the first to suggest that even the 25-OH of natural 1,25(OH)2D3 acts as an H-bond acceptor [44].

IV. RATIONALE BASED ON ORGANIC CHEMISTRY A. Conformation Issues The cyclohexane A-ring of 1,25(OH)2D3 is conformationally flexible. A-Ring chair-to-chair interconversion via ring flipping occurs at room temperature, thereby interconverting the α-conformer with its 1-axial-OH and 3-equatorial-OH into the corresponding β-conformer

1412

3

HOeq

GARY H. POSNER AND MEHMET KAHRAMAN

OHax

H OHeq

3

1

H

(Eq. 1)

1

H

OHax α-conformer

OH

7

β-conformer

HO 6

1 3

with its 1-equatorial-OH and 3-axial-OH, and vice versa (Eq. 1). The A-ring conformation of a deltanoid may change when bound to the VDR. In order to use 19F NMR to probe A-ring conformation in VDRcomplexed deltanoids, 4,4-difluoro-1,25(OH)2D3 (14) and 19-fluoro-1,25(OH)2D3 (15a and 15b) were synthesized [45–47]. Using 19F NMR, 4,4-difluorinated deltanoid 14 was shown to be approximately a 1:1 mixture of A-ring α- and β-conformers, but both geometric isomers 15a and 15b of the 19-fluorinated deltanoid exist in only the α-conformer. The seco-B-ring of 1,25(OH)2D3 with its conjugated triene system is also conformationally flexible. As shown in Scheme 1 for the biosynthesis of 1,25(OH)2D3, vitamin D3 exists as two extreme conformers: 6-s-cis and 6-s-trans. The “s” in these designations refers to the “single bond” between carbon atoms 6 and 7; rotational barriers in carbon-carbon single bonds generally are low, thereby allowing easy interconversions at room temperature between the 6-s-cis and 6-s-trans single bond conformers of vitamin D3 and also of 1,25(OH)2D3. The Riverside group has synthesized and evaluated some deltanoids (e.g. 16) in which the 6,7-single bond is locked into only the 6-s-cis conformation; such exclusive 6-s-cis deltanoids have been pivotal in suggesting that the 6-s-cis conformer

H

OH

H R R′

X HO

OH

4,4-difluorocalcitriol (X = CF2, R = R′ = H, 14) (E)-19-fluorocalcitriol (X = CH2, R = F, R′ = H, 15a) (Z)-19-fluorocalcitriol (X = CH2, R = H, R′ = F, 15b)

STRUCTURES 14 AND15

OH (6-s-cis-locked deltanoid)

STRUCTURE 16

seems responsible for a deltanoids’ nongenomic rapid biological action (Chapter 23), whereas the 6-s-trans conformer seems responsible for genomic responses [48–50].

B. Stereoid Precursors Considerable pharmaceutical company research relating chemical structure to biological activity (structure-activity relationship, SAR) has produced the potent 22-oxa deltanoids Leo KH-1060 (Chapter 84) and Chugai OCT (Chapter 86) (Scheme 3). Such SAR generalizations quantify the antiproliferative potency advantage (Scheme 3, Leo KH-1060, increases in antiproliferative activity are in parentheses) gained by altering the skeleton of 1,25(OH)2D3 via C-20 epimerization (27X), C-22 oxygenation (10X), and side chain homologation at C-24 and C-27 (2.5X) [3]. Both 22-oxa deltanoids Leo KH-1060 and Chugai OCT are prepared typically from steroid precursors (Scheme 3). Thus, C-20 ketone 17 is derived from C-22 aldehyde 18, which is prepared via ozonolysis of triene-protected steroidal vitamin D2 followed ultimately by allylic 1-hydroxylation [51]. Likewise, C-20 alcohol 19 is prepared from the steroid dehydroepiandrosterone (DHEA) [52]. The major advantage of preparing new deltanoids from almost entire steroid precursors is that most of the deltanoid skeleton, including absolute stereochemistry, is available without needing costly and time-consuming multistep synthesis. The major advantage of using small parts of steroids for construction of new deltanoids is versatility. A shown in Scheme 4, oxidative cleavage and then in situ reduction of inexpensive and readily available vitamin D2 (ergocalciferol) produces versatile, enantiomerically pure deltanoid building block 20 [53]. This C,D-ring unit 20 can be converted easily into diastereomeric

1413

CHAPTER 80 Overview: Rational Design of 1α,25-Dihydroxyvitamin D3 Analogs (Deltanoids)

10X 2.5X

26 22

27X

O

24a

20

H

H

HO 3β

24b

22

OH

O

25

20

27

OH

H

H

10X

HO 3β

1α OH

Lexacalcitol (LeoKH-1060)

1α OH

Maxacalcitol (Chugai OCT)

O

20

H OH

20

H

R3SiO

H

1

H

3

R3SiO 19

R3SiO

OSiR3 17 O

O 22 H

H H HO

H

DHEA

R3SiO

OSiR3 18 22

H

H Vitamin D2

HO

SCHEME 3

1414

GARY H. POSNER AND MEHMET KAHRAMAN

C-22 aldehydes 21 and 22 (differing only by C-20 stereochemistry) and then into C-20-epi C-22 alcohol 23 and into C-20 ketone 24 [53]. Ketone 24 can be converted into C-17 ketone 25 via Baeyer-Villiger oxidation, hydrolysis, and then C-17 oxidation

(Scheme 4). These enantiomerically pure C,D-ring building blocks allow versatile attachment of diverse side chains and versatile attachment via Lythgoe coupling [54] also of A-ring portions. Two examples follow.

22 H OH H

H

1. O3 2. NaBH4 H OH 20

HO Vitamin D2

1. Et3SiOTf 2. Swern Oxidation O O

20

H 8

1. HN

20

O

2. 1O2

H Et3SiO

22 H H

H Et3SiO

24

21

nBu4NOH O O

20

17

22 H H

8 H

H

Et3SiO

Et3SiO

25

22 NaBH4

20

Et3SiO 23

SCHEME 4

H

22 OH H

1415

CHAPTER 80 Overview: Rational Design of 1α,25-Dihydroxyvitamin D3 Analogs (Deltanoids)

20

1. Me3SiOTf

H

20

Me3SiO

SO2t-Bu

O

O

22

O

O

S

H

2. Et3SiH

8

H

H

RO

AcO

24 R = Et3Si

24a R = Ac

O

20

P(O)Ph2

O

O

22

S 26

1.

H

R3SiO

H

OSiR3

O

O

22 20

O

S

H

2. n-Bu4NF

H O HO 3β

1α OH 26

SCHEME 5

O

OH

HOOC

OH

HO

t-BuMe2SiO

2

OSiMe2t-Bu

OH (-)-quinic acid

20 H

25 OH

1. n-BuLi 2. 20 H

8 8

H

28 O H 3. Cation exchange resin

H

HO

2

OSiEt3

H

t-BuMe2SiO

OH

P(O)Ph2

2

OSiMe2t -Bu 27

29

SCHEME 6

1416

GARY H. POSNER AND MEHMET KAHRAMAN

C-20 Ketone 24a has been converted via reductive etherification into nonclassical side chain sulfone deltanoid 26 (Scheme 5), a powerfully antiproliferative and transcriptionally active but low-calcemic new deltanoid [53]. Also, natural enantiomerically pure (-)quinic acid has been converted into A-ring phosphine oxide 27 that was joined with (20S)-25-oxy Grundmann C-8 ketone 28 via Lythgoe coupling [54] to produce 2-methylene-19-nor-20-epi-1α,25-dihydroxyvitamin D3 (29) (Scheme 6) [55]. This new deltanoid potently and selectively induces bone formation in rats [56] (see Chapter 87).

C. Multistep Synthesis Multistep synthesis of key building blocks that can be joined to form new deltanoids has been reviewed [54,57]. Structurally and stereochemically complex deltanoid building blocks still often offer serious synthetic

challenges for organic chemists. Hybrid deltanoids, modified in two remote regions, are best prepared by joining small parts of natural stereoids (typically the enantiomerically pure C,D-ring) with a nonclassical side chain and then with a nonclassical A-ring. Both side chain and Aring are usually prepared via multistep syntheses. Such convergent synthesis allows for much versatility in selecting components to join and, in principle, allows for small libraries of new deltanoids to be produced [58]. Three examples of multistep syntheses of therapeutically desirable hybrid deltanoids follow. The Théramex 14-epi-19-nor-23-yne hybrid deltanoid 33 is in advanced clinical trials for treatment of psoriasis [59,60]. It has been prepared via convergent coupling of A-ring cyclopropane aldehyde 31 (prepared from precursor cyclohexanediol 30 in 9 steps) and C,D-ring bromo-olefin 32 (prepared in about 7 steps from vitamin D2) (Scheme 7). 1α-Fluoro-16,23-diene-20-epi hybrid deltanoid Ro 26-9228 is in human clinical trials for treatment of

Vitamin D2

7 steps

OEt H H

O

14

14

H Br

7

H

32 6

+ 6

7 H

CHO

A HO

H OH

A TX 522

OSiMe2t-Bu 31

33

9 steps

CO2Me

OH

HO 30

SCHEME 7

OH

1417

CHAPTER 80 Overview: Rational Design of 1α,25-Dihydroxyvitamin D3 Analogs (Deltanoids)

23 CHO

20 16 H R3SiO 34, 20-natural 35, 20-epi

F 20

OSiMe3

OSiMe3 16

8

8 O

24

20

23

F

H

H

O 38

36

P(O)Ph2

P(O)Ph2

1

1

t-BuMe2SiO

t-BuMe2SiO

F

39

37 F 20

23

OH 16

H

H 7

7

1

1 HO

F 24

20 OH

16

8

OSiMe2t-Bu

F

HO

Ro 26-9228

CH2OH(5)

Hopkins QW 1624F2-2

SCHEME 8

osteoporosis [61,62]. It has been prepared via convergent coupling of 16,23-diene 8-ketone 36 (prepared from 20-epi-23-aldehyde 35) and A-ring 1-fluorinated unit 37 (Scheme 8). Also, Hopkins QW-1624F2-2 5, a low-calcemic cancer chemopreventive hybrid deltanoid [22], has been synthesized by coupling 24-difluorinated C,D-ring 8-ketone 38 (prepared from 20-natural 23-aldehyde 34) with A-ring 1-homologated unit 39 (Scheme 8) [21].

Scheme 9 is provided to illustrate the multistep preparation from vitamin D2 of diastereomeric C,D-ring building blocks 34 and 35 used in Scheme 8 to prepare hybrid deltanoids Ro 26-9228 and Hopkins QW-1624F2-2 5. Wittig olefination of C-17 ketone 25 produces 17-ethylidene derivative 40-Z that can be isomerized into 40-E [63]. Allylic oxygenation at C-16 produces allylic alcohol 41-Z and separately allylic alcohol 41-E. Then, concerted and highly

1418

GARY H. POSNER AND MEHMET KAHRAMAN

O 17

PPh3

17 16

8

17 16

8

8 H

H

H R3SiO

R3SiO

R3SiO

25

40-E

40-Z SeO2

steps

SeO2

Vitamin D2 16

16

OH

OH

8

8

H

H R3SiO

R3SiO

41-E

41-Z

23 CHO

20 16 8

23 CHO

20 16 8 H

H R3SiO

R3SiO

34, 20-natural

35, 20-epi

SCHEME 9

stereocontrolled [3,3]-sigmatropic rearrangement of the corresponding allyl vinyl ethers produces the required 16-ene C-23 aldehydes 34 and 35 (Scheme 9) [64,65]. Although not a hybrid deltanoid, conceptually new 16-ene-25-oxime deltanoid 44 is low-calcemic but antiproliferatively and transcriptionally potent. Oxime deltanoid 44 has been prepared also from key 16-ene23-aldehyde building block 34 (Scheme 10) [66]. Noteworthy is the regiospecific joining of the A-ring nucleophile 43 with only the C-8 ketone group in C-8,C-25-diketone 42, and noteworthy also is oximation of the C-25 ketone group without disturbing the sensitive conjugated triene unit. An important advance in synthetic methodology for efficient construction of A-ring cyclohexane units directly via C-10 to C-5 bond formation in one operation from noncyclic precursors has been reported (Eq. 2) [67]. This organometallic palladium-mediated

approach to a C-6 vinylpalladium intermediate that couples in situ with C-7 vinylic bromides has been used successfully in some deltanoid syntheses (Eq. 2) [68].

V. Conclusions Rational design of new generations of deltanoids will continue to rely on understanding the metabolism and the molecular biology of existing deltanoids and also on the power of modern organic chemistry for synthesis of complex molecules. Progress will be facilitated especially by good communication and cooperation between molecule-makers and molecule-testers [69]. Identifying the best new deltanoid drug candidates ideally will involve direct head-to-head comparisons of leading deltanoid new chemical entities, irrespective of their industrial or academic origin.

1419

CHAPTER 80 Overview: Rational Design of 1α,25-Dihydroxyvitamin D3 Analogs (Deltanoids)

23 CHO

20

25

20

O 16

16

8

8

H

Et3SiO

H

O

42 34, 20-natural P(O)Ph2

t-BuMe2SiO

OSiMe2t-Bu 43

25

N MeO H

HO

OH 44

SCHEME 10

H Br

7

H

OH

H

7

6

2. n-Bu4NF 6 H

(Eq. 2)

32 % overall yield

5

t-BuMe2SiO

H

1. Catalytic (Ph3P)4Pd

+

3 HO

10 3 1 2

2

O OSiMe2t-Bu

O HO

t-BuMe2SiO

OH

1 OH

1420 References 1. Reichle H, Koeffler HP, Norman AW 1989 The role of the vitamin D endocrine system in health and disease. N Engl. J Med 320:980–991. 2. Vieth R 1990 The mechanisms of vitamin D toxicity. Bone Miner 11:267–272. 3. Bouillon R Okamura WH, Norman AW 1995 Structure-function relationships in the vitamin D endocrine system. Endocr Rev 16:200–257. 4. Jones G, Strugnell SA, DeLuca HF 1998 Current understanding of the molecular actions of vitamin D. Physiol Rev 78: 1193–1231. 5. Barger-Lux MJ, Heaney RP, Lanspa SJ, Healy JC, DeLuca HF 1995 An investigation of sources of variation in calcium absorption efficiency. J Clin Endocrinol Metab 80:406–411. 6. Rasmussen H, Wong M, Bikle D, Goodman DBP 1972 Hormonal control of the renal conversion of 25-hydroxycholecalciferol to 1,25-dihydroxycholecalciferol. J Clin Invest 51:2502–2504. 7. Chen TC, Schwartz GC, Burnstein KL, Lokeshwar BL, Holick MF 2000 The in vitro evaluation of 25-hydroxyvitamin D3 and 19-nor-1α,25-dihydroxyvitamin D2 as therapeutic agents for prostate cancer. Clin Cancer Res 6:901–908. 8. Crocker JS, Muhtadie SF, Hamilton DC, Cole DE 1985 The comparative toxicity of vitamin D metabolites in the weanling mouse. Biol Appl Pharmacol 80:119–126. 9. Gallagher JC, Bishop CW, Knutson JC, Mazess RB, DeLuca HF 1994 Effects of increasing doses of 1α-hydroxyvitamin D2 on calcium homeostasis in postmenopausal osteopenic women. J Bone Miner Res 9:607–614. 10. Slatopolsky E, Finch J, Ritter C, Denda M, Morrissey J, Brown A, DeLuca HF 1995 A new analog of calcitriol, 19-nor-1,25(OH)2D2, suppresses parathyroid hormone secretion in uremic rats in the absence of hypercalcemia. Am J Kidney Dis 26:852–860. 11. Mehta RG, Moriarty RM, Mehta RR, Penmasta R, Lazzaro G, Constantinou A, Guo L 1997 Prevention of preneoplastic mammary lesion development by a novel vitamin D analog, 1α-hydroxyvitamin D5. J Natl Cancer Inst 89:212–218. 12. Posner GH, Crawford KR, Yang HW, Kahraman M, Jeon HB, Li H, Lee JK, Suh BC, Hatcher MA, Labonte T, Usera A, Dolan PM, Kensler TW, Peleg S, Jones G, Zhang A, Korczak B, Saha U, Chuang SS 2004 Potent, Low-Calcemic, Selective Inhibitors of CYP24 Hydroxylase: 24-Sulfone Analogs of the Hormone 1α,25-Dihydroxyvitamin D3. J Steroid Biochem Mol Biol, 89–90:5–12. 13. Reddy GS, Clark JW, Tserng K-Y, Uskokovic MR, McLane JA 1993 Metabolism of 1,25(OH)216-ene D3 in kidney: Influence of structural modification of D-ring on side chain metabolism. Bioorg Med Chem Lett 3:1879–1884. 14. Posner GH, Guyton KZ, Kensler TW, Barsony J, Lieberman ME 1993 1-(Hydroxyalkyl)-25-hydroxyvitamin D3 analogs of calcitriol 2. Preliminary biological evaluation. Bioorg Med Chem Lett 3:1835–1840. 15. Peleg S, Uskokovic M, Ahene A, Vickery B, Avnur Z 2002 Cellular and molecular events associated with the boneprotecting activity of the noncalcemic vitamin D analog Ro-26-9228 in osteopenic rats. Endocrin 143:1625–1636. 16. Hansen CM, Hamberg KJ, Binderup E, Binderup L 2000 Seocalcitol (EB 1089): A vitamin D analog of anticancer potential. Background, design, synthesis, preclinical, and clinical evaluation. Curr Pharm Design 6:803–828.

GARY H. POSNER AND MEHMET KAHRAMAN

17. Smith MB, March J. “Advanced Organic Chemistry.” WileyInterscience:New York, NY, Fifth Edition, 2001, Chapter 1. 18. Banks RE, Smart BE, Tatlow JC (Eds.) “Organofluorine Chemistry: Principles and Commercial Applications” Plenum Press: New York, 1994. 19. Okamoto S, Tanaka Y, DeLuca HF, Kobayashi Y, Ikekawa N 1983 Biological activity of 24,24-difluoro-1,25-dihydroxyvitamin D3. Am J Physiol 244:E159–E163. 20. Gill HS, Londowski JM, Corradino RA, Zinsmeister AR, Kumar R 1990 Synthesis and biological activity of novel vitamin D analogs: 24,24-Difluoro-25-hydroxy-26,27-dimethylvitamin D3 and 24,24-difluoro-1α,25-dihydroxy-26,27dimethylvitamin D3. J Med Chem 33:480–490. 21. Posner GH, Lee JK, Wang Q, Peleg S, Burke M, Brem H, Dolan PM, Kensler TW 1998 Noncalcemic, antiproliferative, transcriptionally active, 24-fluorinated hybrid analogs of the hormone 1α,25-dihydroxyvitamin D3. Synthesis and preliminary biological evaluation. J Med Chem 41:3008–3014. 22. Kensler TW, Dolan PM, Gange SJ, Lee J-K, Wang Q, Posner GH 2000 Conceptually new deltanoids (vitamin D analogs) inhibit multistage skin tumorigenesis. Carcinogen 21:1341–1345. 23. Siu-Caldera ML, Clark JW, Santos-Moore A, Peleg S, Liu YY, Uskokovic MR, Sharma S, Reddy GS 1996 1α,25-Dihydroxy24-oxo-16-ene vitamin D3, a metabolite of a synthetic vitamin D3 analog, 1α,25-dihydroxy-16-ene vitamin D3, is equipotent to its parent in modulating growth and differentiation of human leukemic cells. J Steroid Biochem Mol Biol 59: 405–412. 24. Brown AJ, Ritter C, Slatopolsky E, Muralidharan KR, Okamura WH, Reddy GS 1999 1α,25-Dihydroxy-3-epivitamin D3, a natural metabolite of 1α,25-dihydroxyvitamin D3, is a potent suppressor of parathyroid hormone secretion. J Cell Biochem 73:106–113. 25. Peleg S, Posner GH 2003 Vitamin D analogs as modulators of vitamin D receptor action. Curr Trends Med Chem 3:1555–1572. 26. Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Molecular Cell 5:173–179. 27. Tocchini-Valentini G, Rochel N, Wurtz JM, Mitschler A, Moras D 2001 Crystal structures of the vitamin D receptor complexed to superagonist 20-epi ligands. Proc Natl Acad Sci USA, 98:5491–5496. 28. Verboven C, Rabijns A, De Maeyer M, Van Baelen H, Bouillon R, De Ranter C 2002 A structural basis for the unique binding features of the human vitamin D–binding protein. Nat Struct Biol 9:131–136. 29. Choi M, Yamamoto K, Masuno H, Nakashima K, Taga T, Yamada S 2001 Ligand recognition by the vitamin D receptor. Bioorg Med Chem 9:1721–1730. 30. Zhou X, Zhu G-D, Van Haver D, Vandewalle M, De Clercq PJ, Verstuyf A, Bouillon R 1999 Synthesis, biological activity, and conformational analysis of four seco-D-15,19-bisnor-1α,25dihydroxyvitamin D analogs, diastereomeric at C17 and C20. J Med Chem 42:3539–3556. 31. Wu Y, Sabbe K, De Clercq P, Vandewalle M, Bouillon R, Verstuyf A 2002 Vitamin D3: Synthesis of seco C-9,11,21trisnor-17-methyl-1α,25-dihydroxyvitamin D3 analogs. Bioorg Med Chem Lett 12:1629–1632. 32. Verstuyf A, Verlinden L, Van Baelen H, Sabbe K, D’Hallewyn C, De Clercq P, Vandewalle M, Bouillon R 1998 The biological activity of nonsteroidal vitamin D hormone analogs lacking both the C- and D-rings. J Bone Miner Res 13:549–558.

CHAPTER 80 Overview: Rational Design of 1α,25-Dihydroxyvitamin D3 Analogs (Deltanoids) 33. Bauer FW, Barbier P, Mohr P, Phister T, Pirson W, Theil F-P 2000 Preclinical profile of the cyclohexanediol Ro 65-2299, a potential oral antipsoriatic. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D Endocrine System: Structural, Biological, Genetic, and Clinical Aspects. University of California, Riverside: Printing and Reprographics, pp. 615–618. 34. Swann SL, Bergh JJ, Farach-Carson MC, Koh JT 2002 Rational design of vitamin D3 analogs, which selectively restore activity to a vitamin D receptor mutant associated with rickets. Org Lett 4:3863–3866. 35. Swann SL, Bergh J, Farach-Carson MC, Ocasio CA, Koh JT 2002 Structure-based design of selective agonists for a ricketsassociated mutant of the vitamin D receptor. J Am Chem Soc 124:13795–13805. 36. Masuno H, Yamamoto K, Wang X, Choi M, Ooizumi H, Shinki T, Yamada S 2002 Rational design, synthesis, and biological activity of novel conformationally-restricted vitamin D analogs, (22R)- and (22S)-22-ethyl-1,25-dihydroxy-23,24didehydro-24a,24b-dihomo-20-epivitamin D3. J Med Chem 45:1825–1834. 37. Yamada S, Yamamoto K, Masuno H, Ohta M 1998 Conformation function relationship of vitamin D: Conformational analysis predicts potential side-chain structure. J Med Chem 41:1467–1475. 38. Yamada S, Yamamoto K, Masuno H 2000 Structure function analysis of vitamin D and VDR model. Curr Pharm Des 6:733–748. 39. Gabriëls S, Van Haver D, Vandewalle M, De Clercq P, Verstuyf A, Bouillon R 2001 Development of analogs of 1α,25-dihydroxyvitamin D3 with biased side-chain orientation: Methylated des-C, D-homo analogs. Chem Eur J 7:520–532. 40. Okamura WH, Palenzuela JA, Plumet J, Midland MM 1992 Vitamin D: Structure function analyses and the design of analogs. J Cell Biochem 49:1–9. 41. Figadere B, Norman AW, Henry HL, Koeffler HP, Zhou J-Y, Okamura WH 1991 Studies of vitamin D (calciferol) and its analogs. 39. Arocalciferols: synthesis and biological evaluation of aromatic side-chain analogs of 1α,25-dihydroxyvitamin D3. J Med Chem 34:2452–2463. 42. Posner GH, Wang Q, Han G, Lee JK, Crawford K, Zand S, Brem H, Peleg S, Dolan P, Kensler TW 1999 Conceptually new sulfone analogs of the hormone 1α,25-dihydroxyvitamin D3: Synthesis and preliminary biological evaluation. J Med Chem 42:3425–3435. 43. Unpublished results of Drs. H. Masuno and S. Yamada, Institute of Biomaterials and Bioengineering, Tokyo Medical and Dental University, Tokyo, Japan. We thank Drs. Masuno and Yamada for these docking data. 44. Ostrem VK, Lau WF, Lee SH, Perlman K, Prahl J, Schnoes HK, DeLuca, HF 1987 Induction of monocytic differentiation of HL-60 cells by 1,25-dihydroxyvitamin D analogs. J Biol Chem 262:14164–14171. 45. Shimizu M, Iwasaki Y, Ohno A, Yamada S 1997 Synthesis of fluorovitamin D analogs for conformational analysis of ligand bound to vitamin D receptor. In: Norman AW, Bouillon R, Thomasset, M (eds.) Vitamin D: Chemistry, Biology, and Clinical Applications of the Steroid Hormone: Proceedings from the Tenth Workshop on Vitamin D, Strasbourg, France— May 24–29. University of California, Riverside: Printing and Reprographics, pp. 24–25. 46. Shimizu M, Iwasaki Y, Ohno A, Yamada S 1999 4,4-Difluoro1α,25-dihydroxyvitamin D3: Analog to probe A-ring conformation in vitamin D–receptor complex. Tett Lett 40:1697–1700. 47. Shimizu M, Ohno A, Iwasaki Y, Yamada S, Ooizumi H, DeLuca HF 2000 A-ring conformation and biological activity: Based on studies of fluorovitamin D analogs. In: Norman AW,

48. 49. 50.

51. 52.

53.

54. 55.

56. 57. 58. 59.

60.

61. 62.

63.

1421

Bouillon R, Thomasset M (eds.) Vitamin D Endocrine System: Structural, Biological, Genetic and Clinical Aspects: Proceedings from the Eleventh Workshop on Vitamin D, Nashville, TN, USA—May 27–June 1. University of California, Riverside: Printing and Reprographics, pp. 45–48. Okamura WH, Do S, Kim H, Jeganathan S, Vu T, Zhu G-D, Norman AW 2001 Conformationally-restricted mimics of vitamin D rotamers. Steroids 66:239–247. Hayashi R, Fernández S, Okamura WH 2002 An 8π-electron electrocyclization leading to a 9,19-methano-bridged analog of 1α,25-dihydroxyvitamin D3. Org Lett 4:851–854. Norman AW, Bishop JE, Bula CM, Olivera CJ, Mizwicki MT, Zanello LP, Ishida H, Okamura WH 2002 Molecular tools for study of genomic and rapid signal transduction responses initiated by 1α,25(OH)2 vitamin D3. Steroids 67:457–466. Andrews DR, Barton DHR, Hesse RH, Pechet MM 1986 Synthesis of 25-hydroxy- and 1α,25-dihydroxyvitamin D3 from vitamin D2 (calciferol). J Org Chem 51:4819–4828. Kubodera N, Watanabe H, Kawanishi T, Matsumoto M 1992 Synthetic studies of vitamin D analogs. XI. Synthesis and differentiation-inducing activity of 1α,25-dihydroxy22-oxavitamin D3 analogs. Chem Pharm Bull 40:1494–1499. Posner GH, Crawford K, Siu-Caldera M-L, Reddy GS, Sarabia SF, Feldman D, Etten EV, Mathieu C, Gennaro Vouros P, Peleg S, Dolan PM, Kensler TW 2000 Conceptually new 20epi-22-oxa sulfone analogs of the hormone 1α,25-dihydroxyvitamin D3: Synthesis and biological evaluation. J Med Chem 43:3581–3586. Dai H, Posner GH 1994 Synthetic approaches to vitamin D. Synthesis 1383–1398. Sicinski RR, Prahl JM, Smith CM, DeLuca HF 1998 New 1α,25dihydroxy-19-norvitamin D3 compounds of high biological activity: Synthesis and biological evaluation of 2-hydroxymethyl, 2-methyl and 2-methylene analogs. J Med Chem 41:4662–4674. Shevde NK, Plum LA, Clagett-Dame M, Yamamoto H, Pike JW, DeLuca HF 2002 A potent analog of 1α,25-dihydroxyvitamin D3 selectively induces bone formation. Biochem 99:13487–13491. Zhu G, Okamura WH 1995 Synthesis of vitamin D (calciferol). Chem Rev 95:1877–1952. Hijikuro I, Doi T, Takahashi T 2001 Parallel synthesis of a vitamin D3 library in the solid-phase. J Am Chem Soc 123: 3716–3722. Wu Y, Zhao Y, Tian H, Clercq PD, Vandewalle M, Berthier M, Pellegrino G, Maillos P, Pascal J-C 2001 A practical synthesis of 14-epi-19-nor-1α,25-dihydroxyvitamin D3 analogs and their A-ring epimers. Eur J Org Chem 3779–3788. Verlinden L, Verstuyf A, Camp MV, Marcelis S, Sabbe K, Zhao X-Y, Clercq PD, Vandewalle M, Bouillon R 2000 Two novel 14-epi-analogs of 1α,25-dihydroxyvitamin D3 inhibit the growth of human breast cancer cells in vitro and in vivo. Cancer Res 60:2673–2679. Kabat MM, Radinov R 2001 The practical synthesis of vitamin D analogs: A challenge for process research. Curr Opin Drug Discovery Dev 4:808–833. Kabat MM, Garofalo LM, Daniewski AR, Hutchings SD, Liu W, Okabe M, Radinov R, Zhou Y 2001 Efficient synthesis of 1α-fluoro A-ring phosphine oxide, a useful building block for vitamin D analogs, from (S)-carvone via a highly selective palladium-catalyzed isomerization of dieneoxide to dieneol. J Org Chem 66:6141–6150. Blæhr LKA, Bjorkling F, Calverley MJ, Binderup E, Begtrup M 2003 Synthesis of Novel Hapten Derivatives of 1α,25dihydroxy vitamin D3 and its 20-epi analogs. J Org Chem 68:1367–1375.

1422 64. Uskokovic MR, Studzinski GP, Gardner JP, Reddy GS, Campbell MJ, Koeffler HP 1997 The 16-ene vitamin D analogs. Curr Pharm Des 3:99–123. 65. Hatcher MA, Posner GH 2002 [3,3]-Sigmatropic rearrangements: Short, stereocontrolled synthesis of functionalized vitamin D3 side-chain units. Tett Lett 43: 5009–5012. 66. Posner GH, Halford BA, Peleg S, Dolan PM, Kensler TW 2002 Conceptually new low-calcemic oxime analogs of the hormone 1α,25-dihydroxyvitamin D3: Synthesis and biological testing. J Med Chem 45:1723–1730. 67. Trost BM, Dumas J, Villa M 1992 New strategies for the synthesis of vitamin D metabolites via Pd-catalyzed reactions. J Am Chem Soc 114:9836–9845.

GARY H. POSNER AND MEHMET KAHRAMAN

68. Kittaka A, Suhara Y, Takayanagi H, Fujishima T, Kurihara M, Takayama H 2000 A concise and efficient route to 2α-(ωhydroxyalkoxy)-1α,25-dihydroxyvitamin D3: Remarkably high affinity to vitamin D receptor. Org Lett 2:2619–2622. 69. Posner GH, Jeon HB, Sarjeant A, Riccio ES, Doppalapudi RS, Kapetanovic IM, Dolan P, Kensler TW 2004 Low-calcemic, efficacious, 1α,25-dihydroxyvitamin D3 analog QW-1624F2-2: Calcemic dose-response determination, preclinical genotoxicity testing, and revision of A-ring stereochemistry, Steroids, submitted.

CHAPTER 81

Analog Metabolism GLENVILLE JONES

Departments of Biochemistry and Medicine, Queen’s University, Kingston, Ontario, Canada

I. General Considerations II. Examples of the Metabolism of Analogs of Vitamin D

I. GENERAL CONSIDERATIONS A. Vitamin D Metabolism No mention of the metabolism of vitamin D analogs can ignore the rich and varied history of the metabolism of vitamin D itself over the last few decades [1]. (See Chapter 1 of this treatise.) Metabolic investigations over the last few decades have revealed not only the nature of the hydroxylation/oxidation steps important in the activation of vitamin D but also much about the cytochrome P450-based enzymes involved. In fact, it was the elucidation of the metabolism of vitamin D that sparked the synthesis of the first vitamin D analogs back in the early 1970s. See Chapter 2 for details of natural metabolites and Chapters 4, 5, 6 for details of the 25-, 1α-, and 24-hydroxylase enzymes, since this knowledge is important background for a fuller appreciation of the metabolism of vitamin D analogs. Any review of the metabolism of vitamin D analogs should make an important distinction between:1) that metabolism providing activation of the analog which thereby leads to a more biologically active molecule and 2) that metabolism, otherwise known as catabolism, which deactivates the molecule leading to its destruction and excretion. In fact, this distinction allows for a basis for a classification of vitamin D analogs into two distinct groups: a) Prodrugs (analogs) requiring one or more step(s) of activation (e.g. one step: 1α-OH-D2, 1α-OH-D3, 1α-OH-D4, 1α-OH-D5, or 25-OH-D3; multiple steps: vitamin D2 and dihydrotachysterol) b) Analogs of 1α,25-dihydroxyvitamin D3 (calcitriol) requiring NO activation (e.g. calcipotriol, OCT, EB1089, KH1060, 19-nor-1α,25-(OH)2 D3) It is clear that many of the early generations of vitamin D analogs were prodrugs, which were designed VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

III. Important Implications Derived from Analog Metabolism Studies References

to take advantage of the enzymes already in place for the metabolic activation of vitamin D itself. Most of these analogs showed a close resemblance to vitamin D since they required sufficient similarity to the natural substance to ensure activation. More recently, chemists in the pharmaceutical industry and university domain have designed more exotic vitamin D analogs, which deviate significantly from the basic vitamin D/calcitriol structure. In the majority of these cases, it is the vitamin D side chain which has been changed in these analogs, but there now exists a significant number of examples stemming from excursions into A-ring or C/D-ring modification. Three reviews [2,3,148] and Chapters 80 and 84–88 offer comprehensive lists of the analogs of vitamin D synthesized to date. In Tables I and II, we have selected, respectively, some of the more interesting prodrugs and calcitriol analogs that are currently marketed or for which metabolic data are available. Prodrug activation (e.g., for vitamin D2 or 1α-OH-D3) may involve the same cytochrome P450-based enzymes (namely the 25- and 1α-hydroxylases) as for vitamin D3, and consequently this allows us to explore the substrate specificity of these vitamin D–related enzymes. Current dogma suggests that 25-hydroxylation of endogenous vitamin D3 is carried out by a bifunctional enzyme also capable of the 25- and 27-hydroxylation steps within pathways of bile acid metabolism [4]. Indeed, this liver mitochondrial cytochrome P450, known as CYP27A (see Chapter 4), when transfected into COS-1 cells has also been shown to be capable of the 25-hydroxylation (and other hydroxylations at C-24 and C-27) of a number of vitamin D analogs (including vitamin D3), when provided at high substrate concentrations [5]. However, there remains some doubt that CYP27A exclusively carries out the 25-hydroxylation of vitamin D3 at the substrate concentrations observed in the human in vivo [6]. There is some experimental data [collected in ref. 6] to support the idea that another enzyme CYP2R1, in the liver microsomal fraction, is also responsible for Copyright © 2005, Elsevier, Inc. All rights reserved.

1424

GLENVILLE JONES

Table I

Vitamin D Prodrugs

[3]

[2]

[1] R

CH2 3

R

H3C

Side chain structure R 21

1αOHD3 [3]

24

22

CH2 3

HO

Vitamin D prodrugs [ring structure]

R

3 OH

HO

1 OH

Possible target diseases

Company

Mode of delivery

Ref.

27

Leo

Osteoporosis

Systemic

[134]

Bone Care Int.

Hyperparathyroidism

Systemic

[135]

Bone Care Int.

Psoriasis Cancer

Topical Systemic

[56]

1αOHD5 [3]

NCI

Cancer

Systemic

[57]

Dihydrotachysterol [2]

Solvay-Duphar

Renal failure

Systemic

[29]

Vitamin D2 [1]

Various

Rickets Osteomalacia

Systemic

[136]

20

23

25

26

28

1αOHD2 [3]

1αOHD4 [3]

28

28

29

25-hydroxylation of low concentrations of endogenous vitamin D3 [6a]. Similarly, 1α-hydroxylation of vitamin D analogs lacking a 1α-hydroxyl function may not be the exclusive domain of the renal 1α-hydroxylase enzyme (containing CYP27B) described in Chapter 5. It is possible that the extra-renal forms of the 1α-hydroxylase [7] (see also Chapter 79) or other CYPs may activate vitamin D analogs (e.g. dihydrotachysterol), though this reviewer knows of no example where this is the underlying strategy of a clinical treatment protocol.

B. Calcitriol Catabolism The other side of metabolism is catabolism and this is where the majority of the emphasis of this chapter will be placed. Calcitriol is subject to catabolism by two different pathways. 1. C-24 OXIDATION PATHWAY TO CALCITROIC ACID

The C-24 oxidation pathway seems to predominate in most target cells because of the inducible nature of

1425

CHAPTER 81 Analog Metabolism

TABLE II [1]

[2]

Analogs of 1,25(OH)2D3

[3]

[4]

R

[5]

[6]

[7]

R

R

R

R

R

R 16

CH2

H3C

3 HO

CH2 3

3 OH

CH2

1

HO

HO

OH

2

OH

O

Vitamin D analog [ring structure] 1α,25(OH)2D3 [3]

Side chain structure (R)

20

23

OH

HO

OH

10 HO

OH

OH

Possible target diseases

Mode of delivery

Roche, Solvay-Duphar

Hypocalcemia Psoriasis

Systemic Topical

[137]

Sumitomo-Taisho

Osteoporosis Hypoparathyroidism

Systemic Systemic

[107]

OH

Abbott

Hyperparathyroidism

Systemic

[116]

OH

Chugai

Hyperparathyroidism Psoriasis

Systemic Topical

[88]

Leo

Psoriasis

Topical

[59]

Roche

Leukemia

Systemic

[138]

Leo

Cancer

Systemic

[78]

Leo

Immune diseases

Systemic

[62]

Leo

Immune diseases

Systemic

[63]

OH

25

26

CF3 OH CF3

26,27-F6-1α,25(OH)2D3 [3]

HO

Company

27

24

22

21

CH2

10

Ref.

28

19-Nor-1α,25(OH)2D2 [5]

O

22-Oxacalcitriol (OCT) [3]

OH

Calcipotriol (MC903) [3]

1α,25(OH)2-16-ene-23-yne-D3 (Ro 23-7553) [6]

OH

27a OH

EB1089 [3]

24a

OH

20-Epi-1α,25(OH)2D3 [3]

KH1060 [3]

26a

O

OH

Continued

1426

GLENVILLE JONES

TABLE II Vitamin D analog [ring structure]

Analogs of 1,25(OH)2D3—Cont’d

Side chain structure (R)

Possible target diseases

Company

Mode of delivery

Refs.

O

KH1650 [3] H3C

Leo

Psoriasis

Topical

[97]

CH3 OH

2-µethylene-19-Nor20-Epi-1α,25(OH)2D3 [7]

OH

Deltanoids

Osteoporosis

Systemic

[64]

ED-71 [4]

OH

Chugai

Osteoporosis

Systemic

[105]

Bone Care Int.

Psoriasis

Systemic

[47]

Teijin

Psoriasis

Topical

[139]

OH

1α,24(S)(OH)2D2 [3]

OH

1α,24R(OH)2D3 (TV-02) [3]

the mitochondrial cytochrome P450 involved, known as CYP24, which is the substrate-binding component of the 24-hydroxylase complex [8]. The C-24 oxidation pathway comprises 5 enzymatic steps (Fig. 1) involving successive hydroxylation/oxidation reactions

at C-24 and C-23 followed by cleavage of the molecule between C-23 and C-24 and oxidation of the resultant truncated product to calcitroic acid [9,10]. Targeted disruption of the murine CYP24 gene gives 50% lethality at weaning due to hypercalcemia and

R OH

1α,25(OH)2D3

OH

1α,24,25(OH)3D3

OH

24-oxo-1α,25(OH)2D3

OH

24-oxo-1α,23,25(OH)3D3

OH

R O CH2 HO

O

OH

OH

side-chain cleavage enzyme

24,25,26,27-Tetranor1α,23(OH)2D3

FIGURE 1

CH2OH

COOH

Calcitroic acid (1α,23–Acid)

C24-Oxidation pathway. (Reproduced from ref. 9 with permission.)

1427

CHAPTER 81 Analog Metabolism

substrate preferences of CYP24 or other enzyme(s) involved.

nephrocalcinosis; survivors showing much reduced capacity to clear a bolus of 1α,25-(OH)2D3 from the bloodstream [11,12]. These results suggest an important role for CYP24 in catabolism. Reconstitution assays using recombinant CYP24 produced in E. coli and bacculovirus systems have shown that the first three enzyme activities of the pathway, and possibly more, reside in a single cytochrome P450 chain [13–15]. The fact that this C-24 oxidation pathway has been connected to a known excretory product of calcitriol, in the form of calcitroic acid, adds credence to the view that the pathway is catabolic in nature. Consistent with this viewpoint is the finding that the mRNA for CYP24 and C-24-hydroxylation activity have been found in classical vitamin D target tissues including intestine, kidney, bone, as well as a variety of primary cells and cultured cell lines such as CaCo-2 (colon), UMR-106 (bone), LLC-PK1 (kidney), HPK1A-ras (keratinocyte) [16–19]. Depending upon the structure of their side chain, vitamin D analogs can also be metabolized by these same C-24 oxidation pathway enzymes, usually to molecules with reduced biological activity. Therefore, studies concerned with metabolism of vitamin D analogs have the dual role of defining metabolic products of the analog and allowing for exploration of the

2. C-26 HYDROXYLATION/26,23-LACTONE FORMATION

The role of 26-hydroxylation of vitamin D compounds is unknown. Equally obscure is the importance or the function of the lactone pathway to either 25-OH-D3-26,23-lactone or 1α,25-(OH)2D3-26,23lactone derived from 25-OH-D3 or 1α,25-(OH)2D3, respectively. The evidence for the formation of these metabolites is irrefutable; the pathway to the lactone well defined [20] (see also Fig. 2); and recent evidence suggests that CYP24 is again responsible since the recombinant enzyme catalyzes the five-step process to the lactone, which starts with the 23-hydroxylated metabolite: 23,25-(OH)2D3 or 1α,23,25-(OH)3D3 [12,21] On the other hand, it is not clear why the alternative metabolites 25,26-(OH)2D3 and 1α,25,26-(OH)3D3 are formed at all since 26-hydroxylated derivatives do not seem to be the initial precursors to either observed lactone [22]. Pathway idiosyncrasies aside, it seems clear that both 26-hydroxylation and 26,23-lactone formation result in molecules with reduced biological activity, suggesting but not conclusively proving that they represent catabolites.

25(OH)D3

OH

I

OH

HO

HO 23S,25(OH)2D3

OH

OH OH

II

HO

H

III′ HO

23S,25R,26(OH)3D3

OH OH HO HO O

IV′ HO

FIGURE 2

OH O

O

O

O

IV HO (23S,25R)-25(OH)D326,23-lactone

OH OH

III HO (23S,25R)-25(OH)D326,23-lactol

26,23-Lactone pathway. (Reproduced from ref. 22 with permission).

1428

GLENVILLE JONES

C. Nonvitamin D–Related Metabolism and General Methodology As the vitamin D molecule is increasingly modified by the organic chemist, it becomes more and more susceptible to metabolism by enzymes that are distinct from those specifically involved in vitamin D metabolism. It is the analogs of calcitriol that are the most likely to be subject to metabolism by nonvitamin D–related enzyme systems. This is because analogs of calcitriol incorporate the greatest number of structural changes and are modified mainly in that part of the molecule that is subject to metabolic alteration in vivo, specifically the side chain. In the specific examples of vitamin D analogs that follow, note that there is strong evidence for involvement of other (thus far undefined) cytochrome P450-based enzyme systems. Given the uncertainty about the identity of the enzymes involved in the metabolism of vitamin D analogs or even their tissue source, it is not surprising that most investigations involve a cross-section of metabolic systems in order to identify the metabolic products and to focus TABLE III

Biological Systems Used in the Study of the Metabolism of Vitamin D and Its Analogs

Biological system Isolated perfused organ Rat liver Rat kidney Primary cells Chick kidney Bovine parathyroid Mouse neonatal keratinocyte from wildtype αCYP24 null mice Cultured cell linesa CaCo-2, human colon (“intestinal”) UMR-106, rat osteosarcoma LLC-PK1, pig kidney HPK1A-ras, human keratinocyte SW900, human lung HD-11, chick myelomonocyte Broken cell systems Rat liver mitochondria Rat kidney mitochondria Rat, human, and minipig postmitochondrial supernatant Reconstituted system with cytochrome P450, ferridoxin, reductase Transfected cell systems CYP27A transfected into COS-1 cells CYP24 transfected into E. coli CYP24 transfected into insect cells CYP24 transfected into V79 Chinese hamster lung cell line b aOnly b79

on the enzymes responsible [23]. Metabolic studies utilize a variety of species and in vivo and in vitro models from broken cell to intact cell systems (Table III). The predominant models used for metabolic studies are liver-based systems and have revealed that this organ is a primary site of vitamin D analog metabolism. As with calcitriol, many vitamin D analogs can also be metabolized by vitamin D–target cells. The cloning of most if not all of the vitamin D–related cytochromes P450 has enabled us to generate cell-free and cell expression systems that allow us to study the CYPs in isolation. This has been achieved for CYP24, CYP27A, CYP2R1, and CYP27B [13–15]. These novel metabolic systems are valuable in that they allow the study of developing specific inhibitors for any of the known CYPs, or alternatively, of studying the substrate preferences of the target cell enzymes in order to design better catabolism-sensitive or catabolism-resistant vitamin D analogs for use as drugs. Methodology for studying vitamin D analog metabolism not only involves a wide variety of biological systems (Table III), but also usually depends upon the

a few examples are given. host cell line lacks endogenous CYP24 or CYP27A or CYP27B.

Enzyme activity

Ref.

25-Hydroxylation 1α- and 24-hydroxylation

[140] [141,142]

1α- and 24-hydroxylation 24-Oxidation pathway 24-Oxidation pathway (with and without CYP24)

[143] [25] [15]

24-Oxidation pathway 24-Oxidation pathway 24-Oxidation pathway 24-Oxidation pathway Extra-renal 1-hydroxylation Extra-renal 1-hydroxylation

[17] [18] [144] [19] [149] [35,36]

25-Hydroxylation 1α- and 24-hydroxylation General metabolism 1α- and 25-hydroxylation

[33] [145] [28] [129] [5]

25-Hydroxylation 24-Oxidation pathway 24-Oxidation pathway 24-Oxidation pathway

[13] [14] [15]

1429

CHAPTER 81 Analog Metabolism

availability of a suitably-radioactive vitamin D analog. This radioactive analog must possess a sufficiently-high specific activity to allow detection of nanomolar concentrations of analog in addition to a label location that is in a metabolically-resistant region of the molecule. This usually means that [3H]-labeling is the preferred isotope, and a nuclear-location of the radioactive tag is best if the label is not to be lost. Such is the case with [1β-3H] 1α,25-(OH)2D3 where the label is retained by the molecule even on truncation of the side chain to [1β-3H] calcitroic acid, which can be conveniently separated from other metabolites by a simple Bligh and Dyer extraction [24]. Occasionally, the susceptibility of the [3H]-label to metabolic attack is judiciously used as an indicator of the type of metabolism occurring. An excellent example of this is the ingenuity of Chugai chemists working with Slatopolsky’s group to compare the metabolism of both [26-3H]OCT and [2β-3H]OCT in their biological systems. Using such a combination of labels, these researchers were able to confirm the loss of a portion of the side chain of OCT during metabolism, thereby suggesting the formation of side-chain truncated metabolites of OCT [25,26]. Although the availability of the costly radioactive analog has not been a significant problem when studying most of the wellestablished compounds developed to date, it remains a significant barrier to the widespread screening of the hundreds of vitamin D analogs currently available. If radioactive analogs are not readily available, another technique involving detection of metabolites by diode-array spectrophotometry can be used, but at the disadvantage of being forced to employ high substrate concentrations [27]. Although this technique is restricted to high concentrations of analogs, the analysis is quicker and cheaper than using radioactive analogs, results in the generation of larger amounts of metabolites, and this, in turn, permits a more rigorous identification of the metabolic products. Our laboratory has used such a procedure very effectively over the past 10 years to identify the products of both natural and synthetic vitamin D compounds incubated with a range of in vitro biological systems. The quantitative and qualitatative answers that diode-array spectrophotometry provides are, in large part, consistent with those answers resulting from studies employing radioactive vitamin D analogs (cf. Sorensen et al. [28] and Masuda et al. [19]). Most recently, metabolic studies of vitamin D analogs have been revolutionized by the emergence of a new analytical technique for the detection of all metabolites based upon liquid chromatography/mass spectrometry (LC/MS) [15]. Though this technique lacks the exquisite sensitivity offered by radioisotopicallylabeled molecules, it is still one or two orders of magnitude more sensitive than conventional LC detectors (e.g. diode-array detector). Thus, LC/MS-based method

can detect as little as 10 pg of a vitamin D analog, whereas most LC detectors have a detection limit of 1 nanogram. Furthermore, LC/MS-based methods are quick and convenient, providing a selectivity unmatched by conventional LC detectors by incorporating spectral focusing routines (e.g. MRM—focus on transitions between major MS1 ions and their daughter ions), which dramatically reduce the background noise. As a consequence, LC/MS-based technologies are rapidly taking over as the method of choice in metabolic studies in all small molecule research, including vitamin D. Thus, while there are few current examples of the use of LC/MS with vitamin D, it can be safely assumed that in the future, LC/MS-based methods will become metabolic screening tools for all nonradioactive vitamin D analogs. The specific metabolic studies described below will provide examples of the use of all three of these approaches to the study of the metabolism of vitamin D analogs.

II. EXAMPLES OF THE METABOLISM OF ANALOGS OF VITAMIN D A. Dihydrotachysterol This example of a vitamin D prodrug represents the oldest vitamin D analog and was developed in the 1930s as a method of stabilizing the triene structure of one of the photoisomers of vitamin D. The structure of dihydrotachysterol2 shown in Table I contains an A-ring rotated through 180°, a reduced C10-19 double bond, and the side chain structure of ergosterol/ vitamin D2. This side chain is depicted because the clinically approved drug form of dihydrotachysterol is dihydrotachysterol2. However, it should be noted that dihydrotachysterol3 (DHT3) can also be synthesized with the side chain of vitamin D3. The metabolism of both dihydrotachysterol2 (DHT2) and dihydrotachysterol3 (DHT3) have been extensively studied over the past three decades [29–32]. Initial studies performed in the early 1970s showed that both DHT2 and DHT3 are efficiently converted to their 25-hydroxylated metabolites [33]. The effectiveness of DHT to relieve the hypocalcemia of chronic renal failure in the absence of a functional renal 1α-hydroxylase led to the hypothesis [34] that 25OH-DHT might represent the biologically-active form of DHT, by virtue of its 3β-hydroxy group being rotated 180° into a “pseudo 1α-hydroxyl position.” It was thus believed that 1α-hydroxylation of 25-OH-DHT was unnecessary. This viewpoint prevailed for at least a decade, but debate was renewed when Bosch et al. [30] were able to provide evidence for the existence of a mixture of 1α- and 1β-hydroxylated products of 25-OH-DHT2 in the blood of rats dosed

1430

GLENVILLE JONES

subject to 1-hydroxylation by an extra-renal source of CYP27B of bone marrow origin in vivo. Though the enzymes involved in the activation of DHT, especially the 1-hydroxylation step, have an altered specificity toward this molecule, the enzymes involved in the catabolism of DHT3 appear to treat the molecule as they would 25-OH-D3 or 1α,25-(OH)2D3. Side chain hydroxylated derivatives of both 25-OH-DHT3 and 1,25-(OH)2DHT3 have been identified and appear to be analogous to intermediates of the C-24 oxidation and 26,23-lactone pathways of vitamin D3 metabolism [37,38]. The major difference between the catabolism of DHT3 and vitamin D3 is that the 26,23-lactone formation from DHT3 appears exaggerated, suggesting that DHT3 is either a better substrate for the enzymes involved in 26,23-lactone formation, or else it is discriminated against by CYP24 or other enzymes of the alternative C-24 oxidation pathway.

with DHT2. Studies involving the perfusion of kidneys from vitamin D–deficient rats with an incubation medium containing 25-OH-DHT3 and using diode-array spectrophotometry to analyze the extracts showed this molecule to be subject to extensive metabolism by renal enzymes, but failed to give the expected 1-hydroxylated metabolites, opening up the possibility that the 1α- and 1β-hydroxylated metabolites observed by Bosch et al. might be formed by an extra-renal 1-hydroxylase [31] (see also Fig. 3). Following the synthesis of appropriate authentic standards, subsequent research [35] has rigorously confirmed the in vivo formation and identity of 1α,25-(OH)2DHT and 1β,25-(OH)2DHT in both rat and human. The ability of these 1α-hydroxylated forms of both DHT2 and DHT3 to stimulate a VDRE-inducible growth hormone reporter system exceeded that of 25-OH-DHT, and in the process established 1α,25(OH)2DHT and 1β,25-(OH)2DHT as the most potent derivatives of DHT identified to date. The formation of these metabolites also brings into question the importance of the “pseudo 1α-hydroxyl group” hypothesis, although current findings do not rule out that the biological activity of DHT might be due to the collective action of a group of metabolites including 25-OH-DHT, 1α,25-(OH)2DHT and 1β,25-(OH)2DHT. The latest information on the site of biosynthesis of 1-hydroxylated DHTs comes from studies using the cultured chicken myelomonocytic cell line, HD-11 [35]. This cell line, which has been documented previously as a rich source of the extra-renal 1α-hydroxylase [36], has also been shown to be capable of the 1-hydroxylation of 25-OH-DHT [35]. These results are consistent with DHT being 25-hydroxylated in the liver and then

0.3

B. Vitamin D2 Derivatives Though vitamin D2 can be synthesized naturally by irradiation of ergosterol, little finds its way into the human diet unless it is provided as a dietary supplement. Thus, one could make a case for considering vitamin D2 as a prodrug. The complex metabolism of vitamin D2 has been included as part of Chapter 2 and will not be repeated here. However, it is worth noting that vitamin D2 gives rise to several analogous metabolites to those of vitamin D3 in the form of 25-OH-D2 [39], 1α,25(OH)2D2 [40], and 24,25-(OH)2D2 [41], as well as several unique metabolites including 24-OH-D2 [42],

(AU)

(AU)

0.06

(AU)

0.04

0 200 250 DHT3

25OHDHT3

A

D E

H

300

K

350 0

5 Time 0–10 min

FIGURE 3

10

12

14

16

18

20

22

24

Time 10–33.04 min

26

28

30

32

Time 25–33.08 min

In vivo metabolism of dihydrotachysterol3 in the rat. Diode-array HPLC of the plasma extract of a rat administered 1 mg DHT3 18 hr prior to sacrifice. Metabolites are labeled 25-OH-DHT3 and peaks A–L. All possess the distinctive tricuspid UV spectrum (λmax 242.5 nm, 251nm, and 260.5 nm). Metabolites A–L were subsequently identified as side chain modified compounds analogous to vitamin D metabolites of the C-24 oxidation and 26,23-lactone pathways depicted in Figs. 1 and 2. (Reproduced from ref. 31 with permission.)

1431

CHAPTER 81 Analog Metabolism

1α,24S-(OH)2D2 [43], 24,26-(OH)2D2 [44], and 1α,25, 28-(OH)3D2 [45]. These differences in the metabolism of vitamin D2 have been exploited by pharmaceutical companies synthesizing and using the metabolites unaltered or else creating slightly modified versions (e.g. Roche compound 1α,25,28-(OH)3D2; Bone Care compound 1α,24S-(OH)2D2). Furthermore, features of the vitamin D2 side chain, namely the C22-23 double bond or the C-24 methyl group (see Table I), have been successfully incorporated into the structure of other analogs (e.g. calcipotriol). Even a prodrug based upon vitamin D2 has been designed in the form of 1α-OH-D2 [46]. This molecule is a valuable tool in studying hydroxylation reactions in the liver. At low substrate concentrations, 1α-OH-D2, like 1α-OH-D3, is 25-hydroxylated by liver hepatomas, Hep3B and HepG2, producing the well-established, biologically-active compound 1α,25-(OH)2D2. However, when the substrate concentration is increased to micromolar values, the principal site of hydroxylation of 1αOH-D2 becomes the C-24 position, the product being 1α,24S-(OH)2D2, another compound with significant biological activity [47,48]. This metabolite has been previously reported in cows receiving massive doses of vitamin D2 [43]. Transfection studies using the liver cytochrome P450, CYP27A, expressed in COS-1 cells suggest that 1α,24S-(OH)2D2 is a product of this cytochrome [5]. Whether the formation of this unique metabolic product of 1α-OH-D2 is the reason for the relative lower toxicity of 1α-OH-D2, as compared to 1α-OH-D3 [49], has not been established definitively. Active vitamin D2 compounds, such as 1α,25(OH)2D2, 19-nor-1α,25-(OH)2D2, and 1α,24S-(OH)2D2, are also subject to further metabolism, although it differs from that of calcitriol, essentially because on the face of it, the modifications in the vitamin D2 side chain prevent the C23-C24 cleavage observed during calcitroic acid production. Instead, the principal products are more polar tri- and tetra-hydroxylated metabolites such as: 1α,24,25-(OH)3D2, 1α,25,28-(OH)3D2, and 1α,25,26(OH)3D2 from 1α,25-(OH)2D2 [45,50,51]; 19-nor1α,24,25-(OH)3D2, 19-nor-1α,24,25,28-(OH)4D2, and 19-nor-1α,24,25,26-(OH)4D2 from 19-nor-1α,25(OH)2D2 [52]; and 1α,24,26-(OH)3D2 from 1α,24S(OH)2D2 [48]. These poly-hydroxylated metabolites are accompanied by side-chain cleaved products [52a], and there are even suggestions that calcitroic acid can be formed from 1α,25-(OH)2D2 [53]. This is not surprising when one considers that recent data using recombinant protein suggests that CYP24 is again responsible for these multiple hydroxylations of the D2 side chain [54,55]. However, at least in the case of 1α,24S(OH)2D2 metabolism by CYP24, the rate of metabolism appears slower than that of 1α,25-(OH)2D3 [48].

Some D2 catabolites retain considerable biological activity and at least one, 1α,25,28-(OH)3D2, is patented for use as a drug. See Chapter 2 for further discussion of vitamin D2 metabolism. It is interesting to note that two newer families of vitamin D compounds under development are based upon minor modifications of the vitamin D2 side chain. These are the D4 series, which retain the C24(C28)-methyl group but lack the C22
C23 double bond [56], and the D5 series, which possess a C24(C28,29)-ethyl group but lack the C22
C23 double bond [57].

C. Cyclopropane-ring Containing Analogs of Vitamin D These analogs are modified in their side chains such that C-26 is joined to C-27 to give a cyclopropane ring consisting of C-25, C-26, and C-27. The simplest member of this series is MC 969, which possesses the vitamin D3 side chain except for the presence of the cyclopropane ring together with a 1α-OH-D nucleus [58]. The best-known member of this group of compounds is MC 903 or calcipotriol, the structure of which is shown in Table II. In addition to the cyclopropane ring, calcipotriol features a C22
C23 double bond and a 24S-hydroxyl group [59]. As is presented in Chapter 101, calcipotriol was the first vitamin D analog to be approved for topical use in psoriasis, and is currently used worldwide for the successful control of this skin lesion [60]. When MC 969 is incubated with the hepatoma Hep3B, it is hydroxylated, not at the C-25 position as is 1α-OH-D3, but at the C-24 position, as is 1α-OH-D2, and then further oxidized to a 24-ketone [58]. 25-Hydroxylation of vitamin D analogs containing cyclopropane ring structures is feasible, and indeed the molecule has been synthesized chemically, but it is not produced enzymatically from MC969. It thus appears that the cyclopropane ring directs the hydroxylation site to the C-24 position. Of course, the other cyclopropane-ring containing analog, calcipotriol, contains a pre-existing C-24 hydroxyl, which has been proposed to act as a surrogate C-25 hydroxyl in interactions of the molecule with the VDR. Pharmacokinetic data acquired for calcipotriol showed that it had a very short t1/2, in the order of minutes; results that are consistent with the lack of a hypercalciuric/hypercalcemic effect when administered in vivo [61]. The first metabolism studies [28] revealed that calcipotriol was rapidly metabolized by a variety of different liver preparations from rat, minipig, and human to two novel products. Sorensen et al. [28] were able to isolate and identify the two principal products as a C22
C23 unsaturated, 24-ketone

1432

GLENVILLE JONES

enzymes, and then by vitamin D-related pathways to a side-chain cleaved molecule. The catabolites are produced in a variety of tissues and appear to have lower biological activity than the parent molecule. The reduction of the C22
C23 double bond during the earliest phase of calcipotriol catabolism was an unexpected event, given that the C22
C23 double bond in vitamin D2 compounds is extraordinarily stable to metabolism. It thus appears that metabolism of calcipotriol provides evidence that the C-24 methyl group in the vitamin D2 side chain must play a stabilizing role, preventing the formation of the 24-ketone, which facilitates the reduction of the C22
C23 double bond.

(MC1046) and a C22-C23 reduced, 24-ketone (MC1080). Coincidentally, this is the same product as is formed from MC969. These results were confirmed and extended by Masuda et al. [19], who showed that calcipotriol metabolism was not confined to liver tissue, but could be carried out by a variety of cells including those cells exposed to topically-administered calcipotriol in vivo, namely keratinocytes. Furthermore, Masuda et al. [19] proposed further metabolism of the 24-ketone in these vitamin D target cells to side-chain cleaved molecules including calcitroic acid (Fig. 4). The main implications of this work are that calcipotriol is subject to rapid metabolism initially by nonvitamin D–related

A

C 0.2 3 (MC 903)

Absorbance (265 nm)

1

R

OH

R MC 903

0.15 2

4 O

0.1

CH2 MC 1046 HO

OH

O

0.05 5

6

MC 1080

7 O

0 0

4

12

8

16

20

24

O

28

Time (min)

OH MC 1439 (23S)

B

OH MC 1441(23R)

OH

0.2

OH

Absorbance (265 nm)

4 OH MC 1577 (23R,24S)

0.15

O MC 1586 (24S)

1

0.1

OH

OH MC 1575 (23S,24S)

0.05 2

3

5

6 7 CH2OH

24,25,26,27Tetranor– 1α,23(OH)2D3

COOH

Calcitroic acid

0 0

4

8

12

16

Time (min)

20

24

28

FIGURE 4 In vitro metabolism of calcipotriol (MC903) by HPK1A-ras cells. HPLC of lipid extracts following incubation of MC903 with (A) HPK1A human keratinocytes and (B) HPK1A-ras human keratinocytes. Peak 1 = MC1080; Peak 2 = MC1046; Peak 3 = MC903 (calcipotriol); Peak 4 = mixture of MC1439 and MC1441; Peak 5 = Tetranor-1α,23(OH)2D3; Peak 6 = MC1577; Peak 7 = MC1575. (C) Proposed pathway of calcipotriol metabolism in cultured keratinocytes. (Reproduced from ref. 19 with permission.)

1433

CHAPTER 81 Analog Metabolism

However, it is still unknown which enzyme is responsible for this reduction in the side chain of calcipotriol.

D. 20-Epi- and 20-Methyl Analogs In the early 1990s, Leo organic chemists were first to change the stereochemistry of the side chain at the C-20 position [62,63]. As a result, they were in a position to synthesize a novel class of compounds, which were 20-epimers of existing analogs, the simplest being 20-epi-1α,25-(OH)2D3 and the most complex being KH1060 (see Table II for structures). Some of these 20-epi-analogs are extremely potent in cell differentiation and anti-proliferation assays, and are thus under development for use in hyperproliferative conditions. Another 20-epi analog 2-methylene-19nor-20S-1α,25-(OH)2D3 (known as 2MD) is reported to

B 3000

2000 MC1288 1000 1α,25(OH)2D3

0 0

1

10

100

Radioactivity bound to VDR (cpm)

Radioactivity in Aq. layer (cpm)

A

be highly bone-specific and is thus under development as an anti-osteoporosis drug [64; see Chapter 87]. These 20-epi-molecules have been particularly well studied, not only for their susceptibility to metabolism, but also for their ability to transactivate model genes (see Chapter 83). Dilworth et al. [65] showed that 20-epi-1,25-(OH)2D3 (MC 1288) has virtually no DBP binding, slightly improved affinity for the bovine thymus VDR, and an altered rate of metabolism through certain steps of the C-24 oxidation pathway. Dilworth et al. [65] concluded that all of these factors: DBP affinity, VDR affinity, and rate of catabolism in target cells contribute to the biological activity advantages that 20-epi-1α,25-(OH)2D3 appears to possess over 1α,25-(OH)2D3 in vitro (Fig. 5A–D). The influence of DBP (in FCS) on 20-epi-1,25-(OH)2D3 and 1,25(OH)2D3-induced-hGH reporter gene expression is illustrated in Fig. 5C and D. Though the curve for 20-epi-1,25-(OH)2D3 remains unchanged by the presence

1000

1200

1α,25(OH)2D3

600 MC1288

0

0

Amount of competitor (mg)

1000

[Ligand] (pg/mL)

C

D 250 1α,25(OH)2D3 (ED601.6 × 10−9)

MC1288 (ED505.3×10−11)

40

0 10−12

10−11

10−10

10−9

10−8

Induction of hGH production (ng/mL)

80 Induction of hGH production (ng/mL)

100

10

MC1288 (ED506.4 × 10−11)

125 1α,25(OH)2D3 (ED506.6 × 10−10)

0

10−12

10−11

10−10

10−9

10−8

[Analog] (M)

FIGURE 5 Biological parameters for 20-epi-1α,25-(OH)2D3. (A) Ability of 20-epi-1α,25-(OH)2D3 to compete for C24-oxidation pathway enzymes. (B) VDR affinity of 20-epi-1α,25-(OH)2D3 compared to 1α,25-(OH)2D3. (C and D): Effect of fetal calf serum (containing DBP) on reporter gene induction by 20-epi-1α,25-(OH)2D3 in the COS-1 cell line. (Reproduced from ref. 65 with permission.)

1434 of DBP, the curve for 1,25-(OH)2D3 is reduced by an order of magnitude, exaggerating the potency difference between the two molecules. Dilworth et al. [65] postulated that the remaining difference in transactivation activity is due in part to metabolic differences and in part to affinity differences at the transcriptional level (i.e., VDR; VDR-RXR; VDR-RXR-VDRE). However, it should be noted that the slower rate of metabolic inactivation of 20-epi-1α,25-(OH)2D3 has been challenged by some workers [66] and an alternative theory proposed. In this alternative hypothesis, it is proposed that 20-epi-1α,25-(OH)2D3 is metabolized at a similar rate to 1α,25-(OH)2D3 but that the metabolic intermediates, particularly the biologically active form 20-epi24-oxo-1α,25-(OH)2D3, are not broken down as efficiently and thus accumulate [66]. It is interesting to note that Peleg et al. [67] (also see Chapter 83) have extended this work to study the precise details of the gene transactivation events invoved in 20-epi-1α,25-(OH)2D3 action. They have found that two 20-epi-compounds, 20-epi-1α,25-(OH)2D3 and KH1060, are unusual in that they are able to form highly stable, protease-resistant RXR-VDR-VDRE heterodimeric complexes in vitro. Whether the stability of these transactivating complexes is related to the increased transactivation ability in model reporter genes is still to be established. However the work shows promise. Also interesting is the apparent correlation between the stability of 20-epi-1α,25-(OH)2D3-containing transactivation complexes (observed by Peleg et al. in ref 67) and the metabolic stability of 20-epi-1α,25-(OH)2D3 to target cell cytochrome P450 hydroxylation (observed by Dilworth et al. in ref 65). Whether this correlation holds for other potent, metabolically-stable vitamin D analogs has yet to be investigated. The 20-epi-compound KH1060 will be discussed later in this chapter under Section G. “Oxa-group Containing Analogs.” 20-Methyl vitamin D analogs are a group of biologically active compounds that have been synthesized mainly by the Schering company [118]. As with the 20-epi series, the presence of the 20-methyl group imposes a different conformation of the side chain and thus creates interesting problems for CYP24, as well as the VDR. Metabolic studies with 20-methyl-1, 25-(OH)2D3 and related compounds [68] suggest that the 20-methyl group makes 23-hydroxylation difficult and though 24-oxidation products can be detected by such techniques as LC/MS (Fig. 6; taken from M. Kaufmann and G. Jones, unpublished data), the 24-oxidation pathway intermediates build up in an analogous way to that proposed for 20-epi-1,25(OH)2D3 [66]. In fact, another 20-methyl-1,25-(OH)2D3 analog with a 23-ene group is resistant to metabolism and is even more potent than 20-methyl-1,25-(OH)2D3 itself [69].

GLENVILLE JONES

E. Homologated Analogs This modification involves the insertion of carbon atoms into the vitamin D side chain. It can take two different forms: a) insertion of the carbon atoms in the main side chain between carbons 22–25 such that additional carbons are numbered 24a, 24b, 24c, etc.; b) insertion of extra carbon atoms on the terminal methyl groups making them into dimethyl (ethyl) groups (likened to lengthening the “claws” on a crab). These carbons are 26a, 27a, etc. Homologated compounds were first developed in the early 1980s by Ikekawa and colleagues and initially tested for biological activity by DeLuca’s group [70–72]. Stern’s group showed that these analogs possess increased biological activity compared to 1α,25-(OH)2D3 when assayed in a cultured bone model in vitro [73]. Whether metabolism of these homologated analogs is different from that of 1α,25-(OH)2D3 or might play a role in the increased biological activity was initially unknown. The effect of lengthening the side chain poses interesting problems for the enzymes involved in hydroxylation of the side chain. The active site of the cytochrome P450 is forced to accommodate a longer side chain, which might alter the efficiency of hydroxylation and depending upon the way the side chain is anchored might change the site of hydroxylation (termed regioselectivity). Dilworth et al. [74] examined this systematically by studying the effect on metabolism of adding one, two, or three carbons to the main vitamin D side chain. Homologs synthesized by the Leo chemist, Martin Calverley, and used by Dilworth et al. [74] with the 25-hydroxyl group already in place included: MC 1127 (24a-homo-1α,25-(OH)2D3), MC 1147 (24a,24b-dihomo-1α,25-(OH)2D3), and MC 1179 (24a,24b,24c-trihomo-1α,25-(OH)2D3). Only one compound was synthesized without the 25-hydroxyl group, and this was the 1α-OH-D3 homolog with two extra carbon atoms: MC 1281 (24a,24b-dihomo-1α-OH-D3). The results obtained from metabolic studies using HPK1A-ras keratinocytes as a source of target cell 23- and 24-hydroxylase enzymes (presumably CYP24) and using hepatoma cells as a source of 25-hydroxylase (presumably CYP27A) suggested that both cytochrome P450 isoforms continued to efficiently hydroxylate all homologs provided. Somewhat surprisingly, CYP24 maintained its hydroxylation sites at C-23 and C-24 despite the extension of the side chain by up to three carbons and seemingly preferring not to

1435

CHAPTER 81 Analog Metabolism

20-methyl-24-oxo-1α,25-(OH)2D3 100

Relative abundance (%)

MW: 444 [M+H]+: 445 [M+Na]+: 467 [M+H−H2O]+: 427

467

H3C CH

3

O

CH3

MS1 H

OH

H CH2

0

427

OH

HO

444

543

575 m/z

60

80

100

120

100

140

160

180

200

220

240

260

300 320

280

340

360

380

400

420

440

460

480

500

520

540

560

580

600

135

Relative abundance (%)

H3C CH

3

CH3 H

CH2 117

60

80

100

120

140

160

180

fragments of m/z 427

−H2O

135

OH

HO 0

MS2

OH 153

H

O

200

220

240

260

280

300

320

340

360

380

400

420

440

100

Relative abundance (%)

12 h 8h MRM

4h

transition of m/z 427 to 135

0h

0 15

16

17

18

19

20

21

22

23

Retention Time (minutes)`

FIGURE 6 LC/MS of metabolites of 20-methyl-1α,25-(OH)2D3. Analysis of the purified metabolite by MS1 revealed several charactteristic ions which helped putatively identify the metabolite 20-methyl-24-oxo-1α,25-(OH)2D3 based upon molecular weight. In MS2 mode, fragmentation of the dehydration product (m/z 427) produced a specific cis-triene cleavage fragment (m/z 135). In MRM mode, the transition from m/z 427 to 135 formed a sensitive and selective means to quantitate the production of this metabolite in HPK1A-ras cells when incubated with 20-methyl-1α,25-(OH)2D3 over a time course. (Kaufmann M. and Jones G., unpublished results.)

1436 move down the side chain to C-24a, C-24b, or C-24c to be adjacent to the tertiary carbon, C-25. On the other hand, CYP27A hydroxylated MC1281 terminally at C-25 and C-27, appearing to ignore the longer internal side chain. These homologs therefore offer interesting insights into hydroxylation site selection by the vitamin D–specific cytochromes. Dilworth et al. [74] postulated that CYP24 must be directed to its hydroxylation site by the distance from the vitamin D ring structure, whereas CYP27 is directed by the distance of the hydroxylation site from the end of the side chain. Dilworth et al. [75] have also examined the effect of introducing the terminal 26,27-dimethyl groups into the side chain of 1α,25-(OH)2D3 to make the analog 26,27-dimethyl-1α,25-(OH)2D3 (MC1548). They found that MC1548 was metabolized at the same rate as 1α,25(OH)2D3 by the keratinocyte cell line HPK1A-ras. Products included metabolites of the C-24 oxidation pathway, and this was confirmed by the observation that MC1548 and 1α,25-(OH)2D3 were equally effective in blocking the metabolism of [1β-3H]1α,25-(OH)2D3 to [1β-3H] calcitroic acid. However, the products of MC1548 also included 26a-OH-MC1548, suggesting that the introduction of the dimethyl group into the side chain makes it susceptible to attack at a new terminal location by keratinocyte enzymes. Although MC1548 is the simplest molecule in the dimethyl homologated series, it represents a valuable tool in understanding the relative importance of various modifications within complex homologated molecules such as EB1089 and KH1060 (see also Chapter 84). These latter molecules have greatly increased biological activity over 1α,25-(OH)2D3 in cell-differentiating systems in vitro, and elucidating the importance of metabolism to this increased potency in vitro should provide useful insights into vitamin D analog action [76]. The metabolism of EB1089 and KH1060 will be discussed in Sections: F. “Unsaturated Analogs” and G. “Oxa-group Containing Analogs,” respectively.

F. Unsaturated Analogs The idea of introducing double bond(s) into the side chain of vitamin D analogs arose from experience with vitamin D2. Vitamin D2 metabolites have similar biological activity to those of vitamin D3 so that it appears that the introduction of the double bond is not deleterious. As mentioned already in Sections B. “Vitamin D2 Derivatives” and C. “Cyclopropane-ring Containing Analogs of Vitamin D” in this chapter, the metabolism of the side chain is significantly altered by this relatively minor change.

GLENVILLE JONES

The modification has not been confined to the introduction of a C22
C23 double bond. Roche has developed molecules with two novel modifications: a) introduction of a C16-C17 double bond and b) introduction of a C23-C24 triple bond that when combined produce the highly successful 16-ene, 23-yne analog of 1α,25-(OH)2D3 [77] (see Table II for structure). As alluded to earlier, Leo Pharmaceuticals has introduced the promising unsaturated analog EB1089 that contains a conjugated double-bond system at C22-C23 and C24-C24a, in addition to both main side chain and terminal dimethyl types of homologation [78] (see Table II for structure). These two series of Roche and Leo compounds have shown strong antiproliferative activity both in vitro and in vivo [77,79,80]. The metabolism of the 16-ene compound by the perfused rat kidney has been studied by Reddy et al. [81]. Reddy et al. [81] found that the introduction of the C16-C17 double bond reduces 23-hydroxylation of the molecule, and the implication is that the D-ring modification must alter the conformation of the side chain sufficiently to subtly change the site of hydroxylation by CYP24, the cytochrome P450 thought to be responsible for 23- and 24-hydroxylation. It is worth noting that Dilworth et al. [65] also noted the absence of measurable 23-hydroxylation of the analog 20-epi-1α,25(OH)2D3 in their studies, reinforcing the view that modifications around the C17-C20 bond profoundly influence the rate of 23-hydroxylation. The metabolism of the 16-ene, 23-yne analog of 1α,25-(OH)2D3 by WEHI-3 myeloid leukemic cells has been studied by Satchell and Norman [82]. Though one might predict that because this molecule is blocked in the C-23 and C-24 positions that it must be stable to C-24 oxidation pathway enzyme(s), it was found experimentally that the 16-ene,23-yne analog has the same t1/2 as 1α,25-(OH)2D3 when incubated with this cell line (6.8 hr). The main product of [25-14C]1α,25-(OH)2-16-ene-23-yne-D3 was not identified by these workers, but appeared to be more polar than the starting material. It will be interesting to see in any possible follow-up work if the metabolite that they have isolated has lost its C23-C24 triple bond or is simply further hydroxylated at some alternative site in its side chain (e.g. at C-26). Based upon knowledge emerging from other analogs, this reviewer favors the latter. Another unsaturated analog, which one might predict would be relatively metabolically stable, is EB1089 with its conjugated double-bond system. However, as pointed out earlier, EB1089 contains three structural modifications: the conjugated double-bond system is

CHAPTER 81 Analog Metabolism

accompanied by two types of side chain homologation. Nevertheless, as expected, the conjugated double-bond system dominates the metabolic fate of EB1089, there being no C-24 oxidation activity due to the blocking action of the conjugated diene system. When metabolism is studied with either in vitro liver cell systems or the cultured keratinocyte cell line, HPK1A-ras, disappearance of EB1089 is much slower than that of 1α,25(OH)2D3 [83]. Such data are consistent with the fairly long t1/2 in pharmacokinetic studies in vivo [84]. Since the conjugated system of EB1089 blocks C-24 oxidation reactions, it is not surprising that a different site in the molecule becomes the target for hydroxylation. Diode array spectrophotometry has allowed for the identification of the principal metabolic products of EB1089 as 26- and 26a-hydroxylated metabolites (Fig. 7) [85,86]. Note also that these metabolites of EB1089 have been chemically synthesized and shown to retain significant biological activity in cell differentiation and antiproliferative assays [86]. Again, it is interesting to note that with EB1089 and other molecules blocked in the C-23 and C-24 positions, such as 1α,24S-(OH)2D2 [48], the terminal carbons C-26 and C-26a become the alternative sites of further hydroxylation. However, it should also be noted

1437 that even in molecules not blocked in the C-23 and C-24 positions but containing the terminal 26- and 27-dimethyl homologation such as 26,27-dimethyl1α,25-(OH)2D3 (MC1548) [75], there seems to be significant terminal 26a-hydroxylation occurring. Therefore, the hydroxylation of EB1089 at C-26 and C-26a may be in part a consequence of the introduction of the conjugated double-bond system and in part a consequence of the introduction of the terminal homologation. When the C22
C23 double bond is present in the side chain in the absence of a C-24 methyl group, as in calcipotriol, the double bond appears vulnerable to reduction. As pointed out earlier, the principal metabolites of calcipotriol are reduced in the C22
C23 bond except for one, the C22 -C23 unsaturated, 24-ketone (MC1046) [19,28]. This suggests a C-24 ketone must be present to allow for this reduction to occur. Work of Wankadiya et al. [87] using the Roche compound, 22∆-1α,25-(OH) D , an analog that contains the C22
C23 2 3 double bond but lacks a C-24 subsituent, tends to indirectly support this theory. When incubated with the chronic myelogenous leukemic cell line, RWLeu-4, this molecule, like 1α,25-(OH)2D3, is converted, presumably via 22∆-1α,24,25-(OH)3D3 and 22∆-24-oxo-1α,25(OH)2D3, metabolites analogous to intermediates in the

FIGURE 7 In vitro metabolism of EB1089 by HPK1A-ras cells. Diode array HPLC of lipid extracts following incubation of EB1089 (10 µM) with the human keratinocyte, HPK1A-ras for 72 hr. In addition to the substrate at 8.5 min, two metabolites showing the characteristic UV chromophore of EB1089 (λmax 235 nm, shoulder 265nm) are visible in the part of the HPLC profile reproduced here (8–20 min). Metabolite peaks at 15.03 min and 16.55 min were isolated by extensive HPLC and identified [68,69,71] by comparison to synthetic standards on HPLC, GC-MS, and NMR. The identifications are: Peak A at 15.03 min = 26-OH-EB1089; Peak B at 16.55 min = 26a-OH-EB1089. (Reproduced from Shankar et al. [85] with permission.)

1438 C-24 oxidation pathway, to the side chain truncated product 24,25,26,27-tetranor-1,23-(OH)2D3, a molecule which lacks the C22
C23 double bond. However, Wankadiya et al. [87] did not identify intermediates in this process, and thus do not know at which stage C22
C23 reduction occurred.

G. Oxa-group Containing Analogs These compounds involve the replacement of a carbon atom (usually in the side chain) with an oxygen atom. The best known of these are the 22-oxa analogs including 22-oxa-calcitriol (OCT) [88] and KH1060 [89]. Lesser known analogs include 23-oxa-series [90–92, 97] and 24-oxa-1α-hydroxyvitamin D3 [93,94]. All of these molecules are metabolically fascinating to study because the oxa-atom makes the molecule inherently unstable should it be hydroxylated at the adjacent carbon atom. The hydroxylation at an adjacent carbon generates an unstable hemi-acetal, which spontaneously breaks down to eliminate the carbons distal to the oxa-group. In the case of: a) the 22-oxa compounds the expected product(s) would be C-20 alcohol/ ketone; b) the 23-oxa compounds the expected product(s) would be C-22 alcohol/ketone; and c) the 24-oxa compounds the expected product(s) would be C-23 alcohol/ ketone/acid. The metabolism of OCT has been extensively studied in a number of different biological systems, including primary parathyroid [25] and keratinocyte cells [95], as well as cultured osteosarcoma, hepatoma, and keratinocyte cell lines [26] (see also Chapter 86). In all these systems, OCT is rapidly broken down. As outlined earlier, the use of two different radioactive labels in [26-3H]OCT and [2β-3H]OCT enabled Brown et al. [25] to suggest that the side chain was truncated, though definitive proof of the identity of the products was not immediately forthcoming. It was not until the work of Masuda et al. [26] that the principal metabolites were unequivocably identified by GC-MS as 24-OH-OCT, 26-OH-OCT, and hexanor-1α,20-dihydroxyvitamin D3. In the case of the keratinocyte HPK1A-ras, an additional product, hexanor-20-oxo-1α-hydroxyvitamin D3, is formed. These latter two truncated products are suggestive of hydroxylation of OCT at the C-23 position to give the theoretical unstable intermediate postulated at the beginning of the studies. Though all of these products were isolated from in vitro systems, there is evidence that the processes also occur in vivo because Kobayashi et al. [96] have generated data that suggests that the biliary excretory form of OCT in the rat is a glucuronide ester of the truncated 20-alcohol. The 23-oxa derivative KH1650 (see Table II) is broken down rapidly to the expected 22-alcohol [97].

GLENVILLE JONES

The metabolism of 24-oxa-1α-hydroxyvitamin D3 (MC1090) has been studied at micromolar concentrations using the hepatoma, Hep3B [93]. As expected, 24-oxa-1α-OH-D3 was found to be converted in high yield to two truncated products: tetranor-1,23-(OH)2D3 and calcitroic acid; again suggesting hydroxylation at the C-25 position adjacent to the 24-oxa group which results in an unstable intermediate. The products were again identified by GC-MS. Both the above examples of simple oxa-analogs provide useful knowledge which can help in predicting the metabolic fate of a complex oxa-analog such as KH1060. This highly potent compound which possesses in vitro cell-differentiating activity exceeding that of any other analog synthesized to date, has four different modifications to the side chain of 1α,25-(OH)2D3 namely: 1) a 22-oxa group; 2) the 20-epi side chain stereochemistry; 3) 24a-homologation; and 4) 26- and 27-dimethyl homologation (see Table II for structure). Since all of these changes are known to affect biological activity in vitro and in vivo as well as side chain metabolism, it comes as no surprise that the metabolism of KH1060 is extremely complex. KH1060 has a very short t1/2 in pharmacokinetic studies in vivo [84], giving a metabolic profile with at least 16 unknown metabolites. Recently, Dilworth et al. [98] reported the first in vitro study using micromolar concentrations of KH1060 incubated with the keratinocyte cell line HPK1A-ras. Dilworth et al. [98] were able to discern 22 different metabolites after multiple HPLC steps and assigned structures to 12 of these metabolites (see Fig. 8). As would be expected from consideration of the studies of other oxa-compounds, two of these were truncated products identical to the molecules formed from another 22-oxa compound, OCT. As would be expected from consideration of the studies of other homologated compounds (see Section E. “Homologated Analogs”), other products are hydroxylated at specific carbons of the side chain including C-26 and C-26a. As with EB1089 and MC1548, the presence of dimethyl groups in the terminus of the side chain appears to attract hydroxylation to this site in KH1060. One novel metabolite found only for KH1060 is 24a-OH-KH1060, observed both in broken cell and intact cell models [99,100]. One important facet of this complex metabolic profile is that rather than simplifying our understanding of the mechanism of action of KH1060, these data complicate it. This is because biological assays of each of the metabolic products have shown that several of the principal and long-lived metabolic products of KH1060 retain significant vitamin D–dependent gene-inducing activity in reporter gene expression systems [100]. This point will be discussed further under Section III.B. Implications for Mechanism of Action of Vitamin D Analogs.

1439

CHAPTER 81 Analog Metabolism

FIGURE 8 In vitro metabolism of KH1060 by HPK1A-ras cells. Diode array HPLC of lipid extracts following incubation of KH1060 (10 µM) with the human keratinocyte, HPK1A-ras for 72 hr. Nine peaks showing the characteristic UV chromophore of vitamin D (λmax 265 nm, λmin 228 nm) are visible in the part of the HPLC profile reproduced here (11–25 min). Rechromatography of these peaks on a second HPLC system resulted in the further resolution of these 9 peaks into 22 separate metabolites. Many of these metabolites were identified (ref. 82) by comparison to synthetic standards on HPLC and GC-MS. Identifications include: Peak at 13.39 min = 24a-OH-KH1060; Peak at 22.14 min = 26-OH-KH1060. (Reproduced from Dilworth et al. ref. 98 with permission.)

H. Other Analogs There are several other analogs synthesized to date that do not fit readily into the groups discussed thus far and yet have been studied metabolically. These are listed briefly below. 1. 1α,24R-(OH)2D3

This compound is marketed as an antipsoriatic drug by Teijin, and its structure is shown in Table II. The metabolism of [1β-3H]1α,24R-(OH)2D3 by the isolated perfused rat kidney was described by Reddy et al. [101]. Interestingly, Reddy et al. found two distinct pathways of metabolism, a major one involving direct C-24 oxidation (24-oxidation, 23-hydroxylation, side chain cleavage), resulting in calcitroic acid formation, and a second minor pathway, involving first 25-hydroxylation to 1α,24R,25-(OH)3D3, followed by the same C-24 oxidation steps occurring in the catabolism of 1α,25-(OH)2D3 (see Fig. 1). Since the authors used both pharmacological as well as “physiological” concentrations of the analog, they were able to generate sufficient quantities in order to identify their products by mass spectrometry. Their finding of 25-hydroxylase activity

in the kidney is not a complete surprise since others [102] have detected the enzyme activity there, and CYP27A mRNA is detected in rat kidney tissue [103]. However, a follow-up study by the same researchers [104] suggests that the renal 25-hydroxylase which is able to hydroxylate 1α,24R-(OH)2D3 is unable to 25-hydroxylate the compounds vitamin D3 and 1α-OH-D3. The implication of this work is that 1α,24R-(OH)2D3 is metabolized by a nonspecific enzyme to intermediates of the C-24 oxidation pathway of 1α,25-(OH)2D3. If this is the case in vivo, one might anticipate the potential for the drug 1α,24R-(OH)2D3 to interfere with normal 1α,25-(OH)2D3 catabolism. 2. ED-71

The structure of this compound is also shown in Table II. It is unusual in that it represents one of the few A-ring modified analogs whose metabolism has been studied to date (see Chapter 86). It possesses a unique 2β-hydroxypropoxy group, in addition to the usual 1α-, 3β- and 25-hydroxy groups of 1α,25-(OH)2D3. The extra bulky group at the 2β-position has the effects of improving the affinity of ED-71 for the plasmabinding protein, DBP, but as a consequence may make

1440 it more difficult for the analog to enter target cells. There is some optimism that these unique properties may make ED-71, a “long-lived vitamin D” and therefore show the necessary properties to be suitable as an antiosteoporosis drug [105]. Since ED-71 has the normal side chain of 1α,25-(OH)2D3, it should be susceptible to the same C-24 oxidation sequence and other pathways as the natural hormone. Indeed, initial studies have provided evidence for the formation of several of the same 24- and 26-hydroxylated and 24-oxidized products as 1α,25-(OH)2D3, but at a much reduced rate [106]. There is some possibility for metabolism of 2β-group by nonspecific enzymes, though Masuda et al. [106] failed to observe such catabolites. Additional consideration of this compound can be found in Chapter 86. 3. 26,26,26,27,27,27-HEXAFLUORO-1α,25-(OH)2D3 (26,27-F6-1α,25-(OH)2D3)

This analog was first synthesized in the early 1980s [107], along with a number of other side chain fluorinated analogs, to test the importance of certain key hydroxylation sites (e.g. C-23, C-24, C-25, C-26(27), C-1) to biological activity. It was noted immediately that 26,27-F6-1α,25-(OH)2D3 was extremely potent (10-fold higher than 1α,25-(OH)2D3) in calcemia assays both in vitro and in vivo [108–110]. Lohnes and Jones [111] presented evidence using a bone cell line, UMR106, that 26,27-F6-1α,25-(OH)2D3 had a longer t1/2 inside target cells due to the apparent lack of 24-hydroxylation of 26,27-F6-1α,25-(OH)2D3. At around the same time, Morii’s group noted the appearance of a metabolite of 26,27-F6-1α,25-(OH)2D3, which they have identified as 26,27-F6-1α,23,25-(OH)3D3 [112]. This compound has excellent calcemic activity in its own right, but whether this derivative is in part responsible for the biological activity of 26,27-F6-1α,25(OH)2D3 is not conclusively proven. Nonetheless, 26,27-F6-1α,25-(OH)2D3 has undergone clinical trials for hypocalcemia associated with hypoparathyroidism and uremia [113,114]. 4. OTHER ANALOGS

As further generations of analogs emerge, they will likely be studied metabolically. Each major new modification must be tested for its impact upon the metabolic machinery, as well as on biological activity. There are some compounds listed in Table I that are currently untested or for which only preliminary results have been published. These include the aromatic analog KH1650 [97], A-ring modified analogs such as 2-methyl-1α,25-(OH)2D3 and 2-methylene19-nor-20-epi-1α, 25-(OH)2D3 [64], as well as the side chain modified analogs 1α-OH-D5 [57] and the sulphone family [115].

GLENVILLE JONES

III. IMPORTANT IMPLICATIONS DERIVED FROM ANALOG METABOLISM STUDIES A. Correlations with Pharmacokinetic Information The susceptibility of a vitamin D analog to metabolism and excretion undoubtedly plays a significant role in determining the biological activity of that analog in vivo. The quickest and easiest way to acquire such knowledge is by pharmacokinetic analysis. From a classical vitamin D outlook that this author prefers to take, pharmacokinetic data reflect a few important parameters regarding each analog including: a) The affinity of the vitamin D analog for DBP in the bloodstream. b) The rate of target cell uptake and metabolism by target cell enzymes. c) The rate of liver cell uptake, hepatic metabolism, and biliary clearance. d) The rate of storage depot uptake and release. Metabolism, whether target cell or liver, is reflected in only two components of this list of factors measured by pharmacokinetics. Therefore, it cannot be expected that in vitro metabolic parameters would exactly correlate with in vivo pharmacokinetic parameters. Nevertheless, a comparison of the two might be worthwhile. In the case of some of the analogs shown in Tables I and II, pharmacokinetic data [84,119,120] are available and can be compared to the data provided by in vitro metabolic studies. In Table IV an attempt is made to compare pharmacokinetic data from these sources with DBP binding data and target cell metabolic data. As pointed out by Kissmeyer et al. [84], compounds segregate into at least two categories (perhaps more) on the basis of their pharmacokinetic parameters. Calcemic Analogs (Strong or Weak) Those analogs with a long t1/2 which is either a function of strong DBP binding or a reduced rate of metabolism (or both). The analog ED-71 has a strong DBP binding affinity. There appear to be a group of analogs in which a long t1/2 is correlated with a slower rate of metabolism (e.g. EB1089, ED-71, 1α,24S-(OH)2D2). With the exception of ED-71, most of these active analogs bind DBP poorly. Noncalcemic Analogs Those analogs with a short t1/2, which is either a function of poor DBP binding or a rapid rate of metabolism (or both). Examples include calcipotriol, KH1060, and OCT.

1441

CHAPTER 81 Analog Metabolism

TABLE IV

Compounda 25OHD3 1α,25(OH)2D3 1β,25(OH)2D3 Calcipotriol MC 1127 EB 1089 CB 966 KH 1139 KH 1060 KH 1049

Compoundd 1α,25(OH)2D3 OCT

Compounda 1α,25(OH)2D3 1α,25(OH)2D2 1α,24(OH)2D2

Serum concentration at t = 5 min (ng/ml) 2040 2429 2912 121 545 152 176 154 103 104

t1/2 (hr) > 2.8 2.2 >4 0.2 1.6 2.1 1.8 0.7 0.4 0.5

t1/2 (hr) 7.0 2.5

Pharmacokinetic Data on Vitamin D Analogs

AUC∞ (ng/ml × hr) 9596 7355 13,228 27 1216 255 267 142 46 40

Serum clearance (ml/hr/kg)

Binding affinity for DBP (M)

Relative binding affinity for DBPb

Rate of metabolismc

21 27 15 7407 167 784 693 1408 4348 5000

9 × 10−9 1.5–6.0 × 10−7 1.7 × 10−8 1.7 × 10−6 5.2 × 10−6 7.9 × 10−6 3.2 × 10−5 6.5 × 10−5 n.b. n.b.

33 1 17 0.1 0.1 0.03 0.02 0.007 0 0

Very slow Fast – Very fast Fast Slow – – Fast –

Metabolic clearance rate (ml/min)

Relative binding affinity for DBPe

Rate of metabolismf

1 266

Fast Very fast

5.0 48.2

Baseline (pg/ml)

AUC∞ t1/2 (hr)

67.0 <10 <10

5.8 5.1 4.9

Relative binding (pg/ml × hr) 3690 2676 659

Rate of affinity for DBPh 1 – 14

Rate of metabolismi Fast – Slow

was measured following a single intravenous dose of 200 µg/kg to rats. From Kissmeyer et al. [84]. human DBP, the relative numbers are compared to the 1α,25(OH)2D3 value. n.b., No binding. From Kissmeyer et al. [84]. c From Masuda et al. [19], Dilworth et al. [65, 98], and Shankar et al. [85]. d Following a single intravenous dose of 100 ng/animal to dogs. From Dusso et al. [120]. e Using rat DBP, the relative numbers are compared to the 1α,25(OH) D value. From Dusso et al. [120]. 2 3 f From Masuda et al. [26]. g Following a single oral dose of 0.39 ug/kg to rats. From Knutson et al. [146]. h Using rat DBP. From Strugnell et al. [47]. i From Jones et al. [48]. a Activity b Using

It should be noted that though these classifications are used in the vitamin D literature, they are somewhat artificial since pure “noncalcemic” analogs do not yet exist. All “noncalcemic” analogs will cause hypercalcemia if their concentration is raised sufficiently. The crucial issue is whether systemically administered, “weakly calcemic” or “noncalcemic” analogs can produce their anticell proliferation/pro-cell differentiation effects in vivo at concentrations lower than that required to produce calcemia. Various in vivo clinical trials currently in progress will be the acid test for this question.

B. Implications for Mechanism of Action of Vitamin D Analogs There is currently tremendous interest in defining the mechanism of action of vitamin D analogs, particularly for clarifying the difference between “calcemic” or “noncalcemic” analogs. Chapters 82 and 83 will discuss other aspects of analog action in detail. It is obvious from the amount of space committed to this mechanism that the problem is not simple but multifactorial. Therefore, the next section of this chapter

1442 will focus on the importance of metabolism to the complex picture. Metabolism can have an impact on the mode of action in a few different ways: 1) Lack of high affinity binding to DBP in the blood can make the analog vulnerable to liver enzymes, which may lead to deactivation and excretion of the analog. 2) Target cell enzymes may activate or deactivate the administered analog: i) to metabolites which possess increased, equivalent, or slightly decreased biological activity at target genes, and which may have an extended t1/2 inside the cell. ii) to metabolites which possess much reduced biological activity. 3) The rate of metabolism by target cell enzymes may be influenced by the rate of entry of “free” analog from the cell exterior and the association and dissociation rates of VDR-RXR-DNA complexes [121]. Chapter 82 will stress the importance of DBP binding and pharmacokinetics. It is clear that point 1) hepatic metabolism above relates mainly to this chapter. Chapter 83 will focus on the mechanisms by which analogs interact with VDR-RXR heterodimeric complexes and ultimately attract coactivators into the transcriptional machinery associated with the vitamin D– dependent gene in order to regulate gene expression. Points 2) and 3) target cell metabolism relate mainly to Chapters 82 and 83. It should be noted with regard to molecular mechanisms of action at the target cell level that metabolism is often disregarded or given too little emphasis. Furthermore, certain metabolic assumptions are made when testing biological activity that are not always valid. These include: i) the analog is biologically active as administered. ii) the analog is stable in the in vitro target cell model used, whether in vitro organ culture, cultured target cell, or host cell/reporter gene construct. The validity of this approach is made even more tenuous when data acquired with different in vitro models where metabolic considerations may or may not apply are compared to data acquired in vivo where metabolic considerations definitely apply. The reader is cautioned that invalid comparisons of in vivo and in vitro data abound in this field. In the opinion of this reviewer, metabolism will turn out to be a key parameter, but not the only important parameter in vitamin D analog action. It is our view that only when we consider all of the parameters which

GLENVILLE JONES

can influence analog action within the overall equation will we be in a position to fully understand the molecular mechanisms underlying their “noncalcemic” or “calcemic” actions. From this it seems unlikely that there will be two sets of such parameters providing perfect “noncalcemic” and “calcemic” analogs, as is presented in Section III.A, but rather several different permutations of the same parameters giving rise to analogs with slightly different applications.

C. Implications for Future Drug Design A case for the importance of metabolism within vitamin D analog action has been presented throughout this chapter. Much has been learned about those modifications to the vitamin D molecule, which change metabolism but in the process also improve biological activity. For some applications of vitamin D, this involves the concept of making “metabolism-resistant analogs” (e.g. those blocked in the C-23, C-24, C-26, or C-27 positions), which possess enhanced calcemic activity. For other applications of vitamin D, this involves the concept of making “metabolism-sensitive analogs” (e.g. those with oxa-groups at key side chain positions or a C22
C23 juxtaposed with a 24-hydroxyl) to localize the biological effect to the site of analog administration. Over the immediate future we can anticipate: 1) A search for additional novel synthetic modifications to the vitamin D side chain and ring structure; 2) Continuation of the trend to combine proven modifications in order to synthesize newer generations of so-called “hybrid analogs”; 3) Synthesis of “smart” molecules where metabolic and structure activity information gained from earlier generations of molecules is used to improve existing analogs. 4) A search for novel chemical entities included “nonsteroidal” vitamin D analogs that mimic certain actions of the vitamin D molecule (see Chapter 88). 5) Development of CYP24 inhibitors which mimic the action of 1,25-(OH)2D3 by blocking its catabolism [15,122]. One can envision that the VDR binding pocket studies [123,124] and cytochrome P450-substrate binding pocket studies [128] will provide particularly valuable information for the design of further generations of vitamin D analogs. The reader is referred to Chapter 15 for the latest information on VDR structure and modeling. The final section of this chapter outlines our

1443

CHAPTER 81 Analog Metabolism

progress in the area of modeling of vitamin D relatedcytochrome P450 isoforms.

D. Cytochrome P450 Isoform Modeling Studies The cytochrome P450 superfamily constitute a group of over 100 proteins subdivided into microsomal and mitochondrial isoforms which are responsible for the metabolism (e.g. hydroxylation) of endogenous and exogenous (xenobiotic) compounds [125]. Their structure is well conserved across the superfamily with domains for heme-binding, ferredoxin-binding, O2-binding, and substrate binding. These proteins are membrane-associated and are thus not easily studied by X-ray crystallographic means. For several mammalian steroidal cytochrome P450 isoforms (e.g. aromatase, cholesterol side-chain cleavage enzyme, 17-hydroxylase, rat 2B1, human 2D6) modeling studies (e.g. 126,127) have begun based upon information derived from crystal structures of soluble prokaryotic cytochromes P450 (CAM, BM-3, TERP, EryF). Thus, for the mammalian cytochromes P450, this work is in its infancy. Nevertheless, the approach appears highly promising. Such models, despite being crude first approximations, allow for identification of putative active-site residues suitable for site-directed mutagenesis studies. Refinements of the model derived from mutant proteins then follow. In the case of CYP27A [128], CYP27B, and CYP24 [129] such an approach can now be undertaken not only using the information derived from the primary amino-acid sequence [130] and other modeled cytochrome P450s, but also using the information derived from natural mutations of CYP27A and CYP27B resulting in the human diseases, cerebrotendinous xanthomatosis and vitamin D–dependency rickets [131,132] (see Chapter 71). Note no human disease has currently been recognized involving mutations of CYP24. Modeling of the CYP proteins, substrate preferences (e.g. Ref 5) and mutated analyses of the CYPS should permit a much clearer picture of the analog/cytochrome P450 interactions occurring in vivo. It is expected that knowledge generated using this approach will eventually result in more rational vitamin D analog design in the future [133].

Acknowledgements The author wishes to thank F. Jeffrey Dilworth and David Prosser in compilation of the references, tables, and figures contained within this document. Some of the work cited here is supported through grants to the

author from the Canadian Institutes of Health Research/ Medical Research Council of Canada. The collaborative research reviewed here required the input of some excellent research trainees: Drs. David Lohnes, Fuad Qaw, Stephen Strugnell, Sonoko Masuda, and David Prosser, as well as the interdisciplinary involvement of talented scientists from around the world. I gratefully acknowledge the essential contributions to our analog work of Hugh LJ Makin, Martin Calverley, Joyce Knutson, Charles Bishop, Noboru Kubodera, AnneMarie Kissmeyer, Richard Kremer, Mark R. Haussler, and Hector F. DeLuca.

References 1. DeLuca HF 1988 The vitamin D story: A collaborative effort of basic science and clinical medicine. FASEB J 2:224–236. 2. Bouillon R, Okamura WH, Norman AW 1995 Structurefunction relationships in the vitamin D endocrine system. Endocr Rev 16:200–257. 3. Calverley MJ, Jones G 1992 Vitamin D. In: RT Blickenstaff (ed) Antitumor Steroids. Academic Press: San Diego, pp. 193–270. 4. Okuda KI, Usui E, Ohyama Y 1995 Recent progress in enzymology and molecular biology of enzymes involved in vitamin D metabolism. J Lipid Res 36:1641–1652. 5. Guo Y-D, Strugnell S, Back DW, Jones G 1993 Transfected human liver cytochrome P-450 hydroxylates vitamin D analogs at different side-chain positions. Proc Natl Acad Sci USA 90:8668–8672. 6. Jones G, Strugnell S, DeLuca HF 1998 Current understanding of the molecular actions of vitamin D. Physiological Reviews 78:1193–1231. 6a. Cheng JB, Motola DL, Mangelsdorf DJ, Russell DW 2003. De-orphanization of cytochrome P450 2R1: a microsomal vitamin D 25-hydroxylase. J Biol Chem 278:38084–38093. 7. Adams JS, Gacad MA 1985 Characterization of 1α-hydroxylation of vitamin D3 sterols by cultured alveolar macrophages from patients with sarcoidosis. J Exp Med 161:755–765. 8. Ohyama Y, Noshiro M, Okuda K 1991 Cloning and expression of cDNA encoding 25-hydroxyvitamin D3 24-hydroxylase. FEBS Lett 278:195–198. 9. Makin G, Lohnes D, Byford V, Ray R, Jones G 1989 Target cell metabolism of 1,25-dihydroxyvitamin D3 to calcitroic acid. Evidence for a pathway in kidney and bone involving 24-oxidation. Biochem J 262:173–180. 10. Reddy GS, Tserng K-Y 1989 Calcitroic acid, end product of renal metabolism of 1,25-dihydroxyvitamin D3 through C-24 oxidation pathway. Biochemistry 28:1763–1769. 11. St-Arnaud R, Arabian A, Travers R, Barletta F, Raval-Pandya M, Chapin K, Depovere J, Mathieu C, Christakos S, Demay MB, Glorieux FH 2000 Deficient mineralization of intramembranous bone in vitamin D-24-hydroxylase-ablated mice is due to elevated 1,25-dihydroxyvitamin D and not to the absence of 24,25-dihydroxyvitamin D. Endocrinology 141:2658–2666. 12. Jones G, Byford V, Arabian A, St-Arnaud R 2000. Altered Pharmacokinetics of 1α,25-(OH)2D3 in Blood and Tissues of the CYP24-null Mouse. J Bone Miner Res 15: abstract 1246, S199.

1444 13. Akiyoshi-Shibata M, Sakaki T, Ohyama Y, Noshiro M, Okuda K, Yabusaki Y 1994 Further oxidation of hydroxycalcidiol by calcidiol 24-hydroxylase—A study with the mature enzyme expressed in Escherichia coli. Eur J Biochem 224:335–343. 14. Beckman M, Prahl JM, DeLuca HF 1995 Sidechain metabolism of [26,27-3H]-25-hydroxycholecalciferol (25-OH-D3) by recombinant human 24-hydroxylase (p450cc24) expressed in baculovirus infected Spodoptera frugiperda (SF21) insect cells. J Bone Miner Res 10:S395 (Abstract M573). 15. Masuda S, Gao M, Zhang A, Kaufmann, Jones G 2003 Importance of Cytochrome P450-mediated metabolism in the mechanism of action of vitamin D analogs. In: J Reichrath, M Friedrich, W Tilgen (eds) Vitamin D Analogs in Cancer Prevention and Therapy. Recent Results in Cancer Research Volume 164. Springer-Verlag: Germany, pp. 189–202. 16. Shinki T, Jin CH, Nishimuar A, Nagai Y, Ohyama Y, Noshiro M, Okuda K, Suda T 1992 Parathyroid hormone inhibits 25-hydroxyvitamin D3-24-hydroxylase mRNA expression stimulated by 1α,25-dihydroxyvitamin D3 in rat kidney but not in intestine. J Biol Chem 267:13757–13762. 17. Tomon M, Tenenhouse HS, Jones G 1990 Expression of 25-hydroxyvitamin D3-24-hydroxylase activity in CaCo-2 cells. An in vitro model of intestinal vitamin D catabolism. Endocrinology 126:2868–2875. 18. Lohnes D, Jones G 1987 Side-chain metabolism of vitamin D in osteosarcoma cell line UMR-106: Characterization of products. J Biol Chem 262:14394–14401. 19. Masuda S, Strugnell S, Calverley MJ, Makin HLJ, Kremer R, Jones G 1994 In vitro metabolism of the antipsoriatic vitamin D analog, calcipotriol, in two cultured human keratinocyte models. J Biol Chem 269:4794–4803. 20. Yamada S, Nakayama K, Takayama H, Shinki T, Takasaki Y, Suda T 1984 Isolation, identification, and metabolism of (23S,25R)-25-hydroxyvitamin D3-26,23-lactol: A biosynthetic precursor of (23S,25R)-25-hydroxyvitamin D3-26, 23-lactone. J Biol Chem 259:884–889. 21. Sakaki T, Sawada N, Komai K, Shiozawa S, Yamada S, Yamamoto K, Ohyama Y, Inouye K 2000. Dual metabolic pathway of 25-hydroxyvitamin D3 catalyzed by human CYP24. Eur J Biochem 267:6158–6165 22. Horst RL, Wovkulich PM, Baggiolini EG, Uskokovic MR, Engstrom GW, Napoli JL 1984 (23S)-l,23,35-Trihydroxyvitamin D3: Its biological activity and role in 1α,25-dihydroxyvitamin D3-26,23-lactone biosynthesis. Biochemistry 23: 3973–3976. 23. Jones G, Lohnes D, Strugnell S, Guo Y-D, Masuda S, Byford V, Makin HLJ, Calverley MJ 1994 Target Cell Metabolism of Vitamin D and Its Analogs. In: AW Norman, R Bouillon, M Thomasset (eds) Vitamin D. A Pluripotent Steroid Hormone: Structural Studies, Molecular Endocrinology and Clinical Applications, de Gruyter: Berlin, pp. 161–169. 24. Bligh EG, Dyer WJ 1957 A rapid method of total lipid extraction and purification. Can J Biochem 37:911–917. 25. Brown AJ, Berkoben M, Ritter C, Kubodera N, Nishii Y, Slatopolsky E 1992 Metabolism of 22-oxacalcitriol by a vitamin D–inducible pathway in cultured parathyroid cells. Biochem Biophys Res Commun 189:759–764. 26. Masuda S, Byford V, Kremer R, Makin HLJ, Kubodera N, Nishii Y, Okazaki A, Okano T, Kobayashi T, Jones G 1996 In vitro metabolism of the vitamin D analog, 22-oxacalcitriol, using cultured osteosarcoma, hepatoma, and keratinocyte cell lines. J Biol Chem 271:8700–8708.

GLENVILLE JONES

27. Jones G 1986 Elucidation of a new pathway of 25-hydroxyvitamin D3 metabolism. Meth Enzymol 123:141–154. 28. Sorensen H, Binderup L, Calverley MJ, Hoffmeyer L, Rastrup Anderson N 1990. In vitro metabolism of calcipotriol (MC903), a vitamin D analog. Biochem Pharmacol 39:391–393. 29. Suda T, Hallick RB, DeLuca HF, Schnoes HK 1970 25-Hydroxydihydrotachysterol3 synthesis and biological activity. Biochemistry 9:1651–1657. 30. Bosch R, Versluis C, Terlouw JK, Thijssen JHH, Duursma SA 1985 Isolation and identification of 25-hydroxydihydrotachysterol2, 1α,25-dihydroxydihydrotachysterol2 and lβ,25dihydroxydihydrotachysterol2. J Steroid Biochem 23:223–229. 31. Jones G, Edwards N, Vriezen D, Porteous C, Tafford DJH, Cunningham J, Makin HLJ 1988 Isolation and identification of seven metabolites of 25-hydroxydihydrotachysterol3 formed in the isolated perfused rat kidney: A model for the study of side-chain metabolism of vitamin D. Biochemistry 27:7070–7079. 32. Qaw F, Calverley MJ, Schroeder NJ, Trafford DJH, Makin HLJ, Jones G 1993 In vivo metabolism of the vitamin D analog, dihydrotachysterol. Evidence for formation of 1α,25and 1β,25-dihydroxydihydrotachysterol metabolites and studies of their biological activity. J Biol Chem 268:282–292. 33. Bhattacharyya MH, DeLuca HF 1973 Comparative studies on the 25-hydroxylation of vitamin D3 and dihydrotachysterol3. J Biol Chem 248:2974–2977. 34. Wing RM, Okamura WH, Pirio MP, Sine SM, Norman AW 1974 Vitamin D in solution: Conformations of vitamin D3, 1,25-dihydroxyvitamin D3 and dihydrotachysterol3. Science 186:939–941. 35. Qaw F, Schroeder NJ, Calverley MJ, Maestro M, Mourino A, Trafford DJH, Makin HLJ, Jones G 1992 In vitro synthesis of 1,25-dihydroxydihydrotachysterol in the myelomonocytic cell line, HD-11. J Bone Miner Res 7:S161 (Abstract 274). 36. Shany S, Ren S-Y, Arbelle JE, Clemens TL, Adams JS 1993 Subcellular localization and partial purification of the 25-hydroxyvitamin D-1-hydroxylation reaction in the avian myelo-monocytic cell line HD-11. J Bone Miner 8:269–276. 37. Qaw FS, Makin HLJ, Jones G 1992 Metabolism of 25-hydroxy-dihydrotachysterol3 in bone cells in vitro. Steroids 57:236–243. 38. Schroeder NJ, Qaw F, Calverley MJ, Trafford DJH, Jones G, Makin HLJ 1992 Polar metabolites of dihydrotachysterol3 in the rat: Comparison with in vitro metabolites of 1α,25dihydroxydihydrotachysterol3. Biochem Pharmacol 43: 1893–1905. 39. Suda T, DeLuca HF, Schnoes HK, Blunt JW 1969 Isolation and identification of 25-hydroxyergocalciferol. Biochemistry 8:3515–3520. 40. Jones G, Schnoes HK, DeLuca HF 1975 Isolation and identification of 1,25-dihydroxyvitamin D2. Biochemistry 14:1250–1256. 41. Jones G, Rosenthal A, Segev D, Mazur Y, Frolow F, Halfon Y, Rabinovich D, Shakked Z 1979 Isolation and identification of 24,25-dihydroxyvitamin D2 using the perfused rat kidney. Biochemistry 18:1094–1101. 42. Jones G, Schnoes HK, Levan L, DeLuca HF 1980 Isolation and identification of 24-hydroxyvitamin D2 and 24,25-dihydroxyvitamin D2. Arch Biochem Biophys 202:450–457. 43. Horst RL, Koszewski NJ, Reinhardt TA 1990 1a-Hydroxylation of 24-hydroxyvitamin D2 represents a minor physiological pathway for the activation of vitamin D2 in mammals. Biochemistry 29:578–582.

1445

CHAPTER 81 Analog Metabolism

44. Koszewski NJ, Reinhardt TA, Napoli JL, Beitz DC, Horst RL 1988 24,26-Dihydroxyvitamin D2: A unique physiological metabolite of vitamin D2. Biochemistry 27:5785–5790. 45. Reddy GS, Tserng K-Y 1986 Isolation and identification of 1,24,25-trihydroxyvitamin D2, 1,24,25,28-tetrahydroxyvitamin D2, and 1,24,25,26-tetrahydroxyvitamin D2: New metabolites of 1,25-dihydroxyvitamin D2 produced in the rat kidney. Biochemistry 25:5328–5336. 46. Lam HY, Schnoes HK, DeLuca HF 1974 1α-Hydroxyvitamin D2: A potent synthetic analog of vitamin D2. Science 186:1038–1040. 47. Strugnell S, Byford V, Makin HLJ, Moriarty RM, Gilardi R, LeVan LW, Knutson JC, Bishop CW, Jones G 1995 1α,24(S)-dihydroxyvitamin D2: A biologically active product of 1α-hydroxyvitamin D2 made in the human hepatoma, Hep3B. Biochem J 310:233–241. 48. Jones G, Byford V, Kremer R, Makin HLJ, Rice RH, deGraffenreid LA, Knutson JC, Bishop CA 1996 Antiproliferative activity and target cell catabolism of the vitamin D analog, 1α,24(S)-dihydroxyvitamin D2 in normal and immortalized human epidermal cells. Biochem Pharmacol 52:133–140. 49. Sjoden G, Smith C, Lindgren V, DeLuca HF 1985 1α-hydroxyvitamin D2 is less toxic than 1α-hydroxyvitamin D3 in the rat. Proc Soc Exp Biol Med 178:432–436. 50. Horst RL, Koszewski NJ, Reinhardt TA 1988 Species variation of vitamin D metabolism and action: Lessons to be learned from farm animals. In: AW Norman, K Schaefer, H-G Grigoleit, D von Herrath (eds) Vitamin D: Molecular, Cellular, and Clinical Endocrinology, de Gruyter: Berlin, pp. 93–101. 51. Clark JW, Reddy GS, Santos-Moore A, Wankadiya KF, Reddy GP, Lasky S, Tserng K-Y, Uskokovic MR 1993 Metabolism and biological activity of 1,25-dihydroxyvitamin D2 and its metabolites in a chronic myelogenous leukemia cell line, RWLEU-4. Bioorg Med Lett 3:1873–1878. 52. Shankar VN, Propp AE, Schroeder NS, Surber BW, Makin HLJ, Jones G 2001 In vitro metabolism of 19-nor1α,25-(OH)2D2 in cultured cell lines: inducible synthesis of lipid- and water-soluble metabolites. Arch Biochem Biophys 387:297–306. 52a. Jones G, Byford V, Kremer R, Makin HLJ, Knutson JC, Bishop CW 1994. Novel cleavage of vitamin D2 side chain during catabolism by keratinocyte cell line. Ninth Workshop on Vitamin D, Orlando, Florida, June 1994, abstract. 53. Zimmerman DR, Reinhardt TA, Kremer R, Beitz DC, Reddy GS, Horst RL 2001 Calcitroic acid is a major catabolic metabolite in the metabolism of 1α-dihydroxyvitamin D2. Arch Biochem Biophys 392:14–22. 54. Horst RL, Omdahl JA, Reddy S 2003 Rat cytochrome P450C24 (CYP24) does not metabolize 1,25-dihydroxyvitamin D2 to calcitroic acid. J Cell Biochem 88:282–285. 55. Masuda S, Arabian A, McCaig J, Kaufmann M, Strugnell SA, Knutson JC, St-Arnaud R, Jones G 2002 CYP24-Null keratinocytes demonstrate that CYP24 is responsible for activation and inactivation of 1α(OH)D2. J Bone Mineral Research 17:S396, abstract SU459. 56. Byford V, Strugnell S, Coldwell R, Schroeder N, Makin HLJ, Knutson JC, Bishop CW, Jones G 2002. Use of vitamin D4 analogs to investigate differences in hepatic and target cell metabolism of vitamins D2 and D3. Biochim Biophys Acta 1583:151–166. 57. Mehta RG, Hussain EA, Mehta RR, Das Gupta TK 2003 Chemoprevention of mammary carcinogenesis by

58.

59. 60. 61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

1α-hydroxyvitamin D5, a synthetic analog of vitamin D. Mutat Res 523–524:253–264. Strugnell S, Calverley MJ, Jones G 1990 Metabolism of a cyclo-propane-ring-containing analog of 1α-hydroxyvitamin D3 in a hepatocyte cell model: Identification of 24-oxidized metabolites. Biochem Pharmacol 40:333–341. Calverley MJ 1987 Synthesis of MC903, a biologically active vitamin D metabolite analog. Tetrahedron 43:4609–4619. Jones G, Calverley MJ 1993 A dialogue on analogs: Newer vitamin D drugs for use in bone disease, psoriasis, and cancer. Trends Endocrinol Metab 4:297–303. Binderup L 1988 MC903 A novel vitamin D analog with potent effects on cell proliferation and cell differentiation. In: AW Norman, K Schaefer, H-G Grigoleit, D von Herrath (eds) Vitamin D: Molecular: Cellular and Clinical Endocrinology de Gruyter: Berlin, pp. 300–309. Calverley MJ, Binderup E, Binderup L 1991 The 20-epi modification in the vitamin D series: Selective enhancement of “nonclassical” receptor-mediated effects. In: AW Norman, R Bouillon, M Thomasset (eds) Vitamin D: Gene Regulation, Structure-Function Analysis and Clinical Application, de Gruyter: Berlin, pp. 163–164. Binderup L, Latini S, Binderup E, Bretting C, Calverley M, Hansen K 1991 20-Epi-vitamin D3 analogs: A novel class of potent regulators of cell growth and immune responses. Biochem Pharmacol 42:1569–1575. Shevde NK, Plum LA, Clagett-Dame M, Yamamoto H, Pike JW, DeLuca HF 2002 A potent analog of 1α,25-dihydroxyvitamin D3 selectively induces bone formation. Proc Natl Acad Sci USA 99:13487–13491. Dilworth FJ, Calverley MJ, Makin HLJ, Jones G 1994 Increased biological activity of 20-epi-l,25-dihydroxyvitamin D3 is due to reduced catabolism and altered protein binding. Biochem Pharmacol 47:987–993. Siu-Caldera ML, Sekimoto H, Peleg S, Nguyen C, Kissmeyer AM, Binderup L, Weiskopf A, Vouros P, Uskokovic MR, Reddy GS 1999 Enhanced biological activity of 1α,25-dihydroxy-20-epi-vitamin D3, the C-20 epimer of 1α,25-dihydroxyvitamin D3, is in part due to its metabolism into stable intermediary metabolites with significant biological activity. J Steroid Biochem Mol Biol 71:111–121. Peleg S, Sastry M, Collins ED, Bishop JE, Norman AW 1995 Distinct conformational changes induced by 20-epi analogs of 1α,25-dihydroxyvitamin D3 are associated with enhanced activation of the vitamin D receptor. J Biol Chem 270:10551–10558. Shankar VN, Byford V, Prosser DE, Schroeder NJ, Makin HLJ, Wiesinger H, Neef G, Steinmeyer A, Jones G 2001 Metabolism of a 20-methyl substituted series of vitamin D analogs by cultured human cells: Apparent reduction of 23-hydroxylation of the side chain by 20-methyl group. Biochem Pharmacol 61:893–902. Danielsson C, Nayeri S, Wiesinger H, Thieroff-Ekerdt R, Carlberg C 1996. Potent gene regulatory and antiproliferative activities of 20-methyl analogs of 1,25 dihydroxyvitamin D3. J Cell Biochem 63:199–206. Ostrem VK, Lau WF, Lee SH, Perlman K, Prahl J, Schnoes HK, DeLuca HF, Ikekawa N 1987 Induction of monocytic differentiation of HL-60 cells by 1,25-dihydroxyvitamin D analogs. J Biol Chem 262:14164–14171. Ostrem VK, Tanaka V, Prahl J, DeLuca HF, Ikekawa N 1987 24- and 26-homo-l,25-dihydroxyvitamin D3: Preferential activity in inducing differentiation of human leukemic cells HL-60 in vitro. Proc Natl Acad Sci USA 84:2610–2614.

1446 72. Perlman K, Kutner A, Prahl J, Smith C, Inaba M, Schnoes HK, DeLuca HF 1990 24-homologated 1,25-dihydroxyvitamin D3 compounds: Separation of calcium and cell differentiation activities. Biochemistry 29:190–196. 73. Paulson SK, Perlman K, DeLuca H, Stern PH 1990 24- and 26-homo-l,25-dihydroxyvitamin D3 analogs: Potencies on in vitro bone resorption differ from those reported for cell differentiation. J Bone Miner Res 5:201–206. 74. Dilworth FJ, Scott I, Green A, Strugnell S, Guo Y-D, Roberts EA, Kremer R, Calverley MJ, Makin HLJ, Jones G 1995a Different mechanisms of hydroxylation site selection by liver and kidney cytochrome P450 species (CYP27 and CYP24) involved in vitamin D metabolism. J Biol Chem 270:16766–16774. 75. Dilworth FJ, Scott I, Calverley MJ, Makin HLJ, Jones G 1995b Enzymes of side chain oxidation pathway not affected by addition of methyl groups to end of the vitamin D3 side chain. J Bone Miner Res 10:S388 (Abstract M546). 76. Binderup L, Carlberg C, Kissmeyer AM, Latini S, Mathiasen IS, Mork-Hansen C 1993 The need for new vitamin D analogs: Mechanisms of action and clinical applications. In: AW Norman, R Bouillon, M Thomasset (eds) Vitamin D. A Pluripotent Steroid Hormone: Structural Studies, Molecular Endocrinology and Clinical Applications, de Gruyter: Berlin, pp. 55–63. 77. Zhou J-Y, Norman AW, Chen D-L, Sun G, Uskokovic M, Koeffler HP 1990 l,25-Dihydroxy-16-ene-23-yne-vitamin D3 prolongs survival time of leukemic mice. Proc Natl Acad Sci USA 87:3929–3932. 78. Binderup E, Calverley MJ, Binderup L 1991 Synthesis and biological activity of 1a-hydroxylated vitamin D analogues with poly-unsaturated side chains. In: AW Norman, R Bouillon, M Thomasset (eds) Vitamin D: Gene Regulation, StructureFunction Analysis and Clinical Application, de Gruyter: Berlin, pp. 192–193. 79. Colston KW, Mackay AG, James SY, Binderup L, Chandler S, Coombes RC 1992 EB1089: A new vitamin D analog that inhibits the growth of breast cancer cells in vivo and in vitro. Biochem Pharmacol 44:2273–2280. 80. James SY, Mackay AG, Binderup L, Colston KW 1994 Effects of a new synthetic analog, EB1089, on the estrogenresponsive growth of human breast cancer cells. J Endocrinol 141:555–563. 81. Reddy GS, Clark JW, Tserng K-Y, Uskokovic MR, McLane JA 1993 Metabolism of l,25-(OH)2-16-ene D3 in kidney: Influence of structural modification of D-ring on side chain metabolism. Bioorg Med Lett 3:1879–1884. 82. Satchell DP, Norman AW 1996 Metabolism of the cell-differentiating agent l,25-(OH)2-16-ene-23-yne vitamin D3 by leukemic cells. J Steroid Biochem Mol Biol 57:117–124. 83. Shankar VN, Makin HLJ, Schroeder NJ, Trafford DJH, Kissmeyer A-M, Calverley MJ, Binderup E, Jones G 1995 Metabolism of the antiproliferative vitamin D analog, EB1089, in a cultured human keratinocyte model. Bone 17:326 (abstract). 84. Kissmeyer A-M, Mathiasen IS, Latini S, Binderup L 1995 Pharmacokinetic studies of vitamin D analogs: Relationship to vitamin D–binding protein (DBP). Endocrine 3:263–266. 85. Shankar VN, Dilworth FJ, Makin HIJ, Schroeder NJ, Trafford DAJ, Kissmeyer A-M, Calverley MJ, Binderup E, Jones G 1997 Metabolism of the vitamin D analog EB1089 by cultured human cells: Redirection of hydroxylation site to distal carbons of the side chain. Biochem Pharmacol 53: 783–793.

GLENVILLE JONES

86. Kissmeyer A-M, Binderup E, Binderup L, Hansen CM, Andersen NR, Schroeder NJ, Makin HLJ, Shankar VN, Jones G 1997 The metabolism of the vitamin D analog EB1089: Identification of in vivo and in vitro metabolites and their biological activities. Biochem Pharmacol 53:1087–1097. 87. Wankadiya KF, Uskokovic MR, Clark J, Tserng K-Y, Reddy GS 1992 Novel evidence for the reduction of the double bond in ∆22-l,25-dihydroxyvitamin D3. J Bone Miner Res 7:S171 (Abstract 315). 88. Murayama E, Miyamoto K, Kubodera N, Mori T, Matsunaga I 1986 Synthetic studies of vitamin D analogs. VIII. Synthesis of 22-oxavitamin D3 analogs. Chem Pharm Bull (Tokyo) 34:4410–4413. 89. Hansen K, Calverley MJ, Binderup L 1991 Synthesis and biological activity of 22-oxavitamin D analogs. In: AW Norman, R Bouillon, M Thomasset (eds) Vitamin D: Gene Regulation, Structure-Function Analysis and Clinical Application, de Gruyter: Berlin, pp. 161–162. 90. Kubodera N, Miyamoto K, Akiyama M, Matsumoto M, Mori T 1991 Synthetic studies of vitamin D analogs. 9. Synthesis and differentiation-inducing activity of 1α,25-dihydroxy-23oxa-vitamin D3, thia-vitamin D3 and aza-vitamin D3. Chem Pharm Bull (Tokyo) 39:3221–3224. 91. Baggiolini EG, Zacobelli JA, Hennessy BM, Batcho AD, Sereno JF, Uskokovic MR 1986 Stereocontrolled total synthesis of 1α,25-dihydroxycholecalciferol and 1α,25-dihydroxyergocalciferol. J Org Chem 51:3098–3108. 92. Allewaert K, Van Baelen H, Bouillon R, Zhao X-Y, De Clercq P, Vanderwalle M 1993 Synthesis and biological evaluation of 23-oxa-, 23-thia- and 23-oxa-24-oxo-1α, 25-dihydroxyvitamin D3. Bioorg Med Lett 3:1859–1862. 93. Calverley MJ, Strugnell S, Jones G 1993 The seleno-acetal route to 1α-hydroxy-vitamin D analogs: Synthesis of 24-oxa1α-hydroxyvitamin D3, a useful vitamin D metabolism probe. Tetrahedron 49:739–746. 94. Sarandeses LA, Valles MJ, Castedo L, Mourino A 1993 Synthesis of 24-oxa-vitamin D3 and 1α-hydroxy-24-oxavitamin D3. Tetrahedron 49:731–738. 95. Bikle DD, Abe-Hashimoto J, Su MJ, Felt S, Gibson DFC, Pillai S 1995 22-Oxa calcitriol is a less potent regulator of keratinocyte proliferation and differentiation due to decreased cellular uptake and enhanced catabolism. J Invest Dermatol 105:693–698. 96. Kobayashi T, Tsugawa N, Okano T, Masuda S, Takeuchi A, Kubodera N, Nishii Y 1994 The binding properties with blood proteins and tissue distribution of 22-oxa-1α,25-dihydroxyvitamin D3, a noncalcemic analog of 1α,25-dihydroxyvitamin D3 in rats. J Biochem (Tokyo) 115:373–380. 97. Jones G, Byford V, Moore B, Logsted-Nielsen J, Calverley MJ, Hansen K, Kissmeyer A-M 2000. A novel aromatic vitamin D analog is rapidly metabolized in cultured cell lines in vitro. J Bone Mineral Research 15: Abstract M507, S576. 98. Dilworth FJ, Calverley MJ, Kissmeyer A-M, Binderup E, Makin HLJ, Jones G 1996 KH1060, a potent vitamin D analog, is degraded in cultured keratinocytes via several different pathways. Bone Miner Res 11:S424 (abstract T500). 99. Rastrup Anderson N, Buchwald FA, Grue-Sorensen G 1992 Identification and synthesis of a metabolite of KH1060, a new potent 1α,25-dihydroxyvitamin D3 analog. Bioorg Med Chem Lett 2:1713–1716. 100. Van Den Bemd G, J-C M, Dilworth FJ, Makin HLJ, Prahl JM, DeLuca HF, Jones G, Pols HAP, Van Leeuwen JPTM 2000 Contribution of several metabolites of the vitamin D analog 20-epi-22-oxa-24a,26a,27a-trihomo-1,

1447

CHAPTER 81 Analog Metabolism

101.

102. 103.

104.

105. 106.

107.

108.

109.

110.

111. 112.

113.

114.

115.

25-(OH)2vitamin D3 (KH1060) to the overall biological activity of KH1060 by a shared mechanism of action. Biochem Pharmacol 59:621–27 Reddy GS, Ishizuka S, Wandkadiya KF, Tserng K-Y, Yeung B, Vouros P 1992 Metabolism of 1α,24(R)-dihydroxyvitamin D3 into calcitroic acid in rat kidney through two different metabolic pathways. J Bone Miner Res 7:S170 (abstract 312). Tucker G, Gagnon RE, Haussler MR 1973 Vitamin D3-25hydroxylase: Tissue occurrence and lack of regulation. Arch Biochem Biophys 155:47–57. Axen E, Postlind H, Wikvall K 1995 Effects of CYP27 mRNA expression in rat kidney and liver by 1α,25-dihydroxyvitamin D3, a suppressor of renal 25-hydroxyvitamin D3-1α-hydroxylase activity. Biochem Biophys Res Commun 215:136–141. Weinstein EA, Siu-Caldera M-L, Ishizuka S, Reddy GS 1995 Evidence of 25-hydroxylation in the rat kidney for 1α,24(R)-dihydroxyvitamin D3 only but not for vitamin D3 or 1α-hydroxyvitamin D3. J Bone Miner Res 10:S497 (abstract T570). Nishii Y, Sato K, Kobayashi T 1993 The development of vitamin D analogs for the treatment of osteoporosis. Osteoporosis Int 1 (Suppl):S190–193. Masuda S, Makin HLJ, Kremer R, Okano T, Kobayashi T, Sato K, Nishii Y, Jones G 1994 Metabolism of 2β-(3-hydroxypropoxy)-1α,25-dihydroxyvitamin D3 (ED-71) in cultured cell lines. J Bone Miner Res 9:S289 (abstract B238). Kobayashi Y, Taguchi T, Mitsuhashi S, Eguchi T, Ohshima E, Ikekawa N 1982 Studies on organic fluorine compounds. XXXIX. Studies on steroids. LXXIX. Synthesis of 1α,25dihydroxy-26,26,26,27,27,27-hexafluorovitamin D3. Chem Pharm Bull (Tokyo) 30:4297–4303. Koeffler HP, Armatruda T, Ikekawa N, Kobayashi Y, DeLuca HF 1984 Induction of macrophage differentiation of human normal and leukemic myeloid stem cells by 1α,25dihydroxyvitamin D3 and its fluorinated analogs. Cancer Res 44:6524–6528. Inaba M, Okuno S, Nishizawa Y, Yukioka K, Otani S, Matsui-Yuasa I, Morisawa S, DeLuca HF, Morii H 1987 Biological activity of fluorinated vitamin D analogs at C-26 and C-27 on human promyelocytic leukemia cells, HL-60. Arch Biochem Biophys 258:421–425. Kistler A, Galli B, Horst R, Truitt GA, Uskokovic MR 1989 Effects of vitamin D derivatives on soft tissue calcification in neonatal and calcium mobilization in adult rats. Arch Toxicol 63:394–400. Lohnes D, Jones G 1992 Further metabolism of 1α,25-dihydroxyvitamin D3 in target cells. J Nutr Sci Vitaminol (Tokyo) Special Issue: 75–78. Inaba M, Okuno S, Nishizawa Y, Imanishi Y, Katsumata T, Sugata I, Morii H 1993 Effect of substituting fluorine for hydrogen at C-26 and C-27 on the side chain of 1α,25-dihydroxyvitamin D3. Biochem Pharmacol 45:2331–2336. Nakatsuka K, Imanishi Y, Morishima Y, Sekiya K, Sasao K, Miki T, Nishizawa Y, Katsumata T, Nagata A, Murakawa S 1992 Biological potency of a fluorinated vitamin D analog in hypoparathyroidism. Bone Miner 16:73–81. Nishizawa Y, Morii H, Ogura Y, DeLuca HF 1991 Clinical trial of 26,26,26,27,27,27-hexafluoro-1α,25-dihydroxyvitamin D3 in uremic patients on hemodialysis: Preliminary report. Contrib Nephrol 90:196–203. Posner GH, Crawford K, Siu-Caldera ML, Reddy GS, Sarabia SF, Feldman D, van Etten E, Mathieu C, Gennaro L, Vouros P, Peleg S, Dolan PM, Kensler TW 2000

116.

117. 118.

119.

120.

121.

122.

123. 124.

125. 126. 127. 128.

129.

130.

131.

Conceptually new 20-epi-22-oxa sulfone analogs of the hormone 1α,25-dihydroxyvitamin D3: synthesis and biological evaluation. J Med Chem 43:3581–3586. Perlman KL, Sicinski RR, Schnoes HK, DeLuca HF 1990 1α,25-Dihydroxy-19-nor-vitamin D3, a novel vitamin D– related compound with potential therapeutic activity. Tetrahedron Lett 31:1823–1824. Posner GH, Dai H 1993 l-(Hydroxyalkyl)-25-hydroxyvitamin D3 analogs of calcitriol-I. Synthesis. Bioorg Med Chem Lett 2:1829–1834. Neef G, Kirsch G, Schwarz K, Wiesinger H, Menrad A, Fahnrich M, Thieroff-Ekerdt, Steinmeyer A 1994 20-Methyl vitamin D analogs. In: AW Norman, R Bouillon, M Thomasset (eds) Vitamin D. A Pluripotent Steroid Hormone: Structural Studies, Molecular Endocrinology and Clinical Applications, de Gruyter: Berlin, pp. 97–98. Bouillon R, Allewaert K, Xiang DZ, Tan BK, Van Baelen H 1991 Vitamin D analogs with low affinity for the vitamin D– binding protein: Enhanced in vitro and decreased in vivo activity. J Bone Miner Res 6:1051–1057. Dusso AS, Negrea L, Gunawardhana S, Lopez-Hilker S, Finch J, Mori T, Nishii Y, Slatopolsky E, Brown AJ 1991 On the mechanisms for the selective action of vitamin D analogs. Endocrinology 128:1687–1692. Cheskis B, Lemon BD, Uskokovic MR, Lomedico PT, Freedman LP 1995 Vitamin D3-retinoid X receptor dimerization. DNA binding, and transactivation are differentially affected by analogs of 1,25-dihydroxyvitamin D3. Mol Endocrinol 9:1814–1824. Schuster I, Egger H, Astecker N, Herzig G, Schussler M, Vorisek G 2001 Selective inhibitors of CYP24: mechanistic tools to explore vitamin D metabolism in human keratinocytes. Steroids 66:451–462. Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2001. The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179. Tocchini-Valentini G, Rochel N, Wurtz JM, Mitschler A, Moras D 2001 Crystal structures of the vitamin D receptor complexed to superagonist 20-epi ligands. Proc Natl Acad Sci USA 98:5491–5496. Guengerich FP 1991 Reactions and significance of cytochrome P-450 enzymes. J Biol Chem 266:10019–10022. Graham-Lorence S, Amarneh B, White RE, Peterson JA, Simpson ER 1995 A three-dimensional model of the aromatase cytochrome P450. Protein Sci 4:1065–1080. Vijayakumar S, Salerno JC 1992 Molecular modeling of the 3-D structure of cytochrome P-450SCC. Biochim Biophys Acta 1160:281–286. Prosser DE, Dakin KA, Donini OAT, Weaver DF, Jia Z, Jones G 1996 A three-dimensional model of the cytochrome P450, CYP27A and its vitamin D binding site. J Bone Miner Res 11:S313 (abstract M525). Omdahl JL, Bobrovnikova EV, Annalora A, Chen P, Serda R 2003 Expression, structure-function, and molecular modeling of vitamin D P450s. J Cell Biochem 88(2): 356–362. Cali JJ, Russell DW 1991 Characterization of human sterol 27-hydroxylase. A mitochondrial cytochrome P450 that catalyzes multiple oxidation reactions in bile acid biosynthesis. J Biol Chem 266:7774–7778. Cali JJ, Hsieh C-L, Francke V, Russell DW 1991 Mutations in the bile acid biosynthetic enzyme sterol 27-hydroxylase underlie cerebrotendinous xanthomatosis. J Biol Chem 266:7779–7783.

1448 132. St-Arnaud R, Messerlian S, Moir JM, Omdahl JL, Glorieux FH. 1997 The 25-hydroxyvitamin D 1α-hydroxylase gene maps to the pseudovitamin D–deficiency rickets (PDDR) disease locus. J Bone Miner Res 12:1552–1559. 133. Schuster I, Egger H, Nussbaumer P, Kroemer RT 2003 Inhibitors of vitamin D hydroxylases: Structure-activity relationships. J Cell Biochem 88:372–380. 134. Barton DH, Hesse RH, Pechet MM, Rizzardo E 1973 A convenient synthesis of 1α-hydroxy-vitamin D3. J Am Chem Soc 95:2748–2749. 135. Paaren HE, Hamer DE, Schnoes HK, DeLuca HF 1978 Direct C-l hydroxylation of vitamin D compounds: Convenient preparation of lα-hydroxyvitamin D3, lα,25-dihydroxyvitamin D3 and 1α-hydroxyvitamin D2. Proc Natl Acad Sci USA 75:2080–2081. 136. Fraser D, Kooh SW, Kind P, Holick MF, Tanaka Y, DeLuca HF 1973 Pathogenesis of hereditary vitamin D–dependency rickets. N Engl J Med 289:817–822. 137. Baggiolini EG, Wovkulich PM, Iacobelli JA, Hennessy BM, Uskokovic MR 1982 Preparation of 1-alpha hydroxylated vitamin D metabolites by total synthesis. In: AW Norman, K Schaefer, D von Herrath, H-G Grigoleit (eds) Vitamin D: Chemical, Biochemical and Clinical Endocrinology of Calcium Metabolism, de Gruyter: Berlin, pp. 1089–1100. 138. Baggiolini EG, Partridge JJ, Shiuey S-J, Truitt GA, Uskokovic MR 1989 Cholecalciferol 23-yne derivatives, their pharmaceutical compositions, their use in the treatment of calcium-related diseases, and their antitumor activity, US 4,804,502 [abstract]. Chem Abstr 111:58160d. 139. Morisaki M, Koizumi N, Ikekawa N, Takeshita T, Ishimoto S 1975 Synthesis of active forms of vitamin D. Part IX. Synthesis of 1α,24-dihydroxycholecalciferol. J Chem Soc Perkin Trans 1(1):1421–1424. 140. Fukushima M, Suzuki Y, Tohira Y, Nishii Y, Suzuki M, Sasaki S, Suda T 1976 25-Hydroxylation of lα-hydroxyvitamin D3 in vivo and in the perfused rat liver. FEBS Lett 65:211–214.

GLENVILLE JONES

141. Rosenthal AM, Jones G, Kooh SW, Fraser D 1980 25-Hydroxyvitamin D3 metabolism by the isolated perfused rat kidney. Am J Physiol (Endocrinol Metab) 239:E12–E20. 142. Reddy GS, Jones G, Kooh SW, Fraser D, DeLuca HF 1983 Effects of metabolites and analogs of vitamin D3 on 24(R),25-dihydroxyvitamin D3 synthesis. Am J Physiol 235:E359–E364. 143. Henry HL 1979 Regulation of the hydroxylation of 25-hydroxyvitamin D3 in vivo and in primary cultures of chick kidney cells. J Biol Chem 254:2722–2729. 144. Chandler JS, Chandler SK, Pike JW, Haussler MR 1984 1,25-Dihydroxyvitamin D3 induces 25-hydroxyvitamin D324-hydroxylase in a cultured monkey kidney cell line (LLC-MK2) apparently deficient in the high affinity receptor for the hormone. J Biol Chem 259:2214–2222. 145. Vieth R, Fraser D 1979 Kinetic behavior of 25-hydroxyvitamin D-1-hydroxylase and -24-hydroxylase in rat kidney mitochondria. J Biol Chem 254:12455–12460. 146. Knutson JC, LeVan LW, Valliere CR, Bishop CW 1997 Pharmacokinetics and systemic effect on calcium homeostasis of 1α,24-dihydroxyvitamin D2 in rats: Comparison with lα,25-dihydroxyvitamin D2, calcitriol and calcipotriol. Biochem Pharmacol 53:829–837. 147. Jones G 2002 Part III Pharmacological mechanisms of therapeutics: Vitamin D and analogs. In: J Bilezikian, L Raisz, G Rodan (eds) Principles of Bone Biology, Second Edition. Academic Press: San Diego, pp. 1407–1422. 148. Masuda S, Jones G 2003.Vitamin D Analogs: Drug design based upon proteins involved in vitamin D signal transduction. J Current Drug Targets: Immune, Endocrine and Metabolic Disorders 3:43–67. 149. Jones G, Ramshaw H, Zhang A, Cook R, Byford V, White J, Petkovich M 1999 Expression and activity of vitamin D–metabolizing cytochrome P450s (CYP1alpha and CYP24) in human non small cell lung carcinomas. Endocrinology 140:3303–3310.

CHAPTER 82

Mechanisms for the Selective Actions of Vitamin D Analogs ALEX J. BROWN AND EDUARDO A. SLATOPOLSKY Renal Division, Department of Medicine, Washington University School of Medicine, St. Louis, Missouri

I. Identification of Selective Vitamin D Analogs II. The in Vivo Selectivity Vitamin D Analogs Is Determined by multiple Protein Interactions

III. Concluding Remarks References

Vitamin D was first identified as an essential factor for normal mineral metabolism and skeletal development. These actions are now known to be mediated by metabolites of vitamin D, primarily 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] but perhaps 24,25-dihydroxyvitamin D3 as well. 1,25(OH)2D3 acts as a steroid hormone and binds to a well-characterized intracellular receptor that is localized in the nucleus and regulates gene transcription by binding to specific motifs within the target genes (see Chapters 11–22). There is also in vitro evidence that 1,25(OH)2D3 and 24,25(OH)2D3 can interact with cell surface receptors and rapidly stimulate signaling from the cell membrane [1] (see Chapter 23). At present, the potential roles for these nongenomic actions and their relevance in vivo are less well understood. Research from many laboratories has demonstrated that the actions of the vitamin D system extend beyond a role restricted to bone and mineral metabolism. Many of these activities suggested potential therapeutic applications for 1,25(OH)2D3. The native vitamin D hormone is currently in use for the treament of secondary hyperparathyroidism in renal failure patients [2], psoriasis [3], and X-linked hypophosphatemic rickets [4]. In addition, the ability of 1,25(OH)2D3 to block proliferation of many cell types, including neoplastic cells, in vitro has indicated the potential of this compound for treating various types of cancer [5] (see Chapters 88–97). However, a major limitation to 1,25(OH)2D3 therapy is its potent calcemic and phosphatemic activities. Studies to date suggest that the doses of 1,25(OH)2D3 necessary to block cell proliferation in vivo may also produce profound hypercalcemia and hyperphosphatemia.

I. IDENTIFICATION OF SELECTIVE VITAMIN D ANALOGS

VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

A major breakthrough in vitamin D therapeutics has been the development of vitamin D analogs that retain many of the potential clinically useful activities of 1,25(OH)2D3 for the newly discovered indications, but have much lower calcemic and phosphatemic activities in vivo [6,7]. The promising analogs identified initially displayed high differentiating and antiproliferative activities in vitro, but were found to have low calcemic activity in vivo [8–12]. One explanation for these results could have been that the analogs were relatively less active than 1,25(OH)2D3 in vivo due to reduced bioavailability. Although this may be the case for some analogs, several “noncalcemic” analogs [actually, the analogs are all calcemic but to a lesser degree than 1,25(OH)2D3] have been found to retain some of the activities of 1,25(OH)2D3 in animal models. Selective actions of several analogs have now been demonstrated in vivo. These analogs are discussed in detail in Chapters 79–87, and their uses for specific indications are discussed in Chapters 88–104. Secondary hyperparathyroidism in chronic renal failure has been treated with 1,25(OH)2D3 or 1α(OH)D3 for many years, but the potent calcemic activities of these compounds often produce hypercalcemia. Several new analogs have been developed that retain the PTH suppressive effect of 1,25(OH)2D3 but with lower calcemic activity. Four analogs are currently available to patients: l,25(OH)2-19-nor-D2 (Abbott Laboratories) and 1α(OH)D2 (Bone Care International) in the United States and 22-oxa-l,25(OH)2D3 (OCT; Chugai Pharmaceuticals) and 1,25(OH)2-26,27-F6-D3 in Copyright © 2005, Elsevier, Inc. All rights reserved.

1450

ALEX J. BROWN AND EDUARDO A. SLATOPOLSKY

PTH, % suppression vs. vehicle

Japan [13] (see Chapters 75, 85, 86, and 103). Studies in animals have revealed that OCT effectively suppressed parathyroid hormone (PTH) levels in hyperparathyroid uremic rats at doses that had minimal effects on serum calcium and phosphate levels [14–16]. Data from the study of Hirata et al. [16] demonstrating the selectivity of OCT are illustrated in Fig. 1. Similarly, 1,25(OH)219-nor-D2 was found to be about three times less active than 1,25(OH)2D3 in suppressing PTH in patients, but ten times less calcemic and phosphatemic [17–21]. Renal failure patients receiving 1,25(OH)2-19-nor-D2 have been found to have lower rates of mortality than those receiving 1,25(OH)2D3 [22]; the mechanism is unclear but may involve the lower calcemic and phosphatemic activities of the hormone. Selective effects in vivo of 1α(OH)D2 and 1,25(OH)2-26,27-F6-D3 on the parathyroid glands have not been reported. Other analogs shown to display parathyroid selectivity in animal models include 1,25-dihydroxy-dihydrotachysterol [23] and the 20-epi analogs CB1093, EB1213, 40

1,25(OH)2D3

and GS1725 [24]. Clearly, it is possible to develop analogs with selectively for suppression of PTH. Analogs are also available for treatment of psoriasis. Calcipotriol, the first of the “noncalcemic” analogs to be approved for clinical use, is effective in treating psoriatic lesions by topical application, and has minimal calcemic activity, even when administered systemically [10,25]. 1,24(OH)2D3 and OCT are also available in Japan for psoriasis [26,27]. Although toxicity of topically applied vitamin D compounds is less compared to systemic administration, excessive application can cause hypercalcemia and therefore the lower calcemic activities of the new analogs provide a safer means to treat psoriasis. The greatest activity for vitamin D analog therapy is in the treatment of various types of cancer, as discussed in greater detail in Section IX. 22-Oxa-l,25(OH)2D3 and the analog EB1089 (Leo Pharmaceuticals) were shown to suppress PTH-related peptide in cancer cells in vivo [28,29], suggesting their use in the treatment of

OCT

50 60 70 80 90 100 0.01

0.038

0.1

0.317

1.0

10.0 Dose (µg/kg)

15

Serum Ca (mg/dl)

14 13 1,25(OH)2D3

12

OCT 11.5 mg/dl

11 10 9 8 0.01

FIGURE 1

0.083

0.1

1.0

3.954 10.0 Dose (µg/kg)

Selectivity of OCT for suppression of PTH in uremic rats. Uremic rats were treated with OCT or 1,25(OH)2D3 at the specified IV doses every other day for 2 weeks, and PTH and calcium were determined 24 hours after the final injection. The doses of each compound that produced a 40% suppression of PTH or an increase in calcium to 11.5 mg/dl are shown. OCT was 8.3 times less potent in suppressing PTH, but 47.6 times less calcemic, indicating that the therapeutic window is approximately 5.7 times wider for OCT than for 1,25(OH)2D3. Adapted from Hirata et al. [16].

CHAPTER 82 Mechanisms for the Selective Actions of Vitamin D Analogs

hypercalcemia of malignancy. 22-Oxa-l,25(OH)2D3 [30,31], calcipotriol [32], and EB1089 [33] (Leo Pharmaceuticals), l,25-(OH)2-16-ene-23-yne-26,27-F6vitamin D3 [34] (Hoffmann-LaRoche), and 1α(OH)D5 [35] have been shown to inhibit the in vivo proliferation of breast cancer cells. The analogs KH1060, EB1089, and Ro 26-9114 were shown to control prostate cancer cell (LNCaP) growth in nude mice more effectively than 1,25(OH)2D3, inducing tumor necrosis and calcification, but with no hypercalcemia [36]. The analog l,25-(OH)2-16-ene-23-yne-26,27-F6-vitamin D3 may be effective in control of androgen-induced carcinoma of the prostate and seminal vesicles [37,38]. 22-Oxa-1,25(OH)2D3 [39] and l,25-(OH)2-16-ene23-yne-26,27-F6-vitamin D3 [40], and 1,25(OH)2-16ene-19-nor-24-oxo-D3 [41] can inhibit growth of experimentally induced tumors of the small and large intestine. Another derivative, l,25-(OH)2-16-ene-23-ynevitamin D3 (Hoffmann-LaRoche), can prolong the survival of mice injected with leukemia cells [42]. EB1089 was found to give a greater reduction in tumor size than 1,25(OH)2D3 in a mouse model of head and neck squamous cell carcinoma without producing hypercalcemia [43]. Many other analogs with low calcemic activity that have been shown to inhibit proliferation of cancer cells in vitro have not yet been demonstrated to exert selectivity in vivo. The immunomodulatory actions of vitamin D compounds have suggested therapeutic applications for autoimmune disorders and transplantation. Modulation of the immune system with little hypercalcemia has now been demonstrated for several analogs. Abe et al. found that OCT was 50 times more potent than 1,25(OH)2D3 in augmenting a primary immune response in mice, but was 100 times less calcemic [44]. Lemire et al. demonstrated that 1,25(OH)2-16-ene-24-oxo-D3 was more potent than 1,25(OH)2D3 or 1,25(OH)2-16ene-D3 in suppressing experimental autoimmune encephalomyelitis, but, unlike the other two compounds, did not increase serum calcium [45]. KH1060 was shown by Mathieu et al. to reduce the incidence of diabetes in NOD mice at doses that did not increase serum or urinary calcium [46]. Most recently, Zugel et al. showed that the vitamin D analog ZK191784 effectively inhibited contact hypersensitivity in mice [47]. Although it was approximately 100 times less active than 1,25(OH)2D3, its effect on urinary calcium was more than 3000 times lower. These studies clearly illustrate the selectivity of the analogs on the immune system. Osteoporosis is another therapeutic target for vitamin D analogs. 1,25(OH)2D3 or its synthetic analog 1α(OH)D3 have been used with some success, but analogs are being developed that are more effective in increasing bone mineral. One such analog with

1451

clinical potential is 1,25(OH)2-2-(3-hydroxypropoxy)D3 (ED-71) from Chugai Pharmaceuticals. This analog appears to be much more effective than 1,25(OH)2D3 and 1α(OH)D3 in stimulating bone mineralization in ovariectomized rats [48,49] and in corticosteroidtreated rats [50]. ED-71 is currently in clinical trials in Japan. A very promising new analog for osteoporosis is Ro-26-9228. This highly modified compound (1F,25(OH)-20-epi-23-ene-25,26-dimethyl-D3) is bone protective in vivo, and has been shown to be very active in bone but not in the duodenum [51]. Similar selectivity was noted for 2-methylene-19-nor-(20S)1,25(OH)2D3 (2MD), which showed greater relative activity in bone than in the intestine. In OVX rats, 2MD increased bone mass whereas 1,25(OH)2D3 only prevented loss of bone mineral [52]. The mechanism responsible for the cell/tissue specificity of these analogs is under investigation. In general, the low calcemic activity of these analogs permitted the use of doses sufficiently high to obtain the desired effect. Quantification of the degree of selectivity (i.e., potency of the desired effect versus calcemic activity) is difficult to determine in many cases because of lack of data for the relative dose responses for 1,25(OH)2D3 and the analog. In some studies, such data cannot be gathered due to the hypercalcemic toxicity of 1,25(OH)2D3 at the doses required for the desired activity. The novel aspect of these analogs is their differential actions, compared to 1,25(OH)2D3, in vivo. In fact, as these analogs have relatively high affinity for the vitamin D receptor (VDR), usually within one order of magnitude, it is not unexpected that they are able to mimic many of the actions of 1,25(OH)2D3 in vivo. Their unique feature is the ability to efficiently support some but not all l,25(OH)2D3-associated activities. Most commonly, the analogs display decreased potency in enhancing intestinal calcium absorption and/or bone mobilization. In some cases, the analogs have relatively high calcemic effects, but tend to produce even higher activities in other specific cells or tissues. The selectivity is not always cell- or tissue-specific, but can be gene- or process-specific within the same tissue. For example, several analogs [l,25(OH)2-22ene-24-dihomo-D3, l,25(OH)222-ene-24-trihomo-D3, and 1,25,28(OH)3D2] have been shown to induce the vitamin D–dependent calcium-binding protein in the intestine without increasing intestinal calcium transport [53,54]. Similarly, 20-epi-l,25(OH)2D3 was found to be 1000 times more potent than 1,25(OH)2D3 in inducing cell differentiation in a human leukemia cell line (HL-60), but it was equipotent in activating transcription of a vitamin D–responsive reporter construct in the same cells [55]. The remainder of this chapter

1452

ALEX J. BROWN AND EDUARDO A. SLATOPOLSKY

As a general concept, the actions of vitamin D compounds are determined by their interactions with five classes of proteins: the nuclear vitamin D receptor, transport proteins (primarily the serum vitamin D binding protein), metabolic enzymes (the vitamin D-24-hydroxylase and the 3-epimerase), cell surface membrane receptors, and the newly discovered intracellular vitamin D–binding proteins. The following sections discuss the structural modifications in vitamin D analogs that can affect these interactions and, ultimately, the biological profile of the compounds.

discusses potential mechanisms for the selective actions of vitamin D analogs, presenting examples when available.

II. THE IN VIVO SELECTIVITY OF VITAMIN D ANALOGS IS DETERMINED BY MULTIPLE PROTEIN INTERACTIONS The potential mechanisms through which selectivity could be achieved are summarized schematically in Fig. 2. These include altered systemic transport via DBP (1) or lipoproteins (2) that could influence cellular uptake (3); interaction with intracellular vitamin D binding proteins that target the compounds for metabolism or to the VDR; metabolism to active (4) or inactive (5) compounds; altered binding to the plasma membrane receptor (6); induction by the analog or its metabolites of an altered conformational change in the VDR (7) that could influence heterodimerization with RXR (8), binding to DNA (9) and subsequent recruitment of other components to the transcriptional initiation complex (10). The factors that contribute to selectivity of a few vitamin D analogs have been identified, and in many cases multiple factors may be involved.

1

A. Vitamin D Receptor (VDR) Most of the biological activities of vitamin D compounds are mediated by a nuclear receptor that binds 1,25(OH)2D3 with high affinity and specificity. The observation that some vitamin D analogs exerted selective actions in vivo (i.e., therapeutic activity similar to or higher than that of 1,25(OH)2D3, but with lower calcemic activity) led to early speculation that there could be multiple forms of the VDR that may recognize some but not all vitamin D compounds. It was proposed that the intestine and bone may express a form of the

Analog

DBP

7 mVD

2

ase

R

ID BP

3

OH

44 2

5 RNApol VDR

FIGURE 2

6

VDR

RXR

Potential sites of differential actions of 1,25(OH)2D3 and its analogs. Possible steps in the vitamin D activation pathways at which differences in vitamin D analog action could lead to selective activities in vivo are shown. The steps diagrammed include: (1) interactions with DBP or other serum proteins including lipoproteins, (2) cellular uptake and interaction with intracellular binding proteins, (3) intracelluar metabolism to active intermediary metabolites or (4) to inactive end-products, or (5) nuclear uptake and VDR binding (6) formation of the VDR-RXR complex, binding to the activated complex to DNA and formation of the preinitiation complex RNA polymerase II (RNApol) and (7) activation of the nongenomic pathway through a putative membrane vitamin D receptor (mVDR).

CHAPTER 82 Mechanisms for the Selective Actions of Vitamin D Analogs

VDR that discriminates against the noncalemic analogs. Presently there is no evidence that multiple forms of the VDR are responsible for the altered actions of certain analogs. However, the report by Crofts et al. [56] revealed that the VDR gene is driven by multiple promoters that incorporate different 5′ exons into the VDR mRNA, which could result in VDR proteins with different N-termini. The expression of the bulk of these transcripts was very low, however, and varied with cell type and tissue. At present, it is unclear whether cell-specific expression of these forms of the VDR account for the differential effects of vitamin D analogs observed in vivo. Vitamin D ligands induce a conformational change in the receptor upon binding. This is discussed in detail in several other chapters in this volume. The altered structures of vitamin D analogs likely contact the VDR differently than 1,25(OH)2D3, perhaps leading to subtle changes in the active VDR conformation that could potentially produce cell or gene-specific selectivity [7]. Steroid hormone receptors are complex molecules that interact with small molecules (ligands), many other proteins (kinases, motor proteins, other transcription factors, and components of the transcriptional initiation complex), and very specific DNA sequences. Not surprisingly, VDR and the members of this receptor superfamily contain multiple functional domains. The locations and functions of these domains in the VDR molecule are discussed in detail elsewhere in this volume (see Section II). The following sections present potential mechanisms by which differential interactions of vitamin D analogs with the VDR could lead to selective actions at the cellular or gene level. 1. LIGAND BINDING

Therapeutically useful vitamin D analogs must bind well to the VDR, at least in the clinically relevant target tissue(s) or cell(s). The key structural portion of the vitamin D compound for VDR binding is the A-ring containing the 1α-hydroxyl group. Until recently, all evidence indicated that the 1α-hydroxyl group was essential for VDR binding and activation. Modification of other parts of the molecule, however, appears capable of restoring biological activity in the absence of the 1α-hydroxyl group. Gardner and co-workers [57] reported that hexafluorination of carbons 26 and 27 can partially restore the HL-60 differentiating activity of several 1-desoxy analogs. The basis for these findings is unclear, but 1-hydroxylation of the analogs by these cells cannot be excluded. However, this explanation cannot account for the observation by Peleg et al. [58] that epimerization of carbon 20 can restore the transcriptional activity of 1β-hydroxymethyl-1,25(OH)2D3.

1453

Thus, other modifications appear to be capable of compensating for the lack of a functional 1α-hydroxyl group. Other portions of the molecule, notably the side chain, can be greatly modified with minimal effect on VDR binding. In fact, a recently reported analog lacking the entire side chain ((20S)-1α-hydroxy-2methylene-19-nor-bishomopregnacalciferol) is able to effectively suppress PTH in vivo with virtually no calcemic activity [59]. Other analogs that lack the C ring have been shown to retain high VDR affinity and activity [60]. However, as discussed below, the strength of the interaction between the VDR and its ligand is not the only determinant of VDR-mediated activation. The affinity of vitamin D compounds for the VDR is often a good predictor of their activity, but many exceptions can be found. Notably, several analogs with the 20-epi configuration have much higher activity than expected based on their equilibrium binding affinity (Kd). It is important to remember that the Kd is simply a ratio of the rate constants for dissociation and association (Kd = kdissoc/kassoc). Further kinetic analysis revealed that several (and perhaps all) 20-epi analogs have low rates of dissociation, implying that the rate of association is also low. The major effect of a slow dissociation is greater stabilization of the VDR; the impact of this is discussed below in more detail. This example illustrates the potential limitations of estimating the functional potential of an analog on the basis of its equilibrium binding. VDR affinity may also differ between cell types in vivo. Koike et al. [61] performed autoradiography on tissues from rats injected with varying doses of [3H]OCT. They observed saturation of the tritium localization to the parathyroid glands at lower concentrations than in other tissues. The mechanism for this lower apparent Kd for the VDR in the parathyroid gland is unclear. The VDR in the parathyroid glands is identical to that in other tissues, and therefore the higher apparent affinity is likely due to other factors such as the intracellular binding proteins that appear to play a role in the delivery of ligands to the VDR (see below). The ligand-binding domain (LBD) is located in the carboxy-terminal half of the VDR. At the C-terminal end of the LBD is the activation function-2 domain (AF-2). This terminal α-helix undergoes a dramatic conformational shift upon binding of the ligand, a crucial step in receptor activation. As discussed below and elsewhere in this volume (see Section II), several vitamin D analogs have been shown to induce a different VDR conformation than 1,25(OH)2D3. As a result, each of the following ligand-dependent processes may be differentially affected by vitamin D analogs.

1454 2. TRANSLOCATION OF THE VDR TO THE NUCLEUS

Ligand binding leads to nuclear translocation of the VDR from the cytoplasm to the nucleus, a process discussed in detail in Chapter 22. Two potential nuclear translocation motifs have been identified in the VDR molecule. One is a bipartate motif consisting of basic residues at each end of the sequence spanning amino acids 79 to 105 in the human VDR. The other is a basic sequence of seven amino acids [49–55] that is unique to the VDR. Point mutations in these sequences impair VDR translocation and activity. The VDR is rapidly shuttled to the nucleus along microtubules in response to 1,25(OH)2D3 in cultured fibroblasts and monocytes [62,63]. This was confirmed by following, in real time, the ligand-dependent translocation of a fusion protein of VDR with green fluorescent protein (GFP) [64]. Structural analysis revealed the requirement of the AF-2 domain of the VDR. It is likely that the VDR interacts with a motor protein that directs it to the nucleus. However, it is unclear whether this interaction is stimulated by a ligand-induced conformational change in the VDR or by activation of the process by ligand-dependent signaling via a membrane receptor (e.g. stimulation of kinases that activate the motor proteins). By either mechanism, vitamin D analogs could differentially affect the rate of nuclear uptake of the VDR and, therefore, its ability to control gene transcription. 3. VDR PHOSPHORYLATION

The VDR becomes hyperphosphorylated in response to the ligand [65] (see Chapter 13), but the role of phosphorylation on VDR activity in vivo is not fully understood. The major ligand-dependent phosphorylation sites have been identified as Ser51 and Ser208, although other potential phosphorylation sites may be present. As for stimulating nuclear translocation, it is unclear whether the ligand induces VDR phosphorylation by changing the conformation of the receptor to allow access to protein kinases or by stimulating specific kinase cascades through activation of the membrane receptor. The effect of phosphorylation on VDR activity appears to vary with the site modified. In vitro studies have shown that phosphorylation of Ser51 by protein kinase C (PKC) [66] leads to a decrease in VDR activity [67,68]. 1,25(OH)2D3 also has been shown to rapidly activate PKC through interaction of the membrane receptor, suggesting that activation of PKC may operate to attenuate the genomic activity of hormone. The VDR also contains a consensus site for protein kinase A [69], but the role of phosphorylation of this site is unclear. Initial studies report enhancement [70,71] and inhibition [69] of transcriptional activity. In support of the latter observation, co-expression of the catalytic subunit of PKA with the VDR in HeLa and Saos-2

ALEX J. BROWN AND EDUARDO A. SLATOPOLSKY

cells inhibited transactivation by 1,25(OH)2D3, but mutation of the PKA site on the VDR did not affect the inhibitory effect of PKA, suggesting that phosphorylation of other proteins, not the VDR, is responsible [72]. Phosphorylation of Ser208 by casein kinase-II has been shown to increase the transcriptional activity of the VDR [73]. The overall role of serine phosphorylation of the VDR appears to be stimulatory. Okadaic acid, an inhibitor of the serine/threonine protein phosphatase-1, enhances VDR-mediated transactivation [74,75]. Tyrosine phosphorylation of the VDR has also been reported. In skeletal muscle cells, 1,25(OH)2D3 has been shown to rapidly activate Src kinase leading to Src interaction with and phosphorylation of the VDR [76]. The effect of tyrosine phosphorylation on VDR activity is not clear, but it may allow interaction of the receptor with proteins involved in rapid signaling. Thus, phosphorylation of the VDR may regulate its activity. At present there is no evidence that vitamin D analogs differentially affect VDR phosphorylation. However, given that 1,25(OH)2D3 may alter the activity of the kinases responsible for VDR phosphorylation by stimulation of nongenomic pathways mediated by membrane receptors with clearly different ligand specificities, it seems likely that analogs would vary in their abilities to alter the phosphorylation state, and therefore activity, of the VDR independently of their actual affinity for the VDR. 4. HETERODIMERIZATION OF THE VDR WITH RXR

Like the other members of the steroid/thyroid receptor superfamily, the liganded VDR binds to DNA motifs (vitamin D response elements or VDREs) not as a monomer, but as part of a multiprotein complex. Binding to most known VDREs is greatly enhanced by the ligand-dependent interaction of the VDR with another nuclear transcription factor, the retinoid X receptor (RXR), but some VDREs may recognize VDR complex lacking RXR [77]. It has been postulated that VDR can also form homodimers as well, but this has been demonstrated only in vitro with much higher, and likely nonphysiological concentrations of VDR. The portion of the VDR that interacts with RXR involves the ligandbinding domain, and the interaction could be greatly affected by the type of conformational change induced by 1,25(OH)2D3 or its analogs or by steric hindrance by the ligand itself. Liu et al. reported that 20-epi1,25(OH)2D3 is more potent than 1,25(OH)2D3 in promoting VDR-RXR dimerization, in keeping with the higher activity of this analog [78]. Enhanced promotion by 20-epi analogs of VDR homodimerization has also been reported [79]. While the role of VDR homodimers is uncertain, this provides a potential mechanism for

CHAPTER 82 Mechanisms for the Selective Actions of Vitamin D Analogs

analog-enhanced activation of genes with VDREs that respond to the homodimer. 5. DNA BINDING

The ligand-activated VDR interacts with specific DNA motifs in the promoters of target genes termed vitamin D response elements (VDREs). Promoter analysis of genes regulated transcriptionally by 1,25(OH)2D3 has identified many VDREs with similar but distinct structure [80–82] (see Section II). The most common VDRE type, designated DR3, contains two direct repeats of 6 nucleotide bases separated by a 3-nucleotide spacer. The sequence of the hexameric repeats, or half-sites, varies considerably, but a general consensus sequence of AGGTCA has been established. In general, the RXR binds to the upstream half-site and the VDR to the downstream half-site. Another type of VDRE, the IP9, consists of two inverted palindromic sequences separated by 9 base pairs [83]. Vitamin D ligands have been shown to influence both the strength and specificity of the interaction of the VDR/RXR heterodimer with DNA. There is now considerable evidence that 1,25(OH)2D3 analogs with the 20-epi configuration induce a stronger interaction between the VDR and RXR in vitro, leading to a potentially more stable association of the heterodimer with VDREs. Studies by Peleg and co-workers [84] demonstrated that several 20-epi vitamin D analogs have higher activity than predicted from their VDR affinities. Using electrophoretic mobility shift assays (EMSA), which measure binding of proteins to small DNA molecules, these authors found increased binding of VDR/RXR to the osteocalcin VDRE in nuclear extracts from cells treated with 20-epi analogs than in extracts from cells treated with 1,25(OH)2D3. Furthermore, they provided evidence that the 20-epi analogs produced a conformation in the VDR that was distinct from that induced by 1,25(OH)2D3, using an in vitro protease-clipping assay in which recombinant, 35S-labeled VDR bound to various ligands is treated with a protease and the fragments are resolved by electrophoresis. The 20-epi analogs protected the VDR from proteolysis in a way that is different from that of 1,25(OH)2D3, suggesting an altered conformational state. As stated above, the C-terminus of the VDR (AF-2 domain) is essential for binding of other critical proteins and for transcriptional activity. Peleg et al. found that elimination of AF-2 domain prevented VDR activation by 1,25(OH)2D3 but not by the 20-epi analogs [85]. Again, these findings point to a distinct conformation induced by the 20-epi compounds that enhances binding of these other proteins, including RXR, in a manner that is independent of the AF-2 domain. These findings were confirmed by Ryhanen et al. [86]

1455

who reported that the 20-epi analogs MC1288 and KH1060 are more potent inducers of alkaline phosphatase and osteocalcin in osteoblastic cells than 1,25(OH)2D3 and that the effects were of longer duration. They demonstrated more stable DNA binding of VDR/RXR from nuclear extracts from analog-treated than from 1,25(OH)2D3-treated osteoblasts. Similar findings have been reported for other 20-epi analogs [87]. At least part of this enhanced stability induced by the 20-epi analogs may be attributed to stabilization of the VDR as discussed below. While these properties of the 20-epi analogs could endow them with greater potency in stimulating VDRmediated responses, they do not necessarily confer selectivity since the VDR is responsible for both antiproliferative and calcemic activities of vitamin D compounds. For example, despite the extremely potent immunosuppressive potential of KH1060 observed in vitro, the analog did not prevent renal allograft rejection or the development of hypercalcemia [88]. The vitamin D ligand affects not only the strength of VDR binding to the VDRE, but also its specificity. Analogs containing a 20-methyl group appear to preferentially activate VDR binding to the IP9 type of VDRE [89]. The analog EB1089 was shown to promote preferential interaction of the VDR with the IP9 type of VDRE, while CB1093 promoted binding preferentially to the DR3 type [90]. This could account for the differential effects of the two analogs on cell growth. While the two compounds are equipotent inhibitors of cell proliferation, CB1093 is 10 times more active in inducing apoptosis. Thus, an analog may determine the activity of the VDR on a gene-specific basis by altering the interaction with specific VDRE motifs. 6. RECRUITMENT OF OTHER COMPONENTS COMPLEX

OF THE INITIATION

Once the heterodimer is bound to DNA, it recruits other components of the transcriptional initiation complex (e.g. TFIIB [91]). In addition, steroid receptor coactivators interact with the VDR in a ligand-dependent fashion to further enhance transcriptional activity. The interaction of the VDR with these proteins, and hence its transcriptional activity, could also be influenced by the different conformation states induced by vitamin D analogs. Several vitamin D analogs have been shown to differ from 1,25(OH)2D3 in their recruitment of these coactivators. The higher potencies of 20-epi analogs in the induction of the cell cycle inhibitor p21 have been attributed, at least in part, to their enhanced abilities to recruit the DRIP coactivator complex to the initiation complex [92]. Again, this could be due to enhanced stabilization of the VDR. On the other hand, the 20-epi

1456 analogs were found to be equipotent to 1,25(OH)2D3 in the recruitment of the coactivators SRC-1 and GRIP-1 [93]. Differential effects on ligand-induced coactivator binding to VDR have been reported for 1,25(OH)2-22oxa-D3 (OCT) as well [94]. While 1,25(OH)2D3 could promote interaction of the VDR with SRC-1, TIF2 and AIB1, OCT supported only interaction with TIF2. Issa et al. investigated 12 analogs for their abilities to mediate recruitment of the coactivators GRIP1 and RAC3 in vitro [95]. There was considerable ligand-specific variability in the strength of the VDR-coactivator interaction that was not correlated with the affinity of the ligands for the VDR. Altered recruitment of coactivators by analogactivated VDR could provide a mechanism for genespecific effects. Full activation (or repression) of gene transcription by the VDR may require the full complement of transcriptional components. The AF-2 domain (helix 12) of the VDR is known to be critical for interaction with many of the components of the transcriptional complex, and mutations in this region abrogate the activity of 1,25(OH)2D3 by influencing both heterodimerization with RXR and the coactivator binding [93,96]. Although these mutations do not affect VDR-RXR heterodimerization induced by the 20-epi-1,25(OH)2D3, they hamper SRC-1 binding and reduce transcriptional activity. 7. LIGAND-DEPENDENT VDR REGULATION

The liganded VDR has been shown to be much more resistant to intracellular degradation, as discussed above. Masuyama and MacDonald [97] demonstrated that the VDR is degraded primarily by the proteasome complex, a large macromolecular structure that degrades most cellular proteins. Inhibition of proteasome activity prevented VDR turnover and increased VDR content. However, control of VDR levels by vitamin D compounds appears to be complex, since ligands may not only stabilize the VDR, but also target the receptor for degradation. There is evidence that ligands for steroid hormone receptors that dissociate more slowly have been shown to better protect the receptor from degradation [98]. Vitamin D analogs have differing abilities to stabilize the VDR, and in a few cases, this has been attributed to altered dissociation rates. The best-studied examples are the 20-epi compounds. Van den Bemd et al. [99] reported that KH1060 was much more effective than 1,25(OH)2D3 in slowing the rate of VDR degradation in osteoblastic cells. Consistent with this was the report that while nuclear extracts of cells treated with the 20-epi analog CB1093 showed more VDRE binding than nuclear extracts from 1,25(OH)2D3-treated cells, binding of the two vitamin D compounds to VDR

ALEX J. BROWN AND EDUARDO A. SLATOPOLSKY

in vitro produced equivalent results in the EMSA. Thus, it appears that the higher potency of 20-epi analogs cannot be attributed directly to the distinct conformational change in the VDR, but rather to changes observed only in whole cells, most likely through the increased ability to stabilize the VDR. Evidence that slowly dissociating vitamin D ligands are superior in stabilizing the VDR was reported by Peleg et al. [100]. They reported that the analog 1β-hydroxymethyl-3-epi-1,25(OH)2D3 is more sensitive to changes in the AF-2 domain, tends to dissociate more readily from the VDR, and is less potent than 1,25(OH)2D3. Modifying this analog to contain 1β-hydroxymethyl and dimethyl groups at carbons 26 and 27 enhanced transcriptional activity, increased the stability of the VDR to proteolysis, and reduced the dissociation rate. Addition of a 16-ene to the molecule enhanced these properties even further. Stabilization leads to an increase in VDR content in the cell, and this would largely explain the greater activity of certain analogs, notably those with the 20-epi configuration. Clearly, the degree of VDR up-regulation would be related to the activity of the degradative pathway in the cell. In cells with a very high rate of VDR turnover, ligand-dependent stabilization would produce a larger increase in VDR content. Thus, analogs with slow dissociation rates would be expected to be more active in these cells. This could provide a mechanism for cell-selective actions of vitamin D analogs. A further complexity in VDR regulation was revealed in the recent report by Masuyama and MacDonald [97]. The VDR binds in a ligand-dependent manner to SUG1, a component of the proteasome complex. Overexpression of SUG1 increased VDR degradation and reduced 1,25(OH)2D3-mediated activity. These results suggest that vitamin D ligand, by promoting interaction with SUG1, may target the VDR for degradation. The ligand specificity for SUG1 binding has not been rigorously examined, but since this interaction requires the AF-2 domain of the VDR [97], it is likely that analogs with altered interaction with this domain would differentially promote SUG1 binding. Thus, there appear to be counteracting effects of vitamin D compounds on VDR regulation, and the relative abilities of vitamin D compounds to stabilize the VDR or target the receptor for degradation may determine the VDR content, and hence biological activity, in a specific cell.

B. Interaction with the Serum Vitamin D Binding Protein (DBP) and Other Transporters Vitamin D compounds are relatively hydrophobic with little solubility in aqueous solution and must be

1457

CHAPTER 82 Mechanisms for the Selective Actions of Vitamin D Analogs

Plasma levels, ng/ml

10 8 1,25D3

6 4 2 OCT 0

VDR complexes, pmol/mg DNA

transported in vivo attached to proteins. The major carrier of vitamin D compounds in the circulation is the serum vitamin D–binding protein (DBP). This protein binds all of the natural vitamin D metabolites with high affinity and circulates at a concentration of 5 µM, compared to picomolar levels of 1,25(OH)2D3 (see Chapters 8 and 9). Other abundant proteins, such as albumin and lipoproteins, may bind lesser amounts of the natural vitamin D compounds, but with much lower affinity. However, they may play a role in transporting analogs that bind poorly to DBP. In the case of 1,25(OH)2D3, over 99% is protein bound, mostly to DBP. The structural elements of vitamin D compounds that affect DBP binding are different than for interaction with the VDR. The 1α-hydroxyl group is not required for DBP binding, and 1-desoxy compounds have a higher affinity than their 1-hydroxylated counterparts. On the other hand, DBP affinity is greatly affected by changes in the side chain, the most common site of modification in the therapeutically important vitamin D analogs. This altered DBP binding has been shown to play a critical role in the selectivity of several vitamin D analogs. DBP performs two major functions with respect to vitamin D compounds: it enhances their circulating half-life and, importantly in the case of 1,25(OH)2D3 and its analogs, it decreases tissue accessibility. In this way DBP acts as a reservoir and plays a key role in guarding against vitamin D intoxication. Thus, the DBP affinity will affect both the clearance rate and tissue uptake of vitamin D analogs. Altered pharmacokinetics plays a central role in determining the unique biological profile of several analogs. The side chain modifications of most of the vitamin D compounds in development reduce DBP affinity and cause the analog to be rapidly cleared or poorly absorbed into the circulation. The best studied example of an analog that exerts its selectivity through this pharmacokinetic mechanism is OCT. The affinity of OCT for DBP is about 500 times lower than that of 1,25(OH)2D3 [101]. As a result, this analog is cleared from the circulation more rapidly than 1,25(OH)2D3 and achieves lower peak levels following injection as shown in Fig. 3, upper panel [102,103]. Despite the lower peak levels of OCT in the blood, the peak levels of the analog were greater than those of 1,25(OH)2D3 in most target tissues, including the intestine as shown in Fig. 3, lower panel. This increased tissue content was short-lived, falling rapidly as the analog was cleared from the circulation. This “pulse” of OCT in the intestine elicited only a transient increase in calcium transport which fell to basal levels soon after OCT disappeared from the circulation (Fig. 4, upper panel) [103].

OCT 3

2

1

0

1,25D3

0

1

2

3 4 5 Time post-injection, h

6

7

8

FIGURE 3

Plasma levels and intestinal localization of [3H]1, 25(OH)2D3 (1,25D3) and [3H]OCT following a single intraperitoneal injection. Upper Panel: Plasma levels of HPLC-purified [3H]1,25(OH)2D3 and [3H]OCT at various times postinjection. Lower Panel: Duodenal VDR content of [3H]1,25(OH)2D3 and [3H]OCT at various time postinjection. From Brown et al. [103].

The effects of OCT on bone were also short-lived compared to those of 1,25(OH)2D3 (Fig. 4, lower panel). Further evidence for a pharmacokinetics being responsible for the low calcemic activity of OCT were the sustained increases in intestinal calcium transport and bone mobilization with constant infusion of the analog, and the rapid return (within 24 hours) to baseline with cessation of the infusion [103]. OCT has been approved in Japan for treatment of secondary hyperparathyroidism in renal failure patients, and is in development for other applications including cancer and psoriasis. In parathyroid glands, OCT treatment produces a prolonged suppression of parathyroid hormone (PTH) gene transcription. The initial findings of Kobayashi et al. [102] indicated that injected [3H]OCT rapidly disappears from the parathyroid glands once the analog is cleared from the circulation, similar to the time course observed in the intestine and other tissues. However, the recent autoradiography study of Koike et al. [104] found that [3H]OCT persists in parathyroid cell nuclei. Furthermore, the nuclear binding of [3H]OCT in the parathyroid glands

1458

ALEX J. BROWN AND EDUARDO A. SLATOPOLSKY

Ca absorbed, % control

400 1,25D3

300

200 OCT 100

Serum calcium, mg/dl

0 1,25D3

8 7 6 OCT 5 Vehicle 4

0

8

16 24 32 Hours after injection

40

48

FIGURE 4

Effects of OCT and 1,25(OH)2D3 on intestinal calcium transport and bone mobilization. Vitamin D–deficient rats were fed a 0.02% Ca diet for two days and then injected IP with 250 ng of 1,25(OH)2D3 or OCT or vehicle. Intestinal calcium absorption was determined by the isolated duodenal loop method using 45Ca. Bone calcium mobilization was estimated by the increase in serum calcium. From Brown et al. [103].

was saturated at a lower dose than the other tissues examined [61]. These findings would support a pharmacokinetic mechanism for the selectivity of OCT in the parathyroid glands. This rapidly cleared analog exploits the differences in the half-lives of the therapeutic response (prolonged PTH suppression) and the toxic responses (short-lived stimulation of intestinal calcium transport and bone mobilization). The molecular basis for the different durations of these responses is unclear. Enhancement of intestinal calcium transport and bone mobilization appear to require continuous exposure to vitamin D compounds, possibly because the inducible proteins mediating these response have relatively short half-lives. Alternatively, the transient responses could be due to a cessation of stimulation by OCT of the nongenomic pathway of vitamin D action as discussed below. The relatively long-lasting effects of all vitamin D compounds, including OCT, on PTH gene expression may be attributable to their slower metabolism in parathyroid cells as discussed in the next section.

The same pharmacokinetic mechanism may explain the effectiveness of OCT in the treatment of cancer. Abe et al. showed that OCT could effectively inhibit the growth of breast cancer cells in mice without producing hypercalcemia [30]. The short half-life of OCT did not prevent its beneficial antiproliferative activity. It is likely that this pharmacokinetic mechanism for analog selectivity will apply, with varying degrees, to all analogs with low DBP affinity. This mechanism may also apply to VDR ligands with nonvitamin D structures. Boehm et al. [105] screened a chemical library and found a nonsteroidal biphenyl compound that could activate the VDR. This compound was very potent in inhibiting cancer cell growth in vitro, but had very low calcemic activity in vivo. Further analysis showed that the new vitamin D ligand did not bind to DBP. Therefore, it is likely that the compound is rapidly cleared and unable to sustain calcemic effects in the intestine and bone. On the other hand, analogs with higher DBP affinity than 1,25(OH)2D3 tend to have longer circulating half-lives and less accessibility to target tissues. An example is ED-71 [2-(3-hydroxypropoxy)1,25(OH)2D3], which is in development for osteoporosis [106] (see Chapter 85). This analog produces a smaller stimulation of bone mobilization and intestinal calcium transport than 1,25(OH)2D3 due to its decreased uptake, but its calcemic effects are more prolonged due to its longer half-life. These altered pharmacokinetics may be responsible, in part, for the ability of ED-71 (but not 1,25(OH)2D3) to produce net bone formation in ovariectomized rats. It is important to note that there is a limit to the therapeutically beneficial effects of low DBP affinity. Analogs that bind extremely poorly to DBP may have little activity in vivo. An example of this is the analog 1,25(OH)2-24-trihomo-22-ene-D3, which we tested for its suitability for secondary hyperparathyroidism (unpublished data). This compound was as potent as 1,25(OH)2D3 in suppressing PTH in cell culture and had virtually no calcemic activity in vivo. However, the analog was virtually inactive in suppressing PTH in vivo. Subsequent characterization showed that the DBP affinity of this compound was more than 10,000 times lower than that of 1,25(OH)2D3. It is likely that the analog did not achieve effective concentrations in the blood and was completely inactive in vivo. Therefore, DBP affinity is a critical parameter that requires testing early in the development of therapeutic vitamin D analogs. A second consideration with respect to DBP affinity is its influence on analog activity in cells cultured in serum-containing medium. Analogs with lower DBP

CHAPTER 82 Mechanisms for the Selective Actions of Vitamin D Analogs

affinity than 1,25(OH)2D3 will be taken up more readily in the presence of serum and will be more active under these conditions. This is often the explanation for why some analogs have higher potency than predicted by their VDR affinity. A number of studies have documented this serum effect on the relative activities of 1,25(OH)2D3 and its analogs [107–109]. Correct interpretation of the potencies of vitamin D analogs will require the determination of DBP affinity. Another consequence of low DBP binding is the freedom of the analogs to associate with other serum proteins. This could potentially alter the delivery of vitamin D analogs to various target tissues. Currently, there is considerable disagreement as to the degree of binding of vitamin D metabolites and analogs to the various potential carriers. Okano et al. [101] and Kobayashi et al. [110] reported differences in the association of 22-oxa-l,25(OH)2D3 and 1,25-(OH)2D3 with serum lipoproteins. When 22-oxa-l,25(OH)2D3 was mixed in vitro with human plasma, it bound almost exclusively (99%) with the lipoprotein fraction, mainly with chylomicrons and low density lipoproteins (LDL), whereas less 1,25(OH)2D3 was bound to lipoprotein (60%). A similar study by Teramoto et al. [111] reported much less binding of these two compounds to lipoprotein fractions. The significance of the differences in lipoprotein binding in the selectivity of 22-oxa-l,25(OH)2D3 is unclear, as the role of carrier proteins in the delivery of vitamin D compounds has received only slight attention [112–115]. At present, a comparison of binding of 22-oxa-1,25(OH)2D3 and 1,25(OH)2D3 to lipoproteins has been done only in human plasma and only in vitro. Therefore, the relevance to ligand selectivity observed in experimental animals is not clear. Better evidence for a mechanism for selectivity involving carrier protein distribution would be the demonstration of differences in the binding of vitamin D analogs and 1,25(OH)2D3 to rat lipoproteins in vivo.

C. Cellular Metabolism by the Vitamin D-24-hydroxylase and Other Enzymes Target cell metabolism has been shown to play an important role in several steroid hormone systems [116]. In mineralocorticoid-responsive tissues, the receptor that mediates the action of aldosterone binds glucocorticoids with equal affinity. Selectivity is achieved in these tissues by efficient degradation of glucocorticoids, rendering them unavailable to the receptor. Alternatively, cells can convert hormones to more active metabolites as in the conversion of T4 to T3 or estrone (E1) to estradiol (E2).

1459

Vitamin D compounds are metabolized primarily, but not exclusively, by the vitamin D-24-hydroxylase, which catalyzes a series of oxidations at carbons 24 and 23 in the side chain of the molecules (see Chapters 6 and 80). Oxidative cleavage between C23 and C24 yields calcitroic acid, probably via an aldehyde intermediate. The 24-hydroxylase is highly induced by 1,25(OH)2D3 and its analogs, and it is generally believed that side chain metabolism has an attenuating effect on vitamin D action. Intermediary metabolites of 1,25(OH)2D3 produced by this pathway have lower VDR affinity and are usually less active than the parent, but exceptions have been noted and are discussed below. Since the most common site of modification of vitamin D analogs is in the side chain, it would be expected that their metabolism would differ from that of 1,25(OH)2D3. In fact, differences in the rates of catabolism and the end-products of metabolism have been shown to account, at least in part, for the unique properties of several vitamin D analogs. Analogs that are quickly degraded within target cells would have lower overall biological activity. This effect was illustrated in a study by Zhao et al. in cell culture [117]. They examined the antiproliferative effects of 1,25(OH)2D3 and several analogs on cultured MCF-7 breast cancer cells (Fig. 5). The ED50 values for each analog were reduced by cotreatment with ketoconazole, a cytochrome P450 inhibitor that blocks 24-hydroxylase activity. The higher ED50 in the absence of the inhibitor was due to the influence of target cell catabolism of the vitamin D compounds. The degree of the reduction of the ED50 by ketoconazole cotreatment varied for each analog, likely due to the differential rates of catabolism of the compounds. Other examples of analogs whose activity may be influenced by their catabolic rate include EB1089 and 20-epi-1,25(OH)2D3. These compounds have higher biological activity than 1,25(OH)2D3 despite similar VDR affinities. The disparity has been attributed, at least in part, to their very slow rates of intracellular catabolism [108,118–120]. The recent report of Lin et al. showed that EB1089 produced the same changes in gene expression as 1,25(OH)2D3 in squamous carcinoma cells, but that the effects were more prolonged [121]. The difference was abolished when 1,25(OH)2D3 was co-incubated with ketoconazole. It is important to note that rapid disappearance of an analog in the target cell could produce the same differential effects as accelerated clearance as described for analogs with low DBP affinity. The selectivity of vitamin D analogs in vivo may also be attributed to cell-specific differences in the rates of catabolism of an analog. For example, OCT

1460

[3H]Thymidine incorporation (% control)

ALEX J. BROWN AND EDUARDO A. SLATOPOLSKY

KH1060 100 + keto 80

60

1,25D3

KH1060 + keto

40

Analog

− keto

1,25(OH)2D3

1.5

KH1060

1.6

ZXY 835

2.5

CD99

1.2

RO23-6010

1.8

20 1,25D3 + keto 10−12

10−11

10−10 10−9 Analog, M

10−8

10−7

FIGURE 5

Influence of catabolism on the biological activities of vitamin D analogs. Left Panel: The antiproliferative effects of 1,25(OH)2D3 and its analog KH1060 on MCF-7 breast cancer cells was measured in the absence or presence of the cytochrome P450 inhibitor ketoconazole. Blocking the catabolic enzyme (24-hydroxylase) enhanced the antiproliferative activities. Right Panel: The ratio of the ED50 in the presence (+keto) versus absence (−keto) of ketoconazole. The activity is enhanced by ketoconazole in all cases, but the variation between analogs likely reflects differences in their rates of catabolism. From Zhao et al. 1996 [117].

appears to be degraded at the same rate as 1,25(OH)2D3 in parathyroid cells [122], more rapidly in keratinocytes [123] and more slowly by monocytes [124]. The rates of catabolism are consistent with the similar activities of OCT and 1,25(OH)2D3 in parathyroid cells [122] and lower activity of OCT in keratinocytes [123]. The explanation for this cell-specific catabolism is unclear, but certainly deserves further study since exploitation of the catabolic differences between vitamin D compounds could produce highly selective therapeutic agents. There are also differences in the amount of catabolic activity in various cell types. In the study by Zhao et al. described above [117], ketoconazole had no effect on the activities of the analogs in the osteoblastic cell line MG-63, indicating that the catabolic activity in these cells is very low. This variation in catabolic activity could lead to cell-specific differences in the actions of 1,25(OH)2D3 and its analogs. In cells with high catabolic activity, analogs that are better substrates for the 24-hydroxylase would be rapidly degraded and less active, whereas ligands that are poor substrates would be more active. By this mechanism alone, each analog could have a unique biological profile. Structural modifications in some vitamin D analogs prevent the completion of the side chain cleavage pathway and yield stable active intermediates that may accumulate in target cells. One example is 1,25(OH)2-16-ene-D3. This analog is catabolized by the

24-hydroxylase to the 24-oxo intermediate [125], but further oxidation occurs very slowly. The 1,25(OH)216-ene-24-oxo-D3 retains significant biological activity in vitro [126] and in vivo [45]. In a murine model of autoimmune encephalitis, 1,25(OH)2-16-ene-24-oxo-D3 was as active as 1,25(OH)2-16-ene-D3 and 1,25(OH)2D3 in suppressing the immune response, but it was substantially less calcemic. A similar interruption in side chain catabolism was noted for 20-epi-1,25(OH)2D3. The 24-oxo metabolite of this analog accumulates in target cells and may be responsible in part for the high biological activity of the parent 20-epi compound [108]. KH1060, which has been shown to be much more active than 1,25(OH)2D3 in slowing cell proliferation, is rapidly metabolized to 24- and 26-hydroxylated metabolites that are also more potent than 1,25(OH)2D3 [127]. These two metabolites have potencies similar to the parent KH1060 in the induction of osteocalcin expression by osteoblast-like cells, in enhancing VDR stability by increasing resistance to proteolysis, and in binding of the VDR to the osteocalcin VDRE [128]. Thus, the 24- and 26-hydroxy metabolites contribute significantly to the overall effects of the parent KH1060. Komuro et al. [129] found that 1,25(OH)2-26,27-F6-D3 is converted to 1,23,25(OH)3-26,27-F6-D3, which is resistant to further metabolism and retains significant activity. Other metabolic enzymes may play a key role in analog selectivity. Reddy and his colleagues recently

CHAPTER 82 Mechanisms for the Selective Actions of Vitamin D Analogs

discovered that 1,25(OH)2D3 can undergo epimerization at carbon 3 to change the 3β-hydroxyl to the 3α configuration [130]. The enzyme responsible for 3-epimerization is not ubiquitously expressed. This reaction has been documented in keratinocytes, parathyroid cells, osteoblastic cells, and colon cancer (Caco-2) cells, but not by the kidney or myeloid leukemia (HL-60) cells. In parathyroid cells, 3-epi-1,25(OH)2D3 has nearly the same potency as 1,25(OH)2D3 in suppressing PTH, but is catabolized more slowly by the 24-hydroxylase and may accumulate in these cells [Brown, 1999 #22670]. The slower inactivation, if it occurs in parathyroid glands in vivo, could be responsible, at least in part, for the prolonged effects of 1,25(OH)2D3 on PTH secretion. An example of cell-specific metabolism was reported by Kamao et al. [131]. OCT was metabolized mainly by the 24-hydroxylase in intestinal (Caco-2) cells and renal (LLC-PK1) cells, whereas in osteoblastic (UMR-106) cells, OCT was metabolized by the 3-epimerization and 25-dehydration pathways. Differences between 1,25(OH)2D3 and its analogs in their relative rates of metabolism through these pathways could lead to preferential action in one cell over the other. The structural requirements for substrates of the enzyme responsible for 3-epimerization is under investigation. Reddy et al. found that the rate of 3-epimerization of 1,25(OH)2-16-ene-23-yne-20-epi-D3 10 times faster than for 1,25(OH)2-16-ene-23-yne-D3 [132]. It is not clear if the 20-epi modification facilitates 3-epimerization of other analogs as well, but structural alterations that facilitate the conversion to the 3-diastereomer may further slow the rate of metabolism in cells with high 3-epimerase activity and therefore enhance the analog action in a cell-specific manner.

D. Nongenomic Activity Mediated by a Cell-surface Receptor Vitamin D compounds can activate a number of signaling pathways, perhaps by activating a distinct receptor at the plasma membrane [1,82,133] (see Chapter 23). These effects are observed within seconds to minutes following exposure to 1,25(OH)2D3, too quickly to involve changes in gene transcription. One of the best characterized rapid actions is the ex vivo stimulation of calcium movement across the intestinal epithelium, a process termed transcaltachia. Nemere and co-workers perfused chick duodenum with 1,25(OH)2D3 and observed increased calcium movement from the lumen to the perfusate within minutes [134]. Others have noted rapid stimulation by 1,25(OH)2D3 of PKC activation and translocation, phosphate fluxes, alkaline phosphatase,

1461

cGMP, and phosphoinositide metabolism in cell culture. 1,25(OH)2D3 increases cytosolic calcium within minutes in a number of cell types, including osteoblasts, parathyroid cells, human myeloid leukemia cells, enterocytes, and myocytes. 1,25(OH)2D3 has been shown to rapidly activate mitogen-activated protein kinase (MAPK) [135,136] and stimulate opening of chloride channels [137]. The nature of the receptor(s) that mediates these rapid actions is unclear. Early evidence suggested that the rapid effects of calcitriol may be mediated by the VDR. 1,25(OH)2D3 cannot rapidly increase cGMP in skin fibroblasts from patients with vitamin D–resistance rickets that lack a functional VDR. In addition, calcitriol has also been shown to stimulate phosphoinositide metabolism in isolated enterocytes from adult rats, but not in those from 16-day-old rats that do not express intestinal vitamin D receptors [138]. However, the finding that 1,25(OH)2D3 cannot stimulate transcaltachia in vitamin D–deficient chick duodenum suggested that this process, and perhaps other nongenomic actions, requires vitamin D–inducible gene products. Evidence now indicates that nongenomic responses may be carried out by a distinct receptor [139–141]. The ligand specificity for the rapid actions is different than that for the genomic response [142–144]. Stimulation of the nongenomic actions does not require the 1α-hydroxyl group; 25(OH)-23-yne-D3 and its 16-ene counterpart can stimulate transcaltachia, but do not bind the VDR [142]. On the other hand, 1,25(OH)-16-ene-23-yne-2D3 and calcipotriol, which have high VDR affinity, do not produce transcaltachia [142]. Further evidence for the distinct nature of the nuclear and membrane receptors is the finding that OCT has genomic, but not nongenomic activity, in rat osteoblastic cells [145]. More recent studies have demonstrated that vitamin D analogs that are locked in the 6-cis configuration act only as agonists for the putative membrane receptor and do not bind to the VDR [143]. The functions of the nongenomic actions of 1,25(OH)2D3 in most cell types and their relevance in vivo are still unclear. In the intestine, it is well established that exposure of the basolateral membrane to 1,25(OH)2D3 stimulates transcaltachia [142]. In chondrocytes, nongenomic actions of 1,25(OH)2D3 and 24,25(OH)2D3 alter membrane lipid turnover, prostaglandin production, and protease activity that lead to modification of bone matrix and calcification [146]. In other cells, the nongenomic events have been proposed to modulate the genomic actions of 1,25(OH)2D3 [147,148], but this remains controversial. Numerous studies have presented evidence that the nongenomic actions may not be critical for 1,25(OH)2D3-mediated gene activation [73,149–152] or inhibition of cell

1462 proliferation [151,153]. However, nongenomic stimulation of protein kinases could potentially influence the VDR-mediated effects of 1,25(OH)2D3. As introduced earlier, genomic activity of the VDR is decreased by PKC, which is stimulated by 1,25(OH)2D3 activation of the membrane receptor. Thus, differential effects of vitamin D analogs on the membrane receptor could influence its genomic actions as well as the VDR-independent effects of the signaling pathways. The low calcemic activity of two analogs, 1,25(OH)216-ene-23-yne-D3 and calcipotriol, could be due to their inability to stimulate transcaltachia in the intestine [142]. Several other analogs, including 1,25,28-trihydroxyvitamin D2, 1,25(OH)2-24-dihomo-22-ene-D3, 1,25(OH)224-trihomo-22-ene-D3, that stimulate genomic responses (calbindin D9k) in the intestine but not calcium transport [53,54] may lack nongenomic activity, but these analogs have not been tested. On the other hand, OCT has been shown to stimulate transcaltachia in the intestine [145]. The inability of OCT to sustain a high rate of calcium transport after its disappearance from the circulation (Fig. 3) could be due to the loss of stimulation of the nongenomic pathway(s). The role of the nongenomic pathway in bone is even less clear. As described above, the nongenomic effects of 1,25(OH)2D3 and 24,25(OH)2D3 may play a role in bone formation [146]. 1,25(OH)2D3 is known to stimulate both the production and phosphorylation of an osteoblast-derived protein, osteopontin. Safran et al. [154] found that the analog 25(OH)-16-ene23-yne-D3, which is known to stimulate only the nongenomic pathway (no VDR affinity), cannot induce osteopontin but does promote its phosphorylation. Clearly, non-genomic activities could play a role in the vitamin D effects on bone formation. On the other hand, the role of nongenomic effects on bone resorption is not known. It is possible that analogs that are incapable of activating or sustaining a nongenomic response are unable to efficiently induce or maintain bone mobilization. A plasma membrane protein from chick duodenum was isolated that binds vitamin D analogs with affinities that correlate with their activation of transcaltachia [139]. Antibodies raised against the protein recognize primarily a 66 kDa peptide by immunoblot analysis [140]. Most importantly, this antibody inhibits the ability of 1,25(OH)2D3 to stimulate PKC activity in chondrocyte membranes. However, this protein has not been identified or cloned. Another membrane receptor has been identified in ROS 24/1 rat osteosarcoma cells that do not express the VDR [141]. A plasmalemma protein of approximately 36 kDa was shown to be covalently cross-linked to 3-bromoacetylated 1,25(OH)2-[14C]D3. The labeled

ALEX J. BROWN AND EDUARDO A. SLATOPOLSKY

protein was isolated by isoelectric focusing, and a partial sequence identified it as annexin II. Polyclonal antibodies to annexin II decreased binding of 1,25(OH)2-[14C]D3 and blocked the increase in cytosolic calcium by 1,25(OH)2D3 [155]. The role of annexin II as a recpeptor for vitamin D analogs is unclear. The difference in the sizes of the chick duodenal membrane receptor and annexin II suggests that there may be multiple membrane receptors for 1,25(OH)2D3 that are structurally and functionally distinct. The relative ligand specificities and tissue localization of the receptors are not yet known. Despite questions concerning the physiologic relevance of the nongenomic pathways, these membrane-binding sites could potentially offer new pharmacologic targets that could produce cell-specific or even process-specific effects of vitamin D analogs.

E. Intracellular Binding Proteins Proteins of the heat shock-70 family are now known to bind vitamin D compounds and appear to play a determining role in their metabolism and actions [156–163]. These proteins, discovered during the investigation of the vitamin D resistance observed in new world primates, were found to be overexpressed in a marmoset B-lymphoblast cell line B95-8. Unfractionated B95-8 cytosol containing a mixture of intracellular vitamin D–binding proteins (IDBPs) was used in a competitive binding assay with [3H]25(OH)D3 to assess binding affinities of various vitamin D ligands. 25(OH)D3 and 25(OH)D2 bound most avidly. 1,25(OH)2D3 and 1,25(OH)2D2 had about 1/3 the affinity of 25(OH)D3, and 1α (OH)D3, which lacks a 25-hydroxyl group, did not bind at all [156,160]. Overexpression of IDBP-1 stimulated the transactivation by 1,25(OH)2D3 [161] and enhanced the conversion of 25(OH)D3 to 1,25(OH)2D3 [162]. More recently, it was reported that IDBP-1 predominantly directs delivery of ligands to the VDR, while IDBP-3 facilitates mitochondrial delivery of substrates via an N-terminal targeting signal for the inner mitochondrial membrane where it interacts directly with the vitamin D-1α-hydroxylase [163]. It would seem likely the mitochondrial delivery by IDBP-3 of active vitamin D compounds (i.e., VDR ligands) would accelerate their metabolism and inactivation. A more thorough discussion of the IDBP and their roles in vitamin D metabolism and action can be found in Chapter 21. Only a few vitamin D analogs have been tested for their binding to the IDBPs. Calcipotriol, EB1089, and KH1060 were found to have no apparent binding affinity for the IDBPs in unfractionated supernatants from B95-8 cells [159] when competing with [3H]25(OH)D3

CHAPTER 82 Mechanisms for the Selective Actions of Vitamin D Analogs

as ligand. These three analogs are modified in their side chains, indicating the importance of this portion of the vitamin D structure in binding to the IDBPs. In addition, the initial studies used unfractionated cytosol as a source of a mixture of IDBPs; the interaction of the analogs with the individual IDBPs has not been reported. The finding that distinct IDBPs may be involved in enhancement of transactivation and the facilitation of metabolism suggest a complex means for potentially affecting analog selectivity. High affinity binding to IDBP-1 would appear to increase the potency, whereas binding to IDBP-3 would facilitate metabolic inactivation. Relative differences in the interactions of analogs in binding to the IDBPs along with cell-specific differences in the relative expressions of the IDBPs could contribute to the unique biological profiles of the analogs. Future studies are required to assess the impact of the IDBPs on the actions of 1,25(OH)2D3 and its analogs.

III. CONCLUDING REMARKS Discovery of the nonclassical actions of 1,25(OH)2D3 has spawned new interest in vitamin D therapy for the treatment of hyperproliferative diseases such as cancer and psoriasis, and endocrine disorders including hyperparathyroidism, as well as for immunosuppression to prevent autoimmunity and transplant rejection. Hypercalcemia has limited or precluded the use of 1,25(OH)2D3 for most of these applications. Several new vitamin D analogs that retain the desired therapeutic activity but with less toxic (calcemic) side effects have now reached the marketplace. Calcipotriol is now available for psoriasis, and three analogs, 19-nor1,25(OH)2D2, OCT, and 1α(OH)D2, have been approved for secondary hyperparathyroidism. Many other analogs are in development for other applications. Perhaps the most critical of these is cancer. The current approach for developing therapeutic vitamin D analogs involves brute force testing of hundreds if not thousands of compounds. A more thorough understanding of the factors that determine the biological profile of an analog could greatly reduce development time. Our current knowledge of vitamin D physiology and biochemistry indicates that the overall activities of vitamin D compounds are determined by the combined interactions with several key proteins: the nuclear vitamin D receptor, the serum vitamin D binding protein, the 24-hydroxylase, and a membrane receptor. A major goal is to understand how the integrated interactions lead to the in vivo actions of the analogs. This requires an understanding of both the

1463

structure-activity relationships for each of these interactions and the role that these proteins play in the desired and undesired activities. As our knowledge of these two areas increases, so does the reality of designing vitamin D analogs with precise target specificity.

References 1. Norman AW, Bishop JE, Bula CM, Olivera CJ, Mizwicki MT, Zanello LP, Ishida H, Okamura WH 2002 Molecular tools for study of genomic and rapid signal transduction responses by 1α,25-(OH)2D3. Steroids 67:457–466. 2. Slatopolsky E, Weerts C, Thielan J, Horst R, Harter H, Martin KJ 1984 Marked suppression of secondary hyperparathyroidism by intravenous administration of 1,25-dihydroxy-cholecalciferol in uremic patients. J Clin Invest 74:2136–2243. 3. Smith EL, Pincus SH, Donovan L, Holick MF 1988 A novel approach for the evaluation and treatment of psoriasis. Oral or topical use of 1,25-dihydroxyvitamin D3 can be a safe and effective therapy for psoriasis. J Am Acad Dermatol 19:516–528. 4. Drezner MK, Lyles KW, Haussler MR, Harrelson JM 1980 Evaluation of a role for 1,25-dihydroxyvitamin D3 in the pathogenesis and treatment of X-linked hypophosphatemic rickets and osteomalacia. J Clin Invest 66:1020–1032. 5. Studzinski GP, McLane JA, Uskokovic MR 1993 Signaling pathways for vitamin D–induced differentiation: implications for therapy of proliferative and neoplastic diseases. Crit Rev Eukaryot Gene Expr 3:279–312. 6. Brown AJ 2001 Therapeutic uses of vitamin D analogs. Am J Kidney Dis 38:S3–S19. 7. Pike JW, Yamamoto H, Shevde NK 2002 Vitamin D receptormediated gene regulation mechanisms and current concepts of vitamin D analog selectivity. Adv Ren Replace Ther 9:168–174. 8. Murayama E, Miyamoto K, Kubodera N, Mori T, Matsunaga I 1986 Synthetic studies of vitamin D3 analogs. VIII. Synthesis of 22-oxavitamin D3 analogs. Chemical & Pharmaceutical Bulletin 34:4410–4413. 9. Ostrem VK, Tanaka Y, Prahl J, DeLuca HF, Ikekawa N 1987 24- and 26-homo-1,25-dihydroxyvitamin D3: preferential activity in inducing differentiation of human leukemia cells HL-60 in vitro. Proc Natl Acad Sci USA 84:2610–2614. 10. Binderup L, Bramm E 1988 Effects of a novel vitamin D analog MC903 on cell proliferation and differentiation in vitro and on calcium metabolism in vivo. Biochem Pharmacol 37:889–895. 11. Zhou JY, Norman AW, Lubbert M, Collins ED, Uskokovic MR, Koeffler HP 1989 Novel vitamin D analogs that modulate leukemic cell growth and differentiation with little effect on either intestinal calcium absorption or bone mobilization. Blood 74:82–93. 12. Chen TC, Persons K, Uskokovic MR 1992 An evaluation of 1,25-dihydroxyvitamin D3 analogs on the proliferation and differentiation of cultured human keratinocytes, calcium metabolism and differentiation of HL-60 cells. J Nutr Biochem 4:49–57. 13. Brown AJ, Coyne DW 2002 Vitamin D analogs: New therapeutics for secondary hyperparathyroidism. Treatments in Endocrinol 1:313–327.

1464 14. Brown AJ, Ritter CR, Finch JL, Morrissey J, Martin KJ, Murayama E, Nishii Y, Slatopolsky E 1989 The noncalcemic analog of vitamin D, 22-oxacalcitriol, suppresses parathyroid hormone synthesis and secretion. J Clin Invest 84:728–732. 15. Naveh-Many T, Silver J 1993 Effects of calcitriol, 22-oxacalcitriol, and calcipotriol on serum calcium and parathyroid hormone gene expression. Endocrinol 133:2724–2728. 16. Hirata M, Endo K, Katsumata K, Ichikawa F, Kubodera N, Fukagawa M 2002 A comparison between 1,25-dihydroxy22-oxavitamin D3 and 1,25-dihydroxyvitamin D3 regarding suppression of parathyroid hormone and calcemic action. Nephrology, Dialysis Transplantation 17(Suppl 10):41–45. 17. Slatopolsky E, Finch J, Ritter C, Denda M, Morrissey J, Brown A, DeLuca H 1995 A new analog of calcitriol, 19-nor1,25-(OH)2D2, suppresses parathyroid hormone secretion in uremic rats in the absence of hypercalcemia. Am J Kidney Dis 26:852–860. 18. Brown AJ, Finch JL, Takahashi F, Slatopolsky E 2000 The calcemic activity of 19-nor-1,25(OH)2D2 decreases with duration of treatment. J Am Soc Nephrol 11:2088–2094. 19. Finch JL, Brown AJ, Slatopolsky E 1999 Differential effects of 1,25-dihydroxyvitamin D3 and 19-nor-1,25-dihydroxyvitamin D2 on calcium and phosphorus resorption in bone. J Am Soc Nephrol 10:980–985. 20. Brown AJ, Finch J, Slatopolsky E 2002 Differential effects of 19-nor-1,25-dihydroxyvitamin D2 and 1,25-dihydroxyvitamin D3 on intestinal calcium and phosphate transport. J Lab Clin Med 139:279–284. 21. Coyne DW, Grieff M, Ahya SN, Giles K, Norwood K, Slatopolsky E 2002 Differential effects of acute administration of 19-nor-1,25-dihydroxyvitamin D2 and 1,25-dihydroxyvitamin D3 on serum calcium and phosphorus in hemodialysis patients. Am J Kidney Dis 40:1283–1288. 22. Teng M, Wolf M, Lowrie E, Ofsthun N, Lazarus M, Thadhani R 2003 Survival of patients undergoing hemodialysis with paricalcitol or calcitriol therapy. N Engl J Med 349:446–456. 23. Fan SLS, Schroeder NJ, Calverley MJ, Burrin JM, Makin HLJ, Cunningham J 2000 Potent suppression of the parathyroid glands by hydroxylated metabolites of dihydrotachysterol(2). Nephrol Dial Transplant 15:1943–1949. 24. Hruby M, Urena P, Mannstadt M, Schmitt F, Lacour B, Drueke TB 1996 Effects of new vitamin D analogs on parathyroid function in chronically uremic rats with secondary hyperparathyroidism. Nephrol Dial Transplant 11:1781–1786. 25. Kragballe K 1995 Calcipotriol: a new drug for topical psoriasis treatment. Pharmacol & Toxicol 77:241–246. 26. Matsunaga T, Yamamoto M, Mimura H, Ohta T, Kiyoki M, Ohba T, Naruchi T, Hosoi J, Kuroki T 1990 1,24(R)-dihydroxyvitamin D3, a novel active form of vitamin D3 with high activity for inducing epidermal differentiation but decreased hypercalcemic activity. J Dermatol 17:135–142. 27. Morimoto S, Imanaka S, Koh E, Shiraishi T, Nabata T, Kitano S, Miyashita Y, Nishii Y, Ogihara T 1989 Comparison of the inhibitions of proliferation of normal and psoriatic fibroblasts by 1 alpha,25-dihydroxyvitamin D3 and synthetic analogs of vitamin D3 with an oxygen atom in their side chain. Biochem Int 19:1143–1149. 28. Endo K, Ichikawa F, Uchiyama Y, Katsumata K, Ohkawa H, Kumaki K, Ogata E, Ikeda K 1994 Evidence for the uptake of a vitamin D analog (OCT) by a human carcinoma and its effect of suppressing the transcription of parathyroid hormonerelated peptide gene in vivo. J Biol Chem 269:32693–32699. 29. Haq M, Kremer R, Goltzman D, Rabbani SA 1993 A vitamin D analog (EB1089) inhibits parathyroid hormone-related

ALEX J. BROWN AND EDUARDO A. SLATOPOLSKY

30.

31.

32.

33.

34.

35.

36.

37.

38. 39.

40.

41.

42.

peptide production and prevents the development of malignancy-associated hypercalcemia in vivo. J Clin Invest 91:2416–2422. Abe J, Nakano T, Nishii Y, Matsumoto T, Ogata E, Ikeda K 1991 A novel vitamin D3 analog, 22-oxa-1,25-dihydroxyvitamin D3, inhibits the growth of human breast cancer in vitro and in vivo without causing hypercalcemia. Endocrinol 129: 832–837. Oikawa T, Yoshida Y, Shimamura M, Ashino-Fuse H, Iwaguchi T, Tominaga T 1991 Antitumor effect of 22-oxa-1 alpha,25-dihydroxyvitamin D3, a potent angiogenesis inhibitor, on rat mammary tumors induced by 7,12-dimethylbenz[a]anthracene. Anticancer Drugs 2:475–480. Colston KW, Chander SK, Mackay AG, Coombes RC 1992 Effects of synthetic vitamin D analogs on breast cancer cell proliferation in vivo and in vitro. Biochem Pharmacol 44:693–702. Colston KW, Mackay AG, James SY, Binderup L, Chander S, Coombes RC 1992 EB1089: a new vitamin D analog that inhibits the growth of breast cancer cells in vivo and in vitro. Biochem Pharmacol 44:2273–2280. Anzano MA, Smith JM, Uskokovic MR, Peer CW, Mullen LT, Letterio JJ, Welsh MC, Shrader MW, Logsdon DL, Driver CL, et al. 1994 1 alpha,25-Dihydroxy-16-ene-23-yne-26,27-hexafluorocholecalciferol (Ro24-5531), a new deltanoid (vitamin D analog) for prevention of breast cancer in the rat. Cancer Res 54:1653–1656. Mehta RG, Hussain EA, Mehta RR, Das Gupta TK 2003 Chemoprevention of mammary carcinogenesis by 1α-hydroxyvitamin D5, a synthetic analog of vitamin D. Mutat Res 523–524:253–264. Vegesna V, O’Kelly J, Said J, Uskokovic M, Binderup L, Koeffler HP 2003 Ability of potent vitamin D3 analogs to inhibit growth of prostate cancer cells in vivo. Anticancer Res 23:283–289. Lucia MS, Anzano MA, Slayter MV, Anver MR, Green DM, Shrader MW, Logsdon DL, Driver CL, Brown CC, Peer CW, Roberts AB, Sporn MB 1995 Chemopreventive activity of tamoxifen, N-(4-hydroxyphenyl)retinamide, and the vitamin D analog RO24-5531 for adrogen-promoted carcinomas of the rat seminal vesicle and prostate. Cancer Res 55: 5621–5627. Schwartz GG, Hill CC, Oeler TA, Becich MJ, Bahnson RR 1995 1,25-Dihydroxy-16-ene-23-yne-vitamin D3 and prostate cancer cell proliferation in vivo. Urology 46:365–369. Otoshi T, Iwata H, Kitano M, Nishizawa Y, Morii H, Yano Y, Otani S, Fukushima S 1995 Inhibition of intestinal tumor development in rat multi-organ carcinogenesis and aberrant crypt foci in rat colon carcinogenesis by 22-oxa-calcitriol, a synthetic analog of 1 alpha, 25-dihydroxyvitamin D3. Carcinogenesis 16:2091–2097. Wali RK, Bissonnette M, Khare S, Hart J, Sitrin MD, Brasitus TA 1995 1 alpha,25-Dihydroxy-16-ene-23-yne-26, 27-hexafluorocholecalciferol, a noncalcemic analog of 1 alpha,25-dihydroxyvitamin D3, inhibits azoxymethaneinduced colonic tumorigenesis. Cancer Res 55:3050–3054. Huerta S, Irwin RW, Heber D, Go VL, Koeffler HP, Uskokovic M, Harris DM 2002 1α,25(OH)2D3 and its synthetic analog decrease tumor load in the Apc(min) mouse. Cancer Res 62:741–746. Zhou JY, Norman AW, Chen DL, Sun GW, Uskokovic M, Koeffler HP 1990 1,25-Dihydroxy-16-ene-23-yne-vitamin D3 prolongs survival time of leukemic mice. Proc Natl Acad Sci USA 87:3929–3932.

CHAPTER 82 Mechanisms for the Selective Actions of Vitamin D Analogs

43. Prudencio J, Akutsu N, Benlimame N, Wang T, Bastien Y, Lin R, Black MJ, Alaoui-Jamali MA, White JH 2001 Action of low calcemic 1a,25-dihydroxyvitamin D3 analog EB1089 in head and neck squamous cell carcinoma. J Natl Cancer Inst 93:745–753. 44. Abe J, Takita Y, Nakano T, Miyaura C, Suda T, Nishii Y 1989 A synthetic analog of vitamin D3, 22-oxa-1 alpha,25-dihydroxyvitamin D3, is a potent modulator of in vivo immunoregulating activity without inducing hypercalcemia in mice. Endocrinol 124:2645–2647. 45. Lemire JM, Archer DC, Reddy GS 1994 1,25-Dihydroxy-24oxo-16ene-vitamin D3, a renal metabolite of the vitamin D analog 1,25-dihydroxy-16-ene-vitamin D3, exerts immunosuppressive activity equal to its parent without causing hypercalcemia in vivo. Endocrinol 135:2818–2821. 46. Mathieu C, Waer M, Casteels K, Laureys J, Bouillon R 1995 Prevention of type I diabetes in NOD mice by nonhypercalcemic doses of a new structural analog of 1,25-dihydroxyvitamin D3, KH1060. Endocrinol 136:866–872. 47. Zugel U, Steinmeyer A, Giesen C, Asadullah K 2002 A novel immunosuppressive 1α,25-dihydroxyvitamin D3 analog with reduced hypercalcemic activity. J Invest Dermatol 119: 1434–1442. 48. Nishii Y, Sato K, Kobayashi T 1993 The development of vitamin D3 analogs for the treatment of osteoporosis. Osteoporos Int 3:190–193. 49. Uchiyama Y, Higuchi Y, Takeda S, Masaki T, Shira-Ishi A, Sato K, Kubodera N, Ikeda K, Ogata E 2002 ED-71, a vitamin D analog, is a more potent inhibitor of bone resorption than alfacalcidol in an estrogen-deficient rat model of osteoporosis. Bone 30:582–588. 50. Tanaka Y, Nakamura T, Nishida S, Suzuki K, Takeda S, Sato K, Nishii Y 1996 Effects of a synthetic vitamin D analog, ED-71, on bone dynamics and strength in cancellous and cortical bone in prednisolone-treated rats. J Bone Min Res 11:325–336. 51. Peleg S, Uskokovic M, Ahene A, Vickery B, Avnur Z 2002 Cellular and molecular events associated with the bone-protecting activity of the noncalcemic vitamin D analog Ro-26-9228 in osteopenic rats. Endocrinol 143: 1625–1636. 52. Shevde NK, Plum LA, Clagett-Dame M, Yamamoto H, Pike JW, DeLuca HF 2002 A potent analog of 1α,25-dihydroxyvitamin D3 selectively induced bone formation. Proc Natl Acad Sci USA 99:13487–13491. 53. Krisinger J, Strom M, Darwish HD, Perlman K, Smith C, DeLuca HF 1991 Induction of calbindin-D 9k mRNA but not calcium transport in rat intestine by 1,25-dihydroxyvitamin D3 24-homologs. J Biol Chem 266:1910–1913. 54. Wang YZ, Li H, Bruns ME, Uskokovic M, Truitt GA, Horst R, Reinhardt T, Christakos S 1993 Effect of 1,25,28trihydroxyvitamin D2 and 1,24,25-trihydroxyvitamin D3 on intestinal calbindin-D9K mRNA and protein: is there a correlation with intestinal calcium transport? J Bone Min Res 8:1483–1490. 55. Elstner E, Lee YY, Hashiya M, Pakkala S, Binderup L, Norman AW, Okamura WH, Koeffler HP 1994 1 alpha, 25-Dihydroxy-20-epi-vitamin D3: an extraordinarily potent inhibitor of leukemic cell growth in vitro. Blood 84: 1960–1967. 56. Crofts LA, Hancock MS, Morrison NA, Eisman JA 1998 Multiple promoters direct the tissue-specific expression of novel N-terminal variant human vitamin D receptor gene transcripts. Proc Natl Acad Sci USA 95:10529–10534.

1465

57. Gardner JP, Zhang F, Uskokovic MR, Studzinski GP 1996 Vitamin D analog 25-(OH)-16,23E-diene-26,27-hexafluorovitamin D3 induces differentiation of HL60 cells with minimal effects on cellular calcium homeostasis. J Cell Biochem 63:500–512. 58. Peleg S, Liu YY, Reddy S, Horst RL, White MC, Posner GH 1996 A 20-epi side chain restores growth-regulatory and transcriptional activities of an A-ring-modified hybrid analog of 1 alpha, 25-dihydroxyvitamin D3 without increasing its affinity to the vitamin D receptor. J Cell Biochem 63:149–161. 59. Slatopolsky E, Finch J, Plum LA, Clagett-Dame M, Sicinski RF, DeLuca HF 2003 A new vitamin D analog without calcemic and phosphatemic effects. J Am Soc Nephrol 14:(abstract). 60. Verstuyf A, Verlinden L, Van Baelen H, Sabbe K, D’Hallewyn C, De Clercq P, Vandewalle M, Bouillon R 1998 The biological activity of nonsteroidal vitamin D hormone analogs lacking both the C- and D-rings. J Bone Min Res 13:549–558. 61. Koike N, Hayakawa N, Kumaki K, Stumpf WE 1998 In vivo dose-related receptor binding of the vitamin D analog [3H]-1,25-dihydroxy-22-oxavitamin D3 (OCT) in rat parathyroid, kidney distal and proximal tubules, duodenum, and skin, studied by quantitative receptor autoradiography. J Histochem Cytochem 46:1351–1358. 62. Barsony J, McKoy W 1992 Molybdate increases intracellular 3′,5′-guanosine cyclic monophosphate and stabilizes vitamin D receptor association with tubulin-containing filaments. J Biol Chem 267:24457–24465. 63. Kamimura S, Gallieni M, Zhong M, Beron W, Slatopolsky E, Dusso A 1995 Microtubules mediate cellular 25-hydroxyvitamin D3 trafficking and the genomic response to 1,25dihydroxyvitamin D3 in normal human monocytes. J Biol Chem 270:22160–22166. 64. Racz A, Barsony J 1999 Hormone-dependent translocation of vitamin D receptors is linked to transactivation. J Biol Chem 274:19352–193560. 65. McDonnell DP, Scott RA, Kerner SA, BW OM, Pike JW 1989 Functional domains of the human vitamin D3 receptor regulate osteocalcin gene expression. Mol Endocrinol 3:635–644. 66. Hsieh JC, Jurutka PW, Galligan MA, Terpening CM, Haussler CA, Samuels DS, Shimizu Y, Shimizu N, Haussler MR 1991 Human vitamin D receptor is selectively phosphorylated by protein kinase C on serine 51, a residue crucial to its transactivation function. Proc Natl Acad Sci USA 88:9315–9319. 67. Yu XP, Mocharla H, Hustmyer FG, Manolagas SC 1991 Vitamin D receptor expression in human lymphocytes. Signal requirements and characterization by Western blots and DNA sequencing. J Biol Chem 266:7588–7595. 68. Krishnan AV, Feldman D 1991 Activation of protein kinase-C inhibits vitamin D receptor gene expression. Mol Endocrinol 5:605–612. 69. Jurutka PW, Hsieh JC 1993 Phosphorylation of the human 1,25-dihydroxyvitamin D3 receptor by cAMP-dependent protein kinase, in vitro, and in transfected COS-7 cells. Biochemical Biophysical Research Communications 191: 1089–1096. 70. Darwish HM, Burmester JK, Moss VE, DeLuca HF 1993 Phosphorylation is involved in transcriptional activation by the 1,25-dihydroxyvitamin D3 receptor. Biochem Biophys Acta 1167:29–36. 71. Matkovits T, Christakos S 1995 Ligand occupancy is not required for vitamin D receptor and retinoid receptor-mediated transcriptional activation. Mol Endocrinol 9:232–242.

1466 72. Nakajima S, Yamagata M, Sakai N, Ozono K 2000 Effect of cyclic adenosine 3′,5′-monophosphate and protein kinase A on ligand-dependent transactivation via the vitamin D receptor. Mol Cell Endocrinol 159:45–51. 73. Jurutka PW, Terpening CM, Haussler MR 1993 The 1,25dihydroxyvitamin D3 receptor is phosphorylated in response to 1,25-dihydroxyvitamin D3 and 22-oxacalcitriol in rat osteoblasts, and by casein kinase II, in vitro. Biochemistry 32:8184–8192. 74. Jurutka PW, MacDonald PN, Nakajima S, Hsieh JC, Thompson PD, Whitfield GK, Galligan MA, Haussler CA, Haussler MR 2002 Isolation of baculovirus-expressed human vitamin D receptor: DNA responsive element interactions and phosphorylation of the purified receptor. J Cell Biochem 85:435–457. 75. Barletta F, Freedman LP, Christakos S 2002 Enhancement of VDR-mediated transcription by phosphorylation: correlation with increased interaction between the VDR and DRIP205, a subunit of the VDR-interacting protein coactivator complex. Mol Endocrinol 16:301–314. 76. Buitrago C, Vazquez G, De Boland AR, Boland RL 2000 Activation of Src kinase in skeletal muscle cells by 1,25(OH)2-vitamin D3 correlates with tyrosine phosphorylation of the vitamin D receptor (VDR) and VDR-Src interaction. J Cell Biochem 79:274–281. 77. Chen S, Wu J, Hsieh JC, Whitfield GK, Jurutka PW, Haussler MR, Gardner DG 1998 Suppression of ANP gene transcription by liganded vitamin D receptor: involvement of specific receptor domains. Hypertension 31:1338–1342. 78. Liu YY, Nguyen C, Ali Gradezi SA, Schnirer I, Peleg S 2001 Differential regulation of heterodimerization by 1α,25dihydroxyvitamin D3 and its 20-epi analog. Steroids 66: 203–212. 79. Koszewski NJ, Reinhardt TA, Horst RL 1999 Differential effects of 20-epi vitamin D analogs on the vitamin D receptor homodimer. J Bone Min Res 14:509–517. 80. Haussler MR, Whitfield GK, Haussler CA, Hsieh JC, Thompson PD, Selznick SH, Dominiquez CE, Jurutka PW 1998 The nuclear vitamin D receptor: biological and molecular regulatory properties revealed. J Bone Min Res 13:325–349. 81. Strugnell SA, De Luca HF 1997 The vitamin D receptor— structure and transcriptional activation. Proc Soc Exp Biol Med 215:223–228. 82. Norman AW 1998 Receptors for 1α,25(OH)2D3—Past, present, and future. J Bone Min Res 13:1360–1369. 83. Schrader M, Kahlen JP, Carlberg C 1997 Functional characterization of a novel type of 1 alpha,25-dihydroxyvitamin D3 response element identified in the mouse c-fos promoter. Biochem Biophys Res Commun 230:646–651. 84. Peleg S, Sastry M, Collins ED, Bishop JE, Norman AW 1995 Distinct conformational changes induced by 20-epi analogs of 1 alpha,25-dihydroxyvitamin D3 are associated with enhanced activation of the vitamin D receptor. J Biol Chem 270:10551–10558. 85. Liu YY, Collins ED, Norman AW, Peleg S 1997 Differential interaction of 1alpha,25-dihydroxyvitamin D3 analogs and their 20-epi homologs with the vitamin D receptor. J Biol Chem 272:3336–3345. 86. Ryhanen S, Mahonen A, Jaaskelainen T, Maenpaa PH 1996 Synthetic 20-epi analogs of calcitriol are potent inducers of target-gene activation in osteoblastic cells. Eur J Biochem 238:97–103. 87. Mathiasen IS, Grue-Sorensen G, Hansen CM, Binderup L, Bjorkling F 1998 Studies on the interaction between the

ALEX J. BROWN AND EDUARDO A. SLATOPOLSKY

88.

89.

90.

91.

92.

93.

94.

95. 96.

97.

98.

99.

100.

101.

102.

vitamin D receptor and the radiolabelled 20-epi vitamin D analog GS 1500. Biochem Biophys Res Commun 250:283–286. Lewin E, Olgaard K 1994 The in vivo effect of a new, in vitro, extremely potent vitamin D3 analog KH1060 on the suppression of renal allograft rejection in the rat. Calcif Tissue Int 54:150–154. Danielsson C, Nayeri S, Wiesinger H, Thieroff-Ekerdt R, Carlberg C 1996 Potent gene regulatory and antiproliferative activities of 20-methyl analogs of 1,25 dihydroxyvitamin D3. J Cell Biochem 63:199–206. Danielsson C, Mathiasen IS, James SY, Nayeri S, Bretting C, Hansen CM, Colston KW, Carlberg C 1997 Sensitive induction of apoptosis in breast cancer cells by a novel 1,25-dihydroxyvitamin D3 analog shows relation to promoter selectivity. J Cell Biochem 66:552–562. Masuyama H, Jefcoat SC Jr, MacDonald PN 1997 The N-terminal domain of transcription factor IIB is required for direct interaction with the vitamin D receptor and participates in vitamin D–mediated transcription. Mol Endocrinol 11:218–228. Yang W, Freedman LP 1999 20-Epi analogs of 1,25-dihydroxyvitamin D3 are highly potent inducers of DRIP coactivator complex binding to the vitamin D3 receptor. J Biol Chem 274:16838–16845. Liu YY, Nguyen C, Peleg S 2000 Regulation of ligandinduced heterodimerization and coactivator interaction by the activation function-2 domain of the vitamin D receptor. Mol Endocrinol 14:1776–1787. Takeyama K, Masuhiro Y, Fuse H, Endoh H, Murayama A, Kitanaka S, Suzawa M, Yanagisawa J, Kato S 1999 Selective interaction of vitamin D receptor with transcriptional coactivators by a vitamin D analog. Mol Cell Biol 19:1049–1055. Issa LL, Leong GM, Sutherland RL, Eisman JA 2002 Vitamin D analog-specific recruitment of vitamin D receptor coactivators. J Bone Min Res 17:879–890. Masuyama H, Brownfield CM, St-Arnaud R, MacDonald PN 1997 Evidence for ligand-dependent intramolecular folding of the AF-2 domain in vitamin D receptor-activated transcription and coactivator interaction. Mol Endocrinol 11:1507–1517. Masuyama H, MacDonald PN 1998 Proteasome-mediated degradation of the vitamin D receptor (VDR) and a putative role for SUG1 interaction with the AF-2 domain of VDR. J Cell Biochem 71:429–440. Zhou Z-X, Lane MV, Kemppainen JA, French FS, Wilson EM 1995 Specificity of ligand-dependent androgen receptor stabilization: receptor domain interactions influence ligand dissociation and receptor stability. Mol Endocrinol 9:208–218. van den Bemd GC, Pols HA, Birkenhager JC, van Leeuwen JP 1996 Conformational change and enhanced stabilization of the vitamin D receptor by the 1,25-dihydroxyvitamin D3 analog KH1060. Proc Natl Acad Sci USA 93:10685–10690. Peleg S, Nguyen C, Woodard BT, Lee JK, Posner GH 1998 Differential use of transcription activation function 2 domain of the vitamin D receptor by 1,25-dihydroxyvitamin D3 and its A-ring modified analogs. Mol Endocrinol 12:525–535. Okano T, Tsugawa N, Masuda S, Takeuchi A, Kobayashi T, Nishii Y 1989 Protein-binding properties of 22-oxa-1α,25dihydroxyvitamin D3, a synthetic analog of 1α,25-dihydroxyvitamin D3. J Nutr Sci Vitaminol 35:529–533. Kobayashi T, Okano T, Tsugawa N, Masuda S, Takeuchi A, Nishii Y 1991 Metabolism and transporting system of 22-oxacalcitriol. Contrib Nephrol 91:129–133.

CHAPTER 82 Mechanisms for the Selective Actions of Vitamin D Analogs

103. Brown AJ, Finch J, Grieff M, Ritter C, Kubodera N, Nishii Y, Slatopolsky E 1993 The mechanism for the disparate actions of calcitriol and 22-oxacalcitriol in the intestine. Endocrinol 133:1158–1164. 104. Koike N, Hayakawa N, Tokuda K, Saito K, Stumpf WE 2002 In vivo time-course of receptor binding in the parathyroid gland of the vitamin D analog [3H]1,25-dihydroxy-22-oxavitamin D3 compared to [3H]1,25-dihydroxyvitamin D3, determined by micro-autoradiography. Nephrol Dial Transplant 17(Suppl 10):53–57. 105. Boehm MF, Fitzgerald P, Zou A, Elgort MG, Bischoff ED, Mere L, Mais DE, Bissonnette RP, Heyman RA, Nadzan AM, Reichman M, Allegretto EA 1999 Novel nonsecosteroidal vitamin D mimics exert VDR-modulating activities with less calcium mobilization than 1,25-dihydroxyvitamin D3. Chemistry & Biology 6:265–275. 106. Okano T, Tsugawa N, Masuda S, Takeuchi A, Kobayashi T, Nishii Y 1991 A novel synthetic vitamin D3 analog, 2-beta(3-hydroxypropoxy)-calcitriol (ED-71): its biological activities and pharmacological effects on calcium metabolism. Contrib Nephrol 91:116–122. 107. Dusso AS, Negrea L, Gunawardhana S, Lopez-Hilker S, Finch J, Mori T, Nishii Y, Slatopolsky E, Brown AJ 1991 On the mechanisms for the selective action of vitamin D analogs. Endocrinol 128:1687–1692. 108. Dilworth FJ, Calverley MJ, Makin HL, Jones G 1994 Increased biological activity of 20-epi-1,25-dihydroxyvitamin D3 is due to reduced catabolism and altered protein binding. Biochem Pharmacol 47:987–993. 109. Bouillon R, Allewaert K, Xiang DZ, Tan BK, van Baelen H 1991 Vitamin D analogs with low affinity for the vitamin D binding protein: enhanced in vitro and decreased in vivo activity. J Bone Min Res 6:1051–1057. 110. Kobayashi T, Tsugawa N, Okano T, Masuda S, Takeuchi A, Kubodera N, Nishii Y 1994 The binding properties, with blood proteins, and tissue distribution of 22-oxa-1α,25-dihydroxyvitamin D3, a noncalcemic analog of 1α, 25-dihydroxyvitamin D3, in rats. J Biochem 115:373–380. 111. Teramoto T, Endo K, Ikeda K, Kubodera N, Kinoshita M, Yamanaka M, Ogata E 1995 Binding of vitamin D to low-density-lipoprotein (LDL) and LDL receptor-mediated pathway into cells. Biochem Biophys Res Commun 215:199–204. 112. Haddad JG, Jennings AS, Aw TC 1988 Vitamin D uptake and metabolism by perfused rat liver: influences of carrier proteins. Endocrinol 123:498–504. 113. Haddad JG, Aden DP, Aw TC 1989 Plasma carriers influence the uptake of cholecalciferol by human hepatoma-derived cells. J Bone Min Res 4:243–247. 114. Keenan MJ, Holmes RP 1991 The uptake and metabolism of 25-hydroxyvitamin D3 and vitamin D–binding protein by cultured porcine kidney cells (LLC-PK1). Inter J Biochem 23:1225–1230. 115. Esteban C, Geuskens M, Ena JM, Mishal Z, Macho A, Torres JM, Uriel J 1992 Receptor-mediated uptake and processing of vitamin D–binding protein in human B-lymphoid cells. J Biol Chem 267:10177–10183. 116. Stewart PM, Shappard MC 1992 Novel aspects of hormone action: intracellular ligand supply and its control by a series of tissue-specific enzymes. Mol Cell Endocrinol 83: C13–C18. 117. Zhao J, Tan BK, Marcelis S, Verstuyf A, Bouillon R 1996 Enhancement of antiproliferative activity of 1α,25-dihydroxyvitamin D3 (analogs) by cytochrome P450 enzyme

118.

119.

120.

121.

122.

123.

124.

125.

126.

127.

128.

1467

inhibitors is compound- and cell-type specific. Steroid Biochem Mol Biol 57:197–202. Shankar VN, Dilworth FJ, Makin HL, Schroeder NJ, Trafford DJ, Kissmeyer AM, Calverley MJ, Binderup E, Jones G 1997 Metabolism of the vitamin D analog EB1089 by cultured human cells: redirection of hydroxylation site to distal carbons of the side-chain [published erratum appears in Biochem Pharmacol 1997 Jun 15;53(12):1946]. Biochem Pharmacol 53:783–793. Hansen CM, Maenpaa PH 1997 EB1089, a novel vitamin D analog with strong antiproliferative and differentiationinducing effects on target cells. Biochem Pharmacol 54: 1173–1179. Kissmeyer AM, Binderup E, Binderup L, Mork Hansen C, Andersen NR, Makin HL, Schroeder NJ, Shankar VN, Jones G 1997 Metabolism of the vitamin D analog EB 1089: identification of in vivo and in vitro liver metabolites and their biological activities. Biochem Pharmacol 53: 1087–1097. Lin R, Nagai Y, Sladek R, Bastien Y, Ho J, Petrecca K, Sotiropoulou G, Diamandis EP, Hudson TJ, White JH 2002 Expression profiling in squamous carcinoma cells reveals pleiotropic effects of vitamin D3 analog EB1089 signaling on cell proliferation, differentiation, and immune system regulation. Mol Endocrinol 16:1243–1256. Brown AJ, Berkoben M, Ritter C, Kubodera N, Nishii Y, Slatopolsky E 1992 Metabolism of 22-oxacalcitriol by a vitamin D–inducible pathway in cultured parathyroid cells. Biochem Biophys Res Commun 189:759–764. Bikle DD, Abe-Hashimoto J, Su MJ, Felt S, Gibson DF, Pillai S 1995 22-Oxa calcitriol is a less potent regulator of keratinocyte proliferation and differentiation due to decreased cellular uptake and enhanced catabolism. J Invest Dermatol 105:693–698. Kamimura S, Gallieni M, Kubodera N, Nishii Y, Brown AJ, Slatopolsky E, Dusso A 1993 Differential catabolism of 22-oxacalcitriol and 1,25-dihydroxyvitamin D3 by normal human peripheral monocytes. Endocrinol 133: 2719–2723. Reddy GS, Clark JW, Tserng KY, Uskokovic MR, McLane JA 1993 Metabolism of 1,25(OH)2-16-ene-D3 in kidney: influence of structural modification of D-ring on side chain metabolism. Bioorg Med Chem Lett 3: 1879–1884. Siu-Caldera ML, Clark JW, Santos-Moore A, Peleg S, Liu YY, Uskokovic MR, Sharma S, Reddy GS 1996 1alpha,25-dihydroxy-24-oxo-16-ene vitamin D3, a metabolite of a synthetic vitamin D3 analog, 1alpha,25-dihydroxy16-ene vitamin D3, is equipotent to its parent in modulating growth and differentiation of human leukemic cells. J Steroid Biochem Mol Biol 59:405–412. Dilworth FJ, Williams GR, Kissmeyer AM, Nielsen JL, Binderup E, Calverley MJ, Makin HL, Jones G 1997 The vitamin D analog, KH1060, is rapidly degraded both in vivo and in vitro via several pathways: principal metabolites generated retain significant biological activity. Endocrinol 138:5485–5496. van den Bemd GC, Dilworth FJ, Makin HL, Prahl JM, DeLuca HF, Jones G, Pols HA, van Leeuwen JP 2000 Contribution of several metabolites of the vitamin D analog 20-epi-22-oxa-24a,26a,27a-trihomo1,25(OH)2-vitamin D3 (KH1060) to the overall biological activity of KH1060 by a shared mechanism of action. Biochem Pharmacol 59:621–627.

1468 129. Komuro S, Kanamaru H, Nakatsuka I, Yoshitake A 1998 Distribution and metabolism of F6-1,25(OH)2 vitamin D3 and 1,25(OH)2 vitamin D3 in the bones of rats dosed with tritiumlabeled compounds. Steroids 63:505–510. 130. Sekimoto H, Siu-Caldera ML, Weiskopf A, Vouros P, Muralidharan KR, Okamura WH, Uskokovic M, Reddy GS 1999 1α,25Dihydroxyvitamin D3: in vivo metabolite of 1α,25-dihydroxyvitamin D3 in rats. FEBS Lett 448:278–282. 131. Kamao M, Tatematsu S, Hatakeyama S, Ozono K, Kubodera N, Reddy GS, Okano T 2003 Two novel metabolic pathways of 22-oxacalcitriol (OCT). C-25 dehydration and C-3 epimerization and biological activities of novel OCT metabolites. J Biol Chem 278:1463–1471. 132. Reddy GS, Rao DS, Siu-Caldera ML, Astecker N, Weiskopf A, Vouros P, Sasso GJ, Manchand PS, Uskokovic M 2000 1α25-Dihydroxy-16-ene-23-yne-vitamin D3 and 1α,25-dihydroxy-16-ene-23-yne-20-epi-vitamin D3: analogs of 1α,25dihydroxyvitamin D3 that resist metabolism through the C-24 oxidation pathway are metabolized through the C-3 epimerization pathway. Arch Biochem Biophys 383: 197–205. 133. Nemere I, Farach-Carson MC 1998 Membrane receptors for steroid hormones—a case for specific cell surface binding sites for vitamin D metabolites and estrogens. Biochem Biophys Res Commun 248:443–449. 134. Nemere I, Yoshimoto Y, Norman AW 1984 Calcium transport in perfused duodena from normal chicks: enhancement within fourteen minutes of exposure to 1,25-dihydroxyvitamin D3. Endocrinol 115:1476–1483. 135. Beno DWA, Brady LM, Bissonnette M, Davis BH 1995 Protein kinase C and mitogen-activated protein kinase are required for 1,25-dihydroxyvitamin D3-stimulated Egr induction. J Biol Chem 270:3642–3647. 136. Song X, Bishop JE, Okamura WH, Norman AW 1998 Stimulation of phosphorylation of mitogen-activated protein kinase by 1α,25-dihydroxyvitamin D3 in promyelocytic NB4 leukemic cells: a structure-function study. Endocrinol 139:457–465. 137. Zanello LP, Norman AW 1996 1α,25(OH)2-vitamin D3mediated stimulation of outward anionic currents in osteoblast-like ROS 17/2.8 cells. Biochem Biophys Res Commun 225:551–556. 138. Lieberherr M, Grosse B, Duchambon P, Drueke T 1989 A functional cell surface type receptor is required for the early action of 1,25-dihydroxyvitamin D3 on the phosphoinositide metabolism in rat enterocytes. J Biol Chem 264: 20403–20406. 139. Nemere I, Dormanen MC, Hammond MW, Okamura WH, Norman AW 1994 Identification of a specific binding protein for 1α,25-dihydroxyvitamin D3 in basal-lateral membranes of chick intestinal epithelium and relationship to transcaltachia. J Biol Chem 269:23750–23756. 140. Nemere I, Schwartz Z, Pedrozo H, Sylvia VL, Dean DD, Boyan BD 1998 Identification of a membrane receptor for 1,25-dihydroxyvitamin D3, which mediates the rapid activation of protein kinase C. J Bone Min Res 13:1353–1359. 141. Baran D, Quail J, Ray R, Leszyk J, Honeyman T 1998 Identification of the membrane protein that binds 1α,25dihydroxyvitamin D3 and is involved in the rapid actions of the hormone. Bone 23:S176. 142. Zhou LX, Nemere I, Norman AW 1992 1,25-Dihydroxyvitamin D3 analog structure-function assessment of the rapid stimulation of intestinal calcium absorption (transcaltachia). J Bone Min Res 7:457–463.

ALEX J. BROWN AND EDUARDO A. SLATOPOLSKY

143. Norman AW, Song X, Zanello L, Bula C, Okamura WH 1999 Rapid and genomic biological responses are mediated by different shapes of the agonist steroid hormone, 1α,25 (OH)2vitamin D3. Steroids 64:120–128. 144. Bouillon R, Okamura WH, Norman AW 1995 Structurefunction relationships in the vitamin D endocrine system. Endocr Rev 16:200–257. 145. Farach-Carson MC, Abe J, Nishii Y, Khoury R, Wright GC, Norman AW 1993 22-Oxacalcitriol: dissection of 1,25(OH)2D3 receptor-mediated and Ca2+ entry-stimulating pathways. Am J Physiol 265:F705–F711. 146. Boyan BD, Dean DD, Sylvia VL, Schwartz Z 1994 Nongenomic regulation of extracellular matrix events by vitamin D metabolites. J Cell Biochem 56:331–339. 147. Baran DT 1994 Nongenomic actions of the steroid hormone 1alpha,25-dihydroxyvitamin D3. J Cell Biochem 56:303–306. 148. Baran DT, Sorensen AM, Shalhoub V, Owen T, Stein G, Lian J 1992 The rapid nongenomic actions of 1α,25-dihydroxyvitamin D3 modulate the hormone-induced increments in osteocalcin gene transcription in osteoblast-like cells. J Cell Biochem 50:124–129. 149. Khoury R, Ridall AL, Norman AW, Farach-Carson MC 1994 Target gene activation by 1,25-dihydroxyvitamin D3 in osteosarcoma cells is independent of calcium influx. Endocrinol 135:2446–2453. 150. Khoury RS, Weber J, Farach-Carson MC 1995 Vitamin D metabolites modulate osteoblast activity by Ca2+ influx-independent genomic and Ca2+ influx-dependent nongenomic pathways. J Nutr 125:1699S–1703S. 151. Norman AW, Bouillon R, Farach-Carson MC, Bishop JE, Zhou LX, Nemere I, Zhao J, Muralidharan KR, Okamura WH 1993 Demonstration that 1β,25-dihydroxyvitamin D3 is an antagonist of the nongenomic but not genomic biological responses and biological profile of the three A-ring diastereomers of 1α,25-dihydroxyvitamin D3. J Biol Chem 268: 20022–20030. 152. Zhou LX, Norman AW 1995 1 α,25(OH)2-vitamin D3 analog structure-function assessment of intestinal nuclear receptor occupancy with induction of calbindin-D28K. Endocrinol 136:1145–1152. 153. Hedlund TE, Moffatt KA, Miller GJ 1996 Stable expression of the nuclear vitamin D receptor in the human prostatic carcinoma cell line JCA-1: evidence that the antiproliferative effects of 1α,25-dihydroxyvitamin D3 are mediated exclusively through the genomic signaling pathway. Endocrinol 137:1554–1561. 154. Safran JB, Butler WT, Farach-Carson MC 1998 Modulation of osteopontin posttranslational state by 1,25-(OH)2-vitamin D3. Dependence on Ca2+ influx. J Biol Chem 273: 29935–29941. 155. Baran DT, Quail JM, R. R, Leszyk J, Honeyman T 2000 Annexin II is the membrane receptor that mediates the rapid actions of 1α,25-dihydroxyvitamin D3. J Cell Biochem 78:34–46. 156. Gacad MA, Adams JS 1991 Endogenous blockade of 1,25dihydroxyvitamin D–receptor binding in New World primate cells. J Clin Invest 87:996–1001. 157. Gacad MA, Adams JS 1992 Specificity of steroid binding in New World primate B95-8 cells with a vitamin D–resistant phenotype. Endocrinol 131:2581–2587. 158. Gacad MA, Adams JS 1993 Identification of a competitive binding component in vitamin D-resistant New World primate cells with a low affinity but high capacity for 1,25-dihydroxyvitamin D3. J Bone Min Res 8:27–35.

CHAPTER 82 Mechanisms for the Selective Actions of Vitamin D Analogs

159. Gacad MA, Chen H, Arbelle JE, LeBon T, Adams JS 1997 Functional characterization and purification of an intracellular vitamin D–binding protein in vitamin D–resistant New World primate cells. Amino acid sequence homology with proteins in the HSP-70 family. J Biol Chem 272:8433–8440. 160. Gacad MA, Adams JS 1998 Proteins in the heat shock-70 family specifically bind 25-hydroxyvitamin D3 and 17βestradiol. J Clin Endocrinol Metab 83:1264–1267. 161. Wu S, Ren S, Chen H, Chun RF, Gacad MA, Adams JS 2000 Intracellular vitamin D–binding proteins: novel facilitators

1469

of vitamin D–directed transactivation. Mol Endocrinol 14:1387–1397. 162. Wu S, Chun RF, Gacad MA, Ren S, Chen H, Adams JS 2002 Regulation of 1,25-dihydroxyvitamin D synthesis by intracellular vitamin D–binding protein-1. Endocrinol 143: 4135–4138. 163. Adams JS, Chen H, Chun RF, Nguyen L, Wu S, Ren S-Y, Barsony J, Gacad MA 2003 Novel regulators of vitamin D action and metabolism: Lessons learned at the Los Angeles Zoo. J Cell Biochem 88:308–314.

CHAPTER 83

Molecular Basis for Differential Action of Vitamin D Analogs SARA PELEG

Department of Endocrine Neoplasia & Hormonal Disorders, The University of Texas, M. D. Anderson Cancer Center, Houston, Texas 77030

I. Introduction II. Structural Requirements for Transactivation of the VDR by Its Natural Ligand III. Differential Activation of the VDR by Synthetic Analogs

IV. Clinical Significance for Selective Modulation of the VDR by Vitamin D Analogs References

I. INTRODUCTION

hereditary vitamin D–resistant rickets (HVDRR), defects in VDR cause severe bone disease and deficient calcium absorption [20]. In the same disease, the growth inhibitory effect of 1,25(OH)2D3 on lectin-induced T-cells is diminished [32]. Finally, a link between normal development of hair follicles and functional VDR is suggested by the frequent presence of hairlessness (alopecia) in rickets patients with defective VDR [20]. Similar phenotypes are found in VDR-ablated mice, confirming the role of VDR in development of bone disease and alopecia [33–35]. In cell culture systems, the growth regulatory responses to 1,25(OH)2D3 are correlated with the levels of VDR [36]. In the heterogeneous human myeloid leukemia cell line HL-60, 1,25(OH)2D3 regulates the growth of clones with high VDR levels, but clones with few or no VDRs do not respond to 1,25(OH)2D3 [37]. In animal models, growth inhibition is induced by 1,25(OH)2D3 only in solid tumors containing VDR [17]. Additional support for the potential role of the VDR in cellular growth and differentiation is suggested from the abnormal development of mammary glands in VDR-null mice. In the same animals, there is also an increase in susceptibility of skin to carcinogen-induced tumorigenesis and hyperplasia of epithelial cells lining the colon [38–40]. In conclusion, VDR appears to be essential for both physiological and pharmacological responses to 1,25(OH)2D3. The wide range of responses transduced through the VDR provide the organic chemist and the biologist with the challenge of developing drugs that enhance or diminish specific VDR-mediated responses, and which might eventually be useful for treatment of disease. This chapter focuses on the molecular mechanisms by which chemically and stereochemically modified analogs of 1,25(OH)2D3 produce different effects with respect to activation of the VDR.

The most prominent physiological role of the hormonally active metabolite of vitamin D3, 1α,25-dihydroxyvitamin D3 [1,25(OH)2D3], is the regulation of calcium and phosphorous homeostasis as well as bone remodeling through its actions in the intestine, kidney, and bone [1–4]. The hormone also contributes to growth and differentiation of epidermal cells [5] and the bone marrow precursors of osteoclasts [6–9]. As discussed in detail in subsequent chapters, at pharmacological concentrations 1,25(OH)2D3 also has immunoregulatory effects [10] and induces differentiation and inhibits growth of psoriatic skin [11] and a variety of malignant cell types [12–17]. Genomic and nongenomic signal transduction pathways are believed to mediate the diverse effects of this hormone. The genomic pathway has been well established and is mediated by the nuclear vitamin D receptor (VDR) [18]. The VDR belongs to a large family of transcription factors that contain two highly conserved zincbinding finger structures in their DNA binding domain [19]. Transcriptional activity of the VDR depends primarily on its binding with 1,25(OH)2D3 [20]. This interaction induces conformational changes in the receptor [20,21] and facilitates heterodimerization with the retinoid X receptor (RXR) [22–24], promotes binding to specific DNA responsive elements [23,25,26], causes recruitment of transcription coactivators of the p160 family and other components of the transcription apparatus, through interaction with the DRIP complex, all of which eventually lead to regulation of gene expression [27–31]. Several lines of evidence suggest that the broad range of physiological and pharmacological activities induced by 1,25(OH)2D3, both those related to regulation of calcium metabolism and those associated with the regulation of growth and differentiation, are mediated by transcriptional activities of VDR. For example, in VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

1472

SARA PELEG

Affinity and transcriptional activity

OH

H

Catabolism & side-chain orientation

D

C H

3β

A

1α

HO

OH Affinity for VDR 1α,25-dihydroxyvitamin D3

FIGURE 1

Structural formula of 1,25(OH)2D3. Shaded areas indicate structural features that regulate ligand binding and transcriptional activity of VDR.

II. STRUCTURAL REQUIREMENTS FOR TRANSACTIVATION OF THE VDR BY ITS NATURAL LIGAND, 1,25(OH)2D3 1,25(OH)2D3-mediated transcriptional activity of VDR depends primarily on the following three structural features of the ligand: the A-ring, the side chain, and the D-ring (Fig. 1) [41]. The structural features of the

A

Protease sensitivity assay

ligand that are important for binding to VDR were initially elucidated through studies of naturally occurring metabolites of vitamin D3 and synthetic analogs of 1,25(OH)2D3 [41,42]. Recent studies on the crystal structure of the VDR bound to its natural hormone confirm the earlier studies and demonstrate how these structural features interact with specific amino acid residues in the ligand-binding pocket and how these interactions contribute to tight binding and to conformational changes in the ligand-binding domain (LBD) that lead to transcriptional activation of the VDR [43]. The LBD of VDR (like that of other steroid receptors) is multifunctional: it regulates dimerization [29], interaction with transcription factors, and general transcriptional activity [28–31]. The structure of VDR’s LBD is similar to that of other ligand-binding nuclear receptors in that it consists of 12 alpha helices organized in an antiparellel “sandwich” [43]. At the extreme C-terminus of the VDR is helix 12, a highly conserved sequence that contributes the core for the transcription activation function of the LBD that is called activation function 2 (AF-2) [44,45]. 1,25(OH)2D3 binding to VDR causes significant structural changes that decrease the sensitivity of the LBD to proteolytic digestion. These changes allow the VDR to form highaffinity interactions with other proteins, including the RXR, transcription coactivators of the p160 family (such as the steroid receptor coactivator 1 and the glucocorticoid receptor-interacting protein) and bridging factors such as the vitamin D receptor-interacting protein (DRIP 205) [28–30] (Fig. 2). These interactions are dependent on the availability of residues in the helix 12/AF-2 core for interaction with these proteins.

B

Binding to GST-SRC-1

34 kDa Intact

28 kDa

VDR Ligand (-LOG M)

FIGURE 2

6

7

8

9 10

11 un

Ligand (-LOG M)

7

8

9

10

11

un

Correlation between stabilization of VDR against proteolytic digestion and its ability to interact with the nuclear receptor interacting domain of the steroid receptor coactivator 1 (SRC-1). (A) In vitro synthesized 35S-labeled human VDR was incubated with the indicated concentrations of 1,25(OH)2D3 and then subjected to limited proteolytic digestion by trypsin. The proteolytic products were separated by SDS-PAGE and detected by autoradiography. (B) Binding of 35S-labeled synthetic human VDR to the fusion protein GST-SRC-1 was assessed by pull-down assay. The synthetic VDR was incubated with the indicated concentrations of 1,25(OH)2D3 and with recombinant GST-SRC-1 and glutathione-Sepharose beads. The VDR bound to SRC was eluted and separated by SDS-PAGE. The amount of bound VDR was assessed by autoradiography of the gels. Note the correlation between ligand-dependent stabilization of VDR conformation and its ability to interact with SRC-1 [135].

1473

CHAPTER 83 Molecular Basis for Differential Action of Vitamin D Analogs

Coactivator binding site LBD

LBD

Coactivator binding site

H1 2

LBD

Coactivator binding site

2 H1

H12

Unoccupied

Agonist

Antagonist

FIGURE 3 Schematic illustrations of the hypothetical conformations of unoccupied, agonist-occupied, and antagonist-occupied VDR LBD. Note the change in the position of helix12/AF2 core with respect to the coactivator binding site. In the unoccupied LBD, helix 12 is extended, and its residues are not associated with the LBD. In the agonistoccupied LBD, helix 12 changes position and is brought near the coactivator binding site. In the antagonist-occupied LBD, helix 12 masks the coactivator binding site and so prevents interaction with transcription coactivators. The conformations of unoccupied and antagonist occupied LBDs were adapted from structural studies of other LBDs [49].

The conformation of the AF-2 core is different in unoccupied, agonist-occupied and antagonist-occupied LBDs of other nuclear receptors (such as the estrogen receptor or progesterone receptor), and these differences have a strong effect on the target-tissue preferences and gene-regulatory events of the various ligand-receptor complexes (Fig. 3) [46–49]. Although only the structure of the agonist-bound VDR LBD has been published so far, we hypothesize that similar changes probably occur in the unoccupied VDR once it binds to agonists or antagonists.

A. The A-ring The backbone of 1,25(OH)2D3 is similar to that of cholesterol and the steroid hormones except that the four-ring structure has been disrupted by ultraviolet irradiation of the second (B) ring between carbons 9 and 10 (Fig. 1) [2–4]. This change makes the already flexible organic compound even more flexible, giving the A-ring the freedom to rotate from the steroid-like (“folded”) conformation to a vitamin (“extended”) conformation [50]. It has been shown that the steroidlike conformer does not interact with VDR, and therefore it was suggested that the vitamin conformer should be the one to bind with high affinity to the VDR [50]. However, the crystal structure of the VDR LBD showed that the position of the A-ring in the binding pocket is in between the cis and trans conformations, thus suggesting that modifications that restrict the flexibility of the A-ring may compromise ligand binding to the VDR.

The high affinity of vitamin D metabolites and analogs for the VDR also depends on the hydroxyl groups in the A-ring, particularly the 1α-hydroxyl [41,42]. For example, the precursor of 1,25(OH)2D3, 25-hydroxyvitamin D3, binds poorly to VDR, but hydroxylation of this metabolite at the lα position induces high affinity for the receptor [41]. Likewise, another natural metabolite of vitamin D3, 24,25-dihydroxyvitamin D3, also binds poorly to the VDR, but its 1α-hydroxyl metabolite has a significantly greater affinity [51]. Similarly, 1,25(OH)2D3 analogs that have modified side chains and lack the 1α-hydroxyl group bind poorly to VDR, whereas addition of hydroxyl or fluoride groups restores binding to VDR [52]. The explanation for these requirements is provided by X-ray crystallography of the LBD, which shows that the natural hormone is anchored in the binding pocket with its A-ring facing the interior and its 1α-hydroxyl group forming hydrogen bonds with arginine 274 and serine 237 [43]. In addition, although structurefunction studies show that the hydroxyl group at the 3β position is less important for high affinity binding to the VDR, this hydroxyl group is also forming hydrogen bonds, one with serine 278 and the other with tyrosine 143. This structural information has been confirmed by laboratory-made or natural mutations in the hVDR. The latter is the R274L mutation that causes almost complete loss of 1,25(OH)2D3 binding to the VDR and also causes hereditary vitamin D–resistant rickets [53]. The structural information has also allowed the manipulation of vitamin D analogs’ affinity for the VDR by introducing chemical modifications that provide additional anchoring points for the A-ring

1474

SARA PELEG

in the binding pocket. The most notable modifications are in C-2, and they include 2α-methyl, 2α-alkyl, and 2α-hydroxyalkyl analogs [54].

B. The Side Chain The ability of 1,25(OH)2D3 to bind, and more importantly to transactivate, the VDR depends on 25-hydroxylation of the side chain. This is proven by the fact that the synthetic prehormone 1α-hydroxyvitamin D3 cannot bind or transactivate the VDR in vitro or in cell culture. Again, structural resolution of the VDR bound to the natural hormone [43] shows that the 25-hydroxyl group forms two hydrogen bonds, one of which (histidine 397) is essential for tight binding of the ligand to the VDR, and the other (histidine 305) is essential for stabilization of the VDR in transcriptionally active conformation. The side chain of 1,25(OH)2D3 is facing the C-terminal region of the VDR and forms contacts with helix 11 and helix 12/AF-2 core. The AF-2 core contains residues that are essential for binding

to coactivators and for assembly of the AF-2 in transcriptionally active conformation. Therefore, it is likely that modifications in the side chain have the potential to alter transcriptional activation of the VDR either by contributing to ligand affinity for the VDR or by changing the conformation of AF-2 to increase or decrease its transcriptional activity. Because certain side chain modifications increase transcriptional activity of the VDR a 100- to 10,000-fold without changing affinity (Table I) suggests that these two functions of the side chain are independent. For example, transcriptional superagonists can be generated by epimerization of carbon 20 [55]. In the 20-epi analogs, this stereochemical change slightly reduces affinity for VDR, but it increases the potency of VDR-mediated transcriptional activity more than a hundredfold [55]. Additional evidence that the side chain contribution to VDR’s transactivation is a separate function from its contribution to ligand affinity for the VDR is provided by studies with hybrid analogs (Table I). These analogs contain A-ring modifications that compromise affinity for the VDR significantly, but in the presence of a potent side chain,

TABLE I Structure-functions Relationships of 1α,25-Dihydroxyvitamin D3 and Its Analogs Ligand formula Parent compound 1α,25(OH)2D3 A-ring modified analogs 25-OH-D3 1β,25(OH)2D3 1β-(hydroxymethyl)-3α-25-OH-D3

Receptor binding a

Transcriptionb

1 × 10−9

2 × 10−9

1.4 × 10−7 2 × 10−7 8 × 10−8

0.5 × 10−7 2 × 10−7 > 10−6

Side chain-modified analogs 1α-OH-D3 20-epi-1α,25(OH)2D3 20-epi-24a,26a,27a-tri-homo-1α,25(OH)2D3 22-oxa-24a,26a,27a-tri-homo-1α,25(OH)2D3 20-epi-22-oxa-24a,26a,27a-tri-homo-1α,25(OH)2D3

> 10−6 3 × 10−9 1 × 10−9 3 × 10−9 3 × 10−9

> 10−6 5 × 10−12 1 × 10−12 1 × 10−10 5 × 10−11

Hybrid analogs 1β-20-epi-24a,26a,27a-tri-homo-25(OH)2D3 1β-(hydroxymethyl)-3α-20-epi-22-oxa-24a,26a,27a-tri-homo-25-OH-D3

1 × 10−7 >10−6

2 × 10−9 1 × 10−8

Analogs with double bond between C16 and C17 1α,25(OH)2-16-ene-D3 1α,25(OH)2-16-ene-23-yne-D3

1 × 10−9 0.8 × 10−9

1 × 10−10 6 × 10−10

a ED 3 50 (in M), ligand concentration required to reach 50% displacement of [ H]1,25(OH)2D3 binding to recombinant human VDR from transfected COS-1 cells. b ED (in M), effective dose required to produce 50% of maximal transcription activation of a reporter gene containing the osteocalcin VDRE in ROS 50 17/2.8 cells in serum-free culture medium. The transcriptional activities of 1,25(OH)2D3 and the last two analogs in the table (16-ene analogs) were also examined in cells grown in 10% serum. Under these conditions, the ED50 of the three compounds were 5 ×10−10 M, 6 ×10−12 M, and 2 × 10−11 M, respectively.

CHAPTER 83 Molecular Basis for Differential Action of Vitamin D Analogs

their transcriptional activity increases manyfold without a change in their affinity for the VDR [56,57].

C. The D-ring The crystal structure of VDR-1,25(OH)2D3 complex show that the D-ring forms several hydrophobic interactions in the ligand-binding pocket, although these interactions are less critical than those contributed by the A-ring and the side chain [43]. A modification that enhances VDR-mediated transcriptional activity of the hormone is insertion of a double bond between carbons 16 and 17 in the D-ring [58]. Compounds containing this modification, with or without additional chemical modifications in the side chain, have transcriptional activity significantly greater than that of 1,25(OH)2D3 without a significant increase in their affinity for VDR. In addition, this group of analogs exhibits diminished calcium regulating activity. The unique properties of these compounds include lower affinity for vitamin D– binding protein and slower catabolism [59,60], both of which may increase potency (Table I). It is not known what the unsaturation at C16-C17 is doing to the mode of ligand interaction with the VDR, but dot maps suggest that this modification restricts the flexibility of the side chain, and therefore has the potential to force it into contact points that are distinguishable from those used by the side chain of the natural hormone [61]. In conclusion, the A-ring regulates high affinity for VDR, the D-ring controls ligand uptake and metabolism and possibly contributes to the flexibility of the side chain, and the side chain regulates both affinity for VDR and its transactivation (Fig. 1).

III. DIFFERENTIAL ACTIVATION OF THE VDR BY SYNTHETIC ANALOGS A large amount of information on the structure of the LBDs of various nuclear receptors bound to natural or synthetic ligands has accumulated in the past few years. These studies provide evidence for the structural flexibility of nuclear receptors and a partial explanation for the mechanism of action of agonists, antagonist, and selective receptor modulators. Additional studies, using biochemical, molecular, and cellular biology approaches, provide ample evidence to the many ways by which nuclear receptor actions can be manipulated and refined by synthetic ligands. Interestingly, seven years ago we had little evidence for similar flexibility and heterogeneity in transcriptional activation of the VDR. However, today we are able to distinguish at least three groups of ligands that modulate the VDR

1475

differently from the natural hormone. These include superaognists (such as the 20-epi analogs mentioned above), antagonists, and tissue- or gene-selective agonists. Unfortunately, structural studies provide only partial explanation for the mechanism of action of superagonist, and structural data is not yet available for VDR’s LBD bound to antagonists or to tissue-selective agonists. Therefore, the text below will describe primarily biological and biochemical experiments that provide evidence for the differential activation of the VDR by these three groups of compounds.

A. Superagonists 1. DEFINITION AND ASSESSMENT OF SUPERAGONISTS

Superagonists are analogs that are significantly more transcriptionally potent or efficacious than 1,25(OH)2D3. Earlier studies of analogs’ actions used primarily ligand-binding assays (to determine affinity for the VDR) and cellular assays (to determine antiproliferative activities) [41]. More recently, the identification and assessment of their activities is accomplished by using common molecular endocrinology methodologies to assess transcriptional activation of genes by steroid hormone receptors. When several DNA binding sites for VDR-RXR complexes in the promoters of vitamin D responsive genes were characterized, the most common tool to assess transcriptional activity induced by analogs became reporter gene assays, in which a fusion gene containing a vitamin D response element attached to a heterologous promoter and a reporter gene is transfected into eukaryotic cells that express the VDR [62–64]. A typical dose-response curve of reporter gene expression in cells treated with the natural hormone shows a gradual induction of reporter gene expression that may reach plateau at 10 nM, with an effective dose for 50% of maximal activity (ED50) of 1–5 nM. A superagonist by definition is an analog that induces transcription with an ED50 significantly lower than that of the natural hormone, or that induces severalfold greater maximal transcription [Fig. 4 and ref. 21]. The advantages of this assay are that, unlike the growthinhibitory assay, it directly measures a single VDRmediated transcriptional activity, it is not subject to other vitamin D–mediated signaling that may take place in cell growth assays, and it is not dependent on vitamin D–mediated effects on RNA stability. It is also faster and more reproducible than the growth inhibitory assays. Its disadvantage is that, in eukaryotic cells (especially mammalian cells), transcriptional activity may still depend on analog uptake (which may vary with differences in binding to DBP in the serum) [65]

1476

SARA PELEG

A

B

Transcription

Dimerization

100

20E-1,25D3

80 60

1,25D3

40 20 0

C

120

% Maximal activity

% Maximal activity

120

D

100 80

1,25D3

60 40

20E-1,25D3

20 0

1,25D3

60 40 20

0 .001 .01 .1 1 10 100 1000 10000 Ligand concentration (nM)

GRIP-1 binding 120

1 10 100 1000 10000 0 .001 .01 .1 Ligand concentration (nM)

% Maximal activity

% Maximal activity

120

20E-1,25D3

80

0

0 .001 .01 .1 1 10 100 1000 10000 Ligand concentration (nM)

SRC-1 binding

100

100 80

20E-1,25D3

60 40

1,25D3

20 0

1 10 100 1000 10000 0 .001 .01 .1 Ligand concentration (nM)

FIGURE 4 Transcriptional potency of the superagonist 20-epi-1,25(OH)2D3 is correlated with an increase in dimerization potency but not with an increase in binding to the p160 coactivators SRC-1 and GRIP. Transcription (A) was assessed by co-transfecting CV-1 cells with the hVDR and a reporter containing the osteocalcin VDRE attached to the minimal thymidine kinase promoter and the growth hormone gene. Dimerization (B) binding to SRC (C) and binding to GRIP (D) were assessed by pull-down assays, using GST-fusion protein to bind the 35S-labeled in vitro synthesized hVDR. Bound VDR was eluted, separated by SDS-PAGE, visualized by autoradiography, and quantified by densitometry scanning. These experiments suggest that the dimerization interface rather than the coactivator binding interface is distinct in the VDR-hormone and VDR-20-epi analog complexes [69].

and on the cellular metabolism of the analog [66,67]. For these reasons, it is necessary to confirm differential activation of the VDR by testing analogs in eukaryotic systems that do not contain vitamin D metabolic enzymes (such as yeast) [68], by cell-free transcriptional assays, or by other in vitro assays that examine individual ligand-dependent events that are essential for transcriptional activity in vivo (binding to DNA or interactions with dimerization partners and with other partners of transcription) (Fig. 4) [21,69]. 2. THE 20-NATURAL SUPERAGONISTS

Using these approaches, several analogs have been identified that have greater transcriptional potency than 1,25(OH)2D3. The common feature of these compounds is that they have side-chain modifications but an unmodified A-ring. These analogs’ modifications, such as unsaturation at C-23 or fluorine atoms at C-24, limit their availability to 24-hydroxylase [41].

Additional modifications, such as substitution of hydrogen for fluorine atoms at positions 26 and 27 or homologation at these positions, alter access of these compounds to the 26-hydroxylase in the kidney [70]. However, these modifications also have enormous effect on antiproliferative activity and transcriptional response as determined by reporter gene assays in cultured cells. As mentioned above, another modification, unsaturation at positions 16–17 of the D ring, also increases the transcriptional potency of these compounds. This modification alone, or in combination with the aforementioned side-chain modifications, has been shown to slow down 24-hydroxylation and cause cellular accumulation of active catabolic intermediates [71]. Therefore, it appears that almost all of the increases in the potencies of compounds from this group are due to altered pharmacokinetics, including a decrease in binding to DBP that in vitro causes the cells to take up the analogs faster than 1,25(OH)2D3 [65]. However, several studies

CHAPTER 83 Molecular Basis for Differential Action of Vitamin D Analogs

have provided in vitro evidence for a true change in their ability to induce receptor-mediated activity. First, Uskokovic et al. have shown that the analogs containing the 16-ene modification plus hexafluorine substitutions at the 26 and 27 positions have a twofold greater affinity for the VDR than the natural hormone [71], although this affinity is not proportional to the increase in transcriptional potency (100- to 1000-fold) [72]. Another study has shown that several fluorinated compounds increase the affinity of the VDR-analog complex for its dimerization partner RXR or increase the affinity of the VDR-RXR complex for DNA [72]. Again, these changes are not proportional to the observed increases in transcriptional activities of these compounds in culture, suggesting that these modifications may affect additional parameters of transcriptional activation of the VDR. 3. THE 20-EPI SUPERAGONISTS

The best-characterized modulators of the VDR are the 20-epi analogs, including 20-epi-1α,25(OH)2D3 (MC-1288) and 20-epi-22-oxa-24a,26a,27a-tri-homo1α,25(OH)2D3 (KH-1060) (Table I) [55]. These compounds, synthesized by Leo Pharmaceuticals, share one distinct modification, a stereochemical change at carbon 20. Dot maps revealed that the energy-minimized side chain conformers of these analogs have mostly northwest orientations, instead of the northeast orientations of side chains of the natural hormone and 20-natural analogs [73]. Without chemical modification of the 20-epi side chain, the analog 20-epi 1,25(OH)2D3 (MC-1288) has a hundredfold greater growth-inhibitory and transcriptional activities than 1,25D3 does. The addition of chemical modifications to the 20-epi side chains in the analogs KH-1060 and MC-1301 (20-epi24a,26a,27a-tri-homo-1α,25(OH)2D3) increased their activities up to 3,000-fold compared with that of 1,25D3 [21,55,74]. These phenomenal improvements in activities are not associated with any increase in affinity for the VDR [21,55]. These compounds act directly by altering VDR actions, as they induce reporter gene expression more effectively than 1,25D3 in many cell lines and through several types of vitamin D–response elements [21]. These results occurred in both the presence and the absence of DBP, thus reducing the probability that altered DBP-modulated delivery of these ligands contributes to their potency in culture. It has been shown that 24-hydroxylase does not effectively catabolize several 20-epi analogs [67,75]. However, that they are still significantly more potent than 1,25(OH)2D3 in cells lacking 24-hydroxylase activity (ROS 17/2.8) [21] excludes differential catabolism as a primary explanation for their potency. The most compelling evidence for the

1477

molecular basis of the enhanced activity of these compounds came from a series of studies demonstrating, in a yeast two-hybrid system, that the 20-epi analog KH-1060 induces dimerization more effectively than 1,25(OH)2D3 does [68]. More importantly, by using cell-free assay systems, it was shown that binding of 20-epi analogs to the VDR induces a conformation distinct from that induced by the natural hormone [21]. This distinct conformation is associated with the enhanced ability of VDR-20-epi analog complexes to dimerize with RXR [69,76]; to bind to specific DNA sequences (VDREs) [21]; to interact with a key component of the transcription apparatus, DRIP 205; and to induce transcription in cell-free systems [77] (Fig. 4). These biochemical findings strongly suggest that the contact points of 20-epi analogs with the VDR are different from those used by the natural hormone and that these differences cause conformational changes that affect the properties or availability of VDR binding domains for transcription partners (Fig. 4). Biochemical evidence that the modes of interaction of the natural hormone and the analogs are different came from studies of site-directed mutagenesis of the VDR and from comparing the effects of mutations on binding of the hormone and the 20-epi analogs [78]. These experiments showed that the ability of the 20-epi analogs to interact with a VDR that lacked contact sites essential for binding of 1,25D3 is not impaired. The difference in binding requirement appears to involve residues in the C-terminal region of the VDR, including the AF-2 core/helix 12. This domain, which is essential for coordinating the interaction of coactivators with the VDR, also has contact points for the side chain of 1,25D3 but appears to be less important for the binding of the 20-epi analogs to VDR. Studies demonstrating that the half-lives of the VDR-20-epi analog complexes were significantly greater than the half-lives of VDR complexes with their 20-natural counterparts suggested that the 20-epi analogs may be buried more deeply in the binding pocket than the 20-natural compounds are [78]. Interestingly, structural analysis of the VDR-20-epi analog complexes by X-ray crystallography did not support the biochemical data, because it did not show that the 20-epi analogs use contact points different from those used by 1,25D3, unless the analogs had chemical modifications in their side chains. Furthermore, the structural studies did not provide evidence for a significant change in the functional surface of the VDR-20-epi analog complexes that would explain the modified interactions of these complexes with dimerization partners and with the bridging factor DRIP 205 in vitro [79]. An explanation for this discrepancy could be that 20-epi analogs shift their position in the ligand-binding pocket when the VDR is associated with transcription

1478 A

SARA PELEG

VDR-20-epi 1,25D3 monomers Trypsin

20-epi 1,25D3 (-LOG M)

NL NL 11 10 9

8 7

VDR

34 kDa 32 kDa 28 kDa

B

VDR-20-epi 1,25D3 heterodimers Trypsin

20-epi 1,25D3 (-LOG M)

NL 11 10 9

8

7 34 kDa 32 kDa 28 kDa

FIGURE 5 The effect of VDR interaction with transcriptional partners on the mode of analog binding. (A) In vitro translated VDR was incubated with 20-epi analog (20-epi-1,25(OH)2D3) and then subjected to trypsin digestion. (B) In vitro translated VDR was incubated with the 20-epi analog, with GST-RXR and gluatathione-Sepharose beads. VDR-RXR complexes were separated from the unbound VDR and then subjected to trypsin digestion. Note the differences in the conformation of VDR-analog complexes with RXR and VDR-analog monomers. Also, note the 100-fold increase in the ability of the analog to stabilize the conformation of the heterodimerized VDR [69].

partners [83] (Fig. 5), and these changes are not reflected in the monomer structure that was analyzed by X-ray crystallography. Additional explanation could be that VDR bound to 20-epi analogs in the cells undergoes modifications that enhance its abilities to bind DNA and partners of transcription, and these modifications do not occur in the cell-free systems [80,81].

B. Low-calcemic Analogs/Selective Agonists 1. ASSESSMENT OF NONCALCEMIC SELECTIVE AGONISTS

Surprisingly, a major misconception exists in the vitamin D field because analogs have been defined as noncalcemic or low-calcemic if they have significant receptor binding activity in vitro and growth-inhibitory activity in culture but when administered to animals do not induce hypercalcemia/hypercalciurea at the concentration range at which 1,25(OH)2D3 does [41]. However, low calcemic activity in vivo may simply be due to pharmacokinetic properties such as rapid clearance

rate and short terminal half-life, which would lead to very poor overall biological activities [70]. Therefore, a more accurate definition for low-calcemic analog is a compound that has a wider safety window than 1,25(OH)2D3 to induce a biological response without inducing hypercalcemia [82]. We define analogs that have these desirable qualities in vivo as “selective” agonists because, by definition, the ability to induce biological response without changing calcium homeostasis requires that the analog has a preference for a given target tissue (e.g., tumor cells, the parathyroid gland, and skin) over calcium-regulating organs such as intestine, kidney, and bone. In this review, we will discuss only those low-calcemic selective agonists that have shown evidence of regulating VDR functions differently from 1,25(OH)2D3 in culture and in cell-free systems. 2. MECHANISM OF ACTION OF SELECTIVE AGONISTS

The best-characterized analogs that can be defined as low-calcemic but biologically active at a reasonable concentration range in vivo and as selective modulators of the VDR in vitro are three structurally unrelated compounds (Fig. 6). One is Leo Pharmaceuticals’ EB-1089, a side chain-modified analog (22,24-diene26,27-bishomo-1,25-dihydroxyvitamin D3) that inhibits tumor growth in vivo with half the calcemic activity of 1,25(OH)2D3. Another of these analogs is Chugai’s OCT (22-oxa-1,25-dihydroxyvitamin D3), also a side chain modified analog that inhibits parathyroid hormone secretion without inducing hypercalcemia. The third analog is Roche’s Ro-26-9228 (1αF-16-ene-20epi-23-ene-26,27-bishomo-25-hydroxyvitamin D3), a hybrid analog that restores bone loss without inducing hypercalcemia at a wide concentration range. Examination of the mechanism of action of these apparent selective agonists has raised three questions: (1) Is there evidence that their mode of interaction with and transcription activation of the VDR is significantly different from that of the natural hormone? (2) Is there evidence that gene expression in cells or tissues that regulate calcium homeostasis is modulated differently by the natural hormone and by these analogs? (3) Is there compelling evidence that the analog has tissue or gene preference different from that of 1,25(OH)2D3? The compounds described below each have several of these features. However, additional studies are required to further evaluate and substantiate the mechanisms for their apparent selective actions. Leo Pharmaceutical developed EB-1089 primarily for use in chemotherapy of malignancies, with a focus on breast cancer [83–85]. In vivo, EB-1089 has a reasonable safety window for inhibiting tumor growth without hypercalcemia, but how that occurs is not

1479

CHAPTER 83 Molecular Basis for Differential Action of Vitamin D Analogs

22

24

22 O

OH

H

H

20 23

OH

H

H

OH

H

1 HO 3β

1αOH 1

EB-1089

OH

HO OCT

HO

F Ro-26-9228

FIGURE 6 Structural formulas of selective agonists. EB-1089 was synthesized by Leo Pharmaceuticals; OCT was synthesized by Chugai, and Ro-26-9228 was synthesized by Roche Bioscience.

known [84,86]. There have been few attempts to distinguish EB-1089’s mechanism of VDR activation from VDR-mediated actions of 1,25(OH)2D3. In vitro assays demonstrated that the electrophoretic mobilities of VDR-EB-1089-DNA complexes and VDR1,25(OH)2D3-DNA complexes are different, suggesting that the conformations of the complexes or their compositions are different [87]. Another study showed that EB-1089-VDR complexes have an apparent binding preference for IP9 (inverted repeat with nine intervening nucleotides) VDRE instead of the classic DR3 (direct repeats with three intervening nucleotides) VDREs, whereas 1,25(OH)2D3-VDR complexes do not. Cell culture studies also revealed that EB-1089, but not 1,25(OH)2D3, has some preference for inducing transcription through the IP9 VDRE of a transfected reporter gene [87,88]. Because the response elements used were synthetic and not present in natural genes, the relevance of these findings is not clear, except that they showed differences in the molecular properties of EB-1089-VDR and 1,25(OH)2D3-VDR complexes. Unfortunately, these studies were not extended to test cause and effect relationship between these phenomena and the pharmacological actions of EB-1089 in vitro and in vivo. The results of cell-culture and in vivo studies to determine the mechanism for the low calcemic activity of EB-1089 have been somewhat equivocal and provide only a partial explanation for its selective activities. Cell culture studies have shown that EB-1089 is as good or better than 1,25(OH)2D3 at inducing bone resorption [89]. In vivo studies demonstrated that EB-1089 is somewhat less effective than 1,25(OH)2D3 at inducing 24-hydroxylase and calbindin D9K mRNAs in the duodenum [90]. On the other hand, EB-1089 and 1,25(OH)2D3 have similar abilities to induce these genes in the kidney. These results suggest that EB-1089 has different preference for the duodenum than

1,25(OH)2D3 and that the lower calcemic activity of EB-1089 may be associated with a lower ability to induce gene expression at a site that regulates calcium absorption [90]. It may also suggest that EB-1089 has a different tissue preference in vivo. In conclusion, these in vitro and cell culture studies provide some evidence that EB-1089 is a selective modulator of the VDR, and that its selectivity may be either at the level of target genes or target tissues, but additional studies in vivo and in vitro are necessary to substantiate these findings. Another side chain modified analog, OCT, has been shown to be potent in animals without inducing hypercalcemia [41]. The pharmacokinetic properties of this analog are significantly different from those of 1,25(OH)2D3. First and foremost, OCT binds very poorly to DBP [91], which may contribute to its short half-life in animals. However, there is strong evidence that this analog has a selective action in vivo, as it induces only brief intestinal calcium absorption but prolonged inhibition of parathyroid hormone secretion. The brief intestinal calcium absorption correlates with transient induction of calbindin D9K, a vitamin D receptor-modulated gene. The brief induction of calbindin D9K by OCT is significantly different from the longer 1,25(OH)2D3-dependent induction of the gene [91–93]. Two interesting features about these differences in gene expression between the two compounds are that the analog is retained in the intestine longer and that its maximal binding to the intestinal VDR is higher than that of 1,25(OH)2D3. These findings suggest that the brief period of gene expression is not due to a short half-life of the intestinal VDR-OCT complexes, but perhaps to transcriptional activation events downstream of the formation of VDR-ligand complexes. That OCT is indeed a selective agonist has been supported by in vitro assays that showed differences in

1480

C. Antagonists 1. ASSESSMENT OF ANTAGONISTS

Antagonists are receptor-binding compounds that inhibit the actions of the respective natural hormone and, on their own, are unable to elicit a transcriptional response through the receptor. Therapeutically, these compounds are exceedingly valuable, as they are used to prevent growth of hormone-responsive malignancies (estrogen antagonists) [96–98] and to regulate reproductive processes (progesterone antagonists) [99]. Only a few of these compounds are pure antagonists, but in many cases they may act as antagonists on certain target genes or in certain tissues and as moderate

% Maximal reporter gene expression

A

Transcription in Caco-2 cells 120

100

ED50 1,25D3 = 2 nM Ro-26-9228 = 120 nM

80

1,25D3

60

40 Ro-26-9228 20 0 .0001 .001 .01

.1

1

10

100 1000

Ligand (nM)

B % Maximal reporter gene expression

recruitment of transcriptional coactivators to VDR-OCT and VDR-1,25(OH)2D3 complexes [94]. However, the assays did not provide direct proof that the coactivator selectivity of VDR-OCT complexes in vitro leads to cell-type selective or gene-selective actions in vivo. Another example of an analog with convincing tissue- and cell-selective properties is the Roche compound Ro-26-9228 [82]. Biochemically it is a “hybrid” analog because it contains modifications in both the A-ring and the side chain. Its affinity for VDR is 8- to 10-fold lower than the affinity of the natural hormone, as would be expected from replacement of the 1α-OH group with a fluorine atom. However, examination of its transcriptional activities in cell culture showed that Ro-26-9228 is equipotent to the natural hormone in osteoblasts, whereas in intestinal cells it is 60 times less potent (Fig. 7 and [82]). Further studies of the VDR from the two cell types has revealed that when it binds the analog in intestinal cells, it does not acquire the ability to interact with dimerization partners and transcription coactivators, whereas when the analog binds the VDR from osteoblasts or synthetic VDR in vitro, it does have these abilities. In contrast, the natural hormone 1,25(OH)2D3 has these abilities in both cell types [95]. This apparent cell selectivity in vitro mimics the tissue preference in vivo: the administration of Ro-26-9228 to rats induces VDR–dependent gene expression in the bone but not in the duodenum (Fig. 8 and [82]). These properties of the analog in vivo are associated with prevention of bone loss in osteopenic rats without induction of hypercalcemia over a very wide concentration range. This suggests that poor recognition of the analog by the VDR in the duodenum spares the animals from hypercalcemia induced by enhanced calcium absorption, whereas the analog’s preference for bone (probably for osteoblasts) promotes bone-remodeling activities that lead to a net bone gain [82].

SARA PELEG

Transcription in MG-63 cells 120

100

ED50 1,25D3 = 3 nM Ro-26-9228 = 5 nM

80

60

1,25D3

40 Ro-26-9228 20

0 0 .0001 .001 .01

.1

1

10

100 1000

Ligand (nM)

FIGURE 7

A selective modulation of VDR-mediated transcription in cultured cells. Human colon carcinoma cells (Caco-2) or human osteosarcoma cells, MG-63 were transfected with a reporter gene containing a minimal thymidine kinase promoter and a vitamin D response element. The cells were treated with the indicated doses of 1,25(OH)2D3 or the analog Ro-26-9228, and reporter gene expression was assessed 48 h later. Note that the analog is equipotent to 1,25(OH)2D3 in the osteoblast-like cells but is 60 times less potent than 1,25(OH)2D3 in the intestinal-like cells [82].

agonists in others. These properties opened numerous therapeutic possibilities that are best represented by the synthetic estrogen receptor-binding ligands termed selective estrogen receptor modulators (SERMs) that include tamoxifen and raloxifene [96,97]. Their antagonistic properties are used to inhibit the growth of estrogen-dependent breast cancer cells, while their agonist activities are used to maintain bone integrity in

1481

CHAPTER 83 Molecular Basis for Differential Action of Vitamin D Analogs

TGF-β2 (Bone)

24-hydroxylase (Duodenum) 8 * 8

6

4

2

0 Sham

OVX

1,25D3 Ro-26-9228 (0.2 µg/kg) (5 µg/kg)

Densitometry units (normalized)

Densitometry units (normalized)

10

*

6

*

4

2

0 Sham

OVX

1,25D3 Ro-26-9228 (0.2 µg/kg) (5 µg/kg)

FIGURE 8

A selective modulation of gene expression in target tissues for vitamin D action by the analog Ro-26-9228, in vivo. Female rats were either sham-operated or ovariectomized for three weeks and then given vehicle, or the indicated dose of the vitamin D compound. Total RNA was isolated from the duodenum and the tibia of these animals 7 h after treatment. The mRNAs were quantified by northern blots (24-hydroxylase) or by semiquantitative RT-PCR (TGF-β2). Note that Ro-26-9228 did not induce a significant gene expression in the duodenum whereas it did so in the bone. In contrast 1,25(OH)2D3 induced expression of these genes in both tissues [82].

postmenopausal women. At the molecular level, binding of antagonists to their respective receptors induces a conformational change that is not permissive for recruitment of coactivators. In fact, antagonist binding prevents coactivator binding in vitro by causing a conformational change in the position of the C-terminal AF-2 core [46]. The cellular conditions that render these complexes transcriptionally active are debated but include changing ratios of coactivators and corepressors, and interaction of another transcription activation domain (the N-terminal AF-1) with the transcription apparatus under conditions that are not permissive for this interaction through the AF-2 domain of the antagonist-bound receptor (Fig. 3 and [98,99]). Numerous vitamin D analogs have been synthesized, but until five years ago none exhibited clear antagonistic activities. One reason could be the remarkable structural flexibility of vitamin D derivatives (unlike the relatively rigid four-ring steroid hormones) [100]. This flexibility may also allow these ligands to adapt their conformations to the binding pocket of the VDR despite many structural modifications, and therefore vitamin D analogs may be less likely to disrupt formation of a functional VDR surface. Another possibility is that, for practical reasons, vitamin D antagonists are not as obviously useful clinically as selective agonists might be, and therefore have not been synthesized strategically or have not been investigated. Despite these limitations, there are two groups of vitamin D analogs with significant

antagonistic activities. The best representative of the first group is the 26,23-lactone TEI 9647, synthesized at the Teijin Institute for Biomedical Research. The other group of antagonists is the carboxylic esters, the most potent representative of which is ZK 159222, synthesized at Schering AG (Fig. 9). 2. THE LACTONES

The lactone analog TEI 9647 (23S-25-dehydro1α-hydroxyvitamin D3-26,23-lactone) has an affinity for VDR which is one-tenth that of 1,25(OH)2D3 [101]. Its mode of interaction with the VDR is different from that of 1,25(OH)2D3, because the analog stabilizes a conformation that is significantly different from that stabilized by 1,25(OH)2D3 [102]. This distinct Antagonists OH O 23

H

O

22

OH TEI 9647

26 O

O

H

HO

24

H H

HO

OH

ZK 159222

FIGURE 9 Structural formulas of antagonists: TEI 9647 was synthesized by Teijin, and ZK 159222 was synthesized by Schering.

1482 conformation is evident in the VDR monomer and in the VDR complexes with RXR and with DNA [103,104]. It is possible that the antagonist activities are due to disruption of the conformation of helix 12 by the bulky lactone group at the side chain of this compound. This disruption causes the inactivation of the AF-2 domain (Fig. 3) and a loss of transcriptional activity of the antagonist-bound VDR. However, X-ray crystallography of the VDR bound to this ligand confirms that hypothesis has not yet been performed. Despite the lack of structural information, this analog can effectively inhibit 1,25(OH)2D3-mediated transcriptional activity in COS-7 cells cotransfected with the human VDR and a reporter gene containing the 24-hydroxylase VDREs [103]. It also inhibits 1,25(OH)2D3-mediated transcription in Saso-2 cells transfected with the same reporter but with the endogenous VDR, and in MCF-7 cells transfected with a reporter gene containing a single DR3-type VDRE [103]. The antagonistic activity of TEI 9647 seems to involve two LBD-mediated actions: the analog appears to inhibit 1,25(OH)2D3-induced VDR-RXR dimerization and to inhibit interaction of VDR with the p160 coactivator SRC-1 [103,104]. Using HL60 cells as a model for the growth inhibitory and differentiating actions of 1,25(OH)2D3, cellular responses that are largely considered to be VDRmediated, it was shown that TEI 9647 did not induce any of the differentiation markers induced by 1,25(OH)2D3 and prevented all aspects of the growth-inhibitory and differentiating actions of 1,25(OH)2D3 [101]. These results underscore the notion that growth inhibition and differentiation induced by 1,25(OH)2D3 are indeed mediated through the nuclear VDR, although some aspects of these cellular processes are thought to be mediated through a distinct membrane receptor for 1,25D3 (see Chapter 23). TEI 9647 was examined for antagonistic and agonistic activities in normocalcemic rats and in vitamin D–and calcium-depleted rats [105]. TEI 9647 has a moderate ability to inhibit the increase in serum calcium in normocalcemic rats injected with pharmacological amounts of 1,25(OH)2D3, but this analog does not inhibit the normal physiological activities of 1,25(OH)2D3. In vitamin D–deficient and calcium-deficient rats, TEI 9647 acts as a poor agonist of intestinal calcium absorption and a somewhat better agonist of bone resorption, and it very effectively inhibits parathyroid hormone secretion. Interestingly, when given together with 1,25(OH)2D3 to the vitamin D–deficient rats, TEI 9647 can inhibit the calcium-absorbing and bone-resorbing activities of 1,25(OH)2D3 as well as the 1,25(OH)2D3-mediated inhibition of parathyroid hormone secretion [105]. These results suggest that under normal physiological conditions in vivo, TEI 9647 is a poor antagonist.

SARA PELEG

In contrast, in vitro it can antagonize a wide range of 1,25(OH)2D3-mediated VDR actions without a target gene or target tissue preferences. Therefore, it is not likely that TEI 9647 has mixed agonist-antagonist activities that would make it useful in a clinical setting. However, all of the in vitro experiments on TEI 9647 were performed with human cells or recombinant human VDR, and the in vivo experiments were performed with rats. One might speculate that the analog might not be an effective antagonist of the rat VDR as it is of the human VDR, and therefore its actions in the rat may differ from those in humans. 3. THE 26-CARBOXYLIC ESTERS

The other group of antagonists is the carboxylic esters ZK 159222 and ZK 168281 [106]. These compounds appear to act as typical antagonists of nuclear receptors in vitro and in cultured cells, and in very high doses they act as poor agonists. In the presence of 1,25D3, they inhibit VDR-mediated transcriptional activities. In vitro they induce a VDR conformation different from that induced by 1,25(OH)2D3, a clear indication of a difference in the mode of interaction with the receptor. These analogs’ binding to VDR does not prevent dimerization and binding to DNA (as with the lactone compounds), but they do not induce interaction with transcription coactivators of the p160 family and they partially inhibit the 1,25(OH)2D3-mediated interaction of VDR with these factors [106,107]. These results suggest that the step in VDR activation that is disrupted by the 26-carboxylic esters is the induction of a VDR conformation that allows interaction with the coactivators. Since this is an AF-2-dependent function, it suggests that the long side chain of these analogs disrupts the agonist conformation of the helix 12/AF2 core in the VDR [108]. Without X-ray crystallography data, however, it is not possible to determine whether the AF-2 core in these VDR-analog complexes assumes an antagonist conformation such as that seen in the AF-2 core in estrogen receptor-tamoxifen complexes (e.g., the coactivators binding site is masked) or simply is not able to properly interact with coactivators because its conformation is similar to that of the AF-2 core in the unoccupied receptor (Fig. 3). So far, tissue- or genespecific activities of these antagonists have not been identified, either in culture or in animal studies.

IV. CLINICAL SIGNIFICANCE FOR SELECTIVE MODULATION OF THE VDR BY VITAMIN D ANALOGS Vitamin D metabolites (Calderol of Organon, Rocaltrol of Hoffmann-LaRoche) and analogs that are

CHAPTER 83 Molecular Basis for Differential Action of Vitamin D Analogs

in fact prehormones of 1,25(OH)2D3 (One-alpha of Leo Pharmaceuticals, Hectoral of Bone Care International) have been used for years for treatment of senile osteoporosis, postmenopausal osteoporosis, secondary hyperparathyroidism, and the skin disease psoriasis [109–116]. These usages suggest that new analogs of vitamin D will be developed and used first and foremost for treatment of these conditions. For instance, the relatively new analog Dovonex (synthesized at Leo Pharmaceuticals) has been used in the past few years for treatment of psoriasis [117–119] and Zemplar (19nor-1,25(OH)2D3, Abbott Laboratories) and maxacalcitol (OCT, synthesized at Chugai) have recently been approved for use in secondary hyperparathyroidism [120–122]. Of these compounds, however, only OCT has been established as a selective modulator of the VDR. Other interesting analogs are in clinical trials for osteoporosis, including ED-71 of Chugai [123,124] and Roche’s selective modulator Ro-26-9228. Both analogs have been carried to phase II clinical trials in postmenopausal osteoporosis and appear to be well tolerated and effective [82]. The analog Ro-26-9228 (also named BXL628) is currently being tested in clinical trials for treatment of benign prostate hyperplasia (BPH), and it will be used in clinical trials for posttransplantation immunosuppression through the next three years (Dr. L. Adorini, personal communication, and unpublished results). Because the side effects common to drugs used for treatment of these two conditions include bone loss, this bone-protecting vitamin D analog may have a dual beneficial effect for BPH and transplantation patients. Until recently, there did not seem to be a specific clinical use for vitamin D antagonists. However, recent studies on Paget’s disease have suggested a specific increase in osteoclasts’ sensitivity to the differentiating action of 1,25(OH)2D3 as the principal underlying mechanism for abnormal bone formation in patients with this disease. These findings suggest that potent vitamin D antagonists might be useful drugs to inhibit the abnormal activation of osteoclasts in this disease [125,126]. Another clinical condition that has not yet been explored is the osteolytic form of metastatic bone disease. This form of bone metastases is common in multiple myeloma and renal carcinoma, and is associated with massive activation of osteoclast actions. Osteoclast activation in osteolytic metastatic bone disease may be coupled to osteoblasts’ production of cytokines that promote osteoclast differentiation and function (primarily RANKL) [127]. Therefore, it remains to be examined whether 1,25(OH)2D3-regulated cytokine production by osteoblasts can be blocked in this form of metastatic bone disease by vitamin D antagonists, without adverse effects on calcium homeostasis and bone turnover.

1483

One of the most attractive pharmacological features of vitamin D analogs is their ability to inhibit malignant cell growth in vitro and in animal models. These effects depend on the ability of these compounds to induce cell differentiation or apoptosis and to inhibit angiogenesis in vivo [85,128,129]. One of the compounds most thoroughly studied in that respect is the selective modulator EB-1089 (secocalcitol of Leo Pharmaceuticals). In preclinical studies, it inhibited various malignancies including breast, colon, and prostate cancers [84,86,130,131]. EB-1089 even has a significant effect on progression of breast cancer cells into bone in nude mice [132]. In culture and in vivo, EB-1089 appears to have significant differentiating and apoptotic effects and its inhibition of breast cancer cells metastasis into bone suggests that it is also effective on angiogenesis, an important parameter of tumor progression. Interestingly, EB-1089 has been successful in clinical trials of advanced liver cancer and is also being tested in advanced pancreatic and breast cancer [133,134]. Although EB-1089 is efficacious in animal models of malignancies and also has a significant effect on several human malignancies, it still exhibits significant calcemic. Consequently activity, its therapeutic window is somewhat limited [130,133]. Therefore, the future development of additional low-calcemic analogs that are potent inhibitors of tumor cell growth and have a wider therapeutic window will be necessary to further establish vitamin D analogs in cancer therapy.

References 1. Norman AW 1979 Vitamin D: The Calcium Homeostatic Steroid Hormone. Academic Press: New York, pp. 1–490. 2. Lawson DEM, Fraser DR, Kodicek E, Morris HR, Williams DH 1971 Identification of 1,25-dihydroxycholecalciferol, a new kidney hormone controlling calcium metabolism. Nature 230:228–230. 3. Holick MF, Schnoes HK, DeLuca HF 1971 Identification of 1,25-dihydroxycholecalciferol, a form of vitamin D3 metabolically active in the intestine. Proc Natl Acad Sci USA 68:803–804. 4. Norman AW, Myrtle JF, Midgett RJ, Nowicki HG, William V, Popjak G 1971 1,25-Dihydroxycholecalciferol: Identification of the proposed active form of vitamin D3 in the intestine. Science 173:51–54. 5. Bikle DD, Pillai S 1993 Vitamin D, calcium, and epidermal differentiation. Endocr Rev 14:3–19. 6. Takahashi N, Akatsu T, Sasaki T, Nicholson GC, Moseley JM, Martin TJ, Suda T 1988 Induction of calcitonin receptor by 1a,25-dihydroxy vitamin D3 in osteoclast-like multinucleated cells from mouse bone marrow cells. Endocrinology 123:1504–1510. 7. Udagawa N, Takahashi N, Akatsu T, Tanaka H, Sasaki T, Nishi-hara T, Koga T, Martin TJ, Suda T 1990 Origin of osteoclasts: Mature monocytes and macrophages are capable of differentiating into osteoclasts under suitable microenvironment

1484

8.

9.

10. 11.

12.

13.

14. 15. 16.

17. 18.

19.

20.

21.

22.

23.

24.

prepared by bone marrow-derived stromal cells. Proc Natl Acad Sci USA 87:7260–7264. Roodman GD, Ibbotson KJ, MacDonald BR, Kuehl TJ, Mundy GR 1985 1,25-Dihydroxy vitamin D3 causes formation of multinucleated cells with several osteoclast characteristics in cultures of primate marrow. Proc Natl Acad Sci USA 82:8213–8217. Takahashi N, Yamana H, Yoshiki S, Roodman GD, Mundy GR, Jones SJ, Boyde A, Suda T 1988 Osteoclast-like cell formation and its regulation by osteotropic hormones in mouse bone marrow cultures. Endocrinology 122:1373–1382. Amento EP 1987 Vitamin D and the immune system. Steroids 49:55–72. Kragballe K 1993 Vitamin D3 analogs in psoriasis—Clinical use and mode of action. In: Bernard BA, Schroot B (eds) Molecular Biology to Therapeutics. Birkhaeuser, Basel, pp. 174–181. Abe E, Miyaura C, Sakagami H, Takeda M, Konno K, Yamazaki T, Yoshiki S, Suda T 1981 Differentiation of mouse myeloid leukemia cells induced by 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 78:4990–4994. Bar-Shavit Z, Teitelbaum SL, Reitsma P, Hall A, Pegg LE, Trial J, Kahn AJ 1983 Induction of monocytic differentiation and bone resorption by 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 80:5907–5911. Walters MR 1992 Newly identified actions of the vitamin D endocrine system. Endocr Rev 13:719–764. Frampton RJ, Omond SA, Eisman JA 1983 Inhibition of human cancer cell growth by 1,25-dihydroxyvitamin D3 metabolites. Cancer Res 43:4443–4447. Saez S, Falette N, Gullot C, Meggouh F, Lefebvre MF, Crepin M 1993 1,25(OH)2D3 modulation of mammary tumor cell growth in vitro and in vivo. Breast Cancer Res Treat 27:69–81. Eisman JA, Barkla DH, Tutton PJM 1987 Suppression of in vivo growth of human cancer solid tumor xenografts by 1,25-dihydroxyvitamin D3. Cancer Res 46:21–25. Haussler MR, Mangelsdorf DJ, Komm BS, Terpenning CM, Yamaoka K, Allegretto EA, Baker AR, Shine J, McDonnell DR, Hughes M, et al. 1988 Molecular biology of the vitamin D hormone. Recent Prog Horm Res 44:263–305. Mangelsdorf DJ, Thummel C, Beato M, Herrlick P. Schutz G, Umensono K, Blumberg B, Kastner P, Mark M, Chambon P, Evans RM 1995 The nuclear receptor superfamily: The second decade. Cell 83:835–839. Pike JW 1992 Molecular mechanism of cellular response to the vitamin D3 hormone. In: Coe FL, Favus MJ (eds) Disorders of Bone and Mineral Metabolism. Raven: New York, pp. 163–193. Peleg S, Sastry M, Collins ED, Bishop JE, Norman AW 1995 Distinct conformational changes induced by 20-epi analogs of 1a,25-dihydroxyvitamin D3 are associated with enhanced activation of the vitamin D receptor. J Biol Chem 270: 10551–10558. Towers TL, Luisi BF, Asianov A, Freedman LP 1993 DNA target selectivity by the vitamin D3 receptor: Mechanism of dimer binding to an asymmetric repeat element. Proc Natl Acad Sci USA 90:6310–6314. Yu V, Delsert C, Anderson B, Holloway JM, Davary OV, Naar AM, Kirn SY, Bouton J-M, Glass CK, Rosenfeld MG 1991 RXR/3: A coregulator that enhances binding of retinoic acid, thyroid hormone, and vitamin D receptors to their cognate response elements. Cell 67:1251–1266. Cheskis B, Freedman LP 1994 Ligand modulates the conversion of DNA-bound vitamin D3 receptor (VDR) homodimers

SARA PELEG

25.

26. 27. 28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

into VDR-retinoid X receptor heterodimers. Mol Cell Biol 14:3329–3338. Kephart DD, Walfish PG, DeLuca H, Butt TR 1996 Retinoid X receptor isotype identity directs human vitamin D receptor heterodimer transactivation from the 24-hydroxylase vitamin D response elements in yeast. Mol Endocrinol 10:407–419. Cheng HJ, Pike JW 1996 Human vitamin D receptordependent transactivation in Saccharomyces cerevisiae requires retinoid X receptor. Mol Endocrinol 10:196–205. Hannah SS, Norman AW 1994 1a,25(OH)2-vitamin D3 regulated expression of the eukaryotic genome. Nutr Rev 52:376–381. Masuyama H, Brownfield CM, St-Arnaud R, MacDonald PN 1997 Evidence for ligand-dependent intramolecular folding of the AF-2 domain in vitamin D receptor-activated transcription and coactivator interaction. Mol Endocrinol 11:1507–1517. Whitfield GK, Hsieh JC, Nakajima S, MacDonald PN, Thompson PD, Jurutka PW, Haussler CA, Haussler MR 1995 A highly conserved region in the hormone-binding domain of the human vitamin D receptor contains residues vital for heterodimerization with retinoid X receptor and for transcriptional activation. Mol Endocrinol 9:1166–1179. Rachez C, Suldan Z, Ward J, Chang CP, Burakov D, Erdjument-Bromage H, Tempst P, Freedman LP 1998 A novel protein complex that interacts with the vitamin D3 receptor in a ligand-dependent manner and enhances VDR transactivation in a cell-free system. Genes Dev 12:1787–1800. Rachez C, Lemon BD, Suldan Z, Bromleigh V, Gamble M, Naar AM, Erdjument-Bromage H, Tempst P, Freedman LP 1999 Ligand-dependent transcription activation by nuclear receptors requires the DRIP complex. Nature 398: 824–828. Koren R, Ravid A, Liberman UA, Hochberg Z, Weisman Y 1985 Defective binding and function of 1,25-dihydroxyvitamin D3 receptors in peripheral mononuclear cells from patients with end-organ resistance to 1,25-dihydroxyvitamin D3. J Clin Invest 76:2012–2015. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia, and growth retardation after weaning. Nat Genet 16:391–396. Van Cromphaut SJ, Dewerchin M, Hoenderop JG, Stockmans I, Van Herck E, Kato S, Bindels RJ, Collen D, Carmeliet P, Bouillon R, Carmeliet G 2001 Duodenal calcium absorption in vitamin D receptor-knockout mice: functional and molecular aspects. Proc Natl Acad Sci USA 98:13324–13329. Chen CH, Sakai Y, Demay MB 2001 Targeting expression of the human vitamin D receptor to the keratinocytes of vitamin D receptor null mice prevents alopecia. Endocrinology 142:5386–5389. Haussler MR, Donaldson CA, Marion SL, Allgretto EA, Kelly MA, Mangelsdorf DJ, Pike JW 1986 Receptors for the vitamin D hormone: Characterization and functional involvement in 1,25-dihydroxyvitamin D3-regulated events. In: Gotto M, O’Malley BW (eds) The Role of Receptors in Biology and Medicine. Raven: New York, pp. 91–104. Mangelsdorf DJ, Koeffler HP, Donaldson CA, Pike JW, Haussler MR 1984 1,25-dihydroxyvitamin D3–induced differentiation in a human promyelocytic leukemic cell line (HL-60): Receptor-mediated maturation to macrophage-like cells. Endocrinology 98:391–398.

CHAPTER 83 Molecular Basis for Differential Action of Vitamin D Analogs

38. Zinser GM, Sundberg JP, Welsh J 2002 Vitamin D3 receptor ablation sensitizes skin to chemically induced tumorigenesis. Carcinogenesis 23:2103–2109. 39. Welsh J, Wietzke JA, Zinser GM, Smyczek S, Romu S, Tribble E, Welsh JC, Byrne B, Narvaez CJ 2002 Impact of the vitamin D3 receptor on growth-regulatory pathways in mammary gland and breast cancer. J Steroid Biochem Mol Biol 83:85–92. 40. Kallay E, Pietschmann P, Toyokuni S, Bajna E, Hahn P, Mazzucco K, Bieglmayer C, Kato S, Cross HS 2001 Characterization of a vitamin D receptor knockout mouse as a model of colorectal hyperproliferation and DNA damage. Carcinogenesis 22:1429–1435. 41. Bouillon R, Okamura WH, Norman AW 1995 Structurefunction relationship in the vitamin D endocrine system. Endocr Rev 16:200–257. 42. Fraser DR 1980 Regulation of the metabolism of vitamin D. Physiol Rev 60:551–613. 43. Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179. 44. Nakajima S, Yamagata M, Sakai N, Ozono K 1998 Characterization of the activation function-2 domain of the human 1,25-dihydroxyvitamin D3 receptor. Mol Cell Endocrinol 139:15–24. 45. Masuyama H, Brownfield CM, St-Arnaud R, MacDonald PN 1997 Evidence for ligand-dependent intramolecular folding of the AF-2 domain in vitamin D receptor-activated transcription and coactivator interaction. Mol Endocrinol 11:1507–1517. 46. Brzozowski AM, Pike AC, Dauter Z, Hubbard RE, Bonn T, Engstrom O, Ohman L, Greene GL, Gustafsson JA, Carlquist M 1997 Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 389:753–758. 47. Shiau AK, Barstad D, Loria PM, Cheng L, Kushner PJ, Agard DA, Greene GL 1998 The structural basis of estrogen receptor/coactivator recognition and the antagonism of this interaction by tamoxifen. Cell 95:927–937. 48. Gangloff M, Ruff M, Eiler S, Duclaud S, Wurtz JM, Moras D 2001 Crystal structure of a mutant hER alpha ligand-binding domain reveals key structural features for the mechanism of partial agonism. J Biol Chem 276:15059–15065. 49. Steinmetz AC, Renaud JP, Moras D 2001 Binding of ligands and activation of transcription by nuclear receptors. Annu Rev Biophys Biomol Struct 30:329–359. 50. Norman AW, Okamura WH, Farach-Carson MC, Allewaret K, Branisteanu D, Nemere I, Raman-Muralidharan K, Boullion R 1993 Structure-function studies of 1,25-dihydroxyvitamin D3 and the vitamin D endocrine system. 1,25-Dihydroxypentadeuterio-previtamin D3 (as a 6-s cis analog) stimulates nongenomic but not genomic biological responses. J Biol Chem 268:13811–13819. 51. Napoli JL, Pramanik BC, Royal PM, Reinhardt TA, Horst RL 1983 Intestinal synthesis of 24-keto 1,25 dihydroxyvitamin D3. J Biol Chem 158:9100–9107. 52. Zhou JY, Norman AW, Akashi M, Chen D-L, Uskokovic MR, Aurrecoechea JM, Dauben WG, Okamura WH, Koeffler HP 1991 Development of a novel l,25(OH)2-vitamin D3 analog with potent ability to induce HL-60 cell differentiation without modulating calcium metabolism. Blood 12:75–82. 53. Kristjansson K, Rut AR, Hewison M, O’Riordan JL, Hughes MR 1993 Two mutations in the hormone-binding domain of the vitamin D receptor cause tissue resistance to 1,25-dihydroxyvitamin D3. J Clin Invest 92:12–16. 54. Takayama H, Kittka A, Fujishima T, Suhara Y 2003 Design, synthesis, and biological studies of the A-ring modified

55.

56.

57.

58.

59.

60.

61.

62.

63.

64. 65.

66.

67.

1485

1,25-dihydroxyvitamin D3 analogs. Recent Results in Cancer Research 164:289–317. Binderup L, Latini S, Binderup E, Bretting C, Calverley M, Hansen K 1991 20-EPI-Vitamin D3 analogs: A novel class of potent regulators of cell growth and immune responses. Biochem Pharmacol 42:1569–1575. Peleg S, Liu YY, Reddy S, Horst RL, White MC, Posner GH 1996 A 20-epi side chain restores growth-regulatory and transcriptional activities of an A-ring modified hybrid analog of 1alpha,25-dihydroxyvitamin D3 without increasing its affinity to the vitamin D receptor. J Cell Biochem 63:149–161. Peleg S, Nguyen C, Woodard BT, Lee JK, Posner GH 1998 Differential use of transcription activation function 2 domain of the vitamin D receptor by 1alpha,25-dihydroxyvitamin D3 and its A-ring modified analogs. Mol Endocrinol 12:525–535. Ferrara J, McCuaing K, Hendy GN, Uskokovic M, White JH 1994 Highly potent transcriptional activation by 16-ene derivatives of 1,25-dihydroxyvitamin D3. J Biol Chem 269:2971–2981. Uskokovic MR, Baggiolini E, Shuiey S-J, Lacobelli J, Hennessy B, Rriegiel J, Daniewski AR, Pizzolato G, Courtney LF, Horst RL 1991 The 16-ene analogs of 1,25-dihydroxycholecalciferol. Synthesis and Biological Activity. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D: Gene Regulation, Structure-Function Analysis, and Clinical Applications, de Gruyter: New York, pp. 139–156. Reddy GS, Clark JW, Tseng K-Y, Uskokovic MR, McLane JA 1993 Metabolism of l,25(OH)2-16-ene D3 in kidney: Influence of structural modification of D-ring on side chain metabolism. Bioorg Med Chem Lett 3:1879–1884. Yamada S, Yamamoto K, Masuno H, Ohta M 1998 Conformation-function relationship of vitamin D: conformational analysis predicts potential side-chain structure. J Med Chem 23:1467–1475. Kerner SA, Scott RA, Pike JW 1989 Sequence elements in the human osteocalcin gene confer basal activation and inducible response to hormonal vitamin D3. Proc Natl Acad Sci USA 86:4455–4459. Noda M, Vogel RL, Craig AM, Prahl J, DeLuca HF, Denhardt DT 1990 Identification of a DNA sequence responsible for binding of the 1,25-dihydroxyvitamin D3 receptor and 1,25-dihydroxyvitamin D3 enhancement of mousesecreted phosphoprotein 1 (SPP-1 or osteopontin) gene expression. Proc Natl Acad Sci USA 87:9995–9999. Zierold C, Darwish HM, DeLuca HF 1995 Two vitamin D response elements function in the rat 1,25-dihydroxyvitamin D 24-hydroxylase promoter. J Biol Chem 270:1675–1678. Imai Y, Pike JW, Koeffler HP 1995 Potent vitamin D3 analogs: their abilities to enhance transactivation and to bind to the vitamin D3 response element. Leuk Res 19: 147–158. Siu-Caldera ML, Rao DS, Astecker N, Weiskopf A, Vouros P, Konno K, Fujishima T, Takayama H, Peleg S, Reddy GS 2001 Tissue-specific metabolism of 1alpha,25-dihydroxy20-epi-Vitamin D3 into new metabolites with significant biological activity: studies in rat osteosarcoma cells (UMR 106 and ROS 17/2.8). J Cell Biochem 82:599–609. Siu-Caldera ML, Sekimoto H, Peleg S, Nguyen C, Kissmeyer AM, Binderup L, Weiskopf A, Vouros P, Uskokovic MR, Reddy GS 1999 Enhanced biological activity of 1alpha,25-dihydroxy-20-epi-vitamin D3, the C-20 epimer of 1alpha,25-dihydroxyvitamin D3, is in part due to its metabolism into stable intermediary metabolites with significant biological activity. J Steroid Biochem Mol Biol 71:111–121.

1486 68. Zhao XY, Eccleshall TR, Krishnan AV, Gross C, Feldman D 1997 Analysis of vitamin D analog-induced heterodimerization of vitamin D receptor with retinoid X receptor using the yeast two-hybrid system. Mol Endocrinol 11:366–378. 69. Liu Y-Y, Nguyen C, Peleg S 2000 Regulation of ligandinduced heterodimerization and coactivator interaction by the activation function-2 domain of the vitamin D receptor. Mol Endocrinol 14:1776–1787. 70. Jones G. Analog metabolism. 1997 In: D Feldman, FH Glorieux, JW Pike (eds) Vitamin D. Academic Press: San Diego CA, pp. 973–994. 71. Uskokovic MR, Studzinski GP, Reddy SG 1997 The 16-ene vitamin D analogs. In: D Feldman, FH Glorieux, JW Pike (eds) Vitamin D. Academic Press, San Diego CA, pp. 1045–1070. 72. Cheskis B, Lemon BD, Uskokovic M, Lomedico PT, Freedman LP 1995 Vitamin D3-retinoid X receptor dimerization, DNA binding, and transactivation are differentially affected by analogs of 1,25-dihydroxyvitamin D3. Mol Endocrinol 9:1814–1824. 73. Yamamoto K, Sun WY, Ohta M, Hamada K, DeLuca HF, Yamada S 1996 Conformationally restricted analogs of 1alpha,25-dihydroxyvitamin D3 and its 20-epimer: compounds for study of the three-dimensional structure of vitamin D responsible for binding to the receptor. J Med Chem 39:2727–2737. 74. Elstner E, Linker-Israeli M, Said J, Umiel T, de Vos S, Shintaku IP, Heber D, Binderup L, Uskokovic M, Koeffler HP 1995 20-epi-vitamin D3 analogs: a novel class of potent inhibitors of proliferation and inducers of differentiation of human breast cancer cell lines. Cancer Res 55: 2822–2830. 75. Dilworth FJ, Calverley MJ, Makin HL, Jones G 1994 Increased biological activity of 20-epi-1,25-dihydroxyvitamin D3 is due to reduced catabolism and altered protein binding. Biochem Pharmacol 47:987–993. 76. Liu Y-Y, Nguyen C, Gardezi AS, Schnirer I, Peleg S 2001 Differential regulation of heterodimerization by 1alpha,25dihydroxyvitamin D3 and its 20-epi analog. Steroids 66:203–212. 77. Yang W, Freedman LP 1999 20-epi Analogs of 1,25-dihydroxyvitamin D3 are highly potent inducers of DRIP coactivator complex binding to the vitamin D3 receptor. J Biol Chem 274:16838–16845. 78. Liu Y-Y, Collins ED, Norman AW, Peleg S 1997 Differential interaction of 1alpha,25-dihydroxyvitamin D3 analogs and their 20-epi homologs with the vitamin D receptor. J Biol Chem 272:3336–3345. 79. Tocchini-Valentini G, Rochel N, Wurtz JM, Mitschler A, Moras D 2001 Crystal structures of the vitamin D receptor complexed to superagonist 20-epi ligands. Proc Natl Acad Sci USA 98:5491–5496. 80. van den Bemd GJ, Jhamai M, Staal A, van Wijnen AJ, Lian JB, Stein GS, Pols HA, van Leeuwen JP 2002 A central dinucleotide within vitamin D response elements modulates DNA binding and transactivation by the vitamin D receptor in cellular response to natural and synthetic ligands. J Biol Chem 277:14539–14546. 81. Yamamoto H, Shevde NK, Warrier A, Plum LA, DeLuca HF, Pike JW 2003 2-Methylene-19-nor-(20S)-1,25-dihydroxyvitamin D3 potently stimulates gene-specific DNA binding of the vitamin D receptor in osteoblasts. J Biol Chem 278:31756–31765. 82. Peleg S, Uskokovic M, Ahene A, Vickery B, Avnur Z 2002 Cellular and molecular events associated with the

SARA PELEG

83.

84.

85.

86.

87. 88.

89.

90.

91.

92.

93.

94.

95.

96. 97.

bone-protecting activity of the noncalcemic vitamin D analog Ro-26-9228 in osteopenic rats. Endocrinology 143:1625–1636. Danielsson C, Mathiasen IS, James SY, Nayeri S, Bretting C, Hansen CM, Colston KW, Carlberg C 1997 Sensitive induction of apoptosis in breast cancer cells by a novel 1,25-dihydroxyvitamin D3 analog shows relation to promoter selectivity. J Cell Biochem 66:552–562. Koshizuka K, Koike M, Asou H, Cho SK, Stephen T, Rude RK, Binderup L, Uskokovic M, Koeffler HP 1999 Combined effect of vitamin D3 analogs and paclitaxel on the growth of MCF-7 breast cancer cells in vivo. Breast Cancer Res Treat 53:113–120. Sundaram S, Chaudhry M, Reardon D, Gupta M, Gewirtz DA 2000 The vitamin D3 analog EB 1089 enhances the antiproliferative and apoptotic effects of Adriamycin in MCF-7 breast tumor cells. Breast Cancer Res Treat 63:1–10. Akhter J, Chen X, Bowrey P, Bolton EJ, Morris DL 1997 Vitamin D3 analog, EB1089, inhibits growth of subcutaneous xenografts of the human colon cancer cell line, Lovo, in a nude mouse model. Dis Colon Rectum 40:317–321. Quack M, Carlberg C 1999 Selective recognition of vitamin D receptor conformations mediates promoter selectivity of vitamin D analogs. Mol Pharmacol 55:1077–1087. Schrader M, Kahlen JP, Carlberg C 1997 Functional characterization of a novel type of 1alpha,25-dihydroxyvitamin D3 response element identified in the mouse c-fos promoter. Biochem Biophys Res Commun 230:646–651. Wiberg K, Ljunghall S, Binderup L, Ljunggren O 1995 Studies on two new vitamin D analogs, EB 1089 and KH 1060: effects on bone resorption and osteoclast recruitment in vitro. Bone 17:391–395. Roy S, Martel J, Tenenhouse HS 1995 Comparative effects of 1,25-dihydroxyvitamin D3 and EB 1089 on mouse renal and intestinal 25-hydroxyvitamin D3-24-hydroxylase. J Bone Miner Res 10:1951–1959. Dusso AS, Negrea L, Gunawardhana S, Lopez-Hilker S, Finch J, Mori T, Nishii Y, Slatopolsky E, Brown AJ 1991 On the mechanisms for the selective action of vitamin D analogs. Endocrinology 128:1687–1692. Brown AJ, Finch J, Grieff M, Ritter C, Kubodera N, Nishii Y, Slatopolsky E 1993 The mechanism for the disparate actions of calcitriol and 22-oxacalcitriol in the intestine. Endocrinology 133:1158–1164. Furuichi T, Kawata S, Asoh Y, Kumaki K, Ohyama Y 1998 Differential time course of induction of 1alpha,25-dihydroxyvitamin D3-24-hydroxylase mRNA expression in rats by 1alpha,25-dihydroxyvitamin D3 and its analogs. Life Sci 62:453–459. Takeyama K, Masuhiro Y, Fuse H, Endoh H, Murayama A, Kitanaka S, Suzawa M, Yanagisawa J, Kato S 1999 Selective interaction of vitamin D receptor with transcriptional coactivators by a vitamin D analog. Mol Cell Biol 19: 1049–1055. Peleg S, Ismail A, Uskokovic MR, Avnur Z 2003 Evidence for tissue- and cell-type selective activation of the vitamin D receptor by Ro-26-9228, a noncalcemic analog of vitamin D3. J Cell Biochem 88:267–273. Gajdos C, Jordan VC 2002 Selective estrogen receptor modulators as a new therapeutic drug group: concept to reality in a decade. Clin Breast Cancer 2:272–281. McDonnell DP, Connor CE, Wijayaratne A, Chang CY, Norris JD 2002 Definition of the molecular and cellular mechanisms underlying the tissue-selective agonist/antagonist

CHAPTER 83 Molecular Basis for Differential Action of Vitamin D Analogs

98. 99.

100.

101.

102.

103. 104.

105.

106. 107.

108.

109.

110.

111. 112. 113.

activities of selective estrogen receptor modulators. Recent Prog Horm Res 57:295–316. Shang Y, Brown M 2002 Molecular determinants for the tissue specificity of SERMs. Science 295:2465–2468. Liu Z, Auboeuf D, Wong J, Chen JD, Tsai SY, Tsai M, O’Malley BW 2002 Coactivator/corepressor ratios modulate PR-mediated transcription by the selective receptor modulator RU486. Proc Natl Acad Sci USA 99:7940–7944. Okamura WH, Midland MM, Hammond MW, Abd Rahman N, Dormanen MC, Nemere I, Norman AW 1995 Chemistry and conformation of vitamin D molecules. J Steroid Biochem Mol Biol 53:603–613. Miura D, Manabe K, Ozono K, Saito M, Gao Q, Norman AW, Ishizuka S 1999 Antagonistic action of novel 1alpha,25-dihydroxyvitamin D3 -26,23-lactone analogs on differentiation of human leukemia cells (HL-60) induce by 1alpha,25-dihydroxyvitamin D3. J Biol Chem 274: 16392–16399. Bula CM, Bishop JE, Ishizuka S, Norman AW 2000 25-dehydro-1alpha-hydroxyvitamin D3-26,23S-lactone antagonizes the nuclear vitamin D receptor by mediating a unique noncovalent conformational change. Mol Endocrinol 14: 1788–1796. Toell A, Gonzalez MM, Ruf D, Steinmeyer A, Ishizuka S, Carlberg C 2001 Different molecular mechanisms of vitamin D3 receptor antagonists. Mol Pharmacol 59:1478–1485. Ozono K, Saito M, Miura D, Michigami T, Nakajima S, Ishizuka S 1999 Analysis of the molecular mechanism for the antagonistic action of a novel 1alpha,25-dihydroxyvitamin D3 analog toward vitamin D receptor function. J Biol Chem 274:32376–32381. Ishizuka S, Miura D, Ozono K, Chokki M, Mimura H, Norman AW 2001 Antagonistic actions in vivo of (23S)-25dehydro-1alpha-hydroxyvitamin D3-26,23-lactone on calcium metabolism induced by 1alpha,25-dihydroxyvitamin D3. Endocrinology 142:59–67. Bury Y, Steinmeyer A, Carlberg C 2000 Structure activity relationship of carboxylic ester antagonists of the vitamin D3 receptor. Mol Pharmacol 58:1067–1074. Herdick M, Steinmeyer A, Carlberg C 2000 Antagonistic action of a 25-carboxylic ester analogue of 1alpha, 25-dihydroxyvitamin D3 is mediated by a lack of ligand-induced vitamin D receptor interaction with coactivators. J Biol Chem 275:16506–16512. Vaisanen S, Perakyla M, Karkkainen JI, Steinmeyer A, Carlberg C 2002 Critical role of helix 12 of the vitamin D3 receptor for the partial agonism of carboxylic ester antagonists. J Mol Biol 315:229–238. Zerwekh JE, Sakhaee K, Glass K, Pak CY 1983 Long-term 25-hydroxyvitamin D3 therapy in postmenopausal osteoporosis: demonstration of responsive and nonresponsive subgroups. J Clin Endocrinol Metab 56:410–413. Shiraki M, Orimo H, Ito H, Akiguchi I, Nakao J, Takahashi R, Ishizuka S 1985 Long-term treatment of postmenopausal osteoporosis with active vitamin D3, 1-alpha-hydroxycholecalciferol (1 alpha-OHD3) and 1,24 dihydroxycholecalciferol (1,24(OH)2D3). Endocrinol Jpn 32:305–315. Holick MF 1989 1,25-dihydroxyvitamin D3 and the skin: A unique application for the treatment of psoriasis. Proc Soc Exp Biol Med 191:246–257. Ritz E, Mehls O 1995 Vitamin D therapy in patients receiving dialysis. Adv Ren Replace Ther 2:14–19. Diaz-Corte C, Cannata-Andia JB 2000 Management of secondary hyperparathyroidism: the gap between diagnosis

114. 115.

116.

117.

118.

119. 120. 121. 122. 123. 124.

125. 126.

127.

128.

129.

1487

and treatment. The renal osteodystrophy multicenter enquiry. Am J Med Sci 320:107–111. Nordin BE, Need AG, Morris HA, Horowitz M 1999 The special role of “hormonal” forms of vitamin D in the treatment of osteoporosis. Calcif Tissue Int 65:307–310. Peacock M, Liu G, Carey M, McClintock R, Ambrosius W, Hui S, Johnston CC 2000 Effect of calcium or 25OH vitamin D3 dietary supplementation on bone loss at the hip in men and women over the age of 60. J Clin Endocrinol Metab 85: 3011–3019. Perez A, Chen TC, Turner A, Raab R, Bhawan J, Poche P, Holick MF 1996 Efficacy and safety of topical calcitriol (1,25-dihydroxyvitamin D3) for the treatment of psoriasis. Br J Dermatol 134:238–246. Green C, Ganpule M, Harris D, Kavanagh G, Kennedy C, Mallett R, Rustin M, Downes N 1994 Comparative effects of calcipotriol (MC903) solution and placebo (vehicle of MC903) in the treatment of psoriasis of the scalp. Br J Dermatol 130:483–487. Molin L, Cutler TP, Helander I, Nyfors B, Downes N 1997 Comparative efficacy of calcipotriol (MC903) cream and betamethasone 17-valerate cream in the treatment of chronic plaque psoriasis. A randomized, double-blind, parallel group multicenter study. Calcipotriol study group. Br J Dermatol 136:89–93. Mason J, Mason AR, Cork MJ 2002 Topical preparations for the treatment of psoriasis: a systematic review. Br J Dermatol 146:351–364. Llach F, Yudd M 2001 Paricalcitol in dialysis patients with calcitriol-resistant secondary hyperparathyroidism. Am J Kidney Dis 38:S45–S50. Goldenberg MM 1999 Paricalcitol, a new agent for the management of secondary hyperparathyroidism in patients undergoing chronic renal dialysis. Clin Ther 21:432–441. Akizawa T, Suzuki M, Akiba T, Nishizawa Y, Kurokawa K 2001 Clinical effects of maxacalcitol on secondary hyperparathyroidism of uremic patients. Am J Kidney Dis 38:S147–S151. Kubodera N, Tsuji N, Uchiyama Y 2002 Drugs in development for the treatment of osteoporosis: active vitamin D analog (ED-71). Nippon Rinsho 60:517–528. Uchiyama Y, Higuchi Y, Takeda S, Masaki T, Shira-Ishi A, Sato K, Kubodera N, Ikeda K, Ogata E 2002 ED-71, a vitamin D analog, is a more potent inhibitor of bone resorption than alfacalcidol in an estrogen-deficient rat model of osteoporosis. Bone 30:582–588. Kurihara N, Reddy SV, Menaa C, Anderson D, Roodman GD 2000 Osteoclasts expressing the measles virus nucleocapsid gene display a pagetic phenotype. J Clin Invest 105:607–614. Menaa C, Barsony J, Reddy SV, Cornish J, Cundy T, Roodman GD 2000 1,25-dihydroxyvitamin D3 hypersensitivity of osteoclast precursors from patients with Paget’s disease. J Bone Miner Res 15:228–236. Kitazawa S, Kajimoto K, Kondo T, Kitazawa R 2003 Vitamin D3 supports osteoclastogenesis via functional vitamin D response element of human RANKL gene promoter. J Cell Biochem 89:771–777. Oades GM, Dredge K, Kirby RS, Colston KW 2002 Vitamin D receptor-dependent antitumor effects of 1,25-dihydroxyvitamin D3 and two synthetic analogs in three in vivo models of prostate cancer. BJU Int 90:607–616. Bernardi RJ, Johnson CS, Modzelewski RA, Trump DL 2002 Antiproliferative effects of 1α,25-dihydroxyvitamin D3 and vitamin D analogs on tumor-derived endothelial cells. Endocrinology 143:2508–2514.

1488 130. Polek TC, Murthy S, Blutt SE, Boehm MF, Zou A, Weigel NL, Allegretto EA 2001 Novel nonsecosteroidal vitamin D receptor modulator inhibits the growth of LnCap xenograft tumors in athymic mice without increased serum calcium. Prostate 49:224–233. 131. Blutt SE, Polek TC, Stewart LV, Kattan MW, Weigel NL 2000 A calcitriol analog, EB1089, inhibits the growth of LnCap tumors in nude mice. Cancer Res 60:779–782. 132. El Abdaimi K, Dion N, Papavasiliou V, Cardinal PE, Binderup L, Goltzman D, Ste-Marie LG, Kremer R 2000 The vitamin D analog EB 1089 prevents skeletal metastasis and prolongs survival time in nude mice transplanted with human breast cancer cells. Cancer Res 60:4412–4418.

SARA PELEG

133. Gulliford T, English J, Colston KW, Menday P, Moller S, Coombes RC 1998 A phase I study of the vitamin D analog EB 1089 in patients with advanced breast and colorectal cancer. Br J Cancer 78:6–13. 134. Evans TR, Colston KW, Lofts FJ, Cunningham D, Anthoney DA, Gogas H, de Bono JS, Hamberg KJ, Skov T, Mansi JL 2002 A phase II trial of the vitamin D analog secocalcitol (EB1089) in patients with inoperable pancreatic cancer. Br J Cancer 86:680–685. 135. Gardezi SA, Nguyen C, Malloy PJ, Posner GH, Feldman D, Peleg S 2001 A rationale for treatment of hereditary vitamin D– resistant rickets with analogs of 1alpha,25-dihydroxyvitamin D3. J Biol Chem 276:29148–29156.

CHAPTER 84

Development of New Vitamin D Analogs LISE BINDERUP, ERNST BINDERUP, WAGN O. GODTFREDSEN, AND ANNE-MARIE KISSMEYER Medicinal Chemistry and Biological Research, LEO Pharma, Ballerup, Denmark

I. Introduction II. Strategy for Development of New Vitamin D Analogs III. Structure-Activity Relationships

IV. Biological Activities V. Clinical Development of LEO Analogs References

I. INTRODUCTION

reviews our main efforts and achievements in this field since the mid-1980s.

The involvement of LEO Pharma in the synthesis and evaluation of new vitamin D analogs and metabolites dates back to the early 1970s, with the development of 1α-hydroxycholecalciferol [1α(OH)D3] for the treatment of renal osteodystrophy and hyperparathyroidism. Our interest was further stimulated in the early 1980s with the appearance of reports describing receptors for 1α,25-dihydroxyvitamin D3 [1,25(OH)2D3] in nonclassic target tissues and the demonstration of the role of 1,25(OH)2D3 in regulating growth and differentiation of various cancer cell lines, in vitro and in vivo [1,2]. At the same time, a number of reports suggested that 1,25(OH)2D3 might also influence various functions of activated lymphocytes and thereby play a role as a physiological regulator of the immune system [3,4]. These findings suggested new therapeutic possibilities for 1α(OH)D3 and 1,25(OH)2D3, especially in neoplastic and immune-mediated diseases. In 1983, LEO took steps to initiate clinical trials with 1α(OH)D3 in leukemia and non-Hodgkin’s lymphomas [5,6]. The therapeutic usefulness of 1α(OH)D3 and 1,25(OH)2D3 was, however, likely to be limited by their potent effects on calcium metabolism, leading to side effects such as hypercalcemia and soft tissue calcifications. It was therefore decided to try to develop new analogs with a more favorable therapeutic profile. In 1985, the preliminary testing of a small series of new synthetic analogs led to the discovery of a promising candidate, MC903, later named calcipotriol. At the same time, clinical observations suggested that 1α(OH)D3 and 1,25(OH)2D3 might exert antipsoriatic effects [7,8]. It was therefore decided to test calcipotriol in patients with psoriasis and to further expand our engagement in vitamin D chemistry. This chapter VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

II. STRATEGY FOR DEVELOPMENT OF NEW VITAMIN D ANALOGS A. Basic Screening Strategy At the start of the program for synthesis and evaluation of new vitamin D analogs, the primary aim was to identify compounds that were potent regulators of cell proliferation and differentiation, but which had a reduced ability to exert the classic effects of 1,25(OH)2D3 on calcium homeostasis. Structure-activity relationships of a relatively large number of analogs (more than 1000) have been studied in detail over the last 15 years. Initially, the effects of all newly synthesized vitamin D analogs were tested in the human histiocytic lymphoma cell line U 937, which expresses the vitamin D receptor (VDR) and responds to 1,25(OH)2D3 with a decrease in proliferation and induction of differentiation along the monocyte-macrophage pathway [9,10]. The analogs were tested in the cell culture system for 4 days, at concentrations ranging from 10−12 to 10−7M, and their effects were compared to those of 1,25(OH)2D3. More recently, the U 937 cell cultures were supplemented with other cell types, notably mammary cancer cells (MCF-7, human breast carcinoma cell line) and skin cells (HaCaT, human keratinocyte cell line). The binding of the vitamin D analogs to the VDR was assessed by displacement of bound 3H-1,25(OH)2D3 from receptor protein obtained from the intestinal epithelium of rachitic chickens [11]. As the binding affinity of many analogs did not show a direct correlation Copyright © 2005, Elsevier, Inc. All rights reserved.

1490 with their biological activities (see also Section IV.A), the VDR binding studies were extended to include studies of the transcriptional activity of the analogs, using the vitamin D responsive element in the promoter region of the osteocalcin gene. To assess the effects of new vitamin D analogs on calcium metabolism, an in vivo model was chosen [10]. The analogs were administered orally to rats, daily for 7 days. Urine was collected daily, and blood was collected by cardiac puncture at the end of the experiment. Metaphyseal bone was prepared from tibiae. Calcium levels were determined in urine and serum samples, and the calcium content in bone was assessed after ashing. To assure detection of even small differences in potency between various analogs, the rats were given a vitamin D–replete diet with a high calcium content (1%), in contrast to many of the older studies in rats and chickens given low calcium and/or low vitamin D diets [12,13]. As a follow-up to the screening system described above, the metabolic stability (serum half-life) of selected analogs was initially tested in vivo after intravenous administration to rats [14]. At a later stage, an in vitro test using the rat liver post-mitochondrial fraction (S9) was introduced. The analogs were tested at a single concentration, and the percentage of intact analog after 1 hour of incubation was assessed. The analogs were classified as unstable (<10% intact analog), medium stable (10–35% intact analog), and stable (>35% intact analog). Analogs with a low metabolic stability were considered as candidates for topical use, and analogs with a high metabolic stability as candidates for systemic use.

B. Synthesis Strategy From the beginning of the project it was decided to concentrate our efforts on the synthesis of analogs of 1,25(OH)2D3 (I) in which the C-17 side chain was modified while the seco-steroid ring system was kept intact. This decision was partly dictated by the fact that this part of the molecule is more easily accessible to chemical manipulation than the ring system, but it was also decisive that the side chain is known to play a crucial role in the binding of 1,25(OH)2D3 to its receptor [15]. It is beyond the scope of this chapter to detail the synthesis of the more than 1000 analogs that have been made in the LEO laboratories, but some general pathways are outlined in the following sections. Our starting material has been the readily available ergocalciferol (II) (Scheme 1), which, according to a method originally devised by Hesse [16,17], can be converted to a 1α-hydroxylated trans vitamin derivative

LISE BINDERUP

20 17

ET AL .

25 OH

H H

3

1

HO

OH (I)

(IIa), conveniently isolated as the pure, crystalline, bis-silyl ether (III) [18]. After protection of the conjugated triene system as the sulfur dioxide adduct (IV), the side chain can be cleaved selectively by ozonolysis with formation of the aldehyde (V). By heating (V) in the presence of NaHCO3, SO2 is expelled, and the key intermediate (VI) can be isolated as a crystalline compound [18]. Compound (VI) has been a cornerstone in our synthetic work. By means of the Wittig reaction, new side chains could be introduced. Scheme 2 illustrates the synthesis of the anticancer drug EB 1089 (IX) (seocalcitol, see Table II), which contains two conjugated double bonds in the side chain [19]. The product of the Wittig reaction (VII) is reacted with ethyl lithium to give (VIII), which is then isomerized to the cis form by ultraviolet irradiation in the presence of the photosensitizer anthracene [20]. Finally, the hydroxyl groups are deprotected with tetrabutylammonium fluoride in tetrahydrofuran to give (IX). A Wittig reaction with (VI) is also a step in the synthesis of the antipsoriatic drug MC 903 (calcipotriol, see Table III) [18]. Another route to new analogs involves NaBH4 reduction of the aldehyde (VI), followed by tosylation of the resulting alcohol to give (X), which is subsequently reacted with a Grignard reagent to form (XI). Finally, (XI) is isomerized and deprotected to provide the analog CB 966 (XII) (see Table II) [21]. Analogs containing an oxygen atom in the 22 position (22-oxa analogs) can be synthesized as shown in Scheme 3, which depicts the synthesis of the C-20 epimeric compounds KH 1139 and KH 1060 (see Table IV) [22]. Oxidation of the aldehyde (VI) with air in the presence of a copper catalyst yields the methylketone (XIII), which on reduction with NaBH4 forms a mixture of the two epimeric alcohols (XIVa and XIVb), where the 20-epi isomer (XIVb) dominates. Alkylation of the two alcohols, followed by isomerization

1491

CHAPTER 84 Development of New Vitamin D Analogs

H

H

H

H

HO

HO (II)

O

SO2

H

H

Si

Si

O

(IIa)

O

Si

(III)

CHO H O

O H

H

O

H

H O3

S Si O

CHO

O

O H S

Si O

Si

H

NaHCO3 O

(IV)

∆

H

Si

Si

O

(V)

O

Si

(VI)

SCHEME 1

and deprotection as described above, yields the cis analogs (XVa) (KH 1139) and (XVb) (KH 1060), respectively. As described in Section III.D, epimerization at C-20 has a profound influence on the biological properties of the analogs. Another route to 20-epi compounds (Scheme 3) starts with an epimerization of the aldehyde (VI) to the 20-epi aldehyde (XVI), which then, by the same

COOCH3

CHO (C6H5)3P

N

sequence of reactions used in the 20-normal series (Scheme 2), can be converted to 20-epi analogs, such as 20-epi-1,25(OH)2D3 (XVII) (MC 1288) (see Table IV) [23]. The tosylate (XVIII), used in the synthesis of MC 1288, has also been used to synthesize the 20-epi23-thia analog (XIX) (GS 1790) (see Table V) [24]. The reactions depicted in Schemes 1–3 are typical of the pathways used in the synthesis of a wide variety

COOCH3 N

(VI)

N=

H

(VII) H

1. NaHBH4 2.TsCl - Pyridine

2 EtLi Si

O

O

Si

OH OTs N (X)

N (VIII) 1. Anthracene/UV-light 2. TBA+ F− OR

H M=

H

OH A (XI) R = Si(Ch3)3 ; A = N (XII) CB 966 (R = H ; A = M)

M (IX) EB 1089

SCHEME 2

HO

OH

1492

LISE BINDERUP

CHO

ET AL .

CHO

OH−

N

N

(VI)

(XVI)

OH M (XVII) MC 1288

2+

O2, Cu , DABCO 2,2′-Bipyridyl OH O

OTs

N

N

(XIII)

(XVIII)

S M (XIX) GS 1790

NaBH4 OH N (XIVa)

+

N (XIVb)

OH

OH

O

OH

O

+

M

M

(XVa) KH 1139

(XVb) KH 1060

SCHEME 3 N and M have the same meaning as in Scheme 2.

of side chain analogs, but it is obvious that many variations have been necessary.

III. STRUCTURE-ACTIVITY RELATIONSHIPS In this section, the effect of systematic chemical modifications of the 1,25(OH)2D3 side chain on various biological parameters is presented. All the analogs discussed here have been tested for calcemic activity, antiproliferative activity, and ability to induce cell differentiation, as described in Section II.A. However, because the antiproliferative and differentiation-inducing properties run parallel, only the antiproliferative potencies are shown in the tables. All values are given in relation to 1,25(OH)2D3.

A. Variation of Chain Length In Table I the biological activities of a number of analogs which differ from 1,25(OH)2D3 with respect to the length of the C-17 side chain are listed [21]. It is seen that if the chain length is increased with one methylene group as in MC 1127, the ability to inhibit proliferation is increased, whereas the calcemic activity is reduced to about one-third of that of 1,25(OH)2D3. This compound has also been described by Ostrem et al. [25]. If the two terminal methyl groups in MC 1127 are replaced by ethyl groups as in CB 966, the antiproliferative potency is increased

and the calcemic activity further reduced. On the other hand, if propyl groups are substituted for the methyl groups, the antiproliferative potency is reduced to about the same level as that of 1,25(OH)2D3. An increase of the 1,25(OH)2D3 side chain with two methylene groups (MC 1147), also described by Kutner et al. [26], causes a further reduction of the calcemic activity, whereas the antiproliferative potency remains the same as in MC 1127. However, with the introduction of one more methylene group (MC 1179), the antiproliferative activity is reduced. In other words, the optimal number of methylene groups between C-20 and the tertiary hydroxyl group seems to be four or five.

B. Introduction of Double and Triple Bonds The effect of introducing one or two double bonds in the C-17 side chain is illustrated in Table II. Whereas the introduction of a ∆22 double or triple bond in CB 966 decreases the antiproliferative activity [23,27], the introduction of a further ∆24 double bond leads to EB 1089 [19], which, with respect to cell proliferation, is the most active in the series, being 100 times more potent than 1,25(OH)2D3. Because the calcemic activity of EB 1089 is three times lower, a substantial separation of the effects has been achieved. EB 1089 has been chosen as candidate for clinical testing in cancer patients (see Section V.B.2). Another analog, CB 1093, is characterized by several modifications in the side chain. In addition to the triple bond, CB 1093 has altered stereochemistry at

1493

CHAPTER 84 Development of New Vitamin D Analogs

TABLE I Variation of Chain Length Compound 1,25(OH)2D3

Inhibition of U937 cell proliferation IC50 (M)

Side chain structure OH

Calcemic activity relative to 1,25(OH)2D3 (%)

3 × 10−8 [1X]a

100

5 × 10−9 [6X]

38

1 × 10−9 [30X]

17

5 × 10−8 [0.6X]

n.d.b

OH

MC 1127

OH

CB 966

OH

CB 973

MC 1147

OH

5 × 10−9 [6X]

4

CB 953

OH

2 × 10−8 [1X]

2

4 × 10−8 [0.7X]

2

OH

MC 1179 aBoldface bNot

figures indicate activity relative to 1,25(OH)2D3. determined.

C-20 and an ethoxy group at C-22. These modifications increase the antiproliferative activity of the compound to the same level as that of EB 1089. CB 1093 has recently been shown to be a potent inducer of apoptosis (see Section IV.C). The last compound in Table II, HEP 187, is a 20-epi-vitamin D analog, in which the terminal side chain hydroxy group has been replaced by a fluorine atom. This compound has a relatively low antiproliferative effect, compared to the other 20-epi-analogs, but it has been shown to exert interesting effects on bone mineral metabolism (see Section IV.E.3).

C. Calcipotriol and Related Analogs One of the first 1,25(OH)2D3 analogs synthesized at LEO for which a clear separation between the calcemic

activity and the effects on cell regulation was achieved was MC 903 [10,18] (Table III), which later received the United States Adopted Name (USAN) calcipotriene and the International Nonproprietary Name (INN) calcipotriol. In this compound, a ∆22 double bond is introduced in the side chain, the 25-hydroxyl is moved to the 24 position (with the indicated stereochemistry), and a cyclopropane ring is substituted for the isopropyl group in 1,25(OH)2D3. Table III shows that calcipotriol has retained the cell-regulating potency of 1,25(OH)2D3, whereas its calcemic activity has been reduced by a factor of 200. As described in Sections IV.E.2 and V.B.1 of this chapter and in Chapter 101, calcipotriol has become an important antipsoriatic drug. Its 24-epimer MC 900 [18] has a considerably lower antiproliferative potency, and the same holds true for MC 1046 and MC 1080, the two main metabolites of calcipotriol [28].

1494

LISE BINDERUP

TABLE II Compound

Double and Triple Bonds Inhibition of U937 cell proliferation IC50 (M)

Side chain structure

1,25(OH)2D3

ET AL .

OH

Calcemic activity relative to 1,25(OH)2D3 (%)

3 × 10−8 [1X]a

100

1 × 10−9 [30X]

17

2 × 10−8 [1.5X]

24

3 × 10−8 [1X]

n.d.b

3 × 10−10 [100X]

31

3 × 10−10 [100X]

24

5 × 10−9 [6X]

34

OH

CB 966

OH

MC 1473 OH

CB 1309

OH

EB 1089 (seocalcitol) O OH

CB 1093 F

HEP 187 aBoldface bNot

figures indicate activity relative to 1,25(OH)2D3. determined.

The effect of the size of the terminal ring was studied by Calverley [29], who synthesized three pairs of 24-epimeric analogs in which the cyclopropyl group in calcipotriol is replaced by cyclobutyl, cyclopentyl, and cyclohexyl groups, respectively. Although the stereochemistry of the 24-hydroxyl group in these compounds has not been rigorously established, their polarities suggest that in MC 1070, MC 1052, and MC 1048 the 24-hydroxyl group has the same stereochemistry as in calcipotriol, whereas in MC 1069, MC 1050, and MC 1033 the stereochemistry of the 24-hydroxyl group is as in MC 900. It was found that the antiproliferative potencies of the cyclobutyl and cyclopentyl analogs MC 1070 and MC 1050 were similar to that of calcipotriol, whereas the cyclohexyl analog MC 1048 was less potent. All these compounds were, however, significantly more calcemic than calcipotriol.

D. Epimerization at C-20 A seemingly minor modification of the side chain in 1,25(OH)2D3 that has a dramatic effect on its biological activities is epimerization at C-20. As Table IV shows, the 20-epimer of 1,25(OH)2D3 (MC 1288) [23] is about 100 times more potent than the natural hormone as an inhibitor of cell proliferation, whereas its calcemic activity has increased by a factor of only 2. Even more pronounced is the effect of 20-epimerization on immunosuppressive properties [30]. In view of these results, we found it mandatory to investigate the effect of 20-epimerization more broadly [22,23]. Table IV shows the activities of pairs of 20epimers. Both the antiproliferative potency and the calcemic activity are generally higher in the 20-epi than in the 20-normal series. A particularly noteworthy compound is the 22-oxa analog KH 1060 (lexacalcitol),

1495

CHAPTER 84 Development of New Vitamin D Analogs

TABLE III Compound

Calcipotriol and Analogs Inhibition of U937 cell proliferation IC50 (M)

Side chain structure

1,25(OH)2D3

Calcemic activity relative to 1,25(OH)2D3 (%)

3 × 10−8 [1X]a

100

2 × 10−8 [1.5X]

0.5

>1 × 10−7 [<0.3X]

<1

>1 × 10−7 [<0.3X]

<1

>1 × 10−7 [<0.3X]

<1

OH OH

MC 903 (calcipotriol) OH

MC 900 O

MC 1046 O

MC 1080 aBoldface

figures indicate activity relative to 1,25(OH)2D3.

which is 3000 times more potent than 1,25(OH)2D3 as a proliferation inhibitor but only slightly more calcemic.

E. Introduction of an Aromatic Ring The effect of introducing a benzene ring in the side chain is illustrated in Table V. The analogs are all 20-epi compounds, where the benzene ring is attached to an oxygen or a sulfur atom in the 23 position and further substituted in the ortho, meta, or para position with a 2-hydroxypropyl or a 3-hydroxypentyl group [24]. Whereas the ortho- and para-substituted analogs have antiproliferative potencies of the same order of magnitude as 1,25(OH)2D3, the meta-substituted analogs (EB 1213, EB 1219, GS 1500, and GS 1730) are considerably more potent. All the compounds tested are at least one order of magnitude less calcemic than 1,25(OH)2D3. From this series of analogs, GS 1790 has recently been selected as an interesting candidate for preclinical studies in experimental models of bone disease (see Section IV.E.3).

IV. BIOLOGICAL ACTIVITIES During the last 10 years, research on the effects of 1,25(OH)2D3 at the cellular and molecular level has

intensified, partly due to the availability of new synthetic analogs. The new findings have led to an increased understanding of the therapeutic potential of the analogs, of which a substantial number has been investigated in vivo, both in experimental animal models and in clinical trials. This section describes the main biological findings with a number of LEO analogs in the fields of receptor activated gene expression, cell growth and regulation, immune regulation, and bone mineral metabolism.

A. VDR Binding and Gene Expression Most of the biological effects of 1,25(OH)2D3 in vitro and in vivo are believed to be mediated via the VDR, but the activity of many vitamin D analogs is not directly correlated with their binding affinity for the receptor. The prevailing theory has been that different analogs may confer different conformational changes to the VDR ligand complex. This may in turn lead to altered stability of the complex and/or an altered ability to interact with DNA-binding sites or protein coactivators and corepressors. The use of vitamin D analogs with altered stereochemistry at C-20, the 20-epi-vitamin D analogs, has allowed the investigation of the relative importance of

1496

LISE BINDERUP

TABLE IV Compound

Effect of 20-Epimerization Inhibition of U937 cell proliferation IC50 (M)

Side chain structure

1,25(OH)2D3

OH

MC 1288

ET AL .

OH

Calcemic activity relative to 1,25(OH)2D3 (%)

3 × 10−8 [1X]a

100

3 × 10−10 [100X]

200

OH

5 × 10−9 [6X]

MC 1127

38

OH

6 × 10−11 [500X]

EB 1231

500

OH

1 × 10−9 [30X]

CB 966

17

OH

MC 1301

3 × 10−10 [100X]

200

2 × 10−10 [150X]

30

1 × 10−11 [3000X]

130

OH O

KH 1139 OH O

KH 1060 (lexacalcitol) aBoldface

figures indicate activity relative to 1,25(OH)2D3.

the various factors that may increase the transcriptional activity of the VDR. Studies on the direct interaction between the VDR and a tritium-labeled 20-epi-analog ([1-3H]GS 1500) showed that GS 1500 (see Table V) bound to the same binding site as 1,25(OH)2D3, but was able to induce the VDR to form a DNA-binding complex at a concentration 100 times lower than that of 1,25(OH)2D3 [31]. Further studies with the biologically very potent 20-epi-analogs MC 1288 and KH 1060 have suggested that these compounds increase the stability of the analogVDR complex [32–34] by making it more resistant to protease degradation than that of the natural ligand 1,25(OH)2D3 [35,36]. This increased stability has been

linked to the induction of conformational changes in the VDR that prevent binding of the proteins involved in the ubiquitin/proteasome mediated degradation of the receptors [37,38]. The recent availability of the crystal structures of the ligand-binding domain of the VDR, complexed to 1,25(OH)2D3 or to the 20-epi-analogs MC 1288 and KH 1060, has permitted a more detailed investigation of the way in which the VDR interacts with agonistic compounds. The results indicate that both 1,25(OH)2D3 and the analogs adapt to the binding pocket, with new points of contact to the protein, but without inducing major conformational changes [39]. Conformational changes may thus be of less importance than previously

1497

CHAPTER 84 Development of New Vitamin D Analogs

TABLE V Analogs with an Aromatic Ring Compound

Side chain structure

1,25(OH)2D3

OH HO

EB 1224

Inhibition of U937 cell proliferation IC50 (M)

Calcemic activity relative to 1,25(OH)2D3 (%)

3 × 10−8 [1X]a

100

2 × 10−8 [1.5X]

3

1 × 10−10 [300X]

7

8 × 10−8 [0.3X]

n.d.b

8 × 10−11 [375X]

6

6 × 10−8 [0.5X]

<1

1 × 10−10 [300X]

<1

5 × 10−8 [0.6X]

<1

1 × 10−9 [30X]

<1

O

O

EB 1213

OH OH

EB 1220

O

O

EB 1219

OH

HO

GS 1780 S

S

OH

GS 1500 OH

GS 1790

S

S

GS 1730

OH

aBoldface bNot

figures indicate activity relative to 1,25(OH)2D3. determined.

thought, and the increased transcriptional activity of the analogs may be primarily related to the increased stability and half-life of the ligand-receptor complex [39,40]. Another consequence of the increased stabilization induced by vitamin D analogs may be enhanced dimerization with RXR or increased recruitment of coactivators that facilitate interaction between nuclear receptors and the basal transcription machinery. In this respect, a close correlation between the ability of

MC 1288 to induce cell differentiation, p21 transactivation, and recruitment of the coactivator protein complex DRIP to the VDR has been observed [41]. Very recent studies have shown that the strength of interaction of coactivators such as GRIP1 and RAC3 with the VDR is enhanced by the 20-epi-analogs MC 1288, KH 1060, and MC 1301, but not by analogs with 20-normal configuration [42]. The interaction of VDR with coactivators may lead to opening of the chromatin by activation of histone acetylases

1498 and permit transcription of vitamin D responsive genes [43]. The most complete analysis of vitamin D–induced gene expression in a single tissue has recently been performed with the 20-normal analog EB 1089. Human head and neck squamous carcinoma cells were incubated with or without the analog for up to 48 hrs and gene expression was studied using DNA microarray screening. More than 150 genes were regulated, including genes involved in cell cycle progression, cell differentiation, cell adhesion, extracellular matrix composition, and cellular functions of the immune system [44].

B. Cell Cycle Regulation The ability of vitamin D analogs to inhibit proliferation and induce differentiation in a large number of different cell types has led to the development of several of these analogs for the treatment of hyperproliferative diseases such as psoriasis and cancer. One mechanism by which vitamin D compounds may exert their effects on cellular growth is by regulating cell cycle progression. Most cells respond to 1,25(OH)2D3 or its analogs by growth arrest, associated with block in the G0/G1 phase [45]. Early studies indicated that growth inhibition of human keratinocytes was linked to dephosphorylation of pRb [46]. Since then, the study of the molecular basis for induction of cell cycle blockade by vitamin D compounds has been intensified. Up-regulation of the two cyclin-dependent kinase inhibitors p21WAF1/CIP1 and p27KIP1 appears to be one of the main mechanisms of the G1-block. The vitamin D analog EB 1089, which is 50–100 times more potent than 1,25(OH)2D3 in inhibiting proliferation of a large number of cancer cell lines, has been used extensively in the study of cell cycle regulation, showing differential regulation of cyclin/cdk activity in various cancer cells. In HL-60 promyelocytic leukemia cells, EB 1089 induced p27WAF1/CIP1 and decreased cdk2 and cdk6 activities [47], whereas in NCI-H929 myeloma cells, EB 1089 induced p27KIP1 and reduced the cdk2, but not the cdk6 activity [48]. These findings were further refined by a study in thyroid carcinoma cells, where EB 1089 similarly induced expression of p27KIP1 and also protected the cdk inhibitor from degradation, possibly by activation of phosphatases like PTEN [49]. A recent study has linked the effects of EB 1089 on cell proliferation directly to those on cell differentiation. Using SW480-ADH colon cancer cells, EB 1089 was shown to induce expression of E-cadherin and sequestration of β-catenin in the plasma membrane, leading to blockade of the β-catenin signaling pathway,

LISE BINDERUP

ET AL .

down-regulation of genes involved in cell cycle regulation (such as cyclin D1 and c-myc), and induction of differentiation [50].

C. Apoptosis Induction of growth arrest by vitamin D compounds is also linked to induction of apoptosis, especially in cancer cells. Apoptotic features have been observed in many types of cancer cells including breast, colon, and prostate cancer cells, after treatment in vitro with 1,25(OH)2D3 or analogs. EB 1089 has been shown to be a more potent inducer of apoptosis than 1,25(OH)2D3 in a study with 5 different colon cancer cell lines [51]. The apoptotic effects of EB 1089 could be increased by combination with TGFβ [52] or antiestrogens [53]. The mechanism(s) by which 1,25(OH)2D3 and its analogs induce apoptosis may vary from one cell type to another, but in most studies induction of apoptosis has been found to be independent of the p53 tumor suppressor status and did not involve caspase activation [54]. Several studies have shown that both 1,25(OH)2D3 and EB 1089 increase intracellular calcium levels, disrupt mitochondrial functions, and induce cytochrome c release, thus activating caspaseindependent cell death [55,56]. This pathway may also include activation of the calcium-dependent cysteine protease, calpain [56]. In addition, EB 1089 has also been shown to down-regulate the antiapoptotic protein bcl-2 [57]. Recently, CB 1093, a vitamin D analog with 20-epi configuration and a triple bond in the side chain (see Table II), has also shown potent apoptosisinducing effects [58,59]. A number of cell lines from tissues such as skin and bone are resistant to induction of apoptosis by vitamin D compounds. In particular, vitamin D analogs such as MC 903 and CB 1093 have been shown to display antiapoptotic effects in osteoblast-like cells [60]. These findings suggest that the ability of vitamin D compounds to stimulate bone formation may, to some extent, be linked to their ability to protect the osteoblasts from undergoing apoptosis.

D. Growth Factors 1,25(OH)2D3 and its analogs interfere with the mitogenic activity of growth factors, such as IGF-I, IGF-II, and EGF [61,62]. These effects are mainly seen in cancer cells with an increased growth factor expression or activity, which contribute to their survival. The analog EB 1089 was able to inhibit growth of IGF-I stimulated human breast cancer cells [63], and both

CHAPTER 84 Development of New Vitamin D Analogs

EB 1089 and CB 1093 were shown to inhibit the secretion of IGF-II in human colon cancer cells [64]. Blocking of the growth-stimulating activity of IGF-I has been associated with suppression of IGF-receptor 1 expression [61] and with up-regulation of IGF-binding proteins [65,66]. The IGFBPs control the availability of the growth factors, and both EB 1089 and CB 1093 up-regulate IGFBP-6 in colon cancer cells [64] and IGFBP-3 in breast cancer cells [67]. Another study has shown up-regulation of IGFBP-5 in breast cancer cells by EB 1089 and KH 1060 [68]. Vitamin D compounds also inhibit cell proliferation by interfering with signaling through the EGF receptor. A recent study shows that MC 903 is able to inhibit the autocrine phosphorylation of the EGF-receptor in squamous carcinoma cells, resulting in a potent inhibition of cell growth and induction of differentiation [69]. In contrast, treatment of hyperproliferative epithelial cells with 1,25(OH)2D3 often results in an increased secretion of TGFβ or an enhanced expression of TGFβ receptors. In these cells, TGFβ acts as a negative growth regulator and up-regulation of TGFβ activity results in growth inhibition. The analogs EB 1089 and MC 903 enhanced TGFβ1 expression and protein secretion in breast cancer cells [70], whereas other studies showed that EB 1089 was able to up-regulate TGFβ receptor II in breast cancer cells [71] and TGFβ receptors I and II in HL-60 myeloid leukemia cells [72]. The role of vitamin D analogs in regulating growth factor expression and proliferation of epithelial cells is an important new area of research, with implications for the treatment of cancer and other hyperproliferative diseases.

E. Preclinical Experience with Vitamin D Analogs 1. CANCER MODELS

The rationale for the use of vitamin D analogs in cancer is based on the findings described previously in this chapter and more fully discussed in Chapters 89–97 of this book. The search for new analogs with potential clinical usefulness has been directed at finding a candidate with a good systemic bioavailability, potent effects on cell proliferation, an ability to activate apoptotic pathways, and, very importantly, a reduced calcemic activity compared to 1,25(OH)2D3. From our screening program, we selected the analog EB 1089 (seocalcitol), which fulfilled these criteria. EB 1089 is characterized by having two double bonds in the side chain, and methyl groups at C26 and C27, which makes it less susceptible to metabolic degradation (see Table II).

1499 The anticancer effects of EB 1089 have been studied in numerous experimental animal studies. Colston et al. were the first to show that EB 1089 inhibited tumor growth in vivo [73]. Rats with nitrosomethylurea (NMU)-induced mammary tumors were treated with EB 1089 at 0.5 µg/kg/day p.o. for 4 weeks. The treatment induced a significant inhibition of tumor growth, without changes in the serum calcium levels. In contrast, treatment with the same dose of 1,25(OH)2D3 did not affect tumor growth but caused hypercalcemia. Interestingly, histological analysis of NMU-induced tumors from rats treated with EB 1089 has shown evidence of apoptotic cell death, with large areas exhibiting loss of cellularity, a low mitotic index, and nuclear DNA fragmentation [74]. Similar results have been obtained in nude mice with breast cancer xenografts treated with EB 1089 [75]. Further support for the potential role of EB 1089 in the treatment of breast cancer came from a study showing that EB 1089 was able to inhibit the development of osteolytic bone metastases from mammary tumors in nude mice [76]. Other studies have shown that EB 1089 can be effectively combined with taxol, tamoxifen, or anti-estrogens [77–79]. In addition, recent findings suggest that EB 1089 may also be active against estrogen-resistant tumors [53]. Another area of interest has been the potential usefulness of EB 1089 in the treatment of prostate cancer. 1,25(OH)2D3 is able to inhibit the growth of human prostate cancer cell lines in vitro, and pilot studies in patients undergoing surgery for prostate cancer have shown that 1,25(OH)2D3 may decrease the rate of rise of the prostate specific antigen, indicating an effect of the treatment on the rate of recurrence of the disease [80]. The analog EB 1089 is a potent inhibitor of prostate cancer cell proliferation in vitro and induces differentiation or apoptosis in several of these cell lines [81–83]. In vivo, EB 1089 inhibited the growth of androgen-resistant metastatic prostate cancer cells in rats and reduced the number of lung metastases with significantly less calcemic toxicity than 1,25(OH)2D3 [84]. Another study has shown that EB 1089 also inhibits prostate cancer xenografts in nude mice, resulting in tumors significantly less vascularized than in control animals [85]. However, no effects were seen with EB 1089 in a study designed to prevent prostate cancer in a model of transgenic mice which develop androgen resistant prostate cancer [86]. Despite the encouraging preclinical evidence obtained with EB 1089 in the animal models described above, broad clinical testing with this analog has not been initiated in patients with breast and prostate cancers. As will be described in the next section, studies on the pharmacokinetic profile and tissue distribution of EB 1089 have indicated that administration of this analog leads

1500 to very high concentrations of active compound in specific tissues, especially in the liver. The recent preclinical and clinical development of EB 1089 has therefore been targeted toward indications such as hepatocellular carcinomas and colon cancer with risk of liver metastasis. In this regard, it is relevant to note that EB 1089 has been shown to inhibit human colon cancer cell xenografts in nude mice [87] and to potently reduce the incidence of spontaneous hepatocellular carcinomas in mice [88]. 2. IMMUNOLOGICAL DISEASES

Besides its classic actions on calcium metabolism and its effects on cell proliferation and differentiation, 1,25(OH)2D3 also modulates a number of immunological functions (see Chapters 98 and 99). The VDR is constitutively expressed in most cell types of the immune system, especially in antigen presenting cells. In T-lymphocytes, up-regulation of the VDR occurs in response to 1,25(OH)2D3 [89]. 1,25(OH)2D3 inhibits antigen-induced T-lymphocyte proliferation and cytokine production, mainly by targeting the subset of T-helper 1 (TH1) cells that preferentially produce IL-2 and interferon-γ (IFN-γ) [90–92]. The rationale for development of vitamin D analogs for psoriasis has been based on the ability of analogs such as calcipotriol to inhibit hyperproliferative skin cells and normalize their aberrant pattern of differentiation, and also on their immune regulatory effects on TH1-type lymphocytes that are implicated in the pathogenesis of this autoimmune disease [93,94]. The beneficial effect of vitamin D analogs on the regulation of the immune system may be enhanced by combination with immunosuppressive agents like cyclosporine A or steroids. A recent study has shown additive effects of combinations of 1,25(OH)2D3 and steroids on the suppression of T-lymphocyte proliferation and INF-γ production, whereas TH2-type responses were not inhibited [95]. Very recently, another study has provided evidence for a synergistic effect of combined steroid and vitamin D treatment. Using 1,25(OH)2D3 and dexamethasone, Barrat et al. [96] have shown that combined treatment induced the appearance of regulatory T-cells producing IL-10, but not TH1-type cytokines. These regulatory T-cells were able to suppress inflammation in an animal model of autoimmunity. Such findings provide the basis for the development of a new type of topical therapy for psoriasis. In this regard, calcipotriol, in combination with betamethasone dipropionate, has recently been shown to have a more effective antipsoriatic activity than either of the agents alone (see Section V.B.1). Besides the use of topical vitamin D analogs in skin diseases, the discovery of the very potent analogs with

LISE BINDERUP

ET AL .

20-epi side chain configuration has led to a marked interest in their potential use for systemic treatment of graft rejection and autoimmune diseases [30]. Extensive studies have been performed in animal models, in particular with the two 20-epi analogs KH 1060 and MC 1288. MC 1288, 20-epi-1,25(OH)2D3, was selected as our candidate for studying the potential of this class of agents in models of graft rejection. MC 1288, at 0.1 µg/kg/day i.p., was shown to prolong the survival of cardiac allografts in rats, with effects comparable to those obtained with cyclosporine A [97,98]. In the same model, combined treatment with MC 1288 and cyclosporine A produced a significantly prolonged survival time of the cardiac grafts, compared with therapy with either agent alone [99]. In contrast, treatment with MC 1288 alone was not able to prevent rejection of mouse-to-rat cardiac xenotransplants, unless combined with low-dose immunosuppressive therapy [100]. MC 1288 has also been studied for its effects on chronic rejection (allograft arteriosclerosis) using aortic allografting in rats [101]. Chronic rejection is the single most important reason for late graft loss. Changes include adventitial inflammation and intimal thickening. Both parameters were suppressed by longterm treatment with MC 1288, in combination with low-dose cyclosporine A. In order to evaluate the potential of MC 1288 as treatment for human transplantation patients, a preclinical study of kidney transplantation in rhesus monkeys was undertaken by LEO (LEO internal report). MC 1288 was administered at the highest possible dose that did not affect serum calcium levels. No significant changes in graft survival time were observed in monkeys treated with MC 1288 as compared to untreated animals. No differences in the pattern of T-lymphocyte subsets or IL-2 serum levels were observed. It was concluded, from the negative results of this study, that monotherapy of graft rejection with vitamin D analogs was not indicated, and that further efforts should be directed toward combination therapy with classic immunosuppressive agents. The 20-epi analogs have also been investigated for immune intervention in other disease areas. KH 1060 was shown to prevent autoimmune type I diabetes in NOD mice [102] and to delay recurrence of diabetes in NOD mice with syngeneic islet grafts [103]. Subsequent use of the analog MC 1288 in the more advanced stage of disease in NOD mice showed that this analog was not able to reduce the incidence of overt diabetes when treatment was delayed until after the onset of insulitis [104]. When treatment with MC 1288 was combined with a short induction course of cyclosporine A, more than 50% reduction in diabetes incidence was seen [104]. Other studies with the 20-epi-analogs have shown positive effects in experimental autoimmune encephalitis

CHAPTER 84 Development of New Vitamin D Analogs

in mice [105], suppression of arthritis in rats with collagen-induced arthritis [106], and inhibitory effects on T-lymphocyte proliferation in cells obtained from patients with ulcerative colitis [107]. Taken together, these studies show that vitamin D analogs regulate autoimmune responses of the TH1type in different disease settings, and also that systemic administration of clinically effective dosages is limited, due to increases in calcium levels. Combination with classic immunosuppressive agents may enhance efficacy and reduce the risk of side effects of both types of administered drugs. 3. BONE DISEASES

The classic physiological role of 1,25(OH)2D3 is to regulate calcium homeostasis and promote bone mineralization. The role of 1,25(OH)2D3 in the treatment of bone diseases such as rickets and renal osteodystrophy is well established, and several reports indicate that 1,25(OH)2D3 and 1α(OH)D3 may also be beneficial for patients with osteoporosis [108,109]. Surprisingly, efforts to develop new synthetic vitamin D analogs for the treatment of bone diseases have been modest over the past years, but with the event of new analogs (maxacalcitol, paricalcitol) for the treatment of hyperparathyroidism and renal osteodystrophy, interest in the treatment of metabolic bone diseases with synthetic analogs seems to have returned. This interest has mainly been focused on two pathways: the design of analogs with modified pharmacokinetic and metabolic profiles and the design of tissue-specific analogs, along the lines that are currently used in the development of estrogen receptor agonists and antagonists. The bone forming cells, the osteoblasts, are the primary target for vitamin D analog therapy. Osteoblastic cell lines express the VDR, whereas the bone resorbing cells, the osteoclasts, have long been considered not to possess VDR. A recent study using human bone biopsy samples has, however, detected expression of VDR in a percentage of osteoclasts and stromal cells, suggesting that these cells may also be direct targets for VDR agonists [110]. Previous studies have shown that a number of the 20-epi-analogs (MC 1288, KH 1060, and CB 1093) potently stimulate the production of type I procollagen and osteocalcin from cultured osteoblastic cell lines [33,111,112]. In addition, a number of these vitamin D analogs have been found to increase the sensitivity of the osteoblasts to treatment with estrogens, as measured by stimulation of the creatine kinase B activity in ROS 17/2.8 cells [113,114]. These findings support a role for the combination of vitamin D compounds and estrogens in the treatment of postmenopausal osteoporosis.

1501 Further understanding of the effects of vitamin D analogs on osteoblast functions comes from a recent study, in which the ability of several vitamin D analogs (MC 1288, EB 1089, CB 1093, and KH 1060) to induce apoptosis in human osteoblast-like MG-63 cells was investigated [60]. No induction of apoptosis was seen, but some of the analogs exerted antiapoptotic effects in MG-63 cells treated with known apoptosisinducing agents. These findings suggest that the ability of vitamin D compounds to stimulate bone formation in vivo may, to some extent, be linked to their ability to protect the osteoblasts from undergoing apoptosis. Animal models provide evidence of the ability of 1,25(OH)2D3 and 1α(OH)D3 to stimulate bone formation and to improve bone mass in animals with experimental osteoporosis. In normal rats, short-term treatment with 1,25(OH)2D3 depressed osteoclast numbers, augmented osteoblast recruitment, and increased bone mass [115]. In ovariectomized rats, a model of estrogen-depletion induced bone loss, 1,25(OH)2D3 dose-dependently increased bone mass, but also increased serum calcium levels [116]. Two new vitamin D analogs have recently been investigated for their ability to prevent bone loss and to increase bone strength in ovariectomized rats. The first of these analogs, HEP 187, is a 20-epi vitamin D analog, in which the side chain hydroxy group that interacts with the VDR, has been replaced by a fluorine atom (Section III, Table II). The introduction of the fluorine atom is intended to increase the metabolic stability of the 20-epi side chain. In vitro, HEP 187 dosedependently increased secretion of osteocalcin and type I procollagen from MG-63 osteoblast-like cells. HEP 187 was tested in vivo in ovariectomized rats, using oral administration of 0.1 and 1.0 µg/kg/day. Treatment increased bone mineral content and bone strength to the same extent as 1,25(OH)2D3, but with lower calcemic effects [117]. Further screening has led to the identification of our present lead compound, GS 1790 (Section III, Table V). GS 1790 is a 20-epianalog with an aromatic ring in the side chain and a sulfur atom at the 23 position. In vitro, GS 1790 potently increased the production of osteocalcin and type I procollagen. In the ovariectomy model, GS 1790 was administered orally at doses from 10 to 50 µg/kg/day for 7 weeks to aging rats. pQTC scanning showed a significantly increased total and trabecular bone mineral density and biomechanical tests showed an increased bone strength (Table VI). In contrast to the rats treated with 1,25(OH)2D3, which showed a reduced local bone mineralization, GS 1790 promoted bone formation and bone strength with normal mineralization patterns and lower calcemic effects [118]. Recently, some fascinating studies have been performed with the vitamin D analog EB 1089 on the effect

1502

LISE BINDERUP

TABLE VI Test compound

Dosage (p.o)

Vehicle (non-OVX)

—

Vehicle (OVX control)

—

1,25(OH)2D3 (OVX)

0.05 µg/kg

GS 1790 (OVX)

10 µg/kg

GS 1790 (OVX)

25 µg/kg

ET AL .

Effects of GS 1790 in Ovariectomized Rats Serum calcium mmol/l 2.55 (0.10) 2.57 (0.09) 2.91b (0.10) 2.62 (0.17) 2.69 (0.06)

Trabecular density mg/cm3 324.8b (93.6) 157.6 (33.5) 290.8a (65.4) 242.2 (49.7) 297.7a (82.9)

Breaking strength N 290.8 (90.7) 187.7 (52.6) 318.9a (60.7) 323.6a (70.8) 327.3a (56.9)

Female Sprague Dawley rats were treated once daily for 28 days. Trabecular density was measured in the proximal tibia and bone breaking strength in the L3 vertebral body. aP < 0.05 vs. OVX control. bP < 0.01 vs. OVX control (Kruskall-Wallis followed by Dunn’s test).

of microgravity on bone cells. Long-term exposure to microgravity, as experienced by astronauts, leads to a substantial loss of bone mass, and a NASA supported program is evaluating potential candidates for bone protection during space flights. In one study, EB 1089 has been shown to selectively stimulate osteoblastogenesis, but not osteoclastogenesis, in bone marrow cell cultures from hind limb suspended rats [119]. Another study with osteoblast-type cells grown in a simulated microgravity environment has shown that treatment with EB 1089 is able to reverse the defects in cell differentiation and VDR expression induced by microgravity conditions [120]. The use of new vitamin D analogs in the treatment of a variety of bone disorders thus seems to attract new attention. The discovery of selective effects of the analogs on bone cell growth, differentiation, and apoptosis, together with new information on signal transduction pathways and genes involved in the regulation of calcium metabolism makes this an attractive area for further investigation.

V. CLINICAL DEVELOPMENT OF LEO ANALOGS A. Pharmacokinetics and Metabolism The research and development program for vitamin D analogs at LEO has yielded a number of candidates for the treatment of skin diseases, cancer, immunological disorders, and bone diseases. Such candidates must meet specific pharmacological requirements, and they

must also have a specific pharmacokinetic/metabolic profile, suitable to the intended route of administration. To minimize the risk of discontinuation due to a poor pharmacokinetic profile [121], investigations of the pharmacokinetic/metabolic properties of drug candidates are performed early in the development program, so that the profile of the compound may be optimized both with regard to efficiency and to safety. The major challenge is related to the bioanalytical capabilities, as vitamin D analogs are active at very low doses, and the concentrations in biological matrices are correspondingly very low. However, with the introduction of new generations of triple quadropole mass spectrometers, it is now possible to quantify vitamin D analogs in the lower pg/mL serum range [122]. In the following, the importance of the pharmacokinetic profile of vitamin D compounds is illustrated, using calcipotriol (MC 903) and seocalcitol (EB 1089) as examples. 1. CALCIPOTRIOL

Calcipotriol has now been used for more than 10 years as an efficacious and safe topical treatment of psoriasis. Investigations of the systemic pharmacokinetics of calcipotriol have shown that calcipotriol is very rapidly eliminated. After intravenous administration to rats, calcipotriol had a serum half-life of less than 10 min., whereas 1,25(OH)2D3 had a half-life of 2.3 hours [123]. Thus, the area under the serum level/ time curve (AUC) was more than 100 times lower for calcipotriol than for 1,25(OH)2D3. Using the postmitochondrial fraction from livers from rats, minipigs, and humans, the two major metabolites of calcipotriol,

1503

CHAPTER 84 Development of New Vitamin D Analogs

MC 1046 and MC 1080 (see Table III), were identified. The metabolic pathway involved oxidation at carbon 24 in the side chain and reduction of the ∆22 double bond. Formation of the 24-oxidized metabolites was found to constitute a deactivation pathway for calcipotriol, as the biological effects of MC 1046 and MC 1080 on cell proliferation and differentiation were much weaker than those exerted by calcipotriol (see Table III). The rapid systemic elimination due to the formation of metabolites with a low biological activity explains, at least partly, the very low calcemic activity of calcipotriol. In contrast to the rapid systemic metabolism, the metabolism of calcipotriol in different human keratinocytes (HPKA1 and HaCaT cells lines) and in vivo in rat skin was found to be very slow [124,125]. Therefore, calcipotriol has the ideal pharmacokinetic/ metabolic profile for a topical drug, being metabolically stable at the target site to exert the maximal therapeutic effect, and thereafter being metabolized into inactive metabolites once it enters the systemic circulation, thus minimizing the risk of side effects. 2. SEOCALCITOL

Important parameters in the selection of the vitamin D analog seocalcitol as a candidate for the treatment of cancer have been its bioavailability and metabolic stability. In vivo pharmacokinetic studies in rats and in vitro studies in postmitochondrial liver fraction from rats, minipigs, and humans have shown that seocalcitol has a half-life in serum comparable to that of 1,25(OH)2D3 and that the analog is metabolically very stable [126]. Although the serum half-lives of seocalcitol and 1,25(OH)2D3 are comparable, their apparent volumes of distribution are very different, as demonstrated by large differences in their serum concentrations after intravenous administration of the same dose [14]. This is explained by the difference in their binding affinity for the serum vitamin D–binding protein (DBP), for which 1,25(OH)2D3 has a 30-fold higher binding affinity. Thus, the tissue distribution of the two compounds is different. From the autoradiogram in Fig. 1 it can be seen that seocalcitol is accumulated in the Brain

Skin

Lung

TABLE VII Effects of EB 1089 (seocalcitol) on Tumor Cell Proliferation In Vitro Cell line

Test compound

MCF-7 (human breast cancer) HT-29 (human colon cancer) B16 (mouse melanoma)

EB 1089 1,25(OH)2D3 Daunomycin EB 1089 1,25(OH)2D3 Daunomycin EB 1089 1,25(OH)2D3 Daunomycin

Caecum

Rectum

Prostate

FIGURE 1 Tissue distribution of seocalcitol in a rat 8 hours after oral administration of 3H-seocalcitol

(200 µg/kg).

2 × 10−10 1 × 10−8 2 × 10−8 8 × 10−10 4 × 10−8 8 × 10−9 6 × 10−11 6 × 10−9 5 × 10−8

liver after a single oral dose to a rat, with concentrations in the liver about 10-fold higher than in serum. Similar results were found in minipigs after oral administration of seocalcitol [127]. Due to the lack of a sensitive bioanalytical assay, very few pharmacokinetic studies with seocalcitol have been performed in humans. However, a pharmacokinetic study was performed in 84 healthy subjects, who were given a single oral dose of 15 µg seocalcitol as enterocoated capsules. The serum concentration versus time profile was characterized by a steady absorption following the lag phase (2 hours). The maximal serum concentration (Cmax) was 57 pg/mL, and it was achieved at a median of 5 hours. The serum half-life was 14 hours (LEO internal report). Assuming that the ratio of the liver/serum concentration of seocalcitol in humans is similar to that in rats and minipigs, the maximal concentration of seocalcitol in the human liver is above 450 pg/g (∼10−9M). As the IC50 for the antiproliferative effect of seocalcitol in many cancer cells is 10−10–10−9M (Table VII), the concentration of seocalcitol in the liver is expected to be sufficiently high to be clinically effective. The metabolic stability and the high affinity of seocalcitol for the liver makes it suitable for systemic treatment, with the liver as an attractive target organ.

Small intestine Large intestine

Liver Stomach

Inhibition of proliferation IC50 (M)

1504

LISE BINDERUP

B. EFFICACY AND SAFETY 1. PSORIASIS

Calcipotriol The rationale for the use of vitamin D analogs in the treatment of psoriasis is described in Section IV.E.2 and in Chapter 101. The efficacy and safety of topically applied calcipotriol to patients with mild to moderate psoriasis was established in doubleblind, placebo-controlled studies initiated in 1987 [128,129]. Calcipotriol was launched in 1991, and numerous studies have since established the usefulness of this treatment of chronic plaque psoriasis (Table VIII). Calcipotriol is at least as effective as potent topical steroids, and is more effective than tazarotene (retinoid), short contact dithranol, and tar in the treatment of mild to moderate plaque psoriasis. Combination Therapy Since the introduction of calcipotriol, the treatment of psoriasis has been further improved by combination therapy. A review of representative studies is included in Table VIII. Light therapy (PUVA and UVB) combined with calcipotriol seems to have a beneficial effect even at reduced UV doses. In moderate to severe psoriasis, calcipotriol has been used in combination with systemic treatments such as cyclosporine A, etretinate, methotrexate, or acitretin. In these studies, it was possible to reduce the dose of the systemic drug, thus reducing the risk of side effects. However, more long-term combination studies are required to evaluate the long-term risk-benefit ratio.

TABLE VIII

ET AL .

Calcipotriol has also been combined with various topical steroids, resulting in an improved efficacy compared to calcipotriol used as monotherapy. Daivobet The beneficial effects of combining calcipotriol with steroids encouraged the development of the combination product Daivobet® (Table VIII). The rationale for this combination therapy is discussed in Section IV.E.2. Four phase III studies including more than 4,500 psoriasis patients were performed with Daivobet ointment, containing 50 µg/g calcipotriol and 0.5 mg/g betamethasone diproprionate (Table VIII). Once daily application was well tolerated and was more effective than twice daily use of calcipotriol or steroid alone, with a more rapid onset of response. 2. CANCER

Seocalcitol The rationale for the use of vitamin D analogs in cancer is briefly described in Section IV.E.1, and more fully discussed in Chapters 89–97. A clinical program for the study of seocalcitol in the treatment of cancer was initiated in 1995. Dosing experience from both phase I and II studies indicated that 15 µg seocalcitol per day was a safe starting dose. The primary toxicity of the drug was associated with its calcemic properties, i.e. hypercalcemia and related clinical symptoms. Phase II studies in leukemia, breast cancer, and pancreatic cancer have been completed without observation of any complete or partial response.

Calcipotriol: Monotherapy and Combination Therapy of Psoriasis

Calcipotriol Monotherapy Comparative Studies Comparative studies of calcipotriol and corticosteroids

References Crosti et al. [138]; Molin et al. [139]; Bruce et al. [140]; Cunliffe et al. [141]; Kragballe et al. [142] Kokelj et al. [143]

Comparative study evaluating efficacy and tolerability of calcipotriol and tazarotene in psoriatic lesions treatment Comparative studies of calcipotriol and short contact dithranol in psoriasis Wall et al. [144]; Berth-Jones et al. [145] vulgaris A comparative study of calcipotriol ointment and tar in chronic plaque psoriasis Tham et al. [146] Calcipotriol in Combination Therapy Studies Comparative studies of calcipotriol in combination with PUVA Comparative studies of calcipotriol in combination with UVB (broad and narrow band) Calcipotriol combined with systemic treatment (cyclosporine, etretinate, methotrexate, or acitretin) Comparative studies of calcipotriol in combination with corticosteroids (betamethasone, clobetasone, or halobetasol) Calcipotriol in combination with betamethasone dipropionate (Daivobet®) Calcipotriol plus short-contact dithranol: A novel topical combination therapy for chronic plaque psoriasis

Youn et al. [147]; Frappaz et al. [148] Ramsay et al. [149]; Molin [150]; Picot et al. [151]; Brands et al. [152]; Kerscher et al. [153] Grossman et al. [154]; Giannetti et al. [155]; de Jong et al. [156]; van de Kerkhof et al. [157] Kragballe et al. [158]; Ruzicka et al. [159]; Ortonne [160]; Lebwohl et al. [161] Kaufmann et al. [162]; Guenther et al. [163]; Douglas et al. [164]; Papp et al. [165] Monastirli et al. [166]

CHAPTER 84 Development of New Vitamin D Analogs

In contrast, a phase II study in hepatocellular carcinoma (HCC) showed encouraging results [130]. HCC is a primary malignant tumor of the liver, accounting for more than 300,000 cases per year [131]. The annual incidence is very high in Southeast Asia, with approximately 20 cases per 100,000 persons. The incidence in Western countries is considerably lower, but increasing. HCC has a very poor prognosis, and none of the drugs developed so far has resulted in an efficacious treatment of this disease. The vitamin D receptor is expressed in normal liver tissue [132,133], with increased expression both in HCC [134] and in colorectal hepatic metastasis [135]. In vitro studies have shown that liver cancer cell lines also express the VDR and are responsive to the antiproliferative effects of 1,25(OH)2D3 and analogs [136,137]. The effects of seocalcitol on HCC have been studied in C3H/Sy mice, a strain with a very high incidence of spontaneous liver tumors. Only 4% of the mice treated with seocalcitol developed HCC, compared to 36% in the control group [88] (see Section IV.E.1). These findings, coupled with the previously described metabolic stability and high affinity of seocalcitol for the liver, supported the decision to undertake the above mentioned phase II study with seocalcitol in patients with inoperable HCC [130]. Out of 33 patients evaluable for tumor response, two complete responses and 12 cases of stable disease have been reported from this study. Tumor tissue from one of the complete responders was tested for the presence of vitamin D receptors by Western blot and was found positive. At present, two phase III studies in patients with HCC are ongoing. One study is evaluating the efficacy of seocalcitol in prolonging survival of patients with HCC not amendable to curative treatment. Another study is evaluating the efficacy of seocalcitol in prolonging the time to relapse, following intended curative resection or percutaneous ablative treatment for HCC. In conclusion, some of the novel, synthetic vitamin D analogs have shown clinical usefulness, and it is to be expected that their number will increase, together with our expanding knowledge of their chemistry and biology. Results from these studies will be made available at the NIH/Vitamin D workshop Meeting, November 2004, Bethesda, Maryland [167].

References 1. DeLuca HF 1988 The vitamin D story: a collaborative effort of basic science and clinical medicine. FASEB J 2:224–236. 2. Minghetti PP, Norman AW 1989 1,25(OH)2-Vitamin D3 receptors: gene regulation and genetic circuitry. FASEB J 2:3043–3053.

1505 3. Rigby WFC 1988 The immunobiology of vitamin D. Immunol Today 9:54–58. 4. Manolagas SC, Hustmyer FG, Yu XP 1990 Immunomodulating properties of 1,25-dihydroxyvitamin D3. Kidney Int 38:9–16. 5. Wieslander SB, Mortensen BT, Binderup L, Nissen NI 1987 1α(OH)D3 (ETALPHA)R treatment and receptor studies in 16 patients with chronic and myeloproliferative disorders. Eur J Haematol 39:35–38. 6. Cunningham D, Gilchrist NL, Cowan RA, Forrest GJ, McArdle C, Soukop M 1985 Alfacalcidol as a modulator of growth of low grade non-Hodgkin’s lymphomas. BMJ 291: 1153–1155. 7. Morimoto S, Kumahara Y 1985 A patient with psoriasis cured by 1α-hydroxyvitamin D3. Med J Osaka Univ 35:51–54. 8. Dikstein S, Hartzshtark A 1984 Cosmetic and dermatological compositions containing 1-alpha-hydroxycholecalciferol. European Patent No.0129003B2. 9. Mezzetti G, Bagnara G, Monti MG, Bonsi L, Brunelli MA, Barbiroli B 1984 1α,25-dihydroxycholecalciferol and human histiocytic lymphoma cell line (U-937): The presence of a receptor and inhibition of proliferation. Life Sci 34: 2185–2191. 10. Binderup L, Bramm E 1988 Effects of a novel vitamin D analog MC 903 on cell proliferation and differentiation in vitro and on calcium metabolism in vivo. Biochem Pharmacol 37: 889–895. 11. Mellon WS, Franceschi RT, DeLuca HF 1980 An in vitro study of the stability of the chicken intestinal cytosol 1,25dihydroxyvitamin D3-specific receptor. Arch Biochem Biophys 202:83–92. 12. Gallagher JA, Beneton M, Harvey L, Lawson DEM 1986 Response of rachitic rat bones to 1,25-dihydroxyvitamin D3: Biphasic effects on mineralization and lack of effect on bone resorption. Endocrinology 119:1603–1609. 13. Norman AW, Wong RG 1972 Biological activity of the vitamin D metabolite 1,25-dihydroxycholecalciferol in chickens and rats. J Nutr 102:1709–1718. 14. Kissmeyer AM, Mathiasen IS, Latini S, Binderup L 1995 Pharmacokinetic studies of vitamin D analogs: relationship to vitamin D–binding protein (DBP). Endocrine 3:263–266. 15. Kream BE, Jose JL, DeLuca HF 1977 The chick intestinal cytosol-binding protein for 1,25-dihydroxyvitamin D3: A study of analog binding. Arch Biochem Biophys 179:462–468. 16. Hesse RH 1983 Intermediates in the synthesis of vitamin D derivatives. European Patent No.0078704. 17. Andrews DR, Barton DHR, Hesse RH, Pechet MM 1986 Synthesis of 25-hydroxy- and 1α,25-dihydroxyvitamin D3 from vitamin D2 (calciferol). J Org Chem 51:4819–4828. 18. Calverley MJ 1987 Synthesis of MC 903, a biologically active vitamin D metabolite. Tetrahedron 43:4609–4619. 19. Binderup E, Calverley MJ, Binderup L 1991 Synthesis and biological activity of 1α-hydroxylated vitamin D analogs with polyunsaturated side chains. In: AW Norman, et al. (eds) “Vitamin D: Gene Regulation, Structure-Function Analysis, and Clinical Application, Proceedings of the 8th Workshop on Vitamin D, Paris, France.” de Gruyter W, Berlin, New York, pp. 192–193. 20. Gielen JWJ, Koolstra RB, Jacobs HJC, Havinga E 1977 Triplet-sensitized interconversion and photooxygenation of vitamin D and trans-vitamin D. Recl Trav Chim Pays-Bas 113:306–311. 21. Bretting C, Calverley MJ, Binderup L 1991 Synthesis and biological activity of 1α-hydroxylated vitamin D3 analogs with hydroxylated side chains, multihomologated in the 24- or 24,26,27 positions. In: AW Norman, et al. (eds)

1506

22.

23.

24.

25.

26.

27.

28. 29.

30.

31.

32.

33.

34.

“Vitamin D: Gene Regulation, Structure-Function Analysis, and Clinical Application, Proceedings of the 8th Workshop on Vitamin D, Paris, France.” de Gruyter W, Berlin, New York, pp. 159–160. Hansen K, Calverley MJ, Binderup L 1991 Synthesis and biological activity of 22-oxa vitamin D analogs. In: AW Norman, et al. (ed) “Vitamin D: Gene Regulation, Structure-Function Analysis, and Clinical Application, Proceedings of the 8th Workshop on Vitamin D, Paris, France.” de Gruyter W, Berlin, New York, pp. 161–162. Calverley MJ, Binderup E, Binderup L 1991 The 20-epi modification in the vitamin D series: Selective enhancement of “non-classical” receptor-mediated effects. In: AW Norman, et al. (eds) “Vitamin D: Gene Regulation, Structure-Function Analysis, and Clinical Application, Proceedings of the 8th Workshop on Vitamin D, Paris, France.” de Gruyter W, Berlin, New York, pp. 163–164. Grue-Sørensen G, Binderup E, Binderup L 1994 Chemistry and biology of 23-oxa-aro- and 23-thia-aro-vitamin D analogs with high antiproliferative and low calcemic activity. In: AW Norman, et al. (eds) “Vitamin D: A pluripotent steroid hormone, Proceedings of the 9th Workshop on Vitamin D, Orlando, Florida.” de Gruyter W, Berlin, pp. 75–76. Ostrem VK, Tanaka Y, Prahl J, DeLuca HF, Ikekawa N 1987 24- and 26-homo-1,25-dihydroxyvitamin D3: Preferential activity in inducing differentiation of human leukemia cells HL-60 in vitro. Proc Natl Acad Sci USA 84:2610–2614. Kutner A, Perlman KL, Lago A, Sicinski RR, Schnoes HK, DeLuca HF 1988 Novel convergent synthesis of side-chain modified analogs of 1α,25-dihydroxycholecalciferol and 1α,25-dihydroxyergocalciferol. J Org Chem 53:3450–3457. Bretting C, Hansen CM, Andersen NR 1994 Chemistry and biology of 22,23-yne analogs of calcitriol. In: AW Norman, et al. (eds) “Vitamin D—A Pluripotent Steroid Hormone: Structural Studies, Molecular Endocrinology, and Clinical Applications. Proceedings of the 9th Workshop on Vitamin D, Orlando, USA.” de Gruyter W, Berlin, New York, pp. 73–74. Sørensen H, Binderup L, Calverley MJ, Hoffmeyer L, Andersen NR 1990 In vitro metabolism of calcipotriol (MC 903), a vitamin D analog. Biochem Pharmacol 39:391–393. Calverley MJ 1992 Novel vitamin D analogs in the calcipotriol (MC 903) series: synthesis and effect of side-chain structure on the selective biological actions. In: Sarel S, Mechoulam R, Agranat I (eds) Trends in Medicinal Chemistry ’90, Blackwell, Oxford, pp. 299–306. Binderup L, Latini S, Binderup E, Bretting C, Calverley M, Hansen K 1991 20-epi-vitamin D3 analogs: A novel class of potent regulators of cell growth and immune responses. Biochem Pharmacol 42:1569–1575. Mathiasen IS, Grue-Sørensen G, Hansen CM, Binderup L, Björkling F 1998 Studies on the interaction between the vitamin D receptor and the radiolabeled 20-epi vitamin D analog GS 1500. Biochem Biophys Res Commun 250:283–286. Peleg S, Sastry M, Collins ED, Bishop JE, Norman AW 1995 Distinct conformational changes induced by 20-epi analogs of 1α,25-dihydroxyvitamin D3 are associated with enhanced activation of vitamin D receptor. J Biol Chem 270:10551–10558. Ryhänen S, Mahonen A, Jääskeläinen T, Mäenpää PH 1996 Synthetic 20-epi analogs of calcitriol are potent inducers of target-gene activation in osteoblastic cells. Eur J Biochem 238:97–103. van den Bemd GJCM, Pols HAP, Birkenhäger JC, van Leeuwen JPTM 1996 Conformational change and enhanced stabilization of the vitamin D receptor by 1,25-dihydroxyvitamin D3 analog KH 1060. Proc Natl Acad Sci USA 93: 10685–10690.

LISE BINDERUP

ET AL .

35. Liu YY, Collins ED, Norman AW, Peleg S 1997 Differential interaction of 1α,25-dihydroxyvitamin D3 analogs and their 20-epi homologues with the vitamin D receptor. J Biol Chem 272:336–345. 36. Väisänen S, Juntunen K, Itkonen A, Vihko P, Mäenpää PH 1997 Conformational studies of human vitamin D receptor by antipeptide antibodies, partial proteolytic digestion, and ligand binding. Eur J Biochem 248:156–162. 37. Li XY, Boudjelal M, Xiao JH, Peng ZH, Asuru A, Kang S, Fisher GJ, Voorhees JJ 1999 1,25-dihydroxyvitamin D3 increases nuclear vitamin D3 receptors by blocking ubiquitin/ proteasome-mediated degradation in human skin. Mol Endocrinol 13:1686–1694. 38. Jääskeläinen T, Ryhänen S, Mahonen A, DeLuca HF, Mäenpää PH 2000 Mechanism of action of superactive vitamin D analogs through regulated receptor degradation. J Cell Biochem 76:548–558. 39. Tocchini-Valentini G, Rochel N, Wurtz JM, Mitschler A, Moras D 2001 Crystal structures of the vitamin D receptor complexed to superagonist 20-epi ligands. PNAS 98:5491–5496. 40. Choi M, Yamamoto K, Masuno H, Nakashima K, Taga T, Yamada S 2001 Ligand recognition by the vitamin D receptor. Bioorganic & Medicinal Chemistry 9:1721–1730. 41. Yang W, Freedman LP 1999 20-epi analogs of 1,25-dihydroxyvitamin D3 are highly potent inducers of DRIP coactivator complex binding to the vitamin D3 receptor. J Biol Chem 274:16838–16845. 42. Issa LL, Leong GM, Sutherland RL, Eisman JA 2002 Vitamin D analog-specific recruitment of vitamin D receptor coactivators. J Bone Miner Res 17:879–890. 43. Freedman LP 1999 Increasing the complexity of coactivation in nuclear receptor signaling Cell 97:5–8. 44. Lin R, Nagai Y, Sladek R, Bastien Y, Ho J, Petrecca K, Sotiropoulou G, Diamandis EP, Hudson TJ, White JH 2002 Expression profiling in squamous carcinoma cells reveals pleiotropic effects of vitamin D3 analog EB 1089 signaling on cell proliferation, differentiation, and immune system regulation. Mol Endocrinol 16:1243–1256. 45. Zhang F, Godyn JJ, Uskokovic M, Binderup L, Studzinski GP 1994 Monocytic differentiation of HL60 cells induced by potent analogs of vitamin D3 precedes the G1/G0 phase cell cycle block. Cell Proliferation 27:643–654. 46. Kobayashi T, Hashimoto K, Yoshikawa K 1993 Growth inhibition of human keratinocytes by 1,25-dihydroxyvitamin D3 is linked to dephosphorylation of retinoblastoma gene product. Biochem Biophys Res Commun 196:487–493. 47. Seol JG, Park WH, Kim ES, Jung CW, Binderup L, Koeffler HP, Kim BK, Lee YY 2000 Effect of a novel vitamin D3 analog, EB 1089, on G1 cell cycle regulatory proteins in HL-60 cells. Int J Oncol 16:315–320. 48. Park WH, Seol JG, Jung CW, Lee CC, Binderup L, Koeffler HP, Kim BK, Lee YY 2000 Cell cycle arrest induced by the vitamin D3 analog EB 1089 in NCI-H929 cells is associated with induction of the cyclin-dependent kinase inhibitor p27. Experimental Cell Research 254:279–286. 49. Liu W, Asa SL, Fantus G, Walfish PG, Ezzat S 2002 Vitamin D arrests thyroid carcinoma cell growth and induces p27 dephosphorylation and accumulation through PTEN/ Akt-dependent and -independent pathways. Am J Pathol 160: 511–519. 50. Pálmes HG, Gonzáles-Sancho JM, Espada J, Berciano MT, Puig I, Baulida J, Quintanilla M, de Herreros AG, Lafarga M, Muñoz A, Cano A 2001 Vitamin D3 promotes the differentiation of colon carcinoma cells by the induction of E-cadherin and the inhibition of β-catenin signaling. J Cell Biol 154:369–387.

CHAPTER 84 Development of New Vitamin D Analogs

51. Díaz D, Paraskeva C, Thomas MG, Binderup L, Hague A 1999 Apoptosis is induced by the active metabolite of vitamin D3 and its analog EB 1089 in colorectal adenoma and carcinoma cells: Possible implications for prevention and therapy. Cancer Res 60:2304–2312. 52. Park WH, Seol JG, Kim ES, Binderup L, Koeffler HP, Kim BK, Lee YY 2002 The induction of apoptosis by a combined 1,25(OH)2D3 analog, EB 1089 and TGF-β1 in NCI-H929 myeloma cells. Int J Oncol 20:533–542. 53. Larsen SS, Heiberg I, Lykkesfeldt AE 2001 Anti-estrogen resistant human breast cancer cell lines are more sensitive towards treatment with the vitamin D analog EB 1089 than parent MCF-7 cells. Br J Cancer 84:686–690. 54. Mathiasen IS, Lademann U, Jäätelä M 1999 Apoptosis induced by vitamin D compounds in breast cancer cells is inhibited by Bcl-2 but does not involve known caspases or p53. Cancer Res 59:4848–4856. 55. Narvaez CJ, Welsh JE 2001 Role of mitochondria and caspases in vitamin D–mediated apoptosis of MCF-7 breast cancer cells. J Biol Chem 276:9101–9107. 56. Mathiasen IS, Sergeev IN, Bastholm L, Elling F, Norman AW, Jäätelä M 2002 Calcium and calpain as key mediators of apoptosis-like death induced by vitamin D compounds in breast cancer cells. J Biol Chem 277:30738–30745. 57. Simboli-Campbell M, Narvaez CJ, Van Weelden K, Tenniswood M, Welsh JE 1997 Comparative effects of 1,25(OH)2D3 and EB 1089 on cell cycle kinetics and apoptosis in MCF-7 breast cancer cells. Breast Cancer Res Treat 42:31–41. 58. Pirianov G, Colston KW 2001 Interaction of vitamin D analogs with signalling pathways leading to active cell death in breast cancer cells. Steroids 66:301–318. 59. Doudina O, Zang C, Waechter M, Facklam M, Possinger K, Afanassiev B, Binderup L, Koeffler HP, Elstner E 2002 Combination of potent 20-epi-vitamin D3 analog (CB 1093) with pioglitazone inhibits clonal growth and induces apoptosis in MCF7 cell line. Proc Am Assoc Cancer Res 43:434. 60. Hansen CM, Hansen D, Holm PK, Binderup L 2001 Vitamin D compounds exert anti-apoptotic effects in human osteosarcoma cells in vitro. J Steroid Biochem Mol Biol 77:1–11. 61. Xie SP, James SY, Colston KW 1997 Vitamin D derivatives inhibit the mitogenic effects of IGF-1 on MCF-7 human breast cancer cells. J Endocrinol 154:495–504. 62. Koga M, Eisman JA, Sutherland RL 1988 Regulation of epidermal growth factor receptor levels by 1,25-dihydroxyvitamin D3 in human breast cancer cells. Cancer Res 48:2734–2739. 63. Vink-van Wijngaarden T, Pols HAP, Buurman CJ, Birkenhäger JC, van Leeuwen JPTM 1996 Inhibition of insulin and insulinlike growth factor-I-stimulated growth of human breast cancer cells by 1,25-dihydroxyvitamin D3 and the vitamin D3 analog EB 1089. Eur J Cancer 32A, 842–848. 64. Oh YS, Kim EJ, Schaffer BS, Kang YH, Binderup L, MacDonald RG, Park JHY 2001 Synthetic low-calcemic vitamin D3 analogs inhibit secretion of insulin-like growth factor II and stimulate production of insulin-like growth factorbinding protein-6 in conjunction with growth suppression of HT-29 colon cancer cells. Mol Cell Endocrinol 183:141–149. 65. Boyle BJ, Zhao XY, Cohen P, Feldman D 2001 Insulin-like growth factor binding protein-3 mediates 1α-dihydroxyvitamin D3 growth inhibition in the LNCaP prostate cancer cell line through P21/WAF1. J Urol 165:1319–1324. 66. Huynh H, Chow PKH, Ooi LP, Soo KC 2002 A possible role for insulin-like growth factor-binding protein-3 autocrine/ paracrine loops in controlling hepatocellular carcinoma cell proliferation. Cell Growth Differ 13:115–122.

1507 67. Colston KW, Perks CM, Xie SP, Holly JMP 1998 Growth inhibition of both MCF-7 and Hs578T human breast cancer cell lines by vitamin D analogs is associated with increased expression of insulin-like growth factor binding protein-3. J Mol Endocrinol 20:157–162. 68. Rozen F, Yang XF, Huynh H, Pollak M 1997 Antiproliferative action of vitamin D-related compounds and insulin-like growth factor binding protein-5 accumulation. JNCI 89: 652–656. 69. Lee EA, Jeon SH, Yi JY, Jin YJ, Son YS 2001 Calcipotriol inhibits autocrine phosphorylation of EGF receptor in a calciumdependent manner, a possible mechanism for its inhibition of cell proliferation and stimulation of cell differentiation. Biochem Biophys Res Commun 284:419–425. 70. Koli K, Keski-Oja J 1995 1,25-dihydroxyvitamin D3 enhances the expression of transforming growth factor β1 and its latent form binding protein in cultured breast carcinoma cells. Cancer Res 55:1540–1546. 71. Wu G, Fan RS, Li W, Srinivas V, Brattain MG 1998 Regulation of transforming growth factor-β type II receptor expression in human breast cancer MCF-7 cells by vitamin D3 and its analogs. J Biol Chem 273:7749–7756. 72. Jung CW, Kim ES, Seol JG, Park WH, Lee SJ, Kim BK, Lee YY 1999 Antiproliferative effect of a vitamin D3 analog, EB 1089, on HL-60 cells by the induction of TGF-β receptor. Leukemia Res 23:1105–1112. 73. Colston KW, Chander SK, Mackay AG, Coombes RC 1992 Effects of synthetic vitamin D analogs on breast cancer cell proliferation in vivo and in vitro. Biochem Pharmacol 44:693–702. 74. James SY, Mercer E, Brady M, Binderup L, Colston KW 1998 EB 1089, a synthetic analog of vitamin D, induces apoptosis in breast cancer cells in vivo and in vitro. Br J Pharmacol 125:953–962. 75. Van Weelden K, Flanagan L, Binderup L, Tenniswood M, Welsh JE 1998 Apoptotic regression of MCF-7 xenografts in nude mice treated with vitamin D3 analog, EB 1089. Endocrinology 139:2102–2110. 76. El Abdaimi K, Dion N, Papavasiliou V, Cardinal PE, Binderup L, Goltzman D, Ste-Marie LG, Kremer R 2000 The vitamin D analog EB 1089 prevents skeletal metastasis and prolongs survival time in nude mice transplanted with human breast cancer cells. Cancer Res 60:4412–4418. 77. Mackay AG, Ofori-Kuragu EA, Lansdown A, Coombes RC, Binderup L, Colston KW 1996 Effects of the synthetic vitamin D analog EB 1089 and tamoxifen on the growth of experimental rat mammary tumours. Endocrine-Related Cancer 3: 327–335. 78. Koshizuka K, Koike M, Asou H, Cho SK, Stephen T, Rude RK, Binderup L, Uskokovic M, Koeffler HP 1999 Combined effect of vitamin D3 analogs and Paclitaxel on MCF-7 breast cancer cells in vivo. Breast Cancer Res Treat 53:113–120. 79. Colston KW, Mackay AG, James SY 1994 The role of 1α25(OH)2D3 and its analogs in breast cancer. In: AW Norman, et al. (eds) “Vitamin D: A pluripotent steroid hormone: Proceedings of the 9th Workshop on Vitamin D, Orlando, Florida.” de Gruyter W, Berlin, pp. 477–484. 80. Gross C, Stamey T, Hancock S, Feldman D 1998 Treatment of early recurrent prostate cancer with 1,25-dihydroxyvitamin D3 (calcitriol). J Urol 159:2035–2040. 81. Skowronski RJ, Peehl DM, Cramer S, Feldman D 1994 Actions of 1,25-dihydroxyvitamin D and synthetic analogs on cultured human prostate carcinoma cells. In: AW Norman, et al. (eds) “Vitamin D: A pluripotent steroid hormone: Proceedings of the 9th Workshop on Vitamin D, Orlando, Florida.” de Gruyter W, Berlin, pp. 520–521.

1508 82. Blutt SE, McDonnell TJ, Weigel NL 1997 Effects of 1,25dihydroxyvitamin D3 on cell cycle distribution and apoptosis of LNCaP cells. Proc Am Assoc Cancer Res 38:436 83. de Vos S, Holden S, Heber D, Elstner E, Binderup L, Uskokovic M, Rude B, Chen DL, Le J, Cho SK, Koeffler HP 1997 Effects of potent vitamin D3 analogs on clonal proliferation of human prostate cancer cell lines. The Prostate 31: 77–83. 84. Lokeshwar BL, Schwartz GG, Selzer MG, Burnstein KL, Zhuang SH, Block NL, Binderup L 1999 Inhibition of prostate cancer metastasis in vivo: A comparison of 1,25-dihydroxyvitamin D (Calcitriol) and EB 1089. Cancer Epidemiology, Biomarkers and Prevention 8:241–248. 85. Blutt SE, Polek TC, Stewart LV, Kattan MW, Weigel NL 1999 A calcitriol analog, EB 1089, inhibits the growth of LNCaP tumor in nude mice. Cancer Res 60:779–782. 86. Perez-Stable CM, Schwartz GG, Farinas A, Finegold M, Binderup L, Howard GA, Roos BA 2002 The Gγ/T-15 transgenic mouse model of androgen-independent prostate cancer: Target cells of carcinogenesis and the effect of the vitamin D analog EB 1089. Cancer Epidemiology, Biomarkers & Prevention 11:555–563. 87. Akhter J, Chen X, Bowrey P, Bolton EJ, Morris DL 1997 Vitamin D3 analog, EB 1089, inhibits growth of subcutaneous xenografts of the human colon cancer cell line, LoVo, in a nude mouse model. Dis Colon Rectum 40:317–321. 88. Sahpazidou D, Stravoravdi P, Toliou T, Geromichalos G, Zafiriou Z, Natsis K, Gigis P 2003 Significant experimental decrease of the hepatocellular carcinoma incidence in C3H/Sy mice after long-term administration of EB 1089, a vitamin D analog. Oncol Res 13:261–268. 89. Veldman CM, Cantorna MT, DeLuca HF 2000 Expression of 1,25-dihydroxyvitamin D3 receptor in the immune system. Arch Biochem Biophys 374:334–338. 90. Lemire JM, Archer DC, Beck L, Spiegelberg HL 1995 Immunosuppressive actions of 1,25-dihydroxyvitamin D3: Preferential inhibition of Th1 functions. J Nutr 125:1704–1708. 91. Bouillon R, Garmyn M, Verstuyf A, Segaert S, Casteels K, Mathieu C 1995 Paracrine role for calcitriol in the immune system and skin creates new therapeutic possibilities for vitamin D analogs. Eur J Endocrinol 133:7–16. 92. Mattner F, Smiroldo S, Galbiati F, Muller M, Di Lucia P, Poliani PL, Martino G, Panina-Bordignon P, Adorini L 2000 Inhibition of Th1 development and treatment of chronicrelapsing experimental allergic encephalomyelitis by a nonhypercalcemic analog of 1,25-dihydroxyvitamin D3. Eur J Immunol 30:498–508. 93. van de Kerkhof PCM 2001 New developments in the treatment of psoriasis. Skin Pharmacol Appl Skin Physiol 14:129–135. 94. Barker JNWN 1998 Psoriasis as a T-cell mediated autoimmune disease. Hospital Medicine 59:530–533. 95. Jirapongsananuruk O, Melamed I, Leung DYM 2000 Additive immunosuppressive effects of 1,25-dihydroxyvitamin D3 and corticosteroids on Th1, but not Th2, responses. J Allergy Clin Immunol 106:981–985. 96. Barrat FJ, Cua DJ, Boonstra A, Richards DF, Crain C, Savelkoul HF, de Waal-Malefyt R, Coffman RL, Hawrylowicz CM, O’Garra A 2002 In vitro generation of interleukin 10producing regulatory CD4(+) T cells is induced by immunosuppressive drugs and inhibited by T helper type 1 (Th1) and Th2-inducing cytokines. J Exp Med 195:603–616. 97. Johnsson C, Tufveson G 1994 MC 1288—A vitamin D analog with immunosuppressive effects on heart and small bowel grafts. Transpl Int 7:392–397.

LISE BINDERUP

ET AL .

98. Johnsson C, Binderup L, Tufveson G 1996 Immunosuppression with the vitamin D analog MC 1288 in experimental transplantation. Transplant Proc 28:888–891. 99. Johnsson C, Binderup L, Tufveson G 1995 The effects of combined treatment with the novel vitamin D analog MC 1288 and cyclosporine A on cardiac allograft survival. Transplant Immunology 3:245–250. 100. Johnsson C, Tufveson G, Binderup L, Karlsson-Parra A 1997 Synergistic actions of the vitamin D analog MC 1288 and 15deoxyspergualin in xenotransplantation. Xenotransplantation 4:186–193. 101. Räisänen-Sokolowski AK, Pakkala IS, Samila SP, Binderup L, Häyry PJ, Pakkala ST 1997 A vitamin D analog, MC 1288, inhibits adventitial inflammation and suppresses intimal lesions in rat aortic allografts. Transplantation 63:936–941. 102. Mathieu C, Waer M, Casteels K, Laureys J, Bouillon R 1995 Prevention of type I diabetes i NOD mice by nonhypercalcemic doses of a new structural analog of 1,25-dihydroxyvitamin D3, KH 1060. Endocrinology 136:866–872. 103. Casteels K, Waer M, Laureys J, Valckx D, Depovere J, Bouillon R, Mathieu C 1998 Prevention of autoimmune destruction of syngeneic islet grafts in spontaneously diabetic nonobese diabetic mice by a combination of a vitamin D3 analog and cyclosporine. Transplantation 65:1225–1232. 104. Casteels KM, Mathieu C, Waer M, Valckx D, Overbergh L, Laureys JM, Bouillon R 1998 Prevention of type I diabetes in nonobese diabetic mice by late intervention with nonhypercalcemic analogs of 1,25-dihydroxyvitamin D3 in combination with a short induction course of cyclosporin A. Endocrinology 139:95–102. 105. Wulff EM, Bramm E, Binderup L 1995 MC 1288—A vitamin D analog with immunosuppressive effect on experimental autoimmune encephalomyelitis (EAE). Inflammation Res 44:S232 106. Arnold WP, van de Kerkhof PCM 1991 The induction of epidermal ornithine decarboxylase following tape stripping is inhibited by a topical vitamin D3 analog (MC 903). Br J Dermatol 125:6–8. 107. Stio M, Bonanomi AG, d’Albasio G, Treves C 2001 Suppressive effect of 1,25-dihydroxyvitamin D3 and its analogs EB 1089 and KH 1060 on T-lymphocyte proliferation in active ulcerative colitis. Biochem Pharmacol 61:365–371. 108. Civitelli R 1995 The role of vitamin D metabolites in the treatment of osteoporosis. Calcif Tissue Int 57:409–414. 109. Gorai I, Chaki O, Taguchi Y, Nakayama M, Osada H, Suzuki N, Katagiri N, Misu Y, Minaguchi H 1999 Early postmenopausal bone loss is prevented by estrogen and partially by 1α-OH-vitamin D3: Therapetic effects of estrogen and/or 1α-OH-vitamin D3. Calcif Tissue Int 65:16–23. 110. Langub MC, Reinhardt TA, Horst RL, Malluche HH, Koszewski NH 2000 Characterization of vitamin D receptor immunoreactivity in human bone cells. Bone 27:383–387. 111. van den Bemd GJCM, Pols HAP, Birkenhäger JC, Kleinekoort WMC, van Leeuwen JPTM 1995 Differential effects of 1,25-dihydroxyvitamin D3 analogs on osteoblastlike cells and on in vitro bone resorption. J Steroid Biochem Molec Biol 55:337–346. 112. Ryhänen S, Jääskeläinen T, Saarela JTA, Mäenpää PH 1998 Inhibition of proliferation and induction of differentiation of osteoblastic cells by a novel 1,25-dihydroxyvitamin D3 analog with an extensively modified side chain (CB 1093). J Cell Biochem 70:414–424.

CHAPTER 84 Development of New Vitamin D Analogs

113. Sömjen D, Waisman A, Weisman J, Kaye AM 1998 Nonhypercalcemic analogs of vitamin D stimulate creatine kinase B activity in osteoblast-like ROS 17/2.8 cells and up-regulate their responsiveness to estrogens. Steroids 63:340–343. 114. Somjen D, Waisman A, Weisman Y, Kaye AM 2000 “Nonhypercalcemic” analogs of 1α,25 dihydroxy vitamin D3 augment the induction of creatine kinase B by estrogen and selective estrogen receptor modulators (SERMS) in osteoblastlike cells and rat skeletal organs. J Steroid Biochem Molec Biol 72:79–88. 115. Erben RG, Scutt AM, Miao D, Kollenkirchen W, Haberey M 1997 Short-term treatment of rats with high dose of 1,25 dihydroxyvitamin D3 stimulates bone formation and increases the number of osteoblast precursor cells in bone marrow. Endocrinology 138:4629–4635. 116. Erben RG, Bromm S, Stangassinger M 1998 Therapeutic efficacy of 1α,25-dihydroxyvitamin D3 and calcium in osteopenic ovariectomized rats: Evidence for a direct anabolic effect of 1α,25-dihydroxyvitamin D3 on bone. Endocrinology 139:4319–4328. 117. Tranholm M, Holm PK, Hansen CM, Ryhänen S, Väisänen S, Binderup L, Thomsen JS, Mäenpää P, Mosekilde L 2000 In vitro and in vivo effects of a 1,25-dihydroxyvitamin D3 analog (HEP 187). Proceedings of the ASBMR 22nd Annual Meeting, Toronto, Canada, September 2000 S563. 118. Gonzales T, Holm PK, Nielsen JL, Thomsen JS, Ryhänen S, Väisänen S, Binderup L 2003 In vitro and in vivo effects of a vitamin D analog, GS 1790, on bone. Abstract at the 12th Workshop on Vitamin D, Maastricht, The Netherlands, July 2003. 119. Narayanan R, Gaddy-Kurten D, Montague DL, Smith CL, Weigel NL 2000 EB 1089 preferentially stimulates osteoblastogenesis in bone marrow cultures of Sprague Dawley rats. J Bone Miner Res 15:S575. 120. Narayanan R, Smith CL, Weigel NL 2002 Vector-averaged gravity-induced changes in cell signaling and vitamin D receptor activity in MG-63 cells are reversed by a 1,25(OH)2D3 analog, EB 1089. Bone 31:381–388. 121. Kennedy T 1997 Managing the drug discovery/development interface. Drug Discovery Today 2:436–444. 122. Kissmeyer AM, Sonne K, Binderup E 2000 Determination of the vitamin D analog EB 1089 (seocalcitol) in human and pig serum using liquid chromatography-tandem mass spectrometry. Journal of Chromatography B 740:117–128. 123. Kissmeyer AM, Binderup L 1991 Calcipotriol (MC 903): Pharmacokinetics in rats and biological activities of metabolites. A comparative study with 1,25(OH)2D3. Biochem Pharmacol 41:1601–1606. 124. Masuda S, Strugnell S, Calverley MJ, Makin HLJ, Kremer R, Jones G 1994 In vitro metabolism of the antipsoriatic vitamin D analog, calcipotriol, in two cultured human keratinocyte models. J Biol Chem 269:4794–4803. 125. Nielsen JL, Hansen CM, Grue-Sørensen G, Kissmeyer AM 2000 The rate of metabolism of calcipotriol, KH1650 and 1α,25(OH)2D3 in vitro in skin cells (HaCaT) and in vivo (rat skin). Abstract at the 11th Workshop on Vitamin D, Nashville, Tennessee, June 2000. 126. Kissmeyer AM, Binderup E, Binderup L, Hansen CM, Andersen NR, Makin HLJ, Schroeder NJ, Shankar VN, Jones G 1997 The metabolism of the vitamin D analog EB 1089: Identification of in vivo and in vitro liver metabolites and their biological activities. Biochem Pharmacol 53: 1087–1097.

1509 127. Kissmeyer AM, Mortensen JT 2000 Pharmacokinetics and metabolism of a vitamin D analog (Seocalcitol) in rat and minipig. Xenobiotica 30:815–830. 128. Kragballe K, Beck HI, Søgaard H 1988 Improvement of psoriasis by a topical vitamin D3 analog (MC 903) in a double-blind study. Br J Dermatol 119:223–230. 129. Staberg B, Roed-Petersen J, Menné T 1989 Efficacy of topical treatment in psoriasis with MC 903, a new vitamin D analog. Acta Derm Venereol 69:147–150. 130. Dalhoff K, Dancey J, Astrup L, Skovsgaard T, Hamberg KJ, Lofts FJ, Rosmorduc O, Erlinger S, Bach Hansen J, Steward WP, Skov T, Burcharth J, Evans TRJ 2003 A phase II study of the vitamin D analog Seocalcitol in patients with inoperable hepatocellular carcinoma. Br J Cancer 89:252–257. 131. Parkin DM, Pisani P, Ferlay J 1993 Estimates of the worldwide incidence of eighteen major cancers in 1985. Int J Cancer 54:594–606. 132. Berger U, Wilson P, McClelland RA, Colston K, Haussler MR, Pike JW, Coombes RC 1988 Immunocytochemical detection 1,25-dihydroxyvitamin D receptors in normal human tissues. J Clin Endocrinol Metab 67:607–613. 133. Gascon-Barré M, Demers C, Mirshahi A, Neron S, Zalzal S, Nanci A 2003 The normal liver harbors the vitamin D nuclear receptor in nonparenchymal and biliary epithelial cells. Hepatology 37:1034–1042. 134. Sakai Y, Fukuda Y, Yamamoto I, Dokoh S, Imura H 1988 Study on 1,25-dihydroxy-vitamin D3 receptor in human liver cancer tissue. Acta Hepatol Japan 29:1612–1618. 135. Demers C, Bolduc J, Bilodeau M, Lapointe R, Roy A, Dagenais M, Létourneau R, Coté J, Nguyen BN, GastonBarré M, Huet PM 2002 High expression of vitamin D nuclear receptors and metabolizing enzymes in colorectal hepatic metastases: A rationale for the use of vitamin D analogs in adjuvent antitumoral treatment. American Association for the Study of Liver Diseases Abstracts 4:397A. 136. Sakai Y, Fukuda Y, Hase K, Yamamoto I, Dokoh S, Imura H 1989 1,25-dihydroxy-vitamin D3 (VD) receptor in human liver cancer cell lines and effect of VD on their proliferation. Acta Hepatol Japan 30:338–344. 137. Akhter J, Lu Y, Finlay I, Pourgholami MH, Morris DL 2001 1α,25-dihydroxyvitamin D3 and its analogs, EB 1089 and CB 1093, profoundly inhibit the in vitro proliferation of the human hepatoblastoma cell line HEPG2. ANZ J Surg 71:414–417. 138. Crosti C, Finzi AF, Mian E, Scarpa C 1997 Calcipotriol in psoriasis vulgaris: a controlled trial comparing betamethasone dipropionate + salicylic acid. Int J Dermatol 36: 537–541. 139. Molin L, Cutler TP, Helander I, Nyfors B, Downes N 1997 Comparative efficacy of calcipotriol (MC 903) cream and betamethasone 17-valerate cream in the treatment of chronic plaque psoriasis. A randomized, double-blind, parallel group multicenter study. Br J Dermatol 136:89–93. 140. Bruce S, Epinette WW, Funicella T, Ison A, Jones EL, Loss R, McPhee ME, Whitmore C 1994 Comparative study of calcipotriene (MC 903) ointment and fluocinonide ointment in the treatment of psoriasis. J Am Acad Dermatol 31:755–759. 141. Cunliffe WJ, Berth-Jones J, Claudy A, Fairiss G, Goldin D, Gratton D, Henderson CA, Holden CA, Maddin WS, Ortonne JP, Young M 1992 Comparative study of calcipotriol (MC 903) ointment and betamethasone 17-valerate ointment in patients with psoriasis vulgaris. J Am Acad Dermatol 26: 736–743.

1510 142. Kragballe K, Karlsmark T, Nieboer C, Gjertsen BT, van de Kerkhof PCM, Roed-Petersen J, Tikjøb G, de Hoop D, Larkö O, Strand A 1991 Double-blind, right/left comparison of calcipotriol and betamethasone valerate in treatment of psoriasis vulgaris. Lancet 337:193–196. 143. Kokelj F, Plozzer C, Doria A, Trevisan G 1999 Comparative study evaluating efficacy and tolerability of calcipotriol and tazarotene in psoriatic lesions treatment. Br J Dermatol 141:987. 144. Wall ARJ, Poyner TF, Menday AP 1998 A comparison of treatment with dithranol and calcipotriol on the clinical severity and quality of life in patients with psoriasis. Br J Dermatol 139:1005–1011. 145. Berth-Jones J, Chu AC, Dodd WAH, Ganpule M, Griffiths WAD, Haydey RP, Klaber MR, Murray SJ, Rogers S, Jurgensen HJ 1992 A multicenter, parallel-group comparison of calcipotriol ointment and short-contact dithranol therapy in chronic plaque psoriasis. Br J Dermatol 127:266–271. 146. Tham SN, Lun KC, Cheong WK 1994 A comparative study of calcipotriol ointment and tar in chronic plaque psoriasis. Br J Dermatol 131:673–677. 147. Youn JI, Park BS, Park SB, Kim SD, Suh DH 2000 Comparison of calcipotriol-PUVA with conventional PUVA in the treatment of psoriasis. J Dermatol Treatment 11: 125–130. 148. Frappaz A, Thivolet J 1993. Calcipotriol in combination with PUVA: a randomized double-blind placebo study in severe psoriasis. Eur J Dermatol 3:351–354. 149. Ramsay CA, Schwartz BE, Lowson D, Papp K, Bolduc A, Gilbert M 2000 Calcipotriol cream combined with twice weekly broad-band UVB phototherapy: A safe effective and UVB-sparing antipsoriatric combination treatment. Dermatology 200:17–24. 150. Molin L 1999 Topical calcipotriol combined with phototherapy for psoriasis. Dermatology 198:375–381. 151. Picot A, Natta P, Thomas P, Beani JC, Leroy D, Cambazard F, Humbert P, Meynadier J 1999 Combination treatment with calcipotriol and narrow-band UVB for psoriasis. J Eur Acad Dermatol Venereol 12:S334–S335. 152. Brands S, Brakman M, Bos JD, de Rie MA 1999 No additional effect of calcipotriol ointment on low-dose narrowband UVB phototherapy in psoriasis. J Am Acad Dermatol 41:991–995. 153. Kerscher M, Volkenandt M, Plewig G, Lehmann P 1993 Combination phototherapy of psoriasis with calcipotriol and narrow-band UVB. Lancet 342:923. 154. Grossman RM, Thivolet J, Claudy A, Souteyrand P, Guilhou JJ, Thomas P, Amblard P, Belaich S, de Belilovsky C, de la Brassinne M, Martinet C, Bazex JA, Beylot C, Combemale P, Lambert D, Ostojic A, Denoeux JP, Lauret P, Vaillant L, Weber M, Pamphile R, Dubertret L 1994 A novel therapeutic approach to psoriasis with combination calcipotriol ointment and very low-dose cyclosporine. Results of a multicenter placebo-controlled study. J Am Acad Dermatol 31:68–74. 155. Giannetti A, Coppini M, Bertazzoni MG, Califano A, Altieri E, Pazzaglia A, Lega M, Lombardo M, Pelfini C, Fornasa CV, Rabbiosi G, Cespa M 1999 Clinical trial of the efficacy and safety of oral etretinate with calcipotriol cream compared with etretinate alone in moderate-severe psoriasis. J Eur Acad Dermatol Venereol 13:91–95.

LISE BINDERUP

ET AL .

156. de Jong EMGJ 1999 Review of calcipotriol in combination with methotrexate, cyclosporine, and systemic retinoids. J Eur Acad Dermatol Venereol 12:S332 157. van de Kerkhof PCM, Cambazard F, Hutchinson PE, Haneke E, Wong E, Souteyrand P, Damstra RJ, Combemale P, Neumann MHAM, Chalmers RJG, Olsen L, Revuz J 1998 The effect of addition of calcipotriol ointment (50 µg/g) to acitretin therapy in psoriasis. Br J Dermatol 138:84–89. 158. Kragballe K, Barnes L, Hamberg KJ, Hutchinson P, Murphy F, Møller S, Ruzicka T, van de Kerkhof PCM 1998 Calcipotriol cream with or without concurrent topical corticosteroid in psoriasis: tolerability and efficacy. Br J Dermatol 139: 649–654. 159. Ruzicka T, Lorenz B 1998 Comparison of calcipotriol monotherapy and a combination of calcipotriol and betamethasone valerate after two weeks’ treatment with calcipotriol in the topical therapy of psoriasis vulgaris: a multicenter, double-blind, randomized study. Br J Dermatol 138:254–258. 160. Ortonne JP 1994 Psoriasis: New therapeutic modality by calcipotriol and betamethasone dipropionate. Nouvelles Dermatologiques 13:746–751. 161. Lebwohl M, Siskin SB, Epinette W, Breneman D, Funicella T, Kalb R, Moore J 1996 A multicenter trial of calcipotriene ointment and halobetasol ointment compared with either agent alone for the treatment of psoriasis. J Am Acad Dermatol 35:268–269. 162. Kaufmann R, Bibby AJ, Bissonnette R, Cambazard F, Chu AC, Decroix J, Douglas WS, Lowson D, Mascaro JM, Murphy GM, Stymne B 2002 A new calcipotriol/betamethasone dipropionate formulation (Daivobet TM) is an effective once daily treatment for psoriasis vulgaris. Dermatology 205:389–393. 163. Guenther L, Cambazard F, van de Kerkhof PCM, Snellman E, Kragballe K, Chu AC, Tegner E, Garcia-Diez A, Springborg J 2002 Efficacy and safety of a new combination of calcipotriol and betamethasone dipropionate (once or twice daily) compared to calcipotriol (twice daily) in the treatment of psoriasis vulgaris: a randomized, double-blind, vehiclecontrolled clinical trial. Br J Dermatol 147:316–323. 164. Douglas WS, Poulin Y, Decroix J, Ortonne JP, Mrowietz U, Gulliver W, Krogstad AL, Larsen FG, Iglesias L, Buckley C, Bibby AJ 2002 A new calcipotriol/betamethasone formulation with rapid onset of action was superior to monotherapy with betamethasone dipropionate or calcipotriol in psoriasis vulgaris. Acta Derm Venereol 82:131–135. 165. Papp KA, Guenther L, Boyden B, Larsen FG, Havina RJ, Guilhou JJ, Kaufmann R, Rogers S, van de Kerkhof PCM, Hansen LI, Tegner E, Burg G, Talbot D, Chu AC 2003 Early onset of action and efficacy of a combination of calcipotriene and betamethasone in the treatment of psoriasis. J Am Acad Dermatol 48:48–54. 166. Monastirli A, Georgiou S, Pasmatzi E, Sakkis Th, Badavanis G, Drainas D, Sagriotis A, Tsambaos D 2002 Calcipotriol plus short-contact dithranol: A novel topical combination therapy for chronic plaque psoriasis. Skin Pharmacol Appl Skin Physiol 15:246–251. 167. Hamberg KJ 2004 Vitamin D and cancer: Status report on phase III trials with seocalcitol in hepatic cancer. J Steroid Biochem Molec Biol (in press).

CHAPTER 85

Gemini: The 1,25-dihydroxy Vitamin D Analogs with Two Side-Chains MILAN R. USKOKOVIC´ HUBERT MAEHR SATYANARAYANA G. REDDY YAN CHUN LI LUCIANO ADORINI MICHAEL F. HOLICK I. II. III. IV.

BioXell, Inc., Nutley, New Jersey BioXell, Inc., Nutley, New Jersey Brown University, Providence, Rhode Island University of Chicago, Chicago, Illinois BioXell, SpA, Milan, Italy Boston University, Boston, Massachusetts

Introduction Synthesis of Gemini 24R-Hydroxy Gemini Metabolite 23-Yne-26,27-Hexafluoro the Gemini Analogs

I. INTRODUCTION In a continuous search for analogs of 1,25-dihydroxy vitamin D3 (1,25(OH)2D3) (1) with improved biological profiles, the chiral center at carbon 20 became one of the focal points in structure-activity studies when it was shown that 20-epi-1,25(OH)2D3 (2) was a thousand times transcriptionally more active than the natural hormone. Subsequent mechanism investigations by Peleg and Norman [1,2], Moras [3], and Carlberg [4] indicated that 2, once bound to the vitamin D receptor (VDR) ligand-binding domain produces the same agonist conformation as 1,25(OH)2D3, but with an increased half-life and transcriptional activity. The transcriptional activities of 1,25(OH)2D3 (ED50 3.0 ± 1.2 × 10−9 M) and its 20-epi analog (10.0 ± 0.5 × 10−12 M) were compared in ROS 17/2.8 cells transfected with a fusion gene containing the osteocalcin vitamin D response elements (VDRE) fused to the thymidine kinase promoter/growth hormone reporter gene. Peleg and co-workers [2] tested the stability of 1 and 2 VDR complexes under non-equilibrium conditions, similar to those developed during prolonged incubation of cell cultures. They found that only 35% of the VDR binding sites remained occupied with 1 three hours after the excess of the ligand was removed from the treated VDR-transfected COS-1 cells. In the case of 2, however, 80% of the VDR binding sites remained occupied three hours after the removal of its excess. This reduced dissociation rate contributes to a longer VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Gemini Analogs as Renin Inhibitors VI. Inhibition of Acute Allograft Rejection by Gemini analogs VII. Rationale for Using Gemini Analogs to Treat Colon Cancer References

half-life of 20-epi-1,25(OH)2D3 – VDR complex and to its increased transcriptional activity (see Fig. 1). Since 1 or its 20-epi analog 2 occupy only 56% of the space in the ligand-binding VDR domain, we were intrigued by the possibility that a compound such as 3 with two side chains emanating at carbon 20 would accommodate one of its side chains according to 1,25(OH)2D3 and the other according to 20-epi analog occupied spaces in the agonist conformation of the ligand-binding VDR domain, thus both contributing to the transcriptional activity.

II. SYNTHESIS OF GEMINI The two-side-chain analog 3 generally referred to as Gemini, was synthesized as outlined in Fig. 2. The protected olefin 6 is easily accessible in high yield from the Inhoffen-Lythgoe diol via 4 and the iodide 5. A double-ene reaction with ethyl propiolate produced the conjugated diester 7. In this two-step process the monoene reaction product is formed rapidly at an early stage of the reaction. Compound 7 was hydrogenated with palladium on carbon catalyst, and the resulting 8 converted to the diztertiary alcohol 9 by a Grignard reaction [5]. The following steps in the synthesis are reminiscent of the protocol used for the synthesis of 1,25(OH)2D3 [6]. Removal of the alcohol protecting group in 9, oxidation, and protection of two tertiary hydroxyls produced the ketone 10. The Lythgoe coupling with the precursor of Copyright © 2005, Elsevier, Inc. All rights reserved.

1512

MILAN R. USKOKOVIC´

HO OH

H

H

OH

H OH

H

OH

HO

OH

HO

1

HO

OH 3

2

FIGURE 1

CO2Et OH

TBSO

EtO2C

I

TBSO

4

TBSO 5

EtO2C

TBSO

6

7

TMSO

HO CO2Et

OTMS

OH

TBSO

O

TBSO 8

10

9 Ph2P O

TBSO

Ph2P O

TBSO

OTBS

OTBS 11

12 HO

HO

OH

OH TBS = t.Bu(Me)2SiTMS = Me3SiHO

OH

HO

OH 3

13

FIGURE 2

ET AL .

1513

CHAPTER 85 Gemini: The 1,25-dihydroxy Vitamin D Analogs with Two Side-Chains

the ring A 11 and deblocking of hydroxyl groups resulted in Gemini 3. Using the 19-nor-ring A precursor 12 instead of 11 gave the 19-nor Gemini 13.

III. 24R-HYDROXY GEMINI METABOLITE Metabolic study of Gemini (3) and 19-nor-Gemini (13) in perfused rat kidney and bone cells indicated formation of a mono-24-hydroxy metabolite in each case, which were stable to further metabolism. Insertion of the 24R-hydroxy group in one of the side chains produces a stereogenic center at C-20, thus one of the two possible 20-epimeric metabolites can be formed in each case. Figure 3 displays the structures of 20-epimeric24R-hydroxy Gemini and 19-nor-Gemini. To determine the C-20 configuration of the 24Rmonohydroxylated metabolites, the four compounds shown in Fig. 3 were needed for a HPLC comparison. They were obtained by synthesis starting from C-20epimeric intermediates 23 and 24 (Fig. 4). Compounds 23 and 24 were prepared starting from previously shown 6, the side chain of which was extended first by an ene reaction to give 18, then by malonate alkylation to form 20. Decarboxylation and the methyl

H

HO

R

Grignard reaction produced the full side chain not bearing the 24R-hydroxy group. Hydroboration and separation completed the formation of 23 and 24 in a 2:3 ratio. The primary alcohol 24 was converted to sulfone 27, which was then alkylated with the known tosylate 28, thus incorporating the required 24R hydroxy group. Reductive removal of the sulfone group and deblocking of hydroxyl groups produced the tetrol 32 representing the complete CD-ring side-chain portion of the target 20(S)-compounds 15 and 17. The completion of their synthesis required temporary protection of the vicinal diol 32, oxidation to 34, replacement of the hydroxy blocking groups as in 36, and the Lythgoe attachment of the A-rings 11 and 12 (Fig. 5). Starting from the 20S-alcohol 23 (Fig. 4) the corresponding 20R-analogs 14 and 16 were prepared (see Fig. 6). Metabolism products of Gemini (3) in the isolated perfused rat kidney were compared with those of 1. The technique of rat kidney perfusion to study the metabolism of various vitamin D3 analogs was previously described in several publications [7–9]. As previously reported, 1 is metabolized into calcitroic acid and more than 50% of substrate 1,25(OH)2D3 had disappeared from the kidney perfusate due to its

OH

H

HO

S

R

OH R

OH

HO

OH

OH

HO

14

OH 15

H

HO

OH

H

HO

OH

OH R

HO

OH

R

S

OH

HO

OH 17

16

FIGURE 3

R

1514

MILAN R. USKOKOVIC´

OH

ET AL .

CO2Me

OTs

CO2Me

TBSO 6

TBSO

18

TBSO

TBSO

19

20 H

HO

CO2Me

TBSO

OH TBSO 21

TBSO

22

S

OH

23 H

HO

R

TBSO

OH

24

FIGURE 4

rapid metabolism through C-24-oxidation pathway. Unlike 1, the Gemini analog resisted its metabolism through C-24 oxidation pathway and only about 20% of the substrate disappeared from the kidney perfusate. Through the analysis of the perfusate for daughter metabolites, we isolated only one polar metabolite. This metabolite was identified by mass-spectrometry and its sensitivity to periodate oxidation as the mono24-hydroxylated metabolite of Gemini (3). Τhus, it appears that Gemini is metabolized in rat kidney at a reduced rate when compared to 1. Also, there appears to be a block to the further metabolism of 24(OH)Gemini metabolite. The studies were repeated using 19-nor Gemini 13. Like the Gemini analog 3, the 19-nor Gemini analog 13 is also metabolized in the rat kidney to a single polar metabolite and this metabolite was identified as the mono 24-hydroxylated metabolite of 19-nor Gemini. The same metabolite was also produced in the rat osteosarcoma cells (UMR 106). As shown in Fig. 7 of the HPLC comparison with the synthetic epimers 16 and 17, the structure and stereochemistry of the 19-nor Gemini metabolite is 20R,24R (16).

IV. THE 23-YNE-26,27-HEXAFLUORO GEMINI ANALOGS To prevent the 24R-hydroxylation of Gemini and increase its genomic activity, the introduction of C23– C24 triple bond and 26,27-hexafluoro substitution were

investigated as shown in Fig. 7. The formaldehyde ene reaction of the olefin 22 (Fig. 4) gave the mixture of the cis and trans homoallylic alcohols 37. Hydrogenation of 37 with Pt-on carbon catalyst gave the clean conversion to the desired 38 and 39 in nearly equal proportion, which were separated quantitatively by chromatography. The extension of the side chain of the epimer 39 led ultimately to the hexafluorohydroxy ketone 44, the structure of which was established as 20(S) by an X-ray analysis. The attachments of the A-rings then produced the desired acetylenic hexafluoro Gemini 45 and its 19-nor analog 46. Using the same pathway as described in Fig. 8 but starting the 20(R)-diol 38, the corresponding 20(R)-hexafluoro analogs 47 and 48 were obtained. The C-20 stereochemistry of the 24R-hydroxy Gemini analogs 14, 15, 16, and 17 rests on the following established chemical connection shown in Fig. 8.

V. GEMINI ANALOGS AS RENIN INHIBITORS As discussed in Chapter 52, 1,25-dihydroxyvitamin D3 is an endocrine inhibitor of the renin-angiotensin system (RAS) in vivo [10]. The RAS plays an essential role in the regulation of blood pressure, volume, and electrolyte homeostasis. Overstimulation of the RAS is among the major causes for the development of hypertension. Therefore, components of the RAS have been important drug targets for the treatment of hypertension [11].

1515

CHAPTER 85 Gemini: The 1,25-dihydroxy Vitamin D Analogs with Two Side-Chains

Ph H

HO

H

RO

R R

SO2

H

RO

OH

SO2Ph OH OH

TBSO

TBSO 26 R = OH 27 R = TMS

24 R = OH 25 R = I

TBSO 29 R = OH 30 R = TMS

OTs

HO 28

OCH3 H

HO

OCH3

OH H

HO

O

H

OH

H

HO

O

O

O

RO

H

HO

O

HO

31 R = TBS 32 R = H

33

34 Ph2P=O

OH

H

Et3SiO

OSiEt3

OH O

+

OSiEt3

OTBS 11

O

35

TBSO

36

+ H

HO

S

OH R

OH

S

Ph2P O

R

OH

TBSO HO

H

HO

OTBS

OH

HO

OH

12

17

15

FIGURE 5

H

HO

S

H

HO

R

OH

OH

H

HO

R

R OH

TBSO

OH

HO

OH 16

14

FIGURE 6

R OH

23

HO

OH

OH

1516

MILAN R. USKOKOVIC´

19-Nor Substrate + Metabolite

ET AL .

19-Nor Substrate + 20R Epimer 16

21.391 0.020

21.374

0.015 AU

AU

0.010 37.589

37.586

0.010

0.005

0.000

0.000 10

20 30 40 Minutes

50

19-Nor Substrate + 20S Epimer 17

10

40

50

19-Nor Substrate + 20R Epimer 16 + 20S Epimer 17 21.339

21.510

0.030

20 30 Minutes

0.020 32.780 0.020 32.992

AU 0.010

AU

37.454

0.010

0.000

0.000 10

20 30 40 Minutes

50

10

19-Nor Substrate + 20R Epimer 16 + Metabolite

30 40 Minutes

50

19-Nor Substrate + 20S Epimer 17 + Metabolite 21.315

21.312 0.020

20

37.330

0.020 AU

AU

32.788 37.418

0.010

0.010

0.000

0.000 10

20 30 Minutes

40

50

10

20 30 40 Minutes

50

FIGURE 7 HPLC profiles of various combinations of mono-24-hydroxylated metabolite of 19-nor Gemini and the two synthetic standards (20R epimer 16 and 20S epimer 17). HPLC analysis was performed using a straight phase HPLC system. Chromatographic conditions: Zorbax Sil column (9.4 mm × 25 cm; 15% 2-propanol in hexane, flow rate 2 ml/min.)

The renin-inhibiting activity of 1 suggests that low calcemic vitamin D analogs may potentially be used as therapeutic agents to control renin production and thus blood pressure. The analogs of 1 were screened for the activity to inhibit renin production, using an in vitro cell culture system, followed by animal testing. The activity was determined by Northern blot and renin promoter luciferase

reporter assays in As4.1 cells stably transfected with human VDR cDNA [12]. Interestingly, of 20 vitamin D analogs that have been screened so far, only gemini compounds display potency equal to or better than 1, whereas the other single side-chain analogs have little inhibitory activity. Why the double-side chain Gemini compounds possess particularly potent inhibitory activity remains to be explored. Table I summarizes the

1517

CHAPTER 85 Gemini: The 1,25-dihydroxy Vitamin D Analogs with Two Side-Chains

H

H

HO

HO

R

OH

OH

OH TBSO

TBSO

22

TBSO

37

38

+ H

HO

H

HO

S

TBSO

TBSO

S

CHO

TBSO

41

H

HO

TBSO

40

HO

CF3 43

H

H

HO

S

S

H

HO

11

12 O

HO

39

CF3 OH

42

CF3 OH CF3

OH

CF3 OH CF3

CF3 OH CF3 44

HO

OH

46

45 H

HO

H

HO

R

R CF3 OH CF3

HO

OH

H

HO CF3 OH CF3

HO

H

HO

CF3 OH CF3

HO

OH 48

OH 47

H

HO H

R

H

HO H

I

TBSO

TBSO 41

25

FIGURE 8

S

1518

MILAN R. USKOKOVIC´

ET AL .

TABLE I Relative Activity of Gemini Compounds to Inhibit Renin Expression in As.4.1 Cells

HO

HO

HO H

OH 3(++)

HO

49(+/−)

HO

OH

H

HO

H

HO

H

R

R OH

15(++)

HO

HO

H

CF3 OH CF3

S OH

HO

H

45(+++)

47(++)

HO

OH

OH

CF3 OH CF3

48(++)

HO

OH

HO R H

CF3 OH CF3

H R

17(--)

HO S

OH

HO

OH

HO R

R

OH

H

OH H

OH

14(+++)

16(++)

OH

HO

OH

HO

HO

OH

R H R

OH

H

OH

HO H

H

13(+++)

F

S

OH

HO

51(+/-)

OH

HO

OH

OH

OH

50(+/−)

HO

HO

HO

H

OH

OH

S

H

CF3 OH CF3

46(+++)

HO

OH

The inhibitory activity was determined by measuring the renin mRNA level with Northern bolt after treating As4.1-hVDR cells with each Gemini compound for 24 hours at 10−8, 10−9, and 10−10 M. The relative activity is based on that of 1,25-dydroxyvitamin D3, which is arbitrarily set at (++).

relative activity of 13 Gemini compounds tested in cell cultures. The in vivo efficacy of the active Gemini compounds was further tested in mice. Of the eight active compounds listed in Table I, the 19-nor Gemini 13 shows the best efficacy to suppress renin expression in vivo. Daily intraperitoneal injection of this compound into normal mice for one week significantly inhibits renin mRNA expression in the kidney. Whether this or

other Gemini analogs can indeed reduce blood pressure in high-renin hypertension requires further investigations. Given the fact that low calcemic vitamin D analogs have been approved for a number of clinical applications [13], Gemini compounds may offer a new source for the long-sought therapeutic renin inhibitors [14,15], which may potentially be developed into another class of anti-hypertensive drugs.

CHAPTER 85 Gemini: The 1,25-dihydroxy Vitamin D Analogs with Two Side-Chains

VI. INHIBITION OF ACUTE ALLOGRAFT REJECTION BY GEMINI ANALOGS Acute allograft rejection is mediated by immunological mechanisms, with dendritic cells (DCs) and T-cells playing a major role. Current immunosuppressive treatments based on small molecules (cyclosporine A, tacrolimus, sirolimus, mycophenolate mofetil) or on biologicals (anti-CD3, anti-CD52, anti-IL2R) target mostly T-lymphocytes rather than antigen presenting cells (APCs). Drugs targeting both cell subsets would therefore represent a useful addition to the available immunosuppressive agents. The immunomodulatory properties of 1 and its analogs suggest the clinical applicability of these hormones also in the treatment of allograft rejection, with the aim of facilitating tolerance induction and preventing chronic graft rejection. VDR ligands have pleiotropic immunoregulatory activities that are able to control allograft rejection. APCs and T-cells can be direct targets of the hormone, leading to the inhibition of pathogenic effector T-cells and enhancing the frequency of T-cells with suppressive properties, largely via induction of tolerogenic DCs. These immunoregulatory activities, coupled with the absence of major side effects once calcemia is under control, have been translated into effective immunointervention in a variety of graft rejection models [16,17]. Induction of tolerance to allografts remains an unfulfilled goal in clinical transplantation, and VDR ligands could be part of a tolerogenic regimen designed to achieve this goal. VDR ligands have indeed gained widespread clinical application, notably in the treatment of secondary hyperparathyroidism and psoriasis, but hypercalcemia is a dose-limiting effect that prevents sustained systemic administration. To overcome this limitation, a number of 1,25(OH)2D3 analogs, with a wider therapeutic window than 1,25(OH)2D3 (1) itself, have been synthetized and shown effective in experimental models of autoimmune diseases and allograft rejection [18].

A. Immunomodulatory Properties of VDR Ligands in the Inhibition of Graft Rejection APCs, and notably DCs, are key targets of VDR ligands, both in vitro and in vivo. 1,25(OH)2D3 (1) and its analogs inhibit the differentiation and maturation of DCs [19], a critical APC in the induction of T-cellmediated immune responses. These studies, performed either on monocyte-derived DCs from human peripheral blood or on bone-marrow derived mouse DCs, have consistenly shown that in vitro treatment of DCs

1519

with 1 and its analogs leads to down-regulated expression of the costimulatory molecules CD40, CD80, CD86 and to decreased IL-12 and enhanced IL-10 production, resulting in decreased T-cell activation. The abrogation of IL-12 production and the strongly enhanced production of IL-10 highlight the important functional effects of 1 and its analogs on DCs and are, at least in part, responsible for the induction of DCs with tolerogenic properties. The prevention of DC differentiation and maturation, as well as the modulation of their activation and survival leading to DCs with tolerogenic phenotype and function, and to T-cell hyporesponsiveness [20], certainly play an important role in the immunoregulatory activity of 1. These effects are not limited to in vitro activity: 1, and its analogs can also induce DCs with tolerogenic properties in vivo, as demonstrated in models of allograft rejection by oral administration directly to the recipient [21] or by adoptive transfer of in vitro-treated DCs [22]. Tolerogenic DCs induced by a short treatment with 1 are probably responsible for the capacity of this hormone to induce CD4+CD25+ regulatory T-cells that are able to mediate transplantation tolerance [21]. 1,25(OH)2D3 (1) in vivo appears primarily to inhibit Th1-cells and, under appropriate conditions, may favor a deviation to the Th2 pathway. These effects could be, in part, a consequence of direct T-cell targeting by 1 and its analogs, but modulation of APC function by VDR ligands certainly plays an important role in shaping the development of T-cell responses. The capacity of VDR ligands to target APCs and T-cells is mediated by VDR expression in both cell types and by the presence of common targets in their signal transduction pathways, such as the nuclear factor NF-kB that is down-regulated in APCs [23] and in T-cells [24]. The immunoregulatory properties of 1 and its analogs have been demonstrated in different models of experimental organ transplantation, both acute and chronic [16]. 1,25(OH)2D3 and its analogs can significantly prolong allograft survival in heart, kidney, liver, pancreatic islets, skin, and small bowel allografts. In general, these effects have been achieved at the maximum tolerated dose, without inducing hypercalcemia, the major side effect of treatment with VDR ligands. In most experimental models, the acute rejection has been further delayed by combining VDR ligands with a suboptimal dose of CsA or other immunosuppressive agents. The induction of tolerogenic DCs by VDR ligands, which leads to an enhanced number of CD4+CD25+ regulatory T cells in vivo [21,25], is likely to play an important role in controlling graft rejection, both acute and chronic, and in favoring the establishment of

1520

MILAN R. USKOKOVIC´

transplantation tolerance. A short treatment with 1 and mycophenolate mofetil, a selective inhibitor of T- and B-cell proliferation that also modulates APCs, induces tolerance to islet allografts associated with an increased frequency of CD4+CD25+ regulatory T-cells able to adoptively transfer transplantation tolerance [21]. Also the direct effects of VDR ligands on T-cells, in particular the inhibition of IL-2 and IFN-γ production, could play a role in inhibiting graft rejection [16]. 1 is believed to inhibit IL-2 secretion by impairing the formation of the transcription factor complex NF-AT and IFN-γ through interaction of the ligand-bound VDR complex with a VDRE in the promoter region of the cytokine. A combination of 1 and low-dose CsA inhibited the expression of IL-2 and IL-12, and increased significantly IL-10 expression levels in kidney allografts. Additional mechanisms could rely on the capacity of 1,25(OH)2D3 to significantly reduce bioactive renal TGF-β1 by interacting with Smad proteins, important regulators of TGF-β signal transduction. Based on the available evidence of a pro-tolerogenic effect and a reduced incidence of chronic rejection, VDR ligands could be added to standard immunosuppressive regimens in the treatment of allograft rejection. Additive and even synergistic effects have been observed between 1 or its analogs and immunosuppressive agents, in particular CsA, tacrolimus, and sirolimus [26]. These effects have been confirmed in models of graft rejection, making VDR ligands potentially interesting as dose-reducing agents for conventional immunosuppressive drugs in clinical transplantation. Another positive feature of adding VDR ligands to standard immunosuppressive regimens is their protective effect on bone loss. A rapid bone loss is usually seen after organ transplantation and is enhanced by some immunosuppressive regimens, in particular those based on tacrolimus and steroids. Administration of 1 has been shown to prevent bone loss in transplanted patients, although standard prophylactic measures may not always be sufficient to prevent loss of bone mass, and 1,25(OH)2D3 analogs with a wider therapeutic window also could serve this function. In addition, the TABLE II

ET AL .

1,25(OH)2D3 analog 22-oxa-1,25(OH)2D3 (OCT) has been shown to exert an anabolic effect on bone reconstruction by vascularized bone allografts in rats [27], indicating a specific advantage of VDR ligand administration in bone transplantation. In addition to avoiding bone loss, the use of VDR ligands to increase transplant survival does not appear to increase opportunistic infections, an important side effect induced by anti-rejection drugs, in particular calcineurin inhibitors and glucocorticoids.

B. Inhibition of Vascularized Heart Allograft Rejection by Treatment with the 19-nor Gemini Analog 13 Heterotopic cardiac transplants in mice were performed between C57BL/6 donors and BALB/c recipient mice as described by Corry [28]. To determine its capacity to inhibit acute allograft rejection in the vascularized heart transplantation model, 19-nor Gemini (13) was administered daily orally at a dose of 10 µg/kg/day starting the day before transplantation until day +1 followed by 3 µg/kg/day from day +2 until +30, while 1 was administered 3 d/week orally at a dose of 5 µg/kg/day starting the day before transplantation until rejection. Vehicle alone (Mygliol 812) was used as control. Administration of the Gemini analog 13 as monotherapy to mice receiving fully MHC-mismatched cardiac allografts resulted in a statistically significant prolongation of graft survival compared to administration of vehicle alone (mean survival 52 ± 29 vs. 9 ± 0.9 days, p = 0.023). Prolongation of graft survival was over three times more sustained following treatment with 19-nor gemini 13 compared to 1 (Table II). This marked inhibition of acute graft rejection was maintained in 2 of the 3 mice tested following treatment withdrawal and was achieved without inducing hypercalcemia after 15 or 20 administrations, whereas 1 induced a slight hypercalcemia. These results show the potent in vivo immunosuppressive effect of Gemini analog 13 in the context of an acute model of

Delayed Acute Rejection Following VDR Ligand Administration

n

Treatment

Dose

Graft survival time

Mean

SD

P value

Ca mg/dl

6 6 3

vehicle 1,25(OH)2D3 (1) 19-nor Gemini (13)

– 5a 10b – 3c

8, 8, 9, 9, 10, 10 12, 12, 14, 15, 16, 17 30, 41, 85

9.00 14.33 52.00

0.9 2.1 29.0

– 0.0022 0.023

8.8 10, 9 (15×) 10 (15×); 10.5 (20×)

per os (3×/w) every other day from d − 1 until d + 30. po (d – 1,0, +1). cµg/kg po (d + 2, until d + 30). aTreatment bµg/kg

CHAPTER 85 Gemini: The 1,25-dihydroxy Vitamin D Analogs with Two Side-Chains

HO

OH

HO

OH 13

allotransplantation, and demonstrate that its efficacy is superior to that of 1.

VII. RATIONALE FOR USING GEMINI ANALOGS TO TREAT COLON CANCER It is recognized that 1,25(OH)2D3 (1) is a potent inhibitor of cellular proliferation and inducer of maturation of a wide variety of cultured cancer cells, including breast cancer, colon cancer, prostate cancer, melanoma, lung cancer, and leukemia. Tanaka et al. [29] were the first to demonstrate that M-1 leukemic cells and HL-60 cells, which had VDR, were responsive to the antiproliferative activity of 1 and its analog 1α-hydroxyvitamin D3 (1α-OH-D3). They demonstrated in vivo that 1α-OH-D3 was effective in treating M-1 leukemia in mice by prolonging the lives of mice that received the analog. Koeffler et al. [30] initiated human clinical trials with 1 in preleukemia patients and found that although there was an initial response, the patients often developed hypercalcemia and ultimately succumbed to their disease. It has been thought that cancer cells developed a resistance to 1 by either decreasing VDR, altering the 1,25(OH)2D3-VDR-RXR interaction with transcriptional factors on the VDRE, and/or enhancing its catabolism by stimulating the 25-hydroxyvitamin D-24-hydroxylase [31]. The effectiveness of the antiproliferative activity of 1 and its analogs has been amply demonstrated for the treatment of the hyperproliferative skin disorder psoriasis [32]. Numerous analogs of 1 have been synthesized and evaluated for their antiproliferative and maturation activity, as well as their calcemic activity [33]. The goal for developing a potent vitamin D analog to treat some common cancers was to develop an analog that had marked antiproliferative activity without having any significant calcemic effects. Four studies have investigated

1521

the in vivo response of vitamin D analogs in prostate cancer. The analogs were either slightly more potent or equally potent to 1,25(OH)2D3 (1) and appeared to have less calcemic action than 1. In a Phase I trial, patients with advanced hormone refractory prostate cancer received 1α-hydroxyvitamin D2 [34]. Twenty-five percent of the patients achieved stabilization of their disease for six months with the main toxicity being hypercalcemia and renal insufficiency. Another strategy is to use 1 in combination with chemotherapy. The combination of a weekly oral high dose (0.5 µg/kg) of 1 and weekly docetaxel was shown to be well tolerated and effective in achieving a significant decrease in prostatic specific antigen in 30 of 37 metastatic androgen-independent prostate cancer patients [35].

A. Antiproliferative Activity of 1,25(OH)2D3 and Its Analogs on Cultured Human and Rodent Colon Cancer Cells Zhao and Feldman [36] demonstrated that 1,25(OH)2D3 (1) inhibited the proliferation of HT-29 human colon cancer cells in culture and induced them to terminally differentiate. This effect appeared to be directly related to the abundance of the VDR in the cells. Tananka et al. [37] reported in severe combined immunodeficient mice that a 1,25(OH)2D3 analog, 22(S)-24-homohexafluoro-1,22,25-trihydroxyvitamin D3, was effective in decreasing the invasion of human HT-29 colon cancer cells in the renal capsule, compared to placebo controlled animals. Serum calcium concentrations and body weights of the treated mice were similar in both groups demonstrating that the vitamin D analog most likely had a direct effect on inhibiting colon cancer cell growth. We have grown both human and a mouse colon cancer cell lines in culture and found that 1 inhibited 3H-thymidine incorporation in both Caco-2 and MC-26 cells (Fig. 9A, B). Cellular extracts of the MC-26 cells were obtained, and the RNA was extracted. Quantitative RT-PCR analyses revealed the presence of the VDR. We have established an in vivo mouse colon cancer model by injecting 10,000 MC-26 cells subcutaneously in the dorsal posterior back of Balb-c mice that were six to eight weeks of age. Two days after tumor implantation, the mice were divided into three groups: control vehicle, 1,25(OH)2D3 (0.1 µg/kg), 1,25-dihydroxy-(20S)-21-(3-methyl-3-hydroxy-4butyl)-23-yne-26,27-hexafluoro-19-nor-vitamin D3 (46) 1.0 µg/kg. The mice were weighed three times a week and their tumors measured every day (see Fig. 9). After 33 days, there was a marked increase in tumor volume in the mice that received placebo compared to

1522

MILAN R. USKOKOVIC´

A

ET AL .

B 100

100

*

[3H] Thymidine incorporation (% of control)

[3H] Thymidine incorporation (% of control)

* 80 60 40 20 0

10−9

80 * 60

**

40 20 0

10−8 10−7 1,25(OH)2D3 (M)

10−8

10−7 1,25(OH)2D3 (M)

10−6

FIGURE 9 (A) Human colon cancer. Effect of 1,25(OH)2D3 on 3H-thymidine incorporation in a human (Caco-2). Results are means ± SEM *p < 0.005. (B) MC-26 mouse colon cancer. Effect of 1,25(OH)2D3 on the 3H-thymidine incorporation in a mouse cancer cell line (MC-26). Results are means ± SEM. *p = 0.0003 **p < 0.0001.

the mice that received either 1 or 46. After sacrificing the animals, the extent of tumor invasion was evaluated. 50% of the mice receiving placebo and 33% of the mice receiving 1,25(OH)2D3 displayed muscle invasion by the tumor, whereas the group that received Gemini analog 46 showed no tumor extension into the adjacent musculature. At the end of the experiment, on day 33, the average tumor volume of the mice treated with placebo vehicle, 1 and 46 were 1,385 mm3, 692 mm3, and 352 mm3, respectively (see Fig. 10). Thus, the Gemini analog 46 reduced tumor volume by more than 75%. None of the mice that received the 46 analog died, whereas 35 and 30% of the mice receiving vehicle and 1 died at the end of the 33 days of the study (see Fig. 11). An evaluation of the serum calciums revealed a significant elevation in the animals treated with 1, whereas the Gemini analog treated animals showed no evidence of hypercalcemia (Spina,

Tangpricha, and Holick, unpublished results) or body weight loss. Therefore, these data suggest that the Gemini vitamin D analog 46 not only markedly decreased metastatic colon cancer activity, but it also prolonged life and caused no untoward consequences on the calcium metabolism in the mice. This study provides a promising new approach for treating colon cancer with this novel vitamin D analog. In summary, a new class of 1,25-dihydroxy vitamin D receptor ligands, the two side-chain Gemini analogs are being investigated as the potential drug candidates for treatment of bone diseases, hypertension, acute allograft rejection, and colon cancer. Promising results have been obtained in the meaningful in vivo models for these indications. The methods for the synthesis of Gemini have been illustrated that allow their largescale preparation. Placebo

Average Tumor Volume

H 2500 H

OH

2000

mm3

1500 OH 1,25(OH )D 2 3

HO 1000

HO H S

500

H

OH F3C CF 3

0 0

10

20

days HO

FIGURE 10

OH

46

1523

CHAPTER 85 Gemini: The 1,25-dihydroxy Vitamin D Analogs with Two Side-Chains

HO H S H

Survival of Mice 100

OH F3C CF3

90 Percent

HO

OH

46 H

80

H 70

OH

60 HO

50 0

10

20

30

OH

Days

1,25(OH2)D3

Placebo

FIGURE 11

TABLE III Gemini (46) A1:1 A1:2 A1:3 A1:4 A2:1 A2:2 Average

8.4 7.5 8.4 9.2 6.8 8.2 8.1

Calcium Levels (mg/dL) 1,25(OH)2D3 B1:1 B1:2 B1:3 B2:1 B2:2 Average

7.2 12 11.4 7.8 9.6 9.6

Control C1:1 C1:2 C1:3 C2:1 C2:2 Average

8.4 9.6 9 8.8 10.8 9.32

References 1. Peleg S, Sastry M, Collins ED, Bishop JE, Norman AW 1995 Distinct conformational changes induced by 20-epi analogs of 1α,25-dihydroxyvitamin D3 are associated with enhanced activation of the vitamin D receptor. J Biol Chem 270:10551–10558. 2. Liu Y-Y, Collins ED, Norman AW, Peleg S 1997 Differential interaction of 1α,25-dihydroxyvitamin D3 analogs and their 20-epi homologs with the vitamin D receptor. J Biol Chem 272:3336–3345. 3. Tocchini-Valentini G, Rochel N, Wurtz JM, Mitschler A, Moras D 2001 Crystal structures of the vitamin D receptor complexed to superagonist 20-epi ligands. PNAS 98:5491–5496. 4. Vaisanen S, Perakyla M, Karkkainen JI, Uskokovic MR, Carlberg C 2003 Structural evaluation of the agonistic action of a vitamin D analog with two side chains binding to the nuclear vitamin D receptor. Mol Pharm 63:1–8. 5. Norman AW, Manchand PS, Uskokovic MR, Okamura WH, Takeuchi JA, Bishop JE, Hisatake J-I, Koeffler HP, Peleg S 2000 Characterization of a novel analog of 1α,25(OH)2-vitamin D3 with two side chains: interaction with its nuclear receptor and cellular actions. J Med Chem 43:2719–2730.

6. Baggiolini EG, Iacobelli JA, Hennessy BM, Batcho AD, Sereno JF, Uskokovic MR 1986 Stereocontrolled total synthesis of 1α,25-dihydroxycholecalciferol and 1α,25-dihydroxyergocalciferol. J Org Chem 51:3098–3108. 7. Reddy GS, Tserng K-Y 1986 Isolation and identification of 1,24,25-trihydroxyvitamin D2, 1,24,25,28-tetrahydroxyvitamin D2, 1,24,25,26-tetrahydroxy-vitamin D2: New metabolites of 1,25 dihydroxyvitamin D2 produced in rat kidney. Biochemistry 25:5328–5336. 8. Reddy GS, Tserng K-Y, Thomas BR, Dayal R, Norman AW 1987 Isolation and identification of 1,23-dihydroxy-24,25,26, 27-tetranorvitamin D3. Biochemistry 26:324–331. 9. Reddy GS, Tserng K-Y 1989 Calcitroic acid, end product of renal metabolism of 1,25-dihydroxyvitamin D3 through C-24 oxidation pathway. Biochemistry 28:1763–1769. 10. Li YC 2004 The renin-angiotensin system. In: Feldman D, Glorieu F, Pike JW (eds) Vitamin D, Second Edition. Elsevier: San Diego. 11. August P 2003 Initial treatment of hypertension. N Engl J Med 348:610–617. 12. Li YC, Kong J, Wei M, Chen ZF, Liu SQ, Cao LP 2002 1,25Dihydroxyvitamin D3 is a negative endocrine regulator of the renin-angiotensin system. J Clin Invest 110:229–238. 13. Malluche HH, Mawad H, Koszewski NJ 2002 Update on vitamin D and its newer analogs: actions and rationale for treatment in chronic renal failure. Kidney Int 62:367–374. 14. Kokubu T, Ueda E, Fujimoto S, Hiwada K, Kato A 1968 Peptide inhibitors of the renin-angiotensin system. Nature 217:456–457. 15. Burton J, Cody RJ, Jr., Herd JA, Haber E 1980 Specific inhibition of renin by an angiotensinogen analog: studies in sodium depletion and renin-dependent hypertension. Proc Natl Acad Sci USA 77:5476–5479. 16. Adorini L 2002 1,25-Dihydroxyvitamin D3 analogs as potential therapies in transplantation. Curr Opin Investig Drugs 3:1458–1463. 17. Becker BN, Hullett DA, O’Herrin JK, Malin G, Sollinger HW, DeLuca H 2002 Vitamin D as immunomodulatory therapy for kidney transplantation. Transplantation 74:1204–1206.

1524 18. Mathieu C, Adorini L 2002 The coming of age of 1,25-dihydroxyvitamin D3 analogs as immunomodulatory agents. Trends Mol Med 8:174–179. 19. Griffin MD, Xing N, Kumar R 2003 Vitamin D and its analogs as regulators of immune activation and antigen presentation. Annu Rev Nutr 20. Penna G, Adorini L 2000 1,25-dihydroxyvitamin D3 inhibits differentiation, maturation, activation, and survival of dendritic cells leading to impaired alloreactive T-cell activation. J Immunol 164:2405–2411. 21. Gregori S, Casorati M, Amuchastegui S, Smirolodo S, Davalli AM, Adorini L 2001 Regulatory T-cells induced by 1α,25-dihydroxyvitamin D3 and mycophenolate mofetil treatment mediate transplantation tolerance 167:1945–1953. 22. Griffin MD, Lutz W, Phan VA, Bachman LA, McKean DJ, Kumar R 2001 Dendritic cell modulation by 1 alpha,25 dihydroxyvitamin D3 and its analogs: A vitamin D receptor-dependent pathway that promotes a persistent state of immaturity in vitro and in vivo. Proc Natl Acad Sci USA 98:6800–6805. 23. D’Ambrosio D, Cippitelli M, Cocciolo MG, Mazzeo D, Di Lucia P, Lang R, Sinigaglia F, Panina-Bordignon P 1998 Inhibition of IL-12 production by 1,25-dihydroxyvitamin D3. Involvement of NF-kappaB down-regulation in transcriptional repression of the p40 gene. J Clin Invest 101:252–262. 24. Barrat FJ, Cua DJ, Boonstra A, Richards DF, Crain C, Savclkoul HF, de Waal-Malefyt R, Coffman RL, Hawrylowicz CM, O’Garra A 2002 In vitro generation of interleukin 10-producing regulatory CD4(+) T cells is induced by immunosuppressive drugs and inhibited by T-helper type 1 (Th1)- and Th2-inducing cytokines. J Exp Med 195:603–616. 25. Gregori G, Giarratana N, Smiroldo S, Uskokovic M, Adorini L 2002 A 1α,25-Dihydroxyvitamin D3 analog enhances regulatory T-cells and arrests autoimmune diabetes in NOD mice. Diabetes 51:1376–1374. 26. van Etten E, Branisteanu DD, Verstuyf A, Waer M, Bouillon R, Mathieu C 2000 Analogs of 1,25-dihydroxyvitamin D3 as

MILAN R. USKOKOVIC´

27.

28. 29.

30.

31. 32. 33. 34. 35. 36. 37.

ET AL .

dose-reducing agents for classical immunosuppressants. Transplantation 69:1932–1942. Merida L, Shigetomi M, Ihara K, Tsubone T, Ikeda K, Yamaguchi A, Sugiyama T, Kawai S 2002 Effects of vitamin D analog, 22-oxa-1,25-dihydroxyvitamin D3, on bone reconstruction by vascularized bone allograft 30:422–427. Corry RJ, Winn HJ, Russell PS 1973 Primarily vascularized allografts of hearts in mice. Transplantation 16:343–350. Tanaka H, Abe E, Miyaura C, et al. 1982 1,25-dihydroxycholeciferol and human myeloid leukemia cell line (HL-60): The presence of cytosol receptor and induction of differentiation. Biochem J 204:713–719. Koeffler HP, Hirjik J, Iti L, the Southern California Leukemia Group. 1985 1,25-dihydroxyvitamin D3: in vivo and in vitro effects on human preleukemic and leukemic cells. Cancer Treat Rep 69:1399–1407. Holick MF 2003 Vitamin D: A millennium perspective. J Cell Biochem 88:296–307. Holick MF 1998 Clinical efficacy of 1,25-dihydroxyvitamin D3 and its analogs in the treatment of psoriasis. Retinoids 14:12–17. Bouillon R, Okamura WH, Norman AW 1995 Structurefunction relationships in the vitamin D endocrine system. Endocrine Rev 16:200–257. Liu G, et al. 2002 Phase I trial of 1α-hydroxyvitamin D2 in patients with hormone refractory prostate cancer. Clin Cancer Res 8:2820–2827. Beer TM, et al. 2003 Weekly high-dose calcitriol and docetaxel in metastatic androgen-independent prostate cancer. J Clin Oncol 21:123–128. Zhao XY, Feldman D 2001 The role of vitamin D in prostate cancer. Steroids 66:293–300. Tanaka Y, Wu AYS, Ikekawa N, Iseki K, Kawai M, Kobayashi Y 1994 Inhibition of HT-29 human colon cancer growth under the renal capsule of severe combined immunodeficient mice by an analog of 1,25-dihydroxyvitamin D3, DD-003. Cancer Res 54:5148–5153.

CHAPTER 86

Development of OCT and ED-71 NOBORU KUBODERA

Chugai Pharmaceutical Co., Ltd., Tokyo, Japan

I. Introduction to OCT and ED-71 II. Development of OCT for Secondary Hyperparathyroidism and Psoriasis Vulgaris

I. INTRODUCTION TO OCT AND ED-71 The active vitamin D3, 1α,25-dihydroxyvitamin D3 [calcitriol, 1,25(OH)2D3], is now well-recognized as a potent regulator of cell proliferation and differentiation processes in addition to possessing regulatory effects on calcium and phosphorous metabolism [1]. Various analogs of 1,25(OH)2D3 have been synthesized to separate differentiation-induction and antiproliferation activities from calcemic activity with the aim of obtaining useful analogs for the medical treatment of psoriasis vulgaris, cancer, etc., without manifestation of hypercalcemia [2]. 1α,25-Dihydroxy-22-oxavitamin D3 (maxacalcitol, 22-oxacalcitriol, OCT) was obtained from such modification studies of the 1,25(OH)2D3 side chain [3–11] (Fig. 1) and OCT has been shown to be highly effective in stimulating monocytic differentiation of human promyelocytic leukemic HL-60 cells, although OCT is less calcemic than 1,25(OH)2D3 [12]. It is well known that 1,25(OH)2D3 binds to the circulating vitamin D transport carrier, vitamin D–binding protein (DBP), whereby it is carried to various tissues that express vitamin D receptors (VDR) and taken up into the cell nucleus to exert its effects [13]. The basic activity of OCT as a vitamin D3 analog is largely characterized by its weaker affinity for DBP and VDR, which is approximately 1/580 and 1/8, respectively, than that of 1,25(OH)2D3 [14–15]. OCT in blood also has been shown to bind to low density lipopropteins (LDL) as well as to DBP, so that it is incorporated into cells or the nucleus via VDR or LDL receptors of the target organs [16]. There is also an intense interest in obtaining analogs more potent than 1,25(OH)2D3 or 1α-hydroxyvitamin D3 (alfacalcidol, 1αOHD3), a clinically important prodrug of 1,25(OH)2D3, in terms of regulatory effects on calcium and phosphorous metabolism, with the aim of treating bone diseases such as osteoporosis. 1α,25-Dihydroxy-2β-(3-hydroxypropoxy)vitamin D3 (2β-(3-hydroxypropoxy)-calcitriol, ED-71) (Fig. 1) was obtained in our modification studies of the A-ring VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

III. Development of ED-71 for Osteoporosis References

of 1,25(OH)2D3 [17–21]. ED-71 is characterized by (i) a hydroxypropoxy substituent at the 2β-position; (ii) a relatively long plasma half-life arising from its strong affinity for DBP which is approximately two times more potent than 1,25(OH)2D3 [22]; (iii) its potential therapeutic effects on bone [23]. In the previous edition of this book, we described the primary characteristics of OCT and ED-71, which were originally recognized during the course of our exploratory research of these analogs [24]. Since then, OCT has been launched in Japan in 2000 as an injection for the treatment of secondary hyperparathyroidism associated with renal insufficiency and, in 2001, as an ointment for treating the skin disease, psoriasis vulgaris. Clinical trials of ED-71 as a promising candidate for the treatment of osteoporosis have been conducted also in Japan. This chapter describes updated information concerning the development of OCT for secondary hyperparathyroidism and psoriasis vulgaris and ED-71 for osteoporosis. A practical synthesis of OCT for industrial scale production along with a convergent, highly versatile synthesis of ED-71 are also discussed below.

II. DEVELOPMENT OF OCT FOR SECONDARY HYPERPARATHYROIDISM AND PSORIASIS VULGARIS A. Practical Synthesis of OCT for Large Scale Production During the development of OCT, a practical synthesis was needed for industrial scale production. An important objective in the synthesis of OCT was the facile introduction of the side chain, which is characterized by the γ-hydroxy ether linkage. Typically, this can be accomplished by the Williamson ether synthesis which involves the reaction between alkyl halides and alkoxides. While this is a well-known and general method, yields of such reactions are not always Copyright © 2005, Elsevier, Inc. All rights reserved.

1526

NOBORU KUBODERA

O OH Differentiation OCT Secondary hyperparathyroidism

OH

Psoriasis vulgaris

OH

HO

OH

OH

HO

Ca regulation

ED-71

1,25(OH)2D3

Osteoporosis OH

HO

OH

O

FIGURE 1 Chemical structure and therapeutic indications of OCT and ED-71.

the 20(S)-alcohol was accomplished via a conjugate addition reaction of N,N-dimethylacrylamide as the Michael acceptor in the presence of sodium hydride. Subsequent nucleophilic addition to the amide moiety with methyl magnesium chloride in the presence of cerium chloride gave the γ-hydroxy ether side chain of OCT in satisfactory yield. Desilylation, irradiation, and thermal isomerization produced OCT (Fig. 2) [25]. An improved one-pot synthesis of the γ-hydroxy side chain unit of OCT has been elucidated for industrial scale production.

satisfactory, particularly in sterically-hindered steroidal cases. Therefore, we investigated the practical introduction of the γ-hydroxy ether moiety of OCT by employing a two-step Michael and nucleophilic addition sequence. In the total synthesis of OCT, we adopted an 11-step sequence as shown in Fig. 2. Dehydroepiandrosterone (DEA) was used as the starting material, which was converted to the 20(S)-alcohol via microbiological 1α-hydroxylation, silylation, 5,7-diene formation, Wittig olefination, and hydroboration. Alkylation of

O

O

O

Microbiological 1α-hydroxylation

HO

tBuMe SiCl 2

2) γ-collidine 65%

94% TBSO

HO

HO

1) NBS

TBSO

DEA OH

O

NMe2 Ph2PEtBr

TBSO

TBSO

9-BBN H2O2/NaOH 84% TBSO

64% TBSO

TBSO O TBSO

NMe2 O

MeCeCl2 TBSO 78%

TBSO

TBSO

TBSO

O NaH 80%

20(S)-Alcohol

O

O OH nBu NF 2

HO

85% HO

FIGURE 2 Practical synthesis of OCT for large scale production.

OH 1)hν 2) ∆ 17%

OCT

CHAPTER 86 Development of OCT and ED-71

1527

B. Secondary Hyperparathyroidism

also demonstrated. These results indicate that OCT suppresses PTH secretion, not only in normal parathyroid cells but also in hyperplastic parathyroid cells in much the same manner as 1,25(OH)2D3. An early study to examine the effect on parathyroid PTH gene expression in normal rats showed that OCT reduced parathyroid pre-pro PTH mRNA expression in 48 hours post-injection as also did 1,25(OH)2D3 [28]. When assessed in a 5/6-nephrectomized rat model of chronic renal failure, intermittent administration of OCT lowered the circulating PTH level without causing an elevation in serum calcium concentration [29–31] (Fig. 4). Northern blot analysis of parathyroid pre-pro PTH mRNA expression in these same rats revealed significant inhibition of pre-pro PTH mRNA expression by OCT as well as by 1,25(OH)2D3. OCT was thus considered to suppress PTH synthesis and secretion via inhibition of PTH gene transcription as seen in the case of 1,25(OH)2D3 [29] (Fig. 5). In the dog model of chronic renal failure, OCT treatment produced a significant decrease in levels of circulating serum PTH at week 5 of treatment, and the PTH level remained reduced by 80% consistently thereafter throughout the treatment period. Serum calcium concentrations rose significantly at week 13 of OCT administration compared to pretreatment levels, although this calcemic effect was inconspicuous [32]. For effects on calcium levels in the intestine, which is of general clinical concern in the use of vitamin D analogs, a comparative assessment of OCT was carried out in comparison to 1,25(OH)2D3 in vitamin D– deficient rats. Calcium binding protein (Calbindin) D9K gene expression was increased with both OCT and 1,25(OH)2D3 until 16 hours after dosing but disappeared rapidly in the case of OCT, unlike 1,25(OH)2D3 [33]. When assessed using the inverted gut sac technique, 1,25(OH)2D3 proved to induce an active intestinal calcium transport that was characterized by a first phase sharp increase from 1 to 6 hours after administration and a second phase continuous rise from 12 hours post-dose onwards. Unlike the response to 1,25(OH)2D3, the active intestinal calcium transport induced by OCT was practically nonexistent in the second phase. Increase in calcium absorption from the intestine associated with the use of OCT is thus considered less lasting as compared to 1,25(OH)2D3 [33]. b. Effects on Bone The slowly progressive renal failure rat has been used as an established animal model that is created through a single injection of glycopeptide isolated from rat renal cortical tissue [34]. In this model, OCT produced a significant decrease in serum PTH level without causing elevation of serum calcium. A histopathological evaluation of the effects of OCT on parathyroid and bone tissues in these

1. PRECLINICAL RESULTS

a. Effects on Parathyroid Hormone The inhibitory effect of 1,25(OH)2D3 on parathyroid hormone (PTH) synthesis and secretion has been clearly elucidated (see Chapter 30). It has been reported that 1,25(OH)2D3 blocks transcription via the negative vitamin D responsive region in the control region of the PTH gene [26] (see Chapter 30). OCT is also thought to exert its inhibitory action on PTH synthesis through the same mechanism as 1,25(OH)2D3. An in vitro study comparing the effects of OCT to 1,25(OH)2D3 was performed using human hyperplastic parathyroid tissue (obtained during surgery for advanced renal hyperparathyroidism) and normal bovine parathyroid glands. OCT suppressed PTH secretion in both nodular hyperplastic parathyroid tissue and normal bovine tissue in a dose-dependent manner, analogs to 1,25(OH)2D3 [27] (Fig. 3). The additive effect of cellular calcium levels on suppression of PTH secretion by OCT and 1,25(OH)2D3 was

: OCT

125

: 1,25(OH)2D3 Percentage of Control (%)

100 * *

*

**

**

75

**

** **

50

25

0

cont

−12

−11

−10

−9

−8

Conc. (logM)

FIGURE 3 Effect of OCT or 1,25(OH)2D3 on PTH secretion in primary cultures of parathyroid gland cells from patients with secondary hyperparathyroidism associated with chronic renal failure. Primary cultures of cells (calcium concentration in medium: 2 nmol/L) were incubated with added OCT or 1,25(OH)2D3 (10−12 to 10−8 mol/L). After 48 hours of incubation the culture supernatant was removed and fresh medium added, followed by re-incubation for 2 hours. The amount of PTH secreted in 2 hours was determined by RIA (HS-PTH), and shown as percent of drug-free control. Data represent the mean ±S.D. (same hereunder; n = 22). Dunnett’s test: control group vs. each treated group *p < 0.01; **p < 0.001;
: OCT; : 1,25(OH)2D3.

1528

NOBORU KUBODERA

A

B ** **

**

# **

250

**

15 *

** *

Serum Ca (pg/ml)

Serum PTH (pg/ml)

200

150

100

10

5

50

0 Sham vehicle

1,25(OH)2D3 OCT 0.25 1.25 6.25 0.025 0.125 0.625 (µg/kg)

0 Sham vehicle

1,25(OH)2D3 OCT 0.25 1.25 6.25 0.025 0.125 0.625 (µg/kg)

FIGURE 4 Effects of intermittent intravenous administration of OCT on serum PTH and calcium (Ca) levels in renal failure rats. 5/6-Nephrectomized rats were injected i.v. with OCT or 1,25(OH)2D3 at various dose levels three times weekly for 15 days. Grouped t-test: sham-operated group vs. diseased control #p < 0.05. Dunnett’s test: diseased control vs. each treated group *p < 0.05; **p < 0.01.

** **

# **

8 Ratio (PTH mRNA/β-Actin mRNA)

*

6

4

2

0

OCT

1,25(OH)2D3

Sham vehicle 0.25 1.25 6.25 0.025 0.125 0.625 (µg/kg)

FIGURE 5 Effects of OCT and 1,25(OH)2D3 on pre-pro PTH mRNA expression in renal failure rats. 5/6-Nephrectomized rats were injected i.v. with OCT or 1,25(OH)2D3 at various dose levels three times weekly for 15 days. Parathyroids were excised 24 hours after the last dose and analyzed by Northern blot technique for prepro PTH mRNA expression, which is presented as its quotient divided by β-actin mRNA expression. Grouped t-test: sham-operated group vs. diseased control #p < 0.05. Dunnett’s test: diseased control vs. each treated group *p < 0.05; **p < 0.01.

animals disclosed parathyroid hyperplasia with chief cell proliferation in virtually all rats of a diseased control group. In contrast, in the OCT-treated groups, hyperplasia was significantly reduced. Treatment with OCT reduced the incidence of ostitis fibrosa of the tibia and lumbar vertebrae in the OCT-treated groups (37.5–57.1%), compared to the diseased control group (66.7%) [35]. A further evaluation of the effects of OCT with respect to bone formation, bone resorption, and relevant dynamic parameters was performed using the third lumbar vertebra. The data revealed significant increases in bone formation rate, mineral apposition rate, mineralized surface and tetracycline double-labeled area, increases in number and area of osteoclasts, and a significant increase in bone resorption rate in the diseased control group. The significance of these histomorphomery parameters is discussed in Chapter 50. In OCT-treated groups, enhanced bone formation rate, mineral apposition rate, and mineralized surface areas and bone formation in tetracycline double-labeled areas were significantly reduced. There was a significant decrease in the increased osteoclast number and area, indicating a potential efficacy of OCT for highturnover bones. These effects suggest that OCT might ameliorate osseous lesions in secondary hyperparathyroidism [35] (Table I).

1529

CHAPTER 86 Development of OCT and ED-71

TABLE I Bone Histomorphometry in the Lumbar Vertebra at the End of the OCT Treatment GN Parameters Bone formation parameters MS/BS % BFS/BS µm3/µm2/year MAR mm/day Bone resorption parameters ES/BS % Rs.R µm3/µm2/year N.Oc/BS/mm

Control

Vehicle

OCT (0.03µg/kg)

OCT (0.15µg/kg)

16.6 + 0.6 39.8 + 2.3 0.66 + 0.03

27.6 + 3.2 116.3 + 14.0a 1.14 + 0.8a

18.0 + 2.8 67.6 + 13.9b 1.00 + 0.05

13.4 + 2.1b 42.2 + 7.9b 0.83 + 0.07b

21.4 + 0.5 0.61 + 0.00 0.65 + 0.12

33.0 + 2.1a 1.67 + 0.29a 1.36 + 0.24

28.2 + 2.2 1.06 + 0.34 0.70 + 0.08

23.5 + 1.5b 0.76 + 0.19 0.67 + 0.2b

Abbreviations are: MS/BS, mineralizing surface; BFS/BS, bone formation rate; MAR, mineral apposition rate; ES/BS, eroded surface; Rs.R, bone resorption rate; N.Oc/BS/mm, osteoclast number. ap < 0.05 vs. Control group. bp < 0.05 vs. Vehicle group.

The effect of OCT on bone lesions was investigated in a dog model with chronic renal failure to clarify the effect of long-term administration of OCT on bone metabolism. Groups of renal failure dogs received either vehicle alone or OCT for one year and were examined for effects on bone changes in secondary hyperparathyroidism. Pre- and post-treatment bone biopsies were performed, and the results showed significant increases in abnormal fibrous osteoid area (Wo.OS/BS, %) and fibrous tissue area (Fb.S/BS, %) in the bones of vehicle-treated dogs with progression to secondary hyperparathyroidism. Trabecular unit activation frequency (Ac.f) and circulating PTH were also significantly increased. In dogs given OCT, in contrast, Wo.OS/BS(%) was found significantly reduced as compared to the diseased control group; hence, a delay in progression of bone changes. There was no significant intergroup difference in Ac.f changes. These studies suggest, therefore, that OCT might retard the progression of bone changes in secondary hyperparathyroidism without inducing low bone turnover [36]. An ensuing, continued eight-month cross-over study demonstrated a significant decrease in Wo.OS/BS(%) in the OCT-treated dogs as compared to pre-treatment values in dogs assigned to the diseased control group whose initial Wo.OS/BS(%) and Fb.S/BS(%) values were high. Again, there was no conspicuous difference between pre- and post-treatment Ac.f values. In the animals that had initially been treated with OCT, serum PTH increased following the discontinuation of OCT administration, but this was not accompanied by any appreciable change in bone lesions; namely, the

corrective effect of OCT on bones lasted even after the withdrawal of OCT [36]. The efficacy of OCT for improving bone tissues in clinical settings was presumed from the demonstrated ameliorative effects of OCT on bone changes in secondary hyperparathyroidism and also in bone remodeling animals. Furthermore, the marked improvement of ostitis fibrosa parameters and the very modest associated changes in bone turnover parameters indicated the possibility that OCT may become a therapeutic agent with merit, having a lower risk of low turnover bones which is of clinical concern for the use of 1,25(OH)2D3. 2. CLINICAL RESULTS WITH SECONDARY HYPERPARATHYROIDISM PATIENTS

Results of the clinical studies conducted in Japan from 1992 to 1997 have raised the expectation that intravenous OCT administration may be an effective treatment for secondary hyperparathyroidism. Intravenous OCT therapy was shown to afford marked clinical efficacy even for severe secondary hyperparathyroidism through a mechanism whereby intravenously administered OCT directly inhibits PTH synthesis and secretion. The drug also can be administered to patients whose serum calcium is somewhat elevated prior to treatment, resulting in a satisfactory therapeutic response (see Chapter 76 for discussion of the use of vitamin D compounds in renal failure). Results of clinical studies in patients with secondary hyperparathyroidism undergoing maintenance hemodialysis showed that serum calcium values increased following administration of OCT in patients receiving dialysis for renal failure. However, this increase was

1530

NOBORU KUBODERA intact-PTH

HS-PTH

1,000

800

*

n = 160−113 *p < 0.05 ***p < 0.001 vs 0 week

n = 160−128 ***p < 0.001 vs 0 week 60,000

***

*** ***

700

****** *** ***

*** *** *** ***

600

***

*** ***

HS-PTH (pg/ml)

intact-PTH (pg/ml)

900

70,000

*** 50,000

***

*** ***

*** ***

***

40,000

500 *** 400

30,000 −2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 +2

−2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 +2

Weeks

Weeks

FIGURE 6 Serum intact-PTH and HS-PTH levels over time following long-term OCT administration.

remarkably modest in contrast to elevations seen with oral vitamin D analog pulse therapies currently under clinical trials. With OCT, a dose adjustment was readily accomplished and significant suppression of PTH in severe secondary hyperparathyroidism was attained. A remarkable level of PTH suppression was seen even in severe cases with intact-PTH levels as high as 1,000 to 3,000 pg/mL. In a phase III long-term treatment study conducted with 160 chronic hemodialysis patients (97 male and 63 female; age range, 27–78 years; mean past duration of dialysis therapy, 155.3 months) having secondary hyperparathyroidism (intact-PTH level, 156-5,290 pg/mL, mean: 811.4 pg/mL; and corrected serum calcium level, 9.0–11.0 mg/dL, mean: 9.96 mg/dL), intravenous OCT injection produced a rapid decrease in serum PTH. PTH suppression was sustained over six months of treatment (Fig. 6). There were slight, yet significant, elevations in serum calcium and plasma ionized calcium values early in the course of treatment, but these elevations did not increase and were within a

controllable range by withdrawal or dosage reduction (Fig. 7). The individual dose of OCT per hemodialysis (HD) showed changes over time as shown in Figure 8. As a guide for the initial dose level of OCT, treatment was started at 10 µg/HD if intact-PTH level was >500 pg/mL or at 5 µg/HD if intact-PTH level was <500 pg/mL. The individual dose distribution showed a decline with time. While there were some cases with high PTH levels requiring dosage increases to 15 or 20 µg/HD, inconspicuous serum calcium elevations were maintained despite long-term treatment. These findings indicated the broad range of severity of secondary hyperparathyroidism. Analysis of such data from clinical studies conducted at many institutions covering a number of patients has revealed a potent clinical effect of OCT in lowering elevated serum PTH. At the same time, the data analysis also disclosed improvement of impaired bone metabolism by the administration of OCT. Marked improvements of abnormalities indicative of high-turnover bones were noted following the treatment Ionized Ca++

Corrected Ca

3

11

10

9

**

*** *** *** ***

*** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ***

**

2.5

*

** ** *** *** *** * ** ** ** ***** ***** *** *** ** ** ****** * * * **

* p < 0.05 *** p < 0.001 vs 0 week

−2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 +2 Weeks

*

N = 24−14

N = 160–109 ** p < 0.01

8

Ca++ (mEq/l)

Adjusted Ca (mg/dl)

12

2

** p < 0.01

*** p<0.001 vs 0 week

−2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 +2 Weeks

FIGURE 7 Changes in corrected serum calcium (Ca) levels and plasma ionized Ca (Ca++) concentration over time.

1531

CHAPTER 86 Development of OCT and ED-71

(%) 100

1

1

5 9

2

13

15

14

11

16

14

6

6

6

29

27

28

27

35

14

12

12

80 54

87

103 60

14 22

15 1 40 39

36

38

24

28

34

53

16 2 1 1

0 1

4 8 4

2

5

30

32

15 4

13

31

39

55 20

2

9

7 9 12

7

6

12

10

13

20

24

9 11 16

3

0 µg

2.5 µg

7.5 µg

12.5 µg

20 µg

1 µg

5 µg

10 µg

15 µg

Number of patients

26 (weeks)

FIGURE 8 Changes in OCT dose level distribution during the course of treatment. Numbers within bars are patients that received the indicated dose of OCT.

with respect to serum/bone metabolism markers, bone resorption indices such as tartrate-resistant acid phosphatase (TRACP) and collagen type I C-terminal telopeptide (ICTP), and abnormal elevations in alkaline phosphatase(ALP)/bone ALP (BAP), procollagen type I C-terminal propeptide (PICP) and osteocalcin (BGP) [37–39] (Fig. 9). A bone histomorphometric evaluation by Kurihara et al. showed reversal of high bone turnover following treatment with OCT in two patients with secondary hyperparathyroidism with “ostitis fibrosa type” skeletal changes. There was a positive increase in bone formation in one patient, hence the drug stimulated bone metabolic turnover, in the cortical bone of 2 other patients whose skeletal involvement was of the low bone turnover form, i.e., “aplastic” or “osteomalacic” type (Fig. 10). The authors stated that the enhanced bone formation observed in the low turnover bone cases was a fact of profound interest that suggested direct actions of the drug upon bone cells. Tsukamoto et al. reported a marked decrease in fibrotic tissue volume and improvement of high turnover bones following OCT therapy even in patients poorly responding with a serum PTH decrease [40] (Fig. 11). These findings suggest that the amelioration of osteopathy by OCT medication in severe secondary hyperparathyroidism may derive from direct actions of the drug on bone metabolism disparate from PTH-suppressive effects.

C. Psoriasis Vulgaris Inflammatory keratoses such as psoriasis vulgaris are chronic disorders of the skin with diverse, characteristic dermatologic manifestations including erythema, thickening, cornification, and scaling (see Chapters 35 and 101). In most cases the cutaneous disorders are not completely cured and are left to symptomatic treatment. The condition of the disease often interferes with comfort and affects the patients’ quality of life. The disease state is thought to be due to inflammation and abnormalities in proliferation and differentiation of skin cells. Topical corticosteroids are commonly used for treatment, and, in severe cases of the disease, therapies such as oral use of cyclosporine A and etretinate, a synthetic retinoid, may be applied. Adverse reactions to these drugs, however, are a major concern. Topical use of active vitamin D analogs as therapeutic agents for psoriasis has been highlighted recently (see Chapter 101). Inhibition of epidermal keratinocyte proliferation, stimulation, and induction of differentiation of those cells, and immunosuppression have been implicated in the mechanisms of action for efficacy of topical active vitamin D analogs. These drugs are advantageous and show much promise as drugs of first choice in the treatment of psoriasis. They are remarkably safe and provide a longer duration time from initial cure to relapse/recrudescence than conventional therapeutic agents.

1532

NOBORU KUBODERA

TRACP

10

N = 160–128 *** ; p < 0.001 vs 0 week

** ***

9

***

*** *** ***

8

ICTP

80

*** ***

N = 160–140 *** ; p < 0.001 vs 0 week

*** 70

ICTP (ng/ml)

TRACP (IU/L/37°C)

11

***

***

60

*** 50

40

7

−2 0

2 4 6

−2 0 2 4 6

8 10 12 14 16 18 20 22 24 26 +2

weeks 70 *

500

BAP

*** ***

***

***

40 *** *** 30

N = 160–128 *; p < 0.05 ***; p < 0.001 vs 0 week

20

400

*** *** ***

300

−2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 +2

weeks

BGP

PICP

260

200 ***

100

N = 160–141 *** ; p < 0.001 vs 0 week

PICP (ng/ml)

240

150

220 **

***

N = 160–140 **; p < 0.01 *** ; p < 0.001 vs 0 week

200

50 0

***

N = 160–128 *** ; p < 0.001 vs 0 week

weeks

250

***

*** ***

0

2 4 6 8 10 12 14 16 18 20 22 24 26 +2

***

200 100

10 −2 0

ALP

* ***

50

8 10 12 14 16 18 20 22 24 26 +2

weeks

ALP (IU/L/37°C)

BAP (IU/L)

60

BGP (ng/ml)

***

180 −2 0

2 4 6 8 10 12 14 16 18 20 22 24 26 +2

weeks

−2 0

2 4 6 8 10 12 14 16 18 20 22 24 26 +2

weeks

FIGURE 9 Changes in levels of bone metabolism markers following treatment with OCT. TRACP: Tartrateresistant acid phosphatase; ICTP: Collagen type 1 C-terminal telopeptide; ALP: Alkaline phosphatase; BAP: Bone alkaline phophatase; BGP: Osteocalcin; PICP: procollagen type 1 C-terminal propeptide.

Morimoto et al. demonstrated in vitro that OCT and 1,25(OH)2D3 similarly inhibited the growth of dermal fibroblasts from normal human skin in a dose-dependent manner. Growth of dermal fibroblasts from psoriatic patient skin was also inhibited by OCT in a dosedependent fashion, but not inhibited by 1,25(OH)2D3 [41]. Efficacy of OCT has been assessed primarily by in vitro studies because there is no established animal

disease model of psoriasis. Preclinical studies including toxicology and pharmacokinetics have been carried out via percutaneous administration. Clinical trials of OCT ointment commenced in 1994. Phase I and phase II studies were conducted in Japan and the UK to evaluate safety and efficacy for topical use. To determine clinical efficacy, phase II double-blind, randomized, left vs. right, concentration-response study was

1533

CHAPTER 86 Development of OCT and ED-71

(% year) 50

BFR/Ct. Ar

40 1 30

2 5 6

20

7 10

0 Before

After

FIGURE 10 Changes in cortical bone formation rate (BFR/Ct.Ar) following administration of OCT. Cases 2 and 7: Ostitis fibrosa (high turnover); Case 5: Slight change; Case 1: Aplastic (low turnover); Case 6: Osteomalacic (low turnover)

performed with once-daily topical OCT in patients with mild to moderate chronic plaque psoriasis. Primary efficacy parameters were psoriasis severity index (PSI) based upon the sum of scores for erythema, scaling, and induration and the investigators’ overall assessment of patients’ response to therapy at eight weeks of treatment. In this study 144 patients participated. All concentrations of OCT ointment (6, 12.5, 25, and 50 µg/g) were significantly more effective

at reducing PSI than placebo (P < 0.01), with the greatest effect noted for OCT 25 µg/g. Calcipotriol ointment 50 µg/g once daily as active comparator had a similar effect (see Chapter 84). Marked improvement or clearance of psoriasis was greatest for OCT 25 µg/g (54.7% of subjects), which compared favorably with calcipotriol (46.2%). Improvement continued throughout the study period, with no plateau at week 8. Investigators’ and patients’ side preference (secondary efficacy parameters) rated OCT more effective than placebo and 25 µg/g OCT better than calcipotriol (P < 0.05 for investigators’ assessment). Twelve patients withdrew from the study due to adverse events, of which four were judged to be due to study medication. This study indicates that once-daily OCT ointment is effective in the management of plaque psoriasis, with the greatest effect noted at a dosage of 25 µg/g. As no response plateau was seen at eight weeks, these data suggest that further benefits might be obtained if OCT ointment were applied for longer. In this study the investigators’ overall assessment and general preference suggest that OCT 25 µg/g may be more effective than once-daily calcipotriol (Fig. 12). Topical 1α,24(R)-dihydroxyvitamin D3 (tacalcitol) preparations were previously launched and marketed in Japan as a local therapeutic agent for psoriasis while topical calcipotriol, which is already in widespread use in Europe and the United States as a drug with greater efficacy, was also marketed in Japan in 2000. Comparative studies with these topical agents have demonstrated the clinical usefulness of OCT ointment [42]. Major adverse reactions of OCT include local irritation commonly seen with topical vitamin D preparations and elevation of serum calcium following general

25 (%)

Fb. V/ TV (%)

20

100

15 54.7

10

50

52.2 46.2

42.9 34.5

5 ∗∗

10

0

−5

3.6 0

8.6

14.3

22.7

21.3

P

6

12.5

25

50

11.3 calcipotriol

OCT (µg/g)

0

24

FIGURE 11 Reduction in bone marrow fibrous tissue volume

following 24-week treatment with OCT. Paired t-test p <0.001; Solid line: Cases showing a ≥50% decrease intact-PTH; Broken line: Cases showing a <30% decrease intact-PTH.

FIGURE 12 Investigators’ overall assessment (primary efficacy variable) of patients’ response to treatment with OCT ointment: between-patient comparison. The histogram shows results for marked improvement and clearance combined. Data for clearance alone are the numbers in white at the base of histogram. P: placebo.

1534

NOBORU KUBODERA

COOH OH

20 steps

3 steps

O

ED-71

HO

HO

α-Epoxide

Lithocholic acid

FIGURE 13 Original linear synthesis of ED-71 using lithocholic acid as a starting material.

known methods. Wittig reaction between the A-ring and the C/D-ring units provided ED-71 in satisfactory overall yield [43–44] (Fig. 14).

topical application. None of the reports, however, has been serious and their occurrence is infrequent. Calcemic responses were in no case serious. Such responses are more frequent with generalized topical use of high doses early in the course of treatment with OCT. Importantly, it has been documented that safety can be secured through monitoring serum calcium levels.

B. Preclinical Results 1. COMPARISON WITH ALFACALCIDOL

Although active vitamin D is used in certain countries for the treatment of osteoporosis, the risk of causing hypercalcemia/hypercalciuria suggests only a narrow therapeutic window. This has precluded worldwide acceptance. The results of our animal studies suggest that the therapeutic effect of active vitamin D on bone loss after estrogen deficiency can be dissociated, at least in part, from the effects of enhancing intestinal calcium absorption and suppressing parathyroid hormone (PTH) secretion [45–46]. In order to test this, we compared the effects of ED-71 with orally administered 1α-hydroxyvitamin D3 (alfacalcidol, 1αOHD3) on bone mineral density (BMD) and the bone remodeling process as a function of calcium metabolism and PTH levels, in an ovariectomized (OVX) model of osteoporosis. ED-71 increased bone mass at the lumbar vertebra to a greater extent than alfacalcidol (Fig. 15A). During this process calcium absorption was enhanced as indicated by urinary calcium

III. DEVELOPMENT OF ED-71 FOR OSTEOPOROSIS A. Convergent Synthesis of ED-71 for Versatile Method The original synthesis of ED-71 involved a linear route using lithocholic acid as a starting material and proceeded through the α-epoxide as a key intermediate [17–18] (Fig. 13). The 23-step synthetic method seems, however, to be inconvenient for the synthesis of highly functionalized related compounds such as postulated metabolites of ED-71. We, therefore, developed a convergent strategy to ED-71 as a versatile route consisting of the A-ring and the C/D-ring fragments. The readily available C2 symmetrical epoxide was converted to the A-ring fragment in 18 steps and the C/D-ring fragment was prepared from the Inhoffen-Lythgoe diol by

P(O)Ph2 BnO

18 steps

O

OBn

C2-Symmetrical epoxide

TBSO O

OTBS OTBS

A-ring fragment ED-71 OH

OH Inhoffen-Lythgoe diol

OTES

O C/D-ring fragment

FIGURE 14 Convergent synthesis of ED-71.

1535

CHAPTER 86 Development of OCT and ED-71

excretion and serum PTH levels was decreased to the same degree as alfacalcidol. ED-71 lowered biochemical and histological parameters of bone resorption more potently than alfacalcidol (Fig. 15B), while maintaining bone formation markers (Fig. 15C). These results suggest that ED-71 exerts an anti-osteoporotic effect by inhibiting osteoclastic bone resorption while maintaining osteoblastic function, and that these anticatabolic/anabolic effects of ED-71 take place independently of its effects on calcium absorption and PTH [47].

inhibitors and bone formation stimulants. Calcitonin preparations, estrogen preparations, and bisphosphonates are defined as bone resorption inhibitors that exert therapeutic effects mainly by inhibiting osteoclastic bone resorption. PTH, on the other hand, is defined as a bone formation stimulant whose effects are mediated chiefly by its ability to stimulate osteoblasts. It is known that, with bone resorption inhibitors, bone formation is also inhibited in association with suppression of bone resorption. With PTH it is generally thought that not only bone formation but also bone resorption is stimulated. In contrast to the mechanisms of action of these drugs, it has been shown that ED-71 may have the remarkable characteristic

2. POSSIBLE MECHANISM OF ACTION OF ED-71

Drugs for the treatment of osteoporosis currently available are grossly classified into bone resorption

A

B

250

25

∗

#

∗

∗ Urinary D-Pyr/Cr (nM/mM)

150

100

∗

15 #

∗

10

∗

∗

1 0. 2

05

0.

4

(µg/kg BW)

hi Ve

hi Ve

alfacalcidol

ED-71

alfacalcidol

0.

e cl

2

1

0.

05

0.

0.

4 0.

1

2

0.

0.

e cl

Sham

0.

0

(µg/kg BW)

Sham

1 0. 2

5

50

0

20

0.

Spinal BMD (mg/cm2)

200

∗ ∗

ED-71

OVX

OVX

C

25

15 # 10

Ve

alfacalcidol OVX

0. 2

5 0. 1

0. 0

0. 4

0. 2

Sham

hi c

0

0. 1

5

le

Serum Osteocalcin (ng/mL)

∗ 20

(µg/kg BW)

ED-71

FIGURE 15 Comparison of ED-71 with alfacalcidol in ovariectomized (OVX) rats. Eight-month-old OVX WisterImamichi rats were treated orally with indicated doses (µg/kg BW) of ED-71 or alfacalcidol, twice or thrice per week, respectively, for 3 months. Bone mineral density (BMD) at the lumbar vertebrae (15A) was measured by dual-energy X-ray absorptiometry. Urinary deoxypyridinoline (D-Pyr) (15B) was measured using a Pyrilinks-D assay kit. Serum osteocalcin (15C) was measured using rat osteocalcin radioimmunoassay (RIA) reagents.

1536

NOBORU KUBODERA

that it inhibits bone resorption and stimulates bone formation. In Figure 16, changes in markers of bone metabolism observed in animals treated with ED-71 are depicted in comparison with changes following administration of 17β-estradiol. The experiment with ED-71 was performed in a therapeutic manner by initiating treatment after three months had elapsed post ovariectomy (OVX). The experiment with 17β-estradiol consisted of a prophylactic study in which the estrogenic hormone was administered s.c. for three months, starting just after an OVX, because an increase in bone mass could not be anticipated in this system. Urinary deoxypyridinoline excretion (Dpyr/Cr), a bone resorption parameter, and osteocalcin (BGP),

a bone formation parameter, both decreased dosedependently to below the sham-operated group level in response to administration of 17β-estradiol [46]. As a result, the decrease in BMD due to ovary anatomy was inhibited, such that the BMD was maintained at the sham-operated group level but showed no further increase. In the animals treated with ED-71, enhanced urinary Dpyr/Cr was lowered to below the sham-operated group level as in the 17β-estradiol treated group while BGP was increased in a dose-dependent fashion. Animals given PTH displayed a dose-dependent increase in BGP but no trend in increasing urinary Dpyr/Cr was observed (data not shown in this figure).

BMD vs. Dpyr/Cr

BMD vs. BGP * p < 0.05 (vs. OVX)

BMD

Dpyr/Cr

(mg/cm2) 250

BGP (ng/ml) 12

BMD

(nM/mM) 60

(mg/cm2) 250

10 40

200

8

200 *

150

100

20 *

*

*

4

20

100

*

4 2

0 sham OVX

6 *

150

100

0 sham OVX

E2 (µg/kg)

4

20

100

E2 (µg/kg)

BMD vs. Dpyr/Cr

BMD vs. BGP * p < 0.05 (vs. OVX)

Dpyr/Cr (nM/mM)

BMD (mg/cm2)

100

250

BGP (ng/ml)

BMD (mg/cm2)

16

250

14

80 200

200

40

150 * 100

12

60

sham OVX 0.05

*

*

0.1

0.2

ED-71 (µg/kg)

10

150

8

20 0

100

sham OVX 0.05

0.1

0.2

6

ED-71 (µg/kg)

FIGURE 16 Changes in bone metabolism parameters in 17β-estradiol (E2) treated groups

and ED-71 treated groups. Data are presented as the mean ± SE, *p < 0.05 (vs. Ovariectomy (OVX), Dunnett’s multiple comparison). BMD: Bone mineral density; Dpyr/Cr: Urinary deoxypyridinoline excretion; BGP: Osteocalcin.

1537

CHAPTER 86 Development of OCT and ED-71

Results of bone histomorphometric analysis made on hard tissue preparations of the third lumbar vertebra from animals are shown in Fig. 17. In the ED-71 treated group, there was a decrease in osteoclast surface (Oc.S/BS) to below the sham-operated group level while bone formation rate (BFR/BS) was increased to above the OVX group level. There was also an increase in mineral apposition rate (MAR) in these animals; thus bone formation was definitely stimulated. Regarding the bone formation stimulant effect of ED-71, Tsurukami et al. reported that ED-71 produced increases in bone mass and bone formation rate without reducing the trabecular unit activation frequency. This was inferred from bone histomorphometric indices for the fifth lumbar vertebrae of OVX rats having been given ED-71 for 12 weeks [48]. When normal rats of which skeletal growth had ceased due to aging were dosed with ED-71 for 12 weeks, subsequent bone histomorphometric examination revealed a significant increase in lumbar vertebral bone density and dosedependent increases in total osteoid surface (OS/BS), MAR, and BFR/BS. These results provide supportive evidence for the stimulation of bone formation by ED-71 in vivo. The above data demonstrate that ED-71 may have a uniquely characteristic mechanism of action whereby it inhibits bone resorption and also stimulates bone formation. This mode of action is unlike that displayed by currently available bone resorption inhibitors, which are known to inhibit bone resorption and also to suppress bone formation. Furthermore, bone formation stimulants such as PTH accelerate not only bone formation but bone resorption as well (Fig. 18). At present, drugs which inhibit bone resorption constitute the primary therapeutic modality in the medical

Oc.s/BS

BFR/BS

1

0

*

*

Sham OVX 0.1 0.2

ED-71,µg/kg twice a week

*

1.2

75 50 25 0

1. PHASE I

Phase I clinical trial of ED-71 was conducted to estimate safe dosage levels and assess the pharmacokinetics of ED-71 in healthy adult male volunteers. There were no findings of clinical concern, including a calcemic effect, resulting from a single-dose administration study at dose levels of 0.01 to 1.0 µg. In a repeated-dose administration study using dosages of 0.1, 0.25, 0.5, and 1.0 µg once daily for 15 days, the following observations were noted: 1. There was a dose-related increase in serum concentration of ED-71, and the concentration reached a steady state by day 13 of administration, showing linearity within a dose range of 0.1 to 1.0 µg.

1.5

MAR (µm)

2

C. Clinical Results

MAR

100 BFR/BS (µm2/year)

Oc.s/BS (%)

3

treatment of osteoporosis. Generally, patients with osteoporosis often have a decrease in bone mass, subjective symptoms, and intercurrent fractures of vertebral bodies. Through the use of drugs which inhibit bone resorption, halting the progression of the disease may be achievable but the diminished bone mass cannot be restored. Active vitamin D analogs also appear to have a superior efficacy in reducing incidence of fractures in osteoporotic patients compared to their efficacy in maintaining/increasing bone mass. It would be probable that the incidence of fractures may be further reduced if an active vitamin D analog possessing potent bone mass-increasing activity were to become available. Since ED-71 appears capable of restoring decreased bone mass via its bone formation inducing activity, it serves as a potentially promising, ideal therapeutic agent for osteoporosis.

*

PTH

0.9 0.6

OVX Bone formation

0.3 Sham OVX 0.1

0.2

ED-71,µg/kg twice a week

0

ED-71

Sham OVX 0.1 0.2

Estrogen

ED-71,µg/kg twice a week

FIGURE 17 Bone histomorphometric data for lumbar vertebrae of ovariectomized (OVX) rats following treatment with ED-71. 8-month-old female Wistar rats were ovariectomized and, upon confirming a sufficient bone mass decrease after 4 months postoperation, they were administered ED-71 twice weekly for 3 months. Data are presented as the mean ± SE. Oc.s/BS: Osteoclast surface; BFR/BS:Bone formation rate; MAR: Mineral apposition rate.

sham

Bone resorption

FIGURE 18 Possible mechanism of action of ED-71.

1538

NOBORU KUBODERA

2. The earliest pharmacological reaction that appeared was an increase in urinary calcium excretion per day. This increase was dose-dependent with no evidence of a change in serum calcium concentrations. 3. Bone resorption markers decreased in a dosedependent manner while bone formation markers showed no significant change. 4. PTH suppression was not observed even at dose levels of the drug that caused a significant decrease in plasma 1α,25-dihydroxyvitamin D concentration. The above results demonstrate that ED-71 effectively inhibits bone resorption without suppressing bone formation at nonhypercalciuric/nonhypercalcemic dosage levels. 2. EARLY PHASE II

In early phase II clinical trials, a randomized controlled study with ED-71 in 109 osteoporotic patients (102 females and 7 males), 49 to 81 years of age (mean 65.0 years) was conducted. The patients were randomly assigned to either 0.25, 0.5, 0.75, or 1.0 µg/day of ED-71 administered orally. They were treated for six

months, and BMD and bone markers were evaluated. ED-71 treatment increased BMD in the lumbar spine (L2-4) in a dose-dependent manner (0.34 ± 0.73, 0.50 ± 0.91, 3.00 ± 0.65 and 2.66 ± 0.71% in the 0.25, 0.5, 0.75 and 1.0 µg groups, respectively, mean ± SE), and the effect of ED-71 reached the peak at 0.75 µg (Fig. 19). The percentages of patients that showed an increase in the L2-4 BMD over 3% after six months also increased dose-dependently (21.7, 26.1, 54.2, and 45.5% in the 0.25, 0.5, 0.75, and 1.0 µg groups, respectively). ED-71 exhibited dose-dependent suppression of urinary deoxypyridinoline and Crosslaps excretion as well as serum bone-type alkaline phosphatase, whereas serum osteocalcin was not suppressed, suggesting maintenance of bone formation with suppression of bone resorption. Serum level of 1α,25-dihydroxyvitamin D was reduced dosedependently and 24,25-dihydroxyvitamin D levels were increased by the treatment of ED-71. These findings reveal that ED-71 treatment inhibits the activation of native vitamin D and accelerates the metabolism of activated vitamin D. Intact PTH levels were not remarkably suppressed by ED-71 compared to the treatment levels. At present we do not know why

6.0

P = 0.088 5.0 P = 0.036

Percent changes (%)

4.0 ** ** 3.0

2.0

1.0

0.0 Dosage (µg) (Patients)

0.25 (23)

0.5 (23)

0.75 (24)

1.0 (22)

FIGURE 19 Percentage change in lumbar spine (L2-4) bone mineral density (BMD) after 24 weeks treatment of ED-71. **p < 0.01: Paired t-test (Dunnett’s test) between groups.

1539

CHAPTER 86 Development of OCT and ED-71

ED-71 is unable to suppress PTH levels, although this might be due to a unique property of ED-71 compared to active vitamin D3 compounds such as 1,25(OH)2D3 and alfacalcidol. ED-71 was well-tolerated without causing hypercalcemia, and no patient exhibited sustained postprandial hypercalciuria over 0.4 mg/dL GF. These results support the need for further longterm clinical studies to examine the effects of ED-71 on bone fracture in osteoporotic patients [49–50]. 3. LATE PHASE II

Based upon the results of the early phase II trials with ED-71, we are now conducting late phase II further clinical studies as with the following protocol: Patients: primary osteoporosis 200 patients Design: randomized double-blinded study Doses: 0, 0.5, 0.75, and 1.0 µg/day Treatment; once a day for successive 48-week treatment 400 or 200 IU vitamin D3 is added depending on initial 25-hydroxyvitamin D levels • End points at 48 weeks: lumbar spine (L2-4) BMD and bone markers

• • • •

The above-mentioned late phase II study will be completed until summer of 2003 and the subsequent phase III study for fracture prevention is scheduled.

6.

7.

8.

9.

10.

11. 12.

13.

Acknowledgments I am grateful to the following people at Chugai Pharmaceutical Co., Ltd., for their kind help and discussion to complete the manuscript: Sadaaki Yokoyama, Koichi Endo, Yasushi Uchiyama, and Naoki Tsuji.

References 1. Abe E, Miyaura C, Sakagami H, Takeda M, Konno K, Yamazaki T, Yoshiki S, Suda T 1981 Differentiation of mouse myeloid leukemia cells induced by 1α,25-dihydroxyvitamin D3. Proc Natl Acad Sci 78:4990–4994. 2. Bouillon R, Okamura WH, Norman AW 1995 Structure-function relationships in the vitamin D endocrine system. Endocrinology Rev 16:200–257. 3. Kubodera N, Miyamoto K, Ochi K, Matsunaga I 1986 Synthetic studies of vitamin D analogs. VII. Synthesis of 20-oxa-21-norvitamin D3 analogs. Chem Pharm Bull (Tokyo) 34:2286–2289. 4. Murayama E, Miyamoto K, Kubodera N, Mori T, Matsunaga I 1986 Synthetic studies of vitamin D3 analogs VIII. Synthesis of 22-oxavitamin D3 analogs. Chem Pharm Bull (Tokyo) 34:4410–4413. 5. Kubodera N, Miyamoto K, Akiyama M, Matsumoto M, Mori T 1991 Synthetic studies of vitamin D analogs. IX. Synthesis

14. 15.

16.

17.

18.

19.

differentiation-inducing activity of 1α,25-dihydroxy-23-oxa-, thia-, azavitamin D3. Chem Pharm Bull (Tokyo) 39:3221–3224. Kubodera N, Miyamoto K, Matsumoto M, Kawanishi T, Ohkawa H, Mori T 1992 Synthetic studies of vitamin D analogs. X. Synthesis and biological activities of 1α,25-dihydroxy-21-norvitamin D3. Chem Pharm Bull (Tokyo) 40:648–651. Kubodera N, Watanabe H, Kawanishi T, Matsumoto M 1992 Synthetic studies of vitamin D analogs. XI. Synthesis and differentiation-inducing activity of 1α,25-dihydroxy-22oxavitamin D3 analogs Chem Pharm Bull (Tokyo) 40:1494–1499. Murayama E, Miyamoto K, Ochi K, Kubodera N 1992 Synthesis and immunoregulating activity of vitamin D analogs bearing pregnane side chains. Bioorg Med Chem Lett 2:1289–1292. Kubodera N, Watanabe H, Miyamoto K, Matsumoto M 1993 Synthesis and differentiation-inducing activity of 1α,24dihydroxy-22-oxavitamin D3 analogs. Bioorg Med Chem Lett 3:1869–1872. Kubodera N, Watanabe H, Miyamoto K, Matsumoto M, Matsuoka S, Kawanishi T 1993 Synthetic studies of vitamin D analogs. XVII. Synthesis and differentiation-inducing activity of 1α,24-dihydroxy-22-oxavitamin D3 analogs and their 20(R)epimers. Chem Pharm Bull (Tokyo) 41:1659–1663. Kawase A, Hirata M, Endo K, Kubodera N 1995 Synthesis of 1α,25-dihydroxy-22-thiavitamin D3 and related analogs. Bioorg Med Chem Lett 5:279–282. Abe J, Morikawa M, Miyamoto K, Kaiho S, Fukushima M, Miyaura C, Abe E, Suda T, Nishii Y 1987 Synthetic analogs of vitamin D3 with an oxygen atom in the side chain skeleton. FEBS Lett 226:58–62. Koike N, Hayakawa N, Kumaki K, Stumpf WE 1998 In vivo dose-related receptor binding of the vitamin D analogue [3H]1,25-dihydroxy-22-oxavitamin D3 (OCT) in rat parathyroid, kidney distal and proximal tubules, duodenum, and skin, studied by quantitative receptor autoradiography. J Histochem Cytochem 46:1351–1358. Kobayashi T, Okano T, Tsugawa N, Masuda S, Takeuchi A, Nishii Y 1991 Metabolism and transporting system of 22-oxacalcitriol. Comtrib Nephrol 91:129–133. Okano T, Tsugawa N, Masuda S, Takeuchi A, Kobayashi T, Nishii Y 1989 Protein-binding properties of 22-oxa-1α,25dihydroxyvitamin D3, a synthetic analogue of 1α,25-dihydroxyvitamin D3. J Nutr Sci Vitaminol 35:529–533. Teramoto T, Endo K, Ikeda K, Kubodera N, Kinoshita M, Yamanaka M, Ogata E 1995 Binding of vitamin D to low-density-lipoprotein (LDL) and LDL receptor-mediated pathway into cells. Biochem Biophys Res Commun 215: 199–204. Miyamoto K, Kubodera N, Murayama E, Ochi K, Mori T, Matsunaga I 1986 Synthetic studies on vitamin D analogs VI. A new synthesis of 25-hydroxycholesterol from lithocholic acid. Syn Commun 16:513––521. Miyamoto K, Murayama E, Ochi K, Watanabe H, Kubodera N 1993 Synthetic studies of vitamin D analogs. XIV. Synthesis and calcium regulating activity of vitamin D3 analogs bearing a hydroxyalkoxy group at the 2β-position. Chem Pharm Bull (Tokyo) 41:1111–1113. Ono Y, Watanabe H, Shiraishi A, Takeda S, Higuchi Y, Sato K, Tsugawa N, Okano T, Kobayashi T, Kubodera N 1997 Synthetic studies of vitamin D analogs. XXIV. Synthesis of active vitamin D3 analogs substituted at the 2β-position and their preventive effects on bone mineral loss in ovariectomized rats. Chem Pharm Bull (Tokyo) 45:1626–1630.

1540 20. Ono Y, Kawase A, Watanabe H, Shiraishi A, Takeda S, Higuchi Y, Sato K, Yamauchi T, Mikami T, Kato M, Tsugawa N, Okano T, Kubodera N 1998 Syntheses and preventive effects of analogs related to 1α,25-dihydroxy-2β-(3-hydroxypropoxy) vitamin D3 (ED-71) on bone mineral loss in ovariectomized rats. Bioorg Med Chem 6:2517–2523. 21. Tsugawa N, Nakagawa K, Kurobe M, Ono Y, Kubodera N, Ozono K, Okano T 2000 In vitro biological activities of a series of 2β-substituted analogs of 1α,25-dihydroxyvitamin D3. Biol Pharm Bull (Tokyo) 23:66–77. 22. Okano T, Tsugawa N, Masuda S, Takeuchi A, Kobayashi T, Takita Y, Nishii Y 1989 Regulatory activities of 2β-(3-hydroxypropoxy)-1α,25-dihydroxyvitamin D3, a novel synthetic vitamin D3 derivative, on calcium metabolism. Biochem Biophys Res Commun 163:1444–1449. 23. Kobayashi T, Okano T, Tsugawa N, Murano M, Masuda S, Takeuchi A, Sato K, Nishii Y 1993 2β-(3-Hydroxypropoxy)1α,25-dihydroxyvitamin D3 (ED-71), preventive and therapeutic effects on bone mineral loss in ovariectomized rats. Bioorg Med Chem Lett 3:1815–1819. 24. Kubodera N, Sato K, Nishii Y 1997 Characteristics of 22oxacalcitriol (OCT) and 2β-(3-hydroxypropoxy)calcitriol (ED-71). In “Vitamin D” (D Feldman, FH Glorieux, JW Pike ed.), pp. 1071–1086. Academic Press, San Diego. 25. Mikami T, Iwaoka T, Kato M, Watanabe H, Kubodera N 1997 Practical formation of γ-keto or γ-hydroxy ether linkage from alcohol. Syn Commun 27:2363–2369. 26. Silver J, Naveh-Many T, Mayer H, Schmelzer HJ, Popovtzer MM 1986 Regulation by vitamin D metabolites of parathyroid hormone gene transcription in vivo in the rat. J Clin Invest. 78:1296–1301. 27. Sato K, Tominaga Y, Ichikawa F, Uchida K, Takezawa J, Tsuchiya T, Kubodera N, Takagi H 1998 In in vitro suppression of parathyroid hormone secretion by 22-oxa-calcitriol in human parathyroid hyperplasia due to uraemia. Nephrology 4:177–182. 28. Brown AJ, Ritter CR, Finch JL, Morrissey J, Martin KJ, Murayama E, Nishii Y, Slatopolsky E 1989 The noncalcemic analogue of vitamin D, 22-oxacalcitriol, suppresses parathyroid hormone synthesis and secretion. J Clin Invest 84:728–732. 29. Ichikawa F, Hirata M, Endo K, Katsumata K, Ohkawa H, Kubodera N, Fukagawa M, Kurokawa K 1998 Attenuated up-regulation of vitamin D-dependent calcium-binding proteins by 22-oxa-1,25-dihydroxyvitamin D3 in uremic rats: a possible mechanism for less-calcemic action. Nephrology 4:391–395. 30. Hirata M, Endo K, Ohkawa H, Kumaki K, Kubodera N, Slatopolsky E, Kurokawa K, Fukagawa M 2002 A comparison between daily and thrice-weekly i.v. administration of 1,25dihydroxy-22-oxavitamin D3 regarding suppression of parathyroid hormone secretion and calcemic action in uraemic rats. Nephrol Dial Transplant 17 (suppl 10):37–40. 31. Hirata M, Endo K, Katsumata K, Ichikawa F, Kubodera N, Fukagawa M 2002 A comparison between 1,25-dihydroxy-22oxavitamin D3 and 1,25-digydroxyvitamin D3 regarding suppression of parathyroid hormone and calcemic action. Nephrol Dial Transplant 17 (suppl 10):41–45. 32. Takahashi F, Furuichi T, Yorozu K, Kawata S, Kitamura H, Kubodera N, Slatopolsky E 2002 Effects of i.v. and oral 1,25dihydroxy-22-oxavitamin D3 on secondary hyperparathyroidism in dogs with chronic renal failure. Nephrol Dial Transplant 17 (suppl 10):46–52. 33. Brown AJ, Finch J, Grieff M, Ritter C, Kubodera N, Nishii Y, Slatopolsky E 1993 The mechanism for the disparate actions of calcitriol and 22-oxacalcitriol in the intestine. Endocrinology 133:1158–1164.

NOBORU KUBODERA

34. Shibata S, Sakaguchi H, Nagasawa T, Naruse T 1972 Nephritogenic glycoprotein. II. Experimental production of membranous glomerulonephritis in rats by a single injection of homologous renal glycopeptde. Lab Invest 27:457–465. 35. Hirata M, Katsumata K, Masaki T, Koike N, Endo K, Tsunemi K, Ohkawa H, Kurokawa K, Fukagawa M 1999 22Oxacalcitriol ameliorates high-turnover bone and marked osteitis fibrosa in rats with slowly progressive nephritis. Kidney Int 56:2040–2047. 36. Monier-Faugere M.-C, Geng Z, Friedler RM, Qi Q, Kubodera N, Slatopolsky E, Malluche HH 1999 22-Oxacalcitriol suppresses secondary hyperparathyroidism without inducing low bone turnover in dogs with renal failure. Kidney Int 55:821–832. 37. Kurokawa K, Akizawa T, Suzuki M, Akiba T, Ogata E, Slatopolsky E 1996 Effect of 22-oxacalcitriol on hyperparathyroidism of dialysis patients: results of a preliminary study. Nephrol Dial Transplant 11 (Suppl 3):121–124. 38. Kinugasa E, Akizawa T, Takahashi J, Nabeshima K, Ogata H, Nakayama F, Ideura T 2002 Effects of 1,25-dihydroxy-22oxavitamin D3 on parathyroid gland function in haemodialysis patients with secondary hyperparathyroidism. Nephrol Dial Transplant 17 (suppl 10):20–27. 39. Akizawa T, Suzuki M, Akiba T, Nishizawa Y, Ohashi Y, Ogata E, Slatopolsky E, Kurokawa K 2002 Long-term effect of 1,25dihydroxy-22-oxavitamin D3 on secondary hyperparathyroidism in haemodialysis patients. One-year administration study. Nephrol Dial Transplant 17 (suppl 10):28–36. 40. Tsukamoto Y, Hanaoka M, Matsuo T, Saruta T, Nomura M, Takahashi Y 2000 Effect of 22-oxacalcitriol on bone histology of hemodialyzed patients with severe secondary hyperparathyroidism. Am J Kid Des 35:458–464. 41. Morimoto S, Imanaka S, Koh E, Shiraishi T, Nabata T, Kitano S, Miyashita Y, Nishii Y, Ogihara T 1989 Comparison of the inhibitions of proliferation of normal and psoriatic fibroblast by 1α,25-dihydroxyvitamin D3 and synthetic analogs of vitamin D3 with oxygen atom in their side chain. Biochem Int 19:1143–1149. 42. Barker JNWN, Ashton RE, Marks R,Harris RI, Berth-Jones J 1999 Topical maxacalcitol for the treatment of psoriasis vulgaris: A placebo-controlled, double-blind, dose-finding study with active comparator. British J Dermatol 141:274–278. 43. Hatakeyama S, Ikeda T, Maeyama J, Esumi T, Iwabuchi Y, Irie H, Kawase A, Kubodera N 1997 Convergent synthesis of 1α,25-dihydroxy-2β-(3-hydroxypropoxy)vitamin D3 (ED-71). Bioorg Med Chem Lett 22:2871–2874. 44. Hatakeyama S, Kawase A, Uchiyama Y, Maeyama J, Iwabuchi Y, Kubodera N 2001 Synthesis and biological characterization of 1α,24,25-trihydroxy-2β-(3-hydroxypropoxy)vitamin D3 (24-hydroxylated ED-71). Steroids 66:267–276. 45. Shira-Ishi A, Higashi S, Ohkawa H, Kubodera N, Hirasawa T, Ezawa I, Ikeda K, Ogata E 1999 Rationale for the advantage of alfacalcidol over native vitamin D in the treatment of osteoporosis. Calcif Tissue Int 65:311–316. 46. Shira-Ishi A, Takeda S, Masaki T, Higuchi Y, Uchiyama Y, Kubodera N, Sato K, Ikeda K, Nakamura T, Matsumoto T, Ogata E 2000 Alfacalcidol inhibits bone resorption and stimulates formation in an ovariectomized rat model of osteoporosis. Distinct action from estrogen. J Bone Miner Res 15:770–779. 47. Uchiyama Y, Higuchi Y, Takeda S, Masaki T, Shira-Ishi A, Sato K, Kubodera N, Ikeda K, Ogata E 2002 ED-71, a vitamin D analog, is a more potent inhibitor of bone resorption than alfacalcidol in an estrogen-deficient rat model of osteoporosis. Bone 30:582–588.

CHAPTER 86 Development of OCT and ED-71

48. Tsurukami H, Nakamura T, Suzuki K, Sato K, Higuchi Y, Nishii Y 1994 A novel synthetic vitamin D analogue, 2β(3-hydroxypropoxy)1α,25-dihydroxyvitamin D3 (ED-71), increases bone mass by stimulating the bone formation in normal and ovariectomized rats. Calcif Tissue Int 54:142–149. 49. Matsumoto T, Kubodera N, The ED-71 Study Group 2000 1α,25-Dihydroxy-2β-(3-hydroxypropoxy)vitamin D3 (ED-71): A promising candidate for the treatment of osteoporosis.

1541 In “Proceedings of Eleventh Workshop on Vitamin D, Nashville Tennessee, USA, May 2000” (AW Norman, R Bouillon, M Thomasset Ed.), pp. 985–992. University of California, Riverside. 50. Kubodera N, Tsuji N, Uchiyama Y, Endo K 2003 A new active vitamin D analog, ED-71, increase in bone mass with preferential effects on bone in osteoporotic patients. J Cell Biochem 88:286–228.

CHAPTER 87

2-Carbon-Modified Analogs of 19-Nor-1α,25-Dihydroxyvitamin D3 HECTOR F. DELUCA, LORI A. PLUM, MARGARET CLAGETT-DAME, NIRUPAMA K. SHEVDE, AND J. WESLEY PIKE Department of Biochemistry, University of Wisconsin-Madison, Madison, WI 53706

RAFAL R. SICINSKI Department of Chemistry, University of Warsaw, ul. L. Pasteura 2, Warsaw 02-093 Poland

I. II. III. IV.

Introduction 19-Nor- and 10,19-Saturated Derivatives of 1,25(OH)2D3 Epimerization of the 20-Carbon Early 2-Carbon Analogs of 19-Nor-1,25(OH)2D3 and 25-OH-D3 V. 2-Methylene and 2α-Methyl and 2β-Methyl Derivatives of 1,25(OH)2D3 VI. 2-Methylene-19-Nor-(20S)-1,25(OH)2D3 (2MD) and 2α-Methyl-19-Nor-(20S)-1,25(OH)2D3: Analogs That Possess Anabolic Activity on the Synthesis of Bone and Appear to be Bone Selective Analogs of 1,25(OH)2D3 VII. 2-Methylene-19-Nor-(20S)-1,25(ΟΗ)2D3: Molecular Mechanisms of Tissue Selectivity and Enhanced Potency

VIII. 2-Methylene-19-Nor-1α-Hydroxypregnacalciferol (2Mpregna), 2-Methylene-19-Nor-1αHydroxyhomopregnacalciferol (2MP) and 2-Methylene-19-Nor-(20S)-1αHydroxybishomopregnacalciferol (2MbisP) IX. 2-Methylene-19-Nor-Pregnacalciferol (2Mpregna), 2-Methylene-19-Nor-1α-Hydroxyhomopregnacalciferol (2MP) and 2-Methylene-19-Nor-(20S)-1αHydroxybishomopregnacalciferol (2MbisP): Molecular Mechanisms of Tissue Selectivity X. Summary References

I. INTRODUCTION

Chugai [5] (see Chapter 86) and the 16-ene derivatives of Hoffmann-La Roche [6] (see Chapter 85). Unfortunately, these deductions were based on in vitro assays in the case of cellular differentiation and in vivo assays in the case of calcium mobilization and serum calcium elevation. Nevertheless, it resulted in compounds that appeared to be much less calcemic in vivo, but that retained activity in cellular differentiation and presumably in other cellular activities. Seizing on this opportunity, Leo Pharmaceuticals successfully developed MC-903 for the topical treatment of psoriasis, providing a highly successful drug, Dovonex™ [7] (see Chapter 101). Of great interest is that these compounds are largely noncalcemic because they are rapidly metabolized and excreted, thereby eliminating their activity in vivo [8,9]. Thus, these derivatives, although active when applied topically, have very low activity when provided systemically. The 22-oxa derivative appears to retain sufficient lifetime in the plasma to suppress parathyroid activity and it has found use as an analog for the treatment of bone disease secondary to kidney failure [10]. Thus, the primary discrimination in the case of these analogs is rapid metabolism and elimination once they enter the circulation. This does not result in a truly noncalcemic analog, but one that is

With the discovery of the many functions of vitamin D beyond calcium homeostasis and phosphate metabolism has come the search for the holy grail of analogs. This search is largely aimed at analogs of 1α,25-dihydroxyvitamin D3 (1,25(OH)2D3) that produce no hypercalcemia or hyperphosphatemia while still retaining activities such as suppression of parathyroid gland proliferation and preproparathyroid hormone expression, the regulation of keratinocyte proliferation and gene expression, inducing of apoptosis in cancer, and suppression of proliferation by malignant cells. This has been the driving force in the chemical synthesis of a large variety of analogs. At least partial success has been achieved for a variety of different reasons. Side-chain modifications have been the particular target of synthesis. In our laboratory, elongating the side chain of the 24-carbon proved to reduce in vivo activity of stimulating intestinal calcium transport and bone calcium mobilization while at the same time retaining or increasing activity in the cellular differentiation assay using HL-60 cells in culture [1–3]. Similar results were obtained with the MC-903 compound of Leo Pharmaceuticals [4] (see Chapter 84), 22-oxa-1α,25-dihydroxyvitamin D3 of VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

1544 discriminated on the basis of metabolism. However, other side-chain alterations have proven to be very useful as described below in the case of some specific analogs. One of the important lessons to be learned from the early work on analogs is that the compounds must be tested in vivo for all activities before firm deductions can be made regarding selectivity of activity. Thus, the quest still remains for a noncalcemic form of 1,25(OH)2D3 that nevertheless can act in vivo to suppress cancer, to suppress parathyroid secretion, and to suppress keratinocyte proliferation. They may also be useful for the treatment of autoimmune diseases as well. The following chapter will be devoted to the development of some specific 2-carbon-modified analogs of 19-nor-1,25(OH)2D3 that show promise for selectivity for bone and which may provide new analogs that are devoid of calcium activity while still functioning in vivo in suppression of parathyroid hormone or acting on the keratinocyte.

II. 19-NOR- AND 10,19-SATURATED DERIVATIVES OF 1,25(OH)2D3 An important ingredient in the development of the 2-carbon-modified analogs that will be described in this chapter was the finding of the lack of importance of the 10,19-methylene carbon for the function of 1,25(OH)2D3. Before the advent of a crystal structure or computational models of the vitamin D receptor (VDR), one interest in our laboratory was to determine whether the cis-triene structure of 1,25(OH)2D3 is essential for function. As a result, 19-nor-1α,25-dihydroxyvitamin D3 was chemically synthesized [11]. The molecule and related derivative were then tested for biological activity in vivo as well as in vitro. These compounds retained the ability to bind to the VDR, did not alter the ability to induce in vitro transcription using reporter genes and the 24-hydroxylase promoter, but appeared to be 1/25th as active as 1,25(OH)2D3 in raising serum calcium and serum phosphorus. Of considerable interest is that when the 19-nor-1,25(OH)2D3 derivative was further derivatized by elongating the side chain by the addition of methyl groups on the 26- and 27-carbons or by the provision of 2-methylene groups providing a 5-membered ring surrounding the 25-hydroxyl, these compounds repossessed their calcemic activity [12]. Thus, it is very clear that the 10,19-methylene group is not required for the functions of vitamin D. Furthermore, if the side chain is either the cholesterol or ergosterol side chain, it has reduced calcemic activity. This resulted in the development of

HECTOR F. DELUCA,

ET AL .

the important new drug, paracalcitol or Zemplar®, for the treatment of renal osteodystrophy [13]. The advantages of this compound over calcitriol itself are that it is much less calcemic, but only slightly less active in suppression of parathyroid tissue and parathyroid secretion. Thus, the use of Zemplar® has provided the nephrologists with an important treatment of secondary hyperparathyroidism of renal osteodystrophy with a markedly increased window of safety (see Chapter 76). The value of this compound has recently been confirmed by an impressive 50,000patient evaluation of its success versus that of calcitriol [14]. Thus, the 19-nor derivatives represent a true in vivo-active compound with reduced calcemic activity. In pursuit of further understanding of the lack of importance of this derivatization to the action of vitamin D, the 10,19-dihydro-1α,25-dihydroxyvitamin D3 compound was prepared. It was found to be much less active in all respects [15]. The methyl group in that position alters significantly the ability of the A-ring to assume an appropriate conformation and thus did not provide the necessary information in regard to the importance of the 10,19-methylene group. Finally, the 6,7-aza-1α,25-dihydroxyvitamin D3 derivative was prepared and was also devoid of biological activity despite the fact that it retained the π electrons that would be provided by the diene structure of the 19-nor1,25(OH)2D3 [16]. It appears that the diene structure is absolutely essential for the function of the vitamin D hormone. Furthermore, this has been confirmed by the exact alignment in crystal structure of the 5–6 and 7–8 double bonds with the tryptophan π electrons and also explains the absorption shift of the tryptophan residue when the ligand is bound.

III. EPIMERIZATION OF THE 20-CARBON The 20-carbon epimerization of 1,25(OH)2D3 has been studied largely by Leo Pharmaceuticals [17]. Furthermore, Wicha and colleagues (unpublished) have studied the epimerization of the 20-carbon of vitamin D. The epimerization of the 20-carbon appears to play little or no role in binding to the receptor, but significantly increases biological activity both in vivo and in vitro [18]. This derivatization does not appear to provide selectivity in terms of the known actions of vitamin D but merely will increase under some circumstances biological activity by severalfold. This epimerization has been incorporated into the new selective 2-carbonmodified analogs described here.

1545

CHAPTER 87 2-Carbon-Modified Analogs

IV. EARLY 2-CARBON ANALOGS OF 19-NOR-1,25(OH)2D3 AND 25-OH-D3 An important breakthrough was that provided by Nishii et al. in which it was shown that the 2βsubstitution of 2β(hydroxy)-3′-propylalcohol (ED-71) was well tolerated in binding to the VDR and provided activity that seemed to favor bone calcium mobilization [19]. Most important, it illustrated that large substitutions on the 2-carbon could be made without interfering with binding to the receptor and subsequent biological activity. This led to an investigation of 19-nor vitamin D compounds in our laboratory which were substituted with the 2α- and 2β-hydroxyls. In this study, the 2β-hydroxylated compound was slightly better than the 2α-hydroxylated compound, but nevertheless they both provided significant in vivo biological activity [20]. This was followed by an investigation into the substitution of larger groups. In particular, the presence of the 2-benzyloxy ether in the 2α-position not only was well tolerated but provided high intestinal calcium transport activity with relatively low bone calcium mobilizing activity [20]. The Chugai compound ED-71 is currently being developed for the treatment of osteoporosis (see Chapter 86). These compounds all led to our investigation of substitutions on the 2-carbon position of 19-nor-1,25(OH)2D3 compounds.

2-methylene-19-nor-(20S)-1,25-dihydroxyvitamin D3 2MD

OH

HO

OH

2α-methyl-19-nor-(20S)-1,25-dihydroxyvitamin D3 2AMD

OH

HO

OH

2β-methyl-19-nor-(20S)-1,25-dihydroxyvitamin D3 2BMD

V. 2-METHYLENE AND 2α-METHYL AND 2β-METHYL DERIVATIVES OF 1,25(OH)2D3 In this laboratory, having learned that the diene system is required and having learned that side-chain modifications except the 20-epimerization diminishes in vivo activity, our focus on modification of 1,25(OH)2D3 turned to the A-ring. By removing the 10,19-methylene group, we learned that we could reduce in vivo calcium activity provided the side chain was not appreciably altered [11,21]. Finally, we began to consider possible derivatizations that could alter the activity of 1α- and 3βhydroxyl functionalities, which appeared to play an important role not only in binding to the VDR but in biological activity. Thus, the series of compounds, the most interesting of which are shown in Fig. 1, were chemically synthesized and studied extensively for biological activity. Of particular importance was the 2α-methyl derivative that appeared to lock the 1-hydroxyl in the axial configuration. This appeared to be interesting because there had been proposals that the binding form to the receptor required this hydroxyl to be in the equatorial form [22].

OH

HO

OH

FIGURE 1 2-Carbon-modified analogs of 1,25(OH)2D3. The 2-methylene-19-nor-(20S)-1α,25-dihydroxyvitamin D3 (2MD) and the 2α-methyl-19-nor-(20S)-1,25-dihydroxyvitamin D3 (2AMD) represent two potent bone selective analogs of 19-nor-1α,25-dihydroxyvitamin D3. The 2β-methyl compound (2BMD) has much weaker activity both in vitro and in vivo.

We, therefore, synthesized the following series of compounds: 2-methylene-19-nor-(20S) and (20R)-1α,25dihydroxyvitamin D3; 2α-methyl-19-nor-(20S) and (20R)-1α,25-dihydroxyvitamin D3; and 2β,19-nor-(20S) and (20R)-1α,25-dihydroxyvitamin D3 and studied their

1546

VI. 2-METHYLENE-19-NOR-(20S)1,25(OH)2D3 (2MD) AND 2α-METHYL19-NOR-(20S)-1,25(OH)2D3: ANALOGS THAT POSSESS ANABOLIC ACTIVITY ON THE SYNTHESIS OF BONE AND APPEAR TO BE BONE SELECTIVE ANALOGS OF 1,25(OH)2D3 The above studies indicated bone selectivity for 2MD and 2αMD. Because of the cost of pursuing both compounds, efforts were placed entirely on 2MD, but there is reason to believe that the 2α-methyl derivative would behave similarly. To pursue the bone selectivity further, Shevde et al. [24] studied osteoclastogenesis in vitro utilizing bone marrow culture. The results

ET AL .

A Serosal Ca (mg/dL)

12 10 8 6

Vehicle 2MD Calcitriol

4 2 0 0.0

0.1

1.0

10.0

Log [mcg drug/kg body weight]

B 12 Serosal Ca/mucosal Ca

biological activities. It became immediately apparent that the 2β-methyl derivative failed to produce an analog of interest since it failed to bind to the VDR and consequently had very low biological activity. However, the 2-methylene and 2α-methyl derivatives both proved to be extremely potent analogs of 1,25(OH)2D3. These derivatives bind to the VDR as well as 1,25(OH)2D3 [23]. They are also able to induce differentiation of HL-60 cells with the 20S derivatives being much more active than the 20R derivatives, both being at least equal or more active than 1,25(OH)2D3. These compounds also proved to be quite active in stimulating intestinal calcium transport; however, their activity in that regard ultimately appeared to approximate that of 1,25(OH)2D3 [24]. However, the major departure from the native hormone is with the 20S derivatives especially but also to some degree with the 20R derivatives. These compounds are much more active in mobilizing calcium from bone. One of these has been tested more fully in this system as shown in Fig. 2. As a result of a log dose study, it is clear that 2-methylene-19-nor-(20S)-1,25(OH)2D3 is equal to the activity of 1,25(OH)2D3 in intestinal calcium transport, but is approximately 30–100 times more active than 1,25(OH)2D3 in mobilization of calcium from bone. In this latter assay, animals are made both vitamin D–deficient and deprived of dietary calcium. Thus, in the absence of calcium, they are unable to synthesize new bone. The administration of vitamin D compounds under these circumstances causes calcium to come from bone to provide for serum calcium concentrations. This, therefore, represents an in vivo measurement of osteoclastic-mediated bone resorption. Thus, especially with the 2-methylene-19-nor(20S)-1,25(OH)2D3, we see true selectivity for activity on the skeleton in vivo. This study was then a forerunner of a series of 2-carbon analogs being developed in our laboratories.

HECTOR F. DELUCA,

10 8 6 Vehicle 2MD Calcitriol

4 2 0 0.0

0.1

1.0

10.0

Log [mcg drug/kg body weight]

FIGURE 2

2MD is approximately 30 times more potent than 1,25(OH)2D3 (calcitriol) in the mobilization of calcium from (A) bone and is equally potent in stimulating (B) intestinal calcium transport. In these experiments, vitamin D–deficient rats were provided the indicated doses i.p. daily for four days and on the 5th day, serum calcium determinations were made and intestinal calcium transport determinations were performed by the everted sac technique. The rise in serum calcium must arise from the skeleton because of the lack of calcium in the diet. This activity is via the stimulation of osteoblasts to produce RANKL that activates osteoclastogenesis.

illustrate that 2MD has an activity two orders of magnitude greater than 1,25(OH)2D3 in osteoclastogenesis, presumably by inducing the synthesis of RANKL. This is truly a remarkable increase in biopotency for a bone system. This led to the natural question of whether all activities of the osteoblast might be stimulated by 2MD. Thus, Shevde et al. [24] incubated 2MD with human osteoblast cultures taken from discarded bone sutures from pediatric cases. In this system, the osteoblasts were incubated with various concentrations of 2MD for 7 days and then incubated with β-glycerol phosphate and ascorbic acid to cause bone nodule formation. The results shown in Fig. 3 show clearly that 2MD is not simply two orders of magnitude more active because it shows this activity whereas 1,25(OH)2D3 under the

1547

CHAPTER 87 2-Carbon-Modified Analogs

A

B

Control

Control

1,25(OH)2D3 (10−8 M)

1,25(OH)2D3 (10−8 M)

2MD (10−12 M)

2MD (10−12 M)

FIGURE 3 2MD induces potent mineralization in osteoblasts in vitro. Primary human osteoblasts were isolated and cultured at a density of 3 × 105 cells/ml. Confluent cultures were treated with 1,25(OH)2D3 or 2MD at the concentrations indicated in the figure on days 0, 3, and 6 followed by treatment with ascorbic acid and β-glycerophosphate on days 9 and 12. Cells were stained on day 14 using the Von Kossa staining technique to detect the presence of calcified matrix/bone. The dark brown to black stain is indicative of calcified bone nodules. Von Kossa-stained cultures are indicated on the left (A) and microscopic images (10 ×) are shown on the right (B).

same circumstances has little or no activity. Thus, in an in vitro system, this analog demonstrates marked bone anabolic activity. Because of the bone selectivity, experiments were initiated early on to determine if 2MD could increase bone mass in the ovariectomized, aged female rat retired breeder. By now, more than five in vivo experiments have been conducted, all with similar results [24]. A representative example is shown in Fig. 4. After 23 weeks of treatment at 5–7 ng/kg/body weight, 2MD causes a 10% increase in total body bone mineral density. In data not shown, this increase is in both

trabecular and cortical bone, each showing in the case of the femur, a 25% increase in trabecular and cortical bone. Bone strength measurements show that this skeleton has the expected strength as related to bone mineral density. Finally, bone histomorphometry has demonstrated that this increase in bone mass is the result of increased bone synthesis. As a result, 2MD is now under development as a pharmaceutical for the treatment of postmenopausal osteoporosis. We believe this compound can be used wherever bone synthesis is required, i.e., osteoporosis or fracture healing or prosthesis with the skeleton.

1548

HECTOR F. DELUCA,

ET AL .

% Increase compared to OVX-Vehicle

12

#

10

# 8 6

*

4 2 0 Sham

5.5 ng 2MD/kg daily

7.8 ng 2MD/kg daily

35 ng 2MD/kg 2X/week

53 ng 2MD/kg 1X/week

555 ng 1,25(OH)2D3/kg 3X/week

Values shown are the average of 8–12 animals/group. Error bars represent standard error. Statistically significant (p< 0.05) values are denoted by pound signs (different from both sham and OVX-vehicle control animals) or asterisks (different from OVX-vehicle control animals).

FIGURE 4 One-year-old female retired breeder rats obtained from Sprague Dawley were either sham-operated or ovariectomized. After 5 wk, the animals were given the indicated oral doses of 2MD or 1,25(OH)2D3 for a period of 23 weeks. The percent increase of total body bone mineral density in the animal as compared to the ovariectomized controls is plotted. Clearly, 5–6 ng of 2MD/kg/day produced an astounding 8–9% increase in total body bone mineral density. Twice weekly dosing was less effective and even less effective was once-a-week dosing. Notice that a very much larger dose of 1,25(OH)2D3 given three times a week only increased the bone mass to about the level of the sham-operated control and thus gave a very modest percent increase over the ovariectomized control. Bone histomorphometry and other measurements have indicated a 25% increase in trabecular bone, and this increase was not the result of an inhibition of resorption but rather resulted from a significant increase in new bone formation.

VII. 2-METHYLENE-19-NOR(20S)-1,25(OH)2D3: MOLECULAR MECHANISMS OF TISSUE SELECTIVITY AND ENHANCED POTENCY The 2 carbon-modified analogs of 19-nor-1,25− (ΟΗ)2D3 and the many additional analogs discussed in various chapters in this book show tremendous potential for the effective treatment of many diseases. Indeed, 2MD is currently in development as a treatment for osteoporosis, and other analogs are also being developed for a variety of other indications. Despite this, the mechanisms that underlie the unique actions of most of these compounds relative to altered potency and/or efficacy, or in some cases tissue selectivity in vivo remain largely unknown. The fundamental unanswered question is whether these analog-specific patterns of biological activity observed in vivo are due to differences in analog activity within target cells (Chapter 83), or are due to unique in vivo differences in compound metabolism, pharmacokinetic behavior, and/or tissue localization (Chapters 81 and 82). Efforts are currently underway in various laboratories to answer these questions for each individual analog. With respect to 2MD, it and likely other 2 carbonmodified 1,25(OH)2D3 analogs are exceptionally

active in the skeleton in vivo, and appear to favor activities that lead to striking net bone formation rather than the modest or, in some cases, neutral effects that 1,25(ΟΗ)2D3 exhibits on bone. As indicated earlier in this chapter, although the overall mechanism through which 2MD functions in this anabolic fashion is unknown, perhaps the most informative discovery is the finding by Shevde et al. [24] that 2MD exerts a unique and highly potent mineralizing effect on differentiating osteoblasts in vitro at concentrations well below those that efficiently stimulate the formation and activation of osteoclasts (Fig. 3 and ref. 24). Since the osteoblast plays a central role in both bone forming and bone resorbing activities and is the principal target of vitamin D action in bone [25], these data suggest that 2MD may activate a gene or genes in osteoblasts that are unique relative to 1,25(ΟΗ)2D3. Studies are in progress to identify these gene targets and to assess their role(s) in bone formation both in vitro and in vivo. Perhaps the most obvious feature of 2MD, and indeed a characteristic typical of several analogs of vitamin D under current investigation, is the compound’s increased biological potency both in vitro and in vivo. Surprisingly, the biological potencies of many vitamin D compounds do not reflect their demonstrated biochemical affinities for the VDR (see Chapters 82–85). This is indeed true for 2MD, which,

1549

CHAPTER 87 2-Carbon-Modified Analogs

seen in the ligand binding assays, however, 2MD is seen in Fig. 5 to display a two log increase in potency in its ability to promote VDR binding to each of the above gene promoters. Further studies revealed that the accumulation of both VDR and RXR on these gene A 40 Relative intensity

despite its increased biological potency, binds to the VDR in vitro with an equilibrium dissociation constant almost equal to that of 1,25(ΟΗ)2D3 [23]. One possible explanation for 2MD’s increased potency in vivo as well as in vitro is the finding that 2MD binds weakly, if at all, to the vitamin D–binding protein (DBP), a serum macromolecule generally believed to transport vitamin D metabolites in the blood (see Chapter 8). This feature (lack of DBP binding) could facilitate the uptake of 2MD into target cells, thus increasing the analog’s apparent potency by promoting elevated tissue localization. Indeed, this appears to be the case experimentally, as 2MD quickly disappears from the blood and rapidly localizes within tissues that include the cells of bone [26]. Lack of DBP binding, however, may also lead to an increase in the metabolism of 2MD, thereby reducing or altering its overall biological profile. Key in vivo studies do not appear to support this possibility, however, suggesting that perhaps 2MD is carried in the blood by other protein components. Despite these potential explanations for the enhanced potency of 2MD in vivo, this analog also stimulates the expression of a number of genes in vitro in cultured cells at very low concentrations, irrespective of serum DBP content. This finding suggests the possibility of a molecular/cellular component to explain the potency of 2MD. The above studies prompted Yamamoto et al. [27] to investigate the mechanism of action of 2MD in osteoblastic cells in vitro and to contrast the analog’s effects with those of 1,25(ΟΗ)2D3. Initial studies explored the possibility that while the relative affinity of 2MD was identical to that observed for 1,25(ΟΗ)2D3 in broken cell preparations, the binding of 2MD to the VDR might differ from that of 1,25(ΟΗ)2D3 in intact living cells. Accordingly, osteoblasts were incubated with radiolabeled 1,25(ΟΗ)2D3 or 2MD and the uptake of the two ligands and their association with the VDR was measured. Interestingly, both time dependent uptake as well as binding of each of the ligands to the VDR was similar, ruling out the possibility that the uptake of the two ligands into the intact cell was different, or that the relative affinities of 1,25(ΟΗ)2D3 and 2MD were different. These results prompted a further comparison of the ability of the two ligands to induce accumulation of the VDR and its heterodimeric partner RXR on specific target genes in living cells using chromatin immunoprecipitation assays. Cells were incubated with increasing concentrations of either 1,25(ΟΗ)2D3 or 2MD and the binding of both VDR and RXR to the vitamin D response elements (VDRE) located in the promoters of genes for both the 25 hydroxyvitamin D3-24 hydroxylase (Cyp24) and osteopontin (OPN) was assessed. Unlike the results

1,25D3 2MD

30 20 10 0 NT

−11

−10

−9

−8

−7

(log M)

NT

B

1,25D3

2MD

αVDR αRXR

No Ab

Input

NT

C

1,25D3

2MD

αVDR αSRC1 No Ab

Input

FIGURE 5 2MD potently stimulates binding of VDR to active promoters in intact cells. A, MC3T3-E1 cells were incubated under serum-free conditions for two hrs with increasing concentrations of 1,25(OH)2D3 or 2MD and then subjected to chromatin immunoprecipitation (ChIP) analysis using anti-VDR or anti-RNA pol II antibodies. Isolated DNA was subjected to PCR using primers designed to amplify a 196 bp fragment of the OPN promoter from −854 to −658 that contained the VDRE at −764 to −748. The data represent a densitometric quantitation of the resulting PCR fragments. B and C, cells were treated with either vehicle, or 0.1 nM 1,25(OH) 2D3 or 2MD for 45 min. and then subjected to ChIP using antibodies to VDR, RXR, or SRC-1 as indicated. The resulting DNA was amplified as in A, and the PCR products are shown directly.

1550 promoters also resulted in secondary consequences. Both SRC-1 and DRIP205, coactivators necessary for transcriptional activation of Cyp24 and OPN, were also recruited to the two genes in response to 2MD in a highly potent manner. Importantly, these data correspond directly with 2MD’s ability to induce the expression of each of these genes in osteoblasts, and indeed other gene targets [27]. These findings suggest that despite similar pools of both 2MD- and 1,25(ΟΗ)2D3bound VDR, those receptors linked to 2MD display an increased capacity to localize on the regulatory elements present in the promoters for Cyp24 and OPN and to activate the expression of those genes. These observations suggest the possibility that 2MD might induce an unusual conformation within the VDR that favors increased accumulation or increased retention of the receptor on VDREs. Can this unique conformational state induced by 2MD be detected in vitro? We evaluated this possibility with both X-ray diffraction studies using purified rat VDR [28] and by subjecting individual ligand receptor complexes to limited tryptic digestion in vitro [27]. Interestingly, while the digestion studies observed in Fig. 6 indicate that 2MD is able to promote a VDR conformation that is differentially proteolyzed relative to 1,25(ΟΗ)2D3, this unique structure is not evident in the crystal structure of the rat VDR ligand binding domain in complex with either 2MD or 1,25(ΟΗ)2D3. Thus, the overall special organization of this portion of the VDR is similar, irrespective of whether it is bound to 1,25(ΟΗ)2D3 or 2MD. This finding is consistent with those of other investigators who have also failed to demonstrate a unique conformational difference within the ligand binding domain of the VDR via X-ray crystallography in the presence of superagonist analogs such as (20S)-1,25(ΟΗ)2D3 or KH1060 [29]. Nevertheless, these results again contrast with the positive results observed with such compounds in proteolytic digestion assays [30]. Thus, it is possible that visualization of these unique ligandinduced conformations at the atomic level will require three-dimensional structures of full-length VDR or cocrystallization of VDR with RXR or perhaps another of its interacting coactivator partners. Since synthetic ligands are known to alter the natural hormone-induced conformations of several different nuclear receptors [31], such discovery should come as no surprise when it is accomplished for the VDR. VDR binding to most VDREs in vitro requires RXR, which forms an active heterodimer in response to 1,25(ΟΗ)2D3 and participates with the VDR in direct DNA binding (see Chapters 11, 13, 18, and references therein). This requirement of the VDR for RXR in VDRE binding in intact cells has recently been

HECTOR F. DELUCA,

A

VDR-wt 1,25(OH)2D3

10% input

ET AL .

NT 5

10

2MD

20 5

10

20 min

34 32 28

B

VDR (R417A/R420A) 1,25(OH)2D3

10% input

NT −8

−7

2MD −8

−7

(log M)

34 32 28

FIGURE 6

2MD induces a unique VDR conformation. [35S]methionine-labeled hVDR or mutant hVDR(R417A/R420A) was incubated with the indicated concentrations of either 1,25(OH)2D3 or 2MD and then treated with trypsin for the indicated time (A) or 10 min (B). Proteins were resolved using SDS-PAGE and autoradiographed. Wildtype hVDR, upper: hVDR(R417A/R420A), lower.

confirmed [27,32]. VDR also functions together with its heterodimer partner to recruit additional protein complexes that are required for the eventual activation of gene expression (see Chapters 11, 13–17, and references therein). These known particulars of VDR action prompted us to explore the possibility that the increase in affinity of the VDR for inducible promoters such as Cyp24 or OPN in response to 2MD might occur as a result of enhanced interaction with either RXR, with coactivators such as DRIP205 or SRC-1, or both. Surprisingly, the dose-dependent abilities of 2MD and 1,25(ΟΗ)2D3 to induce a biochemical interaction between purified VDR and RXR, SRC-1 or DRIP were virtually identical in vitro [27]. This suggests that 2MD is incapable of promoting an enhanced interaction between the VDR and its various interacting partners. A completely different profile, however, was observed when these interactions were explored in intact cells using a mammalian two-hybrid system. In this system, the VDR was linked to the transactivation domain of VP16, and the VDR interacting domains of RXR, SRC-1, and DRIP205 were coupled to the DNA

1551

CHAPTER 87 2-Carbon-Modified Analogs

A

RXR

RLU/ β-gal (103)

12

NT 2MD 1,25(OH)2D3

8

4

0 NT −13

−12 −11

−10

−9

−8

−7

Ligand (log M)

DRIP205

B

RLU/ β-gal (103)

30

20

NT 2MD 1,25(OH)2D3

10

0

NT −13 −12 −11 −10

−9

−8 −7 −6

binding domain of the yeast protein Gal4. Accordingly, 2MD was again discovered as documented in Fig. 7 to be at least two logs more potent in its ability to promote an interaction between VDR and either RXR, SRC-1 or DRIP205. These results support the idea that the increased potency of 2MD resides in its ability to promote both VDR DNA binding and complex formation on active gene promoters. Despite the lack of support for this hypothesis in the VDR crystal structure, it still seems likely that this effect of 2MD is driven through a unique ligand-induced conformation imposed on the VDR during complex formation. The studies described above indicate that 2MD and perhaps other 2 carbon-modified 1,25(ΟΗ)2D3 analogs are highly potent and potentially gene-selective activators of transcription in bone cells. This balance between its highly potent and potentially gene-selective action in bone and its 1,25(ΟΗ)2D3-like activity in the gut makes it an excellent candidate for the therapeutic treatment of osteoporosis where normal intestinal absorption of calcium is required. As indicated above, however, while some understanding of the increased potency of 2MD and perhaps other analogs is emerging, additional work will be necessary to completely elucidate this facet of 2MD activity. Additional efforts will also be necessary to delineate its actions directly on the osteoblast to induce transcription in a gene- and perhaps cell-selective manner.

Ligand (log M)

VIII. 2-METHYLENE-19-NOR1α-HYDROXYPREGNACALCIFEROL (2MPREGNA), 2-METHYLENE-19-NOR-1αHYDROXYHOMOPREGNACALCIFEROL (2MP) AND 2-METHYLENE-19-NOR-(20S)1α-HYDROXYBISHOMOPREGNACALCIFEROL (2MBISP)

SRC1

C

RLU/ β-gal (103)

60

NT 2MD 1,25(OH)2D3

40 20 0 NT −13 −12 −11 −10 −9 −8 Ligand (log M)

−7

−6

FIGURE 7 2MD is a potent inducer of VDR/RXR, VDR/SRC-1 and VDR/DRIP205 interactions assessed by yeast 2-hybrid assays 6. MC3T3-E1 cells were transfected with expression vectors for chimeric VDR-VP16 and chimeric RXRα- (A), SRC-1(B) or DRIP205- (C) Gal4 DNA binding domain (DBD) fusions, and a multimerized Gal4 response element-luciferase reporter vector. Cells were treated with increasing concentrations of either 1,25(OH)2D3 or 2MD and luciferase activity assessed after 24 hr.

The compounds shown in Fig. 8 represent important new analogs that in the case of the bishomo compound 2MbisP and the pregnacalciferol compound 2Mpregna completely lack calcemic activity and have slight intestinal calcium transport activity in the case of the homopregnacalciferol compound (2MP). Note especially that these compounds completely lack the side chain structure of 1,25(OH)2D3 and have no 25-hydroxyl group, possessing only a short hydrocarbon side chain. These compounds bind very well to the VDR; they are slightly less active in binding to the receptor than 1,25(OH)2D3. They all possess activity in cellular differentiation, although not quite as effective as 1,25(OH)2D3. The results shown in Fig. 9 demonstrate that these compounds

1552

HECTOR F. DELUCA,

2-methylene-19-nor-1α-hydroxypregnacalciferol 2-Mpregna

OH

HO

2-methylene-19-nor-1α-hydroxyhomopregnacalciferol 2-MP

HO

OH

2-methylene-19-nor-(20S)-1α-hydroxybishomopregnacalciferol 2-MbisP

HO

OH

FIGURE 8 Structures of the noncalcemic 2-methylene-19-nor1α-hydroxyvitamin D derivatives that possess a markedly shortened hydrocarbon side chain without a side chain hydroxyl.

are able to suppress serum PTH levels in intact animals at doses wherein serum calcium is not increased. In fact, PTH suppression is found at less than 1 µg/kg and not shown here, doses as high as 66 µg/kg will not increase serum calcium. Although these compounds are relatively new on the scene, we believe they have promise of being truly noncalcemic analogs that may be used for the treatment of conditions where a rise in serum calcium is not desirable. Such conditions are the treatment of autoimmune disease, oral treatment of psoriasis, and the treatment of malignancy.

ET AL .

IX. 2-METHYLENE-19-NORPREGNACALCIFEROL (2MPREGNA), 2-METHYLENE-19-NOR-1αHYDROXYHOMOPREGNACALCIFEROL (2MP) AND 2-METHYLENE-19-NOR(20S)-1α-HYDROXYBISHOMOPREGNACALCIFEROL (2MBISP): MOLECULAR MECHANISMS OF TISSUE SELECTIVITY The most striking general feature of some but not all vitamin D analogs is their reduced ability or their apparent inability to induce hypercalcemia via normal physiological actions on the intestine, bone, and kidney while still retaining their ability to activate or suppress the expression of genes which may not be directly involved in calcium homeostasis. This feature stands in stark contrast to that of 1,25(ΟΗ)2D3, and suggests that some analogs may exhibit tissue- or geneselective activities. Clearly, analogs with this type of biological profile would be attractive for such therapeutic indications as autoimmune disease, psoriasis, or cancer, where an increase in serum calcium levels is not desirable and most likely detrimental. The 2-carbon-modified homopregnacalciferols appear to be such compounds. Accordingly, while preliminary studies (Plum and DeLuca, unpublished) indicate that they may promote cancer cell differentiation in vitro and are highly effective in suppressing serum PTH levels in vivo, as seen in Fig. 9, presumably by inhibiting the synthesis/ secretion of PTH from the parathyroid glands, they are correspondingly unable to induce hypercalcemia in normal animals. The underlying mechanism of this feature of these modified pregnacalciferols, as well as other vitamin D analogs with similar properties, is largely unknown. The general biochemical characteristics of these compounds relative to 1,25(ΟΗ)2D3 include a similarity in affinity for the VDR relative to 1,25(ΟΗ)2D3 and at least in cultured cells in vitro, a similarity in biopotency. The 2-carbon modified homopregnacalciferols are clearly not superagonists, however, and thus do not exhibit properties of increased potency such as those manifested by compounds such as 2MD. As with the 2 carbon-modified analogs of 19-nor1,25(ΟΗ)2D3 discussed earlier, the fundamental unanswered question with respect to the 2-carbon modified homopregnacalciferols is whether the apparent tissue selectivity seen in vivo is a function of unique pharmacokinetic properties that are different from those of 1,25(ΟΗ)2D3 or whether this selectivity is manifested at the level of the target cell. A significant limitation in answering this question resides

1553

CHAPTER 87 2-Carbon-Modified Analogs

PTH Serum Ca

12

100

10

80

8

60

6

40

4

20

2

0

Serum Ca (mg/dL)

% PTH suppression

120

0 Vehicle

1,25(OH)2D3 0.2 mcg/kg

2MP - 0.8 mcg/kg

2MP - 2 mcg/kg

2MP - 7 mcg/kg

2MbisP - 0.8 mcg/kg

2MbisP - 2 mcg/kg

2MbisP - 7 mcg/kg

2Mpregna - 2Mpregna - 2 2Mpregna - 7 0.7 mcg/kg mcg/kg mcg/kg

FIGURE 9 Biological potency of 2-MP, 2-MbisP, and 2-Mpregna in vivo. Adult female Sprague Dawley rats (6/group) were fed a 0.47% calcium, 0.3% phosphorus purified diet for one wk. At this point, the animals were given the dose indicated i.p. for seven consecutive days except 1,25(OH)2D3, which was given only for four days. All animals were continued on the 0.47% calcium, 0.3% phosphorus purified diet. Twenty-four hr after the last injection, animals were sacrificed and serum calcium determined by atomic absorption spectrometry. Serum PTH was measured using an ELISA assay (Immutopics, Inc.). The vertical bars represent standard errors of the mean.

both in the gaps that are currently present in our understanding of how calcium is absorbed from the intestine or resorbed from bone, as well as the limitations present in our understanding of the mechanisms by which 1,25(ΟΗ)2D3 regulates those processes (see Chapters 24, 25, and 38). Nevertheless, significant progress is being made in delineating both processes, and has revealed the central importance of both the intestinal epithelial calcium channel proteins (ECAC1 and 2) and calbindin 9K in the absorption of calcium from the gut, and osteoblast-produced receptor activator of NF-κB ligand (RANKL) as a signaling mediator of bone calcium resorption via the osteoclast. Importantly, 1,25(ΟΗ)2D3 regulates the expression of both genes. Accordingly, we explored the ability of the modified homopregnacalciferols to induced ECAC2 and calbindin 9K in the CaCO2 intestinal cell model and RANKL in the MC 3T3-E1 osteoblast cell model. Interestingly, both 1,25(ΟΗ)2D3 and 2MbisP are almost equipotent in their capacity to stimulate the expression of the three genes in their respective cell backgrounds (Watanuki et al., unpublished). With respect to intestinal calcium absorption, however, the failure of 2MbisP to support the transcription of other factors that may be essential to the calcium transport process could still result in the in vivo phenotype of insensitivity to the 2MbisP. Clearly, much more needs to be done before a clearer understanding of the underlying mechanism emerges.

X. SUMMARY The development of several important new analogs of 1,25(OH)2D3 are described wherein the 19-carbon is deleted and replaced by 2 hydrogens. The configuration of carbon-20 has been altered to the S-configuration, and carbon-2 has substituted either a 2-methylene or a 2α-methyl group. The resulting compounds are 2-methylene-19-nor-(20S)-1α,25(OH)2D3 (2MD) and 2α-methyl-19-nor-(20S)-1α,25(OH)2D3 (2AMD) and show bone calcium mobilizing activity 30 times that of the native hormone while showing equal activity in intestinal calcium transport and binding to the VDR. 2MD specifically stimulates osteoblast cultures to form new bone, a property minimally expressed in the case of 1,25(OH)2D3. Further, 2MD markedly increases bone mass of ovariectomized aged female rats while 1,25(OH)2D3 has minimal but significant activity. 2MD is currently being developed as an anabolic bone agent for the treatment of bone loss diseases such as osteoporosis. By replacing the side chain of 2MD with hydrocarbon side chains of 2–4 carbons without an hydroxyl group, true noncalcemic analogs have been produced. These compounds bind to the receptor almost as well as the native hormone, have significant cellular differentiation activity but lack the ability to raise serum calcium concentration either from bone or from intestine. These compounds are currently being studied for the

1554 treatment of disease where an elevation of calcium is undesirable. In vivo these compounds suppress the parathyroid gland while having no effect on serum calcium concentration. All of the above modified forms of the vitamin D hormone are fully active in transcriptional assays using the reporter gene attached to the 24-hydroxylase promoter and current assays suggest that 2MD causes the formation of a stable complex of VDR to VDREs and markedly improves the affinity of that complex for these response elements. Beyond this, little is known concerning the mechanism whereby 2-methylene19-nor-(20S)-1,25(OH)2D3 is not only more potent but much more selective for activity in bone than is 1,25(OH)2D3.

References 1. Perlman K, Kutner A, Prahl J, Smith C, Inaba M, Schnoes HK, DeLuca HF 1990 24-homologated 1,25-dihydroxyvitamin D3 compounds: Separation of calcium and cell differentiation activities. Biochemistry 29:190–196. 2. Ostrem VK, Tanaka Y, Prahl J, DeLuca HF 1987 24- and 26-Homo-1,25-dihydroxyvitamin D3: Preferential activity in inducing differentiation of human leukemia cells HL-60 in vitro. Proc Natl Acad Sci USA 84:2610–2614. 3. Ostrem VK, Lau WF, Lee SH, Perlman K, Prahl J, Schnoes HK, DeLuca HF, Ikekawa N 1987 Induction of monocytic differentiation of HL-60 cells by 1,25-dihydroxyvitamin D analogs. J Biol Chem 262:14164–14171. 4. Calverley MJ 1987 Synthesis of MC 903, a biologically active vitamin D metabolite. Tetrahedron 43:4609–4619. 5. Kubodera N, Watanabe H, Kawanishi T, Matsumoto M 1992 Synthetic studies of vitamin D analogs. XI. Synthesis and differentiation-inducing activity of 1α,25-dihydroxy-22-oxavitamin D3 analogs. Chem Pharm Bull (Tokyo) 40:1494–1499. 6. Uskokovic MR, Baggiolini E, Shiuey SJ, Iacobelli J, Hennessy B, Kiegel J, Daniewski AR, Pizzolato G, Courtney LF, Horst RL 1991 The 16-ene analogs of 1,25-dihydroxycholecalciferol, synthesis, and biological activity. In: AW Norman, R Bouillon, M Thomasset (eds) Vitamin D: Gene Regulation, Structurefunction Analysis and Clinical Application. De Gruyter: Berlin, pp. 139–145. 7. Kragballe K: Psoriasis and other skin diseases. In: D Feldman, FH Glorieux, JW Pike, (eds) 1997 Vitamin D. Academic Press: San Diego, pp. 1213–1225. 8. Binderup L, Kragballe K 1992 Origin of the use of calcipotriol in psoriasis treatment. Rev Contemp Pharmacother 3:401–409. 9. Jones G 1997 Analog metabolism. In: D Feldman, FH Glorieux, JW Pike (eds) Vitamin D. Academic Press: San Diego, pp. 973–994. 10. Brown AJ, Ritter CS, Finch JL, Morrissey J, Martin KJ, Murayama E, Nishii Y, Slatopolsky E 1989 The noncalcemic analog of vitamin D, 22-oxacalcitriol, suppresses parathyroid hormone synthesis and secretion. J Clin Invest 84:728–732. 11. Perlman KL, Sicinski RR, Schnoes HK, DeLuca HF 1990 1α,25-Dihydroxy-20-nor-vitamin D3, a novel vitamin D– related compound with potential therapeutic activity. Tetraehedron Lett 31:1823–1824.

HECTOR F. DELUCA,

ET AL .

12. Sicinski RR, Prahl JM, Smith CM, DeLuca HF 2002 New highly calcemic 1α,25-dihydroxy-19-norvitamin D3 compounds with modified side chain: 26,27-dihomo- and 26,27dimethylene analogs in 20S-series. Steroids 67:247–256. 13. Slatopolsky E, Finch J, Ritter C, Denda M, Morrissey J, Brown A, DeLuca H 1995 A new analog of calcitriol, 19-nor1,25(OH)2D2, suppresses parathyroid hormone secretion in uremic rats in the absence of hypercalcemia. Am J Kidney Dis 26:852–860. 14. Teng M, Wolf M, Lowrie E, Ofsthun N, Lazarus JM, Thadhani R 2003 Survival of patients undergoing hemodialysis with paracalcitol or calcitriol therapy. N Engl J Med 349: 446–456. 15. Sicinski RR, DeLuca HF 1994 Synthesis, conformational analysis, and biological activity of the 1α,25-dihydroxy-10, 19-dihydrovitamin D3 isomers. Bioorg Chem 22:150–171. 16. Sicinski RR, DeLuca HF 1995 Synthesis of 6,7-diaza19-nor-vitamin D compounds. Bioorg Med Chem Lett 4: 899–904. 17. Calverley MJ, Binderup E, Binderup L 1991 The 20-epi modification in the vitamin D series: Selective enhancement of “non classical” receptor-mediated effects. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D: Gene Regulation, Structure-Function Analysis and Clinical application. De Gruyter: Berlin, New York, pp. 163–164. 18. Hansen K, Calverley MJ, Binderup L: Synthesis and biological activity of 22-oxa vitamin D analogs. In: AW Norman, R Bouillon, M Thomasset (eds) 1991 Vitamin D: Gene Regulation, Structure-Function Analysis and Clinical Application. De Gruyter: Berlin, New York, pp. 161–162. 19. Kubodera N, Sato K, Nishii Y: Characteristics of 22-oxacalcitriol (OCT) and 2β-(3-hydroxypropoxy)-calcitriol (ED-71). In: D Feldman, FH Glorieux, JW Pike (eds) Vitamin D. Academic Press: San Diego, pp. 1071–1086. 20. Sicinski RR, Perlman KL, DeLuca HF 1994 Synthesis and biological activity of 2-hydroxy and 2-alkoxy analogs of 1α,25-dihydroxy-19-nor-vitamin D3. J Med Chem 37: 3730–3738. 21. Brown AJ, Slatopolsky ES 1997 Mechanisms for the selective action of vitamin D analogs. In: D Feldman, FH Glorieux, JW Pike (eds) Vitamin D. Academic Press: San Diego. pp. 995–1009. 22. Okamura WH, Norman AW, Wing RM 1974 Vitamin D: Concerning the relationship between molecular topology and biological function. Proc Natl Acad Sci USA 71:4194–4197. 23. Sicinski RR, Prahl JM, Smith CM, DeLuca HF 1998 New 1α,25-dihydroxy-19-nor-vitamin D3 compounds of high biological activity: Synthesis and biological evaluation of 2-hydroxymethyl, 2-methyl and 2-methylene analogs. J Med Chem 41:4662–4674. 24. Shevde NK, Plum LA, Clagett-Dame M, Yamamoto H, Pike JW, DeLuca HF 2002 A potent analog of 1α,25-dihydroxyvitamin D3 selectively induces bone formation. Proc Natl Acad Sci USA 99:13487–13491. 25. Takeda S, Yoshizawa T, Nagai Y, Yamato H, Fukumoto S, Sekine K, Kato S, Matsumoto T, Fujita T 1999 Stimulation of osteoclast formation by 1,25-dihydroxyvitamin D requires its binding to vitamin D receptor (VDR) in osteoblastic cells: studies using VDR knockout mice. Endocrinology. 140: 1005–1008. 26. Plum LA, Grzywacz P, Ma X, Lake E, Clagett-Dame M, DeLuca HF 2003 Distribution and metabolism of 2MD, an analog of 1-alpha,25(OH)2D3 with potent anabolic activity in bone. J Bone Min 18 (Suppl. 2), S296 (Abst).

CHAPTER 87 2-Carbon-Modified Analogs

27. Yamamoto H, Shevde NK, Warrier, A, Plum LA, DeLuca, HF, Pike JW 2003 2-Methylene-19-nor-1,25-dihydroxyvitamin D3 potently stimulates gene-specific DNA binding of the vitamin D receptor in osteoblast. J. Biol Chem 278:31756–31765. 28. Vanhooke JL, Benning MM, Bauer CB, Pike JW, DeLuca, JF 2004 Molecular structure of the rat vitamin D receptor ligandbinding domain complexed with 2-carbon substituted vitamin D3 hormone analogs and an LxxLL-containing coactivator. Biochemistry, 43:4101–4110. 29. Tocchini-Valentini G, Roche N, Wurtz JM, Mitschler A, Moras D 2001 Crystal structures of the vitamin D receptor complexed to superagonist 20-epi ligands. Proc Natl Acad Sci USA 98: 5491–5496.

1555 30. Peleg S, Sastry M, Collins ED, Bishop JE, Norman AW 1995 Distinct conformational changes induced by 20-epi analogs of 1α,25-dihydroxyvitamin D3 are associated with enhanced activation of the vitamin D receptor. J Biol Chem 270: 10551–10558. 31. Brzozowski AM, Pike AC, Dauter Z, Hubbard RE, Bonn T, Engstrom O, Ohman L, Greene GL, Gustafsson JA, Carlquist M 1997 Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 389:753–758. 32. Kim, S, Warrier A, Shevde NK, Pike JW. 2004 1,25Dihydroxyvitmain D3 stimulates dynamic vitamin D receptor/ retinoid X receptor DNA binding and coactivator recruitment in intact osteoblasts. J Bone Min Res, in press.

CHAPTER 88

Nonsteroidal Analogs PETER ORDENTLICH AND RICHARD A. HEYMAN X-Ceptor Therapeutics, Inc. San Diego, California

I. Introduction II. Nonsecosteroid Bisphenol Compounds III. Bisphenol Analogs as Selective Agonists of Mutant VDR

IV. Non-secosteroid CD Ring Modifications V. Perspectives References

I. INTRODUCTION

the premise that they manifest properties inherently different from that of the endogenous hormone or its secosteroid derivatives. These include minimal binding to DBP and reduced metabolism by the 24-hydroxylase. The unique structural features of these compounds that distinguish them from vitamin D secosteroid derivatives highlight the possibility that they will exhibit novel pharmacokinetic and pharmacodynamic profiles, which can be manipulated to create molecules with less calcemic effects than classic vitamin D analogs. This rationale is based on the well-established fact that synthetic ligands for nuclear hormone receptors can induce unique conformations in the receptor that result in altered tissue and gene specificity as compared to the endogenous steroid ligands [5,6]. Examples of this include tamoxifen and raloxifene, which are tissueselective modulators of estrogen receptor activity. There are very few reports of nonsecosteroidal vitamin D mimics reported in the literature, and most of the effort has been directed towards side-chain secosteroid analogs derived from vitamin D3. There are three classes of compounds with biological data published that may be considered nonsecosteroid vitamin D receptor agonists. These include nonsteroidal molecules based on a bisphenol scaffold (i.e., LG190119) [7], molecules in which the C and/or D ring of 1,25(OH)2D3 have been removed [8,9], and the secondary bile acid lithocholic acid (reviewed in Chapter 53; [10]) (see Fig. 2). A review of the structures and activity of the bisphenol and C/D ring modification class of compounds is included below.

Vitamin D is a steroid hormone identified originally as a preventative of the skeletal disease rickets (reviewed in [1]). In addition to key functions in bone health and development, vitamin D has been shown to play a role in regulating differentiation and proliferation of various cell types. The biologically active component of vitamin D is 1α,25-dihydroxyvitamin D3 (1,25(OH)2D3), which is synthesized through a series of tightly regulated metabolic steps in the liver and kidney (see Chapters 1–5) and binds to the nuclear hormone vitamin D receptor (VDR) to activate gene expression (see Chapters 11–22). 1,25(OH)2D3 functions to control calcium and phosphate levels through regulation of intestinal absorption of calcium and phosphate, reabsorption of calcium in the kidney, and mobilization of calcium from bone [2]. Circulating 1,25(OH)2D3 is bound to vitamin D binding protein (DBP), which may serve to transport it to sites of action and to extend its serum half-life (see Chapters 8 and 9). Additional regulation of 1,25(OH)2D3 activity occurs through a feedback loop that senses an increased level of 1,25(OH)2D3 and induces its own degradation via up-regulation of the enzyme 24-hydroxylase (24-OHase) (see Chapters 2 and 6). In the case of greater than physiological levels of vitamin D activity, excess calcium is absorbed via the intestine leading to hypercalcemia, which is the main side effect that limits the use of current vitamin D analogs in many disease indications such as cancer and osteoporosis. The development of vitamin D analogs that exhibit improved therapeutic efficacies stem from analogs that have a reduced binding to DBP, altered metabolic properties, improved pharmacodynamic profiles, and induction of unique receptor conformations [3,4] and Chapters 82 and 83. All but a few of the compounds generated to date are secosteroid analogs that are based on the vitamin D structure (Fig. 1). This chapter will highlight the development of small molecule vitamin D3 mimics whose design is based on VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

II. NONSECOSTEROID BISPHENOL COMPOUNDS A. Identification, Structure, and Synthesis Based on clinical success with estrogen receptor modulators, the identification of synthetic ligands for Copyright © 2005, Elsevier, Inc. All rights reserved.

1558

PETER ORDENTLICH AND RICHARD A. HEYMAN

25

20

through a cell based high-throughput screen of small molecule compound libraries in which VDR was contransfected with a reporter containing a VDR response element from the 24(OH)ase promoter [7]. The original screening hit was optimized for potency (40 nM) and efficacy (90%) to approach the activity of 1,25(OH)2D3 (2–5 nM; 100%). The bisphenols are to date the only VDR ligands with associated in vitro and in vivo published data that are not derived from the 1,25(OH)2D3 core structural scaffold. The characteristics of these compounds are reviewed below.

OH 17 13

D

C 14 8

H

6 19

A HO 3

B. In Vitro Characterization of Bisphenol Analogs

OH

1

1 , 2 5 ( OH )2D3

FIGURE 1

Structure of 1,25(OH)2D3. Schematic illustration of the 1,25(OH)2D3 molecule with the A-, C-, and D-rings marked, and carbon residues numbered for reference. The side chain referred to in the text and modified in many analogs to alter DBP binding is off of C17.

nuclear hormone receptors that mimic the activity of the endogenous hormone, while reducing their unwanted physiological side effects has been the therapeutic target for many drug discovery nuclear receptor programs. The bisphenol structure template was identified

The original screening hit, LG190090, was a bisphenol derivative with an EC50 of 2.2 µM and efficacy of 80 percent compared to 1,25(OH)2D3 (see Table I). The series of published structures indicate a significant improvement in potency (LG190178 at 40 nM) was obtained by modifications including addition of a diethyl group to the bridgehead of LG190090 and hydroxyl group substitution of one of the 3,3-dimethyl2-butanone groups. These compounds were shown to require VDR for activation and did not cross react with other nuclear hormone receptors. The more potent

COOH

O

O

O

HO

O

Lithocholic acid

LG190119

20 17

25

25

OH

14 8

H

1 OH

OH 17 14 8

H

C-Ring analogs

6 19

HO

3

25

13

6 19

3

OH

14 8

6

HO

17

13

13

20

1 OH

D-Ring analogs

19

HO

3

1 OH

E-Ring analogs

FIGURE 2 Structures of various nonsecosteroid vitamin D receptor ligands. The three types of nonsecosteroid molecules are represented by selected structures, including the bisphenols (LG190119), bile acids (lithocholic acid), and C/D ring modifications that disrupt the core secosteroid scaffold or introduce novel structural motifs (E-ring analogs).

1559

CHAPTER 88 Nonsteroidal Analogs

TABLE I Structures and Activities of Selected Bisphenol Nonsecosteriod Compounds Compound 1,25(OH)2D3

CTF1 EC50 (nM)

VDR binding2 Ki (nM)

LNCaP3 EC50 (nM)

HL604 EC50 (nM)

NHEK5 EC50 (nM)

DBP binding6 IC50 (nM)

2–5

0.5

2

8–10

10–30

200(h); 40(r)

2500

>10000

3000

10000

ND7

>10000

2200

>10000

2000

2000

500

>10000

600

>10000

300

800

500

>10000

40

150

20

30–50

30–50

>10000

LG190090 Cl

Cl

O

O

O

O LG190119

O

O

O

O LG190155

O

O

O

O LG190178

OH O O

O OH

1Cotransfection

(CTF) data was obtained using pRShVDR and VDRE(1)-∆MTV-Luc in HepG2 cells. binding affinity was determined by competition binding with yeast expressed hVDR and [3H]-1,25(OH)2D3. 3Potency of LNCaP growth inhibition determined by BrdU incorporation. 4Potency of HL60 differentiation determined by nitroblue tetrazolium assay. 5Potency of normal human epidermal keratinocytes (NHEK) growth inhibition determined by BrdU incorporation. 6Human (h) and rat (r) DBP binding determined by competition binding versus [3H]-1,25(OH) D . 2 3 7ND is not determined. 2VDR

compounds were shown to bind directly to VDR through ligand-binding assays in which 3[H]-1,25(OH)2D3 was displaced by an unlabeled competitor. Interestingly, the binding affinity of the compounds was less than that predicted by the cell-based functional assays, which may be attributed to a slow off-rate of 1,25(OH)2D3. Several cancer cell lines including prostate [11], breast [12,13], and colon cells [14] have been shown to respond to the antiproliferative activity of 1,25(OH)2D3, which is a consequence of cell cycle arrest [13,15] and in some cases programmed cell death [16]. Activity in cell lines is a significant marker that has served as the basis for further study of vitamin D analogs in animal tumor models. More potent antiproliferative activity relative to 1,25(OH)2D3 in these cell lines has been a prime objective in the design of improved vitamin D analogs in order to maximize their efficacy in vivo. Results from testing of the bisphenol compounds

indicate that they retain the antiproliferative activity that would be expected from a VDR agonist [7]. LG190178 inhibited proliferation of human SK-BR-3 breast cancer cells and LNCaP prostate cancer cells at concentrations ranging from 30–50 nM. The antiproliferative potency correlates well with the activity of the compounds in the VDR cotransfection assay (Table I). Further studies demonstrated these compounds could efficiently differentiate HL60 human leukemia cells into macrophages with the most potent compound, LG190178, having an EC50 value of 30 nM (Table I). Another important and well-established role for 1,25(OH)2D3 is the control of keratinocyte proliferation and differentiation [17,18]. To determine if the nonsecosteroid compounds could replicate this aspect of 1,25(OH)2D3 activity, compounds were added to human epidermal keratinocytes and cultured for seven days. Treatment with selected LG compounds resulted in growth inhibition and potencies

1560

PETER ORDENTLICH AND RICHARD A. HEYMAN

that were consistent with the rank order of the compounds in both cotransfection experiments and inhibition of proliferation in other cell types such as LNCaP and HL60. As in the other experiments, the most potent compound was LG190178. The morphology of the treated keratinocytes was also shown to be similar to that reported for 1,25(OH)2D3 [19], supporting the fact that synthetic nonsecosteroid molecules working through the VDR can mimic vitamin D activities. 1,25(OH)2D3 exposure to target tissues has been shown to partly depend on binding to DBP, which controls delivery to site of action and serum half-life [3,20]. Binding to DBP is therefore a major influence on the pharmacokinetic and pharmacodynamic properties of the vitamin D analogs. Structural evaluation of 1,25(OH)2D3 indicates that the regions of the molecule contributing to DBP binding include the C-17 side chain and the D-ring. A correlation has been established between diminished DBP binding and reduced calcemic activity, although there are exceptions such as the analog ED71, which appears to have enhanced DBP binding but reduced calcemic activity [21]. Much of the chemistry on secosteroids has focused on modifications to the C-17 side chain region of the molecule in order to reduce DBP binding. Advanced clinical leads such as EB1089 (Fig. 3 and Chapter 84) display significantly less binding to DBP compared to 1,25(OH)2D3 . One prediction of nonsecosteroids is significantly reduced DBP binding, and binding studies with the bisphenol compounds support this rationale, as none of the compounds tested displayed any detectable DBP binding (Table I) [7].

C. In vivo Characterization of Bisphenol Analogs Many vitamin D secosteroid analogs exhibit promising in vitro profiles, but in vivo retain their ability

OH

HO

OH

FIGURE 3 Structure of EB1089. Schematic illustration of the vitamin D3 analog EB1089 (Seocalcitol).

to induce hypercalcemia. The bisphenol Vitamin D mimics were therefore assayed in vivo to determine their ability to induce VDR target genes and the potential for separation of calcemic activity from transcriptional activity. In the first report [7], LG190119 (10 mg/kg and 30 mg/kg) was administered orally to BALB/c mice and the serum calcium and kidney 24(OH)ase RNA levels were measured. In this study, LG190119 induced kidney 24-OHase expression approximately tenfold at a dose of 30 mg/kg, administered daily for 3–5 days by oral gavage without a significant increase in serum calcium. These results were confirmed in a second study in which (10 mg/kg; 5 day p.o.), LG190119 gave a 30-fold increase in 24(OH)ase levels with no significant increase in serum calcium. In the same experiment, 1,25(OH)2D3 induced 24(OH)ase 25-fold (5 µg/kg; 5 days p.o.), but significantly increased serum calcium. The most potent compound, LG190178, was not included in these studies. The results demonstrate that based on the comparison of induction of 24(OH)ase mRNA levels to plasma calcium levels, the nonsecosteroid bisphenol compounds may possess an improved therapeutic index. This conclusion assumes that 24(OH)ase mRNA serves as a suitable biomarker for VDR activity, and indicates a separation of VDR target gene expression from calcium induction. Further work has examined the ability to translate the antiproliferative activity of these compounds on various cancer cell lines in vitro to in vivo effect in tumor models. These studies [22] were carried out using LG190119 in a prostate cancer model in which LNCaP xenograft tumors are established in athymic mice. Treatment with LG190119 (dosed every other day at 10 mg/kg over a 12 week period) inhibited the growth of these tumors (as measured by tumor volume) without significantly increasing serum calcium levels. The activity of LG190119 was compared to 1,25(OH)2D3 and EB1089, a less calcemic vitamin D analog currently in clinical trials for treatment of various cancers and shown to be threefold more potent in inhibiting LNCaP cell growth than 1,25(OH)2D3 [15,23]. The efficacy results with LG190119 compared favorably to EB1089, which although inhibiting tumor growth, in these studies was hypercalcemic (dosed at 1 µg/kg), and 1,25(OH)2D3 which in these studies did not have an effect on tumor growth at the dose tested (0.5 µg/kg). In summary, the nonsecosteroid bisphenol template has been shown to bind to and transactivate VDR, activate VDR target genes, and inhibit growth of cancer cells. Animal studies with these compounds demonstrate separation of gene induction and calcium increase indicating that a nonsecosteroidal core with inherently unique pharmacologic and molecular properties may translate into vitamin D mimics with

1561

CHAPTER 88 Nonsteroidal Analogs

improved therapeutic profiles in a number of indications, including cancer.

III. BISPHENOL ANALOGS AS SELECTIVE AGONISTS OF MUTANT VDR Hereditary vitamin D resistant rickets (HVDRR) is an autosomal recessive disease characterized by vitamin D resistance in target organs even in the presence of high levels of circulating 1,25(OH)2D3 (see Chapter 72 for a detailed review). The resistance to vitamin D was shown to result from mutations affecting expression and function of VDR. Several missense mutations have been described in the LBD of VDR that affect the binding affinity of 1,25(OH)2D3 leading to impaired transcription of target genes. In a series of elegant experiments [24], Koh and co-workers have applied molecular modeling and structural based rational design to develop vitamin D mimics that would be complementary to the R274L mutation in the VDR ligand-binding domain and activate the mutant receptor efficiently. In addition to further validating this approach for ligand identification, compounds that could activate the mutant receptor with minimal side effects could potentially be useful as treatments for these patients. The options for a starting template for the identification of analogs that could activate mutant VDRs included derivatives of 1,25(OH)2D3 or selecting a nonsecosteroid molecule. The bisphenol scaffold was chosen for several reasons, including reduced calcemic activity compared to 1,25(OH)2D3, and the ease of synthesis of multiple derivatives. Based on the crystal structure of 1,25(OH)2D3 with VDR, computer modeling predicted that the nonsecosteroid LG190155 would

fill the ligand-binding pocket such that one of the 3,3-dimethyl-2-butanone groups is able to hydrogen bond to His305, the amino acid residue contacted by the C22-C25 side chain of 1,25(OH)2D3. This orientation of LG190155 allows the carbonyl oxygen of the other 3,3-dimethyl-2-butanone group to hydrogen bind to the Ser278 and Tyr143 side chains. These amino acid side chains are bound by the 3β-hydroxy group of 1,25(OH)2D3. Computer modeling of LG190155 further predicted that unlike 1,25(OH)2D3, the molecule would not interact directly with Arg274, the amino acid contacted by the 1α-OH of 1,25(OH)2D3. This prediction was supported by functional assays indicating that LG190155 could activate the R274L mutant VDR with approximately the same potency and efficacy as wild-type VDR. Through further use of computer-aided rational design, the nonsecosteroid scaffold was modified such that one of the 3,3-dimethyl-2-butanone groups was substituted with a limited set of alkylating agents. The model of LG190155 bound to the pocket allowed for the selection of 40 potentially useful substitutions, which after additional computational assessment were narrowed down to 13 optimal analogs. This approach to ligand identification was shown to be successful as 7 of the 13 compounds had activity on the R274L VDR better than LG190155 in terms of potency in cotransfection assays with the best compound, A-13 having EC50 of 3.3nM (Fig. 4). As mentioned previously, the bisphenol nonsecosteroid scaffold was chosen for these studies based on ease of synthesis and less calcemic activity as compared to 1,25(OH)2D3. The analogs optimized for activation of R274L were tested for their ability to stimulate Ca2+ influx in preosteoblastic cells. The results indicated that

O O

O

O

O

O O

O LG190155 wtVDR EC50 110 nM mutVDR EC50 85 nM

FIGURE 4

A-13 Koh mutVDR EC50 3.3 nM

Structures of LG190155 and A-13. The structure and activity of the most potent analog, A-13, on the R274L mutant VDR is shown. The parent compound, LG190155, on the wild-type and R274L mutant VDR is also included for reference. The EC50s were determined by transient cotransfection of HEK293 cells with a luciferase reporter plasmid with wild-type or R274L mutant VDR and treatment with compound for 36 hours. It should be noted, that the value of LG190155 is different from that in Table I (110 nM vs. 600 nM), which may be due to differences in the assay format including cell type and response element used in the cotransfection experiments.

1562 while 1,25(OH)2D3 stimulated Ca2+ by 50%, the analogs did not show any significant Ca2+ influx activity. While these data remain to be verified in vivo, they highlight the promise of using a nonsecosteroid VDR agonist that can be quickly and easily optimized, through computer aided drug design, to mutant forms of the receptor while retaining an intrinsically low calcemic activity shown previously for this molecular template.

IV. NON-SECOSTEROID CD RING MODIFICATIONS One approach to identifying novel molecules has been to randomly screen small molecule libraries with a variety of cellular and biochemical based assays. In the case of VDR, the bisphenols are the only published structural class found to date using this approach (see Section II). An alternative approach has been to make modifications of the existing natural ligands that retain the desired biological activities while minimizing side effects. A large number of analogs of the vitamin D molecule have been synthesized through this approach, with relatively few demonstrating improved properties compared to the parent 1,25(OH)2D3. A more radical form of structural modification to the core ring structure of the 1,25(OH)2D3 molecule was carried out to identify regions of the molecule that could be further modified to optimize therapeutic activity. These modifications to the C- and D-rings disrupt the secosteroid structure, and are reviewed here along with the generation of a new structural component referred to as an E-ring [8,9].

A. C- and D-ring Analogs Modifications to the core secosteroid structure were made to explore the role of the C- and D-rings in the maintenance of proper conformation of the A-ring and the C-17 side chain of the vitamin D molecule. Interestingly, the work from Verstuyf et al. [8,9] identified a series of potent and efficacious nonsecosteroidal VDR ligands that lacked either the C- or D-ring in certain cases or introduced a novel 5 membered “E”-ring as a substitution for the C/D rings (Fig. 5). In order to generate these compounds, an acyclic parent compound (KS 018) was made which when tested did not have any VDR activity or binding to DBP. Re-introduction of only the C-ring with the 1,25(OH)2D3 side chain recovered most of the VDR binding activity for most of the analogs reported (see Table II; ZG1368 and CY625) as compared to 1,25(OH)2D3, while maintaining low to no DBP binding.

PETER ORDENTLICH AND RICHARD A. HEYMAN

In the cell lines tested, the antiproliferative activity of C-ring analogs was generally increased with some compounds being 30–60 fold greater than 1,25(OH)2D3 (ZG1310; ZG1441; ZG1423). The serum calcium induction by the C-ring analogs was reduced by 100-fold for some of the active compounds as compared with 1,25(OH)2D3. Additional modifications to these compounds led to a reduced VDR binding affinity (i.e., 19-nor-A ring), or increased antiproliferative activity (i.e., movement of 25-OH group to C-24(R); C-16 to C-17 double bond; C-23 to C-24 triple bond). In order to evaluate further the structural requirements of the C- and D-ring in the core vitamin D secosteroid template, the D-ring alone was reintroduced into the acyclic backbone molecule (Table III). As with the C-ring analogs, the nonsteroidal D-ring compounds had greatly reduced binding to DBP (<10% of 1,25(OH)2D3), although they retained significant VDR affinity (40–80% of 1,25(OH)2D3). The D-ring analogs were generally less potent in antiproliferative assays compared to similar C-ring molecules, except for WU515 (Table III) and WU507 in which fluorination of C-26 and C-27 led to increased activity that was 10–50 percent more potent than 1,25(OH)2D3 depending on cell type.

B. E-ring Analogs Through the exploration of the structural requirements of the C- and D-rings, it was found that placement of a novel 5 member ring linking C13 to C20 could substitute for C- and D-rings and retain some vitamin D activity (Table IV). Referred to as an E-ring, this structural motif is apparently sufficient to introduce the proper spacing and conformation restraints on the A/secoB domains and side chain of 1,25(OH)2D3. Analogous to the structural changes introduced in the C- and D-ring analogs, the most potent E-ring molecules were generated by modifications to the side chain including fluorination of C26 and C27 (CD 503; CD 504; CD 483) or elongation at the C26, C27, and C24 positions. The most active compounds, including CD 503, CD 504, and CD483, had 2–8 times greater antiproliferative activity and less than 5 percent the calcemic activity of 1,25(OH)2D3. Interestingly, introduction of double and triple bonds between C23 and C24 reduced VDR binding affinity to 10% of 1,25(OH)2D3 (compare CD 504 and CD 483 to KS 176), while retaining twice as much antiproliferative activity in MCF-7 cells. The effects of C26,C27 hexafluorination and unsaturation of the side chain in the E-ring analogs are consistent with the effects of the same

1563

CHAPTER 88 Nonsteroidal Analogs

25

20

OH

17 13 14 8 6 19 A HO

1 OH

3

KS 018

20

25

20 OH

8

E 17

17

13

13

8

H

14

H 6

19

19

19

A 1 OH

ZG 1368

HO 3

OH

8

6

A

25

13

D 14

14

6

HO

20 OH

17 C

25

A 1 OH

KS 176

HO 3

1 OH SL 117

FIGURE 5

Identification of C-, D-, and E-ring modifications of 1,25(OH)2D3. The 1,25(OH)2D3 molecule was reduced to a backbone structure that contained the A-ring and C17 side chain (KS018). The C- and D-rings were reintroduced separately to determine their relative contribution to the conformation and positioning of the A-ring and C-17 side chain. Representative molecules ZG 1368 (C-ring), and KS176 (D-ring) are shown. Also shown is SL 117, a representative of E-ring substituted molecules in which a novel 5membered ring replaces the C/D ring of 1,25(OH)2D3.

modifications made to 1,25(OH)2D3. The antiproliferative activity of the E-ring analogs, combined with a relative lack of calcemic activity, provides these compounds with a potentially improved therapeutic profile over most secosteroid analogs, particularly in the treatment of various cancer types. It will be important to determine if these compounds retain their potent in vitro antiproliferative activities when tested in vivo and what effect the ring modifications may have on the pharmacodynamic and metabolic properties of these compounds. The approach described above complements well the search for nonsecosteroidal vitamin D mimics in that fundamental changes to the core structure that generate nonsteroidal molecules result in improved

ratios of antiproliferative effects to calcemic activity. Verstuyf et al. [8,9] investigated the underlying bases for why and how the newly generated C/D ring and E-ring analogs could maintain vitamin D-like activity while minimizing their calcemic potential. The poor binding of all of the analogs to DBP can explain in part the reduced calcemic activity, as all but a few of the vitamin D analogs made to date that exhibit some separation of differentiation and calcemic activity also bind poorly to DBP. As discussed previously, DBP binding contributes to the serum half-life and tissue exposure of vitamin D analogs. Other factors examined included metabolism, receptor affinity, and induction of altered receptor conformations. Only the latter demonstrated a strong correlation with improved

1564

PETER ORDENTLICH AND RICHARD A. HEYMAN

TABLE II Compound 1,25(OH)2D3

Structures and Activities of Selected C-ring Modifications VDR binding1 (%)

DBP binding2 (%)

HL603 (%)

MCF-74 (%)

Kerat5 (%)

Serum Ca+2;6 (%)

100

100

100

100

100

100

60

20

1000

6000

6000

50

45

3

1000

2000

5000

13

80

5

150

200

200

1

70

10

3500

6000

6000

35

35

2

3000

1000

3750

34

OH C

ZG 1368

A OH

HO

OH C

ZG 1423 A OH

HO

OH C CY 625 A OH

HO

OH

C

ZG 1310 V

A OH

HO

OH

C

ZG 1441 B A HO

OH

1Relative percent VDR binding affinity compared to 1,25(OH) D (100%; 0.13 nM) was determined as the concentration of compound needed to 2 3 displace 50% of [3H]-1,25(OH)2D3 from pig intestinal mucosa expressed VDR. 2Relative percent affinity to hDBP (1,25(OH) D at 100%; 220 nM) determined by competition binding versus [3H]-1,25(OH) D 2 3 2 3. 3Relative percent activity of HL60 differentiation measured by nitroblue tetrazolium assay with 100% (29 nM) indicating the EC50 of 1,25(OH) D . 2 3 4Relative percent activity of MCF-7 growth inhibition determined by [3H]thymidine incorporation with 100% (37 nM) indicating the EC50 of 1,25(OH)2D3. 5Relative percent activity of keratinocyte growth inhibition determined by [3H]thymidine incorporation with 100% (5.3 nM) indicating the EC50 of 1,25(OH)2D3. 6Relative percent of serum calcium levels in mice after compound treatment with 100% (14.7 mg/dl) indicating 1,25(OH) D at 0.4 µg/kg/day for 7 days. 2 3

1565

CHAPTER 88 Nonsteroidal Analogs

TABLE III Compound 1,25(OH)2D3

Structures and Activities of Selected D-ring Modifications VDR binding1 (%)

DBP binding2 (%)

HL603 (%)

MCF-74 (%)

Kerat5 (%)

Serum Ca+2;6 (%)

100

100

100

100

100

100

80

10

85

85

90

70

3

1000

5000

3000

OH D

0.3

SL 117 A OH

HO

D

6

WU 515

A HO

CF3 OH CF3

OH

1Relative percent VDR binding affinity compared to 1,25(OH) D (100%; 0.13 nM) was determined as the concentration of compound needed to dis2 3 place 50% of [3H]-1,25(OH)2D3 from pig intestinal mucosa expressed VDR. 2Relative percent affinity to hDBP (1,25(OH) D at 100%; 220 nM) determined by competition binding versus [3H]-1,25(OH) D 2 3 2 3. 3Relative percent activity of HL60 differentiation measured by nitroblue tetrazolium assay with 100% (29 nM) indicating the EC50 of 1,25(OH) D . 2 3 4Relative percent activity of MCF-7 growth inhibition determined by [3H]thymidine incorporation with 100% (37 nM) indicating the EC50 of 1,25(OH)2D3. 5Relative percent activity of keratinocyte growth inhibition determined by [3H]thymidine incorporation with 100% (5.3 nM) indicating the EC50 of 1,25(OH)2D3. 6Relative percent of serum calcium levels in mice after compound treatment with 100% (14.7 mg/dl) indicating 1,25(OH) D at 0.4 µg/kg/day for 7 days. 2 3

antiproliferative activity, in that these compounds were able to induce a receptor conformation that was more resistant to protease digestion than 1,25(OH)2D3. These findings support the notion that synthetic nonsteroidal analogs or mimics of the endogenous receptor ligands could induce novel and distinct receptor conformations and that these in turn would lead to altered biological properties.

V. PERSPECTIVES The application of classic secosteroid VDR ligands to the treatment of many diseases including cancer and osteoporosis has been hampered by the hypercalcemic activity that is observed at the effective pharmacologic doses of these compounds. The identification and characterization of nonsecosteroidal vitamin D mimics indicates that the development of synthetic molecules can lead to differentially regulated VDR activity and

suggests a promising approach to the identification of vitamin D functional analogs with an improved therapeutic index. Surprisingly, among the hundreds of vitamin D analogs that have been generated and reported, there have been very few reports of nonsecosteroid vitamin D ligands. Indeed, there is only one reported structural class, the bisphenol, that is not based on the core 1,25(OH)2D3 template. Recently, another group of nonsecosteroid vitamin D mimics has been disclosed in patent applications (WO 0138320, WO 0138303, WO 02094754), but were not reviewed here due to a lack of available published in vitro and in vivo data. Other nonsteroidal molecules have been generated through gross structural changes to the 1,25(OH)2D3 molecule in which the C- and D-rings have been eliminated and/or replaced with a novel 5 member E-ring. Nonetheless, based on a molecular and physiological understanding of vitamin D action, these nonsecosteroid compounds display potency and efficacy comparable to 1,25(OH)2D3 with reduced calcemic activity.

1566

PETER ORDENTLICH AND RICHARD A. HEYMAN

TABLE IV Compound 1,25(OH)2D3 E

Structures and Activities of Selected E-ring Modifications VDR binding1 (%)

DBP binding2 (%)

HL603 (%)

MCF-74 (%)

Kerat5 (%)

Serum Ca+2;6 (%)

100

100

100

100

100

100

10

19

20

30

10

< 0.1

80

40

100

850

900

3

65

9

80

550

500

2

10

3

25

215

400

0.2

OH

KS 176 A HO

OH

E

CF3 OH CF3

CD 503 A OH

HO

E

CF3 OF CF3

CD 504 A OH

HO

E

CF3 OH CF3

CD 483 A HO

OH

1Relative percent VDR binding affinity compared to 1,25(OH) D (100%; 0.13 nM) was determined as the concentration of compound needed to 2 3 displace 50% of [3H]-1,25(OH)2D3 from pig intestinal mucosa expressed VDR. 2Relative percent affinity to hDBP (1,25(OH) D at 100%; 220 nM) determined by competition binding versus [3H]-1,25(OH) D 2 3 2 3. 3Relative percent activity of HL60 differentiation measured by nitroblue tetrazolium assay with 100% (29 nM) indicating the EC50 of 1,25(OH) D . 2 3 4Relative percent activity of MCF-7 growth inhibition determined by [3H]thymidine incorporation with 100% (37 nM) indicating the EC50 of 1,25(OH)2D3. 5Relative percent activity of keratinocyte growth inhibition determined by [3H]thymidine incorporation with 100% (5.3 nM) indicating the EC50 of 1,25(OH)2D3. 6Relative percent of serum calcium levels in mice after compound treatment with 100% (14.7 mg/dl) indicating 1,25(OH) D at 0.4 µg/kg/day for 7 days. 2 3

CHAPTER 88 Nonsteroidal Analogs

Reasons for the improved biological profile of nonsecosteroid compounds include greatly reduced binding to DBP, potentially altered metabolic properties in the case of the bisphenols, and the induction of distinct and unique receptor conformations. These properties can result in altered tissue and gene specificity as compared to the endogenous steroid ligands so that nonsecosteroid VDR ligands represent a scaffold upon which novel compounds can be developed which exhibit unique molecular, pharmacological, and pharmacodynamic properties with the potential for an improved therapeutic index with respect to currently available vitamin D analogs.

Acknowledgments We thank Dr. Raju Mohan for helpful discussion and comments.

References 1. Bikle DD, Pillai S 1993 Vitamin D, calcium, and epidermal differentiation. Endocr Rev 14:3–19. 2. Bouillon R, Van Cromphaut S, Carmeliet G 2003 Intestinal calcium absorption: Molecular vitamin D–mediated mechanisms. J Cell Biochem 88:332–339. 3. Brown AJ 2000 Mechanisms for the selective actions of vitamin D analogs. Curr Pharm Des 6:701–716. 4. Stein MS, Wark JD 2003 An update on the therapeutic potential of vitamin D analogs. Expert Opin Investig Drugs 12:825–840. 5. Brzozowski AM, Pike AC, Dauter Z, Hubbard RE, Bonn T, Engstrom O, Ohman L, Greene GL, Gustafsson JA, Carlquist M 1997 Molecular basis of agonism and antagonism in the estrogen receptor. Nature 389:753–758. 6. McDonnell DP, Connor CE, Wijayaratne A, Chang CY, Norris JD 2002 Definition of the molecular and cellular mechanisms underlying the tissue-selective agonist/antagonist activities of selective estrogen receptor modulators. Recent Prog Horm Res 57:295–316. 7. Boehm MF, Fitzgerald P, Zou A, Elgort MG, Bischoff ED, Mere L, Mais DE, Bissonnette RP, Heyman RA, Nadzan AM, Reichman M, Allegretto EA 1999 Novel nonsecosteroidal vitamin D mimics exert VDR-modulating activities with less calcium mobilization than 1,25-dihydroxyvitamin D3. Chem Biol 6:265–275. 8. Verstuyf A, Verlinden L, van Etten E, Shi L, Wu Y, D’Halleweyn C, Van Haver D, Zhu GD, Chen YJ, Zhou X, Haussler MR, De Clercq P, Vandewalle M, Van Baelen H, Mathieu C, Bouillon R 2000 Biological activity of CD-ring modified 1alpha,25-dihydroxyvitamin D analogs: C-ring and five-membered D-ring analogs. J Bone Miner Res 15:237–252.

1567 9. Verstuyf A, Verlinden L, Van Baelen H, Sabbe K, D’Hallewyn C, De Clercq P, Vandewalle M, Bouillon R 1998 The biological activity of nonsteroidal vitamin D hormone analogs lacking both the C- and D-rings. J Bone Miner Res 13:549–558. 10. Makishima M, Lu TT, Xie W, Whitfield GK, Domoto H, Evans RM, Haussler MR, Mangelsdorf DJ 2002 Vitamin D receptor as an intestinal bile acid sensor. Science 296:1313–1316. 11. Johnson CS, Hershberger PA, Trump DL 2002 Vitamin D– related therapies in prostate cancer. Cancer Metastasis Rev 21:147–158. 12. Colston KW, Chander SK, Mackay AG, Coombes RC 1992 Effects of synthetic vitamin D analogs on breast cancer cell proliferation in vivo and in vitro. Biochem Pharmacol 44:693–702. 13. Colston KW, Hansen CM 2002 Mechanisms implicated in the growth regulatory effects of vitamin D in breast cancer. Endocr Relat Cancer 9:45–59. 14. Shabahang M, Buras RR, Davoodi F, Schumaker LM, Nauta RJ, Uskokovic MR, Brenner RV, Evans SR 1994 Growth inhibition of HT-29 human colon cancer cells by analogs of 1,25-dihydroxyvitamin D3. Cancer Res 54: 4057–4064. 15. Skowronski RJ, Peehl DM, Feldman D 1995 Actions of vitamin D3, analogs on human prostate cancer cell lines: comparison with 1,25-dihydroxyvitamin D3. Endocrinology 136:20–26. 16. Blutt SE, McDonnell TJ, Polek TC, Weigel NL 2000 Calcitriol-induced apoptosis in LNCaP cells is blocked by overexpression of Bcl-2. Endocrinology 141:10–17. 17. Bikle DD, Tu CL, Xie Z, Oda Y 2003 Vitamin D–regulated keratinocyte differentiation: role of coactivators. J Cell Biochem 88:290–295. 18. Bikle DD, Gee E, Pillai S 1993 Regulation of keratinocyte growth, differentiation, and vitamin D metabolism by analogs of 1,25-dihydroxyvitamin D. J Invest Dermatol 101:713–718. 19. Jones KT, Sharpe GR 1994 Intracellular free calcium and growth changes in single human keratinocytes in response to vitamin D and five 20-epi-analogs. Arch Dermatol Res 286:123–129. 20. Bouillon R, Allewaert K, Xiang DZ, Tan BK, van Baelen H 1991 Vitamin D analogs with low affinity for the vitamin D– binding protein: enhanced in vitro and decreased in vivo activity. J Bone Miner Res 6:1051–1057. 21. Kubodera N, Tsuji N, Uchiyama Y, Endo K 2003 A new active vitamin D analog, ED-71, causes increase in bone mass with preferential effects on bone in osteoporotic patients. J Cell Biochem 88:286–289. 22. Polek TC, Murthy S, Blutt SE, Boehm MF, Zou A, Weigel NL, Allegretto EA 2001 Novel nonsecosteroidal vitamin D receptor modulator inhibits the growth of LNCaP xenograft tumors in athymic mice without increased serum calcium. Prostate 49:224–233. 23. Blutt SE, Polek TC, Stewart LV, Kattan MW, Weigel NL 2000 A calcitriol analog, EB1089, inhibits the growth of LNCaP tumors in nude mice. Cancer Res 60:779–782. 24. Swann SL, Bergh J, Farach-Carson MC, Ocasio CA, Koh JT 2002 Structure-based design of selective agonists for a ricketsassociated mutant of the vitamin D receptor. J Am Chem Soc 124:13795–13805.

CHAPTER 89

Vitamin D: Cancer and Differentiation JOHANNES P. T. M. VAN LEEUWEN AND HUIBERT A. P. POLS Department of Internal Medicine, Erasmus MC, 3000 DR Rotterdam, The Netherlands

I. II. III. IV.

Introduction Vitamin D and Cancer Vitamin D Effects on Tumor Cells Combination Therapy

I. INTRODUCTION The seco-steroid hormone 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] is the most potent natural metabolite of vitamin D3 and is an important regulator of calcium homeostasis and bone metabolism via actions in intestine, bone, kidney, and parathyroid glands. 1,25-(OH)2D3 exerts its effects via an intracellular receptor that is a member of the steroid hormone receptor family (see Chapters 11–20 and 22 in this book). Throughout the last decades, it has become evident that the vitamin D receptor (VDR) is not limited to cells and tissues involved in regulation of calcium and bone metabolism but is also present in a wide variety of other cells and tissues including cancer cells of various origins. This led to a vast series of studies on the role of vitamin D in tumor cell growth regulation, treatment of cancer, and development of potent synthetic vitamin D analogs. Various specialized chapters will discuss in detail the effect of vitamin D on specific cancers (Chapters 89–97) and development and actions of vitamin D analogs (Chapters 80–88). In this chapter we aim to give an overview of the history and current stage and developments on vitamin D and cancer, regulation of tumor cells, possible mechanisms, and clinical applications.

II. VITAMIN D AND CANCER A. Vitamin D Receptor As exemplified in Table I, the VDR has also been demonstrated in a broad range of tumors and malignant cell types. For colon and breast cancer cells, an inverse relationship between VDR level and degree of differentiation has been described by some investigators [1,2]. VDR level is increased in ovarian carcinoma compared to normal ovarian tissue [3]. For colorectal cancer it was shown that VDR expression is associated with a more favorable prognosis in colorectal cancer [4]. A VDR immunoreactivity score showed an increase in VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Resistance and Vitamin D Metabolism VI. Stimulation of Proliferation VII. Conclusions References

breast carcinoma specimens compared to normal breast tissue but no clear relation with proliferative status could be assessed [5]. A later study by the same group showed that VDR expression is not a prognostic factor for breast cancer, but the strong VDR immunoreactivity in the breast cancer specimens supports the evidence for it to be a target for intervention [6]. Also in other studies no associations between VDR and clinical and biochemical parameters of breast cancer were found [7–12]. Albeit that the association studies on VDR expression and predictive and/or prognostic characteristics for cancer are so far not conclusive, the widespread distribution of the VDR in malignant cells indicates that regulation of cancer cell function might be a new target in the action of 1,25-(OH)2D3 and provides a biological basis for the epidemiological observations discussed in the next paragraph. A recent observation put the VDR in relation to cancer in a whole new perspective. It was shown that VDR can function as a receptor for the secondary bile acid lithocholic acid. This compound is hepatotoxic and a potential enteric carcinogenic. Interestingly, both binding of lithocholic acid and vitamin D to the VDR results in induction of CYP3A, the enzyme that detoxifies lithocholic acid in the liver and intestine [13,14]; (see also Chapter 53). It is postulated that vitamin D and lithocholic acid, by binding to the VDR, activate a feed-forward catabolic pathway that increases CYP3A expression leading to detoxification of carcinogenic bile acids. A relation between the presence of VDR and carcinogenesis was recently also shown for the skin. Absence of VDR increased the sensitivity for chemically induced tumorigenesis [15].

B. Epidemiology In 1980 an epidemiological study based on indirect evidence suggested a relationship between vitamin D and cancer. This was derived from analyses of death Copyright © 2005, Elsevier, Inc. All rights reserved.

1572

JOHANNES P. T. M. VAN LEEUWEN AND HUIBERT A. P. POLS

TABLE I VDR in Tumors and Malignant Cell Types Basal cell carcinoma Breast carcinoma Bladder cancer Cervical carcinoma Colonic adenocarcinoma Colorectal carcinoma Gall bladder carcinoma Glioma cells Kaposi sarcoma Lung carcinoma Lymphocytic leukemia Malignant B-cell progenitors Malignant melanoma Medullary thyroid carcinoma

Myeloid leukemia Myeloma Osteogenic sarcoma Ovarian carcinoma Neuroblastoma Non-Hodgkin’s lymphoma Pancreatic carcinoma Parathyroid adenoma Pituitary adenoma Prostate carcinoma Renal cell carcinoma Squamous cell carcinoma Transitional cell bladder carcinoma Uterine carcinosarcoma

rates from colon cancer, which tended to increase with increasing latitude and decreasing sunlight [16]. Later more direct evidence about a relation between vitamin D and colon cancer came from the inverse relationship between levels of serum 25-hydroxyvitamin D3 [a 1,25-(OH)2D3 precursor] and incidence of colonic cancer [17,18]. In addition, a similar relationship between sunlight exposure, vitamin D, and the risk for fatal breast and prostate cancer has been suggested [19–23] (see Chapter 90). The relationship between sunlight exposure and cancer, especially with respect to vitamin D, has been carefully reviewed by Studzinski and Moore [24]. The dual relationship between sunlight and cancer is of interest and remains the subject of continuing studies [25–27]. A relationship between skin type and prostate cancer has been described [28–30] and recently an article on the skin, sunlight, vitamin D, and cancer has been presented from an evolutionary perspective [31]. The relationship between cancer, diet, and calcium intake and vitamin D has been addressed in several studies [32–37] (see Chapter 91). A Canadian study noted similar vitamin D intakes in breast cancer patients and control subjects [38]. Moreover, in a mouse model, no relationship was found between dietary intake of a wide range of doses of calcium or vitamin D and carcinogeninduced skin tumors [39]. A large Finish epidemiological study showed an association of low serum 25-hydroxyvitamin D3 with prostate cancer [40,41]. A study on intake of micronutrients suggested that vitamin D and calcium might interact with antioxidants like vitamin C and E in reducing colorectal cancer risk [42]. It is clear that sunlight exposure, vitamin D intake, and other

dietary components such as calcium and fat should be considered as possibly interacting with one another when the relationship between vitamin D and cancer risk is assessed. The data on VDR as bile acid sensor and its postulated role in detoxification provide a direct biological basis for the relation between increased colon cancer and high-fat diets [43] and that colon cancer occurs in areas with higher prevalence of rickets [36]. In addition, mice lacking VDR have been reported to have a higher proliferation rate in the colon [44,45]. A survey of mutations in the VDR in osteosarcomas, several other sarcomas, nonsmall cell lung cancers, and a large number of cell lines representing many tumor types did not show that mutations or rearrangements in the VDR gene play a role in these cancers [46]. Aspects on sunlight and the epidemiology of vitamin D and calcium will be further discussed in greater detail in Chapters 90 and 91, respectively. In the VDR gene several polymorphisms have been identified and studied in relation to various endpoints (discussed in Chapter 68). Throughout the last years, an increasing number of studies have studied the association of polymorphisms in the VDR and cancer. The first study showed an association between polymorphisms at the 3′ end of the VDR gene and prostate cancer [47]. This was shortly followed by a study showing an association of prostate cancer with variations in the 3′ poly-A stretch in the VDR gene [48]. Interestingly, the Odds Ratio for the VDR polymorphism was about twofold that of the one for the CAG repeat in the androgen receptor. This was followed by several others studies also showing associations of polymorphisms in the 3′ region of the VDR gene and prostate cancer, [49–55] albeit other studies couldn’t confirm this [56–60]. For breast cancer both presence [61–66] and absence of association [67] with polymorphisms in the VDR gene have been reported. Also for colon cancer both presence [68,69] and absence [70] of an association with VDR polymorphisms have been reported. No association was reported with basal cell carcinoma [71]. A single study reported an association with the aggressive renal cell carcinoma [72], malignant melanoma [73], and another study on rectal cancer reported a correlation between VDR gene polymorphisms and erbB-2/HER-2 expression [74]. It should be concluded that so far the studies on VDR gene polymorphisms and cancer are far from conclusive. A major reason might be the limited size of most of the studies. More association studies on VDR gene polymorphisms and specific cancers are needed, which should be followed by a meta-analysis to definitively assess whether there is an association and if so, what is the size of the effect. Also, for studies on VDR gene polymorphisms, it is important to take into account the

1573

CHAPTER 89 Vitamin D: Cancer and Differentiation

impact of environmental factors. Diet, vitamin D intake, and sun exposure may modify the association of polymorphism and cancer risk. Interaction between vitamin D and calcium intake and cancer was also found in some of the VDR gene polymorphism studies [68,75–77]. Some studies reported decreased risk of prostate cancer [75] and colorectal adenomas [76] in those subjects with lower vitamin D levels and a particular VDR gene polymorphism. However, results of these studies are unusual in light of the fact that higher calcium and vitamin D intake are generally associated with a modestly reduced risk of colorectal neoplasia. Finally, most importantly it should be realized that except for the FokI translational start site polymorphism, all polymorphisms analyzed so far are anonymous, and functionality or linkage with functional polymorphisms should be proven. The 3′ polymorphisms have been shown to be in linkage with 3′-UTR polymorphisms, but no relation with VDR mRNA stability could be proven [78]. Detailed discussion of possible functional consequences of VDR gene polymorphisms and impact of vitamin D levels is beyond the scope of this chapter but will be addressed in Chapter 67.

the in vivo observation that 1,25-(OH)2D3 prolongs the survival time of mice inoculated with myeloid leukemia cells [86]. As shown in Table II, over the years 1,25-(OH)2D3 has been shown to have beneficial effects in several other in vivo animal models of various types of cancers [87–109]. An important aspect and limitation of the treatment of cancer with 1,25-(OH)2D3 was revealed by this limited set of clinical trials (see section II.D); to achieve growth inhibition, high doses are needed (confirming the in vitro data), which can cause the side effect of hypercalcemia. This has prompted the development of analogs of 1,25-(OH)2D3 in order to dissociate the antiproliferative effect from the calcemic and bone metabolism effects (see Chapters 80–88) [110,111]. Although the precise mechanism is not completely understood, at the moment several 1,25-(OH)2D3 analogs are available that seem to fulfill these criteria. In Table III the in vivo animal studies using 1,25-(OH)2D3 analogs on various cancer types are summarized [97,103,104,106–109,112–129].

D. Clinical Studies C. Growth and Development In addition to the epidemiological studies and demonstration of vitamin D receptor in tumor cells, since the early 1980s there has also been an increasing amount of cell biological data supporting a role for vitamin D in cancer. Multiple studies have shown that at high concentrations (10−9–10−7 M) 1,25-(OH)2D3 inhibits the growth of tumor cells in vitro. It was demonstrated as early as 1981 that 1,25-(OH)2D3 inhibits the growth of malignant melanoma cells and stimulates the differentiation of immature mouse myeloid leukemia cells in culture [79–81]. 1,25-(OH)2D3 also induces differentiation of normal bone marrow cells (see Chapter 96). Immature bone marrow cells of the monocyte-macrophage lineage are believed to be the precursors of osteoclasts, and 1,25-(OH)2D3 induces differentiation of immature myeloid cells toward monocytes-macrophages and also stimulates the activation and fusion of some macrophages (discussed in Chapter 38). From these results, it has been postulated that 1,25-(OH)2D3 stimulates differentiation and fusion of osteoclast progenitors into osteoclasts [82–84]. Also, in the intestine, 1,25-(OH)2D3 has important effects on cellular proliferation and differentiation [85]. Thus, via stimulation of the differentiation inducing capacity of bone and interstitial cells, 1,25-(OH)2D3 may play an important role in the regulation of calcium and bone metabolism. These in vitro findings were followed by

Considering the calcemic actions of 1,25-(OH)2D3 up to this point in time only a few clinical trials of vitamin D compounds in cancer have been performed. Alfacalcidol (1α-hydroxyvitamin D3; 1α-(OH)D3), which is converted to 1,25-(OH)2D3 in vivo, caused a beneficial response in low-grade non-Hodgkin’s lymphoma patients [130,131]. Also, with alfacalcidol, transient improvement in peripheral blood counts was seen in patients with myelodysplasia; however, half of the patients developed hypercalcemia [132]. Another study reported a sustained hematological response in six myelodysplasia patients treated with high doses of alfacalcidol [133]. These patients were restricted in their dietary calcium intake; nevertheless, four patients developed hypercalcemia due to increased bone resorption. With respect to treatment of cutaneous T-cell lymphoma with a combination of 1,25(OH)2D3 and retinoids, contrasting results have been obtained. It has been suggested that the variability was due to differences in phenotype of the various lymphomas [134–138]. A study on early recurrent prostate cancer showed that daily treatment with 1,25-(OH)2D3 slowed the rise in prostate-specific antigen, but treatment coincided with hypercalcemic affects [139]. Using a regime of weekly treatment with high-dose calcitriol was found to be safe, but didn’t result in a significant reduction in prostate-specific antigen (PSA) in prostate cancer cells [140]. Two studies were specifically designed to examine the route of application and calcemic response in patients with advanced malignancies [141,142].

1574

JOHANNES P. T. M. VAN LEEUWEN AND HUIBERT A. P. POLS

TABLE II Tumor Adenocarcinoma Breast Colon

Kaposi sarcoma Leydig tumor Lung

Melanoma Osteosarcoma Prostate

Retinoblastoma

Walker carcinoma Skin

In Vivo Effects of 1,25-(OH)2D3 and 1α-(OH)D3 in Animal Models of Cancera Model

CAC-8 cells injected in nude mice NMU- and DMBA-induced breast cancer in rats Human colon cell line implanted into nude mice; DMH-induced colon cancer in rats; APCmin mice KS Y-1 cells implanted in nude mice Leydig cell tumor implanted into rats Implantation of lewis lung carcinoma into mice

Human melanoma cells implanted into nude mice Human osteosarcoma cells implanted into nude mice Dunning MAT LyLu rat prostate model; LNCaP xenografts in nude mice; PAIII tumors in Lobund-Wistar rats Retinoblastoma cell line implanted into nude mice; transgenic mice with retinoblastoma Walker carcinoma cells injected in rats DMBA/TPA-induced skin tumors in mice

Effect

Refs.

Reduction in tumor volume Tumor suppression

[107] [93,96]

Tumor suppression; reduction of the incidence of colon adenocarcinomas; decrease in polyp number and tumor load Tumor growth retardation Tumor suppression Reduction of the number of metastases (without suppression of primary tumor); tumor suppression; increased antitumor immunity Tumor suppression

[90,92,95,371]

Tumor suppression

[98]

Reduction in lung metastasis; tumor suppression

[103,104,106,108,109]

Tumor suppression

[91,94]

Tumor suppression Inhibition of tumor formation

[100] [88,89]

[105] [97] [87,99,101,102]

[90]

aThe dosage, duration of treatment, diet, and effects on serum/urinary calcium vary among the studies. NMU, Nitrosomethylurea; DMBA, 7,12dimethylbenz[a]anthracene; DMH, 1,2-dimethylhydrazine dihydrochloride; TPA, 12-O-tetradecanoylphorbol-13-acetate.

Clinical trials using vitamin D analogs have been initiated over the last years. However, these were mostly limited clinical trials focusing on small groups of patients for whom regular treatment has failed. Only data from a few studies has been published. The analog calcipotriol (MC903) has been used for topical treatment of advanced breast cancer; however, several of the patients still developed hypercalcemia [143]. More recent studies have been published on advanced breast cancer [144] and pancreatic cancer [145] but the clinical results were limited. In a single case of Kaposi sarcoma and topical application of calcipotriol (Daivonex/Dovonex/MC903), good success in tumor regression was reported [105]. In Chapter 97 the current clinical status of 1,25-(OH)2D3 and its analogs as therapeutic agents for cancer will be discussed in greater detail.

E. Angiogenesis and Metastasis For the tumor suppressive activity of vitamin D3 compounds in vivo, besides growth inhibition, two

additional actions may be involved. First, angiogenesis is an essential requirement for the growth of solid tumors. Compounds that inhibit angiogenesis might therefore contribute to antitumor therapy. Antiangiogenic drugs may cause inhibition of tumor progression, stabilization of tumor growth, tumor regression, and prevention of metastasis. Antiangiogenic effects may play a role in the tumor suppressive activity of vitamin D3 compounds. Two studies reported an antiangiogenic effect of 1,25-(OH)2D3 and the analog 22-oxacalcitriol using different experimental model systems [115,146]. In addition, it was shown that 1,25-(OH)2D3 inhibits angiogenesis induced by the human papilloma virus type 16 (HPV16)- or HPV18-containing cell lines HeLa. Skv-e2, and Skv-el2 when intradermally injected into immunosuppressed mice [147]. Also, with the nonvirus-transformed human cell lines T47-D (breast carcinoma) and A431 (vulva carcinoma), similar results were obtained [148]. In these studies the mice were treated for 5 days with 1,25-(OH)2D3 prior to the injection of tumor cells. The effect of 1,25-(OH)2D3 on angiogenesis may be due to inhibition of tumor cell proliferation, resulting in fewer angiogenic cells.

1575

CHAPTER 89 Vitamin D: Cancer and Differentiation

TABLE III Analog

In Vivo Effects 1,25-(OH)2D3 Analogs in Animal Models for Cancer a

Model

1,25-(OH)D2 1,25-(OH)D5 CB966 CB1093

Retinoblastoma Breast Breast Prostate

DD-003 EB1089 EB1089 EB1089 EB1089 EB1089 EB1089

Colon Adenocarcinoma Breast Colon Hepatocellular carcinoma Leydig cell tumor Prostate

KH1060 LG190119 OCT OCT OCT OCT MC903 Ro 23-7553 Ro 23-7553 Ro 24-5531 Ro 24-5531 Ro-25-6760 Ro-26-9114 Ro-26-9114

Prostate Prostate Breast Breast Breast Colon Breast Prostate Leukemia Breast Colon Prostate Colon Prostate

Antitumor effect Tumor suppression Tumor suppression Tumor suppression Tumor suppression No effect on angiogenesis Tumor suppression Tumor suppression Tumor suppression Tumor suppression Inhibition of tumor incidence Tumor suppression Tumor suppression Reduction lung metastases No effect on angiogenesis Tumor suppression Tumor suppression Tumor suppression Tumor suppression Tumor suppression Decreased tumor incidence Tumor suppression Tumor suppression Increased survival Decreased tumor incidence Decreased tumor incidence Tumor suppression Decrease in polyp number and tumor load Tumor suppression

Refs. [128] [129] [114] [108] [120] [107] [114,116,125,316] [124] [372] [97] [104,106,108,109,126,127] [109] [106] [113,118] [115] [118] [121] [117] [122] [112] [119] [123] [103] [371] [109]

a MC903, 1,24-dihydroxy-22-ene-24-cyclopropyl-vitamin D ; CB966, 24a,26a,27a-tri-homo-1α,25-dihydroxyvitamin D ; CB1093, 20-epi-22(S)3 3 ethoxy-23yne-24a, 26a,27a-trihomo-1α,25-dihydroxyvitamin D3; DD-003,22(S)-24-homo-26,26,26,27,27,27-hexafluoro-1α,22,25-trihydroxyvitamin D3; EB1089, 22,24-diene-24a,26a,27a-trihomo-1α,25-dihydroxyvitamin D3; OCT, 22-Oxacalcitriol; Ro 23-7553, 1,25-dihydroxy-16-ene-23-yne-vitamin D3; Ro 24-5531, 1,25-dihydroxy-16-ene-23-yne-26,27-hexafluorovitamin D3. Ro 26-9114, 1α,25-(OH)2-16-ene-19-nor-24-oxo-D3.

However, inhibition of angiogenesis could also be observed when the tumor cells were treated in vitro with 1,25-(OH)2D3 and, after cell washing, were injected into mice [148]. Under these conditions both control and 1,25-(OH)2D3-treated mice were injected with similar numbers of cells. Therefore, these data indicate that 1,25-(OH)2D3 inhibits the release of angiogenic factors (vascular endothelium growth factor, transforming growth factor-α, basic fibroblast growth factor, epidermal growth factor, etc.) or stimulates antiangiogenic factors. 1,25-(OH)2D3 treatment caused a reduction in the angiogenic signaling

molecule, angiopoietin-2 in squamous cell carcinoma and radiation-induced fibrosarcoma-1 cells [149]. In retinoblastomas in mice, 1,25-(OH)2D3 has also been shown to reduce angiogenesis [150]. A recent study by Oades et al., however, showed that the 1,25(OH)2D3 analogs EB1089 and CB1093 inhibited tumor growth in two prostate animal models but did not inhibit angiogenesis in a rat aorta assay [108]. Whether this indicates that vitamin D affects angiogenesis in a tumor situation and not in a nonmalignant condition is not clear. This may resemble the effects of endostatin, which inhibits pathological but not normal

1576 vascularization [151,152]. In support of this is the finding that 1,25-(OH)2D3 and its analogs EB1089, Ro-25-6760, and ILX23-7553 potently inhibit growth of endothelial cells derived from tumors, but are less potent against normal aortic or yolk sac endothelial cells [149]. Finally, an interesting observation is deglycosylated vitamin D–binding protein (DBP-maf) has also been reported to inhibit angiogenesis [153,154] and to inhibit growth of pancreatic tumor in nude mice [154]. Whether 1,25-(OH)2D3 may interfere with DBP-maf in tumor growth inhibition and antiangiogenesis remains to be established. Interaction with another factor, interleukin-12, in the inhibition of angiogenesis has been reported [155]. The second mechanism of antitumor activity, which may be related to angiogenesis, is metastasis. Metastasis is the primary cause of the fatal outcome of cancer diseases. A study by Mork Hansen et al. indicated that 1,25-(OH)2D3 may be effective in reducing the invasiveness of breast cancer cells [156]. They showed that 1,25-(OH)2D3 inhibited the invasion and migration of a metastatic human breast cancer cell line (MDA-MB-231) using the Boyden chamber invasion assay. In support of this, it was shown that 1,25-(OH)2D3, KH1060, EB1089, and CB1093 inhibited secretion of tissue-type and urokinase plasminogen activator and increased plasminogen activator inhibitor 1 in the MDA-MB-231 metastatic breast cancer cells [157]. In an in vivo study, it was shown that 1,25-(OH)2D3 reduces metastasis to the lung of subcutaneously implanted Lewis lung carcinoma cells [101]. In two animal models of prostate cancer, 1,25-(OH)2D3 and the analogs EB1089 and RO25-6760 inhibited lung metastases [103,104]. In these models, the tumors were implanted subcutaneously and therefore, in contrast to the model of direct tumor cell injection in the left ventricle [158], no bone metastases occurred. However, a fact to be considered in relation to metastasis is that bone is the most frequent site of metastasis of advanced breast and prostate cancer. There are some indications from clinical studies that bone metastases develop preferentially in areas with high bone turnover [159,160]. In contrast, agents that inhibit bone resorption have been reported to reduce the incidence of skeletal metastasis [161]. As 1,25-(OH)2D3 may stimulate bone turnover, treatment of cancer with 1,25-(OH)2D3 might theoretically increase the risk of skeletal metastases. This aspect of 1,25-(OH)2D3 therapy certainly needs further study. In this aspect, the use of vitamin D3 analogs with reduced calcemic activity or treatment with vitamin D3 in combination with other compounds to reduce bone turnover (see Section IV) may be helpful. The data obtained so far on angiogenesis and metastasis indicate that these two processes are part of the spectrum of mechanisms by which vitamin D3 exerts its anticancer activity.

JOHANNES P. T. M. VAN LEEUWEN AND HUIBERT A. P. POLS

F. Parathyroid Hormone-Related Peptide 1,25-(OH)2D3 and parathyroid hormone (PTH) mutually regulate synthesis and secretion of one another. Production and secretion of PTH are inhibited by 1,25-(OH)2D3 via a transcriptional effect, and a vitamin D responsive element (VDRE) in the promoter of the PTH gene has been identified [162,163] (see Chapter 30). Parathyroid hormone-related peptide (PTHrP) was initially isolated from several carcinomas and is responsible for the humoral hypercalcemia of malignancy syndrome [164]. Although originally identified in carcinomas, PTHrP has also been identified in normal cells (see Chapter 43). In normal human mammary epithelial cells, 1,25(OH)2D3 did not affect basal but inhibited growth factor-stimulated PTHrP expression via an effect on transcription [165]. In normal keratinocytes 1,25-(OH)2D3 had no effect on PTHrP secretion in basal culture conditions [166], but did inhibit growth factor-stimulated PTHrP production as well [167]. Likewise, 1,25(OH)2D3, as well as the analogs 22-oxacalcitriol and MC903, inhibited PTHrP secretion in immortalized human keratinocytes (HPK1A), but this inhibition was less in the more malignant ras-transfected clone HPK1A-ras [168,169]. 1,25-(OH)2D3 and the analogs EB1089 and 22-oxacalcitriol inhibit the PTHrP gene transcription in and release from the squamous cancer cell line NCI H520 [170]. In addition, in the human T-cell lymphotrophic virus type I (HTLV-I)-transfected T-cell line MT-2, 1,25-(OH)2D3 and 22-oxacalcitriol did inhibit PTHrP gene expression and PTHrP secretion [171]. In rat H-500 Leydig tumor cells [172], and PC-3 prostate cancer cells 1,25-(OH)2D3 inhibited PTHrP secretion. It was suggested that this might play a role in the growth inhibition by vitamin D as PTHrP stimulates prostate cancer growth, tumor invasion, and metastasis [173–175]. In vivo observations comparable to these in vitro observations have also been made. When these H-500 Leydig tumor cells were implanted in Fisher rats, treatment with 1,25-(OH)2D3 and the analog EB1089 resulted in reduced levels of tumor PTHrP mRNA and PTHrP serum levels [97]. EB1089 also reduced serum levels of PTHrP in nude mice implanted with squamous cancer cells [176]. In Fisher rats implanted with the Walker carcinoma, 1,25(OH)2D3 caused a decrease in serum PTHrP, but the ratio of PTHrP levels and tumor weight was similar in rats receiving vehicle or 1,25-(OH)2D3. The data point to an indirect effect on PTHrP via growth inhibition. However, the PTHrP mRNA levels appeared to be decreased by 1,25-(OH)2D3 [100]. In nude mice bearing the FA-6 cell line of a pancreas carcinoma lymph node metastasis, 22-oxacalcitriol inhibits PTHrP gene expression, which is related to inhibition of

CHAPTER 89 Vitamin D: Cancer and Differentiation

tumor-induced hypercalcemia [177]. Together, the overall picture that emerges from these studies is that an important additional anticancer effect of vitamin D3 and analogs could be the inhibition of the humoral hypercalcemia of malignancy. In contrast to these inhibitory effects in human tumor cells and tumor models, a stimulatory effect of 1,25(OH)2D3 and EB1089 on PTHrP gene transcription and PTHrP production by a canine oral squamous carcinoma cell line (Sec 2/88) has been observed [178,179]. Also in an in vivo model of canine adenocarcinoma CAC-8 implanted in nude mice, stimulation of PTHrP by 1,25(OH)2D3 and EB1089 was observed [179]. These data indicate that the effect of vitamin D and analogs on canine tumors differs from that on human tumors.

III. VITAMIN D EFFECTS ON TUMOR CELLS A. Cell Cycle It has now been well established that vitamin D inhibits growth of cells by interfering with the cell cycle. Proliferating cells progress through the cell cycle, which comprises the G0/G1 phase (most differentiated, nondividing cells are in the G1 phase), the S phase in which new DNA is synthesized, and the G2 phase, which is followed by mitosis (M phase) whereon the cells reenter the G0/G1 phase. In most of the cells studied so far, treatment with 1,25-(OH)2D3 and its analogs results in a blockade at a specific checkpoint, i.e., the restriction point (R), in the G1 phase limiting the transition of G1 to S and reducing the number of cells in S phase. Some studies also have examined the effect on the G2 phase, but these results are somewhat more diverse. In general it can be concluded that blocking the transition from the G0/G1 phase to the S phase plays an important role in the growth inhibitory effect of 1,25-(OH)2D3. In the regulation of the cell cycle, numerous genes and proteins have been described. It is beyond the scope of this chapter to discuss in detail the regulation of all of the genes/proteins by vitamin D. In Fig. 1, an overview is given of the interacting genes/proteins that are involved in intracellular signaling and regulating the cell cycle. These genes and proteins are part of the cascade of events on which vitamin D exerts its effects. The components shown to be regulated by vitamin D are indicated. Figure 1 is a compilation of data present so far; it is important to realize that probably not all genes/proteins are affected by vitamin D in all tumor cells. However, in this way one gets an overview of the broad range of effects of vitamin D on intracellular signaling pathways involved in regulation of (tumor) cell

1577 growth. More details on the regulation of the cell cycle will be discussed in several other chapters, especially Chapter 92. Besides its effects on cell cycle regulation, vitamin D has recently been implicated to be involved in control of genomic stability [180]. 1,25-(OH)2D3 has been reported to inhibit hepatic chromosomal aberrations and DNA strand breaks [181]. This is supported by the finding that 1,25-(OH)2D3 and EB1089 stimulated the expression of GADD45, which stimulates DNA repair [182] and might be coupled to release of p53 from Mdm2 (see Fig. 1). 1. (ONCO)GENES AND TUMOR SUPPRESSOR GENES

Oncogenes and tumor suppressor genes generally are involved in control of the cell cycle and apoptosis (see Chapter 92). One of the most widely studied oncogenes in relation to vitamin D is c-myc. C-Myc suppresses expression of cell cycle/growth arrest genes gas1, p15, p21, p27, and gadd34, -45, and -153 [183]. C-Myc has been postulated to play an early role in the following cascade of events in G1: cyclins activate cyclin-dependent kinases (CDKs), which in turn can phosphorylate the retinoblastoma tumor suppressor gene product (p110RB), resulting in transition from G1 to S phase (see Fig. 1). In HL-60 cells, breast cancer cells, and several other cell types, 1,25-(OH)2D3 has been reported to decrease c-myc oncogene expression [184–189]. Analysis of HL-60 sublines showed a relation between reduction of c-myc expression and inhibition of proliferation [190]. Similar observations were made for neuroblastoma cells treated with 1,25-(OH)2D3, EB1089, and KH10560 [191]. We did not observe a 1,25-(OH)2D3induced change in c-myc expression in MCF-7 and ZR75.1 breast cancer cells while they were both growth inhibited [192], and a similar observation has been made for the colon-adenocarcinoma CaCo-2 cell line [193]. Nontransformed embryonic fibroblasts are growth inhibited by 1,25-(OH)2D3, whereas c-myc is not changed or is even increased [194,195]. In the MG-63 osteosarcoma cell line, 1,25-(OH)2D3 has been shown to enhance c-myc expression [196], whereas we observed growth inhibition by 1,25-(OH)2D3 [197]. These data show that regulation of c-myc expression may be part of growth inhibition by vitamin D, but that this is not generally applicable to all cells. 1,25-(OH)2D3 has also been reported to regulate expression of other oncogenes, like c-myb, c-fos, c-fms, c-fra1, c-jun, junD, c-Ki-ras, N-ras, c-src [189,198–203]; however, these data are rather limited. Nevertheless, it is clear that 1,25-(OH)2D3 has effects on the expression of various oncogenes. The data so far are not conclusive with respect to which genes are crucial in the growth inhibitory action of 1,25-(OH)2D3. This can be attributed to the fact that these (proto)oncogenes encode for transcription factors,

1578

JOHANNES P. T. M. VAN LEEUWEN AND HUIBERT A. P. POLS

M G2

G1 S

Cell cycle progression

Differentiation Apoptosis

R

Gene transcription

HDAC

E2F

p p p

pRb p p p

PP1

p

pRb p p

E2F

HDAC

pRb

pRb p p

p

E2F

Cyclin E

Cyclin D

p pRb

Cyclin E – Cdk2

HDAC

Cdk4/6

Cdk2

E2F

p16INK4A p19ARF

p p27KIP1

p

p

Cell cycle progression

p21CIP1

p27KIP1 p

P45/SKP2

Cyclin D – Cdk4/6

p21CIP1

E2F

p53 p53

CKS-1 P45/SKP2

See note

p

p27KIP1

DNA- repair GADD45α

Mdm2 Mdm2 R

Decreased protein stability Degradation Proteasome

NF-κB pathway IKK Apoptosis Bad / Bcl2 Procaspase-9 FKHR

p

Cyclin D p

PLD p

GSK-3

AKT

Cyclin D

p PKC

AKT

p

VDR

β-Catenin VDR

PIP3

PIP2

p

GSK-3

p

DG

p

β-Catenin

p

PDK1

β-Catenin

IP3 PTEN PLC

P13-K

SHIP1/2

Raf / MEK / ERK pathway

To E-cadherin in the membrane

CHAPTER 89 Vitamin D: Cancer and Differentiation

growth factor receptors, or components or intracellular signaling cascades. The effects of these may differ between cells dependent on presence or absence of additional cell type specific conditions. Therefore, their postulated role is often complex. For example, increased c-myc expression can be related to induction of apoptosis but also to stimulation of cell cycle progression. In contrast to the oncogenes, the effect of 1,25(OH)2D3 on the retinoblastoma tumor suppressor gene is much clearer. This may be related to the fact that, in contrast to oncogenes, retinoblastoma and p53 take well-defined positions in the control of cell cycle and DNA repair (see Fig. 1). The p110RB retinoblastoma gene product can either be phosphorylated or dephosphorylated. In the phosphorylated form, it can activate several transcription factors and cause transition to S phase and DNA synthesis. In human chronic myelogenous leukemia cells [204], breast cancer cells [205], and HL-60 cells [206,207], 1,25-(OH)2D3 caused a dephosphorylation of p110RB, which is related to growth inhibition and cell cycle arrest in G0/G1 and also in G2 [207]. In leukemic cells, 1,25-(OH)2D3 also caused a reduction in the cellular level of p110RB [204,206]. In nontransformed keratinocytes, 1,25-(OH)2D3 induced dephosphorylation of p110RB as well [208]. The other major tumor suppressor gene is p53. For leukemic U937 cells, it was reported that presence of p53 is important for 1,25-(OH)2D3-induced differentiation [209]. In rat glioma cells, 1,25-(OH)2D3 induces expression of p53 [210]. However, 1,25-(OH)2D3 can inhibit cell growth and induce differentiation in cancer cells with defective p53 [211] and also p53-independent induction of apoptosis by EB1089 has been demonstrated [212]. These latter observations might be explained by the fact that vitamin D also interferes at levels in the cascade of cell cycle control down-stream of p53 (see Fig. 1). Recently, an additional interesting relationship between tumor suppressor genes and vitamin D has recently been shown for the Wilms’ tumor suppressor gene WT1. This zinc-finger containing

1579 transcription factor induces transcription of the VDR gene [213]. Several interesting additional genes and vitamin D targets in cancer treatment should be mentioned. First in 1994 Chen and DeLuca isolated and characterized a vitamin D–induced gene in HL-60 cells [214]. This protein, vitamin D-up-regulated protein (VDUP1), is a thioredoxin-binding protein-2 [215]. Thioredoxin has several roles in processes such as proliferation or apoptosis. It also promotes DNA binding of transcription factors such as NF-κB, AP-1, p53, and PEBP2. In addition, overexpression of thioredoxin suppresses the degradation of IκB and the transactivation of NF-κB, whereas overexpression of nuclear-targeted thioredoxin exhibits the enhancement of NF-κB-dependent transactivation [216]. However, it is only in more recent studies that a relationship between VDUP1 and cancer has been established. The expression of VDUP1 was found to correlate with malignant status of colorectal and gastric cancers [217]. 5-fluorouracil, which is widely used for treatment of colon cancer, induces VDUP1 expression in the SW620 colon cancer cell line [218]. In smooth muscle cells and cardiomyocytes VDUP1 inhibits proliferation and is involved in induction of apoptosis [219,220]. An association with vitamin D effects on cancer is made by two recent studies showing induction of VDUP1 by 1,25(OH)2D3 in tumor cells and that VDUP1 induces cell cycle arrest [221,222]. Moreover, interaction with histone deacetylase (HDAC; see Fig. 1), promyelocytic leukemia zinc-finger (PLZF) was demonstrated. Interestingly and further complicating the story, PLZF inhibits 1,25-(OH)2D3 induced differentiation of U937 leukemic cells by binding to the VDR and inhibiting gene transcription [223,224]. Interestingly, the gene, DRH1, was cloned from hepatocellular carcinoma, and its expression was strongly reduced in cancer tissue compared to normal liver [225]. DRH1 has a 41% homology with VDUP1. Whether this points to a new family of cancer genes remains to be established, but it certainly opens new venues for intervening in cancer cell growth.

FIGURE 1 Schematic representation summarizing the intracellular pathways and signaling pathways involved regulation of the cell cycle shown to be regulated by 1,25-(OH)2D3 and 1,25-(OH)2D3 analogs in regulating cell proliferation. Targets shown to be affected by 1,25(OH)2D3 and/or its analogs are indicated in the bold boxes and ovals. Bold arrows and fine dotted lines indicate stimulation and inhibition, respectively. Coarse dotted lines indicate processing to the proteasome. p indicates phosphorylation. The effects on these cellular targets are not demonstrated in all types of cancer cells but this diagram is aimed to give an overview of demonstrated targets and potential targets. NOTE: Dependent on the site of phosphorylation proteins can either be destabilized or degraded or be stabilized and activated. For example: phosphorylation of p21 at T145 by AKT leads to degradation while phosphorylation of S146 by AKT leads to increased stability. Abbreviation used: AKT (PKB), Protein kinase B; Bad, BCL2-antagonist of cell death; Bcl2, B-cell leukemia/lymphoma 2; Cdk, Cyclindependent kinase; CKS-1, Cyclin kinase subunit 1; DG, Diacylglycerol; E2F, Transcription factor; ERK, Extracellular-signal regulated kinase; FHKR (AFX/FOX), Forkhead family of transcription factors; GSK-3, Glycogen synthase kinase-3; HDAC, Histone deacetylase; IKK, I-κB kinase; IP3, Inositol 1,4,5-trisphosphate; Mdm2, Mouse double minute 2; MEK, Raf-1-MAPK/ERK kinase; PDK1, Phosphatidylinositoldependent kinase 1; PI3-K, Phosphatidyl inositol 3 kinase; PIP2, Phosphatidylinositol (4,5)-phosphate; PIP3, Phosphatidylinositol (3,4,5) phosphate; PKC, Protein kinase C; PLC, Phospholipase C; PLD, Phospholipase D; PP!, Protein phosphatase 1-like protein; pRB, Retinoblastoma protein; PTEN, Phosphatase and tensin homologue; SHIP 1 and 2, Src homology 2 (SH2) containing phosphatases 1 and 2; SKP2, Ubiquitin ligase; VDR, Vitamin D receptor.

1580 Second, an additional therapeutic target for vitamin D compounds might be regulation of enzymes involved in estrogen and androgen synthesis and metabolism [226–229]. Third, telomerase activity provides a mechanism for unlimited cell division. In HL-60 cells, 1,25-(OH)2D3 inhibits telomerase activity [230]. Fourth, the homeobox genes may prove to be a major target for vitamin D action in cancer, but this possibility remains to be elucidated. In a differential expression screen using the human U937 leukemic cells, the HoxA10 gene was shown to be regulated by 1,25-(OH)2D3 [231]. It is to be expected that as a result of the increasing application of large scale microarray gene expression analyses, a vast number of new cell cycle and vitamin D regulated genes will be identified and add to the unraveling and understanding of vitamin D control of cancer cell proliferation [232–235].

B. Apoptosis A block in the cell cycle preventing transition into S phase may cause cells to go either into apoptosis or to enter a specific differentiation pathway (see Chapter 93). What exactly determines the decision of apoptosis or differentiation remains to be elucidated. It is suggested that early G1 phase may be the point at which switching between cell cycle progression and induction of apoptosis occurs [236,237]. Induction of apoptosis, an orderly and characteristic sequence of biochemical, molecular, and structural changes resulting in the death of the cell [238], is a mechanism by which 1,25-(OH)2D3 inhibits tumor cell growth and may contribute to tumor suppression and explain the reduction in tumor volume found in various in vivo animal studies (see Section II.C). 1,25-(OH)2D3 has been shown to regulate expression of apoptosis genes and to induce apoptosis of cancer cells of various origins. For example, 1,25-(OH)2D3 and the analog Ro 25-6760 cause a cell cycle block in HT-29 human colon cancer cells, resulting in growth inhibition and induction of apoptosis [239]. The bcl-2 oncogene decreases the rate of programmed cell death [240,241]. However, protection of HL-60 cells against apoptosis occurred despite down-regulation of bcl-2 gene expression [242]. In several breast cancer cell lines (MCF-7, BT-474, MDA-MB-231) 1,25-(OH)2D3 and the analogs KH1060 and EB1089 decreased bcl-2 expression [211,243]. The analog CB1093 reduced bcl-2 expression in MCF-7 cells associated with the induction of apoptosis [244]. However, only in MCF-7 cells has this change in bcl-2 expression been accompanied by apoptosis. Effects on other genes/proteins have also been reported [245], and microarray gene

JOHANNES P. T. M. VAN LEEUWEN AND HUIBERT A. P. POLS

expression analyses and differential screening will also definitively reveal additional vitamin D targets in regulating apoptosis [246]. A central role for apoptosis in the action of 1,25(OH)2D3 is unclear because growth inhibition of several other breast cancer cells appeared to be independent of apoptosis [211]. Also, MCF-7 cells that showed growth inhibition by 1,25-(OH)2D3 could, after removal of the hormone, again be stimulated to grow, implying transient growth inhibition and not cell death [247]. Stable transfection of leukemic U937 cells with the wild-type p53 tumor suppressor gene resulted in a reduced growth rate and produced cells that can undergo either apoptosis or maturation. In these cells 1,25-(OH)2D3 protects against p53-induced apoptosis and enhances p53-induced maturation [209]. In two independent studies with HL-60 cells, 1,25-(OH)2D3 was found either to protect against or to have no effects on apoptosis [242,248]. Vitamin D protection against apoptosis was also detected in human U937 leukemic cells treated tumor necrosis factor α [249]. Absence of a vitamin D effect on apoptosis might be explained by the expression of the antiapoptotic protein BAG-1 p50 isoform. This protein has been shown to bind to the VDR and block vitamin D–induced transcription [250]. The presence of additional interacting factors might also be important for the eventual effect on apoptosis as in the study with HL-60 cells, which in the presence but not the absence of 9-cis-retinoic acid, 1,25-(OH)2D3 did induce apoptosis [248]. The role of vitamin D interaction with other factors will be discussed in more detail in Section IV. In summary, the data obtained so far show that 1,25-(OH)2D3-induced growth inhibition can be related to apoptosis in some cases, but that growth inhibition is frequently observed to be independent of apoptosis. Possibly in these latter cases, induction of differentiation is more prominent. The factor that decides whether cells undergo apoptosis or differentiation is unclear but is probably dependent on cell cycle stage, presence of other factors, and levels of expression of oncogenes and tumor suppressor genes. An interesting phenomenom to be studied concerning vitamin D and apoptosis is calbindin 28K. Calbindin 28K is a wellknown vitamin D–induced protein that has recently been shown to inhibit apoptosis [251]. It is tempting to speculate that calbindin 28K plays a role in the decision whether vitamin D induces cells to differentiate or to go into apoptosis or that it is involved when 1,25-(OH)2D3 protects against apoptosis (see Chapter 42).

C. Differentiation In addition to proliferation and apoptosis, the third major cellular process is differentiation. As described

CHAPTER 89 Vitamin D: Cancer and Differentiation

above for the classic actions of 1,25-(OH)2D3 related to calcium homeostasis, effects on cell differentiation and proliferation are involved. The coupling between proliferation and differentiation has been most widely studied for cells of the hematopoietic system (Chapter 96) and keratinocytes (Chapter 35). In general, 1,25-(OH)2D3 inhibits proliferation and induces differentiation along the monocyte-macrophage lineage. Rapidly proliferating and poorly differentiated keratinocytes can be induced to differentiate by 1,25-(OH)2D3. A further relationship between the vitamin D3 system and differentiation is demonstrated by the fact that in poorly differentiated keratinocytes 1,25-(OH)2D3 production and vitamin D receptor levels are high, whereas after induction of differentiation these levels decrease [252], and in melanoma cells 1,25-(OH)2D3 stimulates melanin production [253]. Effects on differentiation have also been reported for other cell types. Inhibition of prostate cancer cell proliferation is paralleled by an increased production of prostate specific antigen [254–257]. In the BT-20 breast cancer cells 1,25-(OH)2D3 induced morphological changes indicative for differentiation [258]. In several breast cancer cell lines, the stimulation of differentiation has been established by determining lipid production by the cells [211]. In this study, Elstner et al. demonstrated an uncoupling between effects on proliferation and differentiation. In two breast cancer cell lines, 1,25-(OH)2D3 and various analogs induced differentiation even though the cells were resistant to cell cycle and antiproliferative effects. This finding, together with data obtained with human myelogenous leukemia cells, [204] suggests a dissociation between the cellular vitamin D3 pathways involved in regulation of differentiation and proliferation (see also Section V). For a HL-60 subclone, a similar observation was made [190], and in another HL-60 subclone the induction of differentiation was found to precede the G0/G1 cell cycle block. In contrast to the above-mentioned observations on stimulation of differentiation, 1,25-(OH)2D3 inhibits erythroid differentiation of the erythroleukemia cell line K562 [186], and 1,25-(OH)2D3 inhibits Activin A-induced differentiation of murine erythroleukemic F5-5 cells [259]. Although precise relationships among growth inhibition, cell cycle effects, and apoptosis are unclear, it can be concluded that an important effect of vitamin D3 on both normal and malignant cells is induction of differentiation.

D. Growth Factors and Growth Factor Receptors Besides regulation of cell cycle-related oncogenes and tumor suppressor genes, interaction with tumoror stroma-derived growth factors is important for

1581 growth inhibition. Stimulation of breast cancer cell proliferation by coculture with fibroblasts is inhibited by 1,25-(OH)2D3 [260]. A good candidate to interact with the 1,25-(OH)2D3 action is transforming growth factor-β (TGFβ). TGFβ is involved in cell cycle control and apoptosis [261,262]. TGFβ can interfere with the cascade of events in the GI phase described above and inhibit the ability of cells to enter S phase when the factor is present during the GI phase. TGFβ has been shown to suppress c-myc, cyclin A, cyclin E, and cdk2 and cdk4 expression [262]. In line with this, TGFβ has been reported to inhibit phosphorylation of p110RB [263]. Vitamin D3 compounds induce dephosphorylation of the retinoblastoma gene product, and vitamin D3 growth inhibition of MCF-7 breast cancer cells is inhibited by a TGFβ neutralizing antibody [264]. 1,25-(OH)2D3 and several analogs stimulated the expression of TGFβ mRNA and secretion of active and latent TGFβ1 by the breast cancer cell line BT-20 [154]. 1,25-(OH)2D3 enhanced TGβ1 gene expression in human keratinocytes [265] and the secretion of TGFβ in murine keratinocytes [266]. In both studies, antibodies against TGFβ inhibited the growth inhibitory effect of vitamin D3. Further evidence for a vitamin D3-TGFβ interaction is that bone matrix of vitamin D–deficient rats contains substantially less TGFβ than controls [267]. Therefore, on the basis of these consistent findings, TGFβ is a likely candidate to play a role in the l,25-(OH)2D3-induced growth inhibition [268]. Interactions with the insulin-like growth factor (IGF) system have also been described. IGFs are potent growth stimulators of various cells, and their effect is regulated via a series of IGF binding proteins (IGFBPs). 1,25-(OH)2D3 and the analog EB1089 inhibit the IGF-Istimulated growth of MCF-7 breast cancer cells [269]. In prostate cancer cell lines, 1,25-(OH)2D3 induced expression of IGFBP6 but not IGFBP4 [270]. In human osteosarcoma cell lines, 1,25-(OH)2D3 and the analog 1α-dihydroxy-16-ene-23-yne-26,27-hexafluorocholecalciferol potently stimulated the expression and secretion of IGFBP3 [271–273]. In one study an association has been made between increased IGFBP3 levels and 1,25-(OH)2D3 growth inhibition [271]. Recent observations that antisense oligonucleotides to IGFBP3 prevented growth inhibition of prostate cancer cells by 1,25-(OH)2D3 [235] provided further evidence for an interplay between 1,25-(OH)2D3 and IGFBP3. Interestingly, in the human osteosarcoma cell line MG-63, 1,25-(OH)2D3 and TGFβ synergistically increased IGF-BP-3 secretion [273]. An example of growth factor receptor regulation by 1,25-(OH)2D3 concerns the epidermal growth factor (EGF) receptor. This receptor is down-regulated in T47-D breast cancer cells and up-regulated in BT-20 breast cancer cells. Nevertheless, 1,25-(OH)2D3 inhibits the growth of both

1582 cell lines [274,275]. These data provide evidence that interactions with growth factors are only part of the 1,25-(OH)2D3 action on tumor cells. As described above, it is clear that 1,25-(OH)2D3 has effects on the expression of various oncogenes and tumor suppressor genes and that multiple interactions with various growth factors exist. However, the data on these aspects, separately as well as in combination, are still too limited to define a distinct mechanism of action for the 1,25-(OH)2D3 anticancer effects. However, with respect to growth inhibition, at this time two models of action can be postulated. In the first one, 1,25-(OH)2D3 directly interferes with a crucial gene(s) involved in the control of the cell cycle. In this case, in view of the general pattern of the genes involved in cell cycle control, this mechanism of action will be similar in all types of cancer cells. However, the effect on cell cycle genes will be dependent on the presence or absence of additional growth factors. This will determine, depending on which growth factors are present, the differences in 1,25-(OH)2D3 action between cancer types of different origin but also within cancer types of similar origin. The second model is based on an indirect effect of 1,25-(OH)2D3 on cell cycle progression and tumor growth. In this case 1,25-(OH)2D3 may either inhibit or potentiate the effect of growth stimulatory or inhibitory factors, respectively, via, for example, effects on growth factor production, growth factor binding protein levels, or receptor regulation. It is also conceivable that a combination of both models forms the basis of 1,25-(OH)2D3 regulation of tumor cell growth.

IV. COMBINATION THERAPY The data obtained with 1,25-(OH)2D3 and its analogs on growth inhibition and stimulation of differentiation offer promise for their use as an endocrine anticancer treatment. Single agent treatment with low calcemic 1,25-(OH)2D3 analogs could be useful; however, combination therapy with other tumor effective drugs may provide an even more beneficial effect. Up to now several in vitro and in vivo studies have focused on possible future combination therapies with 1,25(OH)2D3 and 1,25-(OH)2D3 analogs. For breast cancer cells the combination of the presently most widely-used endocrine therapy, the antiestrogen tamoxifen, with 1,25-(OH)2D3 and 1,25(OH)2D3 analogs resulted in a greater growth inhibition of MCF-7 and ZR-75-1 cells than treatment with either compound alone [118,192,247]. In combination with tamoxifen, the cells were more sensitive to the antiproliferative action of 1,25(OH)2D3 and the analogs;

JOHANNES P. T. M. VAN LEEUWEN AND HUIBERT A. P. POLS

that is, the EC50 values of the vitamin D3 compounds in the presence of tamoxifen were lower than those in the absence of tamoxifen. Studies with MCF-7 cells suggested a synergistic effect of 1,25(OH)2D3 and tamoxifen on apoptosis [276]. In addition, in in vivo breast cancer models a synergistic effect of the tamoxifen-l,25(OH)2D3 analogs combination was observed [118,119]. Additional data on the interaction between the estrogen/antiestrogen system and vitamin D comes from studies showing the presence of an estrogen responsive element in the VDR promoter and regulation of VDR by estradiol in breast cancer cells [277]. This is intriguing that the stimulator of breast cancer cell growth induces the expression of the receptor for a growth inhibitor. VDR up-regulation in breast cancer cells and increased transcriptional activity was mimicked by the phytoestrogens resveratrol and genistein and blocked by tamoxifen [278]. In colon cancer also, VDR up-regulation by estradiol has been reported. However, in colon it was hypothesized to contribute to the protective effect of estradiol on chemicallyinduced colon carcinogenesis [279]. These important and complex interactions between the vitamin D and estrogen endocrine system in the regulation of cancer (cells) are promising and warrant further detailed analyses, e.g. regarding tissue(cancer)specific effects. In addition, the estrogen endocrine system may regulate the metabolism of 1,25-(OH)2D3 in cancer cells and thereby affect its action (see Section V). Interaction with another sex steroid, testosterone, has been described for ovarian cancer. Vitamin D inhibits dihydrotestosterone (DHT) and DHT stimulation of ovarian cancer cells [280]. Intriguingly, also here the growth stimulator and growth inhibitor mutually upregulate each others receptors. Also, in prostate cancer cells, it has been shown that 1,25-(OH)2D3, while inhibiting androgen stimulated growth, up-regulates the androgen receptor [281]. Interaction with another steroid in regulating cancer cells had already been reported in 1983. The synthetic glucocorticoid, dexamethasone, and 1,25-(OH)2D3 synergistically induced differentiation of murine myeloid leukemia cells [282]. This was supported by in vitro and in vivo data showing that dexamethasone enhanced the effect of vitamin D on growth inhibition, cell cycle arrest, and apoptosis of squamous carcinoma cells [283,284]. A possible mechanism is the up-regulation of VDR by dexamethasone [283]. An interesting aspect of this combination is not only the direct interaction at cancer cell level, but also in the control of the calcemic action of 1,25-(OH)2D3. Glucocorticoids inhibit intestinal calcium absorption and increase renal calcium excretion and in this way it may limit the hypercalcemic action of 1,25-(OH)2D3 [285].

1583

CHAPTER 89 Vitamin D: Cancer and Differentiation

Combination of vitamin D3 and retinoids has been examined in various systems. A combination of retinoic acid and 1,25-(OH)2D3 resulted in a more profound inhibition of both T47-D breast cancer cells [286] and LA-N-5 human neuroblastoma cells [287]. 9-cisRetinoic acid augmented l,25-(OH)2D3-induced growth inhibition and differentiation of HL-60 cells [288]. Besides growth inhibition and differentiation effects, the combination of 1,25-(OH)2D3 and various isomers of retinoic acid were more potent in reducing angiogenesis than either compound alone [146–148]. The background of the interaction between retinoids and 1,25-(OH)2D3 may be attributed to heterodimer formation of the respective receptors [289]. For several cytokines, interactions with 1,25-(OH)2D3 have been described. Interferon-γ and 1,25-(OH)2D3 synergistically inhibited the proliferation and stimulated the differentiation of HL-60, WEHI-3, and U937 myeloid leukemia cells [290–293]. Treatment of LLC-LN7 tumor cells with 1,25-(OH)2D3 with IFN-γ synergistically reduced tumor granulocyte-macrophage colonystimulating factor (GM-CSF) secretion and a blockage in the capacity of the tumor cells to induce granulocytemacrophage-suppressor cells [99]. In the mouse myeloid leukemia cell line Ml interleukin-4 enhanced 1,25(OH)2 D3-induced differentiation [189,294,295]. Also with interleukin-1β, interleukin-3, interleukin-6, and interleukin-12 interactions with 1,25-(OH)2D3 have been reported [296–298]. 1,25-(OH)2D3 and tumor necrosis factor synergistically induced growth inhibition and differentiation of HL-60 [299]. For MCF-7 cells an interaction between 1,25-(OH)2D3 and tumor necrosis factor has also been reported [298,300]. In the presence of GM-CSF, lower concentrations of 1,25(OH)2D3 could be used to achieve a similar antiproliferative effect in MCF-7 cells [301] and to induce differentiation of U937 myeloid leukemic cells [302]. Other factors shown to interact with 1,25-(OH)2D3 are butyrate [303–305], melatonin [306], EGF [307], and the factors described in Section III.C. Furthermore, combinations of vitamin D3 compounds with cytotoxic drugs, antioxidants, and radiation have been studied. In vivo adriamycin and in vitro carboplatin, cisplatin, and doxorubicin interacted synergistically with 1,25-(OH)2D3 to inhibit breast cancer cell growth [113,308–311]. In a carcinogen-induced rat mammary tumor model, treatment with 1α-(OH)D3 and 5-fluorouracil, however, did not result in enhanced antitumor effects [96]. Recently, interactions with a plant-derived polyphenolic antioxidant, carnosic acid were demonstrated in the differentiation of HL-60 cells, which was related to a decrease in the intracellular levels of reactive oxygen species [312,313]. Also interaction with radiation therapy in breast cancer has been described [314–316].

The data on combinations of 1,25-(OH)2D3 and 1,25-(OH)2D3 analogs with various other anticancer compounds are promising and merit further analyses. The development of effective combination therapies may result in better response rates and lower required dosages, thereby reducing the risk of negative side effects.

V. RESISTANCE AND VITAMIN D METABOLISM Classic vitamin D resistance concerns the disease hereditary vitamin D–resistant rickets, which is characterized by the presence of a nonfunctional VDR and consequently aberrations in calcium and bone metabolism (see Chapter 72). For cancer cells, the presence of a functional VDR is also a prerequisite for a growth regulatory response, and a relationship between VDR level and growth inhibition has been suggested for osteosarcoma, colon carcinoma, breast cancer, prostate cancer cells, and rat glioma [1,2,108,129,205,210,317–321]. Cell lines established from DMBA-induced breast tumors in VDR knockout mice are insensitive to growth arrest and apoptosis by 1,25-(OH)2D3, EB1089 and CB1093 [322]. Albeit that VDR is a prerequisite for tumor cell growth regulation, the presence of the VDR is not always coupled to a growth inhibitory response of 1,25(OH)2D3. Results from studies with transformed fibroblasts [194], myelogenous leukemia cells [190,204,323], transformed keratinocytes [187], and various breast cancer cell lines [211,324] demonstrated a lack of growth inhibition by 1,25(OH)2D3 even in the presence of VDR. In this situation, the designation “resistant” is based on the lack of growth inhibition, even though, as discussed earlier in Section III.C, some of these cells are still capable of being induced to differentiate [204,211]. This points to a specific defect in the growth inhibitory pathway. In the resistant MCF-7 cells, this defect is not located at a very common site in the growth inhibitory pathway of the cell because the growth could still be inhibited with the antiestrogen tamoxifen [324]. For myelogenous leukemia cells, similar observations have been made [325]. For VDR-independent resistance to growth inhibition, the underlying mechanism(s) is unknown. For the resistant MCF-7 clone, this is not related to up-regulation of the P-glycoprotein [324]. Interestingly, these vitamin D–resistant MCF-7 clones can be sensitized to 1,25(OH)2D by activation of protein kinase C, resulting in induction of apoptosis and transcriptional activation, suggesting that alterations in phosphorylation may affect vitamin D sensitivity [326]. An interesting growth

1584 inhibition resistant MCF-7 cell clone was described by Hansen et al. This clone was not growth inhibited while VDR was still present and 24-hydroxylase could still be induced [327]. Other examples of vitamin D resistance are HL60 cells that have been cultured for four years in the presence of 1,25-(OH)2D3 and resulted in clones that are resistant to differentiation inducing and growth inhibition. They became not only resistant to 1,25(OH)2D but also to 5-beta-D-arabinocytosine, suggesting a common metabolic pathway being responsible [328]. Whether this relates to the up-regulation of the multidrug resistance proteins is not clear. In the resistant leukemia JMRD3 cell line, altered regulation and DNA–binding activity of junD as part of the AP-1 complex has been reported [200]. Resistance to growth inhibition in the presence of VDR has also been linked to disruption of the VDR-RXR complex [329] and increased RXR degradation [330]. In addition, other factors, like the acute myeloid leukemia translocation products (e.g. PLZF) may contribute to resistance to vitamin D by sequestering the VDR [223,224]. The 1,25(OH)2D3 sensitive and resistant cell clones provide interesting models to examine the molecular mechanisms of l,25(OH)2D3-induced growth inhibition. For example, lack of p21 results in no cell cycle block [331] and no apoptosis was detected with a mutated p53 [211]. Finally, the recent identification of cellular proteins that are involved in the vitamin D resistance in new world primates might add to the understanding of tumor cell resistance to vitamin D [332,333] (see Chapter 21). At this time, the major mechanism for vitamin D resistance or reduced sensitivity in VDR containing tumor and cancer cells is 1,25-(OH)2D3 catabolism via the C24-hydroxylation pathway. An inverse relationship between cellular metabolism of 1,25-(OH)2D3 via 24-hydroxylation and growth inhibition of prostate cancer cells has been suggested [318]. The latter observation is intriguing, the more so as an inverse relationship between VDR level and induction of 24-hydroxylase (CYP24) activity was reported. In general, there may exist a direct relationship between VDR level and induction of 24-hydroxylase activity [319,334]. An important role in the control of 1,25-(OH)2D3 action on cancer cells was provided by studies with the 1,25(OH)2D3–resistant prostate cancer cell line DU145. It was shown that 1,25-(OH)2D3 did inhibit the growth of these cells when it was combined with the 24-hydroxylase inhibitor Liazorole [335]. Inhibition of 24hydroxylase activity in HL-60 cells also altered the effect of 1,25-(OH)2D3 and 20-epi analogs [336]. The action of the analog EB1089 was also limited by hydroxylation at the C24 position [337]. However, it was

JOHANNES P. T. M. VAN LEEUWEN AND HUIBERT A. P. POLS

suggested that the increased potency of EB1089 is at least partly due to resistance to 24-hydroxylation [234]. Alternatively, 24-hydroxylation of the analog KH1060 has been implicated as one of the mechanisms to explain the potency of this analog. The 24-hydroxylated metabolites of this analog are very stable and are biologically active [338,339]. It has been shown that the naturally occurring 24-hydroxylated metabolite of vitamin D3, 24R,25-(OH)2D3, also has a preventive effect on chemically-induced colon cancer [340]. Interaction between the estrogen system and 24-hydroxylase is also of importance. Recent data have shown that the phytoestrogen genistein inhibits 24-hydroxylase activity in prostate cancer cells and thereby increases the responsiveness to 1,25-(OH)2D3 [341]. A role for 24-hydroxylase as oncogene is suggested by data showing amplification of the CYP24 locus on chromosome 20q13.2 [342]. In contrast to degradation of 1,25-(OH)2D3 by 24-hydroxylase in cancer cells, recently it has become clear that tumor cells contain 1α-hydroxylase activity and thereby are able to generate 1,25-(OH)2D3. Expression of 1α-hydroxylase has been demonstrated in colorectal cancer [343–345]. It was postulated that in early stages tumor cells respond by up-regulating 1α-hydroxylase activity to counteract neoplastic growth while at later stages of tumor development this is lost [343]. Also in prostate cancer [346] and inflammatory myofibroblastic tumor [347] 1α-hydroxylase has been detected, albeit in the latter case the tumor contains large numbers of macrophages. It can be anticipated that in the coming years investigation of the expression of both 24-hydroxylase, 1α-hydroxylase in tumors will add to the understanding of vitamin D in the initiation and progression of cancer.

VI. STIMULATION OF PROLIFERATION Over the years a limited number of studies have demonstrated that, in contrast to growth inhibition, 1,25-(OH)2D3 can also stimulate tumor cell growth and tumor development. In several cells 1,25-(OH)2D3 has been reported to have a biphasic effect, that is, at lower concentrations (<10−9 M) it stimulates proliferation and at higher concentrations (10−9 to 10−7 M) it inhibits proliferation. However, clear growth stimulation can sometimes be observed not only at low concentrations but also at the concentrations generally found to inhibit tumor cell proliferation and tumor development. 1,25(OH)2D3 has been shown to stimulate the growth of a human medullary thyroid carcinoma cell line [348]. Not only cancer cells but also several normal cells, for example, human monocytes [349], smooth muscle

1585

CHAPTER 89 Vitamin D: Cancer and Differentiation

cells [350], and alveolar type II cells [351], are stimulated to grow by 1,25-(OH)2D3. Skin is another organ in which different effects of 1,25-(OH)2D3 have been observed. In vivo studies demonstrated that 1,25-(OH)2D3 and analogs stimulate keratinocyte proliferation in normal mice [352–355] and enhance anchorage-independent growth of preneoplastic epidermal cells [356]. In contrast, other studies showed 1,25-(OH)2D3 inhibition of proliferation of mouse and human keratinocytes [357,358], and 1,25-(OH)2D3 is also effective in the treatment of the hyperproliferative disorder psoriasis [359]. Moreover, in vivo studies demonstrated that, depending on the carcinogen, 1,25-(OH)2D3 can either reduce [88] or enhance the induction and development of skin tumors in mice [360,361]. In addition, 1,25-(OH)2D3 enhances the chemically-induced transformation of BALB 3T3 cells and hamster embryo cells [362,363]. 1,25-(OH)2D3 also enhanced 12-O-tetradecanoylphorbol-13-acetateinduced tumorigenic transformation of mouse epidermal JB6 Cl41.5a cells [364,365]. Another example comes from research on osteosarcoma cells. In 1986 it was shown that 1,25-(OH)2D3 stimulated the growth of tumors in athymic mice inoculated with the ROS 17/2.8 osteosarcoma cell line [366]. Earlier the same group reported growth stimulation in vitro of these osteosarcoma cells at low concentrations of 1,25(OH)2D3, but growth inhibition by 10−8 M [317]. They speculated that this discrepancy resulted from limited in vivo availability of 1,25-(OH)2D3 for the tumor cells, resulting in concentrations shown to be growth stimulatory in vitro. However, in other experiments with nude mice, the availability of 1,25-(OH)2D3 did not seem to be a factor, as growth inhibition was observed

200 ROS 17/2.8

Control growth (%)

175

(see Table II). In particular, in nude mice implanted with human osteosarcoma cells (MG-63), growth inhibition and tumor suppression by 1,25-(OH)2D3 were observed [98]. In two different in vitro studies, growth inhibition of MG-63 and growth stimulation of ROS 17/2.8 cells was reported [367,368]. For smooth muscle cells, it has been demonstrated, for example, that growth inhibition or stimulation can depend on the presence of additional growth factors in the culture medium [350]. We followed up on this concept by comparing the effects of 1,25-(OH)2D3 and analogs on the growth and osteoblastic characteristics of the two osteosarcoma cell lines under identical culture conditions. At concentrations 10−10 to 10−7 M, 1,25-(OH)2D3 caused an increase in cell proliferation by 100% in ROS 17/2.8 cells, whereas the proliferation of MG-63 cells was inhibited (Fig. 2) [197]. In contrast, in both cell lines 1,25(OH)2D3 stimulated osteoblastic differentiation characteristics such as production of osteocalcin and alkaline phosphatase activity [197,367]. Analyses with another steroid hormone demonstrated that glucocorticoids inhibited the growth of both osteosarcoma cell lines [369,370]. These data indicate specific differences between these cell lines, especially with respect to the 1,25-(OH)2D3 growth regulatory mechanisms. Taken together, the data on growth stimulation and tumor development, although detected in only a minority of cancer cells, demonstrate that treatment with 1,25-(OH)2D3 or analogs may not always cause growth inhibition and tumor size reduction. It is therefore of utmost importance to identify the mechanism(s) by which 1,25-(OH)2D3 exerts its inhibitory and stimulatory effects on cell growth. This may provide tools to assess whether treatment of a particular tumor will be beneficial. Moreover, purely from a mechanistic point of view, the presence of growth-stimulated and growth-inhibited cells, like the 1,25-(OH)2D3 sensitive and resistant cells, may provide tools to examine the 1,25-(OH)2D3 mechanism of growth regulation.

150

VII. CONCLUSIONS

125 100 75

MG-63

50 25 0 0

−14 −13

−12

−11 −10

−9

−8

−7

−6

LG[1,25-(OH)2D3] (M)

FIGURE 2

Effect of 1,25-(OH)2D3 on proliferation of the osteosarcoma cell lines ROS 17/2.8 and MG-63. Effects on proliferation were examined as described by van den Bemd et al. [197].

The data obtained so far, on (1) the distribution of the VDR in a broad range of tumors and (2) the inhibition of cancer cell growth, angiogenesis, metastasis, and PTHrP synthesis by 1,25-(OH)2D3, all hold promise for the development of treatment strategies based on vitamin D3 use in a wide range of cancers. Moreover, combination of vitamin D compounds with other antitumor drugs, hormones, or growth factors is an important additional therapeutic option. Throughout the last years data have accumulated on the cellular targets and mechanism of action of 1,25-(OH)2D3–induced cancer

1586 growth inhibition. The clinical application is enhanced by the development of 1,25-(OH)2D3 analogs with potent growth inhibitory actions and reduced hypercalcemic activity. At the moment more clinical studies are needed in order to firmly establish whether 1,25(OH)2D3 and especially vitamin D3 analogs have therapeutic potential. In the meantime it is crucial to further our understanding of the mechanism(s) by which 1,25(OH)2D3 exerts its effects on tumor cell growth so that these drugs may be employed more effectively.

JOHANNES P. T. M. VAN LEEUWEN AND HUIBERT A. P. POLS

13.

14.

15. 16.

References 17. 1. Shabahang M, Buras RR, Davoodi F, Schumaker LM, Nauta RJ, Evans SR 1993 1,25-dihydroxyvitamin D3 receptor as a marker of human colon carcinoma cell line differentiation and growth inhibition. Cancer Res 53:3712–3718. 2. Buras RR, Schumaker LM, Davoodi F, Brenner RV, Shabahang M, Nauta RJ, Evans SR 1994 Vitamin D receptors in breast cancer cells. Breast Cancer Res Treat 31:191–202. 3. Villena-Heinsen C, Meyberg R, Axt-Fliedner R, Reitnauer K, Reichrath J, Friedrich M 2002 Immunohistochemical analysis of 1,25-dihydroxyvitamin D3 receptors, estrogen, and progesterone receptors and Ki-67 in ovarian carcinoma. Anticancer Res 22:2261–2267. 4. Evans SR, Nolla J, Hanfelt J, Shabahang M, Nauta RJ, Shchepotin IB 1998 Vitamin D receptor expression as a predictive marker of biological behavior in human colorectal cancer. Clin Cancer Res 4:1591–1595. 5. Friedrich M, Rafi L, Tilgen W, Schmidt W, Reichrath J 1998 Expression of 1,25-dihydroxy vitamin D3 receptor in breast carcinoma. J Histochem Cytochem 46:1335–1337. 6. Friedrich M, Villena-Heinsen C, Tilgen W, Schmidt W, Reichrat J, Axt-Fliedner R 2002 Vitamin D receptor (VDR) expression is not a prognostic factor in breast cancer. Anticancer Res 22:1919–1924. 7. Eisman JA, Suva LJ, Sher E, Pearce PJ, Funder JW, Martin TJ 1981 Frequency of 1,25-dihydroxyvitamin D3 receptor in human breast cancer. Cancer Res 41:5121–5124. 8. Freake HC, Abeyasekera G, Iwasaki J, Marcocci C, MacIntyre I, McClelland RA, Skilton RA, Easton DF, Coombes RC 1984 Measurement of 1,25-dihydroxyvitamin D3 receptors in breast cancer and their relationship to biochemical and clinical indices. Cancer Res 44:1677–1681. 9. Eisman JA, Suva LJ, Martin TJ 1986 Significance of 1,25dihydroxyvitamin D3 receptor in primary breast cancers. Cancer Res 46:5406–5408. 10. Berger U, Wilson P, McClelland RA, Colston K, Haussler MR, Pike JW, Coombes RC 1987 Immunocytochemical detection of 1,25-dihydroxyvitamin D3 receptor in breast cancer. Cancer Res 47:6793–6799. 11. Berger U, McClelland RA, Wilson P, Greene GL, Haussler MR, Pike JW, Colston K, Easton D, Coombes RC 1991 Immunocytochemical determination of estrogen receptor, progesterone receptor, and 1,25-dihydroxyvitamin D3 receptor in breast cancer and relationship to prognosis. Cancer Res 51:239–244. 12. Goode EL, Dunning AM, Kuschel B, Healey CS, Day NE, Ponder BA, Easton DF, Pharoah PP 2002 Effect of germ-line

18. 19.

20. 21. 22. 23. 24. 25. 26.

27. 28.

29.

30.

31.

genetic variation on breast cancer survival in a populationbased study. Cancer Res 62:3052–3057. Makishima M, Lu TT, Xie W, Whitfield GK, Domoto H, Evans RM, Haussler MR, Mangelsdorf DJ 2002 Vitamin D receptor as an intestinal bile acid sensor. Science 296: 1313–1316. Thompson PD, Jurutka PW, Whitfield GK, Myskowski SM, Eichhorst KR, Dominguez CE, Haussler CA, Haussler MR 2002 Liganded VDR induces CYP3A4 in small intestinal and colon cancer cells via DR3 and ER6 vitamin D responsive elements. Biochem Biophys Res Commun 299:730–738. Zinser GM, Sundberg JP, Welsh J 2002 Vitamin D(3) receptor ablation sensitizes skin to chemically-induced tumorigenesis. Carcinogenesis 23:2103–2109. Garland CF, Garland FC 1980 Do sunlight and vitamin D reduce the likelihood of colon cancer? Int J Epidemiol 9:227–231. Garland CF, Comstock GW, Garland FC, Helsing KJ, Shaw EK, Gorham ED 1989 Serum 25-hydroxyvitamin D and colon cancer: eight-year prospective study. Lancet 2:1176–1178. Garland CF, Garland FC, Gorham ED 1991 Can colon cancer incidence and death rates be reduced with calcium and vitamin D? Am J Clin Nutr 54:193S–201S. Garland FC, Garland CF, Gorham ED, Young JF 1990 Geographic variation in breast cancer mortality in the United States: a hypothesis involving exposure to solar radiation. Prev Med 19:614–622. Schwartz GG, Hulka BS 1990 Is vitamin D deficiency a risk factor for prostate cancer? (Hypothesis). Anticancer Res 10:1307–1311. Gorham ED, Garland FC, Garland CF 1990 Sunlight and breast cancer incidence in the USSR. Int J Epidemiol 19:820–824. Hanchette CL, Schwartz GG 1992 Geographic patterns of prostate cancer mortality. Evidence for a protective effect of ultraviolet radiation. Cancer 70:2861–2869. Ainsleigh HG 1993 Beneficial effects of sun exposure on cancer mortality. Prev Med 22:132–140. Studzinski GP, Moore DC 1995 Sunlight—can it prevent as well as cause cancer? Cancer Res 55:4014–4022. Grant WB 2002 An estimate of premature cancer mortality in the U.S. due to inadequate doses of solar ultraviolet-B radiation. Cancer 94:1867–1875. de Gruijl FR, Longstreth J, Norval M, Cullen AP, Slaper H, Kripke ML, Takizawa Y, van der Leun JC 2003 Health effects from stratospheric ozone depletion and interactions with climate change. Photochem Photobiol Sci 2:16–28. Albert MR, Ostheimer KG 2003 The evolution of current medical and popular attitudes toward ultraviolet light exposure: Part 2. J Am Acad Dermatol 48:909–918. Luscombe CJ, French ME, Liu S, Saxby MF, Jones PW, Fryer AA, Strange RC 2001 Outcome in prostate cancer associations with skin type and polymorphism in pigmentation-related genes. Carcinogenesis 22:1343–1347. Luscombe CJ, French ME, Liu S, Saxby MF, Jones PW, Fryer AA, Strange RC 2001 Prostate cancer risk: associations with ultraviolet radiation, tyrosinase and melanocortin-1 receptor genotypes. Br J Cancer 85:1504–1509. Bodiwala D, Luscombe CJ, French ME, Liu S, Saxby MF, Jones PW, Ramachandran S, Fryer AA, Strange RC 2003 Susceptibility to prostate cancer: studies on interactions between UVR exposure and skin type. Carcinogenesis 24:711–717. Jablonski NG, Chaplin G 2002 Skin deep. Sci Am 287:74–81.

1587

CHAPTER 89 Vitamin D: Cancer and Differentiation

32. Jacobson EA, James KA, Newmark HL, Carroll KK 1989 Effects of dietary fat, calcium, and vitamin D on growth and mammary tumorigenesis induced by 7,12-dimethylbenz(a)anthracene in female Sprague-Dawley rats. Cancer Res 49:6300–6303. 33. Barger-Lux MJ, Heaney RP 1994 The role of calcium intake in preventing bone fragility, hypertension, and certain cancers. J Nutr 124:1406S–1411S. 34. Newmark HL 1994 Vitamin D adequacy: a possible relationship to breast cancer. Adv Exp Med Biol 364:109–114. 35. Kleibeuker JH, van der MR, de Vries EG 1995 Calcium and vitamin D: possible protective agents against colorectal cancer? Eur J Cancer 31A:1081–1084. 36. Garland CF, Garland FC, Gorham ED 1999 Calcium and vitamin D. Their potential roles in colon and breast cancer prevention. Ann NY Acad Sci 889:107–119. 37. McCullough ML, Robertson AS, Rodriguez C, Jacobs EJ, Chao A, Carolyn J, Calle EE, Willett WC, Thun MJ 2003 Calcium, vitamin D, dairy products, and risk of colorectal cancer in the cancer prevention study II nutrition cohort (United States). Cancer Causes Control 14:1–12. 38. Simard A, Vobecky J, Vobecky JS 1991 Vitamin D deficiency and cancer of the breast: an unprovocative ecological hypothesis. Can J Public Health 82:300–303. 39. Pence BC, Richard BC, Lawlis RS, Kuratko CN 1991 Effects of dietary calcium and vitamin D3 on tumor promotion in mouse skin. Nutr Cancer 16:171–181. 40. Ahonen MH, Tenkanen L, Teppo L, Hakama M, Tuohimaa P 2000 Prostate cancer risk and prediagnostic serum 25-hydroxyvitamin D levels (Finland). Cancer Causes Control 11:847–852. 41. Tuohimaa P, Lyakhovich A, Aksenov N, Pennanen P, Syvala H, Lou YR, Ahonen M, Hasan T, Pasanen P, Blauer M, Manninen T, Miettinen S, Vilja P, Ylikomi T 2001 Vitamin D and prostate cancer. J Steroid Biochem Mol Biol 76:125–134. 42. La Vecchia C, Braga C, Negri E, Franceschi S, Russo A, Conti E, Falcini F, Giacosa A, Montella M, Decarli A 1997 Intake of selected micronutrients and risk of colorectal cancer. Int J Cancer 73:525–530. 43. Lipkin M, Reddy B, Newmark H, Lamprecht SA 1999 Dietary factors in human colorectal cancer. Annu Rev Nutr 19:545–586. 44. Lu TT, Makishima M, Repa JJ, Schoonjans K, Kerr TA, Auwerx J, Mangelsdorf DJ 2000 Molecular basis for feedback regulation of bile acid synthesis by nuclear receptors. Mol Cell 6:507–515. 45. Kallay E, Pietschmann P, Toyokuni S, Bajna E, Hahn P, Mazzucco K, Bieglmayer C, Kato S, Cross HS 2001 Characterization of a vitamin D receptor knockout mouse as a model of colorectal hyperproliferation and DNA damage. Carcinogenesis 22:1429–1435. 46. Miller CW, Morosetti R, Campbell MJ, Mendoza S, Koeffler HP 1997 Integrity of the 1,25-dihydroxyvitamin D3 receptor in bone, lung, and other cancers. Mol Carcinog 19:254–257. 47. Taylor JA, Hirvonen A, Watson M, Pittman G, Mohler JL, Bell DA 1996 Association of prostate cancer with vitamin D receptor gene polymorphism. Cancer Res 56:4108–4110. 48. Ingles SA, Ross RK, Yu MC, Irvine RA, La Pera G, Haile RW, Coetzee GA 1997 Association of prostate cancer risk with genetic polymorphisms in vitamin D receptor and androgen receptor. J Natl Cancer Inst 89:166–170. 49. Ingles SA, Coetzee GA, Ross RK, Henderson BE, Kolonel LN, Crocitto L, Wang W, Haile RW 1998 Association of prostate

50.

51.

52.

53.

54.

55.

56.

57. 58. 59.

60.

61.

62.

63.

64.

cancer with vitamin D receptor haplotypes in AfricanAmericans. Cancer Res 58:1620–1623. Correa-Cerro L, Berthon P, Haussler J, Bochum S, Drelon E, Mangin P, Fournier G, Paiss T, Cussenot O, Vogel W 1999 Vitamin D receptor polymorphisms as markers in prostate cancer. Hum Genet 105:281–287. Habuchi T, Suzuki T, Sasaki R, Wang L, Sato K, Satoh S, Akao T, Tsuchiya N, Shimoda N, Wada Y, Koizumi A, Chihara J, Ogawa O, Kato T 2000 Association of vitamin D receptor gene polymorphism with prostate cancer and benign prostatic hyperplasia in a Japanese population. Cancer Res 60:305–308. Hamasaki T, Inatomi H, Katoh T, Ikuyama T, Matsumoto T 2001 Clinical and pathological significance of vitamin D receptor gene polymorphism for prostate cancer which is associated with a higher mortality in Japanese. Endocr J 48:543–549. Hamasaki T, Inatomi H, Katoh T, Ikuyama T, Matsumoto T 2002 Significance of vitamin D receptor gene polymorphism for risk and disease severity of prostate cancer and benign prostatic hyperplasia in Japanese. Urol Int 68:226–231. Medeiros R, Morais A, Vasconcelos A, Costa S, Pinto D, Oliveira J, Lopes C 2002 The role of vitamin D receptor gene polymorphisms in the susceptibility to prostate cancer of a southern European population. J Hum Genet 47:413–418. Xu Y, Shibata A, McNeal JE, Stamey TA, Feldman D, Peehl DM 2003 Vitamin D receptor start codon polymorphism (fokI) and prostate cancer progression. Cancer Epidemiol Biomarkers Prev 12:23–27. Watanabe M, Fukutome K, Murata M, Uemura H, Kubota Y, Kawamura J, Yatani R 1999 Significance of vitamin D receptor gene polymorphism for prostate cancer risk in Japanese. Anticancer Res 19:4511–4514. Blazer DG III, Umbach DM, Bostick RM, Taylor JA 2000 Vitamin D receptor polymorphisms and prostate cancer. Mol Carcinog 27:18–23. Gsur A, Madersbacher S, Haidinger G, Schatzl G, Marberger M, Vutuc C, Micksche M 2002 Vitamin D receptor gene polymorphism and prostate cancer risk. Prostate 51:30–34. Suzuki K, Matsui H, Ohtake N, Nakata S, Takei T, Koike H, Nakazato H, Okugi H, Hasumi M, Fukabori Y, Kurokawa K, Yamanaka H 2003 Vitamin D receptor gene polymorphism in familial prostate cancer in a Japanese population. Int J Urol 10:261–266. Tayeb MT, Clark C, Haites NE, Sharp L, Murray GI, McLeod HL 2003 CYP3A4 and VDR gene polymorphisms and the risk of prostate cancer in men with benign prostate hyperplasia. Br J Cancer 88:928–932. Curran JE, Vaughan T, Lea RA, Weinstein SR, Morrison NA, Griffiths LR 1999 Association of a vitamin D receptor polymorphism with sporadic breast cancer development. Int J Cancer 83:723–726. Lundin AC, Soderkvist P, Eriksson B, Bergman-Jungestrom M, Wingren S 1999 Association of breast cancer progression with a vitamin D receptor gene polymorphism. South-East Sweden Breast Cancer Group. Cancer Res 59:2332–2334. Ingles SA, Garcia DG, Wang W, Nieters A, Henderson BE, Kolonel LN, Haile RW, Coetzee GA 2000 Vitamin D receptor genotype and breast cancer in Latinas (United States). Cancer Causes Control 11:25–30. Bretherton-Watt D, Given-Wilson R, Mansi JL, Thomas V, Carter N, Colston KW 2001 Vitamin D receptor gene polymorphisms are associated with breast cancer risk in a UK Caucasian population. Br J Cancer 85:171–175.

1588 65. Hou MF, Tien YC, Lin GT, Chen CJ, Liu CS, Lin SY, Huang TJ 2002 Association of vitamin D receptor gene polymorphism with sporadic breast cancer in Taiwanese patients. Breast Cancer Res Treat 74:1–7. 66. Schondorf T, Eisberg C, Wassmer G, Warm M, Becker M, Rein DT, Gohring UJ 2003 Association of the vitamin D receptor genotype with bone metastases in breast cancer patients. Oncology 64:154–159. 67. Dunning AM, McBride S, Gregory J, Durocher F, Foster NA, Healey CS, Smith N, Pharoah PD, Luben RN, Easton DF, Ponder BA 1999 No association between androgen or vitamin D receptor gene polymorphisms and risk of breast cancer. Carcinogenesis 20:2131–2135. 68. Ingles SA, Wang J, Coetzee GA, Lee ER, Frankl HD, Haile RW 2001 Vitamin D receptor polymorphisms and risk of colorectal adenomas (United States). Cancer Causes Control 12:607–614. 69. Slatter ML, Yakumo K, Hoffman M, Neuhausen S 2001 Variants of the VDR gene and risk of colon cancer (United States). Cancer Causes Control 12:359–364. 70. Peters U, McGlynn KA, Chatterjee N, Gunter E, Garcia-Closas M, Rothman N, Sinha R 2001 Vitamin D, calcium, and vitamin D receptor polymorphism in colorectal adenomas. Cancer Epidemiol Biomarkers Prev 10:1267–1274. 71. Ramachandran S, Fryer AA, Smith AG, Lear JT, Bowers B, Hartland AJ, Whiteside JR, Jones PW, Strange RC 2001 Basal cell carcinomas: association of allelic variants with a high-risk subgroup of patients with the multiple presentation phenotype. Pharmacogenetics 11:247–254. 72. Ikuyama T, Hamasaki T, Inatomi H, Katoh T, Muratani T, Matsumoto T 2002 Association of vitamin D receptor gene polymorphism with renal cell carcinoma in Japanese. Endocr J 49:433–438. 73. Hutchinson PE, Osborne JE, Lear JT, Smith AG, Bowers PW, Morris PN, Jones PW, York C, Strange RC, Fryer AA 2000 Vitamin D receptor polymorphisms are associated with altered prognosis in patients with malignant melanoma. Clin Cancer Res 6:498–504. 74. Speer G, Dworak O, Cseh K, Bori Z, Salamon D, Torok I, Winkler G, Vargha P, Nagy Z, Takacs I, Kucsera M, Lakatos P 2000 Vitamin D receptor gene BsmI polymorphism correlates with erbB-2/HER-2 expression in human rectal cancer. Oncology 58:242–247. 75. Ma J, Stampfer MJ, Gann PH, Hough HL, Giovannucci E, Kelsey KT, Hennekens CH, Hunter DJ 1998 Vitamin D receptor polymorphisms, circulating vitamin D metabolites, and risk of prostate cancer in United States physicians. Cancer Epidemiol Biomarkers Prev 7:385–390. 76. Kim HS, Newcomb PA, Ulrich CM, Keener CL, Bigler J, Farin FM, Bostick RM, Potter JD 2001 Vitamin D receptor polymorphism and the risk of colorectal adenomas: evidence of interaction with dietary vitamin D and calcium. Cancer Epidemiol Biomarkers Prev 10:869–874. 77. Wong HL, Seow A, Arakawa K, Lee HP, Yu MC, Ingles SA 2003 Vitamin D receptor start codon polymorphism and colorectal cancer risk: effect modification by dietary calcium and fat in Singapore Chinese. Carcinogenesis 24:1091–1095. 78. Durrin LK, Haile RW, Ingles SA, Coetzee GA 1999 Vitamin D receptor 3′-untranslated region polymorphisms: lack of effect on mRNA stability. Biochim Biophys Acta 1453:311–320. 79. Colston K, Colston MJ, Feldman D 1981 1,25-dihydroxyvitamin D3 and malignant melanoma: the presence of receptors and inhibition of cell growth in culture. Endocrinology 108: 1083–1086.

JOHANNES P. T. M. VAN LEEUWEN AND HUIBERT A. P. POLS

80. Abe E, Miyaura C, Sakagami H, Takeda M, Konno K, Yamazaki T, Yoshiki S, Suda T 1981 Differentiation of mouse myeloid leukemia cells induced by 1 alpha,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 78:4990–4994. 81. Miyaura C, Abe E, Kuribayashi T, Tanaka H, Konno K, Nishii Y, Suda T 1981 1 alpha,25-dihydroxyvitamin D3 induces differentiation of human myeloid leukemia cells. Biochem Biophys Res Commun 102:937–943. 82. Suda T, Miyaura C, Abe E, Kuroki T 1986 Modulation of cell differentiation, immune responses and tumor promotion by vitamin D compounds. In: Peck WA (eds). Elsevier, Amsterdam pp. 1–48. 83. Suda T, Takahashi N, Abe E 1992 Role of vitamin D in bone resorption. J Cell Biochem 49:53–58. 84. Suda T, Udagawa N, Nakamura I, Miyaura C, Takahashi N 1995 Modulation of osteoclast differentiation by local factors. Bone 17:87S–91S. 85. Wu JC, Smith MW, Lawson DE 1992 Time dependency of 1,25(OH)2D3 induction of calbindin mRNA and calbindin expression in chick enterocytes during their differentiation along the crypt-villus axis. Differentiation 51:195–200. 86. Honma Y, Hozumi M, Abe E, Konno K, Fukushima M, Hata S, Nishii Y, DeLuca HF, Suda T 1983 1 alpha,25-dihydroxyvitamin D3 and 1 alpha-hydroxyvitamin D3 prolong survival time of mice inoculated with myeloid leukemia cells. Proc Natl Acad Sci USA 80:201–204. 87. Sato T, Takusagawa K, Asoo N, Konno K 1982 Antitumor effect of 1 alpha-hydroxyvitamin D3. Tohoku J Exp Med 138:445–446. 88. Wood AW, Chang RL, Huang MT, Uskokovic M, Conney AH 1983 1 alpha, 25-dihydroxyvitamin D3 inhibits phorbol esterdependent chemical carcinogenesis in mouse skin. Biochem Biophys Res Commun 116:605–611. 89. Chida K, Hashiba H, Fukushima M, Suda T, Kuroki T 1985 Inhibition of tumor promotion in mouse skin by 1 alpha, 25-dihydroxyvitamin D3. Cancer Res 45:5426–5430. 90. Eisman JA, Barkla DH, Tutton PJ 1987 Suppression of in vivo growth of human cancer solid tumor xenografts by 1,25-dihydroxyvitamin D3. Cancer Res 47:21–25. 91. Cohen SM, Saulenas AM, Sullivan CR, Albert DM 1988 Further studies of the effect of vitamin D on retinoblastoma. Inhibition with 1,25-dihydroxycholecalciferol. Arch Ophthalmol 106:541–543. 92. Eisman JA, Koga M, Sutherland RL, Barkla DH, Tutton PJ 1989 1,25-dihydroxyvitamin D3 and the regulation of human cancer cell replication. Proc Soc Exp Biol Med 191: 221–226. 93. Colston KW, Berger U, Coombes RC 1989 Possible role for vitamin D in controlling breast cancer cell proliferation. Lancet 1:188–191. 94. Albert DM, Marcus DM, Gallo JP, O’Brien JM 1992 The antineoplastic effect of vitamin D in transgenic mice with retinoblastoma. Invest Ophthalmol Vis Sci 33:2354–2364. 95. Belleli A, Shany S, Levy J, Guberman R, Lamprecht SA 1992 A protective role of 1,25-dihydroxyvitamin D3 in chemically-induced rat colon carcinogenesis. Carcinogenesis 13:2293–2298. 96. Iino Y, Yoshida M, Sugamata N, Maemura M, Ohwada S, Yokoe T, Ishikita T, Horiuchi R, Morishita Y 1992 1 alphahydroxyvitamin D3, hypercalcemia, and growth suppression of 7,12-dimethylbenz[a]anthracene-induced rat mammary tumors. Breast Cancer Res Treat 22:133–140. 97. Haq M, Kremer R, Goltzman D, Rabbani SA 1993 A vitamin D analog (EB1089) inhibits parathyroid hormonerelated peptide production and prevents the development of

1589

CHAPTER 89 Vitamin D: Cancer and Differentiation

98.

99.

100.

101.

102.

103.

104.

105. 106.

107.

108.

109.

110. 111. 112.

malignancy-associated hypercalcemia in vivo. J Clin Invest 91:2416–2422. Tsuchiya H, Morishita H, Tomita K, Ueda Y, Tanaka M 1993 Differentiating and antitumor activities of 1 alpha,25-dihydroxyvitamin D3 in vitro and 1 alpha-hydroxyvitamin D3 in vivo on human osteosarcoma. J Orthop Res 11:122–130. Young MR, Halpin J, Hussain R, Lozano Y, Djordjevic A, Devata S, Matthews JP, Wright MA 1993 Inhibition of tumor production of granulocyte-macrophage colony-stimulating factor by 1 alpha, 25-dihydroxyvitamin D3 reduces tumor motility and metastasis. Invasion Metastasis 13:169–177. Cohen-Solal ME, Bouizar Z, Denne MA, Graulet AM, Gueris J, Bracq S, Jullienne A, De Vernejoul MC 1995 1,25 dihydroxyvitamin D and dexamethasone decrease in vivo Walker carcinoma growth, but not parathyroid hormonerelated protein secretion. Horm Metab Res 27:403–407. Young MR, Ihm J, Lozano Y, Wright MA, Prechel MM 1995 Treating tumor-bearing mice with vitamin D3 diminishes tumor-induced myelopoiesis and associated immunosuppression, and reduces tumor metastasis and recurrence. Cancer Immunol Immunother 41:37–45. Young MR, Lozano Y, Ihm J, Wright MA, Prechel MM 1996 Vitamin D3 treatment of tumor bearers can stimulate immune competence and reduce tumor growth when treatment coincides with a heightened presence of natural suppressor cells. Cancer Lett 104:153–161. Getzenberg RH, Light BW, Lapco PE, Konety BR, Nangia AK, Acierno JS, Dhir R, Shurin Z, Day RS, Trump DL, Johnson CS 1997 Vitamin D inhibition of prostate adenocarcinoma growth and metastasis in the Dunning rat prostate model system. Urology 50:999–1006. Lokeshwar BL, Schwartz GG, Selzer MG, Burnstein KL, Zhuang SH, Block NL, Binderup L 1999 Inhibition of prostate cancer metastasis in vivo: a comparison of 1,23dihydroxyvitamin D (calcitriol) and EB1089. Cancer Epidemiol Biomarkers Prev 8:241–248. Masood R, Nagpal S, Zheng T, Cai J, Tulpule A, Smith DL, Gill PS 2000 Kaposi sarcoma is a therapeutic target for vitamin D(3) receptor agonist. Blood 96:3188–3194. Polek TC, Murthy S, Blutt SE, Boehm MF, Zou A, Weigel NL, Allegretto EA 2001 Novel nonsecosteroidal vitamin D receptor modulator inhibits the growth of LNCaP xenograft tumors in athymic mice without increased serum calcium. Prostate 49:224–233. Kunakornsawat S, Rosol TJ, Capen CC, Reddy GS, Binderup L, Inpanbutr N 2002 Effects of 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] and its analogs (EB1089 and analog V) on canine adenocarcinoma (CAC-8) in nude mice. Biol Pharm Bull 25:642–647. Oades GM, Dredge K, Kirby RS, Colston KW 2002 Vitamin D receptor-dependent antitumor effects of 1,25-dihydroxyvitamin D3 and two synthetic analogs in three in vivo models of prostate cancer. BJU Int 90:607–616. Vegesna V, O’Kelly J, Said J, Uskokovic M, Binderup L, Koeffle HP 2003 Ability of potent vitamin D3 analogs to inhibit growth of prostate cancer cells in vivo. Anticancer Res 23:283–289. Pols HA, Birkenhager JC, van Leeuwen JP 1994 Vitamin D analogs: from molecule to clinical application. Clin Endocrinol (Oxf) 40:285–292. Bouillon R, Okamura WH, Norman AW 1995 Structurefunction relationships in the vitamin D endocrine system. Endocr Rev 16:200–257. Zhou JY, Norman AW, Chen DL, Sun GW, Uskokovic M, Koeffler HP 1990 1,25-dihydroxy-16-ene-23-yne-vitamin D3

113.

114.

115.

116.

117.

118.

119.

120.

121.

122. 123.

124.

125.

126.

prolongs survival time of leukemic mice. Proc Natl Acad Sci USA 87:3929–3932. Abe J, Nakano T, Nishii Y, Matsumoto T, Ogata E, Ikeda K 1991 A novel vitamin D3 analog, 22-oxa-1,25-dihydroxyvitamin D3, inhibits the growth of human breast cancer in vitro and in vivo without causing hypercalcemia. Endocrinology 129:832–837. Colston K, Mackay AG, Chandler S, Binderup L, Coombes RC 1991 Novel vitamin D analogs suppress tumour growth in vivo. In: Norman AW, Bouillon R, Thomasset M, (eds) Vitamin D: Gene regulation, structure-function analysis and clinical application. De Gruyter: pp.465–466. Oikawa T, Yoshida Y, Shimamura M, Ashino-Fuse H, Iwaguchi T, Tominaga T 1991 Antitumor effect of 22-oxa1 alpha,25-dihydroxyvitamin D3, a potent angiogenesis inhibitor, on rat mammary tumors induced by 7,12-dimethylbenz[a]anthracene. Anticancer Drugs 2:475–480. Colston KW, Chander SK, Mackay AG, Coombes RC 1992 Effects of synthetic vitamin D analogs on breast cancer cell proliferation in vivo and in vitro. Biochem Pharmacol 44:693–702. Colston KW, Mackay AG, James SY, Binderup L, Chander S, Coombes RC 1992 EB1089: a new vitamin D analog that inhibits the growth of breast cancer cells in vivo and in vivo. Biochem Pharmacol 44:2273–2280. Abe-Hashimoto J, Kikuchi T, Matsumoto T, Nishii Y, Ogata E, Ikeda K 1993 Antitumor effect of 22-oxa-calcitriol, a noncalcemic analog of calcitriol, in athymic mice implanted with human breast carcinoma and its synergism with tamoxifen. Cancer Res 53:2534–2537. Anzano MA, Smith JM, Uskokovic MR, Peer CW, Mullen LT, Letterio JJ, Welsh MC, Shrader MW, Logsdon DL, Driver CL, 1994 1 alpha,25-dihydroxy-16-ene-23-yne-26,27hexafluorocholecalciferol (Ro24–5531), a new deltanoid (vitamin D analog) for prevention of breast cancer in the rat. Cancer Res 54:1653–1656. Tanaka Y, Wu AY, Ikekawa N, Iseki K, Kawai M, Kobayashi Y 1994 Inhibition of HT-29 human colon cancer growth under the renal capsule of severe combined immunodeficient mice by an analog of 1,25-dihydroxyvitamin D3, DD-003. Cancer Res 54:5148–5153. Otoshi T, Iwata H, Kitano M, Nishizawa Y, Morii H, Yano Y, Otani S, Fukushima S 1995 Inhibition of intestinal tumor development in rat multi-organ carcinogenesis and aberrant crypt foci in rat colon carcinogenesis by 22-oxa-calcitriol, a synthetic analog of 1 alpha, 25-dihydroxyvitamin D3. Carcinogenesis 16:2091–2097. Schwartz GG, Hill CC, Oeler TA, Becich MJ, Bahnson RR 1995 1,25-dihydroxy-16-ene-23-yne-vitamin D3 and prostate cancer cell proliferation in vivo. Urology 46:365–369. Wali RK, Bissonnette M, Khare S, Hart J, Sitrin MD, Brasitus TA 1995 1 alpha,25-dihydroxy-16-ene-23-yne26,27-hexafluorocholecalciferol, a noncalcemic analog of 1 alpha,25-dihydroxyvitamin D3, inhibits azoxymethaneinduced colonic tumorigenesis. Cancer Res 55:3050–3054. Akhter J, Chen X, Bowrey P, Bolton EJ, Morris DL 1997 Vitamin D3 analog, EB1089, inhibits growth of subcutaneous xenografts of the human colon cancer cell line, LoVo, in a nude mouse model. Dis Colon Rectum 40:317–321. VanWeelden K, Flanagan L, Binderup L, Tenniswood M, Welsh J 1998 Apoptotic regression of MCF-7 xenografts in nude mice treated with the vitamin D3 analog, EB1089. Endocrinology 139:2102–2110. Nickerson T, Huynh H 1999 Vitamin D analog EB1089induced prostate regression is associated with increased gene

1590

127. 128.

129.

130.

131. 132. 133.

134. 135. 136. 137. 138. 139. 140.

141.

142.

143. 144.

145.

expression of insulin-like growth factor binding proteins. J Endocrinol 160:223–229. Blutt SE, Polek TC, Stewart LV, Kattan MW, Weigel NL 2000 A calcitriol analog, EB1089, inhibits the growth of LNCaP tumors in nude mice. Cancer Res 60:779–782. Dawson DG, Gleiser J, Zimbric ML, Darjatmoko SR, Lindstrom MJ, Strugnell SA, Albert DM 2003 Toxicity and dose-response studies of 1-alpha hydroxyvitamin D2 in LH-beta-tag transgenic mice. Ophthalmology 110:835–839. Mehta RG, Hussain EA, Mehta RR, Das Gupta TK 2003 Chemoprevention of mammary carcinogenesis by 1alphahydroxyvitamin D5, a synthetic analog of vitamin D. Mutat Res 523–524:253–264. Cunningham D, Gilchrist NL, Cowan RA, Forrest GJ, McArdle CS, Soukop M 1985 Alfacalcidol as a modulator of growth of low grade non-Hodgkin’s lymphomas. Br Med J (Clin Res Ed) 291:1153–1155. Raina V, Cunningham D, Gilchrist N, Soukop M 1991 Alfacalcidol is a nontoxic, effective treatment of follicular, small-cleaved cell lymphoma. Br J Cancer 63:463–465. Koeffler HP, Hirji K, Itri L 1985 1,25-dihydroxyvitamin D3: in vivo and in vitro effects on human preleukemic and leukemic cells. Cancer Treat Rep 69:1399–1407. Kelsey SM, Newland AC, Cunningham J, Makin HL, Coldwell RD, Mills MJ, Grant IR 1992 Sustained hematological response to high-dose oral alfacalcidol in patients with myelodysplastic syndromes. Lancet 340:316–317. French LE, Ramelet AA, Saurat JH 1994 Remission of cutaneous T-cell lymphoma with combined calcitriol and acitretin. Lancet 344:686–687. Majewski S, Skopinska M, Bollag W, Jablonska S 1994 Combination of isotretinoin and calcitriol for precancerous and cancerous skin lesions. Lancet 344:1510–1511. Thomsen K 1995 Cutaneous T-cell lymphoma and calcitriol and isotretinoin treatment. Lancet 345:1583. French LE, Saurat JH 1995 Treatment of cutaneous T-cell lymphoma by retinoids and calcitriol. Lancet 346:376. Bagot M 1995 Treatment of cutaneous T-cell lymphoma by retinoids and calcitriol. Lancet 346:376–377. Gross C, Stamey T, Hancock S, Feldman D 1998 Treatment of early recurrent prostate cancer with 1,25-dihydroxyvitamin D3 (calcitriol). J Urol 159:2035–2039. Beer TM, Lemmon D, Lowe BA, Henner WD 2003 Highdose weekly oral calcitriol in patients with a rising PSA after prostatectomy or radiation for prostate carcinoma. Cancer 97:1217–1224. Smith DC, Johnson CS, Freeman CC, Muindi J, Wilson JW, Trump DL 1999 A phase I trial of calcitriol (1,25-dihydroxycholecalciferol) in patients with advanced malignancy. Clin Cancer Res 5:1339–1345. Beer TM, Munar M, Henner WD 2001 A phase I trial of pulse calcitriol in patients with refractory malignancies: pulse dosing permits substantial dose escalation. Cancer 91:2431–2439. Bower M, Colston KW, Stein RC, Hedley A, Gazet JC, Ford HT, Combes RC 1991 Topical calcipotriol treatment in advanced breast cancer. Lancet 337:701–702. Gulliford T, English J, Colston KW, Menday P, Moller S, Coombes RC 1998 A phase I study of the vitamin D analog EB 1089 in patients with advanced breast and colorectal cancer. Br J Cancer 78:6–13. Evans TR, Colston KW, Lofts FJ, Cunningham D, Anthoney DA, Gogas H, de Bono JS, Hamberg KJ, Skov T, Mansi JL 2002 A phase II trial of the vitamin D analog

JOHANNES P. T. M. VAN LEEUWEN AND HUIBERT A. P. POLS

146.

147. 148.

149.

150.

151.

152. 153.

154.

155.

156.

157. 158. 159. 160. 161.

162.

Seocalcitol (EB1089) in patients with inoperable pancreatic cancer. Br J Cancer 86:680–685. Majewski S, Szmurlo A, Marczak M, Jablonska S, Bollag W 1993 Inhibition of tumor cell-induced angiogenesis by retinoids, 1,25-dihydroxyvitamin D3 and their combination. Cancer Lett 75:35–39. Bollag W, Majewski S, Jablonska S 1994 Cancer combination chemotherapy with retinoids: experimental rationale. Leukemia 8:1453–1457. Majewski S, Marczak M, Szmurlo A, Jablonska S, Bollag W 1995 Retinoids, interferon alpha, 1,25-dihydroxyvitamin D3 and their combination inhibit angiogenesis induced by nonHPV-harboring tumor cell lines. RAR alpha mediates the antiangiogenic effect of retinoids. Cancer Lett 89:117–124. Bernardi RJ, Johnson CS, Modzelewski RA, Trump DL 2002 Antiproliferative effects of 1alpha,25-dihydroxyvitamin D(3) and vitamin D analogs on tumor-derived endothelial cells. Endocrinology 143:2508–2514. Shokravi MT, Marcus DM, Alroy J, Egan K, Saornil MA, Albert DM 1995 Vitamin D inhibits angiogenesis in transgenic murine retinoblastoma. Invest Ophthalmol Vis Sci 36:83–87. Ding I, Sun JZ, Fenton B, Liu WM, Kimsely P, Okunieff P, Min W 2001 Intratumoral administration of endostatin plasmid inhibits vascular growth and perfusion in MCa-4 murine mammary carcinomas. Cancer Res 61:526–531. Marneros AG, Olsen BR 2001 The role of collagen-derived proteolytic fragments in angiogenesis. Matrix Biol 20:337–345. Kanda S, Mochizuki Y, Miyata Y, Kanetake H, Yamamoto N 2002 Effects of vitamin D(3)–binding protein-derived macrophage activating factor (GcMAF) on angiogenesis. J Natl Cancer Inst 94:1311–1319. Kisker O, Onizuka S, Becker CM, Fannon M, Flynn E, D’Amato R, Zetter B, Folkman J, Ray R, Swamy N, Pirie-Shepherd S 2003 Vitamin D binding proteinmacrophage activating factor (DBP-maf) inhibits angiogenesis and tumor growth in mice. Neoplasia 5:32–40. Nowicka D, Zagozdzon R, Majewski S, Marczak M, Jablonska S, Bollag W 1998 Calcitriol enhances antineoplastic and antiangiogenic effects of interleukin-12. Arch Dermatol Res 290:696–700. Hansen CM, Frandsen TL, Brunner N, Binderup L 1994 1 alpha,25-dihydroxyvitamin D3 inhibits the invasive potential of human breast cancer cells in vitro. Clin Exp Metastasis 12:195–202. Koli K, Keski-Oja J 2000 1alpha,25-dihydroxyvitamin D3 and its analogs down-regulate cell invasion-associated proteases in cultured malignant cells. Cell Growth Differ 11:221–229. Arguello F, Baggs RB, Frantz CN 1988 A murine model of experimental metastasis to bone and bone marrow. Cancer Res 48:6876–6881. Agha FP, Norman A, Hirschl S, Klein R 1976 Paget’s disease. Coexistence with metastatic carcinoma. N Y State J Med 76:734–735. Orr W, Varani J, Gondex MK, Ward PA, Mundy GR 1979 Chemotactic responses of tumor cells to products of resorbing bone. Science 203:176–179. Sasaki A, Boyce BF, Story B, Wright KR, Chapman M, Boyce R, Mundy GR, Yoneda T 1995 Bisphosphonate risedronate reduces metastatic human breast cancer burden in bone in nude mice. Cancer Res 55:3551–3557. Silver J, Naveh-Many T, Mayer H, Schmelzer HJ, Popovtzer MM 1986 Regulation by vitamin D metabolites of

1591

CHAPTER 89 Vitamin D: Cancer and Differentiation

163.

164. 165.

166.

167.

168.

169.

170.

171.

172.

173. 174. 175. 176.

177.

178.

parathyroid hormone gene transcription in vivo in the rat. J Clin Invest 78:1296–1301. Mackey SL, Heymont JL, Kronenberg HM, Demay MB 1996 Vitamin D receptor binding to the negative human parathyroid hormone vitamin D response element does not require the retinoid x receptor. Mol Endocrinol 10:298–305. Moseley JM, Gillespie MT 1995 Parathyroid hormonerelated protein. Crit Rev Clin Lab Sci 32:299–343. Sebag M, Henderson J, Goltzman D, Kremer R 1994 Regulation of parathyroid hormone-related peptide production in normal human mammary epithelial cells in vitro. Am J Physiol 267:C723–C730. Werkmeister JR, Merryman JI, McCauley LK, Horton JE, Capen CC, Rosol TJ 1993 Parathyroid hormone-related protein production by normal human keratinocytes in vitro. Exp Cell Res 208:68–74. Kremer R, Karaplis AC, Henderson J, Gulliver W, Banville D, Hendy GN, Goltzman D 1991 Regulation of parathyroid hormone-like peptide in cultured normal human keratinocytes. Effect of growth factors and 1,25 dihydroxyvitamin D3 on gene expression and secretion. J Clin Invest 87:884–893. Henderson J, Sebag M, Rhim J, Goltzman D, Kremer R 1991 Dysregulation of parathyroid hormone-like peptide expression and secretion in a keratinocyte model of tumor progression. Cancer Res 51:6521–6528. Yu J, Papavasiliou V, Rhim J, Goltzman D, Kremer R 1995 Vitamin D analogs: new therapeutic agents for the treatment of squamous cancer and its associated hypercalcemia. Anticancer Drugs 6:101–108. Falzon M 1997 The noncalcemic vitamin D analogs EB1089 and 22-oxacalcitriol interact with the vitamin D receptor and suppress parathyroid hormone-related peptide gene expression. Mol Cell Endocrinol 127:99–108. Inoue D, Matsumoto T, Ogata E, Ikeda K 1993 22-oxacalcitriol, a noncalcemic analog of calcitriol, suppresses both cell proliferation and parathyroid hormone-related peptide gene expression in human T-cell lymphotrophic virus, type Iinfected T-cells. J Biol Chem 268:16730–16736. Liu B, Goltzman D, Rabbani SA 1993 Regulation of parathyroid hormone-related peptide production in vitro by the rat hypercalcemic Leydig cell tumor H-500. Endocrinology 132:1658–1664. Tovar SV, Falzon M 2002 Regulation of PTH-related protein gene expression by vitamin D in PC-3 prostate cancer cells. Mol Cell Endocrinol 190:115–124. Shen X, Falzon M 2003 Parathyroid hormone-related protein up-regulates integrin expression via an intracrine pathway in PC-3 prostate cancer cells. Regul Pept 113:17–29. Shen X, Falzon M 2003 PTH-related protein modulates PC-3 prostate cancer cell adhesion and integrin subunit profile. Mol Cell Endocrinol 199:165–177. El Abdaimi K, Papavasiliou V, Rabbani SA, Rhim JS, Goltzman D, Kremer R 1999 Reversal of hypercalcemia with the vitamin D analog EB1089 in a human model of squamous cancer. Cancer Res 59:3325–3328. Endo K, Ichikawa F, Uchiyama Y, Katsumata K, Ohkawa H, Kumaki K, Ogata E, Ikeda K 1994 Evidence for the uptake of a vitamin D analog (OCT) by a human carcinoma and its effect of suppressing the transcription of parathyroid hormone-related peptide gene in vivo. J Biol Chem 269:32693–32699. Merryman JI, Capen CC, McCauley LK, Werkmeister JR, Suter MM, Rosol TJ 1993 Regulation of parathyroid

179.

180. 181.

182.

183. 184.

185.

186.

187. 188.

189.

190.

191.

192.

193.

hormone-related protein production by a squamous carcinoma cell line in vitro. Lab Invest 69:347–354. Kunakornsawat S, Rosol TJ, Capen CC, Middleton RP, Hannah SS, Inpanbutr N 2001 Effects of 1,25(OH)2D3, EB1089, and analog V on PTHrP production, PTHrP mRNA expression and cell growth in SCC 2/88. Anticancer Res 21:3355–3363. Chatterjee M 2001 Vitamin D and genomic stability. Mutat Res 475:69–87. Saha BK, Bishayee A, Kanjilal NB, Chatterjee M 2001 1Alpha,25-dihydroxyvitamin D3 inhibits hepatic chromosomal aberrations, DNA strand breaks and specific DNA adducts during rat hepatocarcinogenesis. Cell Mol Life Sci 58:1141–1149. Akutsu N, Lin R, Bastien Y, Bestawros A, Enepekides DJ, Black MJ, White JH 2001 Regulation of gene expression by 1alpha,25-dihydroxyvitamin D3 and its analog EB1089 under growth-inhibitory conditions in squamous carcinoma cells. Mol Endocrinol 15:1127–1139. Gartel AL, Shchors K 2003 Mechanisms of c-myc-mediated transcriptional repression of growth arrest genes. Exp Cell Res 283:17–21. Karmali R, Bhalla AK, Farrow SM, Williams MM, Lal S, Lydyard PM, O’Riordan JL 1989 Early regulation of c-myc mRNA by 1,25-dihydroxyvitamin D3 in human myelomonocytic U937 cells. J Mol Endocrinol 3:43–48. Koizumi T, Nakao Y, Kawanishi M, Maeda S, Sugiyama T, Fujita T 1989 Suppression of c-myc mRNA expression by steroid hormones in HTLV-I-infected T-cell line, KH-2. Int J Cancer 44:701–706. Moore DC, Carter DL, Studzinski GP 1992 Inhibition by 1,25 dihydroxyvitamin D3 of c-myc down-regulation and DNA fragmentation in cytosine arabinoside-induced erythroid differentiation of K562 cells. J Cell Physiol 151:539–548. Sebag M, Henderson J, Rhim J, Kremer R 1992 Relative resistance to 1,25-dihydroxyvitamin D3 in a keratinocyte model of tumor progression. J Biol Chem 267:12162–12167. Saunders DE, Christensen C, Wappler NL, Schultz JF, Lawrence WD, Malviya VK, Malone JM, Deppe G 1993 Inhibition of c-myc in breast and ovarian carcinoma cells by 1,25-dihydroxyvitamin D3, retinoic acid and dexamethasone. Anticancer Drugs 4:201–208. Kasukabe T, Okabe-Kado J, Hozumi M, Honma Y 1994 Inhibition by interleukin 4 of leukemia inhibitory factor-, interleukin 6-, and dexamethasone-induced differentiation of mouse myeloid leukemia cells: role of c-myc and junB protooncogenes. Cancer Res 54:592–597. Taoka T, Collins ED, Irino S, Norman AW 1993 1,25(OH)2vitamin D3 mediated changes in mRNA for c-myc and 1,25(OH)2D3 receptor in HL-60 cells and related subclones. Mol Cell Endocrinol 95:51–57. Stio M, Celli A, Treves C 2001 Synergistic antiproliferative effects of vitamin D derivatives and 9-cis retinoic acid in SH-SY5Y human neuroblastoma cells. J Steroid Biochem Mol Biol 77:213–222. Vink-van Wijngaarden T, Pols HA, Buurman CJ, van den Bemd GJ, Dorssers LC, Birkenhager JC, van Leeuwen JP 1994 Inhibition of breast cancer cell growth by combined treatment with vitamin D3 analogs and tamoxifen. Cancer Res 54:5711–5717. Hulla W, Kallay E, Krugluger W, Peterlik M, Cross HS 1995 Growth control of human colon-adenocarcinoma-derived Caco-2 cells by vitamin–D compounds and extracellular

1592

194.

195.

196.

197.

198.

199.

200.

201.

202.

203.

204.

205.

206.

207.

calcium in vitro: relation to c-myc-oncogene and vitamin-Dreceptor expression. Int J Cancer 62:711–716. Trydal T, Lillehaug JR, Aksnes L, Aarskog D 1992 Regulation of cell growth, c-myc mRNA, and 1,25-(OH)2 vitamin D3 receptor in C3H/10T1/2 mouse embryo fibroblasts by calcipotriol and 1,25-(OH)2 vitamin D3. Acta Endocrinol (Copenh) 126:75–79. Paatero GI, Trydal T, Karlstedt KA, Aarskog D, Lillehaug JR 1994 Time-course study of 1,25-(OH)2-vitamin D3 induction of homologous receptor and c-myc in nontransformed and transformed C3H/10T1/2 cell clones. Int J Biochem 26:367–374. Mahonen A, Pirskanen A, Maenpaa PH 1991 Homologous and heterologous regulation of 1,25-dihydroxyvitamin D-3 receptor mRNA levels in human osteosarcoma cells. Biochim Biophys Acta 1088:111–118. van den Bemd GJ, Pols HA, Birkenhager JC, Kleinekoort WM, van Leeuwen JP 1995 Differential effects of 1,25-dihydroxyvitamin D3-analogs on osteoblast-like cells and on in vitro bone resorption. J Steroid Biochem Mol Biol 55:337–346. Brelvi ZS, Christakos S, Studzinski GP 1986 Expression of monocyte-specific oncogenes c-fos and c-fms in HL60 cells treated with vitamin D3 analogs correlates with inhibition of DNA synthesis and reduced calmodulin concentration. Lab Invest 55:269–275. Huh N, Satoh M, Nose K, Abe E, Suda T, Rajewsky MF, Kuroki T 1987 1 alpha,25-dihydroxyvitamin D3 induces anchorage-independent growth and c-Ki-ras expression of BALB/3T3 and NIH/3T3 cells. Jpn J Cancer Res 78:99–102. Lasky SR, Iwata K, Rosmarin AG, Caprio DG, Maizel AL 1995 Differential regulation of JunD by dihydroxycholecalciferol in human chronic myelogenous leukemia cells. J Biol Chem 270:19676–19679. Khare S, Bolt MJ, Wali RK, Skarosi SF, Roy HK, Niedziela S, Scaglione-Sewell B, Aquino B, Abraham C, Sitrin MD, Brasitus TA, Bissonnette M 1997 1,25 dihydroxyvitamin D3 stimulates phospholipase C-gamma in rat colonocytes: role of c-Src in PLC-gamma activation. J Clin Invest 99:1831–1841. Khare S, Bissonnette M, Wali R, Skarosi S, Boss GR, von Lintig FC, Scaglione-Sewell B, Sitrin MD, Brasitus TA 1999 1,25-dihydroxyvitamin D3 but not TPA activates PLD in Caco-2 cells via pp60(c-src) and RhoA. Am J Physiol 276:G1005–G1015. Johansen C, Kragballe K, Henningsen J, Westergaard M, Kristiansen K, Iversen L 2003 1alpha,25-dihydroxyvitamin D3 stimulates activator protein 1 DNA-binding activity by a phosphatidylinositol 3-kinase/Ras/MEK/extracellular signal regulated kinase 1/2 and c-Jun N-terminal kinase 1-dependent increase in c-Fos, Fra1, and c-Jun expression in human keratinocytes. J Invest Dermatol 120:561–570. Lasky SR, Posner MR, Iwata K, Santos-Moore A, Yen A, Samuel V, Clark J, Maizel AL 1994 Characterization of a vitamin D3–resistant human chronic myelogenous leukemia cell line. Blood 84:4283–4294. Fan FS, Yu WC 1995 1,25-dihydroxyvitamin D3 suppresses cell growth, DNA synthesis, and phosphorylation of retinoblastoma protein in a breast cancer cell line. Cancer Invest 13:280–286. Yen A, Chandler S, Forbes ME, Fung YK, T’Ang A, Pearson R 1992 Coupled down-regulation of the RB retinoblastoma and c-myc genes antecedes cell differentiation: possible role of RB as a “status quo” gene. Eur J Cell Biol 57:210–221. Yen A, Varvayanis S 1994 Late dephosphorylation of the RB protein in G2 during the process of induced cell differentiation. Exp Cell Res 214:250–257.

JOHANNES P. T. M. VAN LEEUWEN AND HUIBERT A. P. POLS

208. Kobayashi T, Hashimoto K, Yoshikawa K 1993 Growth inhibition of human keratinocytes by 1,25-dihydroxyvitamin D3 is linked to dephosphorylation of retinoblastoma gene product. Biochem Biophys Res Commun 196:487–493. 209. Ehinger M, Bergh G, Olofsson T, Baldetorp B, Olsson I, Gullberg U 1996 Expression of the p53 tumor suppressor gene induces differentiation and promotes induction of differentiation by 1,25-dihydroxycholecalciferol in leukemic U-937 cells. Blood 87:1064–1074. 210. Davoust N, Wion D, Chevalier G, Garabedian M, Brachet P, Couez D 1998 Vitamin D receptor stable transfection restores the susceptibility to 1,25-dihydroxyvitamin D3 cytotoxicity in a rat glioma resistant clone. J Neurosci Res 52:210–219. 211. Elstner E, Linker-Israeli M, Said J, Umiel T, de Vos S, Shintaku IP, Heber D, Binderup L, Uskokovic M, Koeffler HP 1995 20-epi-vitamin D3 analogs: a novel class of potent inhibitors of proliferation and inducers of differentiation of human breast cancer cell lines. Cancer Res 55:2822–2830. 212. Pepper C, Thomas A, Hoy T, Milligan D, Bentley P, Fegan C 2003 The vitamin D3 analog EB1089 induces apoptosis via a p53-independent mechanism involving p38 MAP kinase activation and suppression of ERK activity in B-cell chronic lymphocytic leukemia cells in vitro. Blood 101:2454–2460. 213. Lee TH, Pelletier J 2001 Functional characterization of WT1 binding sites within the human vitamin D receptor gene promoter. Physiol Genomics 7:187–200. 214. Chen KS, DeLuca HF 1994 Isolation and characterization of a novel cDNA from HL-60 cells treated with 1,25-dihydroxyvitamin D-3. Biochim Biophys Acta 1219:26–32. 215. Nishiyama A, Matsui M, Iwata S, Hirota K, Masutani H, Nakamura H, Takagi Y, Sono H, Gon Y, Yodoi J 1999 Identification of thioredoxin-binding protein-2/vitamin D(3) up-regulated protein 1 as a negative regulator of thioredoxin function and expression. J Biol Chem 274:21645–21650. 216. Nishiyama A, Masutani H, Nakamura H, Nishinaka Y, Yodoi J 2001 Redox regulation by thioredoxin and thioredoxin-binding proteins. IUBMB Life 52:29–33. 217. Ikarashi M, Takahashi Y, Ishii Y, Nagata T, Asai S, Ishikawa K 2002 Vitamin D3 up-regulated protein 1 (VDUP1) expression in gastrointestinal cancer and its relation to stage of disease. Anticancer Res 22:4045–4048. 218. Takahashi Y, Nagata T, Ishii Y, Ikarashi M, Ishikawa K, Asai S 2002 Up-regulation of vitamin D3 up-regulated protein 1 gene in response to 5-fluorouracil in colon carcinoma SW620. Oncol Rep 9:75–79. 219. Schulze PC, De Keulenaer GW, Yoshioka J, Kassik KA, Lee RT 2002 Vitamin D3 up-regulated protein-1 (VDUP-1) regulates redox-dependent vascular smooth muscle cell proliferation through interaction with thioredoxin. Circ Res 91:689–695. 220. Wang Y, De Keulenaer GW, Lee RT 2002 Vitamin D(3) up-regulated protein-1 is a stress-responsive gene that regulates cardiomyocyte viability through interaction with thioredoxin. J Biol Chem 277:26496–26500. 221. Han SH, Jeon JH, Ju HR, Jung U, Kim KY, Yoo HS, Lee YH, Song KS, Hwang HM, Na YS, Yang Y, Lee KN, Choi I 2003 VDUP1 up-regulated by TGF-beta1 and 1,25-dihydroxyvitamin D(3) inhibits tumor cell growth by blocking cell-cycle progression. Oncogene 22:4035–4046. 222. Song H, Cho D, Jeon JH, Han SH, Hur DY, Kim YS, Choi I 2003 Vitamin D(3) up-regulating protein 1 (VDUP1) antisense DNA regulates tumorigenicity and melanogenesis of murine melanoma cells via regulating the expression of fas ligand and reactive oxygen species. Immunol Lett 86:235–247.

CHAPTER 89 Vitamin D: Cancer and Differentiation

223. Ward JO, McConnell MJ, Carlile GW, Pandolfi PP, Licht JD, Freedman LP 2001 The acute promyelocytic leukemiaassociated protein, promyelocytic leukemia zinc finger, regulates 1,25-dihydroxyvitamin D(3)-induced monocytic differentiation of U937 cells through a physical interaction with vitamin D(3) receptor. Blood 98:3290–3300. 224. Puccetti E, Obradovic D, Beissert T, Bianchini A, Washburn B, Chiaradonna F, Boehrer S, Hoelzer D, Ottmann OG, Pelicci PG, Nervi C, Ruthardt M 2002 AML-associated translocation products block vitamin D(3)-induced differentiation by sequestering the vitamin D(3) receptor. Cancer Res 62:7050–7058. 225. Yamamoto Y, Sakamoto M, Fujii G, Kanetaka K, Asaka M, Hirohashi S 2001 Cloning and characterization of a novel gene, DRH1, down-regulated in advanced human hepatocellular carcinoma. Clin Cancer Res 7:297–303. 226. Jakob F, Homann D, Adamski J 1995 Expression and regulation of aromatase and 17 beta-hydroxysteroid dehydrogenase type 4 in human THP 1 leukemia cells. J Steroid Biochem Mol Biol 55:555–563. 227. Hughes SV, Robinson E, Bland R, Lewis HM, Stewart PM, Hewison M 1997 1,25-dihydroxyvitamin D3 regulates estrogen metabolism in cultured keratinocytes. Endocrinology 138:3711–3718. 228. Hughes PJ, Twist LE, Durham J, Choudhry MA, Drayson M, Chandraratna R, Michell RH, Kirk CJ, Brown G 2001 Up-regulation of steroid sulphatase activity in HL60 promyelocytic cells by retinoids and 1alpha,25-dihydroxyvitamin D3. Biochem J 355:361–371. 229. Enjuanes A, Garcia-Giralt N, Supervia A, Nogues X, Mellibovsky L, Carbonell J, Grinberg D, Balcells S, Diez-Perez A 2003 Regulation of CYP19 gene expression in primary human osteoblasts: effects of vitamin D and other treatments. Eur J Endocrinol 148:519–526. 230. Reichman TW, Albanell J, Wang X, Moore MA, Studzinski GP 1997 Down-regulation of telomerase activity in HL60 cells by differentiating agents is accompanied by increased expression of telomerase-associated protein. J Cell Biochem 67:13–23. 231. Rots NY, Liu M, Anderson EC, Freedman LP 1998 A differential screen for ligand-regulated genes: identification of HoxA10 as a target of vitamin D3 induction in myeloid leukemic cells. Mol Cell Biol 18:1911–1918. 232. Iyer VR, Eisen MB, Ross DT, Schuler G, Moore T, Lee JC, Trent JM, Staudt LM, Hudson J Jr, Boguski MS, Lashkari D, Shalon D, Botstein D, Brown PO 1999 The transcriptional program in the response of human fibroblasts to serum. Science 283:83–87. 233. Whitfield ML, Sherlock G, Saldanha AJ, Murray JI, Ball CA, Alexander KE, Matese JC, Perou CM, Hurt MM, Brown PO, Botstein D 2002 Identification of genes periodically expressed in the human cell cycle and their expression in tumors. Mol Biol Cell 13:1977–2000. 234. Lin R, Nagai Y, Sladek R, Bastien Y, Ho J, Petrecca K, Sotiropoulou G, Diamandis EP, Hudson TJ, White JH 2002 Expression profiling in squamous carcinoma cells reveals pleiotropic effects of vitamin D3 analog EB1089 signaling on cell proliferation, differentiation, and immune system regulation. Mol Endocrinol 16:1243–1256. 235. Krishnan AV, Peehl DM, Feldman D 2003 Inhibition of prostate cancer growth by vitamin D: Regulation of target gene expression. J Cell Biochem 88:363–371. 236. Walker PR, Kwast-Welfeld J, Gourdeau H, Leblanc J, Neugebauer W, Sikorska M 1993 Relationship between apoptosis and the cell cycle in lymphocytes: roles of protein kinase C, tyrosine phosphorylation, and AP1. Exp Cell Res 207:142–151.

1593 237. Denmeade SR, Lin XS, Isaacs JT 1996 Role of programmed (apoptotic) cell death during the progression and therapy for prostate cancer. Prostate 28:251–265. 238. Kerr JF, Wyllie AH, Currie AR 1972 Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br J Cancer 26:239–257. 239. Evans SR, Soldatenkov V, Shchepotin EB, Bogrash E, Shchepotin IB 1999 Novel 19-nor-hexafluoride vitamin D3 analog (Ro 25-6760) inhibits human colon cancer in vitro via apoptosis. Int J Oncol 14:979–985. 240. Hockenbery D, Nunez G, Milliman C, Schreiber RD, Korsmeyer SJ 1990 Bcl-2 is an inner mitochondrial membrane protein that blocks programmed cell death. Nature 348:334–336. 241. Korsmeyer SJ, McDonnell TJ, Nunez G, Hockenbery D, Young R 1990 Bcl-2: B cell life, death, and neoplasia. Curr Top Microbiol Immunol 166:203–207. 242. Xu HM, Tepper CG, Jones JB, Fernandez CE, Studzinski GP 1993 1,25-dihydroxyvitamin D3 protects HL60 cells against apoptosis but down-regulates the expression of the bcl-2 gene. Exp Cell Res 209:367–374. 243. James SY, Mackay AG, Colston KW 1995 Vitamin D derivatives in combination with 9-cis retinoic acid promote active cell death in breast cancer cells. J Mol Endocrinol 14:391–394. 244. Danielsson C, Torma H, Vahlquist A, Carlberg C 1999 Positive and negative interaction of 1,25-dihydroxyvitamin D3 and the retinoid CD437 in the induction of human melanoma cell apoptosis. Int J Cancer 81:467–470. 245. Guzey M, Kitada S, Reed JC 2002 Apoptosis induction by 1alpha,25-dihydroxyvitamin D3 in prostate cancer. Mol Cancer Ther 1:667–677. 246. Baudet C, Perret E, Delpech B, Kaghad M, Brachet P, Wion D, Caput D 1998 Differentially expressed genes in C6.9 glioma cells during vitamin D–induced cell death program. Cell Death Differ 5:116–125. 247. Vink-van Wijngaarden T, Pols HA, Buurman CJ, Birkenhager JC, van Leeuwen JP 1994 Combined effects of 1,25-dihydroxyvitamin D3 and tamoxifen on the growth of MCF-7 and ZR-75-1 human breast cancer cells. Breast Cancer Res Treat 29:161–168. 248. Wallington LA, Bunce CM, Durham J, Brown G 1995 Particular combinations of signals, by retinoic acid and 1 alpha, 25 dihydroxyvitamin D3, promote apoptosis of HL60 cells. Leukemia 9:1185–1190. 249. Wu YL, Jiang XR, Lillington DM, Allen PD, Newland AC, Kelsey SM 1998 1,25-dihydroxyvitamin D3 protects human leukemic cells from tumor necrosis factor-induced apoptosis via inactivation of cytosolic phospholipase A2. Cancer Res 58:633–640. 250. Witcher M, Yang X, Pater A, Tang SC 2001 BAG-1 p50 isoform interacts with the vitamin D receptor and its cellular overexpression inhibits the vitamin D pathway. Exp Cell Res 265:167–173. 251. Christakos S, Barletta F, Huening M, Dhawan P, Liu Y, Porta A, Peng X 2003 Vitamin D target proteins: function and regulation. J Cell Biochem 88:238–244. 252. Bikle DD, Pillai S 1993 Vitamin D, calcium, and epidermal differentiation. Endocr Rev 14:3–19. 253. Hosoi J, Abe E, Suda T, Kuroki T 1985 Regulation of melanin synthesis of B16 mouse melanoma cells by 1 alpha, 25-dihydroxyvitamin D3 and retinoic acid. Cancer Res 45:1474–1478. 254. van Leeuwen JP, van Steenbrugge GJ, Veldscholte J, van den Bemd GJ, Oomen MHA, Pols HA, Birkenhager JC 1994 Growth regulation of three sublines of the human prostatic

1594

255.

256.

257.

258.

259.

260.

261.

262. 263.

264.

265.

266. 267.

268.

269.

carcinoma line LNCaP by an interrelated action of 1,25-dihydroxyvitamin D3 and androgens. In: Norman AW, Bouillon R, Thomasset M, (eds) Vitamin D, pluripotent steroid hormone: Structural studies, molecular endocrinology and clinical application. de Gruyter: 510–511. Esquenet M, Swinnen JV, Heyns W, Verhoeven G 1996 Control of LNCaP proliferation and differentiation: actions and interactions of androgens, 1alpha,25-dihydroxycholecalciferol, all-trans retinoic acid, 9-cis retinoic acid, and phenylacetate. Prostate 28:182–194. Zhao XY, Ly LH, Peehl DM, Feldman D 1997 1alpha, 25-dihydroxyvitamin D3 actions in LNCaP human prostate cancer cells are androgen-dependent. Endocrinology 138: 3290–3298. Hedlund TE, Moffatt KA, Uskokovic MR, Miller GJ 1997 Three synthetic vitamin D analogs induce prostate-specific acid phosphatase and prostate-specific antigen while inhibiting the growth of human prostate cancer cells in a vitamin D receptor-dependent fashion. Clin Cancer Res 3:1331–1338. Frappart L, Falette N, Lefebvre MF, Bremond A, Vauzelle JL, Saez S 1989 In vitro study of effects of 1,25 dihydroxyvitamin D3 on the morphology of human breast cancer cell line BT.20. Differentiation 40:63–69. Nagasaki T, Hino M, Inaba M, Nishizawa Y, Morii H, Otani S 1997 Inhibition by 1alpha,25-dihydroxyvitamin D3 of activin A-induced differentiation of murine erythroleukemic F5-5 cells. Arch Biochem Biophys 343:181–187. Lefebvre MF, Guillot C, Crepin M, Saez S 1995 Influence of tumor-derived fibroblasts and 1,25-dihydroxyvitamin D3 on growth of breast cancer cell lines. Breast Cancer Res Treat 33:189–197. Rotello RJ, Lieberman RC, Purchio AF, Gerschenson LE 1991 Coordinated regulation of apoptosis and cell proliferation by transforming growth factor beta 1 in cultured uterine epithelial cells. Proc Natl Acad Sci USA 88:3412–3415. Alexandrow MG, Moses HL 1995 Transforming growth factor beta 1 inhibits mouse keratinocytes late in G1 independent of effects on gene transcription. Cancer Res 55:3928–3932. Laiho M, Decaprio JA, Ludlow JW, Livingston DM, Massague J 1990 Growth inhibition by TGF-beta linked to suppression of retinoblastoma protein phosphorylation. Cell 62:175–185. Simboli-Campbell M, Welsh J 1995 1,25-dihydroxyvitamin D3 1995 Coordinate regulator of active cell death and proliferation in MCF-7 breast cancer cells. In: Tenniswood M, Michna H, (eds) Apoptosis in hormone dependent cancers. SpringerVerlag, pp. 181–200. Kim HJ, Abdelkader N, Katz M, McLane JA 1992 1,25-dihydroxyvitamin D3 enhances antiproliferative effect and transcription of TGF-beta1 on human keratinocytes in culture. J Cell Physiol 151:579–587. Koli K, Keski-Oja J 1993 Vitamin D3 and calcipotriol enhance the secretion of transforming growth factor-beta 1 and -beta 2 in cultured murine keratinocytes. Growth Factors 8:153–163. Finkelman RD, Linkhart TA, Mohan S, Lau KH, Baylink DJ, Bell NH 1991 Vitamin D deficiency causes a selective reduction in deposition of transforming growth factor beta in rat bone: possible mechanism for impaired osteoinduction. Proc Natl Acad Sci USA 88:3657–3660. Yang L, Yang J, Venkateswarlu S, Ko T, Brattain MG 2001 Autocrine TGFbeta signaling mediates vitamin D3 analoginduced growth inhibition in breast cells. J Cell Physiol 188:383–393. Vink-van Wijngaarden T, Pols HA, Buurman CJ, Birkenhager JC, van Leeuwen JP 1996 Inhibition of insulin- and

JOHANNES P. T. M. VAN LEEUWEN AND HUIBERT A. P. POLS

270.

271.

272.

273.

274.

275.

276. 277.

278.

279.

280.

281.

282.

283.

insulin-like growth factor-I-stimulated growth of human breast cancer cells by 1,25-dihydroxyvitamin D3 and the vitamin D3 analog EB1089. Eur J Cancer 32A:842–848. Drivdahl RH, Loop SM, Andress DL, Ostenson RC 1995 IGF-binding proteins in human prostate tumor cells: expression and regulation by 1,25-dihydroxyvitamin D3. Prostate 26:72–79. Velez-Yanguas MC, Kalebic T, Maggi M, Kappel CC, Letterio J, Uskokovic M, Helman LJ 1996 1 alpha, 25-dihydroxy-16-ene-23-yne-26,27-hexafluorocholecalciferol (Ro24-5531) modulation of insulin-like growth factor-binding protein-3 and induction of differentiation and growth arrest in a human osteosarcoma cell line. J Clin Endocrinol Metab 81:93–99. Moriwake T, Tanaka H, Kanzaki S, Higuchi J, Seino Y 1992 1,25-dihydroxyvitamin D3 stimulates the secretion of insulin-like growth factor binding protein 3 (IGFBP-3) by cultured human osteosarcoma cells. Endocrinology 130:1071–1073. Nakao Y, Hilliker S, Baylink DJ, Mohan S 1994 Studies on the regulation of insulin-like growth factor binding protein 3 secretion in human osteosarcoma cells in vitro. J Bone Miner Res 9:865–872. Koga M, Eisman JA, Sutherland RL 1988 Regulation of epidermal growth factor receptor levels by 1,25-dihydroxyvitamin D3 in human breast cancer cells. Cancer Res 48:2734–2739. Falette N, Frappart L, Lefebvre MF, Saez S 1989 Increased epidermal growth factor receptor level in breast cancer cells treated by 1,25-dihydroxyvitamin D3. Mol Cell Endocrinol 63:189–198. Welsh J 1994 Induction of apoptosis in breast cancer cells in response to vitamin D and antiestrogens. Biochem Cell Biol 72:537–545. Byrne IM, Flanagan L, Tenniswood MP, Welsh J 2000 Identification of a hormone-responsive promoter immediately upstream of exon 1c in the human vitamin D receptor gene. Endocrinology 141:2829–2836. Wietzke JA, Welsh J 2003 Phytoestrogen regulation of a vitamin D3 receptor promoter and 1,25-dihydroxyvitamin D3 actions in human breast cancer cells. J Steroid Biochem Mol Biol 84:149–157. Smirnoff P, Liel Y, Gnainsky J, Shany S, Schwartz B 1999 The protective effect of estrogen against chemically-induced murine colon carcinogenesis is associated with decreased CpG island methylation and increased mRNA and protein expression of the colonic vitamin D receptor. Oncol Res 11:255–264. Ahonen MH, Zhuang YH, Aine R, Ylikomi T, Tuohimaa P 2000 Androgen receptor and vitamin D receptor in human ovarian cancer: growth stimulation and inhibition by ligands. Int J Cancer 86:40–46. Zhao XY, Ly LH, Peehl DM, Feldman D 1999 Induction of androgen receptor by 1alpha,25-dihydroxyvitamin D3 and 9-cis retinoic acid in LNCaP human prostate cancer cells. Endocrinology 140:1205–1212. Miyaura C, Abe E, Honma Y, Hozumi M, Nishii Y, Suda T 1983 Cooperative effect of 1 alpha,25-dihydroxyvitamin D3 and dexamethasone in inducing differentiation of mouse myeloid leukemia cells. Arch Biochem Biophys 227: 379–385. Yu WD, McElwain MC, Modzelewski RA, Russell DM, Smith DC, Trump DL, Johnson CS 1998 Enhancement of 1,25-dihydroxyvitamin D3-mediated antitumor activity with dexamethasone. J Natl Cancer Inst 90:134–141.

CHAPTER 89 Vitamin D: Cancer and Differentiation

284. Bernardi RJ, Trump DL, Yu WD, Mcguire TF, Hershberger PA, Johnson CS 2001 Combination of 1alpha,25-dihydroxyvitamin D(3) with dexamethasone enhances cell cycle arrest and apoptosis: role of nuclear receptor cross-talk and Erk/Akt signaling. Clin Cancer Res 7:4164–4173. 285. Ziegler R, Kasperk C 1998 Glucocorticoid-induced osteoporosis: prevention and treatment. Steroids 63:344–348. 286. Koga M, Sutherland RL 1991 Retinoic acid acts synergistically with 1,25-dihydroxyvitamin D3 or antioestrogen to inhibit T-47D human breast cancer cell proliferation. J Steroid Biochem Mol Biol 39:455–460. 287. Moore TB, Sidell N, Chow VJ, Medzoyan RH, Huang JI, Yamashiro JM, Wada RK 1995 Differentiating effects of 1,25-dihydroxycholecalciferol (D3) on LA-N-5 human neuroblastoma cells and its synergy with retinoic acid. J Pediatr Hematol Oncol 17:311–317. 288. Bunce CM, Wallington LA, Harrison P, Williams GR, Brown G 1995 Treatment of HL60 cells with various combinations of retinoids and 1 alpha,25 dihydroxyvitamin D3 results in differentiation towards neutrophils or monocytes or a failure to differentiate and apoptosis. Leukemia 9:410–418. 289. Kliewer SA, Umesono K, Mangelsdorf DJ, Evans RM 1992 Retinoid X receptor interacts with nuclear receptors in retinoic acid, thyroid hormone, and vitamin D3 signalling. Nature 355:446–449. 290. Matsui T, Takahashi R, Mihara K, Nakagawa T, Koizumi T, Nakao Y, Sugiyama T, Fujita T 1985 Cooperative regulation of c-myc expression in differentiation of human promyelocytic leukemia induced by recombinant gammainterferon and 1,25-dihydroxyvitamin D3. Cancer Res 45: 4366–4371. 291. Morel PA, Manolagas SC, Provvedini DM, Wegmann DR, Chiller JM 1986 Interferon gamma-induced IA expression in WEHI-3 cells is enhanced by the presence of 1,25-dihydroxyvitamin D3. J Immunol 136:2181–2186. 292. Kelsey SM, Makin HL, MacEy MG, Newland AC 1990 Gamma interferon augments functional and phenotypic characteristics of vitamin D3-induced monocytoid differentiation in the U937 human leukaemic cell line. Leuk Res 14:1027–1033. 293. Kelsey SM, Coldwell RD, Makin HL, Syndercombe-Court, Newland AC 1994 Neopterin release by myeloid leukemic cells can be synergistically augmented by 1,25-dihydroxyvitamin D3 in combination with gamma interferon or granulocyte-macrophage colony stimulating factor. J Steroid Biochem Mol Biol 48:95–99. 294. Kasukabe T, Okabe-Kado J, Honma Y, Hozumi M 1991 Interleukin-4 inhibits the differentiation of mouse myeloid leukemia M1 cells induced by dexamethasone, D-factor/ leukemia inhibitory factor and interleukin-6, but not by 1 alpha,25-dihydroxyvitamin D3. FEBS Lett 291:181–184. 295. Kasukabe T, Okabe-Kado J, Honma Y, Hozumi M 1992 Interleukin 4 potentiates the antiproliferative effect of 1 alpha, 25-dihydroxyvitamin D3 on mouse monocytic leukemia cells but antagonizes the antiproliferative effects of interferon alpha, beta and interleukin 6. Leuk Res 16:1191–1196. 296. Hassan HT, Tsiriyotis C, Maurer HR, Spandidos DA 1992 Recombinant human interleukin-3 opposes the effects of vitamins A and D on HL-60 human myeloid leukaemia cells. Anticancer Res 12:821–825. 297. Prechel MM, Lozano Y, Wright MA, Ihm J, Young MR 1996 Immune modulation by interleukin-12 in tumor-bearing mice receiving vitamin D3 treatments to block induction of immunosuppressive granulocyte/macrophage progenitor cells. Cancer Immunol Immunother 42:213–220.

1595 298. Koren R, Rocker D, Kotestiano O, Liberman UA, Ravid A 2000 Synergistic anticancer activity of 1,25-dihydroxyvitamin D(3) and immune cytokines: the involvement of reactive oxygen species. J Steroid Biochem Mol Biol 73:105–112. 299. Katakami Y, Nakao Y, Katakami N, Koizumi T, Ogawa R, Yamada H, Takai Y, Fujita T 1988 Cooperative effects of tumor necrosis factor-alpha and 1,25-dihydroxyvitamin D3 on growth inhibition, differentiation, and c-myc reduction in human promyelocytic leukemia cell line HL-60. Biochem Biophys Res Commun 152:1151–1157. 300. Rocker D, Ravid A, Liberman UA, Garach-Jehoshua O, Koren R 1994 1,25-dihydroxyvitamin D3 potentiates the cytotoxic effect of TNF on human breast cancer cells. Mol Cell Endocrinol 106:157–162. 301. Hassan HT, Eliopoulos A, Maurer HR, Spandidos DA 1992 Recombinant human GM-CSF enhances the antiproliferative activity of vitamin D in MCF-7 human breast cancer clonogenic cells. Eur J Cancer 28A:1588–1589. 302. Kelsey SM, Makin HL, Newland AC 1992 Functional significance of induction of differentiation in human myeloid leukaemic blasts by 1,25-dihydroxyvitamin D3 and GM-CSF. Leuk Res 16:427–434. 303. Rashid SF, Moore JS, Walker E, Driver PM, Engel J, Edwards CE, Brown G, Uskokovic MR, Campbell MJ 2001 Synergistic growth inhibition of prostate cancer cells by 1 alpha,25 dihydroxyvitamin D(3) and its 19-nor-hexafluoride analogs in combination with either sodium butyrate or trichostatin A. Oncogene 20:1860–1872. 304. Gaschott T, Werz O, Steinmeyer A, Steinhilber D, Stein J 2001 Butyrate-induced differentiation of Caco-2 cells is mediated by vitamin D receptor. Biochem Biophys Res Commun 288:690–696. 305. Gaschott T, Wachtershauser A, Steinhilber D, Stein J 2001 1,25-dihydroxycholecalciferol enhances butyrate-induced p21(Waf1/Cip1) expression. Biochem Biophys Res Commun 283:80–85. 306. Bizzarri M, Cucina A, Valente MG, Tagliaferri F, Borrelli V, Stipa F, Cavallaro A 2003 Melatonin and vitamin D(3) increase TGF-beta(1) release and induce growth inhibition in breast cancer cell cultures. J Surg Res 110:332–337. 307. Tong WM, Kallay E, Hofer H, Hulla W, Manhardt T, Peterlik M, Cross HS 1998 Growth regulation of human colon cancer cells by epidermal growth factor and 1,25-dihydroxyvitamin D3 is mediated by mutual modulation of receptor expression. Eur J Cancer 34:2119–2125. 308. Cho YL, Christensen C, Saunders DE, Lawrence WD, Deppe G, Malviya VK, Malone JM 1991 Combined effects of 1,25-dihydroxyvitamin D3 and platinum drugs on the growth of MCF-7 cells. Cancer Res 51:2848–2853. 309. Ravid A, Rocker D, Machlenkin A, Rotem C, Hochman A, Kessler-Icekson G, Liberman UA, Koren R 1999 1,25Dihydroxyvitamin D3 enhances the susceptibility of breast cancer cells to doxorubicin-induced oxidative damage. Cancer Res 59:862–867. 310. Sundaram S, Chaudhry M, Reardon D, Gupta M, Gewirtz DA 2000 The vitamin D3 analog EB 1089 enhances the antiproliferative and apoptotic effects of adriamycin in MCF-7 breast tumor cells. Breast Cancer Res Treat 63:1–10. 311. Chaudhry M, Sundaram S, Gennings C, Carter H, Gewirtz DA 2001 The vitamin D3 analog, ILX-23–7553, enhances the response to adriamycin and irradiation in MCF-7 breast tumor cells. Cancer Chemother Pharmacol 47:429–436. 312. Danilenko M, Wang X, Studzinski GP 2001 Carnosic acid and promotion of monocytic differentiation of HL60-G cells initiated by other agents. J Natl Cancer Inst 93:1224–1233.

1596 313. Danilenko M, Wang Q, Wang X, Levy J, Sharoni Y, Studzinski GP 2003 Carnosic acid potentiates the antioxidant and prodifferentiation effects of 1alpha,25-dihydroxyvitamin D3 in leukemia cells but does not promote elevation of basal levels of intracellular calcium. Cancer Res 63:1325–1332. 314. Gewirtz DA, Gupta MS, Sundaram S 2002 Vitamin D3 and vitamin D3 analogs as an adjunct to cancer chemotherapy and radiotherapy. Curr Med Chem Anti Canc Agents 2:683–690. 315. Polar MK, Gennings C, Park M, Gupta MS, Gewirtz DA 2003 Effect of the vitamin D(3) analog ILX 23–7553 on apoptosis and sensitivity to fractionated radiation in breast tumor cells and normal human fibroblasts. Cancer Chemother Pharmacol 51:415–421. 316. Sundaram S, Sea A, Feldman S, Strawbridge R, Hoopes PJ, Demidenko E, Binderup L, Gewirtz DA 2003 The combination of a potent vitamin D(3) analog, EB 1089, with ionizing radiation reduces tumor growth and induces apoptosis of MCF-7 breast tumor xenografts in nude mice. Clin Cancer Res 9:2350–2356. 317. Dokoh S, Donaldson CA, Haussler MR 1984 Influence of 1,25-dihydroxyvitamin D3 on cultured osteogenic sarcoma cells: correlation with the 1,25-dihydroxyvitamin D3 receptor. Cancer Res 44:2103–2109. 318. Miller GJ, Stapleton GE, Hedlund TE, Moffat KA 1995 Vitamin D receptor expression, 24-hydroxylase activity, and inhibition of growth by 1alpha,25-dihydroxyvitamin D3 in seven human prostatic carcinoma cell lines. Clin Cancer Res 1:997–1003. 319. Hedlund TE, Moffatt KA, Miller GJ 1996 Stable expression of the nuclear vitamin D receptor in the human prostatic carcinoma cell line JCA-1: evidence that the antiproliferative effects of 1 alpha, 25-dihydroxyvitamin D3 are mediated exclusively through the genomic signaling pathway. Endocrinology 137:1554–1561. 320. Jensen SS, Madsen MW, Lukas J, Bartek J, Binderup L 2002 Sensitivity to growth suppression by 1alpha,25-dihydroxyvitamin D(3) among MCF-7 clones correlates with Vitamin D receptor protein induction. J Steroid Biochem Mol Biol 81:123–133. 321. Tokumoto M, Tsuruya K, Fukuda K, Kanai H, Kuroki S, Hirakata H, Iida M 2003 Parathyroid cell growth in patients with advanced secondary hyperparathyroidism: vitamin D receptor and cyclin-dependent kinase inhibitors, p21 and p27. Nephrol Dial Transplant 18(Suppl. 3):III9–III12. 322. Zinser GM, McEleney K, Welsh J 2003 Characterization of mammary tumor cell lines from wild type and vitamin D(3) receptor knockout mice. Mol Cell Endocrinol 200:67–80. 323. Xu HM, Kolla SS, Goldenberg NA, Studzinski GP 1992 Resistance to 1,25-dihydroxyvitamin D3 of a deoxycytidine kinase-deficient variant of human leukemia HL60 cells. Exp Cell Res 203:244–250. 324. Narvaez CJ, VanWeelden K, Byrne I, Welsh J 1996 Characterization of a vitamin D3-resistant MCF-7 cell line. Endocrinology 137:400–409. 325. Iwata K, Kouttab N, Ogata H, Morgan JW, Maizel AL, Lasky SR 1996 Differential regulation of vitamin D receptors in clonal populations of a chronic myelogenous leukemia cell line. Exp Cell Res 225:143–150. 326. Narvaez CJ, Byrne BM, Romu S, Valrance M, Welsh J 2003 Induction of apoptosis by 1,25-dihydroxyvitamin D(3) in MCF-7 Vitamin D(3)–resistant variant can be sensitized by TPA. J Steroid Biochem Mol Biol 84:199–209. 327. Hansen CM, Rohde L, Madsen MW, Hansen D, Colston KW, Pirianov G, Holm PK, Binderup L 2001 MCF-7/VD(R): a new vitamin D resistant cell line. J Cell Biochem 82:422–436.

JOHANNES P. T. M. VAN LEEUWEN AND HUIBERT A. P. POLS

328. Wajchmann HJ, Rathod B, Song S, Xu H, Wang X, Uskokovic MR, Studzinski GP 1996 Loss of deoxcytidine kinase expression and tetraploidization of HL60 cells following long-term culture in 1,25-dihydroxyvitamin D3. Exp Cell Res 224:312–322. 329. Solomon C, Sebag M, White JH, Rhim J, Kremer R 1998 Disruption of vitamin D receptor-retinoid X receptor heterodimer formation following ras transformation of human keratinocytes. J Biol Chem 273:17573–17578. 330. Prufer K, Schroder C, Hegyi K, Barsony J 2002 Degradation of RXRs influences sensitivity of rat osteosarcoma cells to the antiproliferative effects of calcitriol. Mol Endocrinol 16:961–976. 331. Zhuang SH, Burnstein KL 1998 Antiproliferative effect of 1alpha,25-dihydroxyvitamin D3 in human prostate cancer cell line LNCaP involves reduction of cyclin-dependent kinase 2 activity and persistent G1 accumulation. Endocrinology 139:1197–1207. 332. Chen H, Hu B, Allegretto EA, Adams JS 2000 The vitamin D response element-binding protein. A novel dominant-negative regulator of vitamin D–directed transactivation. J Biol Chem 275:35557–35564. 333. Chen H, Hewison M, Hu B, Adams JS 2003 Heterogeneous nuclear ribonucleoprotein (hnRNP) binding to hormone response elements: a cause of vitamin D resistance. Proc Natl Acad Sci USA 100:6109–6114. 334. Chen TL, Hauschka PV, Feldman D 1986 Dexamethasone increases 1,25-dihydroxyvitamin D3 receptor levels and augments bioresponses in rat osteoblast-like cells. Endocrinology 118:1119–1126. 335. Ly LH, Zhao XY, Holloway L, Feldman D 1999 Liarozole acts synergistically with 1alpha,25-dihydroxyvitamin D3 to inhibit growth of DU 145 human prostate cancer cells by blocking 24-hydroxylase activity. Endocrinology 140:2071–2076. 336. Campbell MJ, Drayson MT, Durham J, Wallington L, Siu-Caldera ML, Reddy GS, Brown G 1999 Metabolism of 1alpha,25(OH)2D3 and its 20-epi analog integrates clonal expansion, maturation, and apoptosis during HL-60 cell differentiation. Mol Cell Endocrinol 149:169–183. 337. Quack M, Mork HC, Binderup E, Kissmeyer AM, Carlberg C 1998 Metabolism of the vitamin D3 analog EB1089 alters receptor complex formation and reduces promoter selectivity. Br J Pharmacol 125:607–614. 338. Dilworth FJ, Williams GR, Kissmeyer AM, Nielsen JL, Binderup E, Calverley MJ, Makin HL, Jones G 1997 The vitamin D analog, KH1060, is rapidly degraded both in vivo and in vitro via several pathways: principal metabolites generated retain significant biological activity. Endocrinology 138:5485–5496. 339. van den Bemd GC, Dilworth FJ, Makin HL, Prahl JM, DeLuca HF, Jones G, Pols HA, van Leeuwen JP 2000 Contribution of several metabolites of the vitamin D analog 20-epi-22-oxa-24a,26a,27a-tri-homo-1,25-(OH)(2) vitamin D(3) (KH 1060) to the overall biological activity of KH1060 by a shared mechanism of action. Biochem Pharmacol 59:621–627. 340. Taniyama T, Wanibuchi H, Salim EI, Yano Y, Otani S, Nishizawa Y, Morii H, Fukushima S 2000 Chemopreventive effect of 24R,25-dihydroxyvitamin D(3) in N, N’-dimethylhydrazine-induced rat colon carcinogenesis. Carcinogenesis 21:173–178. 341. Farhan H, Wahala K, Cross HS 2003 Genistein inhibits Vitamin D hydroxylases CYP24 and CYP27B1 expression in prostate cells. J Steroid Biochem Mol Biol 84:423–429. 342. Albertson DG, Ylstra B, Segraves R, Collins C, Dairkee SH, Kowbel D, Kuo WL, Gray JW, Pinkel D 2000 Quantitative

1597

CHAPTER 89 Vitamin D: Cancer and Differentiation

343.

344.

345.

346.

347.

348.

349.

350. 351. 352.

353.

354.

355. 356.

357.

mapping of amplicon structure by array CGH identifies CYP24 as a candidate oncogene. Nat Genet 25:144–146. Cross HS, Bareis P, Hofer H, Bischof MG, Bajna E, Kriwanek S, Bonner E, Peterlik M 2001 25-Hydroxyvitamin D(3)-1alpha-hydroxylase and vitamin D receptor gene expression in human colonic mucosa is elevated during early cancerogenesis. Steroids 66:287–292. Bareis P, Kallay E, Bischof MG, Bises G, Hofer H, Potzi C, Manhardt T, Bland R, Cross HS 2002 Clonal differences in expression of 25-hydroxyvitamin D(3)-1alpha-hydroxylase, of 25-hydroxyvitamin D(3)-24-hydroxylase, and of the vitamin D receptor in human colon carcinoma cells: effects of epidermal growth factor and 1alpha,25-dihydroxyvitamin D(3). Exp Cell Res 276:320–327. Holt PR, Arber N, Halmos B, Forde K, Kissileff H, McGlynn KA, Moss SF, Kurihara N, Fan K, Yang K, Lipkin M 2002 Colonic epithelial cell proliferation decreases with increasing levels of serum 25-hydroxyvitamin D. Cancer Epidemiol Biomarkers Prev 11:113–119. Chen TC, Wang L, Whitlatch LW, Flanagan JN, Holick MF 2003 Prostatic 25-hydroxyvitamin D-1alpha-hydroxylase and its implication in prostate cancer. J Cell Biochem 88: 315–322. Ogose A, Kawashima H, Morita O, Hotta T, Umezu H, Endo N 2003 Increase in serum 1,25-dihydroxyvitamin D and hypercalcemia in a patient with inflammatory myofibroblastic tumor. J Clin Pathol 56:310–312. Baier R, Grauer A, Lazaretti-Castro M, Ziegler R, Raue F 1994 Differential effects of 1,25-dihydroxyvitamin D3 on cell proliferation and calcitonin gene expression. Endocrinology 135:2006–2011. Ohta M, Okabe T, Ozawa K, Urabe A, Takaku F 1985 1 alpha,25-dihydroxyvitamin D3 (calcitriol) stimulates proliferation of human circulating monocytes in vitro. FEBS Lett 185:9–13. Mitsuhashi T, Morris RC Jr, Ives HE 1991 1,25-dihydroxyvitamin D3 modulates growth of vascular smooth muscle cells. J Clin Invest 87:1889–1895. Edelson JD, Chan S, Jassal D, Post M, Tanswell AK 1994 Vitamin D stimulates DNA synthesis in alveolar type-II cells. Biochim Biophys Acta 1221:159–166. Lutzow-Holm C, De Angelis P, Grosvik H, Clausen OP 1993 1,25-dihydroxyvitamin D3 and the vitamin D analog KH1060 induce hyperproliferation in normal mouse epidermis. A BrdUrd/DNA flow cytometric study. Exp Dermatol 2:113–120. Gniadecki R 1994 A vitamin D analog KH 1060 activates the protein kinase C-c-fos signaling pathway to stimulate epidermal proliferation in murine skin. J Endocrinol 143: 521–525. Gniadecki R, Serup J 1995 Stimulation of epidermal proliferation in mice with 1 alpha, 25-dihydroxyvitamin D3 and receptor-active 20-EPI analogs of 1 alpha, 25-dihydroxyvitamin D3. Biochem Pharmacol 49:621–624. Gniadecki R, Gniadecka M, Serup J 1995 The effects of KH 1060, a potent 20-epi analog of the vitamin D3 hormone, on hairless mouse skin in vivo. Br J Dermatol 132:841–852. Hosoi J, Abe E, Suda T, Colburn NH, Kuroki T 1986 Induction of anchorage-independent growth of JB6 mouse epidermal cells by 1 alpha,25-dihydroxyvitamin D3. Cancer Res 46:5582–5586. Kitano Y, Ikeda N, Okano M 1991 Suppression of proliferation of human epidermal keratinocytes by 1,25-dihydroxyvitamin D3. Analysis of its effect on psoriatic lesion and of

358.

359. 360. 361.

362.

363.

364.

365.

366.

367.

368.

369. 370.

371.

372.

its mechanism using human keratinocytes in culture. Eur J Clin Invest 21:53–58. Hosomi J, Hosoi J, Abe E, Suda T, Kuroki T 1983 Regulation of terminal differentiation of cultured mouse epidermal cells by 1 alpha,25-dihydroxyvitamin D3. Endocrinology 113:1950–1957. Kragballe K 1992 Vitamin D3 and skin diseases. Arch Dermatol Res 284 (Suppl. 1):S30–S36. Kuroki T, Sasaki K, Chida K, Abe E, Suda T 1983 1 alpha, 25-dihydroxyvitamin D3 markedly enhances chemicallyinduced transformation in BALB 3T3 cells. Gann 74:611–614. Sasaki K, Chida K, Hashiba H, Kamata N, Abe E, Suda T, Kuroki T 1986 Enhancement by 1 alpha,25-dihydroxyvitamin D3 of chemically induced transformation of BALB 3T3 cells without induction of ornithine decarboxylase or activation of protein kinase C1. Cancer Res 46:604–610. Jones CA, Callaham MF, Huberman E 1984 Enhancement of chemical carcinogen-induced cell transformation in hamster embryo cells by 1 alpha,25-dihydroxycholecalciferol, the biologically active metabolite of vitamin D3. Carcinogenesis 5:1155–1159. Wood AW, Chang RL, Huang MT, Baggiolini E, Partridge JJ, Uskokovic M, Conney AH 1985 Stimulatory effect of 1 alpha, 25-dihydroxyvitamin D3 on the formation of skin tumors in mice treated chronically with 7,12-dimethylbenz[a]anthracene. Biochem Biophys Res Commun 130: 924–931. Chang PL, Prince CW 1993 1 alpha,25-dihydroxyvitamin D3 enhances 12-O-tetradecanoylphorbol-13-acetate-induced tumorigenic transformation and osteopontin expression in mouse JB6 epidermal cells. Cancer Res 53:2217–2220. Chang PL, Lee TF, Garretson K, Prince CW 1997 Calcitriol enhancement of TPA-induced tumorigenic transformation is mediated through vitamin D receptor-dependent and -independent pathways. Clin Exp Metastasis 15:580–592. Yamaoka K, Marion SL, Gallegos A, Haussler MR 1986 1,25-dihydroxyvitamin D3 enhances the growth of tumors in athymic mice inoculated with receptor rich osteosarcoma cells. Biochem Biophys Res Commun 139:1292–1298. Franceschi RT, James WM, Zerlauth G 1985 1 alpha, 25-dihydroxyvitamin D3 specific regulation of growth, morphology, and fibronectin in a human osteosarcoma cell line. J Cell Physiol 123:401–409. Gronowicz G, Egan JJ, Rodan GA 1986 The effect of 1,25dihydroxyvitamin D3 on the cytoskeleton of rat calvaria and rat osteosarcoma (ROS 17/2.8) osteoblastic cells. J Bone Miner Res 1:441–455. Hodge BO, Kream BE 1988 Variable effects of dexamethasone on protein synthesis in clonal rat osteosarcoma cells. Endocrinology 122:2127–2133. Abbadia Z, Amiral J, Trzeciak MC, Delmas PD, Clezardin P 1993 The growth-supportive effect of thrombospondin (TSP1) and the expression of TSP1 by human MG-63 osteoblastic cells are both inhibited by dexamethasone. FEBS Lett 335:161–166. Huerta S, Irwin RW, Heber D, Go VL, Moatamed F, Huerta S, Ou C, Harris DM 2003 Intestinal polyp formation in the Apcmin mouse: effects of levels of dietary calcium and altered vitamin D homeostasis. Dig Dis Sci 48:870–876. Sahpazidou D, Stravoravdi P, Toliou T, Geromichalos G, Zafiriou G, Natsis K, Gigis P 2003 Significant experimental decrease of the hepatocellular carcinoma incidence in C3H/Sy mice after long-term administration of EB1089, a vitamin D analog. Oncol Res 13:261–268.

CHAPTER 90

Vitamin D, Sunlight, and the Natural History of Prostate Cancer GARY G. SCHWARTZ TAI C. CHEN

I. II. III. IV.

Comprehensive Cancer Center, Wake Forest University, Winston-Salem, NC Vitamin D, Skin, and Bone Research Laboratory, Boston University Medical Center, Boston, MA

Introduction and Background Prostate Cancer and the Vitamin D Hypothesis Observational Studies Experimental Studies of the Vitamin D Hypothesis

I. INTRODUCTION AND BACKGROUND In 1990, in a paper titled, “Is vitamin D deficiency a risk factor for prostate cancer (Hypothesis)?”, Schwartz and Hulka proposed that clinical prostate cancer may be caused by vitamin D deficiency [1]. Since that time, the status of “the vitamin D hypothesis” has gone from that of the proverbial “dark horse” to that of a frontrunner in the race to understand—and to alter—the natural history of this common cancer [2]. This chapter reviews our current understanding of the role of vitamin D in the epidemiology of prostate cancer. We begin with an overview of vitamin D synthesis and of the descriptive epidemiology of prostate cancer.

A. Vitamin D Synthesis The synthesis of 1,25(OH)2D begins with the production of vitamin D3 (cholecalciferol) after 7-dehydrocholesterol present in the skin is exposed to UV-B radiation (wavelength 290–315 nm). Alternately, vitamin D3 or vitamin D2 (ergocalciferol, a sterol derived from plants) can be obtained from the diet. To become biologically active, vitamin D must undergo two hydroxylations. The first hydroxylation occurs in the liver at the 25th carbon position, forming 25-hydroxyvitamin D (25-OHD), the prohormone and major circulating form of vitamin D. The second hydroxylation occurs at the 1α position, forming 1,25(OH)2D, the hormonal form of vitamin D [3] (see Holick, Chapter 3 for a review of the photobiology of vitamin D). Serum levels of 25-OHD are the best indicator of an individual’s overall vitamin D status, whereas serum levels of 1,25(OH)2D are VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. 1,25(OH)2D Is an Autocrine Hormone in the Prostate VI. Vitamin D Hypothesis: Conclusions References

useful in evaluating disorders in calcium and bone metabolism [4]. Classically, the hydroxylation of 25-OHD at the 1α position was presumed to occur exclusively or predominantly in the kidney and the function of 1,25(OH)2D was thought to lie in the control of mineral metabolism. However, it is now clear that local synthesis of 1,25(OH)2D occurs in an autocrine or paracrine fashion in many nonrenal tissues [5], including keratinocytes, colon, and prostate cells [6], where 1,25(OH)2D controls key processes involving cell differentiation and proliferation [7] (see [8] and Hewison and Adams, Chapter 79 for reviews). The discovery of this expanded role for vitamin D in the prostate has important implications for prostate cancer prevention (see Section V). Sunlight exposure of the skin is by far the most important source of vitamin D [9]. In the U.S., small quantities of vitamin D are added to milk and to some other foods (e.g., margarine, breakfast cereals) principally to prevent rickets. However, vitamin D is not added to milk in many European countries and the quantity of vitamin D in Western diets generally is negligible. It is therefore misleading to consider cholecalciferol and/or ergocalciferol “vitamins,” i.e., essential dietary nutrients. They are more accurately conceived of as hormonal precursors and function as vitamins only in the absence of sunlight [10].

B. The Descriptive Epidemiology of Prostate Cancer: Effects of Age, Race, and Place Prostate cancer is the most common incident (nonskin) cancer among American men and, after lung cancer, Copyright © 2005, Elsevier, Inc. All rights reserved.

1600 the most fatal, accounting for approximately 221,000 new cases and 29,000 deaths in 2003 [11]. Clinical prostate cancer is strongly age-dependent: mortality rates increase logarithmically with age and are approximately 50% higher among African-Americans than among Caucasians. Other than age and race, the most conspicuous feature of prostate cancer mortality is a striking variation by place: age-adjusted mortality rates vary over twentyfold worldwide, and are highest among African-Americans and northern Europeans [12] (see Fig. 1). The similarity in mortality rates between African-Americans and northern Europeans is an important clue, as it suggests that these populations may share some common factor that underlies their similar mortality experiences. Indeed, whatever theory for the etiology of prostate cancer one proposes, it must answer the question, How are African-Americans and northern Europeans alike?

GARY G. SCHWARTZ AND TAI C. CHEN

A unique feature of the epidemiology of prostate cancer is the high prevalence of “incidental” (also known as “subclinical” or “autopsy”) cancer. Autopsies performed on men who have died from causes other than prostate cancer reveal that approximately 27% of men in their 40s and 34% of men in their 50s have histological prostate cancer [13]. The prevalence of these subclinical cancers reaches 60% in men over the age of 80 and continues to increase with age [14]. Histologically, these lesions are indistinguishable from prostate cancers that are potentially life-threatening and are considered to represent cancers at an earlier stage in their natural history. In contrast to mortality rates, the prevalence of incidental prostate tumors is similar among older men worldwide regardless of their racial or geographic origins [15,16]. The discrepancy between the occurrence of clinical and subclinical prostate cancer suggests that clinical cancers result from factors that govern the growth of the subclinical cancers. Thus, any etiologic hypothesis for prostate cancer must also explain why subclinical prostate cancer is ubiquitous, whereas clinical prostate cancer is not. A related observation concerns the striking effects of migration and of Westernization. In the 1960s, mortality rates for prostate cancer among Japanese living in Japan were 1/15 those of Caucasians in the U.S. These rates quadrupled in first and second generations of Japanese migrants to the U.S. [17]. This suggests either that something in the U.S. promotes the growth of latent cancers among Japanese migrants, or that some factor, more prevalent in Japan than in the U.S., restrains their growth and that upon migration, this restraint is lost. Similarly, age-adjusted mortality rates for prostate cancer among the indigenous Japanese doubled in the time period from 1970 to 1990 [18]. This rapid increase strongly implicates some factor associated with Westernization in the etiology of fatal prostate cancer.

II. PROSTATE CANCER AND THE VITAMIN D HYPOTHESIS A. Vitamin D and Prostate Cancer Risk

FIGURE 1 Worldwide age-adjusted mortality rates for prostate cancer, drawn from data in Kurihara et al. Mortality rates show a twentyfold variation and are highest among African American men and among northern European populations.

In 1990, we hypothesized that vitamin D maintains the differentiated phenotype of prostate cells and that vitamin D deficiency permits subclinical prostate cancer to progress to clinical disease. Classically, vitamin D “deficiency” had been defined in terms of the bony diseases, rickets (in children) and osteomalacia (in adults). Conversely, vitamin D sufficiency was equated with the absence of bony disease. However, we suggested

1601

CHAPTER 90 Vitamin D, Sunlight, and the Natural History of Prostate Cancer

that the levels of vitamin D adequate to maintain a healthy skeleton may be inadequate to maintain a healthy prostate. The vitamin D hypothesis proposed that the major risk factors of the descriptive epidemiology of prostate cancer: increasing age, Black race, and residence at northern latitudes, are related in that each is associated with vitamin D deficiency (see Chapters 47 and 66). Consider the increased risk with age. This is understandable because the prevalence of vitamin D deficiency increases with age. The elderly are commonly vitamin D–deficient for several reasons [19] (see Chapter 50). First, older persons, especially those with limited mobility, often get less solar exposure than younger persons. Second, the thinner epidermis of older individuals contains less 7-dehydrocholesterol than does that of younger individuals and, consequently, less vitamin D3 is formed following solar exposure [20]. This results in lower serum levels of 25-OHD3 in older individuals. It is now clear that vitamin D deficiency is common among the elderly worldwide, especially for housebound persons and geriatric populations [21]. For example, a recent study of centenarians in Northern Italy revealed that 99 of the 104 persons examined (95%) had undetectable levels of 25-OHD (< 5 nmol/liter) [22]. The higher risk for prostate cancer among Blacks is understandable because black (or otherwise densely pigmented) skin blocks ultraviolet rays, making it more difficult for dark-skinned individuals to synthesize vitamin D from ultraviolet light [23]. Consequently, serum levels of 25-OHD in Blacks are often 1/2 or less than those of Caucasians living at similar latitudes [24]. It is in this sense that African-Americans “resemble” European populations living at northern latitudes; both groups typically have low serum levels of 25-OHD. Recent population-based data in the U.S. indicate that the prevalence of vitamin D deficiency is ten times higher among African-Americans than among Caucasians [25]. The low risk among Japanese living in Japan is comprehensible as well. Data on serum 25-OHD levels among the Japanese are among the highest ever recorded [26,27]. These high vitamin D levels reflect the traditional diet which is high in oily fish [28]. For example, tuna and skipjack, two of the most commonly consumed fish in Japan, contain approximately 16,000 and 57,000 IU vitamin D per gram of their oil [29]. (For comparison, milk in the U.S. is supplemented with 400 IU vitamin D/quart.) The protective effect of living in Japan would be expected to wane (and clinical prostate cancer rates later to increase) as Japanese migrate and/or adopt a more Western diet.

TABLE I Risk Factors for Prostate Cancer and Their Interpretation by the Vitamin D Hypothesis, Adapted from Schwartz and Hulka, 1990 Risk factor

Explanation by deficiency hypothesis

Age

The prevalence of vitamin D deficiency increases with age.

Race Blacks Asians

Geography

Melanin inhibits synthesis of vitamin D. Traditional diet high in vitamin D (fish oil) protects against clinical cancer. Protection wanes as migrants adopt a western diet. U.S. mortality rates from prostate cancer are inversely correlated with ultraviolet radiation.

This situation is analogous to the recent epidemic of rickets that occurred when individuals from the sunrich Indian subcontinent migrated to locations in sunpoor northern Europe [30,31]. In summary, the vitamin D hypothesis stemmed from an analogy between the epidemiology of prostate cancer and that of rickets, the “classic” disease of vitamin D deficiency [32]. We reasoned that if vitamin D deficiency could cause one (once-common) disease, it could perhaps cause another, albeit one that presented later in life. Heaney recently labeled such diseases “long-latency deficiency diseases” [33]. In 1990, this hypothesis met with considerable skepticism, as the mechanisms underlying the effects of vitamin D on prostate cells were completely unknown. Although we had predicted the existence of receptors for 1,25(OH)2D (VDR) in human prostate cells (encouraged, in part, by the evidence of VDR in the prostate of the mouse published in 1989 [34]), confirmation of this prediction did not occur until two years later (see Section IV) [35]. The vitamin D hypothesis made many other predictions; in particular, it predicted that, since most vitamin D comes from exposure to sunlight, mortality from prostate cancer should increase as the quantity of available sunlight decreases.

B. Ultraviolet Radiation and Prostate Cancer: Descriptive Studies In 1992, Hanchette and Schwartz tested this prediction cartographically using data on age-adjusted mortality rates for Caucasian men at the level of the county (Fig. 2) and corresponding data on ultraviolet radiation in the 3,073 counties of the contiguous U.S. We used trend surface analysis, a form of geographic analysis that is essentially linear regression over space [36].

1602

GARY G. SCHWARTZ AND TAI C. CHEN

Rate per 100,000 15.0–16.9 17.0–18.9 19.0–20.9 21.0–22.9 23.0–24.9

FIGURE 2 Prostate cancer mortality by county among white men, 1970–1979, in the contiguous U.S. Figure from Cancer, 70, No. 12, 1992, p. 2864. Copyright 1992, American Cancer Society. Reprinted by permission of Wiley-Liss, Inc. A subsidiary of John Wiley & Sons, Inc.

Our analyses demonstrated that ultraviolet radiation and prostate cancer mortality among Caucasian men are significantly inversely correlated (P < 0.0001) and exhibit opposite geographic trend surfaces [37] (Figs. 3 and 4). The latitudinal gradient in prostate cancer mortality was not apparent in choropleth maps (i.e., the type of maps used most commonly) of prostate cancer mortality (see Fig. 2). These findings subsequently have been replicated using different data sets and analytic techniques for prostate cancer mortality rates in the U.S. [38,39] and Italy [40]. A north-south gradient for prostate cancer also has been reported across Europe [41]. The inverse correlations between the availability of ultraviolet radiation and prostate cancer mortality rates added support for the hypothesis that vitamin D deficiency increases the risk for clinical prostate cancer. However, because these data are based on groups, not individuals, we could not validly conclude that individuals with low exposure to sunlight experience lower rates of prostate cancer. (To do so risks committing the “ecologic fallacy,” an error in reasoning in which relationships observed at the level of the group are mistakenly applied to individuals [42]). Testing the hypothesis that lower exposure to sunlight increases the risk of prostate cancer in individuals requires epidemiologic studies on individuals (see Section III).

III. OBSERVATIONAL STUDIES A. Seroepidemiological Studies Numerous seroepidemiologic studies have attempted to “shed light” on the vitamin D hypothesis. In 1993, Corder et al. analyzed data on stored sera from members of the Kaiser Permanente Plan in northern California. Serum levels of 25-OHD and 1,25(OH)2D were measured for 181 men who subsequently were diagnosed with prostate cancer. Although serum levels of 25-OHD were similar among cases and controls, serum levels of 1,25(OH)2D were slightly but significantly lower among cases, with a mean difference of 1.8 pg/ml. The effect was greatest in men over the age of 57 and in men with low serum levels of 25-OHD. Low 1,25(OH)2D levels were associated with palpable and anaplastic tumors, but not with well-differentiated tumors or tumors discovered incidentally at surgery for benign prostatic hyperplasia (BPH). Subsequently, six studies have addressed the subject of prostate cancer risk in association with circulating vitamin D metabolites. A small case-control study by Braun et al. (1995) based on 61 cases failed to find any difference between cases and controls [43]. Gann et al. (1996) studied 232 cases of prostate cancer and 414 agematched controls from participants in the Physicians’ Health Study. Median levels of 25-OHD, 1,25(OH)2D

CHAPTER 90 Vitamin D, Sunlight, and the Natural History of Prostate Cancer

Predicted index 10.0–18.0 18.0–26.0 26.0–34.0 34.0–42.0

FIGURE 3

Linear trend surface map of ultraviolet radiation in 3,073 counties of the contiguous U.S. Figure redrawn from Cancer, 70, No. 12, 1992, p. 2865. Copyright 1992, American Cancer Society. Reprinted by permission of Wiley-Liss, Inc. A subsidiary of John Wiley & Sons, Inc.

and vitamin D–binding protein were indistinguishable between cases and controls. A nested case-control study conducted by Nomura et al. in Hawaii (1998) compared 136 cases to 136 controls and reported that the risk of prostate cancer was reduced, although not significantly, among men with high levels of both 25-OHD and 1,25(OH)2D [44]. Because this study was conducted in Hawaii, very few men had low serum levels of 25-OHD. Conversely, three larger studies have yielded results in support of the vitamin D hypothesis. Ma et al. (1998) examined the associations between serum vitamin D metabolites and polymorphisms in the VDR in relation to prostate cancer risk in the Physician’s Health Study [45]. They observed no significant associations overall among

Predicted rate 19.80–20.10 20.10–20.40 20.40–20.70 20.70–21.00

FIGURE 4

Linear trend surface map of age-adjusted prostate cancer among white men, 1970–1979 in 3,073 counties of the contiguous U.S. Figure redrawn from Cancer, 70, No. 12, 1992, p. 2865. Copyright 1992, American Cancer Society. Reprinted by permission of Wiley-Liss, Inc. A subsidiary of John Wiley & Sons, Inc.

1603

372 incident cases and 591 controls for these VDR polymorphisms and prostate cancer. However, in an analysis restricted to men with plasma 25-OHD levels below the median (i.e., among men with relative vitamin D insufficiency), the relative risk for men with the BB vs. the bb genotype was 0.43 (95% CI = 0.19–0.98). This risk reduction was more pronounced among older men (RR = 0.18, 95% CI = 0.05–0.68). Ahonen and colleagues (2000) conducted a nested case-control study of 19,000 middle-aged Finnish men who attended the first screening of the Helsinki Heart Study and were free of clinical prostate cancer at baseline [46]. After 13 years of follow-up, 149 cases of prostate cancer were identified. Cases were matched to probability-density sampled controls (four per case) and were matched for age, residence, and time of sample retrieval. Prostate cancer risk was inversely related to serum 25-OHD levels at baseline. (Because 25-OHD is the best marker of vitamin D status, levels of 1,25(OH)2D were not measured.) Men with 25-OHD levels below the median (40 nmol/L) had an OR of 1.7 (95% C.I. 1.2–2.6) compared to men above the median, and their mean age at diagnosis of prostate cancer was significantly younger. Recently, Tuohimaa reported the results of a large, nested case-control study among Nordic men (Norway, Finland, and Sweden). They studied serum 25-OHD levels in 622 prostate cancer cases and 1,451 matched controls. They reported a “U–shaped” risk of prostate cancer, with both low (≤ 19 nmol/L) and high (≥ 80 nmol/L) 25-OHD serum levels associated with higher prostate cancer risk. In the Finnish men, an increased risk was seen for the lowest compared to the highest quintile (OR = 1.9, 95% CI = 0.97–3.7), a result concordant with that previously reported by Ahonen et al. for Finns in the Helsinki Heart Study. However, in addition to significantly increased risk at low levels of 25-OHD, the risk of prostate cancer also appeared to increase for men at the highest levels of summertime 25-OHD (compared to normal levels of 25-OHD), especially among Norwegian and Swedish cases. There are at least two possible explanations for the increased risk seen at high levels of 25(OH)D. As noted by Tuohimaa, high dietary vitamin D may be associated with other risk factors. For example, some dietary sources of vitamin D (e.g., supplements of fish liver oil, commonly consumed in Norway) are also very rich in vitamin A. Vitamin A and its metabolites are known to antagonize the prodifferentiating and antiproliferative actions of 1,25(OH)2D [47,48]. An alternative explanation was proposed by Vieth, who noted that winter at high latitudes produces a gradual decline in serum levels of 25-OHD, and that during this decline the autocrine synthesis of 1,25(OH)2D by the prostate

1604 cannot be maintained. This decline would be greatest in men whose summer levels of 25-OHD were highest [49]. Thus, high levels of 25-OHD per se would not increase risk for prostate cancer, but cycles of vitamin D inadequacy would. In summary, the data from seroepidemiologic studies of 25-OHD and 1,25(OH)2D are conflicting. However, several conclusions can be drawn. First, the findings of Corder et al., of lower levels of 1,25(OH)2D prior to the diagnosis of cancer, have not been confirmed. It is possible that these findings reflect confounding by preexisting cancer. The average interval between blood draw and diagnosis in the study by Corder et al. was 11 years, although in some instances was less than 2 years. Because prostate cancer has a very long natural history (20 years or longer), serum levels of 1,25(OH)2D may have been depressed by pre-existing prostate cancer. Serum levels of 1,25(OH)2D are known to decrease with increasing stage of disease in patients with cancer of the breast [50] and of the colon [51]. Second, although the results from three case-control studies were null, results from the largest prospective studies support the hypothesis that lower serum levels of 25-OHD increase prostate cancer risk. It is noteworthy that 2 of these studies occurred in Scandinavia, locations where vitamin D insufficiency may be common, especially during winter. The finding by Tuohimaa that elevated levels of 25-OHD also were associated with increased risk for prostate cancer among Norwegians and Swedes is an isolated finding and, as these authors note, is not consistent with the larger epidemiologic and biological literature. In 1998, Giovannucci hypothesized that dietary calcium should increase the risk of developing prostate cancer because calcium acts to reduce serum levels of 1,25(OH)2D [52]. Whether calcium increases prostate cancer risk is unclear: both positive [53,54] and negative studies [55,56] have been reported (see Chapter 91). If calcium does increase risk, the mechanism is unlikely to involve a decrease in serum levels of 1,25(OH)2D since, with the exception of Corder et al., seroepidemiologic studies do not support an association between decreased serum levels of 1,25(OH)2D and prostate cancer [57]. Alternately, calcium may decrease circulating serum levels of 25-OHD. For example, Bell et al. showed that the addition of a 2,000 mg/day calcium for 4 days blunted the increase in serum 25-OHD caused by a dose of 100,000 IU/day by 50%, from a 24 ng/ml rise in 25-OHD to only 12 ng/ml [58]. Finally, we note that the seroepidemiologic studies of vitamin D and the risk of prostate cancer have important methodological limitations. One limitation is the use of a single serum sample to estimate vitamin D status. This may be insensitive to detect vitamin D

GARY G. SCHWARTZ AND TAI C. CHEN

insufficiency. A second is the timing of this sample vis à vis the natural history of prostate cancer. The timeframe(s) during which vitamin D protects against prostate cancer is unknown. For example, is it 5–10 years before diagnosis, as might be expected if vitamin D slowed the growth of an existing tumor, or does vitamin D exert effects earlier, perhaps in childhood, as might occur if vitamin D acted as a differentiating agent that altered the eventual phenotype of the adult prostate cell? We have shown that exposure of rats to 1,25(OH)2D at birth dramatically changes the phenotype of the rat prostate in adulthood [59]. If similar phenomena occur among men, then studies that sample vitamin D status in adult life may be measuring vitamin D at a time that is of limited relevance to the natural history of prostate cancer. Studies of lifetime vitamin D exposure, for example, as obtained through residential records and validated solar histories, can be especially informative in this regard (see Section III.C).

B. VDR Polymorphism Studies The effects of vitamin D in prostate cells are mediated through the VDR, which is a member of the steroid/ nuclear receptor superfamily. In target cells, the VDR binds 1,25(OH)2D with high affinity and specificity. The interaction initiates a complex cascade of events beginning with the formation of a heterodimeric complex with the retinoid X receptor (RXR) on specific vitamin D response elements (VDREs) within the promoter region of vitamin D–responsive genes, and influences the rate of RNA polymerase II-mediated transcription of these genes [60] (see Chapters 11 and 13). At present, at least 60 genes including those involved in cell-cycle arrest, apoptosis, and differentiation of prostate cells are known to be regulated by 1,25(OH)2D [61]. The recognition that prostate cells express VDR led to a series of studies of prostate cancer in relation to VDR polymorphisms (see Chapter 68). VDR polymorphisms have been identified in multiple sites including exons 2, 8, and 9 of the VDR gene, which involve Fok I, Bsm I, and Taq I restriction fragment length polymorphisms (RFLPs) (respectively). The Fok I RFLP generates a VDR protein with three additional amino acids at the N-terminus, whereas no coding sequence is affected with Bsm I and Taq I RFLPs. A microsatellite polymorphism in the 3′ untranslated region that does not alter the VDR coding sequence also has been identified. The findings of a number of studies have recently been subjected to a meta-analysis [62]. The first report on VDR polymorphisms and prostate cancer appeared in 1997 by Taylor et al. who reported an association between Taq I RFLP and prostate cancer

CHAPTER 90 Vitamin D, Sunlight, and the Natural History of Prostate Cancer

risk in men from North Carolina [63]. Subsequently, numerous studies from Asia, Europe, and the U.S. have shown significant associations between prostate cancer and Taq I [64–67] Bsm I [68,69], Fok I [70], and the poly-A microsatellite [71]. Conversely, a similar number of studies have reported nonsignificant or null findings [72–79]. The inconsistencies in these findings may be due to several factors. First, the Bsm I polymorphism is a poor marker for the VDR 3′ untranslated region in some populations [80], and this may explain contradictory findings using this marker. Second, the existence of associations between prostate cancer and VDR polymorphisms may depend upon the stage of prostate cancer examined and upon the choice of the control group. For example, Hamasaki et al. found that the frequency of the TT genotype was significantly higher among prostate cancer patients with locally advanced or metastatic disease when these were compared to controls without cancer (OR = 3.52, 95% CI 1.59–7.80; TT vs. Tt/tt) but not when they were compared to controls with BPH [67]. Third, variation in skin types that determine vitamin D3 synthesis may influence the outcome of these studies [81]. Most importantly, most studies of VDR polymorphisms have been performed in the absence of serum levels of vitamin D and/or in the absence of data on solar exposures. It is possible that, as in the study by Ma et al., VDR polymorphisms convey risk only when 25-OHD levels are marginal or low. Lastly, variability in 1-α-hydroxylase levels in the prostate likely play an important role in vitamin D responsiveness (see below), but these have yet to be investigated in conjunction with VDR polymorphisms.

C. Studies of Sunlight Exposure As noted above, for most individuals, sunlight is the most important source of vitamin D. A single whole body exposure to a minimal erythema dose of solar radiation can produce approximately 10,000 IU (250 µg) vitamin D3. Exposure of only parts of the body produces smaller but considerable amounts of vitamin D. For example, Barger-Lux and Heaney calculated that for lightly clad men at northern latitudes in the U.S. (Nebraska, Kansas, and North Dakota) summertime work outdoors was equivalent to a daily oral dosing of 69.5 µg vitamin D3 (2780 IU vitamin D3) [82]. Several epidemiologic studies have examined the risk of prostate cancer in relation to exposure to ultraviolet (UV) radiation. Luscombe and colleagues conducted a case-control study of prostate cancer and UV exposure in North Staffordshire, UK. Two hundred ten men with prostate cancer were compared to 155 men with BPH

1605

on various measures of lifetime sunlight exposure. A high sunbathing score was significantly protective for prostate cancer (OR = 0.83, 95% CI 0.76–0.89). Conversely, a low exposure to ultraviolet radiation was associated with a significantly increased risk (OR = 3.03, 95% CI = 1.59–5.78). Interestingly, multiple sunburns during childhood were significantly inversely associated with risk of prostate cancer (OR = 0.18, 95% CI 0.08–0.38). Subsequently this group reported that among men with low levels of ultraviolet exposure, those with an inability to tan (skin type 1 of the Fitzpatrick system, “always burn/never tan” [83]) were significantly protected compared to other skin types (types 2–4). They interpreted these data to indicate that among men with low levels of UV exposure, an inability to tan is advantageous because it permits greater cutaneous synthesis of vitamin D [84]. An important methodological issue in retrospective studies such as Luscombe et al. is recall bias. Due to the popularization of the vitamin D hypothesis in the lay press (e.g. [85,86]), many men with prostate cancer may be familiar with this idea and, consciously or not, may under-report their actual exposures to ultraviolet radiation, leading to a bias in support of the hypothesis. Three studies that circumvent this problem are those of Freedman et al. and Robsahm et al., both of which determined ultraviolet exposures using data from death certificates, and John et al., which used data on sunlight exposures that were collected prospectively. Freedman et al. [87] conducted a death certificatebased case-control study of mortality from prostate in association with residential and occupational exposure to sunlight. Cases were deaths from cancer between 1984 and 1995 in 24 states. Controls were agefrequency matched to cases and were deaths from causes other than cancer and neurological diseases like multiple sclerosis, which are thought to involve sunlight exposure [88]. Occupational exposure to sunlight was classified based on usual occupation listed on the death certificate. Residential exposure to sunlight was classified by state of residence at birth and at death. In order to reduce error from persons with varied solar histories, persons with discrepant places of birth and death (about 25% of the sample) were excluded. These authors found that high residential (but not occupational) exposure to sunlight was associated with a significantly decreased risk of fatal prostate cancer (OR = 0.90, 95% CI 0.87–0.93). Robsahm and colleagues [89] investigated whether variation in ultraviolet radiation influenced the prognosis of prostate cancers diagnosed in Norway. Due to its large spread in latitude (ranging from 58° N to 71° N), Norway experiences large regional and seasonal

1606 differences in ultraviolet radiation [90]. Robsahm et al. used death certificate data to measure occupational sun exposure, residential region in one of eight predefined north-south strata, and season of diagnosis for prostate cancer. No differences were observed for occupational exposure or for case-fatality rates in the different geographic regions. However, significant variation in case-fatality rates were seen according to season of diagnosis. Diagnoses made in the summer and fall, when serum 25-OHD levels are highest, were associated with significantly lower case-fatality rates. The authors concluded that vitamin D levels at the time of diagnosis significantly alter prognosis from prostate cancer. John et al. [91] analyzed data from the First National Health and Nutrition Examination Survey (NHANES I) Epidemiologic Follow-up Study in order to test the hypothesis that sunlight exposure reduces the risk of developing prostate cancer. One hundred and fifty-three men with incident prostate cancer were identified from a cohort of 3,414 white men who completed the dermatologic examination and were followed up to 1992. Age-adjusted relative risks (RR) and 95% confidence intervals (CI) were estimated for various measures of sunlight exposure using Cox proportional hazards. The data were adjusted for the confounding effects of education, income, BMI, height, alcohol consumption, smoking, physical activity, energy intake, and intake of fat and calcium. The state of longest residence in the South (RR = 0.58, CI = 0.38–0.88, p < 0.01) and high solar radiation in the state of birth (RR = 0.48, CI = 0.30–0.76, p < 0.01) were associated with substantial and significant reductions in the risk of prostate cancer. The prospective design utilized by John et al. essentially precludes the possibility of recall bias. Moreover, loss to follow up in NHANES I was very low (~5%), greatly minimizing the possibility of selection bias. These findings are consistent with those of Hanchette and Schwartz at the ecologic level, which showed a 50% mortality difference from northern to southern U.S. counties (Figs. 2 and 3). The findings of John et al. also are consistent with those of Friedman and colleagues with respect to a protective effect of residential sunlight exposure. The protective effects of solar exposure at place of birth observed by John et al. also are consistent with the findings of Luscombe and colleagues that high solar radiation in childhood (i.e., frequent sunburns as a child) exerts a protective effect. In summary, the data from analytic epidemiologic studies support the hypothesis that exposure to sunlight protects against clinical prostate cancer. Moreover, at least some of this protection appears to occur relatively early in life.

GARY G. SCHWARTZ AND TAI C. CHEN

D. Studies of Dietary Vitamin D Recently, several epidemiologic studies of prostate cancer risk have included information on dietary vitamin D (e.g, Kristal et al., 2002 [92]). The quantity of vitamin D typically observed in these studies is small (less than 400 IU/day). Heaney et al. calculated that a dose of 400 IU/d would raise serum levels of 25-OHD3 only modestly, by 7.0 nmol/L (< 3 ng/ml). Given the variability in assays for measuring 25-OHD, this quantity may be too low to significantly alter serum 25-OHD levels reproducibly in individuals [93,94]. Thus, small differences in vitamin D levels, such as those typically associated with Western diets, may be difficult to detect and may be of limited biologic significance with respect to prostate cancer. Conversely, diets that are high in oily fish can contain appreciable quantities of vitamin D. In this regard, several epidemiologic studies have reported that frequent consumption of fish is associated with a reduced risk of prostate cancer. In a case-control study in the UK, Ewings and Bowie compared 159 cases with prostate cancer to 161 men with benign prostatic hypertrophy and 164 nonurological hospital controls. They found no increased risk associated with dietary fat, sexual activity, and farming—the hypotheses originally under investigation, but found a striking effect of fish consumption (0/159 cases reported fish consumption vs. 14/325 controls, for an “undefined” (infinite) OR, 95% CI = 0.00–0.60) [95]. The Health Professional’s Follow-up Study also reported strong inverse associations between fish intake and risk of metastatic prostate cancer. Similarly, a study from the Swedish Twin Registry reported a significant relationship between total fish consumption and prostate cancer mortality [96]. Other positive and negative studies have been reported and are summarized in a recent review [97]. Because the quantity of vitamin D varies greatly among different fishes and even among different parts of the same fish (e.g., livers may contain 100 times the vitamin D3 of skin or viscera) [98], future studies of fish consumption and prostate cancer risk should focus on the type and parts of fish consumed.

IV. EXPERIMENTAL STUDIES OF THE VITAMIN D HYPOTHESIS By far, the greatest influence of the vitamin D hypothesis has been on laboratory investigations. The effects of vitamin D on prostate cells is the subject of numerous recent reviews [99–103] and is discussed in detail in Chapter 94. Here, we note that these studies demonstrate the presence of VDR in prostate cells and establish that

CHAPTER 90 Vitamin D, Sunlight, and the Natural History of Prostate Cancer

vitamin D metabolites exert pleiotropic and often profound anticancer effects upon these cells. The first published laboratory study on the presence of VDR in prostate cancer was performed by Miller and colleagues (1992) and utilized the LNCaP prostate cancer cell line. Although LNCaP cells are immortal (and thus have one essential feature of cancerous cells), they are considered to be a model of well-differentiated prostate cells (e.g., they express androgen receptor and secrete Prostate Specific Antigen). Miller et al. demonstrated the existence of functional VDR in LNCaP cells and showed that exposure of these cells to 1,25(OH)2D stimulated their differentiation. This finding, they concluded, “is consistent with the hypothesis of Schwartz and Hulka in that physiological concentrations of vitamin D3 promote the differentiation of prostatic carcinoma cells.” These results have been replicated and extended by many other groups [104–107]. Subsequently, VDR were demonstrated in seven well-characterized human prostate cancer cell lines [108] and physiological levels of 1,25(OH)2D were shown to inhibit their proliferation [104,108]. Moreover, physiological doses of 1,25(OH)2D were shown to inhibit the proliferation of primary cultures of noncancerous human prostate cells [109] and to exert striking anti-invasive and anti-metastatic effects in vitro and in vivo [110,111]. These findings have led to the active investigation of vitamin D compounds as therapeutic agents in prostate cancer [112], discussed in detail in Chapters 94 and 97.

V. 1,25(OH)2D IS AN AUTOCRINE HORMONE IN THE PROSTATE A. 1
-Hydroxylase Is Present in the Prostate Although laboratory studies confirmed that 1,25(OH)2D inhibits the proliferation, invasion, and metastasis of prostate cancers, a major conceptual problem for the vitamin D hypothesis remained: we had shown that prostate cancer mortality rates are inversely correlated with UV radiation. We interpreted these findings in support of the hypothesis that 1,25(OH)2D maintains the differentiated phenotype of prostate cells and that low levels of 1,25(OH)2D increase the risk for clinical prostate cancer. The conceptual problem was this: Systemic levels of the pro-hormone, 25-OHD, are dependent upon exposure to ultraviolet radiation [113]. However, in normal individuals, systemic levels of the active hormone, 1,25(OH)2D, are tightly regulated and are not correlated with systemic levels of 25-OHD [114]. Furthermore, although serum levels of 25-OHD are

1607

lower among African-Americans than Caucasians, serum levels of the 1,25(OH)2D are not [115]. Thus, how could exposure to ultraviolet radiation result in the exposure of prostate cells to higher levels of 1,25(OH)2D?; and, How could the excess mortality from prostate cancer among African-Americans be explained? We reasoned that these problems would be solved if prostate cells synthesized their own 1,25(OH)2D from circulating levels of 25-OHD. We examined this possibility in 1998. We investigated three well-characterized human prostate cancer cell lines, LNCaP, DU145, and PC-3, and two primary cultures of cells derived from noncancerous human prostates (one normal and one BPH) for their ability to synthesize 1,25(OH)2D (i.e, for evidence of 1α-hydroxylase activity) [6]. The enzymatic reactions were performed in the presence of 1,2-dianilinoethane, an antioxidant and free radical scavenger, and in the presence and absence of clotrimazole, a cytochrome P450 inhibitor. To obtain a definitive answer, the product, 1,25(OH)2D, was determined by two different methods. First, we used a thymus receptor binding assay, which specifically recognizes 1,25(OH)2D. Second, we utilized radioactive 25-OHD3 as the substrate, and the radioactive 1,25(OH)2D produced was analyzed with a high performance liquid chromatographic (HPLC) system, which is capable of separating 1,25(OH)2D from other vitamin D metabolites and from products of nonenzymatic reactions. Our data demonstrated clearly that two of the three human prostate cancer cell lines, PC-3 and DU 145 cells, as well as primary cultures of noncancerous prostatic cells, possess 1α-hydroxylase activity. Furthermore, 1α-hydroxylase activity was severalfold higher in the primary cultures from noncancerous prostate tissue than in the prostate cancer cell lines. These data (which were generated prior to the cloning of the 1-α-hydroxylase by several groups in the 1997) subsequently have been confirmed using reverse transcriptase PCR amplification of the 1-α-hydroxylase gene.

B. 25-Hydroxyvitamin D Exerts Antiproliferative Effects on Prostate Cells The intracellular production of 1,25(OH)2D by prostatic cells suggested that 25-OHD might regulate the differentiation and proliferation of prostate cells. This interpretation would be consistent with the excess prostate cancer mortality observed at higher latitudes and with the excess among African-Americans. We therefore studied the effects of 25-OHD and 1,25(OH)2D on the proliferation of primary cultures of prostatic epithelial cells using [3H]thymidine incorporation into

1608

GARY G. SCHWARTZ AND TAI C. CHEN

25(OH)D3

Radioactivity (cpm)

50000 10000 1,25(OH)2D3

800 600 400

24,25(OH)2D3

200 0 0

1

2

4 5 6 7 8 Elution volume (mL)

9

10

11

25(OH)D3

50000 10000 Radioactivity (cpm)

3

+ P450 inhibitor

800 600

1,25(OH)2D3 24,25(OH)2D3

400 200 0 0

1

2

3 4 5 6 7 Elution volume (mL)

8

9

10

11

FIGURE 5

Human prostate cells in primary culture synthesize 1,25(OH)2D3 from 25(OH)D3. HPLC elution profile of tritium activity of lipid extracts from cultured normal human prostate primary cultures incubated with [3H]25-(OH)-D3. Second passage of normal human prostate cells were incubated with nonradioactive 50 nM 25-(OH)-D3, 0.91 µCi/nmol [3H]25-(OH)-D3, and 10 µM DPPD at 37 °C for 2 h in the presence and absence of 20 µM clotrimazole, P450 inhibitor. The lipid extract was applied to a C-18-OH cartridge. The fraction eluted from the cartridge with 6% isopropanol in hexane was dried down under nitrogen and reconstituted in the normal phase solvent containing methylene chloride:isopropanol (19:1). Aliquots of 30 µl of sample plus 10 µl of each 25-(OH)-D3 (100 ng), 24,25(OH)2D3 (100 ng), and 1,25(OH)2D3 (100 ng) as standards (as indicated) were run simultaneously on a 5-µm particle size Econosphere normal phase silica column using 5% isopropanol in methylene chloride at a flow rate of 0.5 ml/min. Fractions collected were dried down under nitrogen, and counted for radioactivity.

DNA and by counting cell number for cell proliferation assays [116] and also by high density growth and clonal growth assays [117]. 25-OHD and 1,25(OH)2D each inhibited growth in a dose- and time-dependent manner. The potencies of 25-OHD and 1,25(OH)2D were not significantly different. Growth inhibition in both the [3H]thymidine incorporation and the clonal assays was evident at 1 nM of 25-OHD. Importantly, the concentrations of 25-OHD used in the assays are

well within the normal physiologic range of 25-OHD in humans (35–100 nM) [118]. These data indicate that 25-OHD, which previously was considered to have little biological activity, can become a potent antiproliferative hormone for prostatic cells that express 1-α-OHase. Our findings established that there are, in fact, two vitamin D systems in the prostate: an endocrine system in which 1,25(OH)2D is manufactured by the kidney and an autocrine system in which the prostate manufactures its own 1,25(OH)2D. These findings have implications for the interpretation of epidemiologic studies of prostate cancer risk in relation to vitamin D. For example, if prostate cells synthesize their own 1,25(OH)2D in vivo, then systemic levels of 1,25(OH)2D may not reflect levels of 1,25(OH)2D at the level of the prostate cell. Thus, the risk of prostate cancer may be influenced by intraprostatic as well as systemic levels of 1,25(OH)2D. These data suggest that risk of prostate cancer may be more closely associated with serum levels of 25-OHD than with 1,25(OH)2D, a finding consistent with the pattern of results that has emerged from seroepidemiologic studies. The autocrine synthesis of 1,25(OH)2D by prostatic cells also has important implications for the use of vitamin D metabolites in prostate cancer chemoprevention. It is now clear that 1,25(OH)2D exerts pleitropic anticancer effects on normal and cancerous prostate cells. However, 1,25(OH)2D is not suitable as a chemopreventive agent because of the risk of hypercalcemia. Our findings raise the possibility that by increasing the available substrate, supplementation of men with 25-OHD or vitamin D could reduce the risk of cancer by promoting the synthesis of 1,25(OH)2D by prostatic cells. Because the 1,25(OH)2D that is produced within prostatic cells should exert its biological effects within the cell, 1,25(OH)2D would not be released into the systemic circulation and the problem of hypercalcemia would be greatly reduced.

C. 1-
-Hydroxylase Levels Are Lower in Prostate Cancers than in Noncancerous Prostates The findings that prostate cancer cell lines had less 1α-hydroxylase activity than the primary cultures made us wonder whether a decrease in 1α-hydroxylase activity is a characteristic of cancerous prostate cells in general. Consequently, we compared 1α-hydroxylase activity in cells derived from normal, benign prostatic hyperplasia (BPH) and cancerous prostate tissues by HPLC. A comparison among cells derived from four cancer, two BPH, and three normal prostate tissues

1609

CHAPTER 90 Vitamin D, Sunlight, and the Natural History of Prostate Cancer

Control

25-OH-D3

1,25(OH2)D3

1 nM

10 nM

100 nM

FIGURE 6

Clonal growth of prostatic epithelial cells in response to 25(OH)D3 and 1,25(OH)2D3. Six hundred cells were inoculated into 60-mm, collagen-coated dishes containing serum-free media with vehicle of the indicated concentrations of 25(OH)D3 or 1,25(OH)2D3. After 14 days of incubation, the cells were fixed and stained. Each dose was tested on triplicate dishes. Representative dishes from each treatment are shown. Reprinted from Barretto et al., 2000, with permission.

indicated a marked decrease in 1α-hydroxylase activity in the prostate cancer cells, including an undetectable level of activity in LNCaP cells [119]. This finding has been replicated in a larger sample by Hsu et al. (2001) [120]. The differential expression of 1α-hydroxylase activity among these cells is likely regulated at the level of the promoter [121,122]. These findings confirm that loss of 1α-hydroxylase activity, an enzyme that synthesizes the growth-inhibitory hormone, 1α,25(OH)2D, is associated with prostate cancer. However, observational studies cannot determine whether the loss of 1α-hydroxylase activity is a cause or a consequence of prostate carcinogenesis. One approach to investigate whether prostate cancer cells are cancerous (e.g., grow independently from normal growth regulatory signals) because they have lost their 1α-hydroxylase activity due to an inability to turn on its gene, is to transfect prostate cancer cells with 1α-hydroxylase cDNA and expression vectors. This should confer antiproliferative activity to 25-OHD in the transfected cells. Since LNCaP cells have little 1α-hydroxylase activity and their proliferation is not inhibited by 25-OHD but is inhibited by 1α,25(OH)2D, we transfected these cells with 1α-hydroxylase cDNA plasmid. Transient or stable transfection of 1α-OHase cDNA into LNCaP cells increased 1α-hydroxylase

activity from undetectable to 4.95 ± 0.69 and 5.8 ± 0.7 pmol/mg protein/hour (respectively). In response to 25(OH)D, transfected LNCaP cells showed a significant inhibition of 3H-thymidine incorporation (37 ± 6 % and 56 ± 4% at 10−8 M for transiently and stably transfected cells, respectively). These findings confirm an important autocrine role for 1α,25(OH)2D in the prostate. Furthermore, they suggest that, in conjunction with the systemic administration of 25-OHD, the introduction of the 1α-hydroxylase gene to prostate cancer cells could constitute an endocrine form of gene therapy [123].

D. The Prostatic 1α-hydroxylase Is Not Regulated by PTH or Calcium The discovery that noncancerous prostate cells possess 25-hydroxyvitamin D-1α-hydroxylase activity raises the possibility that vitamin D or 25-OHD could be used to chemoprevent prostate cancer. However, in order for the prostatic synthesis of 1α,25(OH)2D to be useful in cancer chemoprevention, the prostatic 1α-hydroxylase must not be under the same tight control as is the renal 1α-hydroxylase. We therefore examined whether the prostate 1α-hydroxylase was regulated

GARY G. SCHWARTZ AND TAI C. CHEN

4

VI. VITAMIN D HYPOTHESIS: CONCLUSIONS

3

Since the vitamin D hypothesis was proposed in 1990, large strides have been made in our understanding of the role of vitamin D in the natural history of prostate cancer. The similarities between the descriptive epidemiology of prostate cancer and vitamin D deficiency have catalyzed investigations in fields as diverse as epidemiology, biochemistry, and experimental therapeutics. Although there is some inconsistency in the observational studies, the results of the larger seroepidemiologic studies, together with the results of studies of sunlight exposure, support the hypothesis that vitamin D insufficiency is causally related to prostate cancer. Experimental studies demonstrate unambiguously that vitamin D metabolites exert prodifferentiating, antiproliferative, antiinvasive, and antimetastatic effects on prostate cells (see Chapter 94). Our discovery that 1,25(OH)2D is an autocrine hormone in the prostate provides the biochemical link between the epidemiologic and experimental data and has opened a new endocrine window on the prostate cell. The recognition of an expanded role for vitamin D in the prostate suggests roles for vitamin D in many stages of the natural history of prostate cancer. The burden of prostate cancer can be conceptualized as an

1

0

ND N1 N2 N3 B1 B2 P1 P2 P3 P4 DU PC3 LNCaP 145 CaP Normal BPH

FIGURE 7 1α-OHase activity in primary cultures of normal, BPH, and prostate cancer (CaP), and in human prostate cancer cell lines, DU 145, PC-3, and LNCaP cells. Bars are standard deviations of three determinations. The four primary prostate cultures were obtained from a 63-year-old Caucasian (P1), a 50-year-old African-American (P2), a 67-year-old Caucasian (P3) and a 53-year-old Caucasian (P4) with prostate cancer. Three normal cultures were obtained from histologically normal prostates of a 21-year-old and a 27-year-old donor and a 42-year-old AfricanAmerican organ donor. BPH cultures were derived from open prostatectomy specimens of a 58-year-old and a 60-year-old Caucasian. Reprinted from Whitlatch et al., 2002, with permission.

by calcium, parathyroid hormone, and 1α,25(OH)2D3, three major regulators of the renal 1α-hydroxylase [124]. Treatment of primary prostate epithelial cells derived from normal prostate tissue for 24 hours with PTH at 10 nM and 100 nM had no significant effect on the 1α-hydroxylase activity. Conversely, enzyme activity decreased to 66 ± 4 and 20 ± 11% of the control in the presence of 10 nM and 100 nM 1α,25(OH)2D3, respectively. Using the transformed noncancerous PZ-HPV-7 cells (cells that were derived from epithelial cells of the peripheral zone of the normal prostate tissue by transfecting with HPV18 DNA [125]), no significant changes in 1α-hydroxylase activity were observed either at 6 or 24 h after media calcium concentration was changed from 0.03 mM to 1.2 mM with EGF. In contrast, 1α,hydroxylase activity in HKC-8 kidney cells, used as a positive control, was inhibited 40% in the presence of 1.2 or 2.4 mM calcium. The demonstration that the intra-prostatic synthesis of 1α,25(OH)2D in cultures is unaffected by PTH and calcium confirms that the prostate 1α-hydroxylase is distinct from the renal enzyme. The lack of regulation by PTH and calcium supports the use of vitamin D and 25-OHD as chemopreventive agents for prostate cancer because their administration should cause an increased synthesis of 1α,25(OH)2D within prostate cells.

120 incorparation (% of Control)

2

3H-Thymidine

1α-OHase Activity (pmol/mg protein/hr)

1610

A

100

B

80 * 60

*

40 20 0

ol

) ) .1 .1 .1R3 e (AS se (S R3 R3 C a C s PC P P Ha -OH le e le α -O tab Stab Has 1α .1-1 S O 1 3 . 3 R α 1 CR PC ntr

Co

P

FIGURE 8 Transfection of 1α-OHase into LNCaP cells confers antiproliferative activity to 25(OH)D. Panel A: Effect of 25(OH)D3 (10−8M) on 3H-thymidine incorporation into DNA of LNCaP cells with or without transient transfection with PCR 3.1 vector, anti-sense (AS), or sense PCR 3.1-1α-OHase cDNA (S). Panel B: Effect of 25(OH)D3 (10−8M) on 3H-thymidine incorporation into DNA of LNCaP cells stably transfected with vector PCR 3.1 or with sense PCR 3.1-1α-OHase cDNA. Bars indicate the standard deviation of 8 determinations. *p < 0.05. Reprinted from Whitlatch et al., 2002, with permission.

1611

CHAPTER 90 Vitamin D, Sunlight, and the Natural History of Prostate Cancer

POTENTIAL INTERVENTIONS WITH VITAMIN D METABOLITES

CONDITION

POSSIBLE THERAPY

• Minimal residual disease

1,25 (OH)2D & Analogs

• Clinical cancer • Prostatic intraepithelial neoplasia

25-OH-D

• Men at risk

Vitamin D

FIGURE 9 The natural history of prostate cancer and possible treatments with vitamin D. At each stage of prostate cancer, there is a corresponding treatment with some form of vitamin D, ranging from the hormonal precursor (vitamin D3) in men at risk for prostate cancer, to active vitamin D, 1,25(OH)2D3 and its analogs, for men with advanced disease.

iceberg (see Fig. 9). The vast majority of prostate cancers are latent or subclinical (i.e., beneath the waterline in the figure). At the apex of the iceberg are men with clinical disease. These men, especially those men with disease that no longer responds to androgen withdrawal, may be candidates for clinical trials using the hormonal form of vitamin D or its less calcemic analogs. Because clinical prostate tumors appear to have less 1α-hydroxylase than noncancerous prostates, men with clinical cancer are unlikely to benefit from prohormonal forms of vitamin D and should be treated with some form of active vitamin D. Men who do not have clinical disease but who are at increased risk for prostate cancer, such as men with histological evidence of prostate intraepithelial neoplasia (PIN), a possible precursor of prostate cancer [126], or men with a positive family history of prostate cancer, may consider supplementation with 25-OHD or vitamin D, as these drugs should be converted to 1,25(OH)2D intraprostatically. Finally, for men at the base of the iceberg, i.e., virtually all other men, especially men who live at extreme geographic latitudes and/or who receive little effective exposure to sunlight, prophylactic supplementation with vitamin D may be prudent. In the 20th century, public health programs of vitamin D supplementation virtually eliminated rickets.

Whether vitamin D can reduce the burden of prostate cancer is a challenge for public health in the 21st century.

Acknowledgments Research described in this chapter has been supported by NCI R03CA 48440, NCI R01CA68565, and NCI R03CA85750 (to GGS) and by U.S. Army DAMD 17-01-1-0025 (to TCC).

References 1. Schwartz GG, Hulka BS 1990 Is vitamin D deficiency a risk factor for prostate cancer (hypothesis)? Anticancer Res 11:1307–1311. 2. Schwartz GG 2000 Prostate cancer and vitamin D: From concept to clinic. A 10-year update. In: AW Norman, R Bouillon, M Thomasset (eds) Vitamin D Endocrine System. University of California Press, pp. 445–452. 3. Holick MF. Photobiology of vitamin D, In: Feldman D, Glorieux FH, Pike JW (eds) Vitamin D, 1st ed. Academic Press: San Diego, CA: 33–39. 4. Holick MF 1990 The use and interpretation of assays for vitamin D and its metabolites. J Nutr 120(Suppl 11): 1464–1469.

1612 5. Zehnder D, Bland R, Williams MC, McNinch RW, Howie AJ, Sterwart PM, Hewison M 2001 Extrarenal expression of 25-hydroxyvitamin D3-1 alpha-hydroxylase. J Clin Endocrinol Metab 86:888–894. 6. Schwartz GG, Whitlatch LW, Chen TC, Lokeshwar BL, Holick MF 1998 Human prostate cells synthesize 1,25-dihydroxyvitamin D3 from 25-hydroxyvitamin D3. Cancer Epidemiol Biomark Prev 7:391–395. 7. Hansen CM, Binderup L, Hamberg KJ, Carlberg C 2001 Vitamin D and cancer: Effects of 1,25(OH)2D3 and its analogs on growth control and tumorigenesis. Front Biosc 6:D820–D848. 8. Omdahl JL, Morris HA, May 2002 Hydroxylase enzymes of the vitamin D pathway: Expression, function, and regulation. Annu Rev Nutr 22: 139–166. 9. Vieth R 1999 Vitamin D supplementation, 25-hydroxyvitamin D concentrations, and safety. Am J Clin Nutr 69:842–856. 10. Glerup H, Mikkelsen K, Poulsen L, Hass E, Overbeck S, Thomsen H, Charles P, Eriksen EF 2000 Commonly recommended daily intake of vitamin D is not sufficient if sunlight exposure is limited. J Intern Med 247:260–268. 11. American Cancer Society, Cancer Facts & Figures, 2003. 12. Kurihara M, Aoki K, Hismamichi S (eds.) 1989 Cancer mortality statistics in the world, 1950–1985. University of Nagoya Press: Japan. 13. Sakr WA, Haas GP, Cassin BF, Pontes JE, Crissman JD 1993 The frequency of carcinoma and intraepithelial neoplasia of the prostate in young adult patients. J Urol 150: 379–385. 14. Breslow N, Chan CW, Dhom G, et al. 1977 Latent carcinoma of the prostate in seven areas. Int J Cancer 20:680–688. 15. Guileyardo JM, Johnson WD, Welsh RA, Akazaki K, Correa P 1980 Prevalence of latent prostate carcinoma in two U.S. populations J Natl Cancer Inst 65:311–316. 16. Yatani R, Chigusa I, Akazaki K, Stermmerman G, Welsh R, Correa P 1984 Geographic pathology of latent prostate carcinoma. Int J Cancer 29: 611–616. 17. Haenzel W, Kurihara M 1968 Studies of Japanese migrants: I. Mortality from cancers and other diseases among Japanese in the United States. J Natl Cancer Inst 40:344–360. 18. Nakata S, Takahashi H, Ohtake N, Takei T, Yamanaka H 2000 Trends and characteristics in prostate cancer mortality in Japan. Int J Urol 7:254–257. 19. Russell RM 2000 The aging process as a modifier of metabolism. Am J Clin Nutr 72 529S–532S. 20. MacLaughlin J, Holick MF 1985 Aging decreases the capacity of human skin to produce vitamin D3. J Clin Invest 76:1536–1538. 21. Lips P 2001 Vitamin D deficiency and secondary hyperparathyroidism in the elderly: Consequences for bone loss and fractures and therapeutic implications. Endocr Rev 22:477–501. 22. Passeri G, Pini G, Triano L, Vescovini R, Sansoni P, Passeri M, Guerresi P, Delsignore R, Pedrazzoni M, Franceschi C 2003 Low vitamin D status, high bone turnover, and bone fractures in centenarians. J Clin Endocrinol Metab 88: 5109–5115. 23. Matsuoka LY, Wortsman J, Haddad JG, Kolm P, Hollis BW 1991 Racial pigmentation and the cutaneous synthesis of vitamin D. Arch Dermatol 127:536–538. 24. Bell NH, Greene A, Epstein S, Oexmann MJ, Shaw S, Shary J 1985 Evidence for alteration of the vitamin D endocrine system in blacks. J Clin Invest 76:470–473.

GARY G. SCHWARTZ AND TAI C. CHEN

25. Nesby-O’Dell S, Scanlon KS, Cogswell ME, Gillespie C, Hollis BW, Looker AC, Allen C, Doughertly C, Gunter EW, Bowman BA 2002 Hypovitaminosis D prevalence and determinants among African American and white women of reproductive age: Third National Health and Nutrition Examination Survey, 1988–1994. Am J Clin Nutr 76:187–192. 26. Nakamura K, Nashimoto M, Hori Y, Muto K, Yamamoto M 1999 Serum 25-hydroxyvitamin D levels in active women of middle and advanced age in a rural community in Japan. Nutrition 15: 870–873. 27. Nakamura K, Nashimoto M, Hori Y, Yamamoto M 2000 Serum 25-hydroxyvitamin D concentrations and related dietary factors in peri- and postmenopausal Japanese women. Am J Clin Nutr 71:1161–1165. 28. Nakamura K, Nashimoto M, Okuda Y, Ota T, Yamamoto M 2002 Fish as a major source of vitamin D in the Japanese diet. Nutrition 18:415–416. 29. Takeuchi A, Okano T, Ayame M, Yoshikawa H, Teroaka S, Murakami Y, Kobayashi T 1984 High-performance liquid chromatographic determination of vitamin D3 in fish liver oils and eel body oils. J Nutr Sci Vitaminol 30:421–430. 30. Henderson JB, Dunnigan MG, McIntosh WB, Abdul Motall AA, Gettinby G, Glekin BM 1987 The importance of limited exposure to ultraviolet radiation and dietary factors in the etiology of Asian rickets: A risk-factor model. Q J Med 63:413–425. 31. Stephens WP, Klimiuk PS, Warrington S, Taylor IL, Berry JL, Mawer EB 1982 Observations on the natural history of vitamin D deficiency among Asian immigrants. Q J Med 202:171–188. 32. Abrams SA 2002 Nutritional rickets: an old disease returns. Nutr Rev 60:111–115. 33. Heaney RP 2003 Long-latency deficiency disease: insights from calcium and vitamin D. Am J Clin Nutrition 78:912–919. 34. Schleicher G, Priovvette TH, Stumpf WE 1989 Distribution of soltriol [1,25(OH)2-Vitamin D3] binding sites in male sex organs of the mouse: An autoradiographic study. J Histochem Cytochem 37:1083–1086. 35. Miller GJ, Stapleton GE, Ferrara JA, Lucia MS, Pfister S, Hedlund, TE, Upadhya P 1992 The human prostatic carcinoma cell line LNCap expresses biologically active, specific receptors for 1 alpha,25-dihydroxyvitamin D3. Cancer Res 52(3):515–520. 36. Meade MS, Earickson RJ 2000 Medical Geography, Guilford Press: New York. 37. Hanchette CL, Schwartz GG 1992 Geographic patterns of prostate cancer mortality. Evidence for a protective effect of ultraviolet radiation. Cancer 70:2861–2869. 38. Kafadar K 1998 Geographic patterns of prostate cancer mortality. Spatial smoothing and the need for adjustment. Annals of Epidemiology. 39. Grant WB 2002 An estimate of premature cancer mortality in the U.S. due to inadequate doses of solar ultraviolet-B radiation. Cancer 94:1867–1875. 40. Di Silverio F, La Pera G, Tenaglia R 1982 Age-adjusted mortality rate and regional distribution for prostatic carcinoma in Italy between 1969 and 1978. Prostate 3:631–636. 41. Vercelli M, Quaglia A, Marani E, Parodi S 2000 Prostate cancer incidence and mortality trends among elderly and adult Europeans. Critical Reviews in Oncology/Hematology, 35:133–144. 42. Schwartz GG 2002 Vitamin D and the Big Three: Cancers of the Colon, Breast, and Prostate. In: MF Holick

CHAPTER 90 Vitamin D, Sunlight, and the Natural History of Prostate Cancer

43. 44.

45.

46. 47.

48.

49. 50.

51.

52. 53.

54.

55.

56.

57. 58. 59.

(ed) Biological Effects of Light, Kluwer: Amsterdam: 255–265. Braun MM, Helsouer KH, Hollis BW, Comstock GW 1995 Prostate cancer and prediagnostic levels of serum vitamin D metabolites. Cancer Causes Control 6:235–239. Nomura AMY, Stemmerman GN, Lee J, Kolonel LN, Chen TC, Turner, A, Holick MF 1998 Serum vitamin D metabolite levels and the subsequent development of prostate cancer (Hawaii, United States). Cancer Causes Control 9:525–432. Ma J, Stampfer MJ, Gann PH, Hough HL, Giovannucci E, Kelsey KT, Hennekens CH, Hunter DJ 1998 Vitamin D receptor polymorphisms, circulating vitamin D metabolites, and risk of prostate cancer in United States physicians. Cancer Epidemiol Biomarkers Prev 7: 385–390. Ahonen MH, Tenkanen L, Teppo L, Hakama M, Tuohimaa P 2000 Prostate cancer risk and prediagnostic serum 25-hydroxyvitamin D levels (Finland). Cancer Causes Control 11:847–852. Oberg F, Botling J, Nilsson K 1993 Functional antagonism between vitamin D3 and retinoic acid in the regulation of CD14 and CD23 expression during monocytic differentiation of U-937 cells. J Immunol 150:3487–3495. Gibson DF, Bikle DD, Harris J 1998 All-trans retinoic acid blocks the antiproliferative prodifferentiating actions of 1,25dihydroxyvitamin D3 in normal human keratinocytes. J Cell Physiol 174:1–8. Vieth R 2004 Enzyme kinetic hypothesis to explain the Ushaped risk curve for prostate cancer vs. 25-hydroxyvitamin D in Scandinavian countries [letter]. Int J Cancer, 111:468. Mawer EB, Walls J, Howell A, Daview M, Ratcliffe WA, Bundred NJ 1997 Serum 1,25-dihydroxyvitamin D may be related inversely to disease activity in breast cancer patients with bone metastases. J Clin Endocrinol Metab 82: 118–122. Niv Y, Sperber AD, Figer A, Igael D, Shany S, Fraser G, Schwartz B 1999 In colorectal carcinoma patients, serum vitamin D levels vary according to stage of the carcinoma. Cancer 86:391–397. Giovannucci E 1998 Dietary influences of 1,25(OH)2D in relation to prostate cancer: A hypothesis. Cancer Causes Control 9:467–472. Chan JM, Giovannucci E, Andersson S-O, Yeun J, Adami H-O, Wolk A 1998 Dairy products, calcium, phosphorus, vitamin D, and risk of prostate cancer. Cancer Causes Control 9:559–566. Chan JM, Stampfer MJ, Ma J, Gann PH, Gaziano JM, Giovannucci EL 2001 Dairy products, calcium, and prostate cancer risk in the Physicians’ Health Study. Am J Clin Nutr 74:549–554. Schuurman AG, van den Brandt PA, Dorant E, Goldbohm RA 1999 Animal products, calcium, and protein and prostate cancer risk in the Netherlands Cohort Study. Br J Cancer 80:1107–1113. Berndt SI Carter HB, Landis PK, Tucker KL, Hsieh LJ, Metter EJ, Platz EA 2002 Calcium intake and prostate cancer risk in a long-term aging study: The Baltimore Longitudinal Study of Aging. Urology 60:1118–1123. Vieth R 2002 Dairy products, calcium, and prostate cancer risk in the Physicians’ Health Study [letter]. Am J Clin Nutr 76:490–491. Bell NH, Shaw S, Turner RT 1987 Evidence that calcium modulates circulating 25-hydroxyvitamin D in man. J Bone Miner Res 2:211–214. Konety BR, Schwartz GG, Acierno JS, Becich MJ, Getzenberg RH 1996 The role of Vitamin D in normal prostatic

60. 61.

62. 63. 64.

65.

66.

67.

68.

69.

70.

71.

72. 73. 74.

1613

growth and development. Cell Growth & Differentiation 7:1563–1570. Christako S, Dhawan P, Liu Y, Peng X, Porta A 2003 New insights into the mechanisms of vitamin D action. J Cell Biochem 88:695–705. Hedlund TE, Moffatt KA, Miller GJ 1996 Stable expression of the nuclear vitamin D receptor in the human prostatic carcinoma cell line JCA-1: Evidence that the antiproliferative effects of 1α,25-dihydroxyvitamin D3 are mediated exclusively through the genomic signaling pathway. Endocrinol 137:1554–1561. Ntais C, Polycarpou A, Ionnidis PA 2003 Vitamin D receptor gene polymorphisms and risk of prostate cancer. Cancer Epidem Biomark Prev 12:1395–1402. Taylor JA, Hirvonen A, Watson M, Pittman G, Mohler JL, Bell DA 1996 Association of prostate cancer with vitamin D receptor genet polymorphism. Cancer Res 56:4108–4110. Correa-Cerro L, Berthon P, Haussler J, Bochum S, Drelon E, Mangin P, Fournier G, Paiss T, Cussenot O, Vogel W 1999 Vitamin D receptor polymorphisms as markers in prostate cancer. Human Genet 105:281–287. Hamasaki T, Inatomi H, Katoh T, Ikuyama T, Matsumoto T 2001 Clinical and pathological significance of vitamin D receptor gene polymorphism for prostate cancer which is associated with a higher mortality in Japanese. Endocr J 48:543–549. Medeiros R Morais A, Vasconcelos A, Costa S, Pinto D, Oliveira J, Lopes C 2002 The role of vitamin D receptor gene polymorphisms in the susceptibility to prostate cancer of a southern European population. J Human Genet 47:413–418. Hamasaki T, Inatomi H, Katoh T, Ikuyama T, Matsumoto T 2002 Significance of vitamin D receptor gene polymorphism for risk and disease severity of prostate cancer and benign prostatic hyperplasia in Japanese. Urol Int 68:226–231. Ingles SA, Coetzee GA, Ross RK, Henderson BE, Kolonel LN, Crocitto KL, Wang W, Haile RW 1998 Association of prostate cancer with vitamin D receptor haplotypes in AfricanAmericans. Cancer Res 58:1620–1623. Habuchi T, Suzuki T, Sasaki R, Wang L, Sato K, Satoh S, Akao T, Tsuchiya N, Shimoda N, Wada Y, Koizumi A, Chihara J, Ogawa O, Kato T 2000 Association of vitamin D receptor gene polymorphism with prostate cancer and benign prostatic hyperplasia in a Japanese population. Cancer Res 60:305–308. Xu Y, Shibata A, McNeal JE, Stamey TA, Feldman D, Peehl DM 2003 Vitamin D receptor start codon polymorphism (FokI) and prostate cancer progression. Cancer Epidemiol Biomark Prev 12:23–27. Ingles SA, Ross RK, Yu MC, Irvine RA, La Pera G, Haile RW, Goetzee GA 1997 Association of prostate cancer risk with genetic polymorphisms in vitamin D receptor and androgen receptor. J Natl Cancer Inst 89:166–170. Kibel AS, Issacs SD, Isaacs WB, Bova GS 1998 Vitamin D receptor polymorphisms and lethal prostate cancer. J Urol 160:1405–1409. Furuya Y, Akakura K, Masai M, Ito H 1999 Vitamin D receptor gene polymorphism in Japanese patients with prostate cancer. Endocr J 46:467–470. Watanabe M, Fukutome K, Murata M, Uemura H, Kubota Y, Kawamura J, Yatani R 1999 Significance of vitamin D receptor gene polymorphism for prostate cancer risk in Japanese. Anticancer Res 19:4511–4514.

1614 75. Blazer DG, Umbach DM, Bostick RM, Taylor JA 2000 Vitamin D receptor polymorphisms and prostate cancer. Mol Carcinogen 27:18–23. 76. Chokklingam AP, McGlynn KA, Gao YT, Pollak M, Deng J, Sesterhenn IA, Mostofi FK, Fraumeni JF, Hsing AW 2001 Vitamin D receptor gene polymorphisms, insulin-like growth factors, and prostate cancer risk: a population-based casecontrol study in China. Cancer Res 61:4333–4336. 77. Gsur A, Madersbacher S, Haidinger G, Schatzl G, Marberger M, Vutuc C, Micksche M 2002 Vitamin D receptor gene polymorphism and prostate cancer risk. Prostate 51:30–34. 78. Suzuki K, Matsui H, Ohtake N, Nakata S, Takei T, Koike H, Nakazato H, Okugi H, Hasumi M, Fukabori Y, Kurokawa K, Yamanaka H 2003 Vitamin D receptor gene polymorphism in familial prostate cancer in a Japanese population. Int J Urology 10:261–266. 79. Tayeb MT, Clark C, Haites NE, Sharp L, Murray GI, McLeod HL 2003 CYP3A4 and VDR gene polymorphisms and the risk of prostate cancer in men with benign prostate hyperplasia. Br J Cancer 88:928–932. 80. Habuchi T, Suzuki T, Sasaki R, Wang L, Sato K, Satoh S, Adao T, Tsuchiya N, Shimoda N, Wada Y, Koizumi A, Chihara J, Ogawa O, Kato T 2000 Association of vitamin D receptor gene polymorphism with prostate cancer and benign prostatic hyperplasia in a Japanese population. Cancer Res 58:1620–1623. 81. Luscombe CJ, French ME, Liu S, Saxby MF, Jones PW, Fryer AA, Strange RC 2001 Prostate cancer risk: associations with ultraviolet radiation, tyrosinase, and melanocortin-1 receptor genotypes. Br J Cancer 85:1504–1509. 82. Barger-Lux MJ, Heaney RP 2002 Effects of above-average summer sun exposure on serum 25-hydroxyvitamin D and calcium absorption. J Clin Endocrinol Metab 87: 4952–4956. 83. Fitzpatrick TB 1988 The validity and practicality of sunreactive skin types I through VI. Arch Dermatol 124:869–871. 84. Bodiwala D, Luscombe CJ, French ME, Liu S, Saxby MF, Jones PW, Ramachandran S, Fryer AA, Strange RC 2003 Susceptibility to prostate cancer: Studies on interactions between UVR exposure and skin type. Carcinogenesis 24:711–717. 85. Arnot B 2000 The Prostate Cancer Protection Plan: The Foods, Supplements and Drugs that Could Save Your Life. Little, Brown & Co.: Boston MA. 86. Walsh PC, Worthington JF 1995 The Prostate: A Guide for Men. Johns Hopkins University Press: Baltimore, MD. 87. Freedman DM, Dosemeci M, McGlynn K 2002 Sunlight and mortality from breast, ovarian, colon, prostate, and nonmelanoma skin cancer: A composite death certificate based case-control study. Occup Environ Med 59:257–262. 88. Schwartz GG 1992 Multiple sclerosis and prostate cancer: What do their similar geographies suggest? Neuroepidemiol 11:244–254. 89. Robsahm TE, Tretli S, Dahlback A, Moan J 2004 Vitamin D3 from sunlight may improve the prognosis of breast, colon, and prostate cancer. Cancer Causes Control 15:149–158. 90. Maon J, Dahlback A, Henricksen T, Magnus K 1989 Biological amplification factor for sunlight-induced nonmelanoma skin cancer at high latitudes. Cancer Res 49: 5207–5212. 91. John EM, Dreon DM, Koo J, Schwartz GG 2004 Residential sunlight exposure is associated with a decreased risk of prostate cancer. Journal of Steroid Biochemistry & Molecular Biology, 89–90:549–552.

GARY G. SCHWARTZ AND TAI C. CHEN

92. Kristal AR, Cohen JH, Qu P, Stanford JL 2002 Associations of energy, fat, calcium, and vitamin D with prostate cancer risk. Cancer Epidem Biomar 11:719–725. 93. Heaney, RP, Daview KM, Chen TC, Holick MF, Barger-Lux MJ 2003 Human serum 25-hydroxycholecalciferol response to extended oral dosing with cholecalciferol. Am J Clin Nutr 77:204–210. 94. Poskitt EM, Cole TJ, Lawson DE 1979 Diet, sunlight, and 25-hydroxyvitamin D in healthy children and adults. Br Med J 1:221–223. 95. Ewings P, Bowie C 1996 A case-control study of cancer of the prostate in Somerset and east Devon. Br J Cancer 74:661–666. 96. Terry P, Lichtenstein P, Feychting M, Ahlbom A, Wolk A 2001 Fatty fish consumption and risk of prostate cancer. Lancet 357:1764–1766. 97. Terry PD, Rohan TE, Wolk A 2003 Intakes of fish and marine fatty acids and the risks of cancers of the breast and prostate and other hormone-related cancers: A review of the epidemiologic evidence. Am J Clin Nutr 77:532–543. 98. Takeuchi A, Okano T, Sayamoto M, Sawamura S, Kobayashi T, Motosugi M, Yamakawa T 1986 Tissue distribution of 7-dehydrocholesterol, vitamin D3, and 25-hydroxyvitamin D3 in several species of fishes. J Nutr Sci Vitaminol 32:13–22. 99. Miller GJ 1998 Vitamin D and prostate cancer: Biologic interactions and clinical potentials. Cancer Metast Rev 17:353–360. 100. Polek TC, Weigel NL 2002 Vitamin D and prostate cancer. J Androl 23:9–17. 101. Leibowitz SB, Kantoff PW 2003 Differentiating agents and the treatment of prostate cancer: Vitamin D-3 and peroxisome proliferators-activated receptor gamma ligands. Sem Oncol 30:698–708. 102. Peehl DM, Krishnan AV, Feldman D 2003 Pathways mediating the growth-inhibitory actions of vitamin D in prostate cancer. J Nutr 133: 2461S–2469S. 103. Chen TC, Holick MR 2003 Vitamin D and prostate cancer prevention and treatment. Trends Endocrinol Metab 14: 423–430. 104. Skowronski RJ, Peehl DM, Feldman D 1993 Vitamin D and prostate cancer: 1,25-dihydroxyvitamin D3 receptors and actions in human prostate cancer cell lines. Endocrinol 132(5):1952–1960. 105. Hsieh T, Ng C, Mallouh C, Tazaki H, Wu JM 1996 Regulation of growth, PSA/PAP, and androgen receptor expression by 1α,25-dihydroxyvitamin D3 in the androgendependent LNCaP cells. Biochem and Biophys Res Comm 223:141–146. 106. Esquenet M, Swinned JV, Heyns W, Verhoeven G 1996 Control of LNCaP proliferation and differentiation: Actions and interactions of androgens, 1α,25-dihydroxycholecalciferol, all-trans-retinoic acid, 9-cis-retinoic acid, and phenylacetate. Prostate 28:182–194. 107. Schwartz GG, Oeler TA, Uskokovic MR, Bahnson RR 1994 Human prostate cancer cell lines: inhibition of proliferation by vitamin D analogs. Anticancer Res 14:1077–1081. 108. Miller GJ, Stapleton GE, Hedlund TE, Mofatt KA 1995 Vitamin D receptor expression, 24-hydroxylase activity, and inhibition of growth by 1 alpha,25-dihydroxyvitamin D3 in seven human prostatic carcinoma cell lines. Clin Cancer Res 1:997–1003. 109. Peehl DM, Skowronski RJ, Leung GK, Wong ST, Stamey TA, Feldman D 1994 Antiproliferative effects of 1,25-dihydroxyvitamin D3 on primary cultures of human prostatic cells. Cancer Res 54(3):805–810.

CHAPTER 90 Vitamin D, Sunlight, and the Natural History of Prostate Cancer

110. Schwartz GG, Wang M-H, Zhang M, Singh RK, Siegal GP 1997 1α,25-Dihydroxyvitamin D (calcitriol) inhibits the invasiveness of human prostate cancer cells. Cancer Epidemiology, Biomarkers & Prevention 6:727–732. 111. Lokeshwar BL, Schwartz GG, Selzer MG, Burnstein KL, Zhuang S-H, Block NL, Binderup L 1999 Inhibition of prostate cancer metastasis in vivo: A comparison of 1,25dihydroxyvitamin D (calcitriol) and EB1089. Cancer Epidemiology, Biomarkers & Prevention, 8:214–248. 112. Chen TC, Holick MF, Lokeshwar BL, Burnstein KL, Schwartz GG 2003 Evaluation of vitamin D analogs as therapeutic agents for prostate cancer. Recent Results Cancer Res 164:273–288. 113. Poskitt EM, Cole TJ, Lawson DE 1979 Diet, sunlight, and 25-hydroxyvitamin D in healthy children and adults. Br Med J 1:221–223. 114. Chesney RW, Rosen JF, Hamstra AJ, Smith C, Mahaffey K, DeLuca HF 1981 Absence of seasonal variations in serum concentrations of 1,25-dihydroxyvitamin D despite a rise in 25-hydroxyvitamin D in summer. J Clin Endocrin Metab 53:139–143. 115. Bell NH, Greene A, Epstein S, Oexman MH, Shaw S, Shary J 1985 Evidence for alteration of the vitamin D-endocrine system in blacks. J Clin Invest 76:74–76. 116. Chen TC, Schwartz GG, Burnstein KL, Lokeshwar BL, Holick MF 2000 The in vitro evaluation of 25-hydroxyvitamin D3 and 19-nor-1,α,25-dihydroxyvitamin D2 as therapeutic agents for prostate cancer. Clin Cancer Res 6:901–908. 117. Barreto AM, Schwartz GG, Woodruff R, Cramer SD 2000 25-Hydroxyvitamin D3, the prohormone of 1,25-dihydroxyvitamin D3, inhibits the proliferation of primary prostatic epithelial cells. Cancer Epidem Biomar 9:265–270. 118. Hughes MR, Baylink D, Jones PG, Haussler MR 1976 Radioligand receptor assay for 25-hydroxyvitamin D2/D3 and 1α,25-dihydroxyvitamin D2/D3. J Clin Invest 58:61–70. 119. Chen TC, Whitlatch LW, Young MV, Flanagan JN, Schwartz GG, Lokeshwar BL, Kong XF, Xue P, Holick MF 2000

120.

121. 122.

123.

124.

125.

126.

1615

Enhancement of 25-hydroxyvitamin D-1-α-hydroxylase activity in prostate cells by gene transfection: A novel approach for the treatment of prostate cancer. In Vitamin D Endocrine System: Structural, Biological, Genetic, and Clinical Aspects. AW Norman, R Bouillon, M Thomasset (eds) University of California, Printing and Reprographics, Riverside CA, pp. 525–528. Hsu J-Y, Feldman D, McNeal JE, Peehl DM 2001 Reduced 1α-hydroxylase activity in human prostate cancer cells correlates with decreased susceptibility to 25-Hydroxyvitamin D3induced growth inhibition. Cancer Res 61:2852–2856. Wang L, Whitlatch LW, Flanagan JN, Holick MF, Chen TC 2003 Vitamin D autocrine system and prostate cancer. Recent Results Cancer Res 164:223–237. Wang L, Flanagan JN, Jamieson DP, Holick MF, Chen TC Regulation of 25-hydroxyvitamin D-1α-hydroxylase by epidermal growth factor in prostate cells. J Steroid Biochem Mol Biol, in press. Whitlatch LW, Young MV, Schwartz GG, Flanagan JN, Burnstein KL, Lokeshwar BL, Rich ES, Holick MF, Chen TC 2002 25-Hydroxyvitamin D-1α-hydroxylase activity is diminished in human prostate cancer and is enhanced by gene transfer. J Steroid Biochem Mol Biol 81:135–140. Young MV, Schwartz GG, Wang L, Jamieson DP, Whitlatch LW, Flanagan JN, Lokeshwar BL, Holick MF, Chen TC 2004 The prostate 25-hydroxyvitamin D-1α-hydroxylase is not influenced by parathyroid hormone and calcium: Implications for prostate cancer chemoprevention by vitamin D. Carcinogenesis, 25:967–971. Weijerman PC, Konig JJ, Wong ST, Niesters HG, Peehl DM 1994 Lipofection-mediated immortalization of human prostatic epithelial cells of normal and malignant origin using human papillomavirus type 18 DNA. Cancer Res 54:5579–5583. Hsieh K, Albertsen PC 2003 Populations at high risk for prostate cancer. Urol Clin North Am 30:669–676.

CHAPTER 91

Epidemiology of Cancer Risk: Vitamin D and Calcium EDWARD GIOVANNUCCI

ELIZABETH A. PLATZ

Departments of Nutrition and Epidemiology, Harvard School of Public Health, Boston, MA, and Channing Laboratory, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD

I. Introduction II. Colorectal Neoplasms III. Prostate Cancer

IV. Conclusion References

I. INTRODUCTION

In addition to its mineral regulatory functions, in vitro and in vivo studies indicate that calcium and vitamin D metabolites, particularly 1,25(OH)2D, participate in the regulation of cellular proliferation and differentiation. These properties have generated the hypothesis that calcium and vitamin D may have some anticancer properties in humans, particularly for colorectal and prostate cancer. Although 25(OH)D is the best indicator of nutritional status of vitamin D, generally the anticancer properties have been attributed to 1,25(OH)2D. However, recent evidence suggests that other tissues, including the prostate [4] and colon [5] besides the kidney can convert 25(OH)D to 1,25(OH)2D, thus raising the possibility that 25(OH)D can have direct anticancer effects itself in addition to being the precursor to 1,25(OH)2D. While clearly our understanding of potential anticancer mechanisms has increased dramatically recently, the role of calcium and vitamin D in the prevention of human cancer remains unsettled. Results from in vitro or short-term intervention studies based on presumed intermediate endpoints such as proliferation do not lend to simple predictions for cancer risk associated with long-term moderate differences in calcium intake or vitamin D levels. Thus, epidemiologic studies remain an important component of our understanding. Over the past several decades, a number of epidemiologic studies for colorectal and prostate cancer have provided important data concerning the vitamin D and calcium hypothesis. Our knowledge in this area has improved, but many questions remain because of various methodologic limitations. The types of studies used

This chapter will review the epidemiological studies of vitamin D and calcium in relation to risk of colorectal and prostate cancers. Vitamin D physiology is discussed in extensive detail throughout this book, so only a brief introduction to issues relevant for epidemiologic studies is provided here. In epidemiologic studies, two metabolites of vitamin D have been typically measured in the blood, 25-hydroxycholecalciferol (25(OH)D) and 1,25-dihydroxycholecalciferol (1,25(OH)2D). Circulating 25(OH)D is formed in the liver from cholecalciferol, or vitamin D, which itself can be made by photoproduction through ultraviolet light conversion of 7-dehydrocholesterol in the skin [1]. Thus, circulating 25(OH)D concentrations vary with exposure to sunlight, in combination with constitutional factors such as skin pigmentation and aging, which both tend to diminish the capacity for this conversion [2]. In addition to photoproduction in the skin, vitamin D is also contributed by foods, such as fish, eggs, butter, and fortified milk products and breakfast cereals, and vitamin D–containing multivitamins and supplements. Circulating 25(OH)D is the best indicator of nutritional vitamin D status [1]. The hormone 1,25(OH)2D, produced from 25(OH)D in the kidney, enhances calcium absorption from the small intestine and is tightly regulated to ensure calcium homeostasis [3]. Low intakes and low circulating calcium concentrations tend to increase the production of 1,25(OH)2D, which then initiates a series of physiologic actions to stabilize circulating calcium concentrations. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

1618

EDWARD GIOVANNUCCI AND ELIZABETH A. PLATZ

have been those that have examined sunlight exposure as a determinant of vitamin D synthesis in the skin, studies of vitamin D intake and of calcium and dairy food intake, studies of circulating vitamin D metabolites, studies of polymorphisms in the vitamin D receptor, and rarely intervention studies. This chapter will critically review the studies of calcium and vitamin D in relation to colorectal and prostate cancers, the cancers that have been most studied, focusing on strengths and weaknesses of the methodology. Recommendations for future studies will be made.

entire populations, but relatively little data exist for sunlight exposure at the individual level in case-control or cohort studies. One exception is a case-control study of 1993 colon cancer cases and 2410 population controls interviewed in Northern California, Utah, and Minnesota that did not find a relationship with reported sunshine exposure [12]. Evidence on sun exposure and cancer is reviewed further in detail in another chapter.

II. COLORECTAL NEOPLASMS

The relationships between colorectal cancer and dietary or supplementary vitamin D were investigated in cohort studies of men [13,14] and women [15–17] or both sexes [18,19], and in case-control studies [12,20–26]. Some studies did not support an association, but more than half suggested inverse associations [13–16,19,21,23,25,26]. Among the studies supporting an inverse association, the relative risk (RR) for colorectal cancer ranged from 0.88 comparing extreme quintiles of intake of dietary and supplemental vitamin D in the Nurses’ Health Study, which consisted of 501 cases among 89,448 women followed for 12 years [16], to 0.5 comparing extreme quartiles of vitamin D intake (P trend < 0.05) in the Western Electric study, which consisted of 47 cases among 1,924 men followed for 19 years [13]. A Swedish population-based casecontrol study of 352 colon and 217 rectal cancers noted a moderate inverse association (extreme quintiles odds ratio (OR= 0.5, 95% CI 0.3–0.9) between rectal cancer only and dietary vitamin D (supplements not assessed) in multivariate, but not age-adjusted analysis [25]. In the Iowa Women’s Study, women in the highest tertile of intake had a moderately lower risk of rectal cancer (OR = 0.76; P trend = 0.20) [17]. In a recent cohort study based on the large Cancer Prevention Study II Nutrition Cohort [19], total vitamin D intake from diet and supplements was inversely associated with risk of colorectal cancer (RR= 0.71, 95% CI 0.51–0.98). Many of these studies controlled for some confounding factors, but almost all dietary vitamin D comes from fortified milk products, fatty fish, and multivitamin supplements, so conclusively attributing the benefit to vitamin D alone is difficult. For example, much of the vitamin D comes from dairy products, which also are the major sources of another putative protective factor, calcium.

A. Colorectal Cancer Colorectal cancer is one of the most common cancers in developed countries. In the United States, approximately 105,500 cases of colon cancer and 42,000 of rectal cancer cases are expected to occur in 2003 [6]. Colorectal cancer is the third leading cause of cancer deaths in each sex and second for both sexes combined, ultimately afflicting approximately 6% of individuals in their lifetime. Approximately half of individuals diagnosed with this disease will die from it. Factors related to a “westernized” lifestyle and dietary pattern, including obesity, sedentary lifestyle, and “western” diet pattern are believed to be among the most important factors related to the etiology of colorectal cancer [7]. Over the past two decades, two additional, though not entirely independent, hypotheses have been that higher calcium intake and higher circulating vitamin D metabolites are associated with a lower risk of colorectal cancer. The evidence for these hypotheses is reviewed here. The effects of Vitamin D on colon cancer cells are discussed in Chapter 95. 1. SUNLIGHT EXPOSURE

The vitamin D hypothesis for colorectal cancer originated over two decades ago when Garland and Garland [8] observed that the states with the highest mean solar radiation, such as New Mexico and Arizona, had the lowest rates of death from colon cancer, whereas states with the lowest mean solar radiation, including New York, New Hampshire, and Vermont, had the highest mortality rates of colorectal cancer. The authors hypothesized that the higher mortality rates of colorectal cancer in the less sunny regions were caused by inadequate vitamin D status (see Chapter 90). After this report, additional studies that have evaluated the hypothesis of a relation between colorectal cancer incidence [9,10] or mortality [11] and solar radiation [9,10] or ultraviolet light-blocking air pollution [11] have generally supported the vitamin D hypothesis. These studies encompass cancer rates in

2. CASE-CONTROL AND COHORT STUDIES OF VITAMIN D INTAKES

3. STUDIES BASED ON PLASMA MEASURES D

OF VITAMIN

The studies summarized above have examined vitamin D exposure based on surrogates of sunlight exposure or reported dietary intakes. Epidemiologic studies have not considered simultaneously dietary

CHAPTER 91 Epidemiology of Cancer Risk: Vitamin D and Calcium

vitamin D and vitamin D contribution from sunlight exposure as direct exposures, but three studies have measured circulating vitamin D concentrations, which integrate dietary sources and endogenous production of vitamin D. Two of the studies were based from the Washington County, Maryland cohort. The first study from this cohort included 34 cases with a lag of months to eight years between blood donation and colorectal cancer diagnosis, and 67 controls. In this study, an inverse relation between 25(OH)D and colon cancer was observed, although the relation was somewhat attenuated in the highest quintile [27]. The second study based on this cohort included 57 cases with 10 to 17 years between sample collection and time of diagnosis and 114 controls; in this study, neither 25(OH)D nor 1,25(OH)2D was related to colorectal cancer risk [28]. The other study was nested in the Finnish AlphaTocopherol, Beta-carotene Cancer Prevention Study. In that study, no relation was observed between serum 1,25(OH)2D concentration and colorectal cancer among the 91 colon and 55 rectal cancer cases and 290 controls, but an inverse relation was suggested for 25(OH)D level, particularly for colorectal cancer (extreme quartiles: OR= 0.3, 95% CI 0.1–1.1, P trend = 0.04) [29]. Thus, the epidemiologic data are limited, but do suggest an inverse association with 25(OH)D. These results are of interest considering recent evidence of 1 α-hydroxylase activity in the large bowel mucosa [5].

B. Colorectal Adenoma Colorectal adenomas are well-established precursor lesions for the majority of cancers [30]. Thus, these have been used as intermediate endpoints for colorectal neoplasia and to examine risk factors for early stages of carcinogenesis. The relation between sunlight exposure and adenoma risk has not been studied.

1619

adenoma in women using the baseline dietary assessment and eight years of follow-up (extreme quintiles: RR = 0.30, 95% CI 0.13–0.71, P trend = 0.005), but not a subsequent dietary assessment and the latter four years of follow-up [36]. Thus, overall the results on vitamin D intake and risk of adenoma are not consistently supportive of an association. 2. STUDIES BASED ON PLASMA MARKERS VITAMIN D

OF

In the past several years, some studies have examined plasma vitamin D levels and adenoma risk. These studies examined distal colorectal adenomas, and no data are available for proximal adenomas. One study of 473 cases with an adenoma and 507 controls with no adenoma at sigmoidoscopy measured only 25(OH)D [33]. The results from this study suggested an inverse, but not statistically significant association with 25(OH)D (OR for high versus low quartile = 0.74, 95% CI 0.51– 1.09). Another study also examined only 25(OH)D with risk of colorectal adenoma among 239 cases and 228 controls among patients who had had sigmoidoscopy [34]. In this study, each 10 ng/mL increase of serum 25(OH)D was associated with a 26% decrease in risk of adenoma (OR= 0.74, 95% CI 0.60–0.92). The lower risk was primarily in the deficiency range. Another study, the only one that was prospective, was nested in the Nurses’ Health Study and examined both 1,25(OH)2D and 25(OH)D in 326 matched case-control pairs [38]. Overall, mean plasma 25(OH)D and 1,25(OH)2D were only slightly nonsignificantly lower in cases than controls. However, women whose plasma 1,25(OH)2D was below 26.0 pg/mL, a level typically considered to be below normal, were at increased risk of distal colorectal adenoma (OR = 1.58, 95% CI 1.03–2.40). Plasma 25(OH)D displayed a U-shaped pattern, with risk lowest in the middle two quartiles. The data are sparse, but are consistent with an inverse association with 25(OH)D and possibly with 1,25(OH)2D.

1. DIET-BASED STUDIES OF VITAMIN D

Six case-control studies [24,31–35] and two cohort studies (published in one report) [36], and one study of recurrent adenomas [37] have examined the relation between vitamin D and the risk of colorectal adenoma. Of these studies, only two suggested an inverse association between vitamin D intake and adenoma risk [24,36]. Furthermore, findings were not consistent across gender or by location within the colon. In one of the case-control studies, which consisted of 154 small and 208 large adenoma cases and 426 polyp-free controls, an inverse association between vitamin D intake was seen only for small adenoma in women (extreme quintiles: OR =0.4, P trend = 0.04) [24]. In the cohort studies, an inverse relation was seen only for rectal

C. Polymorphisms in the Vitamin D Receptor Gene The effects of 1,25(OH)2D are mediated by its binding to the vitamin D receptor (VDR), which transactivates transcription of target genes [39]. A number of polymorphisms in the VDR have been identified (see Chapter 68). These include: a poly-A microsatellite in the 3′-untranslated region, and the following restriction length polymorphisms: BsmI in intron 8 (denoted b), ApaI in intron 8 (denoted a), and TaqI in exon 9, which results in a base, but not amino acid change (C352T and denoted t). All of these polymorphisms are in strong linkage disequilibrium. None of these polymorphisms

1620 affects the amino acid coding sequence of the vitamin D receptor, although the BAt haplotype has been associated with greater vitamin D receptor transcriptional activity or enhanced mRNA stability in artificial gene constructs than the baT haplotype. In addition, individuals possessing the BB genotype have higher circulating 1,25(OH)2D levels than those with the Bb or bb genotypes [40,41], although the mechanism is unknown. Another polymorphism in the VDR has been identified in exon 2 by using the FokI enzyme. The presence of the FokI site (f allele) results in a VDR that is 427 residues long, whereas the absence of this site (F allele) results in a shorter VDR of 424 residues [42]. The F allele has been associated with higher bone mineral density [43] and lower risk of bone fractures [44]. Thus, the activity of the VDR related to the F allele is believed to be greater than that related to the f allele. One study has examined VDR variants in relation to colon cancer risk and three in relation to adenoma risk. In a case-control study of 250 population-based cases of colon cancer and 364 controls, the BAt haplotype (i.e. BBAAtt vs. all others) was related to a lower risk of colon cancer (OR = 0.5, 95% CI 0.3–0.9) [45]. As expected because of their being in linkage disequilibrium, each of the individual genotypes (B,A,t) was similarly related to lower risk. The FokI variant was not related to risk (OR =1.0, 95% CI 0.6–1.8 for ff versus FF). The adenoma study described above [34] that found an inverse relation with 25(OH)D did not find an association with the FokI variant, although risk was nonsignificantly lower for the FF variant (OR = 0.75, 95% CI 0.36–1.58 for FF versus ff). In contrast, another study of 373 colorectal adenoma cases and 394 controls found that compared to the FF genotype, the OR was 0.79 (95% CI 0.44–1.41) for the Ff genotype and 0.32 (95% CI 0.11–0.91) for the ff genotype [46]. In a clinic-based case-control study of 393 and 406 controls, compared with the bb genotype, neither the Bb (OR = 0.86, 95% CI 0.63–1.19) nor the BB genotype (OR = 0.77, 95% CI 0.50–1.18) was appreciably associated with adenoma risk [35]. However, as in the only study on colon cancer described above [45], the BB genotype was in the direction of lower risk.

D. Discussion:Vitamin D and Colorectal Neoplasia The potential influence of vitamin D in relation to colorectal cancer risk has been examined in a variety of ways. Studies based on sunlight exposure first led to the vitamin D hypothesis because individuals living in sunnier regions tend to have a lower risk of colorectal

EDWARD GIOVANNUCCI AND ELIZABETH A. PLATZ

cancer mortality [8]. Although provocative, interpretation of ecologic studies as indicative of a causal relation between colorectal cancer and vitamin D is limited because excluding potentially confounding factors is difficult due to lack of individual-specific exposure information. For example, migration from one region of the country to another prior to time of cancer diagnosis or death, time spent outdoors by season of the year, shading by buildings, use of sunscreen, amount of skin covered by clothing, and skin pigmentation all influence photosynthesis of vitamin D [47] (see Chapter 3). Ecologic studies cannot control for potentially confounding factors; for example, outdoor physical activity, a protective factor for colorectal cancer, or regional variations in red meat and alcohol consumption, which are risk factors for colorectal cancer [7]. An approach that has not been utilized much is the examination of region or sunlight exposure in case-control or cohort studies, which in addition to measuring region, sunlight exposure, and mediating factors such as skin color, could also adjust for potential confounders such as dietary practices that may differ across regions. Finally, indirect measures of exposures at the population level may sometimes give false impressions. For example, in one study that sampled elderly individuals throughout Europe, residents in Northern Europe had circulating 25(OH)D concentrations twice as high as similarly aged individuals living in Southern Europe [48]. This apparent paradox may be explained by the higher milk intake in the northern regions. Thus, the data based on sunlight exposure are interesting and suggestive, but studies focusing on sunlight exposure need to be enhanced with more rigorous study designs. A second line of evidence has been based on studies that have examined dietary intake of vitamin D in relation to risk of colorectal cancer or adenomas. The results based on cancer risk have been mixed and perhaps weakly supportive, whereas those on adenomas have been less supportive. There are many important considerations for these studies. First, studies based solely on dietary intakes, even assuming that vitamin D intake is measured adequately, would still substantially misclassify overall vitamin D status because of lack of information of sunlight exposure. Second, some degree of measurement error is inherent in the measure of vitamin D intake, and the degree of measurement error is not typically quantified in studies. Third, most studies measure intake at one point, and the presumed time relation between vitamin D status and risk of colorectal neoplasia is not well established. Fourth, in most populations studied, high intake of vitamin D represents a combination of high intake of (fortified) dairy products, fatty fish, and multivitamin supplements.

CHAPTER 91 Epidemiology of Cancer Risk: Vitamin D and Calcium

Because other components of these have been hypothesized to reduce risk of colorectal cancer: calcium (dairy products), omega-3 fatty acids (fish), and folic acid (multivitamin supplements), even if an inverse association between vitamin D intake and colorectal cancer risk is observed, the possibility for confounding by other factors is difficult to exclude. Although the dietary studies of vitamin D are moderately supportive at best, many factors could attenuate true associations; thus, the data do not exclude an important effect of dietary vitamin D. Perhaps the strongest approach to evaluate the vitamin D hypothesis is to examine plasma levels of vitamin D. However, in total, the relevant studies are comprised of just over 200 total cases of colorectal cancer. Recently, more data have been available for adenomas. The use of a biomarker has some theoretical advantages. Presumably, a biomarker can integrate the influence of sunlight exposure, dietary intake, and constitutional factors. However, a potential limitation is that most studies rely on a single measurement of 1,25(OH)2D or 25(OH)D. Substantial intra-individual variability could exist; for example, simply the season the blood sample is taken could strongly influence an individual’s vitamin D level. While season could be matched or adjusted for in analyses, seasonal variation will add heterogeneity; for example, a measurement in the winter will not reflect how high vitamin D levels may rise in the summer for an individual. Another limitation of available studies is that the vitamin D–binding protein has not been studied; variation in this binding protein could be an important determinant of bioavailable vitamin D. To date, the evidence is much too sparse and conflicting to make a definitive statement for either circulating 25(OH)D or 1,25(OH)2D and colorectal neoplasia. With the identification of VDR genotypes, another approach to study the relation between vitamin D and risk of colorectal cancer has emerged. This tool has been used in relatively few studies, but may prove fruitful if several conditions are met. First, the functionality of the known and future identified polymorphisms must be established. Second, studies must be well designed and sufficiently large to examine moderate-sized effects. Third, concurrent measures of vitamin D status ideally should be available. Early and limited results are consistent with a possibly lower risk of colorectal neoplasia associated with the BAt haplotype, but more study is needed. Only a few studies have considered the FokI polymorphism and these are conflicting. In summary, the vitamin D hypothesis for colorectal cancer remains viable, and some epidemiologic data are intriguing. Although the data in each area of study

1621

are not compelling on their own, it is noteworthy that some data regarding sunlight exposure, dietary vitamin D, plasma vitamin D, and VDR polymorphisms are supportive of the vitamin D hypothesis. However, the full range of available epidemiologic tools has been hardly utilized. Future large studies that incorporate data on dietary and supplement intake, sunlight exposure, plasma levels of vitamin D metabolites, and VDR polymorphisms are likely to provide a much stronger test of this hypothesis.

E. Calcium Although calcium interacts with the vitamin D axis, the notion of a role of calcium against colorectal cancer originated independently of this relationship with vitamin D [49]. The major hypothesis has been that calcium reduces colorectal cancer risk by binding secondary bile acids and ionized fatty acids to form insoluble soaps in the large bowel lumen [50,51]. If unbound, these compounds are purported to irritate the colorectal mucosa, causing proliferation and repair processes that could be carcinogenic. The role of the vitamin D receptor in this respect is discussed in Chapter 53. Alternatively, calcium was proposed to directly alter colorectal mucosal activity [52,53]. This section deals with the epidemiologic studies of calcium and colorectal neoplasia. 1. ECOLOGIC STUDIES

Ecologic studies have suggested an inverse relation between intakes of calcium [54,55], milk, and other dairy products, the primary dietary calcium source [56,57], and death from colon or rectal cancer. However, the selection of specific populations for the comparisons may have biased the estimate. In contrast to these studies, when a wider array of populations are chosen, a slight positive association between per capita milk and milk product consumption and colon cancer mortality emerges [58]. One exception to this pattern is Finland, where consumption of calcium is high but rate of colon cancer is relatively low. Ecologic studies have important limitations for this association because milk and dairy consumption tend to be high in westernized or economically developed countries, which may have a higher risk for other reasons, including obesity, sedentary lifestyle, and other dietary factors [7]. It is plausible that despite the lack of an inverse correlation across countries (or even a positive association), calcium or milk intake at the individual level could be protective among individuals in a high-risk population. In fact, in one analysis, the correlation between per capita milk and milk product intakes and age-adjusted

1622 colon cancer mortality was 0.40, but after adjustment for per capita animal fat, the correlation reversed to –0.30 [56]. Thus, overall, because of the limitations of the ecologic studies and the alternative interpretations of the results of these studies, these studies provide neither strong evidence for or against a role of calcium in colon cancer risk. 2. CASE-CONTROL AND COHORT STUDIES OF COLORECTAL CANCERS

The association between calcium intake and risk of colorectal cancer has been studied in a number of casecontrol and cohort studies over the past two decades. Among the case-control studies [12,21–25,59–74], the majority [12,22,23,60,61,64,65,68–71,73–75] suggested a protective effect of calcium on colorectal cancer risk, but only in about half [12,22,23,61,64,68–70,73,74] were the findings statistically significant overall, or at least in one subsite or gender. Two of these studies found statistically significant inverse associations between calcium intake and colon [70] and colorectal [61] cancer only in women. In a large case-control study of Chinese men and women and North American Chinese, an inverse relation for calcium was noted for the rectum only in the North Americans [68], while in a case-control study in Italy, an inverse relation was observed for the colon, but not the rectum [23]. A recently published meta-analysis [76] concluded that the results of case-control studies are heterogeneous and on the whole do not provide evidence of an association between total intake of dairy products, milk, cheese, or yogurt and colorectal cancer risk. Although the meta-analysis did not examine explicitly calcium intake, the data are not strongly supportive of the calcium hypothesis because dairy products are the major source of calcium. However, of note, substantial heterogeneity exists, and a good number of the individual studies are consistent with a moderate protective influence. The data from cohort studies have been more supportive of an association [13–20,75,77–82]. Of these, most [13–15,18,19,75,77,79–82] suggested an inverse association between calcium intakes and colorectal cancer risk, although the inverse association was statistically significant in only some of these studies [13,19,77,81,82]. The meta-analysis that did not support a role for dairy products in case-control studies did show a statistically significant inverse association for total dairy products (OR = 0.62, 95% CI 0.52–0.74, and for milk (OR = 0.80, 95% CI 0.68–0.95) in the cohort studies [76]. Notably, no heterogeneity of the risk estimates in the cohort studies was evident. The meta-analysis did not include four recent supportive studies for calcium [18,19,81,82], which warrant further discussion. Two of the studies were

EDWARD GIOVANNUCCI AND ELIZABETH A. PLATZ

conducted in Finland [18,81] and two in the U.S. [19,82], including a cohort study combining two populations in the United States (thus, in essence, these represent five distinct study populations) [82]. The study by Jarvinen et al. was relatively small, with only 38 cases in the colon and 34 in the rectum. A suggestive inverse association was observed for milk and calcium only for colon cancer but not for rectal cancer. In the study by Pietinen et al., also in Finland, an inverse association was observed for calcium; risk was 30–40% lower in quartiles 2–4, compared with the lowest quartile, which had a median calcium intake of 856 mg/day. A similar pattern was observed for milk. In the study by Wu et al. [82], an inverse association was observed only for distal colon cancer both for men of the Health Professionals Follow-Up Study and women of the Nurses’ Health Study. Interestingly, men with low intakes of total calcium were at higher risk, but the incremental benefit of additional calcium beyond approximately 700 mg/day appeared to be minimal, suggestive of a threshold effect. In the study by McCullough et al. [19], risk also seemed to level off around the range of 1000 mg/day or so, and no additional benefit was observed for intakes exceeding 1200 mg/day. In both U.S. studies, an independent beneficial effect of dietary and supplemental calcium was observed. In the study by Wu et al. [82], supplemental calcium was related to lower risk only if dietary calcium was relatively low, supporting both the existence of a threshold effect and a true effect of calcium rather than another factor in dairy products. 3. STUDIES OF COLORECTAL ADENOMA

Some studies have examined calcium intake in relation to adenoma risk. The general design of these studies was to examine the prevalence of adenoma discovered at endoscopy among individuals having a colonoscopy or sigmoidoscopy for diverse reasons. Controls were individuals who did not have an adenoma at endoscopy. Most studies have not observed a relation between calcium intake and adenoma risk [24,31–36,83]. In some studies, the null relation did not vary by source of calcium [36], site within large bowel [36], polyp size [24,36], or gender [24,36]. However, one study of men and women undergoing colonoscopy in North Carolina and consisting of 236 cases of adenomas (27 of which were adenocarcinomas) and 409 controls suggested that calcium intake from diet and supplements was inversely related to adenoma among men, although the trend was not statistically significant (extreme quartiles multivariate OR=0.44, CI 0.15–1.24); among women no clear relation was evident [84]. In the study by Peters et al. a borderline significant inverse association was observed [34].

CHAPTER 91 Epidemiology of Cancer Risk: Vitamin D and Calcium

Two observational studies examined risk of recurrent adenomas in individuals within randomized trials examining other factors [37,85]. In the study by Martinez et al. [37], dietary calcium intake of >1068 versus <698 mg/day was associated with an OR of 0.56 (95% CI 0.39–0.80, P trend = 0.007), and for total calcium >1279 versus <778, the OR was 0.62 (95% CI 0.42–0.90, P trend = 0.005). The study by Hyman et al. [85] showed only a modest inverse association (OR = 0.72, 95% CI 0.43–1.22) for high versus low quintiles of intake, but found a strong trend for deceasing number of adenomas (P trend = 0.005) across increasing calcium intake. The generally supportive results for recurrent adenomas in observational studies are interesting in light of the results from intervention studies of recurrent adenomas, which are discussed next.

F. Human Intervention Studies of Calcium Intake and Colorectal Adenoma The hypothesis that calcium reduces risk of colorectal neoplasia was tested in randomized intervention trials. The strongest support of the hypothesis was from an intervention trial of calcium supplementation (1200 mg of elemental calcium vs. placebo) among 913 participants who previously had colorectal adenomas and were then followed for new occurrences of adenomas (sometimes called recurrent adenomas) [86]. In that study, the investigators found a moderate but statistically significant reduction in risk of adenoma recurrence: the recurrence rate was 31% in the calcium group and 38% (RR = 0.76, 95% CI 0.60–0.96) in the placebo group. Similar results were observed in the European Calcium Fibre Polyp Prevention trial, a smaller study conducted in Europe around the same time [87]. In that study, the RR for recurrent adenomas was 0.66 (95% CI 0.38–1.17), although the lower recurrence rate among the calcium group was not statistically significant (P = 0.16) in this relatively small study. Together, these data support that higher calcium intake may reduce the rate of new adenomas by about 25 to 35%. The results are consistent with those from the two observational studies of recurrent adenomas. Most critically, these studies provide support of the calcium hypothesis from randomized data, which are not prone to biases that may occur in observational studies.

G. Discussion 1. INTERPRETATION OF EVIDENCE

The interpretation of the human studies that examined calcium intake in relation to colorectal cancer and

1623

risk is quite complex. The apparently modest association (approximately 15–30% reduction in risk) strains the detection capacity of epidemiologic methods. In fact, case-control studies of cancer and studies of prevalent adenomas, with a few exceptions, tend to be null. The prospective data for cancer as well as for adenoma recurrence and the randomized data for adenoma recurrence tend to support a modest inverse association. The supportive studies are based on prospectively collected data, which tend to be less prone to methodologic biases possible in case-control and prevalent adenoma data, and on randomized data, which is less susceptible to confounding than are observational data. Also, indirectly supportive of a role of calcium are the results of the meta-analysis that showed an inverse association for total dairy products (OR = 0.62, 95% CI 0.52–0.74) and for milk (OR = 0.80, 95% CI 0.68–0.95) in the prospective studies [76]. Dairy products and milk are the primary source of calcium, although other components of dairy products (including vitamin D) could contribute to the benefit. However, the most recent largest prospective epidemiologic studies, which included three of the large prospective studies of diet in the U.S. (Nurses’ Health Study, Health Professionals Follow-Up Study, American Cancer Society Cohort) found a benefit with supplemental calcium [19,82]. The randomized data on recurrent adenomas are crucial because they provide evidence that calcium directly influences a colorectal neoplastic endpoint in a setting in which confounding is essentially precluded. The epidemiologic data using the cancer endpoint are complementary because they indicate a similar effect on cancer as observed for recurrent adenomas. Alone, the data on recurrent adenomas would be less compelling because only a small fraction of adenomas (around 10%) progress to cancer, and not all colorectal cancers arise from adenomas. Calcium has a relatively small influence (about a 15–30%) reduction in adenoma risk. If, for example, this association was preferential for adenomas that are less likely to progress to cancer (e.g. small tubular adenomas), then the quantitative impact on cancer could be much less than the results observed in the randomized trials. However, the associations for the cancers suggest a similar, if not stronger, magnitude of association. Thus, the results for recurrent adenomas on the whole are probably generalizable to cancers. 2. MECHANISM

The most widely cited hypothesis regarding a benefit of calcium is that calcium might reduce colon cancer risk by binding secondary bile acids and ionized fatty acids to form insoluble soaps in the colonic lumen,

1624 thereby reducing the proliferative stimulus of these compounds on the colonic mucosa [50]. While appealing, a weakness of this hypothesis is the lack of definitive evidence that secondary bile acids and ionized fatty acids underlie the higher risk of colon cancer in Western populations. Although fat is a promoter in animal models of carcinogenesis [88], and production of bile acids and fatty acids is greater with consumption of high fat diets, epidemiologic studies have not found that calcium is more beneficial among people who consume high fat diets [14,16,19,25,59,77,82]. In addition, animal studies find that the protective effect of calcium is not completely dependent on fat intake [49]. A meta-analysis of cohort and case-control studies on calcium and colorectal cancer risk published between 1980 and 1994 found large heterogeneity in estimates not accountable by differences in endpoint, subsite, gender, fat intake, calcium source, and dose, or study design [89]. In recent years, stronger evidence has converged in supporting the role of energy imbalance, obesity, and physical inactivity, rather than fat intake, as being the major “Western” factors related to colon cancer risk [7]. These factors are believed to ultimately increase cell proliferation in the large bowel mediated through insulin and insulin-like growth factors [90]. Interestingly, Ma et al. [91] found that calcium intake was related to lower risk of colorectal cancer primarily in individuals with high circulating IGF-1 concentrations. The insulin/ IGF-1 hypothesis may coincide with the hypothesis that calcium exerts a direct influence on proliferation. The studies that have examined the influence of calcium intake on proliferation in colorectal crypts have been inconsistent, with some showing a reduction [52,53,92–94], and others not [95–100]. Of those studies that were randomized and placebo-controlled [94–96, 98–100], only one showed a benefit of calcium supplementation [94]. One of the best conducted studies showed no change in the proportion of proliferating cells in the crypt overall (i.e., labeling index or LI) in rectal biopsies from patients with sporadic adenoma supplemented daily with 1 or 2 g for six months, but it did demonstrate a statistically significant reduction in the proportion of proliferating cells in the top 40% of the crypt [98]. Bostick et al. suggest that their findings are most compatible with the hypothesis that calcium directly affects progression through the cell cycle, rather than an indirect effect through binding of bile acids and fatty acids. More recent evidence, reviewed by Lamprecht and Lipkin [49], indicates that the protective effect of calcium involves interactions with vitamin D acting on a series of signaling events at various tiers of colonic cell organization. In a recent study, fasting levels of

EDWARD GIOVANNUCCI AND ELIZABETH A. PLATZ

25(OH)D but not 1,25(OH)2D, were inversely associated with whole crypt labeling index and the size of the proliferative compartment [101]. Calcium supplementation influenced this relationship between 25(OH)D and proliferative parameters. These data suggest that there may be a local influence of 25(OH)D through conversion of 25(OH)D to 1,25(OH)2D [5]. 3. IMPLICATIONS AND FUTURE RESEARCH

While the apparent protective effect of calcium on colorectal cancer risk appears relatively modest in magnitude, this effect could have important public health ramifications because of the high prevalence of colorectal cancer. For example, approximately 150,000 cases of colorectal cancer are diagnosed annually in the U.S. If, for example, calcium deficiency contributed to even only 10% of these cases, then potentially about 15,000 cancer cases and 7,500 deaths could be prevented annually through the relatively simple dietary intervention of increasing calcium intake. However, resolving the dose-response is critical. Some of the recent large prospective studies find a threshold effect, with most of the benefit arising through avoiding low intakes (e.g., below 500–700 mg/day). Benefits appear to plateau around 700–1000 mg/day, and whether any benefit occurs beyond 1,000 mg/day or so of calcium remains unclear. The randomized trials did not address this because they examined only one dose. Another important issue to resolve is whether vitamin D is a modifying factor of the apparent benefit of calcium. As summarized above, the data on vitamin D and colo-rectal cancer are suggestive, but not definitive. Finally, whether some highly susceptible individuals could be identified, such as those with high IGF-1 levels or a genetic marker, would be important to establish.

III. PROSTATE CANCER A. International Patterns of Mortality and Incidence Prostate cancer mortality rates around the world vary more than thirtyfold. The lowest rates are observed in the Far East and on the Indian subcontinent, and the highest rates observed in Western Europe, Australia, and North America [102]. Adjusting the rates to the World Health Organization world standard population, the mortality rate in the year 2000 for prostate cancer was approximately 1 per 100,000 men annually in China compared to 17.9 per 100,000 in the U.S. [103]. The contrast in rates is even higher for prostate incidence: in China the rate is 2.9 per 100,000 men

CHAPTER 91 Epidemiology of Cancer Risk: Vitamin D and Calcium

compared with 107.8 and 185.4 per 100,000 men in white and black Americans, respectively [103]. A sharp gradient in the prostate cancer mortality rate is observed between Northern Europe (e.g., Sweden, Norway, and Denmark), where the rates are more than 23 per 100,000 men annually, and Southern Europe (e.g., Greece), where the rates are half that [102]. Some of this wide disparity in prostate cancer incidence rates among countries is likely due to differences in medical practice leading to differential rates of detection of subclinical tumors. The frequency of these latent tumors does not vary substantially across populations [104,105], indicating that factors that cause the growth and progression of prostate tumors most likely account for the marked variation in prostate cancer incidence. Widespread screening using prostatic-specific antigen (PSA) has greatly increased the numbers of preclinical prostate cancers diagnosed. Many dietary hypotheses have been proposed to explain this variation in rates or progression of prostate cancer, but none have been definitively established. The role of vitamin D in regulating prostate cell growth is discussed in Chapter 94.

B. Vitamin D 1. SUNLIGHT EXPOSURE

Schwartz and Hulka have hypothesized that vitamin D protects against prostate cancer based primarily on correlations between regional UV radiation and prostate cancer mortality rates in the U.S. [106]. Hanchette and Schwartz [107] examined the geographic distributions of UV radiation and prostate cancer mortality in 3073 counties in the U.S., and found a prostate cancer mortality gradient, with rates higher in the North and lower in the South. More recently, Grant [108] examined UV-B data for July 1992 and cancer rates in the U.S. between 1970 and 1994 and confirmed these findings for prostate cancer (as well as many other cancer sites). Luscombe et al. [109] conducted a case-control study of 210 cases and 155 controls (with benign prostatic hyperplasia) in the UK and examined whether indicators of UV exposure at the individual level were associated with risk of prostate cancer. They found that childhood sunburn frequency (OR = 0.18, 95% CI 0.08–0.38), regular foreign holidays (OR = 0.41, 95% CI 0.25–0.68), sunbathing score (OR = 0.83, 95% CI 0.76– 0.89) and low exposure to UV radiation (OR = 3.03, 95% CI 1.59–5.78) were associated with the risk of prostate cancer. In addition, cases with low UV exposure developed prostate cancer at a younger median age (67.7 years) than cases with a higher exposure (72.1 years). These results were confirmed in another population by

1625

the same group [110]. In addition, these investigators also found that the risk may be modified by skin pigmentation and genes that influence it [111]. These findings provide for the first time some confirmation of the UV hypothesis at the individual level. These provocative results need to be confirmed in prospective analyses to preclude the possibility of recall bias, although the relatively strong associations observed would tend to argue against such bias. The sunlight hypothesis is discussed further in Chapter 90. 2. DIETARY VITAMIN D

Surprisingly little data has evaluated whether dietary or supplemental vitamin D is related to the risk of prostate cancer, and the limited evidence is not supportive. In the Health Professionals Follow-Up Study, a large cohort of U.S. men [112], no relation was observed between dietary and supplemental vitamin D with intakes ranging from <150 to >800 IU/day and total or advanced prostate cancer. This analysis was based on 1369 cases of prostate cancer. A case-control study of prostate cancer conducted in Sweden also did not support an effect of dietary vitamin D [113]. In addition, a cohort study of Finnish male smokers did not find any association between dietary vitamin D and risk of prostate cancer [114]. A recent case-control study of 605 prostate cancer cases diagnosed at younger ages cancer found no association with dietary vitamin D [115]. As discussed above, the study of dietary and supplemental vitamin D is prone to many limitations. In any case, currently no epidemiologic data support the hypothesis that an increase in dietary vitamin D would lower risk of prostate cancer. 3. CIRCULATING VITAMIN D

Six case-control studies nested in prospective cohort studies have examined circulating vitamin D metabolites in relation to risk of prostate cancer [116–121]. Corder et al. [116] measured 1,25(OH)2D and 25(OH)D levels in blood collected prior to diagnosis in 181 prostate cancer cases and 181 controls nested in the Kaiser-Permanente cohort. Risk of prostate cancer, especially in older men, was significantly lower among those with higher 1,25(OH)2D concentrations. In men over 57 years, compared to the lowest 1,25(OH)2D quartile, the prostate cancer odds ratios were 0.66, 0.53, and 0.37 in the second, third, and fourth quartiles, respectively. Men with the lowest risk of prostate cancer, and particularly of tumors of advanced stage and high grade, were those with high circulating 1,25(OH)2D and low 25(OH)D. A much weaker (nonsignificant) inverse relation between overall prostate cancer and prediagnostic plasma, 1,25(OH)2D concentration was observed among

1626 232 cases and 414 age-matched controls nested in the Physicians’ Health Study [118]. This study was extended to 372 cases and 591 controls with subsequent follow-up [41]. However, some findings were observed in subgroup analyses that are noteworthy because they were similar to those found in the Corder et al. study [116]. When considering jointly 1,25(OH)2D and 25(OH)D levels, reduced risks of prostate cancer were noted for high 1,25(OH)2D and low 25(OH)D levels in the Corder et al. study [116] and in the Gann et al. study [118], particularly for older men and for more aggressive prostate cancer, but the joint association was statistically significant only in the Corder et al. study [116]. In the Physicians’ Health Study, results for vitamin D metabolites also indicated some potentially interesting interactions with VDR polymorphisms (see below). Three studies were essentially nonsupportive of the vitamin D hypothesis. A study by Braun et al. which also evaluated 1,25(OH)2D concentration in blood obtained prior to cancer diagnosis, did not show an inverse relation between prostate cancer and serum 1,25(OH)2D level [117]. This study consisted of only 61 cases, only 34 were nonincidental findings, and only 19 of these were diagnosed in men older than 57 years. Thus, this study had very low power to examine risk of the more aggressive prostate cancers in the older age groups. A study in Hawaii by Nomura et al. [119] generally was nonsupportive of the vitamin D hypothesis. An additional study was recently conducted for 460 prostate cancer cases in the Health Professionals Follow-up Study who were diagnosed through 1998 after providing a blood specimen in 1993/95 [121]. Plasma 1,25(OH)2D and 25(OH)D concentrations were compared to an equal number of age-matched men. In this study, no association was observed between prostate cancer risk and prediagnostic plasma concentrations of 1,25(OH)2D or 25(OH)D. Only one plasma-based study was conducted outside the U.S. This study [120] did not measure 1,25(OH)2D, but did show a 1.7-times higher risk of prostate cancer (95% CI 1.2−2.5) in Finnish men whose plasma 25(OH)D concentrations were below the median compared to at or above. The association was strongest in men who were younger at study entry (<52 years old: OR = 3.1, 95% CI 1.6−6.1) [120]. All the other studies measured 25(OH)D and none supported an association between this metabolite and prostate cancer risk [116–119,121]. The authors of the Finnish study indicated that more than half of their participants had 25(OH)D levels consistent with clinical vitamin D deficiency [120]; their median of 40 nmol/L (16.0 ng/mL) was near the usual cutpoint for vitamin D deficiency of approximately 15 ng/mL. In the other studies, the

EDWARD GIOVANNUCCI AND ELIZABETH A. PLATZ

percentages of controls that had 25(OH)D lower than 15 ng/mL were ~5% [117], 6.5% [118], 11% [121], and 13.3% [116]. The proportion with vitamin D deficiency was not available in the Nomura et al. study, which was conducted in Hawaii, but it is likely low given that their median 25(OH)D concentration was 41 ng/mL in the cases and 41.6 ng/mL in the controls [119]. Thus, a possible explanation for the inconsistent findings for 25(OH)D and prostate cancer among studies is that in individuals with adequate vitamin D levels there is no added benefit of higher circulating levels, but that an incremental increase in 25(OH)D to sufficiency may reduce risk. An additional factor that has not been studied adequately is the vitamin D–binding protein, which could influence the bioavailability of vitamin D metabolites (see Chapter 8). In the only prospective study that examined this factor, measuring the binding protein did not confer much additional information [118]. One small case-control did suggest those with higher binding protein concentrations may receive less benefit from vitamin D [122]; however, this study was limited by a small sample size, retrospective collection of information and use of convenience controls. In summary, only one of five studies that measured 1,25(OH)2D found a statistically significant inverse association with prostate cancer risk [116], and one of six that measured 25(OH)D found a significant inverse association [120]. Some potentially notable tendencies, but hardly consistent, were that vitamin D is more relevant for the older onset cases, for the more aggressive endpoints, when 1,25(OH)2D is high and 25(OH)D is low, and when 25(OH)D is in the deficiency range. If vitamin D is more important for progression, as indicated by associations with more advanced endpoints in some studies, then more recent studies in which cases are predominantly PSA-detected may miss an important effect on progression.

C. Polymorphisms in the Vitamin D Receptor Gene and 1-α-hydroxylase As described above, a number of polymorphisms in the vitamin D receptor have been identified. The following polymorphisms have been evaluated in relation to prostate cancer: a poly-A microsatellite in the 3′-untranslated region, and the following restriction length polymorphisms: BsmI in intron 8 (denoted b), ApaI in intron 8 (denoted a), and TaqI in exon 9 (denoted t). These polymorphisms are in strong linkage disequilibrium, and none of these polymorphisms affects the amino acid coding sequence of the vitamin D receptor [40]. Also, examined in several studies is the

CHAPTER 91 Epidemiology of Cancer Risk: Vitamin D and Calcium

FokI polymorphism in exon 2. The BAt haplotype has been studied relatively extensively for prostate cancer in U.S. and European populations and in Asian populations (mostly in Japan and one study in China). See Chapter 68 for an extensive discussion of the VDR polymorphisms. For the BAt haplotype, 10 studies have been conducted in the U.S. or Europe. Three of the studies found a statistically significant lower risk associated with the BAt haplotype [123–125]. The largest study, and the only prospective nested case-control study, was the study within the Physicians’ Health Study by Ma et al. [41]. This study found only a weak nonsignificant lower risk associated with the BAt haplotype, but did find a strong inverse association in subgroups of older men, and those with 25(OH)D below the median. Other studies were nonsupportive [126–131], although three of these studies were seriously limited by small sample sizes (<100 cases) [126,127,131]. Four Japanese studies examined the BAt haplotype. Two, similar to some of the studies in Western populations, found a statistically significant reduced risk associated with the BAt haplotype [132,133]. The three null studies were limited by small number of cases [134–136]. In summary of the studies examining the BAt haplotype, five found a statistically significant inverse association of prostate cancer with the BAt haplotype. The largest study did not, but subgroup analyses offered some support that this haplotype is associated with lower risk, particularly for more aggressive cancers [41]. Overall, the relative risks ranged from two- to fivefold. Most of the null studies had severe power limitations. All of the studies, except the study by Ma et al. [41] relied on convenience control groups. Because of the limitations of these studies and the inconsistencies, the conclusions drawn need to be tempered. Nonetheless, the data are suggestive of a lower risk, particularly of aggressive cancers, associated with the BAt haplotype. These results are of interest because the BAt haplotype has been reported to have greater vitamin D receptor transcriptional activity or enhanced mRNA stability in artificial gene constructs than the BAt haplotype, and individuals with the BB genotype may have higher circulating 1,25(OH)2D levels than Bb or bb genotypes [40,41]. Only three studies have examined prostate cancer risk in relation with the FokI genotypes [129,137,138]. The results are inconsistent. In addition, no association has been observed between polymorphisms in the gene encoding 25-hydroxyvitamin D 1-α-hydroxylase, the enzyme that catalyzes the conversion of 25(OH)D to 1,25(OH)2D [139], and prostate cancer risk. Given the limited research, more needs to be done for the FokI

1627

and 1-α-hydroxylase genotypes before even tentative statements can be made.

D. Calcium and Dairy Products Relevant to the vitamin D hypothesis is intake of calcium and dairy products, which are the greatest source of calcium and of vitamin D in populations that fortify milk with vitamin D. Until just recently, few studies had examined the relation between dietary and supplemental calcium intake and risk of prostate cancer, although more studies have examined associations with milk and dairy consumption. The general pattern has been in stark contrast to that observed for colon cancer: studies of calcium, milk, and dairy products generally show that these are positively related to prostate cancer risk. This increased risk of prostate cancer associated with higher milk and dairy consumption was first observed for ecologic studies. Countries with greater per capita milk consumption have higher national prostate cancer mortality rates (r = 0.66). This correlation is greater than for other foods high in animal fat (e.g., meats, r = 0.39) [140], suggesting that dairy products impart an adverse effect beyond that of total dietary fat or saturated fat. Milk (r = 0.75) and cheese (r = 0.69) consumption were correlated with regional mortality rates in Italy, even when controlling for other factors characteristic of each region [141,142]. In a more recently conducted ecologic study, based on mortality rates from 41 countries, the nonfat portion of milk had the highest association with prostate cancer mortality rates of dietary factors examined (r2 = 0.73; P < 0.001) [143]. Although these ecologic studies are consistent with the hypothesis that higher intakes of milk and dairy products are correlated with higher risk of prostate cancer mortality, as noted earlier, these studies are limited by lack of information on individualspecific exposures, outcomes, and confounding factors. Thus, evaluating the relation of sources of calcium in the diet with prostate cancer in case-control and cohort studies is important. In case-control studies, men consuming high levels of milk and other dairy products have been observed to be at an either statistically significant increased risk [115,144–150], or borderline significant (P ≤ 0.1) increased risk of prostate cancer [151–154] in most studies. One study reported an association with saturated fat from dairy products [155]. Several studies, however, have not supported an association [156,157], and no study has suggested a protective effect. Overall, among the case-control studies, nine support a statistically significant association with some component of

1628 milk or dairy products, four found nonsignificant but borderline significant associations, and two found no associations. The case-control studies are limited because total energy intake was not controlled in most of them, and they have some potential methodologic limitations, such as the potential for recall and selection biases. Nonetheless, they provide strong evidence against a benefit of milk or dairy products, and the positive association observed in many settings (including diverse ethnic groups in the U.S., Italy, Canada, Sweden, Uruguay, Greece) is noteworthy because methodologic limitations and confounding is unlikely to be uniform across such diverse settings. Most prospective cohort studies, which tend to be less prone to methodologic limitations than casecontrol studies, also tend to support an association between higher intake of milk or dairy products and risk of prostate cancer [158–162]. However, some have not supported an association [114,163–165]. No studies support a protective influence of dairy products. In the case-control studies reporting an elevated risk, the magnitude of the relative risk comparing high to low consumption of milk is around 2 (ranging up to fivefold) [147]. The associations tend to be weaker for prospective studies, with RRs generally in the range of 1.3 to 1.6 between high and low intakes. The weaker associations in prospective studies might reflect biases that exaggerated the magnitude of the association in case-control studies. Alternatively, more of the prospective studies were conducted more recently when the case mix is heavily skewed toward early-stage, relatively indolent lesions, and some studies had long time lags between the dietary assessment and the diagnosis of cancer. If dietary factors are primarily important for disease progression, then weaker associations would be expected to occur in the cohort studies. Many fewer studies have examined calcium intake in relation to prostate cancer risk. The effect of calcium intake on prostate cancer risk was directly evaluated in a case-control study in Sweden [113] and in a cohort study in the U.S. [112], with both reporting a positive relation. The cohort study, which consisted of over 50,000 health professionals, found substantially greater associations between calcium intake with metastatic and fatal prostate cancer. Supplemental (12% of the men were calcium supplement users) and dietary (mostly dairy) calcium sources were independently associated with increased risk of advanced prostate cancer. The risk of metastatic prostate cancer was 5 times greater among men consuming >2000 compared to <500 mg/day of calcium from the diet and supplements. An association was also observed in the prospective Physicians’ Health Study [161]. Another case-control

EDWARD GIOVANNUCCI AND ELIZABETH A. PLATZ

study found a twofold increased risk of regional/ distant prostate cancer associated with higher calcium intake, but no increased risk with local disease [115]. A recent large prospective study found that very high calcium take (>2000 mg/day) was association with an increased risk of prostate cancer [166]. Interestingly, the association was stronger for men not tested by PSA, indicating an effect on progression. Other studies are nonsupportive [114,162,167,168] or only suggestive of an effect of calcium [153,154]. In summary, from several study designs in a variety of populations, the finding of a positive relation of milk and dairy products with prostate cancer, particularly advanced disease, has been observed frequently though not invariably. Evidence from some recent studies showing a positive association between calcium from supplements and advanced prostate cancer supports an effect of calcium apart from its co-occurrence with fat or other components in dairy products.

E. Discussion The human evidence on vitamin D, calcium, and dairy products in relation to prostate cancer risk is quite complex to interpret. Some evidence is moderately supportive of a benefit of vitamin D. Both ecologic and a few case-control studies suggest that greater sunlight exposure is associated with a reduced risk of prostate cancer. The studies of VDR polymorphisms are somewhat suggestive, though inconclusive, that differences in VDR might influence risk. Although the functionality of these polymorphisms remains in question, the presence of an association with the VDR genotypes points to the importance of vitamin D. The plasma-based studies are generally nonsupportive but are suggestive of a possible protective role of high circulating 1,25(OH)2D, or a risk for deficiency in 25(OH)D in the etiology of advanced prostate cancer. Dietary vitamin D studies are limited but nonsupportive. Vitamin D–binding protein concentration is another factor that may modify risk, but the only prospective study did not support a role for this factor; more studies, however, are clearly needed. An additional complexity in prostate cancer epidemiologic studies is that the recent widespread use of PSA has almost eliminated the presence of advanced disease, at least at initial diagnosis. Additionally, if vitamin D influences progression of the disease rather than early stages, null associations in recent studies where the case mix is dominated by early stage, relatively indolent lesions may not necessarily be considered strong evidence against the vitamin D hypothesis.

CHAPTER 91 Epidemiology of Cancer Risk: Vitamin D and Calcium

In future studies, total prostate cancer as the main outcome is unlikely to be an adequate phenotype to study; better indicators of progression or of aggressive behavior are paramount to identify. In addition, some data, though not all, indicate that some modifiable factors may be more relevant for prostate cancers that have an older age-of-onset. Prostate cancer is a uniquely heterogeneous malignancy, so its study is particularly complex and usually does not allow for simple answers. The frequent finding of a positive association between intake of milk, dairy products, and possibly calcium and risk of prostate cancer has been enigmatic. Milk and dairy products are the major sources of calcium and dietary vitamin D (in the U.S.) so this finding appears paradoxical given the other evidence that vitamin D exposure could inhibit carcinogenesis. An attempt to reconcile these apparently conflicting observations was the postulation that high intakes of dairy products, despite providing vitamin D, may actually suppress 1,25(OH)2D production [112]. This hypothesis was directly evaluated in the Physicians’ Health Study, which had shown an increased risk of prostate cancer associated with higher intakes of skim milk, dairy products, and calcium from dairy products [161]. In that study, men who consumed >600 mg calcium per day from skim milk had a mean plasma 1,25(OH)2D concentration of 30.06 pg/mL compared to men who consumed ≤150 mg/day, who had a mean 1,25(OH)2D concentration of 35.64 pg/mL (P = 0.005). This study was the first to show simultaneously that dairy calcium is associated with a lower 1,25(OH)2D and with an increased risk of prostate cancer. An alternative hypothesis is that the increased risk of prostate cancer often seen with higher dairy product consumption is due to some component of the fat content of dairy products; this is a plausible and not necessarily mutually exclusive hypothesis. However, some studies support an equally strong if not more pronounced association with skim/low fat milk than with whole milk [112,115,143,148,161,165]. Another potentially important recently identified factor is in vitro evidence that human prostate cells can synthesize 1,25(OH)2D from 25(OH)D [4] and that 25(OH)D may inhibit the proliferation of primary prostatic epithelial cells [169]. These findings need to be verified in vivo, but suggest that high circulating 25(OH)D would decrease risk of prostate cancer. Although the studies based on sunlight exposure would support this, this hypothesis is difficult to reconcile with the generally null studies on dietary and supplementary vitamin D, the weak findings for plasma-based studies of 25(OH)D, and the frequent finding of a positive association with dairy products.

1629

IV. CONCLUSION In conclusion, the hypothesis that the vitamin D axis plays an important role in carcinogenesis is viable and an active and exciting area of research. Several lines of human and animal evidence support a role for calcium and vitamin D for colorectal cancer. Simple answers are precluded for prostate cancer because of inconsistencies in studies. Clearly, a better understanding of the complex biology that accounts for interactions among dietary factors, UV light, genetic polymorphisms in the VDR and among the vitamin D metabolites will be required to better make sense of the existing epidemiology and to design new studies. Other factors, such as the influence of vitamin D on the insulin-like growth factor axis will have to also be taken into account. Furthermore, effects are likely to differ among the organ sites, as the recent divergent associations for calcium and dairy products for prostate and colorectal neoplasms show. Future epidemiologic studies must be well-designed to avoid methodologic bias, be sufficiently large, and take into account simultaneously factors such as diet, UV exposure, plasma levels, and genetic polymorphisms. This will require large studies and the pooling of data to get more definitive results. In addition, the complexity of the changing prostate cancer case mix cannot be ignored.

References 1. Hollis BW, Kamerud JW, Selvaag SR, Lorenz JD, Napoli JL 1993 Determination of vitamin D status by radioimmunoassay using I125-labelled tracer. Clin Chem 39:529–533. 2. Bouillon RA, Auwerx JH, Lissens WD, Pelemans WK 1987 Vitamin D status in the elderly: seasonal substrate deficiency causes 1,25-dihydroxycholecalciferol deficiency. Am J Clin Nutr 45:755–763. 3. Fraser D 1980 Regulation of the metabolism of vitamin D. Physiol Rev 60:551–613. 4. Schwartz GG, Whitlatch LW, Chen TC, Lokeshwar BL, Holick MF 1998 Human prostate cells synthesize 1,25dihydroxyvitamin D3 from 25-hydroxyvitamin D3. Cancer Epidemiol Biomarkers Prev 7:391–395. 5. Tangpricha V, Flanagan JN, Whitlatch LW, Tseng CC, Chen TC, Holt PR, Lipkin MS, Holick MF 2001 25-hydroxyvitamin D-1alpha-hydroxylase in normal and malignant colon tissue. Lancet 357:1673–1674. 6. Jemal A, Murray T, Samuels A, Ghafoor A, Ward E, Thun MJ 2003 Cancer Statistics. CA Cancer J Clin Med 53:5–26. 7. Giovannucci E, Willett WC 1994 Dietary factors and risk of colon cancer. Ann Med 26:443–452. 8. Garland CF, Garland FC 1980 Do sunlight and vitamin D reduce the likelihood of colon cancer? Int J Epidemiol 9:227–231. 9. Spitz MR, Paolucci MJ, Newell GR 1988 Occurrence of colorectal cancer: focus on Texas. Cancer Bull 40: 187–191.

1630 10. Emerson JC, Weiss NS 1992 Colorectal cancer and solar radiation. Cancer Causes Control 3:95–99. 11. Gorham ED, Garland CF, Garland FC 1989 Acid haze air pollution and breast and colon cancer mortality in 20 Canadian cities. Can J Public Health 80:96–100. 12. Kampman E, Slattery ML, Caan B, Potter JD 2000 Calcium, vitamin D, sunshine exposures, dairy products, and colon cancer risk (United States). Cancer Causes Control 11:459–466. 13. Garland C, Shekelle RB, Barrett-Conner E, Criqui MH, Rossof AH, Paul O 1985 Dietary vitamin D and calcium and risk of colorectal cancer: A 19-year prospective study in men. Lancet 1:307–309. 14. Kearney J, Giovannucci E, Rimm EB, Ascherio A, Stampfer MJ, Colditz GA, Wing A, Kampman E, Willett WC 1996 Calcium, vitamin D, and dairy foods and the occurrence of colon cancer in men. Am J Epidemiol 143:907–917. 15. Bostick RM, Potter JD, Sellers TA, McKenszie DR, Kushi H, Folsom AR 1993 Relation of calcium, vitamin D, and dairy food intake to incidence of colon cancer in older women. Am J Epidemiol 137:1302–17. 16. Martínez ME, Giovannucci EL, Colditz GA, Stampfer MJ, Hunter DJ, Speizer FE, Wing A, Willett WC 1996 Calcium, vitamin D, and the occurrence of colorectal cancer among women. J Natl Cancer Inst 88:1375–1382. 17. Zheng W, Anderson KE, Kushi LH, Sellers TA, Greenstein J, Hong CP, Cerhan JR, Bostick RM, Folson 1998 A prospective cohort study of intake of calcium, vitamin D, and other micronutrients in relation to incidence of rectal cancer among postmenopausal women. Cancer Epidemiol Biomarkers Prev 7:221–225. 18. Jarvinen R, Knekt P, Hakulinen T, Aromaa A 2001 Prospective study on milk products, calcium, and cancers of the colon and rectum. Eur J Clin Nutr 55:1000–1007. 19. McCullough ML, Robertson AS, Rodriguez C, Jacobs EJ, Chao A, Carolyn J, Calle EE, Willett WC, Thun MJ 2003 Calcium, vitamin D, dairy products, and risk of colorectal cancer in the cancer prevention study II nutrition cohort (United States). Cancer Causes Control 14:1–12. 20. Heilbrun LK, Nomura A, Hankin JH, Stemmermann GN 1985 Dietary vitamin D and calcium and risk of colorectal cancer (letter). Lancet 1:925. 21. Benito E, Stiggelbout A, Bosch FX, Obrador A, Kaldor J, Mulet M, Munoz N 1991 Nutritional factors in colorectal cancer risk: a case-control study in Majorca. Int J Cancer 49:161–167. 22. Peters RK, Pike MC, Garabrandt D, Mack TM 1992 Diet and colon cancer in Los Angeles County, California. Cancer Causes Control 3:457–473. 23. Ferraroni M, La Vecchia C, D’Avanzo B, Negri E, Franceschi S, Decarli A 1994 Selected micronutrient intake and the risk of colorectal cancer. Br J Cancer 70:1150–1155. 24. Boutron MC, Faivre J, Marteau P, Couillault C, Senesse P, Quipourt V 1996 Calcium, phosphorus, vitamin D, dairy products, and colorectal carcinogenesis: a French casecontrol study. Br J Cancer 74:145–151. 25. Pritchard RS,ABJ, Gerhardsson de Verdier M 1996 Dietary calcium, vitamin D, and the risk of colorectal cancer in Stockholm, Sweden. Cancer Epidemiol Biomarkers Prev 5:897–900. 26. Marcus PM, Newcomb PA 1998 The association of calcium and vitamin D, and colon and rectal cancer in Wisconsin women. Int J Epidemiology 27:788–793. 27. Garland CF, Comstock GW, Garland FC, Helsing KJ, Shaw EK, Gorham ED 1989 Serum 25-hydroxyvitamin D

EDWARD GIOVANNUCCI AND ELIZABETH A. PLATZ

28. 29.

30. 31. 32.

33.

34.

35.

36.

37.

38.

39. 40. 41.

42.

43.

44.

and colon cancer: eight-year prospective study. Lancet 2: 1176–1178. Braun MM, Helzlsouer KJ, Hollis BW, Comstock GW 1995 Colon cancer and serum vitamin D metabolite levels 10–17 years prior to diagnosis. Am J Epidemiol 142:608–611. Tangrea J, Helzlsouer K, Pietinen P, Taylor P, Hollis B, Virtamo J, Albanes D 1997 Serum levels of vitamin D metabolites and the subsequent risk of colon and rectal cancer in Finnish men. Cancer Causes Control 8:615–625. Lev R 1990 “Adenomatous Polyps of the Colon,” SpringerVerlag, New York. Benito E, Cabeza E, Moreno V, Obrador A, Bosch FX 1993 Diet and colorectal adenomas: a case-control study in Majorca. Int J Cancer 55:213–219. Little J, Logan RFA, Hawtin PG, Hardcastle JD, Turner ID 1993 Colorectal adenomas and diet: a case-control study of subjects participating in the Nottingham fecal occult blood screening programme. Br J Cancer 67:177–184. Levine AJ, Harper JM, Ervin CM, Chen YH, Harmon E, Xue S, Lee ER, Frankel HD, Haile RW 2001 Serum 25-hydroxyvitamin D, dietary calcium intake, and distal colorectal adenoma risk. Nutr Cancer 39:35–41. Peters U, McGlynn KA, Chatterjee N, Gunter E, GarciaClosas M, Rothman N, Sinha R 2001 Vitamin D, calcium, and vitamin D receptor polymorphism in colorectal adenomas. Cancer Epidemiol Biomarkers Prev 10:1267–1274. Kim HS, Newcomb PA, Ulrich CM, Keener CL, Bigler J, Farin FM, Bostick RM, Potter JD 2001 Vitamin D receptor polymorphism and the risk of colorectal adenomas. Cancer Epidemiol Biomarkers Prev 10:869–874. Kampman E, Giovannucci E, van’t Veer P, Rimm E, Stampfer MJ, Colditz GA, Kok FJ, Willett WC 1994 Calcium, vitamin D, dairy foods, and the occurrence of colorectal adenomas among men and women in two prospective studies. Am J Epidemiol 139:16–29. Martinez ME, Marshall JR, Sampliner R, Wilkinson J, Alberts DS 2002 Calcium, vitamin D, and risk of adenoma recurrence (United States). Cancer Causes Control 13:213–220. Platz EA, Hankinson SE, Hollis BW, Colditz GA, Hunter DJ, Speizer FE, Giovannucci E 2000 Plasma 1,25-dihydroxyand 25-hydroxyvitamin D and adenomatous polyps of the distal colorectum. Cancer Epidemiol Biomarkers Prev 9:1059–1065. Reichel H, Koeffler HP, Norman AW 1989 The role of the vitamin D endocrine system in health and disease. N Engl J Med 320:980–991. Morrison AN, Qi CJ, Tokita A 1994 Prediction of bone density from vitamin D receptor alleles. Nature 67:284–297. Ma J, Stampfer MJ, Gann PH, Hough HL, Giovannucci E, Kelsey KT, Hennekens CH, Hunter DJ 1998 Vitamin D receptor polymorphisms, circulating vitamin D metabolites, and risk of prostate cancer in United States physicians. Cancer Epidemiol Biomarkers Prev 7:385–390. Arai H, Miyamoto K, Taketami Y, Yamamoto H, Iemori Y, Morita K, Tonai T, Nishisho T, Mori S, Takeda E 1997 A vitamin D receptor gene polymorphism in the translation initiation codon: effect on protein activity and relation to bone mineral density in Japanese women. J Bone Miner Res 12:915–921. Lucotte G, Mercier G, Burckel A 1999 The vitamin D receptor FokI start codon polymorphism and bone mineral density in osteoporotic postmenopausal French women. Clin Genet 56:221–224. Gennari L, Becherini L, Mansani R, Masi L, Falchetti A, Morelli A, Colli E, Gonnelli S, Cepollaro C, Brandi ML 1999

CHAPTER 91 Epidemiology of Cancer Risk: Vitamin D and Calcium

45. 46.

47. 48.

49. 50. 51. 52.

53. 54. 55. 56. 57. 58. 59.

60. 61. 62.

63.

FokI polymorphism at translation initiation site of the vitamin D receptor gene predicts bone mineral density and vertebral fractures in postmenopausal Italian women. J Bone Miner Res 14:1379–1386. Slattery ML, Yakumo K, Hoffmann M, Neuhausen S 2001 Variants of the VDR gene and risk of colon cancer (United States). Cancer Causes Control 12:359–364. Ingles SA, Wang J, Coetzee GA, Lee ER, Frankl HD, Haile RW 2001 Vitamin D receptor polymorphisms and risk of colorectal adenomas (United States). Cancer Causes Control 12:607–614. Holick MF 1994 McCollum Award lecture, 1994: Vitamin D—new horizons for the 21st century. Am J Clin Nutr 60:619–630. van der Wielen RPJ, Lowik MRH, van den Berg H, de Groot LC, Haller J, Moreiras O, van Staveren WA 1995 Serum vitamin D concentrations among elderly people in Europe. Lancet 346:207–210. Lamprecht SA, Lipkin M 2001 Cellular mechanisms of calcium and vitamin D in the inhibition of colorectal carcinogenesis. Ann NY Acad Sci 952:73–87. Newmark HL, Wargovich MJ, Bruce WR 1984 Colon cancer and dietary fat, phosphate, and calcium: a hypothesis. J Natl Cancer Inst 72:1323–1325. van der Meer R, de Vries HT 1985 Differential binding of glycine- and taurine-conjugated bile acids to insoluble calcium phosphate. Biochem J 229:265–268. Lipkin M, Newmark H 1985 Effect of added dietary calcium on colonic epithelial-cell proliferation in subjects at high risk for familial colonic cancer. N Engl J Med 313:1381–1384. Buset M, Kipkin M, Winawer S, Swaroop S, Friedman E 1986 Inhibition of human colonic epithelial cell proliferation in vivo and in vitro by calcium. Cancer Res 46:5426–5430. Boing H, Martinez R, Frentzel-Beyme R, Oltersdorf U 1985 Regional nutritional pattern and cancer mortality in the Federal Republic of Germany. Nutr Cancer 7:121–130. Sorenson AW, Slattery ML, Ford MH 1988 Calcium and colon cancer: a review. Nutr Cancer 11:135–45. McKeown-Eyssen GE, Bright-See E 1984 Dietary factors in colon cancer: international relationships. Nutr Cancer 6:160–170. Rosen M, Nystom L, Wall S 1988 Diet and cancer mortality in the counties of Sweden. Am J Epidemiol 127:42–49. Martinez ME, Willett WC 1998 Calcium, vitamin D, and colorectal cancer: a review of the epidemiologic evidence. Cancer Epidemiol Biomarkers Prev 7:163–168. Heilbrun LK, Hankin JH, Nomura A, Hankin HJ, Stemmerman GN 1986 Colon cancer and dietary fat, phosphorus, and calcium in Hawaiian-Japanese men. Am J Clin Nutr 43:306–309. Macquart-Moulin G, Riboli E, Cornee J, Charnay B, Berthezene P, Day N 1986 Case-control study on colorectal cancer and diet in Marseilles. Int J Cancer 38:183–191. Kune S, Kune GA, Watson LF 1987 Case-control study of dietary etiologic factors: the Melbourne Colorectal Cancer Study. Nutr Cancer 9:21–42. Tuyns AJ, Haelterman M, Kaaks R 1987 Colorectal cancer and the intake of nutrients: oligosaccharides are a risk factor, fats are not. A case-control study in Belgium. Nutr Cancer 10:181–196. Graham S, Marshall J, Haughey B, Mittelman A, Swanson M, Zielezny M, Byers T, Wilkinson G, West D 1988 Dietary epidemiology of cancer of the colon in western New York. Am J Epidemiol 128:490–503.

1631

64. Slattery ML, Sorenson AW, Ford MH 1988 Dietary calcium intake as a mitigating factor in colon cancer. Am J Epidemiol 128:504–514. 65. Lee HP, Gourley L, Duffy SW, Esteve J, Lee J, Day NE 1989 Colorectal cancer and diet in an Asian population—A casecontrol study among Singapore Chinese. Int J Cancer 43:1007–1016. 66. Freudenheim JL, Graham S, Marshall JR, Haughey BP, Wilkinson G 1990 A case-control study of diet and rectal cancer in western New York. Am J Epidemiol 131:612–624. 67. Negri E, La Vecchia C, D Avanzo B, Franceschi S 1990 Calcium, dairy products, and colorectal cancer. Nutr Cancer 13:255–262. 68. Whittemore AS, Wu-Williams AH, Lee M, Zheng S, Gallagher RP, Jiao DA, Zhou L, Wang XH, Chen K, Jung D 1990 Diet, physical activity and colorectal cancer among Chinese in North America and China. J Natl Cancer Inst 82:915–926. 69. Arbman G, Axelson O, Ericsson-Begodzki A.-B, Fredriksson M, Nilsson E, Sjödahl R 1992 Cereal fiber, calcium, and colorectal cancer. Cancer 69:2042–2048. 70. Meyer F, White E 1993 Alcohol and nutrients in relation to colon cancer in middle-aged adults. Am J Epidemiol 138:225–236. 71. Zaridze D, Filipchenko V, Kustov V, Serdyuk V, Duffy S 1993 Diet and colorectal cancer: results of two case-control studies in Russia. Eur J Cancer 29A:112–115. 72. Kampman E, Van’t Veer P, Hiddink GJ, Van Aken-Schneijder P, Kok FJ, Hermus RJJ 1994 Fermented dairy products, dietary calcium, and colon cancer: a case-control study in the Netherlands. Int J Cancer 59:170–176. 73. Ghadirian P, Lacroix A, Maisonneuve P, Perret C, Potvin C, Gravel D, Bernard D, Boyle P 1997 Nutritional factors and colon carcinoma: a case-control study involving French Canadians in Montreal, Quebec, Canada. Cancer 80:858–864. 74. White E, Shannon JS, Patterson RE 1997 Relationship between vitamin and calcium supplement use and colon cancer. Cancer Epidemiol Biomarkers Prev 6:769–774. 75. Wu AH, Paganini-Hill A, Ross RK, Henderson BE 1987 Alcohol, physical activity, and other risk factors for colorectal cancer: a prospective study. Br J Cancer 55:687–694. 76. Norat T, Riboli E 2003 Dairy products and colorectal cancer. A review of possible mechanisms and epidemiological evidence. Eur J Clin Nutr 57:1–17. 77. Stemmermann GN, Normura A, Chyou PH 1990 The influence of dairy and nondairy calcium on subsite large bowel cancer risk. Dis Colon Rectum 33:190–194. 78. Kampman E, Goldbohm RA, van den Brandt PA, van’t Veer P 1994 Fermented dairy products, calcium, and colorectal cancer in the Netherlands Cohort Study. Cancer Res 54:3186–3190. 79. Kato I, Akhmedkhanov A, Koenig K, Toniolo PG, Shore RE, Riboli E 1997 Prospective study of diet and female colorectal cancer: the New York University Women’s Health Study. Nutr Cancer 28:276–281. 80. Hsing AW, McLaughlin JK, Chow W-H, Schuman LM, Co Chien HT, Gridley G, Bjelke E, Wacholder S, Blot WJ 1998 Risk factors for colorectal cancer in a prospective study among U.S. white men. Int J Cancer 77:549–553. 81. Pietinen P, Malila N, Virtanen M, Hartman TJ, Tangrea JA, Albanes D, Virtamo J 1999 Diet and risk of colorectal cancer in a cohort of Finnish men. Cancer Causes Control 10:387–396. 82. Wu K, Willett WC, Fuchs CS, Colditz GA, Giovannucci E 2002 Calcium intake and risk of colon cancer in women and men. J Natl Cancer Inst 94:437–446.

1632 83. Macquart-Moulin G, Riboli E, Cornée J, Kaaks R, Berthezène P 1987 Colorectal polyps and diet: a case-control study in Marseilles. Int J Cancer 40:179–188. 84. Tseng M, Murray SC, Kupper LL, Sandler RS 1996 Micronutrients and the risk of colorectal adenomas. Am J Epidemiol 144:1005–1014. 85. Hyman J, Baron JA, Dain BJ, Sandler RS, Haile RW, Manbdel JS, Mott LA, Greenberg ER 1998 Dietary and supplemental calcium and the recurrence of colorectal adenomas. Cancer Epidemiol 7:291–295. 86. Baron JA, Beach M, Mandel JS, van Stolk RU, Haile RW, Sandler RS, Rothstein R, Summers RW, Snover DC, Beck GJ, Bond JH, Greenberg ER, and the Calcium Polyp Prevention Study Group 1999 Calcium supplements for the prevention of colorectal adenomas. N Engl J Med 340:101–107. 87. Faivre J, Bonithon-Kopp C, MedicinneCINE FD 1999 A randomized trial of calcium and fiber supplementation in the prevention of recurrence of colorectal adenomas. In: American Gastroenterological Association Annual Meeting. Orlando, FL, pp. A357. 88. Reddy BS, Maeura Y 1984 Tumor promotion by dietary fat in azoxymethane-induced colon carcinogenesis in female F344 rats: influence of amount and source of dietary fat. J Natl Cancer Inst 72:745–750. 89. Bergsma-Kadijk JA, van’t Veer P, Kampman E, Burema J 1996 Calcium does not protect against colorectal neoplasia. Epidemiology 7:590–597. 90. Giovannucci E 2001 Insulin, insulin-like growth factors, and colon cancer: a review of the evidence. J Nutr 131:3109S3120S. 91. Ma J, Giovannucci E, Pollak M, Chan JM, Gaziano JM, Willett W, Stampfer MJ 2001 Milk intake, circulating levels of insulin-like growth factor-I, and risk of colorectal cancer in men. J Natl Cancer Inst 93:1330–1336. 92. Lipkin M, Friedman E, Winawer SJ, Newmark H 1989 Colonic epithelial cell proliferation in responders and nonresponders to supplemental dietary calcium. Cancer Res 49:248–254. 93. Rozen P, Fireman Z, Fine N, Wax Y, Ron E 1989 Oral calcium suppresses increased rectal epithelial proliferation of persons at risk of colorectal cancer. Gut 30:650–655. 94. Wargovich MJ, Isbel G, Shabot M, Winn R, Lanza F, Hochman L, Larson E, Lynch P, Roubein L, Levin B 1992 Calcium supplementation decreases rectal epithelial cell proliferation in subjects with sporadic adenoma. Gastroenterology 103:92–97. 95. Gregoire RC, Stern HS, Yeung KS, Stadler J, Langley S, Furrer R, Bruce WR 1989 Effect of calcium supplementation on mucosal cell proliferation in high-risk patients for colon cancer. Gut 30:376–382. 96. Stern HS, Gregoire RC, Kashtan H, Stadler J, Bruce RW 1990 Long-term effects of dietary calcium on risk markers for colon cancer in patients with familial polyposis. Surgery 108:528–533. 97. Kleibeuker JH, Welberg JW, Mulder NH, van der Meer R, Cats A, Limburg AJ, Kreumer WM, Hardonk MJ, de Vries EG 1993 Epithelial cell proliferation in the sigmoid colon of patients with adenomatous polyps increases during oral calcium supplementation. Br J Cancer 67:500–503. 98. Bostick R, Fosdick L, Wood JR, Grambsch P, Grandits GA, Lillemoe T, Louis TA, Potter JD 1995 Calcium and colorectoral epithelial cell proliferation in sporadic adenoma patients: a randomized, double-blinded, placebo-controlled clinical trial. J Natl Cancer Inst 87:1307–1315.

EDWARD GIOVANNUCCI AND ELIZABETH A. PLATZ

99. Cats A, Kleibeuker JH, van der Meer R, Kuipers F, Sluiter WJ, Hardonk MJ, de Vries EG 1995 Randomized, doubleblinded, placebo-controlled intervention study with supplemental calcium in families with hereditary nonpolyposis colorectal cancer. J Natl Cancer Inst 87:598–603. 100. Alberts DS, Einspahr J, Ritenbaugh C, Aickin M, ReesMcGee S, Atwood J, Emerson S, Mason-Liddil N, Bettinger L, Patel J, Bellapravalu S, Ramanujam PS, Phelps J, Clark L 1997 The effect of wheat bran fiber and calcium supplementation on rectal mucosal proliferation rates in patients with resected adenomatous colorectal polyps. Cancer Epidemiol Biomarkers Prev 6:161–169. 101. Holt PR, Arber N, Halmos B, Forde K, Kissileff H, McGlynn KA, Moss SF, Fan K, Yang K, Lipkin M 2002 Colonic epithelial cell proliferation decreases with increasing levels of serum 25-hydroxy vitamin D. Cancer Epidemiology, Biomarkers and Prevention 11:113–119. 102. American Cancer Society 2003 “Cancer Facts and Figures, 2003,” American Cancer Society, Atlanta, GA. 103. Parkin DM, Whelan SL, Ferlay J, Teppo L, Thomas D 2003 Cancer Incidence in Five Continents, International Agency for Research on Cancer, IARC Scientific Publications No. 155: Lyon, France. 104. Yatani R, Chigusa I, Akazaki K, Stemmermann GN, Welsh RA, Correa P 1982 Geographic pathology of latent prostatic carcinoma. Int J Cancer 29:611–666. 105. Jackson MA, Kovi J, Heshmat MY, Ogunmuyiwa TA, Jones GW, Williams AO, Christian EC, Nkposong EO, Rao MS, Jackson AG, Ahluwalia BS 1980 Characterization of prostatic carcinoma among blacks: a comparison between a low-incidence area, Ibadan, Nigeria, and a high-incidence area, Washington DC. Prostate 1:185–205. 106. Schwartz GG, Hulka BS 1990 Is vitamin D deficiency a risk factor for prostate cancer? (Hypothesis). Anticancer Res 10:1307–1311. 107. Hanchette CL, Schwartz GG 1992 Geographic patterns of prostate cancer mortality. Cancer 70:2861–2869. 108. Grant WB 2002 An estimate of premature cancer mortality in the U.S. due to inadequate doses of solar ultraviolet-B radiation. Cancer 94:1867–1875. 109. Luscombe CJ, Fryer AA, French ME, Liu S, Saxby MF, Jones PW, Strange RC 2001 Exposure to ultraviolet radiation: association with susceptibility and age at presentation with prostate cancer. Lancet 358:641–642. 110. Bodiwala D, Luscombe CJ, Liu S, Saxby M, French M, Jones PW, Fryer AA, Strange RC 2003 Prostate cancer risk and exposure to ultraviolet radiation: further support for the protective effect of sunlight. Cancer Lett 192:145–149. 111. Luscombe CJ, French ME, Liu S, Saxby MF, Jones PW, Fryer AA, Strange RC 2001 Prostate cancer risk: associations with ultraviolet radiation, tyrosinase, and melanocortin-1 receptor genotypes. Br J Cancer 85:1504–1509. 112. Giovannucci E, Rimm EB, Wolk A, Ascherio A, Stampfer MJ, Colditz GA, Willett WC 1998 Calcium and fructose intake in relation to risk of prostate cancer. Cancer Res 58: 442–447. 113. Chan JM, Giovannucci E, Andersson S-O, Yuen J, Adami H-O, Wolk A 1998 Dairy products, calcium, phosphorous, vitamin D, and risk of prostate cancer. Cancer Causes Control 9:559–566. 114. Chan JM, Pietinen P, Virtanen M, Malila N, Tangrea J 2000 Diet and prostate cancer risk in a cohort of smokers, with a specific focus on calcium and phosphorus (Finland). Cancer Causes Control 11:859–867.

CHAPTER 91 Epidemiology of Cancer Risk: Vitamin D and Calcium

115. Kristal AR, Cohen JH, Qu P, Stanford JL 2002 Associations of energy, fat, calcium, and vitamin D with prostate cancer risk. Cancer Epidemiol Biomarkers Prev 11:719–725. 116. Corder EH, Guess HA, Hulka BS, Friedman GD, Sadler M, Vollmer RT, Lobaugh B, Drezner MK, Vogelman JH, Orentreich N 1993 Vitamin D and prostate cancer: a prediagnostic study with stored sera. Cancer Epidemiol Biomarkers Prev 2:467–472. 117. Braun MM, Helzlsouer KJ, Hollis BW, Comstock GW 1995 Prostate cancer and prediagnostic levels of serum vitamin D metabolites (Maryland, United States). Cancer Causes Control 6:235–239. 118. Gann PH, Ma J, Hennekens CH, Hollis BW, Haddad JG, Stampfer MJ 1996 Circulating vitamin D metabolites in relation to subsequent development of prostate cancer. Cancer Epidemiol Biomarkers Prev 5:121–126. 119. Nomura AM, Stemmermann GN, Lee J, Kolonel LN, Chen TC, Turner A, Holick MF 1998 Serum vitamin D metabolite levels and the subsequent development of prostate cancer. Cancer Causes Control 9:425–432. 120. Ahonen MH, Tenkanen L, LT, Hakama M, Tuohimaa P 2000 Prostate cancer risk and prediagnostic serum 25-hydroxyvitamin D levels (Finland). Cancer Causes Control 11:847–852. 121. Platz EA, Leitzmann MF, Hollis BW, Willett WC, Giovannucci E 2004 Plasma 1,25-dihydroxy- and 25hydroxyvitamin D and subsequent risk of prostate cancer. Cancer Causes Control 15:255–265. 122. Schwartz GG 1994 Correspondence re: E. H. Corder et al. 1993 Vitamin D and prostate cancer: a prediagnostic study with stored sera. Cancer Epidemiol Biomarkers Prev 2:467–472 [comment]. Cancer Epidemiol Biomarkers Prev 3:183–184. 123. Taylor JA, Hirvonen A, Watson M, Pittman G, Mohler JL, Bell DA 1996 Association of prostate cancer with vitamin D receptor gene polymorphism. Cancer Res 56:4108–4110. 124. Ingles SA, Ross RK, Yu MC, Irvine RA, La Pera G, Haile RW, Coetzee GA 1997 Association of prostate cancer risk with genetic polymorphisms in vitamin D receptor and androgen receptor. J Natl Cancer Inst 89:166–170. 125. Medeiros R, Morais A, Vasconcelos A, Costa S, Pinto D, Oliveira J, Lopes C 2002 The role of vitamin D receptor gene polymorphisms in the susceptibility to prostate cancer of a southern European population. J Hum Genet 47:413–418. 126. Tayeb MT, Clark C, Haites NE, Sharp L, Murray GI, McLeod HL 2003 CYP3A4 and VDR gene polymorphisms and the risk of prostate cancer in men with benign prostatic hyperplasia. Br J Cancer 88:928–932. 127. Kibel AS, Isaacs SD, Isaacs WB, Bova GS 1998 Vitamin D receptor polymorphisms and lethal prostate cancer. J Urol 160:1405–1409. 128. Correa-Cerro L, Berthon P, Haussler J, Bochum S, Drelon E, Mangin P, Fournier G, Paiss T, Cussenot O, Vogel W 1999 Vitamin D receptor polymorphisms as markers in prostate cancer. Hum Genet 105:281–287. 129. Luscombe CJ, French ME, Liu S, Saxby MF, Jones PW, Fryer AA, Strange RC 2001 Outcome in prostate cancer associations with skin type and polymorphism in pigmentationrelated genes. Carcinogenesis 22:1343–1347. 130. Gsur A, Maderbacher S, Haidinger G, Schatzl G, Marberger M, Vutuc C, Micksche M 2002 Vitamin D receptor gene polymorphism and prostate cancer risk. Prostate 51:30–34. 131. Blazer DG, Umbach DM, Bostick RM, Taylor JA 2000 Vitamin D receptor polymorphisms and prostate cancer. Mol Carcinog 27:18–23.

1633

132. Hamasaki T, Inatomi H, Katoh T, Ikuyama T, Matsumoto T 2001 Clinical and pathological significance of vitamin D receptor gene polymorphism for prostate cancer which is associated with a higher mortality in Japanese. Endocrine J 48:543–549. 133. Habuchi T, Suzuki T, Sasaki R, Wang L, Sato K, Satoh S, Akao AT, Tsuchiya N, Shimoda N, Wada Y, Koizumi A, Chihara J, Ogawa O, Kato AT 2000 Association of vitamin D receptor gene polymorphism with prostate cancer and benign prostatic hyperplasia in a Japanese population. Cancer Res 60:305–308. 134. Furuya Y, Akakura K, Masai M, Ito H 1999 Vitamin D receptor gene polymorphism in Japanese patients with prostate cancer. Endocrine J 46:467–470. 135. Watanabe M, Fukutome K, Murata M, Uemura H, Kubota Y, Kawamura J, Yatani R 1999 Significance of vitamin D receptor gene polymorphism for prostate cancer risk in Japanese. Anticancer Res 19:4511–4514. 136. Suzuki K, Matsui H, Ohtake N, Nakata S, Takei T, Koike H, Nakazato H, Okugi H, Hasumi M, Fukabori Y, Kurokawa K, Yamanaka H 2003 Vitamin D receptor gene polymorphism in familial prostate cancer in a Japanese population. Int J Urology 10:261–266. 137. Xu Y, Shibata A, McNeal JE, Stamey TA, Feldman D, Peehl DM 2003 Vitamin D receptor start codon polymorphism (FokI) and prostate cancer progression. Cancer Epidemiol Biomarkers Prev 12:23–27. 138. Chokkalingam AP, McGlynn KA, Gao YT, Pollak M, Deng J, Sesterhenn IA, Mostofi FK, Fraumeni JF, Hsing AW 2001 Vitamin D receptor gene polymorphisms, insulin-like growth factors, and prostate cancer risk: a population-based casecontrol study in China. Cancer Res 61:4333–4336. 139. Hawkins GA, Cramer SD, Zheng SL, Isaacs SD, Wiley KE, Chang BL, Bleecker ER, Walsh PC, Meyers DA, Isaacs WB, Xu J 2002 Sequence variants in the human 25-hydroxyvitamin D3 1α-hydroxylase (CYP27B1) gene are not associated with prostate cancer risk. Prostate 53:175–178. 140. Rose DP, Boyar AP, Wynder EL 1986 International comparisons of mortality rates for cancer of the breast, ovary, prostate, and colon, and per capita food consumption. Cancer 58:2263–2271. 141. Mezzanotte G, Cislaghi C, Decarli A, Le Vecchia C 1986 Cancer mortality in broad Italian geographical areas, 1975–1977. Tumori 72:145–152. 142. Decarli A, La Vecchia C 1986 Environmental factors and cancer mortality in Italy: Correlational exercise. Oncology 43:116–126. 143. Grant WB 1999 An ecologic study of dietary links to prostate cancer. Altern Med Rev 4:162–169. 144. Talamini R, La Vecchia C, Decarli A, Negri E, Franceschi S 1986 Nutrition, social factors, and prostatic cancer in a Northern Italian population. Br J Cancer 53:817–821. 145. Talamini R, Franceschi S, La Vecchia C, Serraino D, Barra S, Negri E 1992 Diet and prostatic cancer: a case-control study in Northern Italy. Nutr Cancer 18:277–286. 146. Mettlin C, Selenskas S, Natarajan NS, Huben R 1989 Betacarotene and animal fats and their relationship to prostate cancer risk: A case-control study. Cancer 64:605–612. 147. La Vecchia C, Negri E, D’Avanzo B, Franceschi S, Boyle P 1991 Dairy products and the risk of prostatic cancer. Oncology 48:406–410. 148. Chan JM, Stampfer MJ, Giovannucci E, Gann PH, Ma J, Wilkinson P, Hennekens CH, Pollak M 1998 Plasma insulinlike growth factor-I and prostate cancer risk: a prospective study. Science 279:563–566.

1634 149. Jain MG, Hislop GT, Howe GR, Ghadirian P 1999 Plant foods, antioxidants, and prostate cancer risk: findings from case-control studies in Canada. Nutr Cancer 34:173–184. 150. De Stefani E, Fierro L, Barrios E, Ronco A 1995 Tobacco, alcohol, diet, and risk of prostate cancer. Tumori 81:315–320. 151. Rotkin ID 1977 Studies in the epidemiology of prostatic cancer: expanded sampling. Cancer Treat Rep 61:173–180. 152. Schuman LM, Mandel JS, Radke A, Seal U, Halberg F 1982 Some selected features of the epidemiology of prostatic cancer: Minneapolis-St. Paul, Minnesota case-control study, 1976–1979. In: K Magnus (ed) Trends in Cancer Incidence: Causes and Practical Implications. Hemisphere Publishing Corp: Washington, DC, pp. 345–354. 153. Hayes RB, Ziegler RG, Gridley G, Swanson C, Greenberg RS, Swanson GM, Schoenberg JB, Silverman DT, Brown LM, Pottern LM, Liff J, Schwartz AG, Fraumeni JF Jr, Hoover RN 1999 Dietary factors and risks for prostate cancer among blacks and whites in the United States. Cancer Epidemiol Biomarkers Prev 8:25–34. 154. Tzonou A, Signorello LB, Lagiou P, Wuu J, Trichopoulos D, Trichopoulou A 1999 Diet and cancer of the prostate: a casecontrol study in Greece. Int J Cancer 80:704–708. 155. Whittemore AS, Kolonel LN, Wu AH, John EM, Gallagher RP, Howe GR, Burch JD, Hankin J, Dreon DM, West DW, Teh CZ, Paffenbarger RS Jr 1995 Prostate cancer in relation to diet, physical activity, and body size in blacks, whites, and Asians in the United States and Canada. J Natl Cancer Inst 87:652–661. 156. Ewings P, Bowie C 1996 A case-control study of cancer of the prostate in Somerset and East Devon. Br J Cancer 74:661–666. 157. Deneo-Pellegrini H, De Stefani E, Ronco A, Mendilaharsu M 1999 Foods, nutrients, and prostate cancer: a case-control study in Uruguay. Br J Cancer 80:591–597. 158. Snowdon DA, Phillips RL, Choi W 1984 Diet, obesity, and risk of fatal prostate cancer. Am J Epidemiol 120:244–250. 159. Le Marchand L, Kolonel LN, Wilkens LR, Myers BC, Hirohata T 1994 Animal fat consumption and prostate cancer: a prospective study in Hawaii. Epidemiology 5:276–282.

EDWARD GIOVANNUCCI AND ELIZABETH A. PLATZ

160. Michaud DS, Augustsson K, Rimm EB, Stampfer MJ, Willett WC, Giovannucci E 2001 A prospective study on intake of animal products and risk of prostate cancer. Cancer Causes Control 12:557–567. 161. Chan JM, Stampfer MJ, Ma J, Gann PH, Gaziano JM, Giovannucci EL 2001 Dairy products, calcium, and prostate cancer risk in the Physicians’ Health Study (comment). Am J Clin Nutr 74:549–554. 162. Schuurman AG, van den Brandt PA, Dorant E, Goldbohm RA 1999 Animal products, calcium, and protein and prostate cancer risk in the Netherlands Cohort Study. Br J Cancer 80:1107–1113. 163. Hsing AW, McLaughlin JK, Schuman LM, Bjelke E, Gridley G, Wacholder S, Chien HT, Blot WJ 1990 Diet, tobacco use, and fatal prostate cancer: results from the Lutheran Brotherhood Cohort Study. Cancer Res 50:6836–6840. 164. Mills PK, Beeson WL, Phillips RL, Fraser GE 1989 Cohort study of diet, lifestyle, and prostate cancer in Adventist men. Cancer 64:598–604. 165. Veierod MB, Laake P, Thelle DS 1997 Dietary fat intake and risk of prostate cancer: a prospective study of 25,708 Norwegian men. Int J Cancer 73:634–638. 166. Rodriguez C, McCullough ML, Mondul AM, Jacobs EJ, Fakhrabadi-Shokoohi D, Giovannucci EL, Thun MJ, Calle EE 2003 Calcium, dairy products, and risk of prostate cancer in a prospective cohort of United States men. Cancer Epidemiol Biomarkers Prev 12:597–603. 167. Berndt SI, Carter HB, Landis PK, Tucker KL, Hsieh LJ, Metter EJ, Platz EA 2002 Calcium intake and prostate cancer risk in a long-term aging study: the Baltimore Longitudinal Study of Aging. Urology 60:1118–1123. 168. Tavani A, Gallus S, Franceschi S, La Vecchia C 2001 Calcium, dairy products, and the risk of prostate cancer. Prostate 48:118–121. 169. Barreto AM, Schwartz GG, Woodruff R, Cramer SD 2000 25-Hydroxyvitamin D3, the prohormone of 1,25-dihydroxyvitamin D3, inhibits the proliferation of primary prostatic epithelial cells. Cancer Epidemiol Biomarkers Prev 9: 265–270.

CHAPTER 92

Differentiation and the Cell Cycle GEORGE P. STUDZINSKI UMD-New Jersey Medical School, Newark, New Jersey

MICHAEL DANILENKO Department of Clinical Biochemistry, Faculty of Health Science, Ben-Gurion University fo the Negev, Beer-Sheva, Israel

I. Introduction II. Induction of Differentiation by 1,25(OH)2D3 and Analogs (“Deltanoids”) III. Cell Cycle Consequences of Deltanoid-induced Differentiation

IV. Cell-type Specificity of Inhibition of Cell Proliferation by Deltanoids Without Evidence of Differentiation V. Conclusions References

I. INTRODUCTION

However, even in these cases, there is an eventual slowdown of the cell cycle traverse and cessation of proliferation of differentiated cells. Consequently, numerous attempts are being made to exploit the differentiating actions of vitamin D derivatives (“deltanoids”) to induce proliferative quiescence of neoplastic cells, and thus increase the range of options for optimal therapy of human cancer. This is addressed in other chapters, primarily Chapter 97. The term “differentiation” is often used in a rather specialized way to imply the acquisition of new functional properties by a cell that already appears to be mature and capable of function. Examples are monocytes “differentiating” into macrophages, or naive T lymphocytes becoming helper or cytotoxic cells. Little is known, however, of cell cycle alterations in this form of differentiation.

In general, cell cycle control is extremely wellconserved throughout the eukaryotic species. The basic machinery consists of several cyclin-dependent kinases and cyclins that pair with each other, sometimes changing partners, to drive the cell towards and through mitosis [1–4]. This basic arrangement seems almost monotonously similar in all cells, yet in multicelluar organisms control of cell proliferation must be, and is, cell type specific. The required control is provided by proteins that regulate the kinase activity of Cdk/cyclin pairs, most often in a negative fashion, and usually occurs in response to cues from the environment. Importantly, cell cycle changes are also triggered in cells undergoing differentiation. Differentiation can be considered to be, in essence, a persistent pattern of expression of previously dormant genes, which results in new functional capabilities of the differentiated cell. The new functions require cellular resources that compete with and finally titrate out the resources required for proliferation, and allow an accumulation of negative regulators of the cell cycle (e.g., p27Kip1 [5]), which then predominate over the positive regulators. Thus, there is a reciprocal relationship between cellular differentiation and cell cycle progression/proliferation [6–8], though there is also evidence that differentiation and cycle arrest need not be strictly coupled [9–12]. Cell cycle changes in differentiating cells do not always take place immediately—in some cells there is at first a boost of proliferation—as in normal hematopoiesis, or in HL60 [11–13] and U937 [14] cells differentiating in response to derivatives of vitamin D3. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

II. INDUCTION OF DIFFERENTIATION BY 1,25(OH)2D3 AND ANALOGS (“DELTANOIDS”) A. Principal Models In 1981 the Suda Laboratory reported that an in vitro exposure of M1 mouse myeloid leukemia cells to 1,25(OH)2D3 induces these immature cells to differentiate into functional macrophages [15]. This seminal finding was followed by numerous reports not only confirming that leukemic cells, murine or human [16–19], differentiate in response to treatment with 1,25(OH)2D3 or other deltanoids, but also demonstrating that several other types of neoplastic cells show Copyright © 2005, Elsevier, Inc. All rights reserved.

1636 similar responses. These include colon cancer, breast cancer, prostate cancer, neuroblastoma, osteosarcoma, squamous cell carcinoma, and malignant melanoma. In addition, normally developing immature cells can be induced to differentiate by 1,25(OH)2D3, for example keratinocytes, myoblasts, and perhaps hematopoietic cell precursors. Specific markers by which these phenotypes can be recognized are summarized, and sample references are provided, in Table I. It is apparent that 1,25(OH)2D3 is a powerful differentiating agent that targets diverse cell types, but its physiological role in development remains to be fully determined. Although data from vitamin D receptor (VDR) knockout mice suggest that 1,25(OH)2D3 can function in differentiation of normal cells (Chapter 20), most in vitro studies with normal immature cells used concentrations of 1,25(OH)2D3 in great excess over the physiological levels. While it can be argued that concentrations of 1,25(OH)2D3 in specific tissue niches in which precursor cells differentiate can be higher than the concentrations found in the plasma, a clearer assessment of the physiological role of 1,25(OH)2D3 in normal differentiation may be achieved by studies that utilize low nanomolar or subnanomolar concentrations of 1,25(OH)2D3.

B. Initial Signals for Differentiation Effects of 1,25(OH)2D3 Cells that differentiate when exposed to 1,25(OH)2D3 usually express VDR [20–23]. In several studies, evidence that VDR is strictly required for the 1,25(OH)2D3–induced differentiation was obtained. For instance, transfection of the VDR conferred differentiation responsiveness to 1,25(OH)2D3 in WEHI-3B D+ murine myelomonocytic cells, which lack inducible VDR expression [24]. In VDR knockout mice, 12-Otetradecanoylphorbol-13-acetate, but not 1,25(OH)2D3 [25] or 19-nor-1,25-dihydroxyvitamin D2 [26], induced differentiation of bone marrow-committed myeloid stem cells to monocytes/macrophages. This indicates the requirement of VDR for 1,25(OH)2D3–induced monocyte/macrophage differentiation. Similarly, VDR is required for morphogenesis and negative growth regulation in the mammary gland [27]. VDR is a nuclear protein that may also shuttle to and from the cytoplasm [28] (reviewed in Chapter 22). Thus, liganded VDR, often acting as a heterodimer with one of the three members of the retinoid X receptor (RXR) family is the principal mediator of the effects of deltanoids, as described in detail in Chapter 13. There is, however, intriguing evidence that 1,25(OH)2D3 and especially some other deltanoids have direct

GEORGE P. STUDZINSKI AND MICHAEL DANILENKO

cell membrane effects that modify, or enhance, the VDR-transmitted signals. This is extensively discussed in Chapter 23. Although the membrane effects originally described in the enterocytes are able to account for the very rapid calcium transport in the intestine (“transcaltachia” [29,30]), several reports suggest that the increased therapeutic ratios (i.e., the ratios of differentiation to calcemic potencies) of 1,25(OH)2D3, and some low-calcemic deltanoids may be due, in part, to the auxiliary effects of non-VDR mediated actions of these compounds [31–33]. Nonetheless, it is most likely that the actions of VDR-RXR heterodimers that modulate gene expression transmit the differentiation signals to the basal transcription machinery by interacting with vitamin D response element (VDRE) sequences of the DNA molecule [34,35]. A large assortment of nuclear receptor co-activators, such as DRIP/Mediator and SRC/p160 [36–39] and co-repressors (e.g., SMRT and N-CoR) [38,40–42] has been identified, which provides positive or negative regulation of the VDR transcriptional activity. In addition to the “classic” VDRE sequences that transactivate vitamin D–responsive gene transcription, “negative” VDREs have also been characterized that inhibit transcription of certain genes, e.g., parathyroid hormone gene [43,44]. Interestingly, in myeloid leukemia cells, promyelocytic leukemia zinc finger (PLZF) protein and the chromosomal translocation products, such as promyelocytic leukemia-retinoic acid receptor alpha (PML/RARα) and PLZF/RARα fusion proteins also repress the differentiating action of VDR by binding and sequestering this receptor [45,46]. The regulatory effects of VDR co-modulators are discussed in detail in Chapters 14 and 16. How the initial vitamin D–induced gene expression leads to the acquisition of a new functional phenotype, i.e. differentiation, is one of the current mysteries, as among the known direct target genes of VDR only one, p21Cip1, has a possible relevance to the differentiating actions of 1,25(OH)2D3 in tissues other than bone [14,47]. Thus, with regard to differentiation of most cell types, the links from VDR-initiated events to the downstream targets of 1,25(OH)2D3 still need to be found.

C. Pathways That Participate in 1,25(OH)2D3 Signal Propagation Pharmacologic inhibitors of several signaling pathways reduce, to varying extents, 1,25(OH)2D3-induced differentiation in diverse model systems [13,48–54] encouraging the belief that these pathways participate in differentiation signaling.

1637

CHAPTER 92 Differentiation and the Cell Cycle

TABLE I Examples of Cellular Models of Differentiation Induced by 1,25(OH)2D3 and Deltanoids Differentiation marker

Known function

Comments

Hematopoietic cells (HL60, U937, THP-1, UF-1, WEHI-3) CD14 LPS cell-surface binding protein CD11b Cell surface protein (integrin αM) Nonspecific esterase Cytoplasmic hydrolytic enzyme Superoxide anion Component of oxidative burst Morphologic changesa

References

Early monocytic differentiation General myeloid differentiation Monocytic differentiation Phagocytic activity

[16,113] [220,272,273] [16,256,274] [256,275–277] [102]

Colon cancer cells (Caco-2, primary adenoma and carcinoma lines) Alkaline phosphatase Brush border-associated hydrolase Carcinoembryonic antigen (CEA) Adhesion molecule

Intestinal and placental isozymes Early development protein

[21,89,278–280] [281,282]

Osteoblast-like cells (MG-63, ROS 17/2.8, MC3T3-E1) Osteocalcin Osteoblast-specific noncollagenous protein Alkaline phosphatase Hydrolytic enzyme

Late osteoblastic differentiation Bone mineralization

[283–285] [283,286,287]

Prostate cancer cells (LNCaP, PC-3) Prostate-specific antigen (PSA) Serine protease Prostate-specific acid phosphatase Protein tyrosine phosphatase E-cadherin Calcium-dependent cell adhesion molecule

Secreted by prostate epithelial cells Prostate growth regulating enzyme Major epithelial cadherin

[288–292] [291] [290,293]

Breast cancer cells (MCF-7, T47D, MDA-MB-231, MDA-MB-436, BT-20, SK-BR-3, UISO-BCA-4) Intracellular lipid droplets Storage/precursor material Casein Major milk protein

[294–297] [295,297]

Neuroblastoma (LA-N-5) Acetylcholine esterase Neurite outgrowth Melanoma cells (B16) Tyrosinase

Serine hydrolase

May regulate neurite outgrowth

[298–300] [299,300]

Copper-containing oxidase

Key enzyme in melanin synthesis

[301,302]

Structural skin component Keratinocyte-specific form Cornified cell envelope constituent

[303] [304] [303,305,306]

Calcium-dependent crosslinking enzyme Glutamine-rich transglutaminase substrate Cysteine proteinase inhibitor

Keratinocyte-specific form Cornified cell envelope constituent Cornified cell envelope constituent

[304,307,308] [116,305,308] [309,310] [307,311]

Contractile protein ATP metabolizing enzyme

Late differentiation

[49,312] [49,313]

Squamous cell carcinoma (SCC13, SCC25) Keratin 1 Fibrous scleroprotein Transglutaminase Calcium-dependent crosslinking enzyme Involucrin Glutamine-rich transglutaminase substrate Keratinocytes Transglutaminase Involucrin Cystatin A Cornified envelope formation Muscle cells (C2C12) Myosin Creatine kinase a

Morphologic changes can be recognized in many forms of differentiation.

1. PROTEIN KINASE C

A number of early studies linked several isoforms of PKC to differentiation [55–59]. Following the observation by Martell, Simpson, and Taylor [60] that treatment

of HL60 cells by 1,25(OH)2D3 increases cellular TPA receptors, which implies increased abundance of PKC, the Hannun Laboratory showed that 1,25(OH)2D3 increases the mRNA for isoforms α and β of PKC in

1638 these cells [61]. These results were duplicated in many other laboratories, and PKC inhibitors or antisense oligonucleotides to this enzyme were demonstrated to interfere with 1,25(OH)2D3-induced differentiation [49,51,62], further implying a role for at least some isoforms of PKC in differentiation induced by 1,25(OH)2D3. If its role could be established, PKC would provide a central position in a logically pleasing sequence of events that led from an exposure of a cell to 1,25(OH)2D3 towards differentiation. First, the lipid-soluble 1,25(OH)2D3 may interact with cell membrane lipids or activate membrane-associated phospholipases [63] directly or through the still elusive membrane receptor, to generate a rise in phospholipidderived intracellular calcium ([Ca2+]i). As the result of raised [Ca2+]i and DAG concentrations, several PKC isoforms can be activated [64,65], and the signal can be propagated further by activating regulators of other signaling pathways, such as the mitogen-activated protein kinase (MAPK) pathway. Known examples of such links are the phosphorylation of Raf1 by PKCα in colon cancer CaCo-2 cells [66], and the translocation of MAPK ERK1/2 to the nucleus [67]. Another potential link is the regulation of VDR by PKC activation [68–70]. However, several major difficulties have so far precluded a full assessment of the role of PKC activity in differentiation. These include its presence in multiple isoforms with overlapping properties, and the fact that full inhibition of cellular PKC activity is usually incompatible with cell survival. 2. PHOSPHATIDYLINOSITOL 3-KINASE

Effects of 1,25(OH)2D3 originating in cell membranes are also linked to several other signaling pathways in still ill-defined ways. The phosphatidylinositol 3-kinase (PI3-K)/AKT signaling has been implicated in 1,25(OH)2D3-induced differentiation of THP-1 human leukemia cells [48], HL60 cells [71–73], and keratinocytes, in which the membrane receptor for 1,25(OH)2D3 was reported to be annexin II [54]. 3. MAP KINASE PATHWAYS

Although it is possible that MAPK activation also originates at the cell membrane, particularly in intestinal cells, osteoblasts, and chondrocytes, where the activation of MAPKs is very rapid [74], in other cell systems up-regulation of MAPKs can take place in a matter of hours and days, not only seconds or minutes, and genomic actions of 1,25(OH)2D3 appear to be essential. In these model systems, which include HL60 leukemia cells, three MAPK pathways have been extensively studied. These pathways utilize ERKs, JNKs, and p38 MAPKs as the principal signal transmitters [75], as illustrated in Fig 1.

GEORGE P. STUDZINSKI AND MICHAEL DANILENKO

1,25D3

? An upstream regulator

KSR Raf1

MKK3/6

MKK4

MEK1/2

p38

JNK

ERK1/2

?

PKC

P c-Jun Fos family p38 targets

Differentiation

AP-1

p27 Cell cycle arrest

ERK targets VDR

TRE

FIGURE 1

The MAP kinase pathways, which are up-regulated by 1,25(OH)2D3 in leukemia cells. The ERK and JNK pathways have positive effects on differentiation [13,54], while the p38 MAPK pathway has an inhibitory effect on monocytic, but not granulocytic differentiation of HL60 cells [91,93]. Also shown is the potential role of the AP-1 transcription factor, which may act as an intermediary positive effector of 1,25(OH)2D3 signals by upregulating the expression of VDR [98]. The dotted line depicts the upregulation by 1,25(OH)2D3 of kinase suppressor of Ras (KSR) [86], which is presumably indirect.

The MEK/ERK MAPK module is activated by a sequence of kinase reactions that are initiated at the cell membrane by extracellular signals, which interact with receptors that include growth factor and cytokine receptors [76–79]. This cascade of phosphorylations leads to DNA replication, differentiation, and enhanced cell survival, the outcomes depending on the duration of the signal as well as the activity of other signaling pathways. The current paradigm is that these membrane events activate the small G protein Ras, which sequentially activates Raf1, MEK1, then ERK1/2, which translocates to the nucleus and phosphorylates transcription factors, thus increasing their activity [76–79]. Recently, however, another component, kinase suppressor of the Ras (KSR1), has been shown to be an intermediary between Ras, Raf, and MEK [50,80–82]. Although its function has been argued to be a kinase that phosphorylates and thus activates Raf1 [83,84], or a scaffold that brings Ras, Raf, and MEK together [85], it is clear that KSR facilitates the actions of the ERK MAPK pathway irrespective of its mode of action. Interestingly, KSR is up-regulated by 1,25(OH)2D3 in HL60 cells, and appears to amplify the differentiation signal provided by nanomolar concentrations of 1,25(OH)2D3 [50,86]. Activation of ERK2 by 1,25(OH)2D3 has been found in many differentiation systems. For instance, “rapid

1639

CHAPTER 92 Differentiation and the Cell Cycle

and transient” activation has been reported in NB4 acute promyelocytic leukemia cells [87], normal human keratinocytes [54,88], CaCo-2 colon cancer cells [89], and HL60 human myeloid leukemia cells [13,90]. While all these examples were found to be transient, the scale of “rapidity” varied from 30 sec [87], to several hours [13]. In the case of HL60 cells, Wang and Studzinski [13] found that the period of ERK2 activation following exposure to 1,25(OH)2D3 corresponded to the time that these cells continued to proliferate before 1,25(OH)2D3-induced cell cycle block became apparent. Similarly, Gniadecki [88] reported that in normal human keratinocytes 1,25(OH)2D3 stimulated the activity of MAPKs and DNA synthesis. Together with the finding by Song et al. [87] of MAPK activation by 1,25(OH)2D3 in NB4 leukemia cells in which 1,25(OH)2D3, when administered alone, does not induce differentiation, these observations raise the question regarding the nature of the association between ERK activation and 1,25(OH)2D3-induced differentiation. It is possible that ERK facilitates the early, proliferative phase of differentiation [9,11,12,91], thus increasing the numbers of differentiated cells, rather than being related to the expression of the differentiated phenotype. This is consistent with the observation that the MEK/ERK inhibitor PD98059 substantially reduces, but does not totally prevent, 1,25(OH)2D3 induced differentiation [13], and the report that transformation of immortalized keratinocytes with Ha-RAS oncogene, an upstream regulator of the MEK/ERK pathway, results in resistance to the antiproliferative action of 1,25(OH)2D3 [92]. The stress and pro-inflammatory cytokine-activated p38 MAPK has a complex relationship to 1,25(OH)2D3induced differentiation [75,93–96], as well as to apoptosis [97]. It has been reported that p38-mediated signals are necessary for induction of osteoclast differentiation, but not for osteoclast function [96], and that in keratinocytes 1,25(OH)2D3 inhibits its activation [95]. In contrast, in HL60 cells 1,25(OH)2D3 and other deltanoids activate p38, though inhibition of its activity by specific inhibitors SB202190 or SB203580 actually increases differentiation [93,94,98]. It was postulated that this paradoxical effect is due to the presence of a negative feedback mechanism that regulates several MAPK pathways (Fig. 1), and this explanation is consistent with the finding of increased JNK pathway activity, as well as enhanced differentiation, in HL60 cells treated with the p38 MAPK inhibitors SB202190 or SB203580 and 1,25(OH)2D3 [93,94]. The JNK MAPK pathway activation by 1,25(OH)2D3 in general has a positive effect on differentiation. In addition to HL60 cells [53,93,94], this pathway has been shown to be involved in stimulation of CaCo-2 cell differentiation [89], and, together with p38, in sensitization

of human breast cancer cells MCF-7 to 1,25(OH)2D3– induced growth inhibition [99]. In contrast to ERK MAPK pathway activation which is transient, the JNK pathway activity increases more slowly after an exposure to 1,25(OH)2D3, and in HL60 cells parallels the growth inhibitory effects of 1,25(OH)2D3 [53]. Thus, in HL60 cells, enhanced JNK activity is a feature of late stages of monocytic differentiation, and perhaps is responsible for its maintenance. 4. THE CDK5/P35 PATHWAY

The pathways described above have been implicated in 1,25(OH)2D3–induced differentiation by showing that their inhibition by pharmacological agents and dominant-negative or antisense constructs diminishes differentiation, suggesting that they are necessary for, or contributory to, signaling of differentiation. Whether the enhanced activity of any of these pathways is sufficient to induce the differentiated phenotype has not been unequivocally shown. One exception is the demonstration that transfection of a cyclin-dependent kinase 5 (Cdk5)-expressing plasmid can lead to the expression of markers of early monocytic differentiation (CD14 and NSE) if a cyclin-like protein p35Nck5a is present [100]. Since the expression of both Cdk5 and p35 is upregulated by 1,25(OH)2D3 [100–102], this provides a mechanism by which 1,25(OH)2D3 can signal the early stages of monocytic differentiation. A confirmation of this role of the Cdk5/p35 module was obtained by finding that monocytes in p35 knockout mice were deficient in NSE, a monocyte-specific esterase used as a cell-specific marker [103]. The role of the ERK pathways as the upstream regulator of Cdk5/p35 was studied by Harada et al. [104] in differentiating neurons. The authors reported that the nerve-growth factor-induced differentiation in these cells was associated with a strong, sustained expression of p35 through activation of the ERK pathway, and that constitutive activation of ERK was necessary and sufficient for p35 induction. Furthermore, the mechanism by which activation of ERK induces expression of p35 involves the transcription factor Egr1, which was in these experiments induced by nerve-growth factor through the ERK pathway and appeared to mediate the induction of p35 by ERK [104]. Thus, while it is likely that MAPK pathways are upstream of Cdk5/p35, the evidence is currently not available for deltanoid–induced differentiation.

D. Role of General Transcription Factors in 1,25(OH)2D3-Induced Differentiation In addition to directly activating VDR to heterodimerize with a member of RXR family, which results in

1640 transcription of VDRE-containing genes, several ubiquitous transcription factors appear to be involved, perhaps in a contributory way, in 1,25(OH)2D3–induced differentiation. These may act by interacting with the adjacent VDREs [105], by up-regulating VDR expression through its promoter region (e.g., ref. [99]), and in other ways, many of which remain to be elucidated. 1. ACTIVATOR PROTEIN-1 (AP-1)

ERK and JNK pathways activate members of the Fos and Jun families, as well as some related proteins, which dimerize in various combinations to form the AP-1 transcription factor [106,107]. Thus, AP-1 can integrate and transmit signals transduced by the MAPK pathways previously discussed. Early studies have shown, for instance, that the expression of c-Jun is increased during the 1,25(OH)2D3–induced differentiation of human myeloid cells [53,108], and coordinate occupancy of AP-1 sites and VDRE elements by their cognate transcription factors provide a possible model for the reciprocal relationships between different cellular phenotypes and functional activities such as those that occur during differentiation [109,110]. For example, a composite AP-1 steroid hormone element that responds to 1,25(OH)2D3 mediates differentiationspecific gene expression of human keratin-1 [111]. There is also evidence for functional cooperation between VDR and Ras-activated Ets transcription factors in 1,25(OH)2D3–mediated induction of CYP24 gene expression [112]. Liu and Freedman [105] conducted an extensive study of such transcriptional synergism between VDR and nonreceptor transcription factors, and concluded that its functional basis appears to be at the level of cooperative DNA binding. AP-1 activation by 1,25(OH)2D3 has been described in diverse differentiation systems. These include HL60 cells [113], colon cancer CaCo-2 cells [89], osteoblastic cells [114], keratinocytes [115,116], and breast cancer cells [99]. Interestingly, the pathways that signal AP-1 activation are apparently cell type-specific. For instance, the p38 and JNK MAPK pathways cooperate to activate VDR by c-Jun/AP-1 in breast cancer cells [99], while in keratinocytes and HL60 leukemia cells AP-1 activation is attributed to both the ERK and JNK pathways [54,93]. Thus, AP-1 transcription factor appears to be an important integrator of converging differentiation pathways. 2. SP1 TRANSCRIPTION FACTOR

The Spl, a 95–105 Kd protein, is ubiquitously expressed in growing cells, and, usually in combination with other factors, acts as a transcriptional activator of many housekeeping genes [117–121]. Its role in 1,25(OH)2D3–induced differentiation has been suggested

GEORGE P. STUDZINSKI AND MICHAEL DANILENKO

in human myeloid leukemia cells. Specifically, the DNA binding of Spl was found to be increased following an exposure of HL60 cells to 1,25(OH)2D3 [10,113,122], while in U937 cells Spl may participate in the 1,25(OH)2D3–induced expression of the CD14 monocyte marker, which has several Spl sites in its promoter [123]. It was also shown that up-regulation of p27Kipl in 1,25(OH)2D3-treated U937 cells can be mediated by Spl [124], though more recent work indicates that forkhead transcription factors play a major role in the transcriptional regulation of p27Kip1 expression [125,126]. 3. OTHER TRANSCRIPTION FACTORS

Undoubtedly, many other transcription factors contribute to 1,25(OH)2D3-induced differentiation and the eventual cell cycle arrest and some of these, notably c-Myc, will be discussed relative to cell cycle control. The important distinction between factors, which regulate the expression of new genetic programs, and factors which carry out functions of the differentiated cells is not easy to make at present.

III. CELL CYCLE CONSEQUENCES OF DELTANOID-INDUCED DIFFERENTIATION A. General Features of Cell Cycle Machinery 1. CELL CYCLE COMPARTMENTS AND CHECKPOINTS

The consecutive progression through four distinct phases of the cell cycle called G1, S, G2, and M results in proliferation of eukaryotic cells (Fig. 2). DNA replication occurs during the S phase; chromosome separation (karyokinesis) takes place during the M phase, and is followed by cell division (cytokinesis); G1 and G2 are gap or growth phases. The G1 phase can be further subdivided into early G1, or post-mitotic G1, mid-G1, in which principal cell growth takes place, and late-G1, in which final preparations for DNA replication occur. The G2 phase is thought to be necessary for monitoring of chromosome replication and preparations for mitotic spindle assembly [127–129]. Cells that are not actively dividing may either be permanently removed from these cycling phases by terminal differentiation, senescence or apoptosis, or be temporarily arrested in a noncycling quiescent state known as G0 if the cells have the G1 DNA content, though quiescence can also occasionally take place in the G2 phase (G2 arrest). As mentioned above, specific nuances have been described, but the remarkable feature of the cell cycle is the conservation, from yeast to

1641

CHAPTER 92 Differentiation and the Cell Cycle

Terminal differentiation

Terminal differentiation

Polyploidy

p130Rb2-E2F complex

Licensing factors G0 Cdk1-Cyc B

M p21Cip1 p27Kip1

G2

pRb-P < pRb/E2F Cdk4/6-Cyc D

G1

Cdk2-Cyc A/E S

c-Myc R

Post-RC ORC

Cdk2-Cyc E pRb-P > pRb + E2F

Pre-RC ORC Cdc 6 /18 MCM Cdt1

FIGURE 2 The general concept of the cell cycle, and examples of factors that control DNA replication and the cell cycle traverse. The principal locus for terminal differentiation is in G0, but can occur in G2, with polyploidy. In early G1 the level of phosphorylated retinoblastoma protein (P-pRb) is less (<) than the level of hypophosphorylated pRb protein complexed with transcription factor E2F (pRb/E2F), while in late G1 the level of P-pRb is high, and it is not complexed with E2F. This and the increasing Cdk2/Cyclin E activity allow passage through the Restriction point (R). Additional details of the controls of cell cycle progression are provided in the text and illustrated in Figs. 3–6. RC, replicative complex; ORC, origin recognition complex; MCM, mini-chromosome maintenance complex; licensing factors, ORC, MCM, and other components of the pre-RC complex; Cyc, cyclin.

mammalian cells, of the basic regulatory mechanisms and components. A series of regulatory steps, referred to as checkpoints, control the traverse of the various compartments of the cell cycle [130]. A simple definition of a checkpoint is that it is a mechanism that prevents progression to the next part of the cell cycle unless and until the preceding part has been satisfactorily completed. Such checkpoints operate in each phase of the cell cycle, and although the precise mechanisms are in most cases unclear, DNA damage is known to activate two protein kinases, Chk1 and Chk2, which then mediate cell cycle arrest [131,132]. In addition, several regulators downstream from Chk1 or Chk2, such as p53, have been shown to have importance in checkpoint control. Interestingly, these regulators of cell cycle progression may also influence cellular decisions to differentiate [133]. The forerunner of checkpoints was described as the restriction (R) point in mid to late G1 phase. Based on

work in his own and in other laboratories, Pardee defined a transition in G1 phase which commits a cell to initiate DNA replication [127]. Additional work indicated that phosphorylation of the retinoblastoma susceptibility protein (pRb) may provide the principal mechanism for the transition through the R point [134]. Subsequent passage through the S phase can also be controlled by the S phase checkpoints [135]. The final result of the cell cycle traverse is a faithful replication and accurate partitioning of genetic information. The fidelity of this partitioning is maintained by the G2 and the M phase checkpoints. The G2 phase checkpoints monitor the integrity of DNA and the accuracy of DNA replication, and the M phase checkpoints ensure correct chromosomal segregation and alignment. Each of these checkpoints arrests cell cycle progress to allow editing and repair of genetic information. Overall, the checkpoints assure that each daughter cell receives a full complement of genetic information intact and identical to the parental cell.

1642

GEORGE P. STUDZINSKI AND MICHAEL DANILENKO

2. Mechanisms That Drive Cell Cycle Progression

of cyclin H (regulatory subunit), Cdk7 (catalytic subunit), and MAT1 (assembly factor) [139,140]. This complex phosphorylates threonines (Thr) 161/160 on Cdk1 (formerly Cdc2) and Cdk2, respectively [141]. Phosphorylation of this site is necessary for the cyclinCdk complex to be activated. Cdk inhibitory phosphorylation sites include Thr14 and Tyr15. Regulation of these sites is achieved by a group of proteins that include Cdc25A, B, C (phosphatases), as well as wee1 and mik1 (kinases). The phosphatases activate the cell cycle by cleaving the phosphate groups on Thr14 and Tyr15 residues of the Cdk, while the kinases are inhibitory by phosphorylating the same sites. Cdc25A is involved in the G1/S checkpoint [142], while Cdc25B and C, and human wee1-like kinase, regulate the traverse through the G2/M phases [143–145]. If DNA damage occurs prior to or in the S phase, Chk2 phosphorylates Cdc25A and targets it for degradation, while G2/M transition is regulated by Chk1 and Chk2 by phosphorylation of Cdc25C at Ser216 [146]. Another level of Cdk regulation is provided by the Cdk inhibitory proteins (CDKIs), which prevent Cdk activation, generally by binding to the kinases, thus preventing their activation by cyclins, as described below.

Passage through the restriction points is propelled by the activity of a group of enzymes known as cyclindependent kinases (Cdks). Cdks are usually present throughout the cell cycle, and work in concert with cyclins, which are nuclear proteins whose levels oscillate in a cell cycle-dependent manner [136,137]. In general, there are at least nine levels at which the activity of Cdks can be controlled, as detailed in Fig. 3. The primary regulator of Cdks activity is cyclin binding, because Cdks and cyclins need to form a complex prior to activation. Since the protein levels of the cyclins dramatically change during the cell cycle, the binding of Cdks and cyclins is related to cyclin availability. The abundance of cyclins, like all proteins, is dependent on the balance between their synthesis and their degradation, the latter occurring by ubiquitindependent proteolysis [138]. The activity of the cyclin-Cdk complexes also depends on both activating and inhibitory phosphorylations. A known kinase, which can phosphorylate Cdks is cyclin activating kinase (CAK). CAK is comprised

Synthesis

Degradation

2

CAK

3

CDK7 MAT1 Cyclin

Cyclin H 4

CDKI

CDK

9

P T 160

CDKI

Wee1 Mik1

Cyclin

T 14 CDK

CDK 1

Y 15 P T 160

Cyclin

7 5

CDK

CDC25

Cyclin

6 Chk1/2

8 T 14 P Y 15 P

Degradation

FIGURE 3 The central paradigm for the control of the cycle traverse by cyclin-dependent kinases (Cdk). Cdk activity is primarily activated by cyclin binding, phosphorylation of Thr160 by cyclin activating kinase (CAK), and dephosphorylation of Thr14 and Tyr15 by Cdc25 phosphatases. Phosphorylation of Thr14 and Tyr15 by wee1 and mik1 kinases as well as binding of Cdk inhibitors (CDKIs) result in Cdk inhibition. In this figure, the centrally placed canonical Cdk/cyclin complex is the driving force for cell cycle progression.

1643

CHAPTER 92 Differentiation and the Cell Cycle

B. Regulation of Cell Cycle Progression

Like p21Cip1, p27Kip1 inhibits the activity of the G1/S cyclin Cdk complexes. For instance p27Kip1 participates in G1 arrest produced by the exposure of fibroblasts derived from mink lung to the transforming growth factor β (TGFβ), and by cell-cell contact [150]. In actively dividing cells, p27Kip1 is phosphorylated by cyclin E-Cdk2 complex in the nucleus [151], and its abundance regulated by p45Skp2, also known as SCF/Skp2, which promotes ubiquitin-mediated degradation of p27Kip1 [152,153]. However, cytoplasmic control is also suggested by another, not yet identified ubiquitin ligase, that is present in the cytoplasm and is Thr187 phosphorylation-independent [154,155]. As mentioned above, the forkhead family of transcription factors can also up-regulate expression of p27Kip1 [125,126]. The inducers of p57Kip2 expression are unknown. Another family of regulatory peptides are the INK4 proteins, which include p16(INK4A), p15(INK4B), p18(INK4C), and p19(INK4D). These proteins specifically block cyclin D-Cdk 4/6 activity, leading to a G1 phase arrest [156]. The INK4 proteins inhibit Cdk4

1. THE G1 TO S PHASE TRANSITION

While the cyclin D-Cdk4/6 and cyclin E-Cdk2 complexes control the entry into the S phase, these complexes are in turn controlled by families of G1/S regulatory polypeptides, the CDKIs [136] (Fig. 4). Specifically, cyclin D, E, and A-dependent kinases are negatively regulated by a family of CDKIs that consists of p21Cip1, p27Kip1, and p57Kip2. Although all three of these inhibitors block progression through the G1 phase, each is usually activated by different stimuli. The expression of p21Cip1 can be under the transcriptional control of the p53 tumor suppressor gene, activated by DNA damage [147], but may also be independent of p53 [148]. The increase in p21Cip1 protein levels may lead to the inhibition of cyclin D-Cdk4/6 activity, which contribute to the G1 arrest. A second mechanism of action of p21Cip1 may be related to its ability to bind to the proliferating cell nuclear antigen (PCNA), a molecule involved in DNA replication and repair [149].

p53

Other signals

p21Cip1

Skp2

p27Kip1

Degradation

pRb-E2F CDK4,6

CDK2

INK4 Cyclin D

Cyclin E P

pRb

+ E2F

c-Myc

G1

ODC; Cdc25A; CycD1, E, A

S

FIGURE 4 Control of G1 to S phase transition by the pRb-E2F pathway. Phosphorylation of pRb by active Cdks releases E2F transcription factors, which activate, directly or indirectly, genes whose products are required for DNA replication [159,161]. The activity of Cdks can be controlled by factors shown in Fig. 3, and these include CDKIs belonging to the INK4 and Cip/Kip families. ODC, ornithine decarboxylase; Cyc, cyclin. Only a few examples of over 600 known c-Myc target genes are shown here; this growing list can be accessed at www.myccancergene.org.

1644 and 6 by preventing the binding of cyclin D, but also inhibit the activation of the formed Cdk 4/6-cyclin D complexes. Treatment of human keratinocytes with TGF-β results in an increase in p15(INK4B) expression and its association with Cdk4 and 6 [157]. While there is a large body of evidence showing the regulation of Cip/Kip CDKIs by 1,25(OH)2D3 and its analogs (see below), the control of INK proteins by deltanoids has not been established. An important event for the G1-S transition appears to be the phosphorylation of the tumor suppressor, pRb, as shown in Fig. 4 [158,159]. pRb, and other pRb-like “pocket” proteins (p130/Rb2,p107), are believed to control the entry into the S phase by interacting with a member of the E2F transcription factor family. This family is composed of at least five proteins (E2F 1–5), which are active when they form heterodimeric complexes with one of the E2F-related transcription factors, DP-1, 2, or 3. In its simplest form, the current hypothesis is that hypophosphorylated pRb binds to E2F, preventing cell entry into the S phase. Upon increased level of phosphorylation, pRb frees E2F, which in its heterodimeric complex with DP is capable of activating genes necessary for S phase initiation in part mediated by c-Myc/Max transcription factor [160–164]. However, the actual situation appears to be more complicated. Recent reports [165,166] suggest that gene repression by pRb also involves modulation of chromatin architecture. The proposed mechanism rests on the finding that histone deacetylase HDAC1 physically interacts with pRb through the “pocket” domain and recruits HDAC1 to E2F. The complex of these three proteins binds to E2F target promoters. HDAC1 may then facilitate the removal of highly charged acetyl groups from core histones, leading to a tight association between the nucleosomes, which prevents the access of transcription factors to their cognate elements in the gene promoters [156]. Mitogenic signals (e.g., growth factors, serum compounds) that stimulate cell progress through G1 coincide with increased expression of D cyclins. This extracellular regulation of cyclin D isoforms is not observed with other cyclin proteins. The current classical scenario is that cyclin D then forms a complex with Cdk4 and/or Cdk6, and its activity is regulated by the mechanisms described above. The activated cyclin DCdk 4/6 complexes then phosphorylate pRb and release E2F, or relieve the chromatin configuration constraints described above, leading to G1 progression. The cyclin D complex alone does not dictate control of progression through G1. Cyclin E is another G1 cyclin, which is synthesized later in the cell cycle than cyclin D, peaking at the late G1/S phase boundary. The expression of cyclin E is mitogen-independent, and

GEORGE P. STUDZINSKI AND MICHAEL DANILENKO

cyclin E forms an active complex with Cdk2. One level of regulation for the cyclin E-Cdk2 complex is through protein phosphatase Cdc25A, which cleaves the phosphate groups on the Thr14 and Tyr15 residues of Cdk2, and activates the Cdk2-cyclin E complex [167]. Cdc25A activity may in turn be regulated by phosphorylation by Chk1 [168]. c-Myc has been shown to regulate the expression and phosphorylation, and therefore the activity, of Cdc25A [169]. Cyclin D-Cdk4 and cyclin D-Cdk6 complexes are believed to trigger pRb phosphorylation, but cyclin E-Cdk2 complex also contributes to the phosphorylation of pRb in late G1, leading to cell entry into the S phase, where pRb phosphorylation is maintained by cyclin E/Cdk2, or cyclin A/Cdk2. 2. CONTROL OF DNA REPLICATION LICENSING

The S phase is the period where DNA replication takes place, and the basic machinery for this process is well conserved from yeast to mammals [170]. It is permitted by the rising level of Cdk activity, and is initiated on many sites on chromosomes, designated as “replication origins”, which can exist in two states [171]. In G1 phase, a multiprotein complex, the prereplicative complex (pre-RC), is assembled, but once DNA replication is initiated, the complex has fewer components (“post-RC”), which persists to the end of mitosis, and does not permit re-replication of DNA. At that time point proteolytic activity destroys the cyclins and other nuclear proteins, and Cdk activity becomes low. These two states of replication origins, separated by Cdk activity, generally ensure that the S phase and mitosis alternate. When chromatin becomes competent for DNA replication by the presence of pre-RC on the replication origin, it is considered to be “licensed.” Components of the pre-RC, the “initiator” proteins, include the Origin Recognition Complex (ORC), Cdc6/18, Cdt1 and Mini-Chromosome Maintenance (MCM) proteins. Although the DNA consensus sequences have not been found in organisms other than yeast, ORC proteins have been found in several eukaryotes, and all have ATP binding sites, consistent with the requirement of ATP for the initiation of DNA replication [172]. Cdc6/18 mammalian homologs may be related to M checkpoint control [173], while Cdt1 (Cdc10-dependent transcript 1), which also has peak protein levels at G1/S boundary [174], is associated with DNA replication checkpoint control. The six MCM proteins form a hexameric complex, approximately 600 kD in size, which may function as a replicative helicase [175]. Importantly, Cdt1 binds tightly to a DNA replication initiation inhibitor (Geminin) and this inhibits MCM loading. Licensing in G1 phase is permitted after the end of mitosis, when

1645

CHAPTER 92 Differentiation and the Cell Cycle

Geminin is destroyed by the Anaphase Promoting Complex (APC) – ubiquitin system [176]. Thus, the current model posits that ORC associates with the replication origins throughout the cell cycle, and when the cells exit mitosis Cdc6/18 and Cdt1 are loaded on chromatin, and in turn aid loading of MCM on the pre-RC complex, thus completing licensing [174]. The licensed complex can now be activated for DNA replication by a protein kinase, such as cyclin E/Cdk2 or Dbf4-dependent kinase (DDK) [177], and the DNA replicating machinery (e.g., Cdc45, replication protein A (RPA), DNA polymerase α and ε) is recruited to the initiation sites [178–181]. To further facilitate replication, a SCF-ubiquitination complex, which destroys Cdk inhibitors, can be recruited to the pre-RC by a cyclin-binding site on Cdc6/18 [182]. 3. THE G2 AND M PHASE TRANSITIONS

Once the cell has faithfully replicated its genome, the next cellular function is to segregate this DNA into equivalent, or nearly equivalent, daughter cells. The central regulation for the transition from G2 to mitosis is by the cyclin B-Cdk1 complex, initially called maturation (mitosis) promoting factor, or MPF [183]. In general, the activity of this complex is governed by factors similar to those responsible for the G1-S transition, including Cdk-cyclin association and activating phosphorylation by CAK (Fig. 5). The CDKIs were not known to play a major role in the control of the G2/M traverse, but studies in S. Reed’s laboratory indicate that the situation is more complex than previously believed [184,185]. Regulation of the Cdk1-cyclin B complex does include the G2/M specific phosphatase/kinase cell cycle regulatory proteins Cdc25C and perhaps Wee1-like tyrosine kinase. Cdc25C is a protein phosphatase, which cleaves the inhibitory phosphate groups at both Tyr15 and Thr14 on Cdk1 [186,187]. Cdc25C itself requires phosphorylation to be activated, and recent data support that Cdc25C is phosphorylated and activated by the cyclin B-Cdk1 complex, thus forming a positive feedback loop [188]. On the other hand, Wee1 and Mik1 phosphorylate these same sites, and thus act as inhibitor of the progression into mitosis [189]. In addition to the accurate duplication of the genetic material, cell cycle controls ensure its correct segregation into the daughter cells. This occurs at two levels: regulation of proteins that bind together the two chromatids, and control of centrosome duplication and spindle assembly. The sister chromatids adhere to one another by the adhesive properties of multi-submit complex composed of several proteins, so far best characterized in yeast, collectively called cohesin [190]. Cohesin is

Wee1

Mik1

Inactivating kinases

Cdc25C

Activating phosphatase

CAK Activating kinase

Chk2 Plk1

Y15 T14

T161

Cdk 1

G2

M Cyclin B

Ubiquitination

FIGURE 5 Control of G2 to M phase transition and the completion of mitosis. Cyclin B-Cdk1 complex is a central regulator of the transition from G2 to M. The activity of this complex is governed by factors similar to those responsible for the G1-S transition (see Fig. 4), including Cdk-cyclin association, activating phosphorylation of Thr161 as well as phosphorylation/ dephosphorylation of Thr14 and Tyr15. See text for additional details. CAK, cyclin activating kinase.

destroyed by proteolytic cleavage of one of its subunits, Scc1/Med1, by a calcium-activated cysteine protease, related to caspases, known as “separase” [191]. In animal cells the dissolution of cohesion between the chromatids occurs in two steps, one at prophase, the other at anaphase, and only the latter requires separase [192]. Interestingly, separase is subjected to multiple levels of regulation. These include its phosphorylation by cyclin B-Cdk1 [193], and the inhibition of separase catalytic activity by securin [194]. Since both cyclin B and securin are ubiquitinated by APC and destroyed at the end of mitosis, this ensures orderly and precisely timed separation of the chromosomes at telophase. Polo-like kinase (Plk-1) also regulates chromosome adhesion and other aspects of mitosis, including centrosome maturation and orientation [195]. The recent finding of co-localization of Plk1 and Chk2 suggests that there is a lateral communication between the mitotic checkpoint and the DNA integrity checkpoint [196]. The changes in cell cycle traverse and DNA replication that occur in numerous forms of differentiation have been previously reviewed [8]. Changes that specifically follow exposure to vitamin D derivatives have been less extensively studied, and with this background these will now be described.

1646

GEORGE P. STUDZINSKI AND MICHAEL DANILENKO

C. Modulation of Cell Cycle Events by Deltanoids The inhibition of cell cycle traverse by 1,25(OH)2D3 and analogs has been investigated in normal and malignant keratinocytes [197–200], and in many other types of tumor cells, with myeloid leukemia providing an excellent in vitro model system for this purpose [26,201]. 1. THE G1/S BLOCK

Deltanoids inhibit proliferation of diverse types of mammalian cells by arresting them in the G1/G0 phase of the cell cycle. While the exact sequence of events from VDR activation to G1/G0 arrest remains to be elucidated, and may not be exactly the same in all cell types, several pathways and cell cycle arrest effectors are already known to be involved. These are the up-regulation of protein levels of the Cdk inhibitors p21Cip1 and/or p27Kip1, upregulation of the retinoblastoma gene expression and reduced phosphorylation of pRb protein, and the inhibition of c-Myc expression [75,91,202–204]. a. Upregulation of p21Cip1 and p27Kip1 Elevated protein levels of the Cip/Kip family of CDKIs result

1,25D3 Skp2

Differentiation

?

p27Kip1 protein

p21Cip1 mRNA

pRb–E2F INK4

CDK4,6 Cyclin D

CDK2 Cyclin E pRb

E2F

c-Myc G1

FIGURE 6

S

Inhibitory effects of 1,25(OH)2D3 on the G1 traverse, which result in G1/S block. The key events in leukemia cells include: (1) An early up-regulation of RB gene expression, which sequesters E2F necessary for the expression of cyclin E; (2) Inhibition of p45Skp2 expression, which, together with reduced cyclin E levels and consequent reduced phosphorylation of p27Kip1, led to increased levels p27Kip1; (3) Progressively increasing sequestration of E2F resulting in further increase in p27Kip1 and decreases in c-Myc and other proteins required for DNA replication. The direct activation of p21Cip1 gene may have an effect on other pathways that lead to differentiation, or contribute to the effects of 1,25(OH)2D3 on the pRb/E2F pathway.

from the exposure to 1,25(OH)2D3 and other deltanoids in many cell types (Table II), and may be a near universal feature of the antiproliferative effects of these compounds. In view of the importance of Cdk activity as the driving force for cell cycle progression, it is not difficult to understand that increased levels of CDKIs can target Cdk2 complexed to cyclin D, E, or A, and grind the cell cycle traverse to a halt. However, the mechanisms of the CDKI up-regulation are not entirely clear, and there are subtle differences between the antiproliferative effects of p21Cip1 and p27Kip1, while p57Kip2 is not known to have a role in antiproliferative effects of deltanoids. Considerable excitement was generated when p21Cip1 was found to be up-regulated in a number of differentiation systems, including HL60 cells treated with 1,25(OH)2D3 [148,205]. It was suggested that p21Cip1, and/or p27Kip1, not only promote the G1 arrest but also contribute to differentiation [47,206]. It seems, however, that while these Cdk inhibitors may not be solely responsible for the G1 block, the data regarding their role in differentiation are conflicting. For instance, mice lacking p21Cip1 undergo normal development [207], even though p21Cip1−/− embryonic fibroblasts show impaired arrest in G1 in response to DNA damage. An imbalance between growth and differentiation can be demonstrated in these cells, and in other in vitro cell differentiation systems with p21Cip1 knockouts. Keratinocytes which are p21/Cip1−/−, and to a lesser extent those with p27Kip1 knockouts, have an increased proliferative potential [208]. With regard to differentiation, however, Harvat [209] showed that growth arrest resulting from overexpression of p21Cip1 in mouse primary keratinocytes is not sufficient to induce the expression of markers of differentiation. Further, in malignant counterparts of these cells, the squamous cell carcinoma (SCC) cells, not even growth arrest is clearly linked to p21Cip1, as 1,25(OH)2D3 inhibited growth but reduced p21Cip1 levels in vitro and in SCC tumors [210]. In another system, the myelomonocytic cell line U937, Freedman’s group noted transcriptional activation of the p21Cip1 gene by 1,25(OH)2D3, and suggested that this is linked to differentiation of these leukemia cells [47]. Importantly, they identified a functional vitamin D response element (VDRE) in the promoter of the p21Cip1 gene, and noted that the p21Cip1 transcript can be detected as early as 2 h after 1,25(OH)2D3 addition, consistent with p21Cip1 being a direct mediator of 1,25(OH)2D3 action. However, in this system the up-regulation of p21Cip1 after exposure to 1,25(OH)2D3 is transient and accompanied by a proliferative burst [14], which does not correlate with the onset of the G1 block that is observed 24–48 h

1647

CHAPTER 92 Differentiation and the Cell Cycle

TABLE II CDKI

Examples of Up-Regulation of CDKI Levels by 1,25(OH)2D3 and Deltanoids

Cell type

Functional effect/comment

References

Hematopoietic cells p21Cip1 HL60 p21Cip1 U937 p21Cip1 U937 p21Cip1 U937

Immediate early gene induced during monocytic differentiation/G1 arrest Transcriptional activation of p21/induces monocytic differentiation Cytoplasmic localization/anti-apoptotic effect Antisense to p21 decreases differentiation

[148,205,314,315] [47] [316] [317]

p27Kip1 p27Kip1

HL60 HL60

Proliferation block/G1 arrest Antisense to p27 reverses G1 arrest

[91,204,318,319] [201]

p27Kip1 p27Kip1

U937 U937

Proliferation block Vitamin D receptor-independent upregulation of p27 gene

[319] [124]

p21/p27 p21/p27 p21/p27

HL60 U937 UF-1

G1 arrest/little change in CDK4 activity Early proliferative burst followed by growth arrest and differentiation Granulocytic differentiation/proliferation block/G1 arrest

[320] [14] [220]

Prostate cancer cells p21Cip1 LNCaP p21Cip1 ALVA-31 p21Cip1 PC-3 p21Cip1 DU-145

Proliferation block/G1 arrest Proliferation block/G1 arrest Proliferation block/differentiation Proliferation block

[211–213] [213] [290] [290]

p27Kip1

LNCaP

Increased association of p27 with Cdk2/Stabilization of p27/G1 arrest

[321,322]

p21/p27

LNCaP

Proliferation block/differentiation

[290]

Breast cancer cells p21Cip1 MCF-7 p21Cip1 MCF-7E p21Cip1 MCF7/LCC2

Proliferation block/G1 arrest/apoptosis Proliferation block/G1 arrest Proliferation block/G1 arrest/transient upregulation of com1

[323,324] [214] [325]

p27Kip1 p27Kip1

MCF-7 SK-BR-3

Proliferation block/G1 arrest Proliferation block

[326] [326]

p21/p27 p21/p27 p21/p27 p21/p27

MCF-7 BT20 ZR75 SUM-159PT

Proliferation block/G1 arrest/downregulation of c-Myc Proliferation block/G1 arrest Proliferation block/G1 arrest Proliferation block/Apoptosis

[215,216,296,327,328] [214] [214] [329]

Pancreatic cancer cells p21/p27 BxPC-3 p21/p27 Hs 700T p21/p27 SUP-1

Early transient CDKI upregulation/proliferation block/G1 arrest Early transient CDKI upregulation/proliferation block/G1 arrest Early transient CDKI upregulation/proliferation block/G1 arrest

[218] [218] [218]

Insulinoma cells p21Cip1 Beta TC(3)

G1 arrest/apoptosis

[330]

Colon cancer cells p21Cip1 HT-29

Proliferation block/G1 arrest/apoptosis

[331]

p21/p27

Co-association between Cdk2, p27Kip1 and cyclin E/proliferation block

[332–334]

Caco-2

(Continued)

1648

GEORGE P. STUDZINSKI AND MICHAEL DANILENKO

TABLE II CDKI

Examples of Up-Regulation of CDKI Levels by 1,25(OH)2D3 and Deltanoids— Cont’d

Cell type

Neuroblastoma cells p21Cip1 SH-SY5Y p21Cip1 NB69 p21Cip1 SK-N-AS p21Cip1 IMR5 p21Cip1 CHP134 p21Cip1 NGP

Functional effect/comment Downregulation of Myc and Id2/induction of RARβ/proliferation block Downregulation of Myc and Id2/induction of RARβ/proliferation block Downregulation of Myc/induction of RARβ/proliferation block Downregulation of Myc and Id2/induction of RARβ/proliferation block Downregulation of Id2/induction of RARβ/proliferation block Downregulation of Myc and Id2/induction of RARβ/proliferation block

later, and is accompanied by markedly increased levels of p27Kip1 [47,91,204]. Thus, although p21Cip1 may initiate a cascade of unknown events that lead to the expression of the differentiated monocytic phenotype, it is unlikely to be directly responsible for the G1 arrest in leukemia, or SCC, cells. Such function, however, has been attributed to p21Cip1 in other cells, including prostate cancer [211–213], breast cancer [214–216], and parathyroid cells [217]. The role of binding of p21Cip1 to PCNA [149], the processive factor for DNA replication, remains to be elucidated. The first demonstration that up-regulation of 27Kip1 is associated with the G1 arrest which takes place following 1,25(OH)2D3–induced differentiation was reported by Wang et al. in 1996 [91]. They showed a sustained increase in p27Kip1 protein abundance that coincided with the appearance of the 1,25(OH)2D3– induced G1 block in HL60 cells, and correlated with reduced kinase activity of Cdk6 and Cdk2 [91,204]. Further, reductions of the levels of p27Kip1 by several independent approaches reversed the G1 block, but not the differentiated phenotype [201]. Accordingly, these data clearly show that, at least in HL60 cells, p27Kip controls the 1,25(OH)2D3–induced G1 block, but not the differentiated phenotype. Similar findings have been obtained in several other systems, although the data cannot always be so clearly interpreted. For instance, the up-regulation of p27Kip1 is often accompanied by an up-regulation of p21Cip1 [214,218,219]. However, even in situations where p21Cip1 and p27Kip1 are both up-regulated by 1,25(OH)2D3 or other deltanoids, increased levels of p27Kip1 correlate better with the onset of G1 block than the up-regulation of p21Cip1 [47,91,210,220,221]. This, however, is subject to cell context, a striking example being a recent report that p27Kip1 is essential for the antiproliferative action of 1,25(OH)2D3 on primary, but not on immortalized, mouse embryonic fibroblasts [222]. The role of CDKIs in 1,25(OH)2D3– induced cell cycle arrest is also difficult to assess

References [335,336] [335,336] [335,336] [335,336] [335,336] [335,336]

because, when present at relatively low levels, p21Cip1 and p27Kip1 serve to facilitate complex formation of cyclins D with Cdks, and their transport to the nucleus [223,224], and only high levels of CDKIs are inhibitory [223,225,226]. Thus, one possible explanation for the up-regulation of p21Cip1 that does not correlate with G1 arrest is that p21Cip1 simply serves to facilitate cyclin D-Cdk complex formation. More likely, however, is that elevated levels of p21Cip1 inhibit cyclin E-Cdk2 activity and block cyclin E-Cdk2 phosphorylation of p27Kip1, which leads to its degradation in proliferating cells [151,227,228]. Unlike p21Cip1, which can be directly up-regulated by 1,25(OH)2D3 through a VDRE in p21Cip1 promoter [47], p27Kip1 has no VDR–binding element in its promoter, and may be regulated at both transcriptional and post-transcriptional levels, although control by protein degradation appears to be most important. While the precise mechanism of p27Kip1 up-regulation by 1,25(OH)2D3 is currently not known, several reports focus on this question. One study showed that transcription factors Sp1 and NF-Y can synergistically mediate the 1,25(OH)2D3–induced expression of p27Kip1 in transiently transfected U937 leukemia cells [124]. In these experiments, deletion and mutational analysis revealed that p27Kip1 promoter activation required both GGGCGG (Spl binding) and a CCAAT (NF-Y binding) sequences. As presented above (Section II.D.2.), Sp1 transcription factor is activated in 1,25(OH)2D3–treated leukemia cells, so this could potentially be a plausible mechanism for the induction of G1 arrest by 1,25(OH)2D3, at least in myeloid leukemia cells. However, the difficulty in accepting this scenario is that p27Kip1 mRNA levels are not found to be appreciably increased following treatment with 1,25(OH)2D3 [47,229], and that in a number of cell types p27Kip1 is transcriptionally activated by the forkhead transcription factors, such as AFX (FOX04) [125,126]. More clearly related to 1,25(OH)2D3– induced G1 arrest is the recent report that a deltanoid,

CHAPTER 92 Differentiation and the Cell Cycle

EB1089, inhibits the expression of the F-box protein p45Skp2 and thus prevents its degradation by the proteasome system [230]. In mouse squamous cell carcinoma AT-84 cells EB1089 did not change p27Kip1 mRNA levels, but reduced the mRNAs for p45Skp2, which ubiquitinates p27Kip1, and for Cks1, which targets p45Skp2 to p27Kip1 [230,231]. A similar decrease in p45Skp2 expression and stabilization of p27Kip1 protein was demonstrated in acute promyelocytic leukemia cells [230]. Since these changes become evident at about 48 h of the exposure to the deltanoid, there is good correlation with the onset of the G1 block. The latent period of 24–48 h for p27Kip1 up-regulation may also be needed to inactivate the cyclin E-Cdk2 complex, which phosphorylates Thr187 of p27Kip1, that is required for ubiquitination of p27Kip2 by p45Skp2 [151]. Inhibition of cyclin E-Cdk2 activity following exposure of HL60 cells to 1,25(OH)2D3 has been demonstrated [204], and this may contribute to 1,25(OH)2D3-induced increase in p27Kip1 levels. b. Retinoblastoma Protein Control of DeltanoidInduced G1 Block The suggested placement of the inactivation of the cyclin E-Cdk2 complex upstream of p27Kip1 up-regulation raises the question of how this complex is inactivated in 1,25(OH)2D3-treated cells. One possible answer is provided by the finding that the retinoblastoma (RB) gene is up-regulated early in deltanoid-induced differentiation of HL60 cells [75,98]. The increased levels of pRb can then bind and inactivate E2F transcription factors necessary for the expression of cyclin E, and thus the activity of the cyclin E-Cdk2 complex. Accordingly, the phosphorylation of T187 on p27Kip1 is reduced, allowing the accumulation of this Cdk inhibitor, and a further increase in the hypophosphorylated forms of pRb, also a substrate for the cyclin E-Cdk complex. Hypophosphorylated pRb now further binds E2F, and thus reduces cyclin E expression to the point that p27Kip1 is no longer phosphorylated and degraded, as the result of this positive feedback loop, leading to G1 arrest. It is known that in HL60 cells the expression of pRb normally occurs primarily during G1 phase [232], and can be detected at both mRNA and protein levels within 8–12 h of exposure to 1,25(OH)2D3 [98], although the mechanism of its up-regulation remains to be determined. These in vitro studies are supported by the finding of gross defects in the development of the hematopoietic system in RB knockout mice [233,234], and by the transcriptional studies which show that the RB gene plays a role in normal human adult hematopoiesis [235]. Thus, pRb appears to have a role in the early stage of 1,25(OH)2D3–induced differentiation, and contributes to changes in cellular transcriptional and kinase activities that lead to G1 arrest at a later stage.

1649 c. Down-regulation of c-Myc Expression in Deltanoid-induced Differentiation and G1 Arrest The pRb/E2F pathway also controls the expression of c-Myc, as E2F transcription factors have been reported to up-regulate the c-Myc gene, [236–238] and inhibition of c-Myc expression by 1,25(OH)2D3 may be responsible, at least in part, for the G1 block in differentiating cells (see Figs. 4 and 6). Indeed, the association between c-Myc down-regulation and 1,25(OH)2D3-induced differentiation of human leukemia cells was one of the earliest findings that initiated the studies of the molecular basis of the cellular changes that follow exposure to this hormone [16,202,239,240]. The intense interest in c-Myc as a potential negative regulator of differentiation was fueled largely by its deregulated expression in several types of human neoplasia [241,242]. c-Myc is known to promote cell cycle progression mostly through coordinated transcriptional regulation of target genes [243] (www.myccancergene.org). These include the DNA replication and cell cycle traverse-promoting genes such as ornithine decarboxylase, Cdc25A, and cyclins E and A [241]. Conversely, c-Myc inhibits the transcription of cell cycle inhibitor p21Cip1 [244], and it has been suggested that this is due, at least in part, to the sequestration of the Sp1 transcription factor, which is required for p21Cip1 transcription [245,246]. While the above considerations present an almost complete sequence of events that can explain the G1 arrest induced by 1,25(OH)2D3 (summarized in Fig. 6), an additional level of control of c-Myc expression by 1,25(OH)2D3 is provided by studies of Simpson et al. [247]. They found that in differentiating HL60 cells 1,25(OH)2D3 increased the expression and DNA– binding activity of HOXB4, a product of a homeobox gene, and that HOXB4 binds to the sites in the c-Myc gene, which are involved in blocking by 1,25(OH)2D3 of the elongation of c-Myc transcripts [248]. Further, these authors demonstrated that a HOXB4 antisense oligonucleotide partially inhibited the 1,25(OH)2D3– induced decrease in c-Myc protein levels [249]. While they observed reduction of 1,25(OH)2D3-induced differentiation in these experiments, the effect of HOXB4 antisense on the G1 block was not reported. Nonetheless, these studies are significant, as members of the HOX gene family are known to be involved in hematopoiesis and leukemogenesis [250–252]. Also, other HOX genes participate in 1,25(OH)2D3–induced differentiation; HOX B7 was reported to increase in HL60 cells [253], while in U937 and MCF-7 cells 1,25(OH)2D3 increased expression of HOXA10 [254]. Transcriptional blockage of c-Myc expression was also reported in colon cancer cells following exposure to 1,25(OH)2D3 [255]. In these cells, c-Myc is principally under the control of APC-β catenin/T-cell

1650 factor signaling. Thus, it is possible that 1,25(OH)2D3 regulates c-Myc expression by several different pathways, but all of these appear to exert control on the c-Myc gene at transcriptional level. 2. THE G2/M RETARDATION AND POLYPLOIDIZATION

The occurrence of abnormalities in G2/M transition in 1,25(OH)2D3 and other deltanoid-treated cells has been observed infrequently, with a general consensus that the G1 phase is the principal target of the antiproliferative actions of deltanoids. However, in early studies of 1,25(OH)2D3 action Abe et al. [256] detected an increase in the G2+ M compartment in WEHI murine myelomonocytic cells, also described in HL60 cells by Godyn et al. [257]. The basis for this increase may be a reduction in the levels of Cdk1 in these cells [258], although the roles of cohesin, separase, or Plks remain to be investigated in the light of the recently accumulating knowledge of mitotic controls (see Section III.B.3). One consequence is the higher ploidy of HL60 cells exposed for prolonged periods of time to 1,25(OH)2D3, observed as an increased number of binucleated cells [257], or as nearly doubled DNA content of these cells [113]. Interestingly, polyploidization of 1,25(OH)2D3-treated cells is an alternative to differentiation, as these cells over-ride the antiproliferative actions of 1,25(OH)2D3 and do not express the differentiated phenotype [113]. Thus, HL60 cells can have a 1,25(OH)2D3–induced defect in completion of mitosis that allows one round of DNA endoreduplication. Whether osteoclast, or perhaps megakaryocyte, polyploidization is also influenced by 1,25(OH)2D3 remains a possibility.

IV. CELL-TYPE SPECIFICITY OF INHIBITION OF CELL PROLIFERATION BY DELTANOIDS WITHOUT EVIDENCE OF DIFFERENTIATION While some effects of 1,25(OH)2D3 and other deltanoids can be recognized in a variety of cell types, there is remarkable cell-type specificity in most of such effects, and it is important to realize that only a few generalizations can be made regarding the antiproliferative actions of these compounds. However, it appears to be true that deltanoid-induced differentiation is not a consequence of inhibited proliferation, as differentiation often precedes the G1 block [9,12], and deltanoids can inhibit cell proliferation with only minimal, or absent, evidence of differentiation. Indeed, the antiproliferative effect 1,25(OH)2D3 on cultured melanoma cells was recognized by Colston et al. [259] in 1981, at

GEORGE P. STUDZINSKI AND MICHAEL DANILENKO

the same time as the differentiation-inducing action of 1,25(OH)2D3 was described in myeloid leukemia cells by Abe et al. [15]. Cell specificity of responses to deltanoids is also illustrated by the finding that 1,25(OH)2D3 can cause a G1 block in cultured thyroid carcinoma and pituitary corticotroph, but not lactotroph, cells [260,261]. Interestingly, while in thyroid carcinoma cells the mechanisms of the antiproliferative effects include dephosphorylation of p27Kip1 in a PTEN-dependent manner, leading to a diminished association between p45Skp2 and p27Kip1 with its consequent accumulation [260], in pituitary corticotroph cells the mechanism appears to be a diminished association of p27Kip1 with p45Skp2 and Cdk2, without an involvement of PTEN [261]. This illustrates not only the exquisite cell-type specificity of the mechanisms involved in the antiproliferative actions of deltanoids, but also that the up-regulation of p27Kip1 is unrelated to differentiation, as demonstrated previously in leukemia cells [201]. Another mechanism for the antiproliferative actions of deltanoids on cell types that show only minimal evidence of differentiation is provided by the apoptosisinducing actions of deltanoids, as described in other chapters in this volume (e.g., Chapter 93). Again, cell type determines this response, as in contrast to various carcinomas, e.g., breast cancer cells [262], 1,25(OH)2D3 protects HL60 leukemia cells from apoptosis, as first demonstrated by Xu et al. [263]. Thus, the activity of survival pathways may determine whether differentiation can take place in the presence of deltanoids, or whether a potential default pathway will lead to apoptosis, perhaps as the result of prolonged residence in a cell cycle compartment other than G0. In any case, deltanoids can be effective antiproliferative agents in many cell types that express VDR.

V. CONCLUSIONS Deltanoids present new therapeutic options for treatment of human malignancies due to their demonstrated antiproliferative actions in a wide variety of cell types. While the mechanisms vary, G1 block produced by up-regulation of p27Kip1 is an almost constant feature of the cell cycle effects of deltanoids. Further studies are needed on the mechanisms that upregulate p27Kip1, although the control of its degradation by p45Skp2 that is influenced by p27Kip1-T187 phosphorylation by the E2F-cyclin E-Cdk2 pathway, and inhibition of this pathway by pRb, present exciting possibilities. Also, in view of the ability of cells treated with 1,25(OH)2D3 to develop resistance to its antiproliferative actions [264,265], synergistic effects of

1651

CHAPTER 92 Differentiation and the Cell Cycle

deltanoids combined with other agents should be further explored [266–271], as should the pathways that transmit other extracellular signals to the cell nucleus in concert with vitamin D receptor-initiated signals.

14.

Acknowledgments

15.

We thank Dr. Robert Murray for comments on the manuscript and Ms. Terri McNeil for expert secretarial assistance. We also acknowledge the support for experimental work performed in our laboratories by the National Cancer Institute (Grant RO1-CA44722 to GPS) and the Israel-US BiNational Science Foundation (Grant to MD and GPS).

16.

17. 18.

References 1. Reed SI 1997 Control of the G1/S transition. Cancer Surv 29:7–23. 2. Coffman FD, Studzinski GP 1999 Differentiation-related mechanisms which suppress DNA replication. Exp Cell Res 248:58–73. 3. Sherr CJ 2002 D1 in G2. Cell Cycle 1:36–38. 4. Vermeulen K, Van Bockstaele DR, Berneman ZN 2003 The cell cycle: a review of regulation, deregulation, and therapeutic targets in cancer. Cell Prolif 36:131–149. 5. Cheng M, Sexl V, Sherr CJ, Roussel MF 1998 Assembly of cyclin D–dependent kinase and titration of p27Kip1 regulated by mitogen-activated protein kinase kinase (MEK1). Proc Natl Acad Sci USA 95:1091–1096. 6. Stein GS, Lian JB, Owen TA 1990 Relationship of cell growth to the regulation of tissue-specific gene expression during osteoblast differentiation. FASEB J 4:3111–3123. 7. Lian JB, Stein GS, Bortell R, Owen TA 1991 Phenotype suppression: a postulated molecular mechanism for mediating the relationship of proliferation and differentiation by Fos/Jun interactions at AP-1 sites in steroid responsive promoter elements of tissue-specific genes. J Cell Biochem 45:9–14. 8. Studzinski GP, Harrison LE 1999 Differentiation-related changes in the cell cycle traverse. Int Rev Cytol 189:1–58. 9. Zhang F, Godyn JJ, Uskokovic M, Binderup L, Studzinski GP 1994 Monocytic differentiation of HL60 cells induced by potent analogs of vitamin D3 precedes the G1/G0 phase cell cycle block. Cell Prolif 27:643–654. 10. Studzinski GP, Rathod B, Wang QM, Rao J, Zhang F 1997 Uncoupling of cell cycle arrest from the expression of monocytic differentiation markers in HL60 cell variants. Exp Cell Res 232:376–387. 11. Brown G, Choudhry MA, Durham J, Drayson MT, Michell RH 1999 Monocytically differentiating HL60 cells proliferate rapidly before they mature. Exp Cell Res 253:511–518. 12. Drayson MT, Michell RH, Durham J, Brown G 2001 Cell proliferation and CD11b expression are controlled independently during HL60 cell differentiation initiated by 1,25 alpha dihydroxyvitamin D3 or all-trans-retinoic acid. Exp Cell Res 266:126–134. 13. Wang X, Studzinski GP 2001 Activation of extracellular signal-regulated kinases (ERKs) defines the first phase of

19.

20.

21.

22.

23.

24.

25.

26.

27. 28. 29.

1,25-dihydroxyvitamin D3-induced differentiation of HL60 cells. J Cell Biochem 80:471–482. Rots NY, Iavarone A, Bromleigh V, Freedman LP 1999 Induced differentiation of U937 cells by 1,25-dihydroxyvitamin D3 involves cell cycle arrest in G1 that is preceded by a transient proliferative burst and an increase in cyclin expression. Blood 93:2721–2729. Abe E, Miyaura C, Sakagami H, Takeda M, Konno K, Yamazaki T, Yoshiki S, Suda T 1981 Differentiation of mouse myeloid leukemia cells induced by 1 alpha,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 78:4990–4994. Studzinski GP, Bhandal AK, Brelvi ZS 1985 A system for monocytic differentiation of leukemic cells HL 60 by a short exposure to 1,25-dihydroxycholecalciferol. Proc Soc Exp Biol Med 179:288–295. Suda T 1989 The role of 1 alpha,25-dihydroxyvitamin D3 in the myeloid cell differentiation. Proc Soc Exp Biol Med 191: 214–220. Norman AW, Zhou JY, Henry HL, Uskokovic MR, Koeffler HP 1990 Structure-function studies on analogs of 1alpha,25dihydroxyvitamin D3: differential effects on leukemic cell growth, differentiation, and intestinal calcium absorption. Cancer Res 50:6857–6864. Studzinski GP, McLane JA, Uskokovic MR 1993 Signaling pathways for vitamin D–induced differentiation: implications for therapy of proliferative and neoplastic diseases. Crit Rev Eukaryot Gene Expr 3:279–312. Pan P, Reddy K, Lee S, Studzinski GP 1991 Differentiationrelated regulation of 1,25-dihydroxyvitamin D3 receptor mRNA in human leukemia cells HL-60. Cell Prolif 24: 159–170. Halline AG, Davidson NO, Skarosi SF, Sitrin MD, Tietze C, Alpers DH, Brasitus TA 1994 Effects of 1,25-dihydroxyvitamin D3 on proliferation and differentiation of Caco-2 cells. Endocrinology 134:1710–1717. Solvsten H, Svendsen ML, Fogh K, Kragballe K 1997 Up-regulation of vitamin D receptor levels by 1,25(OH)2 vitamin D3 in cultured human keratinocytes. Arch Dermatol Res 289:367–372. Escaleira MT, Brentani MM 1999 Vitamin D3 receptor (VDR) expression in HC-11 mammary cells: regulation by growth-modulatory agents, differentiation, and Ha-ras transformation. Breast Cancer Res Treat 54:123–133. Li J, Finch RA, Sartorelli AC 1999 Role of vitamin D3 receptor in the synergistic differentiation of WEHI-3B leukemia cells by vitamin D3 and retinoic acid. Exp Cell Res 249: 279–290. O’Kelly J, Hisatake J, Hisatake Y, Bishop J, Norman A, Koeffler HP 2002 Normal myelopoiesis but abnormal T lymphocyte responses in vitamin D receptor knockout mice. J Clin Invest 109:1091–1099. Kumagai T, O’Kelly J, Said JW, Koeffler HP 2003 Vitamin D2 analog 19-nor-1,25-dihydroxyvitamin D2: antitumor activity against leukemia, myeloma, and colon cancer cells. J Natl Cancer Inst 95:896–905. Zinser G, Packman K, Welsh J 2002 Vitamin D3 receptor ablation alters mammary gland morphogenesis. Development 129:3067–3076. Barsony J, Prufer K 2002 Vitamin D receptor and retinoid X receptor interactions in motion. Vitam Horm 65:345–376. Nemere I, Norman AW 1986 Parathyroid hormone stimulates calcium transport in perfused duodena from normal chicks: comparison with the rapid (transcaltachic) effect of 1,25dihydroxyvitamin D3. Endocrinology 119:1406–1408.

1652 30. Nemere I, Norman AW 1987 The rapid, hormonally stimulated transport of calcium (transcaltachia). J Bone Miner Res 2: 167–169. 31. Berry DM, Antochi R, Bhatia M, Meckling-Gill KA 1996 1,25-dihydroxyvitamin D3 stimulates expression and translocation of protein kinase Calpha and Cdelta via a nongenomic mechanism and rapidly induces phosphorylation of a 33-kDa protein in acute promyelocytic NB4 cells. J Biol Chem 271:16090–16096. 32. Ji Y, Wang X, Donnelly RJ, Uskokovic MR, Studzinski GP 2002 Signaling of monocytic differentiation by a nonhypercalcemic analog of vitamin D3, 1,25(OH)2–5,6 trans-16-enevitamin D3, involves nuclear vitamin D receptor (nVDR) and non-nVDR-mediated pathways. J Cell Physiol 191:198–207. 33. Wali RK, Kong J, Sitrin MD, Bissonnette M, Li YC 2003 Vitamin D receptor is not required for the rapid actions of 1,25-dihydroxyvitamin D3 to increase intracellular calcium and activate protein kinase C in mouse osteoblasts. J Cell Biochem 88:794–801. 34. Haussler MR, Haussler CA, Jurutka PW, Thompson PD, Hsieh JC, Remus LS, Selznick SH, Whitfield GK 1997 The vitamin D hormone and its nuclear receptor: molecular actions and disease states. J Endocrinol 154(Suppl):S57–S73. 35. Carlberg C 2003 Molecular basis of the selective activity of vitamin D analogs. J Cell Biochem 88:274–281. 36. Rachez C, Freedman LP 2000 Mechanisms of gene regulation by vitamin D3 receptor: a network of coactivator interactions. Gene 246:9–21. 37. MacDonald PN, Baudino TA, Tokumaru H, Dowd DR, Zhang C 2001 Vitamin D receptor and nuclear receptor coactivators: crucial interactions in vitamin D-mediated transcription. Steroids 66:171–176. 38. Lee JW, Lee YC, Na SY, Jung DJ, Lee SK 2001 Transcriptional coregulators of the nuclear receptor superfamily: coactivators and corepressors. Cell Mol Life Sci 58:289–297. 39. Bikle DD, Tu CL, Xie Z, Oda Y 2003 Vitamin D regulated keratinocyte differentiation: role of coactivators. J Cell Biochem 88:290–295. 40. Tagami T, Lutz WH, Kumar R, Jameson JL 1998 The interaction of the vitamin D receptor with nuclear receptor corepressors and coactivators. Biochem Biophys Res Commun 253:358–363. 41. Banwell CM, Singh R, Stewart PM, Uskokovic MR, Campbell MJ 2003 Antiproliferative signaling by 1,25(OH)2D3 in prostate and breast cancer is suppressed by a mechanism involving histone deacetylation. Recent Results Cancer Res 164:83–98. 42. Ueki N, Hayman MJ 2003 Signal-dependent N-CoR requirement for repression by the Ski oncoprotein. J Biol Chem 278:24858–24864. 43. Mackey SL, Heymont JL, Kronenberg HM, Demay MB 1996 Vitamin D receptor binding to the negative human parathyroid hormone vitamin D response element does not require the retinoid x receptor. Mol Endocrinol 10:298–305. 44. Russell J, Ashok S, Koszewski NJ 1999 Vitamin D receptor interactions with the rat parathyroid hormone gene: synergistic effects between two negative vitamin D response elements. J Bone Miner Res 14:1828–1837. 45. Ward JO, McConnell MJ, Carlile GW, Pandolfi PP, Licht JD, Freedman LP 2001 The acute promyelocytic leukemiaassociated protein, promyelocytic leukemia zinc finger, regulates 1,25-dihydroxyvitamin D3–induced monocytic differentiation of U937 cells through a physical interaction with vitamin D3 receptor. Blood 98:3290–3300. 46. Puccetti E, Obradovic D, Beissert T, Bianchini A, Washburn B, Chiaradonna F, Boehrer S, Hoelzer D, Ottmann OG, Pelicci PG,

GEORGE P. STUDZINSKI AND MICHAEL DANILENKO

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

Nervi C, Ruthardt M 2002 AML-associated translocation products block vitamin D3–induced differentiation by sequestering the vitamin D3 receptor. Cancer Res 62:7050–7058. Liu M, Lee MH, Cohen M, Bommakanti M, Freedman LP 1996 Transcriptional activation of the Cdk inhibitor p21 by vitamin D3 leads to the induced differentiation of the myelomonocytic cell line U937. Genes Dev 10:142–153. Hmama Z, Nandan D, Sly L, Knutson KL, Herrera-Velit P, Reiner NE 1999 1alpha,25-dihydroxyvitamin D3–induced myeloid cell differentiation is regulated by a vitamin D receptorphosphatidylinositol 3-kinase signaling complex. J Exp Med 190:1583–1594. Capiati DA, Tellez-Inon MT, Boland RL 1999 Participation of protein kinase C alpha in 1,25-dihydroxy-vitamin D3 regulation of chick myoblast proliferation and differentiation. Mol Cell Endocrinol 153:39–45. Wang X, Studzinski GP 2001 Phosphorylation of raf-1 by kinase suppressor of ras is inhibited by “MEK-specific” inhibitors PD 098059 and U0126 in differentiating HL60 cells. Exp Cell Res 268:294–300. Shimizu T, Taira N, Senou M, Takeda K 2002 Involvement of diverse protein kinase C isoforms in the differentiation of ML-1 human myeloblastic leukemia cells induced by the vitamin D3 analog KH1060 and the phorbol ester TPA. Cancer Lett 186:67–74. Chae HJ, Jeong BJ, Ha MS, Lee JK, Byun JO, Jung WY, Yun YG, Lee DG, Oh SH, Chae SW, Kwak YG, Kim HH, Lee ZH, Kim HR 2002 ERK MAP Kinase is required in 1,25(OH)2D3–induced differentiation in human osteoblasts. Immunopharmacol Immunotoxicol 24:31–41. Wang Q, Wang X, Studzinski GP 2003 Jun N-terminal kinase pathway enhances signaling of monocytic differentiation of human leukemia cells induced by 1,25-dihydroxyvitamin D3. J Cell Biochem 89:1087–1101. Johansen C, Kragballe K, Henningsen J, Westergaard M, Kristiansen K, Iversen L 2003 1alpha,25-dihydroxyvitamin D3 stimulates activator protein 1 DNA-binding activity by a phosphatidylinositol 3-kinase/Ras/MEK/extracellular signal regulated kinase 1/2 and c-Jun N-terminal kinase 1-dependent increase in c-Fos, Fra1, and c-Jun expression in human keratinocytes. J Invest Dermatol 120:561–570. Zylber-Katz E, Glazer RI 1985 Phospholipid- and Ca2+dependent protein kinase activity and protein phosphorylation patterns in the differentiation of human promyelocytic leukemia cell line HL-60. Cancer Res 45:5159–5164. Todd RF 3rd, Bury MJ, Alvarez PA, Brott DA, Liu DY 1986 Regulation of human monocyte surface antigen expression. I. Up-modulation of Mo3e antigen expression on U-937 and HL-60 cells stimulated by pharmacologic activators of protein kinase C. Blood 68:1154–1161. Ways DK, Dodd RC, Bennett TE, Gray TK, Earp HS 1987 1,25-dihydroxyvitamin D3 enhances phorbol ester-stimulated differentiation and protein kinase C-dependent substrate phosphorylation activity in the U937 human monoblastoid cell. Endocrinology 121:1654–1661. Martell RE, Simpson RU, Hsu T 1988 Effects of protein kinase inhibitors 1(5-isoquinolinesulfonyl)-2-methylpiperazine dihydrochloride (H-7) and N-[2-guanidinoethyl]5-isoquinolinesulfonamide hydrochloride (HA1004) on calcitriol-induced differentiation of HL-60 cells. Biochem Pharmacol 37:635–640. Solomon DH, O’Driscoll K, Sosne G, Weinstein IB, Cayre YE 1991 1 alpha,25–dihydroxyvitamin D3–induced regulation of protein kinase C gene expression during HL-60 cell differentiation. Cell Growth Differ 2:187–194.

1653

CHAPTER 92 Differentiation and the Cell Cycle

60. Martell RE, Simpson RU, Taylor JM 1987 1,25-Dihydroxyvitamin D3 regulation of phorbol ester receptors in HL-60 leukemia cells. J Biol Chem 262:5570–5575. 61. Obeid LM, Okazaki T, Karolak LA, Hannun YA 1990 Transcriptional regulation of protein kinase C by 1,25-dihydroxyvitamin D3 in HL-60 cells. J Biol Chem 265:2370–2374. 62. Gamard CJ, Blobe GC, Hannun YA, Obeid LM 1994 Specific role for protein kinase C beta in cell differentiation. Cell Growth Differ 5:405–409. 63. Bollinger Bollag W, Bollag RJ 2001 1,25-dihydroxyvitamin D3, phospholipase D, and protein kinase C in keratinocyte differentiation. Mol Cell Endocrinol 177:173–182. 64. Wali RK, Baum CL, Sitrin MD, Bolt MJ, Dudeja PK, Brasitus TA 1992 Effect of vitamin D status on the rapid actions of 1,25-dihydroxycholecalciferol in rat colonic membranes. Am J Physiol 262:G945–G953. 65. Wali RK, Baum CL, Bolt MJ, Brasitus TA, Sitrin MD 1992 1,25-dihydroxyvitamin D3 inhibits Na+-H+ exchange by stimulating membrane phosphoinositide turnover and increasing cytosolic calcium in CaCo-2 cells. Endocrinology 131: 1125–1133. 66. Bissonnette M, Tien XY, Niedziela SM, Hartmann SC, Frawley BP, Jr., Roy HK, Sitrin MD, Perlman RL, Brasitus TA 1994 1,25(OH)2 vitamin D3 activates PKC-alpha in Caco-2 cells: a mechanism to limit secosteroid-induced rise in [Ca2+]i. Am J Physiol 267:G465–G475. 67. Marcinkowska E, Wiedlocha A, Radzikowski C, 1997 1,25-dihydroxyvitamin D3 induced activation and subsequent nuclear translocation of MAPK is upstream regulated by PKC in HL-60 cells. Biochem Biophys Res Commun 241: 419–426. 68. van Leeuwen JP, Birkenhager JC, Buurman CJ, van den Bemd GJ, Bos MP, Pols HA 1992 Bidirectional regulation of the 1,25-dihydroxyvitamin D3 receptor by phorbol ester-activated protein kinase-C in osteoblast-like cells: interaction with adenosine 3′,5′-monophosphateinduced up-regulation of the 1,25-dihydroxyvitamin D3 receptor. Endocrinology 130:2259–2266. 69. Reinhardt TA, Horst RL 1994 Phorbol 12-myristate 13-acetate and 1,25-dihydroxyvitamin D3 regulate 1,25-dihydroxyvitamin D3 receptors synergistically in rat osteosarcoma cells. Mol Cell Endocrinol 101:159–165. 70. Krishnan AV, Cramer SD, Bringhurst FR, Feldman D 1995 Regulation of 1,25-dihydroxyvitamin D3 receptors by parathyroid hormone in osteoblastic cells: role of second messenger pathways. Endocrinology 136:705–712. 71. Marcinkowska E, Wiedlocha A, Radzikowski C 1998 Evidence that phosphatidylinositol 3-kinase and p70S6K protein are involved in differentiation of HL-60 cells induced by calcitriol. Anticancer Res 18:3507–3514. 72. Liu Q, Ning W, Dantzer R, Freund GG, Kelley KW 1998 Activation of protein kinase C-zeta and phosphatidylinositol 3′-kinase and promotion of macrophage differentiation by insulin-like growth factor-I. J Immunol 160:1393–1401. 73. Kim SH, Kang SN, Kim HJ, Kim TS 2002 Potentiation of 1,25-dihydroxyvitamin D3–induced differentiation of human promyelocytic leukemia cells into monocytes by costunolide, a germacranolide sesquiterpene lactone. Biochem Pharmacol 64:1233–1242. 74. Boyan BD, Sylvia VL, Dean DD, Schwartz Z 2002 Membrane mediated signaling mechanisms are used differentially by metabolites of vitamin D3 in musculoskeletal cells. Steroids 67:421–427. 75. Ji Y, Kutner A, Verstuyf A, Verlinden L, Studzinski GP 2002 Derivatives of vitamins D2 and D3 activate three MAPK

76. 77. 78. 79.

80. 81. 82.

83.

84. 85. 86.

87.

88.

89.

90.

91.

92. 93.

pathways and up-regulate pRb expression in differentiating HL60 cells. Cell Cycle 1:410–415. Strniskova M, Barancik M, Ravingerova T 2002 Mitogenactivated protein kinases and their role in regulation of cellular processes. Gen Physiol Biophys 21:231–255. Johnson GL, Lapadat R 2002 Mitogen-activated protein kinase pathways mediated by ERK, JNK, and p38 protein kinases. Science 298:1911–1912. Pouyssegur J, Volmat V, Lenormand P 2002 Fidelity and spatio-temporal control in MAP kinase (ERKs) signaling. Biochem Pharmacol 64:755–763. Chang F, Steelman LS, Lee JT, Shelton JG, Navolanic PM, Blalock WL, Franklin RA, McCubrey JA 2003 Signal transduction mediated by the Ras/Raf/MEK/ERK pathway from cytokine receptors to transcription factors: potential targeting for therapeutic intervention. Leukemia 17:1263–1293. Kornfeld K, Hom DB, Horvitz HR 1995 The ksr-1 gene encodes a novel protein kinase involved in Ras-mediated signaling in C. elegans. Cell 83:903–913. Sundaram M, Han M 1995 The C. elegans ksr-1 gene encodes a novel Raf-related kinase involved in Ras-mediated signal transduction. Cell 83:889–901. Stewart S, Sundaram M, Zhang Y, Lee J, Han M, Guan KL 1999 Kinase suppressor of Ras forms a multiprotein signaling complex and modulates MEK localization. Mol Cell Biol 19:5523–5534. Zhang Y, Yao B, Delikat S, Bayoumy S, Lin XH, Basu S, McGinley M, Chan-Hui PY, Lichenstein H, Kolesnick R 1997 Kinase suppressor of Ras is ceramide-activated protein kinase. Cell 89:63–72. Xing HR, Kolesnick R 2001 Kinase suppressor of Ras signals through Thr269 of c-Raf-1. J Biol Chem 276:9733–9741. Morrison DK 2001 KSR: a MAPK scaffold of the Ras pathway? J Cell Sci 114:1609–1612. Wang X, Studzinski GP 2003 Kinase suppressor of RAS (KSR) amplifies the differentiation signal provided by low concentrations of 1,25-dihydroxyvitamin D3. J Cell Physiol 198:333–342. Song X, Bishop JE, Okamura WH, Norman AW 1998 Stimulation of phosphorylation of mitogen-activated protein kinase by 1alpha,25-dihydroxyvitamin D3 in promyelocytic NB4 leukemia cells: a structure-function study. Endocrinology 139:457–465. Gniadecki R 1996 Activation of Raf-mitogen-activated protein kinase signaling pathway by 1,25-dihydroxyvitamin D3 in normal human keratinocytes. J Invest Dermatol 106:1212–1217. Chen A, Davis BH, Bissonnette M, Scaglione-Sewell B, Brasitus TA 1999 1,25-dihydroxyvitamin D3 stimulates activator protein-1-dependent Caco-2 cell differentiation. J Biol Chem 274:35505–35513. Marcinkowska E 2001 Evidence that activation of MEK1,2/ERK1,2 signal transduction pathway is necessary for calcitriol-induced differentiation of HL-60 cells. Anticancer Res 21:499–504. Wang QM, Jones JB, Studzinski GP 1996 Cyclin-dependent kinase inhibitor p27 as a mediator of the G1-S phase block induced by 1,25-dihydroxyvitamin D3 in HL60 cells. Cancer Res 56:264–267. Solomon C, Kremer R, White JH, Rhim JS 2001 Vitamin D resistance in RAS-transformed keratinocytes: mechanism and reversal strategies. Radiat Res 155:156–162. Wang X, Rao J, Studzinski GP 2000 Inhibition of p38 MAP kinase activity up-regulates multiple MAP kinase pathways

1654

94.

95.

96.

97.

98.

99.

100.

101.

102.

103. 104. 105. 106. 107. 108.

109.

and potentiates 1,25-dihydroxyvitamin D3–induced differentiation of human leukemia HL60 cells. Exp Cell Res 258: 425–437. Wang X, Studzinski GP 2001 Inhibition of p38MAP kinase potentiates the JNK/SAPK pathway and AP-1 activity in monocytic but not in macrophage or granulocytic differentiation of HL60 cells. J Cell Biochem 82:68–77. Ravid A, Rubinstein E, Gamady A, Rotem C, Liberman UA, Koren R 2002 Vitamin D inhibits the activation of stressactivated protein kinases by physiological and environmental stresses in keratinocytes. J Endocrinol 173:525–532. Li X, Udagawa N, Itoh K, Suda K, Murase Y, Nishihara T, Suda T, Takahashi N 2002 p38 MAPK-mediated signals are required for inducing osteoclast differentiation but not for osteoclast function. Endocrinology 143:3105–3113. Pepper C, Thomas A, Hoy T, Milligan D, Bentley P, Fegan C 2003 The vitamin D3 analog EB1089 induces apoptosis via a p53-independent mechanism involving p38 MAP kinase activation and suppression of ERK activity in B-cell chronic lymphocytic leukemia cells in vitro. Blood 101:2454–2460. Ji Y, Kutner A, Verstuyf A, Verlinden-L, Studzinski GP 2003 Derivatives of vitamins D2 and D3 activate three MAPK pathways and up-regulate pRb expression in differentiating HL60 cells 12th Workshop on Vitamin D, Maastricht. Qi X, Pramanik R, Wang J, Schultz RM, Maitra RK, Han J, DeLuca HF, Chen G 2002 The p38 and JNK pathways cooperate to trans-activate vitamin D receptor via c-Jun/AP-1 and sensitize human breast cancer cells to vitamin D3-induced growth inhibition. J Biol Chem 277:25884–25892. Chen F, Studzinski GP 2001 Expression of the neuronal cyclin-dependent kinase 5 activator p35Nck5a in human monocytic cells is associated with differentiation. Blood 97:3763–3767. Chen F, Studzinski GP 1999 Cyclin-dependent kinase 5 activity enhances monocytic phenotype and cell cycle traverse in 1,25-dihydroxyvitamin D3-treated HL60 cells. Exp Cell Res 249:422–428. Chen F, Rao J, Studzinski GP 2000 Specific association of increased cyclin-dependent kinase 5 expression with monocytic lineage of differentiation of human leukemia HL60 cells. J Leukoc Biol 67:559–566. Studzinski GP, Harrison JS 2003 The neuronal cyclin-dependent kinase 5 activator p35Nck5a and Cdk5 activity in monocytic cells. Leuk Lymphoma 44:235–240. Harada T, Morooka T, Ogawa S, Nishida E 2001 ERK induces p35, a neuron-specific activator of Cdk5, through induction of Egr1. Nat Cell Biol 3:453–459. Liu M, Freedman LP 1994 Transcriptional synergism between the vitamin D3 receptor and other nonreceptor transcription factors. Mol Endocrinol 8:1593–1604. Karin M, Liu Z, Zandi E 1997 AP-1 function and regulation. Curr Opin Cell Biol 9:240–246. Shaulian E, Karin M 2001 AP-1 in cell proliferation and survival. Oncogene 20:2390–2400. Sherman ML, Stone RM, Datta R, Bernstein SH, Kufe DW 1990 Transcriptional and post-transcriptional regulation of c-jun expression during monocytic differentiation of human myeloid leukemic cells. J Biol Chem 265:3320–3323. Owen TA, Bortell R, Yocum SA, Smock SL, Zhang M, Abate C, Shalhoub V, Aronin N, Wright KL, van Wijnen AJ, et al. 1990 Coordinate occupancy of AP-1 sites in the vitamin Dresponsive and CCAAT box elements by Fos-Jun in the osteocalcin gene: model for phenotype suppression of transcription. Proc Natl Acad Sci USA 87:9990–9994.

GEORGE P. STUDZINSKI AND MICHAEL DANILENKO

110. Schule R, Umesono K, Mangelsdorf DJ, Bolado J, Pike JW, Evans RM 1990 Jun-Fos and receptors for vitamins A and D recognize a common response element in the human osteocalcin gene. Cell 61:497–504. 111. Lu B, Rothnagel JA, Longley MA, Tsai SY, Roop DR 1994 Differentiation-specific expression of human keratin 1 is mediated by a composite AP-1/steroid hormone element. J Biol Chem 269:7443–7449. 112. Dwivedi PP, Omdahl JL, Kola I, Hume DA, May BK 2000 Regulation of rat cytochrome P450C24 (CYP24) gene expression. Evidence for functional cooperation of Ras-activated Ets transcription factors with the vitamin D receptor in 1,25-dihydroxyvitamin D3–mediated induction. J Biol Chem 275:47–55. 113. Studzinski GP, Rathod B, Rao J, Kheir A, Wajchman HJ, Zhang F, Finan JB, Nowell PC 1996 Transition to tetraploidy in 1,25-dihydroxyvitamin D3–resistant HL60 cells is preceded by reduced growth factor dependence and constitutive up-regulation of Sp1 and AP-1 transcription factors. Cancer Res 56:5513–5521. 114. Kveiborg M, Rattan SI, Clark BF, Eriksen EF, Kassem M 2001 Treatment with 1,25-dihydroxyvitamin D3 reduces impairment of human osteoblast functions during cellular aging in culture. J Cell Physiol 186:298–306. 115. Bikle DD, Ng D, Tu CL, Oda Y, Xie Z 2001 Calcium- and vitamin D–regulated keratinocyte differentiation. Mol Cell Endocrinol 177:161–171. 116. Bikle DD, Ng D, Oda Y, Hanley K, Feingold K, Xie Z 2002 The vitamin D response element of the involucrin gene mediates its regulation by 1,25-dihydroxyvitamin D3. J Invest Dermatol 119:1109–1113. 117. Dynan WS, Tjian R 1983 Isolation of transcription factors that discriminate between different promoters recognized by RNA polymerase II. Cell 32:669–680. 118. Gidoni D, Dynan WS, Tjian R 1984 Multiple specific contacts between a mammalian transcription factor and its cognate promoters. Nature 312:409–413. 119. Ikeda K, Nagano K, Kawakami K 1993 Anomalous interaction of Sp1 and specific binding of an E-box-binding protein with the regulatory elements of the Na,K-ATPase alpha 2 subunit gene promoter. Eur J Biochem 218:195–204. 120. Dusing MR, Wiginton DA 1994 Sp1 is essential for both enhancer-mediated and basal activation of the TATA-less human adenosine deaminase promoter. Nucleic Acids Res 22:669–677. 121. Lu J, Lee W, Jiang C, Keller EB 1994 Start site selection by Sp1 in the TATA-less human Ha-ras promoter. J Biol Chem 269:5391–5402. 122. Rao J, Zhang F, Donnelly RJ, Spector NL, Studzinski GP 1998 Truncation of Sp1 transcription factor by myeloblastin in undifferentiated HL60 cells. J Cell Physiol 175:121–128. 123. Zhang DE, Hetherington CJ, Gonzalez DA, Chen HM, Tenen DG 1994 Regulation of CD14 expression during monocytic differentiation induced with 1 alpha,25-dihydroxyvitamin D3. J Immunol 153:3276–3284. 124. Inoue T, Kamiyama J, Sakai T 1999 Sp1 and NF-Y synergistically mediate the effect of vitamin D3 in the p27Kip1 gene promoter that lacks vitamin D response elements. J Biol Chem 274:32309–32317. 125. Medema RH, Kops GJ, Bos JL, Burgering BM 2000 AFX-like forkhead transcription factors mediate cell-cycle regulation by Ras and PKB through p27Kip1. Nature 404:782–787. 126. Stahl M, Dijkers PF, Kops GJ, Lens SM, Coffer PJ, Burgering BM, Medema RH 2002 The forkhead transcription

CHAPTER 92 Differentiation and the Cell Cycle

127. 128. 129. 130. 131.

132.

133. 134. 135. 136. 137. 138. 139. 140. 141.

142. 143. 144. 145. 146. 147.

148.

factor FoxO regulates transcription of p27Kip1 and Bim in response to IL-2. J Immunol 168:5024–5031. Pardee AB 1989 G1 events and regulation of cell proliferation. Science 246:603–608. Nurse P 1990 Universal control mechanism regulating onset of M-phase. Nature 344:503–508. Sherr CJ 1994 G1 phase progression: cycling on cue. Cell 79:551–555. Zhou BB, Elledge SJ 2000 The DNA damage response: putting checkpoints in perspective. Nature 408:433–439. Peng CY, Graves PR, Thoma RS, Wu Z, Shaw AS, Piwnica-Worms H 1997 Mitotic and G2 checkpoint control: regulation of 14-3-3 protein binding by phosphorylation of Cdc25C on serine-216. Science 277:1501–1505. Mailand N, Podtelejnikov AV, Groth A, Mann M, Bartek J, Lukas J 2002 Regulation of G2/M events by Cdc25A through phosphorylation-dependent modulation of its stability. EMBO J 21:5911–5920. Elledge SJ 1996 Cell cycle checkpoints: preventing an identity crisis. Science 274:1664–1672. Dou QP, Levin AH, Zhao S, Pardee AB 1993 Cyclin E and cyclin A as candidates for the restriction point protein. Cancer Res 53:1493–1497. O’Connor PM, Fan S 1996 DNA damage checkpoints: implications for cancer therapy. Prog Cell Cycle Res 2: 165–173. Hunter T, Pines J 1994 Cyclins and cancer. II: Cyclin D and CDK inhibitors come of age. Cell 79:573–582. Morgan DO 1995 Principles of CDK regulation. Nature 374: 131–134. Hoyt MA 1997 Eliminating all obstacles: regulated proteolysis in the eukaryotic cell cycle. Cell 91:149–151. Matsuoka M, Kato JY, Fisher RP, Morgan DO, Sherr CJ 1994 Activation of cyclin-dependent kinase 4 (cdk4) by mouse MO15-associated kinase. Mol Cell Biol 14:7265–7275. Sheaff RJ 1997 Regulation of mammalian cyclin-dependent kinase 2. Methods Enzymol 283:173–193. Bartkova J, Zemanova M, Bartek J 1996 Expression of CDK7/CAK in normal and tumor cells of diverse histogenesis, cell-cycle position, and differentiation. Int J Cancer 66:732–737. Jinno S, Suto K, Nagata A, Igarashi M, Kanaoka Y, Nojima H, Okayama H 1994 Cdc25A is a novel phosphatase functioning early in the cell cycle. EMBO J 13:1549–1556. Gautier J, Solomon MJ, Booher RN, Bazan JF, Kirschner MW 1991 cdc25 is a specific tyrosine phosphatase that directly activates p34cdc2. Cell 67:197–211. Heald R, McLoughlin M, McKeon F 1993 Human wee1 maintains mitotic timing by protecting the nucleus from cytoplasmically activated Cdc2 kinase. Cell 74:463–474. Giles N, Forrest A, Gabrielli B 2003 14–3-3 acts as an intramolecular bridge to regulate cdc25B localization and activity. J Biol Chem 278:28580–28587. Matsuoka S, Huang M, Elledge SJ 1998 Linkage of ATM to cell cycle regulation by the Chk2 protein kinase. Science 282:1893–1897. el-Deiry WS, Tokino T, Velculescu VE, Levy DB, Parsons R, Trent JM, Lin D, Mercer WE, Kinzler KW, Vogelstein B 1993 WAF1, a potential mediator of p53 tumor suppression. Cell 75:817–825. Jiang H, Lin J, Su ZZ, Collart FR, Huberman E, Fisher PB 1994 Induction of differentiation in human promyelocytic HL-60 leukemia cells activates p21, WAF1/CIP1, expression in the absence of p53. Oncogene 9:3397–3406.

1655 149. Waga S, Hannon GJ, Beach D, Stillman B 1994 The p21 inhibitor of cyclin-dependent kinases controls DNA replication by interaction with PCNA. Nature 369:574–578. 150. Polyak K, Kato JY, Solomon MJ, Sherr CJ, Massague J, Roberts JM, Koff A 1994 p27Kip1, a cyclin-Cdk inhibitor, links transforming growth factor-beta and contact inhibition to cell cycle arrest. Genes Dev 8:9–22. 151. Sheaff RJ, Groudine M, Gordon M, Roberts JM, Clurman BE 1997 Cyclin E-CDK2 is a regulator of p27Kip1. Genes Dev 11:1464–1478. 152. Tsvetkov LM, Yeh KH, Lee SJ, Sun H, Zhang H 1999 p27Kip1 ubiquitination and degradation is regulated by the SCF/Skp2 complex through phosphorylated Thr187 in p27. Curr Biol 9:661–664. 153. Carrano AC, Eytan E, Hershko A, Pagano M 1999 SKP2 is required for ubiquitin-mediated degradation of the CDK inhibitor p27. Nat Cell Biol 1:193–199. 154. Hara T, Kamura T, Nakayama K, Oshikawa K, Hatakeyama S 2001 Degradation of p27Kip1 at the G0-G1 transition mediated by a Skp2-independent ubiquitination pathway. J Biol Chem 276:48937–48943. 155. Boehm M, Yoshimoto T, Crook MF, Nallamshetty S, True A, Nabel GJ, Nabel EG 2002 A growth factor-dependent nuclear kinase phosphorylates p27Kip1 and regulates cell cycle progression. EMBO J 21:3390–3401. 156. Serrano M, Hannon GJ, Beach D 1993 A new regulatory motif in cell-cycle control causing specific inhibition of cyclin D/CDK4. Nature 366:704–707. 157. Hannon GJ, Beach D 1994 p15INK4B is a potential effector of TGF-beta-induced cell cycle arrest. Nature 371:257–261. 158. Chen PL, Scully P, Shew JY, Wang JY, Lee WH 1989 Phosphorylation of the retinoblastoma gene product is modulated during the cell cycle and cellular differentiation. Cell 58:1193–1198. 159. Bartek J, Bartkova J, Lukas J 1997 The retinoblastoma protein pathway in cell cycle control and cancer. Exp Cell Res 237:1–6. 160. Magae J, Wu CL, Illenye S, Harlow E, Heintz NH 1996 Nuclear localization of DP and E2F transcription factors by heterodimeric partners and retinoblastoma protein family members. J Cell Sci 109:1717–1726. 161. Pierce AM, Schneider-Broussard R, Philhower JL, Johnson DG 1998 Differential activities of E2F family members: unique functions in regulating transcription. Mol Carcinog 22: 190–198. 162. Campanero MR, Armstrong M, Flemington E 1999 Distinct cellular factors regulate the c-myb promoter through its E2F element. Mol Cell Biol 19:8442–8450. 163. Yamasaki L 1999 Balancing proliferation and apoptosis in vivo: the Goldilocks theory of E2F/DP action. Biochim Biophys Acta 1423:M9–M15. 164. Tamrakar S, Rubin E, Ludlow JW 2000 Role of pRB dephosphorylation in cell cycle regulation. Front Biosci 5:D121–D137. 165. Brehm A, Miska EA, McCance DJ, Reid JL, Bannister AJ, Kouzarides T 1998 Retinoblastoma protein recruits histone deacetylase to repress transcription. Nature 391:597–601. 166. Magnaghi-Jaulin L, Groisman R, Naguibneva I, Robin P, Lorain S, Le Villain JP, Troalen F, Trouche D, Harel-Bellan A 1998 Retinoblastoma protein represses transcription by recruiting a histone deacetylase. Nature 391:601–605. 167. Hoffmann I, Draetta G, Karsenti E 1994 Activation of the phosphatase activity of human cdc25A by a cdk2-cyclin E dependent phosphorylation at the G1/S transition. EMBO J 13:4302–4310.

1656 168. Hassepass I, Voit R, Hoffmann I 2003 Phosphorylation at serine 75 is required for UV-mediated degradation of human Cdc25A phosphatase at the S-phase checkpoint. J Biol Chem 278:29824–29829. 169. Galaktionov K, Chen X, Beach D 1996 Cdc25 cell-cycle phosphatase as a target of c-myc. Nature 382:511–517. 170. Nishitani H, Lygerou Z 2002 Control of DNA replication licensing in a cell cycle. Genes Cells 7:523–534. 171. Diffley JF, Cocker JH, Dowell SJ, Rowley A 1994 Two steps in the assembly of complexes at yeast replication origins in vivo. Cell 78:303–316. 172. Dutta A, Bell SP 1997 Initiation of DNA replication in eukaryotic cells. Annu Rev Cell Dev Biol 13:293–332. 173. Kelly TJ, Martin GS, Forsburg SL, Stephen RJ, Russo A, Nurse P 1993 The fission yeast cdc18+ gene product couples S phase to START and mitosis. Cell 74:371–382. 174. Nishitani H, Lygerou Z, Nishimoto T, Nurse P 2000 The Cdt1 protein is required to license DNA for replication in fission yeast. Nature 404:625–628. 175. Lee JK, Hurwitz J 2001 Processive DNA helicase activity of the minichromosome maintenance proteins 4, 6, and 7 complex requires forked DNA structures. Proc Natl Acad Sci USA 98:54–59. 176. McGarry TJ, Kirschner MW 1998 Geminin, an inhibitor of DNA replication, is degraded during mitosis. Cell 93: 1043–1053. 177. Masai H, Sato N, Takeda T, Arai K 1999 CDC7 kinase complex as a molecular switch for DNA replication. Front Biosci 4:D834–D840. 178. Mimura S, Masuda T, Matsui T, Takisawa H 2000 Central role for cdc45 in establishing an initiation complex of DNA replication in Xenopus egg extracts. Genes Cells 5:439–452. 179. Walter J, Newport J 2000 Initiation of eukaryotic DNA replication: origin unwinding and sequential chromatin association of Cdc45, RPA, and DNA polymerase alpha. Mol Cell 5:617–627. 180. Zou L, Stillman B 2000 Assembly of a complex containing Cdc45p, replication protein A, and Mcm2p at replication origins controlled by S-phase cyclin-dependent kinases and Cdc7p-Dbf4p kinase. Mol Cell Biol 20:3086–3096. 181. Takisawa H, Mimura S, Kubota Y 2000 Eukaryotic DNA replication: from prereplication complex to initiation complex. Curr Opin Cell Biol 12:690–696. 182. Furstenthal L, Swanson C, Kaiser BK, Eldridge AG, Jackson PK 2001 Triggering ubiquitination of a CDK inhibitor at origins of DNA replication. Nat Cell Biol 3:715–722. 183. Masui Y, Markert CL 1971 Cytoplasmic control of nuclear behavior during meiotic maturation of frog oocytes. J Exp Zool 177:129–145. 184. Dulic V, Stein GH, Far DF, Reed SI 1998 Nuclear accumulation of p21Cip1 at the onset of mitosis: a role at the G2/Mphase transition. Mol Cell Biol 18:546–557. 185. Niculescu AB 3rd, Chen X, Smeets M, Hengst L, Prives C, Reed SI 1998 Effects of p21Cip1/Waf1 at both the G1/S and the G2/M cell cycle transitions: pRb is a critical determinant in blocking DNA replication and in preventing endoreduplication. Mol Cell Biol 18:629–643. 186. Sadhu K, Reed SI, Richardson H, Russell P 1990 Human homolog of fission yeast cdc25 mitotic inducer is predominantly expressed in G2. Proc Natl Acad Sci USA 87: 5139–5143. 187. Strausfeld U, Fernandez A, Capony JP, Girard F, Lautredou N, Derancourt J, Labbe JC, Lamb NJ 1994 Activation of

GEORGE P. STUDZINSKI AND MICHAEL DANILENKO

188.

189.

190. 191. 192.

193. 194. 195. 196. 197. 198.

199.

200.

201.

202.

203. 204.

p34cdc2 protein kinase by microinjection of human cdc25C into mammalian cells. Requirement for prior phosphorylation of cdc25C by p34cdc2 on sites phosphorylated at mitosis. J Biol Chem 269:5989–6000. Hoffmann I, Clarke PR, Marcote MJ, Karsenti E, Draetta G 1993 Phosphorylation and activation of human cdc25-C by cdc2—cyclin B and its involvement in the self-amplification of MPF at mitosis. EMBO J 12:53–63. Parker LL, Sylvestre PJ, Byrnes MJ 3rd, Liu F, Piwnica-Worms H 1995 Identification of a 95-kDa WEE1like tyrosine kinase in HeLa cells. Proc Natl Acad Sci USA 92:9638–9642. Nasmyth K, Peters JM, Uhlmann F 2000 Splitting the chromosome: cutting the ties that bind sister chromatids. Science 288:1379–1385. Uhlmann F, Wernic D, Poupart MA, Koonin EV, Nasmyth K 2000 Cleavage of cohesin by the CD clan protease separin triggers anaphase in yeast. Cell 103:375–386. Waizenegger IC, Hauf S, Meinke A, Peters JM 2000 Two distinct pathways remove mammalian cohesin from chromosome arms in prophase and from centromeres in anaphase. Cell 103:399–410. Stemmann O, Zou H, Gerber SA, Gygi SP, Kirschner MW 2001 Dual inhibition of sister chromatid separation at metaphase. Cell 107:715–726. Zachariae W, Nasmyth K 1999 Whose end is destruction: cell division and the anaphase-promoting complex. Genes Dev 13:2039–2058. Donaldson MM, Tavares AA, Hagan IM, Nigg EA, Glover DM 2001 The mitotic roles of Polo-like kinase. J Cell Sci 114:2357–2358. Tsvetkov L, Xu X, Li J, Stern DF 2003 Polo-like kinase 1 and Chk2 interact and co-localize to centrosomes and the midbody. J Biol Chem 278:8468–8475. Hosomi J, Hosoi J, Abe E, Suda T, Kuroki T 1983 Regulation of terminal differentiation of cultured mouse epidermal cells by 1 alpha,25-dihydroxyvitamin D3. Endocrinology 113:1950–1957. Svendsen ML, Daneels G, Geysen J, Binderup L, Kragballe K 1997 Proliferation and differentiation of cultured human keratinocytes is modulated by 1,25(OH)2D3 and synthetic vitamin D3 analogs in a cell density-, calcium- and serumdependent manner. Pharmacol Toxicol 80:49–56. Light BW, Yu WD, McElwain MC, Russell DM, Trump DL, Johnson CS 1997 Potentiation of cisplatin antitumor activity using a vitamin D analog in a murine squamous cell carcinoma model system. Cancer Res 57:3759–3764. Kobayashi T, Okumura H, Hashimoto K, Asada H, Inui S, Yoshikawa K 1998 Synchronization of normal human keratinocyte in culture: its application to the analysis of 1,25dihydroxyvitamin D3 effects on cell cycle. J Dermatol Sci 17:108–114. Wang QM, Chen F, Luo X, Moore DC, Flanagan M, Studzinski GP 1998 Lowering of p27Kip1 levels by its antisense or by development of resistance to 1,25-dihydroxyvitamin D3 reverses the G1 block but not differentiation of HL60 cells. Leukemia 12:1256–1265. Reitsma PH, Rothberg PG, Astrin SM, Trial J, Bar-Shavit Z, Hall A, Teitelbaum SL, Kahn AJ 1983 Regulation of myc gene expression in HL-60 leukemia cells by a vitamin D metabolite. Nature 306:492–494. Studzinski GP 1989 Oncogenes, growth, and the cell cycle: an overview. Cell Tissue Kinet 22:405–424. Wang QM, Luo X, Studzinski GP 1997 Cyclin-dependent kinase 6 is the principal target of p27/Kip1 regulation of the

1657

CHAPTER 92 Differentiation and the Cell Cycle

205. 206.

207. 208.

209.

210.

211.

212.

213.

214. 215.

216.

217. 218.

219.

220.

G1-phase traverse in 1,25-dihydroxyvitamin D3-treated HL60 cells. Cancer Res 57:2851–2855. Steinman RA, Hoffman B, Iro A, Guillouf C, Liebermann DA, el-Houseini ME 1994 Induction of p21WAF-1/CIP1 during differentiation. Oncogene 9:3389–3396. Parker SB, Eichele G, Zhang P, Rawls A, Sands AT, Bradley A, Olson EN, Harper JW, Elledge SJ 1995 p53-independent expression of p21Cip1 in muscle and other terminally differentiating cells. Science 267:1024–1027. Deng C, Zhang P, Harper JW, Elledge SJ, Leder P 1995 Mice lacking p21CIP1/WAF1 undergo normal development, but are defective in G1 checkpoint control. Cell 82:675–684. Missero C, Di Cunto F, Kiyokawa H, Koff A, Dotto GP 1996 The absence of p21Cip1/WAF1 alters keratinocyte growth and differentiation and promotes ras-tumor progression. Genes Dev 10:3065–3075. Harvat BL, Wang A, Seth P, Jetten AM 1998 Up-regulation of p27Kip1, p21WAF1/Cip1 and p16Ink4a is associated with, but not sufficient for, induction of squamous differentiation. J Cell Sci 111:1185–1196. Hershberger PA, Modzelewski RA, Shurin ZR, Rueger RM, Trump DL, Johnson CS 1999 1,25-dihydroxycholecalciferol (1,25-D3) inhibits the growth of squamous cell carcinoma and down-modulates p21Waf1/Cip1 in vitro and in vivo. Cancer Res 59:2644–2649. Zhuang SH, Burnstein KL 1998 Antiproliferative effect of 1alpha,25-dihydroxyvitamin D3 in human prostate cancer cell line LNCaP involves reduction of cyclin-dependent kinase 2 activity and persistent G1 accumulation. Endocrinology 139:1197–1207. Boyle BJ, Zhao XY, Cohen P, Feldman D 2001 Insulinlike growth factor binding protein-3 mediates 1 alpha, 25-dihydroxyvitamin D3 growth inhibition in the LNCaP prostate cancer cell line through p21WAF1. J Urol 165: 1319–1324. Moffatt KA, Johannes WU, Hedlund TE, Miller GJ 2001 Growth inhibitory effects of 1alpha, 25-dihydroxyvitamin D3 are mediated by increased levels of p21 in the prostatic carcinoma cell line ALVA-31. Cancer Res 61:7122–7129. Wu G, Fan RS, Li W, Ko TC, Brattain MG 1997 Modulation of cell cycle control by vitamin D3 and its analog, EB1089, in human breast cancer cells. Oncogene 15:1555–1563. Verlinden L, Verstuyf A, Convents R, Marcelis S, Van Camp M, Bouillon R 1998 Action of 1,25(OH)2D3 on the cell cycle genes, cyclin D1, p21, and p27 in MCF-7 cells. Mol Cell Endocrinol 142:57–65. Jensen SS, Madsen MW, Lukas J, Binderup L, Bartek J 2001 Inhibitory effects of 1alpha,25-dihydroxyvitamin D3 on the G1-S phase-controlling machinery. Mol Endocrinol 15:1370–1380. Cozzolino M, Lu Y, Finch J, Slatopolsky E, Dusso AS 2001 p21WAF1 and TGF-alpha mediate parathyroid growth arrest by vitamin D and high calcium. Kidney Int 60:2109–2117. Kawa S, Nikaido T, Aoki Y, Zhai Y, Kumagai T, Furihata K, Fujii S, Kiyosawa K 1997 Vitamin D analogs up-regulate p21 and p27 during growth inhibition of pancreatic cancer cell lines. Br J Cancer 76:884–889. Hager G, Formanek M, Gedlicka C, Thurnher D, Knerer B, Kornfehl J 2001 1,25(OH)2 vitamin D3 induces elevated expression of the cell cycle-regulating genes P21 and P27 in squamous carcinoma cell lines of the head and neck. Acta Otolaryngol 121:103–109. Muto A, Kizaki M, Yamato K, Kawai Y, Kamata-Matsushita M, Ueno H, Ohguchi M, Nishihara T, Koeffler HP, Ikeda Y 1999

221.

222.

223.

224.

225.

226.

227. 228.

229. 230.

231.

232.

233.

234.

235.

1,25-dihydroxyvitamin D3 induces differentiation of a retinoic acid-resistant acute promyelocytic leukemia cell line (UF-1) associated with expression of p21WAF1/CIP1 and p27KIP1. Blood 93:2225–2233. Prudencio J, Akutsu N, Benlimame N, Wang T, Bastien Y, Lin R, Black MJ, Alaoui-Jamali MA, White JH 2001 Action of low calcemic 1alpha,25-dihydroxyvitamin D3 analog EB1089 in head and neck squamous cell carcinoma. J Natl Cancer Inst 93:745–753. Wade WN, Willingham MC, Koumenis C, Cramer SD 2002 p27Kip1 is essential for the antiproliferative action of 1,25dihydroxyvitamin D3 in primary, but not immortalized, mouse embryonic fibroblasts. J Biol Chem 277:37301–37306. LaBaer J, Garrett MD, Stevenson LF, Slingerland JM, Sandhu C, Chou HS, Fattaey A, Harlow E 1997 New functional activities for the p21 family of CDK inhibitors. Genes Dev 11:847–862. Cheng M, Olivier P, Diehl JA, Fero M, Roussel MF, Roberts JM, Sherr CJ 1999 The p21Cip1 and p27Kip1 CDK “inhibitors” are essential activators of cyclin D-dependent kinases in murine fibroblasts. EMBO J 18:1571–1583. Polyak K, Lee MH, Erdjument-Bromage H, Koff A, Roberts JM, Tempst P, Massague J 1994 Cloning of p27Kip1, a cyclin-dependent kinase inhibitor and a potential mediator of extracellular antimitogenic signals. Cell 78:59–66. Reynisdottir I, Polyak K, Iavarone A, Massague J 1995 Kip/Cip and Ink4 Cdk inhibitors cooperate to induce cell cycle arrest in response to TGF-beta. Genes Dev 9: 1831–1845. Vlach J, Hennecke S, Amati B 1997 Phosphorylation-dependent degradation of the cyclin-dependent kinase inhibitor p27. EMBO J 16:5334–5344. Montagnoli A, Fiore F, Eytan E, Carrano AC, Draetta GF, Hershko A, Pagano M 1999 Ubiquitination of p27 is regulated by Cdk-dependent phosphorylation and trimeric complex formation. Genes Dev 13:1181–1189. Hengst L, Reed SI 1996 Translational control of p27Kip1 accumulation during the cell cycle. Science 271:1861–1864. Lin R, Wang TT, Miller WH Jr., White JH 2003 Inhibition of F-Box protein p45/SKP2 expression and stabilization of cyclin-dependent kinase inhibitor p27KIP1 in vitamin D analog-treated cancer cells. Endocrinology 144:749–753. Spruck C, Strohmaier H, Watson M, Smith AP, Ryan A, Krek TW, Reed SI 2001 A CDK-independent function of mammalian Cks1: targeting of SCF/Skp2 to the CDK inhibitor p27Kip1. Mol Cell 7:639–650. Juan G, Li X, Darzynkiewicz Z 1998 Phosphorylation of retinoblastoma protein assayed in individual HL-60 cells during their proliferation and differentiation. Exp Cell Res 244:83–92. Clarke AR, Maandag ER, van Roon M, van der Lugt NM, van der Valk M, Hooper ML, Berns A, te Riele H 1992 Requirement for a functional Rb-1 gene in murine development. Nature 359:328–330. Lee EY, Chang CY, Hu N, Wang YC, Lai CC, Herrup K, Lee WH, Bradley A 1992 Mice deficient for Rb are nonviable and show defects in neurogenesis and haematopoiesis. Nature 359:288–294. Condorelli GL, Testa U, Valtieri M, Vitelli L, De Luca A, Barberi T, Montesoro E, Campisi S, Giordano A, Peschle C 1995 Modulation of retinoblastoma gene in normal adult hematopoiesis: peak expression and functional role in advanced erythroid differentiation. Proc Natl Acad Sci USA 92:4808–4812.

1658 236. Hiebert SW, Lipp M, Nevins JR 1989 E1A-dependent transactivation of the human MYC promoter is mediated by the E2F factor. Proc Natl Acad Sci USA 86:3594–3598. 237. Roussel MF, Davis JN, Cleveland JL, Ghysdael J, Hiebert SW 1994 Dual control of myc expression through a single DNA binding site targeted by ets family proteins and E2F-1. Oncogene 9:405–415. 238. Ishida S, Shudo K, Takada S, Koike K 1995 A direct role of transcription factor E2F in c-myc gene expression during granulocytic and macrophage-like differentiation of HL60 cells. Cell Growth Differ 6:229–237. 239. Watanabe T, Sariban E, Mitchell T, Kufe D 1985 Human c-myc and N-ras expression during induction of HL-60 cellular differentiation. Biochem Biophys Res Commun 126: 999–1005. 240. Brelvi ZS, Studzinski GP 1986 Changes in the expression of oncogenes encoding nuclear phosphoproteins but not c-Ha-ras have a relationship to monocytic differentiation of HL 60 cells. J Cell Biol 102:2234–2243. 241. Dang CV, Resar LM, Emison E, Kim S, Li Q, Prescott JE, Wonsey D, Zeller K 1999 Function of the c-myc oncogenic transcription factor. Exp Cell Res 253:63–77. 242. Nesbit CE, Tersak JM, Prochownik EV 1999 MYC oncogenes and human neoplastic disease. Oncogene 18:3004–3016. 243. Brelvi ZS, Studzinski GP 1987 Coordinate expression of c-myc, c-myb, and histone H4 genes in reversibly differentiating HL 60 cells. J Cell Physiol 131:43–49. 244. Coller HA, Grandori C, Tamayo P, Colbert T, Lander ES, Eisenman RN, Golub TR 2000 Expression analysis with oligonucleotide microarrays reveals that MYC regulates genes involved in growth, cell cycle, signaling, and adhesion. Proc Natl Acad Sci USA 97:3260–3265. 245. Gartel AL, Ye X, Goufman E, Shianov P, Hay N, Najmabadi F, Tyner AL 2001 Myc represses the p21WAF1/CIP1 promoter and interacts with Sp1/Sp3. Proc Natl Acad Sci USA 98: 4510–4515. 246. Gartel AL, Shchors K 2003 Mechanisms of c-myc-mediated transcriptional repression of growth arrest genes. Exp Cell Res 283:17–21. 247. Simpson RU, Hsu T, Wendt MD, Taylor JM 1989 1,25dihydroxyvitamin D3 regulation of c-myc protooncogene transcription. Possible involvement of protein kinase C. J Biol Chem 264:19710–19715. 248. Pan Q, Simpson RU 1999 c-Myc intron element-binding proteins are required for 1,25-dihydroxyvitamin D3 regulation of c-Myc during HL-60 cell differentiation and the involvement of HOXB4. J Biol Chem 274:8437–8444. 249. Pan Q, Simpson RU 2001 Antisense knockout of HOXB4 blocks 1,25-dihydroxyvitamin D3 inhibition of c-myc expression. J Endocrinol 169:153–159. 250. Shen WF, Largman C, Lowney P, Corral JC, Detmer K, Hauser CA, Simonitch TA, Hack FM, Lawrence HJ 1989 Lineage-restricted expression of homeobox-containing genes in human hematopoietic cell lines. Proc Natl Acad Sci USA 86:8536–8540. 251. Celetti A, Barba P, Cillo C, Rotoli B, Boncinelli E, Magli MC 1993 Characteristic patterns of HOX gene expression in different types of human leukemia. Int J Cancer 53:237–244. 252. Magli MC, Largman C, Lawrence HJ 1997 Effects of HOX homeobox genes in blood cell differentiation. J Cell Physiol 173:168–177. 253. Lill MC, Fuller JF, Herzig R, Crooks GM, Gasson JC 1995 The role of the homeobox gene, HOX B7, in human myelomonocytic differentiation. Blood 85:692–697.

GEORGE P. STUDZINSKI AND MICHAEL DANILENKO

254. Rots NY, Liu M, Anderson EC, Freedman LP 1998 A differential screen for ligand-regulated genes: identification of HoxA10 as a target of vitamin D3 induction in myeloid leukemic cells. Mol Cell Biol 18:1911–1918. 255. Wilson AJ, Velcich A, Arango D, Kurland AR, Shenoy SM, Pezo RC, Levsky JM, Singer RH, Augenlicht LH 2002 Novel detection and differential utilization of a c-myc transcriptional block in colon cancer chemoprevention. Cancer Res 62:6006–6010. 256. Abe J, Moriya Y, Saito M, Sugawara Y, Suda T, Nishii Y 1986 Modulation of cell growth, differentiation, and production of interleukin-3 by 1 alpha,25-dihydroxyvitamin D3 in the murine myelomonocytic leukemia cell line WEHI-3. Cancer Res 46:6316–6321. 257. Godyn JJ, Xu H, Zhang F, Kolla S, Studzinski GP 1994 A dual block to cell cycle progression in HL60 cells exposed to analogs of vitamin D3. Cell Prolif 27:37–46. 258. Harrison LE, Wang QM, Studzinski GP 1999 1,25-dihydroxyvitamin D3-induced retardation of the G2/M traverse is associated with decreased levels of p34/cdc2 in HL60 cells. J Cell Biochem 75:226–234. 259. Colston K, Colston MJ, Feldman D 1981 1,25-dihydroxyvitamin D3 and malignant melanoma: the presence of receptors and inhibition of cell growth in culture. Endocrinology 108: 1083–1086. 260. Liu W, Asa SL, Fantus IG, Walfish PG, Ezzat S 2002 Vitamin D arrests thyroid carcinoma cell growth and induces p27 dephosphorylation and accumulation through PTEN/akt-dependent and -independent pathways. Am J Pathol 160:511–519. 261. Liu W, Asa SL, Ezzat S 2002 Vitamin D and its analog EB1089 induce p27 accumulation and diminish association of p27 with Skp2 independent of PTEN in pituitary corticotroph cells. Brain Pathol 12:412–419. 262. Narvaez CJ, Byrne BM, Romu S, Valrance M, Welsh J 2003 Induction of apoptosis by 1,25-dihydroxyvitamin D3 in MCF-7 Vitamin D3–resistant variant can be sensitized by TPA. J Steroid Biochem Mol Biol 84:199–209. 263. Xu HM, Tepper CG, Jones JB, Fernandez CE, Studzinski GP 1993 1,25-dihydroxyvitamin D3 protects HL60 cells against apoptosis but down-regulates the expression of the bcl-2 gene. Exp Cell Res 209:367–374. 264. Xu HM, Kolla SS, Goldenberg NA, Studzinski GP 1992 Resistance to 1,25-dihydroxyvitamin D3 of a deoxycytidine kinase-deficient variant of human leukemia HL60 cells. Exp Cell Res 203:244–250. 265. Lasky SR, Posner MR, Iwata K, Santos-Moore A, Yen A, Samuel V, Clark J, Maizel AL 1994 Characterization of a vitamin D3–resistant human chronic myelogenous leukemia cell line. Blood 84:4283–4294. 266. Campbell MJ, Park S, Uskokovic MR, Dawson MI, Jong L, Koeffler HP 1999 Synergistic inhibition of prostate cancer cell lines by a 19-nor hexafluoride vitamin D3 analogue and anti-activator protein 1 retinoid. Br J Cancer 79:101–107. 267. Koshizuka K, Koike M, Asou H, Cho SK, Stephen T, Rude RK, Binderup L, Uskokovic M, Koeffler HP 1999 Combined effect of vitamin D3 analogs and paclitaxel on the growth of MCF-7 breast cancer cells in vivo. Breast Cancer Res Treat 53:113–120. 268. Danilenko M, Wang X, Studzinski GP 2001 Carnosic acid and promotion of monocytic differentiation of HL60-G cells initiated by other agents. J Natl Cancer Inst 93:1224–1233. 269. Hershberger PA, McGuire TF, Yu WD, Zuhowski EG, Schellens JH, Egorin MJ, Trump DL, Johnson CS 2002 Cisplatin potentiates 1,25-dihydroxyvitamin D3-induced

CHAPTER 92 Differentiation and the Cell Cycle

270.

271.

272. 273.

274. 275.

276.

277.

278.

279.

280. 281. 282.

283.

284.

apoptosis in association with increased mitogen-activated protein kinase kinase kinase 1 (MEKK-1) expression. Mol Cancer Ther 1:821–829. Ahmed S, Johnson CS, Rueger RM, Trump DL 2002 Calcitriol (1,25-dihydroxycholecalciferol) potentiates activity of mitoxantrone/dexamethasone in an androgen independent prostate cancer model. J Urol 168:756–761. Danilenko M, Wang Q, Wang X, Levy J, Sharoni Y, Studzinski GP 2003 Carnosic acid potentiates the antioxidant and prodifferentiation effects of 1alpha,25-dihydroxyvitamin D3 in leukemia cells, but does not promote elevation of basal levels of intracellular calcium. Cancer Res 63:1325–1332. Griffin JD, Ritz J, Nadler LM, Schlossman SF 1981 Expression of myeloid differentiation antigens on normal and malignant myeloid cells. J Clin Invest 68:932–941. Brackman D, Lund-Johansen F, Aarskog D 1995 Expression of leukocyte differentiation antigens during the differentiation of HL-60 cells induced by 1,25-dihydroxyvitamin D3: comparison with the maturation of normal monocytic and granulocytic bone marrow cells. J Leukoc Biol 58:547–555. Yam LT, Li CY, Crosby WH 1971 Cytochemical identification of monocytes and granulocytes. Am J Clin Pathol 55: 283–290. Olsson I, Gullberg U, Ivhed I, Nilsson K 1983 Induction of differentiation of the human histiocytic lymphoma cell line U-937 by 1alpha,25-dihydroxycholecalciferol. Cancer Res 43:5862–5867. Blair OC, Carbone R, Sartorelli AC 1985 Differentiation of HL-60 promyelocytic leukemia cells monitored by flow cytometric measurement of nitro blue tetrazolium (NBT) reduction. Cytometry 6:54–61. Sly LM, Lopez M, Nauseef WM, Reiner NE 2001 1alpha, 25-dihydroxyvitamin D3-induced monocyte antimycobacterial activity is regulated by phosphatidylinositol 3-kinase and mediated by the NADPH-dependent phagocyte oxidase. J Biol Chem 276:35482–35493. Gum JR, Kam WK, Byrd JC, Hicks JW, Sleisenger MH, Kim YS 1987 Effects of sodium butyrate on human colonic adenocarcinoma cells. Induction of placental-like alkaline phosphatase. J Biol Chem 262:1092–1097. Giuliano AR, Franceschi RT, Wood RJ 1991 Characterization of the vitamin D receptor from the Caco-2 human colon carcinoma cell line: effect of cellular differentiation. Arch Biochem Biophys 285:261–269. Gaschott T, Stein J 2003 Short-chain fatty acids and colon cancer cells: the vitamin D receptor–butyrate connection. Recent Results Cancer Res 164:247–257. Saini K, Steele G, Thomas P 1990 Induction of carcinoembryonic-antigen-gene expression in human colorectal carcinoma by sodium butyrate. Biochem J 272:541–544. Buras RR, Shabahang M, Davoodi F, Schumaker LM, Cullen KJ, Byers S, Nauta RJ, Evans SR 1995 The effect of extracellular calcium on colonocytes: evidence for differential responsiveness based upon degree of cell differentiation. Cell Prolif 28:245–262. Franceschi RT, James WM, Zerlauth G 1985 1 alpha, 25-dihydroxyvitamin D3 specific regulation of growth, morphology, and fibronectin in a human osteosarcoma cell line. J Cell Physiol 123:401–409. van den Bemd GJ, Pols HA, Birkenhager JC, Kleinekoort WM, van Leeuwen JP 1995 Differential effects of 1,25dihydroxyvitamin D3-analogs on osteoblast-like cells and on in vitro bone resorption. J Steroid Biochem Mol Biol 55: 337–346.

1659 285. Fratzl-Zelman N, Glantschnig H, Rumpler M, Nader A, Ellinger A, Varga F 2003 The expression of matrix metalloproteinase-13 and osteocalcin in mouse osteoblasts is related to osteoblastic differentiation and is modulated by 1,25dihydroxyvitamin D3 and thyroid hormones. Cell Biol Int 27:459–468. 286. Pernalete N, Mori T, Nishii Y, Slatopolsky E, Brown AJ 1991 The activity of 22-oxacalcitriol in osteoblast-like (ROS 17/2.8) cells. Endocrinology 129:778–784. 287. Bonewald LF, Kester MB, Schwartz Z, Swain LD, Khare A, Johnson TL, Leach RJ, Boyan BD 1992 Effects of combining transforming growth factor beta and 1,25-dihydroxyvitamin D3 on differentiation of a human osteosarcoma (MG-63). J Biol Chem 267:8943–8949. 288. Skowronski RJ, Peehl DM, Feldman D 1993 Vitamin D and prostate cancer: 1,25-dihydroxyvitamin D3 receptors and actions in human prostate cancer cell lines. Endocrinology 132:1952–1960. 289. Esquenet M, Swinnen JV, Heyns W, Verhoeven G 1996 Control of LNCaP proliferation and differentiation: actions and interactions of androgens, 1alpha,25-dihydroxycholecalciferol, all-trans retinoic acid, 9-cis retinoic acid, and phenylacetate. Prostate 28:182–194. 290. Campbell MJ, Elstner E, Holden S, Uskokovic M, Koeffler HP 1997 Inhibition of proliferation of prostate cancer cells by a 19-nor-hexafluoride vitamin D3 analog involves the induction of p21waf1, p27kip1, and E-cadherin. J Mol Endocrinol 19:15–27. 291. Hedlund TE, Moffatt KA, Uskokovic MR, Miller GJ 1997 Three synthetic vitamin D analogs induce prostate-specific acid phosphatase and prostate-specific antigen while inhibiting the growth of human prostate cancer cells in a vitamin D receptor-dependent fashion. Clin Cancer Res 3:1331–1338. 292. Bauer JA, Thompson TA, Church DR, Ariazi EA, Wilding G 2003 Growth inhibition and differentiation in human prostate carcinoma cells induced by the vitamin D analog 1alpha, 24-dihydroxyvitamin D2. Prostate 55:159–167. 293. Otto T, Rembrink K, Goepel M, Meyer-Schwickerath M, Rubben H 1993 E-cadherin: a marker for differentiation and invasiveness in prostatic carcinoma. Urol Res 21:359–362. 294. Elstner E, Linker-Israeli M, Said J, Umiel T, de Vos S, Shintaku IP, Heber D, Binderup L, Uskokovic M, Koeffler HP 1995 20-epi-vitamin D3 analogs: a novel class of potent inhibitors of proliferation and inducers of differentiation of human breast cancer cell lines. Cancer Res 55:2822–2830. 295. Lazzaro G, Agadir A, Qing W, Poria M, Mehta RR, Moriarty RM, Das Gupta TK, Zhang XK, Mehta RG 2000 Induction of differentiation by 1alpha-hydroxyvitamin D5 in T47D human breast cancer cells and its interaction with vitamin D receptors. Eur J Cancer 36:780–786. 296. Wang Q, Lee D, Sysounthone V, Chandraratna RAS, Christakos S, Korah R, Wieder R 2001 1,25-dihydroxyvitamin D3 and retinoic acid analogs induce differentiation in breast cancer cells with function- and cell-specific additive effects. Breast Cancer Res Treat 67:157–168. 297. Mehta RR, Bratescu L, Graves JM, Green A, Mehta RG 2000 Differentiation of human breast carcinoma cells by a novel vitamin D analog: 1alpha-hydroxyvitamin D5. Int J Oncol 16:65–73. 298. Lanks KW, Dorwin JM, Papirmeister B 1974 Increased rate of acetylcholinesterase synthesis in differentiating neuroblastoma cells. J Cell Biol 63:824–830. 299. Moore TB, Koeffler HP, Yamashiro JM, Wada RK 1996 Vitamin D3 analogs inhibit growth and induce differentiation

1660

300.

301.

302. 303.

304.

305.

306.

307.

308.

309. 310.

311.

312. 313. 314.

in LA-N-5 human neuroblastoma cells. Clin Exp Metastasis 14:239–245. Moore TB, Sidell N, Chow VJ, Medzoyan RH, Huang JI, Yamashiro JM, Wada RK 1995 Differentiating effects of 1,25-dihydroxycholecalciferol D3 on LA-N-5 human neuroblastoma cells and its synergy with retinoic acid. J Pediatr Hematol Oncol 17:311–317. Hosoi J, Abe E, Suda T, Kuroki T 1985 Regulation of melanin synthesis of B16 mouse melanoma cells by 1 alpha, 25-dihydroxyvitamin D3 and retinoic acid. Cancer Res 45: 1474–1478. Mason RS, Pryke AM, Ranson M, Thomas HE, Posen S 1988 Human melanoma cells: functional modulation by calciotropic hormones. J Invest Dermatol 90:834–840. Cho KH, Son YS, Lee DY, Chung EK, Hur KC, Hong SI, Fuchs E 1996 Calcipotriol (MC 903), a synthetic derivative of vitamin D3 stimulates differentiation of squamous carcinoma cell line in the raft culture. Anticancer Res 16:337–347. Rice RH, Green H 1978 Relation of protein synthesis and transglutaminase activity to formation of the cross-linked envelope during terminal differentiation of the cultured human epidermal keratinocyte. J Cell Biol 76:705–711. Murphy GF, Flynn TC, Rice RH, Pinkus GS 1984 Involucrin expression in normal and neoplastic human skin: a marker for keratinocyte differentiation. J Invest Dermatol 82: 453–457. Lee E, Jeon SH, Yi JY, Jin YJ, Son YS 2001 Calcipotriol inhibits autocrine phosphorylation of EGF receptor in a calcium-dependent manner, a possible mechanism for its inhibition of cell proliferation and stimulation of cell differentiation. Biochem Biophys Res Commun 284:419–425. Smith EL, Walworth NC, Holick MF 1986 Effect of 1 alpha,25-dihydroxyvitamin D3 on the morphologic and biochemical differentiation of cultured human epidermal keratinocytes grown in serum-free conditions. J Invest Dermatol 86:709–714. Takahashi H, Ibe M, Kinouchi M, Ishida-Yamamoto A, Hashimoto Y, Iizuka H 2003 Similarly potent action of 1,25dihydroxyvitamin D3 and its analogs, tacalcitol, calcipotriol, and maxacalcitol on normal human keratinocyte proliferation and differentiation. J Dermatol Sci 31:21–28. Steven AC, Steinert PM 1994 Protein composition of cornified cell envelopes of epidermal keratinocytes. J Cell Sci 107:693–700. Takahashi H, Ibe M, Honma M, Ishida-Yamamoto A, Hashimoto Y, Iizuka H 2003 1,25-dihydroxyvitamin D3 increases human cystatin A expression by inhibiting the Raf1/MEK1/ERK signaling pathway of keratinocytes. Arch Dermatol Res. [Epub ahead of print]. Bikle DD, Gee E, Pillai S 1993 Regulation of keratinocyte growth, differentiation, and vitamin D metabolism by analogs of 1,25-dihydroxyvitamin D. J Invest Dermatol 101: 713–718. Strohman RC, Paterson B, Fluck R, Przybyla A 1974 Cell fusion and terminal differentiation of myogenic cells in culture. J Anim Sci 38:1103–1110. Drittanti L, de Boland AR, Boland R 1989 Modulation of DNA synthesis in cultured muscle cells by 1,25-dihydroxyvitamin D3. Biochim Biophys Acta 1014:112–119. Schwaller J, Koeffler HP, Niklaus G, Loetscher P, Nagel S, Fey MF, Tobler A 1995 Posttranscriptional stabilization underlies p53-independent induction of p21WAF1/CIP1/ SDI1 in differentiating human leukemic cells. J Clin Invest 95:973–979.

GEORGE P. STUDZINSKI AND MICHAEL DANILENKO

315. Munker R, Kobayashi T, Elstner E, Norman AW, Uskokovic M, Zhang W, Andreeff M, Koeffler HP 1996 A new series of vitamin D analogs is highly active for clonal inhibition, differentiation, and induction of WAF1 in myeloid leukemia. Blood 88:2201–2209. 316. Asada M, Yamada T, Ichijo H, Delia D, Miyazono K, Fukumuro K, Mizutani S 1999 Apoptosis inhibitory activity of cytoplasmic p21Cip1/WAF1 in monocytic differentiation. EMBO J 18:1223–1234. 317. Asada M, Yamada T, Fukumuro K, Mizutani S 1998 p21Cip1/WAF1 is important for differentiation and survival of U937 cells. Leukemia 12:1944–1950. 318. Asou H, Koike M, Elstner E, Cambell M, Le J, Uskokovic MR, Kamada N, Koeffler HP 1998 19-nor vitamin-D analogs: a new class of potent inhibitors of proliferation and inducers of differentiation of human myeloid leukemia cell lines. Blood 92:2441–2449. 319. Shiohara M, Uskokovic M, Hisatake J, Hisatake Y, Koike K, Komiyama A, Koeffler HP 2001 24-Oxo metabolites of vitamin D3 analogs: disassociation of their prominent antileukemic effects from their lack of calcium modulation. Cancer Res 61:3361–3368. 320. Seol JG, Park WH, Kim ES, Jung CW, Binderup L, Koeffler HP, Kim BK, Lee YY 2000 Effect of a novel vitamin D3 analog, EB1089, on G1 cell cycle regulatory proteins in HL-60 cells. Int J Oncol 16:315–320. 321. Yang ES, Maiorino CA, Roos BA, Knight SR, Burnstein KL 2002 Vitamin D–mediated growth inhibition of an androgenablated LNCaP cell line model of human prostate cancer. Mol Cell Endocrinol 186:69–79. 322. Yang ES, Burnstein KL 2003 Vitamin D inhibits G1 to S phase progression in LNCaP prostate cancer cells through p27Kip1 stabilization and Cdk2 mislocalization to the cytoplasm. J Biol Chem 268:46862–46868. 323. James SY, Mackay AG, Colston KW 1996 Effects of 1,25 dihydroxyvitamin D3 and its analogs on induction of apoptosis in breast cancer cells. J Steroid Biochem Mol Biol 58: 395–401. 324. Narvaez CJ, Welsh J 1997 Differential effects of 1,25-dihydroxyvitamin D3 and tetradecanoylphorbol acetate on cell cycle and apoptosis of MCF-7 cells and a vitamin D3–resistant variant. Endocrinology 138:4690–4698. 325. Bratland A, Risberg K, Maelandsmo GM, Gutzkow KB, Olsen OE, Moghaddam A, Wang MY, Hansen CM, Blomhoff HK, Berg JP, Fodstad O, Ree AH 2000 Expression of a novel factor, com1, is regulated by 1,25-dihydroxyvitamin D3 in breast cancer cells. Cancer Res 60:5578–5583. 326. Koike M, Elstner E, Campbell MJ, Asou H, Uskokovic M, Tsuruoka N, Koeffler HP 1997 19-nor-hexafluoride analog of vitamin D3: a novel class of potent inhibitors of proliferation of human breast cell lines. Cancer Res 57:4545–4550. 327. Hisatake J, Kubota T, Hisatake Y, Uskokovic M, Tomoyasu S, Koeffler HP 1999 5,6-trans-16-ene-vitamin D3: a new class of potent inhibitors of proliferation of prostate, breast, and myeloid leukemic cells. Cancer Res 59:4023–4029. 328. Verlinden L, Verstuyf A, Van Camp M, Marcelis S, Sabbe K, Zhao XY, De Clercq P, Vandewalle M, Bouillon R 2000 Two novel 14-Epi-analogs of 1,25-dihydroxyvitamin D3 inhibit the growth of human breast cancer cells in vitro and in vivo. Cancer Res 60:2673–2679. 329. Flanagan L, Packman K, Juba B, O’Neill S, Tenniswood M, Welsh J 2003 Efficacy of vitamin D compounds to modulate estrogen receptor negative breast cancer growth and invasion. J Steroid Biochem Mol Biol 84:181–192.

CHAPTER 92 Differentiation and the Cell Cycle

330. Galbiati F, Polastri L, Thorens B, Dupraz P, Fiorina P, Cavallaro U, Christofori G, Davalli AM 2003 Molecular pathways involved in the antineoplastic effects of calcitriol on insulinoma cells. Endocrinology 144:1832–1841. 331. Evans SR, Soldatenkov V, Shchepotin EB, Bogrash E, Shchepotin IB 1999 Novel 19-nor-hexafluoride vitamin D3 analog (Ro 25–6760) inhibits human colon cancer in vitro via apoptosis. Int J Oncol 14:979–985. 332. Scaglione-Sewell BA, Bissonnette M, Skarosi S, Abraham C, Brasitus TA 2000 A vitamin D3 analog induces a G1-phase arrest in CaCo-2 cells by inhibiting cdk2 and cdk6: roles of cyclin E, p21Waf1, and p27Kip1. Endocrinology 141: 3931–3939. 333. Gaschott T, Wachtershauser A, Steinhilber D, Stein J 2001 1,25-dihydroxycholecalciferol enhances butyrate-induced

1661 p21Waf1/Cip1 expression. Biochem Biophys Res Commun 283:80–85. 334. Gaschott T, Steinmeyer A, Steinhilber D, Stein J 2002 ZK 156718, a low calcemic, antiproliferative, and prodifferentiating vitamin D analog. Biochem Biophys Res Commun 290:504–509. 335. Gumireddy K, Ikegaki N, Phillips PC, Sutton LN, Reddy CD 2003 Effect of 20-epi-1alpha,25-dihydroxyvitamin D3 on the proliferation of human neuroblastoma: role of cell cycle regulators and the Myc-Id2 pathway. Biochem Pharmacol 65:1943–1955. 336. Gumireddy K, Reddy GS, Ikegaki N, Binderup L, Sutton LN, Phillips PC, Reddy CD 2003 Antiproliferative effects of 20-epi-vitamin-D3 analog, KH1060 in human neuroblastoma: induction of RAR-beta and p21Cip1. Cancer Lett 190:51–60.

CHAPTER 93

Vitamin D and Breast Cancer KAY COLSTON JOELLEN WELSH

Reader in Clinical Biochemistry and Metabolism, St. George’s Hospital Medical School, Cranmer Terrace, London Professor, Department of Biological Sciences, University of Notre Dame, Notre Dame, Indiana

I. Introduction II. Vitamin D Actions on Breast Cancer Cells III. Determinants of Breast Cancer Sensitivity to Vitamin D

IV. Vitamin D Analogs: Preclinical and Clinical Trials V. Vitamin D and Prevention of Breast Cancer VI. Summary and Outstanding Research Questions References

I. INTRODUCTION

trials for various indications, including cancer. However, further development of synthetic ligands for either treatment or prevention of breast cancer requires more accurate understanding of the role(s) of their cognate nuclear receptors in both normal and transformed mammary cells. In this chapter, we will review the extensive literature documenting the effects of 1,25(OH)2D3 (the natural ligand for VDR) and numerous bio-active vitamin D analogs on breast cancer cells and tumors. Furthermore, we will highlight emerging data on the role of the vitamin D endocrine system in the normal mammary gland and the possibility that vitamin D signaling may influence breast cancer development.

Adenocarcinoma of the breast arises when epithelial cells present in the mammary ducts or alveoli become transformed through a series of genetic and epi-genetic events. Considerable evidence indicates that estrogen, which drives mammary epithelial cell proliferation, is intricately involved in the etiology of human breast cancer. Anti-estrogens such as tamoxifen are effective for both treatment and prevention of estrogen dependent breast cancer. Tamoxifen represents the best characterized selective estrogen receptor modifier (SERM), a class of synthetic compounds that interact with the nuclear estrogen receptor (ER) in a cell type specific manner. SERMS antagonize ER signaling in breast tissue but not in bone, thus limiting proliferation of estrogen dependent breast cancer cells without adversely affecting bone mass. Although SERMs are effective in treatment of estrogen responsive disease, only one third of breast tumors are estrogen dependent, and tumor progression is often associated with loss of estrogen sensitivity. Thus, there is a need for alternative therapies that target estrogen independent cells and that minimize the progression of estrogen responsive disease to hormone independence. Other nuclear receptors present in mammary cells, such as the progesterone receptor (PR), retinoid receptors, and the vitamin D receptor (VDR), have emerged as promising therapeutic targets for breast cancer. Based on the importance of nuclear receptors in mediating expression of genes involved in proliferation, differentiation, and apoptosis, synthetic structural analogs of nuclear receptor ligands which exhibit biological properties distinct from the natural ligands represent a feasible approach to manipulate nuclear receptor activity. Structurally, most nuclear receptor ligands have the additional advantage of being orally active. In the case of the VDR, many synthetic analogs with desirable therapeutic profiles have been developed, and some are in clinical VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

II. Vitamin D Actions on Breast Cancer Cells The VDR is expressed in the majority of human breast tumors, thus it represents a potential therapeutic target for established cancer. Extensive research has been directed towards elucidation of the effects of 1,25(OH)2D3 and its synthetic analogs on breast cancer cells, and several reviews on this topic are available [1–3]. In this section, we provide a concise summary of the effects of 1,25(OH)2D3 and several of its structural analogs on proliferation, differentiation, apoptosis, angiogenesis, and invasion of breast cancer cells. Potential targets of VDR signaling involved in these effects are discussed below and summarized in Table I.

A. Effects of 1,25(OH)2D3 on Breast Cancer Cell Proliferation 1. CELL CYCLE PROGRESSION

Treatment of MCF-7 breast cancer cells with 1,25(OH)2D3 at nanomolar concentrations induces cell Copyright © 2005, Elsevier, Inc. All rights reserved.

1664

KAY COLSTON AND JOELLEN WELSH

TABLE I Vitamin D and Therapy of Breast Cancer Model Breast cancer cells in vitro

Observations ● ●

●

●

●

●

Animal models of breast cancer

●

● ●

Clinical Trials

●

●

●

VDR expressed in human breast cancer cells; regulated by hormones and growth factors 1,25(OH)2D3 and analogs induce G1 arrest, apoptosis, and/or differentiation via VDR dependent mechanisms 1,25(OH)2D3 suppresses estrogen, EGF, and IGF-1 mitogenic signaling and enhances growth inhibitory actions of TGFβ Modulation of cell cycle regulatory proteins, oncogenes, and tumor suppressor genes linked to vitamin D mediated anti-tumor effects 1,25(OH)2D3 induces apoptosis via caspase independent mechanisms that involve mitochondria and/or endoplasmic reticulum 1,25(OH)2D3 inhibits angiogenesis and invasion via effects on tumor cells, endothelial cells, and extracellular matrix proteases Vitamin D analogs inhibit growth of carcinogen-induced mammary tumors and human xenografts in absence of weight loss or hypercalcemia Anti-tumor effects in vivo associated with induction of growth arrest and apoptosis EB1089 exerts antimetastatic effects in vivo Topical calcipotriol well tolerated in women with locally advanced or cutaneous metastatic breast cancer Dose escalation study indicated oral EB1089 well tolerated in majority of advanced breast cancer patients Clinical effects of vitamin D analogs in advanced cancer patients to date limited to partial disease stabilization

Summary of effects of natural and synthetic vitamin D compounds on breast cancer cells in vitro and in vivo. Details and references are noted in text.

cycle arrest in G0/G1 [4,5], dephosphorylation of the retinoblastoma protein [5,6], and increases in the cyclin dependent kinase inhibitors p21WAF-1/CIP1 and p27kip1 [7–9] (see also Chapter 92). A vitamin D responsive element (VDRE) in the human p21WAF-1/CIP1 gene promoter suggests that p21WAF-1/CIP1 is a direct transcriptional target of the VDR [10]. Effects of vitamin D compounds on p27kip1 vary with cell type; in some studies p27kip1 is unchanged after treatment with 1,25(OH)2D3 or the synthetic analog EB1089 [4,8,11], whereas in others, p27kip1 expression is increased [4,9,12]. Analysis of the p27kip1 gene promoter suggests that the 1,25(OH)2D3–VDR complex induces transcription of this gene through SP1 and NF-Y transcription factors rather than direct DNA binding [13]. Upregulation of p21WAF-1/CIP1 and p27kip1 by vitamin D compounds is associated with inhibition of cyclin dependent kinase (CDK) activity, including CDK2 associated histone H1 kinase, cyclin D1/CDK4, and cyclin A/CDK2 [4,12]. VDR also interacts with protein phosphatases PP1c and PP2Ac to inactivate the p70 S6 kinase, which is essential for G1/S phase transition [14]. Thus, the net result of vitamin D signaling is to prevent entry into S phase, leading to accumulation in G1. In some breast cancer cells, vitamin D mediated G1

arrest is associated with induction of differentiation markers such as lipid and casein [15–17]. 2. REGULATION OF ONCOGENES AND TUMOR SUPPRESSOR GENES IN BREAST CANCER CELLS BY VITAMIN D

The effect of vitamin D compounds on MCF-7 cells has been further studied at the level of c-myc and c-fos proto-oncogene expression. EB1089 (Chapter 84) decreases c-myc mRNA and transiently increases c-fos expression, being approximately 50 times more potent than 1,25(OH)2D3 [16]. The observation that vitamin D signaling regulates c-myc mRNA is consistent with the presence of a putative VDRE in the human c-myc gene [18]. The p53 tumor suppressor gene plays a crucial role in regulation of growth arrest and apoptosis in response to cellular stress and DNA damage. Growth inhibition of MCF-7 cells, which express wild-type p53, by two vitamin D analogs (EB1089 and KH1060) is associated with increased p53 expression [19,20]. However, 1,25(OH)2D3 does not consistently up-regulate p53 in breast cancer cells [6,21], and vitamin D compounds can inhibit growth of breast cancer cells expressing mutant p53 such as T47D [22–24]. Thus, functional p53 is not required for the antiproliferative

1665

CHAPTER 93 Vitamin D and Breast Cancer

effects of vitamin D. This notion is consistent with data indicating that 1,25(OH)2D3 mediated transactivation of the p21WAF-1/CIP1 gene promoter is p53 independent [10]. BRCA1, a tumor suppressor gene which functions in DNA repair, is also induced by 1,25(OH)2D3 in MCF-7 cells [25]. BRCA1 promoter analysis indicated that the effects of 1,25(OH)2D3 are indirectly mediated by the VDR. Sensitivity to 1,25(OH)2D3 mediated growth inhibition correlates with induction of BRCA1 and is highest in well-differentiated breast cancer cells. These data suggest that hereditary breast cancers that develop in patients with germ line mutations in BRCA1 might be less responsive to vitamin D mediated growth inhibition. 3. ESTROGEN SIGNALING

1,25(OH)2D3 and EB1089 down-regulate ER and suppress estrogen action in MCF-7 cells [5,26–28]. Since estrogen is mitogenic for most breast cancer cells, down-regulation of estrogen regulated pathways may contribute to the antiproliferative effects of vitamin D. Sensitivity to 1,25(OH)2D3 is generally higher in breast cancer cells that express ER, such as MCF-7 and T47D, than in those that do not [25,29]. Downregulation of ER by vitamin D compounds attenuates both the mitogenic effects of estrogen and the induction of target genes such as the PR and pS2 [26,27]. Co-treatment of ER positive breast cancer cells with 1,25(OH)2D3 or EB1089 and the anti-estrogens tamoxifen or ICI 182,780 inhibits proliferation more than either compound alone [26,30–33]. Sequence analysis of the ERα gene promoter has identified a potential VDRE, suggesting a direct regulatory effect of 1,25(OH)2D3 on ERα gene transcription [34]. Under some conditions, 1,25(OH)2D3 inhibits estrogen-induced transcription of the pS2 gene in the absence of a change in ER abundance [35], suggesting that vitamin D compounds may exert multiple effects on ER signaling. However, since 1,25(OH)2D3 and its analogs also inhibit growth of estrogen-independent breast cancer cells, ER signaling is not required for the anti-tumor effects of vitamin D compounds [33,36–39]. Furthermore, in some cases, breast cancer cells selected for anti-estrogen resistance show increased sensitivity to vitamin D [40]. 4. GROWTH FACTOR SIGNALING

Additional evidence suggests that vitamin D modulates secretion, processing, and/or signaling of critical growth factors in breast cancer cells. Vitamin D compounds inhibit mitogenic activity of EGF and IGF-I and induce negative growth regulators, such as TGFβ. Thus, EB1089 attenuates the growth stimulatory effects of EGF [41] and 1,25(OH)2D3 regulates EGF

receptor levels in breast cancer cells [42,43]. In addition, it has recently been reported that the gene encoding amphiregulin, a heparin-binding EGFrelated growth factor, is transcriptionally regulated by 1,25(OH)2D3 [44]. As reviewed by Sachdev and Yee [45], high plasma IGF-I is associated with increased risk of breast cancer and the IGF-I receptor (IGF-IR) is overexpressed in many breast cancer cell lines. Effects of IGF-I reflect both its extracellular concentration and the levels of IGF binding proteins (IGFBPs), which modulate its availability to the IGF-IR. In breast cancer cells, vitamin D compounds block the mitogenic effects of IGF-I, decrease expression of IGF-IR, and induce inhibitory IGFBPs such as IGFBPs-3 and -5 [37,38,46–49]. The net result of vitamin D signaling thus, is attenuation of IGF-1 stimulated mitogenesis and accumulation of IGFBPs which can promote apoptosis. In breast cancer cells, 1,25(OH)2D3 enhances the expression of the negative growth regulator TGFβ1, as well as its latent form binding protein [50,51]. A direct effect of 1,25(OH)2D3 on the TGFβ2 gene is supported by the identification of VDREs in its promoter [52]. The antiproliferative effect of vitamin D compounds is partially abrogated by neutralizing antibodies to TGFβ [12,51,53,54], indicating that TGFβ can be functionally linked to the growth inhibitory effects of vitamin D in vitro.

B. Pro-apoptotic Effects of 1,25(OH)2D3 in Breast Cancer Cells In addition to their antiproliferative effects, 1,25(OH)2D3 and its analogs induce morphological and biochemical features of apoptosis (cell shrinkage, chromatin condensation, and DNA fragmentation) in breast cancer cells [5,7,55,56]. Other markers of apoptosis induced by 1,25(OH)2D3 include reorientation of phosphatidylserine (PS) to the exterior of the cell, PARP cleavage and up-regulation of apoptotic related proteins, such as clusterin, cathepsin B, and TGFβ [5,27,53,56]. Furthermore, 1,25(OH)2D3 exerts additive or synergistic effects in combination with other triggers of apoptosis, such as anti-estrogens, TNFα, radiation, and chemotherapeutic agents [19,33,57–59]. It is not quite clear whether these synergistic effects result from interactions of 1,25(OH)2D3 with agonistspecific signaling pathways or whether 1,25(OH)2D3 impacts on components of a common apoptotic pathway. The intracellular signaling pathways implicated in 1,25(OH)2D3 mediated apoptosis of MCF-7 cells are depicted in Fig. 1. Several independent studies have

1666 reported that sensitivity to 1,25(OH)2D3–mediated apoptosis reflects the relative expression and/or subcellular localization of the Bcl-2 family of pro- and anti-apoptotic proteins. Treatment of MCF-7 cells with 1,25(OH)2D3 or EB1089 induces redistribution of the pro-apoptotic Bcl-2 family member, Bax, from the cytosol to the mitochondria and down-regulates the anti-apoptotic protein Bcl-2 [5,19,60,61]. Furthermore, overexpression of Bcl-2 renders MCF-7 cells resistant to 1,25(OH)2D3 mediated apoptosis [23]. Since Bcl-2 and Bax act antagonistically in the regulation of apoptosis, these data suggest that translocation of Bax in conjunction with down-regulation of Bcl-2 may be necessary for 1,25(OH)2D3–mediated apoptosis. Vitamin D–mediated Bax translocation triggers reactive oxygen species (ROS) generation, dissipation of the mitochondrial membrane potential, and release of cytochrome c into the cytosol [60,61], features of the intrinsic (mitochondrial) pathway of apoptosis [62]. 1,25(OH)2D3 also enhances mitochondrial ROS generation and cytochrome c release in MCF-7 cells treated with TNFα [63]. Of particular interest, neither Bax translocation, ROS generation, mitochondrial membrane potential dissipation, nor cytochrome c release are induced by 1,25(OH)2D3 in MCF-7DRES cells, a variant of MCF-7 cells selected for vitamin D resistance [60,61]. Another pathway recently implicated in vitamin D–mediated apoptosis of MCF-7 cells involves calcium release from the endoplasmic reticulum and activation of µ-calpain; this process can be prevented by either calpain inhibitors or calcium buffering agents such as calbindin D28K [64]. While the specific interactions between the apoptotic pathways depicted in Fig. 1 have yet to be resolved, it is possible that signals generated from both the mitochondria and the endoplasmic reticulum cooperate to induce cell death in response to 1,25(OH)2D3. It is clear, however, that MCF-7 cells undergo cell death in the presence of caspase inhibitors, indicating that the commitment to 1,25(OH)2D3 mediated cell death is caspase independent [23,60].

C. Role of 1,25(OH)2D3 in Regulation of Angiogenesis, Invasion, and Metastasis Metastasis, the process by which tumor cells invade secondary sites, requires degradation of the extracellular matrix and is facilitated by angiogenesis, the growth of new blood vessels into developing tumors. Effects of vitamin D signaling on late stage breast cancer have been studied in ER negative breast cancer cell lines, such as MDA-MB-231 and SUM159PT cells, which are invasive in vitro and metastatic in vivo. In these cell

KAY COLSTON AND JOELLEN WELSH

1,25(OH)2D3-VDR complex

Redox Signals?

BAX

Ca

Mitochondria

Endoplasmic Reticulum

Cyt c

ROS

Ca++

Calpain

Caspases?

APOPTOSIS

FIGURE 1 Potential mechanisms of vitamin D mediated apoptosis in breast cancer cells. VDR signaling, triggered by 1,25(OH)2D3 or synthetic analogs, alter apoptotic pathways present in mitochondria and endoplasmic reticulum, leading to DNA fragmentation and cell death. The direct targets of VDR, as well as the mechanisms by which apoptosis is triggered, are not yet identified. See text for further details.

lines, 1,25(OH)2D3 and EB1089 inhibit invasion as measured by the in vitro Boyden chamber assay [9,65]. Inhibition of invasion by vitamin D compounds can be dissociated from effects on proliferation, and may be linked to regulation of extracellular proteases such as MMP-9, urokinase-type plasminogen activator (uPA), and tissue type plasminogen activator (tPA). In MDAMB-231 cells, these effects may result from vitamin D–mediated up-regulation of protease inhibitors PA inhibitor 1 and MMP inhibitor 1 [66]. The anti-tumor effects of 1,25(OH)2D3 may also involve regulation of angiogenesis, since 1,25(OH)2D3 inhibits angiogenesis in the chick embryo chorioallantoic membrane assay [67] and in tumor cell-induced angiogenesis assays in mice [68]. Moreover, vitamin D analogs reduce angiogenesis of MCF-7 breast tumors overexpressing vascular endothelial growth factor (VEGF) and inhibit VEGF expression in MDA-MB231 xenografts [69,70]. VDR is expressed in endothelial cells [71] and 1,25(OH)2D3 blocks both basal and VEGF-induced endothelial cell sprouting, elongation, and proliferation [70,72]. Collectively, these studies

1667

CHAPTER 93 Vitamin D and Breast Cancer

indicate that vitamin D signaling likely inhibits angiogenesis via VDRs expressed on both the transformed mammary epithelial cells and the endothelial cells within breast tumors.

III. DETERMINANTS OF BREAST CANCER SENSITIVITY TO VITAMIN D A. Ligand Availability Circulating 1,25(OH)2D3 is delivered to cells via the serum vitamin D–binding protein, but little is known about metabolism or half life of 1,25(OH)2D3 in breast cancer cells. Catabolism of 1,25(OH)2D3 is initiated via hydroxylation at the 24 position in the side chain, a reaction catalyzed by the 25(OH)D3 24-hydroxylase (CYP24). Comparative genome hybridization studies have found that CYP24 is amplified in human breast cancer [73], suggesting that enhanced catabolism of 1,25(OH)2D3 by the 24-hydroxylase, leading to reduced ligand availability to the VDR, could contribute to breast cancer. In addition to uptake of 1,25(OH)2D3 from the circulation, it is formally possible that breast cancer cells express the 25(OH)D3 1α-hydroxylase (CYP27B1) and produce 1,25(OH)2D3 from 25(OH)D3. The 1α-hydroxylase mRNA has been detected in MCF7 cells and in human breast tumors [74]; however, MCF-7 cells are not growth inhibited by 25(OH)D3 [75], suggesting that access to or activity of the 1αhydroxylase enzyme is not sufficient to generate growth-inhibitory concentrations of 1,25(OH)2D3. In contrast, nontransformed human mammary epithelial cells express 1α-hydroxylase and are growth inhibited by physiological concentrations of 25(OH)D3 [75], indicating that 1α-hydroxylase activity may be physiologically relevant in normal breast cells. Further studies to assess the metabolism of both 25(OH)D3 and 1,25(OH)2D3 in mammary cells as a function of transformation will be necessary to clarify the role of 1α-hydroxylase in breast cancer.

B. Expression and Regulation of VDR in Breast Cancer Cells Receptors for 1,25(OH)2D3 have been demonstrated in carcinogen-induced rat mammary tumors, human breast tumors and established breast cancer cell lines [29,76,77]. A recent study has clarified the importance of the nuclear VDR in mediating the effects of 1,25(OH)2D3 and its analogs in breast cancer cells. Zinser et al. [78] utilized cell lines derived from carcinogen-induced mammary tumors generated in VDR knockout mice and wild-type control litter mates to

demonstrate that cells lacking VDR fail to respond to vitamin D compounds. In contrast VDR expressing mammary tumor cell lines generated from tumors that developed in wild-type mice were growth inhibited. These data confirm that the nuclear VDR is required to mediate anti-tumor effects and therefore, expression, function, and regulation of VDR in mammary cells are important determinants of sensitivity to vitamin D. VDR abundance is affected by many physiological factors and is achieved through a variety of mechanisms, including alterations in transcription and/or mRNA stability, post-translational effects, and ligand-induced stabilization. Expression of the VDR in cultured cells and in vivo is regulated by many physiological agents, including 1,25(OH)2D3 itself, estrogens, retinoids, and growth factors [79–81]. Thus, breast cancer cell sensitivity to 1,25(OH)2D3 mediated growth regulation may in part reflect the activity of other hormone signaling pathways through their impact on VDR expression. Comparison of a panel of breast cancer cells indicates that ER positive cells tend to express higher levels of VDR than ER negative cells [29]. Furthermore, in vitro studies demonstrate that estrogen up-regulates the VDR, and anti-estrogens such as tamoxifen downregulate the VDR, in ER positive breast cancer cells [75,82]. In MCF-7 and T47D cells, estrogen transcriptionally up-regulates the VDR promoter upstream of exon 1c [24,83]. Collectively, these data support the concept that estrogen and anti-estrogens are important regulators of VDR expression in breast cancer cells, a concept with clinical implications arising from the potential use of SERMs for prevention and/or treatment of breast cancer and osteoporosis. The efficacy and toxicity of vitamin D analogs is determined, in part, by the level of VDR in target tissues, and thus it will be important to determine the degree to which estrogen status influences VDR abundance in different 1,25(OH)2D3 target cells in vivo (i.e., breast, bone, uterus). In this respect, it will also be important to assess whether novel SERMs or phytoestrogens currently utilized by post-menopausal women act as estrogen agonists or antagonists in regulation of VDR expression. The phytoestrogen resveratrol has recently been shown to up-regulate VDR and sensitize breast cancer cells to vitamin D–mediated growth inhibition, offering proof of principle that dietary factors can impact on cellular sensitivity to 1,25(OH)2D3 through regulation of VDR [24].

C. Vitamin D Resistance Although it is clear that the VDR is required for breast cancer cell responsiveness to vitamin D compounds,

1668 a number of established breast cancer cell lines that express VDR fail to respond to the anti-proliferative effects of 1,25(OH)2D3. Data from mammary cell lines suggest that oncogenic transformation with SV40 or ras inhibits VDR signaling and induces resistance to the growth inhibitory effects of 1,25(OH)2D3 [80,84], raising the possibility that breast cancer progression may be facilitated by deregulation of the vitamin D pathway. In an effort to understand the cellular basis for insensitivity to vitamin D, Narvaez et al. [82] selected and characterized 1,25(OH)2D3–resistant subclones of MCF-7 cells. The resulting MCF-7DRES cells express VDR, but do not undergo growth arrest or apoptosis in response to 1,25(OH)2D3. MCF-7DRES cells are selectively resistant to 1,25(OH)2D3 and its structural analogs, and respond to other anti-proliferative agents [21,33,82]. Similar results have been obtained in an independently derived 1,25(OH)2D3–resistant subclone of MCF-7 cells, labeled MCF-7/VDR [8]. The mechanisms underlying vitamin D resistance in these MCF-7 clones are incompletely understood. Theoretically, selective insensitivity to 1,25(OH)2D3 could be secondary to defective VDR, reduced availability of ligand, or uncoupling of a functional vitamin D signaling pathway from growth arrest/apoptosis. While resistance could be associated with elevated expression of the vitamin D 24-hydroxylase enzyme which inactivates 1,25(OH)2D3, this does not appear to be the case for either of the vitamin D–resistant MCF-7 variants. Both MCF-7DRES and MCF-7/VDR cells contain transcriptionally active VDRs when measured with consensus VDREs; however, basal VDR expression is lower in both resistant cell lines than in parental MCF-7 cells. In MCF-7DRES cells, 1,25(OH)2D3 comparably up-regulates the steady state level of the VDR protein in both sensitive and resistant cell lines [21]. MCF-7DRES cells can be sensitized to the growthinhibitory effects of 1,25(OH)2D3 by co-treatment with low concentrations of the phorbol ester TPA, suggesting that phosphorylation pathways may be altered in this cell line [21,61]. Further studies with these interesting cell lines will be necessary to resolve the mechanism(s) of vitamin D resistance. Significantly, the MCF-7DRES cell line retains resistance to vitamin D analogs when grown as xenografts in nude mice [85], providing an important model system for understanding the basis of vitamin D resistance in vivo.

D. Prognostic Significance of Breast Tumor VDR Expression A high proportion (>80%) of breast cancer biopsy specimens contain VDR [86–88], and up-regulation of

KAY COLSTON AND JOELLEN WELSH

VDR protein in breast carcinomas compared to normal breast tissue has been reported [89]. In breast tumors, there is no significant correlation between VDR expression and ER expression, lymph node status, or tumor grade [90,91]. Tumor VDR status does not appear to be related to overall survival [86,87,91] or to survival after relapse [91]. However, in a study of 136 patients with primary breast cancer, it was found that women with VDR negative tumors relapsed significantly earlier than women with VDR positive tumors [91,92].

IV. VITAMIN D ANALOGS: PRECLINICAL AND CLINICAL TRIALS A. Natural Ligands Versus Synthetic Analogs While the beneficial effects of 1,25(OH)2D3 on cancer cells support its use as a therapeutic agent, natural vitamin D metabolites exert potentially toxic effects on calcium handling at the doses required for anti-tumor effects. Thus, the metabolite 1α(OH)D3, which is converted to 1,25(OH)2D3 in vivo, effectively inhibits tumor growth in vivo, but the therapeutic window is extremely narrow [reviewed in 2]. Modifications to the parent vitamin D structure have successfully generated synthetic analogs with enhanced growth regulatory effects and limited calcium mobilizing action (see Section VIII of this book). Several vitamin D analogs have been tested in preclinical and clinical models of breast cancer as described below.

B. Effects of Vitamin D Analogs on Breast Cancer Cells and Tumors 1. CALCIPOTRIOL

Calcipotriol (MC903, Leo Pharmaceutical Products, Denmark) contains a cyclopropyl substitution in the side chain [93] and is equipotent with 1,25(OH)2D3 in inhibition of MCF-7 cell growth in vitro [94]. However, due to rapid inactivation of the analog in vivo, calcipitriol displays calcemic activity 100–200 times less than 1,25(OH)2D3. This compound is marketed for topical treatment of psoriasis (see Chapter 101).The efficacy of topical treatment with calcipotriol was assessed in 19 women with locally advanced or cutaneous metastatic breast cancer [95]. Of 14 patients who completed treatment with topical calcipotriol ointment, three showed partial response and one exhibited a minimal response.

1669

CHAPTER 93 Vitamin D and Breast Cancer

2. MAXACALCITOL (OCT)

22-oxa-1,25(OH)2D3 (Chugai Pharmaceutical Co. Ltd), in which an oxygen atom is substituted for the methyl group at C-22 is described in detail in Chapter 86. OCT displays reduced calcemic activity in vivo and effectively inhibits growth of both ER positive MCF-7 xenografts and ER negative MX-1 tumors in nude mice [36]. In MCF-7 xenografts, OCT exerts synergistic antitumor effects with the anti-estrogen tamoxifen [30]. Furthermore, OCT exerts growth-inhibitory effects on dimethylbenzanthracene (DMBA)-induced rat mammary tumors, alone and in combination with the aromatase inhibitor CGS 16949A [96], and reduces VEGF expression in MDA-MB-231 breast tumors [69]. 3. SEOCALCITOL (EB1089)

Seocalcitol (EB1089), a second generation analog from Leo Pharmaceutical Products is detailed in Chapter 84. EB1089 contains a conjugated double-bond system and is approximately 50 times more potent than 1,25(OH)2D3 in vitro with markedly reduced effects on calcium metabolism in vivo [16,94,97]. Oral administration of EB1089 to rats bearing nitrosomethyl urea (NMU)-induced mammary tumors dose-dependently inhibits tumor growth, and effective low doses do not increase serum calcium [94,98]. Anti-tumor effects of EB1089 are also observed in DMBA-induced mammary tumors and MCF-7 xenografts [41,85]. In both the NMU-induced tumor model and MCF-7 cell xenografts, EB1089 induces tumor regression through inhibition of proliferation and induction of DNA fragmentation indicative of apoptotic cell death [85,99]. The beneficial effects of EB1089 on tumor progression in the MCF-7 xenograft model is enhanced when co-administered with paclitaxel [100], retinoic acid [101], or radiation [59]. EB1089 also inhibits development of bone metastases and increases survival of mice following intracardiac inoculation of ER negative MDA-MB-231 cells [102]. Based on a dose-finding study in 13 healthy volunteers, doses in the range of 5–20 µg/day EB1089, given orally for four consecutive days, can be considered for future use in clinical trials. A phase I trial of oral EB1089 in patients with advanced breast and colorectal cancer has been completed. This trial was an open, noncontrolled single-center study with sequentially assigned dose levels [103]. Twenty-five women with breast cancer and four women and seven men with colorectal carcinoma received EB1089 twice daily for five days with a three week post-dosing follow-up. Twenty patients received compassionate treatment after this post-dosing interval for between 10 and 234 days (mean 90 ± 62 days). On the basis of this study, the estimated maximum tolerated dose of EB1089 for prolonged use is approximately 7 µg/m2/day. Ten patients developed

hypercalcemia, which resolved by seven days after cessation of treatment, and no other serious adverse reactions were observed. Although no clear anti-tumor effects were seen in this study, six patients (two colorectal, four breast cancer) showed disease stabilization for at least three months. 4. OTHER VITAMIN D ANALOGS WITH ANTI-TUMOR EFFECTS ON BREAST CANCER

The 16-ene vitamin D analogs are characterized by the introduction of a double bond at the C16 position in the D ring of the molecule [104] (see Chapter 85). 1α,25-dihydroxy-16-ene-23-yne-cholecalciferol (Ro23-7553 or ILX-23-7553, Hoffmann-LaRoche Ltd) is more potent than 1,25(OH)2D3 in cell growth inhibition and is currently in phase I trials for patients with advanced metastatic cancer. The Hoffmann-LaRoche compound Ro25-6760 is a 19-nor-hexafluoride analog that suppresses growth of human breast cancer cells in vitro and inhibits growth of MCF-7 xenografts in nude mice [105]. Synergistic inhibition of tumor growth is observed in animals given Ro25-6760 in combination with paclitaxel [100]. The 20-epi analogs, including 20-epi(S)-ethoxy23-yne 24a,26a,27a-trihomo1α,25-dihydroxyvitamin D3 (CB1093, Leo Pharmaceuticals) and 20-epi-22oxa24a,26a,27a-tri-homo-1,25(OH)2D3 (KH1060, Leo Pharmaceuticals), potently inhibit growth of breast cancer cells, xenografts, and NMU-induced rat mammary tumors [20,106,107]. Novel analogs with 19-nor and 14epi modifications developed by Bouillon and colleagues also exhibit anti-cancer effects in human breast cancer cells in vitro and in vivo [108]. The analog 1α-hydroxy24-ethyl-cholecalciferol (1α(OH)D5) developed by Mehta and colleagues exerts anti-tumor effects against established human breast cancer cells and directly inhibits preneoplastic lesion development in mouse mammary gland organ culture [109]. Collectively, these studies offer proof of principle that vitamin D analogs can inhibit breast cancer progression with minimal calcemic side effects. While it is clear that most, if not all, vitamin D analogs mediate their growth-inhibitory effects through the VDR [78], further mechanistic studies are required to understand the selective actions of these analogs in vivo (see Chapters 82 and 83).

V. VITAMIN D AND PREVENTION OF BREAST CANCER A. Expression and Role of VDR in Normal Mammary Gland A potential role of vitamin D in breast cancer prevention has been suggested based on animal,

1670

KAY COLSTON AND JOELLEN WELSH

TABLE II Evidence Linking Vitamin D to Prevention of Breast Cancer Approaches Animal models

Observations ● ● ● ●

Epidemiological studies

●

●

Genetic studies

● ● ●

VDR is expressed in normal mammary gland 1,25(OH)2D3 inhibits hormone stimulated ductal growth and branching 1,25(OH)2D3 inhibits carcinogen induced pre-neoplastic lesions in mammary organ culture VDR null mice exhibit accelerated mammary gland development Inverse associations reported between biomarkers of sunlight exposure, dairy products, and/or dietary vitamin D and risk of breast cancer Low serum 1,25(OH)2D3 associated with enhanced breast cancer risk and/or disease activity Amplification of 25-hydroxyvitamin D 24-hydroxylase in breast cancers VDR polymorphisms linked to breast cancer risk and/or metastatic progression Fok 1 and singlet A repeat polymorphisms affect VDR transcriptional activity

Summary of data generated in animal, epidemiological, and genetic studies, see text for details and references.

epidemiological, and genetic studies (Table II). The VDR is present in rabbit, rat, mouse, and human mammary gland [110–113], and its expression is developmentally regulated. VDR expression is high throughout puberty, pregnancy, and lactation, periods of maximal tissue growth and remodeling [78,113,114]. The dynamic regulation of VDR in mammary gland during the reproductive cycle suggests that hormones and/or growth factors that impact on glandular development may modulate VDR expression. Indeed, lactogenic hormones up-regulate VDR in normal mammary gland and nontranformed mammary cells in vitro [80,115]; however, the specific factors responsible for VDR regulation in the normal mammary gland in vivo have yet to be defined. Developmental regulation of VDR in mammary cells implies that vitamin D signaling may be involved in the regulation of glandular function. In vitro, 1,25(OH)2D3 inhibits growth of nontransformed mammary cells as well as breast cancer cells [75,80,81]. However, in contrast to breast cancer cells, nontransformed mammary cells exhibit markers of differentiation rather than apoptosis [15,80,116]. This has led to the suggestion that 1,25(OH)2D3 and the VDR induce a program of genes that inhibit proliferation and maintain differentiation in the normal gland [117]. This suggestion is supported by organ culture studies that demonstrate effects of 1,25(OH)2D3 on calcium transport, casein expression, and branching morphogenesis [113,118,119]. Furthermore, mammary glands from VDR-ablated mice are heavier and exhibit increased ductal extension and branching morphogenesis compared to glands from wild-type control mice [113]. In addition, glands from VDR-ablated mice exhibit enhanced growth in response to estrogen and progesterone, both in vivo

and in organ culture, compared to glands from control mice. In organ culture, 1,25(OH)2D3 inhibits branching of mammary glands from control mice but has no effect on glands from VDR knockout mice. These and other data, reviewed in Welsh et al. [81], provide evidence that 1,25(OH)2D3 and the nuclear VDR exert growth inhibitory effects on normal mammary cells during early development of the gland.

B. Prevention of Breast Cancer by Vitamin D: Preclinical Studies Identification of 1,25(OH)2D3 and the VDR as components of a signaling network that impacts on proliferation and differentiation in the normal mammary gland raises the possibility that optimal vitamin D status may protect against mammary transformation. In support of this suggestion, rats fed diets high in calcium and vitamin D develop fewer mammary tumors in response to the carcinogen dimethylbenzanthracene (DMBA) than mice fed diets low in calcium and vitamin D [120]. However, whether this difference specifically reflects vitamin D signaling is unclear since vitamin D deficiency is associated with multiple metabolic disturbances. Prevention of NMU-induced mammary tumors with vitamin D analogs, including Ro24-5531 [1α,25-dihydroxy-16-ene-23-yne-26-27hexafluorocholecalciferol) and 1α(OH)D5 provide further support that vitamin D may protect against breast cancer [121,122]. A direct effect of 1,25(OH)2D3 and 1α(OH)D5 on the sensitivity of the mammary gland to transformation is suggested by studies indicating that both vitamin D compounds prevent DMBA induced preneoplastic lesions in organ culture [123].

CHAPTER 93 Vitamin D and Breast Cancer

C. Epidemiological Studies on Vitamin D Status and Breast Cancer The majority of women who develop breast cancer are of postmenopausal age, and estrogen deficiency and aging are often associated with vitamin D deficiency. However, few epidemiological studies have examined whether dietary intake of vitamin D per se alters breast cancer incidence in populations (see Chapter 91 for a discussion). An evaluation of the Nurses Health Study [124] found that intakes of dairy products, dairy calcium, and total vitamin D (as measured by food frequency questionnaires) were inversely associated with breast cancer risk in premenopausal, but not postmenopausal, women. These data are consistent with an earlier study that reported an inverse correlation between intake of dairy products and breast cancer risk [125]. Another recent study included evaluation of sunlight exposure in addition to vitamin D from diet and supplements in relation to breast cancer risk [126]. In this study, several measures of sunlight exposure and dietary vitamin D intake were associated with a reduced risk of breast cancer; however, the associations were dependent on region of residence. Correlations between risk of breast cancer and exposure to solar radiation, which increases epidermal synthesis of vitamin D, have also been proposed [127–129] (see Chapter 90). In two studies where vitamin D status was measured in relation to breast cancer, low levels of 1,25(OH)2D3 were found to be associated with increased breast cancer risk or disease progression [130,131].

D. VDR Polymorphisms and Breast Cancer Risk There has been considerable interest in genetically determined differences in the VDR signaling pathway in relation to disease susceptibility. A number of common allelic variants, or polymorphisms, in the human VDR gene have been examined in relation to risk of breast cancer. The best studied VDR polymorphisms include a start codon polymorphism (FokI) in exon 2, BsmI and Apa I polymorphisms in an intronic region between exons VIII and IX, a Taq I variant in exon IX and a singlet (A) repeat in exon IX. An Australian study [132] examined Apa I and Taq I polymorphisms in patients with breast cancer compared to women with no history (family or personal) of breast cancer. Allele frequencies of the Apa I polymorphism showed a significant association with breast cancer risk while the Taq I polymorphism showed a similar trend, but the association was not significant, and allelle frequencies

1671 of the Fok I polymorphism were not significantly different. In a Japanese population study on the Bsm I polymorphism, the bb genotype conferred an almost fourfold increase in the risk of breast cancer [133]. A study in a London Caucasian population also demonstrated a significant association between breast cancer risk and the Bsm I polymorphism, with the odds ratio for bb vs BB genotype over 2.3 [134]. The ‘L’ poly(A) variant was also associated with a similar risk in this study. However, the data are not entirely consistent, as two reports showed no association between the Taq I polymorphism and breast cancer risk [135,136], and an increased (rather than decreased) breast cancer risk was associated with the BB genotype among Latina women in the United States [137] and with the AA genotype in a small study of Taiwanese women [138]. Other studies have shown an association of VDR polymorphisms with disease progression rather than breast cancer risk. Ruggiero et al. [139] suggested that the VDR Bsm I polymorphism was related to development of more aggressive metastatic breast cancer, and a second study reported a significant association between the bb VDR genotype and presence of lymph node metastases [140]. Schondorf et al. [141] reported that breast cancer patients with the AA genotype have a 1.7-fold increased risk of developing bone metastases, whereas patients with the TT genotype have a 0.5-fold risk. Although these findings are certainly intriguing, the underlying basis for an association between VDR polymorphisms and breast cancer susceptibility is currently unclear. Three of the VDR polymorphisms that have been linked to breast cancer susceptibility (BsmI, Apa I, or Taq I variants) do not alter the amount, structure, or function of the VDR protein produced. There is evidence, however, that two VDR polymorphisms (the VDR start codon polymorphism defined by FokI and the singlet (A) repeat in exon IX) may have functional significance. The FokI site dictates which of two potential translation initiation sites is utilized. Individuals lacking the FokI restriction site initiate translation at the first site, and express the full length VDR consisting of 427 amino acids. In contrast, individuals with the FokI restriction site utilize a second ATG site, generating a VDR protein of 424 amino acids. Although no significant differences in ligand affinity, DNA binding, or transactivation activity were found between these two VDR forms when studied independently, when the VDR start codon polymorphism was considered simultaneously with the singlet (A) repeat in exon IX, differences in VDR function were detected in vitro [142]. In transient transfection assays with a vitamin D responsive reporter gene, the shorter VDR variant was shown to interact more strongly with the

1672

KAY COLSTON AND JOELLEN WELSH

transcription factor TFIIB and display higher potency than the longer VDR variant. These data support the concept that functionally relevant polymorphisms in the VDR exist, and further studies will be required to determine whether VDR genotype interacts with other risk factors for breast cancer (also see Chapter 68).

VI. SUMMARY AND OUTSTANDING RESEARCH QUESTIONS VDR and 1,25(OH)2D3, its natural ligand, act through multiple signaling pathways to induce growth arrest, differentiation, and apoptosis in mammary epithelial cells (Fig. 2). Synthetic analogs of 1,25(OH)2D3, which have

Mammary cell exposed to estrogen, progesterone, and/or mitogenic growth factors

Intact VDR signaling G1 Arrest

Differentiation or apoptosis

Normal glandular development; deletion of mutated cells; tumor prevention or regression

FIGURE 2

Deregulated VDR signaling Proliferation

Proliferation

Enhanced proliferation; increased sensitivity to transformation

Model for effects of VDR pathway on normal or transformed mammary epithelial cells. In the presence of functional VDR, 1,25(OH)2D3 attentuates hormone and growth factor stimulated proliferation, inducing G1 arrest and differentiation in normal cells and/or apoptosis in transformed cells. Thus, vitamin D signaling is predicted to maintain normal glandular development, eliminate mutated cells, and/or limit growth of established tumors. In the absence of VDR signaling, cells are more sensitive to the mitogenic effects of hormones and growth factors and may be more likely to undergo transformation.

potent growth inhibitory effects with minimal calcemic activity in vivo, provide proof of principle that vitamin D signaling can inhibit the growth of established tumors in animal models. Studies with VDR null mice indicate a functional role for vitamin D signaling in the normal mammary gland. Clinical studies and epidemiological approaches have provided evidence that vitamin D signaling represents a target for breast cancer prevention. Challenges for the future include better understanding of the transport, uptake, and metabolism of 1,25(OH)2D3 and bioactive analogs in breast cancer cells, the molecular mechanism of action and specific targets of the VDR in mammary gland, and the influence of genetic differences in the VDR on an individual’s response to vitamin D compounds. Such understanding should provide insight into design of vitamin D–based strategies to impact on breast cancer development or therapy.

References 1. Colston KW, Mork Hansen C 2001 Mechanisms implicated in the growth regulatory effects of vitamin D in breast cancer. Endocrine Related Cancer 9:45–59. 2. Mørk Hansen C, Binderup L, Hamberg K, Carlberg C 2001a Vitamin D and cancer: effects of 1,25(OH)2D3 and its analogs on growth control and tumorigenesis. Frontiers in Biosci 6:820–848. 3. Mehta RG, Mehta RJ 2002 Vitamin D and cancer. J Nutr Biochem 13:252–264. 4. Wu G, Fan RS, Li W, Ko T, Brattain MG 1997 Modulation of cell cycle by vitamin D3 and its analog EB1089 in human breast cancer cells. Oncogene 15:1555–1563. 5. Simboli-Campbell M, Narvaez CJ, VanWeelden K, Tenniswood M, Welsh JE 1997 Comparative effects of 1α,25(OH)2D3 and EB1089 on cell cycle kinetics and apoptosis in MCF-7 cells. Breast Cancer Res Treat 42:31–41. 6. Fan FS, Yu W 1995 1,25-dihydroxyvitamin D3 suppresses cell growth, DNA synthesis, and phosphorylation of retinoblastoma protein in a breast cancer cell line. Cancer Invest 13:280–286. 7. James SY, Mackay AG, Colston KW 1996 Effects on 1,25 dihydroxyvitamin D3 and its analogs on induction of apoptosis in breast cancer cells. J Steroid Biochem Molec Biol 58:395–401. 8. Mørk Hansen C, Rhode L, Madsen MW, Hansen D, Colston KW, Pirianov G, Holm PK, Binderup L 2001b MCF-7/VDR: a new vitamin D–resistant cell line. J Cell Biochem 82:422–436. 9. Flanagan L, Packman K, Juba B, O’Neill S, Tenniswood M, Welsh JE 2003 Efficacy of vitamin D compounds to modulate estrogen receptor negative breast cancer growth and invasion. J Steroid Biochem Mol Biol 84:181–192. 10. Liu M, Lee M-H, Cohen M, Bommakanti M, Freedman LP 1996 Transcriptional activation of the cdk inhibitor p21 by vitamin D3 leads to the induced differentiation of the myelocytic cell line U937. Genes & Development 10:142–153. 11. Jensen S, Madsen MW, Lukas J, Binderup L, Bartek J 2001 Inhibitory effects of 1α,25-dihydroxyvitamin D3 on the G1-S phase controlling machinery. Mol Endocrinol 15: 1370–1380.

1673

CHAPTER 93 Vitamin D and Breast Cancer

12. Verlinden L, Verstuyf A, Convents R, Marcelis S, van Camp M, Bouillon R 1998 Action of 1,25(OH)2D3 on the cell cycle genes, cyclin D1, p21, and p27 in MCF-7 cells. Mol Cell Endocrinology 142:57–65. 13. Inoue T, Kamiyama J, Sakai T 1999 Sp1 and NF-Y synergistically mediate the effect of vitamin D(3) in the p27(kip1) gene promoter that lacks vitamin D response elements. J Biol Chem 274:32309–32317. 14. Bettoun DJ, Buck DW 2nd, Lu J, Khalifa B, Chin WW, Nagpal S 2002 A vitamin D receptor-Ser/Thr phosphatasep70 S6 kinase complex and modulation of its enzymatic activities by the ligand. J Biol Chem 277:24847–24850. 15. Lazzaro G, Agadir A, Quing W Poria M, Mehta RR, Moriarty RM, Das Gupta TK, Zhang XK, Mehta RG 2000 Induction of differentiation by 1alpha-hydroxyvitamin D5 in T47D human breast cancer cells and its interaction with vitamin D receptors. Eur J Cancer 36:780–786. 16. Mathiasen IS, Colston KW, Binderup L 1993 EB1089, a novel vitamin D analog, has strong antiproliferative and differentiation inducing effects on cancer cells. J Steroid Biochem Molec Biology 46:365–371. 17. Wang Q, Lee D 2001 Sysounthone V, Chandraratna R, Christakos S, Korah R, Wieder R 2001 1,25-dihydroxyvitamin D3 and retinoic acid analogues induce differentiation in breast cancer cells with function and cell specific additive effects. Breast Cancer Research Treat 67:157–168. 18. Okano K, Usa T, Ohtsuru A, Tsukazaki T, Miyazaki Y, Yonekura A, Namba H, Shindoh H, Yamashita S 1999 Effect of 22-oxa-1,25-dihydroxyvitamin D3 on human thyroid cancer cell growth. Endocrine J 46:243–252. 19. James SY, Mackay AG, Colston KW 1995 Vitamin D derivatives in combination with 9-cis retinoic acid promote active cell death in breast cancer cells. J Mol Endocrinol 14:391–394. 20. Elstner E, Linker-Israli M, Said J, Umiel T, deVos S, Shintaku IP, Heber D, Binderup L Uskokovic M, Koeffler HP 1995 20- epi-vitamin D analogs: a novel class of potent inhibitors of proliferation and inducers of differentiation of human breast cancer cell lines. Cancer Research 55: 2822–2830. 21. Narvaez CJ, Welsh JE 1997 Differential effects of 1,25dihydroxy-vitamin D3 and tetradecanoylphorbol acetate on cell cycle and apoptosis of MCF-7 cells and a vitamin D3–resistant variant. Endocrinology 138: 4690–4698. 22. Eisman JA, Sutherland RI, McMenemy ML, Fragonas JC, Musgrove EA, Pang GYN 1989 Effects of 1,25-dihydroxyvitamin D3 on cell cycle kinetics of T47D human breast cancer cells. J Cell Physiol 138:611–616. 23. Mathiasen IS, Lademann U, Jäättelä M 1999 Apoptosis induced by vitamin D compounds in breast cancer cells is inhibited by Bcl-2 but does not involve known caspases or p53. Cancer Research 59:4848–4856. 24. Weitzke JA, Welsh JE 2003 Phytoestrogen regulation of a vitamin D3 receptor promoter and 1,25-dihydroxyvitamin D3 actions in human breast cancer cells. J Steroid Biochem Mol Biol 84: 149–157. 25. Campbell MJ, Gombart AF, Kwok SH, Park S, Koeffler HP 2000 The antiproliferative effects of 1alpha,25(OH)2D3 on breast and prostate cells are associated with induction of BRCA1 gene expression. Oncogene 19:5091–5097. 26. James SY, Mackay AG, Binderup L, Colston KW 1994 Effects of a new synthetic vitamin D analog, EB1089, on the oestrogen-responsive growth of human breast cancer cells. J Endocrinology 141:555–546. 27. Colston KW, Mackay AG, James SY 1995 Vitamin D3 derivatives and breast cancer. In: M Tenniswood,

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38. 39.

40.

41.

H Michna (eds) Schering Foundation Workshop Volume 14, Springer Verlag: Heidelberg pp. 201–224. Swami S, Krishnan AV, Feldman D 2000 1,25-dihydroxyvitamin D3 down-regulates estrogen receptor abundance and suppresses estrogen action in MCF-7 human breast cancer cells. Clin Cancer Research 6:3371–3379. Buras RR, Schumaker LM, Davoodi F, Brenner RV, Shabahang M, mock Nauta RJ, Evans SR 1994 Vitamin D receptors in breast cancer cells. Breast Cancer Res Treat 31:191–202. Abe-Hashimoto J, Kikuchi T, Matsumoto T, Nishii Y, Ogata E, Ikeda K 1993 Anti-tumor effect of 22-oxa-calctriol, a noncalcemic analog of calcitriol in athymic mice implanted with human breast carcinoma and its synergism with tamoxifen. Cancer Research 53:2534–2537. Vink-van Wijngaarden T, Pols HAP, Buurman CJ, van der Bemd GJCM, Dorssers CJ, Birkenhager JC, van Leeuwen JPTM 1994 Inhibition of breast cancer cell growth by combined treatment with vitamin D analogs and tamoxifen. Cancer Research 54:5711–5717. Love-Schimenti C, Gibson D, Ratman A, Bikle D 1996 Antiestrogen potentiation of antiproliferative effects of vitamin D3 analogs in breast cancer cells. Cancer Research 56: 2789–2794. Nolan E, Donepudi M, VanWeelden K, Flanagan L, Welsh JE 1998 Dissociation of vitamin D3 and anti-estrogen mediated growth regulation in MCF-7 breast cancer cells. Mol Cell Biochem 188:13–20. Stoica A, Saceda M, Fakhro A, Solomon HB, Fenester BD, Martin MB 1999 Regulation of estrogen receptor-α gene expression by 1,25-dihydroxyvitamin D in MCF-7 cells. J Cell Biochem 75:640–651. Demirpence E, Balaguer P, Trousse F, Nicolas J-C, Pons M, Gagne D 1994 Antiestrogenic effects of all-trans retinoic acid and 1,25-dihydroxyvitamin D3 in breast cancer cells occur at the estrogen response element level but through different mechanisms. Cancer Research 54:1458–1464. Abe J, Nakano T, Nishii Y, Matsumoto T, Ogata E, Ikeda K 1991 A novel vitamin D3 analog, 22-oxa-1,25-dihydroxyvitamin D3, inhibits the growth of human breast cancer in vitro and in vivo without causing hypercalcemia. Endocrinology 129:832–837. Colston KW, Perks CM, Xie SP, Holly JMP 1998 Growth inhibition of both MCF-7 and Hs578T human breast cancer cell lines by vitamin D analogs is associated with increased expression of insulin-like binding protein-3. J Mol Endocrinol 20: 157–162. Xie SP, Pirianov G, Colston KW 1999 Vitamin D analogs suppress IGF-I signaling and promote apoptosis in breast cancer cells. Eur J Cancer 35:1717–1723. Flanagan L, VanWeelden K, Ammerman C, Ethier S, Welsh JE 1999 SUM-159PT cells, a novel estrogen independent human breast cancer model system. Breast Cancer Research Treat 58:193–204. Larsen SS, Heiberg I, Lykkesfeldt AE 2001 Anti-estrogen resistant human breast cancer cell lines are more sensitive towards treatment with the vitamin D analog EB1089 than parent MCF-7 cells. Br J Cancer 2001: 84:686–690. Saez S, Meggough F, Lefebvre M-F, Descotes F, Pampile R, Adam L, Crepin M 1994 Potential direct and indirect influence of 1,25(OH)2D3 on the growth of human colonic and breast carcinoma. In: AW Norman, R Bouillon, M Thomasset (eds) Vitamin D, a Pluripotent Steroid Hormone: Structural Studies, Molecular Endocrinology, and Clinical Applications. De Gruyter: Berlin, pp. 469–476.

1674 42. Desprez P-Y, Poujol D, Falette N, Lefebvre M-F, Saez F 1991 1,25-dihydroxyvitamin D3 increases epidermal growth factor receptor gene expression in BT-20 breast carcinoma cells. Biochem Biophys Res Comm 176:1–6. 43. Koga M, Eisman JA, Sutherland RL 1988 Regulation of epidermal growth factor receptor levels by 1,25-dihydroxyvitamin D3 in human breast cancer cells. Cancer Research 48:2734–2739. 44. Akutsu N, Bastien Y, Lin R, Mader S, White JH 2001 Amphiregulin is a vitamin D3 target gene in squamous cell and breast carcinoma. Biochem Biophys Res Commun 281:1051–1056. 45. Sachdev D, Yee D 1994 The IGF system and breast cancer. Endocr Relat Cancer 8:197–209. 46. Rozen F, Yang X-F, Huynh H, Pollak M 1997 Antiproliferative action of vitamin D–related compounds and insulin-like growth factor binding protein 5 accumulation. J Natl Cancer Inst 89:652–656. 47. Vink-van Wijngaarden T, Pols HAP, Buurman CJ, Birkenhager JC, van Leeuwen JPTM 1996 Inhibition of insulin and insulin-like growth factor stimulated growth of human breast cancer cells by 1,25-dihydroxyvitamin D3 and the vitamin D analog EB1089. Eur J Cancer 32A:842–848. 48. Xie S, James SY, Colston KW 1997 Vitamin D derivatives inhibit the mitogenic effects of IGF-I on MCF-7 human breast cancer cells. J Endocrinol 154:495–504. 49. Pirianov G, Colston KW 2001a Interaction of vitamin D analogs with signaling pathways leading to apoptosis in breast cancer cells. Steroids 66:309–318. 50. Koli K, Keski-Oja J 1994 1,25-dihydroxyvitamin D3 has been shown to enhance the expression of transforming growth factor B1 and its latent form binding protein in breast carcinoma cells. Cancer Research 55:1540–1547. 51. Mercier T, Chaumontet C, Gaillard-Sanchez I, Martel P, Heberden C 1996 Calcitriol and lexicalcitol (KH1060) inhibit the growth of human breast adenocarcinoma cells by enhancing transforming growth factor-β production. Biochem Pharmacol 52:505–510. 52. Wu Y, Craig TA, Lutz WH, Kumar R 1999 Identification of vitamin D response elements in the human transforming growth factor beta 2 gene. Biochemistry 38:2654–2660. 53. Simboli-Campbell M, Welsh JE 1995 1,25 dihydroxyvitamin D3: coordinate regulator of active cell death and proliferation in MCF-7 breast cancer cells. In: M Tenniswood, H Michna (eds) Schering Foundation Workshop Volume 14, Springer Verlag: Heidelberg, pp. 191–200. 54. Yang L, Yang J, Venkateswarlu S, Ko T, Brattain MG 2001 Autocrine TGFβ signaling mediates vitamin D3 analoginduced growth inhibition in breast cells. J Cell Physiol 188:383–393. 55. Welsh JE 1994 Induction of apoptosis in breast cancer cells in response to vitamin D and antiestrogens. Biochem Cell Biol 72:537–545. 56. Simboli-Campbell M, Narvaez CJ, Tenniswood M, Welsh JE 1996 1,25-dihydroxyvitamin D3 induces morphological and biochemical markers of apoptosis in MCF-7 breast cancer cells. J Steroid Biochem Mol Biol 58:367–376. 57. Wang Q, Yang W, Uytingo MS, Christakos S, Wieder R 2000 1,25-dihydroxyvitamin D3 and all-trans-retinoic acid sensitize breast cancer cells to chemotherapy-induced cell death. Cancer Research 60:2040–2048. 58. Pirianov G, Colston KW 2001b Interactions of vitamin D analog CB1093, TNFa, and ceramide on breast cancer cell apoptosis. Mol Cell Endocrinol 172:69–78.

KAY COLSTON AND JOELLEN WELSH

59. Sundaram S, Sea A, Feldman S, Strawbridge R, Hoopes PJ, Demidenko E, Binderup L, Gewirtz DA 2003 The combination of a potent vitamin D analog, EB1089, with ionizing radiation reduces tumor growth and induces apoptosis of MCF-7 breast tumor xenografts in nude mice. Clin Cancer Research 9:2350–2356. 60. Narvaez CJ, Welsh JE 2001 Role of mitochondria and caspases in vitamin D–mediated apoptosis of MCF-7 breast cancer cells. J Biol Chem 276:9101–9107. 61. Narvaez CJ, Byrne B, Romu S, Valrance M, Welsh JE 2003 Induction of apoptosis by 1,25-dihydroxyvitamin D3 in MCF-7 vitamin D3–resistant variant can be sensitized by TPA. J Steroid Biochem Mol Biol 84:199–209. 62. Scorrano L, Korsmeyer SJ 2003 Mechanisms of cytochrome c release by proapoptotic BCL-2 family members. Biochem Biophys Res Comm 304:437–444. 63. Weitsman GE, Ravid A, Liberman U, Koren R 2003 Vitamin D enhances caspase dependent and independent TNF α induced breast cancer cell death: the role of reactive oxygen species and mitochondria. Int J Cancer 106:178–186. 64. Mathiasen IS, Sergeev IN, Bastholm L, Elling F, Norman AW, Jäättelä M 2003 Calcium and calpain as key mediators of apoptosis-like death induced by vitamin D3 compounds in breast cancer cells. J Biol Chem 277:30738–30745. 65. Mørk Hansen C, Frandsen TL, Brunner N, Binderup L 1994 1,25-dihydroxyvitamin D3 inhibits the invasive potential of human breast cancer cells in vitro. Clin Exp Metastasis 12:195–202. 66. Koli K, Keski-Oja J 2000 1,25-dihydroxyvitamin D3 and its analogs down-regulate cell invasion-associated proteases in cultured malignant cells. Cell Growth and Differentiation 11:221–229. 67. Oikawa T, Hirotani K, Ogasawara H, Katayama T, Nakamura O, Iwaguchi T, Hiragun A 1990 Inhibition of angiogenesis by vitamin D3 analogs. Eur J Pharmacol 178:247–250. 68. Majewski S, Szmurio A, Marczak M, Jablonska S Bollag W, Jablonska J 1996 Vitamin D3 is a potent inhibitor of tumor cell-induced angiogenesis. J Invest Dermatol Symp Proc 1:97–101. 69. Matsumoto H, Lino Y, Koibuchi Y, Andoh T, Horii Y, Takei H, Horiguchi J, Maemura M, Yokoe T, Morishita Y 1999 Antitumor effect of 22-oxacalcitriol on estrogen receptor negative MDA-MB-231 tumors in athymic mice. Oncol Rep 6:349–352. 70. Mantell DJ, Owens PE, Bundred NJ, Mawer EB, Canfield AE 2000 1alpha,25-dihydroxyvitamin D(3) inhibits angiogenesis in vitro and in vivo. Circ Research 87:214–220. 71. Merke J, Milde P, Lewicka S, Hugel U, Klaus G, Mangelsdorf DJ, Haussler MR, Rauterberg EW, Ritz E 1989 Identification and regulation of 1,25-dihydroxyvitamin D3 receptor activity and biosynthesis of 1,25-dihydroxyvitamin D3. Studies in cultured bovine aortic endothelial cells and human dermal capillaries. J Clin Invest 83:1903–1915. 72. Bernardi RJ, Johnson CS, Modzelewski RA, Trump DL 2002 Antiproliferative effect of 1alpha,25-dihydroxyvitamin D(3) and vitamin D analogs on tumor-derived endothelial cells. Endocrinology 143:2508–2514. 73. Albertson DG, Ylstra B, Segraves R, Collins C, Dairkee SH, Kowbel D, Kuo WL, Gray JW, Pinkel D 2001 Quantitative mapping of amplicon structure by array CGH identifies CYP24 as a candidate oncogene. Nature Genetics 25:144–146. 74. Friedrich M, Reichrath J, Chen TC, Tanpricha V, Gherson I, Tilgen W, Schmidt W, Holick M 2000 Expression of 25-hydroxyvitamin D3-1α-hydroxylase in breast tissue.

CHAPTER 93 Vitamin D and Breast Cancer

75.

76. 77. 78.

79.

80.

81. 82. 83.

84. 85.

86.

87. 88.

89.

90.

In: AW Norman, R Bouillon, M Thomasset (eds) Vitamin D endocrine system: Structural, Biological, Genetic and Clinical Aspects. University of California Riverside Press: Riverside, CA, pp. 189–191. Welsh JE, Weitzke JA, Zinser GM, Smyczek S, Romu S, Tribble E, Welsh JC, Byrne B, Narvaez CJ 2003b Impact of the vitamin D3 receptor on growth regulatory pathways in mammary gland and breast cancer. J Steroid Biochem Mol Biol 83:85–92. Eisman JA, Macintyre I, Martin TJ, Frampton RJ, King RJ 1980 Normal and malignant breast tissue is a target organ for 1,25(OH)2 vitamin D3. Clin Endocrinology 13:267–272. Colston K, Wilkinson JR, Coombes RC 1986 1,25dihydroxyvitamin D3 binding in estrogen responsive rat breast tumor. Endocrinology 119:397–403. Zinser GM, McEleney K, Welsh JE 2003 Characterization of mammary tumor cell lines from wild-type and vitamin D receptor knockout mice. Mol Cell Endocrinology 200: 67–80. Krishnan AV, Feldman D 1991 Stimulation of 1,25-dihydroxy-vitamin D3 receptor gene expression in cultured cells by serum and growth factors. J Bone Mineral Research 6: 1099–1107. Escaleira MT, Brentani MM 1999 Vitamin D3 receptor (VDR) expression in HC-11 mammary cells: regulation by growth modulatory agents, differentiation, and Ha-ras transformation. Breast Cancer Research Treat 54:123–133. Welsh JE, Wietzke JA, Zinser GM, Byrne B, Smith K, Narvaez CJ 2003a Vitamin D3 receptor as target for breast cancer prevention. J Nutrition 133:2425S–2433S. Narvaez CJ, VanWeelden K, Byrne I, Welsh JE 1996 Characterization of a vitamin D3–resistant MCF-7 cell line. Endocrinology 137:400–409. Byrne I, Flanagan L, Tenniswood M, Welsh JE 2000 Identification of a hormone responsive promoter immediately upstream of exon 1c in the human vitamin D receptor gene. Endocrinology 141:2829–2836. Agadir A, Lazzara G, Zheng Y, Zhang X, Mehta RG 1999 Resistance of HBL100 human breast epithelial cells to vitamin D action. Carcinogenesis 20:577–582. VanWeelden K, Flanagan L, Binderup L, Tenniswood M, Welsh JE 1998 Apoptotic regression of MCF-7 xenografts in nude mice treated with the vitamin D analog EB1089. Endocrinology 139:2102–2110. Freake HC, Abeasekekera G, Iwasaki J, Marocci C, MacIntyre I, McClelland RA, Skilton RA, Easton DF, Coombes RC 1984 Measurement of 1,25-dihydroxyvitamin D3 receptors in breast cancer and relationship to biochemical and clinical indices. Cancer Research 44:1677–1681. Eisman JA, Suva LJ, Martin TJ 1986 Significance of 1,25dihydroxyvitamin D3 receptor in primary breast cancers. Cancer Research 46:5406–5408. Berger U, Wilson P, McClelland R, Colston K, Haussler MR, Pike JW, Coombes RC 1987 Immunocytochemical detection of 1,25-dihydroxyvitamin D3 receptor in primary breast cancer. Cancer Research 47:6793–6795. Friedrich M, Villena-Heinsen C, Tilgen W, Schmidt W, Reichrat J, Axt-Fliedner R 2002a Vitamin D receptor (VDR) expression is not a prognostic factor in breast cancer. Anticancer Res 22:1919–1924. Friedrich M, Axt-Fliedner R, Villena-Heinsen C, Tilgen W, Schmidt W, Reichrat J 2002b Analysis of vitamin D receptor (VDR) and retinoid X receptor alpha in breast cancer. Histochem J 34:35–40.

1675 91. Berger U, McClelland RA, Wilson P, Greene GL, Haussler MR, Pike JW, Colston K, Easton D, Coombes RC 1991 Immunocytochemical detection of estrogen receptor, progesterone receptor, and 1,25-dihydroxyvitamin D3 receptor in breast cancer and relation to prognosis. Cancer Research 1991 51:239–244. 92. Colston KW, Berger U, Coombes RC 1989 Possible role for vitamin D in controlling breast cancer cell proliferation. Lancet 1:185–191. 93. Colston KW, Chander SK, Mackay AG, Coombes RC 1992a Effects of synthetic vitamin D analogs in breast cancer cell proliferation in vivo and in vitro. Biochem Pharmacol 44: 1153–1155. 94. Binderup L, Bramm E 1988 Effects of a novel vitamin D analog MC903 on cell proliferation and differentiation in vitro and on calcium metabolism in vivo. Biochem Pharmacol 37:887–895. 94. Colston KW, Mackay AG, James SY, Binderup L, Chander S, Coombes RC 1992b EB1089: a new vitamin D analog that inhibits the growth of breast cancer cells in vivo and in vitro. Biochem Pharmacol 44:2273–2280. 95. Bower M, Colston KW, Stein RC, Hedley A, Gazet J-C, Ford HT, Coombes RC 1991 Topical calcipotriol treatment in advanced breast cancer. Lancet 337:701–702. 96. Andoh T, Lino Y 1996 Usefulness of 22-oxa-1,25dihydroxyvitamin D3 (OCT) as a single agent or combined therapy with aromatase inhibitor (CGS 16949A) on 7,12dimethylbenz[a]anthracene-induced rat mammary tumors. Int J Oncology 9:79–82. 97. Mørk Hansen C, Hamberg KJ, Binderup E, Binderup L 2000 Seocalcitol (EB1089): a vitamin D analog of anti-cancer potential. Background, design, synthesis, preclinical, and clinical evaluation. Curr Pharmaceut Des 6:803–828. 98. Mackay AG, Ofori-Kuragu EA, Lansdown A, Coombes RA, Binderup L, Colston KW 1996 Effects of the synthetic vitamin D analog EB1089 and tamoxifen on the growth of experimental rat mammary tumors. Endocrine Related Cancer 3:327–335. 99. James SY, Mercer E, Brady M, Binderup L, Colston KW 1998 EB1089, a synthetic analog of vitamin D3, induces apoptosis in breast cancer cells in vivo and in vitro. Br J Pharmacol 125:953–962. 100. Koshizuka K, Koike M, Asou H, Cho SK, Stephen T, Rude RK, Binderup L, Uskokovic M, Koeffler HP 1999a Combined effect of vitamin D3 analogs and paclitaxel on the growth of MCF-7 breast cancer cells in vivo. Breast Cancer Research Treatment 53:113–120. 101. Koshizuka K, Kubota T, Said J, Koike M, Binderup L, Uskokovic M, Koeffler HP 1999b Combination therapy of a vitamin D analog and all trans retinoic acid. Effect on human breast cancer in nude mice. Anticancer Research 19:519–524. 102. El Abdaimi K, Dion N, Papavasiliou V, Cardinal P-E, Binderup L, Goltzman D, Ste-Marie L-G, Kremer R 2000 The vitamin D analog EB1089 prevents skeletal metastasis and prolongs survival time in nude mice transplanted with human breast cancer cells. Cancer Research 60:4412–4418. 103. Gulliford T, English J, Colston KW, Menday P, Moller S, Coombes RC 1998 A phase I study of the vitamin D analog EB1089 in patients with advanced breast and colorectal cancer. Br J Cancer 78:6–13. 104. Uskokovic MR, Studzinski GP, Gardner JP, Reddy SG, Campbell MJ, Koeffler HP 1997 The 16-ene vitamin D analogs. Curr Pharmaceut Des 3:99–123.

1676 105. Koike M, Elstner E, Campbell MJ, Asou H, Uskokovic M, Tsuruoka N, Koeffler HP 1997 19-nor-hexafluoride analog of vitamin D3: a novel class of potent inhibitors of proliferation of human breast cell lines. Cancer Research 57:4545–4550. 106. Danielsson C, Mathiasen IS, James SY, Nayeri S, Bretting S, Mørk Hansen C, Colston KW, Carlberg C 1997 Sensitive induction of apoptosis in breast cancer cells by a novel 1,25dihydroxyvitamin D3 analog shows relation to promoter selectivity. J Cell Biochem 66:552–562. 107. Koshizuka K, Koike M, Said J, Binderup L, Koeffler HP 1998 Novel vitamin D3 analog (CB1093) when combined with paclitaxel and cisplatin inhibit growth of MCF-7 human breast cancer cells in vivo. Int J Oncol 13:421–428. 108. Verlinden L, Verstuyf A, van Camp M, Marcelis S, Sabbe K, Zhao X-Y, de Clercq P, Vandewalle M, Bouillon R 2000 Two novel 14-epi analog of 1,25-dihydroxyvitamin D3 inhibit the growth of human breast cancer cells in vitro and in vivo. Cancer Research 60:2673–2679. 109. Hussain EA, Mehta RR, Ray R, Das Gupta TK, Mehta RG 2003 Efficacy and mechanism of action of 1alpha-hydroxy-24ethyl-cholecalciferol (1alpha[OH]D5) in breast cancer prevention and therapy. Recent Results Cancer Res 164:393–411. 110. Findlay DM, Michelangeli VP, Eisman JA, Frampton RJ, Moseley JM, MacIntyre I, Whitehead R, Martin TJ 1980 Calcitonin and 1,25-dihydroxyvitamin D3 receptors in human breast cancer cell lines. Cancer Research 40:4764–4767. 111. Narbaitz R, Sar M, Stumpf WE, Huang S, DeLuca HF 1981 1,25-dihydroxyvitamin D3 target cells in rat mammary gland. Horm Res 15:263–269. 113. Zinser G, Packman K, Welsh JE 2002 Vitamin D3 receptor ablation alters mammary gland morphogenesis. Development 129:3067–3076. 114. Colston K, Berger U, Wilson P, Hadcocks L, Naeem I, Earl H, Coombes R 1988 Mammary gland 1,25-dihydroxyvitamin D3 receptor content during pregnancy and lactation. Mol Cell Endocrinology 60:15–22. 115. Mezzetti G, Barbiroli B, Oka T 1987 1,25-dihydroxycholecalciferol receptor regulation in hormonally-induced differentiation of mouse mammary gland in culture. Endocrinology 120:2488–2493. 116. Kanazawa T, Enami J, Hohmoto K 1999 Effects of 1 alpha, 25-dihydroxycholecalciferol and cortisol on the growth and differentiation of primary cultures of mouse mammary epithelial cells in collagen gel. Cell Biology International 23:481–487. 117. Narvaez CJ, Zinser G, Welsh JE 2001 Functions of 1α,25dihydroxyvitamin D3 in mammary gland: from normal development to breast cancer. Steroids 66:301–308. 118. Bhattacharjee M, Wientroub S, Wonderhaar BK 1987 Milk protein synthesis by mammary glands of vitamin D–deficient mice. Endocrinology 121:865–874. 119. Mezzetti G, Monti M, Casolo L, Piccinini G, Moruzzi M 1988 1,25-dihydroxycholecalciferol-dependent calcium uptake by mouse mammary gland in culture. Endocrinology 122:389–394. 120. Jacobson E, James K, Newmark H, Carroll K 1989 Effects of dietary fat, calcium, and vitamin D on growth and mammary tumorigenesis induced by 7,12-dimethylbenz(a)anthracene in female Sprague-Dawley rats. Cancer Research 49:6300–6303. 121. Anzano MA, Smith JM, Uskokovic MR, Peer CW, Mullen LT, Letterio JJ, Welsh MC, Shrader MW, Logsdon DL, Driver CL, Brown CC, Roberts AB, Sporn MB 1994 1αdihydroxy-16-ene-23-yne-26,27-hexafluorocholecalciferol

KAY COLSTON AND JOELLEN WELSH

122.

123.

124. 125.

126.

127.

128.

129. 130.

131.

132.

133. 134.

135.

136. 137. 138.

(Ro24–5531), a new deltanoid (vitamin D analog) for prevention of breast cancer in the rat. Cancer Research 54: 1653–1656. Mehta R, Hawthorne M, Uselding L, Albinescu D, Moriarty R, Christov K, Mehta R 2000 Prevention of N-methyl-Nnitrosourea-induced mammary carcinogenesis in rats by 1αhydroxyvitamin D5. J Natl Cancer Inst 92:1836–1840. Mehta R, Moriarty R, Mehta R, Penmasta R, Lazzaro G, Constantinou A, Guo L 1997 Prevention of preneoplastic mammary lesion development by a novel vitamin D analog, 1α-hydroxyvitamin D5. J Natl Cancer Inst 89:212–218. Knekt P, Jarvinen R, Seppanen R, Pukkala E, Aromaa A 1996 Intake of dairy products and risk of breast cancer. Br J Cancer 73:687–691. Shin MJ, Holmes MD, Hankinson SE, Wu K, Colditz GA, Willett WC 2002 Intake of dairy products, calcium, and vitamin D and risk of breast cancer. J Natl Cancer Inst. 94:1301–1310. John EM, Schwartz GG, Dreon DM, Koo J 1999 Vitamin D and breast cancer risk: the NHANES I epidemiologic followup study, 1971–1975 to 1992. Cancer Epidemiol Biomarkers Prev 8:399–406. Garland FC, Garland CF, Gorham ED, Young JF 1990 Geographic variation in breast cancer mortality in the United States: a hypothesis involving exposure to solar radiation. Prev Med 19:614–622. Freedman DM, Dosemeci M, McGlynn K 2002 Sunlight and mortality from breast, ovarian, colon, prostate, and nonmelanoma skin cancer: a composite death certificate based case-control study. Occup Environ Med 59:257–262. Grant WB 2002 An ecologic study of dietary and solar ultraviolet-B links to breast carcinoma mortality rates. Cancer 94:272–281. Mawer EB, Walls J, Howell A, Davies M, Ratcliffe WA, Bundred NJ 1997 Serum 1,25-dihydroxyvitamin D may be related inversely to disease activity in breast cancer patients with bone metastases. J Clin Endo Metab 82:118–122. Janowsky EC, Lester GE, Weinberg CR, Millikan RC, Schildkraut JM, Garrett PA, Hulka BS 1999 Association between low levels of 1,25-dihydroxyvitamin D3 and breast cancer risk. Pub Health Nutr 2:283–291. Curran JE, Vaughn T, Lea RA, Weinstein SR, Morrison NA, Griffiths LR 1999 Association of a vitamin D receptor polymorphism with sporadic breast cancer development. Int J Cancer 83:723–726. Yamagata Z, Zhang Y, Asaka A, Kanamori M, Fukutomi T 1997 Association of breast cancer with vitamin D receptor gene. Am J Hum Genet 61:388. Bretherton-Watt D, Given-Wilson R, Mansi JL, Thomas V, Carter N, Colston KW 2001 Vitamin D receptor gene polymorphisms are associated with breast cancer risk in a UK Caucasian population. Br J Cancer 85:171–176. Dunning AM, McBride S, Gregory J, Durocher F, Foster NA, Healy CS, Smith N, Pharoah PDP, Luben RN, Easton DF, Ponder BAJ 1999 No association between androgen or vitamin D receptor gene polymorphisms and breast cancer. Carcinogenesis 20:2131–2135. Newcomb PA, Kim H, Trentham-Dietz A, Farin F, Hunter DKM 2002 Vitamin D receptor polymorphism and breast cancer risk. Cancer Epidemiol Biomarkers Prev 11:1503–1504. Ingles SA, Garcia DG, Wang W, Nieters A, Henderson BE, Coetzee GA 2000 Vitamin D receptor genotype and breast cancer in Latinas (United States). Cancer Causes Contr 11:25–30. Hou MF, Tien YC, Lin GT, Chen CJ, Liu CS, Lin SY, Huang TJ 2002 Association of vitamin D receptor gene polymorphisms

CHAPTER 93 Vitamin D and Breast Cancer

with sporadic breast cancer in Taiwanese patients. Breast Cancer Res Treat 74:1–7. 139. Ruggiero M, Pacini S, Aterini S, Allai C, Ruggiero C, Pacini P 2001 Vitamin D receptor gene polymorphism is associated with metastatic breast cancer. Oncology Research 10:43–46. 140. Lundin AC, Soderkvist P, Eriksson B, Bergmann-Jungestrom M, Wingren S 1999 Association of breast cancer progression with a vitamin D receptor polymorphism. Cancer Research 59:2332–2334.

1677 141. Schondorf T, Eisberg C, Wassmer G, Warm M, Becker M, Rein DT, Gohring UJ 2003 Association of the vitamin D receptor genotype with bone metastases in breast cancer patients. Oncology 64:154–159. 142. Whitfield GK, Remus LS, Jurutka PW, Zitzer H, Oza AK, Dang HT, Haussler CA, Galligan MA, Thatcher ML, Encinas Dominguez C, Haussler MR 2001 Functionally relevant polymorphisms in the human nuclear vitamin D receptor gene. Mol Cell Endocrinol 177:45–59.

CHAPTER 94

Vitamin D and Prostate Cancer ARUNA V. KRISHNAN Department of Medicine, Stanford University School of Medicine, Stanford, CA

DONNA M. PEEHL Department of Urology, Stanford University School of Medicine, Stanford, CA

DAVID FELDMAN Department of Medicine, Stanford University School of Medicine, Stanford, CA

I. II. III. IV. V.

Introduction Prostate as a Target for Vitamin D Inhibition of Prostate Cancer Growth by Vitamin D Vitamin D Analogs Mechanisms of Vitamin D–Mediated Growth Inhibition

I. INTRODUCTION A. Scope of the Problem Adenocarcinoma of the prostate gland is the most commonly diagnosed malignancy in American men, excluding skin cancer [1]. The number of cases may actually be substantially underestimated since clinically silent prostate cancer (PCa) is very common. In men over the age of 50, subclinical PCa is found in as many as 40% of individuals [2]. Although PCa is generally a slow growing malignancy, mortality from this disease is nonetheless considerable. In contrast to many other malignancies, the incidence of PCa has continued to rise each year and currently PCa is the second leading cause of cancer death among U.S. men [3]. Over the last two decades the age-adjusted mortality rate from PCa increased 7% among U.S. Caucasian men. Since PCa rates increase with advancing age, one can expect that PCa will become an even greater problem as life expectancy continues to increase. As a result, PCa has rapidly become a major public health concern not only in the U.S. but also worldwide.

B. Etiology, Treatment, and the Role of Hormonal Factors The etiologic factors associated with PCa are varied and include age and race as well as genetic, dietary, and hormonal influences [4]. Androgens are integrally involved in the regulation of prostate growth and cell proliferation [5]. Although several studies have VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

VI. Vitamin D in Combination with Other Agents VII. Clinical Trials VIII. Summary and Conclusions References

demonstrated a correlation between serum testosterone levels and increased risk of PCa [6], it is unclear if androgen plays a dominant role in neoplastic transformation [5]. Circulating androgens do promote tumor growth and perhaps participate in neoplastic transformation. Indeed, the withdrawal of androgens causes involution and apoptosis of normal as well as malignant prostatic epithelial cells [6,7]. This provides the basis for the use of androgen ablation in the treatment of clinically advanced PCa and androgen deprivation therapy remains the mainstay of PCa treatment [3]. At the outset of this chapter, a brief overview of therapeutic considerations would be useful. Early diagnosis of PCa is often made by screening for prostate specific antigen (PSA), one of the most useful tumor markers. A digital rectal exam (DRE) may also indicate prostate enlargement and/or abnormal texture. When indicated, prostate biopsy is performed to confirm the presence of cancer. If the cancer is felt to be confined to the prostate, therapeutic choices for primary therapy are prostatectomy or radiation using external beam or radioactive seed implantation. Elderly men, especially with less invasive pathology on biopsy or those with comorbidities, may be watched (watchful waiting) with monitoring of PSA levels. If the cancer is restricted to the prostate capsule, surgery or radiation may lead to a cure. If the cancer had already escaped the capsule, primary therapy would fail and eventually the PSA would again be found to rise indicating that the cancer had spread. Androgen ablation therapy almost always leads to cancer regression, whether treating cancer recurrence following primary therapy or in those men with metastatic cancer. Deprivation of androgens can Copyright © 2005, Elsevier, Inc. All rights reserved.

1680 be accomplished by orchiectomy or more commonly today by pharmacological means using gonadotropin releasing hormone (GnRH) analogs such as lupron, zoladex, or other drugs that inhibit luteinizing hormone (LH) release. Additional therapy may include the use of anti-androgen drugs (androgen antagonists), such as flutamide or casodex, that block binding of androgens to the androgen receptor (AR). A recent trial suggested that the 5α-reductase inhibitor finasteride might reduce prostate cancer incidence, although the cancers that did occur might be more aggressive [8]. Initially, almost all patients respond to andorogen ablation therapy. However, many if not all patients eventually fail as the cancer develops the ability to grow in the absence of androgens [9,10]. This is called androgen-independent prostate cancer (AIPC) or hormone refractory prostate cancer (HRPC) [3]. AIPC is the progressive and metastatic form of the disease and unfortunately, it is not amenable to current therapies. This transition of PCa to AIPC remains both a therapeutic as well as an experimental challenge. Growth of the normal or malignant prostate is dependent on androgens. Circulating testosterone is converted to the more potent dihydrotestosterone (DHT) within the prostate by the enzyme 5α-reductase. Both testosterone and DHT act via classical AR to stimulate growth, cell survival, PSA, and other androgen-regulated genes. Most cases of AIPC retain AR. There are several molecular mechanisms underlying AIPC development including: a) amplification of the AR gene, which increases AR sensitivity to even low levels of androgens; b) mutations in the AR gene that broaden ligand specificity so that nonandrogens or even androgen antagonists can stimulate prostate growth; c) androgen-independent activation of the AR by growth factor signaling pathways or activation of AR coactivators; and d) pathways that are independent of AR [10]. Evidence is now accumulating for the etiologic role of AR gene mutations in the pathogenesis of some cases of AIPC [9–11]. The presence of missense mutations in the AR gene renders the AR “promiscuous” and allows the inappropriate activation of the AR by nonandrogen steroids as well as AR antagonists leading to AIPC [9–16]. Nonandrogen steroid hormones such as estrogens, progestins, as well as glucocorticoids may activate AR containing mutations in its ligandbinding domain and thereby stimulate the growth of PCa cells that harbor such promiscuous mutant ARs [17–19]. Some of these AR mutations also enable AR antagonists to acquire agonistic activity leading to a failure of androgen ablation therapy [20,21]. Other steroid hormones, such as progestins and estrogens, may play a role in PCa. Receptors for both progesterone and estrogen have been observed in

ARUNA V. KRISHNAN, DONNA M. PEEHL, AND DAVID FELDMAN

prostatic tumors [22]. The presence of estrogen receptor (ER)-α and -β has been demonstrated in normal prostate, dysplasia, and cancer [23]. Estrogens and selective estrogen receptor modulators (SERMs) have been shown to modulate the growth and even induce apoptosis in PCa cell lines [24]. Estrogens have been shown to stimulate the growth of LNCaP, human PCa cells [25]. A selective ER antagonist ICI 182,780 (faslodex) caused growth inhibition in the human PCa cell lines PC-3 and DU 145, and this growth inhibition was abrogated by an ER anti-sense oligonucleotide [26]. Phytoestrogens, such as genistein, daidzein, and equol, also inhibited the growth of human prostate cancer cells [27]. However, in addition to acting through their own receptors, estrogens, SERMs, progestins, glucocorticoids, and other steroid hormones can activate mutated ARs in PCa cells or tumors harboring promiscuous AR mutations, and thereby modulate cancer cell growth [10]. One of the goals of current research on PCa and AIPC is the identification of new agents that would prevent and/or slow down the progression of this disease and in recent years vitamin D has emerged as a promising therapeutic agent [28–34]. We describe in the following sections several lines of evidence for the potential benefits of vitamin D in PCa.

C. Epidemiology 1. SUNLIGHT EXPOSURE

Based upon epidemiological studies, several risk factors for prostate cancer have been identified including age, race, and genetic factors [7,35]. Environmental factors may also have a strong influence upon the expression of the disease. For example, when compared to Nigerian men, African American men have a sixfold increased risk of developing clinically detectable prostate cancer [7]. It has long been appreciated that solar radiation can decrease the mortality rates of noncutaneous malignancies [36,37]. Age is the strongest risk factor for prostate cancer, and the elderly are frequently vitamin D–deficient due to several factors, including less exposure to UV radiation (see Chapter 3). Of particlular interest is the hypothesis put forward by Schwartz and colleagues [38,39] suggesting a role for vitamin D in decreasing the risk of developing prostate cancer (see Chapter 90). Their hypothesis is based upon the observation that prostate cancer mortality rates in the U.S. are inversely proportional to the geographically determined incident UV radiation exposure from the sun, and that UV light is essential for vitamin D synthesis. This hypothesis is not without precedent, as vitamin D may have a role in the prevention of colon cancer as well (see Chapter 95). It may also offer a

1681

CHAPTER 94 Vitamin D and Prostate Cancer

potential explanation of why African American men have a higher incidence of prostate cancer than Caucasian men [40]. African American individuals have lower serum 25-hydroxyvitamin D [25(OH)D] levels as a result of their darker skin pigmentation because the high melanin levels in darkly pigmented skin block UV radiation and inhibit the formation of vitamin D3 [41] (see Chapters 3 and 47). The findings of a recent study of patients with PCa and benign prostatic hyperplasia (BPH) [42] suggest that susceptibility to PCa is in part determined by the extent of exposure to UV radiation and that the ability to pigment the skin modulates this effect. 2. DIET AND SERUM LEVELS OF VITAMIN D METABOLITES

Dietary forms of vitamin D include vitamin D– supplemented milk and other foods, ergocalciferol (vitamin D2) in plants, and vitamin D3 in animal products. Diet has been proposed as a risk factor for PCa, and the low risk of PCa for indigenous Japanese has been postulated to be related to their traditional diet [43]. This diet, among other attributes, is rich in oily fish, which are an important dietary source of vitamin D3. Some epidemiological studies show that high levels of dietary calcium are a significant risk factor for prostate cancer [44,45] (see Chapter 91). This may be relevant to the relationship of vitamin D and PCa because high levels of serum calcium suppress parathyroid hormone and reduce the renal production of 1,25-dihydroxyvitamin-D, (1,25(OH)2D3). One of these studies [44] also shows that a high intake of fruit is associated with a decreased risk for PCa. The authors suggest that high intake of fruit-derived fructose would lead to hypophosphatemia, which stimulates 1,25(OH)2D production. These observations provide indirect support for the possible protective role of high 1,25(OH)2D levels on PCa. However, another recent study does not find an association between vitamin D intake and PCa risk in a population of American men [46]. Corder et al. [47] undertook a case-control study and determined the levels of vitamin D metabolites in stored sera collected in the San Francisco Bay area between 1964 and 1971 and matched for age, race, and day of serum storage. This study looked at a nested group of sera from men who ultimately developed PCa out of a larger population of 250,000 individual samples. Mean levels of 1,25(OH)2D, the active metabolite of vitamin D, were slightly but significantly lower in men who went on to develop PCa when compared to controls who did not develop cancer. The risk of PCa decreased with higher 1,25(OH)2D concentration, especially in men with low 25(OH)D. In men over the age of 57, serum 1,25(OH)2D concentration was an important risk factor for palpable and anaplastic tumors but not

for incidentally discovered tumors or well-differentiated cancers. Another prediagnostic study carried out later, in a group of 20,305 men in Maryland, by Braun et al. [48] could not confirm these results. A study by Gann et al. [49] also did not find a correlation between levels of circulating vitamin D metabolites and subsequent development of PCa. However in a further analysis of their data, Corder et al. [50] subsequently found that 1,25(OH)2D levels showed a seasonal variation in PCa cases but not in controls, with a nadir in summer months. This may explain the lack of an effect in the Braun study due to the fact that serum collections occurred mainly in the fall, and as a result might have missed a nadir in levels among individuals that went on to develop PCa. A nested case-control study in JapaneseAmerican men conducted in Hawaii [51] also did not find a strong association between serum 1,25(OH)2D levels and the incidence of PCa, possibly due to the lack of sufficient number of study subjects with low vitamin D levels. 3. VITAMIN D BINDING PROTEIN

The vitamin D binding protein (DBP) may modulate vitamin D action by controlling the levels of free 25(OH)D or 1,25(OH)2D available to activate the VDR (see Chapter 8). As such, it may play a role in the etiology of PCa. However, studies in this area have produced divergent results. Corder et al. [50] failed to find an association with DBP and PCa in their cohort of subjects. In a group of 68 men with PCa, Schwartz et al. [52] found that DBP levels were significantly higher in individuals with PCa compared to controls. Individuals with DBP levels >350 mg/L had a greater than fivefold increase in PCa risk.

D. Genetic Factors 1. VDR POLYMORPHISMS AND PCA RISK

As in other target tissues, the mechanism of action of 1,25(OH)2D in the prostate involves hormone action through the classical ligand-dependent activation of genes via the vitamin D receptor (VDR) (see Chapters 11,13). Several polymorphisms have been identified in the VDR gene that may contribute to the risk of osteoporosis [53] (see also Chapter 68). Some of these polymorphisms may contribute to PCa risk as well [54]. Ingles et al. [55] studied the VDR polymorphism due to an increased number of adenosine residues in a microsatellite polyA tract in the noncoding region of exon 9. They found that the presence of the long (L) VDR allele (with polyA of > 18) versus the short allele (S with poly-A < 18), whether in the homozygous state (LL) or heterozygous state (LS), was associated with a four- to fivefold increase in PCa risk. This polymorphism is linked to

1682 the more commonly studied VDR polymorphisms at the BsmI and TaqI sites in intron 8 and exon 9 of the VDR gene. Taylor et al. [56] examined the association between the TaqI polymorphism and PCa risk. Their results showed that men homozygous for the t allele (presence of the TaqI site) had one-third less risk of developing PCa requiring prostatectomy [56]. Based on linkage disequilibrium between the T allele and the L allele, it appears that these VDR polymorphisms are associated with an increased risk of PCa. However, a recent study of the TaqI polymorphism in 400 patients with BPH who have been followed clinically for a median of 11 years did not find an association between the risk of developing PCa and the TaqI variant genotype [57]. Several studies also report a lack of association between PCa risk and the TaqI polymorphism or poly-A tract length [58–61]. Some recent reports show that the presence of the B genotype (lack of BsmI site) in the homozygous (BB) or heterozygous (Bb) state lowers the risk of PCa when compared to the bb genotype [62,63]. However, Suzuki et al. [64] could not confirm any significant association between the BsmI, ApaI, and TaqI VDR polymorphisms and familial PCa risk in a Japanese population. The VDR polymorphisms discussed so far do not alter the amino acid sequence of the VDR protein. However, it is possible that they could alter VDR mRNA expression or stability and thereby affect the abundance of VDR. The FokI polymorphism at the start codon of the VDR gene results in amino acid changes in the VDR protein. The ATG variant (f allele or M1 containing the FokI site) initiates at this site and codes for a VDR three amino acids longer than the ACG variant (F allele or M3 without the FokI site) that initiates at the second ATG, 3 amino acids downstream [65]. Xu et al. [66] examined the association of the FokI genotype with the histopathological characteristics and prognosis of PCa among cancer patients who had undergone radical prostatectomy. They found that subjects with the ff genotype had a significantly lower mean percentage of Gleason grade 4/5 cancer and concluded that the ff genotype was associated with less aggressive histopathological findings than the Ff or FF genotypes [66]. Interestingly, regarding the FokI polymorphism, it appears that the alleles that may be protective against PCa in men may be predictive of low bone mass in some groups of women (see Chapter 68). In a population of PCa patients from Shanghai, Chokkalingam et al. [67] did not find an association between the FokI polymorphism and PCa risk. Interestingly, however, these investigators found that in men with the ff genotype, those in the highest tertile of plasma insulin-like growth factor binding protein-3 (IGFBP-3) had a decreased risk of PCa versus those in

ARUNA V. KRISHNAN, DONNA M. PEEHL, AND DAVID FELDMAN

the lowest tertile. These results suggest that the IGF/IGFBP axis and vitamin D regulatory systems may interact to affect PCa risk. The clinical utility of these VDR polymorphisms in predicting PCa risk is not entirely clear and the subject requires further investigation. Haussler and colleagues [68] have suggested that the F and f alleles have differential ability to transactivate target genes explaining why the FokI site might alter disease risk (see Chapters 13 and 68). This data may rationalize why the F variant may be protective in osteoporosis in women but it is counterintuitive for the increased risk or worse prognosis of prostate cancer in men associated with the F genotype.

II. PROSTATE AS A TARGET FOR VITAMIN D A. Vitamin D Is an Antiproliferative and Prodifferentiation Agent Although the role of vitamin D in maintaining calcium homeostasis has been understood for a long while (see Chapter 24), it is only recently that investigators have begun to understand the broader scope of vitamin D actions [69]. In addition to exhibiting immunomodulatory effects (see Chapter 36), 1,25(OH)2D has been shown to have antiproliferative and prodifferentiating actions in a number of tumors and malignant cells including PCa [29–34,70], raising the possibility of its use as an anti-cancer agent. The epidemiological evidence described above supports the notion that the prostate represents a vitamin D target organ. The finding by Miller et al. [71] of the presence of VDR in LNCaP human PCa cells and the demonstration of VDR and its antiproliferative actions in three different PCa cell lines including LNCaP, PC-3, and DU 145 cells by Skowronski et al. [72] were important early findings that suggested that 1,25(OH)2D might play a direct role in prostate biology.

B. VDR in the Normal Prostate Although the initial description of VDR in prostate and most of the subsequent investigation has centered around PCa cell lines, 1,25(OH)2D also appears to play an important role in normal prostate tissue. Peehl et al. [73] reported the presence of VDR in freshly obtained surgical prostate specimens as well as primary cultures of epithelial and stromal cells of the prostate. Primary cultures from surgical specimens of BPH also demonstrated VDR [73]. Although VDR were present in both epithelial and stromal cells cultured separately, lower

1683

CHAPTER 94 Vitamin D and Prostate Cancer

III. INHIBITION OF PROSTATE CANCER GROWTH BY VITAMIN D A. In Vitro Studies in Prostate Cells 1. CANCER CELL LINES

Miller and co-workers [71] demonstrated the presence of VDR in LNCaP human PCa cells. In their study, 1,25(OH)2D3 at concentrations from 10−11–10−9 M was slightly stimulatory to cell growth, when the cells were cultured in media supplemented with charcoalstripped serum depleted of endogenous androgens, conditions under which LNCaP cells grew very poorly. A subsequent study by Skowronski et al. [72] demonstrated the presence of VDR in LNCaP cells as well as two other human prostate cancer cell lines DU 145 and PC-3. Interestingly, Skowronski and co-workers found 1,25(OH)2D3 to exert growth-inhibitory effects upon these cell lines (Fig. 1). The growth inhibition in LNCaP cells was quite striking (~ 60%) when cultured in the presence of increasing concentrations of 1,25(OH)2D3 up to 100 nM in regular growth medium containing 5% fetal bovine serum (FBS), a medium supporting robust cell growth. The discrepancy between the effect of 1,25(OH)2D3 on the growth of LNCaP cells reported in these studies may be due to differences in the culture conditions. Indeed, as shown in a later study by Zhao et al. [80], androgens present in serum in the growth medium influenced the effect of vitamin D on LNCaP cell growth (see Section III.C below). Esquenet et al. [81]

120

DNA content (% control)

levels of VDR were seen in the stromal fibroblasts compared to cells of the glandular epithelium. The region of origin within the prostate tissue did not influence the abundance of VDR, as both the peripheral zone and central zone cultures had similar amounts of VDR [73]. The presence of VDR has also been demonstrated in the secretory epithelium and stromal cells from human prostate tissue [74] and in human neonatal prostatic epithelial cells transformed with Simian Virus 40 (SV40) [75]. Krill et al. [76] studied VDR expression in normal prostate glands from donors of various age groups and found that VDR expression changed with age with peak levels in the fifth decade and a decline thereafter. Vitamin D has been shown to exert antiproliferative effects on rat neonatal prostatic epithelial cells [77] and human prostatic epithelial cells [73,78]. Konety et al. [77,79] showed that in the rat, exposure of pups in utero to administered 1,25(OH)2D3 influenced prostatic growth and differentiation throughout the life of the animal. In general, all of these studies support a role of vitamin D in normal prostate physiology and growth.

100 80 60 40 20 0 0

0.01

0.1

1

10

100 1000

1,25(OH)2D3 (nM)

FIGURE 1 Dose response effect of 1,25(OH)2D3 on the proliferation of human prostate cancer cell lines. LNCaP (circles), PC-3 (triangles) and DU 145 (squares) cells seeded in 6-well plates were treated for 6 days with vehicle or indicated concentrations of 1,25(OH)2D3 in culture media. Proliferation was assessed by the determination of DNA content at the end of 6 days. Values shown are mean ± SD from two to four experiments done in duplicate. From Skowronski et al. [72] with permission.

also reported significant inhibition of LNCaP cell growth by 1,25(OH)2D3 in the presence of androgens. The study by Skowronski et al. [72] demonstrated growth inhibition by vitamin D in other PCa cells such as PC-3 and DU 145 (Fig. 1). The magnitude of growth inhibition was less in PC-3 cells (~40–50%) when compared to LNCaP cells, and the growth inhibition seen in DU 145 cells was minimal. Interestingly, Skowronski et al. also reported an inverse correlation between 1,25(OH)2D3 induction of 25-hydroxyvitamin D 24-hydroxylase (24-hydroxylase) and 1,25(OH)2D3mediated growth inhibition in these cells. They found that 24-hydroxylase mRNA was induced by 1,25(OH)2D3 maximally in DU 145 and substantially in PC-3 cells while the induction of 24-hydroxylase mRNA was not detected in LNCaP cells (see Section III.B below). So there is limited correlation between growth inhibition and induction of specific genes. 1,25(OH)2D3 has also been shown to inhibit the growth of other prostate cancer cell lines such as ALVA 31, PPC-1 [82], and MDA PCa 2a and 2b cells [83]. 2. PRIMARY PROSTATE CELLS

The use of primary cultures of prostate cancer cells provides an important tool to study cancer biology as they may be more closely related to the clinical setting than established cancer cell lines. Peehl et al. [73] reported the results of their investigation of a series of primary prostate cancer cell cultures. Epithelial cultures were generated from surgical specimens obtained

1684

ARUNA V. KRISHNAN, DONNA M. PEEHL, AND DAVID FELDMAN

120

E-PZ-2 E-PZ-6 E-CZ-3 E-CZ-4 E-BPH-2 E-BPH-3 E-CA-11

Percent of control

100

80

60 40 20

0

0

0.0025

0.025

0.25

2.5

25

1,25-(OH)2 vitamin D3 (nM)

FIGURE 2 1,25(OH)2D3 causes growth inhibition in normal and malignant primary prostatic epithelial cells. Proliferation was determined by clonogenic assays in serum-free medium after 10 days of treatment with increasing doses of 1,25(OH)2D3 as indicated. E-PZ and E-CZ represent normal prostatic epithelial cell lines. E-BPH-2 and -3 are epithelial cell cultures derived from benign prostatic hyperplasia specimens. E-CA-11 refers to an epithelial cell culture generated from prostatic carcinoma. From Peehl et al. [73] with permission.

during prostatectomy. 1,25(OH)2D3 ligand-binding experiments revealed variable levels of VDR abundance among the different cultures. The VDR abundance did not correlate with histological grade of the tumors. As shown in Fig. 2, in clonogenic growth assays, 1,25(OH)2D3 exhibited a substantial inhibition (ED50 = 0.25 − 1 nM) of the growth of these primary epithelial cells derived from the peripheral and central zones of normal prostate as well as from BPH and cancer. Interestingly, the growth inhibition was irreversible after withdrawal of 1,25(OH)2D3 from the medium [73]. In contrast, the growth inhibition seen in LNCaP cells is reversible [81]. Peehl et al. [73] also studied cultures of stromal fibroblasts generated from the prostate surgical specimens and found that these cells also contained VDR and were inhibited by 1,25(OH)2D3. The abundance of VDR was less in the stromal fibroblasts, and the growth inhibition by 1,25(OH)2D3 was less when compared to the epithelial cultures. The significance of the presence of VDR in the fibroblastic component of the prostate is uncertain. However, there is evidence that fibroblasts may influence prostate epithelial cell growth through the production of growth factors that act in a paracrine manner [84,85]. 1,25(OH)2D3 may potentially regulate stromal production of some of these growth factors and thereby control the growth of prostatic epithelium [86,87].

3. VIRALLY TRANSFORMED PROSTATE CELLS

The large T antigen of simian virus 40 (SV40) and the E6 and E7 proteins of human papilloma virus (HPV) cause disruptions in the p53 and retinoblastoma (Rb) genes [88,89] and have been used to transform prostate cells into immortalized cell lines [90,91]. Studies of such virally transformed prostate cell lines are useful models to help gain an understanding of the role of p53 and Rb genes in the pathogenesis of prostate cancer and to potentially elucidate the mechanism of the growth-inhibitory action of 1,25(OH)2D3. Investigators have evaluated VDR and 1,25(OH)2D3 action in both SV40- and HPV-transformed cell lines [90,92,93]. These studies showed that high affinity VDRs were present in the SV40 and HPV transformed cell lines and that 24-hydroxylase mRNA was induced by 1,25(OH)2D3 treatment of these virally transformed cell lines. However, the effect of 1,25(OH)2D3 on growth was different in the SV40- and HPV-transformed cell lines. Whereas the growth of HPV-transformed cell lines was inhibited by 1,25(OH)2D3, the SV40-transformed cell lines were resistant to 1,25(OH)2D3 action in terms of an antiproliferative response. As discussed in the section below, the expression of the large T antigen appears to cause resistance to 1,25(OH)2D3 action on proliferation in prostate as well as breast epithelial cells due to an inhibition of the transcriptional activity of VDR [94]. 4. RESISTANCE OF SOME PROSTATE CELLS GROWTH INHIBITION BY VITAMIN D

TO

While many of the established PCa cell lines or primary cultures of adenocarcinoma-derived cells respond to vitamin D, there is evidence that some PCa cells may become resistant to antiproliferative activity of 1,25(OH)2D3. This resistance can develop at several levels: (i) through the loss or decreased expression of VDR or the retinoid X receptor (RXR), (ii) through VDR polymorphisms that diminish its function, (iii) through elevated expression of VDR co-repressors, (iv) by increased expression of enzymes that metabolize 1,25(OH)2D3, or (v) by other means. An example of a loss of response to vitamin D due to the loss of VDR is seen in the cancer cell line JCA-1 which does not respond to 1,25(OH)2D3 with growth inhibition [95]. VDR is not detectable in these cells and stable transfection of the cells with a VDR cDNA makes these cells sensitive to growth inhibition by vitamin D [95]. As discussed above, transformation of prostatic epithelial cells by SV40 but not HPV results in a loss of growth inhibitory effect of 1,25(OH)2D3 in the transformed cells. Human breast epithelial cells transformed by the SV40 large T antigen have also been shown to become resistant to 1,25(OH)2D3 [94], demonstrating

CHAPTER 94 Vitamin D and Prostate Cancer

the generality of this finding. The authors of this study showed that the expression of the large T antigen strongly inhibited 1,25(OH)2D3–induced VDRE transcriptional activity in a dose-dependent manner and that increasing the VDR concentration could reverse the inhibitory effect of the large T antigen. Indeed, the resistance to 1,25(OH)2D3–mediated growth inhibition seen in the breast epithelial cells could be overcome by the overexpression of VDR in the cells [94]. As shown previously [72], 1,25(OH)2D3 did not inhibit the growth of DU 145 cells despite the presence of a functional VDR in these cells. DU 145 cells, like JCA-1 and PC-3 cells, have relatively low numbers of VDR, and stable transfection of VDR into these cells somewhat restored the growth inhibitory activity of 1,25(OH)2D3 [96]. However, DU 145 cells also have the highest levels of 1,25(OH)2D3–inducible 24-hydroxylase activity among the different prostate cancer cell lines studied so far. 24-Hydroxylase initiates the catabolism of 1,25(OH)2D3 to inactive metabolites. As discussed in detail in the following section (Section III.B), adding an inhibitor of 24-hydroxylase resulted in a significant inhibition of DU 145 cell growth by 1,25(OH)2D3 [97]. One important concept that has emerged from these studies is that different signaling pathways are used by VDR to regulate genes involved in growth control versus those involved in other functions. This is apparent in the ability of cells resistant to growth inhibition by 1,25(OH)2D3 to still exhibit VDR up-regulation [97] and/or 24-hydroxylase induction in response to 1,25(OH)2D3 [72,97].

B. Cellular Responsiveness to 1,25(OH)2D3—Role of Enzymes Involved in Vitamin D Metabolism The key enzymes involved in vitamin D metabolism are 24-hydroxylase (CYP24), which catalyzes the initial step in the conversion of 1,25(OH)2D3 to less active metabolites (see Chapter 6) and 1α-hydroxylase (CYP 27B1), which catalyzes the synthesis of 1,25(OH)2D3 from 25(OH)2D3 (see Chapter 5). As discussed in the following sections the level of expression of these enzymes in target cells such as PCa cells influences the magnitude of the growth-inhibitory responses to vitamin D metabolites. 1. 25-HYDROXYVITAMIN D3 24-HYDROXYLASE (CYP24)

1,25(OH)2D3 induces the expression of the enzyme 25-hydroxyvitamin D3 24-hydroxylase (24-hydroxylase) in target cells, which catalyzes the initial step in the conversion of the active molecule 1,25(OH)2D3 into less active metabolites. An interesting study by

1685 Albertson et al. [98] used comparative genomic hybridization (CGH) to resolve two regions of amplification within an approximately 2 Mb region of recurrent aberration at 20 q13.2 in breast cancer. A known putative oncogene ZNF217 mapped to one peak, and CYP24 (encoding vitamin D 24-hydroxylase), whose overexpression likely leads to the abrogation of growth control mediated by vitamin D, mapped to the other, raising the possibility that the CYP24 gene encoding 24-hydroxylase might be an oncogene. It is clear that in normal as well as malignant cells induction of 24-hydroxylase by 1,25(OH)2D3 would limit its own growth-inhibitory actions. In prostate cells, the degree of growth inhibition of vitamin D has been shown to be inversely proportional to the 24-hydroxylase activity in the cells. Among the human PCa cell lines DU 145, PC-3, and LNCaP, DU 145 cells exhibit the highest level of 24-hydroxylase induction and are the least responsive to 1,25(OH)2D3 in terms of growth inhibition [72,82]. On the other hand, the basal and induced expression of 24-hydroxylase is very low in LNCaP cells, and growth inhibition by 1,25(OH)2D3 is substantial. Ly et al. [97] examined the possibility that inhibition of 24-hydroxylase activity would render DU 145 cells more sensitive to the antiproliferative effect of 1,25(OH)2D3. The results of their investigation show that in DU 145 cells, liarozole (an imidazole derivative that inhibits P450 hydroxylases) causes significant inhibition of 24-hydroxylase activity leading to an increase in 1,25(OH)2D3 half-life in DU 145 cells and thereby allows a substantial antiproliferative effect [97]. In many target cells, 1,25(OH)2D3 has been known to cause homologous up-regulation of its receptor levels [99,100]. Prolongation of 1,25(OH)2D3 half-life (due to the inhibition of 24-hydroxylase by liarozole) therefore resulted in an enhanced up-regulation of VDR in DU 145 cells, making them more responsive to 1,25(OH)2D3. As shown in Fig. 3, the growth inhibition due to the treatment of DU 145 cells with 1,25(OH)2D3 alone or liarozole alone was minimal while the combination of both these agents resulted in a substantial inhibition of cell growth. This increase in growth inhibition could be due to both an extension of the 1,25(OH)2D3 half-life and enhanced VDR up-regulation resulting from the inhibition of 24-hydroxylase by liarozole. Miller et al. [82] also demonstrated that the differences in 1,25(OH)2D3–mediated growth inhibition between various PCa cell lines correlate inversely to 24-hydroxylase expression in these cells. A recent study by Peehl et al. [101] has shown that in primary human PCa cells, the use of the P450 inhibitor ketoconazole potentiates the growth inhibitory effects of 1,25(OH)2D3 or its structural analog EB 1089 by

1686

ARUNA V. KRISHNAN, DONNA M. PEEHL, AND DAVID FELDMAN

120 VD 10 (nM) Liarozole (1 µM)

100

Growth (% of control)

Both 80

60

40

20

0 0

2

4 Time (days)

6

8

FIGURE 3

Effect of 1,25(OH)2D3 and liarozole on DU 145 cell growth. Effect of 1,25(OH)2D3 (10 nM), liarozole (1 µM), or a combination of both on DU 145 cell growth was determined over a time course of 6 days with fresh medium containing the agents being replenished every 2 days. Cell proliferation was estimated using the MTI assay. Data are expressed as mean ± SD of three experiments. *p < 0.05 compared with vehicle treated control group. From Ly et al. [97] with permission.

inhibiting the 24-hydroxylase activity in these cells. Thus, combinations of 1,25(OH)2D3 with inhibitors of 24-hydroxylase such as ketoconazole or liarozole may enhance its anti-tumor effects in PCa therapy. The combination approach may also allow the use of 1,25(OH)2D3 at lower concentrations thereby reducing the hypercalcemic side effects. 2. 25-HYDROXYVITAMIN D3 1α-HYDROXYLASE (CYP27B1)

The active hormone 1,25(OH)2D3 is formed in the kidney by the hydroxylation of 25(OH)D3 at the C-1 position by the enzyme 1α-hydroxylase (see Chapter 5). The kidneys are the major source of circulating 1,25(OH)2D3 in the body. In recent years, however, the presence of extra-renal 1α-hydroxylase has been demonstrated, which contributes to the local production of 1,25(OH)2D3 within various tissues (see Chapter 79). Schwartz et al. [102] showed that normal human prostatic epithelial cells express 1α-hydroxylase. They raised the possibility that treatment with 25(OH)D3 could potentially inhibit the growth of PCa, due to local production of 1,25(OH)2D3 within the prostate, thus avoiding the systemic side effect of hypercalcemia due to 1,25(OH)2D3 administration. The ability of 25(OH)D3 to cause hypercalcemia is much reduced because of its lower affinity for the

VDR. Results of a study by Barreto et al. [103] support this hypothesis by demonstrating the growth inhibitory effect of 25(OH)D3 in primary epithelial cell strains derived from normal human prostatic peripheral zone. A recent study by Hsu et al. [104] quantitated the levels of 1α-hydroxylase in primary prostatic epithelial cells derived from normal tissue, BPH, or cancer as well as in established PCa cell lines. This study shows that epithelial cells from normal prostate have more 1α-hydroxylase activity than those derived from BPH or cancer. The activity in primary cancer cells is lower than BPH, and the PCa cell lines express the lowest 1α-hydroxylase activity. Whitlatch et al. [105] similarly found reduced 1α-hydroxylase activity in prostate cancer cells compared to normal prostatic cells. The decrease in 1α-hydroxylase enzyme activity in PCa cells may arise from a decrease in 1α-hydroxylase gene promoter activity in these cells [106]. Segersten et al. [107] examined 1α-hydroxylase expression by RT-PCR and immunohistochemical analyses and reported that the expression of 1α-hydroxylase is lower in parathyroid carcinomas, compared with normal parathyroid tissue. However, studies on tissues derived from normal colon and colon carcinoma show elevated levels of 1α-hydroxylase in colon carcinoma [108]. The Hsu et al. study [104] also shows that the antiproliferative effect of 25(OH)D3 correlates

1687

CHAPTER 94 Vitamin D and Prostate Cancer

A

80 60 40 20

120 100 80 60 40 20 0 0

140

0.01 0.1 1 10 Concentration (nM)

100 80 60 40 20 0 0

140

E-CA-10

120 100 80 60 40 20 0 0

E-PZ-12

120 100 80 60 40 20 0

1 10 0.01 0.1 Concentration (nM)

DNA content (% of control)

E-CA-6 DNA content (% of control)

DNA content (% of control)

140

1 10 0.01 0.1 Concentration (nM)

120

0

1 10 0.01 0.1 Concentration (nM)

E-CA-12

120 100 80 60 40 20

LNCaP

140 120 100 80 60 40 20 0

0 0.01 0.1 1 10 Concentration (nM)

C DNA content (% of control)

100

140

E-PZ-8 DNA content (% of control)

120

0 0

B

140

E-CZ-2 DNA content (% of control)

DNA content (% of control)

140

0

1 10 0.01 0.1 Concentration (nM)

0

0.01 0.1 1 10 Concentration (nM)

FIGURE 4 Growth inhibition by 25(OH)D3 (solid triangles) in comparison with 1,25(OH)2D3 (open circles) in normal and malignant prostate epithelial cells. Cells were plated and treated with various concentrations of the vitamin D metabolite for 6 days and DNA contents were assayed. A. In normal prostatic epithelial cells (E-CZ-2, E-PZ-8, and E-PZ-12), 25(OH)D3 and 1,25(OH)2D3 were equally growth inhibitory at concentrations between 0.01 and 10 nM. B. In primary cultures of cancer cells (E-CA-6, E-CA-10, E-CA-12), 1,25(OH)2D3 induced ~20–40% more growth inhibition than 25(OH)D3. C. The prostate cancer cell line LNCaP, although fully responsive to 1,25(OH)2D3, was resistant to treatment with 25(OH)D3. From Hsu et al. [104] with permission.

with the endogenous 1α-hydroxylase activity in prostate cells. As illustrated in Fig. 4, the growth of primary epithelial cells from normal tissue or BPH is inhibited by 25(OH)D3 to an extent similar to 1,25(OH)2D3, as it could be converted to 1,25(OH)2D3 by endogenous 1αhydroxylase activity. In contrast, in primary epithelial cells from cancer or in the LNCaP human PCa cell line, with very low endogenous 1α-hydroxylase activity, the antiproliferative action of 25(OH)D3 is much less pronounced in comparison to 1,25(OH)2D3. Importantly, the findings of reduced 1α-hydroxylase in cancerderived prostatic epithelial cells raises the possibility that this difference may endow the malignant cells with an intrinsic growth advantage because of the resultant decrease in the local production of the growth inhibitory agent 1,25(OH)2D3. In addition, local deficiency of 1,25(OH)2D3 may allow cellular de-differentiation and invasion, hallmarks of malignancy. A recent study by Hawkins et al. [109] examined three frequent single nucleotide polymorphisms present in the 1α-hydroxylase gene in PCa patients and control subjects and concluded that this gene does not play a major role in PCa susceptibility. The evidence

from in vitro studies, however, suggests that a decrease in 1α-hydroxylase activity may represent an important mechanism in PCa development and/or progression and that the administration of 25(OH)D3 might be an effective chemopreventive approach while 1αhydroxylase is initially still high within the prostate [104].

C. Vitamin D and Androgen Interactions Androgens acting through the AR regulate prostate growth and play an important role in the development and progression of PCa [5]. In vitro studies have shown that there is cross talk between 1,25(OH)2D3 and androgen signaling in the androgen-responsive PCa cell line LNCaP [80,110]. 1,25(OH)2D3 up-regulates AR gene expression at both mRNA and protein levels and also increases PSA expression in LNCaP cells [111,112]. The secretion of PSA by LNCaP cells is synergistically enhanced when the cells are exposed to a combination of androgens and 1,25(OH)2D3, probably due, in part, to the up-regulation of AR by 1,25(OH)2D3 [110]. The antiproliferative action of 1,25(OH)2D3 in

A 300 Tumor volume (mm3)

LNCaP cells appears to be androgen-dependent as it could be blocked by the AR antagonist casodex [80]. Recent investigations involving cDNA microarray analyses of 1,25(OH)2D3–regulated target genes in LNCaP cells reveal that several of the 1,25(OH)2D3– regulated genes in these cells that modulate cell growth are also androgen responsive genes [113]. However, this androgen-dependent mechanism of 1,25(OH)2D3 action may be specific to LNCaP cells because 1,25(OH)2D3 also inhibits the growth of other PCa cells that do not express the AR [72,73]. Zhao et al. [83] have shown that 1,25(OH)2D3 inhibits the growth and up-regulates AR expression in MDA PCa 2a and 2b cells that were recently established from a bone metastasis in a patient who exhibited advanced AIPC. In contrast to LNCaP, however, the growth-inhibitory action of 1,25(OH)2D3 in the MDA PCa cells appears to be androgen-independent [83]. Yang et al. [114] studied the growth properties of a LNCaP derived androgen-independent subline, LNCaP-104R1, and found that these cells, unlike the parental LNCaP cells, grew well in medium containing charcoal-stripped serum depleted of endogenous androgens. Under these conditions 1,25(OH)2D3 caused significant growth inhibition of these androgen-independent cells that was not abolished by the anti-androgen casodex [114]. Importantly these findings as well as the inhibition of AR negative cells [72,73,83,97] support the potential therapeutic role of vitamin D in the treatment of AIPC.

ARUNA V. KRISHNAN, DONNA M. PEEHL, AND DAVID FELDMAN

200

100 Treatment started

−13 −7

0

4 8 12 14 17 19 Time (days)

B 14 12 10 8 6 4 2 0

C

Sesame oil

EB1089

60 50 Weight (g)

D. In Vivo Studies in Animal Models Although several rodent models of PCa have been developed [115–118], there is still a lack of a perfect animal model for human PCa. Several researchers have developed human prostate tumor xenograft models by transplanting clinical prostate tumors or cultured human PCa cells into immune-deficient mice [119]. Using these animal models investigators have attempted to support the in vitro studies showing inhibition of PCa cells by vitamin D [29–34,70]. As discussed in detail in several sections of this book, the concentrations of 1,25(OH)2D3 required to produce a significant antiproliferative effect in vivo causes hypercalcemia as a side effect. Therefore, investigators have used structural analogs of 1,25(OH)2D3 that exhibit reduced hypercalcemic effects in several in vivo animal studies as well as in clinical trials. Schwartz et al. [120] showed that administration of the vitamin D analog 1,25-dihydroxy-16-ene-23-ynevitamin D3 (Ro23-7553) to mice bearing PC-3 xenografts resulted in a 15% decrease in tumor volume without significant increases in serum calcium levels.

Sesame oil EB1089

0

Serum calcium (mg/dl)

1688

40

Sesame oil EB1089

Treatment

30 20 10 0

−30 −20 −13 −7 0

4

8 12 14 17 19

Time (days)

FIGURE 5

EB 1089 inhibits the growth of LNCaP xenografts in nude mice. One million LNCaP cells were mixed with 100 µl of Matrigel Matrix and injected subcutaneously into one site on the dorsal side of male nude mice (approximately 6 weeks of age). Tumors were allowed to grow to a volume of 150 mm3. Vehicle (sesame oil) or 0.5 µg/kg of EB 1089 were injected intraperitoneally into the mice every other day. Points are an average of treatment groups (n = 10) and bars are mean ± SE. Panel A. Tumor volume. Panel B. Average serum calcium at the end of the study. Panel C. Average body weight in control and treatment groups. From Blutt et al. [121] with permission.

1689

CHAPTER 94 Vitamin D and Prostate Cancer

Blutt et al. [121] showed that intraperitoneal injections of the vitamin D analog EB 1089 reduced growth of LNCaP tumors in a nude mouse xenograft model. As illustrated in Fig. 5, intraperitoneal injections of EB 1089 into athymic male nu/nu mice bearing LNCaP xenografts substantially reduced tumor volume (Fig. 5A) without raising serum calcium levels beyond the normal range (Fig. 5B) or causing weight loss (Fig. 5C). These data suggest that EB1089 is a promising candidate for PCa treatment. A recent study by Koeffler and colleagues reported inhibition of LNCaP xenograft growth in nude mice by three different vitamin D analogs without increases in serum calcium levels [122]. Other models of PCa have been developed in rats, mice, and dogs (reviewed in [118]), in addition to the xenografts of human PCa tumor tissue or cells in immune-compromised hosts. One of the first established and widely used models is the Dunning rat model [123]. The original Dunning R-3327 was a spontaneous rat prostatic adenocarcinoma and a variety of sublines have been developed from it including the highly metastatic MAT LyLu subline which when injected into rats forms highly metastatic tumors [124]. Getzenberg et al. [115] demonstrated that 1,25(OH)2D3 and the analog Ro23-6760 inhibited the growth of the MAT LyLu tumors in rats. Another study of the MAT LyLu tumor model in rats showed that 1,25(OH)2D3 and the analog EB 1089 decreased tumor size and the number of lung metastasis in these animals [125]. EB 1089 was significantly less hypercalcemic than 1,25(OH)2D3 and did not induce severe weight loss [125]. Oades et al. [126] compared the action of 1,25(OH)2D3 and the less hypercalcemic analogs EB1089 and CB1093 in three animal models of prostate cancer, MAT LyLu Dunning prostate model, PAIII tumors in Lobund-Wistar rats, and LNCaP xenografts in nude mice. Although both analogs increased serum calcium levels, the levels were significantly less than in rats treated with 1,25(OH)2D3. Tumor growth was inhibited in male athymic nu/nu mice with LNCaP tumor xenografts. PAIII cells failed to express functional VDR and were insensitive to 1,25(OH)2D3 and its analogs, either in vitro or in vivo. VDRdependent growth inhibition and not the inhibition of angiogenesis was the main mechanism of action of these compounds in vivo [126]. Cumulatively, the results support the notion that the less hypercalcemic analogs of 1,25(OH)2D3 offer a novel therapeutic option for treating prostate cancer. Investigators have attempted to explore the chemopreventive activity, if any, of vitamin D compounds in animal models. Transgenic models of PCa have also been developed in mice (reviewed in [118]). In the Transgenic Adenocarcinoma of Mouse Prostate (TRAMP) model, the prostate-specific rat probasin promoter has been

used to drive the expression of the SV 40 large–T antigen coding region, and mice expressing the transgene display progressive forms of prostatic disease that histologically resemble human PCa [116]. Another model used to test the chemopreventive effect of vitamin D was the G gamma/T-15 transgenic mouse model of AIPC. Although the administration of the vitamin D analog EB 1089 did not alter the onset of tumors in these mice, it slowed down the rate of tumor growth [127]. Xue et al. [128] fed rats a high fat, low calcium and low vitamin D diet, which resulted in the hyperproliferation of the dorsal prostate epithelium which could promote tumorigenesis. Increasing the levels of calcium and vitamin D in the diet inhibited hyperproliferation, providing evidence for the anti-tumor activity of calcium and/or vitamin D in the diet. Some studies have used 1,25(OH)2D3 or its analogs in combination with other therapeutic agents and demonstrated their anti-tumor activity in vivo (for a detailed discussion of the combination approach, see Section VI). Thus in vivo models provide a valuable tool to demonstrate the anti-tumor activity of vitamin D compounds while monitoring their tendency to elevate serum calcium levels and thus validate their use in preparation for clinical trials.

IV. VITAMIN D ANALOGS A. Potency Versus Toxicity The concentrations of 1,25(OH)2D3 used to produce significant antiproliferative effects in cultured PCa cells are often much higher than circulating levels of 1,25(OH)2D3 and whether they can be safely achieved in vivo or not is being investigated (see Chapter 97). Use of high doses of 1,25(OH)2D3 as a treatment for PCa and other cancers predictably results in hypercalcemia and hypercalciuria. This is the only side-effect that has been encountered, but potential hypercalcemia and renal stone formation limit the concentration of 1,25(OH)2D3 that can be administered to patients. Consequently, structural analogs of 1,25(OH)2D3 that effectively activate the VDR but are less hypercalcemic are being developed and evaluated for their potency as antiproliferative agents in vitro and in vivo. Section VIII of this book (Chapters 80–88) extensively reviews the currently available vitamin D analogs and the structure-function relationships that determine the relative hypercalcemic and differentiating potencies of these analogs. Several mechanisms could contribute to the differential activity of analogs on calcium metabolism and antiproliferation and are discussed in detail in Chapters 82 and 83. However, a complete understanding of the mechanistic basis for the separation

1690 of hypercalcemic and antiproliferative activities has not yet been realized.

ARUNA V. KRISHNAN, DONNA M. PEEHL, AND DAVID FELDMAN

also see Fig. 5). It is hoped that some of these vitamin D analogs will emerge as clinically useful anti-cancer agents.

B. Vitamin D Analogs and Prostate Cancer Many investigations have shown that a number of these vitamin D analogs can inhibit the growth of PCa cells in culture at concentrations lower than 1,25(OH)2D3 [129–135]. Several of these analogs have also been shown to slow the growth of prostate xenograft tumors in animals (as discussed in Section III.E). For example, Schwartz et al. [129] showed that the 16-diene analogs were more effective in inhibiting the growth of PCa cells than 1,25(OH)2D3 and unlike 1,25(OH)2D3 the analogs inhibited the growth of DU 145 cells as well. A study by Skowronski et al. [130] demonstrated that the analogs EB 1089, MC903, 22-oxacalcitriol, and Ro24-2637 significantly inhibited the growth of PCa cells with ED50 values lower than for 1,25(OH)2D3. The analogs were also more potent in stimulating the production of PSA regarded as a marker of differentiation in these cells. Similarly, 24-oxo metabolites of vitamin D [136] and fluorine derivatives of the 1,25(OH)2D3 side-chain [131] also exhibited enhanced antiproliferative potencies. Hisatake et al. [137] report that a newly synthesized vitamin D analog [1,25(OH)216-ene-5,6-trans-D3 (Ro 25-4020)], that has a novel 5,6-trans motif, exhibits a 10- to 100-fold increase in antiproliferative activity in breast cancer and PCa cells while exhibiting at least 40-fold less hypercalcemic activity in mice. The higher activity than expected from its binding affinity for VDR is due, in part, to its metabolism to a 24-oxo metabolite, which retains significant biopotency [138]. Derivatives of vitamin D2 that have been shown to be less hypercalcemic than 1,25(OH)2D3 exhibit antiproliferative effects in PCa cells [139,140]. Interestingly, Polek et al. [141] showed that LG190119, one of a series of novel nonsteroidal VDR modulators, also inhibits LNCaP xenograft growth in mice without causing hypercalcemia. Nonsteroidal analogs are discussed in Chapter 88. A recent report [142] showed that the vitamin D analog V (BXL-353) inhibited the growth of BPH cells in culture and stimulated apoptosis. When administered to intact or castrated rats supplemented with testosterone, BXL-353 reduced the androgen effect on ventral prostate weight without causing hypercalcemia or affecting sex hormone secretion, suggesting that it might be useful in the treatment of patients with BPH [142] (see also Chapter 104). Several of the vitamin D analogs have been tested in vivo and found to exhibit tumor inhibitory effects without inducing hypercalcemia (discussed in detail in Section III.E and

V. MECHANISMS OF VITAMIN D– MEDIATED GROWTH INHIBITION Several studies have investigated the molecular mechanisms by which 1,25(OH)2D3 or its analogs exert growth inhibitory effects on cancer cells including PCa cells. As described below, 1,25(OH)2D3 seems to have multiple and diverse actions [143], often cellspecific, including effects on cell cycle arrest, apoptosis, inhibition of metastasis, and angiogenesis (see Chapters 89, 92, and 93). Investigators have also attempted to identify novel 1,25(OH)2D3–target genes that mediate various actions of the hormone, especially the regulation of cell growth.

A. Growth Arrest In many cancer cells, treatment with 1,25(OH)2D3 or its analogs results in the accumulation of cells in the G0/G1 phase of the cell cycle [136], and this has been shown to be the case in LNCaP cells as well [136,144]. Treatment of LNCaP cells with 1,25(OH)2D3 or the analog EB1089 caused an increase in the percentage of cells accumulating in the G1 phase. The combination of 10−8 M 1,25(OH)2D3 and 9-cis retinoic acid resulted in synergistic growth inhibition and caused more cells to accumulate in G1 when compared to 10−8 M 1,25(OH)2D3 alone. There appear to be multiple mechanisms by which 1,25(OH)2D3 causes cell cycle arrest. The retinoblastoma protein (Rb) is a key regulator of G1 to S phase transition. Hyperphosphorylation of the Rb protein by G1 cyclins and their cyclin-dependent protein kinase (CDK) partners inactivates the Rb protein and releases the repression on E2F transcriptional activity allowing cells to progress from G1 to S phase. Zhuang and Burnstein [145] have shown that in LNCaP cells 1,25(OH)2D3 exerts its effects on some of these key steps. 1,25(OH)2D3 treatment of LNCaP cells causes an increase in the expression of the CDK inhibitor p21, a decrease in CDK2 activity leading to a decrease in the phosphorylation of Rb, and repression of E2F transcriptional activity resulting in G1 arrest of the cells. Liu et al. [146] have shown that 1,25(OH)2D3 directly up-regulates p21 gene expression in U937 leukemia cells, acting through a putative vitamin D response element (VDRE) in the promoter of the p21 gene. However, in LNCaP cells, the regulation of p21 gene

1691

CHAPTER 94 Vitamin D and Prostate Cancer

expression appears to be indirect [131,145]. Boyle et al. [147] showed that the induction of insulin-like growth factor binding protein-3 (IGFBP-3) gene expression in these cells by 1,25(OH)2D3 results in increased p21 protein levels. The role of IGFBP-3 in mediating the growth inhibitory effect of 1,25(OH)2D3 is discussed in detail in Section V.D. Moffatt et al. [148] showed that 1,25(OH)2D3 inhibits growth and increases the expression of p21 mRNA and protein levels in ALVA-31 cells. 1,25(OH)2D3–mediated growth inhibition of these cells is abolished by stable transfection of the cells with a p21 antisense construct demonstrating that p21 expression is necessary to mediate the antiproliferative effect of 1,25(OH)2D3 in ALVA-31 cells. A recent study by Polek et al. [149] investigated the role of p53 in 1,25(OH)2D3–mediated cell cycle arrest through increases in p21 levels, as p21 is a known p53 target gene [150]. The findings of this study indicate that p53 is not required to induce growth inhibition or cause accumulation of cells in the G1 phase by 1,25(OH)2D3 in LNCaP cells. However, elimination of p53 function in LNCaP cells reduces G0 arrest as measured by the loss of Ki67 expression, allows the cells to recover from 1,25(OH)2D3-mediated growth arrest, and eliminates the growth inhibitory effects of combinations of 9-cis retinoic acid and 1,25(OH)2D3 [149]. Although a functional Rb plays a key role in cell cycle control, lack of a functional Rb gene in DU 145 cells does not appear to be the critical reason for their reduced sensitivity to growth inhibition by 1,25(OH)2D3. As detailed earlier in Section III.B, a combination of 1,25(OH)2D3 and the 24-hydroxylase inhibitor, liarozole, causes appreciable growth inhibition in these cells [97]. Also, transfection of a functional Rb into DU 145 cells did not render the cells more sensitive to growth inhibition by 1,25(OH)2D3 in vitro [93], even though the Rb-transfected DU 145 cells exhibited reduced tumorigenicity in vivo as xenografts in nude mice [151]. 1,25(OH)2D3 does not increase p21 expression in PC-3 cells, which is consistent with the lack of G1 accumulation of these cells following 1,25(OH)2D3 treatment. Thus, the regulation of cell cycle distribution by 1,25(OH)2D3 appears to be cell-specific and may involve multiple pathways of action.

B. Apoptosis Induction of apoptosis or programmed cell death by 1,25(OH)2D3 is not uniformly seen in all cancer cells (see Chapter 93). In the case of PCa, investigators have mostly focused on LNCaP cells, and the findings have been variable. Zhuang and Burnstein [145] did not detect apoptosis in an adherent population of LNCaP

cells treated with 10 nM 1,25(OH)2D3 using terminal transferase labeling. Hsieh and Wu [152] examined the nonadherent portion of a LNCaP cell population and found a small (35%) increase in hypodiploid cells following 1,25(OH)2D3 treatment that was characteristic of apoptosis. Fife et al. [153] demonstrated DNA fragmentation in LNCaP cells treated with 10 nM 1,25(OH)2D3 for four days. Blutt et al. [154] showed evidence of apoptosis in LNCaP cells treated with 1,25(OH)2D3 for six days and also showed the downregulation of the pro-apoptotic proteins Bcl-2 and Bcl-XL. They went on to demonstrate the involvement of Bcl-2 in 1,25(OH)2D3–mediated apoptosis by stably transfecting the Bcl-2 gene into LNCaP cells and showing the loss of an apoptotic response to 1,25(OH)2D3 in LNCaP cells that overexpressed Bcl-2. A more recent study by the same group [149] shows that p53 is not absolutely required for the induction of apoptosis in LNCaP cells by 1,25(OH)2D3 and that the induction of apoptosis appears to be caspase-dependent. 1,25(OH)2D3 down-regulates Bcl-2 expression even after the elimination of p53 function in LNCaP cells [149]. In LNCaP cells, therefore, 1,25(OH)2D3 stimulates not only growth arrest but also apoptosis, although to a much lesser extent. Recent investigations report inhibition of anti-apoptotic proteins by vitamin D in other PCa cells. Guzey et al. [155] showed that in LNCaP and ALVA-31 cells, 1,25(OH)2D3 decreased the expression of several anti-apoptotic proteins such as Bcl-2, Bcl-XL, and Mcl-1 leading to the activation of the mitochondrial pathway of apoptosis. The vitamin D analog V (BXL-353) has been shown to decrease Bcl-2 expression in DU 145 cells [156]. Induction of apoptosis by 1,25(OH)2D3, however, appears to be cell-specific as it is not uniformly evident in all the cells that respond to 1,25(OH)2D3 with growth inhibition. Even in LNCaP cells, where apoptosis has been demonstrated by some studies, the major action of 1,25(OH)2D3 to inhibit cell growth appears to be cell cycle arrest [154].

C. Differentiation 1,25(OH)2D3 has been shown to induce the differentiation of a number of normal and malignant cells (see Chapter 92). However, as yet there is no strong evidence supporting a role for vitamin D as a differentiation-promoting agent in PCa. Peehl et al. [73] did not find any changes in cell morphology or in the expression of various keratins as markers of epithelial cell differentiation when primary human prostatic epithelial cells were exposed to 1,25(OH)2D3. Konety et al. [157] harvested prostate tissue from castrated rats

1692

ARUNA V. KRISHNAN, DONNA M. PEEHL, AND DAVID FELDMAN

epithelial cells [73]. Expression of autocrine growth factors by the epithelium may contribute to the progression of PCa through the development of independence from epithelial-stromal interactions that modulate the growth and development of the normal prostate gland. Some important growth factors that regulate prostate epithelial growth include epidermal growth factor, keratinocyte growth factor, basic fibroblast growth factor, transforming growth factors (TGFs), and insulin-like growth factors (IGFs). Some of these growth factors also appear to play an important role in the establishment and growth of metastatic PCa cells in bone [159,160]. The following is a discussion of the role played by members of two prominent growth factor families, namely IGFs and TGFs, in the actions of 1,25(OH)2D3 in prostate cells. In PC-3 and ALVA 31 cells, 1,25(OH)2D3 decreases the availability of IGF by increasing the expression of its binding proteins IGFBP-3 and IGFBP-5 [161,162]. Boyle et al. [147] provided evidence that the upregulation of IGFBP-3 expression by 1,25(OH)2D3 is a necessary component of 1,25(OH)2D3–mediated inhibition of LNCaP cell growth. 1,25(OH)2D3 treatment increases IGFBP-3 mRNA levels. Importantly, addition of IGFBP-3 anti-sense oligonucleotides abrogates 1,25(OH)2D3–mediated growth inhibition (Fig. 6). Immunoneutralization of the IGFBP-3 protein similarly

treated with vehicle, testosterone (T), 1,25(OH)2D3, or a combination of T and 1,25(OH)2D3. Histological examination of the prostate tissue revealed a greater degree of epithelial cellular differentiation in rats treated with T and 1,25(OH)2D3 compared to rats treated with T alone. In the PCa cells LNCaP and MDA PCa 2a and 2b, 1,25(OH)2D3 increases the expression of PSA [83,110], which is regarded as a differentiation marker for prostatic epithelial cells. However, the effect appears to be much smaller compared to the induction of PSA expression by androgens, and the synergistic increase in PSA in cells treated with both 1,25(OH)2D3 and androgens is in part due to the up-regulation of AR levels by 1,25(OH)2D3 [83].

D. Growth Factor Actions Growth factors play an important role in the regulation of prostate epithelial cell growth by autocrine and paracrine mechanisms. Prostatic stromal cells are capable of modifying the epithelial cell environment through paracrine production of peptide growth factors that can act on the basal epithelium [9,158]. The stromal cells are androgen-responsive [158], and the expression of VDR has also been demonstrated in the stromal fibroblasts, although at levels lower than the

Antisense 114

Control

140

Sense

DNA content (% of control)

120 100

100

100

100

80 54 60

41

40 20

0 1,25-(OH)2D3 (10 nM)

−

+

−

+

−

+

FIGURE 6 Abrogation of 1,25(OH)2D3 mediated growth inhibition of LNCaP cells by IGFBP-3 anti-sense oligonucleotides. Cells were seeded in 96-well plates and grown in serum-free growth medium for 4 days with 10 nM 1,25(OH)2D3 (+) or ethanol vehicle (−), along with 8 µg/ml of anti-sense or sense IGFBP-3 oligonucleotides. No oligonucleotides were added to the control group. DNA concentrations were determined at the end of the experiment and values in cells treated with vehicle for each group were defined as 100%. Values are shown as mean ± SE of three experiments. From Boyle et al. [147] with permission.

1693

CHAPTER 94 Vitamin D and Prostate Cancer

abrogates growth inhibition by 1,25(OH)2D3. The data suggest that in LNCaP cells the growth inhibitory action of 1,25(OH)2D3 depends on IGFBP-3 up-regulation. This study also demonstrates that the increase in the levels of the CDK inhibitor protein p21 elicited by 1,25(OH)2D3 could be blocked by anti-IGFBP-3 antibodies showing that IGFBP-3 induction is necessary for the up-regulation of p21 expression by 1,25(OH)2D3. IGFBP-3 expression in benign human prostatic epithelial cells has also been shown to be increased by 1,25(OH)2D3. Nickerson and Huynh [163] demonstrated prostate regression in rats following the administration of the vitamin D analog EB 1089, which was associated with increases in the expression of several IGFBPs including IGFBP-3. The authors concluded that the prostate regression was related to alterations in the availability of IGFs as a result of increased production of IGFBPs. In recent studies Peng et al. characterized a VDRE in the promoter region on the IGFBP-3 gene [163a]. TGFβ and IGFBP-3 are pleiotropic factors that play an important role in the regulation of growth and differentiation in many cells [164–166]. They inhibit proliferation and induce apoptosis in prostate epithelial cells [164,166]. In PC-3 human PCa cells TGFβ has been shown to increase the expression of IGFBP-3 leading to growth arrest and apoptosis [165,166]. In NRP-152 cells, a nontumorigenic epithelial line derived from rat dorsolateral prostate, 1,25(OH)2D3 was shown to induce the expression of TGFβ2 and TGFβ3. In these cells TGFβ appears to mediate growth inhibition and certain biological responses due to both retinoic acid and 1,25(OH)2D3 [167]. The presence of VDRE sequences was demonstrated in the promoter of the human TGFβ2 gene [168].

E. Inhibition of Invasion and Metastasis In addition to the inhibition of proliferation in malignant cells, 1,25(OH)2D3 is also believed to play a role in tumor invasion and metastasis. 1,25(OH)2D3 can inhibit the invasiveness of breast and lung carcinoma cells in vitro [169,170]. 1,25(OH)2D3 decreases the tumor size and lung metastasis of the highly metastatic MAT LyLu and R3327-AT-2 Dunning PCa cells in vivo [115,126]. Schwartz et al. [171] showed that 1,25(OH)2D3 and the analog 1,25-dihydroxy-16ene-23-yne-cholecalciferol markedly inhibited the invasiveness of DU 145 human PCa cells through Amgel and also caused selective decreases in the secreted levels of matrix metalloproteinases (MMP)-2 and -9. A study by Sung and Feldman [172] showed that in DU 145 and PC-3 PCa cells, 1,25(OH)2D3

inhibited invasiveness, cell adhesion, and migration to the basement membrane matrix protein laminin due in part to decreasing the expression of α6 and β4 integrins. In LNCaP and PC-3 cells, 1,25(OH)2D3 and its analogs have also been shown to increase the expression of E-cadherin, a tumor suppressor gene whose expression is inversely correlated to the metastatic potential of the cells [131].

F. Angiogenesis Angiogenesis or the process of new blood vessel formation is critical for tumor progression and metastasis. 1,25(OH)2D3 has been shown to inhibit tumor cellinduced angiogenesis in mice [173,174]. Bernardi et al. [175] examined the ability of 1,25(OH)2D3 to modulate angiogenic signaling in tumor-derived endothelial cells and observed a reduction in the angiogenic signaling molecule, angiopoietin-2. These studies support the hypothesis that angiogenesis inhibition plays a role in the antitumor effects of 1,25(OH)2D3 [176].

G. Novel 1,25(OH)2D3 Target Genes 1,25(OH)2D3, acting through the VDR, initiates its effects on cell growth and differentiation by the direct activation or repression of target genes. The identification of target genes is one of the goals of current research on vitamin D. Freedman and co-workers used a differential screening technique to isolate putative 1,25(OH)2D3–inducible target genes during myeloid cell differentiation and identified genes encoding the CDK inhibitor p21 and the homeobox protein HoxA10 as vitamin D targets [143]. Feldman, Peehl, and colleagues recently performed cDNA microarray analyses in primary human prostatic epithelial cells derived from normal prostate or an adenocarcinoma of Gleason grade 3/3 as well as LNCaP cells to identify molecular targets of 1,25(OH)2D3 involved in the regulation of prostate epithelial cell growth [70,113]. Table I lists some of the putative responsive genes in LNCaP cells as well as primary prostatic epithelial cells derived from normal prostate or PCa. Several interesting and noteworthy observations have emerged from these studies. 24-hydroxylase, the classical 1,25(OH)2D3–inducible target gene, is maximally up-regulated in both normal and cancer-derived primary cell cultures. In contrast, in LNCaP cells the basal and induced levels of 24-hydroxylase are extremely low as has been shown earlier by Northern blot analysis [72]. As discussed earlier in Section III.B.1, lower levels of

1694 24-hydroxylase in LNCaP cells makes them sensitive to the growth inhibitory effects of 1,25(OH)2D3. In LNCaP cells, the expression of the IGFBP-3 gene shows the highest fold-increase after 1,25(OH)2D3 treatment, which is in agreement with the previous study by these investigators showing that the up-regulation of IGFBP-3 expression is essential for 1,25(OH)2D3–mediated growth inhibition in these cells [147]. No regulation of IGFBP-3 could be detected

ARUNA V. KRISHNAN, DONNA M. PEEHL, AND DAVID FELDMAN

in primary cells consistent with the fact that primary cells do not express IGFBP-3. Several of the genes regulated by 1,25(OH)2D3 in LNCaP cells, such as NDRG1 and 15-hydroxyprostaglandin dehydrogenase (Table I), have also been demonstrated to be androgenresponsive [177,178]. These findings are in accord with earlier data showing that in LNCaP cells the growth inhibitory action of 1,25(OH)2D3 is androgendependent [80].

TABLE I Summary of Results from cDNA Microarray Analysis of 1,25(OH)2D3 (10 nM) Treatment of LNCaP Cells, and Primary Cultures of Normal and Prostate Cancer Cells Gene LNCaP cells Insulin-like growth factor binding protein-3 N-myc downstream regulated (NDRG1) Liprin β 2 Hydroxy prostaglandin dehydrogenase 15-(NAD) Prostate differentiation factor ATP-binding cassette, subfamily A, member 5 Claudin 4

Fold increase at 6 h

Fold increase at 24 h

2.4 1.1 1.9 0.95 2.6 1.0 2.0

33.2 3.3 2.7 2.7 2.6 2.4 2.2

Normal prostate epithelial cell strain Cytochrome P450, subfamily XXIV (vitamin D 24-hydroxylase) Metallothionein 1H Metallothionein 1L Metallothionein 1G Metallothionein 1E Bone morphogenetic protein 6 Dual specificity phosphatase 10 Apoptosis inhibitor 4 (survivin) Thioredoxin reductase 1 A kinase anchor protein (gravin) 12 Purinergic receptor P2Y, G-protein coupled 2 Carbonic anhydrase II Hydroxy prostaglandin dehydrogenase 15-(NAD)

79 12.7 12 9.2 6.2 3.9 3.1 3.1 3.1 2.8 2.6 2.4 1.2

83 5.3 2.8 6.7 2.4 3.0 2.0 2.9 1.8 1.6 2.3 3.2 2.3

Prostate cancer epithelial cell strain Cytochrome P450, subfamily XXIV (vitamin D 24-hydroxylase) Dual specificity phosphatase 10 Purinergic receptor P2Y, G-protein coupled 2 Bone morphogenetic protein 6 Thioredoxin reductase 1 A kinase anchor protein (gravin) 12 Carbonic anhydrase II Tumor necrosis factor (ligand) superfamily, member 11 MAD (mothers against decapentaplegic, drosophila) homolog 6 Superoxide dismutase 2, mitochondrial

78 3.4 2.9 2.8 2.8 2.5 2.1 1.0 1.0 1.0

46 0.51 0.9 3.2 1.6 −1.4 4.1 4.3 2.1 2.0

CHAPTER 94 Vitamin D and Prostate Cancer

These cDNA microarray analyses have revealed several novel putative vitamin D target genes in primary epithelial cells (Table I). In general, there is an appreciable overlap in the profiles of 1,25(OH)2D3–regulated genes in normal and cancer-derived primary cells. In both of these cell types, the expression of dual specificity phosphatase 10 shows maximal up-regulation. Dual specificity phosphatase 10 inactivates mitogen activated protein kinase (MAPK), suggesting that an important feature of the growth inhibitory activity of 1,25(OH)2D3 in these cells may be an inhibition of the growth-promoting effect of MAPK. Early up-regulation of the kinase anchoring protein gravin that is known to coordinate the localization of protein kinase C (PKC) and protein kinase A (PKA) is of interest and may be related to the recently reported effect of vitamin D on the packaging of PKC in chondrocytes [179]. These data are also supportive of the role of vitamin D as an antioxidant in primary prostate cells (see Chapter 45). Thioredoxin reductase 1, involved in redox balance, is an early response gene in both normal and cancer cells. Up-regulation of superoxide dismutatse 2 is also indicative of protection from oxidative damage. The regulation of the expression of metallothionein genes by 1,25(OH)2D3 is different between the normal and cancer-derived primary cells, the former showing an up-regulation and the latter a significant down-regulation. Metallothioneins constitute the majority of intracellular protein thiols and as such are considered to act as cell survival factors. Up-regulation of metallothioneins in normal prostatic epithelial cells is consistent with the anti-apoptotic effect of 1,25(OH)2D3 in these cells. It is also noteworthy that 1,25(OH)2D3 up-regulates the expression of the anti-apoptotic protein survivin in these cells. Certain metallothioneins have been reported to be over-expressed in PCa [180] and hence a downregulation of their expression in the cancer-derived cells may be therapeutically beneficial. In summary, cDNA microarray analysis is a powerful tool that has revealed biologically important targets of 1,25(OH)2D3 in prostate cells and has provided a starting point for additional investigations into the molecular mechanisms underlying the anti-cancer effect of 1,25(OH)2D3 and its analogs.

VI. VITAMIN D IN COMBINATION WITH OTHER AGENTS The efficacy of 1,25(OH)2D3 in PCa therapy is directly related to the dose of 1,25(OH)2D3 administered. Hypercalcemia becomes more frequent at higher concentrations of 1,25(OH)2D3, which limits the maximum dose that can be given safely. Several vitamin D

1695 analogs (Section VIII of this book) exhibit increased potency as antiproliferative agents and have less hypercalcemic effects and therefore form a class of agents with a higher therapeutic index than the parent compound, 1,25(OH)2D3. Another avenue to increase efficacy and decrease toxicity is to use a combination of agents that act by different mechanisms, at doses that are less than required when the agents are administered individually. This drug combination strategy has the advantage of limiting the toxicity associated with the individual drugs while obtaining additive and potentially synergistic therapeutic effects. Investigators have been studying the antiproliferative effect of vitamin D in prostate cancer in combination with other agents as discussed below.

A. Vitamin D and 24-hydroxylase Inhibitors As detailed in Section III.B, a combination of 1,25(OH)2D3 with inhibitors of 24-hydroxylase such as liarozole [97] (Fig. 3) or ketoconazole [101] increases the half-life of 1,25(OH)2D3 in PCa cells and also leads to VDR up-regulation. The combination of 1,25(OH)2D3 with 24-hydroxylase inhibitors therefore would allow the use of lower doses of 1,25(OH)2D3 in vivo possibly reducing the hypercalcemic side effects. Ketoconazole, an imidazole derivative that inhibits mammalian P 450 enzymes, has been shown to exhibit growth inhibitory activity by itself in PCa cells [181], as well as in primary prostatic epithelial cells [101]. Ketoconazole has been shown to block testicular synthesis of androgens, and therefore has been frequently used to ablate androgen biosynthesis in PCa patients [182]. However, ketoconazole, in addition to blocking 24-hydroxylase [183], also blocks the activity of 1α-hydroxylase (Chapter 5) leading to low levels of 1,25(OH)2D3, a major risk factor for osteoporosis and osteomalacia [69]. Indeed, administration of ketoconazole to normal men has been shown to result in a dose-dependent decrease in serum levels of 1,25(OH)2D3 [184]. Men on this ketoconazole-based androgen ablation therapy would therefore be at a great risk for metabolic bone disease and other side effects of vitamin D deficiency. Adding 1,25(OH)2D3 or analogs to therapeutic regimens that include ketoconazole would be beneficial as it would restore circulating levels of active vitamin D [101]. However, the clinical use of ketoconazole and 1,25(OH)2D3 combination needs to be approached with caution. Ketoconazole inhibits multiple P 450 enzymes including those in the steroidogenic pathways leading to the synthesis of testosterone, cortisol, and aldosterone [182]. Adrenal insufficiency would exacerbate hypercalcemia. Therefore, patients

1696 undergoing this combination therapy require careful monitoring for hypercalcemia and adrenal insufficiency in addition to the assessment of prostate cancer progression.

B. Vitamin D and Retinoids The retinoids are comprised of vitamin A and its derivatives, including all-trans retinoic acid (ATRA), 9-cis retinoic acid (9-cis RA), and 13-cis retinoic acid (13-cis RA). These compounds are ligands for the retinoid receptors (RARs and RXRs) that act as nuclear transcription factors. RXRs are transcription factors that heterodimerize with RAR, VDR, and a variety of other nuclear transcription factors (see Chapters 11 and 13). In turn, these ligand-activated complexes regulate a variety of cellular processes including growth and differentiation. Because of their role in controlling growth and differentiation, retinoids have been examined for potential anti-cancer activity in epithelial and nonepithelial malignancies, and may be particularly active in certain hematologic malignancies [185]. Prostate tissue expresses RARs and RXRs. Interestingly, cancer tissue appears to have lower concentrations of endogenous ATRA than normal prostate tissue, suggesting a possible rationale for a chemopreventive role for retinoids in prostate cancer [186]. Retinoids have also shown efficacy in controlling prostate cancer growth in animal models [187–189]. Several in vitro studies demonstrate synergistic activity between vitamin D and retinoids in prostate cells. ATRA enhanced the growth inhibitory effects of 1,25(OH)2D3 on primary cultures of normal and malignant prostatic epithelial cells [190]. Although Esquenet et al. [81] did not find any cooperativity between ATRA or 9-cis RA and 1,25(OH)2D3 on LNCaP cell growth in the presence of androgens, Blutt et al. [144] reported that 1,25(OH)2D3 and 9-cis RA inhibited the growth of LNCaP cells in a synergistic manner and caused the accumulation of these cells in the G1 phase of the cell cycle. Zhao et al. [110] also showed synergistic inhibitory activity between 1,25(OH)2D3 and 9-cis RA on the growth of LNCaP cells. In LNCaP cells, 1,25(OH)2D3 and 9-cis RA increased the expression of AR mRNA, and the effect of both the hormones on AR expression was additive [110]. Various other retinoids have been shown to sensitize prostate cancer cells to 1,25(OH)2D3 or its analogs [191–193]. These studies provide a rationale for testing the use of combinations of vitamin D compounds with retinoic acid derivatives in PCa therapy.

ARUNA V. KRISHNAN, DONNA M. PEEHL, AND DAVID FELDMAN

C. Vitamin D and Glucocorticoids Glucocorticoids have been widely used in the treatment of advanced PCa and have been part of PCa therapy in several randomized trials (for a review see [194]). Although glucocorticoids do not induce apoptosis in PCa cells, they exhibit growth inhibitory effects [195,196]. Glucocorticoids may exert anti-tumor effects in vivo on PCa by producing a negative feedback on the pituitary gland, leading to a decrease in both testicular and adrenal androgens [194]. They appear to inhibit prostate cell growth in vitro by modulating growth factors such as lipocortin, TGFβ-1, urokinase-type plasminogen activator, and interleukin-6 [194]. However, in PCa cells carrying promiscuous AR mutations (see Section I.B), some glucocorticoids can inappropriately act through the AR and stimulate cell growth [18,19,197]. Combination therapy with glucocorticoids and vitamin D compounds in general enhances the anti-tumor activity of vitamin D. However, Peehl et al. [190] made the observation that hydrocortisone at µM concentrations significantly reduced the antiproliferative effect of 1,25(OH)2D3 in primary prostatic epithelial cells. In a murine squamous cell carcinoma model system, dexamethasone has been shown to reduce 1,25(OH)2D3– mediated hypercalcemia and enhance the anti-tumor activity of 1,25(OH)2D3 in vitro and in vivo, possibly due to the up-regulation of VDR levels [198]. Further investigations using this model system suggest the involvement of Erk and Akt signaling pathways in the antiproliferative effects of 1,25(OH)2D3 and dexamethasone combinations [199]. Ahmed et al. [200] have shown that 1,25(OH)2D3 significantly increases mitoxantrone and dexamethasone mediated growth inhibition in PC-3 PCa cells in vitro, and the combination of 1,25(OH)2D3 plus mitoxantrone and dexamethasone causes significantly greater tumor regression in PC-3 xenograft-bearing mice. These studies have prompted Trump and colleagues to test the combination of 1,25(OH)2D3 and dexamethasone in clinical trials (see also Chapter 97).

D. Vitamin D and Peroxisome Proliferator-activated Receptor (PPAR) Ligands The PPARs, like VDR, RXR, and RARs, are members of the steroid receptor superfamily of proteins. Ligands for PPARγ exhibit anticancer activity against a wide variety of neoplastic cells including PCa cells, and in vivo studies in animal models also demonstrate the anti-cancer and chemopreventive capabilities of

CHAPTER 94 Vitamin D and Prostate Cancer

these ligands [201]. PPARγ ligands have been shown to inhibit the growth of PC-3 human PCa cells [202,203], producing morphological changes consistent with a less malignant phenotype [202]. The PPARγ ligand troglitazone also inhibited the growth of PC-3 xenograft tumors in immunodeficient mice [202]. The PPARγ ligand rosiglitazone has been shown to inhibit the growth of primary prostatic epithelial cells and increase the expression of genes encoding differentiation-associated secretory proteins such as adipophilin in these cells [204]. Hisatake et al. [205] showed that PPARγ ligands inhibited the activation of the PSA gene by androgens in LNCaP cells and thereby suppressed PSA production by these cells. They also demonstrated that oral administration of troglitazone reduced the increase in velocity of PSA levels in a patient with occult recurrent PCa. Mueller et al. [206] conducted a phase II clinical study in patients with advanced PCa and found an unexpectedly high incidence of prolonged stabilization of PSA in patients treated orally with troglitazone. Although combinations of vitamin D compounds and PPARγ ligands have not been tested in cell culture, animal models, or clinical trials, the existing data provide a rationale for studying the effect of this combination in PCa therapy and chemoprevention of PCa.

E. Vitamin D and Soy Isoflavones Soy is a rich source of isoflavones such as genistein and daidzein. An inverse correlation between consumption of soy in the diet and PCa incidence has been shown in epidemiological studies [207]. Soy-derived phytoestrogens have been demonstrated to inhibit the growth of PCa cells in culture [27,208]. Genistein decreases the expression of AR mRNA and protein and inhibits the transcriptional activation of the PSA gene in LNCaP cells [209]. Genistein also decreases AR expression in the dorsolateral prostate of rats [210]. In vivo studies in animal models have demonstrated an inhibitory effect of dietary genistein on the development of PCa [211,212] and a decrease in plasma testosterone levels in genistein-fed rats [213]. Several studies have shown interactions between phytoestrogens and the vitamin D axis in prostate cells. 1,25(OH)2D3 and genistein have been shown to synergistically inhibit the growth of human primary prostatic epithelial cells and LNCaP cells [214]. The combination of these two agents causes arrest of the primary prostate cells in the G0/G1 phase as well as the G2M phase of the cell cycle [214]. Farhan et al. [215] have shown that genistein and other isoflavanoids modulate the availability of 1,25(OH)2D3 in DU 145

1697 cells by inhibiting the activity of 24-hydroxylase and 1α-hydroxylase. The inhibition of 24-hydroxylase appears to be at the transcriptional level while that of 1α-hydroxylase involves deacetylation. A recent study by Weitzke and Welsh [216] reported an up-regulation of the transcription of the VDR promoter as measured by reporter gene activity and also demonstrated an increase in VDR protein expression in breast cancer cells following phytoestrogen treatment. Thus, a combination of vitamin D compounds and phytoestrogens such as genistein may be beneficial in PCa treatment as genistein can potentiate the growth inhibitory activity of vitamin D by the various mechanisms mentioned above.

F. Vitamin D and Chemotherapeutic Drugs Several studies in cell cultures and animal models as well as clinical trials (see Section VII) have demonstrated the potential utility of vitamin D and its analogs as agents that can enhance the antiproliferative and cytotoxic effects of conventional chemotherapeutic drugs. Combined administration of 1,25(OH)2D3 or its analog with platinum compounds such as carboplatin or cisplatin has been shown to result in a marked enhancement of growth inhibition in breast cancer [217] and prostate cancer cells [218] over that seen with the platinum compound alone. Wang et al. [219] showed that pretreatment of breast cancer cells with 1,25(OH)2D3 or all-trans retinoic acid (ATRA) significantly lowers the threshold for cell killing by the chemotherapy drugs paclitaxel and adriamycin. The vitamin D analog EB1089, in combination with paclitaxel, effectively inhibits the growth of MC-7 breast cancer cell growth in vivo [220]. A recent study by Hershberger et al. [221] showed that treatment of murine squamous carcinoma cells and PC-3 PCa cells in vitro with 1,25(OH)2D3 prior to paclitaxel caused significantly greater growth inhibition than either agent alone. In PC-3 tumor-bearing mice, 1,25(OH)2D3 pretreatment similarly enhanced the tumor inhibitory effect of paclitaxel [221]. The molecular basis for the enhanced anti-tumor activity of this combination appears to be the increase in the expression of the cell cycle inhibitor p21 by 1,25(OH)2D3, rendering the cells more susceptible to paclitaxel-induced apoptosis [221]. Paclitaxel cytotoxicity has been similarly shown to be increased in breast and colon cancer cells when p21 expression is specifically perturbed [222,223]. The ability of vitamin D to reduce cell survival signals may explain why it can enhance the anti-tumor activity of mechanistically diverse cytotoxic agents such as platinum compounds and taxol derivatives. These data

1698

ARUNA V. KRISHNAN, DONNA M. PEEHL, AND DAVID FELDMAN

clearly support the combined use of vitamin D compounds and cytotoxic drugs in the treatment of solid tumors such as PCa. A recent in vitro study demonstrates synergistic inhibition of the growth of PCa cells by vitamin D or its analogs in combination with histone deacetylase inhibitors such as sodium butyrate and trichostatin A [224]. Both in vitro and in vivo investigations thus far support the use of vitamin D or analogs in combination with other growth inhibitory agents in clinical trials of PCa prevention and/or treatment.

VII. CLINICAL TRIALS Investigators have undertaken clinical trials in cancer patients to evaluate the safety and efficacy of treatment with vitamin D or its analogs (see Chapter 97 for details). Based on the extensive data developed from cell culture and animal models, there have been several strategies attempted. First, because it is an approved drug, 1,25(OH)2D3 (calcitriol) has been used and administered in as high a dose as tolerated, limited by hypercalciuria or hypercalcemia. Second, trials are attempting to use less hypercalcemic analogs with a wider therapeutic window than calcitriol. Third, calcitriol has been administered intermittently in very high doses where it apparently can still cause its antiproliferative effects and only transient hypercalcemia that, thus far, does not appear to cause toxicity. Fourth, calcitriol or analogs are being used in combination therapy with agents that may enhance the antiproliferative activity while reducing its hypercalcemic tendency.

TABLE II Patient no. 1 2 3 4 5 6 7

Osborn et al. [225] reported a small phase II trial of calcitriol in 13 patients with hormone-refractory metastatic PCa. No objective responses (> 50% reduction in serum PSA levels or >30% reduction in measurable tumor mass) could be seen, and the median time to progression was 10.6 weeks. A pilot study by Gross et al. [226] used increasing concentrations of calcitriol to treat seven patients with early recurrent PCa following radiation or surgery. At the beginning of the trial, the patients showed no evidence of metastasis, and the only sign of recurrent disease was rising levels of serum PSA. Calcitriol doses were limited to 2 µg/day due to hypercalciuria in all patients. The doubling time of serum PSA before and after calcitriol treatment was compared in each patient. As shown in Table II, in all seven patients the rate of PSA rise was substantially decreased by calcitriol, and in the case of one patient, the serum PSA levels substantially decreased, registering a negative doubling time. Due to the formation of small asymptomatic renal stones, calcitriol therapy was discontinued in two of the seven patients. Withdrawal from therapy resulted in the resumption of the rise in PSA with the doubling time returning to the values seen before therapy was initiated (Table II). This study provides evidence that vitamin D could be effective in slowing the progression of PCa, based on PSA values. However, when therapy was discontinued, the resumption of growth at pretherapy levels suggests that calcitriol was only cytostatic and did not kill the cancer cells. These early trials underscore the fact that development of hypercalciuria or hypercalcemia would preclude the use of very high doses of calcitriol and therefore limit its therapeutic benefit that may only be realized at high doses.

PSA Doubling Time in Months Before and After 1,25(OH)2D3 Treatment of Men with Early Recurrent PCa Before therapy

During therapy

p Value

Therapy discontinued

17.1 5.5 12.0 10.8 23.2 12.5 4.1

−3.3 12.5 37.6 15.8 43.0 27.4 9.4

0.005 < 0.0001 0.009 0.002 0.029 0.02 0.0009

10.8 2.1 13.7

Patients completed 6 to 15 months of 1,25(OH)2D3 therapy starting with a dose of 0.5 µg/day administered orally. The dose was increased on 0.25 µg increments weekly to a maximum dose of 2.5 µg/day, depending on individual calcemic and calciuric responses. Serum samples were collected for PSA measurements weekly during the dose escalation phase and then monthly. PSA doubling time was calculated before, during, and after therapy. Adapted from Gross et al. [226] with permission.

1699

CHAPTER 94 Vitamin D and Prostate Cancer

Smith et al. [227] conducted a phase I trial in patients with advanced solid tumors to determine the maximum tolerated dose of calcitriol administered subcutaneously every other day (QOD) and found hypercalcemia occurring at the 10 µg QOD dose. A recent study by Beer et al. [228] tested the effect of calcitriol at a very high oral dose once weekly in patients with a rising PSA after prostatectomy or radiation therapy. This study found the weekly oral administration of 0.5 µg/kg calcitriol to patients on a low calcium diet to be safe and the median PSA doubling time increased from 7.8 months to 10.3 months. This dosage, approximately 70-fold the physiologic replacement dose of calcitriol, suggests that very large bolus intermittent therapy may be a means to enhance the anti-cancer actions of calcitriol while avoiding its side effects. The hypercalcemia returned to normal within 24–36 hours, and thus far renal stones or impairment of kidney function have not been noted. As discussed in Section IV, several structural analogs of calcitriol that are more potent as antiproliferative agents but exhibit less hypercalcemic effects are being developed and characterized. A phase I study of the analog 1α-hydroxyvitamin D2 (paracalcitol) administered orally to patients with advanced hormone refractory PCa recommended a dose of 12.5 µg/day given continuously for use in phase II trials [229]. Investigators have also been studying the efficacy of calcitriol in treatment of men with PCa in combination with other agents as discussed in detail in Section VI. Johnson et al. [230] have initiated phase I and phase II trials of calcitriol, either alone or in combination with other agents. Preclinical data indicate that glucocorticoids potentiate the anti-tumor effects of calcitriol and alleviate some of the hypercalcemic side effects (see Section VI.C). One of the trials initiated by Johnson et al. [230] is a phase II study of calcitriol in combination with dexamethasone in 43 patients with AIPC. Altogether, 80% of the patients have exhibited a slowing in the rate of increase in serum PSA, and 34% have had stable disease or decrease in PSA (> 50% reduction) levels. Patients with bone pain at study entry have had pain relief. As discussed in Section VI.F, cell culture and animal studies indicate that calcitriol and its analogs can enhance the cytotoxic effects of conventional chemotherapeutic drugs, and investigators have begun testing these combinations in phase I and phase II trials. Trump, Johnson, and co-workers are conducting two phase I trials testing 1,25(OH)2D3 in combination with carboplatin or paclitaxel. In the study testing calcitriol and carboplatin, no dose-limiting toxicity has been seen up to calcitriol doses of 13 µg/day orally for three days every four weeks with carboplatin, and the study

continues [230]. In the second trial, paclitaxel is given with escalating doses of calcitriol every day for three days per week for a period of six weeks, and so far no dose limiting toxicity has been encountered [231]. Beer et al. [232] are conducting a phase II study to evaluate weekly oral administration of very high dose calcitriol (0.5 µg/kg) on day 1 followed by intravenous administration of docetaxel on day 2 in patients with AIPC, repeated for six consecutive weeks of each eightweek cycle. In a recent report Beer et al. [233] showed that the combination of a once weekly oral high-dose calcitriol and weekly docetaxel was a well-tolerated regimen for patients with AIPC. Thus far 22 out of 37 patients exhibit a > 75% reduction in serum PSA at the end of an eight-week treatment cycle, confirmed four weeks later. The results of this study indicate that PSA and measurable disease response rates, as well as time to progression and survival, are more promising with the combination approach when compared to phase II studies of the single agent docetaxel in AIPC. To confirm these encouraging findings these investigators have started a placebo-controlled, double-blinded randomized comparison of docetaxel plus calcitriol to docetaxel alone. The result of the various clinical investigations so far support the potential promise of exploiting vitamin D compounds alone or in combination with other agents to control PCa progression and further studies are clearly warranted.

VIII. SUMMARY AND CONCLUSIONS A number of studies have established the role of 1,25(OH)2D3 as an antiproliferative agent in normal and malignant prostate cells. Investigations using animal models have also demonstrated the anti-tumor effects of 1,25(OH)2D3 and its less hypercalcemic analogs. The mechanisms underlying the anti-cancer effects of vitamin D compounds in PCa cells are varied and cell-specific and include growth arrest, apoptosis, pro-differentiation effects, and modification of growth factor activity. Some studies suggest additional effects to inhibit invasiveness and anti-angiogenesis, but there are less data on these issues. 1,25(OH)2D3 interacts with androgen signaling in PCa cells possibly enhancing its prodifferentiating activity. Investigators are making progress in identifying 1,25(OH)2D3-regulated genes and understanding their role in the mediation of the above-mentioned effects. Such studies would unveil novel 1,25(OH)2D3–responsive genes and provide new therapeutic targets. Administration of pharmacological doses of calcitriol to men with PCa results in hypercalcemia and hypercalciuria, which limits the concentration of

1700 calcitriol that can safely be administered to patients. Consequently, structural analogs of calcitriol that effectively activate the VDR but exhibit less hypercalcemic effects are being developed and evaluated as potential anti-cancer agents. A challenging area of research involves defining the mechanisms by which various vitamin D analogs maintain potent growth inhibitory effects and yet are less able to induce hypercalcemia. Intermittent dosing regimens with very high boluses of calcitriol are an interesting approach that apparently inhibits PCa while not causing persistent hypercalcemia. Combinations of calcitriol or its analogs with inhibitors of 24-hydroxylase or other growth-inhibitory molecules such as retinoids, glucocorticoids, or cytotoxic chemotherapeutic drugs permit the use of vitamin D and the other agents at relatively lower doses, thereby avoiding toxic side effects while achieving synergistic antitumor effects. Several investigators are carrying out clinical trials to determine whether calcitriol or analogs singly or in combination with other agents can prevent the progression of PCa. Investigators are also examining the role of vitamin D compounds as chemopreventive agents. We believe that progress in these areas of research will result in rational drug design and lead to the development of more potent and safer vitamin D analogs. It remains to be determined what will be the most effective use of calcitriol and its analogs in terms of dosage regimen, combination with other agents, and phase of the disease in which it will be most useful in the treatment and/or prevention of PCa.

References 1. Jemal A, Murray T, Samuels A, Ghafoor A, Ward E, Thun MJ 2003 Cancer statistics. CA Cancer J Clin 53:5–26. 2. Stamey TA, Freiha FS, McNeal JE, Redwine EA, Whittemore AS, Schmid H-P 1993 Localized prostate cancer. Cancer 71(Suppl):933–938. 3. Hellerstedt BA, Pienta KJ 2002 The current state of hormonal therapy for prostate cancer. CA Cancer J Clin 52:154–179. 4. Nelson WG, De Marzo AM, Isaacs WB 2003 Prostate Cancer. N Engl J Med 349:366–381. 5. Taplin ME, Ho SM 2001 Clinical review 134: The endocrinology of prostate cancer. J Clin Endocrinol Metab 86:3467–3477. 6. Carter H, Coffey D 1988 Prostate cancer: The magnitude of the problem in the United States. In: DS Coffey, MI Resnick, FA Dorr, JP Karr, et al. (eds) A Multidisciplinary Analysis of Controversies in the Management of Prostate Cancer. Plenum Press: New York, pp. 1–9. 7. Kozlowski JM, Grayhack JT 1995 Carcinoma of the prostate. In: JY Gillenwater, IT Grayhack, SS Howards, JW Duckett (eds) Adult and Pediatric Urology, Second Ed. Mosby Year Book: St Louis, pp. 1575–1713. 8. Thompson IM, Goodman PJ, Tangen CM, Lucia MS, Miller GJ, Ford LG, Lieber MM, Cespedes RD, Atkins JN, Lippman SM, Carlin SM, Ryan A, Szczepanek CM, Crowley JJ,

ARUNA V. KRISHNAN, DONNA M. PEEHL, AND DAVID FELDMAN

9. 10. 11. 12.

13.

14.

15. 16.

17.

18.

19. 20. 21. 22. 23.

24.

Coltan CA Jr. 2003 The influence of finasteride on the development of prostate cancer. N Engl J Med 349:215–224. Isaacs JT 1999 The biology of hormone refractory prostate cancer. Why does it develop? Urol Clin North Am 26:263–273. Feldman BJ, Feldman D 2001 The development of androgenindependent prostate cancer. Nature Rev Cancer 1:34–45. Culig Z, Klocker H, Bartsch G, Hobisch A 2001 Androgen receptor mutations in carcinoma of the prostate: significance for endocrine therapy. Am J Pharmacogenomics 1:241–249. Taplin ME, Bubley GJ, Ko YJ, Small EJ, Upton M, Rajeshkumar B, Balk SP 1999 Selection for androgen receptor mutations in prostate cancers treated with androgen antagonist. Cancer Res 59:2511–2515. Nazareth LV, Stenoien DL, Bingman WE 3rd, James AJ, Wu C, Zhang Y, Edwards DP, Mancini M, Marcelli M, Lamb DJ, Weigel NL 1999 A C619Y mutation in the human androgen receptor causes inactivation and mislocalization of the receptor with concomitant sequestration of SRC-1 (steroid receptor coactivator 1). Mol Endocrinol 13:2065–2075. Marcelli M, Ittmann M, Mariani S, Sutherland R, Nigam R, Murthy L, Zhao Y, DiConcini D, Puxeddu E, Esen A, Eastham J, Weigel NL, Lamb DJ 2000 Androgen receptor mutations in prostate cancer. Cancer Res 60:944–949. Buchanan G, Greenberg M, Scher HI, Harris JM, Marshall VR, Tilley WD 2001 Collocation of androgen receptor gene mutations in prostate cancer. Clin Cancer Res 7:1273–1281. James AJ, Agoulnik IU, Harris JM, Buchanan G, Tilley WD, Marcelli M, Lamb DJ, Weigel NL 2002 A novel androgen receptor mutant, A748T, exhibits hormone concentrationdependent defects in nuclear accumulation and activity despite normal hormone-binding affinity. Mol Endocrinol 16: 2692–2705. Veldscholte J, Berrevoets CA, Ris-Stalpers C, Kuiper GG, Jenster G, Trapman J, Brinkmann AO, Mulder E 1992 The androgen receptor in LNCaP cells contains a mutation in the ligand-binding domain which affects steroid binding characteristics and response to antiandrogens. J Steroid Biochem Mol Biol 41:665–669. Zhao XY, Malloy PJ, Krishnan AV, Swami S, Navone NM, Peehl DM, Feldman D 2000 Glucocorticoids can promote androgen-independent growth of prostate cancer cells through a mutated androgen receptor. Nat Med 6:703–706. Chang CY, Walther PJ, McDonnell DP 2001 Glucocorticoids manifest androgenic activity in a cell line derived from a metastatic prostate cancer. Cancer Res 61:8712–8717. Scher HI, Kelly WK 1993 Flutamide withdrawal syndrome: its impact on clinical trials in hormone-refractory prostate cancer. J Clin Oncol 11:1566–1572. Paul R, Breul J 2000 Antiandrogen withdrawal syndrome associated with prostate cancer therapies: incidence and clinical significance. Drug Saf 23:381–390. Wolf RM, Schneider SL, Pontes JE, Englander L, Karr JP, Murphy GP, Sandberg AA 1985 Estrogen and progestin receptors in human prostatic carcinoma. Cancer 55:2477–2481. Leav I, Lau KM, Adams JY, McNeal JE, Taplin ME, Wang J, Singh H, Ho SM 2001 Comparative studies of the estrogen receptors beta and alpha and the androgen receptor in normal human prostate glands, dysplasia, and in primary and metastatic carcinoma. Am J Pathol 159:79–92. Kim IY, Kim BC, Seong DH, Lee DK, Seo JM, Hong YJ, Kim HT, Morton RA, Kim SJ 2002 Raloxifene, a mixed estrogen agonist/antagonist, induces apoptosis in androgenindependent human prostate cancer cell lines. Cancer Res 62:5365–5369.

1701

CHAPTER 94 Vitamin D and Prostate Cancer

25. Castagnetta LA, Miceli MD, Sorci CMG, Pfeffer U, Farruggio R, Oliveri G, Calabro M, Carruba G 1995 Growth of LNCaP human prostate cancer cell is stimulated by estradiol via its own receptor. Endocrinology 136:2309–2319. 26. Lau KM, LaSpina M, Long J, Ho SM 2000 Expression of estrogen receptor (ER)-alpha and ER-beta in normal and malignant prostatic epithelial cells: regulation by methylation and involvement in growth regulation. Cancer Res 60:3175–3182. 27. Mitchell JH, Duthie SJ, Collins AR 2000 Effects of phytoestrogens on growth and DNA integrity in human prostate tumor cell lines: PC-3 and LNCaP. Nutr Cancer 38:223–228. 28. Gross C, Peehl DM, Feldman D 1997 Vitamin D and prostate cancer. In: D Feldman, FH Glorieux, JW Pike (eds) Vitamin D. Academic Press: San Diego, pp. 1125–1139. 29. Blutt SE, Weigel NL 1999 Vitamin D and prostate cancer. Proc Soc Exp Biol Med 221:89–98. 30. Konety BR, Johnson CS, Trump DL, Getzenberg RH 1999 Vitamin D in the prevention and treatment of prostate cancer. Semin Urol Oncol 17:77–84. 31. Miller G 1999 Vitamin D and prostate cancer: biological interactions and clinical potentials. Cancer Metastasis Rev 17:353–360. 32. Feldman D, Zhao XY, Krishnan AV 2000 Vitamin D and prostate cancer [editorial; comment]. Endocrinology 141:5–9. 33. Krishnan AV, Peehl DM, Feldman D, 2003 Inhibition of prostate cancer growth by vitamin D: Regulation of target gene expression. J Cell Biochem 88:363–371. 34. Peehl DM, Feldman D 2003 The role of vitamin D and retinoids in controlling prostate cancer progression. Endocr Relat Cancer 10:131–140. 35. Ruijter E, van de Kaa C, Miller G, Ruiter D, Debruyne F, Schalken J 1999 Molecular genetics and epidemiology of prostate carcinoma. Endocr Rev 20:22–45. 36. Apperly FL 1941 The relation of solar radiation to cancer mortality in North America. Cancer Research 1:191–195. 37. Grant WB 2002 An estimate of premature cancer mortality in the U.S. due to inadequate doses of solar ultraviolet-B radiation. Cancer 94:1867–1875. 38. Schwartz GG, Hulka BS 1990 Is vitamin D deficiency a risk factor for prostate cancer? (Hypothesis). Anticancer Res 10:1307–1311. 39. Hanchette CL, Schwartz GG 1992 Geographic patterns of prostate cancer mortality. Evidence for a protective effect of ultaraviolet radiation. Cancer 70:2861–2869. 40. Studzinski GP, Moore DC 1995 Sunlight—can it prevent as well as cause cancer? Cancer Res 55:4014–4022. 41. Bell NH, Greene A, Epstein S, Oexmann MJ, Shaw S, Shary J 1985 Evidence for alteration of the vitamin D–endocrine system in blacks. J Clin Invest 76:470–473. 42. Bodiwala D, Luscombe CJ, French ME, Liu S, Saxby MF, Jones PW, Ramachandran S, Fryer AA, Strange RC 2003 Susceptibility to prostate cancer: studies on interactions between UVR exposure and skin type. Carcinogenesis 24:711–717. 43. Kolonel LN, Hankin JH, Whittemore AS, Wu AH, Gallagher RP, Wilkens LR, John EM, Howe GR, Dreon DM, West DW, Paffenbarger RS Jr 2000 Vegetables, fruits, legumes, and prostate cancer: a multiethnic case-control study. Cancer Epidemiol Biomarkers Prev 9:795–804. 44. Giovannucci E, Rimm EB, Wolk A, Ascherio A, Stampfer MJ, Colditz GA, Willett WC 1998 Calcium and fructose intake in relation to risk of prostate cancer. Cancer Res 58:442–447. 45. Rodriguez C, McCullough ML, Mondul AM, Jacobs EJ, Fakhrabadi-Shokoohi D, Giovannucci EL, Thun MJ, Calle EE

46. 47.

48.

49.

50.

51.

52. 53.

54. 55.

56. 57.

58. 59. 60.

61. 62.

2003 Calcium, dairy products, and risk of prostate cancer in a prospective cohort of United States men. Cancer Epidemiol Biomarkers Prev 12:597–603. Kristal AR, Cohen JH, Qu P, Stanford JL 2002 Associations of energy, fat, calcium, and vitamin D with prostate cancer risk. Cancer Epidemiol Biomarkers Prev 11:719–725. Corder EH, Guess HA, Hulka BS, Friedman GD, Sadler M, Vollmer RT, Lobaugh B, Drezner MK, Vogelman JH, Orentreich N 1993 Vitamin D and prostate cancer: a prediagnostic study with stored sera. Cancer Epidemiol Biomarkers Prev 2:467–472. Braun MM, Helzlsouer KJ, Hollis BW, Comstock GW 1995 Prostate cancer and prediagnostic levels of serum vitamin D metabolites (Maryland, United States). Cancer Causes Control 6:235–239. Gann PH, Ma J, Hennekens CH, Hollis BW, Haddad JG, Stampfer MJ 1996 Circulating vitamin D metabolites in relation to subsequent development of prostate cancer. Cancer Epidemiol Biomarkers Prev 5:121–126. Corder EH, Friedman GD, Vogelman JH, Orentreich N 1995 Seasonal variation in vitamin D, vitamin D-binding protein, and dehydroepiandrosterone: risk of prostate cancer in black and white men. Cancer Epidemiol Biomarkers Prev 4: 655–659. Nomura AM, Stemmermann GN, Lee J, Kolonel LN, Chen TC, Turner A, Holick MF 1998 Serum vitamin D metabolite levels and the subsequent development of prostate cancer (Hawaii, United States). Cancer Causes Control 9:425–432. Schwartz GG, Hulka, BS, Morris D, Mohler JL 1992 Prostate cancer and vitamin (homone) D: a case control study. J Urol 147(Suppl):294a. Uitterlinden AG, van Leeuwen JPTM, Pols HA 2001 Genetics and genomics of osteoporosis. In: R Marcus, D Feldman, J Kelsey (eds) Osteoporosis, Vol. 1. Academic Press: San Diego, pp. 639–667. Feldman D 1997 Androgen and vitamin D receptor gene polymorphisms: the long and short of prostate cancer risk. J Natl Cancer Inst 89:109–111. Ingles SA, Coetzee GA, Ross RK, Henderson BE, Kolonel LN, Crocitto L, Wang W, Haile RW 1998 Association of prostate cancer with vitamin D receptor haplotypes in African-Americans. Cancer Res 58:1620–1623. Taylor JA, Hirvonen A, Watson M, Pittman G, Mohler JL, Bell DA 1996 Association of prostate cancer with vitamin D receptor gene polymorphism. Cancer Res 56:4108–4110. Tayeb MT, Clark C, Haites NE, Sharp L, Murray GI, McLeod HL 2003 CYP3A4 and VDR gene polymorphisms and the risk of prostate cancer in men with benign prostate hyperplasia. Br J Cancer 88:928–932. Kibel AS, Isaacs SD, Isaacs WB, Bova GS 1998 Vitamin D receptor polymorphisms and lethal prostate cancer. J Urol 160:1405–1409. Blazer DG, 3rd, Ubach DM, Bostick RM, Taylor JA 2000 Vitamin D receptor polymorphisms and prostate cancer. Mol Carcinog 27:18–23. Luscombe CJ, French ME, Liu S, Sazby MF, Jones PW, Fryer AA, Strange RC 2001 Outcome in prostate cancer associations with skin type and polymorphism in pigmentationrelated genes. Carcinogenesis 22:1343–1347. Gsur A, Madersbacher S, Haidinger G, Schatzl G, Marberger M, Vutuc C, Micksche M 2002 Vitamin D receptor gene polymorphism and prostate cancer risk. Prostate 51:30–34. Ma J, Stampfer MJ, Gann PH, Hough HL, Giovannucci E, Kelsey KT, Hennekens CH, Hunter DJ 1998 Vitamin D

1702

63.

64.

65.

66.

67.

68.

69.

70. 71.

72.

73.

74.

75.

76. 77.

receptor polymorphisms, circulating vitamin D metabolites, and risk of prostate cancer in United States physicians. Cancer Epidemiol Biomarkers Prev 7:385–390. Habuchi T, Suzuki T, Sasaki R, Wang L, Sato K, Satoh S, Akao T, Tsuchiya N, Shimoda N, Wada Y, Koizumi A, Chihara J, Ogawa O, Kato T 2000 Association of vitamin D receptor gene polymorphism with prostate cancer and benign prostatic hyperplasia in a Japanese population. Cancer Res 60:305–308. Suzuki K, Matsui H, Ohtake N, Nakata S, Takei T, Koike H, Nakazato H, Okugi H, Hasumi M, Fukabori Y, Kurokawa K, Yamanaka H 2003 Vitamin D receptor gene polymorphism in familial prostate cancer in a Japanese population. Int J Urol 10:261–266. Gross C, Krishnan AV, Malloy PJ, Eccleshall TR, Zhao XY, Feldman D 1998 The vitamin D receptor gene start codon polymorphism: a functional analysis of FokI variants. J Bone Miner Res 13:1691–1699. Xu Y, Shibata A, McNeal JE, Stamey TA, Feldman D, Peehl DM 2003 Vitamin D receptor start codon polymorphism (FokI) and prostate cancer progression. Cancer Epidemiol Biomarkers Prev 12:23–27. Chokkalingam AP, McGlynn KA, Gao YT, Pollak M, Deng J, Sesterhenn IA, Mostofi FK, Fraumeni JF Jr, Hsing AW 2001 Vitamin D receptor gene polymorphisms, insulin-like growth factors, and prostate cancer risk: a population-based casecontrol study in China. Cancer Res 61:4333–4336. Whitfield GK, Remus LS, Jurutka PW, Zitzer H, Oza AK, Dang HT, Haussler CA, Galligan MA, Thatcher ML, Encinas Dominguez C, Haussler MR 2001 Functionally relevant polymorphisms in the human nuclear vitamin D receptor gene. Mol Cell Endocrinol 177:145–159. Feldman D, Malloy PJ, Gross C 2001 Vitamin D: biology, action, and clinical implications. In: R Marcus, D Feldman, J Kelsey (eds) Osteoporosis, vol. 1. Academic Press: San Diego, pp. 257–303. Peehl DM, Krishnan AV, Feldman D 2003 Pathways mediating the growth-inhibitory actions of vitamin D in prostate cancer. J Nutr 133:2461S–2469S. Miller GJ, Stapleton GE, Ferrara JA, Lucia MS, Pfister S, Hedlund TE, Upadhya P 1992 The human prostatic carcinoma cell line LNCaP expresses biologically active, specific receptors for 1α,25-dihydroxyvitamin D3. Cancer Res 52:515–520. Skowronski RJ, Peehl DM, Feldman D 1993 Vitamin D and prostate cancer: 1,25-dihydroxyvitamin D3 receptors and actions in human prostate cancer cell lines. Endocrinology 132:1952–1960. Peehl DM, Skowronski RJ, Leung GK, Wong ST, Stamey TA, Feldman D 1994 Antiproliferative effects of 1,25-dihydroxyvitamin D3 on primary cultures of human prostatic cells. Cancer Res 54:805–810. Kivineva M, Blauer M, Syvala H, Tammela T, Tuohimaa P 1998 Localization of 1α,25-dihydroxyvitamin D3 receptor (VDR) expression in human prostate. J Steroid Biochem Mol Biol 66:121–127. Leman ES, Getzenberg RH 2003 Effects of 1,25-dihydroxyvitamin D3 on the distribution of androgen and vitamin D receptors in human prostate neonatal epithelial cells. J Cell Biochem 88:609–622. Krill D, DeFlavia P, Dhir R, Luo J, Becich MJ, Lehman E, Getzenberg RH 2001 Expression patterns of vitamin D receptor in human prostate. J Cell Biochem 82:566–572. Konety BR, Leman E, Vietmeier B, Arlotti J, Dhir R, Getzenberg RH 2000 In vitro and in vivo effects of vitamin D

ARUNA V. KRISHNAN, DONNA M. PEEHL, AND DAVID FELDMAN

78.

79.

80.

81.

82.

83.

84. 85. 86. 87. 88. 89. 90.

91.

92. 93.

(calcitriol) administration on the normal neonatal and prepubertal prostate. J Urol 164:1812–1818. Krill D, Stoner J, Konety BR, Becich MJ, Getzenberg RH 1999 Differential effects of vitamin D on normal human prostate epithelial and stromal cells in primary culture. Urology 54:171–177. Konety BR, Nangia AK, Nguyen TS, Thomas A, Getzenberg RH 1999 Effect of prenatal vitamin D (calcitriol) exposure on the growth and development of the prostate. Prostate 41:181–189. Zhao XY, Ly LH, Peehl DM, Feldman D 1997 1α, 25-dihydroxyvitamin D3 actions in LNCaP human prostate cancer cells are androgen-dependent. Endocrinology 138: 3290–3298. Esquenet M, Swinnen JV, Heyns W, Verhoeven G 1996 Control of LNCaP proliferation and differentiation: actions and interactions of androgens, 1α,25-dihydroxycholecalciferol, all-trans retinoic acid, 9-cis retinoic acid, and phenyl-acetate. Prostate 28:182–194. Miller GJ, Stapelton GE, Hedlund TE, Moffatt KA 1995 Vitamin D receptor expression, 24-hydroxylase activity, and inhibition of growth by 1α,25-dihydroxyvitamin D3 in seven human prostatic carcinoma cell lines. Clin Cancer Res 1:997–1003. Zhao XY, Peehl DM, Navone NM, Feldman D 2000 l α,25-dihydroxyvitamin D3 inhibits prostate cancer cell growth by androgen-dependent and androgen-independent mechanisms. Endocrinology 141:2548–2556. Chung LWK 1991 Fibroblasts are critical determinants in prostatic cancer growth and dissemination. Cancer Met Rev 10:263–274. Chung LWK 1995 The role of stromal-epithelial interaction in normal and malignant growth. Cancer Surveys 23:33–42. Sung SY, Chung LW 2002 Prostate tumor-stroma interaction: molecular mechanisms and opportunities for therapeutic targeting. Differentiation 70:506–521. Marker PC, Donjacour AA, Dahiya R, Cunha GR 2003 Hormonal, cellular, and molecular control of prostatic development. Dev Biol 253:165–174. Ludlow J 1993 Interactions between SV40 large-tumor antigen and the growth suppressor proteins pRB and p53. FASEB J 7:866–871. Vousden K 1993 Interactions of human papillomavirus transforming proteins with the products of tumor suppressor genes. FASEB J 7:872–879. Lee M-S, Garkovenko E, Yun JS, Weijerman PC, Peehl DM, Chen L-S, Rhim JS 1994 Characterization of adult human prostatic epithelial cells immortalized by polybrene-induced DNA transfection with a plasmid containing an origin-defective SV40 genome. Int J Oncol 4:821–830. Weijerman PC, Konig JJ, Wong ST, Niesters HG, Peehl DM 1994 Lipofection-mediated immortalization of human prostatic epithelial cells of normal and malignant origin using human papillomavirus type 18 DNA. Cancer Res 54: 5579–5583. Peehl DM, Wong ST, Rhim JS 1995 Altered growth regulation of prostatic epithelial cells by human papillomavirusinduced transformation. Int J Oncol 6:1177–1184. Gross C, Skoronski RJ, Plymate SR, Rhim JS, Peehl DM, Feldman D 1996 Simian virus 40-, but not human papillomavirus-, transformation of prostatic epithelial cells results in loss of growth-inhibition by 1,25-dihydroxyvitamin D3. Int J Oncol 8:41–47.

CHAPTER 94 Vitamin D and Prostate Cancer

94. Agadir A, Lazzaro G, Zheng Y, Zhang XK, Mehta R 1999 Resistance of HBL 100 human breast epithelial cells to vitamin D action. Carcinogenesis 20:577–582. 95. Hedlund TE, Moffatt KA, Miller GJ 1996 Stable expression of the nuclear vitamin D receptor in the human prostatic carcinoma cell line JCA-1: evidence that the antiproliferative effects of 1α,25-dihydroxyvitamin D3 are mediated exclusively through the genomic signaling pathway. Endocrinology 137:1554–1561. 96. Zhuang SH, Schwartz GG, Cameron D, Burnstein KL 1997 Vitamin D receptor content and transcriptional activity do not fully predict antiproliferative effects of vitamin D in human prostate cancer cell lines. Mol Cell Endocrinol 126:83–90. 97. Ly LH, Zhao XY, Holloway L, Feldman D 1999 Liarozole acts synergistically with 1α,25-dihydroxyvitamin D3 to inhibit growth of DU 145 human prostate cancer cells by blocking 24-hydroxylase activity. Endocrinology 140:2017–2076. 98. Albertson DG, Ylstra B, Segraves R, Collins C, Dairkee SH, Kowbel D, Kuo WL, Gray JW, Pinkel D 2000 Quantitative mapping of amplicon structure by array CGH identifies CYP24 as a candidate oncogene. Nat Genet 25:144–146. 99. Costa EM, Hirst MA, Feldman D 1985 Regulation of 1,25dihydroxyvitamin D3 receptors by vitamin D analogs in cultured mammalian cells. Endocrinology 117:2203–2210. 100. Krishnan AV, Feldman D 1997 Regulation of vitamin D receptor abundance. In: D Feldman, FH Glorieux, JW Pike (eds) Vitamin D, Vol. 1. Academic Press: San Diego, pp. 179–200. 101. Peehl DM, Seto E, Hsu JY, Feldman D 2002 Preclinical activity of ketoconazole in combination with calcitriol or the vitamin D analog EB 1089 in prostate cancer cells. J Urol 168:1583–1588. 102. Schwartz GG, Whitlatch LW, Chen TC, Lokeshwar BL, Holick MF 1998 Human prostate cells synthesize 1,25-dihydroxyvitamin D3 from 25-hydroxyvitamin D3. Cancer Epidemiol Biomarkers Prev 7:391–395. 103. Barreto AM, Schwartz GG, Woodruff R, Carmer SD 2000 25-hydroxyvitamin D3, the prohormone of 1,25-dihydroxyvitamin D3, inhibits the proliferation of primary prostatic epithelial cells. Cancer Epidemiol Biomarkers Prev 9:265–270. 104. Hsu JY, Feldman D, McNeal JE, Peehl DM 2001 Reduced 1α-hydroxylase activity in human prostate cancer cells correlates with decreased susceptibility to 25-hydroxyvitamin D3–induced growth inhibition. Cancer Res 61: 2852–2856. 105. Whitlatch LW, Young MV, Schwartz GG, Flanagan JN, Burnstein KL, Lokeshwar BL, Rich ES, Holick MF, Chen TC 2002 25-hydroxyvitamin D-1α-hydroxylase activity is diminished in human prostate cancer cells and is enhanced by gene transfer. J Steroid Biochem Mol Biol 81:135–140. 106. Chen TC, Wang L, Whitlatch LW, Flanagan JN, Holick MF 2003 Prostatic 25-hydroxyvitamin D-l α-hydroxylase and its implication in prostate cancer. J Cell Biochem 88: 315–322. 107. Segersten U, Correa P, Hewison M, Hellman P, Dralle H, Carling T, Akerstrom G, Westin G 2002 25-hydroxyvitamin D3-1α-hydroxylase expression in normal and pathological parathyroid glands. J Clin Endocrinol Metab 87:2967–2972. 108. Bareis P, Bises G, Bischof MG, Cross HS, Peterlik M 2001 25-hydroxyvitamin D metabolism in human colon cancer cells during tumor progression. Biochem Biophys Res Commun 285:1012–1017.

1703 109. Hawkins GA, Cramer SD, Zheng SL, Isaacs SD, Wiley KE, Chang BL, Bleecker ER, Walsh PC, Meyers DA, Isaacs WB, Xu J 2002 Sequence variants in the human 25-hydroxyvitamin D3 1-α-hydroxylase (CYP27B1) gene are not associated with prostate cancer risk. Prostate 53:175–178. 110. Zhao XY, Ly LH, Peehl DM, Feldman D 1999 Induction of androgen receptor by 1α,25-dihydroxyvitamin D3 and 9-cis retinoic acid in LNCaP human prostate cancer cells. Endocrinology 140:1205–1212. 111. Hsieh TY, Ng Cy, Mallouch C, Tazaki H, Wu JM 1996 Regulation of growth, PSA/PAP and androgen receptor expression by 1α,25-dihydroxyvitamin D3 in the androgendependent LNCaP cells. Biochem Biophys Res Commun 22:141–146. 112. Zhao X-Y, Skowronski R, Peehl D, Feldman D 1996 Induction of prostate-specific antigen and up-regulation of androgen receptor by 1α,25-dihydroxyvitamin D3 in the LNCaP human prostate cancer cell line. 10th International Congress of Endocrinology, Vol. 1. Endocrine Society Press: San Franciso, p. 155. 113. Krishnan AV, Shinghal R, Raghavachari N, Brooks JD, Peehl DM, Feldman D 2004 Analysis of vitamin D–regulated gene expression in LNCaP human prostate cancer cells using cDNA microarrays. Prostate 59:243–251. 114. Yang ES, Maiorino CA, Roos BA, Knight SR, Burnstein KL 2002 Vitamin D–mediated growth inhibition of an androgenablated LNCaP cell line model of human prostate cancer. Mol Cell Endocrinol 186:69–79. 115. Getzenberg RH, Light BW, Lapco PE, Konety BR, Nangia AK, Acierno JS, Dhir R, Shurin Z, Day RS, Trump DL, Johnson CS 1997 Vitamin D inhibition of prostate adenocarcinoma growth and metastasis in the Dunning rat prostate model system. Urology 50:999–1006. 116. Gingrich JR, Barrios RJ, Kattan MW, Nahm HS, Finegold MJ, Greenberg NM 1997 Androgen-independent prostate cancer progression in the TRAMP model. Cancer Res 57:4687–4691. 117. Lucia MS, Bostwick DG, Bosland M, Cockett AT, Knapp DW, Leav I, Pollard M, Rinker-Schaeffer C, Shirai T, Watkins BA 1998 Workgroup I: rodent models of prostate cancer. Prostate 36:49–55. 118. Navone NM, Logothetis CJ, von Eschenbach AC, Troncoso P 1998 Model systems of prostate cancer: uses and limitations. Cancer Metastasis Rev 17:361–371. 119. van Weerden WM, Romijn JC 2000 Use of nude mouse xenograft models in prostate cancer research. Prostate 43:263–271. 120. Schwartz GG, Hill CC, Oeler TA, Becich MJ, Bahnson RR 1995 1,25-dihydroxy-16-ene-23-yne-vitamin D3 and prostate cancer cell proliferation in vivo. Urology 46:365–369. 121. Blutt SE, Polek TC, Stewart LV, Kattan MW, Weigel NL 2000 A calcitriol analog, EB 1089, inhibits the growth of LNCaP tumors in nude mice. Cancer Res 60:779–782. 122. Vegesna V, O’Kelly J, Said J, Uskokovic M, Binderup L, Koeffler HP 2003 Ability of potent vitamin D3 analogs to inhibit growth of prostate cancer cells in vivo. Anticancer Res 23:283–289. 123. Dunning WF 1963 Prostate cancer in the rat. J Natl Cancer Inst Monogr 12:351–369. 124. Isaacs JT, Yu GW, Coffey DS 1981 The characterization of a newly identified, highly metastatic variety of Dunning R 3327 rat prostatic adenocarcinoma system: the MAT LyLu tumor. Invest Urol 19:20–23. 125. Lokeshwar BL, Schwartz GG, Selzer MG, Burnstein KL, Zhuang SH, Block NL, Binderup L 1999 Inhibition of prostate

1704

126.

127.

128. 129. 130.

131.

132.

133.

134. 135.

136.

137.

138.

139.

cancer metastasis in vivo: a comparison of 1,23-dihydroxyvitamin D (calcitriol) and EB1089. Cancer Epidemiol Biomarkers Prev 8:241–248. Oades GM, Dredge K, Kirby RS, Colston KW 2002 Vitamin D receptor-dependent antitumor effects of 1,25-dihydroxyvitamin D3 and two synthetic analogs in three in vivo models of prostate cancer. BJU Int 90:607–616. Perez-Stable CM, Schwartz GG, Farinas A, Finegold M, Binderup L, Howard GA, Roos BA 2002 The Ggamma/T-15 transgenic mouse model of androgen-independent prostate cancer: target cells of carcinogenesis and the effect of the vitamin D analog EB 1089. Cancer Epidemiol Biomarkers Prev 11:555–563. Xue L, Lipkin M, Newmark H, Wang J 1999 Influence of dietary calcium and vitamin D on diet-induced epithelial cell hyperproliferation in mice. J Natl Cancer Inst 91:176–181. Schwartz GG, Oeler TA, Uskokovic R, Bahnson RR 1994 Human prostate cancer cells: inhibition of proliferation by vitamin D analogs. Anticancer Res 14:1077–1081. Skowronski RJ, Peehl DM, Feldman D 1995 Actions of vitamin D3, analogs on human prostate cancer cell lines: comparison with 1,25-dihydroxyvitamin D3. Endocrinology 136:20–26. Campbell MJ, Elstner E, Holden S, Uskokovic M, Koeffler HP 1997 Inhibition of proliferation of prostate cancer cells by a 19-nor-hexafluoride vitamin D3 analog involves the induction of p21 waf1, p27kip1 and E-cadherin. J Mol Endocrinol 19:15–27. Hedlund TE, Moffatt KA, Uskokovic MR, Miller GJ 1997 Three synthetic vitamin D analogs induce prostate-specific acid phosphatase and prostate-specific antigen while inhibiting the growth of human prostate cancer cells in a vitamin D receptor-dependent fashion. Clin Cancer Res 3:1331–1338. de Vos S, Holden S, Heber D, Elstner E, Binderup L, Uskokovic M, Rude B, Chen DL, Le J, Cho SK, Koeffler HP 1997 Effects of potent vitamin D3 analogs on clonal proliferation of human prostate cancer cell lines. Prostate 31:77–83. Wang X, Chen X, Akhter J, Morris DL 1997 The in vitro effect of vitamin D3 analog EB-1089 on a human prostate cancer cell line (PC-3). Br J Urol 80:260–262. Kubota T, Koshizuka K, Koike M, Uskokovic M, Miyoshi I, Koeffler HP 1998 19-nor-26,27-bishomo-vitamin D3 analogs: a unique class of potent inhibitors of proliferation of prostate, breast, and hematopoietic cancer cells. Cancer Res 58:3370–3375. Campbell MJ, Reddy GS, Koeffler HP 1997 Vitamin D3 analogs and their 24-oxo metabolites equally inhibit clonal proliferation of a variety of cancer cells but have differing molecular effects. J Cell Biochem 66:413–425. Hisatake J, Kubota T, Hisatake Y, Uskokovic M, Tomoyasu S, Koeffler HP 1999 5,6-trans-16-ene-vitamin D3: a new class of potent inhibitors of proliferation of prostate, breast, and myeloid leukemic cells. Cancer Res 59:4023–4029. Swami S, Zhao XY, Sarabia S, Siu-Caldera M-L, Uskokovic M, Feldman D 2003 A low-calcemic vitamin D analog (RO 25-4020) inhibits the growth of LNCaP human prostate cancer cells with increased potency by producing an active 24-oxo metabolite (RO 29-9970). In: J Reichrath, M Friedrich, W Tilgen (eds) Recent Results in Cancer Research, Vol. 164. Springer-Verlag: Berlin, vol. 164, pp. 349–352. Chen TC, Schwartz GG, Burnstein KL, Lokeshwar BL, Holick MF 2000 The in vitro evaluation of 25-hydroxyvitamin D3 and 19-nor-1α,25-dihydroxyvitamin D2 as

ARUNA V. KRISHNAN, DONNA M. PEEHL, AND DAVID FELDMAN

140.

141.

142.

143. 144.

145.

146.

147.

148.

149.

150.

151.

152.

153. 154.

therapeutic agents for prostate cancer. Clin Cancer Res 6:901–908. Bauer JA, Thompson TA, Church DR, Ariazi EA, Wilding G 2003 Growth inhibition and differentiation in human prostate carcinoma cells induced by the vitamin D analog 1 α, 24-dihydroxyvitamin D2. Prostate 55:159–167. Polek TC, Murthy S, Blutt SE, Boehm MF, Zou A, Weigel NL, Allegretto EA 2001 Novel nonsecosteroidal vitamin D receptor modulator inhibits the growth of LNCaP xenograft tumors in athymic mice without increased serum calcium. Prostate 49:224–233. Crescioli C, Ferruzzi P, Caporali A, Mancina R, Comerci A, Muratori M, Scaltriti M, Vannelli GB, Smiroldo S, Mariani R, Villari D, Bettuzzi S, Serio M, Adorini L, Maggi M 2003 Inhibition of spontaneous and androgen-induced prostate growth by a nonhypercalcemic calcitriol analog. Endocrinology 144:3046–3057. Freedman LP 1999 Transcriptional targets of the vitamin D3 receptor-mediating cell cycle arrest and differentiation. J Nutr 129:581S–586S. Blutt SE, Allegretto EA, Pike JW, Weigel NL 1997 1,25dihydroxyvitamin D3 and 9-cis-retinoic acid act synergistically to inhibit the growth of LNCaP prostate cells and cause accumulation of cells in G1. Endocrinology 138:1491–1497. Zhuang SH, Burnstein KL 1998 Antiproliferative effect of 1α,25-dihydroxyvitamin D3 in human prostate cancer cell line LNCaP involves reduction of cyclin-dependent kinase 2 activity and persistent G1 accumulation. Endo-crinology 139:1197–1207. Liu M, Lee MH, Cohen M, Bommakanti M, Freedman LP 1996 Transcriptional activation of the Cdk inhibitor p21 by vitamin D3 leads to the induced differentiation of the myelomonocytic cell line U937. Genes Dev 10:142–153. Boyle BJ, Zhao XY, Cohen P, Feldman D 2001 Insulin-like growth factor binding protein-3 mediates 1α,25-dihydroxyvitamin D3 growth inhibition in the LNCaP prostate cancer cell line through p21/WAF1. J Urol 165:1319–1324. Moffatt KA, Johannes WU, Hedlund TE, Miller GJ 2001 Growth inhibitory effects of 1α,25-dihydroxy-vitamin D3 are mediated by increased levels of p21 in the prostatic carcinoma cell line ALVA-31. Cancer Res 61:7122–7129. Polek TC, Stewart LV, Ryu EJ, Cohen MB, Allegretto EA, Weigel NL 2003 p53 Is required for 1,25-dihydroxyvitamin D3–induced G0 arrest but is not required for G1 accumulation or apoptosis of LNCaP prostate cancer cells. Endocrinology 144:50–60. el-Deiry WS, Tokino T, Velculescu VE, Levy DB, Parsons R, Trent JM, Lin D, Mercer WE, Kinzler KW, Vogelstein B 1993 WAF1, a potential mediator of p53 tumor suppression. Cell 75:817–825. Bookstein R, Shew JY, Chen PL, Scully P, Lee WH 1990 Suppression of tumorigenicity of human prostate carcinoma cells by replacing a mutated RB gene. Science 247:712–715. Hsieh T, Wu JM 1997 Induction of apoptosis and altered nuclear/cytoplasmic distribution of the androgen receptor and prostate-specific antigen by 1α,25-dihydroxyvitamin D3 in androgen-responsive LNCaP cells. Biochem Biophys Res Commun 235:539–544. Fife RS, Sledge GW Jr, Proctor C 1997 Effects of vitamin D3 on proliferation of cancer cells in vitro. Cancer Lett 120:65–69. Blutt SE, McDonnell TJ, Polek TC, Weigel NL 2000 Calcitriol-induced apotosis in LNCaP cells is blocked by overexpression of Bcl-2. Endocrinology 141:10–17.

CHAPTER 94 Vitamin D and Prostate Cancer

155. Guzey M, Kitada S, Reed JC 2002 Apoptosis induction by 1α,25-dihydroxyvitamin D3 in prostate cancer. Mol Cancer Ther 1:667–677. 156. Crescioli C, Maggi M, Luconi M, Vannelli GB, Salerno R, Sinisi AA, Bonaccorsi L, Ferruzzi P, Barni T, Forti G, Serio M 2002 Vitamin D3 analog inhibits keratinocyte growth factor signaling and induces apoptosis in human prostate cancer cells. Prostate 50:15–26. 157. Konety BR, Schwartz GG, Acierno JS Jr, Becich MJ, Getzenberg RH 1996 The role of vitamin D in normal prostate growth and differentiation. Cell Growth Differ 7:1563–1570. 158. Craft N, Sawyers CL 1998 Mechanistic concepts in androgen-dependence of prostate cancer. Cancer Metastasis Rev 17:421–427. 159. Gleave ME, Hsieh JT, von Eschenbach AC, Chung LW 1992 Prostate and bone fibroblasts induce human prostate cancer growth in vivo: implications for bidirectional tumor-stromal cell interaction in prostate carcinoma growth and metastasis. J Urol 147:1151–1159. 160. Bogdanos, J, Karamanolakis D, Tenta R, Tsintavis A, Milathianakis C, Mitsiades C, Koutsilieris M 2003 Endocrine/ paracrine/autocrine survival factor activity of bone microenvironment participates in the development of androgen ablation and chemotherapy refractoriness of prostate cancer metastasis in skeleton. Endocr Relat Cancer 10:279–289. 161. Drivdahl RH, Loop SM, Andress DL, Ostenson RC 1995 IGF-binding proteins in human prostate tumor cells: expression and regulation by 1,25-dihydroxyvitamin D3. Prostate 26:72–79. 162. Huynh H, Pollak M, Zhang JC 1998 Regulation of insulinlike growth factor (IGF) II and IGF binding protein 3 autocrine loop in human PC-3 prostate cancer cells by vitamin D metabolite 1,25(OH)2D3 and its analog EB 1089. Int J Oncol 13:137–143. 163. Nickerson T, Huynh H 1999 Vitamin D analog EB 1089induced prostate regression is associated with increased gene expression of insulin-like growth factor binding proteins. J Endocrimal 160:223–229. 163a. Peng L, Malloy PJ, Feldman D 2004 Identification of a functional vitamin D response element within the human insulin-like growth factor binding protein-3 promotor. Molec Endo 18:1109–1119. 164. Lee C, Sintich SM, Mathews EP, Shah AH, Kundu SD, Perry KT, Cho JS, Ilio KY, Cronauer MV, Janulis L, Sensibar JA 1999 Transforming growth factor-beta in benign and malignant prostate. Prostate 39:285–290. 165. Hwa V, Oh Y, Rosenfeld RG 1997 Insulin-like growth factor binding protein-3 and -5 are regulated by transforming growth factor-beta and retinoic acid in the human prostate adenocarcinoma cell line PC-3. Endocrine 6:235–242. 166. Rajah R, Valentinis B, Cohen P 1997 Insulin-like growth factor (IGF)-binding protein-3 induces apoptosis and mediates the effects of transforming growth factor-beta1 on programmed cell death through a p53- and IGF-independent mechanism. J Biol Chem 272:12181–12188. 167. Danielpour D 1996 Induction of transforming growth factorbeta autocrine activity by all-trans-retinoic acid and 1α,25dihydroxyvitamin D3 in NRP-152 rat prostatic epithelial cells. J Cell Physiol 166:231–239. 168. Wu Y, Craig TA, Lutz WH, Kumar R 1999 Identification of 1α,25-dihydroxyvitamin D3 response elements in the human transforming growth factor beta 2 gene. Biochemistry 38:2654–2660.

1705 169. Hansen CM, Frandsen TL, Brunner N, Binderup L 1994 1α,25-dihydroxyvitamin D3 inhibits the invasive potential of human breast cancer cells in vitro. Clin Exp Metastasis 12:195–202. 170. Young MR, Ihm J, Lozano Y, Wright MA, Prechel MM 1995 Treating tumor-bearing mice with vitamin D3 diminishes tumor-induced myelopoiesis and associated immunosuppression, and reduces tumor metastasis and recurrence. Cancer Immunol Immunother 41:37–45. 171. Schwartz GG, Wang MH, Zang M, Singh RK, Siegal GP 1997 1α,25-Dihydroxyvitamin D (calcitriol) inhibits the invasiveness of human prostate cancer cells. Cancer Epidemiol Biomarkers Prev 6:727–732. 172. Sung V, Feldman D 2000 1,25-dihydroxyvitamin D3 decreases human prostate cancer cell adhesion and migration. Mol Cell Endocrinol 164:133–143. 173. Majewski S, Skopinska M, Marczak M, Szmurlo A, Bollag W, Jablonska S 1996 Vitamin D3 is a potent inhibitor of tumor cell-induced angiogenesis. J Investig Dermatol Symp Proc 1:97–101. 174. Mantell DJ, Owens PE, Bundred NJ, Mawer EB, Canfield AE 2000 1α,25-dihydroxyvitamin D3 inhibits angiogenesis in vitro and in vivo. Circ Res 87:214–220. 175. Bernardi RJ, Johnson CS, Modzelewski RA, Trump DL 2002 Antiproliferative effects of 1α,25-dihydroxyvitamin D3 and vitamin D analogs on tumor-derived endothelial cells. Endocrinology 143:2508–2514. 176. Tosetti F, Ferrari N, De Flora S, Albini A 2002 “Angioprevention” angiogenesis is a common and key target for cancer chemopreventive agents. FASEB J 16:2–14. 177. Ulrix W, Swinnen JV, Heyns W, Verhoeven G 1999 The differentiation-related gene 1, Drg1, is markedly up-regulated by androgens in LNCaP prostatic adenocarcinoma cells. FEBS Lett 455:23–26. 178. Tong M, Tai HH 2000 Induction of NAD(+)-linked 15hydroxyprostaglandin dehydrogenase expression by androgens in human prostate cancer cells. Biochem Biophys Res Commun 276:77–81. 179. Schwartz Z, Sylvia VL, Larsson D, Nemere I, Casasola D, Dean DD, Boyan BD 2002 1α,25(OH)2D3 regulates chondrocyte matrix vesicle protein Kinase C (PKC) directly via G-protein-dependent mechanisms and indirectly via incorporation of PKC during matrix vesicle biogenesis. J Biol Chem 277:11828–11837. 180. Zhang XH, Jin L, Sakamoto H, Takenaka I 1996 Immunohistochemical localization of metallothionein in human prostate cancer. J Urol 156:1679–1681. 181. Blagosklonny MV, Dixon SC, Figg WD 2000 Efficacy of microtubule-active drugs followed by ketoconazole in human metastatic prostate cancer cell lines. J Urol 163:1022–1026. 182. Feldman D 1986 Ketoconazole and other imidazole derivatives as inhibitors of steroidogenesis. Endocr Rev 7:409–420. 183. Loose DS, Kan PB, Hirst MA, Marcus RA, Feldman D 1983 Ketoconazole blocks adrenal steroidogenesis by inhibiting cytochrome P450-dependent enzymes. J Clin Invest 71:1495–1499. 184. Glass AR, Eil C 1986 Ketoconazole-induced reduction in serum 1,25-dihydroxyvitamin D. J Clin Endocrinol Metab 63:766–769. 185. Tallman MS, Wiernik PH 1992 Retinoids in cancer treatment. J Clin Pharmacol 32:868–888. 186. Pasquali D, Thaller C, Eichele G 1996 Abnormal level of retinoid acid in prostate cancer tissues. J Clin Endocrinol Metab 81:2186–2191.

1706 187. Pienta KJ, Nguyen NM, Lehr JE 1993 Treatment of prostate cancer in the rat with the synthetic retinoid fenretinide. Cancer Res 53:224–226. 188. Slawin K, Kadmon D, Park SH, Scardino PT, Anzano M, Sporn MB, Thompson TC 1993 Dietary fenretinide, a synthetic retinoid, decreases the tumor incidence and the tumor mass of ras+myc-induced carcinomas in the mouse prostate reconstitution model system. Cancer Res 53:4461–4465. 189. Stearns ME, Wang M, Fudge K 1993 Liarazole and 13-cis retinoic acid anti-prostatic tumor activity. Cancer Res 53: 3073–3077. 190. Peehl DM, Wong ST, Cramer SD, Gross C, Feldman D 1995 Suramin, hydrocortisone, and retinoic acid modify inhibitory effects of 1,25-dihydroxyvitamin D3 on prostatic epithelial cells. Urol Oncol 1:188–194. 191. Campbell MJ, Park S, Uskokovic MR, Dawson MI, Koeffler HP 1998 Expression of retinoic acid receptor-beta sensitizes prostate cancer cells to growth inhibition mediated by combinations of retinoids and a 19-nor hexafluoride vitamin D3 analog. Endocrinology 139:1972–1980. 192. Campbell MJ, Park S, Uskokovic MR, Dawson MI, Jong L, Koeffler HP 1999 Synergistic inhibition of prostate cancer cell lines by a 19-nor hexafluoride vitamin D3 analog and anti-activator protein 1 retinoid. Br J Cancer 79:101–107. 193. Elstner E, Campbell MJ, Munker R, Shintaku P, Binderup L, Heber D, Said J, Koeffler HP 1999 Novel 20-epi-vitamin D3 analog combined with 9-cis-retinoic acid markedly inhibits colony growth of prostate cancer cells. Prostate 40:141–149. 194. Fakih M, Johnson CS, Trump DL 2002 Glucocorticoids and treatment of prostate cancer: a preclinical and clinical review. Urology 60:553–561. 195. Smith RG, Syms AJ, Norris JS 1984 Differential effects of androgens and glucocorticoids on regulation of androgen receptor concentrations and cell growth. J Steroid Biochem 20:277–281. 196. Carollo M, Parente L, D’Alessandro N 1998 Dexamethasoneinduced cytotoxic activity and drug resistance effects in androgen-independent prostate tumor PC-3 cells are mediated by lipocortin 1. Oncol Res 10:245–254. 197. Krishnan AV, Zhao XY, Swami S, Brive L, Peehl DM, Ely KR, Feldman D 2002 A glucocorticoid-responsive mutant androgen receptors exhibits unique ligand specificity: therapeutic implications for androgen-independent prostate cancer. Endocrinology 143:1889–1900. 198. Yu WD, McElwain MC, Modzelewski RA, Russell DM, Smith DC, Trump DL, Johnson CS 1998 Enhancement of 1,25-dihydroxyvitamin D3-mediated antitumor activity with dexamethasone. J Natl Cancer Inst 90:134–141. 199. Bernardi RJ, Trump DL, Yu WD, McGuire TF, Hershberger PA, Johnson CS 2001 Combination of 1alpha,25-dihydroxyvitamin D3 with dexamethasone enhances cell cycle arrest and apoptosis: role of nuclear receptor cross-talk and Erk/Akt signaling. Clin Cancer Res 7:4164–4173. 200. Ahmed S, Johnson CS, Rueger RM, Trump DL 2002 Calcitriol (1,25-dihydroxycholecalciferol) potentiates activity of mitoxantrone/dexamethasone in an androgen-independent prostate cancer model. J Urol 168:756–761. 201. Koeffler HP 2003 Peroxisome proliferator-activated receptor gamma and cancers. Clin Cancer Res 9:1–9. 202. Kubota T, Koshizuka K, Williamson EA, Asou H, Said JW, Holden S, Miyoshi I, Koeffler HP 1998 Ligand for peroxisome proliferator-activated receptor gamma (troglitazeon) has potent antitumor effect against human prostate cancer both in vitro and in vivo. Cancer Res 58:3344–3352.

ARUNA V. KRISHNAN, DONNA M. PEEHL, AND DAVID FELDMAN

203. Shappell SB, Gupta RA, Manning S, Whitehead R, Boeglin WE, Schneider C, Case T, Price J, Jack GS, Wheeler TM, Matusik RJ, Brash AR, Dubois RN 2001 15S-Hydroxyeicosatetraenoic acid activates peroxisome proliferator-activated receptor gamma and inhibits proliferation in PC3 prostate carcinoma cells. Cancer Res 61:497–503. 204. Xu Y, Iyengar S, Roberts RL, Shappell SB, Peehl DM 2003 Primary culture model of peroxisome proliferator-activated receptor gamma activity in prostate cancer cells. J Cell Physiol 196:131–143. 205. Hisatake JI, Ikezoe T, Carey M, Holden S, Tomoyasu S, Koeffler HP 2000 Down-regulation of prostate-specific antigen expression by ligands for peroxisome proliferatoractivated receptor gamma in human prostate cancer. Cancer Res 60:5494–5498. 206. Mueller E, Smith M, Sarraf P, Kroll T, Aiyer A, Kaufman DS, Oh W, Demetri G, Figg WD, Zhou XP, Eng C, Spiegelman BM, Kantoff PW 2000 Effects of ligand activation of peroxisome proliferator-activated receptor gamma in human prostate cancer. Proc Natl Acad Sci USA 97:10990–10995. 207. Severson RK, Nomura AM, Grove JS, Stemmermann GN 1989 A prospective study of demographics, diet, and prostate cancer among men of Japanese ancestry in Hawaii. Cancer Res 49:1857–1860. 208. Hempstock J, Kavanagh JP, George NJ 1998 Growth inhibition of prostate cell lines in vitro by phytoestrogens. Br Urol 82:560–563. 209. Davis JN, Kucuk O, Sarkar FH 2002 Expression of prostatespecific antigen is transcriptionally regulated by genistein in prostate cancer cells. Mol Carcinog 34:91–101. 210. Fritz WA, Wang J, Eltoum IE, Lamartiniere CA 2002 Dietary genistein down-regulates androgen and estrogen receptor expression in the rat prostate. Mol Cell Endocrinol 186: 89–99. 211. Pollard M, Wolter W 2000 Prevention of spontaneous prostate-related cancer in Lobund-Wistar rats by a soy protein isolate/isoflavone diet. Prostate 45:101–105. 212. Mentor-Marcel R, Lamartiniere CA, Eltoum IE, Greenberg NM, Elgavish A 2001 Genistein in the diet reduces the incidence of poorly differentiated prostatic adenocarcinoma in transgenic mice (TRAMP). Cancer Res 61:6777–6782. 213. Weber KS, Setchell KD, Stocco DM, Lephart ED 2001 Dietary soy-phytoestrogens decrease testosterone levels and prostate weight without altering LH, prostate 5α-reductase, or testicular steroidogenic acute regulatory peptide levels in adult male Sprague-Dawley rats. J Endocrinol 170:591–599. 214. Rao A, Woodruff RD, Wade WN, Kute TE, Cramer SD 2002 Genistein and vitamin D synergistically inhibit human prostatic epithelial cell growth. J Nutr 132:3191–3194. 215. Farhan H, Wahala K, Cross HS 2003 Genistein inhibits vitamin D hydroxylases CYP24 and CYP27B1 expression in prostate cells. J Steroid Biochem Mol Biol 84:423–429. 216. Wietzke JA, Welsh J 2003 Phytoestrogen regulation of a vitamin D3 receptor promoter and 1,25-dihydroxyvitamin D3 actions in human breast cancer cells. J Steroid Biochem Mol Biol 84:149–157. 217. Cho YL, Christensen C, Saunders DE, Lawrence WD, Deppe G, Malviya VK, Malone JM 1991 Combined effects of 1,25dihydroxyvitamin D3 and platinum drugs on the growth of MCF-7 cells. Cancer Res 51:2848–2853. 218. Moffatt KA, Johannes WU, Miller GJ 1999 1α,25dihydroxyvitamin D3 and platinum drugs act synergistically to inhibit the growth of prostate cancer cell lines. Clin Cancer Res 5:695–703.

1707

CHAPTER 94 Vitamin D and Prostate Cancer

219. Wang Q, Yang W, Uytingco MS, Christakos S, Wieder R 2000 1,25-dihydroxyvitamin D3 and all-trans-retinoic acid sensitize breast cancer cells to chemotherapy-induced cell death. Cancer Res 60:2040–2048. 220. Koshizuka K, Koike M, Asou H, Cho SK, Stephen T, Rude RK, Binderup L, Uskokovic M, Koeffler HP 1999 Combined effect of vitamin D3 analogs and paclitaxel on the growth of MCF-7 breast cancer cells in vivo. Breast Cancer Res Treat 53:113–120. 221. Hershberger PA, Yu WD, Modzelewski RA, Rueger RM, Johnson CS, Trump DL 2001 Calcitriol (1,25-dihydroxycholecalciferol) enhances paclitaxel antitumor activity in vitro and in vivo and accelerates paclitaxel-induced apotosis. Clin Cancer Res 7:1043–1051. 222. Barboule N, Chadebech P, Baldin V, Vidal S, Valette A 1997 Involvement of p21 in mitotic exit after paclitaxel treatment in MCF-7 breast adenocarcinoma cell line. Oncogene 15:2867–2875. 223. Stewart ZA, Mays D, Pietenpol JA 1999 Defective G1-S cell cycle checkpoint function sensitizes cells to microtubule inhibitor-induced apoptosis. Cancer Res 59: 3831–3837. 224. Rashid SF, Moore JS, Walker E, Driver PM, Engel J, Edwards CE, Brown G, Uskokovic MR, Campbell MJ 2001 Synergistic growth inhibition of prostate cancer cells by 1α,25dihydroxyvitamin D3 and its 19-nor-hexafluoride analogs in combination with either sodium butyrate or trichostatin A. Oncogene 20:1860–1872. 225. Osborn JL, Schwartz GG, Smith DC, Bahnson R, Day R, Trump DL 1995 Phase II trial of oral 1,25-dihydroxyvitamin D (calcitriol) in hormone refractory prostate cancer. Urol Oncol 1:195–198. 226. Gross C, Stamey T, Hancock S, Feldman D 1998 Treatment of early recurrent prostate cancer with 1,25-dihydroxy-

227.

228.

229.

230. 231.

232. 233.

vitamin D3 (calcitriol) [published erratum appears in J Urol 1998 Sep; 160(3 Pt 1):840]. J Urol 159:2035–2039; discussion 2039–2040. Smith DC, Johnson CS, Freeman CC, Muindi J, Wilson JW, Trump DL 1999 A phase I trial of calcitriol (1,25-dihydroxycholecalciferol) in patients with advanced malignancy. Clin Cancer Res 5:1339–1345. Beer TM, Lemmon D, Lowe BA, Henner WD 2003 Highdose weekly oral calcitriol in patients with a rising PSA after prostatectomy or radiation for prostate carcinoma. Cancer 97:1217–1224. Liu G, Oettel K, Ripple G, Staab MJ, Horvath D, Alberti D, Arzoomanian R, Marnocha R, Bruskewitz R, Mazess R, Bishop C, Bhattacharya A, Bailey H, Wilding G 2002 Phase I trial of 1α-hydroxyvitamin D2 in patients with hormone refractory prostate cancer. Clin Cancer Res 8: 2820–2827. Johnson CS, Hershberger PA, Bernardi RJ, McGuire TF, Trump DL 2002 Vitamin D receptor: a potential target for intervention. Urology 60:123–130; discussion 130–121. Muindi JR, Peng Y, Potter DM, Hershberger PA, Tauch JS, Capozzoli MJ, Egorin MJ, Johnson CS, Trump DL 2002 Pharmacokinetics of high-dose oral calcitriol: results from a phase 1 trial of calcitriol and paclitaxel. Clin Pharmacol Ther 72:648–659. Beer TM, Hough KM, Garzotto M, Lowe BA, Henner WD 2001 Weekly high-dose calcitriol and docetaxel in advanced prostate cancer. Semin Oncol 28:49–55. Beer TM, Eilers KM, Garzotto M, Egorin MJ, Lowe BA, Henner WD 2003 Weekly high-dose calcitriol and docetaxel in metastatic androgen-independent prostate cancer. J Clin Oncol 21: 123–128.

CHAPTER 95

Vitamin D and Colon Cancer HEIDE S. CROSS Department of Pathophysiology, Medical University of Vienna, Austria I. Introduction II. Molecular Basis of Vitamin D Action on Neoplastic Colonocytes III. Vitamin D Metabolism in Normal and Neoplastic Colon Cells

IV. Nutritional Regulation of CYP27B1 and CYP24 V. Conclusion References

I. INTRODUCTION

than 3.75 micrograms vitamin D/day was associated with a 50% reduction in the incidence of colorectal cancer [11]. The same nested case-control study based on a cohort of more than 25,000 individuals demonstrated that a moderately elevated serum concentration of 25-(OH)-D3 (65–100 nmol/l) was associated with a highly significant reduction in colorectal cancer incidence [11]. Several later studies, however, provided ambiguous results: a large 10–17 year retrospective study of Washington County residents did not provide any link between colorectal cancer incidence and serum levels of 25-(OH)-D3, or of 1,25-dihydroxyvitamin D3 (1,25-(OH)2-D3) evaluated prior to disease occurrence [12]. When serum 25-(OH)-D3 concentrations in patients with colonic neoplasia were compared with those of noncancer patients, no correlation with the disease was found either [13]. Tangrea et al. [14], however, did find that the estimated relative risk of large bowel cancer decreased with increasing serum 25-(OH)-D3 levels, and that the association was most pronounced for rectal cancer in a nested case-control study within a Finnish clinical trial cohort. It is highly interesting that in this large study levels of 1,25-(OH)2-D3 again did not correlate at all with colon cancer risk. One of the very few studies demonstrating a correlation between low circulating levels of 1,25-(OH)2-D3 (below 26 pg/ml, a level considered to be below normal) and enhanced risk was found in the prospective Nurses’ Health Study. A higher risk of distal colorectal adenomas was found in individuals with lower than 26 pg 1,25-(OH)2-D3/ml serum [15]. In another study by Niv et al. [16], which was marred by the small number of patients involved, a steady reduction of serum 1,25-(OH)2-D3 was observed in parallel with advancing tumor stages, but not with the biological tumor grade. A plethora of studies has been based on semiquantitative food frequency questionnaires, which often involved not only vitamin D but dairy food intake as well. Therefore, calcium ingestion may be a confounding factor in these evaluations. The consensus is that,

A. Colonic Tumor Prevention by Vitamin D 1. EPIDEMIOLOGICAL EVIDENCE

As early as 1980 Garland et al. [1] proposed that vitamin D may be a protective factor against colorectal cancer. This hypothesis was based on the observation that the geographic distribution of colon cancer mortality in the U.S.A. was highest in regions where the population was least exposed to solar radiation. UV-B is responsible for vitamin D production in the skin, and serum levels of 25-hydroxyvitamin D3 (25-(OH)-D3) are a direct reflection of sunlight exposure, of the use of sun blockers, and of skin pigmentation [2]. Thus, low serum 25-(OH)-D3 levels generally found in African-Americans probably are a reflection of reduced vitamin D synthesis due to high melanin concentrations in the skin [3]. This population segment also has enhanced incidence of colorectal, breast, and prostate cancer [4]. The link between colorectal cancer incidence and solar radiation was later confirmed by Freedman et al. [5] and by several large studies comparing southern and northern parts of the U.S.A. [6]. Recently, Grant [7] suggested that actually 20–30% of colorectal cancer incidence is due to insufficient exposure to sunlight (see Chapters 90 and 91). A follow-up study by Garland et al. [8] in almost 2000 males demonstrated that risk of colorectal cancer correlated inversely with dietary vitamin D and calcium intake. Further studies on dietary vitamin D and calcium (see [9]) appeared to confirm these data. When Garland et al. [10] demonstrated in blood samples from a Washington County population that a serum 25-(OH)-D3 concentration of 20 ng/ml or more decreased threefold the risk of colon cancer, a direct connection between serum vitamin D levels and colon cancer incidence appeared to be established. However, in this study, cancer incidence was evaluated only 1–8 years after blood sampling. In a 19-years prospective study, it was subsequently established that a dietary intake of more VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

1710 if vitamin D consumption has any preventive effect on colorectal cancer incidence, it is a very modest one. However, this protective effect might be augmented by intake of multivitamin supplements (see [17–23]). In this respect, the argument was raised that solar radiation or nutritional vitamin D intake, for instance by consuming fatty fish, would not be sufficient to effectively prevent the occurrence of colorectal tumors. Indeed, in a later study Garland et al. [24] suggested that intake of at least 800 IU (20 micrograms) of vitamin D3 together with 1000 mg calcium would be needed to significantly reduce the incidence and mortality rates of human colon cancer. 2. CONCLUSION

Thus, cumulative epidemiological evidence suggests that there is a direct correlation between reduced colorectal cancer incidence and sunlight exposure, low skin pigmentation, nutritional vitamin D intake, and high serum levels of 25-(OH)-D3. This association could be strengthened by vitamin D supplementation. In a recent human pilot study, Holt et al. [25] demonstrated for the first time that rectal crypt proliferation was inversely correlated with 25-(OH)-D3 levels in serum. This indicates, though in an indirect way, that low 25-(OH)-D3 levels may be indeed associated with colorectal cancer incidence. However, no relationship between serum 1,25-(OH)2-D3 and disease occurrence was convincingly apparent in any of the cited studies, except at very low serum 25-OH-D concentrations.

II. MOLECULAR BASIS OF VITAMIN D ACTION ON NEOPLASTIC COLONOCYTES Since the 1980s, 1,25-(OH)2-D3 has been recognized as a potent cellular antiproliferative and prodifferentiating agent in the colon. More recently, there has been intense interest in its effects on apoptosis, malignant cell invasion, and metastasis. The classical signaling pathway is via a nuclear vitamin D receptor (VDR), which is a transcription factor (see [26]). The existence of a separate “membrane” receptor has also been suggested [27]; however, recent data from a VDR knockout mouse provide good evidence that this purported receptor is of minor importance in the intestine [28].

A. The Vitamin D Receptor in Normal and Malignant Colon Cells In 1982, Frampton et al. [29] provided evidence that the VDR was present in many human cancer cell lines

HEIDE S. CROSS

and also in all colonic cell lines they investigated. Since then, the presence of the VDR has been extensively studied in a variety of colon cancer cell lines. Giuliano et al. [30] showed in Caco-2 cells that the VDR was functional in these colon cancer cells and that expression was increased when Caco-2 cells became confluent and differentiated in culture. Interestingly, Zhao and Feldman [31] demonstrated convincingly that, at least in HT-29 cells which were differentiated by chemical means, VDR abundance was actually decreased with decreased proliferation and increased differentiation. Brehier and Thomasset [32] found no specific binding of 1,25-(OH)2-D3 in differentiated HT-29 cells. Harper et al. [33] also found smaller amounts of the receptor in galactose-grown, i.e. differentiated, HT-29 cells when compared with undifferentiated (glucosegrown) HT-29 cells. In Caco-2 cells, however, there was strong expression of the VDR upon differentiation. Conversely, activation of proliferation in these cells by epidermal growth factor (EGF) resulted in down-regulation of the expression of the high affinity receptor [34]. The question of whether the VDR was more highly expressed in proliferating or differentiated colon cancer cells was further studied by Shabahang et al. [35]. Their conclusion was that the more differentiated the colonic cell lines were, the higher was their VDR expression. They also evaluated presence of the VDR in human malignant colonic tissue. In the majority of these tumors, they found lower expression of VDR than in tumor-adjacent normal mucosa from the same patient; however, the number of cases analyzed was small (12 patients). Lointier et al. [36] investigated VDR expression in 23 human tumor tissue specimens and in adjacent normal-appearing mucosa. They did not observe any difference in VDR distribution between normal right and left colon, or the rectum. However, they did find the receptor in most of the normal tissue specimen, but only in very few of the adenocarcinomas. Notably, all positive adenocarcinomas were of the welldifferentiated (low grade) type. Vandewalle et al. [37] demonstrated significantly higher VDR expression in transformed colon than in normal tissue in the proximal and distal colon, but not in the rectum. These data were accrued by binding studies with tissue homogenates. Cross et al. [38] were the first to demonstrate by immunoblotting that VDR protein expression was increasing during the transition from normal mucosa to polyps and during progression into malignancy. In rather advanced tumor stages, however, expression was diminished or disappeared completely. This suggests that colon cancer cells express the VDR as long as they retain a certain level of differentiation. This could explain why their data differed from those obtained by

1711

CHAPTER 95 Vitamin D and Colon Cancer

TABLE I Semi-quantitative Evaluation of VDR and EGFR mRNA Expression in Epithelial Cells Adenocarcinoma Normal adjacent mucosa VDR EGFR

17.5 ±1.4 58.2 ± 5.5

Low grade

High grade

125.0 ± 19.7 * 122.0 ± 18.1 *

46.3 ± 5.5 142.5 ± 27*

In situ hybridization (IHS) reactivity scores were calculated by multiplying the percentage of receptor positive cells by the average signal intensity. Data are mean ± SEM. * indicates statistically significant difference from respective IHS score in normal adjacent mucosa at p < 0.01 (using Student’s unpaired t test).

some other groups, who did not take into account tumor staging and grading. Further studies by Sheinin et al. [39], on a larger number of human colon tissues demonstrated convincingly, by in situ techniques, that in normal colon VDR expression is weak and positivity is found mainly in luminal cells, i.e. differentiated crypt cells. During tumor progression, the number of VDR-positive colonocytes increased dramatically in parallel with epidermal growth factor receptor (EGFR) expression, and it reached its maximum in low-grade (well to moderately differentiated) tumor tissue, whereas in high grade cancers (G3/4, low differentiation or undifferentiated tissue), VDR expression was very low. In contrast, EGFR expression rose even further in undifferentiated tumors (Table I). When establishing primary cultures from human premalignant and malignant colonic tissue at diverse

stages of tumor progression, Tong et al. [40] have shown a mosaic pattern of VDR expression in colon cells. This demonstrated that a large fraction of cells isolated from human colon tumors, but not all, expressed the VDR and thus could respond to the genomic action of vitamin D compounds. Cross et al. [41] and Kállay et al. [42] demonstrated by immunoblotting and RT-PCR in colon tissue derived from 61 patients that there was indeed little VDR present in normal tissue, and that expression rose during colon tumor progression. In Fig. 1, VDR expression is shown in parallel with a proliferation marker, Ki-67. Barely any VDR expression is found in a (G3/4) tumor with cells at a low differentiation level (Fig. 1C), when it is was compared with expression in normal colon mucosa (Fig. 1A) or in a moderately-differentiated (G2) tumor (Fig. 1B). Low VDR expression in normal mucosa (Fig. 1A) is paralleled by low Ki-67 positivity mainly in the lower crypt area (Fig. 1D). With enhanced VDR expression in low grade cancer (Fig. 1B) there is also markedly increased Ki-67 positivity (Fig. 1E), whereas in high grade tumors with strong proliferation, there is almost no more VDR positivity apparent (Fig. 1C and F). These, and data from other laboratories, led to the suggestion that vitamin D receptor expression could be used as a predictive marker of biological behavior in human colorectal cancer [43]. The importance of the VDR for prevention of colonic hyperproliferation and of potential tumorigenesis was demonstrated by Kállay et al. [44] in the VDR knockout mouse model established by the group of Kato [45]. Complete loss of the VDR resulted in colonic hyperproliferation, cyclin D1 elevation, and

A

B

C

D

E

F

FIGURE 1 Evaluation of vitamin D receptor (A, B, C) and Ki-67 (D, E, F) expression in normal and malignant human colon. A, D: Normal colon mucosa, B, E: Moderately differentiated (G2) adenocarcinoma, and C, F: Adenocarcinoma of the colon at low differentiation (G3/4). (Cross and co-workers, unpublished data).

1712

HEIDE S. CROSS

a dramatic increase of DNA damage mainly in the distal colon (as measured by 8-hydroxy-2′-deoxyguanosine accumulation). This suggests that, at least in this animal model, growth control by 1,25-(OH)2-D3 is highly effective in the distal [44], but not in the proximal colon, and may be essential for maintenance of normal growth conditions of mucosal cells (see also Fig. 2). Recently, data from a population-based case-control study of colon cancer suggested that normal molecular variants of the VDR gene, i.e. polymorphisms of the VDR, might be related to the development of colon cancer [46]. The different variants could have less or more transcriptional activity upon binding of the hormonal ligand (see Chapter 68). Epidemiological studies to

A

Cyclin D1

200

CA CD

% of control

150

100 *

VDR +/+

B

VDR +/−

VDR −/−

8-OHdG

Densitometry units

15000 ** 10000

5000

0

FIGURE 2

B. Effect of 1,25-(OH)2-D3 and of Vitamin D Analogs on Proliferation, Differentiation, and Apoptosis of Colonic Cells The vitamin D receptor is known to exert its antimitotic and prodifferentiating effect by binding vitamin D metabolites and analogs. This not only maintains normal growth of the colonic crypt, but potentially can prevent the progression into premalignancy (see [42]). Interestingly, it has been observed recently [47] that there may be other paths of colon cancer prevention following VDR activation: the VDR can also bind the enteric carcinogen lithocholic acid, and can activate its detoxification via transcriptional induction of cytochrome P450 enzymes (see Chapter 53). The following sections will present, with some selectivity, experimental evidence for the effects of various vitamin D compounds on proliferation, differentiation, and apoptosis of colon cells in in vitro systems, in animal models, and in human studies. 1. IN VITRO MODELS

50

0

evaluate the association of these variants with diet and lifestyle factors are needed.

VDR +/+

VDR +/−

VDR −/−

A: Expression of Cyclin D1 protein by immunoblotting in the ascending (CA) and descending (CD) colon of wild-type (VDR+/+), heterozygous (VDR+/−), and VDR knockout (VDR−/−) mice. Values are expressed as mean ± SD, n = 5 animals per genotype. Statistically significant differences compared to the VDR+/+ group are indicated as * ( p < 0.05) (Student’s t test). B: Immunohistochemical evaluation of 8-OHdG expression in ascending (CA) and descending (CD) colon of wild-type (VDR+/+), heterozygous (VDR +/−), and VDR knockout (VDR−/−), mice. For quantification three 35 mm photo slides were taken randomly from each sample and were analyzed by NIH Image freeware. Values are expressed as mean ± SD, n = 5 animals of each genotype. Statistical significance was indicated as **(p< 0.01). (Student’s t test). 8-OHdG, 8-hydroxy2′-deoxyguanosine.

a. Action of 1,25-(OH)2-D3 In 1987 Lointier et al. [48] demonstrated that 10 nM 1,25-(OH)2-D3 inhibited growth of the LoVo colon cancer cell line under serumfree conditions. Brehier et al. [32], by evaluating induction of brush border hydrolase activity, demonstrated the differentiating effect of 1,25-(OH)2-D3 in HT-29 cells. Harper et al. [33] observed a decrease in the growth rate of HT-29 cells at the low concentration of 10 pM 1,25(OH)2-D3. Cross et al. [49–51] provided extensive evidence that 1–10 nM 1,25-(OH)2-D3 decreased growth and increased activity of the differentiation marker alkaline phosphatase in the human colon adenocarcinoma-derived cell line Caco-2. Induction of hyperproliferation of Caco-2 cells resulted in 100-fold higher sensitivity to the antimitotic prodifferentiating action of the secosteroid hormone [49]. Evidence for a dosedependent reduction of proliferation and increased alkaline phosphatase activity in Caco-2 cells was provided also by Halline et al. [52]. Responsiveness of primary cultures obtained from human normal colon, colon adenomas, and carcinomas to 1,25-(OH)2-D3 was demonstrated for the first time by Tong et al. in 1998 [40]. The proliferative rate of adenoma cells was initially twice that of cells obtained from normal mucosa. When adenoma cells were treated with 10 nM of the secosteroid, their mitotic rate was reduced to that of normal colonocytes in culture. Carcinoma-derived primary cultures responded to 1,25-(OH)2-D3 and vitamin D analogs in

1713

CHAPTER 95 Vitamin D and Colon Cancer

a concentration-dependent manner with respect to proliferation and differentiation. Since colorectal tumors are frequently under mitotic control by EGF or TGF-α, and during human colon tumor progression EGFR expression is up-regulated in parallel to that of the VDR [39] (see also Table I), Tong et al. [53] evaluated the interaction of EGF with the 1,25-(OH)2-D3. Their data demonstrated a mutual modulation of the VDR by the hormones, which resulted in enhanced activity of 1,25-(OH)2-D3 in EGFtreated Caco-2 cells. Bareis et al. [54] pointed out that only well-differentiated colonic cell lines or primary cultures were also able to respond to 10 nM 1,25-(OH)2-D3 by growth reduction. Their data showed that EGF treatment of a differentiated Caco-2 clone, but not of an undifferentiated one, increased VDR expression (see also Fig. 3). This suggested that VDR mediated growth inhibition by 1,25-(OH)2-D3 would be effective only in differentiated human colorectal carcinomas. Franceschi et al. [55] found that 1,25-(OH)2-D3 was able to stimulate fibronectin synthesis in colon cancer

A 0.0 nM

120

10.0 nM % of control

100

**

80 60 40 20 0 Caco-AQ

Caco-2/15

B Caco-2/AQ

Caco-2/15 EGF:

FIGURE 3

0.0 ng

10.0 ng

A: Evaluation of [3H]-thymidine incorporation into DNA in Caco-2/AQ and Caco-2/15 clones. Cells were treated for 48 hours with 10 nM 1,25-(OH)2-D3. Data are accumulated from two separate experiments and are presented in percent of ethanol control (mean ± SD; n = 6 per group; **: p < 0.01). B: Western blot analysis of VDR expression in Caco-2/AQ and Caco-2/15 cells with and without exposure to 10 ng EGF for 48 hours.

cell lines, indicating that the secosteroid may also play a restrictive role in tumor metastasis. Recently, Palmer et al. [56] demonstrated that only VDR–positive clones derived from the human colon carcinoma cell line SW480 were able to respond to the differentiating action of 1,25-(OH)2-D3. Conversely those clones that lacked the VDR did not respond. This vitamin D– induced differentiation process resulted in induction of E-cadherin and of other adhesion proteins, and promoted the translocation of β-catenin from the nucleus to the plasma membrane. 1,25-(OH)2-D3 also repressed β-catenin/TCF-4 transcriptional activity and modulated target genes in a manner opposite to that of βcatenin. Wilson et al. [57] demonstrated in HT-29 cells the significance of this effect with respect to regulation of the c-myc oncogene: elevated β-catenin/TCF signaling due to mutations in the adenomatous polyposis coli (APC) gene resulted in increased transcriptional activity of c-myc, one of the early immediate genes initiating cell cycle traverse. 1,25-(OH)2-D3 induced transcriptional blockage that resulted in decreased c-myc expression. TGF-β, a well recognized growth inhibitor of normal colonocytes, is no longer active in human colon cancer cells. However, 1,25-(OH)2-D3 treatment activated TGF-β signaling in Caco-2 cells, and enhanced abundance of the type 1 TGF-β receptor [58]. Thus, the secosteroid sensitizes Caco-2 cells to the growth-inhibitory action of TGF-β1. According to very recent data by Thompson et al. [59], transcriptional activity of the liganded VDR may result also in induction of cytochrome P450 detoxification enzymes, which may contribute to colonic chemoprotective mechanisms by detoxification of enteric carcinogens. Though several reports have been published on rapid and membrane based signal transduction mechanisms following exposure of colon cancer cells to 1,25-(OH)2-D3 (see [60–62]), none is of specific relevance yet for colon cancer prevention or therapy by 1,25-(OH)2-D3. b. Vitamin D Analogs It is becoming increasingly evident that adjuvant treatment of colorectal cancer patients with 1,25-(OH)2-D3 for its antimitotic prodifferentiating activity would necessitate a pharmacological dose, which would have the classical adverse consequences, namely hypercalcemia, soft tissue calcification, and nephrocalcinosis. Therefore, over the past decades, more than 400 analogs of vitamin D have been chemically synthesized and their biological properties have been systematically explored, calcemic effects have been quantified, and cell-differentiating and antimitotic potential have been evaluated (for reviews see [63,64] and Chapters 80–88). Action of some of these analogs on colon cancer cells will be reviewed here.

1714 Cross et al. [51] evaluated concentration-dependent growth inhibition in relation to hypercalcemic potential of two side chain-modified synthetic vitamin D analogs (Ro23-4319, Ro23-7553). Ro23-7553, a 16ene23yne side chain-modified vitamin D analog, was tenfold more effective than the 1,25-(OH)2-D3 in suppressing growth of Caco-2 cells; however, it was also tenfold more potent in stimulating calcium release from cultured mouse calvariae. With respect to intestinal calcium absorption the analogs were rather less effective than the parent hormone. Several other studies [65–69] evaluated only differential growth inhibition of colon cancer cells by a variety of analogs, and the main conclusion was that side chain-modified analogs improve activity of the secosteroid maximally by a factor 10, while their hypercalcemic activity is still moderately high. Only the 1β-(hydroxymethyl) congeners of the natural hormone, though still possessing significant antimitotic activity in spite of their low affinity for the VDR [70], did not promote osteoclast differentiation in vitro. This suggests that at nanomolar doses they would not cause hypercalcemia in human studies [71]. Oh et al. [72] implied that the antiproliferative activity of EB1089 in HT-29 cells was, at least in part, due to decreased secretion of IGF-II and increased concentrations of the IGF-II binding protein IGFBP-6. Tanaka et al. [73,74] observed enhanced differentiation of HT-29 cells upon treatment with a combination of sodium butyrate and vitamin D analogs and proposed this combination as a differentiation-based therapy for the clinical management of human colon cancer. Evaluation of Ro25-6760 action on HT-29 cells demonstrated significant growth inhibition, apoptosis induction, with enhanced expression of p21Waf1 and G1/Go cell cycle arrest [75]. An increase of the proapoptotic protein BAK induced by the vitamin D analog EB1089 in colon cell lines also suggested a mechanism of action involving apoptosis [76]. Another vitamin D analog was shown to increase expression of the cdk inhibitors p21Waf1 and of p27Kip1 independent of changes in pRB [77]. 2. ANIMAL MODELS FOR COLORECTAL TUMOR PREVENTION BY VITAMIN D COMPOUNDS

As early as 1988, Pence and Buddingh [78] evaluated the effect of 2000 IU vitamin D3/kg diet on 1,2-dimethylhydrazine-induced colon carcinogenesis in male rats. Development of cancer was promoted with 20% corn oil in the diet. Their data suggested that only in animals on a high fat diet (i.e. the promoter) did vitamin D3 significantly reduce tumor incidence. Subsequent work by Kawaura et al. [79], who instilled intrarectally lithocholic acid in rats with N-methyl-N-nitrosourea-induced colonic tumors, demonstrated that 1α-(OH)-D3 as well

HEIDE S. CROSS

was inhibiting promotion of tumorigenesis. Apparently, without exogenous promoters, vitamin D compounds did not affect colonic carcinogenesis and did not interfere with formation of bile acid profiles [80]. Comer et al. [81] also provided evidence that dietary levels between 250–10,000 IU vitamin D3/kg diet did not alter carcinogen-induced tumor incidence without prior promoter treatment, while work by Belleli et al. [82] using 400 ng 1,25-(OH)2-D3 per rat (an exceedingly high dose) suggested a protective effect of the secosteroid if it was delivered before the mutagen. When the protective action of vitamin D in combination with dietary calcium on carcinogen-induced colon tumors in rodents was investigated [83,84], it became apparent that both substances affected cellular kinetic indices, i.e., tumor size and not tumor incidence, and that their mode of action appeared to be a cooperative one. Newmark and Lipkin [85] introduced the concept of a nutritional stress diet for mice designed to represent the human Western diet, which is deficient in calcium and vitamin D, and rich in phosphate and fat. This diet led to hyperproliferation and hyperplasia in the rodent colon and, when fed long enough, to functional and structural modifications in the colon mucosa similar to those found in humans at increased risk for colonic neoplasia [86]. Mokady et al. [87] demonstrated that rats fed the Western stress diet had an enhanced response to tumor induction by a carcinogen, whereas supplementary vitamin D3 abrogated this tumor induction. (See also Chapter 91.) Several studies addressed the question of the efficacy of vitamin D analogs in rodent colon tumor models [88–91]. In these it was claimed that blood calcium levels generally were not raised by the analogs and that the development of aberrant crypt foci, a putative neoplastic lesion, or of carcinogen-induced tumors was significantly reduced by vitamin D compounds. The most pronounced inhibition was found if analogs were administered after carcinogen treatment. The wellknown vitamin D analog EB1089 was used in a human colon cancer LoVo cell xenograft study in a nude mouse model and led to 50% inhibition of tumor growth [92]. Another xenograft study with human colorectal cancer cells differing in VDR content demonstrated that tumor growth of VDR-positive cells was reduced in a concentration-dependent manner by 1,25-(OH)2-D3 and the hexafluorinated analog Ro25-6760, whereas growth of VDR-negative xenografts was not [93]. Tanaka et al. [94] used a novel analog, DD-003, to inhibit HT-29 human colon cancer cell growth under the renal capsule of immunodeficient mice. Interestingly, PKC isoform expression was significantly altered in precancerous lesions in rat colon after treatment with Ro24-5531 [95].

1715

CHAPTER 95 Vitamin D and Colon Cancer

While some of these vitamin D–derived compounds appear to be quite effective in reducing tumor size and tumor burden, hypercalcemia is still sometimes detected. The general consensus is clear: vitamin D analogs do inhibit colon carcinogenesis specifically when administered in the postinitiation phase by reducing colonocyte proliferation. A possible additional mechanism of action for these substances in rodent models might be inhibition of angiogenesis, as was demonstrated by Iseki et al. [96], and by inhibition of metastasis [97]. Recently, Wali et al. [98] demonstrated in rat colon that azoxymethane-induced aberrant crypt foci and tumors expressed enhanced levels of cyclin D1, of cyclooxygenase-2, and of inducible nitric oxide synthase, as well as reduced E-cadherin levels. These changes were significantly inhibited by a fluorinated vitamin D analog. Studies on tumor prevention by 1,25-(OH)2-D3 and its analogs were recently extended to the Apc(min) mouse by Huerta et al. [99]. They observed a significant decrease in total tumor burden while serum calcium in the group treated with analogs was only moderately elevated. These results suggest that such analogs may ultimately have utility as in humans chemopreventive agents at least in population groups at high risk for colon cancer, if serum calcium concentration is constantly monitored.

C. Clinical Studies In 1992, Thomas et al. [100,101] evaluated in vitro crypt cell production rate (CCPR) in rectal tissue obtained from familial adenomatous polyposis (FAP) patients. 1 µM–100 pM 1,25-(OH)2-D3, as well as the synthetic vitamin D analogs MC-903 and EB-1089, reduced CCPR in explant cultures by approximately 50%, whereas EGF increased CCPR by 100% [100]. This indicated for the first time that human colorectal tissue would indeed respond to 1,25-(OH)2-D3 and its analogs in a similar manner to cell and animal experiments. Accumulating experimental evidence from in vitro and from animal studies suggested that the analog EB-1089 was potentially useful for human treatment due to weaker calcemic side-effects while still maintaining high antimitotic, prodifferentiating, and apoptotic activity. In this respect, a promising phase I study with EB-1089 in patients with advanced colon and breast cancer was initiated [102]. An initial evaluation of this study [103] showed that hypercalcemia was still seen in patients receiving 17 microgram/day EB-1089 for 10 to 234 days. Hypercalcemia was reversible by discontinuing administration of the substance or by reducing the amount, and a tolerable dose for most patients was established at 7 microgram/day. There were no complete or partial responses, but 6 out

of 21 patients on treatment for more than 90 days showed stabilization of their disease. The only other human trial was performed on FAP patients who had previously undergone colectomy but had upper gastrointestinal polyps. In this double-blind randomized crossover trial, the effectiveness of sulindac, a specific inhibitor of cyclooxygenase-2, was compared with that of calcium in combination with calciferol. While sulindac treatment resulted in reduction of the crypt cell proliferation index in gastric epithelium but not in duodenal mucosa, calcium and calciferol had no effect whatsoever [104].

D. Conclusion Experimental results show quite clearly that, while 1,25-(OH)2-D3 and also some of its analogs indeed have antimitotic and prodifferentiating activity in colon cancer cells in vitro, their use in vivo, especially in the human patient, has not yet been well explored. While administration at low doses does not appear to be very effective, the high nanomolar concentrations needed to inhibit growth frequently are prohibitively hypercalcemic. Epidemiological data have not reliably supported the hypothesis that in humans serum 1,25-(OH)2-D3 at the highest physiological range showed a negative correlation with colorectal tumor incidence (see Section I.A.1). However, there is apparently a negative correlation between high levels of 25-(OH)-D3, the proliferative index of crypt cells [25], and colorectal cancer incidence. While this obviously favors the population group living at latitudes with high incident sunshine, also those consuming vitamin supplements could have higher levels of the precursor of the active metabolite. This precursor could conceivably be used by colon cells for extrarenal synthesis of 1,25-(OH)2-D3, which may function as an autocrine/paracrine cell cycle regulator in the colon. Evidence for this new concept of organlocalized accumulation of 1,25-(OH)2-D3, which would not influence its serum levels, will be discussed in the following sections.

III. VITAMIN D METABOLISM IN NORMAL AND NEOPLASTIC COLON CELLS A. Human Colon Cancer Cell Lines and Primary Cultures In 1990, Tomon et al. [105] were the first to demonstrate vitamin D catabolism in an in vitro intestinal cell

1716

HEIDE S. CROSS

model, the human colon adenocarcinoma-derived cell line Caco-2. These cells did not exhibit constitutive 25-(OH)-D3-24-hydroxylase (CYP24) activity. Catabolic activity was inducible upon treatment with 100 nM 1,25-(OH)2-D3. While, at this early date, only conjectures about extrarenal synthesis and degradation of 1,25-(OH)2-D3 existed, its role in growth control in the intestine was already well established. Early data by Birge et al. [106] demonstrated, that in rats dosed with vitamin D, mucosal cell turnover was accelerated and there was an approximately 20% increase in villus height, whereas in vitamin D–deficient rats, villus height was blunted. Thus, there is a trophic as well as a differentiating influence on intestinal morphology, which, at the time, was attributed to serum 1,25-(OH)2-D3 that had been synthesized in the kidney from the precursor 25-(OH)-D3. However, in retrospect this growth regulation could also have been attributed to extrarenal intestinal synthesis of 1,25-(OH)2-D3. It was only much later in 1997 when Cross et al. [107] demonstrated in Caco-2 cells the conversion of the precursor 25 OH D3 into 1,25-(OH)2-D3. They found constitutive expression of the 25-D3-1α-hydroxylase (CYP27B1) in almost any growth phase of this cell

C

[H3]25(OH)D3 metabolites (cpm)

[H3]25(OH)D3 metabolites (cpm)

A

line, and the sequential metabolism/catabolism of the secosteroid along the C-24 and C-23 oxidative pathways. It is therefore conceivable that human colon cells can control their growth via 1,25-(OH)2-D3 in an autocrine/ paracrine manner dependent upon presence of the vitamin D receptor. Bischof et al. [108] provided evidence that distinct oxidation pathways for 1,25-(OH)2-D3 catabolism were used by two Caco-2 clones differing in their level of differentiation. Colonic metabolism of vitamin D was subsequently evaluated in a variety of primary cultures, which were established from human colon tumors at different grades and stages. Such cultures are more closely related to human physiology and more appropriate to verify evidence obtained from cell lines. Figure 4 compares 25-(OH)-D3 metabolism in two different human colonic cancer cell types. From high performance liquid chromatography (HPLC) analysis, it was obvious that COGA-13 cells, which were isolated from a G2/3 human adenocarcinoma of the right colon, had no innate 1α-hydroxylase activity, which, however, was pronounced in Caco-2 cells (Fig. 4A and 4B). Caco-2 cells responded with reduced 1α-hydroxylation and enhanced 24-hydroxylation to 1,25-(OH)2-D3 treatment (Fig. 4 C), B

3000

3000

1,25(OH)2-D3

CYP24-Metabolites

2000

2000 1,25(OH)2-3-epi-D3

1000

1000

0

0 5

10

15

20

25

30

5

10

15

20

25

30

5

10

15

20

25

30

D 3000

3000 CYP24-Metabolites 2000

2000

1,25(OH)2-D3

1000

1000

0

0 5

10

15

20

Elution time (min)

25

30

Elution time (min)

FIGURE 4 Comparison of HPLC profiles of 25-(OH)-D3 metabolism in human colonic cancer cells. Cells were incubated with the precursor 25-(OH)-[26,27-methyl-3H]-D3. The 1α-hydroxylated compounds appear after 20–25 minutes elution time, whereas the 24-hydroxylated metabolites are seen much earlier between 5–10 minutes elution time, just after the precursor peak. A: Untreated Caco-2 cells have mainly 1α-hydroxylated metabolite production and almost no 24-hydroxylase activity. B: Untreated COGA-13 cells have constitutively high 24-hydroxylase activity and no 1α-hydroxylation. C: In Caco-2 cells treatment with 10 nM 1,25-(OH)2-D3 results in significant down-regulation of 1α-hydroxylase activity and significant increase of 24-hydroxylation. D: In COGA-13 cells, the same treatment yields no detectable effects on 25-(OH)-D3 metabolism (Cross and co-workers, unpublished data).

1717

CHAPTER 95 Vitamin D and Colon Cancer

whereas COGA-13 cells were insensitive, probably because their maximal 24-hydroxylation capacity had already been reached (Fig. 4D) (see also Section III.A.1). These results support the hypothesis that it could be the degree of differentiation of cells that determines their ability for 1,25-(OH)2-D3 synthesis or catabolism (see also [54]). Even more direct evidence for extrarenal vitamin D metabolism present in cancer patient-derived colon tissue was also provided. Cells were isolated from freshly excised human colon tumors, as well as from the adjacent normal mucosa outside the tumor border. After testing cells for general metabolic activity, they were immediately processed for HPLC assays. Bareis et al. [109] demonstrated unequivocally with this method that in human colon cancer cells there is very active vitamin D metabolism including both 24-hydroxylase and 1α-hydroxylase activity, whereas in the adjacent normal mucosa there is very little of either. Bareis and co-workers also demonstrated by RT-PCR analysis in human colon tumor-derived primary cultures low CYP27B1 mRNA and high CYP24 mRNA expression. The latter, however, was not only present as the wild-type transcript, but also as a splice variant. The relationship between splice variants and enzymatic activity is as yet unknown [109]. There is accumulating evidence that human colon cancer cells express varying levels of the metabolic and catabolic vitamin D hydroxylases, supporting the hypothesis that the human colon tissue has differential capacity to accumulate the active hormonal metabolite. In addition, there are apparently unique regulatory processes that depend upon difficult to define levels of differentiation (see also Chapter 79). 1. REGULATION OF VITAMIN D METABOLISM CATABOLISM BY 1,25-(OH)2-D3 AND EPIDERMAL GROWTH FACTOR IN VITRO

of their 24-hydroxylation pattern after treatment with the active vitamin D metabolite (Fig. 4D). Thus, there is, on the one hand, constitutive expression of this catabolic enzyme in cells derived from a colon tumor at a low differentiation level. On the other hand, in Caco-2 cells even under vitamin D treatment, there is still 1α-hydroxylation though reduced. However, there is also 24-hydroxylating activity. This clearly demonstrates the potential problem of hormone degradation, which might exist in colon tissue following localized synthesis of 1,25-(OH)2-D3. Two different clones derived from the Caco-2 cell line were analyzed (Caco-2/AQ with high proliferation, lower differentiation; Caco-2/15 with low proliferation, high differentiation). When their CYP27B1 expression levels after treatment with 10 nM 1,25-(OH)2-D3 were compared, it became obvious that in Caco-2/AQ cells not only activity but also protein (and mRNA) level of CYP27B1 was down-regulated, whereas in Caco-2/15 cells it was increased [54]. Such up-regulated expression of CYP27B1 appeared to be typical for well-differentiated cell lines, and again suggests the primary importance of the biological grade of cells, which would determine synthesis and degradation of 1,25(OH)2-D3 in the human cancerous colon mucosa under physiological conditions. Regulation of vitamin D metabolic and catabolic enzymes was also seen with EGF [54]. This latter observation may be of some physiological importance for colonic synthesis of vitamin D, since expression of the EGFR is well recognized to increase during colon tumor progression (see Table I and [39]) and transforming growth factor α (TGF-α), also a ligand for the EGFR, is an autocrine growth factor during tumorigenesis. Moreover, as described above in Section II.A.1.a, there is mutual regulation between the VDR and the EGFR in colon cancer cell lines (see also [110]).

AND

If cells in the human colon synthesize and catabolize 1,25-(OH)2-D3 to different extents, this may be due to regulatory factors present in the cellular environment. The active hormone 1,25-(OH)2-D3 could itself be present due to local synthesis, and this could result in regulation similar to the well known pathway in the kidney, i.e. down-regulation of 1α-hydroxylase activity and up-regulation of 24-hydroxylase activity. Consequently, treatment of Caco-2 cells for 48 hours with 10 nM 1,25-(OH)2-D3 resulted in induced 24-hydroxylation and in considerable reduction of the 1α-hydroxylated peak, as measured by HPLC analysis (Fig. 4C). COGA-13 cells isolated from a high-grade colon adenocarcinoma, while expressing CYP24 strongly in the control, did not show any modulation

B. Expression of CYP27B1 and CYP24 in Human Intestine There is increasing evidence that during tumor progression differential cellular regulation of vitamin D metabolism and catabolism occurs possibly similar to the regulation demonstrated by Cross et al. [41] in colon cancer patients with respect to vitamin D receptor expression: VDR is elevated at the mRNA level very early during progression, i.e. in differentiated tumors, while late during progression VDR expression is significantly reduced [38,41]. Such a regulatory pattern points towards a physiological defense mechanism against tumor progression, which may fail during late stages.

1718 1. EVALUATION OF MRNA EXPRESSION CYP27B1 AND CYP24

FOR

Evaluation of tissue specimens from 50 colorectal adenocarcinomas by RT-PCR demonstrated convincingly that CYP27B1 mRNA was elevated in G1 and G2 tumors when compared with adjacent normal mucosa from the same patient, and also in comparison with colon mucosa from noncancer patients. In G3 tumors expression dropped to low levels [41]. This seemed to be true only for early stages of the disease (pT1-pT3), during late stage (pT4) disease both G2 and G3 tumors had low CYP27B1 mRNA expression [109]. Tangpricha et al. [111] studied CYP27B1 mRNA expression by real-time PCR in normal colon and colon tumors, though a quantitative evaluation was not possible due to the small number of patients. Recently, Ogunkolade et al. [112] showed CYP27B1 expression in colonic tissue also by real-time RT-PCR in a larger number of individuals. They did not confirm the increase of CYP27B1 in tumors compared with healthy colon samples from noncancer individuals described by Cross et al. [41] and Bareis et al. [109]. It has to be pointed out, however, that Ogunkolade et al. did not examine colonic tumors with respect to the biological grade of cells as was done in other studies [41,109]. Cross et al. [113] provided very recently the evaluation of CYP27B1 mRNA by real-time PCR in colon tumors and adjacent mucosa from 18 cancer patients, as well as in colon mucosa from 5 noncancer patients. All tumor patients had high to medium differentiated (G1, G2) primary adenocarcinomas. The authors clearly demonstrated that the normal mucosa from tumor as well as from nontumor patients had similarly low levels of CYP27B1 mRNA, whereas this was consistently increased in tumor tissue. The discrepancies in results between different laboratories [41,109,113], and [111,112], are most likely caused by the fact, that (1) G3 tumors frequently have very low expression of CYP27B1 mRNA, and (2) that sometimes the adjacent “normal” mucosa of a G3 tumor displays high expression of CYP27B1 mRNA similar to that found in early colon tumors [42]. Bareis et al. [109] detected a transcript of CYP24 in human cancerous colon lesions as well as in the adjacent mucosa of the same patient. While CYP24 was consistently higher in tumor tissue than in adjacent normal tissue, they also found at least two transcripts of differing size, where the larger one contained an additional sequence with homology to intron 1. Expression levels of the smaller transcript appeared to be highest in late-stage high-grade tumors, whereas the larger one was present in low-grade highly-differentiated tumor tissue.

HEIDE S. CROSS

2. CYP27B1 PROTEIN EXPRESSION IN HUMAN COLON

THE

Zehnder et al. [114] were the first to demonstrate by immunohistochemistry and immunoblotting extrarenal expression of the CYP27B1. Among the tissues investigated was also the human colon where staining was found in epithelial cells and apparently also in parasympathetic ganglia. They suggested that the discrete pattern of staining found in various organs in the human body emphasized a possible role for the hydroxylase as an intracrine modulator of vitamin D function in peripheral tissues. When Cross and co-workers evaluated CYP27B1 protein expression in colon tissue samples (normal mucosa from noncancer patients, adenomas, low- and high-grade tumors and adjacent mucosa from the same patient) from 38 patients by immunofluorescence, they found that at least 50% of tumor patients were positive. This positivity depended completely upon histology of the tissue: As long as there were glandular differentiated structures present, these were positive for CYP27B1, even if the rest of the tissue consisted of cells of low differentiation and was negative for CYP27B1. Fig. 5A shows normal human colon mucosa with barely any positivity. Fig. 5B exemplifies strongly enhanced expression of CYP27B1 in a colon adenoma, while Fig. 5C shows strong expression in the small intestine (ileum). a. Coexpression of CYP27B1, VDR, and Ki-67 Proteins in Human Colon Further evaluation of human colon tissue sections from cancer patients by immunofluorescence showed that many cells positive for the VDR were also positive for CYP27B1 as recently shown by Bises et al. [115]. However, there were also many cells only positive for the VDR, especially in normal and premalignant tissues. This again points to an autocrine/paracrine mode of action for the secosteroid. While the VDR is obviously present also in normal mucosa where it may regulate mucosal function and growth with 1,25-(OH)2-D3 as ligand, it is only during onset of malignant progression that CYP27B1 expression is strongly induced in more than 50% of patients. While the VDR frequently is found also in proliferating cells (there is coexpression with the Ki-67 antigen, see Fig. 1), CYP27B1 positivity is rarely found in Ki-67positive cells (Cross et al., unpublished observations). While CYP27B1 positivity is barely apparent by immunofluorescence in the normal colonic mucosa, it is interesting to observe that in small intestine (ileum) much more positivity is present (Fig. 5C). This observation suggests that in the small intestine there is an innate defense against hyperproliferation due to high

1719

CHAPTER 95 Vitamin D and Colon Cancer

B

A

C

FIGURE 5

Immunofluorescence analysis of CYP27B1 protein in normal human colon mucosa (A), colon adenoma (B), and small intestine (C). (Cross and co-workers, unpublished data).

local synthesis of the 1,25-(OH)2-D3. This could, at least partially, be responsible for the low tumor incidence occurring in the human small intestine.

C. CYP27B1, CYP24, and VDR Expression in a Mouse Model Kállay et al. [44] demonstrated in a mouse model that VDR expression was higher in the proximal than in the distal colon. Their results from a VDR-knockout (VDR-KO) mouse showed that it was mainly the distal colon that was negatively affected by the lack of genomic action of 1,25-(OH)2-D3 [44]. However, CYP27B1 mRNA expression in proximal and distal mouse colon seemed quite similar, whereas CYP24 expression was much higher in the proximal than in the distal colon [116]. This implies that in the proximal colon less active vitamin D metabolite would be potentially available. Recently Kállay et al. [117] demonstrated for the first time by RT-PCR that the regulation of CYP27B1 and CYP24 mRNA in the mouse colon is completely different from that in the kidney. In the VDR-KO mouse kidney, CYP27B1 mRNA expression is highly induced due to lack of the VDR and consequent ineffectiveness of 1,25-(OH)2-D3 action, while, in the colon CYP27B1 expression is significantly downregulated in parallel with enhanced proliferation (as shown by increased PCNA expression) (Fig. 6A, B). These very recent results demonstrate again the unique aspects of extrarenal 1,25-(OH)2-D3 synthesis, the completely different regulation in the colon in contrast to the one in the kidney, and also the potential importance this could have for colorectal cancer prevention.

IV. NUTRITIONAL REGULATION OF CYP27B1 AND CYP24 This new concept of extrarenal synthesis and catabolism of vitamin D in colonocytes should gain increasing importance for tumor prevention as well as for therapy, since 1,25-(OH)2-D3 could act locally and prevent hyperproliferation and dedifferentiation without causing generalized hypercalcemia. However, some cells could respond to local accumulation of 1,25-(OH)2-D3 with increased activity of CYP24 and decreased activity A Kidney

Colon

B

VDR +/+

FIGURE 6

VDR −/−

A: Comparison of the CYP27B1 status in the kidney and colon of wild-type (VDR+/+) and knockout (VDR−/−) mice. B: Expression of the proliferation marker PCNA in the colon of wild-type (VDR+/+) and knockout (VDR−/−) mice.

1720 of CYP27B1. Therefore, the regulation of these colonic enzymes by, for instance, nutritional means to improve localized accumulation of 1,25-(OH)2-D3 might be of some importance for cancer prevention and also for therapy of low-grade, early-stage tumors. Considerable physiological evidence is accumulating for a protective effect of estrogenic substances against colorectal cancer incidence. At all ages, women are less likely than men to develop colon cancer, and postmenopausal hormone replacement therapy even further reduces colon cancer risk by up to 25%. Potter et al. [118] demonstrated lower risk of adenomatous polyps of the large bowel with hormone replacement therapy. In addition, in several colon cancer animal models, male rodents were shown to have higher tumor burden and increased aberrant crypt formation rates (see [119]), the latter being a typical precursor lesion of colorectal cancer. Very recently a large comprehensive study by the Women’s Health Initiative Investigators [120] on physiological effects of hormone replacement therapy (HRT) was stopped, since most parameters that were assumed to be beneficially affected by HRT, were either negatively (for instance, breast cancer) or not at all affected. The only highly significant exceptions were reduction in incidence of colorectal cancer and of osteoporosis. Thus, the reluctance to use HRT for a longer period and for minor problems seems to be well founded. However, the observed reduction of colorectal cancer incidence needs to be explored further. The mechanism of action could conceivably involve modulation of the vitamin D system. In the clinical situation, for instance, plasma 1,25(OH)2-D3 levels are elevated during human pregnancy and remain high postpartum in lactating women [120, 121] even beyond that component explained by increased DBP (see Chapter 51). In animal studies, female rats treated with estradiol benzoate daily for eight days had increased 1,25-(OH)2-D3 concentrations in plasma, gut mucosa, and kidneys [123]. Interactions between vitamin D and estrogen have also been observed in murine colon carcinoma [124]. However, evidence that estrogenic substances indeed regulate vitamin D metabolism was still missing.

A. Regulation of 1,25-(OH)2-D3 Synthesis by Phytoestrogens Dramatically reduced incidence, especially of hormone-related cancers such as mammary and prostate tumors, has been linked to the consumption of a typical Asian diet, which contains high amounts of soy products and thus is rich in phytoestrogens. It is conceivable that these substances, through their potential to

HEIDE S. CROSS

act as selective estrogen receptor modulators, could have an effect on vitamin D–related inhibition of tumor growth in the mammary and prostate gland. Interestingly, isoflavones belonging to the group of phytoestrogens have been shown to down-regulate estrogen receptor (ER) expression, which could lead to reduced estrogenic responses, i.e. protection against deleterious sex hormone effects in hormone-responsive tissues [125]. Foley and co-workers [126] suggested that malignant transformation of the human colon is associated with a marked diminution of ER-β expression, which is widely regarded to be the predominant ER-subtype in normal colonic tissue [127]. Phytoestrogens bind with high affinity to ER-β. This could indicate a possible protective mode of action for phytoestrogens in the colon, even though this requires further experimental evaluation. Kállay et al. [128] demonstrated in a mouse model that soy feeding elevated CYP27B1 mRNA expression both in the proximal and distal sections of the colon. In contrast, CYP24 mRNA expression was considerably reduced by soy consumption. When they administered genistein, a well-known isoflavone, by gavage to mice, a similar effect was observed (Table II). This suggests that genistein, a major constituent of soy, is one of the nutritional substances that could be used to modulate vitamin D metabolism and catabolism. These results imply potentially enhanced colonic synthesis of 1,25-(OH)2-D3 in populations on high nutritional soy consumption, and conceivably enhanced protection against colorectal cancer. While, in breast and prostate tissue, phytoestrogens contained in soy may very well interfere with the action of estrogen itself, they could also stimulate extrarenal vitamin D synthesis

TABLE II Expression of CYP27B1 and CYP24 mRNA in the Murine Colon CYP27B1 Right colon AIN 76A diet Phytoestrogen diet

242 ± 176 468 ± 135*

Left colon

CYP24 Right colon

Left colon

148 ± 76 111 ± 22 33 ± 9.4 460 ± 182* 23 ± 11* 27 ± 6.7

Multiplex RT-PCR, i.e. simultaneous amplification of transcripts specific either for CYP27B1 or for CYP24, and a transcript specific for the epithelial cell marker cytokeratin 8, was carried out for semi-quantitative evaluation of respective mRNA expression levels. PCR products were analyzed on agarose gels. The level of CYP27B1 and CYP24 expression was correlated with that of the epithelial cell marker CK 8. Values are expressed as mean ± SD, n = 8 animals in each group. Statistical significance is indicated as * ( p < 0.05).

1721

CHAPTER 95 Vitamin D and Colon Cancer

in such tissues, enabling the steroid hormone to exert its well recognized antimitotic, prodifferentiating action [129].

Acknowledgment I thank Professor Meinrad Peterlik, PhD., MD, for invaluable suggestions and criticisms, and Dr. Enikö Kállay for critically reviewing the manuscript.

V. CONCLUSION While epidemiology has provided strong support for the concept, that vitamin D could be a prevention factor during colon tumorigenesis, it was only the serum level of the precursor 25-(OH)-D3 and not 1,25-(OH)2-D3 itself that correlated convincingly with human colorectal tumor incidence. While 1,25-(OH)2-D3 is known to prevent proliferation and to induce differentiation and apoptosis in colonocytes, pharmacological doses are necessary, regardless whether it is tested in an in vitro or an in vivo system. Such high concentrations, however, are prohibitively hypercalcemic. While some analogs may maintain their effectiveness as cell cycle regulators at 10–100-fold lower doses than the parent compound, even these concentrations could result in hypercalcemia in patients. It therefore appears unlikely that any of these compounds will ever be used for preventive purposes. However, in late stage colon cancers, they may very well prove to stabilize the disease. Recent data have demonstrated extrarenal synthesis of the secosteroid in the colon. Physiological regulation of vitamin D metabolic and catabolic hydroxylases in normal and malignant human colonic tissue suggests a role for the locally accumulated hormone in prevention of tumor progression. In addition, during low-grade earlystage malignancy, colonic synthesis of 1,25-(OH)2-D3 could potentially provide a block to progression, if 1,25-(OH)2-D3 catabolism could be inhibited. Renal or colonic 1,25-(OH)2-D3 synthesis and catabolism is differentially regulated. While lack of vitamin D action due to absence of the VDR results in elevated expression of CYP27B1 in the kidney, the same enzyme is down-regulated in the colon, probably due to enhanced proliferation of the tissue. This implies that there may exist substances that could enhance extrarenal 1,25-(OH)2-D3 accumulation without affecting renal synthesis. Recent results allude to such action mechanisms for phytoestrogens. Genistein, a phytoestrogen and major component of soy, can induce expression of CYP27B1 and reduce that of CYP24 in the mouse colon. The involvement of phytoestrogens in the regulation of the vitamin D system could conceivably explain the observation that women have less colorectal cancer incidence than men, probably due to their higher estrogenic background. Consumption of phytoestrogens, however, could also be acceptable for men to protect against colorectal cancer incidence.

References 1. Garland CF, Garland FC 1980 Do sunlight and vitamin D reduce the likelihood of colon cancer? Int J Epidemiol 9: 227–231. 2. Tangpricha V, Pearce EN, Chen TC, Holick MF 2002 Vitamin D insufficiency among free-living healthy young adults. Am J Med 112:659–662. 3. Nesby-O’Dell S, Scanlon KS, Cogswell ME, Gillespie C, Hollis BW, Looker AC, Allen C, Doughertly C, Gunter EW, Bowman BA 2002 Hypovitaminosis D prevalence and determinants among African-American and white women of reproductive age: third National Health and Nutrition Examination Survey, 1988–1994. Am J Clin Nutr 76:187–192. 4. Anonymous 1997 Food, nutrition and the prevention of cancer: a global perspective. World Cancer Research Fund and American Institute for Cancer Research, Washington, DC 1997. 5. Freedman DM, Dosemeci M, McGlynn K 2002 Sunlight and mortality from breast, ovarian, colon, prostate, and nonmelanoma skin cancer: a composite death certificate based case-control study. Occup Environ Med 59:257–262. 6. Emerson JC, Weiss NS 1992 Colorectal cancer and solar radiation. Cancer Causes Control 3:95–99. 7. Grant WB 2002 An estimate of premature cancer mortality in the U.S. due to inadequate doses of solar ultraviolet-B radiation. Cancer 94(6):1867–1875. 8. Garland C, Shekelle RB, Barrett-Connor E, Criqui MH, Rossof AH, Paul O 1985 Dietary vitamin D and calcium and risk of colorectal cancer: a 19-year prospective study in men. Lancet 1:307–309. 9. Heilbrun LK, Nomura A, Hankin JH, Stemmermann GN 1985 Dietary vitamin D and calcium and risk of colorectal cancer. Lancet 1:925. 10. Garland CF, Comstock GW, Garland FC, Helsing KJ, Shaw EK, Gorham ED 1989 Serum 25-hydroxyvitamin D and colon cancer: eight-year prospective study. Lancet 2:1176–1178. 11. Garland CF, Garland FC, Gorham ED 1991 Can colon cancer incidence and death rates be reduced with calcium and vitamin D? Am J Clin Nutr 54:193S–201S. 12. Braun MM, Helzlsouer KJ, Hollis BW, Comstock GW 1995 Colon cancer and serum vitamin D metabolite levels 10–17 years prior to diagnosis. Am J Epidemiol 142:608–611. 13. Glass AR, Kikendall JW, Sobin LH, Bowen PE 1993 Serum 25-hydroxyvitamin D concentrations in colonic neoplasia. Horm Metab Res 25:397–398. 14. Tangrea J, Helzlsouer K, Pietinen P, Taylor P, Hollis B, Virtamo J, Albanes D 1997 Serum levels of vitamin D metabolites and the subsequent risk of colon and rectal cancer in Finnish men. Cancer Causes Control 8:615–625. 15. Platz EA, Hankinson SE, Hollis BW, Colditz GA, Hunter DJ, Speizer FE, Giovannucci E 2000 Plasma 1,25-dihydroxyand 25-hydroxyvitamin D and adenomatous polyps of the distal colorectum. Cancer Epidemiol Biomarkers Prev 9:1059–1065.

1722 16. Niv Y, Sperber AD, Figer A, Igael D, Shany S, Fraser G, Schwartz B 1999 In colorectal carcinoma patients, serum vitamin D levels vary according to stage of the carcinoma. Cancer 86:391–397. 17. Bostick RM, Potter JD, Sellers TA, McKenzie DR, Kushi LH, Folsom AR 1993 Relation of calcium, vitamin D, and dairy food intake to incidence of colon cancer among older women. The Iowa Women’s Health Study. Am J Epidemiol 137:1302–1317. 18. Kampman E, Giovannucci E, van ‘t Veer P, Rimm E, Stampfer MJ, Colditz GA, Kok FJ, Willett WC 1994 Calcium, vitamin D, dairy foods, and the occurrence of colorectal adenomas among men and women in two prospective studies. Am J Epidemiol 139:16–29. 19. Kearney J, Giovannucci E, Rimm EB, Ascherio A, Stampfer MJ, Colditz GA, Wing A, Kampman E, Willett WC 1996 Calcium, vitamin D, and dairy foods and the occurrence of colon cancer in men. Am J Epidemiol 143:907–917. 20. Marcus PM, Newcomb PA 1998 The association of calcium and vitamin D, and colon and rectal cancer in Wisconsin women. Int J Epidemiol 27:788–793. 21. Anonymous 1998 Design of the Women’s Health Initiative clinical trial and observational study. Control Clin Trials 19:61–109. 22. Kampman E, Slattery ML, Caan B, Potter JD 2000 Calcium, vitamin D, sunshine exposure, dairy products and colon cancer risk (United States). Cancer Causes Control 11:459–466. 23. Jarvinen R, Knekt P, Hakulinen T, Aromaa A 2001 Prospective study on milk products, calcium and cancers of the colon and rectum. Eur J Clin Nutr 55:1000–1007. 24. Garland CF, Garland FC, Gorham ED 1999 Calcium and vitamin D. Their potential roles in colon and breast cancer prevention. Ann NY Acad Sci 889:107–119. 25. Holt PR, Arber N, Halmos B, Forde K, Kissileff H, McGlynn KA, Moss SF, Kurihara N, Fan K, Yang K, Lipkin M 2002 Colonic epithelial cell proliferation decreases with increasing levels of serum 25-hydroxy vitamin D. Cancer Epidemiol Biomarkers Prev 11:113–119. 26. Hsieh JC WG, Oza AK, Dang HT, Price JN, Galligan MA, Jurutka PW, Thompson PD, Haussler CA, Haussler MR 1999 Characterization of unique DNA-binding and transcriptionalactivation functions in the carboxyl-terminal extension of the zinc finger region in the human vitamin D receptor. Biochemistry 38(49):16347–16358. 27. Norman AW, Bishop, J E, Bula CM, Olivera CJ, Mizwicki MT, Zanello L P, Ishida H, Okamura W H 2002 Molecular tools for study of genomic and rapid signal transduction responses initiated by 1α,25(OH)2-vitamin D3. Steroids 67(6): 457–466. 28. Erben RG, Soegiarto DW, Weber K, Zeitz U, Lieberherr M, Gniadecki R, Moller G, Adamski J, Balling R 2002 Deletion of deoxyribonucleic acid binding domain of the vitamin D receptor abrogates genomic and nongenomic functions of vitamin D. Mol Endocrinol 16:1524–1537. 29. Frampton RJ, Suva LJ, Eisman JA, Findlay DM, Moore GE, Moseley JM, Martin TJ 1982 Presence of 1,25-dihydroxyvitamin D3 receptors in established human cancer cell lines in culture. Cancer Res 42:1116–1119. 30. Giuliano AR, Franceschi RT, Wood RJ 1991 Characterization of the vitamin D receptor from the Caco-2 human colon carcinoma cell line: effect of cellular differentiation. Arch Biochem Biophys 285:261–269. 31. Zhao X, Feldman D 1993 Regulation of vitamin D receptor abundance and responsiveness during differentiation of

HEIDE S. CROSS

32.

33. 34.

35.

36.

37. 38.

39.

40.

41.

42.

43.

44.

45.

HT-29 human colon cancer cells. Endocrinology. 132: 1808–1814. Brehier A, Thomasset M 1988 Human colon cell line HT-29: characterization of 1,25-dihydroxyvitamin D3 receptor and induction of differentiation by the hormone. J Steroid Biochem 29:265–270. Harper KD, Iozzo RV, Haddad JG 1989 Receptors for and bioresponses to 1,25-dihydroxyvitamin D in a human colon carcinoma cell line (HT-29). Metabolism 38:1062–1069. Cross HS, Hulla W, Tong WM, Peterlik M 1995 Growth regulation of human colon adenocarcinoma-derived cells by calcium, vitamin D and epidermal growth factor. J Nutr 125:2004S–2008S. Shabahang M, Buras RR, Davoodi F, Schumaker LM, Nauta RJ, Evans SR 1993 1,25-Dihydroxyvitamin D3 receptor as a marker of human colon carcinoma cell line differentiation and growth inhibition. Cancer Res 53:3712–3718. Lointier P, Meggouh F, Pezet D, Dapoigny M, Dieng PN, Saez S, Chipponi J 1989 Specific receptors for 1,25-dihydroxyvitamin D3 (1,25-DR) and human colorectal carcinogenesis. Anticancer Res 9:1921–1924. Vandewalle B, Adenis A, Hornez L, Revillion F, Lefebvre J 1994 1,25-dihydroxyvitamin D3 receptors in normal and malignant human colorectal tissues. Cancer Lett 86:67–73. Cross HS, Bajna E, Bises G, Genser D, Kallay E, Potzi R, Wenzl E, Wrba F, Roka R, Peterlik M (eds.) 1996 Vitamin D receptor and cytokeratin expression may be progression indicators in human colon cancer. Anticancer Res Vol. 16. Sheinin Y, Kaserer K, Wrba F, Wenzl E, Kriwanek S, Peterlik M, Cross HS 2000 In situ mRNA hybridization analysis and immunolocalization of the vitamin D receptor in normal and carcinomatous human colonic mucosa: relation to epidermal growth factor receptor expression. Virchows Arch 437:501–507. Tong WM, Bises G, Sheinin Y, Ellinger A, Genser D, Potzi R, Wrba F, Wenzl E, Roka R, Neuhold N, Peterlik M, Cross HS 1998 Establishment of primary cultures from human colonic tissue during tumor progression: vitamin-D responses and vitamin-D-receptor expression. Int J Cancer 75:467–472. Cross HS, Bareis P, Hofer H, Bischof MG, Bajna E, Kriwanek S, Bonner E, Peterlik M 2001 25-Hydroxyvitamin D3-1alpha-hydroxylase and vitamin D receptor gene expression in human colonic mucosa is elevated during early cancerogenesis. Steroids 66:287–292. Kallay E, Bareis P, Bajna E, Kriwanek S, Bonner E, Toyokuni S, Cross HS 2002 Vitamin D receptor activity and prevention of colonic hyperproliferation and oxidative stress. Food Chem Toxicol 40:1191–1196. Evans SR, Nolla J, Hanfelt J, Shabahang M, Nauta RJ, Shchepotin IB 1998 Vitamin D receptor expression as a predictive marker of biological behavior in human colorectal cancer. Clin Cancer Res 4:1591–1595. Kallay E, Pietschmann P, Toyokuni S, Bajna E, Hahn P, Mazzucco K, Bieglmayer C, Kato S, Cross HS 2001 Characterization of a vitamin D receptor knockout mouse as a model of colorectal hyperproliferation and DNA damage. Carcinogenesis 22:1429–1435. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Genet 16:391–396.

CHAPTER 95 Vitamin D and Colon Cancer

46. Slatter ML, Yakumo K, Hoffman M, Neuhausen S 2001 Variants of the VDR gene and risk of colon cancer (United States). Cancer Causes Control 12:359–364. 47. Makishima M, Lu TT, Xie W, Whitfield GK, Domoto H, Evans RM, Haussler MR, Mangelsdorf DJ 2002 Vitamin D receptor as an intestinal bile acid sensor. Science 296: 1313–1316. 48. Lointier P, Wargovich MJ, Saez S, Levin B, Wildrick DM, Boman BM 1987 The role of vitamin D3 in the proliferation of a human colon cancer cell line in vitro. Anticancer Res 7:817–821. 49. Cross HS, Huber C, Peterlik M 1991 Antiproliferative effect of 1,25-dihydroxyvitamin D3 and its analogs on human colon adenocarcinoma cells (CaCo-2): influence of extracellular calcium. Biochem Biophys Res Commun 179:57–62. 50. Cross HS, Pavelka M, Slavik J, Peterlik M 1992 Growth control of human colon cancer cells by vitamin D and calcium in vitro. J Natl Cancer Inst 84:1355–1357. 51. Cross HS, Farsoudi KH, Peterlik M 1993 Growth inhibition of human colon adenocarcinoma-derived Caco-2 cells by 1,25-dihydroxyvitamin D3 and two synthetic analogs: relation to in vitro hypercalcemic potential. Naunyn Schmiedebergs Arch Pharmacol 347:105–110. 52. Halline AG, Davidson NO, Skarosi SF, Sitrin MD, Tietze C, Alpers DH, Brasitus TA 1994 Effects of 1,25-dihydroxyvitamin D3 on proliferation and differentiation of Caco-2 cells. Endocrinology 134:1710–1717. 53. Tong WM, Kallay E, Hofer H, Hulla W, Manhardt T, Peterlik M, Cross HS 1998 Growth regulation of human colon cancer cells by epidermal growth factor and 1,25-dihydroxyvitamin D3 is mediated by mutual modulation of receptor expression. Eur J Cancer 34:2119–2125. 54. Bareis P, Kallay E, Bischof MG, Bises G, Hofer H, Potzi C, Manhardt T, Bland R, Cross HS 2002 Clonal differences in expression of 25-hydroxyvitamin D3-1α-hydroxylase, of 25-hydroxyvitamin D3-24-hydroxylase, and of the vitamin D receptor in human colon carcinoma cells: effects of epidermal growth factor and 1α,25-dihydroxyvitamin D3. Exp Cell Res 276:320–327. 55. Franceschi RT, Linson CJ, Peter TC, Romano PR 1987 Regulation of cellular adhesion and fibronectin synthesis by 1α,25-dihydroxyvitamin D3. J Biol Chem 262: 4165–4171. 56. Palmer HG, Gonzalez-Sancho JM, Espada J, Berciano MT, Puig I, Baulida J, Quintanilla M, Cano A, de Herreros AG, Lafarga M, Munoz A 2001 Vitamin D3 promotes the differentiation of colon carcinoma cells by the induction of E-cadherin and the inhibition of beta-catenin signaling. J Cell Biol 154:369–387. 57. Wilson AJ, Velcich A, Arango D, Kurland AR, Shenoy SM, Pezo RC, Levsky JM, Singer RH, Augenlicht LH 2002 Novel detection and differential utilization of a c-myc transcriptional block in colon cancer chemoprevention. Cancer Res 62:6006–6010. 58. Chen A, Davis BH, Sitrin MD, Brasitus TA, Bissonnette M 2002 Transforming growth factor-beta 1 signaling contributes to Caco-2 cell growth inhibition induced by 1,25(OH)2D3. Am J Physiol Gastrointest Liver Physiol 283: G864–G874. 59. Thompson PD, Jurutka PW, Whitfield GK, Myskowski SM, Eichhorst KR, Dominguez CE, Haussler CA, Haussler MR 2002 Liganded VDR induces CYP3A4 in small intestinal and colon cancer cells via DR3 and ER6 vitamin D responsive elements. Biochem Biophys Res Commun 299:730–738.

1723 60. Wali RK, Bolt MJ, Tien XY, Brasitus TA, Sitrin MD 1992 Differential effect of 1,25-dihydroxycholecalciferol on phosphoinositide turnover in the antipodal plasma membranes of colonic epithelial cells. Biochem Biophys Res Commun 187:1128–1134. 61. Bissonnette M, Tien XY, Niedziela SM, Hartmann SC, Frawley BP, Jr., Roy HK, Sitrin MD, Perlman RL, Brasitus TA 1994 1,25OH2 vitamin D3 activates PKC-alpha in Caco-2 cells: a mechanism to limit secosteroid-induced rise in [Ca2+]i. Am J Physiol. 267:G465–G475. 62. Sitrin MD, Bissonnette M, Bolt MJ, Wali R, Khare S, Scaglione-Sewell B, Skarosi S, Brasitus TA 1999 Rapid effects of 1,25(OH)2 vitamin D3 on signal transduction systems in colonic cells. Steroids 64:137–142. 63. Norman AW 1995 The vitamin D Endocrine system: manipulation of structure-function relationships to provide opportunities for development of new cancer chemopreventive and immunosuppressive agents. J Cell Biochem Suppl 22:218–225. 64. Nagpal S, Lu J, Boehm MF 2001 Vitamin D analogs: mechanism of action and therapeutic applications. Curr Med Chem 8:1661–1679. 65. Shabahang M, Buras RR, Davoodi F, Schumaker LM, Nauta RJ, Uskokovic MR, Brenner RV, Evans SR 1994 Growth inhibition of HT-29 human colon cancer cells by analogues of 1,25-dihydroxyvitamin D3. Cancer Res 54:4057–4064. 66. Akhter J, Goerdel M, Morris DL 1996 Vitamin D3 analogue (EB 1089) inhibits in vitro cellular proliferation of human colon cancer cells. Br J Surg 83:229–230. 67. Levy Y, Knutson JC, Bishop C, Shany S 1998 The novel analog 1,24(S)-dihydroxyvitamin D2 is as equipotent as 1,25-dihydroxyvitamin D3 in growth regulation of cancer cell lines. Anticancer Res 18:1769–1775. 68. Opolski A, Wietrzyk J, Chrobak A, Marcinkowska E, Wojdat E, Kutner A, Radzikowski C 1999 Antiproliferative activity in vitro of side-chain analogues of calcitriol against various human normal and cancer cell lines. Anticancer Res 19: 5217–5222. 69. Gaschott T, Steinmeyer A, Steinhilber D, Stein J 2002 ZK 156718, a low calcemic, antiproliferative, and prodifferentiating vitamin D analog. Biochem Biophys Res Commun 290:504–509. 70. Peleg S, Nguyen C, Woodard BT, Lee JK, Posner GH 1998 Differential use of transcription activation function 2 domain of the vitamin D receptor by 1,25-dihydroxyvitamin D3 and its A ring modified analogs. Mol Endocrinol 12:525–535. 71. Hofer H, Ho G, Peterlik M, Uskokovic MR, Lee JK, White MC, Posner GH, Cross HS 1999 Biological effects of 1alphahydroxy- and 1beta-(hydroxymethyl)-vitamin D compounds relevant for potential colorectal cancer therapy. J Pharmacol Exp Ther 291:450–455. 72. Oh YS, Kim EJ, Schaffer BS, Kang YH, Binderup L, MacDonald RG, Park JH 2001 Synthetic low-calcemic vitamin D(3) analogs inhibit secretion of insulin-like growth factor II and stimulate production of insulin-like growth factor-binding protein-6 in conjunction with growth suppression of HT-29 colon cancer cells. Mol Cell Endocrinol 183: 141–149. 73. Tanaka Y, Bush KK, Eguchi T, Ikekawa N, Taguchi T, Kobayashi Y, Higgins PJ 1990 Effects of 1,25-dihydroxyvitamin D3 and its analogs on butyrate-induced differentiation of HT-29 human colonic carcinoma cells and on the reversal of the differentiated phenotype. Arch Biochem Biophys 276: 415–423.

1724 74. Tanaka Y, Bush KK, Klauck TM, Higgins PJ 1989 Enhancement of butyrate-induced differentiation of HT-29 human colon carcinoma cells by 1,25-dihydroxyvitamin D3. Biochem Pharmacol 38:3859–3865. 75. Evans SR, Soldatenkov V, Shchepotin EB, Bogrash E, Shchepotin IB 1999 Novel 19-nor-hexafluoride vitamin D3 analog (Ro 25-6760) inhibits human colon cancer in vitro via apoptosis. Int J Oncol 14:979–985. 76. Diaz GD, Paraskeva C, Thomas MG, Binderup L, Hague A 2000 Apoptosis is induced by the active metabolite of vitamin D3 and its analogue EB1089 in colorectal adenoma and carcinoma cells: possible implications for prevention and therapy. Cancer Res 60:2304–2312. 77. Scaglione-Sewell BA, Bissonnette M, Skarosi S, Abraham C, Brasitus TA 2000 A vitamin D3 analog induces a G1-phase arrest in CaCo-2 cells by inhibiting cdk2 and cdk6: roles of cyclin E, p21Waf1, and p27Kip1. Endocrinology 141: 3931–3939. 78. Pence BC, Buddingh F 1988 Inhibition of dietary fat-promoted colon carcinogenesis in rats by supplemental calcium or vitamin D3. Carcinogenesis 9:187–190. 79. Kawaura A, Tanida N, Sawada K, Oda M, Shimoyama T 1989 Supplemental administration of 1α-hydroxyvitamin D3 inhibits promotion by intrarectal instillation of lithocholic acid in N-methyl-N-nitrosourea-induced colonic tumorigenesis in rats. Carcinogenesis 10:647–649. 80. Oda M, Kawaura A, Tanida N, Sawada K, Maekawa S, Kano M, Shimoyama T 1990 Effects of 1α-hydroxyvitamin D3 on N-methyl-N-nitrosourea-induced colonic tumorigenesis, and on fecal bile acid profiles with respect to soluble and precipitated phases in rats. Tokushima J Exp Med 37:75–81. 81. Comer PF, Clark TD, Glauert HP 1993 Effect of dietary vitamin D3 (cholecalciferol) on colon carcinogenesis induced by 1,2-dimethylhydrazine in male Fischer 344 rats. Nutr Cancer 19:113–124. 82. Belleli A, Shany S, Levy J, Guberman R, Lamprecht SA 1992 A protective role of 1,25-dihydroxyvitamin D3 in chemicallyinduced rat colon carcinogenesis. Carcinogenesis 13: 2293–2298. 83. Sitrin MD, Halline AG, Abrahams C, Brasitus TA 1991 Dietary calcium and vitamin D modulate 1,2-dimethylhydrazineinduced colonic carcinogenesis in the rat. Cancer Res 51: 5608–5613. 84. Beaty MM, Lee EY, Glauert HP 1993 Influence of dietary calcium and vitamin D on colon epithelial cell proliferation and 1,2-dimethylhydrazine-induced colon carcinogenesis in rats fed high fat diets. J Nutr 123:144–152. 85. Newmark HL, Lipkin M 1992 Calcium, vitamin D, and colon cancer. Cancer Res 52:2067s–2070s. 86. Risio M, Lipkin M, Newmark H, Yang K, Rossini FP, Steele VE, Boone CW, Kelloff GJ 1996 Apoptosis, cell replication, and Western-style diet-induced tumorigenesis in mouse colon. Cancer Res 56:4910–4916. 87. Mokady E, Schwartz B, Shany S, Lamprecht SA 2000 A protective role of dietary vitamin D3 in rat colon carcinogenesis. Nutr Cancer 38:65–73. 88. Otoshi T, Iwata H, Kitano M, Nishizawa Y, Morii H, Yano Y, Otani S, Fukushima S 1995 Inhibition of intestinal tumor development in rat multi-organ carcinogenesis and aberrant crypt foci in rat colon carcinogenesis by 22-oxa-calcitriol, a synthetic analogue of 1α,25-dihydroxyvitamin D3. Carcinogenesis 16:2091–2097. 89. Wali RK, Bissonnette M, Khare S, Hart J, Sitrin MD, Brasitus TA 1995 1α,25-dihydroxy-16-ene-23-yne-26,27hexafluorocholecalciferol, a noncalcemic analogue of

HEIDE S. CROSS

90.

91.

92.

93.

94.

95.

96.

97.

98.

99.

100. 101. 102. 103.

1α,25-dihydroxyvitamin D3, inhibits azoxymethane-induced colonic tumorigenesis. Cancer Res 55:3050–3054. Salim EI, Wanibuchi H, Taniyama T, Yano Y, Morimura K, Yamamoto S, Otani S, Nishizawa Y, Morii H, Fukushima S 1997 Inhibition of development of N,N′-dimethylhydrazineinduced rat colonic aberrant crypt foci by pre, post, and simultaneous treatments with 24R,25-dihydroxyvitamin D3. Jpn J Cancer Res 88:1052–1062. Taniyama T, Wanibuchi H, Salim EI, Yano Y, Otani S, Nishizawa Y, Morii H, Fukushima S 2000 Chemopreventive effect of 24R,25-dihydroxyvitamin D3 in N,N′-dimethylhydrazine-induced rat colon carcinogenesis. Carcinogenesis 21:173–178. Akhter J, Chen X, Bowrey P, Bolton EJ, Morris DL 1997 Vitamin D3 analog, EB1089, inhibits growth of subcutaneous xenografts of the human colon cancer cell line, LoVo, in a nude mouse model. Dis Colon Rectum 40:317–321. Evans SR, Schwartz AM, Shchepotin EI, Uskokovic M, Shchepotin IB 1998 Growth inhibitory effects of 1,25-dihydroxyvitamin D3 and its synthetic analog, 1α,25dihydroxy-16-ene-23yne-26,27-hexafluoro-19-nor-cholecalciferol (Ro 25-6760), on a human colon cancer xenograft. Clin Cancer Res 4:2869–2876. Tanaka Y, Wu AY, Ikekawa N, Iseki K, Kawai M, Kobayashi Y 1994 Inhibition of HT-29 human colon cancer growth under the renal capsule of severe combined immunodeficient mice by an analog of 1,25-dihydroxyvitamin D3, DD-003. Cancer Res 54:5148–5153. Wali RK, Bissonnette M, Khare S, Aquino B, Niedziela S, Sitrin M, Brasitus TA 1996 Protein kinase C isoforms in the chemopreventive effects of a novel vitamin D3 analog in rat colonic tumorigenesis. Gastroenterology 111:118–126. Iseki K, Tatsuta M, Uehara H, Iishi H, Yano H, Sakai N, Ishiguro S 1999 Inhibition of angiogenesis as a mechanism for inhibition by 1α-hydroxyvitamin D3 and 1,25-dihydroxyvitamin D3 of colon carcinogenesis induced by azoxymethane in Wistar rats. Int J Cancer 81:730–733. Evans SR, Shchepotin EI, Young H, Rochon J, Uskokovic M, Shchepotin IB 2000 1,25-dihydroxyvitamin D3 synthetic analogs inhibit spontaneous metastases in a 1,2-dimethylhydrazine-induced colon carcinogenesis model. Int J Oncol 16:1249–1254. Wali RK, Khare S, Tretiakova M, Cohen G, Nguyen L, Hart J, Wang J, Wen M, Ramaswamy A, Joseph L, Sitrin M, Brasitus T, Bissonnette M 2002 Ursodeoxycholic acid and F(6)-D3 inhibit aberrant crypt proliferation in the rat azoxymethane model of colon cancer: roles of cyclin D1 and E-cadherin. Cancer Epidemiol Biomarkers Prev 11:1653–1662. Huerta S, Irwin RW, Heber D, Go VL, Koeffler HP, Uskokovic MR, Harris DM 2002 1α,25-(OH)2-D3 and its synthetic analog decrease tumor load in the Apc(min) Mouse. Cancer Res 62:741–746. Thomas MG 1995 Luminal and humoral influences on human rectal epithelial cytokinetics. Ann R Coll Surg Engl 77:85–89. Thomas MG, Tebbutt S, Williamson RC 1992 Vitamin D and its metabolites inhibit cell proliferation in human rectal mucosa and a colon cancer cell line. Gut 33:1660–1663. Hansen CM, Maenpaa PH 1997 EB 1089, a novel vitamin D analog with strong antiproliferative and differentiation-inducing effects on target cells. Biochem Pharmacol 54:1173–1179. Gulliford T, English J, Colston KW, Menday P, Moller S, Coombes RC 1998 A phase I study of the vitamin D analog EB 1089 in patients with advanced breast and colorectal cancer. Br J Cancer 78:6–13.

CHAPTER 95 Vitamin D and Colon Cancer

104. Seow-Choen F, Vijayan V, Keng V 1996 Prospective randomized study of sulindac versus calcium and calciferol for upper gastrointestinal polyps in familial adenomatous polyposis. Br J Surg 83:1763–1766. 105. Tomon M, Tenenhouse HS, Jones G 1990 Expression of 25-hydroxyvitamin D3-24-hydroxylase activity in Caco-2 cells. An in vitro model of intestinal vitamin D catabolism. Endocrinology 126:2868–2875. 106. Birge SJ, Alpers DH 1973 Stimulation of intestinal mucosal proliferation by vitamin D. Gastroenterology 64:977–982. 107. Cross HS, Peterlik M, Reddy GS, Schuster I 1997 Vitamin D metabolism in human colon adenocarcinoma-derived Caco-2 cells: expression of 25-hydroxyvitamin D3-1alpha-hydroxylase activity and regulation of side-chain metabolism. J Steroid Biochem Mol Biol 62:21–28. 108. Bischof MG, Siu-Caldera ML, Weiskopf A, Vouros P, Cross HS, Peterlik M, Reddy GS 1998 Differentiationrelated pathways of 1 alpha,25-dihydroxycholecalciferol metabolism in human colon adenocarcinoma-derived Caco-2 cells: production of 1 alpha,25-dihydroxy-3epi-cholecalciferol. Exp Cell Res 241:194–201. 109. Bareis P, Bises G, Bischof MG, Cross HS, Peterlik M 2001 25-hydroxyvitamin D metabolism in human colon cancer cells during tumor progression. Biochem Biophys Res Commun 285:1012–1017. 110. Tong WM, Hofer H, Ellinger A, Peterlik M, Cross HS 1999 Mechanism of antimitogenic action of vitamin D in human colon carcinoma cells: relevance for suppression of epidermal growth factor-stimulated cell growth. Oncol Res 11:77–84. 111. Tangpricha V, Flanagan JN, Whitlatch LW, Tseng CC, Chen TC, Holt PR, Lipkin MS, Holick MF 2001 25-hydroxyvitamin D-1alpha-hydroxylase in normal and malignant colon tissue. Lancet 357:1673–1674. 112. Ogunkolade BW, Boucher BJ, Fairclough PD, Hitman GA, Dorudi S, Jenkins PJ, Bustin SA 2002 Expression of 25-hydroxyvitamin D-1-alpha-hydroxylase mRNA in individuals with colorectal cancer. Lancet 359:1831–1832. 113. Cross HS, Kallay E, Farhan H, Weiland T, Manhardt T 2003 Regulation of extrarenal vitamin D metabolism as a tool for colon and prostate cancer prevention. Recent Results Cancer Res 164:413–425. 114. Zehnder D, Bland R, Williams MC, McNinch RW, Howie AJ, Stewart PM, Hewison M 2001 Extrarenal expression of 25-hydroxyvitamin d(3)-1 alpha-hydroxylase. J Clin Endocrinol Metab 86:888–894. 115. Bises G, Kállay E, Weiland T, Wrba F, Wenzl E, Bonner E, Kriwanek S, Obrist P, Cross HS 2004 25-hydroxyvitamin D31α-hydroxylase expression in normal and malignant human colon. J Histochem Cytochem 52(7):985–989. 116. Cross HS, Kállay E, Korchide M, Lechner 2003 Regulation of extrarenal synthesis of 1,25-dihydroxyvitamin D3— relevance for colonic cancer prevention and therapy. Molecular Aspects of Medicine. 24(6):459–465.

1725 117. Kállay E, Bajna E, Cross HS 2002 Gender- and sitespecific expression of 25-hydroxyvitamin D3-1α-hydroxylase in the mouse large intestine: relevance for colonocyte hyperproliferation. Proceedings of the 93rd Anual Meeting of the American Association for Cancer Research. San Francisco. 43:127. 118. Potter JD, Bostick RM, Grandits GA, Fosdick L, Elmer P, Wood J, Grambsch P, Louis TA 1996 Hormone replacement therapy is associated with lower risk of adenomatous polyps of the large bowel: the Minnesota Cancer Prevention Research Unit Case-Control Study. Cancer Epidemiol Biomarkers Prev 5:779–784. 119. Ochiai M, Watanabe M, Kushida H, Wakabayashi K, Sugimura T, Nagao M 1996 DNA adduct formation, cell proliferation, and aberrant crypt focus formation induced by PhIP in male and female rat colon with relevance to carcinogenesis. Carcinogenesis 17:95–98. 120. Design of the Women’s Health Initiative clinical trial and observational study. The Women’s Health Initiative Study Group. Control Clin Trials 19:61–109. 121. Kumar R, Cohen WR, Silva P, Epstein FH 1979 Elevated 1,25-dihydroxyvitamin D plasma levels in normal human pregnancy and lactation. J Clin Invest 63:342–344. 122. Lund B, Selnes A 1979 Plasma 1,25-dihydroxyvitamin D levels in pregnancy and lactation. Acta Endocrinol (Copenh) 92:330–335. 123. Baksi SN, Kenny AD 1978 Does estradiol stimulate in vivo production of 1,25-dihydroxyvitamin D3 in the rat? Life Sci 22:787–792. 124. Smirnoff P, Liel Y, Gnainsky J, Shany S, Schwartz B 1999 The protective effect of estrogen against chemically-induced murine colon carcinogenesis is associated with decreased CpG island methylation and increased mRNA and protein expression of the colonic vitamin D receptor. Oncol Res 11:255–264. 125. Sathyamoorthy N, Wang TT 1997 Differential effects of dietary phyto-oestrogens daidzein and equol on human breast cancer MCF-7 cells. Eur J Cancer 33:2384–2389. 126. Foley EF, Jazaeri AA, Shupnik MA, Jazaeri O, Rice LW 2000 Selective loss of estrogen receptor beta in malignant human colon. Cancer Res 60:245–248. 127. Campbell-Thompson M, Lynch IJ, Bhardwaj B 2001 Expression of estrogen receptor (ER) subtypes and ERbeta isoforms in colon cancer. Cancer Res 61:632–640. 128. Kallay E, Adlercreutz H, Farhan H, Lechner D, Bajna E, Gerdenitsch W, Campbell M, Cross HS 2002 Phytoestrogens regulate vitamin D metabolism in the mouse colon: relevance for colon tumor prevention and therapy. J Nutr 132: 3490S–3493S. 129. Farhan H, Wahala K, Cross HS 2003 Genistein inhibits vitamin D hydroxylases CYP24 and CYP27B1 expression in prostate cells. J Steroid Biochem Mol Biol 84: 423–429.

CHAPTER 96

Vitamin D and Hematological Malignancy JAMES O’KELLY ROBERTA MOROSETTI H. PHILLIP KOEFFLER

I. II. III. IV.

Division of Hematology/Oncology, Department of Medicine, Cedars-Sinai Medical Center, UCLA School of Medicine, Los Angeles, California Pediatric Oncology Division, Catholic University of Rome, Rome, Italy Division of Hematology/Oncology, Department of Medicine, Cedars-Sinai Medical Center, UCLA School of Medicine, Los Angeles, California

Overview of Hematopoiesis Vitamin D Receptors in Blood Cells Effects of Vitamin D Compounds on Normal Hematopoiesis Effects of Vitamin D Compounds on Leukemic Cell Lines

I. OVERVIEW OF HEMATOPOIESIS Hematopoiesis is the process that leads to the formation of the highly specialized circulating blood cells from bone marrow pluripotent progenitor stem cells. These stem cells are the most primitive blood cells, and they have the ability to either self-replicate or differentiate. They are regulated by a feedback system and are affected by various stimuli such as bone marrow depletion, hemorrhage, infection, and stress. They produce more mature “committed” cells that are able to proliferate and differentiate into cells of different lineages, acquiring specific functional properties (Fig. 1). The pluripotent stem cell common to granulocytes, erythrocytes, monocytes, and megakaryocytes is called the colony-forming unit-GEMM (CFU-GEMM), and the committed cells giving rise to the lineage specific cells are assayed in vitro as erythroid burst-forming units (BFU-E), megakaryocyte colony-forming units (CFUMK), and granulocyte-monocyte colony-forming units (CFU-GM). Each of these stem cells has cell surface receptors for specific cytokines. Binding of cytokines to these receptors stimulates secondary intracellular signals that deliver a message to the nucleus to stimulate proliferation, differentiation, and/or activation. The CFU-GMs, in the presence of cytokines, undergo a differentiation program progressing to granulocytes and monocytes. The growth factors acting primarily on the granulocyte-macrophage pathway are granulocyte-macrophage colony-stimulating factor (GM-CSF), granulocyte colony-stimulating factor (G-CSF), and macrophage colony-stimulating factor (M-CSF). The GM-CSF also stimulates VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Vitamin D Analogs Effective Against Leukemic Cells VI. Summary and Conclusions References

eosinophils, enhances megakaryocytic colony formation, and increases erythroid colony formation in the presence of erythropoietin (Epo). In vivo, it causes an increase in granulocytes, monocytes, and eosinophils. It can activate these cells to efficiently fight microbes. The G-CSF stimulates the formation of granulocyte colonies in vitro. It is able to act synergistically with interleukin-3 (IL-3), GM-CSF, and M-CSF. This cytokine is active in vivo, stimulating an increase of peripheral blood granulocytes. The M-CSF stimulates the formation of macrophage colonies in vitro. It maintains the survival of differentiated macrophages and increases their anti-tumor activities and secretion of oxygen reduction products and plasminogen activators. This cytokine binds to a receptor that is the product of the protooncogene c-fms. Interleukin-3 has multilineage stimulating activity and acts directly on the granulocyte-macrophage pathway, but also enhances the development of erythroid, megakaryocytic, and mast cells, and possibly T lymphocytes. In synergy with Epo, IL-3 stimulates the formation of early erythroid stem cells, promoting the formation of colonies of red cells in soft gel culture known as BFU-E. In addition, it supports the formation of early multilineage blast cells in vitro. IL-3 also induces leukemic blasts to enter the cell cycle and induces, either alone or in combination with other growth factors, the production of all the myeloid cells in vivo. Stem cell factor (SCF) promotes survival, proliferation and differentiation of hematopoeitic progenitor cells. It synergizes with other growth factors such as IL-3, GM-CSF, G-CSF, and Epo to support the colony growth of BFU-E, CFU-GM, and CFU-GEMM in vitro. Copyright © 2005, Elsevier, Inc. All rights reserved.

1728

JAMES O’KELLY, ROBERTA MOROSETTI, AND H. PHILLIP KOEFFLER

SCF EPO IL-3

BFU-E

EPO

SCF IL -3 BFU-MK

CFU-Meg

SCF IL-3

SCF CFU-GEMM

SCF IL-3 GM-CSF TPO

CFU-M IL-3 GM-CSF CFU-GM G-CSF CFU-G

STEM CELL

Macrophage

Neutrophil

Eosinophil

CFU-Eu CFU-Blast

Platelets

GM-CSF M-CSF Monocyte

IL-3 GM-CSF G-CSF

IL-3 GM-CSF IL-5 SCF

IL -6 TPO Megakaryocyte

IL-3 GM-CSF G-MSF

IL -3 GM-CSF -

Red Cell

Reticulocyte

CFU-E

IL -3 IL -4 CFU-Bas

Basophil IL-1 IL IL-2 IL IL-6 IL IL-7 IL

Pre-T

T lymphocyte

SCF Lymphoid stem cell

SCF IL-7

IL-6 IL-4

Pre-B B lymphocyte

FIGURE 1 Scheme of hematopoiesis. The key progenitor cells and their growth factors are shown. CFUBlast, colony-forming unit-blast; CFU-GEMM, CFU-granulocyte, erythrocyte, megakaryocyte, macrophage; BFU-E, burst-forming unit-erythroid; CFU-E, CFU-erythroid; BFU-MK, BFU-megakaryocyte; CFU-Meg, CFU-megakaryocyte; CFU-GM, CFU-granulocyte-monocyte; CFU-Eo, CFU-eosinophil; CFU-Bas, CFUbasophil; SCF, stem cell factor; IL-3, interleukin-3; GM-CSF, granulocyte-monocyte colony-stimulating factor; EPO, erythropoietin; TPO, thrombopoietin.

Although SCF alone has a modest effect on colony growth, in the presence of other cytokines SCF increases both the size and the number of these colonies. It is a ligand for the c-kit receptor, a tyrosine kinase receptor that is expressed in hematopoeitic progenitor cells. The growth factor Epo stimulates the formation of erythroid colonies (CFU-E) in vitro and is the primary hormone of erythropoiesis in animals and humans in vivo. It binds to a specific receptor (Epo-R). Production of erythroblasts is hormonally regulated by a feedback mechanism mediated by the linear correlation between tissue oxygenation of Epo-producing cells in the kidney mediated by oxygen-carrying hemoglobin in red blood cells. Anemia causes tissue hypoxia, resulting in an increase of serum Epo levels.

II. VITAMIN D RECEPTORS IN BLOOD CELLS The genomic actions of 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] are mediated by the intracellular vitamin D receptor (VDR), which belongs to a large

family of nuclear receptors [1]. VDR forms a heterodimer with the retinoid X receptor (RXR); this complex regulates expression of target genes by binding to vitamin D responsive elements (VDREs) in the promoter regions of their target genes [2]. The mechanism of action of 1,25(OH)2D3 via the VDR is discussed in Chapters 11 and 13. Expression of VDR has been detected in various normal and leukemic hematopoeitic cells. It is expressed constitutively in monocytes, in certain subsets of thymocytes, and after in vitro activation of B and T lymphocytes [3–5]. Expression of VDR is induced in the lymphocytes of patients with rheumatoid arthritis, in human tonsillar lymphocytes, and in pulmonary lymphocytes of patients with tuberculosis and sarcoidosis [6–8]. In addition, lymphocytes of patients with hereditary vitamin D–resistant rickets type II (HVDRR) have various alterations of the VDR [9]. Also, fewer receptors have been detected in the peripheral blood mononuclear cells of patients with X-linked hypophosphatemic rickets [10]. Examination of a large array of myeloid leukemia cell lines blocked at various stages of maturation showed that they all

1729

CHAPTER 96 Vitamin D and Hematological Malignancy

expressed VDR, albeit at different levels [3]. Bone marrow-derived stromal cells express VDR and show a reduction of their proliferation that occurs after their exposure to 1,25(OH)2D3. Both T-helper and T-suppressor lymphocytes express similar concentrations of VDR. In particular, T lymphocytes express high levels of VDR mRNA, whereas resting B lymphocytes express either very low or nondetectable levels of VDR transcripts [3]. Nevertheless, 1,25(OH)2D3 inhibits the synthesis of immunoglobulins (Ig) by B lymphocytes in vitro [11,12]. This suppression, however, could be the result of the inhibition of T-helper activity [12]. Production of lymphokines, including IL-2, is markedly decreased by 1,25(OH)2D3 in activated T lymphocytes, and this could cause the suppression of Ig synthesis [13–16]. The effects of vitamin D on the immune system are discussed in Chapter 36. Studies by us in VDR knockout (KO) mice indicated that expression of VDR is dispensible for normal myeloid development [17]. No difference in the numbers and percentages of red and white cells were found between VDR KO and wild-type (WT) mice. Committed myeloid stem cells from the bone marrow cultured in methylcellulose formed similar numbers of colonies when grown in the presence of either GM-CSF, G-CSF, M-CSF alone or in combination with IL-3. Furthermore, bone marrow cells from VDR KO and WT mice formed a similar number and percentage of granulocyte, macrophage, and granulocyte/macrophage mixed colonies when cultured in methylcellulose with GMCSF and IL-3. Under these conditions, treatment with 1,25(OH)2D3 dramatically increased the percentage of macrophage colonies derived from WT but not VDR KO bone marrow cultures. This observation demonstrates the requirement of VDR expression for 1,25(OH)2D3–induction of bone marrow progenitors into monocytes/macrophages. The proportion of T- and B-cells were normal in the VDR KO mice. However, the antigen-stimulated spleen cells from VDR KO mice produced less IFNγ and more IL-4 than those from WT mice, indicating impaired Th1 differentiation. Additionally, IL-12 stimulation induced a weaker proliferative response in VDR KO splenocytes as compared to WT, and expression of STAT4 was reduced. These results suggest that VDR plays an important role in the Th1-type immune response. The HL-60 myeloblastic cell line cultured in the presence of 1,25(OH)2D3 (10−7 M) has a 50% decrease of VDR protein levels at about 24 hr, which returned to normal levels after 72 hr; no change of VDR mRNA expression occurred [3]. These data suggested that one of the major sites of regulation of VDR expression occurs at the posttranscriptional level. The same cell line exposed to a lower dose of 1,25(OH)2D3 for 12 hr

appeared to have an increased number of VDRs, as determined by immunoprecipitation, which returned to normal levels after 72 hr [18]. The HL-60 myeloblasts cultured with retinoic acid (RA) and dimethyl sulfoxide (DMSO) or 12-Otetra-decanoylphorbol-13-acetate (TPA) terminally differentiate into granulocytes or macrophages, respectively. The differentiation is associated with induction of high expression of VDR mRNA. Also, normal human nondividing macrophages express VDR mRNA, and these levels do not change after exposure to activating factors such as tumor necrosis factor α (TNFα). The expression of VDR mRNA was not detectable in nonproliferating lymphocytes harvested from normal human peripheral blood, but VDR mRNA expression increased in proliferating lymphocytes after a 24 hr exposure to the lectin phytohemagglutinin-A (PHA), suggesting that in lymphocytes a major site of regulation of VDR expression is at the transcriptional level [3, 19]. Moreover, low levels of VDR expression were detected in low-grade non-Hodgkin’s lymphoma (NHL) tumor samples and in the follicular lymphoma B-cell lines SU-DHL4 and SU-DHL5 [20]. The VDR can bind to the osteocalcin response element along with the activator protein-1 (AP1) complexes [21]. In addition, Jun and Fos proto-oncogenes are upregulated by 1,25(OH)2D3 [22]. Jun-D DNA binding activity is increased during cell cycle arrest in the human chronic myelogenous leukemia RWLeu-4 cultured with 1,25(OH)2D3, suggesting that Jun D binding activity may play a role in the regulation of cell proliferation by 1,25(OH)2D3 [21].

III. EFFECTS OF VITAMIN D COMPOUNDS ON NORMAL HEMATOPOIESIS The role of 1,25(OH)2D3 in cell differentiation was first described by Abe et al. [23] in the murine leukemia cell line Ml, which was induced to differentiate into more mature cells by 1,25(OH)2D3. Normal human bone marrow committed stem cells cultured in either soft agar with 1,25(OH)2D3 (10−7 M) or in liquid culture with 1,25(OH)2D3 (5 × 10−9 M for 5 days) and monocytes cultured in serum-free medium with 1,25(OH)2D3 (5 × 10−8 M for 7 days) differentiate into macrophages [24,25]. In further studies, these macrophages were functionally competent and able to release large amounts of TNFα and IL-6 [26]. Furthermore, the terminal differentiation of monocytes into mature macrophages can be obtained in vitro by culturing these cells in the presence of serum or in a serum-free medium with the addition of vitamin D3 compounds [4,26,27].

1730

JAMES O’KELLY, ROBERTA MOROSETTI, AND H. PHILLIP KOEFFLER

As mentioned earlier, 1,25(OH)2D3 is able to inhibit IL-2 synthesis and the proliferation of peripheral blood lymphocytes [12–15]. Indeed, 1,25(OH)2D3 appears to be able to regulate many lymphokines. For example, Tobler et al. [28] showed that expression of the lymphokine GM-CSF is regulated by 1,25(OH)2D3 through VDR by a process independent of IL-2 production. In particular, 1,25(OH)2D3 was able to inhibit both GMCSF mRNA and protein expression in PHA-activated normal human peripheral blood lymphocytes (PBL). The former occurred at least in part by destabilizing and shortening the half-life of the GM-CSF mRNA [28]. The down-regulation of GM-CSF was obtained at 1,25(OH)2D3 concentrations similar to those reached in vivo, with a 50% reduction of GM-CSF activity occurring at 10−10 M 1,25(OH)2D3. In addition, IL-2 did not affect the modulation of GM-CSF production by 1,25(OH)2D3 in the PBL cocultured with 1,25(OH)2D3 (10−10−10−7 M) and high concentrations of IL-2.

IV. EFFECTS OF VITAMIN D COMPOUNDS ON LEUKEMIC CELL LINES All of the studies conducted so far with 1,25(OH)2D3 emphasize the need for new vitamin D3 analogs with greater anti-leukemic effects and less toxicity. In spite of the promising data obtained from in vitro studies, results of clinical trials in leukemia

with 1,25(OH)2D3 are limited in scope and thus far have exhibited only mediocre results. For example, the myelodysplastic syndrome (MDS) is associated with anemia, thrombocytopenia, and leukopenia and an increased number of myeloid progenitor cells in the bone marrow. Some patients with MDS go on to develop acute myeloid leukemia. We treated 18 MDS patients with increasing doses of 1,25(OH)2D3 up to a maximum of 2 µg/day for 12 weeks. Although an improvement of at least one hematologic parameter occurred in 8 patients after more than 4 weeks, the response was not durable and not detectable at the end of the study at 12 weeks [24]. Nine patients developed hypercalcemia, which was the dose-limiting toxicity. In another study, seven MDS patients were treated with 1,25(OH)2D3 (2.5 µg/day, for at least 8 weeks), with no beneficial effects [29]. A major drawback in using 1,25(OH)2D3 is its calcemic effect, which prevents pharmacological doses of the compound from being given. Vitamin D analogs have been synthesized that have enhanced potency to inhibit proliferation and promote differentiation of cancer cells, with less calcemic activity as compared to 1,25(OH)2D3 (see Chapters 80–88). Many of these analogs in vitro are between 10- and a 1000-fold more active than the parental 1,25(OH)2D3 in their growth suppressive activity. A comparison of the relative anti-leukemic potencies of vitamin D compounds is provided in Table I. These analogs could provide

TABLE I Effect of Vitamin D Compounds on Clonal Proliferation of HL-60 Cells in Soft Agar and Calcium Levels in Mice Compound 1,25(OH)2D3 1,25(OH)2-16-ene-D3 1,25(OH)2-16-ene-23-yne-D3 1,25(OH)2-16-ene-5,6-trans-D3 1,25(OH)2-16-ene-24-oxo-D3 1,25(OH)2-16-ene-19-nor-D3 1,25(OH)2-16-ene-24-oxo-19-nor-D3 1,25(OH)2-20-epi-D3 1,25(OH)2-20-epi-22-oxa-24,26,27-trishomo-D3d 1,25(OH)2-diene-24,26,27-trihomo-D3e 19-nor-1,25(OH)2D2f

ED50a (×10−9 mol/l) 4–18 0.015 3 0.03 0.2 0.8 0.1 0.006 0.001 0.23 2.4

aED represents the effective dose achieving 50% growth inhibition of HL-60 cells. 50 bMTD, Maximally tolerated dose; highest dose reported that did not produce hypercalcemia

toneally, three times per week. cND, not done. dLeo Pharmaceutical code name is KH 1060. e Leo Pharmaceutical code name is EB 1089. f Abbott Laboratories code name is Paricalcitol.

MTDb (µg) 0.0625 0.125 2 4 NDc 0.5 6 0.00125 0.0125 0.25 0.1

Reference [102–109] [102] [102, 104] [103] [107] [107] [106] [104, 108, 109] [108] [104] [31]

or other noticeable toxicities in mice when injected intraperi-

1731

CHAPTER 96 Vitamin D and Hematological Malignancy

a larger therapeutic window for the treatment of hematologic malignancies, retaining the useful properties of 1,25(OH)2D3 [30].

A. Celluar Effects of Vitamin D Compounds on Leukemic Cells The various vitamin D compounds have similar effects on inducing differentiation and inhibiting proliferation of several acute myeloid leukemia cell lines such as HL-60, U937, THP-1, HEL, and NB4. In contrast, more immature myeloid leukemia cell lines such as HL-60 blasts, KG1, KGla, and K562 do not respond to the hormone. Vitamin D analogs inhibit cell growth mainly by inducing cell cycle arrest. Many studies have shown that treated leukemic cell lines accumulate in the G0/G1 and G2/M phase of the cell cycle, with a concomitant decrease in the proportion of cells in S-phase [31–33]. These effects of vitamin D compounds on the cell cycle are discussed in Chapter 92. HL-60 cells treated with 1,25(OH)2D3 acquire the morphology and functional characteristics of macrophages. Expression of the cell-surface markers CD14 and CD11b are up-regulated. The cells become adherent to charged surfaces, develop pseudopodia, stain positively for nonspecific esterase (NSE) with a reduction of nitroblue tetrazolium (NET), and acquire the ability to phagocytose yeast during incubation with 1,25(OH)2D3 (10−10−10−7 M for 7 days) [25,34]. In addition, the treated cells acquired the ability to degrade bone marrow matrix in vitro, raising the possibility that the cells may have acquired some osteoclast-like characteristics. Leukemic cells from acute myelogenous leukemia (AML) patients respond to vitamin D compounds when cultured in vitro; however, they are often less sensitive than the cell lines. They are often still able to undergo partial monocytic differentiation as assessed by NBT reduction, morphology, and phagocytic ability. Furthermore, their clonal growth is often inhibited [25,34]. The molecular targets of vitamin D3 compounds in leukemic cells are described in the following section, and are summarized in Table II.

B. Molecular Mechanisms of Action of Vitamin D Compounds Against Leukemic Cells Vitamin D compounds can exert their anti-cancer effects by activating the VDR and modulating the transcription of various target genes. Some of these target genes are associated with inhibition of growth and

TABLE II Molecular Targets of Vitamin D Compounds in Leukemic Cellsa Cell Cycle/Apoptosis Cyclin A ↑ Cyclin D1 ↑ Cyclin E ↑ p15 ↑ p21Waf1 ↑ p27Kip1 ↑ Bcl-2 ↓ Differentiation Markers CD11b ↑ CD14 ↑

Oncogenes c-myc ↓ Dek ↓ Fli ↓ Tumor Suppressors PTEN ↑ BTG ↑ Kinases PI3-K ↑activity p38 MAPK ↑activity ERK 1/2 ↑activity PKC ↑levels

a Regulation of expression or activity may occur either directly or as a consequence of differentiation. See text for details.

induction of differentiation, but this modulation may not be a direct effect, as it may simply reflect the entire process of differentiation. Myeloid leukemic cell lines treated with 1,25(OH)2D3 undergo an initial proliferative burst, which is followed by growth inhibition, terminal differentiation, and subsequent apoptosis [36,37]. Levels of cyclin A, D1, and E increased in U937 cells within 24 hours of 1,25(OH)2D3–treatment, decreasing after 48 hours, although cyclin dependent kinase (CDK) levels did not change [37]. The CDK inhibitors p21 and p27, important regulators of the cell cycle, were elevated during the periods of both proliferation and growth inhibition. A strong correlation appears to exist between early induction of p21 and the beginning of the differentiation program. The marked increase of p21 protein expression in response to 1,25(OH)2D3 may be due to enhanced posttranscriptional stabilization of p21 mRNA [38]. The up-regulation of p21 mRNA occurred independently of de novo protein synthesis, further supporting the hypothesis that p21 is an early response gene. Indeed, the p21 promoter contains a vitamin D response element, and induction requires the presence of VDR. Also, experiments using a variety of cell lines showed that 1,25(OH)2D3 and other differentiating agents could mediate their induction of p21-independent of an intact p53 gene [38]. Using differential hybridization, Liu et al. [39] showed that p21 is differentially expressed in response to 1,25(OH)2D3 in the myelomonocytic cell line U937. Transient overexpression of p21 and p27 in U937 cells

1732

JAMES O’KELLY, ROBERTA MOROSETTI, AND H. PHILLIP KOEFFLER

promoted the appearance of the cell surface differentiation molecules CD14 and CD11b. One series of experiments showed that the p15, p16, p18, p21, and p27 CDKIs were up-regulated in a time-dependent manner after the addition of 1,25(OH)2D3 [39]. This induction occurred within 4 hr of the addition of 1,25(OH)2D3 in the presence of cycloheximide (CHX), suggesting a direct transcriptional activation by VDR. In another study, the protein expression of different G1-phase regulators has been examined in HL-60 cells exposed to different concentrations of 1,25(OH)2D3. A strong up-regulation of p27 protein expression was evident after 72 hr of exposure to the compound, and it was dependent on 1,25(OH)2D3 concentration. This up-regulation was also associated with increased levels of D- and E-cyclins, coinciding with the G1 arrest. These results suggested a prominent role of p27 in mediating the antiproliferative activity of 1,25(OH)2D3 in this cell line [40]. Activation of the protooncogene c-myc by retroviral insertion or chromosomal rearrangement is a typical feature of human leukemias. The HL-60 leukemia cell line is characterized by high levels of expression of cmyc due to gene amplification [41, 42]. Treatment of this cell line with 1,25(OH)2D3 results in a down-regulation of expression of this oncogene related to cell differentiation [43]. Supression of c-myc by 1,25(OH)2D3 and its noncalcemic analogs has been demonstrated to occur at the transcriptional level in HL-60 cells [30,44]. 1,25(OH)2D3 is thought to up-regulate proteins such as the homeobox gene, HoxB4, that binds to the first exon/intron border of c-myc to prevent transcriptional elongation, a process dependent on activation of PKCβ [45,46]. Another homeobox gene, HoxA10, was found by differential display to be a gene transcriptionally induced by 1,25(OH)2D3 during differentiation of U937 cells [47]. 1,25(OH)2D3 has a protective effect against apoptosis in HL-60 cells [48,49]. This effect lends support to the observation that monocytic differentiation interferes with programs leading to apoptotic death. In other cell types, inhibition of apoptosis correlates with elevated levels of Bcl-2, but this does not appear to be the case with myeloid cells. In fact, after culture with 1,25(OH)2D3 a down-regulation of Bcl-2 was observed both at the mRNA and protein levels [49]. Exposure of HL-60 cells to 1,25(OH)2D3 induces the expression of the protooncogene c-fms, which occurs in parallel with the induction of CD14 expression and a block of their cell cycle in G0/G1 phase [50]. In the chronic myelogenous leukemia (CML) cell line RWLeu-4, an inhibition of proliferation was observed after 1,25(OH)2D3 treatment. Moreover, the binding activity of the protooncogene junD was enhanced by 1,25(OH)2D3 in these cells

during their cell cycle arrest, whereas it was not decreased in a 1,25(OH)2D3–resistant variant cell line [21]. Fusion proteins involving the retinoic acid receptor alpha (RAR α) with either the PML or PLZF nuclear proteins are the genetic markers of acute promyelocytic leukemias (APLs). Although APLs with PMLRARα are more sensitive to retinoic acid, expression of either PML-RARα or PLZF-RARα in U937 and HL-60 cells blocks terminal differentiation induced by 1,25(OH)2D3 [51]. Both PML-RARα or PLZF-RARα can bind to VDR in U937 cells and sequester VDR away from its normal sites of localization [52]. Overexpression of VDR overcomes the block in 1,25(OH)2D3–stimulated differentiation caused by the fusion proteins. The cell lines HL-60 and U937 have been used to attempt to identify early response genes directly regulated by VDR. Bories et al. [53] identified a serine protease, myeloblastin, that was down-regulated by phorbol esters in promyelocytic cells, causing growth arrest and cell differentiation. They also isolated cDNAs coding for fructose 1,6-biphosphatase, whose expression is up-regulated by 1,25(OH)2D3 in HL-60 cells and peripheral blood monocytes. Genes regulated during the course of 1,25(OH)2D3–mediated HL-60 cell differentiation have been analyzed using cDNA array analysis [54]. Among the genes shown to be down-regulated were the putative oncogenes Dek and Fli-1, and up-regulated genes included the antiproliferative BTG1. Increasing evidence suggests that both the antiproliferative and differentiation-inducing effects of vitamin D compounds require their modulation of the intracellular kinase pathways, p38 MAPK, ERK, and PI3-K. Activation of PI3-K has been shown to be required for 1,25(OH)2D3–stimulated myeloid differentiation, as determined by induction of CD14 expression [55]. PI3-K was activated by 1,25(OH)2D3 in THP-1 cells within 20 minutes. Pretreatment with the PI3-K inhibitors, LY 294004 and wortmanin, inhibited CD14 induction in response to 1,25(OH)2D3 in THP-1 cells and peripheral blood monocytes. Furthermore, antisense oligonucleotides against PI3-K blocked induction of CD14 expression in THP-1 and U937 cells. Expression of the VDR was required for activation of PI3-K; and interestingly, VDR was found to associate with the active form of the kinase. Inhibitors of PI3-K have also been shown to block the differentiation induced by 1,25(OH)2D3 in HL-60 cells [56]. Exposure of either HL-60 or NB-4 cells to differentiation-inducing concentrations of vitamin D compounds causes activation and nuclear translocation of MAPK [57–59]. In addition, the vitamin D3 analog EB1089 was recently demonstrated to induce apoptosis of B-cell chronic lymphocytic leukemia cells from

CHAPTER 96 Vitamin D and Hematological Malignancy

patients, an event preceded by stimulation of p38 MAPK and suppression of ERK activity [60]. Furthermore, 1,25(OH)2D3 was found to stimulate the transient [24–48h] phosphorylation of ERK1/2, which was followed by growth arrest and differentiation of HL-60 cells [61]. In another study, PD98059, an ERK1/2 inhibitor, blocked the 1,25(OH)2D3–stimulated differentiation of HL-60 cells [62]. Activation of PKC by the phorbol diesters such as TPA, promotes monocyte differentiation of leukemic cell lines [63,64]. Differentiation of HL-60 cells in response to 1,25(OH)2D3 is accompanied by increased levels of PKCβ, and this differentiation can be inhibited by the specific PKC inhibitor, chelerythrine chloride [65]. Other vitamin D analogs have been shown to stimulate expression and translocation of PKCα and delta during NB-4 monocytic differentiation [66]. Some of the effects of vitamin D compounds on the signaling pathways occur within seconds. For example, rapid changes in the phosphorylation status of MAPK (within 30 seconds) have been demonstrated in response to 1,25(OH)2D3 in NB-4 cells [58]. These effects occur too quickly to be attributed to the genomic actions of vitamin D–mediated activated transcription of target genes by VDR. Nonetheless, 1,25(OH)2D3–activated intracellular signaling pathways require the presence of VDR to stimulate monocyte/macrophage differentiation, as demonstrated by studies on bone marrow cells from VDR KO mice [17] and cells from patients with vitamin D–dependent rickets type II [67,68]. The rapid nongenomic activities of vitamin D are described in detail in Chapter 23.

C. Vitamin D Compounds in Combination with Other Agents Because of the potential toxicity of 1,25(OH)2D3 and its analogs at the concentrations required in vivo, various attempts have been made to use them with other compounds that might act synergistically to achieve an anti-leukemic effect capable of promoting cell differentiation, yet with an acceptable toxicity. A range of compounds with different mechanisms of action have been studied. Vitamin D compounds may cooperate with other differentiating agents such as retinoids, tissue plasminogen activator, and interferon (IFN). For example, 1,25(OH)2D3 can potentiate IFN-γ action to induce the expression of CD11b and CD14. We and others have shown that the combination of vitamin D analogs and either all-trans-retinoic acid (ATRA) or 9-cis-retinoic acid (9-cis-RA) can potentiate the terminal differentiation process of HL-60 cells down the monocyte-macrophage pathway [69,70]. These findings have also been demonstrated in other studies [71,72].

1733 Cells cultured in the presence of the combination of 1,25(OH)2D3 and ATRA developed atypically, having a neutrophilic morphology, but in other properties were typical of monocytes (e.g., CD14 expression, ability to bind to bacterial LPS, and ability to develop sodium fluoride-inhibited NSE) [69,70]. The combination of ATRA (10−9 M) and the vitamin D3 analogs l,25(OH)216-ene-23-yne D3 or 1,25(OH)2-23-yne D3 (10−9 to 10−10 M) showed a synergistic effect on the induction of differentiation and inhibition of proliferation of HL-60 cells [73]. A decrease of c-myc expression was also observed in the presence of ATRA and l,25(OH)2-16ene-23-yne D3. This down-regulation of c-myc was stronger than that observed using single agents and correlated with the initiation of differentiation. A synergistic antineoplastic effect of l,25(OH)2-16-ene-23yne D3 and ATRA has been shown in HL-60 cells [73]. A HL-60 clone resistant to ATRA was much more sensitive to inhibition of proliferation by l,25(OH)2-16ene-23-yne D3 as compared with 1,25(OH)2D3. In addition, the induction of differentiation of these cells by l,25(OH)2-16-ene-23-yne D3 was much stronger in these cells in contrast to wild-type HL-60 cells. Another retinoid-resistant acute promyelocyte leukemia cell line (UF-1) was induced towards granulocye differentiation by 1,25(OH)2D3, in association with stimulation of p21Waf1 and p27Kip1 expression [74]. These effects were enhanced by the addition of ATRA. In the promyelocytic cell line NB4, carrying the translocation t(15;17) typical of APL, vitamin D compounds can act as weak inducers of monocytic differentiation [75,76]. Bathia et al. [77] showed that the combination of 1,25(OH)2D3 and TPA resulted in a synergistic response in NB4 cells, causing a complete differentiation to fully functional adherent macrophages with a rapid arrest of cell growth in the first 24 hr. Remarkable inhibition of proliferation and induction of differentiation occurred when NB4 cells were cultured with both 9-cis-RA and KH1060 (a 20-epi-vitamin D3 analog) [76]. ATRA and 1,25(OH)2D3 also synergistically induce monocytic differentiation in the promonocytic cell line U937 [78]. The same group observed that U937 cells exposed to a moderate thermal stress responded with increased differentiation after the addition of 1,25(OH)2D3 and ATRA [79]. Nonsteroidal anti-inflammatory drugs (NSAIDs) have been shown to enhance the differentiation of HL-60 cells in respnse to 1,25(OH)2D3 and its analogs [80,81]. This effect may occur because of the ability of NSAIDs to inhibit an aldoketoreductase; this enzyme supresses the production of natural PPAR γ ligands by blocking the conversion of prostaglandin D2 to prostaglandin J2 [82].

1734

JAMES O’KELLY, ROBERTA MOROSETTI, AND H. PHILLIP KOEFFLER

Vitamin D compounds have also been combined successfully with naturally occurring plant products. One of these is carnosic acid, a plant-derived polyphenol antioxidant recently shown to potentiate the prodifferentiative effects of 1,25(OH)2D3 [83,84]. These studies demonstrated that carnosic acid enhanced the growth arrest and CD11b and CD14 expression induced by 1,25(OH)2D3 in a HL-60 subline, and combining these compounds produced greater stimulation of VDR expression and activity as determined by gelshift analysis. Differentiation was correlated with antioxidant activity, and was associated with activation of the RAF-ERK pathway and increased binding of the AP-1 transcription factor to the promoter of VDR. Furthermore, potentiation of differentiation by carnosic acid with 1,25(OH)2D3 (10−9 M) did not lead to the increase in intracellular calcium concentrations as compared to when the cells were treated with 10−7 M 1,25(OH)2D3 alone, although differentiation levels were equivalent. Combining vitamin D compounds with traditional chemotherapy agents such as cisplatin, etoposide, and doxorubicin has been shown to reduce the concentrations required for their anti-leukemic activities [85, 86]. Studies in other cancers have also shown that a vitamin D compound combined with a chemotherapeutic agent was more effective than either agent alone [87–89]. A recent phase I clinical trial has demonstrated the therapeutic potential of this method, as very high doses of 1,25(OH)2D3 could be given orally with paclitaxel [90]. Another novel approach to circumvent the calcemic side effects of vitamin D compounds has been to give high doses of 1,25(OH)2D3 subcutaneously every other day [91]. In a phase I clinical trial, the maximum tolerated dose administered in this fashion was five times greater than the daily oral dose that had previously been shown to cause hypercalcemia [91].

V. VITAMIN D ANALOGS EFFECTIVE AGAINST LEUKEMIC CELLS The first attempts using analogs focused on the compound 1α-hydroxyvitamin D3 (1αOHD3), a vitamin D3 analog that is efficiently converted to 1,25(OH)2D3 in vivo by D3-25-hydroxylase. This compound was administered to mice previously inoculated with the Ml leukemia cell line and showed greater activity than 1,25(OH)2D3 [92]. Its conversion to the active form resulted in a more prolonged elevation of plasma levels of 1,25(OH)2D3, and the dose (25 pmol, every other day) produced only a slight and not significant elevation of serum calcium. In addition, survival of the leukemic mice was increased by 50–60%; however, the more effective doses produced hypercalcemia. Also, the

administration of 1αOHD3 produced tumor regression in follicular NHLs in rats, but hypercalcemia was the dose-limiting factor [20]. In one study, six patients with MDS were treated with 1αOHD3 at 1 µg/day for a minimum of three months, but neither a good clinical response nor toxicity was observed in these cases [93]. In another clinical study, thirty MDS patients were included in two different groups: one group received 1α-OHD3 at 4–6 µg/day and the other group received placebo; the patients were treated for a median of 17 weeks [94]. An improvement of hematologic parameters was detected in only one patient, but the authors felt the treated group had a greater proportion of patients who did not progress to leukemia as compared to the control group. Hypercalcemia and increased serum creatinine were observed in two patients, and these abnormal measurements regressed with reduction of the dose [94]. A case has been reported of an individual with chronic myelomonocytic leukemia (subtype of MDS) who achieved complete remission with 25-hydroxyvitamin D3 therapy for 15 months; this remission was sustained for 15 months after the end of the treatment [95]. These results are surprising because 25OHD3 has low activity by itself and in vitro has little anti-leukemic activity. However, as substrate for 1α-hydroxylase it may lead to local production of 1,25(OH)2D3 (see Chapter 79). Calcipotriol (MC903) has a cyclopropyl group at the end of the side chain formed by the fusion of C-26 and C-27, a hydroxyl group at C-24, and a double bond at C-22. This compound was equipotent to 1,25(OH)2D3 in inhibiting the proliferation and inducing the differentiation of the monoblastic cell line U937 [96,97]. In bone marrow cultures, the analog promoted the formation of multinucleated osteoclastlike cells, a vitamin D–mediated function. The effects of this compound on the immune system were very similar to those induced by 1,25(OH)2D3. By interfering with T-helper cell activity, calcipotriol reduced immunoglobulin production and blocked the proliferation of thymocytes induced by IL-1 [98,99]. Exposure of the follicular NHL B-cell lines SU-DUL4 and SUDUL5, carrying the t(14;18) translocation characteristic of the disease, to MC903 inhibited of proliferation only at high concentrations of the compound (10 −7 M) [20]. At the same time, calcipotriol was 100-fold less active than 1,25(OH)2D3 in inducing hypercalcemia and mobilizing bone calcium in rats [100]. However, the analog is rapidly inactivated in the intact animal, and therefore has been developed as a topical agent (see Chapter 101). Introduction of a double bond at carbon 16 has proved to be an effective modification of 1,25(OH)2D3 [101].

CHAPTER 96 Vitamin D and Hematological Malignancy

When combined with other motifs the 16 ene modification has led to a series of analogs with potent antiproliferative and differentiation-promoting activities with decreased calcemic effects. Prior studies by us have shown that vitamin D3 analogs having the C-16-ene motif were almost 100-fold more potent than 1,25(OH)2D3 in inhibiting growth of HL-60 leukemia cells, while the calcemic activity was the same or markedly less than 1,25(OH)2D3 [102,103]. Combination of the C-16-double bond and the C-23-triple bond [1,25(OH)2-16-ene-23yne-D3] produces a compound that is a more potent inducer of growth inhibition and differentiation in HL-60 cells than 1,25(OH)2D3, and is 15 times less hypercalcemic in mice [102]. The analog l,25(OH)2-16-ene-23yne D3 has potent antiproliferative and differentiating effects on leukemic cells in vitro [104]. In blocking HL60 clonal growth, l,25(OH)2-16-ene-23-yne D3 has a potency about four times higher than 1,25(OH)2D3. This compound administered to vitamin D–deficient chicks is about 30 times less effective than 1,25(OH)2D3 in stimulating intestinal calcium absorption and about 50 times less effective in inducing bone calcium mobilization [97]. Further experiments have demonstrated the therapeutic potential of l,25(OH)2-16-ene-23-yne D3 by its ability to prolong markedly the survival of mice inoculated with the myeloid leukemic cell line WEHI 3BD+ when treated with a high dose (1.6 µg every other day) of the compound [105]. The 1,25(OH)2-16-ene-19-nor-24-oxo-D3 was synthesized as a result of previous studies that isolated 24-oxo metabolites of potent vitamin D3 analogs, which were formed in a rat kidney perfusion system [106]. We found that these 24-oxo-metabolites had markedly reduced calcemic activity, but possessed at least an equal ability as the unmetabolized analogs to inhibit the clonal growth of breast and prostate cancer cells and myeloid leukemia cells in vitro. Taken together, these findings prompted the chemical synthesis of a series of vitamin D3 analogs with 1,25(OH)2-16ene-19-nor-24-oxo-D3 being one of the more exciting compounds, having the ability to inhibit acute myeloid leukemia cells in the concentration range of 10−10 M [107]. Remarkably, this compound had very little calcemic activity even when 6 µg was administered intraperitoneally to the mice three times a week [107]. The compound l,25(OH)2-20-epi D3 is characterized by an inverted stoichiometry at C-20 of the side chain. The monoblastic cell line U937 cultured with this compound showed a strong induction of differentiation [108]. It was also a potent modulator of cytokinemediated T-lymphocyte activation and exerted calcemic effects comparable to 1,25(OH)2D3 in rats. A study by us suggested that l,25(OH)2-20-epi D3 is the most potent vitamin D3 compound at inhibiting the clonal

1735 growth of HL-60 cells and at inducing cell differentiation. In fact, it was about 2600-fold more potent than 1,25(OH)2D3 in inhibiting the clonal growth of HL-60 cells and about 5000-fold more effective in preventing clonal growth of fresh human leukemic myeloid cells [109]. 1,25(OH)2-20-epi D3 exerts its effects by binding directly to VDR as shown by a T-lymphocytic cell line established from a patient with vitamin D–dependent rickets type II (HVDRR) lacking a functional VDR. Clonal growth was not inhibited after treatment of these cells with high doses of either 1,25(OH)2-20epi D3 or 1,25(OH)2D3 (10−7 M). In contrast, control experiments showed that these compounds [1,25(OH)220-epi D3 > 1,25(OH)2D3] were powerful inhibitors of proliferation of a human T-cell leukemia virus type I (HTLV-I) transformed T-cell line that possessed VDR. KH1060 is a potent vitamin D3 20-epi analog with an oxygen in place of C-22 and three additional carbons in the side chain. It is about 14,000-fold more potent than 1,25(OH)2D3 in inhibiting the clonal growth of the monoblastic cell line U937 [108]. It also has a powerful effect on other leukemic cells [70,108, 109]. However, it has the same hypercalcemic activity and the same receptor binding affinity as 1,25(OH)2D3. One promising new vitamin D analog is paricalcitol (19-nor-1,25-dihydroxyvitamin D2), which has been approved by the Food and Drug Administration for the clinical treatment of secondary hyperparathyroidism. Clinical trials have demonstrated that it possesses very low calcemic activity [110,111]. Studies by us and another group have demonstrated that paricalcitol has antiproliferative, prodifferentiation activities against myeloid leukemia and myeloma cell lines [31,112]. Paricalcitol activity was dependent on the presence of VDR, as it was unable to induce differentiation of mononuclear bone marrow cells from VDR knockout mice, whereas cells from wild-type mice were differentiated towards monocytes/macrophages [31]. Furthermore, paracalcitol was able to inhibit tumor growth without causing hypercalcemia in nude mice. These observations have prompted us to begin a clinical trial of paricalcitol aimed at treating patients with MDS. Potential mechanisms by which vitamin D analogs may have increased biological activity compared to 1,25(OH)2D3 are: reduced affinity to the serum vitamin D–binding protein; decreased catabolism by 24-hydroxylase; retention of biological activities by metabolic products of vitamin D analogs; increased stability of the ligand-VDR complex; increased VDR DNA–binding and dimerization with RXR; and enhanced recruitment of the DRIP coactivator complex. These topics are covered in detail in Chapters 81–83. In conclusion, new vitamin D analogs have potent anti-leukemic activity and lower hypercalcemic

1736

JAMES O’KELLY, ROBERTA MOROSETTI, AND H. PHILLIP KOEFFLER

effects than 1,25(OH)2D3, and should be considered for the treatment of hematologic malignancies either alone or in combination with other differentiating agents. However, more phase I, II, and III trials are still necessary to assess the safety and effectiveness of these treatments.

VI. SUMMARY AND CONCLUSIONS The hormone 1,25(OH)2D3 plays a role in normal hematopoiesis, enhancing the activity of monocytesmacrophages and inhibiting cytokine production by T lymphocytes. It can also inhibit proliferation and induce differentiation of various myeloid leukemia cell lines. Its activity is mediated by vitamin D receptors that belong to the superfamily of steroid-thyroid receptors. However, the anti-leukemic activity of 1,25(OH)2D3 in vivo is associated with high toxicity and the onset of hypercalcemia as the dose-limiting effect. Limited clinical trials have been performed for the treatment of preleukemia with differentiating agents including 1,25(OH)2D3, but the in vitro effective dose caused hypercalcemia in vivo. Since the mid-1980s, many vitamin D analogs have been identified with reduced hypercalcemic activity and high potential to induce cell differentiation and to inhibit proliferation of leukemic cells. Further studies have been performed in vitro and in vivo using these analogs with other differentiating agents such as retinoids, in the hopes that the combination of agents working through different pathways could lead to synergistic activity. Proof of principle that 1,25(OH)2D3 and its analogs are beneficial in cancer has occurred in experiments conducted in vitro and in laboratory animals; however, the results of currently ongoing and future clinical trials in patients using vitamin D analogs will determine their ultimate therapeutic value.

References 1. Mangelsdorf D, Thummel C, Beato M, Herrlich P, Schutz G, Umesono K, Blumberg B, Kastner P, Mark M, Chambon P 1995 The nuclear receptor superfamily: The second decade. Cell 83:835–839. 2. Christakos S, Raval-Pandya M, Wernyj R, Yang W 1996 Genomic mechanisms involved in the pleiotropic actions of 1,25-dihydroxyvitamin D3. Biochem J 316:361–371. 3. Kizaki M, Norman A, Bishop J, Lin C, Karmakar A, Koeffler H 1991 1,25-dihydroxyvitamin D3 receptor RNA: expression in hematopoietic cells. Blood 77:1238–1247. 4. Provvedini D, Deftos L, Manolagas S 1986 1,25-dihydroxyvitamin D3 promotes in vitro morphologic and enzymatic changes in normal human monocytes consistent with their differentiation into macrophages. Bone 7:23–28.

5. Zerwekh J, Yu X, Breslau N, Manolagas S, Pak C 1993 Vitamin D receptor quantitation in human blood mononuclear cells in health and disease. Mol Cell Endocrinol 96:1–6. 6. Manolagas S, Werntz D, Tsoukas C, Provvedini D, Vaughan J 1986 1,25-dihydroxyvitamin D3 receptors in lymphocytes from patients with rheumatoid arthritis. J Lab Clin Med 108:596–600. 7. Provvedini D, Ruiot C, Sobol R, Tsoukas C, Manolagas S 1987 1,25-dihydroxyvitamin D3 receptors in human thymic and tonsillar lymphocytes. J Bone Miner Res 2: 239–247. 8. Biyoudi-Vouenze R, Cadranel J, Valeyre D, Milleron B, Hance A, Soler P 1991 Expression of 1,25(OH)2D3 receptors on alveolar lymphocytes from patients with pulmonary granulomatous diseases. Am Rev Respir Dis 143:1376–1380. 9. Koren R, Ravid A, Liberman U, Hochberg Z, Weisman Y, Novogrodsky A 1985 Defective binding and function of 1,25-dihydroxyvitamin D3 receptors in peripheral mononuclear cells of patients with end-organ resistance to 1,25dihydroxyvitamin D. J Clin Invest 76:2012–2015. 10. Nakajima S, Yamaoka K, Yamamoto T, Okada S, Tanaka H, Seino Y 1990 Decreased concentration of 1,25-dihydroxyvitamin D3 receptors in peripheral mononuclear cells of patients with X-linked hypophosphatemic rickets: effect of phosphate supplementation. Bone Miner 10:201–209. 11. Iho S, Takahashi T, Kura F, Sugiyama H, Hoshino T 1986 The effect of 1,25-dihydroxyvitamin D3 on in vitro immunoglobulin production in human B cells. J Immunol 136:4427–4431. 12. Lemire J, Adams J, Kermani-Arab V, Bakke A, Sakai R, Jordan S 1985 1,25-dihydroxyvitamin D3 suppresses human T helper/inducer lymphocyte activity in vitro. J Immunol 134:3032–3035. 13. Rigby W, Stacy T, Fanger M 1984 Inhibition of T lymphocyte mitogenesis by 1,25-dihydroxyvitamin D3 (calcitriol). J Clin Invest 74:1451–1455. 14. Tsoukas C, Provvedini D, Manolagas S 1984 1,25-dihydroxyvitamin D3: a novel immunoregulatory hormone. Science 224:1438–1440. 15. Reichel H, Koeffler H, Tobler A, Norman A 1987 1α,25dihydroxyvitamin D3 inhibits gamma-interferon synthesis by normal human peripheral blood lymphocytes. Proc Natl Acad Sci USA 84:3385–3389. 16. Tobler A, Miller C, Norman A, Koeffler H 1988 1,25-dihydroxyvitamin D3 modulates the expression of a lymphokine (granulocyte-macrophage colony-stimulating factor) posttranscriptionally. J Clin Invest 81:1819–1823. 17. O’Kelly J, Hisatake J, Hisatake Y, Bishop J, Norman A, Koeffler H 2002 Normal myelopoiesis but abnormal T lymphocyte responses in vitamin D receptor knockout mice. J Clin Invest 109:1091–1099. 18. Lee Y, Inaba M, DeLuca H, Mellon W 1989 Immunological identification of 1,25-dihydroxyvitamin D3 receptors in human promyelocytic leukemic cells (HL-60) during homologous regulation. J Biol Chem 264:13701–13705. 19. Provvedini D, Tsoukas C, Deftos L, Manolagas S 1983 1,25dihydroxyvitamin D3 receptors in human leukocytes. Science 221:1181–1183. 20. Hickish T, Cunningham D, Colston K, Millar B, Sandle J, Mackay A, Soukop M, Sloane J 1993 The effect of 1,25dihydroxyvitamin D3 on lymphoma cell lines and expression of vitamin D receptor in lymphoma. Br J Cancer 68: 668–672. 21. Lasky S, Iwata K, Rosmarin A, Caprio D, Maizel A 1995 Differential regulation of JunD by dihydroxycholecalciferol

1737

CHAPTER 96 Vitamin D and Hematological Malignancy

22. 23.

24.

25. 26. 27.

28.

29.

30.

31.

32.

33. 34.

35. 36.

37.

in human chronic myelogenous leukemia cells. J Biol Chem 270:19676–19679. Gaynor R, Simon K, Koeffler P 1991 Expression of c-jun during macrophage differentiation of HL-60 cells. Blood 77:2618–2623. Abe E, Miyaura C, Sakagami H, Takeda M, Konno K, Yamazaki T, Yoshiki S, Suda T 1981 Differentiation of mouse myeloid leukemia cells induced by 1α, 25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 78:4990–4494. Koeffler H, Amatruda T, Ikekawa N, Kobayashi Y, DeLuca H 1984 Induction of macrophage differentiation of human normal and leukemic myeloid stem cells by 1,25-dihydroxyvitamin D3 and its fluorinated analogs. Cancer Res 12: 5624–5628. Paquette R Koeffler H 1992 Differentiation therapy. In: Myelodysplastic syndromes. Hematol/Oncol Clin North A 6:687–706. Kreutz M, Andreesen R 1990 Induction of human monocyte to macrophage maturation in vitro by 1,25-dihydroxyvitamin D3. Blood 76:2457–2461. Choudhuri U, Adams J, Byrom N, McCarthy D, Barrett J 1990 1,25-dihydroxyvitamin D3 induces normal mononuclear blood cells to differentiate in the direction of monocytemacrophages. Haematologia 23:9–19. Tobler A, Gasson J, Reichel H, Norman A, Koeffler H 1987 Granulocyte-macrophage colony-stimulating factor. Sensitive and receptor-mediated regulation by 1,25-dihydroxyvitamin D3 in normal human peripheral blood lymphocytes. J Clin Invest 79:1700–1705. Richard C, Mazo E, Cuadrado M, Iriondo A, Bello C, Ganda-rillas M, Zubizarreta A 1986 Treatment of myelodysplastic syndrome with 1,25-dihydroxyvitamin D3. Am J Hematol 73:175–178. Zhou J, Norman A, Lubbert M, Collins E, Uskokovic M, Koeffler H 1989 Novel vitamin D analogs that modulate leukemic cell growth and differentiation with little effect on either intestinal calcium absorption or bone mobilization. Blood 74:82–93. Kumagai T, O’Kelly J, Said J, Koeffler H 2003 Vitamin D2 analog 19-nor-1,25-dihydroxyvitamin D2: Antitumor activity against leukemia, myeloma, and colon cancer cells. J Natl Cancer Inst 76:2012–2015. Hisatake J, O’Kelly J, Uskokovic M, Tomayasu S, Koeffler H 2001 Novel vitamin D3 analog, 21-(3-methyl-3-hydroxybutyl)-19-nor-D3, that modulates cell growth, differentiation, apoptosis, cell cycle, and induction of PTEN in leukemic cells. Blood 97:2427–2433. Godyn J, Xu H, Zhang F, Kolla S, Studzinski G 1994 A dual block to cell cycle progression in HL60 cells exposed to analogs of vitamin D3. Cell Prolif 27:37–46. Mangelsdorf D, Koeffler H, Donaldson C, Pike J, Haussler M 1984 1,25-dihydroxyvitamin D3-induced differentiation in a human promyelocytic leukemia cell line (HL-60): receptor-mediated maturation to macrophage-like cells. J Cell Biol 98:391–398. Koeffler H, Hirji K, Itri L 1985 1,25-dihydroxyvitamin D3: in vivo and in vitro effects on human preleukemic and leukemic cells. Cancer Treat Rep 69:1399–1407. Campbell M, Drayson M, Durham J, Wallington L, Siu-Caldera M, Reddy G, Brown G 1999 Metabolism of 1α,25(OH)2D3 and its 20-epi analog integrates clonal expansion, maturation and apoptosis during HL-60 cell differentiation. Mol Cell Endocrinol 149:169–183. Rots N, Iavarone A, Bromleigh V, Freedman L 1999 Induced differentiation of U937 cells by 1,25-dihydroxyvitamin D3

38.

39.

40.

41.

42. 43.

44.

45. 46.

47.

48.

49.

50.

51.

52.

involves cell cycle arrest in G1 that is preceded by a transient proliferative burst and an increase in cyclin expression. Blood 93:2721–2729. Schwaller J, Koeffler H, Niklaus G, Loetscher P, Nagel S, Fey M, Tobler A 1995 Posttranscriptional stabilization underlies p53-independent induction of p21WAF1/C1P1/ SD11 in differentiating human leukemic cells. J Clin Invest 95:973–979. Liu M, Lee M, Cohen M, Bommakanti M, Freedman L 1996 Transcriptional activation of the Cdk inhibitor p21 by vitamin D3 leads to the induced differentiation of the myelomonocytic cell line U937. Genes Dev 10:142–153. Wang Q, Jones J, Studzinski G 1996 Cyclin-dependent kinase inhibitor p27 as a mediator of the G1-S phase block induced by 1,25-dihydroxyvitamin D3 in HL-60 cells. Cancer Res 56:264–267. Obeid L, Okazaki T, Karolak L, Hannun Y 1990 Transcriptional regulation of protein kinase C by 1,25dihydroxyvitamin D3 in HL-60 cells. J Biol Chem 265: 2370–2374. Dalla-Favera R, Wong-Staal F, Gallo R 1982 Onc gene amplification in promyelocytic leukemia cell line HL-60 and primary leukemic cells of the same patient. Nature 299:61–63. Reitsma P, Rothberg P, Astrin S, Trial J, Bar-Shavit Z, Hall A, Teitelbaum S, Kahn A 1983 Regulation of myc gene expression in HL-60 leukemia cells by a vitamin D metabolite. Nature 306:. Simpson R, Hsu T, Begley D, Mitchell B, Alizadeh B 1987 Transcriptional regulation of the c-myc protooncogene by 1,25-dihydroxyvitamin D3 in HL-60 promyelocytic leukemia cells. J Biol Chem 262:4104–4108. Pan Q, Martell R, O’Connell T, Simpson R 1996 1,25dihydroxyvitamin D3–regulated binding of nuclear proteins to a c-myc intron element. Endocrinology 137:4154–4160. Pan Q Simpson R 1999 c-myc intron element-binding proteins are required for 1,25-dihydroxyvitamin D3 regulation of c-myc during HL-60 cell differentiation and the involvement of HOXB4. J Biol Chem 274:8437–8444. Rots N, Liu M, Anderson E, Freedman L 1998 A differential screen for ligand-regulated genes: identification of HoxA10 as a target of vitamin D3 induction in myeloid leukemic cells. Mol Cell Biol 18:1911–1918. Wu Y, Jiang X, Lillington D, Allen P, Newland A, Kelsey S 1998 1,25-dihydroxyvitamin D3 protects human leukemic cells from tumor necrosis factor-induced apoptosis via inactivation of cytosolic phospholipase A2. Cancer Res 58:633–640. Xu H, Tepper C, Jones J, Fernandez C, Studzinski G 1993 1,25-dihydroxyvitamin D3 protects HL60 cells against apoptosis but down-regulates the expression of the bcl-2 gene. Exp Cell Res 209:367–374. Rowley P, Farley B, Giuliano R, LaBella S, Leary J 1992 Induction of the fms proto-oncogene product in HL-60 cells by vitamin D: a flow cytometric analysis. Leuk Res 16: 403–410. Ruthardt M, Testa U, Nervi C, Ferrucci P, Grignani F, Puccetti E, Grignani F, Peschle C, Pelicci P 1997 Opposite effects of the acute promyelocytic leukemia PML-retinoic acid receptor alpha (RAR alpha) and PLZF-RAR alpha fusion proteins on retinoic acid signaling. Mol Cell Biol 17:4859–4869. Puccetti E, Obradovic D, Beissert T, Bianchini A, Washburn BFC, Boehrer S, Hoelzer D, Ottmann O, Pelicci P, Nervi C, Ruthardt M 2002 AML-associated translocation products block vitamin D3–induced differentiation by sequestering the vitamin D3 receptor. Cancer Res 62:7050–7058.

1738

JAMES O’KELLY, ROBERTA MOROSETTI, AND H. PHILLIP KOEFFLER

53. Bories D, Raynal M, Solomon D, Darzynkiewicz Z, Cayre Y 1989 Down-regulation of a serine protease, myeloblastin, causes growth arrest and differentiation of promyelocytic leukemia cells. Cell 59:959–968. 54. Savli H, Aalto Y, Nagy B, Knuutila S, Pakkala S 2002 Gene expression analysis of 1,25(OH)2D3–dependent differentiation of HL-60 cells: a cDNA array study. Br J Haematol 118:1065–1070. 55. Hmama Z, Nandan D, Sly L, Knutson K, Herrera-Velit P, Reiner N 1999 1α,25-dihydroxyvitamin D3–induced myeloid cell differentiation is regulated by a vitamin D receptorphosphatidylinositol 3-kinase signaling complex. J Exp Med 190:1583–1594. 56. Marcinkowska E, Wiedlocha A, Radzikowski C 1998 Evidence that phosphatidylinositol 3-kinase and p70S6K protein are involved in differentiation of HL-60 cells induced by calcitriol. Anticancer Res 18:3507–3514. 57. Marcinkowska E, Wiedlocha A, Radzikowski C 1997 1,25dihydroxyvitamin D3–induced activation and subsequent nuclear translocation of MAPK is up-stream regulated by PKC in HL-60 cells: Biochem Biophys Res Commun 241:419–426. 58. Song X, Bishop J, Okamura W, Norman A 1998 Stimulation of phosphorylation of mitogen-activated protein kinase by 1α,25-dihydroxyvitamin D3 in promyelocytic NB4 leukemia cells: a structure-function study. Endocrinology 139: 457–465. 59. Ji Y, Kutner A, Verstuyf A, Verlinden L, Studzinski G 2002 Derivatives of vitamins D2 and D3 activate three MAPK pathways and up-regulate pRb expression in differentiating HL60 cells. Cell cycle 1:410–415. 60. Pepper C, Thomas A, Hoy T, Milligan D, Bentley P, Fegan C 2002 The vitamin D3 analog EB1089 induces apoptosis via a p53-independent mechanism involving p38 MAP kinase activation and suppression of ERK activity in B-cell chronic lymphocytic leukemia cells in vitro. Blood 101:2454–2460. 61. Wang X, Studzinski G 2001 Inhibition of p38MAP kinase potentiates the JNK/SAPK pathway and AP-1 activity in monocytic but not in macrophage or granulocytic differentiation of HL60 cells. J Cell Biochem 82:68–77. 62. Marcinkowska E 2001 Evidence that activation of MEK1,2/erk1,2 signal transduction pathway is necessary for calcitriol-induced differentiation of HL-60 cells. Anticancer Res 21:499–504. 63. Koeffler H, Bar-Eli M, Territo M 1981 Phorbol ester effect on differentiation of human myeloid leukemia cell lines blocked at different stages of maturation. Cancer Res 41:919–926. 64. Tonetti D, Henning-Chubb C, Yamanishi D, Huberman E 1994 Protein kinase C-beta is required for macrophage differentiation of human HL-60 leukemia cells. J Biol Chem 269:23230–23235. 65. Pan Q, Granger J, O’Connell T, Somerman M, Simpson R 1997 Promotion of HL-60 cell differentiation by 1,25-dihydroxyvitamin D3 regulation of protein kinase C levels and activity. Biochem Pharmacol 54:909–915. 66. Berry D, Meckling-Gill, K 1999 Vitamin D analogs, 20-Epi-22-oxa-24a,26a,27a,-trihomo-1α,25(OH)2-vitamin D3, 1,24(OH)2-22-ene-24-cyclopropyl-vitamin D3 and 1alpha, 25(OH)2-lumisterol3 prime NB4 leukemia cells for monocytic differentiation via nongenomic signaling pathways, involving calcium and calpain. Endocrinology 140:4779–4788. 67. Koeffler H, Bishop J, Reichel H, Singer F, Nagler A, Tobler A, Walka M, Norman A 1991 Lymphocyte cell lines

68.

69.

70.

71.

72.

73. 74.

75. 76.

77.

78.

79. 80.

from vitamin D–dependent rickets type II show functional defects in the 1α,25-dihydroxyvitamin D3 receptor. Mol Cell Endocrinol 70:1–11. Nagler A, Merchav S, Fabian I, Tatarsky I, Weisman Y, Hochberg Z 1987 Myeloid progenitors from the bone marrow of patients with vitamin D–resistant rickets (type II) fail to respond to 1,25(OH)2D3. Br J Haematol 67:267–271. Masciulli R, Testa U, Barberi T, Samoggia P, Tritarelli E, Pustorino R, Mastroberardino G, Camagna A, Peschle C 1995 Combined vitamin D3/retinoic acid induction of human promyelocytic cell lines: enhanced phagocytic cell maturation and hybrid granulomonocytic phenotype. Cell Growth Differ 6:493–503. Elstner E, Linker-Israeli M, Umiel T, Le J, Grillier I, Said J, Shintaku I, Krajewski S, Reed J, Binderup L, Koeffler H 1996 Combination of a potent 20-epi-vitamin D3 analog (KH 1060) with 9-cis-retinoic acid irreversibly inhibits clonal growth, decreases bcl-2 expression, and induces apoptosis in HL-60 leukemic cells. Cancer Res 56:3570–3576. Brown G, Bunce C, Rowlands D, Williams G 1994 All-trans retinoic acid and 1α,25-dihydroxyvitamin D3 cooperate to promote differentiation of the human promyeloid leukemia cell line. Leukemia 8:806–815. Bunce C, Wallington L, Harrison P, Williams G, Brown G 1995 Treatment of HL60 cells with various combinations of retinoids and 1α,25 dihydroxyvitamin D3 results in differentiation towards neutrophils or monocytes or a failure to differentiate and apoptosis. Leukemia 9:410–418. Dore B, Uskokovic M, Monparler R 1993 Interaction of retinoic acid and vitamin D3 analogs on HL-60 myeloid leukemic cells. Leuk Res 17:749–757. Muto A, Kizaki M, Yamato K, Kawai Y, Kamata-Matsushita M, Ueno H, Ohguchi M, Nishihara T, Koeffler H, Ikeda Y 1999 1,25-dihydroxyvitamin D3 induces differentiation of a retinoic acid-resistant acute promyelocytic leukemia cell line (UF-1) associated with expression of p21(WAF1/CIP1) and p27(KIP1). Blood 93:2225–2233. Hu Z, Ma W, Uphoff C, Lanotte M, Drexler H 1993 Modulation of gene expression in the acute promyelocytic leukemia cell line NB4. Leukemia 7:1817–1823. Elstner E, Linker-Israeli M, Le J, Umiel T, Michl P, Said J, Binderup L, Reed J, Koeffler H 1997 Synergistic decrease of clonal proliferation, induction of differentiation, and apoptosis of acute promyelocytic leukemia cells after combined treatment with novel 20-epi vitamin D3 analogs and 9-cis retinoic acid. J Clin Invest 99:349–360. Bhatia M, Kirkland J, Meckling-Gill K 1994 M-CSF and 1,25 dihydroxyvitamin D3 synergize with 12-O-tetradecanoylphorbol-13-acetate to induce macrophage differentiation in acute promyelocytic leukemia NB4 cells. Leukemia 8:1744–1749. Taimi M, Chateau M, Cabane S, Marti J 1991 Synergistic effect of retinoic acid and 1,25-dihydroxyvitamin D3 on the differentiation of the human monocytic cell line U937. Leuk Res 15:1145–1152. Cellier M, Taimi M, Chateau M, Cannat A, Marti J 1993 Thermal stress as an inducer of differentiation of U937 cells. Leuk Res 17:649–656. Bunce C, French P, Durham J, Stockley R, Michell R, Brown G 1994 Indomethacin potentiates the induction of HL60 differentiation to neutrophils, by retinoic acid and granulocyte colony-stimulating factor, and to monocytes, by vitamin D3. Leukemia 8:595–604.

CHAPTER 96 Vitamin D and Hematological Malignancy

81. Sokoloski J, Sartorelli A 1998 Induction of the differentiation of HL-60 promyelocytic leukemia cells by nonsteroidal anti-inflammatory agents in combination with low levels of vitamin D3. Leuk Res 22:153–161. 82. Desmond J, Mountford J, Drayson M, Walker E, Hewison M, Ride J, Luong Q, Hayden R, Vanin E, Bunce C 2003 The aldo-keto reductase AKR1C3 is a novel suppressor of cell differentiation that provides a plausible target for the noncyclooxygenase-dependent antineoplastic actions of nonsteroidal anti-inflammatory drugs. Cancer Res 63: 505–512. 83. Danilenko M, Wang X, Studzinski G 2001 Carnosic acid and promotion of monocytic differentiation of HL60-G cells initiated by other agents. J Natl Cancer Inst 93: 1224–1233. 84. Danilenko M, Wang Q, Wang X, Levy J, Sharoni Y, Studzinski G 2003 Carnosic acid potentiates the antioxidant and prodifferentiation effects of 1α,25-dihydroxyvitamin D3 in leukemia cells but does not promote elevation of basal levels of intracellular calcium. Cancer Res 63: 1325–1332. 85. Torres R, Calle C, Aller P, Mata F 2000 Etoposide stimulates 1,25-dihydroxyvitamin D3 differentiation activity, hormone binding and hormone receptor expression in HL-60 human promyelocytic cells. Mol Cell Biochem 208:157–162. 86. Siwinska A, Opolski A, Chrobak A, Wietrzyk J, Wojdat E, Kutner A, Szelejewski W, Radzikowski C 2001 Potentiation of the antiproliferative effect in vitro of doxorubicin, cisplatin, and genistein by new analogs of vitamin D. Anticancer Res 21:1925–1929. 87. Koshizuka K, Koike M, Kubota T, Said J, Binderup L, Koeffler H 1998 Novel vitamin D3 analog (CB1093) when combined with paclitaxel and cisplatin inhibit growth of MCF-7 human breast cancer cells in vivo. Int J Oncol 13:421–428. 88. Koshizuka K, Koike M, Asou H, Cho S, Stephen T, Rude R, Binderup L, Uskokovic M, Koeffler H 1999 Combined effect of vitamin D3 analogs and paclitaxel on the growth of MCF7 breast cancer cells in vivo. Breast Cancer Res Treat 53:113–120. 89. Yu W, McElwain M, Modzelewski R, Russell D, Smith D, Trump D, Johnson C 1998 Enhancement of 1,25-dihydroxyvitamin D3–mediated anti-tumor activity with dexamethasone. J Natl Cancer Inst. 90:134–141. 90. Muindi J, Peng Y, Potter D, Hershberger P, Tauch J, Capozzoli M, Egorin M, Johnson C, Trump D 2002 Pharmacokinetics of high-dose oral calcitriol: results from a phase 1 trial of calcitriol and paclitaxel. Clin Pharmacol Ther 72:648–659. 91. Smith D, Johnson C, Freeman C, Muindi J, Wilson J, Trump D 1999 A phase I trial of calcitriol (1,25-dihydroxycholecalciferol) in patients with advanced malignancy. Clin Cancer Res 5:1339–1345. 92. Honma Y, Hozumi M, Abe E, Konno K, Fuku S, Hima M, Hata S, Nishii Y, DeLuca H, Suda T 1983 1,25-dihydroxyvitamin D3 prolong survival time mice inoculated with myeloid leukemia cells. Proc Natl Acad Sci USA 80: 201–204. 93. Metha A, Kumaran T, Marsh G 1984 Treatment of myelodysplastic syndrome with alfacalcidol. Lancet 2:761. 94. Motomura S, Kanamori H, Maruta A, Kodama F, Ohkubo T 1991 The effect of 1-hydroxyvitamin D3 for prolongation of leukemic transformation-free survival in myelodysplastic syndromes. Am J Hematol 38:67–68.

1739 95. Mellibovsky L, Diez A, Aubia J, Nogues X, Perez-Vila E, Serrano S, Recker R 1993 Long-standing remission after 25-OH-D3 treatment in a case of chronic myelomonocytic leukemia. Br J Haematol 85:811–812. 96. Binderup L, Bramm E 1988 Effects of a novel vitamin D analogue MC903 on cell proliferation and differentiation in vitro and on calcium metabolism in vivo. Biochem Pharmacol 37:889–895. 97. Brown A, Dusso A, Slatopolsky E 1994 Selective vitamin D analogs and their therapeutic applications. Semin Nephrol 14:156–174. 98. Muller K, Svenson M, Bendtzen K 1988 1,25Dihydroxyvitamin D3 and a novel vitamin D analog MC903 are potent inhibitors of human interleukin 1 in vitro. Immunol Lett 17:361–366. 99. Muller K, Heilmann C, Poulsen L, Barington T, Bendtzenk K 1991 The role of monocytes and T-cells in 1,25-dihydroxyvitamin D3–mediated inhibition of B-cell function in vitro. Immunopharmacology 21:121–128. 100. Rebel V, Ossenkoppele G, van de Loosdrecht A, Wijermans P, Beelen R, Langenhuijsen M 1992 Monocytic differentiation induction of HL-60 cells by MC 903, a novel vitamin D analog. Leuk Res 16:443–451. 101. Uskokovic M, Stuzinski G, Gardner J, Reddy S, Campbell M, Koeffler H 1997 The 16-ene vitamin D analogs. Curr Pharm Design 3:99–123. 102. Jung S, Lee Y, Pakkala S, de Vos S, Elsner E, Norman A, Green J, Uskokovic M, Koeffler H 1996 1,25-(OH)2-16-enevitamin D3 is a potent antileukemic agent with low potential to cause hypercalcemia. Leuk Res 18:453–463. 103. Hisatake J, Kubota T, Hisatake Y, Uskokovic M, Tomoyasu S, Koeffler H 1999 5,6-trans-16-ene-vitamin D3: a new class of potent inhibitors of proliferation of prostate, breast, and myeloid leukemic cells. Cancer Res 59:4023–4029. 104. Pakkala S, de Vos S, Elstner E, Ruder K, Uskokovic M, Binderup L, Koeffler H 1995 Vitamin D3 analogs: Effect on leukemic clonal growth and differentiation, and on serum calcium levels. Leuk Res 19:65–72. 105. Zhou J, Norman A, Chen D, Sun G, Uskokovic M, Koeffler H 1990 1,25(OH)2-16ene-23yne vitamin D3 prolongs survival time of leukemic mice. Proc Natl Acad Sci USA 87:3929–3932. 106. Campbell M, Reddy G, Koeffler H 1997 Vitamin D3 analogs and their 24-oxo metabolites equally inhibit clonal proliferation of a variety of cancer cells but have different molecular effects. J Cell Biochem 66:413–425. 107. Shiohara M, Uskokovic M, Hisitake J, Hisatake Y, Koike K, Komiyama A, Koeffler H 2001 24-oxo metabolites of vitamin D3 analogs: Disassociation of their prominent antileukemic effects from their lack of calcium modulation. Cancer Res 61:3361–3368. 108. Binderup L, Latini S, Binderup E, Bretting C, Calverley M, Hansen K 1991 20-epi-vitamin D3 analogs: A novel class of potent regulators of cell growth and immune response. Biochem Pharmacol 42:1569–1575. 109. Elstner E, Lee Y, Hashiya M, Pakkala S, Binderup L, Norman A, Okamura W, Koeffler H 1994 l,25-dihydroxy-20epi-vitamin D3: An extraordinarily potent inhibitor of leukemic cell growth in vitro. Blood 84:1960–1967. 110. Llach F, Keshav G, Goldblat M, Lindberg J, Sadler R, Delmez J, Arruda J, Lau A, Slatopolsky E 1998 Suppression of parathyroid hormone secretion in hemodialysis patients by a novel vitamin D analog: 19-nor-1,25-dihydroxyvitamin D2. Am J Kidney Dis 32:S48–54.

1740

JAMES O’KELLY, ROBERTA MOROSETTI, AND H. PHILLIP KOEFFLER

111. Martin K, Gonzalez E, Gellens M, Hamm L, Abboud H, Lindberg J 1998 19-nor-1-α-25-dihydroxyvitamin D2 (Paricalcitol) safely and effectively reduces the levels of intact parathyroid hormone in patients on hemodialysis. J Am Soc Nephrol 9:1427–1432.

112. Molnar I, Kute T, Willingham M, Powell B, Dodge W, Schwartz G 2003 19-nor-1α,25-dihydroxyvitamin D2 (paricalcitol): effects on clonal proliferation, differentiation, and apoptosis in human leukemic cell lines. J Cancer Res Clin Oncol 129:35–42.

CHAPTER 97

Clinical Development of Calcitriol and Calcitriol Analogs in Oncology: Progress and Considerations for Future Development* DONALD L. TRUMP JOSEPHIA MUINDI CANDACE S. JOHNSON

PAMELA A. HERSHBERGER I. II. III. IV.

Department of Medicine, Roswell Park Cancer Institute, Buffalo, NY Department of Medicine, Roswell Park Cancer Institute, Buffalo, NY Department of Pharmacology & Therapeutics, Roswell Park Cancer Institute, Buffalo, NY Department of Pharmaceutical Sciences, State University at Buffalo, Buffalo, NY Department of Pharmacology, University of Pittsburgh, Pittsburgh, PA

Introduction Clinical Trials Laboratory-Clinical Extrapolations of Calcitriol Exposure High Dose Intermittent Calcitriol

I. INTRODUCTION 1,25 dihydroxyvitamin D3 (calcitriol), a central factor in bone and mineral metabolism, is a potent antiproliferative agent in a wide variety of malignant cell types [1–12]. As noted in preceding chapters, calcitriol and calcitriol analogs have significant anti-tumor activity in vitro and in vivo in animal and human hematopoietic and epithelial cancer models. Calcitriol enhances the in vitro and in vivo anti-tumor effects of platinum and taxane analogs, as well as antimetabolites (cytosine arabinoside, gemcitabine), topoisomerase inhibitors (etoposide, irinotecan), and alkylating agents [12–14]. Calcitriol as a single agent induces G0/G1 arrest, modulates p27Kipl and p21Waf1/Cipl (the cyclin-dependent kinase (cdk) inhibitors implicated in G1 arrest), induces cleavage of caspase 3, PARP, and the growth-promoting/ prosurvival signaling molecule mitogen-activated protein kinase (MEK) in a caspase-dependent manner [4,9,11,12,15]. In association with these effects, full

*This work is supported by grants from the NCI (CA95045, CA67267, and CA85142) and CaPCURE/The Prostate Cancer Research Foundation. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Calcitriol + Cytotoxic Agent Combinations VI. Calcitriol Analogs VII. The Future References

length MEK, phospho-Erk (P-Erk), and phospho-Akt (P-Akt) are lost. The phosphorylation and expression of Akt, a kinase regulating a second cell survival pathway, is also inhibited after treatment with calcitriol. In contrast to changes that occur during cytotoxic drug-induced apoptosis, the pro-apoptotic signaling molecule MEKK-1 is significantly up-regulated by calcitriol [9]. Enhancement of cytotoxic agent-mediated apoptosis by calcitriol is associated with an increase in PARP-, MEK-, MEKK-1, and caspase-cleavage; P-Erk and P-Akt decrease. In addition, the expression of the p53 homolog, p73, is strongly induced by calcitriol, and p73 can sensitize tumor cells to the cytotoxic effects of platinum and taxanes [20]. Glucocorticoids (GC) potentiate the anti-tumor effect of calcitriol and decrease calcitriol-induced hypercalcemia [16,17]. Both in vitro and in vivo, GC significantly increase vitamin D receptor (VDR) ligand-binding in the tumor while decreasing binding in intestinal mucosa [16], the site of calcium absorption [17]. P-Erk and P-Akt are decreased with calcitriol/GC, compared to either agent alone [16]. These preclinical data support the development of calcitriol-based approaches to cancer therapy. Historically, a limited number of trials in cancer patients have been completed testing vitamin D-based approaches. Copyright © 2005, Elsevier, Inc. All rights reserved.

1742

DONALD L. TRUMP, JOSEPHIA MUINDI, CANDACE S. JOHNSON, AND PAMELA A. HERSHBERGER

While early studies were largely negative, the improved understanding of the molecular changes that occur following calcitriol treatment and new approaches to the use of calcitriol and analogs provide encouraging data on the potential to develop calcitriol-based anti-tumor approaches. This chapter will review the current status of the clinical development of vitamin D analogs in cancer therapeutics, emphasizing important new pharmacokinetic and combination therapy approaches.

II. CLINICAL TRIALS A. Early Trials Oral calcitriol and analogs such as EB 1089 and 1α(OH)D3 have been used in a number of clinical trials. Hematopoietic disorders (myelodysplasia, acute leukemia) have been the most commonly studied diseases. These studies were largely negative and have suffered from a number of limitations: small numbers of patients, lack of control groups, and use of a calcitriol analog in combination with cytotoxic agents in single arm trials, such that conclusions regarding the contribution of the calcitriol analog are difficult [21–26]. Gross and colleagues and Osborn and co-workers conducted straightforward trials of oral daily calcitriol administration in prostate cancer [27,28]. Gross et al. studied patients in whom the prostate-specific antigen was rising following local therapy, and Osborn et al. studied men with androgen-independent prostate cancer. As noted in Chapter 94, both studies were carefully designed, with sufficient power to delineate positive effects of calcitriol. While both studies provided evidence suggesting some positive benefit of calcitriol, both studies illustrate the major shortcomings of all previous studies of calcitriol-based therapeutics: doses of calcitriol and the schedule employed were not reflective of calcitriol exposures achieved in preclinical models in which substantial anti-tumor effects have been demonstrated. Both Gross et al. and Osborn et al. utilized the daily doses of calcitriol employed in the management of benign disease (1.5–2.0 µg). In each study, perturbations of calcium metabolism occurred that led the investigators to limit dose escalation—hypercalciuria (Gross) and hypercalcemia (Osborn). These are the predicted toxicities of vitamin D–based therapies and might have led to investigators abandoning attempts to administer calcitriol in either epithelial or hematopoietic malignancies. Most approaches to overcoming the hypercalcemic effects of calcitriol-based therapies have focused on the development of “nonhypercalcemic” vitamin D analogs. Thousands of such analogs have been synthesized and as will be noted below, some are now entering clinical trials. However, two groups

of investigators have evaluated a different approach to averting the hypercalcemic effects of vitamin D. Reflecting on the fact that in vivo and in vitro experiments utilize high exposure, limited duration treatment, Johnson and Trump as well as Beer and colleagues have evaluated the feasibility of administering high-dose, intermittent regimens of calcitriol.

B. High Dose, Intermittent Calcitriol Regimens Smith and colleagues explored a higher dose subcutaneous regimen of calcitriol, hypothesizing that an every-other-day (QOD) schedule combined with a subcutaneous route of administration might permit safe dose escalation of calcitriol [29]. These investigators were able to administer 8 µg QOD calcitriol safely— this represents a >twofold dose escalation compared to the oral, daily schedule. Beer and colleagues studied oral weekly administration and showed that doses as high as 2.6 µg/kg (approximately 180 µg weekly) could be administered without toxicity. These workers also demonstrated that at doses of >0.5 µg/kg, there appeared to be loss of dose-proportional increase in systemic exposure as doses of oral calcitriol were increased. Importantly, no limiting toxicity was noted in the patients treated by Beer and colleagues. Muindi and co-workers further evaluated these findings of apparent “saturable absorption” during the conduct of a trial of paclitaxel (intravenous, weekly × 6) + dose escalation of oral calcitriol daily for three consecutive days each week (QD × 3, weekly). Calcitriol was administered safely at doses as high as 38 µg QD×3 weekly in combination with paclitaxel [31]. These workers confirmed the loss of dose-proportional increase in calcitriol exposure with increasing dose. Loss of doseproportional increase in exposure appeared to occur at 16–18 µg of calcitriol (Fig. 1). In this figure, baselinesubtracted serum calcitriol AUC0→24hr (area under the concentration-time curve for the 24-hour-period after calcitriol administration) is plotted against dose. A fit to the Michaelis Menten function (AUC = a × dose/ (1 + b × dose) indicates that AUC0→24hr is not proportional to dose (a = 540 ± 140 pg.hr/ml.µg); if AUC were proportional to dose, b would equal 0. The effect of this nonlinearity over the range of doses studied is large; the value of AUC0→24hr at 38 µg is only 4 times that at 4 µg, instead of the 9.5 times expected for a proportional relationship. No deviation from linearity can be detected up to a dose of 17 µg (p =0.4). In this trial, patients with advanced cancer received paclitaxel (80 mg/m2 weekly × 6) + escalating doses of calcitriol, QD×3 weekly ×6. The starting dose of calcitriol was 4 µg po QD × 3 weekly, and patients were entered

1743

CHAPTER 97 Clinical Development of Calcitriol and Calcitriol Analogs in Oncology

A

B Calcitriol AUC0–>24hr (pg.hr/ml)

1400

Cmax (pg/ml)

1200 1000 800 600 400 200

8000

4000

0 0

10

20

30

40

0

10

Calcitriol dose (µg)

20

30

40

Calcitriol dose (µg)

FIGURE 1 (A) Scatter plot of the maximum serum calcitriol concentration (Cmax) vs. calcitriol doses. Closed symbols represent mean values at each dose level. (B) Baseline-subtracted serum calcitriol AUC0 →24hr (area under the concentration-time curve for the 24-hour period after calcitriol administration) plotted against dose, a fit of the Michaelis-Menten function.

Calcitriol AUC0–>24hr (pg.hr/ml)

through the 38 µg dose level. No dose-limiting toxicity was encountered. In this study, the effect of calcitriol on paclitaxel pharmacokinetics was evaluated. No changes in peak concentration, AUC, or t1/2 were noted, indicating the lack of drug-drug interactions between calcitriol and paclitaxel. In these studies as well as those of Beer and colleagues, the commercially available formulation of calcitriol (Rocaltrol®) was used—a formulation available only as 0.25 µg and 0.5 µg caplets. Hence, in these studies, patients were asked to take up to 75–100 caplets at one dose. To investigate whether this apparent limited absorption of calcitriol was related to pharmaceutical limitations posed by multiple caplet ingestion, Muindi and colleagues evaluated patients receiving escalating doses of calcitriol at 14 µg and higher using a liquid formulation of calcitriol (Fig. 2). No change in the curvilinear relationship between dose and AUC was noted.

Taken together these studies clearly indicate that high dose, intermittent administration of calcitriol is safe. Hypercalcemia was transient, and calcium returned to the normal range within approximately 24 hours after dosing. Table I summarizes the doses, concomitant drugs, and limiting toxicities seen in the series of trials conducted by these two groups of investigators. Dose escalation more than tenfold above that achieved with the daily doses of calcitriol is possible— without any apparent toxicity. It is also clear that pharmacokinetic considerations will complicate the use of the current standard formulations of calcitriol. Dose escalation in the studies of Trump and Johnson was ceased when it became clear that dose proportional increase in exposure was not feasible and substantial interpatient variation in exposure was noted (see Fig. 1).

TABLE I Clinical Experience with High Dose Oral Calcitriol Route

Agent1

Schedule

5000

5000

0 14

16 18 Calcitriol dose (µg)

20

Oral, oral SQ, SQ Oral Oral Oral Oral Oral Oral

MDA2

Hypercalcemia

QD

0

2 µg

Yes

QOD QD×3 QD×3 QD× 3 QD× 3 Weekly × 1 Weekly × 6

0 Dex Carboplatin Paclitaxel 0 0 Docetaxel

10 µg 12 µg 24 µg 38 µg 24 µg 2.6 µg/kg 0.5 µg/kg

Yes No No No No No No

FIGURE 2

AUC from patients treated with calcitriol from the phase I trial of calcitriol where pk was determined on day one following calcitriol administration. Patients received either the capsule form (cross symbol) or the liquid form (open symbol).

1 Agents

administered with calcitriol. maximum daily dose administered. QD, every day; QOD, every other day; QD × 3, each day for 3 days/week. 2 MDA,

1744

DONALD L. TRUMP, JOSEPHIA MUINDI, CANDACE S. JOHNSON, AND PAMELA A. HERSHBERGER

III. LABORATORY-CLINICAL EXTRAPOLATIONS OF CALCITRIOL EXPOSURE

100

Among the first questions to arise in attempting to duplicate in the clinic the favorable anti-tumor in vivo effects of calcitriol that can be demonstrated in the laboratory is whether the systemic exposures obtained in the laboratory can be achieved in the clinic. While the initial bias of many has been that administration of high doses of calcitriol is not feasible, the studies of Beer and colleagues and Trump, Johnson, and Muindi clearly show that large doses of calcitriol can be administered safely—the limitation to date is pharmacokinetic, not toxicologic. While realizing that apparently saturable processes of absorption must still be overcome, Muindi and colleagues have characterized the systemic exposure achieved in mice at doses that exhibit anti-tumor and drug potentiating effects in several tumor models. These studies provide a rough “target concentration” that may be necessary to achieve in humans if these anti-tumor effects are to be realized. Figure 3 depicts the plasma concentration-time curve achieved in mice following intraperitoneal administration of “effective doses.” As shown in Table II, at 0.125 µg (the lowest dose to consistently produce significant anti-tumor effects in mice), the AUC0→24hr was 37.3 ng.hr/ml; this compares in man at a 38 µg dose to 7.5 ng.hr/ml. Similarly, in mice the Cmax was 9.2 ng/ml compared to 1.4 ng/ml in man. At the 0.042 µg dose in mice, an anti-tumor effect could be seen but was not consistently observed. Therefore, effective serum calcitriol levels are 5–7 times higher in mice than those achieved at the highest oral dose administered in man (38 µg) in the studies of Muindi and colleagues. The highest AUC0→48hr reported by Beer and colleagues was 47 ng.hr/ml. Beer studied a relatively small number of patients at these high doses. Both groups’ results are in agreement that intermittent high doses of calcitriol can be given safely and that attaining exposure in humans comparable to those required in mice to achieve optimal anti-tumor effects will be difficult with current formulations because they are inconvenient, highly

TABLE II

Plasma calcitriol (ng/ml)

0.125 µg/mouse 0.5 µg/mouse

10

1

0.1

0.01 0

0.042 0.125 0.5

6

9 12 15 Sampling times (hr)

18

21

24

FIGURE 3

Plots of plasma 1,25-D3 I concentration-time curves of normal C3H/HeJ mice; groups of 5–9 mice treated with a single 1,25-D3 i.p. injection of either 0.125 or 0.5 mg dose/mouse. Plasma 1,25-D3 concentrations were measured by RIA. Results are presented as mean ± SD.

variable in absorption, and at the highest doses display apparent saturable absorption characteristics.

IV. HIGH DOSE INTERMITTENT CALCITRIOL Several other studies have confirmed that high-dose intermittent calcitriol is safe and well tolerated.

A. Calcitriol + Dexamethasone Trump and colleagues [33] have completed a 43-patient study of calcitriol in escalating doses to a maximum of 12 µg calcitriol QD ×3 each week together with dexamethasone (4 mg QD ×4, weekly). While it was anticipated that this would be an aggressive calcitriol regimen that would be difficult to administer, this was not the case in practice. No patient ceased therapy because of hypercalcemia. Trimonthly urinary tract radiographs were completed to monitor for urinary

Calcitriol Pharmacokinetic Parameters Mice and Man

Mouse (IP) Dose (µg)

3

Man (PO)

AUC (0→24) (ng.hr/ml)

Cmax (ng/ml)

Dose (µg)

AUC (0→24) (ng.hr/ml)

Cmax (ng/ml)

3.6 37.3 123.9

0.7 9.2 43.4

13 17 38

3.9 ± 1.4 5.4 ± 2.1 7.5 ± 2.1

0.5 ± 0.3 0.5 ± 2.2 1.4 ± 0.9

CHAPTER 97 Clinical Development of Calcitriol and Calcitriol Analogs in Oncology

tract stones. Two patients developed stones; one symptomatic and one asymptomatic. Symptomatic improvement and PSA responses (>50% decline) were seen in 28% of patients, but this response frequency is not clearly greater than one might expect with dexamethasone alone.

B. Single Agent Calcitriol in Androgen-independent Prostate Cancer (AIPC) Following this trial Trump and colleagues undertook a phase I dose escalation trial of calcitriol alone in AIPC to define the maximum tolerated dose and determine response. Dose escalation up to 36 µg QD ×3 weekly was possible without hypercalcemia or urinary tract stones. This study was terminated when it became clear that there was not dose-proportional increase in exposure as drug dose was increased.

C. Single Agent Calcitriol in Androgen-dependent Prostate Cancer Beer and colleagues administered weekly oral calcitriol, 0.5 µg/kg, to 22 men with PSA rising after local therapy (prostatectomy or irradiation) [34]. No toxicity was encountered. No men met the criteria for response established by these investigators. These data clearly indicate that very high intermittent oral doses of calcitriol can be administered safely. Single-agent oral calcitriol therapy (Rocaltrol) appears to have limited activity, at least in prostate cancer, and optimal systemic exposure is limited by variable and incomplete oral absorption. It is important to emphasize that formal bioavailability studies of oral calcitriol have not been conducted to evaluate the exact mechanism of the loss of dose proportional increase in AUC with increasing dose. It is by inference and circumstantial evidence that this observation has been attributed to “decreased absorption.” However, increased rate of metabolism consequent to increased calcitriolinduced 24-hydroxylase activity may also play a role.

V. CALCITRIOL
CYTOTOXIC AGENT COMBINATIONS A. Carboplatin Trump and colleagues initiated a phase I trial of carboplatin + calcitriol, based on the considerable data that

1745

platinum analogs are potentiated by calcitriol [35–37]. Patients with advanced cancer were treated with carboplatin (AUC = 5) every 28 days + escalating doses of calcitriol QD × 3 every 28 days. Calcitriol starting dose was 4 µg QD × 3. Studies were designed such that in each patient, carboplatin was given on day 1 before calcitriol in one of the first two cycles of treatment and on day 3 after two days of high dose calcitriol on the other. This permitted comparison of AUC of carboplatin in the same patient before and after pretreatment with calcitriol. Dose-limiting toxicity was not encountered in this trial. The AUC of carboplatin was higher in patients who received carboplatin following 3 days of calcitriol than in patients in whom carboplatin was administered before calcitriol (mean AUC = 7.6 µg/ml.hr ± 1.8, carboplatin day 3 [DDDC] vs. AUC = 6.6 µg/ml.hr ± 1.4, carboplatin day 1 [CDDD], p = 0.04) (Fig. 4). While no-dose-limiting toxicity has been seen, myelosuppression (% change in platelet count) following the sequence carboplatin →calcitriol (CDDD) was less than that following calcitriol → carboplatin (DDDC), consistent with the change in AUC. No clinically detectable renal impairment was seen with either sequence. These data indicate that potentiation of carboplatin by calcitriol may in part be related to reduced carboplatin clearance. No patients became hypercalcemic. This trial was halted when the concerns regarding predictable and dose proportional exposure became evident in this and other studies of oral calcitriol (Rocaltrol).

B. Taxanes We have discussed the trial of Trump and colleagues evaluating the combination of high dose oral calcitriol + the taxane, paclitaxel. No dose-limiting toxicity was noted at calcitriol doses up to 38 µg QD ×3 weekly. Recently, Beer and colleagues reported a phase I trial demonstrating that patients can tolerate weekly oral dosing of calcitriol at 0.5 µg/kg + docetaxel (36 mg/m2 weekly × 6) without significant toxicity [38]. This group has reported the results of a phase II trial of this regimen in men with androgen-independent prostate cancer. Among 37 men treated with weekly calcitriol (Rocaltrol) + docetaxel, 81% (95% confidence interval [CI] 68–94%) achieved a PSA response rate as measured by a greater than 50% reduction in PSA [39]. This response rate appears greater than response rates of 38–46% reported in phase II studies of single-agent weekly docetaxel. Among 15 patients with measurable tumor masses 8 (53%, Ci 43%–75%) achieved a tumor mass response defined by standard criteria. These are quite encouraging data with respect to anti-tumor effects of calcitriol-based therapy and has led to an

1746

DONALD L. TRUMP, JOSEPHIA MUINDI, CANDACE S. JOHNSON, AND PAMELA A. HERSHBERGER

12 DDDC CDDD 10

Carboplatin AUC

8

6

4

2

0 1

2 4 µg

3

4

5 6 µg

6

7

8 8 µg

9

10

11

11 µg

12

13

14

15

14 µg

16

16 µg

17 24 µg

Patients and calcitriol dose

FIGURE 4 Carboplatin AUC from patients treated at selected doses of calcitriol with either carboplatin on day 1, calcitriol day 1, 2, and 3 (CDDD) (open bars), or calcitriol day 1, 2, and 3 followed by carboplatin on day 3 (DDDC) (closed bars). Carboplatin AUC is significantly greater with the DDDC sequence than the sequence CDDD ( p < 0.04).

ongoing phase III trial to more rigorously evaluate this regimen (see below).

C. Reformulation of Calcitriol In view of the fact that high-dose intermittent calcitriol is safe, feasible—but inconvenient and not dependably absorbed using the currently available and tested formulation Rocaltrol—efforts have been undertaken to develop a more “bioavailable” and more convenient preparation of calcitriol. The express purpose of this effort is to develop a preparation that might allow full testing of calcitriol in high dose alone and with cytotoxic agents. A new formulation, DN 101 (Novocea, Inc.) has completed phase 1 testing; initial results indicate a more linear relationship between dose and exposure. No unusual toxicity or effects have been noted [40]. DN 101 is the calcitriol preparation being used in a randomized trial of docetaxel +/− calcitriol in AIPC. This trial is testing whether the increased PSA response rate seen in the trial of Beer et al. can be confirmed in a randomized trial.

VI. CALCITRIOL ANALOGS While considerable work has been done to demonstrate that high-dose intermittent calcitriol administration is feasible and that blood levels in the range of those found to be effective in animal models have not yet been achieved, great interest remains in the development of vitamin D analogs that retain the antiproliferative and/or prodifferentiative properties of calcitriol with less propensity to cause hypercalcemia. Two analogs have been tested in substantial numbers of cancer patients: EB 1089 or seocalcitol and 1α(OH)D2.

A. EB 1089 (seocalcitol) Phase 1 trials of seocalcitol [(1(S),3(R)-dihydroxy20(R)-(5′-ethyl-5′-hydroxy-hepta-1′(E),3′(E)-dien-1′-yl)9,10-secopregna-5(Z),7(E),10(19)-triene] (see Chapter 84) have been conducted using a daily oral schedule of administration [41]. 7–15 µg/m2m is estimated to be tolerable and all patients who received 17 µg/sqm per day developed hypercalcemia. Preclinical data indicate

CHAPTER 97 Clinical Development of Calcitriol and Calcitriol Analogs in Oncology

seocalcitol is 50–200 times more potent than calcitriol in terms of antiproliferative activity; these phase I data indicate that seocalcitol is 1/7–1/10 as potent in inducing hypercalcemia. As discussed elsewhere in this text, phase II studies have been conducted in breast, pancreatic, colorectal hepatocellular carcinomas (HCC), as well as leukemia [42,43] (Chapter 84). Anti-tumor responses have been seen in HCC, but not in the other diseases.

B. 1(OH)D2 This analog has been developed as one potentially more active and less prone to cause hypercalcemia. This agent is converted to 1,25(OH)2D2 and 1,24(OH)2D2, both of which activate VDR-mediated biologic effects; the 1,24(OH)2D2 metabolite is substantially less potent than 1,25(OH)2D2 or 1,25(OH)2D3 [44,45]. Liu and colleagues conducted a phase I trial of 1α(OH)D2 in prostate cancer patients [46]. Daily dosing from 5 to 15 µg per day was employed. Hypercalcemia with dehydration and azotemia was noted at 15 µg QD; 12.5 µg QD was well tolerated in 3 of 3 patients. Two of 25 patients treated demonstrated evidence of antitumor effect; interestingly both responses were seen at “low” doses (5 µg and 7.5µg). It would appear that 1α(OH)D2 is approximately 1/10 as prone to induce hypercalcemia as a similar dose of calcitriol on a QD-dosing schedule. The anti-tumor effects of seocalcitol and 1α(OH)D2 are modest in the studies conducted. The activity of seocalcitol in HCC is of considerable interest in view of the importance of this disease worldwide.

VII. THE FUTURE These data, in combination with the considerable preclinical information indicating the potential role of vitamin D–based therapies in cancer, continue to stimulate the interest of several research groups. The ongoing study of the new calcitriol formulation + docetaxel in AIPC will be very important in establishing the potential for calcitriol, on this dose and schedule, in the management of prostate cancer. Among the unanswered questions are: 1. What is the proper dose and schedule? Preclinical data indicate that all studies have been conducted at 1/5 – 1/10 the drug exposure that is effective in animal models; and clinical evidence suggests that administration of the exposures effective in preclinical models may be safe and feasible. Trials of intravenous calcitriol and new formulations of calcitriol to establish the maximum possible dose on an intermittent schedule are underway. 2. What is the

1747

“best analog”? While preclinical data suggest advantages for a number of analogs, substantial clinical work remains to evaluate the analogs currently in clinical trials. The greatest preclinical and clinical work has been done with calcitriol. 3. Are there additional novel approaches that may capitalize on modulation of the vitamin D system? Potentiation of the anti-tumor activity of growth factor receptor antagonists (gefitinib), as well as many other cytotoxics, has been well described in preclinical models [47–54]. Vitamin D analogs have been described to have antiangiogenic effects, as well as direct anti-tumor activities; this suggests that combinations and applications in settings where tumor blood vessels are the target may have merit [56–58]. Regional administration has merit. Regional arterial infusions (e.g. hepatic artery) or topical (e.g. cutaneous or bronchial) therapy are being investigated. Regional approaches have the great advantage of permitting the administration of high doses, locally with limited systemic effects. While definitive data regarding the use of vitamin D in the management of cancer remains elusive, preclinical data are persuasive and considerable progress has been made in developing clinical strategies utilizing vitamin D in the treatment of epithelial and hematopoietic cancers.

References 1. Bikle DD, Pillai S 1993 Vitamin D, calcium, and epidermal differentiation. Endocr Rev 14(1):3–19. 2. Reichel H, Koeffler HP, Norman AW 1989 The role of vitamin D endocrine system in health and disease. N Engl J Med 320(15):980–991. 3. McElwain MC, Dettlebach MA, Modezelewski RA, Russell DM, Uskokovic MR, Smith DC, Trump DL, Johnson CS 1995 Antiproliferative effects in vitro and in vivo of 1,25-dihydroxyvitamin D3 and a vitamin D3 analog in a squamous cell carcinoma model system. Mol Cell Diff 3(1):31–50. 4. Getzenberg RH, Light BW, Lapco PE, Konety BR, Nangia AK, Acierno JS, Dhir R, Shurin Z, Day RS, Trump DL, Johnson CS 1997 Vitamin D inhibition of prostate adenocarcinoma growth and metastasis in the Dunning rat prostate model system. Urology 50:999–1006. 5. Mangelsdorf DJ, Koeffler HP, Donaldson CA, Pike JW, Haussler MR 1984 1,25-dihydroxyvitamin D3 induced differentiation in a human promyelocytic leukemia cell line HL-60: receptor mediated maturation to macrophage-like cells. J Cell Biol 98:391–398. 6. Colston KW, Chander SK, Mackay AG, Coombes RC 1992 Effects of synthetic vitamin D analog on breast cancer cell proliferation in vivo and in vitro. Biochem Pharmacol 44:693–702. 7. Shabahang M, Buras RR, Davoodi F, Schumaker LM, Nauta RJ, Uskokovic MR, Brenner RV, Evans SR 1994 Growth inhibition of HT-29 human colon cancer cells by analogs of 1,25dihyroxyvitamin D3. Cancer Res 54:4057–4064. 8. Peehl DM, Skowronski RJ, Leung GK, Wong ST, Stamey TA, Feldman D 1994 Antiproliferative effects of 1,25-dihydroxyvitamin D3 on primary cultures of human prostatic cells. Cancer Res 54(3):805–810.

1748

DONALD L. TRUMP, JOSEPHIA MUINDI, CANDACE S. JOHNSON, AND PAMELA A. HERSHBERGER

9. McGuire TF, Trump DL, Johnson CS 2001 Vitamin D3–induced apoptosis of murine squamous cell carcinoma cells: Selective induction of caspase-dependent MEK cleavage and up-regulation of MEKK-1. J Biol Chem 276(28):26365–26373. 10. Zhou JY, Norman AW, Chen DL, Sun GW, Uskokovic M, Koeffler HP 1990 1,25-dihydroxy-16-ene-23-yne-vitamin D3 prolongs survival time of leukemic mice. Proc Natl Acad Sci USA 87(10):3929–3932. 11. Modzelewski RA 1999 Apoptotic effects of paclitaxel and calcitriol in rat dunning MLL and human PC-3 prostate tumor cells in vitro. Proc Amer Assoc Cancer Res 40:580. 12. Hershberger PA, Yu WD, Modzelewski RA, Rueger RM, Johnson CS, Trump DL 2001 Enhancement of paclitaxel anti-tumor activity in squamous cell carcinoma and prostatic adenocarcinoma by 1,25-dihydroxycholecaciferol (1,25-D3). Clin Cancer Res 7:1043–1051. 13. Light BW, Yu W-D, McElwain MC, Russell DM, Trump DL, Johnson CS 1997 Potentiation of cisplatin anti-tumor activity using a vitamin D analog in a murine squamous cell carcinoma model system. Cancer Res 57(17):3759–3764. 14. Christakos S, Raval-Pandya M, Wernyj RP, Yang W 1996 Genomic mechanisms involved in the pleiotropic actions of 1,25-dihydroxivitamin D3. Biochem J 316(Pt 2):361–371. 15. Hershberger PA, Modzelewski RA, Shurin ZR, Rueger RM, Trump DL, Johnson CS 1999 In vitro and in vivo modulation of p21Wafl/Cip1 and p27Kip1 in squamous cell carcinoma I response to 1,25-dihydroxycholecalciferol (calcitriol). Cancer Res 59:2644–2649. 16. Yu W-D, McElwain MC, Modzelewski RA, Russell DM, Smith DC, Trump DL, Johnson CS 1998 Enhancement of 1,25-dihydroxyvitamin D3–mediated anti-tumor activity with dexamethasone. J Natl Cancer Inst 90(2):134–141. 17. Bernardi RJ, Trump DL, Yu W-D, McGuire TF, Hersherber PA, Johnson CS 2001 Combination of 1α,25-dihydroxyvitamin D3 with dexamethasone enhances cell cycle arrest and apoptosis: Role of nuclear receptor cross-talk and Erk/Akt signaling. Clin Cancer Res 7:4165–4173. 18. Evans RM 1998 The steroid and thyroid hormone receptor superfamily. Science 240:889. 19. Darwish HM, DeLuca HF 1996 Recent advances in the molecular biology of vitamin D action. In: Progress in Nucleic Acid Research and Molecular Biology. Academic Press Inc, Vol 53, pp. 321 . 20. Darwish H, DeLuca HF 1993 Vitamin D–regulated gene expression. Critical Reviews in Eukaryotic Gene Expression 3(2):89–116. 21. Rustin GJ, Quinnell TG, Johnson J, Clarke H, Nelstrop AE, Bollag W 1996 Br J. Trial of isotretinoin and calcitriol monitored by CA 125 in patients with ovarian cancer. Cancer 74(9):1479–1481. 22. Slapak CA, Desforges JF, Fogaren T, Miller KB 1992 Treatment of acute myeloid leukemia in the elderly with lowdose cytarabine, hydroxyurea, and calcitriol. Am J Hematol 41(3):178–183. 23. Petrini M, Caracciolo F, Corini M, Valentini P, Sabbatini AR, Grassi B 1991 Low-dose ARA-C and 1(OH)D3 administration in acute nonlymphoid leukemia: pilot study. Haematologica 76(3):200–203. 24. Hellstrom E, Robert KH, Samuelsson J, Lindemalm C, Grimfors G, Kimby E, Oberg G, Winqvist I, Billstrom R, Carneskog J 1990 Treatment of myelodysplastic syndromes with retinoic acid and 1 alpha-hydroxy-vitamin D3 in combination with low-dose ara-C is not superior to ara-C alone.

25.

26.

27. 28.

29.

30. 31.

32.

33.

34.

35.

36.

37.

38.

39.

Results from a randomized study. The Scandinavian Myelodysplasia Group (SMG). Eur J Haematol 45(5):255–261. Trouillas P, Honnorat J, Bret P, Jouvet A, Gerard JP 2001 Redifferentiation therapy in brain tumors: long-lasting complete regression of glioblastomas and an anaplastic astrocytoma under long-term 1-alpha-hydroxycholecalciferol. J Neurooncol 51(1):57–66. Nozaki T, Kusuzaki K, Takeshita H, Nakamura S, Hashiguchi S, Murata H, Hirasawa Y 2001 Effectiveness of activated vitamin D3 on improving prognosis of osteosarcoma patients. Oncol Rep 8(2):321–324. Gross C, Stamey T, Hancock S, Feldman D 1998 Treatment of early recurrent prostate cancer with 1,25-dihydroxyvitamin D3 (calcitriol). J Urol 159(6):2035–2039. Osborn JL, Schwartz GG, Smith DC, Bahnson R, Day R, Trump DL 1995 Phase II trial of oral 1,25-dihydroxyvitamin D (calcitriol) in hormone refractory prostate cancer. Urol Oncol 1:195–198. Smith DC, Johnson CS, Freeman CC, Muindi J, Wilson JW, Trump DL 1999 A phase I trial of calcitriol (1,25-Dihydroxycholecalciferol) in patients with advanced malignancy. Clin Cancer Res 5:1339–1345. Beer TM, Munar M, Henner WB 2001 A phase I trial of pulse calcitriol in patients with refractory malignancies. Cancer 91:2431–2439. Muindi JR, Peng Y, Potter DM, Hershberger PA, Tauch JS, Capozzoli MJ, Egorin MJ, Johnson CS, Trump DL 2002 Pharmacokinetics of high-dose calcitriol: results obtained during a phase one trial of calcitriol and paclitaxel. Cancer Pharm Therap 72(6):648–659. Muindi JR, Modezlewski RA, Peng Y, Hershberger PA, Trump DL, Johnson CS 2004 Pharmacokinetics of 1α,25dihydroxycholecalciferol pharmacokinetics in normal and tumor-bearing mice. Oncology 66(1):62–66. Trump DL, Serafine S, Brufsky J, Muindi J, Bernardi R, Potter D, Johnson C 2000 High-dose calcitriol (1,25(OH)2 vitamin D3) + dexamethasone in androgen-independent prostate cancer (AIPC). Proceedings Amer Soc Clin Oncol 19:337a. Beer TM, Lemmon D, Lowe BA, Henner WD 2003 High-dose weekly oral calcitriol in patients with a rising PSA after prostatectomy or radiation for prostate carcinoma. Cancer 97(5): 1217–1224. Hershberger PA, McGuire TF, Yu W-D, Zuhowski EG, Schellens JH, Egorin MJ, Trump DL, Johnson CS 2002 Cisplatin potentiates 1,2-dihydroxyvitamin D3–induced apoptosis in association with increased mitogen-activated protein kinase kinase kinase 1 (MEKK-1) expression. Mol Cancer Therapeutics 1(10):821–829. Gong JG, Constanzo A, Yang HQ, Melino G, Kaelin WG Jr, Levrero M, Wang JY 1999 The tyrosine kinase c-Abl regulates p73 in apoptotic response to cisplatin-induced DNA damage. Nature 399:806–809. Johnson CS, Egorin MJ, Zuhowski R, Parise R, Cappozolli M, Belani CP, Long GS, Muindi J, Trump DL 2000 Effects of high-dose calcitriol (1,25-dihydroxyvitamin D3) on the pharmacokinetics of paclitaxel or carboplatin: results of two phase I studies. Proceedings Amer Soc Clin Oncol 19:210a. Beer TM, Eilers KM, Garzotto M, Lowe BA, Henner WD 2002 Androgen-independent prostate cancer (AIPC) treatment with weekly high-dose calcitriol and docetaxel. Proceedings Amer Soc Clin Oncol 21:177a. Beer TM, Eilers KM, Garzotto M, Egorin MJ, Lowe BA, Henner WD 2003 Weekly high-dose calcitriol and docetaxel in

CHAPTER 97 Clinical Development of Calcitriol and Calcitriol Analogs in Oncology

40. 41.

42.

43.

44. 45. 46.

47.

48.

metastatic androgen-independent prostate cancer. J Clin Oncol 21(1):123–128. Beer TM (personal communication). Gulliford T, English J, Colston KW, Menday P, Moller S, Coombes RC 1998 A phase I study of the vitamin D analog EB 1089 in patients with advanced breast and colorectal cancer. Br J Cancer 78(1):6–13. Dalhoff K, Dancey J, Astrup L, Skovsgaard T, Hamberg KJ, Lofts FJ, Rosmorduc O, Erlinger S, Bach Hansen J, Steward WP, Skov T, Burcharth F, Evans TR 2003 A phase II study of the vitamin D analog seocalcitol in patients with inoperable hepatocellular carcinoma. Br J Cancer 89(2):252–257. Evans TR, Colston KW, Lofts FJ, Cunningham D, Anthoney DA, Gogas H, de Bono JS, Hamberg KJ, Skov T, Mansi JL 2002 A phase II trial of the vitamin D analog seocalcitol (EB1089) in patients with inoperable pancreatic cancer. Br J Cancer 86(5):680–685. Sjoden G, Lindgren JU, DeLuca HF 1984 Antirachitic activity of 1alpha-hydroxyergocalciferol and 1alpha-hydroxycholecalciferol in rats. J Nutr 114(11):2043–2046. Sjoden G, Smith C, Lindgren U, DeLuca HF 1985 1alphahydroxyvitamin D2 is less toxic than 1 alpha-hydroxyvitamin D3 in the rat. Proc Soc Exp Biol Med 178(3):432–436. Liu G, Oettel K, Ripple G, Staab MJ, Horvath D, Alberti D, Arzoomanian R, Marnocha R, Bruskewitz R, Mazess R, Bishop C, Bhattacharya A, Bailey H, Wilding G 2002 Phase I trial of 1alpha-hydroxyvitamin D2 in patients with hormone refractory prostate cancer. Clin Cancer Res 8(9):2820–2827. Elstner E, Linker-Israeli M, Umiel T, Le J, Grillier I, Said J, Shintaku IP, Krajewski S, Reed JC, Binderup L, Loeffler HP 1996 Combination of a potent 20-epi-vitamin D3 analog (KH 1060) with 9-cis-retinoic acid irreversibly inhibits clonal growth, decreases bc1–2 expression, and induces apoptosis in HL-60 leukemic cells. Cancer Res 56:3570–3576. Tong WM, Kallay E, Hofer H, Hulla W, Manhardt T, Peterlik M, Cross HS 1998 Growth regulation of human colon cancer cells by epidermal growth factor and 1,25-dihydroxyvitamin D3 is mediated by mutual modulation of receptor expression. Euro J of Can 34:2119–2125.

1749

49. Koga M, Eisman JA, Sutherland RL 1988 Regulation of epidermal growth factor receptor levels by 1,25-dihydroxyvitamin D3 in human breast cancer cells. Can Res 48:2734–2739. 50. Gonzales EA, Disthabanchong S, Kowalewski R, Martin KJ 2002 Mechanisms of the regulation of EGF receptor gene expression by calcitriol and parathyroid hormone in UMR 106-01 cells. Kidney Int 61:1627–1634. 51. Boyle BJ, Zhao XY, Cohen P, Feldman D 2001 Insulin-like growth factor binding protein-3 mediates 1 alpha,25-dihydroxyvitamin D3 growth inhibition in the LNCaP prostate cancer cell line through p21/WAF1. J Urol 165:1319–1324. 52. Rozen F, Yang XF, Huynh H, Pollak M 1997 Antiproliferative action of vitamin D–related compounds and insulin-like growth factor-binding protein 5 accumulation. J Natl Cancer Inst 89:652–656. 53. Rozen F, Pollak M 1999 Inihibition of insulin-like growth factor I receptor signaling by the vitamin D analog EB 1089 in MCF-7 breast cancer cells: A role for insulin-like growth factor binding proteins. Int J Oncol 15:589–594. 54. Danielpour D 1996 Induction of transforming growth factorbeta autocrine activity by all-transretinoic acid and 1 alpha,25dihydroxyvitamin D3 in NRP-152 rat prostatic epithelial cells. J Cell Physiol 166:231–239. 55. Yang L, Yang J, Venkateswarly S, Ko T, Brattain MG 2001 Autocrine TGF beta signaling mediates vitamin D3 analoginduced growth inhibition in breast cells. J Cell Physiol 188:383–393. 56. Mantell DJ, Owens PE, Bundred NJ, Mawer EB, Canfield AE 2000 1 alpha,25-dihydroxyvitamin D3 inhibits angiogenesis in vitro and in vivo. Circ Res 87(3):214–220. 57. Bernardi RJ, Johnson CS, Modzelewski RA, Trump DL 2002 Antiproliferative effects of 1alpha,25-dihydroxyvitamin D3 and vitamin D analogs on tumor-derived endothelial cells. Endocrinology 143(7):2508–2514. 58. Iseki K, Tatsuta M, Uehara H, Iishi H, Yano H, Sakai N, Ishiguro S 1999 Inhibition of angiogenesis as a mechanism for inhibition by 1alpha-hydroxyvitamin D3 and 1,25-dihydroxyvitamin D3 of colon carcinogenesis induced by azoxymethane in Wistar rats. Int J Cancer 81(5):730–733.

CHAPTER 98

Vitamin D3: Autoimmunity and Immunosuppression JACQUES LEMIRE

Division of Pediatric Nephrology, Department of Pediatrics, University of California, San Diego, La Jolla, California

I. Introduction II. Autoimmunity

I. INTRODUCTION Since the mid-1980s, a variety of new properties and applications have been discovered for the hormone 1,25-dihydroxyvitamin D3 [1,25(OH)2D3]. Almost simultaneously the antiproliferative, prodifferentiating, and immunosuppressive activities of this metabolite of vitamin D were defined. It became obvious from the early investigations that in order to achieve maximal immunosuppressive activity in vitro, 1,25(OH)2D3 was required at a concentration higher than that needed to obtain antiproliferative activity. This observation explains in part the early success of the hormone when used for the treatment of psoriasis, whereas, currently, the inherent hypercalcemic properties of 1,25(OH)2D3 still prevent its clinical use for immunosuppression in humans. However, the development of analogs of the active metabolite has now broadened the potential clinical applications of the hormone. The optimal compound would be one that exerts maximal immunosuppressive activity while sparing the recipient of hypercalcemic complications. To achieve that goal, a variety of animal models of autoimmunity have been studied using 1,25(OH)2D3 and related analogs. This work has led to a potential application for the hormone in human autoimmune diseases. This review will concentrate on the mechanisms of action and effectiveness of 1,25(OH)2D3 in animal models of autoimmunity (excluding psoriasis, which is discussed in Chapter 101, and diabetes, which is discussed in Chapter 99) and describe the practical application of the hormone in humans for autoimmunity. Vitamin D regulation of the immune response is covered in Chapter 36.

II. AUTOIMMUNITY A. Immune Mechanisms Operational in Autoimmunity It is beyond the scope of this section to provide an extensive review of mechanisms involved in VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

References

autoimmunity. However, by outlining some key factors described in the dysregulation of the immune system leading to the process, a better understanding of some targets of 1,25(OH)2D3 can be achieved. The description of helper T-cells with different cytokine secretion patterns has opened a new understanding of cell–cell interaction and regulation. Three different T-helper cell subsets have been well described: Th0, Th1, and Th2 cells. The first subset, Th0, represents what appears to be an early precursor or transitory cell that subsequently differentiates into Th1 or Th2 cells on the appropriate stimulus, which is most likely provided by antigen-presenting cells (monocytes, macrophages, B-cells, or dendritic cells). One of these stimuli is interleukin-12 (IL-12), which can be produced by both monocytes, macrophages, and B-cells [1] and promotes Th0 cells to differentiate along the Th1 pathway [2]. The helper T-cell subsets Th1 and Th2 are determined by their cytokine secretion patterns. The Th0 cells produce an unrestricted pattern of cytokines. Th1 cells produce IL-2 and interferon-γ (IFN-γ), whereas Th2 cells produce IL-4, -5, -6, -10, and -13 [3]. Furthermore, Th1 cells can transfer delayed-type hypersensitivity (DTH) [4] and provide help to B-cells to produce the antibody isotype immunoglobulin G2a (IgG2a), whereas Th2 cells help B-cells for IgGi and IgE secretion [5]. Because of their IFN-γ production, Th1 cells can also interact with macrophages to increase bactericidal properties [6]. These Th subsets also cross-regulate one another: IFN-γ produced by Th1 cells can down-modulate Th2 cells, and IL-4 and IL-10 produced by Th2 cells inhibits Th1 cells [7]. Interestingly, IL-10 produced by Th2 cells can indirectly inhibit Th1 cell responses by acting on monocytes that are required by Th1 cells for antigen-specific proliferation and lymphokine secretion [8], most likely by inhibiting IL-12 secretion [9]. The lymphokine IFN-γ produced by Th1 cells also enhances class II antigen expression [10]. This dichotomy between Th1 and Th2 cells has been confirmed in humans. A similar pattern of Copyright © 2005, Elsevier, Inc. All rights reserved.

1754 cytokine secretion for both Th subsets is present [11]. Th1 cells express cytolytic activity against antigenpresenting cells and provide helper function for IgM, IgG, and IgA synthesis at low T-cell/B-cell ratios. At T/B ratios higher than 1:1, a decline in B-cell help is observed, related to the lytic activity of Th1 cells against autologous antigen-presenting B-cells [12]. This downregulation of antibody responses could be operational in vivo. In contrast, Th2 cells develop in response to allergens or parasites, provide help for all immunoglobulin classes including IgE, and lack cytolytic potential [13]. The absence of lytic activity of Th2 cells may account for the long-term IgE responses of patients with atopy or parasitic infections [13]. High efficiency cloning of peripheral blood CD4+ T cells from healthy individuals generates the Th1, Th0, and Th2 cytokine profiles roughly distributed according to a 2:4:1 ratio [13]. The heterogeneity of the cytokine profile in humans is not restricted to CD4+ cells; CD8+ cells, which have the phenotype of cytotoxic and suppressor cells, can also be further defined by analysis of their lymphokine profile [14]. The recognition of the presence of regulatory T-cells leading to peripheral tolerance to extrinsic antigens or autoantigens has helped to understand some of the mechanisms involved in autoimmunity and to provide potential tools for new therapeutic agents [15]. Among those regulatory cells, CD25+T-cells appear to provide a protective effect in the prevention of the autoimmune process [16]. The role of Th2 cells as regulators of physiologic autoimmunity remains unclear. However, a protective effect of Th2 has been shown in situations of induction of autoimmunity such as antigen administration [15]. Further evidence of such a down-regulating role of Th2 cells has been provided by animal models: abrogation of tolerance in IL-4 deficient mice [17] or prevention of diabetes of non-obese diabetic (NOD) mice with administration of Th2 cytokines such as IL-4, IL-10 or IL-13 [18]. Induction of regulatory cells by T-cell vaccination such as T-cell receptor (TCR) peptide vaccination has been accomplished in experimental animal models and provides promising possibilities [19].

B. Mechanisms of Action of 1,25(OH)2D3 in Autoimmunity A significant body of evidence suggests that while the resulting effects of 1,25(OH)2D3 on the immune response are suppressive, its actions are complex. The final response results in part from the interaction of the hormone with both antigen-presenting cells (APC) and T-cells leading to a dual response: suppression of

JACQUES LEMIRE

enhancers and amplification of down-regulators of the immune response. 1. HELPER T-CELLS

From the early discovery of the action of 1,25(OH)2D3 on the immune system, an interest sparked by the discovery of VDR in lymphocytes, a direct antiproliferative effect of the hormone on T-cells has been described [20]. However, its specificity of action resulted from the analysis of cytokine production by T-cells. The sterol inhibits the production of the Th1mediated cytokines IL-2 and IFN-γ, both at transcriptional levels [21,22]. The pro-Th2 cytokine pathway appears to be spared by the sterol, but controversy has recently risen as to the net effect of the hormone on IL4 production. First, 1,25(OH)2D3 leads to an increased production of IL-10 by helper T-cells [23]. By studying human helper T-cell clones, already committed to a specific antigen, the following observations were made after exposure of the cloned T-cells to the sterol. Helper T-cell clones were isolated from atopic patients sensitive to the rye grass antigen Lol pI and characterized as Th0, Th1, and Th2 based on their lymphokine secretion pattern [24]. The Th subsets were activated with Lol pI and antigen-presenting cells in the presence or absence of 1,25(OH)2D3 or the analog 1,25(OH)2-16-ene D3, and the effect of the vitamin D compounds on the lymphokine production was analyzed (Fig. 1). Both 1,25(OH)2D3 and l,25(OH)2-16-ene D3 suppressed the production of IFN-γ by Th1 cells in a dose-dependent manner, but these compounds had minimal effect on IL-4 production by Th2 cells and only at the highest concentrations tested. Interestingly, Th0 cells, producer of both cytokines, showed a profound reduction in IFN-γ in the presence of the vitamin D compounds, whereas IL-4 secretion was less inhibited, suggesting once again a pro-Th1 effect of the hormone and its analog [25]. Further evidence for an immunosuppressive effect of 1,25(OH)2D3 on a Th1-mediated biological activity was provided by the passive transfer of myelin basic protein (MBP)-specific Th1 clones. A characteristic of Th1 cells is their ability to transfer delayed-type hypersensitivity (DTH) [4]. MBP-reactive T-cell clones were activated with syngeneic spleen cells and antigen (MBP) in the presence or absence of 1,25(OH)2D3 before being washed and transferred to the footpads of naive mice. Swelling, as an index of DTH, was measured before and 18 hr after cell transfer using a pressure-sensitive caliper. A complete inhibition of the passive transfer of DTH was observed with 10 nM 1,25(OH)2D3 (Fig. 2) [26]. These results suggested that the hormone could directly interfere with functional activity of Th1 cells.

1755

Th1 and Th2 100 80 60 40 20 0

10−10 10−9 10−8 10−7

% Suppression of IFN (−) and IL-4 (- -)

% Suppression of IFN (−) and IL-4 (- -)

CHAPTER 98 Vitamin D3: Autoimmunity and Immunosuppression

Th0 100 80 60 40 20 0

10−10 10−9 10−8 10−7

FIGURE 1 Effect of 1,25(OH)2D3 ( , ●) and analog 1,25(OH)2-16-ene D3 (
, ■) on lymphokine production by Th1 (left, open symbols), Th2 (left, filled symbols), and Th0 (right) cell clones. Th clones were activated with antigen, APC, and the lymphokines IL-4 and IFN-γ and assayed by radioimmunoassay. The data are expressed as percentage suppression.

More recently, it was shown that 1,25(OH)2D3 could directly affect naïve CD4+ T cells to enhance the development of Th2 cells [27]. When given in vivo, in IL-4–deficient mice, 1,25(OH)2D3 was less effective in its ability to suppress experimental autoimmune encephalitis (EAE) induction [28]. However, other studies have also shown a reduction of [29] or no effect

DTH response (mm×10−10)

12 10 8 6 4 2 1 0

10−10 10−9 10−8

10−7

1,25-Dihydroxyvitamin D3 (M)

FIGURE 2 Delayed-type hypersensitivity (DTH) response in histocompatible recipients of cocultures of MBP-activated, cloned T cells and syngeneic spleen cells exposed to 20 µg/ml MBP. Cocultures were incubated for 5 days in the presence or absence of varying mean ± SD of the DTH response value in five individual mice. An asterisk signifies a significant (p < 0.005) decrease in the DTH response compared to cells not exposed to hormone. From Lemire and Adams [29] with permission.

on [30,31] IL-4 secretion by T-cells, questioning the 1,25(OH)2D3 “switch” to a pro-Th2 response. So far, these studies would suggest a primarily Th1–mediated inhibitory effect of 1,25(OH)2D3 with a resulting enhanced Th2 functional ability. Whether or not IL-4 is affected remains to be clarified in an in vivo situation. 2. SUPPRESSOR, CYTOXIC, AND REGULATORY T-CELLS

The modulation of suppressor/cytotoxic T-cells in the context of autoimmunity by 1,25(OH)2D3 is unclear. Studies in transplantation immunology have suggested the following. In the context of the mixed lymphocyte reaction [MLR] using human cells, when added at the initiation of the MLR, 1,25(OH)2D3 induced suppressor cell activity, reduced the generation of cytotoxic T-cells and the expression of class II antigen [32]. The steroid hormone could also inhibit natural killer cell activity [33–35] and reduce the activity of cytotoxic T cell lines [36]. Further alternative regulatory mechanisms can be attributed to the steroid hormone. 1,25(OH)2D3 could instead promote or allow for the development of regulatory T-cells and to induce tolerance. Naïve human and mouse CD4+T-cells can be induced into regulatory cells producing IL-10 by treatment with 1,25(OH)2D3 [23]. Transplantation tolerance through the induction of CD4+CD25+ regulatory T-cells can be achieved with 1,25(OH)2D3 [37]. Finally, the steroid hormone can interfere with apoptosis by the down-regulation of CD95L, a cell surface molecule not only activating apoptosis but also promoting Th1 activation through antigen-presenting cell maturation [38].

1756 3. ANTIGEN-PRESENTING CELLS

The first evidence of a direct effect of 1,25(OH)2D3 on antigen-presenting cells was provided by the dosedependent inhibition of IL-12, a pro-Th1 cytokine, by the steroid hormone [39]. Recent studies suggest that the resulting in vitro effect of the drug, when administered early, is to interfere with the differentiation and maturation of APC, thereby leading to altered T-cell responsiveness. In vitro treatment with 1,25(OH)2D3 and analogs leads to down-regulated expression of the co-stimulatory molecules CD40, CD80, CD86 and to decreased IL-12 and enhanced IL-10 production, resulting in decreased T-cell activation, and with APC with tolerogenic properties [40]. Some in vivo evidence of such tolerogenic properties of APC induced by 1,25(OH)2D3 and analogs has been shown in allograft rejection models with the sterol given orally [37] or following the passive transfer of 1,25(OH)2D3 treated APC [41].

C. Effects of 1,25(OH)2D3 on Animal Models of Autoimmunity Table I illustrates the most representative studies of 1,25(OH)2D3 and analogs in animal models of autoimmunity (excluding diabetes, Chapter 99). The first one, a long-standing useful model for the study of autoimmune diseases, has been the animal model of multiple sclerosis, EAE. In this model, immunization of susceptible mice or rats with central nervous system proteins will induce a progressive paralysis in the recipients within two weeks. Developments in peptide technology have led to a higher rate of disease induction in the susceptible recipients [42]. There is strong evidence that EAE is a Th1-mediated disease since antigen-specific Th1 cells can transfer disease [43]. Moreover, at the peak of the disease, there is a predominance of Th1 cytokines (IL-2 and IFN-γ) in the central nervous system of the mice; during remission, IL-10 prevails, suggesting a Th2 predominance [44]. In this model, 1,25(OH)2D3 and analogs can prevent the induction and the relapses of the disease [31]. While 1,25(OH)2D3 clearly exerts a Th1 inhibitory effect in favor of a pro-Th2 cytokine secreting effect in vitro, this dichotomy is harder to demonstrate in vivo [45–47]. A particularly interesting model is the experimental lupus of MRL/l mice. Lupus is an autoimmune disorder that leads to the formation and deposition of immune complexes throughout the body. Sites of predilection include the kidneys, causing nephritis, often with renal failure, and the skin, causing rash and inflammation. A potential role of Th1-mediated IgG2a in the pathogenesis of the disease was suggested by

JACQUES LEMIRE

treatment of MRL/1 mice with anti-IgM anti-sera from birth. This resulted in a depletion of IgG2a antibodies and prevented the development of skin, but not glomerular lesions [48]. A significant increase (eightfold) in IgG2a-producing cells is observed in MRL/1 mice between two to five months of age [49]. In response to thymus-dependent antigens, the IgG subclass profiles in all systemic lupus erythematosus (SLE) mice differ from those of normals, with a predominance of IgG2b and IgG2a rather than IgG1. In addition, sera of the majority of MRL/1 mice contain rheumatoid factors that react most strongly with IgG2a [50], The dependence of IgG2a secretion on Th1 cells [5], as well as class II expression secondary to IFN-γ secretion by Th1 cells, would suggest an important role for the Th1 cell subset in the pathogenesis of experimental SLE. The administration 1,25(OH)2D3 from an early age in these mice could completely inhibit the development of skin lesions, characteristics of this animal model [51]. The development of nephritis and resulting proteinuria was not prevented with the same treatment. However, a subsequent study suggested that a diet with a normal to high calcium content (0.87%) administered to MRL/l mice undergoing similar therapy with 1,25(OH)2D3 could even prevent the development of nephritis (proteinuria) [52]. The animal model of inflammatory bowel disease, experimental murine inflammatory bowel disease, provides additional information about the potential mechanisms of action of the sterol besides reducing the severity of the disease. IL-10 knockout mice, made vitamin D–deficient, develop a severe wasting syndrome; treatment with 1,25(OH)2D3 improved symptoms and prevented the progression of existing disease [61]. These observations suggest once again that the primary target of 1,25(OH)2D3 is through inhibition of Th1 cell activity rather than direct stimulation of the Th2 pathway.

D. Effects of 1,25(OH)2D3 in Autoimmunity in Humans The first and most studied application for 1,25(OH)2D3 in an autoimmune disease in humans is psoriasis and the experience is reviewed in Chapter 101. Limited application of the hormone for other autoimmune disease in humans results from the intrinsic hypercalcemic properties of the hormone, restricting the therapeutic potential of 1,25(OH)2D3. Two significant advances have held promising perspectives for the use of the steroid hormone in humans: the synergistic properties of 1,25(OH)2D3 with known immunosuppressive agents, such as corticosteroids [67]

1757

CHAPTER 98 Vitamin D3: Autoimmunity and Immunosuppression

TABLE I 1,25 (OH)2D3 and Analogs in Autoimmunity

Organ

Dose (µg/kg)a

Outcome, treated/ controls (measure)

Serum calcium (mg/dl)

1,25(OH)2D3

5/2d

80%/5% (survival)

9.7

Lemire [53]

5/2d 7.5/2d

1/4 (disease activity) 1/4 (disease activity)

11.2 9.7

Lemire et al. [54]

2.5/2d

3.3/5 (disease activity)

8.7

Lemire et al. [55]

EAE/mouse

1,25(OH)2D3-16-ene D3 1,25(OH)2-24-oxo16-ene D3 1,25(OH)2D3-16-ene-23ene-26,27hexafluroro D3 MC1288

0.2/2d

10.2

Lemire et al. [56]

EAE/mouse

Ro 63-2023

240/2d

10.0

Mattner et al. [31]

Experimental autoimmune thyroiditis mouse Adjuvant arthritis/rat Murine Lyme arthritis Collagen-induced arthritis Collagen-induced arthritis Experimental Murine Inflammatory Bowel Disease Heymann nephritis/ rat

1,25(OH)2D3

0.2/d

25%/92.8% (disease incidence) 100%/50% (survival) 1.4/3.1 (relapses) 50%/85.7% (histologic incidence)

N/Ab

Fournier et al. [57]

1,25(OH)2D3

0.2/d

12.0

Boissier et al. [58]

1,25(OH)2D3

1/d

11.2

Cantorna et al. [59]

1,25(OH)2D3

2.5/d

7.9

Cantorna et al. [59]

MC 1288

0.1/d

2.6 mmol

Larsson [60]

1,25(OH)2D3

0.25/d

11.9/16.9 (arthritic score) 0.2/0.23 cm (ankle size) 0%/100% (disease incidence) 50%/100% (disease incidence) 1.7/3.0 (histology score)

3 mmol

Cantorna [61]

1,25(OH)2D3

0.5/d

11.8

Branisteau et al. [62]

KH1060

0.5/d

Mercury chlorideinduced nephritis/rat

1,25(OH)2D3

0.1/d

KH1060

0.3/d

Nephrotoxic serum nephritis/rat Lupus nephritis/ mouse Lupus nephritis/ mouse Lupus nephritis/ mouse Lupus nephritis/ mouse Lupus/MRL mouse

1,25(OH)2D3

0.5/d

1,24R(OH)2D3

0.1/d

OCT

0.002–0.1/d 10–40%/80% (proteinuria incidence) 5/2d < 4/>6 (urinary protein/ creatinine ratio) 5–10/d 0–0.2/2–2.5 (severity score) 5/2d None/present (skin lesions)

Model species

Nervous EAE/mouse system EAE/mouse EAE/mouse EAE/mouse

Thyroid Joint

Bowel

Kidney

Kidney

Skin

ad,

daily; 2d, every second day. not available.

bN/A:

Vitamin D3 or analog

1,25(OH)2D3 1,25(OH)2D3 1,25(OH)2D3

80/210 (mg urinary protein/day) 210/210 (mg urinary protein/day) 180/780 (mg urinary protein/day) <20/780 (mg urinary protein/day) <50/300 (mg protein/day) 1.0/3.3 (dipstick)

Ref.

11 >14

Lillevang et al. [63]

>14 11

Hattori [64]

N/A

Koizumi et al. [65]

10.4

Abe et al. [66]

8

Lemire et al. [51]

N/A

Deluca [52]

8

Lemire et al. [51]

1758

Oral calcitrol and total skin score Patients with systemis scleroderma

40

Total skin score [mean & SE]

and cyclosporine [57,58,68,69], and the development of 1,25(OH)2D3-analogs with reduced hypercalcemic activity. These exciting possibilities have yet to be applied in the treatment of autoimmune conditions. The ideal application for 1,25(OH)2D3 in humans is a disease in which a significant autoimmune component plays a role and no efficacious treatment is available. Systemic scleroderma might represent such a disease. While the cause of the disease is unknown, the pathogenesis is multifactorial [70]. An early immunologic trigger may lead to expansion of fibrogenic clones of tissue fibroblasts accompanying clinical expansion with early migration of CD4+ and CD8+ cells with a preponderance of CD4+ cells in the skin of affected patients [71,72]. These activated T-cells express HLA-DR molecules, IL-2 receptors, and increased CD4+ markers [73,74]. By precisely suppressing those immune mechanisms, 1,25(OH)2D3 could play a role in modulating the disease. A tolerability and feasibility study of calcitriol in the treatment of systemic sclerosis was done in 10 patients with a diagnosis of systemic sclerosis [75]. One patient withdrew within two weeks of the study secondary to drug intolerance. The 9 patients (M/F ratio of 2/7; age: 44 ± 14 years), with a disease duration of 6.9 ± 4.6 years, completed the six-month trial. After initial assessment and instructions to patients to follow a low calcium diet (800 mg/day), calcitriol was started at low dosage (0.25 or 0.5 mcg/day) and increased on a monthly basis to the maximum tolerable dosage, i.e. urinary calcium excretion of <350 mg/24 hours and/or serum calcium <10.8 mg/dl. Patients were assessed monthly with total skin score (Modified Rodnan Score, 17 body parts, scale 1–3), maximum oral aperture, and fist closure. Vitamin D metabolism was assessed at entry and at termination of the study and included: serum levels of 1,25(OH)2D3, 25-OH-D, intact PTH, serum calcium, and urinary calcium excretion. Safety monitoring included: measurements of serum calcium and urinary calcium excretion within one week following any change in dosage of calcitriol; renal function assessed by clearance of creatinine (24-hour urine collection), and renal ultrasound (to exclude nephrocalcinosis or nephrolithiasis). This was an unblinded study without a placebo control, done primarily for safety reasons and to establish a dose for a placebo-controlled study. The maximum tolerated dosage of calcitriol was 1.4 ± 0.5 µg/day (ranging from 0.5 to 2 µg/day). The mean intrapatient change (mean ± SE) of total skin score from baseline to end of study was −10.8 ± 3.3, a change of significant magnitude. Figure 3 illustrates the mean total skin score of the whole group per month during the course of the study.

JACQUES LEMIRE

32

24

16

8

0 0

1

2

3

4

5

6

Months of therapy

FIGURE 3

Individual total skin score (Modified Rodnan Score, 17 body parts, scale 1–3) of 9 patients with systemic sclerosis from entry to end of the 6-month treatment with calcitriol. Data are expressed as mean total skin score and standard error from the mean. *p < 0.05.

The most pronounced effect of calcitriol was observed on the change in the total skin score but some other physical variables improved while others did not change in therapy. While calcium excretion increased in all patients treated with calcitriol, no patient developed hypercalcemia, nephrocalcinosis, or nephrolithiasis. At the maximal tolerated dose of calcitriol, further reduction of calcium from the diet or awareness to comply to a strict diet improved the degree of hypercalciuria. Table II illustrates the biochemical and radiologic parameters studied.

TABLE II Vitamin D Metabolism and Renal Function of Patients Treated with Calcitriol* Parameters Serum calcium (mg/dl) Urinary calcium (mg/day) Intact PTH (pg/ml) 1,25(OH)2D3 (pg/ml) 25(OH)D (ng/ml) Renal function Serum creatinine (mg/dl) Creatinine clearance (ml/min) Renal ultrasound

Entry

Termination

9.3 ± 0.4 112 ± 47 26.9 ± 12.9 44.4 ±14.2 25.8 ± 10.4

9.1 ± 0.37 270 ± 134** 25.8 ± 27.2 62.7 ± 20.8** 25.1 ± 10.4

0.6 ± 0.1 129 ± 47 normal (9/9)

0.7 ± 0.2 115 ± 37 normal (9/9)

*calcitriol administered at bedtime and blood drawn in the morning. **p < 0.001.

1759

CHAPTER 98 Vitamin D3: Autoimmunity and Immunosuppression

TABLE III

Scleroderma

Patients/ treatment duration

Treatment of Scleroderma with 1,25(OH)2D3 Serum calcium (mg/dl)

Dosage

Outcome

7/6 mo

0.25 µg/d × 1 wk

Decreased induration of slerotic lesions

Localized

7/2 to 24 mo

0.75 µg/d by 3 wks × 6 mo 1.75 µg/d

Localized and systemic Systemic

20 (localized) 0.75 µg/d × 6 mo 7 (systemic)/ 1.25 µg/d × 3 mo 9 mo 11/6 to 36 mo 1.75 µg/d

Systemic

10/7 mo

Localized

Systemic

3/6 mo

0.5 to 0.75 µg/d

0.5 to 0.75 µg/d

N/Ab

Urine calciuma (ratio or mg/d) 0.35

Ref. Elst [76]

(mean, end) Improved skin extensibility Clearing of lesions (4 pts) No difference

N/A

N/A

Humbert et al. [77]

9.2

0.81 (mean, end)

Hulshof [78]

Increased skin extensibility Improved oral aperture and flexion Decreased induration of slerotic lesions

<10.5

<300 mg/d

Humbert et al. [79]

N/A

230 (mg/d) Hulshof [80] (mean, during treatment)

10.8

100–400 (mg/d) Caca [81]

Joint mobility improvement Decreased induration of slerotic lesions Increased skin extensibility Joint mobility improvement

anormal

values. ratio: <0.2. 24 hrs <400 mg/d. bN/A: not available.

This study suggests a beneficial role of calcitriol in this human autoimmune disease. Such a role has been described so far in numerous studies of both localized and systemic scleroderma. Table III summarizes these results. These results suggest that calcitriol can improve some of the manifestations of a severe autoimmune process when administered to patients. The calcium mobilizing properties of the hormone significantly restrict its optimal therapeutic potential since the dosages used are threefold higher than typically used for replacement. This disease, scleroderma, provides further indication that not all organ systems would be equally sensitive to the action of 1,25(OH)2D3 , at least at the tolerated doses. For example, induration and elasticity of the skin is more susceptible to the positive action of calcitriol than internal organs (lungs, gastrointestinal tract, for example). It is therefore imperative that a doubleblind, randomized, placebo-controlled trial of calcitriol in the treatment of systemic sclerosis be initiated to confirm a therapeutic application for 1,25(OH)2D3.

Then, when the development of analogs of 1,25(OH)2D3 with reduced hypercalcemic activity finally expands for other human application, the ultimate potential of the vitamin D compounds in the treatment of autoimmunity could be fully evaluated. In conclusion, calcitriol exerts significant therapeutic as well as preventive actions in many animal models of autoimmunity. A potential role of calcitriol has emerged in some human autoimmune diseases, such as psoriasis (Chapter 101) and scleroderma. The development of fewer calcemic analogs of calcitriol for human application should expand its role in autoimmune diseases.

References 1. D’Andrea A, Rengaraju M, Valiante NM, Chehimi J, Kubin M, Aste M, Chan M, Kobayashi M, Young D, Nickbarg E, Chizzonite R, Wolf SF, Trinchieri G 1992 Production of natural killer cell stimulatory factor (interleukin-12) by peripheral blood mononuclear cells. J Exp Med 176:1387–1398.

1760 2. Manetti R, Parronchi P, Guidizi MG, Piccinni MP, Maggi E, Trinchieri G, Romagnani S 1993 Natural killer cell stimulatory factor [interleukin 12 (IL-12)] induces T-helper type 1 (Th1specific immune responses and inhibits the development of IL-4 producing Th cells. J Exp Med 177:1199–1204. 3. Mosmann TR, Coffman RL 1989 Th1 and Th2 cells: Different patterns of lymphokine secretion lead to different functional properties. Annu Rev Immunol 7:145–173. 4. Cher DJ, Mosmann TR 1987 Two types of murine helper T-cell clone. II. Delayed-type hypersensitivity is mediated by Th1 clones. J Immunol 138:3688–3694. 5. Koizumi T, Nakao Y, Matsui T, Nakagawa I, Matsuda S, Komoriya K 1985 Effects of corticosteroid and l,24Rdihydroxyvitamin D3 administration on lymphoproliferative and autoimmune disorders in MRL/MP-Lpr/lpr mice. Int Arch Allergy Appl Immunol 77:396–404. 6. Gazzinelli R, Oswald I, Hieny S, James S, Sher A 1992 The microbicidal activity of interferon-y treated macrophages against Trypanosoma cruzi involves an L-arginine– dependent, nitrogen oxide-mediated mechanism inhibitable by interleukin-10 and transforming growth factor/3. Eur J Immunol 22:2501–2506. 7. Mosmann TR 1991 Cytokine secretion patterns and crossregulation of T-cell subsets. Immunol Res 10:183–188. 8. Mosmann TR, Moore KW 1991 The role of IL-10 in crossregulation of TH1 and TH2 responses. Immunol Today 12: 49–53. 9. D’Andrea A, Aste-Amerzaga M, Valiante NM, Ma X, Kubin M, Trinchieri G 1993 Interleukin 10 (IL-10) inhibits human lymphocyte interferon −/− production by suppressing natural killer cell stimulatory factor/IL-12 synthesis in accessory cells. J Exp Med 178:1041–1048. 10. Basham TY, Merigan TC 1984 Recombinant interferon-γ increases HLA-DR synthesis and expression. J Immunol 130: 1492–1494. 11. Salgame P, Abrams JS, Clayberger C, Goldstein H, Convitt J, Modlin RL, Bloom BR 1991 Differing lymphokine profiles and functional subsets of human CD4 and CD8 T cell clones. Science 254:279–281. 12. Del Prete G, De Carli M, Ricci M, Romagnani S 1991 Helper activity for immunoglobulin synthesis of T-helper type 1 (Th1) and Th2 human T-cell clones is limited by their cytolytic capacity. J Exp Med 174:809–813. 13. De Carli M, D’Elios MM, Zancuoghi G, Romagnani S, Del Prete G 1994 Human Th1 and Th2 cells: Functional properties, regulation of development, and role in autoimmunity. Autoimmunity 18:301–308. 14. Cox FEG, Liew FY 1992 T-cell subsets and cytokines in parasitic infections. Immunol Today 13:445–448. 15. Bach JF 2002 Immunorégulation et auto-immunité. J de la Société de Biologie 196:255–258. 16. Takahashi T, Tagami T, Yamazaki S, Uede T, Shimizu J, Sakaguchi N, Mak TW, Sakaguchi S 2000 Immunologic selftolerance maintained by CD25(+)CD4(+) regulatory T-cells constitutively expressing cytotoxic T lymphocyte associated antigen 4. J Exp Med 192:303–310. 17. Tisch R, Wang B, Serreze DV 1999 Induction of glutamic acid decarboxylase 65-specific Th2 cells and suppression of autoimmune diabetes at late stages of disease is epitope dependent. J Immunol 163:1178–1187. 18. Bach JF, Chatenoud L 2001 Tolerance to islet autoantigens and type I diabetes. Annu Rev Immunol 19:131–161. 19. Adorini L 2001 Selective immunointervention in autoimmune diseases: lessons from multiple sclerosis. J Chemother 13: 219–234.

JACQUES LEMIRE

20. Lemire JM 1992 Immunomodulatory role of 1,25-dihydroxyvitamin D3. J Cell Biochem 49:26–31. 21. Alroy L, Towers T, Freedman L 1995 Transcriptional repression of the interleukin-2 gene by vitamin D3: direct inhibition NFATp/AP1 complex formation by a nuclear hormone receptor. Mol Cell Biol 15:5789–5799. 22. Takeuchi A, Reddy G, Kobayashi T, Okano T, Park J, Sharma S 1998 Nuclear factor of activated T-cells (NFAT) as a molecular target for 1α,25-dihydroxyvitamin D3-mediated effects. J Immunol 160:209–218. 23. Barrat FJ, Cua DJ, Boonstra A, Richards DF, Crain C, Savelkoul EF et al. 2002 In vitro generation of interleukin 10-producing regulatory CD4(+) T-cells is induced by immunosuppressive drugs and inhibited by T-helper type 1 (Th1)- and Th2-inducing cytokines. J Exp Med 195:603–616. 24. Spiegelberg HL, Beck L, Stevenson DD, Ishioka GY 1994 Recognition of T-cell epitopes and lymphokine secretion by rye grass allergen lolium perenne I-Specific human T-cell clones. J Immunol 152:4706–4711. 25. Lemire JM, Archer DC, Beck L, Spiegelberg HL 1995 Immunosuppressive actions of 1,25-dihydroxyvitamin D3. Preferential inhibition of Th1 functions. J Nutr 125:1704S–1708S. 26. Lemire JM, Adams JS 1992 1,25-dihydroxyvitamin D3 inhibits the passive transfer of cellular immunity by a myelin basic protein specific T-cell clone. J Bone Miner Res 7: 171–177. 27. Boonstra A, Barrat FJ, Crain C, Heath VL, Savelkoul HF, O’Garra A 2001 1α,25-dihydroxyvitamin D3 has a direct effect on naive CD4+ T-cells to enhance the development of Th2 cells. J Immunol 167:4974–4980. 28. Cantorna MT, Humpal-Winter J, DeLuca HF 2000 In vivo upregulation of interleukin 4 is one mechanism underlying the immunoregulatory effects of 1,25-dihydroxyvitamin D3. Arch Biochem Biophys 377:135–138. 29. Staeva-Vieira TP, Freedman LP 2002 1,25-dihydroxyvitamin D3 inhibits IFN-gamma and IL-4 levels during in vitro polarization of primary murine CD4+ cells. J Immunol 168: 181–1189. 30. Nashold FE, Hoag KA, Goverman J, Hayes CE 2001 Rag-1– dependent cells are necessary for 1,25-dihydroxyvitamin D3 prevention of experimental autoimmune encephalomyelitis. J Neuroimmunol 119:16–29. 31. Mattner F, Smiroldo S, Galbiati F, Muller M, Di Lucia P, Poliani PL, Martino G, Panina-Bordignon P, Adorini L 2000 Inhibition of Th1 development and treatment of chronicrelapsing experimental allergic encephalomyelitis by a nonhypercalcemic analog of 1,25-dihydroxyvitamin D3. Eur J Immunol 30:498–508. 32. Meehan MA, Kerman RH, Lemire JM 1992 1,25-dihydroxyvitamin D3 enhances generation of nonspecific suppressor cells while inhibiting the induction of cytotoxic cells in a human MLR. Cell Immunol 140:400–409. 33. Lemire JM, Ince A, Cox P, Kohl S 1990 1,25-dihydroxyvitamin D3 selectively interferes with cellular mechanisms of cytotoxicity. Kidney Intl 37:420. 34. Merino F, Alvarez-Mon M, De la Hera A, Ales JR, Bonilla F, Durantez A 1989 Regulation of natural killer cytotoxicity by 1,25-dihydroxyvitamin D3. Cell Immunol 118:328–336. 35. Leung KH 1989 Inhibition of human natural killer cell and lymphokine-activated killer cell cytotoxicity and differentiation by vitamin D3. Scand J Immunol 30:199–208. 36. Lemire JM 1997 The role of Vitamin D3 in immunosuppression: Lessons from autoimmunity and transplantation In: D Feldman, F Glorieux, JW Pike (eds) Vitamin D. Academic Press: San Diego, pp. 1167–1181.

CHAPTER 98 Vitamin D3: Autoimmunity and Immunosuppression 37. Gregori S, Casorati M, Amuchastegui S, Smiroldo S, Davalli AM, Adorini L 2001 Regulatory T-cells induced by 1α,25dihydroxyvitamin D3 and mycophenolate mofetil treatment mediate transplantation tolerance. J Immunol 167:1945–1953. 38. Rescigno M, Piguet V, Valzasina B, Lens S, Zubler R, French L, et al. 2000 Fas engagement induces the maturation of dendritic cells (DCs), the release of interleukin (IL)-1beta, and the production of interferon gamma in the absence of IL-12 during DC-T-cell cognate interaction: a new role for Fas ligand in inflammatory responses. J Exp Med 192:1661–1668. 39. D’Ambrosio D, Cippitelli M, Cocciolo MG, Mazzeo D, Di Lucia P, Lang R, et al. 1998 Inhibition of IL-2 production by 1,25-dihydroxyvitamin D3. Involvement of NF-kappaB down-regulation in transcriptional repression of the p40 gene. J Clin Invest 101:252–262. 40. Adorini L 2002 Immunomodulatory effects of vitamin D receptor ligands in autoimmune diseases. Intl Immunopharm 2:1017–1028. 41. Griffin MD, Lutz W, Phan VA, Bachman LA, McKean DJ, Kumar R 2001 Dendritic cell modulation by 1α,25-dihydroxyvitamin D3 and its analogs: a vitamin D receptor-dependent pathway that promotes a persistent state of immaturity in vitro and in vivo. Proc Natl Acad Sci USA 22:22. 42. Tuohy VK, Lu Z, Sobel RA, Laursen RA, Lees MB 1989 Identification of an encephalitogenic determinant of myelin proteolipid protein for SJL mice. J Immunol 142:1523–1527. 43. Baron JL, Madri JA, Ruddle NH, Hasbim G, Janeway CA Jr 1993 Surface expression of a4 integrin by CD4 T-cells is required for their entry into brain parenchyma. J Exp Med 177:57–68. 44. Kennedy MK, Torrance DS, Picha KS, Mohler KM 1992 Analysis of cytokine mRNA expression in the central nervous system of mice with experimental allergic encephalomyelitis reveals that IL-10 mRNA expression correlates with recovery. J Immunol 149:2496–2505. 45. Racke MK, Bonomo A, Scott DE, Cannella B, Levine A, Raine CS, et al. 1994 Cytokine-induced immune deviation as a therapy for inflammatory autoimmune disease. J Exp Med 180:1961–1966. 46. Furlan R, Poliani PL, Galbiati F, Bergami A, Grimaldi L, Comi G, et al. 1998 Central nervous system delivery of interleukin 4 by a nonreplicative herpes simplex type 1 viral vector ameliorates autoimmune demyelination. Hum Gene Ther 9: 2605–2617. 47. Bettelli E, Prabhu Das M, Howard ED, Weiner HL, Sobel RA, Kuchroo VK 1998 IL-10 is critical in the regulation of autoimune encephalomyelitis as demonstrated by studies of IL-10- and IL-4-deficient and transgenic mice. J Immunol 161:3299–3306. 48. Berney T, Fulpius T, Shibata T, Reininger L, Van Snick J, Shan H, Weigert M, Marshak-Rothstein A, Izuu S 1992 Selective pathogenicity of murine rheumatoid factors of the cryoprecipitable IgG3 subclass. Int Immunol 4:93–99. 49. Slack JH, Hang LM, Barkley J, Fulton RJ, O’Hoostelaere L, Robinson A, Dixon FJ 1984 Isotypes of spontaneous and mitogen-induced auto-antibodies in SLE-prone mice. J Immunol 132:1271–1275. 50. Theofilopoulos AN, Balderas RS, Hang LM, Dixon FJ 1984 Monoclonal IgM rheumatoid factors derived from arthritic MRL/ Mp-lpr/lpr mice. J Exp Med 158:901–906. 51. Lemire JM, Ince A, Takasima M 1992 1,25-dihydroxyvitamin D3 attenuates the expression of experimental murine lupus of MRL/1 mice. Autoimmunity 12:143–148. 52. DeLuca HF, Cantorna MT 2001 Vitamin D: Its role and uses in immunology. FASEB 15:2579–2585.

1761

53. Lemire JM, Archer DC 1991 1,25-dihydroxyvitamin D3 prevents the in vivo induction of murine experimental autoimmune encephalomyelitis. J Clin Invest 87:1103–1107. 54. Lemire JM, Archer DC, Reddy GS 1994 l,25-dihydroxy-24oxo-16ene-vitamin D3, a renal metabolite of the vitamin D analog 1,25-dihydroxy-16ene-vitamin D3, exerts immunosuppressive activity equal to its parent without causing hypercalcemia in vivo. Endocrinology 135:2818–2821. 55. Lemire JM, Archer DC, Beck L, Reddy GS, Uskokovic MR, Spiegelberg HL 1994 The role and mechanism of vitamin D analogs in immunosuppression. In: Norman AW, Bouillon R, Thorn-asset M (eds) Vitamin D. A Pluripotent Steroid Hormone: Structural Studies, Molecular Endocrinology and Clinical Applications, de Gruyter: Berlin, New York, pp. 531–539. 56. Lemire JM, Archer DC, Binderup L 1997 The vitamin D analog, 20-epi-l,25(OH)2D3 (MC1288), prevents the onset of autoimmunity with reduced calcemic properties. (Unpublished observation.) 57. Fournier C, Gepner P, Sadouk M, Charreire J 1990 In vivo beneficial effects of cyclosporin A and 1,25-dihydroxyvitamin D3 on the induction of experimental autoimmune thyroiditis. Clin Immunol Immunopathol 54:53–63. 58. Boissier MC, Chiocchia G, Fournier C 1992 Combination of cyclosporin A and calcitriol in the treatment of adjuvant arthritis. J Rheumatol 19:754–757. 59. Cantorna MT, Hayes CE, DeLuca HF 1998 1,25-dihydroxycholecalciferol inhibits the progression of arthritis in murine models of human arthritis. J Nutr 128:68–72. 60. Larsson P, Mattson L, Klareskog L, Johnsson C 1998 A vitamin D analog (MC 1288) has immunomodulatory properties and suppresses collagen-induced arthritis (CIA) without causing hypercalcaemia. Clin Exp Immunol 114:277–283. 61. Cantorna MT, Munsick C, Bemiss C, Mahon BD 2000 1,25dihydroxycholecalciferol prevents and ameliorates symptoms of experimental murine inflammatory bowel disease. J Nutr 130:2648–2652. 62. Branisteau DD, Leenaerts P, van Damme B, Bouillon R 1993 Partial prevention of active Heymann nephritis by 1α,25-dihydroxyvitamin D3. Clin Exp Immunol 94:412–417. 63. Lillevang ST, Rosenkvist J, Andersen CB, Larsen S, Kemp E, Kristensen T 1992 Single and combined effects of the vitamin D analog KH1060 and cyclosporin A on mercuric chlorideinduced autoimmune disease in the BN rat. Clin Exp Immunol 88:301–306. 64. Hattori M 1990 Effect of lα,25(OH)2D3 on experimental rat nephrotoxic serum nephritis. Nippon Jinzo Gakkai Shi 32: 147–159. 65. Koizumi T, Nakao Y, Matsui T, et al. 1985 Effects of corticosteroid and l,24.R-dihydroxyvitamin D3 administration on lymphoproliferative and autoimmune disorders in MRL/MPLpr/lpr mice. Int Arch Allergy Appl Immunol 77:396–404. 66. Abe J, Takita Y, Nakano T, Miyaura C, Suda T, Nishii Y 1990 22-Oxa-1a-25-dihydroxyvitamin D3: A new synthetic analog of vitamin D having a potent immunoregulating activity without inducing hypercalcaemia in mice. In: DV Cohn, FH Glorieux, TJ Martin (eds) Calcium regulation and bone metabolism. Elsevier: Amsterdam, pp. 146–151. 67. Jirapongsananuruk O, Melamed I, Leung DY 2000 Additive immunosuppressive effects of 1,25-dihydroxyvitamin D3 and corticosteroids on TH2, but not TH2, responses. J Allergy and Clin Immunol 106:981–985. 68. Van Etten E, Branisteanu DD, Verstuyf A, Waer M, Bouillon R, Mathieu C 2000 Analogs of 1,25-dihydroxyvitamin D3 as dose-reducing agents for classical immunosuppressants. Transplantation 69:1932–1942.

1762 69. Redaelli CA, Wagner M, Günter-Duwe D, Tian YH, Stahel PF, Mazzucchelli L, Schmid RA, Schilling MK 2002 1α,25-dihydroxyvitamin D3 shows strong and additive immunomodulatory effects with cyclosporine A in rat renal allotransplants. Kidney Intl 61:288–296. 70. Furst DE, Clements PJ 1997 Hypothesis for the pathogenesis of systemic sclerosis. J Rheumatol 24:53–57. 71. Kahari VM 1994 Activation of dermal connective tissue in scleroderma. Ann Intern Med 25:511–518. 72. Prescott RJ, Freemont PW, Jones CJ 1992 Sequential dermal microvascular and perivascular changes in the development of scleroderma. J Pathol 166:255–263. 73. Fioccao U, Rosada M, Cozzi L, et al. 1993 Early phenotypic activation of circulating helper memory T-cells in scleroderma: Correlation with disease activity. Ann Rheum Dis 52: 272–277. 74. Umehara H, Kumagai S, Ishida H, Suginoshita T, Macda M, Imura H 1988 Enhanced production of interleukin-2 in patients with progressive systemic sclerosis. Hyperactivity of CD-4 positive cells? Arthritis Rheum 31:401–407. 75. Santana-Sahagun JE, Lemire J, Albani S, Eichenfield L, Weisman MH 1999 Safety and tolerability of calcitriol therapy as a disease-modifying agent in systemic sclerosis: an open pilot study. Arthritis and Rheumatism 42:S187, Abs. 701.

JACQUES LEMIRE

76. Elst EF, van Suijlekom-Smit LWA, Oranje AP 1999 Treatment of 1,25-dihydroxyvitamin D3 (calcitriol) in seven children. Ped Dermatol 16:53–58. 77. Humbert P, Delaporte E, Dupond JL, Rochefort A, Laurent R, Drobacheff C, De Wazieres B, Bergoend H, Agache P 1994 Treatment of localized scleroderma with oral 1,25-dihydroxyvitamin D3. Eur J Dermatol 4:21–23. 78. Hulshof MM, Bouwes Bavinck JN, Bergman W, Masclee AAM, Heickendorff L, Breedveld FC, Dijkmans BAC 2000 Doubleblind, placebo-controlled study of oral calcitriol for the treatment of localized and systemic scleroderma. J Am Acad Dermatol 43:1017–1023. 79. Humbert P, Dupond JL, Agache P, Laurent R, Rochefort A, Drobacheff C, de Wazieres B, Aubin F 1993 Treatment of scleroderma with oral 1,25-dihydroxyvitamin D3: Evaluation of skin involvement using noninvasive techniques. Acta Derm Venereol (Stockh) 73:449–451. 80. Hulshof MM, Pavel S, Breedveld FC, Dijkmans BAC, Vermeer BJ 1994 Oral calcitriol as a new therapeutic modality for generalized morphea. Arch Dermatol 130:1290–1293. 81. Caca-Biljanovska NG, Vlckova-Laskoska MT, Dervendi DV, Pesic NP, Laskoski DS 1999 Treatment of generalized morphea with oral 1,25-dihydroxyvitamin D3. RheumaDerm, edited by Mallia and Uitto, pp. 299–304.

CHAPTER 99

Vitamin D and Diabetes CHANTAL MATHIEU, CONNY GYSEMANS, AND ROGER BOUILLON Onderwijsen Navorsing, U.Z. Gasthuisberg, Leuven, Belgium

I. Introduction II. Vitamin D and the β Cell III. Vitamin D and the Immune System in Type 1 Diabetes Mellitus

IV. Vitamin D Receptor Polymorphism and the Risk for Diabetes V. Clinical Perspectives References

I. INTRODUCTION

osteoblasts from diabetics to 1,25(OH)2D3 in contrast to the near normal responsiveness of the intestinal mucosa [13]. The osteoblast response, however, can be quickly normalized by insulin infusion [19], whereas IGF-I-therapy is partly effective both on 1,25(OH)2D3 concentration [20] and osteoblast function [18]. Insulin and IGF-I thus have marked effects on the kinetics of 1,25(OH)2D3 (production and catabolism) as well as on the sensitivity of classical target tissues (bone, intestine, kidney) to this secosteroid hormone. Indeed, Verhaeghe et al. showed in ovariectomized rats that rhIGF-I (250 µg/d) delivered sc for 28 days increased both plasma IGF-I and 1,25(OH)2D3 concentrations, as well as plasma osteocalcin and urinary pyridinolines [21]. The present chapter will focus mainly on the effects of vitamin D and its activated form, 1,25(OH)2D3, on the susceptibility for diabetes and its role in pathogenesis of the disease itself. Diabetes mellitus is a common disease in the Western world, with an estimated prevalence of 4 to 5%. The majority (95%) of the diabetic patients suffer from type 2 diabetes or noninsulin dependent diabetes, a metabolic syndrome characterized by insulin resistance and relatively inadequate insulin production by the β cell in the islets of Langerhans of the pancreas [22]. In this metabolic syndrome, it is still unclear whether the primary dysfunction is to be situated in the peripheral target organs of insulin (being mainly liver, fat, and muscle) or in the β cell [23]. Insulin resistance is induced by obesity and sedentary lifestyle and is also involved in the pathogenesis of cardiovascular disease. This insulin resistance is probably the major determinant of the disease, but β cell dysfunction is always present and will determine the severity of the clinical presentation. Type 1 diabetes, also known as juvenile or insulin dependent diabetes mellitus, is in etiology a totally different disease.

The influence of insulin on vitamin D and bone homeostasis is complex. Insulin enhances the 1α-hydroxylation of 25-hydroxyvitamin D in vitro [1], whereas insulin-like growth factor (IGF)-I also mediates the effects of hypophosphatemia on renal 1,25(OH)2D3 production [2]. Diabetic rats with either spontaneous diabetes [3,4] or streptozotocin-induced diabetes [5–7] have decreased serum concentrations of 1,25(OH)2D3. Insulin-dependent diabetic children and also poorly controlled diabetic adults have decreased serum concentrations of 1,25(OH)2D3 [8–11], whereas good diabetes control normalizes circulating 1,25(OH)2D3 [9,12]. Most of the effects of insulin deficiency on serum 1,25(OH)2D3 can be explained by decreased concentrations of the vitamin D–binding protein (DBP) [5,13]. Indeed, the DBP-corrected or calculated free concentrations of 1,25(OH)2D3 are either normal or slightly increased in both diabetic animals and humans [13]. Kinetic studies in spontaneously diabetic BB (biobreeding) rats clearly showed increased metabolic clearance (>50%) of 1,25(OH)2D3 but normal production rates. No such studies have been performed in human diabetics. The classical vitamin D target tissues are, however, differently affected by insulin deficiency: the basal intestinal active calcium absorption and duodenal calbindin D concentration are decreased [3,14–16], but calbindin and intestal calcium absorption are still responsive to low calcium intake or 1,25(OH)2D3 therapy [13]. The renal calbindin D-28K (CaBP) is not affected by insulin deficiency. At the bone level, a marked decrease in osteoblast number and function has repeatedly been observed in both diabetic animals and humans [17,18]. Therapy with even high doses of 1,25(OH)2D3 clearly shows marked resistance of the VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

1764

CHANTAL MATHIEU, CONNY GYSEMANS, AND ROGER BOUILLON

It has become clear in recent years that this disease is an autoimmune disorder, related to Graves’ disease, characterized by a destruction of the insulin producing β cells in the pancreas by the body’s own immune system [24]. Whereas type 2 diabetes is a typical disease of the obese and aging patient, type 1 diabetes mainly occurs in children and adolescents. Since receptors for 1,25(OH)2D3, the activated form of vitamin D, have been described in the main cells involved in the pathogenesis of both types of diabetes [25,26], scientific and clinical interest has focused on these molecules with respect to their potential role in the pathogenesis of the diseases, but even more with respect to their therapeutic potential in the prevention of the diseases [25,27,28]. The main effects of vitamin D and its activated form on the β cell, with direct implications for the pathogenesis and prevention of mainly type 2 but also type 1 diabetes, will be described and discussed. Furthermore, the role of vitamin D and 1,25(OH)2D3 in the prevention of type 1 diabetes and its effects on the immune system of the animal models for the disease will be covered. Special attention is also given to the emerging evidence on correlations between vitamin D receptor (VDR) polymorphisms and risk for type 1 and type 2 diabetes. Older and novel data on the possible role of vitamin D deficiency in the pathogenesis of both diseases will be discussed.

(but not the other pancreatic hormones) secretion and induces glucose intolerance, while addition of vitamin D in the diet restores the abnormalities [31,37– 42]. The effects of vitamin D deficiency and repletion on glucose homeostasis in vivo are, however, diverse and do not only involve the β cell. Rachitic animals do not have a normal food intake, lose weight, and cannot maintain a normal plasma calcium level. Moreover, these metabolic changes by themselves impair directly Ca2+ handling in the β cell and provoke β cell dysfunction and glucose intolerance [43,45]. Data from VDR knockout (VDR-KO) mice are conflicting, with some groups reporting impaired glucose tolerance [46], while in other reports no impairment of the glucose metabolism is reported [47]. Here, the background of the mouse in which the VDR-KO is introduced is of critical importance. Figure 1A shows a normal intraperitoneal glucose tolerance test (IPGTT) in VDR-KO mice in a C57Bl/6 background. In the animal model of type 1 diabetes, the NOD (nonobese diabetic) mouse, vitamin D deficiency leads to impaired glucose tolerance, with decreased insulin secretion of the islets in vitro and earlier onset of diabetes in vivo, leading to a final increased diabetes incidence [Giulietti et al., unpublished data] [48]. In this model, however, the β cell as well as immune factors may contribute to the increase in diabetes incidence.

II. VITAMIN D AND THE β CELL

The in vivo observations, where it is unclear whether the effects of vitamin D deficiency on glucose tolerance were direct effects of the deficiency on the β cell, were extended by in vitro data. Islets isolated from rachitic animals show impaired insulin release when cultured in vitro and challenged with glucose [49,50]. The abnormalities induced by vitamin D deficiency in vivo can be abolished by co-culturing the islets in the presence of high concentrations of 1,25(OH)2D3 [31,49–51]. Even more convincing are the data on islets isolated from normal animals where most papers show a stimulation of insulin release upon glucose challenge in the presence of high doses of 1,25(OH)2D3 [34,51,52]. Interpretation of the findings is sometimes obscured by the many different methods used (such as incubation time with 1,25(OH)2D3, animal source of islets, type of glucose challenge), but overall an improved β cell function is observed in the presence of 1,25(OH)2D3. Interestingly, some authors also studied the effects of 1,25(OH)2D3 on insulin synthesis, and even this element contributing to normal β cell function is normalized or increased by 1,25(OH)2D3 [37,53]. Progressively, insight is gained in the mechanism by which 1,25(OH)2D3 might act on insulin secretion: a significant rise in cytosolic calcium levels is observed

A. Metabolic Effects Since the early observations in 1980 [29] by Norman et al. that pancreatic insulin secretion is selectively inhibited by vitamin D deficiency, several reports have demonstrated an active role for vitamin D and especially its activated form, 1,25(OH)2D3, in the regulation of the function of the endocrine pancreas, especially the β cell. Receptors for 1,25(OH)2D3 have been described in β cells [30]. Not only does the classical 1,25(OH)2D3 receptor seem to be present, but interesting observations also suggest the existence of a receptor localized in the membrane [30–35]. Apart from the vitamin D receptors, also the effector part of the vitamin D machinery is present: immune-reactive vitamin D–dependent CaBP has been demonstrated in β cells [36]. 1. EFFECTS OF VITAMIN D DEFICIENCY IN VIVO

The early observations mainly focused on the effects of vitamin D deficiency in vivo in several animal models and humans on insulin secretion and glucose tolerance [29,37–44]. Vitamin D deficiency clearly impairs insulin

2. EFFECTS OF VITAMIN D METABOLITES IN VITRO

1765

CHAPTER 99 Vitamin D and Diabetes

A

B 30000 VDR KO VDR HE VDR WT

300

Insulin secretion (pmol/l)

Glycemia (mg/dl)

400

200

100

25000 20000 15000 10000 5000 0

0 0

15

30

60

90

120

0

5

Time (min)

10 20 Glucose (mM)

30

Insulin content (× 10−4 pmol/islet)

C 25 20 15 10 5 0 0

5

10 20 Glucose (mM)

30

FIGURE 1 Intraperitoneal glucose tolerance test (A) in VDR KO, VDR (HE) and VDR WT mice. Glycemia was measured at different time points after glucose challenge. No difference in glycemia was noted between KO, HE and WT mice after glucose injection. (B & C) In vitro insulin secretion (B) and synthesis (C) of islets isolated from VDR KO, VDR HE, and VDR WT mice after incubation with different concentrations of glucose (0–30 mM). Again no differences were measured between VDR KO, HE and WT mice either for insulin secretion or insulin synthesis.

after incubation with 1,25(OH)2D3, resulting in insulin secretion. Controversy still exists on whether only an influx of extra-cellular calcium is responsible for this rise or whether also mobilization of intracellular calcium reserves is involved [30,34,49,50]. In the older papers, the genomic pathway was mainly studied and effects on calcium mobilization and insulin secretion were only observed after long incubation periods [49–52]. Recently, the non-genomic pathway, involving a putative plasmalemmal receptor inducing rapid calcium fluxes also has been implicated in the effects of vitamin D on insulin secretion [30–34]. The importance of calcium in this system is demonstrated by our own data in NOD mice that were vitamin D–deficient in early life, but maintained normal calcium levels. Here, isolated islets exposed to

glucose in vitro had perfectly normal insulin synthesis and secretion (Fig. 2A and B). In VDR-KO mice again the background on which the KO is introduced seems essential. Whereas some authors report abnormal insulin synthesis and secretion, VDR-KO mice on C57Bl/6 background have normal insulin synthesis and secretion (Fig. 1B and C). 3. EFFECTS OF VITAMIN D METABOLITES IN VIVO—CLINICAL IMPLICATIONS

Based on the observations in vitro and in animal models of vitamin D deficiency in vivo, clinical trials with vitamin D or 1,25(OH)2D3 on glucose metabolism have been performed. There are several reports demonstrating that vitamin D–depleted humans have reduced insulin secretion while, as expected, vitamin D treatment

1766

CHANTAL MATHIEU, CONNY GYSEMANS, AND ROGER BOUILLON

B

A

25 Insulin content (× 10−4 pmol/islet)

Insulin secretion (pmol/l)

30000 ctr NOD vit D def NOD

25000 20000 15000 10000 5000

20 15 10 5 0

0 0

5

10 20 Glucose (mM)

30

0

5

10 20 Glucose (mM)

30

FIGURE 2

In vitro insulin secretion (A) and synthesis (B) of islets isolated from vitamin D–deficient NOD mice and control NOD mice. With the purpose of inducing vitamin D deficiency in NOD mice, 3-wk-old NOD mice were kept in UV-free surroundings and received a vitamin D–depleted diet containing 2% calcium and 1.25% phosphorus. Control mice were kept under similar conditions except their chow was supplemented with 2500 IU/kg vitamin D. From 100 days of age, all mice were fed control chow. No differences were measured between control and vitamin D–deficient mice either for insulin secretion or insulin synthesis.

(and calcium) reverses the vitamin D–deficient state and restores glucose tolerance [54]. However, the dose given and the administration route are often different between the studies. In vitamin D–depleted humans, 1,25(OH)2D3 supplements of 2000 IU per day [55] or a single intramuscular injection of 100,000 IU [56] were administered, which makes comparison between the studies hazardous. Interesting are the studies where the effects of 1,25(OH)2D3 repletion in the relatively 1,25(OH)2D3– deficient state of uremia were investigated [57]. In a study by Allegra et al., uremic patients with low 1,25(OH)2D3 serum levels were treated with 0.5µg of 1,25(OH)2D3 (plus 500 mg of calcium) per day for 21 days, and IVGTTs were performed prior to and after treatment. Interestingly, 1,25(OH)2D3 (plus 500 mg of calcium) caused an increment in glucose-induced insulin response only in the first few minutes of the stimulation (in uremia especially this rapid phase of insulin release is disturbed). In this study, however, repletion of 1,25(OH)2D3 could not completely reverse the glucose intolerance. Orwoll et al. performed an interesting pilot study on possible clinical applications of 1,25(OH)2D3 treatment in a situation of impaired insulin secretion without vitamin D deficiency [58]. Type 2 diabetic patients received 1,25(OH)2D3 (1 µg/d for 4 days) in a double-blinded, placebo-controlled cross-over trial. Clear effects of 1,25(OH)2D3 treatment were noted on parameters of calcium metabolism, but this study was unable to determine if hypovitaminosis D increases the risk of developing type 2 diabetes.

Since this could be due to the short duration of 1,25(OH)2D3 treatment, this approach should not be discarded. Recently, another pilot trial showed beneficial effects of vitamin D supplements on first phase insulin secretion in type 2 diabetic women [59]. However, it was reported by Taylor and Wise that vitamin D supplementation in three cases of British Asians with vitamin D deficiency and type 2 diabetes led to increased insulin resistance and a deterioration of glycemic control [60]. In childhood diabetes, several epidemiological studies describe a correlation between a north-south gradient and the incidence of disease, as well as an inverse correlation between monthly hours of sunshine and the incidence of diabetes [61]. Also, a seasonal pattern of disease onset is well described in type 1 diabetes [62]. Dietary vitamin D supplementation is often recommended in pregnant women and children to prevent vitamin D deficiency. Besides the study of Stene et al., who demonstrated that cod liver oil taken during pregnancy could reduce the risk for type 1 diabetes in their offspring (odds ratio = 0.30), the EURODIAB group suggested an association between vitamin D supplementation in infancy and a decreased risk for type 1 diabetes (odds ratio = 0.69) in a multicenter casecontrol study [63]. Moreover, Hypponen et al. reported that intake of 2000 IU of vitamin D during the first year of life diminished the risk of developing type 1 diabetes [64]. This study also confirmed that suspected rickets was associated with higher incidence for childhood diabetes (odds ratio = 2.6). Protection against

1767

CHAPTER 99 Vitamin D and Diabetes

type 1 diabetes mellitus by vitamin D supplements may be due to a combination of immune effects and β-cell protection. In NOD mice or BB rats, no clear protection against type 1 diabetes can be seen by neonatal and early life supplements of regular vitamin D [65]. However, in NOD mice, clear preservation of insulin content in β-cells was observed.

B. Effects of Vitamin D on β Cell Characteristics In the pathogenesis of type 1 diabetes, β-cell damage by cytokines and other inflammatory agents might play an important role. Sandler et al. demonstrated that 1,25(OH)2D3 and some of its new structural analogs counteract the suppressive effects of IL-1β on β-cell function, such as insulin synthesis and insulin secretion [66]. Similar effects were observed in the case of IFNγ-induced impairment of β-cell function. Hahn et al. demonstrated that 1,25(OH)2D3 and some of its analogs can prevent the IFNγ-induced decrease in insulin synthesis and secretion [53]. These effects were observed at high concentrations of 1,25(OH)2D3 (5–10nM). On the other hand, Mauricio et al. could not demonstrate any effects of 1,25(OH)2D3 on IL-1β induced β-cell dysfunction [67]. The main difference between the papers describing a protective effect and this latter study is the incubation time: in the former paper, incubation periods ranging between 48 and 72 hours were used, while in the latter study islets were only incubated with 1,25(OH)2D3 for 24 hours. Considering observations made in similar conditions [49–52], this incubation timing might be the crucial point in the effect of 1,25(OH)2D3. Not only the effects of cytokines on β-cell function are altered by 1,25(OH)2D3, but also the induction of surface markers by these cytokines appears to be blocked [53,68,69]. When neonatal rat islets were incubated with IFNγ, several surface markers such as MHC class II molecules and adhesion molecules (ICAM-I) were up-regulated as expected, Co-incubating the islets with 1,25(OH)2D3 or one of its analogs decreased markedly the up-regulation of MHC class II molecules after IFNγ stimulation (15% MHC class II positive β-cells versus 30% in the presence of IFNγ only, p < 0.001). Similar results were obtained with an analog of 1,25(OH)2D3, ZXY 1106 [53]. Riachy et al. demonstrated that incubation of human islet cells with IL-1β, IFNγ, and TNFα, decreased the insulin content after 48 hours of culture while leaving islet viability unchanged. However, 6 days after cytokine treatment, islet viability decreased to 60%. 1,25(OH)2D3 treatment of human pancreatic islets incubated with IL-1β, IFNγ, and TNFα increased MHC class I expression, as well as nitric oxide (NO)

and IL-6 release [68]. Recently, this group published that vitamin D3 was able to induce and maintain high levels of A20, an anti-apoptotic protein, in rat RINm5F cells and human islets after treatment with the three inflammatory cytokines [69]. These data showing β-cell protection against inflammatory agents involved in the pathogenesis of type 1 diabetes may have direct implications for the observed in vivo effects of 1,25(OH)2D3 and its analogs in prevention of type 1 diabetes in animal models.

III. VITAMIN D AND THE IMMUNE SYSTEM IN TYPE 1 DIABETES MELLITUS Prevention of juvenile diabetes is one of the major goals in health care for the future. When strategies for prevention of a disease are to be developed, a first requisite is the knowledge of the pathogenesis of the disease. From studies in two animal models for type 1 diabetes, the NOD mouse and the BB rat, but also in humans, it has become clear that type 1 diabetes can be considered an autoimmune disease, where a central role in the destruction of the β-cell is played by the body’s own immune system. In this immune system action, almost all cells (monocytes/macrophages, T lymphocytes, B lymphocytes, NK cells, dendritic cells) play a role. Up to now, most prevention studies have been carried out in the NOD mouse and can be divided into several major categories: pure immune suppression, immune modulation, antigen-(specific) tolerance induction, and β-cell protection. Results in NOD mice are promising for many of these treatments, but many obstacles to human applications still exist. All studies involving long-term immune suppression are inconceivable as strategy for the prevention of a chronic disease striking mainly children, and moreover, the preliminary results of these drugs in recent onset diabetic patients are disappointing. At this moment, a major interest is focused on immune modulation and β-cell protection, two characteristics of 1,25(OH)2D3 and many of its newer analogs. Since 1,25(OH)2D3 can be produced by monocytes/ macrophages/dendritic cells, and since receptors are present in several immune cells, a physiological role for this substance as a messenger or cytokine-like molecule between cells of the immune system is probable and therapeutic possibilities in the prevention of this autoimmune disease are to be expected. Moreover, as described above, clear β-cell protective effects of 1,25(OH)2D3 against several inflammatory agents involved in β-cell destruction have been observed.

1768

CHANTAL MATHIEU, CONNY GYSEMANS, AND ROGER BOUILLON

A. Primary Prevention of Type 1 Diabetes in Animal Models of Type 1 Diabetes by 1,25(OH)2D3 and Its Analogs 1. EARLY INTERVENTION—EFFECTS OF 1,25(OH)2D3 ON DIABETES PREVENTION

Two good animal models for type 1 diabetes in humans have been described up to now: the BB rat and the NOD mouse. Both strains develop a disease quite similar to human diabetes in a spontaneous manner, but the BB rat is characterized by severe immune abnormalities, and therefore most studies on pathogenesis and the development of therapeutic strategies have been performed in the NOD mouse. The NOD mouse was derived from a cataract developing substrain of the out-bred JcI-ICR mouse by selective inbreeding from 1974 to 1980 [70]. Diabetes develops spontaneously between the 12th and the 30th week of age, proceeded by the histological lesion of the disease, called insulitis. Insulitis is the reflection of the infiltration of the islets of Langerhans by a mixture of immune cells. Chronic administration of pharmacological doses of 1,25(OH)2D3 can reduce the incidence of both insulitis and diabetes in NOD mice [71,72] (Table I). When, 1,25(OH)2D3 was administered at a dose of 5 µg per kg body weight every two days from the age of weaning (21 days) until the end of the life of the NOD mice (200 days), insulitis, the histopathological lesion of type 1 diabetes, was reduced from 80% in control mice to 58% (p < 0.05) in 1,25(OH)2D3–treated mice. Diabetes itself was reduced from 56% to 8% (p < 0.001). Although treatment was globally well tolerated, hypercalcemia and bone decalcification were seen. These data were confirmed by other groups [48]. Extensive immunological screening, including FACS analysis of major lymphocyte subsets, evaluation of T-cell proliferation (after CD3 or Con A stimulation), and NK-cell function, was unable to discover major changes in the treated versus control mice. A major finding was the restoration of a well-known defect of the

NOD mouse: the absence of suppressor cell function. The restoration of suppressor cell function could be demonstrated both in vitro and in vivo [72,73]. Adorini et al. demonstrated that the nature of the regulator cell induced by 1,25(OH)2D3 or its analogs is most likely a CD4+CD25+ cell [74,75]. The question remains however whether this restoration of suppressor cells is the main mechanism involved in protection against diabetes by 1,25(OH)2D3, since not only was there protection against diabetes, but also against insulitis, pointing towards interference with the induction of auto-immunity itself. Diabetes occurring in NOD mice after cyclophosphamide administration is believed to be due to an elimination of regulator cells [76,77]. The time needed for recovery of the immune cells after injection of a high dose of cyclophosphamide is different between various T-cell populations. Longlived effector T-cells recover more promptly than the short-lived suppressor or regulator cells [78]. Several potential preventive therapies for diabetes have already been tested in this cyclophosphamide model. Protection against diabetes by therapeutic interventions that are believed to induce suppressor T-cells but that have no effect on autoimmune effector cells can be broken by cyclophosphamide. This was shown for the protection against diabetes achieved by Complete Freund’s Adjuvant (CFA) or for the prevention of recurrence of diabetes in syngeneic islet grafts by BCG [79,80]. On the other hand, cyclophosphamide could not reverse the protection against diabetes in animals treated with the streptococcal preparation, OK432, as this substance prevents diabetes by inhibiting the generation of effector cells for β cell destruction [81]. Casteels et al. have demonstrated that 1,25(OH)2D3 can also prevent diabetes induced by cyclophosphamide in the NOD mouse [82]. Cyclophosphamide induced diabetes (78%) and insulitis (100%) in control animals. When 1,25(OH)2D3 was administered from day 21 until day 69, protection against diabetes (17%, p < 0.01) and insulitis (42%, p < 0.005) was observed. This protection was achieved

TABLE I 1,25(OH)2D3 and its Analogs in Animal Models of Diabetes and Islet Transplantation Animal models of diabetes Diabetes type 1 low-dose streptozotocin-induced Islet transplantation Pancreatic islets

Major effects

References

Prevention of disease, inhibition of insulitis and diabetes Decreased incidence of insulitis and diabetes

[71,72,82,104,133] [88]

Induction of allo-transplantation tolerance, prevention of autoimmune recurrence after syngeneic transplantation, prevention of early xenograft failure

[122,123,134,135]

CHAPTER 99 Vitamin D and Diabetes

despite the total elimination of suppressor cells in the 1,25(OH)2D3– treated group by cyclophosphamide, as shown by co-transfer experiments. 1,25(OH)2D3 treatment also did not interfere with the quantitative and qualitative recovery of the major lymphohematopoietic cells after cyclophosphamide injection. Striking again was the absence of insulitis in most animals treated with 1,25(OH)2D3. The combination of these data (resistance against cyclophosphamide and reduction of insulitis) together with the absence of protection in co-transfer experiments suggested that indeed cyclophosphamide had eliminated all suppressor cells, but that these suppressor cells were not the main protective mechanism in the 1,25(OH)2D3– treated NOD mice. The basis of protection by 1,25(OH)2D3 seems to be more a reshaping of the immune repertoire with elimination of effector cells, but also the direct β-cell protective effects of 1,25(OH)2D3 might play a major role in disease prevention. The reshaping of the immune system involves more specifically a shift of production of T-cell cytokines from predominantly Th1 (IL-2, IFNγ) in control mice to Th2 (IL4, IL10) in 1,25(OH)2D3 or analog-treated mice [83]. Moreover, this shift appeared to be antigen-specific and most probably due to a direct interference of 1,25(OH)2D3 or its analogs with the antigen presenting dendritic cells. Indeed, 1,25(OH)2D3 induces a reshaping of the dendritic cells towards tolerogenic cells [75,84,85]. We even demonstrated that dendritic cells generated in the presence of 1,25(OH)2D3 or an analog can redirect already committed T-cell clones derived from a type 1 diabetic patient towards nonproliferation [85]. In the NOD mouse, the reshaping of the immune system happens centrally, in the thymus, where treatment with 1,25(OH)2D3 restores the sensitivity of T lymphocytes towards apoptosis inducing signals, thus allowing better elimination of auto-immune effector cells [86]. Studies on diabetes prevention in the NOD mouse model, with spontaneous development of type 1 diabetes, are probably the most relevant for direct application of the findings in the human situation. But the fluctuating incidence of diabetes in the stock mouse colonies and the duration of most intervention studies makes this model not optimal for the screening of large groups of new potentially therapeutic agents. Some researchers have therefore looked for an easier model to study and especially a quicker model for screening new drugs. Such a fast model is the multiple low-dose streptozotocin (MLDSZ) model [87]. Streptozotocin is an antibiotic produced by Streptomyces achromogenes that has a specific β-cell toxic effect. A single high dose of streptozotocin (70–250 mg/kg body weight) causes a fast and complete destruction of β-cells in most species.

1769 The administration of multiple subdiabetic doses of streptozotocin causes more subtle β-cell damage, thus triggering a nonspecific inflammatory reaction in the islets that is then followed by insulin deficiency. A major criticism of this model is the fact that the complete scenario of β-cell destruction in this model is unclear and that diabetes is probably the result of nonspecific inflammatory damage of the β-cell together with other islet cells. Therefore, this diabetes model is not a true autoimmune model. However the question remains whether in some humans type 1 diabetes is a true autoimmune disease and not a consequence of nonspecific β-cell destruction by one or other inflammatory conditions (e.g. viral infection). Inaba et al. used the high and the MLDSZ model to test the effect of 1α-hydroxyvitamin D3, a precursor of 1,25(OH)2D3, on diabetes prevention [88]. When diabetes was chemically induced by a single injection (200 mg/kg body weight) of streptozotocin, no protection against diabetes was seen. When, however, multiple low doses of streptozotocin were administered, 1α-hydroxyvitamin D3 reduced the diabetes incidence dramatically in a dose-dependent manner: control mice developed diabetes in 100% of the cases, while administration of 0.4 µg/kg of 1α-hydroxyvitamin D3 reduced the diabetes incidence to 46% (p < 0.01). Administration of 0.3 µg also provided protection (diabetes incidence of (61%, p < 0.025), but administration of 0.2 µg/kg of 1α-hydroxyvitamin D3 did not prevent diabetes. Data on toxicity in this study are unfortunately limited to evolution of body weight (unchanged), and no indication of calcemic or bone effects are given. Histologic examination of the pancreas of the experimental mice demonstrated that 1α-hydroxyvitamin D3 also reduced insulitis in this model. The BB rat is the other spontaneous animal model for type 1 diabetes. Data on immune pathogenesis of type 1 diabetes in the BB rat are less abundant than in the NOD mouse, and observations are hampered by the severe lymphopenia and T-cell dysfunction occurring in these animals [89–91]. The existence of different pathogenetic mechanisms for autoimmune diabetes in animal models suggests the existence of different pathogenetic scenarios in the human situation. This is also suggested by the apparent contradictory findings sometimes seen in different patient populations, e.g., different genetic characteristics between Japanese and Caucasian diabetic patients, differences in immune characteristics between the patients with very early onset of the disease versus later onset of the disease, and finally the differences observed between patients with isolated type 1 diabetes and patients with multiorgan autoimmune involvement.

1770 In the BB rat, no difference in diabetes incidence between control (33%) and 1,25(OH)2D3–treated rats (24%) was observed when 1,25(OH)2D3 was administered from weaning until 120 days (0.8 µg/kg every other day) [92]. These findings in the BB rat again confirm the basic differences in disease pathogenesis that can be found between the two available animal models for type 1 diabetes and also indicate that caution is warranted when transferring findings from either of these models to the human situation. 2. LATE INTERVENTION — EFFECTS OF 1,25(OH)2D3 ON DIABETES PREVENTION

The NOD mouse is considered a good model for human diabetes and allows for testing not only new therapeutic drugs, but also offers the opportunity to elaborate the optimal and achievable treatment modalities. A relevant question, for instance, is whether longterm 1,25(OH)2D3 treatment is necessary or whether a short-term intervention would suffice for disease prevention and if so, when this short-term treatment should be administered. We, therefore, designed an experiment where NOD mice received 1,25(OH)2D3 in different time windows: one group received 1,25(OH)2D3 for their whole lifetime (from weaning until 200 days of age), another group was treated only during their youth (from weaning until 100 days of age), and a final group was treated only from 100 days of age until 200 days, All mice were followed up until 200 days, and the dose of 1,25(OH)2D3 was 5 µg per kg body weight administered every other day in all groups. A control group vehicle received arachis oil. As expected, insulitis incidence at 200 days was lowered from 93% to 54% (p < 0.025) in the long-term treated group while also diabetes incidence was reduced from 86% in this control group to 13% in the 1,25(OH)2D3–treated group (p < 0.001). Mice treated only during youth showed a significantly lower insulitis incidence (46%, p < 0.05 versus the control group), and diabetes occurred in 30% of these mice (p < 0.001). When therapy was initiated at day 100 (when control mice have insulitis in about 75%), insulitis was present in 90% and 80% of these mice developed diabetes by day 200. These values are not significantly different from the control values. When however, not the endpoint of diabetic occurrence but the timing of diabetes onset was compared, the treated animals showed a delayed onset of diabetes compared to the control group (p < 0.01). In the group treated from 100 until 200 days of age, effects on serum calcium levels were comparable to the “long-term” treated group; in the group treated from 21 until 100 days of age, the serum calcium level was normal as expected. Bone turnover (reflected in serum osteocalcin levels) in the animals treated with 1,25(OH)2D3 from 21

CHANTAL MATHIEU, CONNY GYSEMANS, AND ROGER BOUILLON

until 200 days of age was increased as previously demonstrated, while the bone turnover in the two groups with shorter treatment duration was closer to the normal range. In the group receiving the treatment in their youth, these values were still elevated. Bone calcium content determination reflected again the severe impact of 1,25(OH)2D3 treatment in youth since the bone calcium content of these animals even 100 days after stopping the treatment was still significantly decreased compared to the control group (p < 0.001). This period seems to be crucial in bone remodeling, and interference with bone modeling in this period leaves traces for the rest of life. On the other hand, when treatment was only initiated at 100 days of age, important bone loss under 1,25(OH)2D3 treatment was seen, although less than with the long-term treatment (p < 0.01). At the moment, clinical trials are underway where 1,25(OH)2D3 is administered to newly diagnosed type 1 diabetic patients, with residual C-peptide levels, and thus residual β cell function [A. Ziegler, P. Pozzilli, personal communication]. Read-out of the trials is preservation of C-peptide levels. Great caution should observed with this kind of trial since the doses administered are at the edge of toxicity, and, as the NOD model suggests, only a delay of disease progression is to be expected. 3. ANALOGS OF 1,25(OH)2D3

A major obstacle to human application of 1,25(OH)2D3 is its important effect on calcium and bone metabolism. New structural analogs of 1,25(OH)2D3 with less effects on calcium metabolism, but more pronounced immunological effects have been developed, especially through side chain modifications [75,93–97]. The actions of this family of molecules on the immune system are exerted via receptors for 1,25(OH)2D3 (VDR) that are present in several immune cells, such as dendritic cells, monocytes/macrophages, activated T lymphocytes and B lymphocytes. In vivo, these immune effects result in protection against autoimmunity and prolongation of allograft survival (see chapter 36) [71,72,75,98–104]. Several of the most promising of these analogs coming from different chemical laboratories have been tested in the model of spontaneous type 1 diabetes in the NOD mouse [93]. The mechanism of protection against insulitis and diabetes appears to be similar to that of 1,25(OH)2D3. Effects of the analogs on dendritic cell phenotype, regulator cell induction, and β cell protection have been described [75]. Exposure to analogs of 1,25(OH)2D3 enhances both chemotactic and phagocytotic capacity of monocytes/ macrophages, while their antigen-presenting cell function decreases. Moreover, 1,25(OH)2D3 alters significantly the antigen-presenting cell function of dendritic cells. Treatment with analogs of

1771

CHAPTER 99 Vitamin D and Diabetes

1,25(OH)2D3, both during in vitro and in vivo maturation, inhibits the surface expression of MHC class II and of co-stimulatory molecules, such as CD86, CD80, and CD40 [85]. In the search for the optimal analog a combination of β cell protection, immune modulation and low calcemic effects is sought. Until now, several analogs are promising, but before embarking on long-term interventions in high-risk individuals for type 1 diabetes, long-term safety data will have to be gathered. 4. COMBINATION WITH OTHER IMMUNE MODULATORS

In animal models of type 1 diabetes, such as the NOD mouse, disease prevention can be achieved by chronic use of immune suppressants such as cyclospoine CsA ([105]) (Table II). Such an approach cannot be applied to humans because of chronic side TABLE II

effects. Common adverse effects of calcineurin inhibitors like CsA and tacrolimus (FK506) are bowel disturbance, nephrotoxicity, neurotoxicity, as well as serious side effects on pancreatic β cell function, for example, reduced insulin synthesis and β-cell toxicity [106]. Most immunosuppressive agents target T-cells, or some, like mycophenolate mofetil (MMF), both T and B cells. Conversely, no immunomodulatory agent in clinical use specifically targets antigen-presenting cells and in particular dendritic cells, which are known to be involved in T cell activation and tolerance induction. In NOD mice, diabetes can be prevented by 1,25(OH)2D3 and its nonhypercalcemic analogs when treatment is started before insulitis is present [72,104]. A critical question for the applicability of these analogs in the human situation is whether the analogs of 1,25(OH)2D3 can arrest progression to clinically overt

1,25(OH)2D3 and Its Analogs Combined In Vitro and In Vivo with Other Immune Suppressants

Drugs

CsA Rapamycin FK506 MMF Leflunomide Dexamethasone

CsA

Rapamycin MMF IFN-β

References In vitro models Proliferation and IL-2 production of PHA-stimulated PBMC Proliferation and cytokine production in MLR Proliferation of PHA-stimulated PBMC Proliferation of PHA-stimulated PBMC Proliferation and cytokine production in MLR Proliferation of PHA-stimulated PBMC Proliferation of PHA-stimulated PBMC Proliferation and cytokine production of anti-CD3-stimulated PBMC Proliferation, cytokine and chemokine production and T cell activation of dendritic cells Induction of regulatory IL-10 producing T lymphocytes

[141]

In vivo models Experimental autoimmune encephalomyelitis Type 1 diabetes Mercuric-chloride induced autoimmunity Adjuvant arthritis Thyroiditis Transplantation of syngeneic islets Transplantation of xenogeneic islets Transplantation of vascularized renal allografts Transplantation of liver allografts Bone marrow transplantation Transplantation of aorta allografts Transplantation of heart allografts Transplantation of skin allografts Experimental autoimmune encephalomyelitis Experimental autoimmune encephalomyelitis Transplantation of allogeneic islets Transplantation of syngeneic islets

[109,142] [108,133] [143,144] [145] [101] [110,134,146] [135] [147,148] [149] [150] [151] [152] [153] [138] [109] [122] [123]

[108,109,120,136,137] [109,138] [109,120] [109] [109] [130,140]

1772 diabetes if administered when active β cell destruction is already present, which is the situation in prediabetic subjects in whom immune intervention is considered [107]. Casteels et al. demonstrated that some of these analogs, when combined with a short induction course of a classical immunosuppressant such as CsA, can arrest the progression of the disease even when administered after autoimmunity has already started [108]. This approach of combining a short induction course with a classical immunosuppressant and a nonhypercalcemic analog of 1,25(OH)2D3 is very promising and might open new perspectives in the prevention of autoimmune diabetes in humans. Several other combinations of vitamin D analogs and other immune modulators have been tested both in vitro and in vivo. Using the median effect analysis of Chou and Tallaylay, five top candidates for a combination therapy with 1,25(OH)2D3 could be selected: CsA, FK506, rapamycin, leflunomide, and MMF. These therapies had a respective combinatorial index for 50% suppression of PHA induced human T-cell proliferation of 0.16 ± 0.06; 0.27 ± 0.12; 0.36 ± 0.34; 0.39 ± 0.12 and 0.43 ± 0.30, with index <1 representative for synergism [109]. In the model of experimental autoimmune encephalitis (EAE) in SJL mice the in vitro results could be translated to the in vivo situation (see Chapter 98). More novel immune modulators, such as type I interferons (IFNα/β), which have a broad range of immunomodulatory properties (mainly inhibition of T-cell IFNγ production by blocking IL-12 secretion of dendritic cells and restriction of T-cell proliferation in part through down-regulation of IL-12 messages or up-regulation of IL-10 levels), also gave interesting results in combination with nonhypercalcemic analogs of 1,25(OH)2D3 in EAE and models of type 1 diabetes [van Etten et al. unpublished data and see following paragraph].

B. Secondary Prevention of Type 1 Diabetes in the NOD Mouse by Analogs of 1,25(OH)2D3—Prevention of Recurrence of Autoimmune Diabetes after Islet Transplantation Type 1 diabetes is characterized not only by an autoimmune destruction of the body’s own β-cells, but also by the formation of an autoimmune memory. The latter phenomenon is responsible for the destruction of MHC matched or syngeneic β cells, transplanted under the form of isolated β cells, islets, or whole pancreas [110–116]. This disease recurrence explains why in clinical pancreas and islet transplantation in type 1 diabetic patients, extensive immune suppression is needed: relatively high doses of several immune suppressants

CHANTAL MATHIEU, CONNY GYSEMANS, AND ROGER BOUILLON

are needed not only to prevent allograft rejection, but also to break the autoimmune memory [117–119]. Some analogs of 1,25(OH)2D3 have been tested for their capacity to prevent disease recurrence after islet transplantation in spontaneously diabetic NOD mice. The most spectacular results were obtained with a combination of KH1060 (20-epi-22-oxa-24,26, 27-trishomo-1,25-(OH)2D3) (see Chapter 84) and subtherapeutic doses of CsA [120,121]. Administration of high doses of the analog (1 µg/kg/2d) alone was as effective in delaying islet loss as the highest tolerable dose of CsA (15 mg/kg/d), but eventually the disease recurred in all mice. In the group receiving KH1060 (0.5 µg/kg/2d) together with CsA (7.5 mg/kg/d), a synergistic effect between both drugs was seen: 4 of 7 mice maintained a functioning graft for 60 days and more importantly, these animals did not show recurrence for at least 30 days after stopping the treatment. All treatment was administered from the day before transplantation until diabetes recurrence or in the case of normoglycemia until 60 days after transplantation. Insulin content determinations of the graft and native pancreas of the recipient clearly demonstrated that normoglycemia was the result of graft survival and not of recovery of the β-cells of the recipient’s own pancreas. Insulin content in pancreases of recurring and nonrecurring mice was comparable and showed noregeneration of the original β-cells (0.0125 ± 0.012 pmol/mg in recurring versus 0.008 ± 0.004 pmol/mg in nonrecurring mice, NS), while the insulin content in the grafts showed a clear difference between recurring and nonrecurring mice (45 ± 27 pmol/graft in recurring versus 1285 ± 106 pmol/graft in nonrecurring mice, p < 0.00001). The highest dose of KH1060, as well as the highest dose of CsA, had clear toxic effects on the general condition of the animals as reflected by the course of the weight. However, by giving KH1060 in a fractionated way (1 µg/kg/2 days instead of 0.5 µg/kg/d) these, cumulative high doses were well-tolerated for long periods of time. The subtherapeutic doses of KH1060 and CsA were nontoxic and had minor effects on serum calcium and osteocalcin levels. The combination of both drugs was also well-tolerated and had similar effects on serum calcium, but resulted in clear effects on osteocalcin levels, indicating also a synergistic effect on bone remodeling. Bone calcium content in all treatment groups was decreased. In the KH1060 groups, this effect was more important and dose-dependent. Unfortunately, as already reflected by the osteocalcin levels, the combined subtherapeutic doses of KH1060 and CsA also act synergistically on bone remodeling. Caution in interpreting these results is warranted since treatment duration in tested animals in the combination group lasted for 60 days, while the animals tested in the other groups only received treatment for shorter periods (± 20 days).

1773

CHAPTER 99 Vitamin D and Diabetes

Adorini et al. have reported a similar synergism between 1,25(OH)2D3 or an analog and mycophenolate mofetil (MMF) [122]. A novel approach can be found in combinations of analogs of 1,25(OH)2D3 with other natural immune modulators such as IFN β. We demonstrated that subtherapeutic doses of rIFN β alone (1 × 105 IU/d) had minor effects on autoimmune diabetes recurrence after islet transplantation (20.8 ± 14.2 days, NS vs. 10.8 ± 2.9 days in controls). However, interestingly, a combination of rIFNβ with TX527 (14-epi-19-nor-20-epi-23-yne-1,25-(OH)2D3) maintained islet graft function in 100% of mice during treatment (start: the day before islet transplantation and continued until 20 days (rIFNβ) or 30 days (TX527) after transplantation) and resulted in a marked delay of autoimmune diabetes recurrence (61.6 ± 19.6 days, p < 0.005 vs. controls) [123]. We also demonstrated that rIFNβ in combination with TX527 results in an inhibition of the Th1 pathway (IL-12, IL-2 and IFNγ), which is known to be associated with the pathogenesis of organ-specific autoimmune diseases. In addition, enhanced expression of Th2 cytokine IL-110 by rIFNβ in combination-therapy with the TX527 analog was observed.

IV. VITAMIN D RECEPTOR POLYMORPHISM AND THE RISK FOR DIABETES Novel insights on the possible role of vitamin D/ 1,25(OH)2D3 in the pathogenesis of type 1 and type 2 diabetes come from the epidemiological data on correlations between VDR polymorphisms and risk of the disease in certain populations. Vitamin D exerts its genomic effects mainly via the nuclear VDR. VDR is expressed in many tissues, including pancreatic islets [30]. The gene encoding the VDR is located in humans on chromosome 12cen-q12 and shows extensive polymorphism including a FokI polymorphism in exon II, BsmI, and ApaI allelic variants in the intron between exons VIII and IX, a TaqI restriction fragment length polymorphism in exon IX, and a mononucleotide [(A)n] repeat polymorphism in the 3′ untranslated region (see Chapter 68). More than 15 reports in different populations show correlations between some of these polymorphisms and type 1 or type 2 diabetes. McDermott et al. demonstrated that excessive transmission of the BsmI alleles affects South Indian subjects with type 1 diabetes, linking BsmI polymorphism of the VDR to an increased risk for type 1 diabetes [124]. This association was confirmed in different populations [125–128]. A study by Ban et al. revealed an association between VDR initiation codon polymorphism (FokI) and GAD65 positivity in the Japanese

population [129]. The meaning of these correlations between genetic markers of vitamin D metabolism and type 1 diabetes remains unclear. Still, the protective effects of vitamin D on the β cell should not be neglected and might be an explanation of why correlations are observed, not only between VDR polymorphisms and type 1 diabetes but also type 2 diabetes. Indeed, a study on Bangladesh Asians demonstrated that ApaI polymorphisms influence insulin secretion in response to glucose [130]. Also Oh and Barrett-Conner investigated VDR polymorphism and susceptibly for type 2 diabetes in a community-based study of unrelated older adults without known diabetes [131]. Their research suggested that ApaI polymorphism may be associated with higher fasting plasma glucose and prevalence of glucose intolerance. More recently, genotyping for TaqI, ApaI, BsmI, and FokI revealed that BsmI in young males with low physical activity (≤3 h per week) was associated with high levels of fasting glucose [132]. From these data, vitamin D and VDR polymorphisms may have a potential role in the pathogenesis of diabetes. Until now, no clear explanation for these associations has been given. The FokI polymorphism could have functional implications, altering ligand-mediated gene expression in β cells or the immune system. An explanation for associations with the polymorphisms located in introns is more difficult to understand.

V. CLINICAL PERSPECTIVES Clear effects of 1,25(OH)2D3 and its newer analogs have been described on the different major players in the pathogenesis of diabetes mellitus, both type 1 and type 2 diabetes. In vitro as well as in vivo, a modest stimulation of insulin synthesis and insulin secretion by 1,25(OH)2D3 is observed [31,34,49–52]. This positive effect is not only observed upon repletion of 1,25(OH)2D3 in the vitamin D–deficient state [31,49–51], but can also be observed in the vitamin D–replete state [34,49–52]. Moreover, a direct β cell protection by 1,25(OH)2D3 and its analogs against metabolic and inflammatory stress has been demonstrated [66]. On the other hand, major effects on the immune system involved in the pathogenesis of type 1 diabetes have been described in vitro as well as in vivo [25,27,28], and prevention of type 1 diabetes and its recurrence after islet transplantation by 1,25(OH)2D3 and its analogs can be achieved (alone or in combination with other immune modulators) [72,89,104,120]. A major problem with using 1,25(OH)2D3 or the currently available analogs in prevention or cure of diabetes is their hypercalcemic and bone remodeling effects when administered in the doses needed for immune or β-cell protective effects. Future applications

1774 of this therapy in human diabetes are conceivable, since through chemical alterations of the 1,25(OH)2D3 molecule even better analogs, with an optimal dissociation between calcemic and immune modulator effects can be synthesized [120]. A place for these analogs in the treatment (prevention or cure) of diabetes can be conceived first of all as β-cell protective and stimulating agents added to the current treatment modalities of type 2 diabetes. Further, these analogs could play a major role in prevention strategies for type 1 diabetes in humans, because of their ideal profile as β-cell protective and especially immune-active drugs. However, before applying these drugs in humans, more information should be gathered not only on their mechanism of action but especially on the safety of these products in long-term use.

CHANTAL MATHIEU, CONNY GYSEMANS, AND ROGER BOUILLON

12. 13.

14. 15. 16. 17.

18.

References 1. Henry HL 1981 25-OH-D3 metabolism in kidney cell cultures: lack of a direct effect of estradiol. Am J Physiol 240: E119–E124. 2. Condamine L, Vztovsnik F, Friedlander G, Menaa C, Garabedian M 1994 Local action of phosphate depletion and insulin-like growth factor 1 on in vitro production of 1,25dihydroxyvitamin D by cultured mammalian kidney cells. J Clin Invest 94:1673–1679. 3. Nyomba BL, Verhaeghe J, Thomasset M, Lissens W, Bouillon R 1989 Bone mineral homeostasis in spontaneously diabetic BB rats. I. Abnormal vitamin D metabolism and impaired active intestinal calcium absorption. Endocrinology 124:565–572. 4. Verhaeghe J, Van Herck E, Visser WJ, Suiker AMH, Thomasset M, Einhorn TA, Faierman E, Bouillon R 1990 Bone and mineral metabolism in BB rats with long-term diabetes. Diabetes 39:477–482. 5. Nyomba BL, Bouillon R, Lissens W, Van Baelen H, De Moor P 1985 1,25-dihydroxyvitamin-D and vitamin D–binding protein are both decreased in streptozotocin-diabetic rats. Endocrinology 116:2483–2488. 6. Hough S, Russell JE, Teitelbaum SL, Avioli LV 1982 Calcium homeostasis in chronic streptozotocin-induced diabetes mellitus in the rat. Am J Physiol 242:E451–E456. 7. Schneider LE, Schedl HP, McCain T, Haussler MR 1977 Experimental diabetes reduces circulating 1,25-dihydroxyvitamin D in the rat. Science 196:1452–1454. 8. Storm TL, Sorensen OH, Lund B, Christiansen JS, Andersen AR, Lumholtz IB, Parving HH 1983 Vitamin D metabolism in insulin-dependent diabetes mellitus. Metab Bone Dis Relat Res 5:107–110. 9. Nyomba BL, Bouillon R, Bidingij M, Kandjing K, De Moor P 1986 Vitamin D metabolites and their binding protein in adult diabetic-patients. Diabetes 35:911–915. 10. Rodland O, Markesta T, Aksnes L, Aarskog D 1985 Plasma concentrations of vitamin D metabolites during puberty of diabetic children. Diabetology 28:663–666. 11. Verrotti A, Basciani F, Carle F, Morgese G, Chiarelli F 1999 Calcium metabolism in adolescents and young adults with type 1

19. 20.

21.

22. 23. 24. 25. 26.

27.

28. 29. 30.

diabetes mellitus without and with persistent microalbuminuria. J Endocrinol Invest 22:198–202. Auwerx J, Dequeker J, Bouillon R, Geusens P, Nijs J 1988 Mineral metabolism and bone mass at peripheral and axial skeleton in diabetes mellitus. Diabetes 37:8–12. Verhaeghe J, Suiker AMH, Van Bree R, Van Herck E, Jans I, Visser WJ, Thomasset M, Allewaert K, Bouillon R 1993 Increased clearance of 1,25(OH)2D3 and tissue-specific responsiveness to 1,25(OH)2D3 in diabetic rats. Am J Physiol 265:E215–E223. Schneider LE, Schedl HP 1972 Diabetes and intestinal calcium absorption in the rat. Am J Physiol 223:1319–1323. Schneider LE, Wilson HD, Schedl HP 1974 Effects of alloxan diabetes on duodenal calcium-binding protein in the rat. Am J Physiol 227:832–838. Stone LA, Weaver VM, Bruns ME, Christakos S, Welsh J 1991 Vitamin D receptors and compensatory tissue growth in spontaneously diabetic BB rats. Ann Nutr Metab 35:196–202. Bouillon R, Verhaeghe J, Bex M, Suiker A, Visser W 1991 Diabetic bone disease. In: H Rifkin, JA Colwell, SI Taylor (eds) Diabetes. Elsevier Science, Amsterdam Publishers: 1271–1274. Verhaeghe J, Suiker AMH, Visser WJ, Van Herck E, Van Bree R, Bouillon R 1992 The effects of systemic insulin, insulin-like growth factor-I and growth hormone on bone growth and turnover in spontaneously diabetic BB rats. J Endocrinol 134:485–492. Verhaeghe J, Bouillon R 1994 Actions of insulin and the IGFs on bone. NIPS 9:20–22. Binz K, Schmid C, Bouillon R, Froesch ER, Jürgensen K, Hunziker EB 1994 Interactions of insulin-like growth factor I with dexamethasone on trabecular bone density and mineral metabolism in rats. Eur J Endocrinol 130:387–393. Verhaeghe J, van Bree R, Van Herck E, Thomas H, Skottner A, Dequeker J, Mosekilde L, Einhorn TA, Bouillon R 1996 Effects of recombinant human growth hormone and insulinlike growth factor-I, with or without 17 beta-estradiol, on bone and mineral homeostasis of aged ovariectomized rats. J Bone Miner Res 11:1723–1735. De Fronzo RA, Bonadonna RC, Ferrannini E 1992 Pathogenesis of NIDDM. Diabetes Care 15:318–368. Taylor SI, Accili D, Imai Y 1994 Insulin resistance or insulin deficiency — Which is the primary cause of NIDDM? Diabetes 43:735–740. Castano L, Eisenbarth GS 1990 Type-I diabetes: A chronic autoimmune disease of human, mouse, and rat. Annu Rev Immunol 8:647–679. Mathieu C, Adorini L 2002 The coming of age of vitamin D3 as an immunomodulator. Trends Mol Med 8:174–179. Christakos S, Norman AW 1979 Studies on the mode of action of calciferol XVII. Evidence for a specific high-affinity binding protein for 1,25-dihydroxyvitamin D3 in chick kidney and pancreas. Biochem Biophys Res Commun 89:56–63. Bouillon R, Garmyn M, Verstuyf AM, Segaert S, Casteels K, Mathieu C 1995 A paracrine role for 1,25(OH)2D3 in the immune system and skin creates new therapeutic possibilities for vitamin D analogs. Eur J Endocrinol 133:7–16. Casteels K, Bouillon R, Waer M, Mathieu C 1995 Immunomodulatory effects of 1,25(OH)2D3. Current Opinion in Nephrology and Hypertension 4:313–318. Norman AW, Frankel BJ, Heldt AM, Grodsky GM 1980 Vitamin D deficiency inhibits pancreatic secretion of insulin. Science 209:823–825. Lee S, Clark SA, Gill RK, Christakos S 1994 1,25-dihydroxyvitamin D3 and pancreatic β-cell function. Vitamin D receptors, gene expression, and insulin secretion. Endocrinology 134:1602–1610.

1775

CHAPTER 99 Vitamin D and Diabetes

31. Cade C, Norman AW 1987 Rapid normalization/stimulation by 1,25(OH)2-vitamin D3 of insulin secretion and glucose tolerance in the vitamin D–deficient rat. Endocrinology 120:1490–1497. 32. de Boland AR, Norman AW 1990 Influx of extracellular calcium mediates 1,25-dihydroxyvitamin D3–dependent transcaltachia (the rapid stimulation of duodenal Ca2+ transport). Endocrinology 127:2475–2480. 33. Norman AW 1994 Editorial: The vitamin D endocrine system: Identification of another piece of the puzzle. Endocrinology 134:A1601–C1601. 34. Sergeev IN, Rhoten WB 1995 1,25-dihydroxyvitamin D3 evokes oscillations of intracellular calcium in a pancreatic β-cell line. Endocrinology 136:2852–2861. 35. Norman AW, Okamura WH, Bishop JE, Henry HL 2002 Update on biological actions of 1alpha,25(OH)2-vitamin D3 (rapid effects) and 24R,25(OH)2-vitamin D3. Mol Cell Endocrinol 197:1–13. 36. Roth J, Bonner-Weir S, Norman AW, Orci L 1982 Immunocytochemistry of vitamin D–dependent calciumbinding protein in chick pancreas: exclusive localization in B-cells. Endocrinology 110:2216–2218. 37. Chertow BS, Sivitz WI, Baranetsky NG, Clark SA, Waite A, DeLuca HF 1983 Cellular mechanisms of insulin release: The effects of vitamin D deficiency and repletion on rat insulin secretion. Endocrinology 113:1511–1518. 38. Nyomba BL, Bouillon R, De Moor P 1984 Influence of vitamin D status on insulin secretion and glucose tolerance in the rabbit. Endocrinology 115:191–197. 39. Kadowaki S, Norman AW 1984 Dietary vitamin D is essential for normal insulin secretion from the perfused rat pancreas. J Clin Invest 73:759–766. 40. Cade C, Norman AW 1986 Vitamin D3 improves impaired glucose tolerance and insulin secretion in the vitamin D–deficient rat in vivo. Endocrinology 119:84–90. 41. Gedik O, Akalin S 1986 Effects of vitamin D deficiency and repletion on insulin and glucagon secretion in man. Diabetologia 29:142–145. 42. Tanaka Y, Seino Y, Ishida M, Yamaoka K, Yabuuchi H, Ishida H, Seino S, Imura H 1984 Effect of vitamin D3 on the pancreatic secretion of insulin and somatostatin. Acta Endocrinol 105: 528–533. 43. Tanaka Y, Seino Y, Ishida M, Yamaoka K, Satomura K, Yabuuchi H, Imura H 1986 Effect of 1,25-dihydroxyvitamin D3 on insulin secretion: direct or mediated? Endocrinology 118:1971–1976. 44. Isaia G, Giorgino R, Adami S 2001 High prevalence of hypovitaminosis D in female type 2 diabetic population. Diabetes Care 24:1496. 45. Beaulieu C, Kestekian R, Havrankova J, Gascon-Barré M 1993 Calcium is essential in normalizing intolerance to glucose that accompanies vitamin D depletion in vivo Diabetes 42: 35–43. 46. Zeitz U, Weber K, Soegiarto DW, Wolf E, Balling R, Erben RG 2003 Impaired insulin secretory capacity in mice lacking a functional vitamin D receptor. FASEB J 17:509–511. 47. Mathieu C, Van Etten E, Gysemans C, Decallonne B, Kato S, Laureys J, Depovere J, Valckx D, Verstuyf A, Bouillon R 2001 In vitro and in vivo analysis of the immune system of vitamin D receptor knockout mice. J Bone Miner Res. 16:2057–2065. 48. Zella JB, DeLuca HF 2003 Vitamin D and autoimmune diabetes. J Cell Biochem 88:216–222. 49. Billaudel BJL, Faure AG, Sutter BCJ 1990 Effect of 1,25 dihydroxyvitamin D3 on isolated islets from vitamin D3–deprived rats. Am J Physiol 258:E643–E648. 50. Billaudel BJL, Delbancut APA, Sutter BCJ, Faure AG 1993 Stimulatory effect of 1,25-dihydroxyvitamin D3 on calcium

51. 52.

53.

54.

55. 56.

57. 58. 59.

60. 61. 62.

63. 64. 65. 66.

67. 68.

handling and insulin secretion by islets from vitamin D3– deficient rats. Steroids 58:335–341. Kadowaki S, Norman AW 1985 Time course study of insulin secretion after 1,25-dihydroxyvitamin D3 administration. Endocrinology 117:1765–1771. d’Emden MC, Dunlop M, Larkins RG, Wark JD 1989 The in vitro effect of 1α,25-dihydroxyvitamin D3 on insulin production by neonatal rat islets. Biochem Biophys Res Commun 164:413–418. Hahn HJ, Kuttler B, Mathieu C, Bouillon R 1997 1,25Dihydroxyvitamin D3 reduces MHC antigen expression on pancreatic beta-cells in vitro. Transplant Proc 29: 2156–2157. Kumar S, Davies M, Zakaria Y, Mawer EB, Gordon C, Olukoga AO, Boulton AJM 1994 Improvement in glucose tolerance and beta-cell function in a patient with vitamin D deficiency during treatment with vitamin D. Postgrad Med J 70:440–443. Gedik O, Akalin S 1986 Effects of vitamin D deficiency and repletion on insulin and glucagon secretion in man. Diabetologia 29:142–145. Boucher BJ, Mannan N, Noonan K, Hales CN, Evans SJ 1995 Glucose intolerance and impairment of insulin secretion in relation to vitamin D deficiency in east London Asians. Diabetologia 38:1239–1245. Allegra V, Luisetto G, Mengozzi G, Martimbianco L, Vasile A 1994 Glucose-induced insulin secretion in uremia: role of 1α, 25(OH)2 -vitamin D3. Nephron 68:41–47. Orwoll E, Riddle M, Prince M 1994 Effects of vitamin D on insulin and glucagon secretion in noninsulin-dependent diabetes mellitus. Am J Clin Nutr 59:1083–1087. Borissova AM, Tankova T, Kirilov G, Dakovska L, Kovacheva R 2003 The effect of vitamin D3 on insulin secretion and peripheral insulin sensitivity in type 2 diabetic patients. Int J Clin Pract. 57:258–261. Taylor AV, Wise PH 1998 Vitamin D replacement in Asians with diabetes may increase insulin resistance. Postgrad Med J 74:365–366. EURODIAB ACE Study Group 2000. Variation and trends in incidence of childhood diabetes in Europe. Lancet 355: 873–876. Karvonen M, Jantti V, Muntoni S, Stabilini M, Stabilini L, Muntoni S, Tuomilehto J 1998 Comparison of the seasonal pattern in the clinical onset of IDDM in Finland and Sardinia. Diabetes Care 21:1101–1109. Stene LC, Ulriksen J, Magnus P, Joner G 2000 Use of cod liver oil during pregnancy associated with lower risk of type I diabetes in the offspring. Diabetologia 43:1093–1098. Hypponen E, Laara E, Reunanen A, Jarvelin MR, Virtanen SM 2001 Intake of vitamin D and risk of type 1 diabetes: a birthcohort study. Lancet 358:1500–1503. Mathieu C, van Etten E, Gysemans C, Decallonne B, Bouillon R 2002 Seasonality of birth in patients with type 1 diabetes. Lancet 359:1248. Sandler S, Buschard K, Bendtzen K 1994 Effects of 1,25dihydrohyvitamin D3 and the analogs MC903 and KH1060 on interleukin-1 beta-induced inhibition of rat pancreatic islet beta-cell function in vitro. Immunol Lett 41:73–77. Mauricio D, Andersen HU, Karlsen AE, Mandrup-Poulsen T, Nerup J 1995 Effect of steroidal hormones on interleukin-1ß action on rat islets. Diabetologia 38S:A80. Riachy R, Vandewalle B, Belaich S, Kerr-Conte J, Gmyr V, Zerimech F, d’Herbomez M, Lefebvre J, Pattou F 2001 Beneficial effect of 1,25 dihydroxyvitamin D3 on cytokine-treated human pancreatic islets. J Endocrinol 169:161–168.

1776 69. Riachy R, Vandewalle B, Kerr Conte J, Moerman E, Sacchetti P, Lukowiak B, Gmyr V, Bouckenooghe T, Dubois M, Pattou F 2002 1,25-dihydroxyvitamin D3 protects RINm5F and human islet cells against cytokine-induced apoptosis: implication of the antiapoptotic protein A20. Endocrinology 143:4809–4819. 70. Makino S, Kunimoto K, Muraoka Y, Mizushima Y, Katagiri K, Tochino Y 1980 Breeding of a nonobese, diabetic strain of mice. Exp Anim 29:1–13. 71. Mathieu C, Laureys J, Sobis H, Vandeputte M, Waer M, Bouillon R 1992 1,25-Dihydroxyvitamin D3 prevents insulitis in NOD mice. Diabetes 41:1491–1495. 72. Mathieu C, Waer M, Laureys J, Rutgeerts O, Bouillon R 1994 Prevention of type I diabetes in NOD mice by 1,25-dihydroxyvitamin D3. Diabetologia 37:552–558. 73. Meehan MA, Kerman RH, Lemire JM 1992 1,25-dihydroxyvitamin D3 enhances the generation of nonspecific suppressor cells while inhibiting the induction of cytotoxic cells in a human MLR. Cell Immunol 140:400–409. 74. Adorini L, Gregori S, Harrison LC 2002 Understanding autoimmune diabetes: insights from mouse models. Trends Mol Med 8:31–38. 75. Gregori S, Giarratana N, Smiroldo S, Uskokovic M, Adorini L 2002 A 1alpha,25-dihydroxyvitamin D(3) analog enhances regulatory T-cells and arrests autoimmune diabetes in NOD mice. Diabetes 51:1367–1374. 76. Harada M, Makino S 1984 Promotion of spontaneous diabetes in nonobese diabetes-prone mice by cyclophosphamide. Diabetologia 27:604–606. 77. Yasunami R, Bach JF 1988 Antisuppressor effect of cyclophosphamide on the development of spontaneous diabetes in NOD mice. Eur J Immunol 18:481–484. 78. Mitsuoka A, Baba M, Morikawa S 1976 Enhancement of delayed hypersensitivity by depletion of suppressor T-cells with cyclophosphamide in mice. Nature 262:77–78. 79. Qin HY, Sadelain M, Hitchon C, Lauzon J, Singh B 1993 Complete Freund’s adjuvant-induced T-cells prevent the development and adoptive transfer of diabetes in nonobese diabetic mice. J Immunol 150:2072–2080. 80. Lakey J, Singh B, Warnock G, Elliott J, Rajotte R 1995 Long-term survival of syngeneic islet grafts in BCG-treated diabetic NOD mice can be reversed by cyclophosphamide. Transplantation 59:1751–1753. 81. Shintani S, Satoh J, Seino H, Goto Y, Toyota T 1990 Mechanism of action of a streptococcal preparation (OK-432) in prevention of autoimmune diabetes in NOD mice. J Immunol 144:136–141. 82. Casteels K, Waer M, Bouillon R, Depovere J, Valckx D, Laureys J, Mathieu C 1998 1,25-dihydroxyvitamin D3 restores sensitivity to cyclophosphamide-induced apoptosis in nonobese diabetic (NOD) mice and protects against diabetes. Clin Exp Immunol 112:181–187. 83. Overbergh L, Decallonne B, Waer M, Rutgeerts O, Valckx D, Casteels KM, Laureys J, Bouillon R, Mathieu C 2000 1alpha,25-dihydroxyvitamin D3 induces an autoantigenspecific T-helper 1/T-helper 2 immune shift in NOD mice immunized with GAD65 (p524–543). Diabetes 49:1301–1307. 84. Adorini L 2003 Tolerogenic dendritic cells induced by vitamin D receptor ligands enhance regulatory T-cells inhibiting autoimmune diabetes. Ann NY Acad Sci 987:258–261. 85. van Halteren AG, van Etten E, de Jong EC, Bouillon R, Roep BO, Mathieu C 2002 Redirection of human autoreactive T-cells upon interaction with dendritic cells modulated by TX527, an analog of 1,25 dihydroxyvitamin D(3). Diabetes 51:2119–2125.

CHANTAL MATHIEU, CONNY GYSEMANS, AND ROGER BOUILLON

86. Casteels KM, Gysemans CA, Waer M, Bouillon R, Laureys JM, Depovere J, Mathieu C 1998 Sex difference in resistance to dexamethasone-induced apoptosis in NOD mice: treatment with 1,25(OH)2D3 restores defect. Diabetes 47:1033–1037. 87. Kolb H, Kröncke KD 1993 IDDM — Lessons from the lowdose streptozocin model in mice. Diab Rev 1:116–126. 88. Inaba M, Nishizawa Y, Song K, Tanishita H, Okuno S, Miki T, Morii H 1992 Partial protection of 1α-hydroxyvitamin D3 against the development of diabetes induced by multiple low-dose streptozotocin injection in CD-1 mice. Metabolism 41:631–635. 89. Guttmann RD, Colle E, Michel F, Seemayer T 1983 Spontaneous diabetes mellitus syndrome in the rat: T lymphopenia and its association with clinical disease and pancreatic lymphocytic infiltration. J Immunol 130:1732–1735. 90. Georgiou HM, Lagarde AC, Bellgrau D 1988 T-cell dysfunction in the diabetes-prone BB rat. J Exp Med 167:132–147. 91. Elder ME, Maclaren NK 1983 Identification of profound peripheral T lymphocyte immunodeficiencies in the spontaneously diabetic BB rat. J Immunol 130:1723–1731. 92. Mathieu C, Casteels K, Bouillon R 1997 Vitamin D and diabetes. In: D Feldman, F Glorieux, J Pike (eds). Vitamin D. Academic press: San Diego, 1183–1196. 93. Bikle DD 1992 Clinical counterpoint: Vitamin D: New actions, new analogs, new therapeutic potential. Endocrine Reviews 13:765–784. 94. Bouillon RK, Allewaert K, Xiang DZ, Tan BK, Van Baelen H 1991 Vitamin D analogs with low affinity for the vitamin D–binding protein: Enhanced in vitro and decreased in vivo activity. J Bone Mineral Res 6:1051–1057. 95. Bouillon R, Okamura WH, Norman AW 1995 Structurefunction relationships in the vitamin D endocrine system. End Rev 16:200–257. 96. Verstuyf A, Verlinden L, van Etten E, Shi L, Wu Y, D’Halleweyn C, Van Haver D, Zhu GD, Chen YJ, Zhou X, Haussler MR, De Clercq P, Vandewalle M, Van Baelen H, Mathieu C, Bouillon R 2000 Biological activity of CD-ring modified 1alpha,25-dihydroxyvitamin D analogs: C-ring and five-membered D-ring analogs. J Bone Miner Res 15:237–252. 97. Verstuyf A, Segaert S, Verlinden L, Bouillon R, Mathieu C 2000 Recent developments in the use of vitamin D analogs. Expert Opin Investig Drugs 9:443–455. 98. Branisteanu DD, Leenaerts P, Van Damme B, Bouillon R 1993 Partial prevention of active Heymann nephritis by 1a,25-dihydroxyvitamin D3. Clin Exp Immunol 94:412–417. 99. Lemire JM, Archer C 1991 1,25-dihydroxyvitamin D3 prevents the in vivo induction of murine experimental autoimmune encephalomyelitis. J Clin Invest 87:1103–1107. 100. Lemire JM, Adams JS 1992 1,25-dihydroxyvitamin D3 inhibits the passive transfer of cellular immunity by a myelin basic protein-specific T-cell clone. J Bone Mineral Res 7:171–177. 101. Fournier C, Gepner P, Sadouk M, Charriere J 1990 In vivo beneficial effects of cyclosporine A and 1,25-dihydroxyvitamin D3 on the induction of experimental autoimmune thyroiditis. Clin Immunol Immunopathol 54:53–63. 102. Jordan SC, Nigata M, Mullen Y 1988 1,25-dihydroxyvitamin D3 prolongs rat cardiac allograft survival. In: K Schaeffer, AW Norman, HG Grigoleit, DV Herrath (eds). Vitamin D: Molecular, Cellular, and Clinical Endocrinology. W. de Gruyter: Berlin, New York, 334–335. 103. Latini S, Binderup L 1991 KH1060: A new vitamin D analog with potent effects on cell-mediated immune responses in vitro.

CHAPTER 99 Vitamin D and Diabetes

104.

105.

106. 107.

108.

109.

110. 111. 112. 113. 114. 115. 116.

117. 118.

119.

120.

In: AW Norman, R Bouillon, M Thomasset (eds). Vitamin D: Gene Regulation, Structure-Function Analysis and Clinical Application. W. de Gruyter: Berlin, New York, 516–519. Mathieu C, Waer M, Casteels K, Laureys J, Bouillon R 1995 Prevention of type I diabetes in NOD mice by nonhypercalcemic doses of a new structural analog of 1,25(OH)2D3, KH1060. Endocrinology 136:866–872. Mori Y, Suko M, Okudaira H, Matsuba I, Tsuruoka A, Sasaki A, Yokoyama H, Tanase T, Shida T, Nishumira M, Terada E, Ikeda Y 1986 Preventive effects of cyclosporine on diabetes in NOD mice. Diabetologia 29:244–247. Weir MR, Fink JC 1999 Risk for post-transplant diabetes mellitus with current immunosuppressive medications. Am J Kidney Dis 34:1–13. Bach JF 1994 Predictive medicine in autoimmune diseases: from the identification of genetic predisposition and environmental influence to precocious immunotherapy. Clin Immunol Immunopathol 72:156–161. Casteels K, Waer M, Bouillon R, Allewaert K, Laureys J, Mathieu C 1996 Prevention of type I diabetes by late intervention with nonhypercalcemic analogs of vitamin D3 in combination with cyclosporin A. Transplant Proc 28:3095. van Etten E, Branisteanu DD, Verstuyf A, Waer M, Bouillon R, Mathieu C 2000 Analogs of 1,25-dihydroxyvitamin D3 as dose-reducing agents for classical immunosuppressants. Transplantation 69:1932–1942. Weringer EJ, Like AA 1986 Diabetes mellitus in the BB/W rat — Insulitis in pancreatic islet grafts after transplantation in diabetic recipients. Am J Pathol 125:107–112. Mandel TE, Koulmanda M, Loudovaris T, Bacelj A 1989 Islet grafts in NOD mice: A comparison of Iso-, Allo-, and pig xenografts. Transplant Proc 21:3813–3814. Lafferty KJ 1989 Islet transplantation as a therapy for type I diabetes mellitus. Diab Nutr Metab 2:323–332. Bartlett ST, Killewich LA 1991 Recurrent diabetes in pancreas transplants in the BB rat: Evidence for a defect extrinsic to the beta cell. Transplant Proc 23:689–690. Pipeleers D, Pipeleers-Marchal M, Markholst H, Hoorens A, Klöppel G 1991 Transplantation of purified islet cells in diabetic BB rats. Diabetologia 34:390–396. Woehrle M, Pullmann J, Bretzel RG, Federlin K 1992 Recurrence of diabetes after islet transplantation in the BB rat. Transplant Proc 24:999. Woehrle M, Pullmann J, Bretzel RG, Federlin K 1992 Prevention of recurrent autoimmune diabetes in the BB rat by islet transplantation under the renal kidney capsule. Transplantation 53:1099–1102. Lacy PE 1993 Status of islet cell transplantation. Diab Rev 1:76–92. Ricordi C, Tzakis AG, Carroll PB, Zeng Y, Rodriguez Rilo HL, Alejandro R, Shapiro R, Fung JJ, Demetris AJ, Mintz DH, Startzl TE 1992 Human islet isolation and allotransplantation in 22 consecutive cases. Transplantation 53:407–414. Sharp DW, Lacy PE, Santiago JV, McCullough CS, Weide LG, Boyle PJ, Falqui L, Marchetti P, Ricordi C, Gingerich RL, Jaffe AS, Cryer PE, Hanto DW, Anderson CB, Wayne Flye M 1991 Results of our first nine intraportal islet allografts in type I, insulin-dependent diabetic patients. Transplantation 51:76–85. Mathieu C, Bouillon R, Laureys J, Waer M 1994 Prevention of autoimmune destruction of transplanted islets in spontaneously diabetic NOD mice by KH1060, a 20-epi analog of vitamin D—Synergy with cyclosporin A. Transpl Proc 26:3128–3129.

1777 121. Katz I, Epstein S 1992 Post-transplantation bone disease. J Bone Mineral Res 7:123–126. 122. Gregori S, Casorati M, Amuchastegui S, Smiroldo S, Davalli AM, Adorini L 2001 Regulatory T-cells induced by 1 alpha,25-dihydroxyvitamin D3 and mycophenolate mofetil treatment mediate transplantation tolerance. J Immunol 167:1945–1953. 123. Gysemans C, Van Etten E, Overbergh L, Verstuyf A, Waer M, Bouillon R, Mathieu C 2002 Treatment of autoimmune diabetes recurrence in non-obese diabetic mice by mouse interferon-beta in combination with an analog of 1alpha,25dihydroxyvitamin-D3. Clin Exp Immunol 128:213–220. 124. McDermott MF, Ramachandran A, Ogunkolade BW, Aganna E, Curtis D, Boucher BJ, Snehalatha C, Hitman GA 1997 Allelic variation in the vitamin D receptor influences susceptibility to IDDM in Indian Asians. Diabetologia 40:971–975. 125. Chang TJ, Lei HH, Yeh JI, Chiu KC, Lee KC, Chen MC, Tai TY, Chuang LM 2000 Vitamin D receptor gene polymorphisms influence susceptibility to type 1 diabetes mellitus in the Taiwanese population. Clin Endocrinol 52:575–580. 126. Pani MA, Knapp M, Donner H, Braun J, Baur MP, Usadel KH, Badenhoop K 2000 Vitamin D receptor allele combinations influence genetic susceptibility to type 1 diabetes in Germans. Diabetes 49:504–507. 127. Yokota I, Satomura S, Kitamura S, Taki Y, Naito E, Ito M, Nisisho K, Kuroda Y 2002 Association between vitamin D receptor genotype and age of onset in juvenile Japanese patients with type 1 diabetes. Diabetes Care 25:1244. 128. Ogunkolade BW, Boucher BJ, Prahl JM, Bustin SA, Burrin JM, Noonan K, North BV, Mannan N, McDermott MF, DeLuca HF, Hitman GA 2002 Vitamin D receptor (VDR) mRNA and VDR protein levels in relation to vitamin D status, insulin secretory capacity, and VDR genotype in Bangladeshi Asians. Diabetes 51:2294–2300. 129. Ban Y, Taniyama M, Yanagawa T, Yamada S, Maruyama T, Kasuga A, Ban Y 2001 Vitamin D receptor initiation codon polymorphism influences genetic susceptibility to type 1 diabetes mellitus in the Japanese population. BMC Med Genet 2:7. 130. Hitman GA, Mannan N, McDermott MF, Aganna E, Ogunkolade BW, Hales CN, Boucher BJ 1998 Vitamin D receptor gene polymorphisms influence insulin secretion in Bangladeshi Asians. Diabetes 47:688–690. 131. Oh JY, Barrett-Connor E 2002 Association between vitamin D receptor polymorphism and type 2 diabetes or metabolic syndrome in community-dwelling older adults: the Rancho Bernardo Study. Metabolism 51:356–359. 132. Ortlepp JR, Metrikat J, Albrecht M, von Korff A, Hanrath P, Hoffmann R 2003 The vitamin D receptor gene variant and physical activity predicts fasting glucose levels in healthy young men. Diabet Med 20:451. 133. Casteels KM, Mathieu C, Waer M, Valckx D, Overbergh L, Laureys JM, Bouillon R 1998 Prevention of type 1 diabetes in nonobese diabetic mice by late intervention with nonhypercalcemic analogs of 1,25-dihydroxyvitamin D3 in combination with a short induction course of cyclosporine A. Endocrinology 139:95–102. 134. Casteels K, Waer M, Laureys J, Valckx D, Depovere J, Bouillon R, Mathieu C 1998 Prevention of autoimmune destruction of syngeneic islet grafts in spontaneously diabetic NOD mice by a combination of a vitamin D3 analog and cyclosporin A. Transplantation 65:1–8. 135. Gysemans C, Waer M, Laureys J, Bouillon R, Mathieu C 2001 A combination of KH1060, a vitamin D3 analog, and

1778

136. 137.

138. 139.

140.

141.

142.

143.

144.

cyclosporin prevents early graft failure and prolongs graft survival of xenogeneic islets in nonobese diabetic mice. Transplant Proc 33:2365. Gupta S, Fass D, Shimizu M, Vayevegula B 1989 Potentiation of immunosuppressive effects of cyclosporine A by 1alpha,25-dihydroxyvitamin D3. Cell Immunol 121:290–297. Gepner P, Amor B, Fournier C 1989 1,25-dihydroxyvitamin D3 potentiates the in vitro inhibitory effects of cyclosporine A on T-cells from rheumatoid arthritis patients. Arthritis Rheum 32:31–36. Branisteanu DD, Mathieu C, Bouillon R 1997 Synergism between sirolimus and 1,25-dihydroxyvitamin D3 in vitro and in vivo. J Neuroimmunol 79:138–147. Jiraposananuruk O, Melamed I, Leung DY 2000 Additive immunosuppressive effects of 1,25-dihydroxyvitamin D3 and corticosteroids on TH1, but not TH2, responses. J Allergy Clin Immunol 106:981–985. Xing N, Maldonado ML, Bachman LA, McKean DJ, Kumar R, Griffin MD 2002 Distinctive dendritic cell modulation by vitamin D3 and glucocorticoid pathways. Biochem Biophys Res Commun 297:645–652. Barrat FJ, Cua DJ, Boonstra A, Richards DF, Crain C, Savelkoul HF, de Waal-Malefyt R, Coffman RL, Hawrylomicz CM, O’Garra A 2002 In vitro generation of interleukin 10-producing regulatory CD4+ T cells is induced by immunosuppressive drugs and inhibited by T helper type 1 (Th1)- and Th2-inducing cytokines. J Exp Med 195:603–616. Branisteanu DD, Waer M, Sobis H, Marcelis S, Vandeputte M, Bouillon R 1995 Prevention of murine experimental allergic encephalomyelitis: cooperative effects of cyclosporine and 1alpha,25-(OH)2D3. J Neuroimmunol 61:151–160. Lillevang ST, Rosenkvist J, Andersen CB, Larsen S, Kemp E, Kristensen T 1992 Single and combined effects of the vitamin D analog KH1060 and cyclosporine A on mercuricchloride-induced autoimmune disease in the BN rat. Clin Exp Immunol 88:301–306. Vendeville B, Baran D, Gascon-Barre M 1995 Effects of vitamin D3 and cyclosporine A on HgCl2-induced autoimmunity

CHANTAL MATHIEU, CONNY GYSEMANS, AND ROGER BOUILLON

145. 146.

147.

148. 149.

150.

151.

152.

153.

in the brown Norway rats. Nephrol Dial Transplant 10: 2020–2026. Boissier MC, Chiocchia G, Fournier C 1992 Combination of cyclosporine A and calcitriol in the treatment of adjuvant arthritis. J Rheumatol 19:754–757. Mathieu C, Casteels K, Waer M, Laureys J, Valckx D, Bouillon R 1998 Prevention of diabetes recurrence after syngeneic islet transplantation in NOD mice by analogs of 1,25(OH)2D3 in combination with cyclosporine A: mechanism of action involves an immune shift from Th1 to Th2. Transplant Proc 30:541. Redaelli CA, Wagner M, Gunter-Duwe D, Tian YH, Stahel PF, Mazzucchelli L, Schmid RA, Schilling MK 2002 1alpha,25-dihydroxyvitmain D3 shows strong and additive immunoregulatory effects with cyclosporine A in rat renal allotransplants. Kidney Int 61:288–296. Kallio E, Hayry P, Pakkala S 1996 MC1288, a vitamin D analog, reduces short- and long-term renal allograft rejection in the rat. Transplant Proc 28:3113. Redaelli CA, Wagner M, Tien YH, Mazzucchelli L, Stahel PF, Schilling MK, Dufour JF 2001 1alpha,25-dihydroxycholecalciferol reduces rejection and improves survival in rat liver allografts. Hepatology 34:926–934. Pakkala I, Taskinen E, Pakkala S, Raisanen-Sokolowski A 2001 MC1288, a vitamin D analog, prevents acute graft-versus-host disease in rat bone marrow transplantation. Bone Marrow Transplant 27:863–867. Raisanen-Sokolowski AK, Pakkala IS, Samila SP, Binderup L, Hayry PJ, Pakkala ST 1997 A vitamin D analog, MC1288, inhibits adventitial inflammation and suppresses intimal lesions in rat aortic allografts. Transplantation 63:936–941. Johnsson C, Binderup L, Tufveson G 1995 The effects of combined treatment with the novel vitamin D analog MC1288 and cyclosporine A on cardiac allograft survival. Transplant Immunol 3:245–250. Veyron P, Pamphile R, Binderup L Touraine JL 1995 New 20-epi-vitamin D3 analogs: immunosuppressive effects on skin allograft survival. Transplant Proc 27:450.

CHAPTER 100

Vitamin D, A Neuroactive Hormone: From Brain Development to Pathological Disorders PHILIPPE BRACHET, ISABELLE NEVEU, AND PHILIPPE NAVEILHAN INSERM U 643, Centre Hospitalier Universitaire, 30 bd Jean Monnet, 44093 Nantes, France

EMMANUEL GARCION INSERM U 646, 10 rue André Boquel, 49100 Angers, France

DIDIER WION INSERM U318, Centre Hospitalier Michallon, 38043 Grenoble cedex 09, France

I. Introduction II. Vitamin D Receptor and Targets in the Central Nervous System III. Vitamin D Actions in the Central Nervous System

IV. 1,25(OH)2D3 and Brain Tumors V. 1,25(OH)2D3, a Mediator of Neuro-Immune Interactions VI. Conclusion References

I. INTRODUCTION

nuclei, which project axons into the hippocampus and cortex. Glutamate, the major excitatory neurotransmitter in the brain, is known to act as an excitotoxic agent when present in excess, provoking through the N-methyl-Daspartate (NMDA) receptor a massive entry of calcium, and it generates oxidative or nitrative stresses that ultimately cause neuronal death. Glutamate excitotoxicity may be one mechanism that participates in the neurodegeneration of HD and PD (reviewed in [1,2]). However, reactive oxygen species (ROS) are probably broadly involved in the terminal events leading to cell death. In line with this, post-mortem findings support the involvement of ROS in the progression of PD, HD, and AD (reviewed in [3–5]). Moreover, animal models of PD are based on a selective intoxication of dopaminergic neurons with 6-hydroxydopamine, a compound whose catabolism by monoamine-oxidase generates ROS. The same is true for the 1-methyl-4-phenylpyridinium ion (MPP+), which causes a Parkinson-like syndrome in humans. In AD, the amyloid deposits, which form the senile plaques or more diffuse perivascular deposits, are essentially composed of a 40–42 amino-acid peptide referred to as amyloid β peptide (Aβ). Aggregated Aβ peptide has a neurotoxic action, and activates microglial cells, the brain resident macrophages. These cells sustain a protracted pro-inflammatory context within the brain of AD patients, with a liberation of ROS and

The etiology and physiopathology of most neurodegenerative diseases of the central nervous system (CNS) remain poorly understood. In Parkinson’s disease (PD), motor impairments result from the selective death of the dopaminergic neurons of the substantia nigra, resulting in a depletion of dopamine in their target area, the striatum. Huntington’s disease (HD) is a genetic autosomal dominant disorder with a fatal outcome. It is initially characterized by a loss of γ-amino butyric acid-containing (GABAergic) neurons of the striatum. These neurons constitute a complex population, characterized by the presence of various neuropeptides (neuropeptide Y, substance P) and calcium-binding proteins, such as calbindin-D28k or parvalbumin. GABAergic neurons project their axons towards the substantia nigra, the cortex as well as other structures, and their destruction results in both cognitive and motor impairments. Alzheimer’s disease (AD) is a complex pathology characterized by a spatiotemporal extension of two major lesions, the intraneuronal accumulation of paired helical filaments of pathogenic tau proteins and the deposits of amyloid. Both lesions mainly affect the cortex and the hippocampus. Further neurodegenerative phenomena extend in a retrograde manner to other structures, such as to the cholinergic neurons of the septum, diagonal band, and the Meynert’s VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Copyright © 2005, Elsevier, Inc. All rights reserved.

1780 cytotoxic lymphokines such as tumor necrosis factor-α (TNF-α) or interleukin-1 (IL-1) [3–6]. In multiple sclerosis (MS), demyelination occurs in active plaques, in response to a massive infiltration by macrophages, but also by T lymphocytes, including autoreactive T-cells, which react with antigens present in the myelin sheath, such as myelin basic protein (MBP). In this case, too, oligodendrocyte death is ascribed to the production of ROS in a marked pro-inflammatory context (reviewed in [7]). The possible involvement of 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) in MS was proposed on the basis of its action on the immune system, since it was found to have prophylactic effects in experimental autoimmune encephalomyelitis (EAE), a model of MS. Other putative actions of 1,25(OH)2D3 in the CNS have been ascribed to its classical role in the control of calcium metabolism, for instance via the expression of calcium-binding proteins in neurons. However, there is a growing body of evidence which broadens the roles of 1,25(OH)2D3 in the brain. These include regulatory effects on the production of neurotrophic factors or the tuning of the brain detoxifying machinery. Moreover, neurons, but also astrocytes, oligodendrocytes, and microglial cells are cell targets of this hormone. It seems, therefore, that 1,25(OH)2D3 is part of a complex cytokine/hormonal network controlling brain homeostasis and repair.

II. VITAMIN D RECEPTOR AND TARGETS IN THE CENTRAL NERVOUS SYSTEM 1,25(OH)2D3 and other metabolites of vitamin D have been detected in the cerebrospinal fluid, and this hormone has also been shown to cross the blood-brain barrier [8–10]. Furthermore, microglial cells can convert 25-hydroxy vitamin D3 into 1,25(OH)2D3 upon activation with interferon-γ (IFNγ) [11]. Detection of 25-hydroxyvitamin D-1α-hydroxylase transcripts in the brain, including in neurons [12,13], supports the concept that the active metabolite of vitamin D can be synthesized in the CNS. In order to identify the hormone’s target cells, several studies have been performed using radiolabeled 1,25(OH)2D3 and have identified high affinity binding sites, most probably the vitamin D receptor (VDR) in the nervous system of various types of vertebrates. In the CNS, these specific binding sites were found in numerous structures [14]. These include motor trigeminal and hypoglossal nuclei, the stria terminalis or reticular nucleus in the thalamus, the amygdala and CA4 part of the hippocampus, and cholinergic nuclei in the telencephalon. These latter mapping data suggest a role for 1,25(OH)2D3 in

PHILIPPE BRACHET

ET AL .

cognitive processes [15], while the presence of binding sites in hypothalamic nuclei were proposed to underlie an involvement of brain 1,25(OH)2D3 in the neuroendocrine network, for instance through an influence on corticotropin releasing factor-containing neurons and gonadotropin-containing neurons [16]. Similarly, the implication of 1,25(OH)2D3 in biological rhythms has also been proposed, because the density of binding sites in circumventricular structures is notably higher in species that are subjected to seasonal behaviors [17,18]. In the peripheral nervous system (PNS), binding sites have been detected in the motor neurons of the spinal cord and in the trigeminal ganglia, that is, in cell bodies of sensory neurons [19]. Studies performed with antibodies directed against the VDR have confirmed the widespread presence of VDR-immunoreactivity in many structures of the adult brain [20–23], and data collected with both methods are largely concordant. An interesting finding is that VDRimmunoreactivity has been detected both in the hippocampus and in numerous cortical areas, including the enthorinal cortex [20], which constitutes an essential structure in the cortico-hippocampal connectivity, and is a primary target of neurofibrillary degeneration in AD [24]. In the rat hippocampus, VDR-immunoreactive neurons have been detected along the whole rostrocaudal axis, but with regional variations [22,23]. VDRexpressing neurons were present in the pyramidal layer of the cornu Ammonis (CA1-CA4), the molecular layer of the dentate gyrus and the hilus, a structure in which neurogenesis occurs, even during adult life. Functionality of the VDR was supported by electrophoretic mobility shift assays performed with the osteopontin or osteocalcin vitamin D–responsive elements (VDRE) [22]. It is worth noting that although VDR immunoreactivity was essentially allocated to cells with a neuronal morphology, the presence of a few VDR-immunoreactive astrocytes, as defined by their network of glial fibrillary acidic protein (GFAP), was also detected [22]. Besides these studies reporting the presence of the VDR in the CNS and PNS, immunoreactive material was observed in numerous cell types in the rat or human retina [25,26]. The immunohistochemical detection of the VDR has also been extended to fetal development. Use of genetargeted mutant mice showed that the VDR gene is turned on as soon as gestational day 11.5 in the developing mouse brain [27]. In the rat embryo, a pronounced expression of VDR-immunoreactivity takes place during the differentiation of many structures and declines to some extent thereafter [28]. This conclusion is supported by the fact that anti-VDR antibody labeled cells were consistently detected in the neuroepithelium and adjacent differentiating fields of developing areas. This temporary expression of VDR was first detected in

1781

CHAPTER 100 Vitamin D, A Neuroactive Hormone

the spinal cord and nascent dorsal root ganglia. However, it was found at different time-points in the midbrain and diencephalon, and in the telencephalon, including the cortex and hippocampus. As in the case of the hilar region of the dentate gyrus of the adult rat, expression of the VDR was detected in the ventricular epithelium and the subventricular zones, in which neural stem cells continuously generate migrating neuroblasts which differentiate into olfactory neurons. Finally, expression of the VDR gene has been revealed by in situ hybridization, for instance in the pathological human hippocampus [29] or in several areas of the fetal rat brain [28]. This confirms by another method that the VDR is broadly expressed in the nervous system and in the brain, in particular. This expression, however, appears to be tightly regulated by temporal and regional cues. The role of the hormone in the CNS, PNS, or eye remains to be clarified. The fact that VDR immunoreactivity or 1,25(OH)2D3 binding occurs in given neuronal populations does not imply that the hormone directly modulates the neurochemical characteristics or activity of these cells. Rather, 1,25(OH)2D3 might exert a broad action in the brain by influencing basic metabolic functions.

III. VITAMIN D ACTIONS IN THE CENTRAL NERVOUS SYSTEM A. Calbindins and Other Calcium-Binding Proteins It has been proposed that 1,25(OH)2D3 could mediate certain effects in the brain by controlling the calcium buffering capacity of neurons through calcium-binding proteins. This was suggested on the basis of the classical role of the hormone in the intestine, kidney, or bone, together with the presence of a canonical VDRE in the promoter of the calbindin-D28k promoter [30]. This possibility was substantiated by the fact that calbindin-D28k transcripts or protein are substantially reduced in certain brain areas in patients deceased from neurodegenerative diseases [29–33]. This is the case of the substantia nigra in PD, the nucleus basalis, and the hippocampus in AD, and the striatum and the hippocampus in HD. One study showed a correlation between reduced calbindin-D28k transcripts and VDR transcripts in the AD hippocampus [29]. However, null mutant mice for the calbindin-D28k gene have an apparently normal CNS, even though they develop a severe ataxia [34]. Furthermore, susceptibility of midbrain dopaminergic neurons to MPP+ is unaffected in these animals [35]. Another fact that indicates that the relationship between calbindin-D28k and 1,25(OH)2D3 is complex in the brain, is that apart from certain

structures, there is no systematic overlap of the presence of calbindin-D28k and that of nuclear and/or cytoplasmic VDR-immunoreactivity [20,21]. A chronic treatment of rats with 1,25(OH)2D3 had no effect on the brain content of calbindin-D28k, protein S100, calmodulin, and parvalbumin, except for the striatum, in which levels of parvalbumin were somewhat increased [36]. In contrast, another study based on shorter treatments detected an accumulation of calbindin-D28k, parvalbumin, and calretinin immunoreactivities in spinal motor neurons [37]. However, investigations with mice lacking the DNA-binding domain of the VDR did not show any alteration in the abundance of calbindin-D28k transcripts in the brain, in comparison to normal animals. In contrast, transcripts of calbindin-D9k were scarcer in mutant animals and treatment with low doses of 1,25(OH)2D3 increased their levels [27]. Paradoxically, this calcium-binding protein is much less represented in the brain than calbindinD28k, and unlike the latter protein, its expression seems to be subject to nonconventional regulatory mechanisms by the hormone [38]. Hence, these data remain difficult to interpret, especially since these studies used different methodological approaches which preclude their comparison. New clues concerning the role of 1,25(OH)2D3 in the nervous system have been obtained by in vitro approaches, from studies performed on primary cultures of brain-derived cells.

B. In Vitro Investigations Provide New Clues about the In Vivo Actions of 1,25(OH)2D3 Studies performed in vitro have shown that rat primary astrocytes and oligodendrocytes contain VDR transcripts, whose abundance rapidly increased in the presence of 1,25(OH)2D3 [39]. Astrocytes also express transcripts of the 25 (OH) vitamin D3 24-hydroxylase, which do also accumulate in the presence of the hormone [40]. It was concluded that astrocytes have the potential to respond to 1,25(OH)2D3, and that the hormone, which can be synthesized locally [11], also controls to some extent its own catabolism, as in other cells. Likewise, VDR transcripts have been detected in cultures of oligodendrocytes and Schwann cells [41,42], the myelinating cells of the central and peripheral axons, respectively. VDR immunoreactivity has also been identified in cultured hippocampal neurons and in the dopaminergic neurons of the substantia nigra [43,44]. 1. EFFECTS ON NEUROTROPHIC FACTORS

A striking response of primary astrocytes, oligodendrocytes, Schwann cells, and CNS neurons to the presence of 1,25(OH)2D3 is a marked increase in the production of the neurotrophic factor nerve growth

1782 factor (NGF). The hormone enhanced both the levels of NGF transcripts and secreted protein [39,41,42,45]. Existence of a promoter in the NGF gene that is activated by the hormone strongly supports the fact that the up-regulation of NGF synthesis in astrocytes, oligodendrocytes, or neurons is primarily of transcriptional origin [46]. Further studies showed that in astrocyte cultures, 1,25(OH)2D3 also enhanced the abundance of neurotrophin 3 (NT-3) transcripts, a neurotrophic protein structurally related to NGF, but appeared to decrease those of brain-derived neurotrophic factor (BDNF), another member of the neurotrophin family [47]. The action of 1,25(OH)2D3 on NGF synthesis has been reproduced in a large variety of transformed cell lines, such as fibroblasts [48,49], osteosarcoma [50], and tumors of neural origin, including C6 gliomas and neuroblastomas [39,51]. The action of the hormone on the production of NGF has also been reproduced in vivo. Chronic systemic treatment increased both NGF transcripts and protein in two structures, which naturally produce this factor, the cortex and the hippocampus [52]. Conversely, newborn rats that developed in vitamin D–deficient dams had lower levels of NGF in their brain [53]. These observations are relevant to the clinic, since neuroprotection of the cholinergic, NGF-responsive neurons of the basal forebrain could prevent or delay the progression of mild cognitive impairments in normal aging or AD onset. These observations also provide an explanation for the fact that intraperitoneal or intraventricular treatments with 1,25(OH)2D3 enhanced the specific activity of choline acetyltransferase, as forebrain cholinergic neurons are responsive to NGF [54]. Besides its effects on neurotrophins, 1,25(OH)2D3 has been reported to enhance the expression of another potent trophic factor active on several CNS neurons, the glial cell line-derived neurotrophic factor (GDNF). This was initially reported to take place in rat C6 glioma cells [55], but was not supported by data collected in primary cultures of glial cells. These latter studies, however, detected a potent action of thyroid hormone [56]. Nevertheless, lower levels of GDNF were observed in the brain of vitamin D–depleted newborn rats, as for NGF [53]. Conversely, systemic administration of 1,25(OH)2D3 for eight days increased the GDNF content in the adult rat cortex, while reducing cortical infarction following ischemia [57]. A similar treatment also attenuated the toxicity of 6-hydroxydopamine on a classical target of GDNF, the mesencephalic dopaminergic neurons of the substantia nigra [58]. 2. REGULATION OF THE LOW-AFFINITY NEUROTROPHIN RECEPTOR

These in vivo data indicate that 1,25(OH)2D3 has the potential to participate in the modulation of neuronal

PHILIPPE BRACHET

ET AL .

plasticity and survival through an action on the supply of trophic factors, such as NGF and GDNF. However, other mechanisms could operate in parallel. As a mirror image of the effect of 1,25(OH)2D3 on NGF synthesis, the hormone is also involved in the regulation of its low-affinity receptor, referred to as p75NTR. This receptor is not restricted to a single ligand, as it binds various neurotrophins. However, it acts primarily as a co-receptor of the different high affinity neurotrophin receptors, the transmembrane tyrosine-kinases TrkA, B, and C (specific binding sites for NGF, BDNF, or NT3, and NT4/5, respectively). The major trophic actions of neurotrophins, such as cell survival, are mediated by Trk receptors. However, these effects are enhanced when the p75NTR is present. The affinity of NGF binding is increased when p75NTR and TrkA are co-expressed [59]. Furthermore, both receptors reciprocally modulate their transducing machineries in a complex cross-talk [60]. p75NTR also participates in the retrograde axonal transport of neurotrophins [61,62]. In contrast, p75NTR by itself triggers an opposing signal since it promotes cell apoptosis, instead of cell survival via the Trk receptor [63–65]. Up-regulation by 1,25(OH)2D3 of p75NTR transcripts was first reported in C6 gliomas and subsequently in cultured oligodendrocytes [42, 66]. A VDRE was also identified in the promoter of the p75NTR [66]. Treatment of adult rats with high doses of 1,25(OH)2D3 decreased the levels of p75NTR transcripts in the adult spinal cord, thus suggesting a hormone-mediated reduction in the pro-apoptotic potential of this receptor [66]. Conversely, p75NTR-immunoreactivity and transcripts were greatly decreased in several brain areas of rat pups born to vitamin D–depleted mothers. This loss of expression could affect the cell death processes that normally occur during brain development [53]. The frequency of proliferating cells in the dentate gyrus, hypothalamus, and basal ganglia was higher in vitamin D–depleted rats. This effect suggests that 1,25(OH)2D3, which has a marked differentiating and antiproliferative action on a large variety of cells, is actively involved in the control of cell number in the brain. Hence, reduced apoptosis associated with an enhanced growth potential would account for the severe alteration in brain shape in vitamin D–depleted pups [53]. It was argued that these observations are relevant with respect to psychiatric disorders, such as schizophrenia, which could have an increased risk factor upon vitamin D deficiency during pregnancy [67,68]. 3. CALCIUM CHANNELS

The studies that have focused on the intracellular control of calcium concentration have detected evidence for a modulatory action of 1,25(OH)2D3 on L-type voltage-sensitive calcium channels in cultures of

1783

CHAPTER 100 Vitamin D, A Neuroactive Hormone

hippocampal neurons. Prolonged treatment of neurons with low concentrations of the hormone depressed the channel density and mRNA content of the pore forming subunits α1c and α1d. This accounts for the protective effect of the hormone against toxic stimuli such as medium exchange or treatments with NMDA or glutamate [43]. Voltage-sensitive L-type channels are of considerable physiological importance in the function of the hippocampus, including long-term potentiation or gene expression. The regulatory action of 1,25(OH)2D3 provides a molecular explanation for the fact that protracted treatment of rats with the hormone prevents age-related loss of hippocampal neurons [69], and this neuroprotection raises intriguing questions regarding the prevention of the exacerbated degenerative process which operates in the hippocampus of AD patients. 4. 1,25(OH)2D3 AND BRAIN DETOXIFICATION

Another protective action of 1,25(OH)2D3 relies on its regulation of the detoxifying machinery. The hormone was reported to enhance the specific activity of γ-glutamyl transpeptidase in the brain [70]. This enzyme cleaves extracellular glutathione, thus favoring the uptake of catabolic products for the intracellular synthesis of this same compound. In cultures of primary astrocytes, 1,25(OH)2D3 increases the cellular content in glutathione and enhances γ-glutamyl transferase specific activity under severe proinflammatory conditions, without any effect on other enzymes, such as superoxide dismutase or glutathione peroxidase [71]. Type II nitric oxide synthase (NOS II), an inducible enzyme which generates high quantities of NO and which is tightly associated with the physiopathology of MS, is another target of 1,25(OH)2D3. The hormone down-regulates NOS II expression in different models of brain inflammation, such as EAE and lipopolysaccharide (LPS) injection [72,73]. In this latter system, NOS II is first expressed by macrophage/microglia and then by astrocytes. The inhibitory effect of 1,25(OH)2D3 reduces both the abundance of NOS II transcripts and protein. It has been suggested that the regulatory effect is transcriptional and results from the presence of a VDRE in the promoter of the NOS II gene [74]. Since both activated macrophages and microglial cells can synthesize 1,25(OH)2D3 from its precursor, it seems likely that the hormone acts as an endogenous regulator of NOS II in the brain. However, 1,25(OH)2D3 could exert other protective actions against insults generated by ROS. Treatment of mesencephalic neurons with the hormone has been shown to enhance their resistance to 6-hydroxydopamine, MPP+, glutamate, a calcium ionophore, and H2O2 [44,58]. This effect can be achieved following a hormonal pretreatment of at least 24 h and requires protein synthesis. It is not thought to

result from a variation in glutathione levels, but could depend on other thiol-based scavengers, such as thioredoxin or metallothionein [44].

IV. 1,25(OH)2D3 AND BRAIN TUMORS 1,25(OH)2D3 and its analogs are known to inhibit growth and to induce the differentiation of several cancer cells [75] (see section IX). In CNS tumors, VDR expression has been documented in human meningiomas and gliomas [76–78], which are both nonneuronal tumors. In gliomas, VDR expression was found to correlate with the grade of the tumor with the highest level of expression found in untractable glioblastomas [79]. Several studies have reported the responsiveness of gliomas to 1,25(OH)2D3 or to some of its analogs in vitro. Thus, rat or human glioblastoma cell lines respond to the addition of 1,25(OH)2D3, alone or in combination with retinoic acid, by the induction of cell death/differentiation programs [78,80–82]. Hence, it has been suggested that the capacity of activated microglial cells to produce 1,25(OH)2D3 [11] could be part of the mechanisms limiting the extension of some gliomas in vivo. A striking observation, which supports a close relationship between 1,25(OH)2D3 and the growth of malignant gliomas, is that amplification of splice variants of the 25-hydroxyvitamin D-1α-hydroxylase gene is a frequent aberration in these tumors [83,84]. However, the responsiveness to the hormone is complex. The finding that the neurotransmitter noradrenaline inhibits 1,25(OH)2D3-induced glioma cell death suggests that if such a cytotoxic mechanism exists in vivo, it might be inoperative in the close vicinity of noradrenergic neurons [85]. Furthermore, variants of C6 gliomas, which fail to express the VDR, escape the cell death program triggered by the hormone [86]. Studies of the in vivo effects of 1,25(OH)2D3 on gliomas are scarce. However, it is noteworthy that longlasting complete regression of glioblastomas has been obtained in humans under long-term 1-alpha-hydroxycholecalciferol therapy [87]. Regarding the effects of 1,25(OH)2D3 on tumors of neuronal origin (i.e., neuroblastomas), several studies have reported differentiation or antiproliferative effects of the hormone on neuroblastoma cell lines [88,89]. A synergistic effect with retinoids has also been observed, suggesting the potential use of 1,25(OH)2D3 and retinoid analogs in combination in the clinic [90]. Taken together these results suggest that meningiomas, gliomas, and neuroblastomas are potential targets of 1,25(OH)2D3. It can be expected that in the near future, gene expression profiles of brain tumors will tell us whether VDR expression, alone or with other members of the steroid-thyroid

1784 hormone receptor superfamily, is a valuable diagnostic or prognostic parameter for the use of 1,25(OH)2D3 as a curative compound in these malignant diseases. However, recent data indicating that the hormone plays an active role in the control of cell proliferation during normal brain development [53], raise the possibility that it also contributes to the natural brain defense against carcinogenesis.

V. 1,25(OH)2D3, A MEDIATOR OF NEURO-IMMUNE INTERACTIONS MS is a slowly progressive, disabling disease of the CNS characterized by disseminating patches of demyelination in the brain and spinal cord. While the etiology of MS is unknown, it is regarded as a complex multi-causal disease comprising genetic factors, dysfunction of the immune system (auto-immunity), and environmental factors. The concept that vitamin D represents an important factor in MS pathophysiology is quite old [91], but has been re-emphasized in more recent days [92,93]. Both the natural immunomodulatory functions of 1,25(OH)2D3 and its beneficial role in the T-cell mediated MS paradigm, EAE, support the protective involvement of 1,25(OH)2D3 in MS (see Chapters 36 and 98). The hormone was initially described as a preventive factor for EAE when administered at the time of immunization, or to inhibit its passive cellular transfer [94,95]. Further studies showed later that 1,25(OH)2D3 reversibly blocks the progression of EAE when administered after the onset of clinical signs in both rats [96] and mice [97]. Alternatively, vitamin D deprivation aggravated the clinical symptoms of EAE [74,97]. Furthermore, the VDR appeared to be essential for the mediation of the hormonal effects on EAE symptoms [98]. It is now well documented that EAE is mediated by CD4+ Th1 cells that secrete proinflammatory cytokines, and the recovery as well as the remission phase is related to CD4+ T-cells that secrete transforming growth factor-β1 (TGF-β1) [99]. One mechanism explaining the effect of 1,25(OH)2D3 during EAE includes its regulatory actions in peripheral lymphoid organs. 1,25(OH)2D3 inhibits IL-12 production, which normally promotes Th1 polarization [100,101], and it enhances the lymph node content of the anti-encephalitogenic cytokines TGF-β1 and IL-4 [102]. It also favors the differentiation of monocytes into macrophages [103,104], modulates the expression of co-stimulatory signaling receptors [105], and prevents the maturation of dendritic cells, thus leading to impaired T-cell activation [106–109]. 1,25(OH)2D3 also exerts direct effects on T-cells by inhibiting IL-2

PHILIPPE BRACHET

ET AL .

and IFN-γ production [110–113]. Another effect of 1,25(OH)2D3 curative treatments on EAE, is the decrease in expression of the CD4 and MHC class II antigens by brain-infiltrating leukocytes [96]. In parallel to systemic actions, other effects of 1,25(OH)2D3 could directly take place in the CNS. The increase in the anti-encephalitogenic cytokine TGF-β1 in secondary lymphoid organs [102] is probably not associated with a corresponding response in the CNS, since no changes in TGF-β1 secretion were seen in murine microglial cells in vitro after exposure to 1,25(OH)2D3 [114], while TGF-β1 transcripts were unaffected in the CNS of EAE rat treated with the hormone [74]. In contrast, down-regulation of NOS II mRNA and protein by 1,25(OH)2D3 [72–74] is likely to be a direct response, since it sequentially affects microglial cells and astrocytes, which are both VDR-responsive cells [39,114], and also takes place following a direct intrahippocampal injection of LPS [72]. The early down-regulation of MHC class II expression on microglial cells observed during EAE may also result from specific CNS actions of the hormone [74,96]. This effect could either be direct, via the VDR, or indirect, involving, for instance, an up-regulation of NGF synthesis by glial cells or neurons [39,42,45,52], which would in turn inhibit MHC class II molecule induction via a p75NTR receptordependent mechanism [115]. Moreover, 1,25(OH)2D3 down-regulates macrophage-colony stimulating factor (M-CSF) and TNF-α production by cultured astrocytes submitted to an inflammatory stress [116]. This observation also supports a pivotal role for the hormone within the CNS itself during EAE, notably on the degree of activation of microglial cells and their ability to serve as antigen-presenting cells. In parallel with microglial cells and astrocytes, the involvement of oligodendrocytes and neurons as targets of 1,25(OH)2D3 during EAE remains unknown.

VI. CONCLUSION The pleiotropic actions of 1,25(OH)2D3 are summarized in Fig. 1. The hormone clearly behaves as a neuroactive compound largely implicated in the control of brain homeostasis. Because of the potential toxic effects of high doses of 1,25(OH)2D3, several vitamin D analogs (see Section VIII) have been tested in EAE and have been able to reduce the gravity of EAE in a similar fashion to 1,25(OH)2D3, involving notably an inhibition of Th1 development [74,117,118]. Further clinical trials will determine whether these and other analogs exert beneficial effects in resolving MS. Likewise, the neuroprotective effects of the hormone, achieved by its action on the levels of NGF, GDNF,

1785

CHAPTER 100 Vitamin D, A Neuroactive Hormone

Involvement of 1,25(OH)2D3 in Brain Homeostasis Development 1,25(OH)2D3

25(OH)D3 Neurons Neural stem cells

NGF, GDNF p75 NGFR Cell proliferation Control of brain shape

Transit across BBB

Neuroprotection 1,25-hydroxylase Neurons

Neurons Glial cells

NGF, GDNF, BDNF Thiol-based scavengers L-type calcium channels

Activated microglia Other brain cells ?

Oncogenesis Gliomas

1,25(OH)2D3

p75NGFR Cell proliferation Cell death Inflammation

Astrocytes Microglia

NOS II TNFalpha M-CSF MHC II

FIGURE 1 Whilst 1,25(OH)2D3 and its precursor 25(OH)D3 can transit across the blood-brain barrier, an intracerebral synthesis of the active hormone appears possible, since neurons or activated microgial cells express 25-hydroxyvitamin D-1α-hydroxylase (1,25-hydroxylase). The hormone receptor, the VDR, is found both in neurons and glial cells, and the figure summarizes the broad spectrum of responses triggered by the hormone in these cells. Taken together, these responses suggest that 1,25(OH)2D3 contributes to maintaining brain homeostatis, and is also one of the elements conferring special immune status to the CNS.

and the low-affinity neurotrophin receptor support the potential of vitamin D analogs in the treatment of neurodegenerative disorders such as PD and AD. Such analogs could contribute to improving the survival of basal forebrain cholinergic neurons [119] and mesencephalic dopaminergic neurons [120], respectively. Conversely, the contrasting action of 1,25(OH)2D3 on NOS II synthesis and on the detoxifying machinery of brain cells could be beneficial in neurodegenerative disorders or following ischemic injury [1–5,121–124]. An important nutritional issue, however, is to avoid vitamin D deficiency [125–128] in homebound patients or aging persons, as discussed elsewhere in this volume (Chapter 66).

References 1. Plaitakis A, Shashidharan P 2000 Glutamate transport and metabolism in dopaminergic neurons of substantia nigra: implications for the pathogenesis of Parkinson’s disease. J Neurol 247 (Suppl 2):II25–II35.

2. Davies S, Ramsden DB 2001 Huntington’s disease. Mol Pathol 54:409–413. 3. Giasson BI, Ischiropoulos H, Lee VM, Trojanowski JQ 2002 The relationship between oxidative/nitrative stress and pathological inclusions in Alzheimer’s and Parkinson’s diseases. Free Radic Biol Med 32:1264–1275. 4. Halliwell B 2001 Role of free radicals in the neurodegenerative diseases: therapeutic implications for antioxidant treatment. Drugs Aging 18:685–716. 5. Tabner BJ, Turnbull S, El-Agnaf O, Allsop D 2001 Production of reactive oxygen species from aggregating proteins implicated in Alzheimer’s disease, Parkinson’s disease, and other neurodegenerative diseases. Curr Top Med Chem 1:507–517. 6. McGeer EG, McGeer PL 1998 The importance of inflammatory mechanisms in Alzheimer disease. Exp Gerontol 33:371–378. 7. Buntinx M, Stinissen P, Steels P, Ameloot M, Raus J 2002 Immune-mediated oligodendrocyte injury in multiple sclerosis: molecular mechanisms and therapeutic interventions. Crit Rev Immunol 22:391–424. 8. Balabanova S, Richter HP, Antoniadis G, Homoki J, Kremmer N, Hanle J, Teller WM 1984 25-Hydroxyvitamin D, 24,25-dihydroxyvitamin D and 1,25-dihydroxyvitamin D in human cerebrospinal fluid. Klin Wochenschr 62:1086–1090. 9. Gascon-Barre M, Huet PM 1983 Apparent [3H]1,25dihydroxyvitamin D3 uptake by canine and rodent brain. Am J Physiol 244:E266–E271.

1786 10. Pardridge WM, Sakiyama R, Coty WA 1985 Restricted transport of vitamin D and A derivatives through the rat bloodbrain barrier. J Neurochem 44:1138–1141. 11. Neveu I, Naveilhan P, Menaa C, Wion D, Brachet P, Garabedian M 1994 Synthesis of 1,25-dihydroxyvitamin D3 by rat brain macrophages in vitro. J Neurosci Res 38:214–220. 12. Zehnder D, Bland R, Williams MC, McNinch RW, Howie AJ, Stewart PM, Hewison M 2001 Extrarenal expression of 25-hydroxyvitamin D3-1 alpha-hydroxylase. J Clin Endocrinol Metab 86:888–894. 13. Fu GK, Lin D, Zhang MY, Bikle DD, Shackleton CH, Miller WL, Portale AA 1997 Cloning of human 25-hydroxyvitamin D-1 alpha-hydroxylase and mutations causing vitamin D– dependent rickets type 1. Mol Endocrinol 11:1961–1970. 14. Stumpf WE, O’Brien LP 1987 1,25 (OH)2 vitamin D3 sites of action in the brain. An autoradiographic study. Histochemistry 87:393–406. 15. Musiol IM, Stumpf WE, Bidmon HJ, Heiss C, Mayerhofer A, Bartke A 1992 Vitamin D nuclear binding to neurons of the septal, substriatal, and amygdaloid area in the Siberian hamster (Phodopus sungorus) brain. Neuroscience 48:841–848. 16. Stumpf WE, Privette TH 1989 Light, vitamin D, and psychiatry. Role of 1,25 dihydroxyvitamin D3 (soltriol) in etiology and therapy of seasonal affective disorder and other mental processes. Psychopharmacology 97:285–294. 17. Bidmon HJ, Stumpf WE 1992 Choroid plexus, ependyma, and arachnoidea express receptors for vitamin D: differences between “seasonal” and “nonseasonal” breeders. Prog Brain Res 91:279–283. 18. Stumpf WE, Privette TH 1991 The steroid hormone of sunlight soltriol (vitamin D) as a seasonal regulator of biological activities and photoperiodic rhythms. J Steroid Biochem Mol Biol 39:283–289. 19. Stumpf WE, Clark SA, O’Brien LP, Reid FA 1988 1,25(OH)2 vitamin D3 sites of action in spinal cord and sensory ganglion. Anat Embryol 177:307–310. 20. Prufer K, Veenstra TD, Jirikowski GF, Kumar R 1999 Distribution of 1,25-dihydroxyvitamin D3 receptor immunoreactivity in the rat brain and spinal cord. J Chem Neuroanat 16:135–145. 21. Clemens TL, Garrett KP, Zhou XY, Pike JW, Haussler MR, Dempster DW 1988 Immunocytochemical localization of the 1,25-dihydroxyvitamin D3 receptor in target cells. Endocrinology 122:1224–1230. 22. Langub MC, Herman JP, Malluche HH, Koszewski NJ 2001 Evidence of functional vitamin D receptors in rat hippocampus. Neuroscience 104:49–56. 23. Walbert T, Jirikowski GF, Prufer K 2001 Distribution of 1,25-dihydroxyvitamin D3 receptor immunoreactivity in the limbic system of the rat. Horm Metab Res 33:525–531. 24. Delacourte A, David JP, Sergeant N, Buee L, Wattez A, Vermersch P, Ghozali F, Fallet-Bianco C, Pasquier F, Lebert F, Petit H, Di Menza C 1999 The biochemical pathway of neurofibrillary degeneration in aging and Alzheimer’s disease. Neurology 52:1158–1165. 25. Johnson JA, Grande JP, Roche PC, Campbell RJ, Kumar R 1995 Immunolocalization of calcitriol receptor, plasma membrane calcium pump, and calbindin-D28k in the cornea and ciliary body of the rat eye. Ophthalmic Res 27:42–47. 26. Johnson JA, Grande JP, Roche PC, Campbell RJ, Kumar R 1995 Immunolocalization of the calcitriol receptor, calbindin-D28k, and the plasma membrane calcium pump in the human eye. Curr Eye Res 14:101–108.

PHILIPPE BRACHET

ET AL .

27. Erben RG, Soegiarto DW, Weber K, Zeitz U, Lieberherr M, Gniadecki R, Moller G, Adamski J, Balling R 2002 Deletion of deoxyribonucleic acid-binding domain of the vitamin D receptor abrogates genomic and nongenomic functions of vitamin D. Mol Endocrinol 16:1524–1537. 28. Veenstra TD, Prufer K, Koenigsberger C, Brimijoin SW, Grande JP, Kumar R 1998 1,25-Dihydroxyvitamin D3 receptors in the central nervous system of the rat embryo. Brain Res 804:193–205. 29. Sutherland MK, Somerville MJ, Yoong LK, Bergeron C, Haussler MR, McLachlan DR 1992 Reduction of vitamin D hormone receptor mRNA levels in Alzheimer as compared to Huntington hippocampus: correlation with calbindin-28k mRNA levels. Brain Res Mol Brain Res 13:239–250. 30. Gill RK, Christakos S 1993 Identification of sequence elements in mouse calbindin-D28k gene that confer 1,25dihydroxyvitamin D3- and butyrate-inducible responses. Proc Natl Acad Sci USA 90:2984–2988. 31. Maguire-Zeiss KA, Li ZW, Shimoda LM, Hamill RW 1995 Calbindin D28k mRNA in hippocampus, superior temporal gyrus, and cerebellum: comparison between control and Alzheimer disease subjects. Brain Res Mol Brain Res 30:362–366. 32. Sutherland MK, Wong L, Somerville MJ, Yoong LK, Bergeron C, Parmentier M, McLachlan DR 1993 Reduction of calbindin-28k mRNA levels in Alzheimer as compared to Huntington hippocampus. Brain Res Mol Brain Res 18:32–42. 33. Iacopino AM, Christakos S 1990 Specific reduction of calcium-binding protein (28-kilodalton calbindin-D) gene expression in aging and neurodegenerative diseases. Proc Natl Acad Sci USA 87:4078–4082. 34. Airaksinen MS, Eilers J, Garaschuk O, Thoenen H, Konnerth A, Meyer M 1997 Ataxia and altered dendritic calcium signaling in mice carrying a targeted null mutation of the calbindin D28k gene. Proc Natl Acad Sci USA 94:1488–1493. 35. Airaksinen MS, Thoenen H, Meyer M 1997 Vulnerability of midbrain dopaminergic neurons in calbindin-D28k-deficient mice: lack of evidence for a neuroprotective role of endogenous calbindin in MPTP-treated and weaver mice. Eur J Neurosci 9:120–127. 36. de Viragh PA, Haglid KG, Celio MR 1989 Parvalbumin increases in the caudate putamen of rats with vitamin D hypervitaminosis. Proc Natl Acad Sci USA 86:3887–3890. 37. Alexianu ME, Robbins E, Carswell S, Appel SH 1998 1Alpha, 25 dihydroxyvitamin D3–dependent up-regulation of calcium-binding proteins in motoneuron cells. J Neurosci Res 51:58–66. 38. Colnot S, Ovejero C, Romagnolo B, Porteu A, Lacourte P, Thomasset M, Perret C 2000 Transgenic analysis of the response of the rat calbindin-D 9k gene to vitamin D. Endocrinology 141:2301–2308. 39. Neveu I, Naveilhan P, Jehan F, Baudet C, Wion D, De Luca HF, Brachet P 1994 1,25-dihydroxyvitamin D3 regulates the synthesis of nerve growth factor in primary cultures of glial cells. Brain Res Mol Brain Res 24:70–76. 40. Naveilhan P, Neveu I, Baudet C, Ohyama KY, Brachet P, Wion D 1993 Expression of 25(OH) vitamin D3 24-hydroxylase gene in glial cells. Neuroreport 5:255–257. 41. Cornet A, Baudet C, Neveu I, Baron-Van Evercooren A, Brachet P, Naveilhan P 1998 1,25-dihydroxyvitamin D3 regulates the expression of VDR and NGF gene in Schwann cells in vitro. J Neurosci Res 53:742–746.

CHAPTER 100 Vitamin D, A Neuroactive Hormone

42. Baas D, Prufer K, Ittel ME, Kuchler-Bopp S, Labourdette G, Sarlieve LL, Brachet P 2000 Rat oligodendrocytes express the vitamin D3 receptor and respond to 1,25-dihydroxyvitamin D3. Glia 31:59–68. 43. Brewer LD, Thibault V, Chen KC, Langub MC, Landfield PW, Porter NM 2001 Vitamin D hormone confers neuroprotection in parallel with down-regulation of L-type calcium channel expression in hippocampal neurons. J Neurosci 21:98–108. 44. Ibi M, Sawada H, Nakanishi M, Kume T, Katsuki H, Kaneko S, Shimohama S, Akaike A 2001 Protective effects of 1 alpha,25-(OH)2D3 against the neurotoxicity of glutamate and reactive oxygen species in mesencephalic culture. Neuropharmacology 40:761–771. 45. Brown J, Bianco JI, McGrath JJ, Eyles, DW 2003 1,25Dihydroxyvitamin D3 induces nerve growth factor, promotes neurite outgrowth, and inhibits mitosis in embryonic rat hippocampal neurons. Neurosci Lett 343:139–143. 46. Racke MM, Mason PJ, Johnson MP, Brankamp RG, Linnik MD 1996 Demonstration of a second pharmacologically active promoter region in the NGF gene that induces transcription at exon 3. Brain Res Mol Brain Res 41:192–199. 47. Neveu I, Naveilhan P, Baudet C, Brachet P, Metsis M 1994 1,25-dihydroxyvitamin D3 regulates NT-3, NT-4 but not BDNF mRNA in astrocytes. Neuroreport 6:124–126. 48. Wion D, MacGrogan D, Neveu I, Jehan F, Houlgatte R, Brachet P 1991 1,25-Dihydroxyvitamin D3 is a potent inducer of nerve growth factor synthesis. J Neurosci Res 28:110–114. 49. Musiol IM, Feldman D 1997 1,25-dihydroxyvitamin D3 induction of nerve growth factor in L929 mouse fibroblasts: effect of vitamin D receptor regulation and potency of vitamin D3 analogs. Endocrinology 138:12–18. 50. Jehan F, Naveilhan P, Neveu I, Harvie D, Dicou E, Brachet P, Wion D 1996 Regulation of NGF, BDNF, and LNGFR gene expression in ROS 17/2.8 cells. Mol Cell Endocrinol 116: 149–156. 51. Veenstra TD, Londowski JM, Windebank AJ, Brimijoin S, Kumar R 1997 Effects of 1,25-dihydroxyvitamin D3 on growth of mouse neuroblastoma cells. Brain Res Dev Brain Res 99:53–60. 52. Saporito MS, Brown ER, Hartpence KC, Wilcox HM, Vaught JL, Carswell S 1994 Chronic 1,25-dihydroxyvitamin D3–mediated induction of nerve growth factor mRNA and protein in L929 fibroblasts and in adult rat brain. Brain Res 633:189–196. 53. Eyles D, Brown J, Mackay-Sim A, McGrath J, Feron F 2003 Vitamin D3 and brain development. Neuroscience 118: 641–653. 54. Sonnenberg J, Luine VN, Krey LC Christakos S 1986 1,25-Dihydroxyvitamin D3 treatment results in increased choline acetyltransferase activity in specific brain nuclei. Endocrinology 118:1433–1439. 55. Naveilhan P, Neveu I, Wion D, Brachet P 1996 1,25Dihydroxyvitamin D3, an inducer of glial cell line-derived neurotrophic factor. Neuroreport 7:2171–2175. 56. Remy S, Naveilhan P, Brachet P, Neveu I 2001 Differential regulation of GDNF, neurturin, and their receptors in primary cultures of rat glial cells. J Neurosci Res 64:242–251. 57. Wang Y, Chiang YH, Su TP, Hayashi T, Morales M, Hoffer BJ, Lin SZ 2000 Vitamin D3 attenuates cortical infarction induced by middle cerebral arterial ligation in rats. Neuropharmacology 39:873–880.

1787 58. Wang JY, Wu JN, Cherng TL, Hoffer BJ, Chen HH, Borlongan CV, Wang Y 2001 Vitamin D3 attenuates 6-hydroxydopamineinduced neurotoxicity in rats. Brain Res 904:67–75. 59. Hempstead BL, Martin-Zanca D, Kaplan DR, Parada LF, Chao MV 1991 High-affinity NGF binding requires coexpression of the trk proto-oncogene and the low-affinity NGF receptor. Nature 350:678–683. 60. Kaplan DR, Miller FD 2000 Neurotrophin signal transduction in the nervous system. Curr Opin Neurobiol 10:381–391. 61. Curtis R, Adryan KM, Stark JL, Park JS, Compton DL, Weskamp G, Huber LJ, Chao MV, Jaenisch R, Lee KF, et al. 1995 Differential role of the low affinity neurotrophin receptor (p75) in retrograde axonal transport of the neurotrophins. Neuron 14:1201–1211. 62. Kramer BM, Van der Zee CE, Hagg T 1999 P75 nerve growth factor receptor is important for retrograde transport of neurotrophins in adult cholinergic basal forebrain neurons. Neuroscience 94:1163–1172. 63. Frade JM, Rodriguez-Tebar A, Barde YA 1996 Induction of cell death by endogenous nerve growth factor through its p75 receptor. Nature 383:166–168. 64. Yoon SO, Casaccia-Bonnefil P, Carter B, Chao MV 1998 Competitive signaling between TrkA and p75 nerve growth factor receptors determines cell survival. J Neurosci 18: 3273–3281. 65. Rabizadeh S, Oh J, Zhong LT, Yang J, Bitler CM, Butcher LL, Bredesen DE 1993 Induction of apoptosis by the low-affinity NGF receptor. Science 261:345–348. 66. Naveilhan P, Neveu I, Baudet C, Funakoshi H, Wion D, Brachet P, Metsis M 1996 1,25-Dihydroxyvitamin D3 regulates the expression of the low-affinity neurotrophin receptor. Brain Res Mol Brain Res 41:259–268. 67. McGrath J 2001 Does “imprinting” with low prenatal vitamin D contribute to the risk of various adult disorders? Med Hypotheses 56:367–371. 68. McGrath J 1999 Hypothesis: is low prenatal vitamin D a riskmodifying factor for schizophrenia? Schizophr Res 40:173–177. 69. Landfield PW, Cadwallader-Neal L 1998 Long-term treatment with calcitriol (1,25(OH)2 vit D3) retards a biomarker of hippocampal aging in rats. Neurobiol Aging 19: 469–477. 70. Garcion E, Thanh XD, Bled F, Teissier E, Dehouck MP, Rigault F, Brachet P, Girault A, Torpier G, Darcy F 1996 1,25-Dihydroxyvitamin D3 regulates gamma 1 transpeptidase activity in rat brain. Neurosci Lett 216:183–186. 71. Garcion E, Sindji L, Leblondel G, Brachet P, Darcy F 1999 1,25-dihydroxyvitamin D3 regulates the synthesis of gammaglutamyl transpeptidase and glutathione levels in rat primary astrocytes. J Neurochem 73:859–866. 72. Garcion E, Sindji L, Montero-Menei C, Andre C, Brachet P, Darcy F 1998 Expression of inducible nitric oxide synthase during rat brain inflammation: regulation by 1,25-dihydroxyvitamin D3. Glia 22:282–294. 73. Garcion E, Nataf S, Berod A, Darcy F, Brachet P 1997 1,25Dihydroxyvitamin D3 inhibits the expression of inducible nitric oxide synthase in rat central nervous system during experimental allergic encephalomyelitis. Brain Res Mol Brain Res 45:255–267. 74. Garcion E, Sindji L, Nataf S, Brachet P, Darcy F, MonteroMenei, CN 2003 Treatment of experimental autoimmune encephalomyelitis in rat by 1,25-dihydroxyvitamin D3 leads to early effects within the central nervous system. Acta Neuropathol (Berl) 105:438–448.

1788 75. Hansen CM, Binderup L, Hamberg KJ, Carlberg C 2001 Vitamin D and cancer: effects of 1,25(OH)2D3 and its analogs on growth control and tumorigenesis. Front Biosci 6:D820–D848. 76. Magrassi L, Butti G, Silini E, Bono F, Paoletti P, Milanesi G 1993 The expression of genes of the steroid-thyroid hormone receptor superfamily in central nervous system tumors. Anticancer Res 13:859–866. 77. Naveilhan P, Berger F, Haddad K, Barbot N, Benabid AL, Brachet P, Wion D 1994 Induction of glioma cell death by 1,25(OH)2 vitamin D3: towards an endocrine therapy of brain tumors? J Neurosci Res 37:271–277. 78. Magrassi L, Butti G, Pezzotta S, Infuso L, Milanesi G 1995 Effects of vitamin D and retinoic acid on human glioblastoma cell lines. Acta Neurochir 133:184–190. 79. Magrassi L, Bono F, Milanesi G, Butti G 1992 Vitamin D receptor expression in human brain tumors. J Neurosurg Sci 36:27–30. 80. Baudet C, Chevalier G, Chassevent A, Canova C, Filmon R, Larra F, Brachet P, Wion D 1996 1,25-Dihydroxyvitamin D3 induces programmed cell death in a rat glioma cell line. J Neurosci Res 46:540–550. 81. Baudet C, Chevalier G, Naveilhan P, Binderup L, Brachet P, Wion D 1996 Cytotoxic effects of 1α,25-dihydroxyvitamin D3 and synthetic vitamin D3 analogs on a glioma cell line. Cancer Lett 100:3–10. 82. Baudet C, Perret E, Delpech B, Kaghad M, Brachet P, Wion D, Caput D 1998 Differentially expressed genes in C6.9 glioma cells during vitamin D–induced cell death program. Cell Death Differ 5:116–125. 83. Diesel B, Fischer U, Meese E 2003 Gene amplification and splice variants of 25-hydroxyvitamin D3 1,α-hydroxylase (CYP27B1) in glioblastoma multiforme—a possible role in tumor progression? Recent Results Cancer Res 164:151–155. 84. Maas RM, Reus K, Diesel B, Steudel WI, Feiden W, Fischer U, Meese E 2001 Amplification and expression of splice variants of the gene encoding the P450 cytochrome 25-hydroxyvitamin D3 1,α-hydroxylase (CYP27B1) in human malignant glioma. Clin Cancer Res 7:868–875. 85. Canova C, Baudet C, Chevalier G, Brachet P, Wion D 1997 Noradrenaline inhibits the programmed cell death induced by 1,25-dihydroxyvitamin D3 in glioma. Eur J Pharmacol 319: 365–368. 86. Davoust N, Wion D, Chevalier G, Garabedian M, Brachet P, Couez D 1998 Vitamin D receptor stable transfection restores the susceptibility to 1,25-dihydroxyvitamin D3 cytotoxicity in a rat glioma-resistant clone. J Neurosci Res 52:210–219. 87. Trouillas P, Honnorat J, Bret P, Jouvet A, Gerard JP 2001 Redifferentiation therapy in brain tumors: long-lasting complete regression of glioblastomas and an anaplastic astrocytoma under long-term 1-α-hydroxycholecalciferol. J Neurooncol 51:57–66. 88. Moore TB, Sidell N, Chow VJ, Medzoyan RH, Huang JI, Yamashiro JM, Wada RK 1995 Differentiating effects of 1,25-dihydroxycholecalciferol D3 on LA-N-5 human neuroblastoma cells and its synergy with retinoic acid. J Pediatr Hematol Oncol 17:311–317. 89. Gumireddy K, Reddy GS, Ikegaki N, Binderup L, Sutton LN, Phillips PC, Reddy CD 2003 Anti-proliferative effects of 20-epi-vitamin-D3 analog, KH1060 in human neuroblastoma: induction of RAR-beta and p21(Cip1). Cancer Lett 190:51–60. 90. Stio M, Celli A, Treves C 2001 Synergistic antiproliferative effects of vitamin D derivatives and 9-cis retinoic acid in SH-SY5Y human neuroblastoma cells. J Steroid Biochem Mol Biol 77:213–222.

PHILIPPE BRACHET

ET AL .

91. Agranoff BW, Goldberg D 1974 Diet and the geographical distribution of multiple sclerosis. Lancet 2:1061–1066. 92. Fukazawa T, Yabe I, Kikuchi S, Sasaki H, Hamada T, Miyasaka K, Tashiro K 1999 Association of vitamin D receptor gene polymorphism with multiple sclerosis in Japanese. J Neurol Sci 166:47–52. 93. Niino M, Fukazawa T, Yabe I, Kikuchi S, Sasaki H, Tashiro K 2000 Vitamin D receptor gene polymorphism in multiple sclerosis and the association with HLA class II alleles. J Neurol Sci 177:65–71. 94. Lemire JM, Archer DC 1991 1,25-dihydroxyvitamin D3 prevents the in vivo induction of murine experimental autoimmune encephalomyelitis. J Clin Invest 87:1103–1107. 95. Lemire JM, Adams JS 1992 1,25-dihydroxyvitamin D3 inhibits the passive transfer of cellular immunity by a myelin basic protein-specific T-cell clone. J Bone Miner Res 7:171–177. 96. Nataf S, Garcion E, Darcy F, Chabannes D, Muller JY, Brachet P 1996 1,25 dihydroxyvitamin D3 exerts regional effects in the central nervous system during experimental allergic encephalomyelitis. J Neuropathol Exp Neurol 55: 904–914. 97. Cantorna MT, Hayes CE, DeLuca HF 1996 1,25Dihydroxyvitamin D3 reversibly blocks the progression of relapsing encephalomyelitis, a model of multiple sclerosis. Proc Natl Acad Sci USA 93:7861–7864. 98. Meehan TF, DeLuca HF 2002 The vitamin D receptor is necessary for 1α,25-dihydroxyvitamin D3 to suppress experimental autoimmune encephalomyelitis in mice. Arch Biochem Biophys 408:200–204. 99. Swanborg RH 2001 Experimental autoimmune encephalomyelitis in the rat: lessons in T-cell immunology and autoreactivity. Immunol Rev 184:129–135. 100. Lemire JM, Archer DC, Beck L, Spiegelberg HL 1995 Immunosuppressive actions of 1,25-dihydroxyvitamin D3: preferential inhibition of Th1 functions. J Nutr 125: 1704S–1708S. 101. D’Ambrosio D, Cippitelli M, Cocciolo MG, Mazzeo D, Di Lucia P, Lang R, Sinigaglia F, Panina-Bordignon P 1998 Inhibition of IL-12 production by 1,25-dihydroxyvitamin D3. Involvement of NF-κB downregulation in transcriptional repression of the p40 gene. J Clin Invest 101:252–262. 102. Cantorna MT, Woodward WD, Hayes CE, DeLuca HF 1998 1,25-dihydroxyvitamin D3 is a positive regulator for the two anti-encephalitogenic cytokines TGF-beta 1 and IL-4. J Immunol 160:5314–5319. 103. Kreutz M, Andreesen R 1990 Induction of human monocyte to macrophage maturation in vitro by 1,25-dihydroxyvitamin D3. Blood 76:2457–2461. 104. Provvedini DM, Deftos LJ, Manolagas SC 1986 1,25Dihydroxyvitamin D3 promotes in vitro morphologic and enzymatic changes in normal human monocytes consistent with their differentiation into macrophages. Bone 7:23–28. 105. Clavreul A, D’Hellencourt CL, Montero-Menei C, Potron G, Couez D 1998 Vitamin D differentially regulates B7.1 and B7.2 expression on human peripheral blood monocytes. Immunology 95:272–277. 106. Griffin MD, Lutz W, Phan VA, Bachman LA, McKean DJ, Kumar R 2001 Dendritic cell modulation by 1α,25-dihydroxyvitamin D3 and its analogs: a vitamin D receptor-dependent pathway that promotes a persistent state of immaturity in vitro and in vivo. Proc Natl Acad Sci USA 98:6800–6805. 107. Penna G, Adorini L 2000 1α,25-dihydroxyvitamin D3 inhibits differentiation, maturation, activation, and survival of dendritic cells leading to impaired alloreactive T-cell activation. J Immunol 164:2405–2411.

CHAPTER 100 Vitamin D, A Neuroactive Hormone

108. Piemonti L, Monti P, Sironi M, Fraticelli P, Leone BE, Dal Cin E, Allavena P, Di Carlo V 2000 Vitamin D3 affects differentiation, maturation, and function of human monocyte-derived dendritic cells. J Immunol 164: 4443–4451. 109. Berer A, Stockl J, Majdic O, Wagner T, Kollars M, Lechner K, Geissler K, Oehler L 2000 1,25-dihydroxyvitamin D3 inhibits dendritic cell differentiation and maturation in vitro. Exp Hematol 28:575–583. 110. Rigby WF, Denome S, Fanger MW 1987 Regulation of lymphokine production and human T lymphocyte activation by 1,25-dihydroxyvitamin D3. Specific inhibition at the level of messenger RNA. J Clin Invest 79:1659–1664. 111. Reichel H, Koeffler HP, Tobler A, Norman AW 1987 1α,25-dihydroxyvitamin D3 inhibits gamma-interferon synthesis by normal human peripheral blood lymphocytes. Proc Natl Acad Sci USA 84:3385–3389. 112. Takeuchi A, Reddy GS, Kobayashi T, Okano T, Park J, Sharma S 1998 Nuclear factor of activated T-cells (NFAT) as a molecular target for 1α,25-dihydroxyvitamin D3– mediated effects. J Immunol 160:209–218. 113. Alroy I, Towers TL, Freedman LP 1995 Transcriptional repression of the interleukin-2 gene by vitamin D3: direct inhibition of NFATp/AP-1 complex formation by a nuclear hormone receptor. Mol Cell Biol 15:5789–5799. 114. Lefebvre d’Hellencourt C, Montero-Menei CN, Bernard R, Couez D 2003 Vitamin D3 inhibits proinflammatory cytokines and nitric oxide production by the EOC13 microglial cell line. J Neurosci Res 71:575–582. 115. Neumann H, Misgeld T, Matsumuro K, Wekerle H 1998 Neurotrophins inhibit major histocompatibility class II inducibility of microglia: involvement of the p75 neurotrophin receptor. Proc Natl Acad Sci USA 95:5779–5784. 116. Furman I, Baudet C, Brachet P 1996 Differential expression of M-CSF, LIF, and TNF-α genes in normal and malignant rat glial cells: regulation by lipopolysaccharide and vitamin D. J Neurosci Res 46:360–366. 117. Mattner F, Smiroldo S, Galbiati F, Muller M, Di Lucia P, Poliani PL, Martino G, Panina-Bordignon P, Adorini L 2000 Inhibition of Th1 development and treatment of chronicrelapsing experimental allergic encephalomyelitis by a nonhypercalcemic analog of 1,25-dihydroxyvitamin D3. Eur J Immunol 30:498–508.

1789 118. Van Etten E, Branisteanu DD, Overbergh L, Bouillon R, Verstuyf A, Mathieu C 2003 Combination of a 1,25dihydroxyvitamin D3 analog and a bisphosphonate prevents experimental autoimmune encephalomyelitis and preserves bone. Bone 32:397–404. 119. Salehi A, Delcroix JD, Mobley WC 2003 Traffic at the intersection of neurotrophic factor signaling and neurodegeneration. Trends Neurosci 26:73–80. 120. Kordower JH, Emborg ME, Bloch J, Ma SY, Chu Y, Leventhal L, McBride J, Chen EY, Palfi S, Roitberg BZ, Brown WD, Holden JE, Pyzalski R, Taylor MD, Carvey P, Ling Z, Trono D, Hantraye P, Deglon N, Aebischer P 2000 Neurodegeneration prevented by lentiviral vector delivery of GDNF in primate models of Parkinson’s disease. Science 290:767–773. 121. Derex L, Trouillas P 1997 Reversible Parkinsonism, hypophosphoremia, and hypocalcemia under vitamin D therapy. Mov Disord 12:612–613. 122. Damier P, Hirsch EC, Zhang P, Agid Y, Javoy-Agid F 1993 Glutathione peroxidase, glial cells, and Parkinson’s disease. Neuroscience 52:1–6. 123. Iravani MM, Kashefi K, Mander P, Rose S, Jenner P 2002 Involvement of inducible nitric oxide synthase in inflammation-induced dopaminergic neurodegeneration. Neuroscience 110:49–58. 124. Hunot S, Boissiere F, Faucheux B, Brugg B, Mouatt-Prigent A, Agid Y, Hirsch EC 1996 Nitric oxide synthase and neuronal vulnerability in Parkinson’s disease. Neuroscience 72: 355–363. 125. Gloth FM 3rd, Gundberg CM, Hollis BW, Haddad JG Jr Tobin JD 1995 Vitamin D deficiency in homebound elderly persons. JAMA 274:1683–1686. 126. Nieves J, Cosman F, Herbert J, Shen V, Lindsay R 1994 High prevalence of vitamin D deficiency and reduced bone mass in multiple sclerosis. Neurology 44:1687–1692. 127. Sato Y, Oizumi K, Kuno H, Kaji M 1999 Effect of immobilization upon renal synthesis of 1,25-dihydroxyvitamin D in disabled elderly stroke patients. Bone 24: 271–275. 128. Sato Y, Asoh T, Oizumi K 1998 High prevalence of vitamin D deficiency and reduced bone mass in elderly women with Alzheimer’s disease. Bone 23:555–557.

CHAPTER 101

Psoriasis and Other Skin Diseases JÖRG REICHRATH The Saarland University Hospital, Hamburg, Germany

MICHAEL F. HOLICK Boston University, Boston, MA

I. II. III. IV.

Introduction Pathogenesis of Psoriasis The Vitamin D System in Normal and Psoriatic Skin Physiological and Pharmacological Actions of Vitamin D Analogs in Normal and Psoriatic Skin V. Clinical Use of 1,25(OH)2D3 and its Analogs in Psoriasis

VI. Vitamin D Analog Therapy in Other Skin Diseases VII. Perspectives for the Development of New Vitamin D Analogs with Less Calcemic Activity for the Treatment of Hyperproliferative Skin Disorders References

I. INTRODUCTION

II. PATHOGENESIS OF PSORIASIS

Earlier in the last century, vitamin D3 was used in dermatology in huge pharmacological doses for the treatment of scleroderma, psoriasis, lupus vulgaris, and atopic dermatitis. A rationale for the use of vitamin D in psoriasis was the clinical observation that this disease, in general, markedly improves in the summer. In 1936, J. Krafka wrote in The Journal of Laboratory and Clinical Medicine under the headline “A simple treatment for psoriasis”: “A commonly observed fact in the South concerning this disease (psoriasis) is that it generally clears up in the summer sun. This led the author to the hypothesis that it might be cured with viosterol .… A patient with a case of ten years standing, continuous duration was put on viosterol (an oil containing vitamin D2) … . Within sixty days from the beginning of the test, the skin of the patient was entirely clear” [1]. In 1950, H.W. Spier reported results of a study with 94 psoriasis patients performed in 1948–49 [2]. First, patients received orally 30 mg vitamin D2/week for 2–4 weeks, and thereafter 20 mg vitamin D2/week until a total dose of appr. 300 mg vitamin D2 was reached. In this study, patients were treated for 3–4 months; 20% of patients showed a good response, 25% a satisfactory response, 25% a moderate response, and 30% a nonsatisfactory response. But these first attempts to use vitamin D treatment in dermatology were abandoned because of severe vitamin D intoxication resulting in hypercalcemia, hypercalciuria, and kidney stones occurring in patients that received huge pharmacological doses of vitamin D (up to 1000-fold the regular daily need of vitamin D). Also, they were stopped because other new treatments were introduced for the therapy of these diseases.

A. Psoriasis: Pathogenesis, Immunology, and Histology of Skin Lesions

VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

Psoriasis vulgaris is a chronic dermatosis of unknown etiology characterized by hyperproliferation and inflammation of the skin. The most common form of the disease is plaque psoriasis, in which skin develops scaly, raised red lesions. The severity of chronic plaque psoriasis ranges from mild, when the disease has only a mild impact on quality of life, to severe, when patients’ lives are significantly affected [3]. In severe cases, most of the body surface, including the scalp and nails, may be involved. The peak age of onset for this psychologically debilitating and disfiguring disease is the second decade, but psoriasis may first appear at any age from infancy to the aged [4]. It is considered a multifactorial disease and has a prevalence of about 1–2% in the United States and Europe. Population, family, and twin studies clearly demonstrate that there is a strong but very complex genetic component leading to the development of psoriatic skin lesions [5]. Most likely, multiple genes are involved in the pathogenesis of psoriasis. Molecular biology techniques have been developed that allow studies to analyze psoriasis susceptibility genes, but to date, no specific genetic marker of the disease has been identified. Psoriasis has long been known to be associated with certain HLA antigens, particularly HLA-Cw6, although there is no evidence that a psoriasis susceptibility gene exists at this locus [6]. It is unknown what cell types in human skin are primarily affected by the disease. Recent studies support the hypothesis that epidermal hyperproliferation in psoriasis may be mediated by cells of the immune system, most likely T lymphocytes [7,8]. Copyright © 2005, Elsevier, Inc. All rights reserved.

1792 The vast majority of T-cells in psoriatic lesions are situated in the perivascular area in the dermis; many are also found in the epidermis. Activated CD4+ and/or CD8+ T-cells in psoriatic lesions express HLA-DR, the interleukin-2 receptor (CD 25), bear the CLA+ memoryeffector CD45RO+ phenotype, and secrete specific immune mediators and cytokines, such as IL-2 and interferon-γ [8–11]. Thus, psoriasis represents mainly a Th1 profile disease (characterized by T-lymphocyte secretion of IL-2, IL-12, and interferon-γ) [12]. In contrast, atopic dermatitis represents a Th2 profile disease that is characterized by T-cell secretion of IL-4, IL-5, and IL-10 [13]. The activation signal for the development of psoriatic lesions is still unknown, although there is increasing evidence that superantigens such as the N-terminal component of bacterial M-proteins may be of importance for the initiation of T-cell proliferation in psoriasis [8,14]. It has also been hypothesized that psoriasis patients develop an effector-immune response to skin (auto)antigens, which have yet to be specifically identified. According to this model, the immunologic process underlying psoriasis begins with a sensitization-type phase during which the skin dendritic cells migrate to regional lymph nodes where they present these skin antigens to naïve T-cells. This sensitization phase occurs prior to the development of skin lesions [8,15]. When sensitization is obtained, the psoriasis skin lesion may develop as a result of the migration of T-cells in the skin where they are activated by antigen presenting cells, including Langerhans’ cells presenting self-skin antigens [8,15]. The histologic appearance depends on the age of the psoriatic lesion and the site of the biopsy. In general, epidermal hyperplasia is present, in which the granular layer may be lost and the stratum corneum shows parakeratosis. Typical lesions show histological elongation of the dermal papillae, with a relatively thin epidermis at the top of the papillae. The epidermis may show in suprapapillar compartments intercellular edema and infiltration with T-lymphocytes and neutrophiles, which can extend into spongiform pustules of Kogoj or Munro microabscesses [16].

JÖRG REICHRATH AND MICHAEL F. HOLICK

Nongenomic effects of 1,25(OH)2D3 and analogs are related to effects on intracellular calcium [21,22]. In keratinocytes and other cell types, 1,25(OH)2D3 rapidly increases free cytosolic calcium levels [21,22] (Chapter 23). Genomic effects of 1,25(OH)2D3 are mediated via binding to a nuclear receptor protein (vitamin D receptor; VDR) that is present in target tissues and binds 1,25(OH)2D3 with high affinity (KD 10−9−10−10 M) and low capacity [23,24] (Chapters 11,13). The human VDR has been cloned [25] and sequence analysis demonstrated that this protein belongs to the superfamily of trans-acting transcriptional regulatory factors, which includes the steroid and thyroid hormone receptors and the retinoic acid receptors [25] (Chapter 11). Interaction of 1,25(OH)2D3 with VDR results in the phosphorylation of the receptor complex that in turn activates the transcription of 1,25(OH)2D–sensitive target genes, especially genes involved in cellular differentiation and proliferation. VDR requires auxillary factors for sufficient DNAbinding [26]. These auxillary proteins were identified as

III. THE VITAMIN D SYSTEM IN NORMAL AND PSORIATIC SKIN Vitamin D is made in the skin as a result of exposure to solar or artificial ultraviolet B radiation as discussed in Chapter 3 [17,18]. It is now known that the skin itself is a target tissue for the secosteroid hormone 1α,25dihydroxyvitamin D (1,25(OH)2D, calcitriol), the biologically active vitamin D metabolite [19,20]. 1,25(OH)2D3 exerts genomic and nongenomic effects.

FIGURE 1

Immunohistochemical demonstration of VDR in human skin. Notice strong nuclear VDR immunoreactivity in cells of all layers of the viable epidermis (arrows). Labeled avidin-biotin technique using antibody 9A7γ directed against VDR. Original magnification ×400.

1793

CHAPTER 101 Psoriasis and Other Skin Diseases

the retinoid-X receptors (RXR-α,-β,-γ) that were shown to heterodimerize with VDR, and facilitate DNA-binding to the respective vitamin D response elements (VDRE) in the promoter region of target genes [26,27]. In the skin, both VDR (Fig. 1) and RXR-α are expressed in keratinocytes, fibroblasts, Langerhans cells, sebaceous gland cells, endothelial cells, and most cell types related to the skin immune system [28,29]. In vitro studies revealed that 1,25(OH)2D3 is extremely effective in inducing the terminal differentiation and in inhibiting the proliferation of cultured human keratinocytes in a dose-dependent manner [30–32] (Chapter 35). Additionally, 1,25(OH)2D3 acts on many cell types involved in immunologic reactions, including activated T and B lymphocytes, macrophages, and Langerhans cells [33,34] (Chapter 36). Data about the effects of 1,25(OH)2D3 on the melanin pigmentation system are still conflicting, but most studies do not support the possibility that 1,25(OH)2D3 regulates melanogenesis in human skin [35].

IV. PHYSIOLOGICAL AND PHARMACOLOGICAL ACTIONS OF VITAMIN D ANALOGS IN NORMAL AND PSORIATIC SKIN A. Biological Effects of Vitamin D and Analogs in Psoriasis

FIGURE 2

The mechanisms underlying the therapeutic effectiveness of vitamin D and its analogs in psoriasis are still not completely understood (Fig. 2). The analogs are described in Section VIII of this book. Calcipotriol, which is used for the topical treatment of psoriasis, is discussed in Chapter 84. Results from immunohistochemical and molecular biology studies indicate that the antiproliferative effects of topical 1,25(OH)2D3 on epidermal keratinocytes are more pronounced as compared to effects on dermal inflammation (Fig. 3). Modulation of various markers of epidermal proliferation (proliferating cell nuclear antigen, Ki-67 antigen), and differentiation (involucrin, transglutaminase K, filaggrin, cytokeratins 10,16) in lesional psoriatic skin after topical application of vitamin D analogs was shown in situ [36] (Fig. 4). Interestingly, effects of topical treatment with vitamin D analogs on dermal inflammation are less pronounced (CD-antigens, cytokines, HLA-DR, etc.) as compared to effects on epidermal proliferation or differentiation. One reason for this observation may be that the bioavailability of this potent hormone in the dermal compartment may be markedly reduced as compared to the epidermal compartment [36].

Molecular biology studies have demonstrated that clinical improvement in psoriatic lesions treated with 1,25(OH)2D3 correlates with an elevation of VDR mRNA [37]. It is well known that some patients suffering from psoriasis either are resistant to or develop resistance to 1,25(OH)2D3 treatment. It was demonstrated that responders can be distinguished from the nonresponders on the molecular level since nonresponders show no elevation of VDR mRNA in skin lesions during treatment. These data suggest that the ability of 1,25(OH)2D3 to regulate keratinocyte growth is closely linked to the regulation of VDR expression. The target genes of topical 1,25(OH)2D3 that are responsible for its therapeutic efficacy in psoriasis are still unknown. Major candidates for 1,25(OH)2D3 target genes that are responsible for the 1,25(OH)2D3–induced terminal differentiation in keratinocytes are distinct cell cycle associated proteins (i.e., INK4 family), including p21/WAF-1 [38].

(Top) shows the arms of a patient with a long history of plaque psoriasis before treatment. (Bottom) shows the same patient, who applied only petroleum jelly to the forearm on the right and petroleum jelly containing 1,25(OH)2D3 (15 µg/g) to the forearm on the left.

1794

JÖRG REICHRATH AND MICHAEL F. HOLICK

A

B

C

D

FIGURE 3 Histological demonstration of morphological changes in lesional psoriatic skin after six weeks of topical treatment. A. lesional

psoriatic skin before treatment, B. calcitriol 15 µg/g; C. calcipotriol 50 µg/g; D. nonlesional psoriatic skin. Notice strong reduction of epidermal thickness after topical treatment with vitamin D analogs. Hematoxylin-Eosin staining. Original magnification × 200. (See color plate).

1795

CHAPTER 101 Psoriasis and Other Skin Diseases

A

B

C FIGURE 4 Immunohistological detection of transglutaminase K in lesional psoriatic skin. (A) before treatment; (B) lesional psoriatic skin after six weeks of topical treatment with calcipotriol (50 µg/g); and (C) in nonlesional psoriatic skin. Notice strong staining for transglutaminase K in all epidermal cell layers of lesional psoriatic skin before treatment (A, arrows). In contrast, after six weeks of topical treatment with calcipotriol staining in lesional psoriatic skin (B, arrows) is restricted to the upper layers of the viable epidermis, a staining pattern that is characteristic for nonlesional psoriatic skin (C, arrows). Original magnification ×160. (See color plate).

1796 B. VDR Genotypes and Response to Treatment Data analyzing VDR genotype in psoriasis are somewhat conflicting—some studies report a correlation between individual VDR genotypes and the skin eruptions of psoriasis [37,39]. The VDR genotype are described in detail in Chapter 68. While no differences in VDR genotype between controls and psoriasis patients were reported at the Bsm-1 site, some studies reported a significant difference in terms of the Apa-1 [40] and Fok-1 [41] sites. Additionally, it has been shown that VDR genotypes are not associated with clinical response to calcipotriol, at least in Korean psoriasis patients [42]. Kontula et al. [43] and Mee et al. [44] studied the Bsm-1 polymorphism and the response to calcipotriol treatment in psoriatic patients and found no association between them. Recently, it has been shown that the frequency of the F-allele (Fok-1 site) was lower in patients nonresponsive to topical treatment with tacalcitol compared with controls (47 vs. 64%, p = 0.05). Although the number of patients examined in this study was small, this result is intriguing because the F-allele has been demonstrated to be more responsive to 1,25(OH)2D3 than F-allele in vitro (see Chapter 13 and 68). Hutchinson et al. [45] reported that the F-allele was significantly less common in malignant melanoma cases than controls, suggesting consistency with in vitro data. According to Colin et al. [46], the Fok-1 polymorphism was associated with the response to 1,25(OH)2D3, and under conditions of vitamin D insufficiency this finding might have clinical implications.

C. Vitamin D Levels Data concerning serum levels of 1,25(OH)2D or 25-hydroxyvitamin D [25(OH)D] in psoriatic patients are conflicting. Some studies report reduced concentrations of 1,25(OH)2D in patients with manifest disease [47], while others report normal levels [48]. Additionally, the coincidence of pustular psoriasis with hypocalcemia [49] and the exacerbation of psoriasis under chloroquin therapy (thereby reducing 1,25(OH)2D levels via inhibition of the extrarenal 25-hydroxyvitamin D-1α-hydroxylase) are well known [50] and may give rise to speculation about vitamin D levels and severity of psoriasis.

V. CLINICAL USE OF 1,25(OH)2D3 AND ITS ANALOGS IN PSORIASIS A. Topical Use The use of 1,25(OH)2D3 and its analogs for the treatment of psoriasis resulted from two independent lines of investigation. Since psoriasis is a hyperproliferative

JÖRG REICHRATH AND MICHAEL F. HOLICK

skin disorder, it seemed reasonable that the antiproliferative effects of 1,25(OH)2D3 could be used for the treatment of this disease. However, before launching clinical trials in 1985, MacLaughlin and associates reported the observation that psoriatic fibroblasts are partially resistant to the antiproliferative effects of 1,25(OH)2D3 [51]. This observation prompted the authors to speculate that 1,25(OH)2D3 may be effective in pharmacologic dosages for the treatment of psoriasis. The other line of investigation resulted from a clinical observation. In 1985, Morimoto and Kumahara reported that a patient, who was treated orally with 1α-hydroxyvitamin D3 for osteoporosis, had a dramatic remission of psoriatic skin lesions [52]. Morimoto et al. reported a follow-up study, demonstrating that almost 80% of 17 patients with psoriasis who were treated orally with 1α-hydroxyvitamin D3 at a dose of 1.0 µg/day for up to six months showed clinically significant improvement [53]. Numerous studies have reported that various vitamin D analogs, including 1,25(OH)2D3 (calcitriol), calcipotriol (calcipotriene), tacalcitol (1,24-(OH)2D3), hexafluoro-1,25-(OH)2D3 [54], and maxacalcitol (22-oxo-1,25-(OH)2D3) are effective and safe in the topical treatment of psoriasis [55–62]. It has been shown that topical 1,25(OH)2D3 is very effective and safe in the long-term treatment of psoriasis vulgaris [55] (Fig. 3). Applied twice daily topically in amounts of up to 100 g ointment (50 µg calcipotriol/g ointment) per week, calcipotriol, a synthetic analog of calcitriol, was shown to be slightly more effective in the topical treatment of psoriasis than betamethasone 17-valerate ointment [63]. Recently, efficacy of topical treatment with maxacalcitol was compared with topical calcipotriol treatment [58]. In this study, investigators’ overall assessment suggested that maxacalcitol 25 µg/g applied daily may be more effective than once-daily calcipotriol (50 µg/g). It has been reported that a mild dermatitis can be seen in about 10% of patients treated with calcipotriol (50 µg/g), particularly on the face [64]. This side effect (mild dermatitis on the face) is not reported after topical treatment with 1,25(OH)2D3. Allergic contact dermatitis to vitamin D analogs is very rare, however, cases with allergic contact dermatitis to other ingredients of the ointment including propylene glycol have been reported [65–67]. The most common adverse event observed in psoriasis patients treated with maxacalcitol (6–50 µg/g maxacalcitol/g ointment) was burning of the target plaque [58]. In three out of four patients that developed this side effect in one study, symptoms were severe enough to require discontinuation of the treatment [58]. A double-blind, right/left comparison, placebocontrolled evaluation demonstrated efficacy and safety

1797

CHAPTER 101 Psoriasis and Other Skin Diseases

of topical treatment with hexafluoro-1,25(OH)2D3 (50 µg/g) in psoriasis patients [59]. Adverse events included mild irritation. This irritation did not necessitate discontinuation of the study medication. During the large area topical application study period, a cobblestone appearance was initially noted in a few patients. This resolved with continued therapy after 3–4 weeks. Hexafluoro-1,25(OH)2D3–treated plaques also developed very mild perilesional scales as observed with other vitamin D analogs [59]. Efficacy and safety of topical treatment with tacalcitol (4 µg/g and 20 µg/g) has been shown as well [61,62,68]. In one study [62], tacalcitol-treatment was generally well-tolerated, and there were no serious or unexpected adverse events reported. However, discontinuation of treatment as a result of skin irritation was seen in 5.9% of these patients [62]. The greatest frequency of cutaneous side-effects occurred during initial treatment and the incidence decreased markedly and the treatment was well-tolerated with continued use [62]. Recently, the results of four separate studies designed to evaluate specific local-safety parameters of vitamin D analogs including cumulative irritancy, cutaneous contact sensitization, photoallergic contact sensitization, and phototoxicity were analyzed [69]. 1,25(OH)2D3 3 µg/g ointment was classified as nonirritant when compared to calcipotriol, tacalcitol, and white petrolatum. Petrolatum and tacalcitol were slightly irritant and calcipotriol moderately irritant. No sensitization was observed with 1,25(OH)2D3 3 µg/g ointment. With regard to phototoxic potential, sites treated with 1,25(OH)2D3 3 µg/g ointment or vehicle ointment were less irritated than those treated with white petrolatum or those that were untreated. Using standard photoallergenicity testing methodology, there were no skin reactions of a photoallergic nature to the 1,25(OH)2D3 [69]. Patients with psoriasis may need intermittent treatment for the rest of their lives. Vitamin D analogs have been shown not to exhibit tachyphylaxis during treatment of psoriatic lesions and can be continued indefinitely. They are effective and safe for the treatment of skin areas that are usually difficult to treat in psoriatic patients and that respond slowly. Additionally, vitamin D analogs are effective in the treatment of psoriatic skin lesions in children and in HIV-patients.

FIGURE 5

Forty-year-old male with psoriasis for 15 years. Before (top) and 6 months after (bottom) 2 µg each night of oral 1,25(OH)2D3. Reproduced with permission [70].

Serum calcium concentrations and 24 h urinary calcium excretion increased by 3.9% and 148.2%, respectively, but were not outside the normal range. Bone mineral density of these patients remained unchanged. A very important consideration for the use of orally administered calcitriol is the dosing technique. To avoid its effects on enhancing dietary calcium absorption, it is very important to provide 1,25(OH)2D3 at nighttime and with a low calcium diet. Perez et al. [70] showed that as a result of this dosing technique, doses of 2 µg/ to 4 µg/night are well-tolerated by psoriatic patients, although doses over 2 µg caused hypercalcemia in prostate cancer trials (Chapter 94).

B. Oral Use Recently, a long-term follow-up study demonstrated the efficacy and safety of oral 1,25(OH)2D3 in the potential treatment of psoriasis [70] (Fig. 5). Of the 85 patients included in that study who received oral 1,25(OH)2D3, after 36 months 88.0% had some improvement in their disease, 26.5%, 26.3%, and 25.3% had complete, moderate, and slight improvement in their disease, respectively.

C. Specific Therapies 1. TREATMENT OF SCALP PSORIASIS

Recently, a double-blind, randomized multicenter study demonstrated that calcipotriol solution is effective in the topical treatment of scalp psoriasis [71–73]. Forty-nine patients were treated twice daily over

1798

JÖRG REICHRATH AND MICHAEL F. HOLICK

a four-week period [71]. In the study, 60% of the patients on calcipotriol showed clearance or marked improvement vs. 17% in the placebo group. No side effects were reported. 2. TREATMENT OF NAIL PSORIASIS

The occurrence of nail psoriasis has been reported in up to 50% of patients. Nails in general are very difficult to treat and respond slowly to other therapies. Up to now, there has been no consistently effective treatment for psoriatic nails. Recently, it was shown that calcipotriol ointment is effective in the treatment of nail psoriasis [74]. 3. TREATMENT OF FACE AND FLEXURES

Although the use of calcipotriol ointment is not recommended on face and skin flexures due to irritancy, most patients tolerate vitamin D analogs on these sites. Recently, it has been shown that calcitriol ointment (3 µg of calcitriol per gram of petrolatum) was found to be better-tolerated and would appear to be more effective than calcipotriol ointment (50 µg/g of petrolatum) in the treatment of psoriasis in sensitive areas [75]. 4. TREATMENT OF SKIN LESIONS IN CHILDREN

During the last few years, it has been shown that topical application of vitamin D analogs, including 1,25(OH)2D3 ointment (3 µg of calcitriol per gram of petrolatum), is an effective, safe, and reliable therapy to treat psoriatic skin lesions in children [76–78]. 5. TREATMENT OF PSORIATIC LESIONS IN HIV-PATIENTS

We treated an HIV-positive patient suffering from psoriatic skin lesions with topical and oral calcitriol. The patient responded well, and there was no evidence of enhancement in HIV-disease activity or alterations in the number of T lymphocytes or CD4+ and CD8+ cells (Fig. 6).

FIGURE 6 A 46-year-old man with AIDS-associated psoriasis. (top) before, and (bottom) after 3 months of oral 1,25(OH)2D3 treatment.

6. COMBINATION OF VITAMIN D ANALOGS WITH OTHER THERAPIES

Recently, it was reported that efficacy of topical treatment with vitamin D analogs in psoriasis can be increased by combination with other therapies, including methotrexate (MTX), very low-dose oral cyclosporine (2 mg/kg/day), oral acitretin, topical dithranol, topical steroids, PUVA (psoralen plus UVA) and UV-B or narrow band UV-B phototherapy [79–88]. It has been shown that the combination of calcipotriol and MTX is safe and well-tolerated [87]. The combination resulted in lower cumulative dosages of MTX compared with MTX and vehicle. Therefore, the risk of MTX-induced side-effects is substantially decreased [87]. Addition of calcipotriol

ointment to oral application of acitretin (a vitamin A analog) was shown to produce a significantly better treatment response achieved with a lower cumulative dose of acitretin in patients with severe extensive psoriasis, as compared with the group of patients treated with oral acitretin alone [81]. The number of patients reporting adverse events was similar between the two treatment groups [81]. Complete clearing or 90% improvement in psoriasis area and severity index (PASI) was observed in 50% of patients treated with calcipotriol/cyclosporine vs. 11.8% in the placebo/cyclosporine group. No difference was

1799

CHAPTER 101 Psoriasis and Other Skin Diseases

found in that study between the groups in short-time side effects. Kragballe and co-workers reported that efficacy of topical calcipotriol treatment in psoriasis can be improved by simultaneous ultraviolet B phototherapy. Combination therapy of psoriasis with topical calcipotriol and narrow-band UVB has been shown to be very effective for the treatment of psoriatic plaques [80]. It has been shown that the combination of topical treatment with vitamin D analogs and UV radiation does not alter the tolerability or safety of therapy [88]. Vitamin D analogs may be topically applied at any time up to two hours before or immediately after UV radiation [88]. Results of a controlled, right/left study have demonstrated that pretreatment of psoriasis with the vitamin D3 derivative tacalcitol increases the responsiveness to 311-nm ultraviolet B [89]. Additionally, it was shown that tacalcitol ointment (4 µg/g) and 0.1% tazarotene gel are both comparably effective in improving the therapeutic result of PUVA therapy in patients with chronic, plaquetype psoriasis [90]. The treatment requirements to induce complete or near complete clearing were significantly lower for both combination treatments than for PUVA monotherapy (P < 0.01). The median cumulative UVA dose and number of exposures were 30.6 J/cm2 (95% confidence interval, CI 22.5–71.2) and 14 (95% CI 11–16) for tacalcitol plus PUVA, 32.3 J/cm2 (95% CI 22.5–73.8) and 14 (95% CI 11–19) for tazarotene plus PUVA, and 37.0 J/cm2 (95% CI 29.5–83.9) and 16 (95% CI 14–22) for PUVA monotherapy. No difference between the three regimens was observed with regard to duration of remission. Adverse reactions occurred more often with 0.1% tazarotene than with tacalcitol but were, in general, mild and completely reversible upon using a lower concentration of 0.05% tazarotene. It has been concluded that besides accelerating the treatment response, both agents, by virtue of their UVA dose-sparing effect, might also help to reduce possible long-term hazards of PUVA treatment. Previously, a case report described two patients treated with a combination treatment of calcipotriol and both psoralens and ultraviolet A who developed hyperpigmentation at the lesional sites where calcipotriol ointment was applied [91]. Combined topical treatment with calcipotriol ointment (50 µg/g) and betamethasone ointment was shown to be slightly more effective and caused less skin irritation than calcipotriol used twice daily [82]. Recently, a new vehicle has been created with the objective of obtaining optimal stability of both calcipotriol and betamethasone dipropionate in a combination product. Early onset of action and efficacy of a fixed combination of calcipotriol and betamethasone dipropionate in this new vehicle in the treatment of psoriasis has been reported recently [92].

VI. VITAMIN D ANALOG THERAPY IN OTHER SKIN DISEASES A. Vitamin D and Ichthyosis A double-blind, bilaterally paired, comparative study showed the effectiveness of topical treatment with calcipotriol ointment on congenital ichthyoses [93]. Reduction in scaling and roughness on the calcipotrioltreated side was seen in all patients with lamellar ichthyosis and bullous ichthyotic erythroderma of Brocq. The only patient treated with Comel-Netherton syndrome showed mild improvement, while the only patient suffering from ichthyosis bullosa of Siemens that was treated with calcipotriol did not show any change in severity on the calcipotriol-treated as compared to the vehicle-treated side. Recently, it has been reported that topical tacalcitol therapy was ineffective against ichthyoses without epidermal hyperproliferation but with retentive hyperkeratosis, including X-linked ichthyosis (XLI), ichthyosis vulgaris (IV), and acquired ichthyosis [94].

B. Vitamin D Analogs and Scleroderma Previous findings point to the efficacy of vitamin D analogs for the treatment of scleroderma. Humbert et al. [95] reported that oral administration of 1.0–2.5 µg/d calcitriol improves skin involvement, probably via inhibition of fibroblast proliferation and dermal collagen deposition.

C. Vitamin D Analogs and Skin Cancer In vitro studies have demonstrated strong antiproliferative and prodifferentiating effects of vitamin D analogs in many VDR-expressing tumor cell lines, including malignant melanoma, squamous cell carcinoma, and leukemic cells [34,96]. In vivo studies showed that active vitamin D analogs reduced proliferation and tumor progression of epithelial tumors in rats [97], as well as other cancers [98,99] (see Section IX of this book). Little is known regarding the effects of calcitriol on the formation of metastases in patients with malignant melanoma or squamous cell carcinoma of the skin.

D. Vitamin D Analogs and Other Skin Diseases A number of case reports demonstrated positive effects of topical treatment with vitamin D analogs in a variety of skin diseases such as transient acantholytic

1800 dermatosis (Grovers disease), inflammatory linear verrucous epidermal naevus (ILVEN), disseminated superficial actinic parakeratosis, pityriasis rubra pilaris, epidermolytic palmoplantar keratoderma of Vorner, confluent and reticulated papillomatosis (Gougerot-Carteaud syndrome), and Sjögren-Larsson syndrome [100, rev. in 101]. These promising observations will have to be further evaluated in clinical trials.

VII. PERSPECTIVES FOR THE DEVELOPMENT OF NEW VITAMIN D ANALOGS WITH LESS CALCEMIC ACTIVITY FOR THE TREATMENT OF HYPERPROLIFERATIVE SKIN DISORDERS The use of vitamin D analogs in dermatology and other medical fields has been limited, since serious side effects, mainly on calcium metabolism, have occurred at the supraphysiological doses needed to achieve clinical efficacy. The evaluation of new vitamin D analogs with strong immunosuppressive, antiproliferative, and differentiating effects but only marginal effects on calcium metabolism will introduce important new therapies for the treatment of various skin and other diseases. The goal to create new vitamin D analogs with selective biological activity and minimal undesirable side effects has not yet been reached, but recent findings are promising. This subject is extensively discussed in Section VIII of this book. Calcipotriol, a vitamin D analog with similar VDR binding properties compared to calcitriol but low affinity for the vitamin D–binding protein (DBP), is effective, safe, and approved for the topical treatment of psoriasis [56,63]. In vivo studies in rats showed that effects of calcipotriol on calcium metabolism are 100–200-fold lower compared to calcitriol, while in vitro effects on proliferation and differentiation on human keratinocytes are comparable [102]. These differential effects are probably caused by the different pharmacokinetic profiles of calcipotriol and calcitriol (different affinity for DBP). Serum half-life in rats of these vitamin D analogs was shown to be 4 min after treatment with calcipotriol in contrast to 15 min after treatment with calcitriol [102]. However, one has to emphasize that the calcium studies comparing calcitriol and calcipotriol were done in vivo while most studies analyzing proliferation or differentiation were done in vitro. A different approach to create new vitamin D analogs that are effective in the topical treatment of hyperproliferative or inflammatory skin diseases is the goal to

JÖRG REICHRATH AND MICHAEL F. HOLICK

create new synthetic compounds with a high degree of dissociation that are metabolized in the skin and therefore exert only few systemic side effects. New analogs of vitamin D, obtained by a combination of the 20-methyl modification and removal of the C-19 methylene with side-chain modification including triple-bond at C-23 [103] or 2β-substituted 1,25(OH)2D3 [104], are promising candidates. Another interesting approach to enhance the concentration of calcitriol locally in the skin without obtaining systemic side effects are attempts to specifically inhibit the activity of vitamin D metabolizing enzymes, (i.e., 24-hydroxylase for calcitriol) that are present in the skin and responsible for the catabolism of calcitriol [105]. It is known that various active drugs, including other steroidal hormones and cytochrome P450 inhibitors such as ketoconazole, inhibit the activity of 24-hydroxylase in the skin [106]. It may be possible to enhance the concentration of endogenous calcitriol locally in the skin by the topical application of these compounds without obtaining systemic side effects. We speculate that the therapeutic effects of various antimycotic compounds including ketoconazole in the treatment of seborrheic dermatitis may at least in part be due to this mechanism. It is known that VDR requires several accessory proteins for efficient binding to vitamin D response elements in promoter regions of target genes and induction of transactivation [107]. As a consequence, different vitamin D analogs, depending on their chemical structure, may have different affinities for the various homo- or heterodimers of VDR and nuclear cofactors including RXR-α [108]. The synthesis of new vitamin D analogs that activate different vitamin D signaling pathways may lead to the introduction of new therapeutics for the topical or oral treatment of various skin diseases. These new drugs may induce strong effects on target cell proliferation and differentiation in the skin or the immune system, but only marginal effects on calcium metabolism. Another approach to enhance the therapeutic effects of orally or topically administered calcitriol may be the combination with synergistic drugs. The recent discovery of different vitamin D signaling pathways that are determined and regulated by cofactors of VDR including RXR-α and their corresponding ligands, suggests that 9-cis RA or all-trans RA may act synergistically with vitamin D analogs to induce VDR-mediated transactivation and regulate the transcriptional activity of distinct gene networks. Little is known about the effects of the combined application of vitamin D and vitamin A analogs under physiological conditions in vivo. This combination may selectively enhance or block different biological effects of vitamin D analogs

CHAPTER 101 Psoriasis and Other Skin Diseases

that are mediated by different vitamin D signaling pathways. In conclusion, it can be speculated that new vitamin D analogs will introduce new alternatives for the treatment of various skin disorders. If the final goal can be reached to create strong antiproliferative and antiinflammatory vitamin D analogs with only little calcemic activity, these new agents may herald a new era in dermatologic therapy, that possibly can be compared with the introduction of synthetic corticosteroids or retinoids. These new drugs, which activate selective vitamin D signaling pathways but exert only little calcemic activity, will also be effective in the systemic treatment of various malignancies and other diseases in addition to skin diseases. However, since skin diseases can also be treated by topical application of drugs, additional strategies are available to develop analogs with improved safety and efficacy for the treatment of psoriasis and other skin disorders.

References 1. Krafka J 2002 A simple treatment for psoriasis. J Lab Clin Med 1936;21:1147–1148. 2. Spier HW 1950 Vitamin-D2-Behandlung der Psoriasis. Therapeutische Ergebnisse und Untersuchungen über die Vitamin-D2-Wirkung auf den Blutspiegel von Cholesterin, Phosphor, Kalium und Glutathion bei Psoriasispatienten. Hautarzt 1:205–216. 3. Krueger G, Koo J, Lebwohl M, Menter A, Stern RS, Rolstad T 2001 The impact of psoriasis on quality of life: results of a 1998 National Psoriasis Foundation patientmembership survey. Arch Dermatol 137:280–284. 4. Christophers E, Henseler T 1985 Psoriasis of early and late onset: characterization of two types of psoriasis vulgaris. J Am Acad Dermatol 13:450 – 456. 5. Christophers E, Henseler T 1989 Patient subgroups and the inflammatory pattern in psoriasis. Acta Derm Venereol (Stockh) 69:88–92. 6. Elder JT, Henseler T, Christophers E, Voorhees JJ, Nair RP 1994 Of genes and antigens: the inheritance of psoriasis. J Invest Dermatol 103:150S–153S. 7. Valdimarsson H, Baker BS, Jonsdittir I, Fry L 1986 Psoriasis: a disease of abnormal keratinocyte proliferation induced by T lymphocytes. Immunol Today 7:256–259. 8. Nickoloff BJ, Schroder JM, von den Driesch P, Raychaudhuri SP, Farber EM, Boehncke WH, Morhenn VB, Rosenberg EW, Schon MP, Holick MF 2000 Is psoriasis a T-cell disease? Exp Dermatol 9(5):359–375. 9. Lee RE, Gaspari AA, Lotze MT, Chang AE, Rosenberg SA 1988 Interleukin 2 and psoriasis. Arch Dermatol 124: 1811–1815. 10. Barker JN, Jones ML, Mitra RS, Crockett Torab E, Fantone JC, Kunkel SL, Warren JS, Dixit VM, Nickoloff BJ 1991 Modulation of keratinocyte-derived interleukin-8, which is chemotactic for neutrophils and T lymphocytes. Am J Pathol 139:869–876. 11. Gottlieb AB 1990 Immunologic mechanisms in psoriasis. J Invest Dermatol 95:18S–19S.

1801 12. Schlaak JF, Buslau M, Jochum W, Hermann E, Girndt M, Gallati H, Meyer zum Büschenfelde KH, Fleischer B 1994 T-cells involved in psoriasis vulgaris belong to the Th1 subset. J Invest Dermatol 102:145–149. 13. van Reijsen FC, Druijnzeel-Koomen CAFM, Kalthoff FS, Maggi E, Romagnani S, Westland JKT, Mudde GC 1992 Skin-derived aeroallergen-specific T-cell clones of Th2 phenotype in patients with atopic dermatitis. J Allergy Clin Immunol 90:184–192. 14. Leung DY, Walsh P, Giorno R, Norris DA 1993 A potential role for superantigens in the pathogenesis of psoriasis. J Invest Dermatol 100:225–228. 15. Robert C, Kupper TS 1999 Inflammatory skin diseases, T-cells, and immune surveillance. N Engl J Med 341:1817–1828. 16. Chowaniec O, Jablonska S, Beutner EH, Proniewska M, Jarzabek Chorzelska M, Rzesa G 1981 Earliest clinical and histological changes in psoriasis. Dermatologica 163:42–51. 17. Holick MF, MacLaughlin JA, Clark MB, Holick SA, Potts JT, Anderson RR, Blank IH, Parrish JA, Elias P 1980 Photosynthesis of previtamin D3 in human skin and the physiological consequences. Science 210:203–205. 18. Holick MF, MacLaughlin JA, Anderson RR, Parrish J 1982 Photochemistry and photobiology of vitamin D. In: JD Regan, JA Parrish (eds). Photomedicine. Plenum Press: New York, pp. 195–218. 19. Holick MF, Smith E, Pincus S 1987 Skin as the site of vitamin D synthesis and target tissue for 1,25-dihydroxyvitamin D3. Arch Dermatol 123:1677–1682. 20. Holick MF 1991 Photobiology, physiology, and clinical applications for Vitamin D. In: LA Goldsmith (ed). Physiology, Biochemistry, and Molecular Biology of the Skin, 2nd ed. Oxford University Press: New York, pp. 928–956. 21. Bittiner B., Bleehen SS, Mac Neil S 1991 1α-25-(OH)2 Vitamin D3 increases intracellular calcium in human keratinocytes. Br J Dermatol 124:12230–12235. 22. MacLaughlin JA, Cantley LC, Holick MF 1990 1,25(OH)2D3 increases calcium and phosphatidylinositol metabolism in differentiating cultured human keratinocytes. J Nutr Biochem 1:81–87. 23. Haussler MR 1986 Vitamin D receptors: nature and function. Annu Rev Nutr 6:527–562. 24. Haussler MR, Mangelsdorf DJ, Komm BS, Terpening CM, Yamaoka K, Allegretto EA, Baker AR, Shine J, McDonnell DP, Hughes M, Weigel NL, O’Malley BW 1988 Molecular biology of the vitamin D hormone. Recent Prog Horm Res 44:263–305. 25. Baker AR, McDonnell DP, Hughes M, Crisp TM, Mangelsdorf DJ, Haussler MR, Pike JW, Shine J, O’Malley BW 1988 Cloning and expression of full-length cDNA encoding human vitamin D receptor. Proc Natl Acad Sci USA 85:3294–3298. 26. Yu VC, Deisert C, Andersen B, Holloway JM, Devary OV, Näär AM, Kim SY, Boutin JM, Glass CK, Rosenfeld MG 1991 RXRβ: a coregulator that enhances binding of retinoic acid, thyroid hormone, and vitamin D receptors to their cognate response elements. Cell 67:1251–1266. 27. Leid M, Kastner P, Lyons R, Nakshatri H, Saunders M, Zacharewski T, Chen J, Staub A, Garnier J, Mader S, Chambon P 1992 Purification, cloning, and RXR identity of the HeLa cell factor with which RAR or TR heterodimerizes to bind target sequences efficiently. Cell 68:377–395. 28. Milde P, Hauser U, Simon R, Mall G, Ernst V, Haussler MR, Frosch P, Rauterberg EW 1991 Expression of 1,25-dihydroxyvitamin D3 receptors in normal and psoriatic skin. J Invest Dermatol 97:230–239.

1802 29. Reichrath J, Münssinger T, Kerber A, Rochette-Egly C, Chambon P, Bahmer FA, Baum HP 1995 In situ detection of retinoid-X receptor expression in normal and psoriatic human skin. Br J Dermatol 133:168–175. 30. Smith EL, Walworth NC, Holick MF 1986 Effect of 1α-25dihydroxyvitamin D3 on the morphologic and biochemical differentiation of cultured human epidermal keratinocytes grown under serum-free conditions. J Invest Dermatol 86:709–714. 31. Hosomi J, Hosoi J, Abe E, Suda T, Kuroki T 1983 Regulation of terminal differentiation of cultured mouse epidermal cells by 1-alpha 25-dihydroxyvitamin D3. Endocrinol 113:1950–1957. 32. Gniadecki R, Serup J 1995 Stimulation of epidermal proliferation in mice with 1 alpha, 25-dihydroxyvitamin D3 and receptor-active 20-EPI analogs of 1 alpha, 25-dihydroxyvitamin D3. Biochem Pharmacol 49:621–624. 33. Rigby WFC 1988 The immunobiology of vitamin D. Immunol Today 9:54–58. 34. Texereau M, Viac J 1992 Vitamin D, immune system and skin. Europ J Dermatol 2:258–264. 35. Ranson M, Posen S, Mason RS 1988 Human melanocytes as a target tissue for hormones: in vitro studies with 1α,25dihydroxyvitamin D3, alpha-melanocyte stimulating hormone, and beta-estradiol. J Invest Dermatol 91:593–598. 36. Reichrath J, Müller SM, Kerber A, Baum HP, Bahmer FA 1997 Biologic effects of topical calcipotriol (MC 903) treatment in psoriatic skin. J Am Acad Dermatol 36:19–28. 37. Chen ML, Perez A, Sanan DK, Heinrich G, Chen TC, Holick MF 1996 Induction of vitamin D receptor mRNA expression in psoriatic plaques correlates with clinical response to 1,25-dihydroxyvitamin D3. J Invest Dermatol 106:637–641. 38. Missero C, Calautti E, Eckner R, Chin J, Tsai LH, Livingston DM, Dotto GP 1995 Involvement of the cell-cycle inhibitor Cip1/WAF1 and the E1A-associated p300 protein in terminal differentiation. Proc Natl Acad Sci USA 92:5451–5455. 39. Okita H, Ohtsuka T, Yamakage A, Yamazaki S 2002 Polymorphism of the vitamin D3 receptor in patients with psoriasis. Arch Dermatol Res 294(4):159–162. 40. Park BS, Park JS, Lee DY, Youn JI, Kim IG 1999 Vitamin D receptor polymorphism is associated with psoriasis. J Invest Dermatol 112(1):113–116. 41. Saeki H, Asano N, Tsunemi Y, Takekoshi T, Kishimoto M, Mitsui H, Tada Y, Torii H, Komine M, Asahina A, Tamaki K 2002 Polymorphisms of vitamin D receptor gene in Japanese patients with psoriasis vulgaris. J Dermatol Sci 30(2):167–171. 42. Lee DY, Park BS, Choi KH, Jeon JH, Cho KH, Song KY, Kim IG, Youn JI 2002 Vitamin D receptor genotypes are not associated with clinical response to calcipotriol in Korean psoriasis patients. Arch Dermatol Res 294(1–2):1–5. 43. Kontula K, Välimäki S, Kainulainen K, Viitanen AM, KeskiOja J 1997 Vitamin D receptor polymorphism and treatment of psoriasis with calcipotriol. Br J Dermatol 136:147–148. 44. Mee JB, Cork MJ 1998 Vitamin D receptor polymorphism and calcipotriol response in patients with psoriasis. J Invest Dermatol 110:301–302. 45. Hutchinson PE, Osborne JE, Lear JT, Smith AG, Bowers PW, Moeeis PN, Jones PW, York C, Strange RC, Fryer AA 2000 Vitamin D receptor polymorphisms are associated with altered prognosis in patients with malignant melanoma. Clin Cancer Res 6:498–504. 46. Colin EM, Weel AEAM, Uitterlinden AG, Buurman CJ, Birkenhäger JC, Pols HAP, van Leeuwen JPTM 2000 Consequences of vitamin D receptor gene polymorphisms for growth inhibition of cultured human peripheral blood

JÖRG REICHRATH AND MICHAEL F. HOLICK

47. 48. 49.

50. 51.

52. 53. 54.

55.

56. 57. 58.

59.

60.

61.

62.

63.

mononuclear cells by 1,25-dihydroxyvitamin D3. Clin Endocrinol 52:211–216. Staberg B, Oxholm A, Klemp P, Christiansen C 1987 Abnormal vitamin D metabolism in patients with psoriasis. Acta Derm Venereol 67(1):65–68. Perez A, Chen TC, Turner A, Holick MF 1995 Pilot study of topical calcitriol (1,25-dihydroxyvitamin D3) for treating psoriasis in children. Arch Dermatol 131:961–962. Stewart AF, Battaglini-Sabetta J, Millstone L 1984 Hypocalcemia-induced pustular psoriasis of von Zumbusch. New experience with an old syndrome. Ann Intern Med 100(5):677–680. Stone OJ 1985 Chloroquine, ground substance, aggravation of psoriasis. Int J Dermatol 24(8):539. MacLaughlin JA, Gange W, Taylor D, Smith E, Holick MF 1985 Cultured psoriatic fibroblasts from involved and uninvolved sites have partial but not absolute resistance to the proliferation-inhibition activity of 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 82:5409–5412. Morimoto S, Kumahara Y 1985 A patient with psoriasis cured by 1α-hydroxyvitamin D3. Med J Osaka Univ 35:3–4, 51–54. Morimoto S, Yochikawa K, Kozuka T, Kitano Y, Imawaka S, Fukuo K, Koh E, Kumahara Y 1986 An open study of vitamin D3 treatment in psoriasis vulgaris. Br J Dermatol 115:421–429. Chen TC, Holick MF 2000 Hexafluoro-1,25-dihydroxyvitamin D3 has markedly increased potency in inhibiting proliferation of cultured human keratinocytes compared with 1,25-dihydroxyvitamin D3. Br J Dermatol 144:72–78. Perez A, Chen TC, Turner A, Raab R, Bhawan J, Poche P, Holick MF 1996 Efficacy and safety of topical calcitriol (1,25-dihydroxyvitamin D3) for the treatment of psoriasis. Br J Dermatol 134:238–246. Kragballe K, Beck HI, Sogaard H 1988 Improvement of psoriasis by topical vitamin D3 analog (MC 903) in a double-blind study. Br J Dermatol 119:223–230. van de Kerkhof PCM, van Bokhoven M, Zultak M, Czarnetzki BM 1989 A double-blind study of topical 1α-25dihydroxyvitamin D3 in psoriasis. Br J Dermatol 120:661–664. Barker JN, Ashton RE, Marks R, Harris RI, Berth-Jones J 1999 Topical maxacalcitol for the treatment of psoriasis vulgaris: a placebo-controlled, double-blind, dose-finding study with active comparator. Br J Dermatol 141(2):274–278. Durakovic C, Malabanan A, Holick MF 2001 Rationale for use and clinical responsiveness of hexafluoro-1,25-dihydroxyvitamin D3 for the treatment of plaque psoriasis: a pilot study. Br J Dermatol 144(3):500–506. Holick MF, Chen ML, Kong XF, Sanan DK 1996 Clinical uses for calciotropic hormones 1,25-dihydroxyvitamin D3 and parathyroid hormone-related peptide in dermatology: a new perspective. J Invest Dermatol (Symp Proc) 1:1–9. Miyachi Y, Ohkawara A, Ohkido M, Harada S, Tamaki K, Nakagawa H, Hori Y, Nishiyama S 2002 Long-term safety and efficacy of high-concentration (20 microg/g) tacalcitol ointment in psoriasis vulgaris. Eur J Dermatol 12(5):463–468. van de Kerkhof PC, Berth-Jones J, Griffiths CE, Harrison PV, Honigsmann H, Marks R, Roelandts R, Schopf E, Trompke C 2002 Long-term efficacy and safety of tacalcitol ointment in patients with chronic plaque psoriasis. Br J Dermatol 146(3):414–422. Kragballe K, Gjertsen BT, de Hoop D, Karlsmark T, van de Kerhof PCM, Larko O, Nieboer C, Roed-Petersen J, Strand A, Tikjob B 1991 Double-blind right/left comparison of calcipotriol and betamethasone valerate in treatment of psoriasis vulgaris. Lancet 337:193–196.

CHAPTER 101 Psoriasis and Other Skin Diseases

64. Serup J 1994 Calcipotriol irritation: Mechanism, diagnosis, and clinical implication. Acta Derm Venereol (Stockh) Abstract 186:42S. 65. Fisher DA 1997 Allergic contact dermatitis to propylene glycol in calcipotriene ointment. Cutis 60:43–44. 66. Krayenbuhl BH, Elsner P 1999 Allergic and irritant contact dermatitis to calcipotriol. Am J Contact Dermat 10(2):78–80. 67. Park YK, Lee JH, Chung WG 2002 Allergic contact dermatitis from calcipotriol. Acta Derm Venereol 82(1):71–72. 68. Katayama I, Ohkawara A, Ohkido M, Harada S, Tamaki K, Nakagawa H, Hori Y, Nishiyama S 2002 High-concentration (20 µg/g) tacalcitol ointment therapy on refractory psoriasis vulgaris with low response to topical corticosteroids. Eur J Dermatol 12(6):553–557. 69. Queille-Roussel C, Duteil L, Parneix-Spake A, Arsonnaud S, Rizova E 2001 The safety of calcitriol 3 microg/g ointment. Evaluation of cutaneous contact sensitization, cumulative irritancy, photoallergic contact sensitization, and phototoxicity. Eur J Dermatol 11(3):219–224. 70. Perez A, Raab R, Chen TC, Turner A, Holick MF 1996 Safety and efficacy of oral calcitriol (1,25-dihydroxyvitamin D3) for the treatment of psoriasis. Br J Dermatol 134:1070–1078. 71. Green C, Ganpule M, Harris D, Kavanagh G, Kennedy C, Mallett, R, Rustin M, Downes N 1994 Comparative effects of calcipotriol (MC 903) solution and placebo (vehicle of MC 903) in the treatment of psoriasis of the scalp. Br J Dermatol 130:483–487. 72. Koo J 2002 Vitamin D and scalp psoriasis. Cutis 70:21–24. 73. van de Kerkhof PC, Franssen ME 2001 Psoriasis of the scalp. Diagnosis and management. Am J Clin Dermatol 2(3):159–165. 74. Petrow W 1995 Treatment of a nail psoriasis with calcipotriol. Akt Dermatol 21:396–400. 75. Ortonne JP, Humbert P, Nicolas JF, Tsankov N, Tonev SD, Janin A, Czernielewski J, Lahfa M, Dubertret L 2003 Intraindividual comparison of the cutaneous safety and efficacy of calcitriol 3 µg/g ointment and calcipotriol 50 µg/g ointment on chronic plaque psoriasis localized in facial, hairline, retroauricular, or flexural areas. Br J Dermatol 148(2):326–333. 76. Saggese G, Federico G, Battini R 1993 Topical application of 1,25 dihydroxyvitamin D3 (calcitriol) is an effective and reliable therapy to cure skin lesions in psoriatic children. Eur J Pediatr 152:389–392. 77. Perez A, Chen TC, Turner A, Holick MF 1995 Pilot study of topical calcitriol (1,25-dihydroxyvitamin D3) for treating psoriasis in children. Arch Dermatol 131:961–962. 78. Travis LB, Silverberg NB 2001 Psoriasis in infancy: therapy with calcipotriene ointment. Cutis 68:341–344. 79. Grossman RM, Thivolet J, Claudy A, Souteyrand P, Guilhou JJ, Thomas P, Amblard P, Belaich S, de Belilovsky C, de la Brassinne M, et al. 1994 A novel therapeutic approach to psoriasis with combination calcipotriol ointment and very low-dose cyclosporine: a result of a multicenter placebocontrolled study. J Am Acad Dermatol 31:68–74. 80. Kerscher M, Volkenandt M, Plewig G, Lehmann P 1993 Combination phototherapy of psoriasis with calcipotriol and narrow band UVB. Lancet 342:923. 81. Cambazard, van de Kerkhof PCM, Hutchinson PE, and the Calcipotriol Study Group. Proceedings of the 3rd International Calcipotriol Symposium, Munich Germany, 23 March 1996. 82. Ortonne JP 1994 Calcipotriol in combination with bethamethasone diproprionate. Nouv Dermatol 13:736–751. 83. Kragballe K 1990 Combination of topical calcipotriol (MC 903) and UVB radiation for psoriasis vulgaris. Dermatologica 181:211–214.

1803 84. Mascaro JM 2002 Vitamin D and psoralen plus UVA radiation. Cutis 70:13–15. 85. Monastirli A, Georgiou S, Pasmatzi E, Sakkis T, Badavanis G, Drainas D, Sagriotis A, Tsambaos D 2002 Calcipotriol plus short-contact dithranol: a novel topical combination therapy for chronic plaque psoriasis. Skin Pharmacol Appl Skin Physiol 15(4):246–251. 86. van de Kerkhof P 2002 Vitamin D and systemic therapy. Cutis 70:16–20. 87. de Jong EM, Mork NJ, Seijger MM, De La Brassine M, Lauharanta J, Jansen CT, Guilhou JJ, Guillot B, Ostrojic A, Souteyrand P, Vaillant L, Barnes L, Rogers S, Klaber MR, Van De Kerkhof PC 2003 The combination of calcipotriol and methotrexate compared with methotrexate and vehicle in psoriasis: results of a multicenter placebo-controlled randomized trial. Br J Dermatol 148(2):318–325. 88. Kragballe K 2002 Vitamin D and UVB radiation therapy. Cutis 70:9–12. 89. Messer G, Degitz K, Plewig G, Rocken M 2001 Pretreatment of psoriasis with the vitamin D3 derivative tacalcitol increases the responsiveness to 311-nm ultraviolet B: results of a controlled, right/left study. Br J Dermatol 144(3):628–629. 90. Tzaneva S, Honigsmann H, Tanew A, Seeber A 2002 A comparison of psoralen plus ultraviolet A (PUVA) monotherapy, tacalcitol plus PUVA, and tazarotene plus PUVA in patients with chronic plaque-type psoriasis. Br J Dermatol 147(4): 748–753. 91. Glaser R, Rowert J, Mrowietz U 1998 Hyperpigmentation due to topical calcipotriol and photochemotherapy in two psoriatic patients. Br J Dermatol 139(1):148–151. 92. Papp KA, Guenther L, Boyden B, Larsen FG, Harvima RJ, Guilhou JJ, Kaufmann R, Rogers S, van de Kerkhof PC, Hanssen LI, Tegner E, Burg G, Talbot D, Chu A 2003 Early onset of action and efficacy of a combination of calcipotriene and betamethasone dipropionate in the treatment of psoriasis. J Am Acad Dermatol 48(1):48–54. 93. Lucker GP, van de Kerkhof PC, van Dijk MR, Steijlen PM 1994 Effect of topical calcipotriol on congenital ichthyosis. Br J Dermatol 131:546–550. 94. Okano M 2001 Assessment of the clinical effect of topical tacalcitol on ichthyoses with retentive hyperkeratosis. Dermatology 202(2):116–118. 95. Humbert P, Dupond JL, Agache P, Laurent R, Rochefort A, Drobacheff C, de Wazieres B, Aubin F 1993 Treatment of scleroderma with oral 1,25-dihydroxyvitamin D3: evaluation of skin involvement using non-invasive techniques. Results of an open prospective trial. Acta Derm Venereol (Stockh) 73:449–451. 96. Koeffler HP, Hirji K, Itri L 1985 1,25-dihydroxyvitamin D3: in vivo and in vitro effects on human preleukemic and leukemic cells. Cancer Treat Rep 69:1399–1407. 97. Colston KW, Chander SK, Mackay AG, Coombes RC 1992 Effects of synthetic vitamin D analogs on breast cancer cell proliferation in vivo and in vitro. Biochem Pharmacol 44:693–702. 98. Eisman JA, Barkla DH, Tutton PJM 1987 Suppression of in vivo growth of human cancer solid tumor xenografts by 1α,25-dihydroxyvitamin D3. Cancer Res 47:21–25. 99. Chen TC, Schwartz GG, Burnstein KL, Lokeshwar BL, Holick MF 2000 The in vitro evaluation of 25-hydroxyvitamin D3 and 19-nor-1alpha,25-dihydroxyvitamin D2 as therapeutic agents for prostate cancer. Clin Cancer Res 6(3):901–908. 100. Carrozzo AM, Gatti S, Ferranti G, Primavera G, Vidolin AP, Nini G 2000 Calcipotriol treatment of confluent and reticulated papillomatosis (Gougerot-Carteau syndrome). JEADV 14:131–133.

1804 101. Reichrath J, Holick MF 1999 Clinical utility of 1,25dihydroxyvitamin D3 and its analogs for the treatment of psoriasis and other skin diseases. In: MF Holick (ed). Vitamin D. Physiology, Molecular Biology and Clinical Applications. Humana Press: Totowa, New Jersey, pp. 357–374. 102. Binderup L, Latini S, Binderup E, Bretting C, Calverley M, Hansen K 1991 20-epi-vitamin D3 analogs: a novel class of potent regulators of cell growth and immune response. Biochem Pharmacol 42:1569–1575. 103. Neef G, Kirsch G, Schwarz K, Wiesinger H, Menrad A, Fähnrich M, Thieroff-Eckerdt R, Steinmeyer A 1994 20-methyl vitamin D analogs. In: AW Norman, R Bouillon, M Thomasset (eds). Vitamin D. A pluripotent steroid hormone: structural studies, molecular endocrinology, and clinical applications. Walter de Gruyter: Berlin, pp. 97–98. 104. Schönecker B, Reichenbächer M, Gliesing S, Prousa R, Wittmann S, Breiter S, Thieroff-Eckerdt R, Wiesinger H, Haberey M, Scheddin D, Mayer H 1994 2β-substituted calcitriols and other A-ring substituted analogs—synthesis and biological results. In: AW Norman, R Bouillon, M Thomasset (eds). Vitamin D. A pluripotent steroid

JÖRG REICHRATH AND MICHAEL F. HOLICK

105.

106.

107. 108.

hormone: structural studies, molecular endocrinology, and clinical applications. Walter de Gruyter: Berlin, pp. 99–100. Schuster I, Herzig G, Vorisek G 1994 Steroidal hormones as modulators of vitamin D metabolism in human keratinocytes. In: AW Norman, R Bouillon, M Thomasset (eds). Vitamin D. A pluripotent steroid hormone: structural studies, molecular endocrinology, and clinical applications. Walter de Gruyter Berlin, 184–185. Zhao J, Marcelis S, Tan BK, Verstuaf A, Bouillon R 1994 Potentialization of vitamin D (analogs) by cytochrome P-450 enzyme inhibitors is analog- and cell-type sepcific. In: AW Norman, R Bouillon, M Thomasset (eds). Vitamin D. A pluripotent steroid hormone: structural studies, molecular endocrinology and clinical applications. Walter de Gruyter Berlin, pp. 97–98. Carlberg C, Bendik I, Wyss A, Meier E, Sturzenbecker LJ, Grippo JF, Hunziker W 1993 Two nuclear signaling pathways for vitamin D. Nature 361:657–660. Schräder M, Müller KM, Becker-Andre M, Carlberg C 1994 Response element selectivity for heterodimerization of vitamin D receptors with retinoic acid and retinoid X receptors. J Mol Endocrinol 12:327–333.

CHAPTER 102

Muscles and Falls HENNING GLERUP ERIK FINK ERIKSEN

University Department of Gastroenterology and Hepatology, Aarhus University Hospital, DK-8000 Aarhus C, Denmark Lilly Research Laboratories, Lilly Corporate Center, Indianapolis, Indiana

I. II. III. IV.

Introduction Hypovitaminosis D Myopathy (HDM): Symptoms and Signs Muscle Physiology in Relation to HDM Experimental Studies on the Effects of Vitamin D on Striated Muscle V. Clinical Studies on Hypovitaminosis D Myopathy

I. INTRODUCTION For many years, myopathy has been considered part of the osteomalacic symptom complex [1–6], occasionally seen in patients suffering from osteomalacic bone disease caused by severe vitamin D deficiency. In agreement with this notion, the first term suggested for vitamin D deficiency-related myopathy was “osteomalacic myopathy” [5,7]. Recently, however, myopathy has been shown to be a prominent, and common symptom of vitamin D deficiency. Furthermore, new data also indicate that severely impaired muscle function may be present, even before biochemical signs of bone disease develops [8]. Myopathy has been demonstrated in vitamin D deficient “fallers,” even without signs of secondary hyperparathyroidism [9]. Consequently, we suggest the use of the term “hypovitaminosis D myopathy” (HDM) for this condition. The interest in HDM has varied over the years. In the seventies, several publications focused on myopathy in relation to vitamin D deficiency [6,10–17], whereas research on HDM was limited to experimental studies of nongenomic effects of 1,25(OH)2D3 in the eighties and early nineties. These studies are reviewed elsewhere in this book (Chapter 55). In 1992 Chapuy et al. [18] presented a 43% reduction in hip fracture rate in elderly women treated with 800 IU cholecalciferol and 1200 mg calcium for 18 months (see Chapter 66). The rapid reduction in hip fractures demonstrated in this study was difficult to explain by vitamin D-induced improvements in bone mass only, and thus suggested that improvement of muscle tone and coordination after vitamin D might play a role. Ever since this pivotal study, the interest in clinical studies of the muscular effects of vitamin D VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

VI. Clinical Studies on the Risk of Falls and Vitamin D Status VII. Is HDM Caused by Low Levels of 25OHD, 1,25(OH)2D or Elevated PTH? VIII. Other Possible Muscular Effects of Vitamin D IX. Summary References

has been growing. Several studies have demonstrated, that large proportions of hip fracture patients are vitamin D deficient [19–21], and these findings led several groups to suggest that the epidemic of hip fractures among the elderly could at least partly be explained by falls secondary to vitamin D deficiency-related myopathy. Consequently, hip fractures might be prevented by vitamin D supplementation. In support of this hypothesis, Bischoff et al. [9] recently reported a 49% reduction of falls among elderly treated with 800 IU cholecalciferol and 1200 mg calcium for 3 months.

II. HYPOVITAMINOSIS D MYOPATHY (HDM): SYMPTOMS AND SIGNS HDM is often overlooked and misdiagnosed [22]. In this context, it is important to understand that the decrease in muscle strength is a continuous and gradual process, whereas the loss of functional ability is quantal. This means that the patients can have a considerable loss of muscle strength before they complain of muscle weakness. Their main complaint would most likely be fatigue. The patients do not complain of muscle weakness until they are unable to walk or experience difficulties rising from a squatting position unaided. Secondly, many of the symptoms of HDM are rather diffuse (pain, fatigue, weakness, paresthesias, etc.) and thus often lead to alternative diagnoses such as rheumatic disease, polymyalgia, psychoneurotic disorders, fibromyalgia, malignant diseases, etc. In a recent attempt to evaluate HDM-related symptoms among 65 patients with different degrees of vitamin D deficiency and 22 controls, a graded scale questionnaire was used [23,24]. The most prominent Copyright © 2005, Elsevier, Inc. All rights reserved.

1806

HENNING GLERUP AND ERIK FINK ERIKSEN

symptoms were diffuse muscle pain of the proximal muscles of the lower limb, proximal arm muscles and shoulder muscles, followed by deep bone pain, fatigue, difficulty in ascending a staircase/carrying loads, peripheral paresthesias, muscle cramps, and joint pain [5,12,22]. The symptoms all correlated significantly to serum levels of 25OHD, whereas no significant correlation to 1,25(OH)2D levels could be detected [23]. The waddling gait classically associated with HDM [2,4,5] is only seen in severe cases of vitamin D deficiency. The typical “faller” or hip fracture patient will often present muscular symptoms, but most clinicians will attribute such symptoms to aging. However, if patients are asked specifically for symptoms of HDM, they will usually present them. Moreover, muscle strength measured by a dynamometer will often be reduced. HDM typically affects proximal muscle groups, particularly the weight-bearing antigravity muscles of the lower limb [8]. As mentioned above, the symptoms of HDM are rather nonspecific. Thus, assessment of serum 25OHD and PTH is obligatory when the diagnosis HDM

is suspected. Presence of secondary hyperparathyroidism or low levels of 25OHD will strongly support the diagnosis of HDM. Measurement of serum calcium, alkaline phosphatase, and 1,25(OH)2D are of limited value in the diagnosis of HDM, as values can be normal even when severe HDM symptoms are present [8].

III. MUSCLE PHYSIOLOGY IN RELATION TO HDM A detailed review of muscle ultra structure and physiology is beyond the scope of this chapter. However, in order to understand the effects of vitamin D on striated muscle, a short introduction is warranted. Striated muscle cells consist of bundles of myofibrils enveloped in the cell membrane/sarcolemma (see Fig. 1). The myofibrils are surrounded by the sarcoplasmatic reticulum (SR), the calcium reservoir of the cell. The diameter of muscle fibers varies (10–100 µm) [25,26]. From the sarcolemma, tubular invaginations (the T-tubules) penetrate deeply into the cell making

A 3

2

2

6

4 1 3

2

Triad

4

4

5

FIGURE 1 Skeletal muscle. A single muscle fiber surrounded by its sarcolemma (5). The sarcoplasmatic reticulum (2) with its terminal cisternae (3) surrounds each myofibril (1) which is a build up of thick and thin filaments. The T-tubules (4) are invaginations of the cell membrane (sarcolemma) penetrating deep into the muscle fiber. The arrays of mitochondria are situated among the myofibrils (6). Reproduced by permission from Krstic RV: Ultrastructure of the mammalian cell. SpringerVerlag, 1979.

1807

CHAPTER 102 Muscles and Falls

a network around the myofibrils, where they are located very near to the terminal cisternae of SR in the so called “triads.” The network of the T-tubules and the SR are so extensive that a 100 µm diameter fiber with a cell surface of 1 cm2 has a T-tubular surface of 7 cm2 and a SR surface of 135 cm2 [27]. Calcium release from the SR will always take place within 1 µm distance from the binding site on troponin C [28]. The myofibrils consist of regularly arranged myosin A Myosin

Cross-bridge Troponin Tropomyosin

I

C T

Actin

and actin filaments, giving the muscle the characteristic striated appearance under the microscope. A myofibril comprises 100–400 filaments and has a diameter of 1 µm [25]. The filaments consist of three components: actin, tropomyosin, and troponin (see Fig. 2). The myosin molecules of the thick filament are arranged in a helix with myosin heads projecting in a 45° angle from the longitudinal direction of the filament [25–27,29]. The thin filaments possess a binding site for the myosin head, which under relaxation is covered by troponin. Troponin is made up of three subunits: troponin I (binds to actin), troponin T (binds to tropomyosin), and troponin C (has strong affinity for Ca++). A muscle contraction is initiated by a depolarization reaching the muscle from the motor-end-plate via the T-tubules. The depolarization results in liberation of calcium from the terminal cisternae of the SR through activation of voltage-dependent calcium release channels. As the Ca++ concentration rises, Ca++ will bind to troponin C and this results in a conformational change of the actin filament, which allows the myosin head to bind to the binding site (see Fig. 2). In the presence of Mg++ and under consumption of ATP the myosin head bends 45° and subsequently moves along the actin filament (10 nm per bend) [29]. Relaxation occurs as Ca++ is lowered again (see Fig. 3).

A. Calcium Homeostasis in Muscle Cells

B

ADP + pi

Ca2+ Ca2+ ATP

Ca2+

I T

C

Ca2+

FIGURE 2 Myosin heads extending from the thick filaments reach towards binding sites on the thin actin filaments. In relaxation the binding sites are covered by troponin. With binding of Ca++ to troponin C, a conformation change occurs allowing the myosin head to bind to its actin binding site, ATP is consumed and contraction is initiated. Modified by permission from N Engl J Med 1975;293:1184.

As the calcium ion is essential for muscle function, maintenance of calcium homeostasis is a highly prioritized process in the muscle cell. The Ca++ concentration in the cytosol of the resting muscle is very low (10−7–10−8 mol/l), 10,000 times lower than extracellular levels (2 × 10−3 mol/l). An energy dependent pump, in the form of Ca-ATPase, maintains this gradient. This pump is localized both in the outer cell membrane (sarcolemma) pumping Ca++ out of the cell, as well as in the sarcoplasmatic membrane pumping Ca++ into the sarcoplasmatic reticulum (SR). The content of Ca-ATPase of the SR is much higher than in the sarcolemma, which makes SR the most important organelle for regulation of intracellular calcium homeostasis. In the SR, a calcium-binding protein (calsequestrin) keeps the concentration of free Ca++ at a level of 0.5 × 10–3 mol/l. Calsequestrin has the capacity of binding 43 mol Ca++ per mol protein. If all Ca++ were free, the concentration in SR could be as high as 20 × 10–3 mol/l [29]. The calsequestrin is bound to the calcium release channels of the SR, and may be important in the regulation of Ca++ release from the SR [30,31].

1808

HENNING GLERUP AND ERIK FINK ERIKSEN

Z line At rest

B. Energy Supply

A band Sarcoplasmic reticulum

During muscle contraction, the Ca++ level in the cytosol can increase to 2 × 10−4 mol/l. Relaxation of the muscle occurs when Ca++ has been pumped back into the SR. If stimulation stops, cytosolic calcium levels can be normalized within 1/20 second. Thus Ca-ATPase plays a pivotal role in the regulation of calcium homeostasis and function of muscle cells.

Transverse tubule

Extracellular space

Contraction

Muscles demand a steady energy supply as ATP is consumed both during muscle contraction and muscle relaxation. During muscle contraction, the available pool of ATP in the cell is consumed within a few seconds. New ATP is produced by oxidative phosphorylation of glucose, fatty acid, and amino acids through the respiratory chain, and further degradation to CO2 and H2O through the O2-demanding citric acid circle. Muscles also have the capacity of glycogenesis as storage of energy for later use. Under anaerobic conditions, glucose can undergo anaerobic glycolysis leading to the formation of lactic acid and ATP. For periods of high energy demand, ATP is stored as phosphocreatine, which rapidly yields ATP via the reaction: phosphocreatine + ADP ← → creatine + ATP.

C. Fiber Types

Relaxation

FIGURE 3 Muscle contraction—relaxation. During rest, calcium is stored in the terminal cisternae of the sarcoplasmatic reticulum. The action potential spreads via the T-tubules and releases calcium and thus initiates the contraction. Calcium is pumped into the sarcoplasmatic reticulum, and the muscle relaxes. Modified with permission from N Engl J Med 1971;285:31.

Based on histochemical staining, muscle fibers are divided into three types: I, type IIa, and type IIb. The three fiber types differ in their energy metabolism. Type I fibers possess the capacity for oxidative metabolism making them more fatigue resistant. Type IIb fibers are solely dependent on fast energy supply from ATP, creatinephosphate, and from glycolysis of glucose to lactic acid. Type IIa fibers have both oxidative and glycolytic capacity, making them more fatigue-resistant than type IIb. Type I fibers are smaller and weaker than type II; they are more easily excited and are recruited before the type II fibers. Type II fibers are innervated by big motor-neurones and are big fibers suitable for precise and powerful movements, whereas type I are more suitable for maintaining a basal muscle tonus [29,32]. In agreement with type II fibers being the more powerful, the terminal cisternae of type II fibers contain up to 7 times more Ca-ATPase than the type I fibers [33,34]. The fiber composition differs between the different muscle groups of the body. A distal muscle, such as soleus, is mainly built of type I fibers, whereas a proximal muscle like quadriceps has a high percentage of type II fibers [25].

1809

CHAPTER 102 Muscles and Falls

B. Animal and Cell Culture Studies

IV. EXPERIMENTAL STUDIES ON THE EFFECTS OF VITAMIN D ON STRIATED MUSCLE A. Genomic and Nongenomic Effects of Vitamin D on Striated Muscle Striated muscle contain vitamin D receptors (VDR) [35–38], and vitamin D has been shown to stimulate the synthesis of several important muscle proteins (troponin C, actin) in the sarcoplasmatic reticulum and the inner membrane of mitochondria [17,39,40]. In addition to the classical genomic effects of vitamin D, a number of nongenomic effects have been identified [41–43]. The nongenomic effects are mediated through binding of vitamin D to membrane receptors resulting in formation of intracellular second messengers, which elicit rapid regulation intracellular enzyme activity and ion-pump systems. These effects are extensively reviewed elsewhere in this book (Chapter 55).

Single contraction

The importance of vitamin D for muscle function was firmly established, when Curry et al. [44] published their study in Nature in 1974. They showed that the ATP-dependent Ca++ uptake of isolated vesicles of SR was reduced in muscle from vitamin D–depleted rabbits, when compared to vitamin D-repleted animals. They further reported that the administration of three oral doses of vitamin D3 (100 IU) to the vitamin D– depleted animals normalized the Ca++-uptake. A few years later, [14] Rodman and Baker [45] published another important study. They used an in vivo model of vitamin D–depleted and –repleted rats for electrophysiological investigations of the muscle kinetics of the soleus muscle (see Fig. 4). In this model, vitamin D–depleted animals exhibited prolonged timeto-peak contractile tensions and prolonged relaxation half-lives. Again, this could be normalized by a few days of vitamin D treatment prior to muscle testing. Moreover, they demonstrated that the phenomenon was independent of serum levels of Ca++ and phosphate.

Tetanus

5 msec

10 msec

Control

TP 15 msec T1/2 14 msec

T1/2 32 msec

5 msec

10 msec

No vit D High Ca++ and PO4−

TP 21 msec T1/2 21 msec

FIGURE 4

T1/2 53 msec

Contraction and relaxation response in vitamin D–depleted and –repleted rats. Reproduced by permission from Rodman and Baker [45], Kidney International 1978;13:189–193.

1810

HENNING GLERUP AND ERIK FINK ERIKSEN

Birge and Haddad [16] studied the effect of 25OHD on muscle in an in vitro model. They found that 25OHD increased the intracellular content of ATP and phosphate and increased protein synthesis. Vitamin D3 and 1,25(OH)2D were found to have no effects on these mechanisms. The active transport of Ca++ into SR by Ca-ATPase is 1,25(OH)2D–dependent [46–48]. The activity of the pump seems to be regulated by 1,25(OH)2D–stimulated phosphorylation of proteins in the SR-membrane [41]. Furthermore, stereological studies [49] have shown that vitamin D–deficient rats exhibit reduced SR volume and increased mitochondria volume. These changes are also reversed by treatment with 1,25(OH)2D. The myofibrillar content of actin and troponin C is reduced in vitamin D–depleted animals [17,39] and normalized by vitamin D treatment. This normalization is most likely an action of 25OHD, rather than 1,25(OH)2D [17].

In two studies [39,50], the content of Ca-ATPase in vitamin D–depleted animals was investigated. It was found to be normal and unchanged by vitamin D treatment. Also the histochemical and electron microscopic appearance of muscles from vitamin D–depleted animals in the two studies was completely normal. These studies, however, do not rule out an effect of vitamin D on Ca-ATPase synthesis. Type II fibers, usually described as being atrophic in vitamin D–depleted individuals, show a much higher content of Ca-ATPase than other fiber types. Vitamin D3 supplementation of vitamin D–depleted chicks enhances [3H]-leucine labeling of proteins of the inner mitochondrial membrane by 50%, of sarcoplasmatic proteins by 20%, and contractile proteins by 10% [40]. These results indicate that vitamin D exerts important effects on the protein synthesis in muscle. The pronounced effect on the mitochondrial proteins suggests a regulatory effect on energy metabolism.

Ultrastructure of skeletal muscle vitamin D-repleted and vitamin D-deprived rats

B

Deprived TS

TS

M

A

Repleted

T

z

M TS

z

M G

T

M

G TS

T

M

M M

M

Z SR M

G

M

G

M

T 2b

TS

z

T

z

2a

Exchangeable calcium stained by potassium-pyroantimonate in vitamin D-repleted and vitamin D-deprived rats.

C

D

Repleted

Deprived My

TC

My Z

Mi My Z

Z

TS Z

Z

TS

TS TS

Mi Mi TC

TS

Mi

TS

Mi

My Z

3a

FIGURE 5

3b

Z

Ultrastructure of skeletal muscle in vitamin D–repleted (upper left (a)) and vitamin D–depleted (upper right (b)) rats. Notice the disturbed architecture of the muscle in vitamin D–depleted rats. Some of the sarcomeres fail to relax (arrow heads). Z: Z-line, SR: Sarcoplasmic Reticulum, M: Mitochondria, G: Glycogen, T: Triad, TS: T-System. In the lower part of the figure, the functional pool of protein bound exchangeable calcium has been visualized by potassium-pyroantimonate staining. In vitamin D–repleted rats (lower left (c)) high concentration of calcium appears at the T-system (TS) and the terminal cisternae (TC) of the sarcoplasmic reticulum. Irregular dispersed calcium is seen along the myofibrils (My). The calcium staining is less intensive and more irregular in vitamin D–depleted rats (lower right (d)). Reproduced from Toury et al. 1990 Biology of the Cell. 69:179–189 with permission from Elsevier.

1811

CHAPTER 102 Muscles and Falls

Toury et al. [51] investigated the calcium content in subcellular fractions of vitamin D–depleted and –repleted rats. In all subcellular fractions except the cytosol, they found lower Ca++ contents in vitamin D– depleted compared to vitamin D–repleted animals. In the cytosol, the calcium content was higher in vitamin D– depleted animals compared to vitamin D–repleted, indicating lower activity of the Ca-ATPase. In electron microscopic studies of muscle biopsies, exchangeable calcium was visualized with pyroantimonateosmium (see Fig. 5). Vitamin D–replete animals revealed much more intensive staining of the areas around the terminal cisternae of the SR, the T-tubules, and at the active sites of the myofibrils. Thus, the net effect of vitamin D on intracellular calcium homeostasis seems to be augmentation of intracellular calcium fluxes and an increased calcium pool available for muscle contraction.

V. CLINICAL STUDIES ON HYPOVITAMINOSIS D MYOPATHY A. Biopsy Studies Muscle biopsies obtained in severely vitamin D– deficient patients revealed selective atrophy of type II muscle fibers with enlarged interfibrillar spaces and infiltration of fat, increased fibrosis, and glycogen granules [6,11–15,52]. Mitochondria were irregular in size and shape. These findings are not specific for HDM, but can be seen in a number of other endocrine myopathies [53]. However, in neuropathic atrophy, both type I and type II fibers are affected [25]. In immobilization atrophy, type I fibers are primarily affected [25]. Sato et al. [54] recently investigated muscle biopsies from hip fracture patients. The patients were divided into vitamin D–sufficient (25OHD > 39 nmol/l, n = 20) and vitamin D–deficient (serum 25OHD < 39 nmol/l, n = 22). In the vitamin D–deficient group, type II fibers were significantly smaller (15.4 ± 4.2 µm) than in the vitamin D–sufficient group (38.7 ± 8.1 µm) (p < 0.0001). Furthermore, in the vitamin D–deficient group, type II fiber diameters correlated with serum levels of 25OHD (r = 0.714, p = 0.001). No such correlation was demonstrable in the vitamin D–sufficient group. Eight patients with vitamin D deficiency (25OHD = 7.0 ± 0.7 nmol/l) underwent muscle biopsies of the vastus lateralis muscle before and after three months of vitamin D treatment. The Ca-ATPase content increased significantly during treatment (before 3.4 ± 0.4, after 4.7 ± 0.2 nmol/g wet wt, p < 0.02) [23]. This increase is most likely explained by regeneration of type II fibers, which have a higher overall Ca-ATPase content.

The Ca-ATPase content correlated to serum levels of 25OHD (r = 0.5, p < 0.05) but not to 1,25(OH)2D or PTH. During sudden movements (e.g., correction of postural imbalance), the fast and strong type II fibers are the first to be recruited to avoid falling [32]. Thus, the selective type II fiber atrophy seen in HDM may be closely associated with the tendency to fall in vitamin D–deficient individuals [23,55,56].

B. Clinical Studies on Muscle Function In 1975 Skaria et al. [6] reported that 25 of 30 patients with osteomalacia exhibited an abnormal electromyogram, with signs of both myopathy and reduced nerve conduction velocity. Fourteen patients were followed during vitamin D treatment for several months. All but one showed improvement in their electromyograms, whereas no improvement in nerve conduction could be detected. In a prospective study Young et al. [10] followed the progress in muscle function in 12 osteomalacic patients during vitamin D treatment. They used an isometric dynamometer model for measurement of quadriceps muscle strength, and obtained muscle biopsies from the lateral vastus muscle. A significant improvement in both muscle strength and muscle biopsies was demonstrated after three months of treatment, but full restitution of the muscles required treatment for 6 to 12 months. Several studies have shown clinical improvement in muscle strength and function following vitamin D therapy. In two studies α-calcidiol 0.5 µg/d given to frail vitamin D–deficient elderly individuals, improved knee extension strength, walking speed [57], and “time taken to dress” [15]. In 349 elderly individuals (age >70 years) Mowe et al. [58] demonstrated a significant correlation between handgrip-strength and serum levels of 25OHD (r = 0.22, p < 0.001) and an inverse correlation between tendency to fall and 25OHD (r = 0.30, p < 0.05). These findings are in agreement with those of Bischoff et al. [59], who investigated leg extension power in relation to serum levels of vitamin D metabolites in 103 women and 216 men (65–95 years). In men, muscle power correlated to 25OHD (r = 0.24, p < 0.0004) and 1,25(OH)2D (r = 0.14, p < 0.045). In women, no significant correlation to 25OHD was found. However, only 12% of the women exhibited low 25OHD levels (<30 nmol/l). Using a scoring system for daily living activities Gloth et al. [60] were able to show improvement during vitamin D treatment, whereas Corless et al. [61] found no effect of supplementation with ergocalciferol

1812 (9000 IU) to frail old patients. A high prevalence of severe comorbidities in the study group, however, most likely influenced the evaluation of functional performance. The presence of severe comorbidity probably also explains the negative results in the study of Verreault et al. [62]. Boonen et al. [63] investigated the correlation between muscle function and serum levels of 1,25(OH)2D in 245 elderly women (70–90 years). No correlation could be demonstrated. 25OHD levels were not reported in this study. Grady et al. [64] studied the effect of daily supplementation with 0.5 µg calcitriol to 98 elderly (age >69 years). No effect on muscle function was found. However, the study group displayed a normal vitamin D status (25OHD > 60 nmol/l), and thus no HDM would be expected. Both studies are confounded by the age related loss of muscle strength (1.5% pr year) reported by McComas [32], which seems to be obligatory and nonpreventable by vitamin D. In veiled Arab women living in Denmark (n = 55) (25OHD = 6.7 ± 0.6 nmol/l), Glerup et al. [8,65] demonstrated a 34% reduction in muscle power determined by voluntary knee extension (MVC), when compared to controls (N = 22) with normal vitamin D levels. A series of ergocalciferol injections (100,000 IU weekly for one month followed by 100,000 IU monthly for six months) and a daily supplement of 1200 mg calcium were given. MVC increased by 13% after three months and by 24% after six months (p < 0.02). Electrical stimulation of the quadriceps muscle was also performed. The resulting single-twitch measurements (see Fig. 6) allowed estimation of muscle kinetics. Using this experimental setup, the maximal force production is derived from the slope of the ascending limb of the curve, while the slope of the descending limb of the curve expresses the maximal relaxation rate. The vitamin D–deficient group exhibited slower and weaker muscles compared to controls. After six months of treatment, however, muscle kinetics of the quadriceps was completely normalized. These findings corroborate the findings of Rodman and Baker [45] in rats. Glerup et al. also demonstrated that HDM precedes biochemical signs of bone involvement [8]. Further, MVC correlated significantly with serum levels of 25OHD (r = 0.34, p < 0.01) and PTH (r = −0.33, p < 0.001), but not with 1,25(OH)2D (r = −0.14, NS). In a multivariate regression analysis, only 25OHD was found to be significant. Decreased muscle power could be detected when serum levels of 25OHD were below 60 nmol/l (see Fig. 7). Some of the participants underwent investigation with 31P-MR-spectroscopy [66] in order to determine the effects of vitamin D status on energy metabolism. The muscular content of energy rich phosphocreatine and inorganic phosphate were

HENNING GLERUP AND ERIK FINK ERIKSEN

N 80

Single twitch electrical stimulation *p < 10−4 **p < 10−5

NS

70 60

*

50

**

40 30 20 10 0

−5

0

10 × 10 ms

5

D-depleted-baseline 3 months of vit D treatment 6 months of vit D treatment Controls

MPR expressed as the slope in the increasing part of the curve MRR expressed as the slope in the decreasing part of the curve

FIGURE 6

Single-twitch electrical stimulation of the quadriceps muscle was performed in an isometric dynamometer model with the knee in 90° bowed position. 55 vitamin D–deficient patients (25OHD = 6.7 ± 0.6 nmol/l) had performed measurement before and after 3 and 6 months of high-dose vitamin D treatment. The results were compared with those of 22 matched controls with normal vitamin D level. The ascending part of the curve expresses the maximal production rate of the muscle. The descending part expresses the maximal relaxation rate. Reproduced with permission from Calcified Tissue International 2000;66:419–424. [8].

Twitch (N) 120

100

+2SD +1SD

80

−1SD

60

−2SD 40

20

0 0

FIGURE 7

20

40

60 80 25OHD (nmol/l)

100

120

In single-twitch electrical stimulation of the quadriceps muscle, decreased muscle function could be detected at a level of 25-OHD below 60 nmol/l. The solid and dotted lines indicate the range of normal values (mean 74.6 nmol/l ± 2 SD) of controls with normal levels of 25OHD (47.1 ± 4.6 nmol/l) [8].

CHAPTER 102 Muscles and Falls

determined during rest, work, and recovery after work. At rest, vitamin D–deficient patients exhibited significant lower levels of phosphocreatine and inorganic phosphate than controls. During work, the decline in phosphocreatine and increase in inorganic phosphate was higher in vitamin D–deficient patients. Also, the recovery time after contraction was significantly longer in vitamin D–deficient patients than controls (56 s vs. 36 s, p < 0.05). After three months of vitamin D treatment all measurements were normalized. The susceptibility of individuals to vitamin D deficiency may also be genetically determined. Geusens et al. [67] recently reported effects of VDR polymorphisms on muscle function. In nonobese women they found a 23% higher quadriceps strength in genotype bb compared to genotype BB (p < 0.01).

VI. CLINICAL STUDIES ON THE RISK OF FALLS AND VITAMIN D STATUS A. Vitamin D and Fracture Risk As mentioned in the introduction to this chapter, the study by Chapuy et al. [18] inspired many investigators to perform studies on the muscular effects of vitamin D. In a group of 3270 elderly (84 ± 6 years) nursing home residents, they demonstrated a 43% reduction in hip fracture incidence after 18 months of treatment with 800 IU cholecalciferol and 1200 mg calcium daily. This reduction in hip fracture incidence was somewhat surprising since the change in BMD was more modest. BMD of the proximal femur increased by 2.7% in the vitamin D–treated group compared to a 4.6% decrease in the placebo group. Re-evaluation after three years continued to show a 25% reduction in hip fracture incidence. The obvious question was: could the results be partly explained by improvement in muscle function and consequently reduced risk of falls? The relations between muscle function, risk of falls, and vitamin D status have recently been extensively reviewed by Jansen et al. [68] and Pfeifer et al. [69]. They found that 90% of all hip fractures involve a fall [70,71]. Fractures caused by falls occur in about 5% of elderly persons every year, with the hip being involved in 1–2% [72–74]. The hypothesis that vitamin D supplementation may reduce the risk of falls and thus fractures has gained further support from several studies. Dawson-Hughes et al. [75] found a 58% reduction in nonvertebral fractures in 389 elderly (age > 65 years) treated with daily supplementation of 700 IU cholecalciferol and 500 mg calcium. Heikenheimo et al. [76] gave an annual injection of ergocalciferol (150,000–300,000 IU) to

1813 621 elderly (age > 74 years) resulting in a significant reduction in fractures of the upper extremities. Lips et al. [77], on the other hand, found no effect on the incidence of hip or peripheral fractures in a 3.5 years study of 2578 elderly supplied with 400 IU cholecalciferol daily. Meyer et al. [78] gave the same dose of cholecalciferol to 569 elderly nursing home residents for two years. Again no fracture-preventing effect was seen when compared to matched controls. The negative results of the last two studies are most likely explained by the low dose of vitamin D used in these subjects. Furthermore, no calcium supplementation was given. This notion is further corroborated by the fact that the patients in the Lips study still showed signs of secondary hyperparathyroidism at study end-point [77]. Seasonal variation of hip fracture incidence [79,80], as well as a higher incidence of hip fractures at higher latitudes [80,81], has been described. This is most likely caused by seasonal variations in sunlight exposure and resulting serum levels of 25OHD [82]. It is less likely that changes in bone due to seasonal variation in 25OHD could result in increased fracture incidence [83]. In Europe, 36% of men and 47% of women aged > 70 years exhibit wintertime levels of 25OHD < 30 nmol/l [84]. Serum levels of 25OHD have been shown by Chapuy et al. [85] to correlate with latitude (r = −0.7, p < 0.01). More pronounced seasonal variations in 25OHD at higher latitudes might be a contributing factor to the increase in hip fracture incidence during wintertime at high latitude. Exposure to sunlight is quite essential in maintenance of normal vitamin D status and in the prophylaxis against HDM. Normally the skin supplies the body with 80–100% of its requirements of vitamin D [86]. If sunlight exposure of the skin is limited, however, 25OHD will decrease and HDM may develop. A study of veiled Caucasian women living in Denmark showed that lack of direct sunlight exposure required a daily supply of 800 IU vitamin D3 in order to maintain normal vitamin D status [65]. This is in agreement with the study of Holick [86], who investigated vitamin D levels in submariners. A supply of more than 600 IU vitamin D3 was necessary to maintain normal levels of vitamin D during the absence of sunlight for three months. In elderly fallers Dhesi et al. [87] described a relationship between 25OHD levels and the number of times per week the patient went outside. Thus, homebound frail old people have a high risk for development of HDM making them even more immobile. They get less sunlight exposure, and moreover their age-dependent skin atrophy further reduces the ability to produce vitamin D, despite ample sunlight exposure [86]. These data lead to the following conclusion: all individuals

1814

HENNING GLERUP AND ERIK FINK ERIKSEN

at risk of HDM, especially the elderly (> 65 years) who are not regularly exposed to sunlight, should be given a daily supplement of at least 800 IU vitamin D3 in order to avoid HDM.

33% of elderly people experience at least one fall per year [72–74]. Mowe et al. [58] found lower levels of 25OHD among fallers compared to nonfallers with an inverse correlation between serum levels of 25OHD and the risk of falls (r = −0.27, p < 0.001). This is in agreement with the findings of Stein et al. [56], who described a correlation between secondary hyperparathyroidism and the risk of falls. In a study of 4251 elderly Australian women living in residential care (age > 84 years), Flicker et al. [88] recently identified low serum levels of 25OHD as a major risk factor for falls (hazard risk ratio 0.64, p < 0.004). In a “falls clinic” in London, to which patients with at least one fall within the last 8 weeks, were referred, Dhesi et al. [87,89] described severe hypovitaminosis D (25OHD < 30 nmol/l) in 31.8% and moderate hypovitaminosis (25OHD < 50 nmol/l) in 72.8%. Patients with low levels of 25OHD (< 30 nmol/l) displayed significantly impaired psychomotor function measured by a performance test (AFPT), postural sway, choice reaction time (CRT), and isometric quadriceps strength measurement. In a multivariate analysis, 25OHD was identified as an independent variable for AFPT, CRT, and body sway. For quadriceps strength, PTH was found to be an independent variable. Pfeifer et al. [90] investigated the effect on body sway and fall incidence in 148 women (aged 74 ± 1 year) treated with either a daily dose of 1200 mg calcium in combination with 800 IU cholecalciferol or 1200 mg of calcium alone. After one year, a 9% decrease in body sway was seen in the vitamin D–treated group compared to the group treated with calcium only (p = 0.043). The number of falls per subject per year was 0.45 in the calcium-only group compared to 0.24 in the vitamin D–treated group (p = 0.034). Bischoff et al. [9] treated 122 elderly women (mean age 83.3 years) with either 1200 mg calcium in combination with 800 IU cholecalciferol (Cal + D) or calcium only (Cal). Musculoskeletal function was measured by knee extension and flexion strength, grip strength, and timed “up & go” test. After three months, a significant improvement in musculoskeletal function was found in the Cal + D group (p < 0.0094). The risk of falling was reduced by 49% when comparing the Cal + D to the Cal group (p < 0.01) (see Fig. 8). The data presented above lend strong support to the hypothesis, that it is vitamin D supplementation more

Probability

B. Vitamin D and Risk of Falls

0.8 0.7

Cal + D

0.6

Cal

0.5 0.4 0.3 0.2 0.1 0

0

1

2 3 Number of falls

>=4

FIGURE 8 Number of falls and the effect of treatment with 800 IU cholecalciferol + 1200 mg calcium daily (Cal + D) versus calcium alone (Cal). Adjusted probabilities and SEs for having zero, one, or multiple falls for subjects in both treatment groups. SEs are calculated by taking 1 SD above and below the mean rate of falls and calculating the resulting Poisson probabilities. Adjustments have been performed for length of observation in the treatment period, previous falls in pretreatment period, a person who fell in the pretreatment period, age, and baseline 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D. Reproduced from Bischoff HA, et al. J Bone Miner Res 2003;18:343–351 with permission of the American Society for Bone and Mineral Research.

than calcium supplementation that protects against falls. The analysis of these studies suggests that a daily dose of at least 800 IU vitamin D3 should be given in order to ensure vitamin D sufficiency and significant improvement in muscle function and reduction in the incidence of falls.

VII. IS HDM CAUSED BY LOW LEVELS OF 25OHD, 1,25(OH)2D OR ELEVATED PTH? From a theoretical point of view, the most likely effector associated with HDM should be 1,25(OH)2D. VDR has been identified in muscle [35–38] and 1,25(OH)2D shows the highest affinity for the receptor. In agreement with this theory, experimental research has almost exclusively focused on the muscular effects of 1,25(OH)2D. Clinical experience, however, indicates that complaints associated with HDM correlate to serum levels of 25OHD and not to 1,25(OH)2D. The lack of correlation to 1,25(OH)2D levels might be due to increased renal 1-α-hydroxylase activity caused by secondary hyperparathyroidism. PTH, on the other hand, might also exert direct effects on muscle. In the following sections the three “candidates” will be discussed separately.

1815

CHAPTER 102 Muscles and Falls

A. PTH

B. 1,25(OH)2D

It has been suggested that HDM is caused by raised levels of PTH associated with secondary hyperparathyroidism, rather vitamin D deficiency itself [91]. Fatigue and muscular weakness are classically associated with primary hyperparathyroidism [92–95], and successful parathyroid surgery results in significant improvement of these muscular symptoms [92,94,96]. In addition, muscle biopsies obtained from patients with primary hyperparathyroidism reveal type II fiber atrophy, very much like the changes seen in HDM. Treatment with PTH reduces the intracellular content of inorganic phosphate, creatine phosphate, and CaATPase [97], which are exactly the findings we reported after analysis of muscle biopsies and 31P-MR-spectroscopy in patients with HDM [23,66]. Furthermore, mitochondrial oxygen consumption and the activity of creatine phosphokinase and CaATPase are reduced and oxidation of long-chain fatty acids impaired by PTH [98]. Thus, many parallels exist between myopathy caused by primary hyperparathyroidism and HDM. However, muscle weakness is not always present in patients with primary hyperparathyroidism [3,99], and HDM can be present despite normal PTH levels [9]. Further, improvements in muscle strength after surgery for primary hyperparathyroidism do not correlate to postoperative decreases in PTH or calcium [94,100]. If PTH was the only effector on skeletal muscle, reversal of secondary hyperparathyroidism by a high intake of calcium should reverse the muscular symptoms. However, in their rat study Rodman and Baker [45] detected severe perturbations of muscle function in vitamin D-deficient rats despite high serum levels of calcium and phosphate (Fig. 4). Furthermore, in a study on type II fiber size in hip fracture patients, Sato et al. [54] reported a significant correlation between serum levels of 25OHD and fiber size—no correlation to PTH was reported. Generally, clinical studies [8,56,89] report inverse correlations between PTH and muscle symptoms, and positive correlations to 25OHD [8,56,58,59,89]. In a comment accompanying the study of Stein et al. [56], Birge [83] suggested that PTH might be a better biological marker for vitamin D deficiency at the tissue level than serum levels of 25OHD, and this could be the reason why PTH and not 25OHD come out as significant determinants in multiple regression analyses. In this context, however, the significant interaction between 25OHD and PTH has to be taken into account. In conclusion: There is strong evidence for a direct effect of vitamin D on muscle both in clinical and experimental studies. It is possible, however, that secondary hyperparathyroidism may exert additive or synergistic effects on HDM development.

Theoretically, 1,25(OH)2D concentrations should be the most likely effector of vitamin D effects on muscle— and indeed, a large amount of experimental data support this notion (see Chapter 55). In clinical studies, however, this expected relationship finds less support [59]. Glerup et al. [8,23] found no correlation between muscle function and 1,25(OH)2D (r = −0.14, NS), whereas maximal knee extension strength correlated significantly to 25OHD (r = 0.34, p < 0.01). In fact, it is common to see severe symptoms of HDM with normal or even elevated values of 1,25(OH)2D. Furthermore, hypovitaminosis D–related symptoms (diffuse muscle pain, deep bone pain, paresthesia, fatigue, muscle cramps, joint pain) all correlated to 25OHD (KruskalWallis ANOVA: p < 0.001) but not to 1,25(OH)2D (NS). The absence of correlations to 1,25(OH)2D is probably explained by several factors. First, renal 1-αhydroxylase activity is under tight control by PTH levels, resulting in normal or even elevated levels of 1,25(OH)2D, despite very low levels of 25OHD. Second, serum levels of 1,25(OH)2D don’t necessarily tell anything about the intracellular levels of the hormone in muscle cells. Third, Geusens et al. [67] have shown clinical importance of VDR-genotypes, which support an in vivo effect of VDR-mediated effects. Thus, 1,25(OH)2D seems to be involved in the pathogenesis of HDM. One hypothesis may reconcile the inconsistencies outlined above, namely the purported presence of intracellular, autocrine production of 1,25(OH)2D from 25OHD in muscle cells [68]. There is an increasing amount of evidence suggesting the clinical importance of extrarenal 1,25(OH)2D synthesis [101–107] (see Chapter 79). Two features distinguish extrarenal from renal synthesis of 1,25(OH)2D: 1) it is not under control of PTH, but is dependent on the availability of the substrate 25OHD; 2) local 1,25(OH)2D synthesis has been shown to take place in the mitochondria [101]. Muscle has a very high content of mitochondria, which makes muscle a very likely site of extrarenal 1,25(OH)2D synthesis. Intracellular production of 1,25(OH)2D could explain the correlation between 25OHD and the muscular effects of vitamin D. Further, this pathway still requires 1,25(OH)2D to be the final effector of vitamin D’s muscular effects. It has been argued that local 1,25(OH)2D synthesis in muscle should result in increased serum levels of 1,25(OH)2D. Significant local 1,25OH2D production has been identified in other tissues (endothelium [102], prostate [106], bone cells, liver cells, skin, etc. [103]), but these sites do not result in increased serum levels of 1,25(OH)2D. The absence of increased 1,25(OH)2D in these instances is most likely explained by the presence of highly-induced intracellular 24-hydroxylase

1816 activity, ensuring degradation of 1,25OH2D before it reaches the circulation. Also, failure of release of 1,25(OH)2D into the circulation may be an additional factor.

C. 25OHD 25OHD is considered to be the storage and circulating form of vitamin D, and measurement of serum levels of 25OHD best reflect the vitamin D status of the body. 25OHD has been presumed to be biologically inert, but recent data challenge this notion. As already mentioned above, serum levels of 25OHD correlate to the biological effects of vitamin D in vivo. The effects of 25OHD on muscle cells could be mediated in several ways. 25OHD has some affinity for VDR, but the affinity of 1,25(OH)2D for VDR is approximately 1000-fold higher than 25OHD. The serum concentration of 25OHD is about 500–1000 times higher than 1,25(OH)2D, but most is bound to vitamin D binding protein (DBP) (see Chapter 8). Competitive binding of the two vitamin D metabolites to VDR might be possible under some circumstances [108–110]. No specific receptor for 25OHD has been identified. As mentioned in the paragraph above, a more likely explanation is the local synthesis of 1,25(OH)2D from 25OHD as substrate. Finally, it is possible that 25OHD could exert direct effects on muscle via an effector-mechanism, which is still under investigation. Recently, Nykjaer et al. [111–113] (see Chapter 10) identified the cubilinmegalin receptor system as being responsible for renal reuptake of vitamin D metabolites bound to DBP. Muscle tissue possesses receptors of the LDL receptor family, which potentially could be involved in tissue specific uptake of 25OHD, but this still needs to be investigated (personal communication A. Nykjaer).

VIII. OTHER POSSIBLE MUSCULAR EFFECTS OF VITAMIN D A. Insulin Resistance in Vitamin D Deficiency — Due To HDM? Vitamin D deficiency has been reported to increase the risk of developing insulin resistance and abnormal oral glucose tolerance tests (OGTT) [114–117]. Striated muscle is central in the pathogenesis of Type 2 diabetes. GLUT4 is the most important glucose transporter in muscle [118–120]. The GLUT4 content of muscles declines with age [118,120], especially in the fast type II muscle fibers. Furthermore, GLUT4

HENNING GLERUP AND ERIK FINK ERIKSEN

is reduced in type 2 diabetes [119]. More research is necessary to establish a possible effect of vitamin D on the GLUT4 content of the muscle. In type 2 diabetes, serum levels of free fatty acids are elevated [121,122]. Significant perturbations in the energy metabolism of mitochondria in muscle has been described in hypovitaminosis D [66], as well as in the presence of increased levels of PTH [97]. Additional research is warranted on the possible effects of vitamin D and PTH on fatty oxidation in striated muscle.

B. Possible Effects of Vitamin D on Muscle Regeneration During exercise, serum levels of 1,25(OH)2D have been reported to increase temporarily [123–127]. Exercise damages the muscle fibers and induces regeneration and growth of the muscle through enhanced satellite cell proliferation [128,129]. It could be speculated that 1,25OH2D might be of importance in the regeneration process of muscle. Furthermore, reduced IGF-I levels seem to play a role in age-related muscle degeneration. A possible interrelationship between IGF-I levels and vitamin D levels should be investigated [130].

IX. SUMMARY In this chapter we have reviewed the increasing evidence pointing to direct effects of vitamin D on striated muscle, making striated muscle an important target organ for vitamin D. Hypovitaminosis D myopathy (HDM) is a reversible disease that can recover completely, usually with significant improvement within a few weeks to a month after beginning vitamin D treatment [7,131,132]. Full restoration of severe HDM, however, may take 6 to 12 months of treatment with vitamin D [10]. Moreover, there is strong evidence for the prophylactic effects of vitamin D to reduce the risk of falls through improved muscular function and thereby to decrease the incidence of fractures. A daily dose of at least 800 IU (20 µg) cholecalciferol preferably in combination with 1000–1200 mg calcium seems to be the most effective treatment. Consequently combined vitamin D and calcium prophylaxis should be considered to combat hip fractures in the elderly. All patients at risk for vitamin D deficiency (i.e., lack of sunlight exposure) should be suspected to suffer from HDM. Those patients suspected of having HDM should have a blood test performed for measurement of 25OHD and PTH. In severe cases of HDM, treatment

CHAPTER 102 Muscles and Falls

should be initiated with a higher dose of vitamin D in order to speed up recovery. 300,000 IU cholecalciferol or ergocalciferol can be given either as an oral dose or intramuscular injection. This can be given as a single dose or repeated every month for three months. The high dose vitamin D should be combined with a daily supply of calcium. In order to avoid HDM, the serum levels of 25OHD should be kept above 50 nmol/l and PTH levels should be suppressed to the normal range. Maintenance of normal 25OHD levels in the elderly should have a high priority, as hip fractures and disability carry a high cost for society as well as for the individual patients. Treatment of HDM results in significant improvement in quality of life. However, vitamin D is not the solution to every musculoskeletal problem in the aging population. The age-related loss of muscle power (approximately 1.5% per year [32]) seems to be obligatory and unrelated to vitamin D deficiency. The data summarized in this review, lead to new questions, of which the ones, we consider most important are listed below: 1. Do muscle cells have the capacity to synthesize 1,25(OH)2D from 25OHD? 2. Is hydroxylation of 25OHD to 1,25(OH)2D necessary in order to mediate its effect on muscle, or does 25OHD have an effect of its own? 3. How do elevated PTH levels interact with vitamin D in muscle? 4. Finally, is the uptake of 25OHD and 1,25(OH)2D in muscle a matter of simple diffusion, or do muscle cells possess a system for facilitated uptake of the compounds?

References 1. Prineas W, Stuart-Mason A, Henson, RA 1965 Myopathy in metabolic bone disease. BMJ I:1034–1036. 2. Smith R, Stern G 1967 Myopathy, osteomalacia, and hyperparathyroidism. Brain 90:593–602. 3. Smith R, Stern G 1969 Muscular weakness in osteomalacia and hyperparathyroidism. J Neurol Sci 8:511–520. 4. Chalmers J, Conacher WDH, Gardner DL, Scott PJ, 1967 Osteomalacia—A common disease in elderly women. J Bone Joint Surg 49B:403–423. 5. Schott GD, Wills MR, 1976 Muscle weakness in osteomalacia. Lancet 1:626–629. 6. Skaria J, Katiyar BC, Srivastava TP, Dube B 1975 Myopathy and neuropathy associated with osteomalacia. Acta Neurol Scand 51:37–58. 7. Russell JA 1994 Osteomalacic myopathy. Muscle Nerve 17:578–580. 8. Glerup H, Mikkelsen K, Poulsen L, Hass E, Overbeck S, Andersen H, Charles P, Eriksen EF 2000 Hypovitaminosis D myopathy without biochemical signs of osteomalacic bone involvement. Calcif Tissue Int 66(6):419–424.

1817 9. Bischoff HA, Stahelin HB, Dick W, Akos R, Knecht M, Salis C, Nebiker M, Theiler R, Pfeifer M, Begerow B, Lew RA, Conzelmann M 2003 Effects of vitamin D and calcium supplementation on falls: a randomized controlled trial. J Bone Miner Res 18:343–351. 10. Young A, Edwards RHT, Jones DA, Brenton DP, 1981 Quadriceps muscle strength and fiber size during the treatment of osteomalacia. In: IAF Stokes (ed) Mechanical factors and the skeleton. 11. Young A, Brenton, DP, Edwards R 1978 Analysis of muscle weakness in osteomalacia. Clin Sci Mol Med 54(2): p. 31. 12. Mallette LE, Patten BM, Engel WK 1975 Neuromuscular disease in secondary hyperparathyroidism. Ann Intern Med 82:474–483. 13. Dastur DK, Gagrat BM, Wadia NH, Desai M, Bharucha EP 1975 Nature of muscular change in osteomalacia: light- and electron-microscope observations. J Pathol 117:211–228. 14. Yoshikawa S, Nakamura T, Tanabe H, Imamura T 1979 Osteomalacic myopathy. Endocrinol Jpn 26:65–72. 15. Sorensen OH, Lund B, Saltin B, Andersen RB, Hjorth L, Melsen F, Mosekilde L 1979 Myopathy in bone loss of aging: improvement by treatment with 1 alpha-hydroxycholecalciferol and calcium. Clin Sci 56:157–161. 16. Birge SJ, Haddad JG, 1975 25-hydroxycholecalciferol stimulation of muscle metabolism. J Clin Invest 56: 1100–1107. 17. Pointon JJ, Francis MJ, Smith R 1979 Effect of vitamin D deficiency on sarcoplasmic reticulum function and troponin C concentration of rabbit skeletal muscle. Clin Sci 57: 257–263. 18. Chapuy MC, Arlot ME, Duboeuf F, Brun J, Crouzet B, Arnaud S, Delmas PD, Meunier PJ, 1992 Vitamin D3 and calcium to prevent hip fractures in the elderly women. N Engl J Med 327:1637–1642. 19. Lips P, van Ginkel FC, Jongen MJ, Rubertus F, van der Vijgh WJ, Netelenbos JC 1987 Determinants of vitamin D status in patients with hip fracture and in elderly control subjects. Am J Clin Nutr 46:1005–1010. 20. Lund B, Sorensen OH, Melsen F, Mosekilde, L 1982 Vitamin D metabolism and osteomalacia in patients with fractures of the proximal femur. Acta Orthop Scand 53:251–254. 21. LeBoff MS, Kohlmeier L, Hurwitz S, Franklin J, Wright J, Glowacki J 1999 Occult vitamin D deficiency in postmenopausal U.S. women with acute hip fracture. JAMA 281: 1505–1511. 22. Glerup H, Eriksen EF 1999 Osteomalacia and severe vitamin D deficiency—A review of the clinical and paraclinical findings and guidelines for the treatment with vitamin D. Ugeskr Laeger 161:2515–2521. 23. Glerup H 1999 Investigations on the role of vitamin D in muscle function—A study of muscle function in vitamin D– deficient humans and the effect of treatment with vitamin D, PhD Thesis, University of Aarhus, Denmark. 24. Glerup H, Eriksen EF 1999 Acroparaesthesia—a typical finding in vitamin D deficiency. Br J Rheumatol 38:482. 25. Jones DA, Round JM 1992 Skeletal muscle in health and disease—a textbook of muscle physiology. Manchester University Press: Manchester, UK. 26. Gayton AC 1996 Textbook of medical physiology W.B. Saunders: Philadelphia. 27. White A, Handler P, Smith EL 1985 Muscle. Chap 36. Principles of Biochemistry 6th edn. McGraw-Hill. 28. Dulhunty AF, Junankar PR, Eager KR, Ahern GP, Laver DR, 1996 Ion channels in the sarcoplasmic reticulum of striated muscle. Acta Physiol Scand 156:375–385.

1818 29. Murphy RA, Muscle. In: RM Berne, MN Levy (eds) Physiology, Mosby: St. Louis. 30. Yano K, Zarain Herzberg A 1994 Sarcoplasmic reticulum calsequestrins: structural and functional properties. Mol Cell Biochem 135:61–70. 31. Milner RE, Famulski KS, Michalak M 1992 Calcium-binding proteins in the sarcoplasmic/endoplasmic reticulum of muscle and nonmuscle cells. Mol Cell Biochem 112:1–13. 32. McComas AJ 1996 Skeletal muscle—Form and function, Human Kinetics: Champaign, IL. 33. Dulhunty AF, Banyard MR, Medveczky CJ 1987 Distribution of calcium ATPase in the sarcoplasmic reticulum of fast- and slow-twitch muscles determined with monoclonal antibodies. J Membr Biol 99:79–92. 34. Maier A, Leberer E, Pette D 1986 Distribution of sarcoplasmic reticulum Ca-ATPase and of calsequestrin in rabbit and rat skeletal muscle fibers. Histochemistry 86:63–69. 35. Costa EM, Blau HM, Feldman D 1986 1,25-dihydroxyvitamin D3 receptors and hormonal responses in cloned human skeletal muscle cells. Endocrinology 119:2214–2220. 36. Bischoff HA, Borchers M, Gudat F, Duermueller U, Theiler R, Stahelin HB, Dick W 2001 In situ detection of 1,25-dihydroxyvitamin D3 receptor in human skeletal muscle tissue. Histochem J 33:19–24. 37. Boland R, Norman A, Ritz E, Hasselbach W 1985 Presence of a 1,25-dihydroxyvitamin D3 receptor in chick skeletal muscle myoblasts. Biochem Biophys Res Commun 128: 305–311. 38. Simpson RU, Thomas GA, Arnold AJ 1985 Identification of 1,25-dihydroxyvitamin D3 receptors and activities in muscle. J Biol Chem 260:8882–8891. 39. de Boland AR, Albornoz LE, Boland R 1983 The effect of cholecalciferol in vivo on proteins and lipids of skeletal muscle from rachitic chicks. Calcif Tissue Int 35:798–805. 40. de Boland AR, Boland RL 1984 Effects of vitamin D3 on in vivo labeling of chick skeletal muscle proteins with [3H]leucine. Z Naturforsch C 39:1015–1016. 41. Boland R 1986 Role of vitamin D in skeletal muscle function. Endocr Rev 7:434–448. 42. de Boland AR, Boland RL 1987 Rapid changes in skeletal muscle calcium uptake induced in vitro by 1,25-dihydroxyvitamin D3 are suppressed by calcium channel blockers. Endocrinology 120:1858–1864. 43. de Boland, AR, Boland, RL, 1994 Nongenomic signal transduction pathway of vitamin D in muscle. Cell Signal 6:717–724. 44. Curry OB, Basten JF, Francis MJ, Smith R 1974 Calcium uptake by sarcoplasmic reticulum of muscle from vitamin D– deficient rabbits. Nature 249:83–84. 45. Rodman JS, Baker T 1978 Changes in the kinetics of muscle contraction in vitamin D–depleted rats. Kidney Int 13: 189–193. 46. Matthews C, Heimberg KW, Ritz E, Agostini B, Fritzsche J, Hasselbach, W 1977 Effect of 1,25-dihydroxycholecalciferol on impaired calcium transport by the sarcoplasmic reticulum in experimental uremia. Kidney Int 11:227–235. 47. Boland R, de Boland AR, Ritz E, Hasselbach, W 1983 Effect of 1,25-dihydroxycholecalciferol on sarcoplasmic reticulum calcium transport in strontium-fed chicks. Calcif Tissue Int 35:190–194. 48. Boland R, Matthews C, de Boland AR, Ritz E, Hasselbach, W 1983 Reversal of decreased phosphorylation of sarcoplasmic reticulum calcium transport ATPase by 1,25-dihydroxycholecalciferol in experimental uremia. Calcif Tissue Int 35:195–201.

HENNING GLERUP AND ERIK FINK ERIKSEN

49. Sjostrom M, Lorentzon R, Larsson SE, Holmlund, D 1978 The influence of 1,25-dihydroxycholecalciferol on the ultrastructural organization of skeletal muscle fibers. Morphometric analyses on vitamin D–deficient or calcium–deficient growing rats. Med Biol 56:209–215. 50. Pleasure D, Wyszynski B, Sumner A, Schotland D, Feldman B, Nugent N, Hitz K, Goodman DB 1979 Skeletal muscle calcium metabolism and contractile force in vitamin D–deficient chicks. J Clin Invest 64:1157–1167. 51. Toury R, Stelly N, Boisonneau E, Convert M, Dupuis Y 1990 Relationship between vitamin D status and deposition of bound calcium in skeletal muscle of the rat. Biol Cell 69:179–189. 52. Ronin DI, Wu Y, Sahgal V, MacLean IC 1991. Intractable muscle pain syndrome, osteomalacia, axonopathy in longterm use of phenytoin. Arch Phys Med Rehabil 72(10): 755–758. 53. Ruff, RL, Weissmann,J 1988 Endocrine myopathies. Neurol Clin 6:575–592. 54. Sato Y, Inose M, Higuchi I, Higuchi F, Kondo, I 2002 Changes in the supporting muscles of the fractured hip in elderly women. Bone 30:325–330. 55. Pfeifer M, Minne HW 1999 Vitamin D and hip fracture. Trends Endocrinol Metab 10:417–420. 56. Stein MS, Wark JD, Scherer SC, Walton SL, Chick P, Di Carlantonio M, Zajac JD, Flicker L 1999 Falls relate to vitamin D and parathyroid hormone in an Australian nursing home and hostel. J Am Geriatr Soc 47:1195–1201. 57. Verhaar HJ, Samson MM, Jansen PA, de Vreede PL, Manten JW, Duursma SA 2000 Muscle strength, functional mobility, and vitamin D in older women. Aging (Milano) 12:455–460. 58. Mowe M, Haug E, Bohmer T 1999 Low serum calcidiol concentration in older adults with reduced muscular function. J Am Geriatr Soc 47:220–226. 59. Bischoff HA, Stahelin HB, Urscheler N, Ehrsam R, Vonthein R, Perrig-Chiello P, Tyndall A, Theiler R 1999 Muscle strength in the elderly:its relation to vitamin D metabolites. Arch Phys Med Rehabil 80:54–58. 60. Gloth FM, Smith CE, Hollis BW, Tobin JD 1995 Functional improvement with vitamin D replenishment in a cohort of frail, vitamin D–deficient older people. J Am Geriatr Soc 43:1269–1271. 61. Corless D, Dawson E, Fraser F, Ellis M, Evans SJ, Perry JD, Reisner C, Silver CP, Beer M, Boucher BJ 1985 Do vitamin D supplements improve the physical capabilities of elder hospital patients? Age Ageing 14:76–84. 62. Verreault R, Semba RD, Volpato S, Ferrucci L, Fried LP, Guralnik JM 2002 Low serum vitamin D does not predict new disability or loss of muscle strength in older women. J Am Geriatr Soc 50:912–917. 63. Boonen S, Lysens R, Verbeke G, Joosten E, Dejaeger E, Pelemans WK, Flamaing J, Bouillon R 1998 Relation between age-associated endocrine deficiencies and muscle function in elderly women: a cross-sectional study. Age Ageing 27:449–454. 64. Grady D, Halloran B, Cummings S, Leveille S, Wells L, Black D, Byl N 1991 1,25-dihydroxyvitamin D3 and muscle strength in the elderly: a randomized controlled trial. J Clin Endocrinol Metab 73:1111–1117. 65. Glerup H, Thomsen J, Mikkelsen K, Poulsen L, Hass E, Overbeck S, Charles P, Eriksen EF 2000 Commonly recommended daily intake of vitamin D is not sufficient if sunlight exposure is limited. J Intern Med 247: 260–268.

CHAPTER 102 Muscles and Falls

66. Glerup H, Stødkilde-Jørgensen H, Eriksen EF 2001 Hypovitaminosis D myopathy characterized by 31P MR spectroscopy. Bone 28 (Suppl 5):S121. 67. Geusens P, Vandevyver C, Vanhoof J, Cassiman JJ, Boonen S, Raus J 1997 Quadriceps and grip strength are related to vitamin D receptor genotype in elderly nonobese women. J Bone Miner Res 12:2082–2088. 68. Janssen HC, Samson MM, Verhaar HJ 2002 Vitamin D deficiency, muscle function, and falls in elderly people. Am J Clin Nutr 75:611–615. 69. Pfeifer M, Begerow B, Minne HW 2002 Vitamin D and muscle function. Osteoporos Int 13:187–194. 70. Cummings SR, Nevitt MC 1989 A hypothesis: the causes of hip fractures. J Gerontol 44:M107–111. 71. Grisso JA, Kelsey JL, Strom BL, Chiu GY, Maislin G, O’Brien LA, Hoffman S, Kaplan F 1991 Risk factors for falls as a cause of hip fracture in women. The Northeast Hip Fracture Study Group. N Engl J Med 324:1326–1331. 72. Tinetti ME, Williams CS 1997 Falls, injuries due to falls, and the risk of admission to a nursing home. N Engl J Med 337:1279–1284. 73. Blake AJ, Morgan K, Bendall MJ, Dallosso H, Ebrahim SB, Arie TH, Fentem PH, Bassey EJ 1988 Falls by elderly people at home: prevalence and associated factors. Age Ageing 17:365–372. 74. Graafmans WC, Ooms ME, Hofstee HM, Bezemer PD, Bouter LM, Lips P 1996 Falls in the elderly: a prospective study of risk factors and risk profiles. Am J Epidemiol 143:1129–1136. 75. Dawson Hughes B, Harris S, Krall EA, Dallal GE 1997 Effect of calcium and vitamin D supplementation on bone density in men and women 65 years of age or older. N Engl J Med 337:670–676. 76. Heikinheimo RJ, Inkovaara JA, Harju EJ, Haavisto MV, Kaarela RH, Kataja JM, Kokko AM, Kolho LA, Rajala SA 1992 Annual injection of vitamin D and fractures of aged bones. Calcif Tissue Int 51:105–110. 77. Lips P, Graafmans WC, Ooms ME, Bezemer PD, Bouter LM 1996 Vitamin D supplementation and fracture incidence in elderly persons. A randomized, placebo-controlled clinical trial. Ann Intern Med 124:400–406. 78. Meyer HE, Smedshaug GB, Kvaavik E, Falch JA, Tverdal A, Pedersen JI 2002 Can vitamin D supplementation reduce the risk of fracture in the elderly? A randomized controlled trial. J Bone Miner Res 17:709–715. 79. Jacobsen SJ, Goldberg J, Miles TP, Brody JA, Stiers W, Rimm AA 1991 Seasonal variation in the incidence of hip fracture among white persons aged 65 years and older in the United States, 1984–1987. Am J Epidemiol 133:996–1004. 80. Douglas S, Bunyan A, Chiu KH, Twaddle B, Maffulli N 2000 Seasonal variation of hip fracture at three latitudes. Injury 31(1):11–19. 81. Bacon WE, Smith GS, Baker DL 1989 Geographical variation in the occurrence of hip fractures among elderly white US population. Am J Public Health 79(11):1556–1558. 82. Brot C, Vestergaard P, Kolthoff N, Gram J, Hermann AP, Sorensen OH 2001 Vitamin D status and its adequacy in healthy Danish perimenopausal women: relationships to dietary intake, sun exposure, and serum parathyroid hormone. Br J Nutr 86(Suppl 1):S97–S103. 83. Birge SJ 1999 Can falls and hip fracture be prevented in frail older adults? J Am Geriatr Soc 47:1265–1266. 84. van der Wielen RP, Lowik MR, van dB, de GL, Haller J, Moreiras O, van SW 1995 Serum vitamin D concentrations among elderly people in Europe. Lancet 346:207–210.

1819 85. Chapuy MC, Preziosi P, Maamer M, Arnaud S, Galan P, Hercberg S, Meunier PJ 1997 Prevalence of vitamin D insufficiency in an adult normal population. Osteoporos Int 7:439–443. 86. Holick MF 1994 McCollum Award Lecture, 1994: vitamin D— new horizons for the 21st century. Am J Clin Nutr 60: 619–630. 87. Dhesi JK, Moniz C, Close JC, Jackson SH, Allain TJ 2002 A rationale for vitamin D prescribing in a falls clinic population. Age Ageing 31:267–271. 88. Flicker L, Mead K, Nowson CA, Scherer SC, Stein MS, Thomas J, Hopper JL, Wark JD 2001 Risk factors for falls in older women in residential care in Australia. J Bone Miner Res 16(Suppl 1):S170. 89. Dhesi JK, Bearne LM, Moniz C, Hurley MV, Jackson SH, Swift CG, Allain TJ 2002 Neuromuscular and psychomotor function in elderly subjects who fall and the relationship with vitamin D status. J Bone Miner Res 17:891–897. 90. Pfeifer M, Begerow B, Minne HW, Abrams C, Nachtigall D, Hansen, C 2000 Effects of a short-term vitamin D and calcium supplementation on body sway and secondary hyperparathyroidism in elderly women. J Bone Miner Res 15: 1113–1118. 91. McCarty MF 2002 Vitamin D status and muscle strength. Am J Clin Nutr 76:1454–1455. 92. Colliander EB, Strigard K, Westblad P, Rolf C, Nordenstrom J 1998 Muscle strength and endurance after surgery for primary hyperparathyroidism. Eur J Surg 164:489–494. 93. Turken SA, Cafferty M, Silverberg SJ, De La Cruz L, Cimino C, Lange DJ, Lovelace RE, Bilezikian JP 1989 Neuromuscular involvement in mild, asymptomatic primary hyperparathyroidism. Am J Med 87:553–557. 94. Kristoffersson A, Bjerle P, Stjernberg N, Järhult J 1988 Preand postoperative respiratory muscle strength in primary hyperparathyroidism. Acta Chir Scand 154:415–418. 95. Patten MP, Mallette LE, Prince A, Aurbach GD, Belezikian JP, King Engel W 1974 Neuromuscular disease in primary hyperparathyroidism. Ann Intern Med 80:182–193. 96. Deutch SR, Jensen MB, Christiansen PM, Hessov I 2000 Muscular performance and fatigue in primary hyperparathyroidism. World J Surg 24:102–107. 97. Baczynski R, Massry SG, Magott M, el Belbessi S, Kohan R, Brautbar N 1985 Effect of parathyroid hormone on energy metabolism of skeletal muscle. Kidney Int 28: 722–727. 98. Smogorzewski M, Piskorska G, Borum PR, Massry SG 1988 Chronic renal failure, parathyroid hormone, and fatty acids oxidation in skeletal muscle. Kidney Int 33: 555–560. 99. Jansson S, Grimby G, Hagne I, Hedman I, Tisell LE 1991 Muscle structure and function before and after surgery for primary hyperparathyroidism. Eur J Surg 157:13–16. 100. Chou FF, Sheen Chen SM, Leong CP 1995 Neuromuscular recovery after parathyroidectomy in primary hyperparathyroidism. Surgery 117:18–25. 101. Zehnder D, Bland R, Walker EA, Bradwell AR, Howie AJ, Hewison M, Stewart PM 1999 Expression of 25-hydroxyvitamin D3-1alpha-hydroxylase in the human kidney. J Am Soc Nephrol 10:2465–2473. 102. Zehnder D, Bland R, Chana RS, Wheeler DC, Howie AJ, Williams MC, Stewart PM, Hewison M 2002 Synthesis of 1,25-dihydroxyvitamin D3 by human endothelial cells is regulated by inflammatory cytokines: a novel autocrine determinant of vascular cell adhesion. J Am Soc Nephrol 13:621–629.

1820 103. Dusso A, Brown A, Slatopolsky E 1994 Extrarenal production of calcitriol. Semin Nephrol 14:144–155. 104. Dusso A, Lopez-Hilker S, Rapp N, Slatopolsky E 1988 Extrarenal production of calcitriol in chronic renal failure. Kidney Int 34:368–375. 105. Dusso AS, Finch J, Brown A, Ritter C, Delmez J, Schreiner G, Slatopolsky E 1991 Extrarenal production of calcitriol in normal and uremic humans. J Clin Endocrinol Metab 72: 157–164. 106. Schwartz GG, Whitlatch LW, Chen TC, Lokeshwar BL, Holick MF 1998 Human prostate cells synthesize 1,25-dihydroxyvitamin D3 from 25-hydroxyvitamin D3. Cancer Epidemiol Biomarkers Prev 7:391–395. 107. Barreto AM, Schwartz GG, Woodruff R, Cramer SD 2000 25-hydroxyvitamin D3, the prohormone of 1,25-dihydroxyvitamin D3, inhibits the proliferation of primary prostatic epithelial cells. Cancer Epidemiol Biomarkers Prev 9:265–270. 108. Reichel H, Koeffler HP, Norman AW 1989 The role of the vitamin D endocrine system in health and disease. N Engl J Med 320:980–991. 109. Barger-Lux MJ, Heaney RP, Lanspa SJ, Healy JC, DeLuca HF 1995 An investigation of sources of variation in calcium absorption efficiency. J Clin Endocrinol Metab 80:406–411. 110. Barger-Lux MJ, Heaney RP, Dowell S, Bierman J 1996 Relative molar potency of 25-hydroxyvitamin D indicates a major role in calcium absorption. J Bone Miner Res 11:S423. 111. Nykjaer A, Dragun D, Walther D, Vorum H, Jacobsen C, Herz J, Melsen F, Christensen EI, Willnow TE 1999 An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) vitamin D3. Cell 96:507–515. 112. Nykjaer A, Fyfe JC, Kozyraki R, Leheste JR, Jacobsen C, Nielsen MS, Verroust PJ, Aminoff M, de la CA, Moestrup SK, Ray R, Gliemann J, Willnow TE, Christensen EI 2001 Cubilin dysfunction causes abnormal metabolism of the steroid hormone 25(OH) vitamin D3. Proc Natl Acad Sci USA 20:98, 13895–13900. 113. Leheste JR, Melsen F, Wellner M, Jansen P, Schlichting U, Renner-Muller I, Andreassen TT, Wolf E, Bachmann S, Nykjaer A, Willnow TE 2002 Hypocalcemia and osteopathy in mice with kidney-specific megalin gene defect. Faseb J 114. Boucher BJ 1998 Inadequate vitamin D status: does it contribute to the disorders comprising syndrome ‘X’? Br J Nutr 79:315–327. 115. Boucher BJ, Mannan N, Noonan K, Hales CN, Evans SJ 1995 Glucose intolerance and impairment of insulin secretion in relation to vitamin D deficiency in east London Asians. Diabetologia 38:1239–1245. 116. Baynes KCR, Boucher BJ, Feskens EJM, Kromhout D 1999 Vitamin D, glucose tolerance, and insulinanemia in elderly men. Diabetologia 40:344–347. 117. Rudnicki PM, Molsted-Pedersen L 1997 Effect of 1,25-dihydroxycholecalciferol on glucose metabolism in gestational diabetes mellitus. Diabetologia 40:40–44.

HENNING GLERUP AND ERIK FINK ERIKSEN

118. Rittweger J, Rauch F 2000 What is new in musculoskeletal interactions. J Musculoskel Neuron Interact 1:171–176. 119. Gaster M, Staehr P, Beck-Nielsen H, Schroder HD, Handberg A 2001 GLUT4 is reduced in slow muscle fibers of type 2 diabetic patients: is insulin resistance in type 2 diabetes a slow, type 1 fiber disease? Diabetes 50:1324–1329. 120. Gaster M, Poulsen P, Handberg A, Schroder HD, BeckNielsen H 2000 Direct evidence of fiber type-dependent GLUT-4 expression in human skeletal muscle. Am J Physiol Endocrinol Metab 278:E910–E916. 121. Jensen MD 2002 Fatty acid oxidation in human skeletal muscle. J Clin Invest 110:1607–1609. 122. Rasmussen BB, Holmback UC, Volpi E, Morio-Liondore B, Paddon-Jones D, Wolfe RR 2002 Malonyl coenzyme A and the regulation of functional carnitine palmitoyltransferase-1 activity and fat oxidation in human skeletal muscle. J Clin Invest 110:1687–1693. 123. Klausen T, Breum L, Sorensen HA, Schifter S, Sonne B 1993 Plasma levels of parathyroid hormone, vitamin D, calcitonin, and calcium in association with endurance exercise. Calcif Tissue Int 52:205–208. 124. Zittermann A, Sabatschus O, Jantzen S, Platen P, Danz A, Stehle P 2002 Evidence for an acute rise of intestinal calcium absorption in response to aerobic exercise. Eur J Nutr 41:189–196. 125. Bell NH, Godsen RN, Henry DP, Shary J, Epstein S 1988 The effects of muscle-building exercise on vitamin D and mineral metabolism. J Bone Miner Res 3:369–373. 126. Zittermann A, Sabatschus O, Jantzen S, Platen P, Danz A, Dimitriou T, Scheld K, Klein K, Stehle P 2000 Exercisetrained young men have higher calcium absorption rates and plasma calcitriol levels compared with age-matched sedentary controls. Calcif Tissue Int 67:215–219. 127. Nelson ME, Meredith CN, Dawson-Hughes B, Evans WJ 1988 Hormone and bone mineral status in endurance-trained and sedentary postmenopausal women. J Clin Endocrinol Metab 66:927–933. 128. Smith HK, Maxwell L, Rodgers CD, McKee NH, Plyley MJ 2001 Exercise-enhanced satellite cell proliferation and new myonuclear accretion in rat skeletal muscle. J Appl Physiol 90:1407–1414. 129. Zammit PS, Heslop L, Hudon V, Rosenblatt JD, Tajbakhsh S, Buckingham ME, Beauchamp JR, Partridge TA 2002 Kinetics of myoblast proliferation show that resident satellite cells are competent to fully regenerate skeletal muscle fibers. Exp Cell Res 281:39–49. 130. Grounds MD 2002 Reasons for the degeneration of aging skeletal muscle: a central role for IGF-1 signaling. Biogerontology 3:19–24. 131. Prabhala A, Garg R, Dandona P 2000 Severe myopathy associated with vitamin D deficiency in western New York. Arch Intern Med 160:1199–1203. 132. Mingrone G, Greco AV, Castagneto M, Gasbarrini G 1999 A woman who left her wheelchair. Lancet 353:806.

CHAPTER 103

Renal Failure and Secondary Hyperparathyroidism MASAFUMI FUKAGAWA KIYOSHI KUROKAWA

Division of Nephrology & Dialysis Center, Kobe University School of Medicine, Kobe, Japan Research Center for Advanced Science and Technology, The University of Tokyo, Tokyo, Japan

I. Role of Vitamin D in the Development of Hyperparathyroidism in Renal Failure II. Resistance to 1,25(OH)2D as a Cause of Severe Secondary Hyperparathyroidism in Chronic Renal Failure

I. ROLE OF VITAMIN D IN THE DEVELOPMENT OF HYPERPARATHYROIDISM IN RENAL FAILURE The kidney is the main organ for the production of active vitamin D, 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) [1], a process that is catalyzed by 25-hydroxyvitamin D3 1alpha-hydroxylase (1α-hydroxylase) in the proximal tubule cells [2–4]. Activity of this enzyme is attenuated in chronic renal failure due to phosphate load [5,6] as well as to the decreased numbers of viable nephrons [1]. Furthermore, it has recently been shown that fibroblast growth factor-23 (FGF-23) may suppress activation of vitamin D [7]. FGF-23 is a newly discovered phosphaturic factor (see Chapters 26 and 29) and increased serum levels have been reported in patients with renal dysfunction [8]. In addition to the decreased production of 1,25(OH)2D3 in the kidney, the importance of vitamin D deficiency has been recognized again, especially in chronic kidney disease (CKD) stages 3 and 4 [9]. Vitamin D deficiency is reflected by decreased serum concentrations of 25(OH)D [10]. Such decrease of 25(OH)D level may result from the loss of vitamin D-binding protein into the urine [11] as well as malnutrition [12]. In addition, a decrease of megalin on the brush border of proximal tubules has been reported [13], which results in diminished reuptake of filtered 25(OH)D [14] (see Chapter 10). In chronic renal failure, the secretion of parathyroid hormone (PTH) is stimulated by several factors, primarily hypocalcemia and reduced production of 1,25(OH)2D3 [1]. In addition, direct stimulatory VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

III. Management of Severe Hyperparathyroidism Refractory to Medical Therapy IV. Future Roles of Vitamin D Analogs in Chronic Renal Failure References

action of phosphate on parathyroid has recently been demonstrated [15–17]. Thus, secondary hyperparathyroidism develops almost inevitably in patients with chronic renal failure without appropriate therapy [1]. Excess PTH accelerates bone turnover and results in a typical bone abnormality known as osteitis fibrosa [18]. Vitamin D metabolites suppress the secretion of PTH by correcting hypocalcemia and also by direct action on parathyroid cells in patients of chronic renal failure [19,20]. However, it is still difficult to suppress PTH secretion in substantial numbers of patients by vitamin D treatment. Such patients usually have marked parathyroid hyperplasia [21]. Since conventional uses of vitamin D in mild and advanced renal failure, including 1,25(OH)2D3 pulse therapy, are discussed in the Chapter 76 by Dusso, Brown, and Slatopolsky, we will focus on patients with severe disease that are refractory to medical therapy and summarize the new therapeutic uses of vitamin D metabolites.

II. RESISTANCE TO 1,25(OH)2D AS A CAUSE OF SEVERE SECONDARY HYPERPARATHYROIDISM IN CHRONIC RENAL FAILURE A. Resistance to 1,25(OH)2D in Chronic Renal Failure Despite physiological plasma concentrations of 1,25(OH)2D, as well as those of calcium ion obtained by routine treatment, there are still many patients with elevated plasma PTH levels. Some of these patients respond to supraphysiological concentration of 1,25(OH)2D3 Copyright © 2005, Elsevier, Inc. All rights reserved.

1822

MASAFUMI FUKAGAWA AND KIYOSHI KUROKAWA

achieved either by intravenous or oral intermittent high doses of 1,25(OH)2D3, which is also refered to as “1,25(OH)2D3 pulse therapy” [22–24]. These observations suggest that the resistance of parathyroid cells to 1,25(OH)2D may play a major role in the pathogenesis of severe secondary hyperparathyroidism in chronic renal failure [25]. Resistance to physiological concentrations of 1,25(OH)2D may develop during the early phase of chronic renal failure. In rat models of mild renal failure, PTH secretion, synthesis, and parathyroid cell proliferation were all enhanced even in the presence of a normal plasma concentration of calcium and 1,25(OH)2D [26]. Hyperparathyroidism returned to normal with pharmacological doses of 1,25(OH)2D3 without the induction of hypercalcemia. In these rats, 1,25(OH)2D receptor (VDR) density in parathyroid glands, detected by Western blot, was decreased compared to levels seen in normal rats. Such reduction of VDR density in parathyroid glands also has been demonstrated in enlarged parathyroid glands of chronic dialysis patients [27], as well as in animal models of chronic uremia [28,29]. This abnormality, reduced VDR concentration, is currently considered the central feature responsible for the resistance of parathyroid glands to 1,25(OH)2D in chronic renal failure [30]. In addition to the decreased density of VDR, several mechanisms have been proposed (Fig. 1). Decreased density of retinoid receptor X (RXR), which forms heterodimers with VDR, has been suspected [31]; however, the significance of this observation still remains unclear. Hsu and associates have been focused

on the possible inhibition of 1,25(OH)2D action by uremic toxins [32]. They have shown that serum from uremic patients inhibited the interaction between the 1,25(OH)2D-VDR complex and DNA [33], possibly through the formation of a Schiff base [33]. Although they have examined the effects of glyoxylate [35], other uremic toxins responsible for this inhibition still remain to be identified [36]. In addition, calreticulin has been shown to inhibit the binding of the 1,25(OH)2D-VDR complex to the vitamin D responsive element (VDRE) in the PTH gene promoter [37]. Hypocalcemia was found to induce increased concentrations of calreticulin exclusively in parathyroid glands. It is possible that this molecule plays some role in the regulation of VDR function by extracellular calcium [38,39]. Decreased action of 1,25(OH)2D due to these mechanisms finally results in disturbed up-regulation of VDR, which further increases resistance to 1,25(OH)2D in chronic renal failure (Fig. 1) [40–42]. Such a vicious cycle may be prevented by early vitamin D treatment as suggested by the animal models [43].

B. Advantages and Limitations of Intravenous Calcitriol Therapy Considering that these abnormalities cause vitamin D resistance, it is quite reasonable that supraphysiological concentrations of 1,25(OH)2D have been shown effective in suppressing PTH secretion in chronic dialysis patients resistant to conventional oral calcitriol [22–24]. Since the peak concentration of 1,25(OH)2D is more important

Calcium ion 1,25(OH)2D RXR 1,25(OH)2D receptor

1,25 D-receptor complex

Uremic toxins Up-regulation Calreticulin

DNA Nucleus AAAAAA mRNA Up-regulation Calcium-sensing receptor

FIGURE 1 Mechanisms of resistance to 1,25(OH)2D in chronic renal failure. Several steps of 1,25(OH)2D action are disturbed in chronic renal failure, leading to further reduction of VDR.

CHAPTER 103 Renal Failure and Secondary Hyperparathyroidism

for the suppression of PTH secretion than the total dose of calcitriol, as shown in dialysis patients [22] and in experimental animals [44], higher doses of calcitriol theoretically should be more effective. However, high doses of calcitriol often cause hypercalcemia and hyperphosphatemia, resulting in reduction or disconcentration of therapy. High Ca × Pi product leads to metastatic calcification including within blood vessels, which may result in a higher mortality risk [45]. This has been the main reason why less calcemic vitamin D analogs, such as 19-nor-1,25(OH)2D2 [46] and 22-oxa1,25(OH)2D3 [47], have been developed (see Section VIII of this book). Even with these less calcemic vitamin D analogs, PTH secretion is still difficult to control in some patients.

C. Parathyroid Size as a Marker for the Prognosis of Vitamin D Therapy In order to avoid unnecessary vitamin D treatment and metastatic calcification, it is certainly necessary to have a good predictor of the prognosis of vitamin D therapy for parathyroid suppression in these patients. Although recent data suggest that serum FGF-23 levels may be a marker for the future prognosis of hyperparathyroidism [48], parathyroid gland size assessed by ultrasonography is the most simple and useful marker at this time. Marked parathyroid gland hyperplasia is a unique feature of secondary hyperparathyroidism in chronic dialysis patients [49]. Although the size of each of the four glands is usually different, even in the same patient, it has been recognized by experience that the size of the largest gland roughly correlates with the length and severity of uremia and with the degree of prevailing plasma and stimulated peak PTH levels [50,51]. The size also correlates with the degree of abnormal control of PTH secretion [52,53], which may be normalized by calcitriol pulse therapy [54]. Clinical observations of dialysis patients suggest that the size of the largest gland is the critical marker for the long-term prognosis of vitamin D therapy [55]. If the largest gland is larger than 1 cm in diameter or about 0.5 cm3 in volume, it is usually difficult to suppress PTH secretion by calcitriol pulse therapy. In such patients, secondary hyperparathyroidism always persists or relapses even if it initially responded to calcitriol pulse therapy. By contrast, patients with only smaller glands usually respond well to calcitriol pulse therapy, and parathyroid gland function can be controlled then with oral active vitamin D sterols. Thus, the size of the parathyroid gland may have more relevance than plasma PTH levels in assessment of calcitriol pulse

1823

therapy [37]. Furthermore, Tominaga et al. demonstrated in patients treated by surgical parathyroidectomy that autoimplantation of tissue fragments from glands heavier than 0.5 g resulted in frequent relapse of hyperparathyroidism [56]. Thus, the critical size for the management strategy for hyperparathyroidism in chronic dialysis patients seems to be less than 0.5 cm3 in volume. The correlation between the gland size and the resistance to calcitriol can be explained by the degree of decrease of VDR density. VDR density is inversely correlated with the weight of enlarged glands [57]. Large parathyroid glands are usually composed of nodular hyperplasia, a more advanced type of pathology than diffuse hyperplasia seen in small glands [58]. It has been reported that cells in nodular hyperplasia glands have higher proliferative potentials [59,60,62] and more abnormal regulation of PTH secretion [63] than cells in diffuse hyperplasia glands. We and others have clearly shown that the VDR number was decreased more in nodular hyperplasia than in diffuse hyperplasia [57,64]. Since 90% of the glands heavier than 0.5 g were composed of nodular hyperplasia as shown by Tominaga and Takagi [61], the difference in the response to calcitriol that is dependent upon gland size can be explained by the difference in the type of hyperplasia in the larger glands. In nodular hyperplasia, decreased density of the calcium-sensing receptor also has been demonstrated [65,66]. Although it is still controversial whether this decrease is the cause or the result of secondary hyperparathyroidism, a direct correlation between cell proliferation and decrease of calcium-sensing receptor has been suggested [67,68]. Thus, glands with nodular hyperplasia are less responsive to the suppressive effect of ambient calcium. This may partially explain the empirical finding of high PTH levels in the presence of hypercalcemia in patients with nodular hyperplasia. The progression of parathyroid hyperplasia is summarized in Fig. 2. It is of note that some enlarged glands smaller than 0.5 cm3 may be composed of nodular hyperplasia [61]. Nodule formation may be recognized by the shape of the glands detected by the latest models of ultrasonography devices [70]. Furthermore, Onoda et al. recently reported that positive blood supply detected inside the gland was highly suggestive of nodular hyperplasia [71].

III. MANAGEMENT OF SEVERE HYPERPARATHYROIDISM REFRACTORY TO MEDICAL THERAPY Prevention of parathyroid hyperplasia from the early phase of chronic renal failure is the most important

1824

MASAFUMI FUKAGAWA AND KIYOSHI KUROKAWA

Normal parathyroid

Diffuse hyperplasia

Responsive to medical therapy

Early nodularity in diffuse hyperplasia

Point of no-return?

Refractory to medical therapy Nodular hyperplasia

Indication for parathyroid intervention

FIGURE 2 Progression of parathyroid hyperplasia in chronic renal failure. Cells with more severe reduction of VDR receptor and calcium-sensing receptor within diffuse hyperplasia form small nodules leading to nodular hyperplasia.

strategy for the management of secondary hyperparathyroidism in chronic renal failure. This can be achieved by dietary phosphate restriction and the early use of phosphate binders and cautious use of active vitamin D sterols as described in Chapter 76. Recent data suggest that the direct effects of phosphate on parathyroid cells is especially important in the early phase of chronic renal failure [72].

A. Selective Percutaneous Ethanol Injection Therapy (PEIT) Although the introduction of new vitamin D analogs and calcimimetics may be promising in the treatment of secondary hyperparathyroidism, what can be done for patients with nodular hyperplasia? Do they have any choice other than surgical parathyroidectomy [73,74].

The ongoing discussion indicates that small glands composed of diffuse hyperplasia should still be responsive to calcitriol even in such patients. However, what about the patients that have already progressed to nodular hyperplasia? For patients with nodular hyperplasia, two new techniques have been established. The first technique is the selective percutaneous ethanol injection therapy (PEIT) [75–77]. The second technique is direct vitamin D injection therapy (see below). In PEIT, glands with nodular hyperplasia are “selectively” destroyed by ethanol injection under ultrasonographic guidance. Other glands with diffuse hyperplasia are then controlled by medical therapy (Fig. 3). Recently, this technique has become more powerful and safer than ever and has become widely used, especially in Japan. According to the guideline by Japanese Society for Parathyroid Intervention [78],

CHAPTER 103 Renal Failure and Secondary Hyperparathyroidism

1825

before PEIT in patients with more than three glands of critical size. Recurrent nerve palsy due to leakage of ehanol is one of the most important complications of PEIT [77]. It is also suspected that adhesions in the tissue surrounding the parathyroids may be caused by leakage of ethanol. This could be a major problem if surgical parathyroidectomy will be needed in the future. With the routine use of color Doppler flow mapping by ultrasonography, the volume of ethanol used for PEIT has become minimal, leading to the lower rate of such complications [79]. Nevertheless, skilled operators and appropriate equipments are certainly required for successful and safe PEIT [78]. Ethanol vitamin D

B. Direct Vitamin D Injection Therapy

Diffuse hyperplasia Nodular hyperplasia

FIGURE 3 Parathyroid intervention under ultrasonographic guidance. Glands with nodular hyperplasia (hatched circle) are treated by intervention using percutaneous ethanol injection therapy (PEIT). Other glands with diffuse hyperplasia (open circle) are then controlled by medical therapy.

patients having one or two glands with nodular hyperplasia are the best candidates for PEIT. After the initial ethanol injection, PTH levels and recurrence should be monitored carefully. More importantly, after the successful destruction of nodular hyperplasia, residual glands with diffuse hyperplasia should be managed by appropriate medical therapy, including dietary phosphorus restriction. Thus, good compliance of the patients to medical therapy and regular check-ups after PEIT is essential. In the largest study by Kakuta et al., 46 patients were treated by selective PEIT on an outpatient basis followed by appropriate medical therapy. PTH levels in 80% of the patients remained within the target range at one year after initial treatment [79]. Long-term follow-up (three years) after PEIT has been also reported by this group [80]. Failure of PTH suppression despite successful ablation of glands with nodular hyperplasia suggests the existence of another gland containing nodular hyperplasia beyond the reach of ultrasonography [81]. If ectopic glands are recognized before the procedure is performed, initial surgical parathyroidectomy is indicated. Thus, it may be reasonable to search for ectopic glands

The second technique for treating nodular hyperplasia is direct calcitriol injection therapy under ultrasonographic guidance (PCIT) [82]. In this therapy, very high local concentration of 1,25(OH)2D3 or other vitamin D metabolites is achieved exclusively in injected parathyroid glands. As already reported, direct calcitriol injection therapy not only suppressed PTH secretion, but also restored the responsiveness to medical therapy. Such effects have been confirmed by other studies with different protocols [83–85]. Furthermore, direct injections of 22-oxacalcitriol have also been tried with favorable results [86]. In contrast to PEIT, the risk of recurrent nerve palsy is extremely low with PCIT. However, a recent report suggested that direct injection of 22-oxacalcitriol might evoke inflammation, resulting in adhesions of the surrounding tissue [87]. The use of vitamin D analogs is promising and such risks should be avoidable with improvements in technique. Future development may allow direct injection of new calcimimetics and even adenovirus-mediated gene transfer to treat parathyroid hyperplasia, as has been shown recently in animal models [88].

C. Regression of Parathyroid Hyperplasia: Is It Really Possible? The cell cycle of parathyroid cells is usually very slow, even in the hyperplastic glands [89]. Although prevention of parathyroid hyperplasia by several therapeutic modalities has been demonstrated in rat models of chronic kidney disease [90], suppression of PTH secretion may not lead to the complete normalization of parathyroid cell function including proliferation site in patients with secondary hyperparathyroidism [91,92].

1826 Thus, it is still an unsettled issue whether hyperplastic parathyroid glands do regress after proper medical therapy or after kidney transplantation. Regression of parathyroid hyperplasia has been reported in chronic dialysis patients treated by oral calcitriol pulse therapy [93–95], although controversial data have been also reported [96]. Such a regression was observed not only in cases with successful PTH suppression, but also in small glands even in cases without significant suppression of PTH [55,97]. Thus, in our opinion it is reasonable to conclude that glands with diffuse hyperplasia regress after effective medical therapy. In contrast, as discussed above, glands with nodular hyperplasia do not regress except for a few cases in which spontaneous apoplexy of the gland was suspected [98,99]. Due to the lack of established parathyroid cell lines for in vitro studies, mechanisms of regression have not yet been satisfactorily elucidated [100]. In order to achieve regression of hyperplastic glands, suppression of cell proliferation may not be sufficient. Negative cell balance by increased apoptosis may be needed. However, in rats, it has been very hard to demonstrate the apoptosis of parathyroid cells, which takes place in a limited number of cells during cell turnover [101–104]. Moreover, interpretation of apoptotic cells demonstrated in surgically removed parathyroid glands in dialysis patients has been controversial [100]. In a 1977 report by Henry et al. [105], reduction of parathyroid cell number was clearly demonstrated in three-month-old vitamin D–deficient chickens treated with vitamin D replacement. In contrast, 1,25(OH)2D3 treatment suppressed parathyroid cell proliferation, but did not reverse hyperplasia in experimental uremia, as demonstrated by Szabo et al. [106]. In recent animal studies by Lewin et al. [107,108], hyperparathyroidism induced by long-term uremia returned to normal following kidney transplantation. However, parathyroid hyperplasia was persistent. Such a suppression of PTH secretion with persistent hyperplasia has also been demonstrated in rat models of secondary hyperparathyroidism induced by high phosphorus diet, by switching to low phosphorus diet. In these animal models, reversal of reduced VDR or calcium-sensing receptor has not been confirmed at least in the short term. As discussed above, it has been suggested recently that regression of nodular hyperplasia may be induced by direct vitamin D injection therapy, originally performed with calcitriol [82]. By injecting directly into enlarged glands under ultrasonography, very high local concentration of vitamin D or analog can be achieved transiently. Shiizaki et al. recently reported that direct injection of 22-oxa-calcitriol solution into enlarged glands in patients leads to the regression of hyperplasia [84].

MASAFUMI FUKAGAWA AND KIYOSHI KUROKAWA

By repeated parathyroid biopsy before and after the therapy, they clearly demonstrated the induction of apoptosis of parathyroid cells in the injected glands. They also suggested that such a regression was associated with up-regulation of VDR and the calcium-sensing receptor. These data suggest that direct vitamin D injection therapy not only induces apoptosis of parathyroid cells, but also restores the responsiveness of residual parathyroid cells to medical therapy, leading to normalization of parathyroid hyperplasia. It may be possible that such specific effects of vitamin D on parathyroid cells may also be achieved by oral or intravenous preparations, if vitamin D analogs with these specific actions can be designed in the future. It is also of note that increased 25-hydroxyvitamin D3 1α-hydroxylase and reduced 25-hydroxyvitamin D3 24-hydroxylase expression have been reported in parathyroid tumors [109]. Thus, parathyroid is not only a target organ of vitamin D, but it also metabolizes vitamin D. Since parathyroid glands possess 1α-hydroxylase, it may become possible to develop new vitamin D metabolites that use this system to be activated only in parathyroid.

IV. FUTURE ROLES OF VITAMIN D ANALOGS IN CHRONIC RENAL FAILURE A. Design of Vitamin D Analogs with Specific Actions on Specific Tissues in Chronic Renal Failure The parathyroid glands are not the only target organ of vitamin D therapy in patients with chronic renal failure. The skin and the immune system are other examples; however, the role of vitamin D treatment on these systems, as well as other organs, has not been fully clarified yet [110,111]. A recent report suggests that paricalcitol treatment leads to better survival than calcitriol treatment in chronic dialysis patients [112]. It is still unclear whether such a difference is due to the less calcemic effect of paricalcitol or to its effects on other organ systems. Bone is a representative classic target organ of vitamin D. Thus, different effects of 22-oxacalcitriol versus calcitriol on bone turnover have been noted in animal models of chronic renal failure [113] (see Chapter 86). As recently suggested, the different effects of vitamin D analogs on bone, seen in in vitro and in vivo experiments, might be clues that help in the future elucidation of the mechanism for the differential actions of analogs in various tissues [114]. 22-oxacalcitriol was originally developed as an analog

1827

CHAPTER 103 Renal Failure and Secondary Hyperparathyroidism

with potent activity on the differentiation of leukemic cell lines [115]. Such an activity has also been examined with paricalcitol [116]. Recently, it has been demonstrated that different vitamin D analogs utilize specific cofactors for target gene regulation [117]. Different cofactors bind to different genes, evoking different actions on the same cell. Thus, it is expected that design of vitamin D analogs with differential actions in specific organs will become possible [118]. For example, it may be possible to design analogs that specifically induce apoptosis of parathyroid cells as well as less calcemic analogs.

B. Possible Treatment of Chronic Kidney Disease by Vitamin D Analogs The kidney is not only the site of active vitamin D production, but also is its target organ. As intensively discussed in several previous chapters, vitamin D metabolites modulate the activity of the enzymes involved in vitamin D synthesis and degradation. Calcium-binding proteins [119] are also induced by vitamin D in the distal tubules of the kidney. VDRs have been identified in various parts of the kidney and no doubt are regulated by vitamin D metabolites [120]. Inhibition of renal cell proliferation by vitamin D was initially demonstrated in renal cell carcinoma lines [121]. It has also been shown that 1,25(OH)2D3 diminished 3H-thymidine incorporation, cell counts, and TGF-β secretion into the supernatant of cultured proximal tubular cell lines [122–124] and in cultured human mesangial cells [125]. Regulation of mesangial cell smooth muscle phenotype has also been suggested [126]. In vivo, 1,25(OH)2D3 reduced renal weight, protein content, DNA content, and the number of mitoses in the remnant kidney with compensatory hypertrophy after uninephrectomy [127]. On the contrary, 1,25(OH)2D3 may induce type IV collagen synthesis, possibly through up-regulation of TGF-β type II receptor [126] and upregulation of protein-1 [128]. It may be possible that vitamin D ameliorates glomerular injury seen in chronic kidney disease, although the effects may depend on the phase of renal injury. It has recently been shown that 1,25(OH)2D3 inhibited progressive glomerulosclerosis in subtotally nephrectomized rats [129] and reduced proteinuria, glomerular hyper-cellularity and inflammatory infiltration in anti-Thy-1.1 nephritis [130]. Similar suppressive effects have been also demonstrated with retinoic acids [131,132]. These studies suggest the possibility that vitamin D may alter the rate of progression of CKD. In contrast, there has been a concern that oral vitamin D treatment

in CKD patients may increase the risk of accelerating the progression of renal dysfunction by increasing urinary calcium excretion. In this respect, the lesscalcemic vitamin D analogs may be more suitable for this purpose [133]. 22-oxa-calcitriol is one such less-calcemic vitamin D compound used for the treatment of severe secondary hyperparathyroidism in chronic dialysis patients [134], as extensively reviewed in Chapter 86. It has recently been shown that 22-oxa-calcitriol also effectively ameliorated glomerular sclerosis in two rat models of chronic kidney disease without affecting calcium and phosphorus levels [135,136]. Although further studies are needed, a vitamin D analog with such properties may be a promising agent for the treatment of chronic kidney disease in the near future.

Acknowledgements This work was partly supported by grants from Renal Osteodystrophy Foundation and from Renal Anemia Foundation, Japan. Authors are grateful to Ms Tomoko Nii-Kono for her secretarial assistance and art work.

References 1. Coburn JW, Llach F 1994 Renal osteodystrophy. In “Clinical Disorders of Fluid and Electrolyte Metabolism, 5th ed” (R. Narins, Ed), pp. 1299–1377, McGraw Hill Book Co.: New York, New York. 2. Takeyama K, Kitanaka S, Sato T, Kobori M, Yanagisawa J, Kato S 1997 25-Hydroxyvitamin D3 1alpha-hydroxylase and vitamin D synthesis. Science 277:1827–1830. 3. Murayama A, Takeyama K, Kitanaka S, Kodera Y, Kawaguchi Y, Hosoya T, Kato S 1999 Positive and negative regulations of the renal 25-hydroxyvitamin D3 1alphahydroxylase gene by parathyroid hormone, calcitonin, and 1alpha,25(OH)2D3 in intact animals. Endocrinoogy 140: 2224–2231. 4. Miller WL, Portale AA 2003 Vitamin D biosynthesis and vitamin D 1alpha-hydroxylase deficiency. Endocr Dev 6:156–174. 5. Portale AA, Booth BE, Halloran BP, Morris RC Jr. 1984 Effect of dietary phosphorus on circulating concentrations of 1,25-dihydroxyvitamin D and immunoreactive parathyroid hormone in children with moderate renal insufficiency. J Clin Invest 73:1580–1589. 6. Zhang MY, Wang X, Wang JT, Compagnone NA, Mellon SH, Olson JL, Tenenhouse HS, Miller WL, Portale AA 2002 Dietary phosphorus transcriptionally regulates 25-hydroxyvitamin D-1alpha-hydroxylase gene expression in the proximal renal tubule. Endocrinology 143:587–595. 7. Shimada T, Mizutani S, Muto T, Yoneya T, Hino R, Takeda S, Takeuchi Y, Fujita T, Fukumoto S, Yamashita T 2001 Cloning and characterization of FGF23 as a causative factor of tumorinduced osteomalacia. Proc Natl Acad Sci USA 98:6500–6505.

1828 8. Shigematsu T, Kazama JJ, Yamashita T, Fukumoto S, Hosoya T, Gejyo F, Fukagawa M 2004 Possible involvement of circulating fibroblast growth factor-23 in the development of secondary hyperparathyroidism associated with renal insufficiency. Am J Kidney Dis 44:250–256. 9. Cannata-Andia JB, Alonzo CG 2002 Vitamin D deficiency: a neglected aspect of disturbed calcium metabolism in renal failure. Nephrol Dial Transplant 17:1875–1878. 10. Ishimura E, Nishizawa Y, Inaba M, Matsumoto N, Emoto M, Kawagishi T, Shoji S, Okuno S, Kim M, Miki T, Morii H 1999 Serum levels of 1,25-dihydroxyvitamin D, 24,25-dihydroxyvitamin D, and 25-hydroxyvitamin D in nondialyzed patients with chronic renal failure. Kidney Int 55:1019–1027. 11. Mittal SK, Dash SC, Tiwari SC, Agarwal SK, Saxena S, Fishbane S 1999 Bone histology in patients with nephrotic syndrome and normal renal function. Kidney Int 55:1912–1919. 12. Thomas MK, Lloyd-Jones DM, Thadhani RI, Shaw AC, Deraska DJ, Kitch BT, Vamvakas EC, Dick IM, Prince RL, Finkelstein JS 1998 Hypovitaminosis D in medical inpatients. N Engl J Med 338:777–783. 13. Takemoto F, Shinki T, Yokoyama K, Inokami T, Hara S, Yamada A, Kurokawa K, Uchida S 2003 Gene expression of vitamin D hydroxylase and megalin in the remnant kidney of nephrectomized rats. Kidney Int 64:414–420. 14. Nykjaer A, Dragun D, Walther D, Vorum H, Jacobsen C, Herz J, Melsen F, Christensen EI, Willnow TE 1999 An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) vitamin D3. Cell 96:507–515. 15. Slatopolsky E, Finch J, Denda M, Ritter C, Zhong M, Dusso A, MacDonald PN, Brown AJ 1996 Phosphorus restriction prevents parathyroid gland growth. High phosphorus directly stimulates PTH secretion in vitro. J Clin Invest 97:2534–2540. 16. Almaden Y, Canalejo A, Hernandez A, Ballesteros E, Garcia-Navarro S, Torres A, Rodriguez M 1996 Direct effect of phosphorus on PTH secretion from whole rat parathyroid glands in vitro. J Bone Miner Res 11:970–976. 17. Yi H, Fukagawa M, Yamato H, Kumagai M, Watanabe T, Kurokawa K 1995 Prevention of enhanced parathyroid hormone secretion, synthesis and hyperplasia by mild dietary phosphorus restriction in early chronic renal failure in rats: possible direct role of phosphorus. Nephron 70:242–248. 18. Malluche H, Faugere MC 1990 Renal bone disease 1990: An unmet challenge for the nephrologist. Kidney Int 38:193–211. 19. Naveh-Many T, Marx R, Keshet E, Pike JW, Silver J 1990 Regulation of 1,25-dihydroxyvitamin D3 receptor gene expression by 1,25-dihydroxyvitamin D3 in the parathyroid in vitro. J Clin Invest 86:1968–1975. 20. Silver J, Kilav R, Naveh-Many T 2002 Mechanisms of secondary hyperparathyroidism. Am J Physiol Renal Physiol 283:F367–F376. 21. Fukagawa M 1999 Cell biology of parathyroid hyperplasia in uremia. Am J Med Sci 317:377–382. 22. Slatopolsky EA, Weerts C, Thielan J, Horst R, Harter H, Martin KJ 1984 Marked suppression of secondary hyperparathyroidism by intravenous administration of 1,25-dihydroxycholecalciferol in uremic patients. J Clin Invest 74:2136–2143. 23. Andress DL, Norris KC, Coburn JW, Slatopolsky EA, Sherrard DJ 1989 Intravenous calcitriol in the treatment of refractory osteitis fibrosa of chronic renal failure. N Engl J Med 321:274–279. 24. Tsukamoto Y, Nomura M, Takahashi Y, Takagi Y, Yoshida A, Nagaoka T, Togashi K, Kitawada R, Marumo F 1990 The ‘Oral 1,25-dihydroxyvitamin D3 pulse therapy’ in hemodialysis

MASAFUMI FUKAGAWA AND KIYOSHI KUROKAWA

25. 26. 27. 28. 29.

30. 31.

32. 33. 34. 35. 36.

37.

38.

39.

40.

41. 42.

patients with severe secondary hyperparathyroidism. Nephron 56:368–373. Fukagawa M, Fukuda N, Yi H, Kurokawa K 1995 Resistance to calcitriol as a cause of parathyroid hyperfunction in chronic renal failure. Nephrol Dial Transplant 10:316–319. Fukagawa M, Kaname S, Igarashi T, Ogata E, Kurokawa K 1991 Regulation of parathyroid hormone synthesis in chronic renal failure in rats. Kidney Int 39:874–881. Korkor AB 1987 Reduced binding of [3H]1,25-dihydroxyvitamin D3 in the parathyroid glands of patients with renal failure. N Engl J Med 316:1573–1577. Merke J, Hugel U, Zlotowski A, Szabo A, Bommer J, Mall G, Ritz E 1987 Diminished parathyroid 1,25(OH)2D3 receptors in experimental uremia. Kidney Int 32:350–353. Brown A, Dusso A, Lopez-Hilker S, Lewis-Finch J, Grooms P, Slatopolsky EA 1989 1,25(OH)2D receptors are decreased in parathyroid glands from chronically uremic dogs. Kidney Int 35:19–23. Dusso AS 2003 Vitamin D receptor: Mechanisms for vitamin D resistance in renal failure. Kidney Int (Suppl) 85:6–9. Sawaya BP, Koszewski NJ, Qi Q, Langub MC, Monier-Faugere MC, Malluche HH 1997 Secondary hyperparathyroidism and vitamin D receptor binding to vitamin D response elements in rats with incipient renal failure. J Am Soc Nephrol 8:271–278. Hsu CH, Patel SR 1997 Uremic toxins and vitamin D metabolism. Kidney Int (Suppl) 62:S65–S68. Patel SR, Ke HQ, Vanholder R, Koenig RJ, Hsu CH 1995 Inhibition of calcitriol receptor binding to vitamin D response elements by uremic toxins. J Clin Invest 96:50–59. Patel SR, Koenig RJ, Hsu CH 1996 Effect of Schiff base formation on the function of the calcitriol receptor. Kidney Int 50:1539–1545. Patel SR, Xu Y, Koenig RJ, Hsu CH 1997 Effect of glyoxylate on the function of the calcitriol receptor and vitamin D metabolism. Kidney Int 52:39–44. Canalejo A, Almaden Y, De Smet R, Glorieux G, Garfia B, Luque F, Vanholder R, Rodriguez M 2003 Effects of uremic ultrafiltrate on the regulation of the parathyroid cell cycle by calcitriol. Kidney Int 63:732–737. Sela-Brown A, Russell J, Koszewski NJ, Michalak M, Naveh-Many T, Silver J 1998 Calreticulin inhibits vitamin D’s action on the PTH gene in vitro and may prevent vitamin D’s effect in vivo in hypocalcemic rats. Mol Endocrinol 12: 1193–1200. Brown AJ, Zhong M, Finch J, Ritter C, Slatopolsky E 1995 The roles of calcium and 1,25-dihydroxyvitamin D3 in the regulation of vitamin D receptor expression by rat parathyroid glands. Endocrinology 136:1419–1425. Garfia B, Canadillas S, Canalejo A, Luque F, Siendones E, Quesada M, Almaden Y, Aguilera-Tejero E, Rodriguez M 2002 Regulation of parathyroid vitamin D receptor expression by extracellular calcium. J Am Soc Nephrol 13:2945–2952. Strom M, Sandgrin ME, Brown TA, DeLuca HF 1989 1,25-dihydroxyvitamin D3 up-regulates the 1,25-dihydroxyvitamin D3 receptor in vivo. Proc Natl Acad Sci USA 86: 9770–9773. Shvil Y, Naveh-Many T, Brach P, Silver J 1990 Regulation of parathyroid cell gene expression in experimental uremia. J Am Soc Nephrol 1:99–104. Koyama H, Nishizawa Y, Inaba M, Hino M, Prahl JM, DeLuca HF, Morii H 1994 Impaired homologous up-regulation of vitamin D receptor in rats with chronic renal failure. Am J Physiol 266:F706–F712.

CHAPTER 103 Renal Failure and Secondary Hyperparathyroidism

43. Denda M, Finch J, Brown AJ, Nishii Y, Kubodera N, Slatopolsky E 1996 1,25-dihydroxyvitamin D3 and 22-oxacalcitriol prevent the decrease in vitamin D receptor content in the parathyroid glands of uremic rats. Kidney Int 50:34–39. 44. Reichel H, Szabo A, Uhl J, Pesian S, Schmultz A, SchmidtGayk H, Ritz E 1993 Intermittent versus continuous administration of 1,25-dihydroxyvitamin D3 in experimental renal hyperparathyroidism. Kidney Int 44:1259–1265. 45. Block GA, Hulbert-Shearon TE, Levin NW, Port FK 1998 Association of serum phosphorus and calcium × phosphate product with mortality risk in chronic hemodialysis patients: a national study. Am J Kidney Dis 31:607–617. 46. Martin KJ, Gonzalez EA, Gellens M, Hamm LL, Abboud H, Lindberg J 1998 19-Nor-1-alpha-25-dihydroxyvitamin D2 (Paricalcitol) safely and effectively reduces the levels of intact parathyroid hormone in patients on hemodialysis. J Am Soc Nephrol 9:1427–1432. 47. Brown AJ, Ritter CR, Finch JL, Morrissey J, Martin KJ, Murayama E, Nishii Y, Slatopolsky E 1989 The noncalcemic analog of vitamin D, 22-oxacalcitriol, suppresses parathyroid hormone synthesis and secretion. J Clin Invest 84:728–732. 48. Kazama JJ, Gejyo F, Shigematsu T, Fukumoto S, Yamashita T, Nakanishi S, Fukagawa M 2003 Circulating fibroblast growth factor-23 (FGF-23) levels predict the development of secondary hyperparathyroidism in long-term dialysis patients (abstract). J Am Soc Nephrol, 14:209A. 49. Drüeke TB 2000 Cell biology of parathyroid gland hyperplasia in chronic renal failure. J Am Soc Nephrol 11: 1141–1152. 50. Malmaeus J, Glimelius L, Johansson G, Akerstrom G, Ljunghall S 1984 Parathyroid pathology in hyperparathyroidism secondary to chronic renal failure. Scand J Urol Nephrol 18:157–166. 51. McCarron DA, Muther RS, Lenfesty B, Bennett WM 1982 Parathyroid function in persistent hyperparathyroidism: Relationship to gland size. Kidney Int 22:662–670. 52. Wallfelt CH, Larsson R, Gylfe E, Ljunghall S, Rasted J, Akerstrom G 1988 Secretory disturbance in hyperplastic parathyroid nodules of uremic hyperparathyroidism: Implication for parathyroid autoplantation. World J Surg 12:431–438. 53. Felsenfeld AJ, Llach F 1993 Parathyroid function in chronic renal failure. Kidney Int 43:771–789. 54. Delmez JA, Tindra C, Grooms P, Dusso A, Windus DW, Slatopolsky E 1989 Parathyroid hormone suppression by intravenous 1,25(OH)2 vitamin D. A role for increased sensitivity to calcium. J Clin Invest 83:1349–1355. 55. Fukagawa M, Kitaoka M, Yi H, Fukuda N, Matsumoto T, Ogata E, Kurokawa K 1994 Serial evaluation of parathyroid size by ultrasonography is another useful marker for the long-term prognosis of calcitriol pulse therapy in chronic dialysis patients. Nephron 68:221–228. 56. Tominaga Y, Sato K, Tanaka Y, Numano M, Uchida K, Takagi H 1995 Histopathology and pathophysiology of secondary hyperparathyroidism due to chronic renal failure. Clin Nephrol 44:S42–S47. 57. Fukuda N, Tanaka H, Tominaga Y, Fukagawa M, Kurokawa K, Seino Y 1993 Decreased 1,25-dihydroxyvitamin D3 receptor density is associated with a more severe form of parathyroid hyperplasia in chronic uremic patients. J Clin Invest 92:1436–1443. 58. Åkerstrom G, Malmaeus J, Grimelius L, Ljunghall S, Bergstrom R 1984 Histological changes in parathyroid

59.

60.

61. 62.

63.

64.

65.

66.

67.

68. 69.

70. 71.

72.

73. 74.

1829 glands in subclinical and clinical renal disease. Scand J Urol Nephrol 18:75–84. Tominaga Y, Grimelius L, Falkmer UG, Johansson H, Falkmer S 1991 DNA ploidy pattern of parathyroid parenchymal cells in renal secondary hyperparathyroidism with relapse. Analyt Cell Pathol 3:325–333. Tominaga Y, Tanaka Y, Sato K, Numano M, Uchida K, Falkmer U, Grimelius L, Johansson H, Takagi H 1992 Recurrent renal hyperparathyroidism and DNA analysis of autografted parathyroid tissue. World J Surg 16:595–603. Tominaga Y, Takagi H 1996 Molecular genetics of hyperparathyroid disease. Current Opinions Nephrol Hypertens 5:336–341. Tokumoto M, Tsuruya K, Fukuda K, Kanai H, Kuroki S, Hirakata H 2002 Reduced p21, p27 and vitamin D receptor in the nodular hyperplasia in patients with advanced secondary hyperparathyroidism. Kidney Int 62:1196–1207. Brown EM, Wilson RE, Eastman RC, Pallotta J, Marynick SP 1982 Abnormal regulation of parathyroid hormone release by calcium in secondary hyperparathyroidism due to chronic renal failure. J Clin Endocrinol Metab 54:172–179. Carling T, Rastad J, Szabo E, Westin G, Akerstrom G 2000 Reduced parathyroid vitamin D receptor messenger ribonucleic acid levels in primary and secondary hyperparathyroidism. J Clin Endocrinol Metab 85:2000–2003. Gogusev J, Duchambon P, Hory B, Giovannini M, Goureau Y, Sarfati E, Drueke TB 1997 Depressed expression of calcium receptor in parathyroid gland tissue of patients with hyperparathyroidism. Kidney Int 51:328–336. Brown AJ, Ritter CS, Finch JL, Slatopolsky EA 1999 Decreased calcium-sensing receptor expression in hyperplastic parathyroid glands of uremic rats: role of dietary phosphate. Kidney Int 55:1284–1292. Yano S, Sugimoto T, Tsukamoto T, Chihara K, Kobayashi A, Kitazawa S, Maeda S, Kitazawa R 2000 Association of decreased calcium-sensing receptor expression with proliferation of parathyroid cells in secondary hyperparathyroidism. Kidney Int 58:1980–1986. Ritter CS, Finch JL, Slatopolsky EA, Brown AJ 2001 Parathyroid hyperplasia in uremic rats precedes downregulation of the calcium receptor. Kidney Int 60:1737–1744. Canaff L, Hendy GN 2002 Human calcium-sensing receptor gene. Vitamin D response elements in promoters P1 and P2 confer transcriptional responsiveness to 1,25-dihydroxyvitamin D. J Biol Chem 277:30337–30350. Fukagawa M, Kitaoka M, Kurokawa K 1997 Value of parathyroid sonography in secondary hyperparathyroidism (letter). Nephrol Dial Transplant 12:2461. Onoda N, Kurihara S, Sakurai Y, Owada K, Osono E, Adachi H, Suga M, Yoneshima H 2003 Evaluation of blood supply to the parathyroid glands in secondary hyperparathyroidism compared with histopathology. Nephrol Dial Transplant 18(Suppl 3):iii34–iii37. Almaden Y, Hernandez A, Torregrosa V, Canalejo A, Sabate L, Fernandez-Cruz L, Campistol JM, Torres A, Rodriguez M 1998 High phosphate level directly stimulates parathyroid hormone secretion and synthesis by human parathyroid tissue in vitro. J Am Soc Nephrol 9:1845–1852. Fukagawa M, Kazama JJ, Shigematsu T 2002 Management of the patients with advanced secondary hyperparathyroidism: the Japanese approach. Nephrol Dial Transplant 17:1553–1557. Fukagawa M, Tominaga Y, Kitaoka M, Kakuta T, Kurokawa K 1999 Medical and surgical aspects of parathyroidectomy. Kidney Int 56(Suppl 73):S65–S69.

1830 75. Kitaoka M, Fukagawa M, Ogata E, Kurokawa K 1994 Reduction of functioning parathyroid cell mass by ethanol injection in chronic dialysis patients. Kidney Int 46:1110–1117. 76. Giangrande A, Castiglioni A, Sorbiati L, Allaria P 1992 Ultrasound guided percutaneous fine needle ethanol injection into parathyroid glands in secondary hyperparathyroidism. Nephrol Dial Transplant 7:412–421. 77. Kitaoka M 2003 Ultrasonographic diagnosis of parathyroid glands and percutaneous ethanol injection therapy. Nephrol Dial Transplant 18(Suppl 3):iii27–iii30. 78. Fukagawa M, Kitaoka M, Tominaga Y, Akizawa T, Kurokawa K, for Japanese Society for Parathyroid Intervention 2003 Guidelines for percutaneous ethanol injection therapy (PEIT) of the parathyroid glands in chronic dialysis patients. Nephrol Dial Transplant 18(Suppl 3):31–33. 79. Kakuta T, Fukagawa M, Fujisaki T, Hida M, Suzuki H, Sakai H, Kurokawa K, Saito A 1999 Prognosis of parathyroid function after successful percutaneous ethanol injection therapy (PEIT) guided by color Doppler flow mapping in chronic dialysis patients. Am J Kidney Dis 33:1091–1099. 80. Tanaka R, Kakuta T, Fujisaki T, Tanaka S, Sakai H, Kurokawa K, Saito A 2003 Long-term (three years) prognosis of parathyroid function in chronic dialysis patients after percutaneous ethanol injection therapy guided by color Doppler ultrasonography. Nephrol Dial Transplant 18: (Suppl 3):iii58–iii61. 81. Fukagawa M, Nakanishi S 2003 Role of parathyroid intervention in the management of secondary hyperparathyroidism in chronic renal failure. Nephrol Dial Transplant 18 (Suppl 3):23–26. 82. Kitaoka M, Fukagawa M, Kurokawa K 1995 Direct injection of calcitriol into parathyroid hyperplasia in chronic dialysis patients with severe parathyroid hyperfunction. Nephrology 1:563–568. 83. Kitaoka M, Onoda N, Kitamura H, Koiwa M, Tanaka M, Fukagawa M 2003 Percutaneous calcitriol injection therapy (PCIT) for secondary hyperparathyroidism: multicentre trial. Nephrol Dial Transplant 18(Suppl 3):iii38–iii41. 84. Shiizaki K, Negi S, Mizobuchi M, Hatamura I, Narukawa N, Sakaguchi T, Kitabata Y, Sumikado S, Akizawa T 2003 Effect of percutaneous calcitriol injection therapy on secondary hyperparathyroidism in uremic patients. Nephrol Dial Transplant 18(Suppl 3):iii42–iii46. 85. Nakanishi S, Yano S, Nomura R, Tsukamoto T, Shimizu Y, Shin J, Fukagawa M 2003 Efficacy of direct injection of calcitriol into the parathyroid glands in uraemic patients with moderate to severe secondary hyperparathyroidism. Nephrol Dial Transplant 18(Suppl 3):iii47–iii49. 86. Shiizaki K, Hatamura I, Negi S, Narukawa N, Mizobuchi M, Sakaguchi T, Ooshima A, Akizawa T 2003 Percutaneous maxacalcitol injection therapy regresses hyperplasia of parathyroid and induces apoptosis in uremia. Kidney Int 64:992–1003. 87. Yamamoto H, Katoh N, Takeyama H, Ikeda M, Yokoyama K, Shigematsu T, Kawaguchi Y, Hosoya T 2003 Surgical verification of percutaneous maxacalcitriol injection therapy on enlarged parathyroid glands in chronic dialysis patients. Nephrol Dial Transplant 18(Suppl 3):iii50–iii52. 88. Iwasaki Y, Kakuta T, Haruguchi H, Fukuda N, Kurokawa K, Fukagawa M 2003 Adenovirus-mediated functional gene transfer into parathyroid cells in vivo and in vitro. Nephrol Dial Transplant 18(Suppl 3):iii18–iii22. 89. Parfitt AM 1982 Hypercalcemic hyperparathyroidism following renal transplantation: Differential diagnosis, management, and implications for cell population control in the parathyroid gland. Mineral Electrolyte Metab 8:92–112.

MASAFUMI FUKAGAWA AND KIYOSHI KUROKAWA

90. Wada M, Nagano N, Furuya Y, Chin J, Nemeth EF, Fox J 2000 Calcimimetic NPS R-568 prevents parathyroid hyperplasia in rats with severe secondary hyperparathyroidism. Kidney Int 57:50–58. 91. Takahashi F, Denda M, Finch JL, Brown AJ, Slatopolsky E 2002 Hyperplasia of the parathyroid gland without secondary hyperparathyroidism. Kidney Int 6:1332–1338. 92. Ritter CS, Martin DR, Lu Y, Slatopolsky E, Brown AJ 2002 Reversal of secondary hyperparathyroidism by phosphate restriction restores parathyroid calcium-sensing receptor expression and function. J Bone Miner Res 17: 2206–2213. 93. Fukagawa M, Okazaki R, Takano K, Kaname S, Ogata E, Kitaoka M, Harada S, Sekine N, Matsumoto T, Kurokawa K 1990 Regression of parathyroid hyperplasia by calcitriolpulse therapy in patients on long-term dialysis. N Engl J Med 323:421–422. 94. Hyodo T, Koumi T, Ueda M, Miyagawa I, Kodani K, Doi S, Ishibashi M, Takemoto M 1991 Can oral 1,25(OH)2D3 therapy reduce parathyroid hyperplasia? Nephron 59:171–172. 95. Cannella G, Bonucci E, Rolla D, Ballanti P, Moriero E, De Grandi R, Augeri C, Claudiani F, Di Maio G 1994 Evidence of healing of secondary hyperparathyroidism in chronically hemodialyzed uremic patients treated with long-term intravenous calcitriol. Kidney Int 46:1124–1132. 96. Quarles LD, Yohay DA, Carroll BA, Spritzer CE, Minda SA, Bartholomay D, Lobaugh BA 1994 Prospective trial of pulse oral versus intravenous calcitriol treatment of hyperparathyroidism in ESRD. Kidney Int 45:1710–1721. 97. Fukagawa M, Fukuda N, Yi H, Kitaoka M, Kurokawa K 1996 Derangement of parathyroid function in renal failure: Biological and clinical aspects. J Nephrol 9:219–224. 98. Nylen E, Shah A, Hall J 1996 Spontaneous remission of primary hyperparathyroidism from parathyroid apoplexy. J Clin Endocrinol Metab 81:1326–1328. 99. Chaffanjon PC, Chavanis N, Chabre O, Brichon PY 2003 Extracapsular hematoma of the parathyroid glands. World J Surg 27:14–17. 100. Drueke TB, Zhang P, Gogusev J 1997 Apoptosis: background and possible role in secondary hyperparathyroidism. Nephrol Dial Transplant 12:2228–2233. 101. Naveh-Many T, Rahamimov R, Livni N, Silver J 1995 Parathyroid cell proliferation in normal and chronic renal failure rats. The effects of calcium, phosphate, and vitamin D. J Clin Invest 96:1786–1793. 102. Zhang P, Duchambon P, Gogusev J, Nabarra B, Sarfati E, Bourdeau A, Drueke TB 2000 Apoptosis in parathyroid hyperplasia of patients with primary or secondary uremic hyperparathyroidism. Kidney Int 57:437–445. 103. Canalejo A, Almaden Y, Torregrosa V, Gomez-Villamandos JC, Ramos B, Campistol JM, Felsenfeld AJ, Rodriguez M 2000 The in vitro effect of calcitriol on parathyroid cell proliferation and apoptosis. J Am Soc Nephrol 11:1865–1872. 104. Jara A, Gonzalez S, Felsenfeld AJ, Chacon C, Valdivieso A, Jalil R, Chuaqui B 2001 Failure of high doses of calcitriol and hypercalcemia to induce apoptosis in hyperplastic parathyroid glands of azotaemic rats. Nephrol Dial Transplant 16:506–512. 105. Henry HL, Taylor AN, Norman AW 1977 Response of chick parathyroid glands to the vitamin D metabolites, 1,25-dihydroxycholecalciferol and 24,25-dihydroxycholecalciferol. J Nutr 107:1918–1926. 106. Szabo A, Merke J, Beier E, Mall G, Ritz E 1989 1,25(OH)2 vitamin D3 inhibits parathyroid cell proliferation in experimental uremia. Kidney Int 35:1049–1056.

CHAPTER 103 Renal Failure and Secondary Hyperparathyroidism

107. Lewin E, Wang W, Olgaard K 1997 Reversibility of experimental secondary hyperparathyroidism. Kidney Int 52:1232–1241. 108. Lewin E, Garfia B, Recio FL, Rodriguez M, Olgaard K 2003 Persistent down-regulation of calcium-sensing receptor mRNA in rat parathyroids when severe secondary hyperparathyroidism is reversed by an isogenic kidney transplantation. J Am Soc Nephrol 13:2110–2116. 109. Correa P, Segersten U, Hellman P, Akerstrom G, Westin G 2002 Increased 25-hydroxyvitamin D3 1alpha-hydroxylase and reduced 25-hydroxyvitamin D3 24-hydroxylase expression in parathyroid tumors—new perspects for treatment of hyperparathyroidism with vitamin D. J Clin Endocrinol Metab 87:5826–5829. 110. Mathieu C, Van Etten E, Gysemans C, Decallonne B, Kato S, Laureys J, Depovere J, Valckx D, Verstuyf A, Bouillon R 2001 In vitro and in vivo analysis of the immune system of vitamin D receptor knockout mice. J Bone Miner Res 16:2057–2065. 111. Kira M, Kobayashi T, Yoshikawa K 2003 Vitamin D and the skin. J Dermatol 30:429–437. 112. Teng M, Wolf M, Lowrie E, Ofsthun N, Lazarus JM, Thadhani R 2003 Survival of patients undergoing hemodialysis with paricalcitol or calcitriol therapy. N Engl J Med 349: 446–456. 113. Monier-Faugere MC, Geng Z, Friedler RM, Qi Q, Kubodera N, Slatopolsky E, Malluche HH 1999 22-oxacalcitriol suppresses secondary hyperparathyroidism without inducing low bone turnover in dogs with renal failure. Kidney Int 55:821–832. 114. Shibata T, Shira-Ishi A, Sato T, Masaki T, Masuda A, Hishiya A, Ishikura N, Higashi S, Uchida Y, Saito MO, Ito M, Ogata E, Watanabe K, Ikeda K 2002 Vitamin D hormone inhibits osteoclastogenesis in vivo by decreasing the pool of osteoclast precursors in bone marrow. J Bone Miner Res 17:622–629. 115. Abe J, Morikawa M, Miyamoto K, Kaiho S, Fukushima M, Miyaura C, Abe E, Nishii Y 1987 Synthetic analogs of vitamin D3 with an oxygen atom in the side chain skeleton. FEBS Lett 226:58–62. 116. Molnar I, Kute T, Willingham MC, Powell BL, Dodge WH, Schwartz GG 2003 19-nor-1alpha,25-dihydroxyvitamin D2 (paricalcitol): effects on clonal proliferation, differentiation, and apoptosis in human leukemic cell lines. J Cancer Res Clin Oncol 129:35–42. 117. Takeyama K, Masuhiro Y, Fuse H, Endoh H, Murayama A, Kitanaka S, Suzawa M, Yanagisawa J, Kato S 1999 Selective interaction of vitamin D receptor with transcriptional coactivators by a vitamin D analog. Mol Cell Biol 19:1049–1055. 118. Fujishima T, Kittaka A, Yamaoka K, Takeyama K, Kato S, Takayama H 2003 Synthesis of 2,2-dimethyl-1,25-dihydroxyvitamin D3: A-ring structural motif that modulates interactions of vitamin D receptor with transcriptional coactivators. Org Biomol Chem 1:1863–1869. 119. Christakos S, Barletta F, Huening M, Dhawan P, Liu Y, Porta A, Peng X 2003 Vitamin D target proteins: function and regulation. J Cell Biochem 88:238–244. 120. Iida K, Shinki T, Yamaguchi A, DeLuca HF, Kurokawa K, Suda T 1995 A possible role of vitamin D receptors in regulating vitamin D activation in the kidney. Proc Natl Acad Sci USA 92:6112–6116. 121. Nagakura K, Abe E, Suda T, Hayakawa M, Nakamura H, Tazaki H 1986 Inhibitory effect of 1alpha,25-dihydroxyvitamin D3 on the growth of the renal carcinoma cell line. Kidney Int 29:834–840.

1831

122. Weih M, Orth S, Weinreich T, Reichel H, Ritz E 1994 Inhibition of growth by calcitriol in a proximal tubular cell line (OK). Nephrol Dial Transplant 9:1390–1394. 123. Weinreich T, Landolt M, Booy C, Wuthrich R, Binswanger U 1999 1,25-dihydroxyvitamin D3 stimulates transforming growth factor-beta1 synthesis by mouse renal proximal tubular cells. Kidney Blood Press Res 22:99–105. 124. Weinreich T, Muller A, Wuthrich RP, Booy C, Binswanger U 1996 1,25-dihydroxyvitamin D3 and the synthetic vitamin D analog, KH 1060, modulate the growth of mouse proximal tubular cells. Kidney Blood Press Res 19:325–331. 125. Weinreich T, Merke J, Schonermark M, Reichel H, Diebold M, Hansch GM, Ritz E 1991 Actions of 1,25dihydroxyvitamin D3 on human mesangial cells. Am J Kidney Dis 18:359–366. 126. Abe H, Iehara N, Utsunomiya K, Kita T, Doi T 1999 A vitamin D analog regulates mesangial cell smooth muscle phenotypes in a transforming growth factor beta type II receptor-mediated manner. J Biol Chem 274: 20874–20878. 127. Matthias S, Busch R, Merke J, Mall G, Thomasset M, Ritz E 1991 Effects of 1,25(OH)2D3 on compensatory renal growth in the growing rat. Kidney Int 40:212–218. 128. Kobayashi T, Uehara S, Ikeda T, Itadani H, Kotani H 2003 Vitamin D3 up-regulated protein-1 regulates collagen expression in mesangial cells. Kidney Int 64:1632–1642. 129. Schwarz U, Amann K, Orth SR, Simonaviciene A, Wessels S, Ritz E 1998 Effect of 1,25(OH)2 vitamin D3 on glomerulosclerosis in subtotally nephrectomized rats. Kidney Int 53:1696–1705. 130. Panichi V, Migliori M, Taccola D, Filippi C, De Nisco L, Giovannini L, Palla R, Tetta C, Camussi G 2001 Effects of 1,25(OH)2D3 in experimental mesangial proliferative nephritis in rats. Kidney Int 60:87–95. 131. Datta PK, Lianos EA 1999 Retinoic acids inhibit inducible nitric oxide synthase expression in mesangial cells. Kidney Int 56:486–493. 132. Xu Q, Konta T, Kitamura M 2002 Retinoic acid regulation of mesangial cell apoptosis. Exp Nephrol 10:171–175. 133. Hirata M, Katsumata K, Endo K, Fukushima N, Ohkawa H, Fukagawa M 2003 In subtotally nephrectomized rats 22-oxacalcitriol suppresses parathyroid hormone with less risk of cardiovascular calcification or deterioration of residual renal function than 1,25(OH)2 vitamin D3. Nephrol Dial Trasplant 18:1770–1776. 134. Akizawa T, Suzuki M, Akiba T, Nishizawa Y, Ohashi Y, Ogata E, Slatopolsky E, Kurokawa K 2002 Long-term effect of 1,25-dihydroxy-22-oxavitamin D3 on secondary hyperparathyroidism in hemodialysis patients. One-year administration study. Nephrol Dial Transplant 17(Suppl) 10:28–36. 135. Makibayashi K, Tatematsu M, Hirata M, Fukushima N, Kusano K, Ohashi S, Abe H, Kuze K, Fukatsu A, Kita T, Doi T 2001 A vitamin D analog ameliorates glomerular injury on rat glomerulonephritis. Am J Pathol 158:1733–1741. 136. Hirata M, Makibayashi K, Katsumata K, Kusano K, Watanabe T, Fukushima N, Doi T 2002 22-oxacalcitriol prevents progressive glomerulosclerosis without adversely affecting calcium and phosphorus metabolism in subtotally nephrectomized rats. Nephrol Dial Transplant 17: 2132–2137.

CHAPTER 104

Inhibition of Benign Prostatic Hyperplasia by Vitamin D Receptor Ligands

I. II. III. IV.

MARIO MAGGI CLARA CRESCIOLI

Andrology Unit, Department of Clinical Physiopathology, University of Florence, Florence, Italy

LUCIANO ADORINI

BioXell, Milan, Italy

Endocrinology Unit, Department of Clinical Physiopathology, University of Florence, Florence, Italy

Introduction Pathogenesis of BPH Effects of Androgens and Growth Factors on Human BPH Cells Vitamin D Receptor Expression in Prostate Cells

I. INTRODUCTION The prostate is a gland surrounding the male urethra below the neck of the bladder and producing the prostatic fluid, a secretion which contributes 30% to the total ejaculate. The prostatic fluid is rich in fibrinolytic enzymes, such as prostatic-specific antigen (PSA), acid phosphatase, citric acid, and zinc. In humans, the prostate gland is composed of 40 to 50 ducts distributed essentially in three distinct zones: peripheral, central, and transitional or periurethral. While cell transformation in the peripheral zone gives rise to prostate cancer, cell growth in the periurethral zone leads to the most common age-related disease of the male: benign prostatic hyperplasia (BPH). The prostate weight is only a few grams at birth, and it increases during puberty, reaching approximately 20 g in the young adult. In contrast to the pubertal growth phase, which involves the entire gland, during the fifth decade of life, in the majority of men, there is a second growth phase selectively involving the periurethral zone leading to BPH [1]. The prevalence of BPH increases with age so that by age 80, about 90% of men have histological evidence of BPH [1]. In a subset of elderly men (27–35%), BPH can cause lower urinary tract symptoms (LUTS), which may require medical or surgical treatment [2] due to the compression by the enlarged prostate of the prostatic urethra, which decreases bladder outflow. In the earliest stages, this obstruction is compensated by an increased activity of the bladder detrusor muscular system but eventually complete voiding of the bladder is prevented, due to the slackening of the neck musculature. Urinary obstruction VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX

V. Antiproliferative Effects of BXL-353 on Human BPH Cells VI. Inhibition of in Vivo Prostate Growth by BXL-353 VII. Conclusions References

and even renal insufficiency might follow. This may lead to emergency surgery for acute urinary retention with an increased risk of morbidity and even death when compared to elective surgery [3].

II. PATHOGENESIS OF BPH Periurethral prostate overgrowth involves both epithelial and stromal components, including both fibroblasts and smooth muscle cells, in various combinations. One of the earliest events in BPH development is the reduction of the epithelium/stroma ratio, most probably due to an imbalance between growth and death programs. Indeed, in the hyperplastic prostate, the epithelium regularly undergoes apoptosis, whereas stromal cells escape it [4], with a consequent increase in the stromal volume [5]. In addition, stromal growth factors (GFs) induce epithelial overgrowth and glandular hyperplasia [6,7]. All these events are clearly androgen-dependent, as shown by the observation that BPH does not develop in hypogonadal men, and that either surgical or pharmacological castration results in a decrease gland size [8–10]. However, BPH develops mainly in older men, when circulating testosterone, and in particular free testosterone, is progressively decreasing. Therefore, it is possible that sensitivity to androgens, rather than circulating androgen levels, are involved in BPH pathogenesis. A higher transcriptional activity of the androgen receptor (AR) due to a decreased number of CAG repeats in exon 1 has been reported in the majority [11–13], although not in all studies [14,15]. Interestingly, in hypogonadal Copyright © 2005, Elsevier, Inc. All rights reserved.

1834

MARIO MAGGI, CLARA CRESCIOLI, AND LUCIANO ADORINI

patients the effect of androgen substitution on prostate growth was inversely related to the extent of CAG residues [16]. In addition, a recent study has indicated a decreased expression of the AR co-repressor DAX-1 in BPH [17]. These studies support the view that AR activity is up-regulated in the prostate of BPH patients. Therefore, blocking AR activity represents a promising approach in the treatment of BPH. This could be achieved by reducing androgen levels, for example, blocking their formation with GnRH analogs or by antagonizing androgen activity at the receptor level using AR antagonists. Although both these strategies might be indeed effective, in clinical practice they are unacceptable because of the major side effects caused by complete androgen ablation in otherwise healthy individuals. Wilson 1972 [18], first hypothesized that the main androgen inducing prostate hyperplasia was not testosterone (T), but its highly biologically active metabolite dihydrotestosterone (DHT), which is formed locally by two 5α-reducing iso-enzymes (5α-reductase type 1 and 2, the latter being predominant, see [19] for review). Interestingly, intra-prostatic DHT content is not decreased as a function of age [20–22]. According to Wilson’s original hypothesis, blocking DHT formation with a type 2 selective (finasteride) or with a dual (dutasteride) inhibitor of 5α-reductase isoforms is, indeed, an effective treatment for BPH [23,24]. However, prostate size reduction obtained with this strategy is relatively limited (about 25%). Also some men experience sexual side effects related to partial androgen deficiency (decreased libido and impotence) that are not well tolerated, in particular in the ageing male [3,25]. It is possible that the limited clinical response to 5α-reductase inhibitors is due to a compensatory increase in intra-prostatic growth factor (GF) receptors, which follows androgen deprivation [26,27]. Therefore, an alternative strategy to reduce age-related prostate overgrowth is to decrease the activity of androgen-induced prostatic GFs, which are considered to mediate, at least partially, the proliferative activity of sex steroids in the gland [7,28,29]. It is interesting to note that the prostate gland is one of the few androgen targets retaining a proliferative responsiveness to androgens in adulthood. Therefore disrupting androgeninduced, intra-prostatic GF signaling is an attractive option to obtain a selective, and sexual side-effect free, therapy for BPH.

mutual interactions between the two compartments (reviewed in [30]). However, as discussed above, stromal rather than epithelial cells are thought to be primarily involved in the pathogenesis of BPH. As shown in Fig. 1, BPH-derived stromal cells (BPH cells) express the AR gene and protein, with a high affinity for the ligand (Kd = 72 ± 34 pM), as well as both isoforms of 5α-reductase [31]. In addition, they respond to androgens with an increased growth (EC50 = 380 ± 200 pM, [31]). An increase in BPH cell proliferation was also obtained with addition of specific GFs, such as epidermal growth factor (EGF [32]), keratinocyte-GF (KGF [32,33]), and insulin-like growth factor-I (IGF-I [34]). Data in Fig. 2 show the maximal stimulatory activity of KGF (10 ng/ml), Des (1–3) IGF-I (an IGF-I analog which does not bind to binding proteins, 10 ng/ml) and T (10 nM) on BPH cell proliferation. KGF- and Des [1–3] IGF-I-induced proliferation was completely blocked only by specific antibodies, but not by unrelated antibodies or immunoglobulins (Fig. 2). Conversely, testosterone-induced cell growth was completely abolished not only by an AR antagonist (cyproterone acetate) or by a type 2 5α-reductase inhibitor (finasteride), but also by antibodies against the receptors for KGF (KGFR) and IGF-I (IGFR1) (Fig. 2). This indicates that T-induced proliferative activity in BPH cells is at least partially mediated by KGFR and IGFR1. This finding is consistent with data from organ cultures of neonatal rat ventral prostates, in which exogenous administration of KGF completely replaces the requirement of T for prostate growth and branching morphogenesis [35]. In addition, KGF has been also shown to replace androgen in eliciting growth and differentiation of seminal vesicles [36]. Hence, KGF, the predominant fibroblast GF (FGF) in human prostate [37], is considered one of the main prostatic andromedins, that is mediators of androgen-induced growth [38]. Also, the IGF system has been implicated in the pathogenesis of BPH. IGFR1 and IGF-II expression were higher in the periurethral zone of the human prostate than in other zones, and IGF-II levels were strictly correlated with the intra-prostatic androgen level [39]. Patients with the highest circulating levels of IGF-I have an elevated risk of BPH [40] and transgenic mice overexpressing IGF-I protein in the prostate show sign of hyperplasia in the ventral lobe, the most androgen-dependent zone [41].

III. EFFECTS OF ANDROGENS AND GROWTH FACTORS ON HUMAN BPH CELLS

IV. VITAMIN D RECEPTOR EXPRESSION IN PROSTATE CELLS

In the human prostate, the AR is expressed in both the epithelial and the stromal compartments and regulates

The aforementioned experimental and clinical studies indicate that an ideal medical treatment for BPH might be an agent able to disrupt the intra-prostatic

NC

hfPSMC

CHO 1827

MW

hfPSMC NC

541 bp

MW

0.015

hfPSMC NC

893 bp 196 bp

CHO 1829

AR GAPDH

Prostate

5αR-1 BPH cells

Prostate

BPH cells

MW [3H] R1881 B/T

Prostate

B 0.028

BPH cells

A

576 bp

NC

CHO

hfPSMC

Prostate

−6

BPH cells

−8 Log [R1881] (M)

380 bp

CHO hfPSMC

CHO 1829

CHO 1827

VDR

MW NC

−10

BPH cells Prostate

0.002

MW

5αR-2

196 bp

GAPDH

FIGURE 1 Expression of AR, 5α-reductase and VDR in human BPH cells. Panel A: homologous competition curve for [3H]R1881 binding. R1881 binds with high affinity (Kd = 72 ± 34 pM) and low capacity (Bmax = 2.64 ± 0.5 fM) to a single class of sites. Ordinate: B/T, Bound to total ratio for [3H]R1881; Abscissa: Total concentration (molar) of labeled and unlabeled R1881. Inset: RT-PCR detection of the AR gene. Products are derived from total RNA using specific primers for AR (upper panel) and GAPDH (lower panel). Panel B: RT-PCR amplimers from total RNA of BPH cells, prostate tissue, CHO 1827 (transfected with 5α-reductase 1, 5α-R1, gene) or CHO 1829 (transfected with 5α-reductase 2, 5α-R2 gene), CHO cells, human fetal penile smooth muscle cells (hfPSMC), using specific primers for 5α-R1 (upper panel), 5α-R2 (second panel), VDR (third panel) and GAPDH (bottom panel). CHO 1827 or 1829, hfPSMC and human prostate were used as positive controls for 5α-R1, 5α-R2 and AR. GAPDH mRNA amplification was performed to verify the integrity and loading of the extracted total RNA. MW, molecular weight markers; NC, negative control. The blots are representative of three separate experiments. *

160 *

* *

*

*

*

* 120

*

100

Des IGF-I (10 ng/ml)

IgG

Anti-IGFR1

Anti-KGFR

F Cyp

*

F Cyp

BXL-353

IgG

Anti-IGFR1

Anti-KGFR

BXL-353

IgG

Anti-IGFR1

0 KGF (10 ng/ml)

*

* Anti-KGFR

IgG

Anti-IGFR1

Anti-KGFR

20

BXL-353

*

60 40

*

*

BXL-353

80

Control

Cell number (% of control)

140

T (10 nM)

FIGURE 2 Antiproliferative effect of BXL-353 on BPH cell growth induced by GFs or testosterone (T). Incubation for 48 h with KGF (10 ng/ml), Des [1–3] IGF-I (10 ng/ml) or T (10 nM) significantly induced BPH cell proliferation. Anti-androgens such as the 5α-R2 inhibitor, finasteride (F, 1 nM), or the AR antagonist cyproterone acetate (Cyp, 100 nM) completely reverted T-induced stimulation of BPH cell growth, but they did not exert any effect on basal cell growth. Specific antibodies against KGFR (Anti-KGFR, 1 µg/ml) and IGFR1 (AntiIGFR1, 1 µg/ml) blocked cell growth stimulated by their cognate GFs. T-induced cell proliferation was blunted by both types of anti-GF receptor antibodies. Immunoglobulin controls (IgG 1 µg/ml) failed to block either GFor T-stimulated proliferation. BXL-353 (10 nM) was able to block BPH cell growth both in basal condition and in the presence of T or GFs. Results are expressed as percent increase (mean ± SEM) over their relative controls in 4 different experiments performed in quadruplicate (* P< 0.01 vs control; ° P < 0.01 vs GF- or T- treated cells by one-way ANOVA and paired or unpaired Student’s t tests). The data are derived from Crescioli et al., 2003.

1836 cross-talk between AR and GFs but devoid of antiandrogenic properties. Calcitriol analogs might comply with such criteria. The strict inter-relationships between vitamin D and prostate have been extensively described. Vitamin D deficiency has been proposed to be a risk factor for prostate cancer [42,43], because prostate cancer mortality in the USA increases as the availability of ultraviolet light exposure, and therefore of vitamin D formation, decreases [44] (see Chapter 90). Polymorphisms in the VDR gene have also been associated with increased risk of prostate cancer in some studies [45–47] (see Chapter 68). Malignant prostate cells express the VDR, and treatments with calcitriol, or lesshypercalcemic analogs, can inhibit prostate cancer proliferation and invasiveness (see Chapter 94 and ref. [48,49] for review). Interestingly, also epithelial and stromal cells of both human [50,51] (see also Figs. 1 and 2) and rat [51] normal prostate cells express the VDR, and addition of 1,25 (OH)2D3 inhibits cell growth [50].

V. ANTIPROLIFERATIVE EFFECTS OF BXL-353 ON HUMAN BPH CELLS We have extensively studied the antiproliferative effects of VDR ligands, and in particular of 1,25-dihydroxy-16-ene-23-yne D3 (BXL-353 or analog V), a compound 30-fold less calcemic than calcitriol, on human stromal prostate cells (BPH cells). BPH cells were obtained from prostate tissues derived from patients, who underwent suprapubic adenomectomy for BPH and did not receive any pharmacological treatment in the three months preceding surgery [33]. BPH cells showed positive staining for smooth-muscle actin, vimentin, and desmin, suggesting fibromuscular morphological features. Conversely, they were negative for epithelial and endothelial markers such as cytokeratin and factor VIII [33]. As shown in Fig. 2 BXL-353 completely inhibited GF- or T-induced BPH cell proliferation and also decreased the growth of unstimulated cells [31,33,34]. To better understand the antiproliferative effect of BXL-353, we have studied its effects on the cell cycle distribution of partially synchronized BPH cells after a 24 h culture with medium or KGF (10 ng/ml). As shown in Fig. 3, after serum starvation, more than 75% of the cells were in G0/G1-early S phase, as indicated by fluorescence emission of propidium iodide-stained nuclei. Treatment with KGF allowed the cells to progress through the cell cycle with a statistically significant decrease in the proportion of cells accumulated in G0/G1-early S and an increase in cells traversing the G2/M phase. The simultaneous addition of BXL-353 completely antagonized the KGF-induced effects on cell-cycle progression.

MARIO MAGGI, CLARA CRESCIOLI, AND LUCIANO ADORINI

In BPH cells GFs and steroids not only stimulated DNA synthesis and cell proliferation but also prolonged cell survival, via induction of the anti-apoptotic protein Bcl-2 [31,33,34] (see also Fig. 4). Members of the Bcl-2 family are essential mediators of cell survival and apoptosis, and include both anti- and pro-apoptotic intracellular proteins residing at the mitochondrial outer membrane [52–55]. Their classification is based on the presence or absence of Bcl-2 homology (BH) domains: BH1, BH2, BH3, and BH4 [56]. In particular, Bcl-2 and Bcl-XL members, both containing all four BH domains, inhibit apoptosis and promote cell survival [57]. Bcl-2 activity, derived by integrating signals from survival and death stimuli, seems to be regulated by several different mechanisms, like homo- and heterodimerization with other family members, or posttranslational modifications such as phosphorylation and proteolysis [52,58]. BXL-353 not only dramatically reduced GF- or T-induced Bcl-2 overexpression and survival, but also in the presence of these anti-apoptotic factors was able to stimulate a sustained death program (Fig. 4). Hence, in BPH cells, BXL-353 induced a decrease in the progression through the cell cycle and an increase in the rate of programmed cell death. Similar results were observed with calcitriol in breast cancer cells [59], in the androgen-dependent prostate cancer cell line LNCaP [60,61] as well as in metastatic Dunning rat prostate carcinoma [62]. Interestingly, in LNCaP cells, overexpression of Bcl-2 completely blocked calcitriol-induced apoptosis but only partially affected cell cycle arrest [61], indicating that partially independent pathways mediate the effects of calcitriol on cell proliferation and cell death. The inhibitory effect of BXL-353 on prostate growth and survival is at least partially explained by the inhibition of GF-induced receptor activation. We have found that a rapid incubation of both benign [33] and malignant [63] prostate cells with BXL-353 dramatically reduces agonist-induced KGF-R auto-phosphorylation, one of the earliest event of KGF signaling. Because this effect was rapid, induced in a few minutes, and accompanied by an increase in intracellular calcium concentrations [33], we speculated the involvement of a nontranscriptional mechanism which, in agreement with recent results in chick myoblast [64], might involve the same VDR translocated from the nucleus to the microsomal fraction. Other studies have shown that rapid effects of calcitriol are mediated by a binding protein different from the VDR [65–68]. It has also been shown [69] that the VDR, upon ligand binding, physically interacts with the catalytic subunit of protein phosphatases PP1 and PP2Ac, thereby promoting their enzymatic activities with the consequent inactivation of p70S6k, a kinase essential

1837

CHAPTER 104 Inhibition of Benign Prostatic Hyperplasia by Vitamin D Receptor Ligands

A

G0/G1

S

Phase percentage

80

*

70

G2/M

25

30

20

25 *

20

15

15 10

10

60 5 50

KGF

Control

KGF +BXL-353

5 Control

KGF

KGF +BXL-353

Control

KGF

KGF +BXL-353

B 140

140

140 77

Nuclei number

78

64

6 0

0

200

17 400 600 Control

18 0 800 0

200

7

18 400 600 KGF

800

0

0

200

15 400 600 KGF+BXL-353

800

FIGURE 3 Effect of BXL-353 on cell cycle distribution of BPH cells. Panel A shows the effect of 24 h treatment with KGF (10 ng/ml) with or without BXL-353 (10 nM) on partially synchronized (24 h serum starvation) BPH cells. KGF significantly reduced the number of BPH cell accumulated in the G0/G1 phase, thereby increasing the percentage of nuclei in G2/M. Simultaneous treatment with BXL-353 completely blocked the mitogenic activity of KGF. Results are expressed as cell cycle phase percentage. * P < 0.01 vs. control by one-way ANOVA and paired or unpaired Student’s t tests. For assay method see ref. [86]. Panel B shows results from a typical experiment.

in G0/G1 transition. VDR ligands not only induced a prompt decrease in phosphorylated KGFR [33,63], but also in phospho-Erk and phospho-Akt [70,71]. Hence, it is possible that calcitriol and related analogs might activate the catalytic subunit of distinct families of phosphatases, leading to antiproliferative effects by targeting GF signaling. Interestingly, in the human epidermoid A431 cells, overexpressing an autocrine growth loop for EGF, calcitriol not only induced a rapid alteration of EGFR auto-phosphorylation (as previously observed by us on KGFR), but also impaired EGFR membrane trafficking and signaling via the classic VDR-dependent mechanism [72]. In conclusion, it is possible that genomic (nuclear VDR-dependent) as well as rapid or nongenomic (cytoplasmic VDR-dependent?) mechanisms simultaneously contribute to the growthsuppressing activity of VDR ligands on prostate cells.

VI. INHIBITION OF IN VIVO PROSTATE GROWTH BY BXL-353 To investigate whether calcitriol analogs might represent a new opportunity to decrease prostrate cell

overgrowth and, therefore, to treat BPH, we carried out a series of studies using the rat as an experimental model. Taking advantage of the high sensitivity to androgens of the ventral prostate, castrated rats were supplemented with T with or without increasing doses of BXL-353 for various time periods [31]. Changes in ventral prostate volume and morphology, along with measurements of calcemia and hormonal values were studied. BXL-353 has a maximum tolerated dose of 30 µg/Kg, and at any dose tested never caused hypercalcemia. One week treatment with BXL-353 was sufficient to decrease significantly and dose-dependently ventral prostate weight, with an IC50 = 1.5 ± 1 µg/Kg (Fig. 5, panel A). Similar results were obtained with a two-week treatment of BXL-353 (Fig. 5, panel B). A 30% reduction of ventral prostate weight was induced by one month treatment with BXL-353 to intact adult rats (Fig. 5, panel C). It is interesting to note that prostate weight reduction induced by BXL-353 (∼50% decrease) is similar to that obtained in similar experimental models with 10 mg/Kg finasteride [31]. Because we observed that human BPH stromal cells underwent apoptosis even after short-term in vitro exposure to BXL-353, we investigated the fate of rat

A 60 * 50

Apoptotic index %

*

*

40

*

30

20

*

+ BXL-353

*

+ BXL-353

+ BXL-353

Control

10

+ BXL-353

*

0

+KGF (10 ng/ml)

+Des (1–3) IGF (10 ng/ml)

+T (10 nm)

B 45 * *

40

*

30 *

25

*

* 20

Control

5

+ BXL-353

10

+ BXL-353

15 + BXL-353

Bcl-2 expression %

35

*

0 BXL-353 +KGF (10 ng/ml)

FIGURE 4

+Des (1–3) IGF (10 ng/ml)

+T (10 nm)

Effect of BXL-353 on apoptosis and Bcl-2 expression in BPH cells. The effect of 48 h treatment with BXL-353 (10 nM) alone or in combination with GFs (KGF and Des [1–3] IGF-I, 10 ng/ml) or T (10 nM) is reported on apoptosis (panel A) and Bcl-2 expression (panel B). Apoptotic index was obtained from in situ end labeling (ISEL) experiments and represents the number of stained nuclei over total cells in each of at least 5 separate fields per slide. Percentage of Bcl-2 stained cells was calculated by counting the number of immunopositive BPH cells over total cells in each of at least 5 separate fields per slide. Treatment with GFs or T significantly decreased the number of apoptotic cells (panel A), while it increased Bcl-2 positivity (panel B). Simultaneous treatment with GFs or T and BXL-353 significantly blunted the effect of both GFs and T on the number of ISEL (panel A) and Bcl-2 (panel B) positive cells. A 48 h exposure to BXL-353 induced a massive increase in apoptotic index (panel A), while Bcl-2 expression decreased (panel B). *P < 0.01 vs. control; °P < 0.01 vs. GF- or T-treated cells by one-way ANOVA and paired or unpaired Student’s t tests. The data are partially derived from Crescioli et al., 2000; 2002; 2003.

CHAPTER 104 Inhibition of Benign Prostatic Hyperplasia by Vitamin D Receptor Ligands

A

B

IC50 = 1.52 ± 1.05 µg/Kg

IC50 = 7.4361 ± 1.25 µg/Kg

110 100

100

Ventral prostate weight (% of T-treated)

Ventral prostate weight (% of T-treated)

1839

90 80 70 60 50

90 80 70 60

40 100

1000

10000

BXL-353 (µg/Kg)

C

50 1000

10000 BXL-353 (µg/Kg)

IC50 = 4.2041 ± 2.1 µg/Kg

Ventral prostate weight (% of vehicle)

100 95 90 85 80 75 70 65

1000 10000 BXL-353 (µg/Kg)

FIGURE 5 Effect of BXL-353 on rat ventral prostate weight. Panel A: Castrated rats were supplemented with a single injection of T enanthate (30 mg/Kg/week) and orally treated for 4 days with vehicle or increasing concentrations of BXL-353 (1–30 µg/Kg). Ventral prostate weight is expressed as % variation (mean ± SEM) of the weight of T-replaced castrated rats, in two separate experiments. ^P < 0.05 and *P < 0.01 vs T-supplemented vehicle-treated rats by one-way ANOVA and paired or unpaired Student’s t tests. Panel B: Castrated rats were injected with T enanthate (15 mg/Kg/week) and orally treated for 5 day/week for two consecutive weeks with vehicle or increasing concentrations of BXL-353 (3–30 µg/Kg). Ventral prostate weight is expressed as % variation (mean ± SEM) of the weight of T-replaced castrated rats (n = 4). *P < 0.01 vs. T-supplemented vehicle-treated rats by one-way ANOVA and paired or unpaired Student’s t tests. Panel C: Intact adult rats were orally treated for over one month (5 times/week for a total of 27 administrations) with vehicle (control) or increasing concentrations of BXL-353 (3–30 µg/Kg). Ventral prostate weight is expressed as % variation (mean ± SEM) of the weight of control rats (n = 4) (*P < 0.01 vs control rats). Data are derived from Crescioli et al., 2003.

prostate cells after sub-acute (7–14 days, castrated rats) or prolonged (1 month, intact rats) treatment with the analog. By using terminal deoxynucleotidyl transferase (TdT) mediated dUTP nick end-labeling (TUNEL), we observed the typical hallmark of nuclear fragmentation in both the epithelial and stromal cells of the BXL-353-treated prostate in all the experimental protocols studied [31]. In addition, we found that BXL353 treatments induced a dose- and time-dependent up-regulation of clusterin gene and protein. Clusterin (CLU), or testosterone-repressed message 2 (TRPM-2), is an ubiquitous, puzzling protein expressed also in the rat prostate [73–75], which is down-regulated by androgens and up-regulated by growth arrest and cell

death (see in [76]). CLU has different intracellular and extracellular functions. As an extracellular, secretory glycoprotein, it has a chaperone-like role, binding a wide range of unrelated molecules and probably clearing cellular debris. Conversely, the intracellular ∼49 kDa protein, after appropriate stimuli, is transported from the cytoplasm to the nucleus, where it binds DNA helicases, as Ku70/Ku86, thereby reducing DNA repair and allowing cell death [77]. Hence, CLU is generally considered an androgen-regulated, pro-apoptotic protein. We confirmed that in the rat prostate, CLU expression was increased by castration and finasteride administration [73,75], and we found that BXL-353 induced a sustained increase in CLU gene and protein

1840

MARIO MAGGI, CLARA CRESCIOLI, AND LUCIANO ADORINI

expression [31]. Interestingly, CLU-positive cells were more apparent in the glands of BXL-353 treated rats showing the more pronounced features of involution and atrophy. In conclusion, studies in the rat strongly support findings in human BPH cells: BXL-353, similar to finasteride, counteracts the growth promoting effect of T, by inducing growth cell arrest and apoptosis. However, BXL-353, at variance with finasteride, is not an antiandrogen. It does not bind to the AR, as demonstrated by competition studies using the synthetic androgen [3H] R1881on BPH homogenates, and it does not inhibit 5α-reductase activity, as shown by the failure to interfere with DHT formation in CHO cells transfected with type 1 or type 2 5α-reductase iso-enzymes [31]. In addition, BXL-353 did not affect the gonadal or pituitary secretion of testosterone or gonadotrophin [31]. Hence it should act downstream of the AR receptor ligand interaction. The activated AR is a multiple phosphorylated protein and some of its phosphorylation sites (as Ser 650) are required for full transcriptional activity (see ref. [78]). Hence, it is possible that BXL353 might activate the catalytic subunit of distinct families of phosphatases, therefore exerting its antiproliferative effects acting on AR-dependent signaling. Alternatively, BXL-353 may disrupt androgendependent GF-mediated survival pathways, thus hampering T-induced BPH cell growth.

patients using α-blockers [81]. A possible mechanism of action underlying the risk reduction of BPH-related surgery by finasteride is that 5α-reductase inhibitors, by blocking DHT formation, shrink the prostate volume, which, in turn, is shown to be itself an important risk factor for BPH progression and, consequently, BPH-related surgery [80,82]. In fact, although many variables make it difficult to predict an individual’s clinical course, prostatic size is reported to be one of the most important risk factors along with age and prostatic specific antigen (PSA) value [83]. Hence, the ideal treatment for BPH should include a medication that reduces prostate volume without interfering with androgen activity. Actually, patient compliance for finasteride may be limited by consistent sexual side effects, such as decreased libido, altered sexual potency, or ejaculatory dysfunction [84], especially in men with borderline erectile function [85]. Well-tolerated calcitriol analogs, such as BXL-353, might represent such a new class of drugs, because they decrease AR-mediated prostate growth, acting downstream of the AR on the GF-mediated proliferation pathways. Based on the data reviewed here, a double-blind, placebo-controlled phase II study is currently ongoing in Italy to evaluate the effects of a nonhypercalcemic calcitriol analog in patients with BPH.

VII. CONCLUSIONS

1. Berry SJ, Coffey DS, Walsh PC, Ewing LL 1984 The development of human benign prostatic hyperplasia with age. J Urol 132:474–479. 2. Jacobsen SJ, Girman CJ, Guess HA, Panser LA, Chute CG, Oesterling JE, Lieber MM 1995 Do prostate size and urinary flow rates predict health care-seeking behavior for urinary symptoms in men? Urology 45:64–69. 3. Thorpe A, Neal D 2003 Benign prostatic hyperplasia. Lancet 361:1359–1367. 4. Claus S, Berges R, Senge T, Schulze H 1997 Cell kinetic in epithelium and stroma of benign prostatic hyperplasia. J Urol 158:217–221. 5. Shapiro E, Becich MJ, Hartanto V, Lepor H 1992 The relative proportion of stromal and epithelial hyperplasia is related to the development of symptomatic benign prostate hyperplasia. J Urol 147:1293–1297. 6. McNeal JE 1978 Origin and evolution of benign prostatic enlargement. Invest Urol 15:340–345. 7. Serio M, Fiorelli G 1991 Dual control by androgens and peptide growth factors of prostatic growth in human benign prostatic hyperplasia. Mol Cell Endocrinol 78:C77–C81. 8. Peters CA, Walsh PC 1987 The effect of nafarelin acetate, a luteinizing-hormone-releasing hormone agonist, on benign prostatic hyperplasia. N Engl J Med 317:599–604. 9. McConnell JD 1990 Androgen ablation and blockade in the treatment of benign prostatic hyperplasia. Urol Clin North Am 17:661–670. 10. Schroder FH 1994 5 alpha-reductase inhibitors and prostatic disease. Clin Endocrinol 41:139–147.

A large proportion of aging males develop BPH and, until recently, the only options for treatment were surgical intervention or watchful waiting. During the last 10 years, progress in medical therapy of BPH has resulted in effective treatments patterns leading to a significant improvement in the quality of life of affected patients. At present, two different classes of agents are available for BPH treatment: α-blockers and 5α-reductase inhibitors. Although sexual related side effects are more often reported with 5α-reductase inhibitors than with α-blockers [79], the reverse is true for reduction in risk of BPH-related surgeries [80]. A population-based cohort study, conducted in more than 5000 patients, receiving either α-blockers or 5α-reductase inhibitors (finasteride), showed that the incidence of BPH-related surgery was higher in α-blocker-treated patients than in 5α-reductase inhibitor-treated ones [80]. Similar results, obtained from a retrospective analysis of patients’ data, demonstrated that the risk of experiencing serious complication related to BPH progression (catheterization, acute urinary retention, surgery) was significantly lower in finasteride-treated patients compared to

References

CHAPTER 104 Inhibition of Benign Prostatic Hyperplasia by Vitamin D Receptor Ligands

11. Giovannucci E, Stampfer MJ, Chan A, Krithivas K, Gann PH, Hennekens CH, Kantoff PW 1999 CAG repeat within the androgen receptor gene and incidence of surgery for benign prostatic hyperplasia in U.S. physicians. Prostate 39:130–134. 12. Mitsumori K, Terai A, Oka H, Segawa T, Ogura K, Yoshida O, Ogawa O 1999 Androgen receptor CAG repeat length polymorphism in benign prostatic hyperplasia (BPH): correlation with adenoma growth. Prostate 41:253–257. 13. Shibata A, Stamey TA, McNeal JE, Cheng I, Peehl DM 2001 Genetic polymorphisms in the androgen receptor and type II 5 alpha-reductase genes in prostate enlargement. J Urol 166: 1560–1564. 14. Bousema JT, Bussemakers MJ, van Houwelingen KP, Debruyne FM, Verbeek AL, de La Rosette JJ, Kiemeney LA 2000 Polymorphisms in the vitamin D receptor gene and the androgen receptor gene and the risk of benign prostatic hyperplasia. Eur Urol 37:234–238. 15. Azzouzi AR, Cochand-Priollet B, Mangin P, Fournier G, Berthon P, Latil A, Cussenot O 2002 Impact of constitutional genetic variation in androgen/oestrogen-regulating genes on age-related changes in human prostate. Eur J Endocrinol 147: 479–484. 16. Zitzmann M, Depenbusch M, Gromoll J, Nieschlag E 2003 Prostate volume and growth in testosterone-substituted hypogonadal men are dependent on the CAG repeat polymorphism of the androgen receptor gene: a longitudinal pharmacogenetic study. J Clin Endocrinol Metab 88:2049–2054. 17. Agoulnik IU, Krause WC, Bingman WE 3rd, Rahman HT, Amrikachi M, Ayala GE, Weigel NL 2003 Repressors of androgen and progesterone receptor action. J Biol Chem (in press). 18. Wilson JD 1972 Recent studies on the mechanism of action of testosterone. N Engl J Med 287:1284–1291. 19. Carson C, Rittmaster R 2003 The role of dihydrotestosterone in benign prostatic hyperplasia. Urology 61:2–7. 20. Siiteri PK, Wilson JD 1970 Dihydrotestosterone in prostatic hypertrophy. I. The formation and content of dihydrotestosterone in the hypertrophic prostate of man. J Clin Invest 49:1737–1745. 21. Hammond GL, Kontturi M, Vihko P, Vihko R 1978 Serum steroids in normal males and patients with prostatic diseases. Clin Endocrinol (Oxf) 9:113–121. 22. Walsh PC, Lepor H, Eggleston JC 1983 Radical prostatectomy with preservation of sexual function: anatomical and pathological considerations. Prostate 4:473–485. 23. McConnell JD, Bruskewitz R, Walsh P, Andriole G, Lieber M, Holtgrewe HL, Albertsen P, Roehrborn CG, Nickel JC, Wang DZ, Taylor AM, Waldstreicher J 1998 The effect of finasteride on the risk of acute urinary retention and the need for surgical treatment among men with benign prostatic hyperplasia Finasteride Long-Term Efficacy and Safety Study Group. N Engl J Med 338:557–563. 24. Roehrborn CG, Boyle P, Nickel JC, Hoefner K, Andriole G; ARIA3001 ARIA3002 and ARIA3003 Study Investigators 2002 Efficacy and safety of a dual inhibitor of 5-alphareductase types 1 and 2 (dutasteride) in men with benign prostatic hyperplasia. Urology 60:434–441. 25. Steers W 2001 5alpha-reductase activity in the prostate. Urology 58:17–24. 26. Fiorelli G, De Bellis A, Longo A, Natali A, Costantini A, Serio M 1989 Epidermal growth factor receptors in human hyperplastic prostate tissue and their modulation by chronic treatment with a gonadotropin-releasing hormone analog. J Clin Endocrinol Metab 68:740–743. 27. Fiorelli G, De Bellis A, Longo A, Giannini S, Natali A, Costantini A, Vannelli GB, Serio M 1991 Insulin-like growth

28. 29. 30. 31.

32.

33.

34.

35.

36.

37. 38. 39.

40.

41.

42. 43.

1841

factor-I receptors in human hyperplastic prostate tissue: characterization, tissue localization, and their modulation by chronic treatment with a gonadotropin-releasing hormone analog. J Clin Endocrinol Metab 72:740–746. Cunha GR 1996 Growth factors as mediators of androgen action during male urogenital development. Prostate Suppl 6:22–25. Gnanapragasam VJ, McCahy PJ, Neal DE, Robson CN 2000 Insulin-like growth factor II and androgen receptor expression in the prostate. BJU Int 86:731–735. Marcelli M, Cunningham GR 1999 Hormonal signaling in prostatic hyperplasia and neoplasia. J Clin Endocrinol Metab 84:3463–3468. Crescioli C, Ferruzzi P, Caporali A, Mancina R, Comerci A, Muratori M, Scaltriti M, Vannelli GB, Smiroldo S, Mariani R, Villari D, Bettuzzi S, Serio M, Adorini L, Maggi M 2003 Inhibition of spontaneous and androgen-induced prostate growth by a nonhypercalcemic calcitriol analog. Endocrinology 144:3046–3057. De Bellis A, Crescioli C, Grappon C, Milani S, Ghiandi P, Forti G, Serio M 1998 Expression and cellular localization of keratinocyte growth factor and its receptor in human hyperplastic prostate tissue. J Clin Endocrinol Metab 83:2186–2191. Crescioli C, Maggi M, Vannelli GB, Luconi M, Salerno R, Barni T, Gulisano M, Forti G, Serio M 2000 Effect of a vitamin D3 analog on keratinocyte growth factor-induced cell proliferation in benign prostate hyperplasia. J Clin Endocrinol Metab 85:2576–2583. Crescioli C, Villari D, Forti G, Ferruzzi P, Petrone L, Vannelli GB, Adorini L, Salerno R, Serio M, Maggi M 2002 Des (1–3) IGF-I-stimulated growth of human stromal BPH cells is inhibited by a vitamin D3 analogue. Mol Cell Endocrinol 198:69–75. Sugimura Y, Cunha GR, Hayward N, Hayashi N, Arima K, Kawamura J, Rubin J, Aaronson S 1993 Keratinocyte growth factor (KGF) is a mediator of testosterone-induced prostatic development. Mol Cell Differ 1:423–426. Alarid ET, Rubin JS, Young P, Chedid M, Ron D, Aaronson SA, Cunha GR 1994 Keratinocyte growth factor function in epithelial induction during seminal vesicle development. Proc Natl Acad Sci USA 91:1074–1078. Ittman M, Mansukhani A 1997 Expression of fibroblast growth factors (FGFs) and FGF receptors in human prostate. J Urol 157:351–356. Lopez-Otin C, Diamandis EP 1998 Breast and prostate cancer: an analysis of common epidemiological, genetic, and biochemical features. Endocr Rev 19:365–396. Monti S, Di Silveri F, Iraci R, Martini C, Zanzara S, Falasco P, Poggi M, Stigliano A, Sciarra F, Toscano V 2001 Regional variations of insulin-like growth factor I (IGF-I), IGF-II, and receptor type I in benign prostatic hyperplasia tissue and their correlation with intraprostatic androgens. J Clin Endocrinol Metab 86:1700–1706. Chokkalingam AP, Gao YT, Deng J, Stanczyk FZ, Sesterhenn IA, Mostofi FK, Fraumeni JF Jr, Hsing AW 2002 Insulin-like growth factors and risk of benign prostatic hyperplasia. Prostate 52:98–105. Konno-Takahashi N, Takeuchi T, Shimizu T, Nishimatsu H, Fukuhara H, Kamijo T, Moriyama N, Tejima S, Kitamura T 2003 Engineered IGF-I expression induces glandular enlargement in the murine prostate. J Endocrinol 177:389–398. Schwartz GG, Hulka BS 1990 Is vitamin D deficiency a risk factor for prostate cancer? (Hypothesis). Anticancer Res 10:1307–1311. Hanchette CL, Schwartz CG 1992 Geographic patterns of prostate cancer mortality. Evidence for a protective effect of ultraviolet radiation. Cancer 70:2861–2869.

1842

MARIO MAGGI, CLARA CRESCIOLI, AND LUCIANO ADORINI

44. Luscombe CJ, Fryer AA, French ME, Liu S, Saxby MF, Jones PW, Strange RC 2001 Exposure to ultraviolet radiation: association with susceptibility and age at presentation with prostate cancer. Lancet 358:641–642. 45. Taylor JA, Hirvonen A, Watson M, Pittman G, Mohler JL, Bell DA 1996 Association of prostate cancer with vitamin D receptor gene polymorphism. Cancer Res 56:4108–4110. 46. Habuchi T, Liqing Z, Suzuki T, Sasaki R, Tsuchiya N, Tachiki H, Shimoda N, Satoh S, Sato K, Kakehi Y, Kamoto T, Ogawa O, Kato T 2000 Increased risk of prostate cancer and benign prostatic hyperplasia associated with a CYP17 gene polymorphism with a gene dosage effect. Cancer Res 60:5710–5713. 47. Ingles SA, Coetzee GA, Ross RK, Henderson BE, Kolonel LN, Crocitto L, Wang W, Haile RW 1998 Association of prostate cancer with vitamin D receptor haplotypes in AfricanAmericans. Cancer Res 58:1620–1623. 48. Peehl DM, Feldman D 2003 International Congress on Hormonal Steroids and Hormones and Cancer: The role of vitamin D and retinoids in controlling prostate cancer progression. Endocr Relat Cancer 10:131–140. 49. Krishnan AV, Peehl DM, Feldman D 2003 Inhibition of prostate cancer growth by vitamin D: Regulation of target gene expression. J Cell Biochem 88:363–371. 50. Peehl DM, Skowronski RJ, Leung GK, Wong ST, Stamey TA, Feldman D 1994 Antiproliferative effects of 1,25-dihydroxyvitamin D3 on primary cultures of human prostatic cells. Cancer Res 54:805–810. 51. Kivineva M, Blauer M, Syvala H, Tammela T, Tuohimaa P 1998 Localization of 1,25-dihydroxyvitamin D3 receptor (VDR) expression in human prostate. J Steroid Biochem Mol Biol 66:121–127. 52. Gross A, McDonnell JM, Korsmeyer SJ 1999 BCL-2 family members and the mitochondria in apoptosis. Genes Dev 13:1899–1911. 53. Li H, Yuan J 1999 Deciphering the pathways of life and death. Curr Opin Cell Biol 11:261–266. 54. Wang E, Marcotte R, Petroulakis E 1999 Signaling pathway for apoptosis: a racetrack for life or death. J Cell Biochem (Suppl 32–33):95–102. 55. Ashkenazi A, Dixit VM 1999 Apoptosis control by death and decoy receptors. Curr Opin Cell Biol 11:255–260. 56. Fesik SW 2000 Insights into programmed cell death through structural biology. Cell 103:273–282. 57. Huang DC, Strasser A 2000 BH3-only proteins essential initiatiors of apoptotic cell death. Cell 103:839–842. 58. Van Hoof C, Goris J 2003 Phosphatases in apoptosis: to be or not to be, PP2A is in the heart of the question. Biochim Biophys Acta 1640:97–104. 59. Simboli-Campbell M, Narvaez CJ, van Weelden K, Tenniswood M, Welsh J 1997 Comparative effects of 1,25(OH)2D3 and EB1089 on cell cycle kinetics and apoptosis in MCF-7 breast cancer cells. Breast Cancer Res Treat 42:31–41. 60. Blutt SE, Allegretto EA, Pike JW, Weigel NL 1997 1,25dihydroxyvitamin D3 and 9-cis-retinoic acid act synergistically to inhibit the growth of LNCaP prostate cells and cause accumulation of cells in G1. Endocrinology 138:1491–1497. 61. Blutt SE, Polek TC, Stewart LV, Kattan MW, Weigel NL 2000 A calcitriol analog, EB1089, inhibits the growth of LNCaP tumors in nude mice. Cancer Res 60:779–782. 62. Getzenberg RH, Light BW, Lapco PE, Konety BR, Nangia AK, Acierno JS, Dhir R, Shurin Z, Day RS, Trump DL, Johnson CS 1997 Vitamin D inhibition of prostate adenocarcinoma growth and metastasis in the Dunning rat prostate model system. Urology 50:999–1006. 63. Crescioli C, Maggi M, Luconi M, Vannelli GB, Salerno R, Sinisi AA, Bonaccorsi L, Ferruzzi P, Barni T, Forti G, Serio M

2002 Vitamin D3 analog inhibits keratinocyte growth factor signaling and induces apoptosis in human prostate cancer cells. Prostate 50:15–26. Capiati D, Benassati S, Boland RL 2002 1,25(OH)2-vitamin D3 induces translocation of the vitamin D receptor (VDR) to the plasma membrane in skeletal muscle cells. J Cell Biochem 86:128–135. Nemere I, Dormanen MC, Hammond MW, Okamura WH, Norman AW 1994 Identification of a specific binding protein for 1alpha,25-dihydroxyvitamin D3 in basal-lateral membranes of chick intestinal epithelium and relationship to transcaltachia. J Biol Chem 269:23750–23756. Nemere I, Szego CM 1998 Early action of parathyroid hormone and 1,25-dihydroxycholecalciferol on isolated epithelial cells from rat intestine: I. Limited lysosomal enzyme release and calcium uptake. Endocrinology 108:1450–1462. Nemere I, Ray R, McManus W 2000 Immunochemical studies on the putative plasmalemmal receptor for 1,25(OH)2D3. I. Chick intestine. Am J Physiol Endocrinol Metab 278: E1104–E1114. Jia Z, Nemere I 1999 Immunochemical studies on the putative plasmalemmal receptor for 1,25-dihydroxyvitamin D3 II. Chick kidney and brain. Steroids 64:541–550. Bettoun DJ, Buck DW 2nd, Lu J, Khalifa B, Chin WW, Nagpal S 2002 A vitamin D receptor-Ser/Thr phosphatase-p70 S6 kinase complex and modulation of its enzymatic activities by the ligand. J Biol Chem 277:24847–24850. McGuire TF, Trump DL, Johnson CS 2001 Vitamin D3–induced apoptosis of murine squamous cell carcinoma cells. Selective induction of caspase-dependent MEK cleavage and up-regulation of MEKK-1. J Biol Chem 276:26365–26373. Bernardi RJ, Johnson CS, Modzelewski RA, Trump DL 2002 Antiproliferative effects of 1α,25-dihydroxyvitamin D3 and vitamin D analogs on tumor-derived endothelial cells. Endocrinology 143:2508–2514. Cordero JB, Cozzolino M, Lu Y, Vidal M, Slatopolsky E, Stahl PD, Barbieri MA, Dusso A 2002 1,25-dihydroxyvitamin D down-regulates cell membrane growth- and nuclear growthpromoting signals by the epidermal growth factor receptor. J Biol Chem 277:38965–38971. Bettuzzi S, Hiipakka RA, Gilna P, Liao S 1989 Identification of an androgen-repressed mRNA in rat ventral prostate as coding for Sulphated Glycoprotein 2 by cDNA cloning and sequence analysis. Biochem J 257:293–296. Lakins J, Bennett SA, Chen JH, Arnold JM, Morrissey C, Wong P, O’Sullivan J, Tenniswood M 1998 Clusterin biogenesis is altered during apoptosis in the regressing rat ventral prostate. J Biol Chem 273:27887–27895. Astancolle S, Guidetti G, Pinna C, Corti A, Bettuzzi S 2000 Increased levels of clusterin (SGP-2) mRNA and protein accompany rat ventral prostate involution following finasteride treatment. J Endocrinol 167:197–204. Trougakos IP, Gonos ES 2002 Clusterin/apolipoprotein J in human aging and cancer. Int J Biochem Cell Biol 34: 1430–1448. Leskov KS, Klokov DY, Li J, Kinsella TJ, Boothman DA 2003 Synthesis and functional analyses of nuclear clusterin, a cell death protein. J Biol Chem 278:11590–11600. Gelmann EP 2002 Molecular biology of the androgen receptor. J Clin Oncol 20:3001–3015. Kassabian VS 2003 Sexual function in patients treated for benign prostatic hyperplasia. Lancet 361:60–62. Souverein PC, Erkens JA, de la Rosette JJ, Leufkens HG, Herings RM 2003 Drug treatment of benign prostatic hyperplasia and hospital admission for BPH-related surgery. Eur Urol 43:528–534.

64.

65.

66.

67.

68. 69.

70.

71.

72.

73.

74.

75.

76. 77. 78. 79. 80.

CHAPTER 104 Inhibition of Benign Prostatic Hyperplasia by Vitamin D Receptor Ligands

81. Boyle P, Roehrborn CG, Harkaway R, de la Rosette J, Emberton M, Logie J 2002 5-α reductase inhibitors (5ARIs) provide superior benefits to alpha-blockers in preventing acute urinary retention and surgery. BJU Int 90(Suppl S2):18. 82. Marberger MJ, Andersen JT, Nickel JC, Malice MP, Gabriel M, Pappas F et al. 2000 Prostate volume and serum prostate specific antigen as predictors of acute urinary retention. Combined experience from three large multinational placebocontrolled trials. Eur Urol 38:563–568. 83. Djavan B, Nickel JC, de la Rosette J, Abrams P 2002 The urologist view of BPH progression: results of an international survey. Eur Urol 41:490–496.

1843

84. Barry MJ, Roehrborn CG 2001 Benign prostatic hyperplasia. Br Med J 323:1042–1046. 85. Steers W, Guay AT, Leriche A, Gingell C, Hargreave TB, Wright PJ, Price DE, Feldman RA 2001 Assessment of the efficacy and safety of Viagra (sildenafilcitrate) in men with erectile dysfunction during long-term treatment. Int J Impot Res 13:261–267. 86. Maggi M, Bonaccorsi L, Finetti G, Carloni V, Muratori M, Laffi G, Forti G, Serio M, Baldi E 1994 Platelet-activating factor mediates an autocrine proliferative loop in the endometrial adenocarcinoma cell line HEC-1A. Cancer Res 54: 4777–4784.

Index

A AC. See Adenylyl cyclase ACE. See Angiotensin-converting enzyme Acidemia, rickets/osteomalacia features evident in, 984 Acidosis, 1α-hydroxylase inhibited by, 1316 Actin, 142 DBP sequestering, 147–148, 148f Actin-scavenger system, 143 DBP-actin complex implications for, 148–149 Activating domain (AD), AF-2, 279, 280f Activation function (AF-2) domain D analogs changing conformation of, 1452f, 1453 mutations v. D analogs, 1456 NCoA62/SKIP/VDR interaction independent of, 296 in NR LBDs, 279, 280f in 1,25(OH)2D3 transactivation of VDR, 1472 VDR LBD coactivator binding site v., 1473, 1473f WSTF coactivating ligand-induced, 309, 310f Activator protein (AP-1), in 1,25(OH)2D3-induced differentiation, 1640 AD. See Activating domain; Addison’s disease; Alzheimer’s disease Adaptor proteins, receptors requiring, 159 Addison’s disease (AD), VDR polymorphisms v., 1148 Adenylyl cyclase (AC), in 1α,25(OH)2D3/Ca2+ mechanism, 890f, 891 ADH. See Antidiuretic hormone ADHR. See Autosomal dominant hypophosphatemic rickets Adjusted apposition rate (Aj.AR) FP v., 1034 MAR v. osteoid surface in, 956–957 O.Th v., 1035, 1035f Adults acute hypocalcemia therapy for, 1058 D3 dosage considerations for, 1003–1005, 1009t D insufficiency in, 1091–1092, 1091f, 1096–1097 elderly D insufficiency in, 825–826, 1092–1094, 1092f D supplementation v. bone loss in, 825–826 insufficiency v. hip fracture risk in, 1089 intestinal D absorption reduced in, 1087–1088 infant D3 dosage v., 1003 minuscule D3 intake recommendations for, 1008 AF-2. See Activation function domain AFM. See Atomic force microscopy

Africans. See also specific nationalities VDR polymorphism allele frequencies in, 1128t, 1129 VDR polymorphisms v. PCa in, 1139t, 1145 Aging cutaneous D production v., 823–824 D metabolism v., 823–833 D tissue responsiveness in, 831–833 dietary D/intestinal absorption v., 824–826 FCA in response to free 1,25(OH)2D v., 832, 832f 25OHD serum concentration v., 826–827 1,25(OH)2D synthesis/metabolism v., 826–831 25OHD synthesis v., 826 PCa incidence v., 1679, 1680 as PCa risk factor, 1601, 1601t PTH/1,25(OH)2D v., 828, 828f Agonists. See also specific agonists action mechanism of selective, 1478–1480, 1479f assessment of noncalcemic selective, 1478 differential VDR activation by noncalcemic selective, 1478–1480 nonsecosteroid VDR, 1557–1567 AIDS, granulomatous disease incidence increasing with, 1391 AIPC. See Androgen-independent prostate cancer Aj.AR. See Adjusted apposition rate Al. See Aluminum Alaskans, D deficiency in child/infant, 795 Alcohol. See Ethanol Alcoholic liver disease bone disorders associated with, 1306 clinical features of, 1306 D metabolism impaired in, 1306, 1306f management, 1306 Alcoholics, inadequate dietary D/sunlight exposure in, 1266 Alcoholism hypophosphatemia in, 1177 magnesemia in, 1177 Alendronate bone loss prevented by, 1248 Ca absorption/bone formation suppression v., 1244 Alfacalcidol. See 1α-Hydroxyvitamin D3 Algerians, sunlight exposure v. rickets in, 1066–1067 Alien, NR co-repressor, 299 Alkaline phosphatase Ca entry v., 416 mineral properties in KO animals, 479t

1846 Alkaline phosphatase (Continued) in mineralization, 479t in mineralization v. 1α,25(OH)2D3, 716 1α(OH)ase-null mice demonstrating elevated, 438 1,25(OH)2D3 elevating matrix vesicle, 590–591 tissue’s ability to calcify associated with, 580 Alkalosis, serum phosphate depressed by, 1177 All-trans-retinoic acid (ATRA), in leukemia combination therapy, 1733 Allograft rejection Geminis inhibiting, 1519–1521 immunoregulatory mechanisms inhibiting, 642 19-nor Gemini v. vascularized heart, 1520–1521, 1520t, 1521f 1,25(OH)2D3/analogs inhibiting acute, 640–641, 641t orally active VDR ligand treatments for, 643 VDR ligands/1,25(OH)2D3 analogs inhibiting chronic, 642 VDR ligands v., 640–642 Alopecia in HVDRR, 1208, 1208f, 1210 HVDRR associated with, 1229–1230 in HVDRR phenotype, 621 HVDRR severity correlating with, 1210 in VDR-ablated mice with normal mineral ions, 542 VDR locus mutation/expression in, 1404 VDR-null mice having progressive, 346–347 VDR RXRα heteropartner role supported by, 233 Aluminum (Al) D metabolism influenced by, 1255t, 1270–1271 in osteomalacia, 1270 toxicity from chronic renal disease, 979–981 Alzheimer’s disease (AD), 1779–1780 D analogs treating, 1785 Ameloblasts calbindin-D9K/D28K in, 724–725, 730t D-dependent molecules in, 602–604, 603f enamel elaborated by, 599 American Indians, metabolic bone disease v. lactose intolerance in, 917t Analogs. See also Agonists; Antagonists; Deltanoids; Geminis; Superagonists 2-carbon-modified 19-Nor-1α,25-dihydroxyvitamin D3, 1543–1554 aromatic rings in, 1495, 1497t autoimmunity v. 1,25(OH)2D3, animal models of, 1756, 1757t bone diseases v., 1501–1502 breast cancer cell apoptosis induced by 1,25(OH)2D3, 1665–1666, 1666f, 1672f breast cancer cells/tumors v., 1668–1669 breast cancer cells v. 1,25(OH)2D3, 1663–1667, 1664t breast cancer v. 1,25(OH)2D3, 1663–1672 breast cancer v. preclinical/clinical trials of, 1668–1669 C-ring, 1562, 1563f structures/activities of, 1562, 1564t calcemic v. antiproliferative effect of 1,25(OH)2D3, 1573, 1575t cancer models v., 1499–1500 catabolism v. selectivity of, 1459–1461, 1460f cell-specific catabolism of, 1459–1460 cell-surface receptor v. nongenomic activity of, 1461–1462 chronic kidney disease treated with, 1827 chronic renal failure v. action/tissue specificity of, 1826–1827 chronic renal failure v. future roles of, 1826–1827 clinical development of LEO, 1502–1505 clinical studies of cancer v., 1574 colon cancer cells v. antiproliferative activity by, 1521–1522, 1522f, 1523f, 1523t colon cell proliferation/differentiation/apoptosis v.

INDEX

Analogs (Continued) clinical studies of, 1715 in vitro, 1713–1714 in vivo animal models for, 1714–1715 D-ring, 1562, 1563f structures/activities of, 1562, 1565t DBP affinity v. side chain modifications in, 1457 dermal inflammation/epidermal proliferation/differentiation v., 1781 developing new, 1489–1505 basic screening strategy for, 1489–1490 strategy for, 1489–1492 synthesis strategy for, 1490–1492 development of OCT/ED-71, 1525–1539 dialysis patient PTH secretion controlled by, 1822–1823 differential initiation complex recruitment by, 1455–1456 differential VDR activation by, 1475–1482, 1476f differentiating/antiproliferative v. calcemic activities in, 1449 DNA binding v. selectivity of, 1455 E-ring, 1562–1565, 1563f structures/activities of, 1562–1564, 1566t EAE treated with, 1784 efficacy/safety in clinical development of, 1504–1505 20-epi 1,25(OH)2D3, 1494–1495, 1496t, 1511–1522 transcriptional activity of, 1511 epidermal proliferation marker modulation v. topical, 1781, 1783f in future HVDRR therapy, 1229 hyperproliferative skin disorders v. less calcemic, 1788–1789 hypertension v. low calcemic, 1514–1516 IDBP interactions v. selectivity of, 1462–1463 identifying selective, 1449–2900 immunological diseases v., 1500–1501 immunosuppressive agents synergistically influencing 1,25(OH)2D3, 1520 implications for action mechanism of, 1441–1442 kidney stone formation v., 1351–1352 leukemia v., 1734–1736 leukemia v. 20-epi, 1730t, 1735 leukemia v. C-16-ene, 1730t, 1734–1735 leukemic cell lines v., 1730t level of selectivity in, 1451–1452 ligand binding determining selectivity of, 1452f, 1453 mechanisms for selective actions by, 1449–2911 metabolism of, 1423–1443 biological systems in studying, 1428–1429, 1428t drug design implications in, 1442 examples for, 1429–1440 general considerations in, 1423–1429 implications from studying, 1440–1443 invalid comparisons of in vivo/in vitro, 1442 non-D-related enzymes in, 1428–1429 pharmacokinetic information correlating with, 1440–1441, 1441t questionably valid assumptions regarding, 1442 radioactive analogs in studying, 1428–1429 specific analogs exemplifying, 1429–1440 metabolism of 20-epi-/20-methyl, 1433–1434, 1433f, 1435f metabolism of cyclopropane-ring containing, 1431–1433, 1432f metabolism of homologated, 1434–1436 metabolism of oxa-group containing, 1438 metabolism of unsaturated, 1436–1437 molecular basis for differential action of, 1471–1483 non-psoriatic skin diseases v., 1787–1788 nonsecosteroid CD ring modifications in, 1562–1565 nonsteroidal, 1557–1567

INDEX

Analogs (Continued) in normal/psoriatic skin, 1781–1784 1,25(OH)2D3, 1423–1424, 1425t–1426t animal model diabetes prevented by, 1768–1772 biological activities of, 1495–1502 bisphenols v., 1565 clinical trials of, 1742–1743 diabetes/auto-immune diabetes v., 1772–1773 diabetes early intervention using, 1768–1770, 1768t diabetes v. immune modulators combined with, 1771–1772, 1771t early trials of, 1742 oncology using, 1741–1747 secondary hyperparathyroidism in chronic renal failure v., 1331–1332, 1332f tumors v., 1741–1742 VDR binding/gene expression v., 1495–1498, 1511, 1512f 1,25(OH)2D3 side chain, 1490–1492, 1490f, 1491f, 1492f chain length v. structure-activity of, 1492, 1493t diabetes v., 1768t, 1770–1771 double/triple bonds in, 1492–1493 structure-activity relationships in, 1492–1495 1,25(OH)2D3 structure in secosteroid, 1558, 1558f overview of, 1403–1404 PCa treatment clinical trials of, 1698–1699 PCa v., 1689–1690 PD/AD treated with, 1785 pharmacokinetic data describing, 1440–1441, 1441t pharmacokinetics/metabolism in clinical development of, 1502–1503 preclinical experience with, 1499–1502 prodrug, 1423–1424, 1424t prodrug/1α,25(OH)2D3 analog classification of, 1423 protein interactions v. in vivo selectivity of, 1452–1463, 1452f psoriasis treated with, 1450 psoriasis v. biological effects of, 1781, 1781f psoriasis v. clinical use of 1,25(OH)2D3, 1784–1787 psoriasis v. topical 1,25(OH)2D3, 1784–1785 PTH-suppressing/noncalcemic, 1449–2898 quantifying selectivity of, 1451 rat kidney perfusion in studying D3, 1513–1514, 1516f selection criteria for, 1543–1544 serum DBP interacting with, 1456–1459 stabilization/increased VDR v. activity of, 1456 strong/weak calcemic/noncalcemic, 1440–1441 synthesizing 20-epi, 1491, 1491f, 1492f, 1496t, 1497t synthesizing 22-oxa, 1490–1491, 1492f, 1496t target cell enzymes activating/deactivating, 1441–1442 toxicity due to synthetic, 1357–1359 VDR/coactivator interaction enhanced by 20-epi, 1497–1498 VDR conformational change v. selectivity in, 1452f, 1453 VDR interactions v. binding of 20-epi, 1477–1478, 1478f VDR modulated by, 1482–1483 VDR phosphorylation possibly influenced by, 1454 VDR stabilized by, 1456 VDR transcriptional activity v. 1α,25(OH)2D3, 1474–1475, 1474t Androgen human BPH cells v., 1834 in PCa treatment, 1679–1680 PCa v. D interaction with, 1687–1688 Androgen-independent prostate cancer (AIPC) high dose intermittent 1,25(OH)2D3 v., 1745 1,25(OH)2D3/taxane combinations v., 1745–1746 1,25(OH)2D3 v., 1742 Androgen receptor (AR) in human BPH cells, 1834, 1835f in PCa, 1679–1680

1847 Anemia, in hemodialysis patients associated with VDR polymorphism, 246 Angiogenesis, D compounds v., 1574–1576, 1693 Angiotensin-converting enzyme (ACE), in RAS cascade, 871, 871f Anonymous polymorphisms, VDR gene analysis v., 1138 Antagonists assessment of, 1480–1481 26-carboxylic ester, 1481f, 1482 differential VDR activation by, 1480–1482 lactone, 1481–1482, 1481f structure of representative, 1481, 1481f Anthropometry, VDR gene polymorphisms v., 1143 Anticonvulsants, D metabolism influenced by, 1255t, 1264–1265 Antidiuretic hormone (ADH) IMCD water reabsorption increased by, 558 in RAS cascade, 871–872, 871f Antigen presenting cells (APCs) immunosuppressive treatments not targeting, 1519 1,25(OH)2D3/analog autoimmunity v., 1756 Antimycotics, kidney 1α-hydroxylase activity regulated by, 78 Antioxidant, 1,25(OH)2D3 as, 764t, 765–767 Antituberculous agents, D metabolism influenced by, 1255t, 1271 AP-1. See Activator protein Apa polymorphisms, 1131t–1132t, 1135–1137 BMD v., 1142–1143 Apatite hydroxyapatite v., 477, 478–480, 478f vertebrate tissues with largest/smallest crystals of, 477 APCs. See Antigen presenting cells APLs. See Promyelocyticleukemias Apoptosis BXL-353/GFs v. BPH cell, 1836–1837, 1838f calbindin-D28K regulating, 727 calbindin-D28K v. bone cell, 725, 725f 20-epi D analogs inducing, 1498 hypertrophic chondrocyte, 577 1,25(OH)2D3 inhibiting tumor cells through, 1580 1,25(OH)2D3 protecting epidermal keratinocytes from, 766–767 1,25(OH)2D3 protecting HL-60 cells from, 1732 1,25(OH)2D3 v. PCa cell, 1691 oncogenes/tumor suppressor genes in, 1577–1580 AR. See Androgen receptor Arabs D metabolism in United Arab Emirates/non-Gulf, 794 MVC v. D in veiled Danish, 1812, 1812f Area under the curve (AUC), high dose intermittent regimen v. 1,25(OH)2D3, 1742–1743, 1743f Aromatase, 858 D v. expression of, 859 Asian Indians bone histomorphometry of, 793 bone mass in, 793 D metabolism in, 790t, 792–793 low 25OHD in immigrant, 1026 sunlight exposure v. rickets in, 1066–1067 VDR polymorphisms v. diabetes in, 1146 Asians high hVDR gene frequency in, 244–245 metabolic bone disease v. lactose intolerance in, 917t PCa risk in, 1601, 1601t rickets in immigrant Southeast, 968–971 VDR polymorphism allele frequencies in, 1128t, 1129 Astrocytes, in brain antioxidative/detoxification processes, 766 Atomic force microscopy (AFM), bone crystals imaged by, 482 ATP, muscle cell energy from, 1808 ATRA. See All-trans-retinoic acid Australians, D deficiency in, 795

1848

INDEX

Autoimmune disorders calbindin-D28K Ca buffering v., 726 as D analog therapeutic target, 1451 D analogs treating, 1451 orally active VDR ligand treatments for, 643 VDR ligand immunomodulation mechanisms in models of, 636–640, 637t VDR polymorphisms v., 1146–1148 VDR-RXR heterodimer-activating ligands v., 241 Autoimmune polyglandular syndrome type 1, hypocalcemia v., 1053 Autoimmunity, 1753–1759 D3 in, 1753–1759 immune mechanisms in, 1753–1754 1,25(OH)2D3 action mechanisms in, 1754–1756 regulatory T cells in, 1754 Autosomal dominant hypophosphatemic rickets (ADHR), 463t, 467–468 clinical features, 1189–1190 as disorder of phosphate metabolism, 1189–1190 as hypophosphatemic disease, 1163

B B lymphocytes, antigens recognized by, 631 BAF. See Brahma/SWI2-related gene 1 associated factor Basal cell carcinoma, VDR polymorphisms v., 1146 Basic multicellular units (BMUs) conditions for progress by, 504–505 as instrument of bone remodeling, 501–506, 502f, 503f, 504f originations/progressions/terminations in, 503 posttarget progression by, 503, 505f Bedouins HHRH linkage analysis in, 469 osteomalacia/Ca deficiency in, 1077 serum 25OHD in dark-skinned, 794–795 vitamin D deficiency in, 41 Benign prostatic hyperplasia (BPH) 1α-hydroxylase in cells derived from, 1608–1609, 1610f pathogenesis, 1833–1834 VDR ligands inhibiting, 1833–1840 Beri-beri, nutrition v., 3 β cells clinical trials of D metabolites v., 1765–1767 D deficiency v. in vivo, 1764 D influencing characteristics of, 1767 D metabolites v. in vitro, 1764–1765 D v., 1764–1767 D’s metabolic influences on, 1764–1767 BFR. See Bone formation rate Bile, inactive polar D derivatives in, 1295 Bile acids enterohepatic circulation reclaiming, 863–864 fate of, 863 in lipid digestion/absorption, 863 NRs regulating metabolism of, 865–866 production of secondary, 864–865, 864f VDR as sensor for carcinogenic, 863–869 Binding proteins, intracellular VDRE, 351–361 Binding sites proposed for regulatory trans-proteins, 94–95, 95f rat CYP24A1 proximal promoter region, 93–97, 94f Biopsy, bone, 951–952 Bisphenols, 1557–1561 antiproliferative potency of nonsecosteroid, 1559, 1559t

Bisphenols (Continued) identification/structure/synthesis of nonsecosteroid, 1557–1558 metabolic properties altered in, 1567 mutant VDR v. nonsecosteroid, 1561–1562 1,25(OH)2D3 template v., 1565 in vitro characterization of nonsecosteroid, 1558–1560, 1559t in vivo nonsecosteroid, 1560–1561 Bisphosphonates, D metabolism influenced by, 1255t, 1268 Blacks. See also Africans age-adjusted PCa mortality in, 1600, 1600f bone histomorphometry of, 792 bone mass in, 791–792 colon cancer v. skin pigmentation in, 1709–1710 D metabolism in, 790–791, 790t dietary calcium reduction response by, 778 hypertension v. D deficiency in, 899–900 metabolic bone disease v. lactose intolerance in, 917t 25OHD concentration/vitamin D deficiency in, 38–39 osteoporosis/atraumatic fractures in, 792 osteoporosis/Ca absorption in, 816 PCa in Caucasians/Nigerians/American, 1680–1681 PCa mortality/incidence in American, 1625 PCa risk in, 1601, 1601t PTH-stimulated bone resorption in, 778 Blood cells, VDR in, 1728–1729 Blood pressure. See also Hypertension D endocrine system regulating, 291, 873 RAS regulating, 871 regulation v. D3, 999t sunlight/D v., 873–874, 874f in VDR-null mice, 345 BMAR. See Bone mineral apposition rate BMC. See Bone mineral content BMD. See Bone mineral density 2BMD. See 2β-Methyl-19-nor-(20S)-1,25-dihydroxyvitamin D3 BMUs. See Basic multicellular units Bone, 565–768 activation frequency v. turnover, 959, 1034 age-related loss of, 506–508 age v. thickness of trabecular, 500 aging v. D responsiveness by, 833, 833t apposition rates influenced by 1,25(OH)2D3, 656–657 biopsy, 951–952 black/Caucasian formation rate of, 792 Ca in formation of, 411 calbindin-D9K/D28K in, 724–725 cancellous, marrow composition for, 500–501, 501t CaR in, 558 CKD with low 1,25(OH)2D3 or 1,25(OH)2D3/VDR resistance in, 1325 collagen synthesis regulation in, 704–705 D action redundant on, 429 D actions on mineralization of, 565–566 D analog nongenomic pathway in, 1462 D/D metabolites acting on, 565–571 D deficiency impairing resorption of, 776, 777, 777f D-deficient, 487–488 D direct/indirect effects on, 1042–1044 D insufficiency impact on, 1089–1090 D intake v. winter loss of, 1090t D mobilizing mineral in, 665–666, 666f D modulating formation of, 327–328 deformities caused by D deficiency, 972f–973f, 973, 974f density correlated with VDR gene polymorphisms, 184, 200 density influenced by exercise, 796 density seasonal decline v. D/Ca, 998

INDEX

Bone (Continued) dento-alveolar complex of, 599–605 development v. perturbed D metabolism, 107–108 disease in gastrointestinal disorders, 1297–1299 disorders v. gastrointestinal/hepatobiliary disease, 1293–1307 driving transfers of Ca2+ in mineralization of, 774, 775f estrogen/PTH/D metabolism v. age-related loss of, 1107–1109 evolution of D-related disease of, 1035–1039 formation rates, 958 formation selectively induced by deltanoid, 1415f, 1416 fracture aluminum toxicity associated with, 980–981 D3 dose preventing, 997, 998f in Looser’s zones, 973 manifesting osteoporosis, 1101 normal 25OHD v., 1022–1023 rate v. D supplementation, 1094–1096, 1095f, 1096f repair influenced by 24,25(OH)2D, 108 risk v. aging/activity level, 508 risk v. D, 1813–1814 risk v. serum 25OHD, 1370, 1370f VDR polymorphisms v., 1143–1144 in vertebrae v. 12523, 1111–1112 glucocorticoid therapy causing loss of, 1239 growth in central skeleton, 499, 500f health supporting higher D levels, 1370 hydroxide-deficient apatite crystals in, 477 24-hydroxylase enzyme regulation in, 92 irreversible loss of, 506, 507f, 508f length, 499–500 loss influenced by age-related PTH increase, 1107, 1108f partitioned into horizontal/vertical components, 506–507, 509f VDR polymorphisms associated with alveolar, 246 marrow rejection inhibited by 1,25(OH)2D3/analogs, 641t matrix proteins influenced by D, 479t, 484–487 matrix synthesis gene expression v. 1α,25(OH)2D3, 715 metabolism v. ED-71/17β-estradiol, 1536, 1536f mineral content/density v. pregnancy, 840–841 mineral metabolism influenced by HEP 187, 1493, 1494t mineralization assessed by histomorphometry, 954–957 mineralization mechanisms in, 480 mineralization v. vitamin D, 5–6 modeling v. remodeling, 497–498, 498t OCT influence on, 1527–1529, 1529t OCT v. formation rate of cortical, 1531, 1533f 1,25(OH)2D3 regulating type I collagen expression in, 703–705 1,25(OH)2D3 v. mineral homeostasis in, 220–221, 221f 1,25(OH)2D3 v. 24,25(OH)2D3 in endochondral formation of, 583 1α,25(OH)2D3 VDRnuc in, 385t osteocalcin gene expression modulation during, 328–329 osteomalacia v. properties of, 489–490, 490f Pi flux in human, 453–455, 454f Pi transport in, 465–467 progenitor cell commitment v. 1,25(OH)2D3, 650f, 655–656 PTHrP expression in, 739t remodeling, 497–511 activation frequency, 504–505 balance assessment, 959–961 BMUs, 501–506, 502f, 503f, 504f bone age/mineral density v., 503f, 506 cycle in cancellous bone, 506, 507f cycles v. bone loss, 507, 509f focal imbalance, 506, 508f influences mediated by strain, 508 initiated by fatigue microdamage, 503

1849 Bone (Continued) osteonal/hemiosteonal, 502–503 purpose, 500–501 in structural/metabolic bone, 501, 501t remodeling markers, 921t remodeling periods, 959 renal osteodystrophy v. periosteal formation of, 979 resorption/formation v. ED-71, 1537, 1537f, 1538 resorption in GHS rats, 1349 resorption in vitro v. 1,25(OH)2D3, 568–569, 569f resorption in vivo v. D, 569–570, 570f resorption induced by 1,25(OH)2D3, 680–681, 681f resorption influenced by D, 568–570 resorption rates, 958 resorption signal transduced by RANKL/ODF, 672–673 RUNX2∆C-homozygous mice not forming, 335 steroid receptors/actions in, 1239–1241 strength structural determinants, 961 structural/cellular basis for growth of, 498–500 structure assessed by histomorphometry, 961–963 target genes v. D, 566–568 turnover, 497 assessment, 958–959 dissimilarly affected by PTH/1,25(OH)2D3, 510 turnover/density v. D insufficiency in elderly, 825 two types mineralization in, 1030–1031 VDR in growth/maturation/remodeling of, 430f, 433 VDR overexpression strengthening, 488–489, 489f volume v. VDR ablation, 343, 344t Bone cells. See also Osteoblasts; Osteoclasts in bone modeling/remodeling, 497–511 calbindin-D28K v. degeneration of, 725 coordinated activities of, 497, 498t Bone disease. See also specific bone diseases approaching patients with metabolic, 913–928 D analogs v., 1501–1502 gastrointestinal conditions associated with, 1299–1303 gastrointestinal disease with metabolic, 1298–1299 non-Ca/D factors in gastrointestinal disease-associated, 1298 Bone formation rate (BFR) in HVO, 1035–1036 OCT v., 1528, 1529t Bone histomorphometry, 951–963 biopsy, 951–952 adverse effects, 952 indications, 952 procedure, 951–952, 951f bone remodeling indices in derived, 953, 953t primary, 953, 953t bone structure assessment by, 961–963 three-dimensional, 962–963 two-dimensional, 961–962 dynamic indices from mineralizing perimeter/surface, 958 future developments in, 963 limitations, 954 methodology, 952 remodeling balance assessment by, 959–961 bone formation v., 959–960, 959f bone resorption v., 960–961, 960f terminology, 952–954 referents/abbreviations, 952–953, 953t theoretical considerations influencing, 952 in vivo tetracycline labeling in, 954, 955, 955f Bone mass D in GIO treatment v., 1245–1247

1850 Bone mass (Continued) ED-71 v. ovariectomized rat, 1537, 1537f IH patients having low, 1342–1343 loss reduced by D supplementation, 1094 2MD v. ovariectomized rat, 1547, 1548f Bone matrix proteins, 712, 712f 1α,25(OH)2D3 regulating, 712, 713t target genes in, 711–717 VDREs identified in, 712, 713t Bone mineral apposition rate (BMAR), D metabolites influencing, 656–657 Bone mineral content (BMC) maternal D status v. infant, 842–843 1,25(OH)2D3 restoring PDDR, 1202, 1202f Bone mineral densitometry, in evaluating bone metabolic disease, 922–924 Bone mineral density (BMD) alendronate v., 1248 Cdx2 polymorphism v., 1130 corticosteroid/Ca/D v., 1245–1247, 1247f D metabolites v., 1245, 1246t in defining normal serum 25OHD, 1022, 1022f in ED-71 clinical trials, 1538–1539, 1538f intestinal CA absorption/VDR polymorphisms v., 1141–1142 1,25(OH)2D v., 1107 1αOHD3 v., 1247–1248 in osteoporosis/fracture, 1141, 1143–1144 prednisone/Ca/D v., 1245, 1246f VDR gene polymorphism v. lumbar spine, 243 VDR gene polymorphism v. women’s, 243 VDR polymorphisms v., 1142–1143 Bone scintigraphy, skeletal abnormalities uncovered with, 922 Bone sialoprotein (BSP) expression suppressed by 1,25(OH)2D3, 486 1α,25(OH)2D3 suppressing, 715 production v. D, 567–568 BPH. See Benign prostatic hyperplasia BPH cells AR/5α-reductase/VDR in, 1834, 1835f BXL-353/GFs v. apoptosis/Bcl-2 expression in, 1836–1837, 1838f BXL-353 v., 1836–1837, 1837f KGF/IGF-I/T v., 1834, 1835f Brahma/SWI2-related gene 1 associated factor (BAF) components in WINAC, 308, 308f in 1,25(OH)2D3-liganded VDR-RXR heterodimer transactivation, 238, 239f Brain detoxification v. 1,25(OH)2D3, 1783 development influenced by D3, 999t homeostasis v. 12523, 1784, 1785f 1,25(OH)2D3 antioxidant activities in, 764t 1,25(OH)2D3 in development/disorders of, 1779–1785, 1785f 1,25(OH)2D3 v. tumors in, 1783–1784 PTHrP expression in, 739t Breast cancer angiogenesis/invasion/metastasis v. 1,25(OH)2D3, 1666–1667 cell proliferation v. D analogs, 1451 D3 protecting against, 999t D sensitivity, 1667–1668 EB1089 v., 1499 epidemiology of D v., 1671 natural ligands v. synthetic analogs treating, 1668 1,25(OH)2D3/analog antiproliferative effects v., 1459, 1460f 1,25(OH)2D3/analogs v., 1663–1672 1,25(OH)2D3 prooxidant activities in, 764–765, 764t

INDEX

Breast cancer (Continued) 1,25(OH)2D3/VDR modulating cell proliferation/apoptosis in, 857–858 preclinical studies of D preventing, 1670 prevention and D, 1669–1672, 1670t prognosis v. tumor VDR expression, 1668 PTHrP expression in, 739t PTHrP production v., 742 PTHrP v. osteolytic bone metastasis in, 744 risk, for nationality, 245 risk v. VDR polymorphisms, 1671–1672 VDR polymorphisms v., 1145 Breast cancer cells apoptosis-independent 1,25(OH)2D3 inhibition of, 1580 apoptosis induced by 1,25(OH)2D3/analogs in, 1665–1666, 1666f, 1672f D compounds v. oncogenes/tumor suppressor genes in, 1664–1665 D resistance in, 1667–1668 EB1089 inhibiting IGF-I-stimulated, 1498 EB1089 v., 1669 1,25(OH)2D3/analog actions on, 1663–1667, 1664t 1,25(OH)2D3/deltanoid-induced differentiation modeled in, 1637t 1,25(OH)2D3/EB1089 arresting progression of, 1663–1664, 1672f 1,25(OH)2D3/EB1089 v. estrogen action in MCF-7, 1665 1,25(OH)2D3 v. proliferation of, 1663–1665, 1672f VDR expression/regulation in, 1667 Breast milk Ca absorption fraction from, 813, 814t maternal D/Ca intake v. D/Ca in, 846–847 neonatal hypocalcemia incidence v., 841–842, 842f substitutes v. rickets risk, 1066 Brown tumors, in hyperparathyroidism, 978f, 979 Brush border membrane lipids v. Ca entry, 415–416 membrane Na/Pi transport reduced in Npt2a KO mice, 462 membrane Pi transport, 455 permeability increased by D, 414 Bsm polymorphisms, 1131t–1132t, 1135–1137 BMD v., 1142–1143 BSP. See Bone sialoprotein Burns, hypophosphatemia associated with, 1177 BXL-353. See 1,25-Dihydroxy-16-ene-23-yne D3

C C-3 epimerization, 1,25(OH)2D3, 24, 25f C-24 oxidation pathway, to calcitroic acid, 1424–1427, 1426f C-cells, CaR stimulating CT secretion by, 556 c-Fos, osteoclastogenesis v. RANKL-induced expression of, 677–678 c-Src family proteins, osteoclast differentiation v., 677, 677f C27 sterol 27-hydroxylase, 54–55 mitochondrial D3 25-hydroxylase is, 55 C-terminal extension (CTE), hVDR DBD/VDRE alternative positions, 231f, 232 Ca. See Calcium Caffeine, D metabolism influenced by, 1255t, 1271–1272 Calbindin, 417–419, 721–730 Ca absorption v., 418 Ca entry process initiation v., 418–419 in cytosolic Ca diffusion, 419 distribution, 730t EF-hand HLH structural motif characterizing, 722, 722f

INDEX

Calbindin (Continued) functions of, 418–419 gene expression regulation, 728–730 localization/functional significance, 722–728 1,25(OH)2D3 v. in brain, 1781 synthesis, 418 VDRnuc responsible for production of, 384–385 Calbindin-D9K, 523–525, 721–730 in placenta/yolk sac/uterus, 726, 730t structure, 524 Calbindin-D28K, 523–525, 721–730 EF hands contained in, 722, 723f in egg shell gland, 726 enzymes activated by, 727–728 in nervous system, 726–727 PMCa pump regulated by, 524 regulatory function v. apoptosis, 727 structure, 524 Calbindin-D9K gene genomic organization, 729 1,25(OH)2D3 regulating, 729 steroids/other factors regulating, 729–730 Calbindin-D28K gene genomic organization, 728 1,25(OH)2D3 regulating, 728 regulation, 728–729 steroids/other factors regulating, 728–729 Calcidiol. See 25-Hydroxycholecalciferol Calcimimetics, CaR activated by, 554 Calcipotriol breast cancer cells/tumors v., 1668 discovery, 1489 face/skin flexures treated with, 1786 leukemic cells v., 1734 metabolism, 1431–1433, 1432f monotherapy/combination therapy v. psoriasis, 1504, 1504t as noncalcemic analog, 1440, 1441t pharmacokinetics/metabolism in clinical development of, 1495t, 1502–1503 psoriasis treated with OCT v., 1533, 1533f psoriasis treated with topical, 1784 psoriatic lesions v. topical, 1450 scalp psoriasis treated with, 1785–1786 skin irritation in psoriasis treatment with, 1785 structure-activity relationships, 1493–1494, 1495t toxicity, 1358–1359 Wittig reaction in synthesis of, 1490, 1491f, 1495t Calcitonin (CT) C cells producing CGRP and, 689 CaR stimulating C-cell secretion of, 556 CYP24A1 expression regulated by, 93, 94f CYP24A1 promoter expression increased by, 100 D metabolism influenced by, 1254t, 1256–1257 hypocalcemia from overzealous use of, 1057 induction during CYP24A1 promoter mutational analysis, 96–97, 97f intestinal CYP24 expression v., 21 kidney 1α-hydroxylase activity regulated by, 77 levels regulated by D3, 689–691 PTHrP expression/production stimulated by, 741t renal Pi excretion decreased by, 516t Calcitonin (CT) gene D influencing thyroid C cell, 687–697 model system for 1,25(OH)2D3 v. transcription of, 690–691 transcription down-regulated by D, 693–696 transcriptional regulation, 689, 690f, 691–693

1851 α-Calcitonin gene-related peptide (CGRP), C cells producing CT and, 689 Calcitriol. See 1,25-Dihydroxyvitamin D3 Calcitroic acid C-24 oxidation pathway to, 1424–1427, 1426f in calcipotriol metabolism, 1432, 1432f 1,25(OH)2D2 metabolism to, 25–26 Calcium (CA), 1,25(OH)2D3 v. intestinal transport of, 679f Calcium (Ca). See also Calciuria; Eucalcemia; Familial hypocalciuric hypercalcemia; Hereditary hypophosphatemic rickets with hypercalciuria; Hypercalcemia; Hypercalciuria; Hypocalcemia; Normocalcemia; Transcaltachia; Vascular calcification; Williams syndrome absorption by paracellular path, 421–422 absorption by VDR-null mice functional aspects of, 433–435, 434f molecular aspects of, 435–437 absorption enhancers in food, 818 absorption in infancy/childhood, 811–818 absorption in Npt2-null/Hyp mice, 440 absorption in postmenopausal/age-related osteoporosis, 1110 absorption increased by alendronate, 1244 absorption influenced by 1,25(OH)2D3/VDR binding, 220–221, 221f absorption location/timing in gut, 778–779, 779f absorption v. IH, 1341, 1341f absorption v. intake, 779–781, 779f, 780f absorption v. prednisone/glucocorticoids, 1244–1245, 1245f absorption v. serum 1,25(OH)2D in IH, 1342, 1343f absorption v. VDR polymorphism/BMD relationship, 1141–1142 absorption v. vitamin D3, 5 absorptive efficiency v. transit time, 778 absorptive input, 778–782 active/passive absorption of, 781–782 adolescents v. inadequate intake of, 816 adult human body, 773 aging v. intestinal absorption of, 831–832 aging v. total/ionized, 829, 829t balance disordered in sarcoidosis, 1381–1387 balance v. lactation, 4 binding proteins in brain v. 1,25(OH)2D3, 1781 biological importance of, 411 in cancer risk epidemiology, 1617–1629 case-control/cohort studies of colorectal cancer v., 1622 CKD impairing 1α-hydroxylase induction by, 1315–1316 colorectal adenoma v., 1622–1623 colorectal cancer v., 1621–1624 cytosolic transfer of, 417–419 calbindins in, 419 D deprivation/repletion v. muscle, 1810f, 1811 D enhancing paracellular absorption of, 412f, 421–422 D3 improving absorption of, 999t D/25OHD3/1,25(OH)2D3 influencing renal handling of, 518–519 in dairy products v. PCa, 1627–1629 demand response of ECR, 776–778 dietary deficiency of, 1074–1076 duodenal/ileal transcellular/paracellular absorption of, 422–424, 423f ECaC1/2 expression in fine-tuning transport of, 432 in eggshells/embryos, 851 entry step v. absorption, 416–417, 417f entry step v. calbindins, 418–419 extrusion across basolateral membrane, 419–421, 420f falls v. D3 and, 1814, 1814f fetuses/neonates v. low maternal intake of, 843, 843f food fortification with, 817–818

1852 Calcium (Continued) fractional absorption from human milk/formula, 813, 814t GHS rat response to low dietary, 1349, 1350f homeostasis in muscle cells, 1807-1808, 1808f homeostasis v. caffeine, 1271–1272 homeostasis v. cardiovascular disease, 901 in HVDRR therapy, 1227–1228 IH decreasing renal reabsorption of, 1342 IH v. restricted dietary, 1351 intake/absorption v. age, 1103–1106 intake v. absorption, 412–413, 412f, 413f, 782, 782f, 783–784, 784f intervention studies of colorectal adenoma v., 1623–1624 evidence interpretation in, 1623 implications of, 1624 mechanism in, 1623–1624 intestinal absorption of estrogens’ genomic effects v., 444–445 1,25(OH)2D3 v., 411–424 VDR WT/KO mouse gestation and, 440–442, 441f, 441t, 442f VDR WT/KO mouse lactation and, 441f, 441t, 442–444, 442f, 443f jejuno-ileal bypasses reducing serum, 1303 in keratinocyte differentiation, 615–619, 615f keratinocyte differentiation v. intracellular, 616 kidney/parathyroid VDR regulated by, 522–523, 525t KO mice/human intestinal absorption of, 429–437, 446t long-term OCT administration v., 1530–1531, 1530f low Ca intake v. total body, 781 macrophages unresponsive to, 1382–1383, 1383f malabsorption in gastrointestinal disease, 1297 maternal Ca intake v. breast milk, 847, 847f mechanisms of age-related changes in, 1104–1106 metabolism during lactation/weaning, 843–846 metabolism during pregnancy, 839–840 metabolism v. IH therapeutics, 1351 molecular bases for entry of, 415–416 2MP/2MbisP/2Mpregna v. serum, 1552–1553, 1553f nuclear import regulated by, 371–374 obligatory losses of, 774–776 OCT v. renal failure rat, 1527, 1528f 1,25(OH)2D3 promoting absorption of, 291 1α,25(OH)2D3 regulating muscle, 886–887 1,25(OH)2D synthesis influenced by, 828 1,25(OH)2D3 v. intestinal transport of, 678–680 oral supplements, 1059t PTH defending serum, 1050 PTH/1,25(OH)2D3 in maternal regulation of, 859 PTH v., 552, 553f PTHrP expression/production stimulated by, 741t reabsorption by kidney, 515–516 regulation in ECF, 774–778, 775f renal reabsorption, 6 reproduction v. active absorption of, 440–445 restriction/24-hydroxylase catabolism in D toxicity, 1365–1367, 1366t restriction v. renal 1α-hydroxylase in D toxicity, 1365, 1365t serum 25OHD values v. intake of, 1023–1024 supplementation in metabolic bone disease, 927 systemic/intracellular homeostasis of, 751 transcellular absorption of, 414–421 CA entry v., 415–417, 417f calbindin v., 418 cytosolic transfer, 417–419 D in initiating, 414 extrusion across basolateral membrane in, 419–421, 420f

INDEX

Calcium (Continued) in intestine, 414 model for, 423f, 424 thermodynamic parameters, 414 transport in GHS rats, 1348, 1348f, 1348t VDR ablation v., 341 VDR expression affected by, 202t, 204–205 vesicular transport of, 422 VSCCs inactivated by, 755–756, 756f Calcium channel blockers, D metabolism influenced by, 1255t, 1269 Calcium deficiency, hypocalcemia due to dietary, 1056 Calcium economy, 773–778 body calcium compartments in, 773–774 D in, 773–785 fetal bone mineral accretion v., 841, 841f HVDRR Ca absorption efficiency v., 782 Calcium-sensing receptor (CaR), 551–559 amino acids allosterically activating, 554 as “calciostat,” 552–554 fetal role of, 854 intracellular signaling, 551–554 pathways modulated by, 554 isolation, 551 keratinocytes’ alternately spliced variant of, 616 in keratinocytes’ Ca response, 615f, 616–617, 620f in kidney, 556–558 1,25(OH)2D3 regulating, 543 parathyroid cell proliferation v., 545 in parathyroid gland, 554–556 parathyroid VDR interacting with, 556, 556t in placental Ca transport, 853–854 predicted structure, 551–552, 552f PTH secretion and, 537–538 Calcium transport ECaC1/2 expression in fine-tuning, 432, 432f, 432t genes, VDR-null mouse Ca absorption v. expression of, 435–437, 436t, 437t in GHS rats, 1348 1,25(OH)2D3 v. intestinal, 678–680, 679f vesicular, 422 Calciuria, calbindin-D28K in, 724 Calcospherulites, 1030 Calmodulin Ca entry v., 416 in keratinocyte growth/differentiation, 619 Calreticulin, 543 increased in hypocalcemic rats, 544 1,25(OH)2D3 PTH gene regulation v., 543–544 VDR DBD mutations in binding site of, 1220 Cancer. See also specific types of cancer clinical studies of D analogs v., 1574 clinical studies of D compounds v., 1573–1574 D analog therapy v., 1450–1451 D v., 1571–1577 epidemiology of, 1571–1573 diet/Ca intake/D relationship with, 1572 EB1089 v., 1499–1500 efficacy/safety of D analogs treating, 1504–1505 growth/development v. 1,25(OH)2D3, 1573, 1574t, 1575t Hopkins QW-1624F2-2 hybrid deltanoid v., 1417, 1417f hVDR gene polymorphisms v., 245 models v. D analogs, 1499–1500 nonsecosteroidal D mimics v., 1565 1,25(OH)2D3/analog therapeutic potential v., 169 1,25(OH)2D3/carboplatin combinations v. advanced, 1745, 1746f

INDEX

Cancer (Continued) risk epidemiology v. D/Ca, 1617–1629 soy consumption/CYP24A1 expression v., 98 sunlight/vitamin D v., 42 VDR in, 1571, 1572t VDR polymorphisms v., 1145–1146 Cancer cells D resistance/metabolism in, 1583–1584 EB1089 v. PTHrP production in, 744 24-hydroxylase enzyme regulation in, 93 1α-hydroxylase v. 1,25(OH)2D3 PTHrP inhibition in, 744, 745f 1,25(OH)2D3 inducing apoptosis of, 1580 PTHrP expressed by normal/cancer, 739–740, 739t PTHrP stimulators/inhibitors in normal/cancer, 740–742, 741f CaP. See Prostate cancer CaR. See Calcium-sensing receptor CaR gene, abnormalities v. hypocalcemia, 1053 Carbonate, in bone apatite, 480 Carboplatin, 1,25(OH)2D3 in combination with, 1745, 1746f Cardiomyocytes, D signaling in, 903 Cardiovascular disease, VDR polymorphisms v., 1148 Cardiovascular medicine, 899–905 D dose/response curve, 905, 905f Cardiovascular system D signaling in health of, 899–901 epidemiology of, 899–900 functions v. D, 874–875 indirect D actions on, 901–902 VDR genetics v. disease of, 900–901 Cartilage. See also Growth plates bone forming from calcified, 575 Ca/Phosphate supply v. D metabolites in, 581 calbindin-D9K/D28K in, 730t changes during maturation, 577–578 embryonic variations in, 575 genomic/nongenomic regulation by 1,25(OH)2D3/24,25(OH)2D3, 575–591 metabolism regulated by D, 579–582, 581t mineralization mechanisms in, 480 1,25(OH)2D3 v. 24,25(OH)2D3 in, 582–583 Cartilage oligomeric matrix protein (COMP), D metabolites bound by, 579 Catabolism, 1,25(OH)2D3, 1424–1427 C-26 hydroxylation/26,23-lactone formation in, 1427, 1427f Caucasians. See also Whites; specific nationalities age-adjusted PCa mortality in, 1600, 1600f bone mass in, 793 Cdx2 polymorphism in, 1125f, 1126 D deficiency in American juvenile, 1067 graphical LD display across VDR gene in, 1127, 1128f low hVDR gene frequency in, 244–245 osteoporosis/atraumatic fractures in, 792 osteoporosis/Ca absorption in, 816 UV v. PCa in American, 1601–1602, 1602f, 1603f VDR gene association studies in, 1141 VDR polymorphism allele frequencies in, 1128t, 1129 VDR polymorphisms v. breast cancer in, 1145 VDR polymorphisms v. colon cancer in, 1145–1146 VDR polymorphisms v. diabetes in, 1146 VDR polymorphisms v. PCa in, 1139t, 1145 VDR polymorphisms v. psoriasis in, 1146 CB1093, metastasis v., 1576 CB 966, synthesis, 1490, 1491f, 1494t CC. See Chief complaint Cdk5, in monocytic differentiation marker expression, 1639

1853 Cdk inhibitory (CDKI) proteins Cdk regulated by, 1643–1644, 1643f 1,25(OH)2D3/deltanoids up-regulating, 1646–1649, 1647t–1648t Cdk5/p35 pathway, in 1,25(OH)2D3 differentiation signal propagation, 1639 CDKIs. See Cdk inhibitory proteins Cdks. See Cyclin-dependent kinases Cdx2 polymorphism, 1130, 1131t bone fracture risk v., 1144 in Caucasians, 1125f, 1126, 1130 in Japanese, 1126, 1130 polymorphism functionality parameters associated with, 1135f, 1137 Celiac disease Ca malabsorption in, 1297 clinical features of, 1300–1301 D deficiency development in, 1301 management, 1301 Cell cycle apoptosis v. differentiation at block in, 1580 compartments/checkpoints, 1640–1641, 1641f D influencing, 1577–1580, 1578f–1579f deltanoid-induced differentiation v., 1640–1650 deltanoids modulating events in, 1646–1650 differentiation v., 1635–1651 20-epi D analogs regulating, 1498 G1 arrest v. c-Myc expression down-regulation, 1649–1650 G1 block controlled by retinoblastoma protein, 1649 G2 retardation v. polyploidization, 1650 G1/S block, 1646–1650, 1646f machinery features, 1640–1642 1,25(OH)2D3/EB1089 in arrest of breast cancer, 1663–1664, 1672f oncogenes/tumor suppressor genes in, 1577–1580 Cell cycle progression G2/M phase transition in, 1645, 1645f G1/S phase transition in, 1643–1644, 1643f mechanisms driving, 1642, 1642f regulation of, 1643–1645 S phase DNA licensing in, 1644–1645 Cell cycle traverse Cdks controlling, 1642, 1642f checkpoints controlling, 1641, 1641f 1,25(OH)2D3 inhibiting, 1646–1649, 1646f Cells. See also specific types of cells antioxidant mechanisms of, 762 D in oxidative stress response of, 761–768 DBP associated with, 123 higher D levels in health of, 1370–1371 HVDRR studies using various, 1218 1,25(OH)2D3 v. growth/differentiation of cells, 696 PTHrP expressed by normal/cancer, 739–740, 739t PTHrP stimulators/inhibitors in normal/cancer, 740–742, 741f redox state of, 761–762 ROS causing death of, 761 cementoblasts, cementum elaborated by, 599 Cementum bone sharing matrix components with, 602 D bioinactivation causing hypomineralization in, 602 in tooth root/periodontium, 601–602 Central nervous system (CNS) neurodegenerative disease etiology/physiopathology in, 1779–1780 1,25(OH)2D3 actions in, 1781–1783 1,25(OH)2D3 v. tumors in, 1783–1784 VDR/1,25(OH)2D3 targets in, 1780–1781

1854 Cerebrotendinous xanthomatosis (CTX), CYP27A1 expression v., 58–59 CGRP. See α-Calcitonin gene-related peptide Checkpoints, cell cycle traverse controlled by, 1641, 1641f CHF. See Congestive heart failure Chickens D deficiency v. fertility in, 854–855 embryonic development/egg hatchability v. D in, 851–852 Chief complaint (CC), in metabolic bone disease diagnosis, 915 Children. See also Infants; Puberty adolescent Ca absorption in, 815–817 factors influencing, 815–816 early hypocalcemia in, 1036t, 1037 inadequate Ca/D intake in, 816 nutritional rickets in, 968, 969f soda v. Ca intake by, 817 Ca absorption in prepubertal, 815 Ca/D-fortified foods for, 817–818 CA intake by developing countries’, 817 D deficiency and nutritional rickets in, 1065–1077 D deficiency/Ca absorption in, 811–818 deficiency in breast-fed, 1067 PTH resistance in rachitic, 1069 Chinese bone mass in, 793 Bsml RFLP v. nephrolithiasis in, 1142 Buddhist vegetarians v. metabolic bone disease, 917t Ca v. colorectal cancer in, 1622 D metabolism in, 793 PCa mortality/incidence in, 1624–1625 sunlight exposure v. rickets in, 1066–1067 VDR polymorphism v. PCa risk in, 1682 VDR polymorphisms v. colon cancer in, 1145 Chloride channel (ClC)-5, in Ca homeostasis, 159 Chloride channels, 1α,25(OH)2D3 opening ROS 17/2.8 cell, 393, 394f Cholecalciferol. See Vitamin D3 Cholesterol bile acids converted from, 863, 864, 864f D3 derived from, 931 D endocrine system having low concentration of, 1001, 1001f Chondrocytes in bone growth, 498–499, 499f cartilage produced by, 575 cell maturation specific Ca2+ ion kinetics in, 585 chickens/rats modeling growth plate, 577 D metabolite action mechanisms in, 587–588, 587f, 588f lineage of, 575–577 maturation of growth plate, 108 maturation rates of, 575–576 membrane signaling in, 586–587 mineralization v. hypertrophic, 576–577, 576f 1,25(OH)2D3-deficient matrix vesicles produced by, 580 1,25(OH)2D hormone v. differentiation/function of, 112–113 phospholipids altering membrane fluidity in, 585 stereospecific membrane receptors in, 586 zone of maturation v. D3 metabolite response of, 582 Chondrodysplasias, rickets v. differential diagnoses for, 984, 987f Chromatin architecture in nucleus, 328, 329f, 330f ATP-dependent remodeling complexes for, 267–268 binding v. NR co-repressors, 292, 293f in comodulator activity integrated model, 300, 300f Mediator-D HAT activity v. remodeling, 267, 295, 299–300 osteocalcin gene organized by, 330–332

INDEX

Chromatin (Continued) PBAF repressing, 238–240, 239f remodeling facilitating promoter accessibility/regulatory integration, 329–332 remodeling in VDR promoter/VDR targeting, 305–312, 306f remodeling in VDR transcription model, 268–269, 269f remodeling v. DNA repair/replication, 307 RUNX elements in remodeling, 332 WINAC in reorganizing, 310, 311f WINAC inducing ATP-dependent remodeling of, 308–309, 309f Chromatin immunoprecipitation (ChIP) coactivator/transcription factor assembly details from, 299–300 NCoA62SKIP/VDR-activated transcription influence shown by, 297 OC gene promoter organization remodeling v., 332 VDR/WINAC association shown by, 309 Chronic kidney disease (CKD). See also Chronic renal failure abnormal 1,25(OH)2D3/VDR activity in, 1320–1322, 1320f bone loss in, 1325 D bioactivation to 1,25(OH)2D in, 1313–1317 decreased renal mass/GFR v., 1313–1314, 1314f PTH/Ca induction of renal 1α-hydroxylase v., 1315–1316 renal 1α-hydroxylase substrate availability v., 1315 defective homologous VDR up-regulation in, 1317–1318 mechanisms impairing 1,25(OH)2D3/VDR transcriptional activity, 1320–1321, 1320f megalin expression/1,25(OH)2D3 resistance in, 1319 1,25(OH)2D/VDR action altered in, 1317–1322 TGFα/EGFR expression v. parathyroid hyperplasia in, 1323–1324, 1323f, 1324f uremia v. 1,25(OH)2D3/VDR transcriptional activity in, 1320–1321, 1321f VDR polymorphisms v. expression/function in, 1318–1319 Chronic renal failure D therapy in, 1327–1332 1,25(OH)2D3 v. secondary hyperparathyroidism in, 1327–1331, 1330f, 1331–1332, 1332f Chugai OCT cell-specific catabolism of, 1459–1460 DBP affinity v. plasma levels, 1457–1458, 1457f hypercalcemia of malignancy v., 1450–1451 mouse primary immune response v., 1451 PTH suppression/low calcemic activity by, 1449–2898, 1450f SAR in design of, 1412, 1413f Cimetidine, D metabolism influenced by, 1255t, 1270 Cirrhosis, hypocalcemia/secondary hyperparathyroidism in primary biliary, 1056 CKD. See Chronic kidney disease ClC-5. See Chloride channel-5 CNS. See Central nervous system Coagulation cascade, cardiovascular disease v. D regulating blood, 902 Cod liver oil historical importance of, 70 rickets v., 4 Collagen in bone matrix scaffolding, 485 D influencing, 485–486, 486f factors modulating synthesis of, 703–704 molecular mechanisms regulating, 705–706 1,25(OH)2D3 inhibiting organ culture synthesis of, 704 1,25(OH)2D3 regulating expression of type I, 703–705 1,25(OH)2D3 regulating matrix proteins in type I, 703 synthesis regulation in bone, 704–705 Colon dietary Ca absorption in, 424

INDEX

Colon (Continued) 1α-hydroxylase expression in human, 1718–1719, 1719f 1α-hydroxylase/VDR/Ki-67 coexpression in, 1718–1719, 1719f 1α,25(OH)2D3-mediated rapid response in, 386t 1α,25(OH)2D3 VDRnuc in, 385t VDR’s clinical relevance to, 246–249 Colon cancer D supplementation v. fat-promoted, 246 D v., 1709–1721 D/VDR v., 866–867 EB1089 v., 1500 20-epi D analogs v. IGF-II in, 1498–1499 high-fat diet v., 866–867, 1572 rationale for Gemini analogs treating, 1521–1522 sunlight/latitude v. death rate from, 1571–1572 sunlight v. incidence of, 246 VDR/EGFR expression in, 1710–1711, 1711f, 1711t VDR gene polymorphisms v., 1619–1620 VDR polymorphisms v., 1145–1146 VDR polymorphisms v. development of, 1712 Colon cancer cells D metabolism in lines/primary cultures of, 1715–1717, 1716f metabolic/catabolic D hydroxylases expressed in, 1717 1,25(OH)2D3/analog antiproliferative activity v., 1521–1522, 1522f, 1523f, 1523t 1,25(OH)2D3/deltanoid-induced differentiation modeled in, 1637t 1,25(OH)2D3/Ro-26–9228 v., 1480, 1480f Colon cells D metabolism in normal/neoplastic, 1715–1719 1,25(OH)2D3 action on neoplastic, 1710–1715 1,25(OH)2D3/D analogs v. proliferation/differentiation/apoptosis of, 1712–1715 clinical studies of, 1715 in vitro, 1712–1714 in vivo animal models for, 1714–1715 VDR in normal/malignant, 1710–1712 Colorectal adenoma Ca v., 1622–1623 D plasma markers v., 1619 dietary D v., 1619 intervention studies of Ca v., 1623–1624 evidence interpretation in, 1623 implications of, 1624 mechanism in, 1623–1624 VDR gene polymorphisms v., 1619–1620 Colorectal cancer, 1618–1619 Ca v., 1621–1624 case-control/cohort studies of Ca v., 1622 circulating D v., 1618–1619 D dietary/supplementary intake v., 1618 dietary Ca v., 1621–1622 risk epidemiology v. D/Ca, 1617–1629 sunlight exposure v., 1618 Colorectal neoplasms, 1618–1624 Combination therapy advanced cancer v. 1,25(OH)2D3/carboplatin, 1745, 1746f in animal PCa models, 1689 calcipotriol in psoriasis, 1504, 1504t cancer cell differentiation v., 1582–1583 falls v. D3/Ca, 1814, 1814f leukemia v. D compounds in, 1733–1734 PCa, 1695–1698 PCa v. 1,25(OH)2D3/taxane, 1745–1746 psoriasis, 1504, 1504t COMP. See Cartilage oligomeric matrix protein

1855 Competitive chemiluminescence immunoassay (CLIA), 25OHD quantitation using 125I-RIA v. automated, 939, 939f, 940f, 940t Competitive protein binding assays (CPBAs) for D/25OHD, 931–932 25OHD assay consistency v., 1023 Computerized tomography (CT), tumors causing TIO v., 988 Congestive heart failure (CHF), VDR polymorphisms v., 1148 Cortical thick ascending limbs (CTAL), CaR controlling Ca/Mg reabsorption in, 557, 557f Corticosteroids active intestinal Ca absorption v., 445 D metabolism influenced by, 1255t, 1265–1266 CPBAs. See Competitive protein binding assays Crohn’s disease bone disease in, 1302 Ca malabsorption in, 1297 D/25OHD/Ca malabsorption in, 1302 hypercalcemia/D hypersensitivity in, 1361 management of, 1302 VDR polymorphisms v., 1147 CT. See Calcitonin; Computerized tomography CT/CGRP gene alternative RNA processing of, 689, 689f cAMP-induced enhancer, 691 cAMP/neuroendocrine enhancers in 1,25(OH)2D3-inhibited transcription of, 693, 695f cAMP responsive enhancer, 690f mapping negative VDRE of, 694-696, 695f neuroendocrine-specific enhancer, 690f neuroendocrine-specific HLH enhancer, 691–692 MAP kinase stimulation/sumatriptan repression, 692, 692f 1,25(OH)2D3 transcription repression mechanism v., 696, 696f, 696t CT gene. See Calcitonin gene CTAL. See Cortical thick ascending limbs CTE. See C-terminal extension CTX. See Cerebrotendinous xanthomatosis Cubilin, 158–159, 158f Cyclin-dependent kinases (Cdks), cell cycle restriction points v., 1642, 1642f CYP24. See 24-Hydroxylase CYP27. See 25-Hydroxylase CYP1α mRNA v. 1α,25(OH)2D3, 75, 75t mRNA v. PTH, 76, 76t promoter activity v. 1α,25(OH)2D3, 76, 76t promoter activity v. PTH, 76–77, 77f CYP24A1. See 24-Hydroxylase CYP24A1 human, 91 rat, 91 cyp24A1, 24-hydroxylation v., 107–109 CYP27A1. See 25-Hydroxylase CYP27A1 intraacinar localization, 58–59 1,25(OH)2D3 administration v., 57–58, 58f, 59f regulation/enzyme activity, 56–58 transcription v. bile acids, 57 cyp27A1, hepatic 25-hydroxylation v., 105–106 CYP1α gene, extrarenal 1,25(OH)2D synthesis v., 1380 CYP24A1 promoter, rat, 93–97 mutational analysis v., 95–97, 96f, 97f CYP27B1. See 1α-Hydroxylase CYP27B1, structure, 74, 74f cyp27B1, 1α-hydroxylation v., 109–113 CYP2C11. See 25-Hydroxylase

1856

INDEX

CYP2D25. See 25-Hydroxylase CYP2R1. See 25-Hydroxylase Cytochrome P4501α, 1442 cloning/gene structure, 73–75, 74f Cytochromes P450 function of, 88 isoform modeling studies of, 1443 molecular modeling of, 88–89, 89f nomenclature for, 105, 106t in 25OHD3 1α-hydroxylation, 71–72, 71f ROS generation implicating, 762 Cytokines in keratinocyte growth/differentiation, 614 in macrophage 1α-hydroxylase regulation, 1384–1385 Cytoplasm, protein import retarded by docking in, 374 Cytotoxic T cells, 1,25(OH)2D3 autoimmunity v., 1755 Cytotoxic therapy, hyperphosphatemia in, 1178

D D. See Vitamin D D2. See Vitamin D2 D3. See Vitamin D3 Daivobet, in combination therapy of psoriasis, 1504, 1504t DBDs. See DNA binding domains DBP. See Vitamin D binding protein DBP-A, DBP-B structure compared with, 135–136 DBP-actin complexes, 122t, 123, 142–149 other actin complexes compared with, 147 structure of, 143–147, 143f β/γ-actin v., 145t, 146–147 hydrophobic DBP residue interaction v., 144t–145t, 146, 146f structures reported for, 149 tissue necrosis/cell disruption increasing, 126, 142 DBP-B, DBP-A structure compared with, 135–136 DBP-vitamin D complex, D binding site structure in, 137–142, 139f–140f biological implications of, 142 VDR D pocket structure v., 142 DCs. See Dendritic cells DCT. See Distal convoluted tubule Deficiency, 968–974 abnormal fetal organ development v., 852–853 adolescent, 816 age-dependent signs/symptoms of, 916t animal fertility reduced by, 854–855, 854t bisphosphonate antiresorptive effect v., 1114 in breast-fed children, 1067 Ca deprivation increasing susceptibility to, 1051–1052, 1052f cartilage v., 579 resorption of, 581 in celiac disease, 1301 classifying states of, 1024–1026 correction in osteoporosis treatment, 1110–1111 D absorption/input v. risk of, 784 development of acquired, 1296, 1296f development of depletion and, 1293–1296 diet and contemporary, 777–778 ECF [Ca2+] demand response mediated by, 776–777, 777f in elderly v. D synthesis/lifestyle, 823–824 epidemic in industrialized countries, 42 epidemiology of nutritional rickets/D, 1066–1067 HDM v. insulin resistance in, 1816 immune abnormalities associated with, 1389 in infancy/childhood, 811–818

Deficiency (Continued) insufficiency v., 1085–1086 intrinsic/extrinsic, 1029 in last trimester of pregnancy, 803–804 many diseases causing, 968 mild/moderate/severe stages of, 1024, 1025t mineralization v., 487–488 muscle myopathies in, 893 myopathy in, 1805 neonatal Ca metabolism v. maternal, 841 nutritional rickets in children v., 1065–1077 radiographic abnormalities of rickets v. treating, 970f, 971, 974f, 982f RAS v. blood pressure/cardiovascular function in, 875 risk groups, 1026 spiral of developing, 1294, 1295f ubiquity of, 796, 1024–1026, 1025f in vivo β cells v., 1764 7-Dehydrocholesterol (7-DHC) D3 biosynthesis from, 1405, 1408f decreased in elderly subjects, 823 keratinocytes producing D3 from, 609, 610f Dehydroepiandrosterone (DHEA) OCT production using, 1526 as steroid precursor in deltanoid synthesis, 1412, 1413f Delayed-type hypersensitivity (DTH), 1,25(OH)2D3 v. passive transfer of, 1754, 1755f Deltanoids c-Myc v. differentiation/G1 arrest induced by, 1649–1650 catabolism-inhibiting, 1406–1408 CDKIs up-regulated by, 1646–1649, 1647t–1648t cell cycle events modulated by, 1646–1650 cell cycle v. differentiation induced by, 1640–1650 cell type v. proliferation inhibition by, 1650 cellular models of differentiation induced by, 1635–1636, 1637t differentiation induced by, 1635–1640 generalized structure of, 1405, 1407f metabolic rationale guiding development of, 1405–1408 molecular biology rationale in development of, 1408–1411, 1410f multistep synthesis of hybrid, 1416–1418, 1416f, 1417f, 1418f, 1419f nearby structural changes v. catabolism of, 1407–1408, 1407f, 1410f neoplastic cell proliferative quiescence induced by, 1635 1,25(OH)2D3 hypercalcemia/calcification v., 1405 organic chemistry rationale in development of, 1411–1418 physiologically active metabolites of, 1408 potency of 22-ethyl, 1409, 1411f prodrugs/indications in, 1405, 1406f rational design of, 1405–1418 remote structural changes v. catabolism of, 1407, 1410f 6-s-cis-locked, 1412, 1412f SAR in design of 22-oxa, 1412 steroid precursors in constructing, 1412–1416, 1413f, 1414f, 1415f Dendritic cells (DCs) in acute allograft rejection, 1519 adaptive immune responses mediated by, 631 in generating effector/regulatory T cells, 634, 634f immunointervention targeting, 632–633, 632t 1,25(OH)2D3 inducing tolerogenic, 1519 VDR ligand immunoregulation of, 633–635 Dengue fever, VDR polymorphism v. hemorrhagic form of, 246 Dentin characteristics, 600–601 odontocytes elaborating, 599 in tooth root/periodontium, 601–602

INDEX

Dentin phosphoproteins (DPPs), high degree of phosphorylation in, 600–601 Dentin sialoprotein (DSP), in dentin, 601 Dento-alveolar bone complex, 599–605 formation/functions, 599–602 Depression, D3 preventing, 999t DEXA. See Dual energy x-ray absorptiometry Dexamethasone high dose intermittent 1,25(OH)2D3 with, 1744–1745 WT mouse intestinal Ca absorption v., 445, 447f 7-DHC. See 7-Dehydrocholesterol DHEA. See Dehydroepiandrosterone Diabetes in animal models v. 1,25(OH)2D3/analogs, 1768–1772 calbindin-D28K Ca buffering v. type I, 726 calbindin-D28K v. apoptotic cell death in, 727 D/immune system in type 1, 1767–1773 D signaling in regulating, 899 D v., 1763–1774 early intervention with 1,25(OH)2D3/analogs, 1768–1770, 1768t incidence v. sunlight exposure, 1766 insulin v. 1α-hydroxylation in, 1259 late 1,25(OH)2D3 intervention v., 1770 1,25(OH)2D3 analogs v., 1772–1773 1,25(OH)2D3/analogs with immune modulators v., 1771–1772, 1771t 1,25(OH)2D3 v. NOD mouse, 1404 risk v. D3, 999t risk v. VDR polymorphism, 1773 VDR ligand treatment v. type 1, 637–638, 637t VDR polymorphism associated with, 246 VDR polymorphism v., 1146–1147 Diabetic ketoacidosis, hypophosphatemia in, 1177 Diet age-related 25OHD decrease v., 1101 bone mass v. macrobiotic/vegetarian, 795 Ca deficient, 1074–1076 Ca in paleolithic/contemporary, 777–778 celiac disease v., 1300–1301 colon cancer v. high-fat, 866–867, 1572 colorectal cancer v. Ca from, 1621–1622 D hormonal system dependent on high Ca, 780–781, 781f D insufficiency v., 1087–1088, 1087f D3 intake from, 1026 D metabolism influenced by, 789–796 D metabolism v. vegetarian/omnivorous, 795 D negligible in Western, 1599 D status v. normal, 1066 D supplied by, 1294, 1294f, 1617 D toxicity from, 1356 ECF [Ca2+] demand response mediated by, 777–778 elderly v. D intake from, 824 groups/sects/ethnicities with vegetarian, 917t 25OHD half-life reduced by high fiber, 1076 osteoporosis v. Ca/D supplements to, 1112–1113, 1113t as PCa risk factor, 1681 PCa v. Ca/D from dairy products in, 1627–1629 PCa v. D from, 1606, 1625 rickets v. macrobiotic/vegetarian, 1067, 1077 VDR expression affected by, 204-205 vegetarian, 795, 917t, 1067, 1077 Differentiation cell cycle v., 1635–1651 cell cycle v. deltanoid-induced, 1640–1650 cellular models of 1,25(OH)2D3/deltanoids-induced, 1635–1636, 1637t

1857 Differentiation (Continued) 1,25(OH)2D3/deltanoids inducing, 1635–1640 1,25(OH)2D3 influencing tumor cell, 1580–1581 1,25(OH)2D3 signal propagation pathways v., 1636–1639 1,25(OH)2D3 synthesis/catabolism v., 1717 1,25(OH)2D3 v. PCa cell, 1691–1692 signals for 1,25(OH)2D3 influences on, 1636 TF role in 1,25(OH)2D3-induced differentiation, 1639–1640 DiGeorge sequence, hypocalcemia v., 1052–1053 Digital rectal exam (DRE), in PCa diagnosis, 1679 Dihydrotachysterol, metabolism, 1429–1430, 1430f 1,25-Dihydroxy–16-ene-23-yne D3 (BXL-353) BPH cell apoptosis/Bcl-2 expression v., 1836–1837, 1838f BPH cells v. antiproliferative influence of, 1836–1837, 1837f in vivo prostate growth v., 1837–1840, 1839f 10,19-dihydro–1α,25-dihydroxyvitamin D3, activity, 1544 19-Nor-1,25-dihydroxyvitamin D3, early 2-carbon analogs of, 1545 19-Nor-1α,25-dihydroxyvitamin D3 2-carbon-modified analogs of, 1543–1554 synthesis, 1544 23(S),25(R)1,25-Dihydroxyvitamin D3-26,23-lactone, in C-23 oxidative pathway discovery, 23 2α-Methyl-19-nor-(20S)-1,25-dihydroxyvitamin D3 (2αMD), 1545–1546 bone selectivity by, 1546–1547 structure, 1545f 1α,24R-Dihydroxyvitamin D3 [1α,24R(OH)2D3], 1425t–1426t, 1439 C24 oxidation steps in catabolism of, 1426f, 1439 metabolism, 1439 psoriasis treatment using OCT v., 1533 skin irritation in psoriasis treatment with, 1785 2β-Methyl-19-nor-(20S)-1,25-dihydroxyvitamin D3 (2BMD), 1545–1546 structure, 1545f 2-Methylene-19-nor-(20S)-1,25-dihydroxyvitamin D3 (2MD), 1545–1546 bone anabolic activity of, 1546–1547, 1547f bone selectivity by, 1546–1547 1,25(OH)2D3 v., 1546, 1546f ovariectomized rat bone mass v., 1547, 1548f structure, 1545f tissue selectivity/enhanced potency, 1548–1551 VDR binding to promoters stimulated by, 1549–1550, 1549f VDR conformation induced by, 1550, 1550f VDR interactions promoted by, 1550–1551, 1551f 1,25-Dihydroxyvitamin D [1,25(OH)2D], 1389–1390 age v., 1102–1103 as autocrine hormone in prostate, 1607–1610 Ca absorption v. age-related decrease in, 1104 Ca absorption v. age-related resistance to, 1104–1105, 1105f CKD altering VDR-mediated actions of, 1317–1322 CKD v. D bioactivation to, 1379–1383 clearance v. age, 1103 clinical evidence for dysregulated overproduction of, 1381 concentration having clinical relevance, 948t, 949 cutaneous production of, 609–613 CYP1α gene cloning v. extrarenal synthesis of, 1380 detecting, 942–947 competing approaches, 946–947, 946f, 947f issues/improvements v., 946–947 diseases v. hypercalcemia/hypercalciuria mediated by, 1390–1392, 1390t estimating human serum, 942, 943t Geminis as analogs of, 1511–1522 gene expression regulated by, 1212, 1213f in GHS rats, 1348, 1348f

1858 1,25-Dihydroxyvitamin D [1,25(OH)2D] (Continued) granulomatous disease inflammation sites accumulating, 1389 as HDM cause, 1815–1816 health v. changed concentration of, 1102 heightened immunoreactivity elevating, 1387f, 1389 hormonal regulation of, 610–611 hypercalcemia in lymphoma v., 1362 hypocalcemia due to hereditary resistance to, 1056 IH elevating, 1341–1342, 1342f inflammatory arthritis/RA inflammation sites accumulating, 1389–1390 inhibiting its own production, 20 in keratinocytes, 611 intracrine/autocrine action on monocytes/macrophages, 1387–1388, 1387f intravenous therapy v. renal failure, 1822–1823 levels v. Ca absorption in IH, 1342, 1343f macrophages lacking 24-hydroxylase activity directed by, 1383–1384 mechanism of action, 1210–1212 mechanisms of age-related changes in, 1102–1103 metabolism kinetics, 828 in muscle regeneration, 1816 25OHD hydroxylation producing, 1599 25OHD plasma half-life v. concentration of, 1104, 1104f 25OHD seasonality tracked by, 1019–1020, 1020f paracrine suppression of lymphocytes, 1387f, 1388–1389 PCa v., 1610–1611, 1611f PCa v. serum, 1602–1604 in PDDR, 1198–1199, 1199f plasma concentration, 28t production v. diseases, 1359–1362 psoriasis v. serum, 1784 resistance causing secondary hyperparathyroidism in renal failure, 1821–1823 resistance mechanisms in chronic renal failure, 1822, 1822f RIA methodology for detecting, 945–947, 945f assay calibrator preparation for, 945 sample/calibrator extraction/pretreatment in, 945 solid-phase extraction/silica purification chromatography in, 945 RRA methodology for detecting, 943–945, 944f calf thymus VDR preparation for, 943 RRA in, 944f sample extraction for, 943 solid-phase extraction/purification chromatography in, 943, 944f SCC keratinocytes producing, 612 serum concentration/PR/MCR v. aging, 828, 828f substrate-dependent synthesis of, 1019–1020, 1020f trophic factors influencing synthesis of, 828–831 in vitro correlates for dysregulated overproduction of, 1381–1387 1,25-Dihydroxyvitamin D2 [1,25(OH)2D2], isolation/identification, 19 1,25-Dihydroxyvitamin D3 [1,25(OH)2D3], 167–184, 412 A-ring in VDR transactivation, 1472f, 1473–1474 action mechanisms in autoimmunity, 1754–1756 actions in CNS, 1781–1783 actions in VSMCs, 902–903 as active form of vitamin D, 19, 782 in analog synthesis, 1490, 1490f analogs in oncology, 1746–1747 angiogenesis v., 1574–1576, 1693 in animal cancer models, 1573, 1574t animal model diabetes prevented by, 1768–1772 as antioxidant, 764t, 765–767 AUC for lowest anti-tumor dose in mice, 1744, 1744t

INDEX

1,25-Dihydroxyvitamin D3 [1,25(OH)2D3] (Continued) autoimmunity v., 1753–1759 animal models of, 1756, 1757t human, 1756–1759 avian parathyroid PTH/VDR gene expression regulated by, 541–542 β cells v., 1764–1767 basis for 25OHD interaction with, 783 biological effects in psoriasis, 1781, 1781f biphasic effects on osteoblasts, 650 blood pressure inversely associated with, 874, 874f bone apposition rates influenced by, 656–657 brain calbindins/Ca-binding proteins v., 1781 in brain detoxification, 1783 in brain homeostasis control, 1784, 1785f in breast cancer angiogenesis/invasion/metastasis, 1666–1667 breast cancer cell apoptosis induced by, 1665–1666, 1666f, 1672f breast cancer cell proliferation v., 1663–1665, 1672f breast cancer cells acted on by, 1663–1667, 1664t breast cancer sensitivity to, 1667–1668 breast cancer v., 1663–1672 C-/D-/E-ring modifications of, 1562–1565, 1563f C-3 epimerization, 24, 25f Ca/P demand regulating, 566 calbindin-D9K gene expression regulated by, 729 calbindin-D28K gene expression regulated by, 728 calcium channels regulated by, 1782–1783 in cancer combination therapy, 1582–1583 cancer growth/development v., 1573, 1574t, 1575t 20-carbon epimerization, 1544 cartilage differentiation requiring, 582 cartilage regulated by, 575–591 catabolism, 1406, 1409f, 1424–1427 C-26 hydroxylation/26,23-lactone formation, 1427, 1427f CDKIs up-regulated by, 1646–1649, 1647t–1648t cell cycle influenced by, 1577–1580, 1578f–1579f cell growth/differentiation v., 696 cell type v. proliferation inhibition by, 1650 cellular effects on leukemic cells, 1731, 1731t cellular models of differentiation induced by, 1635–1636, 1637t CKD v. catabolism of, 1316–1317 clinical use in psoriasis, 1784–1787 CNS activity v. in vitro investigations, 1781–1783 CNS tumors v., 1783–1784 colon cancer cells v. antiproliferative activity by, 1521–1522, 1522f, 1523f, 1523t colon cancer v., 1709–1721 epidemiology of, 1709–1710 colon cell proliferation/differentiation/apoptosis v. clinical studies of, 1715 in vitro, 1712–1714 in vivo animal models for, 1714–1715 colon tumors prevented by, 1709–1710 concentration v. immunosuppression/hypercalcemia, 1753 conformationally flexible seco-B-ring of, 1408f, 1412 CT/CGRP gene transcription repression mechanism of, 696, 696f, 696t CT levels regulated by, 689–691 CYP24A1 expression induced by, 93–97 cytotoxic agents combined with, 1745–1746 D endocrine system VSCCs influenced by, 751–757 D mediation v., 1040–1042 D-ring in VDR transactivation, 1475 in diabetes early intervention, 1768–1770, 1768t in diabetes late intervention, 1770 diabetes v., 1763–1774

INDEX

1,25-Dihydroxyvitamin D3 [1,25(OH)2D3] (Continued) diabetes v. immune modulators combined with, 1771–1772, 1771t differential cell response to, 651 differentiation induced by, 1635–1640 differentiation signal propagation pathways, 1636–1639 DNA motifs mediating stimulatory effects of, 705–706 dose escalation using QOD schedule, 1742–1743, 1743t early clinical trials of, 1742 epidermal keratinocytes/dermal inflammation v., 1781, 1782f 20-epimerization, 1494–1495, 1496t factors influencing concentration of, 517t fat/bone progenitor cell commitment v., 650f, 655–656 formation, 516–517 G1 block v. transient p21Cip1 up-regulation by, 1646–1648, 1646f GADD45 expression stimulated by, 1577, 1578f–1579f genomic actions’ molecular basis, 313–314 GFR v. serum, 1314, 1314f growth inhibition effect requiring 1α-hydroxylase, 744, 745f high dose intermittent, 1743f, 1743t, 1744–1745 high dose intermittent regimens of, 1742–1743 hormone/GF influence on osteoblasts v., 654–655 human PCa cell line proliferation v., 1683, 1683f hypercalcemic effect limiting use, 542 hyperparathyroidism v. intravenous, 1330–1331 IGF system interacting with, 1581–1582 immunosuppressive agents synergistically influencing, 1520 insulin secretion influenced by, 1764–1765 in intestinal Ca absorption, 411–424 intestinal calbindin increased by, 722–723 isolation/identification, 291 keratinocyte growth/VDR expression regulation linked in, 1781 kidney/parathyroid VDR regulated by, 522–523, 525t laboratory/clinical extrapolations of exposure to, 1744, 1744f, 1744t leukemic cell lines v., 1730–1731, 1730t ligand availability v. breast cancer sensitivity, 1666–1667 low-affinity neurotrophin receptor regulated by, 1781 2MD v., 1546, 1546f membrane-initiated Ca2+ responses to, 753–754 metabolically influencing β cells, 1764–1767 metastasis v., 1576 2-methylene/2α-methyl/2β-methyl derivatives of, 1545–1546, 1545f molecular mechanisms for leukemic cell influence by, 1731–1733 in MS pathophysiology, 1784 multiple membrane receptors for, 1462 as negative endocrine regulator of RAS, 875–878 as neuroactive hormone, 1779–1785, 1785f normal hematopoiesis influenced by, 1729–1730 NOS II v., 1783 novel target genes of, 1693–1695, 1694t oncogene expression regulated by, 1577–1579 oncology using 1,25(OH)2D3 analogs and, 1741–1747 in osteoblast differentiation/activity, 649–658 osteoblast differentiation pathway/status in vitro v., 649–653, 650f osteoblast heterogeneity not caused by, 652 osteoblast proliferation/differentation-associated genes v., 653–654 in osteoporosis treatment, 1111 parathyroid glands influenced by low, 1324–1325, 1325f parathyroid size correlating with resistance to, 1823 PCa cell apoptosis v., 1691 PCa cell growth arrested by, 1690–1691 PCa cell growth inhibition mechanisms, 1690–1695

1859 1,25-Dihydroxyvitamin D3 [1,25(OH)2D3] (Continued) PCa cell invasion/metastasis inhibited by, 1693 PCa cell resistance to antiproliferative effect of, 1684–1685 PCa growth inhibited in vitro by, 1683–1685 PCa treatment clinical trials, 1698–1699, 1698t PCa v. androgen interaction with, 1687–1688 phytoestrogens regulating synthesis of, 1720–1721, 1720t pleiotropism, 1403 in postmenopausal osteoporosis treatment, 1111–1112 potency/toxicity in PCa treatment, 1689–1690 pre-pro PTH mRNA v., 1527, 1528f primary PCa cell cultures v., 1683–1684, 1684f as prooxidant, 763–765, 764t prooxidant/antioxidant activities, 764t, 768 proposed hormonal mechanism, 167, 168f prostate cells synthesizing, 1607, 1608f prostate cells v. antiproliferative influence of, 1607–1608, 1609f psoriasis v. oral, 1785, 1785f psoriasis v. topical, 1784–1785 PTH gene expression regulated by, 539–543 PTH in chronic renal failure v. intravenous, 1329–1330, 1330f PTH-induced Ca2+ influx primed by, 755, 755f PTHrP production inhibited by, 741t, 743, 1576–1577 PTHrP production v. EB1089 analog of, 744 rapid action on growth-zone chondrocytes, 587–588, 587f reformulation for high dose intermittent regimens, 1746 renal Ca/P handling influenced by, 518–519 renin activity inversely associated with, 874, 874f resistance in breast cancer cells, 1667–1668 resistance mechanisms besides VDR mutation, 1226–1227 resistance v. uremia in CKD, 1320–1321, 1321f retinoids/thyroid hormones effects interdependent with, 655 ROS levels/redox-associated molecules v., 767, 767t Runx2/Cbfal TF modulated by, 566–567 19-nor/10,19-saturated derivatives of, 1544 secondary hyperparathyroidism in chronic renal failure v., 1327–1331, 1330f self-induced metabolism, 22 sex steroids synergistically stimulating synthesis of, 1260 side chain in VDR transactivation, 1474–1475 side effects v. medical applications, 142 signals of differentiation influences by, 1636 in skin, 1780 structural requirements for VDR transactivation by, 1472–1475, 1472f structure, 1472f, 1490f, 1558f supplementation safety, 1112–1113 targets in CNS, 1780–1781 TFs in differentiation induced by, 1639–1640 Th subset lymphokine production v., 1754, 1755f therapeutic use v. calcemic/phosphatemic activities, 1449 therapy in chronic renal failure, 1327–1332 tissue v. abnormal VDR function/low, 1322–1327 toxicity, 1357 tumor cell differentiation v., 1580–1581 tumor growth factors/receptors v., 1581–1582 tumor suppressor gene expression regulated by, 1579 VDR expression affected by, 201–204, 202t VDR mediated growth inhibition v. differentiation, 1713, 1713f virally transformed prostate cells v., 1684 in vitro D metabolism/catabolism v., 1717 in vitro osteoblasts influenced by, 649–656 VSCCs Ca2+/transcriptional responses to, 756–757 VSCCs v., 753–754 1α,25-Dihydroxyvitamin D3 [1α,25(OH)2D3], 381 biological actions mediated by VDRnuc, 384–385, 386t

1860 1α,25-Dihydroxyvitamin D3 [1α,25(OH)2D3] (Continued) biological response generation pathways, 383, 385f biosynthesis, 1408f bonds in 16-ene, 23-yne analog of, 1425t–1426t, 1436 bone matrix protein target genes, 711–717 conformational flexibility, 382f, 383, 388, 391, 399–400, 399f deltanoids emulating conformational flexibility of, 1411–1412, 1412f dual role in osteoblast differentiation, 713–714, 714f 20-epi-1α,25(OH)2D3 biological activity advantages over, 1433–1434, 1433f 20-epi compound analogs v., 285–287, 286f Geminis v., 287–288, 878 growth plate effects of 24R,25(OH)2D3 v., 582–583 kidney 1α-hydroxylase activity regulated by, 75–76, 75t, 76t muscle Ca homeostasis regulated by, 886–887 muscle cell proliferation/differentiation influenced by, 887–889 osteoblasts as source of, 717 production of extrarenal, 69–70 renal, 69, 70f rapid response v. 6-s-cis-shaped, 391, 398f rapid responses mediated by, 381–400 molecular tools for study of, 392, 393f schematic model of, 399–400, 399f structure-function evaluation of, 392–397 rational design of analogs for, 1405–1418 side chain modifications v. VDR transcription, 1474–1475, 1474t skeletal homeostasis v., 711–712 structure, 381–383, 382f VDR binding to, 313 24,25-Dihydroxyvitamin D3 [24,25(OH)2D3] bone mineralization stimulated by, 16 cartilage regulated by, 575–591 rapid action on resting-zone chondrocytes, 588–589, 588f 24,25-Dihydroxyvitamin D [24,25(OH)2D] concentration lacking clinical relevance, 948t, 949 detecting, 940–942, 941t preparative normal-phase HPLC in, 941 RIA for, 941–942, 942f sample extraction for, 941 solid-phase extraction chromatography in, 941 25,26-Dihydroxyvitamin D3 [25,26(OH)2D3], R/S isomer mixture in natural, 24 24R,25-Dihydroxyvitamin D3 [24R,25(OH)2D3] growth plate effects of 1α,25(OH)2D3 v., 582–583 in quail/chicken embryonic development/hatchability, 851 Disabled-2, megalin function in PCT requiring, 159 Diseases extrarenal 1α-hydroxylase activity in human, 1379–1394 extrarenal D metabolite overproduction v. human, 1390–1392 VDR gene association analysis v. phenotypes of, 1138 VDR polymorphism association analysis v. states of, 1137–1148 VDR polymorphisms v. hyperproliferative, 1145–1146 VDR polymorphisms v. immune-related, 1146–1148 VDR polymorphisms v. risk of, 1121–1149 Distal convoluted tubule (DCT), transcellular Ca/Mg reabsorption in, 557–558 DNA chromatin packaging genomic, 329, 329f VDR binding, 171 DNA binding domains (DBDs) dimerizing in presence of DNA, 314 DNA binding role v point mutations in, 233, 234f as homodimers bound to mouse osteopontin VDRE, 223t, 230, 231f

INDEX

DNA binding domains (Continued) hVDR nuclear translocation mediated by, 231f, 232–233 hVDR zinc finger, 229–233, 231f HVDRR-related mutations in VDR, 1218–1220 Arg50Gln, 1219, 1219f Arg80Gln, 1219, 1219f description of, 1218–1220, 1219f Gly46Asp, 1219–1220, 1219f His35Gln, 1219, 1219f Lys45Glu, 1219, 1219f structural analysis of, 1220 sequence similarity of hVDR, 229–230, 231f VDR, 1210–1211, 1211f DNA licensing, control of, 1644–1645 DPPs. See Dentin phosphoproteins DRE. See Digital rectal exam DRIP coactivator complex. See Mediator-D coactivator complex Drugs D analog metabolism v. design of, 1442–1443 D metabolism influenced by, 1253, 1255t, 1263–1274 deltanoids as, 1405, 1406f deltanoids as candidate, 1405, 1407f leukemia combination therapy with chemotherapeutic, 1734 mineral/skeletal homeostasis influenced by, 915t, 917 PCa combination therapy with chemotherapeutic, 1697–1698 DSP. See Dentin sialoprotein DTH. See Delayed-type hypersensitivity Dual energy x-ray absorptiometry (DEXA), BMD assessment with, 922–924 Duodenum Ca absorption by VDR-null mouse, 435 high Ca absorption rate in, 778 transcellular/paracellular Ca absorption in, 422–424, 423f Dutch dietary Ca intake of French v., 1096 hypovitaminosis D in institutionalized elderly, 1094

E EAE. See Experimental allergic encephalomyelitis Early neonatal hypocalcemia (ENH), premature infants affected by, 805–806 EB1089 antiproliferative activity v. IGF-II, 1714 bone mass loss in microgravity v., 1501–1502 breast cancer cell progression arrested by, 1663–1664, 1672f breast cancer cells/tumors v., 1669 cancer models v., 1499–1500 cell cycle regulated by, 1498 double bond system, 1425t–1426t, 1436 early clinical trials of, 1742 efficacy/safety of D analogs treating, 1504–1505 GADD45 expression stimulated by, 1577, 1578f–1579f hypercalcemia of malignancy v., 1450–1451 IGF system interacting with, 1581–1582 LNCaP xenograft growth inhibited by, 1688–1689, 1688f mechanism of action, 1478–1479 metastasis v., 1576 as 1,25(OH)2D3 analog in cancer/tumor treatment, 1746–1747 1,25(OH)2D3 v., 744 PCa v., 1690 pharmacokinetics/metabolism in clinical development of, 1503, 1503f, 1503t PTHrP expression/production inhibited by, 741t structure, 1479f, 1560, 1560f

INDEX

EB1089 (Continued) as superagonist, 1477–1478 VDR transcriptional activity v., 1474t in vitro metabolism, 1436–1437, 1437f in vivo colon cells influenced by, 1714 as weak calcemic analog, 1440, 1441t Wittig reaction introducing side chains to, 1490, 1491f, 1494t EBT. See Electron beam tomography ECaC. See Epithelial calcium channel ECaC1, functional properties, 431–432 ECaC2 expression in WT/VDR KO mice, 434f, 436t, 438 fast initial phase in Ca-dependent inactivation of, 432 functional properties, 431–432 tissue distribution, 431 ECD. See Extracellular domain ecologic fallacy, PCa risk v., 1602 ECs. See Endothelial cells ED-71 action mechanism, 1535–1537, 1537f convergent synthesis, 1534, 1534f development, 1525, 1534–1539 17β-estradiol bone metabolism influence v., 1536, 1536f linear synthesis, 1534, 1534f metabolism, 1439–1440 osteoporosis treatment using 1αOHD3 v., 1534–1535, 1535f ovariectomized rat bone mass v., 1537, 1537f as strong calcemic analog, 1440, 1441t structure, 1425t–1426t, 1439, 1525, 1526f EDAX. See Energy dispersive X-ray analysis EGF. See Epidermal growth factor EGF receptor (EGFR), 1710–1712 D compounds v. signaling through, 1499 expression in colon cancer, 1710–1711, 1711t EGFR. See EGF receptor EKG. See Electrocardiogram Electrocardiogram (EKG), changes in D toxicity v. myocardial ischemia, 1368 Electron beam tomography (EBT), coronary artery calcification v., 979 Electron microscopy, crystal size/shape shown by, 481 Enamel ameloblasts producing, 599, 601 apatite crystal size/impurities in, 480 D bioinactivation causing dysplasia/hypomaturation/hypomineralization in, 602 Endothelial cells (ECs), 1,25(OH)2D3 responses by, 903–904 Endothelin-1 (ET-1), parathyroid cell proliferation v., 545 Energy dispersive X-ray analysis (EDAX) growth plate ion concentrations monitored by, 487–488 mineral composition information from, 481 English, winter hypovitaminosis D in adult, 1091, 1091f ENH. See Early neonatal hypocalcemia Enterocyte Ca extrusion across basolateral membrane of, 419–421, 420f structure v. Ca flux, 413–414 vesicular CA transport in, 422 Enterohepatic circulation bile acids reclaimed through, 863–864 D/D metabolites in, 1295–1296 Enzymes D analog selectivity v. metabolic, 1460–1461 metabolic future study directions for, 112–113 mutant mouse models of, 105–114 microsomal, 51t–52t

1861 Enzymes (Continued) mitochondrial, 54–59 sex hormones regulating, 61 Epidermal differentiation, 1,25(OH)2D hormone likely to influence, 112 Epidermal growth factor (EGF) 20-epi D analogs interfering with, 1498–1499 keratinocyte proliferation stimulated by, 614 mitogenic activity inhibited by D compounds, 1665 PTHrP expression/production stimulated by, 741t VDR expression/cell differentiation v., 1713, 1713f in vitro D metabolism/catabolism v., 1717 Epidermis microanatomy of, 613, 613f 1,25(OH)2D3 antioxidant activities in, 764t regulatory Ca gradient in, 609, 613, 615 vitamin A influencing development of, 614 Epithelial calcium channel (ECaC), 527–528 architecture, 430, 431f in CA absorption, 415–416 Ca influx v., 430–432 molecular architecture, 431f, 432 nomenclature, 415n 1,25(OH)2D3 regulating expression of, 558 S100A10-annexin 2 required for functional expression of, 432 TRP superfamily/TRPV family, 430 VDREs in human, 527 Epithelial cells, mouse/human gastrointestinal ECaC2 expression restricted to, 431 ER. See Estrogen receptor EREs. See Estrogen response elements Ergocalciferol. See Vitamin D2 ERKOα hull mice, reduced duodenal ECaC2 mRNA expression in, 441f, 445 ERKOβ hull mice, duodenal gene expression unchanged in, 441f, 445 Estradiol, D metabolism influenced by, 1254t, 1259–1261 Estrogen. See also Estradiol Ca absorption v. deficiency of, 1105 deficiency increasing PTH with age, 1108–1109, 1108f intestinal Ca absorption v., 441f, 441t, 444–445, 444f 1,25(OH)2D v. deficiency of, 1103 PTHrP expression/production stimulated by, 741t serum PTH influenced by, 1024 VDR expression affected by, 202t, 206 Estrogen receptor (ER) localization in plasma membrane caveolae, 400 signaling down-regulated by 1,25(OH)2D3/EB1089, 1665 signaling v. breast cancer cell proliferation, 1663, 1672f Estrogen response elements (EREs) hVDR promoter, 200 optimal VDREs resembling half-sites of, 222 ET-1. See Endothelin-1 Ethanol D metabolism influenced by, 1255t, 1266 hypocalcemia/hypomagnesemia v., 1057 hypomagnesemia associated with abuse of, 1054 Ethiopians climate v. D in, 1026 low serum 25OHD in, 1026 sunlight exposure v. rickets in, 1066–1067 Eucalcemia Ca/D supplements for stable, 1059, 1059t, 1060t D in long-term maintenance of, 1050–1051 Europeans. See also specific nationalities D deficiency in, 1025–1026

1862

INDEX

Europeans (Continued) hypovitaminosis D in institutionalized elderly, 1093, 1093f PCa mortality in northern, 1600, 1600f PCa mortality/incidence in, 1625 Every-other-day (QOD) schedule, 1,25(OH)2D3 dose escalation permitted by, 1742–1743, 1743t Exercise bone density influenced by, 796 D metabolism influenced by, 789–796 Exosome, hnRNP role in, 359 Experimental allergic encephalomyelitis (EAE) D analogs treating, 1784 1,25(OH)2D3 down-regulating NOS II in, 1783 1,25(OH)2D3 v., 1755, 1784 VDR ligand treatment v., 637t, 638–639 Export receptors, nuclear export mediated by, 366–368, 367f Extracellular domain (ECD), CaRs’ large amino-terminal, 551, 552f Extracellular fluid (ECF) [Ca2+] demand response, 776–778 Ca2+ homeostasis in, 553f, 554–555, 555f Ca2+ obligatory loss from, 774–776 [Ca2+] regulation in, 774–778, 775f driving transfers of Ca2+ in, 774–776

F 26,27-F6-1α,25-(OH)2D3. See 26,26,26,27,27,27-Hexafluoro1α,25-(OH)2D3 Falls, D status in risk of, 1814, 1814f Familial hypocalciuric hypercalcemia (FHH), heterozygous inactivating CaR mutations causing, 552–553 Family history (FH), in approaching metabolic bone disease, 918 Famine, 777 Fanconi syndrome (FS) phosphate homeostasis in, 1174–1175 treatment, 1175 type I, 1174 type II, 1174–1175 Farnesoid X receptor (FXR), bile acid synthesis/flow regulated by, 865 Fat, progenitor cell commitment v. 1,25(OH)2D3, 650f, 655–656 Fatty acids, VDR expression affected by, 202t, 205 FCA. See Fractional calcium absorption Ferredoxin mitochondrial characteristics of, 72 hormones regulating, 72–73 in 25OHD3 1α-hydroxylation, 71f, 72–73 Fertility D in, 854–855 in mice with poor D function, 852 Fetus Ca homeostasis in mother and, 853–854 CaR in, 854 D synthesis v. development of, 852–853 low maternal D/Ca intake v., 841–843 maternal Ca economy v. bone mineralization by, 841, 841f FFAs. See Free fatty acids FGF-23. See Fibroblast growth factor-23 FGF-23 gene ADHR caused by, 1190 FLAG-tagged native/R176Q mutant, 1190, 1191f FH. See Family history FHH. See Familial hypocalciuric hypercalcemia

Fibroblast growth factor (FGF)-23 in ADHR, 468, 1164–1165 biological activity, 1165 in OHO pathogenesis, 468 as PHEX substrate, 1166–1167 as phosphatonin, 1164–1166 in XLH, 1193 Fibroblasts, in HVDRR studies, 1212–1217 Fibromyalgia, D3 mitigating, 999t Fibrous dysplasia, pathogenesis of hypophosphatemia in, 1192 FISH. See Fluorescent in situ hybridization Fish D3 in fatty, 1026 in diet v. PCa, 1601, 1606 Fluorescence resonance energy transfer (FRET) cytoplasmic VDR-RXR heterodimer shown by, 364–365, 365f GFP-VDR/RXR-BFP focal binding v., 374–375, 375f Fluorescent in situ hybridization (FISH), VDR gene mapping with, 1122, 1124f Fluoride, D metabolism influenced by, 1255t, 1273 FokI RFLP, 1130–1135, 1131t, 1133t, 1134t BMD v., 1142–1143 bone fracture risk v., 1143–1144 Japanese young adult height v., 1143 polymorphism functionality parameters associated with, 1135f, 1137 Foods Ca absorption enhancers in, 818 Ca/D fortification of, 817–818 rationale for, 817–818 D content of selected, 996t–997t D fortification of, 1065–1066 D toxicity v. natural/D-fortified, 1356 targeted/untargeted fortification of, 1073–1074 Formation period (FP) in osteoid indices, 1034 prolonging, 1034–1035 Fos family proteins, in osteoclastogenesis, 677–678 Founder effect PDDR, 1200 Tyr295stop mutation v., 1221 FP. See Formation period Fractal analysis, in bone structure assessment, 962 Fractional calcium absorption (FCA), response to 1,25(OH)2D v. aging, 832, 832f Free fatty acids (FFAs), DBP-A/B binding sites for, 136 French dietary Ca intake of Dutch v., 1096 hypovitaminosis D in adult urban, 1091, 1092t hypovitaminosis D in elderly, 1092–1093 hypovitaminosis D in institutionalized elderly, 1094 FRET. See Fluorescence resonance energy transfer FS. See Fanconi syndrome FXR. See Farnesoid X receptor

G G protein-couple receptor (GCPR) superfamily CaR as member of, 551–552 large ECDs in, 551–552 Gastrointestinal disease acquired bone disease in, 1297–1299 metabolic bone disease development in, 1298–1299 metabolic disturbances in, 1293–1297

INDEX

Gastrointestinal disease (Continued) non-Ca/D bone disease factors in, 1298 secondary hyperparathyroidism accompanying, 1297 type of acquired bone disease in, 1297–1298 Gc-globulin. See Vitamin D binding protein GDNF. See Glial cell-derived neurotrophic factor Gelsolin in actin-scavenger system, 143 growth nucleation v. DBP, 147 Geminis, 1511–1522 colon cancer treatment with, 1521–1522 20-epimeric-24R-hydroxy/19-nor, 1513, 1513f 23-yne-26,27-hexafluoro, 1514, 1514f, 1516f, 1517f 24R-hydroxy metabolite of, 1513–1514 configuration of, 1513, 1514f hypertension v., 1518 1α,25(OH)2D3 v., 287–288, 878 renin expression suppressed by 19-nor, 1518, 1518t as renin inhibitors, 1514–1518 synthesis of, 1511–1513, 1512f vascularized heart allograft rejection v. 19-nor, 1520–1521, 1520t, 1521f Gene expression 20-epi D analogs v., 1495–1498 VDR-mediated control mechanisms influencing, 235–243 Gene inactivation, tissue-specific, Cre/lox strategy for, 112, 113f Genetic hypercalciuric stone-forming (GHS) rats bone resorption in, 1349 establishment of colony of, 1347 as IH animal model, 1346–1350 intestinal Ca transport in, 1348, 1348t low Ca diet response by, 1349, 1350f mineral balance in, 1348 pathogenesis in, 1350, 1350f serum 1,25(OH)2D in, 1348, 1348f serum/urine chemistries in, 1347–1348 VDR in, 1348–1349, 1349f Geography adult D insufficiency v., 1091–1094, 1091f, 1092f D metabolism influenced by, 790–795 D nutrition/acquired bone disease v., 1297–1298 diabetes incidence v., 1766 hip fracture risk v., 1813 hypertension/stroke v., 873 low D intake v., 1087–1088, 1087f as PCa risk factor, 1601, 1601t rickets/colon cancer similarity in, 866 serum 25OHD v. Argentine, 795 Gestation, VDR WT/KO mouse intestinal Ca absorption during, 440–442, 441f, 441t, 442f GFR. See Glomerular filtration rate GFs. See Growth factors GH. See Growth hormone GHS rats. See Genetic hypercalciuric stone-forming rats GIO. See Glucocorticoid-induced osteoporosis Glass, vitamin D photosynthesis v., 41, 42f Glial cell-derived neurotrophic factor (GDNF) in C cell migration, 687–688 RET/GDNFR-α signaling system, 688, 688f Glomerular filtration rate (GFR), 1,25(OH)2D3 v., 1380, 1380f Glucocorticoid-induced osteoporosis (GIO) D as treatment for, 1243–1248 D metabolism v., 1239–1248 D preparations’ efficacy v., 1247–1248 studies of serum D metabolites in, 1241, 1242t

1863 Glucocorticoid receptor (GR), 1239–1240 actions in bone, 1240 Glucocorticoid response elements (GREs), hVDR promoter, 200 Glucocorticoids Ca absorption v., 1244–1245 CYP27A1 regulated by, 57 D metabolism influenced by, 1241–1243 genomic/nongenomic effects of, 1239 GIO v. effects mediated by, 1239 mineral/skeletal metabolic indices v., 1243, 1243f non-GR TFs mediating, 1239–1240 1,25(OH)2D3 influenced by, 1741 osteoblasts/osteoblast progenitors influenced by, 1240 in PCa combination therapy, 1696 PKC activity/CYP24A1 expression v., 99 proximal tubular Pi transport v., 461t PTHrP expression/production inhibited by, 741t renal Pi excretion decreased by, 516t type I collagen synthesis v., 704 VDR expression affected by, 202t, 205–206 Glucose, serum phosphate depressed by insulin and, 1177 Gluten, celiac disease v., 1300–1301 Goats, Ca balance v. sunlight in lactating, 4 Granulomatous diseases extrarenal D metabolite overproduction v., 1390–1391, 1390t 1,25(OH)2D accumulation in, 1389 1,25(OH)2D production v., 1359–1361 Grave’s disease D deficiency v. subtotal thyroidectomy for, 1262 VDR polymorphisms v., 1148 GREs. See Glucocorticoid response elements Growth, VDR ablation v., 341–345, 342f Growth factor receptors, 1,25(OH)2D3 influencing tumor, 1581–1582 Growth factors (GFs) 20-epi D analogs interfering with, 1498–1499 human BPH cells v., 1834 1,25(OH)2D3 influencing tumor, 1581–1582 1,25(OH)2D3 v. signaling by, 1665 PCa v. actions of, 1692–1693 PTHrP expression/production stimulated by, 741t PTHrP inhibition v. signaling interference in, 743 VDR expression affected by, 207–208 Growth hormone (GH) acromegalic/normal 1,25(OH)2D3 v., 1257–1258 D metabolism influenced by, 1254t, 1257 D metabolism influenced by IGF-I and, 1254t, 1258 renal Pi excretion decreased by, 516t VDR expression affected by, 208 Growth plates, 108 calbindin-D9K/D28K in, 724, 730t chondrocyte differentiation in, 576, 576f chondrocytes changing extracellular matrix composition of, 576f, 577–578 cyp24A1-null mouse, 108–109, 108f D actions influencing, 581–582, 581t dietary Ca increasing thickness of, 112 formation of, 575 in growing skeleton, 1031 horizontal/vertical organization of, 498, 499f 24,25(OH)2D3 targeting cartilage in, 582–583 24R,25(OH)2D3 v. 1α,25(OH)2D3 in, 582–583 rickets v. disorganization of, 1071, 1071f subperiosteal erosions near, 976f, 977 VDR ablation affecting, 343, 343f GS 1500, VDR DNA-binding complex induced by, 1496, 1497t

1864

INDEX

GS 1790, bone loss/strength v., 1497t, 1501, 1502t Gy mice, XLH modeled by, 463, 463t

H Habitus bone histomorphometry v., 795–796 bone mass v., 795 D metabolism influenced by, 789–796, 795 Hair cycle D endocrine system in, 291 regulation of, 621–622 factors implicated in, 621 VDR’s different role in, 622 VDR mediating, 234–235 VDR-RXR triggering mechanism unknown, 249 VDR triggering telogen/anagen transition in, 247f, 248–249 Hair follicles development of mouse, 347 1α,25(OH)2D3 VDRnuc in, 385t VDR ablation effect mouse on, 346–348, 346f VDR in function of, 224–225 VDR’s clinical relevance to, 246–249 Hairless gene (hr), as NR co-repressor, 299 Haplotypes polymorphic variation/disease risk v., 1126–1127 polymorphic variation encapsulated by, 1125f, 1127, 1128f VDR gene v. gene-wide, 1135–1137, 1136f Harrison’s sulcus, 1067–1068, 1068f HAT coactivator complexes. See Histone acetyltransferase coactivator complex HD. See Hemodialysis; Huntington’s disease HDM. See Hypovitaminosis D myopathy Health claiming D influence on, 999–1000 D dose/response curve for cardiovascular, 905, 905f D3 nutrition effects anticipated on, 1009t D signaling in cardiovascular, 899–901 D synthesis/absorption v., 1294 FGF-23 in, 1193–1194 serum 25OHD directly related to, 825 sunlight/dietary vitamin D affecting, 42 VDR-mediated signaling implications for, 243–249 Heart D supplementation benefiting, 874 19-nor Gemini v. allograft rejection in, 1520–1521, 1520t, 1521f 1,25(OH)2D3/analog immunoregulatory properties in, 1519 Heat shock-70 proteins (hsp70s), in macrophage 1α-hydroxylase regulation, 1386–1387 Helix-loop-helix (HLH) proteins, in CT gene transcription, 691 Helper T cells interaction/regulation by Th subsets, 1753–1754 1,25(OH)2D3 in autoimmunity v., 1754–1755 1,25(OH)2D3 v. lymphokine production by Th, 1754, 1755f Hematological malignancy, D v., 1727–1736 Hematopoiesis, 1727–1728, 1728f bone remodeling resembling, 505–506 D compounds influencing, 1729–1730 Hemochromatosis, hypoparathyroidism/hypocalcemia v., 1053 Hemodialysis (HD), long-term OCT dose in, 1530–1531, 1531f HEP 187 bone loss/strength v., 1494t, 1501 bone mineral metabolism influenced by, 1493, 1494t Heparin, D metabolism influenced by, 1255t, 1269–1270 Hepatic osteodystrophy, 1303–1304

Hereditary 1,25-dihydroxyvitamin D-resistant rickets. See Hereditary vitamin D-resistant rickets Hereditary hypophosphatemic rickets with hypercalciuria (HHRH), 469 genetic defect underlying, 1173–1174 pathophysiology, 1173 phosphate homeostasis in, 1169t, 1173–1174 treatment, 1174 XLH/ADHR/OHO v., 469 Hereditary vitamin D-resistant rickets (HVDRR), 224, 1207–1231 alopecia associated with, 1229–1230 arginine altered to leucine in, 233, 234f bisphenol analogs v., 1561–1562, 1561f Ca absorption efficiency in, 782 Ca/P/lactose v., 111–112, 111f cases described, 1212–1213, 1214t–1216t cell lines suppressing RXR-VDR transactivation/VDRE binding, 354f, 355–356 cellular basis of, 1212–1218 clinical/biochemical findings in, 1208–1209 clinical features, 1208–1210 cultured fibroblasts in initial studies of, 1212–1217 D-resistant human with signs of, 355 D v. Ca/phosphate absorption in, 566 defective mineralization/rickets in, 1208, 1208f environmental causes v. disease resembling, 1226–1227 first recognition of, 1212 first report of, 1208 functional domains, 1197, 1198f glutamate 420 altered to lysine in, 233, 234f ligand contact point disruption as basis for, 1224 molecular basis for, 1218–1227 non-VDR protein mutations causing, 1226–1227 25OHD 1α-hydroxylase enzyme activity v. PDDR, 1199f 1,25(OH)2D/1,25(OH)2D3-regulated processes in, 1210 pathophysiology of, 1209 prenatal diagnosis of, 1228 R391C mutant hVDR-RXR heterodimerization impaired in, 233, 234f recognition/semantics of, 1197 serum biochemistry levels in, 1208, 1209t severity correlating with alopecia, 1210 spontaneous healing of Tyr295stop/Arg73Gln, 1228 terminology, 1207 therapy for, 1227–1229 Ca in, 1227–1228 D analogs in future, 1229 D in, 1227 VDR/1α(OH)ase ablation causing symptoms of, 429 VDR arginine/cysteine mutation in, 233, 234f VDR gene defects causing, 224, 429, 1207 Herodotus, sunlight effects v. bone recognized by, 565 Heterogeneous nuclear ribonucleoproteins (hnRNPs), 357–359 classical view of, 357 HVDRR v. abnormal expression of, 1226 as multifunctional proteins, 357–359 26,26,26,27,27,27-Hexafluoro–1α,25-(OH)2D3 [26,27-F61α,25-(OH)2D3], potency/metabolism, 1440 HHM. See Humoral hypercalcemia of malignancy HHRH. See Hereditary hypophosphatemic rickets with hypercalciuria High-performance liquid chromatography (HPLC) D/25OHD analyzed with, 932 D v. quantitative reversed-phase, 934, 934f deficiency/depletion evaluated by, 1086

INDEX

High-performance liquid chromatography (Continued) 24R-monohydroxylated Gemini metabolite configuration determined by, 1513–1514, 1514f, 1515f 25OHD assay consistency v., 1023 Hispanics bone mass in, 794 D metabolism in, 793–794 metabolic bone disease v. lactose intolerance in, 917t Histone acetyltransferase (HAT) coactivator complex enzymatic activity in SRC/p160 family, 264–265 VDR linkage to, 176, 264–265 in VDR transcription model, 268–270, 269f History of present illness (HPI) in approaching metabolic bone disease, 916 nutritional factors in, 916, 917t HLH proteins. See Helix-loop-helix proteins hnRNP A (hnRNPA) family direct repeat half-site specificity in, 357 New World primate VDRE-BPs in, 354 REBiP similar to proteins in, 355 hnRNPs. See Heterogeneous nuclear ribonucleoproteins Hodgkin’s disease, extrarenal D metabolite overproduction in, 1390t, 1391–1392 Homer, rickets described by, 1065 Hopkins QW-1624F2-2, 1410f 24F2-1,25(OH)2D3 v., 1407–1408, 1410f large-scale synthesis/availability of, 1408 Hormone resistance, concept evolution, 1207 Hormones. See also specific hormones D metabolism influenced by, 1253–1263, 1254t proximal tubular Pi transport regulated by, 461–462, 461t HPI. See History of present illness HPLC. See High-performance liquid chromatography HPLCs. See Human periodontal ligament cells hr. See Hairless gene HSA. See Human serum albumin hsp70s. See Heat shock-70 proteins Human periodontal ligament cells (HPLCs), osteoblast-like differentiation in, 602 Human physiology, 773–905 Human serum albumin (HSA), DBP structural similarity to, 135–136, 136–137, 136f, 137f, 138f Human VDR (hVDR) ablated in rodents/humans, 224–225 HVDRR v. point mutations in, 233–235, 234f putative NLSs in, 368–369, 368f structure of superagonist ligands complexed to, 285–287, 286f Humans D requirements from evolution of, 995 D resistance in, 355–356 D supply in genomic selection of, 1006 intestinal Ca absorption in, 429–437 Humoral hypercalcemia of malignancy (HHM), Ca reabsorption in, 1256 Huntington’s disease (HD), 1779 hVDR∆ biological properties, 280–281 LBD topology, 281, 283f ligand-binding pocket, 281–282, 283f radii of gyration, 281, 282f solution studies, 281 hVDR∆−1α,25(OH)2D3 complex, crystal structure, 281–282 HVDRR. See Hereditary vitamin D-resistant rickets HVO. See Hypovitaminosis D osteopathy Hydroxyapatite apatite v. natural, 477, 478–480, 478f

1865 Hydroxyapatite (Continued) chondrocyte matrix supporting, 577 crystal formation in matrix vesicles, 578 crystal growth in dentin, 600 crystal growth in enamel, 601 lattice structure in bone mineral, 1030 2-Methylene-19-nor-(20S)-1α-hydroxybishomopregnacalciferol (2MbisP), 1551–1552, 1552f tissue selectivity, 1552–1553, 1553f 25-Hydroxycholecalciferol, in HVO evolution, 1038t, 1041–1042 2-Methylene-19-nor-1α-hydroxyhomopregnacalciferol (2MP), 1551–1552, 1552f tissue selectivity, 1552–1553, 1553f 25-Hydroxylase, 47–62 activity v. sex, 61 D molecule interaction model, 56 human/mouse, 54 identifying novel types of, 106–107 microsomal, 50t–51t expression sites for, 48, 49t porcine, 53–54, 54f sex differences in, 52–53 mitochondrial, 51t–52t expression sites for, 48, 49t ontogeny, 60 specificity, 55–56 vitamin D activated by, 17–19 26-Hydroxylase, 24 24-Hydroxylase (CYP24), 20–21, 85–100 activity in kidney, 517–518 catabolic pathway/Ca restriction in D toxicity, 1365–1367, 1366t cellular expression, 86–87 in CYP24A1 KO/transgenic animals, 90–91 cellular metabolism by, 1459–1461, 1460f distribution in kidney, 528 enzyme functional in PDDR patients, 1199 kinetic analysis, 89–90 metabolic analysis, 89 spectral analysis, 90 structure/function, 88–90 enzyme pathways, 87–88, 88f enzyme structure/function, 88–90 expression regulators, 93, 94f function/regulated expression, 85–86, 86f human intestine expressing, 1717–1719 in 23-hydroxylation, 23 inhibitors in PCa combination therapy, 1695–1696 keratinocyte differentiation v. activity of, 611–612, 611f macrophages lacking 1,25(OH)2D-directed activity of, 1383–1384 molecular aspects, 93–100 mouse proximal/distal colon expressing, 1719 mRNA expression evaluation for, 1718 nutritional regulation, 1719–1721 1,25(OH)2D actions terminated by, 22 25OHD3/1,25(OH)2D3-, 528 1,25(OH)2D3 regulating Ca deficiency v., 21 intestinal., 21 PCa growth-inhibitory responses influenced by, 1685–1686, 1686f physiological role, 21–22 properties, 87 responsiveness v. VDREs, 316t, 320 terminal reaction, 87, 87f ubiquity, 20–21

1866 1α-Hydroxylase (CYP27B1), 69–78 age v. activity of renal, 1103, 1109 in breast cancer, 1667 CKD directly inhibiting, 1316 CKD v. PTH/Ca induction of, 1315–1316 cytokines stimulating macrophage, 1384–1385 in D toxicity v. restricted Ca diet, 1365, 1365t, 1366t deficiency v. HVDRR, 1208, 1209t enzyme v. 24-hydroxylase enzyme, 1199 factors influencing activity of, 517t functional diversity, 1392 hsp70s regulating macrophage, 1386–1387 human disease/extrarenal activity of, 1379–1394 human intestine expressing, 1717–1719 immune cell regulators v. macrophage, 1384–1387 importance of phosphorus in regulating, 1161 keratinocyte differentiation v. activity of, 611–612, 611f levels in PCas/noncancerous prostates, 1608–1609 LPS amplifying macrophage, 1385–1386, 1385f in macrophages, 1381–1382, 1382f mouse proximal/distal colon expressing, 1719, 1719f mRNA expression evaluation for, 1718 new developments in disease v., 1379 NO regulating macrophage, 1386 nutritional regulation, 1719–1721 1,25(OH)2D3 growth inhibition requiring, 744, 745f PCa growth-inhibitory responses influenced by, 1686–1687, 1687f PCa v., 1626–1627 in prostate, 1607 PTH/Ca not regulating prostatic, 1609–1610 renal phosphate transport/phosphorus v., 1161, 1163f species distribution, 70–71 substrate availability/1,25(OH)2D bioactivation in CKD, 1315 TLR expression/signaling v. extrarenal, 1385 in vitamin D metabolism, 19–20 23-Hydroxylation, vitamin D3, 23 2-Methylene-19-nor-1α-hydroxypregnacalciferol (2Mpregna), 1551–1552, 1552f tissue selectivity, 1552–1553, 1553f 1,25-Hydroxyvitamin D3, urinary loss in megalin KO mice, 154, 155t 23(S),25(R)25-Hydroxyvitamin D3-26,23-lactone, in C-23 oxidative pathway discovery, 23 1α-Hydroxyvitamin D2 [1αOHD2] as 1,25(OH)2D3 analog in cancer/tumor treatment, 1747 toxicity, 1358 1α-Hydroxyvitamin D3 [1αOHD3] early clinical trials of, 1742 in GIO treatment, 1247–1248 leukemic cells v., 1734 osteoporosis treatment using ED-71 v., 1534–1535, 1535f PDDR treatment with, 1201t, 1202 toxicity, 1357–1358 25-Hydroxyvitamin D3/DBP complexes cubilin as endocytic receptor for, 158–159, 158f glomerular filtration of, 153–154, 154f megalin as endocytic receptor for, 154–156, 157–158, 157f, 158f megalin influencing cellular uptake of, 156–157, 156f receptors/co-receptors for, 157–159 two-receptor model for, 159, 159f 25-Hydroxyvitamin D3 [25OHD3] basis for 1,25(OH)2D3 interaction with, 783 biological static/dynamic methods v. normal serum, 1020t BMD v. normal, 1022, 1022f Bone fracture risk defining normal, 1022–1023

INDEX

25-Hydroxyvitamin D3 [25OHD3] (Continued) Ca intake v. serum, 1023–1024 colon cancer v., 1709–1710 CYP27A1 disruption v., 60 D3/metabolite plasma concentrations v., 1362, 1363t desirable target plasma concentration of, 1009t dose-response relationship with D3, 1003, 1005f early 2-carbon analogs of, 1545 endocytic pathways, 153–160, 160f as hormonally active form of vitamin D3, 16 interlaboratory assay variability v., 1023, 1023f, 1023t, 1025 intestinal Ca absorption efficiency correlated with, 783 metabolism in human colon cancer cells, 1716–1717, 1716f neoplastic colonocytes acted on by, 1710–1715 nutritional status improvement v., 1003–1005, 1004t PDDR treatment with, 1201–1203, 1201t, 1202f, 1203f proteins in 1α-hydroxylation of, 71–73 renal endocytosis, 153–159 renal handling of Ca/P influenced by, 518–519 secondary hyperparathyroidism in defining normal, 1020–1022 serum concentration indicating nutritional status, 1003, 1004t serum concentration of D status clinical outcomes v. normal, 1019 defining normal, 1019–1026, 1020t sampling biases v. reference value, 1019 strategies increasing concentration of, 1003, 1004t urinary loss in megalin KO mice, 154, 155t variables influencing normal values of, 1023–1024 25-Hydroxyvitamin D [25OHD]. See also 25-Hydroxyvitamin D3 age v., 1101–1102, 1102f age v. synthesis of, 826, 1617 automated instrumentation CLIA methodology for detecting, 939–940, 939f, 940f, 940t bone fracture risk v. serum, 1370, 1370f Ca absorption role of, 1106 clearance v. age, 1102 clinical implications of keratinocytes producing, 612–613 concentration having clinical relevance, 948t, 949 cytokines regulating, 610f, 612 deficiency caused by defective CYP27A1, 61–62 deficiency v. age, 1103, 1109 detecting, 935–939 endogenous production in hypervitaminosis D, 1359 as HDM cause, 1816 HPLC methodology for detecting, 936 quantitative normal-phase HPLC in, 936, 937f sample extraction for, 936 silica cartridge chromatography in, 933f, 936 solid-phase extraction chromatography in, 936 hydroxylation, 1599 keratinocyte differentiation regulating, 611–612 kidney in metabolism of, 516–518 liver production, 17–19 mechanisms of age-related decrease in, 1101–1102 metabolism/identification of products, 609–610 muscle power/MVC v., 1812–1813, 1812f 1,25(OH)2D concentration v. plasma half-life of, 1296, 1296f parent D/25OHD toxicity due to, 1362 PCa v., 1610–1611, 1611f PCa v. serum, 1602–1604 plasma concentration, 28t poor assay performance v., 939–940, 940t prostate cells v. antiproliferative influence of, 1607–1608, 1609f psoriasis v. serum, 1796 RIA methodology for detecting, 936–938 assay calibrator preparation for, 936

INDEX

25-Hydroxyvitamin D [25OHD] (Continued) other D compounds v., 937–938, 938f, 938t, 939t radioimmunoassay in, 937, 938f sample/calibrator extraction for, 936 seasonal variation, 1019–1020, 1020f, 1069 serum concentration influenced by aging, 826–827 Hyp mice abnormal 24-hydroxylase regulation in, 465 XLH modeled by, 463, 463t, 464f, 465 Hypercalcemia. See also Humoral hypercalcemia of malignancy calcipotriol/related analogs v., 1493–1494 causes of, 1355, 1356t clinical manifestations of D toxicity from, 1368 D-induced, 568–569, 986, 988f D metabolite use limited by, 1403 D single dose rickets therapy v., 1072, 1073 D toxicity causing, 1355–1372 dehydration worsening pathophysiology of, 1368 early detection/prevention of, 1393 EB1089 v., 744 20-epi 1,25(OH)2D3 analogs v., 1494–1495, 1496t excessively fortified milk causing, 1065 identifying patients at risk for, 1393 lowest D3 dose causing, 1005f, 1008–1009 nonsecosteroid analogs v., 1565–1567 in Npt2a KO mice, 462 1,25(OH)2D3 allograft rejection therapy v., 1518 1,25(OH)2D3/analogs influencing, 542–543 24F2-1,25(OH)2D3 and Hopkins QW-1624F2-2 v., 1407–1408, 1410f 1,25(OH)2D3 causing, 744 preventing, 1393 SCCs associated with, 612 screening patients at risk for, 1393 stanniocalcin v., 462 therapeutic 1,25(OH)2D3 inducing, 1449 treating, 1393–1394 Hypercalciuria. See also Idiopathic hypercalciuria D metabolite use limited by, 1403 early detection/prevention of, 1393 as early sign of D toxicity, 1368 identifying patients at risk for, 1393 IH diagnosis v. causes of normocalcemic, 1340, 1340t in Npt2a KO mice, 462 1,25(OH)2D3 causing, 744 screening patients at risk for, 1393 treating, 1393–1394 Hyperesthesia, hypovitaminosis D associated with, 1090 Hyperkalemia, hypocalcemia v., 1049 Hypermagnesemia CaR mutation causing, 553–554 hypocalcemia masked by, 1049 Hyperparathyroidism. See also Secondary hyperparathyroidism D insufficiency in elderly v. mild, 825 D supplementation v., 1094 high-Ca diet preventing KO mouse, 429 intracortical bone resorption in, 975f, 977 intravenous 1,25(OH)2D3 v., 1330–1331 management of severe/refractory, 1823–1826 in megalin KO mice, 154–156, 155t models, 544 in renal osteodystrophy, 976–979 resorbed distal phalangeal tufts v. treated, 975f, 977 subperiosteal erosions in, 975f, 976–977 VDR ablation v., 342, 343f in VDR KO mice, 433, 488 VDR polymorphisms v., 1146

1867 Hyperphosphatemia in altered phosphate load disorders, 1179–1180 dystrophic calcification associated with, 478 1α-hydroxylase inhibited by, 1316 hypocalcemia induced by, 1056–1057 1,25(OH)2D inhibited by, 1161 Pi vs., 454–455 therapeutic 1,25(OH)2D3 inducing, 1449 Hyperproliferative skin disorders, VDR-RXR heterodimeractivating ligands v., 241 Hypertension D analogs v., 878 D signaling in regulating, 899 Gemini compounds v., 1518 inappropriate RAS activation leading to, 871 low calcemic D analogs v., 1514–1516 Hypervitaminosis D2, hypervitaminosis D3 v., 1364–1365 Hypervitaminosis D3 cartilage/bone v., 582 23-hydroxylation in, 23 hypervitaminosis D2 v., 1364–1365 Hypocalcemia. See also Early neonatal hypocalcemia; Late neonatal hypocalcemia biochemical changes induced by, 1051–1052 breast milk/formula v. neonatal, 841–842, 842f CYP27B1 gene mutation v., 109 in CYP27B1-null mice, 703 dermopathy in chronic, 920, 920f differential diagnosis/D therapy v., 1049–1060 differential diagnosis of, 1052–1058 early/late neonatal, 1055 fasting/famine/drought causing, 777 of IDM, 807 long-term treatment of, 1058–1060 management of acute, 1058 in megalin KO mice, 154–156, 155t in 1α(OH)ase-null mice, 438, 489 1,25(OH)2D3 production increased by, 19–20 1,25(OH)2D3 v. HVDRR patient’s, 357 parathyroid cell proliferation stimulated by, 545 physiology of, 1049–1050 PTH gene expression increased by, 544 PTH increasing 1,25(OH)2D3 production in, 703 rickets/osteomalacia v. chronic, 1049 signs/symptoms, 919–920, 919t skeletal homeostasis/D metabolism/illness causing, 1054–1058 therapy for, 1058–1060 uncalcified osteoid v. 1,25(OH)2D3 treatment of, 1041, 1041f in VDR KO mice, 433, 488 Hypokalemia, hypocalcemia masked by, 1049 Hypolipidemics, D metabolism influenced by, 1255t, 1267–1268 Hypomagnesemia in alcoholics, 1177 CaR mutation causing, 553–554 hypocalcemia v., 1049, 1053–1054, 1054 Hypoparathyroidism hypocalcemia v. idiopathic, 1053 in Npt2a KO mice, 462 Hypophosphatemia in altered phosphate load disorders, 1178–1179 bone mineralization abnormalities in, 488 causes of chronic, 924, 924t correction v. therapeutic objectives, 928, 928f in CYP27B1-null mice, 703 dental defects associated with familial, 602 GH regulating 1,25(OH)2D3 in, 1258

1868

INDEX

Hypophosphatemia (Continued) in Npt2a KO mice, 462 in 1α(OH)ase-null mice, 438 1,25(OH)2D increased by, 1161 pathogenesis in fibrous dysplasia, 1192 Pi vs., 454 radiology of renal tubular defect, 981–984 in stage II rickets, 1070 TIO/XLH/ADHR characterized by, 1162–1163 transcellular shift in, 1176–1177 Hypophosphatemic bone disease, FGF-23 gene mutation detection in, 1190 Hypophosphatemic diseases, 1162–1176 common pathway of pathogenesis in, 1163–1168 FGF-23 activity v. phenotypic abnormalities of, 1166 Hypovitaminosis D, urinary loss of 25OHD3 causing, 153 Hypovitaminosis D myopathy (HDM), 1805–1817 clinical studies of, 1811–1813 biopsies in, 1811 muscle function in, 1811–1813 D deficiency insulin resistance v., 1816 misdiagnosis, 1805–1806 muscle physiology v., 1806–1808, 1806f, 1807f 25OHD/1,25(OH)2D/PTH causing, 1814–1816 sunlight v., 1813–1814 symptoms/signs, 1805–1806 Hypovitaminosis D osteopathy (HVO), 1035 biochemical evolution of, 1036–1039, 1038t D metabolism in, 1039–1040 pathogenesis of, 1029–1044 stages of, 1036, 1036t

I IBD. See Inflammatory bowel disease Ichthyosis, D therapy v., 1787 IDBPs. See Intracellular vitamin D binding proteins Idiopathic hypercalciuria (IH), 1339–1346 Ca absorption v. 1,25(OH)2D in, 1342, 1343f diagnosis v. causes of normocalcemic hypercalciuria, 1340, 1340t dietary Ca restriction v., 1351 elevated 1,25(OH)2D in, 1341–1342, 1342f genetic hypercalciuric rats as animal model of, 1346–1350 human genetic, 1346f, 1350–1351, 1351t inheritance, 1340–1341, 1340f intestinal Ca absorption in, 1341, 1341f low bone mass associated with, 1342–1343 nephrolithiasis in, 1142, 1339–1352 overview, 1339–1341 pathogenesis of human, 1341–1346 pathogenetic models of, 1343–1346, 1344f external Ca balance v., 1344–1345, 1345f fasting PTH/urine Ca v., 1344 1,25(OH)2D excess v., 1344f, 1345–1346, 1347f tests of, 1344–1346 urine Ca/Ca balance/low Ca diet v., 1345, 1345f, 1346f renal Ca reabsorption decreased by, 1342 therapeutics v. Ca metabolism, 1351 thiazides v., 1351 IDM. See Infants of diabetic mothers IFNγ. See Interferon-γ IGF. See Insulin-like growth factor IGF-I D metabolism influenced by GH and, 1254t, 1258 20-epi D analogs interfering with, 1498–1499

IGF-II in EB1089 antiproliferative activity, 1714 20-epi D analogs interfering with, 1498–1499 IH. See Idiopathic hypercalciuria Ileum high Ca absorption quantity in, 778 transcellular/paracellular Ca absorption in, 422–424, 423f Ilium, fracture suggesting D deficiency, 973 IMCal. See Intestinal membrane calcium-binding protein IMCD. See Inner medullary collecting duct Immune diseases D analogs v., 1500–1501 VDR polymorphisms v., 1146–1148 Immune responses cardiovascular disease v. D modulating, 902 innate/adaptive layers of, 631 selective intervention in, 631–633, 632t VDR ligands mediating, 631–643 mechanisms involved in, 643, 643f Immune system CKD/1,25(OH)2D3 deficiency/resistance in, 1326–1327 D-deficiency-associated abnormalities in, 1389 D/diabetes v., 1767–1773 D endocrine system involved in, 291 D metabolite local regulatory effects on, 1387–1390 DBP role in, 126–127 VDR ablation effect on, 346 Immunosuppressants D analogs combined with, 1500 D metabolism influenced by, 1255t, 1272–1273 Import receptors, nuclear import mediated by, 366, 367f Importins. See Import receptors Infants acute hypocalcemia therapy for, 1058 Ca absorption in human milk/formula-fed, 813, 814t D actions in perinatal, 803–808 D deficiency/Ca absorption in, 811–818 D3 dosage considerations for, 1002–1003, 1009t D supplementation for formula-fed, 808 D supplementation for low-birth-weight, 808 early/late neonatal hypocalcemia in, 1055 infant/maternal D supplementation v. 25OHD in, 846–847, 846f low maternal D/Ca intake v., 841–843 maternal Ca intake v. BMC in, 843, 843f normal term, 804–805 Ca absorption in, 813–814 Ca intake recommendations for, 813, 813t D deficiency/Ca absorption in, 813–814 D in, 814 supplemental D sources for, 814, 814t premature, 805–807 D deficiency/Ca absorption in, 812–813 D in, 806, 806f early neonatal hypocalcemia v., 805–806 nutritional rickets in, 968 osteopenia risk criteria for, 812, 812t postnatal D supplementation for, 806–807 rickets and Chinese, 793 term growth-retarded, 805 VDDR v. black, 791 Infants of diabetic mothers (IDM), pathogenic factors in, 807 Infections children with rickets v., 1068 VDR polymorphisms v., 1148 Inflammation, 126 DBP role in, 126–127

INDEX

Inflammatory bowel disease (IBD). See also Crohn’s disease clinical/biochemical features of, 1302 VDR ligand treatment v., 637t Infrared spectroscopy, tissue mineralization quantified with, 481, 482–483, 482f Inner medullary collecting duct (IMCD), vasopressin-stimulated water reabsorption v. CaR, 558 Inorganic phosphate (Pi) aging v. serum concentration of, 455 chemistry, 453 circadian rhythm in serum concentration of, 455 disorders associated with renal wasting of, 467–470 common metabolic pathway hypothesized in, 469–470 distribution in body, 453 extracellular homeostasis of, 453–455, 454f homeostasis, 453–469 regulation, 1159–1161 v. NPT2a-related signaling, 1161 1α-hydroxylase activity v. restricted, 464, 464f intestinal absorption of, 455–457 cellular aspects in, 455 molecular mechanisms in, 456, 456t NPT2b gene expression v., 455–456, 456t regulation of, 455–456 kidney excreting, 516, 516t kidney reabsorbing, 516 NPT2a regulating reabsorption of, 459, 461t renal transport of, 457–463 cellular aspects of, 457–458, 458f molecular aspects of, 456t, 458–459 physiology/tubular localization in, 457 supplementation in metabolic bone disease, 927 transport in bone, 465–467 type II cotransporter in handling, 1160 Insufficiency in adults/elderly, 1085–1097 consequences of, 1088–1090 deficiency v., 1085–1086 determinants, 1086–1088 elderly v. consequences of, 825–826 medical causes of, 1088 muscle weakness in, 1090 prevalence, 1090–1094 preventing, 1094–1096, 1097 Insulin D metabolism influenced by, 1254t, 1259 EGF keratinocyte proliferation stimulation enhanced by, 614 proximal tubular Pi transport v., 461t renal Pi excretion decreased by, 516t resistance in D deficiency v. HDM, 1816 secretion influenced by 1,25(OH)2D3, 1764–1765 secretion stimulated by 1α,25(OH)2D3, 394–395, 395f serum phosphate depressed by glucose and, 1177 synthesis/secretion in NOD mice, 1765, 1766f synthesis/secretion in VDR KO mice, 1765, 1765f type I collagen synthesis increased by, 704 Insulin-like growth factor (IGF) mitogenic activity inhibited by D compounds, 1665 in 1,25(OH)2D3 actions on prostate cells, 1692–1693, 1692f system interacting with 1,25(OH)2D3/EB1089, 1581–1582 Integument D’s role in, 609–622 VDR ablation effect on, 346–348, 346f Interferon-γ (IFNγ) CYP24A1 up-regulation inhibition v., 100 D metabolism influenced by, 1254t, 1263

1869 Intestinal membrane calcium-binding protein (IMCal), Ca entry v., 416 Intestine. See also Colon; Duodenum; Ileum active Ca transport in, 414 age v. 1,25(OH)2D resistance/Ca absorption by, 1104–1105, 1105f calbindin-D9K/D28K in, 722–724, 730t CaR in, 559 CKD with low 1,25(OH)2D3 or 1,25(OH)2D3/VDR resistance in, 1322 corticosteroids influencing Ca absorption in, 445 CYP27B1/CYP24 expression in, 1717–1719 D analogs v. induced tumors in, 1451 D deficiency/depletion v., 1294 D-dependent Ca absorption mechanisms in, 433–440 dexamethasone influencing Ca absorption in, 445, 447f ECaC2/calbindin-D9K mediating CA absorption by, 430f, 440 estrogens v. Ca absorption in, 444–445 gestation v. Ca absorption in, 440–442, 441f, 441t, 442f 24-hydroxylase enzyme regulation in, 92 lactation v. Ca absorption in, 441f, 441t, 442–444, 442f, 443f LCA detoxification in, 867–868, 868f 1,25(OH)2D3 and Ca absorption by, 411–424 1,25(OH)2D3 antioxidant activities/CA absorption in, 764t, 767 25OHD correlated with Ca absorption efficiency in, 783 paracellular path in Ca absorption, 421–422 Pi absorption in, 455–457 cellular aspects of, 455 molecular mechanisms of, 456, 456t NPT2b gene expression v., 456–457, 457t regulation of, 455–456 segments v. Ca absorption, 413, 422–424, 423f transcellular/paracellular Ca absorption in, 429, 430f Intoxication Ca/24-hydroxylase catabolic pathway in, 1365–1367, 1366t Ca v. renal 1α-hydroxylase in, 1365, 1365t, 1366t clinical manifestations of, 1368 D absorption/input v. risk of, 784 D2 causing most cases of, 1008 DBP/free metabolite level in, 1367–1368 diagnosis of, 1369 diagnosis of endogenous D, 1392–1393 diagnosis/prevention/treatment v. endogenous, 1392–1394 excessively fortified milk causing, 1009t–1010, 1066 forms of exogenous, 1355–1359 hypercalcemia due to, 1355–1372 metabolic bone disease patient SH v., 917 myocardial ischemia v. EKG changes in, 1368 sarcoidosis associated with endogenous D, 1379–1380 target tissue response to 25OHD in, 1364 TPTX rat 24-hydroxylase activity v., 1366–1367, 1367t treatment of, 1369–1370 VDR in, 1362–1365 Intracellular receptor gene family, VDR in, 172–173 Intracellular vitamin D binding proteins (IDBPs) D analog selectivity influenced by, 1462–1463 in D-resistant New World primates, 359–360 hsp70 family members homologous to, 359, 360f in intracellular D trafficking model, 360 in 25OHD3 1α-hydroxylation, 73 “sink”/“swim” hypotheses for, 359–360, 360f Intragenic interaction, VDR gene analysis v., 1136f, 1138 Irish, winter hypovitaminosis D in adult, 1091, 1091f Isoflavones, PCa combination therapy with soy, 1697 Italians Ca v. colorectal cancer in, 1622

1870

INDEX

Italians (Continued) climate v. D in, 1026 hypovitaminosis D in elderly, 1092–1093 winter hypovitaminosis D in adult, 1091, 1091f

J Japanese Buddhist vegetarians v. metabolic bone disease, 917t Cdx2 polymorphism in, 1126 CKD/VDR polymorphisms/expression in, 1319 FokI RFLP v. height in young adult, 1143 hypovitaminosis D in adult, 1091 PCa risk for indigenous, 1681 PCa risk in, 1601 VDR gene polymorphism v. BMD in, 243 VDR polymorphism v. PCa risk in, 1682 VDR polymorphisms v. diabetes in, 1146 VDR polymorphisms v. MS in, 1147 VDR polymorphisms v. PCa in, 1139t, 1145 VDR polymorphisms v. psoriasis in, 1146 VDR polymorphisms v. renal cell carcinoma in, 1146 westernization v. PCa in, 1600, 1600f Jejuno-ileal bypass bone disorders associated with, 1302–1303 clinical features of, 1302–1303 management, 1303 Jejunum, high Ca absorption quantity in, 778 Jews metabolic bone disease v. lactose intolerance in, 917t serum 25OHD in light-skinned, 794–795 JG cells. See Juxtaglomerular cells JNK. See Jun-N-terminal kinase Jun-N-terminal kinase (JNK), in RANKL-induced osteoclastogenesis, 678 Juxtaglomerular (JG) cells, renin synthesized/secreted by, 871, 872f

K Kellgren score, OA diagnosed with, 1144 Keratinocyte-GF (KGF) BPH cells v. BXL–353 and, 1836–1837, 1837f human BPH cells v., 1834, 1835f Keratinocytes Ca regulating proliferation/differentiation of, 609 Ca sensing mechanism of, 615–616, 615f Ca switch inducing changes in, 616 Ca switch stimulating phosphoinositide metabolism in, 617–618 clinical implications of 1,25(OH)2D production by, 612–613 DRIP205/p160 recruitment in differentiation of, 273 epidermal layers of, 613, 613f growth/differentiation regulators of, 613–615 24-hydroxylase enzyme regulation in, 93 1,25(OH)2D3/deltanoid-induced differentiation modeled in, 1637t 1,25(OH)2D3 inhibiting differentiation of, 1781 1,25(OH)2D3 inhibiting PTHrP production in, 741t 1,25(OH)2D production by transformed, 612 1,25(OH)2D production regulated by differentiation of, 611–612 1,25(OH)2D3 protecting epidermal, 764t, 766–767 1,25(OH)2D regulating differentiation of, 619–621, 620f regulation of differentiation by, 613–621 retinoic acid receptors identified in, 614 VDRs in mammalian/lamprey, 228

Ketoconazole D metabolism influenced by, 1255t, 1266–1267 kidney 1α-hydroxylase activity regulated by, 78 KGF. See Keratinocyte-GF KH1060 analog-VDR complex stabilized by, 1496 autoimmune type I diabetes prevented by, 1500 breast cancer cells/tumors v., 1669 in leukemia combination therapy, 1733 leukemic cells v., 1735 metabolism, 1438, 1439f metastasis v., 1576 as noncalcemic analog, 1440, 1441t structure v. 1α,25(OH)2D3, 285–287, 286f synthesis, 1490–1491, 1492f, 1496t Kidney aging v. D responsiveness by, 833 Ca handling by, 515–516 calbindin-D9K/D28K in, 724, 730t CaR in, 556–558 CKD with low 1,25(OH)2D3 or 1,25(OH)2D3/VDR resistance in, 1325–1326 D analogs treating chronic disease of, 1827 D-dependent protein distribution/regulation in, 519–528 D-responsive proteins in, 520t D v., 515–528 DBP uptake in megalin-deficient, 154, 154f, 155f dystrophic mineral deposits in, 477–478, 478t ECaC1 expression restricted to mouse, 431 function deteriorating with aging, 827 glucocorticoids directly affecting, 1243, 1243f 1α-hydroxylase/24-hydroxylase expression in fetal, 852–853, 853f 24-hydroxylase distribution in, 528 24-hydroxylase enzyme regulation in, 91–92 24-hydroxylase expression in, 86–87 24-hydroxylation in, 21–22 JG apparatus producing renin in, 871 keratinocyte 1,25(OH)2D negative feedback loop v., 611 mass/CKD v. 1,25(OH)2D bioactivation, 1379–1380 mouse models with defective phosphate transport by, 462–463 1,25(OH)2D3/analog immunoregulatory properties in, 1519 1α,25(OH)2D3-mediated rapid response in, 386t in 25OHD metabolism, 516–518 1,25(OH)2D3 produced by, 782 1α,25(OH)2D3 produced in, 69, 70f 1,25(OH)2D v. phosphate transport in, 1161–1162, 1162f 1,25(OH)2D3/VDR-mediated signal termination in, 221f, 222 1α,25(OH)2D3 VDRnuc in, 385t Pi flux in human, 453–455, 454f Pi handling by, 516, 516t Pi homeostasis arbitrated by, 1159 Pi transport in, 457–463 cellular aspects of, 457–458, 458f molecular aspects of, 456t, 458–459 physiology/tubular localization of, 457 PTHrP expression in, 739t rejection inhibited by 1,25(OH)2D3/analogs, 641t tubular defects and hypophosphatemia, 981–984 VDR in, 520–523 polyclonal antibodies detecting, 521, 521f Kidney disease 1,25(OH)2D secretion v., 1313 secretory control/hormonal interactions v., 1313, 1314f Klotho gene, kidney 1α-hydroxylase activity v., 78

1871

INDEX

Knockout (KO) mice alopecia in, 1229–1230 c-src, osteopetrosis in, 677 calbindin D28K, urinary Ca/creatinine ratio in, 525 cyp24A1 growth plate architecture in, 108–109, 108f 24-hydroxylation in, 107–109 phenotype of, 107–109, 107t cyp27A1 hepatic 25-hydroxylation in, 105–106 phenotype of, 106 cyp27B1 D-dependent gene expression in, 110 1α-hydroxylation in, 109–113 hypocalcemia/hypophosphatemia/secondary hyperparathyroidism in, 703 phenotype of, 109–110 phenotype rescue in, 110–112 DBP, 149 serum D metabolites v. bone abnormalities in, 489f viable immune function in, 127 ERα/ERβ, 441f, 445 FGF-23, phosphate homeostasis in, 1193–1194 full-length CaR, 616–617 intestinal Ca absorption in, 429–437, 446t JNK1, RANKL-induced osteoclastogenesis in, 678 megalin, 154–156 Npt2 Ca absorption in, 440 generation/characteristics of, 439–440 phosphate homeostasis in, 439–440 Npt2a intrinsic osteoclast defect in, 467 Npt2c expression in NHERF and, 1162 renal phosphate transport defects in, 462 renal Pi reabsorption v. P450c1α gene expression in, 464f, 465 1α(OH)ase ECaC1 expression reduced in, 438 generating, 438 hypocalcemia/rickets/osteomalacia in, 489 lower calbindins-D expression in, 525 mineral homeostasis in, 438–439 PDDR in, 438–439, 489 prehypertrophic chondrocytes showcased by, 578 RXRα conditional, alopecia in, 233 SRC coactivator family models in, 294–295 VDR, 224–225, 341–348 alopecia in, 620, 621, 665 Ca absorption in, 433–437, 434f Ca absorption v. Ca transporter gene expression in, 435–437, 436t, 437t Ca entry v. Ca transport in, 417 Ca/P ameliorating skeletal abnormalities in, 1403–1404 calbindin D28K expression v. age in, 525 D-dependent active intestinal Ca absorption in, 433–438 dietary intervention v. Ca absorption in, 434f, 436t, 437–438 estrogen deficiency in, 441f, 445 generating, 433 gestation/intestinal Ca absorption in, 440–442, 441f, 441t, 442f HVDRR/alopecia link in, 621 hypocalcemia in, 665 infertility in, 665 intraperitoneal glucose tolerance test in, 1765, 1765f lactation/intestinal Ca absorption in, 441f, 441t, 442–444, 442f, 443f renin expression/Ang II production in, 875–877, 876f

KO mice. See Knockout mice Koreans VDR polymorphisms v. psoriasis in, 1146 VDR polymorphisms v. RA in, 1147 Kuwaitis, sunlight exposure v. rickets in, 1066–1067

L Lactation BMC/BMD during, 843f, 844–845 Ca homeostasis v., 204 D/Ca metabolism during, 843–844 D metabolism in, 839, 843–847 maternal Ca economy v., 845, 845f PRL v. D metabolism during, 1258 VDR WT/KO mouse intestinal Ca absorption during, 441f, 441t, 442–444, 442f, 443f Lactose intolerance, Ca intake v., 1026 Lampreys, VDRs in, 227–228, 279 Late neonatal hypocalcemia (LNH), 807–808 Latinas, hVDR polymorphisms v. breast cancer risk in U.S., 245 LBDs. See Ligand binding domains LC/MS. See Liquid chromatography/mass spectrometry LCA. See Lithocholic acid LD. See Linkage disequilibrium Lebanese climate v. D in, 1026 D metabolism in, 794 hypovitaminosis D in, 1087 low serum 25OHD in, 1026 Leo KH-1060, SAR in design of, 1412, 1413f Leprosy extrarenal D metabolite overproduction v., 1390t, 1391 hypercalcemia/D hypersensitivity in, 1361 Leukemia D analogs effective against, 1734–1736 D compound combination therapy v., 1733–1734 RUNX TF localization in acute myelogenous, 335 Leukemia cells bisphenol compounds v. differentiation of, 1559, 1559f calcipotriol v., 1734 D compounds v., 1730–1734, 1730t, 1731, 1731t molecular mechanisms of D compounds v., 1731–1733 1,25(OH)2D3 v. myeloid, 1731 1αOHD3 v., 1734 Leukocytes, antigenic stimulants activating phagocytic function in, 127 Libyans, rickets in infant, 795 Ligand binding domains (LBDs) coactivator binding v. HVDRR-related mutations in VDR, 1225–1226 Glu420Lys, 1226 connecting region poorly conserved in, 279–280 crystal structure of D NR, 279–288, 280f D analogs inducing conformational change in VDR, 1452f, 1453, 1550, 1550f deltanoids v. nuclear VDR, 1408–1411, 1410f 16-ene-24-sulfone deltanoid v. rickets mutant VDR, 1411, 1411f human/rat VDR, 174–176, 175f HVDRR/alopecia from Glu329Lys/366delC mutations in VDR, 1221f, 1226 HVDRR-related mutations in VDR, 1222–1226 hypothetical conformations of VDR, 1473, 1473f ligand binding changing conformation of VDR, 176, 279, 280f, 293, 294f, 313, 321

1872 Ligand binding domains (Continued) NRs dimerizing via, 314 1,25(OH)2D3 binding v. HVDRR-related mutations in VDR, 1222–1224, 1223f Arg274Leu, 1222–1223, 1223f Cys190Trp, 1223f, 1224 His305Gln, 1223, 1223f Ile314Ser, 1223, 1223f Ile268Thr, 1223f, 1224 structural analysis of, 1224 Trp286Arg, 1223f, 1224 1,25(OH)2D3 stabilizing VDR, 313 proteins with 1α,25(OH)2D3, 387–392 putative alternative VDR, 400 “squelching” in NR, 291–292 structures in hVDR∆/zVDR, 284–285, 284f topology of hVDR∆, 281, 283f VDR, 1211–1212, 1211f in VDR/DNA binding, 178 VDR-RXR heterodimerization v. HVDRR-related mutations in VDR, 1224–1225 Arg391Cys, 1221f, 1224 Gln259Pro, 1221f, 1225 Phe251Cys, 1221f, 1225 structural analysis of, 1225 VDRnuc v. DBP, 387, 387f zVDR, 284, 284f zVDR-Gemini channel showing adaptability in, 287–288 Ligands allograft tolerance induced using VDR, 1519 BPH inhibited by VDR, 1833–1840 classes of nonsecosteroid D, 1565 genomic/nongenomic action specificity of, 1461 hVDR∆ binding, 280–281 identification of ODF, 670–673 immune responses regulated by VDR, 631–643 immunoregulation by VDR, 633, 643, 643f immunosuppressive agents combining with VDR, 642 nonsecosteroid VDR, 1557, 1558f novel mammalian VDR, 225–227, 227f potential uncharacterized novel VDR, 234f, 235 regulatory T cells enhanced by, 636 structure of hVDR complexed to superagonist, 285–287, 286f tissue selectivity in VDR, 271 VDR immunomodulatory mechanisms in autoimmune disease models, 636–640, 637t VDR signaling activated by, 235–237, 236f Linkage disequilibrium (LD) FokI RFLP surrounded by small, 1128f, 1130 polymorphisms predicted by, 1126–1127 Lipids, bile acids in digesting/absorbing, 863 Lipopolysaccharide (LPS) in macrophage 1α-hydroxylase amplification, 1385–1386, 1385f 1,25(OH)2D3 down-regulating NOS II in injection of, 1783 Liquid chromatography/mass spectrometry (LC/MS), 20-methyl-1α,25(OH)2D3 metabolite, 1434, 1435f Lithium, D metabolism influenced by, 1255t, 1274 Lithocholic acid (LCA) detoxification amplified by supplemental D, 246, 247f in ED-71 synthesis, 1534, 1534f intestinal detoxification of, 867–868, 868f as nonsecosteroid VDR agonist, 1557, 1558f VDR regulating CYP3A-dependent detoxification of, 867 Liver D 25-hydroxylation in, 17–19 in D metabolism, 1294–1296

INDEX

Liver (Continued) D metabolites secreted in, 1295 D3 uptake by, 47–48 DBP production in, 121 enterohepatic circulation in, 863–864, 1295–1296 1,25(OH)2D3/analog immunoregulatory properties in, 1519 1α,25(OH)2D3-mediated rapid response in, 386t PTHrP expression in, 739t rejection inhibited by 1,25(OH)2D3/analogs, 641t Liver disease. See also specific liver diseases bone disorders associated with, 1303–1306 Liver X receptor (LXR), cholesterol/bile acid levels controlled by, 865 LNH. See Late neonatal hypocalcemia Locke, John, rickets reported by, 967 Looser’s zones impaired mineralization ambiguously shown by, 1043 in osteomalacia, 971–973, 972f–973f radionuclide bone scans detecting, 986–988, 989f in severe rickets, 1071 in XLH patients, 981, 982f, 984f LPS. See Lipopolysaccharide LXR. See Liver X receptor Lymphocytes, paracrine 1,25(OH)2D suppression of, 1387f, 1388–1389 Lymphoma, hypercalcemia in, 1361–1362 Lysosomes, DBP degradation in, 157, 157f Lythgoe coupling, in steroid precursors, 1414–1416, 1414f, 1415f

M Macrophages 24-hydroxylase enzyme regulation in, 92 D-1α-hydroxylase in, 1381–1382, 1382f 1α-hydroxylase v. immune cell regulators in, 1384–1387 intracrine/autocrine 1,25(OH)2D activation of, 1387–1388, 1387f 1,25(OH)2D-directed 24-hydroxylase activity in, 1383–1384 PTH/Ca/Phosphate responsiveness lacking in, 1382–1383, 1383f Magnesium (Mg). See also Hypermagnesemia; Hypomagnesemia CaR v., 553–554 jejuno-ileal bypasses reducing serum, 1303 in PTH secretion, 1053 supplementation in acute hypocalcemia therapy, 1058 Magnetic resonance imaging (MRI) acid phosphate distribution/crystal structure from, 483 in evaluating bone metabolic disease, 924 tumors causing TIO v. whole body, 988 Malignant hyperthermia, hyperphosphatemia in, 1178 Malnutrition hypocalcemia due to, 1052f, 1055–1056 in last trimester of pregnancy, 803 Mammary gland D in, 857–858 development role of D endocrine system, 291 development v. VDR, 345–346 VDR expression/role in normal, 1669–1670, 1670t MAP kinase. See Mitogen-activated protein kinase MAR. See Mineral apposition rate MARRS. See Membrane-associated rapid response steroid binding protein Matrix attachment regions (MARs) chromatin units between, 314, 315f genomic domain/nuclear scaffold association mediated by, 329f Matrix extracellular phosphoglycoprotein (MEPE), as phosphatonin/minhibin, 1164, 1166

INDEX

Matrix Gla protein (MGP), in mineralization v. 1α,25(OH)2D3, 716 Matrix metalloproteinase–9 (MMP–9) gene, 1,25(OH)2D3 targeting, 566, 567 Matrix vesicles D metabolites modulating PKC in, 586 extracellular matrix growth factors activated by, 589–590 genomically controlled production of, 589 in matrix calcification, 589 nongenomic regulation of, 589–591 proposed mechanism for, 590–591, 590f 1,25(OH)2D3 accelerating crystal formation in, 591 1,25(OH)2D3/24,25(OH)2D3 affecting matrix calcification through, 590–591 Pi transport in, 466 Maturation (mitosis) promoting factor (MPF), cell cycle G2/M transition regulated by, 1645, 1645f Maxacalcitol. See 22-Oxa-calcitriol Maximum voluntary contraction (MVC) D v. veiled Danish Arabs’, 1812, 1812f 25OHD v., 1812 2MbisP. See 2-Methylene-19-nor-(20S)-1αhydroxybishomopregnacalciferol MC1288 analog-VDR complex stabilized by, 1496 chronic graft rejection inhibited by, 642 structure v. 1α,25(OH)2D3, 285–287, 286f as superagonist, 1477–1478 VDR transcriptional activity v., 1474t MC–903 cellular differentiation assay v., 1543 psoriasis v., 1543 McCollum, E. V., in vitamin A/B/D discoveries, 4 McCune-Albright syndrome, fibrous dysplasia as component of, 1192 MCM proteins. See Mini-chromosome maintenance proteins MCR. See Metabolic clearance rate 2MD. See 2-Methylene-19-nor-(20S)-1,25-dihydroxyvitamin D3 2αMD. See 2α-Methyl-19-nor-(20S)-1,25-dihydroxyvitamin D3 Mediator-D coactivator complex, 265–267 in comodulator activity integrated model, 300, 300f D analogs recruiting, 1455 functionality, 266–267 HAT activity absent in, 267, 295 identification, 265–266 MC1288 enhancing recruitment of, 272–273 multiple binding motifs in, 269 other complexes closely related to, 266 subunit composition, 265t transcription factors interacting with, 269–270 unified nomenclature v. DRIP, 295 VDR linkage to, 176 VDR-RXR heterodimers interacting with, 296, 296f in VDR transcription model, 268–270, 269f Medullary thick ascending limbs (MTAL), CaR inhibiting NaCl reabsorption in, 557 Medullary thyroid carcinoma (MTC) 5-HT1 receptor activation v. Ca in, 693, 694f C cells v., 688 D analogs v., 696 hereditary/sporadic, 688 Megalin, 157–158, 157f, 158f expression v. 1,25(OH)2D3 resistance in CKD, 1319 in 25OHD3 1α-hydroxylation, 73 as 25OHD3/DBP complex endocytic receptor, 154–156 Megalin-DBP complexes, in 25OHD3/1,25(OH)2D3 conversion, 125 Melanin, vitamin D photosynthesis v., 38–39, 40–41 Melanoma, VDR polymorphisms v., 1146

1873 Membrane-associated rapid response steroid binding protein (MARRS), odontocytes expressing, 599, 600f Membrane VDR (VDRmem), 391–392 membrane-initiated responses v. 1α,25(OH)2D3 interacting with, 383 rapid responses v. 1α,25(OH)2D3 interacting with, 399–400, 399f Men. See also Prostate cancer D intake v. PCa risk in American, 1681 normative histomorphometric data for, 956t 1,25(OH)2D concentration/MCR/PR in young/elderly, 827, 827f PCa in American, 1679 testosterone v. 1,25(OH)2D3 in hypogonadal, 1261 Metabolic bone disease adverse influences on mineral/skeletal homeostasis in, 915, 915t approaching patients with, 913–928 biochemical investigation of, 924–926 diagnostic evaluation of, 913–926 epidemic, 913 histopathological assessment of, 926 HPI in addressing, 916 laboratory testing for, 921, 921t medical history in evaluating, 914, 915–918, 915t patient complaints in addressing, 915, 916t physical examination for, 916t, 919–921, 919t radiological examination for, 921–924 review of systems in addressing, 918 treatment, 926–928 Metabolic clearance rate (MCR), young/elderly men’s 1,25(OH)2D, 827, 827f Metabolism, 15–29, 1000f, 1423–1424 aging v., 823–833 aging v. 1,25(OH)2D, 826–831 Al v., 1255t, 1270–1271 analog, 1423–1443 biological systems in studying, 1428–1429, 1428t drug design implications of, 1442–1443 examples of, 1429–1440 general considerations in, 1423–1429 implications from studying, 1440–1443 invalid in vivo/in vitro comparisons in, 1442 non-D-related, 1428–1429 pharmacokinetic information correlating with, 1440–1441, 1441t questionably valid assumptions in, 1442 radioactive analogs in studying, 1428–1429 anticonvulsants v., 1255t, 1264–1265 antituberculous agents v., 1255t, 1271 bisphosphonates v., 1255t, 1268 Ca channel blockers v., 1255t, 1269 caffeine v., 1255t, 1271–1272 calcitonin v., 1254t, 1256–1257 cimetidine v., 1255t, 1270 corticosteroids v., 1255t, 1265–1266 cyclopropane ring-containing analogs’, 1431–1433, 1432f D, 15–29, 16–17 mutant mouse metabolic enzyme models v., 105 overview, 15–17 species variation in, 27–29, 28t D2 D3 metabolism v., 16–17 pathways of, 18f unique aspects in, 25–26 D3, 69–70, 70f, 85–86, 86f, 1000–1002, 1000f pathways of, 16–17, 17f D3 abundance v., 1000f, 1001 D2 derivative, 1430–1431

1874 Metabolism (Continued) dihydrotachysterol, 1429–1430, 1430f drugs influencing, 1255t, 1263–1274 20-epi-/20-methyl analog, 1433–1434, 1433f, 1435f estradiol v., 1254t, 1259–1261 ethanol v., 1255t, 1266 exogenous/stimulated PTH v., 1253–1255, 1254t fluoride v., 1255t, 1273 gastrointestinal disease v., 1293–1297 GH/IGF-I v., 1254t, 1258 GH v., 1254t, 1257 GIO v., 1239–1248 heparin v., 1255t, 1269–1270 homologated analog, 1434–1436 hormones influencing, 1254t, 1263–1274 in HVO, 1039–1040 D-24-hydroxylase in, 1459–1461, 1460f hypolipidemics v., 1255t, 1267–1268 IFNγ v., 1254t, 1263 immunosuppressants v., 1255t, 1272–1273 insulin v., 1254t, 1259 ketoconazole v., 1255t, 1266–1267 kinetics of 1,25(OH)2D, 828 lithium v., 1255t, 1274 liver’s roles in, 1294–1296 in normal/neoplastic colon cells, 1715–1719 olestra v., 1255t, 1273 orlistat v., 1255t, 1274 oxa-group containing analog, 1438 phosphate homeostatic disorders v. paradoxical regulation of, 1161 Pi-dependent modulation of, 1161–1162 progesterone v., 1254t, 1261–1262 prolactin v., 1254t, 1258 prostaglandins v., 1254t, 1262–1263 PTHrP v., 1254t, 1255–1256 reciprocal regulation of phosphate homeostasis, D, 1159 sex steroids v., 1254t, 1259–1262 testosterone v., 1254t, 1261 theophylline v., 1255t, 1272 thiazide diuretics v., 1255t, 1268–1269 thyroid hormone v., 1254t, 1262 TNFα v., 1254t, 1263 unsaturated analog, 1436–1437 in vitro regulation by 1,25(OH)2D3/EGF, 1717 Metabolites, early evidence of converted, 6–7 Metastasis, D compounds v., 1576 Metastatic calcification in arteries/around joints, 979, 980f chronic renal failure causing, 979 treating, 979, 980f Mg. See Magnesium MGP. See Matrix Gla protein MI. See Mineralization index; Myocardial infarction Mice. See also specific types of mice in coculture system recruiting osteoclasts, 666–667, 667f 1,25(OH)2D3 v. diabetes/MS in NOD, 1404 Michaelis-Menten equation Ca flux described by, 412 high dose intermittent 1,25(OH)2D3 fitted to, 1742, 1743f Miconazole, kidney 1α-hydroxylase activity regulated by, 78 Microcomputerized tomography (µCT), bone structural data provided by, 481 Microvillar membrane. See Brush border Middle Easterners low 25OHD in immigrant, 1026 rickets in immigrant, 968–971

INDEX

Migraine, 1,25(OH)2D3 repression of CT/CGRP expression v., 690 Mineral apposition rate (MAR) bone formation indices derived from, 955–956 ED-71 v. ovariectomized rat, 1537, 1537f in Mlt, 1033 OCT v., 1528, 1529t Mineral homeostasis, 411–559, 453–469 adverse influences on, 915, 915t clinical disorders of phosphate, 1159–1180, 1164t ECaCs v. Ca, 430–432 factors causing disorders in phosphate, 1162, 1164t fetal/maternal Ca, 853–854 kidney in, 515–528 maternal D status v. neonatal Ca, 843 metabolic bone disease diagnosis v., 914 1,25(OH)2D3 regulating Ca/P, 1403 Pi, 453–469 extracellular, 453–455, 454f PTHrP in fetal C9, 3–4 regulation of Pi, 1159–1161 systemic/intracellular Ca2+, 751 VDR ablation v., 341–345 VDR-null mice skin change v., 346f, 347 VDR-null mice with normal, 344 Mineralization, 477–490 bone histomorphometry assessing, 954–957 complex mineral transport in, 1044 D/D metabolites influencing, 478–480, 565–571 D-deficiency decreasing, 484, 485f D-deficiency v., 487–488 D influencing, 483–487, 565–566 D v. cell/matrix molecule, 479t, 484–487 D v. pathogenesis of impaired, 1040–1044 definitions/terminology, 477–478 dynamic indices of, 955–957 epitaxial/heterogeneous crystal formation in, 477 gene expression v. 1α,25(OH)2D3 during, 715–716 histology, 1036t hypocalcemia due to accelerated skeletal, 1057 maternal D status v. infant, 842–843 microscopic examination of in situ, 1029–1030 in model systems with D alterations, 487–490 variations/dependencies, 487 morphologic/biochemical aspects of, 1029–1031 normocalcemic/D-deficient, 1043 1α,25(OH)2D3 v. available Ca in, 711 osteoblasts influencing, 1044 osteoid indices in recognizing impaired, 1034–1035 as phase transformation, 1030 phosphatonins/minhibins regulating, 1163 physical chemistry of, 478, 479t possible D targets influencing, 509–510 precipation in, 483–484 primary/secondary, 1032 quantifying tissue, 480–483 ash weight/density in, 481 mineral characterization in, 481–483, 482f questions in, 480 radiographic methods showing, 480–481 two types of bone, 1030–1031 type I collagen required for bone matrix, 703 unexpected/dystrophic, 477–478 VDR alteration v., 488–489, 489f in vitro, 483 Mineralization index (MI), osteomalacia diagnosed with, 1035–1036

1875

INDEX

Mineralization lag time (Mlt), 957 in osteoid accumulation/osteomalacia pathogenesis, 1033–1034, 1033f osteomalacia defined using, 1035–1036 Mini-chromosome maintenance (MCM) proteins, in DNA licensing/replication, 1644–1645 Mitogen-activated protein (MAP) kinase cascade up-regulation, 888 CGRP transcription controlled by, 692–693, 692f pathways in 1,25(OH)2D3 differentiation signaling, 1638–1639, 1638f Mitosis, MPF regulating cell cycle transition to, 1645, 1645f Mixed-function oxidases, in 25OHD3 1α-hydroxylation, 71–73, 71f Mlt. See Mineralization lag time MMP–9 gene. See Matrix metalloproteinase–9 gene Monocytes 24-hydroxylase enzyme regulation in, 92 intracrine/autocrine 1,25(OH)2D interaction with, 1387–1388, 1387f Moroccans, serum 25OHD/PTH in dark-skinned Dutch, 794 2MP. See 2-Methylene-19-nor-1α-hydroxyhomopregnacalciferol MPF. See Maturation (mitosis) promoting factor 2Mpregna. See 2-Methylene-19-nor-1α-hydroxypregnacalciferol MRI. See Magnetic resonance imaging MS. See Multiple sclerosis MTAL. See Medullary thick ascending limbs MTC. See Medullary thyroid carcinoma µCT. See Microcomputerized tomography Multiple sclerosis (MS), 1780 D3 v., 999t EAE role supporting 1,25(OH)2D3 involvement in, 1784 1,25(OH)2D3 v. NOD mouse, 1404 VDR polymorphisms v., 1147 Muscle biopsies in HDM clinical studies, 1811 Ca v. D deprivation/repletion, 1810f, 1811 cell culture/animal studies of D, 1809–1811 contraction, 1807, 1807f, 1808f contraction/relaxation v. D, 1809, 1809f D influencing, 883–894 D influencing striated, 1809–1811 D3 metabolites v. phosphate uptake by, 887 dystrophic mineral deposits in, 478t falls v. D’s influence on, 1805–1817 fiber types, 1808 filament components, 1807, 1807f function in HDM clinical studies, 1811–1813 function v. VDR polymorphisms, 1813 genomic/nongenomic D influence on striated, 1809 1α,25(OH)2D3 mechanisms in, 889–893 genomic, 889–890 nongenomic, 890–893, 890f 1α,25(OH)2D3 regulating Ca homeostasis in, 886–887 1α,25(OH)2D3-stimulated signaling in growth of, 888–889, 889 physiology v. HDM, 1806–1808, 1806f, 1807f power v. serum 25OHD, 1812–1813, 1812f proliferation/differentiation influenced by 1α,25(OH)2D3, 887–889 regeneration v. 1,25(OH)2D, 1816 tyrosine phosphorylation mediating 1α,25(OH)2D3 in growth of, 888 VDR in, 885–886 VDR polymorphisms v. strength of, 1143 weakness with D insufficiency, 1090

Muscle cells ATP supplying energy to, 1808 Ca homeostasis in, 1807–1808, 1808f 1,25(OH)2D3/deltanoid-induced differentiation modeled in, 1637t MVC. See Maximum voluntary contraction Myelodysplasia, 1α-(OH)D3 v., 1573 Myocardial infarction (MI), VDR polymorphisms v., 1148 Myopathies animal model studies of, 884–885 clinical background of, 883–884 D-dependent, 883–885, 893

N Na. See Sodium Na/Pi cotransporters, type I/II/III, 1159–1160 NaCl. See Sodium chloride NADPH-ferredoxin reductase, in 25OHD3 1α-hydroxylation, 71f, 73 Native Americans. See American Indians NCoA62/SKIP coactivator complex, 296, 297f in comodulator activity integrated model, 300, 300f transcription/mRNA slicing coupled by, 298 VDR gene expression comodulated by, 296–297, 297f VDR interaction independent of AF-2 domain, 296 NCoR, NR co-repressor, 298–299 NDOs. See Non-digestible oligosaccharides Necrosis, as inflammatory process component, 126 Negative VDRE (nVDRE) as complex VDR binding site, 321–322 positive VDREs v., 321–322 VDR-RXR heterodimer repressing transcription through, 309–310, 310f Neonatal severe hyperparathyroidism (NSHPT), homozygous inactivating CaR mutations causing, 553 Nephrolithiasis in IH, 1142, 1339–1352 renal histopathology in Ca oxalate, 1341 Nephron Ca reabsorption in, 519, 520f 25OHD3 in, 519 Nerve growth factor (NGF), 1,25(OH)2D3 stimulating, 1781–1782 Nervous system calbindin-D28K in, 726–727, 730t CKD/1,25(OH)2D3 deficiency/resistance in, 1327 Neurotrophic factors, 1,25(OH)2D3 influencing, 1781–1782 Neurotrophin receptor, 1,25(OH)2D3 regulating low-affinity, 1782 New World primates biochemical phenotype of rachitic, 352–353, 353f D hormone resistance in, 352, 354–355 early evolution of, 351, 352f high 1,25(OH)2D3 levels in, 353, 353f IDBP function in, 359–360, 360f IDBPs over-expressed by, 359 VDP carrying 25OHD in, 352 NF-κB, activation in osteoclast differentiation, 678 NGF. See Nerve growth factor Nigerians, sunlight exposure v. rickets in, 1066–1067 Nitric oxide (NO), in macrophage 1α-hydroxylase regulation, 1386 NLSs. See Nuclear localization sequences NMR. See Nuclear magnetic resonance NO. See Nitric oxide Non-digestible oligosaccharides (NDOs), enhancing Ca absorption from foods, 818

1876 Non-Hodgkin’s lymphoma extrarenal D metabolite overproduction in, 1390t, 1391–1392 1α-(OH)D3 v., 1573 PTHrP associated with end-stage, 740 Nongenomic response, steroid hormones in, 98–99 Normocalcemia, CaR in restoring, 551 NOS II. See Type II nitric oxide synthase NPC. See Nuclear pore complex NPT2a, not responsible for HHRH, 469 NPT2a gene, expression/protein production regulation in kidney, 1160–1161, 1160f NPT2a protein mediating Na/Pi cotransport, 459 Na/Pi transport changes v., 1160–1161 PTH/AKAP79/RAP retrieving, 461 regulation, 459–462, 461t signaling mechanisms in insertion/retrieval of, 1161 structure-function, 459, 460f NPT2b protein, molecular mechanisms regulating, 456–457, 457t NRs. See Nuclear receptors NSHPT. See Neonatal severe hyperparathyroidism Nuclear localization sequences (NLSs) calcitriol-induced transcription v. mutant, 372f hVDR putative, 368–369, 368f RXR, 369, 370f Nuclear magnetic resonance (NMR), acid phosphate distribution/crystal structure from, 483 Nuclear pore complex (NPC) nucleocytoplasmic macromolecule transport through, 365–374, 367f VDR/RXR exiting nucleus through, 366f, 369 Nuclear receptors (NRs), 279 apo/holo forms of, 279 bile acid metabolism regulated by, 865–866 co-repressors in, 292, 293f, 298–299 Alien, 299 hr, 299 SMRT/NCoR class of, 298–299 comodulators linking VDR/PIC, 292, 293f crystal structures of LBDs in, 279–288, 280f DNA response element recognition by, 230, 231f everted repeats in, 317–319, 318f export of, 368 first described defect in superfamily of, 1218 human, 225, 226f–227f interaction domains/coactivators in, 235–237 N-terminal variants of, 195 1,25(OH)2D3 inducing focal accumulation of, 374–375, 375f protein interactions facilitating export of, 374 RNA processing linked to coactivators in, 297–298 RXR as heterodimeric partner for, 291, 292f VDRs in superfamily of, 225–228 Nuclear VDR (VDRnuc) 1α,25(OH)2D3 gene transcription mediated by, 383 deltanoids selectively modulating, 1408–1411, 1410f LBD of human, 387–388, 387f, 389f, 390f–391f, 398f LBD v. DBP LBD, 387, 387f 1α,25(OH)2D3 biological actions mediated by, 384–385, 386t structure v. DBP structure, 391, 392t tissue distribution of 1α,25(OH)2D3, 384–385, 385t Nucleosome, hnRNP insertion in, 358 Nucleus acceptor sites in, 334 architectural parameters in, 335–336 gene expression in architecture of, 328–329, 329f, 330f gene expression regulatory component scaffolding in, 332–333

INDEX

Nucleus (Continued) import/export mechanisms in, 365–368, 367f protein export retarded by docking in, 374 signal pathway targeting/integration in compartments of, 335 skeletal gene expression regulatory machinery organization in, 327–336 skeletal regulatory factor trafficking in, 334–335 transcription machinery/factor complexes compartmentalized in, 332 transcriptionally active compartments in, 334–335 VDR translocation to, 1452f, 1454 Nutrition, early research in, 3 Nutritional recovery syndrome, hypophosphatemia in, 1177 nVDRE. See Negative VDRE

O OA. See Osteoarthritis Obesity bone histomorphometry in, 795–796 bone mass v., 795 D metabolism v., 795 OC. See Osteocalcin OC gene chromatin organization of, 330–332 nuclear/gene expression relationship represented by, 328–329, 331f OCIF. See Osteoclastogenesis inhibitory factor OCIF/OPG characterization, 668–670 discovery, 668–670 OCT. See 22-Oxa-calcitriol ODF. See Osteoclast differentiation factor Odontoblasts D-dependent molecules in, 602–604, 603f dentin elaborated by, 599 PHEX expressed in, 467 VDR gene/MARRS expressed in, 599, 600f Odontogenesis, process/functions, 599–602 1,25(OH)2D. See 1,25-Dihydroxyvitamin D 1,25(OH)2D2. See 1,25-Dihydroxyvitamin D2 1,25(OH)2D3. See 1,25-Dihydroxyvitamin D3 1α,25(OH)2D3. See 1α,25-Dihydroxyvitamin D3 25OHD. See 25-Hydroxyvitamin D 25OHD3. See 25-Hydroxyvitamin D3 25,26(OH)2D3. See 25,26-Dihydroxyvitamin D3 OHO. See Tumor-induced osteomalacia Olestra, D metabolism influenced by, 1255t, 1273 Omt. See Osteoid maturation time Oncogenes in cell cycle/apoptosis control, 1577–1580 1,25(OH)2D3/EB1089 regulating breast cancer cell, 1664–1665 1,25(OH)2D3 regulating expression of, 1577–1579 Oncogenic hypophosphatemic osteomalacia (OHO). See Tumor-induced osteomalacia Oncology, 1,25(OH)2D3/1,25(OH)2D3 analogs in, 1741–1747 OPG. See Osteoprotegerin OPN. See Osteopontin ORC proteins. See Origin recognition complex proteins Origin recognition complex (ORC) proteins, in DNA licensing/replication, 1644–1645 Orlistat, D metabolism influenced by, 1255t, 1274 OS/BS. See Osteoid surface per unit of bone surface Osteoarthritis (OA) VDR polymorphisms v., 1144–1145 in XLH patients, 983

INDEX

Osteoblast genes, D in vivo response in, 566 Osteoblasts in BMU termination, 503 BMUs assembling teams of, 501–502 bone formation by, 498 calbindin-D9K/D28K in, 724–725, 730t as D analog therapy target, 1501 D compounds v. apoptosis of, 1498 D-dependent molecules in, 602–604, 603f D influencing differentiation/activity of, 649–658 death of, 506 different dexamethasone/1,25(OH)2D3 influences on, 652 differentiation of, 712–713 dual role of 1α,25(OH)2D3 in, 713–714, 714f gene expression sequence v., 328 nuclear architecture v. transcription during, 335 1α,25(OH)2D3 v., 712–714, 712f factors stimulating Na/Pi transport in, 466, 466t glucocorticoids increasing differentiation of, 1240 L-type VSCC v. Ca2+ influx into, 754–755, 755f 2MD inducing mineralization in, 1546–1547, 1547f Mlt v. age of, 1033–1034, 1033f Na/Pi transport in renal proximal tubular cells v., 461t, 466, 466t nonadherent marrow cells in 1,25(OH)2D3 influencing, 652–653 1,25(OH)2D3 causing gene up-regulation in mature, 650f, 652–653 1,25(OH)2D3 deficiency v. mineralization by, 1043–1044 1,25(OH)2D3 directly affecting, 703 1,25(OH)2D3 having biphasic effects on, 650 1,25(OH)2D3 influencing, 1240–1241 1,25(OH)2D3 influencing in vitro, 649–656 1,25(OH)2D3 influencing proliferation of precursors to, 653–654 1,25(OH)2D3 modifying hormone/GF responsiveness of, 654–655 1,25(OH)2D3 “pushing” maturity of, 651–652 1,25(OH)2D3 regulating genes influencing proliferation/ differentiation of, 653–654 1,25(OH)2D3 regulating products of, 654 as 1α,25(OH)2D3 source, 717 1,25(OH)2D3 target proteins with in vitro, 566 1,25(OH)2D3 v. heterogeneity of, 652 1,25(OH)2D3 v. in vitro differentiation pathway/status of, 649–653, 650f 1α,25(OH)2D3 v. gene expression in proliferating, 714–715 osteocalcin promoter remodeling v. phenotype development in, 331f osteoclast differentiation/activation regulated by, 677, 677f PHEX expressed in, 467 site-dependent differential activity of, 599 VSCC/VICC interactions in, 755, 755f Osteocalcin (OC), 176 D deficiency rickets v., 1070 expression stimulated by 1,25(OH)2D3, 486–487 human gene promoter induced by 1,25(OH)2D3, 176–177 human/rat 1,25(OH)2D3/PTH v., 567, 567f mineral properties in KO animals, 479t in mineralization, 479t in mineralization v. 1α,25(OH)2D3, 715–716 1,25(OH)2D3 inducing/glucocorticoids suppressing, 1240 1,25(OH)2D3 regulating matrix protein expression in, 703 regulation as VDR-regulated transcription model, 176–178 VDR target gene polymorphism associated with, 246 Osteoclast differentiation factor (ODF) ligand identification, 670–673 molecular cloning, 670–672, 671f OPG-binding protein as candidate, 670 in osteoclastogenesis hypothesis, 667–668, 668f

1877 Osteoclastogenesis, 665–681 factors in molecular mechanism of, 668–676 RANK as signaling receptor in in vitro, 673–674, 673f RANKL-RANK signal transduction pathways in, 676–678 Osteoclastogenesis inhibitory factor (OCIF), isolation/cloning, 668 Osteoclasts in BMU origination, 503 bone resorption by, 498 molecular mechanisms forming/activating, 674, 676f mouse coculture system recruiting, 666–668, 667f Na/Pi transport in, 466–467 OC recruiting, 487 1,25(OH)2D3 v bone resorption by, 678–680, 679f RANKL/ODF differentiating osteoclast progenitors to, 672 Osteocytes, PHEX expressed in, 467 Osteoid accumulation indices, 1034–1035 accumulation of unmineralized, 1031–1035 in growing skeleton, 1031 histomorphometric measurement of, 954–955, 955f kinetics, 1031 Mlt in understanding accumulation of, 1033–1034, 1033f Osteoid-bone interface, 1031–1032 Osteoid maturation time (Omt), 957 human Mlt v., 1032f, 1033 Osteoid seams, 1032 in adult skeleton, 1031–1034 in bone formation, 1032–1033, 1032f D metabolites’ influence on BMAR v., 657 width measurement of, 955f Osteoid surface per unit of bone surface (OS/BS) osteoid seam life span/FP determining, 1034 O.Th v., 1035, 1035f Osteoid thickness (O.Th) in Mlt, 1033 osteomalacia defined using, 1035–1036 Osteoid volume per unit of bone volume (OV/BV), aging/osteoporosis increasing, 1034 Osteomalacia, 967. See also Tumor-induced osteomalacia Al toxicity in, 980, 1270 biochemical evolution of, 1036–1039 causes of, 925–926, 925t–926t characteristics in CYP27B1-null mice, 703 D3 curing, 999t D deficiency causing, 566 D/intestinal Ca absorption v., 703 D metabolism in, 1039–1040 focal, 958 generalized, 957–958, 957f histological diagnosis of, 957–958 histological evolution/kinetic definition of, 1035–1036 histomorphometric diagnosis of, 955 hypovitaminosis D causing, 154, 156t Looser’s zones in, 971–973, 972f–973f, 981, 982f, 984f mean Mlt/mean O.Th in defining, 1035–1036 mineralization index in diagnosing, 1035–1036 mineralization v. bone properties in, 489–490, 490f Mlt in understanding pathogenesis of, 1033–1034, 1033f in 1α(OH)ase-null mice, 489 osteoid seam width/Mlt in, 957 pathogenesis, 1029–1044 in postgastrectomy bone disease, 1299, 1300 radiology of, 967–990 radiology of D-deficiency, 971–973 rickets defined v., 1031 risk v. race/population, 789, 790t

1878 Osteomalacia (Continued) secondary hyperparathyroidism in, 973, 975f VDR ablation v., 342–343 Osteopenia atrophic rickets characterized by, 1071 D deficiency v., 510 premature infant risk criteria for, 812, 812t renal disease/bone resorption causing, 979 Osteopetrosis bone formation/resorption imbalance causing, 665 in c-src KO mice, 677 Osteophytosis, VDR in, 1144–1145 Osteopontin gene, VDR homodimer-mediated activation v., 179–180 Osteopontin (OPN) mineral properties in KO animals, 479t in mineralization, 479t 1,25(OH)2D3 increasing synthesis of, 567–568 1,25(OH)2D3 regulating matrix protein expression in, 703 1,25(OH)2D3 up-regulating, 487 Osteoporosis, 1101–1114. See also Glucocorticoid-induced osteoporosis in African-Americans/Caucasians, 816 bone formation/resorption imbalance causing, 665 calbindin-D28K v. apoptotic cell death in, 727 D in treatment of established, 1109–1114 D supplementation in treatment of, 1113–1114 D v. mineralization in, 489–490 D3 v., 999t ED-71 v. clinical results using, 1537–1539, 1538f development of, 1525, 1534–1539 planned studies of, 1539 preclinical results using, 1534–1537 safe dosage for, 1537–1538 increased net bone resorption in type I, 1109, 1109f 1,25(OH)2D3 in treating, 1111 1,25(OH)2D safety margin v., 998 PTH/25OHD in type II, 1109 rationale/principles of D treatment of, 1109–1111 renal disease/bone resorption causing, 979 Ro 26–9228 hybrid deltanoid v., 1416–1417, 1417f VDR gene association studies in, 1141–1144 VDR gene polymorphisms predicting, 184 VDR target gene polymorphism associated with, 246 Osteoprotegerin (OPG) bone-resorbing factors regulating, 675–676, 676t clinical trials, 676 discovery, 668 as RANKL decoy receptor, 568 regulating expression of, 674–676 structure of human, 668, 669f in TNF receptor family, 668 Osteosclerosis radiology of, 981, 982f in renal disease/secondary hyperparathyroidism, 977, 977f O.Th. See Osteoid thickness OV/BV. See Osteoid volume per unit of bone volume Ovarian cancer, VDR in, 856 Ovary D in, 856 VDR ablation reducing aromatase activity in, 345 22-Oxa-calcitriol (OCT) adverse reactions to topical, 1533–1534 angiogenesis v., 1574 BFR v., 1528, 1529t

INDEX

22-Oxa-calcitriol (Continued) bone influenced by, 1527–1529, 1529t bone metabolism marker changes v., 1531, 1532f breast cancer cells/tumors v., 1669 cortical bone formation rate v., 1531, 1533f development, 1525–1534 MAR v., 1528, 1529t mechanism of action, 1479–1480 metabolism, 1438 as noncalcemic analog, 1440, 1441t PCa v., 1690 psoriasis vulgaris v., 1531–1534, 1533f PTH v., 1527, 1527f, 1528f PTHrP gene expression/secretion inhibited by, 1576 reduced parathyroid VDR v., 1317, 1318f secondary hyperparathyroidism v., 1527–1531 structure, 1479f, 1525, 1526f synthesis for large-scale production, 1525–1526, 1526f Oxidation-reduction (redox) homeostasis D/cellular response to, 761–768 ROS v., 761–763

P P. See Phosphorus PAD. See Peripheral arterial vascular disease Paget’s disease of bone, calcitonin treating, 1257 Pakistanis, D metabolism in, 790t, 792–793 Pancreas calbindin-D28K in, 725–726, 730t CKD/1,25(OH)2D3 deficiency/resistance in, 1326 Pancreatic disease, metabolic bone disease v., 1301 Pancreatitis, hypocalcemia/tetany in, 1057 Paracellular path, CA absorption by, 421–422 D-dependency of, 412f, 421–422 thermodynamic parameters of, 421 Parathyroid gland. See also Hyperparathyroidism; Hypoparathyroidism; Neonatal severe hyperparathyroidism Ca2+ elevation v., 553f, 554–555, 555f CaR activity v. D effect on, 543 CaR in, 554–556 cell proliferation v. secondary hyperparathyroidism, 544–547 CKD with low 1,25(OH)2D3 or 1,25(OH)2D3/VDR resistance in, 1322–1325, 1323f, 1324f, 1325f D in, 537–547 hyperplasia v. TGFα/EGFR expression in CKD, 1323–1324, 1323f, 1324f low 1,25(OH)2D3 v., 1324–1325, 1325f model of factors regulating, 546f 1,25(OH)2D3 deficiency v. VDR content in, 1317, 1318f as 1,25(OH)2D3 target organ, 539, 540f physiological insights from homozygous null mice, 545–547 PTHrP expression in, 739t removal v. rat 1,25(OH)2D synthesis, 463–464 VDR/CaR interactions in, 556, 556t Parathyroid hormone (PTH) age/glomerular filtration rate v. serum, 1024 age v., 1106–1107 biosynthesis, 537 blacks showing decreased skeletal sensitivity to, 791 bone loss/D deficiency v., 509–510 bone loss v. age-related increase in, 1107, 1108f Ca channel blockers v., 1269 Ca reabsorption in D-replete/deficient rats v., 518–519, 519f Ca v., 552, 553f

INDEX

Parathyroid hormone (Continued) CaR mutations v., 555 CaR v., 537–538 CKD impairing 1α-hydroxylase induction by, 1315–1316 CYP24A1 expression regulated by, 93, 94f CYP24A1 promoter expression/CYP24A1 activity v., 99–100 D bone resorption inhibition v., 570 as D deficiency/insufficiency/repletion index, 1085–1086 D insufficiency increasing serum, 1088–1089 D metabolism influenced by exogenous/stimulated, 1253–1255, 1254t degradation in PTH control, 544 24,25-dihydroxyvitamin D3 suppressing secretion of, 16 diuretics v. serum, 1024 early D deficiency associated with normal, 1069 ECF [Ca2+] demand response mediated by, 776 effector mechanism independence/redundancy, 778 estrogen deficiency v. age-related increase in, 1108–1109, 1108f estrogen influencing serum, 1024 function v. age, 1106 as HDM cause, 1815 hypocalcemia due to abnormal availability of, 1052–1054 hypocalcemia due to resistance to, 1054 hypocalcemia v. defective secretion of, 1053–1054 hypocalcemia v. postsurgery reduction in, 1053 intravenous 1,25(OH)2D3 in renal failure v., 1329–1330, 1330f kidney 1α-hydroxylase activity regulated by, 76–77, 76t, 77t kidney/parathyroid VDR regulated by, 522–523 long-term OCT administration v., 1530–1531, 1530f macrophages unresponsive to, 1382–1383, 1383f mechanisms of age-related increases in, 1106–1107 2MP/2MbisP/2Mpregna v., 1552–1553 noncalcemic D analogs suppressing, 1449–2898 OCT v., 1527, 1527f in 1,25(OH)2D3 feedback loop, 539 1,25(OH)2D3 production increased by, 19–20 1,25(OH)2D synthesis influenced by, 830–831, 830f 25OHD v., 1088, 1088f osteoblast Ca2+ entry stimulated by, 755, 755f osteoblast Na/Pi transport stimulated by, 466, 466t during pregnancy, 839 renal Pi excretion decreased by, 516t renal Pi reabsorption regulated by, 460–461, 461t resistance in rachitic children, 1069 secretion control in dialysis patients, 1822–1823 secretion suppression set-point v. age, 1107 secretion v. cardiovascular disease, 901–902 serum Ca defended by, 1050 serum 25OHD v. serum, 1020–1022, 1021f type I collagen synthesis inhibited by, 704 VDR expression affected by, 208 VDR level 1,25(OH)2D3 regulation of, 540f, 541–542 Parathyroid hormone-related peptide (PTHrP), 737–746 antisense RNA technology inhibiting production of, 743 D compounds inhibiting expression/secretion of, 1576–1577 D therapy v. production of, 741t, 743–746 gene/gene products, 737–738 hypercalcemic cancer patient survival v., 740–742, 740f immunotherapy neutralizing, 742–743 mechanism, 738–739 normal/cancer cells/tissues expressing, 739–740, 739t production regulation, 739–742 SCCs elaborating, 612 stimulators/inhibitors in normal/cancer cells/tissues, 740–742, 741f therapeutic strategies inhibiting production of, 742–746, 742f

1879 Parathyroid hormone-related protein (PTHrP) D metabolism influenced by exogenous/stimulated, 1254t, 1255–1256 in placental Ca transport, 853 synthetic molecules’ actions v., 1256 Parathyroid hyperplasia in D therapy prognosis, 1823, 1824f regression of, 1825–1826 Parkinson’s disease (PD), 765, 1779 D analogs treating, 1785 1,25(OH)2D3 antioxidant activities in, 764t, 766–767 Past medical history (PMH), in approaching metabolic bone disease, 917 PBAF. See Polybromo- and BAF-containing complex PBC. See Primary biliary cirrhosis p450c24. See 24-Hydroxylase PCa. See Prostate cancer PCIT. See Percutaneous calcitriol injection therapy PCR. See Polymerase chain reaction PCTs. See Proximal convoluted tubules PD. See Parkinson’s disease PDDR. See Pseudo-vitamin D-deficiency rickets PEIT. See Percutaneous ethanol injection therapy PENK. See Proenkephalin Peptide growth factors, in keratinocyte growth/differentiation, 614–615 Peptide hormones, VDR expression affected by, 207–208 Percutaneous calcitriol injection therapy (PCIT), nodule hyperplasia v., 1825 Percutaneous ethanol injection therapy (PEIT) nodule hyperplasia v. selective, 1824–1825 severe hyperparathyroidism v. selective, 1825f Perinatal period, 803 D actions, 803–808 recommended D intake in, 808 Periodontium cementum in, 601–602 D influencing, 604–605 dentin in, 601–602 Peripheral arterial vascular disease (PAD), D deficiency associated with, 900 Peripheral quantitative computerized tomography (pQCT), peripheral cortical/trabecular density measured with, 481 Peroxisome proliferator-activated receptor (PPAR), ligands in PCa combination therapy, 1696–1697 Phagocytic cells DBP allele activation v. recruiting, 127 tissue damage causing recruitment/activation of, 126–127 Pharmacology, 995–1010 D3 dosage considerations in, 1002–1006, 1009t D3 nutrition questions in, 1009t DBP in, 1002 pharmacokinetic principles in, 1006–1007 PHEX, in XLH/TIO/ADHR pathogenesis model, 1165–1166 PHEX gene, responsible for XLH, 463t, 467 Phorbol ester PKC activity/CYP24A1 expression v., 99 PTHrP expression/production stimulated by, 741t Phosphate. See also Autosomal dominant hypophosphatemic rickets; Hereditary hypophosphatemic rickets with hypercalciuria; Hyperphosphatemia; Hypophosphatemia; Inorganic phosphate (Pi); Phosphaturia; Tumor-induced osteomalacia; X-linked hypophosphatemia absorption v. 1,25(OH)2D3/VDR binding, 220–221, 221f D3 metabolites modifying muscle uptake of, 887 deficiency-related disorders, 467–470

1880 Phosphate (Continued) common metabolic pathway in, 469–470 deprivation, 1176 disorders in homeostasis of, 1159–1180, 1164t disorders in metabolism of, 1189–1194 disorders in renal transport of, 1162–1175 disorders of decreased, 1176 disorders of increased, 1177–1178 disorders related to altered, 1176–1180 gastrointestinal malabsorption of, 1176 homeostasis FGF-23 in maintaining, 1193–1194 in Npt2-null/Hyp mice, 439–440 homeostasis v. cardiovascular disease, 901 macrophages unresponsive to, 1382–1383, 1383f mineralization v. low plasma, 1042 mouse models with defective transport of, 462–463 1,25(OH)2D3 production v. plasma, 20 1,25(OH)2D3 promoting absorption of, 291 1,25(OH)2D v. renal transport of, 1161–1162, 1162f phosphatonins/minhibins regulating renal transport of, 1163 renal D metabolism regulation v., 463–465, 464f retention from chronic renal failure, 979 Phosphatidylinositol 3-kinase (PI3-K), in 1,25(OH)2D3 differentiation signal propagation, 1638 Phosphatonin/minhibin protein candidate biological activity, 1167–1168, 1167t sFRP–4 as, 1164 Phosphatonin/minhibin proteins FGF-23 as, 1164–1166 MEPE as, 1164, 1166 postulated, 1163 Phosphaturia in 1α(OH)ase-null mice, 438 1,25(OH)2D3 v., 519 Phosphorus (P) absorption v. vitamin D3, 5 abundance/ubiquity in tissues, 1159 aging v., 829, 829t altered phosphate load disorders causing abnormal, 1178–1180 combined mechanisms decreasing serum, 1177 D/25OHD3/1,25(OH)2D3 influencing renal handling of, 518–519 disorders related to transcellular shift of, 1176–1177 kidney 1α-hydroxylase activity regulated by dietary, 77 1,25(OH)2D synthesis influenced by, 828–830, 829f PTH infusion/recovery v., 830, 831f renal reabsorption of, 6 treatment of abnormal serum, 1180 VDR ablation v., 342 VDR expression affected by, 202t, 205 Photobiology, 37–43 history, 37–38 Photosynthesis, 38–41, 38f latitude/season/time v., 39, 40f ultraviolet B-absorbing materials v., 40–41, 41f, 42f Phytoestrogens, 1,25(OH)2D3 synthesis regulated by, 1720–1721, 1720t Pi. See Inorganic phosphate PI3-K. See Phosphatidylinositol 3-kinase PIC. See Preinitiation complex PKA. See Protein kinase A PKC. See Protein kinase C PKCα, 1α,25(OH)2D3 muscle stimulation translocating, 890f, 891 Placenta Ca transport using PTHrP/CaR in, 853–854 calbindin-D9K in, 726, 730t

INDEX

Placenta (Continued) 1,25(OH)2D3 production in, 859 PTHrP expression in, 739t Plants, vitamin D2 synthesized by, 15 Plasma membrane calcium (PMCa) pump, 419–420, 525–527 activity modulators, 420–421 CA transported across basolateral membrane by, 419–420 calbindins stimulating, 420 distribution, 527, 527t in duodenal/ileal Ca absorption, 422–424, 423f epitopes in distal tubular cell basolateral membrane, 525–527, 526f Plasma membrane sodium-calcium exchanger, CA transported across basolateral membrane by, 421 Plasma membranes Ca2+ response to 1,25(OH)2D3 initiated by, 754–755 resting potential “left-shifting” in, 754–755, 755f PMCa pump. See Plasma membrane calcium pump PMH. See Past medical history Pneumonia hypercalcemia/D hypersensitivity in pneumocystis carinii, 1361 resistance influenced by D3, 999t Polybromo- and BAF-containing complex (PBAF) in 1,25(OH)2D3-liganded VDR-RXR heterodimer transactivation, 238–240, 239f in VDR transcriptional function, 268 Polymerase chain reaction (PCR), in HVDRR studies, 1218 Polynesians bone mass in, 794 D metabolism in, 794 Postgastrectomy bone disease, 1299–1300 biochemistry, 1300 bone turnover in, 1299 clinical features, 1299–1300 D metabolism of, 1299 management, 1300 osteomalacia in, 1299 pattern of, 1299 PPAR. See Peroxisome proliferator-activated receptor pQCT. See Peripheral quantitative computerized tomography PR. See Production rate Prednisone, Ca absorption v., 1244–1245, 1245f Pregnancy bone mineral content/density during, 840–841 Ca homeostasis v., 204 D/Ca metabolic adaptations during, 839–841 D metabolism in, 839–843 last trimester, 803–804 D deficiency in, 803–804 D supplementation in, 804, 808 malnutrition in, 803 1,25(OH)2D v. Ca absorption during, 839–840, 840f Pregnane X receptor (PXR) accumulated toxins v., 865–866 VDR ligand binding/heterodimerization/transactivation domains compared with, 233–235, 234f Preinitiation complex (PIC) in comodulator activity integrated model, 300, 300f VDR contacts with, 291–292 Previtamin D3, vitamin D3 produced from, 38, 38f, 39f Primary biliary cirrhosis (PBC) bone disorders associated with, 1304–1306 clinical features of, 1304 D metabolism in, 1305, 1305f management of, 1305–1306 Primates. See New World primates PRL. See Prolactin

INDEX

Production rate (PR), young/elderly men’s 1,25(OH)2D, 827, 827f Proenkephalin (PENK), 1,25(OH)2D3/other osteotropic hormones regulating, 655 Progesterone, D metabolism influenced by, 1254t, 1261–1262 Prolactin (PRL), D metabolism influenced by, 1254t, 1258 Promyelocyticleukemias (APLs), RAR α fusion proteins marking, 1732 Prooxidant, 1,25(OH)2D3 as, 763–765, 764t Prostaglandins, D metabolism influenced by, 1254t, 1262–1263 Prostate D endocrine/autocrine systems in, 1608 as D target, 1682–1683 1α-hydroxylase in, 1607 1,25(OH)2D as autocrine hormone in, 1607–1610 1,25(OH)2D3 synthesized from 25OHD3 in, 1607, 1608f 1,25(OH)2D3 v. virally transformed cells of, 1684 VDR in, 1682–1683 Prostate cancer cells D analogs v. growth of, 1451 enzymes in D metabolite response by, 1685–1687 growth factor actions in regulating, 1692–1693 mechanisms of D-mediated growth inhibition v., 1690–1695 1,25(OH)2D3 arresting growth of, 1690–1691 1,25(OH)2D3/deltanoid-induced differentiation modeled in, 1637t 1,25(OH)2D3 inhibiting invasion/metastasis by, 15 1,25(OH)2D3 v. apoptosis in, 1691 1,25(OH)2D3 v. differentiation in, 1691–1692 1,25(OH)2D3 v. primary cultured, 1683–1684, 1684f 1,25(OH)2D3 v. proliferation of human, 1683, 1683f resistance to 1,25(OH)2D3 antiproliferative effect in, 1684–1685 Prostate cancer (PCa) animal models/in vivo studies of, 1688–1689 Ca/dairy products v., 1627–1629 circulating D v., 1625–1626 clinical trials evaluating 1,25(OH)2D3/analogs, 1698–1699 combination therapy v., 1695–1698 D analogs v., 1689–1690 D3 protecting against, 999t D/sunlight/natural history v., 1599–1611 D v., 1679–1700 dietary D v., 1625 D’s antiproliferative/prodifferentation actions in, 1682 EB1089 v., 1499 epidemiology, 1680–1681 epidemiology v. age/race/place, 1599–1600 etiology/treatment/hormonal factors, 1679–1680 genetic factors v., 1681–1682 high dose intermittent 1,25(OH)2D3 v. androgen-dependent, 1745 high incidence of incidental, 1600 mortality/incidence, 1624–1625 1,25(OH)2D3 inhibiting in vitro growth of, 1683–1685 1,25(OH)2D3 v., 1742 25OHD v. risk of, 1000f, 1002 risk epidemiology v. D/Ca, 1617–1629 risk factors v. vitamin D hypothesis, 1601, 1601t risk v. D, 1600–1601 sunlight v., 1617, 1625 VDR gene polymorphisms/1α-hydroxylase v., 1626–1627 VDR genetic polymorphisms associated with, 1261, 1681–1682 VDR polymorphisms v., 1139t, 1145 vitamin D hypothesis, 1600–1602, 1610–1611, 1611f dietary D studies, 1606 observational studies, 1602–1606 seroepidemiological studies, 1602–1604 sunlight exposure studies, 1605–1606 VDR polymorphism studies, 1604–1605

1881 Prostate cells BXL–353 v. in vivo growth of, 1837–1840, 1839f VDR expression in, 1834–1836 Prostate specific antigen (PSA) doubling time v. 1,25(OH)2D3 PCa treatment, 1698, 1698t in early PCa diagnosis, 1679 Protein kinase A (PKA) PKC interaction v. VDR regulation by, 209–210 VDR expression affected by, 209 Protein kinase C (PKC) activity v. 1α,25(OH)2D3 muscle stimulation, 890f, 891 D metabolites modulating matrix vesicle, 586 glucocorticoid/CYP24A1 expression v., 99 hVDR phosphorylation v. nuclear localization/DNA binding, 231f, 232–233 keratinocyte differentiation mechanism unclear in, 618–619 in keratinocyte growth/differentiation, 618–619, 620f in 1,25(OH)2D3 differentiation signal propagation, 1637–1638 in 1α,25(OH)2D3-mediated rapid responses, 386t phorbol ester/CYP24A1 expression v., 99 PKA interaction v. VDR regulation by, 209–210 VDR expression affected by, 209 Protein kinases, 1,25(OH)2D3/mitogen-activated, 97–98 Protein sulfhydryl (-SH) groups, Ca entry v., 416 Protusio acetabulae, 972f–973f, 973 Provitamin D3, photolysis, 38, 39f Proximal convoluted tubules (PCTs), in 25OHD3/1,25(OH)2D3 conversion, 153 PSA. See Prostate specific antigen Pseudo-vitamin D-deficiency rickets (PDDR), 1197–1204 biochemical findings in, 1198–1199 clinical manifestations, 1197–1198 CYP27B1 gene causing, 109 D/1,25(OH)2D3 v. 1,25(OH)2D3 in, 1199, 1199f defective 25OHD3/1,25(OH)2D3 conversion in, 1197, 1198f founder effect in Northeastern Quebec, 1200 genetic studies, 1199–1200 HVDRR v., 1208, 1209t 1α-hydroxylase gene defects in, 1200–1201, 1201f hypocalcemia in, 1056 in 1α(OH)ase-null mice, 438–439, 489 1,25(OH)2D in, 16 1,25(OH)2D replacement therapy v., 110 1,α(OH)D replacement therapy v., 111 1,25(OH)2D3 therapy v. rickets phenotype in, 1202, 1203f placenta studies, 1199, 1200f recognition/semantics, 1197 tooth enamel dysplasia in, 1202, 1203f treatment, 1201–1203 Pseudohypoparathyroidism, hypocalcemia v., 1054 Psoriasis, 1791–1801 calcipotriol/betamethasone dipropionate treating, 1500 calcipotriol treating scalp, 1785–1786 clinical use of 1,25(OH)2D3/analogs v., 1784–1787 D/analog biological effects in, 1781, 1781f D analogs treating, 1450 D analogs v., 744 efficacy/safety of D analogs treating, 1504, 1504t MC–903 v., 1543 OCT development for, 1525–1534 OCT v., 1531–1534, 1533f pathogenesis of, 1791–1792 safe VDR ligand treatments for, 643 serum 1,25(OH)2D3/25OHD in, 1784 skin lesion histology in, 1791–1792 specific therapies for, 1785–1787

1882

INDEX

Psoriasis (Continued) Théramex hybrid deltanoid v., 1416, 1416f therapies increasing topical D analog efficacy in, 1786–1787 topical 1,25(OH)2D3/analogs v., 1784–1785 topical 1,25(OH)2D3 v., 1785, 1785f treatment response v. VDR genotypes, 1784 VDR ligand treatment v., 637t, 640 VDR polymorphisms v., 1146 PTH. See Parathyroid hormone PTH gene, 538–539 abnormalities v. hypocalcemia, 1053 expression v. VDREs, 539 1,25(OH)2D3 regulating, 539 calreticulin v., 543–544 in vivo studies of, 539, 540f promoter sequences, 538–539 regulation, 538 structure, 538 PTHrP. See Parathyroid hormone-related peptide; Parathyroid hormone-related protein PTHrP gene, 737–738 organization/structure-function relationships, 737–738, 738f Puberty, Ca absorption efficiency/utilization during, 815–816, 815f, 816f Purdah, rickets incidence v., 1067 PXR. See Pregnane X receptor

Q qBEI. See Quantitative backscattered electron imaging QCT. See Quantitative computed tomography QOD schedule. See Every-other-day schedule Quantitative backscattered electron imaging (qBEI) localized mineral measurement with, 483 old bone mineralization measured with, 489 Quantitative computed tomography (QCT), BMD assessment with, 922–924 Québecois, PDDR in, 1199–1200

R RA. See Retinoic acid; Rheumatoid arthritis 9-cis-RA, in leukemia combination therapy, 1733 RAA axis. See Renin-angiotensin-aldosterone axis Race. See also Blacks; Whites; specific nationalities D metabolism influenced by, 790–795, 790t as PCa risk factor, 1601, 1601t rickets/osteomalacia risk v., 789, 790t Radiation, hypocalcemia v., 1053 Radioimmunoassays (RIAs) 125I-labeled tracers coupled with, 932 improved antibody for, 946, 946t metabolites assayed as 1,25(OH)2D by, 946–947, 947f 25OHD assay consistency v., 1023 1,25(OH)2D measured by RRA v., 946, 946f 25OHD/1,25(OH)2D quantitated with, 932 Radiology imaging, 986–990 nuclear medicine in, 986–988 plain radiographs in, 986 rickets/osteomalacia v., 967–990 Radioreceptor assays (RRAs), for 1,25(OH)2D, 932, 942–947, 943t Raman spectroscopy, tissue mineralization quantified with, 481, 482–483, 482f

RANK. See Receptor activator of NFκB RANKL. See Receptor activator of NFκB ligand RANKL/ODF bone resorption signal transduced by, 672–673 cloning/structure, 670–672, 671f osteoclast progenitors/osteoclasts differentiated by, 672 receptor identified as RANK, 673–674 as sole in vivo RANKL receptor, 674 RANKL-RANK signaling, alternations v. skeletal phenotype, 678, 679t Rapid response, 381 6-s-cis-shaped 1α,25(OH)2D3 optimal agonist for, 391, 398f case studies, 392–397, 397t 1α,25(OH)2D3-mediated, 381–400, 386t molecular tools for study of, 392, 393f structure-function evaluation of, 392–397 in ROS 17/2.8 cells, 392–394, 394f schematic model, 399–400, 399f steroid hormones generating, 386 VSMC v. 1α,25(OH)2D3-mediated, 395–397, 396f RAREs. See Retinoic acid response elements RAS. See Renin-angiotensin system Rat calvaria cells biphasic 1,25(OH)2D3/osteoblast relationship confirmed in, 651 1,25(OH)2D3/dexamethasone stimulating adipogenesis in, 655–656 1,25(OH)2D3/osteoblast relationship modeled by, 649–650, 650f Rat osteoblastic osteosarcoma (ROS), 1,25(OH)2D3 inhibiting collagen synthesis in, 704 Rats. See also Genetic hypercalciuric stone-forming rats D deficiency v. fertility of male/female, 855 vitamin D3 discriminated against by, 29 RDA. See Recommended dietary allowance Reactive oxygen species (ROS), 761 generation/degradation of, 762 Janus face of, 761 in neurodegenerative diseases, 1779–1780 redox homeostasis v., 761–763 as signaling cascade messengers, 762–763 REBiP. See Response element binding protein Receptor activator of NFκB ligand (RANKL), 568 bone-resorbing factors regulating, 675–676, 676t mRNA/serum Ca v. 1,25(OH)2D3, 680, 680f 1α,25(OH)2D3 inducing expression of, 713 osteoblasts regulating osteoclasts through inducing, 568 RANK as sole in vivo receptor for, 674 regulating expression of, 674–676 Receptor activator of NFκB (RANK) RANKL/ODF receptor identified as, 673–674 as RANKL/ODF signaling receptor in in vitro osteoclastogenesis, 673–674, 673f Receptor interacting domain (RID), NCoA62/SKIP, 296, 297f Recommended dietary allowance (RDA), 1357 calls for increasing, 1357 Redox homeostasis. See Oxidation-reduction homeostasis Redox state, cellular, 761–762 Regulator of G protein signaling (RGS)–2, 1,25(OH)2D3 inversely modulating, 654 Regulatory T cells in autoimmunity, 1754 1,25(OH)2D3 autoimmunity v., 1755 Renal cell carcinoma, VDR polymorphisms v., 1146 Renal endocytosis, 25OHD3, 153–159 cell biology of, 156–157 molecular biology of, 157–159 physiology of, 153–156

1883

INDEX

Renal failure D analog action/tissue specificity v. chronic, 1826–1827 D in, 1313–1333 future D analog roles in chronic, 1826–1827 hyperparathyroidism v. D in, 1821 intravenous 1,25(OH)2D therapy v., 1822–1823 mechanisms of 1,25(OH)2D resistance in chronic, 1822, 1822f OCT in dog model of chronic, 1527 1,25(OH)2D resistance causing secondary hyperparathyroidism in, 1821–1823 1,25(OH)2D resistance in chronic, 1821–1822 parathyroid hyperplasia in chronic, 1823, 1824f secondary hyperparathyroidism v., 1821–1827 Renal osteodystrophy, 974 aluminum toxicity in, 979–981 metastatic calcification in, 979 1,25(OH)2D3/VDR action in, 1322–1326 periosteal new bone formation in, 979 radiology of, 974–981, 975f Renin 19-nor Gemini suppressing expression of, 1518, 1518t D suppression v. other mechanisms regulating, 877 Gemini analogs inhibiting, 1514–1518 RAS cascade rate limited by, 871 synthesis/secretion control, 872–873 Renin-angiotensin-aldosterone (RAA) axis, D regulating, 901 Renin-angiotensin system (RAS), 871–879 CKD/1,25(OH)2D3 deficiency/resistance in, 1326 components in hypertension treatment, 1514 D endocrine system interaction with, 877–878, 878f Gemini compounds inhibiting, 1516–1518, 1518t 1,25(OH)2D3 as negative endocrine regulator of, 875–878 animal studies evaluating, 875–877 hypothesis of, 875 physiological implications of, 877–878 overview, 871–872, 871f Renin gene expression regulation, 873 expression suppressed by D, 877 Reproduction active Ca absorption during, 440–445 VDR ablation affecting, 345–346 Reproductive organs, D’s role in, 851–860 Resistance humans having New World primate-like, 355–356 hypocalcemia due to hereditary, 1056 index case in humans, 355 in New World primates, 352 biochemical nature of, 354–355 VDRE-BP-2 causing, 355 Response element binding protein (REBiP) binding in cis, 356, 356f D-resistant human patient over-expressing, 355 human, 356–357 overexpression v. HVDRR patient 1,25(OH)2D3 resistance, 356–357, 356f Restriction fragment length polymorphisms (RFLPs) HVDRR-related VDR gene mutation v., 1220 in prenatal VDR gene mutation diagnosis, 1228 VDR gene, 1124, 1125f Retinoblastoma protein, deltanoid-induced G1 block controlled by, 1649 Retinoic acid (RA), VDR expression affected by, 202t, 206–207 Retinoic acid response elements (RAREs), human VDR promoter, 200

Retinoid X Receptor (RXR) import/export receptors interacting with, 368–371 intranuclear trafficking of, 374–376 as NR/VDR heterodimeric partner, 291, 292f nucleocytoplasmic trafficking regulation, 371–374 “piggyback” nuclear import of, 369, 371f putative NLSs in, 369, 370f shuttling v. transcription, 371 subcellular trafficking, 363–376 in VDR/DNA binding, 178–180, 179f, 220, 220f Retinoids in PCa combination therapy, 1696 VDR expression affected by, 206–207 RFLPs. See Restriction fragment length polymorphisms RGS-2. See Regulator of G protein signaling–2 Rhabdomyolysis, hyperphosphatemia in, 1178 Rheumatoid arthritis (RA) 1,25(OH)2D accumulation in, 1389–1390 subperiosteal erosions simulating, 975f, 976 VDR ligand treatment v., 637, 637t VDR polymorphisms v., 1147 RIAs. See Radioimmunoassays Rickets, 967. See also Osteopenia; specific types of rickets Al toxicity in, 980 biochemical abnormalities in nutritional, 1069–1070 biochemical evolution of, 1036–1039 bone turnover markers elevated in nutritional, 1069–1070 Ca2+ availability v. mineralization in, 579 Ca v. D deficiency in, 1077, 1077f cartilage prehypertrophic/hypertrophic zones increased in, 579, 580f causes of, 925–926, 925t–926t Chinese incidence of infantile, 793 in cities, 967, 1065 classical features of, 1067–1068, 1068f clinical presentation of nutritional, 1067–1069 D3 curing, 999t D deficiency and children’s nutritional, 1065–1077 D deficiency causing, 566 D deficiency impairing bone resorption in, 777, 777f D/intestinal Ca absorption v., 703 D metabolism in, 1039–1040 D single dose therapy v., 1072 deltanoids v. VDR mutants associated with, 1409, 1411f dental phenotype of, 602, 602f dietary Ca deficiency causing, 1074–1075, 1075f epidemiology of D deficiency/nutritional, 1066–1067 in exclusively breast-fed children, 777 geographical distribution of colon cancer and, 866 high-Ca diet preventing KO mouse, 429 history, 4, 37–38, 967, 1065–1066 HVOii/HVOiii v. infantile, 1038 in Los Angeles Zoo New World primates, 352–353, 353f mineralization in healing, 1071–1072, 1071f oncogenic, 983–984 pathogenesis, 1029–1044, 1076–1077, 1077f pathophysiological progression of D-deficiency, 1070 PDDR v. nutritional, 1197 premature infants v. nutritional, 812, 812t prevention of nutritional, 1072–1074 radiology of, 967–990, 981, 982f, 1070–1072 radiology of D-deficiency, 968–971, 969f, 970f, 971f restricted definition of, 1031 risk v. race/population, 789, 790t sunlight v. incidence of, 246 symptoms caused by defective CYP27A1, 61–62

1884

INDEX

Rickets (Continued) treatment of nutritional, 1072 ultraviolet/sunlight v., 4–5 VDR ablation v., 342–343 in VDR-null mice v. mineral homeostasis, 433 vitamin D3 activity v. vitamin D-dependency, 7 RID. See Receptor interacting domain Ro-26–9228 mechanism of action, 1480 osteoporosis v., 1451 potency in intestinal cells, 1480, 1480f structure, 1479f target tissue gene expression v., 1480, 1481f ROS. See Rat osteoblastic osteosarcoma; Reactive oxygen species RRAs. See Radioreceptor assays RUNX proteins chromatin remodeling due to, 332 gene expression suppressed by, 333 intranuclear targeting signal v. function of, 335 localization, 334 1,25(OH)2D3 regulation dependent on osteoblast maturity, 651 promoter element/co-regulatory protein interactions by, 332 promoter regulatory complex organization determined by, 333

S SAGE. See Serial analysis of gene expression SAR. See Structure-activity relationship Sarcoidosis active D metabolite produced extrarenally in, 1380 disordered Ca balance pathophysiology in, 1381–1387 dysregulated overproduction of 1,25(OH)2D in, 1381 endogenous D intoxication associated with, 1379–1380 extrarenal D metabolite overproduction v., 1390, 1390t hypercalcemia/D hypersensitivity in, 1359–1360 VDR polymorphisms v., 1147 Saudi Arabians bone mass in, 794 D metabolism in, 794 low serum 25OHD in, 1026 sunlight exposure v. rickets in, 1066–1067 winter hypovitaminosis D in adult, 1091 Scandinavians hypovitaminosis D in adult, 1091, 1091f hypovitaminosis D in healthy elderly, 1092, 1092f hypovitaminosis D in institutionalized elderly, 1093, 1093f Scanning small-angle X-ray scattering (scanning-SAXS), mineral particle thickness/alignment from, 481–482 SCCs. See Squamous cell carcinomas Scleroderma D analog therapy v., 1787 1,25(OH)2D3 treating, 1758–1759, 1758f, 1758t, 1759t SCP. See Start codon polymorphism Scurvy, nutrition v., 3 Secondary hyperparathyroidism bone turnover due to, 509–510 CaR expression v. primary/uremic, 556 in chronic renal failure v. 1,25(OH)2D3 analogs, 1331–1332, 1332f CYP27B1 gene mutation v., 109 CYP27B1-null mice developing, 703 D insufficiency v. senile, 1088, 1088f gastrointestinal diseases associated with, 1297 normal serum 25OHD v., 1020–1022 OCT ameliorating osteopathy in, 1531, 1533f

Secondary hyperparathyroidism (Continued) OCT development for, 1525–1534 OCT v., 1527–1531 clinical results of, 1529–1531 preclinical results of, 1527–1529 1α(OH)ase-null mice demonstrating severe, 438 1,25(OH)2D3 deficiency/CKD v., 1324–1325, 1325f parathyroid cell proliferation v., 544–547 parathyroid gland hyperplasia in, 1823 radiology of D deficiency, 973–974, 975f, 980f renal disease stimulating, 974–976 renal failure and, 1821–1827 treatment in chronic renal failure before dialysis, 1327–1328 during hemodialysis, 1328–1331, 1330f Secreted frizzle-related protein (sFRP)–4, as phosphatonin/ minhibin, 1164 Selective estrogen receptor modulators (SERMs), in VDR ligand tissue selectivity, 270–271 Selective progesterone receptor modulators (SPRMs), in VDR ligand tissue selectivity, 270–271 Seocalcitol. See EB1089 Sepsis, hypocalcemia in acute, 1057 Serial analysis of gene expression (SAGE), TIO tumors v., 1192 SERMs. See Selective estrogen receptor modulators Sex steroids D metabolism influenced by, 1254t, 1259–1262 1,25(OH)2D synthesis influenced by, 828 sFRP–4. See Secreted frizzle-related protein–4 SH. See Social history Shwachman-Diamond syndrome, rickets v. differential diagnoses for, 984–985, 987f Side-chain cleavage, vitamin D2/D3, 25–26 Side-chain oxidation vitamin D2, 25 vitamin D3, 17f Simian bone disease, 351–352 D deficiency v., 351–352 New World primates susceptible to, 351–352 Single dose therapy, 1066 hypercalcemia v., 1072, 1073 patient compliance problem avoided by, 1072 Skeletal genes controlling in vivo expression of, 327–328 intranuclear organization of D-mediated regulatory machinery for, 327–336 Skeletal muscle, as D target tissue, 883–894 Skin D analog actions in normal/psoriatic, 1781–1784 D system in normal/psoriatic, 1792–1793 dystrophic mineral deposits in, 478t 1,25(OH)2D3/analog immunoregulatory properties in, 1519 psoriasis/other diseases of, 1791–1801 rejection inhibited by 1,25(OH)2D3/analogs, 641t structure/function deteriorating with age, 823 VDR in human, 1792f, 1793 Skin cancer D analog therapy v., 1787 D photosynthesis v. recommendations for reducing, 1087 vitamin D photosynthesis v., 42–43 Skin lesions in children v. 1,25(OH)2D3 ointment, 1786 in HIV patients v. oral 1,25(OH)2D3, 1786, 1786f Skull, subperiosteal erosions causing “pepper pot,” 975f, 977 SLE. See Systemic lupus erythematosus SMRT, NR co-repressor, 298–299

1885

INDEX

Social history (SH) in approaching metabolic bone disease, 917–918 vegetarianism in, 917t, 918 vertebral crush deformity correction guided by, 918, 918f Soda, v. Ca intake by adolescents, 817 Sodium chloride (NaCl), CaR inhibiting MTAL reabsorption of, 557 Sodium (Na), excretion v. 1,25(OH)2D3 in TPTX dogs, 518 Software, VDRE screening, 322 Sp1 transcription factor, in 1,25(OH)2D3-induced differentiation, 1640 Spanish, hypovitaminosis D in elderly, 1092–1093 Spermatogenesis, SRC-2 deletion resulting in, 295 Spliceosome, hnRNPs associated with, 358–359 SPRMs. See Selective progesterone receptor modulators Squamous cell carcinomas (SCCs), 1,25(OH)2D produced by keratinocytes from, 612 SRC-1 See Steroid receptor coactivator 1 Stanniocalcin hypercalcemia v., 462 proximal tubular Pi transport regulated by, 461t Star volume, in bone structure assessment, 961–962, 962f Start codon polymorphism (SCP), 1124 Steroid hormones calbindin-D28K/D in producing, 726 calbindin-D9K regulated by, 729–730 calbindin-D28K regulated by, 728–729 CYP24A1 transcription regulated by, 97 D structure v., 381, 382f kidney 1α-hydroxylase activity regulated by, 77–78 nongenomic actions reported for, 98–99 VDR expression affected by, 205–206 Steroid precursors, in deltanoids, 1412–1416, 1413f, 1414f, 1415f Steroid receptor coactivator 1 (SRC-1), VDR stabilization v. interaction with, 1472–1473, 1472f Sterols. See also Steroid hormones equilibrium of bound/free, 124 interpretation/relevance of measurements of antirachitic, 947–949 plasma proteins v. transport/function of, 124–125 Stomach, Ca absorption v. aging, 1105–1106 Store-operated Ca2+ (SOC) anti-INAD antibody v. 1α,25(OH)2D3-dependent influx from, 893, 893f INAD-based signaling complexes in 1α,25(OH)2D3-modulated influx from, 893, 894f muscle influx mediated by TRPC3 proteins/VDR, 892–893, 892f Stosstherapie. See Single dose therapy Structure-activity relationship (SAR), in deltanoid design, 1412 Strut analysis, in bone structure assessment, 961, 962f Sunlight alcoholics lacking exposure to, 1266 blood pressure v. D and, 873–874, 874f colon cancer death rate v., 1571–1572 colon cancer v., 1709–1710 D from UVB component of, 1006–1007 D intake from food v., 995, 996t–997t D nutrition/acquired bone disease v., 1297–1298 D supplementation required by high latitude, 784–785 deprivation determining D insufficiency, 1086–1087 diabetes incidence v., 1766 exposure increasing D3 status, 1009t exposure v. age, 1102, 1109 exposure v. colorectal cancer, 1618 fatal breast/prostate cancer v., 1572 glass v. D synthesis induction by, 823

Sunlight (Continued) HDM v., 1813–1814 high rickets incidence despite, 1066–1067 25OHD v. excessive, 1356 PCa v., 1599–1611, 1617, 1625 PCa v. exposure to, 1680–1681 rickets/colon cancer incidence v., 246 rickets v., 777 vitamin D photosynthesis regulated by, 38–39 Sunscreen, vitamin D photosynthesis v., 40–41, 41f Superagonists, 1475–1476, 1476f differential VDR activation by, 1475–1478, 1476f 20-epi, 1477–1478 20-natural, 1476–1477 Suppressor T cells, 1,25(OH)2D3 autoimmunity v., 1755 Swiss, hypovitaminosis D in adult, 1091 Systemic lupus erythematosus (SLE), VDR ligand treatment v., 637t, 639–640

T T cell response, VDR KO mice showing abnormal, 246 T cells. See also specific types of T cells in acute allograft rejection, 1519 immunosuppressive therapy v. pathogenic, 631–632, 632t VDR ligand immunoregulation of, 635–636, 635t VDR ligands enhancing regulatory, 636 T-cells, in psoriasis, 1791–1792 T lymphocytes, antigens recognized by, 631 Tacalcitol. See 1α,24R-Dihydroxyvitamin D3 Taq polymorphisms, 1131t–1132t, 1135–1137 BMD v., 1142–1143 Taxanes, 1,25(OH)2D3 in combination with, 1745–1746 TC. See Tumoral calcinosis TEI 9647, 1481–1482, 1481f Testis calbindin-D28K in, 726 D in, 855–856 1α,25(OH)2D3 VDRnuc in, 385t SRC-2 deletion causing defects in, 295 VDR ablation reducing aromatase activity in, 345 Testosterone, D metabolism influenced by, 1254t, 1261 Tetany in hypocalcemia, 920, 1049 in PDDR, 1197 TFs. See Transcription factors TGF. See Transforming growth factor TGFα, keratinocytes producing, 614 Theophylline, D metabolism influenced by, 1255t, 1272 Thiazide diuretics D metabolism influenced by, 1255t, 1268–1269 as IH therapeutics, 1351 Thyroid C cells, 687 D controlling CT gene in, 687–697 embryonic development of, 687–688 neoplasia of, 688 origin/function of, 687–688 Thyroid gland C cells in normal adult, 688 PTHrP expression in, 739t Thyroid hormone, D metabolism influenced by, 1254t, 1262 Thyroid receptor-associated proteins (TRAP) coactivator complex. See Mediator-D coactivator complex Thyroparathyroidectomized (TPTX) dogs, 1,25(OH)2D3 v. phosphate/Ca/Na excretion in, 518

1886 Thyroparathyroidectomized (TPTX) rats 24-hydroxylase activity v. D toxicity in, 1366–1367, 1367t 1,25(OH)2D3 v. Ca excretion in, 518, 518f TIO. See Tumor-induced osteomalacia TLRs. See Toll-like receptors TNF. See Tumor necrosis factor TNF receptor-associated factor (TRAF) family proteins, in osteoclastogenesis, 676–677 TNF receptor family. See Tumor necrosis factor receptor family TNFα, D metabolism influenced by, 1254t, 1263 Toddlers, Ca absorption in, 815 Toll-like receptors (TLRs) LPS-induced 1,25(OH)2D production supported by, 1385 pathogens recognized by, 631 Tooth crown influenced by D, 604 D in formation/mineralization of, 601–602 dentin/cementum in root of, 601–602 eruption delayed by rickets, 1068 eruption delayed in PDDR, 1198 Tooth enamel hypoplasia in HVDRR, 1208 1,25(OH)2D3 partially correcting PDDR, 1202, 1203f Total parenteral nutrition (TPN) bone disease caused by Al in, 1270, 1303 bone disorders associated with, 1303 clinical features of, 1303 management, 1303 Toxicity, 26–27. See also Intoxication adipose tissue loss v., 1007, 1010 arterial dystrophic calcification induced by, 478 biological markers for monitoring, 1009t Ca/D supplementation v., 1059 cardiovascular, 899, 904–905 D2/D3, 26–27, 1356–1357 D effects on bone v. hypercalcemia in, 510 D/25OHD, 1356–1357 D osteolytic response/hypercalcemic effects in, 569 EAE v. 1,25(OH)2D3, 1784 factors affecting, 27 high dose intermittent 1,25(OH)2D3, 1743, 1743t increased serum phosphorus associated with, 1177–1178 mechanisms of D, 1362–1368 1,25(OH)2D3, 1357 1,25(OH)2D3 concentration v., 1008 overview, 26 PCa treatment v. 1,25(OH)2D3, 1689–1690 pharmacological issues of safety and, 1007–1010 psoriasis treatment v., 1791 radiology of, 986, 988f synthetic analog, 1357–1359 thresholds in adults/infants, 1357 topically applied/systemically administered D compound, 1450 TPN. See Total parenteral nutrition TPTX dogs. See Thyroparathyroidectomized dogs Trabecular bone pattern factor, in bone structure assessment, 962 TRAF family proteins. See TNF receptor-associated factor family proteins Transactivation coregulators in VDR/RXR, 180–181 HVDRR cell lines suppressing RXR-VDR-mediated, 354f, 355–356 1,25(OH)2D3-liganded VDR-RXR, 237–243, 239f REBiP squelching hormone-directed, 358, 358f RXR’s direct involvement in, 181 squelching VDR-directed VDRE-reporter-driven, 355, 355f

INDEX

Transcaltachia cell-surface receptor in analog-stimulated, 1461 structure-function summary analysis, 397, 397t Transcription apparatus, REBiP in, 358, 358f Transcription factors (TFs) intranuclear pathways directing, 335–336 in 1,25(OH)2D3-induced differentiation, 1639–1640 osteoblast differentiation status varying, 653 as regulatory component scaffolding, 332–333 Transcription start sites (TSSs), 1,25(OH)2D3 target gene VDREs near, 314, 315f, 316t Transforming growth factor (TGF), in 1,25(OH)2D3 actions on prostate cells, 1692–1693 Transient Receptor Potential (TRP) channel superfamily, Ca2+ influx v., 430 muscle, 892–893, 892f Transient Receptor Potential Vanilloid (TRPV) family, Ca influx v. TRPV5/TRPV6 members of, 430 Translation, hnRNPs as ribosome recognition proteins in, 359 Transport receptors in nucleocytoplasmic VDR/RXR trafficking, 365–376 Ran-GTPase regulating, 365–366, 367f TRAP coactivator complex. See Mediator-D coactivator complex TRP channel superfamily. See Transient Receptor Potential channel superfamily TRPV5. See ECaC1 TRPV6. See ECaC2 TRPV family. See Transient Receptor Potential Vanilloid family TSSs. See Transcription start sites Tuberculosis D3 preventing, 999t extrarenal D metabolite overproduction v., 1390–1391, 1390t hypercalcemia/D hypersensitivity in, 1360 VDR polymorphisms v., 1148 Tumor cells D influences on, 1577–1582 D influencing, 1577–1582 D resistance/metabolism in, 1583–1584 1,25(OH)2D3 stimulating proliferation of, 1574t, 1584–1585, 1585f Tumor-induced osteomalacia (TIO), 463t, 468, 983–984 benign/malignant tumors in, 983, 986f as disorder of phosphate metabolism, 1190–1192 genes overexpressed in, 1192 phosphatonin secretion resulting in, 1190 tumors eluding detection in, 983–984, 989f Tumor necrosis factor (TNF) causing apoptosis v. calbindin-D28K/osteoblasts, 725, 725f type I collagen synthesis inhibited by, 704 Tumor necrosis factor (TNF) receptor family nomenclature, 674, 675f OPG in, 668 Tumor suppressor genes in cell cycle/apoptosis control, 1577–1580 1,25(OH)2D3/EB1089 regulating breast cancer cell, 1664–1665 1,25(OH)2D3 regulating expression of, 1579 Tumoral calcinosis (TC), hyperphosphatemic, 1175–1176 Tumors D/analogs differentially influencing canine/human, 1577 D preventing colon, 1709–1710 1,25(OH)2D3/1,25(OH)2D3 analogs v., 1741 1,25(OH)2D3 stimulating development of, 1574t, 1584–1585, 1585f prognosis v. VDR expression in breast cancer, 1668 VDR in, 1572t VDR-RXR heterodimer-activating ligands v., 241

1887

INDEX

Turks serum 25OHD/PTH in dark-skinned Dutch, 794 sunlight exposure v. rickets in, 1066–1067 sunlight exposure v. serum 25OHD in, 794 Type II nitric oxide synthase (NOS II), 1,25(OH)2D3 v., 1783

U UL. See Upper limit Ultraviolet (UV) light age/serum D v., 823, 824f PCa mortality v., 1601–1602, 1602f, 1603f PCa v., 1605–1606 rickets v., 565 serum 25OHD directly related to, 825 Upper limit (UL), conservative safety margin of, 1008 Uremia, 1,25(OH)2D3/VDR-mediated transcription v., 1320–1321, 1321f Uterus calbindin-D9K/D28K in, 726, 730t D in, 856–857 PTHrP expression in, 739t 3’UTR polymorphisms, 1131t–1132t, 1135–1137 BAt/baT haplotype expression in, 1131t–1132t, 1135 UV light. See Ultraviolet light UVB in combination psoriasis therapy, 1504, 1504t as D3 dose, 1006–1007 D status in elderly improved by, 1087

V Vascular calcification, 899 calcitropic hormones v., 904–905 D regulating, 904–905 overview, 904 regression v. treatment, 979 Vascular endothelial growth factor (VEGF) gene, 1,25(OH)2D3 targeting, 566, 567 Vascular inflammation, D signaling in regulating, 899 Vascular smooth muscle cells (VSMCs) 1,25(OH)2D3 actions in, 902–903 1α,25(OH)2D3 promoting migration of, 395–397, 396f Vasculature, direct D actions in, 902–904 Vasopressin. See Antidiuretic hormone VDCCs. See Voltage-dependent calcium channels VDR. See Vitamin D receptor VDR gene, 182–184 arrangement, 1210, 1211f complexity v. polymorphism identification, 1125f, 1129 defects causing HVDRR, 1207 exon-intron structure/polymorphism position, 1124, 1125f gene deletion prematurely terminating, 1222 genomic mapping to chromosome 12q13.1, 1122, 1123f haplotype importance in, 1135–1137, 1136f HVDRR caused by mutated, 111 LD measured across, 1125f, 1127 LD strength display across Caucasian, 1127, 1128f locus, 194–201 structure, 194–195, 195f nephrolithiasis v., 1142 odontoblasts expressing, 599 organization, 182–183, 183f polymorphism association analysis in disease states, 1137–1148

VDR gene (Continued) polymorphism in human, 172, 183–184, 200–201, 200f, 244f disease risk/functional consequences v., 243–246 ethnic variation v., 1127–1129, 1128t undiscovered/functionally significant, 245 polymorphisms/1α-hydroxylase v. PCa, 1626–1627 polymorphisms associated with cancer, 1572–1573 polymorphisms v. colorectal cancer/adenoma, 1619–1620 polymorphisms v. disease risk, 1121–1149 polymorphisms v. sequence comparisons, 1124–1126, 1125f premature termination v. HVDRR-related mutations, 1220–1222 promoters v. D analog selectivity, 1452f, 1453 12q13 locus, genomic structure, 1122, 1123f, 1124f RFLPs, 1124, 1125f sequence variations near anonymous markers, 1125f, 1135 splice site mutations prematurely terminating, 1222 Glu92fs, 1221f, 1222 Leu233fs, 1221f, 1222 stop mutations prematurely terminating, 1220–1222, 1221f Arg30stop, 1221–1222, 1221f Arg73stop, 1221, 1221f Gln152stop, 1221, 1221f Gln317stop, 1221f, 1222 Tyr295stop, 1220, 1221f Tyr295stop ochre, 1214t–1216t, 1220–1221 structural complexity, 193 structure, 194–196 structure/polymorphisms, 1122–1137 study size v. analysis of, 1137–1138 VDR homodimers D analog selectivity v., 1454–1455 RXR-independent 1,25(OH)2D3 signaling v., 319 VDR promoters, 196–200 human, 197–200, 198f, 199f nonhuman, 197, 198f targeting VDR through chromatin remodeling complex, 305–312, 306f VDR-RXR heterodimers 9-cis RA stimulating, 241 allosteric model of, 235–237, 236f cyclic model for transactivation by 1,25(OH)2D3-liganded, 237–243, 239f limits of, 241 cytoplasmic dimerization of, 369, 371f D ligands influencing DNA interaction with, 1455 DNA binding v. hexameric core binding motifs, 319 DNA complex formation of, 318f everted repeats, 317–319, 318f FRET experiments showing cytoplasmic, 364–365, 365f hVDR∆ in, 281 intestinal CYP enzymes regulated by, 246–248, 247f Mediator-D complex interacting with, 296, 296f 1,25(OH)2D3 signaling mediated by, 319 in 1,25(OH)2D3/VDR control of D responsive genes, 1320, 1320f structure/function of, 230–236, 231f, 234f therapeutic potential of ligands activating, 241 VDR signaling v. 1,25(OH)2D3/9-cis RA binding in, 237 VDR superagonist ligand v. enhanced formation of, 271–272, 272f VDRE binding sites for, 222–224, 223t VDRE interacting with, 539–541 VDRM-like activity-originating mechanisms in, 273 VDRE. See Vitamin D response element VDRE-BP. See Vitamin D response element binding protein VDRmem. See Membrane VDR VDRnuc. See Nuclear VDR

1888 Vegetarians diet for bone mass v. macrobiotic/vegetarian, 795 D metabolism v. vegetarian/omnivorous, 795 groups/sects/ethnicities, 917t rickets v., 1067, 1077 metabolic bone disease v. Chinese Buddhist, 917t VEGF gene. See Vascular endothelial growth factor gene Vertebrates, vitamin D3 synthesized by, 15 VICCs. See Voltage-independent calcium channels Vitamin A deficiency inducing squamous metaplasia, 614 in keratinocyte growth/differentiation, 614 McCollum/Davis discovering, 3–4 Vitamin B, McCollum/Osborne/Mendel discovering, 4 Vitamin D autocrine system, in prostate, 1608 Vitamin D binding protein (DBP), 117–128 actin-binding property conserved in vertebrates, 146 analog interaction with, 1456–1459 asymmetric unit having DBP-A/B molecules, 135 binding v. D compound structure, 1457 C-/D-/E-ring analogs binding poorly with, 1563–1565, 1567 changes with age, 1102–1103 concentration v. ecological factors, 126 D analogs’ affinity for, 1440–1441, 1441–1442, 1441t D3 metabolite/analog affinity for, 138, 141t in D3 pharmacology, 1002 D toxicity/free metabolite level v., 1367–1368 20-epi-1,25(OH)2D3/1,25(OH)2D3-induced hGH reporter gene expression v., 1433–1434, 1433f functional features, 121–127 gene, 117–121 allele distribution, 120–121 chromosomal location/linkages, 117–119 evolution within gene family, 118f, 119–120, 119f polymorphisms, 120–121 structural features, 117, 118f transcriptional orientation, 119, 119f hepatic failure/multiple trauma risk v., 148–149 interactions v. free hormone theory, 125–126 isoforms v. disease susceptibility, 121 LBD v. VDRnuc LBD, 387, 387f molecular interactions, 122–123, 122t as multifunctional protein, 124 overview, 117 in PCa etiology, 1681 physiological roles, 124–127 polymorphisms, 120–121 primate D hormone movement from, 354–355 proteolysis after receptor/ligand scavenging, 125 species variation in, 29 structural features, 117, 388–391, 390f–391f, 398f structure explaining unique functions, 149, 150f structure v. VDRnuc structure, 391, 392t synthesis/turnover, 121–122 three-dimensional structure, 135–149, 136f in vertebrate evolution, 120 Vitamin D (D). See also Hypovitaminosis D absorption v. age, 1101–1102 adolescents v. inadequate intake of, 816 analog development, 1489–1505 basic screening strategy, 1489–1492 strategy, 1489–1492 synthesis strategy, 1490–1492 analog selectivity mechanisms, 1449–2911 animal/cell culture muscle influenced by, 1809–1811

INDEX

Vitamin D (Continued) aromatase expression v., 859 β cell characteristics influenced by, 1767 β cells v., 1764–1767 benefits of higher levels of, 1370 in bone fracture clinical trials, 1112, 1113t bone fracture risk v., 1813–1814 bone mass influenced by, 1245–1247 cancer/differentiation v., 1571–1586 in cancer risk epidemiology, 1617–1629 cancer v., epidemiology of, 1571–1573 in cardiovascular medicine, 899–905 cell cycle influenced by, 1577–1580, 1578f–1579f chemistry/metabolism/circulation, 15–160 chicken embryonic development/egg hatchability v., 851–852 colon cancer v., 1709–1721 epidemiology of, 1709–1710 colon cancer v. protective role of, 866 colorectal adenoma v. dietary, 1619 colorectal adenoma v. plasma, 1619 colorectal cancer v. dietary/supplementary, 1618 colorectal cancer v. plasma, 1618–1619 compounds v. leukemic cell lines, 1730–1734, 1730t concentration lacking clinical relevance, 947–948, 948t controlling factors in supply of, 1293–1294 cutaneous production v. age, 823–824 definitions/models for studying rapid actions of, 583–585 detecting, 933–935 methodology for, 933–934, 935f sample extraction for, 933 silica cartridge chromatography in, 933–934, 934f detecting D metabolites and, 931–949 diabetes v., 1763–1774 dietary intake by elderly, 824 discovery, 4, 291 disease states v. D metabolites and, 948t diseases/conditions prevented by, 998–1000, 999t drug-oriented perspective on, 995–997 ED-71 as “long-lived,” 1440 epidemiology of breast cancer v., 1671 estimated requirement v. recommended intake, 785 estimating human serum, 933, 933t factors influencing metabolism of, 789–796 in fertility, 854–855 fetuses/neonates v. low maternal intake of, 841–843 FGF-23 in homeostasis of, 1193 food fortification with, 817–818 genomic/nongenomic influence on striated muscle, 1809 as GIO treatment, 1243–1248 in granuloma-forming disease, 1379–1380 hematological malignancy v., 1727–1736 high latitude summer generation of, 1087 historical perspective on, 3–8 in human physiology, 773–905 in HVDRR therapy, 1227 hydroxylation, 1599 25-hydroxylation, 17–19 in hyperparathyroidism development in renal failure, 1821 hypocalcemia due to malabsorption of, 1056 immune system in type 1 diabetes v., 1767–1773 insufficiency v. low intake of, 1087–1088, 1087f intestinal absorption by elderly, 824–825 intracellular trafficking in IDBP model, 360 ligands, interacting with DBP, 122–123 in mammary gland, 857–858 maternal D intake v. breast milk, 846–847

INDEX

Vitamin D (Continued) mechanism, 167–400 metabolically influencing β cells, 1764–1767 metabolism, 789–790 during pregnancy, 839–840 metabolism during lactation/weaning, 843–846 metabolism in pregnancy/lactation, 839–847 metabolism v. race/geography, 790–795 metabolites influencing β cells in vitro, 1764–1765 metabolites v. β cells in clinical trials, 1765–1767 mineralization influenced by, 478–480 molecular mechanisms in leukemia v., 1731–1733 mouse models lacking function of, 852 muscle contraction/relaxation v., 1809, 1809f muscle influenced by, 883–894 naming conventions, 15 non-bone effects of, 998–1000 not a vitamin, 4–5 optimal status of, 782–785, 784f osteoblast differentiation/activity influenced by, 649–658 in osteoporosis, 1101–1114 in ovary, 856 in parathyroid gland, 537–547 pathogenesis of impaired mineralization v., 1040–1044 in PCa, 1599–1611 PCa risk v. circulating, 1625–1626 PCa v., 1610–1611, 1611f, 1679–1700 PCa v. dietary, 1625 PDDR treatment with, 1201t perinatal actions, 803–808 pharmacology, 995–1010 photobiology, 37–43 physiological sources for activity of, 782–783 plasma half-life, 27 preparations v. GIO, 1247–1248 prostate targeted by, 1682–1683 rapid actions/nongenomic mechanisms, 583–589 in renal failure, 1313–1333 renal handling of Ca/P influenced by, 518–519, 518f replete/deficient state classification, 1024–1026 in reproductive organs, 851–860 resistance in breast cancer cells, 1667–1668 in rickets management, 1072 striated muscle influenced by, 1809–1811 supplementation/fortification, 784–785 supplements for related agents and, 1060t synthesis, 1599 synthesis v. fetal development, 852–853 target organs/actions, 565–768 in testis, 855–856 therapy in chronic renal failure, 1327–1332 tissue responsiveness/role in aging, 831–833 tumor cells influenced by, 1577–1582 in uterus, 856–857 winter oral intake v. photosynthesis of, 1087 Vitamin D2 (D2), 5f, 16f assays not monitoring therapeutic, 939–940, 940t clinical use, 11, 12t D intoxication v., 1108 derivatives’ metabolism, 1430–1431 in diet, 1599 isolation/identification, 5 molecular structure, 931 1,25(OH)2D3 analog synthesis from, 1490, 1491f 25OHD/VDR concentration/hypercalcemia v. supraphysiological, 1364, 1364f, 1364t

1889 Vitamin D2 (Continued) PDDR treated with, 1199f, 1201 skin diseases treated with, 1791 Vitamin D3 (D3), 5f, 16f. See also Hypervitaminosis D3 activation, 7, 8f in autoimmunity, 1753–1759 benefits overlooked, 1000 C-25 hydroxylation, monooxygenase activity, 48 clinical use of D2 v., 1005, 1006t D3/metabolite plasma concentrations v., 1362, 1363t deltanoids v. interconverting 6-s-cis/trans, 1408f, 1412, 1412f derivatives functionalized at C-24, 22–23 dietary/photosynthesized, 1599 differentiation/proliferation regulation pathways dissociated, 1581 distribution, 1006–1007 dose-response relationship with 25OHD3, 1003, 1005f dose v. tissue stores, 1007 falls v. Ca and, 1814, 1814f half-life, 1007, 1009t indications/clinical use, 997–1000, 1009t isolation/identification, 5, 7 metabolic transformations, 931, 932f metabolism in vertebrates, 220–222, 221f metabolism/regulation, 1000–1002, 1000f metabolite/analog formulas/DBP affinity, 141t metabolites, 8 molecular structure, 931 as new drug, 995–997 “normal” disease prevalence v. increased, 999t, 1001 24R,25(OH)2D3 as inactive metabolite of, 851–852 25OHD/VDR concentration/hypercalcemia v. supraphysiological, 1364, 1364f, 1364t in osteoclastogenesis, 665–681 as prosteroid hormone, 15 recommendations v. public health, 1026 serum 25OHD levels ensured by, 1003, 1004t skin diseases treated with, 1791 storage/25OHD conversion, 1007 supplementation safety margin, 1010 Vitamin D-dependent rickets type II. See Hereditary vitamin Dresistant rickets Vitamin D endocrine system, 383–386 adapting to 1,25(OH)2D3 concentration, 1000f, 1001 calbindin-D28K/VDR colocalization in understanding, 721 description of, 383, 384f discovery, 7 implicated in OA, 1144 osteoporosis/fracture v., 1141–1144 physiological process role, 291 pleiotropic effects v. VDR gene association analysis, 1138–1141, 1139t–1140t in prostate, 1608 RAS interaction with, 877–878, 878f VSCCs v. 1,25(OH)2D3, 751–757 Vitamin D hormone receptors, new functions indicated by, 6 Vitamin D hypothesis colorectal cancer v., 1618 experimental studies, 1606–1607 observational studies v., 1602–1606 PCa risk factors in, 1601, 1601t PCa v., 1599–1611, 1611f studies v., 1626 Vitamin D pseudodeficiency. See Pseudo-vitamin D-deficiency rickets (PDDR) Vitamin D-receptor interacting protein coactivator complex. See Mediator-D coactivator complex

1890 Vitamin D receptor (VDR), 167–184, 219–250, 1210–1212. See also Human VDR (hVDR); VDR gene; specific types of VDR ablated in rodents/humans, 224–225 ablation in mice, 341–348 absent in tibial dyschondroplasia, 579 activation functions, 176 affinity v. D analog activity, 1452f, 1453 allosteric model of signaling activation in, 235–237, 236f analog activity v. increased, 1456 analog selectivity in ligand-dependent regulation of, 1452f, 1456 analog selectivity v. RXR heterodimerization of, 1452f, 1454 analogs modulating, 1482–1483 antagonist/partial agonist response paradigms, 271 basal gene transcription apparatus contacting, 263 baT haplotype allele v. nephrolithiasis, 1142 as bile acid receptor, 866–867 binding sites in classical, 314–319 complex, 319–322 binding v. 20-epi D analogs, 1495–1498 binding v. deltanoid A-ring conformation, 1412, 1412f binding v. 1,25(OH)2D3 20-carbon epimerization, 1544 biochemical properties, 170–171, 170t bisphenol analogs as agonists of mutant, 1561–1562, 1561f in blood cells, 1728–1729 in cancer, 1571, 1572t in cancer cell growth regulatory response, 1583–1584 cardiovascular disease and genetics of, 900–901 cellular/tissue distribution, 168–170, 169t characterization, 168–171 chromatin remodeling v. transcriptional control by, 305–307, 306f CKD altering 1,25(OH)2D3 mediation by, 1317–1322 cloning, 171–172 in CNS, 1780–1781 coactivators associating with, 242–243, 264–268, 293–298 different means of recruitment for, 273 Ets-1, 298 interaction mechanism for, 293, 294f ligand-induced/tissue-selective recruitment of, 270–271 Mediator-D, 265–267, 265t multifunctional HAT activity assemblies of, 264–265 NCoA62/SKIP, 265–267 Smad 3, 298 SRC family of, 293–295, 295f SRC/p160 family of, 264 TFIIA/TFIIB/TFIID, 298 WSTF, 309–310, 310f cofactor complexes affecting, 263–274 ATP-dependent remodeling, 267–268 colonic hyperproliferation/tumorigenesis v., 1711–1712, 1712f comodulators, 242–243, 291–300 integrated model for activity by, 299–300, 300f in cultured metanephros, 521–522, 524f cyclic dynamics, 181–182, 181f, 182f D analog selectivity determined by interaction with, 1452–1456, 1452f D endocrine system v. calbindin colocalizing with, 721 D pocket structure, 142 defects causing hypocalcemia, 1056 deltanoids v. rickets-associated mutants of, 1409, 1411f in developing rodent kidney, 521–522, 522f, 523f differential activation by analogs, 1475–1482, 1476f differential activation by antagonists, 1480–1482 differential activation by noncalcemic selective agonists, 1478–1480 differential activation by superagonists, 1475–1478, 1476f

INDEX

Vitamin D receptor (Continued) discovered, 8–9, 167–168, 219–220 DNA binding, 171, 313–314 capacity in normal/renal failure rats, 543 endogenous gene promoters binding to, 177–178, 178f ethnic variation in polymorphisms of, 1127–1129, 1128t evolutionary insights from, 227–228 expression/abundance regulation, 193–210, 202t heterologous, 204–210 homologous, 201–204 expression in colon cancer, 1710–1711, 1711f, 1711t expression in prostate cells, 1834–1836 expression/regulation in breast cancer cells, 1667 expression/role in normal mammary gland, 1669–1670, 1670t functional analysis, 176–182 Gemini/1α,25(OH)2D3 binding efficiency, 287 gene targets/biological actions, 220–225 genomic structure surrounding, 1122, 1123f genotype responses as serum marker differences, 1133t, 1135–1136 genotypes v. psoriasis treatment response, 1784 in GHS rats, 1348–1349, 1349f GR sharing coactivators/corepressors with, 1240 group 1I/1H cholesterol derivatives recognized by, 227f, 229 similarities in, 229, 229f growth inhibition v. differentiation, 1713, 1713f helix 12’s intramolecular interactions in, 281, 283f homologous up-regulation defective in CKD, 1317–1318 in human BPH cells, 1834, 1835f import/export receptors interacting with, 368–371 import v. coactivators, 369, 370f interactions v. 20-epi analog binding, 1477–1478, 1478f intranuclear trafficking of, 374–376 in kidney, 520–523 lamprey, 227–228, 279 LBD hypothetical conformation, 1473, 1473f ligand immunomodulation v. graft rejection, 1519–1520 ligand-triggered protein-protein interactions, 313 ligands as immunoregulatory agents, 633 ligands enhancing regulatory T cells, 636 ligands inhibiting BPH, 1833–1840 localization, 363–365, 364f models, 363–365, 364f in shuttle model, 364–365, 364f localization in plasma membrane caveolae, 400 2MD inducing unique conformation of, 1550, 1550f 2MD promoting interactions by, 1550–1551, 1551f 2MD stimulating promoter binding by, 1549–1550, 1549f mineralization v. altered, 488–489, 489f mouse proximal/distal colon expressing, 1719 muscle, 885–886 muscle SOC influx v. TRPC3 proteins and, 892–893, 892f myeloid development v. expression of, 1729 neocytoplasmic trafficking, 365–374 nonsecosteroidal D mimics differentially regulating, 1565 nonsteroidal analogs inducing unique conformation of, 1567 normal/leukemic hematopoeitic cells expressing, 1728–1729 in normal/malignant colon cells, 1710–1712 novel co-regulatory complexes interacting with, 307–308, 307f nuclear export, 369–370, 372f nuclear export v. transcription, 371, 372f nucleocytoplasmic trafficking regulation, 371–374 1,25(OH)2D3 A-ring in transactivation of, 1472f, 1473–1474 1,25(OH)2D3 D-ring in transactivation of, 1475 1,25(OH)2D3 deficiency reducing parathyroid, 1317, 1318f

1891

INDEX

Vitamin D receptor (Continued) 1,25(OH)2D3 inducing focal accumulation of, 374–375, 375f 1,25(OH)2D3 ligand binding by v. hair cycle regulation, 234–235 in 1α,25(OH)2D3-modulated SOC influx, 893, 894f 1,25(OH)2D3 regulating PTH gene at, 540f, 541–542 1,25(OH)2D3 side chain in transactivation of, 1474–1475 1,25(OH)2D3 up-regulation, 201–204, 202t parathyroid CaR interacting with, 556, 556t phenotype v. fracture risk, 1143–1144 phosphorylation possibly regulating activity, 1454 PIC linked to, 291–292 “piggyback” nuclear import of, 369, 371f polymorphism association analysis in disease states, 1137–1148 polymorphism functionality, 1129–1137, 1131t–1134t polymorphism testing levels, 1129, 1135f polymorphism v. diabetes risk, 1773 polymorphisms, 1122–1126 polymorphisms influencing breast cancer risk, 1671–1672 polymorphisms influencing muscle function, 1813 polymorphisms/intestinal Ca absorption v. BMD, 1141–1142 polymorphisms v. cancer/hyperproliferative disease, 1145–1146 polymorphisms v. colon cancer development, 1712 polymorphisms v. disease risk, 1121–1149 polymorphisms v. PCa, 1604–1605 polymorphisms v. PCa risk, 1681–1682 polymorphisms v. structure/function in CKD, 1318–1319 in prostate, 1682–1683 shuttling v. transcription, 371 skeletal homeostasis not requiring, 344 species having characterized, 279–280 stabilization v. SRC–1 interaction, 1472–1473, 1472f structural domains, 173–176, 173f DNA binding, 174 Ligand binding, 174–176, 175f PXR v. ligand binding/heterodimerization/transactivation, 233–235, 234f zinc finger DNA binding, 229–233, 231f structural gene, 171–176 structural organization, 171 structural requirements for 1,25(OH)2D3 transactivation of, 1472–1475, 1472f structure/function, 229–236 subcellular distribution, 168f, 170 subcellular trafficking, 363–376 superagonist/selective coactivator recruitment paradigms, 271–273 target gene diversity, 313–322 tissue distribution, 193–194, 228 tissue effects of low 1,25(OH)2D3/abnormal, 1322–1327 tissue selective ligands in, 270–273 established paradigms for, 270–271 tissue source/species conservation, 172 as toxic bile acid sensor, 863–869 in toxicity, 1362–1365 transcription regulated by integrated pathways, 268–270, 269f transcription v. 1α,25(OH)2D3/analog side chain modification, 1474–1475, 1474t translocation mechanisms determining analog selectivity, 1452f, 1454 variants, 195–196, 195f, 196f VDR promoter targeting, 305–312, 306f VDRE interaction v. interference footprinting protocols, 542 Vitamin D response element binding protein (VDRE-BP) dominant negative action, 354, 354f compensation for, 359–360 New World primate, 354–355

Vitamin D response element (VDRE) analogy selectivity v. genes with, 1454–1455 binding proteins in intracellular, 351–361 in bone proteins, 712, 713t clusters, 320–321 complex, 316t as complex/multiple TF binding site structures, 320 direct 1,25(OH)2D3 osteoblast modulation v., 654 DNA bend induced by VDR-RXR bonding to, 541 DNA binding polarity, 180, 317–318 DR6/DR3-type, 316t, 317 DR3-type, 314–315 classical VDRE structure in, 314 DBD-DBD distance in, 318, 318f multiple signaling pathways v., 316–317 strongest VDR-RXR heterodimer binding in, 314–315 DR4-type, 316t, 317 ER9-type, 316t DBD-DBD distance in, 318, 318f in human ECaC promoter region, 527 human VDR promoter, 200 known natural types of, 316t location/sequence of positive natural, 177, 177t mapping CT/CGRP gene, 694–696, 695f neuroendocrine-specific/cAMP enhancers in, 693–696 nonhuman VDR promoter, 197 in 1,25(OH)2D3 regulation of PTH gene expression, 539–541 1,25(OH)2D3 target gene promoter regions having, 314 positive/negative/optimal, 222–224, 223t rat CYP24A1 proximal promoter region, 93–97, 94f, 95f simple v. complex, 319–320 software v. identifying complex, 322 VDR interaction v. interference footprinting protocols, 542 VDR-regulated genes originating, 222–224, 223t VDR/RXR recognition by PTHrP, 743–744 VDR structure supporting specific binding to, 230–232, 231f whole-genome screening for putative, 322 Vitamins, discovery of, 3–4 Voltage-dependent calcium channels (VDCCs), in nongenomic 1α,25(OH)2D3 actions in muscle, 890–893, 890f Voltage-independent calcium channels (VICCs), Osteoblast VSCCs interacting with, 755, 755f Voltage-sensitive calcium channels (VSCCs), 751–753 α1 subunit types/functions of, 752, 752t L-type Ca2+ inactivating, 755–756, 756f 1,25(OH)2D3 v. open time in, 754–755, 755f pore-forming α1 subunit transmembrane organization in, 752, 752f subunit structure of, 752, 752f membrane/nuclear action cross-talk in, 757 1,25(OH)2D3 Ca2+/transcriptional responses by, 756–757 1,25(OH)2D3 regulating, 1782–1783 1,25(OH)2D3 v., 753–754 1,25(OH)2D3 v. D endocrine system, 751–757 VSCCs. See Voltage-sensitive calcium channels VSMCs. See Vascular smooth muscle cells

W Weaning BMC/BMD after, 845–846, 845f D/Ca metabolism after, 845 maternal Ca economy after, 846, 846f

1892 Weight, serum 25OHD v., 1007 Whites, dietary calcium reduction response by, 778 Wild-type (WT) mice dexamethasone influencing Ca absorption in, 445, 447f dietary intervention v. Ca absorption in, 434f, 436t, 437–438 intraperitoneal glucose tolerance test in, 1765, 1765f low-Pi diet v. Hyp and, 464f, 465 Williams syndrome defective chromatin remodeling complex WINAC in, 238–240 hypervitaminosis D/hypercalcemia in, 1359 1,25(OH)2D3 repression of CT/CGRP expression v., 690 WSTF gene deleted in patients with, 267–268 Williams syndrome transcription factor (WSTF) as VDR interactant, 307–308, 307f VDR ligand-induced transactivation coactivated by, 309–310, 310f WINAC/VDR interaction through, 268 Wilson’s disease, hypoparathyroidism/hypocalcemia v., 1053 WINAC coactivator complex components, 308, 308f cooperative function with co-regulator complexes, 311–312 nucleosome arrays disrupted by, 309, 309f, 310, 311f purification/identification, 308, 308f VDR promoter targeting mechanism, 310–311, 310f, 311f in VDR transcription model, 268–270 Williams syndrome associated with defective, 238–240 WSTF in VDR interaction with, 268 Women. See also Latinas BMD/insufficiency/secondary hyperparathyroidism in, 1089 bone remodeling markers in young/elderly, 1089–1090, 1090t Ca absorption v. intake in, 779–780, 779f, 780f Ca absorption v. load in, 779–780, 780f Ca v. age-related 1,25(OH)2D resistance in, 1104–1105, 1105f cervical cell carcinoma common in, 857 D deficiency in black, 791 D metabolism in Iranian, 794 D supplementation reducing bone mass loss in, 1094 D v. risk of falls in postmenopausal, 1114 estrogen v. 1,25(OH)2D3 in postmenopausal, 1260 estrogen v. PTH/bone resorption in, 1108–1109, 1108f normative histomorphometric data for, 956t 1,25(OH)2D3 v. postmenopausal osteoporosis in, 1111–1112 25OHD v. age in, 1101, 1102f 25OHD v. PTH in French post-menopausal, 1021–1022, 1021f OPG v. breast carcinoma-related bone metastases in, 676 postmenopausal OPG clinical trials in, 676 VDR target gene associated with osteoporosis in, 246 rickets prevention strategies v., 1072 serum 1,25(OH)2D in, 827 winter hypovitaminosis D in young/elderly, 1093, 1093t

INDEX

WSTF. See Williams syndrome transcription factor WT mice. See Wild-type mice

X X-linked hypophosphatemia (XLH), 467, 981 bone modeling abnormalities in, 982, 983f extraskeletal ossification in, 982–983, 984f, 985f FGF-23 in, 1193 genetic defect underlying, 1170–1172 Hyp/Gy mouse models for, 463, 463t Hyp mouse models for, 464f, 465 in original D-resistant rickets case, 1207 osteoarthritis in, 983 pathogenesis, 1172 pathophysiology, 1168–1170 phosphate homeostasis in, 1168–1173, 1169t radiography of rickets v., 922, 923f, 924f reinforcing FGF-23 as phosphatonin, 1165 treatment, 467, 1172–1173 X-linked recessive hypophosphatemia (XLRH), phosphate homeostasis in, 1175 X-ray diffraction, bone mineral presence determined by, 481, 482f X-rays D-deficient osteomalacia revealed by, 481 discovery of, 967 metabolic bone disease evaluated with, 914, 921–922, 923f, 924f Xenobiotic detoxification, VDR in, 228, 229f XLH. See X-linked hypophosphatemia XLRH. See X-linked recessive hypophosphatemia

Y Yeast higher eukaryote complex diversity v., 266 Mediator-D/related complex function studied in, 266

Z Zebrafish VDR (zVDR) bound to 1α,25(OH)2D3, 284–285, 284f characterized, 279 ZK 159222, 1481f, 1482 zVDR. See Zebrafish VDR zVDR-1α,25(OH)2D3 complex, crystal structure, 284–285, 284f zVDR-Gemini complex channel extending original pocket in, 287–288 structure, 287–288 zVDR-1α,25(OH)2D3 structure v., 287